Expanding the scope of atropisomeric monodentate P-donor ligands in asymmetric catalysis: Hydrogen-transfer reduction of α,β-unsaturated acid derivatives by rhodium/Ph-binepine catalysts

Article abstract of DOI:10.1002/hlca.200690169

A range of α,β-unsaturated acids and esters have been selectively reduced to the corresponding saturated acid derivatives by hydrogen transfer. As the reducing agent, formic acid was used in the presence of Rh ¹ complexes formed with the powerful chiral ligand Ph-binepine (1), an axially chiral binaphthalene-type monodentate P-donor ligand. Very high stereoselectivities (up to 97% ee) were obtained in the case of itaconic acid (2a).

Full text of DOI:10.1002/hlca.200690169

1716
Helvetica Chimica Acta – Vol. 89 (2006)
Expanding the Scope of Atropisomeric Monodentate P-Donor Ligands in
Asymmetric Catalysis: Hydrogen-Transfer Reduction of a,b-Unsaturated
Acid Derivatives by Rhodium/Ph-binepine Catalysts
by Elisabetta Albericoa), Ilenia Nieddub), Rossana Tarasb), and Serafino Gladiali*b)
a) C.N.R. – Istituto di Chimica Biomolecolare, Regione Baldinca, Li Punti, I-07040 Sassari
b) Dipartimento di Chimica, Università di Sassari,Via Vienna 2, I-07100 Sassari
(phone: +39-079-229546; fax: +39-079-229559; e-mail: gladiali@uniss.it)
Dedicated to Professor Giambattista Consiglio, a good friend and excellent chemist, for his remarkable
contributions to the advancement of stereoselective catalysis
A range of a,b-unsaturated acids and esters have been selectively reduced to the corresponding satu-
rated acid derivatives by hydrogen transfer. As the reducing agent, formic acid was used in the presence
of RhI complexes formed with the powerful chiral ligand Ph-binepine (1), an axially chiral binaphthalene-
type monodentate P-donor ligand. Very high stereoselectivities (up to 97% ee) were obtained in the case
of itaconic acid (2a).
Introduction. – The reduction of unsaturated functional groups by means of transi-
tion-metal-mediated hydrogen transfer from a suitable hydrogen donor such as i-PrOH
or HCOOH has experienced increasing success over the last years [1]. This mild meth-
odology represents a viable alternative to catalytic hydrogenation by molecular H2 due
to its operational simplicity and reduction of the risks associated with the use of an
easily inflammable gas of high diffusibility. Significant advances in this area have
been achieved in the asymmetric reduction of ketones and imines with catalysts of
excellent activity/selectivity based on Ru and Rh complexes with monotosyl-substi-
tuted diamines or amino alcohols as chiral modifiers [1a].
Less efforts have been devoted to the transfer hydrogenation of C=C bonds.
Although this process is thermodynamically favored, even when alcohols are used as
H-donors, only the reduction of conjugated C=C bonds is of practical significance,
while simple alkenes and dienes are poorly reactive. A perusal of the literature
shows that transfer hydrogenation of conjugated acid derivatives such as itaconic
acid and a-(acetamido)cinnamic acid selectively takes place at the C=C bond, and ster-
eoselectivities higher than 90% can be sometimes attained in this reaction when per-
formed in the presence of Ru [2] or Rh complexes [3] with chiral bidentate diphos-
phines such as binap [2], bppm [3a–f], or deguphos [3g]. No example of the application
of monodentate phosphines as chiral inducers in this reaction has been reported thus
far, in spite of the excellent performances displayed by binaphthalene-templated
monodentate P-ligands [4], such as phosphonites [5], phosphites [6], and phosphorami-
A
dites [7], in the Rh-catalyzed hydrogenation of a wide variety of substrates.
© 2006 Verlag Helvetica Chimica Acta AG, Zürich

Helvetica Chimica Acta – Vol. 89 (2006)
1717
Pursuing our long-lasting interest in the application of axially chiral phosphacyclic
P-donor ligands in asymmetric catalysis [8], in the course of the last years, we have
focused our attention on the scope of the ‘phenyl binaphthophosphepine’ (Ph-bine-
pine; 1)1) ligand in transition-metal-catalyzed stereoselective reactions (Fig. 1). The
monodentate phosphine ligand 1 had been synthesized for the first time in 1994 by
one of us [9]. In the meantime, Beller and co-workers [10], Zhang and Chi [11], and
ourselves [12] have exploited this ligand with remarkable success in a range of
metal-catalyzed reactions. These results have prompted us to expand the scope of
RhI complexes with 1 to the hydrogen transfer reduction of conjugated acid derivatives.
The results of this study are reported in this paper.
Fig.1. Structures of the phosphepine ligand Ph-binepine ((S)-1) and of its oxide
Results and Discussion. – The reduction of itaconic acid (=2-methylidenebutane-
dioic acid; 2a) with Et3
mark reaction of this investigation (Scheme 1). The commercially available azeotrope
of TEAF (Et3N/HCOOH 5 :2) was consistently used as the reducing agent throughout
N/HCOOH (TEAF) in DMSO has been selected as the bench-
AHCTREUNG
this study, and all catalytic experiments were performed with a substrate/Rh ratio of
66.7:1 and a substrate/HCOOH ratio of 1:5. These conditions allowed us to apply
strictly homogeneous conditions.
Under these conditions, the Rh complex selected as the catalyst precursor,
[Rh(nbd)Cl]2 (nbd=2,5-norbornadiene), displayed a modest, but not negligible, cata-
lytic activity, and itaconic acid was converted into methylsuccinic acid in 15% yield
after 24 h at 408 (Table 1, Entry 1). No separation of metallic Rh was noticed. While
this background reaction has no practical influence on the results of the stereoselective
experiments reported hereafter, its presence points out that DMSO has some protect-
ing effect towards the Rh complex, likely due to the formation of some Rh/DMSO spe-
cies [13].
Scheme 1. Transfer Hydrogenation of Itaconic Acid (2a)
1
)
Systematic name of 1: 4,5-dihydro-4-phenyl-3H-dinaphtho[2,1-c:1’,2’-e]phosphepine.

1718
Helvetica Chimica Acta – Vol. 89 (2006)
Table 1. Hydrogen-Transfer Reduction of 2a as a Function of Ligand-to-Metal Ratio. Conditions: in situ
generated catalyst from [Rh(nbd)Cl]2 and 1 in DMSO, 24 h, 408.
Entry
Ligand
Ligand/Rh ratio
Conversion [%]
ee [%]
Configuration
1
2
3
4
5
None
(S)-1
(R)-1
(S)-1
(S)-1(O)
–
15
81
98
91
15
–
54
82
82
0
–
(S)
(R)
(S)
racemic
1.1:1
2.1:1
4.2 :2
2.1:1
Addition of 1 equiv. of Ph-binepine (1) to a solution containing the complex
[Rh(nbd)Cl]2 strongly increased the catalytic activity, and methylsuccinic acid was
obtained in 81% yield, with 54% enantiomeric excess (ee) (Table 1, Entry 2). Both con-
version and stereoselectivity could be further improved, up to 98 and 82%, respectively,
when the reaction was run at a ligand/Rh ratio of 2 :1 (Entry 3). However, when this
ratio was further increased to 4 :1, the yield decreased to 91%, the ee values basically
remaining unchanged (Entry 4). These results point out that the ligand exerts a bene-
ficial effect on the activity of the metal (ligand-accelerated catalysis) [14]. Thereby,
the catalytically active species most likely features two monodentate P ligands per
Rh center.
Because of the possibility that oxidation of Ph-binepine (1) may occur due to the
long contact of the ligand with DMSO, a test was performed in the presence of the
phosphepine oxide (S)-1(O) (Fig. 1) in place of (S)-1. This led to a 15% yield of racemic
methylsuccinic acid (Entry 5), just as in the absence of ligand. This result, thus, confirms
that no phosphepine oxide ligand is involved at any time in the catalytic process.
Table 2 summarizes the results of the reduction of 2a (Scheme 1) in a range of dif-
ferent temperatures. As it frequently occurs, a decrease of the reaction temperature is
accompanied by a net increase of the stereoselectivity. Indeed, the ee value increased
from 82 to 94% when the reaction was performed at 22 instead of 408 (Table 2, Entries 1
vs. 2). Notably, this increase did not occur at the expense of the conversion, which
remained almost quantitative, even when the reaction time was reduced from 24 to 4
h. A further decrease in temperature, however, diminished the rate of reaction well
below practical application.
Table 2. Hydrogen-Transfer Reduction of 2a as a Function of Temperature and Solvent
Entry
Ligand
Solvent
T (8)
Conversion [%]
ee [%]
Configuration
1a)
2b)
3a)
4b)
(R)-1
(S)-1
(S)-1
(S)-1
DMSO
DMSO
–
–
40
22
40
22
98
>99
87
82
94
53
91
(R)
(S)
(S)
(S)
>99
a) Catalyst formed in situ from [Rh(nbd)Cl]2 and 2 equiv. of 1; 24 h. b) With preformed [Rh(nbd){(S)-1}2]+
CF3
SOÀ3 ; 4 h
AHCTREUNG
When the above reaction was run in neat TEAF (in the absence of DMSO), a
slightly lower, but still respectable, ee value of 91% was achieved at 228 (Table 2,

Helvetica Chimica Acta – Vol. 89 (2006)
1719
Scheme 2. Preparation of Cationic Rh Complexes with Ligand 1
Entry 4). Note, however, that there is a beneficial effect of the co-solvent on the life-
time of the catalyst: when the reaction was run in neat TEAF at 408, both conversion
and ee were considerably lower than in the presence of DMSO as a co-solvent (Entry 1
vs. 3). Because of this, DMSO was used in all subsequent experiments.
There is increasing evidence that the activity, lifetime, and stability of a charged
transition-metal catalyst, as well as the corresponding product selectivity, are signifi-
cantly influenced by the nature of the counter-ion [15a]. For instance, it has been
shown that in the asymmetric hydrogenation of imines [15b] and olefins [15c] promoted
by cationic Rh diphosphine catalysts, both activity and enantioselectivity change on
varying the nature of the counter-ion. To probe the influence of weakly coordinating
anions [16] on the efficiency of transfer hydrogenation, a range of Rh complexes of
the type [Rh{(S)-1}
C
G
Since the 1H- and 31P-NMR spectra of the three complexes showed only minor var-
iations due to the counter-ion, the discussion of the NMR data will be limited to the
spectra of 3b.
In the 31P-NMR spectrum (in CDCl3 solution), coordination of the ligand to the metal was confirmed
by the presence of a doublet at ca. d(P) 35, with a downfield shift of nearly 27 ppm compared to the free
ligand. The splitting is due to the coupling between the coordinated P- and 103Rh-atoms, with a 1J(P,Rh)
coupling constant of ca. 155 Hz. In the corresponding 1H-NMR spectrum, four signals, each accounting
for two H-atoms, were observed for the coordinated norbornadiene (nbd). The allylic (d(H) 3.60) and
olefinic (d(H) 3.91, 5.58) signals were observed as broad singlets, and the bridging nbd CH2 group reson-
ated as a sharp singlet at d(H) 1.33. Four signals, each integrating for two H-atoms, were assigned to the
four benzylic CH2 groups on each coordinated ligand: d(H) 1.90 (d, J=12.1), 2.51 (d, J=14), 3.18 (d,
J=14), 2.35 (t-like m). In the spectrum of the free ligand 1, the diastereoisotopic H-atoms of the CH2
groups give rise to two sets of eight lines for each methylene group at d(H) 2.66 and 2.82. Upon coordi-
nation to Rh, the chemical shifts of the latter signals are spread over a wider range (d(H) 1.90 and 3.18).
Selective decoupling of the three doublets shows that there is a 1J(H,H) coupling, and that coupling to the
P-atom is no longer present2). For the resonance at d(H) 2.35, on the contrary, irradiation of the corre-
2
)
Geminal P,H coupling constants have a Karplus-like dependence from the HÀCÀP electron-pair
dihedral angle. Upon coordination to the metal, a change of the electron density and of the hybrid-

1720
Helvetica Chimica Acta – Vol. 89 (2006)
sponding geminal H-atoms turned the multiplet into a doublet whose splitting arises from a residual cou-
pling. This is most probably due to an agostic interaction of one pseudo-axial H-atom with the metal, as
pointed out by inspection of molecular models. Several multiplets, roughly corresponding to eight dou-
blets and four triplets at d(H) 6.8 and 8.7, and integrating for 34 H-atoms, finally accounted for the aro-
matic H-atoms. In summary, these data support C2-symmetric [Rh{(S)-1}2
(nbd)]+ XÀ complexes.
A
Whichever cationic complex 3 was used, the reduction of itaconic acid (2a) at room
temperature was found to be complete within 4 h (Table 3). However, there was a sig-
nificant influence of the nature of the counter-ion on stereo-induction The ee values
increased from 82 to 89 to 94 and, finally, to 97% for X=ClÀ [18], [3,5-
(CF3)2C6H3]BÀ, CF3SOÀ, and BFÀ, respectively (Table 3). Comparison of the enantiose-
A
lectivities obtained at 3408 with 4the in situ generated catalyst [Rh(nbd)Cl]2
(Table 3,
G
Entry 4; 82% ee) and with preformed [Rh{(S)-1}
firms the role of the counter-ion in determining the stereoselectivity. As the coordinat-
ing ability of the counter-ions decreases in the order ClÀ >CF3SO3À >BF4À >[3,5-
(CF3)2C6H3]4
BÀ, a direct correlation between this property and the trend in stereoselec-
tivity is not immediately apparent.
2(nbd)]+ BFÀ4 (Entry 5; 89% ee) con-
AHCTREUNG
AHCTREUNG
Table 3. Hydrogen-Transfer Reduction of 2a with Catalysts of the Type [Rh(nbd){(S)-1}2]+XÀ as a
Function of Counter-Ion (XÀ), Temperature, and Reaction Time
Entry
XÀ
T [8]
t [h]
Conversion [%]
ee [%]
Configuration
BFÀ4
1
2
3
4
5
22
22
22
40
40
4
4
4
24
24
>99
>99
>99
>99
>99
97
94
89
82
89
(S)
(S)
(S)
(S)
(S)
CF
A
[3,5-(CF3)2C6H3]4
BÀ
a
)
BFÀ4
a) Catalyst made from 1/2 [Rh(nbd)Cl]2
and two equiv. of (S)-1.
G
This behavior is in strong contrast with what was observed in the reduction of the
same substrate with molecular H2, where changing the counter-ion has a dramatic effect
on the activity of the catalyst [19]. When the hydrogenation was performed in CH2Cl2,
conversions of 39%, 84%, and >99% within 1 h were observed for 3c, 3a, and 3b,
respectively. The effect on enantioselectivity was otherwise negligible.
For the sake of comparison with other monodentate P-donor ligands of similar
design, the transfer hydrogenation of itaconic acid (2a) was also performed with in
situ prepared catalysts with a range of chiral monodentate ligands such as the phosphite
(R)-4 [6], the phosphonite (R)-5 [5], the phosphoramidite (R)-6 (MonoPhos) [7], and
the dithiaphosphepine (R)-73) (Fig. 2). The first three ligands are, in general, as effi-
cient as Ph-binepine (1) in Rh-catalyzed asymmetric hydrogenations of a,b-unsatu-
rated acid derivatives (ee values of 90–99%).
In Table 4, the results obtained by screening the above different ligands are
reported, together with the best results scored with Ru and Rh catalysts modified by
ization at the P-atom give rise to a change in the bond angles, which is reflected in a change of the
associated coupling constants [17].
This ligand has been prepared in our laboratory. Its synthesis will be reported elsewhere.
3
)

Helvetica Chimica Acta – Vol. 89 (2006)
1721
Fig. 2. Structures of mono-and bidentate P ligands tested in transfer hydrogenations of 2a
the chiral diphosphine ligands binap [2b] and bppm [3c], respectively. Although the
results are not fully comparable because of somewhat different experimental condi-
tions (longer reaction times were necessary in some cases for a definite conversion to
be observed), the superiority of ligand 1 over the other monodentate ligands in the
reduction of 2a is clearly evident. With ligands (R)-4, (R)-5, and (R)-7, the conversions
were generally poor, with negligible or even null stereoselectivities. This might be due
to ligand degradation. At variance, the Rh complex formed with (R)-6, which is known
to be more stable than the complexes with the other ligands [7b], afforded quantitative
conversions (Table 4, Entries 5 and 6), thus indicating that the relevant catalyst can tol-
erate these reaction conditions. However, (R)-6 was found to be a poor chiral inducer,
the ee values not exceeding 27%.
The excellent performance (97% ee) of Ph-binepine (1) in the reduction of 2a
(Table 4, Entry 1) was challenged only by some catalysts containing bidentate bisphos-
Table 4. Hydrogen-Transfer Reduction of 2a with Different Chiral Ligands
Entry
Conditionsa)
Ligand
Conversion [%]
ee [%]
Configuration
1
2
3
4
5
6
7
8
9
A, B
A, C
D, C
D, C
A, C
D, C
A, C
E
(S)-1
(R)-4
(R)-4
(R)-5
(R)-6
(R)-6
(R)-7
(2S,4S)-bppm
(S)-binap
>99
35
9
15
>99
>99
28
>99
>99
97
5
2
18
24
27
0
92
97
(S)
(R)
(R)
(R)
(R)
(R)
racemic
(S)
(R)
F
a) A: Preformed Rh complex with BFÀ4 ; B: 4 h, 228; C: 24 h, 408; D: Rh complex formed in situ; E:
[Rh2(OAc)4] formed in situ, 5 h, 278; F: [RuH{(S)-binap}2]PF6, i-PrOH, 24 h, 80–838.

1722
Helvetica Chimica Acta – Vol. 89 (2006)
phines such as binap (97% ee) and bppm (92% ee) (Entries 8 and 9, resp). While in the
first case the reaction conditions are completely different from ours (Ru catalyst, i-
PrOH as H-donor), in the second case, both metal and H-donor were the same as in
our case. Unfortunately, bppm is structurally not related at all to Ph-binepine (1),
which prevents any meaningful comparison.
In the aim to explore the scope of this new catalyst and to get some insight into the
reaction mechanism, a range of different substrates (Fig. 3) was tested in the hydrogen-
transfer reduction in the presence of preformed or in situ generated 3a (X=BFÀ4 ).
Fig. 3. Structures of substrates tested in transfer hydrogenations
First, we investigated the outcome of the reaction of the esters 2b–d of itaconic
acid, and the results are summarized in Table 5. It is known since long that these esters
display different ligating properties towards Rh, and that the mode they bind the metal
depends on the location of the substituents on the C=C bond. It is also recognized that
the binding mode of these substrates is a critical factor in steering chiral induction by
hydrogenation with molecular H2 [20].
Table 5. Transfer Hydrogenation of Itaconic and Butenedioic Acid Derivatives
Entry
Ligand
Substrate
Conditionsa)
Conversion [%]
ee [%]
Configuration
1
2
3
4
5
6
(S)-1
(S)-1
(S)-1
(S)-1
(R)-1
(R)-1
2a
2b
2c
2d
8
A
A
A
A
B
B
>99
57
99
75
97
97
13
81
28
2
(S)
(R)
(S)
(R)
(R)
(R)
9
98
12
a) A: Preformed Rh complex with BFÀ4 , 228, 4 h; B: Rh complex formed in situ, 408, 24 h.

Helvetica Chimica Acta – Vol. 89 (2006)
1723
Under the screening conditions used, the diacid 2a and the a-monomethyl ester 2c
were quantitatively reduced, with excellent (97% ee) and good (81% ee) stereoselec-
tivities, respectively, providing the corresponding (S)-configured methylsuccinates
(Table 5, Entries 1 and 3). The reduction of the b-monomethyl and the dimethyl esters
2d and 2b, respectively, proceeded at a slower rate, with conversions as low as 75 and
57% after 4 h. At variance with the previous cases, the reduction products had the
(R)-configuration, and their enantiomeric purities were modest (28 and 13% ee,
resp.; Entries 2 and 4). In addition, during the reaction, partial double-bond migration
in the diester 2b from the terminal to the internal position was observed, resulting in the
formation of the dimethyl ester of mesaconic acid (9), whose concentration was found
to build up steadily in solution because of its low reactivity towards reduction.
A similar dichotomic behavior of itaconic acid derivatives had been observed sev-
eral years ago in a Rh-catalyzed asymmetric hydrogenation [20]. It was attributed to
the ability of the b-COOH group to support better the bidentate binding of the sub-
strate to the metal through chelate coordination of the carboxylate arm. Leitner and
Lange [3f] came to a similar conclusion when they performed transfer hydrogenations
of the same substrates with the Rh catalyst formed with bppm as ligand. Based on label-
ing experiments, they pointed out the pivotal role of the free COOH group in address-
ing the addition of hydrogen across the C=C bond.
Further confirmation of the importance of this structural feature came from the
results obtained in the hydrogen-transfer reduction of citraconic acid (8) and mesaconic
acid (9). Both substrates gave rise to almost racemic products (Table 5), thus corrobo-
rating the strategic role of the b-COOH group in itaconic acid derivatives.
The results obtained in the transfer hydrogenation of a range of precursors of either
a- or b-amino acids are collected in Table 6. Both a-(acetamido)acrylic acid (10a) and
its methyl ester 10b were found to be poor substrates for this reaction. Within 4 h at
room temperature, the acid failed to react at all, and the ester suffered from low con-
version (35%) and poor ee (33%) (Table 6, Entry 4). Although a higher temperature
(408) and longer reaction time (24 h) led to complete hydrogenation of both substrates
(Entries 1 and 2), poor ee values were obtained (18 and 32%, resp., for 10a and 10b).
Slightly better results were achieved with a-(acetamido)cinnamic acid (11a) and its
methyl ester 11b. These substrates are more reactive than the corresponding acrylic
acid congeners, and fair conversions of 40 and 57%, respectively, were observed
(Entries 5 and 6) after 4 h even at 228. Stereoinduction was slightly better than before,
but still remained far from being exciting (58 and 46% ee, resp.). Compared with the
itaconic acid derivatives 2, the derivatives of both 10 and 11 are less reactive, give
lower ee values and, with the exception of methyl a-(acetamido)acrylate (10b), give
rise to the enantiomer having the same configuration of the inducing ligand (Table 6).
Notably, regardless whether the reducing agent is molecular H2 or HCOOH, the
reduction of itaconic acid (2a) with Rh catalysts containing Ph-binepine (1) as ligand
gave rise to methylsuccinic acid of the same configuration and of comparable enantio-
meric purities. This supports the view that these processes rely on a common intermedi-
ate, and that the differences between the two processes must be restricted to the earliest
stages of the reaction path, outside the catalytic cycle.
At variance, in the reduction of the substrates 10 and 11, the configuration of the
reduction product depends on the mode of reduction. With the sole exception of methyl

1724
Helvetica Chimica Acta – Vol. 89 (2006)
a-(acetamido)acrylate, the handedness of the reaction was found to be ‘opposite’
(inducing ligand and reaction products having opposite configurations) with molecular
H2 [10c], but ‘same’ (inducing ligand and product with identical configurations) in the
case of HCOOH as reducing agent. This is a clear-cut indication that the chemistry
involved in these two reductive processes, albeit similar, is far from being identical,
and that the behavior of itaconic acid (2a) in hydrogen-transfer reduction is special
compared with other structurally related substrates.
Enantioselective transfer hydrogenation has been applied for the first time to the
reduction of the b-(acetamido)acrylates 12 and 13 with some success (Table 6). These
are precursors of b-amino acids, a class of compounds of increasing interest for the
pharmaceutical industry [21].
Table 6. Transfer Hydrogenation of Precursors of a-and b-Amino Acids with Preformed [Rh(nbd){(S)-
1}2]+CF3SOÀ3 . Conditions: DMSO, 4 h, 228.
Entry
Substrate
Conversion [%]
ee [%]
Configuration
1a)
2a)
3
10a
10b
10a
>99
>99
0
18
32
–
(S)
(R)
–
4
5
6
7
8
9
10
10b
11a
11b
(Z)-12
(E)-12
(E)-13
(Z/E)-13b)
35
40
57
>99
97
33
58
46
12
46
4
(R)
(S)
(S)
(S)
(S)
(R)
(R)
36
78
2
a) In situ formed Rh complex, 408, 24 h. b) (Z)/(E) 71:29.
The Me-substituted derivatives 12 were reduced almost quantitatively within 4 h at
228, giving rise to the hydrogenated product of (S)-configuration (same), irrespective of
the geometry of the C=C bond of the substrates. For the (E)-isomer, the ee was higher
(46%) than for the (Z)-isomer (12%). This behavior contrasts with that observed by us
in the hydrogenation of these substrates with molecular H2 by Rh/Ph-binepine com-
plexes, where the (E)-isomer is quite unreactive, while the (Z)-isomer gives the satu-
rated product in more than 90% ee [19]. Notably, in this case, the configuration of
the product does not change in moving from hydrogen transfer to hydrogenation, the
relation being same in both cases. From this point of view, the behavior of b-amino
acid precursors contrasts with that of a-(acylamino) acid precursors.
The configurational isomers of the phenyl-substituted substrates 13 displayed a
quite different reactivity pattern: whereas (E)-13 was found to be much less reactive
than both (Z)-13 and the corresponding Me derivative (E)-12, the hydrogenation
rates for (Z)-13 and (Z)-12 were comparable. Whichever the geometry of the C=C
bond of 13, the stereoselectivity was miserable.
Recently, it has been shown that the combination of two different monodentate P-
donor ligands, either both chiral, or one chiral and the other achiral, may, in some cases,
produce a more-stereoselective and/or more-active catalyst [22]. This new strategy has

Helvetica Chimica Acta – Vol. 89 (2006)
1725
been exploited with some success in Rh-catalyzed asymmetric hydrogenations using
various combinations of phosphines, phosphites, phosphonites, and phosphoramidites.
This ‘mixed-ligand approach’ may change the outcome of the reaction when at least
two monodentate ligands are bound to the metal of the active catalyst in the transition
state.
We have attempted to apply this mixed-ligand approach to the hydrogen-transfer
reduction of methyl a-(acetamido)acrylate (10b), and the results are shown in Table
7. This substrate has been selected for a brief screening of two binary mixtures made
up by the combination of 1 with one achiral or one chiral monodentate ligand, triphe-
nylphosphine or (R)-6, respectively.
Table 7. Transfer Hydrogenation of 10b Using the ‘Mixed-Ligand Approach’ (see text)
Entry
Catalyst system
Conditionsa)
Conversion [%]
ee [%]
Configuration
1
2
3
4
5
6
7
8
Rh/(S)-1 1:2
Rh/(R)-6 1:2
A
B
C
D
D
D
E
E
35
n.d.b)
>99
72
8
43
33
–
–
23
3
11
3
(R)
–
–
(R)
(S)
(R)
(R)
(R)
Rh/Ph3
A
Rh/(S)-1/(R)-6 1:1:1
Rh/(R)-1/(R)-6 1:1:1
Rh/(S)-1/(R)-6 1:1:2
Rh/(S)-1/Ph
Rh/(S)-1/Ph
U
>99
>99
8
a) A: Preformed complex with CF3SO3À, 228, 4 h; B: complex formed in situ, 228, 45.5 h. C: complex
U
formed in situ, 228, 2 h; D: complex formed in situ, 228, 24 h; E: complex formed in situ, 228, 4 h. b) Not
detected.
For the ligand combination (S)-1/Ph3P, the results of two runs at different ligand
N
ratios are in keeping with the simultaneous presence of two catalysts containing two
equivalents of the same ligand, each one contributing to the outcome of the reaction
for its relative extent (Table 7, Entries 7 and 8 vs. 1 and 3). The situation is different
in the case of the combination (S)-1/(R)-6. Here, the sharp difference in the rate of
reduction observed for the matched pair (S)-1/(R)-6 and for the mismatched one
(R)-1/(R)-6 (Entries 4 and 5) provides evidence that the complex featuring a ‘hetero-
type combination’ of ligands is the exclusive or the prevailing catalyst in solution.
This species is, however, poorly stereoselective, and the ee values were, in all experi-
ments, lower than that obtained with the complex made from (S)-1 alone.
Conclusions. – We have demonstrated that the ligand Ph-binepine (1) is a powerful
chiral inducer, even in asymmetric hydrogen-transfer reductions of C=C bonds, a reac-
tion in which, so far, there has been no precedent for the use of chiral monodentate P-
donor ligands. Stereoselectivities as high as 97% have been attained with the most
appropriate substrate, itaconic acid (2a). This value is slightly higher than the best
ones previously obtained in this reaction with bidentate chelating diphosphines, and
supports further the view that axially chiral monodentate P-donor ligands can equal
or even outperform the more-popular and long-standing bidentate counterparts.

1726
Helvetica Chimica Acta – Vol. 89 (2006)
The catalytic behavior of Rh/Ph-binepine cationic complexes towards itaconic acid
derivatives stresses the importance of the presence of a free b-COOH group in the sub-
strate. This structural feature not only dictates the configuration of the reduction prod-
uct, but is as well mandatory for quantitative conversions and high stereoselectivities.
This observation provides evidence that the b-carboxylate anion dictates the mode
the substrate binds to Rh, and strongly supports the formation of a neutral adduct
where the COOÀ moiety provides for the bidentate chelate coordination of the sub-
strate to the metal. The intermediacy of such a species has been invoked in asymmetric
hydrogenations [20] and hydrogen-transfer reductions [3f] of these derivatives with Rh-
based catalyst containing chiral diphosphine ligands. Our results are absolutely in keep-
ing with this hypothesis.
The good performance of Ph-binepine (1) in the hydrogen transfer reduction of a,b-
unsaturated acid derivatives expands further the scope of this ligand to a new process,
which adds to the previous ones, where this ligand has already shown its efficiency. It is
worth to stress that, by now, the list of reactions where this chiral inducer has led to ee
values >90% (or where it has outperformed by far any other competitor) is fairly long,
covering a wide variety of organic reactions catalyzed by different transition-metal
complexes of Rh [10a–c], Ru [10d,e], Pt [12b], and Pd [12a]. The results of this
work emphasize further the versatility of Ph-binepine (1) as a chiral inducer of general
utility, and confirm the comparably wider scope of this ligand in respect to all the other
binaphthalene-core monodentate P-donor ligands of similar design.
Experimental Part
1. General. All reactions were carried out under anh. N2 using standard Schlenk techniques, unless
otherwise specified. Solvents were dried by standard procedures, distilled, and stored under N2. Pur-
chased chemicals (Aldrich, Lancaster, Strem) were used without further purification. Ligand 1 and its
oxide, 1(O) [10c], [Na{[3,5-(CF3)2C6H3]4}B] [23a,b], [Rh(nbd)2]+[BF4]À, and [Rh(nbd)2]+[CF3SO3]À
[24], (E)- and (Z)-12 as well as (E)- and (Z)-13 [25] were all prepared according to literature procedures.
Acids obtained from transfer hydrogenation of the corresponding unsaturated substrates were converted
to the corresponding esters with (trimethylsilyl)diazomethane before analysis by chiral GC [26]. NMR
spectra were recorded on a Varian Mercury Plus spectrometer operating at 400 MHz (1H), 100 MHz
(13C), and 162 MHz (31P) at r.t. Chemical shifts d are reported in ppm rel. to Me4Si (1H, 13C) or 85%
G
H3PO4 (31P) as external standard. The enantiomeric excess (ee) was determined by chiral GC (see below).
2. Determination of Enantiomeric Excess. The ee values of the following products were determined
by chiral GC or optical rotation. GC Retention times (tR) are given in min.
2.1. 2-Methylsuccinic Acid Dimethyl Ester (=Dimethyl 2-Methylbutanedioate). Column, DACTBS-
GAMMA-OV 1701 (25 m, i.d. 0.25 mm, ft 0.25 mm); temp. 508, 28/min, 1708, 20 min; PTV injector
(0–4008), detection at 2508, N2 carrier gas (15 psi, 2 ml/min); tR 17.39 (S), 18.29 (R).
2.2. N-Acetylalanine Methyl Ester (=Methyl 2-(Acetamido)propanoate). Column, Chirasil-Val-L (50
m, i.d. 0.32 mm, ft 0.25 mm); temp. 808, 10 min, 18/min, 1108, 20 min, 28/min, 2008, 30 min; PTV injector
(0–4008), detection at 2508, N2 carrier gas (20 psi, 2 ml/min); tR 10.68 (R), 12.10 (S).
2.3. N-Acetylphenylalanine Methyl Ester (=Methyl 2-(Acetamido)-3-phenylpropanoate). Column,
Chirasil-Val-L (50 m, i.d. 0.32 mm, ft 0.25 mm); temp. 1108, 5 min, 1.58/min, 1808, 1 min, 28/min, 2008,
30 min; PTV injector (0–4008), detection at 2508, N2 carrier gas (20 psi, 2 ml/min); tR 36.49 (R), 36.96 (S).
2.4. 3-Acetamidobutyrate Ethyl Ester (=Ethyl 3-(Acetamido)butanoate). Column, Mega DEttBuSi-
BETACDX (25 m, i.d. 0.25 mm, ft 0.25 mm); temp. 1208, PTV injector (0–4008), detection at 2508, N2 car-
rier gas (17 psi, 1 ml/min); tR 2.82 (S), 3.89 (R).

Helvetica Chimica Acta – Vol. 89 (2006)
1727
2.5. 3-Acetamido-3-phenylpropionate Ethyl Ester (=Ethyl 3-(Acetamido)-3-phenylpropanoate). The
ee and the configuration of the prevailing enantiomer were assigned based on reported optical-rotation
data [27].
3. Syntheses of the Complexes 3. 3.1. [Rh(nbd){(S)-1}2]+ BF4À (3a). To a soln. of [Rh(nbd)2]+ [BF4]À
(41 mg, 0.11 mmol) in MeOH/CH2Cl2 1:1 (10 ml) was added (S)-1 (100 mg, 0.257 mmol), and the mixture
was stirred at r.t. for 1 h. The solvent was removed at reduced pressure, and the residue was dissolved in
CH2Cl2 (2 ml). Addition of Et2
ACHTREUNG
washed with Et2O, dried in vacuo, and recrystallized from CH2Cl2/Et2
A
ACHTREUNG
(400 MHz, CDCl3): 1.39 (s, 2 H, nbd); 1.87 (d, J=12.4, CH2); 2.30–2.37 (m, CH2); 2.51 (d, J=14.0,
CH2); 3.10 (d, J=13.6, CH2); 3.60 (br. s, 2 H, nbd); 3.95 (br. s, 2 H, nbd); 5.53 (br. s, 2 H, nbd); 6.70
(d, J=8.4, 2 arom H); 7.12 (d, J=8.4, 2 arom. H); 7.17–7.21 (m, 4 arom. H); 7.35–7.39 (m, 2 arom.
H); 7.42–7.48 (m, 4 arom. H); 7.50–7.53 (m, 12 arom. H); 7.76 (d, J=8.8, 2 arom. H); 7.97 (d, J=8.4,
2 arom. H); 8.04 (d, J=8.4, 2 arom. H); 8.30 (d, J=8.0, 2 arom. H). 31P-NMR (162 MHz, CDCl3):
35.74 (d, J(P,Rh)=153.2). Anal. calc. for C63H50BF4P2Rh (1058.76): C 71.47, H 4.76; found: C 71.68, H
4.51.
3.2. [Rh(nbd){(S)-1}2]+ CF3SO3À (3b). To a soln. of [Rh(nbd)2]+[CF3SO3]À (56 mg, 0.128 mmol) in
MeOH/CH2Cl2 1:1 (10 ml) was added (S)-1 (100 mg, 0.257 mmol), and the mixture was stirred at r.t.
for 1 h. The solvent was removed at reduced pressure, and the residue was dissolved in CH2Cl2 (2 ml).
Addition of Et
N
G
dried in vacuo, and recrystallized from CH2Cl2/Et2
AHCTREUNG
1.33 (s, 2 H, nbd); 1.90 (d, J=12.1, CH2); 2.32–2.35 (m, CH2); 2.51 (d, J=14.0, CH2), 3.18 (d, J=14.0,
CH2); 3.60 (br. s, 2 H, nbd); 3.91 (br. s, 2 H, nbd); 5.58 (br. s, 2 H, nbd); 6.71 (d, J=8.6, 2 arom. H);
7.11 (d, J=8.2, 2 arom. H); 7.15–7.20 (m, 4 arom. H); 7.22–7.28 (m, 2 arom. H); 7.34–7.38 (m, 4
arom. H); 7.42–7.53 (m, 12 arom. H); 7.74 (d, J=8.2, 2 arom. H); 7.96 (d, J=8.2, 2 arom. H); 8.00 (d,
J=8.2, 2 arom. H); 8.28 (d, J=8.6, 2 arom. H). 31P-NMR (162 MHz, CDCl3): 35.79 (d, J(P,
Rh)=155.1). Anal. calc. for C64H50F3O3P2SRh (1121.02): C 68.57, H 4.54; found: C 68.34, H 4.65.
3.3. [Rh(nbd){(S)-1}2]+ {[3,5-(CF3)2C6H3]4B}À (3c) [23c]. To a soln. of [Rh(nbd)Cl]2 (29.5 mg,
0.064mmol) in CH2Cl2 was added a soln. of [Na{[3,5-(CF3)2C6H3]4}B] (113.4 mg, 0.128 mmol) in
CH2Cl2 (2 ml). The clear yellow soln. became rapidly cloudy due to precipitation of NaCl. The mixture
was stirred for 45 min, and then filtered over a pad of Na2SO4. To the resulting clear filtrate, (S)-1 (100
mg, 0.257 mmol) was added, which resulted in an orange color. The soln. was stirred at r.t. for 2 h. Addi-
tion of hexane brought about precipitation of the product as a micro-crystalline red powder, which was
filtered off and recrystallized from CH2Cl2/Et2
O. Yield: 59 mg (50%). 1H-NMR (400 MHz, CDCl3): 1.34
A
(s, 2 H, nbd); 1.83 (d, J=12.0, CH2); 2.38–2.44 (m, CH2); 2.54 (d, J=14.4, CH2); 2.79 (d, J=14.0, CH2);
3.69 (br. s, 2 H, nbd); 3.82 (br. s, 2 H, nbd); 5.19 (br. s, 2 H, nbd), 6.56 (d, J=8.8, 2 arom. H); 7.09 (d,
J=8.0, 2 arom. H); 7.14–7.35 (m, 12 arom. H); 7.39–7.56 (m, 10 arom. H); 7.12 (s, 4 arom. H); 7.70
(d, 2 arom. H); 7.72 (s, 8 arom. H); 7.78 (d, J=8.4, 2 arom. H); 8.01 (d, J=8.4, 2 arom. H); 8.12 (d,
J=8.0,
2
arom. H). 31P-NMR (162 MHz, CDCl3): 35.53 (d, J(P,Rh)=155.1). Anal. calc. for
C95H62BF24P2Rh (1834.31): C 62.18, H 3.41; found: C 62.31, H 3.30.
4. General Procedure for the Transfer Hydrogenation of Butenedioic Acids [3c]. The substrate (2
mmol) and the preformed Rh complex (0.030 mmol) were dissolved in DMSO (4 ml) under N2. In the
case of the in situ generated catalyst, the soln. was first stirred for 30 min. Then, HCOOH/Et
A
(10 mmol HCOOH, 4 mmol Et3N) was added, whereupon the color of the soln. changed from orange
AHCTREUNG
to yellow under evolution of gas. After stirring the mixture at the given temp. for the required time
(see Tables), 10% aq. HCl (10 ml) was added. The mixture was filtered, the product was extracted
with AcOEt (4×50 ml), and the combined org. layers were dried (Na2SO4). The solvent was removed
in vacuo to afford the crude product, whose identity was checked by NMR. Conversions and ee values
were determined by GC, as reported above.
5. General Procedure for the Transfer Hydrogenation of Precursors of a-and b-Amino Acids [3b].
The substrate (2 mmol) and the preformed Rh complex (0.030 mmol) were dissolved in DMSO (4 ml)
under N2. In the case of the in situ catalyst, the soln. was first stirred for 30 min. Then, HCOOH/Et3
5 :2 (10 mmol HCOOH, 4 mmol Et3N) was added, whereupon the color of the soln. changed from orange
to yellow under evolution of gas. After stirring the mixture at the given temp. for the required time, the
A
AHCTREUNG

1728
Helvetica Chimica Acta – Vol. 89 (2006)
mixture was made alkaline with 2N aq. NaOH (5 ml), and then filtered. The filtrate was washed with Et2
A
(4×15 ml), acidified with 10% aq. HCl (3 ml), and extracted with Et2O (4×15 ml). The combined org.
AHCTREUNG
layers were dried (Na2SO4), and the solvent was removed in vacuo. The identity of the crude product
was checked by NMR. Conversions and ee values were determined by GC, as reported above.
REFERENCES
[1] a) S. Gladiali, E. Alberico, Chem. Soc. Rev. 2006, 35, 226; b) J. S. M. Samec, Jan-E. Bäckvall, P. G.
Andersson, P. Brandt, Chem. Soc. Rev. 2006, 35, 237.
[2] a) J. M. Brown, H. Brunner, W. Leitner, M. Rose, Tetrahedron: Asymmetry 1991, 2, 331; b) M. Sab-
uri, M. Ohnuki, M. Ogasawara, T. Takahashi, Y. Uchida, Tetrahedron Lett. 1992, 33, 5783.
[3] a) H. Brunner, M. Kunz, Chem. Ber. 1986, 119, 2868; b) H. Brunner, W. Leitner, Angew. Chem., Int.
Ed. 1988, 27, 1180; c) H. Brunner, E. Graf, W. Leitner, K. Wutz, Synthesis 1989, 743; d) H. Brunner,
W. J. Leitner, Organomet. Chem. 1990, 387, 209; e) W. Leitner, J. M. Brown, H. Brunner, J. Am.
Chem. Soc. 1993, 115, 152; f) S. Lange, W. J. Leitner, J. Chem. Soc., Dalton Trans. 2002, 752; g)
A. M. d’A. Rocha Gonsalves, J. C. Bayón, M. M. Pereira, M. E. S. Serra, J. P. R. Pereira, J. Organo-
met. Chem. 1998, 553, 190.
[4] T. Jerphagnon, J.-L. Renaud, C. Bruneau, Tetrahedron: Asymmetry 2004, 15, 2101.
[5] C. Claver, E. Fernandez, A. Gillon, K. Heslop, D. J. Hyett, A. Martorell, A. G. Orpen, P. G. Pringle,
Chem. Commun. 2000, 961.
[6] M. T. Reetz, G. Mehler, Angew. Chem., Int. Ed. 2000, 39, 3889; M. T. Reetz, T. Sell, Tetrahedron
Lett. 2000, 41, 6333.
[7] a) M. van den Berg, A. J. Minnaard, E. P. Schudde, J. van Esch, A. H. M. de Vries, J. G. de Vries, B.
L. Feringa, J. Am. Chem. Soc. 2000, 122, 11539; b) M. van den Berg, A. J. Minnaard, R. M. Haak, M.
Leeman, E. P. Schudde, A. Meetsma, B. L. Feringa, A. H. M. de Vries, C. E. P. Maljaars, C. E. Will-
ans, D. Hyett, J. A. F. Boogers, H. J. W. Henderickx, J. G. de Vries, Adv. Synth. Catal. 2003, 345, 308.
[8] S. Gladiali, A. Dore, D. Fabbri, O. De Lucchi, G. Valle, J. Org. Chem. 1994, 59, 6363; S. Gladiali, D.
Fabbri, Chem. Ber./Recl. Trav. Chim. Pays Bas 1997, 130, 543.
[9] S. Gladiali, A. Dore, D. Fabbri, O. De Lucchi, M. Manassero, Tetrahedron: Asymmetry 1994, 5, 511.
[10] a) K. Junge, G. Oehme, A. Monsees, T. Riermeier, U. Dingerdiessen, M. Beller, Tetrahedron Lett.
2002, 43, 4977; b) K. Junge, G. Oehme, A. Monsees, T. Riemer, U. Dingerdissen, M. Beller, J. Orga-
nomet. Chem. 2003, 675, 91; c) K. Junge, B. Hagemann, S. Enthaler, A. Spannenberg, M. Michalik,
G. Oehme, A. Monsees, T. Riermeier, M. Beller, Tetrahedron: Asymmetry 2004, 15, 2621; d) K.
Junge, B. Hagemann, S. Enthaler, G. Oheme, M. Michalik, A. Monsees, T. Riemeier, U. Dirgendes-
sen, M. Beller, Angew. Chem., Int. Ed. 2004, 43, 5066; e) B. Hagemann, K. Junge, S. Enthaler, M.
Michalik, T. Riemeier, A. Monsees, M. Beller, Adv. Synth. Catal. 2005, 347, 1978.
[11] Y. Chi, X. Zhang, Tetrahedron Lett. 2002, 43, 4849.
[12] a) G. Zanoni, S. Gladiali, A. Marchetti, P. Piccinini, I. Tredici, G. Vidari, Angew. Chem., Int. Ed.
2004, 43, 846; b) L. Charruault, V. Michelet, R. Taras, S. Gladiali, J.-P. Genêt, Chem. Commun.
2004, 850.
[13] R. Dorta, H. Rozemberg, L. J. W. Shimon, D. Milstein, Chem.–Eur. J. 2003, 9, 5237.
[14] D. J. Berrisford, C. Bolm, K. B. Sharpless, Angew. Chem., Int. Ed. 1995, 34, 1059.
[15] a) A. Macchioni, Chem. Rev. 2005, 105, 2039; b) J. M. Buriak, J. A. Osborn, Organometallics 1996,
15, 3161; c) J. M. Buriak, J. C. Klein, D. G. Herrington, J. A. Osborn, Chem.–Eur. J. 2000, 6, 139.
[16] W. Beck, K. Sunkel, Chem. Rev. 1988, 88, 1405; S. H. Strauss, Chem. Rev. 1993, 93, 927; I. Krossing, I.
Raabe, Angew. Chem., Int. Ed. 2004, 43, 2066.
[17] W. G. Bentrude, W. N. Setzer, in ‘31P-NMR Spectroscopy in Stereochemical Analysis’, Eds. J. G. Ver-
kade, L. D. Quin, VCH, 1987, pp. 365–389.
[18] K. Fagnon, M. Lautens, Angew. Chem., Int. Ed. 2002, 41, 26.
[19] S. Gladiali et al., to be published.
[20] W. C. Christopfel, B. D. Vineyard, J. Am. Chem. Soc. 1979, 101, 4406.
[21] J.-A. Ma, Angew. Chem., Int. Ed. 2003, 42, 4290; N. Sewald, Angew. Chem., Int. Ed. 2003, 42, 5794.

Helvetica Chimica Acta – Vol. 89 (2006)
1729
ˇ
[22] M. T. Reetz, T. Sell, A. Meiswinkel, G. Mehler, Angew. Chem., Int. Ed. 2003, 42, 790; D. Pena, A. J.
Minnard, J. A. F. Boogers, A. H. M de Vries, B. L. Feringa, Org. Biomol. Chem. 2003, 1, 1087; R.
Hoen, J. A. F. Boogers, H. Bernsmann, A. J. Minnard, A. Meetsma, T. D. Tiemersma-Wegman,
A. H. M. de Vries, J. G. de Vries, B. L. Feringa, Angew. Chem., Int. Ed. 2005, 44, 4209.
[23] a) M. Brookhart, B. Grant, A. F. Volpe, Organometallics 1992, 11, 3920; b) S. R. Bahr, P. Boudjouk,
J. Org. Chem. 1992, 57, 5545; c) B. Guzel, M. A. Omary, J. P. Fackler Jr., A. Akgerman, Inorg. Chim.
Acta 2001, 325, 45.
[24] R. R. Schrock, J. A. Osborn, J. Am. Chem. Soc. 1971, 93, 3089.
[25] W. D. Lubell, M. Kitamura, R. Noyori, Tetrahedron: Asymmetry 1991, 2, 543; G. Zhu, Z. Chen, X.
Zhang, J. Org. Chem. 1999, 64, 6907; J. You, H.-J. Drexler, S. Zhang, C. Fischer, D. Heller, Angew.
Chem., Int. Ed. 2003, 42, 913.
[26] N. Hashimoto, T. Aoyama, T. Shioiri, Chem. Pharm. Bull. 1981, 29, 1475.
[27] Y.-G. Zhou, W. Tang, W.-B. Wang, W. Li, X. Zhang, J. Am. Chem. Soc. 2002, 124, 4952.
Received March 29, 2006