can be treated independently and summed to give the
predicted rotation for the complete molecule. Brewster’s
empirical model has been one of the most successful.6d-i In
the Brewster model, the contribution of each carbon-carbon
bond is calculated and the components are summed. Since
each conformation contributes to the observed rotation,
application of Brewster’s model to conformationally flexible
molecules requires the contribution of each conformer in a
Boltzmann-weighted average to be summed, which can be
tedious. A general analysis of additivity schemes for more
complex frameworks has been developed by Ruch.9 This
analysis has been applied to allenes10 and stereogenic
carbons11 with some success. Ruch’s analysis underlies our
own work with syn-2,2-dimethyl-1,3-dioxanes.
cyanohydrin acetonide 6 in 71% yield in a standard three-
step sequence.16 The lithium anion of 6 was alkylated with
several electrophiles in good yield (steps d-g, Scheme 1),
Scheme 1. Preparation of syn-2,2-Dimethyl-1,3-dioxanes 8
and 10a
The 2,2-dimethyl-1,3-dioxane (e.g. acetonide 4) structure
is one of the most widely used protecting groups for 1,3-
diols and has been used to assign relative configurations to
diols as the syn and anti isomers are easily distinguished by
13C NMR analysis.12 The conformation of the syn- and anti-
2,6-dialkyl-1,3-dioxanes are well defined, with the former
adopting a chair conformation and the latter adopting a 2,5-
twist-boat conformation.12,13 Furthermore, the two substitu-
ents on a syn acetonide ring are remote in space and unlikely
to interact with each other. Thus we postulate that the specific
conformation of R1 will not depend on the identity or
conformation of R2. The syn isomer 4 is a “category a”
molecular framework as defined by Ruch, in that it has only
two substituents (ligands) and these ligands are exchanged
by a mirror plane.9 Ruch has shown that the optical rotation
of category a skeletons can be predicted with “chirality
functions” in which each ligand has an empirically derived
λ-parameter.10,11 The range of ligands is restricted by the
requirement that a molecule with exclusively identical ligands
must have the symmetry of the skeleton. In the case of
skeleton 4, this restriction excludes the use of R1 or R2 that
contain stereogenic centers. It is obvious on inspection that
4 will be achiral when R1 is identical to R2 and when R1
and R2 do not contain stereogenic centers. We set out to
develop a simple, empirical additivity scheme to predict the
molar rotations of syn acetonide structures.
a The R substituent is defined in Table 1. (a) TMS‚NEt2; (b)
DIBALH, Et2O, -78 °C; (c) (i) TMSCN, 18-crown-6/KCN, (ii)
acetone, 2,2-DMP, CSA; (71%); (d) 2-(tert-butyldimethylsilyloxy)-
ethyl bromide, LDA, THF, -40 °C (73%); (e) LHMDS then allyl
chloride, THF, -78 to -20 °C; (48%); (f) n-BuBr, LDA, THF,
-40 °C (50%); (g) 1-bromo-3-butene, LDA, THF, -40 °C (68%);
(h) Li, NH3, -78 °C, THF or Et2O; (i) 2-phenethylbromide, LDA,
THF, -40 °C (60%); (j) see Supporting Information.
and the cyano group was removed in a stereoselective
reduction.15 Nitrile 9 was alkylated with 2-phenethylbromide
and then treated under dissolving metal conditions to generate
acetonide (6S)-8l in 42% yield over two steps. The 2,2-
dimethyl-1,3-dioxanes (6R)-8a and (6S)-8l were converted
to compounds (6R)-8e-k and (6S)-8m-s by standard
methods.17 Acetonide (6R)-8t (R ) H) was prepared by
DIBALH reduction of 5 and acetonide formation.
MW[R]
100
[Φ] )
(1)
Synthesis of the desired substrates began with â-hydroxy
ester 5 (98% ee)14 and with cyanohydrin 9, which was
derived from L-malic acid.15 Ester 5 was converted to
Optical rotation data was collected for each of the
acetonides, and the molar rotation was calculated according
to eq 1, where MW is molecular weight.18 Molar rotation is
(9) Ruch, E. Acc. Chem. Res. 1972, 5, 49-56.
(10) Ruch, E.; Runge, W.; Kresze, G. Angew. Chem., Int. Ed. Engl. 1973,
12, 20-25.
(15) Rychnovsky, S. D.; Zeller, S.; Skalitzky, D. J.; Griesgraber, G. J.
Org. Chem. 1990, 55, 5550-5551.
(16) Rychnovsky, S. D.; Griesgraber, G. J. Org. Chem. 1992, 57, 1559-
1563.
(11) Richter, W. J.; Richter, B.; Ruch, E. Angew. Chem., Int. Ed. Engl.
1973, 12, 30-36.
(12) Rychnovsky, S. D.; Rogers, B. N.; Richardson, T. I. Acc. Chem.
Res. 1998, 31, 9-17.
(17) Details are provided in Supporting Information.
(18) Rotations were measured at 23-25 °C in CHCl3 (c 0.80-1.40).
Uncertainties in molar rotation derive from errors in mass, volume and
rotation measurements. Estimated uncertainties for the molar rotations in
Table 1 are approximately 10%.
(13) Eliel, E. L.; Knoeber, M. C., Sr. J. Am. Chem. Soc. 1968, 90, 3444-
3458.
(14) Rychnovsky, S. D.; Sinz, C. J. Tetrahedron Lett. 1998, 39, 6811-
6814.
3076
Org. Lett., Vol. 4, No. 18, 2002