TiO₂ has been widely used for photocatalysts because of its good chemical- and photostabilities to convert photon energy to electrical and chemical energies 1. However, due to its wide bandgap, the light absorption is limited only to the ultraviolet (UV) region of the solar spectrum. Hence, sensitizing TiO₂ with small bandgap semiconductors, such as quantum dots or organic dyes, has been extensively studied to harvest more photons in the visible light region of solar spectrum for the applications to quantum dot-sensitized solar cells 234, dye-sensitized solar cells 567, and photoelectrochemical (PEC) cells 8910.

Along with this current research trends, combining TiO₂ with carbonaceous nanomaterials has attracted much interest, and studies on these materials are increasing exponentially these days 11. For instance, high performance photocatalysts such as carbon nanotube-TiO₂ 121314, fullerene-TiO₂ (C₆₀-TiO₂) 151617, and graphene-TiO₂ 1819 composites have been introduced by several groups and have shown enhanced photocatalytic activities. Notably, C₆₀ has shown interesting effects when combined with TiO₂: facilitating the separation of photo-generated charge carriers from TiO₂ to C₆₀ 1516 or sensitizing TiO₂ to absorb visible light 17. However, the band-edge position of C₆₀ is unfavorable for a sensitizer of TiO₂ because the lowest unoccupied molecular orbital (LUMO) level of C₆₀ is lower than the conduction band of TiO₂ 17. From the viewpoint of energy levels, 66-phenyl-C61-butyric acid methyl ester (PCBM) is a better candidate than C₆₀ for the sensitization of TiO₂. We expect that the photo-excited electrons of PCBM can be transferred to TiO₂ more efficiently because the LUMO level of PCBM is slightly higher than the conduction band of TiO₂ 20. In our previous study, we have found that the band-edge positions as well as the bandgap of PCBM can be tuned by electron irradiation at different fluences 21. We believe that electron irradiation technique can be an alternative and unique method to modify the molecular structure and tune the bandgap 2223 compared to the conventional methods such as adjusting the particle size of quantum dots 2425 or modifying the molecular structure of the dyes 26 for larger light absorption. In addition to the bandgap, the band-edge positions can also be tuned by electron irradiation compared to the conventional methods such as ionic adsorption for specific quantum dots 27 or by varying the conjugation linkers in organic dyes 28.

Based on our previous findings, we present here a novel approach to improve the PEC performance of PCBM/TiO₂ electrodes using electron beam irradiation. The photocurrent density and open-circuit potential of PCBM/TiO₂ were respectively improved by 90% and 36% by electron irradiation. The effects of the electron irradiation on the PEC performances of PCBM/TiO₂ were systematically analyzed in this study.

Figure 1 shows the schematic representation of the preparation of PCBM/TiO₂ electrode and subsequent electron irradiation. The as-received TiO₂ nanoparticle paste (DSL 18NR-T, Dyesol Industries Pty Ltd., Queanbeyan, New South Wales, Australia) was deposited on a fluorine-doped tin oxide (FTO) glass substrate (8 Ωm^-2, Dyesol) by a doctor blade technique. Before the deposition of TiO₂ paste, FTO glass substrates were cut by 1.0 × 2.5 cm² in dimension and were sonicated successively in acetone, isopropanol, ethanol, and deionized water for thorough cleaning and dried in N₂ gas stream. After the deposition of TiO₂ paste, subsequent annealing process was performed at 450°C for 30 min with a temperature increase rate of 1°C min^-1. After the annealing, TiO₂ nanoparticle film was formed. The as-prepared TiO₂ electrodes were immersed vertically in a chlorobenzene solution containing 1.5 mM PCBM for 5 h while stirring. PCBM solution was prepared by dissolving PCBM (99.5% purity, Nano-C, Inc., Westwood, MA, USA) powder into chlorobenzene (≥99.5% purity, Sigma-Aldrich, St. Luois, MO, USA) solvent. After the immersion, the electrodes were washed in pure chlorobenzene several times and dried at ambient condition. As a result, PCBM/TiO₂ electrodes, where a thin layer of PCBM was coated on the TiO₂ nanoparticle electrodes, were prepared. Coating process of PCBM was carried out in darkness. The irradiation of an electron beam on PCBM/TiO₂ electrodes was carried out at room temperature and in vacuum lower than 2 × 10^-5 Torr. An electron beam was generated from a thermionic electron gun with electron energy of 50 keV, and current density of the electron beam was 1.6 μA cm^-2. The electron fluence was varied by adjusting the irradiation time. PCBM/TiO₂ electrodes were irradiated by 1, 2, and 4 h which correspond to electron fluence of 3.6 × 10¹⁶, 7.2 × 10¹⁶, and 1.44 × 10¹⁷ cm^-2, respectively. Diffuse reflectance UV-visible (VIS) spectra of electron-irradiated PCBM/TiO₂ powders were measured on a spectrometer (S-4100, SCINCO CO., LTD., Seoul, South Korea) by scratching the nanoparticle film off the FTO glass substrate.

After electron irradiation of PCBM/TiO₂ electrodes, a custom-made PEC cell was constructed to measure the PEC responses of electron-irradiated PCBM/TiO₂ electrodes, which act as photo-anodes of PEC cells. The PEC cell has a three-electrode configuration comprising a photo-anode, a Pt wire as a cathode, and a saturated calomel electrode (SCE) (0.242 V vs. NHE, BAS Inc., West Lafayette, IN, USA) as a reference electrode. An aqueous solution of 1 M NaOH (Junsei Chemical Co., Ltd., Chuo-ku, Tokyo, Japan) was used as a supporting electrolyte after 30 min purging with N₂ gas. The PEC response of the electrodes was recorded on a potentiostat (Model SP-50, BioLogic, Claix, France) by sweeping the potential from -1.2 to 0.5 V (vs. SCE) at a sweep rate of 100 mV s^-1. The photo-anodes were illuminated with a solar simulator (Model LS-150, Abet Technologies, Inc., Milford, CT, USA) equipped with AM 1.5 filter. The illumination power was estimated as 80 mW cm^-2 at the photo-anode surface by a digital photometer (ILT1400-A, International Light Technologies, Inc., Peabody, MA, USA).

Figure 1 displays the schematic representation for the preparation of the PCBM/TiO₂ photo-anodes of PEC cells. TiO₂ nanoparticles (NPs) were firstly deposited to form a film on a FTO glass substrate. A uniform TiO₂ NP film was formed by annealing the as-deposited TiO₂ paste at 450°C for 30 min. The TiO₂ NP film was submerged in a PCBM solution for 5 h, and consequently, the TiO₂ NP film was coated with PCBM. Subsequently, the PCBM/TiO₂ electrodes were irradiated with an electron beam. The energy of the electron beam was 50 keV, and the electron fluence was changed by controlling the irradiation time. These electron-irradiated PCBM/TiO₂ films on FTO glass substrates were used as photo-anodes of PEC cells for water splitting. Figure 2 shows the field emission scanning electron microscopy (FESEM) images of the fabricated PCBM/TiO₂ film. TiO₂ NPs with the diameter of approximately 20 nm were deposited on a FTO glass substrate (see details in the 'Methods' section). As shown in the FESEM image, the TiO₂ NPs were well interconnected with one another, forming a rigid film that is strongly attached to the FTO glass substrate. The thickness of the TiO₂ NP film was approximately 16.5 μm.

We observed that transparent TiO₂ NP film became slightly yellowish after the PCBM coating. The UV-VIS absorption spectra shown in Figure 3 more clearly characterize the optical properties of the TiO₂ NP films. When PCBM was coated on TiO₂, visible light absorption of TiO₂ in the wavelength range of 390 to 800 nm was increased, while absorption of UV in the range of 300 to 360 nm was decreased. In addition, when PCBM/TiO₂ was irradiated with an electron beam, the absorbance in both UV and visible light region decreased gradually as the electron fluence increased. In our previous work, we reported that the bandgap of electron-irradiated PCBM increased as the electron fluence was increased. The modification of the bandgap was attributed to the change in the molecular structure of PCBM by electron irradiation. From these facts, we could conclude that the effective bandgap of electron-irradiated PCBM/TiO₂ also increased as the electron fluence increased (Figure 4).

In order to investigate the band-tuning effect caused by the electron irradiation, we tried to characterize the PEC cell device performances using the electron-irradiated PCBM/TiO₂ electrodes. The measurement results of the PEC responses of bare TiO₂, PCBM/TiO₂, and electron-irradiated PCBM/TiO₂ electrodes are listed on Table 1, and the typical current density-potential curves of the electrodes are shown in Figure 5. The saturated current density at 0 V vs. saturated calomel electrode under dark conditions of all the electrodes was less than 15 μA cm^-2. Under illumination of simulated solar light, bare TiO₂ nanoparticle electrode shows saturated photocurrent density (J_ph) of 176 μA cm^-2 and open-circuit potential (E_ocp) of -0.85 V vs. SCE. After coating PCBM on TiO₂ nanoparticles, the PEC performance was improved: J_ph and E_ocp of PCBM/TiO₂ electrode increased to 234 μA cm^-2 and -1.05 V vs. SCE, respectively. The improvement in J_ph and E_ocp is attributed to the increment of visible light absorption of PCBM compared to that of TiO₂. After electron irradiation of PCBM/TiO₂ electrode at electron fluence of 3.6 × 10¹⁶ cm^-2, J_ph and E_ocp increased from 234 to 306 μA cm^-2 and -1.05 to -1.16 V vs. SCE, respectively. The PEC performance of PCBM/TiO₂ electrode was further improved through electron irradiation of increased electron fluence. Both J_ph and E_ocp of electron-irradiated PCBM/TiO₂ were increased with increasing the electron fluence. J_ph increased to 333 μA cm^-2, and E_ocp increased to -1.16 V vs. SCE at the electron fluence of 7.2 × 10¹⁶ cm^-2.

The fact that the PEC performance of PCBM/TiO₂ electrode was improved by electron fluence is interesting because electron irradiation increases the bandgap of PCBM and accordingly decreases the light absorption. As verified in our previous work, the LUMO level of PCBM shifts upward to the vacuum energy level as electron fluence increases. Since the bandgap of PCBM is much lower than that of TiO₂, electron-hole pairs produced in PCBM can contribute to the increase in the photo-current of TiO₂. However, the energy difference between the LUMO energy level of PCBM and the conduction band edge minimum of pure TiO₂ is 0.2 eV, which might not be high enough for efficient electron transfer from PCBM to TiO₂ 29. Since LUMO energy level of PCBM is up-shifted by electron irradiation, electron-irradiated PCBM provides higher driving force of electron injection from PCBM to TiO₂ 25. This can explain why J_ph of electron-irradiated PCBM/TiO₂ electrodes was increased by increasing the electron fluence. Moreover, the increase in the energy difference between the LUMO energy level of electron-irradiated PCBM and the conduction band edge minimum of TiO₂ provides efficient charge separation of the photo-excited electron-hole pairs, thereby improving E_ocp 30.

However, when the electron fluence was further increased to 1.44 × 10¹⁷ cm^-2, the PEC performance of electron-irradiated PCBM/TiO₂ became worse. As shown in Figure 4, the LUMO energy level of PCBM was constantly up-shifted toward the vacuum energy level as the electron fluence was increased. The up-shift in the LUMO energy level of electron-irradiated PCBM increases the driving force of electron injection from PCBM to TiO₂. With the up-shift in the LUMO energy level, the bandgap of the electron-irradiated PCBM also increases with increasing the electron fluence. The increase in the bandgap reduces the light absorption of PCBM and consequently deteriorates the PEC performance. Therefore, electron irradiation induces the two contradictory effects on the PEC performance of the electron-irradiated PCBM/TiO₂, and this suggests that there is an optimum electron fluence at which the PEC performance is maximized. In our experiments, J_ph increased by approximately 90% and E_ocp increased by approximately 36% compared to bare TiO₂ at an optimum electron fluence at 7.2 × 10¹⁶ cm^-2.

Using the fact that the electronic band structure of PCBM can be modified by electron irradiation, PCBM/TiO₂ electrodes were fabricated and tested in a PEC cell. We observed that electron irradiation on PCBM/TiO₂ electrodes led to an increase in J_ph by approximately 90% and E_ocp by approximately 36% at an optimum electron irradiation condition. These results show that electron irradiation approach can be a good tool to tune the bandgap and the band-edge positions of PCBM and provide an evidence that the approach is useful for PEC device application. We believe that the electron irradiation strategy can also control the electronic band structures of other organic semiconducting materials, and thus, this strategy can improve the performances of PEC and photocatalytic devices.

The authors declare that they have no competing interests.

The work was carried out by the collaboration between all authors. SOC initiated the idea of electron irradiation on PCBM/TiO₂ electrodes. SHY performed the electron irradiation experiments. SHY and GA performed the construction of PEC cell and measurement of PEC responses of electron-irradiated PCBM/TiO₂ electrodes. JMK and SHH carried out the diffuse reflectance UV-VIS spectroscopy measurements of electron-irradiated PCBM/TiO₂. SOC and SHY analyzed the data and suggested the mechanism of improvement of electron-irradiated PCBM/TiO₂ electrodes. All authors read and approved the final manuscript.