Hydrothermal synthesis of vanadium dioxides/carbon composites and their transformation to surface-uneven V₂O₅ nanoparticles with high electrochemical properties

Abstract

Vanadium dioxides/carbon composites composed of vanadium dioxides@carbon core–shell structures and amorphous carbon spheres were successfully synthesized using glucose as the carbon sources by a facile one-step hydrothermal route. Then vanadium dioxides/carbon composites were converted to surface-uneven V₂O₅ nanoparticles by the calcination in air atmospheres. The amorphous carbon reacting with O₂ in the air to release gas results in remaining V₂O₅ nanoparticles possessing broken, rough and poral structures. The electrochemical properties of surface-uneven V₂O₅ nanoparticles as supercapacitor electrodes were measured by cyclic voltammetry (CV) and galvanostatic charge–discharge (GCD) both in the aqueous and organic electrolyte. Surface-uneven V₂O₅ nanoparticles exhibit the specific capacitance of 406 F g⁻¹ at the current density of 0.2 A g⁻¹ and retain 246 F g⁻¹ even at high current density of 10 A g⁻¹. The influence of the calcined temperature and time on the specific capacitance, phase and morphology of the products were discussed in detail. The results revealed that the calcination at 400 °C for 4 h with comparatively low ratio of V⁵⁺/V⁴⁺ are favorable for surface-uneven V₂O₅ nanoparticles with the high electrochemical property. During the cycle performance, the specific capacitances of V₂O₅ nanoparticles after 100 cycles are 13.8% and 98.5% of the initial discharge capacity in the aqueous and organic electrolyte, respectively, indicating the cycle performance is significantly improved in organic electrolyte. It turns out that surface-uneven V₂O₅ nanoparticles are an ideal material for supercapacitor electrode in the present work.

---

1. Introduction

Supercapacitors (SCs), also called electrochemical capacitors or ultracapacitors, have drawn increasing interest because they can bridge the gap between batteries and conventional solid-state and electrolytic capacitors with superior rate capability, excellent power output, exceptional cycling life, etc.^1–9 These specific properties make SCs available as power sources for next-generation flexible and portable electronics such as miniature biomedical devices, roll-up displays and wearable devices.¹⁰ According to the storage mechanism, SCs can be generally classified into two types: non-faradic process (electrochemical double layer capacitors, EDLCs) and faradic process (pseudocapacitors, PCs). EDLCs physically store charges by reversible ion adsorption at the electrode–electrolyte interface, while PCs chemically store charges by redox at the vicinity of the surface (a few nanometers).^1,2,11 Electrode materials for EDLCs compose of high surface area carbon-based materials, while for PCs are conductive polymers and transition metal oxides/hydroxides/sulfides.^1,12 SCs based on transition metal oxides/hydroxides/sulfides electrodes exhibit much higher specific capacitance than carbon materials and better electrochemical stability than conducting polymers.^13–16 However, the energy density (E) of conventional SCs is commonly less than 10 W h kg⁻¹, which is much lower than those of batteries and fuel cells, hampering their practical applications.⁴ As is well-known, E of a SC is governed by the specific capacitance (C) and voltage (V) (E = 1/2CV²). Thus, to fabricate novel electrode materials with high C value is an efficient strategy to improve E of SCs.¹⁷

Recently, vanadium oxides (VO_x) have attracted great attention as SCs electrodes delivering high energy density and allowing efficient ion diffusion because of their unique layered structure, multiple valence states (+5 to +2) and low cost.^7,18–30 It was reported that the electrochemical performance is closely related to the morphology and crystal structure of electrode materials, for examples, VO_x with specific structures have better electrochemical property than common VO_x owing to the charge storage mechanism.³¹ The contact area between electrode materials and electrolyte can be significantly increased by controlling the morphology of electrode materials. Correspondingly, the distance which the ions transit the electrolyte is shortened greatly.¹ Therefore, synthesizing electrode materials with novel structures to improve their specific capacitance is still a challenge and meaningful for materials scientists.

Among of VO_x, vanadium pentoxide (V₂O₅) is considered to be one of the most promising candidates for SCs electrodes.^31–34 Therefore, in the past decades, V₂O₅ with different morphologies have been synthesized for SCs applications, such as nanofibers,¹⁵ nanowires,³⁵ nanoribbons,³⁶ hollow spheres,¹⁷ nano porous structures,³⁷ nanosheets constructing 3D architectures,³¹ interconnected V₂O₅ nanoporous network³³ and so on. However, the cycle performance of V₂O₅ electrode is not very good and the reason is the dissolution of V(V)-bearing vanadates produced during the charging/discharging process, which is a common phenomenon for VO_x used as electrodes for SCs and Li-ion batteries.^4,38–40 For examples, our previous work¹¹ reported that V₂O₅ microspheres displayed an initial specific capacitance as high as 308 F g⁻¹ at 1 A g⁻¹, however, the specific capacitance after 80 cycles is only 65 F g⁻¹ and its corresponding retention after is 23% of the initial discharge capacity. Thus, to improve the cycle performance of VO_x as SCs electrodes is very important and meaningful. From the investigation of literatures, the cycle performance can be achieved by encapsulating VO_x into carbonaceous materials including carbon nanofibers,⁴¹ carbon nanotubes,⁴² graphene,¹⁹ etc. However, owing to the dissolution of VO_x leading to the poor cycle performance, we can choose a electrolyte in which VO_x is insoluble. In this contribution, we chose the propylene carbonate as the organic electrolyte and it was found the cycle performance can by greatly improved.

Herein, we developed a route to successfully synthesize surface-uneven V₂O₅ nanoparticles by the transformation from vanadium dioxides/carbon composites and their electrochemical properties as the SCs electrode were briefly discussed by CV and GCD both in the inorganic and organic electrolyte.

2. Experimental section

2.1. Synthesis of vanadium dioxides/carbon composites

All the chemicals purchased from Sinopharm Chemical Reagent Co., Ltd were with analytical grade and used without any further purification. The precursor vanadium dioxides/carbon composites was synthesized using commercial V₂O₅, glucose and H₂O as the starting materials based on the previous reports^43,44 and modified slightly. In a typical route, 0.15 g of bulk V₂O₅ was added to the glucose solution (0.81 g of glucose and 35 mL of H₂O) with vigorously stirring at the room temperature. After 1 h, they were transferred into a 50 mL Teflon lined stainless steel autoclave, sealed and maintained at 180 °C for 48 h. After the reaction, the products were filtered off, washed with H₂O and ethanol several times to remove any possible residue, and dried in vacuum at 75 °C.

2.2. Converting vanadium dioxides/carbon composites to V₂O₅ nanoparticles

To obtain V₂O₅ nanoparticles, the above vanadium dioxides/carbon composites were heated in a muffle furnace with 5 °C min⁻¹ heating rate under the air atmosphere at 300–500 °C for various times, and cooled to room temperature naturally.

2.3. Materials characterization

The phase and composition of the products was identified by X-ray powder diffraction (XRD, Panalytical X'Pert powder diffractometer at 40 kV and 40 mA with Ni-filtered Cu Kα radiation). Fourier transform infrared spectroscopy (FTIR) pattern of the solid samples was measured using KBr pellet technique (about 1 wt% of the samples and 99 wt% of KBr were mixed homogeneously, and then the mixture was pressed to a pellet) and recorded on a Nicolet 6700 spectrometer from 4000 to 400 cm⁻¹ with a resolution of 4 cm⁻¹. The Raman spectra were obtained using a Thermo Scientific spectrometer, with a 532 nm excitation line. The chemical composition of as-obtained samples was revealed by use of an energy-dispersive X-ray spectrometer (EDS) attached to a scanning electron microscope (SEM, QUANTA450). X-ray photoelectron spectroscopy (XPS) was used to investigate the composition of the products and confirm the oxidation state of vanadium preformed on ESCALAB250Xi, Thermo Fisher Scientific. The morphology and dimensions of the products were observed by field emission scanning electron microscopy (FE-SEM, NOVA NanoSEM 450, FEI) and transmission electron microscopy (TEM, FEI Tecnai F30, FEI). The samples were dispersed in absolute ethanol with ultrasonication before TEM characterization. Surface area was determined by Brunauer–Emmett–Teller (BET) method using Micromeritics ASAP-2020 and the samples were degassed at 150 °C for several hours.

2.4. Electrochemical characterization

Electrochemical tests were performed using a three-electrode cell, in which Ni-grid and saturated calomel electrode (SCE) were used as the counter and the reference electrode, respectively. The working electrodes were comprised of 80 wt% of active material, 10 wt% of carbon black and 10 wt% of polyvinylidene difluoride (PVDF). N-Methyl-2-pyrrolidone (NMP) was used as a solvent. The mixed slurries were coated onto Ni foils and heated at 80 °C overnight to remove the organic solvent. Then these foils were pressed onto Ni-grids at a pressure of 10 MPa. The electrolyte was 1 mol L⁻¹ LiNO₃ solution. Cyclic voltammetry (CV) and galvanostatic charge–discharge (GCD) were used to obtain the electrochemical characteristics. CV data were collected between −0.5 and 0.6 V at different scan rates (5–100 mV s⁻¹) and GCD tests were performed in the potential range of −0.6 to 0.8 V at a current density of 0.2–20 A g⁻¹. To improve the cycle performance, the organic electrolyte propylene carbonate and lithium perchlorate were used.

The specific capacitance (C, F g⁻¹) and energy density (E, W h kg⁻¹) of the active material in the electrode determined using charge–discharge curves can be calculated from the following equations:

where C (F g⁻¹) is the specific capacitance, I (A) is the discharge current, Δt (s) is the discharge time, m (g) is the mass of the active material in the working electrode, ΔV (V) represents the potential drop during the discharge process, and E (W h kg⁻¹) is the energy density. The power density P (W kg⁻¹) can be calculated from the energy density E and the discharge time Δt according to the following equation:

3. Results and discussion

3.1. Characterization of vanadium dioxides/carbon composites

Under the hydrothermal condition, V₂O₅ can be reduced by glucose to form low valance vanadium oxides, and the excess glucose and product can form organic carbon.^43,44 In this work, vanadium dioxides/carbon composites were synthesized using commercial V₂O₅ and glucose solution. Their composition and morphology were characterized by EA, EDS, XRD, FTIR, Raman and FE-SEM, as shown in Fig. 1. The EDS spectrum (Fig. 1b) shows that the precursor consists of three elements V, O and C. The EA reveals that the product contains 20.24 wt% of C and 1.668 wt% of H. All the diffraction peaks from XRD patterns (Fig. 1a) are indexed to the monoclinic crystalline phase of VO₂(B) (JCPDS, no. 34-1438), and some XRD peaks of VO₂(B) are not seen. The above analyses indicate the sample consists of the crystal VO₂(B) phase and the amorphous carbon phase.^43–46 The amorphous phase is not observed in Fig. 1a, and the reason is that usually the amorphous carbon prepared by the hydrothermal route cannot be detected by XRD even with higher amounts.⁴⁴ FTIR (Fig. 1c) and Raman (Fig. 1d) tests were carried out to give structure information about vanadium dioxides/carbon composites. As shown in Fig. 1c, the peaks at 3500–3000 cm⁻¹, 2950–2850 cm⁻¹, 1705 cm⁻¹, 1615 cm⁻¹ and 1405 cm⁻¹ are the characteristic –OH, C–H, C=O, C=C and C–H vibrations, which is corresponding with the concept of aromatization of glucose during hydrothermal reaction.⁴⁷ In addition, the peaks at 992, 804 and 539 cm⁻¹ correspond to the vibrational bands characteristic of V⁴⁺=O, V–O and V–O–V bonds.⁴⁸ The Raman spectrum (Fig. 1d) shows two peaks at 1384 cm⁻¹ (D-band) and 1590 cm⁻¹ (G-band), in agreement with the amorphous carbon.⁴⁴ The Raman spectrum also displays some Raman bands in the range of 0–1100 cm⁻¹ due to the various vibrations of V–O type.⁴⁴ Fig. 1e–h show the FE-SEM and TEM images of vanadium dioxides/carbon composites, which reveal that the sample consists of nanobelts and carbon spheres. From Fig. 1g and h, amorphous carbon coating on the surface of the nanobelts (core–shell structures) can even be observed in agreement with the previous report.⁴³ Therefore, vanadium dioxides/carbon composites were successfully synthesized.

Fig. 1 The characterizations of vanadium dioxides/carbon composites: (a) XRD pattern; (b) EDS; (c) FTIR; (d) Raman; (e) FE-SEM; (f–h) TEM.

3.2. Characterization of V₂O₅ nanoparticles

The composition and structure of the typical sample obtained at 400 °C for 4 h were characterized by XRD, Raman and IR tests. Fig. 2 shows the XRD pattern, all the diffraction peaks are indexed as the orthorhombic crystalline phase of V₂O₅ (space group: Pmmn 59) with a lattice constants values of a = 11.516 Å, b = 3.5656 Å, and c = 4.372 Å (JCPDS, no. 41-1426).⁴⁹ No obvious peaks of any other impurities, such as V₆O₁₃, V₃O₇, VO₂ and V₂O₃, are detected, indicating the as-obtained V₂O₅ with high purity. Furthermore, the EA shows that no carbon is detected in V₂O₅.

Fig. 2 XRD patterns of the as-obtained V₂O₅.

Fig. 3 displays the Raman spectrum and the crystal structure of V₂O₅, which shows that the as-prepared V₂O₅ has prominent Raman vibrational frequencies in the range of 100–1100 cm⁻¹ and these vibrational modes are characteristic of orthorhombic V₂O₅.^50–54 The vibrational peaks at 101.2, 144.6, 195.7 and 302.7 cm⁻¹ are assigned to A_g, B_3g, A_g and A_g modes owing to the stretching vibration mode of (V₂O₂)_n which corresponds to the chain translation. Their detection is strongly associated with [VO₅] units in layered structure (the oxide-layered structure). Particularly, the band at 144.6 cm⁻¹ is very sharp, strong and dominant band, indicating that the crystalline V₂O₅ has a very long-range order.⁵² Two bands observed at 282.5 (B_2g mode) and 404.9 cm⁻¹ (A_g mode) are respectively assigned to the bending vibration of the O₃–V=O and V–O₃–V bonds (Fig. 3b). The band at 481.1 cm⁻¹ (A_g mode) is ascribed to the bending vibration of the bridging V–O₂–V (doubly coordinated oxygen). The band located at 526.4 cm⁻¹ (A_g mode) corresponds to the triply coordinated oxygen (V₃–O) stretching mode. The band at 699.0 cm⁻¹ (B_2g mode) is assigned to the doubly coordinated oxygen (V₂–O) stretching mode. The band at 995.0 cm⁻¹ (A_g mode) is indexed to the stretching vibration mode of the double bond (V=O) involving the terminal unshared O oxygen. The IR spectrum (insert in Fig. 3) reveals that the absorption peaks ranging from 1200–400 cm⁻¹ are assigned to the characteristic of V–O vibration band.^{11,21,48,55,56} The band at 1020 cm⁻¹ is indexed to the symmetric stretching vibration of V⁵⁺=O bond and this band is the characteristic structure of the layered orthorhombic V₂O₅. The absorption peak located at 834 cm⁻¹ is assigned to the vibration of O–(V)₃. The absorption peaks ranging from 650 to 450 cm⁻¹ are attributed to the asymmetric and the symmetric stretching modes of V–O–V bridging bonds. The results of Raman and IR analyses are well corresponding with XRD result.

Fig. 3 (a) Raman spectrum of the as-obtained V₂O₅, insert a IR spectrum; (b) the schematic diagram of the crystal structure of V₂O₅.

The morphology and size of the as-obtained V₂O₅ were observed by FE-SEM and TEM, as shown in Fig. 4. FE-SEM images (Fig. 4a and b) show the as-prepared V₂O₅ dominantly consists of nanoparticles and these nanoparticles are mainly composed of short nanobelts, which are consistent with the morphology of the precursor of vanadium dioxides@carbon composites,^43,44 as shown in Fig. 1e. Besides, the carbon spheres are not observed in Fig. 4a and b, which reveals that these carbon spheres reacts with O₂ to form CO₂ and H₂O gas. TEM images (Fig. 4c and d) confirm the results of FE-SEM and provide some other information. It can be clearly observed that the short nanobelts are broken and the surface of V₂O₅ is very rough. From higher magnified TEM image (Fig. 4d), some holes in the particles are even seen. During the calcining process, the amorphous carbon reacts with O₂ in the air to release gas, which result in remaining vanadium oxides with broken and rough structures. The reason could also be owing to the growth, transformation or aggregation by surface and boundary diffusion during the calcining process.⁵⁷ The above characteristic of as-obtained V₂O₅ suggests that it is has high capacitance used as a supercapacitor electrode.¹¹ The high-resolution TEM (HRTEM) image (Fig. 4e) shows that the as-prepared V₂O₅ is crystallized without the presence of dislocations and defects, and it reveals that the distance between the neighboring planes is 0.341 nm, which is very consistent with the (110) plane of orthorhombic V₂O₅ in the XRD patterns (Fig. 2).

Fig. 4 The morphology of the as-obtained V₂O₅: (a and b) FE-SEM images; (c and d) TEM images; (e) HRTEM image.

The specific surface area and the pore volume of the as-prepared V₂O₅ were measured by N₂ adsorption desorption isotherms, as shown in Fig. 5. The N₂ adsorption–desorption isotherms show a sharp capillary condensation step at high relative pressure and belong to type IV isotherm, characteristic of the presence of mesoporous material according to IUPAC classification.³⁵ The BET results reveal that the specific surface area of the as-prepared V₂O₅ is 32 m² g⁻¹, which is much larger than the value of the commercial V₂O₅ (4.9 m² g⁻¹) reported before.³⁵ The pore size-distribution data calculated with the BJH method from the desorption branch of the nitrogen sorption measurements at 77 K is shown in Fig. 5b. The average pore size of the sample is 47.5 nm, which is in the range of the mesoporous material. The measured pore volume is 0.82 cm³ g⁻¹. It is noted that the pore diameter of the as-prepared V₂O₅ is a wide range owing to the amorphous carbon reacting with O₂ in the air to release gas and calcination, in agreement with the TEM observation discussed in Fig. 4.

Fig. 5 (a) Nitrogen adsorption–desorption isotherms of the as-obtained V₂O₅, insert a curve calculating the BET surface area; (b) pore size-distribution data calculated with the BJH method from the desorption branch of the nitrogen sorption measurements at 77 K.

Fig. 6 CV curves of the as-obtained V₂O₅ at different scan rates.

3.3. Electrochemical properties V₂O₅ nanoparticles in aqueous electrolyte

3.3.1. The pseudocapacitive behavior of the typical V₂O₅ nanoparticles. To display the merits of the as-obtained V₂O₅ nanoparticles, their electrochemical properties as a supercapacitor electrode were studied by CV and GCD methods in a three electrode cell. Fig. 6 shows the CV curves of V₂O₅ nanoparticles recorded at different scan rates from 5 to 100 mV s⁻¹ within a potential window of −0.5 to 0.6 V in 1 mol L⁻¹ LiNO₃ electrolyte. Each CV curve shows two pairs of redox peaks. For examples, at the scan rate of 10 mV s⁻¹ (insert in Fig. 6), two anodic peaks are respectively located at 0.023 and 0.250 V in the positive curve; and their corresponding anodic peaks are located at −0.142 and 0.054 V in the negative curve, respectively. The above characteristic demonstrates the superior faradaic reactions at the V₂O₅ electrode surface and it chemically store charges by redox. The electrochemical redox reaction can be expressed as follows:⁵⁸

The changes of V₂O₅ crystal phases occurred along with the whole process, which were generally designated as α (0 < x < 0.1), ε (0.35 < x < 0.5) and δ (0.9 < x < 1).^17,58

At different scan rates, the CV curves mainly have similar shape as depicted in Fig. 6, which reveals an excellent pseudo-capacitive response of the as-obtained V₂O₅ electrode material. With the scan rate increasing, the response current of the electrode also increases. It demonstrates the excellent kinetics and reversibility of the V₂O₅ electrode.¹¹ Particularly, because of the polarization effect of the V₂O₅ electrode,¹¹ the two oxidation peaks shift positively, and the two reduction peaks shift negatively. The redox peaks still exist even at high scan rate (100 mV s⁻¹) as shown in Fig. 6, which reveals that the as-obtained V₂O₅ material has good rate capability.³⁵

Fig. 7a exhibits the discharge curves of the as-obtained V₂O₅ nanoparticles as a supercapacitor electrode performed at various current densities 0.2–20 A g⁻¹. According to eqn (1), the specific capacitance of the as-obtained V₂O₅ nanoparticles are 406, 400, 385, 370, 343, 296 an 246 F g⁻¹ at the current density of 0.2, 0.5, 1, 2, 5, 10 and 20 A g⁻¹, respectively. The current is higher, the specific capacity is much lower. The reason could be that the insufficient active material is involved in the redox reaction and the incremental voltage drop at a higher current density.^44,59 Furthermore, a low utilization rate of active material during the process of charge–discharge at a high current density is another reason.¹ The attenuation of specific capacitance reduces slowly from the low current density (0.2 A g⁻¹) to the high current density (20 A g⁻¹). 64% of the capacitance value (20 A g⁻¹) can remain compared with the value (1 A g⁻¹). The result reveals that the as-obtained V₂O₅ nanoparticles have good rate capability in agreement with the CV curves (Fig. 6), which is attributed to their high specific surface area and porous structure (Fig. 4 and 5). Table 1 summarizes the electrochemical performance of V₂O₅ nanoparticles in the present study and V₂O₅ materials reported in some important literatures. The specific capacitance of the as-obtained V₂O₅ nanoparticles in this work has better capacity than some other V₂O₅ materials. However, the specific capacitance is slightly lower than hollow V₂O₅ microspheres (479 F g⁻¹),¹⁷ and the reason may be that the specific surface area of V₂O₅ materials in these work is bigger than our present work. Fig. 7b depicts a Ragone plot of the as-obtained V₂O₅ nanoparticles. According to eqn (2) and (3), the calculated E at the current density of 0.2, 0.5, 1, 2, 5, 10 and 20 A g⁻¹ is 245.7, 242.0, 231.7, 223.9, 207.5, 179.1 and 148.8 W h kg⁻¹ and the corresponding P is 0.396, 0.989, 1.980, 3.964, 9.895, 19.776 and 39.688 kW kg⁻¹, respectively, which exhibits the outstanding electrochemical performance of the as-obtained V₂O₅ nanoparticles.

Fig. 7 Galvanostatic discharge curves at different current densities (a) and Ragone plot (b) of the as-obtained V₂O₅.

Table 1 Comparison of specific capacitance of the as-obtained V₂O₅ nanoparticles with V₂O₅ materials reported in the previous literatures

Types of V2O5 material	Electrolyte	Potential range/V	Current densities/A g^−1	Specific capacitance/F g^−1	Reference
a The specific capacitance was obtained by the CV curve.
V2O5·0.6H2O nanoribbons	0.5 mol L^−1 K2SO4	0–1	—	181	36
V2O5 nanobelts, nanoparticles, microspheres	1 mol L^−1 LiNO3	−0.4 to 0.8	1	140, 276, 308	11
Electrospun V2O5 nanofibers	2 mol L^−1 KCl	0–0.9 V	0.1	190	15
Nano porous V2O5	2 mol L^−1 KCl	−0.2 to 0.8	5^a mV s^−1	214	37
V2O5 powders	2 mol L^−1 KCl	−0.2 to 0.7	5^a mV s^−1	262	13
Interconnected V2O5 nanoporous network	0.5 mol L^−1 K2SO4	0.2–0.8 V	0.1	304	33
V2O5 nanowires	1 mol L^−1 LiNO3	−0.4 to 0.8	1	351	35
Hollow spherical V2O5	5 mol L^−1 LiNO3	−0.2 to 0.8	5^a mV s^−1	479	17
V2O5 nanoparticles	1 mol L^−1 LiNO3	−0.5 to 0.6	0.2	406	This work
V2O5 nanoparticles	1 mol L^−1 LiNO3	−0.5 to 0.6	1	385	This work
V2O5 nanoparticles	1 mol L^−1 LiNO3	−0.3 to 0.8	1	380	This work

---

3.3.2. The influence of synthetic parameters on the specific capacitance. To know the influence of the synthetic conditions on the electrochemical properties of the as-obtained V₂O₅ nanoparticles, their corresponding GCD tests were performed. It was found that the calcined temperature and time had significant influence on the specific capacitance, phase and morphology of the final products. Fig. 8 shows the influence of the calcined temperature for 4 h. The discharge curves (Fig. 8A) reveal that the specific capacitance are 151, 385 and 367 F g⁻¹ with the calcination at 300, 400 and 500 °C, respectively. When the calcined temperature is 300 °C, its specific capacitance is only 151 F g⁻¹. The reason is that at the low temperature, main phase of VO₂(B) is prepared (Fig. 8B) and FE-SEM image (Fig. 8C) shows that it still contains some carbon spheres. The above results demonstrate that the precursor vanadium dioxides/carbon composites cannot be converted to V₂O₅ at 300 °C. With the temperature increasing to 400 and 500 °C, V₂O₅ nanoparticles with high specific capacitance (Fig. 8) were obtained.

Fig. 8 The influence of the calcined temperature for 4 h: (A) discharge curves at the current density of 1 A g⁻¹; (B) XRD patterns; (C) FE-SEM image at 300 °C; (D) FE-SEM image at 500 °C.

Fig. 9 depicts the influence of the calcined time at 400 °C. Fig. 9A displays the specific capacitance are 375, 385, 371 and 342 F g⁻¹ with the calcination time for 2, 4, 8 and 12 h, respectively, which indicates that when the reaction time is 2–8 h, the as-obtained products process high specific capacitance. However, the specific capacitance of V₂O₅ nanoparticles obtained at 400 °C for 12 h is comparatively low. XRD patterns (Fig. 9B) show the phase of the samples prepared for these times is pure V₂O₅. FE-SEM image (Fig. 9C) reveals that V₂O₅ obtained at 400 °C for 8 h has similar morphology with the sample synthesized at 400 °C for 12 h (Fig. 4), nevertheless the morphology of V₂O₅ obtained at 400 °C for 12 h (Fig. 9D) becomes particles completely and the short nanobelts are disappeared due to the growth, transformation or aggregation by surface and boundary diffusion with longer calcining time. The above results may be the reason that V₂O₅ obtained at 400 °C for 12 h has lower specific capacitance than the samples prepared with short calcining time. Further information about the electronic structure of the samples were provided by XPS, as shown in Fig. 10. The binding energies obtained in the XPS analysis were corrected for specimen charging by referencing C_1s to 284.81 eV.⁶⁰ The survey spectra (Fig. 10a) indicate that there are only two elements V and O in these four samples, in agreement with XRD observation. These samples have similar V_2p peak. Taking the sample obtained at 4 h for example, the V_2p peak (Fig. 10c) splits off two peaks, including V_2p1/2 and V_2p3/2. It is clearly observed that the sample consists of V⁵⁺ (V_2p1/2: 525.2 eV and V_2p3/2: 517.7 eV) and V⁴⁺ (V_2p1/2: 524.4 eV and V_2p3/2: 516.5 eV).^61–63 As depicted in Fig. 10f, the ratio of V⁵⁺/V⁴⁺ increases with the increasing calcination time, and these values are 12.04, 12.85, 13.76 and 17.58 for the samples obtained at 2, 4, 8 and 12 h, respectively. Therefore, the lower ratio of V⁵⁺/V⁴⁺ is benefit for the improvement of the specific capacitance of V₂O₅ in the present work.

Fig. 9 The influence of the calcined time at 400 °C: (A) discharge curves at the current density of 1 A g⁻¹; (B) XRD patterns; (C) FE-SEM image for 8 h; (D) FE-SEM image for 12 h.

Fig. 10 XPS spectra of the sample obtained at different calcined times: (a) survey spectra; (b–e) V_2p peaks; (f) the ratio of V⁵⁺/V⁴⁺ acquired by XPS.

3.3.3. The cycling performance. Due to their best electrochemical property of all the samples, we chose V₂O₅ nanomaterials obtained at 400 °C for 4 h to investigate the cycling performance, which was evaluated by CV and GCD tests, as shown in Fig. 11. The cycling behavior of V₂O₅ nanoparticles electrodes recorded by CV at 20 mV s⁻¹ scan rate (Fig. 11a) intuitively demonstrates that the specific capacitance quickly decreases with the cycles increasing. Fig. 11b–d provides the results of GCD test at 1 A g⁻¹. The specific capacitances of the as-obtained V₂O₅ nanoparticles are 385, 348, 313, 248, 123 and 53 F g⁻¹ in the 1st, 5th, 10th, 20th, 50th and 100th cycle, respectively. Their corresponding retention in the 5th, 10th, 20th, 50th and 100th cycle are 90.4, 81.3, 64.4, 31.9 and 13.8% of the initial discharge capacity. It was reported that this decrease is related with phase conversion and dissolution of the V₂O₅. As shown in Fig. 11, both CV and GCD results reveal that the redox parks and the characteristic plateaus are obviously observed in all the cycles, which suggests that the phase transition is completely reversible with increasing numbers of cycles and it cannot be the reason for the fast fading of the specific capacitance. Therefore, the dissolution of the electrode material V₂O₅ nanoparticles in aqueous LiNO₃ solution during constant charge/discharge process is the reason. There are two evidences to support the above analyses. First, the electrolyte solution gradually turns yellow color, which is the color of V₂O₅ solution. This phenomenon is a common phenomenon for vanadium oxides used as electrodes for supercapacitors and Li-ion batteries.³⁸ Second, FE-SEM images of the working electrodes before and after the cycles can provide another evidence, as shown in Fig. 12, which can clearly observe that the active materials in the working electrodes are much less after the cycles.

Fig. 11 Cycling behavior of the as-obtained V₂O₅ nanoparticles: (a) partial CV curves at 20 mV s⁻¹; (b) partial GCD curves at 1 A g⁻¹; (c) cyclic stability of the specific capacitance; (d) discharge curves at the 1st and 100th cycle.

Fig. 12 FE-SEM images of the working electrodes V₂O₅ nanoparticles before (a) and after (b) the cycles.

3.4. Improvement of the cycling performance of V₂O₅ nanoparticles in organic electrolyte

Because the dissolution of V₂O₅ leads to the poor cycle performance, the organic electrolyte propylene carbonate was chosen to improve its cycle performance and it was found the cycle performance can be greatly improved, as shown in Fig. 13. The CV curve (Fig. 13a) shows two pairs of redox peak, which is consistent with that in aqueous electrolyte (Fig. 6), demonstrating an excellent pseudo-capacitive response. Fig. 13b–d provides the results of GCD test at 1 A g⁻¹. The initial specific capacitances of the as-obtained V₂O₅ nanoparticles is 380 F g⁻¹. The specific capacitance is essentially constant after 100 cycles (Fig. 13d), revealing the cycle performance is significantly improved. This results also provide a side evidence of the dissolution of V₂O₅ in aqueous electrolyte. Fig. 14 shows the cycling behavior of the as-obtained V₂O₅ nanoparticles at various current densities, which indicate they have good rate capability. The specific capacitance is almost constant after 350 cycles compared with the value in the first 100 cycles at 1 A g⁻¹.

Fig. 13 Cycling behavior of the as-obtained V₂O₅ nanoparticles: (a) CV curve at 20 mV s⁻¹; (b) partial GCD curves at 1 A g⁻¹; (c) discharge curves at the 1st and 100th cycle; (d) cyclic stability of the specific capacitance, insert a comparison.

Fig. 14 Cycling behavior of the as-obtained V₂O₅ nanoparticles at various current densities.

4. Conclusion

In summary, vanadium dioxides/carbon composites were successfully synthesized using glucose as the carbon sources by a facile one-step hydrothermal route, and they consisted vanadium dioxides@carbon core–shell structured nanobelts and amorphous carbon spheres. Then vanadium dioxides/carbon composites were converted to surface-uneven V₂O₅ nanoparticles by the calcination in air atmospheres. The composition, morphology and structure of the samples were characterized by XRD, Raman, IR, BET, FE-SEM and TEM. V₂O₅ nanoparticles possess broken, rough and poral structures owing to the combustion of amorphous in the air during the calcination process. Surface-uneven V₂O₅ nanoparticles exhibit the specific capacitance of 406 F g⁻¹ at the current density of 0.2 A g⁻¹ and retain 246 F g⁻¹ even at high current density of 10 A g⁻¹. The influence of the calcined temperature and time on the specific capacitance, phase and morphology of the products were discussed in detail, and the results revealed that the calcination at 400 °C for 4 h with comparatively low ratio of V⁵⁺/V⁴⁺ are favorable for surface-uneven V₂O₅ nanoparticles with the high electrochemical property. The novel morphology and high specific surface area are the main factors which contribute to high electrochemical performance to surface-uneven V₂O₅ nanoparticles. During the cycle performance, the specific capacitances of V₂O₅ nanoparticles after 100 cycles are 13.8% and 98.5% of the initial discharge capacity in the aqueous and organic electrolyte, respectively, indicating the cycle performance is significantly improved in organic electrolyte. The reason of the specific capacitance of V₂O₅ nanoparticles electrodes quickly fading with the increasing cycles is the dissolution of electrode material in aqueous electrolyte. The significance of this work is that it provides more insights about the effect of structures on the electrochemical properties of vanadium oxides with novel structures and their cycle performance can be significantly improved in organic electrolyte.

Acknowledgements

This work was partially supported by the National Natural Science Foundation of China (Grant No. 21601026, 21271037), the Fundamental Research Funds for the Central Universities (DUT16LK37) and Science research project of Liaoning Province Education Department (L2015123).

References

1. Z. Yu, L. Tetard, L. Zhai and J. Thomas, Energy Environ. Sci., 2015, 8, 702–730.
2. M. Winter and R. J. Brodd, Chem. Rev., 2004, 104, 4245–4270.
3. X.-J. Ma, W.-B. Zhang, L.-B. Kong, Y.-C. Luo and L. Kang, RSC Adv., 2015, 5, 97239–97247.
4. L. Cao, J. Zhu, Y. Li, P. Xiao, Y. Zhang, S. Zhang and S. Yang, J. Mater. Chem. A, 2014, 2, 13136–13142.
5. B. N. Reddy, R. Mukkabla, M. Deepa and P. Ghosal, RSC Adv., 2015, 5, 31422–31433.
6. A. Venkatesan, N. R. Krishna Chandar, A. Kandasamy, M. Karl Chinnu, K. N. Marimuthu, R. Mohan Kumar and R. Jayavel, RSC Adv., 2015, 5, 21778–21785.
7. M. Lee, S. K. Balasingam, H. Y. Jeong, W. G. Hong, H.-B.-R. Lee, B. H. Kim and Y. Jun, Sci. Rep., 2015, 5, 8151.
8. Y. Zhang and C. Meng, J. Alloys Compd., 2016, 674, 259–265.
9. F. Wang, S. Xiao, Y. Hou, C. Hu, L. Liu and Y. Wu, RSC Adv., 2013, 3, 13059–13084.
10. X. Zhou, Q. Chen, A. Wang, J. Xu, S. Wu and J. Shen, ACS Appl. Mater. Interfaces, 2016, 8, 3776–3783.
11. Y. Zhang, J. Zheng, Y. Zhao, T. Hu, Z. Gao and C. Meng, Appl. Surf. Sci., 2016, 377, 385–393.
12. L. L. Zhang and X. S. Zhao, Chem. Soc. Rev., 2009, 38, 2520–2531.
13. Z. J. Lao, K. Konstantinov, Y. Tournaire, S. H. Ng, G. X. Wang and H. K. Liu, J. Power Sources, 2006, 162, 1451–1454.
14. D. Yang, J. Power Sources, 2013, 228, 89–96.
15. G. Wee, H. Z. Soh, Y. L. Cheah, S. G. Mhaisalkar and M. Srinivasan, J. Mater. Chem., 2010, 20, 6720–6725.
16. X. Xia, Y. Zhang, D. Chao, C. Guan, Y. Zhang, L. Li, X. Ge, I. M. Bacho, J. Tu and H. J. Fan, Nanoscale, 2014, 6, 5008–5048.
17. J. Yang, T. Lan, J. Liu, Y. Song and M. Wei, Electrochim. Acta, 2013, 105, 489–495.
18. T. Qian, N. Xu, J. Zhou, T. Yang, X. Liu, X. Shen, J. Liang and C. Yan, J. Mater. Chem. A, 2015, 3, 488–493.
19. Y. Wu, G. Gao and G. Wu, J. Mater. Chem. A, 2015, 3, 1828–1832.
20. Y. Wu, G. Gao, H. Yang, W. Bi, X. Liang, Y. Zhang, G. Zhang and G. Wu, J. Mater. Chem. A, 2015, 3, 15692–15699.
21. Y. Zhang, C. Chen, W. Wu, F. Niu, X. Liu, Y. Zhong, Y. Cao, X. Liu and C. Huang, Ceram. Int., 2013, 39, 129–141.
22. Y. Zhang, M. Zhou, M. Fan, C. Huang, C. Chen, Y. Cao, H. Li and X. Liu, Curr. Appl. Phys., 2011, 11, 1159–1163.
23. Y. Zhang, M. Fan, X. Liu, C. Huang and H. Li, Eur. J. Inorg. Chem., 2012, 2012, 1650–1659.
24. I. E. Rauda, V. Augustyn, L. C. Saldarriaga-Lopez, X. Chen, L. T. Schelhas, G. W. Rubloff, B. Dunn and S. H. Tolbert, Adv. Funct. Mater., 2014, 24, 6717–6728.
25. D. Vernardou, M. Apostolopoulou, D. Louloudakis, N. Katsarakis and E. Koudoumas, Chem. Vap. Deposition, 2015, 21, 369–374.
26. D. Vernardou, D. Louloudakis, N. Katsarakis, E. Koudoumas, I. I. Kazadojev, S. O'Brien, M. E. Pemble and I. M. Povey, Sol. Energy Mater. Sol. Cells, 2015, 143, 601–605.
27. T. D. Manning, I. P. Parkin, R. J. H. Clark, D. Sheel, M. E. Pemble and D. Vernadou, J. Mater. Chem., 2002, 12, 2936–2939.
28. T. D. Manning, I. P. Parkin, M. E. Pemble, D. Sheel and D. Vernardou, Chem. Mater., 2004, 16, 744–749.
29. D. Vernardou, M. E. Pemble and D. W. Sheel, Surf. Coat. Technol., 2004, 188–189, 250–254.
30. D. Vernardou, D. Louloudakis, E. Spanakis, N. Katsarakis and E. Koudoumas, Sol. Energy Mater. Sol. Cells, 2014, 128, 36–40.
31. J. Zhu, L. Cao, Y. Wu, Y. Gong, Z. Liu, H. E. Hoster, Y. Zhang, S. Zhang, S. Yang, Q. Yan, P. M. Ajayan and R. Vajtai, Nano Lett., 2013, 13, 5408–5413.
32. B.-H. Kim, C. H. Kim, K. S. Yang, A. Rahy and D. J. Yang, Electrochim. Acta, 2012, 83, 335–340.
33. B. Saravanakumar, K. K. Purushothaman and G. Muralidharan, ACS Appl. Mater. Interfaces, 2012, 4, 4484–4490.
34. T. Peng, J. Wang, Q. Liu, J. Liu and P. Wang, CrystEngComm, 2015, 17, 1673–1679.
35. N. Wang, Y. Zhang, T. Hu, Y. Zhao and C. Meng, Curr. Appl. Phys., 2015, 15, 493–498.
36. Q. T. Qu, Y. Shi, L. L. Li, W. L. Guo, Y. P. Wu, H. P. Zhang, S. Y. Guan and R. Holze, Electrochem. Commun., 2009, 11, 1325–1328.
37. R. N. Reddy and R. G. Reddy, J. Power Sources, 2006, 156, 700–704.
38. H.-Y. Li, C. Wei, L. Wang, Q.-S. Zuo, X. Li and B. Xie, J. Mater. Chem. A, 2015, 3, 22892–22901.
39. M.-H. Bai, T.-Y. Liu, F. Luan, Y. Li and X.-X. Liu, J. Mater. Chem. A, 2014, 2, 10882–10888.
40. S.-Z. Huang, Y. Cai, J. Jin, Y. Li, X.-F. Zheng, H.-E. Wang, M. Wu, L.-H. Chen and B.-L. Su, J. Mater. Chem. A, 2014, 2, 14099–14108.
41. S. D. Perera, B. Patel, J. Bonso, M. Grunewald, J. P. Ferraris and K. J. Balkus, ACS Appl. Mater. Interfaces, 2011, 3, 4512–4517.
42. G. Yilmaz, C. X. Guo and X. Lu, ChemElectroChem, 2016, 3, 158–164.
43. Y. Zhang, J. Zhang, Y. Zhong, L. Yu, Y. Deng, C. Huang and X. Liu, Appl. Surf. Sci., 2012, 263, 124–131.
44. Y. Zhang, J. Zheng, T. Hu, F. Tian and C. Meng, Appl. Surf. Sci., 2016, 371, 189–195.
45. Y. Zhang, X. Zhang, Y. Huang, M. Xiao, F. Niu, C. Huang, C. Meng and W. Wu, Chem. Lett., 2014, 43, 337–339.
46. Y. Zhang, X. Liu, D. Chen, L. Yu, J. Nie, S. Yi, H. Li and C. Huang, J. Alloys Compd., 2011, 509, L69–L73.
47. X. Sun and Y. Li, Angew. Chem., Int. Ed., 2004, 43, 597–601.
48. T. R. Gilson, O. F. Bizri and N. Cheetham, J. Chem. Soc., Dalton Trans., 1973, 291–294,  10.1039/dt9730000291.
49. R. Enjalbert and J. Galy, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1986, 42, 1467–1469.
50. R. Baddour-Hadjean, M. B. Smirnov, K. S. Smirnov, V. Y. Kazimirov, J. M. Gallardo-Amores, U. Amador, M. E. Arroyo-de Dompablo and J. P. Pereira-Ramos, Inorg. Chem., 2012, 51, 3194–3201.
51. G. A. Horrocks, M. F. Likely, J. M. Velazquez and S. Banerjee, J. Mater. Chem. A, 2013, 1, 15265–15277.
52. R. Baddour-Hadjean, V. Golabkan, J. P. Pereira-Ramos, A. Mantoux and D. Lincot, J. Raman Spectrosc., 2002, 33, 631–638.
53. R. Baddour-Hadjean, A. Marzouk and J. P. Pereira-Ramos, J. Raman Spectrosc., 2012, 43, 153–160.
54. G. Faggio, V. Modafferi, G. Panzera, D. Alfieri and S. Santangelo, J. Raman Spectrosc., 2012, 43, 761–768.
55. B. Saravanakumar, K. K. Purushothaman and G. Muralidharan, J. Electroanal. Chem., 2015, 758, 111–116.
56. W. Chen, L. Mai, J. Peng, Q. Xu and Q. Zhu, J. Mater. Sci., 2004, 39, 2625–2627.
57. Y. Zhang, M. Fan, F. Niu, W. Wu, C. Huang, X. Liu, H. Li and X. Liu, Curr. Appl. Phys., 2012, 12, 875–879.
58. C. Delmas, H. Cognac-Auradou, J. M. Cocciantelli, M. Menetrier and J. P. Doumerc, Solid State Ionics, 1994, 69, 257–264.
59. L. Deng, G. Zhang, L. Kang, Z. Lei, C. Liu and Z.-H. Liu, Electrochim. Acta, 2013, 112, 448–457.
60. Y. Zhang, J. Zhang, X. Zhang, S. Mo, W. Wu, F. Niu, Y. Zhong, X. Liu, C. Huang and X. Liu, J. Alloys Compd., 2013, 570, 104–113.
61. G. Silversmit, D. Depla, H. Poelman, G. B. Marin and R. D. Gryse, J. Electron Spectrosc. Relat. Phenom., 2004, 135, 167–175.
62. C. D. Wagner, W. M. Riggs, L. E. Davis and J. F. Moulder, Handbook of X-Ray Photoelectron Spectroscopy, Perkin-Elmer Corporation, Minnesota, 1979.
63. T. D. Manning, I. P. Parkin, C. Blackman and U. Qureshi, J. Mater. Chem., 2005, 15, 4560–4566.

---