Rational synthesis of asymmetric bicyclic siloxane

Article abstract of DOI:10.1039/b500591d

A rational and versatile method to synthesize bicyclosiloxane of design structures is presented. The method is used to synthesize a new, asymmetric bicyclo[7.5.3]octasiloxane and other bicyclosiloxanes. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b500591d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Rational synthesis of asymmetric bicyclic siloxane{
Wenmei Xue, Mayfair C. Kung and Harold H. Kung*
Received (in Berkeley, CA, USA) 12th January 2005, Accepted 11th February 2005
First published as an Advance Article on the web 7th March 2005
DOI: 10.1039/b500591d
In order to form asymmetric structures, siloxanes containing
A rational and versatile method to synthesize bicyclosiloxane
of design structures is presented. The method is used to
synthesize a new, asymmetric bicyclo[7.5.3]octasiloxane and
other bicyclosiloxanes.
different functional groups of different reactivities need to be used.
This was one of the synthesis strategies investigated by Masamune
and coworkers,17 employing hydrochlorosilane (e.g. a-hydro-
v-chlorooligosiloxane) and hydrosilanol for stepwise linear homo-
logation and branched elongation and for building hyperbranched
polysiloxanes and silicone dendrimers. The building blocks with
different functional groups were synthesized by partial functiona-
lization of a precursor, and consequently with low yield. They did
not extend their investigation to asymmetric cyclic compounds.
Expanding on our earlier work in the synthesis of siloxanes,18
we report here a strategy that exerts control of the reactions by
exploiting the differences in reactivity of different functional
groups. Selective reaction of one functional group of a bifunctional
silane with a different silane would result in quantitative formation
of a desired intermediate without byproducts. The remaining
functional group then can be used in a different step of the
synthetic process as desired. This concept is illustrated in Scheme 1
for the synthesis of an asymmetric bicyclosiloxane. The process is
initiated by reacting chlorodimethylsilane with diphenylsilanediol
(reaction (iii)). The different reactivities of the hydride and chloride
substituents permit selective reaction of chloride with a silanol and
ensure termination of the reaction at that stage,17 thereby
eliminating the formation of a wide range of byproducts due to
simultaneous reaction of both substituents as in the case when
dichlorosilane is used. Using excess diphenylsilane ensures that
Siloxanes are building blocks for organosilicon hybrid materials
and silicone rubber that have found applications in electronic1 and
medical devices.2 Their properties depend on the nature of the
organic ligands, length of the siloxane chain, and positions and
nature of functional groups. However, in spite of the rather large
number of siloxane compounds available, there are relatively few
known siloxane bicyclic compounds. Furthermore, the procedures
employed to synthesize the few known bicyclic compounds exert
little control over the size or make-up of these structural units.
Controlled synthesis of such compounds would enable their
properties to be studied. Here we report a method that permits
synthesis of bicyclosiloxanes where the three siloxane chains can be
of different but controlled lengths.
Ladders and cages of silsesquioxane have been synthesized.
Their preparation typically involves reaction of siloxanes with
multiple but identical functional groups. Thus, when the precursor
siloxanes hydrolytically condense with other siloxanes to form
these structures, the products are always symmetric.3–5 For the
synthesis of highly symmetric bicyclosiloxanes, the reported
methods offer little control of the reactions and/or result in very
low yields because of multiple products and extensive purification
steps.5–11 This is illustrated with reactions (i) and (ii), which involve
reaction of organotrichlorosilane with an organosilanediol to form
a siloxane ring with chloro groups at the potential bridgehead
positions (reaction (i)).12–14 Then, the third bridge was added by
reaction of a silanediol with the bridgehead chloro groups
(reaction (ii)). In addition to the large number of products formed,
the hygroscopic nature of the halosilanes causes additional
difficulties in handling.15,16
Scheme 1 (iii) py, yield 86%; (iv) & (vi) H2O, Pd(OH)2/C, yield 93% for 2
and 98% for 4: (v) Me2SiHCl, py, yield 88%; (vii) PhMeSiCl2, py, yield
74%; (viii) Br2, a mixture of cis- and trans-isomers was obtained; (ix) 10,
py, yield 19% from 5.
{ Electronic supplementary information (ESI) available: Preparatory
procedures and NMR data. See http://www.rsc.org/suppdata/cc/b5/
b500591d/
*hkung@northwestern.edu
2164 | Chem. Commun., 2005, 2164–2166
This journal is ß The Royal Society of Chemistry 2005

View Article Online
only one of the hydroxyl groups in diphenylsilanediol was
condensed. A small amount of bis-condensation side-product (less
than 5%) could be separated by column chromatography.
The resulting organo-H-silanol (e.g. HSiMe2OSiPh2OH 1) can
be oxidized to a linear silanediol (e.g. 1,3-dihydroxy-1,1-dimethyl-
3,3-diphenylsiloxane 2) using Pearlman’s catalyst as described in
the literature (reaction (iv)).19 Since the Si atom containing the
phenyl ligands will serve as a bridgehead, the two bridges that
emanate from it will differ in length by one Si atom in this
example. It can be seen that bridges that differ by other numbers
of Si atoms can be formed readily by using different chlorosilanes
in reaction (iii). For example, using HSi(CH3)2OSi(CH3)2Cl would
result in two bridges that differ by two Si atoms.
was converted to1,5-dibromo-1-phenyl-3,3,5,7,7,9,9-heptamethyl-
cyclopentasiloxane 6 (reaction (viii)). The reaction was quantita-
tive, and only one phenyl group in the Ph2SiO2 unit was converted,
because the electron density of the silicon atom was significantly
decreased by the electron-withdrawing effect of the bromine
atom.21 A mixture of cis and trans isomers was produced, and
could be used directly for subsequent steps without purification,
except for pumping off the excess bromine, although only the cis
isomer will produce the bicyclic compound. 6 is an interesting
intermediate as it contains two reactive groups. Thus, it can be
used for other purposes as well, such as a building unit to form
metal organic frameworks.
A bicyclosiloxane can be synthesized by reacting a dibromocy-
closiloxane, such as 6, with a silanediol to form the third bridge.
The size of the silanediol determines the length of this bridge. For
example, 1,5,5-triphenyl-3,3,7,7,9,11,11,13,13,16,16-undecamethyl-
bicyclo[7.5.3]octasiloxane 7, which contains three bridges of
different lengths, was synthesized by reacting 6 with 1,5-
dihydroxy-1,1,5,5-tetramethyl-3,3-diphenylsiloxane 10 (reaction
(ix)). After purification by evacuation and chromatography, 7
was recovered as a colorless, viscous liquid with a yield of 19%
starting from 5. The major factor that lowers the yield is the
formation of the trans isomer of 6. The purity of 7 was
confirmed by NMR, mass spectrometry, and elemental analysis
(see ESI for details).{ Although the compound contains two
optical centers, because of the constraint of the bridges, only two
of the four stereoisomers are expected. The Si atoms in all five
Me2SiO2 groups yield resonances at different chemical shifts:
219.49, 219.83, 220.49, 220.56 and 221.32 ppm. The Si in
Ph2SiO2 appears at 248.33, in MeSiO3 at 266.65, and in PhSiO3
at 280.52 ppm.
Other siloxane units can be added to compound 2 to lengthen
the branches. For example, one cycle of reaction of R2SiHCl with
2 followed by catalytic oxidation will add one Si unit to each of the
two branches (reactions (v) and (vi)), and multiple cycles can be
used. The reactions can be followed readily with 1H, 13C, and 29Si
NMR, and the products of these reactions are air-stable.
Therefore, work-up is easy and the products could be obtained
with over 90% yield.
We have demonstrated the reactions shown in Scheme 1 to
prepare two branches that differ by one Si atom, forming
1,7-dihydroxy-1,1,5,5,7,7-hexamethyl-3,3-diphenylsiloxane 4. This
compound in the pure form can be stored for several months at
218 uC, although it is sensitive towards self-condensation in
solution. The 29Si resonance of the SiMe2OH end groups appear at
29.10 and 210.28 ppm, and the Si in the SiPh2O2 and SiMe2O2
groups show their typical resonances around 247.67 and
218.90 ppm, respectively. The organo-H-siloxanes 1 and 3 are
both liquid at room temperature. The characteristic signal of their
1H NMR is a quadruple peak at 4.5–5.0 ppm, assigned to Me2SiH.
The terminal silicon resonances are at 23.19 for 1, and 24.08 and
26.26 ppm for 3.
The formation of cyclic 1,1,5-triphenyl-3,3,5,7,7,9,9-hepta-
methylcyclopentasiloxane 5 was achieved by condensation of 4
with dichloromethylphenylsilane in the presence of pyridine as an
HCl acceptor (reaction (vii)). By mixing dilute solutions of the
reactants slowly so as to minimize the formation of
undesired, higher molecular weight products, a 74% yield of a
racemic mixture of 5 was achieved. The mass spectrum of 5
showed the siliconium ion peak at m/z 5 541, which was due to
loss of a methyl group, a well-known behavior of organosiloxane
compounds.20
Two other bicyclosiloxanes, 1,3,3-triphenyl-5,7,7,10,10-penta-
methylbicyclo[3.3.3]pentasiloxane
8,
and
1,5,5-triphenyl-
9,
3,3,7,7,9,11,11,14,14-nonamethylbicyclo[7.3.3]heptasiloxane
Unlike previous preparations, the cyclic compound 5 is stable in
air and contains no hydroxyl or chloride functional groups. Thus,
it can be purified and handled easily. The asymmetric structure
were synthesized to illustrate the versatility of this method
(Scheme 2). In these two compounds, two of the siloxane bridges
are identical, but the third is different. The reaction of 1,5-
dihydroxy-1,1,5,5-tetramethyl-3,3-diphenylsiloxane 1019 with
dichloromethylphenylsilane results in the symmetric cyclosiloxane
1,1,5-triphenyl-3,3,5,7,7-pentamethylcyclotetrasiloxane 11 with
92% yield (reaction (x)). The phenyl groups in the potential
bridgehead positions were functionalized with bromine to form a
mixture of cis and trans 1,5-dibromo-1-phenyl-3,3,5,7,7-penta-
methylcyclotetrasiloxane 12 (reaction (xi)). 1H NMR showed a cis
to trans ratio of 2 : 3. The mixture was used to form 8 (reaction
(xii)), but only the cis isomer was expected to be able to condense
with diphenylsilanediol to form the desired product. Indeed, 8 was
obtained quantitatively from the cis isomer, whereas the trans
isomer reacted with silanediol to produce a mixture of high
1
makes all six methyl groups of Me2SiO2 different, and their H
resonances were found at 0.18, 0.14, 0.11, 0.08, 0.07 and 0.01 ppm.
The 29Si NMR chemical shift of the Me2SiO2 moiety was in the
range 218.73 to 219.92 ppm, the Ph2SiO2 moiety at 247.85, and
the MePhSiO2 group at 234.26 ppm. The detailed preparation
procedure, the yields, and the NMR spectra of these compounds
and others described later are provided in the electronic
supplementary information (ESI).{
The phenyl groups at the potential bridgeheads in the
cyclosiloxanes formed in this manner can be converted to bromide
groups by treatment with bromine at room temperature, thereby
functionalizing the Si atoms to anchor the third bridge.19 Thus, 5
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 2164–2166 | 2165

View Article Online
2 See e.g.: (a) H. S. El-Zaim and J. P. Heggers, Polym. Biomater., ed.
S. Dumitriu, Marcel Dekker, Inc., New York, 2nd edn., 2002, p. 79; (b)
R. Yoda, J. Biomater. Sci., Polym. Ed., 1998, 9, 561; (c) C. R. McMillin,
Elastomerics, 1988, 120, 22.
3 M. Unno, A. Suto, K. Takada and H. Matsumoto, Bull. Chem. Soc.
Japan, 2000, 73, 215.
4 Y. Abe and T. Gunji, Prog. Polym. Sci., 2004, 29, 149.
5 M. Unno, S. B. Alias, M. Arai, K. Takada, R. Tanaka and
H. Matsumoto, Appl. Organomet. Chem., 1999, 13, 303.
6 Y. Abe, M. Kaji, K. Arimitsu, K. Suyama and T. Gunji, Silicon Chem.,
2002, 1, 367.
7 V. E. Shklover, N. G. Bokii, Yu. T. Struchkov, K. A. Andrianov,
M. N. Ermakova and N. A. Dmitricheva, Zhurnal Strukturnoi Khimii,
1974, 15, 864, J. Struc. Chem., 1974, 15, 761.
8 K. A. Andrianov, M. N. Ermakova, N. A. Dmitricheva, V. E. Shklover,
N. G. Bokii and Yu. T. Struchkov, Dokl. Akad. Nauk. SSSR, 1975, 220,
1321.
9 K. A. Andrianov, M. N. Ermakova, N. A. Dmitricheva and
N. P. Mashchenko, Izv. Akad. Nauk. SSSR, Ser. Khim., 1977, 47,
1393, J. Gen. Chem. USSR, 1977, 47, 1277.
10 M. B. Fromberg, Yu. K. Petrashko, V. D. Vozhova and
K. A. Andrianov, Izv. Akad. Nauk. SSSR, Ser. Khim., 1965, 660,
Bull. Acad. Sci, USSR, Div. Chem. Sci., 1965, 638.
Scheme 2 (x) PhMeSiCl2, py, yield 92%; (xi) Br2, yield 98%; (xii)
Ph2Si(OH)2, py, yield 97%; (xiii) 10, py, yield 97%.
11 A. A. Zhdanov, B. A. Izmailov, V. D. Myakushev, M. V. Sobolevskii,
L. V. Sobolevskaya and E. K. Dobrovinskaya, Izv. Akad. Nauk. SSSR,
Ser. Khim., 1981, 30, 1861, Bull. Acad. Sci, USSR, Div. Chem. Sci., 1981,
30, 1530.
12 A. A. Zhdanov, N. A. Dmitricheva, M. N. Ermakova, B. D. Lavrukhin
and T. V. Strelkova, Zhurnal Obshchei Khimii, 1984, 54, 343, J. Gen.
Chem. USSR, 1984, 54, 306.
13 N. N. Makarova, B. D. Lavrukhin, T. V. Timofeeva and
V. N. Zelencheva, Izv. Akad. Nauk SSSR, Ser. Khim., 1985, 34, 1114,
Bull. Acad. Sci, USSR, Div. Chem. Sci., 1985, 34, 1017.
14 K. A. Andrianov, A. I. Nogaideli, N. N. Makarova and
O. B. Mukbaniani, Izv. Akad. Nauk SSSR, Ser. Khim., 1977, 26,
1388, Bull. Acad. Sci, USSR, Div. Chem. Sci., 1977, 26, 1282.
15 N. N. Makarova and B. D. Lavrukhin, Izv. Akad. Nauk SSSR, Ser.
Khim., 1986, 35, 652, Bull. Acad. Sci, USSR, Div. Chem. Sci., 1986, 35,
596.
molecular weight compounds that were easily separated from the
desired product 8 by column chromatography.
9 was synthesized similar to 8 except that a larger silanediol was
used in the reaction with 12 (reaction (xiii)). In this case, 1,5-
dihydroxy-1,1,5,5-tetramethyl-3,3-diphenylsiloxane 10 was used.
Again, the cis isomer of 12 reacted quantitatively to form 9. Thus,
we have demonstrated a general method that can be used to
synthesize bicyclosiloxanes of arbitrary bridge lengths and bridge-
head positions.
This work was supported by Basic Energy Sciences, US
Department of Energy, Department of Science.
Wenmei Xue, Mayfair C. Kung and Harold H. Kung*
Department of Chemical and Biological Engineering, Northwestern
University, Evanston, IL 60208-3120, USA.
16 N. V. Chizhova, T. V. Astapova, P. V. Petrovakii and N. N. Makarova,
Russ. Chem. Bull. Int. Ed., 2000, 49, 1430.
17 H. Uchida, Y. Kabe, K. Yoshino, A. Kawamata, T. Tsumuraya and
S. Masamune, J. Am. Chem. Soc., 1990, 112, 7077.
E-mail: hkung@northwestern.edu; Fax: 1-847-4671018
18 Z. Chang, M. C. Kung and H. H. Kung, Chem. Commun., 2004, 206.
19 J. Beckmann, D. Dakternieks, A. Duthie and R. C. Foitzik, Silicon
Chem., 2003, 2, 27.
Notes and references
1 See e.g.: (a) M. Bauer and J. Schneider, Handbook of Adhesive
Technology, Marcel Dekker, Inc., New York, 2nd edn., 2003, p. 875; (b)
J.-P. Mollie, Plastics for Electronics, ed. M. Goosey, Kluwer, Dordrecht,
The Netherlands, 2nd edn., 1999, p. 25; (c) M. Takahashi, Organosilicon
Chemistry III: From Molecules to Materials, ed. N. Auner and J. Weis,
Wiley-VCH Verlag GmbH, Weinheim, Germany, 1998, p. 555.
20 (a) R. E. Swaim, W. P. Weber, H. G. Boettger, M. Evans and
F. M. Bockhoff, Org. Mass Spectrom., 1980, 15, 304–11; (b) E. I. Mysov,
I. S. Akhrem and D. V. Avetisyan, Org. Mass Spectrom., 1976, 11,
358–63.
21 A. Morikawa, M. Kakimoto and Y. Imai, Macromolecules, 1991, 24,
3469.
2166 | Chem. Commun., 2005, 2164–2166
This journal is ß The Royal Society of Chemistry 2005