10848
J. Am. Chem. Soc. 1999, 121, 10848-10849
Scheme 1
Highly Stereoselective Synthesis of a Chiral Methyl
Group by a Facially Controlled Sigmatropic
[1,5]-Hydrogen Shift
Christoph Dehnhardt,† Matthew McDonald,‡ Sungsook Lee,‡
Heinz G. Floss,‡ and Johann Mulzer*,†
Institut fu¨r Organische Chemie, UniVersita¨t Wien
Wa¨hringer Strasse 38, Vienna, Austria
Department of Chemistry, Box 351700
UniVersity of Washington
Seattle, Washington 98195-1700
ReceiVed July 15, 1999
Chiral methyl groups in the form of chiral acetic acid (R)- and
(S)-1 have been invaluable for the elucidation of numerous
biochemical mechanisms.1 So far, two basically different ap-
proaches to (R)- and (S)-1 have been reported: The first one
(Arigoni strategy)2 uses a cascade of ene-retroene reactions to
introduce the three isotopes of hydrogen at one and the same
carbon atom via a mechanistically defined stereochemical path-
way. Optical activity is achieved by chemical optical resolution
of a suitable alcohol intermediate. The second one (Cornforth,3
Floss,4 Altman5 strategy) hinges on SN2 reactions of anionic
hydrogen donors with configurationally defined epoxides or
primary tosylates. In this case, asymmetry is introduced by optical
resolution (Cornforth), by means of a chiral reagent (Floss) or
by starting from the chiral carbon pool (Altman). Recently, we
reported a synthesis of (R)- and (S)-1 via a base-induced [1,3]-
hydrogen shift,6 but with a disappointingly low ee-value of ca.
40%.
Scheme 2
Two reactive diene conformers exo- and endo-2 are involved,
from which dienes 3 and 4 are generated in a 1.5:1 ratio at 250
°C along pathways A and B, respectively. These dienes are (E/
Z) isomers and contain the stereogenic center at C-5 in opposite
configurations. Thus, although the Woodward-Hoffmann rules
are obeyed by conformers exo- and endo-2 individually, the
overall rearrangement lacks stereocontrol. Even more seriously,
about 27% of the starting material 2 is still present after 2 h at
250 °C. Attempting to drive the rearrangement to completion at
275 °C, Roth discovered that altogether ten products are formed
and the [1,5]-hydrogen shift obviously becomes a mess.
On considering how Roth’s experiment might be adapted to
the synthesis of (S)-1 it occurred to us that dienes (Z)-7a-d might
be suitable candidates for the following reasons. First of all, the
isomerization of 7 to 8 should be irreversible as two trisubstituted
olefinic bonds are generated from a mono- and a disubstituted
one. Additionally, due to the presence of the 2-phenyl substituent,
the π system changes from a cross-conjugated one in 7 to a more
stable linearly conjugated one in 8. Earlier experiments10 have
also shown that the phenyl group stabilizes 7 and 8 toward
polymerization and thermodimerization and prevents further
double bond migrations in 8.
Increasing the bulkiness of substituent R should result in a clear
preference of pathway A over B, but it remained to be tested
whether the required selectivity of >100:1 could be achieved in
this way. At this stage, the configurational purity of the 1,2-double
bond in the rearrangement product, which was easy to check via
analytical GC, could serve as an analytical probe of the cleanness
of the concomitant [1,5] chirality transfer. Hence, dienes 7a-d
were prepared as (E/Z) isomer mixtures from phosphonium
bromide11 5 and racemic aldehydes rac-6a-d as shown in Scheme
2 and separated by HPLC. Isomers (Z)-7a-d were submitted to
thermal rearrangements at 200 °C in benzene. In fact, the reaction
was complete within 4 h and a maximum of two products were
found. (E)- and (Z)-8a-c were formed from (Z)-7a-c and (E)-
8d was the sole product from the rearrangement of (Z)-7d, as
In this communication we want to disclose a novel synthesis
of (S)-1 in high optical purity that is based on the mechanism-
controlled chirality transfer involved in thermal sigmatropic [1,5]-
hydrogen shifts. In contrast to [1,2]-, [2,3]-, and [3,3]-sigmatropic
rearrangements which belong to the standard repertoire of organic
synthesis, [1,5]-migrations, despite intensive mechanistic inves-
tigations in the early seventies,7,8 have found only limited
application in stereoselective transformations. This is mainly due
to the formation of side products, and the reversibility of the
reaction and the lack of absolute stereocontrol, for which Roth’s
reaction8 (Scheme 1) may serve as an illustration. According to
the Woodward-Hoffmann rules9 the [1,5]-hydrogen migration
proceeds suprafacially with respect to the cisoid butadiene
skeleton.
† Universita¨t Wien.
‡ University of Washington.
(1) Floss, H. G.; Lee, S. Acc. Chem. Res. 1993, 26, 116.
(2) Townsend, C. A.; Scholl, T.; Arigoni, D. J. Chem. Soc., Chem. Commun.
1975, 921.
(3) Cornforth, J. W.; Eggerer, H.; Buckel, W.; Donninger, C.; Malaby, C.;
Redmond, J. W. Nature 1970, 226, 517.
(4) Kobayashi, K.; Jadhav, K.; Zydowsky, T. M.; Floss, H. G. J. Org. Chem.
1983, 48, 3510.
(5) Altman, L. J.; Han, D. Y.; Bertolino, A.; Handy, G.; Laungani, D.;
Muller, W.; Schwartz, S.; Shanker, D.; de Wolf, W. H.; Yang, F. J. Am. Chem.
Soc. 1978, 100, 3235.
(6) Mulzer, J.; Wille, G.; Bilow, J.; Arigoni, D.; Martinoni, B.; Roten, K.
Tettrahedron Lett. 1997, 38, 5469.
(7) Hasselmann, D. Houben-Weyl; Vol. E21d, 4421.
(8) Roth, W. R.; Ko¨nig, J. Chem. Ber. 1970, 103, 426.
(9) Woodward, R. B.; Hoffmann R. The ConserVation of Orbital Symmetry;
VCH-Verlagsgesellschaft: Weinheim, 1970.
(10) Mulzer, J.; Ku¨hl, U.; Huttner, G.; Evertz, K. Chem. Ber. 1988, 121,
2231.
(11) Mulzer, J.; Bru¨ntrup, G.; Ku¨hl, U.; Hartz, G. Chem. Ber. 1982, 115,
3453.
10.1021/ja992498e CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/06/1999