Additive effects of amines on asymmetric hydrogenation of quinoxalines catalyzed by chiral iridium complexes

Article abstract of DOI:10.1002/chem.201201366

The additive effects of amines were realized in the asymmetric hydrogenation of 2-phenylquinoxaline, and its derivatives, catalyzed by chiral cationic dinuclear triply halide-bridged iridium complexes [{Ir(H)[diphosphine]} ₂(μ-X)₃]X (diphosphine=(S)-2,2'-bis(diphenylphosphino)- 1,1'-binaphthyl [(S)-BINAP], (S)-5,5'-bis(diphenylphosphino)-4,4'-bi-1,3- benzodioxole [(S)-SEGPHOS], (S)-5,5'-bis(diphenylphosphino)-2,2,2',2'- tetrafluoro-4,4'-bi-1,3-benzodioxole [(S)-DIFLUORPHOS]; X=Cl, Br, I) to produce the corresponding 2-aryl-1,2,3,4-tetrahydroquinoxalines. The additive effects of amines were investigated by solution dynamics studies of iridium complexes in the presence of N-methyl-p-anisidine (MPA), which was determined to be the best amine additive for achievement of a high enantioselectivity of (S)-2-phenyl-1,2,3,4-tetrahydroquinoxaline, and by labeling experiments, which revealed a plausible mechanism comprised of two cycles. One catalytic cycle was less active and less enantioselective; it involved the substrate-coordinated mononuclear complex [IrHCl₂(2-phenylquinoxaline){(S)-BINAP}], which afforded half-reduced product 3-phenyl-1,2-dihydroquinoxaline. A poorly enantioselective disproportionation of this half-reduced product afforded (S)-2-phenyl-1,2,3,4-tetrahydroquinoxaline. The other cycle involved a more active hydride-amide catalyst, derived from amine-coordinated mononuclear complex [IrCl₂H(MPA){(S)-BINAP}], which functioned to reduce 2-phenylquinoxaline to (S)-2-phenyl-1,2,3,4-tetrahydroquinoxaline with high enantioselectivity. Based on the proposed mechanism, an Ir^I-JOSIPHOS (JOSIPHOS=(R)-1-[(S_p)-2-(dicyclohexylphosphino)ferrocenylethyl] diphenylphosphine) catalyst in the presence of amine additive resulted in the highest enantioselectivity for the asymmetric hydrogenation of 2-phenylquinoxaline. Interestingly, the reaction rate and enantioselectivity were gradually increased during the reaction by a positive-feedback effect from the product amines. Copyright

Full text of DOI:10.1002/chem.201201366

FULL PAPER
DOI: 10.1002/chem.201201366
Additive Effects of Amines on Asymmetric Hydrogenation of Quinoxalines
Catalyzed by Chiral Iridium Complexes
Takuto Nagano,[a] Atsuhiro Iimuro,[a] Rino Schwenk,[b] Takashi Ohshima,[c]
Yusuke Kita,[a] Antonio Togni,[b] and Kazushi Mashima*[a]
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Abstract: The additive effects of
amines were realized in the asymmetric
hydrogenation of 2-phenylquinoxaline,
and its derivatives, catalyzed by chiral
cationic dinuclear triply halide-bridged
tive for achievement of a high enantio-
selectivity of (S)-2-phenyl-1,2,3,4-tetra-
hydroquinoxaline, and by labeling ex-
periments, which revealed a plausible
mechanism comprised of two cycles.
One catalytic cycle was less active and
less enantioselective; it involved the
dride–amide catalyst, derived from
amine-coordinated mononuclear com-
plex
[IrCl2HACHTNGUTERNNU(G MPA){(S)-BINAP}],
which functioned to reduce 2-phenyl-
quinoxaline to (S)-2-phenyl-1,2,3,4-tet-
rahydroquinoxaline with high enantio-
selectivity. Based on the proposed
mechanism, an IrI–JOSIPHOS (JOSI-
PHOS=(R)-1-[(Sp)-2-(dicyclohexyl-
phosphino)ferrocenylethyl]diphenyl-
phosphine) catalyst in the presence of
amine additive resulted in the highest
enantioselectivity for the asymmetric
hydrogenation of 2-phenylquinoxaline.
Interestingly, the reaction rate and
enantioselectivity were gradually in-
creased during the reaction by a posi-
tive-feedback effect from the product
amines.
iridium
complexes
[{Ir(H)[diphos-
phine]}2ACHTUNGTRENNUNG(m-X)3]X (diphosphine=(S)-
2,2’-bis(diphenylphosphino)-1,1’-bi-
substrate-coordinated
mononuclear
naphthyl [(S)-BINAP], (S)-5,5’-bis(di-
phenylphosphino)-4,4’-bi-1,3-benzo-
dioxole [(S)-SEGPHOS], (S)-5,5’-bis-
(diphenylphosphino)-2,2,2’,2’-tetra-
fluoro-4,4’-bi-1,3-benzodioxole [(S)-DI-
FLUORPHOS]; X=Cl, Br, I) to pro-
duce the corresponding 2-aryl-1,2,3,4-
tetrahydroquinoxalines. The additive
effects of amines were investigated by
solution dynamics studies of iridium
complexes in the presence of N-
methyl-p-anisidine (MPA), which was
determined to be the best amine addi-
complex [IrHCl2(2-phenylquinoxa-
line){(S)-BINAP}], which afforded
half-reduced product 3-phenyl-1,2-di-
hydroquinoxaline. A poorly enantiose-
lective disproportionation of this half-
reduced product afforded (S)-2-phenyl-
1,2,3,4-tetrahydroquinoxaline.
The
other cycle involved a more active hy-
Keywords: asymmetric catalysis
·
hydrogenation · iridium · nitrogen
heterocycles · reaction mechanisms
Introduction
roles,[24] imidazole,[25] oxazole,[25] and indoles,[26] have been
remarkable. However, only a few mechanistic studies of the
asymmetric hydrogenation of N-heteroaromatics have been
reported. Almost all of the proposed methods for hydroge-
nation of quinolines assume inner-sphere[27] rather than
outer-sphere coordination[4e,15] of the substrate. In the case
of quinoxalines, only one mechanistic study has been report-
ed; Zhou proposed the convergent asymmetric disproportio-
nation of dihydroquinoxalines induced by chiral phosphoric
acid in a Ru/Brønsted acid catalyzed asymmetric hydrogena-
tion of 2-aryl quinoxalines.[21]
A number of published reports document the dramatic
impact of additives on catalytic turnover and enantioselec-
tivity.[28] Additive effects are also crucial for asymmetric hy-
drogenation of N-heteroaromatics. For example, Zhou re-
ported that an iodine source used as an additive to activate
the catalyst for asymmetric hydrogenation of imine sub-
strates to give the corresponding chiral amines[29] enhanced
both the catalytic activity and enantioselectivity.[3a] Since
then, a rapidly increasing number of effective ligands have
been introduced by other groups.[3f,j,4f,g,9,10] Other ap-
proaches to activate the substrate have also been reported.
Zhou studied chloroformate as activators, which acted by
the formation of quinolinium and isoquinolinium salts in
situ; these salts were then smoothly hydrogenated.[3c] Ferin-
ga and co-workers reported that addition of piperidine hy-
drochloride improved the enantioselectivity in the asymmet-
ric hydrogenation of quinolines,[8] and Fan and Zhou inde-
pendently reported that Brønsted acids showed positive ef-
fects in the asymmetric hydrogenation of quinolines.[3i,4e,5b,c]
As part of our continued interest in the asymmetric hy-
drogenation of quinoxalines,[20] we discovered that the addi-
Catalytic asymmetric hydrogenation of prochiral unsaturat-
ed compounds, which include C=C, C=O, and C=N bonds,
has been intensively investigated as a highly versatile and
environmentally benign process for creation of a chiral
carbon center.[1] Asymmetric hydrogenation of N-heteroaro-
matic compounds is considered to be a difficult task due to
aromatic resonance stability. Recent developments in asym-
metric hydrogenation of N-heteroaromatic compounds,[2]
such as 2-substituted quinolines[3–15] and quinoxali-
nes[3h,k,4c,f,11,16–22] (to give the 1,2,3,4-tetrahydroquinoline and
1,2,3,4-tetrahydroquinoxaline derivatives, respectively, in
high enantioselectivity), pyridine derivatives,[3h,4g,23] pyr-
[a] T. Nagano, A. Iimuro, Dr. Y. Kita, Prof. Dr. K. Mashima
Department of Chemistry
Graduate School of Engineering Science
Osaka University, CREST
Toyonaka, Osaka 560-8631 (Japan)
Fax : (+81)6-6850-6245
E-mail: Mashima@chem.es.osaka-u.ac.jp
[b] R. Schwenk, Prof. Dr. A. Togni
Department of Chemistry and Applied Biosciences
Swiss Federal Institute of Technology
ETH Zꢁrich, 8093 Zꢁrich (Switzerland)
[c] Prof. Dr. T. Ohshima
Graduate School of Pharmaceutical Sciences
Kyushu University, CREST
Maidashi Higashi-ku, Fukuoka 812-8582 (Japan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201201366.
&
2
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Asymmetric Hydrogenation of Quinoxalines
FULL PAPER
Table 1. Additive effects of various amines in the asymmetric hydrogena-
tion of 4a.
tion of N-methylaniline derivatives enhanced not only the
catalytic activity but also the enantioselectivity in the gener-
ation
of
1,2,3,4-tetrahydroquinoxaline
derivatives
(Scheme 1). Although the additive effects of amines in
Entry Amine
pKa Conversion [%][a] ee of (S)-5a [%][b]
1
2
–
–
>99
>99
59
63
4.58
3
4
5
4.7
>99
74
72
74
63
5.61
2.57
>99
6
7
5.93
5.85
>99
>99
85
77
Scheme 1. Amine additive effects in the asymmetric hydrogenation of 2-
substituted quinoxalines 4 catalyzed by dinuclear iridium complexes 1–3.
8
9
10.7 no reaction
11.3 no reaction
–
–
asymmetric hydrogenation of C=N or C=C bonds were pre-
viously reported independently by the groups of Tani,[30]
Bartok,[31] and Sugimura,[32] the mechanisms of the additive
effect of the amine have not been clarified. The present
study reveals a dual mechanism that involves equilibrium
between two separate catalytic cycles, in which N-methylani-
line derivatives induce the evolution of the original catalyti-
cally active species to a more active and more enantioselec-
tive metal–amide catalyst generated through coordination of
the amine and elimination of HCl. Herein, we report an effi-
cient hydrogenation of quinoxalines in the presence of an
amine additive and our mechanistic studies on the effects of
the additive.
[a] Determined by 1H NMR analysis of the crude product. [b] Deter-
mined by HPLC analysis (Chiralcel OD-H column).
the reaction with N-methyl-p-anisidine (MPA) as an addi-
tive, enantioselectivity was dramatically improved to
85% ee with complete conversion of 4a (Table 1, entry 6).
Enantioselectivity was similarly enhanced by 4-methoxy-
N,N-dimethylaniline (Table 1, entry 7), but to a lesser extent
than by MPA. The pKa value of these amines was clearly
crucial for activation of the iridium catalyst. Aromatic
amines with a pKa around 5 not only enhanced the catalytic
activity but also improved enantioselectivity for the asym-
metric hydrogenation of 4a, whereas highly basic alkyl
amines with a pKa around 10 seemed to deactivate the iridi-
um species, possibly due to strong coordination to the iridi-
um center. These findings are consistent with reports that
simple pyridines and isoquinolines could not be efficiently
hydrogenated by similar iridium catalyst systems because
their respective hydrogenated cyclic amines would have a
pKa around 10, thus coordinate to, and deactivate, the cata-
lytically active species.[3c,23e,g] Consequently, we selected
MPA as the best amine additive for this reaction.
Next, we evaluated the relationship between the amount
of MPA and the enantioselectivity under the same reaction
conditions; the results are presented in Figure 1. At least
100 mol% of MPA was necessary to adequately enhance the
enantioselectivity (84% ee). Although use of 110 and
200 mol% of MPA afforded 85 and 87% ee, respectively, we
fixed the amount of MPA at 100 mol% for subsequent reac-
tions. These observations are consistent with the time de-
Results and Discussion
Effects of amine additives: We examined the additive effects
of alkyl and aryl amines (110 mol% relative to the sub-
strate) on the asymmetric hydrogenation of 4a promoted by
BINAP catalyst (S)-1a under standard reaction conditions
[(S)-1a (0.5 mol%), 30 atm, 308C, 20 h, 1,4-dioxane] and the
results are summarized in Table 1. Addition of aniline and
tetrahydroquinoline effectively improved the enantioselec-
tivity (63 and 72% ee, respectively; Table 1, entries 2 and 3),
whereas n-hexyl amine and piperidine completely stopped
the hydrogenation of 4a (Table 1, entries 8 and 9). Among
aniline and its derivatives, the effects of electron-poor 4-tri-
fluoromethylaniline did not differ from those of unsubstitut-
ed aniline (Table 1, entries 5 and 2), but electron-rich 4-me-
thoxyaniline slightly increased the enantioselectivity, al-
though the reaction rate was slower (Table 1, entry 4). In
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

K. Mashima et al.
than the analogous bromo- or iodo complexes, which indi-
cated that at least one halide atom was necessary to coordi-
nate to the iridium atom during the reaction. Notably, the
requirement for a halide atom bound to the iridium complex
is consistent with a report that halide-free iridium com-
plexes show low catalytic activity for asymmetric hydrogena-
tion of imines.[29] Such superiority of chloride complexes
contrasted with reports that chiral iodo-iridium complexes
that bear a chiral diphosphine ligand tend to be the best cat-
alysts among the series of halide complexes for asymmetric
hydrogenation of N-heteroaromatic compounds.[3a,g,j,4f,g,9,10]
Consequently, we selected (S)-3a combined with MPA
(100 mol%) as the best catalyst system.
Figure 1. Plot of the amount of additive amine (MPA) versus ee of (S)-5a
in the asymmetric hydrogenation of 4a [(S)-1a (0.5 mol%), H2 (30 atm),
308C, 20 h, 1,4-dioxane].
Under the optimized catalytic conditions [(S)-3a/MPA],
we conducted asymmetric hydrogenation reactions of vari-
ous 2-arylquinoxalines 4 and the results are summarized in
Table 3. Para-substituted phenylquinoxalines 4d and 4g–k,
pendence of the reaction, in which enantioselectivity in-
creased exponentially with the increasing amount of the
amine used (see below).
Table 3. Asymmetric hydrogenation of 2-arylquinoxalines 4 catalyzed by
(S)-3a/MPA.
Halide and diphosphine ligand effects in (S)-BINAP, (S)-
SEGPHOS, and (S)-DIFLUORPHOS complexes: In the
presence of the optimized amine additive MPA (100 mol%),
(S)-BINAP, (S)-SEGPHOS, and (S)-DIFLUORPHOS iridi-
um complexes (S)-1a–c, (S)-2a–c, and (S)-3a–c (Scheme 1)
with different halide ligands (chloride, bromide, and iodide)
were tested as catalysts for the hydrogenation of 4a at 308C
and the results are shown in Table 2. Among the tested atro-
pisomeric chiral chelating diphosphine ligands, (S)-DI-
FLUORPHOS complexes 3 were superior relative to the re-
spective (S)-BINAP and (S)-SEGPHOS complexes 1 and 2.
Chloro complexes (S)-1a, (S)-2a, and (S)-3a afforded the
hydrogenated product (S)-5a with better enantioselectivity
Entry
R
4
Conversion [%][a]
5
ee of (S)-5 [%][b]
1
2
C6H5
o-Me C6H4
4a
4b
>99
72
5a
5b
93
92
À
A
A
À
3
4
5
m-Me C6H4
p-Me C6H4
o-OMe C6H4
4c
4d
4e
>99
>99
69
5c
5d
5e
90
92
63
À
À
(95)[c]
(63)[c]
À
6
7
8
m-OMe C6H4 4 f
p-OMe C6H4
p-NO2 C6H4
p-F C6H4
p-Cl-C6H4
p-Br C6H4
m-Br C6H4
>99
>99
>99
>99
93
5 f
5g
5h
5i
5j
5k
5l
92
92
92
93
91
91
91
À
4g
4h
4i
4j
4k
4l
À
À
9
10
11
12
Table 2. Catalytic performance of chloro, bromo, and iodo iridium com-
plexes with (S)-BINAP, (S)-SEGPHOS, and (S)-DIFLUORPHOS li-
gands.
À
>99
>99
À
[a] Determined by 1H NMR analysis of the crude product. [b] Deter-
mined by HPLC analysis (Chiralcel OD-H column). [c] Reaction time
was 40 h.
which bear an electron-donating group, such as methyl or
methoxy, or an electron-withdrawing group, such as a halide
or nitro group, were efficiently hydrogenated to give tetra-
hydroquinoxaline derivatives (S)-5d and (S)-5g–k, respec-
tively, with high enantioselectivities (91–93% ee; Table 3,
entries 4 and 7–11). Furthermore, hydrogenation of meta-
substituted substrates 4c, 4 f, and 4l proceeded smoothly to
give (S)-5c, (S)-5 f, and (S)-5l, respectively, with high enan-
tioselectivity (90–92% ee; Table 3, entries 3, 6 and 12).
Steric bulkiness at the ortho-position affected the reactivity.
Ortho-methyl and ortho-methoxy substituents reduced the
reactivity such that a prolonged reaction time of 40 h was
required to complete the reactions (Table 3, entries 2 and 5).
Ortho-methyl-substituted substrate 4b resulted in 5b in high
Entry Diphosphine
X
Conversion [%][a] ee of (S)-5a [%][b]
1
2
3
Cl (3a) >99
Br (3b) >99
I (3c)
93
89
81
>99
4
5
6
Cl (2a) >99
Br (2b) >99
I (2c)
88
86
81
>99
7
8
9
Cl (1a) >99
Br (1b) >99
I (1c)
84
75
47
>99
[a] Determined by 1H NMR analysis of the crude product. [b] Deter-
mined by HPLC analysis (Chiralcel OD-H column).
&
4
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Asymmetric Hydrogenation of Quinoxalines
FULL PAPER
enantioselectivity (92% ee; Table 3, entry 2), whereas ortho-
methoxy-substituted substrate 4e remarkably reduced the
enantioselectivity to 63% ee (Table 3, entry 5), presumably
due to chelation of the methoxy group and the nitrogen
atom located at the 1-position of quinoxaline to the iridium
center.
of the dinuclear iridium complex (S)-1a with PA (2.1 equiv)
1
was monitored by H and 31P{1H} NMR spectroscopy, which
confirmed that PA coordinated to the iridium center to give
a single isomer of amine adduct (S)-1a-PA (Scheme 2). The
Mechanistic study: Our strong interest in the mechanism of
such amine additive effects in the asymmetric hydrogenation
of 4a led us to conduct a mechanistic study by using (S)-1a,
together with (S)-2a to isolate the intermediate species (due
to suitability of SEGPHOS for crystallization), with MPA
and p-anisidine (PA) as additive amines. The results are de-
scribed below and a plausible mechanism is illustrated in
Figure 2. Two separate catalytic cycles that afforded reduced
Scheme 2. Reaction of dinuclear iridium complexes with PA (2.1 equiv).
1H NMR spectrum in CD2Cl2 displayed a hydride signal at
d=À20.1 ppm (dd, J
ACHUTGTNREN(UNG H,PAHCTUNGTRNEN)UGN =14.2, 21.4 Hz); the coupling
constant values were consistent with a structure in which hy-
dride is cis to the two phosphorus atoms of the BINAP
ligand. The 31P{1H} NMR spectrum displayed a pair of dou-
blets at d=0.3 and À5.6 ppm (J
ACHTUNGERTN(NUNG P,P)=18.6 Hz). The amine
and methoxy protons of PA in (S)-1a-PA were observed at
d=4.7 (brs) and 3.6 ppm (s), respectively. Complex (S)-1a-
PA was isolated as yellow crystalline solid and further char-
acterized by elemental analysis and mass spectroscopy. No-
tably, the isolated complex (S)-1a-PA had very low catalytic
activity, presumably due to strong coordination of PA to the
iridium center and low solubility in the solvent, consistent
with the observed low catalytic activity (Table 1, entry 4).
Similar treatment of SEGPHOS complex (S)-2a with PA
afforded complex (S)-2a-PA, which was characterized by
Figure 2. Summary of the reaction mechanism of amine additive effects
on the asymmetric hydrogenation of 4a.
NMR spectroscopy [dH =À20.1 ppm (dd,
JACTHNGUTERNNU(G H,P)=13.9,
21.4 Hz); dP =À0.8 and À7.2 ppm (d, J(P,P)=19.1 Hz)], as
AHCTUNGTRENNUNG
well as elemental analysis and mass spectroscopy. Complex
(S)-2a-PA was crystallized from a saturated solution of
hexane and CH2Cl2. Figure 3 shows an ORTEP drawing of
(S)-2a-PA, in which the iridium atom adopts an octahedral
geometry. The hydride atom occupies a position cis to the
two phosphine atoms of SEGPHOS and the nitrogen atom
of PA, and the two other positions of the octahedral iridium
product 5a in low (cycle I) and high (cycle II) enantioselec-
tivity, respectively, were involved in this catalytic system,
and the two catalytic cycles were individually catalyzed by
two different active catalytic species, [(S)-1a-4a and (S)-1a-
MPA], respectively, which were in equilibrium with each
other throughout the catalytic reaction. The amount of
amine used changed the abundance ratio of the two catalyti-
cally active species and large amounts of amine were re-
quired to shift the reaction from cycle I to II to improve
enantioselectivity over the whole catalytic system. In the fol-
lowing mechanistic study, CD2Cl2 was used as the solvent
for the purpose of NMR spectroscopic measurements be-
cause the solubility of iridium precursors (S)-1 and (S)-2
was much better in CH2Cl2 than in 1,4-dioxane. The amine
additive effects for the asymmetric hydrogenation of 4a
were similar in CH2Cl2 and 1,4-dioxane.[33]
Figure 3. Crystal structure and numbering scheme of (S)-2a-PA. Selected
Reaction of (S)-1a and (S)-2a with amines and 4a: First, we
conducted controlled experiments for the addition of MPA
(the best additive amine), as well as for PA, (a moderately
effective amine additive) to a solution of iridium catalyst
(S)-1a in CD2Cl2 under an argon atmosphere. The reaction
À
À
À
bond lengths [ꢂ] and angles [8]: Ir H 1.38(2), Ir Cl(1) 2.435(5), Ir Cl(2)
À
À
À
2.499(5), Ir N 2.181(18), Ir P(1) 2.275(5), Ir P(2) 2.256(5); P(1)-Ir-P(2)
92.71(2), P(1)-Ir-H 83(1), P(2)-Ir-H 88(1), P(1)-Ir-N 169.48(5), P(2)-Ir-N
93.62(5), N-Ir-H 89(1). The angle between the best planes of the back-
bone aromatic carbons of (S)-SEGPHOS=66.028.
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

K. Mashima et al.
center are occupied by two chloride atoms. The bond length
with a pair of signals due to (S)-1a. Similarly, we observed
the formation of (S)-2a-MPA, and an equilibrium between
(S)-2a-MPA and (S)-2a. Because of the equilibrium, the
amine adducts (S)-1a-MPA and (S)-2a-MPA could not be
isolated, however, the starting complexes (S)-1a and (S)-2a
were precipitated.
À
À
of Ir P(1) (2.275(5) ꢂ) was 0.02 ꢂ longer than that of Ir
P(2) (2.256(5) ꢂ) due to the influence of the nitrogen atom
trans to P(1). The anisyl group of PA was oriented upward
toward the hydride atom, pointing to the less-bulky environ-
ment around the iridium center, and surrounded by the
edge and face phenyl groups of the SEGPHOS ligand. Ac-
cordingly, the absolute configuration of (S)-SEGPHOS com-
plex (S)-2a-PA was assigned as OC-6-42-C.
The coordination of PA and MPA to the dinuclear iridium
precursor (S)-1a to give the corresponding mononuclear ad-
ducts (S)-1a-PA and (S)-1a-MPA prompted us to investi-
gate the coordination of 4a to (S)-1a. Use of NMR spectro-
scopy to monitor the mixture of (S)-1a and 4a in CD2Cl2 at
À408C revealed two pairs of signals due to stereoisomers of
1
We then measured the H NMR spectrum of (S)-1a and
the best additive MPA (10 equiv) in CD2Cl2 solution in the
temperature range of À60 to 308C (Scheme 3 and Figure 4).
In the 1H NMR spectrum measured at À608C, the amine
adduct (S)-1a-MPA was characterized by a doublet of dou-
(S)-1a-4a [major: dH =À18.2 ppm (dd,
17.2 Hz); dP =À5.9 (d, J
(P,P)=19.4 Hz); minor: dH =À19.3 ppm (dd, J
21.7 Hz); dP =2.2 (d, J
(P,P)=21.2 Hz) and À4.3 ppm (d, J-
(P,P)=21.2 Hz)] (Scheme 4). The two isomers of the 4a-co-
JACHTUNGTRENNUNG
AHCTUNGTRENNUNG
blets at d=À19.9 ppm (J
N
J
N
ACHTUNGTRENNUNG
dride region, along with a doublet of doublets due to (S)-1a.
The chemical shift value and two coupling constants for (S)-
1a-MPA were in good accord with those of (S)-1a-PA,
which suggests that (S)-1a-MPA is a mononuclear octahe-
dral complex with an OC-6-42-C stereochemistry
(Scheme 3). Upon warming the solution to 308C, the dou-
blet of doublets due to (S)-1a-MPA broadened to give a
signal at d=À20.2 ppm (br) and the relative intensity of the
signal due to (S)-1a increased, which indicated an equilibri-
um between (S)-1a-MPA and (S)-1a with temperature-de-
pendent equilibrium constants that obeyed the van ꢃt Hoff
equation to give DH=À0.34(7) kJmolÀ1. In the
31P{1H} NMR spectrum at À608C, a pair of doublets at d=
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
ordinated iridium complex were assigned as C-(S)-1a-4a,
OC-6-42-C (major), and A-(S)-1a-4a, OC-6-42-A (minor).
Assignment of the major isomer was assumed based on the
crystal structure of (S)-2a-PA and that the coordination of
4a to the iridium center was reversible (see the Supporting
Information).[33] Equilibrium constants KC for (S)-1a and C-
(S)-1a-4a and KA for (S)-1a and A-(S)-1a-4a were applied
to the van ꢃt Hoff equation and the estimated enthalpies
were DHC =À0.67(4) kJmolÀ1 and DHA =À0.42(4) kJmolÀ1,
which indicated that the coordination of 4a was an endo-
thermic process.
3.7 and À3.7 ppm with JACHTUNGTRENUN(NG P,P)=17.9 Hz was observed, along
Scheme 3. Reaction of dinuclear iridium complex with MPA (10 equiv).
Scheme 4. Coordination of 4a to (S)-1a to form iridium complexes C-
(S)-1a-4a and A-(S)-1a-4a.
A competition experiment in which (S)-1a was treated
with 4a (2.1 equiv) and MPA (2.1 equiv) provided a mixture
of the two isomers of (S)-1a-4a together with a minor
amount of (S)-1a-MPA based on 1H NMR spectroscopic
measurements. This indicated that N-heteroaromatic com-
pound 4a coordinated more strongly to the iridium center
than MPA (Scheme 5). Accordingly, the difference in the co-
ordination abilities of 4a and MPA was precisely why such a
large amount (100 mol%) of MPA was needed to shift the
equilibrium in favor of (S)-1a-MPA to improve the enantio-
selectivity in cycle II (Figure 2).
Cycle I, catalyzed by (S)-1a-4a: When the asymmetric hy-
drogenation of 4a promoted by (S)-1a and additive MPA
(100 mol%) was stopped after a short reaction time (4 h),
Figure 4. 1H NMR spectra of the reaction mixture of (S)-1a and MPA
(10 equiv) in the temperature range À60–308C in CD2Cl2. Asterisk indi-
cates an impurity.
&
6
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Asymmetric Hydrogenation of Quinoxalines
FULL PAPER
proportionation, evidenced by
the fact that the same dispro-
portionation[21] occurred when
4a-1,2-H2 was treated with HCl
(Scheme 8b). In contrast, treat-
ment of 4a-1,2-H2 with (S)-1a
(10 mol%) in the presence of
MPA
(100 mol%)
induced
enantioselective disproportiona-
tion to give (S)-5a in 78% ee
(Scheme 8c), although the reac-
tion rate was slower (6 h) than
that of racemic disproportiona-
Scheme 5. Competition reaction of (S)-1a with MPA (2.1 equiv) and 4a (2.1 equiv).
hydrogenated product (S)-5a was obtained in low yield
(24%) and high enantioselectivity (84% ee). Furthermore,
at partial conversion of 4a, 3-phenyl-1,2-dihydroquinoxaline
(4a-1,2-H2) was detected in 41% yield (Scheme 6). When
Scheme 6. Asymmetric hydrogenation of 4a with short reaction times.
the hydrogenation reaction of 4a-1,2-H2 was performed
under D2 (10 atm), (S)-5a-d1 was obtained with the same
enantiomeric excess (78% ee) and the deuterium content at
the 2-position of (S)-5a-d1 was determined to be 74% based
1
on H NMR spectroscopic measurements (Scheme 7b). The
Scheme 7. Asymmetric hydrogenation and asymmetric deuteration of 4a-
1,2-H2 under various reaction conditions.
deuterium content at the 1-position (amine proton) could
not be determined due to scrambling and broadening of the
signal. This partial deuterium incorporation at the 2-position
was explained by the assumption of the existence of an
tion (<1 h). Notably, a small amount of (S)-1a (0.5 mol%)
catalyzed the disproportionation of 4a-1,2-H2 to give (S)-5a
enantioselectively (8% ee), even in the absence of MPA
(Scheme 8d), because the reduced product 5a has an amine
moiety that functions as an amine additive to prevent the
racemic disproportionation catalyzed by the acidic proton.
1H NMR spectroscopy revealed that 4a-1,2-H2 reacted with
(S)-1a to generate an iridium–hydride species, which
seemed to reduce 4a-1,2-H2 to afford (S)-5a by asymmetric
transfer hydrogenation from one molecule of 4a-1,2-H2 to
another. Some reduced N-heteroaromatic compounds are
reported to function as transfer hydrogenation reagents;
they easily generate hydrogen to reduce various unsaturated
bonds.[3k,34,35] Thus, we concluded that the additional amine
worked as a Brønsted base to capture acidic protons, which
retarded the rapid racemic disproportionation of 4a-1,2-H2
in catalytic cycle I.
À
extra hydrogen source, which was determined to be the N
H of the amine additive. In another controlled experiment
under D2 with deuterium-labeled MPA (MPA-d1) the deute-
rium content at the 2-position of (S)-5a-d1 increased to 87%
(Scheme 7c), which indicated that the amine proton of
MPA was involved in the catalytically active species as a hy-
dride ligand to be transferred to the C=N double bond of
4a-1,2-H2. Even under the latter conditions, 13% of the
proton at the 2-position of (S)-5a was not deuterated, most
likely because 4a-1,2-H2 also contains an amine proton.
We conducted further controlled experiments with 4a-1,2-
H2 in the absence of H2, outlined in Scheme 8. Surprisingly,
treatment of 4a-1,2-H2 with (S)-1a (10 mol%) in CD2Cl2 at
room temperature spontaneously gave a 1:1 mixture of 4a
and racemic 5a (Scheme 8a). A trace amount of acid de-
rived from (S)-1a, which was synthesized by the addition of
HCl to the IrI precursor (see Experimental), catalyzed dis-
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

K. Mashima et al.
ed in situ) to quinoxaline 4a and tetrahydroquinoxaline 5a
and, at the same time, the chiral iridium complex might me-
diate slow the asymmetric disproportionation of 4a-1,2-H2
by asymmetric transfer hydrogenation, which might be
mediated by the chiral iridium complex. Accordingly, in the
absence of amine additive, the reduction of 4a proceeded
mainly by the rapid racemic disproportionation pathway to
afford (S)-5a with low enantioselectivity. Accumulation of
amine product 5a as the reaction developed eliminated
acidic impurities, thereby the involvement of the racemic
pathway diminished (see below).
Cycle II, catalyzed by (S)-1a-MPA: In sharp contrast to the
asymmetric hydrogenation of 4a with (S)-1a under mild H2
pressure (10 atm) without any amine additive to give (S)-5a
in low conversion (40%) and low enantioselectivity
(11% ee), the same reaction in the presence of adequate
amounts of MPA (100 mol%) gave (S)-5a in full conversion
with high enantioselectivity (92% ee; Scheme 10), which
strongly indicated that the amine additive acted, not only as
a Brønsted base to prevent racemic reduction, but also as an
activator to improve catalytic activity under the hydrogena-
tion conditions.
Scheme 8. Disproportionation of 4a-1,2-H2 under various reaction condi-
tions.
Based on these experimental outcomes, we propose the
reduction mechanism of catalytic cycle I as shown in
Scheme 9 and suggest that this mechanism plays a role in
the catalytic cycle for the asymmetric hydrogenation of 4a
without any amine additives. The first step of cycle I was as-
sumed to be a reversible 1,2-insertion of the C=N bond of
Scheme 10. Asymmetric hydrogenation of 4a under mild H2 pressure.
À
4a into the Ir H bond of (S)-1a-4a to give (S)-6a, followed
À
by the addition of H2 to the Ir N bond to give (S)-7a, in
which 4a-1,2-H2 is coordinated to the iridium center. Be-
cause 4a-1,2-H2 is spontaneously replaced by 4a, (S)-1a-4a
is regenerated under catalytic conditions. In principle, all of
these steps, except for the replacement of 4a-1,2-H2 by 4a,
are reversible. The acidic proton derived from (S)-1a indu-
ces rapid racemic disproportionation of 4a-1,2-H2 (generat-
To reveal the activating effect of MPA under the hydroge-
nation conditions, we studied the controlled reaction of (S)-
1a and H2 gas in the presence and absence of MPA. Moni-
1
toring the H NMR spectrum of the BINAP–Ir complex (S)-
1a in CD2Cl2 in the absence of MPA under exposure to at-
mospheric H2 pressure for 20 h revealed no reaction. In
sharp contrast, the addition of MPA (10 equiv) to (S)-1a in
CD2Cl2 dramatically changed the reaction under H2 pressure
(1 atm). After 30 min, we observed a new doublet of dou-
blets centered at dH =À13.8 ppm (J
assigned to monohydride complex (S)-8a (Figure 5). The
two J(H,P) coupling constants suggested that a hydride oc-
ACHTUNGTREN(NUNG H,P)=16.6, 196.6 Hz)
AHCTUNGTRENNUNG
cupied the position trans to one phosphorus atom of the
BINAP ligand of (S)-8a and cis to the other phosphorus
atom, in sharp contrast to the general tendency that one hy-
Scheme 9. Proposed mechanism of catalytic cycle I involving dispropor-
tionation of 4a-1,2-H2.
Figure 5. Tentative structure of transient monohydride species (S)-8a.
&
8
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Asymmetric Hydrogenation of Quinoxalines
FULL PAPER
dride atom which is bound to iridium bearing a cis-chelating
diphosphine ligand occupies the position cis to both phos-
phorous atoms of the chelating diphosphine ligand.[37] Be-
cause the hydride complex (S)-8a was a transient species
that gradually decreased after 7 h and finally disappeared
after 33 h, its isolation and characterization were hampered.
However, (S)-8a was tentatively assigned as an amide–hy-
dride species generated by the elimination of HCl from (S)-
1a-MPA due to the quantitative formation of an MPA·HCl
salt.[34]
À22.4 ppm (m; terminal hydride) with the same 1:2 relative
intensity. Similar spectral data were obtained for anti-(S)-
10a–c and syn-(S)-10a–c.[33] Mass spectral data for (S)-9a
and (S)-10c confirmed that one of three bridging halide li-
gands was selectively replaced by one hydride ligand during
the course of the reaction. Figure 7 shows the structure of
anti-(S)-10c, in which two iridium centers are bridged by
two iodide ligands and one hydride and, although the posi-
tions of three hydride moieties could not be located, each
iridium center had a pseudo-octahedral and bifacial struc-
ture.
As the reaction proceeded, we observed two sets of sig-
nals assigned to two trihydride dinuclear complexes, anti-
(S)-9a and syn-(S)-9a, which appeared within 30 min and
gradually increased in intensity (Figure 6). We successfully
Figure 7. Crystal structure and numbering scheme of the cationic part of
À
anti-(S)-10c. Selected bond lengths [ꢂ] and angles [8]: Ir(1) Ir(2)
À
À
À
2.845(5), Ir(1) I(1) 2.781(7), Ir(1) I(2) 2.751(7), Ir(2) I(1) 2.744(6),
À
À
À
À
Ir(2) I(2) 2.800(7), Ir(1) P(1) 2.263(3), Ir(1) P(2) 2.285(2), Ir(2) P(3)
À
2.247(3), Ir(2) P(4) 2.278(3); P(1)-Ir(1)-P(2) 95.59(9), P(1)-Ir(1)-I(1)
1
Figure 6. Time-dependent H NMR spectra of hydride species in the reac-
97.43(7), P(1)-Ir(1)-I(2) 168.96(7), I(1)-Ir(1)-I(2) 82.92(2), P(2)-Ir(1)-I(2)
94.96(7), P(2)-Ir(1)-I(1) 105.16(7), P(3)-Ir(2)-P(4) 94.56(9), P(3)-Ir(2)-
I(1) 167.43(7), P(3)-Ir(2)-I(2) 94.89(7), I(1)-Ir(2)-I(2) 82.70(2), P(4)-Ir(2)-
I(1) 97.93(7), P(4)-Ir(2)-I(2) 107.22(7), Ir(1)-I(1)-Ir(2) 61.99(2), Ir(1)-
I(2)-Ir(2) 61.66(2).
tion of (S)-1a and MPA (10 equiv) under H2 (1 atm) at RT. Asterisk indi-
cates an unidentified species.
isolated and characterized (by spectral measurements and
X-ray crystal structural analysis) the trihydride dinuclear
complex anti-(S)-9a and also its analogue anti-(S)-10c,
which bears (S)-SEGPHOS and iodine ligands (Scheme 11).
A pair of hydride signals centered at dH =À11.1 (tt, J-
The mixture of anti- and syn-(S)-9a was determined to be
catalytically inert because the isolated mixture showed no
catalytic activity for the hydrogenation of 4a under our
standard reaction conditions (catalyst (0.5 mol%), H2
(30 atm), 20 h, 308C). Notably, the presence of 4a complete-
ly retarded the generation of (S)-9a, and reduction of 4a to
(S)-5a occurred preferentially.[33] Based on these results, (S)-
9a was excluded from the catalytic cycle. In addition, the
possibility of any dinuclear complex as a key intermediate
in determining the enantioselectivity was ruled out because
a nonlinear effect was not observed in the reaction; the
enantiopurity of complex (S)-1a and the enantioselectivity
of product (S)-5a had a linear relationship, indicative that a
mononuclear species was the active species.[34]
ACHTUNGTRENNUNG(H,H)=6.3 Hz, JACHTUNGTRENNUNG(H,P)=63.0 Hz; bridging hydride) and
À23.4 ppm (m; terminal hydride) in a 1:2 relative intensity
were determined to be produced by anti-(S)-9a. Hydride
signals of syn-(S)-9a were observed at dH =À11.6 (tt, J-
ACHTUNGTRENNUNG(H,H)=8.5 Hz, JACHTUNGTRENNUNG(H,P)=64.2 Hz; bridging hydride) and
These experimental outcomes led us to propose the re-
duction mechanism of catalytic cycle II, shown in
Scheme 12. When the quantity of MPA was large enough,
MPA coordinated to the iridium center to generate (S)-1a-
MPA, even in the presence of strongly coordinating sub-
strate 4a. Elimination of HCl from (S)-1a-MPA afforded
the amide–hydride species (S)-8a, which hydrogenated the
Scheme 11. Generation of trihydride dinuclear complexes.
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

K. Mashima et al.
2) in the stepwise outer-sphere mechanism, a coordinating
dihydrogen ligand is separated into proton and hydride
to reduce the substrate by sequential proton-transfer and
hydride-transfer reactions.[15]
3) in the inner-sphere mechanism, (S)-8a reacts with 4a
then 4a-1,2-H2 to give the product (S)-5a by 1,2-inser-
À
tion of the substrate into the Ir H bond, followed by
heterolytic cleavage of H2.[27]
Although we could not definitely conclude which of the
three mechanisms underlies the reaction, the formation of
dinuclear trihydride complexes (S)-9a suggests the presence
of amine–dihydride species (S)-11a (outer-sphere mecha-
nism), in which the amine is labile and the nascent dihydride
species might be trapped by (S)-1a or (S)-1a-MPA to give
the dinuclear trihydride complexes (S)-9a. Furthermore, the
assumed steric demand supports the outer-sphere mecha-
nism because the inner-sphere mechanism requires unfavor-
able coordination of sterically hindered substrate 4a-1,2-H2
through its nitrogen atom.[33] In addition, our basic reaction
conditions, which avoid racemic disproportionation, might
exclude the stepwise outer-sphere reaction pathway, which
involves protonation of substrates 4a and 4a-1,2-H2. Thus,
MPA acted not only as a base to eliminate HCl by abstrac-
tion of one of two chloride atoms bound to the iridium
center, but also as an amide ligand to generate a highly re-
active and enantioselective catalytically active species. Al-
though the iridium-catalyzed enantioselective disproportio-
nation of 4a-1,2-H2 might occur in cycle II, the disproportio-
nation is slow enough to be ignored (see Scheme 8 above).
Scheme 12. Proposed mechanism of catalytic cycle II involving amide–hy-
dride species (S)-8a and amine–dihydride species (S)-11a.
substrates 4a and 4a-1,2-H2. There are three possible mech-
anisms:
1) in the outer-sphere bifunctional mechanism, similar to
Noyoriꢄs metal–amide bifunctional mechanism,[37] (S)-8a
reacts with H2 to form a highly catalytically active
amine–dihydride species, (S)-11a, which hydrogenates 4a
and 4a-1,2-H2 to give the tetrahydrogenated product (S)-
5a.
Proposed entire reaction mechanism: Based on the results
of all of the controlled experiments described above, we
postulate the complete catalytic cycle for the asymmetric hy-
drogenation of 4a with (S)-1a as shown in Scheme 13, which
comprises pre-equilibrium among (S)-1a, 4a, and MPA, and
Scheme 13. Proposed complete catalytic cycle for the asymmetric hydrogenation of 4a.
&
10
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Asymmetric Hydrogenation of Quinoxalines
FULL PAPER
catalytic cycles I and II (the details of which are discussed
above). Thus, the key events are the reversible coordination
of substrate 4a (cycle I) and MPA (cycle II) to the iridium
dinuclear complex (S)-1a to give the corresponding mono-
nuclear complexes (S)-1a-4a (cycle I) and (S)-1a-MPA (cy-
cle II). Acidic protons derived from (S)-1a induce the fast
disproportionation of 4a-1,2-H2 to 4a and racemic 5a; at
the same time, the iridium complex (S)-1a induces slow
asymmetric disproportionation to afford (S)-5a, resulting in
low enantioselectivity at an early stage of the hydrogena-
tion. As the amount of product amine 5a increases, the
acid-dependent disproportionation is suppressed. In the
presence of excess amounts of MPA amide–hydride IrIII spe-
cies (S)-8a might be in equilibrium with (S)-1a-MPA, ac-
PHOS became the superior catalyst for asymmetric hydro-
genation of 4a under standard reaction conditions to give
(S)-5a with a 96% ee (Scheme 14). This result is consistent
with our proposed mechanism because the IrI–JOSIPHOS/
MPA system directly generates a highly catalytically active
mononuclear amide–hydride IrIII species analogous to (S)-
8a. The advantages of this system are: 1) no HCl elimina-
tion process, which eliminates hydrogen chloride mediated
racemic disproportionation of the half-reduced product 4a-
1,2-H2, and 2) no retro reaction of the amide–hydride com-
plex analogous to (S)-8a to form a bis-chloride amine–hy-
dride complex analogous to (S)-1a-MPA. Thus, under such
a catalytic system, only cycle II, which involves a highly
active amide–hydride species, spontaneously proceeded.
In addition, we successfully synthesized a chloride-bridged
iridium dinuclear complex [(R)-(S)-12a], with a coordinated
(R)-(S)-JOSIPHOS ligand, to confirm its catalytic activity in
the asymmetric hydrogenation of 4a with (100 mol%) or
without MPA. In absence of MPA, 25% ee was realized,
whereas the addition of MPA drastically increased the enan-
tioselectivity to 93% ee (Scheme 15). These results suggest-
ed that the hydrogenation catalyzed by the (R)-(S)-JOSI-
PHOS complex (R)-(S)-12a proceeded through the same re-
action mechanism as that catalyzed by (S)-1a, which in-
volved a disproportionation of 4a-1,2-H2.
À
companied by dissociation of the HCl salt of MPA. The Ir
N bond of (S)-8a reversibly reacts with H2 to give an
amine–dihydride species (S)-11a, which efficiently catalyzes
the asymmetric hydrogenation of the C=N bond of 4a-1,2-
H2 to give (S)-5a with high enantioselectivity (as well as the
analogous reaction of 4a to give 4a-1,2-H2) through the bi-
functional outer-sphere mechanism. At this stage, acid-cata-
lyzed racemic disproportionation is strongly inhibited by
amines present in the reaction mixture.
Direct generation of amide species: In the asymmetric hy-
drogenation of imines, quinolines, and quinoxalines cata-
lyzed by chiral iridium complexes, iodine is usually used as
an additive for the oxidation of an IrI precursor[3f] (e.g. [IrX-
ACHTUNGTRENNUNG(olefin)2]2/chiral chelating diphosphine ligand) to give a cat-
alyst system that involves an IrIII species. As revealed in the
above sections, the formation of IrIII–amide species proceed-
ed in the reaction of IrIII halide complexes with amines.
Therefore, we used MPA as a reagent to generate an IrIII–
I
Scheme 15. Asymmetric hydrogenation of 4a catalyzed by (R)-(S)-12a.
À
amide species by oxidative addition of the N H bond to Ir ,
but the direct reaction of IrI supported by BINAP, SEG-
PHOS, and DIFLUORPHOS ligands did not produce any
products identifiable by NMR spectroscopy. We then fo-
cused on the JOSIPHOS ligand (Scheme 14), the IrI com-
plex of which is reported to act as a catalyst for asymmetric
hydroamination of C=C double bonds by aniline derivatives
to.[38] In the case of iridium-catalyzed hydroamination, a
monomeric neutral amide–hydride IrIII species is the active
catalyst.[39] Unexpectedly, the addition of MPA to [IrCl-
Positive feedback from the product amine: Large amounts
of amine in the reaction mixture improved, not only the cat-
alytic activity, but also the enantioselectivity, therefore, we
anticipated that hydrogenated product 5a would show simi-
lar additive effects to generate the highly active amide–hy-
dride species described in Scheme 13. Addition of (S)-4-me-
thoxyphenyl-1,2,3,4-tetrahydroquinoxaline
[(S)-5b]
A
(100 mol%, 90% ee) or (R)-5b (100 mol%, 90% ee) to the
reagents for asymmetric hydrogenation of 4a with (S)-1a ef-
ficiently increased or decreased the enantioselectivity of (S)-
5a, respectively, dependent on the match or mismatch of the
stereochemistry to the chiral ligand of (S)-1a and the chiral
product 5b (Scheme 16). This match-and-mismatch effect
observed for the asymmetric hydrogenation of 4a strongly
suggested that the diastereomeric amine–hydride complex
was a key intermediate.
Encouraged by our observations, we added product 5a to
subsequent hydrogenation reactions of 4a. Thus, the first
asymmetric hydrogenation of 4a under H2 (30 atm) at 308C
for 20 h in 1,4-dioxane with (S)-1a (0.5 mol%) gave (S)-5a
in moderate ee (59%). The time dependence of the conver-
Scheme 14. Asymmetric hydrogenation of 4a catalyzed by IrI-Josiphos/
MPA system.
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
11
&
ÞÞ
These are not the final page numbers!

K. Mashima et al.
boxaldehyde was catalyzed by the addition of slightly enan-
tiomeric
enriched
2-methyl-1-(5-pyrimidyl)propan-1-ol,
which operated as the ligand of the zinc catalyst to give
highly enantiomerically enriched 2-methyl-1-(5-pyrimidyl)-
propan-1-ol upon hydrolysis, but, to the best of our knowl-
edge, our iridium catalyst system is the first example of a
transition-metal-catalyzed system.
Conclusion
We report the mild asymmetric hydrogenation of quinoxa-
lines catalyzed by chiral dinuclear triply halide-bridged iridi-
um complexes. An amine additive not only enhanced cata-
lytic activity but also the enantioselectivity. Our air-stable
precursors allow the reaction to be performed without ex-
haustive purification or special handling. Mechanistic studies
indicated a dual mechanism that involved two individual
catalytic cycles in equilibrium. Each of the proposed inter-
mediates in our catalytic cycle was either isolated or spec-
troscopically characterized and a series of stoichiometric re-
actions confirmed their role in the catalytic cycle. The key
species was amide–hydride species 8a, which demonstrated
high catalytic activity and hydrogenated substrate 4a to give
product 5a with high enantioselectivity. Based on the pro-
posed mechanism, the highest selectivity was achieved by
using the JOSIPHOS ligand. In addition, we revealed that
hydrogenated product 5a had similar additive effects, as
well as positive-feedback enhancement, in which the pres-
ence of a partially enantiomerically enriched hydrogenated
product 5a improved both the reaction rate and enantiose-
lectivity upon use in subsequent asymmetric reductions of
4a. This is the first report of positive-feedback enhancement
in asymmetric hydrogenation, rationalized by the proposed
dual mechanism. The new catalyst system of a halo-IrIII pre-
cursor combined with large amounts of amine clearly shows
the important role of the amine, not only as a ligand to gen-
erate the highly reactive and enantioselective iridium–amide
species, but also as a Brønsted base to bypass the acid-cata-
lyzed racemic disproportionation of half-reduced compound
4a-1,2-H2 in the iridium-catalyzed asymmetric hydrogena-
tion of 4a. Our results could lead to the development of
asymmetric hydrogenation of N-heteroaromatic compounds
and other C=N bonds under mild conditions. Further exten-
sion of this new catalyst system is ongoing.
Scheme 16. Additive effect of chiral product 5b and match-and-mismatch
effect.
sion of 4a and ee of (S)-5a (Figure 8a) plotted at 1, 3, 5,
and 7 h revealed that the yield of (S)-5a increased exponen-
tially over time, alongside a dramatic increase in enantiose-
lectivity. Surprisingly, the second reduction of 4a with (S)-
1a in the presence of (S)-5a (100 mol%, 59% ee)—the
product of the first asymmetric hydrogenation—afforded
(S)-5a with an enriched ee (65%); the net enantioselectivity
of the second hydrogenation of 4a was estimated to be
72% ee. We conducted a third asymmetric hydrogenation of
4a with (S)-1a in the presence of (S)-5a (100 mol%,
65% ee) and the reduction afforded (S)-5a with 70% ee
(the net enantioselectivity of the third reaction was estimat-
ed to be 75% ee; Figure 8b). This observation was consis-
tent with the mechanism outlined in Scheme 13 above. The
same enhancement phenomenon was observed for the re-
duction of 4a with SEGPHOS complex (S)-2a under the
same reaction conditions. Thus, we demonstrated positive
feedback of the product amine. Such enhanced positive
feedback to metal-assisted asymmetric reactions by a chiral
product was first reported by Soai (asymmetric autocataly-
sis);[40] the addition of diisopropylzinc to pyrimidine-5-car-
Experimental Section
General procedure for asymmetric hydrogenation of 2-arylquinoxalines
(4): A glass tube was charged with 4 (0.42 mmol), iridium complex
(2.1 mmol, 0.5 mol%), and amine additive (110 mol%) if required. The
glass tube was placed in an autoclave. After three cycles of evacuation/
argon backfilling, solvent (3.0 mL) was added to from the inlet, the mix-
ture was charged with H2, then the hydrogen pressure was increased
(30 atm). The reaction mixture was stirred at 308C for the indicated time.
After release of H2, the solvent was removed by evaporation. The con-
version and ee were determined by 1H NMR spectroscopy and HPLC
*
Figure 8. a) Time course for the conversion of 4a ( ) and enantiomeric
&
excess of (S)-5a ( ) in the asymmetric hydrogenation of 4a catalyzed by
Ir complex (S)-1a. b) Sequential reaction catalyzed by (S)-1a in the pres-
ence of enantioenriched (S)-5a (100 mol%) obtained from the previous
reaction (&=(S)-5a; &=(R)-5a).
&
12
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Asymmetric Hydrogenation of Quinoxalines
FULL PAPER
analysis of the crude product. The product was purified by filtration
through a short pad of silica gel (8:2 hexane/EtOAc), and analyzed by
spectroscopic methods.
na-Vidal, and Dr. T. Ayad (ENSCP, France) for their discussions regard-
ing the asymmetric hydrogenation of quinoxalines. T.N. expresses his spe-
cial thanks for the financial support provided by the JSPS Research Fel-
lowships for Young Scientists. A.I. appreciates the financial support sup-
plied by the Institutional Program for Young Researcher Overseas Visits
(JSPS) for his stay in ETH Zꢁrich.
Complex (S)-1a-PA: A mixture of (S)-1a (13 mg, 7.3 mmol) and p-me-
thoxyaniline (2.1 equiv, 1.9 mg, 15.3 mmol) in CH2Cl2 (0.6 mL) was stirred
for 1 h at RT to give a pale-brown solution. Addition of hexane afforded
(S)-1a-PA as a pink powder, which was recrystallized from Et2O/CH2Cl2.
M.p. 120–1308C (decomp.); 1H NMR (400 MHz, CD2Cl2, 308C): d=8.1–
7.2 (m, 20H; Ar), 7.0–6.2 (m, 16H; Ar), 3.6 (s, 3H; MeO), À20.1 ppm
[1] a) R. Noyori, Asymmetric Catalysis in Organic Synthesis, Wiley, New
York, 1994, Chapter 2, pp. 16–94; b) J. M. Brown, R. L. Halterman,
T. Ohkuma, R. Noyori, H.-U. Blaser, F. Spindler in Comprehensive
Asymmetric Catalysis, Vol. 1 (Eds.: E. N. Jacobsen, A. Pfaltz, H. Ya-
mamoto), Springer, Berlin, 1999, pp. 122–266; c) S. J. Roseblade, A.
Pfaltz, Acc. Chem. Res. 2007, 40, 1402–1411; d) G. Shang, W. Li, X.
Zhang in Catalytic Asymmetric Synthesis, 3rd ed. (Ed.: I. Ojima),
Wiley, Hoboken, 2010, Chapter 7, pp. 343–436; e) D. J. Ager, C. J.
Cobley, P. H. Moran, H.-U. Blaser, M. Lotz, F. Spindler, Y. Chi, W.
Tang, X. Zhang, S. H. L. Kok, T. T.-L. Au-Yeung, H. Y. Cheung,
W. S. Lam, S. S. Chan, A. S. C. Chan, M. van den Berg, B. L. Feringa,
A. J. Minnaard, A. Pfaltz, S. Bell, J. M. Brown, T. Ohkuma, R.
Noyori, A. Mortreux, A. Karim, A. J. Blacker, J. G. de Vries, L.
Lefort, M. Thommen in Handbook of Homogeneous Hydrogenation,
(Eds: J. G. de Vries, C. J. Elsevier), Wiley-VCH, Weinheim, 2007,
Part 4, pp. 745–1324; f) K. Pꢁntener, M. Scalone, W. Bonrath, R.
Karge, T. Netcher, F. Roessler F. Spindler in Asymmetric Catalysis
on Industrial Scale: Challenges, Approaches and Solutions, 2nd ed.
(Eds: H. U. Blaser, H. J. Federsel), Wiley-VCH, Weinheim, 2010.
Chapters 2 and 3, pp. 13–38.
(dd,
308C): d=0.3 (d, J
JACHTUNGTRENNUNG
(H,P)=14.2, 21.4 Hz, 1H; Ir-H); 31P NMR (161 MHz, CD2Cl2,
(P,P)=18.6 Hz), À5.6 ppm (d, J
ACHTUNGTREN(NUNG P,P)=18.6 Hz); ESI-
MS (MeOH, 1808C, 4.5 kV): m/z: calcd: 1008.17 [MÀH]+; found:
1008.15; elemental analysis calcd (%) for C51H42Cl2IrNOP2: C 60.65, H
4.19, N 1.39; found: C 60.95, H 3.72, N, 1.51.
Complex (S)-2a-PA: The title complex was synthesized from (S)-2a as
described for (S)-1a-PA above. M.p. 153–1628C (decomp.); 1H NMR
(400 MHz, CD2Cl2, 308C): d=7.9–7.1 (m, 20H; Ar), 6.5–5.7 (m, 8H; Ar),
5.3 (m, 4H; OCH2O), 4.4 (brs, 2H; NH2), 3.7 (s, 3H; MeO), À20.1 ppm
(dd, JACHTUNGTRENNUNG
(H,P)=13.9, 21.4 Hz, 1H; Ir-H); 31P{1H} NMR (100 MHz, CD2Cl2,
308C): d=À0.8 (d, J
(P,P)=19.2 Hz), À7.2 ppm (d, J
ACHTUNGTREN(NUNG P,P)=19.2 Hz); ESI-
Complex (S)-9a: A brown solution of (S)-1a (50 mg, 28 mmol) and MPA
(10 equiv, 39 mg, 0.28 mmol) in CH2Cl2 (1.0 mL) was exposed to atmos-
pheric pressure of H2 and stirred vigorously for 48 h. Addition of hexane
afforded anti-(S)-9a as yellow powder, which was recrystallized from
Et2O/CH2Cl2.
[2] For reviews, see: a) Y.-G. Zhou, Acc. Chem. Res. 2007, 40, 1357–
1366; b) D.-S. Wang, Q.-A. Chen, S.-M. Lu, Y.-G. Zhou, Chem. Rev.
2012, 112, 2557–2590.
Complex anti-(S)-9a (thermodynamic product): M.p. 175–1858C
(decomp.); 1H NMR (400 MHz, CD2Cl2, 308C): d=7.9–7.3 (m, 40H;
PPh2), 7.1–6.5 (m, 24H; Ar), À11.6 (tt, J
ACHUTNGTRENNUG(H,H)=8.5 Hz, JCAHTUNGTREN(NNGU H,P)=64.2 Hz,
[3] a) W.-B. Wang, S.-M. Lu, P. Y. Yang, X.-W. Han, Y.-G. Zhou, J. Am.
Chem. Soc. 2003, 125, 10536–10537; b) S.-M. Lu, X.-W. Han, Y.-G.
Zhou, Adv. Synth. Catal. 2004, 346, 909–912; c) S.-M. Lu, Y.-Q.
Wang, X.-W. Han, Y.-G. Zhou, Angew. Chem. 2006, 118, 2318–2321;
Angew. Chem. Int. Ed. 2006, 45, 2260–2263; d) W.-J. Tang, S.-F.
Zhu, L.-J. Xu, Q.-L. Zhou, Q.-H. Fan, H.-F. Zhou, K. Lam, A. S. C.
Chan, Chem. Commun. 2007, 613–615; e) X.-B. Wang, Y.-G. Zhou,
J. Org. Chem. 2008, 73, 5640–5642; f) D.-W.; Wang, D.-S. Wang, Q.-
A. Chen, Y.-G. Zhou, Chem. Eur. J. 2010, 16, 1133–1136 Wang, X.-
B. Wang, D.-S. Wang, S.-M. Lu, Y.-G. Zhou, Y.-X. Li, J. Org. Chem.
2009, 74, 2780–2787; g) D.-S. Wang, J. Zhou, D.-W. Wang, Y.-L.
Guo, Y.-G. Zhou, Tetrahedron Lett. 2010, 51, 525–528; h) W. Tang,
Y. Sun, L. Xu, T. Wang, Q. Fan, K.-H. Lam, A. S. C. Chan, Org.
Biomol. Chem. 2010, 8, 3464–3471; i) D.-S. Wang, Y.-G. Zhou, Tetra-
hedron Lett. 2010, 51, 3014–3017; j) D.-Y. Zhang, D.-S. Wang, M.-C.
Wang, C.-B. Yu, K. Gao, Y.-G. Zhou, Synthesis 2011, 2796–2802;
k) Q.-A. Chen, K. Gao, Y. Duan, Z.-S. Ye, Y. Yang, Y.-G. Zhou, J.
Am. Chem. Soc. 2012, 134, 2442–2448; l) D.-W.; Wang, D.-S. Wang,
Q.-A. Chen, Y.-G. Zhou, Chem. Eur. J. 2010, 16, 1133–1136.
[4] a) L. Xu, K. H. Lam, J. Li, J. Wu, Q.-H. Fan, W.-H. Lo, A. S. C.
Chan, Chem. Commun. 2005, 1390–1392; b) K. H. Lam, L. Xu, L.
Feng, Q.-H. Fan, F. L. Lam, W.-H. Lo, A. S. C. Chan, Adv. Synth.
Catal. 2005, 347, 1755–1758; c) L. Qiu, F. Y. Kwong, J. Wu, W. H.
Lam, S. Chan, W.-Y. Yu, Y.-M. Li, R. Guo, Z. Zhou, A. S. C. Chan,
J. Am. Chem. Soc. 2006, 128, 5955–5965; d) S. H. Chan, K. H. Lam,
Y.-M. Li, L. Xu, W. Tang, F. L. Lam, W. H. Lo, W. Y. Yiu, Q. Fan,
A. S. C. Chan, Tetrahedron: Asymmetry 2007, 18, 2625–2631; e) H.
Zhou, Z. Li, Z. Wang, T. Wang, L. Xu, Y. He, Q.-H. Fan, J. Pan, L.
Gu, A. S. C. Chan, Angew. Chem. 2008, 120, 8592–8595; Angew.
Chem. Int. Ed. 2008, 47, 8464–8467; f) W. Tang, L. Xu, Q.-H. Fan, J.
Wang, B. Fan, Z. Zhou, K. H. Lam, A. S. C. Chan, Angew. Chem.
2009, 121, 9299–9302; Angew. Chem. Int. Ed. 2009, 48, 9135–9138;
g) W.-J. Tang, J. Tan, L.-J. Xu, K. H. Lam, Q.-H. Fan, A. S. C. Chan,
Adv. Synth. Catal. 2010, 352, 1055–1062; h) T. Wang, L.-G. Zhuo, Z.
Li, F. Chen, Z. Ding, Y. He, Q.-H. Fan, J. Xiang, Z.-X. Yu, A. S. C.
Chan, J. Am. Chem. Soc. 2011, 133, 9878–9891.
1H; m-H), À22.4 ppm (m, 2H; Ir-H); 31P{1H} NMR (161 MHz, CD2Cl2,
308C): d=5.0 (br), 3.2 ppm (br); ESI-MS (MeOH, 1808C, 4.5 kV): m/z:
calcd: 1703.28 [MÀCl]+; found: 1703.28; elemental analysis calcd: (%)
for C88H67Cl3Ir2P4·2CH2Cl2: C 56.62, H 3.75; found: C 56.76, H 3.28.
Complex syn-(S)-9a (kinetic product): This complex could not be isolat-
ed. The aromatic proton signals are not described because the region was
observed as multiplet peak that overlapped with signals from anti-(S)-9a
and MPA. 1H NMR (400 MHz, CD2Cl2, 308C): d=À11.1 (tt, J
ACHTUNGTRNE(NUNG H,H)=
6.3 Hz,
J
(H,P)=63.0 Hz, 1H; m-H), À23.4 ppm (m, 2H; Ir-H);
31P{1H} NMR (161 MHz, CD2Cl2, 308C): d=6.0 (br), 2.0 ppm (br).
Complex (S)-10c: The title complex was synthesized and purified as de-
scribed above for (S)-9a.
Complex anti-(S)-10c (thermodynamic product): 1H NMR (400 MHz,
CD2Cl2, 308C): d=7.5–7.1 (m, 40H; Ar), 6.6 (dd, J
2H; Ar), 6.2 (d, J(H,H)=8.2 Hz, 2H; Ar), 6.0 (m, 2H), 5.8 (dd, J-
ACHTUNGTRENNUNG(H,H)=1.3, 41.1 Hz, 4H; OCH2O), 5.6 (dd, J (H,H)=8.2, 24.7 Hz, 4H;
ACHTUNGTREN(NUGN H,H)=1.7, 8.2 Hz,
ACHTUNGTRENNUNG
OCH2O), 5.2 (m, 2H) À13.5 (tt, J
ACHUTNGTRENNUG(H,H)=9.0 Hz, JHCATUNGTREN(NUGN H,P)=57.6 Hz, 1H;
m-H), À19.0 ppm (m, 2H; Ir-H); 31P{1H} NMR (161 MHz, CD2Cl2, 308C):
d=À4.1 (br), À5.1 ppm (t, J
A
4.5 kV): m/z: calcd: 1863.05 [MÀI]+; found: 1863.06; elemental analysis
calcd (%) for C76H59I3Ir2O8P4·3CH2Cl2: C 42. 28, H 2.92; found: C 42.41,
H 2.23.
Complex syn-(S)-10c (kinetic product): This complex could not be isolat-
ed. Aromatic proton region is not described because the region was ob-
served as multiplet peak overlapped with anti-(S)-10c and MPA in the re-
action mixture. 1H NMR (400 MHz, CD2Cl2, 308C): d=À13.6 (tt, J-
A
E
31P{1H} NMR (161 MHz, CD2Cl2, 308C): d=À10.9 (br), À12.2 ppm (br).
Acknowledgements
This work was supported by the Core Research for Evolutional Science
and Technology (CREST) Program of the Japan Science and Technology
Agency (JST), Japan. We thank Prof. J.-P. GenÞt, Dr. V. Ratovelomana-
[5] a) Z.-J. Wang, G.-J. Deng, Y. Li, Y.-M. He, W.-J. Tang, Q.-H. Fan,
Org. Lett. 2007, 9, 1243–1246; b) Z.-J. Wang, H.-F. Zhou, T.-L.
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
13
&
ÞÞ
These are not the final page numbers!

K. Mashima et al.
Wang, Y.-M. He, Q.-H. Fan, Green Chem. 2009, 11, 767–769; c) Z.-
W. Li, T.-L. Wang, Y.-M. He, Z.-J. Wang, Q.-H. Fan, J. Pan, L.-J. Xu,
Org. Lett. 2008, 10, 5265–5268; d) T. Wang, G. Ouyang, Y.-M. He,
Q.-H. Fan, Synlett 2011, 939–942.
16, 2036–2039; g) D.-S. Wang, J. Tang, Y.-G. Zhou, M.-W. Chen, C.-
B. Yu, Y. Duan, G.-F. Jiang, Chem. Sci. 2011, 2, 803–806; h) Y.
Duan, M.-W. Chen, Z.-S. Ye, D.-S. Wang, Q.-A. Chen, Y.-G. Zhou,
Chem. Eur. J. 2011, 17, 7193–7197; i) Y. Duan, M.-W. Chen, Q.-A.
Chen, C.-B. Yu, Y.-G. Zhou, Org. Biomol. Chem. 2012, 10, 1235–
1238.
[6] a) C. Deport, M. Buchotte, K. Abecassis, H. Tadaoka, T. Ayad, T.
Ohshima, J.-P. GenÞt, K. Mashima, V. Ratovelomanana-Vidal, Syn-
lett 2007, 2743–2747; b) H. Tadaoka, D. Cartigny, T. Nagano, T.
Gosavi, T. Ayad, J.-P. GenÞt, T. Ohshima, V. Ratovelomanana-Vidal,
K. Mashima, Chem. Eur. J. 2009, 15, 9990–9994.
[27] a) R.-H. Fish, A. D. Thormodsen, G. A. Cremer, J. Am. Chem. Soc.
1982, 104, 5234–5237; b) R. A. Sanchez-Delgado, D. Rondon, A.
Andriollo, V. Herrera, G. Martin, B. Chaudret, Organometallics
1993, 12, 4291–4296; c) M. Rosales, R. Vallejo, J. Soto, L. Bastidas,
K. Molina, P. Baricelli, Catal. Lett. 2010, 134, 56–62; d) C. Bianchi-
ni, P. Barbaro, M. Macchi, A. Meli, F. Vizza, Helv. Chim. Acta 2001,
84, 2895–2923.
[28] E. M. Vogl, H. Grçger, M. Shibasaki, Angew. Chem. 1999, 111,
1672–1680; Angew. Chem. Int. Ed. 1999, 38, 1570–1577.
[29] a) F. Spindler, B. Pugin, H.-U. Blaser, Angew. Chem. 1990, 102, 561–
562; Angew. Chem. Int. Ed. Engl. 1990, 29, 558–559; b) T. Morimo-
to, N. Nakajima, K. Achiwa, Chem. Pharm. Bull. 1994, 42, 1951–
1953; c) K. Satoh, M. Inenaga, K. Kanai, Tetrahedron: Asymmetry
1998, 9, 2657–2662; d) D. Xiao, X. Zhang, Angew. Chem. 2001, 113,
3533–3536; Angew. Chem. Int. Ed. 2001, 40, 3425–3428; e) H.-U.
Blaser, H.-P. Buser, R. Haeusel, H.-P. Jalett, F. Spindler, J. Organo-
met. Chem. 2001, 621, 34–38; f) B. Pugin, H. Landert, F. Spindler,
H.-U. Blaser, Adv. Synth. Catal. 2002, 344, 974–979; g) R. Dorta, D.
Broggini, R. Stoop, H. Ruegger, F. Spindler, A. Togni, Chem. Eur. J.
2004, 10, 267–278.
[7] M. T. Reetz, X. Li, Chem. Commun. 2006, 2159–2160.
ˇ ´
[8] N. Mrsic, L. Lefort, B. Laurent, A. F. Jeroen, A. J. Minnaard, B. L.
Feringa, J. G. de Vries, Adv. Synth. Catal. 2008, 350, 1081–1089.
[9] S.-M. Lu, C. Bolm, Adv. Synth. Catal. 2008, 350, 1101–1105.
[10] M. Eggenstein, A. Thomas, J. Theuerkauf, G. Francio, W. Leitner,
Adv. Synth. Catal. 2009, 351, 725–732.
[11] F.-R. Gou, W. Li, X. Zhang, Y.-M. Liang, Adv. Synth. Catal. 2010,
352, 2441–2444.
[12] M. Rubio, A. Pizzano, Molecules 2010, 15, 7732–7741.
[13] J. L. NfflÇez-Rico, H.-F. Fernµndez-PØrez, J. Benet-Buchholz, A.
Vidal-Ferran, Organometallics 2010, 29, 6627–6631.
[14] M. Rueping, R. M. Koenigs, Chem. Commun. 2011, 47, 304–306.
[15] G. E. Dobereiner, A. Nova, N. D. Schley, N. Hazari, S. J. Miller, O.
Eisenstein, R. H. Crabtree, J. Am. Chem. Soc. 2011, 133, 7547–7562.
[16] S. Murata, T. Sugimoto, S. Matsuura, Heterocycles 1987, 26, 763–
766.
[17] a) C. Bianchini, P. Barbaro, G. Scapacci, E. Farnetti, M. Graziani,
Organometallics 1998, 17, 3308–3310; b) C. Bianchini, P. Barbaro,
G. Scapacci, J. Organomet. Chem. 2001, 621, 26–33.
[18] a) C. J. Cobley, J. P. Henschke, Adv. Synth. Catal. 2003, 345, 195–
201; b) J. P. Henschke, M. J. Burk, C. G. Malan, D. Herzberg, J. A.
Peterson, A. J. Wildsmith, C. J. Cobley, G. Casy, Adv. Synth. Catal.
2003, 345, 300–307.
[30] K. Tani, J. Onouchi, T. Yamagata, Y. Kataoka, Chem. Lett. 1995, 24,
955–956.
[31] a) G. Szollosi, Z. Makra, M. Bartok, React. Kinet. Catal. Lett. 2009,
96, 319–325; b) Z. Makra, G. Szollosi, M. Bartok, Catal. Today
2012, 181, 56–61.
[32] T. Y. Kim, T. Sugimura, J. Mol. Catal. A Chem. 2010, 327, 58–62.
[33] See the Supporting Information.
ˇ ´
[19] N. Mrsic, T. Jerphagnon, A. Minnaard, B. L. Feringa, J. de Varies,
Adv. Synth. Catal. 2009, 351, 2549–2552.
[34] For a review on Hantzsch esters, see: S. G. Ouellet, A. M. Walji,
D. W. C. MacMillan, Acc. Chem. Res. 2007, 40, 1327–1339.
[35] a) M. Rueping, A. P. Antonchick, T. Theissmann, Angew. Chem.
2006, 118, 3765–3768; Angew. Chem. Int. Ed. 2006, 45, 3683–3686;
b) D.-W. Wang, W. Zheng, Y.-G. Zhou, Tetrahedron: Asymmetry
2007, 18, 1103–1107; c) Q.-S. Guo, D.-M. Du, J. Xu, Angew. Chem.
2008, 120, 771–774; Angew. Chem. Int. Ed. 2008, 47, 759–762; d) M.
Rueping, F. Tato, F. R. Schoepke, Chem. Eur. J. 2010, 16, 2688–
2691; e) M. Rueping, T. Theissmann, M. Stoeckel, A. P. Antonchick,
Org. Biomol. Chem. 2011, 9, 6844–6850; f) T. B. Nguyen, H. Bous-
serouel, Q. Wang, F. GuØritte, Adv. Synth. Catal. 2011, 353, 257–
262; g) T. B. Nguyen, Q. Wang, F. GuØritte, Chem. Eur. J. 2011, 17,
9576–9580; h) Y. Zhao, Q. Liu, J. Li, B. Zhou, Synlett 2010, 1870–
1872.
[36] a) M. K. Hays, R. Eisenberg, Inorg. Chem. 1991, 30, 2623–2630;
b) H. E. Selnau, J. S. Merola, Organometallics 1993, 12, 3800–3801;
c) J. D. Feldman, J. C. Peters, T. D. Tilley, Organometallics 2002, 21,
4050–4064; d) D. Carmona, J. Ferrer, M. Lorenzo, M. Santander, S.
Ponz, F. J. Lahoz, J. A. López, L. A. Oro, Chem. Commun. 2002,
870–871; e) C. Tejel, M. A. Ciriano, S. JimØnez, L. A. Oro, C.
Graiff, A. Tiripicchio, Organometallics 2005, 24, 1105–1111; f) T.
Hara, T. Yamagata, K. Mashima, Y. Kataoka, Organometallics 2007,
26, 110–118; g) M. Liniger, B. Gschwend, M. Neuburger, S. Schaff-
ner, A. Pfaltz, Organometallics 2010, 29, 5953–5958.
[20] a) D. Cartigny, T. Nagano, T. Ayad, J.-P. GenÞt, T. Ohshima, K. Ma-
shima, V. Ratovelomanana-Vidal, Adv. Synth. Catal. 2010, 352,
1886–1891; b) D. Cartigny, F. Berhal, T. Nagano, P. Phansavath, T.
Ayad, J.-P. GenÞt, T. Ohshima, K. Mashima, V. Ratovelomanana-
Vidal, J. Org. Chem. 2012, 77, 4544–4556.
[21] Q.-A. Chen, D.-S. Wang, Y.-G. Zhou, Y. Duan, H.-J. Fan, Y. Yang,
Z. Zhang, J. Am. Chem. Soc. 2011, 133, 6126–6129.
[22] J. Qin, Z. Ding, Y.-M. He, L. Xu, Q.-H. Fan, Org. Lett. 2011, 13,
6568–6571.
[23] a) M. Studer, C. Wedemeyer-Exl, F. Spindler, H. U. Blaser, Monatsh.
Chem. 2000, 131, 1335–1343; b) L. Hegedꢁs, V. Hµda, A. Tungler, T.
MµthØ, L. Szepesy, Appl. Catal. A 2000, 201, 107–114; c) N. Douja,
M. Besson, P. Gallezot, C. Pinel, J. Mol. Catal. A Chem. 2002, 186,
145–151; d) N. Douja, R. Malacea, M. Banciu, M. Besson, C. Pinel,
Tetrahedron Lett. 2003, 44, 6991–6993; e) F. Glorius, N. Spielkamp,
S. Holle, R. Goddard, C. W. Lehmann, Angew. Chem. 2004, 116,
2910–2913; Angew. Chem. Int. Ed. 2004, 43, 2850–2852; f) C. Y. Le-
gault, A. B. Charette, J. Am. Chem. Soc. 2005, 127, 8966–8967;
g) A. W. Lei, M. Chen, M. S. He, X. Zhang, Eur. J. Org. Chem. 2006,
4343–4347; h) X.-B. Wang, W. Zeng, Y.-G. Zhou, Tetrahedron Lett.
2008, 49, 4922–4924.
[24] a) R. Kuwano, M. Kashiwabara, M. Ohsumi, H. Kusano, J. Am.
Chem. Soc. 2007, 129, 3802; b) R. Kuwano, Heterocycles 2008, 76,
909–922; c) D.-S. Wang, Z.-S. Ye, Q.-A. Chen, Y.-G. Zhou, C.-B.
Yu, H.-J. Fan, Y. Duan, J. Am. Chem. Soc. 2011, 133, 8866–8869.
[25] R. Kuwano, N. Kameyama, R. Ikeda, J. Am. Chem. Soc. 2011, 133,
7312–7315.
[37] a) R. Noyori, S. Hashiguchi, Acc. Chem. Res. 1997, 30, 97–102;
b) K.-J. Haack, S. Hashiguchi, A. Fujii, T. Ikariya, R. Noyori,
Angew. Chem. 1997, 109, 297–300; Angew. Chem. Int. Ed. Engl.
1997, 36, 285–288; c) M. Yamakawa, H. Ito, R. Noyori, J. Am.
Chem. Soc. 2000, 122, 1466–1478; d) R. Noyori, T. Ohkuma, Angew.
Chem. 2001, 113, 40–75; Angew. Chem. Int. Ed. 2001, 40, 40–73;
e) C. A. Sandoval, T. Ohkuma, K. Muniz, R. Noyori, J. Am. Chem.
Soc. 2003, 125, 13490–13503; f) S. Gladiali, E. Alberico, Chem. Soc.
Rev. 2006, 35, 226–236.
[26] a) R. Kuwano, K. Kaneda, T. Ito, K. Sato, T. Kurokawa, Y. Ito, Org.
Lett. 2004, 6, 2213–2215; b) R. Kuwano, M. Kashiwabara, K. Sato,
T. Ito, K. Kaneda, Y. Ito, Tetrahedron: Asymmetry 2006, 17, 521–
535; c) A. M. Maj, I. Suisse, C. MØliet, F. Agbossou-Niedercorn, Tet-
ˇ
rahedron: Asymmetry 2010, 21, 2010–2014; d) N. MrSíc, T. Jerphag-
non, A. J. Minnaard, B. L. Feringa, J. G. de Vries, Tetrahedron:
Asymmetry 2010, 21, 7–10; e) R. Kuwano, M. Kashiwabara, Org.
Lett. 2006, 8, 2653–2655; f) A. Baeza, A. Pfaltz, Chem. Eur. J. 2010,
[38] R. Dorta, P. Egli, F. Zꢁrcher, A. Togni, J. Am. Chem. Soc. 1997, 119,
10857–10858.
&
14
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Asymmetric Hydrogenation of Quinoxalines
FULL PAPER
[39] a) A. L. Casalnuovo, J. C. Calabrese, D. Milstein, J. Am. Chem. Soc.
1988, 110, 6738–6744; b) J. Zhou, J. F. Hartwig, J. Am. Chem. Soc.
2008, 130, 12220–12221.
584–589; d) Amplification of Chirality (Ed.: K. Soai), Springer-
Verlag, 2008; e) L. Schiaffino, G. Ercolani, Angew. Chem. 2008, 120,
6938–6941; Angew. Chem. Int. Ed. 2008, 47, 6832–6835.
[40] a) K. Soai, T. Shibata, H. Morioka, K. Choji, Nature 1995, 378, 767–
768; b) D. K. Kondepudi, K. Asakura, Acc. Chem. Res. 2001, 34,
946–954; c) D. G. Blackmond, Tetrahedron: Asymmetry 2006, 17,
Received: April 21, 2012
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
15
&
ÞÞ
These are not the final page numbers!

K. Mashima et al.
Hydrogenation
T. Nagano, A. Iimuro, R. Schwenk,
T. Ohshima, Y. Kita, A. Togni,
K. Mashima* .................. &&&& — &&&&
Additive Effects of Amines on Asym-
metric Hydrogenation of Quinoxalines
Catalyzed by Chiral Iridium
Complexes
Additive effects were investigated in
the asymmetric hydrogenation of 2-
substituted quinoxalines catalyzed by a
chiral iridium complex (see scheme).
Catalytic activity and enantioselectivity
were improved by the addition of an
aniline derivatives and product amine.
Asymmetric Catalysis
Additive effects of amines were investigated in asym-
metric hydrogenation of 2-substituted quinoxalines cata-
lyzed by an air-stable chiral dinuclear triply-halide-bridged
iridium complex. Catalytic reactivity and enantioselectivity
were dramatically improved by an addition of amine. A
mechanistic study revealed a dual mechanism that
involved two individual catalytic cycles in equilibrium. The
product amine itself could enhance catalytic activity and
enantioselectivity during the reaction course, demon-
strating the first positive feedback enhancement in asym-
metric hydrogenation. For more information see the Full
Paper by K. Mashima et al. on page && ff.
&
16
&
www.chemeurj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!