Oligomerization of styrenes mediated by cationic allyl nickel complexes containing triphenylstibine or triphenylarsine

Article abstract of DOI:10.1039/b818784c

The cationic complexes [Ni(η³-CH₂C(R)CH ₂)(SbPh₃)₃][BAr′₄] (R = CH₃1a, H 1b; Ar′ = 3,5-C₆H₃(CF ₃)₂), [Ni(η³-CH₂C

Full text of DOI:10.1039/b818784c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Oligomerization of styrenes mediated by cationic allyl nickel complexes
containing triphenylstibine or triphenylarsine†
Manuel Jime´nez-Tenorio,*a M. Carmen Puerta,a Isabel Salcedo,a Pedro Valerga,*a Isaac de los R´ıosa and
Kurt Mereiterb
Received 23rd October 2008, Accepted 28th November 2008
First published as an Advance Article on the web 27th January 2009
DOI: 10.1039/b818784c
3
The cationic complexes [Ni(h -CH2C(R)CH2)(SbPh3)3][BAr¢4] (R = CH3 1a, H 1b;
3
Ar¢ = 3,5-C6H3(CF3)2), [Ni(h -CH2C(R)CH2)(AsPh3)2][BAr¢4] (R = CH3 2a, H 2b),
3
[Ni(h -CH2CHCH2)(PPh3)(L)][BAr¢4] (L = SbPh3 3, AsPh3 4), and the neutral derivatives
3
[Ni(h -CH2C(R)CH2)Br(L)] (L = SbPh3, R = CH3 5a, H 5b; L = AsPh3, R = CH3 6a, H 6b) have
been prepared and characterized. The X-ray crystal structures of 1a-b, 2b, 3, 5a and 6b have been
determined. These complexes are very active catalyst precursors for the low-molecular weight
=
oligomerization of RC6H4CH CH2 to mainly dimers and trimers of styrene (R = H) or 4-methylstyrene
(R = CH3). They also catalyse the oligomerization of a-methylstyrene to dimers and trimers,
or to higher oligomers depending upon the reaction conditions (solvent and temperature). The
oligomerization reactions were carried out at 25 ◦C in most cases, in dichloromethane,
1,2-dichloroethane or fluorobenzene, using a olefin/catalyst ratio equal to 2000. The
oligomerization products were characterised by means of GPC/SEC.
Introduction
The use of catalytic systems for olefin polymerization based in
nickel complexes bearing N-, O-, or P-donor ligands is well
established and it is of wide applicability nowadays.1–5 Our research
group has been working in the synthesis and characterization of
allyl nickel complexes bearing bulky phosphine ligands, and in
their applications as catalyst precursors for styrene oligomeriza-
tion and polymerization reactions.6,7 The catalyst precursors are
CH2C(CH3)CH2)(SbPh3)3][BAr¢4] (Ar¢ = 3,5-C6H3(CF3)2) was
isolated and structurally characterised. The smaller cone angle
of SbPh3 compared to AsPh3 or PPh3 might favour the adop-
tion of higher coordination numbers in stibine complexes, but
its driving force is probably reduced electron donation to the
metal centre, which is compensated by binding more ligands.10
in general pseudo-square-planar cationic complexes of the type
3
[Ni(h -CH2C(R)CH2)(P)2]+ (R = H, Me; (P)2 = two monodentate
phosphine ligands or one bidentate phosphine ligand).6,8 Within
this context, we have observed that replacement of P donor ligands
by either Sb or As donor ligands such as SbPh3 or AsPh3 in
cationic methylallyl complexes of nickel or leads to systems with
display a very high catalytic activity, with turnover frequencies
Nt reaching 2052 min-1.9 The products resulting from these nickel-
catalysed styrene oligomerization reactions consists of mixtures of
low molecular weight oligomers, containing a very high proportion
of the head-to-tail dimer of styrene which may represent as much
as 88% of the total amount.
5
Interestingly, the cationic organonickel–stibine complexes [Ni(h -
C5H5)(SbR3)2][BF4] had been reported years ago, but we found no
reports of their applications in catalysis.11 Very recently, the air-
stable complexes trans-[Ni(C6Cl2F3)2L2] (L = SbPh3, AsPh3) have
been synthesized. Their catalytic activity in insertion polymer-
ization of norbornene was tested, showing a strong dependence
of the yield and molecular mass of the polymer on the ligand
used and the solvent.12 Thus, the highest yield and molecular
mass values were obtained using complexes containing SbPh3
as ligand and non-coordinating solvents. This was attributed to
the relative ease of ligand displacement from NiII. In this work
we describe the preparation and structural characterization of
a series of cationic and neutral allyl and 2-methylallyl nickel
complexes bearing SbPh3 or AsPh3, as well as mixed SbPh3/PPh3
and AsPh3/PPh3 ligands. These compounds have shown to be
very active as catalyst precursors for the low molecular weight
oligomerization of styrene, 4-methylstyrene or a-methylstyrene
without the need of a co-catalyst such as poly(methylalumoxane)
(PMAO). The catalytic reactions lead to mixtures of oligomers
containing carbon atom chains with lengths ranging from
The use of SbPh3 as an alternative to PPh3 was found
to be essential in the modification of the catalyst pre-
3
cursor complex. Thus, the novel tris(stibine) complex [Ni(h -
aDepartamento de Ciencia de Materiales e Ingenier´ıa Metalu´rgica y Qu´ımica
Inorga´nica, Facultad de Ciencias, Universidad de Ca´diz, 11510, Puerto
Real, Ca´diz, Spain. E-mail: manuel.tenorio@uca.es, pedro.valerga@uca.es;
Fax: +34 956016288; Tel: +34 956016340
bInstitute of Chemical Technologies and Analytics (E171), Vienna University
of Technology, Getreidemarkt 9, A-1060, Vienna, Austria
† Electronic supplementary information (ESI) available: NMR data of
oligomers and relevant GPC/SEC chromatograms. CCDC reference
numbers 703649–703653. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/b818784c
1842 | Dalton Trans., 2009, 1842–1852
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
C4–C8 to C24–C40 depending upon the substrate, the catalyst and
the reaction conditions. A preliminary account of this research
has already been published.9
2a–b are very similar to those for 1a–b: 125.9 and 116.2 ppm.
There seem to be slight differences in Dd (difference between
1
the chemical shifts of the 13C{ H} NMR resonance of C(2) and
the resonance of the other two allyl carbon atoms C(1)/C(3))
for five–coordinate stibine complexes (Dd(1a) 65.2 ppm; Dd(1b)
57.8 ppm) and for four-coordinate arsine complexes (Dd(2a)
55.8 ppm; Dd(2b) 45.2 ppm). In any case, these values are never as
small as 33–35 ppm, which is the value considered characteristic
Results and discussion
Preparation of the complexes
3
The formally five-coordinate cationic complexes [Ni(h -
3
for [Ni(h -CH2CHCH2)(P)3]+ complexes.15 This means that this
CH2C(R)CH2)(SbPh3)3][BAr¢4] (R = Me 1a, H 1b) were pre-
spectral feature seems to be restricted to five-coordinate allyl nickel
complexes bearing phosphorus-donor ligands.
3
pared by reaction of [{Ni(h -CH2C(R)CH2)(m-Br)}2] with SbPh3
and NaBAr¢4 in fluorobenzene–diethylether, followed by recrys-
tallisation from fluorobenzene–petroleum ether. The analogous
reactions using AsPh3 instead of SbPh3 led to the formally
The X-ray crystal structures of 1b and 2b have been deter-
mined. The structure of 1a is also known and was already
3
reported by us.9 ORTEP views of the cation complexes [Ni(h -
3
four-coordinate derivatives [Ni(h -CH2C(R)CH2)(AsPh3)2][BAr¢4]
3
CH2CHCH2)(SbPh3)3]+ and [Ni(h -CH2CHCH2)(AsPh3)2]+ are
(R = Me 2a, H 2b). Complexes 1a and 2a had been previously
respectively shown in Fig. 1 and 2, together with the most relevant
bond distances and angles.
3
obtained by reaction of [Ni(h -CH2C(CH3)CH2)(COD)][BAr¢4]
with either SbPh3 or AsPh3 in CH2Cl2.
These complexes were isolated in the form of red-purple (1a,
1b) or yellow (2a, 2b) crystalline solids, air-sensitive in solution.
3
A complex formulated as [Ni(h -CH2CHCH2)(AsPh3)2][PF6] has
been briefly mentioned in the literature.13 In the same report,
3
the yellow-orange material obtained by reaction of [Ni(h -
3
CH2CHCH2)2] with SbPh3 and HPF6 was claimed to be [Ni(h -
CH2CHCH2)(SbPh3)2][PF6]. These species were characterised by
means of elemental analysis only, and their catalytic activity
towards the polymerization of butadiene was studied.13,14 However,
we have never observed in our synthetic work the formation
Fig. 1 ORTEP drawing (50% thermal ellipsoids) of the complex cation
3
of yellow-orange four-coordinate species of the type [Ni(h -
3
+
˚
[Ni(h -CH2CHCH2)(SbPh3)3] in 1b. Selected bond lengths (A) and angles
(◦) with estimated standard deviations in parentheses: Ni–C(2) 2.007(2),
Ni–C(3) 2.115(2), Ni–C(1) 2.089(2), Ni–Sb(1) 2.5059(3), Ni–Sb(2)
2.4627(7), Ni–Sb(3) 2.5684(7), C(1)–C(2) 1.407(3), C(2)–C(3) 1.403(3),
Sb(1)–Ni–Sb(2) 102.22(1), Sb(1)–Ni–Sb(3) 106.38(1), Sb(2)–Ni–Sb(3)
105.33(1), C(1)–C(2)–C(3) 117.5(2).
CH2CHCH2)(SbPh3)2]+. Red-purple five-coordinate 1a or 1b was
always obtained, irrespective of the number of equivalents of
SbPh3 used in the preparation. Complexes 1a–b and 2a–b are
diamagnetic, and exhibit in their 1H NMR spectra well-separated
resonances corresponding to the anti and syn protons of the
2-methylallyl or allyl ligands. No coalescence is observed for the
resonances of the anti and syn protons when the temperature is
The crystal structure of 1b consists of separate anion and cation
moieties, and it contains one dichloromethane solvate molecule.
The complex cation in 1b has a pseudo-tetrahedral structure
considering the three Sb atoms and the central carbon atom
of the allyl unit C(2). It results very similar to the structure
adopted by the methylallyl derivative 1a, already reported by us.9
For 1a we remarked the unsymmetrically-bonded 2-methylallyl
◦
raised to 50 C in CDCl3 for any of these derivatives, suggesting
a relatively high energy barrier for the Hsyn/Hanti exchange process
in all these compounds. This contrast with the dynamic behaviour
observed for the recently reported five-coordinate allyl deriva-
3
tives of nickel [Ni(h -CH2CHCH2)(P)3]+ (P = PMe3, P(OMe)3),
interpreted in terms of partial PMe3 or P(OMe)3 dissociation at
ambient temperature, whereas p–s allyl isomerization was found
to be insignificant.15 It appears that a characteristic spectral feature
˚
˚
(Ni–C(1) 2.228(9) A, Ni–C(3) 2.076(5) A). However, for 1b the
allyl ligand appears much more symmetrically arranged (Ni–C(1)
3
1
of [Ni(h -CH2CHCH2)(P)3]+ complexes is that the 13C{ H} NMR
chemical shift for the central carbon atom of the allyl ligand
(C(2)) is shifted to higher field (d(C) ª 90 ppm) compared to
2.089(2) A, Ni–C(3) 2.115(2) A), with a shorter Ni–C(2) separation
˚
˚
˚
of 2.007(2) A. These differences could be simply attributed to the
fact that the X-ray data collection for 1a was measured at 297 K,
whereas for 1b the data collection was carried out at 100 K. All
the Sb–Ni–Sb bond angles lie in the narrow interval 102◦ to 107◦.
The Ni–Sb separations are similar to those found in 1a, being
the value in four-coordinate [Ni(h -CH2CHCH2)(P)2]+ derivatives
3
(d(C) > 105 ppm).15 In the case of complexes 1a–b, the value of
d for the C(2) atoms is respectively 126.8 and 117.9 ppm. These
values are greater than the reported shifts for the resonance of
˚
slightly shorter than the value of 2.4480(6) A reported for Ni–Sb
C(2) in [Ni(h -CH2CHCH2)(P)3]+. Furthermore, the chemical shift
in [Ni(C6Cl2F3)2(SbPh3)2], the only other Ni–SbPh3 complex struc-
turally characterised to date.12 Recently, the solid-state structures
3
values for the resonance of C(2) in the four-coordinate complexes
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 1842–1852 | 1843
©

View Article Online
mixed phosphine–stibine and phosphine–arsine complexes, we
3
used a different approach. We carried out the reactions of [Ni(h -
CH2CHCH2)Br(PPh3)] with one equivalent of SbPh3 or AsPh3
and NaBAr¢4 in fluorobenzene. From these reactions, crystalline
yellow-orange materials were obtained. Their elemental analyses
3
were not entirely satisfactory considering a formulation as [Ni(h -
1
CH2CHCH2)(PPh3)(L)][BAr¢4] (L = SbPh3, AsPh3). The 31P{ H}
NMR spectra of these materials showed two singlet resonances
of similar intensity, suggesting the presence of two different
1
phosphorus-containing species. The H NMR spectra showed in
each case two sets of allyl proton resonances, in the intensity
ratios 3 : 1 for the SbPh3 derivative, and 1 : 1 for the AsPh3
derivative. Upon recrystallisation, a change in the relative intensity
1
1
of the H and 31P{ H} NMR signal sets was observed. A X-ray
crystal structure analysis of the stibine derivative revealed
that these species can be viewed as compounds of composition
3
[Ni(h -CH2CHCH2)(PPh3)x(L)1-x][BAr¢4] (L = SbPh3 3, AsPh3
4) where x= 0.75. In these species, a number of PPh3 ligands
has been replaced by either SbPh3 or AsPh3 molecules with
little structural distortion, making feasible the crystalization.
This can be rationalised considering that these compounds are
Fig. 2 ORTEP drawing (50% thermal ellipsoids) of the complex cation
3
+
˚
[Ni(h -CH2CHCH2)(AsPh3)2] in 2b. Selected bond lengths (A) and angles
(◦) with estimated standard deviations in parentheses: Ni–C(2a) 1.965(6),
Ni–C(3) 2.029(5), Ni–C(1) 2.022(4), Ni–As(1) 2.2955(9), Ni–As(2)
2.3150(10), C(1)–C(2a) 1.386(8), C(2a)–C(3) 1.307(8), As(1)–Ni–As(2)
102.62(3), C(1)–C(2a)–C(3) 125.6(7).
3
salts containing one [Ni(h -CH2CHCH2)(PPh3)(L)]+ (L = SbPh3,
3
AsPh3) cation, one [Ni(h -CH2CHCH2)(PPh3)2]+ cation, and two
3
[BAr¢4]- anions. Hence, 3 and 4 might be also formulated as [Ni(h -
3
3
of a series of five-coordinate allyl derivatives of nickel [Ni(h -
CH2CHCH2)(PPh3)(L)][Ni(h -CH2CHCH2)(PPh3)2][BAr¢4]2 (L =
CH2CHCH2)(triphos)][BPh4] (triphos = PPh(CH2CH2PPh2)2),16
SbPh3 3, AsPh3 4). Fig. 3 shows an ORTEP view of the cation
3
3
and [Ni(h -CH2CHCH2)(P)3][X] (P = PMe3, P(OMe)3; X =
complex [Ni(h -CH2CHCH2)(PPh3)(SbPh3)]+ in compound 3,
CF3SO3 , PF6 , Br-, I-)15 have been reported and studied in detail.
These structures were interpreted in terms of distorted trigonal-
bipyramidal (TBP) or square-pyramidal (SPY) geometries as a
function of the angular parameter t.17 For 1b the value of t is 0.39,
consistent with a heavily distorted TBP geometry for the complex
cation. In contrast with this, the parameter t has a value of 0.09
for 1a, which indicates that the structure can be better interpreted
in terms of a distorted square pyramid, with Sb(3) in the apical
position. The structure of the complex cation 2b is pseudo-square
planar, very similar to those of the structurally-related cationic
together with the most relevant bond distances and angles for
this species.
-
-
3
phosphine complexes [Ni(h -CH2C(CH3)CH2)(PiPr2Ph)2][BPh4]6
3
or [Ni(h -CH2CHCH2)(PPh3)2][BF4],18 but bearing AsPh3 instead
of tertiary phosphines. The Ni–C and Ni–As separations suggest
a fairly symmetrical arrangement of the arsines and the allyl
ligand. However, the bond distance C(1)–C(2a) is significantly
longer than the corresponding to C(2a)–C(3). The Ni–As bond
distances are very similar to the value observed for the complex
[Ni(C6Cl2F3)2(AsPh3)2] (2.3094(7) A),12 but slightly longer that the
˚
˚
value found for the Ni–As bond distance of 2.2552(10) A in the
cluster compound [(h -MeC5H4)3Mo3S4Ni(AsPh3)][p-TolylSO3].19
5
The angle formed by the least squares planes defined by the allyl
ligand and the plane defined by the NiAs2 moiety is of 59.8◦.
We were interested in preparing complexes containing mixed
phosphine–arsine, phosphine–stibine, or arsine–stibine ligands.
Thus, we carried out the reactions of complexes 1a and 1b
with one equivalent of AsPh3, in an attempt to prepare the
mixed arsine–stibine derivatives. However, the only isolable com-
pounds from these reaction mixtures were the corresponding
bis(triphenylarsine) derivatives 2a or 2b. No reaction was observed
between these compounds and SbPh3, even when an excess
amount of stibine was used. It is clear that AsPh3 is far better
ligand than SbPh3 in nickel complexes. For the preparation of
Fig. 3 ORTEP drawing (50% thermal ellipsoids) of the complex cation
3
[Ni(h -CH2CHCH2)(PPh3)(SbPh3)]+ in 3. Selected bond lengths (A) and
˚
angles (◦) with estimated standard deviations in parentheses: Ni–C(2a)
1.962(12), Ni–C(3) 2.045(7), Ni–C(1) 2.010(8), Ni–Sb(1) 2.370(6), Ni–P(1)
2.1915(19), C(1)–C(2a) 1.400(16), C(2a)–C(3) 1.295(14), Sb(1)–Ni–P(1)
102.9(3), C(1)–C(2a)–C(3) 124.5(14).
3
The structure of the cation [Ni(h -CH2CHCH2)(PPh3)(SbPh3)]+
is pseudo-square planar, and almost identical to that of the cation
3
[Ni(h -CH2CHCH2)(PPh3)2]+ also present in the same crystal. The
SbPh3 and PPh3 are disordered, and their respective occupation
factors were refined to the values of 0.46 for Sb, and 0.54 for
1844 | Dalton Trans., 2009, 1842–1852
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
P. The Ni–P and Ni–Sb bond distances were constrained. The
allyl ligand shows two possible orientations, with two alternative
positions for the central carbon atom C(2). It is interesting to
point out the fact that if SbPh3 replaces PPh3 to give these
crystals, then the steric differences between the two ligands might
be not so great. In summation, the systems 3 and 4 contain the
3
bis(phosphine) allyl complex [Ni(h -CH2CHCH2)(PPh3)2]+ plus
3
the corresponding cation [Ni(h -CH2CHCH2)(L)(PPh3)]+ (L =
SbPh3 or AsPh3). In the case of 3, the two cations crystallize
in the same cell in the ratio 1 : 1. Everything points out that the
triphenylarsine derivative 4 behaves similarly. In this fashion, the
1
presence of two sets of allyl resonances in the H NMR spectra,
1
and of two signals in the 31P{ H} NMR spectra can be easily
rationalised.
3
We have isolated the neutral complexes [Ni(h -
CH2C(R)CH2)Br(EPh3)] (R = Me, E = Sb 5a, As 6a; R =
3
H, E = Sb 5b, As 6b) by reaction of [{Ni(h -CH2C(R)CH2)(m-
Fig. 5 ORTEP drawing (50% thermal ellipsoids) of the complex
3
Br)}2] (R = Me, H) with either SbPh3 or AsPh3 in diethylether.
These compounds were obtained in form of dark red crystalline
materials. These species are very air-sensitive in solution, where
they are readily decomposed. This makes difficult the recording of
good quality NMR spectra due to the formation of paramagnetic
impurities. Therefore, great care has to be taken in the mani-
pulation of these compounds with rigurous oxygen exclusion.
The X-ray crystal structures of 5a and 6a have been determined.
˚
[Ni(h -CH2(CH3)CH2)Br(AsPh3)] (6a). Selected bond lengths (A) and
angles (◦) with estimated standard deviations in parentheses: Ni–C(1)
1.992(3), Ni–C(2) 1.996(3), Ni–C(3) 2.022(3), Ni–As 2.3128(6), Ni–Br
2.3302(6), C(1)–C(2) 1.393(5), C(2)–C(3) 1.390(5), C(2)–C(4) 1.516(6),
As–Ni–Br 97.279(18), C(1)–C(2)–C(3) 114.9(3).
well with those obtained respectively for compounds 1a–b and 2b.
The angle As–Ni–Br in 6a has a value of 97.279(18)◦, whereas
the angle Sb–Ni–Br in 5a is 94.92(4)◦. These slight but significant
differences might be attributed to the lesser steric pressure of the
SbPh3 ligand due to its smaller cone angle when compared to
AsPh3. Bromide abstraction from complexes 5a–b and 6a–b using
NaBAr¢4 in fluorobenzene leads to highly reactive cationic species
which were not isolated, but used in situ as catalytic precursors for
the oligomerization of styrenes.
3
ORTEP views of the complexes [Ni(h -CH2C(CH3)CH2)-
3
Br(SbPh3)] and [Ni(h -CH2C(CH3)CH2)Br(AsPh3)] are res-
pectively shown in Fig. 4 and 5, together with the most relevant
bond distances and angles.
Oligomerization reactions
The cationic allyl and methylallyl derivatives 1 to 4 are ex-
tremely active catalytic precursors for the low-molecular weight
oligomerization of styrene, just as we had already reported for
compounds 1a and 2a.9 The neutral derivatives 5a–b or 6a–b
plus one equivalent of NaBAr¢4 in fluorobenzene are also very
effective catalytic systems for styrene oligomerization. In these
systems, highly reactive 14-electron cationic species of the type
3
[Ni(h -CH2C(R)CH2)Ni(L)]+ (R = Me, H; L = SbPh3, AsPh3)20
are generated in situ. The catalytic reactions are carried out in non-
coordinating solvents such as dichloromethane, fluorobenzene or
1,2-dichloroethane. The reactions are in general highly exothermic
and may cause the solvent to boil. The yields of recovered
oligomer mixtures are high, as shown in Table 1. In this Table
we can appreciate that the oligomer mixtures have a very high
content of head-to-tail styrene dimers, which may represent up
to 79% of the total (entry 1). Dimer contents are much higher
when SbPh3-containing catalyst are used (entries 1, 2, 7 and
8). In the case of AsPh3-containing catalysts, the amounts of
styrene dimer are not as high, and there is a clear increase
in the amounts of styrene trimer and higher oligomers (entries
3, 4, 9 and 10). The behaviour of the mixed-ligand complexes
3 and 4 is clearly different. In contrast with the other entries
in Table 1, the resulting oligomers contain little styrene dimer
and trimer, and the main fraction corresponds to oligomers
having Mn values of 700–726 Da (entries 5 and 6). These
Fig.
4
ORTEP drawing (50% thermal ellipsoids) of the com-
3
˚
plex [Ni(h –CH2(CH3)CH2)Br(SbPh3)] (5a). Selected bond lengths (A)
and angles (◦) with estimated standard deviations in parentheses:
Ni–C(2) 1.990(7), Ni–C(1) 2.000(7), Ni–C(3) 2.012(7), Ni–Sb 2.4558(9),
Ni–Br 2.3189(12), C(1)–C(2) 1.404(11), C(2)–C(3) 1.390(11), C(2)–C(4)
1.503(12), Sb–Ni–Br 94.91(4), C(1)–C(2)–C(3) 114.7(8).
The crystal structures of complexes 5a and 6a consist of discrete
molecules. The co-ordination geometry around nickel is pseudo-
square-planar as expected for monomeric 16-electron h -allyl
complexes. The methylallyl ligands appear fairly symmetrically
bonded, with average Ni–C separations close to 2 A in all cases.
The Ni–Sb and Ni–As bond distances have values that compare
3
˚
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 1842–1852 | 1845
©

View Article Online
Table 1 Oligomerization of styrene catalyzed by complexes 1–6
Table 2 Oligomerization of 4-methylstyrene catalyzed by complexes 1–6
Product distributionb (%)
Yielda
Oligomer (GPC data)b
c
Entry Catalyst
Solvent
CH2Cl2
Yielda (%) Mw
Mn
Ip
%
Entry Catalyst
Solvent (%)
Dimer Trimer Oligomers
1
2
1a
1b
95 566
509 1.11 41
246 1.04 59
699 1.03
438 1.02 25
242 1.04 70
1
2
3
4
5
6
7
1a
1b
2a
2b
3
CH2Cl2 100
79
64
53
57
12
12
70
70
42
47
17
23
27
19
17
15
22
25
27
31
4
13
17
24
71c
65d
8
5
24
22
255
719
445
252
CH2Cl2
98
1,2-C2H4Cl2 90
5
CH2Cl2 100
CH2Cl2 100
CH2Cl2 100
3
2a
CH2Cl2
93
6708 4467 1.50 11
4
CH2Cl2
75
100
95
84
98
854
444
258
754
438
254
1199
399
247
1179
396
243
636
403
237
727
435
251
721
397
235
786
417
244
778 1.10 29
436 1.02 30
250 1.03 30
716 1.05 30
431 1.02 34
245 1.04 36
960 1.25 83
393 1.02 11
5a/NaBAr¢4 C6H5F
5b/NaBAr¢4 C6H5F
6a/NaBAr¢4 C6H5F
6b/NaBAr¢4 C6H5F
8
9
10
4
5
2b
3
1,2-C2H4Cl2 93
Experimental conditions: temperature 25 ◦C, solvent (4 mL),
[styrene]/[Ni]= 2000.a After 45 min. b Determined by GPC/SEC. Esti-
mated by area integration of the distribution peaks in the GPC/SEC
chromatogram. c Corresponds to an oligomer of Mw = 830, Mn = 700,
Mw/Mn = 1.19. d Corresponds to an oligomer of Mw = 854, Mn = 726,
Mw/Mn = 1.18.
CH2Cl2
CH2Cl2
70
63
98
98
94
97
242 1.02
944 1.25 82
390 1.01 11
6
6
4
238 1.02
7
7
5a/NaBAr¢4 C6H5F
5b/NaBAr¢4 C6H5F
6a/NaBAr¢4 C6H5F
6b/NaBAr¢4 C6H5F
611 1.04 16
398 1.01 26
228 1.04 58
690 1.05 14
428 1.02 29
242 1.04 57
666 1.08 36
391 1.01 30
229 1.03 34
724 1.09 39
409 1.02 36
236 1.04 25
values are almost identical to those obtained when the catalytic
3
oligomerization of styrene is carried out using the complex [Ni(h -
8
CH2C(CH3)CH2)(PPh3)2]+ as precursor catalyst.21,22 This result is
not surprising, given the fact that compounds 3 and 4 contain
[Ni(h -CH2CHCH2)(PPh3)2]+ cations in their structures, along
3
9
3
with variable amounts of [Ni(h -CH2CHCH2)(PPh3)(L)]+ species
(L = SbPh3 or AsPh3). This indicates that the presence of SbPh3
or AsPh3 has little effect on the catalytic properties of the PPh3-
containing system, which remain essentially unaltered. We have
also studied the catalytic oligomerization of the styrene derivatives
4-methylstyrene and a-methylstyrene, in order to establish the
effect of methyl substituents on the catalytic activity and in the
resulting oligomer distribution. The details of these reactions are
summarized in Tables 2 and 3. The oligomerization reactions of
4-methylstyrene (Table 2) were carried out in 1,2-dichloroethane
in some instances. Although these reactions parallel in general
the behaviour observed with styrene in terms of activity, there
is a clear increase in the oligomerization degree. 4-Methylstyrene
dimers and trimers are always present. The structure of the dimer
corresponds the head-to-tail isomer as in the case of styrene, as
inferred from NMR spectroscopy.
10
Experimental conditions: temperature 25 ◦C, solvent (4 mL), [4-
methylstyrene]/[Ni]= 2000.a After 45 min. b % estimated by area integra-
tion of the distribution peaks in the GPC/SEC chromatogram. c Ip
Mw/Mn (polydispersity index).
=
weight oligomers results more evident in the oligomerization
reactions of a-methylstyrene (Table 3). At variance with reactions
involving styrene and 4-methylstyrene, we have observed that the
characteristics of the oligomers of a-methylstyrene obtained in
the catalytic reactions depend upon the reaction conditions, most
notably reaction time, temperature and solvent used. Besides, not
all the catalyst precursors studied in the present work showed to be
active towards a-methylstyrene. These have not been included in
Table 3. Clearly, the presence of a methyl group at the a-position
of the styrene has a deactivating effect towards the catalytic
effectiveness of most of the compounds studied. Complexes 1a–b
and 2a–b catalyse the very low molecular weight oligomerization
(mainly dimerization and trimerization) of a-methylstyre◦ne when
the reactions are performed in 1,2-dichloroethane at 50 C for a
period of 24 h (Table 3, entries 1, 3, 5 and 6). At variance with this,
when the reactions were carried out in dichloromethane at 25 ◦C,
higher oligomers were obtained, having Mn in the range 1398–2373
Da and with polydispersities Ip in between 1.85 and 3.36. A similar
product distribution was obtained in the reactions using neutral
catalyst precursors plus NaBAr¢4 in fluorobenzene at 25 ◦C. Hence,
the use of dichloromethane or fluorobenzene at room temperature
favours the formation of higher oligomers composed mainly of
carbon chains containing 24 to 40 carbon atoms (C24–C40). In
sharp contrast with this, the use of 1,2-dichoroethane, higher
temperatures and longer reaction times lead to the formation of
In addition to these, there is evidence for the presence of
higher oligomers having Mn in the range 611 to 960. In one
case (Table 2, entry 3), a small amount of a polymer with
Mn = 4467 has been obtained. This suggests that another
different reaction pathways might be also operating here, leading
to higher molecular weight oligomers or even polymers. The
occurrence of two alternative reaction pathways in the styrene
oligo- and polymerization reactions mediated by cationic allyl or
indenyl nickel phosphine complexes has been already reported
in the recent literature.6,7,18 The formation of higher molecular
1846 | Dalton Trans., 2009, 1842–1852
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 3 Oligomerization of a-methylstyrene catalyzed by complexes 1–6
Polymer (GPC data)a
Mw Mn
b
Entry
Catalyst
1a
Solvent
T/◦C
50
Time
24 h
Yield (%)
100
Ip
%
1
2
1,2-C2H4Cl2
CH2Cl2
357
343
1.04
1.02
1.91
1.08
1.02
1.02
1.01
1.04
3.36
1.05
1.02
1.03
1.04
1.76
1.04
1.03
1.85
1.07
1.03
2.93
73
217
3944
434
213
2062
403
27
68
21
6
1b
25
45 min
90
210
524
207
513
3
1b
1,2-C2H4Cl2
50
24 h
100
3
353
221
349
212
20
73
100
77
33
16
83
69
15
16
65
19
10
100
4
5
2a
2a
CH2Cl2
1,2-C2H4Cl2
25
50
45 min
24 h
57
72
4696
363
214
1398
345
211
6
7
2b
1,2-C2H4Cl2
C6H5F
50
25
24 h
100
58
405
231
395
221
5a/NaBAr¢4
45 min
3400
471
235
1934
451
228
8
5b/NaBAr¢4
C6H5F
25
45 min
56
4385
474
221
2373
441
214
9
6b/NaBAr¢4
C6H5F
25
16 h
60
6453
2204
Experimental conditions: solvent (4 mL), [a-methylstyrene]/[Ni] = 2000. a % estimated by area integration of the distribution peaks in the GPC/SEC
chromatogram. b Ip = Mw/Mn (polydispersity index).
dimers and trimers of a-methylstyrene. It is important to note
here that the structure of the dimer of a-methylstyrene is not
homologous to that of the dimers of styrene or 4-methylstyrene
obtained with the same catalyst precursors. NMR data suggest
that in this case the dimer can be formally viewed as the product
of insertion of one a-methylstyrene into one of the C–H bonds of
the methyl substituent at the a-position:
volves the concomitant evolution of hydrogen as an alternative to
the insertion step proposed in Scheme 1.
One of the key steps in these catalytic processes is the gen-
eration of the highly reactive 14-electron species of the type
3
[Ni(h -CH2C(R)CH2)(EPh3)]+ (R = Me, H; E = Sb, As). These
are generated directly by bromide abstraction from the neutral
complexes 5a–b or 6a–b, or by dissociation from the respective
cationic tris(stibine) or bis(arsine) complexes. Since the bis(arsine)
complexes 2a–b are coordinatively unsaturated 16-electron species,
the dissociation of AsPh3 does not seem in principle imperative
for their participation in the catalytic cycle. However, we have
noticed that addition of AsPh3 to the catalytic reaction mixture
has a noticeable effect on the activity. The yields of isolated
oligomers drop drastically when the amount of added AsPh3 is
increased. With 5 equivalents of AsPh3 per equivalent of catalyst
precursor, the catalytic activity is totally suppressed. Given the fact
that the 2a–b do not add a third AsPh3 molecule to furnish the
corresponding tris(arsine), we assume that the addition of AsPh3
produces a shift in the initial ligand dissociation equilibrium. In
this manner, the formation of the 14-electron species is inhibited
and the catalytic cycle can not be initiated. The addition of SbPh3
to 1a–b has a similar effect on their catalytic activity, this being
also suppressed upon the addition of 5 equivalents of SbPh3.
The observed oligomer distribution can be explained in almost
all cases by the reaction sequence involving insertion of the olefin
into the Ni–H bond and subsequent chain growth, followed
by a b-elimination as shown in Scheme 1. Alternatively, for
a-methylstyrene the insertion would occur into the Ni–C of the
The insertion mechanism proposed for the oligomerization
of styrenes mediated by cationic allyl-phosphine complexes7,8,21
can be easily adapted to explain the oligomerization reactions
using allylnickel complexes bearing SbPh3 or AsPh3 ligands. Such
mechanism involves addition of styrene to a Ni–H bond to yield
a styryl complex (proposed to be bound to the nickel center in an
3
h -benzylic fashion22) which by multiple insertion and subsequent
b-hydride elimination gives rise to head-to-tail polystyrene chains
containing methyl and unsaturated end-groups (Scheme 1).
In the case of a-methylstyrene, the cycle in Scheme 1 has to
be modified in order to account for the observed differences in
the structure of the resulting products. A reasonable explanation
consistent with the experimental observations would be to con-
3
3
sider an alternative structure for the h -styryl complex. It seems
symmetric h -styryl intermediate (Scheme 2). A fast b-elimination
very likely that in the particular case of a-methylstyrene, the
h -styryl intermediate would adopt a symmetric structure as
shown in Scheme 2, which does not require the direct involvement
favours low-molecular weight oligomers (dimers and trimers).
If the chain growth reactions are faster, then higher molecular
weight oligomers should be expected. The balance of the rates
of these reactions is dependant on the nature of the substrate,
judging from our results, but also on the temperature at which
3
of the bonds in the aromatic cycle. The formation of such a species,
3
2
namely [Ni(h -CH2C(C6H5)CH2)(EPh3)(h -CH2C(CH3)Ph)]+, in-
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 1842–1852 | 1847
©

View Article Online
Scheme 1
the reactions are performed. Since higher temperatures favour
the formation of dimers and trimers this indicates that the b-
elimination is faster and hence the controlling step in the process
is the chain growth reaction. Opposite to this, lower temperatures
favour the formation of higher oligomers, indicative of a increase
of the rate of the chain growth reactions relative to the b-
elimination step. Both the insertion and the chain growth reactions
are expected to have negative values of the activation entropy
DS‡, whereas positive values are expected for the b-elimination
reaction. Assuming positive values of DH‡, this means that the
variation of the relative reaction rates with the temperature is just
as expected. This is fully consistent with the observed dependence
of the oligomerization degree with the temperature in our case. The
occurrence of another alternative oligomerization mechanism (i.e.
cationic oligomerization) can not be ruled out in principle, but it
does not seem strictly necessary to invoke it in order to explain
satisfactorily the observed oligomer distribution in these catalytic
reactions.
Conclusions
Triphenylstibine and triphenylarsine constitute interesting alter-
natives to the use of more traditional tertiary phosphine ligands in
cationic allylnickel complexes with applications in catalytic styrene
oligomerization reactions. With SbPh3, formally five-coordinate
3
species of the type [Ni(h -CH2C(R)CH2)(SbPh3)3]+ (R = Me, H)
are formed, in contrast with the formally square-planar species
3
[Ni(h -CH2C(R)CH2)(L)2]+ (R = Me, H; L = AsPh3, PPh3).
The allylnickel complexes containing SbPh3 or AsPh3 as co-
ligands are extremely efficient catalyst precursors for the very low
1848 | Dalton Trans., 2009, 1842–1852
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
(10 mm; 7.8 ¥ 300 mm). The oligomer samples were eluted with
THF at a flow rate of 1 cm3 min-1. The system was calibrated using
a Merck polystyrene standard kit.
Syntheses
[Ni(g3-CH2C(R)CH2)(SbPh3)3][BAr¢4] (R = Me 1a, H 1b). To
a slurry of [Ni(COD)2] (0.22 g. 0.8 mmol) in diethyl ether
(15 mL) cooled to -80 ◦C in a ethanol/liquid N2 bath, either
3-bromo-2-methypropene (83.2 mL, 0.8 mmol) or allyl bromide
(69.9 mL, 0.8 mmol) was added. The mixture was warmed at
room temperature and stirred for 45 min. During this time the
slurry changed from yellow to red. The solvent was removed
in vacuo, and the solids extracted with fluorobenzene. To this
red solution SbPh3 (0.85 g. 2.4 mmol) and NaBAr¢4 (0.7 g.
0.8 mmol) was added. A color change from red to dark purple-
brown was observed. The solution was filtered through Celite,
concentrated to 5 mL and petroleum ether (15 mL) was added.
The dark purplish-red microcrystalline precipitate was filtered off,
washed with petroleum ether, and dried in vacuo. Both com-
plexes were recrystallized from fluorobenzene–petroleum ether
(1a) or dichloromethane–petroleum ether (1b, containing one
dichloromethane solvate molecule). Data for 1a: yield: 1.14 g,
70%. Calcd. for C90H64BF24NiSb3: C, 53.1; H, 3.17%. Found:
C, 52.9; H, 3.06%. NMR: 1H (CD2Cl2, 298 K) d 2.04 (s,
3 H, CH2C(CH3)CH2); 2.94 (s br, 2 H, CHantiC(CH3)CHanti);
4.31 (m, 2 H, CHsynC(CH3)CHsyn); 7.29, 7.34 (m, 45 H, C6H5).
Scheme 2
molecular weight oligomerization of styrenes. In general terms,
the use of SbPh3 as co-ligand leads to a decrease in the degree
of styrene oligomerization (mostly dimers) in comparison with
similar systems with PPh3. The degree of styrene oligomerization
when AsPh3-containing complexes are used as catalytic precursors
can be considered intermediate between those obtained with
SbPh3 and PPh3-containing complexes. The nature of the resulting
oligomerization products is also dependant on the substituents
present on the styrene, as well as the reaction conditions. For
styrene and 4-methylstyrene, an insertion mechanism analogous
to that proposed for analogue phosphine-containing systems
has been suggested. The fast b-elimination reaction step favours
dimers and trimers over higher oligomers. Apparently, this can
be reversed as a function of the reaction temperature. In the case
of a-methylstyrene the reaction pathway is slightly modified. The
1
13C{ H} (CDCl3, 298 K) d 23.9 (s, CH2C(CH3)CH2); 65.2
(s, CH2C(CH3)CH2); 117.9 (s, CH2C(CH3)CH2); 129.8, 130.3,
134.0, 135.2 (C6H5). Data for 1b: yield: 1.28 g, 79%. Anal.
Calcd for C90H64BCl2F24NiSb3: C, 51.3; H, 3.06%. Found: C,
51.6; H, 3.10%. 1H NMR (CD2Cl2, 298 K): d 2.92 (d, 2H,
3
3JHH = 12.8 Hz, CHsynHantiCHCHsynHanti), 4.30 (d, 2H, JHH
=
6.4 Hz, CHsynHantiCHCHsynHanti), 5.71 (psq, 1H, JHH = 6.4 Hz,
3
CHsynHantiCHCHsynHanti), 7.14, 7.26 (m, C6H5). 13C{ H} NMR
1
(CD2Cl2, 298 K): d 60.1 (s, CH2CHCH2),117.9 (s, CH2CHCH2),
130.1, 130.9, 134.1, 135.2 (C6H5).
3
participation of a symmetric h -styryl intermediate is proposed in
[Ni(g3-CH2C(R)CH2)(AsPh3)2][BAr¢4] (R = Me 2a, H 2b). To
a slurry of [Ni(COD)2] (0.22 g. 0.8 mmol) in diethyl ether
(15 mL) cooled to -80 ◦C in a ethanol/liquid N2 bath, either
3-bromo-2-methypropene (83.2 mL, 0.8 mmol) or allyl bromide
(69.9 mL, 0.8 mmol) was added. The mixture was warmed at room
temperature and stirred for 45 min. During this time the slurry
changed from yellow to red. The solvent was removed in vacuo,
and the solids extracted with fluorobenzene. To this red solution
AsPh3 (0.51 g. 1.6 mmol) and NaBAr¢4 (0.7 g. 0.8 mmol) was
added. A color change from red to yellowish brown was observed.
The solution was filtered through Celite, concentrated to 5 mL
and petroleum ether (15 mL) was added. The resulting yellow-
orange microcrystalline precipitate was filtered off, washed with
petroleum ether, and dried in vacuo. Both complexes were recrys-
tallized from fluorobenzene–petroleum ether. Data for 2a: yield:
0.87 g, 68%. Anal. Calcd for C72H49As2BF24Ni: C, 54.4; H, 3.11%.
order to rationalise the structures observed for the oligomerization
products of a-methylstyrene.
Experimental
All synthetic operations were performed under a dry dinitrogen
or argon atmosphere following conventional Schlenk techniques.
Tetrahydrofuran, diethyl ether and petroleum ether (boiling point
range 40–60 ◦C) were obtained oxygen- and water-free from
an Innovative Technology, Inc. solvent purification apparatus.
Toluene and fluorobenzene were of anhydrous quality and used as
received. All solvents were deoxygenated immediately before use.
The complex [Ni(COD)2]was obtained according to the literature.8
Styrene was purified by treatment with CaH2 followed by trap-
to-trap distillation. NMR spectra were taken on a Varian Inova
400 MHz or Varian Gemini 300 MHz equipment. Chemical shifts
1
1
are given in ppm from SiMe4 (1H and 13C{ H}), or 85% H3PO4
Found: C, 54.1; H, 3.15%. H NMR (CD2Cl2, 298 K): d 2.01 (s,
(31P{ H}). Microanalysis was performed on an elemental analyzer
3H, CH2C(CH3)CH2), 2.80 (s, 2H, CHsynHantiC(CH3)CHsynHanti),
3.69 (s, 2H, CHsynHanti C(CH3)CHsynHanti), 7.20, 7.28 (m, C6H5).
1
model LECO CHNS-932 at the Servicio Central de Ciencia y
Tecnolog´ıa, Universidad de Ca´diz. Molecular weight distributions
of oligomers were determined at 25 ◦C by GPC/SEC using Merck-
Lacrom equipment fitted with one column Polystyragel PS40
1
13C{ H} NMR (CD2Cl2, 298 K): d 23.13 (s, CH2C(CH3)CH2),
70.07 (s, CH2C(CH3)CH2), 125.9 (s, CH2C(CH3)CH2), 129.64,
131.15, 131.84, 132.66 (C6H5). Data for 2b: yield: 1 g, 78%.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 1842–1852 | 1849
©

View Article Online
Anal. Calcd for C71H47As2BF24Ni: C, 54.1; H, 3.01%. Found:
most often formed during the reaction. Then the solution was
concentrated to 5 mL and petroleum ether (15 mL) was added.
The red microcrystalline precipitate was filtered off, and washed
with petroleum ether, and dried in vacuo. Both complexes were
recrystallized from fluorobenzene–petroleum ether. Data for 5a:
yield: 0.27 g, 62%. Calcd. for C22H22BrNiSb: C, 48.3; H, 4.06%.
1
C, 54.3; H, 3.04%. H NMR (CD2Cl2, 298 K): d 2.92 (d, 2H,
3
3JHH = 14.8 Hz, CHsynHantiCHCHsynHanti), 4.11 (d, 2H, JHH
=
7.6 Hz, CHsynHantiCHCHsynHanti), 5.66 (psq, 1H, JHH = 7.2 Hz,
3
CHsynHantiCHCHsynHanti), 7.22, 7.31 (m, C6H5). 13C{ H} NMR
1
(CD2Cl2, 298 K): d 70.97 (s, CH2CHCH2),116.2 (s, CH2CHCH2),
129.97, 131.38, 132.55, 133.25 (C6H5).
1
Found: C, 48.2; H, 3.98%. NMR: H (C6D6, 298 K) d 1.71 (s,
3 H, CH2C(CH3)CH2); 2.79 (s br, 2 H, CHantiC(CH3)CHanti); 4.31
(s br, 2 H, CHsynC(CH3)CHsyn); 7.01, 7.45 (m, 15 H, C6H5).
[Ni(g3 -CH2CHCH2 )(PPh3 )(L)][Ni(g3 -CH2CHCH2 )(PPh3 )2]-
3
[BAr¢4]2 (L = SbPh3 3, AsPh3 4). To a solution of [Ni(h -
1
13C{ H} (C6D6, 298 K) d 22.6 (s, CH2C(CH3)CH2); 57.2 (s,
CH2CHCH2)Br(PPh3)] (0.25 g. 0.57 mmol) in fluorobenzene
(15 mL) either SbPh3 (0.220 g. 0.57 mmol) or AsPh3 (0.17 g.
0.57 mmol) and NaBAr¢4 (0.51 g. ca. 0.57 mmol) were added.
The mixture was stirred for 45 min at room temperature.
During this time the solution changed from red to dark red.
Then, the solution was concentrated to 5 mL and petroleum
ether (15 mL) was added. The resulting yellow-orange micro-
crystalline precipitate was filtered off, washed with petroleum
ether, and dried in vacuo. Both complexes were recrystallized
from fluorobenzene–petroleum ether. Data for 3: yield: 0.66 g,
75%. Calcd. for C144H94B2F48Ni2P3Sb: C, 55.97; H, 3.07%.
CH2C(CH3)CH2); 129.0 (s, CH2C(CH3)CH2); 129.2, 136.7 (C6H5).
Data for 5b: yield: 0.25 g, 59%. Calcd. for C21H20BrNiSb: C, 47.4;
H, 3.78%. Found: C, 47.1; H, 3.64%. NMR: 1H (CD2Cl2, 298 K)
3
d 2.36 (s br, 2 H, CHantiCHCHanti); 3.51 (d, JHH = 12 Hz, 2 H,
CHsynCHCHsyn); 5.42 (m, 1 H, CH2CHCH2); 7.40, 7.55 (m, 15 H,
1
C6H5).13C{ H} (CDCl3, 298 K) d 58.4 (s, CH2CHCH2); 106.7 (s,
CH2CHCH2); 129.4, 129.8, 134.7, 136.4 (C6H5).
[Ni(g3-CH2C(R)CH2)Br(AsPh3)] (R = Me 6a, H 6b). These
compounds were prepared by a identical procedure to that for
5a-b, using AsPh3 (0.31 g, 1 mmol) instead of SbPh3. Data for 6a:
yield: 0.23 g, 58%. Calcd. for C22H22AsBrNi: C, 52.9; H, 4.44%.
3
Found: C, 55.8; H, 3.05%. NMR: Selected data for [Ni(h -
1
1
CH2CHCH2)(PPh3)(SbPh3)]+ cation: H NMR (CD2Cl2, 298 K):
Found: C, 52.6; H, 4.24%. NMR: H (C6D6, 298 K) d 1.87 (s,
3 H, CH2C(CH3)CH2); 2.32 (s br, 2 H, CHantiC(CH3)CHanti);
3.25 (s br, 2 H, CHsynC(CH3)CHsyn); 7.22, 7.78 (m, 15 H,
3
d 2.93 (d, 2H, JHH = 14.4 Hz, CHsynHantiCHCHsynHanti), 4.23
3
(d, 2H, JHH = 6.2 Hz, CHsynHantiCHCHsynHanti), 5.56 (m, 1H,
1
C6H5). 13C{ H} (C6D6, 298 K): d 22.7 (s, CH2C(CH3)CH2); 61.8
1
CHsynHantiCHCHsynHanti), 7.15–7.32 (m, C6H5). 31P{ H} NMR
1
(CD2Cl2, 298 K): d 27.66. 13C{ H} NMR (CD2Cl2, 298 K): d 70.63
(s, CH2C(CH3)CH2); 125.9 (s, CH2C(CH3)CH2); 129.4, 130.3,
134.6, 136.4 (C6H5). Data for 5b: yield: 0.23 g, 60%. Calcd. for
C21H20AsBrNi: C, 51.9; H, 4.15%. Found: C, 51.6; H, 4.08%.
NMR: 1H (tetrahydrofuran-d8, 298 K) d 2.40 (d, 3JHH = 11.2 Hz,
2 H, CHantiCHCHanti); 3.39 (s br, 2 H, CHsynCHCHsyn); 5.44
(s, CH2CHCH2), 115.29 (s, CH2CHCH2), 130,10, 130.79, 132.19,
3
135.84 (C6H5). Selected data for [Ni(h -CH2CHCH2)(PPh3)2)]+
cation:1H NMR (CD2Cl2, 298 K): d 2.78 (d, 2H, 3JHH = 13.8 Hz,
CHsynHantiCHCHsynHanti), 3.86 (br, 2H, CHsynHantiCHCHsynHanti),
5.60 (m, 1H, CHsynHantiCHCHsynHanti), 7.22–7.32 (m, C6H5).
1
(m, 1 H, CH2CHCH2); 7.37, 7.58 (m, 15 H, C6H5).13C{ H}
1
1
31P{ H} NMR (CD2Cl2, 298 K): d 26.36. 13C{ H} NMR (CD2Cl2,
298 K): d 74.78 (s, CH2CHCH2), 116.98 (s, CH2CHCH2),
129.51, 129.66, 131.79, 133.54, 133.69 (C6H5). Data for 4: yield:
0.69 g, 80%. Calcd. for C144H94AsB2F48Ni2P3: C, 56.84; H, 3.11%.
(tetrahydrofuran-d8, 298 K) d 61.3 (s, CH2CHCH2); 108.7 (s,
CH2CHCH2); 129.4, 130.3, 134.6, 136.4 (C6H5).
X-Ray structure determinations
3
Found: C, 56.5; H, 3.09%. NMR: Selected data for [Ni(h -
Crystal data and experimental details are given in Table 4. X-Ray
diffraction data for compound 1b were collected on a Bruker AXS
SMART Platform 3-circle diffractometer with CCD area detector
at the Institute of Chemical Technologies and Analytics, Vienna
University of Technology and for compounds 2b, 3, 5a and 6a
on a Bruker SMART APEX 3-circle diffractometer with CCD
area detector at The Servicio Central de Ciencia y Tecnolog´ıa
de la Universidad de Ca´diz, using graphite-monochromated Mo-
CH2CHCH2)(PPh3)(AsPh3)]+ cation: 1H NMR (CD2Cl2, 298 K):
d 2.91 (d, 2H, JHH = 13.8 Hz, CHsynHantiCHCHsynHanti), 4.11
3
(d, 2H, JHH = 7.7 Hz, CHsynHantiCHCHsynHanti), 5.66 (m, 1H,
3
CHsynHantiCHCHsynHanti), 7.14–7.45 (m, C6H5). 31P{ H} NMR
1
(CD2Cl2, 298 K): d 26.87. 13C{ H} NMR (CD2Cl2, 298 K): d 72.65
1
(s, CH2CHCH2), 117.19 (s, CH2CHCH2), 129.87, 131.13, 132.02,
133.30 (C6H5). Selected data for [Ni(h -CH2CHCH2)(PPh3)2)]+
3
cation:1H NMR (CD2Cl2, 298 K): d 2.78 (d, 2H, 3JHH = 13.8 Hz,
CHsynHantiCHCHsynHanti), 3.86 (br, 2H, CHsynHantiCHCHsynHanti),
5.60 (m, 1H, CHsynHantiCHCHsynHanti), 7.14–7.45 (m, C6H5).
˚
Ka radiation (l = 0.71073 A). Hemispheres of the reciprocal
space were measured by omega scan frames with d(w) 0.30◦.
Correction for absorption and crystal decay (insignificant) were
applied by semi-empirical method from equivalents using pro-
gram SADABS.23 The structures were solved by direct methods,
completed by subsequent difference Fourier synthesis and refined
on F2 by full matrix least-squares procedures using the program
SHELXTL.24 All non-hydrogen atoms except some in disordered
groups were anisotropically refined. The hydrogen atoms were
placed at geometric positions and refined using the riding model.
In compound 3 the main disorder was found in the cation,
being approximately in 1 : 1 ratio bis(phosphine) or phosphine
and stibine ligands. Complementary P and Sb atoms were refined
to model this disorder. To avoid an averaged bond distance Ru–P
and Ru–Sb bond lengths were constrained to reasonable chemical
1
1
31P{ H} NMR (CD2Cl2, 298 K): d 26.37. 13C{ H} NMR (CD2Cl2,
298 K): d 74.76 (s, CH2CHCH2), 116.79 (s, CH2CHCH2), 129.51,
129.66, 131.79, 133.54, 133.69 (C6H5).
[Ni(g3-CH2C(R)CH2)Br(SbPh3)] (R = Me 5a, H 5b). To a
slurry of [Ni(COD)2] (0.22 g. 0.8 mmol) in diethyl ether (15 mL)
cooled to -80 ◦C a ethanol/liquid N2 bath, either 3-bromo-2-
methypropene (83.2 mL, 0.8 mmol) or allybromide (69.9 mL,
0.8 mmol) was added. The mixture was warmed at room tem-
perature and stirred for 45 min. Then the solvent was removed
under vacuum. The residue was extracted with fluorobenzene.
To this red solution SbPh3 (0.35 g. 1 mmol) was added. The
solution was filtered through Celite, in order to remove Ni0,
1850 | Dalton Trans., 2009, 1842–1852
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 4 Summary of crystallographic data for 1b, 2b, 3, 5a and 6a
Compound
1b
2b
3
5a
6a
Formula
FW
C89.86H63.71BCl1.71F24NiSb3 C71H47As2BF24Ni
C71H47BF24NiP1.54Sb0.46 C22H22BrNiSb
C22H22AsBrNi
499.94
2094.95
1575.45
1529.56
546.77
T/K
100(2)
100(2)
100(2)
293(2)
293(2)
Crystal size/mm
Crystal system
Space group
Cell parameters
0.65 ¥ 0.55 ¥ 0.45
0.55 ¥ 0.12 ¥ 0.11
0.47 ¥ 0.08 ¥ 0.07
0.45 ¥ 0.40 ¥ 0.17
Monoclinic
P21/c (no. 14)
0.35 ¥ 0.30 ¥ 0.27
Monoclinic
P21/c (no. 14)
Triclinic
Triclinic
Triclinic
¯
¯
¯
P1 (no. 2)
P1 (no. 2)
P1 (no. 2)
˚
˚
˚
˚
˚
˚
˚
˚
˚
˚
˚
a = 13.2682(6) A
a = 13.038(3) A
a = 13.093(3) A
a = 12.160(2) A
a = 12.021(2) A
˚
˚
b = 17.6246(7) A
b = 14.663(3) A
b = 14.560(3) A
b = 10.1466(17) A
b = 9.9974(18) A
˚
˚
c = 19.2441(8) A
c = 17.637(4) A
c = 17.565(4) A
c = 17.556(4) A
c = 17.404(3) A
a = 68.634(1)◦
b = 84.927(1)◦
g = 87.387(1)◦
4174.0(3)
2
a = 74.84(3)◦
b = 81.73(3)◦
g = 88.15(3)◦
3220.4(11)
2
a = 75.09(3)◦
b = 81.42(3)◦
g = 88.28(3)◦
3199.4(13)
2
b = 97.557(8)◦
b = 98.246(3)◦
3
˚
Volume/A
2147.3(7)
4
1.691
3.997
1072
2070.0(6)
4
1.604
4.455
1000
Z
rc/g cm-3
1.677
1.347
2068
0.845–1.000
1.625
1.434
1576
0.736–1.000
1.588
0.647
1537.3
0.908–1.000
m(Mo-Ka)/mm-1
F(000)
Max. and min. transmission
factors
Theta range for data
collection
0.649–1.000
0.762–1.000
2.13◦ < q < 30.00◦
1.21◦ < q < 25.03◦
1.45◦ < q < 23.26◦
1.69◦ < q < 25.00◦ 2.36◦ < q < 24.98◦
Reflections collected
Unique reflections
No. of observed reflections
(I > 2sI)
62783
22658
19259
19720 15210
24062 (Rint = 0.0191)
11163 (Rint = 0.0339) 9068 (Rint = 0.0483)
3771 (Rint = 0.0468) 3645 (Rint = 0.0254)
21741
10142
7736
3516
3209
No. of parameters
1132
923
904
230
228
Final R1, wR2 values (I > 2sI) 0.0304, 0.0737
0.0527, 0.1072
0.0591, 0.1135
+0.675, -0.442
0.0772, 0.1448
0.0937, 0.1535
+1.421, -0.817
0.0503, 0.1051
0.0550, 0.1075
+0.725, -1.053
0.0315/0.0821
0.0368/0.0849
+0.714, -0.744
Final R1, wR2 values (all data) 0.0347, 0.0759
Residual electron density
+1.490, -0.913
-3
˚
peaks/e A
values. The allyl ligand was found disordered in compounds 2b and
3. So as to model this disorder, in both cases the central carbon
atom in the allyl ligand was split in two positions being only refined
anisotropically at the part with site occupation factor > 0.5. In
compounds 1b, 2b and 3 some CF3 groups showed orientation
disorder and were refined with 3 anisotropic major and 3 isotropic
minor sites (no geometric restraints) with complementary site
occupation factors. For compound 1b dichloromethane solvate
was refined in population parameter. The program ORTEP-325
was used for plotting.
in the mixture were established by gel permeation chromatography
(GPC).
General procedure using neutral catalyst precursors. A Schlenk
tube was loaded with 0.02 mmol of the neutral catalyst precursor
(5a–b or 6a–b) and 0.02 mmol of NaBAr¢4. Then, 40 mmol of either
styrene, 4-methylstyrene or a-methylstyrene were added, and after
that, 4 mL of fluorobenzene were added to the vigorously stirred
mixture. Normally, an exothermic reaction takes place in most
cases, but not so violent as with the cationic catalyst precursors.
After stirring for 45 min at 25 ◦C (exceptionally 16 h for the
reaction of 6b with a-methylstyrene), MeOH (5 ml) was added
and the mixture exposed to air. The work-up for the isolation and
characterization of the oligomers was as described above.
Oligomerization reactions
General procedure using cationic catalyst precursors.
A
In all cases, the oligomerization reactions were contrasted
against a blank consisting of a reaction mixture identical to the
system under study, but with no catalyst added.
Schlenk tube was loaded with 0.01 mmol of the cationic catalyst
precursor (1a–b, 2a–b, 3 or 4) and the appropriate solvent
(dichloromethane or 1,2-dichloroethane, 4 mL). Then, 20 mmol
of either styrene, 4-methylstyrene or a-methylstyrene were added.
Normally, a violent, exothermic reaction takes place in a few
seconds, causing the mixture to boil. After stirring for 45 min,
MeOH (5 ml) was added and the mixture exposed to air. Some
reactions with a-methylstyrene were carried out at 50 ◦C for a
period of 24 h. These were also quenched by addition of MeOH
(5 mL) and exposed to air. The volatiles were removed in vacuo.
Then, the residue was extracted with dichloromethane and filtered
through a plug of silica gel. The silica gel was washed with several
portions of dichloromethane, and the washings combined with
the filtrate. Removal of the solvent using a rotary evaporator
first, and a vacuum pump afterwards afforded an oily mixture
of styrene oligomers. The ratios of the different oligomers present
Acknowledgements
We thank the Ministerio de Ciencia y Tecnolog´ıa (DGICYT,
Project CTQ2007 60137/BQU) and Junta de Andaluc´ıa (PAI
FQM188, FQM 00094) for financial support.
References
1 F. Speiser, P. Braunstein and L. Saussine, Acc. Chem. Res., 2005, 38,
784–793.
2 V. C. Gibson and S. K. Spitzmesser, Chem. Rev., 2003, 103, 283–315.
3 S. D. Ittel, L. K. Johnson and M. Brookhart, Chem. Rev., 2000, 100,
1169–1203.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 1842–1852 | 1851
©

View Article Online
4 (a) J. Heinicke, M. Koesling, R. Bru¨ll, W. Keim and H. Pritzkow,
Eur. J. Inorg. Chem., 2000, 299–305; (b) J. Heinicke, M. Z. He, A.
Dal, H. F. Klein, O. Hetche, W. Keim, U. Florke and H. J. Haupt,
Eur. J. Inorg. Chem., 2000, 431–440.
15 A. Rufinska, R. Goddard, C. Weidenthaler, M. Buhl and K.-R.
Po¨rschke, Organometallics, 2006, 25, 2308–2330.
16 W. Clegg, G. Cropper, R. A. Henderson, C. Strong and B. Parkinson,
Organometallics, 2001, 20, 2579–2582.
17 The parameter t is defined as (a - b)/60, where a is the largest and b
is the second largest L–Ni–L angle in a five-coordinate structure. See:
A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn and G. C. Verschoor,
J. Chem. Soc., Dalton Trans., 1984, 1349–1356.
18 D. Gareau, C. Sui-Seng, L. F. Groux, F. Brisse and D. Zargarian,
Organometallics, 2005, 24, 4003–4013.
5 W. Liu, J. M. Malinoski and M. Brookhart, Organometallics, 2002, 21,
2836–2838.
6 M. Jime´nez-Tenorio, M. C. Puerta, I. Salcedo, P. Valerga, S. I. Costa,
L. C. Silva and P. T. Gomes, Organometallics, 2004, 23, 3139–3146.
7 I. Hyder, M. Jime´nez-Tenorio, M. C. Puerta and P. Valerga, Dalton
Trans., 2007, 3000–3009.
8 M. Jime´nez-Tenorio, M. C. Puerta, I. Salcedo and P. Valerga, J. Chem.
Soc., Dalton Trans., 2001, 653–657.
9 M. Jime´nez-Tenorio, M. C. Puerta, I. Salcedo, P. Valerga, S. I. Costa,
P. T. Gomes and K. Mereiter, Chem. Commun., 2003, 1168–1169.
10 (a) W. Levason and G. Reid, Coord. Chem. Rev., 2006, 250, 2565–2594;
(b) N. R. Champness and W. Levason, Coord. Chem. Rev.., 1994, 133,
115–217.
11 N. Kuhn and M. Winter, J. Organomet. Chem., 1983, 243, C83–C86.
12 J. A. Casares, P. Espinet, J. M. Mart´ın-Alvarez, J. M. Mart´ınez-Ilarduya
and G. Salas, Eur. J. Inorg. Chem., 2005, 3825–3831.
13 N. Stransky, R. Herzschuh, J.-P. Gehrke and R. Taube, J. Organomet.
Chem., 1984, 270, 357–363.
19 K. Herbst, P. Zanello, M. Corsini, N. D’Amelio, L. Dahlenburg and
M. Brorson, Inorg.Chem., 2003, 42, 974–981.
3
20 For a discussion on 14-electron cationic species of the type [Ni(h -
CH2C(R)CH2)Ni(L)]+ see: D. Alberti, R. Goddard and K.-R. Porschke,
Organometallics, 2005, 24, 3907–3915.
21 J. R. Ascenso, A. R. Dias, P. T. Gomes, C. C. Roma˜o, I. Tkatchenko,
A. Revillon and Q.-T. Pham, Macromolecules, 1996, 29, 4172–4179.
22 J. R. Ascenso, A. R. Dias, P. T. Gomes, C. C. Roma˜o, Q.-T. Pham, D.
Neibecker and I. Tkatchenko, Macromolecules, 1989, 22, 998–1000.
23 G. M. Sheldrick, SADABS, University of Go¨ttingen, Germany, 2001.
24 G. M. Sheldrick, SHELXTL version 6.10, Crystal Structure Analysis
Package, Bruker AXS, Madison, WI, 2000.
14 R. Taube, J.-P. Gehrke and U. Schmidt, J. Organomet. Chem., 1985,
292, 287–296.
25 L. J. Farruggia, ORTEP-3 for Windows, version 1.076, J. Appl.
Crystallogr., 1997, 30, 565.
1852 | Dalton Trans., 2009, 1842–1852
This journal is
The Royal Society of Chemistry 2009
©