Assembly of silica aerogels within silica nanofibers: towards a super-insulating flexible hybrid aerogel membrane

Abstract

Silica aerogels (SA) have well been recognized as one of the most attractive thermal insulation materials, but the inherent brittleness and moisture sensitivity hinder wide applications of these materials. Here, we report the design and fabrication of robust flexible hybrid silica nanofiber (SNF)–SA membranes with super-insulating properties and improved mechanical properties by formation of an interpenetrating network of mesoporous silica within a flexible SNF scaffold. The hybrid SNF/SA membranes were obtained by impregnating electrospun SNF membranes with silica sol, then aging, solvent exchanging, surface modification, and drying at ambient atmosphere. The resultant highly porous (>90%) hybrid SNF/SA membranes exhibit meso- and macroporosity with average pore diameter less than free path of air molecules, improved mechanical strength (224% increase in tensile strength), good flexibility (stiffness < 337.6 mN), hydrophobicity (water contact angle > 144.2°), while maintaining low thermal conductivity (0.021 W m⁻¹ K⁻¹ under ambient conditions). Such a robust hybrid membrane with remarkable integrated performance will have great potential in special thermal management applications under harsh conditions, such as the aerospace field. This success of interpenetrating porous SA in inorganic nanofibrous scaffolds paves a new avenue for the synthesis of multifunctional hybrid aerogels.

---

Introduction

Silica aerogels (SA) are bulk porous materials with low density, large internal surface area, small pore size, and large pore volume.^1,2 Owing to these attractive properties, they possess very low thermal conductivity due to the minimized conduction (low density and a tortuous path for heat transfer through the solid nanostructure), convection (very small pore sizes minimize convection) and radiation (with added IR absorbing or -scattering dopants), and therefore have been identified as the most promising future insulation materials.^3,4 Emerging challenges, however, for SA are their high production cost, inherent brittleness, and instability toward atmospheric moisture, prevent their widespread potential application.^5,6 In parallel, industrial applications of SA under current attention are back in line with their fundamental properties focusing on thermal insulation with some special harsh requirements, for example, space suits, subsea oil pipes, preservation of biological specimens during transport, and so on.⁷ Thus, exploring a kind of flexible or even foldable super-insulation materials has become an attractive topic for the next generation of the thermal insulation industry.^8,9

Different strategies have been so far used for the efficient reinforcement of SA and usually lies in the following three categories: (1) chemical cross-linking with reactive molecules to covalently bridging the silica nanoparticles (NPs) together,¹⁰ (2) incorporating individualized micro- or nanoscopic objects (e.g. glass fibers, NPs, nanotubes) as a secondary phase into the silica matrix,¹¹ (3) interpenetrating SA within a three-dimensional (3D) fibrous scaffold.^4,12,13 With the latter strategy, the skeleton of the scaffold forms a continuous framework and mechanically supports the fragile mesoporous silica NPs assembly (i.e. SA).³ Significant improvements in the mechanical properties (e.g. strength, flexibility) of SA have been reported, with scaffolds based on nanofibrillated cellulose,^3,14 polyurethane (PU) nanofibers,⁴ glass fibers,¹⁵ carbon,¹⁶ and ceramic fibers.^17,18 Among them, electrospun nanofibers possess the robust mechanical strength, fine flexibility, low density, and ease of scalable fabrication from various materials, holding great promise as an exceptional scaffold for tightly interpenetrating SA within the nanofibrous skeleton.^4,12,19–24 Recently, 3D electrospun polyvinylidene fluoride (PVDF) nanofibers have been proposed as scaffolds for the reinforcement of silica aerogels.¹² The electrospun PVDF nanofibers could greatly improve the mechanical strength and flexibility of the silica aerogels, but thermal conductivity values above 0.026 W m⁻¹ K⁻¹ were observed as a result of the significantly enhanced skeletal conductivity. Additionally, a challenge for polymer-based insulation materials is their poor resistance to high temperature or fire, which limits their applications in some special harsh conditions.^16,25

The inorganic fibers could improve the shielding ability of the aerogels to the heat radiation at high temperature and have potential use in high-temperature insulation as well as other special thermal management applications.^17,26 To our best knowledge, there has no reported work on employing electrospun inorganic fibers as scaffolds for the reinforcement of SA. Moreover, owing to the brittleness of the common inorganic fibers, the aerogel composites reinforced by the inorganic fibers were still brittle and less flexible.¹⁶ Significantly, we have recently reported the facile synthesis of flexible silica nanofibrous (SNF) membranes by employing electrospinning and subsequent calcination process.^27,28 With this method, water-soluble poly(vinyl alcohol) (PVA) was introduced as a template material to blend with silica gel for producing the composite nanofibrous membranes via electrospinning. Followed by calcination to remove the organic component, flexible SNF membranes could be fabricated in large-scale.

In this work, we report the fabrication and characterization of a new class of robust flexible hybrid aerogel-based membranes with super-insulating properties and improved mechanical properties by formation of an interpenetrating network of mesoporous SA within a flexible SNF skeleton (Fig. 1). The structure and stiffness of the scaffold can be controlled by varying the SA contents in the hybrid aerogel membranes achieved by regulating silica sol concentration before ambient drying. The SNF surface is compatibilized with the mesoporous SA network using a trimethylchlorosilane (TMCS) sol as active silylating agent. An interpenetrating hybrid aerogel structure is obtained by impregnating the electrospun SNF scaffold (Fig. 1b and c) with a tetraethoxysilane (TEOS)-based sol. After gelation, aging, surface modification and solvent exchanging, the resulting gel is dried at ambient atmosphere (Fig. 1a), yielding a highly porous and flexible hybrid SNF/SA membrane (Fig. 1d and e). Finally, the potential of hybrid membranes for use in super-insulating materials is discussed in the context of the interplay of the structural, thermal, mechanical, and non-wettable properties. It opens a promising way to improve the mechanical stability of the aerogels while keeping a low thermal conductivity via reinforcing the SA by using electrospun nanofibers.

Fig. 1 (a) Schematic illustration of the fabrication process of hybrid SNF/SA membranes. (b) Typical FE-SEM image and (c) photograph of SNF membranes. (d) Typical FE-SEM image and (e) photograph of as-prepared flexible hybrid SNF/SA membranes.

Experimental

Materials

Poly(vinyl alcohol) (PVA, M_w = 88 000), phosphoric acid (H₃PO₄, 85 wt%), ethyl alcohol (99%, EtOH), N,N-dimethyl-formamide (>98%, DMF), hydrochloric acid (37%, HCl), and ammonia (25%, NH₄OH) were purchased from Shanghai Chemical Reagents Co., Ltd., China. Tetraethyl orthosilicate (TEOS), n-hexane (>99%), and trimethylchlorosilane (>95%, TMCS) were purchased from Lingfeng Chemical Co., Ltd., China. Pure water was obtained from a Millipore system. All chemicals were of analytical grade and were used as received without further purification.

Preparation of flexible SNF

PVA solutions were prepared at concentrations of 10 wt% by dissolving the PVA in pure water at 80 °C with vigorous stirring for 10 h. The silica sol with the molar composition of TEOS : H₂O : H₃PO₄ = 1 : 11 : 0.01 was generated from the hydrolysis and polycondensation by dropwise addition of H₃PO₄ and H₂O into TEOS with stirring at room temperature for 8 h. Then 10 g of resultant silica sol was added to the 10 g PVA solution and stirred for another 8 h. The electrospinning process was performed by using a DXES-03 spinning equipment (Shanghai Oriental Flying Nanotechnology Co., Ltd., China) with an applied high voltage of 25 kV and a controllable feed rate of 1 mL h⁻¹. The temperature and relative humidity during electrospinning were 25 ± 2 °C and 45 ± 3%. Finally, the composite membranes were calcined to 800 °C by gradually increasing the temperature at the heating rate of 5 °C min⁻¹ in air to remove the small molecules and PVA by using a muffle resistance furnace (QXR1400-50a, Shanghai Qiantong Instrument Technology Co., Ltd., China) and then cooled down to room temperature naturally to get the flexible SNF.

Preparation of silica sol

The silica sol–gel process employed in this research was in two stages: (i) acid-catalyzed TEOS hydrolysis and (ii) base-catalyzed gelation. In the first step, different amounts of TEOS were dissolved in 17.66 g of EtOH and the mixture stirred for 5 min at 400 rpm with a magnetic stirrer. The acid catalyst solution prepared by dissolving 0.0085 g of HCl in 1.728 g of H₂O was dropwise added into the TEOS/EtOH solution under magnetically vigorous stirring for 1 h. In the second step, 1.728 g of H₂O and 0.024 g of NH₄OH were added slowly into the solution with vigorous stirring for 10 min. Subsequently, 0.987 g of DMF was added into the solution to form homogeneous nanoporous structure of the aerogels. The final silica concentrations of the sol solution were 1, 2, 5, and 8 wt% and the molar ratio of H₂O : EtOH : DMF : HCl : NH₄OH was 1 : 2 : 0.07 : 4.5 × 10⁻⁴ : 9 × 10⁻⁴.

Synthesis of hybrid SNF/SA membranes

The flexible SNF were cut in to 5 × 5 cm² pieces and dipped in the silica sol as a framework for 5 min. After that, the resultant membranes were transferred into a tightly-sealed container for further solidification. After the gelled membranes were formed in approximately 10–20 min, the wet-gel membranes were allowed to age for 24 h at room temperature to strengthen the gel network structure. The wet-gel membranes were washed three times with ethanol for 8 h each, followed by exchanging them into n-hexane once for 8 h. Hydrophobic treatment of the silica gel surfaces was carried out by impregnating with a solution of 10% vol. TMCS/n-hexane at 25 °C for 8 h. The wet-gel films were washed with n-hexane twice in an 8 h cycle each time to remove nearly all of the ethanol-containing fluid and then dried in an air-circulating oven at 40 °C and 60 °C successively for 5 h, followed by further drying at 100 °C for 5 h. Finally, four different samples with the SA concentrations of ca. 20, 35, 60, and 70 wt% were obtained, which were denoted as SNF/SA-1, SNF/SA-2, SNF/SA-3, and SNF/SA-4, respectively.

Characterization

The planar and cross-sectional images of the as-synthesized hybrid SNF/SA membranes were examined by field emission scanning electron microscopy (FE-SEM, SU5000, Hitachi Ltd., Japan) and scanning electron microscopy (SEM, TM3030, Hitachi Ltd., Japan), all samples were coated with gold for 100 s before analysis. Fourier transform infrared spectroscopy (FT-IR) was performed with a Nicolet 8700 FT-IR spectrometer in the range 4000–400 cm⁻¹. The Brunauer–Emmett–Teller (BET) specific surface area, pore volume, and pore-size distribution of the prepared samples were characterized by N₂ adsorption–desorption isotherms with a surface area analyzer (ASAP2020, Micromeritics Co., USA). The thermal conductivities (λ) at room temperature were measured by a TPS2500S thermal conductivity apparatus (Hot Disk, Germany) using a transient plane heat source method. The thermostability of the samples were measured using a TGA400 thermal gravimetric analyser (TG, PerkinElmer, USA) at the heating rate of 10 °C min⁻¹. Water contact angle of the hybrid SNF/SA membranes were measured by a contact angle goniometer Kino SL200B equipped with a tilting base. The bending rigidity were investigated with a three-point flexural bending method on a softness tester (RRY-1000, Hangzhou Qingtong & Boke Automation Technology Co., Ltd., China) with a loading speed of 5 mm min⁻¹ at room temperature. The tensile mechanical properties were measured on a tensile tester (XQ-1C, Shanghai New Fiber Instrument Co., Ltd., China) with a crosshead speed of 10 mm min⁻¹. The hybrid membranes were cut into strips with width of 3 mm as testing samples, and the results were average value of ten samples from each fibrous membrane. The porosity of hybrid SNF/SA membranes was calculated according to the equation,
where ρ, ρ_f and ρ_a are the bulk density of the SNF/SA composites, the skeletal density of the pure silica nanofibers and the skeletal density of the pure silica aerogel, respectively; and ω_f and ω_a are the mass fractions of silica nanofibers and aerogels in the composites, respectively. Herein, ρ_f and ρ_a were both designed to be 2.1 g cm⁻³, which dues to the same composition of the two materials.

Results and discussion

Morphology and microstructure of hybrid SNF/SA membranes

Introducing SA network into a 3D nanofibrous skeleton is expected to reinforce mechanical and thermal properties to the overall hybrid membranes. Therefore, the present work aims to produce a flexible, mechanically strengthen, and super-insulating hybrid SNF/SA membranes by embedding aerogel in ceramic nanofibrous scaffold which can be dried at ambient pressure, and stable under atmospheric conditions. We designed the flexible hybrid membranes based on three criteria: (1) the SNF membranes must assemble into a flexible interlaced network, (2) the interfacial adhesion force between the SNF scaffold and the SA should be strong enough to endure mechanical forces and thermal shocks during any post-treatment processing, (3) the hybrid SNF/SA membranes should maintain perfect thermal insulations under moist environment. The first requirement was satisfied by a versatile, stabilizer-free and sol–gel electrospinning method, which involved the formation of the homogeneous silica sol, the electrospinning assembly and the final calcination. To satisfy the second criteria – the formation of stable aerogel frameworks – we incorporate the flexible SNF as a reinforcement phase into the mesoporous SA to bridge the silica NPs together with a cross-linking procedure, which dues to the strong hydrogen bonding interactions between the SNF and SA NPs. The third criterion was satisfied by the hydrophobization modification of the SNF/SA membranes with suitable silane coupling agent (such as TMCS), which could react at relatively milder conditions to graft modified with hydrophobic groups with no damage.

The representative FE-SEM image of pristine SNF presented in Fig. 1b indicated a randomly oriented 2D nonwoven membranes with an average fiber diameter of 250 nm. By inclusion of SA in the SNF scaffold, the morphologies of resultant hybrid aerogel membranes were remarkably changed. It is clearly showed that the SA agglomerations were well-anchored onto the SNF surface for SA concentration of 20 wt% (i.e. SNF/SA-1), and negligible amount of SA agglomerations were presented among the voids of nanofibers (Fig. 2a). The tight adhesion between SA agglomerations and web-like SNF scaffold could be attributed to the formation of hydrogen bonds between SNF and silica NPs due to the abundance of surface hydroxyls on them (Fig. 3).^5,29–31 While further increased the SA content to 35 (i.e. SNF/SA-2), 60 (i.e. SNF/SA-3) and 70 wt% (i.e. SNF/SA-4), SA agglomerations were not only well-anchored onto the silica nanofiber surface, but also interconnected and fixed securely within the robust 3D SNF scaffolds (Fig. 2b–d).

Fig. 2 FE-SEM images of (a) SNF/SA-1, (b) SNF/SA-2, (c) SNF/SA-3, and (d) SNF/SA-4. (e) FE-SEM image SNF/SA-4 showing that SNF bridge between the gap of the cracks. (f) The corresponding high-magnification FE-SEM image in (d).

Fig. 3 Schematic illustrating the multiscale-assembled mesoporous SNF/SA interface and chemical bonds between mesoporous SA and SNF surface.

Upon careful scrutiny of the FE-SEM images in Fig. 2b–d, some narrow cracks can be found in the hybrid SNF/SA membranes. More interestingly, SNF bridge between the gaps of the cracks (Fig. 2e), which could contribute the high tensile strength even in case existence of undesired cracks. Moreover, the cracks are mostly vertical to the fiber direction, we suppose that the reason maybe the strong adhesion between the SNF and the SA, the aerogels shrink during drying but the SNF barely shrink (Fig. S1, ESI†), whereas the adhesion on the interface are strong, thus shear stresses are induced in the SA nearby the interface.³²

The unique feature of hybrid SNF/SA membranes is that the introduction of SA NPs created the pristine SNF with highly porous structure (Fig. 2f), thus significantly increasing the porosity and specific surface area. The specific surface area and pore structure of the hybrid SNF/SA membranes was examined with Brunauer–Emmett–Teller (BET) measurements, and the data are summarized in Table 1. A significantly increase in the respective surface areas is noted with increasing mass content of the SA, and the calculated BET surface area of SNF, SNF/SA-1, SNF/SA-2, SNF/SA-3, and SNF/SA-4 were 5.03, 167.74, 332.53, 524.71, and 624.19 m² g⁻¹, respectively, indicating the major contributing role of SA NPs on deciding the specific surface area of the hybrid membranes. As shown in Fig. 4a, the N₂ adsorption/desorption isotherms of hybrid SNF/SA membranes display a type IV curve with a hysteresis loop at P/P₀ > 0.6, revealing characteristics of mesopores (pore diameter between 2 and 50 nm) within as-prepared membranes.³ The sharp increase of isotherms in the high relative pressure region (P/P₀ = 0.95–1.0) of hybrid membranes suggests liquid condensation related to the presence of macropores (pore diameter > 50 nm).^33,34 No excessive nitrogen uptake is noted when P/P₀ < 0.05, indicating that the samples do not contain micropores (pore diameter < 2 nm).³⁵ Pore-size distribution calculated by Barrett–Joyner–Halenda (BJH) method from the desorption branch (Fig. 4b) further convinced that the hybrid SNF/SA membranes exhibited a broad range in meso- and macropore size regions (2–70 nm) with small average pore width (<8.6 nm) and a high pore volume of 1.935 cm³ g⁻¹, which is consistent with the aerogel nature. It should be noted that the pore size was mostly around 8 nm, which was much smaller than the average free path of air molecules, thus it could limit the collision of air molecules and reduce the heat convection.^19,36

Table 1 SA mass fraction, porosity, BET specific surface area, average pore size, and thermal conductivity of the pristine SNF and hybrid SNF/SA membranes

Sample	SA content (wt%)	Porosity (%)	BET surface area (m^2 g^−1)	Average pore size (nm)	BJH pore volume (cm^3 g^−1)	Thermal conductivity (W m^−1 K^−1)
SNF	0	91.5	5.03	5.4	0.009	0.026
SNF/SA-1	20	92.4	167.74	5.9	0.210	0.024
SNF/SA-2	35	91.1	332.53	7.3	0.783	0.023
SNF/SA-3	60	91.5	524.71	8.1	1.475	0.022
SNF/SA-4	70	90.8	624.19	8.6	1.935	0.021

---

Fig. 4 (a) N₂ adsorption–desorption isotherms of SNF, SNF/SA-1, SNF/SA-2, SNF/SA-3, and SNF/SA-4. (b) Pore distribution analysis of relevant hybrid aerogel membranes using BJH method.

Thermal properties of hybrid SNF/SA membranes

Fig. 5a presents the thermal conductivities of the pristine SNF and hybrid SNF/SA membranes at room temperature. As expected, the room temperature thermal conductivity of the SNF was significantly decreased from 0.026 W m⁻¹ K⁻¹ to 0.021 W m⁻¹ K⁻¹ as the electrospun SNF membranes were filled with SA. The SNF/SA-4 exhibited extremely low thermal conductivity (∼0.021 W m⁻¹ K⁻¹), which is twice as much as that of glass wool (0.040 W m⁻¹ K⁻¹) and comparable to that of air (0.026 W m⁻¹ K⁻¹) or PU foam (0.024–0.030 W m⁻¹ K⁻¹).^4,37 In addition, the low thermal conductivity indicates that the narrow cracks do not destroy the excellent adiabatic performance of the hybrid SNF/SA membranes. This excellent thermal insulation performance of hybrid SNF/SA membranes could be explained as follows: once the SNF membranes were immersed into the silica sol, some colloidal silica NPs would interact with the nanofibers and adhere tightly on the SNF scaffolds, and then the subsequently generated silica NPs would assemble into the silica gel network with nano-porous structure through the sol–gel process. The SNF were divided by the nano-porous structures of the SA, which consist of abundant NPs. When the heat flow reaches the hybrid SNF/SA membrane, it exhibits much lower solid thermal conductivity (λ_s) than those of the pristine SNF, which dues to the increase of the heat transfer path by the nanoporous structure on the silica aerogel. Moreover, the average pore diameter is smaller than the mean free path (about 70 nm) of main molecules (N₂, O₂) in the atmosphere, which makes it difficult for the air molecules in the pores to collide with each other. Therefore, the gas thermal conductivity (λ_g) of the SNF/SA membranes is decreased. The radiative thermal conductivity (λ_r) makes little contribution to the total thermal conductivity at room temperature.^6,38,39 The hierarchical multi-porous structure of the SNF/SA membranes played an important role in decreasing the thermal conductivities compared with the pristine SNF membranes.

Fig. 5 (a) Thermal conductivity of hybrid SNF/SA membranes as function of concentration of SA. (b) Schematic diagram of the heat transferring mechanism of the hybrid SNF/SA membranes. (c) Thermal gravimetric analysis of pure SNF and hybrid SNF/SA membranes. (d) Photographs show that the hybrid membranes hold great flexibility in liquid nitrogen as well as fire safety field.

The thermal stability of the pristine SNF and hybrid SNF/SA membranes was further characterized by the TG analysis. As shown in Fig. 5c, no obvious weight loss can be detected for pure SNF through the whole heating process from room temperature to 900 °C. There can be observed two weight loss stages for hybrid SNF/SA membranes. The first weight loss (up to 3%) may be caused by the removal of the absorbed water, impurities, and other small molecule substances. In the second weight loss stage, the methyl groups (–CH₃) and ethoxy (–OC₂H₅) were decomposed at higher temperatures, accompanied by the condensation of the hydroxide groups distributed on the surface of silica backbone.^12,40 The weight loss curves then slowly leveled off after the second weight loss stage and finally at 900 °C reached relative light weight loss of 3%, 4%, 10%, and 12% for hybrid SNF/SA-1, SNF/SA-2, SNF/SA-3, and SNF/SA-4, respectively, indicating that the hybrid membranes possess good thermal stability and can be served as excellent high-temperature thermal insulation materials.

In addition to the outstanding thermal insulation performance and thermal stability, this hybrid SNF/SA membrane also presented excellent cryogenic resistance behavior and superior fire-resistance properties. As shown in Fig. 5d and Movie S1 and S2,† the hybrid SNF/SA membranes could hold a great flexibility and even be folded in liquid nitrogen as well as fire safety field, endowing it with a wide operating temperature range and to be satisfy with various application environments (e.g. cryogenic engineering and aerospace).

Mechanical properties of hybrid SNF/SA membranes

Aside from thermal properties, mechanical properties (e.g. stretching stress and flexibility) of the hybrid SNF/SA membranes are of great importance for practical applications.

The tensile stress–strain curves presented in Fig. 6a exhibited that, the samples showed a linear elastic behavior at first with a robust Young's modulus of about 123 MPa, and fracture occurred immediately after the stress achieved its maximum yield value. The concentration of the SA in hybrid SNF/SA membranes significantly affected the mechanical properties of relevant hybrid membranes. As can be seen from Fig. 6a, pristine SNF possessed a breaking tensile stress of 2.0 MPa and elongation of 17%. Increasing the content of SA in the hybrid membrane results in the enhanced tensile stress. With the increase of SA contents to 70 wt%, the tensile stress of hybrid SNF/SA membranes was remarkably increased from 2.0 MPa (pristine SNF) to 6.5 MPa (SNF/SA-4). Meanwhile, the final elongation at break of SNF/SA was dramatically decreased from 17% to 6%. This phenomenon was attributed to the filling of the voids by high content of the micro-cracked SA sections, which could increase the mechanism of consuming energy to hamper the propagation of cracks during the tensile fracture process. Meanwhile, the continuous phase structure forming by the SNF and SA possessed strong interface binding force, which inhibited the fiber slip along the stress direction.

Fig. 6 (a) Breaking tensile stress and (b) bending rigidity of various hybrid SNF/SA membranes. Insets of (b) show photographs of flexible hybrid SNF/SA membranes with various SA contents.

Flexible aerogel membranes could have wide applications in various thermal insulation systems (e.g. space suits). In dramatic contrast to the inherent fragility of traditional inorganic aerogels, the hybrid SNF/SA membranes could bear a repeated bending as high as 180° and rapidly recover their original shape after the release of the stress (Fig. 6b, Fig. 7a, and Movie S3†), indicating their remarkable flexibility. To quantitative characterize the flexibility of resultant hybrid SNF/SA membranes, a term of “stiffness” was introduced and measured with a three-point flexural bending method on a softness tester (Fig. S2, ESI†). As shown in Fig. 6b, the stiffness was increased from 50 to 337.6 mN with the increased content of SA in the hybrid aerogel membranes. Even so, SNF/SA-4 still exhibited a desirable flexibility (337.6 mN), which was also evidenced from the bending shows (Fig. 6b, inset). To further investigate the flexible mechanism of the hybrid SNF/SA membranes, in situ SEM analysis were introduced to evaluate the morphology of the hybrid membranes during the bending and recovery process. As shown in Fig. 7 and Movie S4,† the hybrid SNF/SA-4 membranes possessed excellent elastic property, though tolerate extreme macroscopic deformation. The improved mechanical properties of hybrid SNF/SA membranes will be beneficial for load-bearing thermal insulators and also make the materials to be easily machined to any possible shapes that required in the thermal protection system.

Fig. 7 (a) Optical images presenting the robust flexibility of hybrid SNF/SA-4 membranes. (b) SEM images of hybrid SNF/SA-4 membranes during the bending and recovery process. (c) Corresponding high-magnification SEM images of (b).

The mechanism of the monolithic hybrid SNF/SA membranes can withstand the ambient drying and still retain their flexibility can be attributed to their microstructure. The SA agglomerations were interconnected and fixed securely within the robust 3D SNF scaffold, which prevented the SA from being fragmented.³¹ Upon careful scrutiny of the SEM image in Fig. 2b–e, some narrow cracks can be found in the hybrid SNF/SA membranes. More significantly, gaps between the cracked aerogel sections are bridged by the nanofibers (Fig. 2e), preventing them from disintegrating. Moreover, in situ cross-sectional SEM analysis of the facile bending and recovering process of the hybrid SNF/SA-4 revealing excellent flexible with no cracks appearing during the process (Fig. 7b). Interestingly, the enlarged SEM images at the point of maximum deformation reveals a sheet-like structure with micro-scale gaps among the sheets (Fig. 7c). The micro-cracks and sheet-like structure in the hybrid aerogels could provide enough space to endure a certain deformation and make them exhibit remarkable flexibility.^13,41 Another aspect for the unexpected flexibility of the hybrid aerogels could be ascribed from the high aspect ratio and entanglement of SNF, which could absorb the tensile stress at a high level by the bending and slipping of individual SNF.²⁶ Although the microstructure of hybrid SNF/SA membranes would deform to some extent under external pressure, the SA agglomerations were confined in the 3D SNF scaffold securely to ensure the hybrid membranes retained their integrity. Consequently, synergic effects of the robust flexible SNF scaffold and SA, along with their unique microstructures, endowed the hybrid membranes a perfect flexibility.

Hydrophobicity of hybrid SNF/SA membranes

Another important criteria for high-performance thermal insulation is hydrophobicity, which could keep the hybrid membranes with perfect thermal insulation even in moist environment.⁴² Herein, the hydrophobicity of the hybrid SNF/SA membranes was achieved by TMCS modification. Interestingly, it was found that the as-prepared hybrid membranes could keep their integrity and flexibility (Fig. 1e) even some small cracks can be observed (Fig. 1d), however, hybrid aerogel membrane without TMCS modification would crack to pieces (Fig. S3, ESI†) and exhibit less flexibility, suggesting the important role played by TMCS. WCAs were measured to quantitatively express the hydrophobicity of the hybrid SNF/SA membranes. As shown in Fig. 8a, by introducing SA into the SNF, the WCAs of these hybrid membranes exhibited a gradually increase in WCAs. With the increased SA contents from 20 wt% to 70 wt%, SNF/SA-1, SNF/SA-2, SNF/SA-3, and SNF/SA-4 presented linear increase in WCAs of 144.2°, 147.8°, 152.5°, 155.1°, respectively. Additionally, hybrid SNF/SA membrane is strongly non-wettable for all of the liquid droplets (i.e. water, milk, orange juice) evaluated (Fig. S4, ESI†). The good hydrophobicity of the hybrid SNF/SA membranes could be ascribed to the following reasons: (1) incorporating SA in SNF membranes results in hierarchical and porous structures onto/in the nanofibers (Fig. 2); (2) the hydrophilic silanol groups on the surfaces of SNF/SA composites were replaced by the hydrophobic methyl groups after TMCS modification. Evidence for the hydrophobization of the hybrid membranes also arose from the FT-IR spectral analysis (Fig. S5, ESI†), the characteristic peak around 2964 cm⁻¹ was assigned to the stretching vibration of –CH₃.⁴⁰

Fig. 8 (a) WCAs and the selected optical profile of water droplets of various hybrid SNF/SA membranes measured as a function of SA contents. (b) WCAs and the corresponding shapes of water droplets for the SNF/SA-4 membrane annealed at different temperature.

We further examined the hydrophobicity of the hybrid SNF/SA membranes (take SNF/SA-4 as an example) at various calcination temperatures. As shown in Fig. 8b, the resultant membranes maintained promising hydrophobicity of 140° even after annealing treatment up to 300 °C. Further increasing the annealing temperature up to 400 °C, the hybrid SNF/SA membranes lost their hydrophobicity, which was ascribed to the complete degradation of –CH₃ at such high temperature. Such phenomenon could be verified from TG analysis (Fig. 5b), which exhibited a sharp drop of weight from 300 °C to 400 °C. Based on these data, we concluded that the working temperature limitation of these hybrid membranes could be as high as 300 °C. Consequently, we believe that nanocomposites with such a thermal-stable hydrophobicity can meet application requirements where extreme heat may occur.

Conclusions

In summary, we have demonstrated a novel strategy for the controllable fabrication of robust hybrid SNF/SA membranes with super-insulating properties and improved mechanical properties, by forming an interpenetrating silica network inside a flexible electrospun SNF scaffold. Due to the synergic effects of the SNF scaffold and SA matrix, the obtained hybrid SNF/SA membranes show excellent robustness and flexibility which overcome the inherent fragility of traditional inorganic aerogels. Moreover, the porous hybrid membranes possessed high specific surface area (up to 624.19 m² g⁻¹), high porosity (>90%), good thermal stability, while maintaining low thermal conductivity (0.021 W m⁻¹ K⁻¹). Further optimization of the controllable parameters of the SNF/SA membranes is expected to produce porous superinsulation hybrid membrane with even greater integrated performance for special thermal management applications under some harsh conditions, such as the insulation layer of aircraft or space suit, cold-protective for cryogenic storage and so on.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (No. 51322304 and U1232116), the Fundamental Research Funds for the Central Universities, the “DHU Distinguished Young Professor Program”, and the project was funded by State Key Laboratory for Modification of Chemical Fibers and Polymer Materials, Donghua University (LK1513).

Notes and references

1. C. A. Morris, M. L. Anderson, R. M. Stroud, C. I. Merzbacher and D. R. Rolison, Science, 1999, 284, 622–624.
2. L. Sorensen, G. F. Strouse and A. E. Stiegman, Adv. Mater., 2006, 18, 1965.
3. S. Y. Zhao, Z. Zhang, G. Sebe, R. Wu, R. V. R. Virtudazo, P. Tingaut and M. M. Koebel, Adv. Funct. Mater., 2015, 25, 2326–2334.
4. L. C. Li, B. Yalcin, B. N. Nguyen, M. A. B. Meador and M. Cakmak, ACS Appl. Mater. Interfaces, 2009, 1, 2491–2501.
5. A. Fidalgo, J. P. S. Farinha, J. M. G. Martinho and L. M. Ilharco, J. Mater. Chem. A, 2013, 1, 12044–12052.
6. B. Wicklein, A. Kocjan, G. Salazar-Alvarez, F. Carosio, G. Camino, M. Antonietti and L. Bergstrom, Nat. Nanotechnol., 2015, 10, 277–283.
7. A. Bang, C. Buback, C. Sotiriou-Leventis and N. Leventis, Chem. Mater., 2014, 26, 6979–6993.
8. H. Q. Guo, M. A. B. Meador, L. McCorkle, D. J. Quade, J. A. Guo, B. Hamilton, M. Cakmak and G. Sprowl, ACS Appl. Mater. Interfaces, 2011, 3, 546–552.
9. M. A. B. Meador, E. J. Malow, R. Silva, S. Wright, D. Quade, S. L. Vivod, H. Q. Guo, J. Guo and M. Cakmak, ACS Appl. Mater. Interfaces, 2012, 4, 536–544.
10. M. A. B. Meador, L. A. Capadona, L. McCorkle, D. S. Papadopoulos and N. Leventis, Chem. Mater., 2007, 19, 2247–2260.
11. M. A. B. Meador, S. L. Vivod, L. McCorkle, D. Quade, R. M. Sullivan, L. J. Ghosn, N. Clark and L. A. Capadona, J. Mater. Chem., 2008, 18, 1843–1852.
12. H. J. Wu, Y. T. Chen, Q. L. Chen, Y. F. Ding, X. Q. Zhou and H. T. Gao, J. Nanomater., 2013, 10, 375093.
13. H. Z. Sai, L. Xing, J. H. Xiang, L. J. Cui, J. B. Jiao, C. L. Zhao, Z. Y. Li and F. Li, J. Mater. Chem. A, 2013, 1, 7963–7970.
14. D. Bendahou, A. Bendahou, B. Seantier, Y. Grohens and H. Kaddami, Ind. Crops Prod., 2015, 65, 374–382.
15. B. Yuan, S. Q. Ding, D. D. Wang, G. Wang and H. X. Li, Mater. Lett., 2012, 75, 204–206.
16. T. Y. Wei, S. Y. Lu and Y. C. Chang, J. Phys. Chem. C, 2009, 113, 7424–7428.
17. A. Karout, P. Buisson, A. Perrard and A. C. Pierre, J. Sol-Gel Sci. Technol., 2005, 36, 163–171.
18. S. Li, C. A. Wang and L. F. Hu, J. Am. Ceram. Soc., 2013, 96, 3223–3227.
19. F. Bai, J. T. Wu, G. M. Gong and L. Guo, J. Mater. Chem. A, 2015, 3, 13198–13202.
20. Y. Si, J. Y. Yu, X. M. Tang, J. L. Ge and B. Ding, Nat. Commun., 2014, 5, 5802.
21. Y. Si, Q. X. Fu, X. Q. Wang, J. Zhu, J. Y. Yu, G. Sun and B. Ding, ACS Nano, 2015, 9, 3791–3799.
22. Z. X. Zhong, M. C. Wingert, J. Strzalka, H. H. Wang, T. Sun, J. Wang, R. K. Chen and Z. Jiang, Nanoscale, 2014, 6, 8283–8291.
23. X. Wang, B. Ding, J. Yu and M. Wang, Nano Today, 2011, 6, 510–530.
24. X. Wang, B. Ding, G. Sun, M. Wang and J. Yu, Prog. Mater. Sci., 2013, 58, 1173–1243.
25. G. Q. Zu, J. Shen, L. P. Zou, W. Q. Wang, Y. Lian, Z. H. Zhang and A. Du, Chem. Mater., 2013, 25, 4757–4764.
26. Y. S. Si, X. Mao, H. X. Zheng, J. Y. Yu and B. Ding, RSC Adv., 2015, 5, 6027–6032.
27. M. Guo, B. Ding, X. H. Li, X. L. Wang, J. Y. Yu and M. R. Wang, J. Phys. Chem. C, 2010, 114, 916–921.
28. F. Zhao, X. F. Wang, B. Ding, J. Y. Lin, J. P. Hu, Y. Si, J. Y. Yu and G. Sun, RSC Adv., 2011, 1, 1482–1488.
29. D. Klemm, B. Heublein, H. P. Fink and A. Bohn, Angew. Chem., Int. Ed., 2005, 44, 3358–3393.
30. B. K. Sea, M. Watanabe, K. Kusakabe, S. Morooka and S. S. Kim, Gas Sep. Purif., 1996, 10, 187–195.
31. S. N. Schiffres, K. H. Kim, L. Hu, A. J. H. McGaughey, M. F. Islam and J. A. Malen, Adv. Funct. Mater., 2012, 22, 5251–5258.
32. J. Z. Feng, C. R. Zhang, J. Feng, Y. G. Jiang and N. Zhao, ACS Appl. Mater. Interfaces, 2011, 3, 4796–4803.
33. X. F. Wang, N. G. Akhmedov, Y. H. Duan, D. Luebke and B. Y. Li, J. Mater. Chem. A, 2013, 1, 2978–2982.
34. Y. Si, T. Ren, B. Ding, J. Y. Yu and G. Sun, J. Mater. Chem., 2012, 22, 4619–4622.
35. K. H. Kim, Y. Oh and M. F. Islam, Adv. Funct. Mater., 2013, 23, 377–383.
36. T. Gao, B. P. Jelle, L. I. C. Sandberg and A. Gustavsen, ACS Appl. Mater. Interfaces, 2013, 5, 761–767.
37. L. Kocon, F. Despetis and J. Phalippou, J. Non-Cryst. Solids, 1998, 225, 96–100.
38. G. Q. Zu, J. Shen, W. Q. Wang, L. P. Zou, Y. Lian, Z. H. Zhang, B. Liu and F. Zhang, Chem. Mater., 2014, 26, 5761–5772.
39. H. Maleki, L. Duraes and A. Portugal, Microporous Mesoporous Mater., 2014, 197, 116–129.
40. S. Y. C. Young-Geun Kwon, J. Mater. Sci., 2000, 35, 6075–6079.
41. T. Shimaz, M. Miura, N. Isu, T. Ogawa, K. Ota, H. Maeda and E. H. Ishida, Mater. Sci. Eng., A, 2008, 487, 340–346.
42. G. M. Gong, K. Gao, J. T. Wu, N. Sun, C. Zhou, Y. Zhao and L. Jiang, J. Mater. Chem. A, 2015, 3, 713–718.

---

Footnotes

† Electronic supplementary information (ESI) available: Detailed figure to describe the synthesis process, three-point flexural bending test, the FE-SEM images of SNF/SA membranes before and after hydrophobization treatment. See DOI: 10.1039/c5ra18137b

‡ These authors contributed equally to this work.

---