"Constrained geometry" catalysts of the rare-earth metals for the hydrosilylation of olefins

Article abstract of DOI:10.1016/j.jorganchem.2006.01.060

A series of yttrium and lutetium alkyl complexes [Ln(η⁵-C₅Me₄ZNR′-κN)(CH₂SiMe₃)(THF)_n] (Ln = Y, Lu) was prepared by reacting the tris(trimethylsilylmethyl) precursor [Ln(CH₂SiMe<

Full text of DOI:10.1016/j.jorganchem.2006.01.060

Journal of Organometallic Chemistry 691 (2006) 4393–4399
www.elsevier.com/locate/jorganchem
‘‘Constrained geometry’’ catalysts of the rare-earth metals
for the hydrosilylation of olefins
*
Dominique Robert, Alexander A. Trifonov, Peter Voth, Jun Okuda
Institute of Inorganic Chemistry, RWTH Aachen University, Landoltweg 1, D-52056 Aachen, Germany
Received 22 November 2005; received in revised form 23 January 2006; accepted 31 January 2006
Available online 20 March 2006
Abstract
A series of yttrium and lutetium alkyl complexes [Ln(g5-C5Me4ZNR0-jN)(CH2SiMe3)(THF)n] (Ln = Y, Lu) was prepared by
reacting the tris(trimethylsilylmethyl) precursor [Ln(CH2SiMe3)3(THF)2] with different linked amino-cyclopentadienes of the type
(C5Me4H)ZNHR0 (Z = SiMe2, CH2SiMe2; R0 = tBu, Ph, C6H4-tBu-4, C6H4-nBu-4). The catalytic activity of these alkyl complexes in
the hydrosilylation of 1-decene and styrene using PhSiH3 as reagent was examined under standard conditions. A significant influence
of the ligand structure on the catalytic property (turnover frequency, regioselectivity) was observed with the yttrium complex [Y(g5-
C5Me4CH2SiMe2NtBu-jN)(CH2SiMe3)(THF)] being the most active for 1-decene hydrosilylation.
Ó 2006 Elsevier B.V. All rights reserved.
Keywords: Rare-earth metal; Hydrosilylation; Olefins; Alkyl complex; Bidentate ligand
1. Introduction
included to probe the influence of the ionic radius on the
catalytic activity.
Rare-earth metal complexes are known to be efficient
homogeneous catalysts for a wide range of organic trans-
formations [1]. While rare-earth metallocenes have been
extensively studied for hydrosilylation of olefins [1,2,5b],
mono(cyclopentadienyl) [3], and non-cyclopentadienyl [4]
complexes that catalyze this reaction are still quite uncom-
mon. In addition, the influence of the ligand structure on
the activity and regioselectivity of the catalyst remains
poorly understood [5]. We report here a systematic study
of the hydrosilylation of 1-decene and styrene catalyzed
by rare-earth alkyl complexes bearing a linked amido-
cyclopentadienyl ligand (‘‘constrained geometry’’ catalyst)
of the general formula [Ln(g5-C5Me4ZNR0-jN)(CH2Si-
Me3)(THF)n] [3c,3d]. Both electronic and steric characteris-
tics of the ligand were varied in order to determine the
factors that govern the efficiency and the regioselectivity
of the catalysts. Different rare-earth metals were also
2. Results and discussion
2.1. General hydrosilylation procedure
All yttrium and lutetium complexes [Ln(g5-C5Me4-
ZNR0-jN)(CH2SiMe3)(THF)n] used in this work have been
straightforwardly synthesized by the reaction of the
corresponding tris(trimethylsilylmethyl) complex [Ln-
(CH2SiMe3)3(THF)2] (Ln = Y, Lu) with the linked
amino-cyclopentadienes (C5Me4H)ZNHR0 in pentane with
concomitant formation of two equivalents of tetramethylsi-
lane [3]. Depending on the ligand characteristics (length of
the link Z and substituent at the amido-nitrogen R0), the
products displayed very different solubility in organic
solvents. Thus, [Ln(g5-C5Me4(CH2)nSiMe2NPh-jN)(CH2-
SiMe3)(THF)2] (Ln = Y or Lu, n = 0 or 1) and [Y(g5-C5-
Me4CH2SiMe2NCH2CH2NMe2-jN,N0)(CH2SiMe3)(THF)]
[15] were found to be sparingly soluble in pentane
(and could be isolated by simple filtration), but highly
soluble in aromatic solvents. The other complexes were
*
Corresponding author. Tel.: +49 0241 894645; fax: +49 0241 8092644.
E-mail address: jun.okuda@ac.rwth-aachen.de (J. Okuda).
0022-328X/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.01.060

4394
D. Robert et al. / Journal of Organometallic Chemistry 691 (2006) 4393–4399
SiPhH2
R
PhSiH3
PhH2Si
+
R
R
Z
Ln
CH2SiMe3
(THF)n
N
Z = SiMe2, CH2SiMe2
R' = tBu, C6H4X (X = H, tBu, nBu),
CH2CH2NMe2
R'
PhSiH3
R'
N
THF
THF
Z
H
Z
0.5
Z
Ln
Ln
Ln
N
(THF)m
H
N
R'
H
R'
Scheme 1.
generally highly soluble both in aliphatic and aromatic
hydrocarbons.
terminal) product in practically quantitative yield by all
complexes under consideration (Table 1). The catalytic effi-
ciency, as judged qualitatively by considering the turnover
frequency (TOF) values in hꢀ1 defined as mol substrate per
mol metal and reaction time shows considerable depen-
dence on the nature of the trimethylsilylmethyl complex
employed. The regioselectivity agrees with the general ten-
dency of a-olefins to give predominantly 1,2-insertion into
the polarized metal–hydride bond [6].
After their isolation, the trimethylsilylmethyl complexes
[Ln(g5-C5Me4ZNR0-jN)(CH2SiMe3)(THF)n] were directly
used for hydrosilylation of 1-decene and styrene in NMR-
scale experiments. The olefin and PhSiH3 were subse-
quently added to the complex dissolved in C6D6 at 25 °C,
and monitoring of the typical signals of PhSiH2CH2SiMe3
1
in the H NMR spectrum in all cases evidenced the imme-
diate and complete formation of the catalytically active
hydrido species [Ln(g5-C5Me4ZNR0-jN)(l-H)(THF)m]2.
The kinetically competent species is assumed to be the
mononuclear hydride species in equilibrium with the dimer
[3a] (Scheme 1).
2.2.1. Structure–activity relationship
2.2.1.1. Length of the link. According to the previous
studies performed on rare-earth metallocenes in compari-
son with rare-earth ansa-metallocenes, the presence of a
link between the two cyclopentadienyl-moieties has been
shown to influence the reactivity and the regioselectivity
of the corresponding catalysts significantly by creating a
well defined space around the active site [7]. In the case
of ‘‘constrained geometry’’ catalysts, the length of the link
2.2. 1-Decene
Under standard conditions, 1-decene was exclusively
transformed into the anti-Markovnikov (1,2-addition or
Table 1
Hydrosilylation of 1-decene with PhSiH3 catalyzed by rare-earth metal alkyl complex with various linked amido-cyclopentadienyl ligands; R = CH2SiMe3
Complex
Turnover frequency, TOF (hꢀ1
)
Regioselectivity (1,2)/(2,1)
[Y(g5-C5Me4SiMe2NtBu-jN)R(THF)]
2.9
0.49
20.0
10.0
4.0
0.45
1.33
0.20
1.33
0.20
0.56
0.19
<0.15
100/0
100/0
100/0
100/0
100/0
100/0
100/0
100/0
100/0
100/0
100/0
100/0
100/0
[Lu(g5-C5Me4SiMe2NtBu-jN)R(THF)]
[Y(g5-C5Me4CH2SiMe2NtBu-jN)R(THF)]
[Lu(g5-C5Me4CH2SiMe2NtBu-jN)R(THF)]
[Y(g5-C5Me4SiMe2NPh-jN)R(THF)2]
[Lu(g5-C5Me4SiMe2NPh-jN)R(THF)2]
[Y(g5-C5Me4CH2SiMe2NPh-jN)R(THF)2]
[Lu(g5-C5Me4CH2SiMe2NPh-jN)R(THF)2]
[Y(g5-C5Me4CH2SiMe2NC6H4-tBu-4-jN)R(THF)2]
[Lu(g5-C5Me4CH2SiMe2NC6H4-tBu-4-jN)R(THF)1.5
[Y(g5-C5Me4CH2SiMe2NC6H4-nBu-4-jN)R(THF)2]
]
[Lu(g5-C5Me4CH2SiMe2NC6H4-nBu-4-jN)R(THF)1.5
[Y(g5-C5Me4CH2SiMe2NCH2CH2NMe2-jN,N0)R(THF)]
]

D. Robert et al. / Journal of Organometallic Chemistry 691 (2006) 4393–4399
4395
20
10
CH2SiMe2
SiMe2
0
LuNtBu
LuNPh
YNtBu
YNPh
Complex
Fig. 1. Turnover frequencies (TOF, hꢀ1) for the 1-decene hydrosilylation by PhSiH3 catalyzed by various rare-earth metal trimethylsilylmethyl complexes.
The effect of the length of the link.
was proposed to influence the equilibrium between the
dimeric and the monomeric forms of the active hydrido
species [3a]. Thus, the longer CH2SiMe2 link would block
more efficiently the space around the metal and allow the
hydride to be sterically encumbered to shift the dissociation
equilibrium towards the monomeric species [8]. We assume
that the monomeric hydride is more catalytically efficient
than the corresponding dimer, although in the case of
metallocene hydrides the dimer was also shown to be active
[5b]. The results of the hydrosilylation as function of the
link are compiled in Fig. 1.
exchanging the SiMe2- for the CH2SiMe2-link. Possibly the
combination of the longer link and the phenyl ring created
too much steric bulk around the metal and hindered the
approach of the substrate.
2.2.1.2. Substituent at the amido-nitrogen. The reactivity
induced by the tBu group being taken as a reference, the
influence of different amido-nitrogen substituents was stud-
ied (Fig. 2). Changing the tBu group against the phenyl ring
had a detrimental effect on the catalytic activity. Because of
its electron withdrawing ability, the phenyl ring renders the
metal center more Lewis acidic (presence of two THF mol-
ecules on [Y(g5-C5Me4SiMe2NPh-jN)(CH2SiMe3)(THF)2]
[15] in contrast to one on [Y(g5-C5Me4SiMe2NtBu-
jN)(CH2SiMe3)(THF)] [12]). Therefore, a higher reactivity
For the tBuN substituent all the catalysts with the
longer CH2SiMe2-bridge displayed a better activity than
those with the shorter SiMe2-link. For the PhN substituent
a decrease in reactivity was observed for all catalysts, when
20
10
tBu
Ph
C6H4-tBu-4
C6H4-nBu-4
0
Lu-
CH2SiMe2
Lu-SiMe2
Y-
Y-SiMe2
CH2SiMe2
Complex
Fig. 2. Turnover frequencies (TOF, hꢀ1) for the 1-decene hydrosilylation by PhSiH3 catalyzed by various rare-earth metal trimethylsilylmethyl complexes.
The effect of the amido-substituent.

4396
D. Robert et al. / Journal of Organometallic Chemistry 691 (2006) 4393–4399
could be expected from the anilido derivatives towards elec-
tron-rich substrates. As the opposite effect was observed, the
increased Lewis acidity of the metal possibly led to a more
stable hydride dimer resulting in lower activity. The substi-
tution of the phenyl ring at the para position by a tBu group
does not significantly influence the catalytic activity; the
tBu-group cannot interact neither with the metal center,
nor with the substrate. When this tBu group is exchanged
for a nBu one, a slight decrease in activity was noted.
The hemilabile CH2CH2NMe2 group in [Y(g5-C5Me4-
CH2SiMe2NCH2CH2NMe2-jN,N0)R(THF)] seemed to
have an even stronger deactivating effect than the anilido
groups, possibly because of the additional donor which
coordinates at the metal center and blocks the active site.
The double bond of the olefin might not be able to displace
this pendant arm to efficiently access the metal center in
order to insert into the Ln–H bond.
the whole range of catalysts that were tested (Table 2).
The ratio of anti-Markovnikov/Markovnikov isomer ran-
ged from 4/96 to 46/54.
2.3.1. Structure–activity relationship
2.3.1.1. Length of the link. For the tBuN substituent, the
trend observed for styrene was exactly the opposite to what
was found for 1-decene, the differences being however less
pronounced: the catalysts bearing the longer CH2SiMe2-
link were not always found to be more active than those
with the shorter SiMe2-link.
As was found for 1-decene for the PhN substituent, a
decrease in reactivity toward styrene was observed for all
catalysts, when exchanging the short link for the long
one (Fig. 3). Only [Y(g5-C5Me4CH2SiMe2NPh-jN)-
(CH2SiMe3)(THF)2] showed
a
higher activity than
[Y(g5-C5Me4SiMe2NPh-jN)(CH2SiMe3)(THF)2] towards
styrene.
2.2.2. Regioselectivity
Even in the case of complexes containing a linked
amido-cyclopentadienyl ligand, the coordination sphere
around the metal appears to be encumbered. This observa-
tion is in accordance with the results obtained by Molander
et al. in their study of the influence of parameters such as
metal radius and ligand sterics on the regioselectivity of
1-decene hydrosilylation by rare-earth metallocenes [7].
They showed that a major condition for the increase in
2,1-regioisomer vs. 1,2-regioisomer was the accessible space
at the active site of the catalyst and that the combination of
a small metal with a sterically hindered ligand gave com-
plete 1,2-regioselectivity. The linked amido-cyclopentadie-
nyl complexes do not appear to offer sufficient space for
the substrate to allow the 2,1-hydrosilylation of 1-decene.
Factors other than steric ones may also play a role. This
may include the nature of the monomer–dimer equilibrium
and electronic factors.
2.3.1.2. Substituent at the amido-nitrogen. For the hydro-
silylation of styrene, the same trend as for 1-decene is
observed: the reactivity globally decreases as the steric bulk
of the substituent increases with increasing electron-accep-
tor character. In general, the anilido derivatives are less
active when compared with the tBuN derivatives (Fig. 4).
The tridentate ligand tested in this series, [C5Me4-
CH2SiMe2NCH2CH2NMe2]2ꢀ, appeared to have a strong
deactivating effect on the reactivity towards 1-decene, but
gave a relatively active catalyst for styrene (total conver-
sion within 7.5 h). Furthermore, the hydrosilylation was
highly regioselective (96% of Markovnikov or branched
regioisomer). This type of activity and regioselectivity are
observed with sterically more ‘‘open’’ catalysts such as
[(g5-C5Me5)2NdCH(SiMe3)2] [9], [Y(g5-C5Me4CH2Si-
Me2NtBu-jN)(CH2SiMe3)(THF)] [3a], [K(THF)2]3[(g5-
C5Me5)6Sm6H15] [3e] or [Me2Si(C5Me4)2SmCH(SiMe3)2]
[5a]. The result obtained with the tridentate ligand [16]
can be correlated with the observation that the product
of styrene insertion into the yttrium–hydride bond shows
no phenyl–metal interaction in [Y(g5-C5Me4CH2Si-
Me2NtBu-jN)(CHCH3Ph)(THF)] [3a], whereas such an
2.3. Styrene
Hydrosilylation of styrene occurred at considerably
lower rates of conversion in comparison to 1-decene for
Table 2
Hydrosilylation of styrene with PhSiH3 catalyzed by rare-earth metal alkyl complex with various linked amido-cyclopentadienyl ligands; R = CH2SiMe3
Complex
Turnover frequency, TOF (hꢀ1
)
Regioselectivity primary/secondary
[Y(g5-C5Me4SiMe2NtBu-jN)R(THF)]
4.0
0.18
2.5
12/88
11/89
40/60
46/54
37/63
31/69
25/75
22/78
27/73
30/70
25/75
20/80
4/96
[Lu(g5-C5Me4SiMe2NtBu-jN)R(THF)]
[Y(g5-C5Me4CH2SiMe2NtBu-jN)R(THF)]
[Lu(g5-C5Me4CH2SiMe2NtBu-jN)R(THF)]
[Y(g5-C5Me4SiMe2NPh-jN)R(THF)2]
0.17
0.22
0.17
0.87
0.17
0.80
0.17
0.59
0.16
2.7
[Lu(g5-C5Me4SiMe2NPh-jN)R(THF)2]
[Y(g5-C5Me4CH2SiMe2NPh-jN)R(THF)2]
[Lu(g5-C5Me4CH2SiMe2NPh-jN)R(THF)2]
[Y(g5-C5Me4CH2SiMe2NC6H4-tBu-4-jN)R(THF)2]
[Lu(g5-C5Me4CH2SiMe2NC6H4-tBu-4-jN)R(THF)1.5
[Y(g5-C5Me4CH2SiMe2NC6H4-nBu-4-jN)R(THF)2]
]
[Lu(g5-C5Me4CH2SiMe2NC6H4-nBu-4-jN)R(THF)1.5
[Y(g5-C5Me4CH2SiMe2NCH2CH2NMe2-jN,N0)R(THF)]
]

D. Robert et al. / Journal of Organometallic Chemistry 691 (2006) 4393–4399
4397
4.0
2.0
CH2SiMe2
SiMe2
0.0
LuNtBu
LuNPh
YNtBu
YNPh
Complex
Fig. 3. Turnover frequencies (TOF, hꢀ1) for the styrene hydrosilylation by PhSiH3 catalyzed by various rare-earth metal trimethylsilylmethyl complexes.
The effect of the length of the link.
4
2
tBu
Ph
C6H4-tBu-4
C6H4-nBu-4
0
Lu-
CH2SiMe2
Lu-SiMe2
Y-
Y-SiMe2
CH2SiMe2
Complex
Fig. 4. Turnover frequencies (TOF, hꢀ1) for the styrene hydrosilylation by PhSiH3 catalyzed by various rare-earth metal trimethylsilylmethyl complexes.
The effect of the amido-substituent.
interaction is found in [Y(g5-C5Me4CH2SiMe2CH2CH2-
NMe2-jN,N0)(CHCH3Ph)] [16] as well as in [Y(g5-C5Me4-
SiMe2NtBu-jN)(CHCH3Ph)] [12b].
radius of the metal used. For metallocenes or ansa-metal-
locenes, the regioselectivity for the Markovnikov product
can be increased up to 99% for the larger metal centers such
as Nd or Sm [5a,9]. For the complex bearing the tridentate
ligand with an additional pendant arm at the amido-nitro-
gen [Y(g5-C5Me4CH2SiMe2CH2CH2NMe2-jN,N0)(CH2-
SiMe3)(THF)], the high selectivity for the Markovnikov
product must be related to the hemilabile nature of the
ligand. It is well known that styrene shows a strong prefer-
ence for secondary insertion into the metal–hydride bonds
[10], potentially leading to an g3-coordination at a Lewis-
acidic/electrophilic metal center. This may be due to the
gn-coordination between the p-electron system of the sty-
rene and the Lewis-acidic/electrophilic metal center. This
2.3.2. Regioselectivity
The formation of both regioisomers suggests a situation,
which is not only governed by the sterics of both catalyst
and substrate, but also by the electronic properties of
styrene. [Y(g5-C5Me4CH2SiMe2NCH2CH2NMe2-jN,N0)-
(CH2SiMe3)(THF)] is the only complex that displayed a
very high regioselectivity for the branched isomer (96%).
The fact that the proportion of Markovnikov regioisomer
did not reach 100% might be due to at least two factors.
These may be the steric bulk of the ligand and the ionic

4398
D. Robert et al. / Journal of Organometallic Chemistry 691 (2006) 4393–4399
R'
THF
N
H
H
THF
Z
Z
Ln
Ln
1/2
N
R'
SiPhH2
R
PhH2Si
R
Z
Ln
N
THF
R = nC8H17, Ph
SiPhH3
H
R'
R = Ph
+
R
SiPhH3
Z
Z
Ln
Ln
N
THF
N
THF
R
?
R'
R'
R
secondary (Markovnikov)
insertion product
primary (anti-Markovnikov)
insertion product
Scheme 2.
stabilization of rare-earth metal by aromatic ring has been
shown to be quite common [5a,11,12b] in benzyl and other
arene-containing complexes. Since the Markovnikov
(primary) regioisomer was observed in non-negligible
quantities in every experiment, a parallel reaction pathway
involving a primary insertion must exist (Scheme 2). Under
kinetic control, this energetically favored interaction results
in the preference for secondary regioisomer. At longer reac-
tion time (in the case of less active catalysts) the primary
hydrosilylation product tends to increase. Interestingly
the related scandium hydride complex [Sc(g5-C5Me4Si-
Me2NtBu-jN)(l-H)(PMe3)]2 [8,17] gave a the bis(inser-
tion) product [Sc(g5-C5Me4SiMe2NtBu-jN){CH(Ph)-
CH2CH2CH2Ph}(PMe3)] as a result of primary insertion
followed by a secondary insertion.
somewhat obscure; the specific role of tert-amido group
in titanium based ‘‘constrained geometry’’ catalysts has
been noted before [18].
The regioselectivty for 1-decene hydrosilylation is per-
fect for anti-Markovnikov (1,2) product, whereas for
styrene both regioisomers are observed, with the Mark-
ovnikov product preferred in all cases examined.
In order to elucidate further the structure–activity rela-
tionship for these ‘‘constrained geometry’’ catalysts,
detailed kinetic analysis is required. However, as was sug-
gested earlier [3a], the rather qualitative concept of ‘‘open-
ing up’’ the reaction site to increase reactivity is much more
intricate than in the case of the metallocenes.
4. Experimental section
3. Conclusion
All experiments were performed under an inert atmo-
sphere of argon using standard Schlenk-line techniques.
Pentane and THF were distilled from sodium/benzophe-
none ketyl. Anhydrous yttrium and lutetium trichloride
(STREM) were used as received. The following products
were prepared according to published procedures: [Y(g5-
C5Me4SiMe2NtBu-jN)(CH2SiMe3)(THF)] [12], [Lu(g5-C5-
Me4SiMe2NtBu-jN)(CH2SiMe3)(THF)] [8], [Y(g5-C5Me4-
CH2SiMe2NtBu-jN)(CH2SiMe3)(THF)] [3a], [Lu(g5-C5-
Me4CH2SiMe2NtBu-jN)(CH2SiMe3)(THF)] [14], [Y(g5-
C5Me4SiMe2NPh-jN)(CH2SiMe3)(THF)2] [15], [Lu(g5-C5-
Me4SiMe2NPh-jN)(CH2SiMe3)(THF)2] [15], [Y(g5-C5Me4-
CH2SiMe2NPh-jN)(CH2SiMe3)(THF)2] [14], [Lu(g5-C5-
Me4CH2SiMe2NPh-jN)(CH2SiMe3)(THF)2] [14], [Y(g5-
C5Me4CH2SiMe2NC6H4-tBu-4-jN)(CH2SiMe3)(THF)2] [14],
[Lu(g5-C5Me4CH2SiMe2NC6H4-tBu-4-jN)(CH2SiMe3)-
The following general trends for the hydrosilylation
activity of olefins using ‘‘constrained geometry’’ catalysts
of the rare earth metals yttrium and lutetium can be
recognized:
The apparently slight decrease in the ionic radius from
Y to Lu (3.75% for coordination number 6) [13] results
in a rather significant drop in hydrosilylation activity
towards both 1-decene and styrene.
The activity trends do not follow simple structure–activ-
ity relationship, evidently due to some subtle effects, both
steric and electronic, on the hydride dimer–monomer equi-
librium. At this time we cannot exclude the possibility that
the hydride dimer itself may insert olefin to give a l-alkyl
species [17]. The effect of amido-substituents remain also

D. Robert et al. / Journal of Organometallic Chemistry 691 (2006) 4393–4399
4399
[14], [Y(g5-C5Me4CH2SiMe2NC6H4-nBu-4-
(b) G.A. Molander, J. Winterfeld, J. Organomet. Chem. 524 (1996)
275–279;
(c) G.A. Molander, W.H. Retsch, J. Am. Chem. Soc. 119 (1997)
8817–8825;
(d) H. Schumann, M.R. Keitsch, J. Winterfeld, S. Muhle, G.A.
¨
Molander, J. Organomet. Chem. 559 (1998) 181–190;
(e) G.A. Molander, C.P. Corrette, Organometallics 17 (1998) 5504–
5512;
(f) H. Schumann, M.R. Keitsch, J. Demtschuk, G.A. Molander, J.
Organomet. Chem. 582 (1999) 70–82.
(THF)1.5
]
jN)(CH2SiMe3)(THF)2] [14], [Lu(g5-C5Me4CH2SiMe2-
NC6H4-nBu-4-jN)(CH2SiMe3)(THF)1.5] [14], [Y(g5C5Me4-
CH2SiMe2NCH2CH2NMe2-jN,N0)(CH2SiMe3)(THF)]
[16]. NMR spectra were recorded on Varian Gemini 200
and Varian Unity 500 spectrometers at 25 °C. Chemical
shifts for H NMR and 13C NMR were reported in ppm
1
relative to tetramethylsilane, using residual solvent reso-
nances as reference. Deuterated solvents were dried over
sodium, distilled and degassed prior to use.
[3] (a) A.A. Trifonov, T.P. Spaniol, J. Okuda, Organometallics 20 (2001)
4869–4874;
(b) A.A. Trifonov, T.P. Spaniol, J. Okuda, Dalton Trans. (2004)
2245–2250;
(c) S. Arndt, J. Okuda, Chem. Rev. 102 (2002) 1953–1976;
(d) J. Okuda, Dalton Trans. (2003) 2367–2378;
(e) Z. Hou, Y. Zhang, O. Tardif, Y. Wakatsuki, J. Am. Chem. Soc.
123 (2001) 9216–9217;
4.1. Typical procedure for complex synthesis
To a pentane suspension of [Ln(CH2SiMe3)3(THF)2]
cooled to ꢀ78 °C was added a pentane solution of one
equivalent of the linked amino-cyclopentadiene. After the
addition, the solution was stirred for 1 h, and then slowly
allowed to warm to 0 °C. Additional stirring at this temper-
ature for 2.5 h followed by filtration (complexes insoluble
or sparingly soluble in pentane) or reduction of the solvent
(pentane soluble complexes) afforded the desired products
after standing for one night at ꢀ32 °C. The compounds
(f) O. Tardif, M. Nishiura, Z. Hou, Tetrahedron 59 (2003) 10525–
10539.
[4] (a) T.I. Gountchev, T.D. Tilley, Organometallics 18 (1999) 5661–
5667;
(b) K. Takaki, K. Sonoda, T. Kousaka, G. Koshoji, T. Shishido, K.
Takehira, Tetrahedron Lett. 42 (2001) 9211–9214;
(c) Y. Horino, T. Livinghouse, Organometallics 23 (2004) 12–14.
[5] (a) P.-F. Fu, L. Brard, Y. Li, T.J. Marks, J. Am. Chem. Soc. 117
(1995) 7157–7168;
(b) A.Z. Voskoboynikov, A.K. Shestakova, I.P. Beletskaya, Organo-
metallics 20 (2001) 2794–2801.
[6] (a) G.A. Molander, M. Julius, J. Org. Chem. 57 (1992) 6347–6351;
(b) K. Takaki, K. Komeyama, K. Takehira, Tetrahedron 59 (2003)
10381–10395.
1
already described in the literature were identified by H
1
NMR, all new complexes were fully characterized by H
and 13C NMR spectroscopy, elemental analysis, and metal
titration.
[7] G.A. Molander, E.D. Dowdy, B.C. Noll, Organometallics 17 (1998)
3754–3758.
4.2. Hydrosilylation procedure
[8] S. Arndt, P. Voth, T.P. Spaniol, J. Okuda, Organometallics 19 (2000)
4690–4700.
[9] T. Sakakura, H.-J. Lautenschlager, M. Tanaka, J. Chem. Soc.,
Chem. Commun. 1 (1991) 40–41.
[10] (a) A. Zambelli, P. Longo, C. Pellecchia, A. Grassi, Macromol. 20
(1987) 2037–2039;
In the glove-box, 25 lmol of the complex were dissolved
in 0.5 mL of C6D6 in an NMR tube equipped with a Young
valve. To that solution was first added 20 equivalents of the
olefin, and then 21 equivalents of phenylsilane. The total
volume of the solution was then increased to 0.75 mL,
and the tube was closed and vigorously shaken. The reac-
(b) B.J. Burger, B.D. Santarsiero, M.S. Trimmer, J.E. Bercaw, J.
Am. Chem. Soc. 110 (1988) 3134–3146;
(c) J.E. Nelson, J.E. Bercaw, J.A. Labinger, Organometallics 8 (1989)
2484–2486;
(d) A.M. Lapointe, F.C. Rix, M. Brookhart, J. Am. Chem. Soc. 119
(1997) 906–917.
1
tion was monitored by H NMR spectroscopy on a 200
MHz spectrometer at 25 °C. The reaction time was defined
as being the total time required for the olefin to be entirely
hydrosilylated by the catalyst. The proportions of Mark-
ovnikov and anti-Markovnikov regioisomers were calcu-
[11] (a) E.A. Mintz, K.G. Moloy, T.J. Marks, V.W. Day, J. Am. Chem.
Soc. 104 (1982) 4692–4695;
(b) W.J. Evans, T.A. Ulibarri, J.W. Ziller, J. Am. Chem. Soc. 112
(1990) 219–223.
[12] (a) K.C. Hultzsch, T.P. Spaniol, J. Okuda, Angew. Chem., Int. Ed.
38 (1999) 227–230;
1
lated by integration of the appropriate signals in the H
NMR spectra. All experiments were repeated at least twice.
Acknowledgements
(b) K.C. Hultzsch, P. Voth, K. Beckerle, T.P. Spaniol, J. Okuda,
Organometallics 19 (2000) 228–243.
[13] R.D. Shannon, Acta Cryst. A 32 (1976) 751–767.
[14] P. Voth, D. Robert, T.P. Spaniol, J. Okuda (in preparation).
[15] (a) M. Nishiura, Z. Hou, Y. Wakatsuki, T. Yamaki, T. Miyamoto,
J. Am. Chem. Soc. 125 (2003) 1184–1185;
This work was supported by the Deutsche Forschungs-
gemeinschaft and the Fonds der Chemischen Industrie.
References
(b) M. Nishiura, Z. Hou, J. Mol. Catal. A 213 (2004) 10–106;
(c) W.-Z. Zhang, M. Nishiura, Z. Hou, J. Am. Chem. Soc. 127 (2005)
16788–16789.
[1] (a) G.A. Molander, Chem. Rev. 92 (1992) 29–68;
(b) G.A. Molander, J.A.C. Romero, Chem. Rev. 102 (2002) 2161–
2185;
(c) R. Anwander, second ed., in: B. Cornils, W.A. Herrmann (Eds.),
Applied Homogeneous Catalysis with Organometallic Compounds,
vol. 2, Wiley-VCH, Germany, 2002, pp. 974–1014.
[2] (a) S. Onozawa, T. Sakakura, M. Tanaka, Tetrahedron Lett. 35
(1994) 8177–8180;
[16] P. Voth, T.P. Spaniol, J. Okuda, Organometallics 22 (2003) 3921–3926.
[17] (a) P. Voth, S. Arndt, T.P. Spaniol, J. Okuda, M.L.H. Green, L.J.
Ackerman, Organometallics 22 (2003) 65–76;
(b) P.J. Shapiro, W.D. Cotter, W.P. Schaefer, J.A. Labinger, J.E.
Bercaw, J. Am. Chem. Soc. 116 (1994) 4623.
[18] J. Okuda, K. Musikabhumma, P.-J. Sinnema, Isr. J. Chem. 42 (2002)
383–392.