Polymer Degradation and Stability 96 (2011) 2064e2070 Contents lists available at SciVerse ScienceDirect Polymer Degradation and Stability journal homepage: www.elsevier.com/locate/polydegstab Cross-link network of polydimethylsiloxane via addition and condensation (RTV) mechanisms. Part I: Synthesis and thermal properties Mohamad Riduwan Ramli, Muhammad Bisyrul Hafi Othman, Azlan Arifin, Zulkifli Ahmad* School of Materials and Mineral Resources Engineering, Engineering Campus, Universiti Sains Malaysia, Seri Ampangan, 14300 Nibong Tebal, Penang, Malaysia a r t i c l e i n f o a b s t r a c t Article history: Received 24 August 2011 Accepted 2 October 2011 Available online 6 October 2011 A series of highly cross-linked polysiloxane was synthesised via hydrosilylation and condensation reaction. Structural identification using Fourier Transform Infrared (FTIR) and 1H-NMR confirmed their chemical structures. Their thermal and, mechanical properties, and crystallinity, were analysed and related to the level of cross-link density. These systems displayed elevated thermal and hardness properties at an increased cross-link density. Furthermore, the level of crystallinity was reduced as displayed by XRD analysis. Along with this observation, the calculated fractional free volume (FFV) showed a decreasing trend leading to the ‘densification’ effect. It was envisaged that the linear polysiloxane chain segments aligned parallel to each other in a triclinic crystal system to generate a crystalline domain. The spacing between these stacking chains was found to be about 7.2 Å as measured from simulated XRD pattern. Ó 2011 Elsevier Ltd. All rights reserved. Keywords: Polysiloxane Cross-link network Thermal Crystallinity Fractional free volume X-ray diffraction 1. Introduction Cross-linked polydimethylsiloxane (PDMS), or silicone elastomer, is of major interest to researchers because of its unique properties. It is generally produced by synthesising reactive PDMS prepolymers, which are subsequently cross-linked to give a highly tough and durable elastomeric material. Crosslinking or network formation of silicone polymer can be obtained by three typical routes: condensation reaction (moisture cure), addition reaction (hydrosilylation cure), or radical reaction, which is normally performed at higher temperatures [1e3]. Crosslinking of polysiloxane via condensation reaction affording thermally stable materials as been reported recently [4e7]. Han et al. [6] studied the effect of cross-link density towards the thermal stability of room temperature vulcanization PDMS by incorporating it with polymethylmethoxysiloxane (PMOS) via hydrolytic condensation reaction. Highly cross-linked PMOS phases were formed in situ and the average cross-link density increased as the loading of PMOS was increased. They suggested that dense PMOS phases could reduce the pyrolysis of PDMS at elevated temperature. Chen et al. [7] investigated the thermal degradation, thermooxidative stability and mechanical properties of hybrid PDMS with an octavinyl-Polyhedral oligomeric silsesquioxanes (POSS) * Corresponding author. Tel.: þ604 599 5099; fax: þ604 594 1011. E-mail addresses: riduwanramli85@yahoo.com (M.R. Ramli), zulkifli@eng.usm. my (Z. Ahmad). 0141-3910/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymdegradstab.2011.10.001 derivative prepared by hydrolytic condensation. They reported that the improvement in thermal properties could be attributed to the effective three-dimensional network structures resulting from the structure of the octavinyl-POSS derivative. Furthermore, the improvement of mechanical properties could be attributed to the synergistic effect of the structure of three-dimensional multi-arm cross-linker vinyl-POSS derivative and the perfect distribution of the vinyl-POSS derivative. Meanwhile the addition reaction of polysiloxane provides a more stable material at elevated temperatures, and was developed for rapid processing and fast curing rate of deep section part [8,9]. The two-part heat curable system, which consists of the vinyl and hydride reactive functional groups, and the addition of a platinum catalyst hydrosilylation cure, provides a fast cure system without any by-product. Furthermore, a longer shelf time, one-component system was developed by premixing a scavenging agent such as oximatosilanes, carbamatosilane or aminosilanes [10]. These scavenging agents react with excess hydroxyl groups whether from methanol, silanol, or water, and would not react with the alkoxy groups to prematurely cross-link the PDMS system. A high flame retardancy effect could be accomplished by a cross-linked structure performed under platinum catalysed hydrosilylation reaction as suggested by Hayashida et al. [11] by suppressing the thermal decomposition. The condensation reaction of PDMS however, requires a longer curing time [12,13]. On the other hand, the addition reaction of PDMS in a one-component system requires less mixing and provide a problem-free production material but requires an elevated curing temperature [1,3]. M.R. Ramli et al. / Polymer Degradation and Stability 96 (2011) 2064e2070 2065 Table 1 Feed molar ratio of hydride terminated PDMS which have been synthesised. a b Mw/Mn cIntrinsic Molar bMn Mw Ratio, r (g/mol) (g/mol) Viscosities Monomer I Monomer II (dL/g) (EC) (D4) Samples Concentration (mmol) A B C D a b c 67.6 67.6 67.6 67.6 20.0 10.1 5.4 2.7 0.30 0.15 0.08 0.04 3858 4697 1.22 5866 8206 1.40 11,104 14,838 1.34 18,971 28,672 1.51 0.0122 0.0404 0.0673 0.1495 Calculated based on molar ratio monomer I and monomer II, r ¼ [EC/D4]. Measured using gel permeation chromatography (GPC). Obtained from Ubbelohde solution viscometer using toluene as the solvent. This work attempted to incorporate both mechanisms to yield a novel elastomeric product having synergistic thermal and mechanical properties. It comprised of investigation into the thermal and physical properties of cured PDMS at varying crosslink densities. The degree of cross-link density was designed based on a series of linear prepolymers prepared as the starting materials. The effect of cross-link density on the level of crystallinity was evaluated and the simulated crystal structure of the polysiloxane was generated based on the XRD pattern. 2. Experimental 2.1. Materials Octamethylcyclotetrasiloxane (D4), trifluoromethane sulfonic acid (triflic acid), Platinum(0)-1,3-divinyl-1,1,3,3-tetramethyldisiloxane complex (Karstedt’s catalyst) and vinyltrimethoxysilane 98% were obtained from ALDRICH. 1,1,3,3-tetramethyldisiloxane was purchased from Fluka. The solvents such as the toluene and diethyl ether were obtained from J.T. Baker and Merck respectively. All materials were used as received. 2.2. Synthesis of a,u-dihydrido-polydimethylsiloxane (PDMS-1) Octamethylcyclotetrasiloxane, I (D4) (20 g) and 1,1,3,3tetramethyldisiloxane II (2.7 g, 20.0 mmol) were charged into a round bottom flask and purged with nitrogen gas. The temperature was increased to 55  C and triflic acid (0.2 g) was slowly added into the reaction. The reaction was equilibrated at 55  C for 72 h. The mixture was dissolved in diethyl ether then neutralized by repeatedly being washed with deionized water. The solution was dried over anhydrous magnesium sulphate for 1 h and subsequently filtered through glass wool. Then, diethyl ether and unreacted monomer were distilled out under reduced pressure at 80  C for 2 h. A clear liquid form was obtained with over 90 wt% yield. Four pre-polymer with different molar ratio of I and II were synthesised (Table 1). 1H-NMR (CDCl3): d ¼ 4.7 (m, 2H, SiH); d ¼ 0.1e0.05 (m, 6H, SiCH3). IR y: 2962, 2907, 2127, 1412, 1258, 1079, 1011, 911, 864, 788, 699 cm 1. Fig. 1. Repeat unit for polysiloxane network used to determine the FFV. The n values for samples RTV-A ¼ 51, RTV-B ¼ 78, RTV-C ¼ 149, RTV-D ¼ 256. 2.3. Synthesis of a,u-bis-(trimethoxysilane)-polydimethylsiloxane (PDMS-2) Vinyltrimethoxysilane IV (4.58 ml, 30 mmol), toluene (10 ml) and Karstedt’s catalyst (0.075 ml) were charged into a 50 ml twoneck round bottom flask equipped with a dropping funnel. PDMS-1 III (10 g) was charged into the dropping funnel and the temperature was increased to 50  C. PDMS-1 was slowly dropped into the reaction mixture, which was maintained at a temperature of 50  C. The temperature was then increased to 75  C for 2 h. After completion of the reaction, the solution was cooled to room temperature and the remaining unreacted monomers were removed using rotary evaporator at 90  C for 3 h to give a clear liquid form with over 90 wt% yield. 1H-NMR (CDCl3): d ¼ 3.5 (s, 18H, eSi(OCH3)3); 1.1 (m, 4H, (CH3O)3SiCH2e); 0.6 (d, 4H, e(CH3)2SiCH2e); 0.1e0.05 (6H, eSi(CH3)2e). IR y: 2962, 2907, 2839, 1412, 1258, 1191, 1079, 1013, 788, 699 cm 1. 2.4. Preparation of highly cross-linked PDMS Room temperature vulcanization PDMS (RTV-PDMS) was prepared via hydrolysis followed by condensation reaction of methoxy groups using triflic acid as catalyst (0.3 phr) under ambient conditions for 24 h. The alcohol that was produced as a side product was eliminated from the reaction system using Dean-Stark apparatus. For comparison, a control sample was provided by Penchem Industry Sdn. Bhd. The control was PDMS system cured with the platinum catalyst at 120  C for 2 h. IR y: 2962, 2907, 1412, 1258, 1079, 1013, 788, 699 cm 1. 2.5. Analytical techniques 1 H-NMR spectra were recorded with a Bruker 400 UltraShield linked to a computer running WIN-NMR software. The frequency used was 400 MHz for 1H. Experiments were performed in CDCl3 with tetramethylsilane (TMS) as an internal reference. Infrared spectra were recorded using a PerkineElmer System 2000 Attenuated Total Reflectance (ATR-FTIR) spectroscopy. The spectra were recorded in 4000e500 cm 1 region using 16 scans. Table 2 Some physical properties of RTV samples. Samples Molar Ratio, r a Density, re (g/cm3) Hardness (Shore A) Optimum Q RTV-A RTV-B RTV-C RTV-D 0.30 0.15 0.08 0.04 0.985 0.982 0.975 0.971 67 63 60 59 1.99 2.79 3.44 4.08 a Were taken using pycnometer. Fig. 2. Synthetic route of highly cross-linked RTV-PDMS. 2066 M.R. Ramli et al. / Polymer Degradation and Stability 96 (2011) 2064e2070 Fig. 3. Schematic structure of three-dimensional cross-linked PDMS produced. The swelling test was carried out in toluene and the weights of swollen samples were measured after being blotted with filter paper to remove the excess toluene. The optimum weights of swollen samples were obtained after 24 h and the degree of swelling (Q) was calculated using Eq. (1) following Mazan et al. [14] works, Q ¼ re m2 m1 þ1 rs m1   (1) where, Q is the volume swollen ratio, m1 is the mass of dry sample after elution in toluene, m2 is the mass of swollen sample, rs is the density of solvent which was taken as 0.87 g/cm3 for toluene and re is the density of samples which can be seen in Table 2. The thermal analyses of samples were performed on a DSC 1 Mettler Toledo apparatus with 40 mL aluminium crucible running Mettler Toledo software. The experiments were run from 140 to 100  C with a ramp of 10  C/min for the first heating. Then the rate was fixed at 20  C/min during cooling and second heating. The samples were equilibrated for 5 min at each turning temperature. Thermal degradation and thermo-oxidative stability were performed using a PerkineElmer 7 thermo-analyser. The samples were heated in aluminium crucibles from 30 to 800  C at a heating rate of 20  C/min in an air conditioned environment. The X-ray diffraction (XRD) was carried out on D8 advance Bragg-Brentano configuration from Bruker with Cu Ka ¼ 1.54056 Å, and the slit size and scan rate were 0.01 and 10 /min, respectively. Densities of samples were obtained using XB220A Precisa pycnometer in deionized water and the hardness was measured using Shore A. Meanwhile fractional free volume (FFV) was calculated using van Krevelen group contribution method [15] based on repeat unit of polymer networks as shown in Fig. 1. 3. Results and discussions 3.1. Synthesis of highly cross-linked RTV-PDMS RTV-PDMS was prepared following a two-steps synthesis. Firstly, the preparation of III through ring opening polymerization [16], and followed by hydrosilylation with IV [17] to obtain V. Subsequently, the cross-link networks of V were generated via condensation process of methoxy groups as depicted by Fig. 2. In the first step, cationic ring opening polymerization of I was carried out with an acidic catalyst. The acidic catalyst was preferred Table 3 1 H-NMR chemical shifts for both PDMS-1 and PDMS-2. Fig. 4. Transmittance ATR-FTIR spectra of a) PDMS-1, b) PDMS-2, c) RTV-PDMS. Abbreviation Species Chemical shift (ppm) a b, g c, h d e f ^SieH ^SieCH3 eO2SieCH3 ^SieOCH3 (CH3O)3SieCH2CH2e eO(CH3)2SieCH2CH2e 4.7 0.1e0.05 0.1e0.05 3.6 1.1 0.6 M.R. Ramli et al. / Polymer Degradation and Stability 96 (2011) 2064e2070 2067 Fig. 5. 1H-NMR spectrum of PDMS-1 (above) and PDMS-2 (below). over a basic catalyst due to the sensitivity of hydride (SieH) reactive groups to the latter which might lead to O-silylation reaction [18]. Siloxane cation was first generated in this system by the reactive triflic acid which propagates with the other cyclic siloxanes. The propagation of linear PDMS-I was then terminated by II to obtain a hydride terminal group. The molecular weight of this linear PDMS chain was tailored by controlling the mole ratio of the 1,3tetramethyldisiloxane as the end-capper. The molecular weight for this series is shown as in Table 1. Some physical properties are shown as in Table 2. The hydrosilylation process that occurred in the second step involves the hydrogen of the hydrosilyl group and the vinyl group from the trimethoxy silane in the presence of Karstedt’s catalyst; this introduced the tri-methoxysilane group to the terminal of the chain. The condensation step was finally performed with the active methoxy groups reacting under atmospheric moisture in the presence of 0.3 phr of triflic acid. During the first two hours of the reaction, the methanol-like smell was detected suggesting that the condensation and hydrolysis of the methoxy groups took place. Three methoxy groups attached at the terminal end of PDMS-2 affording a three dimensionally cross-linked PDMS structure (Fig. 3). The disappearance of methoxy groups represented by CH3eO and SieOeCH3 peaks at 2839 cm 1 and 1190 cm 1 respectively unambiguously proves the formation of crosslinking networks as shown in Fig. 4c. 3.3. Nuclear magnetic resonance NMR analysis for PDMS-1 and PDMS-2 are given as in Table 3. In PDMS-1 spectrum, the proton from SieH groups occurred at d 4.7 ppm (assigned as a in Fig. 5) meanwhile siloxane groups (proton from methyl attached to siloxane groups) represented by b and c were assigned mostly upfield at (d 0.1ed 0.05 ppm) respectively. The chemical shifts for SieH and SieCH3 were consistent with the previous study [17,23]. The integration ratios of these two peaks were taken to determine the value of repeating unit n. In PDMS-2 spectrum, methoxy proton peak occurred mostly downfield at d 3.6 ppm as a sharp singlet represented by d (see 3.2. FTIR spectroscopy Complete formation of SieH bonds in PDMS-I was been observed after purification based on FTIR spectrum. Fig. 4a shows the FTIR spectrum of SieH terminated PDMS with two distinctive bands at 2127 cm 1 and 910 cm 1 stretching represented by A and B respectively [19e21]. These bands can be used as the preliminary evidence of the appearance of hydride groups in III. The disappearance of SieH peaks at 2127 cm 1 and 910 cm 1 indicate that the hydrosilylation process was accomplished in the system (Fig. 4b). This is reiterated further by the appearance of symmetrical CH3eO stretching and SieOeCH3 peaks at 2839 cm 1 and 1190 cm 1 respectively. The peak at 2840 cm 1 (sharp) and 1190 cm 1 can be assigned as SieOCH3 groups as has been reported elsewhere [21,22]. Fig. 6. The swelling ratio of RTV in toluene at 25  C. 2068 M.R. Ramli et al. / Polymer Degradation and Stability 96 (2011) 2064e2070 Fig. 7. Effect of molar ratio (end capper to monomer ratio) on cross-link density and degree of swelling of RTV samples. Table 3) due to the presence of neighbouring electronegative oxygen. The methylene proton e and f occur at d 1.1 and 0.6 ppm respectively. For e, the silicon atom bonded to three methoxy groups which contain three oxygen atoms. The chemical shift therefore, is slightly downfield compared to f of which the adjacent silicone atom is attached to two methyl groups and one oxygen atom. Comparison between spectra PDMS-1 and PDMS-2 shows that peak a in PDMS-1 completely disappeared in the PDMS-2 spectrum suggesting a complete hydrosilylation process had occurred. The appearance of peak d at d 3.6 ppm indicates the successful incorporation of methoxy groups in the PDMS-2 structure. The formation of methylene at peak e and f in PDMS-2 fully support the complete hydrosilylation process. 3.4. Cross-link density Cross-link density of samples were obtained after curing process at room temperature and were determined using Flory-Rehner’s Eq. (2), Fig. 9. TGA curves for RTV samples compared to control sample measured in air atmosphere condition. n¼ h lnð1 V1 V2 Þ þ V2 þ xps V22 1= V2 3 V2 2 ! i (2) where, V1 is molar volume of solvent (106.85 cm3/mol was taken for toluene [24]), V2 is reciprocal of volume swelling ratio Q, xps is polymeresolvent interaction coefficient and was taken as 0.465 [24] for siloxane elastomer and toluene interaction. In Fig. 6, the optimum degrees of swelling for RTV-B, RTV-C, and RTV-D were achieved after 23 h of swelling test. RTV-A attained its optimum degree of swelling after only 6 h of swelling test. Degree of swelling is inversely proportional to the cross-link density since at higher cross-link density, solvent diffusion in the polymer matrix would be less efficient. The order of cross-link density would be RTV-A > RTV-B > RTV-C > RTV-D. It can be seen that, the cross-link density of RTV samples increased correspondingly with the increase of molar ratio of endcapper to the monomer as depicted in Fig. 7. At high molar ratio of end-capper, the molecular weights of the chain will decrease and subsequently provide a higher bulk concentration for the condensation reaction site with the silane cross-linker. The highest crosslink density was obtained at 0.30 ratio while the lowest value is at 0.04 ratio. The end-capper was subsequently replaced by three active methoxy groups during the hydrosilylation process. There are six active methoxy groups per molecule, (i.e. three at each endchain of the molecule), and it was these groups which formed the cross-link network through the condensation reaction between chains. The effect of cross-link densities on hardness were shown in Table 2. The hardness of samples decreased with decreasing cross- Table 4 Effect of cross-link density on free volume and glass transition temperature. Fig. 8. Glass transition temperature (2nd scan) of RTV-PDMS samples run at 20 min.  C/ Samples Cross-link density (mol/cm3) RTV-A RTV-B RTV-C RTV-D 13.48 4.52 2.43 1.51 a     10 10 10 10 4 4 4 4 Calculated using Krevelen method [15]. a Fractional free volume (FFV) 0.186 0.187 0.192 0.195 Tg ( C) 108 112 114 119 M.R. Ramli et al. / Polymer Degradation and Stability 96 (2011) 2064e2070 Fig. 10. XRD spectrum for sample RTV-A until RTV-D and the simulated pattern. link densities. Hardness values of samples produced in this study were superior to the commercial RTV samples. The commercial hardness was in the range of 22e40 Shore A [12]. 3.5. Thermal properties Differential Scanning Calorimetry (DSC) was performed on RTVPDMS to reveal the thermal transition in the polymer. The transitions in Fig. 8 were attributed to the glass transition (Tg) of RTVPDMS samples in the range of 108 to 119  C. This is in close agreement with early studies [25] which reported the Tg of analogous RTV silicone elastomers cured with POSS structures was between 118 and 120  C. The Tg decreased progressively with the decreasing cross-link density of the samples. Thus, sample RTVA displayed the highest Tg compared to other samples due to chain restriction as the result of a high cross-link density. All samples shows a gradual step drop during the glass transition as due to the plasticizing effect of the linear siloxane chains which adjoined the crosslinking points in the network. Fractional free volume (FFV) of RTV samples were calculated using group contribution of van der Waals’ volume method as established by van Krevelen [15]. The values in the range 0.15e0.19 are typical for polymeric systems. RTV-A contains the highest cross-link density when compared to other three samples, thus, exhibiting the lowest free volume. It was established from Positron 2069 Annihilation Lifetime Study that a high cross-link density induces a lower fractional free volume [26]. This is because a high cross-link density would affect a ‘densification’ between chains. A control from a commercial source was used in determination of thermal stability to compare with other RTV samples synthesised in this work (Fig. 9). It can be seen that, RTV-A starts to degrade at 475  C along with the control sample. Meanwhile RTV-B was degraded at around 450  C followed by RTV-C and RTV-D at around 375  C. Thus, the thermo-oxidative stability of RTV samples increased with the cross-link density. Interestingly, the thermal stability of RTV samples was comparable with the control sample, although the control sample is the addition system, which is known to be thermally stabile. This might be due to the high cross-link network achievable in this system. Further, RTV-A showed the highest char yield. It has been shown that char yield is indirectly proportional to the cross-link density in a polysiloxane system. A linear, uncross-linked polydimethylsiloxane heated in an Ar atmosphere was completely degraded at 600  C with no residue left [27]. The decreasing trend of Tg (Table 4) was further investigated by studying the X-ray diffraction (XRD) spectra. Essentially the XRD spectra (Fig. 10) show broad halos representing the highly amorphous nature of all the samples. It can be seen however, that the RTV-A shows broad peaks while RTV-D shows the sharpest peaks. This observation shows that the highly cross-linked sample of RTVA is less crystalline than the low cross-linked sample of RTV-D. This can be explained that as the RTV-A has more cross-link network, chain flexibility is immobilised to certain degree which restrict any possible chain alignment to form crystalline domains. Alternatively, as the cross-link densities decreased, the molecular chains have adequate space to align and rearrange into crystalline domains thus, giving sharper peaks in XRD. However, the peak’s positions do not vary much between samples with the change in the cross-link densities implying that crystal system remains intact. The XRD analysis, therefore, corroborate further the observed degree of cross-link density as discuss in the preceding sections. Simulated crystal structures based on the experimental XRD pattern were generated using Material Studio v4.4 software using Forcite module whose energy minimisation was perform under Dreiding Forcefield. One chain per unit cell was generated and minimised in an initially constrained triclinic crystal system. Towards the end of the simulation cycle, this unit cell was relaxed and no apparent change was detected in either the crystal lattice or the chain configuration. Thus, it was established that the obtained unit cell configuration should be correct. Fig. 11 shows the simulated diffraction pattern as compared to the experimental data. The laterally adjacent chain distance was found to be 7.2 Å. These values Fig. 11. Simulated crystal structure of crystallite domain of the cured RTV material showing the stereo projection (a) and laterally (b) adjacent chain in a unit cell. 2070 M.R. Ramli et al. / Polymer Degradation and Stability 96 (2011) 2064e2070 corresponded very well to the main peak positions at 2q ¼ 12.0 from the experimental data. This distance represents the d-spacing of the linear polysiloxane chain backbone. It is envisaged that the crystalline domain is formed between the linear polysiloxane chains which are at close proximity and undergoes chain arrangement. The polysiloxane region resulting from condensation is unlikely to form any crystalline structure as expected from a highly cross-link network. 4. Conclusion Synthesis of RTV siloxane by a combination of addition and condensation systems has been presented in this paper. Structure identifications were established using FTIR and H-NMR methods. The results showed significant improvement of RTV samples compared with a high temperature vulcanization (control) sample. The greatest contribution to this improvement was from high content of cross-link density and synergistically having combined thermal properties from both the addition and condensation system. The swelling test showed that the lower the chain length of the linear polysiloxane segment, the higher the cross-link density. A high cross-link network affects an improvement in hardness and thermal properties. The free volume is also reduced leading to the effect of ‘densification’ of polymeric chains in the bulk system. The high cross-link density of this system, however, reduced the level of crystallinity, which owed much to the chain immobilisation. Simulated crystalline structure based from XRD analysis established that the crystalline domain was generated from the parallel alignment of linear polysiloxane backbone in triclinic crystal system whose inter-chain spacing is about 7.2 Å. [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] Acknowledgement [20] The authors thank Universiti Sains Malaysia and the USM Fellowship Scheme for the financial support of this work by Research University Grant (No. 814069). [21] References [23] [1] Brook M. Silicon in organic, organometallic, and polymer chemistry. J. Wiley; 2000. [2] Gonzaga F, Yu G, Brook M. Polysiloxane elastomers via room temperature, metal-free click chemistry. Macromolecules 2009;42(23):9220e4. [3] de Buyl F. Silicone sealants and structural adhesives* 1. Int J Adhes Adhes 2001;21(5):411e22. [4] Prado LASA, Sforça ML, de Oliveira AG, Yoshida IVP. Poly(dimethylsiloxane) networks modified with poly(phenylsilsesquioxane)s: synthesis, structural characterisation and evaluation of the thermal stability and gas permeability. Eur Polym J 2008;44(10):3080e6. [5] Chen D, Yi S, Wu W, Zhong Y, Liao J, Huang C, et al. Synthesis and characterization of novel room temperature vulcanized (RTV) silicone rubbers using [22] [24] [25] [26] [27] Vinyl-POSS derivatives as cross linking agents. Polymer 2010;51(17): 3867e78. Han Y, Zhang J, Shi L, Qi S, Cheng J, Jin R. Improvement of thermal resistance of polydimethylsiloxanes with polymethylmethoxysiloxane as crosslinker. Polym Degrad Stab 2008;93(1):242e51. Chen D, Nie J, Yi S, Wu W, Zhong Y, Liao J, et al. Thermal behaviour and mechanical properties of novel RTV silicone rubbers using divinyl-hexa [(trimethoxysilyl)ethyl]-POSS as cross-linker. Polym Degrad Stab 2010; 95(4):618e26. Cai D, Neyer A. Cost-effective and reliable sealing method for PDMS (PolyDiMethylSiloxane)-based microfluidic devices with various substrates. Microfluid Nanofluid; 2010:1e10. Parbhoo B, O’Hare LA, Leadley SR. Fundamental aspects of adhesion technology in silicones. In: Chaudhury M, Pocius AV, editors. Surfaces, chemistry and applications. Amsterdam: Elsevier Science B.V.; 2002. p. 677e709. White MA, Beers, Melvin D, Lucas, Gary M, Smith, Robert A, Swiger, Roger T. One package, stable, moisture curable, polyalkoxy-terminated organopolysiloxane compositions and method for making. United States: General Electric Company (Waterford, NY); 1983. Hayashida K, Tsuge S, Ohtani H. Flame retardant mechanism of polydimethylsiloxane material containing platinum compound studied by analytical pyrolysis techniques and alkaline hydrolysis gas chromatography. Polymer 2003;44(19):5611e6. Jerschow P. Silicone elastomers. Rapra Technology Ltd; 2002. Reilly B, Bruner S. Silicones as a material of choice for drug delivery applications; 2004. Mazan J, Leclerc B, Galandrin N, Couarraze G. Diffusion of free polydimethylsiloxane chains in polydimethylsiloxane elastomer networks. Eur Polym J 1995;31(8):803e7. Van Krevelen DW, Te Nijenhuis K. Volumetric properties. Properties of polymers. 4th ed. Amsterdam: Elsevier; 2009. pp. 71e108. Desmurs J, Ghosez L, Martins J, Deforth T, Mignani G. Bis (trifluoromethane) sulfonimide initiated ring-opening polymerization of octamethylcyclotetrasiloxane. J Organomet Chem 2002;646(1e2):171e8. Cai G, Weber W. Synthesis of terminal Si-H irregular tetra-branched star polysiloxanes. Pt-catalyzed hydrosilylation with unsaturated epoxides. Polysiloxane films by photo-acid catalyzed cross-linking. Polymer 2004;45(9): 2941e8. Pawlenko S. Organosilicon chemistry. W. de Gruyter; 1986. Cai D, Neyer A, Kuckuk R, Heise HM. Raman, mid-infrared, near-infrared and ultraviolet-visible spectroscopy of PDMS silicone rubber for characterization of polymer optical waveguide materials. J Mol Struct 2010; 976(1e3):274e81. Satoh T, Hiraki A. Detailed study of SieH stretching modes in mc-Si: H film through second derivative IR spectra. Jpn J Appl Phys 2 1985;24(7):L491e4. Smith A. Infrared spectra-structure correlations for organosilicon compounds. Spectrochim Acta 1960;16(1e2):87e105. Crompton T. Analysis of organosilicon compounds. The Chemistry of organic silicon compounds. New York: Wiley; 1989. pp. 393e444. Louis E, Jussofie I, Kühn F, Herrmann W. Karstedt catalyst-catalyzed stepgrowth co-polyaddition of 1,9-decadiene and 1,1,3,3-tetramethyldisiloxane. J Organomet Chem 2006;691(9):2031e6. Hauser RL, Walker C, Kilbourne FL. Swelling of silicone elastomers. Ind Eng Chem 1956;48(7):1202e8. Chen D, Yi S, Fang P, Zhong Y, Huang C, Wu X. Synthesis and characterization of novel room temperature vulcanized (RTV) silicone rubbers using octa [(trimethoxysilyl)ethyl]-POSS as cross-linker. React Funct Polym 2011;71(4): 502e11. Xu J, Chen B, Zhang Q, Guo B. Prediction of refractive indices of linear polymers by a four-descriptor QSPR model. Polymer 2004;45(26):8651e9. Schiavon M, Redondo S, Pina S, Yoshida I. Investigation on kinetics of thermal decomposition in polysiloxane networks used as precursors of silicon oxycarbide glasses. J Non-Cryst Solids 2002;304(1e3):92e100.