A. Activated carbon

AC is a popular and most considerable research material as electrode for EC and/or battery due to its high specific surface area, high electronic conductivity, and low cost. A surface area of 2000 m²/g can be achieved by controlling distribution of pores during the activation process. Theoretically, the larger the specific surface area of an AC, the greater the C _sp is. However, in practice, the condition is more complex, in which smaller surface area of some AC gives a greater C _sp than those with a high surface area (Fig. 1).^{Reference Jin, Yan, Yu and Zhao36}

FIG. 1.

The effect of surface area and activation time to the specific capacitance of AC (reprinted from Ref. Reference Jin, Yan, Yu and Zhao36, with permission from Royal Society of Chemistry).

The C _sp can be varied over a wide range of 40–220 F/g, depending on the process applied and the type of electrolyte used.^{Reference Zhang, Li, Liu, Sun, Mérel and Désilets37,Reference Yu, Chen, Zheng, Zhao, Li and Xu38} Yu et al.^{Reference Yu, Chen, Zheng, Zhao, Li and Xu38} studied the effects of AC surface group and morphology structure on the electrochemical performance of supercapacitor. In the paper, there were two types of surface functional groups, which had battery-like and supercapacitor-like behaviors, respectively, exposed.

According to Xu et al.,^{Reference Xu, Wu, Chen, Zhang, Cao and Yang39} fabrication of AC fiber cloth electrodes gives a good cycle stability and high value of the electrochemical double layer capacitor (EDLC) capacitance using carbon dioxide activation. The C _sp of 168 F/g for the electrochemical configuration tested in aqueous solution was achieved at temperature of 900 °C. But, the higher C _sp of 208 F/g was achieved at the lower temperature treatment (∼600 °C).

B. Graphene

Graphene is generally formed from one atom sheet of graphite with the arrangement of atoms in a regular hexagonal pattern. The two-dimensional carbon material structure of graphene is considered as one of good carbon materials for the supercapacitor application due to its good electronic conductivity, superior mechanical properties, excellent chemical stability, and enhanced charging and discharging.^{Reference Yan, Liu, Fan, Wei and Zhang40,Reference Liu, Yu, Neff, Zhamu and Jang41} The graphene with a surface area of 2630 m²/g can theoretically perform to a capacitor with 550 F/g.

Yan et al. fabricated graphene as electrode materials in high performance EDLC. It was confirmed that the folding of the graphene can reduce the agglomeration and stacking, which results in improved specific surface area. A high C _sp of 349 F/g was achieved from electrode tested in 6 M KOH solution at a scan rate of 2 mV/s. Then, the C _sp of the electrode was 8% increased after 5000 cycles—an evidence of good electrochemical stability.^{Reference Yan, Liu, Fan, Wei and Zhang40}

Liu et al. investigated graphene based EC at room temperature and 80 °C with a specific energy density of 85.6 and 136 Wh/kg, respectively, at a current density of 1 A/g, which is equivalent to that of nickel metal hydride battery. The curved graphene can reduce the agglomeration and restacking, thus make full consumption of the C _sp and specific surface area.^{Reference Liu, Yu, Neff, Zhamu and Jang41}

C. Carbon black

Carbon black is usually produced by the thermal decomposition of hydrocarbon or partial combustion (basically gases, distillates, and oils) in the gas phase.^{Reference Taylor, Marsh, Heintz and Rodriguez-Reinoso42,Reference Donnet, Bansal and Wang43} The particular properties of carbon black are considered to be surface chemistry, porosity, structure (aggregate size), and fineness (primary particle size). Typically, the conductivity of carbon black is in the range of 10⁻¹ to 10² 1/Ω cm and it is controlled by the capability of electron to cross the gap between graphitic conduction by touching aggregates and by closely spaced aggregates (electron tunneling).^{Reference Donnet, Bansal and Wang43} The surface area of carbon black has generally covered a wide range, which is from 10 to 1500 m²/g.^{Reference Kinoshita44,Reference Taberna, Simon and Fauvarque45} Hence, the porosity in carbon black also differs from small surface pore to the definite hollowing out of particles. In addition, the surface area of carbon black is typically considered more accessible for ion transportation than other structures of large surface area carbon.^{Reference Taberna, Simon and Fauvarque45}

The fabrication of high surface area carbon black, containing a binder, has been investigated by Beck et al.^{Reference Beck, Dolata, Grivei and Probst19} on the EC electrodes with a C _sp of 250 F/g by using an aqueous electrolyte with the specific areas ranging from 28 to 1690 m²/g. However, Nasibi et al. applied mechanical pressing technique to fabricate carbon black electrodes. This technique homogeneously mixed carbon black with 10 wt% of polytetrafluoroethylene (PTFE) which acts as a binder in ethanol to produce a slurry and then presses on the Ni foil. The C _sp of 33.58 F/g was achieved in 2 M KCl electrolyte.^{Reference Nasibi, Rashed and Golozar46}

Another excellent performance of carbon black was reported by Gnanamuthu et al. in Li-ion battery. The fabrication starts with the mixing of Super P carbon black powder and polyvinylidene difluoride (PVDF) (as a binder) in n-methyl-2-pyrrolidone solvent to form a paste. Then, the paste has been spread onto the copper foil using the doctor blade method. The discharge capacity reached up to 1256 mAh/g in organic electrolyte.^{Reference Gnanamuthu and Lee47}

D. Carbon nanotubes

CNTs are typically produced by the catalytic decomposition of hydrocarbons. By carefully controlling various processing parameters, it is possible to engineer the nanostructures in miscellaneous conformations and control their crystalline order.^{Reference Azam, Mohamed, Shikoh and Fujiwara48} There is considerable attention on the application of CNTs as electrode materials for EC and other energy storage devices.^{Reference Hu, Wang, Wang, Zhu and Li49–Reference An, Kim, Park, Moon, Bae, Lim, Lee and Lee53} Compared to other carbon materials, CNT is very good for EC and/or battery due to their unique properties,^{Reference Zhu, Weber, Yang, Konuma, Starke, Kern and Bittner54,Reference Sakamoto and Dunn55} such as highly accessible surface area,^{Reference Guittet, Aria and Gharib56} electrical charge storage ability,^{Reference Azam, Isomura, Fujiwara and Shimoda57} low mass density,^{Reference Du and Pan58} great chemical stability,^{Reference Du, Yeh and Pan59} interconnected network structure,^{Reference Azam, Talib, Mohamad, Kasim and Rashid60} and excellent electrical conductivity.^{Reference Mohamed, Azam, Shikoh and Fujiwara61} Fabrication of single-walled CNTs (SWCNTs) and multiwalled CNTs (MWCNTs) as electrode materials has been actively investigated in both aqueous and non-aqueous electrolytes.

The C _sp of CNTs published by Frackowiak et al. was discussed to be highly dependent on their purity and morphology structure.^{Reference Frackowiak, Jurewicz, Szostak, Delpeux and Beguin20} For purified MWCNTs, the C _sp changes generally from 15 to 80 F/g with surface area ranging from ∼120 to 400 m²/g. The surface area of MWCNT is mostly mesoporous and is assorted with the outer of the tubes, with such cavity come from the entangled MWCNT. Also, C _sp can be increased up to ∼130 F/g by applying acid treatment, such as nitric and sulfuric acid. This simple process modifies the surface structure of the MWCNTs and thus introduces the additional functional group, for a storage contribution from pseudocapacitance.^{Reference Frackowiak, Delpeux, Jurewicz, Szostak, Cazorla-Amoros and Beguin62–Reference Frackowiak, Metenier, Bertagna and Beguin64}

Meanwhile, Hu et al. presented a flexible SWCNT supercapacitor; in which the electrode fabrication was done using wire wound root coating technique and then integrated into the commercial paper of single sheets.^{Reference Hu, Wu and Cui65} The C _sp and high specific power of 33 F/g and 250 kW/kg, respectively, were obtained with 1 M lithium hexafluorophosphate (LiPF₆) in ethylene carbonate and diethylene carbonate (1:1 v/v) electrolyte. By using the same CNT thin film electrode fabrication technique, Hu et al. integrated the silver nanowire and SWCNTs together, purposely to reduce the sheet resistance of the electrodes.^{Reference Hu, Choi, Yang, Jeong, La Mantia, Cui and Cui66} This technique provides a device in sulfuric acid electrolyte with a C _sp of 200 F/g, a specific power of 200 kW/kg, a specific energy of 30–47 Wh/kg, and stable cycling life over 40,000 cycles. The flexible paper fabricated with SWCNT has great potential to replace current collector in Li-ion battery, due to the advantages of high conductivity, light weight, and low cost.

A. Deposition kinetics and mechanism

The EPD technique was successfully developed in numerous applications, however the precise mechanism that could entirely explain the trend of the deposition on the substrate is not well understood.^{Reference Van der Biest and Vandeperre77} There are four principle mechanisms; the flocculation by particle deposition, the neutralization of particle charge, the electrochemical particle coagulation, and the electrical double layer distortion.^{Reference Shi and Zhitomirsky70,Reference Boccaccini and Zhitomirsky72,Reference Besra and Liu73,Reference Van der Biest and Vandeperre77,Reference Fukada, Nagarajan, Mekky, Bao, Kim and Nicholson78} Bouyer et al. suggested that EPD has a two-step process, which comprised the transfer of charged particle to the oppositely charged electrode and the particle coagulation and deposited on the electrode surface to form a mass deposit.^{Reference Bouyer and Foissy79}

Hamaker^{Reference Hamaker75} and Zhitomirsky and Petric^{Reference Zhitomirsky and Petric76} discussed that the stable suspension is an important basis for the excellent formation of the deposition by EPD technique. The formation of the deposition by EPD technique was related with the formation of the sediment by gravitation. This mechanism is called flocculation by particle accumulation. Grillon et al. suggested the neutralization of particle charge mechanism, which exhibited that the charged particles might be neutralized when they come in contact with the electrode and then became static. However, this mechanism is possible for the particle and monolayer deposition.^{Reference Grillon, Fayeulle and Jeandin80}

A. Energy storage mechanism

EC can be divided into two categories, EDLC and pseudocapacitor, due to their different charge storage mechanisms.^{Reference Conway81,Reference Conway, Birss and Wojtowicz83,Reference Qu and Shi84} The Faradaic reactivity of the surface functional group, tending to EDLC, will possibly demonstrate 1–5% of pseudocapacitance out of its total capacitance, whereas is generated by the pretreatment or preparation process of the electrode carbon materials. The pseudocapacitors typically show 5–10% of electrostatic double-layer capacitance, because of the proportionality of the electrochemically accessible interface area.

B. Electrochemical double-layer capacitor

An EDLC develops non-Faradaic capacitance evolving from the electrostatic charge stored at the double-layer, and the ion transfer access has to be confirmed on the charge and discharge processes. Effectiveness of the EDLC, no composition or chemical phase changes are required in the charging and discharging processes; therefore, EDLC possesses unlimited cyclability and is highly reversible without significant degradation. The double-layer capacitance can be determined by using the following equation^{Reference Lufrano and Staiti82}:

(3)$$C = {{A\varepsilon } \mathord{\left/ {\vphantom {{A\varepsilon } {4\pi d}}} \right. \kern-\nulldelimiterspace} {4\pi d}}\quad ,$$

where A represents the surface area of the electrode, ε represents the electrolyte dielectric constant, and d represents the thickness of the double-layer. The charge stored is transferred across the double-layer, which includes a Helmholtz layer with the thickness range of 0.5–0.6 nm, corresponding to the diameters of the solvent ions staying inside it, and an extensive region of diffusion layer with dimensions of 1–100 nm, relying on the concentration of the electrolyte.^{Reference Qu and Shi84}

Shih et al. explained that CNT possessed excellent chemical stability and good conductivity when directly grown onto graphene film as one electrode for EDLC. The formation of bundle arrays offered a higher surface area to perform with electrolyte. Thus, resulting in higher C _sp of 128.4 F/g.^{Reference Shih, Lee and Huang85} Furthermore, Fang et al. successfully fabricated MWCNT/α-MnOOH films by EPD technique. The films with a low mass loading and high areal capacitance of 0.05 mg/cm² and 0.2 F/cm², respectively, were obtained. Electrochemical measurement show that the films perform high C_sp of 327 F/g in 0.1 M Na₂SO₄ aqueous solution.^{Reference Fang, Zhang, Wu, Liu, Wen, Du and Jiang86}

C. Pseudocapacitors

In contrast to the mechanism of EDLC, pseudocapacitor ideally stores charge faradaically via the charge transfer between the electrode and electrolyte. This is realized due to the reduction–oxidation reactions, intercalation processes, and electrosorption. The pseudocapacitor may perform higher energy densities and capacitance properties than EDLC through the presence of Faradaic processes.^{Reference Dubal, Lee, Kim, Kim and Lokhande87} Metal oxides and conducting polymers generally are the two different electrode materials used to store charge in pseudocapacitor. A pseudocapacitor consists of reduction–oxidation (redox) reactions during charging process, which occur at different electrodes. Also, during discharging processes, the ion transfer is in the opposite direction through the electrolyte. The charging and discharging processes of a supercapacitor are much faster than a battery because of the faradaic energy accumulation with faster redox reactions.^{Reference Wei, Nagarajan and Zhitomirsky25}

Chen et al. reported that excellent pseudocapacitor electrode of MWCNT/manganese oxide (MnO) successfully deposited by EPD technique. The MWCNT/MnO films presented a porous structure with some rupture MnO particles incorporated and acquired a high surface area for the redox reactions, resulting the high C _sp of 260 F/g.^{Reference Chen, Chien, Chan, Lin and Wang88} Meanwhile, Ghasemi et al. investigated that the graphene was deposited onto stainless steel as a conductive substrate via EPD technique. Then, hybrid nickel–cobalt hexacyanofarrate nanoparticles were deposited on the graphene electrode using bipotential method. The films indicate the higher C_sp of 411 F/g.^{Reference Ghasemi, Ojani and Ausi89}

A. Temperature effects and thermal properties

For the minimum performance effect over the lifetime of the application, it is required to maintain the supercapacitor core temperature within the rated range of temperature of the device. The lower the temperature is retained, the better for lifetime considerations. Suggesting that, the supercapacitor must be provided with an electrically insulating shrink folder around the body. This is thus the most effective for cooling of the supercapacitor.^{Reference Hahn, Barbieri, Gallay and Kotz96}

B. Lifetime

The life of a supercapacitor is mostly influenced by a combination of operating temperature and voltage. When considering the supercapacitor life, the data sheets show the change in performance, which is the increase in resistance and decrease in capacitance. The supercapacitor's life will not end, however the performance will slightly degrade upon the excess usage of it.^{Reference Kotz, Ruch and Cericola97}

C. Frequency response

Supercapacitor has a standard time of ∼1 s. One time constant gives the time required for 63.2–36.8% full charge or discharge of a supercapacitor. The time constant of a supercapacitor is more higher than that of an electrolytic supercapacitor. Consequently, it is not possible to expose supercapacitor to a continuous moving current as overheating may result. The supercapacitor can react to short pulse power demands, however, the efficiency is reduced due to the time constant.

D. Charging

The charge/discharge of the supercapacitor that can happen at the same time due to the energy storage mechanism is not a chemical reaction. Hence, the rated current for the supercapacitor is utilized for both charge and discharge. Practically, the efficiency of charge/discharge is the same. There are many types of methods that are possible for charging of the supercapacitors, such as through constant current or power charging by using direct current or alternate current source.

E. Efficiency

Dissimilar with batteries, the supercapacitors have the same efficiency through charge/discharge. This allows the supercapacitor to be quickly recharged without current restrictive as long as the current is in the rated current for the supercapacitor. The only efficiency losses related with supercapacitor are due to the internal resistance of the device during cycling, resulting in IR drop. Mostly, the uses of the supercapacitor are 98% and exceed of the efficiency. The efficiency is reduced when using high power or current pulse, but is still greater than 90%.^{Reference Hahn, Kotz, Gallay and Siggel95,Reference Kotz, Ruch and Cericola97}