Homologation of allylic alcohols. An approach to cyclic and acyclic polyoxygenated compounds

Article abstract of DOI:10.1016/S0040-4039(99)02263-7

The combination of the Sharpless asymmetric epoxidation reaction with a sulfur ylide mediated synthesis of allylic alcohols from epoxides provides a powerful iterative process for the production of polyoxygenated compounds. The alkene installed in the sulfur ylide reaction has also been used in a number of ring closing metathesis reactions to produce highly oxygenated cyclic compounds. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(99)02263-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 1255–1259
Homologation of allylic alcohols. An approach to cyclic and
acyclic polyoxygenated compounds
Ryan J. Davoille, David T. Rutherford and Steven D. R. Christie ∗
Department of Chemistry, University of Loughborough, Loughborough, LE11 3TU, UK
Received 12 October 1999; accepted 7 December 1999
Abstract
The combination of the Sharpless asymmetric epoxidation reaction with a sulfur ylide mediated synthesis
of allylic alcohols from epoxides provides a powerful iterative process for the production of polyoxygenated
compounds. The alkene installed in the sulfur ylide reaction has also been used in a number of ring closing
metathesis reactions to produce highly oxygenated cyclic compounds. © 2000 Elsevier Science Ltd. All rights
reserved.
The synthesis of highly oxygenated organic molecules continues to be an area of interest for synthetic
chemists. The stereo- and regio-controlled placement of oxygen represents a particular challenge. Recent
examples of iterative placement of oxygen have been shown by Hanessian,1 Lipshutz,2 McDonald,3
Sweeney4 and Rainer.5 Of particular relevance to the chemistry described herein is the use of silylated
epoxides to produce 1,2-diols in an iterative fashion employed by Wicha.6 We wished to combine two
reactions in order to produce a new protocol for directed oxygen placement on a growing alkyl chain. The
Sharpless asymmetric epoxidation7 (SAE) is a well known reaction and has been widely used in organic
synthesis for a number of years. It provides reliable, and importantly, predictable levels of enantiomeric
excess in the epoxidation of allylic alcohols. Mioskowski has reported the conversion of epoxides to
allylic alcohols using a sulfur ylide.8 The ylide reacts regioselectively at the less hindered end of the
epoxide, transfering a methylene group with concomitant loss of dimethyl sulfide. We envisaged that
these two reactions could be used in tandem: one produces epoxides from allylic alcohols, the other
allylic alcohols from epoxides. The result is then a simple protocol for homologation of allylic alcohols.
In addition, we also hoped to use the alkene installed in the sulfur ylide reaction to provide an entry to a
number of substrates suitable for ring closing metathesis reactions. This would extend the scope of the
protocol to cyclic as well as acyclic targets.
∗
Corresponding author. E-mail: s.d.christie@lboro.ac.uk (S. D. R. Christie)
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02263-7
tetl 16191

1256
In order to test this proposal, we started with racemic glycidol (Scheme 1). Protection of the primary
alcohol as its TBDMS ether (1) was followed by the first of the sulfur ylide reactions. As reported
by Mioskowski,8 treatment of trimethylsulfonium iodide with n-butyllithium produced the requisite
dimethylsulfonium methylidene. Reaction of three equivalents of this with the epoxide gave the allylic
alcohol (2) in 66% yield.9 In these early experiments, we used an alternative epoxidation methodology
to the Sharpless protocol. Thus, treatment with MCPBA gave the epoxide (3) in 61% yield in a
diastereoisomeric ratio of ca. 1:1. In order to test the substrate directing effect of the hydroxyl group
already in place, we then used VO(acac)2/tBuOOH. This produced the same product, but with an
increased diastereomeric ratio of ca. 4:1. The epoxy alcohol product was now ready to be subjected to a
second cycle of the iterative procedure. At this juncture, we were concerned that the acidic proton of the
free hydroxyl group may quench the sulfur ylide, and possibly also facilitate a Payne rearrangement of the
epoxide from the resultant alkoxide. To prevent this possibility, we decided to protect the alcohol. This
protection would also allow for an orthogonal protection strategy on each hydroxyl as the chain grows.
Thus, if we were to produce a polyol with this methodology, selective deprotection of the individual
alcohols would allow manipulation of selected sites as they were unmasked. At this stage, we chose MEM
as a suitable group. A subsequent sulfur ylide reaction produced the allylic diol (5) in good yield. Having
established that the protocol was viable, we then sought to achieve the sequence in a stereocontrolled
manner.
Scheme 1. (a) Me3SI, BuLi, THF, −25°C; (b) MCPBA, DCM; (c) VO(acac)2, tBuOOH; (d) MEMCl, Hunig’s base
Starting with benzyl protected racemic glycidol (6) (Scheme 2), a sulfur ylide reaction gave the allylic
alcohol (7), as expected, which was then subjected to a Sharpless kinetic resolution epoxidation. Using
standard conditions,10 one enantiomer of the alcohol was transformed into its epoxide (8), while the other
was untouched. The enantiomeric excess of the desired product was found to be 96% by HPLC.11 The

1257
free alcohol was again protected as its MEM ether (9) before opening the epoxide with the sulfur ylide
to give the allylic alcohol (10) in 83% yield.
Scheme 2. (a) Me3SI, BuLi, THF, −25°C; (b) Ti(OiPr)4, (+)-diisopropyl tartrate, tBuOOH; (c) MEMCl, Hunig’s base; (d) TiCl4,
0°C, DCM; (e) 2,2-dimethoxypropane; (f) ozone, DCM
Removal of the MEM ether protection and conversion of the diol (11) into its dimethyl acetal was
carried out before ozonolysis transformed the alkene into the expected carbonyl functionality (13). The
yield for the ozonolysis was a moderate 43%, but it allowed us to confirm the stereochemistry of the
product since this short scheme comprises a synthesis of a protected tetrose.12 In this short synthetic
scheme, we have used Sharpless epoxidation chemistry and the sulfur ylide reaction to build up the
molecule in an iterative fashion by homologation of the allyl alcohol functionality. The only other
reactions employed in this scheme are protections and deprotections, and the final ozonolysis. It should
be noted that this procedure could easily be extended to other longer polyol chains by repeating the
iterative cycle an appropriate number of times.
In order to diversify this procedure, and to test the scope of the protocol, we wished to investigate
the potential for transformations into other types of skeleton. Since the sulfur ylide reaction introduces a
terminal alkene group, we envisaged using a ring closing metathesis13 to cyclise the chain into oxacycles
of varying sizes. In this respect, we took the allylic alcohol (7) (Scheme 3) and allylated the hydroxyl
group. This diene (14) was then cleanly metathesised with Grubbs’ catalyst to produce the dihydrofuran
(15) in an excellent 91% yield.
Scheme 3. (a) NaH, THF, allyl bromide; (b) Grubbs’ catalyst, DCM, room temp.
To make a dihydropyran, we extended the chain, as shown in Scheme 4. Allylation of the alcohol
(16) was achieved at low temperature in order to avoid a Payne rearrangement. The second sulfur ylide
reaction proceeded as expected to produce the diene (18). Exposing this to Grubbs’ catalyst gave the
dihydropyran (19) in good yield upon stirring at room temperature.
We have also looked at the possibility of attempting the sulfur ylide reaction on a bis epoxide, since
this would allow the homologation to proceed in two directions at once (Scheme 5). Epoxidation of 1,5-
hexadiene with MCPBA proceeded smoothly to give the expected product (21) in good yield. Reaction
with excess sulfur ylide gave the diol (22) in an unoptimised 35% yield, however, each reaction is

1258
Scheme 4. (a) NaH, THF, allyl bromide, 0°C; (b) Me3SI, BuLi, −25°C; (c) Grubbs’ catalyst, DCM, room temp.
therefore occurring in at least 60% yield. With two terminal alkenes having been placed in the chain, we
subjected the diol to a metathesis reaction. Stirring the diene in dichloromethane with Grubbs’ catalyst
resulted in quantitative conversion to a mixture of the syn and anti isomers of cyclohexene-1,4-diol (23)
which were isolated in 71% yield. We were particularly pleased with this result, as it has been reported
that Grubbs’ catalyst is sensitive to allylic substitution.14 Here we have two allylic alcohols, and have
successfully cyclised the diene without requiring protection of the hydroxyl groups.
Scheme 5. (a) MCPBA, DCM; (b) Me3SI, BuLi, THF, −25°C; (c) Grubbs’ catalyst, DCM, room temp.
Conclusion: We have illustrated a new procedure for homologation of allylic alcohols. The scheme
uses two simple reactions to produce a convenient route to densely functionalised organic molecules. We
have demonstrated how this can be used to furnish straight chain polyols. In addition, we have utilised
the terminal alkene that is installed by the sulfur ylide reaction, and used this in a metathesis approach to
oxygenated dihydrofuran, dihydropyran and cyclohexene diol compounds. Application of this protocol
to natural product synthesis is underway, and will be reported in due course.
Acknowledgements
The authors wish to thank the EPSRC (Quota award to DTR) and the Nuffield Foundation (Undergra-
duate Bursary to RJD) for funding. SDRC wishes to thank Loughborough University and Pfizer Central
Research for additional funding.
References
1. Hanessian, S.; Wang, W.; Gai, Y.; Oliver, E. J. Am. Chem. Soc. 1997, 119, 10034.
2. Lipshutz, B. H.; Barton, J. C. J. Org. Chem. 1998, 53, 4495.
3. Bowman, J. L.; McDonald, F. E. J. Org. Chem. 1998, 63, 3680.
4. Bange, J.; Haughan, A. F.; Knight, J. R.; Sweeney, J. J. Chem. Soc., Perkin Trans. 1 1998, 1039.
5. Rainer, J. D.; Allwein, S. P. J. Org. Chem. 1998, 63, 5310.
6. (a) Raubo, P.; Wicha, J. Tetrahedron: Asymmetry 1996, 7, 763. (b) Achmatowicz, B.; Wicha, J. Tetrahedron: Asymmetry
1993, 4, 399.
7. Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102, 5974. For recent reviews, see: (a) Johnson, R. A.; Sharpless, K.
B. In Comprehensive Organic Synthesis; Trost, B. M.; Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 7, pp. 389–436. (b)
Johnson, R. A.; Sharpless, K. B. In Catalytic Asymmetric Synthesis; Ojima, I., Ed.; VCH: New York, 1993; pp. 103–158.
For stereoselective dihydroxylation of allylic alcohols, see: Cha, J. K.; Christ, W. J.; Kishi, Y. Tetrahedron 1984, 40, 2247.
8. Alcarez, L.; Hamett, J. J.; Mioskowski, C.; Martel, J. P; Le Gall, T.; Shin, D.-S.; Falck, J. R. Tetrahedron Lett. 1994, 35,
5449. See also: Baylon, C.; Heck, M.-P.; Mioskowski, C. J. Org. Chem. 1999, 64, 3354.

1259
9. Preparation of 1-[(phenylmethyl)oxy]but-3-en-2-ol. To a solution of dimethylsulfonium iodide (7.46 g, 36.58 mmol) in
tetrahydrofuran (150 ml) at −25°C was added n-butyllithium (14.14 ml, 35.35mmol). The reaction mixture was stirred for
30 minutes prior to the addition of benzylglycidol (2 g, 12.19 mmol) and then allowed to warm to 0°C over approximately
30 minutes. The reaction mixture was then stirred for a further 4 h, cooled to 0°C quenched with water and extracted with
diethylether. The organic extracts were combined and dried with magnesium sulfate prior to the solvent being removed under
reduced pressure. The crude material was then purified by silica gel chromatography (eluent petrol:diethylether 4:1) to yield
1.41 g (65%) of the title compound as a colourless oil. IR (neat film) 3426, 2860, 1645, 1453, 1104 cm−1. 1H NMR (250
MHz, CDCl3) δ 7.23–7.27 (m, 5H, Ar), 5.80 (ddd, 1H, J=17, 10, 5, -CH=CH2), 5.32 (dt, 1H, J=17, 1, -CH=CH2), 5.15 (dt,
1H, J=10, 1, -CH=CH2), 4.52 (s, 2H, ArCH2), 4.34–4.30 (m, 1H, -CHOH), 3.48 (dd, 1H, J=9, 3, -OCH2), 3.34 (dd, 1H, J=9,
8, -OCH2). 13C NMR (250 MHz, CDCl3) δ 138.00, 137.00, 136.71, 128.40, 127.75, 116.27, 74.03, 73.29, 71.40. HRMS
(EI) calcd for C11H14O2178.09938, found 178.09920.
10. Martin, V. S.; Woodard, S. S.; Katsui, T.; Yamada, Y.; Ikeda, M.; Sharpless, K. B. J. Am. Chem. Soc. 1981, 103, 6237.
11. Chiracel OD column, eluent hexane:isopropanol 97:3, flow rate 0.75 ml/min, retention time 44.38 min.
12. [α]D=−21.2° (c=1.3, CHCl3 25°C) (lit.15 –22.3°).
13. For a recent review, see: Grubbs, R. H.; Chang, S. Tetrahedron 1998, 39, 4413.
14. For examples, see: (a) Sturino, C. F.; Wong, J. C. Y. Tetrahedron Lett. 1998, 39, 9623. (b) Sellier, O.; Van de Weghe, P.;
Eustache, J. Tetrahedron Lett. 1999, 40, 5859.
15. Munier, P.; Guidicelli, M.-B.; Picq, D.; Anker, D. J. Carbohydr. Chem. 1996, 15, 739.