Journal Name REVIEW

Please do not adjust margins

Journal Name REVIEW Revitalizing Carbon Supercapacitor Electrodes with Hierarchical Porous Structures Received 00th January 20xx, Accepted 00th January 20xx DOI: 10.1039/x0xx00000x www.rsc.org/

a

b

a

Tianyu Liu, Feng Zhang, Yu Song and Yat Li

a

Carbon materials, owing to their excellent electrical conductivity, tailorability, inexpensiveness and versatility, have been extensively studied as electrode materials for supercapacitors. The capacitance of carbon-based supercapacitor electrodes remained at a mediocre level between 100 and 200 F g-1 for decades. Until recently, a new family of carbon materials termed hierarchical porous carbons has pushed the capacitance to new benchmark values beyond 300 F g-1, and has revitalized the exploration of carbon materials for supercapacitors. Hierarchical porous carbons contain different scales of pores (from micropores to macropores) inter-connected together and assembled in hierarchical patterns. Experimental studies coupled with theoretical investigations have elucidated that the presence of micropores are responsible for offering large surface area to enhance charge storage capability, whilst mesopores, macropores and the hierarchical structure improve electrolyte infiltration and facilitate ion diffusion. This review will start by introducing different pore types and the definition of hierarchical porous structures, followed by discussion and exemplification of major synthesis strategies. In addition, recent molecular-level understandings of the relationship between pore size, functionalities inside pores, pore spatial distribution and capacitive performance are presented. At last, challenges and future opportunities associated with hierarchical porous carbons for supercapacitors are discussed.

1. Introduction With the ever growing population, diminishing supplies of fossil fuels, and proliferating environmental pollutions, there is an urgent need for clean and sustainable energy. Extensive research efforts have been devoted for technology development to harvest and convert sustainable energies to electricity that can be readily used for electronics. However, due to the intermittent nature, unevenly regional distribution and unstable natural availability of these sustainable energy resources, reliable and efficient energy storage technology is required for full utilization of sustainable energy. Batteries and supercapacitors (a.k.a. electrochemical capacitors or ultra-capacitors) are two major energy storage devices. Figure 1a shows the Ragone plot that compares the energy density and 1 power density of supercapacitors and various batteries. Supercapacitors distinguish from batteries by their ultrahigh power density, i.e., the ability to be rapidly charged and discharged, at a timescale of seconds. Therefore, supercapacitors target applications where high power uptake or delivery and pulsed energy are needed: immediate power supply and recovery of stop-and-go systems, industrial energy management systems and electric vehicles (mostly 2, 3 used in conjunction with lithium-ion batteries), etc. Figure 1b depicts the major components of a supercapacitor, which consists a.

Department of Chemistry and Biochemistry, University of California, Santa Cruz, 1156 High Street, Santa Cruz, California 95064, United States. Email: [email protected] b. Key Laboratory for Advanced Technology in Environmental Protection of Jiangsu Province, Yancheng Institute of Technology, Yancheng 224051, People’s Republic of China

of a positive electrode (a.k.a. cathode), a negative electrode (a.k.a. anode), a separator in between the two electrodes to prevent electric short circuit, and an electrolyte filled inside the package, serving as an ion reservoir. The superior power density of supercapacitors is rooted in their charge storage mechanism. Two mechanisms are involved for supercapacitors (Figure 1c). The first category is electrical double layer (EDL) capacitance. In this case, charges are stored as ions adsorbed at electrode/electrolyte interfaces via electro-static interactions, forming the EDL. The second category is pseudocapacitance from pseudo-capacitive reactions (including underpotential depositions, near-surface redox reactions and fast ion 4 insertion and extrusion that trigger no phase transitions). Supercapacitors that store charges based primarily on EDLs and pseudo-capacitive reactions are called EDL capacitors and pseudocapacitors, respectively. No matter charges are stored by EDLs or pseudo-capacitive reactions, they are managed to accumulate and dissipate rapidly because they only interact with the electrode surface (for the EDL capacitance, under-potential depositions and near-surface redox reactions) or diffuse swiftly in bulk materials (for fast ion intercalation-deintercalation). This property renders 1 supercapacitors the superior rate performance. One of the ultimate goals for supercapacitor research is to achieve high charge storage capacity (measured by capacitance or energy density) at ultra-high scan rates or current densities. Carbon materials are promising candidates for realizing such a goal owing to their high electrical conductivities. The small electrical resistance enables fast electron transfer and transport, which is essential for high-rate performance. However, the gravimetric capacitance of

This journal is © The Royal Society of Chemistry 20xx

J. Name., 2013, 00, 1-3 | 1


Please do not adjust margins REVIEW -1

carbon had remained at a mediocre level between 100 and 200 F g 5 for decades. Previous studies found that the relatively small ionaccessible area of carbon materials is the main cause for the limited capacitance and energy density. Factors that lead to the limited ionaccessible area include: i) inadequate inherent surface area and/or ii) restricted ion accessibility to a large amount of surface. For example, carbon cloth woven by micron sized carbon fibers only 2 -1 have a specific surface area of ~5 m g [determined by the 6 Brunauer-Emmett-Teller (BET) method]. Activated carbon powder, 2 -1 though has a ultra-high surface area surpassing 1000 m g , its abundant micro-pores are barely accessible for ions. As a result, the typical gravimetric capacitance of activated carbons is only around -1 7 100 F g . The capacitance decreased to an even lower level when increasing scan rate or current density, due to the sluggish ion diffusion in the micro-pore dominated structures. Therefore, to enlarge the ion-accessible area of carbon materials while maintaining charge diffusion pathways becomes crucial to achieve high capacitance at ultrafast charging rates. The aforementioned challenge is being addressed. Capacitances -1 that exceed 300 F g have been achieved by a new class of carbon materials termed hierarchical porous carbons (cf. Table 1). They possess multi-scale pores interconnected together and assembled into hierarchical patterns. The hierarchical porosity facilitates electrolyte infiltration and ion diffusion, thus improves ion 8 accessibility of the entire electrode. We will begin this review by introducing the concept of hierarchical porosity, followed by sections covering various synthetic strategies for creating hierarchical porous carbon-based supercapacitor electrodes, and molecular-level understandings of charge storage behaviors in hierarchical pore systems. We will conclude the review with comments on future challenges and opportunities for both experimental works and theoretical studies. In 2014, Dutta et al. published a comprehensive review on the polymer and bio-mass derived hierarchical porous carbons for carbon dioxide storage, 9 photonic crystals, lithium-sulfur batteries, and supercapacitors. Here we will concentrate our efforts on recent advance of hierarchical porous carbons as supercapacitor electrodes since 2014.

2. Hierarchical Porosity The basic requirement for hierarchical porosity is that a porous system must contain multi-scale pores, i.e., must have at least two kinds of pores. In 1985, the International Union of Pure and Applied Chemistry (IUPAC) classified pores according to pore width. The pore width is defined as pore diameter, or specifically for slitshaped pores (i.e., space between two layers), inter-layer distance. 10 According to IUPAC, pores are classified into three categories:  Macropores, with pore widths larger than 50 nm;  Mesopores, with pore widths smaller than 50 nm but larger than 2 nm;  Micropores, with pore widths smaller than 2 nm. With the development of nanomaterials, the IUPAC elaborated 11 their classifications in 2015 to introduce three new pore types:  Nanopores, with pore widths smaller than 100 nm;  Supermicropores, a.k.a. large micropores, with pore widths smaller than 2 nm but larger than 0.7 nm;  Ultramicropores, a.k.a. small micropores, with pore widths smaller than 0.7 nm.

Based on the new definitions, nanopores include micropores, mesopores and macropores but with an upper limit of pore width of 100 nm. Supermicropores and ultramicropores are two subtypes of micropores. The classifications of these six pore types are summarized in Figure 2a. Porous systems that solely contain different sizes of pores are not necessarily qualified as hierarchical porous systems. The notion of “hierarchy” in a material requires a clear presentation of interplay between all different pores and formation of a hierarchical network. For supercapacitors that rely on ion diffusion for operation, different types of pores should be at least interconnected with each other and be accessible for ions. Figure 2b presents an illustration of a typical ion diffusion path in a hierarchical porous structure: ions enter first in largest pores and then flow into smaller ones that are subdivided from the larger pores. This pattern continues until ions reach the smallest pores. The most commonly used characterization methods for hierarchical porous structures are gas sorption and electron microscopy imaging techniques. Gas sorption measurements can gauge different pore widths and corresponding quantities, while electron microscopy imaging techniques enable direct observations on how different pores are engaged in a hierarchical pattern. An extensive review on textural characterizations of hierarchical porous structures is beyond the scope of this review, but it is 8 available in a recently published article.

3. Synthetic Strategies 3.1 Hard Template Methods Hard template methods have been commonly used in producing porous materials. The following procedure is often used for fabricating hierarchical porous carbons. First, hard templates with well-defined structures (e.g., particles and nanorods) are incorporated into carbons or carbon precursors. If carbon precursors are used, they need to be carbonized at elevated temperatures in inert atmosphere. Then the incorporated hard templates are removed by either chemical etching or dissolution, leaving behind the negative carbon replicas of the templates. These replicas typically contain macropores and mesopores depending on the size of templates used. In some cases involving carbon precursors, a small amount of micropores can also be generated during the pyrolysis step as a result of de-hydration and de12 hydrogenation of organic compounds. To increase the amount of micropores, chemical activation processes are usually carried out by mixing the obtained carbon replicas with carbon-corrosion agents, such as potassium hydroxide (KOH), and annealing at temperatures o beyond 600 C. The controlled carbon etching will introduce ample amount of micropores (and some mesopores) into the structure. This activation step has been widely applied to achieve hierarchical porosity. The detailed mechanism of chemical activations 13 (especially for KOH activation) has been reviewed by Wang et al. Here we will review the synthesis of hierarchical porous carbons via hard template methods, using silica, metal oxides, inorganic salts and sublimable organic compounds as templates.

2 | J. Name., 2012, 00, 1-3



Please do not adjust margins REVIEW

Journal Name

Figure 1. (a) Ragone plot compares the specific energy density and specific power density of supercapacitors and other batteries. 1 Reproduced with permission. Copyright 2015, American Chemical Society. (b) Structure schematic of a supercapacitor. (c) Schematic illustration of two possible charge storage mechanisms of supercapacitors: (c1) EDL consists of the inner Helmholtz plane (IHP, compact and ordered layer of ions) and outer Helmholtz plane (OHP, diffusive and loose layer of ions); (c2) surface redox reactions of ruthenium dioxide (RuO2, a typical pseudocapacitive material) in protonic electrolyte.

Figure 2. (a) IUPAC classifications of pores based on pore width. (b) Schematic illustration of an ion diffusion pattern in a typical hierarchical porous structure.


J. Name., 2013, 00, 1-3 | 3



Journal Name -1

3.1.1 Silica Templates Silica is the most popular hard template for producing hierarchical porous carbons, as it can be easily produced in large quantity and readily tailored to different sizes and morphologies 14-19 20-25 including: silica spheres, mesoporous silica templates, silica 26-28 29 30 31 powder, ultrafine silica moieties, silica fibers, zeolite and 32 silica shells . For example, Lin et al. reported the synthesis of nitrogen-doped, ordered mesoporous few-layer carbon (OMFLC-N) using porous silica with ordered mesoporosity as the template 20 (Figure 3a). They first deposited a thin layer of nitrogen-doped graphene by a Ni-assisted chemical vapor deposition method onto the wall of the silica template and then etched silica with hydrofluoric acid (HF). As shown in Figure 3b, the high angle annular dark-field transmission electron microscopy (TEM) image proves the existence of mesopores channels with microporous walls. BET measurements confirm that both mesopores and mircropores are presented (Figure 3c), in consistent with the TEM image. This hierarchical porous carbon structure is able to deliver an ultrahigh -1 -1 -1 capacitance of 855 F g at a current density of 1 A g and 615 F g -1 at 40 A g in 1 M sulfuric acid (H2SO4) aqueous electrolyte, showing its ultra-high capacitance and exceptional rate handling ability. -2 Besides, increasing mass loading up to 10 mg cm , reaching the commercial level, does not detrimentally affect the capacitance of the electrode (Figure 3d). The excellent capacitive performance was ascribed to the high accessible surface to electrolyte and the efficient ion diffusion with negligible geometric and electric hindrance inside the mesopore channels. Moreover, the electrode was reasonably stable in at least two different aqueous electrolytes (Figure 3e).

Figure 3. (a) Schematic illustration of the synthetic strategy. (b) High angle annular dark-field TEM image and (c) pore-size distribution diagrams obtained by nitrogen sorption experiment. OMC: ordered mesoporous carbon; OMCFL: ordered mesoporous few-layer carbon. (d) Gravimetric and volumetric capacitance evaluated at a current

density of 1 A g with various mass loadings in 0.5 M H2SO4 and 2 M lithium sulfate (Li2SO4) aqueous electrolyte. (e) Sustaining stability (tested by potential-holding) and cycling stability performance (tested by cyclic voltammetry) of a symmetric supercapacitor assembled with two identical OMCFL-N electrodes in two aqueous 20 electrolytes. Reproduced with permission. Copyright 2015, American Association for the Advancement of Science. 3.1.2 Metal Oxide Templates Dissolution of silica templates needs acute toxic chemicals such as HF or highly corrosive concentrated and hot alkaline solutions. These drawbacks have shifted researchers’ attention to metal oxide-based hard templates. Metal oxides that can be easily dissolved in diluted acids are promising alternatives to silica. Zinc 33-35 36 oxide (ZnO) and anodized alumina (Al2O3) film are two good examples. Wang et al. prepared flower-like hierarchical porous carbon spheres (FHPC) with flower-shaped ZnO nanosheet 33 clusters as the hard template (Figure 4a). Pitch was used as a carbon precursor. The synthesized hierarchical porous structure composes of numerous microporous and mesoporous carbon sheets that are assembled radially to form slit-shaped macropores (Figure 4b and 4c). As shown in Figure 4d, the pore width of FHPC spans from ultramicropores to mesopores, with double pore size peaks located at 1.3 nm and 4 nm. This FHPC electrode delivered a large specific -1 -1 capacitance of 294 F g at a scan rate of 2 mV s with 71% -1 capacitance retained at 500 mV s , which are substantially higher than the values obtained from its non-activated counterpart at the same scan rates (Figure 4e and f).

Figure 4. (a) The fabrication scheme of FHPC. (b) Scanning electron microscopy (SEM) image of a single carbon cluster. (c) High resolution transmission electron microscopy (HR-TEM) image showing the micropores on each carbon sheet. Comparison of (d) pore size distribution, (e) cyclic voltammograms (CVs), (f) gravimetric capacitances collected for FPC and FHPC. Electrochemical performance was evaluated in 6 M KOH aqueous 33 electrolyte. Figures reproduced with permission. Copyright 2014, Elsevier. 3.1.3 Salt Templates


J. Name., 2013, 00, 1-3 | 4


Please do not adjust margins Journal Name

REVIEW

Inorganic and organic salts are another groups of hard templates used for hierarchical porosity generation. Incorporation of salt templates is mostly carried out with carbon precursors. Salts are more versatile than silica and metal oxide templates given that they can react differently during the pyrolysis steps. According to their reactions in pyrolysis, salt templates can be classified into three categories: (1) Stable salt templates that remain unchanged; (2) Decomposable salt templates that yield metal oxides or metals and (3) Self-salt templates that directly produce carbon materials embedded with templates. Stable salt templates Ionic compounds with high thermal stability are grouped into 37-39 this category, such as sodium chloride (NaCl), potassium 38 37, 39 chloride (KCl), sodium silicate (Na2SiO3), sodium carbonate 37 40 (Na2CO3) and potassium hydroxide (KOH). These templates can be uniformly distributed if they are soluble in carbon precursors. As a result, pores generated by these salts have improved uniformity compared to templates that are not able to form a single phase with carbon precursors. For example, Zhu et al. synthesized 3D hierarchical porous carbon powders with multi-scale pores (HPCBMS) using glucose as the carbon precursor, and sodium chloride (NaCl), sodium carbonate (Na2CO3) and sodium silicate (Na2SiO3) as 37 salt templates. As shown in Figure 5a, by mixing these salts with glucose solution and followed by freeze-drying, the three salt templates were self-assembled into a 3D hierarchical pattern. NaCl crystals fused together and formed micrometer-sized aggregations. The walls of these aggregations were coated with Na2CO3 and Na2SiO3 particles with smaller size. The HPC-BMS with interconnected carbon network and highly porous walls (Figure 5bd) was obtained after carbonization of glucose and dissolution of salt templates in water. To elucidate the exact roles of different salt templates, the authors prepared control samples with different dominating pore sizes by eliminating one or two of three templates. They concluded that NaCl, Na2CO3 and Na2SiO3 templates were responsible for pore generation with size of 1-2 μm, 50-100 nm and 5-10 nm, respectively. Figure 5e shows the rate capability performance of three carbon structures with different porous structures. Among them, the HPC-BMS achieved the highest -1 -1 gravimetric capacitance of 320 F g at 0.5 A g , and more importantly, it retained a gravimetric capacitance as high as 126 F g 1 -1 at an ultrahigh current density of 200 A g . Eliminating any size of pores leaded to severe capacitance loss at large current densities, indicating that nanopores play an important role in rate capability.

Figure 5. (a) Schematic demonstration of the fabrication of HPCBMS. (b) Low magnification, and (c) high magnification SEM image of the macroporous network with highly porous walls and (d) HRTEM image of HPC-BMS showing the presence of mesopores and micropores. Scale bars in (b), (c) and (d) are 1 μm, 500 nm and 10 nm, respectively. (e) Gravimetric capacitances evaluated by galvanostatic charge-discharge experiments in 6 M KOH aqueous solution. HPC-B and HPC-BM represent carbon powders synthesized with NaCl+Na2CO3 and NaCl templates, respectively. Figures 37 reproduced with permission. Copyright 2015, Royal Society of Chemistry. Decomposable salt templates Decomposable salt templates can either decompose to metal oxides during pyrolysis of carbon precursors, or further react with the as-formed carbon and be converted to metals. The yielded metal oxides or metals serve as hard templates that can be readily dissolved by acidic or alkaline solutions to create macropores or mesopores. Particularly, carbonates, nitrates, organics and ammonium salts are preferred to yield lightweight porous carbons owing to the vast amount of gaseous products (e.g., carbon dioxide) generated during the carbonization step. These gaseous products are able to exfoliate or activate the as-formed carbon structures according to the following chemical reaction: CO2 (g) + C(s) → 2CO(g)

(1) 41, 42

both can significantly decrease the mass density of products. Decomposable salt templates that have been demonstrated 43 recently are magnesium acetate (converted to MgO), nickel 44, 45 hydroxide [Ni(OH)2, converted to Ni], , sodium iodide (converted 46 47 to sodium metal), sodium chloroacetate (decompose to NaCl), tetraethylorthosilicate (TEO, decompose to silica), calcium


J. Name., 2013, 00, 1-3 | 5


Please do not adjust margins REVIEW 48-50

carbonate (decompose to CaO) , zinc nitrate (decompose to 51 ZnO) and potassium acid phthalate (decompose to K, K2O and 52 K2CO3). Liang et al. adopted a hydrothermal method to synthesize nitrogen-doped hierarchical hollow nest-like carbon (NHHNC) with 44 Ni(OH)2 and zinc chloride (ZnCl2) as a salt templates (Figure 6a). The polysaccharide (carbon precursor) was first incorporated with Ni(OH)2 and then mixed with ZnCl2 powder. The mixture was o annealed at 700 C in nitrogen atmosphere. During this process, the formed nickel-zinc alloys constituents and nickel metal particles were removed by the hydrothermal reaction in hydrochloric acid solution to generate a porous bird-nest like morphology (Figure 6b). The porous structure containing micro-, meso- and macro-pores -1 (Figure 6c and 6d). The achieved gravimetric capacitance at 1 A g -1 -1 reached a remarkable value of 322 F g , with 175 F g capacitance -1 remained at 20 A g (Figure 6e). Self-salt templates Unlike all the aforementioned templates that need to be manually incorporated into carbon precursors prior to 53 carbonization, metal complexes such as magnesium citrate, , zinc 43 54 52 salicylate, sodium gluconate, , potassium acid phthalate, 55 56 poly(acrylamide-co-acrylic acid) potassium salt, sodium alginate 57-61 and metal organic frameworks (MOFs) have been utilized as precursors that result in hard templates incorporated carbons directly upon carbonization. Metal complexes are composed of metal ions or clusters coordinated with organic ligands. Upon pyrolysis, the metal ion or cluster cores are decomposed to either metal or metal oxide particles, and the ligands are reduced to carbon. The obtained hard templates incorporated carbons are subjected to the same treatments as other hard template methods to synthesize hierarchical porous carbons. Because these complexes result in both templates and carbons without addition of other templates and carbon precursors, they are called self-templates. The 3D hierarchical porous graphitic carbon networks 62 (HPGC) synthesized by Ma et al. is one example. They first built egg-box shaped 3D pectin coordinated magnesium composite foams relying on electrostatic attraction between negatively charged pectin bio-polymer backbones and positively charged magnesium ions (Figure 7a). The asprepared complex foams were then annealed in nitrogen gas o at 500 C to carbonize pectin and decompose magnesium core to MgO, followed by KOH activation. The activated 3D carbon/MgO foams were washed with diluted hydrochloric acid to remove MgO and any inorganic impurities, resulting in 3D carbon foam-like frameworks formed with partiallygraphitic carbon flakes (Figure 7b and c). Besides macropores that are clearly observed in the SEM image, abundant micropores and mesopores were also detected by BET method (Figure 7d). The incorporated MgO template was believed to play dual roles in forming the observed morphology: i) the large MgO particles were responsible for macroporosity generation and the small MgO crystallites served as catalysts for partially-graphitic carbon sheets growth. The HPGC showed -1 -1 a gravimetric capacitance of 274 F g at 1 A g and more -1 significantly, sustained a large capacitance of 234 F g at 50 A

-1

-1

g (Figure 7e). The gravimetric capacitance achieved at 50 A g is substantially higher than most of other reported values for graphene-based supercapacitor electrodes. 3.1.4 Organic Templates

Besides inorganic materials, some organic materials which are soluble in non-polar organic solvents or thermal 63 decomposable such as tetramethylammonium oxalate and 64 polyurethane , have been studied as hard templates to produce hierarchical porous carbons. The most widely used 17, 65-67 organic hard template is poly-styrene (PS). Incorporated PS templates can be easily washed away by hot acetone or burned away in air. PS spheres and silica particles were employed by Chaudhari et al. to synthesize the ordered 17 multimodal porous carbon (OMPC, Figure 8a). First the PS beads with diameters around 400 nm were self-packed into a hexagonal pattern with silica particles (ca. 20 nm in diameter) filled the interstitial sites. Subsequent pyrolysis of the PS/silica composite in air eliminated PS spheres and simultaneously sintered silica particles. Afterwards, phenolic resin (the carbon precursor) was deposited onto the silica network followed by high-temperature annealing to produce carbon. The silica particles were eliminated by immersing in 2 M NaOH solution. The yielded OMPC contains hexagonal patterned and interconnected macropores with walls full of mesopores, as shown in both SEM and TEM (Figure 8b and 8c), rendering a 2 -1 large BET surface area of 1161.6 m g . The OMPC exhibited much higher capacitance than activated carbon with comparable surface area (Figure 8d). Specifically, at a large -1 current density of 10 A g , the OMPC achieved a capacitance -1 of 152 F g , about three times higher than the capacitance of -1 activated carbon obtained at the same current density (59 F g , Figure 8e). The capacitance enhancement was ascribed to the facilitated ion diffusion between mesopores via the ordered and interconnected macropores.

6 | J. Name., 2012, 00, 1-3




Journal Name

Figure 6. (a) Schematic illustration of the synthesis of NHHNC. (b) SEM image showing the porous bird nest-like morphology of NHHNC and (c) TEM image highlighting the porous wall. Inset of (c) shows the mesopores and micrpores on the carbon surface. (d) Pore size distribution of NHHNC (red) and NHHNC without ZnCl2 (NHHNC-na, black). (e) Rate capability performance of NHHNC, NHHNC-na and 44 carbon spheres. Figures reproduced with permission. Copyright 2015, Royal Society of Chemistry.


J. Name., 2013, 00, 1-3 | 7



Figure 7. (a) Schematic diagram showing fabrication process of the HPGC. (b) SEM image and (c) HR-TEM image of the as-formed 3D hierarchical porous carbon framework. Inset of (c) presents the selected electron diffraction pattern collected from regions highlighted by the dotted ellipses, revealing weak crystallinity of the partially-graphitic carbon sheets. (d) Pore size distribution of HPGC0.6-700 and HPGC0800. (e) Gravimetric capacitances of a single electrode tested in a two-electrode system. Note: HPGC0.6-700 is the HPGC. HPGC0-800 is a control 62 sample without KOH activation. Figures reproduced with permission. Copyright 2016, Royal Society of Chemistry.

8 | J. Name., 2012, 00, 1-3




REVIEW

Figure 8. (a) Schematic representation of the fabrication of OMPC. (b) SEM image and (c) TEM image of the OMPC. (d) CVs of OMPC and -1 activated carbon collected at 10 mV s in 1.0 M H2SO4 aqueous electrolyte. (e) Rate capability performance of OMPC and activated carbon. 17 Figures reproduced with permission. Copyright 2014, Royal Society of Chemistry 3.1.5 Ice Template Use of ice templates to prepare hierarchical porous carbons is usually associated with a freeze-drying process, a.k.a. lyophilization. A general protocol of ice template methods involves the following steps: (1) Aqueous solutions containing carbon precursors are prepared and frozen. In some cases, the aqueous solutions will be converted to hydrogels by adding gelation agents; (2) The frozen samples are transferred into a freeze-dryer, in which the chamber is vacuumed to a pressure below the pressure of water’s triple point (611.65 Pa); (3) Slowly increase the chamber temperature to room temperature and ice will start to sublimate. After this step, aerogels (or cryogels) will be obtained; (4) Carbonize the aerogels to obtain the porous carbons. Step 3 and 4 sometimes are done altogether. Since ice crystals are usually within micron size, porous carbons obtained from ice templates usually contain abundant macropores but little mesopores and micropores. Therefore, subsequent

chemical activations are usually required to create hierarchical porosity. Li et al. synthesized hierarchical porous carbon foams as 2+ supercapacitor electrodes with Ca -crosslinked sodium alginate 68 aerogels as precursors. Hu et al. synthesized hierarchical porous N-doped carbons by freeze drying cellulose hydrogels and 69 carbonizing corresponding aerogels in ammonia gas. You et al. prepared boron and nitrogen co-doped hierarchical porous carbon networks by freeze drying agarose hydrogels containing tris base, boric acid and ethylenediaminetetraacetic acid followed by 70 carbonization and KOH activation. Zhang et al. recently developed porous carbon foams (PCFs) from glutaraldehyde-crosslinked and K2CO3-embedded chitosan hydrogels using freeze-drying and 42 subsequent carbonization in nitrogen atmosphere (Figure 9a). The PCFs consist of 3D carbon networks with interconnected porous thin carbon sheets (Figure 9b). These PCFs possess large BET surface 2 -1 area of 1013.0 m g due to the presence of abundant micropores and mesopores (Figure 9c). The 3D inter-connected carbon network was created during the freeze-drying process as a result of eliminating ice templates (Figure 9d). The formation of smaller


J. Name., 2013, 00, 1-3 | 9


Please do not adjust margins REVIEW macropores and the meso-/micro-pores on each carbon sheet were ascribed to phase separation and chemical activation by K2CO3, respectively. The CVs (Figure 9e) and galvanostatic charge-discharge profiles (Figure 9f) of PCF all exhibited near-ideal characteristic EDL capacitive behavior. The gravimetric capacitance of PCF calculated -1 -1 from the discharge time at 0.5 A g was 246.5 F g . It also retained -1 67.5% of the capacitance (166.3 F g ) when current density -1 increased to 100 A g (Figure 9g). These excellent electrochemical performances were attributed to the rapid ion diffusion in the hierarchical porous structure formed by thin nanosheets. In addition, the highly conductive carbon matrix enabled fast electron transfer that also contributed positively to the outstanding rate capability performance. Later, Zhang et al. further improve the rate capability of carbon foam by introducing multiscale pore network (CF-MSP) through the 71 combination of ice and silica template methods. The self-standing carbon aerogel possess large micron-sized pores lined with highly porous carbon sheets. Each sheet is decorated with ~200 nm cavities formed by removal of silica spheres (Figure 10a and 10b). Furthermore, KOH activation creates a large number of micropores (Figure 10c). All the three types of pores interconnect with each other and construct a hierarchical porous structure with extremely 2 -1 large surface area of near 3000 m g (determined by BET). In addition, the carbon foam is inherently doped with nitrogen heteroatoms that are able to improve the wettability of the electrode and contribute pseudo-capacitance. Owing to the large surface area, unique hierarchical porous network, small electrical resistance, and improved electrolyte wettability, CF-MSP exhibited -1 a remarkable gravimetric capacitance of 374.7 ± 7.7 F g at a -1 current density of 1 A g , and exception rate capability with -1 capacitance of 235.9 ± 7.5 F g still retained at an ultrahigh current -1 density of 500 A g (Figure 10e). Another important family of hierarchical porous carbons made from ice template methods is porous graphene aerogels. Graphene aerogel is a form of graphene monolith with 3D networks built up 72 with many randomly oriented graphene sheets. Colloidal graphene oxide suspensions, made by deep oxidation of graphite powder, primarily serve as precursors of graphene aerogels. Graphene hydrogels are assembled by hydrothermally treated -1 concentrated graphene oxide solution (> 2 mg mL ) at o 72 temperatures above 100 C. Followed by freeze-drying the asprepared graphene hydrogels, graphene aerogels are obtained. The volume and shape of the graphene aerogels can be readily adjusted by the amount of precursor suspensions and shape of the reactors used for hydrothermal reactions. Similar as other materials produced from freeze-drying, graphene aerogels contain a large number of macropores. Two methods have been commonly used to create hierarchical porous structures in graphene aerogels. The first method is to mix porogens (the chemical agents that can introduce pores) together with graphene oxide suspensions before hydrothermal reactions. Demonstrated porogens include hydrogen 73, 74 75 76 peroxide (H2O2), carbon dioxide, Schiff base, concentrated 77 78 nitric acid, , ammonia solution and carbon nanofibers/carbon 79 nanotubes. The second method is to chemically activate the graphene aerogels after freeze-drying. For example, Yun et al. demonstrated that abundant mesopores and micropores can be 75 introduced on each graphene sheets by carbon dioxide activation.

Xu et al. developed an alternative H2O2-assisted hydrothermal method to prepare holey graphene aerogels 74 (HGFs, Figure 11a). During the hydrothermal reaction, H2O2 corroded the defective sites of graphene oxide and created abundant in-plane pores with diameters of a few nanometers. As a result, the as-prepared holey graphene aerogel frameworks (HGFs) possessed inter-connected 3D macroporous network with mesoporous and microporous graphene sheets (Figure 11b-d). The continuous network of open channels allows efficient ion transport between adjacent graphene layers, as well as across the entire structure (Figure 11e). When the HGF is compressed to a piece of self-standing thin film and tested in 1-ethyl-3-methylimidazolium tetrafluoroborate/acetonitrile electrolyte, it exhibited nearrectangular CV (Figure 11f) and sloping galvanostatic chargedischarge profiles (Figure 10g), indicating the HGF electrode has near-deal capacitive behavior and very low internal resistance. Significantly, the HGF electrode showed a −1 gravimetric capacitance of 298 F g at a current density of

10 | J. Name., 2012, 00, 1-3




Journal Name

Figures 9. (a) Schematic illustration of the synthesis of PCF. (b) SEM images of PCF. (c) Comparison of pore size distribution of PCF and carbon foam with non-porous carbon sheets (CF). (d) Schematic illustration of the formation mechanism of PCF. Step A: Freezing. The expansion of liquid water squeezes the chitosan “cages” along the directions indicated by red arrows. Step B: Freeze-drying. Ice “islands” sublimate and the macro-pores are left. SEM image shows the porous structure created by ice sublimation. (e) CVs collected at various scan rates and (f) galvanostatic charge-discharge profiles collected at various current densities in 3 M KOH aqueous electrolyte. (g) Gravimetric 42 capacitance and volumetric capacitance of PCF plotted as a function of current density. Figures reproduced with permission. Copyright 2016, Springer.


J. Name., 2013, 00, 1-3 | 11



Figure 10. (a) Top-view SEM image of CF-MSP and (b) side-view SEM image of a piece of carbon sheet of CF-MSP. Scale bars: (a) 25 μm and (b) 1 μm. Inset of (a) is a digital picture of a piece of self-standing CF-MSP (scale bar: 1 mm). (c) Pore size distribution calculated from the absorption-desorption isotherm using Barrett-Joyner-Halenda method (for pore size ranges between 2 nm and 300 nm) and Saito-Foley method (for pore size < 2 nm). (d) N 1s XPS spectrum. The black lines are experimental data that can be de-convoluted into several synthetic peaks (dashed curves). The solid curve is the summation of all the synthetic peaks. Percentages of different carbon species and Nfunctionalities are evaluated based on area of synthetic peaks. N-Q: quaternary nitrogen; N-6: pyridine-like nitrogen. (e) Gravimetric capacitance measured at different current densities. The error bars represent standard deviations evaluated based on data collected in 71 triplicate. Figures reproduced with permission. Copyright 2017, American Chemical Society.

12 | J. Name., 2012, 00, 1-3




REVIEW

Figure 11. (a) Schematic illustration of the synthetic strategy of HGFs and HGF films. (b) A digital picture showing a HGF hydrogel. (c) SEM image showing the 3D continuous network made of graphene sheets. Scale bar is 1 μm. (d) TEM image of a piece of holey graphene sheet of HGFs. Scale bar is 10 nm. (e) Diagram illustrating the presence nanopores facilitates ion transport between adjacent graphene layers. (f) CVs and (g) galvanostatic charge-discharge profiles of HGF in 6 M KOH aqueous electrolyte. (h) Comparison of specific capacitance of HGF 74 and non-holey graphene electrodes evaluated at different current densities. Figures reproduced with permission. Copyright 2014, Nature Publishing Group. −1

−1

1 A g . Increasing the current density to 100 A g , a high −1 gravimetric capacitance of 202 F g was retained (Figure 11h). 3.2 Soft Template Methods Unlike hard templates, soft templates are “soft” because they do not have solid shapes. Soft templates are mainly organic molecules or block copolymers. In water, soft templates are typically self-assembled with each other into bundles, a.k.a. micelles, with their charged and hydrophilic tales facing out. These charged tales attract nearby carbon precursors via electro-static interactions. These carbon precursors are then linked to the soft templates through covalent bonds by co-polymerizations, forming rigid organic micelles wrapped with carbon precursors. The obtained carbon precursor-micelle composites are then carbonized in inert atmosphere. This annealing process leads to thermal decomposition or evaporation of the soft templates and pyrolysis of the carbon precursors. Because organic molecules are assembled orderly when forming micelles, hierarchical porous carbons generated by these soft template methods

usually contain ordered mesopores. The pattern of these ordered pores depend on shapes of micelle: cylinder micelles tend to form hexagonal patterns while spherical micelles result 80 in cubic patterns. Besides, soft template methods save the template removal step after pyrolysis is another major advantage compared to hard template methods. 3.2.1 Surfactant Templates Owing to their bi-polar structure and ability to form micelles in the solvent, a number of surfactants have been chosen as soft templates for preparation of hierarchical porous carbons. Surfactants direct carbon precursors with opposite charges to form specific configurations through electro-static interactions with carbon precursors. For example, Liu et al. manipulated the positively charged hexadecyl trimethylammonium ions and negatively charged resol-silicate composite particles in water to assemble spherical micelles. The micelles were subsequently converted to nitrogen-doped 81 porous spheres. Another example is the carbon nanotube (CNT)-bridged graphene 3D building blocks prepared by Pham


J. Name., 2013, 00, 1-3 | 13


Please do not adjust margins REVIEW et al. using hexadecyl trimethylammonium bromide (CTAB) as 82 the soft templates. They first mixed CNTs and CTAB in deionized water. The interaction between intrinsically positive charged CNTs and negatively charged ends of CTAB surfactants induced graft of CTAB molecules on CNT surface, with the positively charged ends of CTAB sticks outward. A third component, graphene oxide with inherent negative charges was then introduced into the CNT-CTAB mixture. Electrostatic attraction favors the binding of graphene oxide sheets onto the positive ends of CTAB, eventually forming a CNT pillared graphene oxide supermolecular structure (Figure 12a and 12b). The resultant structure was then mixed with KOH (activating agent), vacuum-filtered and annealed in nitrogen gas to obtain the hierarchical porous CNT-bridged graphene electrode with the slit-shaped pores and nanopores on each graphene sheet (Figure 12c). By examining the morphology using SEM (Figure 12d) and polarized Raman spectroscopy, the authors claimed that most CNTs stood vertically in between graphene layers. Owing to the hierarchical porosity, the electrode exhibited 199 -1 -3 F g (equivalent to a volumetric capacitance of 211 F cm ) at a -1 -1 -3 current density of 0.5 A g , and retained 199 F g (105 F cm ) -1 at 20 A g in an ionic liquid electrolyte (Figure 12e).

Another group of soft templates is organic polymers. Different from surfactants, organic polymers bind to the carbon precursors through hydrogen bonds and/or react with carbon precursors to form covalent bonds. The most popular polymer template used is a triblock copolymer called poly(ethylene oxide)-b-poly(propyleneoxide)-b83-92 poly(ethylene oxide), (F-127). F-127 is composed of a central hydrophobic block of polypropylene glycol and two hydrophilic blocks of polyethylene glycol at two ends. All oxygen atoms on the polymer chain are able to form hydrogen bonds with hydrogen atoms of carbon precursors. The resultant F-127 (the carbon precursor) micelles are spherical and yield hollow carbon spheres with ordered mesoporous walls. Besides F-127, other co-polymers such as poly(ethylene 93 glycol)-poly(propylene glycol)-poly(ethylene glycol), poly- (494 vinylpyridine)-b-poly-(ethylene glycol), and poly(styrene-r95 methylacrylic acid) as well as homogeneous polymers 96-98 including poly(vinylpyrrolidone) and poly(methyl 99 methacrylate) are alternatives when synthesizing hierarchical porous carbon-based supercapacitor electrodes. 3.3 Template-free Methods Template-free methods are gaining popularity in synthesizing hierarchical porous carbons due to their facile nature. Biomasses and pre-engineered organic matters with intrinsic macro-pores represent two major precursors. A typical protocol of template-free methods is alike any template methods but without the need of template removal step. First, carbon structures with primarily macropores are produced by pyrolysis of selected carbon precursors. Second, a large amount of mesopores and micropores are introduced by chemical activation to construct hierarchical porous structures. Sometimes these two steps can be combined into one step by mixing carbon precursors with chemical activating agents. Based on the origin of precursors, hierarchical porous carbons produced by template-free methods can be classified into two groups: the bio-mass derived carbons and synthetic materialderived carbons. 3.3.1 Biomass-derived Hierarchical Porous Carbon Electrodes

Figure 12. (a) The ingredients used and (b) a schematic illustration of the CNT-bridged graphene supermolecular architecture. (c) A digital picture of a self-supporting film electrode. (d) Cross-sectional SEM image revealing the vertically aligned CNTs between graphene layers. (e) Rate capability performance of the CNT-bridged graphene thin film evaluated in an ionic liquid electrolyte of 1-ethyl3-methylimidazolium tetrafluoroborate. Red data points and black data points represent capacitance of KOH-activated sample and non-activated sample, respectively. Figures reproduced with 82 permission. Copyright 2015, American Chemical Society. 3.2.2 Organic Polymer Templates

A variety of natural biomasses have been selected as precursors for hierarchical porous carbon. Four kinds of bio-masses have been demonstrated recently: 100, 101 41 (a) plant-derived materials: cotton fibers, sucrose, 41, 102 41, 103 104 105 glucose, cellulose, nitrocellulose, wood fibers, rice 106 107 108 109 110 husk, corn husk, loofah, bamboo, water bamboo, 111, 112 113 114 115 bagasse, tobacco rods, wheat flour, watermelons, 116 117 118 pomelo peels, sun-flower heads, filter papers, water 119 120 121 122 123 hyacinth, , willow catkins, algae, white clover, hemp, 124 125 Jujun grass and sisal leaves . 126 127-129 (b) animal-derived materials: artemia cyst, gelatin, 130 131 132 41, 133 silk, bee honey, shrimp shell, chitosan/chitin, and 134 sheep manure . 135 (c) fungi 84, 136 (d) sewage sludge Biomass-derived carbons inherently contain hetero-atoms such as N and S. These hetero-atoms are believed to be pseudo-

14 | J. Name., 2012, 00, 1-3




REVIEW

capacitive active and can contribute additional pseudo-capacitance. As a result, bio-mass derived carbons usually exhibit higher capacitance than carbons with limited amount of hetero-atoms. Zhang et al. synthesized 3D hierarchical porous carbon powders (3DHPC) by direct pyrolysis of sheep manure, a 134 renewable bio-mass waste. The prepared carbon powders after KOH activation were composed of graphene-like sheets with nano- and micro-tubular pores (Figure 13a and 13b). The combination of micro-, meso- and macro-pores rendered 2 -1 3DHPC large surface area (ca. 1000 m g ) and small ion diffusion resistance, which are preferable for supercapacitor electrodes. In addition, the 3DHPC contained heteroatoms N, O, and S. These heteroatoms not only enhanced the surface wettability, but also contributed pseudo-capacitance. As a result, the 3DHPC achieved outstanding capacitive performance. The constant current charge/discharge profile -1 collected at an ultrahigh current density of 50 A g showed negligible IR drop and sloping feature (Figure 13c). As shown in Figure 13d, CVs maintain near-rectangular shape up to 500 mV -1 s . These results indicate that the electrode has ultra-small electrical resistance. Significantly, as depicted in Figure 13e, -1 -1 the gravimetric capacitance of 3DHPC at 1 A g and 50 A g -1 -1 was 486 F g and 411 F g , respectively, which are among the state-of-the-art capacitive performance achieved by carbonbased supercapacitor electrodes. Besides, the porous carbon structure is extremely stable. Less than 1.5% capacitance loss was detected for 10000 cycles (Figure 13f).

Figure 13. (a) Low magnification and (b) high magnification SEM images of 3DHPC showing its hierarchical porous structure. Scale bars in (a) and (b) are 10 μm and 500 nm, respectively. (c) Galvanostatic charge-discharge profiles collected at 10, 20 and 50 A -1 g . (c) CVs recorded with various scan rates ranging from 5 to 500 -1 mV s . (e) Rate capability performance and (f) long-term cycling

stability of 3DHPC. All electrochemical characterizations were carried out in 6 M KOH aqueous electrolyte. Figures reproduced 134 with permission. Copyright 2016, Wiley. 3.3.2 Synthetic Materials-derived Hierarchical Porous Carbon Electrodes An intrinsic limitation of bio-masses derived hierarchical porous carbons is difficult to control pore size and pore distribution. To circumvent this disadvantage, synthetic organic materials with preengineered porous structure or tunable structures are selected as carbon precursors. Such precursors can be divided into two categories: (a) Synthetic polymers which demand high temperature annealing to convert to carbons. Examples are proton exchange 137, 138 139 membranes, poly(p-phenylenediamine), poly(vinylidene 140 141 fluoride), pyromellitic dianhydride, 142 poly(methylmethacrylate)/poly(acrylonitrile), 143-145 146 147, poly(acrylonitrile), poly(urethane–amide), poly(aniline), 148 149 poly(acrylic acid)/methylated melamine-formaldehyde resins, 150, 151 152 phenolic resins, resorcinol–formaldehyde resins, 153 154 phloroglucinol, sodium polyacrylate and Schiff base networks 155 . (b) Carbon-based materials that can be directly used as active 156 materials. Examples are commercial graphene, graphene 146, 157 158-160 26, 161oxide, carbon nanotubes, reduced graphene oxide, 164 162 165 159 carbon black, urea and activated carbon . Zhu et al. has recently demonstrated a direct ink writing process that was able to engineer graphene aerogels into 3D supercapacitor electrodes with ordered macropores and microporous structures 26 (Figure 14). The three-stage process included preparation of printing inks, direct ink writing and post-treatments (Figure 14a). Unlike conventional non-porous thick electrodes that suffer from sluggish ion diffusion in the interior space, the 3D-printed graphene aerogels (with a thickness of ca. 1 mm) exhibited outstanding rate capability that is comparable to other highly conductive carbon electrodes 10-100 times thinner (Figure 14c). These results unambiguously illustrate that ordered pore distribution is beneficial for facilitating ion diffusion within electrode interior space. Following this work, Liu et al. further demonstrated that treating the 3D-printed electrodes in lithium perchlorate electrolyte by a two-step ion intercalation strategy can double the gravimetric capacitance without sacrificing their superior rate capability and 164 structural integrity. Most recently and inspired by Gogotsi group’s pioneering works on two-dimensional layered metal carbide/nitride compounds, or 166, 167 MXenes, their carbon derivatives - hierarchical porous carbon sheet stacks are developed as supercapacitor electrodes. MAX, a group of electrically conductive ceramic materials (where M and A represent metal elements and X stands for carbon or nitrogen), have strong in-plane chemical bonds but weak out-of-plane bonds. This graphite-like interaction enables selectively etching of metal layers to produce carbon sheets. The obtained carbon sheets possess large slit-shaped interlayer pores and various nanopores on each carbon sheet surface, forming a continuous hierarchical porous structure. For example, Ding et al. developed a two-step top-down method to


J. Name., 2013, 00, 1-3 | 15



Journal Name

Figure 14. (a) Schematic illustration of the 3D printing process to fabricate graphene aerogel electrodes with ordered macro-porosity. (b) SEM image showing the surface of a filament consisted of the graphene aerogel (scale bar is 1 μm). The inset shows the macro-pores (scale 26 bar: 250 μm). Figures reproduced with permission. Copyright 2016, American Chemical Society. synthesize hierarchical porous carbon sheet stacks using a ternary 168 layered carbide, Ti3AlC2 (Figure 15a). They first extracted the aluminum layers with HF. Then the titanium layers were then removed by annealing Ti3C2 in chlorine gas at temperatures higher o 168, 169 than 700 C, according to the following reaction: Ti3 C2 (s) + 6Cl2 (g) → 3TiCl4 (g) + 2C(s) (2) This hierarchical porous structure is expected to enable ultrafast ion diffusion throughout the entire structure (Figure 15b). The as-obtained carbon nanosheets inherited the precursor’s sheet morphology (Figure 15c and 15d). Each carbon nanosheet contains abundant in-plane micropores and mesopores. The carbon sheet stacks exhibited a specific -1 -1 capacitance of 220 F g at 0.5 A g , much higher than the values obtained from control groups. Meanwhile, when the -1 current density was increased up to 20 A g , the carbon sheets could still retain as high as ca. 79% of its initial value, reaching -1 158 F g (Figure 15e). Besides, the electrode could retain ca. 95% capacitance retention over 4000 cycles at a current -1 density of 2 A g (Figure 15f).


J. Name., 2013, 00, 1-3 | 16



REVIEW

Figure 15. (a) Schematic illustration of the two-step etching strategy for synthesis of 2D ultrathin carbon nanosheet stacks. (b) Schematic illustration of rapid ion transport through the ultrathin carbon nanosheets. SEM images of (c) the Ti3C2 precursor and (d) carbon sheets product. (e) Comparison of specific capacitances at different current densities (legend: MDC-700: Ti3C2-derived carbon sheets o with an annealing temperature of 700 C; CDC-900: Ti3AlC2-derived carbon sheets; MDC-900: Ti3C2-derived carbon sheets with an o annealing temperature of 900 C). (f) Cycling stability performances -1 168 of MDC-900 at 2 A g . Figures reproduced with permission. Copyright 2016, Royal Society of Chemistry.

4. Molecular-level Understandings of Charge Storage Behaviors in Hierarchical Porous Carbons Understanding ion behaviors and charge storage mechanism in hierarchical porous carbons is central to the success of pushing the performance of carbon electrode performance. For theoretical studies, the key is to understand how ions diffuse and are stored in different porous structures. It is generally accepted that the presence of macropores and large mesopores facilitates ion diffusion, whilst small mesopores and micropores increase ionaccessible surface area and thus contribute significantly to capacitance. However, the role of ultramicropores has been controversial. They are not preferred because ion diffusion to such tiny pores was believed to be extremely difficult, which can lead to decreased capacitance. However, in 2006, Chmiola et al. observed an anomalous capacitance enhancement when pore size became 170 smaller than ions. This observation spurred a variety of follow-up researches aiming to reveal the mechanism. Soon ion de-solvation 171 172 theory and superionic state theory were proposed. The two theories are correlated with each other: based on molecular-level simulations, the ion de-solvation theory states that ions need to deshell their solvation shells before entering into pores with sizes comparable or smaller to themselves (bare ion size). Due to the desolvation effect, ions could stand closer to each other and were able to be packed more densely, reaching a superionic state. As an increasing amount of ions being stored, the capacitance increases. Since 2014, theoretical studies have continued to verify the aforementioned picture. Efforts can be mainly divided into three directions: (a) Elucidating the relationship between capacitance and pore width (especially the size of nanopores); (b) Understanding the effect of surface functionalities anchored on inner walls of micro-pores; (c) Explaining the influence of pore shapes and pore spatial distributions on overall capacitive performance. 4.1 Effects of Pore Width 4.1.1 Capacitance Novel electrode materials with ultramicropores are developed as platforms to study ion behaviors in tiny pores. One example is carbide-derived carbons (CDCs), which can be obtained by selectively removal of metal layers of precursors, as discussed in Section 3.3.2. These carbons equip with rich micro- and meso-pores. -2 They achieved large areal capacitance (around 300 mF cm ) due to their large surface area, but also excellent rate capability 169 performance. Such rate capability performance is counter-

intuitive as ion diffusion in micropores is believed to be sluggish. Extensive efforts have been devoted to explain this “abnormal” observation. One of the greatest achievements is experimental proof to the ion de-solvation theory. The direct evidence was captured by electrode weight different at different applied potentials by an electrochemical quartz crystal microbalance (EQCM). By comparing the mass change of two CDCs with average pore size of 1 nm and 0.65 nm (denoted as CDC-1 and CDC-0.65 respectively), Tsai et al. observed that the mass change of CDC-0.65 was smaller than CDC-1 at any identical applied bias. EQCM measurements revealed that the mass difference equals the mass 173 of solvent taken by ions. Because pores in CDC-0.65 are smaller than those in CDC-1, ions entering smaller pores tend to de-solvate. De-solvation reduced overall ion mass and hence decreased the amplitude of mass change during charge and discharge processes. 7 Consistently, with the aid of Raman spectroscopy and Li nuclear 7 magnetic resonance ( Li NMR), Urita et al. calculated the solvation number of lithium ions in micropores and compared that in mesopores with propylene carbonate (PC) as the solvent. They observed a significantly reduction in solvation number when lithium 174 ions diffused from mesopores to micropores. Most recently, Prehal et al. quantified the degree of ion de-solvation using in-situ 175 small-angle X-ray scattering. They concluded that the tight attachment of the aqueous solvation shell effectively prevent complete de-solvation of ions in carbons with sub-nanometer pores. The de-solvation theory has been applied to explain the ultrahigh capacitance achieved by nanoporous carbons. Galhena et al. studied the capacitance change of a piece of GO paper with the number of charge-discharge cycle in 1 M tetraethylammonium 176 tetrafluoroborate (TEABF4) electrolyte dissolved in PC. Despite the monotonically increase in interlayer distance of GO planes triggered by irreversible intercalation of PC molecules, the corresponding capacitance change followed a different pattern: it initially increased to a peak value and then drastically decreased when further enlarging the inter-layer distance. Using in-situ X-ray diffraction technique, they found that the peak capacitance was achieved when the inter-layer distance is comparable to the bare tetraethylammonium cations. Their experimental observations again confirmed the ion de-solvation theory. Additionally, the authors claimed that the de-solvation could bring ions closer to electrode surface and thus reduced the distance between ions and electrode surface. Since capacitance is inversely proportional to the ion-electrode surface distance, the capacitance is therefore expected to increase. It was considered as an elaboration on the super-ionic state theory. Besides, Bañuelos et al. utilized fully atomistic molecular dynamics simulations to probe the concentration of ions in micropores and in bulk electrolyte. Ion concentration in micropores was found to be significantly higher 177 than that in bulk electrolyte. It validated the claim of super-ionic state theory. However, it is noteworthy that the “abnormal” capacitance increment can only be observed when pore size is comparable to ion size. If the pore width is larger than bare ions, decreasing pore size will result in capacitance loss due to inaccessibility of solvated ions to pore interior space. The peak value centered at where pore size matches bare ion size. Zhi et al. extended this trend to mesopore region. This trend is illustrated in 178 Figure 16a. Governed by a trade-off relationship between surface


J. Name., 2013, 00, 1-3 | 17


Please do not adjust margins REVIEW area and utilization efficiency of pore walls, capacitance reached a peak value when the mesopore width was approximately 2.5-3 times larger than the diameter of solvated ions. On the other hand, the relationship between pore size and capacitance still remains controversial. In 2011, using density functional theory (DFT), Centeno et al. calculated the specific capacitance of 28 porous carbons with comparable surface area but different pore sizes ranging from 0.7 nm to 15 nm. Surprisingly, all the carbons exhibited nearly constant areal capacitance of -2 179 0.094±0.011 F cm . In an extension of their work, the authors again experimentally confirmed their calculation results (Figure 16b). Capacitance was only directly related to the solvated ion 180 accessible surface contributed from pores larger than 0.63 nm. They believed that the reported capacitance enhancement was an unreliable result because electrode surface area was measured by BET method. The probe molecules (usually di-nitrogen, N2) used for the BET test was not able to accurately probe the surface inside ultramicropores, due to inaccessibility and incomplete formation of 181 adsorption layer. As a result, BET results may under-estimate surface area of electrode material, and thus, capacitance will be over-estimated when it is normalized to the measured surface area. This argument was supported by some other researchers. For example, Wu et al. believed the presence of ultra-small pores cannot be probed by nitrogen or argon molecules (two commonly 182 used probe molecules) at cryogenic temperature. Furthermore, they pointed out that the accuracy of surface area tested at the ultra-low temperature (e.g., 77 K, the liquid nitrogen temperature) should be re-examined, since low temperature did not reflect the true testing environment of supercapacitor electrodes (mostly at or close to room temperature). Another observation that neither agrees with the 183 aforementioned two mechanisms were reported by Jiang et al.. They performed a classical DFT calculation and observed that for a nanopore with pore width 1-10 times the ion diameter, the theoretical capacitance of the nanopore oscillated as the pore width increased (Figure 16c). The origin of the oscillation was ascribed to overlapping of EDLs: due to the narrow pore size, the EDLs from the two charged walls behave as two standing waves, and can superimpose with each other. The maxima in capacitance appear where the interference of the two EDLs is most constructive, i.e., the density of one kind of ion (anion or cation) from one pore wall overlaps most with that of the same-charged ion from the other wall. Although the ions were simulated as hard spheres in their study, modifications of the spheres did not greatly alter the overall picture. It should be noted that the prerequisite for observing such an oscillation pattern is an extremely narrow pore width distribution, which is challenging to be achieved by experimental methods.

perturbation, they inspected that the relatively large supermicropores were first filled by ions and then ions diffused to relatively small supermicropores and ultramicropores in proximity of the initially filled pores. Other small pores which were far away from the large pores were filled at last. This observation suggested that besides macropores, large supermicropores and possibly mesopores could also serve as electrolyte reservoirs. The full charging time of the electrode was directly related to the average pore size. Following this work, the same group further estimated the time-scale for various steps associated with ion diffusion inside 185 nanopores with and without applied external bias (Figure 17). Under zero potential, ion de-solvation happened much faster than ion adsorption (ca. 1 ps vs. ca. 150 ps), and the in-pore diffusion was the slowest (ca. 1 ns). In addition, the diffusion rate inside nanopores is one magnitude order lower than that in the bulk electrolyte. Moreover, when nanopores were electrically charged with a potential of -1 V, anions entered into pores first followed by filling of cations. This phenomenon is called “ion slashing” and its 186 detailed mechanism was explained by Qiao and coworkers. Under bias, the counter ion diffusion time became roughly four times longer compared to the uncharged state. But in both cases, ion diffusion was fast enough to ensure the excellent rate capability performance. This picture is in good agreement with other experimental studies carried out by Prehal et al. where they experimentally justified the “ion slashing” using in-situ X-ray 175 scattering measurements. While Pean et al.’s studies suggested that ion diffusion becomes more sluggish when confined in a limited space, He et al. showed 187 that this is not always true. They observed that ion diffusion coefficient inside narrow space could exceed that in the bulk electrolyte depending on how small the pore width is. They simulated the diffusion behavior of both cations and anions in an ionic liquid by varying the size of a slit-shaped pore. When the pore was so narrow that only allowed a single layer of ions to pass through, the diffusion coefficient is significantly larger than that of -9 2 -1 bulk electrolyte (reaching ca. 5×10 m s ). Once the pore size is sufficient big enough to accommodate multi-layer of ions, the ion diffusion abruptly slowed down because of the relatively high ion packing density that causes “traffic jam”. Most recently, Forse et al. studied the ion-electrode interactions using in-situ diffusion nuclear magnetic resonance 188 spectroscopy. They concluded that the pores wider than 2 nm are able to alleviate ion diffusion congestion inside micropores, and consequently promoting outstanding rate capability performance. This finding is in consistent with an experimental study carried out 189 by Wei et al.. The authors concluded that a wide micropore size distribution was beneficial for achieving outstanding rate capability performance.

4.1.2 Rate Capability

4.2 Influence of Functionalities

In addition to the efforts on understanding how the pore width dictates capacitance, researchers also interested in understanding the relationship of ion diffusion behavior and the rate capability performance of hierarchical porous carbons. Pean et al. simulated the charging dynamics of a nanoporous electrode with 1.1 nm in thickness (similar as the CDC structure) in an ionic liquid 184 electrolyte. Upon applying a transient electric potential

The surface of experimentally synthesized porous carbons is always anchored with inherent or intentionally introduced functionalities, such as oxygen-containing moieties and nitrogen dopants. Understanding the role of these surface functional groups on capacitive performance is therefore critical. However, theoretical studies involving surface functionalities are rarely reported. Dyatkin et al. compared the configurations of planar

18 | J. Name., 2012, 00, 1-3




Journal Name

Figure 16. (a) Plot of theoretical specific capacitance of porous carbons with pore size ranging from micro-pore region to mesopore region (left). Illustrations of ions residing in pores with R/r=1.5, 2 and 4, where R stands for pore radius and r is ion radius. The interior surface of mesopores is able to be gradually fully covered by solvated ions, meaning the utilization efficiency of pore inner surface increases with 178 increasing mesopore size. Figure reproduced with permission. , Copyright 2014, Royal Society of Chemistry. (b) Variation of areal capacitance of the carbon monoliths (M1, M2, M1-A), with other porous carbons and CDCs added for comparison. Figure reproduced with 180 permission. Copyright 2015, Royal Society of Chemistry. (c) Capacitance (the y-axis) of a nanopore as a function of the pore width in terms of nm (the top x-axis) or the ion diameter (the bottom x-axis, σ = 0.5 nm). All the peaks are labelled in the unit of σ. Figure 183 reproduced with permission. Copyright 2011, American Chemical Society.


J. Name., 2013, 00, 1-3 | 19



Figure. 17 Characteristic times extracted from molecular simulations for CDC-like porous carbon electrodes held at 0 V (right) and -1 V (left). The black regions represent carbon walls; red spheres, green spheres and blue ellipses are cations, anions and solvent molecules, 185 respectively. Figure reproduced with permission. Copyright 2015, American Chemical Society.

Figure. 18 (a, b) Cross-sectional (left) and top-view (right) scheme of (a) de-oxygenated pore and (b) oxygenated pore filled with ions. Gray, white, red, purple, and blue spheres represent carbon, hydrogen, oxygen, fluorine, nitrogen and sulfur atoms, respectively. Figures 190 reproduced with permission. Copyright 2016, American Chemical Society. (c) Schematic illustration of ion accessibility to the internal space of single-walled carbon nano-horns. Blue spheres represent ions. Red solid arrows and blue dashed arrows represent rapid and 193 sluggish ion diffusion direction, respectively. Figures reproduced with permission. Copyright 2015, American Chemical Society. cations and planar anions in an ionic liquid when confined in oxygenated pores and de-functionalized pores (Figure 18a and 190 18b). Simulation results showed that the de-functionalized pore allowed ions to pack near-perpendicularly to the pore surface and parallel to each other due to weak interaction between carbon walls and ions. This compact packing arrangement allows a large number of ions to be stored in pores, but also create traffic congestion that slows down ion transport. The oxygenated pore, on the contrary, has stronger ionophicity that managed to attract ions (especially the anions) to the closest of the surface, aligning parallel to the surface. This configuration led to decreased ion packing density but a relatively clear transport channel. These proposed mechanisms are consistent with the different capacitive performance obtained by the group: oxygen-rich carbons have lower capacitance but better rate capability than their defunctionalized counterparts under the same degree of ion confinement. In addition, they suggested that the electrolyte is easier to be decomposed in oxidized pores via a dimerization-

191

induced breakdown mechanism, preventing a device reaching large potential windows. Similar conclusions were obtained by Li et 7 al. on studying the shape of ions. Zhan et al. studied the effects of nitrogen-doping on the performance of graphene by probing the electronic density of state 192 near the Fermi level. They found that while the pyrrolic-N was unable to alter the electronic structure of graphene, the graphitic-N and the pyridinic-N can shift the “Dirac point” to lower energy and higher energy, respectively. In both cases, shifting the “Dirac point” resulted in increased population of energy states near the Fermi level, which consequently enhanced quantum capacitance, and hence, overall capacitance of the doped graphene. 4.3 Influence of Pore Spatial Distribution Spatial distribution of different types and sizes of pores is another critical factor influencing capacitive performance. Pore distribution determines the capacitance mainly by controlling the ion accessibility. As shown in Figure 18c, neither small pores on

20 | J. Name., 2012, 00, 1-3




REVIEW

surface nor large pores buried in the interior space can easily be reached by ions. Yang et al. studied how pore distribution on single 193 walled carbon nano-horns affected capacitive performance. They found that pores on top of each horn with slightly larger size than solvated ions can be more easily reached by ions compared to tiny on-tip pores and deep large pores. Electrochemical impedance studies carried out by Ervin also suggested that dense graphene films can achieve large capacitance at high frequencies if ordered 163 inverse-cone shaped pore arrays are engineered into the films. Zhu et al. further elaborated on Ervin’s conclusion and proposed that ordered distributed pores can facilitate ion diffusion in thick electrodes by reducing structural tortuosity according to the 26 following equations: 𝐷𝑒𝑓𝑓 = 𝑘 𝜏=

𝐷 𝜖

𝐿2 𝐷𝑒𝑓𝑓

(3) (4)

Where Deff is the diffusion coefficient of a fluid diffuses through a porous structure, D is diffusion coefficient in pores, k is porosity, ε is tortuosity, τ is diffusion time that is inversely proportional to diffusion rate, L is ion transport length. Ordered porous structures have much smaller ε values than structures with randomly distributed pores, owing to the well-aligned diffusion channels that reduce tortuosity. As a result, ordered porous structures have much smaller τ than their stochastic porous counterparts (assuming all other variables remain unchanged), meaning that ion diffusion in ordered channels structures is expected to be faster than in stochastically pore networks.

5. Conclusions Hierarchical porous carbons, with their unique connectivity of various scales of pores, have already displayed superior performance over traditional carbon materials. A number of hierarchical carbon-based materials have achieved outstanding -1 gravimetric capacitance beyond 300 F g . Table 1 selectively summarizes the structural parameters and capacitive performance of hierarchical porous carbons with maximum capacitances higher -1 than 300 F g . Clearly, the capacitances at small current densities have reached a new level. Yet, an undeniable fact is that most hierarchical porous carbons fail to retain their outstanding capacitance at rapid charge and discharge rates, i.e., they exhibit poor rate capability performance. This is particularly unfavorable for supercapacitors, a charge storage device expected to have large power density. Overcoming this bottleneck requires long-term joint efforts from both theoretical and experimental studies. Molecular simulation is a powerful tool for investigating ion behaviors in complex electrochemical systems and expected to provide experimentalists the important insights on how to design and construct the “silver bullet”. Previous studies have been primarily focused on modelling ion diffusion within ultra-small pores, while revealing the collaborative roles of hierarchical pores is equally important but rarely studied. For example, the roles of mesopores and macropores in facilitating ion diffusion deserve further investigation, possibly including the influences from heteroatoms. Also, as can be seen from Table 1, the correlation between total pore/micro-pore volumes and capacitances seems rather ambiguous. Simulations on hierarchical porous systems that take

different pore sizes, structures and distributions into consideration could possibly solve the puzzle. Another direction that needs to be explored is simulating ion behaviors in aqueous electrolytes. To date, most theoretical studies choose ionic liquids as electrolyte mainly because ions in ionic liquids have comparable size as pore widths; however, aqueous electrolytes are more popular in supercapacitor community owing to safety consideration, ease of fabrication and fast ion diffusion rate. Changing ionic liquids to aqueous electrolytes will bring significant changes in simulations: presence of solvation shell, different ion size and diffusion coefficients, and possible pseudo-capacitive contributions from doped heteroatoms, etc. The new results might alter our current understandings on ion-pore interactions and possibly induce unprecedented discoveries that inspire experimentalists to design novel structures to further push the performance of supercapacitors. Once optimal structures are obtained by simulation, the next big challenge is to explore synthetic strategies that can achieve the proposed electrode architecture and experimentally confirm the simulation prediction. As shown in Table 1, at this stage, most highperformance carbons are derived from direct pyrolysis of biomasses due to the hierarchical porosity and rich amount of heteroatoms, indicating bio-mass derived carbons are promising carbon precursor candidates. However, template-free methods have limitations in tuning the pore structures. On the other hand, both hard and soft template methods are able to change pore structures simply by tailoring the shape and number of templates and how they are incorporated in host materials. Therefore, a promising synthetic strategy to achieve the optimal structures is to combine these methods. For example, if the ion-desolvation theory holds true for aqueous electrolytes, electrodes with comparable + + + pore size to ultra-small de-solvated ions (e.g., H , Li , Na ) will perhaps achieve much higher capacitance than the state-of-the-art level. It is because ions in aqueous electrolytes are generally much smaller than ions in ionic liquids, resulting in more compact ionpacking. However, how to acquire these ultra-fine pores remains challenging and deserves extensive efforts in material synthesis. Finally, it is extremely critical to develop a set of fair and consistent metrics for evaluating performance of porous carbon materials. First, gravimetric capacitance, areal capacitance and volumetric capacitance are three equally important performance metrics. In terms of portability and assuming the volume of electrodes or devices is not a concern, gravimetric capacitance is the most meaningful among the three. Whilst for applications with limited space for installing energy storage devices, areal capacitance and volumetric capacitance are more significant than gravimetric capacitance. Given that most hierarchical porous carbons are ultralight (low mass density), they are promising electrodes for light-weight charge storage devices but are disadvantageous in areal and volumetric capacitance. Therefore, it is unreasonable to judge their performance by merely taking their areal or volumetric capacitance into consideration. Second, it should be noted that high gravimetric/areal/volumetric capacitance will become less meaningful if the corresponding active mass loading/working area/total volume is ignored. Taking gravimetric capacitance as an example, the gross capacitance, or total capacitance (CT), is


J. Name., 2013, 00, 1-3 | 21


Please do not adjust margins REVIEW evaluated by multiplying gravimetric capacitance (CS) by active material mass loading (m): 𝐶T = 𝐶S × 𝑚 (5) CT will be ultra-small if the mass loading is tiny. It infers that a supercapacitor with limited amount of active material, though might exhibit outstandingly high gravimetric capacitance, will still be impractical as it can only store a small amount of charges. Similar arguments can be made for areal capacitance and volumetric capacitance. Moreover, because capacitance is a function of not only scan rate or current density, but also mass loading, arbitrarily comparing capacitance without considering mass loading is unfair for high active mass loaded electrodes. It is therefore critical for researchers to be aware of this issue, and present mass loading, area or volumetric capacitance for evaluation and comparison. It is more meaningful to push performance at high mass loadings rather than to pursue large numbers with ultra-small amount of active materials.

22 | J. Name., 2012, 00, 1-3




Journal Name

Table 1. Structural and electrochemical properties of selected hierarchical porous carbons (sorted ascendingly based on the highest capacitance reported). Electrode Beehive-like Hierarchical Nanoporous Carbon Foam ZnCl2-activated and Filter Paperderived Carbon Three-Dimensional Porous Nitrogen-Doped Carbon Networks Loofah Sponge-derived Carbon Foam Resol-derived Porous Carbon Holey Graphene Aerogel Mantis Shrimp Shell-derived Carbon Powder Three-dimensional Beehive-like Hierarchical Porous Polyacrylonitrile-based Carbons Microporous Carbon Nanofibers Hierarchical Porous Carbon Microbead Bagasse Wastes-derived Hierarchical Porous Carbon Salt-assembly-derived Carbon Powder Glucose-derived Nitrogen-doped Hierarchical Hollow Nest-like Carbon Nanostructures

Synthetic Method

Electrolyte

Surface Area 2 -1 a) (m g )

Mircro-pore 3 -1 Volume (cm g )

Total Pore Volume 3 -1 (cm g )

Template Free (Biomass)

6 M KOH

1472

0.48

0.61

Template Free (Biomass) Template Free (Synthetic Material) Template Free (Biomass) Hard Template (Salt) Hard Template (Ice) Template Free (Biomass) Template Free (Synthetic Material) Hard Template (Salt) Template Free (Synthetic Material) Template Free (Biomass) Hard Template (Salt) Hard Template (Salt)


6 M KOH 6 M KOH 6 M KOH

2232 1442 1733

-

c)

1.49

-

1.13

-

1.05

6 M KOH

2753

-

1.54

6 M KOH

1560

-

-

6 M KOH

1223

-

1.07

Highest Capacitance -1 (F g ) -1 301 F g -1

(100 A g )

-1

-1

302.3 F g -1

-

1.48

6 M KOH

2164

0.81

1.01

6 M KOH

2269

0.65

1.97

6 M KOH 6 M KOH 6 M KOH

2296 1088 707

0.58

0.86

-

1.32

0.28

0.82

-1

(50 A g )

-1

-1

226 F g

-1

(20 A g )

-1

-1

304 F g

(0.5 A g ) 304 F g

-1

-1

183 F g

-1

(1 A g )

(50 A g )

-1

-1

307 F g

-1

(1 A g ) 310 F g

-1

(1 A g )

178 F g

-1

(40 A g ) -1

(100 A g ) -1

(20 A g )

-1

-1

237 F g

-1

(20 A g )

-1

210 F g

314 F g

(0.5 A g ) 314 F g

-1

(20 A g )

-1

193 F g

320 F g

-1

(0.05 A g )

-1

-1

(100 A g ) 227 F g

-1

(50 A g )

-1

-1

320 F g 320 F g

-1

(0.5 A g )

-1

126 F g

-1

(200 A g ) 175 F g

-1

(20 A g )

(1.0 A g )

55.8

118

74.3

139

60.2

108

58.0

40

76.5

74

87.2

132

75.5

142

66.5

27

60.3

194

70.9

111

39.4

37

54.3

44

-1

-1

322 F g

109

-1

-1

(0.5 A g )

63.8

-1

-1

(0.5 A g )

Ref.#

b)

-1

272.6 F g

-1

(0.3 A g )

Rate Capability (%)

-1

237 F g

312.6 F g

J. Name., 2013, 00, 1-3 | 23


168.8 F g

(1 A g )

-1

2085

-1

(0.1 A g )

-1

6 M KOH

Lowest Capacitance -1 (F g ) -1 192 F g

-1

-1

Please do not adjust margins REVIEW Nitrogen- and Oxygen-enriched 3D Hierarchical Porous Carbon Fibers Graphene/N-enriched Carbon Aerogel Nitrogen-doped Porous MultiNano-Channel Nanocarbon Jujun Grass-derived Hierarchical Porous Carbon Sandwich-type Graphene Based Ndoped Carbon

Journal Name Template Free (Synthetic Material) Hard Template (Ice) Soft Template (Polymer) Template Free (Biomass)

KOH 6 M KOH 1 M H2SO4 6 M KOH

2177 424 840 2532

0.32 0.2 0.52

0.52 0.6 1.97 1.08

Template Free (Synthetic Material)

6 M KOH

1749

-

0.41

Honeycomb-like Porous Carbon


6 M KOH

2725

-

1.28

Sunflower Head-derived Oxygenenriched Hierarchical Porous Carbon


6 M KOH

1032

-

0.60

Shrimp Shell-derived Nitrogendecorated Hierarchical Porous Carbon

Hard Template

6 M KOH

1343

-

0.68

Artemia Cyst-derived Oxygen-Rich Hierarchical Porous Carbon N‑Doped Carbon Aerogels Three-Dimensional Hierarchical Porous Graphene/Carbon Composite N-doped Carbon Spheres Corn Husk-derived Carbon Powder Polyaniline-derived Porous Carbon Tubes Highly N-doped Worm-like Carbon Composites

24 | J. Name., 2012, 00, 1-3

(Salt) Template Free (Biomass)

6 M KOH

1758

0.54

0.76

Hard and Soft Template

6 M KOH

1626

-

1.69


30 wt% KOH

2034

0.28

1.78

Soft Template (Polymer) Hard Template (Salt) Template Free (Synthetic Material) Hard and Soft Template

-1

223 F g

-1

(20 A g )

-1

-1

130 F g

-1

(10 A g )

-1

-1

150 F g

-1

(35 A g )

-1

-1

222 F g

-1

(10 A g )

-1

-1

203 F g

-1

(50 A g )

-1

212 F g

329 F g

(0.1 A g ) 335 F g

(0.1 A g ) 335 F g

(0.25 A g ) 336 F g

(1.0 A g ) 340 F g

(0.5 A g ) 342 F g

0.38

0.49

928

0.37

0.51

6 M KOH

3253 2563

1.08 0.61

1.72 1.52

-1

(20 A g )

-1

290 F g

-1

-1

(2 A g )

-1

-1

324 F g

-1

(10 A g )

-1

-1

160 F g

-1

(10 A g )

348 F g

(0.05 A g ) 349 F g

(0.5 A g ) 354 F g

(0.2 A g )

-1

-1

-1

-1

-1

(0.2 A g ) 356 F g

-1

(1.0 A g ) 365.9 F g

(50 A g )

66.1

124

59.7

102

62.0

116

80.9

117

83.3

49

92.8

126

45.2

91

52.6

150

55.1

86

84.3

107

60.0

147

72.5

21

-1

-1

-1

272 F g

-1

(10 A g )

375 F g

67

-1

300 F g-1 (10 Ag ) 219.5 F -1g (10 A g )

(0.5 A g )

44.8

-1

196 F g-1 (10 Ag )

-1

(0.1 A g )

76

-1

186.5 F g

-1

38.8

-1

-1

356 F g

144

-1

354.8 F g

(0.1 A g )

67.8

-1

-1

(1.0 A g )



-1

279 F g

-1

6 M KOH

-1

-1

345 F g

-1

6 M KOH

-1

(20 A g )

(0.2 A g )

-1

1200

-1

-1

-1

6 M KOH

-1

-1


REVIEW -1

Multiscale Porous Carbon Foam Schiff Base-derived Carbon Powder 3D Hierarchically Porous AllCarbon Foams 3D Honeycomb-like Carbon Powder

Hard Template Template Free (Synthetic Material) Soft Template (Polymer) Template Free (Biomass)

3 M KOH

2905

0.93

4.53

6 M KOH

1377

0.34

0.64

6 M KOH

1286

-

1.1

1 M Na2SO4

2839

-

2.65

-

-

Graphitized Hierarchical Porous Carbon Nano-spheres


6 M KOH

1665

0.19

0.84

Arch-shaped Nano-carbons


1 M H2SO4

619

-

0.66

(0.2 A g ) 379 F g

-1

(0.2 A g ) 379 F g

-1

(0.5 A g ) 383.2 F g

-1

(1.0 A g ) -1

402.5 F g

-1

(1.0 A g ) -1

421 F g

-1

(0.5 A g ) -1

3D Hierarchical Carbon Nanotubes/Activated Carbon Powder


6 M KOH

2278

0.52

1.39

GO-induced Layered Porous Carbon


6 M KOH

1476

-

0.058


6 M KOH

1313

-

-

Sheep Manure-derived Hierarchical Porous Carbon Powder

-1

-1

1816

Template Free

377 F g

-1

6 M KOH

(Synthetic Material)

(1.0 A g )

-1


Sandwiched Porous Carbon/Graphene Hybrids

-1

-1

3D Nitrogen-doped Activated Graphene-like Nanosheets

Flour-derived Honeycomb-like Carbon Foam

375 F g

6 M KOH

2927

1.23

1.78


6 M KOH

1000

-

0.61

Nitrogen-Enriched Hierarchically Porous Carbons


6 M KOH

1580

0.45

0.73

Three-dimensional Hierarchical Porous N-rich Graphitic Carbon


1 M H2SO4

409.15

-

1.66

440 F g

-1

(1.0 A g )

-1

236 F g -1 (100 Ag )

275 F g

-1

313 F g

-1

(20 A g )

-1

-1

-1

641.6 F g -1

-1

-1

411 F g

-1

(50 A g ) -1

(50 A g )

-1

ca. 600 F g

-1

(2 A g ) -1 g-1

75.0

152

65.3

98

97.0

159

48.6

129

58.1

114

65.1

195

84.6

134

69.0

76

84.5

165

-1

443 F g

(1 A g ) 710 F g

138

-1

-1

(1 A g )

84.7

-1

(20 A g )

-1

136

-1

426.8 F-1g (5 Ag )

-1

486 F g

52.8

-1

275 F g-1 (20 Ag )

-1

481 F g

89

-1

301.9 F -1g (50 A g )

(3 A g )

(0.5 A g )

29.3

-1

324.4 F -1g (20 A g )

-1

(0.5 A g )

155

-1

200 F g-1 (50 Ag )

221 F g

473 F g

59.4

-1

111 F g-1 (20 Ag )

-1

(0.5 A g )

71

-1

224 F g-1 (30 Ag )

-1

455 F g

60.0

-1

-1

(20 A g )

A Few-layer Nitrogen-doped Hard Template 715 F 0.5 M H2SO4 1900 2.2 20 Mesoporous Carbon (Silica) (1 A g ) a) Notes: Surface areas are determined by BET method unless otherwise stated. b) Rate capability is defined as the ratio of lowest capacitance to highest capacitance. c) “-” means not reported or not applicable.


J. Name., 2013, 00, 1-3 | 25



Conflict of interest There are no conflicts to declare.

Acknowledgements Y. Li acknowledges support by National Aeronautics and Space Administration (NASA) grant no. NNX15AQ01. T. Liu thanks the financial support from the Chancellor’s Dissertation-year Fellowship awarded by University of California, Santa Cruz. F. Zhang extends gratitude to Natural Science Foundation of Jiangsu Province (no. BK20141262).

Author Biographies

Tianyu Liu received his BS degree of Chemistry from the University of Science and Technology, Beijing, China in 2012 and joined the Prof. Yat Li’s laboratory at University of California, Santa Cruz, United States thereafter. He is now a Ph.D. in chemistry. His research focuses on development of functional materials for sustainable energy harvesting, conversion and storage with an emphasis on supercapacitors.

Yu Song received his B.S. in chemistry from Jilin University, China and received his Ph.D. in physical chemistry from Northeastern University in China. During 2015-2017, he joined Prof. Yat Li’s laboratory as a visiting student at University of California, Santa Cruz, United States. His research interest focuses on graphene-, metal oxide- and conducting polymer-based energy storage and conversion devices.

Yat Li received his B.S. and Ph.D. in chemistry from the University of Hong Kong. He was a postdoctoral research fellow at Harvard University from 2003 to 2007 under the supervision of Prof. Charles M. Lieber. He joined the University of California, Santa Cruz in 2007 and is now an associate professor of chemistry. His research focuses on the design and synthesis of semiconductor nanostructures, investigation of their fundamental properties, and exploration of their potential for energy conversion and storage.

References Feng Zhang received his BS in Material Chemistry in 2000 from Nanjing University of Science and Technology; MS in Applied Chemistry in 2005 from Shantou University, and Ph.D. in Macromolecular Chemistry and Physics in 2009 from Fudan University. He joined Yancheng Institute of Technology in 2000 and he is now an Associate Professor of Chemistry. His current research focuses on the design and synthesis of functional polymer nanocomposites and carbon-based materials, investigation of their functional properties and exploration of their potential applications in energy storage.

1 2 3

4 5 6 7 8 9

B. D. McCloskey, J. Phys. Chem. Lett., 2015, 6, 3592-3593. X. Lu, M. Yu, G. Wang, Y. Tong and Y. Li, Energy Environ. Sci., 2014, 7, 2160-2181. R. R. Salunkhe, Y.-H. Lee, K.-H. Chang, J.-M. Li, P. Simon, J. Tang, N. L. Torad, C.-C. Hu and Y. Yamauchi, Chem.-Eur. J., 2014, 20, 13838-13852. V. Augustyn, P. Simon and B. Dunn, Energy Environ. Sci., 2014, 7, 1597-1614. L. L. Zhang and X. S. Zhao, Chem. Soc. Rev., 2009, 38, 2520. G. Wang, H. Wang, X. Lu, Y. Ling, M. Yu, T. Zhai, Y. Tong and Y. Li, Adv. Mater., 2014, 26, 2676-2682. Y. Li, S. Roy, T. Ben, S. Xu and S. Qiu, Phys. Chem. Chem. Phys., 2014, 16, 12909. K. A. Cychosz, R. Guillet-Nicolas, J. García-Martínez and M. Thommes, Chem. Soc. Rev., 2017, 46, 389-414. S. Dutta, A. Bhaumik and K. C.-W. Wu, Energy Environ. Sci., 2014, 7, 3574-3592.


J. Name., 2013, 00, 1-3 | 26


Please do not adjust margins Journal Name 10

11

12 13 14

15 16 17 18 19 20 21 22

23 24 25 26

27 28 29 30 31

32

33

REVIEW

K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pieroti, J. Rouquerd and T. Siemieniewska, Pure Appl. Chem., 1985, 57, 603-609. M. Thommes, K. Kaneko, A. V. Neimark, J. P. Olivier, F. Rodriguez-Reinoso, J. Rouquerol and K. S. W. Sing, Pure Appl. Chem., 2015, 87, 9-10. X. Yu, J. Zhao, R. Lv, Q. Liang, C. Zhan, Y. Bai, Z.-H. Huang, W. Shen and F. Kang, J. Mater. Chem. A, 2015, 3, 18400-18405. J. Wang and S. Kaskel, J. Mater. Chem., 2012, 22, 23710-23725. Y. Sun, R. B. Sills, X. Hu, Z. W. Seh, X. Xiao, H. Xu, W. Luo, H. Jin, Y. Xin, T. Li, Z. Zhang, J. Zhou, W. Cai, Y. Huang and Y. Cui, Nano Lett., 2015, 15, 3899-3906. S. Wang, L. Zhang, F. Han, W.-C. Li, Y.-Y. Xu, W.-H. Qu and A.-H. Lu, ACS Appl. Mater. Inter., 2014, 6, 11101-11109. J. Tang, T. Wang, R. R. Salunkhe, S. M. Alshehri, V. Malgras and Y. Yamauchi, Chem.-Eur. J., 2015, 21, 17293-17298. S. Chaudhari, S. Y. Kwon and J.-S. Yu, RSC Adv., 2014, 4, 38931-38938. X. Fang, J. Zang, X. Wang, M.-S. Zheng and N. Zheng, J. Mater. Chem. A, 2014, 2, 6191-6197. Y. Zhang, Y. Wang and A. Hu, RSC Adv., 2015, 5, 70297-70301. T. Lin, I.-W. Chen, F. Liu, C. Yang, H. Bi, F. Xu and F. Huang, Science, 2015, 350, 1508-1513. G. Sun, L. Ma, J. Ran, B. Li, X. Shen and H. Tong, Electrochim. Acta, 2016, 194, 168-178. M. Enterría, A. Castro-Muñiz, F. Suárez-García, A. MartínezAlonso, J. M. D. Tascón and T. Kyotani, J. Mater. Chem. A, 2014, 2, 12023. M.-H. Kim, K.-B. Kim, S.-M. Park and K. C. Roh, Sci. Rep., 2016, 6, 21182. C.-W. Lee, S.-B. Yoon, H.-K. Kim, H.-C. Youn, J. Han, K. C. Roh and K.-B. Kim, J. Mater. Chem. A, 2015, 3, 2314-2322. H. S. Kim, Y. H. Kim, K. C. Roh and K.-B. Kim, Nano Res., 2016, 9, 2696. C. Zhu, T. Liu, F. Qian, T. Y.-J. Han, E. B. Duoss, J. D. Kuntz, C. M. Spadaccini, M. A. Worsley and Y. Li, Nano Lett., 2016, 16, 3448-3456. J. Sheng, C. Ma, Y. Ma, H. Zhang, R. Wang, Z. Xie and J. Shi, Int. J. Hydrogen Energy, 2016, 41, 9383-9393. D. Liu, W. Zhang, H. Lin, Y. Li, H. Lu and Y. Wang, RSC Adv., 2015, 5, 19294-19300. H. Zhou, Y. Peng, H. B. Wu, F. Sun, H. Yu, F. Liu, Q. Xu and Y. Lu, Nano Energy, 2016, 21, 80-89. D. Jia, X. Yu, H. Tan, X. Li, F. Han, L. Li and H. Liu, J. Mater. Chem. A, 2017, 5, 1516-1525. R. Ruiz-Rosas, M. J. Valero-Romero, D. Salinas-Torres, J. Rodríguez-Mirasol, T. Cordero, E. Morallón and D. CazorlaAmorós, ChemSusChem, 2014, 7, 1458-1467. T. Yang, R. Zhou, D.-W. Wang, S. P. Jiang, Y. Yamauchi, S. Z. Qiao, M. J. Monteiro and J. Liu, Chem. Commun., 2015, 51, 2518-2521. Q. Wang, J. Yan, Y. Wang, T. Wei, M. Zhang, X. Jing and Z. Fan, Carbon, 2014, 67, 119-127.

34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

J. Chen, Y. Liu, W. Li, H. Yang and L. Xu, RSC Adv., 2015, 5, 98177-98183. G. Zhang, Y. Song, H. Zhang, J. Xu, H. Duan and J. Liu, Adv. Funct. Mater., 2016, 26, 3012-3020. Y. Xue, Y. Ding, J. Niu, Z. Xia, A. Roy, H. Chen, J. Qu, Z. L. Wang and L. Dai, Sci. Adv., 2015, 1, 1400198. S. Zhu, J. Li, C. He, N. Zhao, E. Liu, C. Shi and M. Zhang, J. Mater. Chem. A, 2015, 3, 22266-22273. D. Tang, S. Hu, F. Dai, R. Yi, M. L. Gordin, S. Chen, J. Song and D. Wang, ACS Appl. Mater. Inter., 2016, 8, 6779-6783. S. Zhu, J. Li, L. Ma, L. Guo, Q. Li, C. He, E. Liu, F. He, C. Shi and N. Zhao, ACS Appl. Mater. Inter., 2016, 8, 11720-11728. M. Li, C. Liu, H. Cao, H. Zhao, Y. Zhang and Z. Fan, J. Mater. Chem. A, 2014, 2, 14844-14851. J. Deng, T. Xiong, F. Xu, M. Li, C. Han, Y. Gong, H. Wang and Y. Wang, Green Chem., 2015, 17, 4053-4060. F. Zhang, T. Liu, G. Hou, T. Kou, L. Yue, R. Guan and Y. Li, Nano Res., 2016, 9, 2875-2888. Z. J. Zhang, D. H. Xie, P. Cui and X. Y. Chen, RSC Adv., 2014, 4, 6664-6671. J.-Y. Liang, C.-C. Wang and S.-Y. Lu, J. Mater. Chem. A, 2015, 3, 24453-24462. J. Liang, S. Chen, M. Xie, Y. Wang, X. Guo, X. Guo and W. Ding, J. Mater. Chem. A, 2014, 2, 16884-16891. Z. Zuo, T. Y. Kim, I. Kholmanov, H. Li, H. Chou and Y. Li, Small, 2015, 11, 4922-4930. Y. Zhu, H. Cui, X. Meng, J. Zheng, P. Yang, L. Li, Z. Wang, S. Jia and Z. Zhu, ACS Appl. Mater. Inter., 2016, 8, 6481-6487. L. Zhang, Y. Jiang, L. Wang, C. Zhang and S. Liu, Electrochim. Acta, 2016, 196, 189-196. F. Gao, J. Qu, C. Geng, G. Shao and M. Wu, J. Mater. Chem. A, 2016, 4, 7445-7452. W. Lin, B. Xu and L. Liu, New J. Chem., 2014, 38, 5509-5514. C. Wang, M. J. O’Connell and C. K. Chan, ACS Appl. Mater. Inter., 2015, 7, 8952-8960. H. Luo, Y. Yang, B. Mu, Y. Chen, J. Zhang and X. Zhao, Carbon, 2016, 100, 214-222. X. Yu, J. Zhao, R. Lv, Q. Liang, Y. Bai, Z.-H. Huang, W. Shen and F. Kang, RSC Adv., 2015, 5, 75043-75410. A. B. Fuertes and M. Sevilla, ACS Appl. Mater. Inter., 2015, 7, 4344-4353. D. Puthusseri, V. Aravindan, S. Madhavi and S. Ogale, Energy Environ. Sci., 2014, 7, 728-735. D. Liu, Z. Jia and D. Wang, Carbon, 2016, 100, 664-677. J.-K. Sun and Q. Xu, Chem. Commun., 2014, 50, 13502-13505. Y. Zhu and Y. Tao, RSC Adv., 2016, 6, 28451-28457. A. J. Amali, J.-K. Sun and Q. Xu, Chem. Commun., 2014, 50, 1519-1522. L. Wang, Y. Zheng, Q. Zhang, L. Zuo, S. Chen, S. Chen, H. Hou and Y. Song, RSC Adv., 2014, 4, 51072-51079. P. Zhang, F. Sun, Z. Shen and D. Cao, J. Mater. Chem. A, 2014, 2, 12873-12880. F. Ma, D. Ma, G. Wu, W. Geng, J. Shao, S. Song, J. Wan and J. Qiu, Chem. Commun., 2016, 52, 6673-6676.


J. Name., 2013, 00, 1-3 | 27


Please do not adjust margins REVIEW 63 64 65 66

67 68 69 70 71 72 73

74 75 76 77 78 79 80 81 82

83 84 85 86 87 88 89

S. Perananthan, J. S. Bonso and J. P. Ferraris, Carbon, 2016, 106, 20-27. J. Xu, Z. Tan, W. Zeng, G. Chen, S. Wu, Y. Zhao, K. Ni, Z. Tao, M. Ikram, H. Ji and Y. Zhu, Adv. Mater., 2016, 28, 5222-5228. K. Liu, Y.-M. Chen, G. M. Policastro, M. L. Becker and Y. Zhu, ACS Nano, 2015, 9, 6041-6049. J.-K. Ewert, D. Weingarth, C. Denner, M. Friedrich, M. Zeiger, A. Schreiber, N. Jäckel, V. Presser and R. Kempe, J. Mater. Chem. A, 2015, 3, 18906-18912. P. Ramakrishnan and S. Shanmugam, ACS Sustainable Chem. Eng., 2016, 4, 2439-2448. Y. Li, Q. Liu, D. Kang, J. Gu, W. Zhang and D. Zhang, J. Mater. Chem. A, 2015, 3, 21016-21022. Y. Hu, X. Tong, H. Zhuo, L. Zhong, X. Peng, S. Wang and R. Sun, RSC Adv., 2016, 6, 15788-15795. B. You, F. Kang, P. Yin and Q. Zhang, Carbon, 2016, 103, 9-15. F. Zhang, T. Liu, M. Li, M. Yu, Y. Luo, Y. Tong and Y. Li, Nano Lett., 2017, DOI: 10.1021/acs.nanolett.1027b00533. Y. Xu, K. Sheng, C. Li and G. Shi, ACS Nano, 2010, 4, 4324-4330. Y. Xu, C.-Y. Chen, Z. Zhao, Z. Lin, C. Lee, X. Xu, C. Wang, Y. Huang, M. I. Shakir and X. Duan, Nano Lett., 2015, 15, 46054610. Y. Xu, Z. Lin, X. Zhong, X. Huang, N. O. Weiss, Y. Huang and X. Duan, Nat. Commun., 2014, 5, 4554. S. Yun, S.-O. Kang, S. Park and H. S. Park, Nanoscale, 2014, 6, 5296-5302. X. Yang, X. Zhuang, Y. Huang, J. Jiang, H. Tian, D. Wu, F. Zhang, Y. Mai and X. Feng, Polym. Chem., 2015, 6, 1088-1095. J.-H. Chang, Y.-H. Hung, X.-F. Luo, C.-H. Huang, S. Jung, J.-K. Chang, J. Kong and C.-Y. Su, RSC Adv., 2016, 6, 8384-8394. Z.-Y. Sui, Y.-N. Meng, P.-W. Xiao, Z.-Q. Zhao, Z.-X. Wei and B.-H. Han, ACS Appl. Mater. Inter., 2015, 7, 1431-1438. Q. Zheng, Z. Cai, Z. Ma and S. Gong, ACS Appl. Mater. Inter., 2015, 7, 3263-3271. D. Gu and F. Schüth, Chem. Soc. Rev., 2014, 43, 313-344. C. Liu, J. Wang, J. Li, M. Zeng, R. Luo, J. Shen, X. Sun, W. Han and L. Wang, ACS Appl. Mater. Inter., 2016, 8, 7194-7204. D. T. Pham, T. H. Lee, D. H. Luong, F. Yao, A. Ghosh, V. T. Le, T. H. Kim, B. Li, J. Chang and Y. H. Lee, ACS Nano, 2015, 9, 20182027. D. Liu, G. Cheng, H. Zhao, C. Zeng, D. Qu, L. Xiao, H. Tang, Z. Deng, Y. Li and B.-L. Su, Nano Energy, 2016, 22, 255-268. K. Wang, J. Zhang, W. Xia, R. Zou, J. Guo, Z. Gao, W. Yan, S. Guo and Q. Xu, J. Mater. Chem. A, 2015, 3, 18867-18873. J. Wang, L. Shen, P. Nie, X. Yun, Y. Xu, H. Dou and X. Zhang, J. Mater. Chem. A, 2015, 3, 2853-2860. Z. Wang, L. Sun, F. Xu, X. Peng, Y. Zou, H. Chu, L. Ouyang and M. Zhud, RSC Adv., 2016, 6, 1422-1427. J. Zhu, J. Yang, R. Miao, Z. Yao, X. Zhuang and X. Feng, J. Mater. Chem. A, 2016, 4, 2286-2292. D. Liu, C. Zeng, D. Qu, H. Tang, Y. Li and B.-L. Su, J. Power Sources, 2016, 321, 143-154. B. You, J. Jiang and S. Fan, ACS Appl. Mater. Inter., 2014, 6, 15302-15308.

90 91 92

93 94 95 96 97 98 99 100 101 102 103 104

105 106 107 108 109 110 111 112 113 114 115

L. Wan, J. Wang, L. Xie, Y. Sun and K. Li, ACS Appl. Mater. Inter., 2014, 15583-15596. J. Zhang, G. Chen, Q. Zhang, F. Kang and B. You, ACS Appl. Mater. Inter., 2015, 7, 12760-12766. J. Wang, J. Tang, B. Ding, V. Malgras, Z. Chang, X. Hao, Y. Wang, H. Dou, X. Zhang and Y. Yamauchi, Nat. Commun., 2017, 8, 15717. J. Chen, J. Xu, S. Zhou, N. Zhao and C.-P. Wong, Nano Energy, 2016, 25, 193-202. S. Wang, R. Liu, C. Han, J. Wang, M. Li, J. Yao, H. Li and Y. Wang, Nanoscale, 2014, 6, 13510-13517. F. Ran, X. Zhang, Y. Liu, K. Shen, X. Niu, Y. Tan, L. Kong, L. Kang, C. Xuc and S. Chen, RSC Adv., 2015, 5, 87077-87083. C. Zhan, Q. Xu, X. Yu, Q. Liang, Y. Bai, Z.-H. Huang and F. Kang, RSC Adv., 2016, 6, 41473-41476. G. Wang, B. Qian, Y. Wang, Q. Dong, F. Zhan and J. Qiu, New J. Chem., 2016, 40, 3786-3792. P. Ramakrishnan, S.-G. Park and S. Shanmugam, J. Mater. Chem. A, 2015, 3, 16242-16250. J. Zeng, Q. Cao, B. Jing and X. Peng, RSC Adv., 2016, 6, 1532015326. Y. Liu, Z. Shi, Y. Gao, W. An, Z. Cao and J. Liu, ACS Appl. Mater. Inter., 2016, 8, 28283-28290. P. Cheng, T. Li, H. Yu, L. Zhi, Z. Liu and Z. Lei, J. Phys. Chem. C, 2016, 120, 2079-2086. H. Luo, Z. Liu, L. Chao, X. Wu, X. Lei, Z. Chang and X. Sun, J. Mater. Chem. A, 2015, 3, 3667-3675. J. Huang, J. Wang, C. Wang, H. Zhang, C. Lu and J. Wang, Chem. Mater., 2015, 27, 2107-2113. X. Zhang, K. S. Ziemer, K. Zhang, D. Ramirez, L. Li, S. Wang, L. J. Hope-Weeks and B. L. Weeks, ACS Appl. Mater. Inter., 2015, 7, 1057-1064. D. Liu, S. Yu, Y. Shen, H. Chen, Z. Shen, S. Zhao, S. Fu, Y. Yu and B. Bao, Ind. Eng. Chem. Res., 2015, 54, 12570-12579. E. Y. L. Teo, L. Muniandy, E.-P. Ng, F. Adam, A. R. Mohamed, R. Jose and K. F. Chong, Electrochim. Acta, 2016, 192, 110-119. S. Song, F. Ma, F. Ma, D. Ma, W. Geng and J. Wan, J. Mater. Chem. A, 2015, 3, 18154-18162. Y. Luan, L. Wang, S. Guo, B. Jiang, D. Zhao, H. Yan, C. Tian and H. Fu, RSC Adv., 2015, 5, 42430-42437. W. Tian, Q. Gao, Y. Tan, K. Yang, L. Zhu, C. Yang and H. Zhang, J. Mater. Chem. A, 2015, 3, 5656-5664. J. Li and Q. Wu, New J. Chem., 2015, 39, 3859-3864. H. Feng, H. Hu, H. Dong, Y. Xiao, Y. Cai, B. Lei, Y. Liu and M. Zheng, J. Power Sources, 2016, 302, 164-173. P. Hao, Z. Zhao, J. Tian, H. Li, Y. Sang, G. Yu, H. Cai, H. Liu, C. P. Wong and A. Umar, Nanoscale, 2014, 6, 12120-12129. Y.-Q. Zhao, M. Lu, P.-Y. Tao, Y.-J. Zhang, X.-T. Gong, Z. Yang, G.-Q. Zhang and H.-L. Li, J. Power Sources, 2016, 307, 391-400. X. Wu, L. Jiang, C. Long and Z. Fan, Nano Energy, 2015, 13, 527-536. Y. Ren, J. Zhang, Q. Xu, Z. Chen, D. Yang, B. Wang and Z. Jiang, RSC Adv., 2014, 4, 23412-23419.

28 | J. Name., 2012, 00, 1-3




REVIEW

116 Q. Liang, L. Ye, Z.-H. Huang, Q. Xu, Y. Bai, F. Kang and Q.-H. Yang, Nanoscale, 2014, 6, 13831-13837. 117 D. Ma, G. Wu, J. Wan, F. Ma, W. Geng and S. Song, RSC Adv., 2015, 5, 107785-107792. 118 B. Chang, Y. Guo, Y. Li and B. Yang, RSC Adv., 2015, 5, 7201972027. 119 K. Wu, B. Gao, J. Su, X. Peng, X. Zhang, a. Jijiang Fu, S. Peng and P. K. Chu, RSC Adv., 2016, 6, 29996-30003. 120 L. Xie, G. Sun, F. Su, X. Guo, Q. Kong, X. Li, X. Huang, L. Wan, W. song, K. Li, C. Lv and C.-M. Chen, J. Mater. Chem. A, 2016, 4, 1637-1646. 121 W. Yu, H. Wang, S. Liu, N. Mao, X. Liu, J. Shi, W. Liu, S. Chen and X. Wang, J. Mater. Chem. A, 2016, 4, 5973-5983. 122 G. Ma, Z. Zhang, K. Sun, H. Peng, Q. Yang, F. Ran and Z. Lei, RSC Adv., 2015, 5, 107707-107715. 123 W. Sun, S. M. Lipka, C. Swartz, D. Williams and F. Yang, Carbon, 2016, 103, 181-192. 124 Y. Liu, B. Huang, X. Lin and Z. Xie, J. Mater. Chem. A, 2017, 5, 13009-13018. 125 Y. Li, Q. Zhang, J. Zhang, L. Jin, X. Zhao and T. Xu, Sci. Rep., 2015, 5, 14155. 126 Y. Zhao, W. Ran, J. He, Y. Song, C. Zhang, D.-B. Xiong, F. Gao, J. Wu and Y. Xia, ACS Appl. Mater. Inter., 2015, 7, 1132-1139. 127 L.-L. Zhang, H.-H. Li, Y.-H. Shi, C.-Y. Fan, X.-L. Wu, H.-F. Wang, H.-Z. Sun and J.-P. Zhang, ACS Appl. Mater. Inter., 2016, 8, 4233-4241. 128 H. Fan and W. Shen, ACS Sustainable Chem. Eng., 2016, 4, 1328-1337. 129 X. Zhang, Y. Jiao, L. Sun, L. Wang, A. Wu, H. Yan, M. Meng, C. Tian, B. Jiang and H. Fu, Nanoscale, 2016, 8, 2418-2427. 130 J. Hou, C. Cao, F. Idrees and X. Ma, ACS Nano, 2015, 9, 25562564. 131 J. Li, G. Zan and Q. Wu, New J. Chem., 2015, 39, 8165-8171. 132 W. Yang, Z. Du, Z. Ma, G. Wang, H. Bai and G. Shao, RSC Adv., 2016, 6, 3942-3950. 133 P. Hao, Z. Zhao, Y. Leng, J. Tian, Y. Sang, R. I. Boughton, C. P. Wong, H. Liu and B. Yang, Nano Energy, 2015, 15, 9-23. 134 C. Zhang, X. Zhu, M. Cao, M. Li, N. Li, L. Lai, J. Zhu and D. Wei, ChemSusChem, 2016, 9, 932-937. 135 J. Wang and Q. Liu, RSC Adv., 2015, 5, 4396-4403. 136 H. Feng, M. Zheng, H. Dong, Y. Xiao, H. Hu, Z. Sun, C. Long, Y. Cai, X. Zhao, H. Zhang, B. Lei and Y. Liu, J. Mater. Chem. A, 2015, 3, 15225-15234. 137 H. Peng, G. Ma, K. Sun, J. Mu, Z. Zhang and Z. Lei, ACS Appl. Mater. Inter., 2014, 6, 20795-20803. 138 Z. Li, B. Li, Z. Liu, D. Li, H. Wang and Q. Li, Electrochim. Acta, 2016, 190, 378-387. 139 H. Peng, G. Ma, K. Sun, J. Mu, Z. Zhang and Z. Lei, J. Phys. Chem. C, 2014, 118, 29507-29516. 140 H. Zhang, L. Zhang, J. Chen, H. Su, F. Liu and W. Yang, J. Power Sources, 2016, 315, 120-126. 141 P. Zhang, Z.-A. Qiao, Z. Zhang, S. Wan and S. Dai, J. Mater. Chem. A, 2014, 2, 12262-12269.

142 L. Yao, G. Yang, P. Han, Z. Tang and J. Yang, J. Power Sources, 2016, 315, 209-217. 143 L. Yao, G. Yang and P. Han, RSC Adv., 2016, 6, 43748-43754. 144 Y. Li, C. Lu, S. Zhang, F.-Y. Su, W. Shen, P. Zhou and C. Ma, J. Mater. Chem. A, 2015, 3, 14817-14825. 145 F. Miao, C. Shao, X. Li, K. Wang, N. Lu and Y. Liu, J. Mater. Chem. A, 2016, 4, 5623-5631. 146 T. K. Shruthi, M. S. Kumar, M. Arjunan, A. Pratap and N. Chandrasekaran, RSC Adv., 2015, 5, 93423-93432. 147 T. Zhu, J. Zhou, Z. Li, S. Li, W. Si and S. Zhuo, J. Mater. Chem. A, 2014, 2, 12545-12551. 148 J. Han, G. Xu, B. Ding, J. Pan, H. Dou and D. R. MacFarlane, J. Mater. Chem. A, 2014, 2, 5352-5357. 149 Y. Qiang, J. Jiang, Y. Xiong, H. Chen, J. Chen, S. Guan and J. Chen, RSC Adv., 2016, 6, 9772-9778. 150 X. J. Li, W. Xing, J. Zhou, G. Q. Wang, S. P. Zhuo, Z. F. Yan, Q. Z. Xue and S. Z. Qiao, Chem.-Eur. J., 2014, 20, 13314-13320. 151 J. Zhou, Z. Zhang, Z. Li, T. Zhu and S. Zhuo, RSC Adv., 2015, 5, 46947-46945. 152 B. Chang, a. Guo, Y. Li, H. Yin, S. Zhang, B. Yang and X. Dong, J. Mater. Chem. A, 2015, 3, 9565-9577. 153 X. Li, J. Zhou, W. Xing, F. Subhan, Y. Zhang, P. Bai, B. Xu, S. Zhuo, Q. Xue and Z. Yan, Electrochim. Acta, 2016, 190, 923931. 154 D. Zhu, Y. Wang, W. Lu, H. Zhang, Z. Song, D. Luo, L. Gan, M. Liu and D. Sun, Carbon, 2017, 111, 667-674. 155 J.-S. Wei, H. Ding, Y.-G. Wang and H.-M. Xiong, ACS Appl. Mater. Inter., 2015, 7, 5811-5819. 156 X. Han, M. R. Funk, F. Shen, Y.-C. Chen, Y. Li, C. J. Campbell, J. Dai, X. Yang, J.-W. Kim, Y. Liao, J. W. Connell, V. Barone, Z. Chen, Y. Lin and L. Hu, ACS Nano, 2014, 8, 8255-8265. 157 T. Akhter, M. M. Islam, S. N. Faisal, E. Haque, A. I. Minett, H. K. Liu, K. Konstantinov and S. X. Dou, ACS Appl. Mater. Inter., 2016, 8, 2078-2087. 158 V. Sahu, S. Shekhar, R. K. Sharma and G. Singh, ACS Appl. Mater. Inter., 2015, 7, 3110-3116. 159 F. Zhou, Q. Liu, D. Kang, J. Gu, W. Zhang and D. Zhang, J. Mater. Chem. A, 2014, 2, 3505. 160 J. Patino, N. Lopez-Salas, M. C. Gutierrez, D. Carriazo, M. L. Ferrer and F. d. Monte, J. Mater. Chem. A, 2016, 4, 1251-1263. 161 X. Wang, C. Lu, H. Peng, X. Zhang, Z. Wang and G. Wang, J. Power Sources, 2016, 324, 188-198. 162 W. Ma, S. Chen, S. Yang and M. Zhu, RSC Adv., 2016, 6, 5011250118. 163 M. H. Ervin, Nanotechnology, 2015, 26, 234003. 164 T. Liu, C. Zhu, T. Kou, M. Worsley, F. Qian, C. Condes, E. Duoss, C. Spadaccini and Y. Li, ChemNanoMat, 2016, 2, 635-641. 165 M. Yang, Y. Zhong, J. Bao, X. Zhou, J. Wei and Z. Zhou, J. Mater. Chem. A, 2015, 3, 11387-11394. 166 M. Naguib, M. Kurtoglu, V. Presser, J. Lu, J. Niu, M. Heon, L. Hultman, Y. Gogotsi and M. W. Barsoum, Adv. Mater., 2011, 23, 4248-4253. 167 B. Anasori, M. R. Lukatskaya and Y. Gogotsi, Nat. Rev. Mater., 2017, 2, 16098.


J. Name., 2013, 00, 1-3 | 29


Please do not adjust margins REVIEW 168 B. Ding, J. Wang, Y. Wang, Z. Chang, G. Pang, H. Dou and X. Zhang, Nanoscale, 2016, 8, 11136-11142. 169 P. Huang, C. Lethien, S. Pinaud, K. Brousse, R. Laloo, V. Turq, M. Respaud, A. Demorti E Re, B. Daffos, P. L. Taberna, B. Chaudret, Y. Gogotsi and P. Simon, Science, 2016, 351, 691695. 170 J. Chmiola, G. Yushin, Y. Gogotsi, C. Portet, P. Simon and P. L. Taberna, Science, 2006, 313, 1760-1763. 171 C. Merlet, C. Péan, B. Rotenberg, P. A. Madden, B. Daffos, P. L. Taberna, P. Simon and M. Salanne, Nat. Commun., 2013, 4, 2701. 172 S. Kondrat, N. Georgi, M. V. Fedorov and A. A. Kornyshev, Phys. Chem. Chem. Phys., 2011, 13, 11359-11366. 173 W.-Y. Tsai, P.-L. Taberna and P. Simon, J. Am. Chem. Soc., 2014, 136, 8722-8728. 174 K. Urita, N. Ide, K. Isobe, H. Furukawa and I. Moriguchi, ACS Nano, 2014, 8, 3614-3619. 175 C. Prehal, D. Weingarth, E. Perre, R. T. Lechner, H. Amenitsch, O. Paris and V. Presserbd, Energy Environ. Sci., 2015, 8, 17251735. 176 D. T. L. Galhena, B. C. Bayer, S. Hofmann and G. A. J. Amaratunga, ACS Nano, 2016, 10, 747-754. 177 J. L. Bañuelos, G. Feng, P. F. Fulvio, S. Li, G. Rother, S. Dai, P. T. Cummings and D. J. Wesolowski, Chem. Mater., 2014, 26, 1144-1153. 178 J. Zhi, Y. Wang, S. Deng and A. Hu, RSC Adv., 2014, 4, 4029640300. 179 T. A. Centeno, O. Sereda and F. Stoeckli, Phys. Chem. Chem. Phys., 2011, 13, 12403-12406. 180 A. Garcia-Gomez, G. Moreno-Fernandez, B. Lobatob and T. A. Centeno, Phys. Chem. Chem. Phys., 2015, 17, 15687-15690. 181 M. Salanne, B. Rotenberg, K. Naoi, K. Kaneko, P. L. Taberna, C. P. Grey, B. Dunn and P. Simon, Nat. Energy, 2016, 1, 16070. 182 X. Wu, Wei Xing, J. Florek, J. Zhou, G. Wang, S. Zhuo, Q. Xue, Z. Yan and F. Kleitz, J. Mater. Chem. A, 2014, 2, 18998-19004. 183 D.-e. Jiang, Z. Jin and J. Wu, Nano Lett., 2011, 11, 5373-5377. 184 C. Pean, C. Merlet, B. Rotenberg, P. A. Madden, P.-L. Taberna, B. Daffos, M. Salanne and P. Simon, ACS Nano, 2014, 8, 15761583. 185 C. Pean, B. Daffos, B. Rotenberg, P. Levitz, M. Haefele, P.-L. Taberna, P. Simon and M. Salanne, J. Am. Chem. Soc., 2015, 137, 12627-12632. 186 Y. He, J. Huang, B. G. Sumpter, A. A. Kornyshev and R. Qiao, J. Phys. Chem. Lett., 2015, 6, 22-30. 187 Y. He, R. Qiao, J. Vatamanu, O. Borodin, D. Bedrov, J. Huang and B. G. Sumpter, J. Phys. Chem. Lett., 2016, 7, 36-42. 188 A. C. Forse, J. M. Griffin, C. Merlet, J. Carretero-Gonzalez, A.-R. O. Raji, N. M. Trease and C. P. Grey, Nat. Energy, 2017, 2, 16216. 189 X. Wei, X. Jiang, J. Wei and S. Gao, Chem. Mater., 2016, 28, 445-458. 190 B. Dyatkin, Y. Zhang, E. Mamontov, A. I. Kolesnikov, Y. Cheng, H. M. Meyer, P. T. Cummings and Y. Gogotsi, J. Phys. Chem. C, 2016, 120, 8730-8741.

191 B. Dyatkin and Y. Gogotsi, Farad. Discuss., 2014, 172, 139-162. 192 C. Zhan, Y. Zhang, P. T. Cummings and D.-e. Jiang, Phys.Chem.Chem.Phys., 2016, 18, 4668-4674. 193 C.-M. Yang, Y.-J. Kim, J. Miyawaki, Y. A. Kim, M. Yudasaka, S. Iijima and K. Kaneko, J. Phys. Chem. C, 2015, 119, 2935-2940. 194 Y. Ma, B. Yu, Y. Guo and C. Wang, J. Solid State Electrochem., 2016, 20, 2231-2240. 195 M. Bai, T. Liu, F. Luan, Y. Li and X. Liu, J. Mater. Chem. A, 2014, 2, 10882-10888.

30 | J. Name., 2012, 00, 1-3

