of R1 slightly enhances the energy difference between both
configurations, whereas introduction of aromatic side groups
in 9d has no influence. We then turned our attention on the
effect of further substitution of the 1,3-oxazolidine moiety
(entry 5). It is necessary that substituents at carbons 4 and 5
of the heterocycle are in a trans relationship.8 However, this
variation only led to dramaticly less selectivity in combina-
tion with a lowered rotational barrier as can be seen from
the lower coalescence temperature of 9e. To our delight,
increasing the steric bulk of the arenesulfonyl group shows
substantial impact on the diastereoselectivity; a ratio of 98:2
is observed in ent-9f.
We finally varied the size of the substituent at the
phosphorus atom (entries 7-10). On going from flat phenol
to branched alcohols we expected stronger steric interaction
of the heterocycles with the residue at the P-atom. Introduc-
tion of an isopropoxy group led to the described behavior
of compound 9g in the VT NMR spectra (vide supra). With
this strongly enhanced selectivity of more than 95:5 in hand,
we hoped that an even more branched substituent (compound
9i) would lead to higher selectivity and clear spectra. In
contrast, we obtained spectra with a more developed
coalescence phenomenon (more equal populations of the
interchanging signals furnishes more broadened signals at
TC) and a selectivity below our expectations of 95:5. We
therefore further investigated this behavior by slightly
increasing the size of the alkyl residues R1 in the hetero-
cycles. Isopropoxy compound 9h (entry 8) showed spectra
comparable to phosphite 9g. Finally, the highly branched
phosphite 9j again displayed a decrease in selectivity.
Interestingly, the ratio of 94:6 is slightly lower than that of
compound 9i bearing a sterically less demanding alkyl group
R1. This makes us assume that in phosphites 9i and 9j the
molecule might be somewhat sterically overcrowded, leading
to a decrease in selectivity.
Scheme 4. Michael Addition of Diethylzinc to
2-Cyclohexenone
The enantioselectivities obtained are found in Table 2. In
general, (2R,4R)-configured ligands furnish (R)-12.10 On one
hand, ligand 9g, which displays one of the best ratios
concerning the chiral axis, delivers the best result in the
model reaction, but on the other hand, there is no general
trend between the diastereoselectivity within the free ligand
and the ee in the Michael product 12. Nevertheless, these
results demonstrate the potential that lies within our highly
modular ligands.
In conclusion, we presented the synthesis of novel and
highly modular (four variable sites) bisphenol-based ligands.
All synthetic steps proceed in good to excellent yield and
can be performed on a gram scale. To the best of our
knowledge, this is the first bisphenol-based phosphorus
ligand reported that bears the chiral information at the 3,3′-
positions. The 1,3-oxazolidinyl substituents are relatively far
away from the chiral biaryl axis and have an unexpectedly
high impact on the diastereomeric ratio. The selectivity of
up to 98:2 for this type of compounds is the best so far
reported. The strong influence of the arenesulfonyl group
on the diastereomeric ratio renders the free hydroxy oxazo-
lidines 8 an interesting subject of research suitable for the
use as diol ligands, complementing BINOL compounds. This
project is currently under way, as well as further investiga-
tions in the field of phosphites 9.
Acknowledgment. Generous support by the Fonds der
Chemischen Industrie is gratefully acknowledged.
To demonstrate the potential use of our novel bisphenol
compounds as ligands in asymmetric catalysis, we applied
them in the copper-mediated Michael-type addition of
diethylzinc to 2-cyclohexenone (Scheme 4).9
Supporting Information Available: Representative pro-
cedures, spectral data of new compounds, and crystal-
lographic data (CIF) for compound 10f. This material is
(8) In our work using chiral N-arenesulfonyl-1,3-oxazolidines as chiral
auxiliaries, nor-pseudoephedrine derivatives often provided superior selec-
tivity; see, for example: Bru¨ggemann, M.; Fro¨hlich, R.; Wibbeling, B.;
Holst, C.; Hoppe, D. Tetrahedron 2002, 58, 321.
OL060528U
(9) For reviews on copper-catalyzed Michael additions, see: (a) Alexakis,
A.; Benhaim, C. Eur. J. Org. Chem. 2002, 3221. (b) Krause, N.; Hoffmann-
Ro¨der, A. Synthesis 2001, 171.
(10) The absolute configuration of 12 was determined by derivatiza-
tion: Alexakis, A.; Frutos, J. C.; Mangeney, P. Tetrahedron: Asymmetry
1993, 4, 2431.
2458
Org. Lett., Vol. 8, No. 12, 2006