Nickel and iron complexes with N,P,N-type ligands: Synthesis, structure and catalytic oligomerization of ethylene

Article abstract of DOI:10.1039/b802009d

The N,P,N-type ligands bis(2-picolyl)phenylphosphine (1), bis(4,5-dihydro-2-oxazolylmethyl)phenylphosphine (2), bis(4,4-dimethyl-2- oxazolylmethyl)phenylphosphine (3) and bis(2-picolyloxy)phenylphosphine (4) were used to synthesize the corresponding pentacoordinated Ni(ii) complexes [Ni{bis(2-picolyl)phenylphosphine}Cl₂] (6), [Ni{bis(4,5-dihydro-2- oxazolylmethyl)phenylphosphine}Cl₂] (7), [Ni{bis(4,4-dimethyl-2- oxazolylmethyl)phenylphosphine}Cl₂] (8) and [Ni{bis(2-picolyloxy) phenylphosphine}Cl₂] (9), respectively. The hexacoordinated iron complexes [Fe{bis(2-picolyl)phenylphosphine}₂][Cl ₃FeOFeCl₃] (10), [Fe{bis(4,5-dihydro-2-oxazolylmethyl) phenylphosphine}₂][Cl₃FeOFeCl₃] (11) and the tetracoordinated complex [Fe{bis(4,4-dimethyl-2-oxazolylmethyl)phenylphosphine} Cl₂] (abbreviated [FeCl₂(NPN^Me2-N,N)]) were prepared by reaction of FeCl₂?4H₂O with ligands 1-3, respectively. The crystal structures of the octahedral complexes 10 and 11, determined by X-ray diffraction, showed that two tridentate ligands are facially coordinated to the metal centre with a cis-arrangement of the P atoms and the dianion (μ-oxo)bis[trichloroferrate(iii)] compensates the doubly positive charge of the complex. The cyclic voltammograms of 10 and 11 showed two reversible redox couples attributed to the reduction of the dianion (Fe ₂OCl₆)^2- (-0.24 V for 10 and -0.20 V for 11vs. SCE) and to the oxidation of the Fe(ii) ion of the complex (0.67 V for 10 and 0.52 V for 11vs. SCE). The cyclic voltammogram of [FeCl₂(NPN ^Me2-N,N)] showed a reversible redox couple at -0.17 V vs. SCE assigned to the oxidation of the Fe(ii) atom and an irreversible process at 0.65 V. The complexes 6, 8-11 and [FeCl₂(NPN^Me2-N,N)] have been evaluated in the catalytic oligomerization of ethylene with AlEtCl ₂ or MAO as cocatalyst. The nickel complex 6 proved to be the most active precatalyst in the series, with a turnover frequency (TOF) of 61 800 mol_C2H₄ mol_Ni^-1 h ^-1 with 10 equiv. of AlEtCl₂ and 12 200 mol _C2H₄ mol_Ni^-1 h^-1 with 200 equiv. of MAO. Precatalysts 8 and 9 were the most selective in butenes, up to 90% with 6 equiv. of AlEtCl₂ and 89% with 2 equiv. of AlEtCl ₂, respectively, and up to 92% butenes with 400 equiv. of MAO and 91% butenes with 200 equiv. MAO, respectively. The best selectivities for 1-butene were provided by 8 and AlEtCl₂ (up to 31% with 6 equiv.) and 9 with MAO (up to 72% with 200 equiv.). The iron complexes were not significantly active with AlEtCl₂ or MAO as cocatalyst. The Royal Society of Chemistry 2008.

Full text of DOI:10.1039/b802009d

View Article Online / Journal Homepage / Table of Contents for this issue
An international journal of inorganic chemistry
www.rsc.org/dalton
Number 22 | 14 June 2008 | Pages 2901–3016
ISSN 1477-9226
HOT ARTICLE
Braunstein et al.
PERSPECTIVE
Bo and Maseras
Nickel and iron complexes with
N,P,N-type ligands
QM/MM methods in inorganic
chemistry
1477-9226(2008)22;1-5

View Article Online
PAPER
www.rsc.org/dalton | Dalton Transactions
Nickel and iron complexes with N,P,N-type ligands: synthesis, structure and
catalytic oligomerization of ethylene†
Anthony Kermagoret, Falk Tomicki and Pierre Braunstein*
Received 4th February 2008, Accepted 20th March 2008
First published as an Advance Article on the web 1st May 2008
DOI: 10.1039/b802009d
The N,P,N-type ligands bis(2-picolyl)phenylphosphine (1), bis(4,5-dihydro-
2-oxazolylmethyl)phenylphosphine (2), bis(4,4-dimethyl-2-oxazolylmethyl)phenylphosphine (3) and
bis(2-picolyloxy)phenylphosphine (4) were used to synthesize the corresponding pentacoordinated
Ni(II) complexes [Ni{bis(2-picolyl)phenylphosphine}Cl2] (6), [Ni{bis(4,5-
dihydro-2-oxazolylmethyl)phenylphosphine}Cl2] (7), [Ni{bis(4,4-dimethyl-
2-oxazolylmethyl)phenylphosphine}Cl2] (8) and [Ni{bis(2-picolyloxy)phenylphosphine}Cl2] (9),
respectively. The hexacoordinated iron complexes [Fe{bis(2-picolyl)phenylphosphine}2][Cl3FeOFeCl3]
(10), [Fe{bis(4,5-dihydro-2-oxazolylmethyl)phenylphosphine}2][Cl3FeOFeCl3] (11) and the
tetracoordinated complex [Fe{bis(4,4-dimethyl-2-oxazolylmethyl)phenylphosphine}Cl2] (abbreviated
[FeCl2(NPNMe2-N,N )]) were prepared by reaction of FeCl2·4H2O with ligands 1–3, respectively. The
crystal structures of the octahedral complexes 10 and 11, determined by X-ray diffraction, showed that
two tridentate ligands are facially coordinated to the metal centre with a cis-arrangement of the P
atoms and the dianion (l-oxo)bis[trichloroferrate(III)] compensates the doubly positive charge of the
complex. The cyclic voltammograms of 10 and 11 showed two reversible redox couples attributed to the
reduction of the dianion (Fe2OCl6)2− (−0.24 V for 10 and −0.20 V for 11 vs. SCE) and to the oxidation
of the Fe(II) ion of the complex (0.67 V for 10 and 0.52 V for 11 vs. SCE). The cyclic voltammogram of
[FeCl2(NPNMe2-N,N )] showed a reversible redox couple at −0.17 V vs. SCE assigned to the oxidation of
the Fe(II) atom and an irreversible process at 0.65 V. The complexes 6, 8–11 and [FeCl2(NPNMe2-N,N )]
have been evaluated in the catalytic oligomerization of ethylene with AlEtCl2 or MAO as cocatalyst.
The nickel complex 6 proved to be the most active precatalyst in the series, with a turnover frequency
(TOF) of 61 800 molC molNi h−1 with 10 equiv. of AlEtCl2 and 12 200 molC molNi h−1 with
−1
−1
H
H
4
2
4
2
200 equiv. of MAO. Precatalysts 8 and 9 were the most selective in butenes, up to 90% with 6 equiv. of
AlEtCl2 and 89% with 2 equiv. of AlEtCl2, respectively, and up to 92% butenes with 400 equiv. of MAO
and 91% butenes with 200 equiv. MAO, respectively. The best selectivities for 1-butene were provided by
8 and AlEtCl2 (up to 31% with 6 equiv.) and 9 with MAO (up to 72% with 200 equiv.). The iron
complexes were not significantly active with AlEtCl2 or MAO as cocatalyst.
Introduction
There is considerable current academic and industrial interest in
catalytic ethylene oligomerization, in particular for the production
of linear a-olefins in the C4–C10 range whose demand is growing
fast. The need for identifying and fine-tuning the parameters
which influence the activity and selectivity of metal catalysts
is generating much effort at the interface between ligand de-
sign, coordination/organometallic chemistry and homogeneous
catalysis.1 We reported recently the synthesis of various mono- and
dinuclear nickel complexes bearing five-, six-, or seven-membered
P,N-chelating ligands in which the P donor function was of the
phosphine, phosphonite or phosphinite-type and the N donor
Scheme 1 Some Ni(II) complexes with P,N-type ligands.
was part of a pyridine or oxazoline heterocycle (Scheme 1).1e,2
Laboratoire de Chimie de Coordination, Institut de Chimie (UMR 7177
CNRS), Universite´ Louis Pasteur, 4 rue Blaise Pascal, F-67070, Strasbourg
Ce´dex, France. E-mail: braunstein@chimie.u-strasbg.fr
† CCDC reference numbers 676912 and 676913. For crystallographic data
in CIF or other electronic format see DOI: 10.1039/b802009d
The coordination geometry about the metal was found to strongly
depend on the ligand substituents and ranged from square-planar
or tetrahedral for mononuclear complexes to square-pyramidal or
trigonal bipyramidal for dinuclear complexes.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 2945–2955 | 2945
©

View Article Online
In the presence of even small amounts of AlEtCl2 or MAO as
cocatalyst, these Ni(II) complexes turned out to be highly active
and selective precatalysts for ethylene dimerization.1e,2
Investigations with related tridentate N,P,N-type ligands to-
ward the formation of new oligomerization catalysts led to
the synthesis of the pentacoordinated mononuclear nickel com-
plex [NiCl2(NOPONMe2-N,P,N )] (NOPONMe2 = bis(4,4-dimethyl-
2-oxazolyldimethylmethoxy)phenylphosphine, Scheme 2), which
is an effective precatalyst leading to selectivities in C4 olefins higher
4 -
The complexes [PdCl2(NPNMe2-P, N )] and [Pd(NPNMe2
3
Scheme
N,P,N )(NCMe)](BF4)2.
than 90% when activated with AlEtCl2.
[FeCl2(4,4-dimethyl-2-oxazolyldimethylmethanolate)]2 was iso-
lated which contains the N,O ligand bonded in
a l-
1
2
g (O):g (N,O) manner (the complex is abbreviated as [FeIII{l-
1
2
g (O):g (N,O)}Cl2]2) (Scheme 5).3 In contrast, the reaction be-
tween FeCl2·4H2O and the ligand NPNMe2 yielded the complex
[FeCl2(NPNMe2-N,N )] in which the ligand displays a N,N coordi-
nation mode (Scheme 5), similar to that in [CoCl2(NPNMe2-N,N )].4
Scheme 2 The complex [NiCl2(NOPONMe2-N,P,N )].
The use of the NOPONMe2 ligand was extended to the formation
of the new Co(II) complex [CoCl2(NOPONMe2-P, N )] (Scheme 3)
which also catalyzes ethylene oligomerization in the presence
of AlEtCl2 as a cocatalyst.4 In contrast to the pentacoordina-
tion observed in [NiCl2(NOPONMe2-N,P, N )], [CoCl2(NOPONMe2
-
P, N )] is tetracoordinated and displays bidentate P,N-coordination
of the ligand with one pendant oxazoline arm. A comparative
study of the behaviour of related phosphonite and phosphine
N,P,N ligands toward Co(II) precursors led to the synthesis of the
new complex [CoCl2(NPNMe2-N,N )] (NPNMe2 = bis(4,4-dimethyl-
2-oxazolylmethyl)phenylphosphine (3)) (Scheme 3). Its structural
determination revealed an unexpected N,N-chelation mode of the
ligand and the lack of phosphorus coordination contrasted with
the situation in [CoCl2(NOPONMe2-P, N )].4
1
2
Scheme 5 The complexes [FeIII{l-g (O):g (N,O)}Cl2]2 and [FeIICl2-
(NPNMe2-N,N )].
This diversity of results prompted us to study further the
coordination behaviour of tridentate N,P,N-type ligands with iron
and nickel complexes. Only a few nickel complexes coordinated by
N,P,N ligands have been synthesized,6 although they might have
promising applications in the catalytic oligomerization of ethylene.
Results and discussion
Synthesis of the ligands
The phosphino-pyridine ligand
1 was prepared according
to the literature7 but the yield can be increased if 2-
[(trimethylsilyl)methyl]pyridine is isolated by distillation before
PPhCl2 is added [eqn (1)]. The pure ligand was obtained in yields
Scheme
3
The complexes [CoCl2(NOPONMe2-P, N )] and [CoCl2-
(NPNMe2-N,N )].
up to 68% and it has been characterized by 1H, 31P{ H} and
1
Because of the different P,N- and N,N-coordination modes
observed in the Co(II) complexes [CoCl2(NOPONMe2-P, N )] and
[CoCl2(NPNMe2-N,N )], respectively, we wished to compare the
structure of the products obtained from NOPONMe2 or NPNMe2
and different Pd(II) precursors. The nature of the counteranion was
found to play a crucial role on the ligand coordination mode:4,5
1
13C{ H} NMR spectroscopic methods.
halides led to a P,N-coordination mode of the ligands NOPONMe2
(1)
−
and NPNMe2, whereas non-coordinating anions, such as BF4
,
resulted in N,P,N-coordination (Scheme 4).
The phosphino-oxazoline ligands 2 and 3 were prepared as de-
scibed by deprotonation of the corresponding oxazoline precursor
followed first by the addition of chlorotrimethylsilane and then
addition of 0.5 equiv. of PPhCl2 at −78 ◦C [eqn (2)].4,8
During attempts to synthesize iron complexes with the
NOPONMe2 ligand by its reaction with anhydrous FeCl3, cleavage
of the P–O bond was observed and the dinuclear complex
2946 | Dalton Trans., 2008, 2945–2955
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
the presence of uncoordinated pyridine or oxazoline functions and
this suggested a tridentate behaviour for ligands 1–4, similar to
that found in [NiCl2(NOPONMe2-N,P,N )].3 The strong IR band
at 1270 cm−1 for 7 shifts to 1320 cm−1 for the methyl substituted
derivative 8. The nickel complexes 6–9 are paramagnetic in solu-
tion and their magnetic moments, determined in CD2Cl2 by the
Evans method,10–13 were 3.1, 2.7, 2.8 and 2.7 lB, respectively. These
values are in the range found for Ni(II) complexes coordinated by
P,N type ligands.3,13–16 Complexes 6–9 were characterized by IR
spectroscopy and elemental analysis and their green colour in
the solid state is comparable with that of related five-coordinated
complexes.3 It is noteworthy that bidentate P,N chelates, such as
Ph2PCH2ox where ox represents the oxazoline featured in 2 were
recently shown to give rise to the formation of interconvertible
mono- and tetranuclear Ni(II) complexes.17
(2)
Owing to the presence of the prochiral P centre, the CH2 protons
in aposition to the phosphorus atom of ligand 2 display an ABMX
spin system (A = B = M = H, X = P)8 and the signals of the PCH2
protons of 1 and 34 correspond to ABX spin systems (A = B = H,
X = P).
The sensitivity of phosphonites to the presence of HCl3
prevents an efficient synthesis of 4 by direct reaction of PPhCl2
with (pyridine-2-yl)methanol, which leads in addition to 4
to (pyridin-2-yl)methyl hydrogen phenylphosphonate 5 and 2-
(chloromethyl)pyridine by-products despite the presence of NEt3
[eqn (3)].
Synthesis of the iron complexes
In scheme 6 are summarized the syntheses of the Fe(II) complexes
10, 11 and [FeCl2(NPNMe2-N,N )],4 which were obtained by reaction
of one equiv. of FeCl2·4H2O with the ligands 1–3, respectively.
During their synthesis, complexes 10 and 11 precipitated instan-
taneously from CH2Cl2 and, after work-up, they were isolated as
red powders in 62% and 58% yield, respectively. The IR spectrum
(3)
of complex 10 was similar to that of the corresponding Ni complex
However, the reaction of two equiv. of pyridine alcoholate with
−1
=
6. The strong IR band for the m(C N) vibration of 11 at 1639 cm
one equiv. of PPhCl2 at −78 ◦C in THF yielded the desired
is shifted towards lower wavenumbers compared to that of its
corresponding nickel complex 7 (1658 cm−1). Slow diffusion of
heptane into a CH2Cl2–MeCN (1 : 1) solution of the complexes
10 and 11 afforded crystals suitable for X-ray diffraction. ORTEP
views of their crystal structures are presented in Fig. 1 and 2 and
selected bond distances and angles are given in Tables 1 and 2,
respectively.
1
tridentate pyridine-phosphonite ligand 4 [eqn (4)]. Its H NMR
spectrum contains an ABX spin system (A = B = H, X = P) for
the CH2 protons.
(4)
The X-ray structural analyses established the fac-coordination
of each tridentate ligand and the mutual cis-position of the P
donors in these octahedral complexes. Interestingly, the reaction
of 2 with [RuCl2(DMSO)4] or [RuCl2(PPh3)3] has previously led to
the formation of the dichloro complexes 12 and 13, respectively, in
which there is only one tridentate N,P,N ligand coordinated to the
Synthesis of the nickel complexes
The Ni(II) bis(2-picolyl)phenylphosphine complex 6 and the
bis(2-oxazolylmethyl) phenylphosphine complexes 7 and 8 were
prepared in MeOH by reaction between equimolar amounts of
NiCl2 with the N,P,N ligands 1–3, respectively, and isolated in
yields of 96%, 61% and 77%, respectively.
Complex 9 was obtained by stirring one equiv. of ligand with
one equiv. of [NiCl2(DME)] (DME = 1,2-dimethoxylethane) in
CH2Cl2 at room temperature. This Ni(II) precursor was preferred
over NiCl2·6H2O9 because of its better solubility in CH2Cl2 and
methanol should be avoided as a solvent with ligands containing
a P–OR bond.2 Coordination of the ligand to the metal centre
was confirmed in IR spectroscopy by the strong absorption
Fig. 1 ORTEP view of the structure of 10 in 10·0.5CH2Cl2 with thermal
ellipsoids drawn at the 50% probability level, symmetry operator for the
generation equivalent position: −x + 1/2, −y + 1/2, z (ꢀ) and −x + 1/2,
−y + 3/2, z (ꢀꢀ).
=
corresponding to the m(C N) vibrations at 1600, 1658, 1636 and
1603 cm−1 for 6–9, respectively. The IR spectra of 6–9 did not show
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 2945–2955 | 2947
©

View Article Online
Scheme 6 Comparative synthesis of the iron complexes 10, 11 and [FeCl2(NPNMe2-N,N )].
◦
a
˚
Table 1 Selected bond distances (A) and angles ( ) in 10·0.5CH2Cl2
Fe1–P
Fe1–N1
Fe1–N2
N1–C8
2.165(2)
2.097(4)
2.022(4)
1.355(7)
1.506(8)
81.5(2)
177.5(2)
85.6(2)
96.4(2)
C7–P
P–C13
C13–C14
C14–N2
1.831(6)
1.823(6)
1.506(8)
1.356(7)
1.754(2)
C8–C7
Fe2–O1
P–Fe1–N1
P–Fe1–N1ꢀ
P–Fe1–N2
P–Fe1–N2ꢀ
P–Fe1–Pꢀ
N1–Fe1–N1ꢀ
N1–Fe1–N2
N1–Fe1–N2ꢀ
N1–Fe1–Pꢀ
N2–Fe1–N1ꢀ
N2–Fe1–N2ꢀ
N2–Fe1–Pꢀ
N1ꢀ–Fe1–N2ꢀ
N1ꢀ–Fe1–Pꢀ
N2ꢀ–Fe1–Pꢀ
Fe2–O1–Fe2ꢀꢀ
177.5(2)
96.7(2)
176.9(2)
96.4(2)
81.3(2)
81.5(2)
85.6(2)
154.1(5)
99.33(9)
97.9(3)
81.3(2)
96.7(2)
a Symmetry operator for the generation equivalent position: −x + 1/2, −y
+ 1/2, z (ꢀ) and −x + 1/2, −y + 3/2, z (ꢀꢀ).
Fig. 2 ORTEP view of the structure of 11 in 11·4MeCN with thermal
ellipsoids drawn at the 50% probability level, symmetry operator for the
generation equivalent position: −x, y, −z +1/2 (ꢀ) and −x + 1, y, −z + 1/2
(ꢀꢀ).
◦
a
˚
Table 2 Selected bond distances (A) and angles ( ) in 11·4MeCN
Fe1–P
2.1842(6)
2.042(2)
1.989(2)
1.276(3)
1.486(3)
82.95(5)
170.61(5)
82.40(5)
98.25(5)
96.26(3)
99.34(9)
88.44(7)
90.92(7)
C7–P
P–C11
1.841(2)
1.836(2)
1.474(3)
1.274(3)
1.7600(5)
Fe1–N1
Fe1–N2
C11–C12
metal centre, also in a fac-coordination mode, and their octahedral
geometry was achieved by further coordination of one molecule
of DMSO or PPh3, respectively.8,18
N1–C8
C8–C7
C12–N2
Fe2–O3
P–Fe1–N1
P–Fe1–N1ꢀ
P–Fe1–N2
P–Fe1–N2ꢀ
P–Fe1–Pꢀ
N1–Fe1–N1ꢀ
N1–Fe1–N2
N1–Fe1–N2ꢀ
N1–Fe1–Pꢀ
N2–Fe1–N1ꢀ
N2–Fe1–N2ꢀ
N2–Fe1–Pꢀ
N1ꢀ–Fe1–N2ꢀ
N1ꢀ–Fe1–Pꢀ
N2ꢀ–Fe1–Pꢀ
Fe2–O3–Fe2ꢀꢀ
170.62(5)
88.44(7)
179.0(1)
98.25(5)
90.92(7)
82.95(5)
82.40(5)
162.0(2)
a Symmetry operator for the generation equivalent position: −x, y, −z
+1/2 (ꢀ) and −x + 1, y, −z + 1/2 (ꢀꢀ).
2948 | Dalton Trans., 2008, 2945–2955
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Complexes 10 and 11 adopt a distorted octahedral geometry as
shown by the values of the P–Fe1–N1ꢀ and N2–Fe1–N2ꢀ angles
(177.5(2) and 176.9(2)◦, respectively, for 10 and 170.61(5) and
179.0(1)◦, respectively, for 11 (Tables 1 and 2). The phosphorus
atoms are coordinated to the Fe(II) ion trans to a nitrogen atom
and the dications have a C2 symmetry axis passing through the
the formation of metal complexes with a N,N coordination of the
ligand.
Electrochemistry
The complexes 10, 11 and [FeCl2(NPNMe2-N,N )] have been studied
by cyclic voltammetry (Fig. 3 and 4). The cyclic voltammograms of
10 and 11 (Fig. 3) showed two reversible redox couples attributed
respectively to the reduction of the dianion (Fe2OCl6)2− (−0.24 V
for 10 and −0.20 V for 11 vs. SCE) and to the oxidation of the Fe(II)
centre of the complex (0.67 V for 10 and 0.52 V for 11 vs. SCE).
Coordinated by pyridine functions, the more electron-rich 10 has
a redox potential for the Fe(II) centre shifted to more positive
values compared to 11 and this indicates a slightly higher stability
of Fe(II) in 10 than in 11. However, compared to Fe(II) complexes
coordinated by two N,N,N-tridentate ligands containing pyridine
functions and with redox potentials between 0.78 and 1.06 V vs.
SCE, the redox potential of 10 is at less positive values.40,41
˚
metal centre. For 11, the Fe1–N1 bond distance (2.042(2) A),
˚
slightly longer than Fe1–N2 (1.989(2) A), results from the trans
influence of the phosphorus atom on N1. This is also noted in
˚
complex 10 with the slightly longer Fe1–N1 distance (2.097(4) A)
˚
compared to Fe1–N2 (2.022(4) A).
The doubly positive charge of the metal complex in 10 or
11 is compensated by the dianion (l-oxo)bis[trichloroferrate(III)]
(Fe2OCl6)2− which results from the partial oxidation of the Fe(II)
precursor.19–22 Complexes where a Fe(II) centre is coordinated
by two tridentate ligands have been described with a (FeCl4)2−
counter ion.23–27 Since its first description in 1978, this ferric
dianion has been studied for its magnetic or structural properties
and its tetraalkylammonium derivatives (R4N)2(Fe2OCl6) are
convenient starting materials for the synthesis of polynuclear
iron complexes.28–32 Iron complexes containing (Fe2OCl6)2− have
been reported and formation of this dianion could occur via
an intermediate Fe(II) anion which rapidly autoxidizes.19–22,32–36
The overall stoichiometry of three iron atoms for two ligands in
complexes 10 and 11 suggests the loss of one third of the ligands
1 and 2.
Despite the similar Fe2–O distances in the dianion present for
˚
10 and 11 (1.754(2) and 1.760◦0(5) A, respectively), the Fe2–O1–
Fe2ꢀꢀ bond angle in 10 (154.1(5) ) is smaller than the corresponding
Fe2–O3–Fe2ꢀꢀ angle in 11 (162.0(2)◦). According to the literature,
the Fe–O distance in the dianion (Fe2OCl6)2− is generally found
˚
around 1.76 A but the value of the Fe–O–Fe angle varies between
Fig. 3 Cyclic voltammograms of 10 and 11 in anhydrous MeCN (0.1 M
147.7 and 180.0◦.30,35
N(n-Bu)4PF6) at a scan rate of 100 mV s−1
.
The white complex [FeCl2(NPNMe2-N,N )] was obtained in
87% yield4 and its magnetic moment of 4.2 lB has now been
determined in CD2Cl2 by the Evans method.10–13 This value is
slightly lower than expected for Fe(II) high spin complexes.37,38
Its crystal structure showed a distorted tetrahedral geometry
with the N,N-coordination of ligand 3 without the expected
coordination of the phosphorus to the iron centre.4 The N,N-
chelating behaviour of 3 in [FeCl2(NPNMe2-N,N )] compared to the
N,P,N-tridentate behaviour of 2 in 11 appears to be related to the
steric effects generated by the presence of methyl substituents on
the oxazoline ring. Depending on the nature of the counteranion
present, ligand 3 has been found to adopt either P,N or N,P,N
coordination modes in Pd(II) complexes and phosphorus coor-
Fig. 4 Cyclic voltammogram of [FeCl2(NPNMe2-N,N )] in anhydrous
MeCN (0.1 M N(n-Bu)4PF6) at a scan rate of 100 mV s−1
1
dination to the metal centre was confirmed by 31P{ H} NMR
.
spectroscopy.4 The role of the counter-anion on the coordination
mode of ligand 2 has also been studied with cationic Ru(II)–
benzene complexes and P,N coordination was observed with
monocationic complexes and N,P,N coordination with dicationic
complexes.8 The ease of conversion between the different coor-
dination modes of 2 and 3 is consistent with their hemilabile
properties.4,39
The cyclic voltammogram of [FeCl2(NPNMe2-N,N )] showed a
reversible redox couple at −0.17 V vs. SCE attributed to the
oxidation of the Fe(II) centre and an irreversible process at 0.65 V
(Fig. 4). This potential is slightly more negative than in Fe(II)
complexes coordinated by a-diimine ligands.42
The crystal structure of 10 suggested that ligands 1 and 2 have
similar coordination behaviours despite the different nature of
their heterocyclic nitrogen donor atom. It will be interesting to
examine whether the addition of substituents on the pyridine
function, to increase the steric component, could also lead to
Catalytic oligomerization of ethylene
The complexes 6, 8–11 and [FeCl2(NPNMe2-N,N )] have been
evaluated in the catalytic oligomerization of ethylene with
different amounts of AlEtCl2 or MAO as cocatalyst. The
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 2945–2955 | 2949
©

View Article Online
complex [NiCl2{P(n-Bu)3}2], which is a typical oligomerization
precatalyst,43 was used as a reference. The Ni complexes 6, 8 and 9
were active when 2, 4, 6 or 10 equiv. of AlEtCl2 (Table 3) or 200 or
400 equiv. of MAO (Table 4) were used as cocatalyst (Fig. 5–9). In
contrast to 6, complexes 8 and 9 did not have significant catalytic
activities with 100 equiv. of MAO. Complex 6 was the most active,
with a turnover frequency (TOF) of 61 800 molC2 molNi h−1
−1
H
4
with 10 equiv. of AlEtCl2 (Fig. 5) and of 12 200 molC molNi−1 h−1
H
2
4
with 200 equiv. of MAO (Fig. 7). With AlEtCl2 as cocatalyst, 8
showed the best selectivities for butenes (up to 90% with 6 equiv.
of AlEtCl2 and up to 31% 1-butene) whereas with MAO, 9 was
the most selective, leading to up to 92% butenes with 400 equiv. of
MAO and up to 72% 1-butene with 200 equiv. of MAO. Similarly
to Ni(II) complexes with P,N-type ligands, complexes 6 and 9
coordinated by pyridine functions afforded with AlEtCl2 or MAO
Fig. 8 Selectivity of the complexes 6, 8 and 9 for C4 compounds using
MAO as cocatalyst (1-butene and 2-butene).
Fig. 5 Catalytic activities of the complexes 6, 8 and 9 in the oligomeriza-
tion of ethylene using AlEtCl2 as cocatalyst, Ref: [NiCl2{P(n-Bu)3}2].
Fig. 9 Selectivity of the complexes 6, 8 and 9 for 1-butene using MAO as
cocatalyst.
as cocatalyst better activities than the oxazoline-based systems 8
and [NiCl2(NOPONMe2-N,P,N )], respectively.3
The nature of the cocatalyst has an impact on the catalytic
results: MAO leads to systems less active but more selective in 1-
butene and in C4 olefins than AlEtCl2. Increasing the amount
of cocatalyst makes the system more active but less selective
in C4 olefins and in 1-butene. As shown by 6 with AlEtCl2 as
cocatalyst, very active systems favored the isomerization of 1-
butene in 2-butene (selectivity in 1-butene less than 9%, Fig. 6)
and the reinsertion of 1-butene and 2-butene to form C6 oligomers
(selectivity in C6 oligomers up to 43%, Table 3).
A comparison of the catalytic results of 6 and 9 with AlEtCl2
as cocatalyst with those obtained with 14 and 15 indicates that
precatalysts coordinated by a P,N chelate lead to slightly higher
activities than those coordinated by a N,P,N tridentate ligand.2
However, this does not significantly affect the selectivities in C4
olefins and in 1-butene. With MAO as cocatalyst, 15 was more
active but significantly less selective in 1-butene than 9, despite
similar catalytic results observed for 6 and 14.2
Fig. 6 Selectivity of the complexes 6, 8 and 9 for 1-butene using AlEtCl2
as cocatalyst, Ref: [NiCl2{P(n-Bu)3}2].
Some tetrahedral iron(II) complexes have shown good activities
in oligomerization and polymerization of ethylene44,45 and it was
therefore interesting to test [FeCl2(NPNMe2-N,N )] with different
Fig. 7 Catalytic activities of the complexes 6, 8 and 9 in the oligomeriza-
tion of ethylene using MAO as cocatalyst.
2950 | Dalton Trans., 2008, 2945–2955
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 3 Comparative catalytic data for complexes 6, 8 and 9 in the oligomerization of ethylene with AlEtCl2 as cocatalysta
Selectivity (mass%)
Productivity/gC gNi h−1
TOF/molC molNi h−1
a-Olefin (C4)b (%)
ka
−1
−1
c
AlEtCl2 (equiv.)
C4
C6
C8
2 H4
2 H4
6
6
8
8
9
9
9
6
10
6
10
2
4
6
53
54
90
76
89
77
53
53
60
43
42
10
23
11
21
41
43
35
4
26 400
29 500
4 200
18 000
1 600
14 900
23 800
27 800
38 800
55 600
61 800
8 800
37 700
3 400
31 300
49 900
58 200
81 400
8
9
0.57
4
<1
1
<1
2
6
4
0.55
<0.10
0.20
<0.10
0.18
0.51
0.57
31
11
11
11
13
8
9
10
6
Refd
5
3
0.40
a Conditions: T = 30 ◦C, 10 bar of C2H4, 35 min, 4 × 10−5 mol Ni complex; solvent 12 or 10 mL chlorobenzene and 3 or 5 mL of cocatalyst toluene solution
for 6 or 10 equiv. of AlEtCl2, respectively (total volume 15 mL). b Within the C4 fraction. c ka = hexenes (mol)/butenes (mol). d Ref = [NiCl2{P(n-Bu)3}2].
Table 4 Comparative catalytic data for complexes 6, 8 and 9 in the oligomerization of ethylene with MAO as cocatalysta,b
Selectivity (mass%)
Productivity/gC gNi h−1
TOF/molC molNi h−1
a-Olefin (C4)c (%)
ka
−1
−1
d
MAO (equiv.)
C4
C6
C8
2 H4
2 H4
6
6
8
8
9
9
100
200
200
400
200
400
71
69
91
92
91
77
26
25
8
7
8
3
3900
5800
500
700
700
9300
12200
1200
1500
1500
3400
24
24
56
59
72
62
0.24
0.24
<0.10
<0.10
<0.10
0.17
6
<1
<1
<1
3
20
1600
◦
a Conditions: T = 30 C, 10 bar of C2H4, 35 min, 4 × 10−5 mol Ni complex; complex dissolved in 10 mL chlorobenzene to which 4, 8 or 16 mL of
cocatalyst solution in toluene is added for 100, 200 or 400 equiv. of MAO, respectively. b Traces of C10 oligomers were detected when using 400 equiv. of
MAO. c Within the C4 fraction. d ka = hexenes (mol)/butenes (mol).
cocatalysts. However, with 200 equiv. of MAO or 6 equiv.
of AlEtCl2, [FeCl2(NPNMe2-N,N )] was poorly active and only
traces of butenes were detected by gas chromatography. The
hexacoordinated mononuclear iron complexes 10 and 11 showed
no activity with MAO or AlEtCl2 as cocatalyst, probably because
of the saturated coordination sphere of the iron centre which
inhibits ethylene coordination to the metal centre. Similarly, iron
complexes containing two C,N,C pincer ligands were also found
inactive.46,47
vs. SCE) for the oxidation of the Fe(II) centre and an irreversible
process (0.65 mV).
Complex 6 is the most active in ethylene oligomerization
within the series, with a TOF of 61 800 molC2 molNi h−1
−1
H
4
in the presence of 10 equiv. of AlEtCl2 as cocatalyst and of
12 200 molC molNi−1 h−1 with 200 equiv. of MAO. Precatalysts 8
H
2
4
and 9 were the most selective in butenes, up to 90% with 6 equiv.
of AlEtCl2 and 89% with 2 equiv. of AlEtCl2, respectively, and
up to 92% butenes with 400 equiv. of MAO and 91% butenes
with 200 equiv. MAO, respectively. The more active systems
favored the isomerization of 1-butene and the reinsertion of C4
olefins to form C6 oligomers. In all cases, the catalytic activity
increased when more cocatalyst was used and AlEtCl2 led to more
active but less selective systems than MAO. The use of tridentate
N,P,N ligands instead of bidentate P,N chelates resulted in slightly
less active catalysts but similarly selective in C4 olefins and 1-
butene with AlEtCl2 as cocatalyst. The iron complexes 10, 11 and
[FeCl2(NPNMe2-N,N )] did not show significant activities in ethylene
oligomerization, either with AlEtCl2 or MAO as cocatalyst.
Conclusion
The phosphino-pyridine 17 and the phosphino-oxazoline ligands
28 and 34 and ligand 4, synthesized by reaction of PPhCl2
with 2 equiv. of its corresponding pyridine alcoholate at low
temperature, were used to prepare the nickel complexes 6–9 in
good yields by reaction with NiCl2 in methanol or [NiCl2(DME)]
in CH2Cl2. The IR spectra of 6–9 suggested the coordination of
the two nitrogen atoms of the tridentate ligands 1–4. Ligands 1–
3 were also used for the preparation of the iron complexes 10
and 11 in which the dicationic Fe(II) centre is hexacoordinated
and of the tetrahedral Fe(II) complex [FeCl2(NPNMe2-N,N )].4 The
crystal structures of 10 and 11 also established the formation of the
dianion (l-oxo)bis[trichloroferrate(III)] to compensate the doubly
positive charge. Whereas the cyclic voltammograms of 10 and
11 showed two reversible redox couples for the reduction of the
dianion (Fe2OCl6)2− and the oxidation of the Fe(II) ion, that of
[FeCl2(NPNMe2-N,N )] showed a reversible redox couple (−0.17 mV
Experimental section
General considerations
All reactions were performed under purified nitrogen. Solvents
were purified and dried under nitrogen by conventional methods.
1
The 1H NMR spectra were recorded at 300 MHz, 31P{ H} NMR
1
spectra were recorded at 121.5 MHz, and 13C{ H} NMR spectra
were recorded at 75.5 MHz on a FT Bruker AC300 instrument. IR
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 2945–2955 | 2951
©

View Article Online
spectra in the range of 4000–400 cm−1 were recorded on a Bruker
IFS28FT. Gas chromatographic analyses were performed on a
Thermoquest GC8000 Top Series gas chromatograph using a HP
Pona column (50 m, 0.2 mm diameter, 0.5 lm film thickness).
Magnetic moments were determined by the Evans method in
CD2Cl2 using a solution of CH3NO2 in CD2Cl2 (20 : 80, v/v) as
reference.9–13 Mass spectra were recorded with a Bruker Daltonics
microTOF (ESI; positive mode; capillary voltage: 4.8 kV; nebulizer
pressure: 0.2 bar; desolvation temperature: 180 ◦C; desolvation
gas flow rate: 4.5 L min−1). Ligands 28 and 3,4 and the complexes
[NiCl2(DME)]2 and [FeCl2(NPNMe2-N,N )]4 have been synthesized
according to the literature.
was added to precipitate LiCl. The yellow solution was filtered
with a cannula and the solvents were removed under reduced
pressure. The resulting yellow oil was dried under vacuum at
60 ◦C overnight (Yield: 1.75 g, 5.4 mmol, 86%). 1H NMR
(300 MHz, CDCl3): d ABX spin system (A = B = H, X = P)
4.90 (2H, dd, JAB = 13.8 Hz, 2JXB = 7.2 Hz), 5.07 (2H, dd, JAB
=
13.8 Hz, 2JXA = 8.4 Hz), 7.15–7.17 (2H, m, aromatic H), 7.44–7.46
(5H, m, aromatic H), 7.62–7.67 (2H, m, aromatic H), 7.71–7.76
1
(2H, m, aromatic H) 8.50–8.52 (2H, m, CHCN); 13C{ H} NMR
(75.5 MHz, CDCl3): d 69.3 (d, 2JPC = 9.7 Hz, OCH2), 121.1 (s, CH
of pyridyl), 122.4 (s, CH of pyridyl), 128.45 (d, 3JPC = 5.5 Hz, m-
CH of aryl), 129.8 (s, p-CH of aryl), 130.2 (d, 2JPC = 23.9 Hz, o-CH
of aryl), 136.7 (s, CH of pyridyl), 140.15 (d, 1JPC = 19.6 Hz, P-C of
aryl), 149.1 (s, CH of pyridyl), 158.4 (d, 2JPC = 5.4 Hz, NC-CH2
Syntheses
1
of pyridyl); 31P{ H} NMR (121.5 MHz, CDCl3): d 161.91.
Synthesis of bis(2-picolyl)phenylphosphine 17. A n-BuLi solu-
tion (96.0 mmol, 1.6 M in hexane) was added dropwise to a
degassed solution of 2-picoline (8.94 g, 96.0 mmol) in 100 mL
of THF at −78 ◦C. After complete deprotonation, the red
anion precipitated and after further stirring for 1 h at −78 ◦C,
degassed chlorotrimethylsilane (10.43 g, 96.0 mmol) was added
dropwise to the solution. The brown mixture was allowed to reach
room temperature overnight and the THF was evaporated under
reduced pressure. The residue was distilled at 120 ◦C and 12 mbar
to obtain the pure, liquid 2-[(trimethylsilyl)methyl]pyridine (Yield:
6.61 g, 40.0 mmol, 42%). 1H NMR (300 MHz, CDCl3): d 0.00 (s,
Synthesis of (pyridin-2-yl)methyl hydrogen phenylphosphonate
5. To a THF solution of (pyridin-2-yl)methanol (3.21 g,
29.4 mmol) containing 10 mL of NEt3 was added PPhCl2 (2.64 g,
14.7 mmol) at −78 ◦C. The mixture was stirred overnight at room
temperature. All volatiles were removed under reduced pressure
and the residue was dissolved in diethyl ether. The solution was
filtered and the solvent was removed under reduced pressure to
yield a yellow oil (Yield: 40% evaluated by 1H NMR monitoring
1
of the chemical shifts of the CH2 protons of 4 and 5). H NMR
(300 MHz, CDCl3): d ABX spin system (A = B = H, X = P)
2
5.16 (1H, dd, JAB = 13.0 Hz, 2JXB = 9.1 Hz), 5.26 (1H, dd, JAB
=
9H, SiCH3), 2.32 (s, 2H, SiCH2), 6.93 (t, JH-H = 7.65 Hz, 2H,
2
2
1
13.0 Hz, 2JXA = 9.1 Hz), 7.15–8.51 (9H, m, aromatic H); 31P{ H}
aromatic H), 7.45 (td, 1H, JH-H = 1.85 Hz, JH-H = 7.65 Hz,
3
1
aromatic H), 8.39 (d, 1H, JH-H = 4.17, CHCN). 13C{ H} NMR
(CDCl3, 75.5 MHz): d −1.7 (s, SiCH3), 30.2 (s, SiCH2), 119.0 (s, C
of pyridyl), 122.1 (s, C of pyridyl), 135.7 (s, C of pyridyl), 148.9 (s,
C of pyridyl), 161.3 (s, C of pyridyl).
NMR (121.5 MHz, CDCl3): d 27.6 (s).
Synthesis of [Ni{bis(2-picolyl)phenylphosphine}Cl2] 6. A solu-
tion of 1 (1.40 g, 4.8 mmol) in methanol (50 mL) was stirred
while NiCl2 (0.62 g, 4.8 mmol) in methanol (50 mL) was added
dropwise to the solution. After several hours, the solvent was
evaporated and 10 mL of CH2Cl2 was added to the red oil. 50 mL
of petroleum ether (boiling point 35–60 ◦C) was added to the
solution to precipitate the green complex. After the suspension
was sonicated for 1 h, the complex was filtered off and washed
several times with diethyl ether. The complex was dried under
vacuum (Yield: 2.00 g, 4.6 mmol, 96%). IR (KBr): 1600 (vs), 1566
(m), 1477 (s), 1435 (vs), 1386 (w), 1311 (w), 1155 (m), 1107 (m),
1055 (w), 1016 (w), 820 (w), 745 (s), 694 (m), 482 (m) cm−1 (m).
Anal. calcd for C18H17Cl2N2NiP: C, 51.24; H, 4.06; N, 6.64. Found:
C, 51.70; H, 4.57; N, 6.32%.
Pure picolyltrimethylsilane (6.61 g, 40.0 mmol) was dissolved
in a mixture of 50 mL THF and 50 mL diethyl ether cooled
down to −78 ◦C and PPhCl2 (3.58 g, 20.0 mmol) was added.
The reaction mixture was allowed to reach room temperature
overnight, the solvents were evaporated under reduced pressure
and the resulting orange oil was dried under vacuum at 60 ◦C
1
(Yield: 4.02 g, 13.6 mmol, 68%). H NMR (300 MHz, CDCl3):
d ABX spin system (A = B = H, X = P) 3.24 (2H, dd, JAB
=
13.3 Hz, 2JXB = 1.8 Hz), 3.31 (2H, d, JAB = 13.3 Hz), 6.93–6.89 (4H,
m, aromatic H), 7.15–7.22 (3H, m, aromatic H), 7.30–7.39 (4H,
1
m, aromatic H), 8.33–8.35 (2H, m, aromatic H); 13C{ H} NMR
(75.5 MHz, CDCl3): d 37.65 (d, 1JPC = 18.4 Hz, P-CH2), 120.8 (s,
CH of pyridyl), 123.6 (d, 3JPC = 5.1 Hz, CH of pyridyl), 128.25 (d,
3JPC = 6.9 Hz, m-CH of aryl), 129.1 (s, p-CH of aryl), 132.6 (d,
2JPC = 19.9 Hz, o-CH of aryl), 136.0 (s, CH of pyridyl), 136.7 (d,
1JPC = 18.4 Hz, C-P of aryl), 149.15 (s, CH of pyridyl), 158.1 (d,
Synthesis of [Ni{bis(4,5-dihydro-2-oxazolylmethyl)phenylphos-
phine}Cl2] 7. A solution of 2 (0.90 g, 3.25 mmol) in methanol
(50 mL) was stirred while NiCl2 (0.42 g, 3.25 mmol) in methanol
(50 mL) was added dropwise to the solution. After several hours
the solvent was evaporated and 10 mL of dried and degassed
CH2Cl2 was added to the red oil. 50 mL of petroleum ether
(boiling point 35–60 ◦C) was added to the solution to precipitate
the green complex. After the suspension was sonicated for 1 h,
the complex was filtered off and washed several times with diethyl
ether. The green complex was dried under vacuum (Yield: 0.80 g,
2.0 mmol, 61%). IR (KBr): 1658 (vs), 1478 (w), 1437 (m), 1402 (s),
1374 (m), 1339 (w), 1270 (s), 1168 (s), 1108 (w), 1034 (s), 999 (w),
972 (w), 932 (m), 868 (w), 749 (m), 694 (m) cm−1. Anal. calcd for
C14H17Cl2N2NiO2P: C, 41.43; H, 4.22; N, 6.90. Found: C, 40.43;
H, 4.42; N, 6.52%.
1
2JPC = 5.7 Hz, NC-CH2 of pyridyl); 31P{ H} NMR (121.5 MHz,
CDCl3): d −12.55. HRMS : Mass calcd for C18H18N2P : 293.1208.
Found. 293.1247 [(N,P,N) + H]+.
Synthesis of bis(2-picolyloxy)phenylphosphine 4. Liquid
(pyridin-2-yl)methanol (1.39 g, 12.7 mmol) was dissolved in
30 mL of THF and cooled at −78 ◦C. n-BuLi (1 equiv., 12.7 mmol
in 8 mL of hexane) was added dropwise and the solution was
stirred for 1 h. Then 0.5 equiv. of PPhCl2 (1.15 g, 6.35 mmol)
was added dropwise at −78 ◦C and the mixture was stirred
overnight to room temperature. The solvents were removed under
reduced pressure and 20 mL of a mixture diethyl ether–CH2Cl2
2952 | Dalton Trans., 2008, 2945–2955
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Oligomerization of ethylene
Synthesis of [Ni{bis(4,4-dimethyl-2-oxazolylmethyl)phenylphos-
phine}Cl2] 8. A solution of 3 (2.42 g, 7.3 mmol) in methanol
(50 mL) was stirred while NiCl2 (0.94 g, 7.3 mmol) in methanol
(50 mL) was added dropwise to the solution. After several hours
the solvent was evaporated and 10 mL of dried and degassed
CH2Cl2 was added to the red oil. 50 mL of petroleum ether
(boiling point 35–60 ◦C) was added to the solution to precipitate
the green complex. After the suspension was sonicated for 30 min,
the complex was filtered off and washed several times with diethyl
ether. Green complex 8 was dried under vacuum (Yield: 2.60 g,
5.6 mmol, 77%). IR (KBr): 1636 (vs), 1464 (m), 1437 (m), 1403
(m), 1369 (s), 1320 (vs), 1162 (s), 1107 (m), 1028 (w), 999 (s),
955 (s), 863 (w), 840 (w), 747 (s), 695 (s) cm−1. Anal. calcd for
C18H25Cl2N2NiO2P: C, 46.80; H, 5.45; N, 6.06. Found: C, 46.51;
H, 5.58; N: 5.77%. Mass calcd for C18H25ClN2NiO2P : 425.0696.
Found. 425.0641 [Ni(N,P,N)Cl]+.
All catalytic reactions were carried out in a magnetically stirred
(900 rpm) 145 mL stainless steel autoclave. A 125 mL glass
container was used to protect the inner walls of the autoclave
from corrosion. The preparation of the catalytic solution of the
precatalyst is dependent on the nature and the amount of the
cocatalyst.
With AlEtCl2, 4 × 10−2 mmol of Ni complex were dissolved in
14, 13, 12 or 10 mL of cholorobenzene depending on the amount
of cocatalyst and injected in the reactor under an ethylene flux.
Then 1, 2, 3 or 5 mL of a cocatalyst solution, corresponding to 2,
4, 6 or 10 equiv. respectively, are added to the reactor to achieve a
total volume of 15 mL with the precatalyst solution.
With MAO, 4 × 10−2 mmol of Ni complex were dissolved
in 10 mL in chlorobenzene and injected into the reactor under
an ethylene flux. Then 4, 8 or 16 mL of a cocatalyst solution,
corresponding to 100, 200 or 400 equiv. of MAO respectively,
were added.
Synthesis of [Ni{bis(2-picolyloxy)phenylphosphine}Cl2] 9. To
a solution of 4 (0.42 g, 1.3 mmol) in CH2Cl2 (30 mL) was
added [NiCl2(DME)] (0.30 g, 1.3 mmol). The dark green solution
was stirred overnight at room temperature. The solution was
concentrated to 5 mL and filtered. 50 mL of petroleum ether was
added to the solution to precipitate the complex. After filtration by
cannula, the green complex was washed with 20 mL of diethyl ether
and dried under vacuum overnight (Yield: 0.51 g, 1.2 mmol, 87%).
IR (KBr): 1603 (vs), 1571 (m), 1483 (s), 1463 (s), 1445 (vs), 1385
(w), 1315 (s), 1231 (m), 1164 (m), 1115 (m), 1026 (vs), 995 (vs sh),
836 (s), 800 (s), 770 (s), 734 (s), 713 (m), 691 (m), 626 (s), 548 (s), 485
(m), 431 (m) cm−1. Anal. calcd for C18H17Cl2N2NiO2P: C, 47.63;
H, 3.77; N, 6.17. Found: C, 47.98; H, 4.21; N, 5.85%. HRMS :
Mass calcd for C18H17ClN2NiO2P : 417.0064. Found. 417.0070
[Ni(N,P,N)Cl]+.
◦
All catalytic tests were started between 25 and 30 C, and no
cooling of the reactor was done during the reaction. After injection
of the catalytic solution and of the cocatalyst under a constant
low flow of ethylene, the reactor was pressurized to 10 bar. A
temperature increase was observed which resulted solely from the
exothermicity of the reaction. The 10 bar working pressure was
maintained during the experiments through a continuous feed
of ethylene from a reserve bottle placed on a balance to allow
continuous monitoring of the ethylene uptake. At the end of each
test (35 min) a dry ice bath, and in the more exothermic cases also
liquid N2, was used to rapidly cool down the reactor, thus stopping
the reaction. When the inner temperature reached 0 ◦C the ice bath
was removed allowing the temperature to slowly rise to 10 ◦C. The
gaseous phase was then transferred into a 10 L polyethylene tank
filled with water. An aliquot of this gaseous phase was transferred
into a Schlenk flask, previously evacuated, for GC analysis. The
products in the reactor were hydrolyzed in situ by the addition of
ethanol (1 mL), transferred into a Schlenk flask, and separated
from the metal complexes by trap-to-trap evaporation (20 ◦C,
0.8 mbar) into a second Schlenk flask previously immersed in
liquid nitrogen in order to avoid any loss of product.
Synthesis of [Fe{bis(2-picolyl)phenylphosphine}2][Cl3FeOFeCl3]
10. To a solution of 1 (2.40 g, 8.2 mmol) in CH2Cl2 (50 mL) was
added a CH2Cl2 solution of FeCl2·4H2O (1.63 g, 8.2 mmol) and the
mixture was stirred at room temperature overnight. After reaction
the brown precipitate was filtered off and washed with diethyl
ether. Complex 10 was dried for several hours under vacuum
(Yield: 1.70 g, 1.70 mmol, 62%). IR (KBr): 1600 (s), 1471 (vs),
1434 (vs), 1396 (w), 1311 (m), 1155 (m), 1106 (s), 861 (vs), 820 (w),
804 (m), 753 (s), 744 (s sh), 707 (m), 692 (m), 603 (w), 504 (m), 492
(s) cm−1. Anal. calcd for C36H34Cl6Fe3N4OP2: C, 44.08; H, 3.49; N,
5.71. Found: C, 43.96; H, 4.02; N, 6.09%.
Electrochemical measurements
Electrochemical experiments were performed with a three-
electrode system consisting of a platinum working electrode, a
platinum-wire counter electrode, and a silver wire as pseudo-
reference. The redox potentials are reported versus the saturated
calomel electrode after correction using ferrocene. All measure-
ments were carried out under Ar, in degassed and distilled MeCN,
using 0.1 M [N(n-Bu)4]PF6 solutions as the supporting electrolyte.
An EG & G Princeton Applied Research Model 273A potentiostat
connected to a computer (Programme Research Electrochemistry
Software) was used.
Synthesis of [Fe{bis(4,5-dihydro-2-oxazolylmethyl)phenyl-
phosphine}2][Cl3FeOFeCl3] 11. To a solution of 2 (0.43 g,
1.55 mmol) in CH2Cl2 (50 mL) was added a CH2Cl2 solution
of FeCl2·4H2O (0.31 g, 1.55 mmol) and the mixture was stirred
overnight at room temperature. The violet precipitate was filtered
off and washed with diethyl ether. Complex 11 was dried under
vacuum (Yield: 0.28 g, 0.3 mmol, 58%). IR (KBr): 1639 (s), 1481
(w), 1435 (w), 1406 (m), 1373 (m), 1269 (vs), 1171 (m), 1108 (w),
999 (w), 933 (m), 858 (w), 745 (m), 694 (m) cm−1. Anal. calcd for
C28H34Cl6Fe3N4O5P2: C, 35.44; H, 3.61; N, 5.91. Found: C, 35.76;
H, 4.39; N, 5.98%. HRMS: Mass calcd for (C28H34FeN4O4P2)/2 :
304.0697. Found: 346.0673 [Fe(N,P,N)]2+.
Crystal structure determinations
Diffraction data were collected on a Kappa CCD diffractometer
˚
using graphite-monochromated Mo-Karadiation (k = 0.71073 A)
(Table 5). Data were collected using phi-scans and the structures
were solved by direct methods using the SHELX 97 software,48,49
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 2945–2955 | 2953
©

View Article Online
Table 5 X-Ray experimental data for 10·0.5CH2Cl2 and 11·4MeCN
Crystal data
10·0.5CH2Cl2
11·4MeCN
Chemical formula
Mr
C36H34FeN4P2·0.5(CH2Cl2)·Fe2OCl6
C28H34FeN4O4P2·4(C2H3N)·Fe2Cl6O
1023.33
1113.00
Crystal system
Space group
T/K
Orthorhombic
Pccn
173(2)
12.6010(3)
16.6710(5)
20.3170(6)
90.00
4268.0 (2)
4
Monoclinic
C2/c
173(2)
15.3270(7)
21.8930(12)
15.4880(9)
110.368(2)
4872.1 (4)
4
˚
a/A
˚
b/A
˚
c/A
b/◦
3
˚
V/A
Z
Dx/Mg m−3
1.593
Mo Ka
1.55
1.517
Mo Ka
1.32
Radiation type
l/mm−1
Crystal form, colour
Crystal size/mm
F(000)
Prism, orange
0.12 × 0.10 × 0.10
2068
9257, 4894, 3202
0.070
Prism, red
0.14 × 0.12 × 0.10
2272
12328, 7114, 5370
0.033
No. of measd, indep. and obsd reflect.
Rint
◦
hmax
/
27.5
30.0
R[F2 > 2r(F2)], wR(F2), S
0.088, 0.168, 1.15
4894
0.050, 0.090, 1.06
7114
No. of relections
No. of parameters
247
272
and the refinement was by full-matrix least squares on F2.
No absorption correction was used. All non-hydrogen atoms
were refined anisotropically with H atoms introduced as fixed
8 P. Braunstein, M. D. Fryzuk, F. Naud and S. J. Rettig, J. Chem. Soc.,
Dalton Trans., 1999, 589.
9 O. Ku¨hl, P. C. Junk and E. Hey-Hawkins, Z. Anorg. Allg. Chem., 2000,
626, 1591.
10 (a) L. M. Venanzi, J. Chem. Soc. A, 1958, 719; (b) D. F. Evans, J. Chem.
Soc. A, 1959, 2003.
˚
contributors (dC–H = 0.95 A, U11 = 0.04).
11 J. L. Deutsch and S. M. Poling, J. Chem. Educ., 1969, 46, 167.
12 J. Lo¨liger and R. Scheffold, J. Chem. Educ., 1972, 49, 646.
13 S. K. Sur, J. Magn. Reson., 1989, 82, 169.
14 F. Speiser, P. Braunstein, L. Saussine and R. Welter, Inorg. Chem., 2004,
43, 1649.
Acknowledgements
We thank the Centre National de la Recherche Scientifique
(CNRS), the Ministe`re de lꢀEducation Nationale et de la
Recherche (Paris) and the Institut Franc¸ais du Pe´trole (IFP)
for support. We are grateful to the Erasmus-Socrates exchange
programme between the University Duisburg-Essen and ULP
(F. T.). We also thank Dr A. DeCian and Prof. R. Welter (ULP
Strasbourg) for the crystal structure determinations, Dr Jean-Paul
Collin (ULP Strasbourg) for the electrochemical measurements
and M. Mermillon-Fournier (LCC Strasbourg) for technical
assistance.
15 F. Speiser, P. Braunstein, L. Saussine and R. Welter, Organometallics,
2004, 23, 2613.
16 F. Speiser, P. Braunstein and L. Saussine, Organometallics, 2004, 23,
2633.
17 A. Kermagoret, R. Pattacini, P. C. Vasquez, G. Rogez, R. Welter and
P. Braunstein, Angew. Chem., Int. Ed., 2007, 46, 6438.
18 P. Braunstein, F. Naud, A. Pfaltz and S. J. Rettig, Organometallics,
2000, 19, 2676.
19 S. Ameerunisha, J. Schneider, T. Meyer, P. S. Zacharias, E. Bill and G.
Henkel, Chem. Commun., 2000, 2155.
20 B. C. Childs and H. A. Goodwin, Aust. J. Chem., 2001, 54, 685.
21 Y. Chen, C. Qian and J. Sun, Organometallics, 2003, 22, 1231.
22 (a) A. Machkour, D. Mandon, M. Lachkar and R. Welter, Inorg. Chem.,
2004, 43, 1545; (b) F. Li, M. Wang, C. Ma, A. Gao, H. Chen and L.
Sun, Dalton Trans., 2006, 2427–2434.
23 L. P. Wu, Y. Yamagiwa, I. Ino, K. Sugimoto, T. Kuroda-
Sowa, T. Kamikawa and M. Munakata, Polyhedron, 1999, 18,
2047.
24 Y. Nakayama, Y. Baba, H. Yasuda, K. Kawakita and N. Ueyama,
Macromolecules, 2003, 36, 7953.
25 F. Pelascini, M. Wesolek, F. Peruch, A. De Cian, N. Kyritsakas, P. J.
Lutz and J. Kress, Polyhedron, 2004, 23, 3193.
26 A. S. Ionkin, W. J. Marshall, D. J. Adelman, B. Bobik Fones,
B. M. Fish and M. F. Schiffhauer, Organometallics, 2006, 25,
2978.
References
1 For recent reviews, see: (a) G. J. P. Britovsek, V. C. Gibson and D. F.
Wass, Angew. Chem., Int. Ed., 1999, 38, 428; (b) S. D. Ittel, L. K.
Johnson and M. Brookhart, Chem. Rev., 2000, 100, 1169; (c) Late
Transition Metal Polymerization Catalysis, ed. B. Rieger, L. Saunders
Baugh, S. Kacker and S. Striegler, Wiley-VCH, New York, 2003;
(d) V. C. Gibson and S. K. Spitzmesser, Chem. Rev., 2003, 103, 283;
(e) F. Speiser, P. Braunstein and L. Saussine, Acc. Chem. Res., 2005, 38,
784; (f) C. Bianchini, G. Giambastiani, I. G. Rios, G. Mantovani, A.
Meli and A. M. Segarra, Coord. Chem. Rev., 2006, 250, 1391; (g) S. Jie,
S. Zhang, K. Wedeking, W. Zhang, H. Ma, X. Lu, Y. Deng and W.-H.
Sun, C. R. Chim., 2006, 9, 1500.
27 A. A. Danopoulos, N. Tsoureas, J. A. Wright and M. E. Light,
Organometallics, 2004, 23, 166.
28 M. G. B. Drew, V. McKee and S. M. Nelson, J. Chem. Soc., Dalton
Trans., 1978, 80.
29 P. C. Healy, B. W. Skelton and A. H. White, Aust. J. Chem., 1983, 36,
2057.
30 M. James, H. Kawaguchi and K. Tatsumi, Polyhedron, 1997, 16, 4279.
31 D. B. Dell’amico, F. Calderazzo, L. Labella, F. Marchetti and E. G. F.
Quadrelli, Gazz. Chim. Ital., 1997, 127, 103.
2 A. Kermagoret and P. Braunstein, Organometallics, 2008, 27, 88.
3 F. Speiser, P. Braunstein and L. Saussine, Dalton Trans., 2004, 1539.
4 A. Kermagoret and P. Braunstein, Dalton Trans., 2008, 585.
5 P. Braunstein, F. Naud, A. Dedieu, M.-M. Rohmer, A. DeCian and
S. J. Rettig, Organometallics, 2001, 20, 2966.
6 D. Fenske, B. Maczek and K. Maczek, Z. Anorg. Allg. Chem., 1997,
623, 1113.
7 E. Lindner, H. Rauleder and W. Hiller, Z. Naturforsch., B: Anorg.
Chem. Org. Chem., 1983, 38, 417.
2954 | Dalton Trans., 2008, 2945–2955
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
32 K. R. Dunbar, J. J. Longridge, J. M. Rawson, J.-S. Sun, K. S. Hagen
and B. Do, Inorg. Synth., 2002, 33, 103.
33 W. M. Reiff, E. H. Witten, K. Mottle, T. F. Brennan and A. R. Garafalo,
Inorg. Chim. Acta, 1983, 77, L83.
34 H. Weiss and J. Stra¨hle, Z. Naturforsch., B: Anorg. Chem. Org. Chem.,
1984, 39, 1453.
35 N. Kuhn, H. Kotowski, C. Maichle-Moessmer and U. Abram, Z. Anorg.
Allg. Chem., 1998, 624, 1653.
36 M. Bolte, H. W. Lerner and S. Scholz, Acta Crystallogr., Sect. E: Struct.
Rep. Online, 2001, 57, m216.
37 G. J. P. Britovsek, V. C. Gibson, S. K. Spitzmesser, K. P. Tellmann,
A. J. P. White and D. J. Williams, J. Chem. Soc., Dalton Trans., 2002,
1159.
41 A. Cabort, A. Michel, B. Therrien, H. Stoeckli-Evans, K. Bernauer, G.
Suss-Fink, A. F. Williams and G. Stupka, Inorg. Chim. Acta, 2003, 350,
193.
42 V. C. Gibson, R. K. O’Reilly, W. Reed, D. F. Wass, A. J. P. White and
D. J. Williams, Chem. Commun., 2002, 1850.
43 N. Bergem, U. Blindheim, O. T. Onsager and H. Wang, Fr. Pat., Fr
1 519 181, 1968.
44 L. P. Spencer, R. Altwer, P. Wei, L. Gelmini, J. Gauld and D. W. Stephan,
Organometallics, 2003, 22, 3841.
45 W.-H. Sun, X. Tang, T. Gao, B. Wu, W. Zhang and H. Ma,
Organometallics, 2004, 23, 5037.
46 D. M. Dawson, D. A. Walker, M. Thornton-Pett and M. Bochmann,
J. Chem. Soc., Dalton Trans., 2000, 459.
38 K. P. Tellmann, V. C. Gibson, A. J. P. White and D. J. Williams,
Organometallics, 2005, 24, 280.
39 P. Braunstein and F. Naud, Angew. Chem., Int. Ed., 2001, 40,
680.
47 D. S. McGuinness, V. C. Gibson and J. W. Steed, Organometallics, 2004,
23, 6288.
48 Kappa CCD Operation Manual, Nonius BV, Delft, The Netherlands,
1997.
40 P. S. Braterman, J. I. Song and R. D. Peacock, Inorg. Chem., 1992, 31,
49 G. M. Sheldrick, SHELXL97, Program for the refinement of crystal
structures, University of Go¨ttingen, Germany, 1997.
555.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 2945–2955 | 2955
©