Steric influences in cyclopentadienyl-monophenoxide complexes of titanium(iv) arising from ortho-substitution of the phenoxide ligand

Article abstract of DOI:10.1016/j.poly.2005.11.016

Reaction of [CpTiCl₃] with 1 equiv. of Me₃SiOC₆H₄CMe₃-4 in benzene or thermalisation with an excess of neat Me₃SiOC₆H₄CMe₃-4 gives good yields of crystalline [CpTiCl₂(OC₆H₄CMe₃-4)] (1) for which an X-ray crystal structure determination shows a distorted tetrahedral geometry. Reaction of [CpTiCl₃] and Me₃SiOC₆H₃CMe₃-2-Me-4 via the benzene solvent method or the direct thermalisation gives rise to crystalline [CpTiCl₂(OC₆H₃CMe₃-2-Me-4)] (2) for which an X-ray crystal structure determination shows a distorted tetrahedral geometry little changed from complex 1 but with the phenoxide ligand ring rotated so that two of the tert-butyl group methyls straddle one chloro ligand. Refluxing [CpTiCl₃] and Me₃SiOC₆H₄Ph-4 in toluene gave crystalline [CpTiCl₂(OC₆H₄Ph-2)] (3) which X-ray crystallography shows has a distorted tetrahedral geometry little changed from 1 or 2. The phenoxide ligand ring is rotated less in 1 than in 2 or 3 with the 2-phenyl substituent sited between the chloro and Cp ligands. The complexes differ only in the interplanar angle between the Cp and phenoxide ligand phenyl rings [6.4(1)° for 1, 29.4(1)° for 2 and 22.2(1)° for 3]. DFT calculations carried out on 2 are in good agreement with the X-ray crystal structure. A relaxed PES scan shows that the tert-butyl group in the complex cannot freely rotate past the adjacent chloro ligand and is essentially locked in close proximity to it. There is a shallow minimum (2 kJ/mol within a range of 60 degrees) for the Ti-O-C-C dihedral angle indicating there is sufficient torsional motion within the molecule to accommodate the tert-butyl group in the observed position.

Full text of DOI:10.1016/j.poly.2005.11.016

Polyhedron 25 (2006) 1729–1736
www.elsevier.com/locate/poly
Steric influences in cyclopentadienyl-monophenoxide complexes
of titanium(iv) arising from ortho-substitution of the phenoxide ligand
*
Alastair J. Nielson , John A. Harrison, Chaohong Shen, Joyce M. Waters
Chemistry, Institute of Fundamental Sciences, Massey University at Auckland, Private Bag 102904, North Shore Mail Centre, Auckland, New Zealand
Received 30 June 2005; accepted 11 November 2005
Available online 22 December 2005
Abstract
Reaction of [CpTiCl3] with 1 equiv. of Me3SiOC6H4CMe3-4 in benzene or thermalisation with an excess of neat Me3SiOC6H4CMe3-4
gives good yields of crystalline [CpTiCl2(OC6H4CMe3-4)] (1) for which an X-ray crystal structure determination shows a distorted tet-
rahedral geometry. Reaction of [CpTiCl3] and Me3SiOC6H3CMe3-2-Me-4 via the benzene solvent method or the direct thermalisation
gives rise to crystalline [CpTiCl2(OC6H3CMe3-2-Me-4)] (2) for which an X-ray crystal structure determination shows a distorted tetra-
hedral geometry little changed from complex 1 but with the phenoxide ligand ring rotated so that two of the tert-butyl group methyls
straddle one chloro ligand. Refluxing [CpTiCl3] and Me3SiOC6H4Ph-4 in toluene gave crystalline [CpTiCl2(OC6H4Ph-2)] (3) which X-ray
crystallography shows has a distorted tetrahedral geometry little changed from 1 or 2. The phenoxide ligand ring is rotated less in 1 than
in 2 or 3 with the 2-phenyl substituent sited between the chloro and Cp ligands. The complexes differ only in the interplanar angle
between the Cp and phenoxide ligand phenyl rings [6.4(1)ꢀ for 1, 29.4(1)ꢀ for 2 and 22.2(1)ꢀ for 3]. DFT calculations carried out on
2 are in good agreement with the X-ray crystal structure. A relaxed PES scan shows that the tert-butyl group in the complex cannot
freely rotate past the adjacent chloro ligand and is essentially locked in close proximity to it. There is a shallow minimum (2 kJ/mol
within a range of 60 degrees) for the Ti–O–C–C dihedral angle indicating there is sufficient torsional motion within the molecule to
accommodate the tert-butyl group in the observed position.
ꢁ 2005 Elsevier Ltd. All rights reserved.
Keywords: Ti(iv); Cyclopentadienyl-monophenoxides; Steric influences; Crystal structures
1. Introduction
ene). Ligand-based steric influences of phenoxide ligands
may also be important in tuning stoichiometric and cata-
lytic reactions in organic synthesis [5]. We have therefore
looked at producing [CpTiCl2(OAr)], complexes in which
the phenoxide ligand is capable of exerting a steric influence
on one of the chloro ligands. As the cost of preparing poten-
tial catalysts is another important factor influencing usage
[6], in order to prepare bulk quantities of complexes for cat-
alytic studies we have chosen 2-tert-butyl or 2-phenyl
substituted phenols which are obtainable cheaply from
commercial sources. We report here results of preparative
and structural studies for complexes containing these
ligands and indicate through DFT calculations how a
tert-butyl group in the ortho position of the phenyl ring
sterically influences the molecule.
Cyclopentadienyl-monophenoxide complexes of tita-
nium(iv), [CpTiCl2(OAr)], which have been known for
some time [1] have found use in inorganic synthesis [2] olefin
polymerizations [3] and organic synthesis [4]. In all these
applications the nature of the phenoxide ligand plays an
important role in whether the reaction is successful or
not. For example, in olefin polymerizations when a cocata-
lyst such as MAO (methyl aluminoxane) is present, the
activity is highest when the 2,6-diisopropylphenoxide ligand
is used in conjunction with Cp* (pentamethyl cyclopentadi-
*
Corresponding author.
E-mail address: a.j.nielson@massey.ac.nz (A.J. Nielson).
0277-5387/$ - see front matter ꢁ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2005.11.016

1730
A.J. Nielson et al. / Polyhedron 25 (2006) 1729–1736
2. Results and discussion
mass when the first formed oil is pumped on or washed
with petroleum spirit and then allowed to stand.
2.1. Syntheses and X-ray crystallography
We have also looked at producing complex 1 directly
from TiCl4 using a one-pot reaction. CpSiMe3 and TiCl4
were reacted in CH2Cl2 to produce [CpTiCl3] and the reac-
tion solution treated with Me3SiOC6H4CMe3-4. However,
NMR spectroscopy indicated that the reaction did not go
to completion with more than 50% of the reaction mixture
being [CpTiCl3].
We have concentrated our studies on cyclopentadienyl
complexes since the Cp ligand is easily prepared in compar-
ison with pentamethylCp (Cp*). The normal starting mate-
rial for titanium Cp complexes is [CpTiCl3], which can be
prepared by the reaction of CpSiMe3 with TiCl4 in dichlo-
romethane [7]. We have found that to obtain good yields of
pure product using this method, high purity TiCl4 must be
used otherwise the reaction material is contaminated with a
dark product which is difficult to remove. Rather than dis-
till the TiCl4 we have opted to purify the [CpTiCl3] which
can be easily carried out by sublimation (see Section 4).
[CpTiCl2(OAr)] complexes are reported to be prepared
in good yield by the addition of 1 equiv. of LiOAr (Ar =
An X-ray crystal structure determination of 1 was car-
ried out in order to establish the structural features present
in the complex when no ortho-substituent is present on the
phenoxide ligand aromatic ring. The structure shows a dis-
torted tetrahedral geometry (Fig. 1). The Ti–O bond length
˚
[1.772(2) A] (Table 1) is similar to that found in [CpTi-
˚
Cl2(OC6H3(CHMe2)2-2,6)] [1.760(4) A] [3l] and the more
bulky pentamethylcyclopentadienyl complexes [Cp*TiCl2-
i
OC6H2Me3-2,4,6; OC6H3 Pr2-2,6; OC6H2CMe3-2-Me2-4,6)
to [CpTiCl3] in diethyl ether (Eq. (1)) [3k,l]
[CpTiCl3] + LiOAr ꢀ! [CpTiCl2(OAr)] + LiCl
ð1Þ
However in developing a high yield route to [CpTi-
Cl2(OC6H4CMe3-4)] (1) which was to be used for compar-
ative studies, it was found that the reaction of
LiOC6H4CMe3-4 and [CpTiCl3] did give the complex but
two minor products were always present even when the
reaction was carried out under carefully controlled condi-
tions. Further, the complex proved very difficult to crystal-
lize when prepared in this manner. Thermalisation using
the phenol (Eq. (2)) also gave a similar result.
[CpTiCl3] + HOAr ꢀ! [CpTiCl2(OAr)] + HCl
ð2Þ
Fig. 1. ORTEP diagram, at the 50% probability level, of the molecular
structure of [CpTiCl2(OC6H4CMe3-4)] (1). Hydrogen atoms have been
omitted for clarity. Selected bond distances and angles are given in Table 1.
When the phenol was reacted with [CpTiCl2Me)] generated
in situ from [CpTiCl3] and 1 equiv. of MeLi in diethyl ether
there was one major contaminating product but the desired
complex still proved difficult to crystallise. Reaction of
LiCp with [TiCl3(OC6H4CMe3-4)] [8] produced the com-
plex but the reaction also formed another product which
reduced the yield. We have found that use of the O-tri-
methyl silyl (TMS) phenol in a thermalisation reaction
(Eq. (3)) gives a better result, producing initially a red oil
which solidifies after pumping to give the required crystal-
line solid in high yield.
Table 1
˚
Selected bond distances (A) and angles (ꢀ) for 1, 2 and 3
Complex
1
2
3
˚
Bond distances (A)
Ti–O
Ti–Cl(1)
Ti–Cl(2)
O–C(1)
1.7718(18)
2.2542(9)
2.2594(8)
1.368(3)
1.7805(15)
2.2581(8)
2.2517(8)
1.372(2)
1.7824(12)
2.2572(6)
2.2631(6)
1.366(2)
[CpTiCl3] + Me3SiOC6H4CMe3-4
Bond angles (ꢀ)
Ti–O–C(1)
O–Ti–Cl(1)
O–Ti–Cl(2)
Cl(1)–Ti–Cl(2)
O–C(1)–C(2)
ꢀ! [CpTiCl2(OC6H4CMe3-4)] (1) + Me3SiCl
ð3Þ
153.77(16)
103.84(7)
104.39(7)
102.89(4)
119.8(2)
152.75(14)
101.82(6)
103.47(5)
104.02(3)
117.35(19)
120.79(19)
122.82(19)
122.0(2)
110.13(19)
110.90(18)
111.28(19)
110.0(2)
155.84(11)
102.91(5)
102.56(4)
102.73(2)
117.82(16)
120.56(15)
121.74(15)
120.90(16)
120.87(16)
The reaction is best carried out by stirring a slight excess of
the TMS phenol with [CpTiCl3] in benzene solution for ca.
18 h after which the solvent is removed and the residue
pumped on or washed with cold petroleum spirit to remove
the unreacted TMS phenol. The complex then crystallises
to a solid mass. NMR spectroscopy indicated that only
10% of the [CpTiCl3] remained at this stage and that a
short period of reflux (ca. 2 h) removed essentially all the
[CpTiCl3]. Only very small amounts of another product
were formed if the refluxing period was extended to 6–
8 h. Even in this case the product crystallises to a solid
O–C(1)–C(6)
119.9(2)
C(1)–C(6)–C(7)
C(5)–C(6)–C(7)
C(6)–C(7)–C(8)
C(6)–C(7)–C(9)
C(6)–C(7)–C(10)
C(8)–C(7)–C(9)
C(8)–C(7)–C(10)
C(9)–C(7)–C(10)
109.2(4)
110.5(4)
104.7(4)
106.7(2)
107.71(19)

A.J. Nielson et al. / Polyhedron 25 (2006) 1729–1736
1731
(OC6H3Me2-2,6)] and [Cp*TiCl2(OC6H3(CHMe2)2-2,6)]
to that shown by complex 1 and suggests that the phenox-
ides are electronically equivalent in both complexes.
An X-ray crystal structure determination of 2 shows a
distorted tetrahedral geometry similar to complex 1 with
the phenoxide phenyl ring rotated so that two of the
2-tert-butyl substituent methyl groups straddle Cl(2)
˚
[Ti–O bond lengths 1.786(2) and 1.773(3) A, respectively]
[3l] and fð1; 3-Bu2t CpÞTiCl2½OC6H3ðCHMe2Þ -2; 6ꢁg [Ti–O
2
˚
bond length 1.773(2) A] [9]. The Ti–Cl bond lengths in 1
[2.254(1) and 2.259(1) A] are not significantly different
˚
from one another and are also the same length as in [CpTi-
˚
˚
Cl2(OC6H3(CHMe2)2-2,6)] [Ti–Cl bond lengths 2.261(1) A].
(Fig. 2). The Ti–O bond length [1.781(2) A], C–O bond
˚
˚
Also the C–O bond length in 1 [1.368(3) A] is no different to
length [1.372(2) A] and Ti–Cl bond lengths [2.252(1) and
˚
˚
the other phenoxide complexes [range 1.366(3)–1.368(6) A].
2.258(1) A] are no different from those found in complex
It can thus be seen that based on the bond length data,
there is little electronic or steric effect in this type of com-
plex whether ortho-substituents such as iso-propyl groups
are present or not. In the mono phenoxide complex {TiCl3-
[OC6H3(CMe3)2-2,6-Me-4]} where the phenoxide ligand is
the dominant p-donor and has no competition from a
p-donor Cp ligand, the Ti–O bond length has the shorter
1 which indicates that the 2-tert-butyl substituent does
not cause significant steric pressure and force an electronic
change to the Ti–O or Ti–Cl bonds in the solid state. Sim-
ilarly the O–Ti–Cl(1), O–Ti–Cl(2) and Cl(1)–Ti–Cl(2) bond
angles in the molecule are also little different from those for
complex 1 (see Table 1) and there is little difference in the
Ti–O–C bond angles [152.8(1)ꢀ for 2, 153.8(2)ꢀ for 1]. How-
ever the interplanar angle between the Cp and phenyl rings
in 2 [29.4(1)ꢀ] is now larger than in 1 [6.4(1)ꢀ] and this
would appear to be a consequence of the tert-butyl methyls
moving away from the Cp ligand to avoid a clash and to
also be accommodated about the Cl ligand.
˚
value of 1.750(2) A [8] showing stronger phenoxide p-
donation than would be expected for the Cp-phenoxide
complexes. The Ti–O–C bond angle in 1 [153.8(2)ꢀ] (Table 1)
is significantly smaller than those found in the 2,6-dimethyl
or diisopropyl complexes quoted above [range 162.3(2)–
173.6(3)ꢀ] so that the 2,6-substituents do have an effect
on this angle but where they are not present, this angle
can go down to ca. 154ꢀ without having any electronic
effect on the Ti–O bond as indicated by the similar Ti–O
bond lengths.
In 1 the phenoxide ligand phenyl ring is rotated so that
the interplanar angle between the Cp and phenyl rings is
only 6.4(1)ꢀ which apparently allows maximum p-overlap
in the Ti–O–C system [5]. Some comparison of the strength
of the p-bonding component in the phenoxide ligand can
be obtained from the 13C NMR spectrum which shows
the ipso-carbon resonance at d 166.2. This is further upfield
from that observed in the monophenoxide complex
[TiCl3(OC6H4CMe3-4)] (d 169.6) and for a variety of this
type of complex (d range 169.6–174.9) where there is a need
for tight O(2p)-to-Ti(d) orbital donation in the presence of
three chloro ligands which are less efficient p-donors than a
phenoxide ligand [8]. However in the bis-phenoxide com-
plex [TiCl2(OC6H4CMe3-4)2] where this need is relaxed,
the ipso-carbon resonance is at a similar position (d 166.3
[5]) to that observed for 1 indicating that the p-donor
requirement is similar whether the other ligand is Cp or
OAr. This would then give some experimental evidence
for the ‘‘isolobal’’ analogy [10] used to compare these dif-
ferent type of ligands systems [11].
The Cl(2)ꢂ ꢂ ꢂC(8) and Cl(2)ꢂ ꢂ ꢂC(9) contacts in 2 are 3.98
˚
and 4.12 A, respectively, so that with this frozen structure,
the tert-butyl group would have difficulty passing over
Cl(2). However the 1H NMR spectrum shows a singlet
for the tert-butyl methyl groups indicating rotation or at
least some movement so that there is clearly some flexibility
in the molecule in solution. For example, in the configura-
tion shown by the crystal structure, bending back of the
Ti–O–C bond further than observed [152.8(1)ꢀ] would
clearly allow the tert-butyl methyls to rotate but angles
smaller than this may be difficult to attain [8]. There is
nothing unusual about the angles associated with the
tert-butyl group either. The O–C(1)–C(6) and C(5)–C(6)–
C(7) angles [120.8(2)ꢀ and 122.0(2)ꢀ] are little different from
the ideal sp2 angle and all the angles associated with C(7)
are close to the ideal sp3 angle. In particular the C(8)–
C(7)–C(9) angle which might be expected to widen if
The reaction of [CpTiCl3] and a slight excess of
Me3SiOC6H3CMe3-2-Me-4 in benzene does not go to com-
pletion when the reaction is stirred at room temperature,
but does so when the mixture is refluxed for a short period
(ca. 2 h) giving rise to [CpTiCl2(OC6H3CMe3-2-Me-4)] (2).
The complex forms as a solid mass on standing after the
excess of TMS phenol is pumped off or removed by wash-
ing. It can then be crystallised from petroleum spirit. The
NMR spectra show a single set of resonances for the Cp
and phenoxide ligands and in the 13C NMR spectrum the
ipso-carbon resonance appears at d 166.4 which is similar
Fig. 2. ORTEP diagram, at the 50% probability level, of the molecular
structure of [CpTiCl2(OC6H3CMe3-2-Me-4)] (2). Hydrogen atoms have
been omitted for clarity. Selected bond distances and angles are given in
Table 1.

1732
A.J. Nielson et al. / Polyhedron 25 (2006) 1729–1736
seriously effected by proximity to Cl(2) is 110.0(2)ꢀ. As to
whether the phenoxide ring can rotate so that the tert-butyl
group passes over the two chloro ligands and also allow
rotation as it does so (geared rotation [12]), is further com-
plicated as there would be an interaction of the phenoxide
ring ortho-proton with the Cp ligand hydrogens to restrict
Ti–O–C bond bending. This effect would be even more pro-
nounced if there were a methyl group in the phenoxide ring
ortho-position or if pentamethyl Cp were present.
NMR spectral analysis of the reaction between
[CpTiCl3] and Me3SiOC6H4Ph-4 showed that after 4 h of
refluxing only 8% of the [CpTiCl3] remained and [CpTi-
Cl2(OC6H4Ph-2)] (3) could be obtained as a crystalline
solid in high yield.
The complex had a single set of resonances for the var-
ious hydrogens and carbons in the NMR spectra with the
ipso-carbon resonance of the phenoxide ring at d 165.4
which is the same as tert-butyl complex 2. An X-ray crystal
structure determination showed a distorted tetrahedral
array (Fig. 3.) with bond length and bond angle features lit-
tle different from those found in complexes 1 and 2 (see
Table 1).
The phenoxide ligand phenyl ring is rotated so that the
phenyl substituent sits over and between Cl(1) and the Cp
ring so that the interplanar angle between the Cp and phen-
oxide ligand phenyl ring is 22.2(1)ꢀ. This compares with
29.4(1)ꢀ for 2-tert-butylcomplex 2 indicating there is less
need for rotation for the phenyl substituent which does
not need to position itself about Cl(1) as does the tert-butyl
group.The phenyl substituent thus covers Cl(1) to a much
lesser extent than the tert-butyl group in complex 2 where
the two of the methyl groups straddle Cl(1) to minimize
interactions.
Given the high yield route to complexes 1, 2 and 3 using
the TMS phenols, we have looked at carrying out the reac-
tion in the absence of solvent to see if Ôgreen chemistryÕ prin-
ciples could be applied to the reactions. When [CpTiCl3]
and Me3SiOC6H4CMe3-4 were heated together at 100 ꢀC
for 6 h. NMR spectroscopy indicated that only about 7%
of the [CpTiCl3] remained unconverted and complex 1
could also be obtained as a crystalline solid. Similarly, a
reaction between [CpTiCl3] and neat Me3SiO-C6H3CMe3-
2-Me-4 at 90 ꢀC for 6 h gave a high conversion of the
[CpTiCl3] and a good yield of complex 2. However with
Me3SiOC6H4Ph-2 at 160–170 ꢀC the [CpTiCl3] conversion
was only about 40% whereas the [CpTiCl3] conversion in
benzene solution was 80% after work-up.
2.2. Theoretical calculations
Complexes 2 and 3 are asymmetric (Cs symmetry) in the
solid state with one of the chloro ligands biased by the sub-
stituent at the 2-position of the phenoxide ligand ring. DFT
calculations [13] were carried out to assess the ability of the
phenoxide ligands to rotate over the Cp or chloro ligands
and maintain the asymmetry. In particular the structure
of [CpTiCl2(OC6H3CMe3-2-Me-4)] (2) was investigated as
the X-ray structure indicated the greater degree of influence
of the ortho-substituent. The calculations gave good agree-
ment with the crystal structure with the greatest deviation
occurring for the Cp–Ti parameters as the Cp ligand was
rotated slightly (4ꢀ) in comparison with the experimental
structure. Bond distances between the heavy elements
(i.e., excluding H) for the other degrees of freedom (i.e.,
non Cp–Ti bond distances) were all within 1% of the crys-
tal structure parameters and the angles agreed within 2% or
better.
An NBO analysis [14] was performed. The NBO analy-
sis estimates the interaction [E(2)] between each donor and
acceptor NBO by second order perturbation theory. This
allows a qualitative understanding of the energetic impor-
tance of the delocalisation of electrons from the various
filled (i.e., high occupancy) orbitals into the empty (i.e.,
low occupancy) orbitals. Using this analysis one can assess
the relative importance of the various ligand (donor) to
metal (acceptor) interactions. Analysis of the NBOÕs
according to this procedure for the ligand to metal interac-
tions indicates that there is strong interaction between the
lone pairs on the Cl atoms and the metal d orbitals (rang-
ing up to 268 kJ/mol). The Cp to Ti interactions are dom-
inated [E(2) = 1224 kJ/mol] by the interaction between a
lone pair (3% 2s 97% 2p character) orbital on C25 with
one of the partially occupied 3d orbitals on the Ti. A num-
ber of interactions [E(2) ranging up to 108 kJ/mol] between
the C21–C22 and C23–C24 p-bonds (donors) and the Ti 3d
orbitals (acceptors) are also clearly evident.
There are a total of nine interactions with an E(2) in the
range between 41 and 214 kJ/mol between the various
oxygen valence orbitals and the partially occupied Ti 3d
orbitals. The results clearly imply that a simple interpreta-
tion in terms of a single lone pair donation to the Ti centre
is an inadequate one. Similarly, a first principles model
might predict a strong donor–acceptor interaction bet-
ween the oxygen valence orbitals and the C1–C2, C1–C6
p-antibonding orbitals, and this indeed seems to be the case
Fig. 3. ORTEP diagram, at the 50% probability level, of the molecular
structure of [CpTiCl2(OC6H4Ph-2)] (3). Hydrogen atoms have been
omitted for clarity. Selected bond distances and angles are given in Table 1.

A.J. Nielson et al. / Polyhedron 25 (2006) 1729–1736
1733
with the O lone pair/C1–C2 p* E(2) calculated to be 90 kJ/
mol and the O lone pair/C1–C6 E(2) weaker at 30 kJ/mol.
There are also weaker interactions between a lone pair on
each of the Cl1 and Cl2 atoms (donor) with empty O (pre-
dominantly 3s/3p: 12%s, 88%p character) orbitals with
E(2) = 21 kJ/mol for each of these interactions.
as a consequence of a twisting motion of the (Cl)2–Ti–Cp
group which is coupled to the change in Ti–O–C1–C6 dihe-
dral. Examination of the normal mode harmonic vibra-
tional frequencies calculated for 2 yields six frequencies
lower than 100 cmꢀ1, at least three of which involve a sig-
nificant component of Ti–O–C1-C6 torsion. Despite this,
these motions are bounded at one extreme by the difficulty
of rotating the Cl atom past the tert-butyl group methyls
and at the other by the decrease in distance between the
phenoxide ring ortho-proton with the Cp protons (the dis-
The Ti to ligand interactions are dominated by the inter-
action between the partially filled valence 4s (88%) + 3d
(12%) Ti hybrid with the empty O 3s(12%) + 3p(88%)
hybrid with an E(2) of 95 kJ/mol. Other significant interac-
tions occur between this Ti orbital and the antibonding O–C
orbital (E(2) = 39 kJ/mol) and one of the vacant O hybrid
(involving 4s, 4p, 3d O basis functions) orbitals (E(2) =
37 kJ/mol). The only other significant Ti to ligand interac-
tions are between the partially occupied valence 3d orbitals
(on Ti) and the p*-antibonding orbitals between C21/C22
(E(2) = 75 kJ/mol) and C23/C24 (E(2) = 60 kJ/mol).
Examination of the charges calculated from the Natural
Population Analysis (NPA) for the two protons closest to
the Cl atom (+0.23 and +0.25) indicates that there is no
significant difference to the other protons on these two
methyl groups (+0.23). The NPA analysis yields for the
other atoms of interest: +1.26 for Ti, ꢀ0.32 for Cl1,
ꢀ0.31 for Cl2, and ꢀ0.71 for the O atom.
˚
tance decreases from an average of 3.00 A at the minimum
˚
energy structure to 2.77 A at +15ꢀ) as suggested above. The
net result of this discussion is that the motion of Cl2
appears to be sterically encumbered by the close proximity
of the tert-butyl group rather than by any electronic
constraints
3. Conclusion
The complexes [CpTiCl(OAr)] [Ar = C6H4CMe3-4 (1),
C6H4CMe3-2-Me-4 (2) and C6H4Ph-2 (3)] are conveniently
prepared by the reaction of the relative TMS phenol with
[CpTiCl3]. X-ray crystallography shows that there is not
much difference in the structural parameters for the three
complexes indicating that 2-tert-butyl or 2-phenyl substitu-
ents on the phenyl ring of the phenoxide ligand are easily
accommodated. The main effect of the substituent is to
change the interplanar angle between the Cp and phenoxide
ligand phenyl rings. For complex 1 in which there is no
ortho-substituent, the angle is only 6.4(1)ꢀ which most likely
allows a maximum of O(2p)-to-Ti(d) and O(2p)-to-C(p)
orbital interaction if intermolecular effects in the crystal
structure are not important. For complex 2 (ortho-tert-butyl
group), the angle increases to 29.4(1)ꢀ and for complex 3
(ortho-phenyl group) the effect is somewhat smaller at
22.2(1)ꢀ. DFT calculations indicate that for 2 the tert-butyl
group cannot freely rotate past the adjacent chloro ligand
and is essentially locked in the position shown by the solid
state structure. However, there is a shallow minimum asso-
ciated with the Ti–O–C–C dihedral angle (2 kJ/mol within a
range of 60ꢀ) so that there is sufficient flexibility in the mol-
ecule to allow torsional motion which accommodates the
tert-butyl group in the observed position.
A potential energy surface (PES) scan was performed by
scanning the Ti–O–C1–C6 dihedral angle and allowing all
of the other internal degrees of freedom to relax to their
minimum energy values. The results of this relaxed PES
scan are shown in Fig. 4 for a variation of the dihedral
angle from +15ꢀ to ꢀ95ꢀ (minimum at ꢀ27ꢀ). Outside this
range the calculations failed to converge. The results indi-
cate that there is a shallow minimum where the energy
change is a mere 2 kJ/mol within a range of 60ꢀ for the
dihedral. Further inspection indicates that despite the large
change in angle the tert-butyl group does not rotate past
the Cl2 atom, but moreover, the change in dihedral is
accompanied by a distortion in other dihedral coordinates.
The most noticeable change being that as the Ti–O–C1–C6
angle is displaced away from its equilibrium value the Cl
atom straddled by the two tert-butyl CH3 groups is pushed
further toward the tert-butyl group central C. This occurs
Relaxed PES Scan
14
4. Experimental
12
10
8
4.1. General methods and materials
All preparations and manipulations were carried out
under dry oxygen-free nitrogen using standard bench-top
techniques for air-sensitive substances. TiCl4 and the phe-
nols were used as received from commercial sources.
CpSiMe3 was prepared by the reaction of CpNa with
Me3SiCl in [15]. Light petroleum (b.p. 40–60 ꢀC) and tolu-
ene were distilled from sodium wire and dichloromethane
from freshly ground CaH2. 1H and 13C–{1H} NMR spectra
were recorded at 400 and 100 MHz, respectively, on a
6
4
2
0
-100
-80
-60
-40
-20
0
20
Ti-O-C1-C6 Dihedral angle
Fig. 4. Relaxed PES scan for the Ti–O–C1–C6 dihedral.

1734
A.J. Nielson et al. / Polyhedron 25 (2006) 1729–1736
Bruker AM400 spectrometer. CDCl3 was dried over, and
distilled from freshly ground CaH2 and benzene-d6 from
sodium wire. C, H and N analyses were determined by
Dr. A. Cunninghame and associates, University of Otago,
New Zealand.
(80 cm3), and chlorotrimethylsilane (5.5 cm3, 39.4 mmol)
added dropwise as for the reaction with to 4-tert-butylphe-
1
nol described above. Yield 7.85 g, (91%). H NMR d: 0.32
(s, 9H, SiMe3); 1.35 (s, 9H, CMe3); 2.26 (s, 3H, Me); 6.71
3
3
4
[d, J(HH) 8.0, 1H, o-H]; 6.93 [dd, J(HH) 8.0, J(HH)
1.8,1H, m-H]; 7.13 [d, J(HH) 1.8, 1H, m-H]. 13C NMR
4
4.2. Synthesis
d; 0.7 (SiMe3); 20.8 (Me); 29.6 (CMe3); 34.4 (C); 118.3
(o-C); 126.9 and 127.7 (m-Cs); 151.9 (ipso-C).
4.2.1. Preparation of [CpTiCl3]
This complex was prepared by the reaction of CpSiMe3
and TiCl4 in CH2Cl2 [7]. The resulting grey-green solid was
purified prior to use by subliming some of the material
from a 250-cm3 round-bottom flask into an unjacketed
air condenser and scraping the required amount of subli-
mate into another flask.
4.2.5. [CpTiCl2(OC6H3CMe3-2-Me-4)] (2)
Method A: Me3SiOC6H3CMe3-2-Me-4 (0.93 g, 3.93
mmol) in benzene (25 cm3) was added to [CpTiCl3]
(0.86 g, 3.9 mmol) in benzene (25 cm3) and the mixture stir-
red for 1 h and then refluxed for 3 h. The solution was fil-
tered and the solvent removed to give an oil which formed
a crystalline mass on standing. Crude yield 1.28 g (94%)
The solid was dissolved in hot petroleum spirit (120 cm3),
the solution filtered and the volume reduced giving the
complex as long orange-yellow crystals on standing. Yield
0.85 g, (62.4%). The complex was characterized by X-ray
4.2.2. Me3SiOC6H4CMe3-4
Triethylamine (6.5 cm3, 46.6 mmol) was added dropwise
to 4-tert-butylphenol (5.0 g, 33.3 mmol) in diethyl ether
(80 cm3), the mixture was cooled to 0 ꢀC and chlorotri-
methylsilane (5 cm3, 39.4 mmol) added dropwise via a syr-
inge. The mixture was stirred for 14 h, refluxed for 3 h, the
solution filtered and the remaining triethylamine hydro-
chloride washed with diethyl ether (50 cm3). The solvent
was removed from the combined filtrate and extract to give
1
crystallography using a single crystal of this product. H
NMR d: 1.39 (s, 9H, CMe3); 2.33 (s, 3H, Me); 6.78 (s,
3
3
5H, Cp); 6.83 [d, J(HH) 7.8, 1H, o-H]; 6.97 [bd, J(HH)
3
7.8, 1H, m-H]; 7.08 [m, J(HH) 7.8, 1H, m-H]. 13C NMR
d; 21.2 (Me); 29.9 (CMe3); 34.7 (C); 123.5 (o-CH); 127.0
(m-CH); 127.8 (m-CH); 134.1 (p-C); 137.1 (o-C); 165.4
(ipso-C).
1
the TMS phenol as a colourless oil. Yield 7.2 g, (97%). H
NMR d: 0.41 (s, 9H, SiMe3); 1.44 (s, 9H, CMe3); 6.92 [d,
3
3J(HH) 8.8, 2H, o-Hs]; 7.38 [d, J(HH) 8.8, 2H, m-Hs].
Method B: Me3SiOC6H3CMe3-2-Me-4 (1.1 g, 4.6 mmol)
was added to [CpTiCl3] (1.0 g, 4.6 mmol) the mixture stir-
red at 90 ꢀC for 6 h and then allowed to stand at room tem-
perature overnight. On standing further, a solid crystalline
mass was formed which was washed with chilled (0 ꢀC)
petroleum spirit to give the complex as an orange solid
which had identical NMR spectra with the sample pre-
pared under method A.
13C NMR d: 0.27 (SiMe3), 31.6 (CMe3), 34.0 (C), 119.3
(o-C), 126.1 (m-C), 143.8 (p-C), 152.8 (ipso-C).
4.2.3. [CpTiCl2(OC6H4CMe3-4)] (1)
Method A: Me3SiOC6H4CMe3-4 (0.90 g, 4.0 mmol) in
benzene (30 cm3) was added dropwise over 20 m to
[CpTiCl3] (0.86 g, 3.9 mmol) in benzene (25 cm3) and the
mixture was stirred for 18 h and refluxed for 2 h. The solu-
tion was filtered and the solvent removed to give a gum
(1.4 g) which formed a crystalline mass on standing. The
solid was broken-up and washed with chilled (0 ꢀC) petro-
leum spirit (30 cm3) and dried under vacuum. Yield 1.22 g,
(94%). The complex was characterized by X-ray crystallog-
raphy using a single crystal from the mass. 1H NMR d: 1.24
4.2.6. Me3SiOC6H4Ph-2
Triethylamine (6.0 cm3, 43.0 mmol) was added dropwise
to 2-phenylphenol (7.2 g, 42.4 mmol) in diethyl ether
(80 cm3), and chlorotrimethylsilane (5.5 cm3, 39.4 mmol)
added dropwise as for the reaction with 4-tert-butylphenol
described above. The oil obtained was dissolved in petro-
leum spirit (50 cm3), the solution filtered to remove a small
amount of white solid and the solvent removed giving the
3
(s, 9H, CMe3); 6.70 (s, 5H, Cp); 6.86 [d, J(HH) 8.4, 2H,
3
o-Hs]; 7.23 [d, J(HH) 8.4, 2H, m-Hs]. 13C NMR d; 31.4
1
(CMe3), 34.5 (C), 117.8 (o-C), 120.7 (Cp), 126.1 (m-C),
147.6 (p-C), 166.2 (ipso-C).
TMS phenol as a colourless liquid. Yield 9.6 g, (93%). H
NMR d: 0.10 (s, 9H, SiMe3); 6.98 [dd, 3J(HH) 7.2,
3
4
Method B: Me3SiOC6H4CMe3-4 (0.96 g, 4.4 mmol) was
added to [CpTiCl3] (0.97 g, 4.4 mmol) and the mixture was
stirred at 100 ꢀC for 6 h. On cooling and standing a solid
crystalline mass was formed which was washed with chilled
(0 ꢀC) petroleum spirit to give the orange-red complex
which had identical NMR spectra to the sample prepared
under by Method A.
4J(HH) 1.1, 1H, o-H]; 7.11 [td, J(HH) 7.2, J(HH) 1.1,
1H, m-H]; 7.20 [td, 3J(HH) 7.2, 4J(HH) 1.1, 1H,
m-H]; 7.25-7.61 [m, 6H, aromatic-Hs]. 13C NMR d: 120.6
(CH); 121.9 (CH); 126.7 (CH); 127.8 (m-CHs); 128.4
(CH); 129.6 (o-CHs); 130.7 (CH); 133.3 (C); 139.0 (o-C);
152.4 (ipso-C).
4.2.7. [CpTiCl2(OC6H4Ph-2)] (3)
4.2.4. Me3SiOC6H4CMe3-2-Me-4
Triethylamine (6.0 cm3, 43.0 mmol) was added dropwise
to 2-phenylphenol (7.2 g, 42.4 mmol) in diethyl ether
Me3SiOC6H4Ph-2 (1 g, 4.13 mmol) in toluene (25 cm3)
was added to [CpTiCl3] (0.88 g, 4.0 mmol) in toluene
(25 cm3) and the mixture stirred for 1 h and then refluxed

A.J. Nielson et al. / Polyhedron 25 (2006) 1729–1736
1735
Table 2
Crystallographic data for complexes 1, 2 and 3
Empirical formula
Colour, habit
C15H18Cl2OTi
orange, plate
0.44 · 0.18 · 0.11
monoclinic
P21/n
6.5596(1)
10.9940(3)
22.4839(6)
90
93.259(1)
90
1618.83(7)
4
C16H20Cl2OTi
C17H14Cl2OTi
orange, plate
0.38 · 0.22 · 0.07
monoclinic
P21/n
10.1346(2)
10.3240(3)
16.7649(3)
90
104.777(1)
90
1697.60(7)
4
yellow, plate
0.42 · 0.16 · 0.08
monoclinic
P21/c
15.3821(2)
7.2149(1)
15.563(1)
90
111.783(1)
90
1603.97(3)
4
Crystal size (mm)
Crystal system
Space group
˚
a (A)
˚
b (A)
˚
c (A)
a (ꢀ)
b (ꢀ)
c (ꢀ)
3
˚
V (A )
Z
˚
k (A)
0.71073
203(2)
0.71073
200(2)
0.71073
150(2)
T (K)
h Range for data collection
No. of data collected
No. of unique data, Rint
Completeness to hmax (%)
No. of parameters refined
R1, data with I > 2r(I)
wR2, all data
1.81–27.50
9875
3562, 0.0371
95.8
182
0.0542, 2730
0.1354, 3562
2.14–26.74
9935
3535, 0.0213
98.1
181
0.0380, 2806
0.0908, 3535
1.43–27.10
9605
3407, 0.0166
96.5
190
0.0291, 2904
0.0770, 3407
for 3 h. The solution was filtered and the solvent removed
to give a gum which solidified to a crystalline mass on
standing. The solid was broken up and washed with hot
light petroleum spirit (50 cm3) leaving a crystalline solid.
Yield 1.15 g (80%). The complex was characterized by X-
ray crystallography using a single crystal from the solid.
1H NMR d: 6.30 (s, 5H, Cp); 7.13 [dd, 3J(HH) 8.1,
set for C and H. This resulted in 404 basis functions con-
tracted from the 764 primitive Gaussians. Vibrational fre-
quency calculations yielded sets of all real frequencies for
the minimum energy structure.
5. Supplementary data
3
4
4J(HH) 1.2, 1H, o-H]; 7.17 [td, J(HH) 7.6, J(HH) 1.2,
Crystallographic data for the structural analyses have
been deposited with the Cambridge Crystallographic Data
Centre, CCDC 272194–272196 for compounds 1, 2 and 3.
Copies of this information may be obtained free of charge
from The Director, CCDC, 12 Union Road, Cambridge,
CB2 1EZ, UK, fax: +44 1223 366 033, e-mail: deposit@ccdc.
ac.uk or on the web www: http://www.ccdc.cam.ac.uk.
1H, m-H]; 7.23-7.33 (m, 2H, m, p-Hs); 7.47 (2t, 3H,
3
4
m-and p-Hs); 7.56 [dd, J(HH) 8.1, J(HH) 1.2, 2H, o-H].
13C NMR d: 120.0 (CH); 121.0 (Cp); 124.7 (CH); 127.5
(CH); 128.5 (m-CHs); 130.2 (o-CHs); 130.4 (CH); 130.7
(C); 152.4 (o-CH)); 165.4 (ipso-C).
4.3. X-ray crystallography
References
Crystallographic and refinement data are given in Table
2. Data collected, with graphite monochromated Mo Ka
X-radiation, were measured on a Siemens SMART CCD
area detector diffractometer with the omega scan method.
The data were corrected for Lorentz and polarisation
effects and for absorption by the multi scan method [16].
The structures were solved by direct methods [17] and
refined by full-matrix least-squares methods [18] on F2.
Hydrogen atoms were placed in calculated positions (phe-
[1] W. Skupinski, A. Wasilewski, J. Organomet. Chem. 220 (1981) 39.
[2] (a) W. Skupinski, K. Nicinski, P. Maksimowski, M. Wasek, J. Mol.
Catal. A: Chem. 178 (2002) 73;
(b) E. Witt, D.W. Stephan, Inorg. Chem. 40 (2001) 3824;
(c) A.V. Firth, E. Witt, D.W. Stephan, Organometallics 17 (1998) 3716;
(d) A.V. Firth, D.W. Stephan, Inorg. Chem. 37 (1998) 4726;
(e) A.V. Firth, D.W. Stephan, Organometallics 16 (1997) 2183;
(f) I.M.M. Fussing, D. Pletcher, R.J. Whitby, J. Organomet. Chem.
470 (1994) 109;
t
˚
(g) W. Skupinski, A. Wasilewski, J. Organomet. Chem. 282 (1985) 69.
[3] (a) D.J. Byun, A. Fudo, A. Tanaka, M. Fujiki, N. Kotohiro,
Macromolecules 37 (2004) 5520;
nyl ring C–H 0.94, Bu C–H 0.97 A) and were refined with
a riding model (Uiso = 0.08). Some disorder was noted for
the tert-butyl group in 1 (with a site occupancy ratio of ca.
8:2). Diagrams were prepared by ORTEP [19].
(b) Y. Qian, H. Zhang, X. Qian, J. Huang, C. Shen, J. Mol. Catal. A:
Chem. 192 (2003) 25;
(c) K. Nomura, K. Oya, Y. Imanishi, J. Mol. Catal. A: Chem. 174
(2001) 127;
(d) W. Skupinski, K. Nicinski, Appl. Organomet. Chem 15 (2001) 635;
(e) Y. Imanishi, K. Nomura, J. Polym. Sc., Part A: Polym. Chem. 38
(2000) 4613;
4.4. Theoretical calculations
All structures were fully optimized using GAUSSIAN 98
[13] at the B3LYP [20] level utilising the 6-311 + G(d) basis
set for Ti, O, and Cl in conjunction with the 6–31G(d) basis
(f) K. Nomura, T. Komatsu, I. Takashi, Y. Imanishi, Macromolecules
33 (2000) 8122;

1736
A.J. Nielson et al. / Polyhedron 25 (2006) 1729–1736
(g) K. Nomura, T. Komatsu, Y. Imanishi, J. Mol. Catal. A: Chem.
159 (2000) 127;
(h) K. Nomura, T. Komatsu, Y. Imanishi, J. Mol. Catal. A: Chem.
152 (2000) 152;
(i) A.V. Firth, J.C. Stewart, A.J. Hoskin, D.W. Stephan, J. Organo-
met. Chem. 591 (1999) 185;
(j) J. Liu, H. Ma, J. Huang, Y. Qian, Eur. Polym. J. 36 (2000) 2055;
(k) K. Nomura, N. Naga, M. Miki, K. Yanagi, Macromolecules 31
(1998) 7588;
(l) K. Nomura, N. Naga, M. Miki, K. Yanagi, A. Imai, Organomet-
allics 17 (1998) 2152;
[13] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgometry Jr., R.E.
Stratmann, J.C. Burant, S. Dapprich, J.M. Millam, A.D. Daniels,
K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi, V. Barone, M.
Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford,
J. Ochterski, G.A. Petersson, P.Y. Ayala, Q. Cui, K. Morokuma, P.
Salvador, J.J. Dannenberg, D.K. Malick, A.D. Rabuck, K. Raghav-
achari, J.B. Foresman, J. Cioslowski, J.V. Oritz, A.G. Baboul, B.B.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y.
Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson,
W. Chen, M.W. Wong, J.L. Andres, C. Gonzalez, M. Head-Gordon,
E.S. Replogle, J.A. Pople, Gaussian 98 (Revision A.11), Gaussian,
Inc., Pittsburgh, PA, 2001.
(m) R.E. Campbell Jr., T.H. Newman, M.T. Malanga, Macromol.
Symp. 97 (1995) 151.
[4] S.J. Sturla, S.L. Buchwald, Organometallics 21 (2002) 739.
[5] A.J. Nielson, C. Shen, P. Schwerdtfeger, J.M. Waters, Eur. J. Inorg.
Chem. (2005) 1343.
[14] E.D. Glendening, A.E. Reed, J.E. Carpenter, F. Weinhold, NBO
Version 3.1.
[6] D.W. Stephan, Organometallics 24 (2005) 2548.
[7] A.M. Cardoso, R.J.H. Clark, S. Moorhouse, J. Chem. Soc., Dalton
Trans. (1980) 1156.
[15] (a) C.S. Kraihanzel, M.L. Losee, J. Am. Chem. Soc. 90 (1968) 4701;
(b) K. Hafner, H. Kaiser, Org. Synth. 44 (1964) 94.
[16] R.H. Blessing, Acta Cryst. A51 (1995) 33.
[8] A.J. Nielson, P. Schwerdtfeger, J.M. Waters, J. Chem. Soc., Dalton
Trans. (2000) 529.
[9] P. Gomez-Sal, A. Martin, M. Mena, P. Royo, R.J. Serrano, J.
Organomet. Chem. 419 (1991) 77.
[10] R. Hoffmann, Angew. Chem., Int. Ed. Engl. 21 (1982) 711.
[11] V.C. Gibson, J. Chem. Soc., Dalton Trans. (1994) 1607.
[12] (a) A.J. Nielson, M.W. Glenny, C.E.F. Rickard, J.M. Waters, J.
Chem. Soc., Dalton Trans. (2000) 4569;
[17] G.M. Sheldrick, Acta Cryst. A46 (1990) 467.
[18] G.M. Sheldrick, Program for the Refinement of Crystal Structures,
University of Go¨ttingen, Germany, 1997.
[19] L.J. Farrugua, Ortep-3 for Windows, Version 1.06, University of
Glasgow, Scotland, UK.
[20] (a) A.D. Becke, J. Chem. Phys. 98 (1993) 5648;
(b) J.A. Pople, M. Head-Gordon, D.J. Fox, K. Raghavachari, L.A.
Curtiss, J. Chem. Phys. 90 (1989) 5622;
(b) F. Montilla, A. Gallindo, E. Carmona, E. Gutierrez-Puebla, A.
Monge, J. Chem. Soc., Dalton Trans. (1998) 1299.
(c) L.A. Curtiss, C. Jones, G.W. Trucks, K. Raghavachari, J.A.
Pople, J. Chem. Phys. 93 (1990) 2537.