Mechanically robust and shape-memory hybrid aerogels for super-insulating applications

Jinrong Wu *a, Lingping Zeng c, Xiaopeng Huang c, Lijuan Zhao *b and Guangsu Huang a
aState Key Laboratory of Polymer Materials Engineering, College of Polymer Science and Engineering, Sichuan University, Chengdu 610065, China. E-mail: wujinrong@scu.edu.cn
bCollege of Chemistry and Materials Science, Sichuan Normal University, Chengdu 610068, China. E-mail: lijuan_zhao@scinu.edu.cn
cDepartment of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA

Received 28th March 2017 , Accepted 12th June 2017

First published on 12th June 2017


Super-insulating aerogels are promising materials to improve the energy efficiency of buildings. However, fabricating super-insulating yet mechanically robust and shape-memory aerogels remains challenging. Here, we integrate graphene oxide and a block copolymer to fabricate hybrid aerogels with triple networks and systematically study their mechanical and thermal properties. We show that the first network serves as sacrificial bonds and dissipates energy upon deformation, enabling the aerogels to have a high mechanical performance. The second network allows the aerogels to memorize the original permanent shape, and the third network is able to store strain energy and fix the aerogels in a temporary shape by vitrification. Remarkably, the strong phonon-scattering effect generated by the enormous interfaces between the three networks yields an ultra-low solid thermal conductivity of ∼8 mW m−1 K−1. The multi-functionality makes this class of hybrid aerogels particularly suitable for super-insulation applications on complex surfaces and in small spaces of buildings, industry and spacecrafts.


Introduction

Maintaining a comfortable interior environment in buildings is responsible for approximately 15% of both global final energy consumption and total global CO2 emissions.1–3 Reducing the huge energy demand and CO2 emissions is now a compulsory obligation across the world. Achieving this goal requires installation of thermal insulating materials to minimize the heat transfer between the interior and outside of building envelopes. Ideally, thermal insulating materials should have a thermal conductivity (λ) lower than the value of air (25 mW m−1 K−1). However, conventional thermal insulating materials, such as expanded polystyrene, polyurethane, fiberglass and mineral wool, have λ values around 30–45 mW m−1 K−1 under ambient conditions.4 These relatively high λ values often lead to quite thick building envelopes that are not desirable considering the cost-effectiveness factor, architectural design and limited space of retrofitting. Vacuum insulation panels and silica aerogels with λ values below that of air represent the state-of-the-art super-insulating materials.4 The drawback of using vacuum insulation panels is their cost and increasing thermal conductivity with usage time. Silica aerogels are fragile and therefore result in difficulties in processing and building shape adaption. In particular, these materials cannot be fitted in spaces with small openings. An alternative strategy to address this challenge is to develop super-insulating yet mechanically robust and shape-memory aerogels.

During the past few years, significant research efforts have been devoted to developing strong and resilient aerogels.5–18 These aerogels are typically made of strong and flexible building blocks, such as graphene and carbon nanotubes.5–14 Graphene and carbon nanotubes are electrically conductive, and thus carriers can transport in the aerogel networks, increasing their thermal conductivity. Consequently, the λ values of these aerogels are typically on the order of 25–300 mW m−1 K−1.19–23 Hybridizing electrically insulating materials with graphene or carbon nanotubes interrupts the conduction paths of carriers, offering a possibility to reduce the thermal conductivity of the aerogels. For example, freeze-casting a suspension of graphene oxide, cellulose nanofibers and sepiolite nanorods produces an aerogel with a thermal conductivity as low as 15 mW m−1 K−1, yet it is mechanically strong and acts as a fire retardant.24 This aerogel, however, does not show shape-memory properties. To extend the utility of thermal insulating aerogels on irregular surfaces and in small spaces, suitable designing strategies are needed to provide a multi-functionality that not only enables an ultralow thermal conductivity but also drives the recovery of the shape after mechanical deformation or temporary shape fixing. To the best of our knowledge, fabricating aerogels with such multi-functionality has not yet been demonstrated.

In this work, we describe a novel and practical approach for fabricating super-insulating hybrid aerogels with multiple networks. To integrate the multiple networks, we first prepare a physical network of graphene oxide (GO) aerogel (Network I), and then infuse Network I with a solution of triblock copolymer. The functional groups on the triblock copolymer can react with GO, forming a GO crosslinked triblock copolymer network (Network II). After removing the solvent, the triblock copolymer itself forms a physical network (Network III) through phase separation between the hard and soft blocks. We find that Network I serves as sacrificial bonds and dissipates energy upon deformation, resulting in the high mechanical performance of the aerogels. Network II allows the aerogels to memorize the original permanent shape due to its elastic nature, and Network III is able to store strain energy and fix the aerogels in a temporary shape by vitrification. Most importantly, the interfaces between the three networks serve as phonon scattering centers that lead to an ultra-low thermal conductivity of the aerogels. These mechanically strong and shape-memory aerogels with ultra-low thermal conductivity have great potential in a wide variety of thermal-insulating applications by improving the energy efficiency.

Results and discussion

Integrating multiple networks of the hybrid aerogels

To integrate the multiple networks, we use a two-step strategy to fabricate the hybrid aerogels (Fig. 1a). In the first step, we prepare a GO aerogel by using ethylenediamine (EDA) induced gelation of an aqueous GO suspension at 95 °C for 6 hours, followed by lyophilization.25 Scanning electron microscopy (SEM) shows that the GO aerogel has a foam-like structure with interconnected cells of tens of micrometers (Fig. 1b). The cell walls are composed of GO nanosheets functionalized with EDA by reacting the primary amine groups of EDA with the epoxy and carboxyl groups on GO. To confirm this reaction, we conduct Fourier transform infrared spectra (FTIR) measurements on pristine GO and the GO aerogel. We find decreased intensities of the bands at 1733 and 1054 cm−1 corresponding to the carboxyl groups and epoxy groups of pristine GO, while new bands at 1656 and 1563 cm−1 corresponding to the amide groups and amine groups show up in the GO aerogel after the reaction (Fig. S1 of the ESI). In the meanwhile, GO is partially reduced by EDA, leading to the reduction in the O/C ratio, as illustrated by X-ray photoelectron spectroscopy (XPS) measurements in Fig. 1d. During the subsequent freezing process, the partially reduced GO nanosheets are forced to concentrate and align along the ice crystal boundaries;26 this allows GO nanosheets to form a continuous network (Network I) after removing ice by lyophilization. Since the partial reduction enhances the π–π interaction between GO nanosheets, Network I is quite stable and thus can maintain its structural integrity upon further treatment.27 This structural stability allows us to infuse the GO aerogel with another liquid and perform a further lyophilization process.
image file: c7ta02686b-f1.tif
Fig. 1 Fabrication and microstructures of the aerogels. (a) Schematic description of the fabrication process of the hybrid aerogel. An aqueous suspension of GO (①) undergoes gelation with EDA and the gel is then freeze dried to form a GO aerogel (②). The GO aerogel is infused with a para-xylene solution of SEBS (③) and is further lyophilized to form a GO/SEBS hybrid aerogel (④). (b) SEM images of the GO aerogel. (c) SEM images of the GO/SEBS hybrid aerogel. (d) XPS spectra of pristine GO nanosheets, GO aerogel and GO/SEBS-1. (e) Schematic description of the microstructure of the hybrid aerogels.

In the second step, we infuse the GO aerogel with a para-xylene solution of a triblock copolymer, polystyrene-b-poly(ethylene-butylene)-b-polystyrene-g-maleic anhydride (SEBS). The maleic anhydride groups of SEBS react with the primary amine groups on the functionalized GO nanosheets by using 4-dimethylaminopyridine as a catalyst at 65 °C overnight. This reaction is demonstrated by FTIR measurements, which indicate that the band at 1717 cm−1 corresponding to anhydride groups almost disappears, while the intensity of the band at 1656 cm−1 corresponding to the amide groups increases after the reaction (Fig. S1 of the ESI). The reaction allows the GO nanosheets to act as giant crosslinking points for SEBS, which forms covalently crosslinked Network II, as schematically described in Fig. 1e. After removing para-xylene by lyophilization, we obtain a GO/SEBS hybrid aerogel. The hybrid aerogel barely shows shape contraction compared with the pristine GO aerogel. In addition, the hybrid aerogel has nearly the same 3D network structure and cell size as the GO aerogel, as shown by the SEM images in Fig. 1c. This suggests that the GO aerogel is indeed very stable and its structure does not change upon hybridization with SEBS. More importantly, the SEM images further show that the SEBS molecules are adsorbed onto the GO nanosheets. The adsorption not only increases the thickness of the cell walls, but also makes the connecting junctions between cells more strong. The SEBS on the cell walls phase separates into a hexagonal microphase structure with polystyrene (PS) cylinders embedded in the continuous poly(ethylene-butylene) (PEB) phase. This microphase structure is the same as that of bulk SEBS, as illustrated by small-angle X-ray scattering (SAXS) measurements in ESI Fig. S2. The phase separation leads to the formation of a physical network (Network III) in which the PS domains act as crosslinking points (Fig. 1e). The glass transition temperature (Tg) of PS domains is 87 °C, as measured by differential scanning calorimetry in ESI Fig. S3.

We prepare three hybrid aerogels, by infusing the GO aerogel with SEBS solutions of three different concentrations. The samples are designated as GO/SEBS-x, where x represents the concentration of x% (wt/vol) SEBS in p-xylene.

Physical properties

Integrating SEBS with GO changes the physical properties of the aerogels. The pristine GO aerogel is highly porous with an ultralow apparent density (ρ) of 9 mg cm−3, while the apparent density of the hybrid aerogels increases linearly with the concentration of SEBS solution (Fig. 2a). We also find that the ρ value of a hybrid aerogel is nearly equivalent to the sum of the apparent density of the GO aerogel and the weight fraction of the SEBS solution. This again suggests that there is negligible shape shrinkage during the second fabrication step. Analysis of nitrogen gas adsorption reveals that the cell walls of the aerogels are mesoporous. The GO aerogel has an average pore diameter of 18 nm (Fig. 2b). Interestingly, GO/SEBS-1 and GO/SEBS-2 have similar pore diameters of 15 and 17 nm, respectively, despite that they have lower surface areas than the GO aerogel (ESI Fig. S4). However, if the concentration of the SEBS solution is higher than 2%, the pore size of the resulting hybrid aerogel becomes much larger than that of the GO aerogel, due to the dramatically reduced surface area. The increase in the pore diameter is undesirable for thermal insulating applications, considering the Knudsen effect that when the mean free path of the gas molecules is larger than the pore diameter, a gas molecule located inside a pore will hit the pore wall rather than another gas molecule.28
image file: c7ta02686b-f2.tif
Fig. 2 Physical properties of the aerogels. (a) Apparent density of the aerogels as a function of SEBS concentration in para-xylene solution. (b) Average pore diameter of the aerogels. (c) Contact angle of the aerogels. (d) A disc-like sample of GO/SEBS-2 stands on water for one week (left), while the sample of the GO aerogel quickly sinks into water (right).

Integrating SEBS with GO also changes the hydrophobicity of the aerogels. Contact angle measurements show that a drop of water is immediately absorbed into the GO aerogel when they contact with each other (Fig. 2c and ESI Movie 1), suggesting that the GO aerogel is highly hydrophilic. In contrast, the hybrid aerogels are highly hydrophobic, as they have a water contact angle of about 136° (Fig. 2c). In addition, the contact angle is nearly independent of the SEBS fraction in the aerogels, probably due to that GO nanosheets are fully covered with SEBS in all three samples. This high hydrophobicity of the hybrid aerogels allows a disc-like sample to stand on the water for more than one week without getting wet, while the sample of the GO aerogel quickly takes up water and sinks into the water (Fig. 2d). Hydrophobicity is a critical property for thermal insulating materials, as a hydrophilic aerogel otherwise absorbs water from air that greatly increases the thermal conductivity of the materials.

Thermal conductivity

The total thermal conductivity (λ) of aerogels includes contributions from solid (λs), gaseous (λg), radiative (λr), and convective (λc) components.29,30 We measure the solid component λs of the aerogels by using an in-house cooler-bridge steady state (CBSS) platform under vacuum conditions,31 as schematically shown in Fig. 3a. Unlike the conventional steady state method, the CBSS approach uses a thermoelectric cooler (TEC) to cool one end of the sample while keeping the opposite end of the sample at ambient temperature through a resistive heater. Since the heater and the environment are maintained at the same temperature, the heat loss from the heater to the environment through wire conduction and radiation is eliminated. Heat flux through the sample is accurately measured without calibration of the heat loss as in the conventional steady state method. For each sample, we measure its thermal conductivity by sweeping across various temperature differences between the TEC cold side and the ambient. The input power to the resistive heater that maintains the heating side of the sample at ambient temperature is linearly proportional to the temperature difference across the sample. The slope of the input power vs. temperature difference curve is the thermal conductance of the sample, defined as G = λsA/L, where A is the sample cross-sectional area and L is the sample thickness. Consequently, the thermal conductivity λs is determined by regressing the input power with the temperature difference across the heating and cooling ends of the sample, as shown by one representative fitting curve in Fig. 3b.
image file: c7ta02686b-f3.tif
Fig. 3 Thermal conductivity of the aerogels. (a) Measurement of thermal conductivity of the aerogels using an in-house cooler-bridge steady state platform developed in the NanoEngineering Group of the Massachusetts Institute of Technology. (b) The measured heater input power as a function of temperature difference across the heating and cooling ends of an aerogel sample. The dashed line is a linear regression fit to the data. (c) Tested solid thermal conductivity and calculated gaseous thermal conductivity for GO and GO/SEBS aerogels. (d) A comparison of the thermal conductivity and apparent density of the GO/SEBS aerogels with those of other porous materials: (1) GO/SEBS hybrid aerogels; (2) expanded polystyrene; (3) polyurethane foam; (4) mineral wool; (5) carbon based aerogels; (6) silica aerogels; (7) phenolic resin aerogels.4,34 (e) Thermal image of a GO/SEBS-2 aerogel (gray) inserted in a sheet of expanded polystyrene foam (EPF, light blue). Measurement was done by placing the EPF on a heating plate of 48 °C (light brown) under an ambient temperature of 17 °C. The image was a snap-shot at 42 s. (f) Measured temperatures of the two points in (e) as a function of time. The black point is located on PFS, while the red point is located on the GO/SEBS-2 aerogel.

The measured λs values for GO and GO/SEBS aerogels are shown in Fig. 3c. The thermal conduction through the solid in aerogels is limited by the extremely small connections between the thin sheets, and thus the GO aerogel has a low λs of approximately 12 mW m−1 K−1. Hybridization with SEBS further reduces the λs, as shown in Fig. 3c. The GO/SEBS-1 and GO/SEBS-2 hybrid aerogels have extremely low λs values of 8 mW m−1 K−1 and 11 mW m−1 K−1 respectively, despite the fact that they have much higher apparent density than the GO aerogel. Such an abnormal phenomenon is attributed to the interfacial resistance between the multiple networks.32,33 The existing interfaces between the GO, polybutadiene phase and PS phase form phonon barriers, where phonons are scattered in a diffusive way, leading to reduction in thermal conductivity. In addition, the GO aerogel is electrically conductive with a resistivity of 105 Ω cm. As a result, the transportation of carriers also contributes to the thermal conductivity. However, after hybridization with SEBS, the hybrid aerogel becomes electrically insulating, with a resistivity of about 1 × 109 Ω cm.

The λs value increases with the SEBS fraction (Fig. 3c), since the apparent density of the aerogels increases (Fig. 2a), which leads to more thermal conduction paths through the solid. In particular, GO/SEBS-4 has an apparent density of 48.7 mg cm−3, which is 5-fold higher than that of the GO aerogel, and thus it has a thermal conductivity higher than that of the GO aerogel.

The gas λg of aerogels is significantly suppressed due to the restricted motion of gas molecules in the ultrafine pores. The λg within aerogels is estimated using the following formula:28

image file: c7ta02686b-t1.tif
where Π is the porosity, λg0 is the gaseous conductivity of free air, β is a constant and for air in aerogels β ≈ 2. Kn is the Knudsen number defined as Kn = lm/δ, where lm and δ are the mean free path of a gas molecule and the pore diameter of the aerogels, respectively. Since lm is about 75 nm in free space at atmospheric pressure, and δ is much smaller than lm (Fig. 2b), we estimate λg to be about 1 mW m−1 K−1 for the GO, GO/SBES-1 and GO/SEBS-2 aerogels at 1 bar, as shown in Fig. 3c. Compared with λs, such an ultralow value of λg does not contribute significantly to the total thermal conductivity.

In addition, the convective λc is negligible since the convection of the gas trapped in aerogels with pore sizes less than 1 mm at ambient pressure is completely suppressed.29 The radiative λr also makes little contribution to the total thermal conductivity of porous materials, especially considering the high infrared-absorbing efficiency of GO in the aerogels.24,30 Due to the insignificance of λr and λc, the total thermal conductivity λ is estimated to be λs + λg. The estimated λ values are from 9 to 17 mW m−1 K−1 for the hybrid aerogels. These values are compared with those of some conventional and state-of-the-art thermal-insulating materials, as shown in Fig. 3d. In terms of both thermal conductivity and apparent density, our aerogels outperform conventional thermal-insulating materials, such as expanded polystyrene, polyurethane foam and mineral wool. Moreover, the thermal conductivities of our aerogels are comparable to the lowest limit of state-of-the-art thermal-insulating materials, such as phenolic resin aerogels and silica aerogels.

To directly visualize the superior thermal insulating properties of the hybrid aerogels, we imbed a cylindrical sample of GO/SEBS-2 in a sheet of expanded polystyrene foam (EPF) with a thickness of 15 mm. The sheet is then placed on a heating plate with a temperature of 48 °C. We use an infrared camera to record the temperature change of the sheet under an environmental temperature of 17 °C, as shown in ESI Movie 2. The GO/SEBS-2 sample has obviously lower temperature than the EPF sheet, as shown in a snap-shot at 42 s in Fig. 3e. We randomly select one point on the EPF sheet (Point 1) and another point on the hybrid aerogel (Point 2), and follow the temperature change of the two points as a function of time, as shown in Fig. 3f. The temperature of both points rises rapidly first and then levels off, but Point 2 takes 158 s to level off, which is more than two-fold higher than 68 s of Point 2. Moreover, the temperature of Point 2 is 3–6 °C lower than that of Point 1. These phenomena demonstrate that the hybrid aerogels indeed outperform conventional thermal insulating materials.

Mechanical properties

Practical thermal-insulating applications require aerogels to be sufficiently strong and resilient to sustain compression loading and recover their original shape from deformation. To assess the mechanical properties of the aerogels, we perform cyclic compression tests under different maximum strains (ε). The GO aerogel shows very poor compressive resilience since it only recovers to 86% and 45% of its original length after compression at ε = 50% and 80%, respectively, as shown in Fig. 4a and b. Such a poor resilience property suggests that the compression results in partial collapse of the GO network.
image file: c7ta02686b-f4.tif
Fig. 4 Mechanical properties of the GO aerogel and the hybrid aerogels. (a) Shape recovery after a compression strain of 80% for the GO aerogel and GO/SEBS-2. (b) Cyclic stress–strain curves of the GO aerogel compressed to different maximum strains. (c) Cyclic stress–strain curves of GO/SEBS-2 compressed to different maximum strains. (d) Consecutive cyclic stress–strain curves of GO/SEBS-2 compressed to a maximum strain of 50%. (e) Energy-dissipation efficiency of the GO aerogel and GO/SEBS-2 compressed to different maximum strains. (f) Stresses at ε = 50% and 80% as a function of apparent density of the aerogels.

Hybridizing the GO network with SEBS introduces Networks II and III. The chemically crosslinked Network II is stable at temperatures lower than the decomposition point, so is the physically crosslinked Network III at temperatures lower than Tg of PS domains. Both Networks II and III are highly elastic due to the intrinsic elasticity of SEBS; this significantly improves the resilience property of the hybrid aerogel. For example, GO/SEBS-2 unfolds almost completely after 50% compression, and recovers to 94% of its original length after 80% compression (Fig. 4a and c). The fully reversible deformation at ε = 50% is further demonstrated by repeated cyclic compression tests, as shown in Fig. 4d. Although the compression stress decreases after the first loading–unloading process, the hybrid aerogel does not show any evident residual strain after 5 cycles of compression.

The cyclic compression loops of the GO aerogel show very large hysteresis, suggesting remarkable dissipation of mechanical energy (Fig. 4b). We quantify the energy-dissipation efficiency by dividing the integrated area of the hysteresis loop by that under the loading curve. The values of energy-dissipation efficiency range from 85 to 92% for the GO aerogel under various maximum strains (Fig. 4f), indicating that most of the mechanical energy is dissipated in the cyclic compression process. This high energy-dissipation efficiency can be attributed to the slide between GO layers and the collapse of the physical network of the GO aerogel. Similarly, Network I in the hybrid aerogels also undergoes a slide between GO layers and collapse of the network structure; this leads to the high energy dissipation of the hybrid aerogels. Quantitatively, the energy-dissipation efficiency varies from 58% to 82% for the hybrid aerogels as the maximum strain changes from 25% to 80%. This high energy dissipation caused by the structural change in Network I may lead to redistribution of stress concentration in a way similar to sacrificial bonds, while Networks II and III keep the integrity of the materials, thereby avoiding macroscopic failure of the hybrid aerogels.35–38 As a result, the hybrid aerogels are highly compressible and tough.

The hybrid aerogels are also stronger than the GO aerogel, as they show higher compression stress under the same strain. For example, GO/SEBS-2 has a compression stress of 32 kPa at ε = 50% and 151 kPa at ε = 80% that are nearly 3- and 4-fold higher than the corresponding values of the GO aerogel, respectively. This suggests that introducing Networks II and III reinforces the hybrid aerogels. By plotting the compression stresses at ε = 50% and 80% as a function of apparent density (Fig. 4f), we find that the stress at ε = 50% increases linearly with apparent density with a slope of 1.1, while that at ε = 80% shows a stronger linear dependence on the apparent density. However, when the apparent density is larger than 30 mg cm−3, the stress at ε = 80% shows a sudden increase; this may be attributed to the occurrence of densification at this strain, as the cell walls are easier to contact with each other at higher apparent densities.

Shape memory

In addition to the aforementioned requirements, thermal insulating materials usually need to be adapted onto surfaces with a complex shape or fitted into an otherwise compact space. It is advantageous if a thermal insulating aerogel has shape-memory properties for these practical applications. However, at present, it is still challenging to fabricate shape-memory aerogels. Integrating multiple networks may offer an alternative strategy to address this challenge. Indeed, the hybrid aerogels show a great ability of shape memory, as shown in Fig. 5a. To demonstrate the shape memory, we first heat a cylindrical sample of GO/SEBS-2 to a deformation temperature of 90 °C, slightly higher than the Tg of PS domains in SEBS. Subsequent cooling under a constant compression strain of 51.3% to room temperature causes the PS domains to adopt a glassy state that immobilizes Network III, thereby fixing the deformation as latent strain energy in Network II. Upon unloading at room temperature, the sample can maintain a strain of 48.7% since only a minor recovery occurs. To quantify the shape fixing, we define the percentage of shape fixity as Rf = εu/εm, where εu and εm represent the strains after unloading and the temporal strain achieved by deformation, respectively.39 A Rf value of 94.9% suggests near-complete shape fixing of the hybrid aerogels. In addition, the aerogels show zero Poisson's ratio since there is no radial expansion when the cylindrical sample is compressed along the axial direction to different maximum strains (ESI Fig. S5). The zero Poisson's ratio leads to linear volume shrinkage as a function of compression strain,6,40 thus allowing the aerogels to adopt a smaller volume upon temporal shape fixing.
image file: c7ta02686b-f5.tif
Fig. 5 Shape memory properties of the hybrid aerogels. (a) Shape fixation and recovery of GO/SEBS-2. The dashed line represents the temporal strain achieved by deformation. (b) Shape recovery of GO/SEBS-2 as a function of time under IR light. L0 is the initial length of the cylindrical sample, and L is the length after shape fixation or shape memory.

After applying a thermal stimulus (heating to T ≥ 90 °C), the permanent, “memorized” shape is recovered, as the strain energy in Network II is released. The residual strain is as small as 4.8%. We define the figure of merit of shape memory as Rr = (εuεp)/(εmεp), where εp is the permanent strain after heat-induced recovery.39 A Rr value of 94.4% indicates the high efficiency of shape memory for the hybrid aerogels. Consecutive shape memory cycling tests show that the shape memory can be repeated for at least five cycles without apparent deterioration in both Rf and Rr (Fig. 5a).

For outdoor thermal insulating applications, the thermal stimulus generated by direct heating is usually inaccessible. It requires that the shape memory can be triggered by a remote stimulus. Since GO is highly absorptive to infrared irradiation (IR), it is reasonable to use an IR stimulus to trigger shape memory. To demonstrate this idea, IR light with a power of 125 W is shed onto a temporarily fixed sample of GO/SEBS-2 that has a fixing strain of 51.7% after unloading at room temperature. Shape recovery is observed to increase with time and saturates with a residual strain of 4.5% at 60 s, as shown in Fig. 5b. The figure of merit of shape memory triggered by IR illumination is nearly the same as that activated by direct heating. The rapid shape recovery and high recovery efficiency open the potential for the practical outdoor applications.

Conclusions

We present that integrating multiple networks is a practical strategy for the fabrication of aerogels with multi-functionality. For a proof of concept, we integrate GO with SEBS to acquire hybrid aerogels. In the hybrid aerogels, the connecting GO nanosheets form Network I, the covalent bonds between GO and SEBS form Network II, and the phase separation of SEBS forms Network III. The triple networks are the key factor, which not only enable a solid thermal conductivity as low as ∼8 mW m−1 K−1, but also result in high mechanical properties, high hydrophobicity, shape memory and zero Poisson's ratio for the hybrid aerogels. The ultra-low thermal conductivity makes them suitable for an array of potential applications. For example, the fabricated hybrid aerogels may be utilized as next-generation building materials to assist in reducing energy consumption significantly by minimizing the energy exchange between the building indoor space and the outdoor ambient. This is particularly attractive for small space usage given the multi-functionality of the aerogels.

Experimental

Fabrication of hybrid aerogels

GO was synthesized by Hummers' method from graphite flakes with a mesh size of 325 (Alfa Aesar),41 and then dispersed in deionized water with a concentration of 5 mg ml−1 by vigorous agitation. 5 ml of aqueous GO dispersion was mixed with 20 μl of ethylenediamine; the mixture was sealed in a glass vial and then heated at 95 °C for 6 h to synthesize a GO hydrogel. The GO hydrogel was freeze dried to form a GO aerogel. Cylindrical samples of the GO aerogel were immersed in para-xylene solutions of SEBS with concentrations of 1% (wt/vol), 2% (wt/vol) and 4% (wt/vol) under vacuum. The solutions containing 10 μg ml−1 of 4-dimethylaminopyridine as a catalyst were heated at 65 °C overnight to allow the reaction between the maleic anhydride groups of SEBS and the amine groups on the functionalized GO nanosheets. Subsequently, cylindrical samples were taken out from the solutions and subjected to lyophilization to form GO/SEBS hybrid aerogels.

Characterization

The microstructure of the aerogels was characterized using a scanning electron microscope (SEM, Zeiss Ultra 55) at an acceleration voltage of 2 to 5 kV. Before measurements, the samples were sputter coated with a thin layer of Pt/Pb (8/2) film of 1 nm in thickness. Chemical structure changes were characterized by Fourier transform infrared spectroscopy (FTIR, Perkin Elmer FTIR spectrometer) with a mode of attenuated total reflectance. Elements of the aerogels were measured by X-ray photoelectron spectra (XPS, Thermo Scientific K-Alpha). Apparent densities were calculated by dividing the weight by volume for the aerogels. The mesopore structure of the aerogels was tested by nitrogen physisorption using a Micromeritics ASAP 2020 HD88 automated system. The samples of 0.1–0.2 g were first degassed at 100 °C for 12 h prior to nitrogen adsorption at −196 °C. BET analysis was carried out for a relative vapor pressure of 0.01–0.3 at −196 °C. Contact angles were measured using a contact angle measurement system (KSV CAM 101) at room temperature with a relative humidity of ∼20%. An infrared camera (FLIR T420) was used to record the temperature change of a cylindrical sample of GO/SEBS-2 imbedded in a sheet of expanded polystyrene foam with a thickness of 15 mm. The sheet was placed on a heating plate with a temperature of 48 °C and an environmental temperature of 17 °C. Uniaxial compression measurements were performed on an Instron 3342 with a 100 N load cell. The measurements were conducted at room temperature in air using a crosshead velocity of 2 mm min−1 to different maximum strains, including 25%, 50% and 80%. Further details of characterizations are available in the ESI.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (grant No. 51303116 and 51673120) and Sichuan University. We appreciate Prof David A. Weitz and Prof. Gang Chen for providing some of the research facilities.

Notes and references

  1. L. Pérez-Lombard, J. Ortiz and C. Pout, Energ. Build., 2008, 40, 394–398 CrossRef.
  2. D. Ürge-Vorsatz, N. Eyre, P. Graham, D. Harvey, E. Hertwich, Y. Jiang, C. Kornevall, M. Majumdar, J. E. McMahon, S. Mirasgedis, S. Murakami and A. Novikova, in Global Energy Assessment – Toward a Sustainable Future, Cambridge University Press, International Institute for Applied Systems Analysis, Cambridge, UK, New York, USA, Laxenburg, Austria, 2012, pp. 649–760 Search PubMed.
  3. Y. Gao, S. Wang, L. Kang, Z. Chen, J. Du, X. Liu, H. Luo and M. Kanehira, Energy Environ. Sci., 2012, 5, 8234–8237 CAS.
  4. B. P. Jelle, Energ. Build., 2011, 43, 2549–2563 CrossRef.
  5. M. M. Biener, J. Ye, T. F. Baumann, Y. M. Wang, S. J. Shin, J. Biener and A. V. Hamza, Adv. Mater., 2014, 26, 4808–4813 CrossRef CAS PubMed.
  6. Y. Wu, N. Yi, L. Huang, T. Zhang, S. Fang, H. Chang, N. Li, J. Oh and J. A. Lee, et al. , Nat. Commun., 2015, 6, 6141 CrossRef CAS PubMed.
  7. H. Sun, Z. Xu and C. Gao, Adv. Mater., 2013, 25, 2554–2560 CrossRef CAS PubMed.
  8. Z. Xu, Y. Zhang, P. Li and C. Gao, ACS Nano, 2012, 6, 7103–7113 CrossRef CAS PubMed.
  9. K. H. Kim, Y. Oh and M. Islam, Nat. Nanotechnol., 2012, 7, 562–566 CrossRef CAS PubMed.
  10. C. Li, L. Qiu, B. Zhang, D. Li and C. Y. Liu, Adv. Mater., 2016, 28, 1510–1516 CrossRef CAS PubMed.
  11. H. W. Liang, Q. F. Guan, L. F. Chen, Z. Zhu, W. J. Zhang and S. H. Yu, Angew. Chem., Int. Ed., 2012, 51, 5101–5105 CrossRef CAS PubMed.
  12. R. Li, C. Chen, J. Li, L. Xu, G. Xiao and D. Yan, J. Mater. Chem. A, 2014, 2, 3057–3064 CAS.
  13. C. Zhu, T. Y.-J. Han, E. B. Duoss, A. M. Golobic, J. D. Kuntz, C. M. Spadaccini and M. A. Worsley, Nat. Commun., 2015, 6, 6962 CrossRef CAS PubMed.
  14. S. Barg, F. M. Perez, N. Ni, P. do Vale Pereira, R. C. Maher, E. Garcia-Tuñon, S. Eslava, S. Agnoli, C. Mattevi and E. Saiz, Nat. Commun., 2014, 5, 4328 CAS.
  15. Y. Kobayashi, T. Saito and A. Isogai, Angew. Chem., Int. Ed., 2014, 53, 10394–10397 CrossRef CAS PubMed.
  16. J. Biener, M. Stadermann, M. Suss, M. A. Worsley, M. M. Biener, K. A. Rose and T. F. Baumann, Energy Environ. Sci., 2011, 4, 656–667 CAS.
  17. J. Li, J. Li, H. Meng, S. Xie, B. Zhang, L. Li, H. Ma, J. Zhang and M. Yu, J. Mater. Chem. A, 2014, 2, 2934–2941 CAS.
  18. H. Huang, P. Chen, X. Zhang, Y. Lu and W. Zhan, Small, 2013, 9, 1397–1404 CrossRef CAS PubMed.
  19. S. N. Schiffres, K. H. Kim, L. Hu, A. J. McGaughey, M. F. Islam and J. A. Malen, Adv. Funct. Mater., 2012, 22, 5251–5258 CrossRef CAS.
  20. K. J. Zhang, A. Yadav, K. H. Kim, Y. Oh, M. F. Islam, C. Uher and K. P. Pipe, Adv. Mater., 2013, 25, 2926–2931 CrossRef CAS PubMed.
  21. K. H. Kim, Y. Oh and M. F. Islam, Adv. Funct. Mater., 2013, 23, 377–383 CrossRef CAS.
  22. G. Li, X. Zhang, J. Wang and J. Fang, J. Mater. Chem. A, 2016, 4, 17042–17049 CAS.
  23. A. Mikhalchan, Z. Fan, T. Q. Tran, P. Liu, V. B. Tan, T.-E. Tay and H. M. Duong, Carbon, 2016, 102, 409–418 CrossRef CAS.
  24. B. Wicklein, A. Kocjan, G. Salazar-Alvarez, F. Carosio, G. Camino, M. Antonietti and L. Bergström, Nat. Nanotechnol., 2015, 10, 277–283 CrossRef CAS PubMed.
  25. H. Hu, Z. Zhao, W. Wan, Y. Gogotsi and J. Qiu, Adv. Mater., 2013, 25, 2219–2223 CrossRef CAS PubMed.
  26. R. Zheng, J. Gao, J. Wang and G. Chen, Nat. Commun., 2011, 2, 289 CrossRef PubMed.
  27. L. Qiu, J. Z. Liu, S. L. Chang, Y. Wu and D. Li, Nat. Commun., 2012, 3, 1241 CrossRef PubMed.
  28. X. Lu, M. Arduini-Schuster, J. Kuhn, O. Nilsson, J. Fricke and R. Pekala, Science, 1992, 255, 971–972 CAS.
  29. L. W. Hrubesh and R. W. Pekala, J. Mater. Res., 1994, 9, 731–738 CrossRef CAS.
  30. O. J. Lee, K. H. Lee, T. Jin Yim, S. Young Kim and K. P. Yoo, J. Non-Cryst. Solids, 2002, 298, 287–292 CrossRef CAS.
  31. D. Kraemer and G. Chen, Rev. Sci. Instrum., 2014, 85, 025108 CrossRef CAS PubMed.
  32. G. Chen, Phys. Rev. B: Condens. Matter Mater. Phys., 1998, 57, 14958 CrossRef CAS.
  33. G. Pernot, M. Stoffel, I. Savic, F. Pezzoli, P. Chen, G. Savelli, A. Jacquot, J. Schumann, U. Denker and I. Mönch, Nat. Mater., 2010, 9, 491–495 CrossRef CAS PubMed.
  34. M. Aegerter, N. Leventis and M. M. Koebel, Aerogels handbook, Springer Science & Business Media, 2011 Search PubMed.
  35. J. P. Gong, Soft Matter, 2010, 6, 2583–2590 RSC.
  36. E. Ducrot, Y. Chen, M. Bulters, R. P. Sijbesma and C. Creton, Science, 2014, 344, 186–189 CrossRef CAS PubMed.
  37. H. Y. Li, L. Yang, G. S. Weng, W. Xing, J. R. Wu and G. S. Huang, J. Mater. Chem. A, 2015, 3, 22385–22392 CAS.
  38. M.-C. Luo, J. Zeng, X. Fu, G. Huang and J. Wu, Polymer, 2016, 106, 21–28 CrossRef CAS.
  39. P. T. Mather, X. Luo and I. A. Rousseau, Annu. Rev. Mater. Res., 2009, 39, 445–471 CrossRef CAS.
  40. X. Xu, Q. Zhang, Y. Yu, W. Chen, H. Hu and H. Li, Adv. Mater., 2016, 28, 9223–9230 CrossRef CAS PubMed.
  41. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339 CrossRef CAS.

Footnotes

Electronic supplementary information (ESI) available: Mechanically robust and shape-memory hybrid aerogels for super-insulating applications. See DOI: 10.1039/c7ta02686b
These authors contributed equally to this work.

This journal is © The Royal Society of Chemistry 2017