Organometallics
Communication
to be explored. However, we expect that such a system would
have several advantages. For example, two kinds of chirogenic
donors can be generated from a single chiral source.
Furthermore, the aromatic backbone would enable construc-
tion of a rigid chiral environment, wherein the rigidity of the
ligand backbone has already proven to be an important factor
Scheme 2. Coordination Study with Pd(cod)Cl2
7
to attaining high selectivities in asymmetric reactions. We
herein report the synthesis of such chiral amino-phosphine
ligands, and their coordination studies with palladium metal as
a model case to elucidate the N-centered configuration
generated upon metal coordination.
Imamoto et al. previously reported the development of
various bisphosphine ligands possessing the chirogenic
PtBuMe unit wherein they employed a phosphine−borane,
a
1
1
1a
Estimated by H- H NOESY.
(
S)-PtBuMeH·BH , as a key precursor. This phosphine−
3
borane can also be readily introduced into an electron-deficient
N-containing aromatic core to produce air-stable chiral
8
phosphine ligands. Inspired by these reports, we selected
group is oriented syn relative to the tBu group on the
phosphorus atom. Because of the SP configuration of the
phosphine, the configuration of the nitrogen center would be
the PtBuMe group as a chirogenic phosphorus group and
pyridine as a ligand backbone. Importantly, we anticipated that
the P-tBu group would be sufficiently large to take part in
through-space interactions with the N-substituent. On the
basis of such a ligand design, N-substituents such as Me, iPr,
and tBu were examined for their ability to control the
configuration of the nitrogen center.
Thus, the target amino-phosphines were readily prepared
from the commercially available 3-amino-2-chloropyridine
Scheme 1). Alkylation of the amine group provided alkyl-
R .
N
In contrast, ligands 1b and c formed a mixture of two
isomeric complexes in molar ratios of 68:32 and 80:20,
respectively. In complex 4b, the major and minor species were
assigned as the R ,S and S ,S isomers, respectively, on the
N
P
N
P
1
1
basis of the H− H NOESY spectrum. Variable-temperature
NMR studies revealed that coalescence of the signals derived
from each isomer was observed at ca. 135 °C. After the
solution was recooled, the obtained spectrum was comparable
rapidly underwent inversion on the NMR time scale at ca. 135
(
Scheme 1. Syntheses of the P-Chirogenic Amino-Phosphine
°
C, and the diastereomeric ratio observed in solution was
thermodynamically determined.
Unfortunately, the structure of complex 4c could not be fully
elucidated by NMR spectroscopy due to the similar chemical
shifts of the signals derived from the P-Me and P-tBu groups.
It should also be noted here that, when an excess of ligand
1
c was employed, the formation of a new species was observed.
On the basis of an X-ray crystallographic analysis, the structure
of this new species was determined to be the complex
PdCl (1c) , in which two phosphorus atoms are coordinated
2
2
1
substituted compounds 2. Lithium phosphide, prepared by the
has been well documented in various amino-phosphine
9
deprotonation of (S)-PtBuMeH·BH with BuLi, was smoothly
ligands.
3
reacted with compounds 2 to afford borane-protected
The structures of the complexes in the solid state were
unambiguously determined by X-ray crystallographic analysis
compounds 3. After deprotection, the desired (R )-amino-
P
10
phosphines were obtained in good yields.
(Figure 2). In complex 4a, the phosphorus and nitrogen
With these ligands in hand, we then examined the syntheses
of palladium complexes by ligand exchange with Pd(cod)Cl2
atoms coordinated toward palladium to form a slightly
distorted square planar palladium complex (∑° = 360°).
Pd
(
Scheme 2). Thus, ligands 1a−c were treated with 1 equiv of
In good agreement with the NMR results, the N-Me group was
located syn relative to the P-tBu group; the complex exhibited
an RN configuration at the nitrogen center and an SP
configuration at the phosphorus atom. Another characteristic
point is that the Pd−Cl bond trans to the phosphine (2.402(3)
Å) is longer than that of the Pd−Cl bond trans to the nitrogen
atom (2.286(3) Å). This difference in bond lengths results
from the large trans influence of the phosphorus atom in
comparison to that of nitrogen.
Pd(cod)Cl in CH Cl . Subsequently, diethyl ether or hexane
2
2
2
was added, and the resulting pale yellow solid was collected to
provide the desired complexes in yields of 93−96%.
Characterization of the complex structures was first
attempted by NMR spectroscopy using CDCl as a solvent.
3
1
31
In the case of complex 4a, the obtained H and P NMR
spectra showed two sets of signals in a ratio of 98:2, implying
that a single diastereomer was formed with high selectivity.
1
1
Furthermore, in the H− H NOESY spectrum, a cross peak
was observed between the proton signals of the N-Me and P-
tBu moieties, and so we tentatively concluded that the N-Me
Similar structural features were also observed in complexes
4b and c. Each crystal contained only one isomer, which
exhibited the R configuration at the nitrogen center. Again, a
N
B
Organometallics XXXX, XXX, XXX−XXX