Ethylene polymerisation and oligomerisation with arene-substituted phenoxy-imine complexes of titanium: Investigation of multi-mechanism catalytic behaviour

Article abstract of DOI:10.1039/c3dt32183e

A range of unsubstituted (1,2) and 6-substituted (3-5) ortho-phenoxy-imine ligands have been prepared and converted to their silyl ether derivatives (6-10). Reaction of silyl ethers with TiCl₄(thf)₂ in the case of the unsubstituted species yields bis-ligated complexes while the substituted species react cleanly to yield complexes of the form [Ti(O^NR)Cl₃(thf)]. In most cases the complexes have been characterised by X-ray crystallography. Testing of the complexes for ethylene oligomerisation and polymerisation has been undertaken employing alkylaluminium co-catalysts (AlEt₃, MAO). In all cases the predominant product formed is polyethylene however careful analysis of the liquid phase reveals a complex process by which 1-butene is most likely formed via Cossee mechanism while 1-hexene results from a metallacyclic process.

Full text of DOI:10.1039/c3dt32183e

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Ethylene polymerisation and oligomerisation with
arene-substituted phenoxy-imine complexes of
titanium: investigation of multi-mechanism catalytic
behaviour†
Cite this: Dalton Trans., 2013, 42, 4185
James A. Suttil,a David S. McGuinness,*a Michael G. Gardinera and Stephen J. Evansb
A range of unsubstituted (1,2) and 6-substituted (3–5) ortho-phenoxy-imine ligands have been prepared
and converted to their silyl ether derivatives (6–10). Reaction of silyl ethers with TiCl4(thf)2 in the case of
the unsubstituted species yields bis-ligated complexes while the substituted species react cleanly to yield
complexes of the form [Ti(O^NR)Cl3(thf)]. In most cases the complexes have been characterised by X-ray
crystallography. Testing of the complexes for ethylene oligomerisation and polymerisation has been
undertaken employing alkylaluminium co-catalysts (AlEt3, MAO). In all cases the predominant product
formed is polyethylene however careful analysis of the liquid phase reveals a complex process by which
1-butene is most likely formed via Cossee mechanism while 1-hexene results from a metallacyclic process.
Received 20th September 2012,
Accepted 30th January 2013
DOI: 10.1039/c3dt32183e
www.rsc.org/dalton
One such example is the highly active and selective ethylene
trimerisation catalyst described by Hessen and coworkers in
Introduction
The oligomerisation and polymerisation of olefins into higher 2001 (I). While selective ethylene dimerisation systems based
olefin feedstocks and polyolefin resins continues to be an area on titanium were previously known, this was the first reported
that receives much attention from both industry and selective trimerisation system for this metal.29 Early reports by
academia. Of particular interest is the selective oligomerisation Pellecchia et al. showed that Cp*TiMe3-B(C6F5)3 in toluene
of ethylene to co-monomer grade linear alpha olefins (LAOs, generated butyl-branched polyethylene along with 1-hexene
such as 1-butene, 1-hexene and 1-octene), a process by which when exposed to ethylene, demonstrating the catalyst’s ability
excessive formation of lower value, higher molecular weight to catalyse both trimerisation and co-polymerisation.30,31 In
fractions can be avoided.1–8 As such, there is significant drive light of this result, Hessen and coworkers recognised the
in developing new selective oligomerisation technologies and ability of the aromatic solvent toluene to stabilise the catalytic
understanding the mechanism by which they function.2,9–28 In species yielding trimerisation selectivity. By modifying the
recent years, early transition metal complexes, in combination cyclopentadienyl ring to incorporate a pendant arene function-
with alkyl aluminium activators such as methylaluminoxane ality, they were able to efficiently convert the mono(cyclopenta-
(MAO), that selectively di-, tri- and tetramerise ethylene have dienyl) polymerisation system into an ethylene trimerisation
become well established in the literature.1–6 While the most system yielding selectivity of up to 93% 1-hexene with excellent
prevalent metal employed for selective trimerisation is activity.32 Through a series of systematic ligand modifications,
undoubtedly Cr, more examples of well characterised selective the authors were able to demonstrate that a careful balance
Ti complexes are being reported.
between steric influences of the ligand, the nature of the
Cp/arene bridging atom and electronics is required to achieve
both high selectivity and activity.32–35 It has been hypothesised
that the exceptional selectivity is a result of the arene function-
ality’s ability to interact with the metal centre in a hemilabile
aSchool of Chemistry, University of Tasmania, Private Bag 75, Hobart, 7001,
Australia. E-mail: david.mcguinness@utas.edu.au; Fax: +61 (0)362262858;
Tel: +61 (0)362263783
fashion yielding
a number of coordination geometries
bSasol Technology Ltd, PO Box 1183, Sasolburg, 1947, South Africa.
that work to stabilise the various stages of the metallacycle
mechanism.10,13,20,35–37 Since this initial discovery, several
other reports have been published investigating the effect of
electron withdrawing groups on the arene ring, varying the
arene functionality to thienyl, thioether and ether substituted
E-mail: Stephen.evans@sasol.com
†Electronic supplementary information (ESI) available: Details of 1-butene and
1-hexene isotopomer analysis and X-ray structure of 13 and dichlorobis(3-tert-
butyl-2-oxybenzoyl) titanium(IV). CCDC 902037–902043. For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/c3dt32183e
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 4185–4196 | 4185

View Article Online
Paper
Dalton Transactions
derivatives and changing the metal to zirconium and hafnium; of the complexes seems to be highly dependent upon the
although benchmarks equal to that of the original systems choice of starting reactants and the steric bulk of both the
have not been achieved.38–41
6-position of the phenol and the nitrogen substituent.44–51
More recently, researchers from Mitsui Chemicals have The most broadly applicable and consistent of the literature
demonstrated the conversion of the highly efficient phenoxy- methods was first reported by Lancaster and coworkers
imine (FI)42 ethylene polymerisation motif to give selective in 2003 and involves the silylation of the salicylaldimine pre-
ethylene trimerisation systems capable of exceptionally high cursor and subsequent dehalosilylation reaction with
activities.43 The switching of selectivity was achieved by TiCl4(thf)2.45,46 This methodology has since been widely
employing only a mono-ligated titanium centre, instead of the employed to prepare heteroligated titanium precatalysts
normal bis-arrangement that yields polymerisation, and incor- bearing at least one salicylaldiminato ligand; such catalysts are
poration of an additional donor atom to yield a tridentate of interest in ethylene/alkene co-polymerisation reactions as
complex (II).
they combine the exceptional activities of FI catalysts with the
higher alkene incorporation of other systems.52–62 Addition-
ally, the resulting complexes should be facile towards acti-
vation for ethylene oligomerisation and polymerisation
employing alkyl aluminium activators, as they contain easily
exchangeable chloro ligands. As such, this methodology has
been employed to prepare a series of novel (salicylaldiminato)-
titanium complexes.
The starting salicylaldimines 1–5 and the subsequent silyl
ethers 6–10 were prepared in moderate to good yields via a
variety of modified literature procedures.45,46 These species
were characterised by 1H and 13C NMR spectroscopy and in
most cases MS; where the silyl ether pre-ligands bore no sub-
stituent on the 6-position only the hydrolysed species could be
observed by MS indicating substantial capacity for hydrolysis.
Initial complex synthesis focused on the pre-ligands 6 and 7
which have an unsubstituted 6-position in the phenol. In both
cases addition of the ligand to TiCl4(thf)2 yield dark red solu-
tions after stirring for 24 hours. Upon work up bis(salicylaldi-
minium)titanium complex 11 and bis(salicylaldiminato)
titanium complex 12 (Scheme 1) were obtained regardless of
an initial 1 : 1 metal to ligand ratio. The 1H and 13C NMR
spectra of 11 shows a single set of signals for the two sym-
metry equivalent ligands. The aldimine –CHvN(H)– resonance
Given these more recent findings, we were interested in
combining aspects of both Hessen’s catalyst and the Mitsui
system, namely the introduction of an arene functionality to
phenoxy-imine catalysts, to yield novel precatalysts with poten-
tially interesting ethylene oligomerisation properties. Previous
reports on other mono-ligand phenoxy-imine catalysts bearing
a phenyl substituent on the nitrogen have shown they have a
propensity for polymer formation, although none of these
reports comment on analyses of the oligomers that may have
formed in the liquid phase.44–46 Given these previous findings,
our focus was on ligand systems employing a bridging atom
between the nitrogen and arene functionalities, akin to the
Hessen system (III, X = ethyl, α,α-dimethylethyl). Herein we
report the preparation and characterisation of a range of both
bis- and mono(salicylaldiminato) Ti complexes as well as their
oligomerisation and polymerisation behaviour when activated
with MAO.
Results and discussion
Synthesis and structures
The preparation of mono(salicylaldiminato) titanium com-
plexes has been previously demonstrated for many com-
binations of either salicylaldimines or silyl ethers of
salicylaldimines with titanium tetrachloride or titanium tetra-
chloride tetrahydrofuran adduct. The resulting stoichiometry Scheme 1 Synthesis of complexes 11 and 12.
4186 | Dalton Trans., 2013, 42, 4185–4196
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
1
is observed in the H NMR spectra as a doublet at δ 8.03 (3J =
7.8 Hz) while the broad vN(H)– resonance is present at δ
12.97. These NMR features are consistent with the previously
reported analogue tetrachlorobis(2,6-diisopropylphenylsalicyl-
aldiminium)titanium(IV) and confirm that the proton is for-
mally bound to the nitrogen atom yielding an overall neutrally
charged ligand with zwitterionic character.63 Crystals suitable
for X-ray diffraction were obtained by slow diffusion of pet-
roleum spirits into
a dichloromethane solution of the
complex. The solid state structure shows a distorted octahedral
geometry with the oxygen atoms arranged trans-bound to the
titanium atom with the chloride ligands adopting the equator-
ial positions (Fig. 1). Key bond length and angles are in
general agreement with the analogous structure reported by
Strauch et al.63 Various reports of such compounds have pre-
viously been published in the literature, however they are typi-
cally prepared by reaction of salicylaldimines with titanium
tetrachloride.63–68 Given that the solvents were rigorously dried
and stored, the source of the proton in our case is currently
unclear.
Attempts to isolate complex 12 initially gave an intractable
mixture of products but upon standing for an extended period
crystallises from a mixture of tetrahydrofuran and petroleum
spirits as dark red crystals. 1H and 13C NMR spectroscopy
(CD2Cl2) show only a single set of resonances indicating the
formation of only a single isomer. X-ray crystallography data
again reveals an octahedral complex wherein the anionic
oxygens adopt a trans-geometry while the remaining nitrogen
donors and chloro ligands are in a cis-arrangement to each
other (Fig. 2). The geometry, bond lengths and angles are
Fig. 2 Molecular structure of 12. Thermal ellipsoids are shown at the 50%
probability level. All methyl, methylene and aromatic-ring hydrogen atoms are
omitted for clarity. Selected bond lengths (Å) and angles (°): Ti1–O1,O2,N1,N2,
Cl1,Cl2 1.884(2), 1.851(2), 2.200(3), 2.181(3), 2.3006(10), 2.3211(11), O1,
O2–Ti1–N1,N2chelate ring 82.61(10), 82.85(10), X–Ti1–Ycis(acyclic),trans 83.49(10)–99.61
(4), 168.88(11)–172.94(8), C1–O1–Ti1 138.1(2), C16–O2–Ti1 141.9(2).
broadly consistent with a previously reported titanium FI
analogue bearing a benzyl N-substituent.69 Furthermore, this
geometry is in complete agreement with previously reported
computational predictions on FI polymerisation catalysts,
which suggest that for analogous complexes with low steric
bulk that a trans-arrangement of the oxygen atoms and cis-
arrangement of the nitrogen donors and chloro ligands is
the most energetically favourable isomer for such a 2 : 1
complex.70
Given these results, focus was shifted towards ligands
bearing sterically bulky groups on the phenol 6-position. Com-
plexes 13–15 were prepared by the method published by
Lancaster and coworkers as red to dark red solids (Schemes 2
and 3) which were all moisture and air sensitive but thermally
stable.45,46 The crystal structure of a phenoxyaldehyde hydro-
lysis product are shown in the ESI.† All the complexes have been
1
characterised by H and 13C NMR and microanalysis and in all
cases could be recrystallised from tetrahydrofuran–petroleum
spirit mixtures to yield crystals suitable for X-ray diffraction.
Complex 14 adopts a distorted octahedral geometry with the
electronically favourable mutually trans-oxygen arrangement
while the chloride ligands are meridionally arranged around
the titanium centre (Fig. 3). This geometry is consistent with
previously reported low steric bulk mono(salicylaldiminato)
titanium complexes.45,46 Complex 13 is structurally similar to
14 and is shown in the ESI.† In both 13 and 14 the N-substitu-
ent is tertiary. Interestingly, complex 15 adopts an arrange-
ment in which the ligating oxygen atoms now form a cis-
arrangement while the chloride ligands are facially arranged
(Fig. 4). Previously, it has been reported that this geometry is
favoured for complexes with sterically bulky nitrogen substitu-
ents however in our case the bulk at the α-C (primary
Fig. 1 Molecular structure of 11 (’ denotes symmetry operator 1 − x, −y, −z).
Thermal ellipsoids are shown at the 50% probability level. Diffuse lattice solvent
was removed in the refinement. All methyl and aromatic-ring hydrogen atoms
are omitted for clarity. Selected bond lengths (Å) and angles (°): Ti1–O1,Cl1,Cl2
1.868(2), 2.3567(7), 2.3768(8), C2–C7 1.441(4), C7–N1 1.270(4), N1–H1 0.90(4),
Cl1–Ti1–Cl2,Cl2’ 88.13(3), 91.87(3), O1–Ti1–Cl 88.60(6)–91.40(6), C1–O1–Ti1
152.35(18), C2–C7–N1 127.8(3).
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 4185–4196 | 4187

View Article Online
Paper
Dalton Transactions
Scheme 2 Synthesis of complexes 13 and 14.
Fig. 4 Molecular structure of 15. Thermal ellipsoids are shown at the 50%
probability level. Disorder in the THF is omitted for clarity, as are all methyl,
methylene and aromatic-ring hydrogen atoms. Selected bond lengths (Å) and
angles (°): Ti1–O1,O2 1.776(4), 2.095(4), Ti1–N1 2.158(5), Ti1–Cl 2.242(2)–
2.2900(18), O1–Ti1–N1 82.55(18), X–Ti1–Ycis(acyclic),trans 80.25(17)–101.98(8),
167.43(14)–172.42(15), C1–O1–Ti1 143.4(4).
may be indicative of a dynamic process in solution. This, com-
bined with the decreased sterics at the α-C position, may
suggest a less rigid N/Ph ethyl linker and in turn a more facile
dynamism between different conformers of this group (relative
to 13 and 14). Alternatively a rapid isomerisation between the
facial and meridional ligand arrangements could also account
for this observation; this has not been explored any further
however.
Scheme 3 Synthesis of complexes 15 and 16.
In addition, a small amount of crystalline bis-ligated
complex 16 was isolated as a by-product of the preparation of
15. X-ray crystallography reveals the expected trans-oxygen
arrangement with mutually cis-nitrogen and chloride ligands
(Fig. 5).70 This structure confirms the computational predic-
tions of Repo and co-workers.69
Ethylene oligomerisation and polymerisation
The bis-ligated and mono(salicylaldiminato) titanium com-
plexes 11–15 were screened for ethylene oligomerisation/poly-
merisation activity in combination with methylaluminoxane
(MAO). Table 1 shows that in all cases (entries 1–5), upon acti-
vation, ethylene polymerisation is the dominant mode of cata-
lysis with polyethylene selectivity ranging from 95.9 to 99.9%
polymer and with moderate productivity. For precatalysts 11
and 12 the moderate productivity (entries 1 and 2) is somewhat
surprising given that they are direct analogues of the highly
active FI catalysts. However, this result may be attributed to
the lack of steric bulk at the ortho-position of the phenol ring;
which has been shown to significantly improve productivity.71
It was also suggested by a referee that ligand transfer to Al
might be more facile without this steric protection, although
we did not investigate this possibility. Compounds 13–15 show
productivities (entries 3–5) broadly comparable with those pre-
viously published by Lancaster and coworkers and Owiny
et al., but, as these authors have previously highlighted, these
Fig. 3 Molecular structure of 14. Thermal ellipsoids are shown at the 50%
probability level. All methyl, methylene and aromatic-ring hydrogen atoms are
omitted for clarity. Selected bond lengths (Å) and angles (°): Ti1–O1,O2 1.7806
(16), 2.2020(16), Ti1–N1 2.3028(19), Ti1–Cl 2.2944(7)–2.3073(8), O1–Ti1–N1
84.42(7), X–Ti1–Ycis(acyclic),trans 82.86(5)–96.85(3), 166.78(3)–179.29(5), C1–O1–
Ti1 145.33(14).
substitution) from the nitrogen has significantly decreased
bulk in comparison to complexes 13 and 14 (tertiary substi-
1
tution).46 Complex 15 shows a somewhat broadened H NMR
spectrum, particularly for the tetrahydrofuran signals, which
4188 | Dalton Trans., 2013, 42, 4185–4196
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
formation of 1-butene is somewhat retarded compared to
longer LAO’s. We have previously reported a similar trend for a
series of tridentate carbene/pyridine/carbene chromium com-
plexes which oligomerise via an extended metallacycle mech-
anism (it should be noted that this observation is not a result
of butene loss prior to analysis, see Experimental section).72 In
this and other studies, it has been proposed that for a metalla-
cyclic mode of oligomerisation the sterically strained metalla-
cyclopentane cannot adopt the required geometry for
β-hydrogen shift and as such ring growth is kinetically
favoured.10,13,20,73–75 However, given the presence of a signifi-
cant amount of C6 isomers, most likely the result of ethylene/
1-butene co-dimerisation which has been recently shown to
result from a Cossee mechanism of dimerisation,76 a process
by which 1-butene and 1-hexene are formed by different
modes of catalysis cannot be ruled out (discussed below).
Conversely, for precatalyst 13, analysis of the liquid phase
shows 14% 1-butene, 65% 1-hexene, 4% C6 isomers, 5%
1-octene and the remainder as trace higher oligomers (C10+).
As previously discussed, studies by Hessen and co-workers
have shown that cumyl-bearing mono(cyclopentadienyl) tita-
nium complexes can undergo hemilabile metal–arene ring
interactions that can effectively stabilise the key steps in the
metallacycle mechanism which in turns yields a highly active
and selective ethylene trimerisation catalysts.29,32,34 While 13,
which also contains a nitrogen bearing cumyl functionality,
does yield almost exclusively polymer, it is possible that a
similar Ti–arene interaction is occurring within this system to
give the liquid phase selectivity to 1-hexene.
Fig. 5 Molecular structure of 16. Thermal ellipsoids are shown at the 50%
probability level. All methyl, methylene and aromatic-ring hydrogen atoms are
omitted for clarity. Selected bond lengths (Å) and angles (°): Ti1–O1,O2,N1,N2,
Cl1,Cl2 1.844(2), 1.856(2), 2.205(3), 2.192(3), 2.3144(10), 2.3087(10), O1,O2–
Ti1–N1,N2chelate ring 81.63(9), 81.35(10), X–Ti1–Ycis(acyclic),trans 79.12(11)–98.57(4),
166.51(8)–170.41(11), C1–O1–Ti1 145.5(2), C20–O2–Ti1 144.3(2).
Given that 15 yielded the highest amount of oligomers
(Table 1, entry 5) and that the oligomeric distribution it dis-
plays is representative of the majority of catalysts discussed
herein, it was selected for detailed mechanistic analysis. Here
again we see evidence for metallacyclic oligomerisation. While
the majority of the liquid phase C6 fraction was 1-hexene
(88%), a number of other products were determined by
GC-FID and GC-MS. 3-Ethylbut-1-ene (5%), 3-methylpent-1-ene
(2%), an unidentified internal hexene (2%), a unidentified hexa-
diene (1%), 3-methylpentane (1%), methylcyclopentane (1%)
and trace hexane were also quantified. The majority of the C6
products can be rationalised by the co-dimerisation of one
unit of ethylene with one unit of 1-butene, as has been pre-
viously reported for other group IV catalysts such as the indust-
rially relevant Alphabutol system (Ti[OBu]4/AlEt3).1,2 However,
from a mechanistic view point, this distribution can be readily
ratified by either a metallacycle (Scheme 4) or a Cossee mech-
anism (Scheme 5). The presence of a hexadiene and methyl-
cyclopentane are harder to reconcile. Hexadienes have previously
been reported, in conjunction with the other co-dimers, as
part of the Alphabutol co-dimer distribution (which occurs via
a Cossee mechanism).76,77 While methylcyclopentane has
been shown to form during ethylene tri- and tetramerisation
with PNP-Cr catalysts (via metallacycles).21,78 Additionally, the
alkanes 3-methylpentane and hexane most likely result from
chain transfer from Ti to Al and then quenching during hydro-
lytic work up of the catalytic mixture. Finally, a proportion of
Table 1 Ethylene oligomerisation/polymerisation with 11–15 and MAOa
Entry
Catalyst
Productivityb
PE (wt%)
Olig (wt%)
1
2
3
4
5
6c
11
12
13
14
15
15
7.6
6.0
29.2
7.7
9.8
0.1
99.9
99.9
99.6
99.6
95.9
83.0
0.1
0.1
0.4
0.4
4.1
17.0
a Conditions: 20 μmol catalyst loading, toluene (50 mL), 300 equiv.
MAO, 10 bar ethylene, 30 °C, 30 min. b kg product {(mol metal) bar}−1
.
c 75 μmol catalyst loading, toluene (50 mL), 5 equiv. AlEt3 in place of
MAO, 20 bar ethylene, 55 °C, 30 min.
productivities are still orders of magnitudes less than their
group IV metallocene or FI counterparts.44–46
Even though these precatalysts clearly act predominantly as
polymerisation systems; careful analysis of the liquid oligomer
phase has been undertaken and yielded some interesting
mechanistic insights. Precatalysts 11, 12, 14 and 15 all show a
Schulz–Flory type distribution of even-carbon numbered linear
alpha olefins in addition to some C6 isomers and branched
C10 products when activated with MAO. However, the Schulz–
Flory constant for 1-butene/1-hexene ranges from 0.50–0.69
while the constants for 1-hexene/1-octene and 1-octene/
1-decene are in the range of 0.15–0.38; this suggests that the
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 4185–4196 | 4189

View Article Online
Paper
Dalton Transactions
Scheme 4 Co-dimerisation of ethylene and 1-butene via
a metallacycle
Fig. 6 Experimental and predicted mass spectrum of 1-butene produced from
C2H4/C2D4 with 15/MAO. Data normalised to mol. Ion m/z 60.
mechanism.
Fig. 7 Experimental and predicted mass spectrum of 1-hexene produced from
C2H4/C2D4 with 15/MAO.
work, we have tested the co-oligomerisation of 1 : 1 H2CvCH2
and D2CvCD2 with 15 in conjunction with MAO focusing on
the 1-butene and 1-hexene formed. In the case of 1-butene, the
theoretical mass spectrum resulting from a Cossee mechanism
Scheme 5 Co-dimerisation of ethylene and 1-butene via a Cossee mechanism.
branched C10 oligomers consisting of two isomers of 4-ethyl- and the experimentally observed ion distribution are shown in
octene and 5-methylnon-1-ene were detected; such species Fig. 6 (see ESI† for the derivation of the theoretical mass spec-
have previously been observed for other Ti and Cr catalysts and trum). A very good match between the theoretical and experi-
result from the co-trimerisation of ethylene/1-hexene via a mental distribution exists, indicating that the formation of
metallacycle mechanism.14,29,32,43 It is clear that this product 1-butene with this catalytic system occurs by a Cossee mechan-
distribution, which cannot be fully accounted for by either a ism. Significant over expression does result for ions m/z 64 and
Cossee or metallacycle mechanism, is highly suggestive of mul- 62, which is consistent with m/z for fully deuterated 1-butene
tiple modes of catalysis or multiple catalytically active species. and loss of one deuteron from this species. This has been con-
Indeed, such a hypothesis has been previously suggested firmed as a d8-1-butene impurity in the perdeuteroethylene.
for polymerisation catalysts including mono(salicylaldimi- Conversely, GC-MS analysis of the 1-hexene formed from the
nato),44,46 FI catalysts79 and other analogous systems80,81 in same experiment is shown in Fig. 7. Comparison of the theo-
combination with MAO.
retical mass spectrum for a metallacycle mechanism with the
Given these results we were interested in exploring the possi- experimentally observed data shows a strong correlation indi-
ble mechanisms involved in the formation of various oligo- cating that this oligomer most likely results from a metalla-
mers. The simplest method for probing oligomerisation cyclic mechanism (see ESI† for the derivation of the theoretical
catalysts involves the co-oligomerisation of a 1 : 1 mixture of mass spectrum). These results show that a myriad of catalytic
ethylene : perdeuteroethylene.9,19 Under these conditions a species can form from a well characterised precatalyst system
Cossee based mechanism will yield an isotopomer distribution upon activation, each yielding a different mode of catalysis
containing H/D scrambling as a result of the β-hydride elimi- and hence a different selectivity for ethylene conversion.
nation and product release. Conversely, no H/D transfer
between oligomers occurs in the metallacycle mechanism, and formed with
as such no H/D scrambled isotopomers are formed. In our Alphabutol-type catalytic system, and as we have shown that
As a significant amount of dimers and co-dimers were
a
distribution consistent with that of an
4190 | Dalton Trans., 2013, 42, 4185–4196
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
the mechanism of dimerisation of our system is the same, it in an MBraun glovebox unless otherwise stated. Solvents,
was of interest to see how 15 would compare under similar excluding dichloromethane, were purified by passage through
conditions used in the Alphabutol process (AlEt3 activation) an Innovative Technologies purification system and, where
and that if by changing the co-catalyst a more selective system appropriate, stored over a sodium mirror. Anhydrous dichloro-
could be realised. Testing of 15 with triethylaluminium co- methane was purchase from Sigma-Aldrich and stored over 3 Å
catalyst (Table 1, entry 6) gave high percentages of polymer molecular sieves. CP grade ethylene (BOC gases) was purified
(83%) and a poor productivity. The 17% yield of liquid phase by passage through a column of activated 3 Å molecular sieves
oligomers consisted of only 1-butene (85%) and co-dimers and alumina. NMR spectra were recorded on a Varian Mercury
(15%) and no higher fractions. In the light of this result it is Plus NMR spectrometer operating at 300 MHz (1H) or 75 MHz
plausible to suggest that in earlier runs the 1-butene and co- (13C) at room temperature. MAO was supplied by Ablemarle as
dimers most likely result from activation with the free tri- a 10% solution in toluene.
methylaluminium in MAO, yielding a species that dimerises by
Preparation of 2-(OH)C6H4CvN(cumyl) (1). Cumyl amine
a Cossee mechanism. The 1-hexene, and also the higher LAOs, (0.63 g, 4.67 mmol) taken up in 15 mL of diethyl ether and
which are not present in the absence of MAO, most likely stirred over molecular sieves. Salicylaldehyde (0.57 g,
result from activation with MAO and the probable formation of 4.67 mmol) in 15 mL of diethyl ether was added as a stream.
a cationic active species.
The reaction mixture was stirred overnight. Diethyl ether was
removed via canula filtration and the molecular sieves washed
with 2 × 8 mL of diethyl ether. The supernatant and washings
were combined and the volatiles removed in vacuo to yield a
Conclusions
A series of Ti(IV) complexes bearing either one or two phenoxy- yellow residue. The residue was recrystallised from 10 mL pet-
imine ligands have been prepared and characterised. X-ray roleum spirits at −20 °C to give the title compound as long
crystallography on a range of complexes has shown structural yellow needles which were dried under reduced pressure with
features in agreement with previously reported complexes.
an 87% yield (0.98 g, 4.07 mmol). 1H NMR (CD2Cl2,
Activation with MAO gives catalytic species with moderate 299.89 MHz): δ 13.99 (s, 1H, OH), 8.31 (s, 1H, CHvN),
activities for ethylene oligomerisation and polymerisation. 7.23–7.42 (m, 7H, aryl-H), 6.84–6.94 (m, 2H, aryl-H), 1.69
While it is clear that polyethylene formation is the dominant (s, 6H, C(CH3)2). 13C NMR (CD2Cl2, 75.41 MHz): δ 162.4
process, with over 95% polymer formation in all cases of MAO (CvN), 161.9 (C–OH), 147.3, 119.6 (aryl-Cipso), 117.3, 118.9,
activation, close examination of the short chained oligomers 126.5, 127.3, 128.9, 132.2, 132.6 (aryl-C), 63.1 (C(CH3)2), 30.0
reveals a variety of operative mechanisms. Complexes 11, 12, (C(CH3)2). MS (electron ionisation): m/z 239 [M]+.
14 and 15 all show Schulz–Flory distributions with reduced
Preparation of 2-(OH)C6H4CvN(CH2CH2Ph) (2). To 2-phen-
1-butene formation while complex 13 shows a similar distri- ethylamine (2.53 g, 20.7 mmol) was added salicylaldehyde
bution but significantly enriched in 1-hexene. We have (2.51 g, 20.7 mmol) at 120 °C with stirring. The resulting
suggested that this enrichment could result from a metallacyc- yellow liquid was stirred overnight and then extracted with
lic mechanism, aided by an arene–Ti interaction similar to 40 mL petroleum spirits. The volatiles were removed under
systems reported by Hessen and co-workers.
reduced pressure and the residue flash distilled under full
Complexes 11, 12, 14 and 15 also show a series of C6 pump vacuum at 190 °C (oil bath temperature) to give the title
isomers consistent with the co-dimerisation of ethylene and product as a yellow liquid, in 86% yield (4.03 g, 17.9 mmol),
1-butene. Deuterium labelling studies with complex 15 have which solidified upon standing. 1H NMR (CD2Cl2,
shown that the 1-butene formed by the catalyst most likely 299.89 MHz): δ 13.36 (s, 1H, O–H), 8.25 (s, 1H, NvCH),
results from a Cossee mechanism while the isotopomer dis- 7.19–7.34 (m, 7H, aryl-H), 6.85–6.92 (m, 2H, aryl-H), 3.84 (td,
tribution of 1-hexene from the same experiment supports a J = 1.2, 6.90 Hz, 2H, NCH2), 3.00 (t, J = 6.90 Hz, 2H, PhCH2).
metallacycle mechanism. These experiments clearly show that 13C NMR (C6D6, 75.41 MHz): δ 165.7 (CvN), 162.4 (C–OH),
even from a well characterised precatalyst species, upon acti- 140.0, 119.6 (aryl-Cipso), 117.9, 118.8, 126.9, 129.0, 129.5, 131.8,
vation, a range of catalytic species can form each with a 132.8 (aryl-C), 61.6 (NCH2), 37.8 (PhCH2). MS (electron ionis-
different mode of catalysis. These results show that certain ation): m/z 225 [M]+.
classes of titanium catalysts have a propensity for metallacyclic
Preparation of 2-(OH)-3-(t-butyl)C6H3CvN(cumyl) (3). To
oligomerisation, as has been established for chromium, but 3-t-butylsalicylaldehyde (0.46 g, 2.57 mmol) in 8 mL of diethyl
the key to selectivity is suppression of polymer formation. ether over 3 Å molecular sieves was added cumylamine (0.35 g,
However, it is still not well understood which factors control 2.57 mmol) in 8 mL of diethyl ether with stirring. The bright
the switch from selective oligomerisation to polymerisation.
yellow reaction mixture was stirred overnight. The resulting
solution was separated via canula filtration and the residue
extracted with 2 × 10 mL of diethyl ether. The organic portions
were combined and concentrated under reduced pressure to
yield a pale yellow solid. The product was purified by recrystal-
Experimental
General
All manipulations were performed under an atmosphere of lisation from hot petroleum spirits followed by flash distilla-
UHP argon (BOC gases) using standard Schlenk techniques or tion at 240 °C (oil bath temperature) under full vacuum to give
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 4185–4196 | 4191

View Article Online
Paper
Dalton Transactions
1
3 as a pale yellow solid in 68% yield (0.52 g, 1.74 mmol). H Attempts to obtain mass spectroscopy of the product yielded
NMR (CD2Cl2, 299.89 MHz): δ 14.60 (s, 1H, OH), 8.29 (s, 1H, only the hydrolysed species. 1H NMR (CD2Cl2, 299.89 MHz):
NvCH), 7.07–7.44 (m, 6H, aryl-H), 7.08 (dd, J = 1.80, 7.50 Hz, δ 8.53 (s, 1H, CHvN), 8.00 (dd, 1H, J = 1.8, 6.0 Hz, aryl-H),
1H, aryl-H), 6.79 (t, J = 7.50, 1H, aryl-H), 1.70 (s, 6H, C(CH3)2), 7.20–7.45 (m, 6H, aryl-H), 7.02 (t, 1H, J = 7.5 Hz, aryl-H), 6.83
1.42 (s, 9H, C(CH3)3). 13C NMR (CD2Cl2, 75.41 MHz): δ 163.2 (dd, 1H, J = 1.2, 6.9 Hz, aryl-H), 1.63 (s, 6H, C(CH3)2), 0.20 (s,
(NvCH), 119.2, 137.7, 147.2, 161.1 (aryl-Cipso), 118.0, 126.4, 9H, Si(CH3)3). 13C NMR (CD2Cl2, 75.41 MHz): δ 148.8, 155.7,
127.2, 128.8, 129.6, 130.4 (aryl-C), 62.7 (C(CH3)2), 35.1 162.5 (aryl-Cipso), 153.9 (CvN), 120.7, 122.2, 126.8, 127.5,
(C(CH3)3), 29.9 (C(CH3)2), 29.5 (C(CH3)3). MS (electron ioni- 128.6, 128.9, 131.9 (aryl-C), 63.3 (C(CH3)2), 30.0 (C(CH3)2), 0.4
sation): m/z 295 [M]+.
(Si(CH3)3).
Preparation of 2-(OH)-3-(t-butyl)C6H3CvN(CMe2CH2Ph)
Preparation of 2-(OSiMe3)C6H4CvN(CH2CH2Ph) (7). Com-
(4). To 3-t-butylsalicylaldehyde (1.06 g, 5.92 mmol) in 3 mL of pound 2 (4.03 g, 17.9 mmol) in 80 mL of petroleum spirits was
petroleum spirits was added 2-methyl-3-phenylpropyl-2-amine cooled to −95 °C with stirring and n-butyllithium (11.2 mL, 1.6
(0.88 g, 5.92 mmol) with stirring to generate a bright yellow M in hexanes, 17.9 mmol) was added dropwise. The white sus-
solution. The mixture was heated to 80 °C for 1 hour and then pension was allowed to return to room temperature and stirred
increased to 120 °C overnight. The bright yellow residue was for five hours. The supernatant was removed via canula fil-
extracted with 20 mL of petroleum spirits, concentrated under tration and the remaining off-white solid was washed with 2 ×
reduced pressure and purified by flash distillation at 216 °C 10 mL of petroleum spirits and dried briefly in vacuo. The
under full vacuum to give 4 as a bright yellow liquid in 72% white solid was taken up in 50 mL of tetrahydrofuran at 0 °C
yield (1.31 g, 4.25 mmol). 1H NMR (CD2Cl2, 299.89 MHz): and 5 mL trimethylsilyl chloride added. The mixture was
δ 14.76 (s, 1H, OH), 8.13 (s, 1H, NvCH), 7.04–7.32 (m, 7H, heated to 55 °C overnight. The volatiles were removed under
aryl-H), 6.77 (t, J = 8.10 Hz, 1H, aryl-H), 2.91 (s, 2H, CH2), 1.44 reduced pressure and the residue extracted with 3 × 10 mL of
(s, 9H, C(CH3)3), 1.32 (s, 6H, C(CH3)2). 13C NMR (CD2Cl2, petroleum spirits. The extracts were condensed and the
75.41 MHz): δ 161.7 (NvCH), 119.1, 137.7, 138.1, 161.3 (aryl- remaining residue was flash distilled under full pump vacuum
Cipso), 117.8, 126.7, 128.1, 129.4, 130.1, 131.1 (aryl-C), 60.4 at 220 °C (oil bath temperature) to give 7 as a colourless liquid
(C(CH3)2), 50.0 (CH2), 35.1 (C(CH3)3), 29.5 (C(CH3)3), 27.0 in 63% yield (3.35 g, 11.3 mmol). Attempts to obtain mass
(C(CH3)2). MS (electron ionisation): m/z 309 [M]+.
spectroscopy of the product yielded only the hydrolysed
1
Preparation of 2-(OH)-3-(t-butyl)C6H3CvN(CH2CH2Ph) (5). species. H NMR (CD2Cl2, 299.89 MHz): δ 8.53 (s, 1H, NvCH),
To 3-t-butylsalicylaldehyde (0.44 g, 2.46 mmol) was added 7.92 (dd, J = 1.80, 6.00 Hz, 1H, aryl-H), 7.19–7.33 (m, 6H, aryl-
2-phenethylamine (0.30 g, 2.46 mmol) with stirring at 120 °C. H), 7.01 (t, J = 7.80 Hz, 1H, aryl-H), 6.84 (dd, J = 1.20, 6.90 Hz,
The resulting bright yellow liquid was stirred overnight and 1H, aryl-H), 3.86 (td, J = 1.2, 7.20 Hz, 2H, NCH2), 3.00 (t, J =
then purified via flash distillation at 215 °C under full vacuum 7.20 Hz, 2H, PhCH2), 0.26 (s, 9H, Si(CH3)3). 13C NMR (CD2Cl2,
to give the title compound 5 in 73% yield (0.50 g, 1.78 mmol). 75.41 MHz): δ 140.9, 155.6, 165.7 (aryl-Cipso), 157.8 (NvCH),
1H NMR (CDCl3, 299.89 MHz): δ 13.58 (s, 1H, OH), 8.28 (s, 1H, 120.6, 122.2, 126.5, 127.8, 128.8, 129.5, 132.0 (aryl-C), 64.0
NvCH), 7.16–7.34 (m, 7H, aryl-H), 6.79 (t, J = 7.80, 1H, aryl-H), (NCH2), 38.1 (PhCH2), 0.47 (Si(CH3)3).
3.76 (t, J = 7.20 Hz, 2H, NCH2), 2.99 (t, J = 7.20 Hz, 2H,
Preparation of 2-(OSiMe3)-3-(t-butyl)C6H3CvN(cumyl) (8).
PhCH2), 1.41 (s, 9H, C(CH3)3). 13C NMR (CDCl3, 75.41 MHz): Compound 3 (1.13 g, 3.83 mmol) was taken up in 20 mL of
δ 166.3 (NvCH), 118.7, 126.7, 128.8, 129.1, 129.5, 131.0, petroleum spirits and cooled to −95 °C with stirring. n-Butyl-
160.6, 164.9 (aryl-C), 100.1 (NCH2), 37.1 (PhCH2), 35.1 lithium (2.4 mL, 1.6 M in hexanes, 3.83 mmol) was added
(C(CH3)3), 29.6 (C(CH3)3). MS (electrospray ionisation): m/z dropwise, upon complete addition the mixture was allowed to
382.2 [M]+.
return to room temperature and stirred for six hours. The
Preparation of 2-(OSiMe3)C6H4CvN(cumyl) (6). Compound resulting white solid was isolated by canula filtration and
1 (0.98 g, 4.07 mmol) was taken up in 40 mL of petroleum dried under reduced pressure. The solid was then taken up in
spirits and cooled to −95 °C with stirring. n-Butyllithium 15 mL of tetrahydrofuran and 2.6 mL of trimethylsilyl chloride
(2.6 mL, 1.6 M in hexanes, 4.07 mmol) was added dropwise. was added before heating the mixture to 60 °C overnight.
The mixture was allowed to return to room temperature and Upon cooling, the volatiles were removed in vacuo and the
stirred for four hours. The supernatant was removed via yellow residue was extracted with 20 mL of petroleum spirits to
canula filtration and the precipitated lithium salt washed with remove any lithium chloride. The organic phase was concen-
10 mL of petroleum spirits. The white solid was dried under trated and the resulting yellow residue recrystallised from
reduced pressure and then taken up in 30 mL of tetrahydro- 7 mL of petroleum spirits at −40 °C to give 8 as a pale yellow
furan. Trimethylsilyl chloride (5 mL) was added and the solid in 72% yield (1.01 g, 2.76 mmol). 1H NMR (CD2Cl2,
mixture heated to 55 °C overnight. The volatiles were removed 299.89 MHz): δ 8.50 (s, 1H, NvCH), 7.90 (dd, J = 2.01, 7.20 Hz,
in vacuo and the remaining residue extracted with 3 × 5 mL 1H, aryl-H), 7.21–7.44 (m, 6H, aryl-H), 6.95 (t, J = 7.20 Hz, aryl-
petroleum spirits. The resulting yellow solution was concen- H), 1.64 (s, 6H, C(CH3)2), 1.39 (s, 9H, (C(CH3)3), 0.22 (s, 9H, Si
trated and the resulting yellow residue purified by flash distil- (CH3)3). 13C NMR (CD2Cl2, 75.41 MHz): δ 129.4, 141.4, 149.1,
lation at 240 °C under full vacuum to give the title compound 154.8 (aryl-Cipso), 154.7 (NvCH), 121.6, 126.5, 126.6, 126.7,
as a colourless liquid in 52% yield (0.66 g, 2.12 mmol). 128.5, 129.9 (aryl-C), 63.2 (C(CH3)2), 35.2 (C(CH3)3), 30.8
4192 | Dalton Trans., 2013, 42, 4185–4196
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
(C(CH3)3), 30.1 (C(CH3)2), 1.98 (Si(CH3)3). MS (electron ioni- 1.20, 6.90, 2H, aryl-H), 2.10 (s, 12H, C(CH3)2). 13C NMR
sation): m/z 352 [M]+.
(CD2Cl2, 75.41 MHz): δ 115.2, 141.7, 170.5 (aryl-Cipso), 166.4
Preparation of 2-(OSiMe3)-3-(t-butyl)C6H3CvN(CMe2CH2Ph) (CvN), 121.2, 126.9, 127.3, 129.5, 129.8, 135.7, 140.4 (aryl-C),
(9). Compound 9 was prepared according to the procedure 66.1 (C(CH3)2), 29.6 (C(CH3)2), 0.4 (Si(CH3)3). Anal. calcd
outlined for compound 10, upon extraction with petroleum for C32H34N2O2TiCl4: C 57.51, N 4.19, H 5.13. Found: C 57.40,
spirits and removal of the volatiles in vacuo a white viscous N 3.93, H 5.43.
liquid was isolated and identified as the title compound. 52%
Preparation of complex 12. TiCl4(thf)2 (0.70 g, 2.10 mmol)
yield (0.85 g, 2.23 mmol). 1H NMR (CD2Cl2, 299.89 MHz): δ was taken up in 15 mL of dichloromethane and cooled to
8.27 (s, 1H, NvCH), 7.88 (dd, J = 1.80 Hz, 6.00 Hz, 1H, aryl-H), −95 °C with stirring. Compound 7 (0.62 g, 2.10 mmol) in
7.38 (dd, J = 1.80 Hz, 6.00 Hz, 1H, aryl-H), 7.10–7.22 (m, 5H, 15 mL of dichloromethane was added dropwise and upon
aryl-H), 6.96 (td, J = 0.90 Hz, 7.65 Hz, 1H, aryl-H), 2.91 (s, 2H, complete addition the resulting orange solution was allowed
CH2), 1.39 (s, 9H, C(CH3)3), 1.26 (s, 6H, C(CH2)2), 0.22 (s, 9H, to warm to room temperature and stirred overnight. The vola-
Si(CH3)3). 13C NMR (CD2Cl2, 75.41 MHz): δ 129.5, 139.2, 141.4, tiles were removed from the resulting deep red solution under
154.8 (aryl-Cipso), 153.0 (NvCH), 121.6, 126.3, 126.5, 127.9, reduced pressure to yield a dark red honeycomb which was
129.7, 131.2 (aryl-C), 61.3 (C(CH3)2), 50.5 (CH2), 35.1 (C(CH3)3), crushed and washed with 3 × 10 mL of petroleum spirits. The
30.8 (C(CH3)3), 27.3 ((C(CH3)2), 2.06 (Si(CH3)3). MS (electron resulting red solid was dried in vacuo for 3.5 hours. The solid
ionisation): m/z 382 [M − H]+.
was then taken up in 3 mL of tetrahydrofuran and petroleum
Preparation of 2-(OSiMe3)-3-(t-butyl)C6H3CvN(CH2CH2Ph) spirits added until precipitation began, the mixture was then
(10). Compound 5 (0.50 g, 1.78 mmol) was taken up in 6 mL cooled to −20 °C to generate a red oil. The mother liquor was
of petroleum spirits and cooled to −95 °C with vigorous stir- decanted, concentrated and upon standing generated the
ring. n-Butyllithium (1.12 mL, 1.6 M in hexanes, 1.78 mmol) title compound in 20% yield (0.12 g, 0.21 mmol) as a dark red
was added dropwise to generate a white suspension. The crystalline solid suitable for X-ray crystallography. 1H NMR
mixture was allowed to return to room temperature and stirred (CD2Cl2, 299.89 MHz): δ 7.87 (s, 2H, NvCH), 7.61 (t, J = 3.60
for two hours. The supernatant was removed via canula fil- Hz, 2H, aryl-H), 6.96–7.32 (m, 16H, aryl-H), 3.82 (m, 4H,
tration and the remaining white solid was dried in vacuo. The NCH2), 3.07 (m, 4H, PhCH2). 13C NMR (CD2Cl2, 75.41 MHz): δ
solid was taken up in 5 mL of tetrahydrofuran and 1.2 mL tri- 166.5 (NvCH), 117.0, 122.5, 124.3, 127.0, 128.9, 129.4, 129.9,
methylsilyl chloride added before heating to 55 °C overnight. 134.6, 136.8, 138.7 (aryl-C), 65.4 (NCH2), 37.6 (PhCH2). Anal.
The volatiles were removed under reduced pressure and the calcd for C30H28N2O2TiCl2: C 46.13, N 3.36, H 5.81. Found:
remaining yellow residue was extracted with 2 × 3 mL of pet- C 46.39, N 3.31, H 5.68.
roleum spirits. The organic extracts were concentrated and the
Preparation of complex 13. Compound
8
(0.04 g,
residue recrystallised at −80 °C from petroleum spirits to give 1.09 mmol) was taken up in 15 mL of tetrahydrofuran and
1
10 as a white solid in 32% yield (0.20 g, 0.57 mmol). H NMR added to TiCl4(THF)2 (0.36 g, 1.09 mmol) in 15 mL of tetra-
(CD2Cl2, 299.89 MHz): δ 8.48 (s, 1H, NvCH), 7.65 (d, J = 7.80 hydrofuran at −95 °C with vigorous stirring. The reaction
Hz, 1H, aryl-C), 7.38 (d, J = 7.80 Hz, 1H, aryl-C), 7.16–7.32 (m, mixture was allowed to return to room temperature and stirred
5H, aryl-C), 6.94 (t, J = 7.80 Hz, 1H, aryl-C), 3.85 (t, J = 7.50 Hz, overnight. The volatiles were removed under reduced pressure
2H, NCH2), 3.01 (t, J = 7.50 Hz, 2H, PhCH2), 1.39 (s, 9H, from the dark red reaction mixture to yield a dark red honey-
C(CH3)3), 0.25 (s, 9H, Si(CH3)3). 13C NMR (CD2Cl2, 75.41 MHz): comb. The honeycomb was washed with 15 mL of petroleum
δ 159.8 (NvCH), 121.7, 126.4, 128.7, 129.1, 129.4, 129.9, 140.8, spirits and then precipitated from tetrahydrofuran with pet-
141.4 (aryl-C), 63.7 (NCH2), 38.0 (PhCH2), 35.1 (C(CH3)3), 30.6 roleum spirits. The supernatant was decanted and the remain-
(C(CH3)3), 1.90 (Si(CH3)3). MS (electrospray ionisation): m/z ing dark red solid dried in vacuo to give 13 in 57% yield
354.3 [M]+.
(0.33 g, 0.63 mmol). Crystals suitable for X-ray diffraction were
Preparation of complex 11. Compound
6
(0.21 g, prepared by recrystallisation of 13 from hot a mixture of
0.66 mmol) was taken up in 6 mL of dichloromethane and petroleum spirits and tetrahydrofuran. 1H NMR (CD2Cl2,
added dropwise to TiCl4(thf)2 (0.22 g, 0.66 mmol) in 6 mL of 299.89 MHz): δ 8.22 (s, 1H, NvCH), 7.14–7.37 (m, 8H, aryl-H),
dichloromethane at −95 °C with stirring. The reaction mixture 3.92 (m, 4H, O(CH2)2(CH2)2), 2.08 (s, 6H, C(CH3)2), 1.87 (m,
was stirred overnight at room temperature. The volatiles were 4H, O(CH2)2(CH2)2), 1.52 (s, 9H, C(CH3)3). 13C NMR (CD2Cl2,
removed from the dark red solution at reduced pressure and 75.41 MHz): δ 169.4 (NvCH), 127.5, 137.4, 148.6, 168.0 (aryl-
the resulting red solid was washed with 2 × 10 mL of pet- Cipso), 125.7, 126.9, 128.1, 129.3, 133.5, 134.1 (aryl-C), 72.7
roleum spirits and 12 mL of dichloromethane. The remaining (C(CH3)2), 70.4 O(CH2)2(CH2)2), 35.4 (C(CH3)3), 31.3
red powder was dried in vacuo to give 11 in 26% yield (0.06 g, O(CH2)2(CH2)2), 29.6 (C(CH3)3), 25.9 (C(CH3)2). Anal. calcd
0.07 mmol). The complex was crystallised from a dichloro- for C24H32NO2TiCl3: C 55.36, N 2.69, H 6.19. Found: C 55.38,
methane solution by slow diffusion of petroleum spirits to give N 2.43, H 5.91.
large red needles suitable for X-ray diffraction. 1H NMR
Preparation of complex 14. To TiCl4(thf)2 (0.66 g,
(CD2Cl2, 299.89 MHz): δ 12.97 (br s, 2H, N+–H–O−), 8.03 (d, J = 1.99 mmol) in 15 mL of tetrahydrofuran was added dropwise
7.80 Hz, 2H, aryl-H), 7.84 (s, 2H, CHvN), 7.42–7.79 (m, 10H, compound 9 (0.76 g, 1.99 mmol) in 15 mL of tetrahydrofuran
aryl-H), 7.22 (dd, J = 1.80, 6.30 Hz, 2H, aryl-H), 7.00 (td, J = at −95 °C with vigorous stirring. The reaction was gradually
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 4185–4196 | 4193

View Article Online
Paper
Dalton Transactions
allowed to return to room temperature and stirred overnight. the reactor. During the reaction, the pressure was kept con-
The resulting dark red solution was concentrated and pet- stant with a replenishing flow of ethylene. After 30 minutes
roleum spirits added to precipitate a dark brown powder. The run time the replenishment of ethylene was ceased and the
powder was isolated by filtration and dried under reduced reactor cooled to <10 °C before venting of excess ethylene to
pressure. The dark brown powder was recrystallised from hot atmospheric pressure and injection of a weighed amount of
tetrahydrofuran/petroleum spirits to give the title product in nonane standard. We have previously72 established that negli-
35% yield (0.37 g, 0.69 mmol) as dark red crystals suitable for gible 1-butene is lost under these conditions; as such analysis
1
x-ray diffraction. H NMR (CD2Cl2, 299.89 MHz): δ 7.70 (s, 1H, of the vented ethylene for 1-butene is unnecessary. Residual
NvCH), 7.72 (dd, J = 1.50, 7.95 Hz, 1H, aryl-H), 7.14–7.29 (m, MAO was quenched with ∼10% HCl solution and samples of
6H, aryl-H), 7.00 (dd, J = 1.20, 7.65 Hz, 1H, aryl-H), 3.91 (m, the reaction mixture were taken for analysis and quantification
4H, O(CH2)2(CH2)2), 3.23 (s, 2H, CH2), 1.88 (m, 4H, of soluble analytes via GC and where appropriate GC-MS. Any
O(CH2)2(CH2)2), 1.76 (s, 6H, C(CH3)2), 1.52 (s, 9H, C(CH3)3). 13
C
polymer formed was removed via filtration and washed with
NMR (CD2Cl2, 75.41 MHz): δ 165.4 (NvCH), 125.9, 127.6, 10% HCl and methanol before drying at ∼60 °C for 3 days.
128.4, 128.7, 129.4, 131.0, 131.6, 133.5, 133.8, 136.9 (aryl-C), CH2CH2/CD2CD2 studies were performed via the method out-
70.3 (C(CH3)2), 68.2 (O(CH2)2(CH2)2), 48.5 (CH2), 35.5 lined above utilising
(C(CH3)3), 29.8 ((C(CH3)2), 29.2 (O(CH2)2(CH2)2), 26.0 deuteroethylene.
(C(CH3)3). Anal. calcd for C25H34NO2TiCl3: C 56.27, N 2.63, H
a
mixture of 1 : 1 ethylene–
X-ray crystallography
6.42. Found: C 56.12, N 2.61, H 6.37.
Preparation of complex 15. Compound 10 (0.37 g, Data for 11 were collected at −173 °C on crystals mounted on a
1.04 mmol) was taken up in 7 mL of dichloromethane and Hampton Scientific cryoloop at the MX2 beamline of the Aus-
added dropwise to TiCl4(thf)2 (0.35 g, 1.04 mmol) in 6 mL of tralian Synchrotron, while the remaining structures were col-
dichloromethane at −95 °C with vigorous stirring. The reaction lected on MX1.82 Data completeness is limited by the single
mixture slurry was allowed to return to room temperature and axis goniometer on the MX beamlines at the Australian Syn-
stirred overnight. The volatiles were removed under reduced chrotron. The structures were solved by direct methods with
pressure and the resulting dark red honeycomb was crushed SHELXS-97, refined using full-matrix least-squares routines
and washed with 3 × 15 mL portions of petroleum spirits. The against F2 with SHELXL-97,83 and visualised using X-SEED.84
residual solvent was removed in vacuo and the resulting red All non-hydrogen atoms were refined anisotropically. The
solid was taken up in 5.5 mL dichloromethane with one drop structure of 15 featured a conformational disorder in the THF
of tetrahydrofuran added. 4 mL of petroleum spirits was ring involving the remote carbon atoms that was modelled as a
added to precipitate a red solid which was isolated via canula two site complementary occupancy with the use of EADP/EXYZ
filtration and dried under reduced pressure to give 15 in 49% cards to assist in modelling of the associated hydrogen atom
yield (0.26 g, 0.51 mmol). Large red block crystals suitable for disorder that extends over most of the THF ring. Disordered
X-ray diffraction were grown by vapour diffusion of petroleum DCM lattice solvent was apparent for 11 that could not be ade-
spirits into a tetrahydrofuran solution of 15. Additionally, quately modelled and required the use of SQUEEZE to remove
layering of the supernatant with petroleum spirits yielded a its contribution. Details of the disorder modelling are provided
small quantity of dark red needles which were suitable for in the cif files and summarised in the figure captions.
X-ray diffraction and determined to be the bis-ligand complex Iminium and aldehyde protons were located and positionally
1
16. H NMR (CD2Cl2, 299.89 MHz): δ 7.95 (s, 1H, NvCH), 7.61 refined. All other hydrogen atoms were placed in calculated
(dd, J = 1.80, 5.70 Hz, 1H, aryl-H), 7.09–7.31 (m, 7H, aryl-H), positions and refined using a riding model with fixed C–H dis-
4.27 (t, J = 7.20 Hz, 2H, NCH2), 4.18 (bs, 4H, O(CH2)2(CH2)2), tances of 0.95 Å (sp2CH), 0.99 Å (CH2), 0.98 Å (CH3). The
3.20 (t, J = 7.20 Hz, 2H, PhCH2), 1.94 (bs, 4H, O(CH2)2(CH2)2), thermal parameters of all hydrogen atoms were estimated as
1.51 (s, 9H, C(CH3)3). 13C NMR (CD2Cl2, 75.41 MHz): δ 167.1 Uiso(H) = 1.2Ueq(C) except for CH3 where Uiso(H) = 1.5Ueq(C).
(NvCH), 129.5, 137.9, 138.9 (aryl-Cipso), 124.3, 127.1, 129.0, A summary of crystallographic data given below.
129.7, 132.7, 133.8 (aryl-C), 74.3 (NCH2), 65.9 (O(CH2)2(CH2)2),
Crystal data for 11: C32H34Cl4N2O2Ti, M = 668.31, trigonal,
37.8 (PhCH2), 35.3 (C(CH3)3, 29.7 (O(CH2)2(CH2)2), 25.9 a = 22.5011(13), c = 18.1552(13) Å, U = 7960.5(9) Å3, T = 100 K,
ˉ
(C(CH3)3). Anal. calcd for C23H30NO2TiCl3: C 54.52, N 2.76, space group R3 (no. 148), Z = 9, 56 709 reflections measured,
H 5.97. Found: C 54.57, N 2.61, H 5.95.
5600 unique (Rint = 0.0832), 4234 > 4σ(F), R = 0.0787 (observed),
Rw = 0.2290 (all data). Crystal data for 12: C30H28Cl2N2O2Ti,
M = 567.34, monoclinic, a = 7.1820(12), b = 25.1880(16), c =
Ethylene polymerisation/oligomerisation
A 0.3 L stainless steel Parr 5500 Compact Mini Reactor was pre- 14.674(4) Å, β = 90.750(3)°, U = 2654.3(8) Å3, T = 100 K, space
heated to 120 °C and flushed with four vacuum/argon cycles group P21/n (no. 14), Z = 4, 23 158 reflections measured, 3535
and two ethylene purges. The reactor was cooled to the appro- unique (Rint = 0.0887), 3227 > 4σ(F), R = 0.0696 (observed), Rw
=
priate temperature and charged with toluene (total solvent 0.1733 (all data). Crystal data for 13: C24H32Cl3NO2Ti, M =
volume of 50 mL). The catalyst (20 μmol) in 10 mL of toluene 520.76, orthorhombic, a = 28.069(3), b = 8.6610(12), c =
was then activated, in a Schlenk to observe any colour 20.8230(11) Å, U = 5062.2(9) Å3, T = 100 K, space group Pna21
changes, with 300 equivalents of MAO before injection into (no. 33), Z = 8, 80 972 reflections measured, 12 185 unique
4194 | Dalton Trans., 2013, 42, 4185–4196
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
(Rint = 0.0643), 11 468 > 4σ(F), R = 0.0957 (observed), Rw
0.2533 (all data). Crystal data for 14: C25H34Cl3NO2Ti, M =
=
10 A. N. J. Blok, P. H. M. Budzelaar and A. W. Gal, Organo-
metallics, 2003, 22, 2564.
534.78, orthorhombic, a = 25.105(3), b = 10.6280(12), c = 11 A. Brückner, J. K. Jabor, A. E. C. McConnell and
9.7480(15) Å, U = 2600.9(6) Å3, T = 100 K, space group Pna21
P. B. Webb, Organometallics, 2008, 27, 3849.
(no. 33), Z = 4, 42 641 reflections measured, 6238 unique (Rint 12 P. H. M. Budzelaar, Can. J. Chem., 2009, 87, 832.
= 0.0477), 5874 > 4σ(F), R = 0.0372 (observed), Rw = 0.0961 (all 13 T. J. M. de Bruin, L. Magna, P. Raybaud and H. Toulhoat,
data). Crystal data for 15: C23H30Cl3NO2Ti, M = 506.73, ortho-
rhombic, a = 20.373(3), b = 16.4210(15), c = 13.9580(16) Å, U = 14 L. H. Do, J. A. Labinger and J. E. Bercaw, Organometallics,
4669.6(9) Å3, T = 100 K, space group Pbca (no. 61), Z = 8, 50 366
2012, 31, 5143.
reflections measured, 3624 unique (Rint = 0.0899), 3310 > 4σ(F), 15 W. Janse van Rensberg, C. Grove, J. P. Steynberg,
Organometallics, 2003, 22, 3404.
R = 0.0786 (observed), Rw = 0.1741 (all data). Crystal data for
16: C38H44Cl2N2O2Ti, M = 679.55, monoclinic, a = 14.1400(10),
K. B. Stark, J. J. Huyser and P. J. Steynberg, Organometallics,
2004, 23, 1207.
b = 7.6160(10), c = 31.9980(19) Å, β = 100.824(2)°, U = 3384.6(5) 16 W. Janse van Rensberg, J.-A. van den Berg and
Å3, T = 100 K, space group P21/n (no. 14), Z = 4, 18 966
P. J. Steynberg, Organometallics, 2007, 26, 1000.
reflections measured, 5287 unique (Rint = 0.0330), 4501 > 4σ(F), 17 I. Y. Skobelev, V. N. Panchenko, O. Y. Lyakin,
R = 0.0507 (observed), Rw = 0.1203 (all data). Crystal data
for dichlorobis(3-tert-butyl-2-oxybenzoyl) titanium(IV):
K. P. Bryliakov, V. H. Zakharov and E. P. Talsi, Organometal-
lics, 2010, 29, 2943.
C22H26Cl2O4Ti, M = 473.23, monoclinic, a = 9.9030(9), b = 18 J. A. Suttil, D. S. McGuinness, M. Pichler, M. G. Gardiner,
22.683(2), c = 10.1360(10) Å, β = 90.781(2)°, U = 2276.6(4) Å3,
D. H. Morgan and S. J. Evans, Dalton Trans., 2012, 41, 6625.
T = 100 K, space group P21/n (no. 14), Z = 4, 30 459 reflections 19 T. Agapie, S. J. Schofer, J. A. Labinger and J. A. Bercaw,
measured, 4589 unique (Rint = 0.0559), 4390 > 4σ(F), R = 0.0340
(observed), Rw = 0.0913 (all data).
J. Am. Chem. Soc., 2004, 126, 1304.
20 S. Tobisch and T. Ziegler, Organometallics, 2003, 22,
5392.
21 J. M. Overett, K. Blann, A. Bollmann, J. T. Dixon,
D. Hassbroek, E. Killian, H. Maumela, D. S. McGuinness
and D. H. Morgan, J. Am. Chem. Soc., 2005, 127, 10723.
22 S. J. Schofer, M. W. Day, L. M. Henling, J. A. Labinger and
J. E. Bercaw, Organometallics, 2006, 25, 2743.
23 A. J. Rucklidge, D. S. McGuinness, R. P. Tooze,
A. M. Z. Slawin, J. D. A. Pelletier, M. J. Hanton and
P. B. Webb, Organometallics, 2007, 26, 2782.
24 A. Jabri, C. B. Mason, Y. Sim, S. Gambarotta, T. J. Burchell
and R. Duchateau, Angew. Chem., Int. Ed., 2008, 47, 9717.
25 I. Vidyarante, G. B. Nikiforov, S. I. Gorelsky, S. Gambarotta,
R. Duchateau and I. Korobkov, Angew. Chem., Int. Ed., 2009,
48, 6552.
Acknowledgements
The Australian Research Council is thanked for financial
support through grant FT100100609 to DSM and an Australian
Postgraduate Award to JAS. We thank Sasol Technology Ltd for
funding and permission to publish this work. Albemarle Asia
Pacific is thanked for donating the MAO solution. Noel Davies
and Thomas Rodemann of the UTas Central Science Labora-
tory are thanked for MS and elemental analysis measurements
respectively. Data for the structures of complexes 11–16 were
obtained on the MX1/MX2 beamlines at the Australian
Synchrotron, Victoria, Australia.
26 D. S. McGuinness, D. B. Brown, R. P. Tooze, F. M. Hess,
J. T. Dixon and A. M. Z. Slawin, Organometallics, 2006, 25,
3605.
27 A. Jabri, C. Temple, P. Crewdson, S. Gambarotta,
I. Korobkov and R. Duchateau, J. Am. Chem. Soc., 2006, 128,
9238.
Notes and references
1 A. W. Al-Sa’doun, Appl. Catal., A, 1993, 105, 1.
2 Y. Chauvin and H. Olivier, in Applied Homogeneous Catalysis 28 Y. Yang, Z. Liu, L. Zhong, P. Qiu, Q. Dong, R. Cheng,
with Organometallic Compounds, ed. B. Cornils and J. Vanderbilt and B. Liu, Organometallics, 2011, 30, 5297.
W. A. Herrmann, VCH, Weinheim, 1996, vol. 1, 29 P. J. W. Deckers, B. Hessen and J. H. Teuben, Angew. Chem.,
pp. 258–268. Int. Ed., 2001, 40, 2516.
3 J. T. Dixon, M. J. Green, F. M. Hess and D. H. Morgan, 30 C. Pellecchia, D. Pappalardo and G.-J. Gruter, Macro-
J. Organomet. Chem., 2004, 689, 3641.
molecules, 1999, 32, 4491.
4 D. S. McGuinness, Chem. Rev., 2011, 111, 2321.
5 P. W. N. M. van Leeuwen, N. D. Clément and
M. J.-L. Tschan, Coord. Chem. Rev., 2011, 255, 1499.
6 T. Agapie, Coord. Chem. Rev., 2011, 255, 861.
7 G. P. Belov, Pet. Chem., 2012, 52, 139.
8 D. F. Wass, Dalton Trans., 2007, 816.
9 T. Agapie, J. A. Labinger and J. A. Bercaw, J. Am. Chem. Soc.,
2007, 129, 14281.
31 C. Pellecchia, D. Pappalardo, L. Oliva, M. Mazzeo and
G.-J. Gruter, Macromolecules, 2000, 33, 2807.
32 P. J. W. Deckers, B. Hessen and J. H. Teuben, Organometal-
lics, 2002, 21, 5122.
33 H. Hagen, W. P. Kretschmer, F. R. van Buren, B. Hessen
and D. A. van Oeffelen, J. Mol. Catal. A: Chem., 2006, 248,
237.
34 B. Hessen, J. Mol. Catal. A: Chem., 2004, 213, 129.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 4185–4196 | 4195

View Article Online
Paper
Dalton Transactions
35 E. Otten, A. A. Batinas, A. Meetsma and B. Hessen, J. Am. 60 Y. Y. Long, Y. X. Wang, J. Y. Liu, X. F. Li and Y. S. Li,
Chem. Soc., 2009, 131, 5298. J. Polym. Sci., Part A: Polym. Chem., 2011, 47, 4626.
36 T. J. M. de Bruin, P. Raybaud and H. Toulhoat, Organome- 61 Y. Y. Long, W. P. Ye, P. Tao and Y. S. Li, Dalton Trans., 2011,
tallics, 2008, 27, 4864.
40, 1610.
37 H. Hagen, Ind. Eng. Chem. Res., 2006, 45, 3544.
38 C. Wang and J. Huang, Chin. J. Chem., 2006, 24, 1397.
39 J. Huang, T. Wu and Y. Qian, Chem. Commun., 2003, 2816.
40 T. Wu, Y. Qian and J. Huang, J. Mol. Catal. A: Chem., 2004,
214, 227.
41 J. A. Suttil, D. S. McGuinness and S. J. Evans, Dalton Trans.,
2010, 39, 5278.
42 M. Mitani, J. Saito, S. Ishii, Y. Nakayama, H. Makio,
62 Y. Y. Long, Y. X. Wang, J. Y. Liu, X. F. Li and Y. S. Li,
J. Polym. Sci., Part A: Polym. Chem., 2012, 50, 638.
63 J. Strauch, T. H. Warren, G. Erker, R. Frolich and
P. Saarenketo, Inorg. Chim. Acta, 2000, 300–302, 810.
64 V. A. Kogan, O. A. Osipov, V. I. Minkin and V. P. Sokolov,
Russ. J. Inorg. Chem., 1965, 10, 45.
65 V. P. Sokolov, V. A. Kogan, O. A. Osipov and
M. I. Rumyantseva, Zh. Obshch. Khim., 1968, 38, 2489.
N. Matsukawa, S. Matsui, J. Mohri, R. Furuyama, H. Terao, 66 D. Negoiu and A. Kriza, An. Univ. Bucuresti, Chim., 1970, 19,
H. Bando, H. Tanaka and T. Fujita, Chem. Rec., 2004, 4,
137.
113.
67 D. Negoiu and A. Kriza, An. Univ. Bucuresti, Chim., 1971, 20,
43 Y. Suzuki, S. Kinoshita, A. Shibahara, S. Ishii,
1971.
K. Kawamura, Y. Inoue and T. Fujita, Organometallics, 2010, 68 A. Sousa, F. de la Fuente, M. Gayoso and R. Bastida, An.
29, 2394. Quim., Ser. B: Quim. Inorg., 1982, 78, 46.
44 D. Owiny, S. Parkin and F. T. Ladipo, J. Organomet. Chem., 69 A. Pärssinen, T. Luhtanen, M. Klinga, T. Pakkanen,
2003, 678, 134. M. Leskelä and T. Repo, Organometallics, 2007, 26, 3690.
45 D. A. Pennington, D. L. Hughes, M. Bochmann and 70 S. Matsui, M. Mitani, J. Saito, Y. Tohi, H. Makio,
S. J. Lancaster, Dalton Trans., 2003, 3480.
46 D. A. Pennington, W. Clegg, S. J. Coles, R. W. Harrington,
M. B. Hursthouse, D. L. Hughes, M. E. Light,
N. Matsukawa, Y. Takagi, K. Tsuru, M. Nitabaru,
T. Nakano, H. Tanaka, N. Kashiwa and T. Fujita, J. Am.
Chem. Soc., 2001, 123, 6847.
M. Schormann, M. Bochmann and S. J. Lancaster, Dalton 71 R. Furuyama, J. Saito, S. Ishii, M. Mitani, S. Mitsui, Y. Tohi,
Trans., 2005, 561.
47 S. A. Pardhy, S. S. Deshpande, S. Gopinathan and
H. Makio, N. Matsukawa, H. Tanaka and T. Fujita, J. Mol.
Catal. A: Chem., 2003, 200, 31.
C. Gopinathan, Indian J. Chem., Sect. A: Inorg., Bio-inorg., 72 D. S. McGuinness, J. A. Suttil, M. G. Gardiner and
Phys., Theor. Anal. Chem., 1981, 21, 173. N. W. Davies, Organometallics, 2008, 27, 4238.
48 S. J. Coles, M. B. Hursthouse, D. G. Kelly and A. J. Toner, 73 R. Emrich, O. Heinemann, P. W. Jolly, C. Krüger and
Polyhedron, 2000, 19, 177. G. P. J. Verhovnik, Organometallics, 1997, 16, 1511.
49 M. Goyal, S. Mishra and A. Singh, Synth. React. Inorg. Met.- 74 J. X. McDermott, J. F. White and G. M. Whitesides, J. Am.
Org. Chem., 2001, 31, 1705. Chem. Soc., 1973, 95, 4451.
50 M. R. Salata and T. J. Marks, Macromolecules, 2009, 42, 75 J. X. McDermott, J. F. White and G. M. Whitesides, J. Am.
1920.
Chem. Soc., 1976, 98, 6521.
51 T. Xu, J. Coord. Chem., 2011, 64, 3652.
52 A. F. Mason and G. W. Coates, J. Am. Chem. Soc., 2004, 126,
10798.
76 J. A. Suttil and D. S. McGuinness, Organometallics, 2012,
31, 7004.
77 A. Bre, Y. Chauvin and D. Commereuc, Nouv. J. Chim.,
1986, 10, 535.
53 D. A. Pennington, S. J. Coles, M. B. Hursthouse,
M. Bochmann and S. J. Lancaster, Chem. Commun., 2005, 78 A. Bollmann, K. Blann, J. T. Dixon, F. M. Hess, E. Killian,
3150.
H. Maumela, D. S. McGuinness, D. H. Morgan, A. Neveling,
S. Otto, M. Overett, A. M. Z. Slawin, P. Wasserscheid and
S. Kuhlmann, J. Am. Chem. Soc., 2004, 126, 14712.
79 Y. Tohi, H. Makio, S. Matsui, M. Onda and T. Fujita, Macro-
molecules, 2003, 36, 523.
54 D. A. Pennington, S. J. Coles, M. B. Hursthouse,
M. Bochmann and S. J. Lancaster, Macromol. Rapid
Commun., 2006, 27, 599.
55 D. A. Pennington, R. W. Harrington, W. Clegg,
M. Bochmann and S. J. Lancaster, J. Organomet. Chem., 80 R. K. J. Bott, D. L. Hughes, M. Schormann, M. Bochmann
2006, 691, 3183. and S. J. Lancaster, J. Organomet. Chem., 2003, 665, 135.
56 K. P. Bryliakov, E. A. Kravtsov, L. Broomfield, E. P. Talsi and 81 D. Liguori, R. Centore, A. Tuzi, F. Grisi, I. Sessa and
M. Bochmann, Organometallics, 2007, 26, 288. A. Zambelli, Macromolecules, 2003, 36, 5451.
57 L. M. Broomfield, Y. Sarazin, J. A. Wright, D. L. Hughes, 82 T. M. McPhillips, S. E. McPhillips, H. J. Chiu, A. E. Cohen,
W. Clegg, R. W. Harrington and M. Bochmann, J. Organo-
met. Chem., 2007, 692, 4603.
58 Y. Nie, J. Sun, Z. Cheng and L. Wang, J. Wuhan Univ.
Technol., Mater. Sci. Ed., 2008, 821.
59 Y. Y. Long, W. P. Ye, X. C. Shi and Y. S. Li, J. Polym. Sci.,
Part A: Polym. Chem., 2009, 47, 6072.
A. M. Deacon, P. J. Ellis, E. Garman, A. Gonzalez,
N. K. Sauter, R. P. Phizackerly, S. M. Soltis and P. Kuhn,
Synchotron Radiat., 2009, 9, 401.
83 G. M. Sheldrick, SHELX97, Progreams for Crystal Structure
Analysis, Universität Göttingen, Germany, 1998.
84 L. J. Barbour, J. Supramol. Chem., 2001, 1, 189.
4196 | Dalton Trans., 2013, 42, 4185–4196
This journal is © The Royal Society of Chemistry 2013