Large onset potential improvement with an epitaxial GaAs/Si photocathode for solar H2 production.
Résumé
III-V compounds semiconductors are excellent candidates for unassisted solar water splitting due to their appropriate band edge energies and their superior optical and transport properties [1,2]. The most significant limitations of III-Vs for photoelectrochemical (PEC) applications are the high substrate cost, their huge overpotentials and their reduced lifetime under operation. Previous studies suggest that the use of photoelectrodes made of thin III-Vs layer deposited on Si substrates with adapted protection strategies can be used to resolve corrosion, cost and resources constraints [3-6]. To address these issues, in this work, we report on the photocathode performance of 1 µm-thick GaAs layers grown on a low-cost Si p-doped substrate by MBE (Molecular Beam Epitaxy) and compare them to those of GaAs:p doped wafers. The photocathodes were investigated in 0.2 M H2SO4 (aq) electrolyte under 1 sun (100 mW/cm2 ) illumination. The onset potential (Vonset) of bare GaAs/Si and bare GaAs:p wafer is quite comparable, at around -0.2 V vs reversible hydrogen electrode (RHE). A significant reduction of surface states density with a sulfur passivation (S-passivation) is demonstrated but with very limited changes on the Vonset value, for both samples. In contrast, the deposition of thin Pt catalyst layers by electroless deposition technique leads to a large positive shift Vonset. The Vonset of GaAs/Si and GaAs:p wafer photocathodes are increased by up to 0.34 and 0.15 V vs RHE respectively. We show that the Vonset of GaAs/Si can be further improved by combining the Pt catalyst and the S-passivation process, reaching 0.4 V vs RHE, a record value for GaAs-based Schottky-like photocathodes, despite the presence of numerous crystalline defects. STEM-EDX analysis, XPS measurements were conducted to further understand this phenomenon. The results show that the Pt layer on GaAs:p wafer is homogeneous and shows an As0-rich Pt surface that can explain the weaker catalytic activity in comparison with GaAs/Si. Finally, the stability of Pt/GaAs/Si photocathodes was evaluated, and a lifetime larger than 112 h was established. These findings provide guidance for further studies towards the fabrication of scalable, cost-efficient and stable unassisted PEC cells for green hydrogen production.
References:
[1] S. Tiwari and D. J. Frank, “Empirical fit to band discontinuities and barrier heights in III–V alloy systems,” Appl. Phys. Lett., vol. 60, no. 5, pp. 630–632, Feb. 1992, doi: 10.1063/1.106575.
[2] G. Siddiqi, Z. Pan, and S. Hu, “III–V Semiconductor Photoelectrodes,” in Semiconductors and Semimetals, vol. 97, Elsevier, 2017, pp. 81–138. doi: 10.1016/bs.semsem.2017.03.002.
[3] M. Alqahtani et al., “Photoelectrochemical water oxidation of GaP 1−x Sb x with a direct band gap of 1.65 eV for full spectrum solar energy harvesting,” Sustain. Energy Fuels, vol. 3, no. 7, pp. 1720–1729, 2019, doi: 10.1039/C9SE00113A.
[4] L. Chen et al., “Epitaxial III–V/Si Vertical Heterostructures with Hybrid 2D-Semimetal/Semiconductor Ambipolar and Photoactive Properties,” Adv. Sci., vol. 9, no. 2, p. 2101661, 2022, doi: 10.1002/advs.202101661.
[5] M. Piriyev et al., “Dual bandgap operation of a GaAs/Si photoelectrode,” Sol. Energy Mater. Sol. Cells, vol. 251, p. 112138, Mar. 2023, doi: 10.1016/j.solmat.2022.112138.
[6] M. Piriyev et al., “Photoelectrode/electrolyte interfacial band lineup engineering with alloyed III–V thin films grown on Si substrates,” J. Mater. Chem. C, vol. 12, no. 3, pp. 1091–1097, Jan. 2024, doi: 10.1039/D3TC02556J.