Communications
to similar catalytic activity but lower enantioselectivity
(Table 1, entries 3–6). Considering the crucial role of the
additive in the asymmetric hydrogenation of heteroaromatic
compounds,[5] we also evaluated the effects of additives. For
example, only 28% conversion and 32% ee were obtained in
the absence of the iodine additive (Table 1, entry 7). Replac-
ing iodine with other additives gave lower ee values or even
racemic product.[11] With iodine as the additive and THF as
the solvent, further investigations focused on the effect of
hydrogen pressure and temperature. Increasing the hydrogen
pressure had no effect on the reactivity and enantioselectivity
(Table 1, entry 8), but decreasing the hydrogen pressure led to
a lower ee value (83%; Table 1, entry 9). Lowering the
reaction temperature resulted in a marked increase in
enantioselectivity, and the best ee value of 93% was obtained
at À58C (Table 1, entries 10 and 11). To the best of our
knowledge, this ee value represents the highest enantioselec-
tivity attained so far in the catalytic asymmetric hydrogena-
tion of 2-methylquinoxaline.[3]
Having established a highly enantioselective hydrogena-
tion of 1a, we turned to examine the catalyst loading (Table 1,
entries 12–16). Pleasingly, we were able to decrease the
catalyst loading to 0.005 mol% without any loss in enantio-
selectivity. Even with 0.005 mol% of Ir catalyst, the reaction
proceeded smoothly in only slightly lowered conversion and
with the same enantioselectivity (Table 1, entry 15). Remark-
ably, under the same reaction conditions, 1a was hydro-
genated in 1 h to give 28% conversion, providing a TON of
18140 and a TOF of 5620 hÀ1 (Table 1, entry 16). Notably, this
TOF value is the best result reported so far in the asymmetric
hydrogenation of heteroaromatic compounds.[3–9]
Table 1: Optimization of the reaction conditions for asymmetric hydro-
genation of 2-methyl-quinoxaline 1a.[a]
Entry Solvent H2 [psi]; T [8C]; S/C T [h] Conv. [%][b] ee [%][c]
1
THF
THF
700; 20; 100
700; 20; 100
20
20
20
20
20
20
20
1
100
100
100
100
100
100
28
89(S)
83(R)
70(S)
80(S)
87(S)
61(S)
38(S)
89(S)
83(S)
91(S)
93(S)
93(S)
93(S)
93(S)
93(S)
93(S)
2[d]
3
toluene 700; 20; 100
4
5
6
CH2Cl2
DCE
MeOH
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
700; 20; 100
700; 20; 100
700; 20; 100
700; 20; 100
1500; 20;100
300; 20; 100
700; 0; 100
700; À5; 100
700; À5; 1000
700; À5; 5000
700; À5; 10000
700; À5; 20000
700; À5; 20000
7[e]
8
100
100
9
1
10
11
12
13[f]
14[f]
15[g]
16[g]
1.5 100
1.5 100
20
20
20
20
1
100
100
100
91
28
[a] All reactions were carried out with 2-methylquinoxaline (0.15 mmol),
I2 (2 mol%), solvent (0.6 mL). [b] The conversion was determined by
1H NMR spectroscopy of the crude reaction mixture. [c] The enantio-
meric excess was determined by HPLC on a chiral stationary phase
according to previously reported methods.[3b] [d] (R)-SDPO was used as
the ligand. [e] Without I2 as an additive. [f] 0.3 mmol of 2-methylquinoxa-
line was used. [g] 0.6 mmol of 2-methylquinoxaline was used. DCE=1,2-
dichloroethane, S/C=substrate/catalyst molar ratio, THF=tetrahydro-
furan.
Next, we explored the scope of the iridium-catalyzed
asymmetric hydrogenation of substituted quinoxalines under
the optimized reaction conditions. The results are listed in
Table 2, which showed that all the substrates were smoothly
reduced with good enantioselectivities and full conversions
even at an S/C ratio of 5000. It was found that the reaction
system was sensitive to steric effect, and the presence of a less
sterically demanding alkyl group at the 2-position led to
better enantioselectivities (compare Table 2, entries 1–4 with
entry 5). The presence of substituents at the 6- and 7-positions
of the quinoxaline framework slightly lowered the enantio-
selectivity (Table 2, entries 6 and 7). Lower ee values were
also observed in the case of aryl-substituted quinoxalines
(Table 2, entries 8 and 9). In the hydrogenation of 2-styryl-
substituted quinoxalines, the substituents on the phenyl ring
showed no influence on the catalytic activity, but slightly
affected the enantioselectivity (Table 2, entries 10–14). The
best results were obtained with substrates bearing phenyl or
p-tolyl groups (Table 2, entries 10 and 13).
the excellent performance of the commercially available
chiral bidentate phosphine ligands in the iridium-catalyzed
asymmetric hydrogenation of quinolines,[5] lower enantiose-
lectivities of tetrahydroquinoxaline were observed when Ir/
binap (18% ee; binap = 2,2’-bis(diphenylphosphino)-1,1’-
binaphthyl),
Ir/MeO-biphep
(59% ee;
biphep = 2,2’-
bis(diphenylphosphino)-1,1’-biphenyl), Ir/P-Phos (49% ee;
P-Phos = 2,2’,6,6’-4,4’-bis(diphenylphosphino)-3,3’-bipyri-
dine), or Ir/synphos (77% ee; synphos = (5,6),-(5’,6’)-bis-
(ethylenedioxy)-2,2’-bis(diphenylphosphino)-1,1’-biphenyl)
were used. In addition, the other sterically demanding (R)-H8-
binapo derivatives bearing substituents on the 3- and 3’-
positions of the ligand framework resulted in much lower
enantioselectivities.[11]
The absolute configuration of the 2-substituted styryl
tetrahydroquinoxalines (2j) was determined to be S based on
single-crystal X-ray analysis of 4-N-tosyl-2-styryl-tetrahydro-
quinoxaline (2o; Scheme 1).[11] The configurations of the
other compounds are proposed by analogy.
With the Ir/H8-binapo catalyst in hand, a systematic study
of the hydrogenation of 1a was performed to establish the
optimum reaction conditions. The solvent effect was exam-
ined, and changing the solvent from THF to other solvents led
Finally, we applied this new protocol to the synthesis of
compound 2b, an inhibitor of cholesteryl ester transfer
protein,[1f] as an example of the quinoxaline class of biolog-
ically active compounds. Asymmetric hydrogenation of 2-
9136
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 9135 –9138