Investigation on Pesudocapacitance Mechanism of Magnéli Oxide Ti₄O₇ in Aqueous Electrolyte

Abstract

Possessing high electronic conductivity and the nature of chemical inertness, the Magnéli phase titanium oxide Ti₄O₇ is a promising material for various electrochemical applications. Herein, the Ti₄O₇ electrode in aqueous Li₂SO₄ electrolyte is characterized for its supercapacitor applications. The oxide electrode exhibits pseudocapacitive behavior over a wide potential range of ±1.0 V (vs. Ag/AgCl), showing a specific capacitance of 105 F g⁻¹, equivalent to 85 µF cm⁻²–oxide, along with outstanding high-rate performance and cycle stability (96 % capacitance retention after 5000 cycles). In situ X-ray absorption near-edge spectroscopy analysis on the Ti K-edge absorption reveals that the pseudocapacitance does not involve the redox reaction of the oxide electrode material. A pseudocapacitance mechanism attributed to the reversible redox reactions of the hydrogen and oxygen atoms adsorbed on the oxide surface is proposed.

1. Introduction

Within the escalated development of electrochemical energy storage (EES) technologies, electrochemical capacitors, also known as supercapacitors, can play an important role because of their high power density, excellent long-term cycling stability, and fast charge-discharge performances.^1–3 In particular, pseudocapacitive oxides, which are of high densities and capable of delivering high areal capacitances, offer the advantage of high volumetric energy and power densities as compared with the carbon-based electric-double-layer capacitors (EDLCs). Considering environmental sustainability, low cost, environmental benignity, and safety are additional important issues to be faced in selecting electrode materials for large-format EES devices. Titanium oxides mostly satisfy these considerations and have been suggested for electrode material in several energy conversion and storage studies.^4–6 However, because of insufficient electronic conductivity, the TiO₂ electrode typically revealed a redox behavior of a battery electrode within a narrow operation window^7–11 and is unsuitable for the supercapacitor application. On the other hand, the so-called Magnéli phase (MP) Ti sub-oxides, having a general formula Ti_nO_2n−1, where n ranges from 3 to 10, are unique among oxides for having substantially high electronic conductivities comparable to those of metals and simultaneously possessing the nature of high chemical robustness.^12–14 As a result, they have been demonstrated for various potential electrochemical conversion and storage applications.^15–22

Among these MP Ti sub-oxides, we have previously demonstrated the promising potential of Ti₄O₇ as a supercapacitor electrode.²³ In an aqueous electrolyte of Li₂SO₄, the Ti₄O₇ electrode showed an operating potential window within ±1.0 V (versus Ag/AgCl/saturated KCl reference electrode) and a specific capacitance exceeding 100 F g⁻¹. In this study, we employ in situ synchrotron X-ray absorption spectroscopy (XAS) to reveal the underlying capacitance mechanism of the Ti₄O₇ supercapacitor electrode. By monitoring and analyzing the variations in the Ti valence, it is concluded that the pseudocapacitance does not originate from the redox reaction of the oxide electrode material itself. A mechanism involving reversible adsorbed oxygen and hydrogen formation on the oxide surface is proposed.

2. Experimental

Ti₄O₇ particles were obtained using the combination of sol-gel and vacuum-carbothermic processes as previously described.²³ Briefly, Ti hydroxy oxide (Ti(OH)_xO_y) was first obtained using acid-catalyzed condensation of Ti(IV) tetra isopropoxide (TTIP, 97 % Sigma-Aldrich), and then redispersed in an aqueous solution of glucose to have a carbon-to-Ti ratio of 1.0. The solution was dried and the resulting powder was subjected to heat treatment at 1000 °C in a vacuum (<10⁻³ torr) for 2.5 hr before quenching.

Microstructural characterization was conducted using scanning electron microscopy (SEM; JSM-7600F, JEOL) and a transmission electron microscope (TEM) (FEI Tecnai TF20, Philips). Powder X-ray diffraction (XRD) was performed using a diffractometer (X-pert/Philips) with CuK_α radiation. Synchrotron X-ray absorption near-edge spectroscopy (XANES) analysis was conducted using the beam-line 17-C1 at the National Synchrotron Radiation Research Center in Taiwan. In situ XANES analysis was performed using porous stainless steel as the current collector and by placing the electrode in an acrylic cell, equipped with a platimum counter electrode and an Ag/AgCl (sat. KCl) reference electrode. The two sides of the cell were perforated and sealed using Kapton foils to allow the probing beam to pass through the cell. The scan rate for the analysis was 2 mV s⁻¹, and the scan was stopped at each selected potential for 10 min before collecting data.

To prepare the working electrode, the oxide, carbon black (Aldrich), multi-wall carbon nanotube (Scientech Corporation), and polytetrafluoroethylene (Aldrich) with weight ratios of 75/8/5/12 were mixed with ethanol to produce a slurry, which was then cast into a film on a Ti foil and dried at 50 °C for 60 min. The electrochemical properties of the electrodes were characterized using cyclic voltammetry (CV) on a potentiostat (608, CH Instruments; Eco Chemie PGSTAT30, AUTOLAB) in 2 M Li₂SO₄ aqueous solution. The fundamental electrochemical properties of the electrodes were characterized in a three-electrode configuration, which consists of a working electrode, platinum foil counter electrode, and Ag/AgCl/saturated KCl (EG&G, 197 mV vs. NHE at 25 °C) reference electrode. The specific capacity (Q_s) was calculated by integrating the current over an entire CV cycle:

  

\begin{equation} Q_{s} = \left(\oint I\,dV\right)/(\nu \times m \times 2), \end{equation} (1)

where I is the current, ν is the potential scan rate, and m is the mass of the active material.

3. Results and Discussion

As shown in Fig. 1a, Ti₄O₇ particles exhibited irregular shapes having rounded edges with sizes ranging from a few tens to hundreds of nanometers. High-resolution TEM analysis revealed clear lattice fringes extending across the entire particles without grain boundary, indicating these particles were single crystals (Fig. 1b). X-ray diffraction (XRD) analysis gave reflections consistent with those of the Magnéli phase (MP; PDF# 00-050-0787) (Fig. 1c).

Figure 1.

Material characterization of Ti₄O₇ powder: (a, b) TEM micrographs and (c) XRD pattern.

Electrochemical measurements were carried out between −1.0 and 1.0 V in an aqueous electrolyte of 2 M Li₂SO₄. As shown in the CV analysis (Fig. 2a), the Ti₄O₇ electrode exhibited nearly rectangular current profiles at scanning rates from 20 to 400 mV s⁻¹ over a very wide potential window of 2 V. Such a wide operating potential, particularly toward the lower potential end, in aqueous electrolytes is rather rare among pseudocapacitive oxides, and it is primarily owing to the low Ti⁴⁺ ↔ Ti³⁺ redox potential (ca. −1.097 V versus Ag/AgCl (sat. KCl)).²⁴ An anodic current wave above 0.8 V came from the oxidation of H₂O. The specific capacitance was 105 F g⁻¹ at 20 mV s⁻¹ and maintained 54 F g⁻¹, more than 50 % retention, even when the scan rate was increased by 20 times to 400 mV s⁻¹ (Fig. 2b). In addition, the electrode demonstrated remarkable cycle stability, retaining 96 % capacitance after 5000 CV cycles at 20 mV s⁻¹ (Fig. 2c).

Figure 2.

Electrochemical characterization of Ti₄O₇ electrode: (a) cyclic voltammograms at different potential scan rates; (b) specific capacitance versus potential scan rate; (c) capacitance retention versus CV cycle number under a scan rate of 20 mV s⁻¹ within ±0.8 V (inset: voltammograpms of selected cycles).

Having a specific surface area of 123 m² g⁻¹, the Ti₄O₇ electrode demonstrated an areal specific capacitance of nearly 85 µF per cm² of the oxide surface, which is substantially higher than that (<10 µF cm⁻²) typical of the electric double-layer capacitance and hence is pseudocapacitive in nature. The charge-storage mechanism was further investigated using in situ XANES analysis. Figure 3a plots the Ti K-edge absorption profiles of the Ti₄O₇ electrode acquired during a cathodic scan from 0.8 to −0.2 V. The profiles showed a pre-edge peak at around 4970 eV, followed by a major absorption front toward higher energy. The absorption edge energy is conventionally taken as the energy of the first inflection point of the major absorption front, and the inflection point can be unequivocally determined by taking the maximum point of the derivative plot of the profile, as shown in Fig. 3b. The edge energy thus determined is 4977.6 eV. The K-edge absorption energy generally increases with approximately a linear correlation with the cation valence. Based on the correlation established using the edge energies of Ti metal (Ti^o) and TiO₂ (Ti⁴⁺), the edge energy of the Ti₄O₇ electrode corresponds to a valence of 3.58 (Fig. 3c), which is very close to the theoretical value of 3.50. Furthermore, from Figs. 3a and 3b, it is clear that the Ti valence did not change at all. This is rather different from the cases of other well-known pseudocapacitive oxides, such as MnO₂ and RuO₂,^25,26 of which the pseudocapacitance involves the redox of the oxide electrode materials and is shown to be accompanied by valence variations, indicated by the shift in absorption edge energy, of the transition metal cations during the charge-discharge processes.

Figure 3.

In situ XANES study of Ti₄O₇ electrode during a cathodic scan: (a) Ti K-edge absorption plots taken at 0.8 V, 0.5 V, 0.2 V, 0 V and −0.2 V, showing overlapped absorption profiles; (b) the derivative curves of the absorption plots shown in (a); the black line marks the inflection point of each major absorption front to determine the edge-energy, and the line indicates no change in the edge energy during the scan; (c) determination of Ti valence of Ti₄O₇ (blue circle) based on linear correlation using the edge energies of the Ti^o (Ti metal) and Ti⁴⁺ (TiO₂) (red circles).

The XANES results indicate that the pseudocapacitance of Ti₄O₇ electrode does not originate from the redox reactions of the oxide electrode material itself. This is consistent with the fact that the Ti⁴⁺ ↔ Ti³⁺ redox potential is outside of the scanned potential range. Alternative redox reactions related to the species present in the electrolyte might be operative. Considering all the species in the electrolyte, the only elements that likely undergo redox reactions within the potential range under consideration are hydrogen and oxygen. It is inferred that the pseudocapacitance of Ti₄O₇ may be attributable to the reversible redox reactions of the hydrogen and oxygen atoms adsorbed on the surface active sites of the oxide particles, including

  

\begin{equation} \text{H}^{ + }{}_{(\text{ad})} + e^{ - } \leftrightarrow \text{H}_{(\text{ad})}, \end{equation} (2)

toward the lower potential end and

  

\begin{equation} 2\text{OH}^{ - }{}_{(\text{ad})} \leftrightarrow \text{O}_{(\text{ad})} + \text{H$_{2}$O} + 2e^{ - }. \end{equation} (3)

toward the higher potential end.

Pseudocapacitance arising from the redox reaction of adsorbed species, but not of the bulk electrode material, is not unprecedented. It has, for example, been shown for the magnetite pseudocapacitor in aqueous Na₂S₂O₃ electrolyte.²⁷ Furthermore, Reaction (2) is analogous to underpotential hydrogen deposition (UHD), which is known for the pseudocapacitance of metal electrodes.²⁸ Reaction (3), on the other hand, is in line with the observation of an anodic current rise associated with oxygen evolution above 0.8 V as mentioned earlier (Fig. 2a). Further experimental and/or theoretical investigations will be needed to validate these hypothesized surface reactions.

4. Conclusion

In summary, Ti₄O₇ has been demonstrated to be a promising pseudocapacitive electrode material in aqueous electrolyte, showing a specific capacitance of 105 F g⁻¹, equivalent to 85 µF cm⁻²–oxide, within a wide potential window of approximately ±1.0 V (vs. Ag/AgCl). The low-potential stability is unique among the transition metal pseudocapacitive oxides. In addition, the oxide electrode exhibits high-rate performance and outstanding cycle stability. In situ XANES analysis has revealed that the pseudocapacitance of Ti₄O₇ does not involve the redox of Ti. It is proposed that the pseudocapacitance originates from the reversible redox reactions of the adsorbed hydrogen and oxygen atoms on the oxide surface.

Acknowledgment

This work is financially supported by the “Advanced Research Center for Green Materials Science and Technology” from The Featured Area Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (112L9006) and the National Science and Technology Council in Taiwan under the grants of NSTC-110-2221-E-002-015-MY3 and 112-2923-E-011-005. The authors also thank Ms. S. J. Ji and Ms. C. Y. Chien (NSTC; NTU) for their assistance in performing electron microscopy analyses.

CRediT Authorship Contribution Statement

Yu-Ting Weng: Data curation (Lead), Formal analysis (Lead), Investigation (Lead), Methodology (Lead), Validation (Lead), Writing – original draft (Lead)

Tsung-Yi Chen: Data curation (Supporting), Methodology (Supporting)

Jeng-Lung Chen: Data curation (Supporting), Methodology (Supporting)

Nae-Lih Wu: Conceptualization (Lead), Formal analysis (Supporting), Funding acquisition (Lead), Project administration (Lead), Supervision (Lead), Writing – review & editing (Equal)

Conflict of Interest

The authors declare no conflict of interest in the manuscript.

Funding

National Science and Technology Council: NSTC-110-2221-E-002-015-MY3

National Science and Technology Council: NSTC-112-2923-E-011-005

Ministry of Education, Taiwan: 112L9006

Footnotes

Associated with the presentation at the 7th International Conference on Advanced Capacitors (ICAC2023) meeting, program number: 3-IL-4.

References

• 1)  M. Pershaanaa, S. Bashir, and S. Ramesh, J. Energy Storage, 50, 104599 (2022).
• 2)  Y. Wang, L. Zhang, H. Hou, W. Xu, G. Duan, S. He, K. Liu, and S. Jiang, J. Mater. Sci., 56, 173 (2021).
• 3)  N. Okita, E. Iwama, and K. Naoi, Electrochemistry, 88, 83 (2020).
• 4)  S. Liang, X. Wang, Y.-J. Cheng, Y. Xia, and P. Müller-Buschbaum, Energy Storage Mater., 45, 201 (2022).
• 5)  D. Saliu, M. Mamo, P. Ndungu, and J. Ramontja, J. Energy Storage, 49, 104155 (2022).
• 6)  J. Chen, J. Li, L. Sun, Z. Lin, Z. Hu, H. Zhang, X. Wu, D. Zhang, G. Cheng, and R. Zheng, J. Energy Chem., 59, 736 (2021).
• 7)  R. Huang, J. Zhang, Z. Dong, H. Lin, and S. Han, ACS Appl. Energy Mater., 6, 1673 (2023).
• 8)  R. D. Kumar, S. Balachandran, A. J. Kumar, C. Brundha, M. Kumar, and M. Lee, Mater. Lett., 335, 133812 (2023).
• 9)  S. Sharma, P. N. Sidhartha, and K. N. Chappanda, J. Electron. Mater., 52, 483 (2023).
• 10)  M. Z. U. Shah, M. Sajjad, H. Hou, S. ur Rahman, and A. Shah, J. Energy Storage, 55, 105651 (2022).
• 11)  H. Zhu and J. Zhang, Inorg. Chem. Commun., 145, 110027 (2022).
• 12)  J. R. Smith and F. C. Walsh, J. Appl. Electrochem., 28, 1021 (1998).
• 13)  M. Woydt, Tribol. Lett., 8, 117 (2000).
• 14)  L. Liborio, G. Mallia, and N. Harrison, Phys. Rev. B, 79, 245133 (2009).
• 15)  S. O. Ganiyu, N. Oturan, S. Raffy, G. Esposito, E. D. van Hullebusch, M. Cretin, and M. A. Oturan, Electrochim. Acta, 242, 344 (2017).
• 16)  P. Geng, J. Su, C. Miles, C. Comninellis, and G. Chen, Electrochim. Acta, 153, 316 (2015).
• 17)  G. Hasegawa, T. Sato, K. Kanamori, K. Nakano, T. Yajima, Y. Kobayashi, H. Kageyama, T. Abe, and K. Nakanishi, Chem. Mater., 25, 3504 (2013).
• 18)  A. Kitada, G. Hasegawa, Y. Kobayashi, K. Kanamori, K. Nakanishi, and H. Kageyama, J. Am. Chem. Soc., 134, 10894 (2012).
• 19)  Z. Luo, S. Sang, Q. Wu, and S. Liu, ECS Electrochem. Lett., 2, A21 (2013).
• 20)  L. Wang, P. Lettenmeier, U. Golla-Schindler, P. Gazdzicki, N. A. Cañas, T. Morawietz, R. Hiesgen, S. S. Hosseiny, A. S. Gago, and K. A. Friedrich, Phys. Chem. Chem. Phys., 18, 4487 (2016).
• 21)  C. Yao, F. Li, X. Li, and D. Xia, J. Mater. Chem., 22, 16560 (2012).
• 22)  P. K. Shen, C. He, S. Chang, X. Huang, and Z. Tian, J. Mater. Chem. A, 3, 14416 (2015).
• 23)  S. S. Huang, Y. H. Lin, W. Chuang, P. S. Shao, C. H. Chuang, J. F. Lee, M. L. Lu, Y. T. Weng, and N. L. Wu, ACS Sustainable Chem. Eng., 6, 3162 (2018).
• 24)  M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, National Association of Corrosion Engineers, Houston (1974).
• 25)  S. L. Kuo and N. L. Wu, J. Electrochem. Soc., 153, A1317 (2006).
• 26)  S. L. Kuo and N. L. Wu, J. Electrochem. Soc., 154, A34 (2007).
• 27)  S. Y. Wang, K. C. Ho, S. L. Kuo, and N. L. Wu, J. Electrochem. Soc., 153, A75 (2006).
• 28)  G. Jerkiewicz, Electrocatalysis, 1, 179 (2010).