Bis(phosphinimino)methanide rare earth amides: Synthesis, structure, and catalysis of hydroamination/cyclization, hydrosilylation, and sequential hydroamination/hydrosilylation

Article abstract of DOI:10.1002/chem.200601510

A series of yttrium and lanthanide amido complexes [Ln{N-(SiHMe ₂)₂}₂{CH(PPh₂NSiMe₃) ₂}] (Ln = Y, La, Sm, Ho, Lu) were synthesized by three different pathways. The title compounds can be obtained either from [Ln{N(SiHMe ₂)₂}₃(thf)₂] and [CH ₂(PPh₂NSiMe₃)₂] or from KN-(SiHMe₂)₂ and [Ln(CH(PPh₂NSiMe ₃)2}-Cl₂]₂, while in a third approach the lanthanum compound was synthesized in a one-pot reaction starting from K{CH(PPh₂NSiMe₃)₂}, LaCl₃, and KN-(SiHMe₂)₂. All the complexes have been characterized by single-crystal X-ray diffraction. The new complexes, [Ln{N(SiHMe ₂)₂}₂{CH(PPh₂NSiMe₃) ₂}], were used as catalysts for hydroamination/cyclization and hydrosilylation reactions. A clear dependence of the reaction rate on the ionic radius of the center metal was observed, showing the lanthanum compound to be the most active one in both reactions. Furthermore, a combination of both reactions - a sequential hydroamination/hydrosilylation reaction - was also investigated.

Full text of DOI:10.1002/chem.200601510

DOI: 10.1002/chem.200601510
Bis(phosphinimino)methanide Rare Earth Amides: Synthesis, Structure, and
Catalysis of Hydroamination/Cyclization, Hydrosilylation, and Sequential
Hydroamination/Hydrosilylation
Marcus Rastätter, Agustino Zulys, and Peter W. Roesky*[a]
Abstract: A series of yttrium and lan-
thanide amido complexes [Ln{N-
(SiHMe2)2}2{CH(PPh2NSiMe3)2}] (Ln=
Y, La, Sm, Ho, Lu) were synthesized
by three different pathways. The title
compounds can be obtained either
K{CH(PPh2NSiMe3)2}, LaCl3, and KN-
(SiHMe2)2. All the complexes have
been characterized by single-crystal X-
ray diffraction. The new complexes,
actions. A clear dependence of the re-
action rate on the ionic radius of the
center metal was observed, showing
the lanthanum compound to be the
most active one in both reactions. Fur-
thermore, a combination of both reac-
tions—a sequential hydroamination/hy-
drosilylation reaction—was also inves-
tigated.
ACHTREUNG
[Ln{N(SiHMe2)2}2{CH(PPh2NSiMe3)2}],
G
were used as catalysts for hydroamina-
tion/cyclization and hydrosilylation re-
from
[CH2(PPh2NSiMe3)2] or from KN-
(SiHMe2)2 and [Ln{CH(PPh2NSiMe3)2}-
Cl2]2, while in a third approach the lan-
thanum compound was synthesized in
[Ln{N
G
(thf)2]
and
ACHTREUNG
Keywords: catalysis · hydroamina-
tion · hydrosilylation · N,P ligands ·
rare earth metals
ACHTREUNG
a
one-pot reaction starting from
Introduction
try,[19–27] into yttrium and lanthanide chemistry as a cyclo-
pentadienyl replacement.[28–30] The obtained complexes were
used as homogenous catalysts for a number of different cat-
In lanthanide chemistry, alkyl, amido, and hydrido cyclopen-
tadienyl complexes, especially metallocenes such as [Ln-
alytic
(PPh2)2}Cl] species are active catalysts for the ring-opening
polymerization of e-caprolactone and the polymerization of
methyl methacrylate,[30] whereas [Ln{CH(PPh2NSiMe3)2}(h8-
applications.
Thus,
[Ln{CH(PPh2NSiMe3)2}{N-
(C5Me5)2R] (R=CH
(SiMe3)2, N
A
ACHTREUNG
to be highly efficient catalysts[1] for a variety of olefin trans-
formations including hydrogenation,[2–5] polymerization,[6–8]
hydroamination,[4,9] hydrosilylation,[10,11] hydroboration,[12]
and hydrophosphination.[13] In addition to the well establish-
ed cyclopentadienyl complexes, a number of non-cyclopen-
tadienyl lanthanide complexes based on amido and alkoxide
ligands are today also known to be active in hydroamina-
tion/cyclization catalysis.[14–18] The first non-cyclopentadienyl
lanthanide catalyst for the hydroamination/cyclization reac-
tion was developed by us,[14] while recently we have intro-
duced the use of the bis(phosphinimino)methanide
{CH(PPh2NSiMe3)2}ꢀ, previously employed by a number of
research groups in main group and transition metal chemis-
ACHTREUNG
C8H8)] was used as catalyst for the hydroamination/cycliza-
tion reaction.[28]
Very recently, Doye et al. have reported on Ti-based cata-
lysts for sequential hydroamination/hydrosilylation,[31] while
we have recently communicated the synthesis—based on
these results—of [La{N(SiHMe2)2}2{CH(PPh2NSiMe3)2}] as a
R
non-cyclopentadienyl lanthanide catalyst useable for the hy-
droamination/cyclization and the hydrosilylation reac-
tions.[32] Furthermore, we also studied this system for se-
quential hydroamination/hydrosilylation catalysis, and our
catalyst turned out to be significantly more active than the
previously reported systems.[31] In this contribution we now
present a full account of the reaction scope, substrate selec-
tivity, lanthanide ion size effect, and kinetic/mechanistic as-
pects of hydroamination/cyclization, hydrosilylation, and se-
quential hydroamination/hydrosilylation reactions catalyzed
[a] Dipl.-Chem. M. Rastätter, M.Sc. A. Zulys, Prof. Dr. P. W. Roesky
Institut für Chemie und Biochemie
Freie Universität Berlin
Fabeckstrasse 34–36, 14195 Berlin (Germany)
Fax : (+49)30-838-52440
E-mail: roesky@chemie.fu-berlin.de
by [Ln{N(SiHMe2)2}2{CH(PPh2NSiMe3)2}]. As well as the
C
easy preparation of a sophisticated catalyst, the main focus
of this contribution is to study the catalytic activity of a
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
3606
ꢀ 2007 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 3606 – 3616

FULL PAPER
robust and extremely efficient complex that may find broad
application in organic synthesis
Results and Discussion
Metal complex synthesis: The starting materials [Ln{N-
A
N
Lu (1e)] (Anwander amides)[33] were prepared in a modified
synthesis similar to that published by Anwander et al., al-
though in a one-pot synthesis, in contrast to the published
two-step procedure, which first required the preparation of
Figure 1. Solid-state structure of 1d, showing the atom labeling scheme
and omitting hydrogen atoms. Selected bond lengths [Š] or angles [8]:
ꢀ
ꢀ
ꢀ
ꢀ
the THF solvates [LnCl
3(thf)x]. Treatment of LnCl3 (Ln=Y,
Ho N1 2.270(2), Ho N2 2.235(2), Ho N3 2.278(2), Ho O1 2.385(2),
ꢀ
ꢀ
ꢀ
Ho O2 2.404(2), Ho Si4 3.2960(9), Ho Si6 3.3050(10); N1-Ho-N2
110.69(10), N1-Ho-N3 135.14(10), N2-Ho-N3 114.05(10), N1-Ho-O1
88.00(7), N2-Ho-O1 95.32(8), N3-Ho-O1 84.81(8), N1-Ho-O2 87.32(8),
N2-Ho-O2 101.70(8), N3-Ho-O2 86.96(8), O1-Ho-O2 162.92(8).
La, Sm, Lu) with KN(SiHMe2)2 (2.9 equiv) gave 1a–e in
AHCTREUNG
high yields. Compounds 1a–c and 1e have been described
previously,[33–35] while the new compound 1d was character-
ized by Raman spectroscopy and elemental analysis. The
ꢀ
characteristic Si Hvibrations
of 1d were observed in the
Raman spectrum at 2013 and
2076 cmꢀ1, while its solid-state
structure was established by
single-crystal X-ray diffraction.
Compound 1d, which in the
solid state adopts a distorted
trigonal bipyramidal geometry
(Figure 1), crystallizes in the
monoclinic space group P21/c
with four molecules in the unit
cell and is thus isostructural to
the previously reported com-
pounds 1a–c and 1e.[33–35]
The title compounds [Ln{N-
A
Me3)2}] [Ln=Y (2a), La (2b),
Sm (2c), Ho (2d), Lu (2e)] can
be obtained by three different
synthetic approaches. To study
the lanthanide ion size effect,
the complexes of the smallest
(Lu) and the largest (La) lan-
thanide ion, as well as those of
some selected elements in
Scheme 1.
between, were synthesized. The first approach starts
from compounds 1a–e, which were treated with
[CH2(PPh2NSiMe3)2] in boiling toluene to give 2a–e through
amine eliminations (Scheme 1A). This method gave the
highest yields and pure products and so was applied for the
synthesis of all five complexes. In contrast with the previous-
The most convenient approach would seem to be a one-
pot reaction sequence in which the potassium methanide
complex K{CH(PPh2NSiMe3)2} is treated with anhydrous
lanthanum trichloride and KN(SiHMe2)2 in a 1:1:2 molar
G
ratio in THF. Unfortunately, the yields are lower than those
obtained in the other methods and the product has to be pu-
rified (Scheme 1C).
ly
(dicyclohexylamide) and [CH2(PPh2NSiMe3)2],[51] which
under similar reaction conditions afford [Sm{C(Ph2P=
NSiMe3)2}(NCy2)
(thf)],[36] no carbene-type complex was ob-
described
reactions
between
samarium
tris-
E
Compounds 2a–e were characterized by standard analyti-
cal/spectroscopic techniques and the solid-state structures of
all five compounds were established by single-crystal X-ray
A
ACHTREUNG
1
served when 1a–e were used as starting materials.
diffraction. The HNMR spectra of 2a, 2b, and 2e show the
Compounds 2a and 2c were also obtained by treatment
of [Ln{(Me3SiNPPh2)2CH}Cl2]2 (Ln=Y, Sm, 1 equiv) with
characteristic signals for the two different substituents. Sin-
glets for the SiMe3 groups [d=0.12 (2a), 0.13 (2b), or
0.12 ppm (2e)] and triplets for the PCHP groups [d=2.36
KN(SiHMe2)2 (2 equiv, Scheme 1B).
A
Chem. Eur. J. 2007, 13, 3606 – 3616
ꢀ 2007 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
3607

P. W. Roesky et al.
(2a), 2.65 (2b), or 2.77 ppm (2e)] are thus observed for the
{CH(PPh2NSiMe3)2}ꢀ ligands, whereas doublets [d=0.60
(2a), 0.56 (2b), or 0.64 ppm (2e)] and septets [d=5.50 (2a),
5.44 (2b), or 5.45 ppm (2e)] are seen for the {N
(SiHMe2)2}ꢀ
G
ligands. In the 31P{1H} NMR spectra the characteristic sig-
nals of the {CH(PPh2NSiMe3)2}ꢀ ligands are observed. As a
result of the influence of the paramagnetic samarium center,
the observed chemical shift in the 31P{1H} NMR spectrum of
2c (d=26.3 ppm) differs from those in the diamagnetic com-
pounds 2a, 2b, and 2e [d=20.1 (2a), 17.1 (2b), 20.6 ppm
(2e)].
The solid-state structures of complexes 2a–e were estab-
lished by single-crystal X-ray diffraction (Figure 2). The co-
ordination polyhedra are formed by the {N
(SiHMe2)2}ꢀ
A
Figure 2. Perspective ORTEP view of the molecular structure of 2a.
Thermal ellipsoids are drawn to encompass 50% probability. Selected
bond lengths [Š] or angles [8] (also given for isostructural 2b–2e): 2a:
groups and the {CH(PPh2NSiMe3)2}ꢀ ligands. Six-membered
metallacycles (N1-P1-C1-P2-N2-Ln) are formed by chelation
of the two trimethlysilylimine groups to the lanthanide
atom. The rings all adopt twist boat conformations, in which
the central carbon atoms and the lanthanide atoms are dis-
placed from the N2P2 least-squares planes. The distances be-
tween the central carbon atom (C1) and the lanthanide
atom [2.697(4) (2a), 2.875(4) (2b), 2.771(3) (2c), 2.685(3)
(2d), 2.647(4) Š (2e)] are longer than average Ln–C distan-
ces; however, a resultant tridentate coordination of the
ligand is observed as previously.[28–30] In contrast with the re-
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Y C1 2.697(4), Y N1 2.403(3), Y N2 2.393(3), Y N3 2.250(4), Y N4
ꢀ
ꢀ
ꢀ
2.260(3), Y Si3 3.2922(15), Y Si4 3.4610(14), Y Si6 3.1331(13); N1-Y-N2
119.91(12), N1-Y-C1 64.42(12), N2-Y-C1 64.56(12), Y-N4-Si5 134.6(2), Y-
ꢀ
ꢀ
ꢀ
N4-Si6 103.5(2). 2b: La C1 2.875(4), La N1 2.536(2), La N2 2.536(3),
ꢀ
ꢀ
ꢀ
ꢀ
La N3 2.415(3), La N4 2.397(3), La Si3 3.4440(12), La Si4 3.4768(13),
ꢀ
La Si6 3.2848(12); N1-La-N2 112.05(9), N1-La-C1 60.75(9), N2-La-C1
60.49(9), La-N4-Si5 130.7(2), La-N4-Si6 105.6(2). 2c: Sm C1 2.771(3),
Sm N1 2.456(2), Sm N2 2.443(2), Sm N3 2.309(3), Sm N4 2.318(3),
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Sm Si3 3.3271(11), Sm Si4 3.4828(11), Sm Si6 3.1935(11); N1-Sm-N2
116.50(8), N1-Sm-C1 62.69(8), N2-Sm-C1 62.63(8), Sm-N4-Si5 132.5(2),
ꢀ
ꢀ
ꢀ
Sm-N4-Si6 104.49(13). 2d: H oC1 2.685(3), Ho N1 2.392(3), Ho N2
ꢀ
ꢀ
ꢀ
ꢀ
1
2.386(2), Ho N3 2.252(3), Ho N4 2.257(3), Ho Si3 3.3011(11), Ho Si4
sults obtained by HNMR for the diamagnetic compounds
ꢀ
3.4381(11), Ho Si6 3.1342(9); N1-Ho-N2 119.43(9), N1-Ho-C1 64.21(9),
N2-Ho-C1 64.57(9), Ho-N4-Si5 134.69(15), Ho-N4-Si6 103.59(13). 2e:
2a, 2b, and 2e, the two {N
alent in the solid state. The {N
to the sterically more hindered side (cis to C1) are symmet-
rically oriented, whereas the other {N
(SiHMe2)2}ꢀ groups
A
(SiHMe2)2}ꢀ groups attached
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Lu C1 2.647(4), Lu N1 2.356(3), Lu N2 2.342(4), Lu N3 2.207(4), Lu
ꢀ
ꢀ
ꢀ
N4 2.220(4), Lu Si3 3.3042(13), Lu Si4 3.3953(13), Lu Si6 3.1034(14);
N1-Lu-N2 121.25(12), N1-Lu-C1 65.11(12), N2-Lu-C1 65.64(13), Lu-N4-
Si5 135.0(2), Lu-N4-Si6-Lu 103.8(2).
AHCTREUNG
(trans to C1) are asymmetric and their geometries can best
be compared to that found in [(C5HPh4)2La{N-
(SiHMe2)2}].[37] One of the silicon atoms in this kind of
groups (Si6) approaches more closely to the metal center
than the others [Ln–Si6 3.1331(13) (2a), 3.2848(12) (2b),
3.1935(11) (2c), 3.1342(9) (2d), 3.1034(14) Š (2e), vs Ln–
Si3 3.2922(15) (2a), 3.4440(12) (2b), 3.3271(11) (2c),
3.3011(11) (2d), 3.3042(13) Š (2e) and Ln–Si4 3.4610(14)
(2a), 3.4768(13) (2b), 3.4828(11) (2c), 3.4381(11) (2d),
(2e) and Ln–N2 2.393(3) (2a), 2.536(3) (2b), Š 2.443(2)
(2c), 2.386(2) (2d), 2.342(4) Š (2e)].
Catalysis: Since we were interested in studying sequential
hydroamination/hydrosilylation we first tested the two reac-
tions separately. The mechanism of the lanthanide metallo-
cene-catalyzed hydroamination/cyclization reaction was es-
tablished some years ago by Marks et al.[9] The scope and
the limitations of using 2a–e as catalysts in the intramolecu-
lar hydroamination/cyclization reaction were tested with dif-
ferent non-activated terminal aminoalkynes and aminoal-
kenes (Scheme 2). We first investigated the dependence of
the rate on the ion radius of the center metal (Ln3+, coordi-
nation number 6)[39] by using 1-(prop-2-enyl)-cyclohexane-
3.3953(13) Š (2e)].
A similar effect is observed in
[(C5HPh4)2La{N(SiHMe2)2}] (3.261(2) vs 3.472(2) Š for one
ACHTREUNG
of the two independent molecules in the unit cell).[37] Con-
comitant with this effect, a decrease in the Ln-N-Si angle
[Ln-N4-Si6 103.5(2) (2a), 105.6(2) (2b), 104.49(13) (2c),
103.59(13) (2d), 103.8(2)8 (2e), vs 134.6(2) (2a), 130.7(2)
(2b), 132.5(2) (2c), 134.69(15) (2d), 135.0(2)8 (2e) for Ln-
N4-Si5] is observed in 2a–e. This is indicative of a weak b-
Si–Hmonoagostic interaction, as has been discussed for
other dimethylsilylamido rare earth metal complexes.[37,38]
The metal–nitrogen bond lengths in the {N
(SiHMe2)2}ꢀ
G
groups [Ln–N3 2.250(4) (2a), 2.415(3) (2b), 2.309(3) (2c),
2.252(3) (2d), 2.207(4) (2e) and Ln–N4 2.260(3) (2a),
2.397(3) (2b), 2.318(3) (2c), 2.257(3) (2d), 2.220(4) Š (2e)]
are close to those found in the starting material [Ln{N-
A
N
distance in the {CH(PPh2NSiMe3)2}ꢀ ligand [Ln–N1 2.403(3)
(2a), 2.536(2) (2b), 2.456(2) (2c), 2.392(3) (2d), 2.356(3)
Scheme 2.
3608
www.chemeurj.org
ꢀ 2007 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 3606 – 3616

Bis(phosphinimino)methanide Rare Earth Amides
FULL PAPER
spectroscopy. The observed NMR signal for the
{CH(PPh2NSiMe3)2}ꢀ ligand remains in the region of the
starting material (d=17.1 ppm) and does not shift to the
range of the free ligand {CH2(PPh2NSiMe3)2} (d=
3.39 ppm),[40]
which
clearly
indicates
that
the
{CH(PPh2NSiMe3)2}ꢀ ligand stays attached to the metal
center during the catalytic conversions.
In our substrate screening we focused on non-activated
aminoalkenes and aminoalkynes (Tables 1 and 2). The rigor-
ously anaerobic reaction between the catalyst and dry, de-
gassed aminoalkene and aminoalkyne proceeded regiospe-
cifically (Tables 1 and 2) and it turned out that all substrates
were converted into the cyclic products under moderate re-
action conditions. The aminoalkynes could be cyclized in
high yields at room temperature (Table 1, entries 1, 4, and
5), at 608C (Table 1, entry 3), or at 1008C (Table 1, entry 2).
Substrates bearing bulky geminal substituents in the b-posi-
tion to the amino group (Thorpe–Ingold effect)[41] could be
cyclized with reasonable catalyst/activator loadings of
Figure 3. Plot of reaction rates for the hydroamination/cyclization of 1-
(prop-2-enyl)-cyclohexanemethanamine (8a) with compounds 2a–e as
catalysts vs ionic radii of the center metals (Ln3+, C.N. 6).[39]
methanamine (8a) as test substrate (Figure 3). As observed
for the metallocene catalysts[9] of the lanthanides, the rates
increase with increasing ion radius, showing the lanthanum
compound 2b to be the most active one.
Kinetic studies were under-
1
taken by in situ HNMR spec-
troscopy. The reaction behavior
of a 50-fold molar excess of the
substrate with constant catalyst
concentration was monitored
until complete substrate con-
sumption and the decrease in
the substrate peak was integrat-
ed in relation to the product
signals. Figure 4 presents kinet-
ics data for the cyclization of 1-
(prop-2-enyl)-cyclohexaneme-
thanamine (8a) as substrate
with 2a as catalyst, which indi-
cates zero-order kinetics in sub-
strate concentration, in analogy
to the results for the metallo-
Figure 4. Ratio of conversion to lanthanide concentration as a function of time for the hydroamination/cycliza-
tion of 1-(prop-2-enyl)-cyclohexanemethanamine (8a) with compound 2a as catalyst in benzene.
cene catalysts. When the initial
concentration of the aminoal-
kene was held constant and the
concentration of the catalyst
was varied over
range,
ln(rate) versus ln
a
plot
[catalyst] in-
fivefold
of
a
G
G
dicates the reaction to be first
order in catalyst concentration
(Figure 5). The rate law can be
expressed as in [Eq. (1)] and is
identical to that for the lantha-
nide metallocene-catalyzed hy-
droamination/cyclization reac-
tions.[5]
0
1
ð1Þ
u¼k ½substrateŠ ½catalystŠ
The catalytic reaction was also
monitored by 31P{1H} NMR
Figure 5. Determination of the reaction order in catalyst concentration for the hydroamination/cyclization of
1-(prop-2-enyl)-cyclohexanemethanamine 8a using compound 2a as catalysts in benzene.
Chem. Eur. J. 2007, 13, 3606 – 3616
ꢀ 2007 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
3609

P. W. Roesky et al.
Table 1. Hydroamination/cyclization of terminal aminoalkynes catalyzed by 2b.
temperature. As observed for
the aminoalkynes, the Thorpe–
Ingold effect also has a signifi-
cant influence on to the reac-
tivity: compounds 8a, 9a, and
10a are thus rapidly converted
into the pyrrolidines with rea-
sonable catalyst loadings of 1–
2 mol% (Table 2, entries 1–3).
Substrate 10a was the most re-
active of the aminoalkenes,
giving the corresponding pyr-
rolidine within 1.5 h. Com-
pound 13a, which is known to
be a difficult substrate for this
reaction, could be converted
Entry
Substrate
Product
Cat. [mol%]
T [8C]
t [h]
Yield [%][a]
1.0
1.0
RT
60
120
1
quant
quant
1
2
3
4
5
2.0
2.0
2.0
2.0
100
60
4
4
95
99
RT
RT
30
40
99[b]
99
into
2,5-dimethylpyrrolidine
(13b) in good yield and in a
4:1 trans/cis ratio. The forma-
tion of a six-membered ring by
use of our catalyst represents a
very promising result (Table 2,
[a] Calculated by 1HNMR, with HN
dard.
A
1–2 mol% in good reaction times at room temperature
(Table 1, entries 4 and 5). As already reported earlier, hy-
droamination of aminoalkenes shows significantly lower
turnover frequencies than that of aminoalkynes (Table 2).[9]
Most of the substrates could be cyclized at 608C reaction
entry 5). It can be concluded that the rate of cyclization for
aminoalkenes follows the order 5>6 (Table 2, entries 2 and
5), consistently with classical, stereoelectronically controlled,
cyclization processes. In earlier studies we had investigated
the homoleptic amides [Ln{N
droamination catalysis.[14b]
comparison of [La{N(SiMe3)2}3]
A
A
AHCTREUNG
Table 2. Hydroamination/cyclization of terminal aminoalkenes catalyzed by 2b.
with catalyst 2b for substrate
3a shows a higher activity for
the latter catalyst. Obviously,
the rates for the hydroamina-
tion are slower than those ob-
served for the metallocene cat-
alyst compounds, yet the cur-
rent catalytic system has the
Entry
Substrate
Product
Cat [mol%]
T [8C]
t [h]
Yield [%][a]
1.1
1.1
RT
60
150
3
91[b]
quant
1
2
3
4
1.3
2.0
60
60
6
quant
quant
advantage of being a very
easily synthesized and robust
catalyst. Indeed, compounds
2a–e are far more easily acces-
sible than metallocene cata-
lysts. The CH2(PPh2NSiMe3)2
can be made in few hours with-
out solvent from commercial
available educts. The title com-
plexes are also accessible in a
one-step procedure. Moreover,
compounds 2a–e are much
more robust towards moisture
and air than the corresponding
metallocenes. It is well estab-
lished that leaving groups
1.5
82
1.7
2.2
60
60
36
22
(trans/cis 79:21)
5
6
quant
–
81
3.0
3.0
60
100
50
35
other than HN(SiHMe2)2 may
A
(trans/cis 80:20)
give higher activities for hydro-
amination,[17b,42] but attaching
these groups would result in
[a] Calculated by 1HNMR, with HN
dard.
(SiHMe2)2 as internal standard. [b] Ferrocene as additional internal stan-
3610
www.chemeurj.org
ꢀ 2007 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 3606 – 3616

Bis(phosphinimino)methanide Rare Earth Amides
FULL PAPER
other disadvantages. In general, lanthanide alkyls are much
more sensitive towards moisture and air and much less read-
ily accessible, while other amido groups give carbene-type
complexes (e.g., NCy2)[36] or are catalytically inactive (e.g.,
NPh2).[28]
(Table 3, entry 6). In this context, we also used compound
1b as a catalyst for the hydrosilylation reaction. Compound
1b is less active than compound 2b, but we did not investi-
gate the catalytic activity of 1b in more detail, because we
and others had shown previously that the homoleptic
amides of the lanthanides have only limited application as
catalysts for the hydroamination reaction,[14b,15b] so com-
pound 1b is obviously not a good catalyst in the desired se-
quential hydroamination/hydrosilylation.
As well as hydroamination, intermolecular hydrosilylation
catalyzed by 2a–e was also investigated (Scheme 3, Tables 3
As substrates for a full screening of compound 2b we
used terminal olefins and dienes that had been employed as
“benchmark” substrates earlier (Table 4).[45] All substrates
were treated with PhSiH3 at room temperature, together
with 1.5 mol% catalyst, to give the corresponding silanes
quantitatively. With regard to reaction temperature, catalyst
loading, and yields, the activity of compound 2b is compara-
ble to that reported for metallocene catalysts.[9] The aliphat-
ic substrates (Table 4, entries 1–3) were converted into the
corresponding anti-Markovnikov products with very high re-
gioselectivity. In contrast, with styrene as substrate a prod-
uct mixture was obtained (Table 4, entry 4). Additionally,
the cyclization/silylation of hexa-1,5-diene (18a) to afford
the corresponding cyclopentane 18b (Table 4, entry 5) was
investigated. Quantitative yields and high regioselectivity
were observed.
Scheme 3.
Table 3. Hydrosilylation of hex-1-ene (15a) with phenylsilane catalyzed
by 2a–c, 2e, and [La{N
(SiMe3)2}3].[a]
A
Entry
Catalyst
Cat. [mol%] T [8C] t [h]
Yield [%][b]
(ratio n/iso)
1
2
3
4
5
6
2a
2b
2c
2e
1b
1.5
1.5
1.5
1.5
3
100
RT
100
100
RT
RT
26
16
6
99
A
99
A
99
A
32
20
40
99
A
99
Finally, we were interested in combining the hydroamina-
tion/cyclization and the hydrosilylation reactions to provide
a sequential hydroamination/hydrosilylation reaction cata-
lyzed by the diamagnetic complexes 2a and 2b (Scheme 4).
Since no mechanistic studies involving sequential hydroami-
A
[La{N
(SiMe3)2}3]
3
98[a]
A
1
[a] Ref. [45]. [b] Calculated by HNMR.
and 4). Hydrosilylation of alkenes is one of the most versa-
tile and efficient methods for the synthesis of alkylsilanes
and their derivatives.[43] Usually, the highly air- and mois-
ture-sensitive lanthanide alkyl and hydride complexes have
been used as catalysts for hydrosilylation,[44] and there is
only one example in which the more stable lanthanide
Scheme 4.
amido
catalyst
[La{N-
ACHTREUNG
Table 4. Hydrosilylation reaction of terminal olefins and dienes catalyzed by 2b.
hydrosilylation catalyst (see
also Table 3, entry 6).[45] Again
we focused on the dependence
of the rate on the ionic radius
of the center metal, using hex-
1-ene and PhSiH3 as substrates
(Table 3). As observed in the
hydroamination, the rates in-
crease with increasing ionic
radius, showing the lanthanum
compound 2b to be the most
active one (Table 3, entry 2).
Compound 2b can be used at
room temperature to give
quantitative yields, reacting ap-
proximately twice as fast as
Entry
Substrate
Product
Cat. [mol%]
T [8C]
t [h]
Yield [%][a]
(ratio n/iso)
99
(99:1)
1
2
1.5
1.5
RT
RT
16
24
G
99[b]
(99:1)
G
99
(99:1)
3
1.5
RT
22
A
99
(35:65)
4
5
1.5
1.5
RT
RT
30
36
A
99
(99:1)
ACHTREUNG
[a] Calculated by 1HNMR, with HN
dard.
A
[La{N(SiMe3)2}3] with use of
half of the catalyst loading
G
Chem. Eur. J. 2007, 13, 3606 – 3616
ꢀ 2007 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
3611

P. W. Roesky et al.
nation/hydrosilylation reactions have been reported in lan-
thanide chemistry we were interested in studying the reac-
tion by multinuclear NMR, which is perfectly possible with
the {CH(PPh2NSiMe3)2}ꢀ ligand. Unlike Doye et al., who
did not isolate the hydrosilylation product but instead per-
formed a hydrolytic workup to isolate the corresponding
secondary amine, we were interested in obtaining the silicon
species.[31] The hydrogenation[46] and hydrosilylation[47] of
imines had been reported previously, but not in sequential
reactions. As substrates, six different phenyl- and alkyl-
only be characterized by NMR techniques. For full charac-
terization we hydrolyzed these products by exposure to air
to give the corresponding amines and then characterized
these amines.
To the best of our knowledge, compounds 2a and 2b are
the first rare earth catalysts for sequential hydroamination/
hydrosilylation reactions to be reported. Moreover, com-
pound 2b is more active than any other catalyst reported so
far for this reaction. Whereas for the titanium-based systems
a 10 mol% loading of catalyst was reported, we used only
2 mol%.[31] Moreover, the second step, the hydrosilylation,
involves working at room temperature with 2b whereas a
reaction temperature of 1058C was reported for the titanium
catalysts. To the best of our knowledge only two more com-
plexes have been reported as catalysts for sequential hydro-
amination/hydrosilylation reactions,[48] and those catalysts,
which are iridium complexes, were used for one reaction
only. The catalyst loadings were compared to those used by
us in a similar range, but the reaction temperature for the
second step was reported to be 608C.
The general advantage of the reported reactions is that
cyclic amines can be obtained either by direct hydroamina-
tion of the corresponding aminoalkene or by a stepwise hy-
droamination/hydrosilylation reaction sequence involving
the corresponding aminoalkynes. Since the hydroamination
of aminoalkenes is significantly slower than the hydroamina-
tion of aminoalkynes, the reported sequential hydroamina-
tion/hydrosilylation is an attrac-
A
complete selectivity with catalyst loadings of 2 mol%
(Table 5). As also observed in the other reactions, the lan-
thanum compound 2b was more active than the yttrium
complex 2a, and, as observed in the non-sequential reac-
tions, the hydroamination is always the slower reaction. The
hydrosilylation reaction was thus always complete within 4 h
at room temperature with the use of 2b as catalyst, while
quantitative conversions in the slower hydroamination reac-
tion in the presence of 2b in short reaction times were ob-
served at room temperature with the substrates 6a and 7a,
at 608C with the substrates 3a and 5a, and at 1008C with
the substrates 4a and 19a. As described above, the Thorpe–
Ingold effect has a significant influence on to the reactivity
of the substrates. As products, both five-membered (3c, 5c,
6c, and 7c) and six-membered rings (4c, 19c) were formed.
The silylamides shown in Table 5 are viscous oils that could
tive alternative pathway for the
Table 5. Hydroamination/hydrosilylation of aminoalkynes catalyzed by 2a and by 2b.[a]
synthesis of cyclic amines. The
products shown in Table 5 are
silylamides and can be easily
hydrolyzed by exposure to air
to provide the corresponding
amines.
Entry
Substrate
Product
Cat.[a]
T [8C]
t [h]
Yield [%][b]
60
60
60
4
4
4
4
99
99
2a
1
99
2b
RT
70[c]
100
RT
100
RT
16
4
4
99
99
99
99
2a
2b
2
Conclusions
4
In summary, it should be em-
phasized that a series of yttrium
and lanthanide amido com-
60
60
60
16
4
16
4
99
99
99
99
2a
2b
3
4
5
RT
plexes [Ln{N
A
G
N
Ph2NSiMe3)2}] [Ln=Y (2a), La
(2b), Sm (2c), Ho (2d), Lu
(2e)] have been synthesized by
three different pathways. These
RT
RT
RT
RT
30
4
16
4
99
99
99
99
2a
2b
RT
RT
RT
RT
40
4
16
4
99
99
99
99
[Ln{N
A
2a
2b
AHCTREUNG
catalysts for the hydroamina-
tion/cyclization and the hydrosi-
lylation reactions. A clear de-
pendence of the reaction rates
on the ionic radii of the center
metals was observed, showing
the lanthanum compound to be
the more active one in both re-
100
60
100
RT
24
12
4
96
95
99
99
2a
2b
6
4
[a] Conditions: 2 mol% of catalyst in C6D6, first step hydroamination, second step hydrosilylation. [b] Calcu-
lated by HNMR. [c] Isolated yield; reaction was performed on a 2 mmol scale.
1
3612
www.chemeurj.org
ꢀ 2007 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 3606 – 3616

Bis(phosphinimino)methanide Rare Earth Amides
FULL PAPER
[Sm{N
C
ACHTREUNG
actions. Furthermore, a combination of both reactions—a
sequential hydroamination/hydrosilylation reaction—was
also investigated. Compound 2b is the first lanthanide cata-
lyst studied for the sequential hydroamination/hydrosilyla-
tion reaction. Even the rates for the hydroamination are
slower than the ones observed for the metallocene catalyst
compounds 2a–e, which are far more easily accessible than
the metallocene catalysts. Compound CH2(PPh2NSiMe3)2
can be made in a few hours without solvent from commer-
cial available educts and the title complexes are also accessi-
ble in a one-step procedure. Moreover, compounds 2a–e are
much more robust than the corresponding metallocenes to-
wards moisture and air. Since no mechanistic studies involv-
ing sequential hydroamination/hydrosilylation reactions in
lanthanide chemistry have been reported, we were interest-
ed in studying the reaction by multinuclear NMR, which is
perfectly possible with the {CH(PPh2NSiMe3)2}ꢀ ligand.
(0.2 g, 0.78 mmol) and KN
AHCTREUNG
88%; FT Raman (single crystal): 399 (w), 615 (s), 681 (m), 760 (w), 926
(w), 1240 (w), 1427 (w), 2013 (w, nSi-H), 2075 (w, nSi-H), 2896 (s),
2951 cmꢀ1 (s, nCH); elemental analysis calcd (%) for C20H58N3O2Si6Sm
(691.57): C 34.73, H8.45, N 6.08; found: C 34.04, H8.87, N 5.70.
[Ho{N
(0.400 g, 1.47 mmol) and KN
AHCTREUNG
1.360 g, 97%. Crystals suitable for X-ray crystallography were obtained
from a concentrated THF solution of 1d. FT Raman (single crystal): n˜ =
400 (w), 625 (s), 681 (m), 761 (w), 926 (w), 1240 (w), 1427 (w), 2013 (w,
nSi-H), 2076 (w, nSi-H), 2896 (s), 2895 (m), 2952 cmꢀ1 (s, nCH); elemen-
tal analysis calcd (%) for C20H58N3OSi6Ho (706.14): C 34.02, H 8.28, N
5.95; found: C 33.89, H8.92; N 5.86.
[Lu{N
(0.500 g, 1.78 mmol) and KN
0.808 g, 63%; 1HNMR (C 6D6, 400 MHz, 208C): d=0.39 (d, 3J
3.04 Hz, 36H; SiH(CH3)2), 1.25 (m, 8H; O-CH2), 3.85 (m, 8H; CH2),
4.94 ppm (sept, 3J (CH3)2); 13C{1H} NMR (C6D6,
(H,H)=2.98 Hz, 6H; SiH
100.4 MHz, 208C): d=3.2 (SiH(CH3)2), 25.3 (O-CH2), 72.1 ppm (CH2).
[Ln{N(SiHMe2)2}2{CH(PPh2NSiMe3)2}] [Ln=Y (2a), La (2b), Sm (2c),
A
AHCTREUNG
ACHTREUNG
AHCTREUNG
ACHTREUNG
AHCTREUNG
AHCTREUNG
Ho (2d), Lu (2e)]
Method A: In
a
general procedure, 1a–e (1 equiv) and
[CH2(PPh2NSiMe3)2] (1 equiv) were dissolved in toluene (20 mL). The
clear solution was heated at reflux for 36 h and was then reduced to dry-
ness, extracted with pentane, and filtered. The filtrate was concentrated
and cooled to ꢀ788C, and a white (2a, 2b, 2e) to slightly yellow (2c) or
yellow/pink (2d) precipitate was formed.
Experimental Section
General considerations: All manipulations of air-sensitive materials were
performed with rigorous exclusion of oxygen and moisture in flame-dried
Schlenk-type glassware either on a dual manifold Schlenk line, interfaced
Method B: In a salt elimination reaction, [Ln{(Me3SiNPPh2)2CH}Cl2]2
to
a
high vacuum (10ꢀ4 Torr) line, or in an argon-filled glove box
(Ln=Y, Sm) (1 equiv) and KN(SiHMe2)2 (2 equiv) were dissolved in
G
(M. Braun). THF was predried over Na wire and distilled under nitrogen
from K and benzophenone ketyl prior to use. Hydrocarbon solvents (tol-
uene and n-pentane) were distilled under nitrogen from LiAlH4. All sol-
vents for vacuum line manipulations were stored in vacuo over LiAlH4 in
resealable flasks. Deuterated solvents were obtained from Chemotrade
Chemiehandelsgesellschaft mbH(all ꢁ99 atom% D) and were degassed,
dried, and stored in vacuo over Na/K alloy in resealable flasks. NMR
THF and stirred for 24 h. The solution was then reduced to dryness, ex-
tracted with pentane, and filtered. The filtrate was concentrated and
cooled to ꢀ788C, and a white precipitate was formed.
Method C: In a one-pot reaction, LaCl3 ( equiv), [K{CH(PPh2NSiMe3)2}]
(1 equiv), and KN(SiHMe2)2 (2 equiv) were dissolved in THF and stirred
G
for 48 h. The solution was then reduced to dryness, extracted with pen-
tane, and filtered. The filtrate was concentrated and cooled to ꢀ788C,
and a white precipitate was formed.
spectra were recorded with
a JNM-LA 400 FT-NMR spectrometer.
Chemical shifts are referenced to internal solvent resonances and are re-
ported relative to tetramethylsilane and 85% phosphoric acid
(31P NMR). Elemental analyses were carried out with an Elementar
[Y{N
A
Method A: This complex was prepared from Y{N
N
N
vario EL instrument. KN
(SiHMe2)2,[49] LnCl3,[50] {CH2(PPh2NSiMe3)2},[51]
E
(0.150 g, 0.24 mmol) and [CH2(PPh2NSiMe3)2] (0.134 g, 0.24 mmol).
Yield: 0.102 g, 79%.
[28]
[K{CH(PPh2NSiMe3)2}],[28] and [LnCl2{CH(PPh2NSiMe3)2}]2 were pre-
pared by literature procedures.
Method B: This complex was prepared from [YCl2{CH(PPh2NSiMe3)2}]2
[Ln{N
E
G
(0.331 g, 0.23 mmol) and KN(SiHMe2)2 (0.158 g, 0.92 mmol). Yield:
T
(1e)]: The compounds were prepared in a modified synthesis similar to
0.371 g, 88%. Crystals suitable for X-ray crystallography were obtained
that published by Anwander et al.[33] LnCl3 (Ln=Y, La, Sm, Lu, Ho) and
1
from a hot pentane solution of 2a. HNMR (C 6D6, 400 MHz, 208C): d=
KN(SiHMe2)2 (2.9 equiv) were placed in a flask. THF was then con-
E
0.12 (s, 18H; Si
2.63 (t, 3J(H,P)=6.83 Hz, 1H; CH), 5.50 (sept, 3J
SiH
(CH3)2), 6.8–7.5 ppm (m, 20H; Ph); 31P{1H} NMR (C6D6, 161.7 MHz,
208C): d=20.1 ppm (d, J(P,Y)=6.00 Hz); FT Raman (single crystal): n˜ =
A
G
A
densed onto the mixture at ꢀ1968C and the suspension was stirred over-
night (20 h) at ambient temperature. The solvent was then evaporated in
vacuo and the residue was extracted with toluene (Ln=Y, Lu) or pen-
tane (Ln=La, Sm, Ho). The extract was filtered, and the solvent was
taken off in vacuo. The remaining microcrystalline powder was washed
with a small amount of pentane at ꢀ788C.
A
ACHTREUNG
AHCTREUNG
2
A
406 (w), 622 (s), 661 (w), 679 (w), 999 (s, nPC), 1027 (m), 1109 (w), 1159
(w), 1575 (w), 1590 (m, nC=C), 2044 (w, nSi-H), 2140 (w, nSi-H), 2896
(m), 2950 (s, nCH), 3061 cmꢀ1 (m, nC=C-H); elemental analysis calcd
(%) for C39H67N4P2Si6Y (961.34): C 51.40, H7.41, N 6.15; found: C 51.15,
H7.18, N 5.43.
[Y{N
(0.500 g, 2.56 mmol) and KN
1.033 g, 66%; 1HNMR (C 6D6, 400 MHz, 208C): d=0.38 (d, 3J
2.92 Hz, 36H; SiH(CH3)2), 1.30 (m, 8H; O-CH2), 3.84 (m, 8H; CH2),
4.98 ppm (sept, 3J (CH3)2); 13C{1H} NMR (C6D6,
(H,H)=2.94 Hz, 6H; SiH
100.4 MHz, 208C): d=3.3 (SiH(CH3)2), 25.3 (O-CH2), 71.15 ppm (CH2).
[La{N(SiHMe2)2}3(thf)2] (1b): This complex was prepared from LaCl3
(0.400 g, 0.82 mmol) and KN(SiHMe2)2 (0.810 g, 2.37 mmol). Yield:
(SiHMe2)2}3
ACHTREUNG
AHCTREUNG
AHCTREUNG
[La{N
A
ACHTREUNG
Method A: This complex was prepared from [La{N
N
N
A
ACHTREUNG
(0.100 g, 0.15 mmol) and [CH2(PPh2NSiMe3)2] (0.082 g, 0.15 mmol).
Yield: 0.133 g, 94%.
AHCTREUNG
A
ACHTREUNG
Method C: This complex was prepared from LaCl3 (0.123 g, 0.5 mmol),
AHCTREUNG
[K{CH(PPh2NSiMe3)2}] (0.300 g, 0.5 mmol) and KN(SiHMe2)2 (0.171 g,
A
0.789 g. A second extraction with pentane gave an additional yield of
1
1 mmol). Yield: 0.390 g, 66%. Crystals suitable for X-ray crystallography
0.240 g. Overall yield 1.029 g, 93%; HNMR (C 6D6, 400 MHz, 208C): d=
were obtained from a concentrated pentane solution of 2b. 1HNMR
0.38 (d, 3J
C
ACHTREUNG
(C6D6, 400 MHz, 208C): d=0.13 (s, 18H; Si
2.92 Hz, 24H; SiH (H,P)=6.33 Hz, 1H; CH), 5.44
(CH3)2), 2.35 (t, 3J
(sept, 3J
(H,H)=2.92 Hz, 4H; SiH(CH3)2), 6.8–7.5 ppm (m, 20H; Ph);
A
N
A
ACHTREUNG
13C{1H} NMR (C6D6, 100.4 MHz, 208C): d=3.2 (SiH
CH2), 70.8 ppm (CH2).
U
A
ACHTREUNG
A
ACHTREUNG
Chem. Eur. J. 2007, 13, 3606 – 3616
ꢀ 2007 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
3613

P. W. Roesky et al.
31P{1H} NMR (C6D6, 161.7 MHz, 208C): d=17.1 ppm; FT Raman (single
crystal): n˜ =407 (w), 626 (s), 662 (w), 678 (w), 999 (s, nPC), 1027 (m),
1107 (w), 1159 (w), 1578 (w), 1590 (m, nC=C), 2044 (w, nSi-H), 2147 (w,
nSi-H), 2895 (m), 2949 (s, nCH), 3062 cmꢀ1 (m, nC=C-H); elemental anal-
ysis calcd (%) for C39H67LaN4P2Si6 (961.34): C 48.73, H7.02, N 5.83;
found: C 48.72, H7.33, N 5.34.
as described below. All substrates were dried by stirring over LiAlH4.
The substrates for the hydrosilylation (14a–18a) are commercially avail-
able.
Substrate synthesis
[1-(But-2-ynyl)cyclohexyl]methanamine (6a): Cyclohexanecarbonitrile
(7.5 mL, 64 mmol) was added dropwise at 08C to a solution of lithium
diisopropylamide (65 mmol) in THF (200 mL), followed by stirring at
08C for 2 h. 1-Bromobut-2-yne (65 mmol) was then added and the mix-
ture was stirred for 3 h at 08C. The prepared 1-(but-2-ynyl)cyclohexane-
carbonitrile was purified by distillation under reduced pressure. An unsa-
turated nitrile (40 mmol) was added under nitrogen at room temperature
with stirring to a suspension of LiAlH4 (2.28 g, 60 mmol) in dry ether
(200 mL). The mixture was stirred, heated at reflux for 2 h, and then
cooled in an ice bath. A THF/water mixture 5:1 (9.2 mL) and then aque-
ous NaOH(15%, 10 mL) were added dropwise to the reaction mixture
with vigorous stirring and cooling. After filtration of the dry granular
precipitate and washing with THF, the filtrate was condensed to leave a
colorless oil 6a, which was purified by distillation. Yield 67%; 1HNMR
(C6D6, 400 MHz, 258C): d=0.64 (brs, 2H; (NH2)), 1.20–1.32 (m, 10H),
[Sm{N
A
Method A: This complex was prepared from [Sm{N
N
N
(0.100 g, 0.15 mmol) and [CH2(PPh2NSiMe3)2] (0.082 g, 0.15 mmol).
Yield: 0.133 g, 94%. Crystals suitable for X-ray crystallography were ob-
tained from a concentrated pentane solution of 2c.
Method B: This complex was prepared from [SmCl2{CH(PPh2NSiMe3)2}]2
(0.331 g, 0.21 mmol) and KN(SiHMe2)2 (0.145 g, 0.85 mmol). Yield:
G
0.335 g, 80%. Crystals suitable for X-ray crystallography were obtained
from a hot pentane solution of 2c. 31P{1H} NMR (C6D6, 161.7 MHz,
208C): d=26.3 ppm (br); FT Raman (single crystal): n˜ =402 (w), 618 (s),
659 (w), 680 (w), 999 (s, nPC), 1027 (m), 1108 (w), 1159 (w), 1575 (w),
1590 (m, nC=C), 2047 (w, nSi-H), 2122 (w, nSi-H), 2895 (m), 2948 (s,
nCH), 3060 cmꢀ1 (m, nC=C-H); elemental analysis calcd (%) for
C39H67N4P2Si6Sm (972.80): C 48.15, H6.94, N 5.76; found: C 49.80, H
6.83, N 5.13.
1.55 (t,
J
N
J
(1H,1H)=2.56 Hz, 2H),
U
[Ho{N
A
Method A: This complex was prepared from [Ho{N
N
N
[1-(Pent-2-ynyl)cyclohexyl]methanamine (7a): Compound 7a was syn-
thesized in a manner analogous to that used for 6a with 1-bromopent-2-
yne in place of 1-bromobut-2-yne. Yield 67%; HNMR (C 6D6, 400 MHz,
(0.500 g, 0.71 mmol) and [CH2(PPh2NSiMe3)2] (0.396 g, 0.71 mmol).
Yield: 0.404 g, 58%. Crystals suitable for X-ray crystallography were ob-
tained from a concentrated pentane solution of 2d; FT Raman (single
crystal): n˜ =405 (w), 623 (s), 661 (w), 679 (w), 999 (s, nPC), 1027 (m),
1108 (w), 1159 (w), 1575 (w), 1590 (m, nC=C), 2045 (w, nSi-H), 2137 (w,
nSi-H), 2895 (m), 2950 (s, nCH), 3061 cmꢀ1 (m, nC=C-H); elemental anal-
ysis calcd (%) for C39H67HoN4P2Si6 (987.37): C 47.44, H6.84, N 5.67;
found: C 47.56, H6.89, N 5.28.
1
258C): d=0.64 (brs, 2H; (NH2)), 0.98 (t, J
1.32 (m, 10H), 2.00 (q, J
A
(s, 2H); 13C{1H} NMR (C6D6, 100 MHz, 258C): d=12.8, 14.6, 22.0, 22.3,
26.7, 33.0, 37.5, 49.1, 77.3, 88.4 ppm; MS (EI): m/z (%): 179 [M]+ (14),
164 [MꢀCH3]+ (11), 30 [CH2NH2+] (100).
Products: The spectra of compounds 3b,[9c] 4b,[9c] 5b,[9h] 8b,[9i] 9b,[9b]
10b,[9i] 11b,[9b] 12b,[9b] 13b,[9b] 14b,[11] 15b,[11] 16b,[11] 17b,[11] and 18b[10c]
were consistent with the literature. As a result of their high air sensitivi-
ties the silyamides were not obtained analytically pure and so were only
characterized by NMR. For better characterization they were then hydro-
lyzed to afford the corresponding amines. The silyamides 3c, 4c, 5c, 6c,
7c, and 19c in air gave the corresponding amines 2-benzylpyrrolidine
(3d),[9i] 2-benzylpiperidine (4d),[9i] 2-pentylpyrrolidine (5d),[11] 6d, 7d,
and 2-methylpiperidine (19d),[9i] which were characterized by 1Hand
13C{1H} NMR spectroscopy and MS.
[Lu{N
A
Method A: This complex was prepared from [Lu{N
N
N
(0.150 g, 0.24 mmol) and [CH2(PPh2NSiMe3)2] (0.134 g, 0.24 mmol).
Yield: 0.102 g, 79%. Crystals suitable for X-ray crystallography were ob-
tained from a concentrated pentane solution of 2e. 1HNMR (C 6D6,
400 MHz, 208C): d=0.12 (s, 18H; SiH
2.92 Hz, 24H; Si (H,P)=6.65 Hz, 1H; CH), 5.45 (sept,
(CH3)3), 2.77 (t, 3J
3J
(H,H)=2.90 Hz, 4H; SiH(CH3)2), 6.8–7.5 ppm (m, 20H; Ph);
A
G
A
ACHTREUNG
A
ACHTREUNG
31P{1H} NMR (C6D6, 161.7 MHz, 208C): d=20.6 ppm; FT Raman (single
crystal): n˜ =407 (w), 626 (s), 662 (w), 677 (w), 999 (s, nPC), 1027 (m),
1108 (w), 1159 (w), 1575 (w), 1590 (m, nC=C), 2044 (w, nSi-H), 2150 (w,
nSi-H), 2895 (m), 2949 (s, nCH), 3062 cmꢀ1 (m, nC=C-H); elemental anal-
ysis calcd (%) for C39H67LuN4P2Si6 (997.40): C 46.96, H6.77, N 5.62;
found: C 47.26, H6.65, N 4.62.
2-Benzyl-1-(phenylsilyl)pyrolidine (3c): 1HNMR (C 6D6, 400 MHz,
258C): d=1.34–1.47 (m, 4H), 2.32 (m, 1H), 2.72 (m, 1H), 2.92 (m, 2H),
3.51 (m, 1H), 5.05 (m, 2H), 6.92–7.55 ppm (m, 10H; 2”Ph);
13C{1H} NMR (C6D6, 100 MHz, 258C): d=26.0, 32.1, 44.1, 48.9, 62.4,
126.2, 128.3, 128.4, 128.5, 129.7, 135.1, 136.0, 140.2 ppm.
2-Benzyl-1-(phenylsilyl)piperidine (4c): 1HNMR (C 6D6, 400 MHz,
258C): d=1.07–1.25 (m, 6H), 2.37 (m, 1H), 2.55 (m, 1H), 2.70–2.76 (m,
2H), 3.05 (m, 1H), 4.70 (m, 2H), 6.76–7.30 ppm (m, 10H; 2”Ph);
13C{1H} NMR (C6D6, 100 MHz, 258C): d=20.8, 27.5, 30.0, 38.3, 43.3, 56.8,
126.2, 128.3, 128.5, 129.6, 130.2, 135.1, 136.1, 140.7 ppm.
General for the hydroamination and hydrosilylation reactions (NMR
scale reaction): The catalyst was weighed into an NMR tube under argon
gas. C6D6 (ꢂ0.7 mL) was condensed into the NMR tube, and the mixture
was frozen to ꢀ1968C. The reactant was injected onto the solid mixture,
and the whole sample was melted and mixed just before the insertion
into the NMR probe (t0). The ratio of reactant to product was exactly cal-
culated by comparison of the integrations of the corresponding signals.
2-Pentyl-1-(phenylsilyl)pyrrolidine (5c): 1HNMR (C 6D6, 400 MHz,
The N
measurements. By use of ferrrocene as an independent standard in some
reactions it was shown that HN(SiHMe2)2 is formed in stoichiometric
A
258C): d=0.82 (t, J
(m, 4H), 1.56–1.71 (m, 1H), 2.97 (m, 2H), 3.25 (m, 1H), 5.16 (q,
(1H,1H)=10 Hz, 2H), 7.31–7.60 ppm (m, 5H; Ph); 13C{1H} NMR (C6D6,
A
AHCTREUNG
J
ACHTREUNG
ratios. The sequential hydroamination/hydrosilylation reaction was per-
formed in a one-pot sequence. After completion of a hydroamination re-
action, PhSiH3 was just added to the reaction mixture.
100 MHz, 258C): d=14.4, 23.0, 26.6, 26.7, 32.3, 32.9, 37.9, 49.0, 60.6,
128.3, 130.0, 135.2, 136.0 ppm.
3-Ethyl-2-azaspiro
d=1.10 (t, J
(1H,1H)=7.44 Hz, 3H), 1.21–1.31 (m, 10H), 1.97 (s, 2H),
2.10 (q, J
(1H,1H)=7.44 Hz, 2H), 3.58 ppm (s, 2H); 13C{1H} NMR (C6D6,
A
For the intramolecular hydroamination/cyclization reactions, the sub-
strates 5-phenylpent-4-yn-1-amine[9c] (3a), 6-phenylhex-5-yn-1-amine[9c]
(4a), non-4-yn-1-amine[9h] (5a), (1-allylcyclohexyl)methanamine[9i] (8a),
2,2-dimethylpent-4-en-1-amine[9b] (9a), 2,2-diphenylpent-4-en-1-amine[9i]
(10a), 2-phenylpent-4-en-1-amine[9b] (11a), 2,2-dimethylhex-5-en-1-
amine[9b] (12a), hex-5-en-2-amine[9b] (13a), and hex-5-yn-1-amine[9h] (19a)
were synthesized by literature procedures. The new compounds [1-(but-2-
ynyl)cyclohexyl]methanamine (6a) and [1-(pent-2-ynyl)cyclohexyl]me-
thanamine (7a) were synthesized by modifications of literature methods
AHCTREUNG
AHCTREUNG
100 MHz, 258C): d=10.7, 24.0, 26.2, 27.4, 33.5, 37.4, 42.5, 73.1,
176.0 ppm; MS (EI): m/z (%): 166 [M+H]+ (100).
3-Ethyl-2-(phenylsilyl)-2-azaspiro
400 MHz, 258C): d=0.72 (t, J
(1H,1H)=7.44 Hz, 3H), 0.98 (m, 1H), 1.27–
1.21 (m, 10H), 1.62 (m, 3H), 2.72 (d, J
(1H,1H)=9.6 Hz, 1H), 2.85 (d,
(1H,1H)=9.6 Hz, 1H), 3.20–3.25 (m, 1H), 5.14 (s, 2H), 7.60–7.31 ppm
A
AHCTREUNG
AHCTREUNG
J
ACHTREUNG
3614
www.chemeurj.org
ꢀ 2007 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 3606 – 3616

Bis(phosphinimino)methanide Rare Earth Amides
FULL PAPER
(m, 5H; Ph); 13C{1H} NMR (C6D6, 100 MHz, 258C): d=10.6, 23.5, 24.5,
26.6, 31.0, 35.4, 37.2, 44.1, 60.9, 128.3, 130.0, 135.1, 135.4 ppm.
1996, 15, 1765–1784; d) P. W. Roesky, U. Denninger, C. L. Stern,
T. J. Marks, Organometallics 1997, 16, 4486–4492.
[5] M. Giardello, V. P. Conticello, L. Brard, M. R. GagnØ, T. J. Marks, J.
Am. Chem. Soc. 1994, 116, 10241–10254.
3-Ethyl-2-azaspiro
d=0.76 (t, J
(1H,1H)=7.08 Hz, 3H), 0.87 (m, 1H), 1.14–1.31 (m, 12H),
1.57 (m, 1H), 2.44 (d, J
(1H,1H)=9.6 Hz, 1H), 2.59 (d, J(1H,1H)=10.8 Hz,
[4.5]decane (6d): 1HNMR (C 6D6, 400 MHz, 258C):
ACHTREUNG
[6] a) P. L. Watson, J. Am. Chem. Soc. 1982, 104, 337–339; b) P. L.
Watson, G. W. Parshall, Acc. Chem. Res. 1995, 28, 51–56; c) E. B.
Coughlin, J. E. Bercaw, J. Am. Chem. Soc. 1992, 114, 7607–7608;
d) P. J. Shapiro, E. E. Bunel, W. P. Schaefer, J. E. Bercaw, Organo-
metallics 1990, 9, 867–876; e) S. Hajela, J. E. Bercaw, Organometal-
lics 1994, 13, 1147–1154; f) M. A. Giardello, Y. Yamamoto, L.
Brard, T. J. Marks, J. Am. Chem. Soc. 1995, 117, 3276–3277.
[7] a) E. E. Bunel, B. J. Burger, J. E. Bercaw, J. Am. Chem. Soc. 1988,
110, 976–981; b) W. E. Piers, P. J. Shapiro, E. E. Bunel, J. E. Bercaw,
Synlett 1990, 2, 74–81.
A
N
1H), 2.75–2.82 ppm (m, 1H); 13C{1H} NMR (C6D6, 100 MHz, 258C): d=
11.4, 23.4, 23.6, 25.9, 29.0, 36.7, 38.0, 42.9, 59.9 ppm; MS (EI): m/z (%):
167 [M]+ (14), 138 [MꢀC2H5]+ (100).
3-Propyl-2-azaspiro
258C): d=0.89 (t, J
2H), 1.98 (s, 2H), 2.12 (t,
(1H,1H)=1.72 Hz, 2H); 13C{1H} NMR (C6D6, 100 MHz, 258C): d=14.2,
[4.5]dec-2-ene (7b): 1HNMR (C 6D6, 400 MHz,
ACHTREUNG
J
ACHTREUNG
J
ACHTREUNG
19.9, 24.0, 26.2, 36.2, 37.4, 42.4, 50.0, 73.0, 175.1 ppm; MS (EI): m/z (%):
179 [M]+ (44), 151 [MꢀC2H4]+ (100), 136 [MꢀC3H7]+ (15).
[8] G. Jeske, H. Lauke, H. Mauermann, P. N. Swepston, H. Schumann,
T. J. Marks, J. Am. Chem. Soc. 1985, 107, 8091–8103.
2-(Phenylsilyl)-3-propyl-2-azaspiro
400 MHz, 258C): d=0.79 (t, J
1.31 (m, 12H), 1.62 (m, 3H), 2.73 (d, J
C
AHCTREUNG
[9] a) M. R. GagnØ, T. J. Marks, J. Am. Chem. Soc. 1989, 111, 4108–
4109; b) M. R. GagnØ, C. L. Stern, T. J. Marks, J. Am. Chem. Soc.
1992, 114, 275–294; c) Y. Li, T. J. Marks, J. Am. Chem. Soc. 1996,
118, 9295–9306; d) P. W. Roesky, C. L. Stern, T. J. Marks, Organo-
metallics 1997, 16, 4705–4711; e) Y. Li, T. J. Marks, J. Am. Chem.
Soc. 1998, 120, 1757–1771; f) V. M. Arredondo, F. E. McDonald,
T. J. Marks, J. Am. Chem. Soc. 1998, 120, 4871–4872; g) V. M. Arre-
dondo, S. Tian, F. E. McDonald, T. J. Marks, J. Am. Chem. Soc.
1999, 121, 3633–3639; h) S. Hong, T. J. Marks, J. Am. Chem. Soc.
2002, 124, 7886–7887; i) J.-S. Ryu, G. Y. Li, T. J. Marks, J. Am.
Chem. Soc. 2003, 125, 12584–12605.
[10] a) S. P. Nolan, M. Porchia, T. J. Marks, Organometallics 1991, 10,
1450–1457; b) T. Sakakura, H.-J. Lautenschläger, M. Tanaka, J.
Chem. Soc. Chem. Commun. 1991, 40–41; c) G. A. Molander, P. J.
Nichols, J. Am. Chem. Soc. 1995, 117, 4414–4416; d) G. A. Moland-
er, W. A. Retsch, Organometallics 1995, 14, 4570–4575.
[11] P.-F. Fu, L. Brard, Y. Li, T. J. Marks, J. Am. Chem. Soc. 1995, 117,
7157–7168.
[12] a) G. A. Molander, D. Pfeiffer, Org. Lett. 2001, 3, 361–363; b) K. N.
Harrison, T. J. Marks, J. Am. Chem. Soc. 1992, 114, 9220–9221;
c) E. A. Bijpost, R. Duchateau, J. H. Teuben, J. Mol. Catal. A 1995,
95, 121–128.
[13] a) M. R. Douglass, T. J. Marks, J. Am. Chem. Soc. 2000, 122, 1824–
1825; b) M. R. Douglass, C. L. Stern, T. J. Marks, J. Am. Chem. Soc.
2001, 123, 10221–10238.
[14] a) M. R. Bürgstein, H. Berberich, P. W. Roesky, Organometallics
1998, 17, 1452–1454; b) M. R. Bürgstein, H. Berberich, P. W.
Roesky, Chem. Eur. J. 2001, 7, 3078–3085.
[15] a) Y. K. Kim, T. Livinghouse, J. E. Bercaw, Tetrahedron Lett. 2001,
42, 2933–2935; b) Y. K. Kim, T. Livinghouse, Angew. Chem. 2002,
114, 3797–3799; Angew. Chem. Int. Ed. 2002, 41, 3645–3647;
c) Y. K. Kim, T. Livinghouse, Y. Horino, J. Am. Chem. Soc. 2003,
125, 9560–9561.
G
(m, 5H; Ph); 13C{1H} NMR (C6D6, 100 MHz, 258C): d=14.5, 19.9, 23.5,
24.5, 26.7, 35.4, 37.3, 40.8, 44.2, 59.3, 128.3, 130.0, 135.1, 136.0 ppm.
3-Propyl-2-azaspiro
d=0.87 (t, J
(1H,1H)=7.2 Hz, 3H), 1.18–1.31 (m, 13H), 1.43 (m, 4H),
1.50 (m, 1H), 2.48 (d, J
(1H,1H)=10.72 Hz, 1H), 2.67 (d, J(1H,1H)=
A
ACHTREUNG
A
ACHTREUNG
10.72 Hz, 1H), 2.84–2.91 ppm (m, 1H); 13C{1H} NMR (C6D6, 100 MHz,
258C): d=14.5, 20.8, 23.9, 24.2, 26.4, 37.1, 38.4, 38.7, 43.1, 58.4 ppm; MS
(EI): m/z (%): 181 [M]+ (12), 138 [MꢀC3H7]+ (100).
2-Methyl-1-(phenylsilyl)piperidine (19c): 1HNMR (C 6D6, 400 MHz,
258C): d=1.05 (d, J
A
1.48 (m, 2H), 2.70 (m, 1H), 2.94 (m, 1H), 3.08 (m, 1H), 5.09 (d,
J
(1H,1H)=2.80 Hz, 2H; PhSiH2), 7.10–7.58 ppm (m, 5H; Ph);
R
13C{1H} NMR (C6D6, 100 MHz, 258C): d=20.4, 23.1, 27.6, 34.3, 47.0, 51.9,
128.3, 129.9, 134.7, 136.0 ppm.
X-raycrystallographic studies of 1d and 2a–e : Crystals of 1d suitable for
X-ray crystallography were obtained from a concentrated THF solution.
Crystals of 2a–e were grown from saturated pentane solutions. A suitable
crystal was in each case covered in mineral oil (Aldrich) and mounted
onto a glass fiber. The crystal was transferred directly to the ꢀ738C cold
N2 stream of a Stoe IPDS 2T or a Bruker CCD Apex 1000 diffractometer.
Subsequent computations were carried out on an Intel Pentium IV PC.
CCDC-624685
(1d),
-286336
(2b),
-624686
(2a),
-624687 (2c), -624688 (2d), and -624689 (2e) contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgements
[16] P. N. OꢁShaughnessy, P. D. Knight, C. Morton, K. M. Gillespie, P.
Scott, Chem. Commun. 2003, 1770–1771.
This work was supported by the Deutsche Forschungsgemeinschaft.
[17] a) D. V. Gribkov, K. C. Hultzsch, F. Hampel, Chem. Eur. J. 2003, 9,
4796–4810; b) K. C. Hultzsch, D. V. Gribkov, Chem. Commun. 2004,
730–731; c) K. C. Hultzsch, F. Hampel, T. Wagner, Organometallics
2004, 23, 2601–2612.
[18] S. Hong, S. Tian, M. V. Metz, T. J. Marks, J. Am. Chem. Soc. 2003,
125, 14768–14783.
[1] Review: a) F. T. Edelmann in Comprehensive Organometallic Chem-
istry II, Vol. 4 (Eds.: E. W. Abel, F. G. A. Stone, G. Wilkinson), Per-
gamon Press, Oxford, 1995, pp. 11–210; b) F. T. Edelmann, Top.
Curr. Chem. 1996, 179, 247–276; c) G. A. Molander, Top. Organo-
met. Chem. 1999, 2, 119–154; d) G. A. Molander, J. A. C. Romero,
Chem. Rev. 2002, 102, 2161.
[19] P. Imhoff, J. H. Guelpen, K. Vrieze, W. J. J. Smeets, A. L. Spek, C. J.
Elsevier, Inorg. Chim. Acta 1995, 235, 77–88.
[20] a) M. W. Avis, M. E. van der Boom, C. J. Elsevier, W. J. J. Smeets,
A. L. Spek, J. Organomet. Chem. 1997, 527, 263–276; b) M. W. Avis,
C. J. Elsevier, J. M. Ernsting, K. Vrieze, N. Veldman, A. L. Spek,
K. V. Katti, C. L. Barnes, Organometallics 1996, 15, 2376–2392;
c) M. W. Avis, K. Vrieze, H. Kooijman, N. Veldman, A. L. Spek,
C. J. Elsevier, Inorg. Chem. 1995, 34, 4092–4105; d) P. Imhoff, R.
van Asselt, J. M. Ernsting, K. Vrieze, C. J. Elsevier, W. J. J. Smeets,
A. L. Spek, A. P. M. Kentgens, Organometallics 1993, 12, 1523–
1536.
[2] W. J. Evans, I. Bloom, W. E. Hunter, J. L. Atwood, J. Am. Chem.
Soc. 1983, 105, 1401–1403.
[3] a) H. Mauermann, P. N. Swepston, T. J. Marks, Organometallics
1985, 4, 200–202; b) G. Jeske, H. Lauke, H. Mauermann, H. Schu-
mann, T. J. Marks, J. Am. Chem. Soc. 1985, 107, 8111–8118.
[4] a) V. P. Conticello, L. Brard, M. A. Giardello, Y. Tsuji, M. Sabat,
C. L. Stern, T. J. Marks, J. Am. Chem. Soc. 1992, 114, 2761–2762;
b) D. Stern, M. Sabat, T. J. Marks, J. Am. Chem. Soc. 1992, 114,
9558–9575; c) C. M. Haar, C. L. Stern, T. J. Marks, Organometallics
Chem. Eur. J. 2007, 13, 3606 – 3616
ꢀ 2007 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
3615

P. W. Roesky et al.
[21] M. T. Gamer, P. W. Roesky, Z. Anorg. Allg. Chem. 2001, 627, 877–
881.
[35] G. W. Rabe, G. P. A. Yap, Z. Kristallogr. New Cryst. Struct. 2000,
215, 457–458.
[22] R. P. Kamalesh Babu, K. Aparna, R. McDonald, R. G. Cavell, Inorg.
Chem. 2000, 39, 4981–4984.
[36] A. Kasani, M. Furguson, R. G. Cavell, J. Am. Chem. Soc. 2000, 122,
726–727.
[23] a) C. M. Ong, P. McKarns, D. W. Stephan, Organometallics 1999, 18,
4197–4208; b) P. Wei, D. W. Stephan, Organometallics 2002, 21,
1308–1310; c) P. Wei, D. W. Stephan, Organometallics 2003, 22,
601–604.
[37] M. G. Kimpel, H. W. Gçrlitzer, M. Tafipolsky, M. Spiegler, W. Scher-
er, R. Anwander, J. Organomet. Chem. 2002, 647, 236–244.
[38] J. Eppinger, M. Spiegler, W. Hieringer, W. A. Herrmann, R. Anwan-
der, J. Am. Chem. Soc. 2000, 122, 3080–3096.
[24] a) A. Kasani, R. P. Kamalesh Babu, R. McDonald, R. G. Cavell, Or-
ganometallics 1999, 18, 3775–3777; b) K. Aparna, R. McDonald, M.
Fuerguson, R. G. Cavell, Organometallics 1999, 18, 4241–4243.
[25] M. J. Sarsfield, H. Steele, M. Helliwell, S. J. Teat, Dalton Trans.
2003, 3443–3449.
[26] W.-P. Leung, C.-W. So, J.-Z. Wang, T. C. W. Mak, Chem. Commun.
2003, 248–249.
[27] A related ligand: a) M. S. Hill, P. B. Hitchcock, Dalton Trans. 2003,
4570–4571; b) M. S. Hill, P. B. Hitchcock, Chem. Commun. 2003,
1758–1759.
[39] R. D. Shannon, C. T. Prewitt, Acta Crystallogr. Sect. B 1969 25, 925–
945.
[40] C. M. Ong, D. W. Stephan, J. Am. Chem. Soc. 1999, 121, 2939–2940.
[41] a) R. M. Beesley, C. K. Ingold, J. F. Thorpe, J. Chem. Soc. 1915, 107,
1080–1106; b) C. K. Ingold, J. Chem. Soc. 1921, 119, 305–329;
c) A. J. Kirby, Adv. Phys. Org. Chem. 1980, 17, 183–278; d) M. E.
Jung, J. Gervay, J. Am. Chem. Soc. 1991, 113, 224–232.
[42] P. N. OꢁShaughnessy, P. Scott, Tetrahedron: Asymmetry 2003, 14,
1979–1983.
[43] B. Marciniec, Comprehensive Handbook on Hydrosilylation, Perga-
mon Press, Oxford, 1992.
[44] A. A. Tifonov, T. Spaniol, J. Okuda, Dalton Trans. 2004, 2245–2250.
[45] Y. Horino, T. Livinghouse, Organometallics 2004, 23, 12–14.
[46] Y. Obora, T. Ohta, C. L. Stern, T. J. Marks, J. Am. Chem. Soc. 1997,
119, 3745–3755.
[28] M. T. Gamer, S. Dehnen, P. W. Roesky, Organometallics 2001, 20,
4230–4236.
[29] a) A. Zulys, T. K. Panda, M. T. Gamer, P. W. Roesky, Chem.
Commun. 2004, 2584–2585; b) A. Zulys, T. K. Panda, M. T. Gamer,
P. W. Roesky, Organometallics 2005, 24, 2197–2202.
[30] M. T. Gamer, P. W. Roesky, M. Rastätter, A. Steffens, M. Glanz,
Chem. Eur. J. 2005, 11, 3165–3172.
[47] K. Takaki, T. Kamata, Y. Miura, T. Shishido, K. Takehira, J. Org.
Chem. 1999, 64, 3891–3895.
[31] A. Heutling, F. Pohlki, I. Bytschkov, S. Doye, Angew. Chem. 2005,
117, 3011–3013; Angew. Chem. Int. Ed. 2005, 44, 2951–2954.
[32] M. Rastätter, A. Zulys, P. W. Roesky, Chem. Commun. 2006, 874–
876.
[48] L. D. Field, B. A. Messerle, S. L. Wren, Organometallics 2003, 22,
4393–4395.
[49] J. Eppinger, E. Herdtweck, R. Anwander, Polyhedron 1998, 17,
1195–1201.
[33] R. Anwander, O. Runte, J. Eppinger, G. Gerstberger, E. Herdtweck,
M. Spiegler, J. Chem. Soc. Dalton Trans. 1998, 847–858.
[34] W. A. Herrmann, R. Anwander, F. C. Munck, W. Scherer, V.
Dufaud, N. W. Huber, G. R. J. Artus, Z. Naturforsch. B 1994, 49,
1789–1797.
[50] M. D. Taylor, C. P. Carter, J. Inorg. Nucl. Chem. 1962, 24, 387–391.
[51] R. Appel, I. Ruppert, Z. Anorg. Allg. Chem. 1974, 406, 131–144.
Received: October 24, 2006
Published online: March 20, 2007
3616
www.chemeurj.org
ꢀ 2007 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 3606 – 3616