A comparison of MOP-phosphonite ligands and their applications in Rh(i)- and Pd(II)-catalysed asymmetric transformations

Article abstract of DOI:10.1039/c6dt02390h

Six chiral MOP-phosphonites have been synthesised and compared via experimental and computational methods in an effort to quantify their differing structural and electronic profiles. They were found to be electron-poor ligands in comparison to their arylphosphine analogues and have a larger trans influence in square planar Pt(ii) complexes. Four [Rh(L^P)(η²:η²-cod)Cl] complexes were synthesised and characterised by NMR, HRMS and X-ray crystallography. Two [Rh(L^P)₂]BF₄ complexes were prepared where one ligand acts as a chelating P,C-π-donor; detailed NMR studies demonstrated a hemilabile η⁶-coordination mode, which in one case was confirmed by X-ray crystallography. Rh(i) complexes were used as catalysts in asymmetric hydrogenation and hydroformylation reactions and in the addition of phenyl boronic acid to an isatin. Pd(ii) complexes were successfully employed in asymmetric Suzuki-Miyaura cross-coupling reactions yielding binaphthyl products. Two [Pd(L^P)₂Cl₂] complexes were synthesised and characterised by X-ray crystallography, both adopting cis orientations, with one of the complexes crystallising as two pseudo-polymorphs.

Full text of DOI:10.1039/c6dt02390h

Dalton
Transactions
View Article Online
View Journal
PAPER
A comparison of MOP-phosphonite ligands and
their applications in Rh(I)- and Pd(II)-catalysed
asymmetric transformations†
Cite this: DOI: 10.1039/c6dt02390h
James T. Fleming, Corinne Wills, Paul G. Waddell, Ross W. Harrington and
Lee J. Higham*
Six chiral MOP-phosphonites have been synthesised and compared via experimental and computational
methods in an effort to quantify their differing structural and electronic profiles. They were found to be
electron-poor ligands in comparison to their arylphosphine analogues and have a larger trans influence in
P
2
2
square planar Pt(II) complexes. Four [Rh(L )(η :η -cod)Cl] complexes were synthesised and characterised
P
by NMR, HRMS and X-ray crystallography. Two [Rh(L )
2
]BF
4
complexes were prepared where one ligand
6
acts as a chelating P,C-π-donor; detailed NMR studies demonstrated a hemilabile η -coordination mode,
which in one case was confirmed by X-ray crystallography. Rh(I) complexes were used as catalysts in
asymmetric hydrogenation and hydroformylation reactions and in the addition of phenyl boronic acid to
Received 15th June 2016,
Accepted 26th August 2016
an isatin. Pd(II) complexes were successfully employed in asymmetric Suzuki–Miyaura cross-coupling
P
reactions yielding binaphthyl products. Two [Pd(L )
2
2
Cl ] complexes were synthesised and characterised
DOI: 10.1039/c6dt02390h
by X-ray crystallography, both adopting cis orientations, with one of the complexes crystallising as two
pseudo-polymorphs.
www.rsc.org/dalton
tate or hemilabile fashion, via air-stable, chiral primary
phosphines (Scheme 1).
Introduction
4
Organophosphorus compounds are a popular choice of ligand
Chiral phosphonite complexes are proven catalysts for a
in homogeneous catalysis because their electronic and steric range of enantioselective reactions, including nickel-catalysed
1
5
6
properties, and those of derived complexes, can be tuned.
hydrocyanations, palladium-catalysed hydrovinylations and
7
8
Chiral monophosphorus ligands have found applications in a rhodium-catalysed hydroformylations and hydrogenations.
variety of organometallic catalysis, however they are often Furthermore, our MOP-phosphonites in Scheme 1 were shown
neglected in favour of the more established diphosphorus to be effective ligands in the palladium-catalysed asymmetric
2
4a
ligand class. Therefore, we are keen to explore the potential of hydrosilylation of styenes. In a follow-up communication we
such ligands, especially those with unusual functionality on prepared the ligands in Chart 1, and showed that despite the
3
the phosphorus. Phosphonites are valuable ligands used in presence of only one element of chirality compared to those
homogeneous catalysis, with characteristic properties of being ligands in Scheme 1, they were still highly effective in the
4
c
poor σ-donors but strong π-acceptors; the energy of their σ* palladium-catalysed hydrosilylation of substituted styrenes.
orbitals is lowered compared to those of related arylpho- In this work we investigate the ramifications of this structural
sphines due to the electronegative substituents, allowing the modification on the stereoelectronic properties of the ligands,
ligand to better stabilise transition metals in low oxidation their coordination chemistry and the catalytic properties of the
states. We reported a convenient synthetic route to enantio-
pure MOP-phosphonite ligands, which can act in a monoden-
School of Chemistry, Bedson Building, Newcastle University, Newcastle upon Tyne,
NE1 7RU, UK. E-mail: lee.higham@ncl.ac.uk
†
Electronic supplementary information (ESI) available: Full experimental and
crystallographic data. CCDC 1483130 (R)-6b, 1483131 (S)-8a, 1483132 (R)-8b,
1
483133 (S)-9a, 1483134 (R)-9b, 1483871 (R)-10b, 1483135 (S)-11a (polymorph 1),
483136 (S)-11a (polymorph 2) and 1483137 (R)-11b. For ESI and crystallographic
1
Scheme 1 Our original synthesis of MOP-phosphonite ligands, con-
data in CIF or other electronic format see DOI: 10.1039/c6dt02390h
taining two chiral elements.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
ca. 50 ppm and downfield for the arylphosphines MeO-MOP
and H-MOP, but larger (ca. 65–75 ppm) and upfield for the
1
31 77
phosphonites. The magnitude of the J ( P– Se) coupling
can provide an indication of the σ-donor strength of the
parent phosphorus lone pair, which is inversely correlated
9
to the coupling constant. The data implies that our MOP-
phosphonites (906–930 Hz) are relatively weak σ-donors compared
to their electron-rich arylphosphine analogues (720–723 Hz) due
to the increased electron-withdrawing effects of the oxygen-
containing substituents, which results in the phosphorus lone
pair having a larger s- and lower p-character, and a decreased σ-
Chart 1 Phosphonite ligands (S)-1a, (R)-1b, (S)-2a, (R)-2b, (S)-3a and
(
R)-3b prepared in this study.
1
basicity (Table 1). The JPSe values for the MOP-phosphonites 1–3
resultant transition metal complexes. Specifically the ligands
S)-1a, (R)-1b, (S)-2a, (R)-2b, (S)-3a and (R)-3b (Chart 1) were
are also larger than those we recently reported for analogous MOP
phosphonites with two methoxy substituents bound to the phos-
(
prepared, as well as their corresponding selenides, and our
subsequent investigations allowed us to make quantitative
comparisons of these donors, capturing their steric and elec-
tronic properties. We used DFT calculations to probe the fron-
tier molecular orbitals of the ligands and we also present
calculations which support our experimental observations that
rotation about the C–C bond in the biphenyl moiety of (S)-2a,
3
phorus (858 and 860 Hz), indicating that replacing two alkyoxy
functionalities with aryloxy groups results in ligands that are
weaker σ-donors in this selenide series.
It has been demonstrated that the presence of bulky substi-
tuents on the phosphorus atom can result in a lower value for
the coupling constant (i.e. a decrease in the s-character of the
1
0
lone pair), which may explain why the bulky dimethyl-
(
R)-2b, (S)-3a and (R)-3b is unrestricted under typical con-
1
4
c
biphenoxy phosphonites (S)-3a and (R)-3b have lower
values than the biphenoxy phosphonites (S)-2a and (R)-2b.
J
PSe
ditions.
The rhodium(I) and palladium(II) coordination
chemistry of the ligands was then investigated in some detail
by multinuclear NMR spectroscopy and X-ray crystallography.
Finally the performance of the MOP-phosphonites as ligands
in asymmetric hydrogenation, hydroformylation and Suzuki–
Miyaura cross-coupling reactions was assessed and compared
to that of the analogous phosphines MeO-MOP and H-MOP.
Rhodium(I) Vaska-type complexes. Another important indi-
cator of the net-donor properties of PR ligands is the sym-
metric carbonyl stretching frequency (ν(CO)) in the IR
3
P
11
spectrum of trans-[Rh(L )
2
(CO)Cl] complexes (Table 1). The
complexes are prepared by reaction of four equivalents of the
1
2
phosphorus ligand with [Rh(CO) Cl] , and do not exhibit the
2
2
high toxicity associated with the nickel(0) carbonyl complexes
traditionally used, although the data does correlate well with
Results and discussion
11,13
the Tolman values.
A low value for ν(CO) indicates that the
Ligand properties
ligand is a strong electron donor, conversely, a high value for
ν(CO) indicates low net donor properties. The values for our
phosphonites (1997–2006 cm− ) indicate overall low net-donor
properties compared to the MOP-phosphine ligands
(1968–1974 cm− ) (Table 1), as would be expected for this
ligand class which are poor σ-donors but strong π-acceptors;
the reduced electron density on the rhodium atom results in
less π-back-bonding to the carbonyl group, resulting in a stron-
P
Organophosphorus selenides. We synthesised L (Se) by
reacting the corresponding phosphorus ligand (L ) with an
1
P
3
1
excess of potassium selenocyanate in THF at 50 °C. The
P
1
1
{
H} NMR of the reaction product consists of a singlet with
7
7
1
Se (I = , natural abundance 7.6%) satellites (Table 1); the
2
P
P
difference in chemical shift between L (Se) and
L
is
1
1b
ger CO bond and hence a relative increase in ν(CO).
Platinum(II) complexes. (S)-H-MOP, (R)-MeO-MOP and each
phosphonite was reacted with the platinum precursor [Pt(PEt
Cl2]2 in CD Cl to generate the unsymmetrical platinum(II)
3
1
1
P
Table 1 Selected P{ H} NMR spectroscopic data for L (Se) com-
P
3
)
pounds and IR spectroscopic data for trans-[Rh(L )
2
(CO)Cl] complexes
2
2
P
c
1
P
d
e
f,g
Ligand
δ (L (Se))
J
PSe (L (Se))
Δδ
ν(CO)
complexes trans-[Pt((S)-H-MOP)(PEt
MeO-MOP)(PEt )Cl ] (R)-4b, trans-[Pt((S)-1a)(PEt
trans-[Pt((R)-1b)(PEt )Cl ] (R)-5b, trans-[Pt((S)-2a)(PEt )Cl ] (S)-
)Cl
] (S)-4a, trans-[Pt((R)-
3
2
a
3
2
3
)Cl ] (S)-5a,
2
(
(
(
(
(
(
(
(
S)-H-MOP
36.6
38.3
89.4
723
720
908
906
926
930
918
915
+50.2
+52.0
−65.4
−63.9
−74.5
−75.6
−67.0
−70.3
1968
1974
2006
2002
2005
1997
2006
2005
b
R)-MeO-MOP
3
2
3
2
a
S)-1a
6a, trans-[Pt((R)-2b)(PEt
3
)Cl
2
] (R)-6b, trans-[Pt((S)-3a)(PEt
3 2
)Cl ]
a
R)-1ba
91.7
(S)-7a and trans-[Pt((R)-3b)(PEt
3
)Cl ] (R)-7b via bridge cleavage
2
S)-2a
104.1
104.4
105.4
103.9
a
14
R)-2ba
reactions. We observed exclusively the formation of the trans
3
1
1
S)-3a
isomers in the
(
P{ H} NMR spectra of the compounds
a
R)-3b
15
P
Table 2). The value of δ (L ) for the complexes with the phos-
a
b
NMR measurements were taken at 162 MHz in CDCl
3
.
NMR phonite ligands was shifted ca. 30–50 ppm upfield compared
c
measurements were taken at 202 MHz in CDCl . Chemical shift in
3
to the resonance of the free ligand, the phenoxy derived
ligands having the largest shift and the ortho-methyl substi-
tuted biphenoxy ligands the lowest; the phosphine ligands
d
e
P
P f
ppm. Coupling constant in Hz. Δδ = δ (L (Se)) − δ (L ) in ppm. IR
g
measurements were taken as solutions in DCM. Symmetric carbonyl
stretching frequency in cm−
1
.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
3
1
1
P
Table 2 Selected P{ H} NMR spectroscopic data for trans-[Pt(L )
PEt )Cl ] complexes
(
3
2
P c
1
P d
c
1
d
2
d
Complex
δ (L )
J
PPt (L )
3
δ (PEt )
J
PPt (PEt
3
)
J
PP
a
1
2
3
4
5
6
7
8
(S)-4a
21.5
25.3
2437
2440
3481
3467
3520
3517
3521
3541
13.8
13.7
11.2
9.0
10.0
9.0
2574
2554
2495
2519
2490
2503
2398
2382
473
468
621
618
624
620
620
620
b
(R)-4b
a
(S)-5a
111.8
108.8
140.6
140.5
140.5
142.6
a
(R)-5b
a
(S)-6a
a
(R)-6b
a
(S)-7a
8.6
12.1
a
(R)-7b
a
b
NMR measurements were taken at 162 MHz in CD
2
Cl
2
.
NMR
c
measurements were taken at 202 MHz in CD
ppm. Coupling constant in Hz.
2
Cl
2
.
Chemical shift in
d
Fig. 1 Molecular structure of (R)-6b. Hydrogen atoms have been
omitted for clarity. Selected average bond distances (Å) and angles (°):
were shifted ca. 35–40 ppm downfield (Table 2). The values of Pt1–P1 2.2703(10), Pt1–P2 2.3148(11), Pt1–Cl1 2.3203(10), Pt1–Cl2
2
the JPP coupling between the ligands are relatively high and 2.2873(10), P1–C2 1.813(4), P1–O1 1.612(3), P1–O2 1.607(3); P1–Pt1–Cl1
1
6
85.72(4), P1–Pt1–P2 170.36(4), P1–Pt1–Cl2 94.02(4), C2–C1–C1’–C2’
–87.6(5), C12–C11–C11’–C12’ −43.2(6).
characteristic of a trans geometry. The phosphonite com-
2
plexes show little variance ( J ≈ 620 Hz) and the values
PP
are noticeably larger than those for the phosphine ligands
2
(
JPP ≈ 470 Hz).
We sought to determine the relative trans influence of our
1
It is cautionary to note however, that the J coupling for
PPt
1
31 195
series of MOP-ligands by measuring the J ( P– Pt) coupling
constants for Pt–PEt in the square-planar platinum(II) com-
PEt in (R)-6b (2503 Hz, Table 2, Entry 6) is significantly larger
than reported for the MOP-bis(dimethylamino)phosphine
complex (2365 Hz), which as discussed would imply that the
phosphonite ligand exhibits a weaker trans influence – this
contradiction may be due to steric effects in the solid state. We
are not aware of any analogous bis(dimethylamino)phosphine
complexes for comparison.
3
3
1
95
1
2
14,17
plexes ( Pt, I = , natural abundance 33.8%) (Table 2).
3
We found that the trans influence of our phosphonites
2382–2519 Hz, Table 2, entries 3–8) is larger than that found
(
for the arylphosphine analogues (2554–2574 Hz, Table 2,
1
entries 1 and 2); a smaller
3
JPPt coupling to PEt denotes a
P
larger trans influence from L . The coupling constants for the
ortho-methyl substituted biphenoxy phosphonites (Table 2,
entries 7 and 8) were noticeably lower than those for the
phenoxy and biphenoxy phosphonites (Table 2, entries 3–6),
implying that ligands (S)-3a and (R)-3b have the largest trans
influence in the series.
DFT calculations. We have performed Density Functional
Theory (DFT) calculations on the optimised structures of the
MOP-ligands using the B3LYP functional with a 6-31G* basis
set to calculate the frontier molecular orbital energies
(
Table 3), the Highest Occupied Molecular Orbital (EHOMO)
19
and Lowest Unoccupied Molecular Orbital (ELUMO). EHOMO
typically contains the phosphorus lone pair in tertiary phos-
phines and can be used as an indication of the ligand σ-donor
We were able to obtain single crystals suitable for X-ray
crystallography for one of our platinum(II) complexes, (R)-6b
(
Fig. 1), by slow diffusion of diethyl ether into a CD
solution.
There is a pseudo square-planar geometry around the plati-
num atom; both (R)-2b and PEt show monodentate coordi-
2
Cl
2
20
strength. The values for EHOMO are invariably lower for the
3
nation through the phosphorus atom and are trans to each Table 3 Calculated HOMO and LUMO energies and symmetric defor-
other, with a P–Pt–P bond angle of 170.36(4)°. The Pt–PEt3 mation coordinate values for the MOP-ligands discussed in this study
bond length (2.3148(11) Å) is similar to that reported for
a,b
a,b
a,c
c,d
Ligand
E
LUMO
E
HOMO
S4
S4
an analogous phosphonite complex, containing a fluorine-
substituted derivative of diphenyl phenylphosphonite, which (S)-H-MOP
R)-MeO-MOP
(S)-1a
R)-1b
that in the related MOP-bis(dimethylamino)phosphine and (S)-2a
−1.15
−1.06
−1.29
−1.21
−1.33
−1.21
−1.34
−1.22
−5.51
−5.27
−5.80
−5.51
−5.76
−5.47
−5.76
−5.48
41
42
54
54
58
58
58
58
—
—
—
(
has two independent molecules in the unit cell (2.321(4) and
1
8
2
.319(4) Å). The Pt–PEt
3
bond length in (R)-6b is longer than
e
(
66/70
—
(R)-2b
(S)-3a
(R)-3b
60
—
—
MOP-phosphirane complexes that we recently reported as part
of our study into the stereoelectronic profiles of a series of
MOP-type ligands, which have analogous bond lengths of
a
Calculated at the DFT level of theory using the B3LYP functional with
a 6-31G* basis set. Energies in eV. Symmetric deformation coordi-
nate in degrees. Calculated from X-ray data. The asymmetric unit
comprises two molecules in different conformations.
2
.3050(15) Å and 2.284(3) Å respectively, indicating that the
b
c
phosphonite ligand has the largest trans influence of the
d
e
3
three.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
H-MOP molecules compared to their MeO-MOP analogues for rotation in biphenyl compounds generally occur at 0°, ±90°
Table 3), implying that the latter have stronger σ-donor and 180°, when the two phenyl rings are either coplanar or
(
properties, in agreement with calculations for related ligands orthogonal; the local minima occur near 45° either side of the
3
previously reported. The HOMO energies for the H-MOP secondary maxima at ±90 due to a compromise between the
phosphonites are lower than that for their arylphosphine steric demands of the ortho-substituents and π-electron
2
6
counterpart, and this trend also holds true for the MeO-MOP overlap. The rotational barrier in 1,1′-biphenyl, and non-
ligand family, implying the phosphonites are weaker P-ligand ortho-substituted biphenyls, are comparable at 0° and 90°
1 26
σ-donors in each series. E
may be invoked in back- (1.4–2.0 kcal mol− ). An activation energy barrier of approxi-
LUMO
1
9
−1
bonding, and the lower values for these phosphonites com- mately 23 kcal mol is required to prevent spontaneous room
2
5d
pared to the phosphines suggests they will likely have better temperature torsional-isomerisation of biphenyl derivatives.
π-acceptor properties (Table 3); the H-MOP phosphonites have It is important to note that when calculating barriers to
lower LUMO energies than their 2′-methoxy-substituted ana- rotation the method and basis set employed have a large
2
5c,d
logues. One of the earliest methods developed for quantifying impact on the optimised torsion angles.
the steric effects of phosphorus ligands is the Tolman cone In our study, the torsion angle between the planes of the
1
3
angle (θ); however, for MOP-type ligands where the substitu- two phenyl rings (C12–C11–C11′–C12′) was constrained at
ent groups differ greatly, the values obtained may not best selected values and the optimised geometry calculated. A plot
2
1
reflect the properties of the ligand. An alternative means of of the relative energies versus the dihedral angle is shown in
evaluating the steric bulk of phosphorus ligands is the sym- Fig. 2; the curves in the graph are not mirror images of each
metric deformation coordinate (S4′), introduced by Orpen and other, as the two minima are diastereomers and hence of
2
2
co-workers using data obtained from crystal structures. S4′ unequal energy. We also calculated the idealised angle for
seeks to evaluate the ligands based on a geometric calculation the unconstrained molecule at the two minima: −46.11°
using bond angles from species with the general formula and +45.56° for (S)-3a, and −46.00° and +45.60° for (R)-3b
Z–PR (Z = transition-, main group- or non-metal). It has also (ESI Table 3†). The calculated energy profiles for both (S)-3a
3
been shown that S4′ values can be calculated with sufficient and (R)-3b are very similar with an energy barrier of approxi-
accuracy to the data obtained from X-ray crystallographic ana- mately 7–8 kcal mol− when the biphenyl rings are coplanar, a
1
2
3
−1
lysis of phosphine complexes. We therefore used the DFT value far below the 23 kcal mol barrier reported as being
optimised structures for the series of MOP-ligands in Table 3 required to prevent spontaneous racemisation at room temp-
2
5d
to calculate a modified descriptor of S4′ (S4), calculating Z to erature.
This result concurs with our experimental findings
be a vector perpendicular to the plane containing the three that rotation about the aryl–aryl junction is unrestricted under
2
4
substituents on the phosphorus.
For phosphines, small values for S4 can suggest substitu-
standard conditions (vide infra).
Rhodium(I) complexes and catalysis. Successful examples of
ents with increasing steric bulk. Our values for (S)-H-MOP and rhodium-catalysed asymmetric transformations employing
(
(
R)-MeO-MOP (Table 3) are close to those calculated for PPh
39.6°) by Suresh and co-workers, implying that they have a Shintani, Nozaki and co-workers reported the high
3
MOP-type ligands can be found in the literature. Recently
2
4
similar steric profile. The S4 values for the MOP-phosphonite yielding and enantioselective synthesis of silicon-stereogenic
2
7
ligands are approximately 12–17° larger than their phosphine dibenzosiloles and the intramolecular alkynylsilylation of
2
8
analogues, and show little variation between themselves. alkynes. MeO-MOP and P-stereogenic phosphine derivatives
However, it is more difficult to accurately model compounds
with P–O bonds compared to P–C bonds, since the R groups in
P(OR) compounds are much further from the phosphorus
3
than in PR3 ligands. Therefore, the accuracy of this method
may not be suitable to our MOP-phosphonite ligands due to
2
3
their conformational flexibility.
We recently reported the solid state structures of (R)-1b and
4
c
(
R)-2b, and consequently here we have been able to calculate
S4 from the X-ray crystallographic data and compare it to the
computed structures (Table 3). The S4 value for (R)-1b is con-
siderably larger in the X-ray structure compared to the calcu-
lated one, whereas both are similar for (R)-2b; the smaller S4
value for (R)-2b compared to (R)-1b is more consistent with
what might be expected due to the biphenoxy group having
less conformational flexibility. We then calculated the approxi-
mate barrier to rotation for the ligands (S)-3a and (R)-3b.
Substituted biphenyls show conformational enantiomorphism:
the ease with which isomerisation occurs depending on the
Fig. 2 Calculated conformational energy differences (in kcal mol−
1
)
versus dihedral angle for the ligands (S)-3a and (R)-3b (in °), relative to
the energy at the optimised dihedral angle; (S)-3a (red) and (R)-3b (blue).
The geometry optimisations were performed at the DFT level of theory
2
5
size of the ortho substituents. The peaks in the energy profile using the B3LYP functional with a 6-31G* basis set.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
3
1
1
have shown promise in the rhodium-catalysed asymmetric
addition of phenyl boronic acid to napthaldehyde.
2D phase-sensitive P{ H} EXSY NMR experiments have
been used to study the mechanism for dynamic ligand
2
9
3
1
Asymmetric intramolecular hydroamination of unactivated exchange in chiral rhodium phosphine complexes, as well as
alkenes using a rhodium(I) catalyst has been described by in the quantitative kinetic analysis of terminal ligand redistri-
Shen and Buchwald, who used a series of binaphthyl-based bution in a nickel phosphine complex. An EXSY NMR experi-
monophosphine ligands to give high yields and up to 91% ee ments were recorded in order to detect any dynamic ligand
3
2
3
0
with the substrates tested.
[
exchange in (S)-9a and (R)-9b: measurements were taken at
Rh(L )(η :η -cod)Cl] complexes. Phosphonites (S)-2a, −55 °C with a mixing time of 100 ms (see ESI†). The absence
P
2
2
2
2
(
R)-2b, (S)-3a and (R)-3b were reacted with [Rh(η :η -cod)Cl]
2
in of cross-peaks in the spectra indicates that the three distinct
a 1 : 1, P : Rh ratio, to quantitatively give the chloro(1,5-cyclo- chemical species observed in (S)-9a and the five distinct
2
2
octadiene)rhodium(I)
(
(
complexes
[Rh((S)-2a)(η :η -cod)Cl] chemical species observed in (R)-9b do not participate in
2
2
2
2
S)-8a, [Rh((R)-2b)(η :η -cod)Cl] (R)-8b, [Rh((S)-3a)(η :η -cod)Cl] dynamic exchange at this temperature.
2
2
S)-9a and [Rh((R)-3b)(η :η -cod)Cl] (R)-9b. The analogous reac-
X-ray crystal structures were determined for (S)-8a, (R)-8b,
tions with (S)-1a and (R)-1b were also attempted, however the (S)-9a and (R)-9b (Fig. 3 and ESI Fig. 2–6†). In the asymmetric
resultant complexes were very sensitive to decomposition and unit of both (S)-9a and (R)-9b there are two molecules in
we were unable to isolate them. Three of the four rhodium different conformations – an overlay of the crystallographically
3
1
1
complexes gave broadened P{ H} NMR spectral signals at independent molecules in (R)-9b is provided to highlight that
room temperature, therefore they were analysed by variable the torsion angle of the dimethyl substituted biphenyl moiety
temperature (VT) NMR spectroscopy (see ESI†). For (S)-8a we is of opposite sign in the two molecules (ESI Fig. 10†). There is
observed a broad doublet at 25 °C which became clearer upon a pseudo-square-planar geometry around the rhodium and
1
heating to 50 °C (δ = 162.9 ppm and JPRh = 224 Hz); cooling the Rh–P bond lengths range from 2.2104(13) in (R)-8b to
the sample down to −55 °C resulted in the observation of two 2.2428(13) Å in (R)-9b, consistent with the literature value of
distinct species with relative intensities of ∼65% (δ
=
2.2112(7) Å for an analogous MeO-MOP-phosphonite complex
1
4b
1
J
64.6 ppm and JPRh = 223 Hz) and ∼35% (δ = 161.4 ppm and which we ourselves reported and 2.233(2) Å for a related
1
33
PRh = 221 Hz). For (R)-8b a doublet was observed at 25 °C (δ = complex employing a hydroxy phosphonite ligand.
The
1
1
63.3 ppm and J
= 221 Hz), and cooling down to −55 °C Rh–Cl bond lengths are longer than the Rh–P bond lengths
PRh
resulted in the appearance of an additional low intensity and range from 2.3470(13) Å in (R)-9b (Fig. 3) to 2.3757(12) Å
doublet resonance. The two species which appear in the in (R)-8b (see ESI Fig. 3†), and the corresponding literature
4
b
33
spectra of (S)-8a and (R)-8b at −55 °C most likely correspond values (vide supra) of 2.3661(8) Å and 2.371(2) Å lie within
to conformational isomers with the ligands exhibiting this range. The phosphorus donor shows a stronger trans influ-
restricted rotation about the biphenyl C–C bridge and/or the ence compared to the chloride in each case, hence for the
Rh–P bond. Both (S)-8a and (R)-8b were poorly soluble in alkene trans to the phosphorus, the C–C bond is shorter and
3 2 2
CDCl and CD Cl , which proved to be problematic when per-
forming NMR experiments due to the high number of scans
required; presumably at low temperatures the complexes pre-
cipitate out of the solution, further reducing the concen-
tration. However, we were able to characterise the binaphthyl
moiety in (S)-8a at 50 °C and (R)-8b at 25 °C, but the broad-
1
13
ened H and C resonances attributed to the biphenyl group
could not be assigned for either complex.
For the dimethylbiphenoxy phosphonite complex (S)-9a,
3
1
1
phosphonite coordination resulted in one broad P{ H} NMR
signal at 25 °C which split into a poorly resolved doublet when
heated to 50 °C. Cooling to −55 °C allowed for the resolution
of three signals with approximate ratios of 76% (δ =
1
1
1
70.2 ppm, JPRh = 225 Hz), 14% (δ = 169.4 ppm, JPRh = 224
1
Hz) and 10% (δ = 162.0 ppm, J
= 223 Hz). The methoxy
PRh
analogue (R)-9b shows a broad doublet coupling at 25 °C,
whose resolution improved upon heating to 50 °C; cooling to
Fig. 3 Molecular structure of one of the two independent molecules of
R)-9b (the asymmetric unit comprises two molecules in different con-
(
−
55 °C resulted in the formation of five distinct doublets with
formations). Hydrogen atoms have been omitted for clarity. Selected
1
approximate ratios of 13% (δ = 168.1 ppm,
J
PRh = 227 Hz), average bond distances (Å) and angles (°): Rh2–P2 2.2428(13), Rh2–Cl2
1
2.3470(13), Rh2–C45 2.131(5), Rh2–C46 2.134(5), Rh2–C49 2.272(6),
8
0% (δ = 167.5 ppm, JPRh = 226 Hz), 1% (δ = 165.1 ppm,
1
1
Rh2–C50 2.237(5), P2–C28 1.823(5), P2–O4 1.609(3), P2–O5 1.615(4),
C45–C46 1.399(8), C49–C50 1.330(8); P2–Rh2–Cl2 89.62(5), P2–Rh2–
C49 173.59(16), P2–Rh2–C50 152.03(16), P2–Rh2–C45 92.60(15), P2–
Rh2–C46 95.98(15), C28–C27–C27’–C28’ −104.6(6), C38–C37–C37’–
C38’ −43.3(8).
JPRh = 227 Hz), 3% (δ = 163.4 ppm, J
= 224 Hz) and 3%
PRh
1
(
δ = 162.8 ppm, JPRh = 223 Hz). The multiple species observed
3
1
1
in the P{ H} NMR of (S)-9a and (R)-9b at −55 °C most likely
correspond to rotamers.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
the Rh–C bonds are longer than the alkene bound cis to the
phosphorus atom. In (S)-8a and (R)-8b the lower naphthyl frag-
ment of the binaphthyl backbone is face-to-face with the
rhodium centre, whereas in both independents of (S)-9a and
(
R)-9b it is orientated away from the rhodium centre and faces
the biphenyl moiety.
P
P
+
[
Rh(L )
plexes using our bulky dimethylbiphenoxy ligands (S)-3a and
R)-3b by reacting two equivalents of the phosphonite with
2 4 2
]BF complexes. We next synthesised [Rh(L ) ] com-
(
2
2
[Rh(η :η -cod)
2
]BF
4
to afford [Rh((S)-3a)
2
]BF
4
(S)-10a and
[
Rh((R)-3b) ]BF (R)-10b. (S)-10a and (R)-10b were characterised
2
4
by multinuclear NMR spectroscopy, HRMS and for (R)-10b,
3
1
1
X-ray crystallography. Solution state P{ H} VT NMR studies
on (S)-10a and (R)-10b (see ESI†) revealed that the two phos-
phorus nuclei are inequivalent, due to the appearance of two
resonances with equal intensity, which are shifted slightly
upfield compared to (S)-9a and (R)-9b. Upon cooling we were
able to measure P–P and P–Rh coupling, which gave a doublet
1
of doublets splitting pattern for (S)-10a ( JPRh = 288 and 294 Hz
2
1
and JPP = 27 Hz) and (R)-10b ( J
= 283 and 310 Hz and
PRh
2
J
PP = 32 Hz).
13
Fig. 4 Proposed molecular structure and a fragment showing the
{
C
1
The H resonances in the spectra of (S)-10a and (R)-10b
1
H} NMR coordination chemical shift (ppm) of ligand B (P,C-π-donor)
were slightly broadened and unresolved at room temperature, relative to that of ligand A (P-donor) for (S)-10a (top) and (R)-10b
(
bottom). NMR measurements were taken at 176 MHz in CDCl
3
at 5 °C
however upon cooling to 5 °C and −20 °C respectively, we were
able to observe the expected number of independent aromatic
resonances (38 for (S)-10a and 36 for (R)-10b) and the 4 inde-
pendent methyl resonances from the biphenoxy moieties.
for (S)-10a and at 126 MHz in CDCl at −20 °C for (R)-10b.
3
1
Using two dimensional NMR experiments ( H-COSY,
The X-ray crystal structure of (R)-10b shows there are two
1
H-ROESY, HSQC and HMBC) performed at these reduced crystallographically independent molecules in the asymmetric
temperatures, in addition to H, H{ P} and C{ H} NMR unit, which comprises two cations and two BF
1
1
31
13
1
4
anions (Fig. 6
data, we were able to unambiguously assign the proton and and ESI Fig. 7†). The cations of (R)-10b contain a rhodium
1
3
carbon resonances. The C NMR chemical shift data for (S)- atom with two coordinated phosphonites; both ligands coordi-
0a and (R)-10b reveal that in the solution state, one of the nate via the phosphorus atom and, in each case, one ligand
hemilabile MOP ligands is coordinating to the rhodium atom acts as chelating P,C-π-donor via second binding
1
a
a
6
via an η -binding mode using the lower naphthyl ring, in mode using the lower naphthyl ring (Fig. 6 and ESI Fig. 7†).
addition to bonding through the phosphorus atom (Fig. 4). The Rh–C distances suggest an η -binding mode, consistent
6
6
The carbon atoms in the η -bound C
6
ring from the P,C- with our solution state NMR experiments, where we propose
ring containing the ipso-C (which happens to be
solely P-donor ligand (A), by a magnitude of 8.9 to 26.5 ppm in the closest contact) coordinates to the rhodium atom
π-donor ligand (B) are shifted upfield compared to those in the that the C
6
6
(
S)-10a and 6.5 to 35.7 in (R)-10b (Fig. 4). The resonances side-on via its π-system (vide supra). This η -bonding motif
attributed to C1′ and C4′ in ligand B are the most significantly for a MOP-type ligand is rare, the other example we are aware
shifted and appear as doublets with J values of 11.3 and of is a MOP-phosphonite complex that we previously reported,
1
(
2.8 Hz in (S)-10a and 15.2 and 10.4 Hz in (R)-10b respectively with a BINOL moiety on the phosphorus atom – the Rh–Caryl
4
b
see ESI†).
distances in both complexes are comparable.
Goldberg
6
The result of having a chelating P,C-π-donor is saturation of reported that η -bonding modes can also be found in
the rhodium coordination sphere. This may explain the rhodium complexes with 2-(dicyclohexylphosphino)biphenyl
P
34
increased stability of the [Rh(L ) ]BF species compared to the ligands.
1
2
4
P
2
2
6 electron [Rh(L )(η :η -cod)Cl] complexes, which are prone
Asymmetric hydrogenation (AH) of alkenes. Chiral mono-
to oxidation. The presence of dynamic behaviour in (S)-10a dentate phosphites, phosphoramidites and phosphonites
3
5
36
8
was probed by Rotating frame Overhauser Effect Spectroscopy have proven to be effective ligands in the rhodium-catalysed
1
P
(
ROESY); the 2D phase-sensitive H-ROESY NMR spectra in AH of prochiral alkenes. We therefore tested the [Rh(L )
2 4
]BF
CDCl
3
at 0 °C and −50 °C are shown in Fig. 5. At 0 °C exchange complexes (S)-10a and (R)-10b against a selection of bench-
6
is evident, a result of hemilabile binding of the η -coordinated mark substrates: dimethyl itaconate (DMI), methyl 2-acetami-
aryl which dissociates and is then replaced by the equivalent doacrylate (MAA) and methyl (Z)-2-acetamidocinnamate
face of the second ligand (Fig. 4). Upon cooling to −50 °C this (MAC). In the AH of DMI we were able to obtain reasonable
dynamic behaviour ceases.
conversion after 48 h, however the reaction did not proceed
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Fig. 6 Molecular structure of one of the cations of (R)-10b (the asym-
metric unit comprises two cations and two anions in different confor-
mations). Hydrogen atoms have been omitted for clarity. Selected
average bond distances (Å) and angles (°): Rh1–P1 2.2117(16), Rh1–P2
2
.2008(18), Rh1–C19’ 2.200(6), Rh1–C20’ 2.311(6), Rh1–C21’ 2.312(6),
Rh1–C22’ 2.278(7), Rh1–C27’ 2.445(7), Rh1–C28’ 2.492(6), P1–C2 1.816
6), P2–C20 1.817(7); P1–Rh1–P2 98.62(6), Rh1–P1–C2 111.1(2), Rh1–
P2–C20 105.9(2), C2–C1–C1’–C2’ −84.3(8), C12–C11–C11’–C12’ −37.7
11), C20–C19–C19’–C20’ −92.6(8), C30–C29–C29’–C30’ 40.2(10).
(
(
due to their weak σ-donor but strong π-acceptor properties. We
screened the six MOP-phosphonites and two MOP-phosphines
in the AHF of styrene, using 0.1 mol% of a catalyst prepared
2
in situ from the reaction of [Rh(acac)(CO) ] with the ligand in a
Rh : P ratio of 1 : 2. The reactions were performed at a pressure
of 40 bar syngas over 16 h and at a moderate temperature of
1
Fig. 5 2D phase-sensitive H-ROESY NMR spectra of (S)-10a in CDCl
3
at 0 °C (top) and −50 °C (bottom), with a mixing time of 200 ms. The
diagonal and exchange cross-peaks are phased positive (red) and the
ROE signals are negative (blue).
36 °C (Table 4).
It is clear that the phosphonite ligands (Table 4, entries
–8) far outperformed their phosphine counterparts (Table 4,
3
entries 1 and 2) with respect to conversion. Very high conver-
sions were obtained in each case, and complete conversion
enantioselectively. With the substrate MAA the conversion was
low and with MAC it was negligible, however we did obtain a
moderate ee for MAC when employing (R)-10b as ligand (see
ESI† for full details).
Table 4 Rhodium-catalysed asymmetric hydroformylation of styrene
Asymmetric hydroformylation (AHF) of styrene. The AHF of
styrene analogues is an industrially relevant reaction due to
the pharmacological importance of a number of the 2-aryl pro-
pionic acid derivatives accessible from the products of this
reaction. For example, (S)-naproxen, (S)-ibuprofen, (R)-flurbi-
profen and (S)-ketoprofen are important non-steroidal anti-
a
b
c
Ligand
Conversion
Iso/n
ee
3
7
inflammatory drugs (NSAIDs). However, the use of mono-
dentate ligands in these transformations is underdeveloped.
Beller and co-workers reported the AHF of styrene with a
1
2
3
4
5
6
7
8
(S)-H-MOP
(R)-MeO-MOP
(S)-1a
(R)-1b
(S)-2a
(R)-2b
(S)-3a
(R)-3b
47
53
99
>99
96
97
19
24
18
15
16
15
12
16
6 (S)
6 (S)
3 (R)
4 (S)
6 (R)
1 (R)
1 (S)
12 (S)
38
rhodium-phosphepine complex (48% ee), however the yield
was poor, demonstrating the scope for progress in this area of
research. Substitution of P–C for P–O bonds in the ligand in
hydroformylation catalysts often results in increased rates of
reaction, as this facilitates the dissociation of CO from the
98
>99
a
1
b
%
Conversion determined by H NMR spectroscopy. Ratio deter-
mined by H NMR spectroscopy. % ee determined by chiral GC
Supelco β-dex); absolute configuration was assigned according to lit-
1
c
3
7
metal. Therefore, we speculated that our MOP-phosphonite
(
ligands could be promising candidates for this transformation erature data.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
was achieved when the MeO-MOP phosphonites (R)-1b and carbon bonds is the palladium-catalysed Suzuki–Miyaura
R)-3b were employed as the ligand. The product was a mixture cross-coupling reaction of a boronic acid with an organic
(
4
0
of the linear (n) and branched (iso) isomers, however the halide. In these reactions, electron rich and bulky phosphine
4
1
desired iso-aldehyde was formed in a large excess; iso/n ratios ligands are most often preferred. However, a DFT compu-
ranged from 12–18 for the phosphonite catalysts (Table 4, tational study by Harvey and co-workers into phosphine
entries 3–8) and was slightly higher (19–24) for the phosphine- donor/acceptor properties and steric features found that the
catalysts (Table 4, entries 1 and 2). Unfortunately the enantio- transmetallation process in the catalytic cycle has a lower
selectivity in these reactions was found to be low, the best energy barrier for weak σ-donor and strong π-acceptor
4
2
result being a 12% ee when employing the ligand (R)-3b ligands. Iwasawa and co-workers have reported high yields
(Table 4, entry 8). However, the high conversions and regio- and enantioselectivities (up to 91% and 78% respectively) in
selectivity observed in these non-optimised conditions with a cross-coupling reactions of aryl chlorides employing bulky,
low catalyst loading, does indicate that modified reaction con- chiral phosphonite ligands with low catalyst loadings
3
8
43
ditions and further ligand development could lead to prom- (0.1 mol%). We therefore synthesised palladium(II) dichlor-
ising AHF catalysts. ide complexes of our MOP-phosphonites and tested their
Asymmetric addition of phenyl boronic acid to an isatin. ability to prepare atropisomeric compounds from aryl bro-
Hayashi and co-workers reported the Rh-catalysed asymmetric mides and aryl boronic acids, a field where a number of mono-
addition of arylboronic acids to isatins in order to develop an phosphines have shown high enantioselectivities and
4
4
asymmetric strategy towards the synthesis of biologically active activity. The enantioselective synthesis of molecules with a
compounds containing the 3-aryl-3-hydroxy-2-oxindole struc- binaphthyl backbone from naphthalene derivatives is interest-
3
9
tural motif. They were able to catalyse the reaction using a ing as this can facilitate a route to chiral monophosphine
Rh(I)/(R)-MeO-MOP complex, achieving enantioselectivities of ligands such as those reported here. This was demonstrated by
up to 91%. We therefore tested our series of MOP-phosphonite Ma and Yang who used a cross-coupling reaction as part of
4
5
ligands in the addition of phenyl boronic acid to 1-(p-methoxy- their strategy to obtain optically pure H-MOP. It has also
benzyl)-5-chloroisatin (Scheme 2). We also used the ligand (R)- been reported that MeO-MOP-stabilised palladium nano-
MeO-MOP to allow us to compare our procedure to the litera- particles can catalyse the coupling of 1-bromo-2-methoxy-
ture data (ESI Table 2†) – both the conversion and ee are agree- naphthalene and naphthalene-1-boronic acid, albeit the
3
9
able with the literature values. Ligands (S)-1a and (R)-1b unoptimised reaction gave only moderate conversion (50%)
4
6
failed to provide any conversion, which we suspect is due to and low enantioselectivity (12%). To the best of our knowl-
the reaction conditions that include water – both ligands are edge there are no reports of MOP-phosphonite ligands being
4
c
sensitive to hydrolysis. The biphenoxy derived ligands (R)-2b, employed in this type of reaction.
S)-3a and (R)-3b fared better however, generating conversions Asymmetric Suzuki–Miyaura cross-coupling reactions. We
(
of between 40–43%. Ligand (S)-2a gave approximately half the tested our ligands in the palladium-catalysed reaction between
conversion of (R)-2b, (S)-3a and (R)-3b, however we have found 1-bromo-2-methoxynaphthalene and naphthalene-1-boronic
that this ligand is the most sensitive to decomposition out of acid in order to synthesise the atropisomeric molecule
the biphenoxy series, and this is possibly the reason for the 2-methoxy-1,1′-binaphthyl (Table 5). The catalyst was generated
reduced activity. The H-MOP-phosphonites (S)-2a and (S)-3a by stirring two equivalents of the corresponding ligand with
2 2
gave matching ees of 32%, however the MeO-MOP analogues [Pd(MeCN) Cl ] in DCM for 2 hours at room temperature,
gave very different ees. (R)-3b produced a low value of 11%, before removing the volatiles under vacuum; the catalysis was
with (R)-2b giving the highest value of 39% ee in this biphe- performed under the conditions stated. All of the MeO-MOP
noxy series. We can also report, to the best of our knowledge, type ligands proceeded with almost full conversion. The reac-
the first application of a Rh(I)/(S)-H-MOP complex as a catalyst tion was complete within 30 minutes when (R)-1b was
in the asymmetric addition of phenylboronic acid to 1-(p- employed as the ligand and within 1 hour with (R)-MeO-MOP;
methoxybenzyl)-5-chloroisatin. The reaction gave complete longer reaction times were required to reach ≥95% conversions
conversion and achieved a good ee of 70% (see ESI Table 2† with (R)-2b and (R)-3b. The H-MOP ligands generally gave
for full details of the conversions and enantioselectivities).
lower conversions and the reaction times were longer com-
Palladium(II) complexes and catalysis. One of the most fre- pared to their methoxy analogues. There was no additional
quently applied synthetic methodologies used to form carbon– conversion observed between 16 and 24 hours reaction
times. The reactions with (R)-MeO-MOP and (R)-1b were the
quickest in the series (Table 5, entries 2 and 4), however
the enantioselectivity was negligible. It was markedly
improved when using the biphenoxy based phosphonites
(
R)-2b and (R)-3b, which gave the highest enantioselectivities
obtained (27% and 42% respectively). The diphenoxy derived
phosphonite (S)-1a achieved the highest ee in the H-MOP
ligand series, whereas (R)-3b was the best MeO-MOP ligand in
this regard.
Scheme 2 Rhodium-catalysed asymmetric addition of phenylboronic
acid to 1-(p-methoxybenzyl)-5-chloroisatin.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
Table 5 Asymmetric Suzuki–Miyaura coupling of naphthalene-1- there was no additional conversion between 16 and 24 hours
boronic acid with 1-bromo-2-methoxynaphthalene
reaction times. The ees improved using 1-bromo-2-methyl-
naphthalene compared to 1-bromo-2-methoxynaphthalene –
the value of the ee increased for every ligand used in our
screening (Tables 5 and 6). There is no major variation in
the conversions obtained between analogous ligands in the
H-MOP and MeO-MOP series (Table 6), however there is a con-
a
Conversion
sistent pattern of the ee being higher for the MeO-MOP
ligands in each case. The enantioselectivities for the diphe-
noxy ((S)-1a) and dimethylbiphenoxy ((S)-3a) phosphonites
b
Ligand
30 min
1 h
16 h
ee
1
2
3
4
5
6
7
8
(S)-H-MOP
(R)-MeO-MOP
(S)-1a
(R)-1b
(S)-2a
(R)-2b
(S)-3a
(R)-3b
86
98
66
>99
42
84
94
>99
83
—
51
91
65
90
95
—
94
—
75
95
85
98
4 (R) were similar (Table 6, entries 3 and 7), whereas it was notice-
8 (R)
ably lower for the biphenoxy ligand (S)-2a, which had an ee
1 (S) close to that obtained with the parent phosphine (S)-H-MOP
12 (R) (Table 6, entries 1 and 5). For the MeO-MOP series, the ee
doubled upon changing the ligand from (R)-MeO-MOP to
42 (R) (R)-1b, however this was accompanied by a ∼50% reduction in
the conversion (Table 6, entries 2 and 4). The ligands (R)-2b
21 (R)
27 (R)
11 (S)
54
89
a
1
b
%
Conversion determined by H NMR spectroscopy. % ee deter-
and (R)-3b achieved higher conversions than (R)-1b, with
(R)-2b affording an ee similar to that of (R)-1b. However, out of
all the phosphines and phosphonites tested, (R)-3b achieved
mined by chiral HPLC (Lux Cellulose) with hexane/2-propanol (99 : 1);
absolute configuration was assigned according to literature data.
the highest ee of all (49%).
P
[
Pd(L )
2
Cl
2
] complexes. In an effort to investigate the solu-
After screening the electron donating methoxy-substituted
naphthalene, we were also interested in investigating the
effects of replacing the methoxy functional group with a
weakly donating methyl moiety. The result of the coupling
reactions of 1-bromo-2-methylnaphthalene with naphthalene-
tion and solid state properties of our Suzuki–Miyaura catalysts,
we synthesised the complexes [Pd((S)-3a) Cl ] (S)-11a and
Pd((R)-3b) Cl ] (R)-11b by reacting [Pd(MeCN) Cl ] with two
equivalents of the corresponding phosphonite ligand in DCM.
We were able to obtain single crystals of (S)-11a and (R)-11b,
2
2
[
2
2
2
2
1
-boronic acid to synthesise 2-methyl-1,1′-binaphthyl are given
3
suitable for X-ray crystallographic analysis, from CDCl /hexane
in Table 6. In comparison to the results obtained with the
methoxy substituted substrate, the conversions obtained are
significantly lower in all cases and, with the exception of (S)-1a
and (S)-3a (Table 6, entries 3 and 7), only a minor improve-
ment to the conversion is made by leaving the reactions longer
than 30 minutes. As with the methoxy substituted substrate,
and DCM/diethyl ether respectively. (S)-11a crystallised as two
pseudo-polymorphs. One is in a monoclinic space group as a
solvate with three chloroform molecules, whereas the second
pseudo-polymorph is the same main molecule but in an ortho-
rhombic space group with one chloroform molecule (ESI Fig. 8
and 9†); an overlay image of the two pseudo-polymorphs has
been provided (ESI Fig. 11†). (R)-11b crystallised in a monocli-
nic space group and the asymmetric unit contains two dis-
Table 6 Asymmetric Suzuki–Miyaura coupling of naphthalene-1- ordered molecules of DCM in multiple orientations (Fig. 7).
boronic acid with 1-bromo-2-methylnaphthalene
Both pseudo-polymorphs of (S)-11a and (R)-11b have a square-
planar geometry about the palladium centre and the phospho-
nite ligands are in a cis orientation. The P–Pd–P bond angles
are 99.79(5) and 101.16(3)° for (S)-11a and 100.73(4)° for
(
2
R)-11b. The Pd–P bond lengths range from 2.2234(9)–
.2331(15) Å in (S)-11a and 2.2329(11)–2.2354(11) Å in (R)-11b
a
and the Pd–Cl bond lengths range from 2.3302(9)–2.3485(15) Å
in (S)-11a and 2.3359(11)–2.3379(11) Å in (R)-11b. The X-ray
structure of the complex [Pd((R)-MeO-MOP) Cl ] was reported
2 2
Conversion
b
Ligand
30 min
1 h
16 h
ee
1
2
3
4
5
6
7
8
(S)-H-MOP
(R)-MeO-MOP
(S)-1a
(R)-1b
(S)-2a
(R)-2b
(S)-3a
(R)-3b
50
47
26
24
32
32
27
41
52
48
31
25
33
34
38
42
55
53
38
26
40
38
46
47
19 (S) by Hayashi et al. and reveals that the MOP-phosphine ligands
21 (S)
are in a trans orientation, unlike that seen for our MOP-phos-
phonite ligands; the Pd–P bond lengths in trans-[Pd((R)-
30 (S)
42 (S)
17 (S) MeO-MOP) Cl ] are 2.339(2) and 2.344(15) Å and the P–Pd–P
2
2
47
43 (S)
32 (S)
49 (S)
bond angle is 174.95(8)°. The X-ray structure of a related
complex with the chloride atoms replaced by fumaronitrile
(
trans-1,2-dicyanoethylene) and where the palladium is in a
a
1
b
%
Conversion determined by
H
NMR spectroscopy.
% ee
zero oxidation state, [Pd((S)-MeO-MOP) (fumaronitrile)], was
2
determined by chiral HPLC (Lux Cellulose) with hexane/2-propanol
99 : 1); absolute configuration was assigned according to literature
reported by Pregosin et al.; the molecule lies on a C2 axis
and the MOP-phosphine ligands are in a cis orientation, the
(
data.
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.

View Article Online
Paper
Dalton Transactions
MOP phosphonite ligands were synthesised and characterised
in detail, and they were then shown to be effective as catalysts
in asymmetric Suzuki–Miyaura cross-coupling reactions, with
full conversions and enantioselectivities of up to 49% possible,
depending on the substrates investigated.
Acknowledgements
We acknowledge the financial support of the EPSRC for
a Fellowship award (LJH: EP/G005206/1) and their National
Mass Spectrometry Facility at Swansea. We also thank the
School of Chemistry, Newcastle University for a Studentship
Fig. 7 Molecular structure of (R)-11b. Hydrogen atoms have been
omitted for clarity. Selected average bond distances (Å) and angles (°):
Pd1–P1 2.2354(11), Pd1–P2 2.2329(11), Pd1–Cl1 2.3359(11), Pd1–Cl2
(
JTF).
2.3379(11), P1–C2 1.793(4), P1–O1 1.605(3), P1–O2 1.593(3), P2–C20
1
.806(4), P2–O4 1.586(3), P2–O5 1.595(3); P1–Pd1–P2 100.73(4), P1–
Pd1–Cl1 169.34(4), P1–Pd1–Cl2 84.88(4), C2–C1–C1’–C2’ −89.5(6),
C12–C11–C11’–C12’ 44.0(7), C20–C19–C19’–C20’ −96.0(5), C30–C29–
C29’–C30’ 45.0(7).
Notes and references
1
(a) Phosphorus Ligands in Asymmetric Catalysis, ed.
A. Börner, Wiley-VCH, Weinheim, 2008; (b) Phosphorus(III)
Ligands in Homogeneous Catalysis: Design and Synthesis, ed.
P. C. J. Kamer and P. W. N. M. Van Leeuwen, John Wiley &
Sons, Chichester, 2012.
P–Pd–P bond angle is 114.11(3)° and the Pd–P bond length is
4
8
2
.3588(8) Å.
The Pd–P bond lengths in these MeO-MOP-phosphine com-
2
3
4
F. Lagasse and H. B. Kagan, Chem. Pharm. Bull., 2000, 48,
plexes are significantly longer than those found in our MOP-
phosphonite complexes. A phosphonite complex analogous to
315.
A. Ficks, W. Clegg, R. W. Harrington and L. J. Higham,
Organometallics, 2014, 33, 6319.
(a) A. Ficks, R. M. Hiney, R. W. Harrington, D. G. Gilheany
and L. J. Higham, Dalton Trans., 2012, 41, 3515;
(
S)-11a and (R)-11b, cis-[Pd(P(OMe)
2 2 2
Ph) Cl ], which has a crys-
tallographic two-fold rotation axis, has a Pd–P bond length of
2
.2300(16) Å and a P–Pd–P bond angle of 100.88(2)°, values
4
9
which are very similar to those found in (S)-11a and (R)-11b.
(
b) A. Ficks, R. W. Harrington and L. J. Higham, Dalton
We analysed (S)-11a and (R)-11b in solution by multinuclear
NMR spectroscopy. For (S)-11a we observed two broad peaks in
Trans., 2013, 42, 6302; (c) J. T. Fleming, A. Ficks,
P. G. Waddell, R. W. Harrington and L. J. Higham, Dalton
Trans., 2016, 45, 1886.
3
1
1
the P{ H} NMR spectrum at 25 °C, a large peak at 144.3 ppm
and a smaller peak at 142.8 ppm, which correspond to the cis
and trans isomers. In the case of (R)-11b one broad peak
5
6
7
L. Bini, C. Muller and D. Vogt, Chem. Commun., 2010, 46,
8325.
R. Bayersdörfer, B. Ganter, U. Englert, W. Keim and
D. Vogt, J. Organomet. Chem., 1998, 552, 187.
J. I. Van der Vlugt, J. M. J. Paulusse, E. J. Zijp,
J. A. Tijmensen, A. M. Mills, A. L. Spek, C. Claver and
D. Vogt, Eur. J. Inorg. Chem., 2004, 4193.
3
1
1
was observed at approximately 140 ppm in the P{ H} NMR
spectrum at room temperature.
Conclusions
8
(a) C. Claver, E. Fernandez, A. Gillon, K. Heslop, D. J. Hyett,
A. Martorell, A. G. Orpen and P. G. Pringle, Chem.
Commun., 2000, 961; (b) M. T. Reetz and T. Sell, Tetrahedron
Lett., 2000, 41, 6333; (c) Y. Fu, G.-H. Hou, J.-H. Xie, L. Xing,
L.-X. Wang and Q.-L. Zhou, J. Org. Chem., 2004, 69, 8157.
We have used experimental and computational data to quan-
tify the stereoelectronic donor properties within a MOP-
phosphonite ligand series, and rationalised previously
reported findings that rotation about the aryl–aryl junction in
the biphenyl moiety of (S)-2a, (R)-2b, (S)-3a and (R)-3b is
unrestricted under standard conditions. We prepared Rh(I)
complexes and through detailed NMR and X-ray crystallo-
9 (a) W. McFarlane and D. S. Rycroft, J. Chem. Soc., Chem.
Commun., 1972, 902; (b) W. McFarlane and D. S. Rycroft,
J. Chem. Soc., Dalton Trans., 1973, 2162.
graphic analyses of them, have proven the existence of M–π 10 (a) R. P. Pinnell, C. A. Megerle, S. L. Manatt and
P
+
interactions in complexes of the type [Rh(L )
2
] , with one
P. A. Kroon, J. Am. Chem. Soc., 1973, 95, 977; (b) D. W. Allen
ligand acting as a chelating P,C-π-donor. In the rhodium-
and B. F. Taylor, J. Chem. Soc., Dalton Trans., 1982, 51.
catalysed AHF reaction of styrene, our ligands proved to be 11 (a) S. Vastag, B. Heil and L. Markó, J. Mol. Catal., 1979, 5,
high yielding and regioselective for the branched isomer with
low catalyst loadings. Palladium dichloride complexes of the
189; (b) A. Roodt, S. Otto and G. Steyl, Coord. Chem. Rev.,
2003, 245, 121.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Dalton Transactions
Paper
1
2 R. A. Baber, M. F. Haddow, A. J. Middleton, A. G. Orpen, 29 (a) M. Sakai, M. Ueda and N. Miyaura, Angew. Chem., Int.
P. G. Pringle, A. Haynes, G. L. Williams and R. Papp,
Organometallics, 2007, 26, 713.
3 C. A. Tolman, Chem. Rev., 1977, 77, 313.
Ed., 1998, 37, 3279; (b) E. F. Clarke, E. Rafter, H. Müller-
Bunz, L. J. Higham and D. G. Gilheany, J. Organomet.
Chem., 2011, 696, 3608.
1
1
4 M. Carreira, M. Charernsuk, M. Eberhard, N. Fey, R. van 30 X. Shen and S. L. Buchwald, Angew. Chem., Int. Ed., 2010,
Ginkel, A. Hamilton, W. P. Mul, A. G. Orpen, H. Phetmung
and P. G. Pringle, J. Am. Chem. Soc., 2009, 131, 3078.
5 (a) A. Pidcock, R. E. Richards and L. M. Venanzi, J. Chem.
49, 564.
31 H. Bircher, B. R. Bender and W. von Philipsborn, Magn.
Reson. Chem., 1993, 31, 293.
1
Soc. A, 1966, 1707; (b) S. O. Grim, R. L. Keiter and 32 J. D. Heise, D. Raftery, B. K. Breedlove, J. Washington and
W. McFarlane, Inorg. Chem., 1967, 6, 1133. C. P. Kubiak, Organometallics, 1998, 17, 4461.
6 K. B. Dillon, A. E. Goeta, P. K. Monks and H. J. Shepherd, 33 D. Selent, W. Baumann, R. Kempe, A. Spannenberg,
1
1
1
Polyhedron, 2010, 29, 606.
7 K. M. Anderson and A. G. Orpen, Chem. Commun., 2001,
D. Röttger, K.-D. Wiese and A. Börner, Organometallics,
2003, 22, 4265.
2
682.
34 A. R. O’Connor, W. Kaminsky, D. M. Heinekey and
K. I. Goldberg, Organometallics, 2011, 30, 2105.
8 M. J. Atherton, J. Fawcett, A. P. Hill, J. H. Holloway,
E. G. Hope, D. R. Russell, G. C. Saunders and 35 M. T. Reetz and G. Mehler, Angew. Chem., Int. Ed., 2000, 39,
R. M. J. Stead, J. Chem. Soc., Dalton Trans., 1997, 1137. 3889.
9 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648; 36 (a) M. Van den Berg, A. J. Minnaard, E. P. Schudde, J. van
1
(
b) P. J. Stephens, F. J. Devlin, C. F. Chabalowski and
Esch, A. H. M. De Vries, J. G. De Vries and B. L. Feringa,
J. Am. Chem. Soc., 2000, 122, 11539; (b) Y. Fu, X.-X. Guo,
S.-F. Zhu, A.-G. Hu, J.-H. Xie and Q.-L. Zhou, J. Org. Chem.,
2004, 69, 4648.
M. J. Frisch, J. Phys. Chem., 1994, 98, 11623.
0 (a) H. M. Senn, D. V. Deubel, P. E. Blöchl, A. Togni and
G. Frenking, J. Mol. Struct.: THEOCHEM, 2000, 506, 233;
2
(b) C. H. Suresh and N. Koga, Inorg. Chem., 2002, 41, 1573; 37 R. Franke, D. Selent and A. Börner, Chem. Rev., 2012, 112,
(
2
c) N. Fey, A. G. Orpen and J. N. Harvey, Coord. Chem. Rev.,
009, 253, 704.
1 T. L. Brown and K. J. Lee, Coord. Chem. Rev., 1993, 128, 89.
5675.
38 G. Erre, S. Enthaler, K. Junge, S. Gladiali and M. Beller,
J. Mol. Catal. A: Chem., 2008, 280, 148.
2
2
2 B. J. Dunne, R. B. Morris and A. G. Orpen, J. Chem. Soc., 39 R. Shintani, M. Inoue and T. Hayashi, Angew. Chem., Int.
Dalton Trans., 1991, 653. Ed., 2006, 45, 3353.
2
3 K. D. Cooney, T. R. Cundari, N. W. Hoffman, K. A. Pittard, 40 N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457.
M. D. Temple and Y. Zhao, J. Am. Chem. Soc., 2003, 125, 41 (a) U. Christmann and R. Vilar, Angew. Chem., Int. Ed.,
4
318.
4 J. Mathew, T. Thomas and C. H. Suresh, Inorg. Chem., 2007,
6, 10800.
5 (a) A. Almenningen, O. Bastiansen, L. Fernholt, B. N. Cyvin,
2005, 44, 366; (b) R. Martin and S. L. Buchwald, Acc. Chem.
Res., 2008, 41, 1461.
42 J. Jover, N. Fey, M. Purdie, G. C. Lloyd-Jones and
J. N. Harvey, J. Mol. Catal. A: Chem., 2010, 324, 39.
2
2
4
S. J. Cyvin and S. Samdal, J. Mol. Struct., 1985, 128, 59; 43 T. Kamei, A. H. Sato and T. Iwasawa, Tetrahedron Lett.,
b) S. Charbonnier, S. T. Beguemsi, Y. T. N’Guessan, 2011, 52, 2638.
D. Legoff, A. Proutiere and R. Viani, J. Mol. Struct., 1987, 44 D. Zhang and Q. Wang, Coord. Chem. Rev., 2015, 286, 1.
(
1
3
2
58, 109; (c) F. Grein, J. Phys. Chem. A, 2002, 106, 45 Y.-N. Ma and S.-D. Yang, Chem. – Eur. J., 2015, 21, 6673.
823; (d) E. Masson, Org. Biomol. Chem., 2013, 11, 46 K. Sawai, R. Tatumi, T. Nakahodo and H. Fujihara, Angew.
859.
Chem., Int. Ed., 2008, 47, 6917.
2
2
2
6 E. L. Eliel and S. H. Wilen, Stereochemistry of Organic 47 Y. Uozumi, K. Kitayama, T. Hayashi, K. Yanagi and
Compounds, John Wiley & Sons, New York, 1994. E. Fukuyo, Bull. Chem. Soc. Jpn., 1995, 68, 713.
7 R. Shintani, C. Takagi, T. Ito, M. Naito and K. Nozaki, 48 P. Dotta, P. G. A. Kumar, P. S. Pregosin and A. Albinati,
Angew. Chem., Int. Ed., 2015, 54, 1616. Helv. Chim. Acta, 2004, 87, 272.
8 R. Shintani, H. Kurata and K. Nozaki, Chem. Commun., 49 A. M. Z. Slawin, P. G. Waddell and J. D. Woollins, Acta
015, 51, 11378. Crystallogr., Sect. E: Struct. Rep. Online, 2010, 66, m321.
2
This journal is © The Royal Society of Chemistry 2016
Dalton Trans.