Reaction of hydrogenosilanes with dimethyl Pt(II) complexes promoted by hemilabile P,N-chelating ligands

Article abstract of DOI:10.1016/S0022-328X(03)00732-0

Reaction of [(κ²-P,N)-Me₂N(CH₂) ₃PPh₂]PtMe₂ [(P∩N)PtMe₂] with HSiR₃ (triethylsilane or diphenylmethylsilane) results in the formation of MeSiR₃ and the meth

Full text of DOI:10.1016/S0022-328X(03)00732-0

Journal of Organometallic Chemistry 686 (2003) 183ꢀ191
/
www.elsevier.com/locate/jorganchem
Reaction of hydrogenosilanes with dimethyl Pt(II) complexes
promoted by hemilabile P,N-chelating ligands
Susan M. Thompson, Frank Stohr, Dietmar Sturmayr, Guido Kickelbick,
¨
Ulrich Schubert *
Institute of Materials Chemistry, Vienna University of Technology, Getreidemarkt 9, A-1060 Wien, Austria
Received 3 April 2003; received in revised form 24 July 2003; accepted 24 July 2003
Abstract
Reaction of [(k2-P,N)ꢀ
formation of MeSiR3 and the methyl silyl complexes (PS
(PS N)Pt(Me)H is additionally observed. In the reaction of HSiMePh2, the rearranged complex (PS
formed. The reactivity of the complexes (PS N)PtMe2 is influenced by the kind of chelating PS N ligand and appears to depend on
/
Me2N(CH2)3PPh2]PtMe2 [(PS
/
N)PtMe2] with HSiR3 (triethylsilane or diphenylmethylsilane) results in the
N)Pt(Me)SiR3. In the early stages of the reaction, the hydrido complex
N)Pt(Ph)SiMe2Ph is also
/
/
/
/
/
how easily the amino group is de-coordinated.
# 2003 Elsevier B.V. All rights reserved.
Keywords: Hemilabile ligands; Methyl/silyl exchange; Oxidative addition; Metal silyl complexes
1. Introduction
ample, does not react with HSiR3 at ambient conditions,
(PS N)PtMe2 readily reacts with trimethoxysilane [5]
/
The reactivity of Pt(II) complexes towards organosi-
lanes is greatly enhanced when hemilabile chelating
(Eq. (1a)) or 1,2-bis(dimethylsilyl)benzene [6] (Eq. (1b)).
Reaction with the latter silane yielded the corresponding
chelated bis(silyl) complex. Noteworthy was the fact
that for the complete reaction of the dimethyl complexes
two equivalents of 1,2-bis(dimethylsilyl)benzene must be
employed. While one equivalent of the silane is neces-
sary to form the bis(silyl) ligand, the second equivalent
turns up as 1-dimethylsilyl-2-trimethylsilyl-benzene, i.e.
one methyl ligand of the starting complex is eliminated
as the methylsilane and the other as methane. The
formation of the corresponding bis(silyl) complex and
methylsilane was also observed in the reaction of
ligands R2N-R?ꢀ
/
PPh2 (PS N) are employed, compared
/
with the corresponding bis(phosphine) complexes, and
several new catalytic and stoichiometric reactions have
been observed [1]. For example, PS
complexes activate both CꢀCl and Siꢀ
catalyze the formation of disiloxanes from HSiR3 [3]
and CꢀCl/SiꢀH exchange reactions [4]. The correspond-
/N-substituted Pt(II)
/
/
Cl bonds [2] and
/
/
ing bis(phosphine) complexes undergo none of these
reactions.
A key entry to metal silyl complexes are oxidative
addition reactions of hydrogenosilanes. While (dppe)-
[(k2-P,N)ꢀ
/Me2NCH2CH2PPh2]PtMe2 with HSi(OMe)3,
but the reaction was more complex: (i) only about
half of the starting complex was converted to the
PtMe2 (dppeꢁbis(diphenylphosphino)ethane), for ex-
/
bis(silyl) complex (PS
mainder turned up as the mono-substituted complex
(PS N)Pt[Si(OMe)3]Me. This implies that the first
methyl/silyl exchange reaction at the platinum center is
/
N)Pt[Si(OMe)3]2, while the re-
* Corresponding author. Tel.: ꢂ
58801-15399.
E-mail address: uschuber@mail.zserv.tuwien.ac.at (U. Schubert).
/
43-1-58801-15320; fax: ꢂ43-1-
/
/
0022-328X/03/$ - see front matter # 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0022-328X(03)00732-0

184
S.M. Thompson et al. / Journal of Organometallic Chemistry 686 (2003) 183ꢀ191
/
faster than the second: (ii) when an excess of HSi(OMe)3
was employed, the hydrogenosilane was catalytically
converted into Si(OMe)4.
2.1. Reaction of 1 with HSiEt3
The first silane investigated was HSiEt3 (Eq. (2)). On
(1a)
(1b)
The reactions of trimethoxysilane and 1,2-bis(di-
methylsilyl)benzene were promoted by the high ten-
dency of both silanes towards oxidative addition
addition of one equivalent of HSiEt3 to 1 at room
temperature, the 29Si-NMR spectrum showed the im-
mediate formation of MeSiEt3, and the 31P-NMR
spectrum the presence of complex 2 (see below). About
13% of the phosphorous species seen in the NMR
spectrum was complex 2, and the rest was the reactant
1 (compared with an internal standard which did not
partake in the reaction) [8]. Complex 2 was only
observed during the first hour of reaction.
reactions, due to the activated SiꢀH bond in the case
/
of HSi(OEt)3 and the chelate-assistance in the case of
1,2-bis(dimethylsilyl)benzene, respectively. In the work
presented in this article, the reactions of (PS
/
N)PtMe2
(1) (PS NÄPh2P(CH2)3NMe2) with the less reactive
/
/
silanes diphenylmethylsilane and triethylsilane were
investigated. The course of the reactions was analyzed
by multi-nuclear NMR spectroscopy. Since we were
mainly interested in elucidating the reaction mechanism,
no attempts were made to isolate the products. All
reactions were carried out in a poly(tetrafluoroethylene)
(PTFE) liner to exclude the possibility of metal cata-
Only platinum satellites (1JPtP
2JSiPtP coupling was seen in the 31P-NMR spectrum of
ꢁ/2208 Hz) but no
2, and the chemical shift (dꢁ14.9) was very similar to
/
that of 1. No new 29Si-NMR signal was observed. The
platinum-bonded hydrogen was identified in the 1H-
NMR spectrum with the characteristic upfield shift at
1
17 Hz and JPtH
ꢃ
/
18.9 ppm (2JPH
ꢁ
/
ꢁ1442). Further-
/
lyzed reactions of the silane with adsorbed water or Siꢀ
/
OH groups on the glass surface [7].
more, 2D HP-HMBC experiments showed a coupling
between this proton and the corresponding phosphorus
signal. The intensity of this signal was found to be only
approximately one-twentieth of the Siꢀ/H proton of
unreacted HSiEt3 after 30 min reaction. The Ptꢀ
/
Me
2. Results and discussion
signal could not be identified in the 13C-NMR spectrum
due to the many overlapping signals in the relevant
region. Because the spectroscopic data of this complex
were the same independent of the employed silane (see
As in the reactions of trimethoxysilane or 1,2-bis(di-
methylsilyl)benzene, no reaction was seen between
(PhMe2P)2PtMe2 or (dppe)2PtMe2 and HSiEt3 or HSi-
MePh2. In contrast, when the neat silanes, in various
stoichiometric ratios were added to benzene solutions of
1 at room temperature, slight gas evolution was
observed, and NMR spectra taken immediately after
silane addition indicated an instantaneous start of the
reactions.
below), we postulate that 2 is (PS N)Pt(H)Me. Further
/
evidence is given later in the text. The cis arrangement
of P and H was also supported by DFT calculations (see
below).
A second new complex with a 31P-NMR chemical
shift of 23 ppm and typical platinum satellites was first

S.M. Thompson et al. / Journal of Organometallic Chemistry 686 (2003) 183ꢀ
/191
185
seen after about 45 min reaction at room temperature
and was assigned to the methyl silyl complex (PS
N)Pt(Me)SiEt3 (3a). The stereochemistry of 3a was
confirmed earlier for related complexes [5,9] and can
be explained by the different trans influence of the
nitrogen and phosphorus donor centers.
undergo reaction with a further hydrogenosilane to
give the methyl silyl complex 3a by hydrogen elimina-
tion. This mechanistic suggestion is supported by the
observation that 2 is only present in approximately the
first hour of reaction, after which it was no longer seen
(Scheme 1).
/
(2)
After 4 h of reaction, the strongest 31P-NMR signal
was still that of the reactant complex 1. Small amounts
The proposed reaction pathway was also investigated
by DFT calculations. The two reaction steps shown in
Scheme 2 (energies in kcal molꢃ1), i.e. the formation of
2 from 1 and the formation of 3 from 2, were calculated
with the necessary simplifications to give model systems
suitable for DFT calculations. Each of the shown
structures were optimized, indicating that the structures
are stable. The corresponding energies are also given for
each step.
of de-coordinated PS N ligand, as well as other
/
phosphorus containing species were also evident besides
complex 3a. At this point, HSiEt3 was only detected in
traces in the 29Si-NMR spectrum. Both complex 1 and
the silane were completely consumed after 24 h at room
temperature.
The sequence in which the various compounds
appeared and disappeared was essentially the same
when 1 was reacted with a 20-fold excess of HSiEt3.
Complex 2 and MeSiEt3 were immediately visible in the
NMR spectra. After 4 h of reaction the portion of 2 was
slightly higher (23%), and 1 was completely consumed
after 16 h. Complex 2 was again only identified in the
first hour of reaction. Complex 3a was first identified
within 1 h reaction at room temperature and was present
in approximately the same magnitude as 1 after 3.25 h.
After 18 h, 3a was the only platinum complex present.
Contrary to the 1:1 reaction, there was no evidence of
other phosphorus species. Due to the large excess and
our assumption that the reaction is stoichiometric,
HSiEt3 was never completely consumed. After reaction
times of longer than 3 h, traces of other silicon species
were observed in the 29Si-NMR spectrum besides
HSiEt3 and MeSiEt3. Because of the large excess of
HSiEt3 the other silicon-containing species were difficult
to assign.
Overall it can be seen that both reactions are
9.7 kcal molꢃ1, and
exothermic, reaction (a) with ꢃ
/
reaction (b) with ꢃ
/
6.9 kcal molꢃ1. In each case a six-
coordinate intermediate was also found and its geometry
optimized.
2.2. Reaction of 1 with HSiMePh2
When an equimolar amount of HSiMePh2 was added
(Eq. (3)), complex 1 was not completely consumed
within 30 h of reaction at room temperature, nor with
heating for 2 h at 60 8C. The 29Si- and 1H-NMR spectra
showed the immediate formation of the methylsilane,
Me2SiPh2. The first new platinum complex observed in
the 31P-NMR spectrum immediately after addition of
the silane was again 2. Complex 2 was only visible for
the first hour of the reaction at room temperature. Here
again the platinum bonded hydrogen in complex 2 was
1
identified in the H-NMR spectrum (ꢃ
/
18.9 ppm), with
The experimental findings suggest the following
reaction mechanism. The incoming silane, with the
strongest trans effect of all the ligands involved, causes
phosphorus coupling and platinum satellites. The other
signal (a doublet at 24 ppm with platinum satellites),
also visible in the first 31P-NMR spectrum taken after
addition of the silane, was assigned to the methyl silyl
complex 3b. The coupling constant of ca. 11 Hz is
characteristic for phosphorus silicon coupling, and
the nitrogen atom of the PS N ligand to de-coordinate.
/
Elimination of methylsilane would give the observed
methyl hydride complex. This complex then could
Scheme 1.

186
S.M. Thompson et al. / Journal of Organometallic Chemistry 686 (2003) 183ꢀ
/191
Scheme 2.
being a doublet indicates that only one equivalent silicon
1
is present (Fig. 1). The JPtP coupling constant was also
after 1 h. Further peaks were clearly identified in the
29Si-NMR spectrum at ꢃ
throughout the complete reaction period. These were
/
10.6 ppm and ꢃ21.7 ppm
/
within the expected range for such complexes.
(3)
However, contrary to the HSiEt3 reaction, a third new
complex (4) was also formed. Complex 4, found at 28
ppm in the 31P-NMR spectrum, also had a doublet with
virtually the same coupling constant as 3b, and platinum
satellites with a coupling constant of 1597Hz. This
complex was first observed after 4 h at room tempera-
ture while the reactant platinum complex 1. The
spectroscopic data strongly suggest that 4 has the
composition and structure as shown in Eq. (3) (see
below). Although further smaller unidentified phospho-
rus peaks were found, no free P,N-chelated ligand was
observed. Complexes 3b and 4 were the only complexes
present at the end of the reaction.
assigned to the silyl ligands of 3b and 4. No clear
platinum satellites were visible. However two-dimen-
sional spectra indicated coupling with platinum, and
literature values support the assignment.
When 1 and HSiMePh2 were reacted in a 1:20 ratio,
the outcome of the reaction was essentially the same.
Me2SiPh2 was again formed immediately. The 31P-
NMR spectrum showed that 1 was consumed within
15 min reaction at room temperature. Complex 2 was
only identified in the first spectrum taken and was
consumed within 15 min reaction at room temperature.
Both complex 3b and 4 were seen immediately and still
present at the end of the reaction. After 5 h reaction at
room temperature, and with additional heating at 60 8C,
3b and 4 were the only phosphorus species present in
approximately the same amount. The silane, being used
1
Only traces of HSiMePh2 were seen in the H-NMR
spectrum after reaction for 5 h at room temperature.
After heating at 60 8C the reactant silane had vanished
Fig. 1. Section from the 31P-NMR spectrum of the reaction mixture between 1 and HSiMePh2 (1:20) after 2 h reaction at room temperature.

S.M. Thompson et al. / Journal of Organometallic Chemistry 686 (2003) 183ꢀ
/191
187
in excess, was not completely consumed, even after
heating for 5 h at 60 8C.
cannot be integrated, it was still possible to qualitatively
analyze the spectra and compare the consumption of the
reactant platinum complex and the formation of the
methylsilane.
When comparing the complete consumption of the
reactant platinum complex (through the 31P-NMR
With regard to the identity of complex 4, it must be
emphasized, that due to the complexity of the NMR
spectra the exact identity of the silyl group bonded to
platinum is not completely certain. We cannot totally
rule out that the silyl ligand is SiHPh2 (though no signal
was found in the expected region in the 1H-NMR
spectra) or SiMe2Ph rather than SiMePh2. However,
as will be discussed in the following paragraph, the
mechanistic possibilities leading to SiHPh2 or SiMe2Ph
are rather unlikely to happen.
spectra) at room temperature, the order is 8B
hB7B5B13 hB6. Comparison of the intensity of the
Me2SiR2 signal in the 29Si-NMR spectra after 1 h at
room temperature gave the following order: 1ꢀ8ꢀ7ꢀ
5:6. Both series reflect similar trends and clearly show
that the employed PS N ligand has a significant
/
1B1
/
/
/
/
/
/
/
/
/
/
There are several possibilities how a complex of the
influence on the reactivity of the complexes. These series
also indicate that the formation of the methyl silane is
closely related to the consumption of the reactant
type (PS
some intermediate complex with the solvent (benzene)
can be ruled out, because no phenylꢀsilyl complex was
observed in the reaction of HSiEt3. A further possibility
is the direct addition of a SiꢀPh bond, either of the
/N)Pt(Ph)SiR3 could be formed. Reaction of
/
platinum complex. The PS N-chelating ligand is
/
thought to enable the reaction between the bis(methyl)
complexes and hydrogenosilanes either through the fact
that the methyl groups are differently activated, and/or
through the chelating ligand’s ability to reversibly de-
coordinate.
/
starting silane HSiMePh2 or the product silane Me2-
SiPh2 to some intermediate platinum complex. The first
possibility would result in the formation of (PS
N)Pt(Ph)SiHMePh and the second in the formation of
(PS N)Pt(Ph)SiMe2Ph. Activation of SiꢀPh bonds has
been observed upon reaction of Ph2PCH2CH2SiPh3ꢃn
Men (nꢁ0, 1) with (Ph3P)2Pt(p-C2H4) [10]. In the same
work was shown that the SiꢀMe bonds were not reactive
/
Complexes 1 and 5 have a three carbon chain between
the nitrogen and phosphorus donor centers, but the
phenyl ring in 5 restricts the flexibility of the ligand. The
other complexes, 6, 7 and 8, all have ethylene bridges
between the donor centers, but vary in the organyl
groups at the nitrogen center. From these results it can
be seen that the complexes in which the amino group de-
coordinates most easily had the fastest reaction rates. In
complex 1, the six-membered ring is less favorable than
the five-membered ring in the corresponding complex 6,
and in complex 8 the large iso-propyl groups provide
some steric hindrance. The stiffening of the chelate ring
in 5 results in a less facile de-coordination of the amino
group relative to 1.
/
/
-
/
/
enough to be added to the platinum center. This would
explain why 4 was observed in the reaction of HSi-
MePh2, but no analogous complex in the reaction of
HSiEt3. The possibility of a SiꢀPh oxidative addition is
/
nevertheless rather unlikely in the present case, because
complex 4 is already observed while unreacted HSiPh3 is
still present. Any reactive intermediate capable of
oxidative addition reactions would preferentially react
with the much more reactive Siꢀ/H bonds rather than
SiꢀPh groups.
/
This is, to some extent, also reflected in the structural
The most probable explanation is that 4 is formed by
a rearrangement of 3, i.e. that the composition of the
data of the (PS
and Figs. 2 and 3). In each case, the Ptꢀ
the range of the PtꢀP distance*or even longer*
despite the smaller radius of nitrogen compared to
phosphorus. The weak PtꢀN interaction results in a
strengthening of the PtꢀMe bond trans to N; the Ptꢀ
bond lengths trans to N are distinctly shorter than that
trans to P. The longest PtꢀN distance is found in 8,
probably because of the steric hindrance of the iso-
propyl groups, while the PtꢀN distance in 6 is shortest
in this series. An interesting structural feature with
regard to the reactivity is the large NꢀPtꢀP angle in 1,
/N)PtMe2 complexes 1, 6 and 8 (Table 1
/
N distance is in
complex is(PS N)Pt(Ph)SiMePh2. Scrambling reactions
/
/
/
/
of silicon substituents in metal silyl complexes are not
uncommon and are usually explained by the intermedi-
ate formation of silylene complexes [11]. Tilley et al.
have pointed out that transfer of a substituent from
silicon to platinum is easier in a three, rather than four-
coordinate complex [12]. Hence such a reaction may be
enabled through the presence of the hemilabile ligand.
An analogous reaction was also observed in the reaction
/
/
/C
/
/
of HSi(OMe)3 with PS N-chelated complexes [5].
/
/
/
which is more than 108 wider than in the five-membered
ring systems 6 and 8. This distortion might be the reason
for the higher reactivity of 1 compared to 6. However, a
word of caution is necessary at this point: The structural
variations among the complexes 1, 6 and 8 are more
complex, i.e. the reactivity differences cannot be traced
back to a single structural parameter. For example,
2.3. Comparison of the reactivity of the different P,N-
chelated platinum complexes
The reaction of HSiMePh2 was used to test the
reactivity of (PS
PS N ligands (Scheme 3). A 20-fold excess of the silane
was employed. Though the 31P- and 29Si-NMR spectra
/
N)PtMe2 complexes with different
/

188
S.M. Thompson et al. / Journal of Organometallic Chemistry 686 (2003) 183ꢀ
/191
Scheme 3.
Table 1
Typical bond lengths (in pm) and angles in complexes (PS N)PtMe2
/
PS
/
N
Ph2PCH2CH2NMe2 (6) [6]
Ph2PCH2CH2Ni Pr2 (8)
Ph2PCH2CH2CH2NMe2 (1)
Ptꢀ
Ptꢀ
Ptꢀ
Ptꢀ
Nꢀ
PꢀPtꢀ
NꢀPtꢀ
CꢀPtꢀ
/
C (trans P)
206.9(8)
203.7(9)
221.0(7)
224.3(2)
84.6(1)
96.5(2)
92.0(2)
87.0(3)
210.2(6)
204.5(7)
230.2(5)
224.6(2)
84.0(1)
93.2(3)
96.7(2)
86.0(3)
210.3(9)
203.4(10)
223.9(8)
224.2(2)
95.5(2)
/C (trans N)
/N
/P
/
Ptꢀ
C (cis)
C (cis)
/P
/
/
89.4(4)
/
/
91.0(4)
/
/C
84.1(5)
there is no straightforward explanation for the shorter
PtꢀC (trans to P) distance in 6.
reactivity of the latter silanes, substitution of the second
methyl ligand of the starting complex was not observed,
and the reaction rate of the first step was slowed down.
It appears that substitution of the second methyl group
requires significantly more energy, therefore it can be
inferred that a bis(silyl) complex can only be formed
with an incoming silane which is somehow activated,
which is not the case for diphenylmethylsilane or
triethylsilane. A possible explanation for the lower
reaction rate of the second methyl/silyl exchange could
be the trans effect of the involved ligands. Contrary to
the Pt(IV) intermediate for the first step (Scheme 2), the
second SiR3 group in the corresponding intermediate for
the second exchange must have a different trans ligand.
/
3. Conclusions
The mono-substitution of the platinum complex
occurred relatively easily in all cases irrespective of the
silane used. The order of reactivity was HSiMePh2ꢀ
/
HSiEt3 reflecting the inductive effect of the different
substituents at the silicon atom.
In previous work, we had employed trimethoxysilane
and 1,2-bis(dimethylsilyl)benzene for the reactions with
complexes of the type (PS N)PtMe2. Both silanes have a
/
higher tendency towards oxidative addition reactions
than the silanes used in this work, viz. diphenylmethyl-
silane and triethylsilane. As a result of the lower
Fig. 2. Molecular structure of [(k2-P,N)-Ph2PCH2CH2NiPr2]PtMe2
(8).
Fig.
3. Molecular
Ph2PCH2CH2CH2NMe2]PtMe2 (1).
structure
of
[(k2-P,N)ꢀ
/

S.M. Thompson et al. / Journal of Organometallic Chemistry 686 (2003) 183ꢀ
/191
189
This is obviously less favorable than the exchange of
first SiR3 ligand being trans to CH3, and therefore the
second substitution does not take place for non-acti-
vated silanes.
be particularly important to obtain selectivity in the
catalytic reactions previously observed.
4. Experimental
The stabilization of the mono-substituted complex
was previously also observed in the reaction of (PS
N)PtMe2 with HSnR3 (RꢁBu, Ph), where (PS
/
All reactions were carried out using standard Schlenk
techniques in an argon atmosphere. Argon was purified
˚
using Cu and molecular sieve (4 A). All solvents used
were dried and distilled at least once under argon. All
NMR measurements were carried out in deuterated
benzene (C6D6), the samples were filled and sealed in an
argon atmosphere. Benzene-d6 was pre-dried using a
/
/
N)Pt(SnR3)Me was a major product, along with
MeSnR3 [13]. The main difference between the reaction
of the stannanes and the silanes reported here is that
stannanes are more reactive. The reaction therefore
proceeds by formation of bis(stannyl) complexes and
distannanes.
The slower reaction of diphenylmethylsilane and
triethylsilane compared with trimethoxysilane, 1,2-
bis(dimethylsilyl)benzene and the stannanes allowed us
to study the first methyl/silyl exchange step in more
detail than was hitherto possible. When analyzing the
results obtained for the reaction of 1 with the hydro-
genosilanes, it seems that the methyl silyl complexes are
formed together with the corresponding methylsilane,
˚
molecular sieve (4 A) and then made oxygen-free using
the freezeꢀpumpꢀthaw method. The reactions were
/
/
carried out in a poly(tetra-fluoroethylene) (PTFE) liner
to exclude the possibility of metal catalyzed reactions of
the silane with adsorbed water or SiꢀOH groups of the
/
glass surface. Reactions were also carried out directly in
glass vessels, the results obtained here were as seen with
the use of a PTFE liner, but siloxane species were also
present.
i.e. that the first PtꢀMe ligand is eliminated as a
/
The complexes (PS N)PtMe2 were prepared by the
/
methylsilane. This is different to what we had previously
postulated (methane elimination in the first step) [5]. A
strong evidence for the methylsilane-elimination path-
way in the reactions of the silanes investigated in the
present work, apart from the observed products, is that
1 was never completely consumed in the 1:1 reactions,
while the starting hydrogenosilane was. Thus, more than
one equivalent of the silane is obviously needed for the
formation of the methyl silyl complexes.
Although we did not observe clear gas evolution when
1 and the silanes were reacted in a 1:1 molar ratio, a
competing methane elimination cannot be excluded, at
least as a minor parallel reaction. A competition
between methylsilane and methane elimination from
intermediates containing hydrido, alkyl and silyl ligands
previously described general procedure [6].
NMR experiments were taken on Bruker Avance 300
and 250 MHz spectrometers at room temperature (r.t.),
unless otherwise stated, using an external standard;
1
TMS, 85% H3PO4 or Na2Pt(H2O)6, for H-, 31P- and
195Pt-NMR, respectively. All pulse programs used were
obtained from the standard Bruker software library.
31P-, 195Pt- and 13C-NMR spectra were proton de-
coupled, 29Si was measured using a INEPT sequence,
or through 2D 1H29Si correlation technique HMBC.
Further 2D experiments were performed using a HMBC
technique for 1H31P and 1H195Pt correlations to aid
peak assignments. Although 13C-NMR spectra were
also taken, these data are not reported here, because the
spectra are very complex and do not allow an unequi-
vocal assignment of the peaks.
(and the ratio of both reactions depending on the nature
of the silyl group) was already observed for other
complexes, such as mer-(Me3P)3Ir(SiR3)(H)Me [14] or
ꢄ
The NMR signals seen throughout the course of the
reaction are given in the following paragraphs, including
those which are only seen at specific reaction times.
Further specific information is given in the main text.
For these reasons chemical shifts are given without
integration, as this varies throughout the course of the
reaction. For clarity, the key NMR data are summarized
in Table 2.
(Cp2ScMeꢂMesSiH3) [15].
/
If the first methyl/silyl exchange of (PS N)PtMe2
/
proceeds by methylsilane elimination, then the second
exchange must proceed by methane elimination, because
both methane and methylsilane were clearly identified in
the experiments with trimethoxysilane and 1,2-bis(di-
methylsilyl)benzene (see Eq. (1)) (and because one
methylꢀsilyl group per bis(silyl) complex is formed).
/
4.1. Computational details
This difference between the first and the second
exchange step is probably again due to the different
trans effects in the intermediate Pt(IV) complexes.
The geometry optimizations of the equilibrium geo-
metries and transition structures were performed using
the B3LYP version of DFT, which is comprised of
Becke’s hybrid three-parameter exchange functional and
the correlation functional of Becke, Lee and co-workers
[16,17]. The LanL2DZ basis set implemented in the
The reactions of HSiMePh2 with the complexes (PS
N)PtMe2 having different PS N ligands showed that the
proper choice of the PS N ligand allows to tailor the
reactivity of the complexes towards silanes, which could
/
/
/

190
S.M. Thompson et al. / Journal of Organometallic Chemistry 686 (2003) 183ꢀ
/191
Table 2
Summary of relevant NMR data
Complex/silane
d(31P)
d(29Si)
d(1H) Hꢀ
/Si or Hꢀ/Pt
1
15.23 (1JPtP
14.91 (1JPtP
23.04 (1JPtP
ꢁ
/
2089 Hz)
2208 Hz)
2850 Hz)
ꢀ
ꢀ/
15.7
10.6
21.6
/
ꢀ
ꢃ
ꢀ/
/
1
17 Hz, JPtH
2
3a
ꢁ
/
/
18.22 (2JPH
ꢁ
/
ꢁ/1442 Hz)
ꢁ/
1
3b
4
24.17 (2JSiPtP
28.23 (2JSiPtP
ꢁ
ꢁ
/
10.7 Hz, JPtP
1
10.7 Hz, JPtP
ꢁ
ꢁ/
/
2494 Hz)
ꢃ
ꢃ/
/
ꢀ/
ꢀ/
/
1597 Hz)
1
4 Hz, JSiH
HSiEt3
MeSiEt3
HSiPh2Me
MeSiPh2Me
ꢀ/
ꢀ/
ꢀ/
ꢀ/
0.05
6.6
3.98 (2JSiCH
ꢁ
/
ꢁ/
35 Hz)
ꢀ/
1
4Hz, JSiH
ꢃ/
17.7
4.94 (2JSiCH
ꢁ
/
ꢁ/41 Hz)
7.97
ꢀ/
GAUSSIAN-98 package [18], was used an all atoms for the
calculations. Vibrational mode analyses were carried out
to confirm that on potential energy surfaces optimized
geometries correspond to a local minimum without an
imaginary frequency mode or a saddle point with only
one imaginary mode. Corrections of zero-point vibra-
tional energies were taken into account in stationary
structures obtained. All of the calculations were carried
out with the GAUSSIAN-98 package of electronic struc-
ture programs [18].
was dark brown, and a brown solid was present. After
24 h at r.t. the sample was heated for 60 min at 60 8C.
Excess of silane: 80 mg (0.48 mmol) of HSiPh2Me was
added at r.t. to 10 mg (0.002 mmol) of 1 in 0.4 ml of
C6D6. On addition of the silane, the sample immediately
turned yellow. With increasing reaction time the sample
became darker, and a brown solid was present. After 5 h
at r.t. the sample was heated for 90 min, and NMR
spectra were taken.
1H-NMR: dꢁ
/
ꢃ
/
18.9 (d with satellites, 2JPH
ꢁ17 Hz,
/
2
1JPtH
4Hz, JSiH
ꢁ
/
1482 Hz, 2), 4.94 (q, with satellites, J(SiCH)ꢁ
/
4.2. Reaction of [(k2-P,N)ꢀ
(1) with HSiEt3
/
Me2N(CH2)3PPh2]PtMe2
ꢁ
/
41 Hz, Hꢀ
/
Si); 31P{1H}-NMR: dꢁ
/28.23
1
2
(d, with satellites, JSiPtP
1
10.7 Hz, JPtP
ꢁ
/
ꢁ
/
1597 Hz, 4),
2
1
24.17 (d, with satellites, JSiPtP
ꢁ
/
10.7 Hz, JPtP
ꢁ
/
2494
2089 Hz, 1),
2170 Hz, 2); 29Si-NMR:
7.97 (s, MeSiPh2Me), ꢃ10.6 (s, SiꢀPt, 3b), ꢃ
17.7 (s, HSiMePh2), ꢃ21.6 (s, Siꢀ
Pt, 4); 195Pt-NMR:
dꢁ
4790 (d, 1JPtP
Hz, 4), ꢃ4073 (d, JPtP
1
1:1 molar ratio: 3.04 mg (0.03 mmol) of HSiEt3 was
added at r.t. to 13 mg (0.03 mmol) of 1 in 0.5 ml of
C6D6. On addition of the silane the reaction mixture
turned yellow. With increasing reaction time the sample
became darker in color, and after 24 h at r.t. the sample
was dark brown.
Excess of silane: 56 mg (0.48 mmol) of HSiEt3 was
added at r.t. to 12 mg (0.02 mmol) of 1 in 0.5 ml of
C6D6. On addition of the silane, gas evolution was
observed and the sample turned yellow. After 2 h at r.t.
Hz, 3b), 15.23 (s, with satellites, JPtP
ꢁ
/
1
14.91 (s, with satellites, JPtP
ꢁ
/
dꢁ
/
ꢃ
/
/
/
/
/
/
/
ꢃ
/
ꢁ
/
2260 Hz, 3b), ꢃ
/
4412 (d, 1JPtP
2056 Hz, 1).
ꢁ2164
/
1
/
ꢁ
/
4.4. X-ray structure analyses of 1 and 8
the sample was orange/brown, with some brown solid.
1H-NMR: dꢁ
Data collection (Table 3): the crystals were mounted
on a Siemens SMART diffractometer (area detector).
2
18.22, (d with satellites, JPH
/
ꢃ
/
ꢁ17
/
1
Hz, JPtH
2JSiCH
4 Hz, JSiH
dꢁ23.04 (s, with satellites, JPtP
(s, with satellites, JPtP
ꢁ
/
1442 Hz, 2), 3.98 (quin, with satellites,
Moꢀ
/
Ka radiation (lꢁ71.069 pm, graphite monochro-
/
1
ꢁ
/
ꢁ
/
35 Hz, Hꢀ
/
Si); 31P{1H}-NMR:
2850 Hz, 3a), 15.23
2089 Hz, 1), 14.91 (s, with
2208 Hz, 2), ꢃ15.12 (s, de-coorrdi-
nated P, N ligand); 29Si-NMR: dꢁ
15.7 (SiꢀPt), 6.6
(MeSiEt3),0.05 (HSiEt3);
mator) was used for all measurements. The cell dimen-
sions were refined with all unique reflections. The data
collection (at 293 K) covered a hemisphere of the
reciprocal space, by a combination of three sets of
exposures. Each set had a different f angle for the
crystal, and each exposure took 20 s and covered 0.38 in
v. The crystal-to-detector distance was 4.40 cm. The
reflections were corrected for polarization and Lorentz
1
/
ꢁ
/
1
ꢁ
/
1
satellites, JPtP
ꢁ
/
/
/
/
4.3. Reaction of [(k2-P,N)ꢀ
(1) with HSiPh2Me
/
Me2N(CH2)3PPh2]PtMe2
effects, and an empirical absorption correction (SADABS
)
was employed.
The structure was solved by direct methods (SHELXS
1:1 molar ratio: 4 mg (0.002 mmol) of HSiPh2Me was
added at r.t. to 10 mg (0.002 mmol) of 1 in 0.5 ml of
C6D6. On addition of the silane no sign of reaction was
observed. However, with increasing reaction time at r.t.,
the sample turned yellow, then orange and after 24 h
-
97). The positions of the hydrogen atoms were calcu-
lated according to an idealized geometry. Refinement
was performed by the full-matrix least-squares method
based on F2
(SHELXL-97) with anisotropic thermal

S.M. Thompson et al. / Journal of Organometallic Chemistry 686 (2003) 183ꢀ
/191
191
Table 3
Crystallographic and structural parameters of 1 and 8
References
[1] Account on previous work: U. Schubert, J. Pfeiffer, F. Stohr, D.
¨
Sturmayr, S. Thompson, J. Organomet. Chem. 646 (2002) 53.
1×
/
C6H6
8
[2] F. Stohr, D. Sturmayr, G. Kickelbick, U. Schubert, Eur. J. Inorg.
¨
Chem. (2002) 2305.
Empirical formula
Formula weight
Space group
Unit cell dimensions
a (pm)
C25H22NPPt
562.5
P21/n
C22H34NPPt
538.6
P21/c
[3] J. Pfeiffer, G. Kickelbick, U. Schubert, Organometallics 19 (2000)
957.
[4] F. Stohr, D. Sturmayr, U. Schubert, J. Chem. Soc. Chem.
¨
1595.1(2)
938.29 (9)
1764.2 (2)
112.959(2)
1134.2(5)
845.4(3)
2293.6(9)
90.14(3)
Commun. (2002) 2222.
b (pm)
c (pm)
[5] J. Pfeiffer, U. Schubert, Organometallics 18 (1999) 3245.
[6] J. Pfeiffer, G. Kickelbick, U. Schubert, Organometallics 19 (2000)
62.
b (8)
V (pm3)
2431.2(4)ꢅ
/
106 2199(2)ꢅ106
/
[7] For example see: M.E. van der Boom, J. Ott, D. Milstein,
Organometallics 17 (1998) 4263.
Z
4
1.537
5.846
4
Dcalc (g cmꢃ3
)
1.627
6.458
[8] A word of caution should be given to the percentage values stated
here. The hetero-nuclear spectra cannot be accurately integrated
due to the method used, thus the values here are only quantitative
estimates to give a general trend.
m (mmꢃ1
)
Crystal size (mm3)
0.6ꢅ/0.4ꢅ/0.08 0.50ꢅ
/0.18ꢅ0.05
/
u Range (8)
2.20ꢀ
/
23.26
2.52ꢀ26.39
/
Reflections collected/unique
Data/parameters
GOF
10912/3499
3499/281
1.089
9877/3785
3785/228
1.084
[9] D. Sturmayr, U. Schubert, Monatsh. Chem. 134 (2003) 791.
[10] H. Gilges, U. Schubert, Organometallics 17 (1998) 4760.
[11] (a) For example: M.D. Curtis, P.S. Epstein, Adv. Organomet.
Chem. 19 (1981) 213;
R1 [I ꢀ
/
2s(I)]
0.058
0.150
0.044
0.113
wR2
(b) L.S. Chang, M.P. Johnson, M. Fink, Organometallics 10
(1991) 1219;
Largest difference peak and
ꢃ3
2.229/ꢃ/1.847
1.938/ꢃ1.517
/
˚
hole (e A
)
(c) H. Yamashita, M. Tanaka, M. Goto, Organometallics 11
(1992) 3227;
(d) D.C. Pestana, T.S. Koloski, D.H. Berry, Organometallics 13
(1994) 4173;
parameters for all non-hydrogen atoms. The parameters
of the hydrogen atoms were refined with a riding model.
(e) L.K. Figge, P.J. Carroll, D.H. Berry, Organometallics 15
(1996) 209;
(f) R.S. Simons, J.C. Gallucci, C.A. Tessier, W.J. Youngs, J.
Organomet. Chem. 654 (2002) 224.
[12] G.P. Mitchell, T.D. Tilley, Angew. Chem. Int. Ed. Engl. 39 (1998)
2524.
5. Supplementary material
[13] S.M. Thompson, U. Schubert, Inorg. Chim. Acta 350 (2003) 329.
[14] M. Aizenberg, D. Milstein, J. Am. Chem. Soc. 117 (1995) 6456.
[15] A.D. Sadow, T.D. Tilley, Angew. Chem. 115 (2003) 827; Angew.
Chem. Int. Ed. Engl. 42 (2003) 807.
Crystallographic data (excluding structure factors) for
the structural analysis have been deposited with the
Cambridge Crystallographic Data Centre, CCDC nos.
207246 and 207245 compounds 1 and 8, respectively.
Copies of this information may be obtained free of
charge from The Director, CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK (Fax: ꢂ44-1223-336033; e-
mail: deposit@ccdc.cam.ac.uk or www: http://
www.ccdc.cam.ac.uk).
[16] A.D. Becke, J. Chem. Phys. 18 (1993) 5648.
[17] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[18] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E.Scuseria, M.A.
Robb, J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery,
R.E. Stratmann, J.C. Burant, S. Dapprich, J.M. Millam, A.D.
Daniels, K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi, V.
Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C.
Adamo, S. Clifford, J. Ochterski, G.A. Petersson, P.Y. Ayala, Q.
Cui, K. Morokuma, D.K. Malick, A.D. Rabuck, K. Raghava-
chari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, B.B. Stefanov, G.
Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R.L.
Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A.
Nanayakkara, C. Gonzalez, M. Challacombe, P.M.W. Gill, B.
Johnson, W. Chen, M.W. Wong, J.L. Andres, C. Gonzalez, M.
Head-Gordon, E.S. Replogle, J.A. Pople Gaussian, Inc., Pitts-
burgh, PA, 1998.
/
Acknowledgements
This work was supported by the Fonds zur Forderung
¨
der wissenschaftlichen Forschung (FWF), Austria.