Takashi Hamada1, Yuki Nakanishi2,3, Kenta Okada1,2, Joji Ohshita1,3,4. 1. Collaborative Research Laboratory, Graduate School of Advanced Science and Engineering, Hiroshima University, 1-4-1 Kagamiyama, Higashi-Hiroshima, Hiroshima 739-8527, Japan. 2. Technical Research Center, Mazda Motor Corporation, 3-1 Shinchi, Fuchu-cho, Aki-gun, Hiroshima 730-8670, Japan. 3. Applied Chemistry Program, Graduate School of Advanced Science and Engineering, Hiroshima University, 1-4-1 Kagamiyama, Higashi-Hiroshima, Hiroshima 739-8527, Japan. 4. Division of Materials Model-Based Research, Digital Monozukuri (Manufacturing) Education and Research Center, Hiroshima University, 3-10-32 Kagamiyama, Higashi-Hiroshima, Hiroshima 739-0046, Japan.
Abstract
With the aim of developing an improved strategy for the preparation of ethylene-bridged polysilsesquioxanes as thermal insulator materials, this paper describes the synthesis of a crack- and shrinkage-free ethylene-bridged polysilsesquioxane film by the hydrosilylation reaction of hydrodimethyl-silylated oligomethylsilsesquioxane (MSQ-SiH) and dimethylvinyl-silylated oligomethylsilsesquioxane (MSQ-SiVi) in the presence of Karstedt's catalyst. Polysilsesquioxane precursors were prepared by the sol-gel reaction of triethoxymethylsilane and the successive capping reaction with chlorodimethylsilane and chlorodimethylvinylsilane. The obtained ethylene-bridged polysilsesquioxane film showed lower density and thermal diffusivity (1.13 g/cm3 and 1.15 × 10-7 m2/s, respectively) than a polymethylsilsesquioxane film (1.34 g/cm3 and 1.36 × 10-7 m2/s, respectively). As a result of the introduction of the SiCCSi ethylene bridge, the thermal insulation property of the polysilsesquioxane film was enhanced.
With the aim of developing an improved strategy for the preparation of ethylene-bridged polysilsesquioxanes as thermal insulator materials, this paper describes the synthesis of a crack- and shrinkage-free ethylene-bridged polysilsesquioxane film by the hydrosilylation reaction of hydrodimethyl-silylated oligomethylsilsesquioxane (MSQ-SiH) and dimethylvinyl-silylated oligomethylsilsesquioxane (MSQ-SiVi) in the presence of Karstedt's catalyst. Polysilsesquioxane precursors were prepared by the sol-gel reaction of triethoxymethylsilane and the successive capping reaction with chlorodimethylsilane and chlorodimethylvinylsilane. The obtained ethylene-bridged polysilsesquioxane film showed lower density and thermal diffusivity (1.13 g/cm3 and 1.15 × 10-7 m2/s, respectively) than a polymethylsilsesquioxane film (1.34 g/cm3 and 1.36 × 10-7 m2/s, respectively). As a result of the introduction of the SiCCSi ethylene bridge, the thermal insulation property of the polysilsesquioxane film was enhanced.
Over the past several
decades, sustainable and efficient thermal
insulation technology has been demanded for energy saving and reduction
in greenhouse gas emission. The efficient use of energy is particularly
important in buildings because a large amount of energy is consumed
and carbon dioxide is emitted in buildings worldwide. In this context,
thermal insulator materials are used in walls and roofs to reduce
thermal conductivity for energy saving.[1−3] For example, cellular
plastics and fibrous materials such as glass, rock, cellulose, and
cotton are conventional thermal insulation materials, which are installed
in building envelopes, creating a thick insulation layer. However,
the use of thick thermal insulator materials in building envelopes
is disadvantageous in terms of space and cost saving. Therefore, the
development of enhanced thermal insulation materials is desired for
realizing high energy efficiency and cost reduction.To achieve
low thermal conductivity in thermal insulator materials,
aerogels have emerged as promising materials.[4,5] Aerogels
are highly porous and low-density materials prepared by supercritical
drying of a gel, and their high porosity is attractive for good thermal
insulation materials.[6] Aerogels are typically
classified into inorganic aerogels and organic polymer aerogels. Among
the inorganic aerogels, silica aerogels have been extensively studied
since Kistler reported the first example in 1931.[7] Silica aerogels have a unique nanostructure that renders
them attractive in the field of nanomaterial science. However, most
silica aerogels have several disadvantages such as brittleness, low
mechanical strength, and high cost, which limit their practical application.
Meanwhile, organic polymer aerogels such as polystyrene,[8] polyurethane,[9] cellulose,[10] polyimide,[11] epoxy/amine,[12] and melamine/formaldehyde[13] have also been the focus of research studies. The use of
organic polymers in aerogels allows tuning their properties by choosing
the appropriate monomer for the polymerization, and a number of organic
polymer aerogels with low thermal conductivity have been successfully
prepared. However, the thermal durability of organic polymers is generally
lower than that of inorganic materials, which limits the application
of organic aerogels at high-temperature conditions. For thermal insulation
materials in buildings, the nonflammable property is also important
because the thermal insulator layer delays heat conduction and prevents
flame transfer during a fire.[1] Unfortunately,
typical organic polymers are flammable;[14] for example, an organic aerogel based on carboxymethyl cellulose
produces a bright flame when placed in a flame.[15] Overall, new thermal insulator materials with high mechanical
strength, high thermal property, high flame resistance, and low production
cost are required from the viewpoint of practical application.Owing to the combination of the unique properties of organic and
inorganic components, organic–inorganic hybrid materials have
attracted increased research attention. Among these types of materials,
polysilsesquioxane stands out as a representative example.[16−18] Polysilsesquioxanes have the general formula (RSiO1.5), in which R is an organic group such
as alkyl and aryl, typically, methyl, phenyl, and vinyl group, and
SiO1.5 is the basic unit of the inorganic Si–O–Si
network. On the one hand, the organic group is usually selected depending
on the application to provide enhanced solubility, compatibility,
flexibility, or surface property.[16] On
the other hand, the Si–O–Si network enhances the thermal
and mechanical properties compared with organic polymers. Polysilsesquioxanes
can be easily prepared by the sol–gel reaction of trialkoxysilane
via acid- or base-catalyzed hydrolysis and successive polycondensation.
Many polysilsesquioxanes have been synthesized so far, and organic–inorganic
hybrid materials based on polysilsesquioxanes have been studied as
scratch-resistant materials,[19] barrier
materials,[20] fire-retardant materials,[21] optical materials,[22] low dielectric constant materials,[23] and
insulator materials.[24,25] Since polysilsesquioxane materials
are expected to combine the properties derived from the organic moiety
and the inorganic oxide, a derivative with high mechanical strength,
thermal stability, and high flame resistance can be envisaged as a
good thermal insulator material applicable at high-temperature conditions,
as long as it also exhibits low thermal conductivity.Among
polysilsesquioxanes, bridged polysilsesquioxanes composed
of an organic bridging group between two silanes have a unique structure.
Bridged polysilsesquioxanes containing methylene, ethylene, ethenylene,
ethynylene, and benzene as organic bridging spacers have been prepared
by a sol–gel method.[26] Interestingly,
bridged polysilsesquioxanes have been applied as adsorption materials,[27,28] separation membranes,[29−31] and low dielectric materials[32−34] by taking advantage of the characteristics of the organic bridging
spacer. Considering these backgrounds, we reported a reverse osmosis
(RO) membrane based on bridged polysilsesquioxane for water purification.[35] The rigidity of the organic bridging spacer
increased the membrane pore size, and RO membranes prepared from bis(triethoxysilyl)ethane,
-ethene, and -ethyne showed improved water permeability as the rigidity
of the bridging units increased in the order ethane <ethene <ethyne.
Bridged polysilsesquioxanes are promising materials for thermal insulation
because the organic bridging spacer provides porosity and the polysilsesquioxane
framework enhances the mechanical strength and the thermal property.[26,35,36] We previously reported the preparation
of bridged polysilsesquioxane by the nitrogen flow method,[37] and the molecular weight of ethylene-bridged
polysilsesquioxane was controlled by changing the amount of water,
according to a reported procedure.[37−40] A gel film of ethylene-bridged
polysilsesquioxane was obtained by heating at 80 °C for a week.
However, when subjecting the gel film to heat treatment at 200 °C
over 20 min, many cracks appeared because the elimination of unhydrolyzed
ethoxy groups in ethylene-bridged polysilsesquioxane caused serious
shrinkage of the film (unpublished data). Unfortunately, isolation
of ethylene-bridged polysilsesquioxane without ethoxy groups was hindered
by the tendency of the ethylene-bridged silane monomer having six
ethoxy groups to form a gel during the sol–gel reaction. Using
bis(trimethoxysilyl)ethane instead of bis(triethoxysilyl)ethane may
improve the shrinking problem. However, the gel formation still seems
to be difficult to control.Therefore, a new strategy for the
preparation of ethylene-bridged
polysilsesquioxanes for various applications including thermal insulator
materials is highly desired. In this paper, we report a new synthetic
method for crack- and shrinkage-free ethylene-bridged polysilsesquioxanes
that involves the hydrosilylation reaction of polysilsesquioxane precursors.
The thermal insulation property of the resulting ethylene-bridged
polysilsesquioxane is investigated.
Results and Discussion
Synthesis
of Silylated Oligomethylsilsesquioxane
As
mentioned in Introduction section, the preparation
of an ethylene-bridged polysilsesquioxane without ethoxy groups was
difficult because the ethylene-bridged silane monomer has six ethoxy
groups, and multifunctional polysilsesquioxanes tend to form a gel.
Therefore, we focused on the preparation of an ethylene-bridged polysilsesquioxane
film by hydrosilylation as a powerful tool for the formation of SiCCSi
bonds.[41]We designed a synthetic
route shown in Scheme for the preparation of oligomethylsilsesquioxanes having hydrosilyl
and vinyl groups as precursors of the ethylene-bridged polysilsesquioxane
film. First, according to a modified reported procedure,[42,43] the sol–gel reaction of triethoxymethylsilane was performed
to achieve the complete hydrolysis of the ethoxy groups and to ensure
a high content of silanol groups for the capping reaction. The hydrolysis
of triethoxymethylsilane was conducted using excess hydrochloric acid
and water to completely hydrolyze the ethoxy groups, and the polycondensation
was maintained for 3 h at 0 °C to prevent excess polycondensation
and gel formation. Despite being a highly sticky product, the obtained
oligomethylsilsesquioxane (MSQ-OH) showed high solubility in polar
organic solvents such as tetrahydrofuran (THF), acetone, and chloroform.
Scheme 1
Synthetic Route for the Precursors of Ethylene-Bridged Polysilsesquioxane
Films
To investigate the chemical
structure of MSQ-OH, 1H
nuclear magnetic resonance (NMR), 29Si NMR, and attenuated
total reflection-Fourier transform infrared spectroscopy (ATR-FTIR)
measurements were conducted. As shown in Figure a, multiple signals attributed to the silylmethyl
groups were observed in the 1H NMR spectrum at ∼0.18
ppm after hydrolysis and polycondensation of triethoxymethylsilane.
Besides, three main signals were observed at 1.24, 1.85, and 3.73
ppm, which can be attributed to unhydrolyzed ethoxy groups of the
product, ethanol as a byproduct, and THF as the reaction solvent,
respectively. The integral ratio of the silylmethyl signal at 0.18
ppm and the total methyl protons from the ethoxy groups in the product
and ethanol at 1.24 ppm was determined to be 3:0.52. Considering the
integral ratio, the ethoxy groups were considerably hydrolyzed and
consumed. However, the accurate estimation of the remaining ethoxy
groups in the product was hindered by overlapping of the signals of
the ethoxy groups in the product and in ethanol as a byproduct at
1.24 ppm in the 1H NMR of MSQ-OH. Moreover, the amount
of silanol in the product could not be determined because the signal
for the silanol group is usually weak and broad. Meanwhile, as shown
in Figure a, the 29Si NMR spectrum of MSQ-OH mainly showed multiple sharp signals
at ∼−56.7 ppm, corresponding to a T2 structure,
along with small signals at ∼−65 ppm, corresponding
to a T3 structure, where T refers to (MeSi(OSi)(OR)3–). Therefore, MSQ-OH had the T2 structure
MeSi(OSi)2(OR) containing two siloxane bonds, indicating
that MSQ-OH was a linear and/or cyclic structure containing silanol
or ethoxy groups in the side chain. The weight-average molecular weight
(Mw) and the polydispersity index (Mw/Mn) were estimated
to be 450 and 1.34 by gel permeation chromatography (GPC) (Table ), respectively, suggesting
an oligomeric structure for MSQ-OH with 5–7 repeating units.
In the ATR-FTIR spectrum of MSQ-OH (Figure a), clear peaks at 1269 and 1049 cm–1 corresponding to the silylmethyl group and the siloxane bond were
observed, respectively. Furthermore, a broad peak assigned to silanol
groups was observed at 3285 cm–1. Therefore, it
can be concluded that MSQ-OH was a linear and/or cyclic oligomer having
many silanol groups.
Figure 1
1H nuclear magnetic resonance spectra of the
precursors
(a) MSQ-OH, (b) MSQ-SiH, and (c) MSQ-SiVi. Chemical shifts are referenced
to residual chloroform in CDCl3 (7.26 ppm).
Figure 2
29Si nuclear magnetic resonance spectra of the precursors
(a) MSQ-OH, (b) MSQ-SiH, and (c) MSQ-SiVi. Chemical shifts are referenced
to tetramethylsilane (0 ppm).
Table 1
Results on the Synthesis of Oligomethylsilsesquioxanes
for the Hydrosilylation Reaction
Mwa
Mw/Mna
residual OEt
(%)b
SiH (mol %)b
SiVi (mol %)b
MTES-OH
450
1.34
n.d.c
MTES-SiH
760
1.37
<1
48
MTES-SiVi
860
1.40
<1
52
Estimated by gel permeation chromatography
employing a polystyrene standard as a reference.
Determined by 1H NMR
relative to the SiCH3 group in the oligomethylsilsesquioxane
backbone.
Not determined.
Figure 3
Attenuated
total reflectance Fourier transform infrared spectra
of the precursors (a) MSQ-OH, (b) MSQ-SiH, and (c) MSQ-SiVi.
1H nuclear magnetic resonance spectra of the
precursors
(a) MSQ-OH, (b) MSQ-SiH, and (c) MSQ-SiVi. Chemical shifts are referenced
to residual chloroform in CDCl3 (7.26 ppm).29Si nuclear magnetic resonance spectra of the precursors
(a) MSQ-OH, (b) MSQ-SiH, and (c) MSQ-SiVi. Chemical shifts are referenced
to tetramethylsilane (0 ppm).Attenuated
total reflectance Fourier transform infrared spectra
of the precursors (a) MSQ-OH, (b) MSQ-SiH, and (c) MSQ-SiVi.Estimated by gel permeation chromatography
employing a polystyrene standard as a reference.Determined by 1H NMR
relative to the SiCH3 group in the oligomethylsilsesquioxane
backbone.Not determined.Next, the capping reaction
of MSQ-OH was performed using chlorodimethylsilane
and chlorodimethylvinylsilane, as shown in Scheme . Since MSQ-OH was a sticky product, it was
dissolved in THF for the capping reaction with the chlorosilanes.
After the reaction with chlorodimethylsilane, the morphology of the
product changed to a colorless liquid. Figure b shows the 1H NMR of hydrodimethyl-silylated
oligomethylsilsesquioxane (MSQ-SiH) after the capping reaction with
chlorodimethylsilane. New signals corresponding to hydrosilyl and
silylmethyl groups appeared at 4.71 and 0.08–0.22 ppm, respectively.
Although the sharp signals attributed to the silylmethyl group in
the dimethylsilyl capping agent were almost overlapped with those
of the silylmethyl group in the oligomethylsilsesquioxane backbone,
the integral ratio of the hydrosilyl group and the total number of
silylmethyl groups was 1:12.3, namely, the molar percentage of the
silylmethyl group was 48% per repeat unit of −MeSiO1.5– (Table ).
The 29Si NMR spectra of MSQ-OH changed dramatically after
the capping reaction. As shown in Figure b, a new signal attributed to the dimethylsilyl
group was observed at −5.51 ppm. Then, multiple sharp signals
at −56.7, corresponding to the T2 structure, shifted
to −64.9 ppm, which can be assigned to a T3 structure
of MeSi(OSi)3 after the capping reaction. This shows that
the silanol groups in MSQ-OH were converted to SiOSiMe2H. Indeed, as shown in Figure b, the broad peak attributed to the silanol group completely
disappeared after the capping reaction, and a shoulder peak ascribed
to a siloxane bond and a peak attributed to the hydrosilyl group appeared
at 1105 and 2131 cm–1, respectively, in the ATR-FTIR
spectrum of MSQ-SiH. These results indicate that the capping reaction
of MSQ-OH with chlorodimethylsilane proceeded quantitatively. The Mw and Mw/Mn of MSQ-SiH were estimated to be 760 and 1.37
by GPC (Table ), respectively.
Apart from the expected increase due to the introduction of the dimethylsilyl
group as a capping agent, no substantial increase in the molecular
weight was observed, further supporting that the capping reaction
proceeded quantitatively without the occurrence of intermolecular
reactions of MSQ-OH. It was concluded that the capping reaction was
controlled using excess chlorodimethylsilane at 0 °C to complete
the capping reaction avoiding the siloxane formation. Furthermore,
although slightly observed in the 1H NMR of MSQ-SiH, the
presence of ethoxy groups was almost negligible. Therefore, it can
be concluded that the obtained MSQ-SiH had a hydrosilyl group per
two repeating units of the siloxane bond and should be a good oligomer
for the preparation of the ethylene-bridged polysilsesquioxane film
by the hydrosilylation reaction.The capping reaction of MSQ-OH
with chlorodimethylvinylsilane was
conducted in a similar way, affording dimethylvinyl-silylated oligomethylsilsesquioxane
(MSQ-SiVi) as a colorless liquid. The Mw and Mw/Mn of MSQ-SiVi were estimated to be 860 and 1.40, respectively (Table ). As shown in Figure c, new signals corresponding
to vinyl and silylmethyl groups appeared at 5.72–6.18 and 0.07–0.20
ppm, respectively. The integral ratio of the vinyl group and total
silylmethyl groups was 3:11.7, and the molar percentage of the vinyl
group was determined to be 52% per repeat unit of −MeSiO1.5– (Table ). The calculated molar ratio of vinyl and silylmethyl groups
in the oligomethylsilsesquioxane backbone was roughly 1:2, which was
in agreement with the capping reaction of MSQ-OH with chlorodimethylsilane.
Moreover, the signals corresponding to the T2 structure
were shifted to −64.6 ppm in the 29Si NMR spectrum
of MSQ-SiVi (Figure c), and the broad peak of silanol completely disappeared in the ATR-FTIR
spectrum of MSQ-SiVi (Figure c). Therefore, MSQ-SiVi could be also considered a good precursor
of ethylene-bridged polysilsesquioxane by the hydrosilylation reaction.
Preparation of the Ethylene-Bridged Polysilsesquioxane Film
by the Hydrosilylation Reaction
The appropriate reaction
temperature for the hydrosilylation reaction was determined by performing
a thermogravimetric and differential thermal analysis (TG-DTA) of
a mixture of MSQ-SiH and MSQ-SiVi in the presence of a 5 wt % Karstedt’s
catalyst solution (10 times diluted with toluene) at a heating rate
of 10 °C/min under an air flow of 100 mL/min. Upon increasing
the temperature from room temperature to 1000 °C, three exothermic
peaks were observed at ∼80 °C, 280 °C, and 480 °C,
as shown in Figure . The first strong exothermic peak at 80 °C can be attributed
to the reaction heat of the hydrosilylation of MSQ-SiH and MSQ-SiVi,
and thus, we considered that the hydrosilylation reaction would effectively
proceed at 80 °C. Meanwhile, second and third exothermic peaks
were observed at 280 and 480 °C, along with a weight loss that
started at ∼250 and 400 °C, respectively. According to
the literature, in polyhedral oligomeric silsesquioxanes having vinyl
groups, the decomposition of the vinyl groups due to peroxidation
and polymerization starts at ∼200 °C in air.[44] Bearing this in mind, we performed the hydrosilylation
reaction of MSQ-SiH and MSQ-SiVi in the presence of Karstedt’s
catalyst as follows: First, the mixture was heated at 80 °C for
1 h to start the hydrosilylation reaction efficiently, followed by
heating at 160 °C for 4 h to complete the hydrosilylation reaction
below the decomposition temperature of 200 °C. For the preparation
of the ethylene-bridged polysilsesquioxane film, a mixture of MSQ-SiH
and MSQ-SiVi with a feed molar ratio of 1:1 was added in a poly(tetrafluoroethylene-co-perfluoroalkylvinyl ether) (PFA) vial together with Karstedt’s
catalyst. The mixture was heated at 80, 120, and 140 °C for 1
h at each temperature and then heated at 160 °C for 4 h to obtain
a colorless and sufficiently stiff ethylene-bridged polysilsesquioxane
film with a MSQ-SiH/MSQ-SiVi ratio of 1:1 (hereafter, denoted as EBSQ-1).
It was noted that the remaining Karstedt’s catalyst did not
influence the color of EBSQ-1 because the amount of Karstedt’s
catalyst was low.
Figure 4
Thermogravimetric analysis of MSQ-SiH and MSQ-SiVi in
the presence
of a 5 wt % Karstedt’s catalyst solution (10 times diluted
with toluene) at a heating rate of 10 °C/min under an air flow
of 100 mL/min.
Thermogravimetric analysis of MSQ-SiH and MSQ-SiVi in
the presence
of a 5 wt % Karstedt’s catalyst solution (10 times diluted
with toluene) at a heating rate of 10 °C/min under an air flow
of 100 mL/min.Figure shows the
ATR-FTIR spectra before and after the hydrosilylation reaction of
MSQ-SiH and MSQ-SiVi using Karstedt’s catalyst. In a mixture
of MSQ-SiH and MSQ-SiVi before the hydrosilylation reaction, a clear
peak corresponding to hydrosilyl groups was observed at 2131 cm–1, along with small peaks of vinyl groups at 3053 and
1595 cm–1. These three peaks almost disappeared
after heating at 160 °C for 4 h, indicating that the hydrosilylation
reaction of MSQ-SiH and MSQ-SiVi proceeded quantitatively to form
EBSQ-1 in the film state, as shown in Figure . No signal attributed to ethylene was detected
in the ATR-FTIR spectrum of EBSQ-1 because the peak of SiCCSi was
overlapped with that of the silylmethyl groups in the oligomethylsilsesquioxane
backbone and the capping agent.
Figure 5
Attenuated total reflectance Fourier transform
infrared spectra
of a mixture of MSQ-SiH and MSQ-SiVi before and after the hydrosilylation
reaction.
Figure 6
Schematic illustration of the hydrosilylation
of MSQ-SiH and MSQ-SiVi.
Attenuated total reflectance Fourier transform
infrared spectra
of a mixture of MSQ-SiH and MSQ-SiVi before and after the hydrosilylation
reaction.Schematic illustration of the hydrosilylation
of MSQ-SiH and MSQ-SiVi.The cross-sectional field
emission scanning electron microscopy
(FE-SEM) image depicted in Figure S1 revealed
that EBSQ-1 was an entirely uniform and crack-free film. Therefore,
it was demonstrated that a crack- and shrinkage-free ethylene-bridged
polysilsesquioxane film can be successfully prepared by the hydrosilylation
reaction of oligomethylsilsesquioxanes containing hydrosilyl and vinyl
groups.
Thermal Diffusivity of the Ethylene-Bridged Polysilsesquioxane
Films
As mentioned in Introduction section, the SiCCSi ethylene bridge affords sufficient porosity
to the previously reported ethylene-bridged polysilsesquioxane to
be used as a membrane for water desalination.[35] Therefore, we expected it to be a good thermal insulating material
because of its porosity. Unfortunately, when the ethylene-bridged
polysilsesquioxane film, which was prepared by hydrolysis and polycondensation
of bis(triethoxysilyl)ethane with water, was heated at 200 °C
over 20 min, many cracks appeared in the film due to the shrinkage
arising from the elimination of ethoxy groups. In EBSQ-1, owing to
the lack of shrinkage during the formation of the film, the SiCCSi
ethylene bridge could be expected to improve the thermal insulating
property. Consequently, the thermal diffusivity of EBSQ-1 was measured
to reveal the effect of SiCCSi on the thermal insulating property.
To provide a comparison, the thermal diffusivity of a film prepared
from MSQ-OH without SiCCSi was also evaluated.To obtain the
film from MSQ-OH for the thermal diffusivity measurements, a 20 wt
% MSQ-OH solution in THF was heated at 160 °C for 4 h in a PFA
vial. However, many holes were observed in the freestanding film due
to the rapid evaporation of THF and the water produced by the polycondensation
of silanol groups. Therefore, the 20 wt % MSQ-OH solution in THF was
first heated at 80 °C overnight to slowly remove THF and water,
followed by heating at 160 °C for 4 h. The resulting film prepared
from MSQ-OH (denoted as MSQ) was uniform and colorless, with a thickness
of ∼600 μm, similar to that of EBSQ-1 (∼700 μm).
The averaged thermal diffusivity of MSQ was 1.36 × 10–7 m2/s, which was higher than that of EBSQ-1 (1.15 ×
10–7 m2/s), as shown in Table . This indicates that, as expected,
the thermal insulating property of the ethylene-bridged film was superior
to that of the MSQ film, demonstrating the key role of the SiCCSi
ethylene bridge in providing sufficient void space. The thermal diffusivity
of EBSQ-1 was also lower than that of organic polymers such as polyvinylchloride,
nylon, low-density polyethylene, high impact polystyrene, and polypropylene.[45]
Table 2
Thermal Diffusivity
and 5 wt % (Td5) and 10 wt
% (Td10) Loss Temperature for
MSQ-OH, EBSQ-1, EBSE-2,
and EBSQ-3
MSQ-SiH
MSQ-SiVi
thermal diffusivity (10–7 m2/s)a
density (g/cm3)
Td5 (°C)b
Td10 (°C)b
EBSQ-1
1
1
1.15
1.13
302
285
EBSQ-2
2
1
1.18
1.15
298
383
EBSQ-3
4
1
1.28
1.16
364
418
MSQ
1.36
1.34
436
481
The measurements
for thermal diffusivity
were repeated at least three times, and the averaged thermal diffusivity
was calculated.
Measured
at a heating rate of 10
°C/min under an air flow of 100 mL/min.
The measurements
for thermal diffusivity
were repeated at least three times, and the averaged thermal diffusivity
was calculated.Measured
at a heating rate of 10
°C/min under an air flow of 100 mL/min.If the void space provided by the ethylene-bridged
structure is
essential for the thermal insulating property, the thermal diffusivity
of the ethylene-bridged polysilsesquioxane film can be expected to
depend on the amount of SiCCSi units. To understand the effect of
the SiCCSi ethylene bridge, ethylene-bridged polysilsesquioxanes having
different amounts of SiCCSi units were prepared by changing the feed
molar ratio of MSQ-SiH and MSQ-SiVi to 2:1 and 4:1, respectively.
The hydrosilylation reaction of MSQ-SiH and MSQ-SiVi at feed molar
ratios of 2:1 and 4:1 also proceeded, affording ethylene-bridged polysilsesquioxane
films with a thickness of ∼700 μm denoted as EBSQ-2 and
EBSQ-3, respectively. Again, complete disappearance of the small peaks
of the vinyl groups of the precursors was evidenced in the ATR-FTIR
spectra (Figure S2). Surprisingly, hydrosilyl
peaks were not observed even under an excess of hydrosilyl groups,
which might be due to the side reaction of hydrosilyl groups with
ambient moisture. The averaged thermal diffusivity of EBSQ-2 and EBSQ-3
were determined to be 1.18 × 10–7 and 1.28
× 10–7 m2/s, respectively (Table ), indicating that
the thermal diffusivity of the ethylene-bridged polysilsesquioxane
films decreased with increasing the amount of SiCCSi units (in the
order EBSQ-1 > EBSQ-2 > EBSQ-3). Therefore, EBSQ-1 with the
largest
amount of SiCCSi units exhibited the lowest thermal diffusivity. On
the other hand, EBSQ-3 showed the highest thermal diffusivity because
EBSQ-3 prepared from a feed molar ratio of MSQ-SiH and MSQ-SiVi of
4:1 contained the lowest amount of SiCCSi units. This trend confirms
that the SiCCSi ethylene bridge improves the thermal insulating property
by providing the void space that ensures sufficient porosity, even
though the void space was too small to analyze by FE-SEM. As shown
in Table , the densities
of the films decreased with increasing the amount of SiCCSi units
(the amount of SiCCSi units followed the order EBSQ-1 > EBSQ-2
> EBSQ-3
> MSQ). This supports that sufficient void space was created by
the
formation of SiCCSi bonds.
Thermal Stability of the Ethylene-Bridged
Polysilsesquioxane
Films
Similar to the thermal insulating property, thermal
stability is also a main concern for thermal insulating materials
to be applicable at high temperatures over 200 °C. The thermal
stabilities of the ethylene-bridged polysilsesquioxane films in air
were evaluated by TG-DTA, and the results are summarized in Table . The MSQ film showed
high thermal stability, with 5% weight loss temperature (Td5) and 10% weight loss temperature (Td10) of 436 and 481 °C, respectively.
For EBSQ-1, EBSQ-2, and EBSQ-3, strong exothermic peaks were observed
at ∼300 °C (Figure S3), which
prevented accurately determining their Td5 and Td10 values;
however, the thermal stabilities of EBSQ-1, EBSQ-2, and EBSQ-3 were
apparently lower than that of the MSQ film, decreasing with increasing
the amount of SiCCSi units, as shown in Table . This is due to the degradation of the organic
network, which is similar to periodic mesoporous organosilicas.[46] Unfortunately, it was concluded that the SiCCSi
ethylene bridge decreased the thermal stability in the air condition.Compared with a gel film of ethylene-bridged polysilsesquioxane
prepared by heating at 80 °C for a week, the thermal stability
of EBSQ-1 was quite higher than that of a previously reported gel
film.[37] In the TG curve of the gel film
prepared by the sol–gel method, a rapid weight loss was observed
at 240–300 °C due to the decomposition of the SiCCSi bonds
and the ethoxy groups. Interestingly, the decomposition temperature
of EBSQ-1 shifted to ∼300 °C because EBSQ-1 has no ethoxy
groups. Compared with the organic polymer, EBSQ-1 showed high thermal
stability.[45] The ethylene-bridged polysilsesquioxane
film prepared by the hydrosilylation reaction exhibited high thermal
stability, and cracks were not observed after heating at 200 °C
for at least 5 h. These results demonstrate the feasibility of our
synthetic route to ethylene-bridged polysilsesquioxane films for the
preparation of thermal insulating materials.
Conclusions
We have developed a synthetic route for the preparation of ethylene-bridged
polysilsesquioxane films as thermal insulating materials, which consists
of the hydrosilylation reaction of the oligomeric precursors MSQ-SiH
and MSQ-SiVi in the presence of Karstedt’s catalyst. MSQ-SiH
and MSQ-SiVi were prepared by the capping reaction of MSQ-OH, which
was in turn obtained by the sol–gel reaction of triethoxymethylsilane
with chlorodimethylsilane and chlorodimethylvinylsilane. 1H and 29Si NMR and GPC analyses revealed that MSQ-SiH
and MSQ-SiVi were linear and/or cyclic oligomers having hydrosilyl
and vinyl groups, respectively.The hydrosilylation reaction
of MSQ-SiH and MSQ-SiVi proceeded
quantitatively, according to the disappearance of hydrosilyl and vinyl
groups as confirmed by ATR-FTIR. The resulting ethylene-bridged polysilsesquioxane
film was uniform and free of cracks, as revealed by a cross-sectional
FE-SEM image.The ethylene-bridged polysilsesquioxane film EBSQ-1
showed lower
density (1.13 g/cm3) and thermal diffusivity (1.15 ×
10–7 m2/s) than the polymethylsilsesquioxane
film MSQ, whose density and thermal diffusivity were 1.34 g/cm3 and 1.36 × 10–7 m2/s, respectively.
The density and thermal diffusivity of ethylene-bridged polysilsesquioxane
films with different contents in SiCCSi units was found to decrease
by increasing the amount of SiCCSi, which proved that the thermal
insulating property of the films was enhanced by introducing the SiCCSi
ethylene-bridged structure. Although the ethylene-bridged structure
decreased the thermal stability, EBSQ exhibited high thermal stability,
with Td5 and Td10 exceeding 280 °C.
Experimental
Section
Materials
Triethoxymethylsilane, chlorodimethylsilane,
and chlorodimethylvinylsilane were purchased from Tokyo Chemical Industry
Co., Ltd. (Tokyo, Japan) and used as received. THF, toluene, diethyl
ether, and 6 mol/L hydrochloric acid were purchased from FUJIFILM
Wako Pure Chemical Co., Ltd. (Osaka, Japan). THF and toluene were
distilled over calcium hydride before use. Diethyl ether and 6 mol/L
hydrochloric acid (HCl) were used as received without further purification.
Platinum(0)-1,3-divinyl-1,1,3,3-tetramethyldisiloxane (Karstedt’s
catalyst) was purchased from Sigma-Aldrich and used as received. Water
was purified by Millipore Mill-Q UV system, which had a resistance
of 18.2 MΩ·cm and a total organic carbon content of <10
ppb.
Measurements
1H NMR (400 MHz) measurements
were performed on a Varian 400 MHz spectrometer, and 29Si NMR (100 MHz) measurements were performed on a Varian 500 MHz
spectrometer. The samples were dissolved in deuterium chloroform (CDCl3), and residual chloroform was used as an internal standard
(7.26 ppm) for chemical shift reference in 1H NMR. In 29Si NMR, tetramethylsilane was used as an internal standard
(0 ppm). GPC was performed on Shimadzu LC-20AD system equipped with
an RID-10A detector and triply connected TSKgel G6000H/G4000H/G2000H
columns. THF was used as an eluent at 40 °C at a flow rate of
1 mL/min, and polystyrene standards were used for calibration. FTIR
spectra were recorded on a Shimazu IR Affinity-1 spectrometer equipped
with an ATR unit. Thermogravimetric analysis (TGA) was performed on
an SII EXSTAR TG-DTA6200 thermal analyzer at a heating rate of 10
°C/min under an air flow of 100 mL/min. FE-SEM was performed on a Hitachi S-4800 field emission
scanning electron microscope. The samples were attached to the holder
with carbon tape, and the surface was coated with platinum. Thermal
diffusivity measurements were performed on ai-Phase Mobile M3 type
1 (ai-Phase Co., Ltd., Japan). The measurements were repeated at least
three times, and the averaged thermal diffusivity was calculated.
The density of the film was measured using a helium gas pycnometer
on AccuPyc 1340 Series Pyconometers (Micromeritics, U.S.).
Synthesis
of MSQ-OH
Triethoxymethylsilane (10.698 g,
60 mmol) and THF (24.33 mL, 300 mmol) were added to a 100 mL flask
and stirred in an ice bath for 10 min. Then, 6 mol/L HCl (2.188 g,
12 mmol) and water (14.468 g, 900 mmol) were added dropwise over 5
min at 0 °C, and the mixture was stirred at this temperature
for 3 h. The reaction mixture was extracted with 30 mL of diethyl
ether and washed with 30 mL of water three times. The organic layer
was dried over anhydride magnesium sulfate for 10 min and evaporated
under reduced pressure. Finally, the residue was dried under vacuum
for 1 h at room temperature to prevent the gel formation, and MSQ-OH
(4.492 g) was obtained as a highly sticky product.
Synthesis of
MSQ-SiH by the Capping Reaction of MSQ-OH with
Chlorodimethylsilane
In a 100 mL two-necked flask, 4.215
g of MSQ-OH was dissolved in 10 mL of THF and stirred at 0 °C
for 10 min. Chlorodimethylsilane (33.33 mL, 300 mmol) was added dropwise
over 1 h at 0 °C. After being allowed to warm to room temperature,
the reaction mixture was stirred overnight. The reaction mixture was
then evaporated under vacuum for several hours at room temperature
to remove THF and unreacted chlorodimethylsilane. The residue was
passed through a poly(vinylidene difluoride) (PVDF) membrane filter
with a pore size of 0.45 μm, and MSQ-SiH (4.974 g) was obtained
as a colorless liquid.
Synthesis of MSQ-SiVi by the Capping Reaction
of MSQ-OH with
Chlorodimethylvinylsilane
MSQ-SiVi was synthesized as described
above for MSQ-SiH except using 4.056 g of MSQ-OH and 40.5 mL of chlorodimethylvinylsilane
(300 mmol) as precursors. MSQ-SiVi (5.264 g) was obtained as a colorless
liquid.In a typical preparation,
0.21 g of MSQ-SiH and 0.19 g of MSQ-SiVi in a molar ratio of ∼1:1
were mixed by a rotation/revolution vacuum mixer (V-mini300, EME corporation)
for 10 min at 1600 rpm. Then, a 5 wt % Karstedt’s catalyst
solution (10 times diluted with toluene) was added to the mixture,
followed by mixing again with the rotation/revolution vacuum mixer
for 10 min at 1600 rpm. For the preparation of the film, the mixture
was poured into a PFA vial and heated at 80, 120, and 140 °C
for 1 h at each temperature. Then, the mixture was heated at 160 °C
for 4 h and was removed from the PFA vial. Finally, a freestanding
film with a thickness of ∼700 μm was obtained.
Authors: Rosaria Ciriminna; Alexandra Fidalgo; Valerica Pandarus; François Béland; Laura M Ilharco; Mario Pagliaro Journal: Chem Rev Date: 2013-06-19 Impact factor: 60.622
Authors: Makoto Seino; Wendong Wang; Jennifer E Lofgreen; Daniel P Puzzo; Takao Manabe; Geoffrey A Ozin Journal: J Am Chem Soc Date: 2011-10-26 Impact factor: 15.419