New nickel-containing homogeneous hydrogenation catalysts structures of [Ni(o-MeO-dpppol)Cl2] and [Ni(dcpe)Cl2]

Article abstract of DOI:10.1016/S1381-1169(03)00216-4

The didentate phosphane ligands 1,3-bis(di(ortho-methoxyphenyl)phosphanyl)-2-propanol (o-MeO-dpppol) and 1,2bis(dicyclohexylphosphanyl)ethane (dcpe) have been used in a study towards catalytic hydrogenation using homogeneous nickel catalysts. The nickel halide complexes [Ni(o-MeO-dpppol)X₂] and [Ni(dcpe)X₂] (X = Cl, Br or I) have been synthesized and were characterized using electronic absorption and IR spectroscopy. The structures of the complexes [Ni(o-MeO-dpppol)Cl₂] and [Ni(dcpe)Cl₂] have been determined by X-ray crystallography. The nickel ions in these complexes are in (distorted) square-planar geometries with NiP₂Cl₂ chromophores. The synthesized halide complexes and in situ mixtures of nickel acetate and the ligands were tested on catalytic activity in homogeneous hydrogenation. The ligand dcpe yields very active catalysts; turnover numbers up to 3000 in 1 h have been obtained. Generally, it appeared that the catalytic activities are higher in methanol than in methanol/dichloromethane mixtures. It was found that the catalytic activities observed for complexes containing the ligand o-MeO-dpppol are less reproducible than those of catalysts with the ligand dcpe. This lower reproducibility is probably related to the fact that the former ligand is more readily oxidized.

Full text of DOI:10.1016/S1381-1169(03)00216-4

Journal of Molecular Catalysis A: Chemical 202 (2003) 97–106
New nickel-containing homogeneous hydrogenation catalysts
Structures of [Ni(o-MeO-dpppol)Cl2] and [Ni(dcpe)Cl2]
Ingrid M. Anguloa, Elisabeth Bouwmana,∗, Remy van Gorkuma,
Sandra M. Loka, Martin Lutzb, Anthony L. Spekb
Gorlaeus Laboratories, Leiden Institute of Chemistry, Leiden University, P.O. Box 9502, 2300 RA Leiden, The Netherlands
a
b
Department of Crystal and Structural Chemistry, Bijvoet Center for Biomolecular Research, Utrecht University,
Padualaan 8, 3584 CH Utrecht, The Netherlands
Received 5 February 2003; received in revised form 18 March 2003; accepted 28 March 2003
Abstract
The didentate phosphane ligands 1,3-bis(di(ortho-methoxyphenyl)phosphanyl)-2-propanol (o-MeO-dpppol) and 1,2-
bis(dicyclohexylphosphanyl)ethane (dcpe) have been used in a study towards catalytic hydrogenation using homogeneous
nickel catalysts. The nickel halide complexes [Ni(o-MeO-dpppol)X2] and [Ni(dcpe)X2] (X = Cl, Br or I) have been
synthesized and were characterized using electronic absorption and IR spectroscopy. The structures of the complexes
[Ni(o-MeO-dpppol)Cl2] and [Ni(dcpe)Cl2] have been determined by X-ray crystallography. The nickel ions in these com-
plexes are in (distorted) square–planar geometries with NiP2Cl2 chromophores. The synthesized halide complexes and in situ
mixtures of nickel acetate and the ligands were tested on catalytic activity in homogeneous hydrogenation. The ligand dcpe
yields very active catalysts; turnover numbers up to 3000 in 1 h have been obtained. Generally, it appeared that the catalytic
activities are higher in methanol than in methanol/dichloromethane mixtures. It was found that the catalytic activities observed
for complexes containing the ligand o-MeO-dpppol are less reproducible than those of catalysts with the ligand dcpe. This
lower reproducibility is probably related to the fact that the former ligand is more readily oxidized.
© 2003 Elsevier Science B.V. All rights reserved.
Keywords: Homogeneous catalysis; Hydrogenation; Nickel complexes; Phosphane ligands; Crystal structures
1. Introduction
We have reported that selected nickel(II) diphos-
phane complexes, containing acetate anions, are able
to hydrogenate homogeneously 1-octene to n-octane
Abbreviations: o-MeO-dppe, 1,2-bis(di(ortho-methoxyphenyl)-
[1]. It appeared that in methanol the catalytic hydro-
genation activity of the nickel(II) catalysts contain-
ing the didentate phosphane ligand 1,2-bis(dicyclo-
hexylphosphanyl)ethane (dcpe) or 1,3-bis(di(ortho-
methoxyphenyl)phosphanyl)propane (o-MeO-dppp)
are higher than that of a catalyst containing the
ligand 1,2-bis(di(ortho-methoxyphenyl)phosphanyl)-
phosphanyl)ethane; o-MeO-dppp, 1,3-bis(di(ortho-methoxyphe-
nyl)phosphanyl)propane; o-EtO-dppe, 1,2-bis(di(ortho-ethoxyphe-
nyl)phosphanyl)ethane; o-MeO-dpppol, 1,3-bis(di(ortho-metho-
xyphenyl)phosphanyl)-2-propanol; dcpe, 1,2-bis(dicyclohexylphos-
phanyl)ethane
∗
Corresponding author. Tel.: +31-71-5274550;
fax: +31-71-5274451.
E-mail address: bouwman@chem.leidenuniv.nl (E. Bouwman).
1381-1169/03/$ – see front matter © 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S1381-1169(03)00216-4

98
I.M. Angulo et al. / Journal of Molecular Catalysis A: Chemical 202 (2003) 97–106
Scheme 1.
ethane (o-MeO-dppe) (see Scheme 1 for ligand
structures).
hexyl groups in dcpe prevent the ligand redistribution
reaction.
In an attempt to explain the differences in cat-
alytic activity a systematic study was undertaken to
investigate the nature of species in solution for both
o-methoxy ligands o-MeO-dppe [2] and o-MeO-dppp
[3]. It was found that complexes of the ligand
o-MeO-dppe are involved in a ligand-redistribution
equilibrium, in contrast to complexes of the lig-
and o-MeO-dppp. In the NMR study of the nickel
complexes with the didentate phosphane ligand
o-MeO-dppe, the complex [Ni(o-MeO-dppe)2]2+
was observed in varying amounts, depending on the
solvents and anions used [2]. The influence of this
ligand redistribution on the catalytic hydrogenation
of 1-octene in the presence of nickel(II) complexes
containing the ligands o-MeO-dppe and o-MeO-dppp
have been investigated [4]. It appeared that catalytic
activity was highest in a dichloromethane/methanol
mixture.
The presence of o-methoxy groups on the phenyl
rings creates so-called hemi-labile ligands [5]. The
enhanced enantioselectivity of hydrogenation cat-
alysts containing such methoxy groups has been
attributed to the possible metal–oxygen interactions
[6].
The catalysts containing the ligand o-MeO-dppe
intrinsically form the most active catalysts. How-
ever, due to the ligand redistribution equilibrium,
which does not play a role in catalysts of the ligand
o-MeO-dppp, the latter reach higher turnover num-
bers [4]. To study the effect of the backbone of the
ligand on the catalytic hydrogenation, the new ligand
1,3-bis(di(ortho-methoxyphenyl)phosphanyl)-2-prop-
anol (o-MeO-dpppol) has been synthesized (see
Scheme 1). The reported high catalytic activity of
catalysts containing the ligand dcpe has been further
investigated. It is envisioned that the bulky cyclo-
2. Experimental
2.1. General
Melting points were determined on a Büchi appa-
ratus and are uncorrected. Elemental analyses were
performed on a Perkin-Elmer 2400 Series II ana-
lyzer. Mass spectra were obtained with a Finnigan
MAT TSQ-700 mass spectrometer equipped with a
custom-made electrospray interface (ESI). Spectra
were collected by constant infusion of the analyte dis-
solved in methanol/water with 1% acetic acid. Infrared
spectra were recorded with a Perkin-Elmer Paragon
1000 IR spectrophotometer equipped with a Golden
Gate ATR device, using the reflectance technique
(4000–300 cm−1). Solid-state electronic absorption
spectra were obtained on a Perkin-Elmer Lambda 900
spectrophotometer using the diffuse reflectance tech-
1
nique, with MgO as a reference. The H NMR and
31P {1H} NMR spectra were recorded on a Bruker
WM-300 spectrometer at 300.13 and 121.50 MHz,
respectively. The 1H and 31P chemical shifts are
recorded in δ units relative to TMS and phosphoric
acid, respectively. GC analyses were performed on a
Chrompack 437A chromatograph fitted with a flame
ionization detector and a 50 m cp-sil 5cb column.
All chemicals were reagent grade and were used
as received or synthesized as described below. Re-
actions were carried out under an atmosphere of
dry argon, unless stated otherwise, by using stan-
dard Schlenk techniques. The diphosphane lig-
and dcpe was obtained from Strem Chemicals.
Tris(o-methoxyphenyl)phosphane was synthesized
according to literature procedures [7].

I.M. Angulo et al. / Journal of Molecular Catalysis A: Chemical 202 (2003) 97–106
99
2.2. Synthesis of 1,3-bis(di-ortho-
methoxyphenylphosphanyl)-2-propanol
(o-MeO-dpppol)
all the NiBr2 was dissolved. o-MeO-dpppol (0.16 g,
0.3 mmol) was dissolved in 3 ml chloroform. When
both solutions were mixed a purple powder precipi-
tated. The solution was filtered and the solid was dried
in air. Yield: 0.18 g of a purple powder (80% based on
ligand).
This synthesis was performed according to the
procedure earlier described for the synthesis of
1,3-bis(di(o-methoxyphenyl)phosphanyl)propane [8].
Starting from 30 mmol (10.5 g) P(o-MeO-Ph)3 and
15 mmol (1.39 g) epichlorohydrin, 4.2 g (51% yield)
of a white powder was obtained. mp 73–78 ◦C. Anal.
Calcd. for C31H34P2O5 (548.6): C, 67.88; H, 6.25.
Found: C, 66.67; H, 6.11. 1H NMR (CDCl3): 2.5 (m,
4H, PCH2), 2.93 (s, 1H, OH), 3.67 (s, 6 H, –OCH3),
3.71 (s, 6 H, –OCH3), 3.95 (m, 1H, CHOH), 6.88
(m, 8H, aromatic protons), 7.14 (m, 4H, aromatic
protons), 7.28 (t, 4H, aromatic protons). 31P {1H}
NMR (CDCl3): −43.3 ppm (s). IR (cm−1): 2930 (w),
1584 (m), 1572 (m), 1456 (s), 1430 (s), 1271 (m),
1237 (vs), 1178 (m), 1162 (m), 1131 (m), 1067 (w),
1042 (w), 1020 (vs), 938 (w), 792 (m), 751 (vs), 574
(w), 508 (m), 458 (m), 422 (w). MS (ESI) m/z 549
[M + H].
Anal. Calcd. for C31H34Br2NiO5P2 0.5 CHCl3
(Fw = 826.8): C, 45.76; H, 4.21. Found: C, 45.96;
H, 4.41. IR (cm−1): 1585 (m), 1570 (m), 1475 (m),
1428 (s), 1278 (m), 1248 (s), 1120 (w), 1070 (m),
1018 (s), 1007 (s), 909 (m), 820 (m), 797 (s), 760
(vs), 570 (m), 506 (m), 482 (s), 461 (m), 418 (m),
328 (m). Diffuse reflectance electronic absorption of
the solid: νmax 28,100 and 19,800 cm−1
.
2.3.3. [Ni(o-MeO-dpppol)I2] (3)
Ni(OAc)2·4H2O (0.12 g, 0.5 mmol) was dissolved
in 10 ml ethanol and o-MeO-dpppol (0.28 g, 0.5 mmol)
was dissolved in 5 ml chloroform. Both solutions were
mixed and after addition of 0.125 ml of HI (8 M in
water, 1 mmol) a purple solid precipitated. The solu-
tion was filtered and the solid was dried in air. Yield:
0.15 g of a dark purple solid (35%).
2.3. Synthesis of the complexes
Anal. Calcd. for C31H34I2NiO5P2 (Fw = 861.07):
C, 43.24; H, 3.98. Found: C, 43.03; H, 3.89. IR
(cm−1): 1584 (m), 1570 (m), 1474 (m), 1424 (s), 1276
(m), 1246 (s), 1117 (w), 1068 (m), 1004 (s, br), 908
(m), 818 (m), 797 (s), 756 (vs), 569 (m), 503 (m), 482
(s), 456 (s), 406 (m). Diffuse reflectance electronic
2.3.1. [Ni(o-MeO-dpppol)Cl2] (1)
NiCl2·6H2O (0.18 g, 0.75 mmol) was dissolved
in 5 ml ethanol and o-MeO-dpppol (0.27 mg,
0.5 mmol) was dissolved in 5 ml chloroform. When
both solutions were mixed, a red solid precipi-
tated. The solution was filtered and the solid was
dried in air. Yield: 0.26 g of a red-purple powder
(75% based on ligand). Crystals of 1 suitable for
single-crystal X-ray diffraction were obtained after re-
crystallization from an iso-propanol/dichloromethane
mixture.
absorption of the solid: νmax 27,900 and 18,300 cm−1
.
2.3.4. [Ni(dcpe)Cl2] (4)
2.3.4.1. Method A. Solid Ni(OAc)2·4H2O (0.25 g,
1 mmol) was dried in vacuo to remove most of the
water. The color changed from bright green to light
green. After dissolution in 30 ml methanol, the lig-
and dcpe (0.42 g, 1 mmol) was added. Then, when
a clear orange solution was formed 0.17 ml of con-
centrated hydrogen chloride (12 M in water, 2 mmol)
was added and immediately an orange solid pre-
cipitated. The solid was collected by filtration and
dried in air. Yield: 0.43 g (78%) of a yellow–orange
powder.
Anal. Calcd. for C31H34Cl2NiO5P2 0.35 CHCl3
(Fw = 719.9): C, 52.30; H, 4.81. Found: C, 52.15;
H, 4.82. IR (cm−1): 1585 (w), 1570 (m), 1475 (m),
1428 (s), 1278 (m), 1248 (s), 1120 (m), 1070 (m),
1019 (s), 1009 (s), 909 (m), 821 (m), 797 (s), 760
(vs), 571 (m), 506 (m), 482 (s), 462 (s), 328 (m).
Diffuse reflectance electronic absorption of the solid:
νmax 27,600 and 20,100 cm−1
.
2.3.4.2. Method B. Solid NiCl2·6 H2O (0.36 g,
1.5 mmol) was dried in vacuo. After dissolution in
30 ml ethanol, the ligand dcpe (0.42 g, 1 mmol) was
2.3.2. [Ni(o-MeO-dpppol)Br2] (2)
NiBr2 (0.16 g, 0.75 mmol) was dissolved in 10 ml
ethanol and 1 ml water. The solution was heated until

100
I.M. Angulo et al. / Journal of Molecular Catalysis A: Chemical 202 (2003) 97–106
added. Immediately a dark orange solution was ob-
tained and an orange solid precipitated within a few
minutes. The solid was collected by filtration and
dried in air. The dry powder is stable in air. Yield:
0.39 g (70%) of a yellow–orange powder. Recrys-
tallization from a toluene/dichloromethane mixture
yielded crystals suitable for X-ray diffraction. This
crystallization was performed in an argon atmosphere.
Anal. Calcd. for C26H48Cl2NiP2 (Fw = 552.23): C,
56.55; H, 8.76. Found: C, 56.19; H, 8.69. IR (cm−1):
2929 (vs), 2850 (vs), 1445 (s), 1407 (w), 1270 (m),
1006 (m), 889 (m), 852 (m), 793 (m), 746 (s), 675
(m), 536 (s), 340 (m), 313 (s). Diffuse reflectance elec-
tronic absorption of the solid: νmax 32,400, 21,900 and
646 (s), 536 (s), 462 (s), 398 (m), 358 (m). Diffuse
reflectance electronic absorption of the solid: νmax
27,200, 19,600 cm−1
.
2.4. X-ray crystal structure determinations of
[Ni(o-MeO-dpppol)Cl2] (1) and [Ni(dcpe)Cl2] (4)
X-ray intensities were measured at 150 K on a No-
nius KappaCCD (1) or an Enraf-Nonius CAD4T (4)
diffractometer, both equipped with a rotating anode
(Mo K␣, λ = 0.71073 Å). The structures were solved
with Patterson methods (DIRDIF-97) [9] and refined
with the program SHELXL-97 [10] against F2 of
all reflections up to a resolution of (sin θ/λ)max
=
16,400 (shoulder) cm−1
.
0.59 Å−1. Non-hydrogen atoms were refined freely
with anisotropic displacement parameters, hydrogen
atoms were refined as rigid groups. The drawings,
structure calculations, and checking for higher sym-
metry were performed with the program PLATON
[11]. Further experimental details are given in Table 1.
The hydrogen- and hydroxyl-substituents of carbon
C6 in 1 had to be refined with a disorder model. In
the crystal, the position of one substituent is partially
occupied with a hydrogen atom and partially with a
hydroxyl group. The ratio is 0.65:0.35. The position of
the second substituent logically has the reversed ratio.
2.3.5. [Ni(dcpe)Br2] (5)
Solid Ni(OAc)2·4 H2O (0.06 g, 0.25 mmol) was
dried in vacuo. After dissolution in 30 ml methanol,
the ligand dcpe (0.11 g, 0.25 mmol) was added. Then,
when a clear orange solution was formed 0.06 ml
of concentrated hydrogen bromide (9 M in water,
0.5 mmol) was added and immediately an orange
solid precipitated. The solid was collected by filtra-
tion and dried in air. The dry powder is stable in air.
Yield: 0.12 g (75%) of a red-orange powder. Anal.
Calcd. for C26H48Br2NiP2 (Fw = 641.13): C, 48.71;
H, 7.55. Found: C, 48.36; H, 7.55. IR (cm−1): 2916
(vs), 2843 (vs), 1446 (m), 1405 (s), 1264 (m), 1005
(m), 888 (m), 863 (m), 845 (s), 790 (s), 737 (m),
671 (s), 652 (s), 535 (s), 400 (m), 367 (w). Diffuse
reflectance electronic absorption of the solid: νmax
2.5. Hydrogenation experiments
Catalytic mixtures were prepared in two methods
in a Schlenk tube, either by formation of an in situ
mixture or by dissolving an isolated nickel halide
complex. In a typical in situ experiment, the nickel(II)
acetate was mixed with the ligand in 20 ml of degassed
solvent at room temperature until a clear solution was
obtained. Also in the case of the isolated nickel halide
complexes, 20 ml of degassed solvent was added to
the complex and the mixture was stirred at room tem-
perature to obtain a clear solution. Sometimes a clear
solution was obtained within 10 min, whereas in other
solvents overnight stirring and heating up to 40 ◦C
was needed. When a clear solution was obtained the
substrate, 1-octene, was added and the solution was
stirred for another 5 min. Hydrogenation reactions
were carried out in a 100 ml Parr stainless steel auto-
clave equipped with a glass inner beaker, and a motor
driven stirrer. The autoclave was evacuated and then
filled with argon three successive times, before the
28,200, 21,300 and 16,400 (shoulder) cm−1
.
2.3.6. [Ni(dcpe)I2] (6)
Solid Ni(OAc)2·4 H2O (0.06 g, 0.25 mmol) was
dried in vacuo. After dissolution in 30 ml methanol,
the ligand dcpe (0.11 g, 0.25 mmol) was added. Then,
when a clear orange solution was formed 0.06 ml
of concentrated hydrogen iodide (8 M in water,
0.5 mmol) was added and immediately an orange
solid precipitated. The solid was collected by filtra-
tion and dried in air. The dry powder is stable in air.
Yield: 0.11 g (62%) of a purple powder. Anal. Calcd.
for C26H48I2NiP2 (Fw = 735.13): C, 42.48; H, 6.58.
Found: C, 42.06; H, 6.53. IR (cm−1): 2916 (vs), 2845
(vs), 1436 (s), 1417 (m), 1266 (m), 1002 (s), 887
(m), 850 (s), 818 (m), 788 (m), 740 (m), 667 (vs),

I.M. Angulo et al. / Journal of Molecular Catalysis A: Chemical 202 (2003) 97–106
101
Table 1
Crystal data and structure refinement details for [Ni(o-MeO-dpppol)Cl2] (1) and [Ni(dcpe)Cl2] (4)
1
4
Formula
NiC31H34Cl2O5P2
678.13
Red
0.30 × 0.24 × 0.06
Orthorhombic
Pbca (no. 61)
15.2030(3)
17.1386(4)
22.8829(5)
90
NiC26H48Cl2P2·C7H8·CH2Cl2
729.25
Yellow
Formula weight
Crystal color
Crystal size (mm3)
System
Space group
a (Å)
0.50 × 0.30 × 0.10
Monoclinic
P21/c (no. 14)
12.1784(18)
14.9007(18)
20.0968(19)
93.347(10)
3640.7(8)
b (Å)
c (Å)
β (◦)
V (Å3)
5962.3(2)
8
1.511
Z
4
Dcalc (g/cm3)
1.330
0.94
µ(Mo K␣) (mm−1
)
0.98
Absolute correction
Transmission range
PLATON (ABST)
0.75–0.94
66297/5244
379/7
0.0353/0.0434
0.0860/0.0901
1.062
PLATON (DELABS)
0.52–0.85
13478/6418
370/0
0.0544/0.1154
0.0968/0.1150
0.965
−0.50/0.46
Reflectance measured/unique
Parameters/restraints
R1 (observed/all reflectance)a
wR2 (observed/all reflectance)b
GoF
Resolution density (e/Å3)
−0.40/0.53
ꢀ
ꢁ
ꢀ
a R1
b wR2
=
||Fo| − |Fc||/ |Fo|.
ꢂ1/2
ꢀ
ꢀ
=
w(Fo2 − Fc2)2/ w(Fo2)2
, w = 1/[σ2(Fo2) + (AP)2 + BP] where P = (Fo2 + 2Fc2)/3, A = 0.0582 (1) or 0.0457 (4),
and B = 2.2431 (1) or 1.2627 (4).
prepared solution was introduced via a syringe into
the autoclave. The autoclave was then purged with di-
hydrogen gas for several seconds, before the reaction
pressure of dihydrogen was applied. The temperature
of the autoclave was controlled by an electronic heat-
ing mantle. The autoclave was connected to a recorder
to register the pressure during the reaction. The re-
action was initiated by activation of the stirrer, and
simultaneous activation of the temperature-control
unit for reactions at elevated temperatures. After 1 h,
the autoclave was cooled in an ice-bath and reaction
samples were taken, these were stored at −20 ◦C.
In some cases biphasic mixtures were obtained after
storage at −20 ◦C, these mixtures were homoge-
nized by the addition of toluene. After dilution with
diethyl ether the samples were analyzed by gas chro-
matography. All reported values are the arithmetic
means of two or more experiments. Experiments are
considered reproducible when the deviation of the
obtained results are less than 10% of the arithmetic
mean.
In the hydrogenation experiments, the only
side-reaction observed is the isomerization of 1-octene
to internal alkenes (cis- and trans-2-octene, 3-octene
and 4-octene). The reported selectivity is given for
the hydrogenation to n-octane.
3. Results and discussion
3.1. Synthesis and characterization
The ligand o-MeO-dpppol was prepared using
the method based on the earlier reported reduc-
tive splitting of tris(o-methoxyphenyl)phosphane
(P(o-MeO-Ph)3) using sodium in ammonia [7]. In
contrast to the unsubstituted ligand o-MeO-dppp, the
ligand o-MeO-dpppol is soluble in methanol.
Synthesis of the complexes has been carried out
in inert atmosphere. Especially in solution phosphane
ligands in general, and the ligand dcpe in particular, are
sensitive to oxidation. The halide complexes appeared

102
I.M. Angulo et al. / Journal of Molecular Catalysis A: Chemical 202 (2003) 97–106
Table 2
to be stable in the solid state. Solutions of nickel ac-
etate and the ligand o-MeO-dpppol, however, are quite
sensitive to oxidation (see Section 3.4).
Selected interatomic distances (Å), angles (◦) and torsion angles
(◦) for 1 and 4
1
4
Within a series of the same ligand, the infrared spec-
tra of the nickel(II) complexes are very similar to each
other. Diffuse reflectance absorption spectra of the
halide compounds show, apart from a charge-transfer
transition at 30,000 cm−1, one strong band at around
20,000 cm−1, just as for the [Ni(dppp)X2] [12],
[Ni(o-MeO-dppp)X2] [3], and [Ni(o-MeO-dppe)X2]
[2] complexes. The lower-energy band is ascribed to
the A1 → B2 transition for a square–planar geom-
etry. Both for the dcpe and for the o-MeO-dpppol
complexes, this band shifts to higher energy in the
order I < Br < Cl, which is consistent with the es-
tablished spectrochemical series [13]. Based on the
similarities in the IR spectra and the X-ray structures
described below, the nickel ion in all complexes has
been assigned a square–planar geometry.
Interatomic distances (Å)
Ni1–Cl1
2.2120(8) Ni1–Cl2
2.2261(8) Ni1–Cl3
2.1733(8) Ni1–P4
2.1678(8) Ni1–P7
3.440(2) Ni–H16A
3.484(2) Ni–H22B
2.2080(14)
2.2050(13)
2.1548(14)
2.1643(14)
3.13
3.18
3.04
3.26
Ni1–Cl2
Ni1–P4
Ni1–P8
Ni1–O17
Ni1–O47
Ni1–H26
Ni1–H36
2.91
3.00
Ni–H32B
Ni–H46A
1
1
Angles (◦)
Cl1–Ni1–Cl2
Cl1–Ni1–P4
Cl1–Ni1–P8
Cl2–Ni1–P4
Cl2–Ni1–P8
P4–Ni1–P8
94.71(3) Cl2–Ni1–Cl3
88.12(3) Cl2–Ni1–P4
163.94(3) Cl2–Ni1–P7
166.01(3) Cl3–Ni1–P4
89.87(3) Cl3–Ni1–P7
91.15(3) P4–Ni1–P7
92.81(5)
88.89(5)
176.50(6)
176.53(6)
89.61(5)
88.82(5)
Torsion angles (◦)
Ni1–P4–C11–C12 −62.7(3)
Ni1–P4–C21–C22 173.5(2)
Ni1–P8–C31–C32 169.7(2)
Ni1–P8–C41–C42 −66.5(3)
Ni1–P4–C11–C16
44.6(4)
Ni1–P4–C21–C22 −46.6(4)
3.2. Crystal structure of [Ni(o-MeO-dpppol)Cl2]
Ni1–P7–C31–C32 −44.7(4)
Ni1–P7–C41–C46
P4–C5–C6–P7
54.9(4)
33.9(4)
A molecular plot of the structure of [Ni(o-MeO-
dpppol)Cl2] is given in Fig. 1. Selected interatomic
distances and angles are given in Table 2.
P4–C5–C6–C7
C5–C6–C7–P8
33.9(3)
36.8(3)
Fig. 1. Platon projection of [Ni(o-MeO-dpppol)Cl2] with atom-labelling scheme. Hydrogen atoms are omitted for clarity. The hydroxyl
substituent at C6 is disordered (see Section 2).

I.M. Angulo et al. / Journal of Molecular Catalysis A: Chemical 202 (2003) 97–106
103
The compound crystallizes in the orthorhombic
space group Pbca, with one molecule in the asym-
metric unit. The structure of [Ni(o-MeO-dpppol)Cl2]
resembles that of the related structure of [Ni(o-MeO-
dppp)Cl2] [3]. The latter complex, however, pos-
sesses an exact, crystallographic two-fold symmetry.
The coordination geometry around the nickel(II) cen-
ter is essentially square planar, but with a severe
tetrahedral distortion, as follows from the dihedral
angle between the Cl–Ni–Cl and P–Ni–P planes
(20.65(4)◦) and the rather small trans coordination
angles (163.94(3) and 166.01(3)◦). The bond dis-
tances and bond angles in 1 are quite similar to those
in [Ni(o-MeO-dppp)Cl2]. The bite angle P4–Ni1–P8
is not influenced by the presence of the hydroxyl
group on the central carbon atom of the propy-
lene bridge. It is possible to distinguish the aryl
rings as orientated either axially (atoms C21–C26
and C31–C36) or equatorially (atoms C11–C16 and
C41–C46) with respect to the plane of coordina-
tion. This orientation of the aryl rings is essentially
the same as in [Ni(o-MeO-dppp)Cl2]. The Ni · · · O
distances between the nickel ion and the methoxy
groups of the equatorial aryl rings (3.440(2) Å for
O17 and 3.484(2) Å for O47) are too large to consider
an electrostatic interaction. The o-hydrogen atoms of
the two axial aryl groups are pointing towards the
nickel atom. The Ni · · · H distances of 2.91 Å (H26)
and 3.00 Å (H36) indicate that some weak Ni · · · H–C
contacts are present, however, these interactions are
weaker than those observed for complexes contain-
ing the ligands o-MeO-dppe [2] and o-MeO-dppp
[3].
The major disorder component O3A forms an inter-
molecular hydrogen bond with Cl2 of a neighboring
molecule (O · · · Cl distance of 3.22 Å), thereby link-
ing the molecules in an infinite one-dimensional
chain.
3.3. Crystal structure of [Ni(dcpe)Cl2]
Molecular plots of the structure of [Ni(dcpe)Cl2]
are given in Fig. 2. Relevant bond lengths and angles
are collected in Table 2.
The complex crystallizes in the monoclinic space
group P21/c with one molecule in the asymmetric unit.
Apart from the nickel complex, the crystal contains
co-crystallized toluene and dichloromethane. The
nickel ion is coordinated by the two phosphorus donors
and two chloride ions in a rather exact square–planar
geometry (dihedral angle of 3.99(7)◦). The bite angle
within the didentate phosphane ligand is 88.82(5)◦,
which is as expected. Similar structures have been
reported for [Pd(dcpe)Cl2] [14] and [Pt(dcpe)Cl2]
[15].
The cyclohexyl rings all are in chair conformations.
Closest contacts of o-hydrogen atoms with the nickel
ion are at 3.04–3.26 Å, which are slightly longer than
those found in 1.
3.4. Catalytic hydrogenation
Both the preformed halide complexes and in situ
mixtures of nickel acetate with the ligands have been
tested on their catalytic activity in the hydrogenation
of 1-octene. The results of the catalytic experiments
are collected in Table 3.
The experiments have been performed in pure
methanol and in methanol/dichloromethane mix-
tures. For comparison, the results of the hydro-
genation experiments in pure methanol with the
halide complexes [Ni(o-MeO-dppp)X2] have been
included.
Nickel acetate in combination with the ligand
dcpe renders very active hydrogenation catalysts in
methanol. The activity at room temperature (entries
1 and 2) is significant and at a higher temperature of
50 ◦C a turnover number of 3000 can be reached within
1 h. In a binary solvent mixture of lower polarity, the
in situ formed catalyst is considerably less active. The
activity of the halide complexes of dcpe increases in
The presence of the hydroxyl group does not
have a large effect on the geometry of the com-
plex. A possible coordination of the hydroxyl group
with the nickel ion is not present in the solid state
(shortest intramolecular distance is 4.855(5) Å) and
there is also no effect on the bite angle of the lig-
and. In solution, however, the hydroxyl group may
coordinate to the nickel ion, thereby forcing the dis-
sociation of one of the phosphane donors that is
subsequently oxidized, which may cause the poor
reproducibility of the catalytic experiments (see
below).
A weak intramolecular hydrogen bond interaction
is present between O3B (minor component) and O27
of a methoxy group (O · · · O distance of 3.19 Å).

104
I.M. Angulo et al. / Journal of Molecular Catalysis A: Chemical 202 (2003) 97–106
Fig. 2. PLATON projections of [Ni(dcpe)Cl2] with atom-labelling scheme (a) front view and (b) top view. Hydrogen atoms are omitted
for clarity.
Table 3
The activity of nickel(II) phosphane complexes in the hydrogenation of 1-octenea
c
d
Entry
Catalyst
T (◦C)
Solvent
[Ni] (mM)b
TONoverall
Sn-octane
e
1
2
3
4
dcpe/Ni(OAc)2
25
25
25
50
MeOH
MeOH
CH2Cl2/MeOH
MeOH
2.5f
0.625
2.5f
1740
1820
200
80
90
90
88
0.625
3000
5
6
[Ni(dcpe)Cl2]
[Ni(dcpe)Br2]
[Ni(dcpe)I2]
50
50
MeOH
CH2Cl2/MeOH
1.25
0.625
130
100
100
100
7
8
50
50
MeOH
CH2Cl2/MeOH
0.625
0.625
940
160
72
70
g
9
10
50
50
MeOH
CH2Cl2/MeOH
0.625
0.625
–
290
<50
45
11
12
13
[Ni(o-MeO-dpppol)Cl2]
[Ni(o-MeO-dpppol)Br2]
[Ni(o-MeO-dpppol)I2]
50
50
50
MeOH
MeOH
MeOH
1.25
1.25
1.25
340
960
1800
100
99
97
14
15
16
[Ni(o-MeO-dppp)Cl2]
[Ni(o-MeO-dppp)Br2]
[Ni(o-MeO-dppp)I2]
50
50
50
MeOH
MeOH
MeOH
1.25
1.25
1.25
730
960
1155
99
99
98
a Reaction conditions: 20 ml solvent, 50 mmol 1-octene, t = 1 h, p(H2) = 50 bar.
b Concentration in 20 ml solvent, calculated before addition of substrate.
c Turnover number (mol converted 1-octene/mol Ni after 1 h).
d Selectivity towards n-octane (mol n-octane/mol converted 1-octene).
e In situ mixtures; dcpe/Ni = 1.1.
f 100 mmol 1-octene.
g Not reproducible.

I.M. Angulo et al. / Journal of Molecular Catalysis A: Chemical 202 (2003) 97–106
105
the order Cl < Br < I in a methanol/dichloromethane
4. Concluding remarks
mixture (entries 6, 8, and 10). In methanol a similar
trend in activity is expected. Indeed, the bromide com-
plex (entry 7) is more active than the chloride complex
(entry 5), and for the iodide complex an activity com-
parable to that of the in situ nickel acetate catalyst was
foreseen. The complex [Ni(dcpe)I2], however, is only
poorly soluble in methanol, and unfortunately, the
results of the experiments with this catalyst precursor
therefore were not reproducible. All dcpe-based cat-
alysts show only moderate selectivities to n-octane,
comparable to those of the o-MeO-dppe-based
catalysts.
The activities as well as the selectivities of the
halide complexes of o-MeO-dpppol (entries 11–13) in
methanol are comparable to those of the [Ni(o-MeO-
dppp)X2] complexes (entries 14–16). Similar trends
are observed going from the more strongly coordi-
nating chloride to the more weakly bound iodide,
but the influence of the anion is more pronounced
in the case of o-MeO-dpppol. In contrast, whereas
the activity of the [Ni(o-MeO-dppp)X2] complexes
is slightly enhanced when dichloromethane is added
[4], the measurements for the [Ni(o-MeO-dpppol)X2]
complexes in a mixture of dichloromethane and
methanol were not reproducible. Also, measure-
ments with catalysts formed from in situ mix-
tures of nickel acetate and o-MeO-dpppol were not
reproducible.
Remarkable is the lower activity of all cata-
lysts in dichloromethane/methanol mixtures. In the
case of o-MeO-dppp and o-MeO-dppe, the use of
dichloromethane was needed for the preparation of in
situ catalysts because of the insolubility of these lig-
ands in methanol. Probably because of the presence
of the hydroxy group in o-MeO-dpppol, the ligand is
soluble in methanol and addition of dichloromethane
is not needed. What causes the different trend in sol-
vent dependence for o-MeO-dpppol and dcpe in com-
parison with o-MeO-dppp and o-MeO-dppe, however,
is not yet understood.
For the ligands o-MeO-dppp and o-MeO-dpppol,
both containing a C3 bridge, the selectivity of the
catalysts is hardly influenced by the anion. In con-
trast, for the ligands dcpe and o-MeO-dppe, both
containing a C2 spacer, the selectivity is strongly
affected by the coordinating properties of the
anion.
In a kinetic study for catalysts based on o-MeO-
dppp, it has become clear that the activation of
dihydrogen is the rate-determining step in the hy-
drogenation reaction [16]. The present study seems
to confirm that for catalytic hydrogenation activ-
ity of nickel catalysts a protic solvent is required.
From the suggested mechanism [4,16], it is clear that
solvation of the anion is important during catalytic
hydrogenation.
The selectivity of a catalyst containing a ligand
with a C2 spacer seems always lower than that of a
catalyst containing a ligand with a C3 spacer. Iso-
merization can only occur in case of 2,1-insertion
of the olefin in the nickel-hydride bond. The bulk-
iness of a C3-bridged ligand can either hinder the
formation of the secondary alkyl species or the sub-
sequent ␤-H elimination. Both reactions require more
space at the equatorial positions of the nickel com-
plex than the hydrogenolysis of the primary alkyl
species. The larger bite angle of the ligand with a
C3 spacer as compared to those with a C2 bridge
implies more steric crowding in the plane of coor-
dination giving a likely explanation for the higher
selectivities that are obtained with o-MeO-dppp and
o-MeO-dpppol.
The C2-bridged ligands seem to form the most
active hydrogenation catalysts. For the ligand
o-MeO-dppe it was found that only a small part of the
nickel ions is available for catalysis, and thus the cat-
alyst formed from this ligand is intrinsically higher in
activity than the one formed with o-MeO-dppp. The
catalysts formed with dcpe are very active even at
room temperature. This seems to imply that for the
nickel dcpe complexes ligand redistribution does not
occur and thus inactive [Ni(dcpe)2]2+ complexes are
not formed.
The catalytic activity of the o-MeO-dpppol-based
catalysts is poorly reproducible, especially in the
presence of acetate anions. In an earlier study it
was observed that the positive influence of the
methoxy groups in o-MeO-dppp and o-MeO-dppe
compared with the unsubstituted dppp and dppp is
caused by the prevention of oxidation of the phos-
phorus donor to phosphane oxide [2]. The pres-
ence of the hydroxyl group in o-MeO-dpppol seems
to counterbalance this positive effect. Possibly, in

106
I.M. Angulo et al. / Journal of Molecular Catalysis A: Chemical 202 (2003) 97–106
solution the hydroxyl group may coordinate to the
nickel ion, thereby forcing the dissociation of one
of the phosphane donors, which is subsequently
oxidized. NMR spectra taken of a spent catalyst
have confirmed oxidation of the o-MeO-dpppol
ligand.
References
[1] I.M. Angulo, A.M. Kluwer, E. Bouwman, Chem. Commun.
(1998) 2689.
[2] I.M. Angulo, E. Bouwman, M. Lutz, W.P. Mul, A.L. Spek,
Inorg. Chem. 40 (2001) 2073.
[3] I.M. Angulo, E. Bouwman, S.M. Lok, M. Lutz, W.P. Mul,
A.L. Spek, Eur. J. Inorg. Chem. (2001) 1465.
[4] I.M. Angulo, S.M. Lok, V.F. Quiroga Norambuena, M. Lutz,
A.L. Spek, E. Bouwman, J. Mol. Catal. A 187 (2002) 55.
[5] A. Bader, E. Lindner, Coord. Chem. Rev. 108 (1991) 27.
[6] W.S. Knowles, Acc. Chem. Res. 16 (1983) 106.
[7] L. Brandsma, H.D. Verkruijsse, Synth. Commun. 20 (1990)
2273.
5. Supplementary material
Crystallographic data (excluding structure fac-
tors) for the structures reported in this paper have
been deposited with the Cambridge Crystallo-
graphic Data Center as supplementary publication
no. CCDC-168811 for [Ni(o-MeO-dpppol)Cl2] and
CCDC-168812 for [Ni(dcpe)Cl2]. Copies of the data
can be obtained free of charge on application to
The Director, CCDC, 12 Union Road, Cambridge
CB2 1EZ, UK (fax: +44-1223-336033; e-mail:
deposit@chemcrys.cam.ac.uk).
[8] P.H.M. Budzelaar, J.A. van Doorn, N. Meijboom, Recl. Trav.
Chim. Pays-Bas 110 (1991) 420.
[9] P.T. Beurskens, G. Admiraal, G. Beurskens, W.P.
Bosman, S. Garc´ıa-Granda, R.O. Gould, J.M.M. Smits, C.
Smykalla, DIRDIF-97 Program System, Technical Report.
Crystallography Laboratory, University of Nijmegen, The
Netherlands, 1999.
[10] G.M. Sheldrick, SHELXL-97. Program for Crystal Structure
Refinement. University of Göttingen, Germany, 1997.
[11] A.L. Spek, PLATON. A Multipurpose Crystallographic Tool.
Utrecht University, The Netherlands, 2000.
[12] G.R. van Hecke, W. De Horrocks Jr., Inorg. Chem. 5 (1966)
1968.
Acknowledgements
[13] A.B.P. Lever, Inorganic Electronic Spectroscopy, Elsevier,
Amsterdam, 1983, Section 6.2.
[14] S. Ganguly, J.T. Mague, D.M. Roundhill, Acta Crystallogr.,
Sect. C 50 (1994) 217.
[15] J.T. Mague, M.J. Fink, C.A. Recatto, Acta Crystallogr., Sect.
C 49 (1993) 1176.
[16] I.M. Angulo, E. Bouwman, J. Mol. Catal. A. 175 (2001) 65.
This research has been financially supported by the
Council for Chemical Sciences of the Netherlands
Organization for Scientific Research (CW-NWO).
We thank Dr. J. Reedijk and Dr. E. Drent for fruitful
discussions.