Organic-inorganic hybrid materials containing metal phosphine complexes: Evidence for preferential cis configuration of Pd(II) and Pt(II) complexes within the materials

Article abstract of DOI:10.1016/S0022-328X(01)01489-9

The hydrolysis-polycondensation of palladium (Pd) and platinum (Pt) complexes of phosphines [p-(OiPr)₃SiC₆ H₄]₂PC₆H₅ and [p-(OiPr)₃SiC₆H₄]₃P give

Full text of DOI:10.1016/S0022-328X(01)01489-9

Journal of Organometallic Chemistry 643–644 (2002) 453–460
www.elsevier.com/locate/jorganchem
Organic–inorganic hybrid materials containing metal
phosphine complexes
Evidence for preferential cis configuration of Pd(II) and
Pt(II) complexes within the materials
Jean-Philippe Bezombes, Claude Chuit, Robert R.J. Corriu *, Catherine Reye´
Laboratoire de Chimie Mole´culaire et Organisation du Solide, UMR 5637, Uni6ersite´ Montpellier II, Case 007,
F-34095 Montpellier Cedex 5, France
Received 26 August 2001; accepted 13 December 2001
Abstract
The hydrolysis–polycondensation of palladium (Pd) and platinum (Pt) complexes of phosphines [p-(OiPr)3SiC6H4]2PC6H5 and
[p-(OiPr)3SiC6H4]3P gives rise to xerogels in which the metal complexes are included within a silica matrix with retention of the
arrangement around the metal centre. However, the trans-Pd as well as the trans-Pt complexes incorporated within the xerogels
slowly isomerize into the more stable cis complexes. Direct reaction of (CH3CN)2PdCl2 with xerogels prepared from both
phosphines leads to the incorporation of the Pd salt, the corresponding complexes always adopting a cis arrangement of the
ligands. Treatment with nBu3P of all the materials containing metal centres leads to the complete removal of the metal except for
the materials obtained by hydrolysis of the metal–[p-(OiPr)3SiC6H4]3P complexes. © 2002 Published by Elsevier Science B.V.
Keywords: Hydrolysis; Hybrid material; Sol-gel; Transition metal
sible by bulky metallic reagents [8]. The metal can also
be introduced by co-hydrolysing a mixture of a metal
chelated trialkoxysilane of type (RO)3SiꢀXꢀD0ꢀM and
a tetraalkoxysilane [9,10]. Both methods allow the con-
trol of the quantity of the metal introduced into the
material. Moreover, the textural characteristics of the
materials can be adjusted through the choice of the
experimental conditions for the hydrolysis. The most
frequently used alkoxysilane derivatives are the phos-
phines Ph2P(CH2)nSi(OEt)3 [6,8,11,12] (n=2, 3). Cata-
lysts have also been obtained by hydrolysis–poly-
condensation of complexed phosphines bearing two
hydrolysable Si(OEt)3 groups [13].
Recently, we studied the hydrolysis–polycondensa-
tion (catalysed by 10% HCl) of phosphine 1 [with three
Si(OR)3 groups], phosphine 2 [with two Si(OR)3
groups], and metallic Pd (3), cis-Pt (4), and trans-Pt (5)
complexes giving rise to xerogels X1–X5 (Eq. (1)) [14].
We report in this paper the hydrolysis–polycondensa-
tion of the palladium (Pd) and platinum (Pt) complexes
of phosphine 2 (catalysed by HCl 10%) giving xerogels
1. Introduction
The economic importance of catalysts has led to
much research into catalysts supported on organic
polymers [1,2] or inorganic polymers such as silica. The
anchorage of catalysts on silica is achieved by reacting
the surface OH groups of the silica [3] with trialkoxysi-
lyl derivatives (RO)3SiꢀXꢀD0 where X is an organic
spacer and D0 a metal coordinating group [4]. The
metal can be introduced before or after the reaction of
the trialkoxysilane with the silica [3,4]. However, in
both cases, the metal loading remains rather low.
A more efficient method to anchor catalysts results
from the sol–gel process [5]. It consists in co-hy-
drolysing a mixture of a trialkoxysilane of type
(RO)3SiꢀXꢀD0 and a tetraalkoxysilane [6]. By this
method, the organic moiety is ‘diluted’ into the silica
matrix and the metal coordinating groups are located
outside the polysiloxane matrix [7] and are easily acces-
* Corresponding author.
0022-328X/02/$ - see front matter © 2002 Published by Elsevier Science B.V.
PII: S0022-328X(01)01489-9

454
J.-P. Bezombes et al. / Journal of Organometallic Chemistry 643–644 (2002) 453–460
X6 and X7 in order to compare the structures and
properties of these xerogels with those of X3–X5. We
have also studied the complexation of PdCl2 by the P
atoms incorporated within xerogels X1 and X2, in a
continuation of the study of the reactivity of these
centres [15]. The trans–cis isomerisation of the Pd and
Pt complexes included into these xerogels is shown. In
order to obtain information about the structures of the
xerogels resulting of the complexation of the phosphi-
nes before and after the hydrolysis, the decomplexation
of the phosphorus atoms was also investigated by react-
ing these xerogels with an excess of nBu3P.
coupling constant (3670 Hz) characteristic of a cis
geometry [17] as has 4 (1JPꢀPt=3614 Hz).
The hydrolysis–polycondensation of complexes 6
and 7 was performed under the same conditions
(30 °C, 0.5 M THF solution, stoichiometric amount of
water, 10% HCl as catalyst) as for complexes 3–5 [14].
Solid state NMR and textural data of the obtained
xerogels X6 and X7 as well as of xerogels X3, X4, and
X5, previously prepared [14], are indicated in Table 1.
The 31P CP MAS-NMR spectra of xerogels X6 and X7
display one signal, the chemical shift of which is very
(1)
close to that of phosphine complexes 6 and 7 as were
the chemical shifts of X3–X5 [14]. The stereochemistry
of the Pt complexes included into X7 was inferred from
the chemical shift of the 31P CP MAS-NMR signal
2. Result and discussion
1
2.1. Hydrolysis of the metallic complexes
(10.5 ppm). The JPꢀPt coupling constant could not be
determined due to the width of the signal. Thus, neither
decomplexation of the metal atom, nor isomerisation of
the metallic complexes occurred during the sol–gel
process. The 29Si CP MAS-NMR spectra of all the
xerogels X3–X7 (Table 1) display signals corresponding
to T1, [CꢀSi(OR)2OSi] and T2, [CꢀSi(OR)(OSi)2] sub-
structures (a weak T3 substructure, [CꢀSi(OSi)3], was
observed only for X7). The ratio of the different T
groups was estimated by the method of deconvoluted T
peak [18] which gives a degree of condensation for all
The Pd and Pt complexes 6 and 7 were prepared as
indicated in Scheme 1, but it was not possible to
prepare the trans-Pt complex of phosphine 2. The
31P-NMR spectra of Pd complex 6 displays a single
resonance at 23.6 ppm close to that of 3 (23.7 ppm) [14]
and close to that of trans-(Ph3P)2PdCl2 (23.1 ppm) [16]
allowing attribution of a trans geometry to complexes 3
and 6. The 31P-NMR spectrum of the Pt complex 7
1
displays a resonance at 14.2 ppm and a large JPꢀPt
Scheme 1.

J.-P. Bezombes et al. / Journal of Organometallic Chemistry 643–644 (2002) 453–460
455
Table 1
1
Solid state 31P-NMR (l, ppm, JPꢀPt, Hz in bracket), solid state 29Si-NMR (l, ppm), degree of condensation (~), and textural data for xerogels
X3–X7
Xerogel
31P-NMR
29Si-NMR
~
SBET
Micropore volume
Microporous surface
(%)
(m2 g−1
)
(ml g−1
)
(m2 g−1
)
X3
X4
X5
X6
X7
27.3
−62.8 (T1), −71.7 (T2)
−63.9 (T1), −71.3 (T2)
58
61
510
460
520
100
410
0.225
0.201
0.230
490
437
497
11.3 (3660)
20.8 (2590)
24.9
−61.5 (T1), −68.9 (T2)
56
58
10.5
−62.1 (T1), −70.0 (T2), −78.9 (T3)
0.153
349
Fig. 1. Solid state 31P-NMR spectrum of Pt xerogel X5 as a function of time: (a) after reaction; (b) after one year.
the xerogels of about 60% (Table 1). Some OH and iPr
groups remain into the xerogels, the iPr groups being
detected by 13C CP MAS-NMR spectroscopy and the
OH groups by DRIFTS analysis. The texture of the
xerogels was investigated using nitrogen adsorption–
desorption measurements (Table 1). The absorption–
desorption isotherms of all samples were of type I
(characteristic of microporous material according to the
BDDT) [19]. The N2 Brunauer–Emmett–Teller (BET)
surface areas are rather large (400–500 m2 g−1) except
for X6 (100 m2 g−1).
No change was apparent after a further year. A similar
evolution of the 31P CP MAS-NMR spectra of the
trans-Pd complexes was observed for xerogels X3 and
X6. The original signal of X3 at 27.3 ppm (correspond-
ing to the trans complex) disappears after about one
year to give another signal at 33.8 ppm. An analogous
change of the trans signal was observed for X6 but at a
slower rate as shown in Fig. 2. The initial signal at 24.9
ppm (of the trans complexes) was transformed into two
signals of the same intensity at 25.9 and 31.2 ppm after
8 months, and a major signal at 32 ppm and a weak
signal at 26 ppm was formed after a further year. Thus
the trans-Pd complexes incorporated within X3 and X6
are slowly transformed to the cis complexes with a
chemical shift of about 30 ppm. The attribution for the
chemical shifts for the trans- and cis-Pd complexes are
in agreement with the values found for the trans- and
cis-[(EtO)3Si(CH2)2PPh2]PdCl2 complexes anchored to
silica (respectively 21.3 and 30.0 ppm) [16]. In contrast,
the 31P CP MAS-NMR spectra of xerogels X4 and X7
2.2. Trans–cis isomerisation of the complexes included
into the xerogels
While the 31P CP MAS-NMR spectrum of X5 with a
trans-Pt complexes at first exhibited one resonance at
20.8 ppm (Table 1), after one year two signals were
observed at 20.6 and 13.5 ppm, signals which were
tentatively attributed to the trans (JPꢀPt=2600 Hz) and
cis (JPꢀPt=3900 Hz) complexes as indicated in Fig. 1.

456
J.-P. Bezombes et al. / Journal of Organometallic Chemistry 643–644 (2002) 453–460
Fig. 2. Solid state 31P-NMR spectrum of Pd xerogel X5 as a function of time: (a) after reaction; (b) after 8 months; (c) after 20 months.
incorporating cis-Pt complexes are all the same after
more than one year. Thus, in xerogels X3–X7 the more
stable geometry around the Pd or Pt atoms is the cis
geometry. The trans complexes isomerise slowly into
the cis complexes, the isomerisation of the Pd com-
plexes being easier than that of the Pt complexes. The
same difference is observed in solution, the Pt phos-
phine complexes being isomerised only in presence of
an external nucleophile [20] which is not the case for
the Pd complexes [21]. Thus, in these materials, the
phosphorus atoms are sufficiently mobile to break a
phosphorus metal bond and to build a new phosphorus
metal bond with another phosphorus atom to give a
more stable complex. Isomerisation by complexation–
decomplexation was observed in the ruthenium com-
plexes included into the xerogel resulting from
hydrolysis of the Ru complex [22] obtained from
PhP[(CH2)4Si(OMe)3](CH2OCH3).
reflux for 48 h. In all cases, the yield of incorporation
of PdCl2 determined by solid state 31P-NMR spec-
troscopy is quantitative. The xerogels obtained,
X1%[Pd], X1[Pd], and X2[Pd] were washed with acetoni-
trile, ethanol, acetone, and diethyl ether and dried
under vacuum. The BET surface area of the new xe-
rogels were not different from those of the starting
xerogels except for X1%, the SBET of which increases to
80 m2 g−1. The 31P CP MAS-NMR spectra of the new
xerogels show the disappearance of the signal attributed
to the phosphines moieties (about −5 ppm) and the
appearance of new signals at 31.8 ppm for X1%[Pd] and
X2[Pd] and 29.9 ppm for X1[Pd], corresponding to the
cis complexes. Thus, all the phosphorus atoms have
complexed the PdCl2 salts. However, the microanalysis
data indicate an excess of PdCl2 into the xerogels
X1%[Pd], X1[Pd], and X2[Pd]. The fact that no CꢁN
stretching band was observed by DRIFTS analysis and
that only a very low percentage of nitrogen was found
by elemental analysis, indicate that all the CH3CN
ligands were eliminated. As the excess of PdCl2 salts
was not eliminated by washing, it is likely that the
PdCl2 salts are retained into the xerogels by coordina-
tion. One explanation is part of the PdCl2 salts was
complexed by two phosphorus atoms in cis position
while the rest is complexed by one phosphorus atom
and probably by another nucleophile present in the
xerogel (OH or OR groups), the ratio of both kinds of
complexes depending on the nature of the xerogels (X1,
X1% or X2). Thus, whatever the texture of the xerogels,
all the phosphorus atoms of the xerogels X1%, X1, and
X2 are accessible by (CH3CN)2PdCl2, as they were by
H2O2, S8, and CH3I [15]. We can deduce that the first
step is the coordination of the Pd atom by one phos-
phorus atom, a second nucleophile (P, OH, OR) giving
the cis-Pd complex. The fact that the BET surfaces of
2.3. Study of the accessibility of the phosphorus atoms
within the materials
We have previously shown that the phosphorus
atoms included into xerogels X1 and X2 can react with
different reagents and that the accessibility of the phos-
phorus atoms is easier from the xerogel X2 than from
the xerogel X1 [15]. To complete this work, we studied
the complexing ability of the phosphorus atoms in-
cluded into the two xerogels X1 and X2 as well as into
xerogel X1% (obtained by hydrolysis of 1 catalysed by
1% of p-toluenesulfonic acid). X1 has a rather large
BET surface area (450 m2 g−1) while X1% and X2 have
low BET surface area (B10 m2 g−1).
We found that the best conditions to incorporate
PdCl2 into the solids were the addition of a twofold
excess of (CH3CN)2PdCl2 in acetonitrile heated under

J.-P. Bezombes et al. / Journal of Organometallic Chemistry 643–644 (2002) 453–460
457
Table 2
of the phosphorus atoms during the treatment of the
xerogel with nBu3P. Indeed, the phosphine oxide incor-
porated into X1 has a chemical shift of 30.7 ppm [14].
So, in xerogels X1[Pd] and X2[Pd] the Pd atoms are
easily accessible by nBu3P. The fact that X1[Pd] is
easily demetallated may result from a possible conser-
vation of the texture of the xerogel X1 after reaction
with (CH3CN)2PdCl2 allowing the accessibility of the
Pd atoms by nBu3P.
Thus, the xerogels X3 and X1[Pd], obtained from the
same phosphine 1 by two different routes, have differ-
ent structures which render the accessibility of the Pd
atom easy in X1[Pd], but very difficult in X3. Moreover,
we observed the same difference of reactivity between
the xerogels X6 and X3 as previously observed between
the xerogels X2 and X1 [15].
Solid state 31P-NMR chemical shifts (l, ppm), 31P-NMR yields (in
brackets), and SBET (m2
xerogels before and after treatment with nBu3P
g
−1, in square brackets) of metallated
Xerogel
Starting xerogel Xerogel after treatment with nBu3P
X3
X6
X4
23.7 [510]
24.9 [100]
11.4 [460]
25.3 a (ca. 100%) [B10]
−5.0 (ca. 100%) [30]
4.1 (ca. 60%) and 13.1 b (ca. 40%)
[14]
X5
20.8 [520]
4.2 (ca. 20%) and 21.6 (ca. 80%)
[B10]
X7
X1[Pd]
10.5 [410]
29.9 [380]
−5.5 c (ca. 100%) [20]
−5.6 (ca. 70%) and 29.9 (ca. 30%)
[350]
X2[Pd]
31.2 [B10]
−5.4 (ca. 100%) [B10]
a Two signals at 25.6 and 31.9 are obtained after 9 months.
b A weak unidentified signal was observed at 55 ppm.
c A weak signal (ca. 5%) at 31.3 ppm corresponding to oxidation of
the phosphorus atoms is observed.
3. Conclusion
In conclusion, metallic complexes have been intro-
duced into a silica matrice by two ways:
the metallated xerogels are not very different than those
of the starting xerogels seems to indicate that the
reaction with (CH3CN)2PdCl2 does not greatly alter the
texture of the xerogels.
– Hydrolysis–polycondensation of a Pd (or Pt) phos-
phine complex, the phosphine bearing two or three
hydrolysable (SiOR)3 groups. In these cases, the
geometry of the complex was preserved during the
sol–gel process, but the trans complexes are slowly
isomerised into the cis complexes.
2.4. Study of the demetallation of the complexes
included into the xerogels
– Reaction of (CH3CN)2PdCl2 with the phosphorus
atoms included into the xerogels X1 [obtained from
[p-(OiPr)3SiC6H4]3P] and X2 [obtained from [p-
(OiPr)3SiC6H4]2PPh]. In these cases, the yields of
complexation are quantitative and the geometry of
the complexes is cis. However, the Pd salt is re-
tained not only by phosphines by also by other
nucleophiles such as OH or OR groups.
– The metallic salts can be removed from the xerogels
by treatment with nBu3P. We found the same dif-
ference of reactivity between the xerogels X3 and
X5 [obtained from [p-(OiPr)3SiC6H4]3P] and X6 and
X7 [obtained from [p-(OiPr)3SiC6H4]2PPh] as that
previously found for the reactivity of the phospho-
rus atoms included into xerogels X1 and X2. The
demetallation is very easy from xerogels X6 and X7
but difficult from xerogels X3 and X5. On the con-
trary, the demetallation is very easy from the xe-
rogels obtained by the second route, whatever the
xerogel. Thus, the structure and the texture of the
xerogels prepared by the two ways are different.
The accessibility of the reaction centres is easier
from X1[Pd] than from X3. That means that the
reaction of (CH3CN)2PdCl2 with the P atoms in-
cluded into X1 do not modify appreciably the struc-
ture of the xerogel and the accessibility of the
reactive centres.
To study the accessibility of the metallic atoms
within the xerogels X3–X7 and within the xerogels
X1[Pd] and X2[Pd], we treated these xerogels with an
excess (10 equivalents) of nBu3P in acetonitrile heated
under reflux for 120 h. The results obtained are indi-
cated in Table 2. As shown in this table, no decomplex-
ation occurred for the Pd complexes of X3 and only
partial decomplexation of the Pt of X4 and X5 was
observed. On the contrary, total decomplexation of the
Pd and Pt complexes of xerogels X6 and X7 occurred.
Thus, the metallic atoms included into the xerogels X6
and X7 obtained from phosphine 2 with two hy-
drolysable Si(OR)3 groups are much more easily acces-
sible than the metallic salts included into xerogels
X3–X5 obtained from phosphine 1 with three hy-
drolysable Si(OR)3 groups. The greater number of
SiꢀOꢀSi attached to the organic units, the less easy is
the diffusion into the xerogel, as it was previously
observed with xerogels X1 and X2 [15,23]. For all these
xerogels (X3–X7), there is a dramatic decrease of the
BET specific surfaces after treatment with nBu3P.
In contrast to these results, the decomplexation of
the phosphorus atoms included into xereogels X1[Pd]
and X2[Pd] is easy and quantitative (Table 2). After
treatment with nBu3P an additional signal, at 29.9 ppm
was observed in the 31P CP MAS-NMR spectra of
xerogels X1[Pd]. This signal was attributed to oxidation

458
J.-P. Bezombes et al. / Journal of Organometallic Chemistry 643–644 (2002) 453–460
4. Experimental
(l, 50 MHz, CDCl3): 25.9 (Me), 66.0 (OCH), 128.1–
136.9 (m, aromatic); 31P-NMR (l, 100 MHz, CDCl3):
14.2 [s and d (satellite 195Pt, d, 1JPꢀPt=3670 Hz)];
29Si-NMR (l, 40 MHz, CDCl3): −63.3; MS (FAB+,
NBA): 1571 [(M+Cl)+, 65%], 1535 [(M−2Cl+H)+,
59%]. Anal. Calc. for C72H110Cl2O12P2PtSi4: C, 53.80;
H, 6.85. Found: C, 53.92; H, 6.95%.
All reactions were carried out under Ar by using a
vacuum line. Solvents were dried and distilled just
before use. M.p. were determined with a Gallenkamp
apparatus and are uncorrected. The solution NMR
spectra were recorded on a Bruker AC-200 (29Si),
Bruker DPX-200 (1H and 13C) and Bruker WP-250 SY
(31P). Chemical shifts (l in ppm) were referenced to
Me4Si (1H, 13C, 29Si) or H3PO4 (31P). The CP MAS 29Si
solid state-NMR spectra were recorded on a Bruker
FTAM-300 as well as CP MAS 13C solid state-NMR
spectra in that case by using the TOSS technique. In
both cases the repetition time was 5 and 10 s with
contact times of 5 and 2 ms. The HPDEC MAS 31P
solid state-NMR spectra were recorded on a Bruker
ASX-200 or a Bruker ASX-400 with repetition time of
5 s. FAB mass spectra [matrix, m-nitrobenzyl alcohol
(NBA)] were registered on JEOL JMS-D3000 spec-
trometer. Specific surface areas were determined by the
BET method on Micromeritics Gemini III 2375 analy-
ser. Elemental analysis were carried out by the Service
Central de Micro-Analyse du CNRS.
4.2.1. Xerogel X6
To a solution of 2.00 g (1.32 mmol) of 6 in THF (2.6
ml) placed in a 20 ml flask were added dropwise at
room temperature (r.t.) 2.6 ml (7.8 mmol of H2O) of a
3.0 M aq. 0.1 M HCl solution in THF. After stirring
for 5 min at r.t., the reaction mixture was heated at
30 °C without stirring. An orange gel was formed after
30 min. The wet gel was allowed to age for 5 days at
30 °C after which it was powdered and washed with
EtOH, C3H6O and Et2O. The powdering and washing
were repeated once and the gel was powdered again and
dried under vacuum for 2 h at 120 °C. Orange powder
(1.25 g) was obtained. 13C-NMR (l, 75 MHz, CP
MAS): 24.7 (Me), 67.0 (OCH), 125.1–133.8 (aromatic);
29Si-NMR (l, 60 MHz, CP MAS): −61.5 (T1), −68.9
(T2), −80 (T3, shoulder); 31P-NMR (l, 81 MHz,
HPDEC MAS): 24.9. SBET: 100 m2 g−1. Anal. Calc. for
C36H26Cl2O6P2PdSi4: C, 47.73; H, 2.87; Cl, 7.84; P,
6.85; Pd, 11.71; Si, 12.38. Found: C, 46.22; H, 3.59; Cl,
4.1. trans-Bis{phenyl[bis(p-triisopropyloxysilylphenyl)]-
phosphine}dichloropalladium (6)
A mixture of phosphine 2 (3.74 g, 5.58 mmol) and of
PdCl2 (495 mg, 2.79 mmol) was heated under reflux in
THF (30 ml) for 3 h. After elimination of the solvent
and crystallisation from propan-2-ol, 3.45 g (2.27
mmol, 82%) of 6 were obtained as yellow needles. M.p.
6.83; P, 6.25; Pd, 9.35; Si, 12.50%, i.e. C38.21H35.61-
Cl1.91O9.46P2.00Pd0.87Si4.42
.
4.2.2. Xerogel 7
The same procedure starting from 2.00 g (1.18 mmol)
of 7 led to 1.18 g of a yellow powder. 29Si-NMR (l, 60
MHz, CP MAS): −62.1 (T1), −70.0 (T2), −78.9 (T3,
shoulder); 31P-NMR (l, 162 MHz, HPDEC MAS):
1
(propan-2-ol): 185.5–187 °C; H-NMR (l, 200 MHz,
3
CDCl3): 1.24 (d, JHꢀH=6.1 Hz, 72 H, Me), 4.30 (sept,
3JHꢀH=6.1 Hz, 12 H, OCH), 7.39–7.73 (m, 26 H,
aromatic); 13C-NMR (l, 50 MHz, CDCl3): 25.9 (Me),
66.0 (OCH), 128.3–136.2 (m, aromatic); 31P-NMR (l,
100 MHz, CDCl3): 23.6; 29Si-NMR (l, 40 MHz,
CDCl3): −62.6. Anal. Calc. for C72H110Cl2O12P2PdSi4:
C, 56.95; H, 7.25. Found: C, 56.83; H, 7.35%.
10.5. SBET: 410 m2
g
−1. Anal. Calc. for C36H26Cl2-
O6P2PtSi4: C, 43.46; H, 2.62; Cl, 7.14; P, 6.24; Pt, 19.63;
Si, 11.27. Found: C, 41.67; H, 4.02; Cl, 6.10; P, 5.80; Pt,
16.60;
Si,
10.50%,
i.e.
C37.12H42.97Cl1.84O10.23-
P2.00Pt0.91Si4.01
.
4.2. cis-Bis{phenyl[bis(p-triisopropyloxysilylphenyl)]-
phosphine}dichloroplatinum (7)
4.2.3. Reacti6ity of xerogels X1%, X1, and X2
Before reaction, all the xerogels were systematically
dried again under vacuum for 2 h at 120 °C. The
reagent (CH3CN)2PdCl2 was used in large excess. The
molar mass of the xerogels has been calculated from an
ideal formula (complete condensation). The following
procedure is given as example.
To a warm solution of phosphine 2 (3.00 g, 4.47
mmol) in propan-2-ol (35 ml) was added a solution of
930 mg (2.24 mmol) of K2PtCl4 in water (15 ml). The
pink suspension was heated at 40 °C for 1 h. After
cooling, the product was extracted with Et2O. The
organic layers were dried over MgSO4 and the solvent
was removed under vacuum. After crystallisation from
propan-2-ol, 2.37 g (1.48 mmol, 66%) of 7 were ob-
tained as a white powder. M.p. (propan-2-ol): 108 °C
(dec.); 1H-NMR (l, 200 MHz, CDCl3): 1.22 (d,
4.2.3.1. Xerogel X1%[Pd]. Xerogel X1% (320 mg, 0.77
mmol) and (CH3CN)2PdCl2 (200 mg, 0.77 mmol) were
heated under reflux in MeCN (30 ml) for 24 h. The
orange suspension was then filtered and the precipitate
was washed with MeCN, EtOH, C3H6O and Et2O.
After drying, 368 mg of an orange powder was ob-
tained. 31P-NMR (l, 81 MHz, HPDEC MAS): 31.2.
3
3JHꢀH=6.1 Hz, 72 H, Me), 4.28 (spt, JHꢀH=6.1 Hz,
12 H, OCH), 7.1–7.7 (m, 26 H, aromatic); 13C-NMR

J.-P. Bezombes et al. / Journal of Organometallic Chemistry 643–644 (2002) 453–460
459
SBET
:
80 m2 g−1
.
Elemental Anal. Calc. for
4.3.5. Xerogel X7
C36H24Cl2O9P2PdSi6: C, 42.88; H, 2.38; Cl, 7.05; P,
6.15; Pd, 10.56; Si, 16.67. Found: C, 38.26; H, 4.29; Cl,
7.58; P, 3.92; Pd, 10.56; Si, 12.04%, i.e.
31P-NMR (l, 162 MHz, HPDEC MAS): −5.5. SBET
:
20 m2 g−1. Elemental Anal. Calc. for C18H13O3PSi2: C,
59.34; H, 3.57; P, 8.51; Si, 15.38. Found: C, 57.06; H,
4.82; Cl, 0.83; P, 8.00; Pt, 1.30; Si, 12.80%.
C54.44H67.87Cl3.38O20.58P2.00Pd1.93Si6.80
.
4.3.6. Xerogel X1[Pd]
4.2.3.2. Xerogel X1[Pd]. 31P-NMR (l, 81 MHz,
31P-NMR (l, 81 MHz, HPDEC MAS): −5.6, 29.9.
HPDEC MAS): 29.9. SBET: 380 m2 −1. Elemental
g
SBET
: . Elemental Anal. Calc. for
B10 m2 g−1
Anal. Calc. for C36H24Cl2O9P2PdSi6: C, 42.88; H, 2.38;
Cl, 7.05; P, 6.15; Pd, 10.56; Si, 16.67. Found: C, 36.70;
H, 3.84; Cl, 9.21; P, 4.70; Pd, 13.95; Si, 13.15%, i.e.
C18H12O4.5PSi3: C, 52.05; H, 2.89; P, 7.47; Si, 20.24.
Found: C, 51.55; H, 5.60; Cl, 0.48; P, 4.80; Pd, 0.90; Si,
9.90%.
C40.34H50.65Cl3.42O15.09P2.00Pd1.31Si6.19
.
4.3.7. Xerogel X2[Pd]
4.2.3.3. Xerogel X2[Pd]. 31P-NMR (l, 81 MHz,
HPDEC MAS): 31.2. SBET: 10 m2 g−1. Elemental Anal.
Calc. for C36H26Cl2O6P2PdSi4: C, 47.73; H, 2.87; Cl,
7.84; P, 6.85; Pd, 11.71; Si, 12.38. Found: C, 40.52; H,
3.78; Cl, 9.25; P, 5.70; Pd, 13.70; Si, 10.70%, i.e.
31P-NMR (l, 81 MHz, HPDEC MAS): −5.4. SBET
:
B10 m2 g−1 Elemental Anal. Calc. for C18H13O3PSi2:
C, 59.34; H, 3.57; P, 8.51; Si, 15.20. Found: C, 57.18;
H, 5.41; Cl, 820 ppm, P, 7.10; Pd, 0.10; Si, 12.40%.
C
36.73H41.12Cl2.83O10.98P2Pd1.40Si4.16
.
References
4.3. Xerogels demetallation
[1] F.R. Hartley, P.N. Vezey, Adv. Inorganom. Chem. 15 (1977)
189.
[2] E. Lindner, T. Schneller, F. Auer, H.A. Mayer, Angew. Chem.
Int. Ed. 38 (1999) 2154.
[3] K.G. Allum, R.D. Hancock, I.V. Howell, S. McKenzie, R.G.
Pitkethly, P.J. Robinson, J. Organomet. Chem. 87 (1975) 203.
[4] U. Deschler, P. Kleinsch, P. Panster, Angew. Chem. Int. Ed. 25
(1986) 236.
[5] (a) U. Schubert, N. Hu¨sing, A. Lorenz, Chem. Mater. 7 (1995)
2010;
All the demetallation reactions have been carried out
in the same way, except that the reaction time was of 5
days for X3–X7 and of 2 days for X1[Pd] and
X2[Pd].The following procedure is given as an example.
4.3.1. Xerogel X3
Xerogel X3 (937 mg, 0.93 mmol) and nBu3P (4.20 g,
20.7 mmol) were refluxed in C6H5CH3 (20 ml) for 5
days. The suspension was filtered and the precipitate
washed with C6H5CH3, EtOH, C3H6O and Et2O. After
drying, 987 mg of an orange powder was obtained.
(b) D.A. Loy, K.J. Shea, Chem. Rev. 95 (1995) 1431;
(c) R.J.P. Corriu, D. Leclercq, Angew. Chem. Int. Ed. 35 (1966)
1421;
(d) R. Corriu, Polyhedron 17 (1998) 925;
(e) R. Corriu, C. R. Acad. Sci. Ser. IIc (1998) 83.
[6] U. Schubert, K. Rose, H. Schmidt, J. Non-Cryst. Solids 105
(1988) 165.
31P-NMR (l, 81 MHz, HPDEC MAS): 25.3. SBET
:
B10 m2 g−1
.
Elemental Anal. Calc. for
C18H12O4.5PdSi3: C, 52.05; H, 2.89; P, 7.47; Si, 20.24%.
[7] E. Lindner, A. Bader, H.A. Mayer, Inorg. Chem. 30 (1991) 3783.
[8] U. Schubert, New J. Chem. 18 (1994) 1049.
[9] U. Schubert, C. Egger, K. Rose, C. Alt, J. Mol. Catal. 55 (1989)
330.
[10] (a) E. Lindner, M. Kemmler, H.A. Mayer, Chem. Ber. 125
(1992) 2385;
4.3.2. Xerogel X4
31P-NMR (l, 162 MHz, HPDEC MAS): −4.1, 13.1,
55.0. SBET: 14 m2
g
−1. Elemental Anal. Calc. for
C18H12O4.5PSi3: C, 52.05; H, 2.89; P, 7.47; Si, 20.24.
Found: C, 46.17; H, 5.35; Cl, 3.36; P, 7.55; Pt, 10.50; Si,
11.15%.
(b) E. Lindner, M. Kemmler, H.A. Mayer, Chem. Ber. 620
(1994) 1142;
(c) E. Lindner, M. Kemmler, H.A. Mayer, P. Wegner, J. Am.
Chem. Soc. 116 (1994) 348;
(d) E. Lindner, R. Schreiber, T. Schneller, P. Wegner, H.A.
Mayer, Inorg. Chem. 35 (1996) 514.
[11] O. Kro¨cher, R.A. Ko¨ppel, A. Baiker, J. Chem. Soc. Chem.
Commun. (1996) 1497.
[12] (a) F.G. Young, Ger. Offen. (1974) 2 330 308;
(b) P. Panster, P. Kleinschmit, Ger. Offen (1980) 3 029 599;
(c) R.V. Parish, D. Habibi, V. Mohammadi, J. Organomet.
Chem. 369 (1989) 17.
4.3.3. Xerogel X5
31P-NMR (l, 162 MHz, HPDEC MAS): −4.2, 21.6.
SBET
:
10 m2 g−1
.
Elemental Anal. Calc. for
C18H12O4.5PSi3: C, 52.05; H, 2.89; P, 7.47; Si, 20.24.
Found: C, 43.96; H, 4.81; Cl, 4.77; P, 7.00; Pt, 9.60; Si,
12.00%.
[13] S. Wieland, P. Panster, in: Scaros, M.G., Prunier (Eds.), Cataly-
sis of Organic Reactions, Marcel Dekker, New York, 1995, p.
383.
[14] J.P. Bezombes, C. Chuit, R.J.P. Corriu, C. Reye´, J. Mater.
Chem. 8 (1998) 1749.
4.3.4. Xerogel X6
31P-NMR (l, 81 MHz, HPDEC MAS): −5.0. SBET
:
30 m2 g−1. Elemental Anal. Calc. for C18H13OP2PdSi6:
C, 59.34; H, 3.57; P, 8.51; Si, 15.38. Found: C, 55.82;
H, 4.58; Cl, 0.47; P, 7.90; Pd, 0.80; Si, 12.80%.
[15] J.P. Bezombes, C. Chuit, R.J.P. Corriu, C. Reye´, J. Mater.
Chem. 9 (1999) 1727.

460
J.-P. Bezombes et al. / Journal of Organometallic Chemistry 643–644 (2002) 453–460
[16] R.A. Komoroski, A.J. Magistro, P.P. Nicholas, Inorg. Chem. 25
(1986) 3917.
[17] (a) S.O. Grim, R.L. Keiter, W. McFarlane, Inorg. Chem. 6
(1967) 1133;
(b) W.P. Power, R.E. Wasylishen, Inorg. Chem. 31 (1992) 2176.
[18] H.W. Oviatt, K.J. Shea, J.H. Small, Chem. Mater. 5 (1993)
943.
[20] (a) P. Haake, R.M. Pfeiffer, J. Am. Chem. Soc. 92 (1970) 4996;
(b) D.G. Cooper, J. Powell, J. Am. Chem. Soc. 95 (1973) 1102.
[21] (a) D.A. Redfield, J.H. Nelson, Inorg. Chem. 12 (1973) 15;
(b) D.A. Redfield, J.H. Nelson, R.A. Henry, D.W. Moore, H.B.
Jonassen, J. Am. Chem. Soc. 96 (1974) 6298.
[22] E. Lindner, M. Kemmler, T. Schneller, H.A. Mayer, Inorg.
Chem. 34 (1995) 5489.
[19] S. Brunauer, L.S. Deming, W.E. Deming, E. Teller, J. Am.
Chem. Soc. 62 (1940) 1723.
[23] J.P. Bezombes, C. Chuit, R.J.P. Corriu, C. Reye´, Can. J. Chem.
78 (2000) 1519.