Graphitic C₃N₄ as a powerful catalyst for all-vanadium redox flow batteries

Abstract

A novel carbon felt electrode modified with carbon nitride (C₃N₄) was developed to improve the electrochemical performance with a VO²⁺/VO₂⁺ redox pair. The graphitic C₃N₄ on carbon felt exhibited an excellent performance in an all-vanadium redox flow battery. This work, for the first time, demonstrates the electrocatalytic properties of C₃N₄ for use in VRFBs.

---

1. Introduction

Enhancement of the development of solar energy and hydropower has encouraged people to explore stationary energy storage devices. Thus they invest significant capital in redox flow batteries (RFBs), particularly in a hot field like vanadium redox flow batteries (VRFBs).¹ RFBs have peculiar advantages compared with conventional batteries using solid active materials, such as decoupling of the capacity and the power output by the designing of an electrical stack. VRFBs are well-suited for large-scale utility applications.^2–4 In addition, the energy density in the VRFB is determined by the concentration of the vanadium species and the volume of the reservoirs, and cross-contamination problems could be avoided because of the enlisting of the same element, V, in both the anodic and cathodic electrolytes.⁵ Furthermore, VRFBs have several additional advantages, such as an excellent electrochemical reversibility and high efficiency. The electrodes play an important role in a VRFB system, where the electrochemical reactions of V(IV)/V(V) and V(II)/V(III) redox couples take place. Carbon-based materials are widely used for electrodes to catalyze couple reactions.^5,6 However, these carbon materials have been proven to show poor kinetic reversibility.⁷ Therefore, a series of modification methods have been reported for enhancing the electrochemical properties of these carbon electrodes. Sun et al. reported the chemical modification of graphite felt by loading Ir particles.^8,9

Due to the high cost and limited mineral resources of Ir metal, this method becomes too unwieldy. Wu et al. used microwave-treated graphite felt as the positive electrode with an activated surface.¹⁰ Park et al. employed porous metal foams as the electrodes.¹¹ In our previous work, we have synthesized nitrogen-doped carbon nanotube/graphite felts as electrode materials for VRFBs with excellent performance.¹² However, the harsh synthesis conditions limit the wide large-scale production of VRFBs with these electrode materials. Graphitic carbon nitride (C₃N₄) with a graphite-like structure has drawn plenty of scientific interest due to its excellent chemical and thermal stability,¹³ high in-plane nitrogen content, and appealing electronic structure. C₃N₄ has been widely investigated for a range of applications such as photocatalysis, CO₂ reduction, electrocatalysis, and bio-imaging applications.^14–16 To the best of our knowledge, there are not any reports on the use of this material for VRFBs. Therefore, it is of essential interest to observe the electrocatalytic behavior of graphitic C₃N₄ for V(IV)/V(V) and V(II)/V(III) redox reactions and thus for VRFBs.

2. Experimental

2.1 Synthesis of C₃N₄ on carbon felt

In this work, for the first time, we investigate the possibility of using graphitic C₃N₄ as a novel catalyst to enhance the electrochemical activity of carbon felt electrodes in VRFBs. To prepare the graphitic C₃N₄ on carbon felt (CF), pure CF was added into the precursor solution, which was prepared by dissolving melamine at a concentration of 0.01 M (the solvent was ultrapure water, and melamine was dissolved with magnetic stirring at room temperature), followed by transferring it into a 100 mL Teflon-lined autoclave for reaction in an oven at 80 °C for 16 hours. The obtained hydrothermal products were washed with deionized water and dried at 60 °C. Finally, the treated CF was calcined in an Ar flow at 400 °C for 4 h with a heating rate of 3 °C min⁻¹ to successfully obtain the C₃N₄ anchored CF.

2.2 Characterization

The morphology and microstructure of the C₃N₄ on CF electrode (C₃N₄-CF) were determined using scanning electron microscopy (SEM) performed with a Hitachi S-4800. Fourier transform infrared (FTIR) spectra of the C₃N₄-CF and pure electrode were collected using a Bruker Tensor 27 spectrometer.

2.3 Cyclic voltammetry and electrochemical impedance spectroscopy

The cyclic voltammetry (CV, 0.4–1.4 V) and electrochemical impedance spectroscopy (EIS, 10⁻² to 10⁵ Hz) tests were conducted at room temperature in a standard three-electrode cell with a CHI660D electrochemical workstation and Autolab PGSTAT302N respectively, with a CF (0.5 × 1 cm) film as the working electrode, platinum mesh (1 × 1 cm) as the counter electrode, and a saturated calomel electrode (SCE) as the reference electrode.

2.4 Battery testing

C₃N₄-CF was used as the electrode material in a single cell to evaluate the electrocatalytic performance. The battery was fabricated by sandwiching a Nafion 117 membrane between two CF electrodes (2 × 2.5 cm). 2 M VOSO₄ dissolved in a 3 M H₂SO₄ solution was used as the electrolyte.

3. Results and discussion

By annealing a melamine precursor with carbon felt, graphitic C₃N₄ could be prepared and supported on the surface of the carbon fibers. Fig. 1 shows the morphology of the carbon felt with or without C₃N₄ loading. As seen in Fig. 1a and b, a relatively smooth surface of the carbon fibers was observed for the pure carbon felt. Fig. 1c and d successfully show the presence of graphitic C₃N₄ on the surface of the carbon fibers of the CF. A continuous C₃N₄ film was formed after the thermal annealing with melamine, which is consistent with literature.¹⁷ With increasing concentrations of melamine, more precipitates were found on CFs prepared with the same synthesis time and temperature. Compared with untreated CF (Fig. 1a and b), the carbon fibers in the C₃N₄-CF were uniformly wrapped in a wrinkled-layer structure of C₃N₄, as shown in Fig. 1c and d. The SEM results clearly demonstrate that C₃N₄ could be successfully grown on the CF with this method. The specific surface areas were 3.088 and 6.835 m² g⁻¹ for the pure CF and the C₃N₄-CF, respectively, by means of a BET method, and the reason for the surface area increase is the increased surface roughness with C₃N₄ on the CF.

Fig. 1 SEM images of [(a) and (b)] pristine CF, and [(c) and (d)] a C₃N₄-CF sample electrode.

The functional groups of the C₃N₄-CF samples were monitored using FTIR spectroscopy, and the spectra are presented in Fig. 2. Compared with the pure CF, three new strong absorption bands appeared at 807 cm⁻¹, 1200–1600 cm⁻¹ and 2800–3400 cm⁻¹.^18–20 The peak at 807 cm⁻¹ was assigned to the bending vibration of tri-s-thiazine. Typical absorption bands for C₃N₄ of a triangular C–N(–C)–C unit at 1180 cm⁻¹ and bridging C–NH–C units at 1239 cm⁻¹ could be found for the C₃N₄-CF sample. N–H bending and C–H stretching peaks can be seen at 1386 cm⁻¹ and 1543 cm⁻¹, respectively, while a C=O stretching peak was discovered at 1728 cm⁻¹. In addition, there are three possible sources for the peaks at 3180 cm⁻¹: (1) primary amines, (2) secondary amines, and (3) O–H stretching vibrations of water molecules.¹⁹ All this evidence directly indicates that C₃N₄ was successfully synthesized on the surface of the CF, which is consistent with the SEM results in Fig. 1, Raman results in Fig. S2, and XPS results in Fig. S3.†

Fig. 2 FTIR spectra of the original CF and the C₃N₄-CF electrode.

Cyclic voltammetry (CV) measurements were conducted to measure the electrochemical activity of the C₃N₄-CF toward the VO²⁺/VO₂⁺ couple for VRFBs. As demonstrated in Fig. 3a, redox reactions of the V(IV)/V(V) couple can be detected on the electrodes with a sweeping voltage range of 0.4–1.4 V vs. SCE for the pure CF and C₃N₄-CF. It could be clearly found that C₃N₄-CF shows a better electrochemical performance than the pure CF with a higher peak current density and lower onset potential. The corresponding electrochemical performance parameters are summarized at Table S1.† The anodic and cathodic peak currents (I_pa and I_pc) were 10.5 and 5.48 mA for the pure electrode, and 12.95 and 13.84 mA with the C₃N₄-CF electrode. The peak potential separations of the VO²⁺/VO₂⁺ reactions were 0.59 and 0.16 V for the pure and C₃N₄-CF electrode, respectively. The electrochemical characterization suggests that the electron transfer kinetics for the VO²⁺/VO₂⁺ reaction with carbon felt are significantly enhanced by the presence of the graphitic C₃N₄.²¹ This result clearly demonstrates that C₃N₄ can act as an efficient electrocatalyst to catalyze the VO²⁺/VO₂⁺ redox reactions. From the oxidation process used to assess these electrodes, the onset potentials are 1.18 and 0.91 V for the pure CF and C₃N₄-CF electrodes, respectively, which means the electrocatalytic kinetics of the oxidation process for the reaction of VO²⁺/VO₂⁺ on the electrodes are in the order of C₃N₄-CF > pure.²¹ We could obtain the same result from the reduction process in Fig. 3a. Moreover, the onset potential for both the anodic and cathodic processes was improved after the growth of C₃N₄, which would be favorable for the electron transfer kinetics and beneficial for increasing the energy storage efficiency.²²

Fig. 3 (a) Cyclic voltammograms with the untreated CF and C₃N₄-CF electrodes for the VO²⁺/VO₂⁺ redox couple at 1 mV s⁻¹ in 0.1 M VOSO₄ + 2 M H₂SO₄. (b) Charge–discharge curves of the VRFBs with pristine CF (black) and a C₃N₄-CF sample (red), (c) CE, VE, and overall EE values for the these electrode materials. (d) Continuous charge–discharge cycles of the VRFBs with CF (black) and C₃N₄-CF (red) samples as the electrode materials at a constant current density of 50 mA cm⁻².

The reversibility of the redox reaction can be estimated from the ratio of the peak currents (I_pa/I_pc) and the peak potential separation for the oxidation and reduction (ΔE = V_pa − V_pc). It could be obviously observed that after the growth of C₃N₄, the I_pa increased from 10.5 mA for pristine CF to 12.95 mA for the C₃N₄-CF and the ΔE reduced to 0.43 mV, indicating that the electrochemical properties are improved significantly. Compared with the pure sample, the value of the I_pa/I_pc for the C₃N₄-CF at the positive electrode was lower (0.93), indicating that it is close to the value (1.00) for a reversible redox reaction.²³ Meanwhile, the ΔE associated with polarization decreased from 0.59 to 0.16 V, resulting in an enhanced reversibility for vanadium redox couples using graphitic C₃N₄. The excellent electrochemical performance should be attributed to the unique polymeric melon structure of C₃N₄, with the C and N atoms having lone pair electrons in the p orbitals, these electronic interactions can form a similar π-conjugated structure to benzene rings to form a highly delocalized conjugated system.²⁴

To further study the performance of the C₃N₄-CF electrode in a VRFB, charge–discharge tests for all the electrodes were conducted. Fig. 3b presents the charge–discharge curves of the cells assembled with the pure CF and C₃N₄-CF electrodes at a current density of 50 mA cm⁻². It can be seen that the C₃N₄-CF electrode shows the lowest charge voltage plateau and the highest discharge voltage plateau. The coulombic efficiency (CE), voltage efficiency (VE), and overall energy efficiency (EE) of the VRFBs obtained with the different electrode materials are displayed in Fig. 3c. The difference in the EE could reflect the different electrocatalytic activities of the electrode materials. It could be observed in Fig. 3c that the energy efficiency (EE) for the C₃N₄-CF electrode is 85%, which is 9.1% higher than that of the pure CF. The cell cycling stability is shown in Fig. S1.† These results present that with C₃N₄ grown on CF as the electrode, the performance of the cell is superior to that with pure CF, further indicating the efficiency of the graphitic C₃N₄ catalyst. Fig. 3d shows the continuous charge–discharge cycles of VRFBs with various electrode materials at a constant current density of 50 mA cm⁻². After continuous cycling for 25 cycles, the C₃N₄-CF exhibited a much higher capacity than that of the pure CF, indicating the excellent stability of C₃N₄.

In order to gain additional supporting evidence, EIS measurements of the electrodes (pure CF and C₃N₄-CF electrodes) in a 0.1 M VOSO₄ + 2 M H₂SO₄ solution at the open-circuit potential were recorded to further investigate the catalytic activity, and the corresponding Nyquist plots are shown in Fig. 4. The Nyquist plot of the electrode is composed of a large semicircle in the high frequency region and a sloped line in the low frequency region. In the equivalent circuit (Fig. 4), R_s is used to represent the resistance composed of contact resistance, electrode resistance and the solution resistance. From the data obtained and shown in Table S2,† the R_s (1.64, 1.53 Ω) values for the pristine carbon felt and the modified carbon felt are almost equivalent, which indicates that C₃N₄ does not seriously increase the resistance of the batteries. R_p (53.1, 23.1 Ω) represents the charge transfer resistance across the electrode–electrolyte interface. The fitted R_p for the C₃N₄-CF was lower than that for the pristine CF, which indicates that the charge transfer process was faster with the C₃N₄-CF. The constant-phase element Q_m is representative of the electric double-layer capacitance of the electrode/solution interface, while Q_t represents the diffusion capacitance attributed to the diffusion processes of VO₂⁺ and VO²⁺. As total values Y₁ (0.722 and 1.10 mMho) and Y₂ (0.68 and 1.35 Mho), the larger values occur for the C₃N₄-CF, which is favorable for ion and electron transport, and so the polarization is dramatically alleviated and the reversibility is greatly improved. The impedance analysis is consistent with the CV results and battery performance mentioned above.

Fig. 4 Nyquist spectra from electrochemical impedance spectroscopy of the electrodes, obtained using 0.1 M VOSO₄ + 2 M H₂SO₄ solutions.

In summary, C₃N₄ exhibits an excellent electrocatalytic redox reversibility with positive VO²⁺/VO₂⁺ couples for VRFBs. An enhanced electrochemical performance was demonstrated in terms of the higher peak current density from CV and the value of I_pa/I_pc almost remains constant at about 0.93. Additionally, from another aspect, the charge transfer resistance is significantly reduced due to the introduction of C₃N₄, whose polymeric melon structure could accelerate electron transport and ion diffusion at the electrode/electrolyte interface with CF. The as-prepared C₃N₄ on CF hybrid represents a significant strategy for the development of highly effective VRFB electrode materials.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant No. 51402100 and 21573066), and the Youth 1000 Talent Program of China.

References

1. W. Wang, S. Kim, B. Chen, Z. Nie, J. Zhang, G.-G. Xia, L. Li and Z. Yang, Energy Environ. Sci., 2011, 4, 4068.
2. E. Sum, M. Rychcik and M. Skyllas-kazacos, J. Power Sources, 1985, 16, 85–95.
3. E. Sum and M. Skyllas-Kazacos, J. Power Sources, 1985, 15, 179–190.
4. M. Skyllas-Kazacos, M. Rychcik, R. G. Robins, A. G. Fane and M. A. Green, J. Electrochem. Soc., 1986, 133, 1057.
5. H. Kaneko, K. Nozaki, Y. Wada, T. Aoki, A. Negishi and M. Kamimoto, Electrochim. Acta, 1991, 36, 1191–1196.
6. B. Sun and M. Skyllas-Kazacos, Electrochim. Acta, 1992, 37, 1253–1260.
7. C. Flox, J. Rubio-Garcia, R. Nafria, R. Zamani, M. Skoumal, T. Andreu, J. Arbiol, A. Cabot and J. R. Morante, Carbon, 2012, 50, 2372–2374.
8. B. Sun and M. Skyllas-Kazakos, Electrochim. Acta, 1991, 36, 513–517.
9. W. H. Wang and X. D. Wang, Electrochim. Acta, 2007, 52, 6755–6762.
10. X. Wu, H. Xu, P. Xu, Y. Shen, L. Lu, J. Shi, J. Fu and H. Zhao, J. Power Sources, 2014, 263, 104–109.
11. M. S. Park, N. J. Lee, S. W. Lee, K. J. Kim, D. J. Oh and Y. J. Kim, ACS Appl. Mater. Interfaces, 2014, 6, 10729–10735.
12. S. Wang, X. Zhao, T. Cochell and A. Manthiram, J. Phys. Chem. Lett., 2012, 3, 2164–2167.
13. J. Xu, Y. Wang and Y. Zhu, Langmuir, 2013, 29, 10566–10572.
14. K. Maeda, X. Wang, Y. Nishihara, D. Lu, M. Antonietti and K. Domen, J. Phys. Chem. C, 2009, 113, 4940–4947.
15. Y. Zheng, J. Liu, J. Liang, M. Jaroniec and S. Z. Qiao, Energy Environ. Sci., 2012, 5, 6717.
16. Y. Hou, Z. Wen, S. Cui, X. Guo and J. Chen, Adv. Mater., 2013, 25, 6291–6297.
17. S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2009, 25, 10397–10401.
18. Y. Zhao, F. Zhao, X. Wang, C. Xu, Z. Zhang, G. Shi and L. Qu, Angew. Chem., Int. Ed., 2014, 53, 13934–13939.
19. J. Xu, H. T. Wu, X. Wang, B. Xue, Y. X. Li and Y. Cao, Phys. Chem. Chem. Phys., 2013, 15, 4510–4517.
20. Y. Z. Wang, N. Hao, Q. M. Feng, H. W. Shi, J. J. Xu and H. Y. Chen, Biosens. Bioelectron., 2016, 77, 76–82.
21. M. Park, Y. J. Jung, J. Kim, H. Lee and J. Cho, Nano Lett., 2013, 13, 4833–4839.
22. P. Han, Y. Yue, Z. Liu, W. Xu, L. Zhang, H. Xu, S. Dong and G. Cui, Energy Environ. Sci., 2011, 4, 4710.
23. L. Shi, S. Liu, Z. He, H. Yuan and J. Shen, Electrochim. Acta, 2015, 178, 748–757.
24. Y. Zheng, Y. Jiao, Y. Zhu, L. H. Li, Y. Han, Y. Chen, A. Du, M. Jaroniec and S. Z. Qiao, Nat. Commun., 2014, 5, 3783.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra11381h

---