Group 4 complexes of chelating dianionic [OSO] binaphtholate ligands; synthesis and alkene polymerisation catalysis

Article abstract of DOI:10.1002/ejic.200400174

Mono- and bis(ligand)titanium and -zirconium complexes of a sulfur-bridged binaphtholate [1,1′-S(2-HOC₁₀H₄-tBu ₂-3,6)₂] (H₂L^SN) have been synthesised and characterised. Complexes of the general form [M(L ^SN)X₂] (X = Cl, CH₂Ph) are prepared by protonolysis or metathesis. X-ray crystallographic structural characterisation of the dibenzyl complex [Zr(L^SN)(CH₂Ph)₂] is reported, as well as the structure of its partially oxidised derivative [Zr(L^SN)(CH₂Ph)(OCH₂Ph)]₂. Complexes of the form [M(L^SN)₂] are readily formed; these can be converted by comproportionation reactions into the mono-L^SN derivatives above. The X-ray crystal structure of a bis(sulfur-bridged biphenolate) is presented. The representative complexes have been tested as alkene polymerisation catalysts, in conjunction with a methylalumoxane cocatalyst, and for the first time a zirconium complex of a sulfur-bridged biphenolate-type ligand has been found to be active in the polymerisation of ethene. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejic.200400174

FULL PAPER
Group 4 Complexes of Chelating Dianionic [OSO] Binaphtholate Ligands;
Synthesis and Alkene Polymerisation Catalysis
Louise S. Natrajan,[a] Claire Wilson,[a] Jun Okuda,[b] and Polly L. Arnold*[a]
Keywords: Binaphthol ligands / Polymerisation catalysis / Organometallic / Group 4 elements / Zirconium / Titanium
Mono- and bis(ligand)titanium and -zirconium complexes of
sulfur-bridged binaphtholate [1,1Ј-S(2-HOC10H4-tBu2-
mono-LSN derivatives above. The X-ray crystal structure of a
bis(sulfur-bridged biphenolate) is presented. The represent-
ative complexes have been tested as alkene polymerisation
catalysts, in conjunction with a methylalumoxane cocatalyst,
and for the first time a zirconium complex of a sulfur-bridged
biphenolate-type ligand has been found to be active in the
polymerisation of ethene.
a
3,6)2] (H2LSN) have been synthesised and characterised.
Complexes of the general form [M(LSN)X2] (X = Cl, CH2Ph)
are prepared by protonolysis or metathesis. X-ray crystallo-
graphic structural characterisation of the dibenzyl complex
[Zr(LSN)(CH2Ph)2] is reported, as well as the structure of its
partially oxidised derivative [Zr(LSN)(CH2Ph)(OCH2Ph)]2.
Complexes of the form [M(LSN)2] are readily formed; these
can be converted by comproportionation reactions into the
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
1 Introduction
[Ti(tbmp)Me-(H2CϭCH2)]ϩ {tbmp ϭ 2,2Ј-thiobis(6-tert-
butyl-4-methylphenol), an [OSO] ligand} with respect to the
reactants.[10] Additionally, the calculations suggest that the
unsaturated backbone provided by the phenolate plays an
important role in stabilising the transition states. The ti-
tanium complexes [Ti(tbmp)X2] (X ϭ Cl, OiPr) exhibit high
catalytic activities for the polymerisation of ethene,[9] styr-
ene[11] and dienes[12] when activated with MAO, and can
also co-polymerise ethene and styrene.[13]
There are very few reports of the analogous zir-
conium() complexes. The zirconium derivatives are re-
portedly far more air- and moisture-sensitive than their ti-
tanium counterparts and react less cleanly.[2] In contrast
with zirconocene α-olefin polymerisation catalysts,[14] the
Zr[OSO] complexes reported so far exhibit no catalytic ac-
tivity.[2] We have recently introduced the use of the larger
naphtholate analogues of [OSO] biphenolates, [1,1Ј-S(2-
HOC10H4-tBu2-3,6)2] (H2LSN), in order to stabilise the first
sulfur-bridged bis(aryloxide) complexes of lanthanides;[15]
the thionaphthol [LSN] thus seemed better suited as a sup-
porting ancillary ligand for zirconium().
The continuing search for alternatives to cyclopen-
tadienyl-based ancillary ligand sets for Group 4 olefin pol-
ymerisation catalysts has produced a huge range of new po-
lydentate ligand systems, of which biphenolates have been
identified as good ligands.[1,2] However, aryloxides are not
generally considered to contribute sufficient electron den-
sity to be isolobal with the cyclopentadienyl ligand.[3] Bi-
phenol ligands in which the two phenol rings are joined in
the ortho position by a potential donor group, such as sul-
fur, can provide an increased stabilisation of reactive, elec-
tron-deficient metal centres.[4] A diverse range of complexes
that contain these atom-bridged biphenolates, where the co-
6
ordinating ligand set is denoted [OXO] (X ϭ S5, S2 , SO7,
P8 for example), have now been reported.
Group 4 complexes incorporating [OXO] ligands have
been identified as promising alkene polymerisation systems
in the form {M[OXO]2Cl2}/MAO (MAO ϭ methylalumox-
ane).[9] The bridging heteroatom, if coordinated, helps pro-
vide the stereochemically rigid framework essential for ster-
eoregular polymer formation.[4] Density functional calcu-
lations show that a bridging sulfur atom increases the pol-
ymerisation rate by destabilising the π-bound intermediate
Herein, we report the first structurally characterised zir-
conium() complexes supported by an [OSO] ligand set.
Comparative investigations between H2LSN and tbmp have
led us to synthesise and characterise a new family of Group
4 complexes of the general formula [M(LSN)X2] (X ϭ chlo-
ride, amide, alkyl; M ϭ Ti, Zr). In addition, their stability
and activity towards olefin polymerisation has been studied
and compared with that of relevant tbmp analogues.
[a]
School of Chemistry, The University of Nottingham,
University Park, Nottingham NG7 2RD, UK
Fax: (internat.) ϩ 44-115-951-3563
E-mail: Polly.Arnold@Nottingham.ac.uk
[b]
Institute of Inorganic Chemistry, Aachen University of
Technology (RWTH),
Prof.-Pirlet-Strasse 1, 52074 Aachen, Germany
3724
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200400174
Eur. J. Inorg. Chem. 2004, 3724Ϫ3732

Group 4 Complexes of Chelating Dianionic [OSO] Binaphtholate Ligands
FULL PAPER
1
2 Results and Discussion
High-temperature H NMR spectroscopy of [D8]toluene
solutions of 2 do not show any processes that render the
two aromatic rings of the ligands equivalent. Thus, besides
the structural rigidity of 2 over an accessible temperature
range, the thermal barrier to fac-cis/fac-trans isomerism is
high.
2.1 Halide Complexes Supported by LSN
2.1.1 Reaction of K2LSN with TiCl4·2THF
2.1.2 Reaction of H2LSN with TiCl4
Treatment of the solvent-free dipotassium salt [K2LSN
]
The reaction of [H2LSN] with neat TiCl4 in hexane or
toluene yields the THF-free complex 1a, which is antici-
pated to be dimeric in solution[20] (Scheme 1). If triethyl-
amine is added to precipitate the HCl by-product, only the
bis(ligand) complex is isolated.[2] However, if THF is added,
further conversion of 1 into 2 is slowed down but not elim-
inated completely, so that neither product is formed in good
yield. Analytically pure 2 is most readily obtained in a syn-
thetically useful yield of 70% by the protonolysis reaction
of [H2LSN] with [Ti(OEt)4] (vide infra).
with [TiCl4·2THF] in toluene results in the formation of the
monomeric complex [Ti(LSN)Cl2·THF] (1), albeit in poor
yield (18%). It is found that the bis(ligand) complex
[Ti(LSN)2] (2) is consistently formed by over-reaction,[16]
and is always initially present as one third of the product
mixture irrespective of the reaction stoichiometry em-
ployed (Scheme 1).
The formation of the bis(ligand) complexes analogous
to 2 in the reactions of chelating biphenols only becomes
significant when the addition of ligand to metal is acceler-
ated or conducted in the absence of a donating solvent.[16,17]
Although single crystals of 2 were obtained from a diethyl
ether solution at Ϫ30 °C, they were unsuitable for single-
crystal X-ray diffraction analysis due to rapid loss of lat-
tice solvent.
As an alternative route to 1, a stoichiometric reaction of
H2LSN with [Ti(NMe2)2Cl2] was undertaken. Again, moni-
1
toring the reaction by H NMR spectroscopy reveals the
existence of a relatively short-lived species assigned as
[Ti(LSN)(Cl)NMe2] (3) by comparison with related com-
plexes, but its decomposition into 2 precludes further
characterisation.
1
The H NMR spectrum of 2 reveals the presence of two
sets of aromatic signals for the ligand, indicating the pres-
ence of a single geometrical isomer out of the three possible
ones. The cis-facial conformation of ligands in 2 (C2), rather
than a trans-facial (Ci) or mer-trans (D2d) arrangement, af-
fords two sets of ligand resonances. This has been observed
2.1.3 Preparation of 1 by Comproportionation
The synthesis of
1 is best achieved by compro-
portionation: Treatment of 2 with 1 equiv. of [TiCl4·2THF]
in a 4:1 mixture of dichloromethane/tetrahydrofuran at Ϫ60
°C (Scheme 1) affords, after warming to ambient tempera-
ture, a dark brown solution from which 1 may be isolated
in 96% yield.
previously
in
the
bis[OOO]zirconium
complex
2-ox-
[Zr{(OC6H3tBu-2-Me-4-CH2-6)2)O}2] {[OOO]
ϭ
apropanediyl-1,3-bis(2-tert-butyl-4-methylphenol)},[17] and
in the biphenolato bis[OCO]Zr complex {[OCO] ϭ [2,2Ј-
methylenebis(6-tert-butyl-4-methylphenol)}.[18] In 2, this
apparent C2 symmetry is consistent with the coordination
of both sulfur atoms. In contrast, the structurally character-
ised bis[ONO]titanium() complex [Ti{(OC6H2Me2-
2.1.4 Reactions of [LSN] with ZrCl4·2THF
Treatment of stoichiometric solvent-free [K2LSN] with
2,4)2NPr}2] adopts the more familiar trans-mer confor- [ZrCl4·2THF] in toluene results in the formation of the
mation.[19]
colourless bis(ligand) complex [Zr(LSN)2·THF] (5) in an
Scheme 1. Reactions of H2LSN with TiCl4
Eur. J. Inorg. Chem. 2004, 3724Ϫ3732
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3725

L. S. Natrajan, C. Wilson, J. Okuda, P. L. Arnold
FULL PAPER
Scheme 2. Synthesis of ZrIVϪLSN adducts
isolated yield of 34% (Scheme 2). Single crystals of 5 suit-
able for X-ray diffraction were grown by slow cooling of a
saturated diethyl ether solution of 5 to Ϫ30 °C. The com-
proportionation of 5 and [ZrCl4·2THF] in THF affords the
desired complex [Zr(LSN)Cl2·THF] (4) in 53% yield.
(1)
The
protonolysis
reaction
of
H2LSN
with
[Zr(NMe2)2Cl2·2THF] yields the amido chloride 6 as an in-
termediate. Irrespective of the reaction solvent polarity, this
rapidly undergoes a ligand redistribution reaction to form
the amine-stabilised analogue of 5, [Zr(LSN)2(HNMe2)] (7),
as the major product [Equation (1)]. The coordinated mol-
ecule of HNMe2 is retained in the product even when the
reaction solvent is THF.
Figure 1. Thermal ellipsoid drawing of 5·Et2O (50% probability);
methyl groups and solvent of crystallisation removed for clarity
ure 1) confirms the expected connectivity. Selected in-
teratomic distances and angles are collected in Table 1.
The first significant feature of the molecular structure of
2.2 Single-Crystal X-ray Analysis of [Zr(LSN)2·THF] (5)
Cooling of a concentrated diethyl ether solution of 5 to
Ϫ30 °C affords single colourless blocks suitable for a X-ray 5 is the co-facial arrangement of the two binaphthol li-
diffraction study. The solid-state molecular structure (Fig- gands, which imparts a distorted capped octahedral ge-
˚
Table 1. Selected interatomic distances [A] and angles [°] for 5·Et2O and 9 and 10
5·Et2O
9
10
Zr(1)ϪO(1)
Zr(1)ϪO(2)
Zr(1)ϪO(3)
Zr(1)ϪS(1)
S(1)ϪC(21)
S(1)ϪC(11)
O(1)ϪC(12)
O(2)ϪC(22)
O(1)ϪZr(1)ϪO(2)
O(1)ϪZr(1)ϪO(3)
O(2)ϪZr(1)ϪO(3)
O(1)ϪZr(1)ϪS(1)
O(2)ϪZr(1)ϪS(1)
O(3)ϪZr(1)ϪS(1)
O(2)ϪZr(1)ϪS(1A)
O(3)ϪZr(1)ϪS(1A)
S(1)ϪZr(1)ϪS(1A)
O(2A)ϪZr(1)ϪS(1)
2.002(2)
2.087(2)
2.283(3)
2.7657(8)
1.784(3)
1.799(3)
1.340(3)
1.331(3)
93.71(8)
104.94(6)
75.59(5)
73.60(6)
71.63(6)
146.963(15)
137.09(6)
146.963(15)
66.07(3)
137.09(6)
Zr(1)ϪO(1)
Zr(1)ϪO(2)
Zr(1)ϪO(3)
Zr(1)ϪS(1)
Zr(1)ϪC(40)
Zr(1)ϪC(50)
Zr(1)ϪC(51)
S(1)ϪC(21)
O(1)ϪZr(1)ϪO(2)
O(1)ϪZr(1)ϪO(3)
O(2)ϪZr(1)ϪO(3)
O(1)ϪZr(1)ϪC(40)
O(2)ϪZr(1)ϪC(40)
O(1)ϪZr(1)ϪC(50)
O(2)ϪZr(1)ϪC(50)
C(40)ϪZr(1)ϪC(50)
O(1)ϪZr(1)ϪS(1)
O(2)ϪZr(1)ϪS(1)
O(3)ϪZr(1)ϪS(1)
1.981(2)
2.004(2)
2.259(2)
2.9248(7)
2.296(3)
2.312(3)
2.850(3)
1.789(3)
92.94(8)
166.14 (8)
83.54(8)
92.21(10)
99.02(10)
91.39(11)
140.56(10)
119.96(12)
70.41(6)
69.79(5)
95.84(6)
Zr(1)ϪO(1)
Zr(1)ϪO(2)
Zr(1)ϪO(5A)
Zr(1)ϪS(1)
Zr(1)ϪO(5)
Zr(1)ϪC(40)
Zr(1)ϪZr(1A)
S(1)ϪC(21)
O(1)ϪZr(1)ϪO(2)
O(2)ϪZr(1)ϪO(5A)
O(1)ϪZr(1)ϪO(5A)
O(2)ϪZr(1)ϪC(40)
O(2)ϪZr(1)ϪS(1)
O(1)ϪZr(1)ϪS(1)
O(5A)ϪZr(1)ϪS(1)
O(5)ϪZr(1)ϪS(1)
1.992(3)
1.986(3)
2.136(3)
2.8881(13)
2.166(3)
2.228(5)
3.4904(13)
1.790(5)
95.10(14)
91.63(14)
148.15(13)
97.88(17)
70.13(9)
70.91(9)
82.34(9)
91.14(9)
3726
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 3724Ϫ3732

Group 4 Complexes of Chelating Dianionic [OSO] Binaphtholate Ligands
FULL PAPER
ometry on the seven-coordinate zirconium centre. The four sive,[27,28] the dispersion contribution, which is substantial
naphtholate oxygen atoms describe the equatorial plane, for the large, polarisable sulfur atom, provides an important
while S(1) and S(1A) occupy approximate axial positions. attractive force that is able to compensate the repulsive elec-
The bound THF molecule resides in the apical site, opposite trostatic force. In 5, the hard metal ion binds a molecule of
and in between the two sulfur atoms. However, the THF, and the sulfurϪsulfur dipole interaction is preferred
O(1)ϪZr(1)ϪO(1A) and O(2)ϪZr(1)ϪO(2A) angles are not over the formation of closer dative bonds between the metal
linear,
and
are
150°
and
94°,
respectively. and the soft sulfur atoms.
NaphtholateϪzirconium bond lengths are in the range ex-
pected for zirconiumϪoxygen single bonds (observed range:
2.3 Alkyl Derivatives of LSN
[21]
˚
1.89Ϫ2.06 A).
The C(11)ϪS(1)ϪC(21) bond angle of
2.3.1 Dialkyltitanium Complexes
each thionaphthol ligand [109.6(1)°] is relatively large, and
the naphthol wings are opened out slightly, enabling the
sulfur atom to come into close proximity to the metal ion.
As a result, a sulfurϪzirconium interaction with a distance
Attempts
to
prepare
the
dibenzyl
complex
[Ti(LSN)(CH2Ph)2] by protonolysis of [Ti(CH2Ph)4] with
[H2LSN] resulted only in the recovery of starting materials.
This contrasts with the facile preparation of the biphenolate
complex [(tbmp)Ti(CH2Ph)2].[2] 1H NMR spectroscopy
suggests the formation of [Ti(LSN)(CH2Ph)2] (8) by metath-
esis between 1 and the Grignard reagent PhCH2MgCl, but
no tractable product could be isolated.
[22]
˚
of 2.7657(8) A is established.
The CϪSϪC dihedral angle between the two naphthol
planes of each ligand is 57°, and notably, the intramolecular
˚
SϪS distance is 3.015 A which is considerably shorter than
˚
the sum of the van der Waals radii of two S atoms (3.6 A),
˚
˚
and shorter than 3.33 A (a single SϪS bond plus 1.50 A)
Ϫ it was suggested that this value indicates a nonbonding
interaction between two atoms (Figure 2).[23] This implies a
significant intramolecular sulfur interaction in the solid
state. The nonlinearity of the O(1)ϪZr(1)ϪO(1A) and
O(2)ϪZr(1)ϪO(2A) moieties also suggests an SϪS interac-
tion that bends both fac ligands towards each other.
In an attempt to synthesise
a dimethyl complex
[Ti(LSN)Me2], a hexane solution of 1 was treated with 2
equiv. of LiMe.[29] However, even at temperatures below
Ϫ10 °C, the reaction mixture quickly became dark to yield
an intractable mixture of products. A one-pot reaction at
Ϫ80 °C of H2LSN with 4 equiv. of LiMe, followed by ad-
dition of TiCl4, also yielded no tractable product.
2.3.2 Dibenzylzirconium Complexes
In contrast to the tetrabenzyltitanium reaction, treatment
of 1 equiv. of [H2LSN] with tetrabenzylzirconium in THF
results in the rapid formation of the bis(ligand) THF ad-
duct 5. No intermediates were detectable (by solution NMR
spectroscopy) during the course of this reaction. This route
to 5 gives higher yields than metathesis reactions with zir-
conium tetrachloride.
Figure 2. Suggested SϪS (σϪπ)-(πϪπ) interaction in 5 (for clarity,
only the filled p orbital on each S atom is indicated); the dashed
lines represent the ZrϪS vector on the C2v plane of each binaphthol
ligand and the SϪS interaction
(2)
Nonbonded intramolecular sulfurϪsulfur interactions
are relatively rare but can be energetically favourable. For
example, the S4 moiety in bis[8-(phenylthio)naphthyl] 1,1Ј-
disulfide, which exhibits short nonbonded sulfur distances
A synthetic scale-up of the reaction (about 2.0 g) suffic-
iently reduces the extent of ligand redistribution to permit
the isolation of the dibenzyl complex 9 as a pale yellow
solid in moderate yield (47%) [Equation (2)]. This complex
is significantly less sensitive to light than the parent alkyl
compound, and if the reaction is carried out in the absence
of donating solvents, an intractable product mixture results.
˚
of 2.988(2) A, is aligned in a linear fashion displaying two
np(S)-σ*(SϪS) 3-c-4-e bonds.[24] Nonbonding S interac-
tions in this system are noted to have a greater influence on
the solid-state geometry in the naphthalene complexes than
in the corresponding phenyl disulfide. In the same way, the
p-type lone pair orbital of Se in the RϪSeϪRЈ fragments
(R ϶ RЈ ϭ alkyl/aryl) has also been shown to play an im-
portant role in intramolecular nonbonding interactions.[25]
Ab initio calculations and statistical studies indicate that
S···S close contacts in RϪSϪR fragments (R ϭ H, alkyl,
aryl) are attractive for a range of orientations (i.e. for a
range of dihedral angles between the C2v axis of each frag-
ment), and there is a maximum probability for a colinear
1
The H NMR spectroscopic resonances ([D6]benzene) for
the methylene benzyl moieties appear as a singlet at δ ϭ
2.67 ppm, rather than an AB quadruplet.[30] Low-tempera-
1
ture H NMR studies in [D8]toluene and CD2Cl2 reveal no
fluxional process slower than the NMR timescale that
would render the two methylene protons inequivalent.[31]
2.3.3 Single-Crystal X-ray Analysis of
[Zr(LSN)(CH2Ph)2·THF] (9)
S···S arrangement.[26] Although calculations show that at
The molecular structure of 9 is shown in Figure 3. It is
˚
short distances (3 A) the interactions are mainly repul- evident that the metal atom maintains a mononuclear en-
Eur. J. Inorg. Chem. 2004, 3724Ϫ3732
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3727

L. S. Natrajan, C. Wilson, J. Okuda, P. L. Arnold
FULL PAPER
vironment, and the bound THF molecule occupies an api- is formed, confirming that 10 derives from the reaction with
cal coordination site of a distorted octahedron. The ligand air; but while the yield is greatly improved for this reaction,
facially caps the zirconium atom, with relatively short naph- we were unable to obtain correct elemental analysis data
tholate oxygenϪzirconium distances of 1.981(2) and for this complex. Work is in progress to identify whether
˚
2.004(2) A for O(1) and O(2), respectively. The sulfur atom the oxidisable sulfur bridge is involved in this oxygen atom
insertion into the ZrϪC bond.
is geometrically placed in a trans arrangement to C(40) of
the η1-benzyl group, with
a
correspondingly long
˚
sulfurϪzirconium distance of 2.9248(7) A. As a result, the
mean planes of the two naphthalene rings of the ligands
form a dihedral angle of 78.0°, with an associated
C(11)ϪS(1)ϪC(21) bond angle of 106.3(1)°.
(3)
The molecular structure of 10 (Figure 4) is dinuclear and
unsolvated. The geometry at Zr is essentially octahedral
with a fac-LSN ligand, a small cone angle of the [OSO]
group and
a relatively long Zr(1)ϪS(1) contact of
˚
2.8881(13) A. The remaining benzyl group shows no ad-
ditional agostic interaction with the metal atom. The oxy-
genated benzyl group bridges the zirconium centre and its
symmetry-related equivalent in a nearly symmetrical fa-
shion with an average ZrϪO distance of 2.15 A, relative to
an average ZrϪOnaphtholate distance of 1.99 A, and a
˚
˚
˚
Figure 3. Molecular structure of 9; view along the 100 plane; ellip-
soids set at the 50% probability level; H atoms removed for clarity
ZrϪZrЈ distance of 3.490(1) A.
The two coordinated benzyl groups are clearly dis-
tinguishable in the solid state, and display η1- and η2-bond-
ing modes.[30] The η2-bound benzyl ring [C(50), (C51)] has
an acute Zr(1)ϪC(50)ϪC(51) angle of 95.3(2)°, in contrast
to that of the η1-benzyl group [Zr(1)ϪC(40)ϪC(41)
118.8(2)°]. The acuteness of this angle results in the close
proximity of the phenyl ring to the zirconium atom, and
in particular gives rise to a short ZrϪCipso interaction of
[30]
˚
2.850(3) A.
2.3.4 Reactivity of 9 with Air
Exposure of dibenzyl compound 9 to air affords three
compounds, which can be identified by 1H NMR spec-
troscopy [Equation (3)]. To date, we have been unable to
isolate the analytically pure major product by recrystallis-
ation, but were able to obtain single crystals suitable for
structure analysis to confirm the identity. The complex is
formally the product of the insertion of one oxygen atom
per metal atom into the ZrϪC bond of a benzyl group. The
controlled O atom insertion into a bound alkyl group at an
early transition metal is unusual,[32] and to the best of our
knowledge, this is the first time a benzyl group has been
selectively oxygenated. Usually, reaction with an oxygen
atom source such as pyridine N-oxide or trimethylamine N-
oxide, or air or water, forms oxo adducts containing the
Figure 4. Molecular structure of 10; ellipsoids set at the 50% prob-
ability level; solvent and Me groups removed for clarity
Interestingly, the ZrϪS distances in both [Zr(LSN)1X2]
ZrϪOϪZr unit.[33] Treatment of the dibenzyl complex 9 complexes are longer than those in the sterically more con-
with trimethylamine N-oxide shows that the same product gested [Zr(LSN)2] complex, giving further weight to the at-
3728
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 3724Ϫ3732

Group 4 Complexes of Chelating Dianionic [OSO] Binaphtholate Ligands
Table 2. Results for the polymerisation of ethene and propene
FULL PAPER
Activity[a]
Mw (ϫ 105)
Mn
Mw/Mn
[b]
[b]
[b]
Complex (mmol)
Yield [g]
Time [min]
1 [Ti(LSN)Cl2·THF] (0.027)[c]
4 [Zr(LSN)Cl2·THF] (0.026)[c]
9 [Zr(LSN)(CH2Ph)2·THF] (0.023)[c]
[(tbmp)TiCl2] (0.02)[d]
7.92
0.26
5.95
7.90
0.55
inactive
4.22
60
60
60
5
29
1
26
1580
220
2.04
8.36
0.22
3.5
46900
28000
4300
29000
8000
4.4
30
5.25
11.9
2.2
[(tbmp)TiCl2]2 (0.0025)[e]
[(tbmp)ZrCl2] (0.02)[f]
60
0.12
1 (0.027)[g]
60
15
0.02
510
4.1
[a]
[b]
[c]
Activities quoted in kg PE molϪ1[catalyst] barϪ1 hϪ1
.
PDI determined by GPC, results are an average of two runs. Conditions
[d]
used: 20 °C, 20 mg of catalyst, 1000 equiv. of MAO in 150 mL of toluene, 10 bar of ethene/propene, 1 h.
Results taken from ref.[2]
,
run at 40 °C, 3 bar of ethene, 500 equiv. of MAO in 200 mL of toluene. Result taken from ref.[20], run at 20 °C, 1 bar of ethene, 1000
[e]
equiv. of MAO in toluene. Results taken from ref.[2], run at 20 °C, 3 bar of ethene, 500 equiv. of MAO in 200 mL of toluene. 10 bar
[f]
[g]
of propene.
tractive SϪS interaction’s dominance of the local geometry isolated from the 1/MAO/ethene run suggest that this a
of the ligands in the bis(ligand) complex 5.
well-defined single-site catalyst. This is in contrast to the
higher PDI reported for the [(tbmp)TiCl2]/MAO system run
for a brief period of time.[2] The polyolefin from the 1/
MAO/propene run was isolated as a viscous oil, with a low
average molecular weight for the polymer. The molecular
weight distribution appears bimodal, suggesting an ill-de-
fined active catalyst species and/or the presence of more
than one active site. Again, a bimodal distribution is appar-
ent in the 4/MAO/ethene system.
2.4 Polymerisation of Ethene and Propene
Three suitable complexes were chosen for the polymeris-
ation studies, [Ti(LSN)Cl2·THF] (1), [Zr(LSN)Cl2·THF] (4)
and [Zr(LSN)(CH2Ph)2·THF] (9). The results are collated in
Table 2; for comparison, the analogous tbmp complexes
have been included. A significant exotherm was observed
during the first 10 min of each polymerisation run (except
4/MAO), attaining a maximum temperature of 50 °C. For
the remainder of each polymerisation run, the autoclave
temperatures remained almost constant (within 5 °C).
The titanium dichloride system 1/MAO and the bis(ben-
2.5 Conclusions
Titanium() and zirconium() complexes supported by
zyl)zirconium complex 9/MAO exhibit moderate activities the binaphtholate LSN dianion have been synthesised and
for the polymerisation of ethene and produce polyethene structurally characterised. The formation of bis(LSN) com-
with high molecular weights, whereas the corresponding zir- plexes renders the direct syntheses of titanium() and zir-
conium chloride system 4/MAO shows a much lower pro- conium() dichloride complexes difficult, with possible
pensity to polymerise ethene. Under identical conditions, 1/ further stabilisation of the zbis(LSN)irconium complex by
MAO also shows moderate activity for the polymerisation an SϪS interaction. However, unlike aminebis(phenolate)
of propene. Theory suggests that the Ti-based catalysts are [ONO] derivatives of Group 4 metals, the bis[OSO] com-
more active than their zirconium counterparts due to a plexes undergo disproportionation reactions to form heter-
slightly lower activation energy barrier;[10] the complete in- oleptic adducts.
activity of the (tmbp)Zr species is therefore surprising.
The ethene polymerisation activity of 1/MAO (28.94 kg plexes have been isolated. The bis(OSO)zirconium THF ad-
PE molϪ1[Ti] barϪ1 Ϫ1) compares well with the dimeric duct exhibits a co-facial arrangement of the naphthol li-
complex [(tbmp)TiCl2]2, which was tested at a lower catalyst gands, a distorted capped octahedral geometry at Zr and
concentration (220 kg PE mol[Ti]Ϫ1 barϪ1 Ϫ1).[20] How- close SϪS distances, which suggest an unusual noncovalent
The first crystallographically characterised Zr[OSO] com-
h
h
ever, the binaphthol-based catalyst remains an order of S···S interaction in the solid state. The zirconium [OSO]
magnitude less active than the corresponding biphenol sys- complexes are kinetically more labile than related aminebis-
tem. Theoretical studies also suggested that the catalyst ac- (phenolate) [ONO] adducts, but less labile than their ti-
tivity is influenced by the extent of conjugation in the li- tanium counterparts, and heteroleptic complexes are easier
gand backbone and that a CϭC bridged system should be to isolate. The dibenzylzirconium complex reacts with air
more active than a CϪC system.[2,10] The naphthol bridge to insert an O atom rather than decompose to a µ-oxodizir-
in these systems, which has a lower aromaticity than a phe- conium complex. Preliminary reactions with the oxygen
nol, supports this argument.
atom source trimethylamine N-oxide show that the same
However, the Zr naphtholate systems are better catalysts product is formed. Structural studies of the complexes have
than their phenolate counterparts. In view of the inactivity shown that the smaller Group 4 metal cations accommo-
of the reported Zr-based biphenolates towards ethene pol- date the symmetrical fac conformation of the LSN ligand
ymerisation, these results are encouraging. Narrow molecu- much more readily than the twisted conformation observed
lar weight distributions and polydispersities for the sample in the corresponding lanthanide chemistry.[15,34]
Eur. J. Inorg. Chem. 2004, 3724Ϫ3732
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3729

L. S. Natrajan, C. Wilson, J. Okuda, P. L. Arnold
FULL PAPER
1
[(LSN)Ti(µ-Cl)Cl]2 (1a): H NMR/C6D6: δ ϭ 8.61 (br., 1 H, 8-H),
The complexes 1 and 9 are effective ethene polymeris-
ation pre-catalysts, whereas the dichloride 4 displays a sur-
prisingly low activity towards ethene polymerisation. The
catalyst system 1/MAO is also moderately active for the
polymerisation of propene.
7.38 (m, 3H 4-H, 5-H, 7-H), 1.48 (s, 9 H, 3-tBu), 1.18 (s, 9 H, 6-
tBu) ppm.
[(LSN)2Ti] (2): A solution of H2LSN in THF (0.2 g, 0.37 mmol, 15
mL) was added dropwise to a cooled solution of Ti(OEt)4 in THF
(19.0 µL, 0.185 mmol, 10 mL) at Ϫ30 °C over a period of 20 min,
and the mixture was warmed to room temperature over 12 h, by
which time 2 had precipitated from the reaction mixture. The
supernatant was decanted from the precipitate, and the crude pow-
der washed with 3 ϫ 10 mL cold hexane to yield analytically pure
3. Experimental Section
1
2 in 70% yield (0.146 g). H NMR (C6D6): δ ϭ 8.93 (br., 1 H, 8-
General Procedures: All experimental procedures were performed
under dry, oxygen-free argon or dinitrogen, with standard Schlenk
techniques (mechanical pump for vacuum 10Ϫ4 mbar). Subsequent
manipulations of isolated compounds were carried out in a glove
box (MBraun Unilab) under dry dinitrogen. All 1H and 13C NMR
spectra were recorded with a Bruker DPX 300 spectrometer, op-
erating frequency variable temperature unit at 300 K, unless other-
wise stated. Chemical shifts are reported in parts per million and
referenced to residual proton resonances in [D6]benzene calibrated
against external TMS. Mass spectra (EI and FAB) were run by
Dr. Ali Abdul-Sada with a Kratos MS80RF instrument. Elemental
analyses were determined by Dr. Stephen Boyer at London Metro-
politan University and by Mr. Trevor Spencer at the University of
Nottingham. Polymerisation runs were conducted in a purpose-
built rig in the laboratory of Professor F. G. N. Cloke, at the Univer-
sity of Sussex, and GPC analysis was performed by Rapra Technol-
ogy Ltd. A glass autoclave with a magnetically coupled stirrer was
charged with a toluene solution of each complex, 1000 equiv. of
MAO, and the vessel pressurised at 20 °C with 10 bar of ethene or
10 bar of propene for 1 h. Samples from the 1/MAO/ethene and 4/
MAO/ethene were found to contain a significant amount of insol-
uble (in 1,2,3-trichlorobenzene) material and were difficult to filter.
The chromatogram peak area for 4/MAO/ethene is less than 20%
that of the 1/MAO/ethene and 9/MAO/ethene samples, suggesting
that the bulk of this sample remained insoluble. The insolubility of
the polymer could be due to an exceptionally high molecular weight
or a high degree of cross-linking. All solvents used (toluene, THF,
diethyl ether, hexane, dichloromethane) were either degassed and
purified by passage through activated alumina towers prior to use,
or were freshly distilled from the appropriate alkali metal under
dinitrogen, and thoroughly degassed prior to use. NMR spectro-
scopic grade [D6]benzene and [D8]toluene were refluxed over pot-
assium metal, and CD2Cl2 refluxed over CaH2 and vacuum-dis-
tilled before use. All chemicals were either obtained from Aldrich
Chemicals, except nBuLi and sulfur dichloride (Acros), or were
synthesised according to standard literature procedures.[35] Litera-
ture procedures for the preparation of [1,1Ј-S(2-HOC10H4tBu2-
H), 8.87 (br., 1 H, 8-H), 7.59 (m, 6 H, 4-H, 5-H, 7-H), 1.67 (s, 9
H, 3-tBu), 1.59 (s, 9 H, 3-tBu), 1.23 (s, 9 H, 6-tBu) ppm. EI-MS:
m/z (%)ϭ 1128 (24) [Mϩ]. C72H88O4S2Ti (1128): calcd. C 76.56, H
7.85; found C 76.48, H 7.80.
[(LSN)Ti(µ-NMe2)Cl]2 (3a): 1H NMR (C6D6): δ ϭ 8.48 (d, 3JH,H ϭ
8.82 Hz, 1 H, 8-H), 7.58 (m, 3H 4-H, 5-H, 7-H), 1.78 (s, 3 H,
NMe2), 1.16 (s, 9 H, 3-tBu), 1.31 (s, 9 H, 6-tBu) ppm.
3
[(LSN)Ti(µ-Cl)NMe2]2 (3b): 1H NMR: δ ϭ 8.79 (d, JH,H
ϭ
8.70 Hz, 1 H, 8-H), 7.58 (m, 3H 4-H, 5-H, 7-H), 1.78 (s, 3 H,
NMe2), 1.26 (s, 9 H, 3-tBu), 1.71 (s, 9 H, 6-tBu) ppm.
[(LSN)ZrCl2·THF] (4): A slurry of ZrCl4·2THF in toluene (21 mg,
0.056 mmol, 7 mL) was added dropwise over a period of 20 min to
a solution of 5 in toluene (70 mg, 0.056 mmol, 7 mL),which was
cooled to Ϫ60 °C. The resultant solution was allowed to slowly
attain room temperature over 4 h and stirred at room temperature
for a further 10 h. All volatiles were removed under reduced press-
ure, and the crude product extracted into hexane (15 mL). Cooling
to Ϫ30 °C afforded a colourless precipitate that was isolated and
washed with cold hexane (3 ϫ 5 mL) to give analytically pure 4 in
53% yield (0.021 g). 1H NMR (C6D6): δ ϭ 8.83 (br., 1 H, 8-H),
7.58 (m, 3 H, 4-H, 5-H, 7-H), 3.70 (br., 4 H,CH2 THF), 1.58 (br.,
9 H,3-tBu), 1.35 (br., 4 H, CH2 THF), 1.23 (s, 9 H, 6-tBu) ppm.
C36H44Cl2O2SZr·THF (775.1): calcd. C 61.98, H 6.76; found C
61.47, H 7.03.
[(LSN)2Zr·THF] (5): In the absence of light, a solution of H2LSN in
THF (500 mg, 0.92 mmol, 20 mL) was added dropwise to a solu-
tion of Zr(CH2Ph)4 in THF (210 mg, 0.46 mmol, 15 mL), which
was cooled to Ϫ60 °C over 30 min. Upon completion of the ad-
dition of the ligand, the reaction mixture was slowly warmed to
room temperature and stirred at ambient temperature for a further
10 h. All volatiles were removed under reduced pressure, and the
crude solid extracted into 5 mL of diethyl ether. Subsequent cool-
ing to Ϫ30 °C afforded single crystals suitable for X-ray diffraction
analysis in 45% yield (258 mg). 1H NMR (C6D6): δ ϭ 8.80 (d,
3JH,H ϭ 8.97 Hz, 1 H, 8-H), 7.64 (s, 1 H, 4-H), 7.57 (s, 1 H, 5-H),
3,6)2] (H2LSN [15]
(K2LSN [36]
have previously been described.
)
and [1,1Ј-S(2-KOC10H4tBu2-3,6)2(THF)2]
)
3
7.11 (d, JH,H ϭ 8.97 Hz, 1 H, 7-H), 4.10 (br., 2 H, CH2 THF),
[(LSN)TiCl2·THF] (1): A solution of TiCl4·2THF in THF (30 mg,
0.09 mmol, 10 mL) was added dropwise to a solution of 2 in a 1:1
mixture of THF/dichloromethane (0.1 g, 0.09 mmol, 30 mL) cooled
to Ϫ78 °C. The reaction mixture was left to warm slowly to ambi-
ent temperature, upon which it changed colour from orange to dark
brown. After 12 h, all volatiles were removed under reduced press-
ure. The crude product was extracted into hexane and cooled to
Ϫ30 °C. Analytically pure 1 was isolated as a microcrystalline
1.57 (s, 9 H, 3-tBu), 1.36 (br., 2 H, CH2 THF) 1.23 (s, 9 H, 6-tBu)
ppm. C72H88O4S2Zr·THF (1245.0): calcd. C 73.37, H 7.77; found
C 73.19, H 7.54.
1
[(LSN)Zr(µ-NMe2)Cl]2 (6a): H NMR (C6D6): δ ϭ 8.81 (br., 2 H,
8-H), 7.56 (m, 6 H, 4-H, 5-H, 7-H), 3.85 (s, 3 H, µ-NMe2), 1.66 (s,
18 H, 3-tBu), 1.23 (s, 18 H, 6-tBu) ppm.
[(LSN)Zr(µ-Cl)NMe2]2 (6b): 1H NMR (C6D6): δ ϭ 8.76 (d, 3JH,H ϭ
8.76 Hz, 2 H, 8-H), 7.56 (m, 6 H, 4-H, 5-H, 7-H), 1.61 (s, 3 H,
NMe2), 1.58 (s, 18 H, 3-tBu), 1.19 (s, 18 H, 6-tBu) ppm.
1
brown solid in 96% yield (62 mg). H NMR (C6D6): δ ϭ 8.66 (d,
3JH,H ϭ 8.85 Hz, 1 H, 8-H), 7.51 (s, 3H 4-H, 5-H, 7-H), 3.81 (br.,
2 H, CH2 THF), 1.62 (s, 9 H, 3-tBu), 1.32 (br., 2 H, CH2 THF),
1.19 (s, 9 H, 6-tBu) ppm. C36H44Cl2O2STi·THF (731.71): calcd. C [(LSN)2Zr·HNMe2] (7): 1H NMR (C6D6): δ ϭ 8.67 (br., 2 H, 8-H,),
3
65.66, H 7.16; found C 65.67, H 7.39.
7.59 (s, 2 H, 4-H), 7.55 (br., 2 H, 7 H), 7.44 (d, JH,H ϭ 8.74 Hz,
3730
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 3724Ϫ3732

Group 4 Complexes of Chelating Dianionic [OSO] Binaphtholate Ligands
FULL PAPER
2 H, 7-H), 2.66 (br., 3 H, HNMe2) 1.53 (s, 18 H, 3-tBu), 1.23 (s,
18 H, 6-tBu) ppm.
of 10 suitable for single-crystal X-ray diffraction were isolated. b)
Under N2, a glass vial was charged with a [D6]benzene solution of
9 (20 mg, 22.6 µmol), and a [D6]benzene solution of Me3NO
(1.6 mg, 22.6 µmol). The resultant mixture was filtered through
glass wool into a Young’s PTFE tap equipped NMR tube, and a
1H NMR spectrum recorded immediately. The reaction was further
monitored by 1H NMR spectroscopy. t ϭ 0 h: The 1H NMR spec-
troscopic resonances of 9 have completely given way to a new tran-
sient species (not identified) and 10 in a 1:1 ratio. 1H NMR (C6D6)
3
[(LSN)Ti(CH2Ph)2] (8): 1H NMR (C6D6): δH ϭ 8.69 (d, JH,H
ϭ
9.03 Hz, 1 H, 8-H), 7.51 (m, 3H 4-H, 5-H, 7-H), 7.05 (m, 5 H,
CH2Ph) 2.10 (s, 2 H, CH2Ph), 1.62 (s, 9 H, 3-tBu), 1.17 (s, 9 H, 6-
tBu) ppm.
[(LSN)Zr(CH2Ph)2·thf] (9): A solution of H2LSN in THF (2.0 g,
3.69 mmol, 25 mL) was added dropwise to
a solution of
3
of 10: δ ϭ 8.98 (d, JH,H ϭ 8.46 Hz, 1 H, 8-H), 7.61 (m, 9 H, Ar),
Zr(CH2Ph)4 in THF (1.68 g, 3.69 mmol, 15 mL), which was cooled
to Ϫ60 °C over 45 min. Upon completion of addition of the ligand,
the reaction mixture was slowly warmed to room temperature and
stirred for a further 15 h. All volatiles were removed under reduced
pressure, and the crude solid extracted into 25 mL of diethyl ether.
Cooling of the solution to Ϫ30 °C resulted in the precipitation of
analytically pure 10 as a yellow powder, which was washed with 3
ϫ 5 mL cold diethyl ether, and isolated in 47% yield (1.54 g). Single
crystals suitable for X-ray diffraction studies were grown from slow
concentration of the supernatant at Ϫ30 °C. 1H NMR (C6D6): δ ϭ
3.59 (br., 1 H, OCH2), 3.25 (br., 1 H, CH2), 1.73 (s, 9 H, 3-tBu),
1.26 (s, 9 H, 6-tBu) ppm. C50H58O3SZr (830.3): calcd. C 72.32, H
7.04; found C 67.01, H 7.96. Yield, isolated from a mixture with 9
1
3
(42 mg, 1.4%). H NMR (C6D6) of transient: δ ϭ 9.16 (d, JH,H ϭ
8.94 Hz, 1 H, 8-H), 7.41 (m, 9 H, Ar), 2.94 (br., 1 H, CH2), 1.68
1
(s, 9 H, 3-tBu), 1.45 (s, 9 H, 6-tBu) ppm. t ϭ 1 h: The H NMR
spectroscopic resonances due to 10 have grown considerably, while
those for the transient species have decreased to a ratio of 5:2.
t ϭ 24 h: 1H NMR spectrum contains a few more resonances of
insignificant integral value but is essentially unchanged. t ϭ 168 h:
3
8.94 (d, JH,H ϭ 8.76 Hz, 1 H, 8-H), 7.62 (s, 1 H, 4-H), 7.60 (s, 1
1
No change in H NMR spectrum.
H, 7-H), 7.57 (s, 1 H, 5-H), 7.55 (m, 2 H, CH2Ph), 6.87 (m, 3 H,
CH2Ph), 3.53 (br., 2 H, CH2 THF), 2.67 (s, 2 H, CH2Ph), 1.56 (s,
9 H, 3-tBu), 1.24 (s, 9 H, 6-tBu), 1.01 (br., 2 H, CH2 THF) ppm.
13C NMR (C6D6): δ ϭ 166.68, 130.00, 129.67, 129.28 (quaternary
C), 145.49, 140.10, 137.85, 133.39, 127.68, 126.17, 125.65 (CH),
65.48 (CH2Ph) 31.34, 30.08 (tBu), 73.18, 25.05 (THF) ppm.
C50H54O2SZr·thf (882.4): calcd. C 73.22, H 7.46; found C 73.49,
H 7.69.
Crystallographic Data: X-ray data (Table 3) were collected using
˚
Mo-Kα radiation (λ ϭ 0.71073 A) with a Bruker SMART1000
CCD area detector diffractometer using ω scans. Structure solution
and refinement was carried out with the SHELX suite of pro-
grams.[37] CCDC-217721 (5), -217722 (9) and -217723 (10) contain
the supplementary crystallographic data for this paper. These data
can be obtained free of charge at www.ccdc.cam.ac.uk/conts/
retrieving.html [or from the Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: (internat.)
ϩ 44-1223-336-033; E-mail: deposit@ccdc.cam.ac.uk].
[(LSN)Zr(OCH2Ph)(CH2Ph)2] (10). a) A Schlenk tube containing
powdered 9 was exposed to air for 2 min. The resulting colourless
powder was recrystallised from diethyl ether, from which crystals
Table 3. Crystallographic data and structure refinement for 5·Et2O and 9 and 10
Empirical formula
Formula mass
C80H106O6S2Zr (5·Et2O)
1319.07
150(2)
C54H66O3SZr (9)
886.35
293(2)
C54H68O4SZr (10)
1808.73
150(2)
Temperature [K]
˚
Wavelength [A]
0.71073
0.71073
0.71073
Crystal system
Space group
Unit cell dimensions
monoclinic
C2/c
a ϭ 33.175(2) A
b ϭ 11.2138(8) A
c ϭ 21.746(2) A
monoclinic
P21/n
a ϭ 16.5541(11) A
b ϭ 11.1071(8) A
c ϭ 25.9090(18) A
monoclinic
P21/n
a ϭ 14.815(3) A
b ϭ 21.612(4) A
c ϭ 16.183(3) A
˚
˚
˚
˚
˚
˚
˚
˚
˚
β ϭ 105.1690(10)°
7808.0(10)
4
1.137
0.242
β ϭ 91.2190(10)°
β ϭ 108.630(4)°
4910.1(15)
2
1.223
0.309
3
˚
Volume [A ]
Z
4762.8(6)
4
Dcalcd. [Mg/m3]
Absorption coefficient [mmϪ1
F(000)
1.236
0.316
1880
]
2856
1920
Crystal size [mm]
θ range for data collection [°]
Index ranges
0.27 ϫ 0.20 ϫ 0.12
1.92Ϫ26.37
Ϫ42 Յ h Յ 44
Ϫ14 Յ k Յ 14
Ϫ28 Յ l Յ 29
35011
0.4 ϫ 0.15 ϫ 0.10
2.21Ϫ29.04
Ϫ22 Յ h Յ 22
Ϫ13 Յ kՅ 15
Ϫ34 Յ l Յ 35
31092
0.22 ϫ 0.20 ϫ 0.14
1.73Ϫ28.77
Ϫ18 Յ h Յ 9
Ϫ26 Յ k Յ 27
Ϫ7 Յ l Յ 21
16297
Reflections collected
Independent reflections
Completeness to θ ϭ 26.37°
Absorption correction
Max. and min. transmission
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R int [I Ͼ 2σ(I)]
7969 [R(int) ϭ 0.049]
99.8%
Semi-empirical from equivalents Semi-empirical from equivalents None
11709 [R(int) ϭ 0.053]
99.2%
8782 [R(int) ϭ 0.0691]
76.2%
0.904 and 0.864
1.000 and 0.870
Full-matrix least squares on F2
7969/38/419
Full-matrix least squares on F2
11709/0/532
Full-matrix least squares on F2
8782/0/486
0.895
1.016
1.054
R1 ϭ 0.0476, wR2 ϭ 0.1293
R1 ϭ 0.0581, wR2 ϭ 0.1142
R1 ϭ 0.0570, wR2 ϭ 0.1302
Eur. J. Inorg. Chem. 2004, 3724Ϫ3732
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3731

L. S. Natrajan, C. Wilson, J. Okuda, P. L. Arnold
FULL PAPER
[19]
[20]
[21]
E. Y. Tshuva, I. Goldberg, M. Kol, Z. Goldschmidt, Inorg.
Chem. 2001, 40, 4263.
Acknowledgments
The authors thank the EPSRC (studentship for L. S. N., Advanced
Fellowship for P. L. A.) and the Royal Society for funding, Pro-
fessor F. Geoffrey N. Cloke for use of a catalyst testing apparatus,
and the Deutsche Forschungsgemeinschaft and the Fonds der
Chemischen Industrie for financial support.
Y. Nakayama, K. Waatanabe, N. Ueyama, A. Nakamura, A.
Harada, J. Okuda, Organometallics 2000, 19, 2498.
E. Y. Tshuva, I. Goldberg, M. Kol, Organometallics 2001, 20,
3071; M. G. Thorn, Z. C. Etheridge, P. E. Fanwick, I. P.
Rothwell, J. Organomet. Chem. 1999, 591, 148.
D. W. Stephan, T. T. Nadasdi, Coord. Chem. Rev. 1996, 147,
147.
[22]
[23]
[1]
˚
V. C. Gibson, S. K. Spitzmesser, Chem. Rev. 2003, 103, 283.
A. Van der Linden, C. J. Schavarien, N. Meijboom, C. Ganter,
A single SϪS bond is 1.83 A; A. Bondi, J. Phys. Chem. 1964,
[2]
68, 441; International Tables of Crystallography, Kluwer Aca-
demic, Dordrecht, 1992, vol. C.
G. Orpen, J. Am. Chem. Soc. 1995, 117, 3008.
[3]
[24]
[25]
V. C. Gibson, J. Chem. Soc., Dalton Trans. 1994, 1607.
W. Nakanishi, S. Hayashi, T. Arai, Chem. Commun. 2002, 2416.
W. Nakanishi, S. Hayashi, S. Toyota, Chem. Commun. 1996,
371; W. Nakanishi, S. Hayashi, S. Toyota, J. Org. Chem. 1998,
63, 8790; S. Hayashi, W. Nakanishi, J. Org. Chem. 1999, 64,
6688.
C. Rovira, J. J. Novoa, Chem. Eur. J. 1999, 5, 3689.
T. N. Guru Row, R. Parthasarathy, J. Am. Chem. Soc. 1981,
103, 477.
[4]
G. J. P. Britosvek, V. C. Gibson, D. F. Wass, Angew. Chem. Int.
Ed. 1999, 38, 428.
S. Fokken, T. P. Spaniol, H.-C. Kang, W. Massa, J. Okuda,
[5]
Organometallics 1999, 15, 5069; L. Porri, A. Ripa, P. Colombo,
E. Miano, S. Capelli, S. V. Meille, J. Organomet. Chem. 1996,
514, 213; F. Amor, S. Fokken, T. Kleinheim, T. P. Spaniol, J.
Okuda, J. Organomet. Chem. 2001, 621, 3; Y. Takashima, Y.
Nakayama, H. Yasuda, A. Nakamura, A. Harada, J. Or-
ganomet. Chem. 2002, 654, 74.
[26]
[27]
[28]
[29]
I. Dance, New. J. Chem. 2003, 27, 22.
S. Fokken, F. Reichwald, T. P. Spaniol, J. Okuda, J. Organomet.
Chem. 2002, 663, 158, for comparison, the activity of the di-
methyltitanium complex [(tbmp)TiMe2] towards ethene polym-
[6]
J. Okuda, S. Fokken, T. Kleinhenn, T. P. Spaniol, Eur. J. Inorg.
Chem. 2000, 1321; C. Capacchione, A. Proto, H. Ebeling, R.
Muelhaupt, K. Moeller, T. P. Spaniol, J. Okuda, J. Am. Chem.
Soc. 2003, 125, 4964.
erisation is calculated to be 12.52 kg PE g [Ti]Ϫ1 barϪ1 hϪ1
.
[30]
[31]
[32]
[33]
M. Bochmann, S. J. Lancaster, M. B. Hursthouse, K. M. Abdul
Malik, Organometallics 1994, 13, 2235; S. J. Lancaster, M.
Bochmann, Organometallics 1993, 12, 663.
F. Guerin, D. H. McConville, J. J. Vittal, G. A. P. Yap, Or-
ganometallics 1998, 17, 5172; R. M. Gauvin, J. Kress, J. Mol.
Catal. A: Chem. 2002, 182/183, 411.
For an earlier example of oxygen insertion into a TiϪMe bond,
see: G. J. Erskine, G. J. B. Hurst, E. L. Weinberg, B. K. Hunter,
J. D. McCowan, J. Organomet. Chem. 1984, 267, 265.
M. J. Carney, P. J. Walsh, F. J. Hollander, R. G. Bergman, Or-
ganometallics 1992, 11, 761; S. J. Kim, I. N. Jung, B. R. Yoo,
S. Cho, J. Ko, S. H. Kim, S. O. Kang, Organometallics 2001,
20, 1501; J. Scholz, G. A. Hadi, K. H. Thiele, H. Gorls, R.
Weimann, H. Schumann, J. Sieler, J. Organomet. Chem. 2001,
626, 243; Y. Wielstra, S. Gambarotta, A. L. Spek, W. J. J.
Smeets, Organometallics 1990, 9, 2142; H. G. Helmut, C. E.
Denner, J. Organomet. Chem. 1990, 391, 53.
[7]
J. Okuda, S. Fokken, H.-C. Kang, W. Massa, Polyhedron 1998,
17, 943; C. R. Cornman, K. M. Geiser-Bush, J. W. Kampf,
Inorg. Chem. 1999, 38, 4303.
R. Siefert, T. Weyhermuller, P. Chaudhuri, J. Chem. Soc., Dal-
[8]
ton Trans. 2000, 4656.
T. Miyatake, K. Mizunuma, Y. Seki, M. Kakugo, Macromol.
[9]
Chem. Rapid Commun., 1989, 10, 349; T. Miyatake, K. Mizun-
uma, M. Kakugo, Macromol. Chem. Macromol. Symp., 1993,
66, 203; F. G. Sernetz, R. Muelhaupt, S. Fokken, J. Okuda,
Macromolecules 1997, 30, 1562; M. Kakugo, T. Miyatake, K.
Mizunuma, Stud. Surf. Sci. Catal., 1999, 56, 517; M. Kakugo,
T. Miyatake, K. Mizunuma, Y. Yagi, Sumitomo Chemical Co.
U.S. 5043, 408.
[10]
R. D. J. Froese, D. G. Musaev, T. Matsubara, K. Morokuma,
J. Am. Chem. Soc. 1997, 119, 7190; R. D. J. Froese, D. G. Mu-
saev, K. Morokuma, Organometallics 1999, 18, 373.
[11]
[34]
[35]
J. Okuda, E. Masoud, Macromol. Chem. Phys. 1998, 199, 543.
L. S. Natrajan, J. J. Hall, A. J. Blake, C. Wilson, P. L. Arnold,
J. Solid State Chem. 2003, 171, 90.
[12]
Y. Takashima, Y. Nakayama, K. Watanabe, T. Itono, N. Uey-
ama, A. Nakamura, H. Yasuda, A. Harada, Macromolecules
2002, 35, 7538.
F. G. Sernetz, R. Muelhaupt, S. Fokken, J. Okuda, Macromol-
TiCl4·2thf and ZrCl4·2thf: L. E. Manzer, Inorg. Synth. 1985,
21, 135; [Zr(CH2Ph)4]: U. Zucchini, E. Albizzati, U. Giannini,
J. Organomet. Chem. 1971, 357; [Ti(NMe2)4]: A. A. Naiini, S.
L. Ringrose, Y. Su, R. A. Jacobsen, J. G. Verkade, Inorg. Chem.
1993, 32, 1290; [Zr(NMe2)4]: H. Schreider, G. T. Puchta, F. A.
Kaul, G. Raudaschl-Sieber, F. Lefebvre, J. Mol. Catal. A,
Chem. 2001, 170, 127.
P. L. Arnold, J. J. Hall, L. S. Natrajan, A. J. Blake, C. Wilson,
Dalton Trans. 2003, 1053.
SHELXTL 5.10, Bruker AXS Inc., Madison, Wisconsin, USA,
1997; G. M. Sheldrick, Acta Crystallogr., Sect. A 1990, 46, 467.
G. M. Sheldrick, SHELXL-97, University of Goettingen, Ger-
many, 1997.
[13]
ecules 1997, 30, 1562.
W. Kaminsky, J. Chem. Soc., Dalton Trans. 1998, 1413; X. Yan,
[14]
A. Chernega, M. L. H. Green, J. Sanders, J. Souter, T. Ushioda,
J. Mol. Catal. A: Chem. 1998, 128, 119; G. W. Coates, Chem.
Rev. 2000, 100, 1223.
[36]
[37]
[15]
P. L. Arnold, L. S. Natrajan, J. J. Hall, S. J. Bird, C. Wilson, J.
Organomet. Chem. 2002, 647, 205.
M. C. W. Chan, K.-H. Tam, Y.-L. Pui, N. Zhu, J. Chem. Soc.,
[16]
Dalton Trans. 2002, 3085.
V. Reimer, T. P. Spaniol, J. Okuda, H. Ebeling, A. Tuchbreiter,
[17]
R. Muelhaupt, Inorg. Chim. Acta 2003, 345, 221.
Received February 24, 2004
Early View Article
Published Online July 22, 2004
[18]
J. Okuda, S. Fokken, H.-C. Kang, W. Massa, Chem. Ber. 1995,
128, 221.
3732
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 3724Ϫ3732