Morphology control and supercapacitor performance of resorcinol–formaldehyde-based carbon particles upon Ni loading in an inverse emulsion system

Abstract

Ni-doped carbon particles with controlled morphology were synthesized by the carbonization of [Ni(H₂O)₆](NO₃)₂-doped resorcinol–formaldehyde aerogel particles extracted from an inverse emulsion polymerization system. Ni was introduced into the reaction system by directly adding nickel salt into the resorcinol–formaldehyde solution. The morphology and structure of the Ni-doped carbon particles were investigated by TEM, SEM, XRD, FI-IR and BET measurements. The size distribution of the Ni-doped carbon particles can be controlled by the stirring speed of the inverse emulsion polymerization system, and the morphology of the prepared carbon particles can be adjusted to be spheres, semispheres, irregular semispheres and capsules by changing the nickel salt concentration. The Ni particles were distributed uniformly in the carbon network. This study illustrates that [Ni(H₂O)₆](NO₃)₂ changed the resorcinol–formaldehyde inverse emulsion polymerization mechanism. Upon ambient drying, the Ni-doped carbon particles exhibit mainly microporosity and the BET surface area of the samples can reach 487 m² g⁻¹ with a corresponding pore volume of 0.229 cm³ g⁻¹. The electrochemical performance was tested using these carbon particles as the electrode material for supercapacitors. The prepared carbon capsules displayed a stabilized capacity of 154 F g⁻¹ after 1200 cycles with an increasing trend, which indicates that the materials have good electrochemical performance.

---

1. Introduction

Carbon particles with different morphologies have many applications such as composite or reinforcing materials,^1,2 chemical electrodes and supercapacitors,^3–5 catalysts or catalytic agents^6,7 and H-storage materials.^8,9 Among the different morphologies, carbon hollow particles, carbon hemispheres and capsules have attracted a great deal of attention for their unique properties such as light weight, low density and derivatives-matrix used as templates or shells for new materials.^10–13 Carbon hemispheres/semispheres are generally synthesized via carbonization of their corresponding precursors derived from solvent released organic hollow spheres,^14–16 and carbon hollow particles have been synthesized using various methods, including the template method,^17–20 high temperature pyrolysis,^21–23 hydrothermal or chemical reduction^24–26 and carbonization of polymer precursors.^27–29 Most of these typical chemical or physical approaches are limited and less controllable in industrial production because of the complex requirements for pressure and temperature.

In the recent years, resorcinol–formaldehyde (RF) particles with various morphologies or metal doped structures have been widely developed as a new category of precursor for morphology and structure controlled carbon particles.^30–35 For example, Yi-Feng Lin et al.³⁰ successfully developed tri-functional mesoporous composite γ-Fe₂O₃/α-Fe₂O₃/carbon aerogel structures. Miller et al.³¹ explored the structure–property relationships of Ru/carbon aerogel composite materials by changing the Ru(ac)₃ concentration using a novel two-step metal vapor impregnation method. Maldonado-Hódar et al.³² investigated the synthesis and pore structure characteristics of supercritical dried Ag, Pd or Pt containing organic aerogels and their carbonized derivatives, proving that the textural characteristics of the transition-metal-containing activated carbon aerogels depended on the nature of the metal. Horikawa et al.³³ synthesized size controlled supercritical dried RF carbon aerogel particles by adjusting the apparent viscosity of the RF sol. Zhang et al.³⁴ synthesized carbon particles with various morphologies, including hollow spheres, bowl-like hollow structures and microcapsules by adjusting the pH values of the RF precursor before adding it to an double oil phase inverse-emulsion system. Sharma et al.³⁵ synthesized a variety of dense and open-architecture amorphous carbon xerogel microspheres and folded fractal-like structures by controlling the synthesis parameters, including stirring time, resorcinol/catalyst ratio and surfactant concentration. All the abovementioned studies were aimed at either preparing modified RF carbon aerogels or control over the morphology of RF spheres using complex conditions, including high temperature requirements, composite solvents, double oil phase and supercritical drying or freezing drying, making it more difficult and expensive to control the system, and limits the application of the carbon aerogels.

Herein, we reported a simple strategy to prepare RF-based carbon particles with controllable morphologies, including carbon semispheres, irregular semispheres and capsules by introducing a certain amount of [Ni(H₂O)₆](NO₃)₂ into the RF solution in an inverse emulsion system. The morphology of the prepared carbon particles can be adjusted to be spheres, semispheres, irregular semispheres and capsules by changing the [Ni(H₂O)₆](NO₃)₂ concentration. The electrochemical properties of the synthesized particles were tested as the electrode material for supercapacitors. When the initial molar ratio of resorcinol to [Ni(H₂O)₆](NO₃)₂ is 20 : 1, the morphology of the prepared RF based carbon particles is capsules, which have a corresponding BET surface area of 487 m² g⁻¹ and a stabilized capacity of 154 F g⁻¹. The materials proved to be applied as a promising candidate in electrode materials.

2. Experimental section

2.1. Sample preparation

The [Ni(H₂O)₆](NO₃)₂ doped RF aerogel (Ni/RFA) precursor was synthesized using inverse-emulsion polymerization. The proportion of resorcinol, formaldehyde, [Ni(H₂O)₆](NO₃)₂, catalyst (Na₂CO₃) and water were decided by the molar ratio, and the amount of resorcinol used for each sample is 0.04 mol. The molar ratio of n(resorcinol) : n(formaldehyde), n(resorcinol) : n(Na₂CO₃), n(resorcinol) : n(water) and n(resorcinol) : n(Ni) was 1 : 2, 100 : 1, 1 : 200 and 4 : 1, respectively. The synthesis procedure of Ni/RFA was as follows: resorcinol was dissolved in formaldehyde under a magnetic stirring apparatus for 15 minutes and then, [Ni(H₂O)₆](NO₃)₂ was added into the RF solution. After 20 minutes, Na₂CO₃ was dissolved in water and added into the RF–Ni system. The reaction system was stirred for 20 minutes and added to an oil phase composed of 10 mL Span-80 and 90 mL cyclohexane in a 250 mL three-necked flask under a stirring speed of 400 rpm. After 72 h stirring at 25 °C, the mixed solution was washed with acetone and dried under ambient pressure and normal temperature for 3 days. During the preparation of the Ni/RFA semispheres, the RF–Ni system was found to solidify when kept stirring for 40 minutes at room temperature without adding the Na₂CO₃ solution as a catalyst and Span-80 solution as the surfactant. After solidification, the product was washed 3 times with 50 mL deionized water to remove the free Ni²⁺ and NO₃⁻, and then dried at 80 °C in a drying oven for 24 h. The directly solidified product derived from the RF–Ni system is named solidified-Ni/RFA (s-Ni/RFA).

Carbonization of the resultant Ni/RFA particles was carried out at 700 °C for 2 h under a N₂ atmosphere in a tubular furnace with a heating rate of 2 °C min⁻¹ to obtain nickel doped RF carbon particles (Ni/CRF). The abovementioned synthetic procedure is illustrated in Fig. 1. During carbonization, Ni(NO₃)₂ was decomposed to NiO and then reduced to Ni particles. Before being tested for their electrochemical performance, the samples were treated with 1 mol L⁻¹ HCl at 80 °C for 4 h to remove the Ni nanoparticles distributed in the resorcinol–formaldehyde carbon (CRF) skeleton.

Fig. 1 Synthetic procedure to prepare the Ni/CRF particles.

The n(resorcinol) : n(Ni) molar ratio was changed from 4 : 1 to 20 : 1 to inspect the influence of [Ni(H₂O)₆](NO₃)₂ on the morphologies of the prepared samples. The stirring speed of the inverse emulsion system was adjusted from 400 rpm to 1000 rpm to prepare size controlled Ni/CRF particles.

2.2. Characterization

The morphologies and structures of the samples were studied by scanning electron microscopy (SEM, ZEISS SUPRA 55) and transmission electron microscopy (TEM, Hitachi H-800). Thermogravimetric analysis (TG) and differential scanning calorimetry (DSC) were carried out on a NETZSCH STA 449C simultaneous thermal instrument. The sample (about 10 mg) was heated at a heating rate of 5 °C min⁻¹ from room temperature to 1000 °C under an argon atmosphere. X-ray diffraction (XRD) patterns were recorded on a Rigaku D/max-2500B2+/PCX system operating with Cu Kα radiation (λ = 1.5406 Å). Nitrogen adsorption–desorption isotherms were measured by means of an ASAP 2020 Micromeritics Instrument at 77 K. The samples were degassed at 250 °C for 6 h before measurement. Specific surface areas were calculated using the Brunauer–Emmett–Teller (BET) equation in a relative pressure range of P/P₀ = 0.05–0.3. The pore size distribution was estimated from the desorption branch of the isotherm using both the Barrett–Joyner–Halenda (BJH) method and the Non-Local Density Functional Theory (NLDFT) method to calculate the mesoporous and microporous size distributions.

The electrochemical study was carried out in a standard three-electrode setup using 30 wt% aqueous KOH solution as the electrolyte. The prepared electrode, nickel foil and Hg/HgO were applied as the working, the counter, and the reference electrodes, respectively. A galvanostatic charge–discharge test was performed using a CT2001A Battery Program Controlling Test System (China-Land Com. Ltd) with a potential ranging from 0.01 to 0.9 V, which was used to calculate the specific capacitance of the GNS-SG electrodes from the slope of the discharge curves (dV/dt). Cyclic voltammetry and electrochemical impedance spectroscopy (EIS) were carried out on a CHI 660B electrochemical working station. For the cyclic voltammetry measurements, the sweep rate was varied from 1 to 300 mV s⁻¹ within a potential range of −0.9 to 0 V. The EIS measurements were performed with a frequency range of 1 Hz to 10 kHz.

3. Results and discussion

3.1 Effects of Ni content on the morphologies of Ni/CRF particles

The effects of Ni content on the morphologies of Ni/CRF particles were investigated and the results are shown in Fig. 2. Compared with the spheres formed in a pure RF system without [Ni(H₂O)₆](NO₃)₂ (Fig. 2a), Fig. 2b–f show that with the resorcinol and Ni molar ratio increasing from 4 : 1 to 20 : 1, the morphologies of the prepared Ni/CRF particles change from semispheres (Fig. 2b and c, n(resorcinol) : n(Ni) = 4 : 1, 6 : 1), irregular semispheres with concave surface (Fig. 2d and e, n(resorcinol) : n(Ni) = 10 : 1, 15 : 1) to capsules (Fig. 2f, n(resorcinol) : n(Ni) = 20 : 1). The morphologies of the samples were influenced by the initial [Ni(H₂O)₆](NO₃)₂ concentration; a higher [Ni(H₂O)₆](NO₃)₂ concentration engenders more regular semispheres and lower sphericity, and hollow particles with sunken surface were the intermediate product during the conversion of hollow particles to semispheres. The detailed description of the conversion progress will be discussed later in the mechanism section. The inset of Fig. 2b is the image of the corresponding Ni/RFA before carbonization. The semispherical morphology shown in the inset of Fig. 2b indicates that the morphology of organic Ni/RFA was preserved in the corresponding Ni/CRF particles. Fig. 2g shows the fractured edge of the semispheres prepared at n(resorcinol) : n(Ni) = 4 : 1 and that the Ni/CRF semispheres have a double-wall structure. Fig. 2 suggests that (a) a higher [Ni(H₂O)₆](NO₃)₂ concentration engenders more regular semispheres and (b) the morphologies of Ni/RFA is preserved in the Ni/CRF particles and these Ni/RFA particles are proved to be thermosetting. According to the abovementioned conclusions, it is speculated that [Ni(H₂O)₆](NO₃)₂ changes the synthesis mechanism of the RF aerogel particles when introduced into the inverse emulsion polymerization system.

Fig. 2 SEM images of Ni/CRF with different [Ni(H₂O)₆](NO₃)₂ concentrations: (a) n(Ni) = 0; (b) and (g) n(resorcinol) : n(Ni) = 4 : 1; (c) 6 : 1; (d) 10 : 1; (e) 15 : 1 and (f) 20 : 1.

Fig. 3a–c show the TEM images of Ni/CRF of the semispheres, concave semispheres and capsules, respectively. Fig. 3a shows that Ni particles with size between 15 and 50 nm were uniformly encapsulated in the CRF semispheres framework, which was proved more clearly in Fig. 3d, the TEM image of the fractured surface of the Ni/CRF semispheres. Fig. 3b and c show that the amount of Ni encapsulated in the semispheres is the highest, the second is that of the irregular semispheres and the least is that of the concave capsules. Fig. 3 proves that the Ni/CRF particles semispherical morphology regularity decreased with a decrease in the [Ni(H₂O)₆](NO₃)₂ concentration. Fig. 3f shows the microstructure of the semispheres acidized by 1 mol L⁻¹ HCl at 80 °C for 4 h and the amount of Ni is significantly less than that of the unacidized semispheres, as shown in Fig. 3a. Fig. 3d and e also show that the Ni/CRF skeleton is formed surrounding the Ni particles, suggesting that in the RF solution, the sol–gel process is carried out surrounding the reaction center formed by the [Ni(H₂O)₆](NO₃)₂–RF nucleus. Combined with the SEM images, it can be concluded as follows: during the process of ambient drying, samples with a regular semisphere structure could be subjected to shrinkage stress and form semispheres with a smooth surface, suggesting that the strength of the RF framework with a higher [Ni(H₂O)₆](NO₃)₂ content was stronger than that with a lower [Ni(H₂O)₆](NO₃)₂ content.

Fig. 3 TEM images of (a) semispheres; (b) irregular semispheres; (c) capsules; (d) edge of the semispheres; (e) fractured surface of the semispheres and (f) acidized semispheres.

3.2 Effects of the stirring rate on controlling the particle size in the inverse emulsion system

Fig. 4 shows the images of the Ni/CRF semispheres (n(resorcinol) : n(Ni) = 4 : 1) prepared under an increasing stirring rate. The particle size of the samples was the average value of 50–100 particles in the SEM images. The calculated particle size under a stirring speed of 400 rpm, 600 rpm, 800 rpm and 1000 rpm of the prepared Ni/CRF semispheres were 20 μm, 8 μm, 1.2 μm and 1.8 μm, respectively, and the corresponding SEM images are shown in Fig. 4a–d. The abovementioned result is summarized in Fig. 4f, the Ni/CRF particle size-stirring rate curve. Fig. 4f shows that when the stirring speed is less than 800 rpm, the particle size decreased with a higher stirring speed; on the contrary, the particle size increases.

Fig. 4 SEM images of Ni/CRF under a stirring rate of (a) 400 rpm, (b) 600 rpm, (c) 800 rpm and (d) 1000 rpm. (e) Ni/CRF particle size-stirring rate curve.

Under the function of stirring, the dispersed phase (RF–[Ni(H₂O)₆](NO₃)₂ solution in this study) was covered by the continuous phase (oil phase), and the sol–gel process was limited to the small units cut by the stirring shearing stress. This phenomenon is typical in the emulsion polymerization or inverse emulsion polymerization systems;^36,37 in a certain low stirring range, a higher stirring rate induced smaller reaction units and particle size to maintain the balance between shearing stress and surface tension. When the stirring speed reaches a certain value, the system stability decreases and small particles tend to assimilate to reduce the surface energy and achieve a higher system stability.

It can be seen in Fig. 2 that the particle size distribution became dispersive and the morphologies became irregular with a decrease in the Ni concentration. When n(resorcinol) : n(Ni) = 4 : 1 and 6 : 1, the morphologies are regular semispheres and the particle size distributions are narrow. When n(resorcinol) : n(Ni) = 10 : 1 and 15 : 1, a low Ni concentration decreased the strength and stability of the RF-sol before being added into the oil phase; therefore, the particle size decreased and the size distribution became dispersive. For the samples n(resorcinol) : n(Ni) = 20 : 1, the morphology is more closer to Fig. 1a, pure RF spheres without Ni. Furthermore, different Ni concentrations engendered differing pH because of the hydrolysis of nickel salt and influenced the sol–gel process and RF sol particle size before being adding into the oil phase. The influence of stirring speed discussed above is consistent with the tendency of a typical inverse emulsion system; therefore, the stirring speed is the main factor considered for controlling the particle size with different concentrations of Ni.

3.3 Morphology controlling mechanism

3.3.1 XRD diffraction. The preparation of s-Ni/RFA without the addition of Na₂CO₃ suggests that [Ni(H₂O)₆](NO₃)₂ can play the catalytic role for the original RF gelation. Fig. 5 shows that the XRD pattern of s-Ni/RFA is consistent with that of [Ni(H₂O)₆](NO₃)₂. The patterns of the Ni/RFA semispheres show wide diffraction peaks at 15.9°, corresponding to the (0 2 0) crystal face of [Ni(H₂O)₆](NO₃)₂. A large amount of nickel salt distributed in the Ni/RFA was removed during the washing of the oil phase with acetone, leading the nickel salt to transform from crystals to an amorphous phase. As in the s-Ni/RFA samples, the short solidification process maintained most of [Ni(H₂O)₆](NO₃)₂ in the s-Ni/RFA skeleton, and therefore [Ni(H₂O)₆](NO₃)₂ was still crystalline. The pattern for Ni/CRF shows the diffraction peaks of crystal carbon at 26.4° and nickel at 44.8°, 51.9° and 76.5°, indicating the decomposition and reduction of the nickel salt. After 2 hours of calcination in air at 350 °C, the XRD pattern of Ni/CRF was not changed, proving that the Ni particles were wrapped by a carbon skeleton and not oxidized, as previously shown in Fig. 3a and d.

Fig. 5 XRD patterns of (a) Ni/RFA semispheres, s-Ni/RFA and Ni/CRF semispheres. (b) The standard spectrum of [Ni(H₂O)₆](NO₃)₂.

3.3.2 Thermogravimetric and differential thermal analysis. TG and DSC tests were used to investigate the interactions between [Ni(H₂O)₆](NO₃)₂ and the RF solution. The TG and DTG curves of (a, a′) [Ni(H₂O)₆](NO₃)₂, (b, b′) RFA spheres and (c, c′) Ni/RFA semispheres are shown in Fig. 6A and the corresponding DSC curves are shown in Fig. 6B. In Fig. 6A-a, the mass loss of [Ni(H₂O)₆](NO₃)₂ (72%) is composed of the total loss of crystal water (0–270 °C) and Ni(NO₃)₂ to NiO (288–434 °C), and it is close to the theoretical value of 74%. Compared with the RFA spheres TG and DTG curves (Fig. 6A-b, b′), the Ni/RFA semispheres curve (Fig. 6A-c, c′) shows a faster weight loss before 162 °C, corresponding to the weight loss of partly volatilized crystal water of [Ni(H₂O)₆](NO₃)₂ in Ni/RFA before this temperature. From 162 to 280 °C, [Ni(H₂O)₆](NO₃)₂ has no impact on the weight loss value of the Ni/RFA TG and DTG curves, implying that the remained crystal water was constrained by the hydrogen-bond interaction between [Ni(H₂O)₆](NO₃)₂ and the RF structure.³⁸ The following differences between Fig. 6A-c and b in the range of 280–320 °C and 320–435 °C correspond to the crystal water loss dropped by the broken hydrogen-bonds and the decomposition of Ni(NO₃)₂ to NiO. The slight weight loss rate differences between Fig. 6A-c and Fig. 6A-b ranging from 550 to 681 °C was due to the reduction of NiO to Ni by carbon,^39,40 also leading to a lower residual weight of Ni/RFA. The endothermic and exothermic peaks before 700 °C in the DSC curves shown in Fig. 6B correspond to the weight change in Fig. 6A very well, and the exothermic peak in Fig. 6B-c may be caused by the preliminary crystallization of carbon promoted by the transition metal, Ni.⁴¹

Fig. 6 (A) TG and DTG, and (B) DSC curves for (a, a′) [Ni(H₂O)₆](NO₃)₂, (b, b′) RFA spheres and (c, c′) Ni/RFA semispheres.

3.3.3 Possible mechanism of morphology controlled Ni/CRF. As mentioned above, [Ni(H₂O)₆](NO₃)₂ changed the RF polymerization mechanism. The optical images of the Ni/RFA capsules during the drying process are shown in Fig. 7, which show the proposed formation mechanism model of the Ni/CRF particles. During the drying process, the spherical degree of the given sample decreased with prolonged drying time. The morphologies of the samples are based on the hollow particles formed in the O/W inverse emulsion system. Combined with Fig. 2g, the RF hollow particles collapse when drying under normal temperature and pressure because of the release of solvent inside the hollow structure. The abovementioned conclusion reveals that the shell of the hollow particles is formed by the RF sol–gel process⁴² and the shell structure of the hollow particles was influenced by the concentration of [Ni(H₂O)₆](NO₃)₂, because the hydrolysis of Ni²⁺ changes the pH value of the sol–gel solution and affects the aerogel structure.⁴³ Before being added into the oil phase, the RF sol–gel achieves a higher cross-linking degree, differing from the typical alkaline RF polymerization due to the chelation of [Ni(H₂O)₆](NO₃)₂.³⁸ Tiny droplets of the oil phase surrounded by the cross-linked structure provide a nucleus for the hollow particles and the differential of [Ni(H₂O)₆](NO₃)₂ concentration between the oil phase and the tailored water phase drives the metal salt to the O/W interface, and a higher concentration of metal salt catalyzes the interfacial reaction, promoting the formation of the hollow structure. Fig. 7 proves that the hollow particles show weak mechanical strength. During drying in air, the relatively strong capillary force caused by removing the solvent from the small pores causes the collapse of the initially formed hollow particles to a semi-spherical structure. When the hollow particles collapse, the flexible shell tends to form poly-directional and irregular collapsing, leading to an irregular morphology. Fig. 2d–f and 3b and c show that the Ni/CRF particles with a lower Ni concentration tend to have an irregular morphology, proving that the Ni concentration is a factor for the mechanical strength of the Ni/RFA hollow particles. Moreover, during the release of solvent in the hollow particles, small particles need to overcome the capillary force and have a long contracting process, bringing about a non-uniform surface stress and the formation of an irregular morphology. This metal related Ni/RFA formation mechanism is generalized by the collapse theory of metal doped hollow RF spheres.

Fig. 7 (a) The synthetic mechanism model for the Ni/CRF particles and (b) optical images of the Ni–RFA capsules during the drying process.

3.4 Specific surface area and pore structure analysis

Fig. 8a and b show the N₂ adsorption–desorption isotherms and the pore size distributions of the typical Ni/CRF semispheres (n(resorcinol) : n(Ni) = 4 : 1) irregular semispheres (irre-semi, n(resorcinol) : n(Ni) = 10 : 1) and capsules (n(resorcinol) : n(Ni) = 10 : 1). According to Fig. 8a, the H₂-type hysteresis loop in the isotherm indicates a 3D cage-like pore structure of a mesoporous material.^44,45 Table 1 illustrates the pore structure of the samples, and the normal CRF spheres without Ni show mainly mesoporosity and the samples containing Ni exhibit mainly microporosity. During the preparation of Ni/RFA, the pH of [Ni(H₂O)₆](NO₃)₂–RF was tested 15 min after dissolving [Ni(H₂O)₆](NO₃)₂ and the pH of the Ni/RFA system in this study is in the range of 2.5–6.9, differing by the original [Ni(H₂O)₆](NO₃)₂ concentration. Small sol–gel microparticles were yielded under this weakly acidic conditions, leading the aerogels to have a more developed microporosity network and greater surface area.⁴⁶ The specific surface area and average pore size of the Ni/CRF samples is significantly larger than normal CRF spheres without Ni, proving that [Ni(H₂O)₆](NO₃)₂ enhanced the RF skeleton and brought more resistance to shrinkage stress during the drying step. Before the electrochemical tests, the samples were acidized to remove the Ni nanoparticles and it was found that acidification increased the specific surface area of the semispheres from 445 to 487 m² g⁻¹ and enlarged the pore width of mesopores to 30 nm because of the removal of Ni. According to Fig. 8b, the NLDFT pore distribution analysis, the NLDFT pore size distribution maximum of prepared Ni/CRF concentrated at 0.5–0.7 nm was mainly microporous with some mesoporosity concentrated at 3–4 nm (inset of Fig. 8b).

Fig. 8 (a) Nitrogen adsorption–desorption isotherms for Ni/CRF and (b) incremental pore volume vs. pore width plots obtained by applying the NLDFT method (N₂, 77 K). The inset of (b) shows the incremental pore volume vs. pore width plots obtained from the BJH model.

Table 1 Specific surface area and pore structure analysis of the prepared CRF particles

Samples	Specific surface area^a/m^2 g^−1	t-plot micropore area^b/m^2 g^−1	t-plot external surface area^b/m^2 g^−1	Total pore volume^c/cm^3 g^−1	t-plot micropore volume^c/cm^3 g^−1	Mesopore and macropore volume/cm^3 g^−1
a Calculated using the BJH model according to the sorption data (relative pressure, 0.05–0.3).b Calculated using the t-plot method with the Harkins and Jura standard isotherm with a thickness range of 3.5–5.0 Å.c Estimated from the adsorbing capacity at P/P0 = 0.995.
CRF spheres	396	56	340	0.242	0.028	0.214
Semispheres	445	386	59	0.256	0.201	0.055
Irre-semi	451	388	63	0.265	0.202	0.063
Capsules	485	424	61	0.281	0.221	0.06
Acidized semispheres	487	440	47	0.256	0.229	0.027

---

3.5 Specific capacitance of the CRF electrodes

Fig. 9a shows the cyclic performance of the carbon electrodes from normal CRF spheres, semispheres, irre-semi and capsules under a current density of 0.1 A g⁻¹ (cycle 1–200), 0.2 A g⁻¹ (cycle 201–400), 0.5 A g⁻¹ (cycle 401–600) and 1 A g⁻¹ (cycle 601–1000) in a potential window between 0 and 0.9 V, and Fig. 9c gives the capacitance retention of the CRF electrodes under a current density of 0.1–1 A g⁻¹. From Fig. 9a, we can see that after 1000 cycles, the normal CRF spheres exhibit a reversible capacity of 81.8 F g⁻¹, while the semispheres, irre-semi, and capsules show a higher specific capacitance of 97.4, 97.1 and 111 F g⁻¹, respectively. The introduction of metal salts into the sol–gel process obviously increased the capacitance of the CRF due to the higher surface area engendered by more stable pore structure formed during the ambient drying process and the intermediates with a hollow structure. In the range of experimental conditions, Ni/CRF shows the best electrochemical performance when n(resorcinol) : n(Ni) = 20 : 1, but the semispheres (n(resorcinol) : n(Ni) = 4 : 1) have the most stable performance due to the strong skeleton supported by the Ni nanoparticles. Fig. 9b shows the potential–t curves of the irre-semi CRF from cycle 2 to 11 and the curve is composed of isosceles triangles with good symmetry, indicating that the irre-semi electrodes behave as an electric double layer capacitor and give a high value of charge–discharge efficiency of nearly 97%. The inset of Fig. 9b shows the potential–t curve of the second charge–discharge process of the CRF electrode with different morphologies. The same charge and discharge properties with the irre-semi CRF can be observed; therefore, the Ni/CRF particles prepared in this experiment are suitable electrode materials for EDLCs.

Fig. 9 (a) Capacitance–cycle number curves for the CRF spheres, semispheres, irre-semi and capsules under a current density of 0.1 A g⁻¹, 0.2 A g⁻¹, 0.5 A g⁻¹ and 1 A g⁻¹ and (b) the potential–t curve for the irre-semi from cycle 2 to 10. The inset of (b) shows the potential–t curve for the second charge–discharge cycle of the CRF electrode. (c) Capacitance retention of the CRF electrodes under a large current density of 0.1–1 A g⁻¹. (d) Cyclic voltammograms curves for the CRF particles in 30% KOH at a scan rate of 5 mV s⁻¹. (e) The complex plane impedance plot of the CRF electrodes with different morphologies.

The cyclic voltammograms for the samples with different morphologies recorded in a 30% aqueous KOH solution at a scan rate of 5 mV s⁻¹ are shown in Fig. 9d. Following the trend of the specific surface area shown in Table 1, the calculated specific capacitance of the CRF sphere, semispheres, irre-semi and capsules are 77 F g⁻¹, 80 F g⁻¹, 147 F g⁻¹ and 154 F g⁻¹, respectively. The abovementioned capacitance values are also consistent with the result of constant current charge and discharge (Fig. 9a and c). The larger pore volume is also responsible for the higher specific capacitance of the capsules and irre-semi CRF particles. It should be pointed out that the specific capacitances in this study are superior to the latest reported traditional CRF materials.^47–49 For example, Ann Laheäär⁴⁷ et al. reported that the capacitance values calculated for a supercritical dried carbon aerogel (CAG)-based system was ∼55 F g⁻¹ and the value for a carbide-derived carbon C(Mo₂C)-based system was ∼125 F g⁻¹. Yoon Jae Lee⁴⁸ et al. prepared carbon aerogels with a high BET surface area (706 m² g⁻¹) under ambient drying and the specific capacitance was found to be 81 F g⁻¹. Zulamita Zapata-Benabithe⁴⁹ et al. prepared supercritical CO₂ dried Cu and Ag doped carbon aerogels and the corresponding specific capacitance were 100 F g⁻¹ and 76 F g⁻¹, respectively.

Furthermore, the calculated surface area specific capacity of the CRF sphere, semispheres, irre-semi and capsules are 0.18 F m⁻², 0.18 F m⁻², 0.33 F m⁻² and 0.32 F m⁻², respectively. If the initial surface of the pores in the network can be fully used, the theoretical surface area specific capacity of the irregular semispheres will be 0.36.⁵⁰ As in the prepared irre-semi and capsules, the test value is nearly the same as the theoretical value. The high surface area specific capacity of the irre-semi and capsules may be caused by the electrolyte infiltrated in the hollow core and fold of the collapsed area, which can reduce the internal resistance, being responsible for the higher capacity. The semispheres show the same surface area specific capacity with the spheres with the same solid structure without the advantage of storage in the electrolyte. Fig. 9e shows the electrochemical impedance plots for the CRF samples with different morphologies. The inset represents the high frequency semicircles. The Nyquist plots are composed of a high frequency semicircle, which represents the charge transfer process at the electrode/electrolyte interface and a straight line at low frequency, indicating the diffusion limiting step in the electrochemical processes. In the low frequency region, the CRF particles electrode shows ideal capacitive behavior with a near vertical line parallel to the imaginary axis.⁵¹ In the high frequency region, the Nyquist plots show that the high frequency impedance of the CRF spheres, semispheres, irre-semi, and capsule is 0.304, 0.324, 0.308, 0.310 Ω. The morphology of semisphere may reduce the contact between the particles, which engenders higher internal resistance when used as a capacitor electrode.

4. Conclusions

In this study, morphology-controlled Ni/CRF was prepared by carbonization of Ni/RFA particles derived from the RF inverse-emulsion system in cyclohexane/Span-80 (oil phase). The polymerization and drying processes were carried out under ambient conditions. Morphologies, including semispheres, irregular semispheres and capsules, were controlled by changing the [Ni(H₂O)₆](NO₃)₂ concentration. The particle size of the samples could be controlled by changing the stirring rate. The BET surface area of the ambient dried Ni/CRF particles can reach 485 m² g⁻¹ and the corresponding pore volume turned out to be 0.28 cm³ g⁻¹. The prepared CRF carbon particles displayed a stabilized capacity of 77–154 F g⁻¹ after 1200 cycles with an increasing tendency, differing by the morphologies. [Ni(H₂O)₆](NO₃)₂ changed the reaction mechanism of the RF polymerization and brought in new structures and morphology features. The Ni/CRF particles are promising materials in supercapacitors, potential material for magnetic separated catalysts, desalination materials or templates for chemical reactions.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (51272016 and 51272019).

Notes and references

1. M. Nishizawa, H. Koshika and I. Uchida, J. Phys. Chem. B, 1999, 103, 192.
2. E. Baumeister and S. Klaeger, Adv. Eng. Mater., 2003, 5, 673.
3. B. Chang, Y. Wang, K. Pei, S. Yang and X. Dong, RSC Adv., 2014, 4, 40546.
4. W. M. Zhang, J. S. Hu, Y. G. Guo, S. F. Zheng, L. S. Zhong, W. G. Song and L. J. Wang, Adv. Mater., 2008, 20, 1160.
5. D. Qu and H. Shi, J. Power Sources, 1998, 74, 99.
6. Z. Wen, Q. Wang, Q. Zhang and J. Li, Electrochem. Commun., 2007, 9, 1867.
7. C. Bianchini and P. K. Shen, Chem. Rev., 2009, 109, 4183.
8. A. F. Gross, J. J. Vajo, S. L. van-Atta and G. L. Olson, J. Phys. Chem. C, 2008, 112, 5651.
9. L. Wang and R. T. Yang, J. Phys. Chem. C, 2009, 113, 21883.
10. X. W. Lou, C. M. Li and L. A. Archer, Adv. Mater., 2009, 21, 2536.
11. Y. Wang, E. Hosono, K. Wang and H. Zhou, Angew. Chem., Int. Ed., 2008, 47, 7461.
12. S. B. Yoon, K. Sohn, J. Y. Kim, C. H. Shin, J. S. Yu and T. Hyeon, Adv. Mater., 2002, 14, 19.
13. F. Böttger-Hiller, P. Kempe, G. Cox, A. Panchenko, N. Janssen, A. Petzold, T. Thurn-Albrecht, L. Borchardt, M. Rose, S. Kaskel, C. Georgi, H. Lang and S. Spange, Angew. Chem., Int. Ed., 2013, 52, 6088.
14. Z. Yan, Z. Hu, C. Chen, H. Meng, P. K. Shen, H. Ji and Y. Meng, J. Power Sources, 2010, 195, 7146.
15. J. H. Kim, B. Fang, S. B. Yoon and J. S. Yu, Appl. Catal., B, 2009, 88, 368.
16. G. S. Chai, S. B. Yoon, J. H. Kim and J. S. Yu, Chem. Commun., 2004, 2766.
17. X. Chuan and S. Zhou, Carbon, 2011, 49, 3708.
18. G. Hu, D. Ma, M. Cheng, L. Liu and X. Bao, Chem. Commun., 2002, 1948.
19. B. You, J. Yang, Y. Sun and Q. Su, Chem. Commun., 2011, 47, 12364.
20. R. J. White, K. Tauer, M. Antonietti and M. M. Titirici, J. Am. Chem. Soc., 2010, 132, 17360.
21. Z. Y. Jin, C. Gao, W. K. Hsu, Y. Zhu, A. Huczko, M. Bystrzejewski, M. Roe, C. Y. Lee and S. Acquah, Carbon, 2005, 43, 1944.
22. Y. Xia and R. Mokaya, Adv. Mater., 2004, 16, 886.
23. Y. Y. Song, Y. Li and X. H. Xia, Electrochem. Commun., 2007, 9, 201.
24. X. Yang, C. Li, W. Wang, B. Yang, S. Zhang and Y. Qian, Chem. Commun., 2004, 342.
25. S. Y. Sawant, R. S. Somani, B. L. Newalkar, N. V. Choudary and H. C. Bajaj, Mater. Lett., 2009, 63, 2339.
26. Y. Ni, M. Shao, Y. Tong, G. Qian and X. Wei, J. Solid State Chem., 2005, 178, 908.
27. C. Scherdel, T. Scherb and G. Reichenauer, Carbon, 2009, 47, 2244.
28. F. L. Wang, L. L. Pang, Y. Y. Jiang, B. Chen, D. Lin, N. Lun, H. L. Zhu, R. Liu and X. L. Meng, Mater. Lett., 2009, 63, 2564.
29. W. Li, G. Reichenauer and J. Fricke, Carbon, 2002, 40, 2955.
30. Y. F. Lin and C. Y. Chang, RSC Adv., 2014, 4, 28628.
31. J. Miller and B. Dunn, Langmuir, 1999, 15, 799.
32. F. J. Maldonado-Hódar, M. A. Ferro-García, J. Rivera-Utrilla and C. Moreno-Castilla, Carbon, 1999, 37, 1199.
33. T. Horikawa, J. Hayashi and K. Muroyama, Carbon, 2004, 42, 169.
34. H. Zhang, H. Xu and C. Zhao, Mater. Chem. Phys., 2012, 133, 429.
35. C. S. Sharma, M. M. Kulkarni, A. Sharma and M. Madou, Chem. Eng. Sci., 2009, 64, 1536.
36. M. Nomura, M. Harada, W. Eguchi and S. Nagata, J. Appl. Polym. Sci., 1972, 16, 835.
37. S. A. Bon, M. Bosveld, B. Klumperman and A. L. German, Macromolecules, 1997, 30, 324.
38. O. Celik and O. Dag, Angew. Chem., Int. Ed., 2001, 40, 3800.
39. J. Criado, A. Ortega and C. Real, React. Solids, 1987, 4, 93.
40. W. Brockner, C. Ehrhardt and M. Gjikaj, Thermochim. Acta, 2007, 456, 64.
41. R. W. Fu, T. F. Baumann, S. Cronin, G. Dresselhaus, M. S. Dresselhaus and J. H. Satcher, Langmuir, 2005, 21, 2647.
42. R. Pekala and D. Schaefer, Macromolecules, 1993, 26, 5487.
43. K. Barral, J. Non-Cryst. Solids, 1998, 225, 46.
44. J. R. Matos, M. Kruk, L. P. Mercuri, M. Jaroniec, L. Zhao, T. Kamiyama, O. Terasaki, T. J. Pinnavaia and Y. Liu, J. Am. Chem. Soc., 2003, 125, 821.
45. M. Sevilla and A. B. Fuertes, Carbon, 2006, 44, 468.
46. C. Moreno-Castilla, F. J. Maldonado-Hódar and A. F. Perez-Cadenas, Langmuir, 2003, 19, 5650.
47. A. Laheäär, A. L. Peikolainen, M. Koel, A. Jänes and E. Lust, J. Solid State Electrochem., 2012, 16, 2717.
48. Y. J. Lee, J. C. Jung, J. Yi, S. H. Baeck, J. R. Yoon and I. K. Song, Curr. Appl. Phys, 2010, 10, 682.
49. Z. Z. Benabithe, F. C. Marín and C. M. Castilla, Mater. Chem. Phys., 2013, 138, 870.
50. S. Mayer, R. Pekala and J. Kaschmitter, J. Electrochem. Soc., 1993, 140, 446.
51. L. Li, H. Song and X. Chen, Microporous Mesoporous Mater., 2006, 94, 9.

---