Scheme 2. Radical Cyclization via Deoxygenation
Scheme 3. Synthesis of Radical Precursora
oxalate 6 underwent a transannular cyclization (path a) before
conformational interconversion (path b), enantioenriched
products would be obtained (Figure 1).
a Reaction conditions: (a) 1-tert-butylthio-1-(trimethylsilyl) oxo-
ethene, (R)-BINOL, Ti(OiPr)4, 4 Å MS, ether, -35 °C, 89%, 89%
ee. (b) TBSOTf, 2,6-lutidine, CH2Cl2, 100%. (c) DIBALH, CH2Cl2,
-25 °C, 93%. (d) I2, PPh3, CH2Cl2, imidazole, 0 °C, 94%. (e) NaH,
allyl dibenzylmalonate, THF, 0 °C to rt, 96%. (f) 20 mol % 16,
CH2Cl2, 1 mM, 40 °C, 16 h, 88%. (g) TBAF, THF, 0 °C to rt,
95%.
such as those developed by Grubbs.7 Initial studies employing
15 as a catalyst (1 mM, CH2Cl2, 40 °C) did not provide the
desired 10-membered carbocycle; instead, only acyclic dimer
products were observed. However, it was found that Grubbs’
second-generation catalyst 16 was effective in forming the
strained ring in good yield (88%) under high dilution
conditions (1 mM in substrate).
The ring-closing metathesis product was isolated as a
mixture of alkene products (ca. 1:1.7), initially assigned as
the E- and Z-cyclodecenes. This result was expected since
ring-closing metathesis in the synthesis of large rings
generally gives mixtures of olefin isomers.7b,8 However,
attempts to separate the alkene products, both as the silyl
ethers and the deprotected alcohols, were unsuccessful by a
variety of chromatographic methods. This result and the
inability to enrich one isomer over the other by chemical
means9 led us to consider the alkene products were not olefin
isomers but rather conformational isomers. Variable tem-
perature NMR studies were conducted to confirm the
presence of two different conformers. 1H NMR experiments
carried out from -30 to 100 °C in toluene-d8 showed
coalescence of the doubled peaks at temperatures at and
above 37 °C (Figure 2). This coalescence temperature
corresponds to an interconversion barrier of ca. 15.5 kcal/
Figure 1. Racemization pathway.
Racemization would occur through a disfavored chair/boat
interconversion. This is an example of memory of chirality
because the chiral information is held by the conformation
of the radical intermediate and not the radical center.
We set out to synthesize the radical precursor 4 through a
ring-closing metathesis of the bisolefin 14 (Scheme 3). The
synthesis of bisolefin 14 began with Mukaiyama-Keck aldol
condensation of 5-hexenal with the trimethylsilyl ketene
acetal derived from tert-butyl thioacetate to afford thioester
11 in 89% yield and 89% ee. Silyl ether formation followed
by reduction of the ester provided alcohol 12 in 93% over
two steps. Conversion of the resultant alcohol to the iodide
13 and coupling with anionic allyl dibenzylmalonate fur-
nished the requisite bisolefin 14 (90%, 2 steps).
(7) (a) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999,
1, 953-956. (b) Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54, 4413-
4450. (c) Grubbs, R. H.; Kirkland, T. A. J. Org. Chem. 1997, 62, 7310-
7318.
(8) (a) Lee, C. W.; Grubbs, R. H. Org. Lett. 2000, 2, 2145-2147. (b)
Furstner, A.; Muller, T. Synlett 1997, 1010-1011. (b) Furstner, A.;
Langemann, K. Synthesis 1997, 792-803.
(9) Selective dihydroxylation with R-AD-mix (60% conversion) returned
the starting cyclodecene of the same isomeric ratio.
With the cyclization precursor 14 in hand, we investigated
the ring-closing metathesis utilizing ruthenium alkylidenes,
2714
Org. Lett., Vol. 6, No. 16, 2004