Chapter 7: Porous Carbon and Carbon‐based Materials via Ice Templating and Freeze‐drying – Ice Templating and Freeze-Drying for Porous Materials and Their Applications

Porous Carbon and Carbon‐based Materials via Ice Templating and Freeze‐drying

7.1 Introduction

Porous carbon materials exhibit high chemical stability and high surface area and are used in a wide range of applications. A well‐known example of carbon materials are activated carbons that are usually fabricated by carbonization of polymers or biomass, followed by activation. Activated carbons are microporous materials (pore size <2 nm) and have been extensively used as adsorbents, as catalyst supports, in gas storage, etc. Like for other types of porous materials, the tuning of pore size, porosity, and surface functionality is crucial to optimize the potential of porous carbons for use in various applications. For example, while the micropores provide high surface area and large number of active sites, mesopores (2–50 nm) are important for mass transport of molecules, and macropores (>50 nm) may be essential for viscous systems, large molecules or tissue engineering.

Porous carbons may be fabricated by hard templates (e.g. porous silica, alumina, silica spheres), soft templates (e.g. surfactant assemblies, ionic liquids, liquid droplets, polymeric structures), or non‐templating methods (e.g. sol–gel process, hydrothermal process, self‐assembly) [17]. The pore surface of carbon materials can be specifically functionalized by incorporating heteroatoms in the synthesis, subsequent surface oxidation and activation, halogenation, and sulfonation, or via the coating of polymers and deposition of nanoparticles [8]. These carbon materials may be readily prepared as powders, monoliths, spheres [9], membranes [10] or fibres/nanofibres [11, 12]. Porous carbons can be produced from high carbon‐content synthetic polymers [10, 11], sustainable natural polymers/biomass [1, 6], polymeric gels [7], or metal organic frameworks [13]. Carbon nanotubes (CNTs) and graphene (and graphene oxides) have been intensively investigated since their discovery not long ago. This is attributed to their excellent individual properties and great potential in various applications [1416]. Importantly, they have also been used as building blocks to fabricate films, fibres, spheres, and three dimensional (3D) porous scaffolds [1720]. Composites and hybrid materials containing CNTs and/or graphene have also been produced in order to improve mechanical properties and induce synergy effects, and also to enhance their properties such as thermal or electrical conductivity [2125]. These carbon structures have found exciting applications in energy storage (e.g. rechargeable batteries, supercapacitors) [12, 19, 2628], sensing [29], fuel cells and solar cells [26, 28, 30], pollution management [18, 31, 32], catalysis [22, 33], and biomedical applications [20, 24, 34, 35].

Ice templating has been used as a simple but effective route to preparing a wide range of porous materials and nanostructures [36, 37]. As the name ‘ice templating’ suggests, the ice crystals formed during the freezing stage are used as templates, which are usually removed by a freeze‐drying process. Ice templating is highly effective for solutions, suspensions, and lightly crosslinked gels where molecules/particles/gel structures are excluded from the growing ice crystals. However, for strong gels or highly crosslinked wet polymeric or inorganic structures, it is very difficult for the growing ice crystals to change the original pore structure to a noticeable degree and hence the effect of ice templating is minimal (although this does not exclude the potential templating effect from the solvated solvent). For these types of gels or structures, the freeze‐drying process is employed to minimize the surface tension that is exerted on the pore structures during sublimation of the frozen solvent, prevent or significantly reduce pore structure collapse, and produce highly porous materials. For example, the freeze‐drying of resorcinol–formaldehydes is performed to produce highly porous cryogels, with porosity on a similar level as supercritical fluid drying but usually much better than air‐dried xerogels, unless a dense porous structure is required [38]. The organic cryogels can then be carbonized to produce carbon cryogels [39].

In this chapter, for each category of porous carbon materials described, both the freeze‐drying‐induced cryostructures and the ice‐templated structures are covered. The relevant applications and particularly the advantages arising from the freeze‐drying or ice‐templating procedures are discussed. Because of the intensive investigations of porous carbon materials for supercapacitors and rechargeable batteries, key properties and performance data are summarized in Tables 7.1 and 7.2 respectively.

Table 7.1 Supercapacitor performance of porous carbon and hybrids by ice templating/freeze‐drying.

MaterialsFabrication methodCapacitySurface areas/pore sizes/pore volumeReferences
Tannin‐F gelCarbon cryogel100 F g−1 in 4 M H2SO4 at scan rate of 2 mV s−1 by CV1420 m2 g−1, micropore volume 0.28 cm3 g−1[40]
S‐doped carbonCryogel from 2‐thiophene carboxaldehyde and resorcinol94.4 F g−1 in organic electrolyte, by charge/discharge1105 m2 g−1
Pore size 4.2 nm
Mesopore volume 0.46 cm3 g−1
Total pore volume 0.86 cm3 g−1
N‐doped carbonCryogel from melamine–formaldehyde 59.23 F g−1 in 5 M KOH by CV at scan rate of 10 mV s−1
62.13 F g−1 in 30% KOH by charge/discharge at the rate of 20 mA
88.44 m2 g−1 mesopore volume 0.18 cm3 g−1
10.23% N
N‐doped graphene carbonFor the above, 5% graphene oxide incorporated and thermally reduced 297 F g−1 by CV
313 F g−1 by charge/discharge
261 m2 g−1 mesopore volume 0.36 cm3 g−1
14.29% N
Porous carbonIce‐templated polyamic acid248 F g−1 at 0.5 A g−1 in 6 M KOH 2038 m2 g−1
0.9% N, Mesopore volume 0.26 cm3 g−1, total pore volume 0.86 cm3 g−1, pore size 0.7 nm
Porous carbonIce‐templating, SiO2 colloid templating and CO2 activation221 F g−1 at 2 mV s−1 in 1 M H2SO42096 m2 g−1 pore volume 3.0 cm3 g−1[44]
CNF microspheresCarbonized PDI gel284 F g−1 at 1A g−1 in 6 M KOH, 67 retaining after 1000 cycles at 4 A g−1 427 m2 g−1
3.61% N
Hg intrusion volume 10.43 cm3 g−1
CNF networksCarbonized PDI gel 192 F g−1 at 1 A g−1
346 F g−1 at 1 mV s−1
in 2 M H2SO4
520 m2 g−1[46]
Graphene aerogelGO reduced to form gel by L‐ascorbic acid and freeze‐dried128 F g−1 at 50 mA g−1 in 6 M KOH 512 m2 g−1,
2.48 cm3 g−1
Graphene aerogelGO reduced and functionalized by hydroquinone to form a gel and then freeze‐dried441 F g−1 at 1 A g−1 in 1 M H2SO4 297 m2 g−1,
Pore volume 0.95 cm3 g−1,
Pore sizes 2–70 nm
Graphene aerogel on Ni foamFreeze‐drying from soaked Ni foam366 F g−1 at 2 A g−1 in 6 M KOH[49]
Graphene paperInfiltration assembly and freeze‐drying137 F g−1 at 1 A g−1 in PVA/H2SO4 solid electrolyte148 m2 g−1[50]
VOPO4–grapheneIce templating, then reduced by hydrazine vapour528 F g−1 at 0.5 A g−1 in 6 M KOH[51]
NiCo2S4 nanotube@Ni–Mn LDH@grapheneIn situ growth in porous graphene1740 mF cm−2 at 1 mA cm−2[52]
Pristine graphene/MWCNTRoom temperature gelation and drying100 F g−1 at 100 A g−1 or 167 F g−1 at 1 A g−1, organic electrolyte700 m2 g−1, 10–200 nm, peaked at 83 nm[53]

Table 7.2 Porous carbon and hybrids as electrodes for rechargeable batteries (the batteries are LIBs, unless stated otherwise, SIBs for sodium‐ion batteries).

MaterialsFabrication methodCapacitySurface areas/pore volume/pore sizesReferences
V2O5–carbon cryogelElectrodeposition in cryogel280 mAh g−1 at 100 mA g−1, based on V2O5, >250 mAh g−1 after 20 cycles402 m2 g−1
Mesopore volume 0.42 cm3 g−1
Pore size 6 nm
SnO2–carbon cryogelImpregnation and thermal decomposition in cryogels590 mAh g−1 at 100 mA g−1 after 50 cycles 153 m2 g−1
Pore volume 0.514 cm3 g−1
Pore size 3.19 nm
S‐carbon compositesS infiltration into porous carbon prepared by ice templating, SiO2 colloidal templating and CO2 activation747 mAh g−1 at the C‐rate of 1C with organic electrolyte, for Li-S battery 2339 m2 g−1
Pore volume 3.8 cm3 g−1,
initial 1 : 1 C:S in mass
CNF webPyrolysed and activated cellulose fibres858 mAh g−1 at 100 mA g−1 after 100 cycles 1236 m2 g−1
Total pore volume 1.03 cm3 g−1
Graphene paperInfiltration assembly and freeze‐drying420 mAh g−1 at 2 A g−1148 m2 g−1[50]
Graphene/CNT spongeFreeze‐drying and thermal annealing436 mAh g−1 at 50 mA g−1 after 100 cycles for SIBs 498 m2 g−1
Pore volume 1.51 cm3 g−1
WS2/CNT–rGO aerogelFreeze‐drying the gel and thermal annealing 749 mAh g−1 at 100 mA g−1 for LIBs
311.4 mAh g−1 at 100 mA g−1 and 253 mAh g−1 at 200 mA g−1 after 100 cycles for SIBs

7.2 Carbon Cryogels and Ice‐templated Carbons

7.2.1 Carbon Cryogels

In order to produce carbon cryogels, polymeric (organic) gels are first formed, followed by a freeze‐drying process and a pyrolysis procedure. The polymeric gels are usually prepared via controlled hydrolysis and polycondensation reactions using different precursors. Figure 7.1 illustrates the molecular structures of some commonly used precursors. Resorcinol and formaldehyde are the mostly used system [7, 39]. Some other systems are also investigated, either to tune pore structures, to induce doping elements, or to improve sustainability by using sustainable precursors. These include phenol–formaldehyde [60], melamine–formaldehyde [42], resorcinol–2‐thiophenecarboxaldehyde [41], and tannin–formaldehyde [40].

Figure 7.1 Molecular structures of the common organic precursors used for the preparation of carbon cryogels.

For the resorcinol (R)–formaldehyde (F) system, the solvent can be water (W) or organic solvents such as acetone or alcohols of different chain lengths (e.g. methanol, ethanol, propanol, butanol). The catalyst (C) can be a base or an acid. Sodium carbonate (Na2CO3) seems to be the most commonly used base catalyst [7]. Sodium hydroxide (NaOH) is also frequently used. Hexamine can be used to introduce N into the carbon cryogel [39]. The R/F molar ratio of 1 : 2 is the widely used condition. The R/C ratios are usually in the range of 50–300. In general, lower R/C ratios lead to the formation of smaller polymer nanoparticles that are connected to form a porous structure with large necks, giving the fibrous gel. For higher R/C ratios, large polymer particles (∼15–200 nm) may be formed, resulting in a ‘string‐of‐pearls’ gel structure [7]. The C/W or R/W ratios can have a substantial effect on the pore size and porosity of carbon cryogels [7]. Particularly with hexamine as catalyst in an alcoholic solvent, the addition of water can decompose hexamine into formaldehyde and ammonia, resulting in a decrease of about 20% in porosity [39]. The type of solvents in RF gels can have a big impact on the pore structure of the resulting dry gels. If water is used as the solvent, it is usually replaced with a polar organic by solvent exchange, for reducing the damages on the pore structure during the drying procedure. When the freeze‐drying procedure is intended, tertiary‐butanol (t‐butanol) is usually used to replace other solvents before freeze‐drying. This is due to its high melting point and relatively high vapour pressure.

Tamon and co‐workers have extensively investigated the fabrication of carbon cryogels from RF gels [6164]. The reactant solutions (R/C = 25 or 200, R/W (g cm−3) = 0.125 or 0.250, R/F = 0.5) with sodium carbonate as catalyst were gelled at 25 °C for 1 day, 50 °C for 1 day, and then 90 °C for 3 days [61]. The RF gels were washed with water or t‐butanol and then freeze‐dried. The pyrolysis procedure was performed after aging the RF gels at 250 °C for 8 h, following a procedure of heating to 250 °C at 4.2 °C min−1, dwelling for 2 h, heating to 1000 °C at 4.2 °C min−1, dwelling for 4 h and then cooling down to room temperature, under a N2 flow of 80 cm3 min−1. Carbon cryogels with surface areas >800 m2 g−1 and mesopore volumes >0.55 cm3 g−1 were formed [61]. It was further found that freeze‐drying of the RF gels with t‐butanol could generate more mesopores, compared to freeze‐drying the RF gels with water [62]. The mesoporosity could also be controlled by changing the polycondensation conditions. The C/W ratio was found to be related to the mesoporosity. The mesopore volume, mesopore size, and surface area decreased with the increase of C/W when C/W was in the region of ≤70 mol m−3. It was hard to generate mesopores in the carbon cryogels when C/W was >70 mol m−3 [63]. In addition to cryogel monoliths, carbon cryogel microspheres were prepared via an inverse emulsion polymerization approach [64]. Aqueous reaction mixtures (R/C = 400 or 100, R/W = 0.25 or 0.50) were emulsified in cyclohexane with Span 80 as the surfactant. The resulting emulsions were stirred at 25 or 60 °C for 5–10 h in order to allow the gelation in the aqueous droplets [64].

Traditionally, it has been very difficult to prepare carbon cryogels with mesopores >40 nm. This is attributed to the low pH required for the synthesis and the difficulty in reproducing carbon cryogels when a small amount of Na2CO3 is used. This problem was addressed by the use of buffer solutions with pHs in the range of 10.0–12.0 (by adding 0.1 M HCl or 0.1 M NaOH into Na2CO3 solution). Carbon cryogels with large mesopores (14–108 nm) were successfully produced [65]. HCl aging of frozen RF gels could markedly increase the hydrophobicity of carbon cryogels [66]. This was achieved by immersing the frozen RF gels in 1 N aqueous HCl solution for 1–20 days at room temperature before washing with t‐butanol. The resulting carbon exhibited similar hydrophobicity as the coal‐derived activated carbon and pure β‐zeolite [66]. Effect of pyrolysis temperature and partial oxidation by oxygen were also investigated, which had an obvious impact on the adsorption of Pb2+ but minimal impact on the adsorption of methylene blue [67].

Modification or doping of carbon cryogels can be achieved by direct incorporation of chemical precursors in the synthesis mixture, impregnation of target entities into the porous carbon cryogels or chemical modification of carbon surface [7, 8, 39]. For example, N‐doped carbon cryogels were prepared by the reaction between melamine and formaldehyde [42] or with hexamine as catalyst [39]. The organic gels from resorcinol and 2‐thiophenecarboxaldehyde could be used to fabricate S‐rich carbon cryogels [41]. When ammonia borane was added to the RF gel during the solvent exchange stage, the prepared B,N‐doped carbon cryogel showed increased surface area and larger mesopore volume [68]. SnO2‐incorporated carbon cryogels were fabricated by impregnating the cyrogel with SnSO4‐ethnaol solution, followed by freeze‐drying and pyrolysis in N2 at 500 °C for 1 h. The diameters of the SnO2 particles were approximately 15 nm [55]. VOSO4 solution (pH 1.8) was infused into porous carbon cryogel films. The films were dried under vacuum at 80 °C and then subjected to potentiodynamic electrodeposition to produce V2O5‐incorporated composites [54]. Sulfonated carbon cryogels as acid catalysts were generated by treating the pyrolysed carbon with concentrated sulfuric acid at 80 °C for 10 h and subsequently 150 °C for 5 h under a dry N2 atmosphere [69].

Figure 7.2 Examples of carbon cryogels with different pore structures. (a) Randomly fibre‐like pore structure. (b) Aligned porous carbon structure. (c) The carbon cryogel with microhoneycomb structure.

Source: Yoshida et al. 2016 [142. Reprinted with permission from Springer Nature.

Source: Minović et al. 2015 [140. Reprinted with permission from Elsevier.

Source: Sepehri et al. 2009 [136. Reprinted with permission from Elsevier.

Carbon cryogels have been used for adsorption (water treatment), electrochemical energy storage, and catalyst support. These applications are enhanced by high surface area, large number of mesopores, and ice‐templated macropores in the carbon cryogels. Figure 7.2 shows the pore structures of some carbon cryogels. The ice‐templated aligned structure (Figure 7.2b) and the honey‐comb structure (Figure 7.2c) in the gels are the results of directional freezing of relatively lightly crosslinked RF gels.

Carbon cryogel microspheres were prepared via inverse emulsion polymerization with surface areas up to 752 m2 g−1, mesopore volume 0.97 cm3 g−1, and micropore volume 0.23 cm3 g−1. These microspheres were packed into a column for high performance liquid chromatography (HPLC) [64]. A similar approach was also employed to prepare carbon microspheres for adsorption of phenol and reactive dyes. The size of the mesopores in the microspheres did not impact the adsorption of phenol. This was likely due to the small size of phenol molecules. For a similar reason, the adsorption of Black 5 and Red E (larger size of molecules) was higher than activated carbon and increased with the size and the volume of mesopores [72]. However, the use of monolithic adsorbents may be easier for the adsorption of pollutants because the monoliths can be easily collected from the solution. Mesoporous carbon cryogels prepared from RF gels was used for the adsorption of As3+ (aqueous sodium arsenite solution). The adsorption capacity (∼2.1 mg g−1 carbon) was similar in the pH range of 2–11 [70]. The carbon cryogels with large mesopores 14–108 nm, prepared by controlling the pH via the use of buffer solution, could be used to distinguish proteins of different sizes. This was achieved by protein adsorption of different molecular weights into the mesopores of different sizes [65]. The hydrophobic carbon cryogels prepared via HCl aging was used to recover 1‐butanol from dilute aqueous solution (135 mM) at 37 °C. A high adsorption capacity of 3.13 mol kg−1 was achieved [66]. In general, the adsorption capacity increased with 1‐butanol concentration but changed slightly when the concentration was >200 mM [66]. The carbon monolith with a microhoneycomb structure (with straight channels of 25–25 μm in diameter) (Figure 7.2c) was fixed into a heat shrinkable tube. The tube was connected to a set‐up containing an HPLC pump and a UV–vis detector for the flow adsorption of phenol [71]. The highest adsorption capacity achieved was around 160 mg g−1 from aqueous phenol solution of 100 ppm [71]. The carbon microspheres obtained from phenolic resin were embedded in a poly(vinyl alcohol) cryogel and evaluated for the removal of pesticides atrazine (32 mg l−1) and malathion (16 mg l−1) in aqueous solutions. The flow through tests showed that the maximum adsorption capacities were 641 mg (atrazine) and 591 mg (malathion) g−1 carbon [73].

Carbon and doped‐carbon cryogels obtained from organic resins are widely used as porous electrodes for supercapacitors. Carbon cryogels obtained from phenol–formaldehyde (PF) resin were tested in 4 M H2SO4 aqueous electrolyte and specific capacitances of around 100 F g−1 were achieved. There was no direct link between the capacitance and the surface area/micropore volume. The highest capacitance obtained was related to mesopores and not too narrow micropores [60]. For the carbon cryogels prepared from tannin–formaldehyde resin, similar specific capacitance of about 100 F g−1 was achieved. However, this study found that the capacitance could be correlated to the micropore volume [40]. S‐rich carbon obtained from 2‐thiophenecarboxaldehyde–resorcinol resin could improve the wetting property of the electrode in organic electrolyte (1 M tetraethylammonium tetrafluoroborate in 50–50 propylene carbonate–dimethyl carbonate). Complete wetness could be achieved for the carbons with 2.5 at.% S. The best capacitance achieved was 94.4 F g−1 when the freeze‐dried resin was pyrolysed at 1000 °C [41]. N‐doped carbon from melamine–formaldehyde showed a specific capacitance of 59 F g−1 at scan rate of 10 mV s−1 in 5 M KOH aqueous electrolyte by cyclic voltammetry or 62 F g−1 at the current rate of 20 mA in 30% KOH solution by the charge/discharge approach. However, with the incorporation of 5% graphene oxide (then reduced to graphene by a thermal treatment), the specific capacitance could be increased to 279 F g−1 and 313 F g−1, respectively [42]. For the N,B‐doped carbon cryogels with higher mesoporosity, higher surface area, and more uniform mesopore size distribution, the specific capacitance (tested in the organic electrolyte as used in Ref. [41], up to ∼120 F g−1) could be correlated with the surface area [68].

Incorporation of metal oxide nanoparticles into carbon cryogels has been used to enhance the performance of lithium ion batteries (LIBs). For example, hydrous vanadium pentoxide (V2O5·nH2O) was formed in porous carbon cryogel by electrodeposition. Owing to the highly interconnected 3D porous structure, the composite electrode reached a discharge capacity of 280 mAh g−1 (based on the mass of V2O5) at a current density of 100 mA g−1. These values were about twice that of V2O5·nH2O on Pt foil. The capacity was still higher than 250 mAh g−1 after 20 cycles [54]. SnO2 nanoparticles (∼15 nm) were incorporated into carbon cryogel via impregnation and thermal decomposition of stannous sulfate. When used as an electrode for LIBs, a discharge capacity of 590 mAh g−1 after 50 cycles at a current density of 100 mA g−1 was recorded. This value was much higher than pure SnO2 nanoparticles, carbon cyrogel and their physical mixture. Again, this performance was attributed to the 3D porous structure and the stabilization of SnO2 particles in the mesopores [55].

Sulfonated carbon cryogels with ice‐templated microhoneycomb structure were used as solid acid catalysts for liquid phase reactions [69]. These catalysts exhibited surface areas of 570–670 m2 g−1, mesopore volumes of 0.2–1.5 cm3 g−1, concentrations of sulfuric acid group in the rage of 0.62–0.97 mmol g−1, depending on R/C (50, 200) and pyrolysis temperature (673 K, 1273 K). The catalyst prepared from R/C = 200 and pyrolysis temperature of 1273 K exhibited the best properties in the combination of 570 m2 g−1, 1.5 cm3 g−1, and 0.62 mmol g−1. By using these catalysts, higher reaction rate and higher selectivity (>90%) than Amberlyst‐15 were observed for esterification of acetic acid with ethanol and oleic acid with methanol, and particularly for the condensation reaction of furfural and 2‐methylfuran [69]. It was believed that the highly interconnected mesopores, larger mesopore volume, and the surface hydrophobicity contributed to the superior catalytic activities.

7.2.2 Ice‐templated Carbons

This section covers polymer solutions (instead of gels) that can be processed to produce porous carbons. Both aqueous solutions and organic solutions may be utilized. However, the selection of polymer is very important. The polymers should have high carbon contents and are not sublimed or melting during the pyrolysis process. Polymers that have been used to fabricate porous carbon or carbon membranes may be used in an ice‐templating process [10, 74]. Some of the commonly used polymers are given in Figure 7.3. Polyimides are formed by condensation reactions of dianhydrides and diamines. There are a range of polyimides. They usually show high‐melting point and high thermal stability. Because of aromatic rings in the polyimide molecules, these polymers are rigid and can be good precursors for production of carbon materials. Polyetherimide is illustrated in Figure 7.3. Another commonly used commercial polyimide is Kapton produced by DuPont [10]. Biomass is also widely used as a source of carbon materials. However, they are not covered in this section because they are insoluble in most of the solvents although alkaline lignin can be readily dissolved in water.

Figure 7.3 Molecular structures of commonly used polymers for ice‐templated carbons.

Polyacrylonitrile (PAN) is the mostly used polymer in the preparation of porous carbon or carbon nanofibres (CNFs). PAN was dissolved in dimethyl sulfoxide (DMSO) and the resulting solutions were frozen and freeze‐dried to produce aligned porous polymer. It was found that stable aligned porous monoliths could be formed when the concentrations were not less than 100 mg ml−1. It was essential to anneal the freeze‐dried PAN in air at 280 °C for 1 h before the pyrolysis process at 800 °C [75]. The aligned macroporous structure was retained in the carbon materials. N2 sorption study showed the surface areas to be below 200 m2 g−1 and lack of mesopores in the resulting carbon materials. The advantage of this method is that doping elements can be easily added in the initial PAN solution. This was demonstrated by incorporation of melamine and graphene oxides to produce N‐doped graphene–carbon composites, which resulted in a substantial increase of reversible capacity for LIBs (300 mAh g−1 at a high current density of 10 A g−1) [75]. After freezing the PAN–DMSO solution, the usual freeze‐drying step could be replaced with a solvent exchange process [76]. The ice‐templated structures were retained (Figure 7.4). Owing to relatively low surface areas after carbonization, varying amounts of KOH were incorporated into the porous PAN, followed by the same annealing and pyrolysing procedure. The incorporation of KOH was simply achieved by soaking porous PAN in aqueous KOH solutions. The highly interconnected macroporosity in PAN facilitated the impregnation of KOH. The activated ice‐templated carbons showed hierarchical porosity (Figure 7.4) with a surface area up to 2206 m2 g−1 and varying percentage of N content. Great potential was demonstrated for the uptake of CO2 and H2 by these N‐doped ice‐templated carbon materials. The carbon material with a surface area of 1049 m2 g−1 and 14.9% N gave a CO2 uptake of 16.12 mmol g−1 at 298 K and 10 bar while the carbon with the surface area of 2206 m2 g−1 and 1.29% N showed the adsorption of H2 (2.66 wt%) at 77 K and 1.2 bar [76].

Figure 7.4 Pore structure of the ice‐templated carbon prepared from polyacrylonitrile at various magnifications. (a) ×300. (b) ×700. (c) ×1400. (d) ×3700. (e) ×6500. (f) ×30k. Scale bar = 10 μm unless otherwise stated.

Source: Roberts et al. 2017 [152. Reprinted with permission from Royal Society of Chemistry.

Other polymers have also been used. One example is sodium poly(4‐styrenesulfonate) (PSS). PSS can be dissolved in water and the resulting aqueous solution can be conveniently freeze‐dried. In the study by Roberts et al. [77], it was found that a single step of thermal treatment at 800 °C under Ar was sufficient to produce ice‐templated porous carbon monoliths. Remarkably, the in situ generated Na2SO4 could act as an activating agent during pyrolysis which was further transformed to Na2S·9H2O. Following washing with acidic solution, sulfur was formed in situ within the porous carbon. By varying degassing temperature, the amount of sulfur removed from the carbon materials could be adjusted. Accordingly, the pore volume and the surface area of the carbon materials could be changed. A high specific surface area of 1051 m2 g−1 was achieved. The S content could be increased from 17.07 wt% to 39.74 wt% by incorporating additional Na2SO4 into porous PSS prior to pyrolysis [77]. Polyamic acid (PAA) is an intermediate polymer to produce polyimide. By carbonization of freeze‐dried PAA from its solution in 1,4‐dioxane, hierarchically porous carbon with ice‐templated macropores was produced [43]. The N content decreased with the increase of carbonization temperature. The carbon obtained by carbonization at 1000 °C exhibited a porosity of 90.8%, a surface area of 2038 m2 g−1 and a N content of 0.9% (by elemental analysis). When used as an electrode, a specific capacitance of 248 F g−1 was reached at 0.5 A g−1 in 6 M KOH solution [43]. Chitosan and benzoxamine (and montmorillonite also added to reinforce the aerogel) were freeze‐dried and then allowed for thermal crosslinking by stepwise treatment from 100 to 175 °C. The carbonized materials (at 800 °C under N2) showed excellent CO2 adsorption (up to 6.25 mmol g−1 at 298 K and 2 bar) [78].

Hard colloidal templating and ice templating have been combined to fabricate hierarchically porous carbon monoliths containing micropores, well‐defined mesopores and macropores. This was demonstrated by directional freezing and freezing‐drying of silica colloids–glucose suspension, followed with carbonization and HF etching to remove silica colloids. Different sizes of silica colloids, e.g. 4, 8, 12 nm, were used. This resulted in the well‐defined mesopores templated by silica colloids [44]. The specific surface area and the extent of micropores could be further increased by CO2 activation. The porous carbon, prepared from 4 nm silica colloids, 1 : 1 ratio of silica to glucose and 4 h of CO2 activation, showed a BET surface area of 2096 m2 g−1 and pore volume of 3.0 cm3 g−1 [44]. The large pores and high pore volumes of such carbon materials facilitated the impregnation of polyethyleneimine, which was then used for CO2 capture. A capacity of 4.2 mmol g−1 could be achieved for CO2 capture after 45 min. When evaluated for their use in supercapacitors, a specific capacitance of 221 F g−1 was recorded at a scan rate of 2 mV s−1 in a three‐electrode set‐up with 1 M H2SO4 as the electrolyte [44]. This method was also applied to the suspension of silica colloids (4 nm) and sucrose. The highest surface area of 2339 m2 g−1 and total pore volume of 3.8 cm3 g−1 were obtained for the porous carbon with 1 : 1 ratio of silica to sucrose and 2 h CO2 activation [56]. Sulfur was infiltrated into the porous carbon via melt infusion. The carbon‐S composite was tested for Li-S battery with 1 M LiTFSi and 1 wt% LiNO3 dissolved in a mixture of 1,3‐dioxlane and dimethoxyethane (1 : 1 v/v) as the electrolyte. A high‐specific capacitance of 747 mAh g−1 was obtained at the rate of 1C. This electrode was highly stable. High capacitances of 647 mAh g−1 and 503 mAh g−1 were still recorded after 200 cycles at high C‐rates of 2C and 5C [56].

Figure 7.5 (a) Poly(acrylonitrile) fibres prepared by electrospinning into liquid nitrogen. The inset shows the end of a broken fibre, clearly indicating a highly porous structure. (b) Porous carbon fibres obtained by carbonization of the poly(acrylonitrile) fibres.

Source: McCann et al. 2006 [158. Reprinted with permission from American Chemical Society.

7.3 Carbon Nanofibres

CNFs may be directly prepared by chemical vapour deposition (CVD) [12]. But the CVD approach requires high‐temperature and high‐vacuum equipment and it is extremely difficult to produce a large amount of CNFs. Carbonization of polymer fibres has been an effective approach to prepare CNFs. The polymer fibres may be fabricated by traditional spinning (e.g. melt spinning) but this method can only form fibres of larger diameters (>5 μm) [12]. Electrospinning has been extensively used to fabricate polymer fibres and CNFs [11, 12]. Control of fibre morphology and fibre diameter is highly important for the electrospinning method. This may be achieved by adjusting solution viscosity, surface tension, flow rate, electric conductivity and voltage [12]. The commonly used polymers for electrospun CNFs include PAN, pitch, lignin, PI and PVDF. Because of their low carbon yields (<15%), polyvinylpyrrolidone (PVP) and PVA are used mainly for metal oxide/CNF composites [12]. The electrospinning process does not usually involve a freezing step. However, instead of a conductive collection plate, highly porous fibres can be fabricated by electrospinning into a cryogenic liquid (e.g. a liquid nitrogen bath) [79]. As shown in Figure 7.5, the ice‐templated porous structure can be observed in the fibres and the highly porous CNFs are generated after thermal annealing and carbonization [79]. Aqueous PVA solution containing SnCl2 was electrospun. The as‐prepared sample was frozen in an icebox at −12 °C for 2 days. Subsequently, the frozen fibres were heated at 110 °C for 2 h and then 550 °C for 3 h in Ar/H2 atmosphere. This led to the formation of Sn/SnOx nanoparticles in the larger ice‐templated pores within the CNFs [80]. The volume of the particles was found to be less than half of the pore volume, which provided space to accommodate the volumetric change of Sn/SnOx nanoparticles during the LIB test. This contributed to the high capacitance of 735 mAh g−1 in the 1st cycle and 510 mAh g−1 after 40 cycles at the current density of 30 mA g−1 [80].

CNFs may be produced by hydrothermal and solvothermal reactions but quite often the products are powders or microspheres [6]. Using nanowires as template can facilitate the formation of CNFs in the form of monoliths. This was demonstrated by the use of Te nanowires in the hydrothermal carbonization of glucose with the production of large‐scale CNF cryogels [81]. Because this nanowire‐templating approach includes a step of removing the template, it may be more convenient to process polymer nanofibres directly and followed by carbonization. For example, aqueous chitin or wood cellulose nanofibre (∼10 nm) suspensions in t‐butanol at a concentration of 0.05 wt% were freeze‐dried and subsequently carbonized to produce CNFs with surface areas of 279 m2 g−1 [82]. Aqueous cellulose nanofibres suspension could be directly freeze‐dried. After pyrolysis at 600 °C, CNFs with diameters of ∼20 nm were formed [57]. CNF webs were prepared from freeze‐drying and pyrolysis of bacterial cellulose and were activated by KOH (700 °C under N2 1 h) to improve the surface area and induce mesopores to the CNFs [83]. The highest surface area achieved was 1236 m2 g−1 and pore volume 1.029 cm2 g−1 for the activated CNFs was obtained by KOH:cellulose = 6 (by mass). When evaluated as an anode for LIBs, a specific capacity of 857.6 mAh g−1 was recorded at a current density of 100 mA g−1 after 100 cycles [83].

It is possible to generate polymeric nanofibres by direct freeze‐drying of aqueous polymer solution [84, 85]. The key to the success of this approach is the use of diluted polymer solutions, e.g. in the range of 0.05–1 wt%, depending on the type and molecular weight of the polymers used. The diameters of the fibres are usually in the range of 100–800 nm. A directional freezing procedure can be applied to fabricate aligned polymer nanofibres. When a water‐soluble polymer of high carbon content, such as lignin, is used, CNFs can be readily formed by carbonization of the polymer fibres [86]. This was convincingly demonstrated by a study from Spender et al. [45]. More importantly, by injecting the lignin solution (0.2 wt%) onto a rotating drum cooled by a pressurized liquid nitrogen tank, they showed it was possible to prepare lignin nanofibres continuously and hence the synthesis of large amount of CNFs [45].

Figure 7.6 The scheme shows the synthesis of perylene diimide derivatives (PDI‐COOH) and the preparation steps for nanofibrous microspheres. The fibrous monoliths can be readily formed via the gelation of a bulk solution (see Figure 7.7).

Source: Liu et al. 2015 [92. Reprinted with permission from Royal Society of Chemistry.

Figure 7.7 The morphologies of nanofibrous gel and carbon monoliths prepared from dissolving the PDI‐COOH in water with triethyl amine. (a) and (b) The gel monolith and the carbon monolith. (c) and (d) N‐doped gel and carbon monoliths by adding melamine in the PDI‐COOH solution. Insets show the photos of the relevant monoliths, scale bars = 5 mm.

Source: Liu et al. 2015 [174. Reprinted with permission from Royal Society of Chemistry.

Low‐density polymeric gels may exhibit nanofibrous networks, which can serve as templates for CNFs. However, there are two limiting issues: (i) quite often these polymers have low carbon content and are thus unsuitable for carbonization; (ii) the diameters of the fibres are not uniform, particularly at the fibre junctions. Zhang and co‐workers proposed to form organic nanofibres via pH‐triggered self‐assembly of perylene diimide derivatives (PDIs) and then carry out carbonization to produce CNF networks [46, 87]. PDIs are multi‐aromatic molecules with an extended quadrupolar π system and strong hydrophobicity. The pH‐triggered self‐assembly is driven by hydrophobic interaction, π–π stacking and interaction of side chains. Figure 7.6 illustrates the synthesis of PDI‐COOH that is used as a precursor to form nanofibrous gels. The PDI‐COOH was dissolved in basic water, assisted by the addition of triethylamine. Glucono‐δ‐lactone (GdL) or freshly prepared aqueous GdL solution was added to the PDI solution. The slow hydrolysis of GdL reduced the solution pH and initiated the gelation. This helped to form a uniform hydrogel, compared to the direct addition of acid solution. A monolithic gel could be readily formed, showing a dark red colour, with the uniform fibre diameters around 25 nm. A washing procedure in the order of water, acetone and cyclohexane and the subsequent freeze‐drying resulted in red nanofibrous monoliths (Figure 7.7). This treatment also facilitated the preparation of CNF monoliths (Figure 7.7). An advantage of this method is that different precursors may be included in the gelation procedure to prepare modified CNFs. For example, melamine was added to make N‐doped CNFs (Figure 7.7c and d). K2SnO3 was used to replace triethylamine to make the solution basic and also introduce SnOx into the CNF monoliths. Non‐ionic surfactant F‐127 was added to improve the surface area of the CNFs (up to 520 m2 g−1). When tested as the electrode material for a supercapacitor, a specific capacitance of 346 F g−1 at the scan rate of 1 mV s−1 or 192 F g−1 at the current density of 1 A g−1 in 2 M H2SO4 was achieved. Remarkably, the specific capacitance was increasing with the test cycles, achieving a capacitance of 226 F g−1 at 4 A g−1 after 1000 charge/discharge cycles [87]. Unlike the electrospinning method, this gelation approach can be combined with emulsification to fabricate nanofibrous microspheres (Figure 7.6). Aqueous PDI solution containing GdL was emulsified into o‐xylene with Span 80 as the surfactant and polystyrene was included as stabilizer. The gelation was initiated in the aqueous droplets to form gel microspheres. The gel microspheres were then collected by centrifuging/filtering and were washed, dried and carbonized to produce nanofibrous carbon microspheres [46]. These microspheres were also evaluated as the electrode for supercapacitors. A high‐specific capacitance of 284 F g−1 at 1A g−1 and good rate capability (retaining 57% after 1000 cycles at 4 A g−1) were recorded [46].

Figure 7.8 Different forms of carbon nanomaterials.

Source: Mauter and Elimelech 2008 [62. Reprinted with permission from American Chemical Society.

7.4 Carbon Nanotubes (CNTs) and CNT‐based Materials

7.4.1 Introduction to CNTs

There are different forms of carbon nanomaterials, from nanodiamonds with tetrahedron sp3 configuration to graphene with the planar sp2 conformation (Figure 7.8) [31]. Theoretically, an sp2‐hybridized CNT can be constructed by rolling up a hexagonal graphene sheet. This can lead to chiral and non‐chiral arrangement of carbon atoms [88]. The non‐chiral CNT can take the armchair or zig‐zag configuration, as illustrated in Figure 7.9 [21]. The transformation induced by the stress along the axial direction of the armchair tube, resulting in the structure with two pentagons and two heptagons in pair, is called the Stone–Wales transformation. This transformation contributes to ductile fracture in armchair CNTs.

Figure 7.9 Schematic representation of the C arrangement in a carbon nanotube. (a) The armchair structure. (b) The zig‐zag structure.

Source: Thostenson et al. 2001 [42. Reprinted with permission from Elsevier.

There are single‐walled, double‐walled, and multi‐walled CNTs. CNTs may be produced by arc discharge and laser ablation and the widely used CVD [89]. CVD is the main approach used to manufacture CNTs, by cracking gaseous hydrocarbon over transition metal catalysts. The carbon is dissolved in the metal and then grows from the metal to form CNTs and also CNFs. The commonly used hydrocarbons are methane, ethylene, acetylene, benzene and xylene while the catalysts are mostly Fe, Co and Ni due to their high solubility and high diffusion rate to carbon atoms [89]. CNTs have excellent mechanical, electronic and thermal properties. For example, a Young's modulus of 0.9 TPa and tensile strength of 150 GPa, electrical conductivity of 107 S m−1 and thermal conductivity of 3500 W (m K)−1 at room temperature, have been reported [26]. Therefore, due to the nanoscale feature and excellent properties, CNTs have been used and have great potential in a wide range of applications such as adsorption of metal ions and environmental applications [18, 90], energy conversion and energy storage [26, 91], catalysis [89] and biomedical applications [24, 91]. Some of the applications are highly promising and, therefore, may be exploited for commercial applications, e.g. CNT yarns and sheets for supercapacitors, sensing and actuators [92, 93].

7.4.2 CNT Aerogels/Cryogels

CNTs and CNT films may be directly produced by the CVD approach. However, solution‐based CNT processing approaches are much more widely used in the preparation of CNT films, yarns and 3D structures [28]. However, surfactant and/or polymer stabilizer are usually required to form stable CNT dispersion, which is critical to produce uniformly distributed CNT structures. Therefore, the direct processing of CNT‐only dispersion is rare. Usually, after the CNT structures are formed, the polymer or surfactant may be removed by washing or thermal treatment. For the term ‘aerogels’, they usually indicate ultralight and highly porous materials fabricated by either supercritical fluid drying or freeze‐drying [94].

The ice‐templating approach has been used to fabricate CNT aerogels or foams. For example, CNT aerogels were prepared with sodium dodecylbenzene sulfonate [94]. Macroporous CNT foams with a porosity of 97% were produced by freeze‐drying. Sodium carboxymethylcellulose (SCMC) was used to stabilize the CNTs in the aqueous dispersion [95]. However, these freeze‐dried CNT structures are mechanically very weak. Chemically crosslinking the CNTs in the 3D structures is an effective route to improving mechanical stability. Ajayan and co‐workers used the Suzuki cross‐coupling reaction between aryl halide and boronic acid derivatives to crosslink the CNTs [96]. Before the reaction, the CNTs were washed with concentrated HNO3 under reflux to remove impurities and introduce the -COOH groups. The CNT‐COOH was then converted to CNT‐COCl by reacting with SOCl2, which could then be crosslinked by 1,4‐phenyldiboronic acid. The crosslinked CNT gels were then freeze‐dried to produce the covalently connected CNT networks [96]. The CNT‐COCl could also be functionalized by allylamine monomers and then crosslinked by a free radical polymerization initiated by azobisisobutyronitrile (AIBN). A freeze‐drying process was then employed to fabricate the chemically crosslinked CNT networks [97]. Mechanically stable porous CNT beads were produced by dropping aqueous CNT dispersions (containing CNT‐COOH, glutaraldehyde, resorcinol, borax) into liquid nitrogen and subsequently freeze‐drying [98].

7.4.3 CNT‐based Porous Materials

CNT–polymer materials are mainly fabricated by the ice‐templating approach. One of the main advantages is to enhance mechanical stability. For example, the freeze‐dried CNT aerogels were fragile, just enough for careful handling while the PVA‐enhanced CNT aerogel could support 8000 times of its own weight [94].

The introduction of polymer may also offer some properties that promote the applications of CNT‐based materials, as has been demonstrated by the CNT–chitosan structures. Chitosan is probably the most widely used polymer in making porous CNT–polymer structures. CNT may be directly dispersed in aqueous chitosan solution. Alternatively, CNTs may be first purified and functionalized by HNO3 washing. In the latter case, the -COOH groups or other oxygen‐containing groups can have stronger interaction with the positive amine groups on chitosan. This facilitates the formation of stable porous CNT–chitosan structures. CNT–chitosan materials have been investigated as scaffolds for cell culturing and tissue engineering. CNT toxicity, e.g. asbestos‐like behaviour, has been a big challenge for biomedical applications. The incorporation of biocompatible chitosan into CNT structures has certainly benefited their biomedical applications [24]. Multi‐wall carbon nanotubes (MWCNTs)–chitosan scaffolds showed good cell adhesion, viability and proliferation when evaluated with C2Cl2 cell line (myoblastic mouse cell). From the in vivo tests, the adsorption potential of MWCNTs was exploited to incorporate rhBMP‐2 into the scaffold, which promoted cell colonization and tissue growth [99]. This scaffold was also compared with chondroitin sulfate–MWCNTs and gelatin–MWCNTs, based on the evaluations using three types of mammalian cells (L929 fibroblasts, Saos‐2 osteoblasts, ECPC cells). It was concluded that the preparation parameters and scaffold properties could be optimized to serve better for the target cells [100]. Chondroitin sulfate is a major regulatory component of nerve tissue and has been used to fabricate ice‐templated scaffolds with MWCNTs [101, 102]. Viable cultures enriched in neuron cells for up to 20 days could be formed, displaying calcium transients and active mitochondria [101]. It was found that the biocompatible MWCNT–chitosan scaffolds could promote the immobilization and proliferation of bacteria such as Escherichia coli within the ice‐templated pores [103]. This kind of materials may be used as electrodes in microbial fuel cells (MFCs). This was demonstrated by the colonization of electroactive bacteria Geobacter sulfurreducens within the 3D microchannelled MWCNT–chitosan scaffold; the pore structure of the scaffold shown in Figure 7.10. The resulting materials were used as a flow through bioanode for the MFC and produced acetate oxidation current densities of up to 24.5 A m−2 and a maximum powder density of 2.87 W m−2 [104].

Figure 7.10 (a) Macroscopic monoliths of different shapes composed of MWCNT/chitosan fabricated by the ice‐templating process. And the pore structures at different magnifications; scales are 1 mm (b), 200 μm (c), 20 μm, (d) and 5 μm (e).

Source: Katuri et al. 2011 [208. Reprinted with permission from Royal Society of Chemistry.

The hierarchical porosity and highly exposed surface of CNTs in CNT–chitosan scaffolds facilitate adsorption of pollutants or other target components [105]. Despite the high porosity, CNT–chitosan scaffolds or compressed powders can be highly conductive due to the presence of CNTs. A percolation threshold was found to be ∼2.5 wt% MWCNTs [106]. The conductivity of such scaffolds facilitates their use as electrodes. For example, a macroporous MWCNT–chitosan scaffold prepared by the ice‐templating process, with surface‐decorated Pt nanoparticles, was used as the anode for methanol fuel cell. The electric conductivity was recorded as 2.5 S cm−1, resulting from a high content of MWCNTs. Current densities of up to 242 mA cm−2 were obtained [107]. As expected, CNT–chitosan composite foams showed enhanced elastic property. The recovery ability of the composite foams increased with the increasing CNT content until a critical point. After this point, the recovering ability after cyclic compression tests was decreased dramatically. This was attributed to the inadequate chitosan presence and the aggregation of CNTs [108].

Other polymers and additives have also been employed to produce CNT‐based materials, achieving improved mechanical stability while still maintaining the electrical and thermal properties. Cotton cellulose was dissolved in aqueous basic CNT suspension with Brij 76 used as a stabilizer. The Young's modulus of the freeze‐dried composites was measured to be up to 90 MPa. The electrical conductivity was in the range of 2.3 × 10−4–2.2 × 10−2 S cm−1 for the composites containing 3–10 wt% CNTs [109]. MWCNT–silk fibroin scaffolds with randomly porous and aligned porous structures were produced by gelation and flash freezing and directional freezing of the suspensions, respectively [110]. The aligned MWCNT–silk fibroin scaffolds showed superior thermal stability and electrical conductivity [110]. Aqueous suspensions containing MWCNT and polyurethane were processed via the directional freezing approach to fabricate light and aligned porous composites [111]. In addition to the enhanced mechanical properties (compared to porous CNT and polyurethane), the composites showed high electromagnetic interference (EMI) shielding effectiveness, exceeding 50 or 20 dB in the X‐band with the density only being 126 or 20 mg cm−3, respectively [111]. V2O5 powders were suspended in aqueous H2O2 solution, which formed a gel upon continuously stirring. Aqueous CNT dispersion was then added to the gel with vigorous stirring. The sheet‐like V2O5–CNT composite formed after freeze‐drying was treated at 300 °C for 1 h and then evaluated as a cathode for LIBs. A discharge capacity of 240 mAh g−1 was obtained at the rate of 5C, and 71% of the capacity was retained after 300 cycles [112]. NiFe2O4 nanoparticles (20 nm in diameter) were dispersed in aqueous chitosan and CNT dispersion. The freeze‐drying process was then employed to prepare the composite sponges. After dropping glucose oxidase solution onto the sponge to prepare a glucose biosensor, two large linear ranges (0–3.0 and 3.2–12.4 mM) with sensitivities of 84.1 and 24.6 μA (mM cm2)−1 were obtained [113].

The polymer phase may be infiltrated into a freeze‐dried porous CNT scaffold to make a composite. For example, liquid epoxy resin was infiltrated into aligned porous CNT scaffold. The curing of epoxy resin was completed at 140 °C for 12 h. The lamellar structure of the original CNTs rendered the composites highly conductive. An electrical conductivity of 0.158 S cm−1 at a CNT loading of 1.31 wt% was observed while the Young's modulus and yield strength were improved at 25.5% and 12.2%, respectively, compared to pure epoxy resin [114]. PAN‐based carbon fibre cloth was soaked with CNT suspension and then freeze‐dried. The freeze‐dried cloth was thermally treated at 400–650 °C under Ar. Subsequently, epoxy resin was impregnated into the CNF/CNT structure to produce the composite with good electrical conductivity and mechanical strength [115]. Liquid polydimethylsiloxane (PDMS) precursor was infiltrated into an MWCNT foam at room temperature under vacuum. The curing of the liquid precursor was completed to produce the CNT–PDMS composite. The composite showed a high thermal conductivity of 0.82 W (m K)−1, about 455% that of pure PDMS and 300% higher than that of the physically blended composites. Both the tensile strength and elongation at break were superior to those of the blended composite as well. This property made it a good candidate for thermal management of electronic devices [116].

7.5 Porous Graphene Networks and Graphene‐based Materials

7.5.1 Graphene and 3D Graphene Networks

Graphene exhibits excellent properties such as very high Young's modulus (∼1100 GPa), fracture strength (125 GPa), thermal conductivity (∼5000 W (m K)−1), mobility of charge carriers (200 000 cm2 (V s)−1), specific surface area (calculated value of 2630 m2 g−1) and the quantum Hall effect [15, 16]. Graphene may be produced by four different methods: (i) CVD and epitaxial growth; (ii) mechanical exfoliation of graphite or the well‐known ‘Scotch tape’ method; (iii) epitaxial growth substrates and (iv) preparation of graphene suspensions [16]. Among the four methods mentioned, the first three methods can produce high‐quality graphene but the quantity is very small. Preparation of graphene suspension is an effective route to addressing this issue with regard to large‐scale synthesis and application. Graphene is the atom‐thick sheet of hexagonal sp2‐bonded carbon network (Figure 7.8). Owing to the high surface area, hydrophobic interaction and π–π stacking potential, graphene tends to aggregate. So it is impractical to prepare graphene suspension directly. Usually, graphite is oxidized by strong oxidants to produce aqueous graphene oxide dispersion. Graphene oxide is basically a graphene sheet with O‐containing groups bonded to the carbon atoms, including hydroxyl, epoxide, carboxylic and carbonyl groups, as illustrated in Figure 7.11 [117]. Graphene oxide (GO) sheets are hydrophilic and negatively charged, which promotes their stable dispersion in water. Polymers or surfactants are usually required to stabilize GO in water. GO can be chemically or thermally reduced to produce graphene (or called chemically converted graphene) under the stabilization of polymer/surfactant usually in a gel or solid state [17, 19, 29]. However, aqueous graphene dispersion can be produced as demonstrated by hydrazine reduction in the presence of ammonia [118].

Figure 7.11 The chemical structure of graphene oxide.

Source: He et al. 1998 [234. Reprinted with permission from Elsevier.

Porous graphene or 3D graphene networks combine the excellent properties of graphene, highly interconnected porosity, and the easy manipulation or manufacturing for a wide range of applications [17, 19, 29, 119, 120]. Various approaches have been developed to fabricate 3D graphene materials including hydrothermal gelation [17], self‐assembly under chemical reduction [17], modified GO and subsequent crosslinking/polymerization [17, 18], template‐directed assembly [19, 29] and 3D printing [121]. Graphene fibres can be fabricated by wet‐spinning. The anodized aluminium oxide (AAO) may be used as a template to prepare graphene tubes while polymer colloidal particles are used as templates to prepare graphene spheres [19]. In addition, the ice‐templating or freeze‐casting approach has been used to fabricate porous graphene with some unique properties [1719, 119, 120]. This is the focus in the following sections.

7.5.2 Porous Graphene by Ice Templating/Freeze‐drying

Pristine graphene may be used directly to prepare graphene aerogels. For example, some solvents that are solid at room temperature but have relatively high vapour pressure were used to disperse graphene flakes at elevated temperatures. When cooling down to room temperature, the dispersions turned into solids. The solvents were conveniently removed by sublimation at room temperature [53]. Phenol (melting point 40.5 °C, vapour pressure 47 Pa at room temperature) and camphene (melting point 51.5 °C, vapour pressure 400 Pa) can be used as such solvents. The produced graphene aerogels showed good elasticity and electrochemical properties [53].

However, the preparation of most porous graphene structures starts with aqueous GO suspensions that are usually prepared by a modified Hummers method [16]. The control of freezing processes and preparation parameters that have been employed for other ice‐templated materials can be similarly applied for the preparation of porous graphene materials [36, 37]. Based on the methods reported in literature, the use of ice templating or freeze‐drying for graphene materials can be classified into five categories as described below.

The first approach is by directly freezing the GO suspensions and subsequent freeze‐drying [122126]. Directional freezing, flash freezing, or slow freezing may be employed to produce graphene materials with either aligned porous structures or randomly porous structures. The GO may be partially reduced in the suspension before freezing, which leads to partial self‐assembly of GO sheets in the suspension and better elasticity and stability of the freeze‐dried graphene structures [119, 120]. Figure 7.12 shows the scheme and the typical pore structures of cellular graphene monolith prepared by freezing a partially reduced GO suspension. It was found that the amount of oxygen‐containing groups on GO could significantly influence the interactions between GO sheets. The superelastic porous graphene monolith could be produced with the C/O atomic ratio of around 1.93 in the partially reduced GOs [127]. Note that in Figure 7.12 there is a thawing procedure that facilitates the annealing and assembly of GO sheets. Indeed, multiple freeze‐drying procedures may be utilized to improve the properties of graphene materials. For example, the thermally reduced GO cryogel was soaked in KOH solution, freeze‐dried, activated at 973 K under Ar, washed with dilute HCl and freeze‐dried again. This produced an aligned porous graphene aerogel with additional mesopores and micropores [128]. The freeze‐dried graphene structure may be directly formed in a porous scaffold or on a porous grid. This can be simply achieved by dipping the template in a GO suspension and subsequently freeze‐drying the GO‐soaked scaffold. An example is the formation of graphene aerogel in nickel foam [49].

Figure 7.12 Porous graphene materials fabricated by directional freezing of GO suspension and subsequent freeze‐drying. (a)–(c) Representative pore structures. (d) The preparation scheme.

Source: Qiu et al. 2012 [254. Reprinted with permission from Nature Publishing Group.

The second method is the freeze‐drying of GO hydrogels. Such an example is given in Figure 7.13. The GO hydrogel was formed by heating aqueous dispersion of GO with hydroquinone at 100 °C for 12 h. The hydrogel was then freeze‐dried to generate a highly porous graphene aerogel [48]. Similarly, a GO gel was formed by heating its aqueous dispersion containing ascorbic acid (1 : 1, w/w) at 70 °C for 4 h and then freeze‐dried [129]. Indeed, all kinds of GO hydrogels formed by hydrothermal treatment or chemical reductions [1719] may fall into this category.

Figure 7.13 Porous graphene monolith fabricated by freeze‐drying of a graphene (GO) gel. (a) Photos of GO dispersion, functionalized GO hydrogel (FGH) in water and picked up using a tweezer, and the freeze‐dried aerogel. (b) and (c) The pore structure of the aerogel.

Source: Xu et al. 2013 [48]. Reprinted with permission from John Wiley and Sons.

The third method is to combine ice templating and emulsion templating to fabricate porous graphene. This approach has the potential to generate porous materials with very high pore volume and systematically tuned porosity [130, 131]. Figure 7.14 illustrates how this approach is employed to prepare porous graphene materials. An oil phase (e.g. toluene) is emulsified into aqueous chemically‐modified graphene (CMG, or GO) dispersion containing PVA and sucrose (stabilizer and facilitating the freezing process) to form an oil‐in‐water emulsion. This emulsion can be directly frozen and freeze‐dried to produce dry porous graphene (Figure 7.14a). The emulsion‐templated cellular pores can be clearly observed in Figure 7.14b. Highly porous fibres may be formed by injecting the emulsion and then freeze‐dried (Figure 7.14c) [132]. Air bubbles or air‐in‐water emulsion may be formed with surfactants (e.g. sodium dodecyl sulfate, or non‐ionic polymeric surfactant) [133] or a conventional detergent [134] and then utilized as templates to fabricate porous graphene. After freeze‐drying, a thermal annealing procedure as well as the removal of surfactant may be necessary to retain the graphene's excellent properties [133].

Figure 7.14 (a) Porous graphene scaffolds by freeze‐drying of oil‐in‐water emulsions with aqueous GO (or chemically modified graphene, CMG) dispersion as the continuous phase. (b) The overview of the pore structure. (c) Porous wires by injecting the emulsion.

Source: Barg et al. 2014. Reprinted with permission from Ref. [264.

The fourth method is the preparation of porous graphene paper via filtration assembly of GO sheets and freeze‐drying (Figure 7.15). GOs may be partially reduced before filtration (Figure 7.15) [135] or directly infiltrated with a membrane under vacuum [50]. The freeze‐dried GO paper can be thermally reduced and further reduced with ascorbic acid solution to generate graphene paper.

Figure 7.15 Porous graphene papers by freeze‐drying the wet film formed by filtration assembly of GO suspensions. (a) The schematic representation of the filtration assembly process. (b) The pore structure of the graphene paper.

Source: Shao et al. 2016 [270. Reprinted with permission from John Wiley and Sons.

The final approach is to fabricate porous graphene microspheres by electrospraying aqueous GO suspension into a cryogenic liquid bath (e.g. liquid nitrogen, cooled n‐hexane), as schematically described in Figure 7.16 [136]. The pore structure may be tuned by the concentration and reduced state of GOs while the diameters of the microspheres can be adjusted by the flow rate of electrospraying. It was observed that higher flowing rate resulted in larger microspheres [136]. Similarly to spraying suspensions or emulsions into liquid nitrogen to make porous microspheres [130], aqueous GO suspensions or emulsions may be directly sprayed or atomized into liquid nitrogen (or other cryogenic liquids). The frozen droplets can then be collected and freeze‐dried to generate GO microspheres that may be further thermally annealed and reduced to form porous graphene microspheres. The size of the nozzles and the atomizing pressure can be varied to change the size of the microspheres. Higher atomizing pressure usually leads to the formation of smaller spheres.

Figure 7.16 Porous graphene spheres by electrospraying into a cryogenic liquid. (a) The schematic preparation process. (b) The optical microscopic image of graphene spheres.

Source: Liao et al. 2016 [272. Reprinted with permission from Royal Society of Chemistry.

Porous graphene monoliths show superelastic behaviour. The cellular monoliths prepared from partially reduced GOs could maintain the structural integrity under loads that were >50 000 times their own weight and rapidly recover from >80% compression [127]. The graphene networks fabricated from both ice‐templating and emulsion‐templating showed 98% and 95% recoverable deformation after 1 cycle and 10 cycles of compression, depending on the density of the materials. The Young's modulus increased with the increasing density, reaching a few megapascals [132]. The air‐bubbled graphene foams exhibited reliably compressive stress of 5.4 MPa and strain as high as 99% at least for 1000 cycles during the compress/release test [133]. Unlike conventional rigid materials and CNT‐based polymeric materials, the electrical resistance of the superelastic graphene networks decreases with the compressive strain increments of up to 60%, in a linear fashion. This property makes graphene foam an excellent candidate for strain sensing, for example, in human health monitoring [124]. This piezoresistive response can be ultrafast and provides instantaneous and high fidelity electrical response to dynamic pressures from quasi‐static to 2000 Hz or ultralow pressure of 0.082 Pa [137]. This is illustrated in Figure 7.17. It can be seen that the voltage response is spontaneous to the change of stress at frequencies of 20 and 1000 Hz (Figure 7.17a and b). The response sensitivity is generally inversely proportional to the density. The high response sensitivity is achieved from the cellular graphene with a density of 0.54 mg cm−3 (Figure 7.17c). This material can detect a very low pressure applied (Figure 7.17d).

Figure 7.17 Dynamic piezoresistive behaviour of the superelastic cellular graphene material and the relation of sensitivity versus density. (a) and (b) High‐frequency response of the graphene material with a density of 0.54 mg cm−3. (c) Electrical response of the graphene materials with different densities. (d) An example of the graphene material with a density of 0.54 mg cm−3 detecting ultralow pressure (1 mg weight on an area of 113 mm2 is equivalent to 0.082 Pa).

Source: Qiu et al. 2016 [274. Reprinted with permission from John Wiley and Sons.

The high electrical conductivity and the interconnected porosity of porous graphene materials have made them excellent electrodes for supercapacitor. A GO suspension was reduced with L‐ascorbic acid to form a hydrogel, which was freeze‐dried (or supercritical CO2 dried) to produce a graphene aerogel. This aerogel exhibited a BET surface area of 512 m2 g−1, pore volume of 2.48 cm3 g−1 and electrical conductivity of ∼100 S m−1 and could support 14 000 times more than its own weight. When evaluated as electrode for supercapacitor, it gave a specific capacity of 128 F g−1 at the rate density of 50 mA g−1 [47]. Hydroquinone‐functionalized graphene aerogel showed a specific capacitance of 441 F g−1 at 1 A g−1 in 1 M H2SO4 aqueous electrolyte. When using this material to build a flexible solid‐state supercapacitor with H2SO4–PVA gel as the electrolyte, a similar performance was obtained, i.e. 412 F g−1 at 1 A g−1, 74% capacitance retention at 20 A g−1 and 87% capacitance retention over 10 000 cycles [48]. Flexible freestanding graphene paper showed a specific capacitance of 137 F g−1 at 1 A g−1 in a solid‐state supercapacitor with H2SO4–PVA as gel electrolyte [50]. The cellular graphene paper was found to be highly effective for ultrahigh‐power‐density supercapacitors, generating power densities in the range of 7.8–14.3 kW kg−1. When increasing the loading of the active materials, the supercapacitor could produce energy density up to 1.11 W h l−1 [135]. The pristine graphene aerogels fabricated by freeze gelation at ambient temperature showed better performance as electrode for supercapacitors under high current density [53]. The aerogels made from glucose‐assisted reduced GO could produce a capacitance of 157 F g−1 at the optimized glucose content of 1 wt% [138]. GO and melamine were mixed to fabricate N‐doped graphene, which showed an improved capacitance of 217 F g−1 at a scan rate of 5 mV s−1, three times higher than that of the non‐doped graphene gel [126]. The 3D graphene aerogel on nickel foam was evaluated as a binder‐free supercapacitor electrode. The tests produced a specific capacitance of 366 F g−1 at a current density of 2 A g−1 in 6 M KOH. The retained capacitance was 60% after 2000 cycles of tests [49]. The graphene aerogels have also been used as electrodes for recharged batteries or fuel cells. For example, when the graphene paper was used as anode for LIBs, a discharge capacity of 420 mAh g−1 was obtained at a current density of 2 A g−1 [50]. Graphene sponge was examined as the anode for MFCs. The set‐up could generate a maximum power density of 427 W m−3, higher than the value of carbon felt at 395 W m−3 [139].

Porous graphene materials exhibit high pore volume, highly interconnected porosity, superelasticity and hydrophobicity. These properties make them excellent candidates for oil/water separation or absorption of organic liquids. Anisotropic porous graphene aerogels could absorb organic solvents (such as hexane, heptane to a mass increase of ∼140 times, acetone and ethanol to a mass increase of ∼160 times) and vegetable oil/pump oil to a mass increase of around 190 times. As tested with the absorption of hexane, the absorbed hexane could be removed by burning, distillation, or more conveniently by squeezing (due to superelasticity) to recycle the graphene aerogels [129]. The hierarchically porous graphene prepared by ice templating and emulsion templating could take up organic liquids 113–276 times their own weight. For a lighter graphene cellular network with a density of 1.5 mg cm−3, the absorption of motor oil could be as high as 605 g/g [131]. The porous graphene spheres showed impressive absorption of organic solvents (60–210 g/g depending on type and density) and oils (∼65–105 g/g from gasoline to lubricating oil) [136]. By placing a porous graphene aerogel on the flame of a wax candle and rotating for 3 s, soot was formed on the outer surface of the aerogel. This procedure increased the surface hydrophobicity and enhanced the absorption of organic liquid by 8–10 times. The weight gains were in the range of 140 g/g for soybean oil to 620 g/g for chloroform [140]. When used for absorption of pollutants from wastewater directly, GO scaffolds may perform better. The hydrophilic nature of the GO scaffold may facilitate wetting by water and enhance the interaction with polar pollutants and ions. One study showed the absorption of trichlorophenol in the concentration range of 1–20 mg l−1, achieving an absorption capacity of 21.08 mg g−1 [141].

Porous graphene or GO scaffolds have been used for other applications as well. One of them is for catalysis. Graphene nanoscroll networks [142], N/S‐doped [143] and N‐doped graphene porous structures [144] have been used as efficient electrocatalysts for oxygen reduction reactions. Porous GO scaffolds can be used to catalyse the oxidation of SO2 to SO3 [123] and as substrates for neural cell growth [125]. The high thermal conductivity of vertically aligned graphene membranes has been employed for the generation of clean water from seawater via the use of solar energy. The average water evaporation rates could reach 1.62 and 6.25 kg (m2 h)−1 under 1 and 4 sun illumination [122].

7.5.3 Graphene‐based or Graphene‐containing Materials by Ice Templating

This section is divided into two categories: graphene (or GO)–polymer composites and graphene–inorganic nanostructured hybrids.

The graphene (or GO)–polymer composites may be further classified into two types. The first type is the composites with polymer as the matrix or the major component. The main purpose of incorporating graphene is to reinforce the mechanical stability of the freeze‐dried porous polymers. GO, due to its surface functionality and good dispersibility in water, has been usually used as the additive. For example, GO was dispersed with epoxy precursor. After freeze‐drying and a conventional curing process, the porous GO–epoxy composites (with GO content 3–9 wt%) exhibited very low density (0.09 g cm−3), good mechanical strength (0.231 MPa) and high elasticity [145]. GO was added to the cellulose solution to induce gelation. The composite aerogel was obtained after freeze‐drying. With the addition of 0.1 wt% GO, the compressive strength and Young's modulus were increased by about 30% and 90%, respectively, compared to that of cellulose aerogel [146]. By combining cellulose nanofibres, GO, and sepiolite nanorod, the anisotropic porous composite (10% GO) showed excellent combustion resistance and a thermal conductivity (15 mW (m K)−1) only half that of expanded polystyrene. The composite showed high strength in the axial direction and low thermal conductivity in the radial direction, making it an excellent candidate for thermal insulation and as a fire‐retardant material [147]. With the incorporation of GO at 0.5–3 wt%, the Young's modulus of the gelatin–PVA–GO composite was improved by 67–133%. The porous composite was biocompatible as investigated with MC3T3‐E1 pre‐osteoblast murine cell line [148]. GO may be reduced and then processed to make reinforced graphene–polymer composites. For example, GO in aqueous suspension was chemically reduced in the presence of poly(sodium 4‐styrenesulfonate) (PSS). A freeze‐drying process was employed to fabricate PSS‐graphene/PVA composites (1 : 50, w/w). The compressional modulus was nearly an order of magnitude higher than the PVA‐only sample [149]. In another study, GO was reduced by hydrazine and then used to prepare the GO–chitosan composite, which was used to detect glucose by electrocatalysis [150].

The second type is the composites where graphene (or GO) is the matrix or has the continuous or skeletal structure. Like the inorganic platelets in nacre (see Chapter 6), the graphene sheets can be used as building blocks, in combination with polymer, to fabricate nacre‐like strong and tough composite materials. Porous silk fibroin (15 wt%)–GO composites could be prepared by freeze‐drying. When the mixing suspension gel was cast as film and air‐dried, the resulting film exhibited a tensile strength of 221 ± 16 MPa, a failure strain of 1.8 ± 0.4%, and a modulus of 17.2 ± 1.9 GPa [151]. Aqueous GO and PVA suspensions were processed by a bidirectional freezing approach, followed by reduction with hydroiodic acid and hot pressing at 95 °C. The electrical conductivities were in the range of 16.5–249 S m−1, corresponding to the GO content of about 44–71 wt%. Remarkably, the composite film with 58 wt% GO showed a tensile strength of 150.9 MPa, a fracture strain of 8.84%, toughness of 7.15 MJ m−3 and Young's modulus of 2.84 GPa [152].

Quite often, 3D porous graphene networks are fabricated first and the polymer phase is incorporated in the second stage. The resulting composites usually show good mechanical stability and superelasticity while the excellent electrical conductivity and thermal conductivity from graphene networks are retained. A robust, conductive, temperature‐responsive binary hydrogel was formed by soaking graphene cellular networks in N‐isopropylacrylamide solution containing a crosslinker and initiator and followed by polymerization [153]. Polyborosiloxane (PBS) exhibits an intrinsic self‐healing behaviour due to its dynamic dative triple and quadruple bonds formed between boron and oxygen in the Si-O groups. PBS is a highly viscous liquid at low strain rates but behaves like a solid at high strain rates. A self‐healing composite may be fabricated by incorporating PBS into porous graphene networks. This was demonstrated by Saiz and co‐authors, by infiltrating liquid PDMS and boron oxide (which react to form the PBS polymer in situ) into ice‐templated graphene networks. 6–8 healing cycles were demonstrated with no structural loss. This composite was used to sensor pressure change based on the piezoresistive response [154]. When epoxy resin was incorporated into vertically aligned porous graphene networks with a low loading of 0.92 vol%, the obtained composite showed a high thermal conductivity (2.13 W (m K)−1), a reduced coefficient of thermal expansion (∼37.4 ppm K−1) and an increased glass transition temperature (135.4 °C) [155].

For the graphene–inorganic hybrid structure, noble metal nanoparticles are usually formed within porous graphene (or GO) networks while inorganic nanostructures may be blended with graphene or grow from the graphene surface. Metal salts can be mixed with aqueous GO suspension and the resulting suspension is freeze‐dried. Metal nanoparticles are formed in situ when the GO scaffold is reduced to graphene. This has been demonstrated by preparation of Pt and Ag nanoparticles within 3D porous graphene. These materials gave ultralow density, good electrical conductivity and excellent EMI shielding effectiveness, and also enhanced electrocatalytic performance [156]. Aqueous suspension containing Nafion, GO and chloroplatinic acid was freeze‐dried to produce a highly porous structure. This structure could be reduced by hydrazine or monosodium citrate to form graphene‐supported Pt nanoparticles on a Nafion scaffold, which combined ionic conductivity (from Nafion), electronic conductivity (from graphene) and catalytic activity (from Pt nanoparticles) [157]. In another study, HAuCl4 was added to aqueous GO suspension. After further addition of ascorbic acid and a heating–freezing–heating process, gold nanoparticle/graphene gel was formed. After washing and freeze‐drying, dry porous graphene structures with Au nanoparticles were produced. Pt and Pd nanoparticles were also produced using a similar procedure. The composites showed excellent catalytic performance in the reduction of 4‐nitrophenol and methylene blue by NaBH4 [158].

Vertically porous composites of layered VOPO4 and graphene were constructed by the directional freezing process [51]. When tested as electrodes for supercapacitors, a high capacitance of 528 F g−1 at 0.5 A g−1 was obtained with 6 M KOH as electrolyte. For the asymmetric supercapacitor built from this composite as the cathode and vertically porous graphene as the anode, a high cell voltage of 1.6 V and energy density of 108 W h kg−1 were achieved [51]. NiCo2S4 nanotube@Ni–Mn layered double hydroxides were formed on graphene sponge by in situ growth via a multi‐step hydrothermal synthesis. When evaluated for supercapacitors, the composite electrode exhibited a specific capacitance of 1740 mF cm−2 at 1 mA cm−2 and 1268 mF cm−2 at 10 mA cm−2 [52]. C3N4–GO aerogel was fabricated directly by freeze‐drying the mixing suspension. Excellent visible‐light photocatalytic activity was demonstrated by the degradation of dyes (Rhodamine B, methyl orange, methylene blue) and the oxidation of NO at a ppb level [159]. The absorption wavelength of the composite could be extended to about 481 nm from the wavelength of about 453 nm from C3N4 powder [159]. Polyethylene glycol was incorporated into an ice‐templated boron nitride (BN)/GO scaffold. With a BN content of 19.2 wt%, the composite exhibited a high thermal conductivity of 1.84 W (m K)−1. It was further demonstrated to possess great potential as phase change materials to realize efficient light‐to‐thermal and light‐to‐electric energy conversion and storage [160].

7.6 Porous Graphene/CNT Hybrid Structures

Graphene (or GO) and CNTs can be used as building blocks together to fabricate porous carbon materials. This may avoid or reduce the dilution effect in CNT or graphene composite materials on thermal and electrical conductivity, where additives such as polymers are included to improve mechanical properties. Porous graphene/CNT hybrid structures prepared by the ice‐templating approach have shown enhanced properties such as surface area, elasticity, and compressive strength, compared to the individual components. The CNTs can act as structural support and as a separator to prevent graphene (GO) nanosheets from stacking or aggregating. This can result in a higher surface area in the resulting hybrid porous structure [53]. The strong interaction between the CNTs and graphene (via π–π stacking and hydrophobic interaction, or the interaction between the remaining functional groups on the surface of CNT and graphene) may bond the graphene nanosheets together and reduce the sliding under compression and thereby enhance the elasticity and compression strength [161, 162]. The long and flexible CNTs may also serve as threads to ‘bundle’ the nanosheets together [58].

Pristine graphene (PG)–MWCNT and reduced graphene oxide (rGO)–MWCNT aerogels were fabricated via a room‐temperature approach [53]. Higher specific capacitances were achieved for the hybrid aerogels. Between PG–MWCNT and rGO–MWCNT, the specific capacitance (305 F g−1) was higher for rGO–MWCNT than for PG/MWCNT (167 F g−1) at a current density of 1 A g−1 with nonaqueous 1.0 M tetraethyl ammonium tetrafluoroborate/propylene carbonate. This was attributed to the higher surface area in the rGO–MWCNT aerogel. However, PG–MWCNT showed a higher specific capacitance (100 F g−1) at a fast scan rate of 100 A g−1 than those of rGO–MWCNT (72 F g−1) and rGO aerogel (0.9 F g−1). This was attributed to the lower internal resistance of PG than that of rGO [53]. Aqueous GO/CNT dispersion was freeze‐dried and then thermally annealed at 800 °C under N2 for 3 h. The porous hybrid sponge with 20 wt% CNTs showed the highest surface area (498 m2 g−1) and pore volume (1.51 cm3 g−1). When tested by cyclic voltammetry (CV) at 10 mV s−1 in 1 M NaCl solution, a specific capacitance of 203 F g−1 was obtained. When evaluated as the capacitive deionization electrode, a high electrosorption capacity of 18.7 mg g−1 was achieved [163]. This composite material was also used as anode for sodium ion batteries, the highest charge capacity was achieved at 436 mAh g−1 after 100 cycles with a current density of 50 mA g−1. A capacity of 195 mAh g−1 was maintained at 10 A g−1 after 7440 cycles [58]. However, not all the graphene–CNT structures produce better performance for supercapacitors. For example, both graphene aerogel and graphene–CNT aerogel were formed on nickel foams. The hybrid material as electrode showed a lower capacitance (207 F g−1 at 2 A g−1) than that of graphene aerogel on nickel foam (366 F g−1 at 2 A g−1) [49]. The freeze‐dried GO–SWCNT scaffold was annealed at 800 °C under Ar for 2 h and evaluated as a counter electrode for dye‐sensitized solar cells. A photovoltaic conversion efficiency of 8.70% was achieved [164].

Figure 7.18 The pore morphologies and structures of (a)–(c) 3D porous graphene aerogel and (d)–(f) the graphene/CNT aerogel with similar initial density.

Source: Lv et al. 2016 [324. Reprinted with permission from Elsevier.

Oxidized CNTs (by refluxing in the mixture of concentrated H2SO4 and HNO3 (3 : 1, v/v)) were dispersed in water with GO. The gel was formed after a hydrothermal treatment (180 °C for 12 h) and then freeze‐dried to form the aerogel. Figure 7.18 shows the porous structures of graphene and graphene–CNT aerogels. Both of them are highly porous. At higher magnifications, the entangled CNTs can be seen to cover the ligaments and the cellular walls. The hybrid aerogel showed a high compressibility with the compression strain as high as 80%. The thermal conductivity increased with the density, achieving the highest at 88.5 W (m K)−1 with a density of 85 mg cm−3 (GO:CNT ratio at always 3 : 1). For the thermal interface testing, by locating the aerogels between two Cu blocks with high temperature and low temperature, a low thermal interface resistance of 13.6 mm2 KW−1 was obtained [162]. The oil‐in‐water emulsion with hexane as oil droplets and GO/oxidized CNTs in the continuous aqueous phase was hydrothermally treated at 180 °C for 10 h to form a gel. The hybrid porous scaffold was produced after freeze‐drying of the gel and then infiltrated with epoxy resin. The resulting composite with a thickness of 9 mm could absorb up to 80% of the incident radiation when evaluated as a microwave absorbing material [165].

The rGO/SWCNT aerogels prepared from a freezing–thawing–freeze drying process were assessed for their thermoelectric (TE) properties [166]. The TE performance is usually evaluated by a dimensionless figure of merit, ZT = S2 T/k, where S is the Seebeck coefficient, T is the temperature and k is the thermal conductivity. The composite aerogel with 37.5 wt% of SWCNT showed an enhanced TE performance with ZT = 8.03 × 10−3 [166]. This was attributed to the interconnected porosity, the energy‐filtering effect, and the phonon scattering at interfaces/joints. The same material, also due to its hydrophobicity, exhibited the high capacity to absorb organic liquids. The absorption capacities varied depending on the solvent, dichloromethane 139 g/g, N‐methyl‐2‐pyrrolidone 98 g/g, and n‐hexane 130 g/g. For n‐hexane, after 5 cycles, about 92% of the absorption capacity was retained [166]. The graphene–CNT aerogels have also been used for the removal of oil or pollutant by other researchers [159, 166]. The hybrid aerogel obtained by freeze‐drying the gel formed by hydrothermal treatment in the presence of ethylenediamine (120 °C, 12 h) could absorb organic liquids in the range of 100–270 times their own weight (100 g/g for n‐hexane). The recycling tests by combustion or squeezing to remove n‐hexane showed nearly no decrease in absorption for 10 cycles [167]. In another study, the hybrid aerogels were demonstrated as superior absorbents to remove pollutants, oxytetracycline 1729 mg g−1, diethyl phthalate 680 mg g−1, methylene blue 685 mg g−1, Cd2+ 235 mg g−1 and diesel 421 g/g [161]. The graphene–CNT aerogels usually have a good combination of compressibility and superelasticity, as shown by squeezing the solvent‐absorbed aerogel for the recycling purpose. Also, due to the piezoresistive behaviour, the graphene–CNT aerogel could be used as pressure sensors, e.g. in fields related to artificial skin [168].

A third component may be introduced into the graphene–CNT aerogels to improve their applications. For example, transition metal dichalcogenide nanosheets are known as active materials for LIBs and SIBs. Although MoS2 has been extensively used as the anode for LIBs and SIBs, WS2 may have greater potential because its intrinsic electronic conductivity is higher than that of MoS2 [59]. (NH4)2WS4 powders were dissolved in the GO and CNT dispersion in DMF. The obtained suspension was solvothermally treated at 200 °C for 10 h to produce a gel. The gel was washed, freeze‐dried, and thermally annealed at 500 °C for 2 h in H2/N2 to generate the hierarchically porous WS2/CNT–rGO aerogel. As the anode, this hybrid material delivered a specific capacitance of 749 mAh g−1 at 100 mA g−1 for LIBs and 311.4 mAh g−1 at 100 mA g−1 and 252.9 mAh g−1 at 200 mA g−1 after 100 cycles for SIBs [59]. Hexagonal nanorods of hydroxyapatite were formed on flexible porous graphene/SWCNT membrane [169]. This hybrid membrane was found to promote the attachment and proliferation of human foetal osteoblast osteoprogenitor cells and enhance in vitro biomineralization in simulated body fluid [169].

7.7 Summary

Porous carbon materials can be prepared by carbonization of porous polymers/polymeric gels or by fabrication directly from carbon building blocks, namely CNTs and graphenes (or graphene oxides). Conventionally, carbon aerogels are produced by carbonization of freeze‐dried highly crosslinked polymeric gels, with the typical example being resorcinol‐formaldehyde gel. The freeze‐drying process ensures the formation of highly porous dry polymeric gels. When the crosslinking density is low, the ice‐templated pore structure may be generated in carbon aerogels as well. However, in order to fully realize the ice‐templating potential, polymer solutions (where the polymers have high carbon content to facilitate the production of carbon materials by carbonization) are processed by controlled freezing and freeze‐drying to generate porous polymers, which can be subsequently pyrolysed under inert atmosphere to generate porous carbon materials. Additional colloidal templating may be included to induce high porosity and mesopores in the carbon materials. CNFs can be produced by carbonization of polymer nanofibres produced by freeze‐drying of dilute polymer solution or nanofibrous organic gels generated by pH‐triggered gelation.

More excitingly, CNTs and graphene/graphene oxide nanosheets are used as carbon building blocks, via the ice‐templating approach, to produce porous carbon structures, composite structures and CNTs/graphene hybrid structures. Most of these materials show good electrical conductivity, mechanical stability and high surface area. Remarkably, graphene and graphene oxide nanosheets can be used to fabricate porous networks with superelasticity and piezoresistivity.

Owing to the inert chemical stability, electrical conductivity, high surface area and surface hydrophobicity, porous carbon materials have been widely used as absorbents for water treatment, adsorbents for gas uptake, oil/water separation, electrode materials and catalyst supports. Applications for each category of the carbon materials are described in this chapter. The ice‐templated anisotropic pore structures can bring additional advantages for certain applications, e.g. anisotropic conductivity and enhanced mass transport in aligned channels. Particularly, the carbon materials made from CNTs and graphenes have been extensively used as electrodes for rechargeable batteries and supercapacitors. Owing to its unique nanosheet structure, graphenes and graphene oxides can be used to fabricate superelastic networks and conductive strong composites, which have been uniquely used for sensing pressure and detecting structure integrity. It is envisaged that the ice‐templating and freeze‐drying approach can be used widely to fabricate various functional carbon materials for advanced applications.


  1. 1 Dutta, S., Bhaumik, A., and Wu, K.C. (2014). Hierarchically porous carbon derived from polymers and biomass: effect of interconnected pores on energy applications. Energy Environ. Sci. 7: 3574–3592.
  2. 2 Liang, C., Li, Z., and Dai, S. (2008). Mesoporous carbon materials: synthesis and modification. Angew. Chem. Int. Ed. 47: 3696–3717.
  3. 3 Lee, J., Kim, J., and Hyeon, T. (2006). Recent progress in the synthesis of porous carbon materials. Adv. Mater. 18: 2073–2094.
  4. 4 Zhang, H. and Cooper, A.I. (2005). Synthesis and applications of emulsion‐templated porous materials. Soft Matter 1: 107–113.
  5. 5 Fechler, N., Fellinger, T.‐P., and Antonietti, M. (2013). “Salt templating”: a simple and sustainable pathway toward highly porous functional carbons from ionic liquids. Adv. Mater. 25: 75–79.
  6. 6 Titirici, M.‐M., White, R.J., Brun, N. et al. (2015). Sustainable carbon materials. Chem. Soc. Rev. 44: 250–290.
  7. 7 Al‐Muhtaseb, S.A. and Ritter, J.A. (2003). Preparation and properties of resorcinol‐formaldehyde organic and carbon gels. Adv. Mater. 115: 101–114.
  8. 8 Stein, A., Wang, Z., and Fierke, M.A. (2009). Functionalization of porous carbon materials with designed pore architecture. Adv. Mater. 21: 265–293.
  9. 9 Liu, J., Wickramaratne, N.P., Qiao, S.Z., and Jaroniec, M. (2015). Molecular‐based design and emerging applications of nanoporous carbon spheres. Nat. Mater. 14: 763–774.
  10. 10 Saufi, S.M. and Ismail, A.F. (2004). Fabrication of carbon membranes for gas separation – a review. Carbon 42: 241–259.
  11. 11 Huang, Z.‐M., Zhang, Y.‐Z., Kotaki, M., and Ramakrishna, S. (2003). A review on polymer nanofibers by electrospinning and their applications in nanocomposites. Compos. Sci. Technol. 63: 2223–2253.
  12. 12 Zhang, B., Kang, F., Tarascon, J.‐M., and Kim, J.‐K. (2016). Recent advances in electrospun carbon nanofibers and their application in electrochemical energy storage. Prog. Mater. Sci. 76: 319–380.
  13. 13 Chaikittisilp, W., Ariga, K., and Yamauchi, Y. (2013). A new family of carbon materials: synthesis of MOF‐derived nanoporous carbons and their promising applications. J. Mater. Chem. A 1: 14–19.
  14. 14 Tasis, D., Tagmatarchis, N., Bianco, A., and Prato, M. (2006). Chemistry of carbon nanotubes. Chem. Rev. 106: 1105–1136.
  15. 15 Geim, A.K. and Novoselov, K.S. (2007). The rise of graphene. Nat. Mater. 6: 183–191.
  16. 16 Park, S. and Ruoff, R.S. (2010). Chemical methods for the production of graphenes. Nat. Nanotechnol. 4: 217–224.
  17. 17 Lv, W., Zhang, C., Li, Z., and Yang, Q. (2015). Self‐assembled 3D graphene monolith from solution. J. Phys. Chem. Lett. 6: 658–668.
  18. 18 Dasgupta, A., Rajukumar, L.P., Rotella, C. et al. (2017). Covalent three‐dimensional networks of graphene and carbon nanotubes: synthesis and environmental applications. Nano Today 12: 116–135.
  19. 19 Cao, X., Yin, Z., and Zhang, H. (2014). Three‐dimensional graphene materials: preparation, structures and application in supercapacitors. Energy Environ. Sci. 7: 1850–1865.
  20. 20 Nardecchia, S., Carriazo, D., Ferrer, M.L. et al. (2013). Three dimensional macroporous architectures and aerogels built of carbon nanotubes and/or graphene: synthesis and applications. Chem. Soc. Rev. 42: 794–830.
  21. 21 Thostenson, E.T., Ren, Z., and Chou, T.‐W. (2001). Advances in the science and technology of carbon nanotubes and their composites: a review. Compos. Sci. Technol. 61: 1899–1912.
  22. 22 An, X. and Yu, J.C. (2011). Graphene‐based photocatalytic composites. RSC Adv. 1: 1426–1434.
  23. 23 Hu, K., Kulkarni, D.D., Choi, I., and Tsukruk, V.V. (2014). Graphene‐polymer nanocomposites for structural and functional applications. Prog. Polym. Sci. 39: 1934–1972.
  24. 24 Serrano, M.C., Gutiérrez, M.C., and del Monte, F. (2014). Role of polymers in the design of 3D carbon nanotube‐based scaffolds for biomedical applications. Prog. Polym. Sci. 39: 1448–1471.
  25. 25 Nieto, A., Bisht, A., Lahiri, D. et al. (2017). Graphene reinforced metal and ceramic matrix composites: a review. Int. Mater. Rev. 62: 241–302.
  26. 26 Yang, Z., Ren, J., Zhang, Z. et al. (2015). Recent advancement of nanostructured carbon for energy applications. Chem. Rev. 115: 5159–5223.
  27. 27 Wood, K.N., O'Hayre, R., and Pylypenko, S. (2014). Recent progress on nitrogen/carbon structures designed for use in energy and sustainability applications. Energy Environ. Sci. 7: 1212–1249.
  28. 28 Hu, L., Hecht, D.S., and Grüner, G. (2010). Carbon nanotube thin films: fabrication, properties, and applications. Chem. Rev. 110: 5790–5844.
  29. 29 Qiu, H., Guan, Y., Luo, P., and Wang, Y. (2017). Recent advances in fabricating monolithic 3D porous graphene and their applications in biosensing and biofuel cells. Biosens. Bioelectron. 89: 85–95.
  30. 30 Chang, H., Joo, S.H., and Pak, C. (2007). Synthesis and characterization of mesoporous carbon for fuel cell applications. J. Mater. Chem. 17: 3078–3088.
  31. 31 Mauter, M.S. and Elimelech, M. (2008). Environmental applications of carbon‐based nanomaterials. Environ. Sci. Technol. 42: 5843–5859.
  32. 32 Ren, X., Chen, C., Nagatsu, M., and Wang, X. (2011). Carbon nanotubes as adsorbents in environmental pollution management: a review. Chem. Eng. J. 170: 395–410.
  33. 33 Yu, J.‐S., Kang, S., Yoon, S.B., and Chai, G. (2002). Fabrication of ordered uniform porous carbon networks and their application to a catalyst supporter. J. Am. Chem. Soc. 124: 9382–9383.
  34. 34 Kumar, S., Rani, R., Dilbaghi, N. et al. (2017). Carbon nanotubes: a novel material for multifaceted applications in human healthcare. Chem. Soc. Rev. 46: 158–196.
  35. 35 Chung, C., Kim, Y.‐K., Shin, D. et al. (2012). Biomedical applications of graphene and graphene oxide. Acc. Chem. Res. 46: 2211–2224.
  36. 36 Qian, L. and Zhang, H. (2011). Controlled freezing and freeze drying: a versatile route for porous and micro‐/nano‐structured materials. J. Chem. Technol. Biotechnol. 86: 172–184.
  37. 37 Deville, S. (2013). Ice‐templating, freeze casting: beyond materials processing. J. Mater. Res. 28: 2202–2219.
  38. 38 Job, N., Théry, A., Pirard, R. et al. (2005). Carbon aerogels, cryogels and xerogels: influence of the drying method on the textural properties of porous carbon materials. Carbon 43: 2481–2494.
  39. 39 Candelaria, S.L., Chen, R., Jeong, Y.‐H., and Cao, G. (2012). Highly porous chemically modified carbon cryogels and their coherent nanocomposites for energy applications. Energy Environ. Sci. 5: 5619–5637.
  40. 40 Amaral‐Labat, G., Szczurek, A., Fierro, V. et al. (2012). Pore structure and electrochemical performances of tannin‐based carbon cryogels. Biomass Bioenergy 39: 274–282.
  41. 41 Zhou, Y., Candelaria, S.L., Liu, Q. et al. (2014). Sulfur‐rich carbon cryogels for supercapacitors with improved conductivity and wettability. J. Mater. Chem. A 2: 8472–8482.
  42. 42 Khalid, B., Meng, Q., Li, J., and Cao, B. (2014). Nitrogen rich graphene‐cross‐linked melamine formaldehyde carbon cryogels for supercapacitors. Electrochim. Acta 142: 101–107.
  43. 43 Choi, Y.‐M., Singh, K.P., Park, J.D. et al. (2016). Hierarchically microcellular microporous carbon from polyamic acid cryogel and its electrochemical capacitance. Energy Technol. 4: 278–287.
  44. 44 Estevez, L., Dua, R., Bhandari, N. et al. (2013). A facile approach for the synthesis of monolithic hierarchical porous carbons‐high performance materials for amine based CO2 capture and supercapacitor electrode. Energy Environ. Sci. 6: 1785–1790.
  45. 45 Spender, J., Demers, A.L., Xie, X. et al. (2012). Method for production of polymer and carbon nanofibers from water‐soluble polymers. Nano Lett. 12: 3857–3860.
  46. 46 Liu, X., Ahmed, A., Wang, Z., and Zhang, H. (2015). Nanofibrous microspheres via emulsion gelation and carbonization. Chem. Commun. 51: 16864–16867.
  47. 47 Zhang, X., Sui, Z., Xu, B. et al. (2011). Mechanically strong and highly conductive graphene aerogel and its use as electrodes for electrochemical power sources. J. Mater. Chem. 21: 6494–6497.
  48. 48 Xu, Y., Lin, Z., Huang, X. et al. (2013). Functionalized graphene hydrogel‐based high‐performance supercapacitors. Adv. Mater. 25: 5779–5784.
  49. 49 Ye, S., Feng, J., and Wu, P. (2013). Deposition of three‐dimensional graphene aerogel on nickel foam as a binder‐free supercapacitor electrode. ACS Appl. Mater. Interfaces 5: 7122–7129.
  50. 50 Shu, K., Wang, C., Li, S. et al. (2015). Flexible free‐standing graphene paper with interconnected porous structure for energy storage. J. Mater. Chem. A 3: 4428–4434.
  51. 51 Lee, K.H., Lee, Y.‐W., Lee, S.W. et al. (2015). Ice‐templated self‐assembly of VOPO4–graphene nanocomposites for vertically porous 3D supercapacitor electrodes. Sci. Rep. 5: 13696.
  52. 52 Wan, H., Liu, J., Ruan, Y. et al. (2015). Hierarchical configuration of NiCo2S4 nanotube@Ni‐Mn layered double hydroxide arrays/three‐dimensional graphene sponge as electrode materials for high‐capacitance supercapacitors. ACS Appl. Mater. Interfaces 7: 15840–15847.
  53. 53 Lin, Y., Liu, F., Casano, G. et al. (2016). Pristine graphene aerogels by room‐temperature freeze gelation. Adv. Mater. 28: 7993–8000.
  54. 54 Pan, A., Liu, D., Zhou, X. et al. (2010). Enhanced lithium‐ion intercalation properties of coherent hydrous vanadium pentoxide–carbon cryogel nanocomposites. J. Power Sources 195: 3893–3899.
  55. 55 Zhang, M., Li, Y., Uchaker, E. et al. (2013). Homogenous incorporation of SnO2 nanoparticles in carbon cryogels via the thermal decomposition of stannous sulfate and their enhanced lithium‐ion intercalation properties. Nano Energy 2: 769–778.
  56. 56 Sahore, R., Estevez, L.P., Ramanujapuram, A. et al. (2015). High‐rate lithium‐sulfur batteries enabled by hierarchical porous carbons synthesized via ice templation. J. Power Sources 297: 188–194.
  57. 57 Jazaeri, E., Zhang, L., Wang, X., and Tsuzuki, T. (2011). Fabrication of carbon nanofiber by pyrolysis of freeze‐dried cellulose nanofiber. Cellulose 18: 1481–1485.
  58. 58 Yan, D., Xu, X., Lu, T. et al. (2016). Reduced graphene oxide/carbon nanotubes sponge: a new high capacity and long life anode material for sodium‐ion‐batteries. J. Power Sources 316: 132–138.
  59. 59 Wang, Y., Kong, D., Shi, W. et al. (2016). Ice templated free‐standing hierarchically WS2/CNT‐rGO aerogel for high‐performance rechargeable lithium and sodium ion batteries. Adv. Energy Mater. 6. doi: 10.1002/aenm.201601057.
  60. 60 Szczurek, A., Jurewicz, K., Amaral‐Labat, G. et al. (2010). Structure and electrochemical capacitance of carbon cryogels derived from phenol‐formaldehyde resins. Carbon 48: 3874–3883.
  61. 61 Tamon, H., Ishizaka, H., Yamamoto, T., and Suzuki, T. (1999). Preparation of mesoporous carbon by freeze drying. Carbon 37: 2049–2055.
  62. 62 Tamon, H., Ishizaka, H., Yamamoto, T., and Suzuki, T. (2000). Influence of freeze‐drying conditions on the mesoporosity of organic gels as carbon precursors. Carbon 38: 1099–1105.
  63. 63 Yamamoto, T., Nishimura, T., Suzuki, T., and Tamon, H. (2001). Control of mesoporosity of carbon gels prepared by sol–gel polycondensation and freeze drying. J. Non‐Cryst. Solids 288: 46–55.
  64. 64 Yamamoto, T., Sugimoto, T., Suzuki, T. et al. (2002). Preparation and characterization of carbon cryogel microspheres. Carbon 40: 1345–1351.
  65. 65 Hamano, Y., Tsujimura, S., Shirai, O., and Kano, K. (2014). Control of the pore size distribution of carbon cryogels by pH adjustment of catalyst solutions. Mater. Lett. 128: 191–194.
  66. 66 Ogino, I., Kazuki, S., and Mukai, S.R. (2014). Marked increase in hydrophobicity of monolithic carbon cryogels via HCl aging of precursor resorcinol‐formaldehyde hydrogels: application to 1‐butanol recovery from dilute aqueous solutions. J. Phys. Chem. C 118: 6866–6872.
  67. 67 Hotová, G. and Slovák, V. (2015). Effect of pyrolysis temperature and thermal oxidation on the adsorption properties of carbon cryogels. Thermochim. Acta 614: 45–51.
  68. 68 Sepehri, S., García, B.B., Zhang, Q., and Cao, G. (2009). Enhanced electrochemical and structural properties of carbon cryogels by surface chemistry alteration with boron and nitrogen. Carbon 47: 1436–1443.
  69. 69 Ogino, I., Suzuki, Y., and Mukai, S.R. (2015). Tubing the pore structure and surface properties of carbon‐based acid catalysts for liquid‐phase reactions. ACS Catal. 5: 4951–4958.
  70. 70 Minović, T., Gulicovski, J.J., Stoiljković, M.M. et al. (2015). Surface characterization of mesoporous carbon cryogel and its application in arsenic (III) adsorption from aqueous solutions. Microporous Mesoporous Mater. 201: 271–276.
  71. 71 Yoshida, S., Iwamura, S., Ogino, I., and Mukai, S.R. (2016). Adsorption of phenol in flow systems by a monolithic carbon cryogel with a microhoneycomb structure. Adsorption 22: 1051–1058.
  72. 72 Kim, S.‐I., Yamamoto, T., Endo, A. et al. (2006). Adsorption of phenol and reactive dyes from aqueous solution on carbon cryogel microspheres with controlled porous structure. Microporous Mesoporous Mater. 96: 191–196.
  73. 73 Busquets, R., Ivanov, A.E., Mbundi, L. et al. (2016). Carbon‐cryogel hierarchical composites as effective and scalable filters for removal of trace organic pollutants from water. J. Environ. Manag. 182: 141–148.
  74. 74 Roberts, A.D., Li, X., and Zhang, H. (2014). Porous carbon spheres and monoliths: morphology control, pore size tuning and their applications as Li‐ion battery anode materials. Chem. Soc. Rev. 43: 4341–4356.
  75. 75 Roberts, A.D., Wang, S., Li, X., and Zhang, H. (2014). Hierarchical porous nitrogen‐rich carbon monoliths via ice‐templating: high capacity and high‐rate performance as lithium‐ion battery anode materials. J. Mater. Chem. A 2: 17787–17796.
  76. 76 Roberts, A.D., Lee, J.S.M., Wong, S.Y. et al. (2017). Nitrogen‐rich activated carbon monoliths via ice‐templating with high CO2 and H2 adsorption capacities. J. Mater. Chem. A 5: 2811–2820.
  77. 77 Roberts, A.D., Li, X., and Zhang, H. (2015). Hierarchically porous sulfur‐containing activated carbon monoliths via ice‐templating and one‐step pyrolysis. Carbon 95: 268–278.
  78. 78 Alhwaige, A.A., Ishida, H., and Qutubuddin, S. (2016). Carbon aerogels with excellent CO2 adsorption capacity synthesized from clay‐reinforced biobased chitosan‐polybenzoxazine nanocomposites. ACS Sustain. Chem. Eng. 4: 1286–1295.
  79. 79 McCann, J.T., Marquez, M., and Xia, Y. (2006). Highly porous fibers by electrospinning into a cryogenic liquid. J. Am. Chem. Soc. 128: 1436–1437.
  80. 80 Zou, L., Gan, L., Lv, R. et al. (2011). A film of porous carbon nanofibers that contain Sn/SnOx nanoparticles in the pores and its electrochemical performance as an anode material for lithium ion batteries. Carbon 49: 89–95.
  81. 81 Liang, H.‐W., Guan, Q.‐F., Chen, L.‐F. et al. (2012). Macroscopic‐scale template synthesis of robust carbonaceous nanofiber hydrogels and aerogels and their applications. Angew. Chem. Int. Ed. 51: 5101–5105.
  82. 82 Nogi, M., Kurosaki, F., Yano, H., and Takano, M. (2010). Preparation of nanofibrillar carbon from chitin nanofibers. Carbohydr. Polym. 81: 919–924.
  83. 83 Wang, W., Sun, Y., Liu, B. et al. (2015). Porous carbon nanofiber webs derived from bacterial cellulose as an anode for high performance lithium ion batteries. Carbon 91: 56–65.
  84. 84 Qian, L., Willneff, E., and Zhang, H. (2009). A novel route to polymeric sub‐micron fibers and their use for inorganic structures. Chem. Commun. 3946–3948.
  85. 85 Qian, L. and Zhang, H. (2010). Green synthesis of chitosan‐based nanofibers and their applications. Green Chem. 12: 1207–1214.
  86. 86 Fang, W., Yang, S., Wang, X.‐L. et al. (2017). Manufacture and application of lignin‐based carbon fibers (LCFs) and lignin‐based carbon nanofibers (LCNFs). Green Chem. 19: 1794–1827.
  87. 87 Liu, X., Roberts, A., Ahmed, A. et al. (2015). Carbon nanofibers by pyrolysis of self‐assembled perylene diimide derivative gels as supercapacitor electrode materials. J. Mater. Chem. A 3: 15513–15522.
  88. 88 Terrones, M. (2003). Science and technology of the twenty‐first century: synthesis, properties, and applications of carbon nanotubes. Annu. Rev. Mater. Res. 33: 419–501.
  89. 89 Yan, Y., Miao, J., Yang, Z. et al. (2015). Carbon nanotube catalysts: recent advances in synthesis, characterization and applications. Chem. Soc. Rev. 44: 3295–3346.
  90. 90 Ihsanullah, Abbas, A., Al‐Amer, A.M. et al. (2016). Heavy metal removal from aqueous solution by advanced carbon nanotubes: critical review of adsorption applications. Sep. Purif. Technol. 157: 141–161.
  91. 91 Schnorr, J.M. and Swager, T.M. (2011). Emerging applications of carbon nanotubes. Chem. Mater. 23: 646–657.
  92. 92 De Volder, M.F.L., Tawfick, S.H., Baughman, R.H., and Hart, A.J. (2013). Carbon nanotubes: present and future commercial applications. Science 339: 535–539.
  93. 93 Lee, J., Ko, S., Kwon, C.H. et al. (2016). Carbon nanotube yarn‐based glucose sensing artificial muscle. Small 12: 2085–2091.
  94. 94 Bryning, M.B., Milkie, D.E., Islam, M.F. et al. (2007). Carbon nanotube aerogels. Adv. Mater. 19: 661–664.
  95. 95 Thongprachan, N., Nakagawa, K., Sano, N. et al. (2008). Preparation of macroporous solid foam from multi‐walled carbon nanotubes by freeze‐drying technique. Mater. Chem. Phys. 112: 262–269.
  96. 96 Ozden, S., Narayanan, T., Tiwary, C.S. et al. (2015). 3D macroporous solids from chemically cross‐linked carbon nanotubes. Small 11: 688–693.
  97. 97 Ozden, S., Tsafack, T., Owuor, P.S. et al. (2017). Chemically interconnected light‐weight 3D–carbon nanotube solid network. Carbon 119: 142–149.
  98. 98 Koizumi, R., Hart, A.H.C., Brunetto, G. et al. (2016). Mechano‐chemical stabilization of three‐dimensional carbon nanotube aggregates. Carbon 110: 27–33.
  99. 99 Abarrategi, A., Gutiérrez, M.C., Moreno‐Vicente, C. et al. (2008). Multiwall carbon nanotube scaffolds for tissue engineering purposes. Biomaterials 29: 94–102.
  100. 100 Nardecchia, S., Serrano, M.C., Gutiérrez, M.C. et al. (2013). Modulating the cytocompatibility of tridimensional carbon nanotube‐based scaffolds. J. Mater. Chem. B 1: 3064–3072.
  101. 101 Serrano, M., Nardecchia, S., García‐Rama, C. et al. (2014). Chondroitin sulphate‐based 3D scaffolds containing MWCNTs for nervous tissue repair. Biomaterials 35: 1543–1551.
  102. 102 Nardecchia, S., Serrano, M.C., García‐Argüelles, S. et al. (2017). Ice as a green‐structure‐directing agent in the synthesis of macroporous MWCNTs and chondroitin sulphate composites. Materials 10: 355.
  103. 103 Gutiérrez, M.C., García‐Carvajal, Z.Y., Hortigüela, M.J. et al. (2007). Biocompatible MWCNT scaffolds for immobilization and proliferation of E. coli. J. Mater. Chem. 17: 2992–2995.
  104. 104 Katuri, K., Ferrer, M.L., Gutiérrez, M.C. et al. (2011). Three‐dimensional microchanelled electrodes in flow‐through configuration for bioanode formation and current generation. Energy. Environ. Sci. 4: 4201–4210.
  105. 105 Ouyang, A., Gong, Q., and Liang, J. (2015). Carbon nanotube–chitosan composite beads with radially aligned channels and nanotube‐exposed walls for bilirubin adsorption. Adv. Eng. Mater. 17: 460–466.
  106. 106 Lau, C., Cooney, M.J., and Atanassov, P. (2008). Conductive macroporous composite chitosan‐carbon nanotube scaffolds. Langmuir 24: 7004–7010.
  107. 107 Gutiérrez, M.C., Hortigüela, M.J., Amarilla, J.M. et al. (2007). Macroporous 3D architectures of self‐assembled MWCNT surface decorated with Pt nanoparticles as anodes for a direct methanol fuel cell. J. Phys. Chem. C 111: 5557–5560.
  108. 108 Yan, J., Wu, T., Ding, Z., and Li, X. (2016). Preparation and characterization of carbon nanotubes/chitosan composite foam with enhanced elastic property. Carbohydr. Polym. 136: 1288–1296.
  109. 109 Qi, H., Mäder, E., and Liu, J. (2013). Electrically conductive aerogels composed of cellulose and carbon nanotubes. J. Mater. Chem. A 1: 9714–9720.
  110. 110 Kwon, S.‐M., Kim, H.‐S., and Jin, H.‐J. (2009). Multiwalled carbon nanotube cryogels with aligned and non‐aligned porous structure. Polymer 50: 2786–2792.
  111. 111 Zeng, Z., Jin, H., Chen, M. et al. (2016). Lightweight and anisotropic porous MWNCT/WPU composites for ultrahigh performance electromagnetic interference shielding. Adv. Funct. Mater. 26: 303–310.
  112. 112 Cheng, J., Gu, G., Guan, Q. et al. (2016). Synthesis of a porous sheet‐like V2O5‐CNT nanocomposite using an ice‐templating ‘bricks‐and‐motar’ assembly approach as a high‐capacity, long cycle life cathode material for lithium‐ion batteries. J. Mater. Chem. A 4: 2729–2737.
  113. 113 Li, Y., Zhao, M., Chen, J. et al. (2016). Self‐assembled NiFe2O4/carbon nanotubes sponge for enhanced glucose biosensing application. Appl. Surf. Sci. 362: 115–120.
  114. 114 Mei, H., Xia, J., Han, D. et al. (2016). Dramatic increase in electrical conductivity in epoxy composites with uni‐directionally oriented laminae of carbon nanotubes. Chem. Eng. J. 304: 970–976.
  115. 115 Dong, L., Hou, F., Li, Y. et al. (2014). Preparation of continuous carbon nanotube networks in carbon fiber/epoxy composite. Compos. Part A 56: 248–255.
  116. 116 Cho, E.‐C., Chang‐Jian, C.‐W., Hsiao, Y.‐S. et al. (2016). Three‐dimensional carbon nanotube based polymer composites for thermal management. Compos. Part A 90: 678–686.
  117. 117 He, H., Klinowski, J., Forster, M., and Lerf, A. (1998). A new structural model for graphite oxide. Chem. Phys. Lett. 287: 53–56.
  118. 118 Li, D., Müller, M.B., Gilje, S. et al. (2008). Processable aqueous dispersions of graphene nanosheets. Nat. Nanotechnol. 3: 101–105.
  119. 119 Huang, X., Yin, Z., Wu, S. et al. (2011). Graphene‐based materials: synthesis, characterization, properties, and applications. Small 7: 1876–1902.
  120. 120 Sun, Y., Wu, Q., and Shi, G. (2011). Graphene based new energy materials. Energy Environ. Sci. 4: 1113–1132.
  121. 121 Zhu, C., Han, Y., Duoss, E.B. et al. (2015). Highly compressible 3D periodic graphene aerogel microlattices. Nat. Commun. 6: 6962.
  122. 122 Zhang, P., Li, J., Lv, L. et al. (2017). Vertically aligned graphene sheets membrane for highly efficient solar thermal generation of clean water. ACS Nano 11: 5087–5093.
  123. 123 Long, Y., Zhang, C., Wang, X. et al. (2011). Oxidation of SO2 to SO3 catalyzed by graphene oxide foams. J. Mater. Chem. 21: 13934–13941.
  124. 124 Kuang, J., Liu, L., Gao, Y. et al. (2013). A hierarchically structured graphene foam and its potential as a large‐scale strain‐gauge sensor. Nanoscale 5: 12171–12177.
  125. 125 Serrano, M.C., Patino, J., Garcia‐Rama, C. et al. (2014). 3D freeze‐standing porous scaffolds made of graphene oxide as substrates for neural cell growth. J. Mater. Chem. B 2: 5698–5706.
  126. 126 Kota, M., Yu, X., Yeon, X.‐H. et al. (2016). Ice‐templated three dimensional nitrogen doped graphene for enhanced supercapacitor performance. J. Power Sources 303: 372–378.
  127. 127 Qiu, L., Liu, J.Z., Chang, S.L.Y. et al. (2012). Biomimetic superelastic graphene‐based cellular monoliths. Nat. Commun. 3: 1241.
  128. 128 Wang, S., Wang, Z., Futamura, R. et al. (2017). Highly microporous‐graphene aerogel monolith of unidirectional honeycomb macro‐textures. Chem. Phys. Lett. 673: 38–43.
  129. 129 Liu, T., Huang, M., Li, X. et al. (2016). Highly compressible anisotropic graphene aerogels fabricated by directional freezing for efficient absorption of organic liquids. Carbon 100: 456–464.
  130. 130 Zhang, H., Wang, D., Butler, R. et al. (2008). Formation and enhanced biocidal activity of water‐dispersible organic nanoparticles. Nat. Nanotechnol. 3: 506–511.
  131. 131 Qian, L., Ahmed, A., Foster, A. et al. (2009). Systematic tuning of pore morphologies and pore volumes in macroporous materials by freezing. J. Mater. Chem. 19: 5212–5219.
  132. 132 Barg, S., Perez, F.M., Ni, N. et al. (2014). Mesoscale assembly of chemically modified graphene into complex cellular networks. Nat. Commun. 5: 4328.
  133. 133 Lv, L., Zhang, P., Cheng, H. et al. (2016). Solution‐processed ultraelastic and strong air‐bubbled graphene foams. Small 12: 3229–3234.
  134. 134 Zhang, R., Cao, Y., Li, P. et al. (2014). Three‐dimensional porous graphene sponges assembled with the combination of surfactant and freeze‐drying. Nano Res. 7: 1477–1487.
  135. 135 Shao, Y., El‐Kady, M.F., Lin, C.‐W. et al. (2016). 3D freeze‐casting of cellular graphene films for ultrahigh‐power‐density supercapacitors. Adv. Mater. 28: 6719–6716.
  136. 136 Liao, S., Zhai, T., and Xia, H. (2016). Highly adsorptive graphene aerogel microspheres with center‐diverging microchannel structures. J. Mater. Chem. A 4: 1068–1077.
  137. 137 Qiu, L., Coskun, M.B., Tang, Y. et al. (2016). Ultrafast dynamic piezoresistive response of graphene‐based cellular elastomers. Adv. Mater. 28: 194–200.
  138. 138 Cheng, J.‐H., Chen, Y.‐H., Yeh, Y.‐S. et al. (2016). Enhancement of electrochemical properties by freeze‐dried graphene oxide via glucose‐assisted reduction. Electrochim. Acta 197: 146–151.
  139. 139 Chen, W., Huang, Y.‐X., Li, D.‐B. et al. (2014). Preparation of a macroporous flexible three dimensional graphene sponge using an ice‐ template as the anode material for microbial fuel cells. RSC Adv. 4: 21619–21624.
  140. 140 Bi, H., Xie, X., Yin, K. et al. (2014). Highly enhanced performance of spongy graphene as an oil sorbent. J. Mater. Chem. A 2: 1652–1656.
  141. 141 Wang, J., Gao, X., Wang, Y., and Gao, C. (2014). Novel graphene oxide sponge synthesized by freeze‐drying process for the removal of 2,4,6‐ trichlorophenol. RSC Adv. 4: 57476–57482.
  142. 142 Shin, Y.‐E., Sa, Y.J., Park, S. et al. (2014). An ice‐templated, pH‐tunable self‐assembly route to hierarchically porous graphene nanoscroll networks. Nanoscale 6: 9734–9741.
  143. 143 Su, Y., Zhang, Y., Zhuang, X. et al. (2013). Low‐temperature synthesis of nitrogen/sulfur co‐doped three‐dimensional graphene frameworks as efficient metal‐free electrocatalyst for oxygen reduction reaction. Carbon 62: 296–301.
  144. 144 Wang, J., Wang, H.‐S., Wang, K. et al. (2014). Ice crystals growth driving assembly of porous nitrogen‐doped graphene for catalyzing oxygen reduction probed by in situ fluorescence electrochemistry. Sci. Rep. 4: 6723.
  145. 145 Ye, S., Feng, J., and Wu, P. (2013). Highly elastic graphene oxide–epoxy composite aerogels via simple freeze‐drying and subsequent routine curing. J. Mater. Chem. A 1: 3495–3502.
  146. 146 Zhang, J., Cao, Y., Feng, J., and Wu, P. (2012). Graphene‐oxide‐sheet‐induced gelation of cellulose and promoted mechanical properties of composite aerogels. J. Phys. Chem. C 116: 8063–8068.
  147. 147 Wicklein, B., Kocjan, A., Salazar‐Alvarez, G. et al. (2015). Thermally insulating and fire‐retardant lightweight anisotropic foams based on nanocellulose and graphene oxide. Nat. Nanotechnol. 10: 277–283.
  148. 148 Ionita, M., Crica, L.E., Tiainen, H. et al. (2016). Gelatin–poly(vinyl alcohol) porous biocomposites reinforced with graphene oxide as biomaterials. J. Mater. Chem. B 4: 282–291.
  149. 149 Vickery, J.L., Patil, A.J., and Mann, S. (2009). Fabrication of graphene‐polymer nanocomposites with higher‐order three‐dimensional architecture. Adv. Mater. 21: 2180–2184.
  150. 150 Qian, L. and Lu, L. (2014). Three dimensional porous graphene–chitosan composites from ice‐induced assembly for direct electron transfer and electrocatalysis of glucose oxidase. RSC Adv. 4: 38273–38280.
  151. 151 Huang, L., Li, C., Yuan, W., and Shi, G. (2013). Strong composite films with layered structures prepared by casting silk fibroin–graphene oxide hydrogels. Nanoscale 5: 3780–3786.
  152. 152 Zhao, N., Yang, M., Zhao, Q. et al. (2017). Superstretchable nacre‐mimetic graphene/poly(vinyl alcohol) composite film based on interfacial architectural engineering. ACS Nano 11: 4777–4784.
  153. 153 Qiu, L., Liu, D., Wang, Y. et al. (2014). Mechanically robust, electrically conductive and stimuli‐responsive binary network hydrogels enabled by superelastic graphene aerogels. Adv. Mater. 26: 3333–3337.
  154. 154 D'Elia, E., Barg, S., Ni, N. et al. (2015). Self‐healing graphene‐based composites with sensing capabilities. Adv. Mater. 27: 4788–4794.
  155. 155 Lian, G., Tuan, C.‐C., Li, L. et al. (2016). Vertically aligned and interconnected graphene networks for high thermal conductivity of epoxy composites with ultralow loading. Chem. Mater. 28: 6096–6104.
  156. 156 Sahoo, P.K., Aepuru, R., Panda, H.S., and Bahadur, D. (2015). Ice‐templated synthesis of multifunctional three dimensional graphene/noble metal nanocomposites and their mechanical, electrical, catalytic, and electromagnetic shielding properties. Sci. Rep. 5: 17726.
  157. 157 Estevez, L., Kelarakis, A., Gong, Q. et al. (2011). Multifunctional graphene/platinum/nafion hybrids via ice templating. J. Am. Chem. Soc. 133: 6122–6125.
  158. 158 Sahoo, P.K., Panigrahy, B., Thakur, D., and Bahadur, D. (2017). Ice‐templating synthesis of macroporous noble metal/3D–graphene nanocomposites: their fluorescence lifetimes and catalytic study. New J. Chem. 41: 7861–7869.
  159. 159 Wan, W., Yu, S., Dong, F. et al. (2016). Efficient C3N4/graphene oxide macroscopic aerogel visible‐light photocatalyst. J. Mater. Chem. A 6: 7823–7829.
  160. 160 Yang, J., Tang, L.‐S., Bao, R.‐Y. et al. (2016). An ice‐templated assembly strategy to construct graphene oxide/boron nitride hybrid porous scaffolds in phase change materials with enhanced thermal conductivity and shape stability for light‐thermal‐electric energy conversion. J. Mater. Chem. A 4: 18841–18851.
  161. 161 Shen, Y., Zhu, X., Zhu, L., and Chen, B. (2017). Synergistic effects of 2D graphene oxide nanosheets and 1D carbon nanotubes in the constructed 3D carbon aerogel for high performance pollutant removal. Chem. Eng. J. 314: 336–346.
  162. 162 Lv, P., Tan, X.‐W., Yu, K.‐H. et al. (2016). Super‐elastic graphene/carbon nanotube aerogel: a novel thermal interface material with highly thermal transport properties. Carbon 99: 222–228.
  163. 163 Xu, X., Liu, Y., Lu, T. et al. (2015). Rational design and fabrication of graphene/carbon nanotubes hybrid sponge for high performance capacitive deionization. J. Mater. Chem. A 3: 13418–13425.
  164. 164 Ma, J., Shen, W., Li, C. et al. (2017). Graphene cryogel‐based counter electrode materials freeze‐dried using different solution media for dye‐sensitized solar cells. Chem. Eng. J. 319: 155–162.
  165. 165 González, M., Baselga, J., and Pozuelo, J. (2016). High porosity scaffold composites of graphene and carbon nanotubes as microwave absorbing materials. J. Mater. Chem. C 4: 8575–8582.
  166. 166 Tan, D., Zhao, J., Gao, C. et al. (2017). Carbon nanoparticle hybrid aerogels: 3D double‐interconnected network porous microstructure, thermoelectric, and solvent‐removal functions. ACS Appl. Mater. Interfaces 9: 21820–21828.
  167. 167 Wan, W., Zhang, R., Li, W. et al. (2016). Graphene‐carbon nanotube aerogel as an ultralight, compressible and recyclable highly efficient absorbent for oil and dyes. Environ. Sci.: Nano 3: 107–113.
  168. 168 Kuang, J., Dai, Z., Liu, L. et al. (2015). Synergistic effects from graphene and carbon nanotubes endow ordered hierarchical structure foams with a combination of compressibility, super‐elasticity and stability and potential application as pressure sensors. Nanoscale 7: 9252–9260.
  169. 169 Zhang, R., Metoki, N., Sharabani‐Yosef, O. et al. (2016). Hydroxyapatite/mesoporous graphene/single‐walled carbon nanotubes freestanding flexible hybrid membranes for regenerative medicine. Adv. Funct. Mater. 26: 7965–7974.