High-performance asymmetric supercapacitor based on vanadium dioxide and carbonized iron-polyaniline electrodes

Supercapacitors (SCs), also called electrochemical capacitors (ECs) have been on the forefront of research owing to their relatively high power density (15 kW kg^-1), low specific energy (<10 W h kg^-1) and a long lifetime.^1–6 To date, the scientific community is working towards increasing the specific energy of SCs by using ingenious device design.⁷ SCs can be classified into three types of capacitors based on their charge storage mechanism: (i) The electrical double-layer capacitors (EDLCs), where charge build-up at the boundary between the electrode and the electrolyte is responsible for the energy storage and the common materials used are carbon-based materials.^2,8 (ii) The faradaic capacitor which as their names suggest, involve redox or faradaic reaction and is mainly displayed in transition metal oxides, metal hydroxides, metal sulfides and conducting polymers.^9–13 (iii) The hybrid capacitors which are the combination of both EDLC and faradaic materials. A subclass of hybrid capacitors is the asymmetric supercapacitor (ASC) which are composed of a positive and a negative electrodes with dissimilar charge storage mechanisms. Generally, the positive electrodes are the faradaic materials^14–21 while the negative electrodes are mostly made of carbon-based materials.^22–26 The Hybrid capacitors have been proposed and considered as a promising solution to improve the low specific capacitance from carbon-based materials and the low conductivity and poor cycle stability of the transition metal oxides/hydroxides.²⁷ 

Carbon-based materials such as activated carbon,²⁸ carbon nanotube²⁹ and graphene³⁰ have been demonstrated to be a good electrode materials in supercapacitor due to their excellent conductivity combined with their good stability.³¹ 

The transition metal oxide used as supercapacitor electrode materials exhibit a high specific capacity as compared to carbon-based materials owing to its multiple oxidations states.

Amongst the low-cost metal oxides, vanadium oxides (e.g. VO₂, V₂O₅, V₂O₃, and V₄O₇) have received recent attention^32–40 which is linked to their abundant sources, and ability to exist in variable oxidation states.⁴¹ 

Vanadium dioxide (VO₂) has an exciting phase with a rich polymorphic stable and metastable forms included VO₂ (A), VO₂ (M), VO₂ (R), VO₂ (B), VO₂ (T) and VO₂ (bcc).^42,43 VO₂ (A), VO₂ (M) and VO₂ (B) are the most attractive due to their tuneable and the relatively easy synthesis process.^42,44 The VO₂ materials change from monoclinic (at a temperature about 68°C) to tetragonal structure reversibly (at a temperatures higher than the 68 °C) and can undergo semiconductor-to-metal transition.^45–50 VO₂ (B) electrodes with a metastable monoclinic structure is a potential electrode material in supercapacitor.⁵¹ 

As compared to vanadium pentoxide (V₂O₅), there are few report on vanadium dioxide for asymmetric supercapacitor. For instance, Wang et al. synthesized a graphene/VO₂ composite material for a positive and a negative electrodes. They assembled a symmetric supercapacitor (graphene/VO₂//graphene/VO₂) using 0.5 M Na₂SO₄ as an aqueous electrolyte. The graphene/VO₂//graphene/VO₂ symmetric device showed a specific energy of 21.3 W h kg⁻¹ at 1 A g^-1. The graphene/VO₂ composite showed a cycling stability with 92% after 5000th cycles at 10 A g⁻¹.⁵² Similarly Ma et al.⁴⁸ prepared a vanadium dioxide electrode using for a symmetric supercapacitor in 1 M Na₂SO₄. The VO₂//VO₂ symmetric device exhibited a specific energy of 21.3 W h kg⁻¹ corresponding to a specific power of 207.2 W kg⁻¹ at a gravimetric current of 0.25 A g⁻¹. They reported a cycling efficiency of 78.7% after 5.000 cycles at a specific current of 0.5 A g⁻¹.⁴⁸ 

In our previous study, we synthesized the vanadium dioxide monoclinic (VO₂ (B)) through a solvothermal method. In a three electrode configuration the VO₂ (B) displayed a specific capacity of 49.28 mAh g^-1 at current density of 0.5 A g^-1 in aqueous electrolyte (6 M KOH).⁵³ 

The present work reports the fabrication of a novel asymmetric supercapacitor (ASC) based on VO₂ (B) monoclinic as a positive electrode and carbonized iron-polyaniline (C-FP) as a negative electrode. The VO₂//C-FP ASC tested in aqueous electrolyte (6 M KOH) was able to reach a potential window of 1.6 V. The asymmetric device exhibited a specific energy and power of 30 W h kg⁻¹ and 713 W kg⁻¹ respectively at 1 A g⁻¹. In addition, the ASC showed an 89% energy efficiency after 10000 galvanostatic charge-discharge cycles with a large improvement after ageing test at a gravimetric current of 10 A g^-1.

Vanadium (V) oxide (V₂O₅, purity ≥98%), oxalic acid dehydrate (C₂H₂O₄·2H₂O purity 99%), iron (III) nitrate (Fe(NO₃)₃·9H₂O purity 99-100%), ammonium peroxydisulfate ((NH₄)₂S₂O₈) purity 98%), aniline hydrochloride (C₆H₅NH₂·HCl purity ≥99%) and propan-2-ol (CH₃CHOHCH₃ purity 99.5%) were purchased from Sigma-Aldrich. Potassium hydroxide (KOH) and ethanol (C₂H₆O purity 99%) were purchased from Merck (South Africa). Polycrystalline nickel foam (thickness of 1.6 mm) was obtained from Alantum (Munich, Germany).

The synthesis of the VO₂ material was carried out using solvothermal method. Initially, 1.2 g of V₂O₅ powder and 2.49 g of H₂C₂O₄·2H₂O was added to 40 mL of deionized water and stirred for 3 h, thereafter, a 6 mL of the homogeneous solution was added to 60 mL of isopropanol under continuous stirring for 20 min.

The solution was transferred into a Teflon-lined stainless steel autoclave and kept at 200 °C for 6 h. The recovered powder was washed several times with deionized water followed by ethanol and dried at 60 °C in an electric oven.⁵³ 

0.2 M aniline hydrochloride (C₆H₅NH₂·HCl) (2.59 g dissolved in 50 mL deionized water) was added to 0.25 M ammonium peroxydisulfate (NH₄)₂S₂O₈) (5.71 g in 50 mL deionized water) and mixed overnight.

Briefly, 0.2 g of Fe(NO₃)₃·9H₂O and 0.0125 g of PANI were dissolved in 50 ml of ethanol and sonicated in the ultra-sonication bath.

After ethanol was almost completely evaporated, the mixture was coated on a nickel (Ni) foam acting as a current collector and pyrolyzed for 2 h under the N₂ atmosphere at 850 °C. The full detailed description of the C-FP can be found in our previous paper.⁵⁴ 

The structural properties of the samples were analysed by X-ray diffraction (XRD) powder using an XPERT-PRO diffractometer (PANalytical BV, The Netherlands) with theta/2theta configuration. The morphology of the materials synthesized was characterized by a high-resolution Zeiss Ultra plus 55 field emission scanning electron microscope (FE-SEM), operated at a voltage of 2.0 kV and a JEOL JEM-2100F transmission electron microscope (TEM). The selected area electron diffraction (SAED) pattern were taken with a JEOL JEM-2100F transmission electron microscope (TEM) and were used to evaluate the elemental composition of the produced materials. The X-ray photoelectron spectroscopy (XPS, K-alpha, Thermo Fisher) was used to analyse the elemental composition of the materials with a monochromatic Al-Kα radiation.

Three - and two electrode configurations were adopted to study the electrochemical properties of the VO₂ and C-FP electrodes.

The electrochemical characterizations were carried out using a Bio-Logic VMP-300 (Knoxville TN 37,930, USA) potentiostat monitored by the EC-Lab® V10.37 software. In the three-electrode configuration, Ag/AgCl (KCl saturated) served as the reference electrode, a glassy carbon plate as the counter electrode and 6 M KOH as the electrolyte. The VO₂ electrode was prepared as follows: 85 wt% of the active material was added to 10 wt% of carbon black as conducting additive and 5 wt% of polyvinylidene difluoride (PVDF) binder in an agate mortar. Few drops of 1-methyl-2-pyrrolidinone (NMP) were added to the mixture to form a slurry, which was pasted on nickel foam (NiF) acting as a current collector and dried at 60 °C in the electric oven for overnight.

Thereafter, the asymmetric supercapacitor consisting of VO₂ positive electrode and C-FP negative electrode was coupled in a coin cell with glass micro-fiber filter paper was used as a separator.

Figure 1 display the X-ray diffraction patterns of VO₂ and C-FP materials. The diffraction peaks of the VO₂ materials are indexed as VO₂(B) monoclinic structure as shown in Fig. 1(a). It also shows that vanadium dioxide synthesized at 200 °C has a space group of C 1 2/m 1.⁵³ XRD pattern of C-FP powder material without pasting on the substrate (Ni foam) is shown in Fig. 1(b). The diffraction peaks of C-FP material are indexed to orthorhombic structures of Fe₃C and FeS with a space group of P nma.^54,55

Figure 2 present the SEM micrographs of the as-prepared VO₂ and C-FP materials at low and high magnifications.

Figure 2(a) shows the SEM micrographs of the VO₂ which reveals the nanosheets-like structure on the microspheres surface. In Fig. (2b), the micrographs of VO₂ exhibits a vertically grown sheet-like structure. Figure 2(c–d) shows SEM micrographs of the C-FP which unveiled agglomerated nanograin morphology. The micrographs of the C-FP materials were showed lattice fringes attributed to the Fe cations on PANI and have been discussed in References 51 and 54.

The morphologies and the elemental composition of VO₂ materials were further studied with transmission electron microscope (TEM) and the selected area electron diffraction (SAED) analysis.

Figure 3(a) displays the TEM micrograph of VO₂ at high magnification which reveals clearly the nanosheets structure as shown in Fig. 2(b). The SAED pattern of the VO₂(B) nanosheets in Fig. 3(b) exhibits the presence of well-defined rings, indicates the poly-crystallinity of the VO₂ monoclinic.

To further evaluate the surface characterization of VO₂ material, the X-ray photoelectron spectroscopy (XPS) was used to determine the chemistry of the material. The core level spectrum of V2p reveals two chemical states of vanadium which are related to excitations of electrons from the V2p_3/2 and V2p_1/2 core levels, respectively, as shown in Fig. 4(a).

The predominant peak located at 516.5 eV in the V2p_3/2 binding energy suggests a vanadium oxidation state of 4⁺ which confirms the formation of VO₂.⁵⁶ Futhermore, as presented in Fig. 4(b), the core level spectrum of O1s displayed the main peak located at 529.7 eV which is ascribed to the component associated to oxygen in VO₂.⁵⁷ 

To construct the asymmetric hybrid supercapacitor of VO₂//C-FP, we, firstly, evaluated the electrochemical performance of the positive VO₂ and negative C-FP electrodes in a three-electrode system using 6 M KOH electrolyte with Ni foam and Ag/AgCl (KCl saturated) as a current collector and reference electrode, respectively.

Figure 5(a) shows the cyclic voltammogram (CV) profile of the VO₂ electrode at different sweep rates (from 5 to 100 mV s^-1) within a potential window range of 0.0 - 0.5 V. The appearance of a pair of redox peaks associated with an anodic peak at ∼0.13 V and cathodic peak ∼0.23 V at 5mV s^-1 reveal a faradaic material. As observed in Fig. 5(a), these peaks are broader compared to those exhibited by battery-like material which is typically narrower and indicative of the occurrence of a redox reaction at a constant potential.⁵⁸ 

The broadness of the peak in faradaic materials is expected as a result of the presence of non-standard sites and defects in the polycrystalline structure. This agrees with the low crystallinity of the VO₂ as recorded from XRD diffraction pattern.⁵⁸ 

Figure 5(b) shows the charge-discharge (CD) of the vanadium dioxide curve at different specific currents. Each discharge curve displays a non-linear curve confirming the faradaic behavior of this electrode material.

Moreover, even at a low specific current of 1 A g^-1, the discharge profile does not show an extended plateau as is the case for batteries.⁵⁹ Figure 5(c) shows the CV curve of the C-FP electrode at different sweep rates from 5 to 100 mV s^-1 in a negative potential window range of -1.2 - 0.0 V.

These CV curves show non-rectangular shapes with no apparent redox peaks. However, Fig. 5(d) which shows the charge-discharge curves at different specific currents in the voltage window of -1.2 to 0.0 V of the C-FP electrode, depicting a non-linear charge-discharge, suggesting a pseudocapacitive activity in this electrode material.

From the chronopotentiometry profile of the VO₂ and C-FP electrodes, the specific capacity, Q (measured in mA h g^-1) of the VO₂ and C-FP electrodes was determined using:

where I_d is defined as the specific current measured in A g^-1 and $tD$ is the time in second (s) for a complete discharge cycle.

Figure 5(e) depicts the values of the specific capacity for the VO₂ and C-FP electrodes as a function of increasing specific current. The specific capacity values of 49.3 and 107 mA h g^-1 were recorded for the VO₂ and C-FP material electrodes respectively, at a gravimetric current of 0.5 A g^-1. This can be related to the thin nanosheets structure of VO₂, which will ensure faster ion and electron transport. Also, the high capacitive characteristic observed in the C-FP can be attributed to the conductive framework, which allows an excellent electric contact and consequently enhances the capacitance performance. Additionally, it can be observed that these two materials (VO₂ and C-FP) are stable in each of its potential windows.

With the aim of optimizing the performance of ASC, the device was assembled using VO₂ as positive and C-FP as negative electrodes, respectively, in 6 M KOH.

The charge equilibrium (⁠$QVO2B=Qc−FP$⁠) was used to balance the masses of both electrodes in the asymmetric cell. This generates equation 2 and 3 which were used to balance the masses:

where $mVO2B,mc−FP,QVO2B,Qc−FP$ describes the mass loading and total charge of the VO₂ and C-FP electrodes, I is given as the applied current and $tD$ is the time of discharge to 0 V. The mass ratio of the VO₂ to C-FP was adopted as 2:1 and the mass loading per unit area of the VO₂ and C-FP electrodes was recorded as 2.24 and 1.12 mg cm^-2, respectively) According to the charge balance, the mass loading of active VO₂ and C-FP on the current collector were measured as 4 and 2 mg, respectively in line with equation 3 above.

Figure 6(a) shows the CV curves of VO₂ and C-FP measured in the stable working potential window at a sweep rate of 50 mV s^-1, a working potential window of 1.7 V could be predicted for the asymmetric device.

Figure 6(b) shows the CV graphs of the VO₂//C-FP asymmetric device at different sweep rates (5 to 200 mV s^-1). However, the maximum working potential limit of the VO₂//C-FP device was recorded to be 1.6 V.

There is no apparent current leap within the operating cell potential window of 1.6 V, suggesting the stability of the device within this potential window. The CD curves of VO₂//C-F at different specific currents (1 to 10 Ag^-1) are shown in Fig. 6(c). The CD curves exhibit faradaic behavior owing to the high redox activity observed from the CV curves of the asymmetric device. The specific capacity of the VO₂//C-FP device was calculated using equation (1) and is shown in Fig. 6(d) as a function of specific current.

The specific capacity of the VO₂//C-FP device reaches a value of 47 mA h g^-1 at a gravimetric current of 1 A g^-1. This value is well positioned between those obtained for VO₂ and C-FP electrodes from the three electrode measurements, at the same specific current. In other words, the specific capacity value of the hybrid device is much higher than that of VO₂ (43.4 mA h g^-1) and lower than that of C-FP (79 mA h g^-1) calculated in the three-electrode configuration at a gravimetric current of 1 A g^-1. This shows a good synergistic improvement by combining these two materials to form a hybrid device.

Figure 6(e) displays the Ragone plot presenting the specific power versus the specific energy of the asymmetric device obtained at different specific currents. The specific energy and the specific power of the device were obtained using equations (4) and (5) respectively.⁶⁰ 

where E_d (W h kg^-1) and P_d (W kg^-1) are the total specific energy and specific power respectively. I_d is the specific current in A g^-1, t is the discharge time (s) and V is the working potential window (V) of the V0₂//C-FP device.

The maximum specific energy value of 30 Wh kg^-1 was recorded for the VO₂//C-FP device with an associated specific power value of 713 W kg^-1 at a 1 A g^-1 specific current. This is maintained at 9.1 W h kg⁻¹ for a specific power of 7.9 kW kg⁻¹ at 10 A g^-1. The high specific energy and specific power of the ASC are attributed to a high specific capacity and device wide operating voltage. This is also related to the good stability, fast kinetics of charge/discharge process⁶¹ and the high ionic conductivity of the electrolyte ions, i.e., 73.5 and 198 Scm² mol^-1 for K⁺ and OH⁻, respectively.²⁶ 

In order to study the stability of the device, it was subjected to 10000 cycles at the high gravimetric current of 10 A g^-1 and the results are shown in Fig. 7(a).

An energy efficiency of the device was calculated using equation (6)

where $ηE$⁠, $Ed$ and $Ec$ are energy efficiency, discharge energy and charge energy from the charge-discharge curve of the VO₂//C-FP device respectively.

The energy efficiency of 89% is obtained with good capacity retention of 78.5% at the 10 000th constant charge-discharge cycle, signifying good electrochemical stability of the device. The further additional stability test was performed after cycling measurement on the cell using the voltage holding test (also called floating test).⁶² It has the ability to determine a direct insight into the possible effect and degradation phenomena which might occur during the electrochemical process.⁶³ The voltage holding test was designed to analyse the device specific capacity at each 10 h period of the potential holding step for up to 70 h. This is following by three GCD to exhibit any change in the cell device with floating over a time of 70 h. The specific capacity as a function of floating time is presented in Fig. 7(b) for 70 h at 10 A g^-1.

It exhibits an increase in the specific capacity value for up to 30 h period of voltage holding time before becoming constant. The increase in the specific capacity could be linked to the evolution of accessible redox sites during the ageing experiment.

This improvement is even more striking when the specific energy was calculated after each voltage holding as shown in Fig. 7(c). Within the first 10 h of voltage holding, the specific energy increases by 32% to finally stabilize after 30 h of floating test, at 15 W h kg^-1, corresponding to an impressive increase of 65% from the original 9.1 W h kg^-1 at 10 A g^-1. It shows that the cell voltage (1.6 V) is stable using 6 M KOH. As compared to other hybrid devices our group has reported this increase is better than that of Co₃(PO₄)₂·4H₂O/GF//C-FP (2.2% from the original value of 9.1 Wh Kg⁻¹)⁶⁴ as earlier reported for hybrid asymmetric capacitors. Thus, the floating test should be considered as a viable option for optimizing the properties of this cell.

The electrochemical impedance spectroscopic (EIS) measurement of the device was performed in an open voltage from 0.01 Hz to 100 kHz frequencies. The Nyquist plot of the asymmetric device (VO₂//C-FP), before stability, after the 10.000 constant galvanostatic cycles and after voltage holding are shown in Fig. 7(d). The equivalent series resistance (ESR) value of the asymmetric device (VO₂//C-FP) was 1.55 Ω before and after 10000th cycles. However, after voltage holding, the ESR decreased to 1 Ω followed by a shorter diffusion length of the electrolyte ions. This low value of ESR confirm the good contact between the electrolyte and the surface of the electrode materials. Thus, any degradation of the cell has been not observed after the voltage holding. More explicitly, no change in the equivalent series resistance was noticed after stability. Two main changes in the impedance could explain the electrochemical improvement of the cell after voltage holding. However, the diffusion was reduced after the stability test. Upon voltage holding, the diffusion length is markedly reduced followed by reduction of the solution resistance. These reductions can significantly enhance the performance of the cell by a fast collection of charges.

Table I compares the asymmetric VO₂//C-FP device with some others devices reported in the literature. The cell shows higher values when compared with other devices.^4,48,52,65 This demonstrates the excellent choice of tandem materials for this asymmetric device.

We have successfully synthesized VO₂(B) nanosheets by a solvothermal method and C-FP material by pyrolysis of an iron-PANI mixture under nitrogen atmosphere. An ASC cell was fabricated from VO₂ adopted as positive and C-FP as the negative electrodes operated with an aqueous 6 M KOH electrolyte. The asymmetrical device exhibited a specific capacity of 47 mA h g^-1 with a high specific energy of 30 W h kg^-1 and the corresponding specific power of 713 W kg^-1 at 1 A g^-1 with 1.6 cell potential. These values are far better as compared to those studies previously published for related devices as indicated in Table I above. The excellent stability performance of the VO₂//C-FP device was demonstrated up to 10000 cycles at a specific current of 10 A g^-1. In addition, the voltage holding data obtained after testing for a period of 70 h shows a significant improvement in device specific capacity and energy after a period of 10 h at 10 A g^-1. This result confirms that the performance of the VO₂//C-FP device increase after the voltage holding test. This asymmetric supercapacitor from VO₂//C-FP exhibits impressive electrochemical performance and hence making the device excellent for energy storage applications.

This research was supported by the South African Research Chairs Initiative (SARChI) of the Department of Science and Technology and the National Research Foundation (NRF) of South Africa (Grant No. 61056). Any idea, finding, conclusion or recommendation expressed in this material is that of the author(s). The NRF does not accept any liability in this regard. N. M. Ndiaye thanks Organization for Women in Science for the Developing World (OWSD) and Swedish International Development Cooperation Agency (Sida), NRF through SARChI in Carbon Technology and Materials and the University of Pretoria for financial support.