COMMUNICATION
and 5 in diverse hydrogenation reactions.[16] In an effort to
broaden the scope of these catalysts, unsaturated phospho-
nates 1 were chosen as substrates. Studies were initiated by
the development of a convenient procedure for the synthesis
of a set of olefins 1. These derivatives can be easily prepared
from readily available b-ketophosphonates by simple treat-
ment with sodium hydride and an acylating agent (either
BzCl or Bz2O see the Supporting Information). Notably, this
reaction yields olefins 1 as the Z isomers exclusively. This
feature is of practical interest, as separation of geometric
isomers is avoided, and confers a challenging aspect to the
hydrogenation of these olefins, since structurally related Z-
b-N-acylaminoacrylates usually give slower and less enantio-
selective reactions than their E counterparts.[17] In addition,
steric effects, resulting from the size of the phosphonate
group, can further reduce the reactivity of these alkenes.[18]
enantioselectivity (Table 1, entry 2). Interestingly, catalysts
based on ligands 4a and 4b had low activity in the hydroge-
nation of related a-acyloxy
Following the pleasing result obtained in the reduction of
1a, the scope of catalyst precursor [Rh(cod)(4b)]BF4 was
examined. A variety of b-acyloxyphosphonates were hydro-
phosphonates.[16b]
ACHTREUNG
A
ACHTREUNG
AHCTREUNG
genated with very high enantioselectivities using this cata-
lyst. Thus, b-alkyl product 2b was obtained with high che-
moselectivity and 99% ee (Table 2, entry 2). Conversely,
Table 2. Hydrogenation of 1 with [Rh
(4b)]BF4.[a]
ACHRTUNEG(cod)ACHTREUGN
Entry Substrate Product distribution Yield 2 ee 2 Configuration
[%]
[%][b] [%][c]
2
6
1
2
3
4
5
6
7
8
1a
1b
1c
1d
1e
1 f
1g
1h
98
98
68
98
90
7327
96
2
2
32
2
90
82
55
9399
85
99
99
99
R
R
R
R
R
R
R
R
A set of rhodium precatalysts of formulation [RhACTHERU(NG cod)AHCTRE(UGN P–
OP)]BF4 (cod=cycloocta-1,5-diene, P–OP=4, 5; Scheme 2)
10
99
99
9573
4
386
87
97
99
[a] Conditions: 4 atm H2, 258C, CH2Cl2, S/C=100, 24 h. Full conversion,
determined by 1H NMR spectroscopy, was detected in all reactions.
[b] Yield of isolated product. [c] Enantiomeric excess determined by
chiral HPLC. Absolute configuration was determined in 2g by X-ray
crystallography. This configuration has tentatively been assigned to the
rest of compounds of the series, assuming the same stereochemical
course of the reaction in all substrates.
nBu-substituted 1c offered a lower selectivity for the de-
sired 2c, although this compound was also obtained with
very high enantioselectivity. A range of b-aryl derivatives
were also examined and, with exception of the p-anisyl de-
rivative 2 f , which showed a moderate chemoselectivity
(73%, Table 2, entry 6), gave high selectivities for the chiral
derivatives 2d–h. Moreover, for all aromatic substrates in-
vestigated, the reaction proceeds with very high enantiose-
lectivity, from 95% ee for 2 f to greater than 99% ee for the
other substrates.
We next examined the configuration of compounds 2a–h.
Firstly, X-ray crystallography, performed on derivative 2g,
established an absolute R configuration for this com-
pound.[19] Moreover, the same configuration has been as-
signed to b-alkyl 2a by comparison of the specific rotation
of the corresponding alcohol 3a (see below) with the value
reported in the literature.[8c] It is interesting to compare the
configuration of products resulting from hydrogenation of
a- and b-acyloxyvinylphosphonates.[20] According to a previ-
ous study, catalysts based on P–OP ligands with an S phos-
phite fragment produce S products from a-acyloxy deriva-
tives.[16b] Therefore, product configuration indicates that hy-
drogen addition occurs on opposed faces of the unsaturated
phosphonates (Scheme 4).
Scheme 3. Hydrogenation of 1. cod=cycloocta-1,5-diene, P–OP=phos-
phane–phosphite ligand 4 or 5.
(P–OP)]BF4.[a]
Table 1. Hydrogenation of 1a with [RhACTHERU(NG cod)ACHTREUGN
Entry P–OP Conv[b] Product distribution [%] ee 2a Configuration
[%][c]
2a[d]
6a[d]
1
2
3
4
(S)-4a 46
(S)-4b 100
(S)-5a 41
(R)-5b 100
36
98
33
32
10
2
8
33
99
31
59
R
R
R
S
68
[a] Conditions: 4 atm H2, 258C, CH2Cl2, S/C (substrate/catalyst ratio)=
100, 24 h. [b] Conversion was determined by 1H NMR spectroscopy.
[c] Enantiomeric excess determined by chiral HPLC. Absolute configura-
tion assigned by comparison of optical rotation of alcohol 3a (resulting
from debenzylation of 2a), with literature value.[8c] [d] Based on starting
material 1a.
has been examined in the hydrogenation of the representa-
tive substrate 1a (Scheme 3, Table 1). Notably, this kind of
hydrogenation produces, in addition to the desired saturated
phosphonate 2a, a second product 6a, generated by elimina-
tion of the OBz group. Interestingly, both the amount of this
by-product and the enantioselectivity of the reaction can be
optimized by an appropriate choice of P–OP ligand. Thus,
among the catalysts investigated, that derived from phos-
phane–phosphite 4b produced excellent results, giving full
conversion and affording 2a with excellent chemo- and
We then explored if compounds 2 can be deprotected to
the corresponding alcohols 3. A set of representative com-
pounds were converted into the desired alcohols without
loss of enantiomeric purity by a simple treatment with
Na2CO3 in methanol (Scheme 5).
Chem. Eur. J. 2008, 14, 9856 – 9859
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9857