Reactivity of AlMe3 with titanium(IV) Schiff base complexes: X-ray structure of [Ti{(μ-Br)(AlMe2)}{(μ-Br)(AlMe2X)}(salen)] · C7H8 (X = Me or Br) and reactivity studies of mono-alkylated [Ti(Me)X(L)] complexes

Article abstract of DOI:10.1016/S0022-328X(98)01167-X

[TiCl₂(salen)] (1) reacts with AlMe₃ (1:2) to give the heterometallic Ti(III) and Ti(IV) complexes [Ti{(μ-Cl)(AlMe₂)}{(μ-Cl)(AlMe₂X)}(salen)] (X = Me or Cl) (2) and [TiMe{(μ-Cl)(AlCl₂Me)}(salen)] (3). Addition of diethyl ether to 3 affords [Ti(Me)Cl(salen)] (4). The analogous reaction of [TiBr₂(salen)] (5) gives the crystallographically characterised [Ti{(μ-Br)(AlMe₂)}{(μ-Br)(AlMe₂X)}(salen)] (X = Me or Br) (6) and [Ti(Me)Br(salen)] (7) in a single step, whilst the comparable reaction of [TiCl₂{(3-MeO)₂salen}] (8) with AlMe₃ yields [Ti(Me)Cl{(3-MeO)₂salen}] (9) with no evidence of titanium(III) species. Reactivity of both halide and methyl groups of 4 has been probed using magnesium reduction, SbCl₅ and AgBF₄ halide abstraction and SO₂ insertion reactions. Hydrolysis of [Ti(Me)X(L)] complexes affords μ-oxo species [TiX(L)]₂(μ-O) [X = Cl, L = salen (13); X = Br, L = salen (14); X = Cl, L = (3-MeO)₂salen (15)].

Full text of DOI:10.1016/S0022-328X(98)01167-X

Journal of Organometallic Chemistry 580 (1999) 304–312
Reactivity of AlMe3 with titanium(IV) Schiff base complexes: X-ray
structure of [Ti{(m-Br)(AlMe2)}{(m-Br)(AlMe2X)}(salen)] · C7H8
(X=Me or Br) and reactivity studies of mono-alkylated [Ti(Me)X(L)]
complexes
Simon J. Coles a, Michael B. Hursthouse a, David G. Kelly b,*, Andrew J. Toner b,
Neil M. Walker b
a Department of Chemistry, Uni6ersity of Wales College of Cardiff, PO Box 912, Cardiff CF1 3TB, UK
b Department of Chemistry and Materials, The Manchester Metropolitan Uni6ersity, Chester Street, Manchester M1 5GD, UK
Received 10 September 1998
Abstract
[TiCl2(salen)] (1) reacts with AlMe3 (1:2) to give the heterometallic Ti(III) and Ti(IV) complexes [Ti{(m-Cl)(AlMe2)}{(m-
Cl)(AlMe2X)}(salen)] (X=Me or Cl) (2) and [TiMe{(m-Cl)(AlCl2Me)}(salen)] (3). Addition of diethyl ether to 3 affords
[Ti(Me)Cl(salen)] (4). The analogous reaction of [TiBr2(salen)] (5) gives the crystallographically characterised [Ti{(m-
Br)(AlMe2)}{(m-Br)(AlMe2X)}(salen)] (X=Me or Br) (6) and [Ti(Me)Br(salen)] (7) in a single step, whilst the comparable reaction
of [TiCl2{(3-MeO)2salen}] (8) with AlMe3 yields [Ti(Me)Cl{(3-MeO)2salen}] (9) with no evidence of titanium(III) species.
Reactivity of both halide and methyl groups of 4 has been probed using magnesium reduction, SbCl5 and AgBF4 halide
abstraction and SO2 insertion reactions. Hydrolysis of [Ti(Me)X(L)] complexes affords m-oxo species [TiX(L)]2(m-O) [X=Cl,
L=salen (13); X=Br, L=salen (14); X=Cl, L=(3-MeO)2salen (15)]. © 1999 Elsevier Science S.A. All rights reserved.
Keywords: Titanium; Alkylation; Schiff base
lics, such as tetrabenzylzirconium [4]. Whilst simple
alkylated species can result from such reactions, re-
1. Introduction
duced metal complexes, ligand-alkylated complexes and
The structure and reactivity of Group 4 metal com-
partially characterised fluxional species have also been
plexes containing Schiff base ligands and alkyl groups
observed. Continued interest in the area stems from the
has been explored in several publications. Two general
apparent dependence of product distribution on both
approaches have been applied to such syntheses; viz the
reaction constituents and conditions. Our own studies
alkylation of Schiff base/chloride complexes with
Group 1 and 2 metal alkyls [1–3] and the direct reac-
have considered the complexes obtained by trimethyla-
luminium alkylation. In a preliminary communication
tion of Schiff base ligands with Group 4 organometal-
we noted the formation of trimetallic Al/TiIII/Al and
bimetallic Al/TiIV complexes from the reaction of
[TiCl2(salen)] and AlMe3 [5]. This work has been ex-
panded to consider the reactivity of structurally related
titanium(IV) Schiff base complexes with AlMe3, and
the products derived from reaction of the novel mono-
alkylated [Ti(Me)Cl(salen)] with a series of substrates.
* Corresponding author. Tel.: +44-161-247-1425; fax: +44-161-
247-1438.
E-mail address: d.g.kelly@mmu.ac.uk (D.G. Kelly)
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(98)01167-X

S.J. Coles et al. / Journal of Organometallic Chemistry 580 (1999) 304–312
305
1. Whilst the isolation of 2 and 3 has been previously
reported [5], the importance of reaction stoichiometry
and conditions is elaborated here. Thus, increasing the
reaction time to ca. 14 h at ambient temperature pro-
vided access to the reduced complex 2 in quantitative
yield. Conditions cannot be modified to afford 3 as the
sole product since reducing the reaction temperature to
ca. −78°C produced 2 and 3 at lower rates, whilst
using an equimolar ratio of AlMe3 produced the same
product distribution together with ca. 45% unreacted 1.
Increasing the Al:Ti stoichiometry does not affect the
distribution of 2 and 3. 2 is stable under inert atmo-
spheres in the solid state and at −20°C in non-polar
solvents, but reacts with THF or Et2O, producing an
immediate purple to blue colour change followed by
rapid decomposition. Crystallographic characterisation
[5] confirms the structure of 2 and indicates the pres-
ence of a substitutionally disordered site on the
AlMe2X moiety; crystallographic refinement being opti-
mised by a 0.65:0.35 Me:Cl distribution. Paramag-
netism of 2 is confirmed by magnetic susceptibility
measurements, Table 1. ESR spectra (77 K,
dichloromethane) are indicative of a d1 configuration,
although no hyperfine structure could be observed.
[TiMe{(m-Cl)(AlCl2Me)}(salen)] (3), which is insolu-
ble in hexanes and only moderately soluble in toluene,
Scheme 1. The activation of Group 4 Schiff base complexes with
trimethylaluminium.
1
is readily separated from 2. H-NMR is in accord with
the assignment of methyl moieties to titanium and
aluminium centres on the basis of resonances at l 0.98
and l −0.36, respectively [7,8]. Further evidence for
this formulation is derived from the reaction of 3 with
Et2O; on treatment of 3 with Et2O, the pendant
AlMeCl2 group is lost as its highly reactive solvent
adduct, Et2O · AlMeCl2, to afford [Ti(Me)Cl(salen)] (4)
in high yield (95%), Scheme 1. Et2O · AlMeCl2 can be
purified by distillation, 101–105°C at 0.1 mmHg [9] to
yield a colourless, highly pyrophoric, viscous liquid,
2. Results and discussion
2.1. Reaction of [TiX2(L)] and AlMe3
The reaction of [TiCl2(salen)] (1) [6] with AlMe3 (1:2)
in toluene/hexane affords the trimetallic titanium (III)
species
[Ti{(m-Cl)(AlMe2)}{(m-Cl)(AlMe2X)}(salen)]
(X=Me or Cl) (2), and the bimetallic titanium(IV)
complex [TiMe{(m-Cl) (AlCl2Me)}(salen)] (3), Scheme
Table 1
Analytical and physical data for products of the reaction TiX2(L)+2AlMe3
Complex
Colour
w(CꢀN)a
(cm−1
1H-NMR
(TiꢁCH3)/l
veff/vB
Analysis (%)
C
)
H
N
Cl
2
Purple
1624, 1598
1615
1.84
48.1
(47.7)
45.8
(45.2)
55.3
(55.9)
39.4
(40.2)
46.8
4.6
(5.2)
4.5
(4.5)
4.5
(4.7)
4.7
(4.6)
3.7
(4.0)
4.5
4.9
(5.4)
6.1
(5.9)
7.6
(7.6)
4.6
(4.5)
6.4
(6.2)
5.7
3
Orange–brown
Red–brown
Purple
−0.36, 0.98
4
1616
0.67
9.8
(9.7)
6
1628, 1598
1612
1.89
7 · 0.5CH2Cl2
9 · 0.5CH2Cl2
Red
0.79
0.78
(46.5)
50.9
Red
(50.2)
(4.7)
(6.0)
a Nujol mull.

306
S.J. Coles et al. / Journal of Organometallic Chemistry 580 (1999) 304–312
which is authenticated by comparison with the pub-
lished 1H-NMR data [10]. The presence of a single
1
methyl group in 4 is confirmed by the l 0.67 H-NMR
resonance. 4 shows remarkable thermal stability dis-
playing no propensity for alkyl migration to the Schiff
base ligand even after prolonged reflux in toluene or
dichloromethane. This behaviour contrasts markedly
with the rapid alkyl migration at ambient temperature
reported for the corresponding dimethyl complex [1].
The reaction of [TiBr2(salen)] (5) [6] with trimethyla-
luminium (1:2) in toluene:hexane (50:50) at ambient
temperature for ca. 12 h affords, after work-up, [Ti{(m-
Br)(AlMe2)}{(m-Br)(AlMe2X)}(salen)] (X=Me or Br)
(6) and [Ti(Me)Br(salen)] (7) in yields of 10 and 60%,
respectively, Scheme 1. Elemental analysis supports the
formulation of both complexes, whilst respective mag-
netic susceptibility and 1H-NMR data are consistent
with these structures, Table 1. Although representing
single crystal rather than bulk analysis, the formulation
is also consistent with the X-ray structure of 6 discussed
in Section 2.3. The latter refines the structural disorder
in the AlMe2X moiety as 0.74:0.26 Me:Br.
Recrystallisation of 7 from dichloromethane at −
20°C yielded solvated [Ti(Me)Br(salen)] as deep red
needles. Spectroscopic and physical properties of 7 are
entirely consistent with its chloride analogue, Table 1.
Scheme 2. Examples of the reactivity of [Ti(Me)Cl(salen)] (4).
1
solid recrystallised from dichloromethane to give sol-
Even crude materials show no H-NMR evidence of
vated [Ti(Me)Cl{(MeO)2salen}] (9).
high-field resonances indicative of terminal or bridging
aluminium methyl groups. Moreover, washing 7 with
diethyl ether does not result in the isolation of the
reactive adduct Et2O · AlMeBr2. Thus, the formation of
5 occurs either entirely without the formation of
[TiMe{(m-Br)(AlMeBr2)}(salen)] or with the latter rep-
resenting only a transient solution intermediate.
Qualitatively it is apparent that a reaction does occur
between [TiF2(salen)] [6] and trimethylaluminium under
analogous conditions to those used for the chloro and
bromo derivatives. The formation of an initial red–or-
ange solution from the yellow fluoride complex and the
absence of a deep purple product is suggestive of the
formation of titanium(IV) complex(es). Storage of the
filtered mother liquor at −20°C consistently gives a
deep blue solution and an intractable oily blue residue
after several days, perhaps indicative of slow reduction
to Ti(III). However, purification and acceptable charac-
terisation of these materials has thus far not proved
possible.
2.2. Reacti6ity of [Ti(Me)Cl(salen)]
Other workers have observed considerable instability
for Group 4 alkyls in the presence of co-ordinated
Schiff bases, often resulting in complex mixtures of
paramagnetic/fluxional species. The unprecedented
thermal stability 4 provides the opportunity to investi-
gate its reactivity without the complications of in-
tramolecular reactivity.
Stirring a heterogeneous mixture of 4 and magnesium
powder in THF for ca. 4 h gave a dark green coloured
solution,
Scheme
2.
Following
purification,
[Ti(THF)Me(salen)] (10) is isolated as a green micro-
crystalline solid in ca. 55% yield, Table 1. The complex
is stable at ambient temperature under an inert atmo-
sphere but is extremely susceptible to oxidation and/or
hydrolysis both in solution and the solid state. The
corresponding phenyl complex has been isolated from
the direct alkylation/reduction of 1 with PhLi [2].
Since ethylene polymerisation is believed to be
catalysed by cationic Group 4 complexes, synthetic
strategies have been developed to generate cationic
metallocenes [11]. It has also been possible to synthesise
cationic Schiff base complexes of zirconium, although
the latter are not catalytically active [4]. In an attempt
to abstract the chloride ligand from 4 and prepare the
first base-free cationic alkyl–titanium(IV) complex con-
Adding trimethylaluminium (two equivalents) to a
toluene:hexane
(50:50)
suspension
of
[TiCl2-
{(MeO)2salen}] (one equivalent) (8) [6] at ambient tem-
perature immediately gave a red–orange precipitate,
but no coloured solutions diagnostic of a titanium(III)
analogue to 2 and 6. Removal of the solvent in vacuo
and addition of THF to the dried solid immediately
resulted in a red–brown slurry. The mother liquor was
removed after ca. 10 min and the remaining red–brown

S.J. Coles et al. / Journal of Organometallic Chemistry 580 (1999) 304–312
307
taining a Schiff base ligand, one equivalent of antimo-
ny(V) chloride was added to a dichloromethane solu-
tion of 4. This method of chloride abstraction has been
utilised by Willey et al. in sequential halide abstractions
from TiCl4, [Cp2TiCl2] and [CpTiCl3] in acetonitrile
[12], and it was anticipated that such a system, incorpo-
rating the non-coordinating SbCl6− counterion would
allow direct investigation of the chemistry of the key
[TiMe(salen)]+ cation. However, the reaction leads to
the immediate precipitation of [Ti{(m-Cl)(SbCl5)}Cl-
(salen)] (11), a unique antimony–titanium complex,
Scheme 2. The structure of 11, which is based on
analytical and spectroscopic data, can be considered as
the [TiCl2(salen)] (1) unit acting as a Lewis base
through a chloride bridge to a pendant antimony pen-
tachloride moiety, Table 2. This type of bonding is
observed in 2 and 6 and has been postulated for 3.
Allowing a CH2Cl2 solution of 11 to stand in air results
in the precipitation of a reddish solid and an uniden-
tified white solid. Washing the former with ace-
tone:water (50:50) gave analytically pure 1. 1 is also
isolated from the reaction of 4 with AgBF4 (1:1) after
reaction in the absence of light.
[14,15]. However, the oxophilicity of titanium and its
inclination to coordinate to hard ligands would suggest
the preferential insertion of SO2 to form an O-sulphi-
nate complex. Such reactivity can be inferred from the
IR spectrum of 12; of the three bands not attributable
to the Schiff base ligand at 853, 975 and 1032 cm−1
;
the lowest stretch suggests the presence of a new tita-
nium–oxygen bond in 12. Such an assignment is in
agreement with those reported previously [15,16], and
indicative of partial multiple bond character [17]. The
bands at 975 and 1032 cm−1 can be assigned to ws(SꢁO)
and was(SꢁO), respectively, which confirms 12 as an
h1-O-sulphinate complex [18].
Oxo-bridged dimeric and polymeric complexes based
on Group 4 Schiff base complexes have previously been
obtained by the partial hydrolysis of titanium(IV)
halides [19] or the oxidation of titanium(III) halides
[20]. The asymmetric halide/methyl substitution offers a
rational and high yield route to dimeric complexes.
Thus, if a CH2Cl2 solution of 4 is allowed to stand in
the air, [TiCl(salen)]2(m-O) (13) is isolated as a bright
red crystalline solid in quantitative yield. Similarly,
[TiBr(salen)]2(m-O) (14) and [TiCl{(MeO)2salen}]2(m-O)
(15) are isolated from solutions of 7 and 9, respectively.
Over several hours, a CH2Cl2 solution of 12 also under-
goes hydrolysis, which results in the precipitation of
[TiCl(salen)]2(m-SO4) (16) as an air stable solid. A struc-
ture for 16 is tentatively proposed in Scheme 2. The
chemistry of titanium(IV) with the sulphate ion is lim-
ited to polymeric species, e.g. TiOSO4 · H2O. There has
been a single report of a titanium complex containing
SO24− but the bonding mode of the sulphate ion in
polymeric [CpTi(m-O)(SO4)0.5(H2O)]n was not estab-
lished [21].
Sulfur dioxide insertion into metal-alkyls represents
an extensively investigated form of reactivity [13]. Thus,
when a deep red CH2Cl2 solution of 4 is exposed to an
atmosphere of sulphur dioxide at ambient temperature
the solution gradually becomes a pale red–orange in
colour, Scheme 2. Work-up and storage at −20°C
affords [Ti(SO2Me)Cl(salen)] · 0.5CH2Cl2 (12) as pale
1
red crystals. H-NMR of 12 displays a singlet at l 2.26
that is typically diagnostic of a sulphur bound methyl
group and is comparable with others reported for tita-
nium methanesulphinate complexes, e.g. [Cp2Ti(O2-
SMe)2] at l 2.48 and [CpTiMe2(O2SMe)2] at l 2.55
Table 2
Analytical and physical data for products of the reactions of [Ti(Me)Cl(salen)]
Complex
Colour
w(CꢀN)a
w(TiꢁO)
veff/vB
Analysis (%)
C
(cm−1
)
(cm−1
)
H
N
Cl
10
Green
Orange
Pale red
Red
1608
1631
1619
1633
1620
1624
1630b
1.76
63.1
(62.8)
27.6
(27.9)
44.6
(44.6)
53.5
(53.7)
44.1
(44.5)
50.9
(51.7)
48.7
6.4
(6.2)
2.0
(2.0)
3.5
(3.8)
4.0
(3.9)
3.4
(3.4)
4.3
(4.3)
3.4
7.3
(7.0)
4.1
(4.1)
6.3
(6.0)
7.6
(7.8)
6.2
(6.3)
7.3
(6.8)
7.2
11
35.9
(36.1)
12 · 0.5CH2Cl2
853
760
735
754
13
14 · CH2Cl2
Red
15
16
Red
Orange
(48.3)
(3.5)
(7.0)
a Nujol mull.
b w(SꢁO) 971, 1044 and 1090 cm−1
.

308
S.J. Coles et al. / Journal of Organometallic Chemistry 580 (1999) 304–312
Fig. 1. The molecular structure of 6 showing the atomic numbering scheme.
Treatment of 4 in toluene at −78°C with phenyl-
lithium gave a dark brown solution, which after filtra-
isostructural with 2; being demonstrated by the com-
parable bond lengths and angles within the two com-
plexes that only differ significantly as a consequence of
the inclusion of the larger bromide ion, Table 3. Initial
characterisation of 2 might have suggested that such a
trimetallic species was one of several possible titaniu-
m(III) species. Moreover, the single disordered site
might have again been indicative of crystal packing
selection rather than bulk properties. However, since 6
is also trimetallic and exhibits the same single site of
disorder, it must be postulated that 2 and 6 represent
stable and structurally representative samples of their
bulk. Furthermore, it must be speculated that the mech-
anism or stoichiometry of complex formation or post-
formation reactivity limits exchange of aluminium
substituents to a single site. Since the Al(1) moiety is
conformationally locked by the halide bridge and the
O(2)ꢁAl(1), interaction substitution exchange should be
mechanistically difficult. In contrast, the Al(2) contain-
ing group, whilst also four-coordinate, can undergo free
rotation and presumably associative or dissociate ex-
change.
tion and standing at −20°C overnight, yielded dark
1
brown crystals. The H-NMR spectrum (C6D6) of this
material indicated complex solution state chemistry
similar to that observed from the reaction of 1 with
alkylating agents [22]. The absence of an imino proton
resonance and the disappearance of the CꢀN IR
stretching vibration indicates a transformation involv-
ing alkyl migration(s) and reduction of the imino
groups of the Schiff base. Evidently, the trans chloride
in 4 more effectively stabilises the single methyl present
more readily than a second alkyl/aryl group.
2.3. Solid state structures of 2 and 6
2 and 6 can be recrystallised from toluene/hexane to
produce crystals suitable for X-ray diffraction. The
structure of 2 has been briefly reported [5]; however,
full data for 6 are presented with detailed comparisons
between the two structures. The structure of 6 indicates
a formulation of [Ti{(m-Br)(AlMe2)}{(m-Br)(AlMe2X)}-
(salen)], Fig. 1, which can be considered as aluminiu-
m(III)–titanium(III)–aluminium(III) moieties of [Me2-
Al(1)], [Ti(salen)] and [Al(2)Me2X (X=Br/Me)]. These
substructures are linked by m2-bromide ions that bond
into the apical positions of the central titanium atom.
The group X contains disordered methyl and bromide
groups modelled on an occupancy of 0.26 (Br) and 0.74
(Me). It is notable that this structure is completely
The distortion resulting from aluminium–oxygen da-
tive bonding is crucial to the anomalous bond lengths
observed with 2 and 6 and to the significant distortions
from the idealised octahedral bonding for titanium.
Within both complexes, the m2-bonded oxygen shows
elongation of the TiꢁO(2) bond beyond that reported in
simple Ti(III) and Ti(IV) Schiff base complexes, Table
3. Shortening of the trans TiꢁN(1) is a natural and

S.J. Coles et al. / Journal of Organometallic Chemistry 580 (1999) 304–312
309
obvious consequence of this elongation. The asymmetry
in the N2O2 salen ligand is also reflected in the signifi-
cantly different CꢀN distances observed in both 2 and
6, although both distances show the increased bond
distance from the free salen ligand [26]. The alu-
minium–oxygen interactions in both complexes should
not be considered trivial since both lie between the
AlꢁO distances reported in the reduced Schiff base
complex [(AlMe)salphan(AlMe2)2] [salphan=N,N%-
bis(o-hydroxybenzyl)-1,3-diaminopropane] [27] and
those of [Al(salen)Et] [28]. The ability of coordinated
Schiff base ligands to act as Lewis bases with a second
metal is not without precedent. This facility has been
observed with both iron(III) and nickel(II), although
similar reactivity was not recorded with titanium(III) or
(IV) [29,30].
salen}]; reactivity that may relate to the greater electron
density afforded to the TiIV centre by the p-donating
fluoride or the electron donating MeO groups. Probing
the reactivity of mono-methylated titanium(IV) com-
plexes using 4, produced a series of conventional prod-
ucts from zinc reduction and sulphur dioxide insertion.
Attempts to produce cationic Schiff base complexes
leads to the regeneration of [TiCl2(salen)]. Hydrolysis of
[Ti(Me)X(salen)] complexes occurs specifically at the
methyl group affording the dimeric m-oxo complexes.
Such reactivity affords a high yield route to dimeric
complexes and potentially through further alkylation to
the rational synthesis of oligomeric complexes.
4. Experimental
3. Conclusion
4.1. Crystal structure determination of 6
All the titanium(IV) complexes considered appear to
react with AlMe3, although characterisable products
have not been obtained in every case. Product distribu-
tion is dependent on both the halide and Schiff base
ligand present in the titanium(IV) precursor. Reduction
appears to occur less readily for the fluoride analogue,
[TiF2(salen)], and is not observed for [TiCl2{(MeO)2-
Recrystallisation of 6 from toluene:hexane (50:50)
produced at −20°C over ca. 4 weeks air-sensitive
crystals of 6 · C7H8. Crystal dimensions 0.07×0.14×
0.14 mm3. Intensity data were recorded on a FAST TV
area detector diffractometer at 150 K following previ-
ously described procedures [31]. The structure was
solved by direct methods (SHELXS) [32] and refined by
Table 3
Important bond lengths and angles in complexes 2 and 6
Bonding featurea
Complex 2
Complex 6
Literature comparisonsb
˚
Bond lengths (A)
TiꢁX(1)
TiꢁX(3)
TiꢁO(1)
TiꢁO(2)
TiꢁN(1)
TiꢁN(2)
Al(1)ꢁO(2)
Al(1)ꢁX(1)
Al(2)ꢁX(3)
2.579(5)
2.512(5)
1.849(4)
2.057(3)
2.127(4)
2.174(5)
1.914(4)
2.287(3)
2.359(4)
2.670(3)
2.589(3)
1.849(4)
2.045(3)
2.106(4)
2.186(5)
1.872(4)
2.421(2)
2.530(3)
2.148 in [Ti(py)Cl(salen)]c
1.835 in [TiCl2(acen)]d
1.867 in [TiF2(salen)]e
2.138 in [Ti(py)Cl(salen)]c
2.141 in [TiCl2(salen)] · THFf
Bond angles (°):
Ti-centered pseudo-Oh angles
Minimum
Maximum
Al(1)-centered pseudo-Td angles
N(2)ꢁTiꢁN(1)=76.6(2)
O(2)ꢁTiꢁO(1)=118.25(13)
N(2)ꢁTiꢁN(1)=76.9(2)
O(2)ꢁTiꢁO(1)=117.4(2)
Minimum
Maximum
Metal bridging halide angles
O(2)ꢁAl(1)ꢁCl(1)=90.0(2)
C(21)ꢁAl(1)ꢁC(20)=123.9(3)
O(2)ꢁAl(1)ꢁBr(1)=89.49(14)
C(17)ꢁAl(1)ꢁBr(1)=112.8(2)
TiꢁX(1)ꢁAl(1)
TiꢁX(3)ꢁAl(2)
82.17(11)
119.46(13)
77.82 (8)
114.72(7)
a X=Cl (2), X=Br (6).
b
˚
Average bond lengths (A).
c [23].
d [24].
e [6].
f [25].

310
S.J. Coles et al. / Journal of Organometallic Chemistry 580 (1999) 304–312
1
full-matrix least-squares on F2
(
SHELX-93), with all
3: H-NMR (C6D6): l −0.36 (s, 3H, AlꢁMe), 0.98 (s,
3H, TiꢁMe), 3.08 (dd, 2H, CH2) 3.70 (dd, 2H, CH2),
6.60 (m, 4H, Ph), 7.00 (m, 4H, Ph) and 7.70 (s, 2H,
CH).
non-hydrogen atoms being refined anisotropically [33].
Data were corrected for absorption effects using DI-
FABS with maximum and minimum correction factors
of 1.08 and 0.93, respectively [34].
Crystal data. C27.74H36.21Al2Br2.26N2O2Ti, M=712.30,
4.4. [Ti(Me)Cl(salen)] (4)
˚
monclinic, a=10.186(9), b=14.936(2), c=21.044(3) A,
3
˚
i=95.05(11)°, V=3188.9(30) A ,
Dcalc. =1.484 g
Diethyl ether (ca. 30 cm3) was decanted onto 3 and
immediately the solid darkened to a red–brown. After
stirring for 10 min, the almost colourless Et2O was
decanted from the solid. After washing with Et2O and
drying in vacuo, 2.2 g (95%) of analytically pure 4 was
collected. Removing the solvent from the filtrate in
vacuo yielded an oily residue that was distilled, 101–
105°C at 0.1 mmHg, to give Et2O · AlCl2Me as a highly
pyrophoric, colourless liquid.
cm−3, T=150(2) K, space group P21/c, Z=4, v(Mo–
Ka)=31.85 mm−1, unique total number of data=
6995, unique number of observed data=1916
[I\2|(I)], R1=0.0454, wR=0.1031.
4.2. Synthetic procedures
Inert atmosphere glovebox and Schlenk-line tech-
niques were used throughout the preparative proce-
dures unless otherwise indicated. Nitrogen and argon
were purified prior to use by passage through two
Et2O · AlCl2Me: 1H-NMR (C6D6): l −0.58 (s, 3H,
AlꢁMe), 0.64 (t, 6H, Me) and 3.26 (q, 4H, CH2).
˚
columns containing MnO and 4 A molecular sieves,
respectively. Solvents were pre-dried, then refluxed over
the appropriate drying agent, and subsequently distilled
under nitrogen or argon. Benzene-d6 was pre-dried over
[Ti(Me)Cl(salen)] (4): 1H-NMR (CH2Cl2): l 0.67 (s, 3H,
Me), 4.00 (dd, 2H, CH2), 4.37 (dd, 2H, CH2), 6.90–7.10
(m, 4H, Ph), 7.50–7.70 (m, 4H, Ph) and 8.50 (s, 2H,
1
˚
4 A molecular sieves, degassed, and stored over
CH). H-NMR (C6D6): l 1.07 (s, 3H, Me), 3.08 (dd,
sodium–potassium alloy. Dichloromethane-d2 and
chloroform-d1 were refluxed over calcium hydride or
phosphorus pentoxide and stored in ampoules over 4 A
molecular sieves. NMR spectra were recorded at 270
MHz. Magnetic susceptibility measurements were made
in sealed tubes using a Johnson Matthey Faraday bal-
ance. Elemental analysis and important characterisa-
tion data are given in Table 1. Syntheses of 1, 5, 8 and
[TiF2(salen)] are previously reported [6].
2H, CH2), 3.76 (dd, 2H, CH2), 6.60–6.85 (m, 4H, Ph),
7.03–7.30 (m, 4H, Ph) and 7.78 (s, 2H, CH). EI+ MS
(70 eV): m/z 365 (10%, [Ti(Me)Cl(salen)]+,
protonated).
˚
4.5. [Ti{(v-Br)(AlMe2)}{(v-Br)(AlMe2X)}(salen)]
(X=Me or Br) (6) and [Ti(Me)Br(salen)] (7)
In a similar preparation to 3, [TiBr2(salen)] (5) was
treated with a toluene solution of AlMe3 over ca. 10
min, and stirred for a further 12 h until all 5 was
consumed. The resulting purple solution was filtered
from a brown precipitate. The solid residue was then
washed successively with hexane (3×20 cm3) and these
extracts were added to the original mother liquor until
they were colourless. This was placed at −20°C and
over a period of ca. 4 weeks gave 6 · C7H8 (ca. 10%) as
deep purple pyrophoric crystals. Following an identical
preparative procedure, the purple toluene solution of 6
was removed by filtration, layered with hexane and
allowed to stand at −20°C overnight to give analyti-
cally pure 6. The remaining solid was dried in vacuo for
ca. 4 h to give crude [Ti(Me)Br(salen)] (7) (ca. 60%).
Several recrystallisations from dichloromethane gave
[Ti(Me)Br(salen)] · 0.5CH2Cl2 as dark red moisture-sen-
sitive needles.
4.3. [Ti{(v-Cl)(AlMe2)}{v-Cl)(AlMe2X)}(salen)] (2)
(X=Me or Cl) and [TiMe{(vCl)(AlCl2Me)}(salen)] (3)
A toluene:hexane (1:1) suspension of 1 (7.0 g, 18
mmol) was treated with a toluene solution of AlMe3
(2.0 M, 18.2 cm3, 36 mmol) over a period of ca. 20 min.
Immediately, a red–purple solution was observed. The
reactants were stirred for a further 30 min at ambient
temperature and an orange–brown solid precipitated
from solution. The deep purple solution was filtered
and the solid residue was then washed successively with
hexane (ca. 6×30 cm3) and these extracts were added
to the original supernatant until they were colourless.
After storage at −20°C, 500 mg (13%) of 2 · C7H8 were
collected as large purple pyrophoric crystals. Solvent-
free 2 was isolated from less polar toluene–hexane
mixtures (25:75) as purple needles. The remaining or-
ange–brown solid was dried in vacuo for ca. 6 h to give
3.2 g (55%) of the air- and moisture-sensitive 3. Recrys-
tallisation of 3 from a dilute toluene solution at −20°C
gave an analytically pure orange–brown microcrys-
talline solid.
1
(7): H-NMR (CDCl3): l 0.79 (s, 3H, Me), 4.11 (dd,
2H, CH2), 4.30 (dd, 2H, CH2) 6.87–7.10 (m, 4H, Ph)
7.50–7.80 (m, 4H, Ph) and 8.51 (s, 2H, CH). FAB+
MS (NOBAH matrix): m/z 394 (5%, [TiBr(salen)]+).

S.J. Coles et al. / Journal of Organometallic Chemistry 580 (1999) 304–312
311
4.6. [Ti(Me)Cl{(MeO)2salen}] (9)
In similar preparation to 3, AlMe3 and
orange supernatant was decanted. Precipitated 11 was
then washed successively with copious quantities of dry
CH2Cl2 (ca. 250 cm3) and these extracts were added to
the original. After standing in air for 12 h a white
insoluble material was removed by filtration. The solvent
was removed in vacuo to yield [TiCl2(salen)] (1) (0.8 g,
85%).
a
[TiCl2{(MeO)2salen}] (8) yielded a light red–orange
solution and a red–orange precipitate within ca. 5 min.
After stirring for a further 30 min, the mother liquor was
removed by decantation. The solid was dried in vacuo
and THF added, resulting in the formation of a red–
brown solid. The solvent was removed after ca. 10 min
and the residue was extracted into CH2Cl2 (ca. 3×20
cm3), filtered and stored at −20°C to give 9 · 0.5CH2Cl2
(2.0 g, 52%).
1: 1H-NMR (CDCl3): l 4.25 (s, 4H, CH2), 6.85–7.13 (m,
4H, Ph), 7.33–7.61 (m, 4H, Ph) and 8.39 (s, 2H, CH).
Found: C, 49.8; H, 3.6; N, 7.5. Calc. for 1,
C16H14N2O2Cl2Ti: C, 49.8; H, 3.6; N, 7.3. w(CꢀN) 1612
cm−1
.
9: 1H-NMR (CDCl3): l 0.78 (s, Me), 3.45 (m, CH2), 4.00
(s, MeO), (m, CH2), 6.95–7.15 (m, Ph) and 8.45 (s, CH).
4.10. [Ti(SO2Me)Cl(salen)] (12)
A CH2Cl2 solution (ca. 50 cm3) of 4 (500 mg, 1.4
mmol) was exposed to SO2 and over a period of ca. 10
min and the solution turned a red–orange in colour.
Solvent was reduced to ca. 20 cm3 and placed at −20°C.
[Ti(SO2Me)Cl(salen)] · 0.5CH2Cl2 (400 mg, 70%) was
collected by filtration and dried in vacuo.
4.7. [Ti(THF)Me(salen)] (10)
A CH2Cl2 solution (40 cm3) of 4 (0.5 g, 1.4 mmol) was
added to an excess of magnesium powder at room
temperature and the resulting heterogeneous mixture
was stirred vigorously. After ca. 30 min, a dark green
solution was apparent, stirring was continued for a
further ca. 3.5 h. The solution was filtered, anhydrous
1,4-dioxane was added, and the mixture cooled to
−20°C. The green mother liquor was then filtered from
a small quantity of insoluble material, the volume
reduced to ca. 20 cm3 in vacuo. 10 (0.30g, 55%) was
collected as a green microcrystalline solid and dried in
vacuo.
12: 1H-NMR (CDCl3): l 2.26 (s, 3H, Me), 3.87 (dd, 2H,
CH2), 4.32 (dd, 2H, CH2), 6.59–7.05 (m, 4H, Ph),
7.41–7.67 (m, 4H, Ph) and 8.16 (s, 2H, CH).
4.11. [TiX(salen)]2(v-O) (13) (14) (15) (16)
A CH2Cl2 solution of 4 (200 mg, 5.49 mmol) was
allowed to stand in air and after 12 h yielded red
crystalline 13 (200 mg, ca. 96%), which was collected,
washed with sodium dried Et2O and dried in vacuo. A
similar preparation yielded 14 · CH2Cl2, 15 and 16.
14 · CH2Cl2: 1H-NMR (CDCl3) l 4.30 (s, 4H, CH2),
6.85–7.20 (m, 4H, Ph), 7.45–7.70 (m, 4H, Ph) and 8.46
(s, 2H, CH).
4.8. [TiCl{(v-Cl)(SbCl5)}(salen)] (11)
A CH2Cl2 solution of antimony pentachloride (1.0 M,
2.74 cm3, 2.74 mmol) was added, with stirring, to a
CH2Cl2 solution of 4 (1.0 g, 2.74 mmol). Instantly, a tan
solid precipitated from solution to leave a very pale
orange supernatant. Stirring was continued for a further
20 min with no observed changes. The supernatant was
then removed by filtration and the solid residue was
washed with CH2Cl2 (ca. 3×20 cm3). The solid was then
dried in vacuo to give ca. 1.2 g (60%) of analytically pure
11. The combined CH2Cl2 extracts were filtered and
layered with toluene. After standing at −20°C overnight
11 · 1.5C7H8 was collected as an orange crystalline mate-
rial (ca. 300 mg).
15: 1H-NMR (CDCl3): l 3.90 (s, 6H, MeO), 4.23 (s, 4H,
CH2), 7.07 and 7.11 (m, 6H, Ph) and 8.36 (s, 2H, CH).
16: 1H-NMR could not be obtained as a consequence of
the insolubility of 16 in all common solvents.
References
1
[1] E. Solari, C. Floriani, A. Chiesi-Villa, C. Rizzoli, J. Chem. Soc.
Dalton Trans. (1992) 3670.
[2] C. Floriani, E. Solari, F. Corazza, A. Chiesi-Villa, C. Guastini,
Angew. Chem. Int. Ed. Engl. 28 (1989) 64.
[3] M. Mazzanti, S. Gambarotta, C. Floriani, A. Chiesi-Villa, C.
Guastini, Inorg. Chem. 25 (1986) 2308.
[4] E.B. Tjaden, D.C. Swenson, R.F. Jordan, J.L. Petersen,
Organometallics 14 (1995) 371.
[TiCl{(m-Cl)(SbCl5)}(salen)] (11): H-NMR (CD2Cl2): l
4.57 (s, 4H, CH2), 7.11–7.42 (m, Ph), 7.45–7.54 (m, Ph)
and 8.90 (s, 2H, CH). Some phenyl resonances are
obscured due to the presence of co-crystallised toluene.
Removal of the solvent in vacuo gave an analytical
sample.
[5] D.G. Kelly, A.J. Toner, N.M. Walker, S.J. Coles, M.B. Hurst-
house, Polyhedron 15 (1996) 4307.
4.9. Hydrolysis of 11
[6] D.G. Kelly, A.J. Toner, N.M. Walker, S.J. Coles, M.B. Hurst-
house, J. Chem. Soc. Dalton Trans. (1998) 3489.
11 was prepared (1.0 g scale) and the resulting pale

312
S.J. Coles et al. / Journal of Organometallic Chemistry 580 (1999) 304–312
[7] J.J. Eisch, A.M. Piotrowski, S.K. Brownstein, E.J. Gabe, F.L.
Lee, J. Am. Chem. Soc. 107 (1985) 7219.
[8] (a) J. Eisch, J. Chem. Res. (S) (1992) 214. (b) F.N. Tebbe, G.W.
Parshall, G.S. Reddy, J. Am. Chem. Soc. 100 (1978) 3611.
[9] S. Takeda, R. Tarao, Bull. Chem. Soc. Jpn. 38 (1965) 1567.
[10] T. Mole, Aust. J. Chem. 17 (1964) 1050.
[21] L. Saunders, L. Spirer, Polymer 6 (1965) 635.
[22] C. Floriani, Polyhedron 8 (1989) 1717.
[23] M. Pasquali, F. Marchetti, A. Landi, C. Floriani, J. Chem. Soc.
Dalton Trans. (1978) 545.
[24] J.-M. Rosset, C. Floriani, M. Mazzanti, A. Chiesi-Villa, C.
Guastini, Inorg. Chem. 29 (1990) 3991.
[11] M. Bochmann, S.J. Lancaster, Angew. Chem. Int. Ed. Engl. 92
(1980) 396.
[25] G. Gilli, D.W.J. Cruickshank, R.L. Beddoes, O.S. Mills, Acta
Crystallogr. Sect B. 28 (1972) 1889.
[12] (a) P.P.K. Claire, G.R. Willey, M.G.B. Drew, J. Chem. Soc.
Chem. Commun. 1100 (1987). (b) G.R. Willey, M.L. McPartlin,
M. Butcher, I.J. Scowen, J. Chem. Soc. Dalton Trans. (1994) 305.
[13] W.A. Schenk, Angew. Chem. Int. Ed. Engl. 26 (1987) 98.
[14] P.C. Wailes, H. Weigold, A.P. Bell, J. Organomet. Chem. 33
(1971) 181.
[26] N. Bresciani-Pahor, M. Calligaris, G. Nardin, L. Randaccio,
Acta. Crystallogr. Sect. B. 34 (1978) 1360.
[27] D.A. Atwood, J.A. Jegier, K.J. Martin, D. Rutherford,
Organometallics 14 (1995) 1453.
[28] S.J. Dzugan, V.L. Goedken, Inorg. Chem. 25 (1986)
2858.
[15] R.J.H. Clark, J.A. Stockwell, J.D. Wilkins, J. Chem. Soc. Dalton
Trans. (1976) 120.
[29] E. Solari, F. Corazza, C. Floriani, A. Chiesi-Villa, C. Guastini,
J. Chem. Soc. Dalton Trans. (1990) 1345.
[16] (a) R.J.H. Clark, M.A Coles, J. Chem. Soc. Chem. Commun.
(1971) 1587. (b) A.T.T. Hsieh, Inorg. Nucl. Chem. Lett. 6 (1970)
767.
[17] N.S. Biradar, V.H. Kulkarni, J. Inorg. Nucl. Chem. 33 (1971)
3847.
[18] A. Wojcicki, Adv. Organometall. Chem. 12 (1974) 32.
[19] K. Nakajima, C. Sasaki, M. Kojima, T. Aoyama, S. Ohba, Y.
Saito, J. Fujita, Chem. Lett. (1987) 2189.
[20] M. Mazzanti, J. Rosset, C. Floriani, A. Chiesi-Villa, C. Guastini,
J. Chem. Soc. Dalton Trans. (1989) 953.
[30] K. Dey, A.K. Biswas, A.K.S. Roy, Indian J. Chem. 20 (1981)
848.
[31] J.A. Darr, S.R. Drake, M.B. Hursthouse, K.M.A. Malik, Inorg.
Chem. 32 (1993) 5704.
[32] G.M. Sheldrick, Acta. Crystallogr. Sect. A. 46 (1990) 467.
[33] G.M. Sheldrick, SHELX-93. University of Go¨ttingen, Germany,
1993.
[34] N.P.C. Walker, D. Stuart, Acta Crystallogr. A39 (1983) 158;
adapted for FAST geometry by A. Karaulov, University of Wales,
1991.
.