Synthesis and application of modular phosphine-phosphoramidite ligands in asymmetric hydroformylation: Structure-selectivity relationship

Article abstract of DOI:10.1002/chem.200902238

A series of hybrid phos- phine-phosphoramidite ligands has been designed and synthesized in moderate yields from chiral BINOL (1, 1′- bi-2-naphthol) or NOBIN (2-amino-2′- hydroxy-1, 1′-binaphthyl). They have achieved highly regio- and enantiose- lectivities in Rh-catalyzed asymmetric hydroformylations of styrene derivatives (branched/linear ratio up to 56.6, ee up to 99%), vinyl acetate derivatives (up to 98% ee), and allyl cyanide (up to 96 % ee). Systematic variation of ligand structure showed that the steric factor on the phsophoramidite moiety determined the performance of the ligand. With the increased hindrance, the branched/linear ratio rose, while the ee value dropped in the hydrofor- mylation of styrene. However, the N- substituents did not influence the selec- tivities much.

Full text of DOI:10.1002/chem.200902238

FULL PAPER
DOI: 10.1002/chem.200902238
Synthesis and Application of Modular Phosphine–Phosphoramidite Ligands
in Asymmetric Hydroformylation: Structure–Selectivity Relationship
Xiaowei Zhang,[a, b] Bonan Cao,[a] Yongjun Yan,[b] Shichao Yu,[a] Baoming Ji,[b] and
Xumu Zhang*[a]
Abstract: A series of hybrid phos-
phine–phosphoramidite ligands has
been designed and synthesized in mod-
erate yields from chiral BINOL (1,1’-
bi-2-naphthol) or NOBIN (2-amino-2’-
hydroxy-1,1’-binaphthyl). They have
achieved highly regio- and enantiose-
lectivities in Rh-catalyzed asymmetric
hydroformylations of styrene deriva-
tives (branched/linear ratio up to 56.6,
ee up to 99%), vinyl acetate derivatives
(up to 98% ee), and allyl cyanide (up
to 96% ee). Systematic variation of
ligand structure showed that the steric
factor on the phsophoramidite moiety
determined the performance of the
ligand. With the increased hindrance,
the branched/linear ratio rose, while
the ee value dropped in the hydrofor-
mylation of styrene. However, the N-
substituents did not influence the selec-
tivities much.
Keywords: asymmetric catalysis
·
enantioselectivity hydroformyla-
·
tion · phosphine–phosphoramidite
ligands · rhodium
Introduction
enantioselective hydroformylations are usually carried out
at low temperature with relatively low reaction rate and
conversion. Second, it is difficult to achieve good reactivity,
regio-, and enantioselecvtivities under the same conditions.
Finally, racemization of the aldehyde products, particularly
those from styrene derivatives, is observed under hydrofor-
mylation reaction conditions.[1b]
Hydroformylation is one of the most important reactions
still widely used in industry that provides aldehydes directly
from alkenes and syngas (CO/H2) in one single step. Mil-
lions of tons of oxo products produced worldwide per year
lead to it being regarded as the largest industrially homoge-
neous catalytic process.[1] In particularly, asymmetric hydro-
formylation (AHF) has attracted much attention as an
atom-economic method to convert olefins into enantiomeri-
cally pure aldehydes, which can be used as important precur-
sors for synthesizing a variety of biologically active products
and fine chemicals.[1–3] Although AHF offers promising ap-
plication, it is still seldom utilized by industry mainly be-
cause of several well-known challenging issues. First, highly
Although new ligands that are able to provide chiral alde-
hydes at reasonably high temperatures without sacrificing
their selectivities are highly desirable, only a few successful
examples were documented in the past two decades
(Scheme 1). Bidentate phosphite ligands proved to be active
in AHF reactions, such as (2R, 4R)-chiraphite (1)[4] and (S,
S)-kelliphite (2).[5] The former ligand shows high enantiose-
lectivity (nearly 90% ee) for the hydroformylation of sty-
rene and the latter one is effective for allyl cyanide [75%
ee, b/l (branched/linear ratio)=16] and vinyl acetate (88%
ee, b/l=56) at low temperature. Another class of successful
ligands is the C2-symmetric bis-phospholane; for example,
(S, S)-esphos (3),[6] reported by Wills and co-workers, pro-
vides a high selectivity for vinyl acetate (90% ee, b/l=16),
but nearly no enantioselectivity for styrene. Landis, Klosin,
and co-workers reported the diazaphospholane ligand 4 and
its analogues,[7] which were applied in the AHF of styrene,
vinyl acetate, and allyl cyanide with high enantioselectivities
(82, 96, 87% ee, respectively) and regioselectivities (b/l=7,
4, 37, respectively) even at elevated temperature. Recently,
a class of C2-symmetric bidentate phosphonite ligands, re-
[a] X. Zhang, B. Cao, Dr. S. Yu, Prof. Dr. X. Zhang
Department of Chemistry and Chemical Biology and
Department of Pharmaceutical Chemistry
Rutgers, The State University of New Jersey
Piscataway, New Jersey 08854 (USA)
Fax : (+1)732-445-6312
E-mail: xumu@rci.rutgers.edu
[b] X. Zhang, Dr. Y. Yan, Dr. B. Ji
Department of Chemistry
The Pennsylvania State University
University Park, Pennsylvania 16802 (USA)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200902238.
Chem. Eur. J. 2010, 16, 871 – 877
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
871

of nitrogen (3.04) is less than
that of oxygen (3.44). Sterically,
the N-substituent on the phos-
phoramidite group can make
the active catalytic complex
show a deeper and more closed
chiral pocket than that of bina-
phos, based on the models from
CAChe MM2 calculation.[11]
Takaya and co-workers have
concluded that the configura-
tion matched (R,S and S,R)-bi-
naphos showed better selectivi-
ty than the mismatched (R,R
and S,S)-binaphos. As ana-
logues of binaphos, the new
hybrid phosphine–phosphora-
midite ligands were synthesized
with two enantiomerically op-
posite binaphthyl groups. First,
we synthesized ligands 9a, 9b
(named as YanPhos),[10] and 9c
with a methyl, ethyl, and benzyl
group, respectively, attached to
the nitrogen atom to investigate
the influence of the N-substitu-
ents (Scheme 2). Afterward,
Scheme 1. Representative examples of chiral ligands for AHF reactions and the structure of the phosphine–
phosphoramidite ligand.
ported by Dingꢁs group, also displayed high selectivity in
Rh-catalyzed AHF of the above olefins.[8] Among all the li-
gands reported hitherto, hybrid ligands bearing a different
phosphorus structure have been regarded as the best in
AHF. The major breakthrough in this area was made in
1993, when Takaya, Nozaki and co-workers reported (R,S)-
binaphos (6),[9] which offered generally high enantioselectiv-
ities in the AHF of a variety of prochiral olefins (up to 94%
ee, b/l=7.3 for styrene). However, with binaphos as the
ligand, chiral aldehyde products can undergo racemization
under certain conditions.[9b] It is still highly desirable to de-
velop new ligands for highly enantioselective hydroformyla-
tion without racemization. Herein, we report the synthesis
of a new family of hybrid phosphine–phosphoramidite li-
gands (as shown in Scheme 1) as well as their applications in
the Rh-catalyzed AHF of styrene, vinyl acetate, allyl cya-
nide, and their derivatives with good to excellent regio- and
enantioselectivities (up to 99% ee). The modular character
allows systematic variation on the ligand structure, which
was utilized to investigate the relationship between the
ligand structure and their control of enantioselectivity. We
envision that understanding of the structure–selectivity rela-
tionship can provide useful guidance for ligand design and
help for optimizing the current theoretical models for AHF.
with an ethyl group as the N-substituent, the binaphthyl
group of the phosphoramidite part was decorated with
methyl groups at the 3,3’-position (ligand 9d) or with the
sterically more bulkyl di-tert-butyl-substituted biphenyl
groups to determine the steric effect of the phosphoramidite
moiety (ligand 9e and 9 f, Scheme 2). Those bulkyl substitu-
ents spatially adjacent to the P atom were expected to make
the chiral pocket more closed and further define the asym-
metric environment around the catalytic center.
Ligands (S,R)-9a, (S,R)-9b, and (S,R)-9c were synthesized
from commercially available (S)-BINOL (1,1’-bi-2-naph-
thol). Following the literature procedure,[12] the phosphine-
amines (7a–c) were easily prepared by a sequence of well-
established steps. Followed by deprotonation with nBuLi
and quenching with the phosphorochloridite (8a, prepared
from (R)-BINOL), the desired ligands (S,R)-9a, (S,R)-9b,
and (S,R)-9c were obtained in moderate yield (33–42%).
An alternative concise pathway to afford N-ethyl phos-
phine-amine (7b) starts from NOBIN (2-amino-2’-hydroxy-
1,1’-binaphthyl) according to a known procedure.[13] As
shown in Scheme 2, the following deprotonation and cou-
pling with the corresponding phosphorochloridites afforded
the desired ligands (R,S)-9b, (R,S)-9d, (R,S)-9e, and (R)-9 f.
It is worthwhile to note that all these ligands are air-stable
solids.
Before applying these novel phosphine–phosphoramidite
ligands in Rh-catalyzed asymmetric hydroformylation, the
optimized reaction conditions were obtained by using (R,S)-
9b as a representative ligand and styrene (10) as a standard
substrate. The AHF reactions were carried out with
Results and Discussion
Compared to binaphos, the phosphine–phosphoramidite
ligand is more electron-donating since the electronegativity
872
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 871 – 877

Asymmetric Hydroformylation
FULL PAPER
entries 3, 5–8). Increasing the
reaction temperature led to
higher conversion but lower ee
values (Table 1, entries 1, 9, and
10). As high as 99% ee was ob-
tained when the reaction was
run at 408C with 25% conver-
sion, while the ee value drop-
ped to 81% at 808C. The best
temperature for this catalyst
system is 60 8C where full con-
version and 98% ee were ach-
ieved. The syngas pressure did
not influence the enantioselec-
tivity but impacted the reactivi-
ty dramatically (Table 1, en-
tries 3, 11, and 12). The higher
CO/H2 pressure resulted in
lower conversion, which is
mainly because the equilibrium
of CO coordination to the Rh
center is shifted more towards
Scheme 2. Synthesis of phosphine–phosphoramidite ligands.
carbonyl-bound Rh species at
high pressure. A longer reac-
0.1 mol% of catalyst loading and 1:1 CO/H2 gas. The cata-
lyst was prepared in situ by mixing [RhACTHNUTRGNEN(UG acac)(CO)2] with
tion time led to only a slight decrease in the enantioselectiv-
ity (Table 1, entries 3, 13, and 14). The racemization of the
Rh/9b-catalyzed AHF was markedly lower than that of bi-
naphos.[9b] All of the above hydroformylation reactions pro-
vided high chemoselectivities (no hydrogenation product
was detected) and good regioselectivities, whereas the enan-
tioselectivities strongly depended on reaction conditions.
Entry 3 in Table 1 represents the optimized reaction condi-
tions for phosphine–phosphoramidite ligands.
Using the optimized reaction conditions, we systematically
investigated the structure–selec-
ligand 9b. The ligand/Rh ratio significantly influenced the
hydroformylation reaction. As shown in Table 1, entries 1–4,
increasing the ligand/Rh ratio from 1:1 to 4:1 improved both
the regioselectivity (b/l from 3.0 to 7.3) and enantioselectivi-
ty (from 21 to 98% ee), but further increasing the ratio to
6:1 did not result in any further improvement. Screening the
reaction solvent showed that nonpolar solvents, such as ben-
zene and toluene, offered high enantioselectivities (Table 1,
tivity relationship of phos-
Table 1. Optimization of Rh-catalyzed asymmetric hydroformylation of styrene with (R,S)-9b.[a]
phine–phosphoramidite ligands.
Three of the most commonly
used standard substrates: sty-
rene (10), vinyl acetate (13),
and allyl cyanide (16) were uti-
lized to examine the regio- and
enantioselectivities of this series
of ligands 9a to 9 f in Rh-cata-
lyzed AHF (Table 2). The
impact of the N-substituent was
examined by comparing the
performance of ligands 9a, 9b,
and 9c (Table 2, entries 1–3).
Increasing the steric bulk of the
N-substituent from a methyl to
a benzyl group slightly elevated
the regioselectivity and de-
creased the enantioselectivity.
An N-ethyl-substituted ligand,
(S,R)-9b, provided hitherto the
best enantioselectivities for Rh-
Entry
9b/Rh
Solvent
T
[8C]
CO/H2
[atm]
Time
[h]
Conv.
[%][b]
b/l[b]
ee
G
[%][c]
1
2
3
4
5
6
7
8
1:1
2:1
4:1
6:1
4:1
4:1
4:1
4:1
4:1
4:1
4:1
4:1
4:1
4:1
PhH
PhH
PhH
PhH
PhMe
CH2Cl2
THF
EtOAc
PhH
PhH
PhH
PhH
PhH
60
60
60
60
60
60
60
60
40
80
60
60
60
60
10/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
10/10
20/20
30/30
10/10
10/10
24
24
24
24
24
24
24
24
24
24
24
24
12
36
99
99
99
99
99
93
95
98
25
99
91
83
87
99
3.0
5.7
7.3
7.3
7.2
10.1
7.3
8.1
10.1
5.7
8.0
7.3
8.1
7.3
21(R)
54(R)
98(R)
98(R)
98(R)
97(R)
78(R)
84(R)
99(R)
81(R)
98(R)
98(R)
99(R)
97(R)
9
10
11
12
13
14
PhH
[a] All reactions were carried out with substrate/Rh=1000. [b] Conversions and branched/linear ratio (b/l)
were determined on the basis of H NMR spectroscopy. [c] Determined by converting the aldehyde to the cor-
responding alcohol with NaBH4 followed by GC analysis (Supelcoꢁs Beta Dex 225). The absolute configura-
tion (R) was assigned by comparing the sign of the optical rotation of the resulting alcohol with (R)-2-phenyl-
propan-1-ol. The entries in boldface highlight key results.
1
Chem. Eur. J. 2010, 16, 871 – 877
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
873

X. Zhang et al.
Table 2. Rh-catalyzed AHF of styrene, vinyl acetate, and allyl cyanide with phosphine–phosphoramidite ligands.[a]
Entry
Ligand
10
b/l[b]
13
b/l[b]
16
b/l[b]
Conv. [%][b]
ee [%][b]
Conv. [%][b]
ee [%][b]
Conv. [%][b]
ee [%][c]
1
2
3
4
5
6
7
N
99(24)[d]
99(22)[d]
99(22)[d]
99(22)[d]
96(13)[d]
99(27)[d]
99(29)[d]
6.6
7.2
7.1
7.2
8.0
27.6
56.6
97(S)
98(S)
95(S)
98(R)
91(R)
66(R)
75(R)
76
76
77
76
69
88
95
12.1
14.0
16.5
13.5
65.2
22.9
7.5
95(R)
96(R)
95(R)
96(S)
84(S)
65(S)
66(S)
97
99
99
99
89
95
98
3.8
96(S)
96(S)
93(S)
96(R)
90(R)
65(R)
69(R)
4.0
4.1
4.0
3.0
2.5
2.9
[a] All reactions were carried out at 608C in toluene with L:Rh=4:1, substrate/Rh=1000, 20 bar 1:1 CO/H2, and a reaction time of 24 h for stryene,
vinyl acetate and a reaction time of 18 h for allyl cyanide. The entries in boldface highlight key results. [b] Conversions, branched/linear ratio and ee
values were determined by GC analysis (Supelcoꢁs Beta Dex 225). The absolute configuration for the products 11, 14, and 17 were assigned by comparing
the sign of the optical rotations with those in the literature.[14] [c] Determined by converting the aldehyde to the corresponding acid and then reacting
with aniline to afford corresponding amide followed by HPLC analysis. [d] The number in parentheses represents the conversion of a reaction carried
out for 3 h.
catalyzed AHF with 98% ee for styrene and 96% ee for
vinyl acetate. Notably, with (S,R)-9b as ligand, the hydrofor-
mylation of allyl cyanide afforded as high as 96% ee, which
is higher than the previously best enantioselectivity (94% ee
at 49% conversion) achieved by binapine,[15] a ligand devel-
oped in our group for hydrogenation. Overall, the variation
of N-substituent does not exert much influence on the selec-
tivities of phosphine–phosphoramidite ligands.
these conditions to our knowledge. The overall trend is such
that increasing the steric bulkiness of the phosphoramidite
moiety diminishes the enantioselectivity, but benefits the re-
gioselectivity for the AHF of styrene.
To investigate the effect of ligand structure on the catalyt-
ic activities, we applied ligands 9a–f in the AHF of styrene
with a reaction time of 3 h (as shown in parentheses in
Table 2). It was found that the N-substituents did not influ-
ence the ligand activity very much. The conversions of sty-
rene with ligands 9a–c were 24, 22, and 22%, respectively.
The activity of 9d was much slower than that of the other li-
gands, which is possibly due to its poor solubility in toluene.
Ligands 9e and 9 f resulted in higher activity with 27% and
29% conversion, respectively.
As a counter enantiomer, (R,S)-9b showed regio- and
enantioselectivities for all the three substrates almost equal
to those achieved with (S,R)-9b as ligand, except for the op-
posite absolute configuration of the products (Table 2, en-
tries 2 and 4). By fixing an N-ethyl group to the phosphine-
amine moiety, we investigated the steric effect of the phos-
phite part on the regio- and enantioselectivities of phos-
phine–phosphoramidite ligands. (R,S)-9d, with two methyl
groups at the 3,3’-position of binaphthyl part, did not display
a higher enantioselectivity than the corresponding (R,S)-9b
as expected (Table 2, entries 4 and 5). Unlike the case of bi-
naphos, where the two additional methyl groups led to a
much higher reactivity and enantioselectivity,[9c] (R,S)-9d re-
sulted in decreased reactivity and ee values, albeit with
higher regioselectivities for styrene and vinyl acetate (b/l=
8.0 and 65.2, respectively). Although possessing the even
more sterically bulky phosphoramidite fragment, (R,S)-9e
proved to be less effective in asymmetric induction than
(R,S)-9d with only moderate ee values for the three sub-
strates (66, 65, and 65%, respectively, Table 2, entry 6). The
improvement was the higher regioselectivity for styrene (b/
l=27.6). To investigate the role of the chirality of the phos-
phoramidite unit, (R)-9 f was prepared and used in AHF. In
contrast to (R,S)-9e, the loss of one chiral center did not
have a negative effect on the enantioselectivity, but led to
slightly higher ee values (75, 66, and 69%, respectively,
Table 2, entry 7). Notably, (R)-9 f afforded a regioselectivity
up to b/l=56.6, which is the best result obtained under
To explain our above observation, we took advantage of
the models from CAChe MM2 calculations with ligand 9b
as a reference. Based on the mechanistic studies by Takaya
and co-workers, it is believed that the active catalyst
[RhH(CO)ACTHNUTRGNEUG{N (R,S)-9b}] coordinates with olefin to form a
trigonal-bypyramidal complex, in which the phosphine occu-
pies an equatorial position and the phosphate is located at
an axial position that is trans to the hydrido ligand.[9b] Herr-
mann and co-workers proposed a semiquantitative theoreti-
cal model that can successfully elucidate the origin of ste-
reodifferentiation in Rh-binaphos-catalyzed AHF.[1d,16] Be-
cause of the structural and electronegative similarity to bi-
naphos (as shown in Scheme 1), ligand 9b is proposed to
proceed by a similar catalytic cycle to that of binaphos. On
the basis of Herrmannꢁs model and CAChe MM2 calcula-
tions, we extrapolate that there are two possible transition
states (TS I and TS II, as shown in Figure 1) in the process-
ing of styrene (labeled as green), which represents a typical
À
olefin substrate herein, insertion into the Rh H bond of
[RhH(CO)
A
Rh center afford the same configuration (R)-2-phenylpropa-
nal. From the stick models for TS I and TS II based on the
874
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 871 – 877

Asymmetric Hydroformylation
FULL PAPER
periment, for styrene, a trend was observed that the ligand
with more bulky substituent provides better regioselectivity.
Encouraged by the successful application of ligand 9b in
the Rh-catalyzed AHF of styrene and vinyl acetate, a series
of their derivatives was hydroformylated by utilizing the
Rh-(R,S)-9b catalyst under the optimized reaction condi-
tions. All the styrene derivatives displayed good regioselec-
tivities and excellent enantioselectivities (up to 99% ee,
Table 3, entries 1–6). The halogen-substituted styrene deriv-
Table 3. Rh-catalyzed AHF of styrene and vinyl acetate derivatives with
(R,S)-9b.[a]
Entry
R
Conv. [%][b]
b/l[b]
ee [%][c]
1
2
3
4
5
6
7
8
p-Me-Ph
p-F-Ph
o-F-Ph
98
99
99
99
97
98
67
53
56
69
40
69
7
7
10
7
6
99(R)
98(R)
98(R)
98(R)
98(R)
98(R)
93(S)
94(S)
94(S)
96(S)
98(S)
93(S)
p-Cl-Ph
Figure 1. a) Models for two possible transition states, TS I and TS II, for
p-MeO-Ph
p-iBu-Ph
CH3CH2COO
À
styrene insertion into the Rh H bond of [RhH(CO)
G
8
models for the corresponding transition states TS I and TS II based on
CAChe MM2 calculations (black rectangle represents naphthyl fragments
from the backbone, dashed rectangles represent naphthyl from the phos-
phoramidite moiety).
24
16
16
16
16
24
CH
CH
CH
3A
3A(CH2)6COO
3A(CH2)8COO
9
CHTUNGTRENNUNG
10
11
12
CHTUNGTRENNUNG
tBuCOO
PhCOO
[a] All reactions were carried out at 608C in benzene with L:Rh=4:1,
substrate/catalyst=1000, 20 bar 1:1 CO/H2, and 24 h. [b] Conversions,
branched/linear ratio were determines based on 1H NMR spectroscopy.
[c] Determined by GC analysis. The absolute configuration were assigned
by comparing the sign of the optical rotations with those in the litera-
ture.[14]
results of CAChe MM2 calculations, it is concluded that the
enantioselectivity arises from the steric repulsion between
the phenyl group of styrene and one of the naphthyl frag-
ments of (R,S)-9b (marked with black rectangle in
Figure 1).[17] In TS I, it is the naphthyl from naphthylamine,
while it is the one from phosphoramidite in TS II. This
model can rationalize our experiment results well. As shown
in Figure 1, the ethyl group on the N atom stretches back
atives in particular achieved high ee values at high conver-
sion (Table 3, entries 2-4). Notably, 98% ee was achieved for
the para-isobutyl styrene (Table 3, entry 6). Its aldehyde
product could be oxidized into ibuprofen, one of the most
widely used nonsteroidal anti-inflammatory drugs. As shown
in Table 3, entries 7–12, a series of vinyl acetate derivatives
was hydroformylated with this catalyst system. Remarkably,
the AHF of a 2,2-dimethylpropionic acid vinyl ester, bearing
a bulky alkyl residue on the carboxyl group, demonstrates
the highest enantioselectivity (98% ee, Table 2, entry 6). In
general, all of the styrene and vinyl acetate derivatives ach-
ieved high regio- and enantioselectivities in the Rh-9b-cata-
lyzed AHF, which makes this methodology potentially inter-
esting to industrial application.
À
away from the plane of the Rh H bond and does not inter-
act with the styrene at all. Thus changing the N-substituent
did not exert a significant influence on the enantioselectivity
of the phosphine–phosphoramidite ligands (Table 2, en-
tries 1–3). Increasing the steric bulkiness of the phsophor-
amidite fragment (marked as a dashed rectangle in Figure 1)
will increase the repulsion to the phenyl ring of styrene and
diminish the energy gap between the Re and Si binding of
styrene to the Rh center in TS I, although it is somewhat
beneficial to the differentiation of two enantiofaces in TS II.
The overall effect will damage the enantioselectivity of our
ligands. Hereby, (R,S)-9d provided lower enantioselectivity
for all the three substrates than (R,S)-9b. Similarly, (R,S)-
9e, with two bulky tert-butyl groups on the phosphoramidite
fragment, has even worse enantioselectivity (Table 2, en-
tries 4–6).
Conclusions
It is hard to predict the regioselectivities of the phos-
phine–phosphoramidite ligands with this theoretical model.
Generally, the regioselectivity of the hydroformylation reac-
tion is mainly determined by the functional group on the
olefin, for example, the phenyl group in styrene.[1a] The
higher chelation stability with the Rh center the functional
group has, the more branched aldehyde will form. In our ex-
In summary, a series of hybrid phosphine–phosphoramidite
ligands has been developed and systematically applied in
the Rh-catalyzed AHF of styrene, vinyl acetate, allyl cya-
nide, and their derivatives with high regio- and enantioselec-
tivities under mild conditions. With ligand 9b, 99% ee for
styrene derivatives, 98% ee for vinyl acetate derivatives, and
Chem. Eur. J. 2010, 16, 871 – 877
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
875

X. Zhang et al.
122.21, 41.05, 14.99 ppm; 31P NMR (162 MHz, CDCl3): d=141.00 (d, J=
59.1 Hz), À13.48 ppm (d, J=59.1 Hz); HRMS (ESI): m/z: calcd for
C54H40NO2P2 ([M+H+]): 796.2534; found: 796.2536.
96% ee for allyl cyanide were achieved, which represents
the best results to date. The relationship between the sub-
stituent and the enantioselectivity of the ligands was de-
duced, which was successfully rationalized by Herrmannꢁs
theoretical model by CAChe MM2 calculations. Further
studies aimed at a better understanding of the origin of the
selectivity of phosphine–phospharamidite ligands and their
application in other metal-catalyzed transformations are in
progress and will be reported in due course.
AHCTUNGTRENNUNG
(S,R)-9c: Yield=42%; white solid; [a]2D0 =+32.5 (c=0.3, CHCl3);
1H NMR (500 MHz, CDCl3): d=8.17 (d, J=8.5 Hz, 1H), 8.15 (d, J=
8.0 Hz, 1H), 7.97 (d, J=8.0 Hz, 1H), 7.92 (d, J=8.0 Hz, 1H), 7.74–7.60
(m, 5H), 7.43–7.01 (m, 23H), 6.87–6.83 (m, 2H), 6.79–6.75 (m, 2H),
6.45–6.42 (m, 1H), 6.24 (d, J=8.5 Hz, 1H), 5.93 (d, J=8.5 Hz, 1H), 3.82
(d, J=14.5 Hz, 1H), 3.21 (d, J=14.5 Hz, 1H); 13C NMR (125 MHz,
CDCl3): d=150.09, 150.05, 149.84, 142.04, 141.91, 138.75, 128.24, 127.92,
137.81, 135.61, 135.42, 133.97, 133.56, 133.43, 132.01, 131.81, 131.75,
131.61, 130.72, 130.50, 130.35, 129.87, 129.81, 128.83, 128.74, 128.56,
128.48, 128.40, 128.37, 128.33, 128.21, 128.11, 128.04, 127.89, 127.84,
127.79, 127.40, 127.23, 127.20, 127.04, 126.88, 126.71, 126.08, 126.03,
125.32, 124.88, 124.63, 122.78, 122.55, 122.10, 51.32 ppm; 31P NMR
(202 MHz, CDCl3): d=138.41 (d, J=78.6 Hz), À11.86 ppm (d, J=
78.6 Hz); HRMS (ESI): m/z: calcd for C59H42NO2P2 ([M+H+]): 858.2691;
found: 858.2692.
Experimental Section
General methods: All reactions and manipulations that were sensitive to
moisture or air were performed in a nitrogen-filled glovebox or using
standard Schlenk techniques, unless otherwise noted. Solvents were dried
with standard procedures and degassed with N2. Column chromatography
was performed by using 200–400 mesh silica gel supplied by Natland In-
ternational Corp. Thin-layer chromatography (TLC) was performed on
EM reagents 0.25 mm silica 60-F plates. 1H, 13C, and 31P NMR spectra
were recorded in CDCl3 or CD2Cl2 on a Bruker Avance 400 MHz spec-
trometer or a Varian Mercury 500 MHz FT-NMR spectrometer. Optical
rotation was obtained on a Perkin–Elmer 341 MC polarimeter. HRMS
were recorded on a Thermo LTQ Orbitrap hybrid mass spectrometer.
GC analysis was carried out on Hewlett–Packard 7890 gas chromato-
graph using chiral capillary columns. Compounds (S)-7a–(S)-7c,[12] (R)-
7b,[10] 8a,[9b] 8d,[9d] 8e,[18] and 8 f[18] were synthesized according to the cor-
responding literature procedures.
A
1H NMR (400 MHz, CD2Cl2): d=8.04–7.96 (m, 3H), 7.86–7.79 (m, 3H),
7.72 (d, J=8.0 Hz, 1H), 7.68 (d, J=8.4 Hz, 1H), 7.53–7.45 (m, 2H), 7.39–
7.28 (m, 5H), 7.26–7.05 (m, 11H), 6.96–6.93 (m, 2H), 6.86–6.81 (m, 2H),
6.61–6.57 (m, 1H), 6.33 (d, J=8.4 Hz, 1H), 2.94–2.83 (m, 1H), 2.58 (s,
3H), 2.50–2.45 (m, 1H), 1.66 (s, 3H), 0.70 ppm (t, J=7.2 Hz, 3H);
13C NMR (100 MHz, CD2Cl2): d=150.19, 150.12, 149.84, 142.23, 141.90,
138.93, 138.61, 138.57, 138.34, 137.72, 135.86, 135.34, 135.12, 134.41,
134.11, 133.94, 133.75, 133.74, 132.18, 132.01, 131.75, 131.68, 131.23,
131.07, 130.80, 130.06, 129.49, 129.09, 128.91, 128.58, 128.48, 128.29,
128.21, 127.93, 127.64, 127.29, 127.11, 127.02, 126.66, 125.80, 125.56,
125.32, 125.23, 125.07, 124.73, 121.81, 41.54, 17.66, 14.76 ppm; 31P NMR
(162 MHz, CD2Cl2): d=139.00 (d, J=61.6 Hz), À14.60 ppm (d, J=
61.6 Hz); HRMS (ESI): m/z: calcd for C56H44NO2P2 ([M+H+]): 824.2847;
found: 824.2843.
A typical procedure for the preparation of ligand 9: The synthesis of 9b
was reported in previous work.[10] Ligands 9a, 9c, and 9d were prepared
following a similar procedure. A typical procedure for (S,R)-9a is as fol-
lows: nBuLi (1.2 mmol, 0.48 mL of 2.5m hexane solution) was added
dropwise to a solution of (S)-7a (480 mg, 1.0 mmol) in anhydrous THF
(10 mL) at À78 8C under an N2 atmosphere. The reaction mixture turned
deep red and was stirred for 4 h at that temperature. Then (R)-8a
(454 mg, 1.3 mmol) in THF (6 mL) was added dropwise. After addition,
the reaction mixture was allowed to warm to room temperature and
stirred overnight. The volatiles were removed under vacuum. CH2Cl2
(5 mL) was added to the residue, and the mixture was filtered to remove
the inorganic salt. The filtrate was concentrated and subjected to flash
chromatography on silica gel (eluted with hexane/EtOAc/NEt3 100:10:1)
to afford pure ligand (S,R)-9a (257 mg) as a white solid in 33% yield.
[a]2D0 =À32.6 (c=0.3, CHCl3); 1H NMR (500 MHz, CDCl3): d=8.04–8.01
(m, 3H), 7.85 (t, J=8.5 Hz, 2H), 7.78 (d, J=8.5 Hz, 2H), 7.68 (d, J=
9.0 Hz, 1H), 7.61–7.57 (m, 2H), 7.38–6.93 (m, 21H), 6.59 (dd, J=8.5,
7.0 Hz, 1H), 6.49 (d, J=8.5 Hz, 1H), 6.31 (d, J=8.5 Hz, 1H), 2.45 ppm
(s, 3H); 13C NMR (125 MHz, CDCl3): d=150.45, 150.41, 149.61, 142.76,
142.49, 138.60, 138.49, 137.88, 137.79, 136.63, 135.25, 135.07, 134.08,
133.16, 133.03, 132.65, 131.61, 131.50, 130.73, 130.28, 129.94, 129.87,
128.84, 128.48, 128.31, 128.28, 128.20, 128.02, 127.75, 127.56, 127.42,
127.24, 127.16, 126.99, 126.53, 126.09, 125.69, 125.35, 124.86, 124.64,
124.12, 124.08, 122.28, 122.20, 35.67, 35.64 ppm; 31P NMR (202 MHz,
CDCl3): d=141.09 (d, J=45.2 Hz), À12.67 ppm (d, J=45.2 Hz); HRMS
(ESI): m/z: calcd for C53H38NO2P2 ([M+H+]): 782.2578; found: 782.2374.
AHCTUNGTRENNUNG
(R,S)-9e: Yield=38%; white solid; [a]2D0 =+91.7 (c=0.5, CHCl3);
1H NMR (400 MHz, CDCl3): d=7.98 (d, J=8.8 Hz, 1H), 7.93 (d, J=
8.4 Hz, 1H), 7.89 (d, J=8.4 Hz, 1H), 7.83 (d, J=8.0 Hz, 1H), 7.72 (d, J=
8.8 Hz, 1H), 7.63 (dd, J=8.4, 2.4 Hz, 1H), 7.50–7.46 (m, 1H), 7.24–7.12
(m, 6H), 7.06–6.87 (m, 9H), 6.65 (t, J=8.0 Hz, 1H), 3.58–3.50 (m, 1H),
2.80–2.73 (m, 1H), 2.23 (s, 3H), 2.20 (s, 3H), 1.78 (s, 3H), 1.76 (s, 3H),
1.34 (s, 9H), 0.81 (t, J=7.2 Hz, 3H), 0.74 ppm (s, 9H); 13C NMR
(100 MHz, CDCl3): d=148.95, 148.84, 147.74, 147.69, 142.72, 142.39,
138.85, 138.69, 138.46, 137.80, 137.77, 136.43, 135.12, 134.89, 134.57,
134.31, 134.05, 133.68, 133.29, 133.23, 133.13, 133.07, 131.80, 131.73,
131.25, 131.05, 130.76, 129.55, 129.38, 128.13, 128.06, 128.03, 127.72,
127.64, 127.46, 127.25, 127.20, 126.64, 126.55, 125.75, 125.15, 124.70, 40.05,
34.63, 34.07, 31.29, 31.26, 30.05, 20.19, 16.80, 16.31, 13.06 ppm; 31P NMR
(162 MHz, CDCl3): d=130.50 (d, J=97.2 Hz), À14.89 ppm (d, J=
97.2 Hz); HRMS (ESI): m/z: calcd for C58H60NO2P2 ([M+H+]):
864.4099; found: 864.4105.
(R)-9 f: Yield=52%; white solid; [a]2D0 =+40.7 (c=0.5, CHCl3);
1H NMR (400 MHz, CDCl3) d: 7.91 (d, J=7.2 Hz, 1H), 7.83 (d, J=
8.4 Hz, 2H), 7.73 (d, J=8.4 Hz, 2H), 7.50 (dd, J=7.2, 2.8 Hz, 1H), 7.39
(t, J=7.2 Hz, 1H) 7.24–7.00 (m, 11H), 6.97-6.93 (m, 4H), 6.85 (t, J=
7.6 Hz, 2H), 6.51–6.47 (m, 1H), 6.27 (d, J=8.4 Hz, 1H), 3.39–3.32 (m,
1H), 3.06–2.97 (m, 1H), 1.24 (s, 9H), 1.23 (s, 9H), 1.04–0.98 (m, 18H),
0.82 ppm (t, J=6.4 Hz, 3H); 13C NMR (100 MHz, CDCl3): d=147.50,
147.40, 146.82, 146.77, 144.32, 144.26, 143.68, 143.63, 138.82, 138.49,
137.93, 137.77, 136.99, 136.85, 133.97, 133.74, 133.48, 133.42, 133.12,
132.29, 132.22, 132.02, 131.97, 131.85, 131.80, 131.48, 130.24, 129.84,
129.73, 128.01, 127.79, 127.24, 127.05, 126.96, 126.87, 126.79, 126.68,
126.63, 126.39, 126.19, 126.14, 125.95, 125.77, 125.40, 125.04, 124.28,
124.14, 123.98, 122.77, 122.74, 37.77, 34.09, 33.67, 33.53, 33.45, 30.53,
30.49, 29.73, 29.34, 20.43 ppm; 31P NMR (162 MHz, CDCl3): d=136.29
(br), À14.93 ppm (d, J=62.4 Hz); HRMS (ESI): m/z: calcd for
C62H68NO2P2 ([M+H+]): 920.4725; found: 920.4749.
Ligands 9b–f were synthesized in moderate yields following the above
procedure. Their characterization data are summarized as follows.
A
1H NMR (400 MHz, CDCl3): d=8.07–7.98 (m, 3H), 7.90 (t, J=7.3 Hz,
2H), 7.78 (d, J=8.2 Hz, 2H), 7.64–7.57 (m, 4H), 7.38–6.99 (m, 16H),
6.96 (t, J=6.8 Hz, 2H), 6.85 (t, J=7.1 Hz, 2H), 6.55 (t, J=7.7 Hz, 1H),
6.38–6.29 (m, 2H), 2.75–2.67 (m, 1H), 2.37–2.29 (m, 1H), 0.65 ppm (t,
J=7.0 Hz, 3H); 13C NMR (100 MHz, CDCl3): d=150.29, 150.22, 149.94,
142.34, 141.95, 138.57, 138.36, 138.27, 138.20, 135.44, 135.14, 134.10,
133.57, 133.36, 131.68, 130.50, 129.88, 129.11, 128.66, 128.59, 128.55,
128.49, 128.46, 128.42, 128.30, 128.12, 127.56, 127.19, 127.12, 127.03,
126.66, 126.29, 126.17, 125.71, 125.53, 125.06, 124.76, 122.49, 122.24,
General procedure for asymmetric hydroformylation: In
filled with nitrogen, to a 2 mL vial equipped with a magnetic bar was
added ligand 9 (0.004 mmol), [Rh(acac)(CO)2] (0.001 mmol in 0.10 mL
a glovebox
AHCTUNGTRENNUNG
876
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 871 – 877

Asymmetric Hydroformylation
FULL PAPER
solvent), dodecane (50 mL, as a GC internal standard, if applicable), and
substrate (1.0 mmol), additional solvent was charged to bring the total
volume of the reaction mixture to 1.0 mL. After the mixture had been
stirred for 10 min, the vial was transferred into an autoclave and taken
out of the glovebox. Carbon monoxide (10 atm) and dihydrogen (10 atm)
were charged in sequence. The reaction mixture was stirred at 608C (oil
bath) for 24 h. The reaction mixture was cooled and the pressure was
carefully released in a well-ventilated hood. For analysis of the products
of styrene and vinyl acetate, the conversion, regioselectivity, and enantio-
meric excesses were determined following the reported method with a
Supelcoꢁs Beta Dex 225 column.[5b] For analysis of the products of allyl
cyanide, the conversion and regioselectivity were determined by GC with
a Supelcoꢁs Beta Dex 120 column.[8] The enantiomeric excesses of prod-
uct 17 were determined by oxidation with Jones reagent to afford the cor-
responding carboxylic acid, followed by reaction with aniline to give the
corresponding amide, which was analyzed by HPLC (Column: Chiralcel
AS; solvent: hexane/iPrOH=80:20; flow: 1.0 mLminÀ1; 254 nm; (S) en-
antiomer: tR =7.75 min, (R) enantiomer: tR =9.74 min). For styrene and
vinyl acetate derivatives, the conversion and regioselectivity were deter-
mined by 1H NMR spectroscopy from the crude reaction mixture. The
enantiomeric excesses of the hydroformylation products were determined
by GC with Supelcoꢁs Beta Dex 225 column (for details see Supporting
Information).
Barner, Chemical Industries, Vol. 89, Marcel Dekker, New York,
2003, pp. 359–367.
[5] a) C. J. Cobley, K. Gardner, J. Klosin, C. Praquin, C. Hill, G. T.
Whiteker, A. Zanotti-Gerosa, J. L. Petersen, K. A. Abboud, J. Org.
Chem. 2004, 69, 4031–4040; b) C. J. Cobley, J. Klosin, C. Qin, G. T.
Whiteker, Org. Lett. 2004, 6, 3277–3280.
[6] S. Breeden, D. J. Cole-Hamilton, D. F. Foster, G. J. Schwarz, M.
Wills, Angew. Chem. 2000, 112, 4272–4274; Angew. Chem. Int. Ed.
2000, 39, 4106–4108.
[7] a) T. P. Clark, C. R. Landis, S. L. Freed, J. Klosin, K. Abboud, J.
Am. Chem. Soc. 2005, 127, 5040–5042; b) A. T. Axtell, C. J. Cobley,
J. Klosin, G. T. Whiteker, A. Zanotti-Gerosa, K. A. Abboud, Angew.
Chem. 2005, 117, 5984–5988; Angew. Chem. Int. Ed. 2005, 44, 5834–
5838; c) P. J. Thomas, A. T. Axtell, J. Klosin, W. Peng, C. L. Rand,
T. P. Clark, C. R. Landis, K. A. Abboud, Org. Lett. 2007, 9, 2665–
2668.
[8] B. Zhao, X. Peng, Z. Wang, C. Xia, K. Ding, Chem. Eur. J. 2008, 14,
7847–7857.
[9] a) N. Sakai, S. Mano, K. Nozaki, H. Takaya, J. Am. Chem. Soc.
1993, 115, 7033–7034; b) K. Nozaki, N. Sakai, T. Nanno, T. Higashi-
jima, S. Mano, T. Horiuchi, H. Takaya, J. Am. Chem. Soc. 1997, 119,
4413–4423; c) T. Horiuchi, T. Ohta, E. Shirakawa, K. Nozaki, H.
Takaya, J. Org. Chem. 1997, 62, 4285–4292; d) T. Horiuchi, E. Shira-
kawa, K. Nozaki, H. Takaya, Organometallics 1997, 16, 2981–2986;
e) K. Nozaki, T. Matsuo, F. Shibahara, T. Hiyama, Adv. Synth. Catal.
2001, 343, 61–63; f) F. Shibahara, K. Nozaki, T. Hiyama, J. Am.
Chem. Soc. 2003, 125, 8555–8560.
Acknowledgements
[10] Y. Yan, X. Zhang, J. Am. Chem. Soc. 2006, 128, 7198–7202.
[11] For figures of the space-filling and stick models (based on CAChe
MM2 calculations) see reference [10].
[12] a) K. Sumi, T. Ikariya, R. Noyori, Can. J. Chem. 2000, 78, 697–703;
b) T. Morimoto, N. Mochizuki, M. Suzuki, Tetrahedron Lett. 2004,
45, 5717–5722.
We thank the National Institutes of Health (GM58832) and Merck &
Co., Inc. for financial support. The Bruker 400 MHz NMR spectrometer
used in these studies was purchased with grant No. 1S10RR023698-01A1
from the National Center for Research Resources (NCRR), a component
of NIH.
[13] S. Vyskocil, M. Smrcina, V. Hanus, M. Polasek, P. Kocovsky, J. Org.
Chem. 1998, 63, 7738–7748.
[1] For reviews, see: a) C. Claver, P. W. N. M. van Leeuwen in Rhodium
Catalyzed Hydroformylation, Vol. 22 (Eds.: C. Claver, P. W. N. M.
van Leeuwen), Kluwer Academic Publishers, Dordrecht, 2000; b) F.
Agbossou, J. F. Carpentier, A. Mortreux, Chem. Rev. 1995, 95,
2485–2506; c) W. A. Herrmann, B. Cornils, Angew. Chem. 1997,
109, 1074–1095; Angew. Chem. Int. Ed. Engl. 1997, 36, 1048–1067;
d) K. Nozaki, I. Ojima in Catalytic Asymmetric Synthesis, 2nd ed.
(Ed.: I. Ojima), Wiley-VCH, Weinheim, 2000, Chapter 7; e) B.
Breit, W. Seiche, Synthesis 2001, 1–36; f) B. Breit, Angew. Chem.
2005, 117, 6976–6986; Angew. Chem. Int. Ed. 2005, 44, 6816–6825.
[2] For recent reviews on AHF, see: a) M. Diꢂguez, O. Pꢃmies, C.
Claver, Tetrahedron: Asymmetry 2004, 15, 2113–2122; b) C. Claver,
M. Dieguez, O. Pamies, S. Castillon, Top. Organomet. Chem. 2006,
18, 35–64; c) J. Klosin, C. R. Landis, Acc. Chem. Res. 2007, 40,
1251–1259.
[14] a) G. Consiglio, P. Pino, L. I. Flowers Jr. , C. U. Pittman, J. Chem.
Soc. Chem. Commun. 1983, 612–613; b) S. G. Cohen, J. Crossley, E.
Khedouri, R. Zand, L. H. Klee, J. Am. Chem. Soc. 1963, 85, 1685–
1691; lit c>Z.-F. Xie, H. Suemune, K. Sakai, Tetrahedron: Asymme-
try 1993, 4, 973–980.
[15] A. T. Axtell, J. Klosin, K. A. Abboud, Organometallics 2006, 25,
5003–5009.
[16] D. Gleich, R. Schmid, W. A. Herrmann, Organometallics 1998, 17,
2141–2143.
[17] This theoretical model does not take account of the chelation of the
olefin functionality with the Rh center, which can also influence the
enantioface selection during or prior to the insertion step somewhat.
Meanwhile, we cannot rule out the possibility that both the steps of
Rh–alkyl and Rh–acyl formation are reversible. In such a situation,
the enantioselectivity will be affected in the final product-forming
Rh–acyl hydorgenolysis step.
[3] a) J. K. Still in Comprehensive Organic Synthesis, Vol. 4 (Eds.: B. M.
Trost, I. Flemming, L. A. Paquette), Pergamon Press, Oxford, 1991,
[18] C. J. Cobley, R. D. J. Froese, J. Klosin, C. Qin, G. T. Whiteker, K. A.
Abboud, Organometallics 2007, 26, 2986–2999.
ˇ
pp. 913–950; b) T. Smejkal, B. Breit, Angew. Chem. 2008, 120, 317–
321; Angew. Chem. Int. Ed. 2008, 47, 311–315.
[4] a) J. E. Babin, G. T. Whiteker (Union Carbide Corporation), WO
9303839, 1993; b) G. T. Whiteker, J. R. Briggs, J. E. Babin, B. A.
Received: August 11, 2009
Published online: November 30, 2009
Chem. Eur. J. 2010, 16, 871 – 877
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
877