The use of a photoelectrochemical device is an efficient method of converting solar energy into hydrogen fuel via water splitting reactions. One of the best photoelectrode materials is Si, which absorbs a broad wavelength range of incident light and produces a high photocurrent level (~44 mA·cm−2). However, the maximum photovoltage that can be generated in single-junction Si devices (~0.75 V) is much lower than the voltage required for a water splitting reaction (>1.6 V). In addition, the Si surface is electrochemically oxidized or reduced when it comes into direct contact with the aqueous electrolyte. Here, we propose the hybridization of the photoelectrochemical device with a thermoelectric device, where the Seebeck voltage generated by the thermal energy triggers the self-biased water splitting reaction without compromising the photocurrent level at 42 mA cm−2. In this hybrid device p-Si, where the surface is protected by HfOx/SiOx bilayers, is used as a photocathode. The HfOx exhibits high corrosion resistance and protection ability, thereby ensuring stability. On applying the Seebeck voltage, the tunneling barrier of HfOx is placed at a negligible energy level in the electron transfer from Si to the electrolyte, showing charge transfer kinetics independent of the HfOx thickness. These findings serve as a proof-of-concept of the stable and high-efficiency production of hydrogen fuel by the photoelectrochemical-thermoelectric hybrid devices.
Solar water-splitting systems that directly convert solar energy into storable and transportable hydrogen fuel are a promising approach to sustainably providing cost-effective renewable energy1,2,3,4,5. In order to realize self-biased solar water splitting, intensive efforts have been made in the past decades to exploit designed prototype systems including photovoltaic-electrolysis combination cells6,7,8,9, tandem photoelectrochemical (PEC) cells using the dual-semiconductor light absorber10,11,12,13,14, and colloid-based “Z-scheme” photocatalytic cells15,16,17,18,19,20,21,22,23,24,25. These systems are normally better suited to generating an output voltage higher than the sum of thermodynamic potential (1.23 V) and kinetic overpotential (>0.3 V), which is required for a practical water-splitting reaction. However, it compromises the photocurrent level because a tradeoff relationship exists between the photovoltage and photocurrent in systems that operate based on the photovoltaic effect. This fundamental obstacle has limited the solar-to-hydrogen efficiency.
As a new approach to overcoming the aforementioned limitation, recently, the hybridization of the PEC cell with other energy-harvesting devices such as thermoelectric (TE) and piezoelectric generators—which generate an output voltage through the conversion of thermal and mechanical energy resources, respectively—has been proposed26,27,28. In particular, the PEC-TE hybrid device is expected to be a commercially viable approach because it can be operated with absorbed solar energy in the form of photons (using the PEC component) and phonons (using the TE component)26,29. The PEC-TE hybrid system—in which the PEC device is electrically connected with the TE device in series—could provide an additional output Seebeck voltage (VTE) as a result of a thermal gradient in the TE in order to overcome the potential barrier required for the water-splitting reaction. In this hybrid system, the overall current is determined by the high photocurrent level of the PEC cell (i.e., the TE is only used as a voltage booster). This feature of the PEC-TE hybrid system can decouple the tradeoff relation between the voltage and current, thus maximizing the H2 power generation (PMax).
Various light-absorbing semiconductors including narrow bandgap (Si, InP, and GaAs) and wide bandgap (TiOx, CuOx, FeOx, and BiVOx) materials have been used as photoelectrodes of PEC cell30. In the PEC-TE hybrid systems, because the output Seebeck voltage of the TE can further applied to drive the water splitting reaction, the narrow bandgap semiconductors with a higher photocurrent relative to the wide bandgap semiconductors are effective in maximizing the H2 power generation. Thus, Si with comparatively narrow bandgap (1.1 eV) is capable of generating a high photocurrent of ~44 mA·cm−2 via harnessing a large portion of the solar spectrum and is a suitable material as a light-absorbing semiconductor photoelectrode31. However, the poor corrosion-resistance of Si is an inherent drawback in terms of the long-term stable operation of the PEC reaction32. One suitable strategy for offsetting this drawback is to use a high-atomic-density insulating material as a protective layer that would enable the complete blocking of the electrolyte permeation33,34,35,36,37. In general, a high-atomic-density insulator of significant thickness provides excellent protection ability but results in an increase in the transfer resistance of the light-induced charge carriers owing to the tunneling barrier of the insulator38. Therefore, for Si PEC devices, it is of paramount importance to design a protection layer that guarantees both long-term stability and facile charge-transfer kinetics.
In this study, we present a PEC-TE hybrid system with a HfOx/SiOx bilayer protected Si photocathode to ensure its long-term stability without the deterioration of the charge-transfer kinetics. The PEC-TE hybrid device based on an antireflective Si nanostructure photocathode enables the self-biased PEC water-splitting reaction at a high photocurrent level of ~42 mA·cm−2, which has a maximum H2 power generation of 55 mW·cm−2. The HfOx, which has strong corrosion resistance in an acidic electrolyte and a relatively higher atomic density (9.68 g·cm−3) than other dielectric materials such as SiOx (2.65), AlOx (3.95), and TiOx (4.23), physically prevents the permeation of the electrolyte into the Si surface, thereby resulting in long-term stable PEC operation for 200 h. Moreover, on applying VTE, the electrical tunneling barrier of HfOx becomes irrelevant to the charge transport because the potential drop caused by the application of VTE mainly occurs in the SiOx layer, which induces the tunneling barrier of HfOx to be located at a level lower than the conduction band of Si. Therefore, VTE is not only applicable for driving a water-splitting reaction but also for reducing the kinetic overpotential required for charge transfer.
H-terminated and SiOx-grown p-Si (100) wafers were prepared through the treatment of diluted hydrofluoric acid and hot-water-oxidation, respectively, and these wafers were deposited with an amorphous HfOx thin film using the atomic layer deposition (ALD) process39,40,41. The chemical composition of HfOx was confirmed using energy-dispersive X-ray spectroscope (EDXS) mapping and X-ray photoelectron spectroscopy (XPS) spectra (See Supplementary Figs 1 and 2). Figure 1a,b show the high-resolution transmission electron microscope (HRTEM) images for the ALD HfOx thin film deposited on the H-terminated and SiOx-grown Si wafers. The HfOx deposited on H-terminated Si was observed to have a 3-nm-thick HfSiOx interfacial layer, which is caused by a diffusion of the Si atoms during the ALD (see Fig. 1a)42. In contrast, the interfacial layer was not found in the SiOx-grown Si. Figure 1b shows the TEM image for the atomically thin SiOx interlayer inserted between the HfOx and Si without the formation of a HfSiOx layer. This difference is likely due to the bond strength of Si–O (≈190 kcal·mol−1)43 being higher than that of Si–H (≈80 kcal·mol−1)44. The stronger Si–O binding formed by SiOx growth prevents the Si diffusion on the Si surface, thereby creating the HfOx/SiOx bilayer without formation of HfSiOx. The chemical quality of the interface between HfOx and Si is clearly confirmed by the angle-resolved X-ray photoelectron spectroscopy (ARXPS) results. For the Si2p XPS spectra at the low incident angle that represents the interface information, the H-terminated Si wafer has one dominant state of Si3+ originating from the HfSiOx interfacial layer (Fig. 1c), whereas the SiOx-grown Si wafer has various sub-oxide states of Si1+, Si2+, Si3+, and Si4+ (corresponding to Si2O, SiO, Si2O3, and SiO2, respectively; see Fig. 1d)45,46.
Linear sweep voltammetry (LSV) responses for the Si photocathodes protected with HfOx, SiOx, and HfOx/SiOx were obtained in 0.5-M H2SO4 to compare the charge transfer kinetics for the PEC water-splitting reaction (Fig. 2a). Two electrode systems consisting of a Si photocathode and Pt anode were used to determine the potential requirement for the self-biased water-splitting reaction. With a thermodynamic potential barrier of 1.23 eV, additional kinetic overpotential is required for the water-splitting reaction. Owing to the sluggish kinetics of the hydrogen evolution reaction (HER) and oxygen evolution reaction (OER), the application of a minimum potential of ~2.8 V vs. Pt is required for realizing a high photocurrent level of 30 mA·cm−2. The deposition of a 5-nm-thick HfOx results in an additional overpotential of 1.1 V at 30 mA·cm−2 owing to the tunneling barrier of HfOx, and the increase in the thickness to ~9 nm further increases the overpotential by 1.2 V (indicated by blue curves in Fig. 2a). Interestingly, on inserting a thin SiOx interlayer between the HfOx and Si (creating a HfOx/SiOx bilayer), the additional overpotential is completely reduced regardless of the HfOx thickness (indicated by the pink curves in Fig. 2a). The same behavior was observed in the dark LSV characteristics for the degenerated doped n-type (n+-Si) wafer, which is only attributed to the charge transfer kinetics (Supplementary Fig. 3). This result indicates that the insertion of SiOx makes the charge transport independent of the tunneling barrier of the HfOx, which is also verified by the electrochemical impedance spectra (Supplementary Fig. 4).
The negligible tunneling barrier can be explained by the effective engineering of an energy band after the insertion of SiOx under the application of a voltage for driving the PEC water-splitting reaction. When applying potential to the electrolyte/insulator/semiconductor configuration, the electric field can be built up across the insulator, thereby inducing a potential drop at the insulator layers47. When the potential is applied across the HfOx/SiOx bilayer, the potential drop can be expressed as
where E and t are the electric field and thickness, respectively. ESiOx and EHfOx are related as
The electric field is dominantly applied in the SiOx layer rather than in HfOx owing to a dielectric constant of SiOx (3.9) being much lower than that of HfOx (11)47. This allows a predominant potential drop across the SiOx layer. The potential drop across the SiOx layer under the application of potential results in a lying conduction band of HfOx that is relatively lower than that of Si, as described in Fig. 2b, which shows the band diagram of the HfOx/SiOx bilayer. As a result, a tunneling barrier of SiOx only affects the transport of the charged carrier in the HfOx/SiOx bilayer, and consequently, the effective tunneling distance becomes equal to the thickness of the SiOx monolayer. In contrast, the potential drop for the HfOx monolayer occurred entirely in the HfOx monolayer, thus creating a triangular barrier as described in Fig. 2c48. Because of the triangular barrier, the tunneling resistance is differentiated with the current density. In the case of a low current level (<~10 mA·cm−2), the charge carriers are mostly transferred through the upper region of the triangular barrier in order to minimize the tunneling distance and resistance. Consequently, in the low-current level, the overpotential induced by the tunneling resistance is independent of the thickness. However, in the case of a high photocurrent region (>~10 mA·cm−2), charge carriers transfer through all the regions of the barrier and suffer from a large tunneling distance and resistance. Therefore, the overpotential in the high-current level is greatly increased with the thickness.
It is also worth noting that the formation of the HfSiOx interfacial layer in the HfOx monolayer induces a hysteresis phenomenon owing to a negative charge trapping/detrapping mechanism at the interfacial layer49; the LSV curve is shifted into the anodic direction during the 10 cycles of the LSV scan (see Supplementary Fig. 5). This suggests that the HfSiOx interfacial layer with a high defect density acts as a recombination site for the charge carrier and also causes an increase in the charge transfer resistance. By contrast, the HfOx/SiOx bilayer prevented the formation of HfSiOx interfacial layer and the hysteresis. To observe the effect of the SiOx interlayer on the PEC performance of the bilayer protected Si photocathodes, the LSV responses were characterized for the samples with the wet-chemically-grown SiOx interlayer with oxidation times of 10 min (thin-SiOx) and 30 min (thick-SiOx) and with a thermally grown 1.8-nm-thick SiO2 interlayer (Supplementary Fig. 6). A wet-chemically-grown thin-SiOx shows the lowest overpotential as compared to the thicker one and the thermally grown SiO2 interlayer.
The engineering of the HfOx barrier height under the application of a voltage gives a further advantage in the PEC-TE hybrid system based on the HfOx/SiOx bilayer protected Si photocathode. As demonstrated in previous works, a hybrid system in which a PEC device is electrically combined in series with a TE device could promote a self-biased solar water-splitting reaction. The Seebeck voltage generated in the TE device owing to the thermal gradient (VTE) causes the Fermi level of the counter electrodes to be lower than the oxygen evolution level, as depicted in Fig. 3a. This can effectively offset the potential required for driving the self-biased water-splitting reaction while the overall current is determined by the photocurrent of the PEC cell (i.e., the TE is only used as a voltage source)26. This was proven by the LSV responses with the driven VTE; the LSV curve for the bare Si photocathode (with DHF treatments) was shifted in the anodic direction in proportion to the applied VTE (see Supplementary Fig. 7). Interestingly, in the case of the HfOx/SiOx bilayer protected Si photocathode, an anodic shift degree is larger than the applied VTE, and the overpotential is further reduced by 350 mV at a 70 °C temperature gradient (see Fig. 3b). To clearly differentiate the shifts, the anodic shift degrees of the LSV curve (reducing overpotential) at a photocurrent of 30 mA·cm−2 were plotted as a function of ΔT along with the VTE (Fig. 3c). It should be noted that the slope of the linear plot for VTE is the Seebeck coefficient and is represented by the relation VTE = SΔT. It is apparent that the slop for the HfOx/SiOx bilayer protected Si photocathode (41 mV·K−1) is much greater than that for VTE (35 mV·K−1). For the purpose of comparison, we also characterized the slope for the Si photocathodes grown with monolayers of HfOx and SiOx (see Supplementary Fig. 8 for the LSV curves). The SiOx has a slope (36 mV·K−1) that is analogous with that of VTE. In contrast, the slope for the HfOx is much greater than that for VTE but smaller than that for the HfOx/SiOx bilayer. These results are obtained owing to a different tunneling barrier height and potential drop under the applied VTE (see Supplementary Fig. 9 for the band diagrams).
As a result of applying VTE, a saturated photocurrent of 34 mA·cm−2 was obtained at a zero potential (i.e., external bias-free). In the PEC-TE hybrid device, an improvement in the photocurrent level is directly correlated with the solar-to-hydrogen conversion efficiency. Thus, we used antireflective Si nanostructures on Si photocathode, which completely absorbed the incident light50,51 and greatly improved external quantum efficiency (EQE) for the light-induced charge carrier, maximizing the photocurrent of 42 mA·cm−2 (see Fig. 3d and Supplementary Figs 10 and 11). While the PEC-TE hybrid system with planar Si provided the maximum H2 power generation (PMax) of 42 mW·cm−2, the antireflective Si nanostructure photocathode achieved a PMax of 55 mW·cm−2, which is four times greater than that of a photovoltaic-electrolysis combination (PMax of ~12 mW·cm−2 indicated by the grey line in Fig. 3d).
The long-term stabilities of the HfOx/SiOx bilayer protected Si photocathode were performed using chronoamperometry at a photocurrent level of 10 mA cm−2 and using chronopotentiometry at an applied potential of −3 V. Due to the HfOx thin film which has an (electro)chemical stability in the acidic electrolyte and transparency for all wavelength region of the incident light, the Si photocathode exhibited a stable water-splitting reaction for over 200 h at high photocurrent level of 33 mA/cm2 (Fig. 4a). The stability achieved at high photocurrent by HfOx thin film is the best result compared to other metal oxide thin films (see Supplementary Table 1). TiOx thin films chemically stable in the acidic and alkaline electrolytes have been widely used as a protection layer and have demonstrated long-term stability over 960 hr52. However, it has been recently reported ultraviolet (UV) light absorbed in TiOx causes the electrochemical reduction of TiOx53. Because of this issue, the TiOx protected Si photocathode operates reliably only when there is no UV irradiation, resulting in low photocurrent. Under the stable operation, the H2 evolution was characterized by measuring a volume of the H2 gas collected under a simulated AM 1.5 G 1-Sun illumination while operating under a certain photocurrent level of 32 mA·cm−2 for 50 min (Supplementary Fig. 12). The measured H2 volumes were corresponded well with the theoretically calculated values at a 100% Faraday efficiency based on the total charge passed. After the PEC operation for 200 h, interestingly, the potential was finally shifted in the anodic direction by 200 mV, which was identified by the anodic shift of the LSV curve (see Fig. 4b). This was found to be the result of the crystallization of the as-deposited a-HfOx during the PEC water-splitting reaction as observed in the TEM images and X-ray diffraction (XRD) results (Supplementary Fig. 13) of the HfOx layer before and after the 200 h stability test (see Fig. 4c,d). Because of a bulk defect state in the a-HfOx that trapped the charge carrier, a substantial resistance was induced for the charge carrier across the a-HfOx, as depicted in Fig. 4e. In contrast, the crystallized HfOx significantly reduced the bulk defect density in the a-HfOx in order to improve the charge transfer kinetics. The crystallization during the solar water splitting reaction at the room temperature condition is an intriguing phenomenon, as it is known to crystallize a-HfOx in the monoclinic phase at temperature of 500 °C54. This can be attributed to an electroforming phenomenon wherein the transition metal oxide can be reformed under the electrical field55. This has been observed in transition-metal-oxide-based solid-state electronic devices. This finding presents a new method of improving the insulator quality used in PEC devices.
We have presented the PEC-TE hybrid device using HfOx/SiOx bilayer protected the Si photocathode in which the HfOx thin film has a strong corrosion-resistance against the acidic electrolyte. The application of the Seebeck voltage generated in the TE device to the PEC device not only drove the self-biased solar water-splitting reaction but also promoted the ease of charge transfer through the HfOx tunneling barrier. On applying the Seeback voltage to the HfOx/SiOx bilayer/Si photocathode, the conduction band of HfOx reached an energy level lower than the conduction band of Si, thereby allowing the electron transfer to become independent of the HfOx tunneling barrier. Because of the kinetic benefit, the voltage required for the self-biased water-splitting reaction of the Si photocathode is lowered by 0.35 V. As a result, the planar-Si-photocathode-based PEC-TE hybrid realized the self-biased water-splitting reaction with the evolved H2 power of 40 mW cm−2 at a temperature difference of 70 °C. By adopting antireflective Si nanostructures that improve the photocurrent level to 42 mA·cm−2, the stored H2 power reached ~52 mW·cm−2. The HfOx thin-film protection ensured long-term stable PEC operation for over 200 h. During the stability test, we observed the phase transformation of HfOx from the amorphous structure into a crystalline structure, which reduces the bulk defect density and thus the kinetic overpotential for the charge transfer through the HfOx layer.
Preparation of HfOx thin-film protected Si photocathodes. A single-crystalline Si wafer with a resistivity of 1–10 Ω cm and degenerately doped n+-type Si (100) wafer with a resistivity of 0.001–0.002 Ω cm were used to characterize the PEC response and charge transfer kinetics, respectively. Prior to the deposition of HfOx, the Si wafer was dipped into diluted hydrofluoric acid to form an H-terminated surface while removing the native oxide. For growing the SiOx thin film on the surface of the Si wafer, the H-terminated Si wafer is oxidized through treatment with hot water at 90 °C for 10–30 min. ALD of the HfOx thin film was performed for both the H-terminated and SiOx-grown Si wafers to obtain the HfOx monolayer and HfOx/SiOx bilayer coated Si wafers, respectively. An HfOx thin film was deposited at 280 °C in a 4-inch traveling-wave type ALD reactor (CN-1 Co.) using Tetrakis (ethylmethylamino) hafnium (TEMAHf) and H2O precursors with a carrier gas of high purity N2 (99.999%). The TEMAHf, N2, and H2O were sequentially injected for 2.5, 30, and 1.5 s, respectively.
Characterization of HfOx thin films. A HRTEM (JEOL, JEM-2100F, Japan) equipped with an EDXS operated at an accelerating voltage of 200 kV was used to observe the morphology and elemental composition of the HfOx-thin-film-coated Si wafers. The thicknesses of the HfOx and SiOx thin films were measured using spectroscopic ellipsometry (Sopra GES 5E, fitted to a Tauc–Lorentz function using the Cauchey model). The chemical states of the HfOx and SiOx thin films were investigated using ARXPS equipped with a monochromatic Al Kα (1.486 eV) source.
Setup of PEC-TE hybrid devices. The PEC device consists of the Si photocathode and Pt mesh anode for the hydrogen evolution reaction (HER) and oxygen evolution reaction (OER), respectively. The back of the Si photocathode was treated with DHF soaking and subsequent scribing using an In–Ga eutectic alloy (Sigma–Aldrich) to fabricate back contacts with a Cu electrode. PEC-TE hybrid systems were used by electrically connecting the PEC device with a TE device in series; the Si photocathode is connected to an anode of the TE device, and a cathode of the TE device is connected to the Pt anode. TE devices with a Seebeck coefficient of 35 mV·K−1 and internal resistances of 2.1 Ω, which are commercially available from Kryotherm (Saint-Petersburg, Russia), were employed.
Characterizations of PEC performances. The PEC properties of the Si photocathodes were measured using a potentiostat (Iviumstat, Eindhoven, Netherlands) with a two-electrode configuration (Si photocathode and Pt mesh anode). The PEC responses were characterized using various methods of LSV, chronopotentiometry, and chronoamperometry in 0.5-M sulfuric acid under a 100-mW·cm−2 illumination (Xenon lamp equipped with AM 1.5 G filter) and calibrated by a Si photodiode standard cell (PV Measurements, Inc.). The LSV responses were measured under dark and illuminated conditions at a scan rate of 50 mV·s−1. The stability and H2 evolution measurements were performed using chronopotentiometry and chronoamperometry. Home-made heating and cooling systems were fabricated to control the temperature and were used for applying a constant temperature difference between the front and back of the TE device. The temperature difference was monitored using K-type thermocouples (Center 306 Data Logger, New Taipei City Taiwan). The quantity of hydrogen evolved at the Si photocathode under illumination and the Seebeck voltage application was measured using the volume displacement method wherein the volume of a water-filled flask changes with the evolution of hydrogen gas. A Pt counter electrode covered with a quartz cylinder and proton exchange membrane was used in order to separately collect hydrogen and oxygen gas. All the PEC-TE hybrid system is shown in Supplementary Fig. 14 as a digital image.
The data supporting the findings of this study are available within the article and its Supplementary Information files. All other relevant source data are available from the corresponding author upon reasonable request.
Fujishima, A. & Honda, K. Electrochemical photolysis of water at a semiconductor electrode. Nature 238, 37–38 (1972).
Grätzel, M. Photoelectrochemical cells. Nature 414, 338–344 (2001).
Walter, M. G. et al. Solar Water Splitting Cells. Chem. Rev. 111, 5815–5815 (2011).
Tachibana, Y., Vayssieres, L. & Durrant, J. R. Artificial photosynthesis for solar water-splitting. Nat. Photon. 6, 511–518 (2012).
Kudo, A. & Miseki, Y. Heterogeneous photocatalyst materials for water splitting. Chem. Soc. Rev. 38, 253–278 (2009).
Jia, J. et al. Solar water splitting by photovoltaic-electrolysis with a solar-to-hydrogen efficiency over 30%. Nat. Commun. 7, 13237 (2016).
Cox, C. R., Lee, J. Z., Nocera, D. G. & Buonassisi, T. Ten-percent solar-to-fuel conversion with nonprecious materials. Proc. Natl. Acad. Sci. USA 111, 14057–14061 (2014).
Luo, J. et al. Water photolysis at 12.3% efficiency via perovskite photovoltaics and earth-abundant catalysts. Science 345, 1593–1596 (2014).
Kim, S. et al. A highly versatile and adaptable artificial leaf with floatability and planar compact design applicable in various natural environments. Adv. Mater. 29, 1702431 (2017).
Reece, S. Y. et al. Wireless solar water splitting using silicon-based semiconductors and earth-abundant catalysts. Science 334, 645–648 (2011).
Jang, J.-W. et al. Enabling unassisted solar water splitting by iron oxide and silicon. Nat. Commun. 6, 7447 (2015).
Brillet, J. et al. Highly efficient water splitting by a dual-absorber tandem cell. Nat. Photon. 6, 824–828 (2012).
Abdi, F. F. et al. Efficient solar water splitting by enhanced charge separation in a bismuth vanadate-silicon tandem photoelectrode. Nat. Commun. 4, 2195 (2013).
Kempler, P. A., Gonzalez, M. A., Papadantonakis, K. M. & Lewis, N. S. Hydrogen evolution with minimal parasitic light absorption by dense Co–P catalyst films on structured p-Si photocathodes. ACS Energy Lett. 3, 612–617 (2018).
Chen, J. Q. et al. Recent progress in enhancing solar-to-hydrogen efficiency. J. Power Sources 280, 649–666 (2015).
Iwase, A., Ng, Y. H., Ishiguro, Y., Kudo, A. & Amal, R. Reduced grapheme oxide as a solid-state electron mediator in Z-scheme photocatalytic water splitting under visible light. J. Am. Chem. Soc. 133, 11054–11057 (2011).
Wang, Q. et al. Scalable water splitting on particulate photocatalyst sheets with a solar-to-hydrogen energy conversion efficiency exceeding 1%. Nat. Mater. 15, 611–615 (2016).
Liu, J. et al. Metal-free efficient photocatalyst for stable visible water splitting via a two-electron pathway. Science 347, 970–974 (2015).
Yang, X. et al. Interfacial optimization of g-C3N4-based Z-scheme heterojunction toward synergistic enhancement of solar-driven photocatalytic oxygen evolution. Appl. Catal. B Environ. 244, 240–249 (2019).
Liu, Q., Shen, J., Yang, X., Zhang, T. & Tang, H. 3D reduced graphene oxide aerogel-mediated Z-scheme photocatalytic system for highly efficient solar-driven water oxidation and removal of antibiotics. Appl. Catal. B Environ. 232, 562–573 (2018).
Liu, Q. et al. Unveiling the origin of boosted photocatalytic hydrogen evolution in simultaneously (S, P, O)-codoped and exfoliated ultrathin g-C3N4 nanosheets. Appl. Catal. B Environ. 248, 84–94 (2019).
Liu, W., Shen, J., Yang, X., Liu, Q. & Tang, H. Dual Z-scheme g-C3N4/Ag3PO4/Ag2MoO4 ternary composite photocatalyst for solar oxygen evolution from water splitting. Appl. Surf. Sci. 456, 369–378 (2018).
Lian, Z. et al. Plasmonic silver quantum dots coupled with hierarchical TiO2 nanotube arrays photoelectrodes for efficient visible-light photoelectrocatalytic hydrogen evolution. Sci. Rep. 5, 10461 (2015).
Lian, Z. et al. Photoelectrocatalytic reduction of CO2 to methanol over a photosystem II-enhanced Cu foam/Si-nanowire system. J. Environ. Sci. 60, 108–113 (2017).
Wang, W., Li, F., Zhang, D., Leung, D. Y. C. & Li, G. Photoelectrocatalytic hydrogen generation and simultaneous degradation of organic pollutant via CdSe/TiO2 nanotube arrays. Appl. Surf. Sci. 362, 490–497 (2016).
Shin, S.-M., Jung, J.-Y., Park, M.-J., Song, J.-W. & Lee, J.-H. Catalyst-free hydrogen evolution of Si photocathode by thermovoltage-driven solar water splitting. J. Power Sources 279, 151–156 (2015).
Wei, A. et al. Triboelectric nanogenerator driven self-powered photoelectrochemical water splitting based on hematite photoanodes. ACS Nano 12, 8625–8632 (2018).
Li, T. et al. Boosting photoelectrochemical water splitting by TENG‐charged Li‐ion battery. Adv. Energy Mater. 7, 1700124 (2017).
Park, K.-T. et al. Lossless hybridization between photovoltaic and thermoelectric devices. Sci. Rep. 3, 2123 (2013).
Kim, J. H., Hansora, D., Sharma, P., Jang, J.-W. & Lee, J. L. Toward practical solar hydrogen production – an artificial photosynthetic leaf-to-farm challenge. Chem. Soc. Rev. 48, 1908–1971 (2019).
Sun, K. et al. Enabling silicon for solar-fuel production. Chem. Rev. 114, 8662–8719 (2014).
Bae, D., Seger, B., Vesborg, P. C., Hansen, O. & Chorkendorff, I. Strategies for stable water splitting via protected photoelectrodes. Chem. Soc. Rev. 46, 1933–1954 (2017).
Hu, S. et al. Amorphous TiO2 coatings stabilize Si, GaAs, and GaP photoanodes for efficient water oxidation. Science 344, 1005–1009 (2014).
Choi, M. J. et al. Long-term durable silicon photocathode protected by a thin Al2O3/SiOx layer for photoelectrochemical hydrogen evolution. J. Mater. Chem. A 2, 2928–2933 (2014).
Sun, K. et al. Stable solar-driven water oxidation to O2(g) by Ni-oxide-coated silicon photoanodes. J. Phys. Chem. Lett. 6, 592–598 (2015).
Yu, Y. et al. Enhanced photoelectrochemical efficiency and stability using a conformal TiO2 film on a black silicon photoanode. Nat. Energy 2, 17045 (2017).
Kast, M. G. et al. SnO2/TiO2 as a base-stable protective layer and antireflective coating for microtextured buried-junction H2-evolving Si photocathodes. ACS Appl. Mater. Inter. 6, 22830–22837 (2014).
Scheuermann, A. G. et al. Design principles for maximizing photovoltage in metal-oxide-protected water-splitting photoanodes. Nat. Mater. 15, 99–105 (2016).
Lau, C., Srimani, T., Bishop, M. D., Hills, G. & Shulaker, M. M. Tunable n-type doping of carbon nanotubes through engineered atomic layer deposition HfOX films. ACS Nano 12, 10924–10931 (2018).
Tian, H. et al. Cost-effective, transfer-free, flexible resistive random access memory using laser-scribed reduced graphene oxide patterning technology. Nano Lett. 14, 3214–3219 (2014).
Carlos, E. Boosting electrical performance of high-κ nanomultilayer dielectrics and electronic devices by combining solution combustion synthesis and UV irradiation. ACS Appl. Mater. Inter. 9, 40428–40437 (2017).
Wang, L., Xue, K., Xu, J. B., Huang, A. P. & Chu, P. K. Control of interfacial silicate between HfO2 and Si by high concentration ozone. Appl. Phys. Lett. 88, 072903 (2006).
Seidel, H., Csepregi, L., Heuberger, A. & Baumgartel, H. Anisotropic etching of crystalline silicon in alkaline solutions. I. Orientation dependence and behavior of passivation layers. J. Electrochem. Soc. 137, 3612–3626 (1991).
Cicero, R. L., Linford, M. R. & Chidsey, C. E. D. Photoreactivity of unsaturated compounds with hydrogen-terminated silicon(111). Langmuir 16, 5688–5695 (2000).
Oniki, Y., Iwazaki, Y., Hasumi, M., Ueno, T. & Kuroiwa, K. HfO2/Si and HfSiO/Si structures fabricated by oxidation of metal thin films. Jpn. J. Appl. Phys. 48, 05DA01 (2009).
Miyata, N., Abe, Y. & Yasuda, T. Effect of interfacial Si oxidation on interface dipoles in HfO2/Si structures. J. Phys. D: Appl. Phys. 46, 315304 (2013).
Chakrabarti, B. et al. Investigation of tunneling current in SiO2/HfO2 gate stacks for flash memory applications. IEEE Trans. Electron Devices 58, 4189–4195 (2011).
Lana-Villarreal, T., Straboni, A., Pichon, L. & Alonso-Vante, N. Photoelectrochemical characterization of p-type silicon electrodes covered with tunnelling. Thin Solid Films 515, 7376–7381 (2007).
Shen, J. et al. Hysteresis-free HfO2 film grown by atomic layer deposition at low temperature. Thin Solid Films 519, 7723–7726 (2011).
Jung, J.-Y. et al. Photoelectrochemical water splitting employing a tapered silicon nanohole array. J. Mater. Chem. A 2, 833–842 (2014).
Jung, J.-Y. et al. A strong antireflective solar cell prepared by tapering silicon nanowires. Opt. Express. 18, A286–A292 (2010).
Bae, D. et al. Carrier-selective p- and n-contacts for efficient and stable photocatalytic water reduction. Catal. Today 15, 59–64 (2016).
Wang, T. et al. Transparent Ta2O5 protective layer for stable silicon photocathode under full solar spectrum. Ind. Eng. Chem. Res. 58, 5510–5515 (2019).
Park, P. K. & Kang, S. W. Enhancement of dielectric constant in HfO2 thin films by the addition of Al2O3. Appl. Phys. Lett. 89, 192905 (2006).
Kim, W. G. et al. Dependence of the switching characteristics of resistance random access memory on the type of transition metal oxide: TiO2, ZrO2, and HfO2. J. Electrochem. Soc. 158, H417–H422 (2011).
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. 2017R1A2B3006941), and the International Collaborative Energy Technology R&D Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (No. 20168520011370).
The authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Jung, JY., Woong Kim, D., Kim, DH. et al. Seebeck-voltage-triggered self-biased photoelectrochemical water splitting using HfOx/SiOx bi-layer protected Si photocathodes. Sci Rep 9, 9132 (2019). https://doi.org/10.1038/s41598-019-45672-4
Strategies and implications of atomic layer deposition in photoelectrochemical water splitting: Recent advances and prospects
Nano Energy (2021)
Comprehensive view on recent developments in hydrogen evolution using MoS2 on a Si photocathode: from electronic to electrochemical aspects
Journal of Materials Chemistry A (2021)
Nano-Micro Letters (2020)
Design Guidelines of Insulator for Improving Stability and Performance of Nanoelectrocatalyst/Insulator/Semiconductor Photoelectrochemical Cells
ACS Applied Energy Materials (2020)
Scientific Reports (2020)