Reactivity studies, structural characterization, and thermolysis of cubic titanosiloxanes: Precursors to titanosilicate materials which catalyze olefin epoxidation

Article abstract of DOI:10.1021/ic034317m

The cubic titanosiloxane [RSiO₃Ti(OPr′)]₄ (R = 2,6-Pr₂′C₆H₃NSiMe₃) (1) is found to be relatively inert in its attempted reactions with alcohols and other acidic hydrogen containing compounds. The reaction of 1 with silanol (Bu^tO)₃-SiOH however proceeds over a period of approximately 3 months to result in the hydrolysis of (Bu^tO)₃SiOH and yield the transesterification product [RSiO₃Ti(OBu^t)]₄ (2) rather than the expected [RSiO₃Ti(OSi(OBu^t)₃)]₄. Products 1 and 2 have been characterized by elemental analysis, thermal analysis, and spectroscopic techniques (IR, El-MS, and NMR). The solid-state structures of both 1 and 2 have been determined by single-crystal X-ray diffraction studies. Compounds 1 and 2 are isomorphous and crystallize in a cubic space group with a central cubic Ti₄Si₄O₁₂ core. Solid state thermolysis of 1 was carried at 450, 600, 800, 900, 1000, and 1200 °C in air, and the resulting titanosilicate materials 1a-f were characterized by spectroscopic (IR and DR UV), powder XRD, and electron microscopic methods. While, the presence of Ti-O-Si linkages appears to be dominant in the samples prepared at lower temperatures (450-800 °C), phase separation of anatase and rutile forms of TiO₂ occurs at temperatures above 900 °C as revealed by IR spectral and PXRD studies. The presence of octahedral titanium centers was observed by DR UV spectroscopy for the samples heated at higher temperatures. The use of new titanosilicate materials as catalysts for olefin epoxidation has been investigated. The titanosilicate materials produced at temperatures below 800 °C with a large number of Ti-O-Si linkages (or tetrahedral titanium centers) were found to be more active catalysts compared to the materials produced above 900 °C. The observed conversion in the epoxidation reactions was found to be somewhat low although the selectivity of the epoxide formation over the other possible oxidized products was found to be very good.

Full text of DOI:10.1021/ic034317m

Inorg. Chem. 2003, 42, 4696−4706
Reactivity Studies, Structural Characterization, and Thermolysis of
Cubic Titanosiloxanes: Precursors to Titanosilicate Materials Which
Catalyze Olefin Epoxidation†
Ramaswamy Murugavel,* Paul Davis, and Vivekanand S. Shete
Department of Chemistry, Indian Institute of TechnologysBombay, Powai, Mumbai-400 076, India
Received March 25, 2003
The cubic titanosiloxane [RSiO Ti(OPr)]4 (R ) 2,6-Pr2iC H NSiMe3) (1) is found to be relatively inert in its attempted
i
3
6 3
t
reactions with alcohols and other acidic hydrogen containing compounds. The reaction of 1 with silanol (BuO)3-
SiOH however proceeds over a period of approximately 3 months to result in the hydrolysis of (Bu O)3SiOH and
yield the transesterification product [RSiO Ti(OBu)]4 (2) rather than the expected [RSiO Ti(OSi(OBu)3)]4. Products
t
t
t
3
3
1 and 2 have been characterized by elemental analysis, thermal analysis, and spectroscopic techniques (IR, EI-
MS, and NMR). The solid-state structures of both 1 and 2 have been determined by single-crystal X-ray diffraction
studies. Compounds 1 and 2 are isomorphous and crystallize in a cubic space group with a central cubic Ti4Si4O
12
core. Solid state thermolysis of 1 was carried at 450, 600, 800, 900, 1000, and 1200 °C in air, and the resulting
titanosilicate materials 1a−f were characterized by spectroscopic (IR and DR UV), powder XRD, and electron
microscopic methods. While, the presence of Ti−O−Si linkages appears to be dominant in the samples prepared
at lower temperatures (450−800 °C), phase separation of anatase and rutile forms of TiO occurs at temperatures
2
above 900 °C as revealed by IR spectral and PXRD studies. The presence of octahedral titanium centers was
observed by DR UV spectroscopy for the samples heated at higher temperatures. The use of new titanosilicate
materials as catalysts for olefin epoxidation has been investigated. The titanosilicate materials produced at
temperatures below 800 °C with a large number of Ti−O−Si linkages (or tetrahedral titanium centers) were found
to be more active catalysts compared to the materials produced above 900 °C. The observed conversion in the
epoxidation reactions was found to be somewhat low although the selectivity of the epoxide formation over the
other possible oxidized products was found to be very good.
Introduction
materials.9 Different approaches have been employed by
various workers in this area to achieve this goal. For example,
it is possible to tailor the silicon:metal ratio in these systems
Silicon-oxygen linkage is the most common chemical
bond in nature, and the heterosiloxane linkage is the most
prominent structural moiety in silicate materials.1 Silanols
containing one or more -OH groups on silicon2 have been
extensively used in the synthesis of heterosiloxanes which
have utility both as model compounds for larger silicate
structures3-8 and as single-source precursors for silicate
(3) (a) Murugavel, R.; Voigt, A.; Walawalkar, M. G.; Roesky, H. W.
Chem. ReV. 1996, 96, 2205. (b) Murugavel, R.; Chandrasekhar, V.;
Roesky, H. W. Acc. Chem. Res. 1996, 29, 183.
(4) Beckmann, J.; Jurkschat, K. Coord. Chem. ReV. 2001, 215, 267.
(5) (a) Lorenz, V.; Fischer, A.; Giessmann, S.; Gilje, J. W.; Gun’ko, Y.;
Jacob, K.; Edelmann, F. T. Coord. Chem. ReV. 2000, 206-207, 321.
(b) King, L.; Sullivan, A. C. Coord. Chem. ReV. 1999, 189, 19.
(6) Feher, F. J.; Budzichowski, T. A. Polyhedron 1995, 14, 3239.
(7) Duchateau, R. Chem. ReV. 2002, 102, 3525.
* To whom correspondence should be addressed. E-mail: rmv@iitb.ac.in.
Fax: +(22) 2572 3480. Tel: +(22) 2576 7163.
(8) Abbenhuis, H. C. L. Chem.sEur. J. 2000, 6, 25.
† Dedicated to Professor C. N. R. Rao on the occasion of his 70th
birthday.
(9) (a) Fujdala, K. L.; Oliver, A. G.; Hollander, F. J.; Tilley, T. D. Inorg.
Chem. 2003, 42, 1140. (b) Brutchey, R. L.; Goldberger, J. E.; Koffas,
T. S.; Tilley, T. D. Chem. Mater. 2003, 15, 1040. (c) Tilley, T. D. J.
Mol. Catal., A 2002, 182-183, 17. (d) Fujdala, K. L.; Tilley, T. D.
Chem. Mater. 2002, 14, 1376. (e) Lugmair, C. G.; Fujdala, K. L.;
Tilley, T. D. Chem. Mater. 2002, 14, 888. (f) Fujdala, K. L.; Tilley,
T. D. J. Am. Chem. Soc. 2001, 123, 10133. (g) Kriesel, J. W.; Sander,
M. S.; Tilley, T. D. AdV. Mater. 2001, 13, 331.
(1) Walawalkar, M. G.; Murugavel, R.; Roesky, H. W. Tailor made Silicon
Oxygen Compounds-From Molecules to Materials; Corriu, R., Jutzi,
P., Eds.; Vieweg: Braunschweig, Germany, 1996; p 61.
(2) (a) Lickiss, P. D. AdV. Inorg. Chem. 1995, 42, 147. (b) Murugavel,
R.; Chandrasekhar, V.; Voigt, A.; Roesky, H. W.; Schmidt, H.-G.;
Noltenmeyer, M. Organometallics 1995, 14, 5298.
4696 Inorganic Chemistry, Vol. 42, No. 15, 2003
10.1021/ic034317m CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/02/2003

Molecular Titanosiloxanes
Chart 1. Molecular Titanosiloxanes Derived from Different Types of
Silanols with Varying Si:Ti Ratios Given in Parentheses
by suitably choosing a silanol with the desired number of
hydroxyl groups.
The reactions of silanols (such as R3SiOH, R2Si(OH)2,
RSi(OH)3) with a series of titanium alkoxides, alkyls, halides,
and amides have been investigated very intensively with the
aim to produce model compounds for well-known Ti-based
silicate and silicalite materials,10a which show interesting
catalytic applications in several organic transformations
including oxidation of olefins to epoxides.10 The range of
titanosiloxanes that have been synthesized thus far using a
molecular approach show very interesting silicon-to-titanium
ratio in the final products obtained (Chart 1).9a,11-14,15a
Particularly interesting among these are the cubic titano-
siloxanes G14a and H13a which have been derived from
(10) (a) Murugavel, R.; Roesky, H. W. Angew. Chem., Int. Ed. Engl. 1997,
36, 477. (b) Laha, S. C.; Kumar, R. J. Catal. 2002, 208, 339. (c) Wang,
X.-S.; Guo, X. W.; Li, G. Catal. Today 2002, 74, 65. (d) Laha, S. C.;
Kumar, R. J. Catal. 2001, 204, 64. (e) Schmidt, I.; Krogh, A.;
Wienberg, K.; Carlsson, A.; Brorson, M.; Jacobsen, C. J. H. Chem.
Commun. 2000, 2157. (f) Bhaumik, A.; Tatsumi, T. Chem. Commun.
1998, 463.
(11) (a) Horacek, M.; Stepnicka, P.; Kubista, J.; Gyepes, R.; Cisarova, I.;
Petrusova, L.; Mach, K. J. Organomet. Chem. 2002, 658, 235. (b)
Hofmann, M.; Malisch, W.; Schumacher, D.; Lager, M.; Nieger, M.
Organometallics 2002, 21, 3485. (c) Sobota, P.; Przybylak, S.; Ejfler,
J.; Kobylka, M.; Jerzykiewicz, L. B. Inorg. Chim. Acta 2002, 334,
159. (d) Zemanek, J.; Horacek, M.; Thewalt, U.; Stepnicka, P.; Kubista,
J.; Cejka, J.; Petrusova, L.; Mach, K. Inorg. Chem. Commun. 2001,
4, 520. (e) Dell’Amico, D. B.; Calderazzo, F.; Ianelli, S.; Labella, L.;
Marchetti, F.; Pelizzi, G. J. Chem. Soc., Dalton Trans. 2000, 4339.
(f) Johnson, B. F. G.; Klunduk, M. C.; Martin, C. M.; Sankar, G.;
Teate, S. J.; Thomas, J. M. J. Organomet. Chem. 2000, 596, 221. (g)
Ciruelos, S.; Cuenca, T.; Gomez-Sal, P.; Manzanero, A.; Royo, P.
Organometallics 1995, 14, 177.
(12) (a) Murugavel, R.; Shete, V. S.; Baheti, K.; Davis, P. J. Organomet.
Chem. 2001, 625, 195. (b) Voigt, A.; Murugavel, R.; Montero, M. L.;
Wesel, H.; Liu, F.-Q.; Roesky, H. W.; Uso´n, I.; Albers, T.; Parisini,
E. Angew. Chem., Int. Ed. Engl. 1997, 36, 1001. (c) Beckmann, J.;
Jurkschat, K.; Mu¨ller, D.; Rabe, S.; Schu¨rmann, M. Organometallics
1999, 18, 2326. (d) Ho¨bbel, D.; Nacken, M.; Schmidt, H.; Huch, V.;
Veith, M. J. Mater. Chem. 1998, 8, 171. (e) Lazell, M.; Motevalli,
M.; Shah, S. A. A.; Simon, C. K. S.; Sullivan, A. C. J. Chem. Soc.,
Dalton Trans. 1996, 1449. (f) Hossain, M. A.; Hursthouse, M. B.;
Ibrahim, A.; Mazid, M.; Sullivan, A. C. J. Chem. Soc., Dalton Trans.
1989, 2347. (g) Hossain, M. A.; Hursthouse, M. B.; Mazid, M. A.;
Sullivan, A. C. J. Chem. Soc., Chem. Commun. 1988, 1305. (h) Fandos,
R.; Otero, A.; Rodriguez, A.; Ruiz, M. J.; Terreros, P. Angew. Chem.,
Int. Ed. 2001, 40, 2884. (i) Liu, F.-Q.; Schmidt, H. G.; Noltemeyer,
M.; Freire-Erdbru¨gger, C.; Sheldrick, G. M.; Roesky, H. W. Z.
Naturforsch. 1992, 47B, 1085. (j) Haoudi-Mazzah, A.; Mazzah, A.;
Schmidt, H. G.; Noltemeyer, M.; Roesky, H. W. Z. Naturforsch. 1991,
46B, 587. (k) Haoudi, A.; Dhamelincourt, P.; Mazzah, A.; Drache,
M.; Conflant, P.; Lazraq, M. J. Mater. Chem. 2000, 1001.
(13) (a) Voigt, A.; Murugavel, R.; Chandrasekhar, V.; Winkhofer, N.;
Roesky, H. W.; Schmidt, H.-G.; Uso´n, I. Organometallics 1996, 15,
1610. (b) Winkhofer, N.; Voigt, A.; Dorn, H.; Roesky, H. W.; Steiner,
A.; Stalke, D.; Reller, A. Angew. Chem., Int. Ed. Engl. 1994, 33, 1352.
(c) Lindemann, H. M.; Schneider, M.; Neumann, B.; Stammler, H.-
G.; Stammler, A.; Jutzi, P. Organometallics 2002, 21, 3009. (d) Nolte,
J. O.; Schneider, M.; Neumann, B.; Stammler, H. G.; Jutzi, P.
Organometallics 2003, 22, 1010. (e) Day, V. W.; Klemperer, W. G.;
Mainz, V. V.; Millar, D. M. J. Am. Chem. Soc. 1985, 107, 6262 (f)
Hossain, M. A.; Hursthouse, M. B.; Malik, K. M. A. Acta Crystallogr.,
Sect. B 1979, B35, 2258.
Cp′TiCl3 or Ti(OR)4 and an incompletely condensed silses-
quioxane or silanetriol, respectively. However, the framework
Si:Ti ratio obtained in each of these two types of cubic
titanosiloxanes is different; while the use of silsequioxane
results in a titanosiloxane G with a 7:1 Si:Ti ratio,14a the
use of silanetriol produces the titanosiloxane H with 1:1 Si:
Ti ratio.13a While it can be argued that the former compound
G is an excellent model for the metal incorporated silica
surfaces, the latter compound has the potential of being used
as precursor for the preparation of titanosilicate materials
with high titanium content, if the separation of TiO2 phases
could be avoided during its conversion to materials.
The TiO2 phase separation from titanosilicate materials is
normally expected in cases where either simple mononuclear
precursors such as Ti(OR)4 are used as titanium source or
when the titanium content is in excess of 2-5% (e.g.
synthesis of TS-1 and TS-2 type materials using Ti(OPri4)).10a
Hence, it was thought in this study that a preformed
titanosiloxane cage such as H with an exceptionally stable
Ti4Si4O12 core would in fact be converted into titanium-rich
(14) (a) Feher, F. J.; Budzichowski, T. A.; Rahimian, K.; Ziller, J. W. J.
Am. Chem. Soc. 1992, 114, 3859. (b) Duchateau, R.; Cremer, U.;
Harmsen, R. J.; Mohamud, S. I.; Abbenhuis, H. C. L.; van Santen, R.
A.; Meetsma, A.; Thiele, S. K.-H.; van Tol, M. F. H.; Kranenburg,
M. Organometallics 1999, 18, 5447. (c) Duchateau, R.; Abbenhuis,
H. C. L.; van Santen, R. A.; Meetsma, A.; Thiele, S. K.-H.; van Tol,
M. F. H. Organometallics 1998, 17, 5663. (d) Duchateau, R.;
Abbenhuis, H. C. L.; van Santen, R. A.; Thiele, S. K.-H.; van Tol, M.
F. H. Organometallics 1998, 17, 5222. (e) Edelmann, F. T.; Giessmann,
S.; Fischer, A. J. Organomet. Chem. 2001, 620, 80. (f) Edelmann, F.
T.; Giessmann, S. Fischer, A. Chem. Commun. 2000, 2153.
(15) (a) Coles, M. P.; Lugmair, C. G.; Terry, K. W.; Tilley, T. D. Chem.
Mater. 2000, 12, 122. (b) Terry, K. W.; Tilley, T. D. Chem. Mater.
1991, 3, 1001. (c) Jarupatrakorn, J.; Tilley, T. D. J. Am. Chem. Soc.
2002, 124, 8380.
Inorganic Chemistry, Vol. 42, No. 15, 2003 4697

Murugavel et al.
i
Scheme 1. Synthesis and Reactivity of [RSiO3Ti(OPri)]4 (1) and Its Conversion to [RSiO3Ti(OBut)]4 (2) (R ) 2,6-Pr2 C6H3NSiMe3)
titanosilicates without breaking the central core structure
during decomposition at relatively low temperatures (<500
°C). Somewhat similar examples, using preformed titano-
siloxanes as precursors for the preparation of titanium-rich
titanosilicates, have been described recently. For example,
Tilley et al. have demonstrated the applicability of the
mononuclear titanium siloxide complex A in preparing TiO2‚
4SiO2 materials by a thermolysis route and their use in olefin
oxidation reactions.15 Similarly, Fujiwara and Roesky have
shown that catalytically useful titanosilicate materials can
be derived from a combination of cubic titanosiloxanes H
and ∼100-fold tetraethyl orthosilicate through sol-gel routes
in acetic anhydride medium.16 The utility of titanosilsesqui-
oxane G as both homogeneous and heterogeneous oxidation
catalysts has also been investigated.17
tions of cubic titanosiloxane [RSiO3Ti(OPri)]4 (R ) 2,6-
i
Pr2 C6H3NSiMe3) (1) with monosilanol (ButO)3SiOH and
isolated a transesterification product which has been obtained
through the hydrolysis of the monosilanol used. Further we
have also investigated, in detail, the thermal decomposition
of 1 leading to the isolation of titanium-rich titanosilicate
materials. The use of these materials as epoxidation catalysts
has also been probed. The details of these studies are
presented in this paper.
Results and Discussion
Reactivity Studies. To study the reactivity of cubic
titanosiloxanes, we chose [RSiO3Ti(OPri)]4 (R ) 2,6-
i
Pr2 C6H3NSiMe3) (1)13a for the reasons of high yields of the
reaction and the relatively stable nature of the parent
silanetriol.2b This compound also has eight reactive func-
tionalities at the eight corners of the cubic structure. While
it has been well documented that the N-Si group of the
silanetriol cleaves over a period of time in the presence of
either an acidic/basic or highly polar medium,20 the Ti-OR
linkage is also known to undergo substitution as well as
hydrolytic cleavage reactions.13c,16b Considering the reactive
nature of these linkages, it can be expected that any acidic
hydrogen containing compound (alcohols, diols, phenols,
carboxylic acids, etc.) would react with 1 to give a substituted
compound at the titanium center. However these cubic
molecules are highly stable under laboratory conditions and
can be easily handled in air, suggesting that these hydrolyz-
able groups require forcing conditions for modifications at
the corners of the cube.
Although the cubic titanosiloxanes derived from discrete
silanetriols were reported nearly a decade ago by Roesky et
al.,13b there has been no studies on further functionalization
of these cubic structures at titanium and silicon corners.18
In the present investigation, continuing our studies on metal
siloxanes and phosphates,19 we have investigated the reac-
(16) (a) Fujiwara, M.; Wessel, H.; Park, H. S.; Roesky, H. W. Chem. Mater.
2002, 14, 4975. (b) Fujiwara, M.; Wessel, H.; Park, H. S.; Roesky,
H. W. Tetrahedron 2002, 58, 239.
(17) (a) Abbenhuis, H. C. L.; Krijnen, S.; van Santen, R. A. Chem. Commun.
1997, 331. (b) Krijnen, S.; Abbenhuis, H. C. L.; Hanssen, R. W. J.
M.; van Santen, R. A. Angew. Chem., Int. Ed. 1998, 37, 356. (c) Wada,
K.; Nakashita, M.; Bundo, M.; Ito, K.; Kondo, T.; Mitsudo, T. Chem.
Lett. 1998, 659.
(18) While this work was under progress, Jutzi and co-workers have
reported on the reactions of Cp’Si(OH)3-derived cubic titanosiloxanes
with a range of reagents such as acetyl acetone, triphenyl silanol, etc.
(see refs 13c,d).
(19) (a) Sathiyendiran, M.; Murugavel, R. Inorg. Chem. 2002, 41, 6404.
(b) Murugavel, R.; Walawalkar, M. G.; Prabusankar, G.; Davis, P.
Organometallics 2001, 20, 2639. (c) Murugavel, R.; Prabusankar, G.;
Walawalkar, M. G. Inorg. Chem. 2001, 40, 1084. (d) Murugavel, R.;
Sathiyendiran, M. Chem. Lett. 2001, 1, 84. (e) Murugavel, R.;
Sathiyendiran, M.; Walawalkar, M. G. Inorg. Chem. 2001, 40, 427.
(f) Walawalkar, M. G. Organometallics 2003, 22, 879.
On the other hand, the modification of these cubic
structures with more exo-cubic silicon centers would also
aid the usage of these molecules as single-source precursors
(20) Voigt, A.; Walawalkar, M. G.; Murugavel, R.; Roesky, H. W.; Parisini,
E.; Lubini. P. Angew. Chem., Int. Ed. Engl. 1997, 36, 2203.
4698 Inorganic Chemistry, Vol. 42, No. 15, 2003

Molecular Titanosiloxanes
Figure 1. Molecular structure of [RSiO3Ti(OBut)]4 (2).
for the preparation of new titanosilicate materials. Hence,
with this aim, the reaction of 1 with (ButO)3SiOH was carried
out (Scheme 1). Although, it was originally expected that
the -OPri groups on titanium will be exchanged with -OSi-
(OBut)3 groups, the final product obtained from the reaction
mixture, after its crystallization over a period of 3 months,
turned out to be [RSiO3Ti(OBut)]4 (2) instead of expected
[RSiO3Ti(OSi(OBut)3)]4.
Figure 2. Core of molecule [RSiO3Ti(OPri)]4 (1).
Table 1. Selected Bond Lengths (Å) and Bond Angles (deg) for 1a
Ti(1)-O(4)
Ti(1)-O(1)
Ti(1)-O(2)
Ti(1)-O(3)
Si(1)-O(3)#1
1.755(5)
1.807(5)
1.791(5)
1.807(5)
1.619(5)
Si(1)-O(1)#2
Si(1)-O(2)
Si(1)-N(1)
Si(2)-N(1)
1.621(6)
1.618(5)
1.702(6)
1.756(6)
A control experiment was carried out to evaluate the
reactivity of titanosiloxane 1 with (ButO)3SiOH in a petro-
leum ether/toluene mixture for a period of 24 h at room
temperature. At the end of the reaction the solvent was
removed and the resultant residue was examined by elec-
trospray LC-MS. The result obtained indicated that the silanol
remains unreacted at the end of the reaction. Hence, it is
reasonable to assume that formation of 2 proceeds via the
cleavage of Si-OBut group by hydrolysis from (ButO)3SiOH
to form tert-butyl alcohol rather than from a preformed
[RSiO3Ti(OSi(OBut)3)]4. In fact we could not obtain any
spectral evidence for the formation of [RSiO3Ti(OSi-
(OBut)3)]4 during the course of the reaction. We have also
performed an independent synthesis of 2 starting from 1 and
a stoichiometric amount of ButOH. The formation of product
O(4)-Ti(1)-O(1)
O(4)-Ti(1)-O(2)
O(1)-Ti(1)-O(2)
O(4)-Ti(1)-O(3)
O(1)-Ti(1)-O(3)
O(2)-Ti(1)-O(3)
O(3)#1-Si(1)-O(2)
109.3(3)
112.3(3)
108.7(3)
111.4(3)
109.2(2)
106.1(2)
111.4(3)
O(1)#2-Si(1)-O(2)
O(3)#1-Si(1)-N(1)
O(1)#2-Si(1)-N(1)
O(2)-Si(1)-N(1)
Si(1)#3-O(1)-Ti(1)
Si(1)-O(2)-Ti(1)
Si(1)#1-O(3)-Ti(1)
106.7(3)
108.3(3)
113.9(3)
109.7(3)
148.0(3)
154.5(3)
150.3(3)
a Symmetry transformations used to generate equivalent atoms: #1, -y
+ 11/4, x - 1/4, -z + 9/4; #2, -x + 3, -y + 5/2, z; #3, y + 1/4, -x +
11/4, -z + 9/4.
Table 2. Selected Bond Lengths (Å) and Bond Angles (deg) for 2a
Ti(1)-O(4)
Ti(1)-O(1)
Ti(1)-O(2)
Ti(1)-O(3)
Si(1)-O(3)#1
Si(1)-O(1)#2
1.741(8)
1.789(9)
1.797(9)
1.810(9)
1.626(9)
1.627(9)
Si(1)-O(2)
Si(1)-N(1)
Si(2)-N(1)
O(1)-Si(1)#3
O(3)-Si(1)#1
1.627(9)
1.709(10)
1.752(10)
1.627(9)
1.626(9)
1
2 was verified in this case by H NMR spectroscopy.
O(4)-Ti(1)-O(1)
O(4)-Ti(1)-O(2)
O(1)-Ti(1)-O(2)
O(4)-Ti(1)-O(3)
O(1)-Ti(1)-O(3)
O(2)-Ti(1)-O(3)
O(3)#1-Si(1)-O(2)
109.2(4)
O(1)#2-Si(1)-O(2)
O(3)#1-Si(1)-N(1)
O(1)#2-Si(1)-N(1)
O(2)-Si(1)-N(1)
Si(1)#3-O(1)-Ti(1)
Si(1)-O(2)-Ti(1)
Si(1)#1-O(3)-Ti(1)
107.6(5)
107.4(5)
114.2(5)
109.6(5)
150.7(5)
149.2(6)
151.2(6)
Compound 2 was characterized by analytical, spectro-
scopic, and single-crystal X-ray diffraction studies. The
crystals of 2 obtained directly from the reaction mixture were
found to be analytically and spectroscopically pure. The
characteristic infrared stretching frequency for Ti-O-Si
linkages was observed at 960 cm-1. The aromatic and
aliphatic stretching vibrations were observed in the region
111.6(4)
108.7(4)
110.9(4)
109.5(4)
106.8(4)
112.4(5)
a Symmetry transformations used to generate equivalent atoms: #1, -x
+ 3, -y + 5/2, z; #2, -y + 11/4, x - 1/4, -z + 9/4; #3, y + 1/4, -x +
11/4, -z + 9/4.
1
3050-2850 cm-1. The H NMR spectrum showed absence
of the septet corresponding to the -CH group of the -OPri
in 1 (δ 4.27 ppm); instead a new peak corresponding to the
-OBut appeared at (δ 1.6 ppm). The molecular ion peak
observed for 2 in EI-MS spectrum (m/z 1782) confirms the
exchange of the isopropyl group with the tert-butyl group.
isomorphous. A perspective view of the molecular structure
of [RSiO3Ti(OBut)]4 (2) is shown in the Figure 1. The cubic
core structure of 1 is shown in Figure 2. Selected bond
lengths and bond angles found in 1 and 2 are listed in Tables
1 and 2, respectively. A comparison of the selected structural
features of 1 and 2 with those of related titanosiloxanes is
presented in Table 3.
The molecular structures of both 1 and 2 are made up of
a Ti-O-Si cubic framework with alternating titanium and
silicon centers in the core (Figure 1). The Ti4Si4O12
X-ray Structures of 1 and 2. To explore the reactivity
pattern and the effect of substituents on the structure of the
cubic titanosiloxanes, we have carried out the single-crystal
X-ray diffraction studies of compounds 1 and 2. The
compounds crystallize in cubic space group I4h3d and are
Inorganic Chemistry, Vol. 42, No. 15, 2003 4699

Murugavel et al.
Table 3. Comparison of Key Structural Parameters in Cubic Titanosiloxanes Derived from Silanetriolsa
compd
Ti-O, Å
Si-O, Å
Ti-Si, Å
body diagonal, Å
Si-O-Ti, deg
ref
13b
13a
13c
13d
13d
13d
[RSiO3Ti(MeC5H4)]4, R ) But
1.79
1.80
1.78
1.80
1.79
1.77
1.80
1.79
1.61
1.63
1.62
1.63
1.62
1.63
1.62
1.62
1.60
1.61
3.32
3.32
3.33
3.30
3.31
3.33
3.32
3.33
3.09c
3.10c
5.70
5.75
5.74
5.73
5.73
5.72
5.75
5.72
5.38
5.38
145-155
145-155
136-158
137-163
138-165
141-158
148-154
146-154
145-151d
144-151d
i
[RSiO3Ti(OEt)]4, R ) 2,6-Pr2 C6H3NSiMe3
[RSiO3Ti(OBut)]4, R ) Cp*
[RSiO3Ti(OBut)]4, R ) 9-(Me3Si)fluorenyl
[RSiO3Ti(OPri)]4, R ) 9-(Me3Si)fluorenyl
[RSiO3Ti(OBut)]4, R ) But(Me3Si)fluorenyl
[RSiO3Ti(OPri)]4, R ) 2,6-Pr2 C6H3NSiMe3
this work
this work
13e
i
[RSiO3Ti(OBut)]4, R ) 2,6-Pr2 C6H3NSiMe3
i
b
[(MeO)SiO1.5
]
8
b
[PhSiO1.5
]
8
13f
a Average values found in the core are listed. b This entry aids a direct structural comparison between cubic titanosiloxanes and cubic siloxanes. c The
distance of the Si-Si edge. d Si-O-Si angle.
polyhedron (Figures 1 and 2) is completely enclosed in a
hydrophobic sheath of alkyl-substituted aryl and SiMe3
groups, which explains the high solubility of the cubic
structures in hydrocarbon solvents. Each Si-Ti edge of the
cube is bridged by an oxygen atom in a µ2 fashion. There
are six puckered Si2Ti2O4 eight-membered rings that define
the cube. Each of these rings adopts a pseudo-C4 crown
conformation. The four oxygen atoms on any of these six
eight-membered rings are in a plane while the Ti and Si
atoms form an almost parallel plane.
The average Ti-O bond length in the cubic framework
(1.790(5) Å) is longer than the exocyclic Ti-O distance of
1.755(5) Å (Tables 1 and 2). These values are comparable
Figure 3. TGA trace for [RSiO3Ti(OPri)]4 (1).
to the Ti-O distances found in other cubic titanosiloxanes
average length of the Si-Ti edge of the cube is 3.32 Å
(Table 3). The average Si-O distance found in 1 (1.619(9)
whereas the body diagonal of the Si4Ti4 cube is 5.75 Å (Table
Å) is comparable with the corresponding distances found in
3). These distances compare very well with those observed
other cubic titanosiloxanes (Table 3). The Si(1)-N(1)
for other titanosiloxanes. The framework Si-O-Ti angles
distance of 1.702(6) Å is considerably shorter than the Si-
in 1 (148.0(3), 154.5(3), and 150.3(3)°) are also comparable
(2)-N(1) distance of 1.756(6) Å. This observation is
to the values found in other cubic titanosiloxanes. The Ti
consistent with the presence of four electronegative atoms
and Si atoms adopt nearly ideal tetrahedral geometries
attached to Si(1). The observed short Si(1)-N(1) and Si-
(average angles around each of them being 109.5°). The sum
(1)-Ocage distances could also be attributed to the multiple
of the angles around the nitrogen atom in 1 (359.9°)
bonding resulting from negative hyperconjugative interac-
corresponds to a trigonal-planar geometry, which is also an
tions (Nl.pfSi-X σ* and Ol.pfSi-X σ* type interactions
expected requisite for negative hyperconjugative interactions
where X ) O or N).21-23 Such cumulatiVe interactions
leading to short Si-N distances. A comparison of the
leading to short C-O distances in the case of orthocarbonates
observed structural features of cubic titanosiloxanes with
C(OR)4 has been demonstrated using extensive crystal-
those of [(MeO)SiO1.5]813e and [PhSiO1.5]813f reveals that the
lographic investigations as well as ab initio quantum me-
core of the all-silicon cube in these molecules is ap-
chanical calculations.23 Similar effects have also been
proximately 10% smaller than their titanosiloxane counter-
observed for phosphinimines and iminophosphonium ions.21b
parts (Table 3).
The SiO3N coordination environments around the framework
silicon atoms of the cube are very similar to orthocarbonates
and phosphinimes, and hence, the observed Si-N and Si-O
distances are attributable to enhanced cumulative hypercon-
jugative interactions.24
It is of interest to calculate and compare the dimensions
of the central cubic framework with related siloxanes. The
Thermal Decomposition Studies. To find the suitability
of the cubic titanosiloxanes as single-source precursors for
the preparation of titanium-rich titanosilicate materials, the
thermogravimetric (TG) analysis of 1 was carried out. The
initial TG analysis of 1 in air showed an onset temperature
of decomposition at 140 °C which continued till 650 °C in
stages (Figure 3), although no specific assignments could
(21) (a) Murugavel, R.; Krishnamurthy, S. S.; Chandrasekhar, J.; Nethaji,
M. Inorg. Chem. 1993, 32, 5447. (b) Murugavel, R.; Kumaravel, S.
S.; Krishnamurthy, S. S.; Nethaji, M.; Chandrasekhar, J. J. Chem. Soc.,
Dalton Trans. 1994, 847. (c) Murugavel, R.; Palanisami, N.; Butcher,
R. J. J. Organomet. Chem. 2003, 675, 65. (d) Murugavel, R.; Pothiraja,
R. New J. Chem. 2003, 27, 968.
be made to the stepwise loss of the various organic
substituents present on the surface of 1. The ceramic yield
obtained after heating 1 to 650 °C (46.1%) corresponds to
the formation of TiO2‚2SiO2 (theoretical yield 46.3%). This
(22) (a) Lehn, J. M.; Wipff, G. J. J. Chem. Soc., Chem. Commun. 1975,
800. (b) Reed, A. E.; Schleyer, P, v. R. J. Am. Chem. Soc. 1990, 112,
1434.
(23) Narasimhamurthy, N.; Manohar, H.; Samuelson, A. G.; Chandrasekhar,
J. J. Am. Chem. Soc. 1990, 112, 2937.
(24) A manifestation of such cumulative interactions leading to short Si-O
distances can also be seen in the silanol series bearing one, two, and
three Si-OH groups: (PhCH2)3SiOH, 1.70(1) Å; (But)2Si(OH)2, 1.65-
(3) Å; ButSi(OH)3, 1.622(4) Å.
4700 Inorganic Chemistry, Vol. 42, No. 15, 2003

Molecular Titanosiloxanes
observation clearly suggests that, apart from the cubic
framework silicon atoms in 1, the exo-cubic -SiMe3 groups
are also converted into SiO2 during thermal decomposition.
The TGA showed no further weight loss until 1000 °C,
revealing the complete removal of all organic groups in the
temperature range 500-600 °C. Using the data obtained from
TG studies, a bulk decomposition of the cubic titanosiloxane
1 was carried out at 450 (1a), 600 (1b), 800 (1c), 900 (1d),
1000 (1e), and 1200 °C (1f). The initial thermal decomposi-
tion at 450 °C produced a beige-colored material, whose
elemental analysis showed the presence of carbon (0.5%)
and hydrogen (1.3%) impurities. However, the same sample
on calcination in air at 600 °C produced a white powder.
The elemental analysis of this material showed no residual
carbon. The presence of small quantities of hydrogen (0.7%)
in this sample is attributed to the adsorption of water
molecules on the titanosilicate material after calcination
(which is also confirmed by independent IR studies; vide
infra). There was no change in color or overall composition
of the material on further calcination at temperatures up to
1200 °C.
At this point, it was necessary to ascertain the exact
composition of the material and also verify the material yield
obtained by TGA corresponding to the composition TiO2‚
2SiO2. Hence, the titanium and silicon content in the material
were determined using the acid digestion method recently
reported by Gunji et al. (see Experimental Section).25 The
results obtained by this procedure revealed the titanium-to-
silicon ratio to be 1:2.
The calcined samples at various temperatures (1a-f) were
further investigated by powder X-ray diffraction (Figure 4)
and infrared spectroscopic studies (Figure 5). While PXRD
was very useful in assessing the crystalline nature of the
material and the identification of the phases, IR spectroscopy
has provided useful information about the presence or
absence of relevant linkages in the material.26 For example,
the strong absorption in the IR spectrum observed around
960 cm-1 in titanosiloxane molecules as well as titanosilicate
materials has been attributed to the presence of a Ti-O-Si
linkage in the system.13a,26 In fact, the intensity of this
absorption is often related to the extent of titanium incor-
poration into the framework of a titanosilicate material with
Ti-O-Si linkages,26b,c especially in the case of titanosilicate
material with higher titanium content (e.g. >5%). Since, our
TGA and solid-state decomposition studies point to the
formation of TiO2‚2SiO2 material, it is of interest to study
the intensity of 960 cm-1 absorption. The IR spectra of all
the calcined samples, shown in Figure 5, clearly contain this
absorption. However, it is instructive to note that the intensity
of this absorption decreases with increase in the calcination
temperature indicating a probable phase separation of TiO2
from the titanosilicate materials at higher temperature. Thus,
Figure 4. Powder XRD patterns of 1a-f. The peaks due to anatase-TiO2
are marked $ while the peaks due to rutile form are labeled #. The un-
indexed peaks found in 1f are due to a crystalline titanosilicate material.
it can be concluded that when the decomposition is carried
out at relatively lower temperature, the presence of Ti-O-
Si linkages in the ceramic material is observed, and on
increasing the temperature above 800 °C, the Ti-O-Si
linkages break down to form TiO2 phases. This observation
also suggests that the compound 1 does not retain its central
cubic core in the titanosilicate materials, at least those
produced at high temperatures.
The presence of a strong broad absorption ca. 3400 cm-1
in samples 1a-f is attributed to the adsorption of water
molecules on the titanosilicate material during sample
preparation for spectral studies. The powder X-ray diffraction
of 1a (Figure 4) does not show any peaks revealing the
complete amorphous nature of the material. Subsequent
heating of the sample at 600 °C (1b) also failed to introduce
any kind of crystallinity to the material. The material heated
at 800 °C for 4 h (1c) showed four broad peaks and a broad
band overlapping with the highest peak. This could be
attributed to the crystallization of titanosilicate material (vide
infra). To study the crystallization behavior in detail, the
(25) Gunji, T.; Kubota, K.; Kishiki, S.; Abe. Y. Bull. Chem. Soc. Jpn. 2002,
75, 537.
(26) (a) Uguina, M. A.; Ovejero, G.; van Grieken, R.; Serrano, D. P.;
Camacho, M. J. Chem. Soc., Chem. Commun. 1994, 27. (b) Hutter,
A.; Dutoit, D. C. M.; Mallat, T.; Schneider, M.; Baiker, A. J. Chem.
Soc., Chem. Commun. 1995, 163. (c) Dutoit, D. C. M.; Mallat, T.;
Baiker, A. J. Catal. 1995, 153, 177.
Inorganic Chemistry, Vol. 42, No. 15, 2003 4701

Murugavel et al.
Figure 5. Infrared spectrum of [RSiO3Ti(OPri)]4 (1) and the calcined samples (in KBr). The peaks at 960 cm-1 due to the Ti-O-Si vibration are shown
with arrows.
sample was subsequently heated at 900 (1d), 1000 (1e), and
1200 °C (1f). The powder XRD pattern of both 1d,e, show
the sharpening of the broad peak (2θ ) 25.4°) and also the
emergence of four new distinct peaks, which can be indexed
to the anatase phase of TiO2 (2θ ) 25.3 and 37.8°). However,
the sample 1f is fully crystalline, gave five distinct major
peaks in powder XRD analysis. The peaks observed at 2θ
) 25.5, 37.9, 48.2, and 55.3° can be indexed to the anatase
phase,27a whereas the peaks at 2θ ) 27.7, 36.3, 39.4, 41.5,
44.3, 54.0, 54.5, and 56.8° correspond to the rutile phase of
TiO2.27b This is clearly indicative of the phase separation
occurring in the sample due to the breakdown of Ti-O-Si
linkages of 1 during the calcination process. It appears that
the unindexed lines in the XRD pattern of the sample 1f,
shown in Figure 4, correspond to a titanosilicate phase, which
also accounts for the residual Ti-O-Si absorption at 960
cm-1 in its IR spectrum (Figure 5).
UV-vis spectroscopy has also been used to characterize
the materials 1a-f. The electronic absorption observed in
(27) (a) Rausch, N.; Burte, E. P. J. Electrochem. Soc. 1993, 140, 145. (b)
Ritala, M.; Leskala, M.; Nykanen, E.; Soininen, P.; Niinisto, L. Thin
Solid Films 1993, 225, 288.
4702 Inorganic Chemistry, Vol. 42, No. 15, 2003

Molecular Titanosiloxanes
Figure 6. Top: TEM bright-field micrographs of the samples 1c (A), 1d (B), 1e (C) and 1f (D). Bottom: Corresponding electron diffraction pattern of a
chosen particle from the respective samples.
titanosilicates could be due to the charge transfer between
the oxide ions of the lattice and an empty d-orbital of the
Ti(IV).26,28 The position of the absorption maximum of
tetrahedral and octahedral Ti(IV) is normally expected at
45 000-48 000 and 42 000 cm-1, respectively. The absor-
bance maxima for the titanium containing materials 1a-f
were broadened and shifted to lower wavelengths (33 100
cm-1). This is consistent with the presence of octahedral Ti
sites. In fact, the presence of octahedral sites within the silica
matrix is expected to be from higher titanium content (>ca.
8 mol % as described in the literature).26,28
6; sample C) is from a highly crystalline phase. The TEM
images of sample 1f show a ring as well as a spot pattern at
different regions (Figure 6; sample D). This reveals the
presence of amorphous, polycrystalline, and single-crystalline
titanosilicates in the same material. The energy-dispersive
X-ray analyses (EDAX) of the samples 1a-f show the
presence of Ti and Si in an approximate ratio of 1:2, further
confirming our results obtained for titanium content through
the acid digestion method (vide supra). Further work is
necessary to clearly understand the exact microstructure of
these materials and the surface characteristics such as surface
area and porosity.
The surface and morphology of the materials 1a-f were
examined using scanning electron microscopy (SEM). The
SEM micrograph of sample 1a depicts irregularity in the
crystalline nature of the surface. As the calcination temper-
ature is increased, the particles become more crystalline. For
example, sample 1e shows the maximum amount of regular-
ity of the surface. The microstructures of the materials 1c-f
were further studied using transmission electron microscopy
(TEM) (Figure 6). The particle size of sample 1c varies from
35 to 80 nm in diameter. The ring pattern obtained from
TEM electron diffraction study for sample 1c reveals the
polycrystalline nature of the material. The material 1d also
shows a similar behavior (Figure 6; sample B). Sample 1e,
which is heated at 1000 °C, is the most crystalline product,
and its particle size ranges from 85 to 115 nm. The
microdiffraction of the material shows both ring pattern as
well as spot pattern at various regions. Ring pattern is due
to the presence of polycrystalline material or a thick sample
containing single crystals, whereas the spot pattern (Figure
Cyclohexene Epoxidation. Titania-silica mixed oxides
and titanosilicates such as TS-1 and TS-2 are effective
catalysts for epoxidation and selective oxidation reactions
using peroxides as oxidants.29 The active sites are assumed
to be the isolated tetrahedral Ti4+ centers.29c The presence
of polar solvents greatly retards the reaction by competing
for the active coordination sites.30 Highly dispersed titania-
silica is very effective for olefin epoxidation with alkyl
hydroperoxides. The formation of Ti-O-Si bonds is very
crucial as demonstrated by the much lower activities of TiO2
supported on other oxides and the physical mixtures of TiO2
and SiO2.28,29b,c,30
(29) (a) Reddy, J. S.; Kumar, R.; Ratnasamy, P. App. Catal. 1990, 58, L1.
(b) Maschmeyer, T.; Klunduk, M. C.; Martin, C. M.; Shephard, D.
S.; Thomas, J. M.; Johnson, B. F. G. Chem. Commun. 1997, 1847.
(c) Sheldon, R. A.; van Doorn, J. A. J. Catal. 1973, 31, 427. (d) Notari,
B. Catal. Today 1994, 18, 163.
(30) (a) Gao, X.; Wachs, I. E. Catal. Today 1999, 51, 233. (b) Blasco, T.;
Corma, A.; Navaroo, M. T.; Pere´z-Pariente, J. J. Catal. 1995, 156,
65. (c) Thomas, J. M. Chem.sEur. J. 1997, 3, 1557. (d) Beghi, M.;
Chiurlo, P.; Costa, L.; Palladino, M.; Pirini, M. F. J. Non-Cryst. Solids
1992, 145, 175. (e) Klein, S.; Weckhuysen, B. M.; Martens, J. A.;
Maier, W. F.; Jacobs, P. A. J. Catal. 1996, 163, 489.
(28) (a) Shraml-Marth, M.; Walther, K. L.; Wokaun, A.; Handy, B. E.;
Baiker, A. J. Non-Cryst. Solids 1992, 143, 93.
Inorganic Chemistry, Vol. 42, No. 15, 2003 4703

Murugavel et al.
Table 4. Conversion of Cyclohexene to Cyclohexene Oxide
titanosilicates (irrespective of the temperature used for the
decomposition). Alternatively, the low conversion could also
be attributed to rapid catalytic peroxide decomposition.
Hence, it is necessary to investigate the catalysis aspects of
1a-f further in detail and also understand the surface
characteristics of these materials before any meaningful
comparison could be made with the results obtained for other
epoxidation catalysts based on titanosilicates. Hence, further
experiments are necessary to optimize olefin epoxidation
reactions in terms of the choice of substrate, choice of the
oxidant, reaction temperature, and reaction time. Work in
this direction is currently underway.
no.
catal
% conversion
1
2
3
4
5
1a
15
11
11
1
1b
1c
1d
1e
1
a Used as a homogeneous catalyst.
Owing to the presence of a number Ti-O-Si linkages
with tetrahedral titanium, the catalytic performances of the
molecular precursor, [RSiO3Ti(OPri)]4 (1), and titania-silica
materials (1b-e) were studied toward the epoxidation of
cyclohexene. Compound 1a was not used in the catalytic
studies due to its residual carbon content while the catalytic
activity of 1f was not investigated because of the larger TiO2
phase separation in this sample. All epoxidation reactions
were carried out in toluene at 65 °C with cumenyl hydro-
peroxide as oxidant. The conversion of cyclohexene to
cyclohexene oxide was followed by GC-MS. To compare
the activities, the results were standardized with respect to
the mass of the catalyst used. In the control experiments
carried out with 1 in the absence of hydroperoxide, a
negligible amount of cyclohexene oxide product was ob-
served by GC analysis after 23 h. Similarly, the heteroge-
neous catalysts 1a-f also in the absence of hydroperoxide
did not show any appreciable activity.31
The molecular precursor 1 was found to be an active
homogeneous catalyst for the epoxidation of cyclohexene
(15% conversion) after 23 h (Table 4). Epoxidation using
heterogeneous titania-silica materials 1b-e was also carried
out under similar conditions. It is found that the samples
1b,c show ∼11% conversion. Cyclohexene oxide was the
sole product in all these cases, thus showing a very high
selectivity in the epoxidation process. However, the samples
1d,e gave a conversion of only <1%. The drastic decrease
in catalytic performance in the latter case is attributed to the
phase separation of TiO2 occurring in the material at elevated
temperatures. This observation is also consistent with the
reduction in the intensity of Ti-O-Si vibration at 960 cm-1
(Figure 5) due to the formation of significant amount of TiO2.
Powder XRD and DR UV measurements also support this
TiO2 phase separation hypothesis for the reduced catalytic
activity of 1d,e. At low temperatures of calcination, the
tetrahedral titanium center predominates whereas at higher
temperatures of calcination octahedral titanium centers were
formed.
Conclusion
It has been shown in this contribution that the modification
of the periphery of cubic titanosiloxanes occurs only at the
titanium center over a period of time although the original
objective of increasing the silicon content on the cubic
titanosiloxane was unsuccessful. The compound [RSiO3Ti-
(OPri)]4 undergoes in situ transesterification to form isomor-
phous molecule [RSiO3Ti(OBut)]4. We have shown that it
is possible to prepare titanium-rich titanosilicate materials
from soluble titanosiloxane molecules. The solid state
thermolysis at relatively lower temperatures (<800 °C) leads
to the formation of Ti-O-Si materials, which are shown to
be potential catalysts for olefin epoxidation. However, the
calcination of the titania-silica material at high temperatures
(>800 °C) produced a mixture of titanosilicates along with
phase-separated TiO2, which shows lesser activity toward
olefin epoxidation. Further work in our laboratory is centered
on further functionalization of cubic titanosiloxanes as well
as optimization of olefin epoxidation reactions using 1a-f.
Experimental Section
Material and Methods. All manipulations were performed under
an atmosphere of nitrogen using standard Schlenk techniques.
Solvents used were freshly distilled prior to use. The silanetriol,
RSi(OH)3 (R ) 2,6-Pr2 C6H3NSiMe3),2b and silanol, (ButO)3SiOH,32
i
have been synthesized by following the reported procedures. Ti-
(OPri)4 obtained from E-Merck was used as received. Cumenyl
hydroperoxde was procured from Lancaster (80% tech).
Chemical analysis has been carried out at the Analytical
Laboratories of the Department of Chemistry, IIT-Bombay, Bom-
bay, on a Carlo Erba model 1106 instrument. 1H NMR spectra were
recorded on a Varian VXR 400S spectrometer. Infrared spectrum
of the samples were obtained on a Nicolet Impact 400 spectrometer.
The DR UV spectra of the calcined samples were recorded on a
Shimadzu UV-260 UV-vis spectrophotometer in the reflectance
mode for the well-ground solid samples. The thermal analysis of
the cubic titanosiloxane 1 has been carried out in air with the heating
rate of 10 °C min-1 on a DuPont model 2100 thermal analyzer .
The powder X-ray diffraction studies were carried out on a Philips
PW1729 X-ray diffractometer using Cu KR radiation. SEM work
was done on a JEOL JSM-840A. Samples were prepared by
ultrasonication in acetone and then dispersed on a copper stub. TEM
micrographs were obtained on a Philips CM 200 microscope. As
the samples show crystalline behavior only above 800 °C, the TEM
measurements were carried out only for the samples 1c-f. Progress
The results obtained from the above-described reactions
are in fact difficult to understand since the increased titanium
content in 1a-f did not lead to a better conversion of
cyclohexene to its epoxide. Possible reasons for the decreased
catalytic efficiency could be either due to the presence of
too many tetrahedral titanium centers in close proximity
(crowding) or the coexistence of small amounts of TiO2
phase with octahedral titanium centers along with the
(31) Another control experiment carried out for the oxidation of cyclohexene
in the presence of oxidant alone also produced no appreciable amounts
of cyclohexene oxide.
(32) Abe, Y.; Kijima, I. Bull. Chem. Soc. Jpn. 1969, 42, 1118.
4704 Inorganic Chemistry, Vol. 42, No. 15, 2003

Molecular Titanosiloxanes
Table 5. Crystal Data and Structure Refinement for 1 and 2
1
2
empirical formula
fw
C72H132N4O16Si8Ti4
1726
C76H140N4O16Si8Ti4
1782
temp, K
293(2)
200(2)
wavelength, Å
0.710 73
0.710 73
cryst system, space group
unit cell dimens
V, Å3
cubic, I4h3d
cubic, I4h3d
a ) 31.702
a ) 31.606
31 860.7
31 572.5
Z
12
1.080
0.430
11 040
12
1.125
0.436
11 424
calcd density, Mg/m3
abs coeff, mm-1
F(000)
θ range for data collcn, deg
1.57-24.99
3.53-25.16
index ranges
0 e h e 37, 0 e k ) 37, 0 e l e 37
14 237/2482 [R(int) ) 0.0698]
99.9
0 e h e 37, 0 e k ) 26, 0 e l e 37
4897/2478 [R(int) ) 0.1292]
98.6
reflcns collcd/unique
completeness to 2θ, %
refinement method
data/restraints/params
goodness-of- fit on F2
final R indices [I > 2σ(I)]
R indices (all data)
largest diff peak and hole, e Å-3
full-matrix LS on F2
2482/0/245
full-matrix LS on F2
2478/0/255
1.067
1.083
R1 ) 0.0520, wR2 ) 0.1353
R1 ) 0.0919, wR2 ) 0.1622
0.425 and -0.238
R1 ) 0.0879, wR2 ) 0.1753
R1 ) 0.1525, wR2 ) 0.2307
0.693 and -0.720
of the epoxidation reactions was monitored on a Perkin-Elmer
Clarus GC-MS system. The metal content in the sample was
determined by following the procedure of Gunji et al.25
of fume ceases), the residue was weighed (A). Independently, 0.2
g of the sample was weighed in a flask and heated in the presence
of ammonium sulfate (1 g), ammonium nitrate (1 g), and sulfuric
acid (20 mL) for 48 h. The residue was poured into 100 g of ice,
followed by filtration. The residue was subjected to firing in a
crucible, followed by weighing (B). The content of SiO2 was
calculated from the content (B), and the content of TiO2 was
calculated from the difference between the contents (A and B).
Epoxidation of Cyclohexene. Appropriate catalyst (50 mg),
toluene (5 mL), and cyclohexene (2.5 mL, 24.7 mmol) were taken
in a round-bottom flask. Reaction mixture was heated in an oil
bath and allowed to equilibrate at 65 °C for 10 min. Cumenyl
hydroperoxide (1 mL) was then added with stirring. The mixture
was stirred for 23 h and allowed to attain room temperature. It
was then centrifuged to remove the suspended particles of the
catalyst. The samples were analyzed using GC-MS. A similar
procedure was adopted for all other samples.
Synthesis of [RSiO3Ti(OPri)4] (R ) 2,6-Pri2C6H3NSiMe3)
(1).13a RSi(OH)3 (1 g, 3.3 mmol) was suspended in petroleum ether
(30 mL), and Ti(OPri)4 (1 mL, 3.3 mmol) was added slowly with
stirring. The solution became clear on stirring for 4 h after which
solvent was removed in vacuo, and the residue obtained was dried
for 12 h and redissolved in minimum amount of petroleum ether.
X-ray-quality crystals were grown at room temperature. The product
was separated by decantation. Yield: 82% (1.02 g). Anal. Calcd
C72H132N4O16Si8Ti4: C, 50.1; H, 7.7; N, 3.3. Found: C, 50.2; H,
8.4; N, 3.3. IR (KBr): 1321, 1248, 1183, 1097, 1064, 966, 913,
1
841, 802 cm-1. H NMR (300 MHz, CDCl3): δ 0.09 (s, 36H,
SiMe3), 0.96 (d, 24H, Me2), 1.16 (d, 24H, Me2), 1.24 (d, 24H, Me2),
3.52 (septet, 8H, CH), 4.27 (septet, 4H, OCH), 6.99 (s, 12H, Ph)
ppm.
Synthesis of [RSiO3Ti(OBut)4] (2). [RSiO3Ti(OPri)]4 (1) (1 g,
0.58 mmol) was dissolved in a 1:1 mixture of toluene/petroleum
ether (20 mL), and (ButO)3SiOH (0.448 g, 2.32 mmol) in petroleum
ether (10 mL) was slowly added to the above solution. The reaction
mixture was stirred for 24 h. The solvent was concentrated to half
the volume. Colorless X-ray-quality crystals were grown at room
temperature over a period of 90 days. Yield: 40% (332 mg based
on [RSiO3Ti(OPri)]4). Anal. Calcd For C76H140N4O16Si8Ti4; C,
51.22; H, 7.91; N, 3.14. Found: C, 51.20; H, 7.85; N, 2.49. IR
(KBr): 1443, 1255, 1083, 974, 909, 853, 802 cm-1. 1H NMR (300
MHz, C6D6): δ 0.05 (s, 36H, SiMe3), 0.2 (d, 36H, Me2), 1.16 (d,
24H, Me2), 1.6 (d, 24H, Me2), 3.52 (sep. 8H, CH), 6.99 (s, 12H,
Ph) ppm. Mass spectrum (EI): m/e 1782 (M+).
Thermolysis of [RSiO3Ti(OPri)]4 (1a-f). A sample of [RSiO3-
Ti(OPri)]4 (1) (1 g) was weighed in a platinum crucible and heated
in air at 450 °C at a rate of 10 °C min-1 for 4 h. The sample was
allowed to cool to room temperature affording (0.45 g, 45%) a
beige-colored solid. The material was subsequently calcined at
various temperatures to get the different materials (600 °C (1b),
800 °C (1c), 900 °C (1d), 1000 °C (1e), and 1200 °C (1f)).
Estimation of Ti and Si. The titanium and silicon contents were
determined by following a reported literature procedure.25 For the
estimation of TiO2 and SiO2, 0.2 g of the respective sample was
weighed in a crucible and sulfuric acid (0.5 mL) was added. After
gradual heating in a Kjeldahl flask overnight (until the evolution
X-ray Structure Determination of 1 and 2. Single crystals of
1 and 2 suitable for X-ray structure analysis were grown using the
procedure described vide supra. Intensity data for [RSiO3Ti(OPri)]4
(1) were collected on Nonius MACH-3 diffractometer. A Siemens-
STOE AED2 four-circle diffractometer using Mo KR radiation at
room temperature was used for 2. Intensity data collection and cell
determination for both the compounds were carried out using a
monochromatized Mo KR radiation in the both diffractometers. Data
were suitably processed, and the space group determination was
carried out by examining the systematic absences. The structure
solution in each case was achieved using SHELXS-97.33 The
refinement of the structure was carried out SHELXL-97.33,34 The
positions of hydrogen atoms were geometrically fixed and refined
using a riding model. All non-hydrogen atoms were refined
anisotropically. Other details pertaining to data collection, structure
solution, and refinement are given in Table 5. Atomic coordinates,
complete bond distances and bond angles, and anisotropic thermal
parameters of all non-hydrogen atoms are deposited as Supporting
Information in the form of CIF files.
Acknowledgment. We acknowledge generous financial
support by the DST, New Delhi, for carrying out this work
(33) Sheldrick, G. M. SHELXS-97: Program for Crystal Structure Solution;
University of Go¨ttingen: Go¨ttingen, Germany, 1997.
(34) Sheldrick, G. M. SHELXL-97: Program for Crystal Structure Refine-
ment; University of Go¨ttingen: Go¨ttingen, Germany, 1997.
Inorganic Chemistry, Vol. 42, No. 15, 2003 4705

Murugavel et al.
in the form of a Swarnajayanti Fellowship Project to R.M.
This work was also supported in part by a DAE Young
Scientist Research Award to R.M. P.D. thanks the CSIR,
New Delhi, for a senior research fellowship. The authors
thank the DST-funded Single Crystal X-ray Diffractometer
Facility at IIT-Bombay for the X-ray diffraction data for
compound 1 and the RSIC, IIT-Bombay, for the TGA and
TEM measurements. We thank the Department of Metal-
lurgical Engineering and Materials Science, IIT Bombay, for
powder XRD and the Department of Earth Sciences, IIT
Bombay, for SEM studies. The authors thank the reviewers
of this manuscript for their comments, which were useful in
revising the paper. R.M. thanks Prof. H. W. Roesky for
introducing him to silanetriol chemistry and constant en-
couragement.
Supporting Information Available: X-ray data (including
atomic coordinates, complete bond lengths and angles, anisotropic
thermal parameters, and hydrogen atom coordinates) in CIF format
for compounds 1 and 2. This material is available free of charge
via the Internet at http://pubs.acs.org.
IC034317M
4706 Inorganic Chemistry, Vol. 42, No. 15, 2003