Imine hydrogenation catalyzed by iridium complexes comprising monodentate chiral phosphoramidites and N-donor ligands

Article abstract of DOI:10.1016/j.jorganchem.2006.08.032

The relatively inexpensive chiral monodentate phosphoramidite (S)-MONOPHOS may be used in combination with pyridines to prepare iridium complexes effective for catalysis of asymmetric imine hydrogenation with comparable enantioselectivity to some of those containing more costly chiral bidentate phosphines. [Ir(cod)((S)-MONOPHOS)(L)]BArF (cod = 1,5-cyclooctadiene; L = 3-methylisoquinoline, acridine, 2,6-lutidine, acetonitrile, or 2,3,3-trimethylindolenine; BArF = tetrakis[3,5-bis(trifluoromethyl)phenyl]borate) are efficient catalysts for the asymmetric hydrogenation of 2,3,3-trimethylindolenine. An important observation is that the catalyst containing acridine is more enantioselective than the catalyst derived from 2,3,3-trimethylindolenine which suggests that the other N-donor ligands are not readily displaced by the substrate during the catalytic cycle.

Full text of DOI:10.1016/j.jorganchem.2006.08.032

Journal of Organometallic Chemistry 691 (2006) 4945–4955
www.elsevier.com/locate/jorganchem
Imine hydrogenation catalyzed by iridium complexes comprising
monodentate chiral phosphoramidites and N-donor ligands
*
J.W. Faller , Suzanna C. Milheiro, Jonathan Parr
Department of Chemistry, Yale University, P.O. Box 208107, New Haven, CT 06520-8107, United States
Received 2 June 2006; received in revised form 15 August 2006; accepted 15 August 2006
Available online 24 August 2006
Abstract
The relatively inexpensive chiral monodentate phosphoramidite (S)-MONOPHOS may be used in combination with pyridines to pre-
pare iridium complexes effective for catalysis of asymmetric imine hydrogenation with comparable enantioselectivity to some of those
containing more costly chiral bidentate phosphines. [Ir(cod)((S)-MONOPHOS)(L)]BArF (cod = 1,5-cyclooctadiene; L = 3-methyliso-
quinoline, acridine, 2,6-lutidine, acetonitrile, or 2,3,3-trimethylindolenine; BArF = tetrakis[3,5-bis(trifluoromethyl)phenyl]borate) are
efficient catalysts for the asymmetric hydrogenation of 2,3,3-trimethylindolenine. An important observation is that the catalyst contain-
ing acridine is more enantioselective than the catalyst derived from 2,3,3-trimethylindolenine which suggests that the other N-donor
ligands are not readily displaced by the substrate during the catalytic cycle.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Hydrogenation; Catalysis; Imine; Asymmetric catalysis
1. Introduction
moderately successful for acyclic imines, but suffered from
low catalyst activity [10,11].
The asymmetric hydrogenation of imines is an impor-
tant route to a-chiral amines [1]. Although there are many
examples of highly enantioselective catalysts for olefin and
ketone reduction, asymmetric imine hydrogenation is still a
challenge in terms of both the turnover frequency and the
lifespan of the active catalyst. This is due to the fact that
C–N double bonds have certain traits, such as their pre-
ferred mode of binding and the strong donor character of
the nitrogen, that are unfavorable for homogeneous cata-
lytic hydrogenation [2,3]. Although early examples of
asymmetric imine hydrogenation were achieved with poor
enantioselectivity [4], some success has been achieved with
rhodium [5–8] and ruthenium [9] catalysts with chiral bis-
phosphine ligands. Another notable system is Buchwald’s
ansa-titanocene catalyst that was particularly successful
for the asymmetric hydrogenation of cyclic imines, and
Recently chiral bisphosphine iridium catalysts have
shown greater success for asymmetric imine hydrogenation
[12–18], although they often require the presence of halide
ions or other additives such as amines and acids. Most of
these systems are not applicable to a wide range of sub-
strates, and require high pressures, long reaction times,
and stoichiometric amounts of chiral material or additives
that tend to be substrate specific. Most of the effective cat-
alysts comprise chiral chelating bisphosphine ligands and it
is to these ligands that the efficacy of the catalysts is usually
ascribed. There are, however, examples that do not have
ligands of this general class, such as the iridium complexes
of chiral bidentate [P,N] ligands, originally reported by
Pfaltz and co-workers [19–22] and the chiral tridentate
[P,N,P] ligands reported by Sablong and Osborn [23].
Crabtree’s catalyst, [Ir(cod)(py)(PCy3)]X is particularly
effective for olefin hydrogenations [24,25] and recently,
[Ir(cod)(py)(PBz3)]PF6 was shown to be more efficient than
[Ir(cod)(PBz3)2]PF6 for the hydrogenation of an imine
derived from aniline [26]. Furthermore, chiral monodentate
*
Corresponding author. Fax: +1 203 432 6144.
E-mail address: jack.faller@yale.edu (J.W. Faller).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.08.032

4946
J.W. Faller et al. / Journal of Organometallic Chemistry 691 (2006) 4945–4955
phosphoramidites have also proved to be effective ligands for
enantioselective hydrogenations of olefins [27]. Based on
these considerations, we set out to determine whether biden-
tate ligands were really required, and if a chiral catalyst anal-
ogousto Crabtree’s catalyst employing a monodentate chiral
phosphorus donor could act as an effective enantioselective
imine hydrogenation catalyst. We chose to test these systems
on 2,3,3-trimethylindolenine (tmi), an imine substrate which
has been shown to be difficult to hydrogenate with good
enantioselectivity in the past [8,14,23].
Although the use of a binaphthol based monodentate
chiral P-donor ligand has been reported previously for
imine hydrogenation, no enantioselectivity was observed
[28]. We report here rare examples of an iridium catalyst
containing a chiral monodentate phosphoramidite and var-
ious N-donor ligands for asymmetric imine hydrogenation.
The commercially available, inexpensive chiral monoden-
tate phosphoramidite (S)-MONOPHOS may be used in
combination with N-donor ligands to form the complexes
[Ir(cod)((S)-MONOPHOS)(L)]BArF (cod = 1,5-cyclooct-
adiene; L = 3-methylisoquinoline, acridine, 2,6-lutidine,
acetonitrile, 2,3,3-trimethylindolenine; BArF = tetrakis[3,5-
bis(trifluoromethyl)phenyl]borate) which are efficient cata-
lysts for the asymmetric hydrogenation of 2,3,3-trimethy-
lindolenine (Scheme 1).
ilar monodentate ligands adopt a cis-configuration.
Complexesꢀ used were of the general formula [Ir(cod)-
(P*)(L)]+X where (P*) is a chiral monodentate phos-
phoramidite, (L) is a nitrogen-donor ligand and Xꢀ is a
non-coordinating anion, as shown in Scheme 2 for
MONOPHOS.
An important consideration with complexes of this type
in comparison to those involving chelates is their propen-
sity for disproportionation. The potential for ligand dis-
placement is also an important factor in the stability of
the intermediates that are formed during the catalytic cycle.
This would be expected to be a complication in imine
hydrogenation wherein the imine substrate might displace
some of the other ligands. Furthermore, whereas the spe-
cies involved in rhodium-catalyzed hydrogenations are
fairly well-understood [29–32], the case for iridium-cata-
lyzed hydrogenations is more nebulous. In the case of rho-
dium catalysis, cycles involve Rh(I)/Rh(III) cycles wherein
a four-coordinate Rh(I) species is converted to the Rh(III)
dihydride. Recent theoretical calculations for P,N and car-
bene–C,N iridium systems have suggested catalytic cycles
involving an Ir(III)/Ir(V) pathway with pseudo-octahedral
or octahedral intermediates [33–35]. Generally one would
anticipate that cis-dihydrides would be formed in the Ir(III)
species, but the disposition of the other ligands is not clear.
Our efforts here are not aimed at resolving the details of the
mechanism, but to suggest that the higher oxidation state
complexes of iridium should resist ligand exchange and
thus provide some rationale for the notion that the phos-
phoramidite and the pyridine ligands initially present
2. Results and discussion
The systems studied are related to complexes of
heterobidentate ligands, particularly when the two dissim-
1% cat, 40 bar H2
rt, 24 h, neat
CH3
H
H
CH3
+
N
H
N
H
N
tmi
(R)
(S)
Scheme 1. The hydrogenation of tmi.
X
L
Me N
O
2
NaBArF or AgSbF6
Me N
O
O
2
P
O
P
CH2Cl2
Ir
Ir
Cl
L
L =
N
N
N
N
X= BArF, SbF6
Scheme 2. Preparation of [Ir(cod)((S)-MONOPHOS)(L)]X.

J.W. Faller et al. / Journal of Organometallic Chemistry 691 (2006) 4945–4955
4947
may be retained on the iridium complex during the cata-
lytic cycle.
2.1. Properties of [Ir(cod)((S)-MONOPHOS)(L)]+
[Ir(cod)((S)-MONOPHOS)Cl], prepared from [Ir(cod)-
Cl]2 and (S)-MONOPHOS, was used to prepare
[Ir(cod)((S)-MONOPHOS)(L)]BArF by combination with
the appropriate N-donor ligand and NaBArF. Character-
ization, however, was more readily carried out with SbF6
complexes. With the C2 symmetric acridine, 2,6-lutidine,
and acetonitrile ligands only one isomer is possible. The
X-ray structure of the 2,6-lutidine complex is shown in
Fig. 1 and the plane of the lutidine is nearly perpendicular
(93.4ꢁ) to that of P–Ir–N plane. One might note that the
tilting of the pyridine plane is induced by the chirality of
the MONOPHOS and represents another chiral element
in the catalytic intermediates derived from these complexes
that may affect enantioselectivity. With ligands such as 3-
methylisoquinoline and tmi, however, diastereomers are
produced since coordination to the metal center produces
an axis of chirality along the M–N bond (Scheme 3). The
rate of rotation about the Ir–N bond for these complexes
is slow on the NMR time scale at room temperature and
tmi complex exists as a mixture of diastereomers in a
ꢁ1:1 ratio in solution upon equilibration. The 3-methyliso-
quinoline complex, however, was found to have been
formed in 64% de. Upon crystallization a single isomer of
the tmi complex was isolated, as shown in Fig. 2. Dissolu-
tion of this isomer showed a gradual appearance of the
O1
N2
Ir1
P1
N1
O2
Fig. 2. ORTEP drawing of the cation in [Ir(cod)((S)-MONOPHOS)(tmi)]
SbF6.
other diastereomer over a period of several days. This ori-
entation of the unsymmetrical ligands, as well as the angle
of twist, can be induced by the chirality of the MONO-
PHOS and represents still another chiral element in cata-
lytic intermediates derived from these complexes that can
affect enantioselectivity.
The complex [Ir(cod)(3-methylisoquinoline)Cl] was pre-
pared in order to estimate the potential variation in ligand
rotational barriers in these compounds. The 1H NMR
spectrum at 20 ꢁC shows four cod olefin proton resonances
in C6D6 solution. This demonstrates that rotation of 3-
methylisoquinoline is slow on the NMR timescale at
20 ꢁC. The two olefin proton resonances at d 5 broaden
and become a single broad resonance at 60 ꢁC and an addi-
tional broadening from exchange of the resonances of
ꢁ2 Hz was observed at 40 ꢁC. This corresponds to a rate
of ꢁ6 sꢀ1 and a rotation barrier of ꢁ16.7 kcal molꢀ1. Thus
it is clear that in systems with greater steric interactions,
barriers greater than 20 kcal molꢀ1 can be expected.
O2
Ir1
P1
O1
2.2. Asymmetric imine hydrogenation
N1
Initial studies of the hydrogenation of tmi using
[Ir(cod)((S)-MONOPHOS)(L)]X (X = SbF6, BArF; L =
3-methylisoquinoline) showed that when using methanol
solvent, the BArF salts of these catalysts were superior
to the corresponding SbF6 salts. This is in agreement with
the reports of Pfaltz and co-workers [36,37] and Burgess
and co-workers [38] which noted that the BArF anion
increases catalyst activity in the iridium-catalyzed hydro-
genation of olefins, and with reports of increased selectiv-
ity for imine hydrogenation with the BArF anion [22].
Although high ee’s were obtained in methanol solution
in some trials, the results were highly variable. This vari-
ability has been observed previously in enantioselective
imine hydrogenation [5] and any attempts to improve
N2
Fig. 1. ORTEP drawing of the cation in[Ir(cod)((S)-MONOPHOS)(2,6-
lutidine)]SbF6.
R
N
R'
S
S
N
S
L
S
L
(Sa)
(Ra)
R'
R
Scheme 3. The enantiomers formed due to axial chirality. (R0 > R).

4948
J.W. Faller et al. / Journal of Organometallic Chemistry 691 (2006) 4945–4955
Table 2
Effect of ligand
reproducibility using distilled solvents or drying agents for
methanol led to poor ee’s.
L
% Conversion
% ee
The problem of reproducibility was resolved by employ-
ing a solvent-less system. The catalyst slowly dissolved in
the imine substrate and using a 1% catalyst loading under
40 bar H2 for 24 h at RT, ee’s of 46% ee were achieved with
3-methylisoquinoline as the ligand. These results were
reproducible and unless otherwise specified, these same
conditions were used throughout the hydrogenation trials.
Since the product is formed enantioselectively, over time
the solvent potentially contains a product which is a chiral
non-racemic ligand that could alter the selectivity. In order
to examine the importance of this potential effect, the reac-
tion was quenched at different times and the ee determined.
These results are summarized in Table 1. The variability in
the ee was small and did not correlate with the time of the
reaction. We therefore concluded that the enantioselectiv-
ity was effectively constant throughout the reaction. There
was also little effect of pressure on the enantioselectivity.
The reaction was conducted at 40, 20, and 10 bar of hydro-
gen (Table 1) and although the conversion decreased from
65% to 40% to 20% respectively, only a small increase in ee
was seen as the pressure was decreased from 40 bar to
10 bar (46–51% respectively). Furthermore, it was found
that the reaction could be carried out with 0.5 mol% cata-
lyst without loss of enantioselectivity.
MeCN
3-Methylisoquinoline
Acridine
47
47
56
80
50 (+)
46 (+)
58 (+)
50 (+)
2,6-Lutidine
Attempts utilizing either the BArF or SbF6 salt of the
bis-monophos complex led to variable results (Table 3),
but all with low conversions and enantioselectivities (43%
conversion and only 15% ee in the best trial). A second
experiment was conducted with the further addition of an
equivalent of (S)-MONOPHOS to the reaction mixture.
Excess (S)-MONOPHOS should encourage the formation
of the bis-(S)-MONOPHOS complex and would increase
the enantioselectivity of the reaction if the bis-(S)-MONO-
PHOS complex were the active species. Although the cata-
lytic activity (50% conversion) was unaffected by the
presence of additional phosphoramidite, less than 1% ee
was observed in the product amine. Conversely, when
excess 3-methylisoquinoline was added to the mixture, only
a small decrease in conversion (33%) and enantioselectivity
(36%) was observed. It was also determined that the neutral
[Ir(cod)((S)-MONOPHOS)Cl] compound is inactive for
imine hydrogenation under these conditions.
While the above results suggest that a bis-(S)-MONO-
PHOS complex is not the species responsible for enantiose-
lectivity, the modest, but significant effect that the N-donor
ligand has on the enantioselectivity shows that a complex
involving one (S)-MONOPHOS is likely to be principally
involved in controlling the chirality of the product.
Although the (S)-MONOPHOS may be the major determi-
nant of enantioselectivity, the N-donor ligand must still
play some role in the selectivity. The observed variations
in ee show that the N-donor ligands have not been com-
pletely displaced by the imine substrate.
The complex [Ir(cod)((S)-MONOPHOS)(tmi)]BArF
was prepared. This complex catalyzed the reduction of
tmi with similar enantioselectivity (44% ee) to complexes
having a pyridine type ligand but with a greater rate
(75% conversion in 24 h). Only the 2,6-lutidine complex
showed a similar rate with 80% conversion in 24 h. This
may be because the pyridine ligands compete for binding
with the substrate and therefore decrease the rate, while
The use of acridine as a ligand in place of the 3-methyl-
isoquinoline increased the ee from 46% to 58%. Other N-
donor ligands were tested for this reaction, but the
observed ee’s were consistently within the range of 46–
58%. The acetonitrile complex was prepared, and tested
for asymmetric imine hydrogenation and with this complex
47% conversion and 50% ee were achieved. Therefore, a
number of experiments were performed under similar con-
ditions to further investigate the significance of the N-
donor ligand.
From examination of Table 2, it can be seen that the
nature of the N-donor ligand only has a modest effect on
the reaction. One possible explanation for the small effect
the nature of L has on the enantioselectivity is that the
ligand dissociates and scrambling of the ligands occurs to
give a bis-(S)-MONOPHOS complex. If this complex is
the active catalytic species, then one would expect a similar
result with any ligand that would allow the formation of
the bis-(S)-MONOPHOS complex under the catalytic con-
ditions. One would also expect a previously prepared bis-
(S)-MONOPHOS complex to give the same results.
Table 3
Importance of N-donor ligand
L
Anion
% Conversion
% ee
Table 1
(S)-MONOPHOS
(S)-MONOPHOS
(S)-MONOPHOS
(S)-MONOPHOS
3-Methylisoquinoline
3-Methylisoquinoline
Cl
BArF
BArF
BArF
SbF6
BArF
BArF
None
43
35
33
17
50
33
<1
15 (+)
4 (+)
Effect of time and pressure (L = 3-methylisoquinoline)
Pressure (bar)
Time (h)
% Conversion
% ee
11 (+)
4 (ꢀ)
40
40
40
40
20
10
48
24
10
5
24
24
65
47
33
25
40
20
48 (+)
46 (+)
50 (+)
45 (+)
48 (+)
51 (+)
<1% (+)a
36 (+)b
NA
a
Excess (S)-MONOPHOS added.
b
Excess 3-methylisoquinoline added.

J.W. Faller et al. / Journal of Organometallic Chemistry 691 (2006) 4945–4955
4949
2,6-lutidine may compete poorly for binding since the N is
sterically hindered. An important feature is that the pres-
ence of the pyridine ligands contributes to a measurable
increase in ee relative to having the tmi alone.
ous molar ratios of [Ir(cod)((S)-MONOPHOS)(acri-
dine)]SbF6 to tmi ranging from 3:1 to 1:3 Ir:tmi, with the
Ir complex concentration constant at 0.003 M, were pre-
pared in CD2Cl2 and the ligand exchange was monitored
1
Acridine, which ultimately was the most successful
ligand in terms of enantioselectivity, was then investigated
as an additive to [Ir(cod)((S)-MONOPHOS)(tmi)]BArF as
the catalyst (Table 4). Acridine was tested at 0.5%, 1% and
2% relative to the substrate, with the catalyst concentration
held constant at 1%. Although our results varied substan-
tially over different trials, the general trend was that any
added acridine increased conversion and decreased or
had little effect on enantioselectivity relative to the case
where no acridine was added. The important feature is that
the results do not reflect the substantially increased enanti-
oselectivity observed when the starting complex contains
acridine as the ligand rather than tmi. The results with the
tmi complex were variable in the presence of acridine,
but consistently gave lower ee’s than the complex that orig-
inally contained the acridine ligand. They also consistently
gave greater conversions than any system lacking added
acridine. This is in contrast to the result observed with
the catalyst bearing the 3-methylisoquinoline ligand, where
excess 3-methylisoquinoline decreased the activity of the
catalyst in terms of conversion.
by H and 31P NMR. The equilibration took several days,
but resulted in effectively complete replacement of the acri-
dine by tmi in the solutions containing an equivalent or
excess of tmi (Ir:tmi, 1:3, 11 days; 1:2, 15 days; 1:1 15 days).
Also, after 15 days, the two solutions with less tmi than
1
iridium show very little free tmi in the H NMR and were
consistent with full displacement of acridine. The Ir–acri-
dine:Ir–tmi ratio was found to be approximately 2:1 when
Ir:tmi was 3:1 and 1:1 where Ir:tmi was 2:1. Acridine was
then added to the 1:3 Ir:tmi solution after it had fully con-
verted to the tmi complex. Enough acridine was added to
give a concentration equal to the tmi so that Ir:tmi:acridine
was 1:3:3. After four days, no evidence of the acridine com-
plex was observed by NMR.
Since the catalytic system has a 1:100 ratio of acridine to
tmi, the above data shows that the acridine will eventually
be displaced by tmi. In order to determine if it could occur
on the timescale of the reaction, a solution in CD2Cl2 of
[Ir(cod)((S)-MONOPHOS)(acridine)]SbF6 and tmi in a
1:10 ratio was prepared. This ratio gives a significant excess
of tmi while still allowing observation of the iridium com-
1
plex. The H and 31P NMR spectra of the mixture were
2.3. Ligand displacement studies
monitored frequently. Although not strictly pseudo-first
order, a half-life of ꢁ1 h is observed for a concentration
of iridium complex of 0.022 M. The replacement of acri-
dine occurred over several hours with only small amounts
of the acridine complex present after 4.5 h. Our catalytic
system was investigated with a 1:100 ratio of complex:tmi
and this situation could in fact allow for complete displace-
ment of the ligand by the substrate early in the reaction if
similar or faster rates of displacement were involved. In
fact, when a 0.03 M solution of [Ir(cod)((S)-MONO-
PHOS)(acridine)]BArF in tmi was prepared, the 31P
NMR showed only two resonances (d 115.99 and 115.91)
consistent with the diastereomers of the tmi complex.
Therefore, one can expect tmi to displace acridine upon
dissolution of the catalyst in the absence of H2. Since the
enantioselectivity does depend on the pyridine ligand, it
is apparent that the ligand is not displaced to a great degree
before the catalyst precursor is converted into a less labile
form under the hydrogen pressure. This is consistent with a
mechanism involving Ir(III), but not involving Ir(I) with
lifetimes consistent with ligand displacement. It also sug-
gests that pressurization of the reaction vessel immediately
after addition of the substrate is essential to avoid degrada-
tion of the enantioselectivity of the catalyst.
In order to better understand the active catalyst, it was
necessary to determine if the pyridine ligands remain
bound to iridium in the presence of the imine substrate.
The 2,6-lutidine complex was combined with an equimolar
1
amount of the imine substrate (tmi). The H NMR of the
0.015 M solution showed free tmi and the 31P NMR
showed a single resonance consistent with that of the 2,6-
lutidine complex. The 31P resonances of the two diastereo-
mers of the tmi complex grew in with nearly equal intensities
and although the equilibration took several days, the tmi
complex was ultimately strongly favored over the 2,6-
lutidine complex. This was as expected since 2,6-lutidine
should be a poor competitor for binding with tmi owing
to steric hindrance.
A similar experiment was undertaken for the acridine
complex. The increased ee observed with the acridine com-
plex may be due to its ability to remain bound to iridium
throughout the reaction. A number of solutions with vari-
Table 4
Effect of acridine as an additive to [Ir(cod)((S)-MONOPHOS)(tmi)]BArF
% Acridine
% Conversion
% ee
0
73
83
87
96
94
96
98
44 (+)
37 (+)
26 (+)
44 (+)
32 (+)
45 (+)
42 (+)
3. Conclusion
0.5
0.5
0.5
1
1
2
We have found that the relatively inexpensive chiral
monodentate phosphoramidite (S)-MONOPHOS may be
used in combination with pyridines for asymmetric imine
hydrogenation of 2,3,3-trimethylindolenine with compara-

4950
J.W. Faller et al. / Journal of Organometallic Chemistry 691 (2006) 4945–4955
ble enantioselectivity to some of the more costly chiral biden-
tate bisphosphines. With few exceptions, such as the iridium
complexes prepared from BCPM [15], Josiphos-type ligands
[17], and BICP [16] which have achieved ee’s of 91%, 93%,
and 95.1% with the aid of specific additives, the ee’s obtained
with this system promise to be competitive with those con-
taining many other more difficultly prepared ligands.
CDCl3) d: 8.07–7.92 (4H, m, aromatic-H); 7.50–7.29 (7H,
m, aromatic-H); 7.20 (1H, m, aromatic-H); 5.42 (2H, m,
cod-CH); 3.44 (1H, m, cod-CH); 2.84 (6H, d, N(CH3)2,
3JPH = 10.8 Hz); 2.54 (1H, m, cod-CH); 2.29 (1H, m,
cod-CH2); 2.21 (1H, m, cod-CH2); 2.04 (1H, m, cod-
CH2); 1.91 (2H, m, cod-CH2); 1.39 (3H, m, cod-CH2).
31P NMR (162 MHz, CDCl3) d: 117.4. 13C NMR
(100 MHz, CDCl3) d: 132.9-125.6, 124.3, 121.5 (20C, aro-
matic C); 103.7 (1C, d, cod-CH, JPC = 20.0 Hz); 103.1
(1C, d, cod-CH, JPC = 18.5 Hz); 55.7 (1C, cod-CH); 52.7
We are currently examining other systems of this type,
but in preliminary experiments have found that more
highly modified analogues of MONOPHOS ((S)-(+)-(2,6-
dimethyl-3,5-dioxa-4-phospha-cyclohepta[2,1-a;3,4-a0]dina-
phthalen-4-yl)dimethylamine; (S)-(+)-(3,5-dioxa-4-phos-
pha-cyclohepta[2,1-a;3,4-a0]dinaphthalen-4- yl)bis[(1S)-1-
phenylethyl]amine) did not significantly improve the
enantioselectivity, although they did enhance the rate of
hydrogenation. Since the [Ir(cod)((S)-MONOPHOS)(L)]-
BArF (L = 3-methylisoquinoline, acridine, and 2,6-luti-
dine) complexes are more effective catalysts for the
asymmetric hydrogenation of 2,3,3-trimethylindolenine than
the [Ir(cod)((S)-MONOPHOS)(tmi)]BArF complex itself,
it follows that the pyridine ligand remains coordinated in
the catalytically active species. Indirectly this implies that
the species involved in the catalytic cycle are dominated
by the presence of higher oxidation state iridium species
which would be less prone to exchange than Ir(I) interme-
diates. Our results are not inconsistent with the fleeting
existence of an Ir(I) species, but would appear to favor
the more recent mechanisms of the type which have been
suggested that follow an Ir(III)/Ir(V) pathway.
2
(1C, cod-CH); 38.6 (2C, d, N(CH3), JPC = 10.0 Hz); 34.6
(1C, d, cod-CH2, JPC = 3.0 Hz); 33.0 (1C, d, cod-CH2,
JPC = 3.0 Hz); 29.4 (1C, d, cod-CH2, JPC = 3.6 Hz); 29.1
(1C, d, cod-CH2, JPC = 3.6 Hz). Anal. Calc. for
C30H30Cl1Ir1N1O2P1 Æ CH2Cl2: C, 47.75; H, 4.14; N, 1.80.
Found: C, 47.83; H, 4.33; N, 1.56%.
4.2.2. [Ir(cod)(3-methylisoquinoline)Cl]
[Ir(cod)Cl]2 (25 mg, 0.037 mmol) was added to a flame-
dried flask under N2 and dissolved in CH2Cl2 (5 mL). To
the orange solution was added 3-methylisoquinoline
(10.7 mg, 0.074 mmol). The resulting yellow solution was
stirred at RT for 30 min, then the solvent removed in vacuo
to leave a yellow solid. Yellow crystals of the compound
(91% yield) were obtained from a CH2Cl2 solution by
vapor diffusion of ether.
1H NMR (400 MHz, CDCl3) d: 9.42 (1H, s, NCH); 7.96
(1H, d, isoq-H, J = 8.4 Hz); 7.77 (2H, m, isoq-H); 7.65
(2H, m, isoq-H); 4.56 (2H, m, cod-CH); 3.25 (3H, s,
CH3); 3.18 (2H, m, cod-CH); 2.51–2.27 (4H, m, cod-
CH2); 1.80 (1H, m, cod-CH2); 1.62 (2H, m, cod-CH2);
1.52 (1H, m, cod-CH2). 13C NMR (125 MHz, CD2Cl2) d:
152.7, 151.3, 135.5, 132.0, 127.7, 127.5, 127.2, 125.7,
122.5 (9C, aromatic C); 69.8 (1C, cod-CH); 68.5 (1C,
cod-CH); 60.6 (1C, cod-CH); 56.5 (1C, cod-CH); 32.7
(1C, cod-CH2); 31.6 (1C, cod-CH2); 31.5 (1C, cod-CH2);
30.6 (1C, cod-CH2); 25.2 (1C, CH3). Anal. Calc. for
C18H21Cl1Ir1N1: C, 45.13; H, 4.42; N, 2.92. Found: C,
44.91; H, 4.40; N, 2.79%.
4. Experimental
4.1. General methods
All synthetic manipulations were performed under a
nitrogen atmosphere using standard Schlenk techniques.
CH2Cl2 was dried by distillation over CaH2 and 2,3,3-
trimethylindolenine was distilled prior to use. [Ir(cod)Cl]2
[39] and NaBArF [40] were prepared by published meth-
ods. (S)-MONOPHOS, AgSbF6, 3-methylisoquinoline,
acridine, 2,6-lutidine, acetonitrile, pentane, and diethyl
ether were used as received. NMR spectra were recorded
on Bruker 400 or 500 MHz instruments and the chemical
shifts reported in ppm calibrated by reference to solvent
resonances. Enantiomeric excesses were determined by
4.2.3. [Ir(cod)((S)-MONOPHOS)(3-methyl-
isoquinoline)]BArF
[Ir(cod)((S)-MONOPHOS)Cl] (29 mg, 0.036 mmol) was
added to a flame-dried flask under N2. It was dissolved in
CH2Cl2 (5 mL). NaBArF (31.9 mg, 0.036 mmol) was added
to the orange solution, turning it dark orange, then 3-meth-
ylisoquinoline (5.4 mg, 0.036 mmol) was added yielding a
bright red solution. The mixture was stirred at RT for
4 h and then filtered through Celite. The solvent was
removed by rotary evaporation to yield a red oil which
was washed with pentane and stored under vacuum until
it solidified. The product was collected in 89% yield and
66% de.
using
(R)-(ꢀ)-2,2,2-trifluoro-1-(9-anthryl)-ethanol
as
chemical shift reagent in CDCl3. Complexes containing
the BArF counter ion did not provide satisfactory elemen-
tal analyses. In general, compositions of these complexes
were determined by comparison of NMR with the analo-
gous hexafluoroantimonates.
4.2. Synthesis of iridium compounds
Major diastereomer: 1H NMR (400 MHz, CDCl3) d:
8.39 (1H, d, aromatic-H, J = 8.8 Hz); 8.24 (1H, d, aro-
matic-H, J = 8.8 Hz); 8.02 (1H, d, aromatic-H,
J = 8.8 Hz); 7.75 (8H, s, BArF-H); 7.54 (4H, s, BArF-H);
4.2.1. [Ir(cod)((S)-MONOPHOS)Cl]
This compound was prepared by analogy to
1
[Ir(cod)((R)-MONOPHOS)Cl] [41]. H NMR (400 MHz,

J.W. Faller et al. / Journal of Organometallic Chemistry 691 (2006) 4945–4955
4951
7.94–7.18 (11H, complex, aromatic-H); 7.00 (2H, m, aro-
matic-H); 6.93 (1H, d, aromatic-H, J = 8.4 Hz); 5.79 (1H,
d, aromatic-H, J = 8.0 Hz); 4.81 (1H, m, cod-CH); 4.60
(1H, m, cod-CH); 4.35 (1H, m, cod-CH); 3.87 (1H, m,
cod-CH); 3.07 (3H, s, CH3); 2.75 (6H, d, N(CH3)2,
3JPH = 11.2 Hz); 2.59–2.15 (8H, complex, cod-CH2). 31P
NMR (162 MHz, CD2Cl2) d: 110.7. Minor diastereomer:
1H NMR (400 MHz, CDCl3) d: 9.41 (1H, s, aromatic-H);
9.25 (1H, s, aromatic-H); 8.97 (1H, s, aromatic-H); 8.17
(1H, d, aromatic-H, J = 8.8 Hz); 8.05 (1H, d, aromatic-
H, J = 7.6 Hz); 7.75 (8H, s, BArF-H); 7.54 (4H, s, BArF-
H); 7.94–7.18 (12 H, complex, aromatic-H); 6.80 (1H, d,
aromatic-H, J = 8.8 Hz); 4.50 (1H, m, cod-CH); 3.94
(1H, m, cod-CH); 3.79 (2H, m, cod-CH); 2.72 (6H, d,
aromatic-H); 8.45 (1H, d, aromatic-H, J = 8.0 Hz); 8.20
(1H, d, aromatic-H, J = 8.8 Hz); 8.08–7.14 (14 H, com-
plex, aromatic-H); 7.04 (1H, t, aromatic-H, J = 7.2 Hz);
5.08 (1H, m, cod-CH); 4.58 (1H, m, cod-CH); 4.42
(1H, m, cod-CH); 3.97 (1H, m, cod-CH); 2.82 (6H, d,
3
N(CH3)2, JPH = 11.1 Hz); 2.54–2.12 (8H, complex, cod-
CH2); 1.95 (3H, s, CH3). 31P NMR (162 MHz, CDCl3)
d: 117.9. Diastereomeric mixture: 13C NMR (126 MHz,
CD2Cl2) d: 154.3–147.0, 136.2–120.0 (58C, aromatic-C);
106.7 (1C, d, cod-CH, JPC = 16.2 Hz); 100.5 (1C, d,
cod-CH, JPC = 14.6 Hz); 99.4 (1C, d, cod-CH,
JPC = 17.3 Hz); 98.8 (1C, d, cod-CH, JPC = 17.3 Hz),
68.9 (1C, cod-CH); 67.3 (1C, cod-CH); 66.4 (1C, cod-
CH); 64.7 (1C, cod-CH); 37.5 (2C, d, N(CH3)2,
3
2
N(CH3)2, JPH = 11.2 Hz); 2.59–2.15 (8H, complex, cod-
2JPC = 6.2 Hz); 37.3 (2C, d, N(CH3)2, JPC = 7.6); 33.4,
CH2); 1.97 (3H, s, CH3). 31P NMR (162 MHz, CD2Cl2)
33.0, 32.4, 30.5, 30.1–29.7, 29.1, 25.5, 23.5 (10C, cod-
CH2, CH3). Anal. Calc. for C40H39F6Ir1N2O2P1Sb1 Æ
CH2Cl2: C, 43.85; H, 3.68; N, 2.50. Found: C, 43.94;
H, 3.77; N, 2.39%.
d: 116.9. Diastereomeric mixture: 13C NMR (126 MHz,
1
CD2Cl2) d: 162.1 (8C, q, BC, JBC = 51.0 Hz); 153.1–
46.9, 135.4–117.8 (114C, aromatic-C, CF3); 106.6 (1C, d,
cod-CH, JPC = 16.1 Hz, major); 105.1 (1C, d, cod-CH,
JPC = 15.6 Hz, minor); 99.6 (1C, d, cod-CH, JPC
=
4.2.5. [Ir(cod)((S)-MONOPHOS)(acridine)]BArF
To a flame-dried flask under N2 was added [Ir(cod)((S)-
MONOPHOS)Cl] (35 mg, 0.05 mmol). The solid was dis-
solved in dry CH2Cl2 (3 mL) to give a pale orange solution.
NaBArF (44.6 mg, 0.05 mmol) was added turning the solu-
tion darker orange, then acridine (9.0 mg, 0.05 mmol) was
added. After an hour, precipitate had formed in the red
solution. The mixture was filtered through Celite and the
solvent removed by rotary evaporation. A red oil formed
which was washed with pentane and yielded and orange
solid under vacuum. The product was collected in 81%
yield.
18.9 Hz, minor); 98.5 (1C, d, cod-CH, JPC = 17.2 Hz,
major); 68.9 (1C, cod-CH, minor); 67.3 (1C, cod-CH,
major); 66.6 (1C, cod-CH, major); 65.0 (1C, cod-CH,
2
minor); 37.4 (2C, d, N(CH3)2, JPC = 6.4 Hz, major); 37.2
2
(2C, d, N(CH3)2, JPC = 8.2 Hz, minor); 34.9 (1C, d, cod-
CH2, JPC = 3.6 Hz, minor); 33.1 (1C, d, cod-CH2,
JPC = 3.6 Hz, major); 32.7 (1C, d, cod-CH2, JPC = 2.9 Hz,
major); 31.8 (1C, d, cod-CH2, JPC = 2.9 Hz, minor); 30.6
(1C, d, cod-CH2, JPC = 2.8 Hz, major); 30.0 (1C, d, cod-
CH2, JPC = 3.9 Hz, minor); 29.2 (1C, d, cod-CH2,
JPC = 2.8 Hz, minor); 29.0 (1C, d, cod-CH2, JPC = 2.8 Hz,
major); 25.4 (1C, CH3, major); 23.5 (1C, CH3, minor).
The complex was analyzed by mass spectrometry in a
solution of methanol. Peaks at 837.79 m/z, 839.90 m/z,
840.89 m/z, and 841.95 m/z were consistent with the iso-
tope model for [Ir(cod)((S)-MONOPHOS)(acridine)]+.
1H NMR (400 MHz, CDCl3) d: 9.51 (1H, d, aromatic-
H, J = 8.5 Hz); 8.87 (1H, s, aromatic-H); 8.35 (1H, d,
aromatic-H, J = 8.5 Hz); 8.16–8.04 (3H, m, aromatic-H);
7.81–7.63 (7H, m, aromatic-H); 7.75 (8H, s, BArF-H);
7.54 (4H, s, BArF-H); 7.40 (1H, m, aromatic-H); 7.25
(2H, m, aromatic-H); 7.16 (1H, d, aromatic-H,
J = 8.6 Hz); 6.84 (1H, d, aromatic-H, J = 8.6 Hz); 6.72
(1H, m, aromatic-H) 6.42 (1H, d, aromatic-H,
J = 8.6 Hz); 6.23 (1H, m, aromatic-H); 4.98 (1H, m,
cod-CH); 4.87 (1H, m, cod-CH); 4.62 (1H, m, cod-CH);
4.12 (1H, m, cod-CH); 2.72 (6H, d, N(CH3)2,
3JPH = 11.5 Hz); 2.54–2.12 (4H, m, cod-CH2); 1.34–0.99
(4H, m, cod-CH2). 31P NMR (122 MHz, CD2Cl2) d:
116.1. 13C NMR (100 MHz, CDCl3) d: 162.1 (4C, q,
4.2.4. [Ir(cod)((S)-MONOPHOS)(3-methyl-
isoquinoline)]SbF6
[Ir(cod)((S)-MONOPHOS)Cl] (50 mg, 0.072 mmol) was
added to a flame-dried flask under N2 in the absence of
light, and dissolved in CH2Cl2 (5 mL). AgSbF6 (24.7 mg,
0.072 mmol)
and
3-methylisoquinoline
(10.3 mg,
0.072 mmol) were added. The mixture was stirred for 1 h
at RT, then filtered through Celite. The solvent was
removed by rotary evaporation, and the resulting orange
residue was washed with pentane and dried in vacuo. The
product was collected in 98% yield with 32% de. Red crys-
tals were obtained from a CH2Cl2 solution by vapor diffu-
sion of ether.
Major diastereomer: 1H NMR (400 MHz, CDCl3) d:
8.54 (1H, d, aromatic-H, J = 9.2 Hz); 8.33 (1H, d, aro-
matic-H, J = 8.0 Hz); 8.08–7.14 (14 H, complex, aro-
matic-H); 6.89 (1H, t, aromatic-H, J = 7.2 Hz); 6.17
(1H, d, aromatic-H, J = 8.8 Hz); 4.95 (1H, m, cod-CH);
4.74 (1H, m, cod-CH); 4.36 (1H, m, cod-CH); 3.80
(1H, m, cod-CH); 3.19 (3H, s, CH3); 2.78 (6H, d,
1
BC, JBC = 50.2 Hz); 148.3–117.8 (61C, aromatic C,
CF3); 103.6 (1C, d, cod-CH, JPC = 15.2 Hz); 102.1 (1C,
d, cod-CH, JPC = 17.6 Hz); 66.6 (1C, cod-CH); 64.8
2
(1C, cod-CH); 37.6 (2C, d, N(CH3)2, JPC = 7.4 Hz);
3
N(CH3)2, JPH = 11.6 Hz); 2.54–2.12 (8H, complex, cod-
34.0 (1C, d, cod-CH2, JPC = 3.2 Hz); 32.3 (1C, d, cod-
CH2, JPC = 3.2 Hz); 30.2 (1C, cod-CH2); 28.6 (1C, cod-
CH2).
CH2). 31P NMR (162 MHz, CDCl3) d: 111.6. Minor dia-
stereomer: 1H NMR (400 MHz, CDCl3) d: 9.70 (1H, s,

4952
J.W. Faller et al. / Journal of Organometallic Chemistry 691 (2006) 4945–4955
4.2.6. [Ir(cod)((S)-MONOPHOS)(acridine)]SbF6
To a flame-dried flask under N2 in the dark was added
[Ir(cod)((S)-MONOPHOS)Cl] (40 mg, 0.058 mmol). It
was dissolved in dry CH2Cl2 (3 mL). AgSbF6 (20 mg,
0.058 mmol) was added to the clear yellow solution. Acri-
dine (10.4 mg, 0.058 mmol) was then added resulting in a
cloudy red solution. The reaction was stirred for 1 h at
RT and filtered through Celite. The solvent was removed
by rotary evaporation. The resulting red solid was washed
with pentane and the excess solvent was removed under
vacuum. The complex was collected in 92% yield. Crystals
were obtained by vapor diffusion of ether into a CH2Cl2
solution of the complex.
JPC = 16.5 Hz); 101.4 (1C, d, cod-CH, JPC = 15.6 Hz);
64.6 (1C, cod-CH); 64.2 (1C, cod-CH); 37.3 (2C, d,
N(CH3)2, 2JPC = 7.9 Hz); 32.7 (1C, d, cod-CH2,
JPC = 2.9 Hz); 32.2 (1C, d, cod-CH2, JPC = 4.0 Hz); 28.9
(1C, d, cod-CH2, JPC = 2.8 Hz); 28.6 (1C, d, cod-CH2,
JPC = 2.2 Hz); 26.1 (1C, lut-CH3); 23.8 (1C, lut-CH3).
4.2.8. [Ir(cod)((S)-MONOPHOS)(2,6-lutidine)]SbF6
To a flame-dried flask under N2 in the dark was added
[Ir(cod)((S)-MONOPHOS)Cl] (25 mg, 0.036 mmol). It
was dissolved in CH2Cl2 (3 mL). AgSbF6 (12.4 mg,
0.036 mmol) was added to the clear yellow solution. A
solution of 2,6-lutidine (38.6 mg, 0.036 mmol) in CH2Cl2
was then added resulting in a cloudy red solution. The reac-
tion was stirred for 1 h at RT and filtered through Celite.
The solvent was removed by rotary evaporation. The
resulting red solid was washed with pentane and the excess
solvent was removed under vacuum. The complex was
obtained in 94% yield. It was crystallized by vapor diffusion
of ether into a solution of the complex in CH2Cl2, and ana-
lyzed by X-ray crystallography.
1H NMR (400 MHz, CDCl3) d: 9.57 (1H, d, aromatic-
H, J = 8.8 Hz); 8.99 (1H, d, aromatic-H, J = 8.8 Hz);
8.41 (2H, m, aromatic-H); 8.22 (3H, m, aromatic-H);
7.82–7.62 (6H, m, aromatic-H); 7.38 (1H, t, aromatic-H,
J = 7.2 Hz); 7.23 (2H, m, aromatic-H); 7.14 (1H, m, aro-
matic-H); 6.83 (1H, d, aromatic-H, J = 8.4 Hz); 6.74 (1H,
t, aromatic-H, J = 7.6 Hz); 6.49 (1H, d, aromatic-H,
J = 9.2 Hz); 6.32 (1H, m, aromatic-H); 4.88 (2H, m, cod-
CH); 4.59 (1H, m, cod-CH); 4.12 (1H, m, cod-CH); 2.90–
2.50 (4H, m, cod-CH2); 2.73 (6H, d, N(CH3)2
3JPH = 10.8 Hz); 2.38 (2H, m, cod-CH2); 2.23 (2H, m,
cod-CH2). 31P NMR (162 MHz, CD2Cl2) d: 116.6. 13C
NMR (126 MHz, CDCl3) 147.8–146.9, 140.7, 133.8–119.6
(33C, aromatic C); 102.9 (1C, d, cod-CH, JPC = 16.6 Hz);
102.2 (1C, d, cod-CH, JPC = 15.6 Hz); 66.4 (1C, cod-
CH); 64.5 (1C, cod-CH); 37.3 (2C, d, N(CH3)2,
2JPC = 6.5 Hz); 33.3 (1C, d, cod-CH2, JPC = 3.9 Hz); 32.3
(1C, d, cod-CH2, JPC = 3.6 Hz); 29.6 (1C, m, cod-CH2);
28.5 (1C, m, cod-CH2). Anal. Calc. for C43H39F6Ir1-
N2O2P1Sb1: C, 48.06; H, 3.66; N, 2.61. Found: C, 47.91;
H, 3.68; N, 2.53%.
1H NMR (500 MHz, CD2Cl2) d: 8.24 (1H, d, aromatic-
H, J = 9.0 Hz); 8.10 (1H, d, aromatic-H, J = 8.0 Hz); 7.98
(2H, t, aromatic-H, J = 9.0 Hz); 7.67 (1H, d, aromatic-H,
J = 9.0 Hz); 7.59 (2H, m, aromatic-H); 7.52 (1H, m, aro-
matic-H); 7.38–7.25 (5H, complex, aromatic-H); 7.15
(1H, m, aromatic-H); 6.79 (1H, d, aromatic-H,
J = 7.5 Hz); 4.71 (1H, m, cod-CH); 4.61 (1H, m, cod-
CH); 4.43 (1H, m, cod-CH); 3.77 (1H, m, cod-CH); 3.25
3
(3H, s, CH3); 2.76 (6H, d, N(CH3)2, JPH = 11.0 Hz);
2.78–2.21 (8H, complex, cod-CH2); 1.90 (3H, s, CH3). 31P
NMR (162 MHz, CDCl3): d 116.6. 13C NMR (125 MHz,
CD2Cl2) d: 158.7, 157.9, 148.3, 147.6, 139.2, 132.9, 132.7–
120.4 (25C, aromatic-C); 102.5 (1C, d, cod-CH,
JPC = 16.2 Hz); 102.0 (1C, d, cod-CH, JPC = 16.0 Hz);
65.0 (1C, cod-CH); 64.6 (1C, cod-CH); 37.8 (2C, d,
N(CH3)2, 2JPC = 7.7 Hz); 33.1 (1C, d, cod-CH2,
JPC = 2.9 Hz); 32.7 (1C, d, cod-CH2, JPC = 3.9 Hz); 29.4
(1C, d, cod-CH2, JPC = 2.8 Hz); 29.1 (1C, d, cod-CH2,
JPC = 2.9 Hz); 26.6 (1C, lut-CH3); 24.3 (1C, lut-CH3).
Anal. Calc. for C37H39F6Ir1N2O2P1Sb1: C, 44.31; H, 3.92;
N, 2.79. Found: C, 44.01; H, 3.88; N, 2.67%.
4.2.7. [Ir(cod)((S)-MONOPHOS)(2,6-lutidine)]BArF
[Ir(cod)((S)-MONOPHOS)Cl] (15.5 mg, 0.022 mmol)
was added to a flame-dried flask under N2 and dissolved
in CH2Cl2 (3 mL). NaBArF (19.8 mg, 0.022 mmol) and
2,6-lutidine (2.4 mg, 0.022 mmol) were added. The mixture
was stirred for 1 h at RT, then filtered through Celite. The
solvent was removed by rotary evaporation, and the result-
ing orange residue was washed with pentane and yielded an
orange solid under vacuum. The product was collected in
4.2.9. [Ir(cod)((S)-MONOPHOS)(2,3,3-trimethyl-
indolnine)]BArF
1
43% yield. H NMR (400 MHz, CDCl3) d: 8.16 (1H, m,
aromatic-H); 8.04 (1H, m, aromatic-H); 7.92 (2H, m, aro-
matic-H); 7.74 (8H, s, BArF-H); 7.55 (4H, s, BArF-H);
7.81–7.19 (8H, complex, aromatic-H); 7.04 (2H, m, aro-
matic-H); 6.68 (1H, m, aromatic-H); 4.57 (1H, m, cod-
CH); 4.51 (1H, m, cod-CH); 4.40 (1H, m, cod-CH); 3.72
(1H, m, cod-CH); 3.12 (3H, s, CH3); 2.72 (6H, d,
[Ir(cod)((S)-MONOPHOS)Cl] (29.4 mg, 0.042 mmol)
was added to a flame-dried flask under N2 and dissolved
in CH2Cl2 (5 mL). NaBArF (37.1 mg, 0.042 mmol) was
added to the solution which became brown and cloudy.
Upon the addition of a CH2Cl2 solution of 2,3,3-trimethy-
lindolenine (96.7 mg, 0.042 mmol), the solution became
dark red. It was stirred at RT for 6 h, filtered through Cel-
ite, and the solvent removed by rotary evaporation. The
oily residue was washed with ether and pentane and yielded
an orange powder under vacuum. The product was col-
lected in 90% yield. Two diastereomers formed in a 1:1
ratio.
3
N(CH3)2, JPH = 10.4 Hz); 2.68–2.06 (8H, complex, cod-
CH2); 1.87 (3H, s, CH3). 31P NMR (162 MHz, CDCl3): d
116.3. 13C NMR (125 MHz, CD2Cl2) d: 161.7 (4C, q, BC,
1JBC = 50.0 Hz); 158.4, 157.2, 154.9, 147.8, 147.1, 142.5,
138.7, 134.7, 132.5–123.4, 122.0, 121.3, 120.7, 119.7,
117.4 (53C, aromatic C, CF3); 101.9 (1C, d, cod-CH,

J.W. Faller et al. / Journal of Organometallic Chemistry 691 (2006) 4945–4955
4953
1H NMR (400 MHz, CDCl3) d: 8.31 (1H, m, aromatic-
H); 8.19 (1H, d, aromatic-H, J = 8.8 Hz); 8.13 (1H, d, aro-
matic-H, J = 8.4 Hz); 8.02 (3H, m, aromatic-H); 7.94 (1H,
m, aromatic-H); 7.87 (4H, m, aromatic-H); 7.75 (16H, s,
BArF-H); 7.55 (8H, s, BArF-H); 7.66–7.18 (14H, complex,
aromatic-H); 6.94 (3H, m, aromatic-H); 6.78 (1H, d, aro-
matic-H, J = 9.2 Hz); 6.58 (1H, t, aromatic-H,
J = 7.6 Hz); 6.40 (1H, t, aromatic-H, J = 8.0 Hz); 5.56
(1H, t, aromatic-H, J = 7.6 Hz); 4.96 (2H, m, cod-CH);
4.81 (1H, m, cod-CH); 4.74 (1H, m, cod-CH); 4.89 (2H,
m, cod-CH); 3.98 (1H, m, cod-CH); 3.88 (1H, m, cod-
J = 6.7 Hz); 7.26 (3H, m, aromatic-H); 7.10 (1H, d, aro-
matic-H, J = 8.9 Hz); 7.02 (1H, d, aromatic-H,
J = 7.6 Hz); 6.90 (1H, d, aromatic-H, J = 8.9 Hz); 6.62
(1H, t, aromatic-H, J = 7.6 Hz); 6.47 (1H, d, aromatic-H,
J = 7.6 Hz); 5.62 (1H, m, aromatic-H); 5.03 (1H, m, cod-
CH); 4.86 (1H, m, cod-CH); 4.55 (1H, m, cod-CH); 3.95
3
(1H, m, cod-CH); 2.80 (6H, d, N(CH3)2, JPH = 11.6 Hz);
2.78 (3H, s, CH3); 2.74–2.23 (8H, complex, cod-CH2);
1.34 (3H, s, CH3); 1.27 (3H, s, CH3). 31P NMR
(162 MHz, CD2Cl2) d: 115.5. Second diastereomer: 1H
NMR(400 MHz, CD2Cl2) d: 8.28 (1H, m, aromatic-H,
J = 8.9 Hz); 8.12 (2H, m, aromatic-H); 8.04 (1H, m, aro-
matic-H); 7.93 (2H, m, aromatic-H); 7.73 (1H, d, aro-
matic-H, J = 8.9 Hz); 7.17–7.67 (8H, m, aromatic-H);
6.84 (1H, d, aromatic-H, J = 8.9 Hz); 5.05 (1H, m, cod-
CH); 4.89 (1H, m, cod-CH); 4.54 (1H, m, cod-CH); 4.04
(1H, m, cod-CH); 2.79 (3H, s, CH3); 2.77 (6H, d,
3
CH); 2.74 (6H, d, N(CH3)2, JPH = 11.6 Hz); 2.73 (6H, d,
3
N(CH3)2, JPH = 11.6 Hz); 2.65 (3H, s, CH3); 2.58 (3H, s,
CH3); 2.14–2.68 (16H, m, cod-CH2); 1.23 (3H, s, CH3);
1.18 (3H, s, CH3); 1.12 (3H, s, CH3); 1.05 (3H, s, CH3).
31P NMR (162 MHz, CDCl3) d: 115.23, 115.28. 13C
NMR (126 MHz, CD2Cl2) d: 191.7; (1C, N@C); 190.0
1
3
(1C, N@C); 161.7 (8C, q, BC, JBC = 49.9 Hz); 149.9–
N(CH3)2, JPH = 11.6 Hz); 2.74–2.29 (8H, complex, cod-
142.7, 134.7–117.4 (108C, aromatic C, CF3); 102.5 (1C, d,
cod-CH, JPC = 15.1 Hz); 102.4 (1C, d, cod-CH,
JPC = 16.4 Hz); 101.4 (1C, d, cod-CH, JPC = 16.4 Hz);
101.3 (1C, d, cod-CH, JPC = 16.4 Hz); 68.7 (1C, cod-
CH); 68.5 (1C, cod-CH); 67.3 (1C, cod-CH); 67.1 (1C,
cod-CH); 54.7 (1C, C(CH3)2); 54.6 (1C, C(CH3)2); 37.1
CH2); 1.19 (3H, s, CH3) 1.13 (3H, s, CH3). 31P NMR
(162 MHz, CD2Cl2) d: 115.8. Diastereomeric mixture. 13C
NMR (100 MHz, CD2Cl2) d: 191.8; (1C, N@C); 190.7
(1C, N@C); 150.0–146.9, 143.8, 142.9, 133.0–131.0,
128.6–125.6, 124.0–117.7 (52C, aromatic C); 102.8 (1C, d,
cod-CH, JPC = 16.3 Hz, minor); 102.6 (1C, d, cod-CH,
2
(2C, d, N(CH3)2, JPC = 7.5 Hz); 37.0 (2C, d, N(CH3)2,
JPC = 16.3 Hz, major); 101.6 (1C, d, cod-CH, JPC =
2JPC = 7.3 Hz); 32.9 (1C, d, cod-CH2, JPC = 3.6 Hz); 32.8
(1C, d, cod-CH2, JPC = 3.6 Hz); 31.9 (2C, d, cod-CH2,
JPC = 3.7 Hz); 31.8 (2C, d, cod-CH2, JPC = 3.4 Hz); 29.7
(1C, d, cod-CH2, JPC = 2.6 Hz); 29.6 (1C, cod-CH2); 28.8
(1C, cod-CH2); 28.7 (1C, d, cod-CH2, JPC = 2.6 Hz); 22.7
(1C, C(CH3)2); 22.4 (1C, C(CH3)2); 22.2 (1C, C(CH3)2);
22.0 (1C, C(CH3)2); 17.5 (1C, CH3); 15.1 (1C, CH3).
16.7 Hz, major); 101.5 (1C, d, cod-CH, JPC = 15.9 Hz,
minor); 68.8 (1C, cod-CH, minor); 68.7 (1C, cod-CH,
major); 67.2 (1C, cod-CH, minor); 67.0 (1C, cod-CH,
major); 54.7 (1C, C(CH3)2, major); 54.6 (1C, C(CH3)2,
2
minor); 37.2 (2C, d, N(CH3)2, JPC = 8.1 Hz, major); 37.1
2
(2C, d, N(CH3)2, JPC = 7.3 Hz, minor); 33.2 (1C, d, cod-
CH2, JPC = 3.2 Hz); 32.9 (1C, d, cod-CH2, JPC = 3.6 Hz);
32.0 (1C, d, cod-CH2, JPC = 3.9 Hz); 31.7 (1C, d, cod-
CH2, JPC = 2.9 Hz); 29.9 (1C, d, cod-CH2, JPC = 3.7 Hz);
29.7 (1C, d, cod-CH2, JPC = 3.2 Hz); 28.9 (1C, d, cod-
CH2, JPC = 2.0 Hz); 28.8 (1C, d, cod-CH2, JPC = 2.0 Hz);
22.9 (1C, C(CH3)2); 22.5 (1C, C(CH3)2); 22.3 (1C,
C(CH3)2); 22.0 (1C, C(CH3)2); 17.8 (1C, CH3); 15.2 (1C,
CH3). Anal. Calc. for C41H43F6Ir1N2O2P1Sb1: C, 46.69;
H, 4.11; N, 2.66. Found: C, 46.33; H, 4.19; N, 2.70%.
4.2.10. [Ir(cod)((S)-MONOPHOS)(2,3,3-trimethyl-
indolenine)]SbF6
[Ir(cod)((S)-MONOPHOS)Cl] (50 mg, 0.072 mmol) was
added to a flame-dried flask under N2 in the dark and dis-
solved in CH2Cl2 (10 mL). AgSbF6 (24.7 mg, 0.072 mmol)
was added to the solution which turned brown and cloudy.
Upon the addition of a CH2Cl2 solution of 2,3,3-trimethy-
lindolenine (11.5 mg, 0.072 mmol), the solution became
dark red. It was stirred at RT for 1 h, filtered through Cel-
ite, and the solvent removed by rotary evaporation. The
resulting red solid was washed with ether and dried in
vacuo. The compound was collected in 97% yield. The com-
plex formed in a 1:1 mixture of diastereomers. Crystals of a
single diastereomer were obtained by vapor diffusion of
ether into a CH2Cl2 solution of the complex. This diaste-
reomer is shown in Fig. 2. Obtaining the NMR of freshly
dissolved crystals initially showed a set of resonances pre-
dominantly of one diastereomer and allowed the assign-
4.2.11. [Ir(cod)((S)-MONOPHOS)(MeCN)]BArF
To a flame-dried flask under N2 was added [Ir(cod)((S)-
MONOPHOS)Cl] (40 mg, 0.058 mmol). It was dissolved in
dry CH2Cl2 (5 mL). NaBArF (51.0 mg, 0.058 mmol) was
added to the clear yellow solution. A solution of acetoni-
trile (2.4 mg, 0.058 mmol) in CH2Cl2 was then added
resulting in a cloudy red solution. The reaction was stirred
for 1 h at RT and filtered through Celite. The compound
was recrystallized from a CH2Cl2 solution layered with
pentane and the resulting red powder was dried in vacuo
resulting in 30% yield. The low yield is a result of the diffi-
culty in crystallizing BArF complexes.
1
ment of the H resonances to a given diastereomer.
1
Diastereomer from solid: H NMR (400 MHz, CD2Cl2)
d: 8.37 (1H, d, aromatic-H, J = 8.9 Hz); 8.19 (1H, d,
aromatic-H, J = 8.4 Hz); 7.93 (2H, t, aromatic-H,
J = 8.4 Hz); 7.81 (1H, d, aromatic-H, J = 8.9 Hz); 7.64
(1H, t, aromatic-H, J = 7.2 Hz); 7.48 (1H, t, aromatic-H,
1H NMR (500 MHz, CDCl3) d: 8.06–7.87, 7.54, 7.25,
7.38 (12H, aromatic-H); 7.68 (8H, s, BArF-H); 7.51 (4H,
s, BArF-H); 5.15 (1H, m, cod-CH); 5.09 (1H, m, cod-
CH); 4.21 (1H, m, cod-CH); 3.27 (1H, m, cod-CH); 2.68

4954
J.W. Faller et al. / Journal of Organometallic Chemistry 691 (2006) 4945–4955
(6H, d, N(CH3)2, 3JPH = 11.5 Hz); 2.30 (3H, s, CH3); 2.26–
1.98, 1.78–1.62 (8H, m, cod-CH2). 31P NMR (202 MHz,
CDCl3) d: 111.5. 13C NMR (101 MHz, CDCl3) d: 162.7
uration of the monophos and by inverting the coordinates
which gave large increases in R factors. Detailed informa-
tion is provided in the supporting information.
1
(4C, q, BC, JBC = 49.8 Hz); 134.8–117.5 (49C, aromatic
C, CN, CF3); 101.9 (1C, cod-CH); 100.1 (1C, cod-CH);
70.1 (1C, cod-CH); 67.6 (1C, cod-CH); 37.4 (2C, d,
N(CH3)2, 2JPC = 8.7 Hz); 32.9 (1C, d, cod-CH2,
JPC = 3.4 Hz); 32.2 (1C, d, cod-CH2, JPC = 3.4 Hz); 29.2
(1C, d, cod-CH2, JPC = 3.4 Hz); 29.0 (1C, d, cod-CH2,
JPC = 3.4 Hz); 0.76 (1C, CH3).
4.4. Imine hydrogenation
A representative procedure for the hydrogenation of
2,3,3-trimethylindolenine is given. Freshly distilled 2,3,3-
trimethylindolenine (200 mg, 1.26 mmol) was transferred
to an oven-dried glass liner, and [Ir(cod)((S)-MONO-
PHOS)(acridine)]BArF (21.4 mg, 0.0126 mmol) was added.
The glass liner was placed in the pressure vessel, which was
purged with H2. The H2 pressure was then brought to
40 bar and the pressure vessel sealed. The reaction was
run for 24 h. The glass liner was then removed from the
pressure vessel and the reaction quenched immediately with
pentane. The pentane solution was then filtered through
Celite and the pentane removed in vacuo to leave the
imine/amine mixture. The conversion and the ee were
4.3. X-ray structure determination and refinement
Crystals of [Ir(cod)((S)-MONOPHOS)(lutidine)]SbF6
and [Ir(cod)((S)-MONOPHOS)(tmi)]SbF6 were obtained
by vapor diffusion of ether into methylene chloride solu-
tions of the complexes. Important data are presented in
Table 5 and the ORTEP diagrams are shown in Figs. 1
and 2. Data were collected on a Nonius KappaCCD
(Mo Ka radiation) diffractometer and were not specifically
corrected for absorption other than the inherent correc-
tions provided by Scalepack [42]. The structures were
solved by direct methods (SIR92) [43] and refined on F for
all reflections [44]. Non-hydrogen atoms were refined with
anisotropic displacement parameters. Hydrogen atoms
were included at calculated positions.
1
determined by H NMR using the chiral shift agent (R)-
(ꢀ)-2,2,2-trifluoro-1-(9-anthryl)-ethanol.
Appendix A. Supplementary data
Crystallographic data for the structural analysis has
been deposited with the Cambridge Crystallographic Data
Centre, CCDC Nos. 609789 and 609790 for the lutidine
and tmi complex respectively compound. Copies of this
information may be obtained free of charge from: The
Director, CCDC, 12 Union Road, Cambridge, CB2 1EZ
UK; fax: (int code) +44 (1223) 336 033; or email: deposit@
In the case of the tmi complex a disordered diethyl ether
molecule was found. The data for this structure were
adjusted using program SQUEEZE in PLATON [45,46] to
remove the scattering from this moiety. Absolute configu-
rations were determined by reference to the known config-
Table 5
Crystallographic data
[Ir(cod)((S)-MONOPHOS)(lut)]SbF6
Orange, block
37H39IrN2O2P, SbF6
[Ir(cod)((S)-MONOPHOS)(tmi)]SbF6
Color, shape
Red, block
C41H43IrN2O2P, C4H10O, SbF6
1128.86
Mo Ka (monochromatic) 0.71073
123
Empirical formula
Formula weight
C
1002.66
˚
Radiation/A
Mo Ka (monochromatic) 0.71073
173
T/K
Crystal system
Space group
Unit cell dimensions
Orthorhombic
P212121 (no. 19)
Monoclinic
P21 (no. 4)
˚
a/A
11.394(2)
14.8727(4)
21.5025(5)
90
3643.7(6)
4
1.828
45.09
0.10 · 0.10 · 0.12
8297, 4648
0.073
13.4736(7)
11.8511(4)
13.8736(6)
92.553(3)
2213.1(2)
2
1.694
37.24
0.08 · 0.08 · 0.20
9756, 5677
0.048
˚
b/A
˚
c/A
b/ꢁ
3
˚
V/A
Z
Dcalc/g cmꢀ3
l/cmꢀ1 (Mo Ka)
Crystal size/mm
Total no. of reflections, unique reflections
Rint
No. of observed data (I > 3r(I))
Parameters, constraints
Ra, Rwb, GOF
5742
451, 0
0.035, 0.028, 1.25
ꢀ1.01 < 0.78
4384
485, 0
0.036, 0.035, 2.39
ꢀ1.80 < 2.86
ꢀ3
˚
Resolved density/e A
P
P
a
R1 = jjFoj ꢀ jFcjj/ jFoj, for all I > 3r(I).
P
P
b
Rw = [ [w(jFoj ꢀ jFcj)2]/ [w(Fo)2]]1/2
.

J.W. Faller et al. / Journal of Organometallic Chemistry 691 (2006) 4945–4955
4955
[22] C. Blanc, F. Agbossou-Niedercorn, G. Nowogrocki, Tetrahedron:
Asymmetry 15 (2004) 2159.
[23] R. Sablong, J.A. Osborn, Tetrahedron Lett. 37 (1996) 4937.
[24] P.G. Nell, Synlett (2001) 160.
ccdc.cam.ac.uk or www:http://www.ccdc.cam.ac.uk. The
metrical parameters are also available in the supporting
information. Supplementary data associated with this arti-
cle can be found, in the online version, at doi:10.1016/
j.jorganchem.2006.08.032.
[25] R. Crabtree, Acc. Chem. Res. 12 (1979) 331.
[26] V.R. Landaeta, B.K. Munoz, M. Peruzzini, V. Herrera, C. Bianchini,
R.A. Sanchez-Delgado, Organometallics 25 (2006) 403.
[27] 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. Willans, D. Hyett, J.A.F. Boogers, H.J.W. Hender-
ickx, J.G. de Vries, Adv. Synth. Catal. 345 (2003) 308.
[28] A. Martorell, C. Claver, E. Fernandez, Inorg. Chem. Commun. 3
(2000) 132.
[29] S. Feldgus, C.R. Landis, J. Am. Chem. Soc. 122 (2000) 12714.
[30] C.R. Landis, J. Halpern, J. Am. Chem. Soc. 109 (1987) 1746.
[31] J.M. Brown, D. Parker, Organometallics 1 (1982) 950.
[32] J.M. Brown, P.A. Chaloner, Chem. Commun. (1980) 344.
[33] C. Mazet, S.P. Smiidt, M. Meuwly, A. Pfaltz, J. Am. Chem. Soc. 126
(2004) 14176.
References
[1] H.U. Blaser, Adv. Synth. Catal. 344 (2002) 17.
[2] B.R. James, Catal. Today 37 (1997) 209.
[3] M.D. Fryzuk, W.E. Piers, Organometallics 9 (1990) 986.
[4] A. Levi, G. Modena, G. Scorrano, Chem. Commun. (1975) 6.
[5] S. Vastag, J. Bakos, S. Toros, N.E. Takach, R.B. King, B. Heil, L.
Marko, J. Mol. Catal. 22 (1984) 283.
[6] C. Lensink, J.G. Devries, Tetrahedron: Asymmetry 3 (1992) 235.
[7] M.J. Burk, J.E. Feaster, J. Am. Chem. Soc. 114 (1992) 6266.
[8] R. Sablong, J.A. Osborn, Tetrahedron: Asymmetry
3059.
7 (1996)
[34] P. Brandt, C. Hedberg, P.G. Andersson, Chem.-Eur. J. 9 (2003)
339.
[35] Y.B. Fan, X.H. Cui, K. Burgess, M.B. Hall, J. Am. Chem. Soc. 126
(2004) 16688.
[9] C.J. Cobley, J.P. Henschke, Adv. Synth. Catal. 345 (2003) 195.
[10] C.A. Willoughby, S.L. Buchwald, J. Am. Chem. Soc. 114 (1992)
7562.
[11] C.A. Willoughby, S.L. Buchwald, J. Am. Chem. Soc. 116 (1994)
8952.
[36] A. Lightfoot, P. Schnider, A. Pfaltz, Angew. Chem., Int. Ed. 37
(1998) 2897.
[12] F. Spindler, B. Pugin, H.U. Blaser, Angew. Chem., Int. Ed. 29 (1990)
558.
[37] D. Drago, P.S. Pregosin, A. Pfaltz, Chem. Commun. (2002) 286.
[38] M.C. Perry, X.H. Cui, M.T. Powell, D.R. Hou, J.H. Reibenspies, K.
Burgess, J. Am. Chem. Soc. 125 (2003) 113.
[39] J.L. Herde, J.C. Lambert, C.V. Senoff, Inorg. Synth. 15 (1974) 18.
[40] S.R. Bahr, P. Boudjouk, J. Org. Chem. 57 (1992) 5545.
[41] B. Bartels, C. Garcia-Yebra, F. Rominger, G. Helmchen, Eur. J.
Inorg. Chem. (2002) 2569.
[42] Z. Otwinowski, W. Minor, in: M.G. Rossman, E. Arnold (Eds.),
International Tables for Crystallography, vol. F, 2001.
[43] A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, J. Appl.
Crystallogr. 26 (1993) 343.
[13] T. Morimoto, K. Achiwa, Tetrahedron: Asymmetry 6 (1995) 2661.
[14] Y.N.C. Chan, J.A. Osborn, J. Am. Chem. Soc. 112 (1990) 9400.
[15] T. Morimoto, N. Nakajima, K. Achiwa, Synlett (1995) 748.
[16] G.X. Zhu, X.M. Zhang, Tetrahedron: Asymmetry 9 (1998) 2415.
[17] H.U. Blaser, H.P. Buser, R. Hausel, H.P. Jalett, F. Spindler, J.
Organomet. Chem. 621 (2001) 34.
[18] D.M. Xiao, X.M. Zhang, Angew. Chem., Int. Ed. 40 (2001) 3425.
[19] S. Kainz, A. Brinkmann, W. Leitner, A. Pfaltz, J. Am. Chem. Soc.
121 (1999) 6421.
[20] P. Schnider, G. Koch, R. Pretot, G.Z. Wang, F.M. Bohnen, C.
Kruger, A. Pfaltz, Chem. Eur. J. 3 (1997) 887.
[21] A. Trifonova, J.S. Diesen, C.J. Chapman, P.G. Andersson, Org.
Lett. 6 (2004) 3825.
[44] TEXSAN for Windows version 1.06: Crystal Structure Analysis
Package, Molecular Structure Corporation (1997–9), 1999.
[45] A.L. Spek, Acta Crystallogr. Sect. A 46 (1990) C34.
[46] A.L. Spek, J. Appl. Crystallogr. 36 (2003) 7.