Chiral metallocenes: The synthesis and X-ray crystal structures of TiCl2(η5:η5-C5Me 4SiMe2C5H3R*) (R* = menthyl or neomenthyl) and related compounds

Article abstract of DOI:10.1139/v01-034

The syntheses of the chiral ansa-metallocene complexes TiCl₂(η⁵:η⁵-C₅ Me₄SiMe₂C₅H₃R*) (R* = menthyl (4a) or neomenthyl (4b)) are reported; initially 4a was obtained as a 3:1 mixture of (R:S) diastereoisomers, which differ only in which face of the asymmetrically substituted cyclopentadienyl ring is bonded to the titanium (chiral descriptor shown). The major diastereomer 4a_R was crystallized out optically pure from the initial reaction mixture, whereas the 4a_S diastereoisomer was isolated after isomerizing a racemic mixture of 4a to a 1:3 mixture of (R:S) diastereoisomers using UV irradiation. The corresponding neomenthyl complex 4b was obtained as a 1.3:1 mixture of diastereoisomers that could not be separated. The optically pure 4a_R was converted stereoselectively into the corresponding (R)-TiMe₂(η⁵:η⁵-C₅Me ₄SiMe₂C₅H₃R*) (R* = menthyl). Syntheses of the related indenyl ligand system C₅Me₄SiMe₂C₉H₇ (3) is reported but complexation to titanium proved to be problematic although ZrCl₂(η⁵:η⁵-C₅Me ₄SiMe₂C₉H₆) (6) was isolated. The crystal structures of both 4a_R and 4a_S are reported and compared. Crystals of 4a_R are orthorhombic, a = 27.857(11), b = 9.985(5), and c = 9.596(4) A, Z = 4, space group P2₁2₁21 (D₂⁴, No. 19), and those of 4a_S are monoclinic, a = 8.5810(10), b = 38.679(4), and c = 8.5842(10) A β = 113.001(2)°, Z = 4, space group P2₁ (C₂², No. 4). The structures were solved by the Patterson method and 4a_R was refined by blocked-cascade least-squares procedures to R = 0.0628 (R_ω = 0.0503) for 902 reflections with |F|/σ(|F|) ≥ 3.0, whereas 4a_S was refined by full-matrix least-squares procedures to R = 0.0646 (wR₂ = 0.1829) for 5734 reflections with |F|/σ(|F|) ≥ 4.0. Both diastereomers of 4a catalyze hydrosilylation of ketones, but as expected from a comparison of the two crystal structures, the 4a_R isomer is the more stereoselective catalyst, i.e., hydrosilylation of acetophenone followed by hydrolysis gives 82% enantiomeric excess (ee) of (S)-PhCH(Me)OH with 4a_R whereas only 16% ee of (R)-PhCH(Me)OH with 4a_S.

Full text of DOI:10.1139/v01-034

731
Chiral metallocenes: the synthesis and X-ray crystal
structures of TiCl2(η5:η5-C5Me4SiMe2C5H3R*) (R* =
menthyl or neomenthyl) and related compounds
Paul Beagley, Philip Davies, Harry Adams, and Colin White
Abstract: The syntheses of the chiral ansa-metallocene complexes TiCl2(η5:η5-C5Me4SiMe2C5H3R*) (R* = menthyl (4a)
or neomenthyl (4b)) are reported; initially 4a was obtained as a 3:1 mixture of (R:S) diastereoisomers, which differ
only in which face of the asymmetrically substituted cyclopentadienyl ring is bonded to the titanium (chiral descriptor
shown). The major diastereomer 4aR was crystallized out optically pure from the initial reaction mixture, whereas the
4aS diastereoisomer was isolated after isomerizing a racemic mixture of 4a to a 1:3 mixture of (R:S) diastereoisomers
using UV irradiation. The corresponding neomenthyl complex 4b was obtained as a 1.3:1 mixture of diastereoisomers
that could not be separated. The optically pure 4aR was converted stereoselectively into the corresponding (R)-
TiMe2(η5:η5-C5Me4SiMe2C5H3R*) (R* = menthyl). Syntheses of the related indenyl ligand system C5Me4SiMe2C9H7 (3)
is reported but complexation to titanium proved to be problematic although ZrCl2(η5:η5-C5Me4SiMe2C9H6) (6) was iso-
lated. The crystal structures of both 4aR and 4aS are reported and compared. Crystals of 4aR are orthorhombic, a =
27.857(11), b = 9.985(5), and c = 9.596(4) Å, Z = 4, space group P212121 (D42 , No. 19), and those of 4aS are
monoclinic, a = 8.5810(10), b = 38.679(4), and c = 8.5842(10) Å, β = 113.001(2)°, Z = 4, space group P21 (C22 ,
No. 4). The structures were solved by the Patterson method and 4aR was refined by blocked-cascade least-squares pro-
cedures to R = 0.0628 (Rω = 0.0503) for 902 reflections with |F|/σ(|F|) ≥ 3.0, whereas 4aS was refined by full-matrix
least-squares procedures to R = 0.0646 (wR2 = 0.1829) for 5734 reflections with |F|/σ(|F|) ≥ 4.0. Both diastereomers of
4a catalyze hydrosilylation of ketones, but as expected from a comparison of the two crystal structures, the 4aR isomer
is the more stereoselective catalyst, i.e., hydrosilylation of acetophenone followed by hydrolysis gives 82%
enantiomeric excess (ee) of (S)-PhCH(Me)OH with 4aR whereas only 16% ee of (R)-PhCH(Me)OH with 4aS.
Key words: titanium, metallocene, chiral, structure, catalyst. 741
Résumé : On a réalisé la synthèse des complexes ansa-métallocènes chiraux TiCl2(η5:η5-C5Me4SiMe2C5H3R*) (R* =
menthyle (4a); R* = néomenthyle (4b)). Originalement 4a a été obtenu sous la forme d’un mélange 3:1 des diastéréo-
mères (R) et (S) qui ne diffèrent que par quelle face du noyau cyclopentadiényle substitué de façon asymétrique est liée
au titane (on explicite le descripteur chiral). Le diastéréomère principal, 4aR a été retiré optiquement pur par cristallisa-
tion du mélange réactionnel initial; le diastéréomère 4aS n’a été isolé qu’après avoir isomérisé un mélange racémique
de 4a en un mélange 1:3 des diastéréomères (R) et (S) par irradiation UV. Le complexe néomenthylique correspondant
4b a été obtenu sous la forme d’un mélange 1:4:1 de diastéréomères qui n’a pas pu être séparé. Le produit 4aR a été
optiquement pur a été transformé stéréosélectivement en (R)-TiMe2(η5:η5-C5Me4SiMe2C5H3R*) (R* = menthyle) corres-
pondant. On rapporte aussi la synthèse du système de ligand indényle apparenté C5Me4SiMe2C9H7 (3), mais sa com-
plexation avec le titane s’est avérée être problématique même si on a pu isoler le ZrCl2(η5:η5-C5Me4SiMe2C9H6) (6).
On a déterminé les structures cristallines des composés 4aR et 4aS et on en a fait une comparaison. Les cristaux de 4aR
sont orthorhombiques, groupe d’espace P212121 (D42 , No. 19), avec a = 27,857(11), b = 9,985(5) et c = 9,596(4) Å et
Z = 4 alors que ceux de 4aS sont monocliniques, groupe d’espace P21 (C22 , No. 4), avec a = 8,5810(10), b = 38,679(4)
et c = 8,5842(10) Å, β = 113,001(2)° et Z = 4. Les structures ont été résolues par la méthode de Patterson et celle de
4aR a été affinée par des méthodes de moindres carrés à cascade bloquée jusqu’à des valeurs de R = 0,0628 (Rω
=
0,0503) pour 902 réflexions avec |F|/σ|F| ≥ 3,0 alors que celle de 4aS a été affinée par la méthode des moindres carrés
(matrice complète) jusqu’à des valeurs de R = 0,0646 (wR2 = 0,1829) pour 5734 réflexions avec |F|/σ|F| ≥ 4,0. Les
deux diastéréomères de 4a catalyzent l’hydrosilylation des cétones mais, comme on peut s’y attendre sur la base d’une
comparaison des deux structures cristallines, l’isomères 4aR est plus stéréosélectif; par exemple l’hydrosilylation de
l’acétophénone, suivie d’une hydrolyse, conduit à un mélange avec 82% d’ee du (S)-PhCH(Me)OH avec 4aR alors que
l’on n’obtient que 16% d’ee de (R)-PhCH(Me)OH avec 4aS.
Received September 12, 2000. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on June 30, 2001.
Dedicated to Professor Brian James on the occasion of his 65th birthday.
P. Beagley, P. Davies, H. Adams, and C. White.1 Department of Chemistry, The University Sheffield S3 7HF, England.
1Corresponding author (telephone: +44 (0)114-222-9310; fax: +44 (0)114-273-8673; e-mail: Colin.white@sheffield.ac.uk).
Can. J. Chem. 79: 731–741 (2001)
© 2001 NRC Canada
DOI: 10.1139/cjc-79-5/6-731

732
Can. J. Chem. Vol. 79, 2001
Mots clés : titane, métallocène, chiral, structure, catalyseur.
[Traduit par la Rédaction] Beagley et al
Introduction
Chiral metallocenes have attracted considerable attention
in the last decade (1) fueled by their spectacular successes in
catalyzing stereoselectively a wide range of organic transfor-
mations (2), including the stereospecific polymerization of
α-olefins (3). We were particularly impressed by the report
of Marks and co-workers (4) that lanthanide complexes, con-
taining a chiral ansa-bis(cyclopentadienyl)2 ligand (1), cata-
The corresponding indenyl analogue (3) was synthesized us-
ing similar procedures (Scheme 1).
Metal complexes
Complexation of ligands 2a and 2b to titanium was car-
ried out as shown in Scheme 1. The dilithium salt of the ap-
propriate ligand precursor was formed via double
deprotonation with two equiv of n-butyllithium in THF and
was then reacted in situ with one equiv of TiCl3·3THF in
refluxing THF for 12 h. The resulting Ti(III) complex was
then oxidized to the desired Ti(IV) complex by stirring with
concentrated hydrochloric acid, open to the air, and worked
up, followed by removal of solvent to give the crude prod-
ucts 4a and 4b as red solids. The ligands 2a and 2b are
chiral by virtue of the menthyl or neomenthyl substituent,
but in addition, the two faces of the asymmetrically substi-
tuted cyclopentadiene ring are different and, therefore,
complexation to titanium introduces an element of planar
chirality into the system. Hence, the titanium complexes 4a
and 4b are obtained as a mixture of diastereoisomers. This is
lyzed the hydrogenation of nonfunctionalized alkenes in up
to 96% enantiomeric excess (ee). To put this in context, the
best result that we know of for the hydrogenation of a non-
functionalized alkene using a rhodium(I)–chiral bis(phos-
phine) catalyst is 48% ee (5), and results of 0–20% ee are
typical even with well-known phosphines such as Diop (6)
and Binap (7). Organolanthanide complexes are, however,
notoriously moisture-sensitive, and synthesizing and handling
them requires special techniques. We, therefore, wondered if
the chiral ansa-bis(cyclopentadienyl) ligand developed by
Marks and co-workers (4), and related ligands, could be
complexed to less moisture-sensitive transition metals and
still give impressive results in enantioselective synthesis. We
report herein our studies involving the syntheses of the cor-
responding titanium complexes. Our choice of titanium was
influenced by the fact that two other research groups had re-
ported impressive results for the hydrogenation of
nonfunctionalized alkenes using titanocene systems, but with
C2-symmetric chiral ligands (8). We should also point out
that, after the start of this work, hydrogenation of non-
functionalized alkenes with ee’s of up to 98% were reported
using an iridium – chiral phosphanodihydrooxazole system
(9) and also using a cationic zirconocene catalyst (10).
1
readily apparent from the H NMR spectra; the asymmetry
of the complex renders all four methyl groups on the
tetramethyl-substituted cyclopentadienyl ring chemically
inequivalent and the four methyl signals corresponding to
each diastereoisomer are readily assigned from their relative
intensity. Thus, the diastereoisomeric ratio can be easily
measured via integration of a major and its corresponding
minor peak. In this way the diastereoisomeric ratios were
found to be 3.0:1 for 4a and 1.3:1 for 4b, i.e., 75 and 57%
diastereoselectivities, respectively. In contrast, Marks and
co-workers (4) found that both these ligands complexed to
lanthanides with much higher diastereoselection (>80%) un-
der similar reaction conditions. In both cases ligand
complexation takes place to the metal in the +3 oxidation
state, and we had expected that complexation to the smaller
titanium atom would have led to an increase in the diastereo-
selection compared to that observed for the lanthanides. We
can only presume that as a result of solvent coordination the
lanthanides are more sterically congested and hence more
stereo-differentiating.
In the case of the menthyl derivative 4a, which was
formed in a diastereomeric ratio of 3:1, cooling a hexane so-
lution to –60°C resulted in the selective crystallization of the
major isomer, presumed to be the kinetically favoured iso-
mer. By X-ray crystallography (see below) we were able to
show that this was the 4aR diastereomer where the R chiral
descriptor refers to the planar chirality of the asymmetric
cyclopentadienyl ring. There are several reports of photo-
induced racemate–meso interconversions of group IV ansa-
metallocene dichlorides, which it is generally agreed, pro-
ceed via photolytic homolytic cleavage of a Ti-(η5-cyclo-
pentadienyl) bond, rotation of the cyclopentadienyl
radical about the linking-Cp bond, and recombination of the
Ti(III) centre with the opposite face of the cyclopentadienyl
Results and discussion
Ligand syntheses
The ansa-cyclopentadiene ligands consisting of a 2,3,4,5-
tetramethylcyclopentadienyl ring linked via a dimethylsilyl
bridge to a menthyl- (2a) or neomenthyl- (2b) cyclopenta-
dienyl moiety were synthesized by a modification of the pro-
cedure reported by Marks and co-workers (4) (Scheme 1). In
our experience, the key to successful synthesis of these lig-
ands is to use freshly prepared 1-(chlorodimethylsilyl)-
2,3,4,5-tetramethylcyclopentadiene; if this is not possible
then this reagent should be distilled immediately before use.
2 Prefix ansa Latin for bent handle, attached at both ends.
© 2001 NRC Canada

Beagley et al
733
Scheme 1.
ligand (11). We were, therefore, interested to see the effect
of photolysis upon these chiral titanium complexes.
Photolyzing the mother liquor left from the recrystallization
of the 4aR diastereomer, which now consisted of an approxi-
mately equal mixture of diastereomers, with UV radiation at
–30°C slowly led to a change in this isomeric ratio with
preferential formation of the most thermodynamically stable
product, i.e., the 4aS diastereomer, which was initially the
minor isomer. Thus, after irradiating for 1 week, the solution
contained a 1:3 ratio of 4aR:4aS, and upon cooling to
–60°C the 4aS diastereomer selectively crystallized out.
Hence, both diastereomers could be obtained and the crystal
structures of both were determined.
In contrast, the separation of the two diastereoisomers of
the neomenthyl complex 4b by recrystallization was not pos-
sible because of the high solubility of both diastereoisomers
in hydrocarbon solvents. For example, no crystals were
formed even after cooling a hot, saturated hexane solution of
4b to –30°C for a number of days. Attempts to significantly
increase the diastereomeric ratio by UV irradiation were also
unsuccessful. Thus, the initial diastereomeric ratio of 1.3:1
changed to only 1:1.4 after UV irradiation for 1 week at
© 2001 NRC Canada

734
Can. J. Chem. Vol. 79, 2001
Fig. 1. Molecular structure of TiCl2{η5:η5-C5Me4SiMe2C5H3(menthyl)},
(R)-diastereoisomer 4aR. Hydrogen atoms are omitted for clarity.
–30°C. Further irradiation had no effect upon this ratio.
Crystallization from pentane was attempted, but no single
diastereoisomer was isolated by this method.
The use of silanized silica as a stationary phase in liquid
chromatography has proved successful for the purification of
ansa-bis(η5-cyclopentadienyl)dichlorotitanium complexes (12).
Therefore, silica gel was modified via treatment with chloro-
trimethylsilane (see Experimental), and the resulting silanized
silica gel was used as a stationary phase for the attempted
separation of the two diastereoisomers of 4b. Unfortunately,
no separation of the two diastereoisomers occurred, although
there were no visible signs of decomposition either. Attempts
to separate the diastereoisomers by converting the dichloride
complex into a less soluble derivative by reaction with either
1,2-benzenedimethanol or (R)-(–)-acetylmandelic acid failed
to yield stable characterizable products.
From the diastereoselectivity observed in the initial for-
mation of the titanium complexes it is clear that the differ-
ence in free energy between the two menthyl diastereomers
is greater than the difference in free energy between the two
neomenthyl diastereomers. The initial diastereomeric ratio is
determined by kinetic control but under UV irradiation the
diastereomers are in equilibrium and the ratio is determined
by their relative thermodynamic stabilities. Hence, it is pos-
sible to estimate the difference in free energy between the
two diastereoisomers of the menthyl substituted complex
(4aR and 4aS) at –30°C as 2.2 kJ mol–1 and that of the
neomenthyl diastereomers as 0.7 kJ mol–1.
failed to yield a stable complex. More successful was the
synthesis of the corresponding zirconium indenyl complex
ZrCl2(η5:η5-C5Me4SiMe2C9H6) (6), although this too proved
to be unstable in solution.
The NMR spectra of the titanium complexes 4a and 4b
are worthy of comment; the three cyclopentadienyl protons
Comparison of the X-ray crystal structures of the
diastereoisomers 4aR and 4aS
are grouped together in
a similar manner in both
diastereoisomers of both catalysts with one proton exhibiting
a greater degree of deshielding than the other two protons
adjacent to the silicon bridge. This proton sits directly over
one of the chloride ligands, and the magnetic anisotropy of
the halogen atom has been ascribed as the origin of the low-
field shift of such protons (13).
Crystallographic data are presented in Table 1 and atomic
cooordinates are listed in Table 2.3 The crystal stucture of
the menthyl complex with R planar chirality (4aR) is shown
in Fig. 1; the corresponding diastereoisomer with S planar
chirality crystallized as two independent molecules per unit
cell and these are shown as 4aS and 4aS′ in Figs. 2a and b,
respectively. As expected, the bond distances in all the
diastereoisomers are essentially similar taking into account
that the face of the menthyl-Cp is reversed on going from
4aR to 4aS and 4aS′ (Table 3). In all the structures each
cyclopentadienyl ring is asymmetrically pentahapto bonded
to the titanium (Ti—C in the range 2.35–2.59 Å), with the
greater asymmetry for the menthylated ring in which the
menthylated carbon C(8) is always more distant from the ti-
tanium than any other ring carbon. All these Ti—C distances
are within the range reported for similar titanium complexes
(17). In each complex both five-membered rings are planar
with the four methyl groups of the tetramethylcyclopenta-
dienyl ring bent away from the titanium by up to 0.24 Å
from the mean plane of the cyclopentadienyl ring. A conse-
quence of the short bridging unit and the bulky menthyl
substituent is that the angle between the planes of the two
Cp rings (i.e., the bite angle) is relatively large at 58.0 and
59.2° for 4aR and 4aS, respectively, (Table 4) compared to
The optically pure dichloro-complex 4aR was reacted with
methylmagnesium bromide to give the corresponding (R)-
TiMe2(η5:η5-C5Me4SiMe2C5H3R*) (R* = menthyl) (5R). We
were surprised and disappointed to find that this compound
was rather unstable, like the corresponding Ti(C5H5)2Me2
(14), as compared to the more robust Ti{(C5H4)2SiMe2}Me2
(15). Similarly, although synthesis of the related indenyl
ligand system C5HMe4SiMe2C9H7 (3) proved to be relatively
straightforward, complexation of this ligand was not. The re-
action of TiCl3·3THF with the dilithium salt of 3 using simi-
lar reaction conditions as those successfully used for the
syntheses of the titanium complexes 4a and 4b repeatedly
failed to yield TiCl2(η5:η5-C5Me4SiMe2C9H7). Titanium
indenyl complexes are notoriously difficult to synthesize or
are unstable and, therefore, these are often prepared and re-
duced in situ to the corresponding more stable tetrahydro-
indenyl titanium complexes using hydrogen in the presence
of a PtO2 catalyst (12a, 16); unfortunately, even this strategy
3 Supplementary material may be purchased from the Depository of Unpublished Data, Document Delivery, CISTI, National Research
Council Canada, Ottawa, ON K1A 0S2, Canada. For information on obtaining material electronically go to
http://www.nrc.ca/cisti/irm/unpub_e.shtml. Crystallographic information has been deposited with the Cambridge Crystallographic Data Cen-
tre. Copies of the data can be obtained, free of charge, on application to the Director, CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K.
(fax: 44-1223-336033 or e-mail: deposit@ccdc.cam.ac.uk).
© 2001 NRC Canada

Beagley et al
735
Table 1. Crystallographic data for diffraction studies of 4aR and 4aS.
4aR
4aS
Crystal parameters
Empirical formula
Formula weight
Color
C26H40Cl2SiTi
499.49
Red
C26H40Cl2SiTi
499.49
Red
Solvent
Crystal size (mm)
Habit
Crystal system
Space group
a (Å)
b (Å)
c (Å)
β (°)
V (Å3)
n-Pentane
n-Pentane
0.07 × 0.04 × 0.03
Red plates
Monoclinic
P21 (C22 , No. 4)
8.5810(10)
38.679(4)
8.5842(10)
113.001(2)
2622.6(5)
4
0.80 × 0.20 × 0.05
Elongated thin plates
Orthorhombic
P212121 (D42 , No. 19)
27.857(11)
9.985(5)
9.596(4)
2669.1(18)
4
1.243
5.73
Z
Dcalcd (Mg m–3)
µ (cm–1)
F(000)
1.265
5.87
1064
1064
Intensity data
Diffractometer
λ (Mo Kα radiation) (Å)
Monochromator
Reflections measured
2θ range (°)
Temperature (K)
Stoe Stadi-2 2-circle
0.71073
Graphite, incident beam
+h, +k, ±l
6.5–50
293
Nicolet R3 4-circle
0.71073
Graphite, incident beam
+h, +k, ±l
2.11–23.32
150(2)
Scan type
ω
ω
Standard reflections
No. of reflections collected
No. of reflections used
Acceptance criterion
Absorption coefficient (mm–1)
Transmissionn coefficient (min)
Transmissn coefficient (max)
Absorption correction
2 every 50 reflections
2701
902
|F|/σ(|F|) > 3.0
0.573
0.882
0.972
Gaussian integration
3 every 100 reflections
11784
5734
|F|/σ(|F|) > 4.0
0.587
0.9601
0.9826
Semi-empirical
Structure solution
Method
Programs
Scattering factors
Patterson–Fourier
SHELXTL (25)
Ref. 26
Patterson–Fourier
SHELXTL (25)
Ref. 26
Refinement method
R
Blocked-cascade least-squares
0.0628
Full-matrix least-squares on F2
0.0646
wR2
0.1829
Rω
0.0503 (266 parameters)
Weighting scheme
Absolute structure parameter
Largest diff. peak and hole (e Å–3)
H refinement
ω–1 = [σ2(F) + 0.00021(F)2]
ω–1 = [σ2(F02 )+ (0.1148P)2 + 0.3205P]
0.00(5)
+0.494 and –0.536
Riding mode
+0.30 and –0.37
Riding mode
other ansa-dichlorotitanocenes; for example, [TiCl2{(η5:η5-
C5H4)2SiMe2}] has a bite angle of 56.2° (17a).
occupying the top face of the complex it can be seen that the
bottom face of 4aR is more sterically crowded because the i-Pr
group of the menthyl substituent points across this face,
whereas in 4aS the menthyl group is orientated such that the
i-Pr group points away from this face. This steric crowding
presumably accounts for the smaller Cl-Ti-Cl bond angle of
95.3(2)° found in 4aR compared to that of 97.75(8)° found in
4aS. If the structures of the diastereoisomers are similar in
solution this would imply that the diastereoisomer 4aR
The Cl(1)-Ti-C(11) and Cl(1)-Ti-C(16) bond angles are
131(4) and 131.1(3)°, respectively, for 4aR, whereas the
coresponding angles for 4aS are 124.1(2) and 123.11(19)°,
showing the different relative orientations of the ligands
about the titanium. However, the differences between the
two diastereoisomers are best viewed by comparison of the
space-filling models (Fig. 3). With the bulky C5Me5 ligand
© 2001 NRC Canada

736
Can. J. Chem. Vol. 79, 2001
Table 2. Atomic coordinates (1 x 104) and equivalent isotropic
displacement parameters (Ueq) (1 x 103) for the two
Table 2 (concluded).
Atom
x
y
z
Ueq (D2)
diastereoisomers of [TiCl2(η5:η5-C5Me4SiMe2C5H3(menthyl)].
C(16)
C(17)
562(8)
603(8)
3884(2)
3993(2)
4167(2)
4156(2)
3979(2)
3967(2)
4362(2)
4333(3)
3929(3)
3492(3)
3541(3)
1527(3)
1631(5)
1665(5)
1335(3)
1215(2)
1168(3)
822(2)
–1089(8)
540(8)
35(2)
33(2)
35(2)
35(2)
36(2)
40(2)
47(2)
52(2)
55(3)
68(3)
55(2)
74(3)
111(6)
114(6)
76(3)
50(2)
55(2)
37(2)
41(2)
43(2)
43(2)
40(2)
74(3)
108(5)
125(6)
80(3)
37(2)
40(2)
36(2)
35(2)
33(2)
53(2)
54(2)
50(2)
42(2)
67(3)
56(2)
Atom
x
y
z
Ueq (D2)
C(18)
C(19)
C(20)
C(21)
C(22)
C(23)
C(24)
C(25)
2167(9)
3139(9)
2176(9)
–766(9)
2652(10)
4776(9)
2762(10)
–967(12)
1437(9)
418(8)
4aR
Ti(1)
Cl(1)
Cl(2)
Si(1)
C(1)
C(2)
C(3)
C(4)
C(5)
C(6)
C(7)
C(8)
C(9)
C(10)
C(11)
C(12)
C(13)
C(14)
C(15)
C(16)
C(17)
C(18)
C(19)
C(20)
C(21)
C(22)
C(23)
C(24)
C(25)
C(26)
4aS
3419(1)
3908(2)
2932(2)
3200(1)
5008
5498
5746
5455
4955
4689
4054
4207
3814
3467
3604
4767
4680(5)
5060(4)
5701(4)
3016(4)
3327(4)
3153(6)
2739(5)
2649(4)
3740(5)
3335(6)
2415(5)
2224(4)
3512(4)
2723(5)
3009(3)
1180(4)
1940(6)
6117(5)
4031(3)
4427(5)
2376(4)
4854(4)
53(1)*
92(2)*
113(2)*
55(2)*
54(6)*
66(7)*
64(7)*
53(7)*
70(7)*
48(5)*
42(6)*
44(6)*
52(4)
57(6)*
46(6)*
53(6)*
68(6)*
67(7)*
82(8)*
39(5)*
39(6)*
48(6)*
58(7)*
62(6)*
68(7)*
84(7)*
106(9)*
88(8)*
77(7)*
89(7)*
–1085(8)
1194(9)
3067(9)
852(10)
–2523(8)
–4605(11)
–2472(11)
1498(12)
1859(14)
3752(15)
4747(14)
4297(9)
2392(10)
503(9)
3654(15)
3008(17)
3136(18)
2532(14)
3225(16)
3053(15)
4672(14)
3773(16)
3754(14)
4674(14)
5277(14)
3519(12)
2087(14)
4259(14)
2694(15)
4455(13)
3600(15)
2341(15)
2308(14)
3617(15)
4040(13)
1159(13)
1045(15)
4044(17)
7278(13)
7099(16)
1535(11)
1617(15)
3004(15)
4157(14)
4182(15)
2756(13)
3928(16)
2851(14)
1921(13)
2303(13)
3590(13)
135(12)
–352(12)
–976(13)
5549(12)
5595(12)
6376(14)
6397(14)
5508(15)
5083(13)
7288(14)
7299(13)
5306(16)
4236(15)
6011(15)
4014(14)
C(26)
–3064(9)
3246(14)
5067(15)
6121(16)
6078(13)
4279(10)
3292(12)
651(9)
1600(10)
610(10)
–947(10)
–990(9)
2086(15)
2710(17)
1690(20)
7045(14)
–1075(9)
–1092(9)
409(10)
C(1A)
C(2A)
C(3A)
C(4A)
C(5A)
C(6A)
C(7A)
C(8A)
C(9A)
C(10A)
C(11A)
C(12A)
C(13A)
C(14A)
C(15A)
C(16A)
C(17A)
C(18A)
C(19A)
C(20A)
C(21A)
C(22A)
C(23A)
C(24A)
C(25A)
C(26A)
1014(2)
1023(2)
854(2)
2000(9)
2980(10)
2066(9)
469(9)
741(2)
1538(3)
1338(4)
1917(5)
1394(3)
52(2)
–379(11)
–1521(13)
–1034(16)
6664(13)
564(8)
–40(2)
2184(9)
3102(8)
2159(9)
600(8)
–216(2)
–230(2)
–61(2)
1424(9)
534(9)
–2517(10)
824(11)
3032(10)
1214(10)
–4631(10)
–2454(11)
13(3)
2773(10)
4798(10)
2666(11)
–774(9)
–971(12)
–3057(9)
–391(3)
–426(2)
–30(2)
Ti(1)
Ti(1A)
Si(1)
Si(1A)
Cl(1)
Cl(1A)
Cl(2)
Cl(2A)
C(1)
C(2)
C(3)
C(4)
C(5)
C(6)
C(7)
C(8)
C(9)
C(10)
C(11)
C(12)
C(13)
C(14)
C(15)
2610(1)
1003(1)
–864(2)
–2409(2)
3090(3)
3853(2)
5432(2)
1311(2)
1476(13)
1848(14)
3785(16)
4743(15)
4280(10)
2408(11)
517(10)
3540(1)
398(1)
3538(1)
401(1)
3568(1)
370(1)
3511(1)
426(1)
2413(3)
2306(5)
2274(5)
2594(3)
2720(3)
2769(2)
3117(2)
2924(2)
2912(2)
3082(2)
3196(2)
2399(3)
2602(4)
2023(5)
2538(3)
1003(1)
2610(1)
32(1)
31(1)
41(1)
40(1)
48(1)
48(1)
51(1)
51(1)
71(3)
114(6)
112(5)
75(3)
54(2)
50(2)
39(2)
39(2)
42(2)
41(2)
39(2)
76(3)
101(5)
124(6)
75(3)
436(3)
396(3)
–2404(2)
–866(2)
3851(2)
3082(2)
1308(2)
5433(2)
3246(13)
5067(16)
6124(16)
6109(12)
4279(11)
3333(11)
656(9)
1585(9)
620(10)
–950(9)
–998(9)
2089(14)
2782(17)
1700(20)
7040(13)
aUeq is defined as one third of the trace of the orthogonalized Uij tensor.
should be a more stereoselective catalyst. Preliminary results
from hydrosilylation experiments (Table 5) are in keeping
with this statement (18).
Experimental
All reactions of moisture sensitive reagents were per-
formed under nitrogen. THF was heated under reflux over
sodium benzophenone ketyl and distilled under nitrogen.
Hexane, toluene, and diethyl ether were heated under reflux
over sodium and freshly distilled under nitrogen prior to use.
Menthyltosylate [(1R, 2S, 5R)-(–)-2-isopropyl-5-methyl-
cyclohexyl tosylate] and neomenthyltosylate [(1S, 2S, 5R)-
(+)-2-isopropyl-5-methylcyclohexyl tosylate] were prepared
by modifications of the literature procedures (19, 20) as we
have previously described (21). (+)-Neomenthyl tosylate is
thermally unstable, therefore, it is best stored at –10°C.
1992(9)
2964(10)
2068(10)
482(9)
–399(12)
–1511(14)
–1011(17)
6647(13)
© 2001 NRC Canada

Beagley et al
737
Table 3. Selected bond lengths (Å) with estimated standard devi-
ations for the two diastereoisomers of [TiCl2(η5:η5-
C5Me4SiMe2C5H3(menthyl)].
Table 4. Bond angles (°) with estimated standard deviations for
the two diastereoisomers of [TiCl2(η5:η5-C5Me4SiMe2C5H3(menthyl)].
4aR 4aS 4aS′
95.3(2) 97.75(8) 97.86(8)
4aR
4aS
4aS′
Cl(1)-Ti(1)-C1(2)
Cl(1)-Ti(1)-C(8)
Cl(2)-Ti(1)-C(8)
Cl(1)-Ti(1)-C(11)
Cl(2)-Ti(1)-C(11)
Cl(1)-Ti(1)-C(16)
Cl(2)-Ti(1)-C(16)
C(11)-Ti(1)-C(16)
C(11)-Si(1)-C(16)
C(6)-C(8)-C(7)
Ti(1)—Cl(1)
Ti(1)—Cl(2)
Ti(1)—C(7)
Ti(1)—C(8)
Ti(1)—C(9)
Ti(1)—C(10)
Ti(1)—C(11)
Ti(1)—C(16)
Ti(1)—C(17)
Ti(1)—C(18)
Ti(1)—C(19)
Ti(1)—C(20)
Si(1)—C(11)
Si(1)—C(16)
C(7)—C(8)
2.310(5)
2.346(5)
2.3171(18)
2.3349(19)
2.359(8)
2.530(9)
2.484(8)
2.358(7)
2.366(7)
2.370(7)
2.381(7)
2.508(8)
2.516(8)
2.392(7)
1.856(8)
1.866(9)
1.416(11)
1.441(10)
1.387(11)
1.424(11)
1.417(11)
1.448(10)
1.432(10)
1.428(11)
1.426(10)
1.409(11)
2.3201(19)
2.3352(17)
2.347(8)
2.535(9)
2.477(9)
2.370(8)
2.367(7)
2.367(7)
2.391(8)
2.500(8)
2.509(8)
2.398(7)
1.852(9)
1.882(8)
1.433(11)
1.432(11)
1.410(11)
1.418(11)
1.425(10)
1.441(10)
1.437(11)
1.397(11)
1.402(10)
1.416(11)
78.7(4)
109.2(3)
131.3(4)
116.2(4)
131.1(3)
115.6(3)
68.4(4)
78.52(18)
102.76(19)
124.1(2)
121.8(2)
123.11(19)
122.77(18)
68.4(3)
78.26(19)
102.77(17)
123.7(2)
122.3(2)
122.64(19)
122.89(18)
68.5(3)
2.428(14)
2.585(14)
2.421(13)
2.352(13)
2.360(14)
2.366(12)
2.340(14)
2.479(14)
2.468(14)
2.448(13)
1.855(14)
1.877(14)
1.434(21)
1.431(19)
1.413(19)
1.383(19)
1.426(18)
1.430(18)
1.409(18)
1.347(22)
1.435(21)
1.393(20)
90.8(6)
91.3(3)
91.1(3)
126.8(12)
129.7(12)
125.2(7)
127.6(7)
125.2(7)
128.3(7)
C(6)-C(8)-C(9)
Symmetry transformations used to generate equivalent atoms:
Fig. 2. Molecular structure of the two independent molecules of
TiCl2{η5:η5-C5Me4SiMe2C5H3(menthyl)}, (S)-diastereoisomer 4aS.
Hydrogen atoms are omitted for clarity.
C(7)—C(11)
C(8)—C(9)
C(9)—C(10)
C(10)—C(11)
C(16)—C(17)
C(16)—C(20)
C(17)—C(18)
C(18)—C(19)
C(19)—C(20)
Trichlorotris(tetrahydrofuran)titanium was synthesized by
the procedure reported by Manzer (22).
Preparation of silanized silica gel
Dichloromethane (1500 cm3) was added to silica gel
(150 g) and the resulting slurry rapidly stirred by an over-
head stirrer. Chlorotrimethylsilane (150 cm3) was added and
the slurry stirred for a further 1 h. The slurry was washed
with saturated sodium hydrogencarbonate solution (100 cm3)
and then with water until the washings were neutral. The re-
sulting silanized silica gel was washed with acetone and
dried in an oven at 100°C overnight.
Menthylcyclopentadiene
This was prepared by the following modification of the
literature procedure (23). NaH (9.6 g, 0.4 mol, i.e., 19.2 g
50% dispersion) was washed with hexane, and freshly dis-
tilled THF (150 cm3) was added. Freshly cracked
cyclopentadiene (26 g, 0.4 mol) in THF (50 cm3) was added
slowly over a 2 h period and the mixture stirred until the
bubbling stopped and the NaH was consumed (approxi-
mately 30 min). The resulting pale red solution of sodium
cyclopentadienide was added to a solution of (1S, 2S, 5R)-5-
methyl-2-(2-propyl)-cyclohexyl tosylate (63.4 g, 0.2 mol) in
THF (200 cm3) at 0°C via cannula. This mixture was heated
under reflux for 6 h producing a sodium tosylate precipitate.
Then the mixture was cooled, filtered, and the solvent re-
moved in vacuo. The brown viscous residue was redissolved
in diethyl ether and washed with water (5 × 100 cm3). After
drying the organic layer over Na2SO4, the solvent was
© 2001 NRC Canada

738
Can. J. Chem. Vol. 79, 2001
Table 5. Catalytic hydrosilylations of ketones in toluene.
Substrate
Catalyst 4aR
Catalyst 4aS
Temperature
(°C)
Conversion
(%)
Conversion
(%)
ee (%)a
ee (%)a
X = H
X = F
X = OMe
25
65
65
100b
100c
100c
82 (S)
66 (S)
65 (S)
100b
16 (R)
12 (R)
23 (R)
38c
100c
aAbsolute configurations of the product alcohols are shown in parentheses.
bAfter 24 h.
cAfter 1 h.
Fig. 3. The structures of the two diastereoisomers 4aR and 4aS
compared.
combined, washed with a saturated sodium carbonate solu-
tion (2 × 50 cm3), and dried over Na2SO4. Removal of the
solvent in vacuo gave a yellow liquid that was distilled to
produce the title product as colourless liquid (30–35°C,
0.5 mm Hg, 6.59 g, 37%). (EI) GC–MS m/z (%): 122 (M+,
1
60). H NMR (250 MHz, CDCl3) δ: 1.80 (br s, 6H, 2CH3),
1.95 (br s, 6H, 2CH3), 2.75 (br s, 2H, C5Me4H2).
Preparation of 1-(chlorodimethylsilyl)-2,3,4,5-
tetramethylcyclopentadiene
The literature procedure (4) was modified as follows.
1,2,3,4-Tetramethylcyclopentadiene (6 g, 0.05 mol) was dis-
solved in dry petroleum ether (bp 40–60°C) and a solution
of n-butyllithium (2.5 M, 20 cm3, 0.05 mol) was added at
0°C over 1 h. The mixture was stirred for 15 h at room temp
to produce a milky white suspension which was thinned by
the addition of THF (45 cm3). A solution of dichloro-
dimethylsilane (9.7 g, 0.075 mol) in THF (30 cm3) was
added at 0°C over 1 h and the mixture stirred for 15 h. The
yellow solution was filtered under nitrogen through Celite
previously dried at 100°C under vacuum to remove the lith-
ium chloride precipitate. The solvent was removed in vacuo
to give the crude product as a yellow-brown oil which was
distilled to give the title product as a yellow-green liquid;
bp0.1 60–65°C (lit. bp0.1 56–60°C), 7.05 g (67%). EI-MS m/z
(%): 214 (M+, 60), 121 ([M – SiMe2Cl]+), 97, 106 ([M –
SiMe2Cl –Me]+), 100). IR (neat) cm–1: 2961s, 2916s, 2861s,
1445m, 1380m, 1258s, 1058br, 844s, 811br, 656m. 1H NMR
(250 MHz, CDCl3) δ: 0.30 (s, 6H, 2Si-CH3), 1.9 (s, 6H,
2CH3), 2.06 (s, 6H, 2CH3) 3.06 (br s, 1H, Cp-H).
removed in vacuo to yield a brown oil which was distilled4
to yield the product as a colourless liquid (bp0.1 70°C,
1
3
23.84 g, 58%). H NMR (250 MHz, CDCl3) δ: 0.71 (d, JH,H
3
= 6 Hz, 3H, CH3), 0.84 (d, JH,H = 6 Hz, 3H, CH3), 0.88 (d,
3JH,H = 6 Hz, 3H, CH3), 0.95–1.84 (m, 9H, CH2’s and
CH’s), 2.31–2.48 (m, 1H, CH), 2.83–2.98 (m, 2H, allylic),
5.97–6.54 (m, 3H, vinylic). 95% pure by GC.
Preparation of neomenthylcyclopentadiene
An analagous procedure to that employed for the prepara-
tion of menthylcyclopentadiene but using menthyltosylate
yielded the title product as a colourless liquid (bp0.01 70°C,
1
42%, 95% pure by GC). H NMR (250 MHz, CDCl3) δ:
0.79–0.90 (m, 9H, CH3), 0.90–1.86 (m, 9H, CH2’s and CH’s),
2.90–3.12 (m, 3H, CH and allylic), 6.08–6.58 (m, 3H, vinylic).
Preparation of 1-(2,3,4,5)-tetramethylcyclopentadiene)-
1′-(3′(-1R, 2S, 5R)-5-methyl-2-(2-propyl)-
cyclohexylcyclopentadiene)dimethylsilane (2a)
The synthesis was based on the procedure reported by
Preparation of 1,2,3,4-tetramethylcyclopentadiene
The synthesis was based on the method reported by Marks
and co-workers (4). 2,3,4,5-Tetramethylcyclopent-2-enone
(24) (18.8 g, 0.13 mol) in diethyl ether (50 cm3) was added
over 1 h to a suspension of lithium aluminium hydride
(2.5 g, 0.07 mol) in diethyl ether (200 cm3) and the mixture
was stirred for 2 h. Aqueous hydrochloric acid (1.0 M, 300 cm3)
was added slowly and the mixture stirred for 16 h in air. The
organic layer was separated and the aqueous layer washed
with diethyl ether (3 × 100 cm3). Both organic layers were
Marks and co-workers (4).
A
solution of lithium
menthylcyclopentadienide was prepared by adding a solution
of n-butyllithium (2.5 M, 5.6 cm3, 0.014 mol) to a solution
of (1R, 2S, 5R)-5-methyl-2-(2-propyl)-cyclohexylcyclopenta-
diene (2.81 g, 0.014 mol) in THF (50 cm3) at 0°C and stir-
ring the mixture for 30 min at room temp. The solution was
then transferred via cannula at room temp to a solution of
freshly distilled 1-(chlorodimethylsilyl)-2,3,4,5-tetramethyl-
cyclopentadiene (3.8 g, 0.018 mol) in THF (25 cm3). The
4 The vacuum distillation must be performed at high vacuum since the product decomposes at higher temperatures and this is probably the
main reason for the low yields. A two stage distillation and washing the condenser after the CpH–(CpH)2 is distilled at 30°C is also recom-
mended to ensure no CpH contaminates the product.
© 2001 NRC Canada

Beagley et al
739
Isolation of (R)-dichloro[1-(η5-3′-(1R, 2S, 5R)-5-methyl-
mixture was heated under reflux for 12 h, cooled to room
temp, and filtered to remove the lithium chloride formed.
Solvent was removed in vacuo to yield a brown residue
which was redissolved in diethyl ether, washed with water
(3 × 10 cm3), dried over Na2SO4,and the solvent removed in
vacuo to yield a yellow-brown oil. Excess 1-(chlorodimeth-
ylsilyl)-2,3,4,5-tetramethylcyclopentadiene was removed by
distillation (heating to 80°C, 0.01 mm Hg) to yield the iso-
mers of the desired menthyl product 2a as a brown residue.
Yield 3.55 g (67%). EI-MS m/z (%): 382 (M+, 10), 261
([M – TMCp]+, 31), 204 ([M – TMCpSiMe2]+, 100). 1H
NMR (250 MHz, CDCl3) δ:–0.35–0.10 (m, 6H, Si-CH3),
2-(2-propyl)cyclohexyl-cyclopentadienyl)-1′-(η5-2,3,4,5-
tetramethylcyclopentadienyl)dimethylsilane]titanium (4aR)
The above oil was dissolved in the minimium of boiling
hexane and crystallized at –30°C, to yield the title product as
red crystals. Yield 0.75g (19%), mp 162–164°C. [α]20 (c
0.032, DCM): –62.5° (589 nm), –103.1° (578 nm), and –256.3°
(549 nm). EI-MS m/z (%): 498 (M+, 10%), 463 ([M – Cl]+,
1
100). H NMR (250 MHz, CDCl3) δ: 0.66 (m, 3H, CH3),
0.76 (m, 3H, CH3), 0.79 (m, 3H, CH3), 0.82 (m, 3H, CH3),
0.83 (m, 3H, CH3), 0.88–1.76 (m, 9H, CH’s, CH2’s), 1.84 (s,
3H, Cp-CH3), 1.87 (s, 3H, Cp-CH3), 2.00 (s, 3H, Cp-CH3),
3
3
4
0.62 (d, JH,H = 6 Hz, 3H, CH3), 0.75 (d, JH,H = 6 Hz, 3H,
2.10 (s, 3H, Cp-CH3), 2.69–2.81 (m, 1H, CH), 5.37 (t, JH,H
3
= 3 Hz, 1H, Cp-H), 5.67 (t, 3JH,H = 4JH,H = 3 Hz, 1H, Cp-H),
CH3), 0.80 (d, JH,H = 6 Hz, 3H, CH3), 0.85–1.60 (m, 9H,
3
4
6.50 (t, JH,H JH,H = 3 Hz, 1H, Cp-H). 13C NMR (63 MHz,
CDCl3) δ: 155.2, 145.7, 139.1, 137.4, 128.7, 103.4, 96.8 (Cp
quarterary), 126.7, 122.6, 114.3 (C1, C3, C4), 48.6 (C12),
41.7, 41.2 (C6, C7), 35.2, 32.3 (C10, C8), 27.0 (C13), 24.8
(C11) 22.8, 21.8, 16.7, 16.3, 16.2, 13.8, 12.8 (TMCp CH3’s
and menthyl CH3’s), 0.36, –1.5 (SiMe2). Anal calcd. for
C26H40Cl2SiTi: C 62.5, H 8.1, Cl 14.2; found: C 62.4, H 8.1,
Cl 14.1.
CH2’s), 1.60–1.75 (m, 6H, Cp-CH3), 1.78 (s, 3H, Cp-CH3),
1.90 (s, 3H, Cp-CH3), 2.29 (m, 1H, menthyl CH), 2.61–3.21
(m, 2H, allylic), 5.8–6.60 (m, 3H, Cp-H). 13C NMR
(63 MHz, CDCl3) δ: 151.9, 150.4, 145.5, 136.1, 133.5 (Cp
tertiary), 143.0, 142.0, 133.9, 138.5, 131.1, 131.0, 130.4,
127.3, 126.8, 126.6 (Cp quaternary), 47.4, 46.6, 41.9, 33.0,
27.9, 27.7, 22.6, 21.6, 15.3, 14.7, 14.2, 11.3, 11.1 (CH3’s
and CH’s), 43.7, 35.4, 24.5, (CH2’s), 1.0, –4.0 (Si-CH3).
Isolation of (S)-dichloro[1-(η5-3′-(1R, 2S, 5R)-5-methyl-2-
(2-propyl)cyclohexyl-cyclopentadienyl)-1′-(η5-2,3,4,5-
tetramethylcyclopentadienyl)dimethylsilane]titanium (4aS)
A 1:1 diastereomeric mixture (2 g) of (R)- and (S)-
dichloro{1-[η5-3′-(menthylcyclopentadienyl)-1′-[η5-2,3,4,5-
tetramethylcyclopentadienyl)dimethylsilane]}titanium was
dissolved in toluene (200 cm3), cooled to –30°C, and irradi-
ated with UV light using a Hanovair medium pressure
125 W lamp. After 7 d the solvent was removed in vacuo
and the mixture filtered through a silanized silica column,
eluting with 4% diethyl ether in pet ether (40–60 oC), result-
Preparation of 1-(2,3,4,5-tetramethylcyclopentadienyl)-
1′-(3′-(1S, 2S, 5R)-5-methyl-2-(2-propylcyclohexyl-
cyclopentadienyl)dimethylsilane (2b)
An analogous procedure to that employed for the preparation
of 1-(2,3,4,5-tetramethylcyclopentadienyl)-1′-[3′-menthylcyclo-
pentadienyl]dimethylsilane but employing neomenthylcyclo-
pentadiene yielded the desired neomenthyl product 2b as a
yellow oil (70%) found to be 91% pure by GC. (EI) GC–MS
m/z (%): 382 (M+, 12), 261 ([M – TMCp]+, 78), 179 ([M –
1
nmCp]+, 100). H NMR (250 MHz, CDCl3) δ:–0.05–0.33
(m, 6H, Si-CH3), 0.90–1.11 (m, 9H, neomenthyl CH3), 1.07–
1.85 (m, 9H, CH2’s and CH’s), 1.85–2.20 (m, 12H, Cp-
CH3’s), 3.00–3.40 (m, 3H, allylic, neomenthyl CH), 6.15
(br s, 1H, olefinic), 6.36–6.48 (m, 1H, olefinic), 6.61–6.72
(m, 1H, olefinic). 13C NMR (63 MHz, CDCl3) δ: 156.5,
147.7, 147.3, 142.6, 136.1, 135.2, 133.2, 132.4, 129.0, 119.7
(Cp’s), 48.0, 47.9, 37.8, 37.7 (CH’s), 42.4, 35.8, 26.1
(neomenthyl CH2’s), 30.2, 26.5, 22.8, 21.4, 21.3, 14.7, 11.3
(CH3’s), 1.1, –2.6 (SiMe2).
1
ing in a viscous red oil (1.1 g) which was shown by H
NMR spectroscopy to contain a S:R ratio of 3:1. The red oil
was dissolved in boiling pentane and cooled to –30°C; the ti-
tle compound crystallized as red-brown microcrystals which
were collected by suction filtration (520 mg, 26%). FAB-
1
MS m/z (%): 498 (M+, 24), 463 ([M – Cl]+, 100). H NMR
(250 MHz, CDCl3) δ: 0.70 (s, 3H, CH3), 0.72 (m, 3H, CH3),
0.81 (m, 3H, CH3), 0.83 (m, 3H, CH3), 0.86 (m, 3H, CH3),
0.85–1.60 (m, 8H, CH’s, CH2’s), 1.69 (s, 3H, Cp-CH3), 1.83
(s, 3H, Cp-CH3), 2.05 (s, 3H, Cp-CH3), 2.13 (s, 3H, Cp-
CH3), 2.25–2.88 (m, 1H, CH), 2.65–2.78 (m, 1H, CH), 5.05
Synthesis of (R/S)-dichloro[1-(η5-3′-(1R, 2S, 5R)-5-methyl-
2-(2-propyl)cyclohexyl-cyclopentadienyl)-1′-(η5-2,3,4,5-
tetramethylcyclopentadienyl)dimethylsilane]titanium
4
4
3
4
(t, JH,H = JH,H = 3 Hz, 1H, Cp-H), 5.47 (t, JH,H = JH,H
=
3
4
3 Hz, 1H, Cp-H), 6.81 (t, JH,H = JH,H = 3 Hz, 1H, Cp-H).
13C NMR (63 MHz, CDCl3) δ: 148.3, 145.3, 141.6, 132.7,
127.8, 101.8, 96.5 (Cp quaternary), 134.9, 117.7, 115.0 (C1,
C3, C4), 50.1 (C12), 41.8, 40.3 (C6, C7), 35.6 (C10), 32.7
(C8), 27.3 (C13), 24.9 (C11), 22.8, 21.7, 16.3, 15.7 (TMCp
CH3’s and menthyl CH3’s), 0.0, 1.0 (SiMe2). Anal. calcd. for
C26H40Cl2SiTi: C 62.5, H 8.1, Cl 14.2; found: C 62.8, H 8.3,
Cl 14.3.
(4aR/S
)
1-(2,3,4,5-Tetramethylcyclopentadiene)-1′-[3′-menthylcyclo-
pentadiene]dimethylsilane (3.02 g, 7.9 mmol) was dissolved
in THF (45 cm3). A solution of n-butyllithium (2.5 M,
6.3 cm3, 15.8 mmol) in hexanes was added at 0°C and the
mixture stirred at room temp for 30 min. The solution was
cooled to –78°C and TiCl3·3THF (2.9 g, 7.9 mmol) was
added rapidly against a stream of nitrogen and the mixture
was allowed to warm to room temp before heating under re-
flux for 6 h. After cooling to room temp, chloroform
(45 cm3) and concentrated aqueous hydrochloric acid (1 M,
30 cm3) were added and the mixture stirred at room temp in
air for 30 min. The organic layer was separated, dried, and
the solvent removed in vacuo to yield a red viscous oil
Preparation of (R/S)-dichloro[1-(η5-2,3,4,5-tetramethyl-
cyclopentadienyl)-1′-(η5-3′-(1S, 2S, 5R)-5-methyl-2-(2-propyl)-
cyclohexylcyclopentadienyl)dimethylsilane]titanium (4bR/S
)
1-(2,3,4,5-Tetramethylcyclopentadienyl)-1′-[3′-neomenthyl-
cyclopentadienyl]dimethylsilane (3.07 g, 8 mmol) was dis-
solved in THF (45 cm3). A solution of n-butyllithium (2.5 M,
1
which was shown by H NMR to be a 3:1 mixture of R:S.
© 2001 NRC Canada

740
Can. J. Chem. Vol. 79, 2001
6.4 cm3, 16 mmol) in hexanes was added at 0°C via syringe
and the mixture stirred at room temp for 30 min. The solu-
tion was cooled to –78 C and TiCl3·3THF (3.1 g, 8 mmol)
122.3, 119.8, 111.3 (C1, C3, C4), 50.7, 47.5 (Ti-CH3), 41.6,
35.6, 25.2 (neomenthyl CH2’s), 47.9, 41.2, 33.0, 27.5, 23.0,
22.0, 16.3, 15.6, 15.4, 12.2, 12.1 (CH’s, CH3’s), 1.35, –1.9
(Si-CH3).
o
was added rapidly against a stream of nitrogen; the mixture
was maintained at –78°C for 5 h then allowed to warm to
room temp and heated under reflux for a further 6 h. After
cooling to room temp, chloroform (45 cm3), together with
concentrated aqueous hydrochloric acid (30 cm3), was added
and the mixture stirred at room temp in air for 30 min. The
organic layer was separated, dried over Na2SO4, and the sol-
vent removed in vacuo to yield a red, viscous oil. This was
chromatographed on a silanized silica column eluting with
hexane to remove organic residues and then a 4% diethyl
ether – hexane mixture to elute a red band. Removal of sol-
vent in vacuo yielded the title product (2.74 g, 68%), which
was found to consist of a 1.3:1 mixture of (R) and (S)
diastereoisomers, mp 122–130°C. [α]20 (c 0.026, CH2Cl2):
+7.6° (589 nm), +30.8° (546 nm), and –61.5° (378 nm).
FAB-MS m/z (%): 499 ([M + 1]+, 20), 463 ([M – Cl]+ 100).
1H NMR (250 MHz, CDCl3) δ: 0.05 (s, 6H, Si-CH3), 0.79–
0.96 (m, 9H, neomenthyl CH3’s), 0.96–1.72 (m, 9H, neo-
menthyl CH’s, CH2’s), 1.77 (s, CH3 (minor)), 1.79 (s, CH3
(major)), 1.87 (s, CH3 (major)), 1.93 (s, CH3 (minor)), 2.09
(s, CH3 (minor)), 2.12 (s, CH3 (major)), 2.13 (s, CH3 (ma-
jor)), 2.18 (s, CH3 (minor)), 3.50–3.60 (m, CH (minor)),
3.63–3.71 (m, CH (major)), 5.20–5.29 (m, 1H, Cp-H), 5.53–
5.61 (m, 1H, Cp-H), 6.85–7.01 (m, 1H, Cp-H). 13C NMR
(63 MHz, CDCl3) δ: 150.1, 146.4, 144.9, 144.0, 142.0,
141.6, 134.0, 132.0, 129.8, 128.9, 102.5, 102.1, 97.5, 97.1
(Cp quaternary), 137.0, 132.1, 119.1, 118.0, 116.5, 115.7
(Cp tertiary), 51.1, 49.6, 37.9, 37.2, 29.4, 29.1, 28.7, 28.0,
22.8, 22.6, 22.3, 21.9, 20.8, 16.3, 16.2, 16.0, 13.8, 13.7,
13.6, 13.3, 13.2, 1.0 (CH’s, CH3’s), 40.2, 38.8, 35.8, 35.7,
26.0, 25.0 (neomenthyl CH2’s), 0.4, –0.3, –0.7, –1.0 (Si-
CH3’s). Anal. calcd. for C26H40Cl2SiTi: C 62.5, H 8.1;
found: C 61.2, H 8.2.
Synthesis of 1-indenyl-1′-(2,3,4,5-
tetramethylcyclopentadienyl)dimethylsilane (3)
Freshly distilled indene (48°C, 0.5 mm Hg, 4.3 cm3,
37 mmol) was dissolved in THF (100 cm3), cooled to 0°C,
and then n-butyllithiium (2.5 M, 14.8 cm3, 37 mmol) in hexanes
was added over 30 min. The resulting orange-red solution
was stirred at room temp for 2 h. 1-(Chlorodimethylsilyl)-
2,3,4,5-tetramethyl-cyclopentadiene (10.0 g, 46 mmol) was
dissolved in THF (75 cm3) and the indenide solution trans-
ferred to this solution via cannula. The reaction mixture was
heated to reflux for 18 h. The suspension was cooled to
room temp and transferred via cannula through Celite, under
nitrogen, to remove lithium chloride. The solvent was re-
moved in vacuo from the filtrate yielding a red-brown oil
which was redissolved in diethyl ether and washed with wa-
ter (3 × 20 cm3). The organic layer was dried over Na2SO4,
filtered, and the solvent removed in vacuo to give an oil,
which was heated to 80°C under 0.01 mm Hg to remove ex-
cess starting material and to leave the title product as an or-
ange oil (7.29 g, 67.0%; 98% pure by GC). Attempts to
separate isomers by recrystallization from pentane at –30°C
were unsuccessful. Similarly, chromatography on alumina,
eluting with pentane, failed to separate the isomers but did
yield a yellow oil. (EI) GC–MS m/z (%): major, 294 (M+,
33), 179 ([M – indenyl]+, 44), 173 ([M – TMCp]+, 100); mi-
nor, 294 (M+, 52), 179 ([M – indenyl]+, 49 ), 173 ([M –
TMCp]+, 100). 1H NMR (250 MHz, CDCl3) δ: –0.61,
−0.21, –0.05 (s, 12H, 2SiCH3 (major), 2SiCH3 (minor)), 1.59
(s, 6H, 2Cp′′CH3 (major)), 1.71 (s, 6H, 2Cp′′CH3 (major)),
1.82 (s, 6H, 2Cp′′CH3 (minor)), 1.89 (s, 6H, 2Cp′′CH3 (mi-
nor)), 2.86 (br s, 2H, Cp′′ipso-CH (major)), 3.05 (br s, 1H,
Cp′′ ipso-CH (minor)), 3.26 (d, J = 3 Hz, 2H, indenyl CH2
(minor)), 3.45 (d, J = 3 Hz, 1H, indenyl SiCH (major)), 6.39
(dd, J = 11, 3 Hz, 2H, indenyl olefinic CH (major)), 6.49 (t,
J = 3 Hz, 1H, indenyl olefinic CH (minor)), 6.69–6.75 (m,
2H, indenyl olefinic CH (major)), 6.95–7.15 (m, 6H, 2 ×
indenyl CH (major) and 2 × indenyl CH (minor)), 7.12–7.39
(m, 6H, 2 × indenyl CH (major) and 2 × indenyl CH (mi-
nor)). 13C NMR (32.44 MHz, CDCl3) δ: –5.9, -5.1, –3.7
(SiCH3), 11.1, 11.3 (Cp′′CH3 (major)), 14.7, 14.8 (Cp′′CH3
(minor)), 40.8 (indenyl CH2 (minor)), 44.4 (indenyl ipso-C
(major)), 54.2 (Cp′′ ipso-C (minor)), 54.6, (Cp′′ ipso-C (ma-
jor)), 120.8 (indenyl CH (major)), 122.2 (indenyl CH (mi-
nor)), 122.9 (indenyl CH (major)), 124.4 (indenyl CH
(minor)), 123.7 (indenyl CH (major)), 126.0 (indenyl CH
(minor)), 124.8 (indenyl CH (major)), 128.8 (indenyl CH
(minor)), 129.1 (indenyl olefinic CH (major)), 135.7
(olefinic indenyl CH (major)), 144.3 (indenyl ipso-C (mi-
nor)), 144.4 (olefinic indenyl CH (minor)), 133.1, 135.1,
136.4, 136.5 (Cp′′), 144.5, 144.9, 148.0 (indenyl).
Synthesis of (R)-dimethyl[1-(η5-3 ′-(–)-
menthylcyclopentadienyl)-1′-(η5-2,3,4,5-tetramethyl-
cyclopentadienyl)dimethylsilane]titanium (5)
Methylmagnesium bromide (0.76 cm3, 2.4 mmol, 3M in
diethyl ether) was added to a red solution of 4aR in diethyl
ether (15 cm3) and the mixture stirred overnight in the dark.
The diethyl ether was removed in vacuo from the resulting
yellow solution and the product extracted by washing the
residue with hexane (5 × 1 cm3). The solvent was removed
in vacuo. No further purification was attempted since previ-
ous attempts to crystallize the product or purify on a
silanized silica column had failed. The product was stored
under Ar(g) in a lightshielded flask at –30°C. FAB-MS m/z:
459 ([M + 1]+), 429 ([M + 1 – 2CH3]+). 1H NMR (250 MHz,
C6D6) δ: –0.29 (s, 3H, Ti-CH3), –0.19 (s, 3H, Ti-CH3), 0.02
(s, 3H, Si-CH3), 0.28, (s, 3H, Si-CH3), 0.75 (d, J = 7 Hz, 3H,
Me), 0.81 (d, J = 7 Hz, 3H, Me), 0.92 (d, J = 7 Hz, 3H, Me),
0.7–1.65 (m, 8H, menthyl-CH’s, CH2’s), 1.49 (s, 3H, TMCp
CH3), 1.59 (s, 3H, TMCp CH3), 1.75 (s, 3H, TMCp CH3),
2.11 (s, 3H, TMCp CH3), 2.40 (m, 1H, menthyl CH), 2.70
Synthesis of dichloro[(η5-1-indenyl)-1′-(η5-2,3,4,5-tetra-
methylcyclopentadienyl)dimethyl silane]zirconium (6)
1-Indenyl-1′-(2,3,4,5-tetramethylcyclopentadienyl)dimethyl-
silane (1.01 g, 3.45 mmol) was dissolved in THF (10.0 cm3)
and the solution cooled to –78°C. n-Butyllithium (2.3 M,
4
4
(m, 1H, menthyl CH), 4.94 (t, JH,H = 2 Hz, JH,H = 2 Hz,
3
4
3
1H), 5.47 (t, JH,H = 2 Hz, JH,H = 2 Hz,1H), 6.93 (t, JH,H
=
2 Hz, 4JH,H = 2 Hz, 1H). 13C NMR (63 MHz, C6D6) δ: 143.3,
132.3, 129.9, 128.5, 122.7, 99.7, 94.3 (quarternary Cp),
© 2001 NRC Canada

Beagley et al
741
3.00 cm3, 6.89 mmol) in hexanes was slowly added and the
reaction mixture stirred at –78°C for 10 min, then at room
temp for a further 1 h, giving a dark red solution. Tetra-
chloro-bis(tetrahydrofuran)zirconium (1.30 g, 3.45 mmol), was
dissolved in THF (25 cm3), cooled to 0°C, and transferred
via cannula onto the dianion solution at 0°C. The reaction
mixture was stirred at room temp for 20 h giving a yellow
suspension in an orange solution. The reaction was heated to
40°C for 72 h causing the mixture to become increasingly
yellow. The mixture was filtered and washed with toluene
yielding a bright yellow solid (1.41 g, 90.0% crude). Solu-
tions were prepared immediately prior to running NMR spectra
since the solid was unstable in solution. FAB-MS m/z (%):
Müller. Angew. Makromol. Chem. 223, 155 (1994); (f) K.B.
Sinclair and R.B.Wilson. Chem. Ind. (London) 857 (1994).
4 (a) V.P. Conticello, L. Brard, M.A. Giardello, Y. Tsuji, M.
Sabat, C.L. Stern, and T.J. Marks. J. Am. Chem. Soc. 114,
2761 (1992); (b) T.J. Marks, M.A. Giardello, V.P. Conticello, L.
Brard, and M. Gagne. J. Am. Chem. Soc. 116, 10241 (1994).
5 T. Hayashi, M. Tanaka, and I. Ogata. Tetrahedron Lett. 295
(1977).
6. (a) W. Dumont, J.C. Poulin, T.P. Dang, and H.B. Kagan. J.
Am. Chem. Soc. 95, 8295 (1973); (b) O. Samuel, R.
Couffignal, M. Lauer, S.Y. Zhang, and H.B. Kagan. Nouv. J.
Chim. 5, 15 (1981).
7. Q. Jiang, D. Xiao, Z. Zhang, P. Cao, and X. Zhang. Angew.
Chem. Int. Ed. 38, 516 (1999).
1
455 (M+, 5), 419 ([M – Cl]+, 14), 247 (24), 173 (80). H
8. (a) R.L. Halterman, K.P.C. Vollhardt, and M.E. Welker. J. Am.
Chem. Soc. 109, 8105 (1987); (b) R.D. Broene and S.L.
Buchwald. J. Am. Chem. Soc. 115, 12569 (1993).
9. A. Lightfoot, P. Schnider, and A. Pfaltz. Angew. Chem. Int.
Ed. 37, 2897 (1998).
NMR (400 MHz, CDCl3) δ: 0.95, 1.15 (s, 6H, 2SiCH3),
1.89, 1.93, 1.45, 1.47 (s, 12H, 4Cp′′CH3), 5.98, 7.24 (d, J =
3 Hz, 2H, indenyl), 7.06, 7.33, 7.50, 7.70 (m, 4H, indenyl).
Catalytic studies
10. M.V. Troutman, D.H. Appella, and S.L. Buchwald. J. Am.
Chem. Soc. 121, 4916 (1999).
Hydrosilylations were carried out by adding n-BuLi in to-
luene (1 cm3, 0.08 M) to the titanium complex (20 mg,
0.04 mmol) at room temp. After stirring the mixture rapidly
for 5 min., phenylsilane (108 µL, 0.88 mmol) was added fol-
lowed 5 min later by the ketone (0.8 mmol). The mixture
was then placed in a thermostatic bath and the reaction mon-
itored by GC. The product alcohols were obtained by adding
acetone and 0.1 M HCl to the mixture; after bubbling had
ceased, excess HCl was neutralized with a saturated solution
of NaHCO3 and the product extracted into diethyl ether and run
through a short plug of anhydrous MgSO4. Enantiomeric ex-
cess was determined using chiral GC using a β-Dex support.
11. (a) R.W. Harrigan, G.S. Hammond, and H.B. Gray. J. Organo-
met. Chem. 81, 79 (1974); (b) Z.T. Tsai and C.H. Brubaker Jr.
J. Organomet. Chem. 166, 199 (1979); (c) P.B. Brindley, A.G.
Davies, and J.A.-A. Hawari. J. Organomet. Chem. 250, 247
(1983); (d) K. Schmidt, A. Reinmuth, U. Rief, J. Diebold, and
H.H. Brintzinger. Organometallics, 16, 1724 (1997); (e) S.
Miyake, L.M. Henling, and J.E. Bercaw, Organometallics, 17,
5528 (1998).
12. (a) T.K. Hollis, A.L. Rheingold, N.P. Robinson, J. Whelan, and
B. Bosnich. Organometallics, 11, 2812 (1992); (b) J.A. Smith
and H.H. Brintzinger. J. Organomet. Chem. 218, 159 (1981).
13. H.H. Brintzinger, S. Gutmann, and M.H. Prosenc. J.
Organomet. Chem. 397, 21 (1990).
Acknowledgments
14. G.J. Wilson, D.A. Erskine, and J.D. McCowan. J. Organomet.
Chem. 114, 119 (1976).
15. R. Gomez, T. Cuenca, P. Royo, and E. Hovestreydt. Organo-
metallics, 10, 2516 (1991).
16. (a) R.L. Halterman, D. Combs, and M.A. Khan. Organometallics,
17, 3900 (1998); (b) H.H. Brintzinger, A. Schafer, E. Karl, L.
Zsolnai, and G. Huttner. J. Organomet. Chem. 328, 87 (1987).
17. (a) J.L. Petersen, W.R. Tikkanen, and C.S. Bajgur. Inorg.
Chem. 24, 233 (1985); (b) H.H. Brintzinger, G. Huttner, and J.
von Seyerl. J. Organomet. Chem. 288, 69 (1985); (c) N.P.
Rath, J.Y. Corey, and R.M. Shalton. J. Organomet. Chem. 503,
205 (1995).
18. C. White, P. Beagley, P. Davies, J.A. Blacker, and B.J. Mellor.
PCT Int. Appl. WO 9 911 648 to Zeneca Ltd. 1999 (March 11).
19. H.J. Phillips. Chem. Soc. 127, 2566 (1925).
20. S. Winstein, B.K. Morse, E. Grunwald, H.W. Jones, J. Corse,
D. Trifan, and H. Marshall. J. Am. Chem. Soc. 74, 1127 (1952)
and refs. therein.
21. J.A. Ramsden, D.A. Milner, H. Adams, N.A. Bailey, A.J.
Smith, and C. White. Organometallics, 14, 2575 (1995).
22. L.E. Manzer. Inorg. Synth. 21, 137 (1982).
23. H.B. Kagan, E. Cesarotti, R. Goddard, and C. Kruger. J.
Organomet. Chem. 162, 297 (1978).
We thank the EPSRC for postgraduate studentships (to
P.D. and P.B) and I.C.I. Specialties for financial support. Ad-
vice and constant encouragement from David J. Milner
(I.C.I.) is gratefully acknowledged.
References
1. (a) R.L. Halterman. Chem. Rev. 92, 965 (1992);
(b) Metallocenes: synthesis, reactivity, applications. Vol. 1 and
2. Edited by A. Togni and R.L. Halterman. Wiley-VCH, New
York. 1998; (c) N. Long. Metallocenes: an introduction to
sandwich complexes. Blackwell Science Inc., Malden. 1998.
2 (a) X. Verdaguer, U.E.W. Lange, M.T. Reding, and S.L.
Buchwald. J. Am. Chem. Soc. 118, 6784 (1996); (b) C.A. Wil-
loughby and S.L. Buchwald. J. Am. Chem. Soc. 116, 8952
(1994); (c) R.L. Halterman, T.M. Ramsey, and Z.L. Chen. J.
Org. Chem. 59, 2642 (1994); (d) Z.L. Chen and R.L.
Halterman. J. Am. Chem. Soc. 114, 2276 (1992); (e) R.L.
Halterman, K.P.C. Vollhardt, M.E. Welker, D. Blaser, and R.
Boese. J. Am. Chem. Soc. 109, 8105 (1987); (f) P. Pino, P.
Cioni, and J.J. Wei. J. Am. Chem. Soc. 109, 6189 (1987).
3. (a) L. Resconi, L. Cavallo, A. Fait, and F. Piemontesi. Chem.
Rev. 100, 1253 (2000); (b) M.A. Giardello, M.S. Eisen, C.L.
Stern, and T.J. Marks. J. Am. Chem. Soc. 117, 12114 (1995);
(c) H.H. Brintzinger, D. Fischer, R. Mülhaupt, B. Rieger, and
R.M. Waymouth. Angew. Chem. Int. Ed. Engl. 34, 1143
(1995); (d) R. Mülhaupt and B. Rieger. Chimia, 49, 486 (1995);
(e) F. Langhauser, J. Kerth, M. Kersting, P. Kölle, D. Lilge, and P.
24. F.X. Kohl and P. Jutzi. J. Organomet. Chem. 243, 119 (1983).
25. G.M. Sheldrick. SHELXL TL, an integrated system for solv-
ing, refining and displaying crystal structures and diffraction
data. University of Gottingen, Germany. 1978; revision 5.1, 1985.
26. International tables for X-ray crystallography. Vol IV. Kynoch
Press, Birmingham, U.K. (present distributor Kluwer Aca-
demic Publishers, Boston, U.S.A.) 1974.
© 2001 NRC Canada