Homogeneous hydrogenation and isomerization of 1-octene catalyzed by nickel(II) complexes with bidentate diarylphosphane ligands

Article abstract of DOI:10.1021/ic400973t

A systematic library of 24 nickel(II) complexes with bidentate diphosphane ligands was synthesized, and the solid-state structures of five of them were determined with X-ray crystallography. The compounds C1-C3 are common P ₂Ni^IIX₂-type complexes, while C4 contains a unique [P₂Ni^II(NH₃)(OAc)]⁺ square-planar structure with a P₂NO donor set and C5 constitutes a rare [(P₂Ni^II)₂(μ-OH)₂] ²⁺ dinuclear compound. The catalytic activity of all complexes was tested in the hydrogenation and/or isomerization of 1-octene in a CH ₂Cl₂/CH₃OH reaction medium. Catalyst precursors bearing ligands with o-alkoxy aryl rings selectively hydrogentate 1-octene to n-octane, while catalytic systems comprising ligands without the o-alkoxy functionality selectively isomerize the substrate to a mixture of internal alkenes, mostly cis- and trans-2-octene. The conversion is enhanced by equipping the ligand aryl rings with electron-donating alkoxy groups, by increasing the steric bulk of the backbone and/or the aryl rings, by employing relatively noncoordinating anions, and by adding a base as the cocatalyst. Using the compound [Ni(L3X)I₂] as the catalyst precursor and upon application of standard hydrogenation conditions, full conversion of the substrate was achieved in 1 h to isomerization products only (TON = 1940). When a catalytic amount of the base is added, a similar result is obtained even in the absence of H₂. A maximum TON of 4500 in 1 h with 96% selectivity for n-octane was achieved by employing [Ni(oMeO-L3X)(NH₃)(OAc)]PF₆ as the catalyst precursor.

Full text of DOI:10.1021/ic400973t

Article
pubs.acs.org/IC
Homogeneous Hydrogenation and Isomerization of 1‑Octene
Catalyzed by Nickel(II) Complexes with Bidentate Diarylphosphane
Ligands
Tiddo J. Mooibroek,† Erica C. M. Wenker,† Wietse Smit,† Ilpo Mutikainen,‡ Martin Lutz,§
and Elisabeth Bouwman*,†
†Gorlaeus Laboratories, Leiden Institute of Chemistry, Leiden University, P.O. Box 9502, 2300 RA Leiden, The Netherlands
‡Laboratory of Inorganic Chemistry, Department of Chemistry, University of Helsinki, P.O. Box 55 (A.I. Virtasen aukio 1), 00014
Helsinki, Finland
§Bijvoet Center for Biomolecular Research, Crystal and Structural Chemistry, Utrecht University, Padualaan 8, 3584 CH Utrecht, The
Netherlands
S
* Supporting Information
ABSTRACT: A systematic library of 24 nickel(II) complexes with bidentate
diphosphane ligands was synthesized, and the solid-state structures of five of them
were determined with X-ray crystallography. The compounds C1−C3 are common
P2NiIIX2-type complexes, while C4 contains a unique [P2NiII(NH3)(OAc)]+ square-
planar structure with a P2NO donor set and C5 constitutes a rare [(P2NiII)2(μ-
OH)2]2+ dinuclear compound. The catalytic activity of all complexes was tested in
the hydrogenation and/or isomerization of 1-octene in a CH2Cl2/CH3OH reaction
medium. Catalyst precursors bearing ligands with o-alkoxy aryl rings selectively hydrogentate 1-octene to n-octane, while catalytic
systems comprising ligands without the o-alkoxy functionality selectively isomerize the substrate to a mixture of internal alkenes,
mostly cis- and trans-2-octene. The conversion is enhanced by equipping the ligand aryl rings with electron-donating alkoxy
groups, by increasing the steric bulk of the backbone and/or the aryl rings, by employing relatively noncoordinating anions, and
by adding a base as the cocatalyst. Using the compound [Ni(L3X)I2] as the catalyst precursor and upon application of standard
hydrogenation conditions, full conversion of the substrate was achieved in 1 h to isomerization products only (TON = 1940).
When a catalytic amount of the base is added, a similar result is obtained even in the absence of H2. A maximum TON of 4500 in
1 h with 96% selectivity for n-octane was achieved by employing [Ni(oMeO-L3X)(NH3)(OAc)]PF6 as the catalyst precursor.
1-octene as the substrate, a cumulative TON of 3000 has been
achieved with 88% selective hydrogenation to n-octane.9
The current study is aimed at expanding the overall
mechanistic understanding of the P2NiII-catalyzed hydro-
genation/isomerization of 1-octene with the hope of finding
even more active and/or selective catalytic systems. Thus, a
large and systematic library of nickel complexes has been
synthesized and characterized. For some complexes, the solid-
state structure was elucidated using X-ray diffraction. The
catalytic activity of the nickel compounds was tested in the
hydrogenation and isomerization of 1-octene.
INTRODUCTION
■
The hydrogenation and isomerization of alkenes are useful and
versatile chemical conversions. Hydrogenation reactions are
carried out on a megaton scale in the fine chemical, commodity
chemical, pharmaceutical, and food industries; a common
process is the hydrogenation of vegetable oils into semisolid
fats for the production of margarine.1,2 Isomerization of
terminal alkenes to internal ones can be useful to obtain
branched aldehydes via a consecutive isomerization/hydro-
formylation synthetic pathway.3,4 Internal alkenes may also be
used to obtain branched copolymers of, e.g., 2-butene and
ethylene.5
Active homogeneous hydrogenation catalysts have been
developed that are based on iridium and rhodium [turnover
numbers (TON) exceeding 104].6 However, these metals are
2−4 orders of magnitude more expensive than the more
abundant nickel.7 Nature uses nickel in the active site of the
enzyme hydrogenase, which can activate molecular hydrogen at
ambient temperature and atmospheric pressure.8 Inspired by
these hydrogenases, we initiated the development of nickel(II)
compounds supported by bidentate diarylphosphane ligands as
catalysts for homogeneous hydrogenation reactions.9−12 Using
RESULTS
■
Complex Synthesis and Characterization. An overview
of the ligands used in this study is shown in Table 1. Because
we have shown that [P2Ni(anion)2] complexes with C2-bridged
phosphane ligands form inactive bischelate [Ni(P2)2]2+
complexes in solution,13 a generic C3-backbone (indicated by
“L3”) was selected for the present study. Ligands with a
butylene-type backbone were not used because these can form
Received: April 19, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Schematic Representation of the Ligands Used in This Study and an Overview of the Nickel Compounds Prepared and
Used in Catalysis
a
b
c
d
e
If not specified, X = C. If not specified, R = H. × = prepared in this study. Used as an in situ formed catalyst. ‘OH’ = [{Ni(L)}2(μ-
OH)2](PF6)2; ‘NH3’ = [Ni(oMeO-L3X)(NH3)(OAc)]PF6.
Table 2. Details of the X-ray Crystal Structure Determinations for Complexes C1−C5
C1
C2
C3
C4
C5
empirical formula
C33H38Cl2NiO4P2 + disordered C35H42Cl2NiO4P2·2CH2Cl2 C39H50I2NiO4P2
C35H44F6NNiO6P3
C66H78F12Ni2O10P6 + disordered
solvent
solvent
a
a
fw
690.18
956.97
957.24
red
840.33
yellow
1647.45
cryst color
cryst size [mm3]
red
red
orange
0.36 × 0.09 × 0.06
0.18 × 0.14 × 0.11
0.30 × 0.10 ×
0.05
0.39 × 0.36 × 0.12 0.30 × 0.24 × 0.24
cryst syst
space group
a[A]
monoclinic
P21/c
monoclinic
P21/n
monoclinic
C2/c
triclinic triclinic
P1
̅
P1
̅
14.1806(4)
18.1253(7)
34.5607(12)
10.966(2)
19.026(4)
21.044(4)
20.816(2)
11.3200(10)
17.623(2)
12.8383(3)
13.1291(3)
13.6642(3)
67.872(1)
63.753(1)
78.577(1)
1912.17(8)
2
11.9615(3)
15.2613(7)
22.5847(9)
97.836(2)
101.560(1)
96.226(2)
3962.1(3)
2
b[A]
c[A]
α [deg]
β [deg]
γ [deg]
V[Å3]
112.3386(14)
98.37(3)
100.130(10)
8216.4(5)
8
4343.8(15)
4
4087.9(7)
4
Z
a
a
dcalc [g cm−3
]
1.116
1.463
1.555
1.456
1.381
a
a
μ [mm−1
sin θ/λ [Å−1
]
0.71
1.05
2.10
0.71
0.74
]
0.59
0.64
0.64
0.65
0.61
reflns (measd/
unique)
64249/14457
10004/7905
4682/4140
47353/8767
41917/12857
abs corr
multiscan
0.45−0.96
769/0
multiscan
0.83−0.89
512/57
multiscan
0.57−0.90
221/0
multiscan
0.75−0.92
488/0
multiscan
0.69−0.83
907/0
abs corr range
no. of param/
restraints
R1/wR2 [I >
2σ(I)]
0.0618/01551
0.887/0.1673
0.0397/0.0978
0.0582/0.1088
0.0199/0.0469
0.0272/0.0512
0.0255/0.0643
0.0293/0.0664
0.0569/0.1409
0.0920/0.1538
R1/wR2 [all
reflns]
S
0.966
1.083
1.128
1.030
1.056
Δρmin/max [e Å−1
]
−0.70/1.63
−0.776/0.659
−1.004/0.340
−0.37/0.32
−0.70/1.41
a
The derived values do not contain the contribution of the disordered solvent.
inactive trans complexes with the small NiII ion.14,15 The length
of the backbone was slightly increased in one case by
substitution of the central C atom into the backbone by the
larger Si atom (indicated by “XSi”). In some instances, the
backbone is more rigid by substitution of the H atoms on the
central atom by methyl (“X”) or ethyl (“X*”) groups; such a
rigid backbone ensures instantaneous complex formation of in
situ formed catalysts16 and leads to more stable chelates.17
The aryl rings of several ligands have been functionalized by
methoxy substitution in the ortho position relative to P. To
allow discrimination between the electronic and steric effects of
this substitution, ligands were equipped with a p-methoxy
B
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. ORTEP projections of C4: (a) front view; (b) top view. ORTEP projections of C5: (c) side view; (d) top view. Ellipsoids are presented at
the 50% probability level. All H atoms (except aryl o-H, NH3, and OH), uncoordinated anions, and solvent molecules are omitted for clarity. Color
code: green, Ni; orange, P; red, O; blue, N; black, C; open sphere, H. The variation of the displacement ellipsoids indicates some minor disorder,
which has not been resolved (C5).
functionality (pMeO-L3), an o-methyl group (oMe-L3), or an
o-ethoxy moiety (oEtO-L3X and oEtO-L3X*). In another
ligand, both an o-methoxy and a m-methyl moiety were
introduced (oMeOmMe-L3) because it has been shown that
this ligand forms a sterically crowded complex with palladium.18
Using the ligands shown in Table 1, a series of nickel(II)
halide compounds of the general formula [Ni(L)X2] (X = Cl,
Br, I) has been synthesized using procedures adapted from the
literature (see Table 1).11 In attempts to make mononuclear
and IR and (1H, 13C{1H},31P{1H}, and 19F) NMR spectros-
copy.
Characterization by 31P{1H} NMR was not possible for
complexes of sterically demanding ligands; such steric crowding
causes the ideal square-planar geometry to become distorted to
a more tetrahedral-like environment, giving paramagnetic
(high-spin) character to NiII, which quenches the NMR signal
of the nearby (i.e., P, C, and H) atoms.21,22 Because of this
paramagnetic behavior, most compounds are 31P NMR silent
and the signal of the central CH2 protons of the propylene
bridge of several compounds is rather broad or not observed at
−
nickel compounds with noncoordinating PF6 counterions
similar to the reported compound [Ni(oMeO-L3)(H2O)2]-
(PF6)2,20 unexpectedly three new dinuclear hydroxide-bridged
complexes with the formula [{Ni(L)}2(μ-OH)2](PF6)2 were
isolated (indicated with ‘OH’ in Table 1). The compound
[Ni(oMeO-L3X)(NH3)(OAc)]PF6 (‘NH3’) was surprisingly
obtained in an attempt to exchange both acetate anions of the
in situ formed [Ni(oMeO-L3X)(OAc)2] for PF6− anions by the
addition of NH4PF6.20 All complexes were obtained as
microcrystalline or crystalline material in good-to-excellent
(nonoptimized) yields. The identity and purity of the isolated
complexes were confirmed by (C, H, N, S, and Ni) elemental
analysis, electrospray ionization mass spectrometry (ESI-MS),
1
all in the H NMR spectra.
The axial or equatorial orientation of the ligand aryl rings can
be distinguished using 1H{31P} NMR: the anagostic interaction
between the Ni and o-H atoms of the axial aryl rings causes
these o-H resonances to shift downfield, whereas an overall
upfield shift is observed for the equatorially aligned o-H
atoms.13,16,18 At room temperature, dynamic flipping of the
backbone causes the aryl rings to alternate between an axial and
an equatorial orientation, resulting in a time-average signal. At
low temperature, this flipping can be frozen to resolve the
resonances of the axial and equatorial rings, as we indeed
C
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 3. Selected Interatomic Distances, Angles, and Torsion Angles for Complexes C1−C5
C1
C2
C3
C4
C5
X1
X2
Cl1
Cl2
Cl1
Cl2
I1
OAc (O41)
NH3 (N31)
OH (O31)
OH (O41)
Distances [Å]
Ni···Ni
2.8968(7)
1.887(3)
Ni−X1
2.2171(10)
2.2181(11)
2.1723(12)
2.1705(11)
2.2095(5)
2.2109(4)
2.1725(4)
2.1760(4)
2.5382(2)
2.1835(1)
1.9072(9)
1.9589(12)
2.1682(3)
2.1830(3)
2.8187(11)
2.75
Ni−X2
1.889(3)
Ni−P1
2.1440(12)
2.1444(12)
Ni−P2
Ni−O42
Ni···H102
Ni···H202
Ni···O506
Ni···O606
2.80
2.9792(6)
3.0599(8)
3.285(5)
2.930(2)
2.9636(4)
3.609(1)
2.78
2.90
2.87
2.95
3.405(3)
3.589(3)
3.4054(11)
3.6635(11)
3.658(4)
3.106(3)
Angles [deg]
X1−Ni−X2
P1−Ni−X1
P1−Ni−X2
P2−Ni−X1
P2−Ni−X2
P1−Ni−P2
92.79(4)
88.76(4)
169.17(4)
168.15(4)
88.98(4)
91.70(4)
122.46
92.79(2)
89.90(2)
166.36(3)
164.51(3)
89.82(2)
91.15(2)
118.8(1)
119.1(1)
91.97(1)
90.55(2)
163.70(2)
88.39(5)
177.86(3)
91.92(4)
87.95(3)
176.29(4)
91.715(13)
121.40
79.74(13)
93.70(9)
169.23(12)
166.32(13)
95.55(10)
92.71(4)
121.03
91.54(2)
119.6(1)
Ni···H102−C102
Ni···H202−C202
120.56
119.49
118.56
Torsion Angles [deg]
Ni−P1−C101−C102
Ni−P1−C501−C502
Ni−P2−C201−C202
Ni−P2−C601−C602
P1−C1−C2−C3
1.9(4)
8.80
4.59(18)
0.04(11)
9.6(4)
−113.3(3)
−2.9(4)
−108.5(3)
−31.7(5)
87.7(4)
−118.77
14.17
109.34(15)
−113.39(10)
5.50(12)
97.8(4)
−9.4(4)
120.8(3)
39.5(5)
−81.7(4)
161.9(3)
−136.01
−38.96
−106.08(11)
−32.61(14)
87.14(12)
34.45(7)
P1−C1−C2−C11
P1−C1−C2−C12
−84.96(19)
155.19(13)
−154.8(3)
−156.45(9)
Dihedral Angle ω between the NiP2 and NiX2 Planes [deg]
20.31 22.78
NiP2−NiX2
15.76(6)
2.23(6)
15.27(16)
typically observed for [Ni(L)X2]-type complexes. This dynamic
behavior is absent in the rigid compound [{Ni(oMeO-
L3X)}2(μ-OH)2](PF6)2, for which a clear diamagnetic NMR
spectrum is obtained despite the relatively large distortion from
a square-planar geometry (see below), in which the axial and
equatorial protons are easily distinguished at room temperature.
To obtain intimate structural knowledge of the complexes
and relate this knowledge to their catalytic performances, the X-
ray structures of some compounds have been determined.
Complex Structures in the Solid State. Single crystals of
the compounds [Ni(oMeO-L3X)Cl2] (C1), [Ni(oMeOmMe-
L3)Cl2] (C2), [Ni(oEtO-L3X*)I2] (C3), [Ni(oMeO-L3X)-
(NH3)(OAc)](PF6) (C4), and [{Ni(oMeO-L3X)}2(μ-OH)2]-
(PF6)2 (C5) were obtained using the solvent-diffusion
technique. Their solid-state structures were determined using
X-ray diffraction. Crystallographic data and details of the
structure refinement are collected in Table 2, and perspective
views of C4 and C5 are shown in parts a,b and c,d of Figure 1,
respectively. Because the structures of complexes C1−C3 are
very similar to those of complexes reported with other
bidentate phosphane ligands9,20,23,24 and that of C4, their
perspective views are given in the Supporting Information
(Figure S1). In Table 3 are listed selected structural
characteristics such as bond distances, angles, and torsion
angles.
the bond distances and angles of the independent molecules are
negligible, the data of only one of them are given in Table 3.
Within the lattice of C2 are two CH2Cl2 solvent molecules, one
of which is disordered. Complex C3 is located on a 2-fold
rotation axis running through the Ni center and the quaternary
C atom in the ligand backbone (C2). The dinuclear compound
C5 cocrystallized with ordered and disordered solvent CH2Cl2
−
molecules. The PF6 anions in C4 and C5 are found in the
second coordination sphere of the complex and are held in
place by several weak N−H···F and C−H···F hydrogen-
bonding interactions.
All NiII ions have a cis-P2X2 (X = anion or NH3) donor set.
The bond distances of the donor atoms to the NiII ion are
about 2.21 Å (Ni−Cl), 2.54 Å (Ni−I), 1.90 Å (Ni−O), and
1.96 Å (Ni−N); the Ni−P distances vary slightly between
2.1388(13) and 2.1835(1) Å. All of these coordination
distances, as well as the C−C, C−H, C−O, and C−P distances,
can be considered as normal.25,26 The distance of 2.897 Å
between the two Ni ions in C5 is well below the sum of their
van der Waals radii (3.26 Å), which is normal in such
(L2Ni)2(μ-OH)2 clusters.27
The coordination geometry in all complexes is essentially
square-planar, although the magnitude of the distortion from
the ideal square-planar geometry varies considerably: the
dihedral angle ω between the P−Ni−P and X−Ni−X planes
ranges from 2.23° in C4 to 22.78° in C3. This difference is
ascribed to the steric constraints that the phosphane ligand and
The asymmetric unit of C1 contains two independent
molecules of [Ni(oMeO-L3X)Cl2]; because the differences in
D
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
anions (or NH3 ligand) impose on each other. For example, ω
becomes smaller when the ionic radius r of X decreases: I (C3;
r = 1.98 Å, ω = 22.78°) > Cl (C2; r = 1.75 Å, ω = 20.31°) > O
(C5; r = 1.52 Å, ω ≈ 16°) > N (C4; r = 1.55 Å, ω = 2.23°).28
Likewise, the dihedral angle in C2 (20.31°) is larger than that in
C1 (16°) because of the bulkiness of the introduced m-methyl
groups.18 The P−Ni−P coordination angle varies slightly from
91.15(2)° to 91.72(2)° in C1−C4, whereas this angle is slightly
larger in C5 (92.6°) due to the restricted O−Ni−O angle.
The two aryl rings attached to a P-donor atom can be
distinguished as oriented either axially or equatorially with
respect to the six-membered NiPCCCP ring. As can be seen in
a catalytic experiment. Identification of 1-octene, the hydro-
genation product n-octane, and the isomerization products cis-
2-octene and trans-2-octene was verified by the injection of
authentic samples. The four other internal alkenes were formed
typically in trace amounts but were not individually identified
and instead were taken together as >2-octenes or taken with cis-
and trans-2-octene as octenes/isomerization products. For
quantification of the analytes, it was assumed that the response
factor for all analytes is unity and that the mass balance of
initially added 1-octene is respected. Indeed, higher-molecular-
weight products were never observed with GLC. Furthermore,
it has been reported that oligomerization/polymerization of
ethylene does not take place with P2NiX2-type complexes under
similar reaction conditions.32
Figure 1, the axial rings have their o-H atoms (H101 − H401
)
pointing toward the Ni center. The Ni···H102−402−C102−402
angles are about 120°, and the Ni···H distances are between
2.75 and 3.06 Å. The Ni···H−C angles in the range of 110−
170° and the M···H−C distances of about 2.3−2.9 Å are
Catalytic Reactions under Standard Conditions Using
[Ni(L)X2] Catalysts. The results using catalyst precursors of the
type [Ni(L)X2], wherein L is a ligand with an unsubstituted
propylene backbone, are shown in Table 4. Using L3 or oMe-
2
characteristic for anagostic interactions between the dz orbital
of a metal ion and the H atom of C−H.29
The equatorial rings have their o-methoxy groups directed
toward the Ni ion, with Ni···O distances in the range of 2.93−
4.17 Å. The Ni···O606/806 distances in C2 [2.930(2) Å] and C5
[3.106(3) and 3.094(3) Å] are shorter than the sum of the van
der Waals radii of the Ni and O atoms (3.15 Å), indicating
electrostatic interactions between the Ni ion and the O atoms
of an alkoxyphenyl ring. Such interactions have never before
been reported in a [P2NiIIAnion2] or [(P2NiII)2(μ-Anion)2]2+
complex, wherein P2 is a bidentate phosphane ligand,30 and are
unknown for complexes with other metals and a bidentate o-
alkoxyaryl-functionalized L3-type ligand.31 Only for the
dicationic compound [Ni(oMeO-L3)(H2O)2](PF6)2 has a
distance of 2.87 Å between one of the methoxy groups and
the NiII ion been reported.20 The other Ni−O distances in the
complexes are well above the sum of the van der Waals radii of
the O and Ni atoms [3.285(5)−4.165(5) Å], which can be
considered as normal for such complexes.9,20
Table 4. Conversion and Selectivity of [Ni(L)X2] Complexes
with L3-Type Ligands for Hydrogenation and Isomerization
a
Products of 1-Octene
selectivity (%)
conversion
(%)
n-
entry
complex
[Ni(L3)Cl2]
TON octane octenes
1
2
3
4
5
6
7
8
0
0
0
0
0
0
0
0
[Ni(L3)Br2]
[Ni(oMe-L3)Cl2]
[Ni(pMeO-L3)Cl2]
[Ni(oMeO-L3)Cl2]
[Ni(oMeO-L3)Br2]
C2
0
0
0
0
b
21
21
49
35
66
410
410
950
680
1280
0
100
0
100
99
100
99
1
0
[Ni(oMeOmMe-L3)
Br2]
1
9
[Ni(oMeOmMe-L3)I2]
100
1940
98
2
a
Reaction conditions: A total of 27.9 μmol of complex (or 27.9 μmol
The crystal packing of C1−C3 and C5 is governed by
common weak intermolecular interactions (weak hydrogen
bonds and CH···π and π···π stacking forces). C4 is packed as a
dimer because of strong intermolecular hydrogen bonding
of NiCl2 and 41.9 μmol of ligand) dissolved in 28.5 mL of a 1.90 M 1-
octene solution in CH3OH/CH2Cl2 (1:1) was heated at 50 °C under
50 bar of H2 for 1 h. [Ni] = 1 mM; 1-octene/Ni = 1940. The TON
b
(mol of substrate converted/mol of Ni) is also given. Catalyst formed
between the acetate and ammonia ligands (N31−H31A···O42
=
in situ.
2.086 Å), while a weaker intramolecular hydrogen bond is
present as well (N31−H31C···O41 = 2.389 Å). The non-
coordinating O atom of the acetate ion in C4 shows a weak
electrostatic interaction with the Ni ion; the intramolecular
Ni···O42 distance is 2.819 Å (van der Waals Ni + O = 3.15 Å).
Analogous P2NiIIX29,20,23,24 and P2PdIIX216,18 complexes have
been reported to exhibit structural features similar to those
described above.
Catalytic Experiments. General Considerations. The
nickel complexes were tested as homogeneous catalysts for
hydrogenation reactions. The catalyst typically was a
presynthesized nickel(II) diphosphane complex. Occasionally,
the complex was generated in situ by dissolving a nickel(II) salt
in combination with 1.5 equiv of the selected diphosphane
ligand in the reaction mixture. A 1:1 mixture of CH2Cl2/
CH3OH was used in this study because we found that the
performance of similar nickel(II) catalysts is best in this solvent
mixture.11 The model substrate chosen for our catalytic
experiments is 1-octene, and a typical 1-octene/nickel ratio of
1940 has been used at a 1.0 mM concentration of nickel.
Gas/liquid chromatography (GLC) analysis of reaction
mixtures was used to identify and quantify the products after
L3 as the supporting ligand did not give any conversion of 1-
octene (entries 1−3). The inactivity of [Ni(oMe-L3)Cl2] can
be ascribed to the poor stability of the complex; the ligand
readily dissociates in a methanolic solution, as was confirmed
with NMR experiments. A conversion of 21% is reached when
the ligand aryl rings are functionalized with electron-donating
methoxy substituents in the para (pMeO-L3, entry 4) or ortho
(oMeO-L3, entry 5) positions. The effect of this substitution
on the selectivity of the catalytic system is drastically different
however; pMeO functionalization results in the formation of
only isomerization products, whereas oMeO substitution
results in the formation of n-octane, as has been reported
earlier.9,10 About 50% conversion to n-octane is reached when
[Ni(oMeO-L3)Br2] (entry 6) is used compared to 21% for the
chloride analogue (entry 5). When the ligand anisyl rings in the
catalyst [Ni(oMeO-L3)X2] are additionally functionalized with
m-methyl moieties (i.e., oMeOmMe-L3, entries 7−9), the
conversion increases roughly by a factor of 1.5; [Ni-
(oMeOmMe-L3)I2] even gives full conversion of the substrate
in 1 h (TON = 1940).
E
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Next, the series of catalyst precursors with substituents on
the propylene backbone was tested, and the results are shown
in Table 5. Surprisingly, when L3X with unsubstituted aryl rings
reported to have the highest activity of a series of nickel
compounds with this ligand,20 attempts were undertaken to
synthesize similar complexes with the ligands oMeO-L3X,
oMeO-L3X*, and L3X. Despite the fact that different types of
compounds were formed with these ligands, their catalytic
activity in the hydrogenation of 1-octene has been investigated;
the results are given in Table 6. Because of the anticipated
Table 5. Conversion and Selectivity of [Ni(L)X2] Complexes
with Substituted L3-Type Ligands for Hydrogenation and
a
Isomerization Products of 1-Octene
conversion
TON
selectivity (%)
Table 6. Conversion and Selectivity for Hydrogenation and
Isomerization Products of 1-Octene Using Cationic Nickel
Complexes
entry
complex
%
n-octane octenes
a
1
[Ni(L3X)Cl2]
[Ni(L3X)Br2]
[Ni(L3X)I2]
C1
6
75
99
26
35
57
95
60
28
41
60
68
100
18
120
1460
1920
500
0
0
100
100
100
2
2
conversion
selectivity (%)
3
0
n-
4a
4b
5
98
96
91
93
96
93
94
90
96
94
100
entry
complex
%
TON octane octenes
b
C1
680
4
1
2
3
[Ni(oMeO-L3X)I2]
37
46
17
3580
4450
820
93
96
99
7
4
1
[Ni(oMeO-L3X)Br2]
[Ni(oMeO-L3X)I2]
[Ni(oMeO-L3XSi)Cl2]
1110
1840
1160
540
9
C4
6
7
b
[{Ni(oMeO-L3X)}2(μ-OH)2]
(PF6)2
7
4
8
[Ni(oMeO-L3X*)Cl2]
[Ni(oMeO-L3X*)Br2]
[Ni(oMeO-L3X*)I2]
7
4
5
[{Ni(oMeO-L3X*)}2(μ-OH)2]
11
530
100
0
9
800
6
(PF6)2
10
11
12
13
1160
1320
1940
3550
10
4
[{Ni(L3X)}2(μ-OH)2](PF6)2
0
0
0
0
b
[Ni(oEtO-L3X*)Cl2]
a
Reaction conditions: A total of 5.6 μmol of complex is dissolved in
28.5 mL of a 1.90 M 1-octene solution in CH3OH/CH2Cl2 (1:1) and
heated at 50 °C under 50 bar of H2 for 1 h. [complex] = 0.2 mM; 1-
octene/Ni = 9670 (entries 1 and 2) and 4830 (entries 3−5). The
TON (mol of substrate converted/mol of Ni) is also given.
C3
6
c
C3
0
a
Reaction conditions: A total of 27.9 μmol of complex (or 27.9 μmol
of NiCl2 and 41.9 μmol of ligand) dissolved in 28.5 mL of a 1.90 M 1-
octene solution in CH3OH/CH2Cl2 (1:1) was heated at 50 °C under
50 bar of H2 for 1 h. [Ni] = 1 mM; 1-octene/Ni = 1940. The TON
(mol of substrate converted/mol of Ni) is also given. Catalyst formed
in situ. 10 times less catalyst was used.
activity, the catalyst loading was 5 times lower [maximum TON
9670 for the mononuclear complex and 4830 for the dinuclear
complexes (based on Ni)]. For comparison, the catalytic
activity of [Ni(oMeO-L3X)I2] was also measured under these
conditions (entry 1). With C4 as the catalyst precursor (entry
2), 46% conversion is reached, corresponding to a TON of
4450, the highest of all complexes tested. The two dinuclear,
dicationic, hydroxide-bridged complexes comprising a ligand
with the oMeO functionality (entries 3 and 4) are about equally
reactive to their [Ni(L)Cl2] analogues. The use of the complex
with L3X (entry 5) does not result in any conversion at all.
Effect of Acid or Base on the Isomerization Activity.
Surprised by our finding that [Ni(L3X)I2] is an efficient
isomerization catalyst (Table 5), it was decided to study this
catalyst in more detail; the results are collected in Table 7. The
isomerization products formed in a standard experiment (entry
1) were identified as cis-2-octene (24%), trans-2-octene (28%),
and other internal alkenes (48%, >2-octenes). In another
reaction, 2 equiv (relative to Ni) of p-toluenesulfonic acid
(HOTs) was added (entry 2). Although the conversion again is
essentially quantitative, the product distribution reveals a higher
selectivity for cis-2-octene (60%). Conversely, the addition of 2
equiv (on Ni) of the base 1,8-bis(dimethylamino)naphthalene
(DMAN, entry 3) results in more trans-2-octene and >2-
octenes, also with full conversion of 1-octene. These three
experiments were repeated in the absence of H2 gas (entries 4−
6). Under a nitrogen atmosphere, approximately 14% 1-octene
is converted to mainly cis-2-octene when no acid or base is
added (entry 1); the addition of HOTs results in complete
suppression of the reactivity (entry 5). Interestingly, full
conversion is reached when DMAN is added (entry 6), yielding
cis-2-octene as the major product.
b
c
is employed as the ligand, active catalytic systems are obtained;
its NiCl2 and NiBr2 complexes give 6% and 75% conversion of
the substrate, yielding only isomerization products (entries 1
and 2). Nearly full conversion of 1-octene to isomerization
products is achieved when employing [Ni(L3X)I2], which thus
almost reached the maximum TON of 1940 within 1 h of
reaction time (entry 3). Catalysts formed with the ligand
oMeO-L3X (entries 4−6) show conversion and selectivity
similar to those of their simpler oMeO-L3 analogues (see Table
4). Interestingly, when the backbone in the ligand is made
slightly longer by substitution of the central (quaternary) C
atom in the ligand backbone with Si, the hydrogenation activity
is roughly doubled from ∼30% for C1 to 60% for [Ni(oMeO-
L3XSi)Cl2] (entries 4 and 7). When the size of the substituents
on the ligand backbone is increased by replacing the methyl
groups in oMeO-L3X with ethyl groups (entries 8−10), the
conversion drops from 95% for [Ni(oMeO-L3X)I2] (entry 6)
to 60% for [Ni(oMeO-L3X*)I2] (entry 10), also giving lower
selectivity for n-octane. On the other hand, when the steric bulk
of the substituent on the phenyl ring is enlarged from a
methoxy group (oMeO-L3X*) to an ethoxy group (oEtO-
L3X*), the conversion is roughly doubled from 30 to 70% for
the chloride complexes (entry 8 vs 11) and from 60 to 100%
for the iodide-containing compounds (entry 10 vs 12). Actually,
the pressure drop observed for the reaction employing C3 as
the catalyst precursor indicated that full conversion is already
reached after about 30 min. When this reaction is repeated
using 10 times less catalyst (entry 13), 18% conversion is
reached in 1 h, corresponding to a TON of 3550.
DISCUSSION
Catalytic Reactions Using Cationic Nickel Catalysts.
Because the mononuclear nickel compound [Ni(oMeO-L3)-
(H2O)2](PF6)2, with noncoordinating counterions, has been
■
Complex Synthesis and Structures. The synthetic
procedures to obtain nickel complexes are straightforward,
F
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 7. Activity of [Ni(L3X)I2] in the Isomerization of 1-Octene under a Hydrogen or Nitrogen Atmosphere and with and
a
without Added Acid (HOTs) or Base (DMAN)
conversion
TON
selectivity (%)
entry
atmosphere
additive
%
n-octane
cis-2-octene
trans-2-octene
>2-octenes
1
2
3
4
5
6
50 bar of H2
50 bar of H2
50 bar of H2
1 bar of N2
1 bar of N2
1 bar of N2
99
100
100
14
1920
1940
1940
270
0
0
0
0
0
0
24
60
13
75
0
28
11
33
8
48
29
54
17
0
2 equiv of HOTs
2 equiv of DMAN
2 equiv of HOTs
2 equiv of DMAN
0
0
0
100
1940
61
11
28
a
Reaction conditions: A total of 27.9 μmol of [Ni(L3X)I2] (and 55.8 μmol of HOTs or DMAN) dissolved in 28.5 mL of a 1.90 M 1-octene solution
in CH3OH/CH2Cl2 (1:1) was heated at 50 °C for 1 h under the indicated atmosphere. [Ni] = 1 mM; 1-octene/Ni = 1940.
and the complexes could be isolated in reasonable-to-good
(nonoptimized) yields and high purity. P2Ni complexes very
similar to C1−C3 have previously been reported.9,20,23,24
Compound C4 was formed in an attempt to exchange both
further support of the homogeneity of the system. The use of
the unsubstituted ligand L3 in combination with nickel salts
does not give any conversion of 1-octene. Functionalization of
the phenyl rings with o-methoxy groups leads to nickel
compounds that are active homogeneous hydrogenation
catalysts, which also produce small amounts of isomerization
products.12 The proposed mechanism for the hydrogenation
and isomerization activity of the nickel diphosphane complexes
is based on monohydridonickel(II) intermediate species and
activation of H2 via heterolytic cleavage, thus avoiding the
intermediacy of highly unstable nickel(0) or nickel(IV)
compounds.10
In this study, the selection of diphosphane ligands has been
extended to include ligands that are substituted at the central C
atom of the propylene linker and/or with additional
functionalization of the ligand aryl rings. Substitution of the
H atoms on the central atom by methyl (L3X) or ethyl (L3X*)
groups results in a more rigid structure; such a rigid backbone
ensures instantaneous complex formation of in situ formed
catalysts16 and leads to more stable complexes.17
In the reaction conditions used in this study, catalysts
comprising the ligand L3 do not show any hydrogenation or
isomerization activity; however, at higher temperatures and
with less-coordinating acetate anions, low conversion to mainly
isomerization products is observed.12 The use of the function-
alized analogues L3X and pMeO-L3 yielded active isomer-
ization catalysts, with cis-2-octene and trans-2-octene being the
main kinetic and thermodynamic isomerization products,
respectively.39 This change in activity is most likely due to
the more electron-donating nature of the ligands, facilitating
dissociation of the anions of the precatalyst to allow substrate
coordination. When the axial positions of the Ni ion are
sterically shielded by o-alkoxy groups on the ligand aryl rings
(i.e., oMeO-L3 and oMeO-L3X), the catalysts are almost
exclusively involved in hydrogenation. This dramatic difference
in selectivity can easily be understood because isomerization
requires the formation of a secondary nickel alkyl complex that
is sterically more crowded than the primary nickel alkyl species;
formation of the latter species will lead to rapid hydrogenation.
When the aryl rings in oMeO-L3 are further functionalized
(i.e., oMeOmMe-L3 vs oMeO-L3) or when the central C atom
in the backbone is replaced by Si (i.e., oMeO-L3XSi vs oMeO-
L3X), more active hydrogenation catalysts are obtained. As
expected, the catalytic activity (either hydrogenation or
isomerization) of the catalytic systems is enhanced by
employing less-coordinating anions, an effect that was reported
earlier.10,11 For each ligand series, the reactivity increases in the
order Cl < Br < I, which may be explained by the more ready
dissociation of the larger halide ions: the I− anion (rvdW = 1.98
−
acetate anions for PF6 using NH4PF6, following a procedure
that has been reported to yield [Ni(oMeOL3)(H2O)2]-
(PF6)2.20 Similarly, three bis(hydroxide)-bridged compounds
were obtained in attempts to make the corresponding bis(aqua)
complexes starting from nickel chloride solutions, using AgPF6
to avoid the formation of the ammonia adduct. At this moment,
we do not know what factors govern the formation of either
[P2Ni(NH3)(OAc)](PF6) or [{P2Ni(OH)}2](PF6)2 types of
complexes instead of [P2Ni(H2O)2](PF6)2 when this synthetic
procedure is applied. Compound C4 is truly unique;
compounds with a P2NiII(NH3)(OAc) or P2NiII(NH3)
coordination sphere thus far have not been reported.33 To
the best of our knowledge, only one example34 of a cis-P2NiNO
donor set has been reported, although in this compound both
the N- and O-donor atoms are anionic and all donor atoms
belong to the same ligand.35 An (L2Ni)2(μ-OH)2 cluster such
as that found in C5 is also rather rare (only 11 examples are
known),27 and merely two structures are known wherein L2 is a
diphosphane ligand.36,37
From the solid-state structures studied, it is clear that the o-
alkoxy-functionalized diphosphane ligands shield the axial
positions of the nickel(II) complex. This is a well-documented
phenomenon for both nickel(II)11,13,20 and palladium(II)16,18
complexes and is known to pertain in solution.13,16,18
Complexes without this o-alkoxy functionality, such as
[Ni(L3)Cl2],23 are known to be relatively unhindered in the
axial positions of nickel(II).
It is also clear that o-alkoxy substituents make the resulting
nickel(II) complex more sterically crowded in the equatorial
positions of the NiII ion, as reflected by distortion of the P2Ni/
NiX2 dihedral angle from the ideal 0°. For example, in
[Ni(L)(Halogen)2]-type complexes, this dihedral angle is about
6° when L = L3 or L3X, but it is 18° when L = oMeO-L3.38
This distortion becomes even larger when the steric bulk on the
ligand aryl rings is further enlarged, as exemplified by the P2Ni/
Ni(anion)2 dihedral angles of 20.31° and 22.78° observed in
respectively C2 and C3 (compare with 17.45° reported for
[Ni(oMeO-L3)Cl2]).20
Catalysis. General Structure−Activity Relationships. We
have earlier reported that the activity of the nickel diphosphane
catalysts for hydrogenation depends crucially on the ligand
structure. Also, under the reaction conditions used in our
studies the nickel catalysts are known to be truly homoge-
neous;12 the different activities and selectivities observed for the
catalysts with various ligands as described in this work are in
G
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Å) has a lower charge density than the Cl− anion (rvdW = 1.75
Å) and generally is less coordinating in nature.
the scarce and expensive catalysts that are currently used for
many reactions.
Catalyst Activation. Our observations employing [Ni(L3X)-
I2] (Table 7), in which added acid can suppress and added base
can promote the isomerization reaction (in the absence or
presence of a hydrogen atmosphere), have important
mechanistic implications. The activation of the catalyst
precursor [P2NiX2] is commonly assumed to proceed via
heterolytic cleavage of H2 by [P2NiX2], leading to [P2NiHX]
and 1 equiv of acid.11 However, as implicated by our results in
the absence of H2, another activation process must also be
possible in our reaction medium [1:1 (v/v) CH2Cl2/CH3OH].
We therefore propose that methanol can be used to form the
first metal hydride [P2NiHX] with the coproduction of HX. In
this reaction, the equilibrium [P2NiX2] + CH3OH ⇆
[P2NiX(OCH3)] + HX is followed by a β-hydrogen abstraction
of coordinated methoxide to form [P2NiH(OCH2)]X, which
may give [P2NiHX] by displacement of formaldehyde by the
anion. Such a process is similar to that proposed for the
[P2PdX2]-catalyzed oxidation of methanol in a CO-reducing
atmosphere when nitrobenzene is used as the oxidant.40−42
Both activation pathways imply that the addition of acid will
hamper the formation of the active species [P2NiHX] by
protonation of the hydride, whereas the addition of base will
facilitate [P2NiHX] formation, in agreement with our
observations.
EXPERIMENTAL SECTION
■
Materials. All chemicals were used as received unless stated
otherwise. Solvents were all of analytical reagent purity and were
obtained from Biosolve. 1-Octene and 1,3-bis(diphenylphosphanyl)-
propane (L3) were purchased from Acros Organics. All other ligands
were generously provided by Shell Global Solutions Amsterdam BV,
where they were synthesized according to literature procedures.43−51
Prior to use, all liquids were distilled under an argon atmosphere over
the appropriate drying agent [sodium metal (1-octene), calcium
hydride (dichloromethane and tetrahydrofuran), potassium carbonate
(acetone), or magnesium metal (methanol)].52 Complex syntheses
were performed under an inert atmosphere of dry argon. The
compounds [Ni(L3)Cl2],23 [Ni(oMeO-L3)Cl2],20 and [Ni(oMeO-
L3)Br2]20 were prepared following literature procedures; their purity
was checked with elemental analysis and NMR spectroscopy.
Analytical Methods. Elemental analyses were performed using a
Perkin-Elmer 2400 series II CHNS/O analyzer. The metal content was
determined with inductively coupled plasma optical emission spec-
troscopy (ICP-OES) using a Varian Vista-MPX CCD simultaneous
ICP-OES spectrometer. Samples were measured as <5% (m/m) metal
dilutions in 3% nitric acid. A Finnigan Aqua mass spectrometer with
ESI was used to record mass spectra. Sample introduction was
achieved through a Dionex ASI-100 automated sample injector with an
eluent flow rate of 0.2 mL min−1. The voltages of the capillary and
aquamax were set at 3 kV and 20 V, respectively. H{31P}, 13C, and
1
31P{1H} NMR spectra were recorded on a Bruker DPX300 (300
MHz) or a Bruker DMX400 (400 MHz) machine. Chemical shifts are
recorded in δ (parts per million) relative to the solvent peak (1H and
13C NMR) or relative to phosphoric acid (31P NMR). IR spectra were
recorded with 4 cm−1 resolution on a Perkin-Elmer Paragon 1000
Fourier transform infrared fitted with a Golden Gate Diamond ATR.
Samples from the catalytic experiments were analyzed with a
Hewlett-Packard 6890 series gas chromatograph equipped with an
autosampler. An AT-1 column (length, 30 m; diameter, 0.25 mm; film
thickness, 1.00 μm) was used as the stationary phase with helium as
the mobile phase. The column was maintained at a temperature of 50
°C, the injection port at 120 °C, and the detector at 250 °C
throughout the whole 40 min run time. An injection volume of 1 μL
was taken with a split ratio of 60:1.
CONCLUSIONS
■
A collection of nickel(II) complexes supported by bidentate
diarylphosphane ligands was synthesized, and of five of them,
the solid-state structure was investigated using X-ray diffraction.
The reactivity of these complexes was investigated for the
hydrogenation and/or isomerization of 1-octene.
It was found that catalysts comprising ligands with o-alkoxy
aryl rings can suppress the otherwise observed isomerization in
favor of hydrogenation, which is most likely due to the steric
constraint that these alkoxy moieties impose on the axial
positions of the P2NiII center. The conversion was found to
increase when the ligand aryl rings are equipped with electron-
donating alkoxy groups, when the steric bulk of the backbone
[CH2 < C(CH3)2 ≈ C(CH2CH3)2 < Si(CH3)2] and/or the aryl
rings (oH < pMeO ≈ oMeO < oMeOmMe < oEtO) is enlarged,
when the anions are noncoordinative in nature, and when a
catalytic amount of base is added.
The most active isomerization catalyst precursor tested is
[Ni(L3X)I2], with which un unprecedented 100% of the
substrate is converted (TON = 1940) solely to isomerization
products within only 1 h of reaction time. Notably, a similar
activity could be achieved under a nitrogen atmosphere at
ambient pressure when a base is added as the cocatalyst. The
novel compound [Ni(oMeO-L3X)(NH3)(OAc)]PF6 is the
most active hydrogenation catalyst precursor tested with an
observed TON of 4500 and 96% selectivity to n-octane within
1 h of reaction time, which is a record for nickel diphosphane
based catalysts.
Structure Determinations. The crystal structures of complexes
C1−C5 were determined using X-ray crystallography. A single crystal
was mounted to a glass fiber using the oil drop method. Reflections
were measured on a Nonius Kappa CCD diffractometer with a rotating
anode and a graphite monochromator (λ = 0.71073 Å). Software
packages used for the intensity integration were Eval1553 (C1 and C4)
and Eval1454 (C5). Absorption correction was performed with
SADABS.55 The structures were solved with SHELXS-9756 (C1−C3
and C5) or SIR-9757 (C4) and refined with SHELXL-9756 against F2 of
all reflections. Non-H atoms were refined freely with anisotropic
displacement parameters. H atoms were included in calculated
positions (C1 and C5) or located in difference Fourier maps (C4).
H atoms of N−H in C4 were refined freely with isotropic
displacement parameters, and all other H atoms were refined with a
riding model. The crystal of C5 was cracked into two fragments, and
only nonoverlapping reflections were used in the refinement
(completeness 87%).
The crystal structures of C1 and C5 contain large voids (2391 Å3/
unit cell in C1 and 499 Å3/unit cell in C5) filled with disordered
solvent molecules. Their contribution to the structure factors was
taken into account using the SQUEEZE routine of PLATON
software,58 resulting in 824 electrons/unit cell in C1 and 104
electrons/unit cell in C5. One of the chloroform solvent molecules in
C3 is disordered in two positions and was thus refined with population
parameters 0.75 and 0.25. Geometry calculations and checking for
higher symmetry were performed with the PLATON program.58
Further details are given in Table 2.
It is our hope that the information collected in this study
opens a venue for the development of even more active
catalytic systems for the hydrogenation and isomerization of
alkenes or for the development of nickel-based catalysts for,
e.g., hydroformylation or hydrocarboxylation. Studies on the
development of catalysts based on the abundant first-row
transition metals are important for the desired replacement of
H
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Catalytic Experiments. Stainless steel autoclaves (100 mL)
heated by a HEL polyBLOCK electrical heating system were used
to perform catalytic experiments. The autoclaves were equipped with
glass liners and magnetic stirbars that were both freshly washed with a
1:3 (v/v) HNO3/HCl mixture. The temperature and pressure were
measured with probes connected to a computer interface, making it
possible to record these parameters throughout the course of the
reaction. An amount of complex (typically 27.9 μmol), and when
appropriate an additive, was weighed into the glass liners. In the case
of in situ experiments, metal salt and ligand were weighed in a 1:1.5
(M/L) ratio. The autoclaves were tightly closed after insertion of the
liners and subsequently filled with argon through a Schlenk system.
Following this, 28.5 mL of a 1.90 M 1-octene stock solution in
dichloromethane/methanol (1:1, v/v) was added to the autoclaves
through one of the valves, resulting typically in a 1 mM nickel solution
with a 1-octene/Ni ratio of 1940. The autoclaves were inserted into
the heating block and put under 50 bar of H2 gas. After heating of the
autoclaves at 50 °C for 60 min (including the heating up of the
system), the autoclaves were cooled. After 30 min of cooling, room
temperature was reached, the remaining hydrogen pressure was
released, and the autoclave was opened to take a sample. Throughout
the reaction and the cooling process, a constant stirring speed of 500
rpm was maintained. Gas chromatography was used to analyze the
contents of the solution (vide supra).
Complex Synthesis. Unless stated otherwise, complexes were
prepared and isolated using the following general procedures:
[Ni(ligand)Cl2] Complexes. NiCl2·6H2O (240 mg, 1.00 mmol) was
dissolved in 6 mL of ethanol and then added dropwise to an equimolar
amount of the ligand dissolved in 6 mL of acetone or chloroform. The
solution changed color immediately, and after a few seconds, a
precipitate was formed. The solution was concentrated (if necessary),
after which the solid was collected by filtration. The product was
washed with diethyl ether and petroleum ether. Crystalline complexes
could be obtained by layering a solution of the complex in
dichloromethane with diethyl ether.
The mixture was refluxed overnight. After partial cooling, 0.5 mL of
hydrochloric acid (1 M in diethyl ether, 0.5 mmol) was added. The
solution turned darker red, but also some white precipitate appeared.
Therefore, more Ni(OAc)2·4H2O (12 mg, 0.05 mmol) and 0.12 mL of
hydrochloric acid (1 M in diethyl ether, 0.12 mmol) were added. After
refluxing for 3 h more, the solution was allowed to cool to room
temperature. Red crystals were formed and isolated by filtration and
washed with methanol and ethanol (58 mg, 0.097 mmol, 48%). Elem
anal. Found (calcd) for C31H34Cl2NiP2: Ni, 9.83 (9.81); C, 62.43
(62.25); H, 6.09 (5.73). ESI-MS. Found (calcd) for [M + H]+: m/z
596.09 (598.15). 1H NMR (300 MHz, CDCl3, 25 °C): δ 2.37 (s, 12H,
−CH3) 7.12 (m, 12H, Ph−H). 31P{1H} NMR (121 MHz, CDCl3, 25
°C): no peaks.
[Ni(oMeOmMe-L3)Cl2] (C2). From NiCl2·6H2O (174.45 mg, 0.734
mmol) and oMeOmMe-L3 (402.18 mg, 0.683 mmol), dark-purple
crystals were obtained (410 mg, 64%). Elem anal. Found (calcd) for
C35H42Cl2NiO4P2·1.5CHCl3·2H2O: C, 47.10 (46.97); H, 5.12 (5.13);
Ni, 6.34 (6.29). ESI-MS. Found (calcd) for [M − Cl]+: m/z 681.17
(681.16). 1H NMR (300 MHz, CD2Cl2, 25 °C): δ 1.65 (br, 2H,
PCH2CH2), 3.06 (vbr, 4H, PCH2CH2), 2.32 (s, 12H, −CH3), 3.86 (s,
12H, OCH3), 7.02 (d, 4H, m-H), 7.25 (d, 4H, p-H), 8.00 (br, 4H, o-
H). 13C NMR (75 MHz, CD2Cl2, 25 °C): δ 19.49 (s, PCH2CH2),
20.67 (s, −CH3), 56.07 (s, −OCH3), 111.56 (s, PCC(OMe)CH),
129.91 (s, PCCHCMe), 133.78 (s, PCCH), 140.89 (s, p-C), 161.09 (s,
PCC(OMe)CH). 31P{1H} NMR (121 MHz, CDCl3, 25 °C): δ 1.41
(vbr). IR (neat, cm−1): 2961 (w, CH), 1572 (w, Ph), 1486 (vs, Ph−
Me), 1471 (m, Ph), 1251 (st, CCO), 1019 (st, CO), 757 (vs, PPh).
[Ni(L3X)Cl2]. Starting from 237 mg (1.00 mmol) of NiCl2·6H2O and
436 mg (1.00 mmol) of L3X, 352 mg (0.62 mmol, 62%) of orange
crystals was isolated. Elem anal. Found (calcd) for C29H30Cl2NiP2: C,
60.14 (61.10); H, 5.52 (5.30). ESI-MS. Found (calcd) for [M − Cl]+:
m/z 534.61 (534.64). 1H NMR (300 MHz, MeOD, 25 °C): δ 1.00 (s,
6H, CH2C(CH3)2CH2), 2.31 (d, 4H, J = 3.2 Hz, CH2C(CH3)2CH2),
7.28 (m, 12H, o-Ph−H and p-Ph−H), 7.38 (m, 8H, m-Ph−H).
31P{1H} NMR (121 MHz, MeOD, 25 °C): no peaks.
Procedure A for [Ni(ligand)Br2] Complexes. NiBr2 (218.94 mg,
1.00 mmol) was dissolved in 0.4 mL of water, after which 10 mL of
ethanol was added. The resulting green solution was degassed and
added to the ligand (1.01 mmol) dissolved in 10 mL of dry and
degassed acetone or chloroform. Over time, a precipitate was obtained
from the resulting dark-red solution. The solid compounds were
collected by filtration and dried in vacuo.
Procedure B for [Ni(ligand)Br2] Complexes. A solution of the
ligand (1.00 mmol) in 10 mL of acetone or chloroform was added to a
suspension of NiBr2 (327 mg, 1.50 mmol) in 10 mL of ethanol. The
mixture was refluxed for 3−4 h and then slowly cooled to room
temperature. After cooling, the crystalline precipitate was collected by
filtration.
[Ni(ligand)I2] Complexes. Ni(OAc)2·4H2O (249 mg, 1.00 mmol)
was dissolved in 10 mL of ethanol and mixed with an equimolar
amount of ligand (1.00 mmol) in 10 mL of acetone or chloroform. To
this solution was added 0.25 mL of concentrated hydrogen iodide
(55−58% in water, 1.80 mmol), and a dark-purple solution was
obtained. If necessary, the solution was refluxed for 3.5 h, after which
the solution was slowly cooled. The crystalline precipitate was isolated
by filtration.
[Ni(oMeO-L3X)Cl2] (C1). Starting from 237 mg (1.00 mmol) of
NiCl2·6H2O and 563 mg (1.00 mmol) of oMeO-L3X, 619 mg (0.90
mmol, 90%) of red crystals was isolated. Elem anal. Found (calcd) for
C33H38Cl2NiO4P2: C, 57.04 (57.43); H, 5.80 (5.55). ESI-MS. Found
1
(calcd) for [M − Cl]+: m/z 654.73 (654.75). H NMR (300 MHz,
CDCl3, 25 °C): δ 0.58 (s, 6H, CH2C(CH3)2CH2), 2.7 (br, 4H,
CH2C(CH3)2CH2), 3.98 (s, 12H, −OCH3), 7.02 (d, 4H, J = 7.8 Hz,
−C(OMe)CH−), 7.10 (t, 4H, J = 6.9 Hz, −PCCHCH−), 7.48 (t, 4H,
J = 7.5 Hz, p-Ph−H), 8.38 (br, 4H, o-Ph−H). 31P{1H} NMR (121
MHz, CDCl3, 25 °C): no peaks.
[Ni(oMeO-L3X*)Cl2]. Starting from 237 mg (1.00 mmol) of
NiCl2·6H2O and 590 mg (1.00 mmol) of oMeO-L3X*, 369 mg
(0.51 mmol, 51%) of red crystals was isolated. Elem anal. Found
(calcd) for C35H42Cl2NiO4P2: C, 58.42 (58.53); H, 5.91 (5.89). ESI-
1
MS. Found (calcd) for [M + H]+: m/z 682.75 (682.80). H NMR
(300 MHz, CDCl3, 25 °C): δ 0.47 (t, 6H, J = 7.5 Hz,
CH2C(CH2CH3)2CH2), 0.83 (q, 4H, J = 7.5 Hz, CH2C-
(CH2CH3)2CH2), 2.5 (vbr, CH2C(CH2CH3)2CH2), 3.95 (s, 12H,
−OCH3), 7.08 (br, 8H, 2 × m-Ph−H), 7.47 (t, 4H, J = 7.8 Hz, p-Ph−
H), 8.41 (br, 4H, o-Ph−H). 31P{1H} NMR (121 MHz, CDCl3, 25
°C): no peaks.
[(Ni(ligand))2(μ-OH)2](PF6)2-Type Complexes. NiCl2·6H2O (178
mg, 0.75 mmol) was dissolved in 5 mL of ethanol and then added
dropwise to an equimolar amount of ligand dissolved in 5 mL of
chloroform. The solution changed color immediately, and after a few
seconds, a precipitate had formed. The precipitate was filtered and
then dissolved in 100 mL of dichloromethane. To this solution was
added a solution of AgPF6 (475 mg, 1.75 mmol) in 10 mL of water.
The two-layer system was stirred for 2 days. Most of the water was
taken from the mixture by using a pipet, and dichloromethane was
dried over magnesium sulfate. Dichloromethane was slowly put under
diethyl ether using a pipet, resulting in a two-layer system. After a few
days, a crystalline compound formed and was isolated by filtration.
[Ni(oMe-L3)Cl2]. Ni(OAc)2·4H2O (51 mg, 0.20 mmol) was mixed
with an equimolar amount of oMe-L3 (94 mg) in 20 mL of ethanol.
[Ni(L3)Br2]. Following procedure A, 475 mg (75%) of a brick-red
solid was obtained from NiBr2 (218.94 mg, 1.00 mmol) and L3 (416.2
mg, 1.01 mmol). Elem anal. Found (calcd) for C27H26Br2NiP2: C,
51.51 (51.40); H, 4.04 (4.15); Ni, 10.26 (9.30). ESI-MS. Found
1
(calcd) for [M − 2Br + OH]+: m/z 488.50 (487.09). H NMR (300
MHz, CD2Cl2, 25 °C): δ −2.95 (br, 2H, PCH2CH2), 2.03 (br, 4H,
PCH2CH2), 2.03 (m, 4H, p-H), 4.18 (m, 4H, m-H), 13.69 (m, 4H, o-
H). 31P{1H} NMR (121 MHz, CD2Cl2, 25 °C): no signal. IR (neat,
cm−1): 3048 (w, CH), 1433 (s, CH), 1098 (s, PPh), 687 (vs, PPh).
[Ni(oMeOmMe-L3)Br2]. Following procedure A and starting from
NiBr2 (226.65 mg, 1.04 mmol) and oMeOmMe-L3 (591.5 mg, 1.00
mmol), 495 mg (61%) of a purple solid was isolated. Elem anal. Found
(calcd) for C35H42Br2NiO4P2: C, 52.43 (52.08); H, 5.90 (5.24); Ni,
7.53 (7.27). ESI-MS. Found (calcd) for [M − Br]+: m/z 726.89
I
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(727.11). 1H NMR (300 MHz, CD2Cl2, 25 °C): δ 1.43 (br, 2H,
PCH2CH2), 2.21 (s, 12H, −CH3), 3.92 (s, 12H, OCH3), 7.09 (m, 8H,
m-H + p-H), 7.87 (br, 4H, o-H). 13C NMR (75 MHz, CD2Cl2, 25 °C):
δ 21.55 (s, −CH3), 57.37 (s, −OCH3), 112.70 (s, PCC(OMe)CH),
130.71 (s, PCCHCMe), 135.21 (s, PCC). 31P{1H} NMR (121 MHz,
CDCl3, 25 °C): no signal. IR (neat, cm−1): 2922 (w, CH), 1575 (w,
Ph), 1486 (vs, Ph-Me), 1247 (vs, CCO), 1014 (s, CO), 755 (m,
PPh).
ESI-MS. Found (calcd) for [M − I]+: m/z 746.61 (746.20). 1H NMR
(300 MHz, CDCl3, 25 °C): δ 0.51 (s, 6H, CH2C(CH3)2CH2), 2.80
(br, 4H, CH2C(CH3)2CH2), 3.97 (s, 12H, −OCH3), 7.02 (d, 4H, J =
8.3 Hz, −C(OMe)CH−), 7.10 (t, 4H, J = 7.4 Hz, −PCCHCH-), 7.46
(t, 4H, J = 7.4 Hz, p-Ph-H), 8.38 (br, 4H, o-Ph-H). 31P{1H} NMR
(121 MHz, CDCl3, 25 °C): no peaks.
[Ni(oMeO-L3X*)I2]. Starting from 249 mg (1.00 mmol) of
Ni(OAc)2·4H2O and 590 mg (1.00 mmol) of oMeO-L3X*, 615 mg
(0.68 mmol, 68%) of dark-purple crystals was isolated. Elem anal.
Found (calcd) for C35H42I2NiO4P2: C, 47.48 (46.65); H, 4.73 (4.70).
ESI-MS. Found (calcd) for [M − I]+: m/z 772.75 (774.25). 1H NMR
(300 MHz, CDCl3, 25 °C): δ 0.46 (t, 6H, J = 7.1 Hz,
CH2C(CH2CH3)2CH2), 0.77 (q, 4H, J = 6.9 Hz, CH2C-
(CH2CH3)2CH2), 2.5 (vbr, 4H, CH2C(CH2CH3)2CH2), 3.95 (s,
12H, −OCH3), 7.11 (br, 8H, 2 × m-Ph−H), 7.45 (br, 4H, p-Ph−H),
8.39 (br, 4H, o-Ph−H). 31P{1H} NMR (121 MHz, CDCl3, 25 °C): no
peaks.
[Ni(L3X)Br2]. Starting from 327 mg (1.50 mmol) of NiBr2 and 441
mg (1.00 mmol) of L3X and employing procedure B, 639 mg (0.97
mmol, 97%) of brown-reddish crystals was isolated. Elem anal. Found
(calcd) for C29H30Br2NiP2: C, 53.30 (52.85); H, 4.98 (4.59). ESI-MS.
1
Found (calcd) for [M + H]+: m/z 659.84 (659.00). H NMR (300
MHz, DMSO-d6, 25 °C): δ 0.93 (s, 6H, CH2C(CH3)2CH2), 2.27 (d,
4H, J = 2.9 Hz, CH2C(CH3)2CH2), 7.26 (m, 12H, o-Ph−H and p-Ph−
H), 7.34 (m, 8H, m-Ph−H). 31P{1H} NMR (121 MHz, DMSO-d6, 25
°C): no peaks.
[Ni(oMeO-L3X)Br2]. Employing procedure B while starting from 327
mg (1.50 mmol) of NiBr2 and 559 mg (1.00 mmol) of oMeO-L3X,
316 mg (0.41 mmol, 41%) of dark-pink crystals was isolated. Elem
anal. Found (calcd) for C33H38Br2NiO4P2: C, 50.01 (50.87); H, 5.06
(4.92). ESI-MS. Found (calcd) for [M − Br]+: m/z 698.65 (699.20)
1H NMR (300 MHz, CDCl3, 25 °C): δ 0.55 (s, 6H, CH2C-
(CH3)2CH2), 3.00 (vbr, 4H, CH2C(CH3)2CH2), 3.99 (s, 12H,
−OCH3), 7.09 (m, 8H, 2 × m-Ph−H), 7.45 (t, 4H, J = 7.8 Hz, p-
Ph−H), 8.37 (br, 4H, o-Ph−H). 31P{1H} NMR (121 MHz, CDCl3, 25
°C): no peaks.
[Ni(oMeO-L3X*)Br2]. Using procedure B and starting from 327 mg
(1.50 mmol) of NiBr2 and 588 mg (1.00 mmol) of oMeO-L3X*, 513
mg (0.64 mmol, 42%) of pink crystals was isolated after
recrystallization from a dichloromethane/diethyl ether mixture. Elem
anal. Found (calcd) for C35H42Br2NiO4P2: C, 53.36 (52.08); H, 5.31
(5.24). ESI-MS. Found (calcd) for [M − Br]+: m/z 726.82 (727.25).
1H NMR (300 MHz, CDCl3, 25 °C): δ 0.47 (t, 6H, J = 7.0 Hz,
CH2C(CH2CH3)2CH2), 0.82 (q, 4H, J = 7.2 MHz, CH2C-
(CH2CH3)2CH2), 2.5 (vbr, CH2C(CH2CH3)2CH2), 3.97 (s, 12H,
−OCH3), 7.14 (br, 8H, 2 × m-Ph−H), 7.44 (t, 4H, J = 7.4 Hz, p-Ph−
H), 8.40 (br, 4H, o-Ph−H). 31P{1H} NMR (121 MHz, CDCl3, 25
°C): no peaks.
[Ni(oEtO-L3X*)I2] (C3). Ni(OAc)2·4H2O (47.9 mg, 0.19 mmol) was
dissolved in 5 mL of ethanol. The resulting green solution was
degassed and added to oEtO-L3X* (112.3 mg, 0.19 mmol) dissolved
in 5 mL of acetone. This resulted in a brown solution. The subsequent
addition of 2 mL of hydroiodic acid (57 wt %) gave a precipitate.
Filtration and drying in vacuo yielded a dark-purple powder (86.00
mg, 50%). Elem anal. Found (calcd) for C39H50I2NiO4P2: C, 48.47
(48.93); H, 5.51 (5.26); Ni, 6.57 (6.13). ESI-MS. Found (calcd) for
[M − 2I + OH]+: m/z 720.33 (719.26). 1H NMR (300 MHz, CD2Cl2,
25 °C): δ 0.47 (t, 6H, PCH2C(CH2CH3)2), 0.78 (q, 4H,
PCH2C(CH2CH3)2), 1.13 (br, 12H, OCH2CH3), 1.49 (br, 4H,
PCH2C(CH2CH3)2), 4.31 (br, 8H, OCH2CH3), 7.20 (br, 8H, m-H),
7.39 (br, 4H, p-H), 8.16 (br, 4H, o-H). 13C NMR (75 MHz, CD2Cl2,
25 °C): 6.94 (s, PCH2C(CH2CH3)2), 14.44 (s, OCH2CH3), 30.83 (s,
PCH2C(CH2CH3)2), 64.29 (s, OCH2CH3), 104.69 (s, PCCHCH),
167.41 (s, PCC(OMe)CH). 31P{1H} NMR (121 MHz, CDCl3, 25
°C): no signal. IR (neat, cm−1): 2978 (w, CH), 1573 (m, Ph), 1480
(m, Ph), 1436 (s, ROR), 1259 (m, CCO), 1031 (s, CO), 750 (vs,
PPh).
[(Ni(L3X))2(μ-OH)2](PF6)2. Starting from 178 mg (0.75 mmol) of
NiCl2·6H2O and 330 mg (0.75 mmol) of L3X, 37 mg (0.03 mmol,
8%) of orange crystals was isolated. Elem anal. Found (calcd) for
C58H62F12Ni2O2P6: C, 53.12 (52.68); H, 5.19 (4.73). ESI-MS. Found
(calcd) for [M]2+: m/z 516.30 [516.20]. 1H NMR (300 MHz, MeOD,
25 °C): δ 0.60 (s, 12H, CH2C(CH3)2CH2), 1.1 (d, 8H, CH2C-
(CH3)2CH2), 7.39 (br, 24H, o-Ph−H and p-Ph−H), 7.63 (br, 16H, m-
Ph−H). 31P{1H} NMR (121 MHz, MeOD, 25 °C): δ 17.3. IR (neat,
cm−1): 731.6 (Ar−H), 830.0 (P−F), 2951.7 (CH2 alkane), 3607.0 (μ-
OH).
[Ni(oMeOmMe-L3)I2]. Ni(OAc)2·4H2O (248.2 mg, 1.00 mmol) and
oMeOmMe-L3 (597.9 mg, 1.00 mmol) were dissolved in 5 mL of
acetone, resulting in a dark-red-brown solution. The subsequent
addition of 3 mL of hydroiodic acid (57 wt %) gives a black solution.
Layered recrystallization of the obtained precipitate in dichloro-
methane/heptane resulted in the formation of deep-purple crystals.
Filtration and drying in vacuo yielded purple crystals (165 mg, 18%).
Elem anal. Found (calcd) for C35H42I2NiO4P2: C, 47.00 (46.65); H,
4.54 (4.70); Ni, 7.41 (6.51). ESI-MS. Found (calcd) for [M − I]+: m/z
[(Ni(oMeO-L3X))2(μ-OH)2](PF6)2 (C5). From 178 mg (0.75 mmol)
of NiCl2·6H2O and 420 mg (0.75 mmol) of oMeO-L3X, 266 mg (0.17
mmol, 45%) of red needlelike crystals was isolated. Elem anal. Found
(calcd) for C66H78F12Ni2O10P6: C, 49.51 (50.73); H, 5.28 (5.03). ESI-
1
773.3 (773.1). H NMR (300 MHz, CD2Cl2, 25 °C): δ 1.56 (br, 2H,
PCH2CH2), 2.30 (s, 12H, −CH3), 4.03 (s, 12H, −OCH3), 7.16 (br,
4H, m-H), 7.16 (br, 4H, p-H), 7.96 (br, 4H, o-H). 13C NMR (75 MHz,
CD2Cl2, 25 °C): δ 16.19 (s, PCH2CH2), 19.79 (s, PCH2CH2), 21.00
(s, −CH3), 57.13 (s, OCH3), 112.22 (s, PCC(OMe)CH), 129.93 (s,
PCCHCMe), 134.60 (s, PCCH), 142.01 (s, PCCH), 163.68
(PCC(OMe)CH). 31P{1H} NMR (121 MHz, CDCl3, 25 °C): no
signal. IR (neat, cm−1): 2980 (w, CH), 1600 (w, Ph), 1495 (vs, Ph−
Me), 1249 (m, Ph), 1249 (vs, CCO), 1010 (s, CO), 750 (s, PPh).
[Ni(L3X)I2]. Starting from 249 mg (1.00 mmol) of Ni(OAc)2·4H2O
and 440 mg (1.00 mmol) of L3X, 727 mg (0.97 mmol, 97%) of dark-
purple crystals was isolated. Elem anal. Found (calcd) for
C29H30I2NiP2: C, 46.05 (46.26); H, 4.11 (4.02). ESI-MS. Found
1
MS. Found (calcd) for [M]2+: m/z 637.33 (636.30). H NMR (300
MHz, MeOD, 25 °C): δ 0.59 (s, 12H, CH2C(CH3)2CH2), 2.20 (t, 8H,
J = 5.7 Hz, CH2C(CH3)2CH2), 3.91 (s, 24H, −OCH3), 7.09 (m, 16H,
2 × m-Ph−H), 7.59 (m, 8H, p-Ph−H), 7.61 (br, 8H, o-Ph−H).
31P{1H} NMR (121 MHz, MeOD, 25 °C): δ 15.7 (s), −143.6 (pentet,
J = 708 Hz, PF6). 19F NMR (282 MHz, CDCl3, 25 °C): δ 3.46 (s,
PF6), 5.36 (s, PF6) ppm. IR (neat, cm−1): 764.3 (Ar−H), 836.3 (P−
F), 1245.9 (Ar−O−CH3), 2946.1 (CH2 alkane), 3613.2 (μ-OH).
[(Ni(oMeO-L3X*))2(μ-OH)2](PF6)2. Starting from 178 mg (0.75
mmol) of NiCl2·6H2O and 442 mg (0.75 mmol) of oMeO-L3X*,
174 mg (0.11 mmol, 29%) of red crystals was isolated. Elem anal.
Found (calcd) for C70H86F12Ni2O10P6: C, 51.86 (51.94); H, 5.77
1
(calcd) for [M + H]+: m/z 751.08 (753.00). H NMR (300 MHz,
CDCl3, 25 °C): δ 0.52 (s, 6H, CH2C(CH3)2CH2), 1.24 (d, 4H, J = 5.8
Hz, CH2C(CH3)2CH2), 6.72 (t, 4H, J = 7.4 Hz, p-Ph−H), 7.62 (d,
8H, J = 7.6 Hz, o-Ph−H), 8.54 (t, 8H, J = 7.3 Hz, m-Ph−H). 31P{1H}
NMR (121 MHz, CDCl3, 25 °C): no peaks.
[Ni(oMeO-L3X)I2]. Starting from 249 mg (1.00 mmol) of
Ni(OAc)2·4H2O and 561 mg (1.00 mmol) of oMeO-L3X, 765 mg
(0.88 mmol, 88%) of dark-purple crystals was isolated. Elem anal.
Found (calcd) for C33H38I2NiO4P2: C, 45.55 (45.40); H, 4.59 (4.39).
1
(5.36). ESI-MS. Found (calcd) for [M]2+: m/z 664.25 (664.35). H
NMR (300 MHz, MeOD, 25 °C): δ 0.38 (t, 12H, J = 7.2 Hz,
CH2C(CH2CH3)2CH2), 0.81 (q, 8H, J = 7.5 Hz, CH2C-
(CH2CH3)2CH2), 2.08 (t, 8H, J = 6.0 Hz, CH2C(CH2CH3)2CH2),
3.80 (s, 24H, −OCH3), 6.98 (t, 16H, J = 8.0 Hz, 2 × m-Ph−H), 7.55
(t, 8H, J = 8.0 Hz, p-Ph−H), 7.80 (br, 8H, o-Ph−H). 31P{1H} NMR
(121 MHz, MeOD, 25 °C): δ 14.3 (s), −143.4 (pentet, J = 711 Hz,
J
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
PF6). IR (neat, cm−1): ν 755.3 (Ar−H), 835.2 (P−F), 1242.8 (Ar−O−
CH3), 2962.8 (CH2 alkane), 3609.2 (μ-OH).
(12) Angulo, I. M.; Kluwer, A. M.; Bouwman, E. Chem. Commun.
1998, 24, 2689−2690.
[Ni(oMeO-L3X)(NH3)(OAc)](PF6) (C4). Ni(OAc)2·4H2O (124 mg,
0.50 mmol) was dissolved in 15 mL of ethanol and mixed with an
equimolar amount of oMeO-L3X (280 mg) in 12.5 mL of chloroform.
After the solutions were mixed, a clear red solution was obtained.
Then, NH4PF6 (245 mg, 1.50 mmol) in 5 mL of ethanol was added,
and a clear orange solution was obtained. The solution was slowly
pipetted under a layer of diethyl ether. After a few days, a yellow/
golden crystalline compound was formed and collected by filtration
(354 mg, 0.42 mmol, 83%). Elem anal. Found (calcd) for
C35H44F6NNiO6P3: C, 50.06 (50.02); H, 5.36 (5.28); N, 1.73
(1.67). ESI-MS. Found (calcd) for [M − NH3]+ and [M − OAc]+,
(13) Angulo, I. M.; Bouwman, E.; Lutz, M.; Mul, W. P.; Spek, A. L.
Inorg. Chem. 2001, 40, 2073−2082.
(14) Gramlich, V.; Consiglio, G. Helv. Chim. Acta 1979, 62, 1016−
1024.
(15) Molander, G. A.; Burke, J. P.; Carroll, P. J. J. Org. Chem. 2004,
69, 8062−8069.
(16) Mooibroek, T. J.; Bouwman, E.; Lutz, M.; Spek, A. L.; Drent, E.
Eur. J. Inorg. Chem. 2010, 298−310.
(17) van Rijn, J. A.; Siegler, M. A.; Spek, A. L.; Bouwman, E.; Drent,
E. Organometallics 2009, 28, 7006−7014.
(18) Mooibroek, T. J.; Lutz, M.; Spek, A. L.; Bouwman, E. Dalton
Trans. 2010, 39, 11027−11034.
1
respectively: m/z 678.88 (678.34), 635.83 (635.32). H NMR (300
MHz, MeOD, 25 °C): δ 0.71 (s, 6H, CH2C(CH3)2CH2), 1.06 (s, 3H,
OAc), 2.39 (br, 4H, CH2C(CH3)2CH2), 3.88 (s, 6H, −OCH3), 3.97
(s, 6H, −OCH3), 7.13 (br, 8H, 2 × m-Ph−H), 7.65 (br, 4H, p-Ph−H),
8.11 (br, 4H, o-Ph−H). 31P{1H} NMR (121 MHz, MeOD, 25 °C): δ
13.8 (s, P trans to NH3), 13.1 (s, P trans to OAc), −143.4 (pentet, J =
705 Hz, PF6). IR (neat, cm−1): 755.3 (Ar−H), 834.2 (P−F), 1246.7
(Ar−O−CH3), 2967.0 (CH2 alkane).
(19) King, R. E.; Miller, S. B.; Knobler, C. B.; Hawthorne, M. F.
Inorg. Chem. 1983, 22, 3548−3554.
(20) Angulo, I. M.; Bouwman, E.; Lok, S. M.; Lutz, M.; Mul, W. P.;
Spek, A. L. Eur. J. Inorg. Chem. 2001, 1465−1473.
(21) Desrochers, P. J.; Telser, J.; Zvyagin, S. A.; Ozarowski, A.;
Krzystek, J.; Vicic, D. A. Inorg. Chem. 2006, 45, 8930−8941.
(22) Hayter, R. G.; Humiec, F. S. Inorg. Chem. 1965, 4, 1701−1706.
(23) Bomfim, J. A. S.; de Souza, F. P.; Filgueiras, C. A. L.; de Sousa,
A. G.; Gambardella, M. T. P. Polyhedron 2003, 22, 1567−1573.
(24) Muth, A.; Reinhard, G.; Huttner, G.; Seitz, T.; Klein, T.;
Zsolnai, L. Z. Naturforsch., B: Chem. Sci. 1994, 49, 889−897.
(25) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen,
A. G.; Taylor, R. J. Chem. Soc., Perkin Trans. 2 1987, S1.
(26) Orpen, A. G.; Brammer, L.; Allen, F. H.; Kennard, O.; Watson,
D. G.; Taylor, R. Dalton Trans. 1989, S1.
ASSOCIATED CONTENT
* Supporting Information
■
S
CIF files, perspective views of C1−C3 (Figure S1), and
activities and selectivities of catalytic experiments (Table S1).
This material is available free of charge via the Internet at
http://pubs.acs.org.
(27) The Cambridge Structure Database, version 5.33 (Feb 2012,
including three updates), was consulted for (L2Ni)2(μ-OH)2 clusters
using ConQuest. A total of 13 CIFs were found, 11 of which contain a
(L2Ni)2(μ-OH)2 motif: BIXBAM (L2 = NN), BIXBEQ (L2 = NN),
GAKDOL (L2 = PP), JOWBOM (L2 = CC), LUVGOY (L2 = CCl),
LUVGOY01 (L2 = CCl), LUVVIH (L2 = NO), MEDTEV (L2 = NN),
SAVVIU (L2 = CP), YOHJOV (L2 = NN), and ZOMMET (L2 = PP);
the other two (FOCKOX and FOCKOX10, L2 = CP) are trinuclear.
The average Ni···Ni distance is 2.867 0.071 Å (2.5%).
(28) Bondi, A. J. Phys. Chem. 1964, 68, 441−451.
AUTHOR INFORMATION
Corresponding Author
*E-mail: bouwman@chem.leidenuniv.nl.
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research has been financially supported by the Council for
Chemical Sciences of The Netherlands Organisation for
Scientific Research. This work has been performed under the
auspices of the joint NIOK Research Graduate School of
Leiden University and six other Dutch Universities. We thank
Shell Global Solutions International BV for providing samples
of diphosphane ligands.
(29) Brookhart, M.; Green, M. L. H.; Parkin, G. Proc. Natl. Acad. Sci.
U.S.A. 2007, 104, 6908−6914.
(30) A total of 19 CIFs were found in the Cambridge Structure
Database (version 5.33, Nov 2011, including four updates) that
contained a P2Ni unit, wherein at least one of the P-donor ligands had
an o-alkoxy substituent. The oxidation state of Ni was 2+ for all
structures except POHMEE (Ni3+) and XOBJOO (Ni0). No short
Ni···O(alkoxyaryl) distances are present in OKUFUV, QOFCAP,
QOFCET, QOSTEX, QOSTIB, XABPIA, and XOBJII (all containing
a bidentate PP ligand) or in XAWVIB, ZAXPEV, ZAXPIZ, ZAXPOF,
and ZAXPUL (all containing a bidentate PO ligand). Short
Ni···O(alkoxyaryl) distances were found in MUPFOT (2.621 and
2.604 Å), MUPFUZ (2.948 Å), and MUPGAG (2.861 Å), which all
contain the (for MUPFUZ and MUPGAG demethylated) tetradentate
P-ligand 1,2,3,4-tetrakis[bis(2-methoxyphenyl)phosphino]cyclobu-
tane. The two short Ni···O(alkoxyaryl) distances (both 2.785 Å) in
POHMAA (containing two bidentate PO ligands) might well be due
to the two very bulky 2,4,6-trimethoxyphenyl substituents on each P-
donor atom. Finally, a short Ni···O(alkoxyaryl) distance (2.866 Å) is
present in QOSTAT, which has a [P2Ni(H2O)2]2+ coordination
sphere.
REFERENCES
■
(1) Singh, D.; Rezac, M.; Pfromm, P. J. Am. Oil Chem. Soc. 2009, 86,
93−101.
(2) Veldsink, J. W.; Bouma, M. J.; Schoon, N. H.; Beenackers, A. A.
̈
̈
C. M. Catal. Rev. 1997, 39, 253−318.
(3) Kuil, M.; Soltner, T.; van Leeuwen, P. W. N. M.; Reek, J. N. H. J.
Am. Chem. Soc. 2006, 128, 11344−11345.
(4) Watkins, A. L.; Hashiguchi, B. G.; Landis, C. R. Org. Lett. 2008,
10, 4553−4556.
(5) Napoli, M.; Mariconda, A.; Immediata, I.; Longo, P. J. Polym.
Chem., Part A-1: Polym. Chem. 2008, 46, 4725−4733.
(6) Crabtree, R. H. The organometallic chemistry of the transition
metals; Wiley-Interscience: New York, 2005.
(7) www.metalprices.com (Feb 2012)
(31) A total of 12 CIFs were found in the Cambridge Structure
Database (version 5.33, Nov 2011, including four updates) containing
a MP(o-alkoxyaryl)2CCCP(o-alkoxyaryl)2 ring (M = any metal except
Ni). Neither BUGVOP, BUGVUV, BUGWAC, JOCJAN, MEKTAY,
MEKTEC, and PEYXEX ([P2Pd2+Anion2] type), nor BUGWEG,
EKUQOR, and EKUQUX ([(P2)2Pd2+]2+ type), nor TUCSUG
([P2Ru2+CpCl] type) exhibit a M···O(alkoxyaryl) interaction. Short
Pd···O(alkoxyaryl) distances (2.811 and 2.825 Å) were found in
(8) Cammack, R. In Bioinorganic Chemistry; Reedijk, J., Ed.; Marcel
Dekker Inc.: New York, 1993; Chapter 7.
(9) Angulo, I. M.; Bouwman, E.; van Gorkum, R.; Lok, S. M.; Lutz,
M.; Spek, A. L. J. Mol. Catal. A: Chem. 2003, 202, 97−106.
(10) Angulo, I. M.; Lok, S. M.; Norambuena, V. F. Q.; Lutz, M.;
Spek, A. L.; Bouwman, E. J. Mol. Catal. A: Chem. 2002, 187, 55−67.
(11) Angulo, I. M. Ph.D. Thesis, Leiden University, Leiden, The
Netherlands, 2001,
K
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
FOQLIH, which contains the tetradentate P-ligand 1,2,3,4-tetrakis[-
bis(2-methoxyphenyl)phosphino]cyclobutane.
(32) Dennett, J. N. L.; Gillon, A. L.; Heslop, K.; Hyett, D. J.;
Fleming, J. S.; Lloyd-Jones, C. E.; Orpen, A. G.; Pringle, P. G.; Wass,
D. F.; Scutt, J. N.; Weatherhead, R. H. Organometallics 2004, 23,
6077−6079.
(33) QAMGUG contains a PNi+(NH3)(η3-allyl) coordination
sphere, which is the only known example that is somewhat similar
to C4.
(34) Scherhag, G.; Spicer, M. D. J. Chem. Soc., Dalton Trans. 2000,
1237−1238.
(35) The Cambridge Structure Database (version 5.33, Nov, 2011
including four updates) contains five Ni2+ complexes with a PPNO
donor set. In BOCTUJ, LOFZOV, and XOMYUU, both O- and N-
donor atoms are anionic, while in KUCXAI and KUCXAI01, the N-
donor atom is charge-neutral. Only LOFZOV has a cis-P2NO donor
set.
(36) Ghilardi, C. A.; Innocenti, P.; Midollini, S.; Orlandini, A. J.
Chem. Soc., Dalton Trans. 1988, 1063−1066.
(37) Reinhard, G.; Soltek, R.; Huttner, G.; Barth, A.; Walter, O.;
Zsolnai, L. Chem. Ber. 1996, 129, 97−108.
(38) The dihedral angle between the P2Ni and Ni(anion)2 planes is
6.72° in [Ni(L3)Cl2] (Polyhedron 2003, 22, 1567), 5.15° in [Ni(L3X)
Br2] (Z. Naturforsch., B: Chem. Sci. 1994, 49, 889), 17.45° in
[Ni(oMeO-L3)Cl2] (Eur. J. Inorg. Chem. 2001, 1465), and 18.65° in
[Ni(oMeO-L3)Br2.
(39) Rogers, D. W.; Dejroongruang, K.; Samuel, S. D.; Fang, W.;
Zhao, Y. P. J. Chem. Thermodyn. 1992, 24, 561−565.
(40) Mooibroek, T. J.; Schoon, L.; Bouwman, E.; Drent, E. Chem.
Eur. J. 2011, 17, 13318−13333.
(41) Mooibroek, T. J.; Bouwman, E.; Drent, E. Eur. J. Inorg. Chem.
2012, 1403−1412.
(42) Mooibroek, T. J.; Bouwman, E.; Drent, E. Organometallics 2012,
31, 4142−4156.
(43) Hobbs, C. F. (Monsanto Company). U.S. Patent 4,120,901,
1987.
(44) Bickelhaupt, F. (Shell Oil Company). U.S. Patent 4,874,897,
1989.
(45) Eilenberg, W. (Shell Oil Company). Eur Patent 364046, 1991.
(46) van Doorn, J. A. (Shell Oil Company). U.S. Patent 4,994,592,
1991.
(47) Drent, E. (Shell Oil Company). U.S. Patent 5,091,587, 1992.
(48) van Ginkel, R. (Shell Oil Company). U.S. Patent 6,548,708,
2003.
(49) Clark, P. W.; Mulraney, B. J. J. Organomet. Chem. 1981, 217,
51−59.
(50) Brown, J. M.; Murrer, B. A. Tetrahedron Lett. 1980, 21, 581−
584.
(51) Wife, R. L.; van Oort, A. B.; van Doorn, J. A.; van Leeuwen, P.
W. N. M. Synthesis 1983, 1, 71−73.
(52) Armarego, W. L. F.; Chai, C. L. L. Purification of laboratory
chemicals; Elsevier: Amsterdam, The Netherlands, 2003.
(53) Schreurs, A. M. M.; Xian, X. Y.; Kroon-Batenburg, L. M. J. J.
Appl. Crystallogr. 2010, 43, 70−82.
(54) Duisenberg, A. J. M.; Kroon-Batenburg, L. M. J.; Schreurs, A. M.
M. J. Appl. Crystallogr. 2003, 36, 220−229.
(55) Sheldrick, G. M. SADABS: Area-Detector Absorption Correction;
Universitat Gottingen: Gottingen, Germany, 1999.
̈
̈
̈
(56) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(57) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.;
Spagna, R. J. Appl. Crystallogr. 1999, 32, 115−119.
(58) Spek, A. L. Acta Crystallogr., Sect. D 2009, 65, 148−155.
L
dx.doi.org/10.1021/ic400973t | Inorg. Chem. XXXX, XXX, XXX−XXX