Supercapacitors, also called electric double-layer capacitors, have attracted widespread interest by virtue of their unique highpower density, long cycle life, and short charging-discharging time [1-3]. These attractive properties make them suitable for many applications where fast response at high rate is essential, such as digital devices, memory back-up systems, energy management and industrial power [4]. The commercial supercapacitor devices generally possess energy density below 10 Wh/kg, which is much lower than the lithium-ion cells (120-170 Wh/kg) [5, 6]. Therefore, an ongoing effort in the field of developing advanced supercapacitors is to further enhance the energy density without sacrificing the high-rate property [6].

Many factors in supercapacitors influence the energy density, such as electrolytes and electrode materials. It is known that the higher operation voltage will give rise to a significant improvement of energy density, because it is proportional to the square of the operation voltage. Therefore, replacing conventional aqueous electrolytes (1 V) with organic electrolytes (2.7 V) is highly desirable for high-energy-density supercapacitors. Meanwhile, electrode materials also play a key role to critically determine the energy density of supercapacitors. The current electrode materials mainly include metal oxides, carbon materials, and conducting polymers. Metal oxides and conducting polymers may provide higher energy density via forming pseudocapacitance [7, 8], but suffer from poor ionic transportation and electronic conductivity, restricting their broad application [9, 10]. Carbon materials, including carbon nanotubes, graphene and porous carbons, are considered as the promising electrode materials because of their tunable surface area, good conductivity, and high chemical and physical stability [11-16]. Among various carbon materials, activated carbons (ACs) bearing plenty of micropores are most widely used, taking account of their large surface areas associated with high capacitance and moderate cost. However, these micropores usually locate on the surface of micron/millimetrescaled carbon particles and are unconnected to each other, leading to slow ionic transfer from bulky solution to interior pores. Thus, the ion accessible pores and surface areas decrease, deteriorating the high rate charging-discharging properties. Recently, three-dimensional (3D) network structured carbons containing interconnected micro-, and meso-/macropores have been shown to exhibit enhanced electrochemical performance, particularly at high current densities when compared with ACs [9, 17-19]. Intrinsically powdery carbon aerogels (PCAs) are a kind of such well-defined 3D interconnected nanonetwork with hierarchical pores, developed most recently in our group [11]. PCAs were found to demonstrate fast ion transport capability as well as a high utilization of pore surface area in aqueous electrolyte. However, PCAs still suffer from a low energy density in aqueous electrolyte, a universal bottleneck for supercapacitors [11]. Thus, the exploration of PCAs in organic electrolytes for high energy density supercapacitors is highly desired.

Herein we report a new class of high-performance organic electrolyte supercapacitors based on our unique PCAs. Fig. 1 gives a structure model of PCA used in this study, which was obtained through utilizing well-defined poly(styrene-co-divinylbenzene) (PSDVB) nanoparticles as building blocks, followed by FriedelCrafts hypercrosslinking and carbonization [11]. The resulting PCA demonstrates several structural advantages from the viewpoint of electrode materials in organic electrolyte supercapacitors. First of all, a large number of micropores whose pore sizes are large enough for organic electrolyte ions to access are formed within the carbon nanonetwork particles, offering large interfaces for the formation of the electric double layer. Moreover, compared with AC, the 3D interconnected meso-and/or macroporous nanonetworks from interstitial sites of nanoparticles in PCA can facilitate ion transport, leading to high charging-discharging property. In addition, PCA exhibits an intrinsically micron-scale powdery form, which is favorable for both easy binding with conducting agent and polymer binder on the current collector and facile electrolyte access during electrochemical charging-discharging process. Therefore, the as-constructed PCA exhibits very attractive electrochemical properties when utilized as an electrode material for organic electrolyte supercapacitors. For example, the PCA has a capacitance as high as 152 F/g and energy density of up to 37 Wh/kg, and demonstrates an excellent cycling stability with 94% capacity retention ratio after 300 cycles. We believe our finding of using PCA for organic electrolyte supercapacitor presented here may pave the way for successfully bridging the gap between normal supercapacitors and batteries.

The precursor of PCA, powdery polymer aerogel (PPA), is in fine powdery form (Fig. 2a). As shown in the scanning electron microscopy (SEM, Fig. 2a) and transmission electron microscopy (TEM, Fig. S1 in Supporting information) images, PPA presents a 3D network morphology stacked by tremendous nanospheres of ~26 nm. Such compact and loose aggregation of these nanoparticles results in an abundant number of mesopores and macropores. Owing to the rigid hypercrosslinked structure of PPA, the 3D network morphology preserves well after carbonization in the resulting PCA (Fig. 2b), demonstrating the good nanostructure inheritability.

Nitrogen adsorption experiments were employed to evaluate the pore characteristics of PCA. As shown in the N₂ adsorptiondesorption isotherm in Fig. 3a, PCA has a steep adsorption uptake at low relative pressure (P/P₀), suggesting the formation of abundant micropores, which can be also observed in highresolution TEM image of PCA (Fig. S2 in Supporting information); and the adsorption amount goes up gradually but still does not get to a plateau near the P/P₀ of 1.0, demonstrating the presence of mesopores and macropores. The measured Brunauer-EmmettTeller surface area (S_BET) is up to 1969 m²/g, much higher than that of precursory PPA (S_BET 538 m²/g). The micropore surface area and the meso-/macropore surface area are calculated to be 1110 and 859 m²/g, respectively, by a t-plot method. The total pore volume (V_t) is measured to be as high as 1.5 cm³/g. According to the pore size distribution curve determined by density functional theory (DFT), the micropores within the network framework are centered at 1.3 nm (Fig. 3b), which is large enough for organic electrolyte ions to access; whereas the meso-and/or macropores are in the range of 10 to 100 nm, facilitating the fast ion transfer/diffusion into the interior pores. These results clearly demonstrate that the selected carbonization creates a well-developed hierarchical porous structure and high surface area in PCA.

Such a unique hierarchical porous structure is expected to endow the PCA with outstanding capacitive performance. Cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS) and galvanostatic charging-discharging measurements were conducted to evaluate the electrochemical properties of PCA by assembling a coin-type supercapacitor. The electrolyte is 1 mol/L tetraethylammonium tetrafluoroborate ((C₂H₅)₄NBF₄) in propylene carbonate (PC) solution, and the operating voltage window is between 0 and 2.7 V. To better show the superior pore structures of PCA in organic electrolyte supercapacitors, a commercially activated carbon (YP-50, 1787 m²/g) was used as reference. Typically, the shape of rectangle in CV curves can be used to reveal the rate of ion diffusion among the carbon structure. The higher the rectangle degree, the faster is the ion diffusion rate. It can be observed that PCA shows a near-rectangular shape at 10 mV/s, whereas YP-50 gives a distorted shape (Fig. 4a). With an increase of sweep rates (e.g., to 50 and 100 mV/s, Figs. 4b and c), the difference in CV curve shape becomes more significant, revealing that the ion diffusion rate within the 3D hierarchical porous network of PCA is much faster than that within the pores of YP-50, particularly during a large current charging-discharging process. The response of ionic behavior in porous structure of electrode can be further revealed by Nyquist plots in Fig. 4d. The linear curve at low frequencies reflects the ion diffusion process. PCA exhibited a much higher line slope than YP-50, demonstrating that PCA is capable of faster ion delivering and diffusion. Based on CV calculation, PCA presents a specific capacitance of up to 152 F/g at a sweep rate of 5 mV/s, and the retention ratio is still up to 61% even at 100 mV/s. Such a PCA-based electrode outperforms many typical carbon electrode materials in organic electrolytes reported previously, such as graphenes (122 F/g) [20], carbon nanotubes (ca. 80 F/g) [21], hierarchical porous carbons (115–120 F/g) [17, 22], and other carbon materials [23-25].

Furthermore, the galvanostatic charging-discharging test was also conducted. The curves show triangular shapes (Fig. 5a), revealing the ideal electrochemical capacitive characteristics. The capacitances calculated from the galvanostatic charging-discharging at current densities are summarized in Fig. 5b. It can be found that PCA possesses a specific capacitance of up to 143 F/g at a current density of 0.05 A/g, far exceeding YP-50 (75 F/g). With increasing the current densities from 0.05 A/g to 10 A/g, PCA shows a higher rate capability than YP-50, further demonstrating a faster ion transport rate in PCA. To evaluate the advantages of PCA in terms of energy and power density, Ragone plots (Fig. 6a) are provided based on the galvanostatic charging-discharging tests. The PCA supercapacitor reaches a gravimetric energy density of 37 Wh/kg at a power output of 34 W/kg. When delivering a power density up to 6750 W/kg, PCA still presents a high energy density of 15 Wh/kg. In sharp contrast, the YP-50 supercapacitor possesses a gravimetric energy density of only 19 Wh/kg at power density of 34 W/kg, which is much lower than that of the PCA, and its energy density drops to 11 Wh/kg at a power density of 6750 W/kg.

To corroborate the usefulness of using organic electrolyte for high energy density, we also conducted PCA supercapacitors in aqueous electrolyte. It is found that PCA only shows energy density of 8 Wh/kg at power density of 25 W/kg (Fig. 6a), significantly lower than the energy density in organic electrolyte. This result reveals that our PCA is very suitable in organic electrolyte for developing higher energy density. Furthermore, the PCA in organic electrolyte shows superior energy density to the other electrode materials, including those based on CNTs [26], LN-porous carbon [27] and K3-900 [25]. Besides their high energy and high-power performances, PCA supercapacitors exhibit outstanding cycling stability. The cyclic stability of the PCA supercapacitor at 5 A/g was studied by the galvanostatic charging-discharging test, as shown in Fig. 6b. After 300 cycles, the PCA still retains more than 94% of the initial capacitance, indicating an excellent long-term electrochemical cycling stability. Apparently, the electrochemical results unambiguously demonstrate that the PCA developed here is very attractive for achieving high energy density without decreasing high power density in electrochemical energy storage using organic electrolyte.

The superior performances of PCA in the organic electrolyte supercapacitor can be ascribed to its well-developed hierarchical porous structure and microsized particle size. The tremendous micropores of PCA are large enough for strongly adsorbing organic electrolyte ions, while the small-sized nanoparticulate morphology and the externally interconnected meso-/macropores together facilitate rapid mass diffusion/transport to access the micropores. Therefore, when utilized as the supercapacitor electrode, PCA can provide robust ion storage capacity and efficient electrochemically active surface to enhance the supercapacitor performances. In sharp contrast, since the isolated micropores in the conventional ACs mainly locate on the surface of large micron/millimeter-scaled carbon particles (e.g., > 10 μm), the ion transfer/diffusion pathways within ACs are long and tortuous and their pore surface utilization is low, leading to inferior supercapacitor performances, especially at high rate operations [9].

In summary, we have successfully employed a new class of PCA as promising electrode materials for organic electrolyte supercapacitors. Benefiting from their unique micro/nanostructure characteristics, 3D interconnected hierarchical porous structures and high surface areas, the resulting PCA provides fast ion transportation pathways during charging-discharging process. Therefore, the as-prepared PCA exhibits superior capacitive behaviors, including large capacitance, excellent rate capability and good cycling stability. We hope that our findings may introduce a new direction in the quest for high-performance porous materials and may pave the way for bridging the gap between traditional supercapacitors and batteries.

Acknowledgments

The authors gratefully acknowledge financial support from the National Natural Science Foundation of China (Nos. 51372280, 51422307, U1601206, 51702262), National Program for Support of Top-notch Young Professionals, Guangdong Natural Science Funds for Distinguished Young Scholar (No. S2013050014408), Tip-top Scientific and Technical Innovative Youth Talents of Guangdong Special Support Program (No. 2014TQ01C337), Fundamental Research Funds for the Central Universities (Nos.15lgjc17, 3102017OQD057), the Key Laboratory of Polymeric Composite & Functional Materials of Ministry of Education (No. PCFM201602), the Project of the Natural Science Foundation of Shaanxi Province (No. 2017JQ5003), the Program of Introducing Talents of Discipline to Universities (No. B08040), and National Key Basic Research Program of China (No. 2014CB932400).

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.cclet.2017.11.024.