The conversion and storage of solar energy into storable forms, such as hydrogen fuel, is a crucial goal for sustainable energy. Photoelectrochemical (PEC) water splitting, which directly converts sunlight into hydrogen and oxygen, offers a promising solution. Silicon (Si), with its ideal bandgap for solar spectrum absorption, is a frequently studied material for PEC applications. However, several challenges hinder Si's effectiveness. These include the need for light-blocking electrocatalysts, Si's limited bandgap insufficient to straddle water's reduction and oxidation potentials, and the conventional design where light absorption, surface protection, and electrolysis all occur on the same side, leading to parasitic light absorption and reduced efficiency. To overcome these limitations, researchers have explored spatial decoupling of light absorption and catalytic activity. Previous approaches, such as using Si micro/nanopillars with selectively integrated catalysts, are complex. Further complicating matters is the need for a potential of ~1.7-1.8 V for unassisted water splitting, exceeding Si's single photoelectrode capabilities. Tandem designs have been proposed, but these are hindered by material compatibility issues. This research proposes a novel back-buried junction (BBJ) design to address these challenges, aiming to create a more efficient and practical solar water-splitting system.

The existing literature highlights the challenges and limitations of using silicon for photoelectrochemical (PEC) water splitting. Researchers have explored various strategies to enhance efficiency, including the use of nanostructures, surface coatings, and tandem cells. While some success has been achieved in improving individual aspects of the PEC process, simultaneously optimizing light absorption, charge separation, and catalysis remains a significant hurdle. The integration of electrocatalysts often leads to light blockage, reducing the overall efficiency. The use of tandem cells, while potentially offering higher voltages, introduces complexities related to material compatibility and fabrication. This study addresses these issues by proposing a novel back-buried junction design that spatially separates light harvesting and catalysis, thus allowing for independent optimization of each component.

The researchers fabricated a back-buried junction (BBJ) PEC cell using a single-crystalline silicon wafer. The front side was optimized for light harvesting with micropyramidal texturing and an anti-reflective (AR) SiNx layer, eliminating the shadowing effect of catalysts. The back side served as the electrocatalysis surface. Boron and phosphorus were implanted in an interdigitated pattern to create a p⁺-Si emitter and an n⁺-Si back surface field (BSF), enhancing carrier collection. A thin P layer served as a front-surface field (FSF) to reduce surface recombination. Interdigitated Pt (for HER) and Ni (for OER) catalysts were deposited on the back side, aligned with the dopant pattern. The device's optical characteristics were analyzed using ultraviolet-visible spectrophotometry, measuring absorbance, reflectance, and transmittance. The PEC performance was characterized using linear sweep voltammetry (LSV) under simulated AM 1.5 G illumination, with potentials calculated relative to the reversible hydrogen electrode (RHE). Experiments were conducted under various pH conditions to assess the cell's versatility. Temperature and angle-of-incidence (AOI) dependence were also studied. A comprehensive current loss analysis was performed to identify optical, electrical, and catalytic losses. Finally, three BBJ-PEC cells were monolithically integrated in series to achieve unassisted water splitting. Gas chromatography (GC) was used to quantify hydrogen production, and long-term stability tests were conducted with and without a protective TiO2 layer deposited via atomic layer deposition (ALD). A Pt-sputtered/Ti-foil was used in the stability test for improved protection.

The BBJ-PEC cell demonstrated exceptional performance. Over 95% of the solar spectrum was absorbed (300-1100 nm). A single BBJ-PEC half-cell achieved a current density of 40.51 mA cm⁻² for HER, with minimal optical (6.11 mA cm⁻²), electrical (1.76 mA cm⁻²), and catalytic (1.67 mA cm⁻²) losses. Three BBJ-PEC cells connected in series produced an open-circuit potential of ~1.83 V, exceeding the thermodynamic requirement for unassisted water splitting. The resulting unassisted solar water splitting achieved a solar-to-hydrogen (STH) efficiency of 15.62%, the highest reported for Si-based unassisted PEC. Using Red Sea water, an STH efficiency of 11.33% was achieved. Temperature and AOI studies showed minor performance degradation (<14% for JH+/H2 and <8% for VOC-E over a wide range). Long-term stability tests showed that with ALD TiO2 surface protection, the system operated for 40 hours. Employing a Pt-sputtered/Ti-foil, a 100-hour stability test showed less than 10% drop in J H+/H2, demonstrating high stability. Gas chromatography measurements confirmed the hydrogen production and high Faradaic efficiency (>90%).

The results demonstrate the success of the spatial decoupling strategy in achieving highly efficient and stable solar water splitting using silicon. The independent optimization of light absorption and catalysis leads to significantly reduced losses compared to conventional PEC designs. The monolithic integration of multiple BBJ cells enables unassisted water splitting, making the technology more practical. The high STH efficiency achieved in this study surpasses previous results for Si-based unassisted PEC, highlighting the potential of this approach for large-scale solar hydrogen production. The analysis of temperature and AOI dependence provides valuable insights into the robustness of the device under real-world conditions. The observed limitations in long-term stability highlight the need for further investigation into surface protection strategies. The demonstration of stable operation over 100 hours with modified surface protection method showcases a clear pathway toward making the technology commercially viable.

This research presents a significant advancement in solar water splitting technology. The novel BBJ design with decoupled light absorption and catalysis significantly improves efficiency and stability. The achieved high STH efficiency (15.62%) for unassisted water splitting is a remarkable milestone, demonstrating the potential for practical solar hydrogen production. Further research could focus on optimizing the electrocatalysts, improving surface protection strategies, and scaling up the device for large-scale applications. The exploration of different semiconductor materials and integration with other energy storage systems would further expand the scope of this technology.

The primary limitation of the study is the long-term stability of the device, particularly under high current densities. While the use of a protective TiO2 layer improved stability, further investigation is needed to develop even more robust surface protection strategies. Another limitation is the use of relatively expensive Pt as the electrocatalyst for HER; future research should explore the use of more cost-effective catalysts. The current study focuses on hydrogen evolution; additional research should explore the simultaneous optimization of both HER and OER for complete water splitting.