Vibrational Spectroscopy of the Adsorbates on Nb-doped TiO₂ Single-crystal Electrodes

Abstract

Titanium dioxide (TiO₂) is one of the candidates for nonplatinum electrocatalysis of the cathode of a fuel cell. Adsorbates on the electrode surface affect the activity of electrochemical reactions. However, adsorbates on TiO₂ single crystal electrodes have not been determined in electrochemical environments. In this study, we have studied the adsorbates on the low index planes of 1 % Nb-doped rutile TiO₂ single crystal electrodes (Nb/TiO₂) in 0.1 M (mol L⁻¹) HClO₄ using nanoparticle surface-enhanced Raman spectroscopy (NPSERS) and infrared reflection absorption spectroscopy (IRAS). A NPSERS band was observed on Nb/TiO₂(110) surface at 585 cm⁻¹ above 0.5 V (vs. RHE). The band was shifted 15 cm⁻¹ to lower frequency in D₂O, indicating that the corresponding adsorbed species contains hydrogen. This band was assigned to a symmetric stretching vibration of Ti–OH–Ti (bridged OH) according to the density functional theory (DFT) calculations. The onset potential of the 585 cm⁻¹ increased as Nb/TiO₂(100) < Nb/TiO₂(110) ∼ Nb/TiO₂(111). However, the order of the activity for the oxygen reduction reaction (ORR) was Nb/TiO₂(100) < Nb/TiO₂(111) < Nb/TiO₂(110). No direct correlation was found between the onset potential of the 585 cm⁻¹ band and the ORR activity. IRAS spectra found adsorbed water on Pd doped Nb/TiO₂(110) (Pd/Nb/TiO₂(110)), the most active plane for the ORR, whereas no adsorbed water was observed on Pd/Nb/TiO₂(100) that has no ORR activity. These findings suggest that adsorbed water, which is undetectable by NPSERS, may induce the ORR activity on Nb/TiO₂.

1. Introduction

A fuel cell is a power generation device that produces electricity through the hydrogen oxidation reaction (HOR) at the anode and the oxygen reduction reaction (ORR) at the cathode. This process generates only electricity, water and heat, making a fuel cell a clean energy conversion system that does not emit carbon dioxide. Polymer electrolyte fuel cell (PEFC) operates at lower temperatures about 80 °C and can achieve high power density. Therefore, PEFC are expected to have broad applications to the power sources for vehicles and portable electronic devices. Although the theoretical efficiency of fuel cells exceeds 80 %, the actual efficiency is lowered by several factors. One of the serious problems is the higher overpotential of the ORR compared to the HOR that results in the significant energy losses. Furthermore, the electrocatalysts of PEFC are costly and scarce platinum materials. Therefore, reducing platinum loading or developing alternative catalyst materials is crucial.

Titanium dioxide (TiO₂), which is abundant in resources, has recently attracted attention as an electrocatalyst of PEFC. Ti₄O₇ exhibited higher electrochemical stability than glassy carbon electrodes and showed less degradation after the accelerated durability tests.¹ Groenenboom et al. investigated various metal dopants in TiO₂.² Nb and Ag dopants reduced the overpotential of the ORR, significantly. Yamamoto et al. predicted that TiO₂ doped with Rh or Pd lowers the ORR overpotential close to zero.³ These findings suggest that metal-doped TiO₂ is more effective for fuel cell reactions compared to pure TiO₂. Therefore, we focus on Nb-doped TiO₂ (Nb/TiO₂).

Kameda et al. reported the ORR activity on the low-index planes of Nb-doped TiO₂ single crystals. The activity increases as Nb/TiO₂(100) < Nb/TiO₂(111) < Nb/TiO₂(110). They also investigated the concentration dependence of doped Nb on the ORR activity and found that 1 wt% (= 0.9 at%) Nb/TiO₂ single crystal electrodes have the highest ORR activity. Furthermore, they demonstrated that the addition of hydrophobic organic molecules, such as melamine and tetrahexylammonium cation (THA⁺), slightly enhances the ORR activity.⁴

Nb atoms are not segregated on the surfaces of Nb/TiO₂ nanoparticles.⁵ When the bulk Nb concentration is 0.9 at%, Nb concentrations at the topmost layers are approximately 2–3 at% on single crystals of rutile Nb/TiO₂.⁶ Therefore, 97–98 % of a surface of Nb/TiO₂ single crystal electrode is composed of Ti and O atoms. Moreover, since NbO₂ and Nb₂O₅ have been shown to exhibit no ORR activity,⁷ the doped Nb is unlikely to serve as an active site.

Takahashi et al. observed OH groups at bridged oxygen sites on TiO₂ surfaces in air at 3279 cm⁻¹ using polarization modulation infrared external reflection spectroscopy (PM-IER). This frequency is approximately 400 cm⁻¹ lower than that observed by high-resolution electron energy loss spectroscopy (HREELS) in vacuum, indicating strong hydrogen bonding with atmospheric water molecules.⁸ Additionally, phonon bands of the crystal lattice appear in the low-frequency region of TiO₂. Raman spectroscopy measurements of rutile TiO₂ in air found phonon bands at 235, 445 and 610 cm⁻¹.⁹

Nanoparticle surface-enhanced Raman spectroscopy (NPSERS) is an effective technique for analyzing surface adsorbates on single crystals.^10–12 NPSERS using Au nanoparticles selectively enhances the Raman scattering signals of surface species via plasmon resonance of the Au nanoparticles, facilitating the detection of surface adsorbates. Unlike conventional surface-enhanced Raman spectroscopy (SERS), this method does not need to roughen the surface, making it applicable to well-defined single crystal electrodes. Sugimura et al. employed NPSERS to observe Pt–(OH) band at 570 cm⁻¹ on Pt single crystal electrodes in perchloric acid aqueous solution. The integrated band intensity of the Pt–(OH) correlated with the ORR activity, indicating that Pt–(OH) is an ORR-inhibiting species.¹³

Shell-isolated nanoparticle-enhanced Raman spectroscopy (SHINERS), which uses SiO₂-coated Au nanoparticles,^14,15 has been widely used to measure adsorbates on single crystal electrodes. The insulating SiO₂ shell suppresses electrochemical reactions on the Au nanoparticles. However, compared to bare Au nanoparticles, the Raman signal enhancement of SiO₂-coated Au nanoparticles is too low to detect adsorbates on Nb/TiO₂ single crystal electrodes. Therefore, we adopted more sensitive NPSERS using bare Au nanoparticles.

There has been no study on surface adsorbates on TiO₂ single-crystal electrodes in electrochemical environments. The lack of information on the adsorbates prevents the design of catalysts of fuel cells utilizing TiO₂. In this paper, we have studied the adsorbates on the low-index planes Nb/TiO₂ in electrochemical environments using NPSERS and infrared reflection absorption spectroscopy (IRAS).

2. Experimental

Perchloric acid (60 %, ultrapure), hydrofluoric acid (40 %, ultrapure) and deuterium oxide (99.8 %) were purchased from Kanto Chemical Co., Inc. Tetrachloroauric (III) acid tetrahydrate (99.9 %) was purchased from Wako Pure Chemical Industries, Ltd. The solutions were prepared using ultrapure water treated with Milli-Q Advantage A10 (Millipore) or deuterium oxide. 1 wt% Nb-doped rutile-type TiO₂ single-crystal plates (10 mm × 10 mm × 0.50 mm) were purchased from Shinkosha Ltd. Pd ion implantation to Nb/TiO₂ single crystal plate was carried out using an ion implantation system at Ion Technology Center, with an acceleration voltage of 70 keV and a dose of 1 × 10¹⁵ atoms cm⁻².

Study on the structural effects on the adsorbates needs atomically flat Nb/TiO₂ single crystal surfaces. Therefore, surface treatment was carried out using the following procedure.^16–19

1. (1)    The Nb/TiO₂ crystal was rinsed with acetone and ultrapure water.
2. (2)    The crystal was annealed in an electric furnace at 600 °C for 2 hours, and then cooled to room temperature.
3. (3)    The crystal was chemically polished in 30 wt% HF solution for 6 hours.

Annealing improves the crystallinity of rutile-type TiO₂. Since TiO₂ gradually dissolves in HF, chemical polishing with HF was used to remove surface roughness. The surface was imaged using atomic force microscope (AFM, MultiMode 8, Bruker) to confirm that the surface was flat at the atomic scale after the surface treatment.

Au nanoparticles used for NPSERS were synthesized by the ultrasonic reduction method.²⁰ The Au nanoparticles had an average diameter of approximately 55 nm. The Au nanoparticles were dispersed on the surface-treated Nb/TiO₂ single crystal using a 10 µL micropipette, followed by vacuum drying. This procedure was repeated three times. The coverage of Au nanoparticle was approximately 0.3.

Cyclic voltammogram (CV) was measured using an electrochemical analyzer (ALS 701CH, BAS) in 0.1 M HClO₄ saturated with Ar. Voltammograms were recorded in the potential range from 0.05 to 1.0 V (vs. RHE) at a scanning rate of 0.050 V s⁻¹.

NPSERS spectra were measured in 0.1 M HClO₄ saturated with Ar using a RamanRxn1 Analyzer (Kaiser Optical Systems, Inc.) equipped with a CCD camera by stepping the potential positively with the interval of 0.1 V. A He–Ne laser (λ = 632.8 nm, laser power 1.5 mW on the sample) was used for the excitation. The laser was focused onto the sample using an objective lens (60× magnification, CFI S Plan Fluor ELWD, Nikon). The single-crystal surface was positioned approximately 0.5 mm from the window to avoid the hindrance of mass transport.

DFT calculations were performed to assign surface adsorbates. All calculations were carried out using Gaussian 16W. The B3LYP density functional was used in combination with the SDD basis set for Nb and Ti atoms, and the 6-31G** basis set for O and H atoms. The spin multiplicity was set to singlet, and water was used as the solvent.

NPSERS and IRAS spectra were also measured on Pd-implanted rutile-type 1 wt% Nb-doped TiO₂ (Pd/Nb/TiO₂) single crystal electrodes in 0.1 M HClO₄ saturated with Ar. Before the measurements, the single-crystal surface was annealed at 800 °C for 12 hours, and then chemically polished in 30 % HF solution for 24 hours to achieve a flat surface at the atomic scale.

NPSERS spectra were measured on the Pd/Nb/TiO₂(110). IRAS measurements were done on Pd/Nb/TiO₂(100) and Pd/Nb/TiO₂(110) using a Fourier transform IR spectrometer (FT-IR-6800, JASCO) equipped with a mercury cadmium telluride (MCT) detector. Sample spectra were collected at 0.5 V and 0.9 V, and reference spectra were measured at 0.1 V. The spectra were averaged using the subtractively normalized interfacial FT-IR spectroscopy (SNIFTIRS) method over 64 × 160 scans. IR window was a CaF₂ prism with an angle of incidence 65°. IRAS spectra were calculated as follows:

  

\begin{equation} \textit{Absorbance} = -{\log}(R_{E\text{s}}/R_{E\text{r}}) \end{equation} (1)

• R_Es: reflected light intensity at sample potential
• R_Er: reflected light intensity at reference potential

3. Results and Discussion

3.1 AFM image

Figure 1 shows AFM images and cross-sectional profiles of the Nb/TiO₂ single-crystal surfaces after the surface treatment. The ionic radius of Ti⁴⁺ and O²⁻ in TiO₂ are 0.075 nm and 0.14 nm, respectively. The terrace roughness examined is within 0.2 nm, showing the surfaces are atomically flat.

Figure 1.

AFM images (a)–(c) and cross-sectional profiles (d)–(f) of Nb/TiO₂ single-crystal surfaces. (a)(d): Nb/TiO₂(100), (b)(e): Nb/TiO₂(110), (c)(f): Nb/TiO₂(111).

3.2 CV

Figure 2 shows CVs of Nb/TiO₂ single crystal electrodes. The CVs agree well with those reported previously.⁴ Broad redox peaks are found around 0.2 V. The origin of this peak remains unidentified.

Figure 2.

CVs of Nb/TiO₂ single crystal electrodes in 0.1 M HClO₄ saturated with Ar. Scanning rate: 50 mV/s.

CVs of Pt single crystal electrodes give the hydrogen adsorption/desorption peaks characteristic of the crystal orientations. No peak corresponding to hydrogen adsorption/desorption appears in the CV of Nb/TiO₂. This is due to the absence of hydrogen adsorption on Nb/TiO₂ surfaces.

CVs of Nb/TiO₂(110) were measured before and after the Au nanoparticle modification to investigate the effects of Au nanoparticle deposition (Fig. S1). The CV after Au nanoparticle modification is similar to that before the modification. This result indicates that the electronic state of the Nb/TiO₂ surface is not altered by the modification of Au nanoparticles.

3.3 NPSERS spectra on the low index planes of Nb/TiO₂

Figure 3a shows absolute NPSERS spectra in 0.1 M HClO₄ saturated with Ar. The dashed lines show the phonon bands that originate from quantized lattice vibrations. Band at 585 cm⁻¹ is observed above 0.5 V (Fig. 3b). In this study, Au nanoparticles were deposited on the electrode surface; there is a concern that these bands originate from Au oxides of Au nanoparticles. However, the Au–O stretching vibration of Au nanoparticles appears at 590 cm⁻¹ above 1.4 V.¹³ Therefore, the band observed at 585 cm⁻¹ is attributed to surface-adsorbed species on Nb/TiO₂.

Figure 3.

NPSERS spectra of Nb/TiO₂(110) in 0.1 M HClO₄/H₂O saturated with Ar: (a) absolute spectra, (b) differential spectra using a spectrum at 0.1 V as a reference.

On Pt electrodes, the NPSERS band of Pt-(OH) shifts to higher frequency at positive potentials.¹³ However, no shift is found on Nb/TiO₂. This difference can be attributed to the semiconducting nature of Nb/TiO₂ that contains fewer free electrons compared to metals. As a result, the electronic state of the adsorbates remains unchanged, resulting in no frequency shift.

NPSERS measurements were also conducted in D₂O solution of 0.1 M HClO₄ saturated with Ar. Figure 4 shows a comparison of the band frequencies in H₂O and D₂O. The band at 585 cm⁻¹ shifts to lower frequency 15 cm⁻¹ in D₂O, indicating that this band originates from a hydrogen-containing adsorbate. The observed shift also excludes the possibility that the band is due to an Au–O species adsorbed on the Au nanoparticles. In addition, Au-OH is observed at approximately 800 cm⁻¹ only in alkaline solutions.²¹

Figure 4.

Comparison of 585 cm⁻¹ band frequency on Nb/TiO₂(110) in Ar saturated 0.1 M HClO₄/H₂O and 0.1 M HClO₄/D₂O at 1.1 V (0.1 V reference difference spectra).

DFT calculations were performed using the structural model shown in Fig. 5 to assign the observed Raman band. The results are summarized in Table 1. According to the DFT calculations, the symmetric stretching vibration of the Ti–OH–Ti (bridged OH) ν_s,Ti-OH-Ti appears at 568 cm⁻¹. When hydrogen is replaced with deuterium to form Ti–OD–Ti, the frequency shifts to lower wavenumber by 12 cm⁻¹. The DFT result is consistent with the experimental result. On the other hand, the DFT calculations in Table S1 show no Nb–OH vibrational modes around 580 cm⁻¹. Therefore, the band at 585 cm⁻¹ is assigned to the symmetric stretching vibration of Ti–OH–Ti (bridged OH).

Figure 5.

A model used for DFT calculations.

Table 1. Results of DFT calculations.

Mode	Frequency (Hydrogen) /cm−1	Raman activity (Hydrogen)	Frequency (Deuterium) /cm−1	Raman activity (Deuterium)
νas,Ti-OH-Ti (asymmetrical)	407	56.6	386	54.0
νs,Ti-OH-Ti (symmetrical)	568	2.59 × 103	556	1.58 × 103
δTi-O-H	804	3.48	611	0.420
νTiO-H	3873	1.69 × 106	2820	9.36 × 105

Figure 6 shows ν_s,Ti-OH-Ti of bridged OH and its adsorption model on the surface. Since the experiments were conducted in an Ar atmosphere, H₂O will be dissociatively adsorbed to form Ti–OH–Ti (bridged OH). Furthermore, it is probable that the adsorbed site of bridged OH is bridged oxygen vacancies.²²

Figure 6.

Mechanism of the bridged OH formation.

NPSERS spectra were also measured on Nb/TiO₂(100) and Nb/TiO₂(111) electrodes using the same procedure as Nb/TiO₂(110). The results are shown in Fig. 7. A NPSERS band intensity depends on the degree of corrugation of Au nanoparticles on the surface. Thus, we compared the NPSERS band intensity between different Pt single crystal electrodes using the band intensity and the oxidation charge of adsorbed CO as internal standard.¹³ However, we cannot compare the band intensity of NPSERS between different Nb/TiO₂ electrodes because CO is not adsorbed on TiO₂.

Figure 7.

Differential NPSERS spectra on (a) Nb/TiO₂(100) and (b) Nb/TiO₂(111) in 0.1 M HClO₄/H₂O saturated with Ar. Reference spectra were collected at 0.1 V.

The 585 cm⁻¹ band also appears on Nb/TiO₂(100), where terminal OH (5-coordinated Ti–OH) are absent.⁸ This result also suggests that the 585 cm⁻¹ band originates from bridged OH rather than terminal OH. Although the detailed reaction pathway on Nb/TiO₂ remains unclear, previous theoretical study on rutile TiO₂(110) predicts a four-electron ORR pathway via adsorbed oxygen molecules. After the formation of OOH, the O–O bond is cleaved and the intermediate is further reduced to produce water.²³ The bridged OH species on the TiO₂ surface might promote O₂ adsorption and contribute to enhanced ORR activity according to the previous report.²³ The onset potentials of bridged OH (585 cm⁻¹ bands) formation are 0.5 V, 0.4 V and above 0.5 V on TiO₂(110), TiO₂(100) and TiO₂(111), respectively. Therefore, the onset potentials of the bridged OH formation increase as Nb/TiO₂(100) < Nb/TiO₂(110) ∼ Nb/TiO₂(111). On the other hand, the ORR activity differs by approximately a factor of two between TiO₂(110) and TiO₂(111).⁴ These results support that the bridged OH does not significantly affect the ORR. On Pt single crystal electrodes, a correlation has been observed between the intensity of the Pt-(OH) band and the activity sequence of the crystal facets, indicating that OH species act as a poisoning adsorbate. The OH effects on the ORR on Nb/TiO₂ single crystal electrodes differ from those on Pt single crystal electrodes. These results indicate that NPSERS inactive adsorbed water contributes to the ORR activity. SERS detected adsorbed water above the hydrogen evolution potential only when ions like Cl⁻ separate adsorbed water from bulk water.²⁴ Therefore analysis using IRAS is necessary.

3.4 NPSERS on Pd/Nb/TiO₂(110)

We could not observe IRAS spectra of adsorbed water on Nb/TiO₂ single crystal electrodes because reflected IR light intensity was too weak. The IRAS spectra of adsorbed water could be measured on Pd doped Nb/TiO₂(110) (Pd/Nb/TiO₂(110)) and Pd doped Nb/TiO₂(100) (Pd/Nb/TiO₂(100)) electrodes after the spectra were averaged over 10240 scans. The ORR activity of Pd/Nb/TiO₂ single crystal electrodes was Pd/Nb/TiO₂(100) ≪ Pd/Nb/TiO₂(110). The marked difference of the ORR activity supports that Pd nanoparticles are not exposed on the surface. NPSERS measurements were also performed on Pd/Nb/TiO₂(110) to confirm that the adsorbates on Pd/Nb/TiO₂(110) are similar to those on Nb/TiO₂(110). The results are shown in Fig. S2. The onset potential of 585 cm⁻¹ band is identical with that on Nb/TiO₂(110). NPSERS quality of Pd/Nb/TiO₂(110) is similar to that of Nb/TiO₂(110).

3.5 IRAS results for Pd/Nb/TiO₂(110)

IRAS spectra were measured on Pd/Nb/TiO₂(100) and Pd/Nb/TiO₂(110) in 0.1 M HClO₄/D₂O, because IR light intensity at O-D stretching vibration ν_OD region is much higher than that of O-H stretching vibration ν_OH region. The results are shown in Fig. 8. Positive-going band of hydrogen-bonded ν_OD is observed around 2600 cm⁻¹ at 0.5 and 0.9 V on Pd/Nb/TiO₂(110). This result shows that the coverages of hydrogen-bonded adsorbed D₂O at 0.5 and 0.9 V are higher than that at 0.1 V. The negative-going band at 2690 cm⁻¹ is assigned to the ν_OD of isolated D₂O adsorbed on the surface. Negative-going band indicates that the coverage of isolated D₂O at 0.1 V is higher than those at the sample potentials. On Pt single crystal electrodes, ν_OD of isolated D₂O was observed as positive-going band,^25–27 showing that the coverages of isolated D₂O at sample potentials are higher than that at reference potential (0.1 V). The polarity of the band of hydrogen-bonded ν_OD on Pd/Nb/TiO₂(110) is also opposite to those on Pt single crystal electrodes.^25–27 The tendency of ν_OD bands on Pd/Nb/TiO₂(110) electrode completely differs from that on Pt single crystal electrodes. We cannot discuss the band intensities and frequencies of the ν_OD bands, because the bipolar bands inhibit the correct analysis of band intensity and frequency. On Pd/Nb/TiO₂(100), considering the signal-to-noise ratio, no adsorbed water is present on the surface. On Pt single crystal electrodes, isolated and hydrogen-bonded water decrease the stability of the blocking species of the ORR (Pt oxides).²⁶ These results suggest that existence of the adsorbed water promotes the ORR activity on Pd/Nb/TiO₂(110) electrodes.

Figure 8.

IRAS spectra on Pd/Nb/TiO₂ in 0.1 M HClO₄/D₂O saturated with Ar. Reference spectra were collected at 0.1 V. All the spectra were averaged over 10240 scans with a resolution of 4 cm⁻¹. (a) Pd/Nb/TiO₂(100), (b) Pd/Nb/TiO₂(110).

4. Conclusion

The Nb/TiO₂ single crystal surfaces were investigated under electrochemical environments using NPSERS and IRAS, and surface adsorbates were identified.

The band at 585 cm⁻¹ was assigned to bridged OH (Ti–OH–Ti) symmetric stretching vibration) based on experimental results using D₂O and DFT calculations.

The onset potentials of the bridged OH formation increase as Nb/TiO₂(100) < Nb/TiO₂(110) ∼ Nb/TiO₂(111). On the other hand, the ORR activity differs by approximately a factor of two between TiO₂(110) and TiO₂(111). These results support that bridged OH does not significantly affect the ORR.

IRAS detected the adsorbed water on Pd/Nb/TiO₂(110) with higher ORR activity, whereas no adsorbed water was found on Pd/Nb/TiO₂(100) that has no ORR activity. These results suggest that adsorbed water promotes the ORR activity on Pd/Nb/TiO₂(110).

Acknowledgment

This work was partially supported by the New Energy and Industrial Technology Development Organization (NEDO).

Data Availability Statement

The data that support the findings of this study are openly available under the terms of the designated Creative Commons License in J-STAGE Data at https://doi.org/10.50892/data.electrochemistry.30293893.

• 1) Titanium dioxide (TiO₂) is one of the candidates for nonplatinum electrocatalysis of the cathode of a fuel cell. Adsorbates on the electrode surface affect the activity of electrochemical reactions. However, adsorbates on TiO₂ single crystal electrodes have not been determined in electrochemical environments. In this study, we have studied the adsorbates on the low index planes of 1 % Nb-doped rutile TiO₂ single crystal electrodes (Nb/TiO₂) in 0.1 M (mol L⁻¹) HClO₄ using nanoparticle surface-enhanced Raman spectroscopy (NPSERS) and infrared reflection absorption spectroscopy (IRAS). A NPSERS band was observed on Nb/TiO₂(110) surface at 585 cm⁻¹ above 0.5 V (vs. RHE). The band was shifted 15 cm⁻¹ to lower frequency in D₂O, indicating that the corresponding adsorbed species contains hydrogen. This band was assigned to a symmetric stretching vibration of Ti–OH–Ti (bridged OH) according to the density functional theory (DFT) calculations. The onset potential of the 585 cm⁻¹ increased as Nb/TiO₂(100) < Nb/TiO₂(110) ∼ Nb/TiO₂(111). However, the order of the activity for the oxygen reduction reaction (ORR) was Nb/TiO₂(100) < Nb/TiO₂(111) < Nb/TiO₂(110). No direct correlation was found between the onset potential of the 585 cm⁻¹ band and the ORR activity. IRAS spectra found adsorbed water on Pd doped Nb/TiO₂(110) (Pd/Nb/TiO₂(110)), the most active plane for the ORR, whereas no adsorbed water was observed on Pd/Nb/TiO₂(100) that has no ORR activity. These findings suggest that adsorbed water, which is undetectable by NPSERS, may induce the ORR activity on Nb/TiO₂.

CRediT Authorship Contribution Statement

Mizuki Kobayashi: Investigation (Lead), Writing – original draft (Lead)

Mizuki Takeno: Investigation (Lead)

Masashi Nakamura: Writing – review & editing (Lead)

Nagahiro Hoshi: Conceptualization (Lead), Funding acquisition (Lead), Project administration (Lead), Supervision (Lead), Writing – review & editing (Lead)

Conflict of Interest

The authors declare no conflict of interest in the manuscript.

Funding

New Energy and Industrial Technology Development Organization: 22101141-0

Footnotes

M. Kobayashi: ECSJ Student Member

M. Nakamura and N. Hoshi: ECSJ Active Members

References

• 1)  M. Hamazaki, A. Ishihara, Y. Kohno, K. Matsuzawa, S. Mitsushima, and K. Ota, Electrochemistry, 83, 817 (2015).
• 2)  M. C. Groenenboom, R. M. Anderson, D. J. Horton, Y. Basdogan, D. F. Roeper, S. A. Policastro, and J. A. Keith, J. Phys. Chem. C, 121, 16825 (2017).
• 3)  Y. Yamamoto, S. Kasamatsu, and O. Sugino, J. Phys. Chem. C, 123, 19486 (2019).
• 4)  N. Kameda, M. Nakamura, and N. Hoshi, Electrochemistry, 89, 1 (2021).
• 5)  D. H. Kim, G. S. Han, W. M. Seong, J. W. Lee, B. J. Kim, N. G. Park, K. S. Hong, S. Lee, and H. S. Jung, ChemSusChem, 8, 2392 (2015).
• 6)  A. Sasahara and M. Tomitori, J. Phys. Chem. C, 117, 17680 (2013).
• 7)  K. Sakai, L. Zhang, and R. R. Adzic, Phys. Chem. Chem. Phys., 10, 159 (2008).
• 8)  K. Takahashi and H. Yui, J. Phys. Chem. C, 113, 20322 (2009).
• 9)  X. Lu, S. Gao, P. Wu, Z. Zhang, L. Zhang, X. Li, and X. Qin, Minerals, 13, 703 (2023).
• 10)  D. L. Jeanmaire and R. P. Van Duyne, J. Electroanal. Chem. Interfacial Electrochem., 84, 1 (1977).
• 11)  A. D. McFarland, M. A. Young, J. A. Dieringer, and R. P. Van Duyne, J. Phys. Chem. B, 109, 11279 (2005).
• 12)  K. Ikeda, N. Fujimoto, H. Uehara, and K. Uosaki, Chem. Phys. Lett., 460, 205 (2008).
• 13)  F. Sugimura, N. Sakai, T. Nakamura, M. Nakamura, K. Ikeda, T. Sakai, and N. Hoshi, Phys. Chem. Chem. Phys., 19, 27570 (2017).
• 14)  J. F. Li, Y. F. Huang, Y. Ding, Z. L. Yang, S. B. Li, X. S. Zhou, F. R. Fan, W. Zhang, Z. Y. Zhou, D. Y. Wu, B. Ren, Z. L. Wang, and Z. Q. Tian, Nature, 464, 392 (2010).
• 15)  U. Zhumaev, A. V. Rudnev, J. F. Li, A. Kuzume, T. H. Vu, and T. Wandlowski, Electrochim. Acta, 112, 853 (2013).
• 16)  A. Imanishi, H. Suzuki, N. Ohashi, H. Kondoh, T. Ohta, and Y. Nakato, J. Phys. Chem. C, 113, 17254 (2009).
• 17)  A. Imanishi, T. Okamura, N. Ohashi, R. Nakamura, and Y. Nakato, J. Am. Chem. Soc., 129, 11569 (2007).
• 18)  A. Imanishi, H. Suzuki, N. Ohashi, T. Ohta, and Y. Nakato, Inorg. Chim. Acta, 361, 778 (2008).
• 19)  R. Nakamura, T. Okamura, N. Ohashi, A. Imanishi, and Y. Nakato, J. Am. Chem. Soc., 127, 12975 (2005).
• 20)  T. Sakai, H. Enomoto, K. Torigoe, H. Sakai, and M. Abe, Colloids Surf. A, 347, 18 (2009).
• 21)  C. Y. Li, J. C. Dong, X. Jin, S. Chen, R. Panneerselvam, A. V. Rudnev, Z. L. Yang, J. F. Li, T. Wandlowski, and Z. Q. Tian, J. Am. Chem. Soc., 137, 7648 (2015).
• 22)  A. Sasahara and M. Tomitori, J. Phys. Chem. C, 120, 21427 (2016).
• 23)  Á. Ganyecz, P. D. Mezei, and M. Kállay, Comput. Theor. Chem., 1168, 112607 (2019).
• 24)  S. H. Macomber, T. E. Furtak, and T. M. Devine, Surf. Sci., 122, 556 (1982).
• 25)  N. Hoshi, K. Saikawa, and M. Nakamura, Electrochem. Commun., 106, 106536 (2019).
• 26)  N. Hoshi, M. Nakamura, R. Kubo, and R. Suzuki, Commun. Chem., 7, 23 (2024).
• 27)  Y. Katagiri, M. Nakamura, and N. Hoshi, J. Phys. Chem. C, 129, 13954 (2025).