Transition metal complexes of triphosphorus macrocycles: A new class of homogeneous olefin polymerisation catalysts

Article abstract of DOI:10.1039/b203002k

Some early transition metal (Ti, V, Cr) complexes of the triphosphorus macrocycle 1,5,9-triethyl-1,5,9-triphospha-cyclododecane, [12]aneP₃Et₃, and related derivatives [12]aneP₃R₃ (R = Ph, C₃H₆

Full text of DOI:10.1039/b203002k

View Article Online / Journal Homepage / Table of Contents for this issue
Transition metal complexes of triphosphorus macrocycles:
A new class of homogeneous olefin polymerisation catalysts
Robert J. Baker and Peter G. Edwards*
Department of Chemistry, Cardiff University, P.O. Box 912, Cardiff, UK CF10 3TB.
E-mail: EdwardsPG@Cardiff.ac.uk
Received 25th March 2002, Accepted 11th June 2002
First published as an Advance Article on the web 5th July 2002
Some early transition metal (Ti, V, Cr) complexes of the triphosphorus macrocycle 1,5,9-triethyl-1,5,9-triphospha-
cyclododecane, [12]aneP3Et3, and related derivatives [12]aneP3R3 (R = Ph, C3H6OMe) have been examined for their
reactivity towards ethene and propene. All complexes of the type [12]aneP3Et3MCl3 display moderate catalytic
activity in the homogeneous polymerisation of ethene when combined with an alkyl aluminium co-catalyst to give
very high molecular weight polymers. Preliminary activation of chromium complexes with alkylating agents was
also studied but these were inactive in the absence of a co-catalyst. Substitution of the P-borne alkyl group with a
pendant ether function was found to switch the catalytic activity of chromium() complexes from polymerisation
to oligomerisation. A nickel halide complex of [12]aneP3Et3 was also found to be active in alkene polymerisation.
phosphine complexes, especially of relatively electropositive
Introduction
metals. Of particular interest is the potential for stabilisation
of relatively robust reaction intermediates, by exploiting the
advantages arising from the macrocyclic ligand structure, which
may support or enhance reactivity for which tertiary phosphine
complexes are not commonly noted. With this in mind, we have
investigated the coordination chemistry of the title ligand
([12]aneP3R3) with the early transition metals.10 In this paper,
we report the catalytic activity of these compounds with respect
to the polymerisation of ethene and α-alkenes. To the best of
our knowledge these are the first compounds of the early trans-
ition metals that contain only phosphine ligands and halides
that polymerise (as opposed to oligomerise) ethene. The few
complexes that have been reported contain π-bonded ligands in
addition to the phosphine.1d
During the last two decades there have been tremendous
advances in the field of catalyst design for alkene polymeris-
ation.1 Since Ziegler and Natta’s discovery that titanium
complexes polymerise ethene2 and propene3 in the presence of
an alkyl aluminium initiator, and the subsequent expansion of
this to the metallocene type catalysts,1a there has been a drive
towards more controlled polymerisation. The result being that
ansa-metallocenes1b and “constrained geometry” catalysts1c
can now polymerise propene in a stereoregular fashion and
display high enough activity to be commercially viable. In
addition to the common metallocenes (mainly Group 4 and 5),
there is a class of heterogeneous catalysts based on chromium;
the Phillips4 and Union Carbide5 systems. These are notable in
so far as they do not require a co-catalyst, and approximately
one third of commercial poly(ethene) is made via the Phillips
method. However, unlike the metallocenes, both the chromium-
based catalysts are heterogeneous and therefore much less
is known about the mechanism of these reactions and the
nature of the active species is still uncertain. Theopold and
co-workers1d have investigated model systems that are homo-
geneous and Gibson et al. have reported a highly active system
based on a bis(imido)Cr() alkyl cation.6 There are still
however, only a handful of well defined homogeneous catalysts
based on chromium. The last few years have seen a dramatic
increase in the logical design of catalyst systems that do not
contain a cyclopentadienyl ligand,1c which also show some of
the highest activities reported for metals other than titanium or
zirconium. This strategy of catalyst design has recently been
extended to iron7 (which had previously shown no polymeris-
ation potential) and aluminium.8
Results and discussion
Addition of a co-catalyst to a solution of [12]aneP3Et3MCl3
(M = Ti, V, Cr) gives rise to low to moderately active polymeris-
ation catalysts. The relevant polymerisation data is shown in
Tables 1 and 2.
Titanium
The addition of MAO (methylaluminoxane) to a solution of
[12]aneP3Et3TiCl3 (1) in CH2Cl2 gives rise to a colour change
from pale blue to light yellow. On addition of monomer, the
system shows some catalytic activity. The reaction of
[12]aneP3Et3TiCl3 (1) with excess Et3Al gives rise to yellow,
diamagnetic solutions (δ 31P{1H} = Ϫ1.29 ppm). The reaction
of this material with 1-hexene (in CDCl3) was followed by
NMR spectroscopy (31P, 1H, 13C). After 1 hour the NMR spec-
trum still shows the presence of monomer and there is a small
broad peak at δ = Ϫ26.6 ppm in the 31P{1H} NMR spectrum
(presumably the metal complex with a growing alkyl chain).
After approx. 2 hours the yellow solution turns deep red in
colour, the peak at δ = Ϫ26.6 ppm disappears from the 31P
NMR spectrum and no monomer is seen in the 1H NMR spec-
trum. The GC-MS results of the 1-hexene reaction (run 12)
show a large number of branched alkanes (C9 and C12 pre-
dominate) which implies that more than one mechanism is
We have previously described the synthesis of the first sym-
metrical triphosphorus macrocyclic ligand in its uncomplexed
state.9 This can be compared to cyclopentadienyl ligands in that
it is a six electron tridentate donor, albeit neutral. These proper-
ties should influence the properties of its complexes (e.g. low
oxidation states should be accessible), and the ligand will be
facially capping in an octahedral environment, forcing the
remaining coordination sites to be mutually cis. In addition, the
macrocyclic coordination effect should help limit phosphine
dissociation, which is prevalent for related acyclic and tripodal
2960
J. Chem. Soc., Dalton Trans., 2002, 2960–2965
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b203002k

View Article Online
Table 1 Polymerisation data for early transition metal complexes
Activity/
Activityd/
Run
Catalysta(mmol)
Monomer
Solventb
Pressure/bar
Initiatorc
Yield/g
gP mmolϪ1 hϪ1
gP mmolϪ1 hϪ1 barϪ1
1
2
LTiCl3 (1)
(0.025)
LTiCl3 (1)
(0.076)
LVCl3 (2)
(0.100)
LCrCl3 (3)
(0.086)
Ethene
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
PhMe
3
6
MAO
(88 : 1)
Et3Al
(78 : 1)
MAO
(30 : 1)
MAO
(92 : 1)
Et3Al
(12 : 1)
MAO
(96 : 1)
Et3Al
(116 : 1)
MAO
(20 : 1)
MAO
(40 : 1)
MAO
(18 : 1)
Et3Al
0.707
0.423
5.421
1.80
32.57
5.56
53.90
20.91
42.60
13.17
8.13
6.51
—
10.85
1.85
17.97
6.97
14.20
4.39
0.81
2.16
—
Ethene
3
Ethene
3
4
Ethene
3
5
LCrCl3 (3)
(0.129)
Ethene
3
5.50
6
L1CrCl3 (5)
(0.821)
Ethene
3
1.081
0.280
0.84
7
LCrCl3 (3)
Ethene
10
3
(0.034)
LCrR3
e
8
Ethene
PhMe
(0.179)
LTiCl3 (1)
(0.076)
LVCl3 (2)
(0.251)
LCrCl3 (3)
(0.086)
LTiCl3 (1)
(0.076)
LCrCl3 (3)
(0.116)
9
Propene
Propene
Propene
1-Hexene
1-Hexene
Norbornene
Ethene
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
3
None
None
0.30
10
11
12
13
14
15
16
17
3
—
—
3
3.48
—
1.16
—
(19 : 1)
Et3Al
—
—
—
3
None
0.845
0.271
0.021
None
0.765
(61 : 1)
MAO
(25 : 1)
MAO
(29 : 1)
Et3Al
(10 : 1)
Et3Al
(2 : 1)
7.27
2.74
—
—
LCrCl3 (3)
(0.099)
—
L2CrCl3 (4)
(0.038)
–
L
Ethene
3
—
—
(0.015)
[LNiBr]Br (6)
(0.17)
Ethene
3
MAO
(300 : 1)
8.85
2.95
a L = [12]aneP3Et3; L1 = [12]aneP3Ph3; L2 = [12]aneP3C3H6OMe. b All reactions carried out in 50 cm3 solvent at 25 ЊC except run 7 (temp. = 70 ЊC)
c Expressed as the ratio Al : Cr. d For comparison with other reported systems see ref. 1e. e The formula designation refers to prior activation of 3 with
alkylating agents rather than an established stoichiometry, R = CH2SiMe3.
Table 2 Properties of soluble polymers
catalytically and none that react with propene. Of the systems
13
reported, (butadiene)2Ti(dmpe)12 and Cp2Ti(PMe3)2 di- and
Run
Mw
Mn
PDI (Mw/Mn)
tri-merise ethene; (butadiene)2Ti(dppe)5 is reported to have
an activity of 0.040 gPE mmolϪ1 hϪ1 and there is a report
of [(Me)2Ti(dmpe)2]14 polymerising ethene at 25ЊC and 4 atm,
but no information on yields and activities was given
(dmpe = 1,2-bis(dimethylphosphino)ethane; dppe = 1,2-bis-
(diphenylphosphino)ethane). The macrocycle complex 1 gives
much higher activities than for any of these previously reported
systems.
3
4
5
6
8
1025000
552000
10400
160500
187500
32200
2190
1615
520
32
252
6.4
310
68.7
2730
involved, and therefore chain transfer to aluminium cannot be
ruled out.11 After work-up of the catalyst solution only a small
amount (<0.1 g) of solid polymer is seen. The complex
[12]aneP3Et3TiCl3 also reacts with ethene (run 1). The polymer
was isolated by removal of the solvent in vacuo but characteris-
ation of the white, waxy solid was hampered by its insolubility
(insoluble in tetrachlorobenzene at 200 ЊC), which implies that
the polymer is cross-linked, highly branched or, in comparison
to the other systems studied, of exceedingly high molecular
weight.
In order to address possible macrocycle ligand transfer from
Ti to Al, treatment of triethyl aluminium with one equivalent
of macrocycle was found to yield an air sensitive oil that is
soluble halocarbons and hydrocarbons. Comparison of NMR
data suggest that this species (31P{1H} δ = Ϫ23.7 ppm) is not
formed in the Ti based catalysis. This aluminium macrocycle
species was also inactive as a polymerisation catalyst under
comparable conditions (run 16). Discrete Ti or Al complexes
could not be isolated from these reaction mixtures and the
nature of the diamagnetic species remains unclear.
Vanadium
Addition of MAO to a solution of [12]aneP3Et3VCl3 (2) in
CH2Cl2 results in a colour change from dark green to red–
brown. Addition of alkylating agents (RLi, R = CH2SiMe3, Me)
to 2 also results in a similar colour change although vanadium
alkyl complexes could not be isolated from these solutions; by
comparison this red colour indicates the formation of V–C
bonds in the reaction with MAO. This system is the most active
of the complexes reported here, although less so than other
non-CpϪ containing catalysts reported recently.15 It is interest-
ing that the vanadium species is more active than titanium,
whereas for the metallocenes the Group 4 metals display
much higher activity. In a recent theoretical study of early trans-
ition metals bearing nitrogen ligands it was found that metal
systems with a d3 configuration showed the most catalytic
potential; the alkene binding energy decreases with increasing d
electron count. This indicates that, for nitrogen ligands, that
chromium() should have higher activity than vanadium().16
A recent investigation using (triazacyclohexane)CrCl3 does
show very high activities.17
Few titanium phosphine complexes in the ϩ (or lower)
oxidation state have been reported which react with ethene
J. Chem. Soc., Dalton Trans., 2002, 2960–2965
2961

View Article Online
The polymer produced from the vanadium catalyst is a
high molecular weight polyethene (Mw ca. 1 million) but the
molecular weight distribution appears to be bimodal. The high
polydispersity indicates that there is more than one catalytic site
or that there are also chain transfer reactions. The polymer
includes material that corresponds to molecular weights of
greater than 10 million, but this is probably cross-linked as
opposed to linear polyethene. Since the polymer is not soluble
in C6H2Cl4 even at 120 ЊC, 13C NMR spectroscopy could not be
used to determine the degree of branching.
the rate of propagation is approximately equal to the rate of
termination, and the larger molecular weight polymers are
formed over a longer period of time. This could also explain the
large polydispersity values seen for some of the polymers.
The 13C NMR data for the soluble polyethenes generated
from runs 4 and 5 show that the samples are completely linear,
with little evidence of branching; this is in agreement with the
presence of straight chain alkanes from the GC-MS data.
In contrast to the chromium complex reported here, the
complex mer-PPPCrCl3 (where PPP is the acyclic, linear tri-
n
The reaction with propene gives no polymer on workup of
the solutions. GC-MS analysis indicates that no oligomers are
formed either.
phosphine, PrP(C2H4PEt2)2) oligomerises ethene to 1-hexene
with 96% selectivity.20 The difference in polymerisation vs.
oligomerisation behaviour between our facially capping phos-
phine compared to the meridonally coordinating phosphine
may be due to the lack of axial steric protection from the linear
phosphine which has been shown to be of importance in
regulating the molecular weight of polymers produced from
late transition metals.21 The only other catalytically active
chromium phosphine complexes are the pendant phosphine
substituted cyclopentadienyl complexes (η1:η5-R2P(CH2)n-
C5H4CrCl2, n = 2, 3 and indenyl analogues) studied by Jolly
et al.22 and the Cr() and Cr() complexes, [Cp*Cr(CH3)-
(dmpe)][PF6] and [Cp*Cr(CH3)(dmpe)],23 however these dmpe
complexes are of low activity and it is believed propagation
requires the dissociation of one of the chelating phosphines to
form η1-dmpe containing intermediates.
Chromium
Addition of MAO or Et3Al to a solution of [12]aneP3Et3CrCl3
(3) in CH2Cl2 results in a colour change from deep blue to dark
red. Attempts to isolate the product from the alkyl aluminium
reaction yields a red, toluene soluble oil that slowly decomposes
at Ϫ30 ЊC for R = Me, for R = Et a red oil that is thermally
stable to 20 ЊC can be obtained. It has been reported that MAO
is a much more effective alkylating agent and forms cationic
species that may be more stable than when using trimethyl
aluminium.18 These relative thermal stabilities could however
account for the fact that the initiation with Et3Al gives a more
active catalyst compared to the MAO initiation (runs 4 and 5,
Table 1). The activity decreases in toluene at elevated temper-
ature (run 7), presumably due to decomposition of the active
species. On addition of ethene a further colour change is seen
and from the resultant brown solution, white polyethene slowly
precipitates (run 4). If this reaction mixture is left for a few
hours, the rate of ethene uptake is reduced to zero and the
solution returns to the original red colour. When more MAO is
added, ethene uptake resumes and more polymer precipitates.
This process may be repeated many times with no apparent
loss of activity. Attempts to identify any complex from this red
reaction mixture have been unsuccessful, the product
decomposing on workup. It is reasonable to propose that this
species contains a chromium centre with at least one alkyl
chain. GC-MS analysis of an aliquot of the catalyst solution
activated by MAO shows low concentrations of C2n ϩ 1 alkanes
(3 ≤ n ≥ 14) and trace amounts of C2n alkenes (4 ≤ n ≥ 14), as
shown in Fig. 1. It is interesting to note that a similar product
distribution is not observed for Et3Al (Fig. 3).
The polymers produced by the different initiators are
markedly different in their physical properties (Fig. 2, Table 2).
Fig. 2 Comparison of the molecular weight distributions of polymers
produced from the Cr catalyst (3) with different alkyl aluminium
activators and with the V analogue (2).
When MAO is used as an initiator (run 4) the resulting polymer
appears to be of higher molecular weight (and of higher poly-
dispersity) than the corresponding reaction with Et3Al (run 5).
This polymer (run 5) is formed as a mixture, the predominant
component being of relatively low molecular weight (330). This
is readily separated and the remaining polymer has a molecular
weight of approx. 500000.
Prior activation of 3 by reaction with three equivalents
of ClMgCH2SiMe3 or LiCH2SiMe3 gives rise to red solutions
(run 8) which do not polymerise alkenes in the absence of an
initiator but do exhibit low activity in the presence of MAO.
This depressed activity may be due to the more sterically
encumbering trimethylsilylmethyl group inhibiting alkyl
abstraction or alkene coordination. The molecular weights
are also lower in this case but the polydispersity decreases
dramatically. These results are in direct contrast to some of the
Cr() alkyls reported by Theopold (e.g. Cp*Cr(CH2SiMe3)2)
which do not require any co-catalyst.24 Gibson et al. have
recently reported the synthesis of a chromium alkyl complex
stabilised by salicylaldiminato ligands that is inactive for the
polymerisation of ethene whereas the parent halide complex
does show some activity.25 Fryzuk et al. recently reported the
complex (PNP)Cr(CH2SiMe3)2, PNP = N(SiMe2CH2PPh2)2,
Fig. 1 GC-MS trace from [12]aneP3Et3CrCl3 ϩ MAO ϩ C2H4
(* denotes toluene solvent; † = alkane; ‡ = alkene).
These hydrocarbons may arise from a mechanism involving
chain transfer of the polymer chain to the aluminium centre,
followed by hydrolysis to give saturated alkanes, as postulated
by Deffieux et al. in the polymerisation of 1-hexene by various
Group 4 metal catalysts,11 and supported by the observations
of Bazan et al. using a Cr() complex.19 The appearance of
roughly equal concentrations of the alkanes also implies that
2962
J. Chem. Soc., Dalton Trans., 2002, 2960–2965

View Article Online
but this is not an active ethene polymerisation catalyst.26 It has
also been reported that alkylchromium complexes of the type
corresponding zirconocene complexes under the same con-
ditions (For example, the zirconium complex Cp2ZrCl2 has an
activity for ethene polymerisation of 1490 kgPE gcatϪ1 hϪ1
whilst the complex TpZrCpCl2 shows a decrease to 76 kgPE
gcatϪ1 hϪ1). The authors ascribe this decrease to the fact that the
Tp ligand is larger and a better electron donor so that the metal
is electronically “overloaded.”30 The relative donating proper-
ties of Cp and Tp have been discussed recently and, for Group 6
metals, Tp was shown to be a better electron donor than Cp.31
Comparison of tritertiaryalkyl phosphine macrocycles with the
analogous triaryl phosphine, [12]aneP3Ph3 are of interest in this
respect since, in principle, the latter should enable manipulation
of the electronic influences of the ligand (as a better π-acceptor)
although IR data for the complexes [12]aneP3R3Mo(CO)3,
(R = Ph, Et) indicate these ligands to have similar electronic
properties (ν(CO) = 1917, 1809 cmϪ1; R = Ph and 1918, 1815
cmϪ1; R = Et).32 The chromium trichloro complex of the
phenyl functionalised macrocycle ([12]aneP3Ph3CrCl3) has not
previously been reported. It is however, readily prepared by
addition of the free macrocycle to suspensions of CrCl3(THF)3
in dichloromethane and is isolated as a blue microcrystalline
solid which is readily soluble in chlorinated hydrocarbons, THF
and toluene and slightly soluble in aliphatic hydrocarbons and
diethyl ether. It is characterised by satisfactory analytical data,
magnetic susceptibility and IR spectroscopy. The reaction of
[12]aneP3Ph3CrCl3 (5) with ethene activated by MAO (run 6,
Table 1) is lower than that of the alkyl substituted macrocyclic
complexes, a difference which may be due to steric factors
arising from the bulkier P–Ph groups rather than any potential
electronic effects. The molecular weight of the polymer also
decreases markedly, again it might be expected that the rate
of chain termination may be enhanced by more sterically
demanding ligands. The low activities shown by all the
macrocycle complexes described may be due to the macrocycle
being generally more electron donating than Tp or Cp ligands.
i
RCrCl2(THF)3 (R = Me, Et, Bu) polymerise ethene at room
temperature, but no details were given.27
We have also previously reported the synthesis of a series
of pendant donor triphosphorus macrocycles.28 We have tested
the activity of the methoxypropyl pendant ether [12]aneP3-
(C3H6OMe)3 complex ([12]aneP3(C3H6OMe)3CrCl3, 4, Fig. 3) in
Fig. 3 [12]aneP3(C3H6OMe)3CrCl3 (4).
ethene polymerisation in order to compare possible influences
of hemi-labile donors. We have previously confirmed the mono-
nuclear structure of 4 by X-ray crystallography10 which also
confirms the facial coordination of the three phosphorus
donors and the absence of any metal–ether interaction. In the
reaction of ethane with 4 activated by Et3Al, no polymer was
produced although GC-MS analysis (Fig. 4) showed the
Late transition metals
There is current interest in polymerisation and co-polymeris-
ation reactions using late transition metals (e.g. Keim’s catalyst
used for oligomerisation in the SHOP process33), especially
based on diimine complexes of nickel and palladium.34 Calcu-
lations suggest that steric protection in the axial position is
essential for producing high molecular weight polymers as
opposed to oligomers.35 The synthesis and structure of cationic
Ni() complexes of [12]aneP3Et3, {[12]aneP3Et3NiBr}Br (6)
and [[12]aneP3Et3Ni(NCMe)2][PF6]2 (7) and the Ni() carbonyl,
[12]aneP3Et3Ni(CO) (8) have recently been reported.36 The
structure of 6 was shown to be heavily distorted from tetra-
hedral in a manner which could facilitate interaction with small
substrate molecules.
The cationic nickel complex 6 polymerises ethene (run 17,
Table 1) although the activities are lower than those reported
in the literature for other systems;1,37 the polymer produced is
highly insoluble and thus presumably of very high molecular
weight. Compound 6 is a very rare example of a Ni compound
containing tertiary phosphine ligands that is active for ethene
polymerisation;38 the majority are based on diimmine ligands
with bulky aryl substituents on the nitrogen or N–O, P–O
chelating ligands. Indeed some catalysts, e.g. Keim’s P–O
chelate ligand, are enhanced by the addition of phosphine
scavenging ligands. The only other catalyst based on a macro-
cyclic ligand is 9[ane]S3PtX4 which also shows relatively low
activities (12 gP mmolϪ1 hϪ1 barϪ1).21 The co-polymerisation of
ethene and CO has also been investigated with the Ni() and
Ni() compounds 7 and 8; under standard conditions reported
in the literature (40 bar ethene, 5 bar CO and 50 ЊC) in dichloro-
methane or acetonitrile, 6 and 7 are inactive in ethane/CO
co-polymerisation as well as alkene polymerisation. Insertions
into Ni–C(alkyl/aryl) bonds, especially by CO, are quite common,39
Fig. 4 GC-MS trace from [12]aneP3(C3H6OMe)3CrCl3 ϩ Et3Al ϩ
C2H4 in CH2Cl2.
presence of even chain oligomers from C8 to C24, (run 15, Table
1). This dramatic difference in selectivity upon simple modifi-
cation of the macrocycle substituents may be due to the ability
of the pendant ether function to compete with alkene binding
which might be expected to suppress propagation, but not
necessarily termination.
The chromium ethyl-macrocycle complex 3 also polymerises
higher α-alkenes and norbornene (runs 11, 13 and 14 in Table 1).
The activities are much lower than for the ethene reactions,
presumably due to steric effects and the stronger M–C bond
of the strained norbornene. No polymerisation or isomeris-
ation of the double bond is observed with higher alkenes (e.g.
1-octadecene). Of the macrocycle adducts, the chromium
complex is the only one found to polymerise propene, suggest-
ing that the smaller titanium and vanadium may be too
sterically encumbered to support a growing chain. The poly-
propene produced is highly crystalline with T g = 102 ЊC and T m
= 206.4 ЊC (∆Hf = 188 J gϪ1). This implies that the polymer
is isotactic but due to the insolubility of the sample, 13C NMR
spectroscopy could not be used to probe the polymer micro-
structure. GPC analysis was also precluded although the
insolubility of the polypropene implies a very high molecular
weight.
It has been suggested that for metallocene catalysts, elec-
tronic effects predominate in the polymerisation of ethene, but
for the polymerisation of propene the activities are more
dependent on steric factors.29 The tris(pyrazolyl)borate (Tp)
ligand can be described as a six electron ligand similar to Cp,
but zirconium Tp complexes are much less active than the
J. Chem. Soc., Dalton Trans., 2002, 2960–2965
2963

View Article Online
but insertion into Ni–CO bonds is less common and this
presumably disfavours the initiation step. Additionally, since 7
is a compound of d10 Ni(), it may be relatively resistant to
coordinating additional ligands. In general, most ethene/CO
co-polymerisation catalysts are cationic palladium complexes
which contain bidentate phosphines and weakly coordinating
anions;40 there are relatively few nickel based co-polymerisation
catalysts known,41 in part due to the facile formation of
Ni(CO)4. Similar reactions were attempted with compound 3
although no activity was observed.
was dissolved in CH2Cl2 (50 cm3) and MAO (2 cm3 of a 10 wt%
solution in toluene, 92 : 1 Al : Cr) added. This was stirred for
15 minutes and transferred to an autoclave via cannula. The
autoclave was evacuated, backfilled with ethene to a pressure of
3 bar and stirred for 1 hour. The reaction was deactivated by the
addition of methanol (50 cm3), the solid isolated by filtration
and an aliquot of the solution was analysed by GC-MS
spectroscopy. The white polymer was then washed with a 5%
HCl solution in MeOH (200 cm3), filtered and dried in vacuo (12
h at 0.1 mmHg). The reaction with the titanium catalyst gave no
solid precipitate; removal of the solvent in vacuo and washing as
above enabled recovery of the polymer. In the case of propene
reactions, again the product polymer did not precipitate from
the reaction mixtures which were worked up by evaporation in
vacuo followed by washing as above.
Conclusion
Electropositive transition metal complexes of macrocyclic tri-
phosphines show activity in the polymerisation of alkenes when
combined with an appropriate initiator. The pendant function
on the macrocyclic ligand significantly influences the reactivity
of the catalyst and the properties of the polymer product. The
stability and activity of these systems is substantially greater
than for related acyclic phosphine ligand complexes and is
presumably due to the macrocyclic coordination effect. Poly-
merisation activity is also extended to nickel compounds which
constitute a new class of homogeneous alkene polymerisation
catalysts.
Reactions with 1-hexene and norbornene
To a solution of the catalyst in an appropriate solvent was
added the alkyl aluminium initiator and the solution stirred for
15 min. The monomer was then added in one portion and
stirred for 1 h. The solvent was removed in vacuo to give a solid
that was washed with a HCl/MeOH solution as above, filtered
and dried in vacuo (12 h at 0.1 mmHg).
Preliminary activation of catalysts by alkylation
Experimental
The following method is typical and was employed for the
reactions of [12]aneP3Et3MCl3 with lithium alkyls (RLi where
for M = Ti, V, Cr; R = Bz, Ph, CH2CMe3, CH2C(Me)2Ph,
CH2SiMe3, CH(SiMe3)2; M = Cr, R = nBu). To a cold (Ϫ78 ЊC)
suspension of [12]aneP3Et3CrCl3 (0.60 g, 1.29 mmol) in THF
(10 cm3) was added LiCH2SiMe3 (3.90 cm3 of a 1.0 M solution
in petrol, 3.9 mmol) and this was allowed to warm to room
temperature over 2 h. After stirring at room temperature for 3 h
the solvent was removed in vacuo and the red residue extracted
with toluene (3 × 20 cm3). The solvent was removed in vacuo to
yield a red oil which was dissolved in petroleum ether (5 cm3)
and cooled to Ϫ35 ЊC. A red oil (0.48 g) was obtained after 3 h
which was isolated by filtration at Ϫ40 ЊC and dried in vacuo.
µeff = 3.58 µB (Evans method). A AgNO3 halide test showed this
material to be halide free, hydrolysis with D2O gave only
All reactions were carried out under nitrogen (purified by pass-
ing over a CrO column) using standard Schlenk line techniques
or in a vacuum atmospheres inert atmosphere box. The
compounds [12]aneP3Et3TiCl3, [12]aneP3Et3VCl3, [12]aneP3R3-
CrCl3, (R = Et, (CH2)3OMe),10 [12]aneP3Ph3,32 ([12]aneP3Et3-
NiBr)(Br), [12]aneP3Et3Ni(CO), ([12]aneP3Et3Ni(NCMe)3)-
(PF6)38 and (C6F5Cu)2ؒdioxane42 were prepared by literature
procedures. Solvents were dried with standard drying agents
and distilled immediately before use. Petroleum ether had bp
40–60 ЊC. 1H and 13C NMR spectra were recorded on a Bruker
DPX400 spectrometer operating at 400.13 MHz and 100 MHz,
respectively. 31P NMR spectra were acquired using a JEOL
FX90Q at 36.2 MHz and referenced to 85% H3PO4. MAO was
used as a 10 wt% solution in toluene. 1-Hexene was dried over
sodium–benzophenone, distilled and stored over molecular
sieves. Ethene and propene polymerisation reactions were
carried out in a 250 cm3 stainless steel autoclave or in a glass
pressure vessel (100 cm3). GC-MS analyses were carried out on
a HP 5890 GC and VGTRIO-1 mass spectrometer using a
BPX-5 column.
n
Me3SiCH2D and for R = Bu, there was no evidence of Cr–H
bonds in the IR spectrum.
Reactions with ethene and CO
In a typical experiment [12]aneP3Et3CrCl3 (0.053 g, 0.11 mmol)
was dissolved in CH2Cl2 (50 cm3) and MAO (2 cm3 of a 10
wt% solution in toluene, 92 : 1; Al : Cr) added. After stirring for
15 minutes this was transferred to an autoclave, evacuated
and backfilled first with CO (3 bar) and then C2H4 to a total
pressure of 6 bar. This was sealed and stirred for 1 h. The excess
gas was vented and MeOH (50 cm3) added. A sample of the
reaction mixture was analysed by GC-MS and the remaining
solvent removed in vacuo.
Synthesis of [12]aneP3Ph3CrCl3 (4)
ToasuspensionofCrCl3(THF)3 (0.04g,0.107mmol)indichloro-
methane (10 cm3), was added a solution of [12]aneP3Ph3
(4.5 cm3, 0.0244 M in hexanes, 0.109 mmol). The suspension
was stirred (12 h) during which time the suspension dissolved
and the solution became deep blue. The resulting solution
was filtered, evaporated to dryness in vacuo and the solid
residue was triturated with hexanes (3 × 10 cm3) to give a pale
blue powder. The blue solid was extracted into warm (30 ЊC)
dichloromethane (30 cm3) giving a green precipitate and a deep
blue solution which was filtered. The blue solution was con-
centrated (10 cm3) and cooled (Ϫ30 ЊC, 12 h) yielding a blue
microcrystalline solid which was filtered, washed with hexane
(10 cm3) and dried in vacuo. Yield: 0.02 g, 0.033 mmol, 31%
based upon CrCl3(THF)3. Analysis found (calc.): %C, 53.8
(53.25); %H, 5.8 (5.42). IR/cmϪ1 (Nujol): 2922, 1463, 1377,
1302, 1157, 1091, 1021, 966, 820, 722, 482, 309. µeff = 3.62 µB
(Evans method).
Acknowledgements
We thank the EPSRC for a studentship (R. J. B.), Mr Robert
Jenkins of this department for GC-MS analysis and RAPRA
Technologies Ltd. for the GPC analysis.
References
1 For recent reviews see (a) M. Bochmann, J. Chem. Soc., Dalton
Trans., 1996, 255; (b) H. H. Brintzinger, D. Fischer, R. Mullhaupt,
B. Rieger and R. M. Weymouth, Angew. Chem., Int. Ed. Engl., 1995,
34, 1143; (c) A. L. McKnight and R. M. Weymouth, Chem. Rev.,
1998, 2587; (d ) K. H. Theopold, Eur. J. Inorg. Chem., 1998, 15; (e)
G. J. P. Britovsek, V. C. Gibson and D. F. Wass, Angew. Chem., Int.
Ed., 1999, 38, 428; ( f ) W. Kaminsky, J. Chem. Soc., Dalton Trans.,
1998, 1413; (g) R. F. Jordan, Adv. Organomet. Chem., 1991, 32, 325.
Reactions with ethene and propene
In a typical reaction [12]aneP3Et3CrCl3 (0.040 g, 0.086 mmol)
2964
J. Chem. Soc., Dalton Trans., 2002, 2960–2965

View Article Online
2 K. Zeigler, E. Holzkamp, H. Breil and H. Martin, Angew. Chem.,
1955, 67, 541; K. Ziegler, Angew. Chem., 1964, 76, 545.
3 G. Natta, Angew. Chem., 1956, 68, 393; G. Natta, Angew. Chem.,
1964, 76, 553.
25 V. C. Gibson, S. Mastroianni, C. Newton, C. Redshaw, G. A. Solan,
A. J. P. White and D. J. Williams, J. Chem. Soc., Dalton Trans., 1999,
147.
26 M. D. Fryzuk, D. B. Leznoff, S. J. Rettig and V. G. Young Jr.,
J. Chem. Soc., Dalton Trans., 1999, 147.
4 M. P. McDaniel, Adv. Catal., 1985, 33, 47.
5 F. J. Karol, G. L. Karapinka, C. Wu, A. W. Dow, R. N. Johnson and
W. L. Carrick, J. Polym. Sci., Part A-1, 1972, 10, 2621.
6 M. P. Coles, C. I. Dalby, V. C. Gibson, W. Clegg and M. R. J.
Elsegood, J. Chem. Soc., Chem. Commun., 1995, 1709.
7 G. J. P. Britovsek, V. C. Gibson, B. S. Kimberley, P. J. Maddox,
S. J. McTavish, G. A. Solan, A. J. P. White and D. J. Williams, Chem.
Commun., 1998, 849; B. L. Small, M. Brookhart and A. M. A.
Bennett, J. Am. Chem. Soc., 1998, 120, 7143.
27 K. Nishimura, H. Kuribayashi, A. Yamamoto and S. Ikeda,
J. Organomet. Chem., 1972, 37, 317.
28 P. G. Edwards, J. S. Fleming and S. S. Liyanage, J. Chem. Soc.,
Dalton Trans., 1997, 193.
29 N. Piccolrovazzi, P. Pino, G. Consiglio, A. Sironi and M. Moret,
Organometallics, 1990, 9, 3098; W. Kaminsky, R. Engehausen,
K. Zoumis, W. Spaleck and J. Rohronann, J. Makromol. Chem.,
1992, 193, 1643.
30 H. G. Alt and A. Koppl, Chem. Rev., 2000, 100, 1205.
31 D. M. Tellers, S. J. Skoog, R. G. Bergman, T. B. Gunnoe and
W. D. Harman, Organometallics, 2000, 19, 2428.
8 R. Duchateau, A. Meetsma and J. H. Teuben, Chem. Commun.,
1996, 223; M. P. Coles and R. F. Jordan, J. Am. Chem. Soc., 1997,
119, 8125.
9 P. G. Edwards, J. S. Fleming and S. S. Liyanage, Inorg. Chem., 1996,
35, 463.
32 D. J. Jones, P. G. Edwards, R. P. Tooze and T. Albers, J. Chem. Soc.,
Dalton Trans., 1999, 1045.
33 W. Keim and M. Peuckert, Organometallics, 1983, 2, 594; W. Keim,
A. Behr, B. Gruber, B. Hoffmann, F. H. Kowaldt, U. Kurschner,
B. Limbacker and F. P. Sistig, Organometallics, 1986, 5, 2356.
34 S. D. Ittel, L. K. Johnson and M. Brookhart, Chem. Rev., 2000, 100,
1169.
35 L. Deng, T. K. Woo, L. Cavallo, D. M. Margl and T. Zeigler, J. Am.
Chem. Soc., 1997, 119, 6177.
36 P. G. Edwards, F. Ingold, S. J. Coles and M. B. Hursthouse, Chem.
Commun., 1998, 545.
10 R. J. Baker, P. C. Davies, P. G. Edwards, R. D. Farley, S. S. Liyanage,
D. M. Murphy and B. S. Yong, Eur. J. Inorg. Chem., 2002, 1975.
11 For a related study of 1-hexene oligomerisation reactions see
H. Q. Liu, A. Deffieux and P. Sigwalt, Polym. Bull., 1990, 24, 355.
12 S. S. Wreford and J. F. Whitney, Inorg. Chem., 1981, 20, 3918.
13 L. B. Kool, M. D. Rausch, H. G. Alt, M. Herberhold, U. Thewalt
and B. Wolf, Angew. Chem., Int. Ed. Engl., 1985, 24, 394.
14 J. A. Jensen, S. R. Wilson, A. J. Schultz and G. S. Girolomi, J. Am.
Chem. Soc., 1987, 109, 8094.
15 For example see N. Desmangles, S. Gambarotta, C. Bensimon,
S. Davis and H. Zahalka, J. Organomet. Chem., 1998, 562, 53;
M. J. R. Brandsma, E. A. C. Brussee, A. Meetsma, B. Hessen and
J. H. Teuben, Eur. J. Inorg. Chem., 1998, 1867; Also ref. 1e.
16 R. Schmid and T. Ziegler, Organometallics, 2000, 19, 2756.
17 R. D. Kohn, M. Haufe, S. Mihan and D. Lilge, Chem. Commun.,
2000, 1927.
18 I. Tritto, S. X. Li, M. C. Sacchi, P. Locatelli and G. Zannoni,
Macromolecules, 1998, 128, 257; I. Tritto, S. Li, M. C. Sacchi and
G. Zannoni, Macromolecules, 1993, 26, 7111.
37 S. Y. Desjardins, K. J. Cavell, J. L. Hoare, B. W. Skelton,
A. A. Sobolev, A. H. White and W. Keim, J. Organomet. Chem.,
1997, 544, 163.
38 N. A. Cooley, S. M. Green, D. F. Wass, K. Heslop, A. G. Orpen and
P. G. Pringle, Organometallics, 2001, 20, 4769.
39 C. P. Kubiak, in Comprehensive Organometallic Chemistry II, vol. 9,
ch. 1, eds. E. W. Abel, F. G. A. Stone and G. Wilkinson, Pergamon
Press, London, 1995.
40 M. C. Gardiner, W. A. Herrmann, C. P. Reisinger, J. Schwarz and
M. Speigler, J. Organomet. Chem., 1999, 572, 239.
19 G. C. Bazan, J. S. Rogers and C. C. Fang, Organometallics, 2001, 20,
2059.
20 F. J. Wu, US Pat., 5,550,305, 1998.
21 S. Timonen, T. T. Pakkanen and J. A. Pakkanen, J. Mol. Catal.,
1996, 111, 267.
22 A. Dohring, V. R. Jensen, P. W. Jolly, W. Thiel and J. C. Weber,
Organometallics, 2001, 20, 2234.
23 B. J. Thomas, S. K. Noh, G. K. Schulte, S. C. Sendlinger and
K. H. Theopold, J. Am. Chem. Soc., 1991, 113, 893.
24 K. H. Theopold, Eur. J. Inorg. Chem., 1998, 15.
41 U. Klabunde, T. H. Tulip, D. C. Roe and S. D. Ittel, J. Organomet.
Chem., 1987, 344, 141; U. Klabunde, R. Mulhaupt, T. Herskovitz,
A. H. Janovicz, J. Calabrese and S. D. Ittel, J. Polym. Sci.: Polym.
Chem., 1987, 25, 1989; U. Klabunde and S. D. Ittel, J. Mol. Catal.,
1987, 41, 123; W. Keim, H. Maas and S. Mecking, Z. Naturforsch.,
Teil B, 1995, 50, 430; U. Klabunde, US Pat., 4,698,403, 1987;
U. Klabunde,US Pat., 4,716,205, 1987; B. Driessen, M. J. Green and
W. Keim, Eur. Pat. Appl., 470,759,A2, 1992; See also ref. 37.
42 A. Cairncross, W. A. Sheppard and E. Wonchoba, Org. Synth., 1979,
59, 122.
J. Chem. Soc., Dalton Trans., 2002, 2960–2965
2965