Supramolecular architecture-directed synthesis of a reactive and purely inorganic ladder polyhydrosilsesquioxane

Article abstract of DOI:10.1039/b904057a

A reactive and purely inorganic high Mw perfect ladder polyhydrosilsesquioxane (H-LPSQ) was first prepared under the direction of two imperative supramolecular architectures: ladder superstructure (H-LS) and donor-acceptor complex (DAC).

Full text of DOI:10.1039/b904057a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Supramolecular architecture-directed synthesis of a reactive and purely
inorganic ladder polyhydrosilsesquioxanew
Zhongjie Ren,a Xinyu Cao,b Ping Xie,a Rongben Zhang,*a Shouke Yan*c and Yongmei Mab
Received (in Cambridge, UK) 26th February 2009, Accepted 1st May 2009
First published as an Advance Article on the web 1st June 2009
DOI: 10.1039/b904057a
A reactive and purely inorganic high Mw perfect ladder
polyhydrosilsesquioxane (H-LPSQ) was first prepared under the
direction of two imperative supramolecular architectures: ladder
superstructure (H-LS) and donor–acceptor complex (DAC).
invalid for the synthesis of H-LPSQ. Furthermore, in
comparison with the synthesis of R-LPSQs, it is more difficult
to synthesize H-LPSQ for the following reasons: (1) Si–H is
highly reactive and unstable; (2) H-LPSQ is purely inorganic
and has extremely low solubility in almost all kinds of
solvents; (3) There is no p–p stacking interaction or utilizable
stereo-effect.
Ladder polymers are of special importance as advanced
materials due to their superior thermo-resistance and mechanical
strength.1 Recently, many studies have focused on high molecular
weight (Mw) ladder organo-grafted polysilsesquioxanes
(R-LPSQs) for their unique hybrid double chain structures.2–4
However, it is very difficult to obtain well ordered high
Mw R-LPSQs from a trifunctional monomer (RSiCl3) due
to the formation of undesirable branched or cross-linked
products. In 1960, Brown et al.5 first reported a high Mw
ladder polyphenylsilsesquioxane (Ph-LPSQ). Nevertheless, its
structure was refuted later by Frey et al.6 Brook4 pointed out
that the high Mw ladder polysilsesquioxanes reported are
generally random networks, but the ladder structure may be
obtained under certain controlled conditions.
Here, we report a supramolecular architecture-directed
approach to the perfect H-LPSQ as shown in Scheme 1. It
includes two steps: (A) precoupling and ladder superstructure
(H-LS) directed synthesis of sacrificial 4,6-bis(octyloxy)benzene-
1,3-diamine-bridged ladder polyhydrosiloxanes (H-DLPS);
(B) a donor–acceptor complex (DAC)-based synchronous
cleavage of the bridge and in situ condensation. 4,6-Bis(octyloxy)-
benzene-1,3-diamine was intentionally chosen as the precoupling
agent because the octyloxy-substitutions are necessary to
increase the solubility of the series precursors. Consequently,
high molecular weight H-DLPS and H-LPSQ can be obtained.
The two imperative supramolecular architectures, H-LS and
DAC, determine the ladder regularity of H-LPSQ, and the
careful control of the reaction conditions plays an important
role in preserving the reactive Si–H groups.
Ladder polyhydrosilsesquioxane (H-LPSQ) is not only an
important material with good dielectric behavior,7 but also an
indispensable precursor for a variety of functional ladder
polymers with photoelectric and liquid-crystalline properties.
Therefore, synthesis of perfect, high Mw H-LPSQ is of
particular significance in both theoretical and practical respects.
Supramolecular template polymerization is an effective
method to synthesize all kinds of polymers with designed
structures.8,9 To synthesize perfect R-LPSQs, Zhang et al.
proposed a supramolecular template strategy named ‘‘supra-
molecular architecture-directed stepwise coupling and
polymerization’’,10 by which a series of well-defined organo-
bridged ladder polysiloxanes (R-OLPSs) and R-LPSQs11,12
Bailey1 has proposed that the most desirable type of
reaction for the formation of a perfect ladder is the one
in which both sides of the ladder are formed simultaneously.
have been prepared. Especially,
a perfect ladder poly-
silsesquioxane with a special triphenylene side group which
has strong p-stacking and enhanced stereo-selective effects was
synthesized for the first time.13 However, this approach is
a State Key Laboratory of Polymer Physics and Chemistry, Beijing
National Laboratory for Molecular Sciences (BNLMS), Institute of
Chemistry, Chinese Academy of Sciences, Beijing, China.
E-mail: zhangrb@iccas.ac.cn; Fax: (+86)10 6255 9373;
Tel: (+86)10 6256 5612
b Laboratory of New Materials, Beijing National Laboratory for
Molecular Sciences, Institute of Chemistry Chinese Academy of
Sciences, Beijing, China
c State Key laboratory of Chemical Resource Engineering, Beijing
University of Chemical Technology, Beijing, China.
E-mail: skyan@mail.buct.edu.cn; Tel: (+86)10 6441 3161
w Electronic supplementary information (ESI) available: Experimental
procedures and details; HNMR, XRD, IR and DSC of new polymers.
See DOI: 10.1039/b904057a
Scheme 1 Synthetic route to H-LPSQ.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 4079–4081 | 4079

View Article Online
In this work, HSiCl3 was transferred into a ladder precursor
(H-I) by precoupling reaction under ꢁ15 1C. Noteworthily,
H-Is can be self-assembled into perfect ladder superstructure
(H-LS) so as to achieve the synchronous growth of the ladder.
The sacrificial perfect H-DLPS is then synthesized via
stoichiometric hydrolysis-dehydrochlorination condensation
reaction of H-LS.
In order to obtain perfect H-LPSQ from H-DLPS,
‘‘synchronous cleavage’’ of the bridge and in situ condensation
is essential. Conventional acid or base catalyzed hydrolysis
can’t realize the ‘‘synchronous cleavage of the bridge’’. It has
been reported that Si–N can be cleaved and transferred into
Si–Cl by acid chloride through a nucleophilic addition-
elimination mechanism.14 To achieve ‘‘synchronous cleavage’’,
isophthalyl dichloride (IPC) was selected as the cleaving agent.
It is proposed that when IPC is added into H-DLPS, it can
form a donor–acceptor complex (DAC) with the bridge by a
synergy of hydrogen bonding between carbonyl and amino
groups, benzene ring p-stacking and dp–pp interaction of the
p-electrons of Cl and the d-orbital of Si. The Cl-atom of IPC
then links to the Si-atom, and breaks the Si–N bonds of the
bridge and further transfers them into two Si–Cl bonds.
The formation of DAC ensures the synchronous breakage of
the two Si–N bonds on the same bridge, so that Si–O–Si can
be formed in situ by hydrolysis and dehydrochlorination
condensation of the two new Si–Cl. In addition, IPC is added
in a batch manner so that the double-chain of H-DLPS
maintains a ladder structure and the in situ condensation can
carry through most effectively.
Fig. 1 A. 29Si NMR spectra of (a) Ladder superstructure (H-LS);
(b) H-DLPS; (c) Ch-LPSQ; B. Solid state CP/MAS 29Si NMR of
H-LPSQ.
regular ladder structure rather than randomly interlaced or
branched structures.
Similarly, the sacrificial H-DLPS was also confirmed to be
the perfect ladder polymer. (Table 1, Fig. 1A-b and Fig. S2 in
the ESIw). Moreover, a high glass transition temperature Tg
(67.6 1C) is attributed to great restriction of the internal
rotation of the double main chain for a highly rigid ladder
structure (see Fig. S3 in the ESIw).
The formation of a stable donor–acceptor complex (DAC)
of IPC with H-DLPS also needs to be confirmed, which is
critical for the ‘‘synchronous cleavage’’. As shown in Fig. 2,
when IPC and H-DLPS were mixed with equal molar amounts
of IPC/the bridge, a low-energy absorbance at 386 nm
emerged, which is different from the characteristic peaks of
IPC and H-DLPS. This suggests the formation of DAC
adducts.
To further confirm that IPC could realize ‘‘synchronous
cleavage of the bridge’’, a model compound containing a Si–H
group, N,N0-bis(dimethylsilyl)-4,6-bis(octyloxy)benzene-1,3-
diamine, was synthesized. Then IPC was added stepwise
to break the bridge of the model compound. 29Si NMR
measurement was conducted to monitor the kinetic process
during the reaction (Fig. 3). When the model compound was
mixed with IPC in a molar ratio of 3 : 2, there were two peaks
at d = ꢁ15.8 and d = 12.5 ppm respectively (Fig. 3b),
indicating existence of two kinds of silicon skeletons. The
peak at d = ꢁ15.8 ppm is assigned to the organo-bridged
model compound, and the peak at d = 12.5 ppm is assigned to
dimethylchlorosilane derived from the break of bridge on the
model compound by IPC. It suggests that the two Si–N bonds
at the same bridge should be broken simultaneously, or a third
peak should appear. While the model compound and IPC were
mixed in a 1 : 1 molar ratio, only one peak of d = 12.5 ppm
was observed (Fig. 3c), indicating that it reacted with IPC
completely and only dimethylchlorosilane could be detected.
To verify the reaction process, the formation of H-LS has to
be confirmed first. The concentration dependent UV spectra of
H-I show that when the concentration of H-I increases, a new
absorbance peak appears at 390 nm and the absorbance peak
at 307 nm shifts to 310 nm. It indicates that H-I molecules
can be self-assembled to give p–p and hydrogen-bonded
adducts, H-LS (see Fig. S1 in the ESIw). Table 1 shows the
characterization data of H-LS. As a supramolecular assembly
by p-stacking aided hydrogen-bonding, XRD analysis
presents two characteristic diffraction peaks corresponding
to ladder width and thickness respectively5 (see Fig. S2 in
the ESIw). The apparent degree of polymerization (DP) was
determined at 40 1C by vapor pressure osmometry (VPO) in
toluene. It indicates that H-LS possesses ca. 14 repeat units of
H-I. The 29Si NMR spectrum of H-I gives only one sharp peak
at d = ꢁ22.2 ppm with baseline-width (D) as small as
ca. 1 ppm (Fig. 1A-a) suggesting that H-LS consists of a very
Table 1 Characterization data of H-LS; H-DLPS; H-LPSQ and
Ch-LPSQ
d
Sample
DPa or Mw W/T by XRD (nm) d/D by 29Si NMRe
H-LS
H-DLPS
14b
0.82/0.40
0.94/0.40
0.45/0.40
0.72/0.41
ꢁ22.2/1
ꢁ19.7/1
—
ꢁ22.6/1
135 230c
H-LPSQ 84 370c
Ch-LPSQ 109 824c
a
b
Degree of polymerization. Determined by vapor pressure
c
d
osmometry. Determined by laser light scattering. W = the ladder
e
width and T = the ladder thickness. d/D = chemical shift and
baseline-width.
Fig. 2 UV spectra of (a) IPC; (b) H-DLPS; (c) H-DLPS and IPC
(IPC and the bridge of H-DLPS are in equal molar amounts).
ꢀc
This journal is The Royal Society of Chemistry 2009
4080 | Chem. Commun., 2009, 4079–4081

View Article Online
character preferably attract each other.16 The helical structure
enhances the de-shielding effect along the parallel direction of
the tightly stacked helical polymer chains, and results in the
remarkable shift.17 In addition, Ch-LPSQ shows a high
Tg (143.8 1C) that is contributed by the highly regular rigid
ladder structure (see Fig. S5 in the ESIw). Because the
conversion of H-LPSQ to Ch-LPSQ doesn’t involve any
breaking of the ladder chains of H-LPSQ, the perfection of
Ch-LPSQ represents the perfection of H-LPSQ. The successful
preparation of Ch-LPSQ also verifies the high reactivity of the
Si–H groups on H-LPSQ.
Fig.
3
29Si NMR spectra of (a) model monomer; (b) model
compound and IPC (3 : 2) and (c) model compound and IPC (1 : 1).
In addition, it can be deduced that the Si–H groups are inert to
IPC under the reaction conditions.
In conclusion, a reactive and perfect, purely inorganic high
Mw ladder polyhydrosilsesquioxane (H-LPSQ) has been
prepared successfully. It includes ladder superstructure
(H-LS)-directed synthesis of sacrificial 4,6-bis(octyloxy)-
benzene-1,3-diamine-bridged ladder polysiloxanes H-DLPS
and then supramolecular complex (DAC)-based synchronous
cleavage and in situ condensation. The two important
supramolecular architectures (H-LS and DAC) and the ladder
polymers (H-DLPS, H-LPSQ and Ch-LPSQ) have been well
characterized. Perfect Ch-LPSQ was obtained from H-LPSQ,
which further verifies the structural perfection and the possible
functionality of H-LPSQ.
The synchronous cleavage of the bridge results in two Si–Cl
groups forming simultaneously. If one of them is hydrolyzed
into Si–OH, it would immediately and conveniently react with
the other Si–Cl in the presence of triethylamine, which is both
catalyst and HCl absorbent, to generate a RSi–O–SiR
bond. Finally, all the bridges of H-DLPS are ruptured and
converted to Si–O–Si bonds. The Mw of the H-LPSQ is
84 370 determined by laser light scattering. In the meantime,
an aromatic polyamide was produced and detected as
by-product.15
Complete conversion of H-DLPS to the perfect ladder
H-LPSQ was also confirmed by elemental analysis, which
shows that H-LPSQ contains no nitrogen or carbon elements
but Si, O and H atoms. The distinguishable absorption of a
Si–H stretching vibration at 2251 cmꢁ1 in the IR spectrum of
H-LPSQ suggests Si–H groups should be well preserved (see
Fig. S6 in the ESIw). The XRD profile of H-LPSQ also gives
two distinct peaks that corresponding to the ladder width and
thickness respectively (see Fig. S4 in the ESIw). Although
solution 29Si NMR is efficient for confirming the ladder
regularity, it can’t be used for H-LPSQ because of its
extremely low solubility (o5 wt%) in organic solvents. Hence,
solid-state 29Si NMR was adopted. As a result, a single peak at
d = ꢁ87.9 ppm was observed in the solid-state 29Si NMR
spectrum (Fig. 1B). It indicates that all the Si atoms are in very
similar microenvironments, namely regular ladder skeletons.
For further confirmation of the perfection of H-LPSQ and
reactivity of its Si–H groups, a solubility-increasing cyclohexyl
group was introduced into the H-LPSQ’s backbone by the
hydrosilylation of cyclohexylene catalyzed by Cp2PtCl2. The
resultant ladder polycyclohexylsilsesquioxane (Ch-LPSQ) also
gives two distinct peaks in the XRD spectrum corresponding
to the ladder width and thickness, respectively. (see Fig. S4 in
the ESIw). Moreover, as shown in Fig. 1A-c, the 29Si NMR
spectrum of Ch-LPSQ shows exceedingly sharp peak at
d = ꢁ22.6 ppm with D of B1 ppm, indicating that all the
Si atoms exist in almost the same chemical environment and
the ladder possesses perfect tacticity. In contrast to the general
silsesquioxane (-SiO3/2) with chemical shift d = ꢁ70 B ꢁ80 ppm
in the 29Si NMR spectrum, the perfect Ch-LPSQ gives
the -SiO3/2 characteristic d = ꢁ22.6 ppm with a remarkably
down-field shift of B60 ppm. Furthermore, the d value of
perfect Ch-LPSQ is independent of the solvent’s polarity
(acetone, toluene). These results indicate that ladder Ch-LPSQ
tends to form a helical structure in solution, which may be
formed by strong intra-molecular dipole–dipole interactions of
the RSi–O–SiR linkages. The Si–O–Si units with 50% ionic
The financial support of NSFC [No. 50073028, 29974036,
20174047, 50521302 and the Outstanding Youth Fund
(No.20425414)] as well as MOST under grant No.
2007CB935902 & 2007CB935904 are gratefully acknowledged.
Notes and references
1 W. J. Bailey, in Concise Encyclopedia of Polymer Science
& Engineering, ed. J. L. Kroschwitz, John Wiley & Sons, Inc,
New York, 1990, pp. 516–528.
2 F. J. Feher, D. A. Newman and J. F. Walzer, J. Am. Chem. Soc.,
1989, 111, 1741–1748.
3 M. Khiterer and K. J. Shea, Nano Lett., 2007, 7, 2684–2687.
4 M. A. Brook, Silicon in Organic, Organometallic, and Polymer
Chemistry, Wiley-Interscience, New York, 2000, pp. 309–331.
5 F. Brown Jr., L. H. Vogt Jr., A. Katchman, J. W. Eustance,
K. M. Kiser and K. W. Krantz, J. Am. Chem. Soc., 1960, 82,
6194–6195.
6 C. L. Frey and J. M. Klosowski, J. Am. Chem. Soc., 1971, 93,
4599–4601.
7 R. H. Baney, M. Itoh, A. Sakakibara and T. Suzuki, Chem. Rev.,
1995, 95, 1409–1430.
8 A. Harada, J. Li and M. Kamachi, Nature, 1993, 364,
516–518.
9 M. Biesalski, Angew. Chem., Int. Ed., 2005, 44, 3297–3301.
10 Q. Zhou, S. Yan, C. C. Han, P. Xie and R. Zhang, Adv. Mater.,
2008, 20, 2970–2976.
11 H. Tang, J. Sun, J. Jiang, X. Zhou, T. Hu, P. Xie and R. B. Zhang,
J. Am. Chem. Soc., 2002, 124, 10482–10488.
12 (a) T. Zhang, K. Deng, P. Zhang, P. Xie and R. B. Zhang,
Chem.–Eur. J., 2006, 12, 3630–3635; (b) Y. Wan, L. Yang,
J. Zhang, P. Zhang, P. Fu, T. Zhang, P. Xie and R. Zhang,
Macromolecules, 2006, 39, 541–547.
13 X. Zhang, P. Xie, Z. Shen, J. Jiang, C. Zhu, H. Li, T. Zhang,
C. C. Han, L. Wan, S. K. Yan and R. B. Zhang, Angew. Chem.,
Int. Ed., 2006, 45, 3112–3116.
14 Y. Oishi, M. Kakimoto and Y. Imai, Macromolecules, 1988, 21,
547–550.
15 A. E. Lozano, J. de Abajo and J. G. de la Campa, Macromolecules,
1997, 30, 2507–2508.
16 C. Eaborn, Organosilicon Compounds, Butterworths Scientific
Publications, London, 1960, pp. 89.
17 H. H. Willard, Instrumental Methods of Analysis, D. Van Nostrand
Company, New York, sixth edn, 1981, pp. 332.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 4079–4081 | 4081