High-frequency fabrication of discrete and dispersible hollow carbon spheres with hierarchical porous shells by using secondary-crosslinking pyrolysis

Abstract

A novel and facile template-free method referred to as “secondary-crosslinking pyrolysis” to fabricate discrete and dispersible hollow carbon spheres (HCSs) with hierarchical porous shells and tailorable shell thicknesses has been successfully developed by using the deformed poly(styrene-co-divinylbenzene) (P(St-co-DVB)) capsules as precursors. The samples are characterized by scanning electron microscopy, transmission electron microscopy, Fourier transform infrared spectroscopy, X-ray diffraction, Raman spectroscopy and N₂ adsorption/desorption. Our experimental results indicate that the carbon particles obtained via the designed “secondary-crosslinking pyrolysis” method present a perfect spherical shape and hollow structure, which is dependent on an interesting phenomenon where the deformed P(St-co-DVB) capsules can recover to hollow spheres through the hypercrosslinking reaction. Moreover the obtained HCSs are uniform, discrete and highly dispersible, and more importantly, they have the hierarchical pore structures that can be used as electrode materials for supercapacitors. And the test results indicate that they exhibit a high specific capacitance up to 192 F g⁻¹ at 5 mV s⁻¹, and the formation mechanism behind these phenomena is discussed.

---

Introduction

Currently, hollow carbon spheres (HCSs) have stimulated increasing interest in terms of both scientific and practical application aspects due to their remarkable properties, including large specific surface area (SSA), low density, electronic conductivity, and large void space, besides excellent chemical and thermal stabilities.^1–5 The subtle combination of these advantages endows HCSs with great potential application for catalysis,^6–10 targeted drug delivery,^11,12 adsorption,^13,14 lithium ion batteries,^15–17 fuel cells,¹⁸ and electrical double-layer capacitors (EDLCs).^19,20 In addition, the large internal volume provides a storage space or an artificial reaction “cell” that can serve many functions.^21,22 The successful applications of HCSs in different fields relies on the availability of discrete and dispersible HCSs with carefully controlled shell thicknesses.

As the electrode materials for EDLCs, the ability of ion transfer/diffusion within the pore texture of the electrode and high ion accessible surface area play crucial roles in achieving high specific capacitance.^23–25 The literature has demonstrated that hierarchical porous carbon (HPC) can just live up to the needs and even enhance electrochemical performance. For instance, the macropores serve as ion-buffering reservoirs and provide a decreased diffusion distance, the mesopores provide transport pathways with a minimized resistance, and the micropores provide the dominant adsorption sites for guest species.^26–29 Benefitting from good pore interconnectivity of micro-, meso-, and macropores, HPC exhibits the advantage of multimodal pores with a synergistic effect during the charging–discharging process and great potential for advanced electrochemical capacitor application. Therefore, it is particularly necessary to prepare HCSs with hierarchical pore structure for ideal electrode materials of supercapacitors.

Considering these aspects mentioned above, the successful application of HCSs in supercapacitors is dependent on the synthesis of discrete and dispersible HCSs with hierarchical pore structure. Until now, HCSs are synthesized through templating methods,^9,30,31 combined with chemical vapor deposition (CVD) technique³² and solvothermal process,^2,33 using various organic or polymer as carbon sources such as sucrose,^34,35 furfuryl alcohol,³⁶ ethylene gas,³² phenol resin,^30,37 and aromatic compounds.^9,38–40 Templating methods for fabrication of HCSs are based on the infiltration and carbonization of precursors in the pores, and synthesis and removal of the template materials. A simple and controllable nanocasting CVD method is used to prepare well-ordered HCSs employing mesoporous silica spheres as the hard template and ethylene as the carbon source.³² Hollow macroporous core and mesoporous shell of the carbon capsules are respectively constructed by silica core and a thin mesoporous silica shell (SCMS) structures.⁴⁰ Nevertheless, the incorporation of the carbon precursor into pore channels of silica shell through infiltration process will inevitably result in the spheres conglutination for the sintering of carbon nanostructures at high-temperature annealing due to the precursor deposition on the outer surface of hard template.⁶ Consequently, it will often result in the nondispersible and conglutinated bulky materials. A positive strategy to fabricate isolated hollow mesoporous carbon spheres is developed by providing electrostatic attraction between the negatively charged glucose-derived carbon precursor generated by hydrothermal treatment and the positively charged solid template (amino-functionalized silica particles with nonporous core–porous shell structure) to induce selective deposition of the precursor into pore system in the template, which effectively prevents conglutination among the carbon spheres.³³ Another templating method referred to as “confined nanospace pyrolysis” to synthesize discrete HCSs is inspired by the structure of an egg consisting of an inorganic outer shell (eggshell), organic inner shell (egg white), and core (yolk).³⁷ The presence of the inorganic outer shell ensures that HCSs remain discrete and retain their original shape. However, these methods have some inherent limitations. For instance, the procedure is tedious and time-consuming due to fabrication, modification, and removal of hard templates with special nanostructures, and template-removing agents (e.g. HF) would lead to serious pollution. Therefore, a simple and effective strategy to produce discrete, dispersible and uniform HCSs with hierarchical pore structure still remains a challenge.

Herein, we describe a new template-free method referred to as “secondary-crosslinking pyrolysis” for the synthesis of discrete, dispersible and uniform HCSs with hierarchical porous shell and tailorable shell thickness using the deformed poly(styrene-co-divinylbenzene) (P(St-co-DVB)) capsules as precursors. Compared with the reported approaches,^32,33,37 the “secondary-crosslinking pyrolysis” method bears multiple prominent advantages: (i) the procedure avoids using the template; (ii) the perfect spherical shape of HCSs derives from the deformed P(St-co-DVB) capsules; (iii) highly crosslinked rigid outer shells produced by the secondary hypercrosslinking reaction prevents the carbon sphere conglutination and collapse during high-temperature calcinations; (iv) the HCSs have discrete, dispersible, highly uniform, and hierarchically porous characters, which are of great importance when used as the electrode materials of supercapacitors, catalyst support, and self-assembling blocks for complex structure. The obtained discrete, dispersible and uniform HCSs with hierarchical pore structure show a high specific capacitance of up to 192 F g⁻¹ at 5 mV s⁻¹ when utilized as electrodes for supercapacitors, which is much more than those of commercialized activated carbon (114 F g⁻¹) and ordered mesoporous carbon (83–112 F g⁻¹).⁴¹

Experimental

Materials

Analytically pure tetrachloromethane (CCl₄) was purchased from Chemical Reagent Factory in Luoyang; analytically pure anhydrous aluminum trichloride (AlCl₃) was supplied by Tianjin Damao Chemical Reagent Factory; analytically pure 1,2-dichloroethane (Cl(CH₂)₂Cl) was supplied by Tianjin Damao Chemical Reagent Factory; chemically pure anhydrous ferric trichloride (FeCl₃) was supplied by Sinopharm Chemical Reagent Co., Ltd; dilute hydrochloric acid (10 wt%) was prepared from concentrated hydrochloric acid (37 wt%), which was offered by Luoyang Haohua Chemical Reagent Company, Ltd; deionized water, analytically pure acetone and ethanol were used as received without any further treatment.

Synthesis of hollow P(St-co-DVB) capsules with mesoporous shell

The monodisperse core@shell P(St-co-DVB) spheres (CSPSs) were synthesized according to our previous work.⁴² The obtained CSPSs were respectively labeled as CSPS-20, CSPS-30 and CSPS-40 when the DVB dosage was 20 wt%, 30 wt%, and 40 wt%. 0.3 g of dried CSPS was put into CCl₄ (30 mL) in a 100 mL three-necked flask in a thermostated water bath of 40 °C. After the mixture was vigorously stirred at a speed of 160 rpm for 10 h, the products were collected and purified by centrifugation and redispersion in acetone by ultrasonic. The hollow P(St-co-DVB) capsules (HPCs) obtained from CSPS-20, CSPS-30 and CSPS-40 were labeled as HPC-20, HPC-30 and HPC-40, respectively. Average molecular weight and its polydispersity of the solution (Table S1†) indicate that the core has been dissolved and removed. Moreover, the mesopores are formed in the shells of HPC (Fig. S1†).

Synthesis of hypercrosslinked hollow P(St-co-DVB) spheres

The hypercrosslinking reaction was conducted in a three-necked flask (100 mL) equipped with a polytetrafluoroethylene-bladed paddle stirrer and a water-cooled reflux condenser. In a typical reaction, 30 mL of CCl₄ and 1.80 g of anhydrous aluminum trichloride (AlCl₃) were put into the flask under vigorous stirring of 160 rpm at room temperature. 20 min later, 0.30 g of HPCs was added into the flask. 40 min later, the flask was put into a water bath of 40 °C, and this moment was regarded as the start time of the reaction. After the reaction continued for 16 h at 40 °C, the products were collected and purified by centrifugation and redispersion in acetone by ultrasonic, and then were washed with dilute hydrochloric acid as well as deionized water three times, respectively. The final products were collected and dried at 45 °C overnight. The hypercrosslinked hollow P(St-co-DVB) spheres (HHPSs) obtained from HPC-20, HPC-30 and HPC-40 were marked as HHPS-T-20, HHPS-T-30 and HHPS-T-40, respectively, among which T represents tetrachloromethane.

The process of the secondary hypercrosslinking reaction was the same as that of the above hypercrosslinking reaction except that Cl(CH₂)₂Cl was employed as the crosslinker and solvent, and FeCl₃ was served as the catalyst. The HHPSs products obtained from HHPS-T-20, HHPS-T-30 and HHPS-T-40 via the secondary hypercrosslinking reaction were respectively labeled as HHPS-T-D-20, HHPS-T-D-30 and HHPS-T-D-40, among which D represents dichloroethane.

Carbonization of HHPS-T-D

Carbonization of HHPS-T-D was carried out in a tube furnace under nitrogen atmosphere. Firstly, the temperature was elevated homogeneously to 700 °C with a heating rate of 100 °C h⁻¹ and then was held at 700 °C for 2 h under a nitrogen flow. After the furnace was naturally cooled down to room temperature, the carbonized products were collected. The hollow carbon spheres (HCSs) obtained from HHPS-T-D-20, HHPS-T-D-30 and HHPS-T-D-40 were, respectively, denoted as HCS-20, HCS-30 and HCS-40.

Characterization

The morphologies and structures of the samples were investigated with a field-emission scanning electron microscope (FE-SEM) (Quanta 200) and a transmission electron microscope (TEM) (FEI Tecnai G2 20) with an accelerating voltage of 200 kV. Fourier transform infrared (FT-IR) spectra were measured in the wavenumber range from 4000 to 400 cm⁻¹ by using a Bruker Optics TENSOR 27 FT-IR spectrophotometer. The solid-state NMR ¹³C {¹H} NMR measurements were carried out using AVANCE AV 400. X-ray diffraction (XRD) patterns were measured on a Y-2000 X-ray diffractometer with Cu Kα radiation (λ = 0.15406 nm) operating at 40 kV and 40 mA. Raman spectra were performed at 532 nm laser excitation on a Renishaw microscope system RM2000 at room temperature. The zeta-potential measurement was conducted on a Malvern ZEN 3690 Zeta PALS at various pH values and a conductance of 200 mS. The thermal stability of the samples were monitored using a thermogravimetric analysis (TGA-50H) with a heating rate of 10 °C min⁻¹ in N₂ flow. Nitrogen adsorption/desorption measurements were carried out in a Micromeritics ASAP 2020 instrument at 77 K. Before the sorption tests, the samples were degassed at 300 °C under vacuum for 8 h. The pore size distribution (PSD) was calculated by density function theory (DFT) and Horvath–Kawazoe (HK) methods. The total pore volume was calculated at a relative pressure of P/P₀ = 0.99.

The capacitance performance of the samples was evaluated in 2 M KOH aqueous solution by a CHI 660D electrochemical workstation (Model 760C) with a standard three-electrode system at room temperature. As-made test electrodes in the form of round sheet were obtained by pressing a mixture film of 80 wt% HCSs, 10 wt% carbon black and 10 wt% poly(tetrafluorethylene) (used as a binder, PTFE 60 wt% dispersion in H₂O, Sigma-Aldrich) into a stainless steel wire current collector (1 × 1 cm²) and dried at 80 °C for 24 h. In addition, platinum foil was used as the counter electrode and Ag/AgCl electrode was used as the reference electrode. The cyclic voltammetry (CV) curves were obtained at various scan rates with voltage ranging from −0.5 V to 0 V. The chronopotentiometry (CP) curves were obtained at various current densities with the same voltage range as CVs. The electrochemical impedance spectroscopy (EIS) was measured in the frequency range of 10 mHz to 10 kHz at the open circuit voltage with an alternate current amplitude of 5 mV.

Results and discussion

The preparation procedure of the discrete, dispersible and uniform HCSs is briefly illustrated in Fig. 1. Firstly, monodisperse deflated soft HPCs with mesoporous shell (Fig. S1†) as the precursors were initially synthesized according to ref. 42. Secondly, the deformed HPCs recover to perfect hollow spheres through the primary hypercrosslinking reaction using AlCl₃ as catalyst, and CCl₄ as solvent and crosslinker, respectively. This interesting experimental phenomenon is a prerequisite for synthesis of perfect HCSs using the deformed HPCs as precursors. In the meantime, hierarchical porous texture composed of micro-, meso-, and macropores are formed in the HHPS-T. Micropores are introduced by the hypercrosslinking reaction,⁴² mesopores derive from mesoporous shell of HPCs (Fig. S1†), and macropores are formed by the stacking of nanospheres. Thirdly, highly crosslinked rigid outer shells are produced by the secondary hypercrosslinking reaction using FeCl₃ as catalyst, and Cl(CH₂)₂Cl as solvent and crosslinker, respectively. It is critical to avoid conglutination among the spheres during high-temperature pyrolysis and obtain discrete and uniform HCSs with perfect spherical shape. Besides, the HHPS-T-D inherits hierarchical porous texture of the HHPS-T. Finally, discrete, dispersible and uniform HCSs with hierarchical pore structure are obtained after carbonizing HHPS-T-D.

Fig. 1 Schematic illustration of the preparation procedure for discrete and dispersible HCSs. (HC is short for hypercrosslinking).

The effect of crosslinking density on the carbon yield of precursor is examined by TGA curves of HPC-30, HHPS-T-30, HHPS-T-D-30 (Fig. 2). At 700 °C, low crosslinked HPC-30 has a yield of 10 wt% and then the yield of HHPS-T-30 greatly increases to 28 wt%, indicating the importance of hypercrosslinking reaction for preparation of HCSs. However, the HCSs obtained from HHPS-T-30 demonstrate serious particle conglutination and deformation (Fig. S2†). After the secondary hypercrosslinking reaction, the thermal stability of HHPS-T-D-30 during carbonization is immensely improved with a yield of up to 38 wt% which is much more than that of HHPS-T-30, demonstrating that high crosslinking density favors good stability of skeleton structure of precursors. Combining with Fig. 3d and f, the secondary hypercrosslinking reaction can prevent HCSs conglutination and deformation.

Fig. 2 TGA curves of HPC-30, HHPS-T-30, HHPS-T-D-30.

Fig. 3 SEM photos of (a) HPC-30, (b) HHPS-T-30, (c) HHPS-T-D-30 and (d) HCS-30. TEM photos of (e) HPC-30, (f) HHPS-T-30, (g) HHPS-T-D-30 and (h) HCS-30.

Fig. 3 shows the SEM and TEM photos of the samples with 30 wt% DVB (HPC-30, HHPS-T-30, HHPS-T-D-30 and HCS-30). Fig. 3a and e show that the HPC-30 (about 340 nm) presents polyhedral sunken structure due to negative pressure produced by the evaporation of solvent and the soft nature of low crosslinked PS shells.⁴³ Moreover, large amount of mesopores exist in the shell of HPC-30 (Fig. 3e and S1†). After the primary hypercrosslinking reaction, the deformed HPCs recover to perfect hollow spheres (HHPS-T-30) with porous structure (Fig. 3b and f). This should be contributed to the swelling and softening effect of solvent on the shell of the deformed HPCs that recover to spherical shape according to the lowest energy principle. And their spherical shape is further fixed by the crosslinked network structure formed during the reaction. Compared with the HHPS-T-30, it is obvious that the size, shape, and porous structure of the HHPS-T-D-30 prepared by the secondary hypercrosslinking reaction, do not change (Fig. 3c and g). After carbonizing the HHPS-T-D-30, the discrete, uniform HCSs (HCS-30, about 260 nm) are successfully obtained. And their shell thickness is about 30 nm (Fig. 3d and h). Besides, plenty of small pores and worm-like channels are homogenously dispersed in the carbon matrix (Fig. 3h). Compared with the seriously conglutinated and deformed HCSs obtained from the HHPS-T-30 without the secondary crosslinking reaction (Fig. S2†), it is suggested that the secondary crosslinking reaction is crucial for the synthesis of discrete HCSs with perfect spherical shape.

The SEM and TEM photos of the samples with 20 wt% DVB and 40 wt% DVB are displayed in Fig. 4 and 5, respectively. Fig. 4a, e and f show that the HPC-20 with porous shell (about 310 nm) seriously deforms due to thin soft shells with low strength induced by low DVB dosage. Fig. 5a, e, and f indicate that a slight extent of sunken structure exists on the HPC-40 with porous shell (about 370 nm), which is different from HPC-20 and HPC-30, and derives from the thick shell with high mechanical strength induced by high DVB dosage that can resist the negative pressure produced by the evaporation of solvent. As the samples with 30 wt% DVB, the deformed HPC-20 and HPC-40 both successfully recover to perfect hollow spheres (HHPS-T-20 and HHPS-T-40) after the primary hypercrosslinking reaction (Fig. 4b and 5b). The secondary hypercrosslinking reaction does not change the size and shape of hollow polymer spheres (Fig. 4c and 5c). Finally, discrete, uniform HCS-20 (about 240 nm) and HCS-40 (about 280 nm) are successfully obtained after carbonization HHPS-T-D (Fig. 4d and 5d). And the shell thickness of HCS-20 and HCS-40 are about 20 nm and 50 nm (Fig. 4h and 5h), respectively. Combining with the above results in Fig. 3, it is concluded that discrete, uniform HCSs with porous shell are successfully synthesized by “secondary-crosslinking pyrolysis” using deformed polymer capsules, providing a new route to synthesize discrete hollow carbon particles. The diameter and shell thickness of HCSs can be controlled by changing DVB dosage of HPCs, and increase with increase of the DVB dosage.

Fig. 4 SEM photos of the (a) HPC-20, (b) HHPS-T-20, (c) HHPS-T-D-20, and (d) HCS-20. TEM photos of (e and f) HPC-20 and (g and h) HCS-20.

Fig. 5 SEM photos of the (a) HPC-40, (b) HHPS-T-40, (c) HHPS-T-D-40, and (d) HCS-40. TEM photos of (e and f) HPC-40 and (g and h) HCS-40.

Dispersible and discrete HCSs are of critical importance for the fundamental study of carbon colloids and practical applications.³⁷ Thus, the zeta potential of the HCSs was measured in aqueous solution as a function of pH value. As can be seen in Fig. 6a, the zeta potentials are all below −30 mV with the total pH ranging from 2 to 13, indicating that HCSs can be stably dispersed in aqueous solution. In addition, it turns out that the surface of HCS-30 is negatively charged. As visually shown in Fig. 6b, the suspension of HCS-30 in water is very stable, with no sign of aggregated precipitation for over 3 days. Such a stable dispersion result mainly derives from electrostatic repulsion without any surfactant or polymer for facilitating the stabilization of the HCSs. In comparison, the HCSs obtained by carbonizing the HHPS-T-30 without the secondary hypercrosslinking reaction fail to stably disperse in water, as evidenced by the visible sediment at the bottom of the sample vial in a short time. The zeta potentials of these HCSs without the secondary hypercrosslinking reaction are not measured because they form sediments in water at different pH values. This behavior is a strong indication that the “secondary-crosslinking pyrolysis” technique ensures the formation of discrete and dispersible HCSs.

Fig. 6 (a) Zeta potentials of the HCS-30 in aqueous solution at different pH values. (b) Digital photos of the dispersed solution of HCS-30 (left) and HCSs by carbonizing the HHPS-T-30 without the secondary hypercrosslinking reaction (right).

To clarify the effect of the crosslinked structure of HHPSs on the morphology of HCSs, the obtained HPC-30, HHPS-T-30 and HHPS-T-D-30 samples were characterized by FT-IR spectroscopy (Fig. 7). It can be seen that the broad characteristic peak at 3500–3400 cm⁻¹ is attributed to the O–H stretching vibration of H₂O, which derives from KBr during the FT-IR measurement. In terms of HPCs (Fig. 7c), the strong peaks between 3100 and 2800 cm⁻¹ are ascribed to the stretching vibrations of C–H and –CH₂– of the PS.^37,44 Similarly, the peaks between 1600 and 1400 cm⁻¹ are assigned to C–H bending vibrations, and the peaks between 900 and 700 cm⁻¹ are mainly related to substituted phenyl rings.^44,45 After the hypercrosslinking reaction using CCl₄ as crosslinking agent, the relative intensity of the peaks at 3028, 2857, and 1492 cm⁻¹ dramatically weaken, and the peaks at 1282 and 1182 cm⁻¹ appear owing to C–Cl bending vibration, which demonstrate the occurrence of Friedel–Crafts alkylation (Fig. 7a and b).^46,47 More importantly, the obvious peak at 1650 cm⁻¹ emerges (Fig. 7b), which arises from the stretching vibrations of the C=O band resulting from the constructed carbonyl (–CO–) crosslinking bridges produced by hydrolysis of –CCl₂–.⁴⁸ This is further verified by the occurrence of the signal at around 195 ppm assigned to carbonyl ¹³C {¹H} NMR spectra of HHPS-T-30 (Fig. 8).⁴⁹ The constructed –CO– crosslinking bridges provide a high crosslinking density, in favor of achieving reinforced framework structure. After the secondary hypercrosslinking reaction using Cl(CH₂)₂Cl as crosslinking agent, it can be clearly seen from Fig. 7I–III (the magnified regions of FT-IR spectra) that the relative intensity of the peaks at 2930, 1450, 795, 760, and 700 cm⁻¹ ascribed to C–H vibrations weakens. Moreover, the relative intensity of the peak at 1282 cm⁻¹ attributed to the C–Cl bending vibration strengthens. It is suggested that Friedel–Crafts alkylation occurs and the –CH₂CH₂– crosslinking bridges are constructed. Solid-state ¹³C {¹H} NMR give further evidence of this reaction (Fig. 8). Compared with HHPS-T-30, the relative intensity of the signals at 45 ppm and 144 ppm respectively attributed to methylene and substituted phenyl rings for HHPS-T-D-30 increases,^49,50 also indicating that the secondary hypercrosslinking reaction happens.

Fig. 7 FT-IR spectra of the (a) HHPS-T-D-30, (b) HHPS-T-30, and (c) HPC-30. I, II, and III show the magnified view of the pink rectangle, respectively.

Fig. 8 ¹³C {¹H} NMR spectra of HHPS-T-30 (red line) and HHPS-T-D-30 (black line).

Considering that the HHPS-T without the secondary hypercrosslinking results in the nondispersible and conglutinated bulky materials at high temperature for the sintering of carbon capsules due to the polymer chain movement in the HHPS-T with relatively low crosslinking density (Fig. 6 and S2†), and on the contrary, the HHPS-T-D with the secondary hypercrosslinking give rise to dispersible and discrete HCSs (Fig. 3–6), highly crosslinked rigid outer shells produced by the secondary crosslinking is vital to fabricate discrete and dispersible HCSs.

The porosity of the obtained samples was measured by nitrogen adsorption–desorption at −196 °C. As shown in Fig. 9a, the nitrogen adsorption–desorption isotherms belong to a characteristic hysteretic-type IV isotherm, indicating the presence of micropores and mesopores. The steep increase in nitrogen uptake at a relative pressure (P/P₀) of below 0.02 is attributed to the filling of micropores, which indicates the presence of a lot of micropores. The uptake from P/P₀ of 0.02–0.2 may be ascribed to the presence of ultramicropores and small mesopores.⁵¹ An obvious hysteresis loop appears during desorption across a large relative pressure range, demonstrating the presence of mesopores with a wide pore size distribution (PSD), which are further proved by the t-plots in Fig. S3.† The clear increase in adsorption quantity at P/P₀ > 0.95 is most probably due to the interstitial spaces among the spheres.⁴² It is obvious that the obtained HCS-20, HCS-30, and HCS-40 all have hierarchical porous structure resulting from that of HHPS-T-D. Besides, hierarchical porous structure is formed in the interior of HHPS-T-30 after the primary hypercrosslinking reaction using CCl₄ as crosslinking agent. The nitrogen adsorption–desorption isotherm of HHPS-T-D-30 is similar to that of HHPS-T-30, which indicates that the secondary hypercrosslinking reaction using ClCH₂CH₂Cl as crosslinking agent hardly changes pore structure of HHPS-T.

Fig. 9 (a) Nitrogen adsorption–desorption isotherms of the samples. (b) PSD of the obtained HCSs with different DVB dosage derived from the DFT method. (c) PSD of the typical samples with 30 wt% DVB derived from the DFT method. (d) PSD of the samples derived from the HK method.

PSD curves derived from the DFT method are shown in Fig. 9b and c. Since nitrogen at its boiling point might be less sensitive to ultramicropores in DFT calculations,⁵² micropore size distribution curves derived from the HK method are shown in Fig. 9d. The detailed regions of micro-, meso-, and macropore size that could be observed in the PSD curves (Fig. 9b and c) are listed in Table 1. As for HCS samples, two regions of micropores with size of 0.6–0.8 nm and 1.0–2.0 nm (or 1.0–1.8 nm) could be observed. The HCS samples also have similar regions of mesopores and macropores except that the HCS-30 is short of the region of mesopores with size of 2.4–7.3 nm. It is suggested that the obtained HCS samples all have hierarchical pore structure and similar pore texture (Fig. 9b and d). In addition, the regions of micro-, meso-, and macropore size are observed for the HHPS-T-30 and are the same to those of HHPS-T-D-30, which indicate that hierarchical pore structure is formed after the primary hypercrosslinking reaction using HPCs with mesoporous shell as precursors, and the secondary hypercrosslinking reaction does not change pore structure of HHPS-T. These results are in good agreement with the nitrogen adsorption–desorption isotherms.

Table 1 PSD regions obtained by the DFT method

Samples	PSD (nm)
HCS-20	0.6–0.8, 1.0–7.3, 8.0–8.7, 9.4–10.5, 12.4–70.0
HCS-30	0.6–0.8, 1.0–2.0, 7.7–8.7, 9.4–10.5, 12.4–62.0
HCS-40	0.6–0.8, 1.0–1.8, 2.4–7.3, 8.0–8.7, 9.4–10.5, 12.4–62.0
HHPS-T-30	0.6–0.8, 1.0–2.0, 2.4–7.3, 8.0–8.7, 9.4–10.5, 12.4–62.0
HHPS-T-D-30	0.6–0.8, 1.0–2.0, 2.4–7.3, 8.0–8.7, 9.4–10.5, 12.4–62.0

---

The textural parameters of the HCS and HHPS samples are summarized in Table 2. As for the HCS samples, as the DVB dosage increases from 20 wt% to 40 wt%, both the specific surface area (SSA) calculated by using the BET model in the relative pressure (P/P₀) range from 0.01 to 0.12 (Fig. S4† and 5), S_micro and V_micro initially increase and then decrease. Moreover, the V_t increases from 0.25 m³ g⁻¹ to 0.39 m³ g⁻¹, and the V_meso and V_macro increase with increase of the DVB dosage. It is obvious that the HCS-30 indicates the biggest S_micro and V_micro. It is noticeable that the SSA (530 m² g⁻¹) of the HHPS-T-30 is about 40 times as much as that (13 m² g⁻¹) of the HPC-30 with mesoporous structure (Fig. S1†), which is due to large amount of micropores constructed by the –CO– crosslinking bridges introduced through the primary hypercrosslinking reaction. Compared with HHPS-T-30, a slight decrease of the V_t for the HHPS-T-D-30 show that the –CH₂CH₂– crosslinking bridges introduced by the secondary hypercrosslinking reaction occupy some pore space of the HHPS-T-30, but hardly change their pore structure. Compared with HHPS-T-D-30, the decrease of SSA and V_t, and disappearance of the region of mesopores with size of 2.4–7.3 nm for the HCS-30 are contributed to collapse of a portion of framework and contraction of HHPS-T-D-30 during high-temperature pyrolysis. Synthesizing the above SEM, TEM, FT-IR, zeta potentials and nitrogen adsorption–desorption results, schematic diagram of the fabrication procedure for discrete, dispersible and uniform HCSs with hierarchical porous shell and the mechanism behind the hypercrosslinking reaction is formed and shown in Fig. 10.

Table 2 Textural parameters of the HCS and HHPS samples

Samples	SBET^a (m^2 g^−1)	Smicro^b (m^2 g^−1)	Vt^c (m^3 g^−1)	Vmicro^d (cm^3 g^−1)	Vmeso^d (cm^3 g^-1)	Vmacro^d​ (cm^3 g^-1)
a Specific surface area (SSA) calculated by the Brunauer–Emmett–Teller (BET) method.b Micropore area calculated by the t-Plot method.c Total pore volume.d Pore volume obtained by the DFT method.
HCS-20	398	247	0.20	0.11	0.08	0.01
HCS-30	432	311	0.39	0.15	0.21	0.03
HCS-40	372	228	0.43	0.13	0.25	0.05
HHPS-T-30	530	216	0.44	0.15	0.24	0.05
HHPS-T-D-30	523	217	0.40	0.13	0.23	0.04

---

Fig. 10 Schematic diagram of the fabrication procedure for discrete, dispersible and uniform HCSs with hierarchical porous shell and the mechanism behind the hypercrosslinking reactions. ① Formation of –CO– crosslinking bridge for HHPS-T with CCl₄ as the crosslinker. ② Formation of –CH₂CH₂– crosslinking bridge for HHPS-T-D with Cl(CH₂)₂Cl as the crosslinker.

The XRD patterns of the HCS powder samples with different DVB dosages are shown in Fig. 11a. Two broad and weak diffraction peaks at around 23.5° and 43° indexed to (002) and (100) planes respectively indicate that these HCSs are amorphous carbon phase, exhibiting that no pronounced graphitization occurs at the carbonization temperature of 700 °C.

Fig. 11 (a) XRD patterns and (b) Raman spectra of the HCS powder samples.

The Raman spectra for the HCSs with different DVB dosages are demonstrated in Fig. 11b. The D-band around 1350 cm⁻¹ is ascribed to the vibrations of carbon atoms with dangling bonds in plane terminations of the disordered graphite and is related to the defects and disorders in structures in HCSs. The G-band around 1600 cm⁻¹ corresponds to an E_2g mode of graphite and is attributed to the vibration of sp²-hybridized carbon atoms in a two-dimensional hexagonal lattice, such as in a graphite layer. The low ratio of I_D/I_G is characteristic of a graphite lattice with perfect two-dimensional order in the graphitic layer.⁵¹ The intensity ratios of I_D/I_G corresponding to HCS-20, HCS-30, and HCS-40 are respectively calculated to be 0.62, 0.60, and 0.70.

Further the supercapacitor performance of the as-prepared HCSs and SCSs was estimated by a three-electrode system using 2.0 M KOH as the electrolyte (Fig. 12 and S6–S8†). As shown in Fig. 12a, all the CV curves of the HCS-30 at different scan rates present a quasi-rectangular shape without a very oblique angle even at a scan rate as high as 400 mV s⁻¹, which demonstrates the typical characteristics of double layer capacitance, a highly capacitive nature and a small equivalent series resistance (ESR) with rapid ion response.^42,53,54 The specific capacitance can be calculated by the equation (see ESI†). Fig. 12b shows the correlation of specific capacitance with scan rates for HCS-30, HCS-20 and HCS-40. The specific capacitance of the HCS-30 is 192 F g⁻¹ at 5 mV s⁻¹, and remains 124 F g⁻¹ at 400 mV s⁻¹, which shows a capacitance retention of 65%, implying a good rate capability. As the scan rate increases, the specific capacitance decreases. The specific capacitance decreases sharply at small scan rate below 25 mV s⁻¹, which can be attributed to the presence of micropores. A slow decrease of the specific capacitance at the scan rates from 25 to 400 mV s⁻¹ may be ascribed to the presence of mesopores.⁴² In comparison, HCS-30 manifests electrochemical performance greatly superior to HCS-20 and HCS-40, which should be benefitting from the largest S_micro, V_micro as well as relatively high V_meso (Table 2) for the micropores provide the dominant adsorption sites and the mesopores provide transport pathways with a minimized resistance. The specific capacitance (192 F g⁻¹) of HCS-30 is higher than that (170 F g⁻¹) of hierarchical porous carbon formed by aggregation and conglutination of SCSs.⁵⁵ It is suggested that discrete HCSs are much more effective than conglutinated SCSs as electrode material, which may be contributed to a decreased ion diffusion distance and more efficient microporous adsorption sites that the discrete HCSs with thin shell can provide. Moreover, the specific capacitance of HCS-30 is also superior to that (166 F g⁻¹) of porous carbon with high micropore rate (97%),⁵⁶ which further suggests the significance of meso- and macropore structure and large internal volume for carbon materials to achieve high electrochemical performance. Charge–discharge behavior of the HCS-30 electrode was tested at current densities from 0.5 to 5.0 A g⁻¹. The galvanostatic charge–discharge curves exhibit almost symmetric triangular curves at a current density of 0.2 A g⁻¹, indicating an outstanding rate performance (Fig. 12c).⁴² Furthermore, it can be seen that the galvanostatic charge–discharge curves remain symmetric triangles without distinct voltage drop (IR) related to the internal resistance even at a current density as high as 5.0 A g⁻¹, suggesting the fast transmission of ions in the hierarchical pore structure. The Nyquist plot of the HCS-30 (Fig. 12d) indicates a straight line in the low-frequency region, which is close to ideal capacitive behavior. Further small semicircular radius in the high-frequency region indicates the minimum ESR, which might facilitate the ion diffusion and electrolyte permeation into the pore channels.⁵⁵

Fig. 12 (a) Cyclic voltammetry (CV) curves of HCS-30 at the scan rates of 5, 25, 50, 100, 200 and 400 mV s⁻¹. (b) The correlation of specific capacitance with scan rates for HCS-20, HCS-30 and HCS-40. (c) Chronopotentiometry (CP) curves of HCS-30 at current densities of 0.5, 1.0, 2.0 and 5.0 A g⁻¹. (d) Nyquist plot of HCS-30 electrode material.

Similar experimental phenomena can be obtained for HCS-20 and HCS-40 (Fig. S6† and 7). These results show that discrete, dispersible HCSs with hierarchical pore structure can promote the mass transfer/diffusion of ions into the pores effectively. Therefore, it can be concluded that “secondary crosslinking pyrolysis” provides an efficient route to obtain ideal HCSs that have potential application in energy storage devices.

Conclusions

In summary, we have successfully fabricated discrete, dispersible and uniform HCSs with hierarchical pore structure by a novel and facile template-free method referred to as “secondary-crosslinking pyrolysis” using the deformed P(St-co-DVB) capsules as precursors. This strategy relies on such a fact that the deformed P(St-co-DVB) capsules become perfect hollow sphere through the hypercrosslinking reaction. The diameter and shell thickness of HCSs could be controlled by changing DVB dosage. Highly crosslinked rigid outer shell produced by the secondary crosslinking reaction is critical to obtain discrete and dispersible HCSs, and resolves conglutination among the carbon particles during high-temperature annealing. Hierarchical pore structure of HCSs derives from that of HHPS constructed by hypercrosslinking reaction of the mesoporous HPC precursors. The HCS-30 shows a high specific capacitance of up to 192 F g⁻¹ at 5 mV s⁻¹, which is greatly superior to conglutinated SCSs and porous carbon materials with high micropore rate. This should be ascribed to high S_micro, V_micro and V_meso, hierarchical pore structure and large internal volume. The secondary-crosslinking pyrolysis method opens a new and efficient avenue for fabrication of ideal HCSs that have potential application in energy storage devices, self-assemble, adsorption, catalysis, drug carriers, inks, and so on.

Acknowledgements

We are grateful for the National Natural Science Foundation of China (No. 51202223, 51173170), the Program for New Century Excellent Talents in University (NCET) and the Key program of science and technology (121PZDGG213) from Zhengzhou Bureau of science and technology.

Notes and references

1. A. H. Lu, W. C. Li, G. P. Hao, B. Spliethoff, H. J. Bongard, B. B. Schaack and F. Schüth, Angew. Chem., Int. Ed., 2010, 49, 1615–1618.
2. R. J. White, K. Tauer, M. Antonietti and M. M. Titirici, J. Am. Chem. Soc., 2010, 132, 17360–17363.
3. X. M. Sun and Y. D. Li, Angew. Chem., Int. Ed., 2004, 43, 597–601.
4. Z. B. Lei, Z. W. Chen and X. S. Zhao, J. Phys. Chem. C, 2010, 114, 19867–19874.
5. L. M. Guo, J. M. Zhang, Q. J. He, L. X. Zhang, J. J. Zhao, Z. Y. Zhu, W. Wu, J. Zhang and J. L. Shi, Chem. Commun., 2010, 46, 7127–7129.
6. Y. Li and J. Shi, Adv. Mater., 2014, 26, 3176–3205.
7. Z. L. Schaefer, M. L. Gross, M. A. Hickner and R. E. Schaak, Angew. Chem., 2010, 122, 7199–7202.
8. T. Harada, S. Ikeda, Y. H. Ng, T. Sakata, H. Mori, T. Torimoto and M. Matsumura, Adv. Funct. Mater., 2008, 18, 2190–2196.
9. R. Liu, S. M. Mahurin, C. Li, R. R. Unocic, J. C. Idrobo, H. Gao, S. J. Pennycook and S. Dai, Angew. Chem., Int. Ed., 2011, 50, 6799–6802.
10. K. J. Lee, Y. J. Sa, H. Y. Jeong, C. W. Bielawski, S. H. Joo and H. R. Moon, Chem. Commun., 2015, 51, 6773–6776.
11. C. Argyo, V. Weiss, C. Brauchle and T. Bein, Chem. Mater., 2014, 26, 435–451.
12. I. I. Slowing, J. L. Vivero-Escoto, C. W. Wu and V. S. Lin, Adv. Drug Delivery Rev., 2008, 60, 1278–1288.
13. L. Guo, L. X. Zhang, J. Zhang, J. M. Zhou, Q. J. He, S. Z. Zeng, X. Z. Cui and J. L. Shi, Chem. Commun., 2009, 25, 6071–6073.
14. Y. X. Zhang, S. C. Xu, Y. Y. Luo, S. S. Pan, H. L. Ding and G. H. Li, J. Mater. Chem., 2011, 21, 3664–3671.
15. N. Jayaprakash, J. Shen, S. S. Moganty, A. Corona and L. A. Archer, Angew. Chem., 2011, 123, 6026–6030.
16. W. M. Zhang, J. S. Hu, Y. G. Guo, S. F. Zheng, L. S. Zhong, W. G. Song and L. J. Wan, Adv. Mater., 2008, 20, 1160–1165.
17. S. B. Yang, X. L. Feng, L. J. Zhi, Q. Cao, J. Maier and K. Müllen, Adv. Mater., 2010, 22, 838–842.
18. J. H. Kim and J. S. Yu, Phys. Chem. Chem. Phys., 2010, 12, 15301–15308.
19. D. R. Dreyer and C. W. Bielawski, Adv. Funct. Mater., 2012, 22, 3247–3253.
20. D. R. Dreyer, S. Murali, Y. W. Zhu, R. S. Ruoff and C. W. Bielawski, J. Mater. Chem., 2011, 21, 3443–3447.
21. W. Meier, Chem. Soc. Rev., 2000, 29, 295–303.
22. F. Caruso, Chem.–Eur. J., 2000, 6, 413–419.
23. A. Izadi-Najafabadi, D. N. Futaba, S. Iijima and K. Hata, J. Am. Chem. Soc., 2010, 132, 18017–18019.
24. A. Kajdos, A. Kvit, F. Jones, J. Jagiello and G. Yushin, J. Am. Chem. Soc., 2010, 132, 3252–3253.
25. H. Zhong, F. Xu, Z. G. Li, R. W. Fu and D. C. Wu, Nanoscale, 2013, 5, 4678–4682.
26. D. W. Wang, F. Li, M. Liu, G. Q. Lu and H. M. Cheng, Angew. Chem., 2007, 120, 379–382.
27. J. Chmiola, G. Yushin, Y. Gogotsi, C. Portet, P. Simon and P. L. Taberna, Science, 2006, 313, 1760–1763.
28. K. S. Xia, Q. M. Gao, J. H. Jiang and J. Hu, Carbon, 2008, 46, 1718–1726.
29. W. Xing, C. C. Huang, S. P. Zhuo, X. Yuan, G. P. Wang, D. HulicovaJurcakova, Z. F. Yan and G. Q. Lu, Carbon, 2009, 47, 1715–1722.
30. G. Z. Yang, H. Cui and C. X. Wang, ChemPhysChem, 2014, 15, 374–381.
31. S. Mezzavilla, C. Baldizzone, K. J. J. Mayrhofer and F. Schüth, ACS Appl. Mater. Interfaces, 2015, 7, 12914–12922.
32. X. C. Chen, K. Kierzek, Z. W. Jiang, H. M. Chen, T. Tang, M. Wojtoniszak, R. J. Kalenczuk, P. K. Chu and E. Borowiak-Palen, J. Phys. Chem. C, 2011, 115, 17717–17724.
33. S. Ikeda, K. Tachi, Y. H. Ng, Y. Ikoma, T. Sakata, H. Mori, H. Harada and M. Matsumura, Chem. Mater., 2007, 19, 4335–4340.
34. D. C. Wu, Z. H. Li, Y. R. Liang, X. Q. Yang, X. H. Zeng and R. W. Fu, Carbon, 2009, 47, 916–918.
35. J. B. Pang, X. Li, D. H. Wang, Z. W. Wu, V. T. John and Z. Z. Yang, Adv. Mater., 2004, 16, 884–886.
36. D. Kawashima, T. Aihara, Y. Kobayashi, T. Kyotani and A. Tomita, Chem. Mater., 2000, 12, 3397–3401.
37. A. H. Lu, T. Sun, W. C. Li, Q. Sun, F. Han, D. H. Liu and Y. Guo, Angew. Chem., Int. Ed., 2011, 50, 1–5.
38. Y. D. Xia and R. Mokaya, Adv. Mater., 2004, 16, 1553–1558.
39. T. W. Kim, I. S. Park and R. R. Prof, Angew. Chem., 2003, 115, 4511–4515.
40. S. B. Yoon, K. Sohn, J. Y. Kim, C. H. Shin, J. S. Yu and T. Hyeon, Adv. Mater., 2002, 14, 19–21.
41. Y. R. Liang, F. X. Liang, H. Zhong, Z. H. Li, R. W. Fu and D. C. Wu, J. Mater. Chem. A, 2013, 1, 7000–7005.
42. J. F. Chen, Z. L. Lang, Q. Xu, J. N. Zhang, J. W. Fu and Z. M. Chen, Phys. Chem. Chem. Phys., 2013, 15, 17786–17792.
43. J. Liang, X. Y. Yu, H. Zhou, H. B. Wu, S. J. Ding and X. W. Lou, Angew. Chem., Int. Ed., 2014, 53, 12803–12807.
44. S. S. Cao, X. Jin, X. H. Yuan, W. W. Wu, J. Hu and W. C. Sheng, J. Polym. Sci., Part A: Polym. Chem., 2010, 48, 1332–1338.
45. M. A. Malik, Ind. Eng. Chem. Res., 2009, 48, 6961–6965.
46. Y. Li, J. F. Chen, Q. Xu, L. H. He and Z. M. Chen, J. Phys. Chem. C, 2009, 113, 10085–10089.
47. K. Kaewtatip and V. Tanrattanakul, Carbohydr. Polym., 2008, 73, 647–655.
48. J. Hradil and E. Králová, Polymer, 1998, 39, 6041–6048.
49. D. Bardelang, A. Brinkmann, C. I. Ratcliffe, J. A. Ripmeester, V. V. Terskikh and K. A. Udachin, CrystEngComm, 2014, 16, 3788–3795.
50. F. W. Ma, H. Zhao, L. P. Sun, Q. Li, L. H. Huo, T. Xia, S. Gao, G. S. Pang, Z. Shi and S. H. Feng, J. Mater. Chem., 2012, 22, 13464–13468.
51. Z. Yang, Y. Xia, X. Sun and R. Mokaya, J. Phys. Chem. B, 2006, 110, 18424–18431.
52. J. Jajiello, C. O. Ania, J. B. Parra, L. Jagiello and J. J. Pis, Carbon, 2007, 45, 1066–1071.
53. K. Xie, X. T. Qin, X. Z. Wang, Y. N. Wang, H. S. Tao, Q. Wu, L. J. Yang and Z. Hu, Adv. Mater., 2012, 24, 347–352.
54. Q. Li, R. R. Jiang, Y. Q. Dou, Z. X. Wu, T. Huang, D. Feng, J. P. Yang, A. S. Yu and D. Y. Zhao, Carbon, 2011, 49, 1248–1257.
55. Y. T. Gong, Z. Z. Wei, J. Wang, P. F. Zhang, H. R. Li and Y. Wang, Sci. Rep., 2014, 4, 6349–6354.
56. Z. H. Li, D. C. Wu, Y. R. Liang, F. Xu and R. W. Fu, Nanoscale, 2013, 5, 10824–10828.

---

Footnote

† Electronic supplementary information (ESI) available: Details of N₂ adsorption–desorption, PSD, SEM, CV curves, CP curves, Nyquist plots and average molecular weight. See DOI: 10.1039/c5ra25932k

---