Organic Letters
Letter
Scheme 1. Regio- and Enantioselective Hydrogenation of
2,3-Double Bond of Conjugated Dienoic Acids
Table 1. Screening Additives for the Asymmetric
Hydrogenation of (2E,4E)-2-Methyl-5-phenylpenta- 2,4-
a
dienoic Acid 1a
b
c
entry
additive (equiv)
conv (%)
ee (%)
1
2
3
4
5
6
7
8
no
16
82
100
63
77
64
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
>99.9
80
Et3N (0.5)
Et3N (1.0)
Et3N (1.5)
DBU
TMG (1.0)
d
e
t-BuOK (1.0)
K2CO3 (0.5)
62
62
a
Reaction conditions: 0.2 mmol scale, [substrate] = 0.2 mol·L−1,
solvent = 1 mL, 1.0 mol % of catalyst ([Rh(NBD)2]BF4:
b
(SC,SC,RFc,RFc,RP,RP)-TriFer = 1:1.1). Determined by 1H NMR
c
of 2,3-double bond of conjugated α-substituted dienoic acids
(B, Scheme 1), despite the tremendous importance of chiral α-
substituted γ,δ-unsaturated acids in organic synthesis.
analysis. Determined by chiral HPLC analysis using a chiral column.
d
e
DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene. TMG = 1,1,3,3-
tetramethylguanidine.
Various ligands has been developed and applied to the
asymmetric hydrogenation of α-substituted α,β-unsturated
carboxylic acids.8 Previously, we developed TriFer and
ChenPhos and applied them in the Rh-catalyzed highly
efficient asymmetric hydrogenation of α-substituted cinnamic
acids8a−c and 2-substituted 2-alkenols.8d Importantly, Rh−
ChenPhos complex could complete differentiation between the
allylic alcohol and the nonfunctionalized double bond in the
asymmetric hydrogenation of geraniol.8d The key to our
success for TriFer and ChenPhos is due to the activation of
substrates through a secondary interaction (ionic bond or
hydrogen bonding) between the ligands and the substrates
(Figure 2). Encouraged by the previous work, we wondered
(100% regioselectivity and >99.9% ee), the catalytic activity
was disappointed, and only 16% conversion was achieved even
by prolonging the reaction time. To our delight, addition of
Et3N brought remarkable improvement of the catalytic activity
(entries 2−4). Both the catalytic activity and enantioselectivity
were significantly dependent upon the base and the amount of
base (entries 2−9). The best result was obtained using 1.0
equiv of Et3N as additive (full conversion, 100% regioselec-
tivity and >99.9% ee, entry 3). The absolute configuration of
2a was assigned (R)-configuration by comparison of its optical
rotation with the reported value.9 When (SC,SC,RFc,RFc,RP,RP)-
TriFer was replaced with (RC,SFc,SP)-ChenPhos, (2S,4E)-2-
methyl-5-phenylpent-4-enoic acid (S)-2a was quantitatively
obtained with 93% ee.
Next, the solvent effect for the hydrogenation was
investigated. The conversion decreased with the increase of
the bulkiness of the alcohols, although the entioselectivity
remained (Table 2, entries 1−3). MeOH with an appropriate
steric factor and pKa value outshined EtOH and i-PrOH. Polar
aprotic solvents such as EtOAc, MTBE, and THF lowered
both activity and enantioselectivity. Nonpolar solvent DCM
was also tested and showed less attractive activity.
With the optimized conditions in hand, we turned our
attention to explore the generality of our catalytic system in the
hydrogenation of various conjugated α-substituted dienoic
acids. The results are summarized in Table 3. A series of
conjugated α-substituted dienoic acids bearing different
substituent groups on the δ and α position of carboxyl group
were hydrogenated well to afford the corresponding chiral α-
substituted γ,δ-unsaturated acids in full conversion with perfect
regio- and enantioselectivity (100% regioselectivity and 97 →
99.9% ee). Importantly, the electronic properties and the steric
size of substituents R (various substituted aryl groups) on the δ
position and R1 (alkyl groups) at the α position had little
influence on the selectivities and activities. This catalyst system
could be applied to a broad scope of substrates.
Figure 2. Structures of TriFer and ChenPhos as well as plausible
substrate catalyst complex in the asymmetric hydrogenation.
whether the challenging regio- and enantioselective hydro-
genation of conjugated α-substituted dienoic acids could be
tackled by employing Rh−TriFer complex as catalyst.
Our investigation was initiated by using (2E,4E)-2-methyl-5-
phenylpenta-2,4-dienoic acid 1a as model substrate. Excellent
regio- and enatioselectivity were obtained with Rh−TriFer
complex, generated in situ from [Rh(NBD)2]BF4 (1.0 mol %;
NBD = 2,5-norbornadiene) and (SC,SC,RFc,RFc,RP,RP)-TriFer
(1.1 equiv with respect to Rh), under a presence of 30 atm of
H2 in MeOH at 20 °C for 24 h (Table 1, entry 1) (some well-
known diphosphine ligands, such as BPPFA, MandyPhos,
Taniaphos, Walphos, and Josiphos, did not show catalytic
Although the regio- and enatioselectivity were almost perfect
The high regioselectivity was confirmed by the further
hydrogenation of (2R,4E)-2-methyl-5-phenyl-4-pentenoic acid
B
Org. Lett. XXXX, XXX, XXX−XXX