Stoichiometric olefin insertion into the Ti-C bond of four-coordinate cationic bis(phenolate) titanium aryl and benzyl complexes

Article abstract of DOI:10.1016/S0022-328X(99)00504-5

A series of five-coordinate bis(phenolate) titanium hydrocarbyl complexes (MBP)Ti(η²-R)R′ was prepared, with MBP=2,2′-methylenebis(4-methyl-6-tert-butylphenolate), R=C₆H₄(o-CH₂NMe₂); R′=Cl, OSO₂CF₃, Me, CH₂CMe₃ or R=CH₂C₆H₄(o-NMe₂); R′=Cl, Me. A structure determination of (MBP)Ti[η²-C₆H₄(o-CH₂NMe ₂)]OSO₂CF₃ showed the metal to have a distorted trigonal bipyramidal coordination geometry. Cationic four-coordinate derivatives [(MBP)Ti(η²-R)]⁺ were generated by reacting the R′=Me derivatives with the Lewis acid B(C₆F₅)₃. These cations were found to undergo stoichiometric insertions of ethene and propene into the Ti-C bond, as seen by 1D and 2D-NMR spectroscopy and quenching reactions with CD₃OD.

Full text of DOI:10.1016/S0022-328X(99)00504-5

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 591 (1999) 88–95
Stoichiometric olefin insertion into the TiꢀC bond of
four-coordinate cationic bis(phenolate) titanium aryl and benzyl
complexesꢀ
Esther E.C.G. Gielens, Tessa W. Dijkstra, Pietro Berno, Auke Meetsma, Bart Hessen *,
Jan H. Teuben
Centre for Catalytic Olefin Polymerisation, Stratingh Institute, Uni6ersity of Groningen, Nijenborgh 4, NL-9747 AG Groningen, The Netherlands
Received 23 June 1999; accepted 15 August 1999
Abstract
A series of five-coordinate bis(phenolate) titanium hydrocarbyl complexes (MBP)Ti(h2-R)R% was prepared, with MBP=2,2%-
methylenebis(4-methyl-6-tert-butylphenolate), R=C6H4(o-CH2NMe2); R%=Cl, OSO2CF3, Me, CH2CMe3 or R=CH2C6H4(o-
NMe2); R%=Cl, Me. A structure determination of (MBP)Ti[h2-C6H4(o-CH2NMe2)]OSO2CF3 showed the metal to have a
distorted trigonal bipyramidal coordination geometry. Cationic four-coordinate derivatives [(MBP)Ti(h2-R)]+ were generated by
reacting the R%=Me derivatives with the Lewis acid B(C6F5)3. These cations were found to undergo stoichiometric insertions of
ethene and propene into the TiꢀC bond, as seen by 1D and 2D-NMR spectroscopy and quenching reactions with CD3OD. © 1999
Elsevier Science S.A. All rights reserved.
Keywords: Olefin insertion; Titanium alkyls; Cationic complexes
quent insertion. It appeared to us that the
bis(phenolate) Ti system would be very interesting and
suitable for studies on well-defined cationic hydrocarbyl
species. For a bis(naphtolate) Zr system a well-defined
cationic hydrocarbyl complex has been generated and
characterised [2], but this appears to be much more
difficult for the Ti systems. Recently, a well-character-
ised cationic Ti species was reported with highly substi-
tuted phenolate ancillary ligands, [(ArO)2TiCH2Ph]-
1. Introduction
Bis(phenolates) of the type 2,2%-X-bis(4-methyl-6-
tert-butylphenolate) (X=CH2, CH2CH2, S) have been
found to be useful dianionic ancillary ligands for
Group 4 transition metals. In particular, it was ob-
served that the bis(phenolate)MCl2/MAO combinations
(M=Ti, Zr) are reasonably active catalysts for the
catalytic polymerisation of ethene, a-olefins and butadi-
ene [1,2], the copolymerisation of ethene and styrene
[1,3], and the polymerisation of styrene to syndiotactic
polystyrene [4]. Recent theoretical studies [5] indicated
that the higher catalytic activities observed for the
bis(phenolates) with a bridging group X that has coor-
dinative ability (e.g. S) may stem from a destabilisation
of the olefin adduct of the bis(phenolate)TiR cation
(which is now four coordinate rather than three coordi-
nate), thus lowering the activation barrier to subse-
[h6-PhCH2B(C6F5)3]
with
ArO=2,6-diphenyl-3,5-
dimethyl-phenolate [6].
We sought to obtain stable cationic four-coordinate
bis(phenolate)TiR species by using aryl or benzyl
groups R with an additional Lewis base functionality
attached, R=C6H4(o-CH2NMe2) and CH2C6H4(o-
NMe2). In this paper, we report the synthesis of the
neutral complexes (MPB)Ti(h2-R)R% (MBP=2,2%-
methylenebis(4-methyl-6-tert-butylphenolate); R%=Cl,
OSO2CF3, Me, CH2CMe3) and their cationic deriva-
tives [(MBP)Ti(h2-R)]+. When studying the reactivity
of the cationic derivatives towards ethene and propene,
we observed stoichiometric single insertion of the olefin
into the titaniumꢀcarbon bond in all cases.
ꢀ Netherlands Institute for Catalysis Research (NIOK) publication
no. RUG 99-4-02.
* Corresponding author. Fax: +31-50-363-4315.
E-mail address: hessen@chem.nug.nl (B. Hessen)
0022-328X/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(99)00504-5

E.E.C.G. Gielens et al. / Journal of Organometallic Chemistry 591 (1999) 88–95
89
to be insensitive to room light. Reaction of 2a with
AgO3SCF3 in THF yielded the triflate (MBP)Ti[h2-
C6H4(o-CH2NMe2)]OSO2CF3 (3), which was struc-
turally characterised. The molecular structure of 3 (Fig.
1, selected interatomic distances and angles in Table 1)
shows the metal to have a distorted trigonal bipyrami-
dal five-coordinate geometry. One of the MBP oxygens,
O(2), and the dimethylamino functionality of the h2-
C6H4(o-CH2NMe2) group are located in the axial posi-
tions, and the triflate group is h1-bound to the metal.
2. Synthesis of (MBP)Ti(h2-R)R% complexes
A convenient starting material for this chemistry is
the bis(phenolate) dichloride (MBP)TiCl2 (1), from
which the dimethyl derivative (MBP)TiMe2 was also
reported [7,8]. Complexes (MBP)Ti(h2-R)Cl (2a, R=
C6H4(o-CH2NMe2); 2b, R=CH2C6H4(o-NMe2)) were
obtained by reaction of 1 with stoichiometric amounts
of the appropriate RLi reagent. The compounds are
yellow solids, thermally stable at ambient temperature
both in the solid state and in solution, and they appear
When compared with the octahedral [2,2%-thiobis-
(4 - methyl - 6 - tert - butylphenolate)]Ti(h2 - C6H4(o-CH2-
NMe2)Cl [9], the TiꢀO and TiꢀC distances in 3 are
somewhat shorter, as expected for the lower coordina-
tion number, although the TiꢀN distances are similar in
both complexes. As the NꢀTiꢀC(24) angle of the h2-R
group in 3 is rather small with 74.89(7)°, the MBP
O(1)ꢀTiꢀO(2) angle (102.43(7)°) and the O(3)ꢀTiꢀC(24)
angle (133.26(8)°) open up somewhat, making 3 consid-
erably more distorted than the tbp five-coordinate com-
plex (MBP)TiCl2(THF) [8], which has an MBP OꢀTiꢀO
angle of 96.2(1)°.
Compounds 2 react with either MeLi or MgMgI to
give the corresponding monomethyl derivatives
(MBP)Ti(h2-R)Me (4a, R=C6H4(o-CH2NMe2); 4b,
R=CH2C6H4(o-NMe2)). These yellow air-sensitive
compounds seem to be stable in the solid state at
ambient temperature, but gradually decompose in solu-
tion, liberating methane. The reaction of 2a with
LiCH2CMe3 resulted in the formation of the neopentyl
derivative (MBP)Ti[h2-C6H4(o-CH2NMe2)]CH2CMe3
(5). The solution behaviour of compounds 2, 4 and 5
was studied by NMR spectroscopy. At or above ambi-
Fig. 1. Molecular structure of 3. Hydrogen atoms are omitted for
clarity. Thermal ellipsoids are drawn at the 50% probability level.
Table 1
,
Selected interatomic distances (A) and angles (°) for 3
Interatomic distances
Ti(1)ꢀO(1)
Ti(1)ꢀO(2)
Ti(1)ꢀO(3)
Ti(1)ꢀC(24)
Ti(1)ꢀN(1)
1.792(2)
1.784(2)
1.999(2)
2.116(2)
2.298(2)
S(1)ꢀO(3)
S(1)ꢀO(4)
S(1)ꢀO(5)
S(1)ꢀC(33)
1.488(2)
1.414(2)
1.426(2)
1.820(3)
1
ent temperature the H-NMR spectra show a Cs sym-
metrically averaged structure with equivalent MBP
tert-butyl and methyl resonances and dimethylamino
methyl groups. At low temperatures the spectra indicate
an asymmetric structure as expected for the tbp-type
complex seen in the solid state for 3. From the coales-
cence temperatures of the MBP methyl resonances, free
energies of activation can be estimated (shown in Table
2). For the observed fluxional behaviour two pathways
are available: (A) dissociation of the NMe2-functional-
ity followed by rotation around the TiꢀC bond (as
Interatomic angles
O(1)ꢀTi(1)ꢀO(2)
O(1)ꢀTi(1)ꢀO(3)
O(2)ꢀTi(1)ꢀN(1)
N(1)ꢀTi(1)ꢀC(24)
O(1)ꢀTi(1)ꢀC(24) 108.47(8)
O(3)ꢀTi(1)ꢀC(24) 133.26(8)
102.43(7)
113.71(7)
161.48(7)
74.89(7)
O(1)ꢀTi(1)ꢀN(1)
O(3)ꢀTi(1)ꢀN(1)
Ti(1)ꢀO(1)ꢀC(1)
Ti(1)ꢀO(2)ꢀC(18)
Ti(1)ꢀO(3)ꢀS(1)
94.00(7)
82.88(7)
137.1(1)
156.1(1)
164.2(1)

90
E.E.C.G. Gielens et al. / Journal of Organometallic Chemistry 591 (1999) 88–95
Table 2
This was attributed to the large steric requirement of
the MBP ligand.
1H-NMR (500 MHz) coalescence temperatures of the MBP methyl
protons of (MBP)Ti(h2-R)R% and derived free energies of activation
(standard deviations in parentheses)
3. Generation and reactivity of (MBP)Ti(h2-R)-catons
Compound
Tc (K)
DG‡Tc (kJ mol−1
)
Compound 4a reacts with the Lewis acid B(C6F5)3 in
2a
2b
4a
4b
5
261
233
227
B200
261
51.3(4)
45.5(5)
44.3(3)
n.d.
bromobenzene-d5 solvent to give the ionic species
{(MBP)Ti[h2-C6H4(o-CH2NMe2)]}[MeB(C6F5)3]
(6a)
cleanly, as seen by NMR spectroscopy. The 19F-NMR
spectrum indicates that the MeB(C6F5)3-anion is non-
53.0(3)
1
coordinating, with Dl(p−m)=2.7 ppm [10]. The H-
and 13C-NMR spectra suggest a Cs symmetry for the
cation, in which the metal centre is likely to be pseudo-
tetrahedrally coordinated. Characteristic differences in
the NMR spectra of 6a relative to neutral 4a include a
1 ppm upfield shift of the MBP methylene protons and
a 6 ppm downfield shift of the TiꢀCipso resonance.
The benzyl complex 4b reacts similarly with B(C6F5)3
in bromobenzene-d5 to give the ionic species
{(MBP)Ti[h2-CH2C6H4(o-NMe2)]}[MeB(C6F5)3] (6b).
suggested for the octahedral [2,2%-thiobis(4-methyl-6-
tert-butylphenolate)]Ti(h2-C6H4(o-CH2NMe2)Cl [9]), or
(B) by a non-dissociative pseudorotation process. It is
clear that the activation barrier for symmetrisation is
dependent both on R and R%. The observations are that
DG‡ is smaller for R%=Me than for R%=Cl, and
smaller for R=CH2C6H4(o-NMe2) than for R=
C6H4(o-CH2NMe2). The latter could be due to a re-
duced Lewis acidity of the NMe2 functionality in the
arylamine relative to the alkylamine (suggesting path-
way A), or due to reduced steric encumbrance of the
metal centre in the benzylic species relative to the s-aryl
species (suggesting pathway B). The dependance on R%
could be due to a reduced Lewis acidity of the metal in
the methyl derivatives (suggesting pathway A). How-
ever, increasing the steric bulk of R to neopentyl slows
down the process, more in line with pathway B, al-
though this could also derive from hindered rotation
around the Tiꢀaryl s-bond in 5 after amine dissocia-
tion. Thus, based on the present set of data, it is
difficult to attribute the observed fluxionality in these
neutral five-coordinate compounds unequivocally to
one specific pathway. It has to be noted that the THF
molecule in (MBP)TiCl2(THF) is readily lost on warm-
ing in vacuo, and that the related dibromide and diio-
dide derivatives do not form isolable THF adducts [8].
1
In contrast to 6a, the H- and 13C-NMR spectra of 6b
suggest some kind of fluxional behaviour (temperature-
dependent broadening of resonances), although this
could not be frozen out above the freezing point of
bromobenzene. This fluxionality may be associated
with a folded envelope geometry of the TiꢀCꢀCꢀCꢀN
ring due to additional interaction of the electron-defi-
cient metal centre with the aryl moiety. Again, the
MeB(C6F5)3 anion appears to be non-coordinating.
Addition of an excess of ethene to a solution of 6a in
bromobenzene-d5, monitored by 1H-NMR spec-
troscopy, initially shows formation of a product with
two new triplet resonances (J=6.7 Hz) at l 3.41 and
2.6 ppm (2H each), suggestive of a single insertion of
ethene into the Tiꢀaryl bond (7a–1, Scheme 1). How-
ever, this product rearranges over a period of 15 min to
an asymmetric species containing a TiꢀCH(Me)ꢀAr
1
moiety (7a–2). This is seen in the H-NMR spectrum
from two new resonances, a doublet (l 1.60 ppm,
Scheme 1.

E.E.C.G. Gielens et al. / Journal of Organometallic Chemistry 591 (1999) 88–95
91
nate species [(MBP)Ti(h2-R)]+ 6a and 6b, it is likely
that the observed olefin insertion into the TiꢀC bonds
of these complexes takes place through initial coordina-
tion of the olefin to the four-coordinate metal centre.
Indeed, from the calculations by Morokuma et al. on
olefin insertion into the TiꢀMe bond of the four-coordi-
nate [2,2%-thiobis(4-methyl-6-tert-butylphenolate)]TiMe-
cation [5], it seems likely that this is a relatively facile
process. However, unlike the case of this thiobis(pheno-
late) system, the position of the Lewis base functional-
ity in the [(MBP)Ti(h2-R)]+ complex is not constrained
relative to the bis(phenolate) ligand. Due to the small
CꢀTiꢀN ‘bite’ angle of the h2-R group in 6 the metal
centre is relatively open, allowing the incoming olefin
easy access. A single olefin insertion causes a significant
increase in this angle. Thus, the approach of a second
olefin molecule to the metal centre of the insertion
products 7 and 8 is expected to be much more difficult.
This, combined with the affinity of the highly Lewis
acidic metal centre for the dimethyamino functionality
(which makes dissociation of the base unfavourable) is
a possible cause for the occurrence of stoichiometric,
rather than catalytic reactivity of the cations 6 with
olefins.
Several cationic zirconocene systems are known that
either will effect only stoichiometric olefin insertion
(e.g. [Cp2Zr(2-pyridyl)(THF)]+ [12] and [Cp2Zr-
(lutidyl)]+ [13]), or will polymerise olefins at or above
ambient temperature, but allow the product of the first
olefin insertion to be observed at lower temperatures,
such as the zwitterionic reaction products of
Cp2Zr(butadiene) with B(C6F5)3 [14]. For more elec-
tron-deficient non-metallocene cationic systems, stoi-
chiometric olefin insertions have so far only been
observed into Mꢀbenzyl bonds, where h6-coordination
to the metal centre of the benzyl aryl group in the first
olefin insertion product quenches the electron deficiency
of the metal centre [6,15]. In the cationic bis(phenolate)
titanium complexes presented here, the metal centre in
the first insertion products 7 and 8 is formally still
sufficiently electron deficient to allow olefin complexa-
tion and insertion without the need for dissociation of
the NMe2-group. There is no evidence for h6-coordina-
tion of the R-aryl moiety in these compounds (a typical
NMR feature of which is the presence of one or several
upfield shifted aryl H resonances with lB6.5 ppm
[6,15]), although it is difficult to exclude interaction of
a part of the aromatic system with the metal centre (e.g.
of the CꢀC bond between the two substituted carbons),
as the aromatic region of the 13C-NMR spectra of 7
and 8 is difficult to analyse. It could be that in this
system the geometry around the metal centre enforced
by the ligands in 7 and 8 (an eight-membered MBPꢀTi
ring and a six- or seven-membered (h2-C2R)-Ti ring) is
sufficient to block further olefin insertion.
Scheme 2.
J=6.4 Hz, 3H) and a quartet (l 3.32 ppm, J=6.4 Hz,
1H), and in the 13C-NMR spectrum by resonances at l
16.80 ppm (q, J=126.7 Hz) and l 110.78 ppm (d,
J=121.3 Hz). No subsequent reaction with ethene was
observed, and quenching with methanol-d4 liberates the
MeCHDC6H4CH2NMe2 fragment, as seen by GC–MS
(164 m/e). Thus, it appears that initial insertion of
ethene into the Tiꢀaryl bond is followed by a rearrange-
ment that is likely to proceed through b-H elimination
and 2,1-reinsertion of the substituted styrene ligand
thus formed. Styrenes are known to have an electronic
preference for secondary insertion (as seen, for exam-
ple, in the catalytic formation of syndiotactic
polystyrene with organo-titanium catalysts [11]).
This rearrangement to give a complex with a sec-
ondary carbon attached to Ti could be a reason why
only stoichiometric, not catalytic, reactivity with ethene
is observed. The observation that this rearrangement is
substantially slower than the initial insertion of ethene
into the Tiꢀaryl bond makes this unlikely, however.
Indeed, upon reaction of the cationic benzyl species 6b
with ethene a stoichiometric reaction is observed
also, producing the (MBP)Ti[h2-CH2CH2CH2C6H4(o-
NMe2)]-cation (7b, Scheme 2). 1H,1H-COSY-NMR
clearly indicates the presence of a linear (CH2)3 moiety,
and in this case no subsequent rearrangement occurs.
This shows that the rearrangement observed in 7a is
driven by the formation of a thermodynamically more
favourable benzylic species.
As with ethene, the reaction of 6a with an excess of
propene also shows the stoichiometric insertion of one
molecule of the olefin into the TiꢀC bond, leading to
the product 8 (Scheme 1). Again, the formation of the
product shown was deduced from the 1H,1H-COSY
and 13C-NMR spectra and from quenching reactions
with methanol-d4 to give CH2DCH(Me)C6H4CH2NMe2
(178 m/e, GC–MS). In this case, the reaction product
of 6a with the olefin does not give a subsequent rear-
rangement. Apparently, the generation after such a
rearrangement of a tertiary carbon centre attached to
titanium is sufficiently unfavourable to offset the for-
mation of a benzylic species here.
4. Discussion
Due to the Lewis acidity and the low coordination
number of the metal centre in the cationic four-coordi-

92
E.E.C.G. Gielens et al. / Journal of Organometallic Chemistry 591 (1999) 88–95
5. Experimental
J=136.2 Hz, NMe2), 35.74 (t, J=127.0 Hz, MBP
CH2), 35.19 (s, CMe3), 30.63 (q, J=126.1 Hz,
C(CH3)3), 21.13 (q, J=126.1 Hz, MBP Me). Anal.
Found: C, 68.50; H, 8.20; Ti, 8.52. Calc. for
C32H42ClNO2Ti: C, 69.12; H, 7.61; Ti, 8.61.
5.1. General considerations
All experiments were performed under nitrogen using
standard Schlenk, glovebox and vacuum-line tech-
niques. Diethyl ether, toluene, benzene-d6 and pentane
were distilled from Na/K or Na prior to use. Bro-
mobenzene-d5 (Aldrich) was degassed, dried on molecu-
5.3. Synthesis of (MBP)Ti[p2-CH2C6H4(o-NMe2)]Cl
(2b)
,
lar sieves (4 A) and stored and used in a dry-box. The
compounds [7,8], LiC6H5(o-CH2NMe2) and
To a mixture of solid 1 (12.36 g, 27.0 mmol) and
LiCH2C6H5(o-NMe2) (4.64 g, 32.9 mmol) was added a
300 ml portion of cold (−80°C) ether. The orange
suspension was warmed to −30°C and stirred for 3.5
h. The mixture was subsequently warmed to ambient
temperature and stirred overnight. The suspension was
filtered, and the solvent was removed in vacuo. Extract-
ing the product with ether and cooling to −80°C
yielded orange crystalline 2b (6.99 g, 12.6 mmol, 47%).
1H-NMR (C6D6, 25°C) l 7.03 and 6.98 (s, 2H each,
MBPꢀAr H), 6.90 (m, 1H, RꢀAr H), 6.8 (m, 2H, RꢀAr
H), 6.73 (m, 1H, RꢀAr H), 4.70 and 3.41 (d, J=13.8
Hz, 1H each, MBP CH2), 3.05 (s, 6H, NMe2), 2.92
(br.s, 2H, TiCH2), 2.12 (s, 6H, MBPꢀCH3), 1.50 (s,
18H, MBPꢀCMe3). 13C-NMR (C6D6, 25°C) l 163.11 (s,
OꢀArCipso),150.73 (s, RꢀAr Cipso), 140.01 (s, RꢀAr
Cipso), 137.11, 136.84 and 131.42 (3×MBPꢀCipso),
129.23 (d, J=158.7 Hz, MBPꢀAr CH), 129.12 (d,
J=158.7 Hz, RꢀAr CH), 127.74 (d, J=158.7 Hz,
RꢀAr CH), 126.32 (d, J=161.1 Hz, RꢀAr CH), 125.77
(d, J=152.6 Hz, MBPꢀAr CH), 117.67 (d, J=156.9
Hz, RꢀAr CH), 79.02 (t, J=131.8 Hz, TiCH2), 49.48
(q, J=138.6 Hz, NMe2), 35.25 (t, J=127.0 Hz, MBP
CH2), 35.18 (s, CMe3), 30.57 (q, J=125.7 Hz,
C(CH3)3), 21.67 (q, J=125.7 Hz, MBP Me). Anal.
Found: C, 68.47; H, 7.60; Ti, 8.50. Calc. for
C32H42ClNO2Ti: C, 69.12; H, 7.61; Ti, 8.61.
1
LiCH2C6H4(o-NMe2) [16] and B(C6F5)3 [17] were pre-
pared according to published procedures. Ethene and
propene (Ucar 99.5%) were used as received. NMR
spectra were recorded on Varian Unity-500 (1H and
13C) and Gemini-200 (19F) spectrometers. GC–MS
analyses were performed using an HP 5973 mass-selec-
tive detector attached to a HP 6890 GC equipped with
a HP-1 dimethylpolysiloxane column. Elemental analy-
ses were performed at the Micro-Analytical Depart-
ment of the University of Groningen. Listed values are
the average of at least two independent determinations.
It was noted that most of the bis(phenolate) hydrocar-
byl species give carbon analyses that are persistently
and reproducibly too low (by up to 1.5%, as in the case
for 4b), whereas correct H and Ti values are obtained.
It is possible that formation of inert Tiꢀcarbides during
the combustion analysis (as seen by the formation of
non-white residual ashes) is responsible for this.
5.2. Synthesis of (MBP)Ti[p2-C6H4(o-CH2NMe2)]Cl
(2a)
To a mixture of solid 1 (6.38 g, 14.0 mmol) and
LiC6H5(o-CH2NMe2) (2.00 g, 14.2 mmol) was added a
150 ml portion of cold (−80°C) ether. The orange
suspension was warmed to −30°C and stirred for 3.5
h. The mixture was subsequently warmed to ambient
temperature and stirred overnight. The suspension was
filtered, and the solvent was removed in vacuo. Extract-
ing the product with an ether–pentane (1:1) mixture
and cooling to −80°C yielded orange crystalline 2a
(5.22 g, 9.39 mmol, 67%). 1H-NMR (C6D6, 25°C) l
7.86 (d, J=7.3 Hz, 1H, RꢀAr H), 7.04 and 6.99 (s, 2H
each, MBPꢀAr H), 6.92 (d, J=7.3 Hz, 1H, RꢀAr H),
6.7 (m, 2H, RꢀAr H), 4.87 and 3.44 (d, J=13.4 Hz, 1H
each, MBPꢀCH2), 3.71 (br.s, 2H, NCH2), 2.76 (s, 6H,
NMe2), 2.10 (s, 6H, MBPꢀCH3), 1.53 (s, 18H,
MBPꢀCMe3). 13C-NMR (C6D6, 25°C) l 196.36 (s,
TiꢀCipso), 161.91 (s, OꢀArCipso), 143.62 (s, NCH2ꢀCipso),
137.4 (br. s, 2×MBPꢀCipso), 132.52 (d, J=158.8 Hz,
R-Ar CH), 131.55 (s, MBPꢀCipso), 130.24 (d, J=161.1
Hz, RꢀAr CH), 129.10 (d, J=158.7 Hz, MBPꢀAr CH),
125.98 (d, J=155.0 Hz, MBPꢀAr CH), 125.74 (d,
J=159.9 Hz, RꢀAr CH), 124.56 (d, J=157.5 Hz,
RꢀAr CH), 67.78 (t, J=137.3 Hz, NCH2), 49.48 (q,
5.4. Synthesis of
(MBP)Ti[p2-C6H4(o-CH2NMe2)]OSO2CF3 (3)
To an orange solution of 2a (4.73 g, 8.51 mmol) in 50
ml of THF, AgOSO2CH3 (2.05 g, 7.98 mmol) was
added at room temperature. The solution turned red
instantaneously, and AgCl precipitated. After stirring
for 2 h, the solution was filtered and the solvent re-
moved in vacuo. The resulting red oil was dissolved in
40 ml of ether. Cooling to −30°C yielded orange
crystalline 3·(Et2O)0.5 (4.80 g, 6.79 mmol, 85%). 1H-
NMR (C6D6, 25°C) l 7.69 (d, J=7.3 Hz, 1H, RꢀAr
H), 6.91 (br s, 4H MBPꢀAr H), 6.85 (t, J=7.6 Hz, 1H,
RꢀAr H), 6.63 (d, J=7.8 Hz, 1H, RꢀAr H), 6.60 (t,
J=7.3 Hz, 1H, RꢀAr H), 4.78 and 3.24 (d, J=13.7
Hz, 1H each, MBP CH2), 3.67 (br.s, 2H, NCH2), 2.68
(s, 6H, NMe2), 2.02 (s, 6H, MBPꢀCH3), 1.45 (s, 18H,
MBPꢀCMe3). 19F-NMR (C6D6, 25°C) l ꢀ77.57. The
diethyl ether in the crystal lattice appears to be rela-

E.E.C.G. Gielens et al. / Journal of Organometallic Chemistry 591 (1999) 88–95
93
tively loosely incorporated. Solvent-free solid 3 for
elemental analysis was obtained by finely grinding the
crystalline material followed by drying overnight at
ambient temperature under diffusion-pump vacuum.
Anal. Found: C, 58.15; H, 6.39; Ti, 7.26. Calc. for
C33H42F3NO5STi: C, 59.19; H, 6.32; Ti, 7.15.
solid LiCH2CMe3 was added. The mixture was stirred
for 2 h, warmed to ambient temperature and then
stirred for a further 20 min. The solvent was removed
in vacuo and residual solvent was removed from the
dark yellow residue by stirring with 30 ml of pentane,
which was subsequently pumped off. Extraction with
pentane and evaporation of the solvent resulted in a
yellow solid that was rinsed with cold pentane and
dried in vacuo. Yield: 1.30 g (2.20 mmol, 59%) of 5.
1H-NMR (C7D8, −40°C) l 8.14 (d, J=7.8 Hz, 1H,
RꢀAr H), 6.8 (m, 3H, Ar H), other aromatic protons
overlapped by solvent, 4.23 and 2.84 (d, J=13 Hz, 1H
each, NCH2), 3.37 and 3.17 (d, J=13.8 Hz, 1H each,
MBP CH2), 2.74 and 2.20 (s, 3H each, NMe2), 2.66 and
1.29 (d, J=10.7 Hz, 1H each, TiCH2), 2.17 and 2.05 (s,
3H each, MBPꢀCH3), 1.74 and 1.65 (s, 9H each,
MBPꢀCMe3), 1.23 (s, 9H, CMe3). 13C-NMR (selected
data, C6D6, 25°C) l 192.22 (s, TiꢀCipso), 108.68 (t,
J=114.2 Hz, TiꢀCH2). Anal. Found: C, 74.31; H, 9.34;
Ti, 7.92. Calc. for C37H53NO2Ti: C, 75.10; H, 9.03; Ti,
8.09.
5.5. Synthesis of (MBP)Ti[p2-C6H4(o-CH2NMe2)]Me
(4a)
To a red suspension of 2a (5.18 g, 8.74 mmol) in 100
ml of ether, cooled to −80°C, 3.3 ml of a 2.7 M MeLi
solution in ether was added dropwise. After 2 h the
mixture was allowed to warm to ambient temperature
and stirred for a further hour. Subsequently, 5 g of
dried Celite was added. The mixture was filtered, and
the solution concentrated to 40 ml. Cooling to −80°C
yielded yellow microcrystalline 4a (3.58 g, 6.70 mmol,
76%). 1H-NMR (C6D6, 25°C) l 7.91 (d, J=7.2 Hz, 1H,
RꢀAr H), 7.05 (br.s, 4H, MBPꢀAr H), 7.00 (d, J=7.4
Hz, 1H, RꢀAr H), 6.8 (m, 2H, RꢀAr H), 4.16 and 3.42
(d, J=13.5 Hz, 1H each, MBP CH2), 3.57 (br.s, 2H,
NCH2), 2.60 (s, 6H, NMe2), 2.14 (s, 6H, MBPꢀCH3),
1.59 (s, 18H, MBPꢀCMe3), 1.46 (s, 3H, TiꢀMe). 13C-
NMR (selected data, C6D6, 25°C) l 191.39 (s, TiꢀCipso),
62.72 (q, J=120.9 Hz, TiꢀCH3). Anal. Found: C,
73.76; H, 8.75; Ti, 8.83. Calc. for C33H45NO2Ti: C,
74.00; H, 8.47; Ti, 8.94.
5.8. Generation of {(MBP)Ti[p2-C6H4(o-CH2NMe2)]}-
[MeB(C6F5)3] (6a)
A solution of 4a (43.2 mg, 80.7 mmol) in 0.25 ml of
bromobenzene-d5 was added to a solution of B(C6F5)3
(45.2 mg, 88.3 mmol) in 0.25 ml of bromobenzene-d5.
The solution immediately turned wine-red, and NMR
spectroscopy showed formation of 6a as the only
product. 1H-NMR (C6D5Br, −35°C) l 7.0 (m, 2H,
RꢀAr H), 6.85 and 6.72 (s, 2H each, MBPꢀAr H), 6.80
(m, 2H, RꢀAr H), 4.11 (br.s, 2H, NCH2), 3.46 and 2.57
(d, J=13.9 Hz, 1H each, MBP CH2), 2.50 (s, 6H,
NMe2), 2.09 (s, 6H, MBPꢀCH3), 1.26 (s, 18H,
MBPꢀCMe3), 1.09 (br.s, 3H, BꢀMe). 13C-NMR
(C6F5 resonances omitted, C6D5Br, −35°C) l 197.30
(s, TiꢀCipso), 162.64 (s, OꢀArCipso), 142.58 (s,
NCH2ꢀCipso), 136.62 (s, MBPꢀCipso), 132.96 (s,
MBPꢀCipso), 132.43 (d, J=138.0 Hz, RꢀAr CH),
131.91 (d, J=167.7 Hz, RꢀAr CH), 128.07 (s,
MBPꢀCipso), 127.67 (d, J=155.0 Hz, RꢀAr CH), three
other aryl resonances overlapped by solvent, 65.21 (t,
J=145.5 Hz, NCH2), 49.53 (q, J=142.4 Hz, NMe2),
35.02 (s, CMe3), 32.63 (t, J=125.0 Hz, MBP CH2),
30.68 (q, J=126.1 Hz, C(CH3)3), 21.16 (q, J=126.2
Hz, MBP Me), 11.21 (br.q, J=115 Hz, BꢀMe). 19F-
NMR (C6D5Br, 25°C) l −167.66 (m−F), −164.96
(p-F), −133.29 (o-F).
5.6. Synthesis of (MBP)Ti[p2-CH2C6H4(o-NMe2)]Me
(4b)
To a red suspension of 2b (1.34 g, 2.41 mmol) in 100
ml of ether, cooled to −80°C, 2.2 ml of a 1.32 M
MeMgI solution in ether was added dropwise. After 3 h
the mixture was allowed to warm to ambient tempera-
ture and stirred for another 2 h. To the yellow–orange
solution 0.25 ml (2.90 mmol) of 1,4-dioxane was added,
precipitating a white solid. After addition of 5 g of
dried Celite the solution was filtered. Concentrating the
solution and cooling to −80°C yielded yellow 4b (0.75
1
g, 1.40 mmol, 58%). H-NMR (C6D6, 25°C) l 7.06 and
7.02 (s, 2H each, MBPꢀAr H), 6.9 (m, 3H, RꢀAr H),
6.85 (m, 1H, RꢀAr H), 4.18 and 3.43 (d, J=13.7 Hz,
1H each, MBP CH2), 2.87 (s, 6H, NMe2), 2.75 (br.s,
2H, TiCH2), 2.16 (s, 6H, MBPꢀCH3), 1.55 (s, 18H,
MBPꢀCMe3), 1.12 (s, 3H, TiꢀMe).13C-NMR (selected
data, C6D6, 25°C) l 69.32 (t, J=130.6 Hz, TiꢀCH2),
56.39 (q, J=120.5 Hz, TiꢀCH3). Anal. Found: C,
72.38; H, 8.45; Ti, 8.95. Calc. for C33H45NO2Ti: C,
74.00; H, 8.47; Ti, 8.94.
5.9. Generation of {(MBP)Ti[p2-CH2C6H4(o-NMe2)]}-
[MeB(C6F5)3] (6b)
5.7. Synthesis of (MBP)Ti[p2-C6H4(o-CH2NMe2)]-
CH2CMe3 (5)
To a red suspension of 2a (2.08 g, 3.74 mmol) in 60
ml of ether, cooled to −80°C, 0.40 g (5.12 mmol) of
A solution of 4b (31.6 mg, 59.0 mmol) in 0.25 ml of
bromobenzene-d5 was added to a solution of B(C6F5)3

94
E.E.C.G. Gielens et al. / Journal of Organometallic Chemistry 591 (1999) 88–95
(31.7 mg, 61.9 mmol) in 0.25 ml of bromobenzene-d5.
The solution immediately turned wine-red, and NMR
spectroscopy showed formation of 6b as the sole
product. 1H-NMR (C6D5Br, 30°C) l 7.35 (m, 2H,
RꢀAr H), 7.12 and 6.93 (s, 2H each, MBPꢀAr H), 7.09
(m, 2H, RꢀAr H), 3.66 (d, J=13.7 Hz, 1H, MBP
CHH), 3.2 (br, 3H, MBP CHH and TiCH2), 2.94 (br.s,
6H, NMe2), 2.20 (s, 6H, MBPꢀCH3), 1.16 (s, 18H,
MBPꢀCMe3), 1.08 (br.s, 3H, BꢀMe). 13C{1H}-NMR
(C6F5 resonances omitted, C6D5Br, 30°C) l 162.29 (s,
OꢀArCipso), 149.11 (s, RꢀAr Cipso), 147.11 (s, RꢀAr
Cipso), 135.95 and 134.39 (2×MBPꢀCipso), 131.81,
121.88, 119.86, 112.86, 90.60, 69.80 (assignment am-
biguous), 46.67 (NMe2), 34.66 (s, CMe3), 33.81 (MBP
CH2), 29.96 (C(CH3)3), 21.01 (MBP Me), 10.92 (br,
BꢀMe). 19F-NMR (C6D5Br, 25°C) l −168.09 (m-F),
−165.46 (p-F), −133.45 (o-F).
J=126.7 Hz, TiCHCH3), 11.16 (br, BꢀMe). Similar
reactions in which an excess of ethene was used did not
lead to subsequent ethene insertions or polymerisation.
5.11. Reaction of 6b with ethene
A solution of 6b in bromobenzene-d5, prepared as
above using 40.8 mg (76.2 mmol) of 4b and 41.4 mg
(80.1 mmol) of B(C6F5)3, was placed in an NMR tube
equipped with a Teflon (Young) valve and attached to
a vacuum line. The solution was frozen in liquid nitro-
gen and evacuated. Subsequently, 76 mmol of ethene
was condensed into the mixture. The valve was closed
and the mixture was thawed out resulting in an or-
ange–red solution. NMR spectroscopy indicated for-
1
mation of 7b. H-NMR (C6D5Br, 45°C) l 7.22 (m, 2H,
RꢀAr H), 7.16 and 6.94 (s, 2H each, MBPꢀAr H), 7.09
(t, J=7.3 Hz, 1H, RꢀAr H), 6.86 (d, J=8.3 Hz, 1H,
RꢀAr H), 3.60 and 3.00 (d, J=13.8 Hz, 1H, MBP
CH2), 3.12 (t, J=6.3 Hz, 2H, TiꢀCH2), 2.81 (s, 6H,
NMe2), 2.65 (m, 2H, ꢀCH2ꢀ), 2.20 (s, 6H, MBPꢀCH3),
2.18 (t, J=6.8 Hz, 2H, RꢀArꢀCH2), 1.24 (s, 18H,
MBPꢀCMe3), 1.05 (br.s, 3H, BꢀMe). 13C-NMR (se-
lected data, C6D5Br, 45°C) l 162.46 (s, OꢀArCipso),
91.73 (t, J=137.1 Hz, TiꢀCH2), 47.51 (q, J=141.8 Hz,
NMe2), 34.61 (s, CMe3), 34.56 (t, J=126.8 Hz, MBP
CH2), 32.37 (t, J=125.0 Hz, ꢀCH2ꢀ), 30.30 (t, J=
133.6, RꢀArꢀCH2), 30.09 (q, J=125.6 Hz, C(CH3)3),
20.93 (q, J=126.2 Hz, MBP Me), 10.68 (br, BꢀMe).
Similar reactions in which an excess of ethene was used
did not lead to subsequent ethene insertions or
polymerisation.
5.10. Reaction of 6a with ethene
A solution of 6a in bromobenzene-d5, prepared as
above using 48.8 mg (91.1 mmol) of 4a and 49.8 mg
(97.3 mmol) of B(C6F5)3, was placed in an NMR tube
equipped with a Teflon (Young) valve and attached to
a vacuum line. The solution was frozen in liquid nitro-
gen and evacuated. Subsequently, 91 mmol of ethene
was condensed into the mixture. The valve was closed
and the mixture was thawed out, resulting in an or-
1
ange–red solution. A H-NMR spectrum was immedi-
ately recorded which showed formation of 7a–1.
1H-NMR (C6D5Br, 25°C) l 7.31 and 7.18 (t, J=7.3
Hz, 1H each, RꢀAr H), 7.14 and 6.95 (s, 2H each,
MBPꢀAr H), 7.06 and 6.99 (d, J=7.3 Hz, 1H each,
RꢀAr H), 3.78 (br.s, 2H, NCH2), 3.59 (d, J=14.2 Hz,
1H, MBP CHH), 3.41 (t, J=6.7 Hz, 2H, CH2), 2.6
(3H, TiꢀCH2 and MBP CHH) 2.44 (s, 6H, NMe2), 2.20
(s, 6H, MBPꢀCH3), 1.30 (s, 18H, MBPꢀCMe3), 1.10
(br.s, 3H, BꢀMe). Upon standing at ambient tempera-
ture, rearrangement of the initial product took place.
After 15 min, conversion to 7a–2 was complete.
H-NMR (C6D5Br, 25°C) l 7.56 and 7.26 (t, J=7.8 Hz,
1H each, RꢀAr H), 7.36 and 7.10 (d, J=7.8 Hz, 1H
each, RꢀAr H), 7.14, 7.04, 7.00 and 6.86 (s, 1H each,
MBPꢀAr H), 4.59 and 3.47 (d, J=13.7 Hz, 1H each,
MBP CH2), 3.53 and 1.81 (d, J=14.2 Hz, NCH2), 3.32
(q, J=6.4 Hz, 1H, TiꢀCH), 2.63 (s, 3H, NMe), 2.2 (s,
6H, NMe and MBPꢀCH3), 2.13 (s, 3H, MBPꢀCH3),
1.60 (d, J=6.4 Hz, 3H TiCHMe), 1.52 and 1.06 (s, 9H
each, MBPꢀCMe3), 1.10 (br.s, 3H, BꢀMe). 13C-NMR
(selected data, C6D5Br, 25°C) l 162.84 and 162.16 (s,
OꢀArCipso), 110.78 (d, J=121.3 Hz, TiꢀCH), 65.35 (t,
J=141.8 Hz, NCH2), 49.15 (q, J=142.7 Hz, NMe),
44.23 (q, J=146.5 Hz, NMe), 35.06 (t, J=125.5 Hz,
MBP CH2), 34.96 and 34.71 (s, CMe3), 30.53 (q, J=
121.3 Hz, C(CH3)3), 29.65 (q, J=126.7 Hz, C(CH3)3),
21.03 and 20.88 (q, J=126.7 Hz, MBP Me), 16.80 (q,
5.12. Reaction of 6a with propene
A solution of 6a in bromobenzene-d5, prepared as
above using 55.5 mg (103.6 mmol) of 4a and 56.3 mg
(110.0 mmol) of B(C6F5)3, was placed in an NMR tube
equipped with a Teflon (Young) valve and attached to
a vacuum line. The solution was frozen in liquid nitro-
gen and evacuated. Subsequently, 104 mmol of propene
was condensed into the mixture. The valve was closed
and the mixture was thawed out, resulting in an deep
red solution. NMR spectroscopy indicated formation of
8a. 1H-NMR (C6D5Br, 25°C) l 7.42 and 7.20 (t, J=7.8
Hz, 1H each, RꢀAr H), 7.24 (d, J=7.8 Hz, 1H, RꢀAr
H), 7.16, 7.12 and 6.90 (s, 1H each, MBPꢀAr H), 7.0
(2H, MBPꢀAr H and RꢀAr H), 4.42 and 3.18 (d,
J=13.7 Hz, 1H each, MBP CH2), 3.61 and 2.72 (d,
J=14.2 Hz, NCH2), 3.51 (m, 1H, CH), 2.95 (dd,
J=11.7 Hz and 3.4 Hz, 1H, TiꢀCHH), 2.63 (ps.t,
J=11.7 Hz, 1H, TiꢀCHH), 2.60 and 2.34 (s, 3H each,
NMe), 2.23 and 2.18 (s, 3H each, MBPꢀCH3), 1.45 and
1.08 (s, 9H each, MBPꢀCMe3), 1.43 (d, J=6.4 Hz, 3H,
CHMe) 1.10 (br.s, 3H, BꢀMe). 13C-NMR (selected
data, C6D5Br, 25°C)
l
163.20 and 162.73 (s,

E.E.C.G. Gielens et al. / Journal of Organometallic Chemistry 591 (1999) 88–95
95
OꢀArCipso), 104.56 (t, J=135.5 Hz, TiꢀCH2), 64.75 (t,
J=143.9 Hz, NCH2), 49.50 (q, J=140.1 Hz, NMe),
46.66 (q, J=140.1 Hz, NMe), 34.85 and 34.51 (s,
CMe3), 34.45 (t, J=127.8 Hz, MBP CH2), 34.26 (d,
J=128.4 Hz, CH), 30.53 (q, J=125.8 Hz, C(CH3)3),
30.09 (q, J=127.1 Hz, C(CH3)3), 23.80 (q, J=129.1
Hz, CHCH3), 20.99 and 20.95 (q, J=126.6 Hz, MBP
Me), 10.84 (br, BꢀMe). Similar reactions in which an
excess of propene was used did not lead to subsequent
propene insertions or polymerisation.
Data Centre, CCDC no. 126019 for compound 3.
Copies of this information may be obtained free of
charge from The Director, CCDC, 12 Union Road,
Cambridge CB2 1EZ (Fax: +44-1223-336-033 or e-
mail deposit@ccdc.cam.ac.uk or http://www.ccdc.
cam.ac.uk).
References
[1] (a) T. Miyatake, K. Mizunuma, M. Kakugo, Makromol. Chem.
Makromol. Symp. 66 (1993) 203. (b) T. Miyatake, K. Mizunima,
Y. Seki, M. Kakugo, Makromol. Chem. Rapid Commun. 10
(1989) 349. (c) T. Miyatake, K. Mizunima, Y. Seki, M. Kakugo,
Chem. Express 2 (1987) 445.
[2] A. van der Linden, C.J. Schaverien, N. Meijboom, C. Ganter,
A.G. Orpen, J. Am. Chem. Soc. 117 (1995) 3008.
[3] F.G. Sernetz, R. Mu¨lhaupt, S. Fokken, J. Okuda, Macro-
molecules 30 (1997) 1562.
5.13. Structure determination of 3·(Et2O)0.5
Suitable single crystals were obtained by crystallisa-
tion from diethyl ether. From one orange crystal with
approximate dimensions 0.20×0.40×0.56 mm, inten-
sity data were recorded on an Enraf–Nonius CAD4-F
diffractometer at 130 K using Mo–Ka radiation with
[4] J. Okuda, E. Masoud, Macromol. Chem. Phys. 199 (1998) 543.
[5] R.D.J. Froese, D.G. Musaev, T. Matsubara, K. Morokuma, J.
Am. Chem. Soc. 119 (1997) 7190.
,
u(Mo–Ka)=0.71073 A.
5.13.1. Crystal data
C33H42F3N1O5STi·(C4H10O)0.5, M=669.63, triclinic,
[6] M.G. Thorn, Z.C. Etheridge, P.E. Fanwick, I.P. Rothwell,
Organometallics 17 (1998) 3636.
[7] C. Floriani, F. Corazza, W. Lesueur, A. Chiesi-Villa, C. Guas-
tini, Angew. Chem. Int. Ed. Engl. 28 (1989) 66.
[8] J. Okuda, S. Fokken, H.-C. Kang, W. Massa, Chem. Ber. 128
(1995) 221.
[9] S. Fokken, T.P. Spaniol, H.-C. Kang, W. Massa, J. Okuda,
Organometallics 15 (1996) 5069.
[10] A.D. Horton, J. de With, A.J. van der Linden, H. van de Weg,
Organometallics 15 (1996) 2672.
[11] A. Grassi, P. Ammendola, A. Zambelli, Makromol. Chem.
Rapid Commun. 8 (1987) 277.
[12] (a) A.S. Guram, R.F. Jordan, Organometallics 10 (1991) 3470.
(b) S. Dagorne, S. Rodewald, R.F. Jordan, Organometallics 16
(1997) 5541 (c) A.S. Guram, R.F. Jordan, Organometallics 9
(1990) 2190.
[13] A.S. Guram, R.F. Jordan, D.F. Taylor, J. Am. Chem. Soc. 113
(1991) 1833.
[14] (a) B. Temme, J. Karl, G. Erker, Chem. Eur. J. 2 (1996) 919. (b)
J. Karl, G. Erker, Chem. Ber. 130 (1997) 1261. (c) J. Karl, M.
Dahlmann, G. Erker, K. Bergander, J. Am. Chem. Soc. 120
(1998) 5643.
(
space group P1, a=11.071(1), b=11.367(1), c=
,
15.185(1) A, h=72.809(4), i=76.394(5), k=
3
,
87.217(5)°, V=1766.7(3) A , Z=2, Dcalc. =1.2588(2) g
cm−3, v(Mo–Ka)=3.5 cm−1. Data (7689 unique
reflections) were collected between 1.45BqB27.0°.
The structure was solved by Patterson methods and
extension of the model was accomplished by direct
methods applied to difference structure factors, as de-
scribed previously [18]. A disordered solvent molecule
was present (with the O-atom located at the inversion
centre), which could not be fitted with a discrete model.
The BYPASS procedure [19] was used to account for this
electron density. Of the molecule of 3 all hydrogen
atoms were located from the difference Fourier map
and included in the refinement with isotropic tempera-
ture factors. Full refinement on Fo converged at RF=
0.038 (wR=0.042, w=1) from 6623 reflections with
I]2.5|(I) and 566 parameters.
[15] (a) C. Pellecchia, A. Grassi, A. Zambelli, Organometallics 13
(1994) 298. (b) C. Pellecchia, A. Immirzi, A. Zambelli, J.
Organomet. Chem. 479 (1994) C9.
[16] J.T.B.H. Jastrebski, G. van Koten, Inorg. Synth. 26 (1989) 150.
[17] A.G. Massey, A. Park, J. Organomet. Chem. 2 (1964) 245.
[18] A.K. Hughes, A. Meetsma, J.H. Teuben, Organometallics 12
(1993) 1936.
[19] P. van der Sluis, A.L. Spek, Acta Crystallogr. Sect. A 23 (1990)
194.
6. Supplementary material
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
.