A ligand influence on the stability of heterobimetallic complexes containing the Ti(μ-O)Al skeleton. Transformation of heterometallic systems to the homometallic Ti(IV) and Al(III) complexes

Article abstract of DOI:10.1039/b710470g

The influence of the ligands on the formation and stability of -oxo-bridged Ti(iv) complexes has been studied. Reaction of LTiCl₃ (1) and LAlMe(OLi) (L = HC(CMeN(2,6-iPr₂C₆H₃)) ₂, "NacNac") afforded i

Full text of DOI:10.1039/b710470g

View Article Online / Journal Homepage / Table of Contents for this issue
An international journal of inorganic chemistry
www.rsc.org/dalton
Number 37 | 7 October 2007 | Pages 4105–4228
ISSN 1477-9226
PAPER
COMMUNICATION
Herbert W. Roesky et al.
A ligand influence on the stability
of heterobimetallic complexes
containing the Ti(µ-O)Al skeleton
Haiping Xia et al.
Synthesis and characterization of a
bimetallic iridium complex with a ten
sp2-carbon chain bridge

View Article Online
PAPER
www.rsc.org/dalton | Dalton Transactions
A ligand influence on the stability of heterobimetallic complexes containing
the Ti(l-O)Al skeleton. Transformation of heterometallic systems to the
homometallic Ti(IV) and Al(III) complexes†‡
Grigory B. Nikiforov,a Herbert W. Roesky,*a Peter G. Jones,b Rainer B. Oswaldc and Mathias Noltemeyera
Received 10th July 2007, Accepted 15th August 2007
First published as an Advance Article on the web 4th September 2007
DOI: 10.1039/b710470g
The influence of the ligands on the formation and stability of l-oxo-bridged Ti(IV) complexes has been
studied. Reaction of LTiCl3 (1) and LAlMe(OLi) (L = HC(CMeN(2,6-iPr2C6H3))2, “NacNac”)
afforded intermediate LTiCl2(l-O)AlMeL (5) in solution, which was converted to LTiCl(l-O)2TiClL (6)
and LAlMeCl within 2 days. The decomposition of 5 was estimated to be thermodynamically favorable.
The interaction of LTiMe3 (3) and LAlMe(OH) yielded the intermediate LTiMe2(l-O)AlMeL (7).
Complex 7 decomposes in solution giving the titanium oxo complex LTiMe(O) (8) and LAlMe2. The
calculated DG298 for this reaction is −128 kJ mol−1. The degradation of LTiMe2(l-O)AlMeL is slow and
follows first order kinetics with k2 = 4.09(7) × 10−7 s−1. The dimeric complex
LTiMe(l-O)2TiMeL–toluene (9a) was isolated from the reaction of 3 with LAlMe(OH) in toluene and
LTiMe(l-O)2TiMeL–hexane (9b) from hexane. The dimerization of 8 yielding LTiMe(l-O)2TiMeL (9)
is marginally endothermic, with a calculated DG298 of +27 kJ mol−1. The formation of the solid 9 is due
to the lattice stabilization. The solid l-oxo-bridged complex 9 and Mes3Ga were obtained from the
reaction of LTiMe3 with [Mes2Ga(OH)]2–THF in toluene, and 9 was also isolated from the reaction of
LTiMe3 with 1 equiv. of H2O in toluene. Compounds LTiCl3 (1), LTiCl(l-O)2TiClL (6), 9a and 9b have
been characterized by X-ray single crystal structure, NMR, IR, EI-MS and elemental analysis.
Complexes 5, 7 and 8 have been characterized by NMR. Compounds 3, 6 and 9 possess moderate
catalytic activity in the polymerization of ethylene.
Although complexes containing the Al(l-OR)Ti bond, includ-
ing TiHal2(salen)AlMe2–AlMe2Br (Hal = Cl, Br, SalenH2 = N,N-
Introduction
bis(salicylidene)ethylenediamine),4 alkoxides,5 complexes sup-
ported by the (R)-H8-BINOL ligand,6 or OSO-type ligands,7
are known, only three metallocene l-oxo-bridged com-
plexes containing the Ti(l-O)Al group have been prepared,
namely Cp2TiMe(l-O)AlMeL,2b Cp2Ti(SH)(l-O)Al(SH)L and
Cp2Ti(SH)(l-O)Al(OH)L.2d These compounds were obtained
using a building block strategy starting from organometallic
hydroxide LAlMe(OH) and Cp2TiMe2, or a sulfide complex
Cp2Ti(l-S)2AlL and H2O. All the reported complexes contain
the cyclopentadienyl ligand. Therefore, an investigation was
undertaken of the effect of various ligands on the stability
and polymerization activity of heterobimetallic l-oxo-bridged
complexes containing the Ti(l-O)Al skeleton. Here, we report
the synthesis, characterization, and catalytic properties of oxygen-
bridged Ti(IV) complexes supported by the bulky b-diketiminato
ligand (L = HC(CMeN(2,6-iPr2C6H3))2, ”NacNac”).
The preparation of new single-site olefin polymerization catalysts
has attracted a great interest in both academia and industry
in the last few decades.1 Our interest has been focused on the
preparation and characterization of oxo-bridged heterometallic
complexes containing group 3 (Al, Ga) and group 4 (Ti, Zr,
Hf) metals.2 These compounds can serve as a model system for
studying the polymerization of olefins and can be considered
as effective single-site homogeneous catalysts possessing high
catalytic activity and requiring a low ratio of cocatalyst to catalyst
precursor, and allowing unprecedented control over the polymer
microstructure, thus generating new polymers with improved
properties.2e In particular, the oxygen-bridged heterobimetallic
complex Cp*2ZrMe(l-O)TiMe2Cp* is highly active and produces
linear low-density polyethylene.3
aInstitut fu¨r Anorganische Chemie, Universita¨t Go¨ttingen, Tammannstrasse
4, D-37077, Go¨ttingen, Germany. E-mail: hroesky@gwdg.de; Fax: +49
(0)551-39-3373
bInstitut fu¨r Anorganische Chemie und Analytische Chemie der Technischen
Universita¨t Braunschweig, Hagenring 30, D-38106, Braunschweig, Germany
Results and discussion
cInstitut fu¨r Physikalische Chemie, Georg-August-Universita¨t Go¨ttingen,
Tammannstrasse 6, D-37077, Go¨ttingen, Germany
Preparation and reactivity of complexes
† Dedicated to Professor Ken Wade on the occasion of his 75th birthday.
The precursor used in this investigation, complex LTiCl3 (1),
was prepared in good yield by metathesis of TiCl4 with LLi in
toluene. Reaction in toluene yielded the major product 1 and
minor amounts of [C(CMeN(2,6-iPr2C6H3))2]2H2 (2) (Scheme 1).8
Formation of 2 can be rationalized as a result of C–C coupling
‡ Electronic supplementary information (ESI) available: IR spectrum of
LH, all experimental details and interpretation of NMR reactions in situ,
NMR spectrum of complex 7, NMR data for LAlMe2 and LAlMe(OLi)
in C6D6 at room temperature, atomic coordinates of the optimized
molecules 7, 8, 9, LAlMe2, details of polymerization experiments. See
DOI: 10.1039/b710470g
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4149–4159 | 4149
©

View Article Online
Table 1 Calculated thermodynamic values (298 K) for the reactions in the system of LTiMe3 (3) with LAlMe(OH)
Reaction
DH298
−51
DG298
−128
−73
27
−174
−229
14
reaction LTiMe(l-O)2TiMeL (9) and LAlMe2 were isolated as
final products at room temperature.12 However, monitoring the
reaction of 3 with 1 equiv. of LAlMe(OH) (see the ESI‡) by
1H NMR spectroscopy revealed a series of products (LTiMe2(l-
O)AlMeL (7), LTiMe(O) (8), 9, LAlMe2 (Table 1) and CH4) as
shown in Scheme 3.
Scheme 1
of two ligands promoted by the Ti(IV) centre. Reaction of 1
with excess of MeLi produced the alkyl complex LTiMe3 (3)9
(Scheme 1), while interaction of 1 with excess of KH in THF
followed by the addition of toluene produced the Ti(III) complex
10
LTiCl2 (4), thus making 1 a potentially useful precursor for the
preparation of Ti(IV) alkyl complexes and Ti(III) derivatives.
The reaction of 1 with LAlMe(OLi) was attempted in order
to prepare the novel l-oxo-bridged titanium–aluminium complex
LTiCl2(l-O)AlClL (5). However, this reaction afforded LAlMeCl11
and LTiCl(l-O)2TiClL (6), the latter being deposited from the
solution. The intermediate complex 5 was observed by means of
NMR, and its relative concentration was always small (see the
ESI‡). The complete transformation of the starting material to
6 and LAlMeCl proceeded within 2 days at room temperature
(Scheme 2).
Scheme 3
A maximum concentration of 7 was achieved from 7 to
36 hours after the beginning of the reaction (Fig. 1 and 2).
However, within this time, complex 7 did not crystallize from the
concentrated toluene or hexane solution,13 but slowly decomposed
with deposition of LTiMe(l-O)2TiMeL (9) and formation of
14
LAlMe2 (Scheme 3).
The formation of 9 and LAlMe2 as the final reaction products
implies migration of the methyl group from the Ti(IV) to the Al(III)
and transfer of the oxygen atom from the Al(III) to the Ti(IV)
centre, followed by dimerization of the intermediate 8. This can be
regarded as a symmetrization of the system. In order to extend our
study, we attempted the reaction of [Mes2Ga(OH)]2–THF and 3
in toluene at room temperature and 100 ◦C, which again results in
the formation of 9, this time with Mes3Ga (Scheme 4), confirming
the oxygen metathesis. The formation of 9 was observed during
the hydrolysis of 3 implying that 9 is highly stable, at least in the
solid state, and that 9 is a final product of various reactions of the
Scheme 2
In order to explore the influence of the ligand on the stability
of the l-oxo-bridged titanium–aluminium system, the reaction of
3 with LAlMe(OH) was investigated. Similarly to the previous
4150 | Dalton Trans., 2007, 4149–4159
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
second order kinetics (Fig. 3). A least squares fit of the linear plot
of 1/c3 against time16 revealed the second order rate constant (k1 =
6.8(2) × 10−3 M−1 s−1).17 No other kinetic data are available for
the interaction of alkyltitanium compounds with organometallic
hydroxides. The reaction of (tBu3PN)TiMe3 with AlMe3 was
determined to be of second order (k1 = 3.9(5) × 10−4 M−1 s−1).18
The decomposition of LTiMe2(l-O)AlMeL (7) with formation of
LTiMe(O) (8) and LAlMe2 was found to follow first order kinetics
(ln(c(LAlMe2)) vs. time and ln(c(7)) vs. time) (Fig. 4). The rate
constant k2 was determined as 4.09(7) × 10−7 s−1.
Fig. 1 Concentration of complexes in the reaction system vs. time.
Fig. 3 Representative data for the decay of LTiMe3 (3) vs. time. Linear
relationship 1/c(LTiMe3) vs. time: Y = 15.1(5) + 6.8(2) × 10−3·X.
Fig. 2 Concentration of complexes in the reaction system vs. time.
Scheme 4
Ti(IV) alkyl complex 3 and compounds that are potential oxygen
donors (e.g. hydroxides, water) (Scheme 5).
Fig. 4 Representative data for the decay of LTiMe2(l-O)AlMeL (7)
vs. time; the linear relationship between ln(c(7)) vs. time is Y
=
−2.758(4)–4.09(7) × 10−7·X.
The decomposition of LTiMe2(l-O)AlMeL (7) resulted in
the formation of LAlMe2 and the titanium complex LTiMe(l-
O)2TiMeL (9) according to the proton NMR spectrum. Calculated
concentrations of complex 9 were always lower than those of
LAlMe2; this might be due to the observed resonance, which
belongs to the monomeric complex LTiMe(O) 8. Its concentration
initially was similar to that of LAlMe2, and after about 33 to 35
(×104) seconds the concentration of 8 starts to decrease relative to
LAlMe2 (see Fig. 1). This behavior of the concentration is common
for monomeric species in dilute solutions, and oligomerization
was induced by increasing the concentration. The additional
resonances of c -H (after 33 × 104 s at 5.42, 5.28 ppm), could
be assigned to the oligomers, although deposition of 9 was not
observed from the solution within the time of the experiment (0 to
80 × 104 s). Additionally, the rate of formation of 8 follows first
order kinetics (k2 = 5.7(8) × 10−7 s−1), similar to that determined
Scheme 5
NMR, kinetic and thermodynamic study of the
LTiMe3–LAlMe(OH) system
The reaction of LTiMe3 and LAlMe(OH) (ratio 1 : 1) was
monitored in detail by 1H NMR. In order to determine the
concentration of a complex, the relative intensities of L (c -H)
or the protons of the Me-group (at the aluminium centre) were
compared. We propose that the relaxation time of the protons
in similar environments for different complexes would be similar,
thus reducing the inaccuracy of the measurement.15 The decay
of the initial species (LAlMe(OH), LTiMe3) was found to follow
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4149–4159 | 4151
©

View Article Online
from ln(c(7)) vs. time with linear regression. For that reason the
presence of 8 in the solution seems feasible.
−92 kJ mol−1, showing that the reaction is thermodynamically
favorable.
The results of the kinetic measurements imply the thermody-
namic instability of LTiMe2(l-O)AlMeL (7). To get a more precise
picture of the reaction involving compound 7, DFT calculations
were carried out for all involved complexes without any simplifi-
cations, aiming at the determination of the reaction mechanism
and reaction energetics. Transformation of 7 to LTiMe(O) (8) and
LAlMe2 or to complex LTiMe(l-O)2TiMeL (9) and LAlMe2 is
thermodynamically favored [eqn (1)–(3), Table 1]. The free energy
of dimerization of 8 to 9 has a small positive value indicating
thermodynamic instability of the dimeric complex in the gas
phase [eqn (2)]. The solvation energies of the neutral molecules in
solvents of low polarity such as benzene and toluene are small and,
hence, we propose that dimerization of 8 is thermodynamically
unfavorable in C6D6 and toluene solution. Obviously, compound
9 is lattice stabilized.
Using this approach and bond dissociation energies Ti–Cl
(388.3 kJ mol−129 or 429.3 kJ mol−130,22) and Al–Cl (430 kJ mol−131
or 420.7 10 kJ mol−122), the enthalpy of the reaction [eqn (5)]
was estimated to be between −71 and −193 kJ mol−1. As
a result, LTiCl2(l-O)AlMeL (5) is thermodynamically unstable
and decomposed to give the thermodynamically favorable 6 and
LAl(Me)Cl.
2 LTiCl2(l-O)AlMeL → LTiCl(l-O)2TiClL + 2LAlMeCl
(5)
5
6
Characterization of titanium complexes 1, 6, 7, 9a and 9b
The structure of LTiCl3 (1) was determined in the space group
Pnma; the structure is isotypic to that of LTiMe3,9 arising from
the comparable size of the Me group and Cl atom (Fig. 5). The
geometry of the complex 1 is also similar to that of LZrCl3,32 (L =
b-diketiminato ligand N(SiMe3)C(tBu)CHC(Ph)N(SiMe3)). The
molecule displays crystallographic mirror symmetry. The Ti(IV)
centre exhibits a distorted square pyramidal environment, with
DH4 ≈ 2E (Ti(l-O)) + 2E (Al-Me) − 2E (Ti-Me)
− 2E(Al(l-O))
(4)
Alternatively, the enthalpy of this transformation can be
estimated from the bond dissociation energies (BDE) [eqn (4)],
provided that the BDEs of Ti(l-O) are similar for the precursor
and the reaction product. In fact, the calculated Ti(l-O) distance in
˚
the Ti atom lying 0.51 A out of the NNClCl plane. The Ti–N
bond length in 1 is in the normal range33 (Table 2).
˚
7 (1.813 A) is close to the Ti(l-O) distance in LTiMe(l-O)2TiMeL–
toluene (9a) and LTiMe(l-O)2TiMeL–hexane (9b) (1.843(2) and
˚
1.8410(13) A), implying similarity of the BDEs for Ti(l-O) in
both compounds.19 The experimentally determined BDEs have
been collected from the literature; E(Ti–C) is 251 kJ mol−1 for
and 190–240 kJ mol−1 for TiR4 (mean E(Al–C)
20,21
25
Cp2TiMe2
255 kJ mol−122 and E(Al–O) is 424 kJ mol−123). The Ti–O BDE
strongly depends on the Ti–O bond length19 and varies from
670( 9.6) kJ mol−1 for [TiO]+24 to E(Ti–OR) 466 kJ mol−1 for
Ti(OR)4 (452 kJ mol−1 for Cp2Ti(OR)2, R = C(O)Ph, C(O)CF3).25
Using various calorimetric reactions, the mean dissociation en-
ergies of Ti–O bonds in alkoxide complexes have been estimated
as 431 kJ mol−1.26 The E(Ti-OAlR*) should be close to E(Ti-OR)
since the Ti–O(Al) bond distances2,27 for the known l-oxo-bridged
Ti(l-O)Al complexes are in the same range as those of the titanium
alkoxides.28 Using these values and eqn (4), the enthalpy of the
reaction [eqn (3)] is estimated to lie in the range from −22 to
Fig. 5 Molecular structure of 1 with selected labelled atoms. Ellipsoids
represent 50% probability levels; carbon radii are arbitrary. “A” indicates
atoms generated by mirror symmetry.
a
˚
Table 2 Selected bond distances (A) and angles (deg) for 1, 6, 9a and 9b
1
6
9a
9b
Ti–N(1)
Ti–Cl(1)
Ti–Cl(2)
N(1)–C(1)
2.062(3)
2.2531(17)
2.1893(19)
1.340(5)
Ti–O(B)
Ti–O
Ti–N(2)
Ti–N(1)
1.8316(8)
1.8589(8)
2.1127(10)
2.1253(10)
2.2434(4)
81.78(4)
148.39(4)
88.78(4)
85.47(4)
Ti–O
Ti–C(1)
Ti–N
O(C)–Ti–O
O–Ti–C(1)
O–Ti–N
C(1)–Ti–N
N–Ti–N(D)
Ti(C)–O–Ti
1.843(2)
2.092(5)
2.149(3)
82.83(16)
104.69(12)
91.35(11)
99.33(14)
84.54(16)
97.17(16)
Ti–O
Ti–O(E)
Ti–C(1)
Ti–N(1)
1.8410(13)
1.8514(13)
2.0808(19)
2.1568(14)
2.1591(15)
82.77(6)
104.89(7)
84.69(6)
155.27(6)
91.83(6)
C(1)–C(2)
1.391(5)
Ti–Cl
Ti–N(2)
N(1)–Ti–N(1)
N(1)–Ti–Cl(2)
N(1)–Ti–Cl(1)
Cl(2)–Ti–Cl(1)
87.13(17)
99.81(10)
87.17(10)
107.84(7)
O–Ti–O(B)
O(B)–Ti–N(2)
O–Ti–N(2)
N(2)–Ti–N(1)
O–Ti–Cl
N(1)–Ti–Cl
Ti(B)–O–Ti
O–Ti–O(E)
O–Ti–C(1)
N(1)–Ti–N(2)
O–Ti–N(1)
O–Ti–N(2)
C(1)–Ti–N(1)
106.43(3)
100.01(3)
98.22(4)
99.83(7)
a Symmetry transformations used to generate equivalent atoms A of 1: x,−y + 1/2, z; to generate equivalent atoms B of 6: −x + 1,−y + 1,−z + 1; C of
9a: −x + 1,−y,−z + 1; D of 9a: x,−y, z; E of 9b: −x + 1,−y,−z + 1.
4152 | Dalton Trans., 2007, 4149–4159
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
The single crystal X-ray structures of 6 (Fig. 6), LTiMe(l-
O)2TiMeL–toluene (9a) and LTiMe(l-O)2TiMeL–hexane (9b)
(Fig. 7) have been determined by X-ray crystallographic methods.§
The structure of 7 was calculated and will be presented below.
LTiCl(l-O)2TiClL (6) contained toluene of solvation. Compound
9 revealed two crystalline forms: 9a is a toluene and 9b a hexane
solvate. Complexes 6, 9a and 9b display imposed crystallographic
¯
¯
symmetries 1, 2/m and 1, respectively. The structures of 6 and
9 consist of two LTi(Cl)(l-O) and LTiMe(l-O) units contain-
ing Ti(IV) centres in a distorted square-pyramidal environment,
formed by two nitrogen atoms, two bridging oxygen atoms and a
carbon atom of the methyl group or chlorine atom. Mean deviation
˚
from the (NNOO) plane is 0.040 A in 6, zero by symmetry
˚
in 9a, and 0.016 A in 9b. The Ti atom lies 0.493, 0.413 and
˚
0.409 A above the (NNOO) plane in 6, 9a and 9b, respectively.
The coordination environment of the Ti atom in {TiN2(l-O)2Me}
and {TiN2(l-O)2Cl} has so far not been observed in the literature,
although a distorted square-planar coordination was reported
for the titanium atom {environment TiN2(l-OMe)2Me} in the
spirosiladiazatitanacyclobutane complex [(cyclo)Si(NtBu)2Ti(l-
Fig. 7 Molecular structure of 9b with selected atoms labelling. Ellipsoids
represent 50% probability levels. Carbon radii are arbitrary.
34
OMe)Me]2 and {environment TiN2(l-OR)2Me} in the chloride
complexes [Ti(Cl)(NMe2)(–OCH2CH(CH2Ph)N(R)−)]2 (where
R = iPr, cyclo-C6H11, 2-adamantyl),35 [(cyclo)Si(NtBu)2Ti(l-
OMe)Cl]2.34 All the bond distances in 6, 9a and 9b are in the
normal range.33,36 The Ti(l-O) bond lengths also fall in the range
for Ti(IV) complexes with the Ti(l-O)2Ti unit.37
the molecular ion for 9 and also a strong cation peak, which
resulted from abstraction of the methyl group. The EI-MS of 6
1
contained strong cations of [LTiOCl]+ and of [LTi-2H]+. The H
and 13C NMR spectra of 6 are consistent either with 6 or with the
monomer LTiCl(O); however, available data do not allow further
detailed assignments.
The complexity of the molecules 6 and 9 prevents accurate
assignment of the IR bands to the individual vibrations. Taking
into account the calculated IR spectrum of 9, we were able to assign
the Ti-Me and Ti-(l-O) vibrations. Observed bands (321 and
340 cm−1, calculated 330 and 352 cm−1) were assigned to the out
of plane vibrations of oxygen of the {MeTi(l-O)2TiMe} unit. A
broad band at564 cm−1 (calculated548, 554, 560 cm−1) is attributed
to the Ti–Me. Very strong overlapped bands with the energies 600
to 694 cm−1 (calculated 585, 609, 620, 639, 642, 661, 664, 680,
693 cm−1) were assigned to the modes of {MeTi(l-O)2TiMe} (in-
plane vibration of oxygen) and the ligand. The contribution of an
in-plane Ti(l-O)2Ti vibration is most substantial for the calculated
bands at 680 and 693 cm−1. Similarly, very strong and overlapping
bands within 620 to 690 cm−1 in the IR spectrum of 6 have been
assigned to the modes with substantial contribution of {ClTi(l-
O)2TiCl} (in-plane vibrations of oxygen) and bands at 309 and
344 cm−1 to the out of plane vibrations of oxygen in {ClTi(l-
O)2TiCl}.
Fig. 6 Molecular structure of 6 with selected atom labelling. Ellipsoids
represent 50% probability levels. Carbon radii are arbitrary.
The structure of LTiMe2(l-O)AlMeL (7) was calculated by DFT
˚
methods (Fig. 8). The Al–(l-O) bond length (1.821 A) in 7 is signif-
icantly longer than that in the l-oxo-bridged titanium–aluminium
Although complexes 6 and 9 are slightly soluble in benzene
complex Cp2TiMe(l-O)AlMeL2c (1.715(3) A) while the Ti–(l-O)
1
and toluene (less than 1 mg in 1 ml), H and 13C NMR spectra
˚
˚
distance in 7 (1.813 A) is similar to that in Cp2TiMe(l-O)AlMeL
were measured. Based on thermodynamical and kinetic data an
assumption was made (see the previous section) that 9 is lattice
stabilized and monomer 8 might be present in the solution of 9.
Hence, NMR resonances in the benzene solution of 9 have been
assigned to the monomeric complex 8. Proton resonance of the
methyl group attached to the titanium atom appears at 1.22 ppm
and 13C resonances at 60.2 ppm for 8. The EI-MS of 9 exhibits
˚
(1.808(3) A), reflecting a higher Ti-(l-O) bond strength (1.01
valence unit in terms of bond valence theory38), than that of Al–
(l-O) (0.58 valence unit). The Ti1–O3–Al2 angle (160.1◦) is also
close to that in Cp2TiMe(l-O)AlMeL (151.7(2)◦).2c
The 1H and 13C NMR spectra are in agreement with the
proposed structure of 7. The proton resonances assigned to 7
have similar relative intensities in all NMR spectra at different re-
action times (in situ reaction). In order to confirm the assignment,
the NMR shifts were calculated with the Gauge-Independent
§ CCDC reference numbers 652772–652774. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/b710470g
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4149–4159 | 4153
©

View Article Online
with B(C6F5)3. Additionally, all complexes were inactive for
styrene polymerization even upon attempted activation with MAO
in toluene at room temperature and 80 ◦C, and 3, 6 and 9a
are inactive in combination with AlMe3 for polymerization of
ethylene. Increase of the cocatalyst to catalyst ratio did not
increase the activities of 3 and 9a. This behaviour is attributable
to the instabilities of the active species, similarly to that of the
LTiMe2–B(C6F5)3 system;40 interaction of the complexes with
MAO resulted in the formation of inactive species (see the ESI‡).
The DSC measurements show that the melting points (Tm) of the
polyethylene produced with 3, 6 and 9a at room temperature are
in the range of 123.4 to 125.8 ◦C, similar to those (120 to 125 ◦C)
produced with Cp*2MeZr(l-O)TiMe2Cp*.3
Fig. 8 Optimized geometry of LTiMe2(l-O)AlMeL (7), hydrogen atoms
˚
are omitted for clarity. Selected bond distances (A) and angles (deg):
Al2–O3 1.821, Al2–C8 1.981, Al2–N6 2.000, Al2–N4 2.011, Ti1–O3 1.813,
Ti1–C9 2.092, Ti1–C10 2.089, Ti1–N7 2.090, Ti1–N5 2.416; O3–Al2–C8
111.95, O3–Al2–N6 116.60, C8–Al2–N6 106.40, C8–Al2–N4 107.32,
N6–Al2–N4 94.73, O3–Ti1–C10 96.25, O3–Ti1–N7 108.48, O3–Ti1–C9
90.61, O3–Ti1–N5 166.44, C10–Ti1–N7 108.05, C10–Ti1–C9 121.32,
C10–Ti1–N5 83.43, N7–Ti1–C9 124.54, C9–Ti1–N5 78.15, Ti1–O3–Al2
160.11.
Conclusions
Interaction of LTiMe3 and LAlMe(OH) gives the target l-
oxo-bridged Ti(IV)–Al(III) complex LTiMe2(l-O)AlMeL (7). This
reaction is determined to be a second order process. While the
previously reported metallocene complex Cp2TiMe(l-O)AlMeL2c
was isolated and characterized, complex 7 is unstable and decom-
poses in solution giving the titanium oxo complex LTiMe(O) (8)
and LAlMe2. Decomposition of 7 is thermodynamically favorable
and the estimated DG298 is −128 kJ mol−1. Decomposition of
LTiMe2(l-O)AlMeL (7) is slow and follows first order kinetics. The
dimeric complex LTiMe(l-O)2TiMeL (9) was isolated from the
reaction system. Dimerization of 8 to 9 is marginally endothermic
(+27 kJ mol−1) and formation of 9 is explained by lattice
stabilization.
1
Atomic Orbital (GIAO) method for 7.39 Calculated H (−0.21
to −0.11 ppm) and 13C (60.6 ppm) chemical shifts of the Ti–Me
resonance are in good agreement with the observed values (0.53
and 56.6 ppm). The Me group at aluminium exhibits a proton
resonance at −0.14 ppm and 13C resonance at −9.5 ppm. The
calculated 13C chemical shift of Al–Me (−9.5 ppm) is in good
agreement but the calculated proton shift fits less well (−1.34 ppm).
The c -H of each NC(Me)CHC(Me)N moiety showed resonances
at 4.25 ppm (calcd 4.28) and 5.03 ppm (calcd 4.46). The upfield
resonance is assigned to the NC(Me)CHC(Me)N attached to
Al(III) and the downfield resonance to the Ti(IV) centre.
The chloride ligand at the Ti(IV) centre does not increase
the stability of a l-oxo-bridged Ti(IV)–Al(III) complex. The
chloride complex LTiCl2(l-O)AlMeL (5) was observed only as a
reaction intermediate and decomposed with deposition of solid
LTiCl(l-O)2TiClL (6) and LAlMeCl. Decomposition of 5 is
thermodynamically favorable; the estimated range of DH is −71
to −193 kJ mol−1.
Polymerization experiments
The trialkyltitanium complex LTiMe3 (3) and oxo-bridged com-
plexes LTiCl(l-O)2TiClL (6) and LTiMe(l-O)2TiMeL–toluene
(9a) showed low activity for polymerization of ethylene, although
the activity of the oxo-bridged complexes 6 and 9a was slightly
higher at 80 ◦C than that at room temperature (Table 3). This
result can be compared with the activity of LTiMe2 complexes.40
Budzelaar’s Ti(III) alkyls were found to be inactive in toluene and
displayed low to moderate activities in benzene upon activation
Complex LTiMe(l-O)2TiMeL (9) was obtained by reaction of
LTiMe3 and [Mes2Ga(OH)]2–THF or alternatively by hydrolysis
of LTiMe3 (3) in toluene, thus confirming the oxygen transfer to
the Ti(IV) centre by interaction of the titanium alkyl complex with
group 13 hydroxides.
Taking the obtained results into account we have proved that
the supporting ligands influence the stability of the l-oxo-bridged
Table 3 Ethylene polymerization data for the complexes 3, 4, 6 and 9a
Complex
Cocatalyst: MAO
AlMe3
25 ◦
Tmd/◦C
25 ◦
25 ◦
C
85 ◦
C
C
80 ◦
C
C
85 ◦
C
c
LTiMe3 (3)
1 × 103b
Not active
1 × 103
1.4 × 104
—
Not active
—
Not active
Not active
—
Not active
—
Not active
Not active
—
123.7
—
125.8
123.4
126.2
—
—
122.2
119.9
123.5
LTiCl2 (4)a
Not active
LTiCl(l-O)2TiClL (6)
LTiMe(l-O)2TiMeL–toluene (9a)
7 × 103
1–4 × 103
4–8 × 103
1 × 104
e
TiBz4
a LTiCl3 (1) and 2 to 4 equiv. AlMe3 formed LTiCl2, therefore only LTiCl2 (4) was tested. b Activity: g (polymer) mol (catalyst)−1 h−1 bar−1, reproducibility
is low as observed by different runs under the same conditions for 9a. Conditions: time 1 h, pressure 1 bar, Al to Ti varied from 500 : 1 to 2000 : 1,
volume 40 ml, 4.3–15.5 lmol of complex was used in different runs (see ESI† for details). c LTiMe3 (3) decomposed on heating in solution thus preventing
variable temperature study. d Determined by DSC. e TiBz4 was studied as a reference under the same conditions. Polymerization of ethylene using TiBz4
42
was studied in ref. 41. The reported activity of TiBz4–AlEt2Cl in n-heptane solution 7 × 103 g (polymer) mol (catalyst)−1 h−1 bar−1
.
4154 | Dalton Trans., 2007, 4149–4159
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Ti(IV)–Al(III) complexes. The alkyl complexes degrade more
slowly than the corresponding chloride complexes. Moreover the
investigated l-oxo-bridged complexes supported by the NacNac
ligand are unstable.
Complexes 3, 6 and 9a possess low catalytic activity in polymer-
ization of ethylene. We explain these phenomena by the instability
of the catalytically-active species.
dark brown. Then, the reaction mixture was stirred 2 h at −78 ◦C
and 16 h at room temperature. The resulting solution was filtered
to remove LiCl and concentrated (10–20 ml), whereupon a black
crystalline solid formed. The resulting mixture was filtered and the
black crystals of 1 washed with n-hexane (2 × 10 ml). Yield of 1
12.80 g (65%). Anal. calcd (found) for C29H40Cl3N2Ti (1): calcd C
◦
60.85 (60.90); H 7.17 (7.20); N 4.90 (4.94). Mp 222–223 C. EI-
MS: m/z (%) = 571 (2.7) [M]+, 535 (100) [M+–Cl], 528 (5.4) [M+-
C3H7], 498 (14.0) [M+–Cl2], 418 (36.7) [L]. 1H NMR (500.2 MHz,
C6D6, r.t.), dH (ppm) = 7.0–7.2 (m, 6 H, C6H3), 5.55 (s, 1H,
C(CH3)CHC(CH3)), 3.14 (sept, 4H, J = 6.75 Hz, CHMe2), 1.51 (s,
6H, C(CH3)CHC(CH3)), 1.42 (d, 12H, J = 6.7 Hz, CHMe2), 1.10
(d, 12H, J = 6.8 Hz, CHMe2). 13C NMR (125.76 MHz, toluene-d8,
r.t.), dC (ppm) = 170.6 (C(CH3)CHC(CH3)), 153.1 (C6H3), 140.3
(C6H3), 137.4 (C6H3), 107.7 (C(CH3)CHC(CH3), 29.1 (CHMe2),
26.3 (C(CH3)CHC(CH3), 24.8 (CH3), 24.5 (CH3), assignment was
done using 1H13C correlation and available references.46 IR (KBr,
cm−1, Nujol mull): 3071 (w), 3057 (w), 1933 (w), 1865 (w), 1797
(w), 1698 (w), 1587 (w), 1542 (vs), 1495 (s), 1464 (vs), 1450 (vs),
1384 (vs), 1361 (vs), 1346 (vs), 1313 (vs), 1282 (vs), 1254 (s), 1242
(vs), 1222 (s), 1178 (m), 1158 (m), 1097 (s), 1058 (m), 1040 (m),
1022 (s), 953 (w), 935 (s), 900 (w), 885 (w), 856 (s), 836 (w), 799
(vs), 763 (vs), 723 (m), 698 (m), 656 (w), 638 (w), 626 (m), 608 (w),
591 (w), 530 (m), 459 (vs), 435 (vs), 393 (vs), 368 (w), 352 (w).
Experimental
All operations were performed under an atmosphere of dry, O2-
free N2 employing both Schlenk line techniques and an MBrown
MB-150B inert atmosphere glove box. Solvents toluene, hexane
and tetrahydrofuran were dried over Na/benzophenone and
distilled under nitrogen prior to use. C6D6 was dried over K
and degassed, H2O was degassed under nitrogen. The compounds
LH with the ligand L = (2,6-iPr2C6H3NC(Me))2CH, NacNac),
LAlMe(OH), [Mes2Ga(OH)]2–THF were prepared by the known
literature methods.2b,43,44,45 Complex LAlMe(OLi) was prepared
from LAlMe(OH) and 1 equiv. of solid MeLi–(Et2O)0.137 at room
temperature in toluene, followed by crystallisation from hexane.
Complex LTiMe3 was prepared from LTiCl3 and 4 equiv. of
solid MeLi–(Et2O)0.137 at 0 ◦C in toluene, followed by filtration,
concentration and washing of the resulted product with n-hexane.9
The MeLi-Li(Et2O)0.137 was prepared from the known amount
(60 ml) of MeLi solution in Et2O (1.6 M) by evaporation of Et2O
in a dynamic vacuum to constant weight. The amount of Et2O
was determined by the weight of the resulting solid and confirmed
by 1H NMR. The purity of all compounds was checked by 1H and
13C NMR spectra. TiCl4 (Aldrich 99%) was distilled prior to use
in a dry and O2-free N2 atmosphere. Solutions of MeLi (1.6 M),
nBuLi (1.6 M) (Acros Organics), MAO (10 wt.%) and AlMe3
Preparation of LTiCl(l-O)2TiClL (6)
Toluene (40 ml) was added to LAlMe(OLi) (0.50 g, 1.0 mmol)
and LTiCl3 (0.60 g, 1.0 mmol). The resulting solution was stirred
for 10 min until all precursors dissolved and was left undisturbed
for 3 days at room temperature. The red block-shaped crystals
of 6 that deposited were filtered from the supernatant solution
and LiCl through a coarse frit. Yield of 6 0.45 g (80%). An
additional crop of 6 (0.1 g) can be obtained by crystallization
of the concentrated (5–8 ml) supernatant solution. Anal. calcd
(found) for C65H90Cl2N4O2Ti2: C 69.33 (69.17); H 8.00 (7.93);
N 4.97 (4.96). Mp 229 ◦C decomp. EI-MS: m/z (%) = 536
(10) [LTi2O2Cl-C3H9]+, 516 (15) [M-2Cl-NiPr2C6H3–iPr2C6H3–
iPr-MeCCHCMe]+, 498 (100) [LTiOOH or M-2Cl-NiPr2C6H3–
were used as received. H and 13C NMR spectra were recorded
1
using Bruker Avance DPX 200, Bruker Avance DRX 500 and
Varian INOVA 600 spectrometers. Chemical shifts are reported
in d units downfield from Me4Si with the solvent as the reference
signal. Mass spectra were recorded using a Finnigan MAT 8230
mass spectrometer, and elemental analyses were carried out at
the Analytical Laboratory of the Institute of Inorganic Chemistry
at the University of Go¨ttingen. Melting points were measured
in sealed capillary tubes under nitrogen and are not corrected.
Melting points of polymers were measured on Netzsch STA 409
PC spectrometer. For the preparation of 1, 2 see the ESI,‡ for
3 see ref. 9 and 4 see ref. 10. Compound LTiCl2(l-O)AlMeL (5)
is unstable and was not isolated (see the ESI, p. 4‡), complex
LTiMe2(l-O)AlMeL (7) observed in solution and characterized by
NMR (see ESI, p. 6–14‡), compound LTiMe(O) (8) was observed
in solution of LTiMe(l-O)2TiMeL (9).
1
HiPr2C6H3–HiPr–MeCCHCMe–CH4]+, 480 (10) [LTiO-H]+. H
NMR (599.74 MHz, C6D6, r.t.), dH (ppm) = 7.28, 7.17, 7.13–
6.99 (m, aryl), 5.19 (s, 2H, C(CH3)CHC(CH3)), 3.24 (sept, 4H,
J = 6.70 Hz, CHMe2), 2.85 (sept, 4H, J = 6.80 Hz, CHMe2),
2.10 (s, 3H, toluene), 1.44 (d, 12H, J = 6.70 Hz, CHMe2),
1.37 (s, 12H, C(CH3)CHC(CH3)), 1.18 (d, 12H, J = 6.70 Hz,
CHMe2), 0.91 (d, 12H, J = 6.70 Hz, CHMe2), 0.85 (d, 12H,
J = 6.80 Hz, CHMe2). 13C NMR (125.707 MHz, C6D6, r.t.),
dC (ppm) = 167.9 (C(CH3)CHC(CH3)), 150.6 (C6H3), 141.9
(C6H3), 140.4 (C6H3), 126.8 (C6H3), 124.6 (C6H3), 124.3 (C6H3),
103.1 (C(CH3)CHC(CH3), 29.1 (CHMe2), 28.1 (CHMe2), 27.4
(CH3), 26.1 (C(CH3)CHC(CH3), 25.2 (CH3), 24.6 (CH3), 24.5
(CH3), assignment was done using 1H13C correlation and available
reference.46
Preparation of LTiCl3 (1)
Dry toluene (250 ml) was added to LH (14.69 g, 35.1 mmol) in a
500 ml Schlenk flask. The mixture was cooled to −78 ◦C and 22 ml
of nBuLi solution in hexane (1.6 M, 35.1 mmol) was added drop by
drop. The reaction solution was stirred for 1 h at −78 ◦C, allowed
to warm to room temperature and stirred o◦vernight at room
temperature. This solution was cooled to −78 C and a solution
of TiCl4 (3.9 ml, 6.71 g, 35.3 mmol) in toluene (20 ml) was added
drop by drop. The color of the solution immediately changed to
Preparation of LTiMe(l-O)2TiMeL–toluene (9a)
Toluene (40 ml) was added to LAlMe(OH) (0.38 g, 0.8 mmol) and
LTiMe3 (0.41 g, 0.8 mmol). The resulting solution was heated
to 90 ◦C and kept at this temperature for 14 h, then filtered
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4149–4159 | 4155
©

View Article Online
and slowly concentrated (over 5 h) under vacuum. The yellow
block-shaped crystals of LTiMe(l-O)2TiMeL–toluene (9a) were
contaminated with LAlMe2 (NMR; see ref. 14), which deposited
while concentrating the solution. The molar relative intensities of
9a and LAlMe2 are nearly 1 : 1. The precipitate was filtered and
washed with toluene (4 × 5 ml) to give pure 9a (NMR). Yield of
9a 0.25 g (25%).
Reaction of LTiMe3 (3) and [Mes2Ga(OH)]2–THF to
LTiMe(l-O)2TiMeL–hexane (9b)
A mixture of LTiMe3 (0.44 g, 0.86 mmol) and [Mes2Ga(OH)]2–
THF (0.62 g, 0.86 mmol) was dissolved in toluene (40 ml).
The resulting solution was kept at 100 ◦C for 14 h under
stirring and finally filtered hot. The solvent was removed under
dynamic vacuum and the residue dissolved in hexane (40 ml). The
hexane solution was concentrated (10 ml) and left at 2 ◦C for
5 d. During this time colourless and yellow crystals deposited
from this solution. Melting point, NMR, EI-MS data for the
colourless crystals were identical with Mes3Ga, the NMR and
EI-MS data for the yellow crystals were identical with those
of 9b.
Preparation of LTiMe(l-O)2TiMeL–hexane (9b)
Alternatively, this reaction can be performed at room temperature
over 14 h. Evaporation of toluene afforded an oily tar, which was
soluble in n-hexane. Compound LTiMe(l-O)2TiMeL–hexane (9b)
was deposited from a concentrated hexane solution together with
LAlMe2 (NMR, ref. 14) over 4 weeks at room temperature, also
see ref. 13. 9b was purified by washing with n-hexane (5 × 5 ml).
Yield 0.42 g, 50%.
Reaction of LAlMe(OH) with 2 equiv. of LTiMe3 in toluene
afforded an oily intractable solid. The NMR spectrum of the
resulting solid in C6D6 showed resonances of 9a, LAlMe2 and
LAlMe(OH).
X-Ray crystallography
The crystallographic data for 1, 6, 9a and 9b are summarized in
Table 4.§ Intensities were collected on a Stoe–Siemens four-circle
diffractometer for 1. A Bruker SMART 1000 CCD diffractometer
was employed for 9a and 9b and a Bruker APEX2 CCD
diffractometer for 6. The data were reduced (SAINT)47 and cor-
rected for absorption (SADABS).48 The structures were solved by
direct methods, and refined against F2 with non-hydrogen atoms
anisotropic; hydrogen atoms were included using a riding model
or rigid methyl groups. All calculations were carried out using
SHELXTL software.49 Exceptions/special features: the toluene of
compound 6 is disordered over an inversion centre and that of
9a over a mirror plane, corresponding to the stoichiometry of a
mono- and disolvate, respectively, per molecule of the dinuclear
complexes; the hexane of compound 9b is ordered (but with high
U values) and lies across an inversion centre (monosolvate). Both
forms of 9 diffracted weakly and were low-temperature sensitive;
crystals (especially 9a) degraded by cracking during the data
collections. The atoms of hexane 9a have very high U values
and there is an improbably-short contact between C97 atoms
of adjacent molecules, implying incomplete occupancy. However,
a free refinement of the hexane occupancy led to a value of
Compound 9a (LTiMe(l-O)2TiMeL–toluene). Anal. calcd
(found) for C67H96N4O2Ti2: C 74.17 (74.20); H 8.86 (8.81); N 5.17
◦
(5.19). Mp 289 C decomp. EI-MS: m/z (%) = 993 (2) [M]+,
1
978 (100) [M+–Me], 963 (2) [M+–C2H6]. H NMR (500.2 MHz,
C6D6, r.t.), dH (ppm) = 7.31, 7.10–6.99 (m, aryl), 5.20 (s, 1H,
C(CH3)CHC(CH3)), 2.99 (sept, 2H, J = 6.80 Hz, CHMe2), 2.90
(sept, 2H, J = 6.80 Hz, CHMe2), 2.10 (s, toluene), 1.47 (s, 6H,
C(CH3)CHC(CH3)), 1.30 (d, 12H, J = 6.80 Hz, CHMe2), 1.22
(d, 12H, J = 6.80 Hz, CHMe2) overlapped with 1.22 (s, 3H,
Ti-Me), 1.00 (d, 12H, J = 6.80 Hz, CHMe2), 0.77 (d, 12H,
J = 6.80 Hz, CHMe2). 13C NMR (125.707 MHz, C6D6, r.t.), dC
(ppm) = 167.6 (C(CH3)CHC(CH3)), 150.1 (C6H3), 142.8 (C6H3),
140.6 (C6H3), 137.8 (C6H3, toluene), 129.3 (C6H3, toluene), 126.2
(C6H3), 125.6 (C6H3, toluene), 124.3 (C6H3), 123.8 (C6H3), 100.5
(C(CH3)CHC(CH3), 60.2 (Ti-Me), 26.8 (CH3), 25.6 (CHMe2),
25.2 (CH3), 24.5 (CH3), 24.3 (CH3), 21.4 (CH3, toluene), as-
˚
1
signment was performed using H13C correlation and available
0.91(1) A. Nevertheless, the description as a single ordered site
1
references.46 EI-MS and H NMR of 9b were similar to those
may be oversimplified. For the X-ray structure of [C(CMeN(2,6-
iPr2C6H3))2]2H2 (2) see ref. 8.
of 9a.
Density functional calculations
Hydrolysis of LTiMe3 (3)
The B3LYP50,51 functional, as implemented in the Gaussian
program package,52 was used throughout all the computations of
the NMR-shifts and the thermodynamic data. Due to the atoms
of the compound, a basis of type LANL2DZ53 for Ti and a 6–
31G(dꢀ,pꢀ) basis54,55 for the remaining atoms was used to achieve
the needed precision. In the first step the compound was fully
optimized to its equilibrium structure (coordinates are given in
the ESI‡) and then, making use of analytical derivatives, the
harmonic vibrational frequencies needed for the thermodynamic
data were calculated. Error of calculation is less than 20 kJ mol−1.
For compound LTiMe2(l-O)AlMeL (7) itself the NMR shifts were
calculated with the Gauge-Independent Atomic Orbital (GIAO)
method39 (see the ESI p. 29‡). Molecules LTiMe2(l-O)AlMeL
(7), LTiMe(O) (8) and LAlMe2 represent a true minimum on the
potential energy surface without negative IR frequencies.
LTiMe3 (0.58 g, 1.1 mmol) was dissolved in toluene (40 ml). Then
the solution was cooled to −79 ◦C and H2O (20.4 lL, 1.1 mmol)
was added with rigorous stirring over a period of 30 min. The
resulting mixture was warmed to room temperature and stirred
for 8 h. This dark green solution was filtered to remove the
colourless solid (0.1 g) and all volatiles were removed in a dynamic
vacuum. The oily residue was dissolved in n-hexane (20 ml), then
the hexane solution concentrated (10 ml) and left undisturbed at
room temperature for 1 d and then at +2 ◦C for 1 week, while
9b deposited (0.06 g) (determined by NMR and EI-MS). The
supernatant mixture was filtered from 9b and concentrated (2–
3 ml) and left at +2 ◦C for 1 week. During this time crystals of 3
and LH deposited (determined by NMR and EI-MS, data for LH
were identical to those in ref. 43).
4156 | Dalton Trans., 2007, 4149–4159
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Table 4 Crystallographic data and structure refinement details‡ for 1, 6, 9a and 9ba
1
6
9a
9b
Empirical formula
Mr
C29H40Cl3N2Ti
570.88
203(2)
C65H90Cl2N4O2Ti2
1126.11
100(2)
C74H98N4O2Ti2
1171.36
133(2)
C66H102N4O2Ti2
1079.32
193(2)
Mo Ka (0.71073)
0.5 × 0.3 × 0.3
Triclinic
¯
P1
8.9759(11)
13.0850(16)
13.5178(16)
88.349(4)
79.207
80.387
1537.7(3)
1
Temperature/K
˚
Radiation used (k)/A
Crystal size/mm
Crystal system
Mo Ka (0.71073)
1.00 × 0.80 × 0.20
Orthorhombic
Pnma
13.678(6)
21.795(14)
9.9457(13)
90
90
90
2965(2)
4
Mo Ka (0.71073)
0.22 × 0.20 × 0.06
Monoclinic
P21/c
13.4999(18)
12.9944(16)
17.350(2)
90
101.761(5)
90
2979.6(6)
2
Mo Ka (0.71073)
0.18 × 0.18 × 0.12
Monoclinic
C2/m
18.7080(17)
20.4316(18)
9.0082(8)
90
110.296(4)
90
3229.5(5)
2
Space group
˚
a/A
˚
b/A
˚
c/A
a (◦)
b (◦)
c (◦)
3
˚
V/A
Z
qcalcd/g cm−3
1.279
0.579
1.255
0.404
1.205
0.296
1.166
0.305
l/mm−1
F (000)
1204
3.52 to 22.52
97.0
−14 ≤ h ≤ 8
0 ≤ k ≤ 23
−10 ≤ l ≤ 0
3227
1204
1.54 to 30.51
100.0
−19 ≤ h ≤ 19
−18 ≤ k ≤ 18
−24 ≤ l ≤ 24
157240
1260
1.53 to 26.36
100.0
−23 ≤ h ≤ 23
−25 ≤ k ≤ 25
−11 ≤ l ≤ 11
20806
586
1.53 to 30.51
99.2
−12 ≤ h ≤ 12
−18 ≤ k ≤ 18
−19 ≤ l ≤ 19
32303
Scan range H (◦)
Completeness to Hmax (%)
Index ranges
Total reflections
Unique reflections
R(int)
Data/restrains/parameters
Goodness-of-fit on F2
R1, wR2 [I > 2r(I)]
R1, wR2 (all data)
1946
0.0425
9104
0.0557
3406
0.1551
9307
0.0317
1946/0/168
1.047
0.0546, 0.1367
0.0682, 0.1493
0.808, −0.698
9104/105/390
1.033
0.0332, 0.0814
0.0450, 0.0875
0.598, −0.566
3406/141/177
1.041
0.0760, 0.1920
0.1165, 0.2132
0.939, −0.634
9307/1/330
1.043
0.0543, 0.1476
0.0755, 0.1659
1.151, −0.621
−3
˚
Maximum/minimum electron density/e A
ꢀ
ꢀ
ꢀ
ꢀ
a wR2 = ( [w(Fo − Fc2)2]/ [Fo4])1/2, R1 = ꢁFo|–|Fcꢁ/ |Fo|, weight = 1/[r2(Fo2) + (A* P)2 + (B* P)], where P = (max (Fo2, 0) +2 * Fc2)/3; for
2
1 A = 0.0767, B = 4.1667; for 6 A = 0.0373, B = 1.6762; for 9a A = 0.105, B = 10.00; for 9b A = 0.091, B = 1.03.
Soc., Dalton Trans., 1996, 255; (q) T. J. Marks, Acc. Chem. Res., 1992,
Acknowledgements
25, 57; (r) R. F. Jordan, Adv. Organomet. Chem., 1991, 32, 325.
2 (a) G. Bai, S. Singh, H. W. Roesky, M. Noltemeyer and H.-G. Schmidt,
Support of the Deutsche Forschungsgemeinschaft and the
Go¨ttinger Akademie der Wissenschaften is gratefully acknowl-
edged. G.B. Nikiforov thanks the Alexander von Humboldt
foundation for a research fellowship.
J. Am. Chem. Soc., 2005, 127, 3449; (b) S. Singh, V. Jancik, H. W.
Roesky and R. Herbst-Irmer, Inorg. Chem., 2006, 45, 949; (c) P. M.
Gurubasavaraj, S. K. Mandal, H. W. Roesky, R. B. Oswald, A. Pal
and M. Noltemeyer, Inorg. Chem., 2007, 46, 1056; (d) V. Jancik and
H. W. Roesky, Angew. Chem., 2005, 117, 5220, (Angew. Chem., Int. Ed.,
2005, 44, 6016); (e) S. Singh and H. W. Roesky, Dalton Trans., 2007,
1360.
References
3 P. M. Gurubasavaraj, H. W. Roesky, P. M. V. Sharma, R. B. Oswald, V.
Dolle, R. Herbst-Irmer and A. Pal, Inorg. Chem., 2007, 46, 4599.
4 S. J. Coles, M. B. Hursthouse, D. G. Kelly, A. J. Toner and N. M.
Walker, J. Organomet. Chem., 1999, 580, 304.
1 (a) G. W. Coates, Chem. Rev., 2000, 100, 1223; (b) G. G. Hlatky, Chem.
Rev., 2000, 100, 1347; (c) H. Braunschweig and F. M. Breitling, Coord.
Chem. Rev., 2006, 250, 2691; (d) J. M. Thomas, R. Raja and D. W. Lewis,
Angew. Chem., 2005, 117, 6614, (Angew. Chem. Int. Ed., 2005, 44, 6456);
(e) S. Gambarotta, Coord. Chem. Rev., 2003, 237, 229; (f) J. R. Severn,
J. C. Chadwick, R. Duchateau and N. Friederichs, Chem. Rev., 2005,
105, 4073; (g) C. R. Baar, C. J. Levy, E. Y.-J. Min, L. M. Henling, M. W.
Day and J. E. Bercaw, J. Am. Chem. Soc., 2004, 126, 8216; (h) E. Y.-X.
Chen, M. V. Metz, L. Li, C. L. Stern and T. J. Marks, J. Am. Chem.
Soc., 1998, 120, 6287; (i) H.-H. Brintzinger, D. Fischer, R. Mu¨lhaupt, B.
Rieger and R. M. Waymouth, Angew. Chem., 1995, 107, 1255, (Angew.
Chem., Int. Ed. Engl., 1995, 34, 1143); (j) G. J. Clarkson, V. C. Gibson,
P. K. Y. Goh, M. L. Hammond, P. D. Knight, P. Scott, T. M. Smit,
A. J. P. White and D. J. Williams, Dalton Trans., 2006, 5484; (k) P. C.
Mohring and N. J. Coville, J. Organomet. Chem., 1994, 479, 1; (l) G. J. P.
Britovsek, V. C. Gibson and D. F. Wass, Angew. Chem., 1999, 111, 448,
(Angew. Chem., Int. Ed., 1999, 38, 428); (m) S. Saccheo, G. Gioia,
A. Grassi, D. E. Bowen and R. F. Jordan, J. Mol. Catal. A: Chem.,
1998, 128, 111; (n) W. Kaminsky, Metalloorganic Catalysts for Synthesis
and Polymerization: Recent Results by Ziegler–Natta and Metallocene
Investigations, Springer-Verlag, Berlin, 1999; (o) W. Kaminsky and M.
Arndt, Adv. Polym. Sci., 1997, 127, 144; (p) M. Bochmann, J. Chem.
5 (a) V. W. Day, T. A. Eberspacher, J. Hao, W. G. Klemperer and B. Zhong,
Inorg. Chem., 1995, 34, 3549, complex {[(C8H12)Rh]2[Ti(OnPr)6]};
(b) R. Gupta, A. Singh and R. C. Mehrotra, Indian J. Chem.,
Sect. A, 1991, 30, 592, complex Fe{Al(OtBu)4}{Ti2(OiPr)9}; (c) L. I.
Vyshinskaya, V. V. Drobotenko, T. P. Samarina and V. K.
Cherkasov, Zh. Obshch. Khim., 1991, 61, 824, complexes TiAl(OtBu)6,
Li[AlTi(OtBu)4Cl3], Li[AlTi(OtBu)4Br3], Ti(OtBu)4·AlBr3; (d) S.
Tudyka, K. Pflanz, F. Aldinger, H. Borrmann, P. Fischer and H. Brun-
ner, Z. Anorg. Allg. Chem., 1997, 623, 1163, complex TiAl2(OiPr)10;
(e) N. B. Sharma, S. Bansal, Y. P. Singh and A. Singh, Indian J. Chem.,
Sect. A, 2006, 45, 1375; (f) A. Rodriguez-Delgado and E. Y.-X. Chen,
Inorg. Chim. Acta, 2004, 357, 3911; (g) M. K. Sharma, A. Singh and
R. C. Mehrotra, Indian J. Chem., Sect. A, 2001, 40, 236; (h) M. K.
Sharma, A. Singh and R. C. Mehrotra, Indian J. Chem., Sect. A, 2000,
39, 410; (i) A. V. Firth, J. C. Stewart, A. J. Hoskin and D. W. Stephan,
J. Organomet. Chem., 1999, 591, 185; (j) C. Krueger, R. Mynott, C.
Siedenbiedel, L. Stehling and G. Wilke, Angew. Chem., 1991, 103,
1714, (Angew. Chem., Int. Ed. Engl., 1991, 30, 1668); (k) C. Blandy,
M. T. Sadani and D. Gervais, Synth. React. Inorg. Met.-Org. Chem.,
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4149–4159 | 4157
©

View Article Online
1978, 8, 381; (l) W. E. Lindsell and R. A. Parr, Polyhedron, 1986, 5,
1259.
6 K.-H. Wu and H.-M. Gau, J. Am. Chem. Soc., 2006, 128, 14808.
7 (a) D. Wisniewska, Z. Janas, P. Sobota and L. B. Jerzykiewicz,
Organometallics, 2006, 25, 6166; (b) Z. Janas, L. B. Jerzykiewicz, P.
Sobota, K. Szczegot and D. Wioniewska, Organometallics, 2005, 24,
3987.
26 D. C. Bradley and M. J. Hillyer, Trans. Faraday Soc., 1966, 62, 2374.
27 (a) The structurally-characterized heterometallic complexes contain-
ing the Al(l-O)Ti bond: TiAl2(OiPr)10 see ref. 5b, d TiBr2(Salen)-
AlMe2AlMe2Br and TiCl2(Salen)AlMe2AlMe2Br (SalenH2 = N,N-
bis(salicylidene)ethylenediamine); (b) S. J. Coles, M. B. Hursthouse,
D. G. Kelly, A. J. Toner and N. M. Walker, J. Organomet. Chem., 1999,
580, 304.
28 (a) K. G. Caulton and L. G. Hubert-Pfalzgraf, Chem. Rev., 1990, 90,
969; (b) D. C. Bradley, R. C. Mehrotra and D. P. Gaur, Metal Alkoxides,
Academic Press, London, 1978, and references cited therein; (c) A. H.
Cowley and R. A. Jones, Angew. Chem., 1989, 101, 1235, (Angew.
Chem., Int. Ed. Engl., 1989, 28, 1208); (d) W. A. Herrmann, N. W.
Huber and O. Runte, Angew. Chem., 1995, 107, 2371, (Angew. Chem.,
Int. Ed. Engl., 1995, 34, 2187).
29 D. L. Hildenbrand, High Temp. Mater. Sci., 1996, 35, 151.
30 Bond dissociation energy of Ti–Cl bond varied from 388 kJ mol−1
forTiCl4 to 508 kJ mol−1 for TiCl2.29 The values for the Ti(IV) complex
taken from ref. 22 and 29 were used in the estimation.
31 (a) M. Kh. Karapetyants and M. L. Karapetyants, Thermodynamic
Constants of Inorganic and Organic Compounds, Humphrey Science
Publ., Ann Arbor, 1970; (b) Table 4 in: A. Haaland, A. Hammel, K.-G.
Martinsen, J. Tremmel and H. V. Volden, J. Chem. Soc., Dalton Trans.,
1992, 2209.
8 Compound 2 was examined by X-ray structural analysis.§ The con-
nectivity of 2 was established, but the quality of the crystal was not
sufficient to refine the structure completely. Crystallographic data for
2 at 133(2) K: Empirical formula C58H81N4, Mr = 833.76, triclinic,
˚
˚
˚
space group P-1, a = 10.985(2) A, b = 23.721◦(5) A, c = 24.504(2) A,
◦
◦
3
˚
a = 60.95(3) , b = 89.70(3) , c = 77.84(3) , V = 5399.4(19) A ,
Z = 4, qcalcd = 1.026 g cm−3, l = 0.059 mm−1, F(000) 1826, index
ranges −9 ≤ h ≤ 12, −27 ≤ k ≤ 27, −28 ≤ l ≤ 28, 21527 reflections
measured (1.67 < h < 24.87◦), 15156 were independent (Rint = 0.0975),
observed reflections [I > 2r(I)] 4974, completeness to h = 24.87◦ 81.0%,
R1 = 0.1537 (for reflections with I > 2r(I)), R1 = 0.2883 (all data),
data/parameters 15156/1157, GOF = 1.187.
9 G. B. Nikiforov, H. W. Roesky, B. C. Heisen, C. Große and R. B.
Oswald, J Am. Chem. Soc., submitted for publishing.
10 G. B. Nikiforov, H. W. Roesky, J. Magull, T. Labahn, D. Vidovic, M.
Noltemeyer, H. G. Schmidt and N. S. Hosmane, Polyhedron, 2003, 22,
2669.
11 H. Zhu, J. Chai, C. He, G. Bai, H. W. Roesky, V. Jancik, H.-G. Schmidt
and M. Noltemeyer, Organometallics, 2005, 24, 380.
12 Complex 9 was deposited from the supernatant solution at room
temperature. Deposition started at r.t. after 15 d, volume of solution
20 ml or after 1 d, volume of solution 2 ml, or at 100 ◦C (1 d) and at
different ratios of the precursors (1 LTiMe3 : 1 equiv. LAlMe(OH) and
1 LTiMe3 : 2 equiv. LAlMe(OH)).
13 The supernatant hexane solution turned oily by concentration. No
deposition of solid was observed at 0 or −30 ◦C during 10 d from
the concentrated hexane solution. Complex 9b deposited from the
concentrated hexane solution over a period of 30 h at r.t.
14 B. Qian, D. L. Ward and M. R. Smith, III, Organometallics, 1998, 17,
3070.
32 P. B. Hitchcock, M. F. Lappert and D. S. Liu, J. Chem. Soc., Chem.
Commun., 1994, 2637.
˚
33 See distances in [LTiCl2(l-Cl)]2 (Ti–N 2.072, 1.986 A, L-benzamidinato
ligand CPh(NSiMe3)2) (a), L2Ti(NCMe3) (Ti–N 2.152(2) and
˚
2.156(3) A L-PhC(NSiMe3)2) (b), L2Ti(CH2SiMe3)2, Ti–N 2.006(2),
˚
˚
2.045(3) A, L2TiCl(CH2SiMe3) Ti–N 2.087(2), 2.091(2) A L-amino-
troponiminate ligand)(c); (a) D. Fenske, E. Hartmann and K. Dehnicke,
Z. Naturforsch., B: Chem. Sci., 1988, 43, 1611; (b) J. R. Hagadorn and
J. Arnold, Organometallics, 1998, 17, 1355; (c) D. P. Steinhuebel and
S. J. Lippard, Inorg. Chem., 1999, 38, 6225.
34 (a) S.-J. Kim, I. N. Jung, B. R. Yoo, S. Cho, J. Ko, S. H. Kim and S. O.
Kang, Organometallics, 2001, 20, 1501; (b) S.-J. Kim, D.-W. Choi, Y.-J.
Lee, B.-H. Chae, J. Ko and S. O. Kang, Organometallics, 2004, 23, 559.
35 (a) C. K. Ho, A. D. Schuler, C. B. Yoo, S. R. Herron, K. A. Kantardjieff
and A. R. Johnson, Inorg. Chim. Acta, 2002, 431, 71; (b) J. R. Petersen,
J. M. Hoover, W. S. Kassel, A. L. Rheingold and A. R. Johnson, Inorg.
Chim. Acta, 2005, 358, 687.
15 We have examined relative intensities of the c -H of L and Al-Me
resonances of the reaction mixture at different recycle delay D1: 1 s, 2, 4
and 6 s. The relative intensities were similar for D1: 2, 4 and 6 s, therefore
D1 2 s was chosen. In addition relative intensities of all resonances were
in agreement with the number of protons.
˚
36 The range of the Ti–C bond distances is 2.074(4)–2.201(7) A; see in ref.
33b and (a) S. Kleinhenz and K. Seppelt, Chem.–Eur. J., 1999, 5, 3573;
(b) K. H. Thiele, H. Windisch, H. Schumann and G. Kociok-Ko¨hn,
Z. Anorg. Allg. Chem., 1994, 620, 523; (c) H. Windisch, K. H. Thiele,
G. Kociok-Ko¨hn and H. Schumann, Z. Anorg. Allg. Chem., 1995, 621,
861.
16 All calculations were performed using OriginPro 7.5 SP5, OriginLab
Corporation,1991–2004.
17 Calculated second order constant using linear relationship 1/c(LTiMe3)
vs. time is similar to that using linear relationship for 1/c(LAlMe(OH))
vs. time first 1.5 h in agreement with the second order of reaction. Slope
of the line 1/c(LAlMe(OH)) vs. time is larger for the whole time period
(8 h), while slope of the line 1/c(LTiMe3) vs. time is nearly similar for the
reaction time period. The explanation is the following: the aluminium
hydroxide might potentially react with the alkyl metal compounds in
the system, including intermediate 7 and product LAlMe2 and complex
9, while LTiMe3 might react only with LAlMe(OH), hence the rate of
consumption of LAlMe(OH) could be higher than that of LTiMe3,
while at the beginning of the reaction LAlMe(OH) reacts preferentially
with LTiMe3, since relative concentration of the reaction products and
intermediates is low.
37 Compound {L2Ti(l-O)2}, L-Schiff base, Ti-(l-O) 1.837(3) and
˚
1.852(2) A:(a) M. L. N. Rao, H. Houjou and K. Hiratani, Chem. Com-
mun., 2002, 420; (b) Compound L2Ti(l-O)2TiL[g1-NC(Ph)N(SiMe3)2]
˚
(L = PhC(NSiMe3)2 Ti-(l-O) 1.790–1.883 A range, see ref. 33b.
38 (a) I. D. Brown, Structure and Bonding in Crystals, ed. M. O’Keefe,
A. Navrotsky, Academic Press, London, 1981; (b) I. D. Brown and D.
Altermatt, Acta Crystallogr., Sect. B, 1985, 41, 244; (c) Values used:
Ti(+4)–O(-2) Ro = 1.815, B = 0.37, Al(+3)–O(-2) Ro = 1.62, B = 0.37.
39 (a) J. L. Dodds, R. McWeeny and A. J. Sadlej, Mol. Phys., 1980, 41,
1419; (b) K. Wolinski, J. F. Hilton and P. Pulay, J. Am. Chem. Soc.,
1990, 112, 8251.
18 J. E. Kickham, F. Guerin and D. W. Stephan, J. Am. Chem. Soc., 2002,
40 P. H. M. Budzelaar, A. B. van Oort and A. G. Orpen, Eur. J. Inorg.
Chem., 1998, 1485.
41 (a) K. Zerche and K. H. Thiele, Z. Anorg. Allg. Chem., 1992, 612, 155;
(b) J. Giesemann, E. Ernst, A. Ernst and J. Ulbricht, Macromol. Chem.
Phys., 1986, 187, 1737; (c) J. Giesemann, A. Brandt and J. Ulbricht,
Acta Polym., 1983, 34, 623.
124, 11486.
19 The relationship E(Ti–O) = (38.493/r2) −7.904, where E is in eV and r
˚
in A is given in: Yu. Ya. Kharitonov and T. V. Gerzha, Zh. Fiz. Khim.,
1978, 52, 235.
20 (a) V. I. Telnoi, I. B. Rabinovich, V. D. Tikhanov, V. N. Latyaeva, L. I.
Vysginskaya and G. A. Razuvaev, Dokl. Akad. Nauk SSSR, 1967, 174,
1374.
42 J. C. W. Chien and J. T. T. Hsieh, J. Polym. Sci.: Polym. Chem. Ed.,
1976, 14, 1915.
21 M. F. Lappert, D. S. Patil and J. B. Pedley, J. Chem. Soc., Chem.
43 E. Krause, Ber. Dtsch. Chem. Ges., 1918, 51, 1447.
44 M. Stender, R. J. Wright, B. E. Eichler, J. Prust, M. M. Olmstead, H. W.
Roesky and P. P. Power, J. Chem. Soc., Dalton Trans., 2001, 3465.
45 J. Storre, A. Klemp, H. W. Roesky, H.-G. Schmidt, M. Noltemeyer, R.
Fleischer and D. Stalke, J. Am. Chem. Soc., 1996, 118, 1380.
46 F. Basuli, B. C. Bailey, L. A. Watson, J. Tomaszewski, J. C. Huffman
and D. J. Mindiola, Organometallics, 2005, 24, 1886.
47 SAINT 6.02, 1997–1999, Bruker AXS, Inc., Madison, Wisconsin, USA.
48 G. Sheldrick, SADABS, 1999, Bruker AXS, Inc., Madison, Wisconsin,
USA.
Commun., 1975, 830.
22 J. E. Huheey, Inorganic Chemistry, Principles of Structure and Reactiv-
ity, Harper and Row Publishers, New York, 1978.
23 D. Chakravorty, Modern Aspects of Solid State Chemistry, Plenum
Press, New York, 1980, 391.
24 E. R. Fisher, J. L. Elkind, D. E. Clemmer, R. Georgiadis, S. K. Loh, N.
Aristov, L. S. Sunderlin and P. B. Armentrout, J. Chem. Phys., 1990,
93, 2676.
25 J. A. M. Simoes and J. L. Beauchamp, Chem. Rev., 1990, 90, 629.
4158 | Dalton Trans., 2007, 4149–4159
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
49 G. Sheldrick, SHELXTL 5.1, 1997, Bruker AXS, Inc., Madison,
Wisconsin, USA.
50 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B, 1988, 37, 785.
51 B. Miehlich, A. Savin, H. Stoll and H. Preuss, Chem. Phys. Lett., 1989,
157, 200.
52 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C.
Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci,
M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E.
Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P.
Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. G. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03
(Revision C.02), Gaussian, Inc., Wallingford, CT, 2004.
53 P. J. Hay and W. R. Wadt, J. Chem.Phys., 1985, 82, 270.
54 G. A. Petersson and M. A. Al-Laham, J. Chem. Phys., 1991, 94, 6081.
55 G. A. Petersson, A. Bennett, T. G. Tensfeldt, M. A. Al-Laham, W. A.
Shirley and J. Mantzaris, J. Chem. Phys., 1988, 89, 2193.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4149–4159 | 4159
©