Cyclometalation of (2 R,5 R)-2,5-diphenylpyrrolidine and 2-phenyl-2-imidazoline ligands with half-sandwich iridium(III) and rhodium(III) complexes

Article abstract of DOI:10.1021/om400626m

Cyclometalation of (2R,5R)-2,5-diphenylpyrrolidine (1) with [(η⁵-Cp)MCl₂]₂ using NaOAc in CH ₂Cl₂ at room temperature, followed by cationization with KPF₆ in CH₃CN, cleanly yield

Full text of DOI:10.1021/om400626m

Article
pubs.acs.org/Organometallics
Cyclometalation of (2R,5R)‑2,5-Diphenylpyrrolidine and 2‑Phenyl-2-
imidazoline Ligands with Half-Sandwich Iridium(III) and Rhodium(III)
Complexes
†
,†
†
‡
‡
Elias Feghali, Laurent Barloy,* Jean-Thomas Issenhuth, Lydia Karmazin-Brelot, Corinne Bailly,
́
,
†
and Michel Pfeffer*
†
Institut de Chimie de Strasbourg, UMR 7177 CNRS/Universite
Blaise Pascal, 67000 Strasbourg, France
́
de Strasbourg, Laboratoire de Synthes
̀
es Met
́
allo-Induites, 4 rue
‡
Institut de Chimie de Strasbourg, UMR 7177 CNRS/Universite
BP296/R8, 67008 Strasbourg Cedex, France
́
de Strasbourg, Service de Radiocristallographie, 1 rue Blaise Pascal,
*
S Supporting Information
ABSTRACT: Cyclometalation of (2R,5R)-2,5-diphenylpyrro-
5
lidine (1) with [(η -Cp*)MCl ] using NaOAc in CH Cl at
2
2
2
2
room temperature, followed by cationization with KPF in
6
CH CN, cleanly yielded the cationic cyclometalated amines
3
2a,b (a, M = Ir; b, M = Rh). This constituted an improvement
with regard to another cyclometalation pathway, which had led
to unwanted oxidation of the ligand to an imine. Compounds
2
a,b were characterized by elemental analysis and X-ray
diffraction of single crystals; the latter gave evidence of an R
configuration of the metal center, and a λ envelope
conformation of the five-membered metallacycle. Successful application of the NaOAc/CH Cl2 methodology to the
2
cycloiridation and cyclorhodation of the imidazolines 3−5 led to the neutral half-sandwich chloro complexes 6−8, which
were completely characterized, with yields ranging from 44% to 82%. In these complexes, the labile NH proton was localized in a
γ position with regard to the metal. From the enantiopure chiral ligands 4 and 5, both diastereoisomers of the four compounds
7
a,b and 8a,b were identified in solution by NMR; they were in equilibrium in solution. Complexes 6b, (S )-7a,b, and (R )-8a,b
M M
were characterized by radiocrystallography, which showed that the metallacycles were planar. Complexes 6−8 were tested as
catalysts for the reduction of acetophenone and imines by hydrogen transfer.
INTRODUCTION
led to some oxidation byproduct (cyclometalated imine)
together with the expected cyclometalated amine (Scheme 1).
Given the interesting catalytic properties of these latter
complexes, we considered it worthwhile to use a synthetic
pathway avoiding this oxidation. Following an initial report o7f
Beck et al. concerning the cyclometalation of oxazolones,
Davies et al. developed a synthetic method using sodium
acetate that was well adapted to the cyclometalation of amines,
imines, and oxazolines with several transition metals (Ru, Rh,
■
The field of cyclometalation using transition metals has
experienced a tremendous development over the latest few
decades, and cyclometalated complexes have found many
applications in the fields of homogeneous catalysis, energy
1
transfer, and antitumoral behavior, inter alia. More recently,
2
the concept of metal−ligand bifunctional catalysis, which was
introduced by Noyori et al., especially in the field of hydrogen
transfer, has been successfully applied to cyclometalated
complexes. We reported in 2007 the synthesis of cationic
8
Ir) through C−H activation. This method has since met great
success and has been applied to the cyclometalation of a wide
cyclometalated ruthenium(II), rhodium(III), and iridium(III)3
5a,9,10
range of nitrogen-containing ligands.
Thorough mecha-
chiral benzylamines in acetonitrile under mild conditions.
nistic investigations have been carried out, showing in particular
that the reaction occurs via ambiphilic metal ligand activation
Indeed, cyclometalated complexes bearing an α-NH function-
ality are efficient catalysts for hydrogen transfer reduction of
ketones and imines, as reported by us and others.
We have shown that cyclometalated secondary amines were
superior to primary amines in terms of enantioselectivity in
11
4
5
(AMLA). In the present paper, we first show that using
NaOAc in CH Cl at room temperature to cyclometalate
2
2
(
2R,5R)-2,5-diphenylpyrrolidine cleanly yields Cp* iridium and
4
c
rhodium amine complexes, almost no oxidation of the amine
having indeed been observed.
such catalytic reactions, with ee values up to 92%; therefore,
we focused our synthetic studies on this category of chiral
6
ligands. In 2011, we showed that the cyclometalation of
III
(
2R,5R)-2,5-diphenylpyrrolidine with [Cp*M Cl ] (M = Ir,
Received: June 28, 2013
2
2
Rh) at room temperature in acetonitrile using NaOH and KPF6
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/om400626m | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 1. Cyclometalation of (2R,5R)-2,5-Diphenylpyrrolidine using NaOH and KPF6
Scheme 2. Cyclometalation of (2R,5R)-2,5-Diphenylpyrrolidine using NaOAc and Subsequent Cationization of the
Cyclometalated Complex
We have also attempted to apply the same method to the
cyclometalation of novel categories of ligands. Actually, Ikariya
et al. have reviewed another kind of metal−ligand cooperative
bifunctional catalysis, which is related to labile protons that are
no longer located on the donor atom linked to the metal but
stimulated by their tremendous catalytic properties in the field
of asymmetric reduction of ketones and imines by hydrogen
transfer, with enantiomeric excesses reaching high levels.
4c
We performed the cyclometalation of enantiopure 1 at room
temperature with a slight excess of NaOAc in CH Cl under
2
2
12
are in β positions (typically pyrazole ligands). Similarly, a γ-
NH activity effect has been reported in ruthenium reduction
catalysts bearing a ligand with one or two benzimidazole
stoichiometric conditions (M:ligand ratio 1:1); evaporation to
dryness in vacuo followed by treatment with KPF in CH CN
6
3
20
at room temperature for cationization gave complexes 2a (M
= Ir) and 2b (M = Rh) in 67% and 71% yields, respectively
1
3
fragments. In 2007, Beller et al. published a paper mentioning
the use of pyridinebisimidazolines in the ruthenium-catalyzed
transfer hydrogenation of ketones, and they also observed a
dramatic NH effect, but this was more related to
(
Scheme 2). As previously reported, only one diastereomer of
complex 2a (respectively 2b) was detected by NMR. It is
interesting to point out that the secondary amine 1 could be
cyclorhodated, whereas the tertiary amine N,N-dimethylbenzyl-
14
enantioselectivity rather than reactivity. Indeed, imidazolines
are now widely used as ligands for manifold metal-catalyzed
8
,10
amine could not,
Rh being reputedly a “difficult” metal for
15
organic transformations. In addition, several examples of
cyclometalation of 2-aryl-substituted imidazoline ligands have
cyclometalation. Only trace amounts of the cyclometalated
pyrrolines were detected by NMR in the crude product;
recrystallization in CH Cl /pentane allowed us to eliminate
16
been reported, most of them with group 10 metals, yet also
2
2
1
7
18
with Ru and Ir; many of the resulting metal complexes have
met successful applications in catalysis as well. We present
herafter a detailed structural and stereochemical analysis of
cyclometalated iridium(III) and rhodium(III) complexes issued
from the cyclometalation of 2-phenyl-2-imidazolines and an
application of these compounds in asymmetric catalysis.
them easily. We had wondered why spontaneous dehydrogen-
ation of the C−N bond happened under the “NaOH/KPF /
6
CH CN” conditions and suggested that M(III) was the
3
6
oxidizing agent. Our new experiment supports this hypothesis,
because it seems reasonable that neutral complexes such as
5
2
[
M(η -Cp*)Cl(κ -OAc)] would be less oxidizing than electron-
5
2+
deficient cationic complexes such as [M(η -Cp*)(CH CN) ] .
3
3
RESULTS AND DISCUSSION
Compounds 2a,b had been characterized by NMR spectros-
copy and mass spectrometry; we have now completed their
characterization with elemental analysis and also crystallog-
raphy, as long as single crystals suitable for X-ray analysis could
■
Throughout this paper, we have used the cyclometalation
method reported by Davies et al., employing sodium acetate in
dichloromethane at room temperature, to generate cyclo-
iridated and cyclorhodated complexes either from (2R,5R)-2,5-
diphenylpyrrolidine or from a series of 2-phenyl-2-imidazo-
be grown from slow diffusion of pentane into a CH Cl
2
2
solution. Both of them display two independent molecules in
the unit cell, which differ in the conformation of the fused N-
containing five-membered rings (vide infra). One of them is
shown in Figure 1 for the iridium compound 2a (which is very
close to the second molecule of 2b in the unit cell) and the
other in Figure 2 for the rhodium compound 2b (which is,
conversely, very close to the second molecule of 2a in the unit
cellsee the Supporting Information). Selected metrical data
corresponding to these figures are given in Table 1 (see Table
S1 of the Supporting Information for selected data of all
crystallized molecules).
1
9
lines.
Cyclometalated (2R,5R)-2,5-Diphenylpyrrolidine. As
we had reported previously, we had not been able to separate
the products of cyclometalation of (2R,5R)-2,5-diphenylpyrro-
lidine (1) with [Cp*MCl ] (M = Ir, Rh) using NaOH and
2
2
KPF , either by chromatography or by any other purification
6
6
method. However, in spite of the complexity of the NMR
spectra of these mixtures, the configuration of the metal center
in all complexes had been unambiguously identified through
2D NMR analysis. Our interest in isolating complexes 2a,b was
B
dx.doi.org/10.1021/om400626m | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
However, in contrast with the cycloruthenated 2,5-
diphenylpyrrolidine, the five-membered metallacycles with Rh
and Ir are now no longer planar but are in a λ envelope
conformation. Probably, some steric interaction between the
bulky Cp* ligand and the unmetalated phenyl substituent
forces the metallacycle to adopt this conformation. Addition-
ally, in this conformation, N1−M1−N2 angles are much larger
(
2a, 92.61(14)°; 2b, 95.32(18)°; vs 85.7(4)°) as an effect of
intramolecular steric repulsion between Ph and the acetonitrile
ligand. It is probable that, in solution, the average conformation
of the complexes lies between these two, in agreement with
NMR data (JH7−NH = 7.3 Hz, slightly lower than the value
measured for the cycloruthenated complex) following a
3
Karplus-like interpretation.
Figure 1. Ortep plot of (R ,R ,R ,R )-2a. Thermal ellipsoids are
drawn at the 50% probability level. Hydrogen atoms and the PF6
anion are omitted for clarity.
C
C
N
Ir
As stated above, the following comments pertaining to the
conformation of 2a also apply to the second molecule of 2b and
vice versa. In 2a, the value of the dihedral angle C1−C6−C7−
N1 reaches −21.7(6)°, whereas it is only −10.4(7)° in 2b; as a
consequence, the angle C7−N1−M1 is smaller in 2a
−
(
108.9(3)°) than in 2b (112.6(3)°). The conformation of the
pyrrolidine ring is envelope in both cases; however, in 2a the
N1, C7, C8, and C9 atoms are coplanar (deviation <0.04 Å
from mean plane), while in 2b the N1, C7, C8, and C10 atoms
are coplanar, as in the cycloruthenated complex.
Cyclometalated 2-Phenyl-2-imidazolines. The cyclo-
metalation of 2-phenyl-2-imidazoline with [Cp*IrCl2]2 in
acetonitrile using NaOH and KPF has already been reported
6
18
by Feringa and de Vries. We have found out that various 2-
phenyl-2-imidazolines could also be readily cycloiridated and
8,10
cyclorhodated using sodium acetate in dichloromethane.
Figure 2. Ortep plot of (R ,R ,R ,R )-2b. Thermal ellipsoids are
drawn at the 50% probability level. Hydrogen atoms and the PF6
anion are omitted for clarity.
This procedure also has the advantage of yielding chloro-
coordinated complexes; in comparison to Rh or Ir cyclo-
metalated complexes coordinated with a labile acetonitrile
ligand, the fluxional processes which involve decoordination−
recoordination of the ancillary ligand are presumably slower,
thus avoiding broad signals on NMR spectra recorded at room
temperature. The imidazoline ligands include the enantiopure
compounds 4 and 5, in which the configuration of the two
carbon atoms is R.
C
C
N
Rh
−
Table 1. Selected Bond Distances (Å) and Angles (deg) for
a
Compounds 2a,b
2
a
2b
M1−C1
M1−N1
M1−N2
M1−centroid
C7−N1
C1−M1−N1
C1−M1−N2
N1−M1−N2
2.048(5)
2.154(4)
2.062(4)
1.834(2)
1.542(6)
77.81(17)
85.69(17)
92.61(14)
108.9(3)
2.021(5)
2.151(4)
2.085(6)
1.834(3)
1.525(7)
80.1(2)
Indeed, cyclometalation of ligands 3−5 with [Cp*MCl ] (M
2
2
b
= Ir, Rh; M:ligand ratio 1:1) at room temperature using a slight
excess of NaOAc in CH Cl led to complexes 6−8 with yields
2
2
ranging from 44% to 82% (Scheme 3), which is roughly the
range of yields that was reported for the cyclometalation of
88.5(2)
10
several nitrogen donor ligands. Kanbara et al. have noticed
that during the cycloruthenation of a pincer ligand with two
imidazoline units, using NH PF in 2-methoxyethanol without
95.32(18)
112.6(3)
C7−N1−M1
a
b
4
6
M = Ir, Rh. Centroid of the five Cp* carbon atoms coordinated to
17
degassing, they were oxidized into imidazoles. This oxidation
coupled with a cyclometalation is reminiscent of our own
the metal center.
Scheme 3. Cyclometalation of 2-Phenyl-2-imidazolines using
Sodium Acetate
The analysis of these X-ray data confirms the conclusions
that have been drawn from NMR data with regard to the
configuration of the amine nitrogen (R), and of the metal ion
6
(
R also). We note hereafter the decreasing priority sequence of
the entities linked to the metal in the pseudotetrahedral
5
environment of the half-sandwich complexes: (1) η -Cp*, (2)
21
CH CN, (3) Namine, (4) Caryl. The geometrical data measured
3
on the five-membered metallacyclic rings (see Table 1) are
roughly similar to those of the cycloruthenated (2R,5R)-2,5-
3b
diphenylpyrrolidine that we reported previously or to those of
related Cp*-coordinated rhodium and iridium cyclometalated
5a,22
benzylic primary amines.
C
dx.doi.org/10.1021/om400626m | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
6
observations in the field of cycloiridation. However, as in the
case of the (2R,5R)-2,5-diphenylpyrrolidine ligand, we observed
that oxidation was also avoided with imidazoline ligands when
the cyclometalation was run in CH Cl at room temperature
Scheme 4. Stereochemistry of the Major R Diastereomer of
Complex 7a
Ir
a
2
2
with NaOAc (even without degassing). During the preparation
of this paper, we became aware of a publication of Xiao et al.
who reported the synthesis of complexes 6a,b, 7a, and 8a,
among others, and the high efficiency of cycloiridated
imidazolines as catalysts for the dehydrogenation of formic
acid. However, we mention hereafter some additional
elements of the characterization of these compounds, especially
with regard to stereochemistry.
2
3
a
The double arrow represents a key NOE contact.
However, given the very close general similarity of our Ir and
Rh complexes, it is highly probable that the RRh diastereomer is
the major species present in solution.
Overall, complexes 6−8 were completely characterized by
1
13
1
D and 2D H and C NMR spectroscopy, electrospray mass
spectrometry, and elemental analysis. The usual characteristics
With regard to the determination of the configuration at the
metal of the diastereomers of complexes 8a,b, NOE analysis
was not conclusive: the proximity of H9 to Cp* was revealed by
intense cross-peaks for both isomers, whereas no Cp*/H8
cross-peaks were seen. The main difference between them was
the H8−H9 coupling constant (major isomer, 11.2 Hz; minor
isomer, 5.5 Hz), but this did not all allow us to determine
directly the configuration of the metal. However, the
diastereomers of an analogous ruthenium(II) complex
coordinated by (4S,5S)-2-(pyridin-2-yl)-4,5-diphenyl-2-imida-
3a,8
of cycloiridated and cyclorhodated aromatic nitrogen ligands
were observed in complexes 6−8: high-frequency H6 (7.78−
.90 ppm) and C1 (Ir complexes, 164.6−165.6 ppm; Rh
7
1
complexes, 178.9−180.3 ppm with
J
= 32−33 Hz)
CRh
1
13
resonances in 1D H and C NMR spectra, and key NOE
Cp*/H6 cross-peaks in 2D ROESY or NOESY spectra. We
also detected in these latter spectra some NH/H3 NOE cross-
peaks which prove that the labile proton is not borne by the
coordinating nitrogen but by the distal one, in a γ position with
regard to the metal; therefore, the CN bond is endo (i.e.
embedded in the metallacycle). This endo selectivity of the
cyclometalation reaction reflects our previous findings with
24
zoline could be identified; the coupling constants between the
benzylic protons were respectively 11.5 Hz for the S ,S ,S
C Ru
isomer and 6.5 Hz for the S ,S ,R isomer. By analogy, we
C
C
C
Ru
6
imine ligands. Along the same line, we note that the
deem it reasonable to assign the R ,R ,R configuration to the
C C M
cyclometalation of (4R,5R)-isoamarine 5 is regioselective: the
exo cyclometalated complexes that would have resulted from
metalation of the phenyl groups at the 4- and 5-positions are
not observed.
major isomer of 8a,b and the R ,R ,S configuration to the
C C M
minor isomer. We also tried to dissolve some single crystals of
(R ,R ,R )-8a (vide infra) in CDCl to confirm this, but
C
C
Ir
3
unfortunately the mixture of isomers was already at equilibrium
when the NMR spectrum was recorded. Indeed, this relatively
fast equilibrium was highlighted by exchange (positive) cross-
peaks on the ROESY or NOESY spectra of 7a,b and 8a,b
Like 2a,b, the half-sandwich complexes 6−8 are pseudote-
trahedral; the priority sequence for the determination of the
5
configuration of the metal center is (1) η -Cp*, (2) Cl, (3) N,
25
(
4) Caryl. The diastereoselectivity of the cyclometalation
between the signals of both isomers. Such EXSY cross-peaks
were also found between both H8 or H9 signals of 6a,b. In the
past, we have also observed that the diastereomers of some
reaction yielding complexes 7a,b and 8a,b has been determined
1
by integration of the respective H NMR signals of each isomer
26
3a
(
Table 2); their diastereomeric ratios are rather low (1.8−3.8),
cyclometalated half-sandwich chloro Ru or Rh complexes
were in equilibrium and under thermodynamic control, through
coordination−decoordination of the chloro ligand. This
conformation instability of the metal is confirmed by the
variation of the diastereomeric ratio of complex 8a with the
temperature (see the Supporting Information). The thermody-
namic data related to the equilibrium
which indicates that the diastereomers are not very much
differentiated by their intramolecular steric strain.
Table 2. Proportions of the Diastereomers of Compounds
a
7
a,b and 8a,b
compd
RM isomer (%)
SM isomer (%)
8
a (minor isomer)⥂8a (major isomer)
7
7
8
8
a
b
a
b
76
79
64
74
24
21
36
26
−1
−1
i.e. ΔH° = −4.7 ± 0.5 kJ mol and ΔS° = −11 ± 2 J mol
−1
K , have thus been determined from a ln K = f(1/T) curve
(
Figure S5, Supporting Information).
Single crystals of complexes 6b, 7a,b, and 8a,b suitable for X-
a
1
Determined by H NMR in CDCl at room temperature.
3
ray analysis were grown from slow diffusion of pentane into a
chloroform solution. Since an X-ray structure of complex 6b
In the ROESY spectrum of complex 7a, a strong NOE Cp*/
H9 cross-peak corresponding to the major diastereomer is
observed, together with a much weaker NOE Cp*/H8 cross-
peak. This fact is in agreement with an R configuration of the
metal center (Scheme 4); conversely, cross-peaks of roughly
equal intensities would have been expected in an SIr
diastereomer, as confirmed by the X-ray structure which is
depicted below. Unfortunately, in the ROESY spectrum of
complex 7b, these NOE cross-peaks are obscured because the
Cp* signals and some signals belonging to the cyclohexenyl
moiety accidentally resonate at the same chemical shift.
23
has been reported by Xiao et al. (although with a different
solvate and a different space group), with no major structural
differences with regard to ours, we provide it in the Supporting
Information. The isolated crystals of 7a,b contained the
R ,R ,S diastereomer (Figures 3 and 4). In contrast, in the
C
C M
case of 8a,b it was the R ,R ,R diastereomer that crystallized
C
C
M
(Figures 5 and 6); therefore, the latter have similarities with the
2
X-ray structure of (S ,S ,S )-[RuCl(C H ){κ (N,N)-(2-(2-
C
C
Ru
6
6
24
pyridyl)-4,5-diphenylimidazoline}]. Note, however, the re-
verse configuration (S ) of the iridium center observed in the
Ir
D
dx.doi.org/10.1021/om400626m | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 3. Ortep plot of (R ,R ,S )-7a. Thermal ellipsoids are drawn at
C
C Ir
the 50% probability level. Hydrogen atoms are omitted for clarity.
Figure 6. Ortep plot of (R ,R ,R )-8b. Thermal ellipsoids are drawn
C
C
Rh
at the 50% probability level. Hydrogen atoms are omitted for clarity.
characteristics of the Ir vs the Rh complex (7a vs 7b; 8a vs
8
b) were very close.
Selected bond distances and angles of the X-ray structures of
a,b and 8a,b are shown in Table 3. The M1−C1 bond
7
Table 3. Selected Bond Distances (Å) and Angles (deg) for
a
7a,b and 8a,b
iridium complex
rhodium complex
7
a
8a
7b
8b
M1−C1
M1−N1
M1−Cl1
M1−centroid
C7−N1
2.057(14)
2.062(10)
2.413(3)
1.824(7)
1.342(17)
1.356(18)
77.2(5)
2.052(6)
2.098(5)
2.4142(18)
1.822(3)
1.311(8)
1.339(9)
77.4(2)
2.111(14)
2.074(12)
2.416(3)
1.829(8)
1.311(17)
1.371(17)
79.1(5)
2.037(9)
2.133(12)
2.405(2)
1.824(5)
1.303(12)
1.346(14)
78.8(4)
Figure 4. Ortep plot of (R ,R ,S )-7b. Thermal ellipsoids are drawn
C
C Rh
b
at the 50% probability level. Hydrogen atoms and CHCl are omitted
3
for clarity.
C7−N2
C1−M1−N1
C1−M1−Cl1
N1−M1−Cl1
89.2(4)
82.6(2)
89.9(4)
84.4(3)
83.4(3)
87.63(16)
116.2(4)
88.5(3)
94.1(2)
C7−N1−M1
116.9(8)
114.3(8)
113.3(7)
a
b
M = Ir, Rh. Centroid of the five Cp* carbon atoms coordinated to
the metal center.
distances (2.03−2.11 Å) and the chelate C1−M1−N1 bite
angles (77−79°) fell approximately in the same range as for
10
Cp* Ir or Rh cyclometalated oxazolines and imidazoles or
6
imines. Similarly, the M1−Cl1 bond distances (2.40−2.42 Å)
were about the same as in the X-ray structures of analogous
10
complexes. The M1−N1 bond distances (2.06−2.13 Å)
compared well with crystallographically characterized ruthe-
24
27
15d,28
nium, rhodium, and iridium
imidazoline complexes;
however, we observed a markedly long bond length in the
isoamarine complexes 8a (2.098(5) Å) and 8b (2.133(12) Å),
due to the steric effect of the adjacent phenyl group on the
ligand. As in the structures of N-unsubstituted or N-alkyl
Figure 5. Ortep plot of (R ,R ,R )-8a. Thermal ellipsoids are drawn at
the 50% probability level. Hydrogen atoms are omitted for clarity.
C
C
Ir
2
4,27
imidazoline metal complexes in the literature,
the C7−N1
crystal structure of the complex resulting from the cyclo-
distances (1.30−1.34 Å) were only slightly shorter than the
C7−N2 distances (1.34−1.37 Å) in the five crystal structures,
which suggested that the π electrons are delocalized within the
amidine function. The endo double-bond character of the C7−
N1 bond was confirmed by the planarity (±0.05 Å) of the five-
membered metallacycles; it was in agreement with the γ
positioning of the NH function revealed by NMR data.
23
metalation of a derivative of (4R ,5R )-isoamarine. In-
C
C
cidentally, two crystallographically independent (but closely
similar) molecules of 7a,b were respectively present within the
asymmetric unit; only one molecule of each compound is
represented in Figures 3 and 4 and discussed in the following
section (see Table S3 of the Supporting Information for
selected data of all crystallized molecules). In general, for a
given imidazoline ligand (4 or 5), the crystallographic
With regard to the steric effects of the substituents in
positions 8 and 9 of the imidazoline ring, the structures of 7a,b
E
dx.doi.org/10.1021/om400626m | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
and 8a,b confirm what could be intuitively anticipated: i.e., the
important bulkiness of the phenyl groups in 8a,b in comparison
to the cyclohexenyl moiety in 7a,b. This is illustrated by a short
Cl1···C21 distance (8a, 3.392(7) Å; 8b, 3.45(1) Å). Although
this remark concerns only one diastereomer, we can conclude
that the proximity of the phenyl group with the chloro ligand
rhodium counterparts 6b−8b were far less efficient (yields 3−
40%).
The enantiomeric excesses were low (1−20%), either with
the ketone or with the imine as substrate. We observed a slight
superiority of the sterically more demanding isoamarine ligand
5 over the cyclohexenyl-fused ligand 4 and, in the case of
acetophenone, of rhodium over iridium. Conversely, we found
no obvious correlation between the ee’s resulting from the
catalytic tests and the diastereomeric ratios of the rhodium and
iridium complexes.
(
supposedly substituted with a hydride during the catalytic
process) may have an influence in asymmetric catalysis (vide
infra).
We have tested the cyclometalated imidazoline iridium and
rhodium complexes as catalysts for the transfer hydrogenation
of acetophenone using refluxing 2-propanol as a reducing agent
CONCLUSION
■
(
Table 4). The activity of the catalysts was disappointing: with
The cyclometalation methodology using sodium acetate in
methylene chloride developed by the Davies group proved to
be efficient for the clean cyclometalation of (2R,5R)-2,5-
diphenylpyrrolidine and of several 1-phenyl-1-imidazoline
ligands. It avoided the unwanted oxidation of the secondary
amine into an imine, a problem we had to face when (2R,5R)-
Table 4. Catalytic Transfer Hydrogenation of
a
Acetophenone
2
,5-diphenylpyrrolidine was cyclometalated in acetonitrile in
reaction time 4 h
reaction time 20 h
the presence of sodium hydroxide and potassium hexafluor-
ophosphate. The stereochemistry of the resulting cyclo-
6
b
c
b
c
d
entry catalyst yield (%)
ee (%)
yield (%)
ee (%)
dr
metalated complexes 2 and 6−8 has been thoroughly
investigated in solution (NMR spectroscopy) and in the solid
state (X-ray diffraction of single crystals). In particular, we
found evidence that the diastereoisomers of 7a,b and 8a,b,
which could be identified, were in equilibrium.
1
2
3
4
5
6
6a
7a
8a
6b
7b
8b
34
27
35
53
63
30
100
67
65
97
90
57
0
1
8
76:24
64:36
17
12
26
10
20
79:21
74:26
It is worth stressing that all cyclometalation reactions
reported in this paper were successful with both iridium(III)
and rhodium(III) complexes, although they are usually easier
with the former than with the latter. Indeed, several
cyclorhodation attempts failed under these conditions, among
which is the cyclorhodation of dimethylbenzylamine; it was
suggested that this could be linked to the absence of
a
Conditions: acetophenone (0.1 M) in 10 mL of 2-propanol,
substrate]:[catalyst]:[tBuOK] = 100:1:5, 82 °C. Determined by
H NMR. Determined by chiral GC; the configuration of the major
enantiomer is R. Diastereomeric ratio of the catalyst (see Table 2).
b
[
1
c
d
10
conversions ranging from 57 to 100% in 20 h at 82 °C, catalysts
delocalization within the metallacyclic ring. Therefore, it is
remarkable that the cyclorhodation of the pyrrolidine ligand
was successful in spite of the absence of such stabilizing
electronic delocalization.
6
−8 were less efficient than previously reported cyclometalated
4c,5a
primary or secondary benzylamines (M = Ir, Rh)
or than
1
3
14
benzimidazole- or imidazoline-coordinated Ru catalysts.
In contrast, the cycloiridated imidazolines 6a and 7a were
remarkably efficient catalysts of the transfer hydrogenation of
N-phenyl-(1-phenylethylidene)amine with formic acid−trie-
thylamine at room temperature, achieving quantitative
conversions of the imine within 2 h (Table 5). The bulkier
complex 8a was slightly less efficient (66% yield). These results
are obviously in agreement with the efficiency of such
In terms of catalytic reactivity in hydrogen transfer, it seems
doubtful that the poorly active cyclometalated imidazoline Rh
and Ir complexes with a γ-NH functionality would behave as
bifunctional hydrogen transfer catalysts as cyclometalated
4
,5
benzylic amines do. Concerning the complexes (2, 7, 8)
synthesized from enantiopure chiral ligands, obviously the steric
influence of the substituents borne by the asymmetric carbon is
much more significant on the pyrrolidine ligand than on the
imidazoline ligands. Indeed, the differentiation between the
diastereomers of the precatalyst (and presumably of the
hydridic active species) is such in the first case that only one
23
complexes as formic-acid-dehydrogenation catalysts. Their
Table 5. Catalytic Transfer Hydrogenation of N-Phenyl-(1-
phenylethylidene)amine
a
6
is present in solution, whereas in the second case both
diastereomers coexist in comparable amounts. Another point is
the enantioselectivities issued from catalytic reductions, which
4c
b
c
d
were remarkable with the pyrrolidine ligand but very low with
the imidazoline ligands. This latter result is in line with our
previous observations that the cyclometalated iridium and
rhodium pyrroline complexes (obtained via oxidation of the
pyrrolidine ligands) were poor catalysts for the enantioselective
reduction of prochiral ketones or imines as far as selectivity was
concerned. Indeed, in these latter compounds the single C−N
bond of the pyrrolidine was replaced by a CN bond, thus
removing the chiral information from the metallacyclic unit. We
may thus anticipate that the catalytic selectivity of the
cyclometalated compounds might well be related to the
existence of a chiral C atom within the metallacyclic unit,
entry
catalyst
yield (%)
ee (%)
dr
1
2
3
4
5
6
6a
7a
8a
6b
7b
8b
100
100
66
40
25
3
3
76:24
64:36
19
3
79:21
74:26
15
a
Conditions: imine (0.4 M) in a mixture of 2 mL CH Cl and 0.5 mL
HCOOH/Et N (1/1), [substrate]:[catalyst] = 100:1, RT, reaction
time 2 h. Determined by H NMR. Determined by HPLC.
Diastereomeric ratio of the catalyst (see Table 2).
2
2
3
b
1
c
d
F
dx.doi.org/10.1021/om400626m | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
since in all cases where this center of chirality did not exist
chiral information was poorly transferred to the reduced
ketones or imines.
NH), 4.0−3.7 (m, 2H, H9), 3.57 (m, 1H, H8), 3.04 (m, 1H, H8), 1.74
5
13
1
(
s, 15H, η -C (CH ) ). C{ H} NMR (100 MHz, CDCl ): δ 177.5
5
3
5
3
(
1
C7), 165.2 (C1), 136.3 (C6), 135.9 (C2), 131.5 (C5), 125.1 (C3),
5
5
22.0 (C4), 87.8 (η -C (CH ) ), 51.6 (C9), 45.1 (C8), 9.5 (η -
5 3 5
C (CH ) ).
5
3 5
+
EXPERIMENTAL SECTION
Analytical data for 7a: ES-MS: m/z (%) 527.20 (100) [M − Cl] .
■
Anal. Calcd for C H ClIrN ·1.5CH Cl : C, 42.67; H, 4.82; N, 4.06.
General Considerations. Experiments were carried out under
argon using a vacuum line and Schlenk techniques. Dichloromethane
and acetonitrile were distilled over calcium hydride under argon
immediately before use. Column chromatography was carried out on
Merck aluminum oxide 90 standardized. The following commercial
reagents were used as received: (2R,5R)-2,5-diphenylpyrrolidine,
sodium acetate, potassium hexafluorophosphate, 2-phenyl-2-imidazo-
line, sodium sulfate, cyclohexane, ethyl acetate, 2-propanol, acetophe-
none, potassium tert-butoxide, formic acid, and triethylamine. The
compounds listed hereafter were synthesized following reported
23 30
2
2
2
1
Found: C, 42.54; H, 4.82; N, 3.84. H NMR (500 MHz, CDCl ):
3
3
R ,R ,R diastereomer (76%), δ 7.78 (d, 1H, H6, J = 7.4), 7.21 (d,
C
C
Ir
3
3
3
1
5
3
1
2
H, H3, J = 7.4), 7.12 (t, 1H, H5, J = 7.4), 6.93 (t, 1H, H4, J = 7.3),
.53 (br, 1H, NH), 3.31 (ddd, 1H, H9, J = 14.2, J = 10.9, J = 3.1),
3
3
3
3
3
3
4
.06 (dddd, 1H, H8, J = 13.8, J = 10.9, J = 3.2, J = 3.0), 2.25 (dq,
2
3
2
3
H, CH , J = 12.0, J = 2.9), 2.06 (dq, 1H, CH , J = 11.1, J = 2.7),
2
2
5
.0−1.2 (m, 6H, CH ), 1.71 (s, 15H, η -C (CH ) ); R ,R ,S
2
5
3
5
C
C
Ir
3
diastereomer (24%), δ 7.90 (d, 1H, H6, J = 7.6), 7.21 (d, 1H, H3,
3
3
J = 7.4), 7.16 (t, 1H, H5, J = 7.5), 6.95 (m, 1H, H4), 5.64 (br, 1H,
29
3
3
3
NH), 3.45 (ddd, 1H, H9, J = 14.3, J = 11.0, J = 3.4), 3.38 (m, 1H,
procedures: [Cp*IrCl2]2 and [Cp*RhCl ] , (3aR,7aR)-2-phenyl-
2
2
3
0
2
3
2
H8), 2.44 (dq, 1H, CH , J = 11.6, J = 3.2), 2.12 (dq, 1H, CH , J =
3
a,4,5,6,7,7a-hexahydro-1H-benzo[d]imidazole (4), (4R,5R)-isoa-
2
2
3
1
32
3
5
1
1.6, J = 3.0), 2.0−1.2 (m, 6H, CH ), 1.74 (s, 15H, η -C (CH ) ).
marine (5), and N-phenyl-(1-phenylethylidene)amine. NMR
2
5
3 5
1
3
1
spectra were recorded at room temperature (unless otherwise stated)
C{ H} NMR (125 MHz, CDCl
3
): R ,R ,RIr diastereomer, δ 178.2
C C
1
on Bruker Avance spectrometers. H NMR spectra were recorded at
(C7), 164.9 (C1), 136.1 (C6), 135.1 (C2), 131.6 (C5), 124.7 (C3),
1
3
1
5
3
00.17, 400.13, or 500.13 MHz and referenced to SiMe4. C{ H}
NMR spectra (broad band decoupled) were recorded at 100.62,
25.77, or 150.92 MHz and referenced to SiMe . The NMR
121.4 (C4), 87.5 (η -C
29.9 (CH ), 25.0 (CH
diastereomer, δ 178.3 (C7), 165.3 (C1), 135.8 (C6), 135.3 (C2),
5
(CH
3 5 2
), 24.4 (CH
) ), 70.0 (C9), 68.2 (C8), 30.7 (CH ),
5
), 9.8 (η -C (CH ) ); R ,R ,SIr
2 5 3 5 C C
2
2
1
4
5
assignments were supported by 2D spectra: COSY, NOESY, and/or
131.3 (C5), 124.6 (C3), 121.5 (C4), 87.5 (η -C
66.2 (C8), 31.2 (CH ), 30.1 (CH ), 24.9 (CH ), 24.5 (CH
(CH ).
Compound 8a. To a solution of [Cp*IrCl
mmol) in CH Cl (10 mL) were added the imidazoline ligand 5 (75
5 3 5
(CH ) ), 71.6 (C9),
1
13
1
5
ROESY for H NMR and HSQC and HMBC for C{ H} NMR.
Chemical shifts are reported in ppm and coupling constants in Hz; the
latter are proton−proton coupling constants unless otherwise stated.
The 13C{ H} signals are singlets unless otherwise stated. For the
adopted numbering scheme, see Scheme 3. Multiplicity: s, singlet; d,
doublet; t, apparent triplet; q, apparent quartet; m, multiplet; br, broad
signal. ES-MS spectra and elemental analyses were carried out by the
2
2
2
), 10.0 (η -
2
C
)
3 5
5
] (100 mg, 0.126
2 2
1
2
2
mg, 0.25 mmol) and NaOAc (25 mg, 0.30 mmol). The reaction
mixture was stirred at room temperature for 4 h and then filtered over
Celite. The filtrate was washed with water (2 × 5 mL), dried over
anhydrous Na SO , and evaporated in vacuo. Compound 8a was
isolated as a yellow-orange powder (135 mg, 82% yield).
corresponding facilities at the Institut de Chimie, Universite
Strasbourg.
́
de
2 4
+
Compound 2a. To a solution of [Cp*IrCl2]2 (90 mg, 0.113
mmol) in CH Cl (9 mL) were added ligand 1 (50 mg, 0.226 mmol)
ES-MS: m/z (%) 625.21 (100) [M − Cl] . Anal. Calcd for
31
1
C
H
32ClIrN
· /
CH
Cl
: C, 55.07; H, 4.81; N, 4.11. Found: C, 55.22;
2
2
2
4
2
2
1
and NaOAc (24 mg, 0.293 mmol). The reaction mixture was stirred at
room temperature for 18 h, and then the resulting orange solution was
evaporated in vacuo. KPF (95 mg, 0.516 mmol) and CH CN (5 mL)
H, 4.98; N, 4.01. H NMR (400 MHz, CDCl ): R ,R ,RIr diastereomer
(64%), δ 7.81 (d, 1H, H6, J = 7.5), 7.52 (d, 2H, HPh ortho, J = 7.3),
3 C C
3
3
3
7.6−7.2 (m, 10H, H3 + H5 + HPh), 7.00 (t, 1H, H4, J = 7.5), 5.68 (br
6
3
3
3
3
were added. The reaction mixture was stirred at room temperature for
h and was then chromatographed over Celite and alumina using
CH CN as eluent. Compound 2a was isolated as a yellow solid after
d, 1H, NH, J = 2), 5.01 (d, 1H, H9, J = 11.3), 4.86 (dd, 1H, H8, J =
3
5
2
11.3, J = 2.1), 1.46 (s, 15H, η -C (CH ) ); R ,R ,SIr diastereomer
5 3 5 C C
3
3
(36%), δ 7.86 (d, 1H, H6, J = 7.7), 7.46 (d, 2H, HPh ortho, J = 7.5),
3
3
recrystallization in CH Cl /pentane (111 mg, 67% yield).
7.6−7.2 (m, 10H, H3 + H5 + HPh), 7.03 (t, 1H, H4, J = 7.5), 5.48 (br,
2
2
3
3
Mp: 78 °C dec. Anal. Calcd for C H F IrN P: C, 45.71; H, 4.66;
1H, NH), 5.14 (d, 1H, H9, J = 5.5), 4.71 (d, 1H, H8, J = 5.6), 1.44
2
8
34
6
2
5
13
1
N, 3.81. Found: C, 46.39; H, 5.05; N, 3.77.
Compound 2b. To a solution of [Cp*RhCl ] (70 mg, 0.113
(s, 15H, η -C (CH ) ). C{ H} NMR (150 MHz, CDCl ): R ,R ,RIr
5 3 5 3 C C
diastereomer, δ 177.6 (C7), 165.6 (C1), 140.1 (CPh ipso), 139.3
(CPh ipso), 136.0 (C6), 134.2 (C2), 132.5 (C5), 129.0 (CHPh), 128.9
2
2
mmol) in CH Cl (9 mL) were added ligand 1 (50 mg, 0.226 mmol)
2
2
and NaOAc (24 mg, 0.293 mmol). The reaction mixture was stirred at
room temperature for 18 h, and then the resulting red-orange solution
was evaporated in vacuo. KPF (95 mg, 0.516 mmol) and CH CN (5
(CHPh), 128.6 (CHPh), 128.3 (CHPh), 128.2 (CHPh), 127.4 (CHPh),
5
124.7 (C3), 121.6 (C4), 87.5 (η -C
(CH
)
), 79.7 (C9), 72.2 (C8),
5
5
3
5
9.3 (η -C (CH ) ); R ,R ,SIr diastereomer, δ 176.7 (C7), 164.6 (C1),
6
3
5 3 5 C C
mL) were added. The reaction mixture was stirred at room
temperature for 2 h and was then chromatographed over Celite and
143.7 (CPh ipso), 142.0 (CPh ipso), 136.3 (C6), 135.0 (C2), 131.5 (C5),
129.1 (CH ), 129.0 (CH ), 128.7 (CH ), 128.2 (CH ), 127.8
Ph
Ph
Ph
Ph
5
alumina using CH CN as eluent. Compound 2b was isolated as an
(CH ), 127.5 (CH ), 124.8 (C3), 121.7 (C4), 88.0 (η -C (CH ) ),
3
Ph Ph 5 3 5
5
orange solid after recrystallization in CH Cl /pentane (103 mg, 71%
79.5 (C9), 72.5 (C8), 9.5 (η -C (CH ) ).
5 3 5
2
2
yield).
Mp: 100 °C dec. Anal. Calcd for C H F N PRh· / CH Cl : C,
Compounds 6b and 7b. To a solution of [Cp*RhCl ] (300 mg,
2 2
1
0.485 mmol) in CH Cl (30 mL) were added the imidazoline ligand
28
34
6
2
3
2
2
2
2
5
0.43; H, 5.18; N, 4.15. Found: C, 50.08; H, 5.41; N, 4.09.
Compounds 6a and 7a. To a solution of [Cp*IrCl ] (300 mg,
(0.970 mmol; 3, 142 mg; 4, 194 mg) and NaOAc (98 mg, 1.19 mmol).
The reaction mixture was stirred at room temperature for 4 h and then
filtered over Celite. The filtrate was washed with water (2 × 50 mL),
dried over anhydrous Na SO , and evaporated in vacuo. The
2
2
0
.377 mmol) in CH Cl (30 mL) was added the imidazoline ligand
2
2
(
0.750 mmol; 3: 110 mg, 4: 150 mg) and NaOAc (74 mg, 0.90 mmol).
2
4
The reaction mixture was stirred at room temperature for 4 h and then
compounds were isolated as red powders (6b, 190 mg, 47% yield;
7b, 230 mg, 50% yield).
filtered over Celite. The filtrate was washed with water (2 × 50 mL),
dried over anhydrous Na SO4 and evaporated in vacuo. The
Analytical data for 6b: Mp: > 250 °C dec. ES-MS: m/z (%) 383.10
2
+
compounds were isolated as yellow-orange powders (6a: 167 mg,
(100) [M − Cl] . Anal. Calcd for C H ClN Rh: C, 54.49; H, 5.77;
19
24
2
1
4
4% yield; 7a: 294 mg, 70% yield).
N, 6.68. Found: C, 54.13; H, 5.88; N, 6.61. H NMR (500 MHz,
CDCl ): δ 7.80 (d, 1H, H6, J = 7.5), 7.30 (d, 1H, H3, J = 7.5), 7.20
(t, 1H, H5, J = 7.5), 6.98 (t, 1H, H4, J = 7.4), 6.08 (br, 1H, NH),
3
3
Analytical data for 6a: Mp: > 250 °C dec. ES-MS: m/z (%) 473.15
3
+
1
3
3
(
100) [M − Cl] . Anal. Calcd for C H ClIrN · / CH Cl : C, 43.29;
19
24
2
3
2
2
1
H, 4.64; N, 5.22. Found: C, 43.46; H, 4.85; N, 4.76. H NMR (300
3.91 (m, 1H, H9), 3.74 (m, 1H, H9), 3.41 (m, 1H, H8), 2.63 (m, 1H,
3
3
5
13
1
MHz, CDCl ): δ 7.80 (d, 1H, H6, J = 7.5), 7.30 (d, 1H, H3, J = 7.5),
H8), 1.65 (s, 15H, η -C (CH ) ). C{ H} NMR (125 MHz, CDCl ):
3
5
3
5
3
3
3
1
2
7.15 (t, 1H, H5, J = 7.5), 6.94 (t, 1H, H4, J = 7.5), 5.79 (br, 1H,
δ 178.9 (d, C1, J = 32), 172.7 (d, C7, J = 3), 136.5 (C6), 135.5
CRh
CRh
dx.doi.org/10.1021/om400626m | Organometallics XXXX, XXX, XXX−XXX
G

Organometallics
Article
5
1
(
=
C2), 130.4 (C5), 124.8 (C3), 122.1 (C4), 94.7 (d, η -C (CH ) , J
determined by HPLC using a chiral column (Daicel Chiralpak OD)
5
3
5
CRh
+
5
7), 51.9 (C9), 44.6 (C8), 9.6 (η -C (CH ) ).
under the following conditions: T = 20 °C, eluent hexane/iPrOH (99/
5
3 5
−1
Analytical data for 7b: ES-MS: m/z (%) 437.15 (100) [M − Cl] .
1), flow rate 0.5 mL min , λ 204 and 244 nm.
2
Anal. Calcd for C H ClN Rh· / CH Cl : C, 53.69; H, 5.96; N, 5.29.
X-ray Crystallography. The crystals of [Ru(C H )(OAc)Cl]
23
30
2
3
2
2
6
6
1
Found: C, 53.33; H, 6.02; N, 5.15. H NMR (500 MHz, CDCl ):
(orange plates), 2a (yellow plates), 2b (yellow needles), 6b (orange
plates), 7a (orange blocks), 7b (orange prisms), 8a (orange prisms),
and 8b (orange needles) were placed in oil; a single crystal was
selected, mounted on a glass fiber, and placed in a low-temperature N2
stream. For the compounds [Ru(C H )(OAc)Cl], 2a,b and 8b, the X-
3
3
R ,R ,R diastereomer (79%), δ 7.79 (d, 1H, H6, J = 7.6), 7.25−7.10
m, 2H, H3 + H5), 6.97 (t, 1H, H4, J = 7.3), 5.53 (br, 1H, NH), 3.39
ddd, 1H, H9, J = 13.6, J = 11.7, J = 3.0), 2.96 (dddd, 1H, H9, J =
4.1, J = 10.9, J = 3.0, J = 2.8), 2.34 (dq, 1H, CH , J = 12.5, J =
.1), 2.08 (dq, 1H, CH , J = 11.8, J = 2.9), 2.0−1.2 (m, 6H, CH ),
.63 (s, 15H, η -C (CH ) ); R ,R ,S diastereomer (21%), δ 7.90 (d,
H, H6, J = 7.7), 7.25−7.10 (m, 2H, H3 + H5), 6.97 (t, 1H, H4, J =
.3), 5.57 (br, 1H, NH), 3.46 (ddd, 1H, H9, J = 14.8, J = 11.1, J =
C
C
Rh
3
(
(
3
3
3
3
3
3
4
2
3
1
3
1
1
7
3
2
6
6
2
3
ray diffraction data collection was carried out on a Bruker APEX II
DUO Kappa-CCD diffractometer equipped with an Oxford
2
2
5
5
3
5
C
C Rh
3
3
Cryosystem liquid N device, using Mo Kα radiation (λ = 0.71073
2
3
3
3
Å). The crystal to detector distance was 38 mm; the cell parameters
3
3
3
4
33
.1), 3.29 (dddd, 1H, H9, J = 13.9, J = 11.3, J = 3.4, J = 2.0), 2.52
dq, 1H, CH , J = 11.6, J = 3.2), 2.12 (m, 1H, CH ), 2.0−1.2 (m,
H, CH ), 1.67 (s, 15H, η -C (CH ) ). C{ H} NMR (125 MHz,
were determined (APEX2 software) from reflections taken from 3
2
3
(
6
sets of 12 frames, each at 10 s exposure. For compounds 6b, 7a,b and
8a, the X-ray diffraction data collection was carried out on a Nonius
Kappa-CCD diffractometer equipped with an Oxford Cryosystem
2
2
5
13
1
2 5 3 5
1
CDCl ): R ,R ,R diastereomer, δ 180.0 (d, C1, J = 33), 174.4 (d,
3
C
C
Rh
CRh
2
C7, J = 3), 137.1 (C6), 135.5 (C2), 130.7 (C5), 124.4 (C3), 122.2
C4), 95.0 (d, η -C (CH ) , J = 7), 69.7 (C9), 67.8 (C8), 30.7
CH ), 30.2 (CH ), 25.1 (CH ), 24.5 (CH ), 9.8 (η -C (CH ) );
R ,R ,S diastereomer, δ 179.9 (d, C1, J 33), 173.5 (d, C7, J
3), 136.6 (C6), 135.4 (C2), 130.6 (C5), 124.3 (C3), 122.2 (C4),
= 7), 72.3 (C9), 66.0 (C8), 31.5 (CH ),
0.1 (CH ), 25.0 (CH ), 24.5 (CH ), 10.1 (η -C (CH ) ).
liquid N device, using Mo Kα radiation (λ = 0.71073 Å). The crystal
2
CRh
5
1
(
(
to detector distance was 36 mm; the cell parameters were determined
5
3
5
CRh
5
34
(Denzo software) from reflections taken from 1 set of 10 frames
2
2
2
2
5
3
5
1
2
(1.0° steps in ϕ angle), each at 20 s exposure. The structures were
C
C
Rh
CRh
CRh
35
=
9
3
solved by direct methods using the program SHELXS-97. The
5
1
5.1 (d, η -C (CH ) , J
refinement and all further calculations were carried out using
5
3
5
CRh
2
5
36
SHELXL-97. The H atoms were included in calculated positions
2
2
2
5
3 5
Compound 8b. To a solution of [Cp*RhCl ] (100 mg, 0.16
and treated as riding atoms using SHELXL default parameters. The
2
2
mmol) in CH Cl (10 mL) were added the imidazoline ligand 5 (96
non-H atoms were refined anisotropically using weighted full-matrix
2
2
2
mg, 0.32 mmol) and NaOAc (34 mg, 0.41 mmol). The reaction
mixture was stirred at room temperature for 4 h and then filtered over
Celite. The filtrate was washed with water (2 × 5 mL), dried over
anhydrous Na SO , and evaporated in vacuo. The solid residue was
then chromatographed over alumina using cyclohexane/ethyl acetate
7/3) as eluent. Compound 8b was isolated as a red powder after
evaporation in vacuo (125 mg, 68% yield).
least squares on F . For the compounds [Ru(C
6
H
6
)(OAc)Cl], 2a,b
and 8b, a semiempirical absorption correction was applied using
3
4
SADABS in APEX2: transmission factors Tmin/Tmax = 0.5383/0.9057
(Ru(C )(OAc)Cl), 0.3230/0.7929 (2a), 0.7126/0.9856 (2b),
H
6
2
4
6
0.8390/0.9308 (8b). For compounds 6b, 7a,b, and 8a, a semiempirical
(
absorption correction was applied using the MULscanABS routine in
3
7
PLATON: transmission factors Tmin/Tmax = 0.72321/0.78263 (6b),
0.39687/0.56892 (7a), 0.77993/0.87298 (7b), 0.37587/0.46721 (8a).
Details of data collection parameters and refinement results are given
in the Supporting Information. For compounds 7a and 8b, the
+
ES-MS: m/z (%) 535.16 (100) [M − Cl] . Anal. Calcd for
C H ClN Rh: C, 65.21; H, 5.65; N, 4.91. Found: C, 65.81; H, 6.09;
31
32
2
1
N, 4.75. H NMR (500 MHz, CDCl ): R ,R ,R diastereomer (74%),
3
C
C
Rh
3
7
3
3
δ 7.82 (d, 1H, H6, J = 7.6), 7.55 (d, 2H, H
(
, J = 7.1), 7.6−7.2
SQUEEZE instruction in PLATON was applied. The residual
electron density was assigned to several disordered molecules of
solvent. Details of data collection parameters and refinement results
are given in the Supporting Information.
Ph ortho
4
3
m, 9H, H5 + H ), 7.19 (dd, 1H, H3, J = 7.5, J = 1.1), 7.01 (td, 1H,
Ph
3
4
3
3
H4, J = 7.5, J = 1.0), 5.67 (br d, 1H, NH, J = 2), 5.09 (d, 1H, H9, J
11.2), 4.75 (dd, 1H, H8, J = 11.2, J = 2.3), 1.41 (s, 15H, η -
C (CH ) ); R ,R ,S diastereomer (26%):, δ 7.90 (d, 1H, H6, J =
.7), 7.6−7.2 (m, 12H, H3 + H5 + H ), 7.05 (td, 1H, H4, J = 7.5, J
1.0), 5.44 (br, 1H, NH), 5.21 (d, 1H, H9, J = 5.6), 4.66 (d, 1H, H8,
J = 5.7), 1.40 (s, 15H, η -C (CH ) ). C{ H} NMR (125 MHz,
3
3
5
=
3
5
3
5
C
C Rh
3
4
7
ASSOCIATED CONTENT
Supporting Information
CIF files and figures and tables giving crystallographic data for
Ph
■
3
=
*
S
3
5
13
1
5
3 5
1
CDCl ): R ,R ,R diastereomer, δ 180.3 (d, C1, J = 32), 173.0 (d,
3
C
C
Rh
CRh
2
[Ru(C
H )(OAc)Cl], 2a,b, 6b, 7a,b, and 8a,b, X-ray diffraction
6 6
C7, J
= 3), 140.3 (CPh ipso), 139.6 (CPh ipso), 137.0 (C6), 134.1
C2), 131.7 (C5), 129.0 (CH ), 128.7 (CH ), 128.5 (CH ), 128.3
CH ), 127.8 (CH ), 127.4 (CH ), 124.2 (C3), 122.3 (C4), 95.0
d, η -C (CH ) , J = 7), 78.9 (C9), 72.0 (C8), 9.4 (η -C (CH ) );
R ,R ,S diastereomer, δ 178.9 (d, C1, J = 33), 172.1 (d, C7,
JCRh = 3), 144.1 (CPh ipso), 142.0 (CPh ipso), 136.8 (C6), 135.0 (C2),
30.9 (C5), 129.1 (CH ), 129.0 (CH ), 128.6 (CH ), 128.2
CRh
data, bond distances and angles, an ortep plot of [Ru(C H )-
(
(
(
6
6
Ph
Ph
Ph
(
OAc)Cl], supplementary details on the crystal structures of
Ph
Ph
Ph
5
1
5
2a,b, 6b, 7a,b, and 8a,b, and a plot of the temperature-
dependent diastereomeric ratio of 8a. This material is available
free of charge via the Internet at http://pubs.acs.org.
5
3
5
CRh
5
3 5
1
C
C
Rh
CRh
2
1
Ph
Ph
Ph
(
(
CH ), 127.6 (CH ), 127.5 (CH ), 124.5 (C3), 122.4 (C4), 95.0
Ph Ph Ph
AUTHOR INFORMATION
Corresponding Authors
L.B.: fax, +33 368 85 15 26; e-mail, barloy@unistra.fr.
*M.P.: e-mail, pfeffer@unistra.fr.
5
1
5
■
*
d, η -C (CH ) , J = 7), 79.7 (C9), 72.2 (C8), 9.6 (η -C (CH ) ).
5
3
5
CRh
5
3 5
Typical Procedure for the Catalytic Transfer Hydrogenation
of Acetophenone. The catalyst (10 μmol) was dissolved in 2-
propanol (10 mL) under argon, and acetophenone (120 mg, 1 mmol)
was added, followed by tBuOK (5.6 mg, 50 μmol); the mixture was
then stirred at 82 °C. Aliquots were filtered over alumina using
CH Cl as eluent. The conversion of the reaction was monitored by
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
2
2
1
H NMR. The ee values were determined by GC using a chiral
capillary column (Restek-βDEXsa, 30 m × 0.25 mm).
Typical Procedure for the Catalytic Transfer Hydrogenation
of Imines. To a stirred solution of the imine substrate (1 mmol) and
the catalyst (10 μmol) in CH Cl (2 mL) was added a freshly prepared
Notes
The authors declare no competing financial interest.
2
2
mixture of HCO H/Et N (1/1 molar ratio, 0.5 mL). The solution was
stirred at room temperature. Aliquots were filtered over alumina using
2
3
ACKNOWLEDGMENTS
We thank Nicolas Renard for technical assistance and
■
CH Cl as eluent, treated with a saturated aqueous Na CO solution,
2
2
2
3
and then washed with brine and dried over MgSO . The conversion of
Christophe Michon and Cel
the ee values of the amines in Table 5. We are grateful to the
́
ine Delabre for having determined
4
1
the reaction was monitored by H NMR. The ee values were
H
dx.doi.org/10.1021/om400626m | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
CNRS, the Laboratory of Excellence (LABEX) “Chemistry of
Complex Systems” (UdS), the FRC (UdS) through the project
X. Eur. J. Inorg. Chem. 2007, 132−139. (c) Liu, H.; Du, D.-M. Adv.
Synth. Catal. 2009, 351, 489−519 and references therein. (d) Franzke,
A.; Pfaltz, A. Chem. Eur. J. 2011, 17, 4131−4144. (e) Mei, L.; Yuan, Z.;
Shi, M. Organometallics 2011, 30, 6466−6475. (f) Yuan, Z.; Mei, L.;
Wei, Y.; Shi, M.; Kattamuri, P. V.; McDowell, P.; Li, G. Org. Biomol.
Chem. 2012, 10, 2509−2513.
“
0
Synergie”, and the ANR (PhotoBioMet, ANR-09-BLAN-0191-
1, postdoc fellowship to J.-T.I.) for financial support of this
work.
(
16) (a) Navarro-Ranninger, C.; Zamora, F.; Lop
Monge, A.; Masaguer, J. R. J. Organomet. Chem. 1996, 506, 149−
54. (b) Weiss, M. E.; Fischer, D. F.; Xin, Z.; Jautze, S.; Schweizer, W.
́
ez-Solera, I.;
REFERENCES
■
1
(
1) (a) Djukic, J.-P.; Sortais, J.-B.; Barloy, L.; Pfeffer, M. Eur. J. Inorg.
Chem. 2009, 817−853. (b) Albrecht, M. Chem. Rev. 2010, 110, 576−
B.; Peters, R. Angew. Chem., Int. Ed. 2006, 45, 5694−5698. (c) Niu, J.-
L.; Hao, X.-Q.; Gong, J.-F.; Song, M.-P. Dalton Trans. 2011, 40, 5135−
623.
5
150 and references therein. (d) Yang, M.-J.; Liu, Y.-J.; Gong, J.-F.;
(
2) (a) Noyori, R.; Yamakawa, M.; Hashiguchi, S. J. Org. Chem. 2001,
Song, M.-P. Organometallics 2011, 30, 3793−3803. (e) Hao, X.-Q.; Xu,
Y.-X.; Yang, M.-J.; Wang, L.; Niu, J.-L.; Gong, J.-F.; Song, M.-P.
Organometallics 2012, 31, 835−846.
6
6, 7931−7944. (b) Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord.
Chem. Rev. 2004, 248, 2201−2237. (c) Ikariya, T.; Blacker, A. J. Acc.
Chem. Res. 2007, 40, 1300−1308.
(17) Maeda, S.; Koizumi, T.; Yamamoto, T.; Tanaka, K.; Kanbara, T.
(
3) (a) Sortais, J.-B.; Pannetier, N.; Holuigue, A.; Barloy, L.; Sirlin,
J. Organomet. Chem. 2007, 692, 5495−5500.
C.; Pfeffer, M.; Kyritsakas, N. Organometallics 2007, 26, 1856−1867.
b) Sortais, J.-B.; Pannetier, N.; Clement, N.; Barloy, L.; Sirlin, C.;
Pfeffer, M.; Kyritsakas, N. Organometallics 2007, 26, 1868−1874.
4) (a) Sortais, J.-B.; Ritleng, V.; Voelklin, A.; Holuigue, A.; Smail,
(
18) Jerphagnon, T.; Gayet, A. J. A.; Berthiol, F.; Ritleng, V.; Mrsic,
̌
́
(
́
N.; Meetsma, A.; Pfeffer, M.; Minnaard, A. J.; Feringa, B. L.; de Vries,
J. G. Chem. Eur. J. 2009, 15, 12780−12790.
(
(19) In addition, the unsuccessful attempt of cycloruthenation of
H.; Barloy, L.; Sirlin, C.; Verzijl, G. K. M.; Boogers, J. A. F.; de Vries,
another ligand under these conditions has led to single crystals of the
complex [Ru(C H )(OAc)Cl] from evaporation of a chloroform
A. H. M.; de Vries, J. G.; Pfeffer, M. Org. Lett. 2005, 7, 1247−1250.
6
6
(
b) Sortais, J. B.; Barloy, L.; Sirlin, C.; de Vries, A. H. M.; de Vries, J.
solution to dryness. Analysis by X-ray diffraction (see the Supporting
G.; Pfeffer, M. Pure Appl. Chem. 2006, 78, 457−462. (c) Pannetier, N.;
Sortais, J.-B.; Issenhuth, J.-T.; Barloy, L.; Sirlin, C.; Holuigue, A.;
Lefort, L.; Panella, L.; de Vries, J. G.; Pfeffer, M. Adv. Synth. Catal.
Information) showed that its structure is closely similar to the reported
8
structure of [Ru(p-cymene)(OAc)Cl]. The synthesis of [Ru(C H )
6
6
(OAc)Cl] has been previously reported: Tocher, D. A.; Gould, R. O.;
2
(
011, 353, 2844−2852.
Stephenson, T. A.; Bennett, M. A.; Ennett, J. P.; Matheson, T. W.;
5) (a) Arita, S.; Koike, T.; Kayaki, Y.; Ikariya, T. Organometallics
Sawyer, L.; Shah, V. K. J. Chem. Soc., Dalton Trans. 1983, 1571−1581.
2
008, 27, 2795−2802. (b) Baratta, W.; Ballico, M.; Del Zotto, A.;
(20) Davies, D. L.; Al-Duaij, O.; Fawcett, J.; Singh, K. Organometallics
Herdtweck, E.; Magnolia, S.; Peloso, R.; Siega, K.; Toniutti, M.;
2
(
010, 29, 1413−1420.
Zangrando, E.; Rigo, P. Organometallics 2009, 28, 4421−4430.
21) (a) Lecomte, C.; Dusausoy, Y.; Protas, J.; Tirouflet, J.;
(
c) Wang, C.; Pettman, A.; Basca, J.; Xiao, J. Angew. Chem., Int. Ed.
Dormond, A. J. Organomet. Chem. 1974, 73, 67−76. (b) Brunner, H.
2
(
010, 49, 7548−7552.
6) Barloy, L.; Issenhuth, J.-T.; Weaver, M. G.; Pannetier, N.; Sirlin,
C.; Pfeffer, M. Organometallics 2011, 30, 1168−1174.
7) Bauer, W.; Prem, M.; Polborn, K.; Sunkel, K.; Steglich, W.; Beck,
W. Eur. J. Inorg. Chem. 1998, 485−493.
8) Davies, D. L.; Al-Duaij, O.; Fawcett, J.; Giardiello, M.; Hilton, S.
T.; Russell, D. R. Dalton Trans. 2003, 4132−4138.
9) (a) Davies, D. L.; Donald, S. M. A.; Al-Duaij, O.; Fawcett, J.;
Little, C.; Macgregor, S. A. Organometallics 2006, 25, 5976−5978.
b) Scheeren, C.; Maasarani, F.; Hijazi, A.; Djukic, J.-P.; Pfeffer, M.;
Zaric, S.; Le Goff, X.-F.; Ricard, L. Organometallics 2007, 26, 3336−
345. (c) Li, L.; Brennessel, W. W.; Jones, W. D. J. Am. Chem. Soc.
Enantiomer 1997, 2, 133−134.
(22) (a) Arita, S.; Koike, T.; Kayaki, Y.; Ikariya, T. Chem. Asian J.
2
008, 3, 1479−1485. (b) Heiden, Z. M.; Rauchfuss, T. B. J. Am. Chem.
Soc. 2006, 128, 13048−13049.
23) Barnard, J. H.; Wang, C.; Berry, N. G.; Xiao, J. Chem. Sci. 2013,
, 1234−1244.
24) Davenport, A. J.; Davies, D. L.; Fawcett, J.; Russell, D. R. J.
Chem. Soc., Perkin Trans. 1 2001, 1500−1503.
25) This equilibrium and a thorough interpretation of 2D NMR
(
̈
(
4
(
(
(
(
(
spectra (COSY, ROESY, HSQC, and HMBC) unfortunately do not
support the interpretation provided in ref 23 concerning the isomers of
compound 8a.
́
3
2
008, 130, 12414−12419. (d) Li, L.; Brennessel, W. W.; Jones, W. D.
(
26) Ritleng, V.; Bertani, P.; Pfeffer, M.; Sirlin, C.; Hirschinger, J.
Inorg. Chem. 2001, 40, 5117−5122.
27) Bastero, A.; Claver, C.; Ruiz, A.; Castillon
Zangrando, E. Chem. Eur. J. 2004, 10, 3747−3760.
28) Guiu, E.; Claver, C.; Benet-Buchholz, J.; Castillo
Tetrahedron: Asymmetry 2004, 15, 3365−3373.
29) White, C.; Yates, A.; Maitlis, P. M. Inorg. Synth. 1992, 29, 228−
30.
30) Tydlitat
934−3940.
31) Busacca, C. A.; Bartholomeyzik, T.; Cheekoori, S.; Grinberg, N.;
Lee, H.; Ma, S.; Saha, A.; Shen, S.; Senanayake, C. H. J. Org. Chem.
008, 73, 9756−9761.
32) Samec, J. S. M.; Bac
33) M86-E01078 APEX2 User Manual; Bruker AXS Inc., Madison,
WI, 2006.
34) Kappa CCD Operation Manual; Nonius BV: Delft, The
Netherlands, 1997.
Organometallics 2009, 28, 3492−3500. (e) Kashiwame, Y.; Kuwata, T.;
Ikariya, T. Chem. Eur. J. 2010, 16, 766−770.
(
́
, S.; Daura, E.; Bo, C.;
(
10) Boutadla, Y.; Davies, D. L.; Jones, R. C.; Singh, K. Chem. Eur. J.
2
(
011, 17, 3438−3448.
11) (a) Davies, D. L.; Donald, S. M. A.; Al-Duaij, O.; Macgregor, S.
A.; Polleth, M. J. Am. Chem. Soc. 2006, 128, 4210−4211. (b) Boutadla,
(
́
n, S.
̈
(
2
Y.; Davies, D. L.; Macgregor, S. A.; Poblador-Bahamonde, A. I. Dalton
Trans. 2009, 5820−5831. (c) Boutadla, Y.; Davies, D. L.; Macgregor,
S. A.; Poblador-Bahamonde, A. I. Dalton Trans. 2009, 5887−5893.
(
̊
̌
́ ̌ ́ ̌
, J.; Bures, F.; Kulhanek, J.; Ruzicka, A. Synthesis 2010,
3
(
(
d) Boutadla, Y.; Al-Duaij, O.; Davies, D. L.; Griffith, G. A.; Singh, K.
Organometallics 2009, 28, 433−440.
(
(
12) Kuwata, S.; Ikariya, T. Chem. Eur. J. 2011, 17, 3542−3556.
13) (a) Zeng, F.; Yu, Z. Organometallics 2009, 28, 1855−1862.
2
(
(
̈
kvall, J.-E. Chem. Eur. J. 2002, 8, 2955−2961.
(
b) Ye, W.; Zhao, M.; Du, W.; Jiang, Q.; Wu, K.; Wu, P.; Yu, Z. Chem.
Eur. J. 2011, 17, 4737−4741.
14) Enthaler, S.; Hagemann, B.; Bhor, S.; Anilkumar, G.; Tse, M. K.;
Bitterlich, B.; Junge, K.; Erre, G.; Beller, M. Adv. Synth. Catal. 2007,
49, 853−860.
15) (a) Busacca, C. A.; Grossbach, D.; Campbell, S. J.; Dong, Y.;
(
(
3
(
(
(
35) Sheldrick, G. M. Acta Crystallogr. 1990, A46, 467−473.
36) Sheldrick, G. M. SHELXL-97; Universitat Gottingen, Gottingen,
̈
̈
̈
Eriksson, M. C.; Harris, R. E.; Jones, P.-J.; Kim, J.-Y.; Lorenz, J. C.;
McKellop, K. B.; O’Brien, E. M.; Qiu, F.; Simpson, R. D.; Smith, L.;
So, R. C.; Spinelli, E. M.; Vitous, J.; Zavattaro, C. J. Org. Chem. 2004,
Germany, 1999.
37) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7−13.
(
6
9, 5187−5195. (b) Bastero, A.; Bella, A. F.; Fernan
́
dez, F.; Jansat, S.;
Claver, C.; Gomez, M.; Muller, G.; Ruiz, A.; Font-Bardía, M.; Solans,
́
I
dx.doi.org/10.1021/om400626m | Organometallics XXXX, XXX, XXX−XXX