NiCl2(1,2-Diiminophosphorane) complexes: A new family of readily accessible and tuneable catalysts for oligomerisation of ethylene

Article abstract of DOI:10.1039/b109992m

1,2-Diiminophosphoranes 1-4 featuring either ethane, benzene, cyclohexane or 1,2-diphenylethane carbon backbones act as tightly bonded 1,4-chelating ligands towards NiCl₂, affording the corresponding paramagnetic complexes 5-8 in high yield. X-Ray diffraction studies performed on compounds 5 and 6 revealed that the conformation of the five-membered metallacycle depends on the rigidity of the carbon backbone. For both complexes, the coordination sphere of the Ni atom is a distorted tetrahedron with bond lengths and angles around nickel similar to those observed for related Ni(II)(α-diimine) complexes. Complexes 5-8 are active for ethylene oligomerisation under mild reaction conditions (0°C, 1.1 bar) upon activation by alkylaluminum derivatives (Et₂AlCl or MAO). The nature of the carbon backbone of the 1,2-diiminophosphorane ligands has a profound impact on the selectivity of the catalytic systems. The selectivity for trimers and higher oligomers varies from 10% (pre-catalyst 8) to 81% (pre-catalyst 5). Effects of varying ethylene pressure, temperature and aluminium co-catalyst/nickel ratios with pre-catalyst 6 are reported. Tailoring the reaction parameters has a modest effect on the oligomer distribution but allows quite high catalytic activities to be achieved with turnover frequencies up to 135 × 10³ h^-1.

Full text of DOI:10.1039/b109992m

View Article Online / Journal Homepage / Table of Contents for this issue
NiCl2(1,2-Diiminophosphorane) complexes: a new family of readily
accessible and tuneable catalysts for oligomerisation of ethylene
Mathieu Sauthier,a Francois Leca,a Roberto Fernando de Souza,*b Katia Bernardo-Gusmao,b
Luiz Fernando Trevisan Queiroz,b Loıc Toupetc and Regis Reau*a
a
´
Organometalliques et Catalyse, Chimie et Electrochimie Moleculaires (CNRS UMR 6509),
Institut de Chimie, CNRS - Universite de Rennes 1, Campus de Beaulieu 35042, Rennes cedex,
´
´
France. E-mail: regis.reau@univ-rennes1.fr
´
Instituto de Quımica, UFRGS, Av. Bento Gonc¸alves 9500, P.O. Box 15003 CEP 91501-970
Porto Alegre, Brazil. E-mail: rfds@if.ufrgs.br
b
c
`
´
´
´
Groupe Matiere Condensee et Materiaux (CNRS UMR 6626), CNRS-Universite de
Rennes 1, 35042, Rennes cedex, France
Received (in Strasbourg, France) 29th October 2001, Accepted 26th December 2001
First published as an Advance Article on the web
1,2-Diiminophosphoranes 1–4 featuring either ethane, benzene, cyclohexane or 1,2-diphenylethane carbon
backbones act as tightly bonded 1,4-chelating ligands towards NiCl2 , affording the corresponding
paramagnetic complexes 5–8 in high yield. X-Ray diffraction studies performed on compounds 5 and 6 revealed
that the conformation of the five-membered metallacycle depends on the rigidity of the carbon backbone. For
both complexes, the coordination sphere of the Ni atom is a distorted tetrahedron with bond lengths and angles
around nickel similar to those observed for related Ni(II)(a-diimine) complexes. Complexes 5–8 are active for
ethylene oligomerisation under mild reaction conditions (0 ꢀC, 1.1 bar) upon activation by alkylaluminum
derivatives (Et2AlCl or MAO). The nature of the carbon backbone of the 1,2-diiminophosphorane ligands has a
profound impact on the selectivity of the catalytic systems. The selectivity for trimers and higher oligomers
varies from 10% (pre-catalyst 8) to 81% (pre-catalyst 5). Effects of varying ethylene pressure, temperature and
aluminium co-catalyst=nickel ratios with pre-catalyst 6 are reported. Tailoring the reaction parameters has a
modest effect on the oligomer distribution but allows quite high catalytic activities to be achieved with turnover
frequencies up to 135 ꢁ 103 hꢂ1
.
Bidentate ligands containing sp2-hybridised nitrogen donor
atoms have attracted much attention in recent years for the
tailoring of transition metal catalysts.1 An important devel-
opment in this field was reported by Brookhart et al. who
showed that a-diimines are efficient ligands for the palladium-
and nickel-catalysed oligomerisation and polymerisation of
ethylene.2 The fact that the steric and electronic properties of
a-diimine ligands can be easily varied has been a key factor in
the optimisation of the catalytic systems.2
Derivatives incorporating ylide moieties have been only very
recently developed as ligands for homogeneous catalysis,3 with
our own recent research focussing on chelating nitrogen
donors featuring iminophosphorane moieties (–N=PR3).4
Iminophosphoranes are readily accessible ylides possessing a
highly polarised P=Nbond. They have found numerous
applications as building blocks in organic synthesis and poly-
mer chemistry.5 Iminophosphoranes coordinate to transition
metals via their approximately sp2-hybridised Natom and
have been incorporated in hetero-6 and homobidentate
ligands. Two types of homobidentate ligands are known,
depending on whether the carbon backbone bridges the two
phosphorus atoms (derivatives A, Scheme 1) or the two
nitrogen atoms (derivatives B, Scheme 1). The coordination
chemistry of type A diiminophosphoranes has been studied in
depth and numerous rhodium, palladium and platinum com-
plexes have been described.7 Nickel(II) complexes bearing
ligands A (Y ¼ CH2 , CH2–CH2) have recently been reported
and it should be noted that these complexes are not active for
the oligomerisation of ethylene.3c The first complexes featuring
Scheme 1 Diiminophosphorane ligands.
type B 1,2-diiminophosphorane ligands date back to 1975,8
but the coordination chemistry of these derivatives has only
very recently been reinvestigated.3d,4 We have shown that, in
spite of their ylide character, type B 1,2-diiminophosphoranes
exhibit a relative ‘‘hardness’’ comparable to that of classical
sp2-hybridised nitrogen ligands, and that their coordination
behaviour is considerably influenced by the nature of the
bridging carbon backbone.4b Furthermore, derivatives B are
expected to exhibit quite different steric and electronic prop-
erties than those of the diiminophosphoranes A. In particular,
they should be more sterically demanding ligands since their
bulky phosphino substituents are close to the metal centre.
630
New J. Chem., 2002, 26, 630–635
DOI: 10.1039/b109992m
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2002

View Article Online
These features prompted us to investigate the use of 1,2-di-
iminophosphoranes 1–4 (Scheme 1) as ligands for the Ni-cat-
alysed oligomerisation of ethylene. Herehein, we report the
synthesis, structural characterisation and catalytic perfor-
mance for ethylene oligomerisation of Ni(II) (1,2-diiminophos-
phorane) complexes. The influence of the ligand backbone and
of reaction conditions (pressure, temperature, nature of alu-
minium cocatalyst) on their catalytic activity and the resultant
product distribution is described.
Results and discussion
Nickel(II)(1,2-diiminophosphorane) complexes
1,2-Diiminophosphoranes 1–4 (Scheme 1) were conveniently
obtained by condensation of commercially available diamines
and Ph3PBr2 , followed by treatment with a strong base (the
Kirsanov reaction).4,5a–c P-Phenyl substituted iminophos-
phoranes were selected for this study since they can be easily
purified and are not air sensitive. The blue–violet complex 5
(Scheme 2) has already been prepared by reacting the 1,2-di-
iminophosphorane 1 with NiCl2 in refluxing CH3CN, followed
by treatment at 120 ꢀC under vacuum.8 As observed for related
a-diimines,2 NiCl2(1,2-dimethoxyethane) reacted at room
temperature in CH2Cl2 with diiminophosphoranes 1–4,
affording the corresponding complexes 5–8 (Scheme 2). Deri-
vatives 5–8 exhibit low solubility in common organic solvents
and, after washing with diethyl ether, are obtained in almost
quantitative yield and high purity. They are paramagnetic,
preventing any NMR spectroscopic analysis. The new com-
plexes 6–8 have been characterised by high-resolution mass
spectrometry and gave satisfactory elemental analyses.
Fig. 1 ORTEP drawing (thermal ellipsoid 50% probability) of
complex 5; the H atoms have been omitted for clarity.
The first monomeric Ni-complex featuring an iminopho-
sphorane ligand [Ph2P(o-C6H4)–N=PMe3] has only been very
recently described6c and Ni-iminophosphorane complexes are
still rare.3c,6e,8,9 Furthermore, few metal complexes bearing
1,2-diiminophosphorane ligands B are known. Solid state
structures have been elucidated for only six complexes pos-
sessing either derivatives 3 (Co,3d Rh,3d Pd4) or 4 (Pd4b) as
ligands. Therefore, Ni(II) complexes 5 and 6 were subjected to
an X-ray diffraction study in order to evaluate the influence of
the carbon backbone rigidity on the coordination behaviour of
1,2-diiminophosphoranes. For both complexes, the Ni atom is
located in a distorted tetrahedral coordination environment
consisting of two Natoms of the 1,2-diiminophosphorane
moieties and two Cl atoms (Fig. 1 and 2, Table 1). The Ni–Cl
bond lengths are typical for this class of compounds, 2c,d,10 and
are almost identical for the two complexes (Table 1). The
geometry around the coordinated nitrogen atoms is essentially
Fig. 2 ORTEP drawing (thermal ellipsoid 50% probability) of
complex 6; the H atoms have been omitted for clarity.
two five-membered metallacycles is very different. The five-
membered metallacycle of complex 5, which contains sp3-
hybridised C atoms, adopts a half-chair conformation [Ni(1)–
N(1)–C(1)–C(2), 34.5ꢀ(8); Ni(1)–N(2)–C(2)–C(1), 35.6ꢀ(6)]
with the two phosphino groups in a mutually trans config-
uration (Fig. 1). Note that this conformation has also been
Table 1 Selected bond lengths and angles for complexes 5 and 6
planar (sum of the angles >357.8ꢀ) and the P–Ndistances
+
5
6
[1.585(5)–1.618(2) A] are normal values for metal-coordinated
iminophosphoranes.3c,d,4 As expected, the conformation of the
Ni(1)–Cl(1)
Ni(1)–Cl(2)
Ni(1)–N(1)
Ni(1)–N(2)
N(1)–C(1)
C(1)–C(2)
C(2)–N(2)
N(1)–P(1)
N(2)–P(2)
2.248(2)
2.258(2)
2.017(5)
1.987(6)
1.481(8)
1.514(10)
1.490(9)
1.585(5)
1.590(6)
2.2811(5)
2.2178(6)
2.0141(16)
1.9960(15)
1.414(2)
1.423(2)
1.413(2)
1.6185(16)
1.6117(16)
Cl(1)–Ni(1)–Cl(2)
N(1)–Ni(1)–Cl(1)
N(1)–Ni(1)–Cl(2)
N(2)–Ni(1)–Cl(1)
N(2)–Ni(2)–Cl(2)
N(2)–Ni(1)–N(2)
Ni(1)–N(1)–C(1)
N(1)–C(1)–C(2)
C(1)–C(2)–N(2)
C(2)–N(2)–Ni(1)
123.23(9)
107.20(17)
112.40(18)
113.01(18)
108.82(17)
86.0(2)
108.7(4)
109.4(6)
109.2(6)
109.3(4)
128.09(2)
104.15(5)
116.55(5)
101.27(5)
113.51(5)
83.60(6)
110.29(11)
125.5(17)
116.23(16)
123.14(13)
Scheme 2 Synthesis of (1,2-diiminophosphorane)NiCl2 complexes.
New J. Chem., 2002, 26, 630–635
631

View Article Online
observed for a Pd(II) complex possessing the 1,2-dipheny-
lethane bridged ligand 4.4b The metallacycle of complex 6,
featuring sp2-carbon atoms, formally shows an envelope con-
formation. The N(1), C(1),C(2) and N(2) atoms lie almost in
the same plane [N(1)–C(1)–C(2)–N(2), 1.3(2)ꢀ] with the Ni
atom being out of this plane [C(1)–C(2)–N(2)–Ni, 10.68(19)ꢀ;
Ni–N(1)–C(1)–C(2), 12.38(19)ꢀ]. However, considering the
small values of the C(1)–C(2)–N(2)–Ni and Ni–N(1)–C(1)–
C(2) dihedral angles, the metallacycle of 6 (Fig. 2) can be
regarded as nearly planar (maximum deviation, 0.06 A). In
+
spite of these structural differences, complexes 5 and 6 share
some important features. Of particular interest, the Ni–N
distances [5, 1.987(6) and 2.017 (5) A; 6, 1.996(2) and 2.014 (2)
+
+
A] and the N(1)–Ni(1)–N(2) bite angles [5, 86.0(2)ꢀ; 6,
Fig. 3 Effect of the carbon backbone of the 1,2-diiminophosphor-
anes on the oligomer distribution.
83.60(6)ꢀ] are comparable for both compounds. It is very
interesting to note that these data are very similar to those
recorded for monomeric (a-diimine)-NiX2 complexes. For
example, (N,N0-di-tert-butylethylenediamine)NiBr2 , which
adopts a distorted tetrahedral geometry with planar coordi-
the ligands has an important impact on the selectivity of the
catalytic systems (Fig. 3). The selectivity in dimers, which are
mainly (E þ Z) internal butenes in all cases, range from 19%
(pre-catalyst 5) to 90% (pre-catalyst 8). Complexes 5–7 afford a
similar amount of trimers (ca. 30%), consisting in a mixture of
hexenes and methyl-2-pentenes, while higher oligomers (>C6)
are the major products with the pre-catalyst 5. The nature of
the carbon backbone influences both the electronic and steric
properties of 1,2-diiminophosphoranes 1–4 and it is difficult to
establish structure-property relationships. However, it is note-
worthy that the less sterically hindered iminophosphorane-
nickel complex 5 gives the highest selectivity (i.e., 54%) for
oligomers incorporating more than 3 monomeric units. Under
similar reaction conditions, we have checked that Ni(acac)2
gives ethylene dimers and bulky a-diimine-nickel systems yield
polymers. These results show that 1,2-diiminophosphoranes
are versatile ligands for the Ni-catalysed ethylene oligomer-
isation; rather high catalytic activities can be achieved and the
selectivity towards dimers, trimers and higher oligomers can be
tailored by varying the nature of the carbon backbone.
The influence of the reaction conditions, including the
addition of phosphine modifiers, on the performance of the
catalytic system has been studied in detail for the most active
pre-catalyst 6. At 0 ꢀC under 1.1 atm of ethylene, the TOF’s
slightly decline for longer reaction times (entries 1 and 2,
Table 3). The catalytic activity decreases as the temperature is
raised from 0 to 30 ꢀC, regardless of the total pressure (entries
1,4 and 3,6, Table 3); this behaviour has been well-documented
and attributed to a probable lowering of ethylene solubility at
higher temperatures.2 The ethylene pressure has a dramatic
effect on the catalytic activity. An increase in the pressure
results in an increase of the TOF from 11 ꢁ 103 (1.1 bar) to
121 ꢁ 103 hꢂ1 (6.0 bar) at 0 ꢀC. This trend is more pronounced
at 30 ꢀC: the same pressure enhancement induces an increase
of the TOF’s from 3 ꢁ 103 to 87 ꢁ 103 hꢂ1 (entries 4 and 6,
Table 3).12
nated Natoms, exhibits Ni–Nbond lengths of 1.996(7) and
+
2.002(8) A while the N–Ni–N angle reaches 82.5(3)ꢀ.10
Catalytic studies
The oligomerisation of ethylene is industrialised on a large
scale and nickel complexes are the most frequently employed
catalysts.11 High performance Ni catalysts that selectively
produce oligomers have been obtained via tailoring of the
surrounding ligands. A major advance in this field arose from
the work of Keim et al.11a–c on anionic P–O ligands that are
used in the prominent Ni-catalysed Shell Higher Olefins Pro-
cess (SHOP) for the synthesis of linear C4–C20 ethylene oli-
gomers. Since 1995, cationic palladium and nickel species
bearing aryl-substituted a-diimines have been developed for
the homologation of olefins.2 The nature of the macro-
molecules obtained depends on reaction conditions and the
ligand structure. Bulky nickel complexes catalyse the poly-
merisation of a-olefins while initiators lacking steric hindrance
oligomerise ethylene to a-olefins.2 It is noteworthy that the
polyethylenes produced by these systems differ from those
made with early metal d0 catalysts by their ‘‘hyperbranched’’
structure.
A convenient method for the generation of catalysts involves
in situ activation of nickel(II) dichloride complexes with an
aluminium co-catalyst.2 The pre-catalysts 5–8 were evaluated
for ethylene homologation in chlorobenzene at 1.1 bar ethy-
lene pressure and 0 ꢀC for 1 h, using Et2AlCl (DEAC) as the
activator (Al=Ni ¼ 70). Under these reaction conditions,
complexes 5–8 are active for ethylene oligomerisation; rela-
tively high turnover frequencies (TOF’s) varying from 8 ꢁ 103
to 11 ꢁ 103 hꢂ1 have been measured (Table 2). The highest
catalytic activity is observed with pre-catalyst 6, bearing a
ligand with an aromatic and rigid carbon backbone.
The most attractive feature of these nickel-iminophos-
phorane catalysts is that the nature of the carbon backbone of
As observed for Ni(II)-a-diimine catalysts,2 the quantity and
nature of aluminium co-catalyst dramatically affects the
Table 2 Ethylene oligomerisation with precatalysts 5–8a
c
c
c
C4
C6
>C6
Pre-catalyst
Yield=g
TOFb=103 hꢂ1
Total (%)
a (%)
Total (%)
Linear (%)
Total (%)
5
6
7
8
7.0
10.0
7.6
8
11
8
19
61
46
90
1
2
1
1
27
31
28
9
23
33
23
59
54
8
26
1
4.5
5
a
Conditions: 1.1 bar of ethylene (constant pressure), 0 ꢀC, 33 mmol of nickel complex, Et2AlCl=Ni ¼ 70, 50 mL of chlorobenzene, 1 h.
b
c
TOF ¼ turnover frequency; mol ethylene converted per mol of Ni catalyst per hour. Weight percent determined by GC analysis.
632
New J. Chem., 2002, 26, 630–635

View Article Online
Table 3 Influence of the reaction parameters on the catalytic behaviour of pre-catalyst 6a
b
C4
b
b
C6
>C6
Aluminum Phosphine
Entry co-catalyst Al=Ni (P=Ni)
TOF=
T=ꢀC P=bar Time=h Yield=g 103 hꢂ1
Total (%) a (%) Total (%) Linear (%) Total (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
Et2AlCl
Et2AlCl
Et2AlCl
Et2AlCl
Et2AlCl
Et2AlCl
Et2AlCl
Et2AlCl
TMA
70
70
70
70
20
70
200
200
200
200
70
—
—
—
—
—
—
—
—
—
—
PPh3 (1)
PCy3 (1)
PCy3 (2)
0
0
0
1.1
1.1
6.0
1.1
6.0
6.0
6.0
11
11
11
1
2
1
1
1
1
1
1
1
1
1
1
1
10.3
13.6
112
3.0
8.0
80
100
125
0.0
17.1
3.4
9.8
8.7
11
7
121
3
61
53
52
70
76
57
52
60
—
77
71
58
78
2
2
14
1
37
23
15
17
—
21
2
31
38
34
30
21
32
36
29
—
18
29
29
19
33
32
53
43
64
56
53
29
—
18
20
12
23
8
9
14
—
3
11
12
11
—
5
—
13
3
30
30
30
30
50
50
50
0
9
87
108
135
0
18
4
MAO
Et2AlCl
Et2AlCl
Et2AlCl
1.1
1.1
1.1
70
70
0
0
10
10
0
9
a
b
Constant ethylene pressure, 33 mmol of nickel complex 6, 50 mL of chlorobenzene. Weight percent determined by GC analysis.
outcome of the oligomerisation reaction. Augmentation of the
Et2AlCl=Ni ratio from 20 : 1 to 70 : 1 gives an increase in the
turnover frequencies from 9 ꢁ 103 to 87 ꢁ 103 hꢂ1 (entries 5
and 6, Table 3). The effect is more modest for further
enhancement of the Al=Ni ratio (entries 6 and 7, Table 3).
Note that the increase in the production of C6 and higher
olefinic fractions is probably due to co-oligomerisation invol-
ving the C4 products accumulated in the reaction vessel (vide
infra). Other alkylaluminium compounds gave rise to catalytic
systems with lower activities. Trimethylaluminum (TMA) is
not an effective co-catalyst. With MAO a TOF of 18 ꢁ 103 hꢂ1
is observed at 11 bar and 50 ꢀC while, for these optimal reac-
tion conditions, a TOF of 135 ꢁ 103 hꢂ1 is recorded with
Et2AlCl (entries 8–10, Table 3).13 It has been observed that the
behaviour of Ni-based catalytic systems can be sensitive to the
addition of phosphine ligands to the reaction media.11f This is
not the case with pre-catalyst 6. A slight decrease of the TOF is
observed upon addition of one equivalent of PPh3 (cf. entries 1
and 11, Table 3) whereas addition of one or two equivalent of
PCy3 has almost no effect on the productivity (cf. entries 1 with
12 and 13, Table 3).
distribution (entries 1,3 and 4,6, Table 3) whereas the pro-
duction of higher oligomers is slightly favoured by lowering
the temperature. For example, under a total pressure of 1.1
bar, pre-catalyst 6 produces 30% trimers and higher oligomers
at 30 ꢀC whereas this proportion reaches 39% at 0 ꢀC (entries 4
and 1, Table 3). The addition of phosphines also has a mar-
ginal effect on the oligomer distribution (entries 1 and 11–13),
suggesting that no coordination of the phosphine to the nickel
centre occurs, probably due to the steric constraints imposed
by the coordinated 1,2-diiminophosphorane. In contrast, the
nature and the quantity of the aluminium co-catalyst do
influence the selectivity. The production of high oligomers is
more efficient with Et2AlCl than with MAO (entries 8 and 10,
Table 3) and is also favoured when the Al : Ni ratio increases
from 20 to 70 (entries 5 and 6, Table 3). These latter data
preclude a chain termination mechanism by transfer to the
alkylaluminum compound.
The C4 fraction contains 1-butene, 2-cis-butene and 2-trans-
butene and in the C6 fraction the products were hexenes and
methylpentenes. Increasing the pressure and lowering the
Al=Ni ratio favour the production of 1-butene; the highest
selectivity (37%) is observed under 6 bar for an Al=Ni ratio of
20 : 1 (entry 5, Table 3). The nature of the C4 products for all
other runs is quite similar. The amount of linear products in
The reaction conditions have a modest influence on the
selectivity of the catalytic system. Variation of ethylene pres-
sure between 1.1 and 6.0 bar does not affect the oligomer
Scheme 3 Proposed mechanism for ethylene oligomerisation.
New J. Chem., 2002, 26, 630–635
633

View Article Online
the C6 fraction increases as the ethylene pressure is raised, this
effect is more pronounced at 0 ꢀC (1.1 bar, 33%; 6.0 bar, 53%,
entries 1 and 3, Table 3) than at 30 ꢀC (1.1 bar, 43%; 6.0 bar,
56%, entries 4 and 6, Table 3).
temperature, to a CH2Cl2 solution (10 mL) of (DME)NiCl2
(0.38 g, 1.72 mmol). The reaction mixture immediately turned
blue and was stirred for 1 h at room temperature. The volatile
materials were removed under vacuum; the residue was washed
with diethyl ether (2 ꢁ 10 mL) and dried under vacuum.
Complex 5 was obtained as a violet solid (1.19 g, yield 98%).
Anal. calcd for C38H34N2P2Cl2Ni: C, 64.26; H, 4.83; N, 3.94.
Found, C, 64.22; H, 4.41; N, 3.81%.
All of these products are expected from a classical oligo-
merisation pathway involving parallel oligomerisation=iso-
merisation reactions (Scheme 3). The mechanism starts with a
hydrido or an alkylnickel species (9 or 11, respectively) being
formed from the dichloro pre-catalyst and the alkylaluminum
co-catalyst. The linear C6 isomers are produced by sequential
coordination and migratory insertion of ethylene into the
nickel-alkyl bond via intermediate 13. Alternatively, prior to
this reaction sequence, the metal alkyl species 12 could undergo
a b-hydride elimination leading to 15, which can give 1-butene
or the intermediate 16 by re-addition. Complex 16 is respon-
sible for the production of internal C4 olefins and branched
alkenes via 17. It should be noted that complex 17 can also be
obtained by coordination and insertion of butenes into the
Ni-alkyl bond of complex 10. The formation of internal C6
alkenes is not presented in Scheme 3 for clarity, but this pro-
cess implies a sequence of b-hydride elimination-insertion steps
starting from complexes 14 and 17. The mechanism presented
in Scheme 3 explains that the amount of linear C6 products
increases when the ethylene pressure is enhanced since a high
quantity of ethylene available in the medium favours the for-
mation of (i) 11 over 17 and (ii) of 13 over 15.
Complex 6. Following the procedure described for complex
5, reaction of diiminophosphorane 2 (0.40 g, 0.63 mmol) and
(DME)NiCl2 (0.14 g, 0.63 mmol) afforded 6 as a red solid (0.41
g, yield 95%). HR-MS (FAB, mNBA): m=z 756.0934 (M)þ
calcd for C42H34N2P2Cl2Ni: 756.0928. Anal. calcd for
C42H34N2P2Cl2Ni: C, 66.53; H, 4.52; N, 3.69. Found, C, 66.21;
H, 4.08; N, 3.75%.
Complex 7. Following the procedure described for complex
5, reaction of a mixture of racemic diiminophosphorane 3
(0.40 g, 0.63 mmol) and (DME)NiCl2 (0.14 g, 0.63 mmol)
afforded 7 as a blue solid (0.24 g, yield 96%). HR-MS (FAB,
mNBA): m=z 727.1712 (M ꢂ Cl)þ, calcd for C42H40N2P2ClNi:
727.1709. Anal. calcd for C42H40N2P2Cl2Ni: C, 66.00; H, 5.28;
N, 3.67. Found, C, 65.76; H, 5.30; N, 3.92%.
Complex 8. Following the procedure described for complex
5, reaction of (1R,2R)-diiminophosphorane 4 (0.20 g, 0.30
mmol) and (DME)NiCl2 (0.07 g, 0.30 mmol) afforded 8 as a
blue solid (0.23 g, yield 95%). HR-MS (FAB, mNBA): m=z
825.1870 (M ꢂ Cl)þ, calcd for C50H42N2P2ClNi : 825.1825.
Anal. calcd for C50H42N2P2Cl2Ni: C, 69.63; H, 4.91; N, 3.25.
Found, C, 69.38; H, 4.60; N, 3.33%.
Conclusion
This study shows that azadiylides are promising N-donor
bidentate ligands for homogeneous catalysis. (1,2-diimino-
phosphorane)NiCl2 complexes possess structural features
similar to a-diimine complexes. Upon activation by Et2AlCl,
they lead to ethylene oligomerisation catalysts exhibiting
rather high catalytic activities under mild reaction conditions.
One of the most interesting features of these new catalytic
systems is that the selectivity is affected by the nature of the
ligand carbon backbone. It is quite surprising that low mole-
cular weight oligomers are obtained in view of the important
steric bulk brought by the triphenylphosphino moieties. Tai-
loring these ligands via modification of the P substituents for
the development of catalytic systems producing heavier oli-
gomers or polymers is under active investigation.
Crystallography
Single crystals of compounds 5 and 6 suitable for a single
crystal X-ray determinations were obtained by slow evapora-
tion from CH2Cl2 solutions at room temperature. The unit cell
constant, space group determination, and the data collection
were carried out on an automatic CAD4 NONIUS (compound
5) or a NONIUS Kappa CCD (compound 6) diffractometer
with graphite monochromated Mo-Ka radiation. The cell
parameters were obtained by fitting a set of 25 high-theta
reflections (compound 5) or with Denzo and Scalepack with 10
frames (psi rotation, 1ꢀ per frame, compound 6). The struc-
tures were solved with SIR-97,14a which reveals the non-
hydrogen atoms of the structure. After anisotropic refinement,
all hydrogen atoms may be found with a Fourier difference.
Compound 6 crystallises with a CH2Cl2 molecule and a
cyclopentane set around the symmetry centre. The entire
structures were refined with SHELXL9714b by full-matrix
least-squares techniques (use of F magnitude; x, y, z, bij for Ni,
P, Cl, C and Natoms, x, y, z in riding mode for the H atoms).
Atomic scattering factors were obtained from the International
Tables for X-ray Crystallography.
Experimental
General consideration
All experiments were performed under an atmosphere of dry
argon using standard Schlenk tube techniques. Solvents were
freshly distilled under argon from sodium–benzophenone
(diethyl ether) or from phosphorus pentoxide (dichloro-
methane). The diiminophosphoranes 1–4 and (DME)NiCl2
were prepared and purified according to literature procedures.
Solids were dried under reduced pressure and chlorobenzene
was distilled over 3 A molecular sieves, immediately prior to
Crystal data for compound 5: C38H34Cl2N2P2NiꢃCH2Cl2 ,
+
M ¼ 795.22, monoclinic, a ¼ 16.921(4), b ¼ 16.112(4), c ¼
+
+
17.597(5) A, b ¼ 115.71(2)ꢀ, U ¼ 4323.5(19) A3, T ¼ 293 K,
use. Diethyl aluminum chloride (DEAC) and trimethylalumi-
num (TMA) were purchased from Aldrich and methylalu-
moxane (MAO) was purchased from Witco, and were used as
received. High-resolution mass spectra were obtained on a
Varian MAT 311 or ZabSpec TOF Micromass at CRMPO,
University of Rennes 1. Elemental analyses were performed by
the Centre de Microanalyse du CNRS at Vernaison, France.
+
space group P2l=c, Z ¼ 4, l(Mo-Ka) ¼ 0.71069 A, Dc ¼ 1.091
Mg mꢂ3, m ¼ 6.70 cmꢂ1, 9418 independent reflections mea-
sured, 4968 observed [I > 2s(F)], 431 variables refined,
R1 ¼ 0.0943, wR2 ¼ 0.2803.
Crystal data for compound 6: C42H34Cl2N2P2NiꢃCH2Cl2 ,
C5H10 , M ¼ 913.19, monoclinic, a ¼ 14.9254(2), b ¼ 15.1776(2),
+
+
c ¼ 18.1566(3) A, b ¼ 92.4895(6)ꢀ, U ¼ 4109.2(1) A3, T ¼ 150
+
K, space group P2l=n, Z ¼ 4, l(Mo-Ka) ¼ 0.71073 A,
Synthesis of nickel complexes
Dc ¼ 1.476 Mg mꢂ3, m ¼ 9.13 cmꢂ1, 9578 independent reflec-
tions measured, 7943 observed [I > 2s(F)], 494 variables
refined, R1 ¼ 0.0367, wR2 ¼ 0.0920.
Complex 5. A CH2Cl2 solution (10 mL) of diiminophos-
phorane (1) (1.00 g, 1.72 mmol) was added, at room
634
New J. Chem., 2002, 26, 630–635

View Article Online
4
5
(a) O. Tardif, L. Toupet and R. Re
´
1998, 661; (b) M. Sauthier, J. Fornie
´
R. Reau, Organometallics, 2000, 19, 553.
au, C. R. Acad. Sci., Ser. IIc,
s-Camer, L. Toupet and
´ ´
CCDC reference numbers 166395 and 182856. See http:==
www.rsc.org=suppdata=nj=b1=b109992m= for crystallographic
data in CIF or other electronic format.
(a) A. W. Johnson, W. C. Kaska, K. A. Stanzewski and D. A.
Dixon, Ylides and Imines of Phosphorus, Wiley-Interscience
Publication, John Wiley and Sons, Inc., New York, 1993; (b) Y. G.
Gololobov and L. F. Kasukhin, Tetrahedron, 1992, 48, 1353;
(c) W. L. Mosby and M. L. Silva, J. Chem. Soc., 1965, 1003;
(d) H. R. Allcock, Science, 1992, 255, 1106.
(a) R. W. Reed, B. D. Santarsiero and R. G. Cavell, Inorg.
Chem., 1996, 35, 4292; (b) C. Y. Liu, D. Y. Chen, M. C. Cheng,
S. M. Peng and S. T. Liu, Organometallics, 1995, 14, 1983; (c) L.
Crociani, F. Tisato, F. Refosco, G. Bandoli, B. Corain and
L. Venanzi, J. Am. Chem. Soc., 1998, 120, 2973; (d) P. Molina,
A. Arques, A. Garcia and M. C. R. de Arellano, Eur. J. Inorg.
Chem., 1998, 1359; (e) W. Keim, R. Appel, A. Storeck, C.
Kruger and R. Goddard, Angew. Chem., Int. Ed. Engl., 1981, 20,
116.
(a) M. W. Avis, K. Vrieze, J. E. Ernsting, C. J. Elsevier, N.
Veldman, A. L. Spek, K. Katti and C. L. Barnes, Organometallics,
1996, 15, 2376; (b) M. W. Avis, K. Vrieze, H. Kooijman,
N. Veldman, A. L. Spek and C. J. Elsevier, Inorg. Chem., 1995, 34,
4092; (c) M. W. Avis, C. J. Elsevier, N. Veldman, H. Kooijman
and A. L. Spek, Inorg. Chem., 1996, 35, 1518; (d) D. M. Hankin,
A. A. Danopoulos, G. Wilkinson, T. K. N. Sweet and M. B.
Hursthouse, J. Chem. Soc., Dalton T rans., 1996, 4063; (e) L. R.
Falvello, S. Fernandez, M. M. Garcia, R. Navarro and E. P.
Urriolabeita, J. Chem. Soc., Dalton T rans., 1998, 3745.
R. Appel and P. Voltz, Z. Anorg. Allg. Chem., 1975, 413, 45.
D. Sellman, F. Geipel and F. W. Heinemann, Eur. J. Inorg. Chem.,
2000, 59.
Oligomerisation reactions
Ethylene oligomerisation reactions have been performed in a
250 mL double-walled stainless steel autoclave equipped with
mechanical stirring, thermocouple and pressure gauge. The
reaction temperature was controlled by a thermostatic bath
circulation. A typical reaction run was performed by intro-
ducing 33 mmol of the nickel(II) complex in 50 mL of chlor-
obenzene. The system was saturated with ethylene and then 1.2
mL of a 1.8 mol Lꢂ1 solution of the alkylaluminum co-catalyst
was added. The ethylene pressure is raised to the desired value
and continuously fed. After 1 h, the reaction was stopped by
adding 3 mL of ethanol and the reactor temperature cooled
down to ꢂ10 ꢀC. Ethylene oligomerisation reactions at atmo-
spheric ethylene pressure have been performed similarly using
a 120 mL double-walled glass reactor magnetically stirred and
continuously fed at 1.1 bar of ethylene. The gas chromato-
graphic analysis of the reaction products has been done on a
Varian 3400CX apparatus with a Petrocol HD capillary col-
umn (methyl silicone, 100 m long, i.d. 0.25 mm and film
thickness of 0.5 mm) working at 36 ꢀC for 15 min and then
heating at 5 ꢀC minꢂ1 up to 250 ꢀC.
6
7
8
9
10 G. B. Jameson, H. R. Oswald and H. R. Beer, J. Am. Chem. Soc.,
1984, 106, 1669.
11 (a) J. Skupinska, Chem. Rev., 1991, 91, 613; (b) W. Keim, Angew.
Chem., Int. Ed. Engl., 1990, 29, 235; (c) W. Keim, A. Behr, B.
Acknowledgements
This work was supported by the ‘‘Ministere de l’Education
Nationale de la Recherche et de la Technologie’’ with doctoral
fellowships for M. S. and F. L., and the CNRS and the CNPq
with an international research program (PICS nꢀ 924).
Limbacker and C. Kruger, Angew. Chem., Int. Ed. Engl., 1983, 22,
¨
503; (d) Y. Chauvin and H. Olivier, in Applied Homogeneous
Catalysis with Organometallic Compounds, eds. B. Cornils and
W. A. Herrmann, Wiley-VCH, New York, 1996, vol. 1, pp. 245–
268; (e) T. R. Younkin, E. F. Connor, J. I. Henderson, S. K.
Friederich, R. H. Grubbs and D. A. Bansleben, Science, 2000,
287, 460; ( f ) B. Bogdanovic, B. Spliethoff and G. Wilke, Angew.
Chem., Int. Ed. Engl., 1980, 19, 622. For catalysts based on cobalt
and iron, see: (g) V. C. Gibson, M. K. P. Tellmann, D. F. Wass,
A. J. P. White and D. J. Williams, Chem. Commun., 2001, 2253;
(h) G. J. P. Britovesk, V. C. Gibson, B. S. Kimberley, G. A. Solan,
A. J. P. White and D. J. Williams, Chem. Commun., 1998, 849;
(i) G. J. P. Britovesk, M. Bruce, V. C. Gibson, B. S. Kimberley,
P. J. Maddox, S. Mastroianni, S. J. McTavish, C. Redshaw, G. A.
References and notes
1
(a) A. Togni and L. M. Venanzi, Angew. Chem., Int. Ed. Engl.,
1994, 33, 497; (b) M. Wills and H. Tye, J. Chem. Soc., Perkin
T rans., 1999, 1109; (c) G. J. P. Britovsek, S. Mastrioanni, G. A.
Solan, S. P. D. Baugh, C. Redshaw, V. C. Gibson, A. J. P. White,
D. J. Williams and M. R. J. Elsegood, Chem. Eur. J., 2000, 6, 2221.
(a) L. K. Johnson, C. M. Killian and M. Brookhart, J. Am. Chem.
Soc., 1995, 117, 6414; (b) C. M. Killian, D. J. Tempel, L. K.
Johnson and M. Brookhart, J. Am. Chem. Soc., 1996, 118, 11 664;
(c) D. P. Gate, S. A. Svejda, E. Onate, C. M. Killian, L. K.
Johnson, P. S. White and M. Brookhart, Macromolecules, 2000,
33, 2320; (d ) S. D. Ittel, L. K. Johnson and M. Brookhart, Chem.
Rev., 2000, 100, 1169; (e) L. H. Shultz and M. Brookhart, Orga-
nometallics, 2001, 20, 3975.
2
Solan, S. Stromberg, A. J. P. White and D. J. Williams, J. Am.
¨
Chem. Soc., 1999, 121, 8728; ( j) B. L. Small, M. Brookhart and
A. Bennet, J. Am. Chem. Soc., 1998, 120, 4049; (k) B. L. Small and
M. Brookhart, J. Am. Chem. Soc., 1998, 120, 7143.
12 Note that the experiments at 1.1 and 6.0 bar are performed in
different reactors (see Experimental), precluding an evaluation of
the order with respect to the ethylene pressure.
13 A diffusional limitation of the TOF is quite likely for this run due
to the high volume (ca. 125 mL) of products accumulated in the
reactor.
14 (a) A. Altomare, M. C. Burla, M. Camalli, G. Cascarano, C.
Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori and
R. Spagna, J. Appl. Crystallogr., 1998, 31, 74; (b) G. M. Sheldrick,
SHELX97-2, Program for the Refinement of Crystal Structures,
3
(a) J. M. Brunel, O. Legrand, S. Reymond and G. Buono, J. Am.
Chem. Soc., 1999, 121, 5807; (b) L. Viau, C. Lepetit, G. Com-
menges and R. Chauvin, Organometallics, 2001, 20, 808; (c) S. Al-
Benna, M. J. Sarsfield, M. Thornton-Pett, D. L. Ormsby, P. J.
Maddox, P. Bres and M. Bochmann, J. Chem. Soc., Dalton T rans.,
2000, 4247; (d ) M. T. Reetz, E. Bohres and R. Goddard, Chem.
Commun., 1998, 935.
University of Gottingen, Germany, 1998.
¨
New J. Chem., 2002, 26, 630–635
635