Preparation and characterization of organic-inorganic hybrid materials incorporating diphosphino moieties. Study of the accessibility of the phosphorus atoms included into the material

Article abstract of DOI:10.1139/v00-091

The hydrolysis-polycondensation by the sol-gel process of the aromatic diphosphines (X₃SiC₆H₄)₂PC₆H ₄P(C₆H₄SiX₃)₂, which are rigid molecules bear

Full text of DOI:10.1139/v00-091

1519
Preparation and characterization of organic-
inorganic hybrid materials incorporating
diphosphino moieties. Study of the accessibility of
the phosphorus atoms included into the material
Jean-Philippe Bezombes, Claude Chuit, Robert J.P. Corriu, and Catherine Reyé
Abstract: The hydrolysis-polycondensation by the sol-gel process of the aromatic diphosphines
(X3SiC6H4)2PC6H4P(C6H4SiX3)2, which are rigid molecules bearing four hydrolysable SiX3 groups (X = OiPr, H), leads
to new organic-inorganic hybrid materials, characterized by solid state 13C, 29Si, and 31P NMR spectroscopies. The
accessibility of the phosphorus centres incorporated into the xerogel obtained from the diphosphine with X = OiPr has
been studied. All the phosphorus atoms reacted quantitatively with H2O2, S8, and CH3I but only 20% with the more
bulky reagent W(CO)5·THF. This result is explained by the rigidity of the inorganic network resulting from the high
number of hydrolysable Si-OiPr groups in the precursor.
Key words : Diphosphines, sol-gel process, xerogels, solid 31P NMR.
Résumé : L’hydrolyse-polycondensation via le procédé sol-gel des diphosphines aromatiques
(X3SiC6H4)2PC6H4P(C6H4SiX3)2, molécules rigides portant quatre groupements SiX3 hydrolysables (X = OiPr, H),
conduit à de nouveaux matériaux hybrides organique-inorganiques qui ont été caractérisés par la RMN à l’état solide
du 13C, du 29Si et du 31P. L’étude de l’ accessibilité des centres phosphorés incorporés dans le xérogel obtenu à partir
de la diphosphine avec X = OiPr est décrite. Tous les atomes de phosphore réagissent quantitativement avec H2O2, S8,
et CH3I, mais seulement 20% réagissent avec W(CO)5·THF réactif plus volumineux que les précédents. Ce résultat est
expliqué par la rigidité du réseau inorganique qui est la conséquence du nombre élevé de groupements hydrolysables
Si-OiPr porté par le précurseur.
Mots clés : diphosphines, procédé sol-gel, xérogels, RMN solide 31P. Bezombes et al. 1525
Introduction
As a continuation of this study, we prepared the aromatic
diphosphines 4 and 5 (with four SiX3 groups) and studied
their hydrolysis and polycondensation as well as the accessi-
bility of the phosphorus centres included into the xerogel X4
obtained from the diphosphine 4. This diphosphine was cho-
sen with rigid substituents around the phosphorus centres as
in the case for the phosphines 1 and 2. This study was un-
dertaken to verify if the high number of Si(OiPr)3 groups in
the precursor 4 gives rise to a low accessibility of the phos-
phorus centres included into the corresponding xerogel as
predicted by our previous study (6).
The sol-gel process allows the preparation of organic–in-
organic hybrid materials under smooth conditions (1–4). In a
recent publication (5) we described the preparation of
phosphines 1, 2, and 3 as well as their hydrolysis and
polycondensation via the sol-gel process, giving rise to the
corresponding xerogels. The accessibility of the phosphorus
centres included in these xerogels was investigated by using
some classical reactions at phosphorus. We found that the
phosphorus centres react quantitatively with H2O2, S8, and
MeI, whilst their reactivity towards W(CO)5·THF (6) varied
depending on the xerogel studied. We noticed that the phos-
phorus atoms became more readily accessible when the
number of Si(OR)3 groups in the precursor is low. Thus the
phosphorus atoms in the xerogel derived from precursor 1
(with two Si(OiPr)3 groups) reacts more readily with
W(CO)5·THF than when included into the xerogels derived
from precursors 2 and 3 (with three SiX3 groups).
Results and discussion
1) Preparation of diphosphines 4 and 5 and of
phosphine derivatives 7–9.
The diphosphine 4 was prepared in 53% yield by reaction
of the previously described functional Grignard reagent 6 (5)
with P,P,P ′,P ′-tetrachloro-p-phenylenediphosphine (7) in
Received July 5, 1999. Published on the NRC Research Press website on November 2, 2000.
This paper is dedicated to Professor Adrian Brook in recognition to his very important contribution in silicon chemistry.
J.-P. Bezombes, C. Chuit, R.J.P. Corriu,1 and C. Reyé. Laboratoire de Chimie Moleculaire et Organisation du Solide. UMR
5637, Université Montpellier II, Case 007, F-34095 Montpellier CEDEX 5, France.
1Author to whom correspondence may be addressed. Telephone : 0467143801. Fax : 0467143852.
e-mail: corriu@crit.univ.montp2.fr
Can. J. Chem. 78: 1519–1525 (2000)
© 2000 NRC Canada

1520
Can. J. Chem. Vol. 78, 2000
Scheme 1.
Scheme 2.
THF (Scheme 1). LiAlH4 reduction of
4
gave the
only one signal at –5.0 ppm which indicates that no
oxidation of the phosphorus atom occurred during the sol-
gel process. The solid state 13C NMR spectrum of X4 dis-
plays one signal at 134.4 ppm attributed to the aromatic car-
bons and two further signals at 24.8 and 66.7 ppm attributed
to remaining OiPr groups. Thus the hydrolysis is incomplete;
however the elemental analysis indicates that only about
10% of the Si(OiPr)3 groups were not hydrolyzed. The solid
state 29Si NMR spectrum of X4 displays two main reso-
nances at –60.5 ppm [substructure T1, C-Si(OR)2OSi] and –
69.1 ppm [substructure T2, C-Si(OR)(OSi)2]. A minor reso-
nance assigned to the substructure T3 [C-Si(OSi)3] was ob-
served while substructure T° was absent. The ratio of the T
groups was estimated from deconvoluted T peaks (8, 9) at
42% for T1, 52% for T2, and 6% for T3 corresponding to a
degree of condensation τ of 55%.2 Thus while the hydrolysis
is important, the polycondensation is rather weak. The ab-
sence of any Q signal corresponding to the SiO4 substructure
(region of –100ppm) is noteworthy indicating that the
diphosphine 5 in good yield (Scheme 1). Treatment of 4 by
methyl iodide, BH3·THF, and W(CO)5·THF gives rise re-
spectively and quantitatively to the diphosphonium salt 7, to
the diphosphine borane complex 8, and to the diphosphine
tungsten pentacarbonyl complex 9 (Scheme 2).
2) Sol-gel processing of diphosphines 4 and 5 and of
diphosphine derivatives 7–9
We previously showed (5) that phosphines 1 and 2 re-
quired an acidic catalyst to be hydrolyzed, therefore
diphosphine 4 was hydrolyzed and polycondensed with 6
equivalents of water in the presence of 10% of HCl in THF
solution (eq. [1] and Table 1). At 30°C, the gelation occurs
within 2 h which is close to the gelation time for phosphine
2 (5). Xerogel X4 was obtained after ageing at 30°C for five
days. It was then powdered and washed twice with ethanol,
acetone, and diethyl ether and dried in vacuo at 120°C for
two hours. The solid state 31P NMR spectrum of X4 displays
2 The degree of condensation was calculated according to the equation: τ = [(0.5)(area T1) + (1.0)(area T2) + (1.5)(area T3)]/1.5.
© 2000 NRC Canada

Bezombes et al.
1521
Table 1. Textural and solid state NMR data for xerogels X4, X5, and X7–X9.
Precursor
Gelification time
SBET (m2 g–1)
31P CP-MAS NMR (δ)
13C CP-MAS NMR (δ)
28Si CP-MAS NMR (δ)
4
7
8
9
5
2 h
460
<10
75
380
570
–5.0
+22.0
–5.8, +22.0 (50/50)
+20.9
–5.1
24.8, 66.7, 134.4
–60.5 (T1), –69.1 (T2)
1 min
15 min
20 min
30 h
–67.7 (T2)
–61.6 (T1), –69.3 (T2), –
77.0 (T3) –50 to –30 (D)
Table 2. Textural data for the materials prepared by hydrolysis and polycondensation of phosphines 1, 2, and 4.
Precursor
Xerogel
SBET (m2 g–1)
τa
Nb
Micropore vol. (mL g–1)
Mean pore dia. (Å)
1 ( f c = 6)
2 ( f c = 9)
4 ( f c = 12)
X1d
X2d
X4d
< 10
400
460
62
57
55
3.7
5.1
6.6
—
0.175
0.182
—
15
15.5
a Degree of condensation, determined by 29Si CP-MAS NMR.
b Average number of Si—O—Si bonds by precursor (N = f × τ).
c Functionality at silicon.
d HCl, 10% as catalyst.
integrity of the organic moiety had been maintained during
the sol-gel process. The surface area (N2 BET measurement)
is of 460 m2/g, value comparable to that found for the hy-
curred probably either by hydrolysis or by reaction of the Si-
OH present in the xerogel or by both processes. The partici-
pation of the Si-OH groups in the cleavage of the P—B
bonds is likely as elemental analysis showed that most of the
B atoms are incorporated in the xerogel after washing, prob-
ably as Si-O-B groups.
Thus the behaviour of diphosphines 4 and 5 and
diphosphine derivatives 7–9 during the hydrolysis process is
very similar to those of phosphines 2 and 3 as well as that of
the corresponding derivatives of phosphine 2 (5).
drolysis of phosphine
2 (Table 2). The adsorption–
desorption isotherms for X4 indicated that the solid is
microporous with a microporous volume of 0.182 mL g–1
and mean pore diameter of 15.5 Å.
The hydrolysis of diphosphine 5 was performed in THF
solution in the presence of
a catalytic amount of
tetrabutylammonium fluoride (10) with 6 equivalents of wa-
ter. The reagents were mixed at 0°C and the solution was al-
lowed to warm to 24°C after 8 min. The gel was formed
after 30 h. After the usual work-up, xerogel X5 was ob-
tained. Its N2 BET surface area was found of 570 m2 g–1,
with a microporous volume of 0.247 mL g–1. There is no ox-
idation of the phosphorus atoms during the hydrolysis as in-
dicated by the solid state 31P NMR spectrum of X5 which
displays only one signal at –5.1 ppm. The solid state 29Si
NMR spectrum of X5 displays a set of three resonances at
–61.6 ppm (substructure T1), –69.3 ppm (substructure T2),
and –77.0 ppm (substructure T3); and a broad resonance
(–50 to –30 ppm) corresponding to CSi(H)O2 substructures.
This is consistent with the FTIR spectrum which shows a
Si—H absorption band at 2160 cm–1. No peak corresponding
to an SiO4 substructure was found around –100 ppm indicat-
ing that no Si—C cleavage occurred during the sol-gel pro-
cess.
Phosphorus derivatives 7–9 were hydrolyzed and
polycondensed under the same conditions as those previ-
ously described for phosphine 4 to give xerogels X7-X9
(eqs. [2]–[4] and Table 1). The solid state 31P NMR spectra
of xerogels X7 and X9 showed that there was neither oxida-
tion nor decomplexation during the polycondensation reac-
tions. In the case of the diphosphine derivative 8 partial
decomplexation took place during the hydrolysis. The P—B
bond cleavage was estimated by 31P NMR spectroscopy to
be about 40% after reaction without drying the xerogel. Af-
ter heating for one night at 120°C, the cleavage of the P—B
bond was increased to 70%. This P—B bond cleavage oc-
3) Reactivity of the xerogel X4 with H2O2, S8, CH3I,
and W(CO)5·THF
To investigate the accessibility of the phosphorus centers
included into X4, we treated it with an excess of the follow-
ing reagents: H2O2, S8, CH3I, and W(CO)5·THF (Scheme 3).
First we studied the oxidation of the phosphorus centres
by reacting X4 with an aqueous solution of H2O2 (60 equiv).
The solid state 31P NMR spectrum of the resulting xerogel
showed the absence of the signal attributed to the phosphino
group (–5.0 ppm) and the presence of a signal at 28.1 ppm
assigned to the phosphine oxide centers (5).
X4 was treated with sulphur (13 equiv) in n-butanol
heated under reflux for 48 h. The solid state 31P NMR spec-
trum of the resulting xerogel exhibited only a signal at
41.3 ppm assigned to the thiophosphoryl centres (5).
Treatment of X4 with an excess of methyl iodide for 48 h
in toluene heated under reflux gave rise to a xerogel, the
solid state 31P NMR spectrum of which displayed a signal at
21.5 ppm, close to that for X7 (δ = 22 ppm). This signal in-
dicates that all the phosphino groups were transformed into
phosphonium salts.
Finally X4 was treated with an excess of W(CO)5·THF (4
equiv) in THF heated under reflux for 96 h. After the usual
work-up, FTIR analysis of the resulting xerogel revealed the
presence of the same CO stretching bands as those observed
in X9, which characterize the expected LW(CO)5. The solid
state 31P NMR spectrum of this xerogel displays a weak sig-
nal at 22.0 ppm corresponding to about 20% of complexed
© 2000 NRC Canada

1522
Can. J. Chem. Vol. 78, 2000
densation τ of the xerogel and the number f of hydrolysable
SiX3 groups of the precursor (N = f × τ) (see Table 2). We
observed that the greater the N value, the more difficult was
the accessibility of the phosphorus centres by the rather
bulky reagent W(CO)5·THF. The present study of the acces-
sibility of the phosphorus centres incorporated into X4 con-
firms this observation.
phosphino centres and a large signal at –7.3 ppm corre-
sponding at about 80% of unchanged phosphino groups
(Fig. 1). Thus, most of the phosphino groups in xerogel X4
are not accessible by W(CO)5·THF under the conditions
used for the reaction.
Discussion
We consider that the number of Si—O—Si bonds should
have an influence on the mobility of the network of the
xerogel. The more important the mobility of the organic
moieties, the easier is the penetration and the diffusion of
the reagents into the hybrid organic-inorganic material (11,
12). Thus the diffusion of the reagent through X1, material
in which the average number of Si—O—Si bonds by organic
moiety is N = 3.7, is much easier than through X4 in which
the organic moieties are more rigid (N = 6.6). This interpre-
tation is supported by 31P NMR data of xerogels X1, X2,
and X4. Indeed as shown on Fig. 2, the broadening of the
31P NMR signals for xerogels increases with the average
number N of Si—O—Si bonds attached to the organic moi-
eties. The less flexible the organophosphorus moieties, the
larger is the 31P NMR signal of the phosphorus atoms (13).
All the phosphorus atoms of the xerogel X4 are accessible
to the small reagents H2O2, S8, and CH3I as was previously
observed for the xerogels X1 and X2 obtained from the
phosphines 1 and 2 respectively (Table 2) (5). In contrast,
the reaction with the rather bulky reagent W(CO)5·THF al-
lowed discrimination between the xerogels X1, X2, and X4.
Indeed, only 20% of phosphorus atoms included into X4
were complexed by W(CO)5 while the complexation was
complete for X1 and was 80% for X2 (Table 3).
In our previous publication (6) we have observed that the
average number (N) of Si-O-Si bonds attached to each
organophosphorous moiety in the material was an important
parameter controlling the accessibility of the phosphino cen-
tres. This number N is a function of both the degree of con-
© 2000 NRC Canada

Bezombes et al.
1523
Fig. 1. Solid state 31P NMR spectrum of the xerogel X4 after re-
Table 3. Solid state 31P NMR chemical shifts (δ in ppm) of
action with W(CO)5·THF
phosphorus atoms after treatment of the xerogels X1, X2, and
X4 with W(CO)5·THF. The percent of each signal is indicated in
parentheses.
31P HPDEC MAS NMR
Xerogel
Complexed centers
Unchanged centers
X1 ( f = 6)
X2 ( f = 9)
X4 ( f = 12)
21.1 (100%)
20.8 (80%)
22.0 (20%)
—
–7.3 (20%)
–5.9 (80%)
Elmer 1600 FTIR spectrophotometer by the DRIFT method.
The solution NMR spectra were recorded on a Bruker AC-
200 (29Si), Bruker DPX-200 (1H and 13C), and Bruker WP
250 SY (31P, 11B). Chemical shifts (δ in ppm) were refer-
enced to Me4Si (1H, 13C, 29Si), BF3·THF (11B), or H3PO4
(31P). The CP MAS solid state 29Si NMR spectra were re-
corded on a Bruker FTAM 300 as well as CP MAS 13C solid
state NMR spectra in that case by using the TOSS tech-
nique. In both cases the repetition time was 5 and 10 s with
Thus these NMR data confirm that the mobility of
organophosphorous moieties decreases as N increases, the
phosphorus atoms being less readily accessible to reagents
owing to a lower swellability of the matrix.
contact times of 5 and 2 milliseconds. The HPDEC MAS 31
P
In conclusion, this study and the preceding one (6) clearly
show that the average number N of Si—O—Si bonds at-
tached to the organic moiety is an important parameter con-
trolling the accessibility of the phosphorus centers included
in an hybrid organic-inorganic material. This shows the
importance of the number of the hydrolysable SiX3 bonds in
the organic precursor. This parameter has been well evi-
denced because the precursors 1, 2, as well as 4 all bear
rigid groups around the phosphorus centres. It is likely that
in the case of phosphines with flexible groups around the
phosphorus atoms, the situation is different.
solid state NMR spectra were recorded on a Bruker FTAM
300, a Brucker ASX 200, or a Brucker ASX 400 with repeti-
tion time of 5 s. FAB mass spectra (matrix, m-nitrobenzyl
alcohol (NBA), thioglycerol (GT)) were registered on Jeol
JMS-D3000 spectrometer. Specific surface areas were deter-
mined by the Brunauer-Emmett-Teller (BET) method on
Micromeritics ASAP 2010 and Micromeritics Gemini III
2375 analysers. Elemental analyses were carried out by the
Service Central de Micro- Analyse du CNRS.
P,P,P ′,P ′-tetrakis(4-triisopropyloxysilylphenyl)p-phenylene-
diphosphine (4): P,P,P′,P′-tetrachloro-p-phenylenediphosphine
(14) (5.28 g, 18.9 mmol) in THF (50 mL) are added
dropwise, at 0°C, to a solution of 4-triisopropyloxysilyl-
phenylmagnesium bromide (77.4 mmol) in THF (200mL).
The reaction mixture was stirred for 12 h at room tempera-
ture and then was heated under reflux for one hour. The
THF was removed under vacuum and pentane (250 mL) was
Experimental
All reactions were carried out under argon by using a vac-
uum line. Solvents were dried and distilled just before use.
Melting points were determined with a Gallenkamp appara-
tus and are uncorrected. IR data were obtained on a Perkin-
Fig. 2. Half line width (∆ν1/2, 1100 Hz for X1, 1400 Hz for X2, 2000 Hz for X4) of the solid state 31P NMR signals of xerogels X1,
X2, and X4 as a function of the average number N of Si-O-Si bonds (3.7 for X1, 5.1 for X2, 6.6 for X4).
© 2000 NRC Canada

1524
Can. J. Chem. Vol. 78, 2000
added. After filtration and removal of the solvent under vac-
uum, 2-propanol (30 mL) was added under stirring to the
oily residue (23.6 g). The resulting powder was
recrystallised from 2-propanol to give 11.7 g, (10.1 mmol,
53%) of white crystals. Mp (2-propanol) 172–175°C; 1H
1263 [(M
–
2BH3
+
H)+, 41%]. Anal. calc. for
C66H110O12B2P2Si4 : C 61.39, H 8.53%; found: C 60.86, H
8.35.
P,P,P ′,P ′-tetrakis(4-triisopropyloxysilylphenyl)p-phenylene-
diphosphine bis tungsten pentacarbonyl (9): A THF solu-
tion of W(CO)5·THF (15) (150 mL, 8.55 mmol) was added
to 2.61 g (2.07 mmol) of 4. After stirring for two hours at
room temperature, the solvent was evaporated under vac-
uum. After elimination of excess of W(CO)5·THF by subli-
mation at 60°C under vacuum, 3.71 g (1.94 mmol, 94%) of
3
NMR (δ, 200 MHz, CDCl3) 1.24 (d, JH-H = 6.1 Hz, 72H,
3
Me), 4.30 (spt, JH-H = 6.1 Hz, 12H, OCH), 7.25–7.36
(m, 12H, aromatic), 7.65–7.70 (m, 8H, aromatic); 13C NMR
(δ, 50 MHz, CDCl3) 25.9 (Me), 65.9 (OCH), 133–139 (aro-
matic); 29Si NMR (δ, 40 MHz, CDCl3) –61.9; 31P NMR
(δ, 100 MHz, CDCl3) –4.9; MS (FAB+, GT) 1263 [(M + H)+,
32%], 1295 [(M + 2 oxygen + H)+, 45%]. Anal. calc. for
C66H104O12P2Si4 : C 62.75, H 8.24%; found: C 62.76, H
8.39.
1
crude 9 was obtained. Mp (2-propanol) 203.6 (decomp.); H
NMR (δ, 200 MHz, CDCl3) 1.25 (d, JH-H = 6.1 Hz, 72H,
3
3
Me), 4.31 (spt, JH-H = 6.1 Hz, 12H, OCH), 7.44–7.80 (m,
20H, aromatic); 13C NMR (δ, 50 MHz, CDCl3) 25.9 (Me),
2
P,P,P ′,P ′-tetrakis(4-trihydrosilylphenyl)p-phenylenediphos-
phine (5): Diphosphine 4 (10.4 g, 8.23 mmol) in diethyl
ether (80 mL) was added dropwise at 0°C to a suspension of
LiAlH4 (1.92 g, 50.5 mmol) in diethyl ether (60 mL). The
reaction mixture was stirred for 3 h at room temperature and
the solvent was then evaporated under vacuum to give a
white solid which was dissolved in pentane (500 mL). After
filtration of the salts and evaporation of the solvent under
vacuum, 4.24 g (7.49 mmol) of raw diphosphine was ob-
66.1 (OCH), 132.4–138.9 (aromatic), 197.4 (s and d JP-C
=
2
6.8 Hz, CO cis), 199.4 (s and d JP-C = 22 Hz, CO trans);
29Si NMR (δ, 40 MHz, CDCl3) –63.3 (d, 5JP-Si = 1.2 Hz); 31
P
5
NMR (δ, 100 MHz, CDCl3) 21.7 (s d, JP-Si = 1.2 Hz); IR
(ν/cm–1, CCl4) 1942, 1980, 2071; MS (FAB+, NBA) 1911
[(M + H)+, 14%], 1587 [(M – W(CO)5 + H)+, 18%], 1263
[(M
–
2W(CO)5
+
H)+, 16%]. Anal. calc. for
C76H104O22P2Si4 W2: C 47.75, H 5.44%; found: C 47.89, H
5.84.
1
tained. Mp 172 (decomp.); H NMR (δ, 200 MHz, CDCl3)
4.23 (s and d (satellite 29Si, 1JSi-H = 202 H), 9H, SiH), 7.23–
7.62 (m, 20H, aromatic ; 13C NMR (δ, 50 MHz, CDCl3)
130–139 (aromatic) ; 29Si NMR (δ, 40 MHz, CDCl3) –58.7 ;
31P NMR (δ, 80 MHz, CDCl3) –4.6 ; IR (ν/cm–1, CCl4) 2160;
MS (FAB+, NBA) 567 [(M + H)+, 40%], 583 [(M + 1 oxy-
gen + H)+, 56%]. Anal. calc. for C30H32P2Si4: C 63.60, H
5.65%; found: C 61.86, H 5.96.
Xerogel X4: To a THF (3.2 mL) solution of 4 (2.00 g,
1.58 mmol) in a 12.5 mL flask were added dropwise at room
temperature 1.6 mL of a 6 M H2O (9.6 mmol) and 0.1 M
HCl (0.16 mmol) solution in THF. The mixture was stirred
for 5 min, then the flask was placed in a water-bath at 30°C
without stirring. Gelation occurred after 3h. The wet whitish
gel was allowed to age for 5 days at 30°C after which it was
powdered and washed with ethanol, acetone and ether. The
powdering and washing were repeated once and the gel was
powdered again and dried in vacuum for 2 h at 120°C to yield
1.14 g of X4 as a white powder. Specific surface area, 31P,
13C, and 29Si CP-MAS NMR are indicated in Table 1. Ele-
mental anal. calc. for C30H20O6PSi4: C 55.38, H 3.07, O
14.77, P 9.54, Si 17.23; found: C 51.80, H 5.13, O 20.87, P 8.05,
Dimethyl[P,P,P ′,P ′-tetrakis(4-triisopropyloxysilylphenyl)p-
phenylene]diphosphonium diiodide (7): Diphosphine 4 (2.38 g,
1.88 mmol) and methyl iodide (0.5 mL, 8.03 mmol) were
heated under reflux in toluene (20 mL) for 3 h. After evapo-
ration of the solvent, 2.87 g (1.86 mmol, 99%) of crude 7
1
was obtained. Mp (2-propanol) 284.6 (decomp.) ; H NMR
3
(δ, 200 MHz, CDCl3) 1.20 (d, JH-H = 6.1 Hz, 72H, Me),
Si 14.15 which corresponds to C33.24H39.51O10.04P2.00Si3.89
.
2
3
3.23 (d, JP-H = 13.3 Hz, 6H, PMe), 4.28 (spt, JH-H
=
6.1 Hz, 12H, OCH), 7.78–8.03 (m, 20H, aromatic); 13C
Xerogel X5: To a THF (5.7 mL) solution of 5 (1.76 g,
3.10 mmol) in a 20 mL test tube was added dropwise at 0°C
in 5 min, under stirring, 3.1 mL of a 6 M H2O (16.2 mmol)
and 0.01 M nBu4NF (0.03 mmol) solution in THF. The reac-
tion mixture was then kept at room temperature without stir-
ring. A release of dihydrogen was observed until gel
formation (30h). The wet gel was left to age for 5 days at
room temperature and the solid was subsequently treated as
it was described before to give 1.93 g of white xerogel X5.
Specific surface area, 31P and 29Si CP-MAS NMR are indi-
cated in Table 1. IR (DRIFT) (ν/cm–1) 2159 (Si-H). Elemen-
tal anal. calc. for C30H20O6PSi4: C 55.38, H 3.07, O 14.77, P
9.54, Si 17.23; found: C 53.82, H 4.36, O 15.87, P 9.25, Si
1
NMR (δ, 50 MHz, CDCl3) 11.8 (d, JP-C = 56.7 Hz, PC),
25.9 (Me), 66.5 (OCH), 112–143 (aromatic) ; 29Si NMR
(δ, 40 MHz, CDCl3) –65.8 ; 31P NMR (δ, 100 MHz, CDCl3)
22.7 ; MS (FAB+, GT) 1293 [(M – 2I)+, 100%]. Anal. calc.
for C68H110O12P2Si4: C 52.78, H 7.11%; found: C 52.01, H
7.41.
P,P,P ′,P ′-tetrakis(4-triisopropyloxysilylphenyl)p-phenylene-
diphosphine diborane (8): A molar solution of BH3·THF in
THF (4.1 mL, 4.1 mmol) was added dropwise, at 0°C, to
2.56 g (2.03 mmol) of 4 in THF (10 mL). After stirring one
hour at room temperature the solvent was removed to give
2.61 g (2.02 mmol, 99%) of crude 8. Mp 163.7 (decomp.);
1H NMR (δ, 200 MHz, CDCl3) 1.25 (d, 3JH-H = 6.1 Hz, 78H,
16.70 which corresponds to C30.06H29.22O6.65P2.00Si3.99
.
3
Me+BH), 4.31 (spt, JH-H = 6.1 Hz, 12H, OCH), 7.54–7.81
Xerogel X7: To a THF (1.3 mL) solution of 2 g (1.29 mmol)
of 7 in a 12.5 mL flask were added dropwise at room tem-
perature 1.3 mL of a 6 M H2O (7.8 mmol) and 0.1 M HCl
(0.13 mmol) solution in THF. The reaction mixture was stirred
and gel formation occurred after one min. After drying,
1.32g of yellow X7 was obtained. Elemental anal. calcd. for
(m, 20H, aromatic); 13C NMR (δ, 50 MHz, CDCl3)
25.9 (Me), 66.1 (OCH), 128–138 (aromatic); 11B NMR
(δ, 80 MHz, CDCl3) –38.5 (broad signal); 29Si NMR
5
(δ, 40 MHz, CDCl3) –63.5 (d, JP-Si = 1.2 Hz); 31P NMR (δ,
100 MHz, CDCl3) 22.1 (broad signal); MS (FAB+, GT)
© 2000 NRC Canada

Bezombes et al.
1525
C32H26I2O6P2Si4 : C 41.11, H 2.78, I 27.20, O 10.28, P 6.64,
Si 11.99; found: C 41.30, H 4.80, I 20.45, O 17.50, P 5.80,
8.96, Si 15.69%; found: C 49.91, H, 4.68, O 18.75, P 6.90, S 6.19,
Si 13.75%, which corresponds to C37.37H42.00O10.53P2.00S1.74Si4.41.
Si 10.15 which corresponds to C36.79H41.30 1.72O11.69P2.00Si3.87.
I
Reactivity of X4 with MeI
334 mg (0.514 mmol) of X4 and 0.25 mL (4.0 mmol) of
methyl iodide were heated under reflux in toluene (10ml) for
48 h. After filtration, the precipitate was washed with etha-
nol, acetone and diethyl ether, and dried to give 398 mg of a
yellow powder. 31P NMR (δ, 162 MHz, HPDEC MAS) 21.4.
SBET < 10 m2/g. Elemental anal. calc. for C36H26I2O6P2Si4:
C 41.11, H 2.78, I 27.20, O 10.28, P 6.64, Si 11.99%; found:
C 43.40, H 4.53, I 15.32, O 17.25, P 7.30, Si 12.20 which
Xerogel X8: To a THF (1.55 mL) solution of 2.0 g (1.55 mmol)
of 8 in a 12.5 mL flask were added dropwise at room tem-
perature 1.55mL of a 6 M H2O (9.3 mmol) and 0.1 M HCl
(0.15 mmol) solution in THF. The reaction mixture was
stirred for 5 min and kept at 30°C without stirring. Gel for-
mation occurred after 15 min. However, after about 10 min a
slight release of gas was observed so that the cork of the
flask was pierced. After ageing, powdering and washing,
1.31 g of white powder was obtained. Drying 2 h at 120°C
of 1.01 g of crude xerogel gave 882 mg of X8. 31P CP MAS
NMR spectra indicated a cleavage of P—B bonds of 50%.
Elemental anal. calcd. for C30H26B2O6P2Si4: C 53.10, H
3.83, B 3.24, O 14.16, P 9.14, Si 16.52; found: C 50.47, H
5.17, B 1.95, O 22.01, P 6.20, Si 14.20 which corresponds to
C42.06H51.7B1.77O13.76P2.00Si5.07. The cleavage of the P—B
bond of the crude xerogel is 40%. It is of 70% if the crude
xereogel is heated one night at 120°C.
corresponds to C30.72H38.47 I1.02 O9.16P2.00Si3.70
.
Reactivity of X4 with W(CO)5·THF
302 mg (0.46 mmol) of X4 and 70 mL of a 0.057 M THF
solution of W(CO)5·THF (4.0 mmol) were heated under re-
flux and stirring for 96 h. The greenish suspension was then
filtered and the precipitate was washed twice with THF (2 ×
30 mL) then with ethanol, acetone, and ethyl ether. After
drying under vacuum for 2 h at 120°C, 344 mg of a pale
green solid was obtained. 31P NMR (δ, 120 MHz, HPDEC
MAS) 20.8 (20%), –5 (80%). IR (DRIFT ν/cm–1, KCl) 1935,
1984, 2072. SBET = 380 m2/g. Elemental anal. calc for
C40H20O16P2Si4W2: C 36.98, H 1.54, O 19.72, P 4.78,
Si 8.63, W 28.35; found: C 43.67, H 3.58, O 23.25, P
Xerogel X9: To a THF (1.0 mL) solution of 2.0 g (1.05 mmol)
of 9 in a 12.5 mL flask were added dropwise at room tem-
perature 1.1mL of a 6 M H2O (6.6 mmol) and 0.1 M HCl
(0.11 mmol) solution in THF. The reaction mixture was
stirred for 5 min and kept at 30°C without stirring. Gel for-
mation occurred after 20min. After drying, 1.38 g of light
beige X9 was obtained. IR (DRIFT ν/cm–1, KCl) 1934, 1986,
2072. Specific surface area and 31P CP-MAS NMR data are
indicated in Table 1. Elemental anal. calcd. for
C40H20O16P2Si4W2: C 36.98, H 1.54, O 19.72, P 4.78, Si
8.63, W 28.30; found : C 30.33, H 2.88, O 28.55, P 3.81, Si 7.58,
5.90, Si 11.80,
W
11.80% which corresponds to
C
38.24H37.62O15.27P2.00Si4.43W0.67
.
References
1. U. Schubert, N. Hüsing, and A. Lorenz. Chem. Mater. 7, 2010
(1995).
2. D.A. Loy and K.J. Shea. Chem. Rev. 95, 1431 (1995).
3. R.J.P. Corriu and D. Leclercq. Angew. Chem. Int. Ed. Engl.
35, 1420 (1996); G. Cerveau and R.J.P. Corriu. Coord. Chem.
Rev. 178–180, 1051 (1998).
W 26.85 which corresponds to C41.13H46.86O29.03P2.00Si4.40W2.37
.
Reactivity of X4 with H2O2
4. R. Corriu. C. R. Acad. Sci. Ser. II 83 (1998); R. Corriu. Poly-
hedron, 5–6, 925 (1998).
5. J.P. Bezombes, C. Chuit, R.J.P. Corriu, and C. Reyé. J. Mater.
Chem. 8, 1749 (1998).
6. J.P. Bezombes, C. Chuit, R.J.P. Corriu, and C. Reyé. J. Mater.
Chem. 9, 1727 (1999).
7. E.M.J. Evleth, L.V.D. Freeman, and R.I. Vagner. J. Org. Chem.
27, 2192 (1962).
8. K.J. Shea, D.A. Loy, and O. Webster. J. Am. Chem. Soc. 114,
6700 (1992).
9. G. Cerveau, R.J.P. Corriu, C. Lepeytre, and P.H. Mutin. J.
Mater. Chem. 8, 2707 (1998).
To a suspension of 325 mg (0.50 mmol) of X4 in CH2Cl2
(20ml) were added dropwise at room temperature 5 mL of a
35% aqueous H2O2 solution (60 mmol). After 13 h stirring
at room temperature, the suspension was filtered off and the
precipitate was washed with water, ethanol, acetone, and di-
ethyl ether. After drying, 296 mg of a white powder were
obtained. 31P NMR (δ, 162 MHz, HPDEC MAS) 28.1. SBET
< 370 m2/g. Elemental anal. calc for C30H20O8P2Si4: C 52.78,
H 2.93, O 18.77, P 9.09, Si 16.42; found: C 47.39, H 4.34,
O
26.07, P 7.35, Si 14.85 which corresponds to
C33.31H36.61O13.74P2.00Si4.47
.
10. R.J.P. Corriu, J.J.E. Moreau, and M. Wong Chi Man. J.
Sol-Gel Sci. Technol. 2, 87 (1994).
11. E. Lindner, M. Kemmler, T. Schneller, and H.A. Mayer. Inorg.
Chem. 34, 5489 (1995).
12. E. Lindner, T. Schneller, H.A. Mayer, H. Bertagnolli, T.S. Ertel,
and W. Hörner. Chem. Mater. 9, 1524 (1997).
13. E. Lindner, R. Schreiber, M. Kemmler, T. Schneller, and H.A.
Mayer. Chem. Mater. 7, 951 (1995).
14. E.M.J. Evleth, L.V.D. Freeman, and R.I. Wagner. J. Org.
Chem. 27, 2192 (1962).
Reactivity of X4 with S8
333 mg (0.51 mmol) of X4, 426 mg (13.3 mmol) of sul-
phur, and 24 mL of n-butanol were heated under reflux for
48 h. The hot suspension was then filtered (to remove the
excess of sulphur) and the precipitate was washed twice with
hot n-butanol (2 × 20 mL) then with acetone and diethyl
ether. After drying under vacuum for 2 h at 120°C, 346 mg
of a whitish solid was obtained. 31P NMR (δ, 120 MHz,
HPDEC MAS) 41.3. SBET = 415 m2/g. Elemental anal. calc.
for C30H20O6P2S2Si4: C 50.42, H 2.80, O 13.45, P 8.68, S
15. R.B. King and N.D. Sadanani. Inorg. Chem. 24, 3136 (1985).
© 2000 NRC Canada