Jinhyeong
Kwon†
a,
Seonmi
Ko†
b,
Hyeonwoo
Kim
b,
Hyo Jin
Park
b,
Changwook
Lee
b and
Junyeob
Yeo
*bc
aLaser-processed Nanomaterials Engineering Lab., Research Institute of Sustainable Development Technology, Korea Institute of Industrial Technology (KITECH), 89 Yangdaegiro-gil, Ipjang-myeon, Seobuk-gu, Chungchengnam-do, Cheonan, 31056, Republic of Korea
bNovel Applied Nano Optics (NANO) Lab, Department of Physics, Kyungpook National University, 80 Daehak-ro, Buk-gu, Daegu, 41566, Republic of Korea. E-mail: junyeob@knu.ac.kr
cDepartment of Hydrogen & Renewable Energy, Kyungpook National University, 80 Daehak-ro, Buk-gu, Daegu, 41566, Republic of Korea
First published on 12th April 2024
The demand for zero-emission energy sources has recently intensified, driven by growing concerns over global warming and climate change. Among various renewable energy technologies, solar water-splitting cells have emerged as a promising avenue for harnessing hydrogen energy by exciting photo-carriers. These cells are typically fabricated using vacuum-based techniques such as physical vapour deposition, atomic layer deposition, chemical vapour deposition, and sputtering. Although these methods deliver moderate performance, they are characterised by extensive production steps and significant costs. This review examines the current vacuum-based fabrication processes for solar water-splitting cells and explores advanced or alternative techniques, with a particular emphasis on optical methods, especially laser processing, for fabricating these cells.
Hydrogen can be sourced from plentiful water or fossil fuels, with the production methods categorised into grey, blue, and green hydrogen routes.14–16 Grey hydrogen is produced through the catalytic chemical reactions between methane, the primary component of natural gas, and high-temperature steam. This process yields hydrogen and carbon dioxide, with approximately 10 kg of carbon dioxide emitted for every 1 kg of hydrogen produced. Blue hydrogen utilises the same production process as grey hydrogen but incorporates carbon capture and storage (CCS) technology to capture and sequester the carbon dioxide generated during production rather than releasing it into the atmosphere.16 This makes blue hydrogen more environmentally friendly than grey hydrogen, resulting in lower carbon dioxide emissions. The CCS technology is relatively mature and competitive, positioning blue hydrogen as the most practical alternative, though it does not completely eliminate carbon dioxide emissions. Green hydrogen, in contrast, is generated by electrolysing water using electrical energy derived from renewable sources, such as solar or wind power, to separate water into hydrogen and oxygen.17 Therefore, it is regarded as the “ultimate environmentally friendly hydrogen” since it does not produce carbon dioxide emissions during its production. However, due to the high costs associated with renewable energy and the inefficiency of current electrolysis facilities, green hydrogen production costs are higher than grey hydrogen. Despite these challenges, numerous efforts are underway to transition to a green hydrogen society, including developing solar water-splitting cells.18,19
This review explores the manufacturing techniques of solar water-splitting cells, focusing on studies from 2020 to 2023. It begins with an overview of vacuum-based fabrication methods, prevalent in most manufacturing techniques, and introduces recent advancements in optical-based processes, especially those involving lasers (Fig. 1). This section underscores the innovative aspects of using lasers in manufacturing and explores potential future developments.
Semiconductor materials typically absorb photons with energies exceeding their bandgaps, leading to the generation of electron–hole pairs. This absorption excites electrons from the valence band (VB) to the conduction band (CB), creating charge carriers while holes remaining in the VB. The excited electrons transition to the CB, becoming free electrons. These free electrons are then transferred to the working electrode (photoanodes for n-type semiconductor materials and photocathodes for p-type semiconductor materials),42 as depicted in Fig. 2. The working electrode is typically modified with materials that facilitate water reduction to hydrogen gas (2H2O → 2H2 + 2e−). Simultaneously, holes in the VB migrate to the counter electrode, contributing to water oxidation and oxygen gas production (2H2O → O2 + 4H+ + 4e−).43 The gases generated can be harvested as renewable energy sources. Enhancing PEC efficiency involves surface modification to prevent electron–hole pair recombination.
Fig. 2 Energy diagrams of solar water-splitting PEC cells for (a) photoanode, (b) photocathode, and (c) combined photoanode and photocathode configurations. |
The electrolyte completes the circuit in the solution by enabling ion flow between the electrodes. Its pH significantly influences PEC efficiency,44 with H+ and OH− concentrations in the electrolyte critical for the acceleration of reduction or oxidation processes.45,46 While numerous factors are at play, the selection of suitable materials and surface processing techniques is key to the overall efficiency of water-splitting PEC reactions. Consequently, substantial efforts focus on employing various nanoscale materials and precision processing tools to advance PEC technology.47 Metal oxides with designed nanostructures are preferred, as nanomaterials offer significant benefits due to their size-dependent properties, including chemical, electrical, and optical characteristics, along with a high surface-to-volume ratio. Complex nanostructured designs provide ample sites for chemical reactions, reduce defects, and shorten electron–hole migration distances. This minimises electron–hole recombination and significantly boosts photoresponse efficiency.48–50
To enhance their surface area, nanomaterials have been intentionally designed in various forms and structures, such as thin films, porous frameworks, nature-mimicking branching systems, and hierarchical configurations. Among these, hierarchical metal-oxide nanostructures have garnered significant attention for electrochemical applications because of their straightforward core-and-shell structure. This has led to an increased focus on developing fabrication methods for the core materials.51–53 Historically, vapour–liquid–solid processes, often conducted under vacuum conditions, were the primary methods for synthesising metal-oxide-based hierarchical nanostructures. However, recent studies have adopted hybrid fabrication techniques combining dry and wet methods to produce these complex structures.
Method | J max (mA cm−2) (at 1.23 V vs. RHE) | Material | Electrolyte | Ref. |
---|---|---|---|---|
a MS-GLAD: magnetron sputtering with glancing angle deposition. b AACVD: aerosol-assisted CVD. c PECVD: plasma-enhanced CVD. | ||||
MS-GLADa | 2.40 (at 1.0 V vs. Ag/AgCl) | WO3/Ag2S | 0.5 M Na2SO4 | 57 |
Sputtering | 4.00 | NiFeOx/Ta3N5/Ta | 0.1 M Na3PO4 | 58 |
CVD | 5.50 (at 1.2 V vs. SCE) | Branched ZnO | 0.1 M Na2SO4 | 64 |
CVD | 0.029 (at 1.2 V vs. SCE) | P doped ZnO | 0.1 M Na2SO4 | 65 |
AACVDb | 1.241 | 40 mol% Fe-doped MoS2/Mo2S3 | 0.35 M Na2SO3 and 0.25 M Na2S | 66 |
PECVDc | 9.00 | RuOx | 1 M NaOH | 67 |
CVD | 0.78 (at 0.8 V vs. Ag/AgCl) | WO3/Fe2O3 | 1 M NaOH | 68 |
ALD | −3.71 (at 0 V vs. RHE) | Cu2O/TiO2/rGO/NiFe | 0.5 M Na3PO4 | 73 |
ALD | W:BiVO4 photoanode: ∼1.23 (at 0.79 V vs. RHE) | W:BiVO4–CuBi2O4/CdS/TiO2/RuOx | 0.3 M K2SO4 and 0.2 M KPi | 74 |
CuBi2O4 photocathode: 0.77 (at 0.67 V vs. RHE) | ||||
ALD | — | WO3/TiO2 | 0.5 M Na2SO4 | 75 |
ALD | 2.38 | CdS@ZnO | 0.1 M Na2SO4 | 76 |
ALD | 0.0026 (at 0.6 V vs. NHE) | Hf–ZnO | 0.5 M Na2SO4 | 77 |
Fig. 3 (a) Schematic representation illustrating the principle of magnetron sputtering technology,55 adapted with permission from ref. 55, copyright 2020 by Elsevier Ltd and Techna Group S.r.l. (b) Diagram of the magnetron sputtering with glancing angle deposition (MS-GLAD) technique,56 indicating that the substrate can be adjusted from positions P0 to P5, adapted with permission from ref. 56, copyright 2021 by Elsevier Ltd and Techna Group S.r.l. |
S. Limwichean et al. used magnetron sputtering with glancing angle deposition (MS-GLAD) and 2-h post-annealing in air to fabricate nanostructured WO3 thin films with diverse morphologies.56 The positioning of the substrate within the MS-GLAD setup significantly impacted the films’ morphology, crystallinity, optical transmittance, and PEC performance. By adjusting substrate positions, the nanorod film layer of WO3 could be precisely engineered, leading to enhanced PEC efficiency and demonstrating the potential of MS-GLAD in improving photoanodes for solar water-splitting cells. J. Yadav and J.P. Singh described a two-step method to create a WO3/Ag2S heterojunction to boost PEC performance. This involved depositing a WO3 thin film via DC sputtering and forming well-separated Ag2S nanorods using MS-GLAD.57 The WO3/Ag2S heterojunction showed superior light absorption and a higher photocurrent density of 2.40 mA cm−2 (at 1.0 V Ag/AgCl) than the bare WO3 thin film (0.34 mA cm−2). The vertically tilted Ag2S nanorods enhanced light trapping, and electrochemical impedance spectroscopy indicated low charge-transfer resistance at the semiconductor–electrolyte interface, reflecting a high flat band potential. This research underscored the importance of the interface between WO3 and Ag2S in water-splitting PEC response.
Meanwhile, T. Higashi et al. introduced a fabrication method that merges sputtering with a solution-based process.58 Initially, Ta3N5, the primary material, was deposited using a radiofrequency magnetron sputtering system. Subsequent surface modification involved applying electrocatalysts such as FeOx, NiOx, CoOx, NiFeOx, and NiFeCoOx to form metal-oxide catalysts, enhancing the PEC oxygen evolution reaction (OER) performance. Among these, the NiFeOx-modified Ta3N5 (NiFeOx/Ta3N5) photoanode showed the most promising onset potential for the OER. Consequently, a solar water-splitting cell with a parallel tandem configuration, incorporating the NiFeOx/Ta3N5 photoanode and an Al-doped La5Ti2Cu0.9Ag0.1S5O7 (LTCA:Al) photocathode, was able to produce stoichiometric hydrogen and oxygen from water splitting without external bias, achieving a solar-to-hydrogen (STH) energy conversion efficiency of 0.2% within the first minute of reaction.
The temperatures involved in CVD are typically higher than those in PVD, which allows for the deposition of a wider variety of materials and the growth of more complex films.61 CVD systems are characterised by their complexity, requiring precise control over gas-phase reactions and often necessitating multiple precursor gases, leading to more elaborate setups and operational considerations.62 CVD is particularly advantageous for applications that demand precise control over film composition and properties, such as semiconductor manufacturing, thin-film deposition, and surface engineering.63 It is frequently used to deposit protective coatings, passivation layers, or catalytic materials, thereby improving the stability, reactivity, and efficiency of solar water-splitting cells. Consequently, there are numerous solar water-splitting cells processed using CVD, with highly cited papers highlighted in this section (Fig. 4).
Fig. 4 (a) Illustration of the growth of both unbranched and branched ZnO nanorods using a CVD system.64 (b) Comparison of photoresponses between (i) unbranched and (ii) branched ZnO nanorods. Reproduced with permission from ref. 64, copyright 2020 by Elsevier Ltd and Techna Group S.r.l. (c) Depicts (i) the RF plasma reactor utilised for the deposition of Ru-based films and (ii) the chemical structure of the Ru precursor.67 Reproduced with permission from ref. 67, copyright 2020 by MDPI. |
In 2021, S. A. Alharbi et al. developed several CVD-based solar water-splitting cells using both branched and unbranched ZnO nanorods.64 The XRD analysis of the CVD-grown ZnO nanorods showed well-ordered, uniformly sized rods with a hexagonal-phase crystalline structure, exhibiting strong light-harvesting capabilities. The notable photoconversion efficiency of these ZnO nanorods indicates their potential in energy storage and conversion applications. Additionally, another study demonstrated the successful synthesis of bare and phosphorous-doped ZnO nanorods on fluorine-doped tin oxide (FTO)-coated glass via CVD. The integration of phosphorous and the hexagonal structure of the nanorods was verified using standard analytical techniques.65 The results revealed that doping the ZnO nanorods with a small amount of phosphorous significantly improved their photoelectrochemical performance.
Furthermore, various researchers have investigated modifications to the CVD process. For example, K.C. Lau et al. utilised aerosol-assisted CVD (AACVD) to create solar water-splitting cells based on mixed iron–molybdenum multi-sulfide composite thin films. These films were deposited at 550 °C for 20 minutes using AACVD, with the concentration of the Cp2Fe dopant varying (10, 20, 40, and 80 mol%).66 PEC measurements showed that the thin film doped with 40 mol% Fe displayed the highest photocurrent density (1.241 mA cm−2), whereas the 80 mol% Fe-doped film had a lower photocurrent density of 0.761 mA cm−2, but exhibited the highest photosensitivity (1.396) due to the longest electron–hole recombination lifetime. Consequently, the amount and oxidation state of the doped iron significantly influenced the PEC performance of the molybdenum multisulfide composites.
Similarly, L. Jozwiak et al. utilised plasma-enhanced CVD (PECVD) to fabricate novel Ru-based thin catalytic films. X-ray photoelectron spectroscopy (XPS) analysis of the RuOx thin films indicated that the as-deposited films were primarily composed of metallic Ru atoms (Ru0) (primarily Ru4+ (RuO2)), which transitioned to higher oxidation states after electrochemical stabilisation.67 These stabilised films demonstrated exceptional catalytic OER activity, as shown by linear scanning voltammetry results at pH = 13.6, with onset potentials and 10 mV overpotentials of approximately 220 mV and 350 mV, respectively. Additionally, a photocurrent density of 9.0 mA cm−2 was obtained at 1.6 V (vs. RHE) in the photoelectrochemical process. Therefore, the plasma-deposited RuOx catalysts emerged as potent alternatives for photoanode materials in solar water-splitting cells.
Y. Zhang et al. utilised dry-based CVD and solution-based processes to fabricate a heterojunction photoanode.68 Specifically, WO3 and Fe2O3 composite heterojunction photoanodes were synthesised on an FTO substrate using hydrothermal and CVD methods, respectively. Compared to the pristine WO3 nanostructure, the WO3/Fe2O3 heterojunction photoanode exhibited a 1.25-fold increase in photocurrent density. The heterojunction formation accelerated the separation and migration of the photogenerated charge carriers and consequently inhibited their recombination. As a result, the PEC performance and stability of the heterojunction photoanode were improved under alkaline conditions.
Fig. 5 (a) Schematic representation of charge transfer during (i) the photoelectrochemical reaction in WO3/TiO2 core–shell nanoplates, developed through hydrothermal treatment and ALD,75 (ii) TEM images of TiO2 layers formed after 200 ALD cycles on WO3. Reproduced with permission from ref. 75, copyright 2020 by American Chemical Society. (b) Synthesis of (i) CdS@ZnO nanorod arrays via the ALD method76 (ii) and the HAADFSTEM image of CdS@ZnO200 core–shell nanorods of Zn and Cd. Reproduced with permission from ref. 76, copyright 2023 by Elsevier B.V. |
ALD was utilised to deposit an ultrathin TiO2 passivation layer on the main material in the study conducted by Y. Wang et al., where the rGO/NiFe-LDH catalyst was applied to a Cu2O photocathode.73 A pristine Cu2O nanostructure was electrodeposited onto an FTO substrate, followed by synthesising the decorated rGO/NiFe-LDH catalyst using hydrothermal methods. The ALD-deposited TiO2 coating layer significantly mitigated photocorrosion and improved the chemical reaction kinetics of the Cu2O photocathode. The current density reached −3.71 mA cm−2, and the photoelectric conversion efficiency was 0.41%, underscoring the effectiveness of this method in enhancing Cu2O applications in PEC water splitting.
In a related effort, A. Song et al. utilised an ALD system to create a passivation layer to prevent photocorrosion in a W:BiVO4–CuBi2O4 tandem solar water-splitting cell, employing a complex structure and various solution-based processes.74 The W:BiVO4 photoanode and CuBi2O4 photocathode were produced through spray pyrolysis. CoPi catalysts were electrodeposited on the W:BiVO4 photoanode. For the photocathode, a CdS buffer layer and TiO2 protection layer were established via chemical bath deposition and ALD, respectively. The RuOx cocatalyst was photoelectrodeposited onto the CuBi2O4/CdS/TiO2 structure. Despite the significant increase in the stability of the CuBi2O4 photocathode after the coating process, the photocurrent onset potential decreased, attributed to the CdS/TiO2 heterojunction layer. Consequently, a tandem cell featuring W:BiVO4/CoPi and CuBi2O4/CdS/TiO2/RuOx demonstrated hydrogen production under illumination with an applied bias of ≥0.4 V.
An ALD system was also utilised for precision coating and to develop a core–shell nanostructuring tool. For instance, using a hydrothermal method and ALD, K. Liu and T. Perng crafted a core–shell flower-like structured WO3/TiO2 nanoplate.75 This core–shell structure exhibited enhanced PEC efficiency compared to pure WO3 under AM 1.5 solar-light irradiation. Moreover, the hybrid structure could absorb substantial light and reduce the recombination rate of electron–hole pairs. In another study, X. Feng et al. reported improved hydrogen evolution efficiency with CdS@ZnO core–shell nanorod arrays developed through ALD in a solar water-splitting cell.76 The CdS@ZnO core–shell nanorod arrays demonstrated a significantly higher H2 evolution rate of 21.64 μmol h−1 cm−2, nearly 9.4 times that of pristine CdS nanorod arrays. This improvement was attributed to the Z-scheme charge-transfer mechanism between the CdS core and ZnO shell, which facilitates the spatial separation of photoinduced electrons and holes and reduces the transfer resistance of charge carriers due to the close contact between the two semiconductors.
Additionally, the ALD process provided a doping effect on the main material in the study conducted by B. Alfakes et al. The researchers precisely introduced a hafnium (Hf) dopant into ZnO through ALD, enhancing its PEC performance.77 Detailed analysis of the doped materials revealed a threefold increase in photocurrent for the best-performing sample (0.75-wt% Hf), attributed to the reduced surface recombination of photocarriers, thereby leading to more efficient PEC water oxidation. They concluded that ALD can be applied to introduce materials into the photoanode, combining unstable and highly active materials with an appropriate protective layer, potentially overcoming stability limitations.
ALD evidently contributes to enhancing the PEC performance of solar water-splitting cells. The findings from these studies underscore that ALD is an essential tool for designing and fabricating PEC electrodes for water splitting.
This section explores various methods for fabricating solar water-splitting cells through laser-related processes, categorised into pulsed laser deposition (PLD), direct laser processing, laser-assisted synthesis, and other optical fabrication techniques. Although the discussion of PLD, typically conducted under vacuum conditions, slightly diverges from the primary focus of this review, which is to summarise advanced fabrication processes employing lasers without vacuum constraints, it is included for a comprehensive overview. Both vacuum and nonvacuum laser-related processes are discussed in this section. Detailed information on solar water-splitting cells fabricated using laser-based methods is compiled in Table 2.
Method | J max (mA cm−2) (at 1.23 V vs. RHE) | Material | Electrolyte | Ref. |
---|---|---|---|---|
a LIPT: laser induced phase transformation. b PLA: pulse laser annealing. c LIPR: laser-induced photoreduction. d LADC: laser-assisted defect control. e LICD: laser-induced catalyst deposition. f PILI: pulsed laser irradiation in liquid. g LSPC: laser synthesis and colloidal treatment. | ||||
PLD | 2.60 | BiVO4 | 0.5 M KPi | 85 |
PLD | 5.70 | Co4O4/pGO/BiVO4/SnOx–Pt/TiOx/PIP/CuOx | 1.0 M KBi | 86 |
PLD | 0.91 | TiO2/Fe2O3 | 1 M KOH | 87 |
LIPTa | 1.146 | Fe2O3 | 1 M NaOH | 88 |
PLAb | 0.484 | Ti:Fe2O3 | 1 M NaOH | 89 |
Laser treatment | 4.43 (at 2.0 V vs. Ag/AgCl) | Cu–TiO2 | 0.5 M Na2SO4 and 0.5 M KOH | 90 |
LIPT | 0.00843 | TiO2 | 1 M KOH | 91 |
LIPRc | 0.747 | Fe2O3 | 1 M NaOH | 92 |
LADCd | 3.21 | WO3/FeNiOOH | 0.5 M Na2SO4 | 93 |
Laser treatment | 0.60 | BiVO4 | 0.1 M KPi | 94 |
Laser treatment | 0.908 | TiO2 | 1 M KOH | 95 |
LICDe | 3.16 | WO3/NC–FeOOH | 0.5 M Na2SO4 | 96 |
PLILf | 1.45 | Fe2O3@TiO2–VTi | 1 M KOH | 97 |
LSPCg | 2.28 | Fe2O3/TiO2–CdTe | 1 M KOH | 98 |
LSPC | 6.22 | LBSO:BiVO4 | 1 M KPi | 99 |
Laser treatment | 6.08 | Mo:BiVO4 | 1 M KPi | 100 |
Photoelectrodeposition | 0.72 (at 1.5 V vs. RHE) | Pt–TiO2 | 1 M KOH | 101 |
Photoelectrodeposition | 0.63 | Fe2O3/NiOOH | 1 M NaOH | 102 |
Photoelectrodeposition | 2.34 | CoOOH/FeOOH/F—Fe2O3 | 1 M KOH | 103 |
Photoelectrodeposition | 4.30 | CoFeBi/BiVO4 | 1 M K3BO3 | 104 |
Photoelectrodeposition | 3.60 | Ni2P2O7–Nd–BiVO4 | 0.5 M Na2SO4 | 105 |
Microwave-assisted synthesis | 1.68 (at 1.0 V vs. Ag/AgCl) | WO3 nanosheets | 0.5 M Na2SO4 | 106 |
Microwave irradiation | 1.37 | Si/Ti:Fe2O3 | 1 M NaOH | 107 |
Fig. 6 (a) Illustration depicting (i) the trio of primary processes occurring in PLD using a BiVO4 target.85 (ii) SEM cross-sectional view of BiVO4 films produced. Inset: An optical image of the FTO/BiVO4 sample. Reproduced with permission from ref. 85, copyright 2020, American Chemical Society. (b) PLD-processed TiO2/Fe2O3 core/shell nanostructure, engineered for enhanced stability and efficiency in visible-light-driven photoelectrochemical water splitting;87 includes (i) a cross-sectional SEM representation, and (ii) a TEM visualisation. Reproduced with permission from ref. 87, copyright 2020 by American Chemical Society. |
M. Kölbach et al. prepared BiVO4 thin films for solar water-splitting cells using PLD,85 originating from a BiVO4 target. The study systematically detailed PLD for BiVO4 films, highlighting discrepancies from an ideal stoichiometric target-to-substrate material transfer. It explored the relationship between the V/Bi ratio of the films, their charge-carrier transport properties, and PEC performance, achieving AM 1.5 sulfite oxidation photocurrents of ∼2.4 mA cm−2 at 1.23 V (vs. RHE) without intentional doping or surface modification. Additionally, the creation of BiVO4 photoelectrodes by alternating ablation of Bi2O3 and V2O5 targets was discussed as an innovative method to control cation stoichiometry, yielding BiVO4 films with AM 1.5 sulfite oxidation photocurrents of up to 2.6 mA cm−2 at 1.23 V (vs. RHE). These findings provide crucial insights into the PLD of ternary oxide semiconductors, facilitating the accelerated synthesis and investigation of new metal-oxide photoelectrodes.
S. Ye et al. developed a BiVO4-based photoanode for an overall unassisted PEC water-splitting system, integrating components like a Co4O4/pGO/BiVO4/SnOx photoanode and a Pt/TiOx/PIP/CuOx photocathode with multimediator modulation.86 PLD was used to form a BiVO4/SnOx layer on the FTO electrode surface. The study revealed orderly electron transport from the BiVO4 photoanode to an organic photocathode, driven by mediator potential differences, a reduced charge recombination rate, and an increased charge-transfer rate within the system. The strategic use of different mediators underscored their roles in enhancing charge transfer and minimising charge recombination. Moreover, combining a fully structured photoanode and photocathode yielded the highest charge-transfer efficiency compared to configurations without mediators. The employment of organic polymers enabled highly complementary light absorption between the organic photocathode and BiVO4 photoanode, significantly boosting solar energy utilisation and leading to an impressive STH conversion efficiency of ∼4.3%. These results underscore the critical importance of thoughtful design and assembly in creating effective dual-photoelectrode devices, particularly highlighting the benefits of complementary light absorption and efficient charge transfer.
Core–shell nanomaterials are pivotal in photoelectrochemistry, as they exhibit enhanced charge separation, improved light absorption, and tailored surface properties, offering versatile and scalable solutions for efficient energy conversion applications. PLD also contributes to synthesising core–shell nanostructures. H. Lu et al. successfully fabricated TiO2/Fe2O3 core–shell heterojunction nanorod arrays using PLD, which was then applied as a photoelectrode in efficient solar water-splitting cells.87 The study extensively analysed the morphology, phase, and carrier transfer mechanisms of both pristine TiO2 and TiO2/Fe2O3 core–shell nanostructures. PEC measurements indicated that the photocurrent density of the TiO2/Fe2O3 core–shell nanostructure nearly doubled compared to that of the pristine TiO2 structures. Additionally, visible-light absorption expanded from 400 to 550 nm, the charge separation rate was enhanced, and exceptional stability was demonstrated, with only a 3% decrease in current density after 14 days of testing. This research illustrates a method for designing efficient nanostructures through a combination of a simple hydrothermal process and a high-quality PLD process with significant potential for versatile expandability.
Fig. 7 (a) Illustration depicting (i) the laser-induced selective phase transformation process applied to hematite nanorods (NRs);88 (ii) combinatorial study of optimal laser power with laser scanning speed for akaganeite NRs on FTO glass substrates; (iii) laser-patterned hematite NRs displaying letters created with the laser scanner. Reproduced with permission from ref. 88, copyright 2020 by American Chemical Society. (b) Schematic showing the laser-based reduction method used for BiVO4 photoanodes.94 Reproduced with permission from ref. 94, copyright 2022 by American Chemical Society. |
Exploring the photothermal effect of laser processing offers an innovative approach to manipulating metal oxides. J. Yeo's research group demonstrated this by fabricating metal-oxide-based solar water-splitting cells with enhanced PEC performance. Specifically, they employed laser processing as a novel post-annealing technique to improve the PEC performance of hematite nanorod photoanodes.88 The focused laser beam created a highly localised high-temperature field, facilitating the transformation of akaganeite (β-FeOOH) into hematite (α-Fe2O3) and the diffusion of Sn from the FTO glass. This laser-induced phase transformation (LIPT) process resulted in exceptional water oxidation performance without damaging the FTO glass substrates, thanks to its precise localisation and instantaneous action. The hematite photoanodes produced through the LIPT process showed significantly enhanced electrical resistance, crystallinity, and optical absorbance, culminating in superior PEC performance.
Similarly, D. Wang et al. introduced a pulse laser annealing (PLA) strategy to activate titanium dopants on hematite photoanodes, aiming for improved PEC performance.89 High-temperature annealing, a common method for activating dopants on metal oxides, often results in considerable thermal damage to the conductive substrate. The laser annealing technique circumvents this issue, activating dopants while minimising thermal damage effectively. The Ti:Fe2O3 photoanode treated with laser exhibited improved crystallinity compared to its untreated counterpart, doubling the photocurrent density. Additionally, the surface charge transfer efficiency more than doubled, and the donor density increased significantly, from 1.28 × 1018 cm−3 to 1.29 × 1019 cm−3. These findings underscore the potential of PLA treatment in enhancing dopant activation and, consequently, the performance of photoanodes for PEC applications.
K. G et al. proposed a method for annealing thin copper-coated TiO2 nanotubes using a laser process to enhance PEC performance.90 This fast and straightforward method enables selective modification of the nanostructures’ surface to achieve desired shapes and sizes. The samples prepared by this laser method (L60-Cu–TiO2 and L120-Cu–TiO2) showed increased absorbance in the visible light spectrum compared to pure TiO2, with the current density observed to improve by approximately sevenfold under both dark and illuminated conditions. This enhancement is ascribable to the synergistic effects of laser-induced morphological and optical property changes, making this approach viable for producing TiO2 nanotubes with outstanding electrochemical activity for the OER.
M. Qiao et al. introduced an innovative approach to transform the phase of TiO2 nanotubes using femtosecond laser processing.91 This method selectively alters the crystal facet phase of the anatase samples, boosting the photocurrent density by more than five times compared to the unaltered anatase TiO2 nanotube photoanode, from 1.59 μA cm−2 to 8.43 μA cm−2. Additionally, the photoluminescence intensity of the TiO2 nanotubes prepared by laser processing was lower than that of the pristine anatase TiO2 nanotube photoanode, indicating improved charge separation efficiency for photogenerated carriers. Thus, the proposed laser annealing processing strategy holds promise not only for PEC applications but also for the development of functional nanomaterials.
In a separate study, J. Yeo's research group investigated the interaction between the laser's photothermal and photothermochemical effects on hematite nanorods. They explored laser-induced photoreduction (LIPR) of hematite nanorods with dual objectives: to create patterned arrays of magnetite nanorods and to enhance the performance of hematite nanorods in PEC water oxidation by engineering oxygen vacancies (OVs).92 A CW laser beam with a wavelength of 532 nm was directed at hematite nanorods immersed in an ethylene glycol (EG) solution, providing the thermal energy necessary for the Fe3+ reduction reaction. They successfully converted hematite nanorods into magnetite using adequate laser power during the LIPR process. OV-rich hematite nanorods were produced at lower laser powers, serving as an intermediary stage before transforming into magnetite. These OV-rich hematite nanorods showed significantly improved donor density and electrical conductivity, attributed to OVs acting as additional electron donors, leading to enhanced PEC water oxidation performance.
Expanding on OV engineering to manipulate the lattice defects of metal oxides, H. Kong et al. reported utilising the laser-assisted defect control (LADC) process on monoclinic tungsten trioxide (m-WO3) to enhance PEC performance.93 The study particularly focused on the correlation between the spatial distribution of OV defects in m-WO3 photoanodes and their PEC performance. It was confirmed that a higher concentration of OVs near the surface of m-WO3, compared to its bulk, was beneficial for PEC performance. Consequently, LADC enabled precise manipulation of the spatial distribution of OVs on the m-WO3 photoanode surface, leading to improved PEC performance.
Similarly, M. Barawi et al. applied laser irradiation to improve the water oxidation properties of BiVO4 photoanodes through surface reduction.94 This technique selectively reduced surface-level V atoms, inducing OV formation without altering the film's bulk structure. The modifications were confirmed by XPS and X-ray absorption spectral analyses. The resulting samples exhibited effective n-doping, marked by a shift in the VB maximum to higher binding energy, which enhanced charge transport and created a more favourable VB position for water oxidation. Additionally, the treated samples showed an extended lifetime for photogenerated holes, reduced charge-transfer resistance, and an accumulation of holes at the surface, collectively reducing recombination and boosting charge extraction. These effects collectively doubled the photocurrent density of the laser-treated samples compared to the untreated ones.
Furthermore, T. Nakajima et al. explored the generation of OVs in TiO2 films95 using laser irradiation, which led to improved PEC performance over pristine TiO2 films. The STH efficiency of the treated TiO2 was 0.52%, approximately 2.6 times higher than that of the untreated TiO2 film. This enhancement was attributed to oxygen deficiency, which increased the density of electron donors in the photoanode, improving charge transport and overall PEC performance. Additionally, laser irradiation altered the cross-sectional morphology of the TiO2 film, increasing the carrier diffusion length and further boosting the PEC properties of the photoanode.
Fig. 8 (a) Schematic diagram of (i) an optical system for LICD. (ii) Schematic diagram of LICD reaction. A CW laser beam with a wavelength of 532 nm was employed to scan the entire sample area. (iii) WO3 NRs are immersed in an aqueous solution containing Ni2+, Co2+ and Fe3+ ions.96 Reproduced with permission from ref. 96, copyright 2023 by The Royal Society of Chemistry. (b) TiO2–VTi before and after PLIL.97 (i) Schematic illustration of the preparation of TiO2–VTi nanocrystals via PLIL. (ii) SEM image of TiO2–VTi raw materials before PLIL (inset: the optical image of TiO2–VTi raw materials dispersed in ethanol before (left) and after PLIL (right)). (iii) STEM image of the TiO2–VTi nanocrystals and (iv) their HRTEM image. Reproduced with permission from ref. 97, copyright 2023 by Elsevier B.V. |
H. Kim et al. employed laser-induced catalyst deposition (LICD) to deposit FeOOH and Ni- and Co-doped FeOOH cocatalysts onto WO3 photoanodes.96 By adjusting the elemental composition and combining Ni, Co, and Fe in the liquid precursor, Ni- and Co-doped FeOOH cocatalysts were effectively formed with a thickness of approximately 5 nm on the WO3 photoanodes via LICD. The WO3 photoanode featuring Ni- and Co-doped FeOOH cocatalysts demonstrated a photocurrent exceeding 3 mA cm−2 at 1.23 V (vs. RHE). The lifespan of photoexcited carriers was analysed through time-resolved photoluminescence, measuring luminescence signals from carrier recombination. Distinguishing luminescent photons with λ < 500 nm and those with λ > 500 nm using short-pass and long-pass optical filters, respectively, it was found that the Ni- and Co-doped FeOOH cocatalysts significantly prolonged the lifetime of photoexcited holes with λ > 500 nm. This suggests that trap-mediated recombination was inhibited, with the cocatalysts providing hole-storing trap states that extend the lifespan of trapped holes, thereby enhancing water oxidation photocurrents.
Similarly, H. Kong et al. conducted a one-step laser deposition process to deposit an FeNiOOH catalyst overlayer on monoclinic m-WO3 nanorods (NRs).93 An FeNiOOH layer ∼2 nm thick was successfully deposited on the WO3 NR. The FeNiOOH-catalyst-deposited monoclinic m-WO3 sample exhibited an increase in photocurrent from 2.13 mA cm−2 to 3.21 mA cm−2 at 1.23 V (vs. RHE), which marks an improvement of approximately 51% over the pristine monoclinic m-WO3 sample. Furthermore, the FeNiOOH catalyst-deposited monoclinic m-WO3 sample displayed a negative shift in the dark current onset potential and enhanced stability.
Meanwhile, F. Li et al. executed nanocrystal synthesis in a liquid precursor via the pulsed laser irradiation in liquid (PLIL) technique.97 To bolster the PEC performance of Fe2O3 photoanodes, the PLIL process was utilised to incorporate sub-5-nm p-type Ti-defected TiO2 nanocrystals into a Fe2O3 matrix. This approach significantly boosted the photocurrent density from 0.27 to 1.45 mA cm−2 at 1.23 V (vs. RHE) and reduced the carrier transport time from 159 to 38 μs. The nano-heterostructures produced through the PLIL process not only expedited the separation of photogenerated electron–hole pairs but also enhanced carrier transport, leading to improved PEC performance.
To enhance the charge transfer efficiency of metal oxides, H. Wang and colleagues applied laser synthesis and colloidal treatment (LSPC) strategies.98 They developed embedded p–n heterojunctions within a hematite photoanode using the LSPC process. Subsequently, introducing a passivation layer (TiO2–CdTe) onto the hematite resulted in a substantial improvement in PEC performance. This enhancement was evidenced by a 12-fold increase in the lifetime of photogenerated carriers, a 3.9-fold increase in the injection efficiency ηinj at 1.23 V (vs. RHE), and an increase in the photocurrent density at 1.23 V (vs. RHE) by 4.4 times. Furthermore, the LSPC process was employed to incorporate laser-generated nanocrystals (LBSO) into the BiVO4 photoanode,99 achieving a photocurrent density increase from 4.01 mA cm−2 to 5.15 mA cm−2 at 1.23 V (vs. RHE) for a single BiVO4 photoanode configuration and 6.22 mA cm−2 at 1.23 V (vs. RHE) for a dual configuration. The LBSO:BiVO4 photoanode, prepared via LSPC, exhibited enhanced charge transfer efficiency alongside increased carrier mobility and lifetime. This research offers an innovative approach to overcoming poor charge transport and augmenting PEC performance.
To mitigate photocorrosion in water splitting, J. Jian and associates introduced advanced techniques for creating a stable and active semiconductor–liquid junction (SCLJ).100 They innovated a dual interfacial layer structure on Mo:BiVO4 (MBVO) photoanodes using laser-generated carbon dots with phenolic hydroxyl groups (LGCDs–PHGs), which protected against photocorrosion and augmented charge separation and transfer, thereby facilitating highly efficient and stable PEC activity. The enhancement in PEC activity was attributed to the improved charge density of the shallow MBVO-OV layer and the effective charge extraction and band alignment facilitated by the LGCD–PHG layer. The uniform application of this dual interfacial layer not only improved interfacial charge-carrier kinetics but also mitigated photocorrosion, leading to increased operational stability. Notably, the FeNiOOH–LGCD-PHG–MBVO photoanode, in a dual configuration, showcased a significant photocurrent density of up to 6.08 mA cm−2 at 1.23 V (vs. RHE) and maintained operational stability over 120 h. The proposed methodology, leveraging laser–matter interaction, proved generally advantageous for employing catecholic molecules to develop an effective SCLJ with enhanced PEC performance.
Fig. 9 (a) Schematic of the experimental steps for the microwave-assisted synthesis of WO3 nanosheets on the W foil.106 (b) HR-TEM images of the area highlighted in (i), with the corresponding lattice fringe distribution depicted in the inset. (ii) Fast Fourier transform (FFT) patterns corresponding to the highlighted blue square in (i). (iii)–(v) EDX elemental mapping of microwave-assisted optimum WO3-MW-3 cycle. (c) (i) J–V and (ii) J–t curves of (1) WO3-MW-1, (2) WO3-MW-3 and (3) WO3-MW-5 cycles, and (4) WO3-MW-3 cycle-non-polished photoanodes annealed at 500 °C, 4 h, respectively. Reproduced with permission from ref. 106, copyright 2022 by Elsevier B.V. |
Similarly, Y. Chen's research group introduced a photoassisted electrodeposition method for efficiently fabricating photoanodes on α-Fe2O3.102 This approach enabled the successful deposition of cocatalysts onto Fe2O3 photoanodes. As a result, the photocurrent density of the α-Fe2O3/NiOOH photoanode prepared by the photoassisted electrodeposition method (α-Fe2O3/NiOOH-ph) reached 626.137 μA cm−2 at 1.23 V (vs. RHE), which is approximately four times that of the pristine α-Fe2O3 photoanode and double that of the α-Fe2O3/NiOOH photoanode prepared by conventional electrodeposition. Additionally, the α-Fe2O3/NiOOH-ph photoanode exhibited reduced photogenerated carrier charge transfer resistance at the interface and a lower onset potential, enhancing charge transfer efficiency compared to the pristine α-Fe2O3 photoanode.
Furthermore, the research groups of G. Hu, Wang, and Jin employed a similar method of metal hydroxide catalyst deposition on the photoanode through photo-assisted electrodeposition. Hu's group demonstrated a strategy that not only improved the sluggish OER kinetics but also enhanced the PEC performance of the photoanode. The CoOOH/FeOOH/F-Fe2O3 photoanode achieved a current density of 2.34 mA cm−2 at 1.23 V (vs. RHE), which is 3.3 times higher than that of the pristine α-Fe2O3 photoanode.103 Wang's group observed an increase in photocurrent for the BiVO4 photoanode, approximately 2.6 times higher than that of the pristine BiVO4,104 along with improved PEC performance, including decreased carrier recombination and increased applied bias photon-to-current and incident-photon-to-electron conversion efficiencies. Jin's group found that the introduction of the Ni2P2O7 cocatalyst accelerated the water oxidation reaction and enhanced the PEC performance of BiVO4.105 The Ni2P2O7–Nd–BiVO4 (3.6 mA cm−2) photoanode prepared by photoassisted electrodeposition exhibited a higher photocurrent density (3.6 mA cm−2) compared to the bare BiVO4 photoanode (1.2 mA cm−2).
M. A. Mahadik et al. explored the PEC performance and the growth mechanism of WO3 nanosheet (NS) photoanodes on tungsten foils using microwave-assisted (MW-assisted) synthesis.106 The growth of the WO3–NS photoanodes was achieved through varying numbers of deposition cycles (one, three, and five times) with a constant microwave synthesis duration of 30 min. The photoanode deposited through three cycles exhibited the highest photocurrent density, reaching 1.68 mA cm−2 at 1.0 V (vs Ag/AgCl). The WO3–NS photoanode, also deposited in three cycles, demonstrated efficient H2 and O2 evolution and significant degradation of the PEC orange II dye within 3 h, indicating its potential for diverse applications such as photovoltaics and electrocatalysis.
Similarly, T. Koh et al. demonstrated an MW-assisted synthesis technique for fabricating an Si and Fe2O3 photoanode (Si/Ti:Fe2O3).107 They addressed high charge recombination and charge transfer kinetics by forming a homojunction. The thickness of the Si/Ti:Fe2O3 photoanode was adjusted by altering the MW irradiation time. The optimised homojunction (Si/Ti:Fe2O3) photoanode achieved a photocurrent density of 1.37 mA cm−2 at 1.23 V (vs. RHE) and exhibited an onset potential shift compared to the bare Ti–Fe2O3 photoanode, significantly enhancing charge separation and carrier transfer. This research suggests a cost-effective method for fabricating homojunction photoanodes using microwave-assisted synthesis. Comprehensive data for various solar water-splitting cells fabricated by other optical-based processes are summarised in Table 2.
The potential of laser-assisted fabrication processes is poised to revolutionise the field through significant advancements in materials science and manufacturing technologies. The exploration of new materials, coupled with the advent of sophisticated laser techniques, is anticipated to substantially enhance the efficiency of solar water-splitting cells. These innovations are not only expected to improve performance through better light absorption and charge separation but also to provide an avenue for integrating these systems with renewable energy sources, thereby moving towards a more sustainable and decentralised energy infrastructure.
In summary, laser-assisted fabrication processes, typically low-temperature and non-vacuum-based, offer several advantages over conventional vacuum-based methods. These advantages include the flexibility in material selection and the feasibility of solution processes, particularly for the deposition of cocatalysts. As technology progresses and researchers devise innovative solutions, laser-based processes may emerge as competitive alternatives for fabricating solar water-splitting cells, promising greater efficiency and scalability. However, ongoing research and development efforts are essential to fully realise their potential in this domain.
Footnote |
† These authors contributed equally to this work. |
This journal is © the Partner Organisations 2024 |