Synergistic interactions between POSS and fumed silica and their effect on the properties of crosslinked PDMS nanocomposite membranes

Abstract

A facile strategy for the synthesis of binary filler nanocomposite membranes containing fumed silica (FS) and octatrimethylsiloxy polyhedral oligomeric silsesquioxane (POSS) nanoparticles was proposed to prepare high performance PDMS–FS–POSS nanocomposite membranes. To fully explore the synergistic effect between the POSS and FS nanoparticles, thermal stability using thermo-gravimetric analysis (TGA) and dispersion quality using scanning electron microscopy (SEM) were investigated, while the crosslinked network was studied using Fourier transformed infrared spectroscopy (FTIR). The results showed that the thermal stability of these novel nanocomposite membranes was much better than that of the neat membrane. Thermodynamically, dipole–dipole interactions between the functional groups are the main parameter leading to better dispersion and thermal stability. Furthermore, it was found that the separation properties of different gases (H₂, C₃H₈, CO₂ and CH₄) across the nanocomposite membranes were enhanced with increasing FS content. All the improvements observed can be attributed to the synergistic interactions between FS and POSS.

---

1. Introduction

Separation of higher hydrocarbons (C₃⁺) from a light gas mixture traditionally requires low temperatures and expensive processes like cryogenic distillation. On the other hand, membrane separation has recently been proposed for the effective recovery of valuable hydrocarbons. The polymeric membranes for this intended application should be organic vapor selective materials such as poly(dimethylsiloxane) (PDMS) or poly(1-trimethylsilyl-1-propyne) (PTMSP). These solubility selective polymers have a strong affinity with hydrocarbons enabling the separation of higher hydrocarbons from a gas mixture.

Recently, progress has been made on the preparation of PDMS nanocomposite membranes. PDMS membranes were selected for their excellent thermo-oxidative stability, high flexibility, low glass transition temperature (T_g), low chemical reactivity, and low surface free energy. Although neat PDMS has excellent chain flexibility, good resistance to thermal degradation and thermo-oxidation in a wide temperature range, it still shows weak performance and cannot satisfy all the requirements for industrial application (service conditions and performance).¹ To overcome these challenges, various inorganic additives have been incorporated into the polymer structure through polymerization or physical blending. It is well recognized that fumed silica (FS) is a potential candidate and has been extensively used to improve the characteristics of PDMS membranes in order to meet the requirements, especially in the rubber industry.^2–4

Recent investigations tried to combine the benefits of FS and polyhedral oligomeric silsesquioxane (POSS) to prepare novel silicone based nanocomposite membranes with improved gas separation and thermal degradation resistance. To date, only the effects of POSS nanoparticles on PTMSP/FS nanocomposite membranes have been investigated and Table 1 describes the most important findings from these studies.

Table 1 Main findings related to binary FS–POSS nanocomposite membranes^a

Membrane		Major findings	Ref.
a PTMSP: poly [1-(trimethylsilyl)-1-propyne]; POSS OL1160: octavinyl-POSS; POSS MS0865: octatrimethylsiloxy-POSS.
PTMSP	POSS OL1160	The incorporation of 10 wt% POSS decreased the membrane permeability by 70%. The permeation was stable during the experiment (up to 27 days). This indicated that physical aging was clearly exhibited by incorporating POSS nanoparticles	28
PTMSP	POSS MS0865	The incorporation of 10 wt% POSS decreased the membrane permeability by 55%, while the FFV and permeability were stable over time for the PTMSP nanocomposite membranes having 10 wt% POSS nanoparticles. POSS props open the larger free volume elements of PTMSP, mitigating the physical aging by avoiding the collapse of larger free volume elements	29 and 30
PTMSP	POSS OL1160	FS and POSS nanoparticles have opposite effects on membrane permeability: FS increases while POSS dramatically decreases the permeability. The gas permeability of the binary FS/POSS filler system appears to be an average of the permeabilities of the respective single-filler systems. Incorporation of POSS filler appears to stop aging in PTMSP and stabilizes the permeability	31

---

PDMS, POSS and FS have similar chemical structures. POSS has a unique cage-like structure with a chemical composition of (RSiO_1.5)_n (n = 6, 8, 10, …), where R is an organic functional group, and owing to the nanosize scale of POSS and its properties, it has attracted interest in membrane studies. The most appropriate POSS for membranes is most certainly octasilsesquioxane (n = 8) which has a cube-shaped Si₈O₁₂ cage with organic groups R at each corner. POSS nanoparticles mainly have a siloxane cage which can be tethered with a wide range of surface functional groups radiating away from the cage. The substitution of the eight trimethylsiloxy [–OSi(CH₃)₃] groups on the POSS surface was found to be appropriate for our research.^5,6

Until now, various types of POSS nanoparticles have been incorporated in different polymers to prepare hybrid membranes and various approaches were proposed to introduce POSS nanoparticles into the polymers: physical blending, covalent bonding with the polymer structure, and incorporation as pendent groups on the polymer matrix.^7–10 The introduction of POSS nanoparticles into a polymer matrix results in enhanced membrane properties such as processability, thermal and mechanical properties, and membrane operational performance.

The unique properties of PDMS–FS and PDMS–POSS hybrid membranes have been separately investigated in several research works, both in conceptual studies targeted at understanding the physical behavior of polymer chains in close contact with nanoparticle surfaces in response to external stimuli such as pressure, temperature, and chemical environments, as well as in applied industry. On the other hand, to the best of our knowledge, there is no related study regarding the effect of FS on the permeation and sorption properties of silicone nanocomposite membranes with POSS. However, the main purpose of this study is to investigate the effect of FS and POSS content on the resistance to thermal degradation and gas separation performance of novel silicone nanocomposite membranes. Emphasis is made on the relationships between the performance (such as thermal stability and operational performance) and the nanostructure, as well as the dispersion state in the nanocomposites. Furthermore, the focus of the current paper is on providing an account of the fundamental science of the functional properties of binary systems. The incorporation of POSS and FS nanoparticles into a polymer matrix was performed through physical blending. Meanwhile, the effect of POSS and FS nanoparticles on the overall membrane structure was investigated to show how surface interactions between polymer chains and both nanofillers (as influenced by means of the nanofiller surface functional groups and their contents) affect the hybrid membranes’ structures. The synergistic effect of POSS and FS addition on the membrane performance properties such as thermal stability and nanoparticle dispersion was also explored. The 3D crosslinked structure of the PDMS based nanocomposite membranes was characterized using FTIR-ATR. It is believed that the outcomes of this study suggest valuable information and propose a roadmap for the synthesis and gas separation of POSS-related polymeric nanocomposite membranes to meet the practical requirements of commercial gas separation applications.

2. Experimental

2.1. Materials

To prepare PDMS-based nanocomposite membranes, DEHESIVE silicone was supplied from Wacker Silicones Corporation (USA). These materials are commonly processed in the form of a 3-component material. Each component can affect the properties of the PDMS membranes. The main components are:

Prepolymer (DEHESIVE 944). Solvent-based addition crosslinkable PDMS (RTV 615A en B, density 1.02 g mL⁻¹, average M_w 25 000, vinyl content: 0.08%) whose end group contains crosslinkable vinyl groups.

Crosslinker V24. Hydrogen polysiloxanes (polymethylhydrogensiloxane, average M_w 2100; hydrogen content: 1.63%) with high levels of reactive Si–H groups for the thermal curing of PDMS. The extent and nature of the crosslinker specifies the mechanical resistance and adhesion strength of the membrane.

Catalyst OL. Platinum complexes (1,3-divinyl-1,1,1,3-tetramethyldisiloxane platinum (Pt) complex solution) for the thermal hardening of PDMS.

The CAB-O-SIL fumed silica TS-530 (FS) with a characteristic dimension of approximately 13 nm was received from Cabot Corporation Products (USA). It is well-known that FS is naturally hydrophilic due to the Si–OH groups on its surface. Therefore, when directly incorporated into hydrophobic PDMS, FS agglomeration is often encountered. The extent of the aggregation can be large enough to cause defects in the nanocomposites. However, FS has a low compatibility with organic components and its surface modification by silane coupling agents has often been performed to improve its compatibility with/dispersion in the PDMS matrix. Accordingly, the modified surface of FS was selected in this study to improve the nanoparticles’ compatibility with the polymer chains. Hence, TS-530 FS was selected since it has a high surface area and has been surface-modified with hexamethyldisilazane. The modification substitutes hydrophilic hydroxyls on the FS surface with hydrophobic trimethylsilyl surface groups resulting in enhanced hydrophobicity. The compatibility of the treated FS with PDMS allows faster incorporation and better dispersion than its untreated counterpart.

Nonporous crystalline octatrimethylsiloxy POSS (MS0865) nanoparticles with a characteristic dimension of approximately 1.2 nm were purchased from Hybrid Plastics (USA).

The structure of the TS-530 FS and POSS nanoparticles are shown in Fig. 1. It can be seen that both TS-530 and POSS have the same surface functional groups. Therefore, they have similar chemical properties and compatibility with PDMS.

Fig. 1 The chemical structures of: (a) surface treated hydrophobic TS 530 FS and (b) POSS MS0865.

Toluene was of analytical purity and purchased from Anachemia Co. (Canada) and used as received. For gas separation measurements, H₂, CO₂, CH₄, and C₃H₈ with purities of 99.0%, 99.998%, 99.0%, and 99.0% respectively, were purchased from Praxair (Canada).

2.2. Nanocomposite membrane preparation

Preparation of nanocomposite membranes is noticeably complex. PDMS–FS–POSS nanocomposite membranes were synthesized using a conventional solution-casting method, as this is the most appropriate technique to form void-free nanocomposite membranes. The flowchart of the preparation methodology for the PDMS–FS–POSS nanocomposite membranes is represented in Fig. 2.

Fig. 2 Processing flowchart for the PDMS–FS–POSS nanocomposite membranes.

The membrane formation is very sensitive to parameters such as the weight proportions of silicon oil to solvent, the crosslinker and the catalyst concentration, the fillers’ moisture uptake, and the evaporation time. The weight ratio of the crosslinker and the catalyst to PDMS and the synthesis conditions were kept constant, only the filler content was changed in the formula to produce the final nanocomposite membranes.

Pretreatment of the POSS and FS nanoparticles was considered as an important parameter influencing the nanocomposite membrane preparation. A more efficient removal of moisture adsorbed in the nanoparticles’ surface can be attained at high temperature. Hence, both POSS and FS nanoparticles were placed in a vacuum oven at 150 °C for 24 h to remove any uptake moisture. The dried powders were then quickly cooled down to room temperature and stored in a desiccator to keep them from any organic vapors or water adsorption.

The nanocomposite membranes were prepared through the physical blending of both fillers and the polymer in the solvent in the same way as used to produce pristine membranes.^11,12 Toluene was used as the solvent for the simultaneous incorporation of the fillers into the polymer chains owing to attractive filler–solvent and PDMS–solvent interactions. The main challenge in property enhancement is the quality of the filler dispersion in the polymer matrix. This is why a sonication treatment of 10 min was carried out to break-up nanoparticle aggregates and improve homogeneity. The V24 crosslinker was subsequently introduced into the solution which was further stirred for 30 min. The catalyst was finally added with continuous stirring for another 15 min. In the final reaction solution, the proportion of DEHESIVE, crosslinker and catalyst was a 10 : 0.1 : 0.05 weight ratio in toluene. The reaction solution was then poured onto a Teflon coated glass plate and kept for curing at ambient temperature for up to 3 days. To prevent dust pollution and to control the evaporation rate, the casting die was covered with a glass plate. The nanocomposite membranes were dried gradually in a vacuum oven for 2 h at 80 °C to remove residual toluene and complete the crosslinking reaction. It is believed that in the nanocomposite membranes, POSS produces extra crosslinking.

3. Membrane characterization

3.1. Scanning electron microscopy (SEM)

Morphological observations were conducted on a JEOL JSM-840A scanning electron microscope at a voltage of 15–20 kV.

3.2. Transmission electron microscopy (TEM)

TEM images were taken using a Philips Tecnai 20 at an accelerating voltage of 200 kV.

3.3. Fourier transform infrared spectroscopy (FTIR)

FTIR is extensively used to evaluate the chemical properties of nanocomposite membranes including chemical bonds, molecular energy levels, molecular orientations, and molecular interactions. FTIR was performed on the membranes as well as the nanoparticles using a Nicolet Magna 850 Fourier transform infrared spectrometer (Thermo Scientific, Madison, WI). All samples were dried in a vacuum oven prior to measurement to remove water. The IR spectra were obtained continuously in the 3600–400 cm⁻¹ wavenumber range at a resolution of 4 cm⁻¹ using Happ-Genzel apodization. The attenuated total reflectance (ATR) method was used to collect the membrane spectra using the FTIR apparatus equipped with a liquid-nitrogen cooled narrow-band mercury/cadmium/telluride (MCT) detector using Golden-Gate (diamond IRE) ATR accessories (Specac Ltd., UK). Suitable background corrections were also performed.

3.4. Thermogravimetric analysis (TGA)

A TA Instruments thermogravimetric analyzer (TGA), model Q5000IR, was used to fully investigate the thermal analysis of the FS and POSS nanoparticles as well as the PDMS–FS–POSS nanocomposite membranes. The onset of decomposition temperature (T_onset) was determined through the weight percent versus temperature curve. The study proposes the use of the derivative thermogravimetric curve (DTG) to analytically explain the nature of the degradation realized in either both nanoparticles or in the nanocomposites. A constant rise of temperature (10 °C min⁻¹ up to 950 °C) under a controlled nitrogen environment was implied for all the membranes. 5 to 10 mg of fine nanoparticles were packed cautiously into the pan to form a uniform layer. The PDMS and PDMS–FS–POSS (10 to 16 mg) nanocomposite films were cut into small rectangular pieces and housed in the pan. Dry nitrogen inert gas was flowed over the balance (10 ml min⁻¹) and the sample chamber (25 ml min⁻¹).

4. Results and discussion

4.1. Morphology of the PDMS–FS–POSS nanocomposite membranes

There are numerous important challenges faced in synthesizing POSS-based nanocomposite membranes such as nanoparticle dispersion, processability, durability, thermostability, oxidation and reliability which are major issues for large-scale production.

The morphological analysis of an inorganic–organic nanocomposite membrane can provide valuable information on the uniformity and distribution of the nanoparticles in the polymer matrix. For instance, in the case of nanoscale inorganic additives, phase separation and agglomeration can be observed in SEM and TEM images. The representative SEM photographs from the cross-sections of PDMS–POSS 2%–FS 5%, PDMS–POSS 2%–FS 10% and PDMS–POSS 2%–FS 15% are shown in Fig. 3 which reveals that a uniform distribution of both FS and POSS within the PDMS matrix was achieved without any voids/defects at the polymer–filler interfaces. The TEM image in Fig. 4 also reveals the well-defined nanometer scale distribution of the fillers.

Fig. 3 Cross-sectional SEM images of the: (a) PDMS–POSS 2%–FS 5%, (b) PDMS–POSS 2%–FS 10%, and (c) PDMS–POSS 2%–FS 15% nanocomposite membranes.

Fig. 4 TEM image of the PDMS–POSS 2%–FS 10% nanocomposite membrane.

The dispersability and easy handling of POSS in an organic solvent and therefore a polymer matrix is the main challenge to get well dispersed POSS in hybrid membranes synthesized by conventional physical blending. The lack of strong attraction between POSS nanoparticles allows a good filler dispersion in the polymer matrix. Also, surface functional groups provide a pathway where the surface chemistry of POSS can improve its stable dispersion in the polymer matrix.

However, it can be speculated that the compatibility of the octatrimethylsiloxy POSS with the PDMS–FS nanocomposite structure is due to the strong dipole–dipole interactions between the functional groups and the polymer. Nanoparticles with similar surface properties as the polymer exhibit better adhesion with the polymer. This results in better contact of the nanoparticles with the polymer as well as better distribution. The larger the difference in the surface energies of the polymer and the nanoparticles, the higher the thermodynamic driving force of the nanoparticles to agglomerate. Hence, a modified nanoparticle surface can prevent nanoparticle agglomeration in the polymer structure. On the basis of a molecular dynamics simulation, Striolo et al.¹³ explored the thermodynamic and transport properties of octamethyl and octahydrido substituted POSS incorporated into PDMS. The findings revealed that POSS nanoparticles do not have a tendency to adsorb on each other when incorporated in PDMS. When the interaction is favorable relative to the POSS–POSS interaction, the POSS nanoparticles are well dispersed and the physical properties of the nanocomposite systems are directly related to the POSS dispersion level.

Polymers interact with both fillers by van der Waals and dipole–dipole interactions, resulting in better stable dispersions of FS and POSS in specific organic solvents and matrices. The main difference is that when the POSS content increases, the FS nanoparticle dispersion became somewhat better as seen in Fig. 3. This decreases the inter-nanoparticle forces and helps dispersion. POSS was chosen for its chemistry leading to a strong attachment to the nanoparticles and organic groups that are compatible with the surrounding matrix. Therefore, POSS is helpful in controlling nanoparticle dispersion in polymer membranes via cooperative surface interactions. The homogeneous distribution of both FS and POSS nanoparticles within the polymer matrix depends on the cooperative surface interactions of POSS functional groups through van der Waals forces and the polarity with the polymer. Octatrimethylsiloxy substituted POSS has a high hydrophobicity due to its trimethylsiloxy groups. POSS compounded in the PDMS improves the structure of the nanocomposites and changes the permeant diffusion through the nanocomposite membranes. As a matter of fact, when POSS is used as a filler in PDMS, it affects the polymer chains’ motion within the nanocomposite membranes. It is believed that POSS distribution in the nanocomposite prevents the molecular motion of the polymer chains causing the gas permeation to change.

Moreover, there was an inevitable trade-off between both POSS and FS nanoparticle loadings. As seen in Fig. 3, the POSS dispersion became better with increasing FS content while increasing the total amount of nanofillers in the PDMS matrix leads to nanofiller agglomeration.

4.2. FTIR characterization

The FTIR spectra were used to explore the structure–property relationship of the PDMS–FS–POSS nanocomposite membranes. Fig. 5 presents the spectra of the neat PDMS and both virgin fillers, as well as the different PDMS–FS–POSS nanocomposite membranes. The ATR-IR spectra for the neat POSS exhibited a strong peak at 1060 cm⁻¹ associated with the symmetric Si–O–Si stretching, thus confirming the silsesquioxane cages.

Fig. 5 FTIR-ATR spectra for the FS and POSS nanoparticles, neat PDMS, and PDMS–FS–POSS nanocomposite membranes.

All the PDMS–FS–POSS hybrid membranes have similar FTIR spectra to the reference membrane. The functional groups on the surface of POSS and FS facilitate synergistic effects between both nanoparticles which were improved by hydrogen bonds and van der Waals forces.^7,8,14 Numerous crosslinked 3D networks were produced during crosslinking and thermal processing. The crosslinked networks of the PDMS nanocomposite membranes were studied using FTIR analysis. All hybrid membranes display virtually identical spectra, as presented in Fig. 5. A peak at 790 cm⁻¹ is attributed to the –CH₃ rocking and the –Si–C– stretching in –Si–CH₃. The bands at 900 cm⁻¹, 1258 cm⁻¹, 1455 cm⁻¹ and 2960 cm⁻¹ are assigned to the –Si–O stretching in –Si–OH, the symmetric –CH₃ deformation in –Si–CH₃, the C–H bending and the –CH₂– stretching in –Si–CH₂–, respectively. Moreover, the bands at 1060 cm⁻¹ and 1005 cm⁻¹ are related to the Si–O–Si stretching vibration of the crosslinked PDMS in these novel nanocomposite membranes. The presence of the Si–O–Si stretching absorption peak in the nanocomposite membranes indicates that POSS is well incorporated into the PDMS molecular chains. The intensity of these bands further increases with POSS content in the nanocomposite membranes.

4.3. Thermal stability of the PDMS–FS–POSS nanocomposite membranes

One major application of TGA is the assessment of the thermal stabilities of the filler content in polymers and nanocomposites. The morphology and nanofiller functionality play a vital role in thermal stability of the nanocomposite membranes. However, there have been numerous studies focusing on the thermal stability of POSS-containing polymer nanocomposites using TGA.^5,6,9,15,16

Fig. 6 shows the TGA and DTG curves of FS and POSS nanoparticles in nitrogen. Within a temperature range of 200–300 °C, POSS is rapidly decomposed with almost complete weight loss which can be associated to the sublimation of the octatrimethylsiloxy groups. Hence, POSS itself cannot exhibit thermal stability in nitrogen. On the other hand, only 10.6% weight loss was observed for FS nanoparticles up to 900 °C showing that FS nanoparticles have better thermal stability than POSS.

Fig. 6 TGA-DTG curves for FS and POSS nanoparticles obtained under a nitrogen atmosphere.

Fig. 7 shows the TGA results of the PDMS–POSS nanocomposite membranes. It is clear that the PDMS thermal stability increases with increasing POSS content. This can be related to the interactions and the tight Si–O frameworks.^7,10,17 This indicates that the intramolecular dipole–dipole interactions between the octatrimethylsiloxy groups improve thermal stability of the nanocomposites.^7,10,17

Fig. 7 TGA curves for neat PDMS and the PDMS–POSS nanocomposite membranes obtained under a nitrogen atmosphere.

The DTG results in Fig. 8 show that there is only one degradation step in the thermal decomposition of all the samples. POSS has a rapid thermal degradation starting at around 250 °C, while the membranes have a slower thermal degradation over a wider range of temperatures where the starting decomposition temperature increases with the increase in the POSS content. This reveals that the combination of POSS with PDMS alters the thermal stability of the membranes and indicates that the incorporation of octatrimethylsiloxy POSS is effective in enhancing the thermal stability of these nanocomposite membranes. However, the PDMS–POSS nanocomposite membrane with 6% wt. POSS exhibits the highest thermal stability with a decomposition range of 500–700 °C, which may be attributed to the strong bonding between the PDMS chains and POSS nanoparticles.

Fig. 8 DTG curves for PDMS and the PDMS–POSS nanocomposite membranes obtained under a nitrogen atmosphere.

In the inert nitrogen environment, the thermodegradation profiles of the PDMS-based nanocomposite membranes composed of FS and POSS nanoparticles were also assessed using TGA, and their behaviors are shown in Fig. 9–11. On the basis of the TGA-DTG results, it can be concluded that all of the PDMS–FS–POSS nanocomposite membranes presented two degradation steps in the DTG curves. The first decomposition peak is at around 440–475 °C, while the second one is at around 680–750 °C, but most of the decomposition occurred at the higher temperature.

Fig. 9 TGA-DTG curves of the PDMS–FS–POSS nanocomposite membranes obtained under a nitrogen atmosphere with 5 wt% FS and varying POSS content.

Fig. 10 TGA-DTG curves of the PDMS–FS–POSS nanocomposite membranes obtained under a nitrogen atmosphere with 10 wt% FS and varying POSS content.

Fig. 11 TGA-DTG curves of the PDMS–FS–POSS nanocomposite membranes obtained under a nitrogen atmosphere with 15 wt% FS and varying POSS content.

PDMS however has a tendency to decompose into cyclic trimers and tetramers at high temperatures as a result of the thermodynamic ring-chain equilibrium through the nanocomposite structures, and this equilibrium is limited when an inorganic filler is introduced into the host matrix.¹⁸

The resistance to thermal decomposition of the PDMS–FS–POSS nanocomposite membranes was significantly enhanced by introducing FS and POSS nanoparticles. This noticeable improvement was contributed to the increased interaction of the host polymer chains and fillers caused by synergistic effect between FS and POSS nanoparticles. This may also be associated with the intrinsically high thermal stability of FS which is higher than 420 °C and when embedded into the stable PDMS–POSS membranes this enhanced their thermostability by increasing the PDMS–filler interactions and the tight inorganic Si–O frameworks.¹⁰ Hence, thermally stable bonding occurs among the inorganic and organic components and the influence of the thermally stable inorganic Si–O networks on the thermal properties of the hybrid membranes was significant at elevated temperatures. From screening the results, however, it can be deduced that octatrimethylsiloxy POSS and FS have a profound enhancement on the thermal properties of the PDMS-based nanocomposite membranes. Therefore, the PDMS–FS–POSS nanocomposite membranes can play an imperative role in high-temperature applications.¹⁰

Consequently, in point of fact, these experimental results on the thermal stability of the nanocomposite membranes could be explained by three reasons: (I) the homogenous distribution of the crosslinked 3D structures originating from the synergistic effect between the POSS and FS can develop the resistance of the PDMS membrane to thermal degradation; (II) the vinyl groups in PDMS could easily crosslink at low temperature when the PDMS membranes thermally degraded, which creates more rigid crosslinked 3D structures; (III) the rigid 3D structures could hinder the random motions of the polymer chains more, and prevent the creation of degrading yields, which could enhance the thermal stability of the loaded PDMS–FS–POSS nanocomposite membranes. The most effect could be the competitive results of the three aforementioned reasons.^{7,8,14,17,19–21} When the filler content in the binary system is high, the combined action of the homogenous distribution of the crosslinked 3D structures and the vinyl groups in these PDMS–FS–POSS nanocomposite membranes dominates during the thermal decomposition process.

Finally, in accordance with the discussion presented above, it can be concluded that when the filler content increases, further synergistic effects between FS and POSS are observed, and distribution of FS and POSS becomes uniform, which also results in the resistance to degradation of the PDMS–FS–POSS nanocomposite membranes as well.

Zhang et al.⁵ studied PDMS with octamethyl substituted POSS and tetraethoxysilanes/dibutyltin dilaurate, while Liu et al.²² studied POSS grafted room temperature vulcanized (RTV-g-POSS) silicone rubbers, which presented similar degradation trends to the POSS-containing nanocomposite membranes. Hong et al.,²³ Beltran et al.,³ Nisola et al.,²⁴ and Song et al.² also found a similar thermal degradation for silica containing nanocomposite membranes. For binary nanocomposite films, Chen et al.^14,19 reported a similar thermal degradation for octa[(trimethoxysilyl)ethyl]-POSS and room temperature vulcanized (RTV) silicone rubbers.

4.4. Effect of pressure and filler content on gas permeation

The detailed description of the gas permeation procedure can be found elsewhere as in our previous study.²⁵Fig. 12 shows the permeability coefficient of H₂, C₃H₈, CO₂ and CH₄ at 35 °C as a function of the FS nanoparticle concentration within the PDMS–POSS 2 wt% matrix at different operating pressures. The results of the inclusion of the binary nanofillers (POSS and FS) in the PDMS nanocomposites on gas permeation were clearly observed.

Fig. 12 Permeability of the penetrants (H₂, CO₂, CH₄ and C₃H₈) through the synthesized PDMS–FS–POSS nanocomposite membranes as a function of feed pressure and FS nanoparticle weight content.

Fig. 12 illustrates the effect of FS nanoparticle content on the permeation coefficient of the PDMS–POSS nanocomposite membranes. As observed, the permeability of CO₂ and C₃H₈ across the PDMS–FS–POSS nanocomposites enhances with enhancing feed pressure while for H₂ and CH₄ it decreases but the pressure has a minor effect on H₂ and CH₄ permeability. The increase in permeability with pressure for CO₂ and C₃H₈ through the PDMS–FS–POSS nanocomposites was more profound than the corresponding decrease for H₂ and CH₄. This trend clearly indicates that the tortuous routes of species penetration and the polymer compactness may affect the transport properties. The order of the magnitude of the permeability coefficient through the PDMS–FS–POSS nanocomposite membranes is as follows:

C₃H₈ ⋙ CO₂ > CH₄ > H₂

The results have shown that the employment of FS in the PDMS–POSS polymer matrix increases the C₃H₈ and CO₂ permeability. Indeed, the nanocomposite performance changes due to the permeating molecules when the polymeric membrane is filled with nanofillers. The nature of the permeating molecules–polymer interaction has a significant effect on the nanofiller–polymer interface. From screening the gas permeation data, it can be deduced that in the presence of C₃H₈ and CO₂, a favorable interaction between the polymer and the functional groups of the nanofillers in the nanocomposites occurs. The functional groups in the nanocomposite structure can interact with the polymer chains and the gas molecules, and the trimethylsiloxy groups at the apex silicon atoms enhanced the hydrophobicity of the nanoparticles. However, one possible reason for this enhancement could be the C₃H₈ and CO₂ molecules’ adsorption at the nanofiller–PDMS interfaces and also at the external surfaces of the fillers’ nanoparticles and this provides a driving force for the C₃H₈ and CO₂ molecules leading to a development in the nanocomposites’ permeability. Consequently, the trimethylsiloxyl functional groups have some interactions with the permanent gases which facilitate their permeation through the nanocomposites.

Higher feed flow pressures, and thus, an enhancement in the permeant concentration in the polymer, causes the membrane affinity to plasticization to increase, particularly for highly condensable gases. An imposed higher pressure on the nanocomposites’ surface could gradually compact the polymer matrix. Further to these dual effects that increase diffusivity, solubility is often enhanced with an increase in feed pressure, predominantly for more soluble species, leading to a higher permeability coefficient.²⁶

Without a doubt, the more condensable gases have a substantial effect on the plasticization of elastomeric membranes, and this effect can be also seen for the nanocomposites. Since the operating pressure rises, the plasticization, with respect to the more condensable gases, intensifies; so that, the gas sorption in the polymer domain increases. Therefore, the gas transport of more condensable components such as C₃H₈ and CO₂ rises at high pressures, whereas for less condensable gases such as H₂ or CH₄ the gas transport decreases.

An overall downward trend in the permeability for H₂ and CH₄ gases with an increase in the FS composition was observed in Fig. 12. H₂ showed a 20.98% reduction in the permeability coefficient as the FS loading increased from 0 to 15 wt% at 8 bar. CH₄ also exhibited a continuous reduction in permeation from 0 to 15 wt% FS composition at 8 bar. The PDMS–POSS 2 wt%–FS 15 wt% nanocomposite membrane exhibited a 21.01% reduction in CH₄ permeation. The nanofiller loading has a smaller effect than the pressure on the H₂ and CH₄ permeability. The incorporation of 15 wt% nano-FS particles increases the permeation of C₃H₈ and CO₂ at 8 bar in the PDMS–POSS–FS nanocomposites by close to 26% and 6%, respectively.

4.5. Effect of pressure and filler content on gas concentration

The detailed description of the gas sorption procedure can be found elsewhere as in our previous study.²⁷ The solubility of CO₂, CH₄, H₂, and C₃H₈ in all PDMS–FS–POSS nanocomposite membranes with various FS loadings at 35 °C is represented in Fig. 13. According to the experimental results presented here, the sequence of the concentration of gases in the neat film is

C₃H₈ ≫ CO₂ > CH₄ > H₂

which is in close agreement with their T_c values. In a series of gases, T_c roughly increases as the penetrant size increases, and therefore sorption enhances as the penetrant size increases.

Fig. 13 Penetrant concentration (H₂, CO₂, CH₄ and C₃H₈) in the synthesized PDMS–FS–POSS nanocomposite membranes as a function of feed pressure and FS nanoparticle weight content.

On the basis of the experimental results, it can be deduced that the penetrant sorption improves in the order PDMS < PDMS–FS (5 < 10 < 15 wt%)–POSS 2 wt%. The order of the sorption isotherms are in accord with increasing FS loading. This can possibly be attributed to the solubility of the gases on the surface of the nanofillers which may improve the sorption capacity of the nanocomposites. Therefore, the higher solubility capacities are mainly due to the additional sorptive sites provided by the extra nanofiller content. The solubilities of CO₂, CH₄ and H₂ are almost constant in the pristine and the PDMS–FS–POSS 2 wt% nanocomposites while the C₃H₈ solubility in all the prepared nanocomposites increases with pressure.

5. Conclusion

Nanoparticles of POSS and FS were introduced into PDMS to fully explore the potential property-improving effects that nanofillers have been shown to have on silicon polymer in intended gas separation applications. POSS nanoparticles have been proposed to be effective for the formulation of nanocomposite membranes with speciality binary nanofillers incorporated into PDMS polymer systems. The nanocomposite membranes containing binary additives show synergism in their physicochemical properties such as dispersion and thermal stability. SEM images showed that nanosized FS and POSS were homogenously distributed in the polymer structure indicating that the trimethylsiloxy functional groups on the surface of the FS and POSS nanospheres are compatible with the PDMS matrix. It can be speculated that the compatibility of the octatrimethylsiloxy POSS with the PDMS–FS nanocomposite structure is due to the high dipole–dipole interaction between the functional groups. Hence, appropriate surface-functionalized nanosized FS and POSS nanoparticles were selected for the PDMS membrane. The experimental results obtained from TGA investigations indicated that the PDMS–FS–POSS nanocomposite membranes present better thermal stability due to the presence of Si–O networks in the structure. Moreover, the gas permeation results also indicated that the PDMS–FS–POSS nanocomposite membrane is very promising for C₃H₈/H₂, C₃H₈/CO₂ and C₃H₈/CH₄ separation and the separation properties were enhanced with the increase of the content of nano-FS.

References

1. M. Rezakazemi, K. Shahidi and T. Mohammadi, Desalin. Water Treat., 2014, 1.
2. Y. Song, J. Yu, D. Dai, L. Song and N. Jiang, Mater. Des., 2014, 64, 687.
3. A. B. Beltran, G. M. Nisola, E. Cho, E. E. D. Lee and W.-J. Chung, Appl. Surf. Sci., 2011, 258, 337.
4. M. Rezakazemi, A. Ebadi Amooghin, M. M. Montazer-Rahmati, A. F. Ismail and T. Matsuura, Prog. Polym. Sci., 2014, 39, 817.
5. Q. G. Zhang, B. C. Fan, Q. L. Liu, A. M. Zhu and F. F. Shi, J. Membr. Sci., 2011, 366, 335.
6. Y. Meng, Z. Wei, L. Liu, L. Liu, L. Zhang, T. Nishi and K. Ito, Polymer, 2013, 54, 3055.
7. D. Chen, Y. Liu and C. Huang, Polym. Degrad. Stab., 2012, 97, 308.
8. D. Chen, S. Yi, W. Wu, Y. Zhong, J. Liao, C. Huang and W. Shi, Polymer, 2010, 51, 3867.
9. D. Yang, W. Zhang, R. Yao and B. Jiang, Polym. Degrad. Stab., 2013, 98, 109.
10. D. Zhang, Y. Liu, Y. Shi and G. Huang, RSC Adv., 2014, 4, 6275.
11. M. Rostamizadeh, M. Rezakazemi, K. Shahidi and T. Mohammadi, Int. J. Hydrogen Energy, 2013, 38, 1128.
12. M. Rezakazemi and T. Mohammadi, Int. J. Hydrogen Energy, 2013, 38, 14035.
13. A. Striolo, C. McCabe and P. T. Cummings, J. Phys. Chem. B, 2005, 109, 14300.
14. D. Chen, S. Yi, P. Fang, Y. Zhong, C. Huang and X. Wu, React. Funct. Polym., 2011, 71, 502.
15. M. L. Chua, L. Shao, B. T. Low, Y. Xiao and T.-S. Chung, J. Membr. Sci., 2011, 385–386, 40.
16. B. Dasgupta, S. K. Sen and S. Banerjee, Mater. Sci. Eng., B, 2010, 168, 30.
17. D. Chen, C. Huang and X. Hu, Polym. Compos., 2013, 34, 1041.
18. R. A. Rhein, Thermally stable elastomers: a review, No. NWC-TP-6372, Naval Weapons Center China Lake CA, 1983.
19. D. Chen, J. Nie, S. Yi, W. Wu, Y. Zhong, J. Liao and C. Huang, Polym. Degrad. Stab., 2010, 95, 618.
20. D. Chen, Y. Liu, H. Zhang, Y. Zhou, C. Huang and C. Xiong, J. Inorg. Organomet. Polym. Mater., 2013, 23, 1375.
21. Y. Zhang, Y. Mao, D. Chen, W. Wu, S. Yi, S. Mo and C. Huang, Polym. Degrad. Stab., 2013, 98, 916.
22. Y. Liu, Y. Shi, D. Zhang, J. Li and G. Huang, Polymer, 2013, 54, 6140.
23. H. Hong, L. Chen, Q. Zhang and Z. Zhang, Polym. Eng. Sci., 2011, 51, 819.
24. G. Nisola, A. Beltran, D. Sim, D. Lee, B. Jung and W.-J. Chung, J. Polym. Res., 2011, 18, 2415.
25. M. Rezakazemi, K. Shahidi and T. Mohammadi, Int. J. Hydrogen Energy, 2012, 37, 14576.
26. A. Singh, B. D. Freeman and I. Pinnau, J. Polym. Sci., Part B: Polym. Phys., 1998, 36, 289.
27. M. Rezakazemi, K. Shahidi and T. Mohammadi, Int. J. Hydrogen Energy, 2012, 37, 17275.
28. W. A. Reinerth, J. J. Schwab, J. D. Lichtenhan, Q. Liu, D. Hilton, B. D. Freeman, L. Toy and H. J. Lee, Abstracts of Papers, 222nd ACS National Meeting (MTLS-020), Chicago, IL, United States, 2001.
29. S. D. Kelman, B. W. Rowe, C. W. Bielawski, S. J. Pas, A. J. Hill, D. R. Paul and B. D. Freeman, J. Membr. Sci., 2008, 320, 123.
30. S. D. Kelman, PhD Dissertation, The University of Texas at Austin, 2008.
31. H. Lin, B. Freeman, L. Toy, T. Merkel and R. Gupta, in 2003 DOE University Coal Research Contractors’ Meeting, available online at: http://www.netl.doe.gov/publications/proceedings/03/ucr-hbcu/Freeman.pdf, 2003.

---