Silicon nanowire–Ta₂O₅–NGQD heterostructure: an efficient photocathode for photoelectrochemical hydrogen evolution

Abstract

Photoelectrochemical (PEC) water splitting has propelled broader research interest for the large-scale and facile entrapment of solar energy in hydrogen fuel. It offers the most favorable and environment-friendly approach to harvest renewable energy under solar radiation for solar-to-hydrogen fuel conversion. However, for superior hydrogen evolution reaction (HER) and enhanced overall efficiency, it is inevitable to design a suitable low-cost, active, scalable and durable photocathode. Although silicon (Si) is the backbone of the photovoltaic industry, an alternate drive is under development to establish its applicability towards PEC-HER. Nevertheless, bare Si with a metal catalyst suffers due to its instability, high reflectance loss, surface oxidation, and sluggish kinetics. These issues can be addressed by passivating the Si surface with the appropriate protection layer anchored with a low-cost metal-free catalyst to design the pertinent photocathode. Herein, vertically aligned p-silicon nanowires (p-SiNWs) with a conformal coating of tantalum pentoxide (Ta₂O₅) passivation layer and N-doped graphene quantum dots (NGQD) as a metal-free catalyst (p-SiNWs–Ta₂O₅–NGQD) have been designed, which impart efficient, stable, and scalable photocathodes for PEC-HER. The photocathode exhibits an applied bias photon-to-current conversion efficiency (ABPE) of ∼21.1%, which is 6-fold higher than that of the p-SiNWs–NGQD matrix with a low overpotential (−449 mV @ 5 mA cm⁻²) and Tafel slope (78 mV dec⁻¹) indicating the superiority of the catalyst. The role of Ta₂O₅ is not only to act as the passivation layer via reducing the lattice mismatch but it also facilitates the charge transfer process from Si to the electrolyte, minimizing the conduction band offset. Moreover, the non-corrosive nature of NGQD, having various N-functionalities, significantly enhances the active site densities and thereby imparts prolonged durability towards the HER performance for at least 10 hours @ 10 mA cm⁻² with negligible potential loss. The simplistic and compatible design strategy can promote a new-fangled pathway for the commercialization of large-scale Si-based metal-free photocathodes towards next-generation green fuel technology.

---

Introduction

Presently, the majority of energy produced worldwide is accomplished by fossil fuels. The rising energy demands and the restricted supply of conventional energy sources have elevated the fear and nervousness regarding the competence of fossil fuels to uphold the regular energy demands.^1–3 In this context, owing to the fast growth of the world population and energy demand, the universal scientific community is forced to pursue the development of alternative clean energy resources to cope-up with the future energy challenges.^1–4 Compared to all clean energy resources (wind energy, tidal energy, bioenergy, etc.), sunlight is an abundant source of energy because of its availability and non-hazardous nature.⁴ In addition, the amount of solar energy reaching the earth's surface is roughly 10 000 times the total world energy consumption,⁵ so its conversion to electricity has become a highly attractive and economical process for providing clean and renewable energy for the present and future, which is termed as solar-driven green fuel technology. Photoelectrochemical (PEC) water splitting has emerged as an efficient technology for producing hydrogen using renewable sources like sunlight and water.^6,7 Hydrogen has been projected as an imperative source of green fuel because of its high energy density and it can be converted to power upon demand through fuel cells.^8,9 However, to be competitive with the renewable energy sources and for large-scale solar-hydrogen production, there is an urgent need to develop low-cost, stable and highly efficient photoelectrodes.² Since it involves the absorption of light, charge carrier generation, separation, and collection via its facile transfer through a chemical medium, the choice of photoelectrodes is very crucial for enhancing the stability and efficiency of the PEC devices. Simultaneously, the suitability of the band-gaps of active materials for adequate broadband light absorption, the energetic feasibility of the energy band edges for efficient water electrolysis, electrode durability, and the overall cost of fabrication are a few major limitations to the commercialization of PEC technology on a global scale.¹⁰ Moreover, water splitting is not a spontaneous process because it requires a minimum of 1.23 V potential for triggering the process.^10,11 Thus, the band edges should straddle the redox potential of water splitting. The material must also be stable enough under harsh aqueous operating conditions. Therefore, the main challenge lies in the development of efficient, reliable and cheap photoelectrodes towards green fuel synthesis to support hydrogen fuel-cell technology.

Among all the existing semiconductors, silicon has attracted the attention of the global research community because of its suitability as an efficient photocathode in direct solar-driven water splitting due to its energetically favorable band edge positions, low carrier recombination velocity, high carrier mobility, earth abundance, low cost, and environmentally gracious nature.^12,13 Despite these advantages, the Si-based PEC devices are still not flourishing because of their instability due to surface oxidation under harsh chemical environments, high reflectance loss and lethargic charge-transfer kinetics at the Si/electrolyte interface owing to small exchange current density, which renders its applicability in the PEC-HER process.^12,14 To overcome the stability issues several approaches have been adopted such as the incorporation of a thin-film protection/passivation layer, surface functionalization, variation of the electrolyte medium, etc.^11,12,15 The charge transfer kinetics are addressed via modifying the electrode surfaces by anchoring with a suitable electrocatalyst.^11,16 In this context, heterojunctions with various metal or metal oxide/sulphide/phosphide-based electrocatalysts have been investigated, which not only protect the silicon surface but they also act as a suitable catalyst.^15,17–20 However, these systems suffer from the drawback of attaining high efficiencies because of the lattice mismatch at the Si/metal or Si/metal composite interfaces and therefore create high interfacial barrier resistance. In general, the protective layer should have minimal band offset and high-quality interface with low lattice mismatch to enhance charge carrier collection.^21,22

It has been reported that the use of a suitable electrocatalyst and/or semiconducting passivation layer can protect the surface corrosion under operational PEC conditions, which can help to improve the stability and lifetime of the photoelectrodes.²³ Recently, Esposito et al.²² demonstrated a noteworthy improvement in the PEC performance of Si-based photoelectrodes by introducing a high-quality thermally grown SiO₂ layer and bilayer metal catalyst. However, the large conduction band offset at the Si/SiO₂ interface inhibited the charge transfer mechanism. There are also several reports on using different transition metal-oxides as hole-selective (molybdenum oxide, tungsten oxide, vanadium oxide and cuprous oxide, etc.) or electron-selective layers (titanium oxide, tantalum oxide and zinc oxide, etc.) with Si and different metal-based catalysts for PEC application.^24–26 Li Ji et al.²¹ proposed the thin epitaxial layer of STO over Si with Ti/Pt as a nanostructured catalyst that can demonstrate a photocurrent density of −35 mA cm⁻² and applied bias photon to current conversion efficiency (ABPE) of 4.9% because of a low lattice mismatch of ∼1.69% (Si-3.84 Å and STO-3.905 Å). Nevertheless, it has been observed that coating the Si surface with TaO_x surprisingly enhances the ABPE to 7.7% of the Si/TaO_x–Pt system due to a low conduction band offset.²⁴ Inspired by this concept, Wang et al.²⁷ demonstrated a current density of −34.7 mA cm⁻² with an ABPE efficiency of 8.1% by using the pn⁺-Si/Ta₂O₅/Pt photoelectrode. They realized that Ta₂O₅ is a better choice as an oxide protection layer for Si because of its very low lattice mismatch of ∼1% at the Si (3.84 Å) and Ta₂O₅ (3.80 Å) interface; it simultaneously enables the proficient transport of electrons from Si to Ta₂O₅ due to a very small conduction band offset, and restrains the flow of holes for a large valence band offset. Besides the modification of the interfacial lattice parameters, as well as engineering the band off-set, another alternate approach can be adopted to enhance the efficiency of the Si-based photocathode via the nano-texturing of the Si surface responsible for high photo-absorption. Currently, 1D nanowires are gaining more attention due to high surface to volume ratio, orthogonalization and anti-reflective properties, thereby allowing efficient catalyst loading at a given active area of the Si-based photoelectrode.^12,13,16,28

Nevertheless, metal-based PEC-HER catalysts suffer from surface poisoning, high activation barrier, high cost, non-durability, and low efficiency, which have inhibited their commercialization.^12,14 In this regard, to surpass the adverse effects of using metal catalysts, researchers are more inclined towards metal-free catalysts like r-GO,²⁹ graphene quantum dots (GQD),^16,30 nitrogen-doped graphene quantum dots (NGQD),^12,31 graphene sheets,³²etc. Among them, NGQD has emerged as an innovative class of materials owing to their large surface areas, high electron mobility, sharp catalytic edges, environmental friendliness, corrosion-resistant properties, excellent transmittance, and tunable band-gap.^12,31 Moreover, their discrete electronic levels allow efficient charge transfer and prolong the excited state lifetime, making them suitable catalysts for PEC applications.^11,33 In addition, nitrogen functionality facilitates a resistance-less path for the rapid electron transfer process and consequently accelerates the HER kinetics.^12,34 Sim et al.^31,32 demonstrated the use of N-doped graphene as a catalyst for both planar and nanostructured Si and reported an ABPE of 2.29%. Furthermore, the large band-gap of NGQD enables its absorption in the UV region making it a suitable material for heterojunctions with Si having NIR absorption thus enabling broadband absorption. Along with this, the configuration of the Si-nanowire/passivation layer/catalyst is capable of demonstrating excellent photoelectrochemical performance with substantial improvement of the lifetime of Si-based photoelectrodes. The bifunctional nature of the catalyst not only helps to improve the catalytic performance but also acts as a potential passivating layer to protect the surface oxidation of silicon in harsh media.^35,36

Taking into consideration the advantages of Si nanowires coupled with Ta₂O₅ and NGQD, the present report demonstrates the investigation of p-SiNWs/Ta₂O₅/NGQD as an efficient photocathode for the PEC-HER. The photocathode exhibits a photocurrent density of −49.5 mA cm⁻² along with ABPE of ∼21.1%. The lower values of the overpotential and Tafel slope also indicate the superiority of the catalyst with prolonged stability for at least 10 hours @ 10 mA cm⁻² with a loss of ∼5% potential. The as-designed photocathode provides a green route with tandem architecture incompatible with Si-process line technology to facilitate broad-band solar absorption for solar-to-hydrogen fuel conversion.

Experimental section

Fabrication of vertically aligned silicon nanowires (SiNWs)

The nanowires were grown on 1 × 1 cm² p-type Si 〈100〉 substrates (B-doped; 1–3 Ω cm) via employing a cost-effective nanosphere lithography technique, followed by metal-assisted chemical etching (MACE), as reported elsewhere.³⁷ In brief, the organic impurities of the Si wafers were cleaned by sonication in acetone and ethanol, followed by purified deionized water (DI). The partially cleaned Si substrates were immersed in piranha solution (H₂SO₄ : H₂O₂ in 3 : 1 ratio) at 90 °C for an hour, followed by rinsing in DI water and dried using a nitrogen gun. Thereafter, the polystyrene (PS) beads were used as a hard mask for the formation of Si nanowires. The solution of PS beads with ethanol in a 1 : 1 ratio was sonicated for 2 hours and the solution was slowly transferred over a mixture of DI water and IPA (50 : 0.002) in a glass Petri-dish to get uniform coverage of the monolayer PS beads. This free-floating hexagonally oriented monolayer film was transferred over the previously cleaned Si substrate. The diameters of the closely packed PS beads were reduced from 450 nm to ∼230 nm by etching under optimized O₂ plasma, and 30 nm gold (Au) thin film was deposited by an e-beam deposition technique to catalyze the growth of SiNWs. The protected PS beads were removed under mild sonication in toluene to develop uniform Au mesh on the silicon surface. The resulting Au/Si samples were immersed in HF : H₂O₂ : H₂O (1 : 0.5 : 5) for 5 minutes to etch the Si underneath, followed by the removal of the Au catalyst in aqua-regia (nitric acid : hydrochloric acid in 1 : 3), and finally resulted in the formation of uniform vertically aligned p-SiNWs.

Deposition of Ta₂O₅ on p-SiNWs by the PLD method

The as-prepared vertically aligned p-SiNWs were coated with the thin film of Ta₂O₅ at a variable thickness by a pulse laser deposition (PLD) system. During deposition, the distance between the target and Si substrate was fixed at 6 mm with the substrate being heated at 350 °C. The deposition was carried out under an oxygen pressure of 1.6 × 10⁻⁶ torr and the rate of deposition was fixed at 0.25 Å per pulse. The laser fluence was optimized and kept at 2 J cm⁻² to avoid droplet formation on the film surface.

Fabrication of nitrogen-doped graphene quantum dots (NGQD)

N-doped graphene quantum dots (NGQD) were synthesized by the pyrolysis of citric acid as reported elsewhere.³⁸ Briefly, 3 g of citric acid were heated at 200 °C for 20 minutes to obtain an orange-coloured liquid. Then, 0.25 mol L⁻¹ NaOH solution was slowly added dropwise into 100 mL of the earlier obtained orange-coloured solution under rigorous stirring until pH ∼ 7. A change in colour from orange to brown and finally faint yellow (almost transparent) indicated the formation of GQD. Thereafter, 10 mL of the as-prepared GQD was mixed with 10 mL of a 6.5 M aqueous solution of urea under constant stirring for 30 minutes, followed by heating at 110 °C for 3 hours in a poly-tetra-fluoro-ethylene-lined autoclave. After cooling to room temperature, an orange homogeneous solution of NGQD was obtained. This solution was then centrifuged for 15 minutes at 10 000 rpm to remove any undissolved materials. The supernatant was further filtered through a 0.22 μm cellulose acetate membrane to remove larger particles and the remaining filtrate containing the desired NGQD was processed for necessary characterization.

Material characterization

The as-prepared samples were physically characterized by employing Field Emission Scanning Electron Microscopy (FE-SEM), (JEOL JSM-7600F) and Transmission Electron Microscopy (TEM) (JEOL JEM 2100) operated at 200 kV for the morphological and structural investigations. The optical properties of NGQD were investigated using a Shimadzu UV-Vis spectrometer (model UV-2600) and the PL spectra were recorded on a FluoroLog 3-221 (HORIBA Scientific) fluorimeter equipped with a 450 W xenon lamp. To investigate the surface functionalization of the NGQD and GQD, Attenuated Total Reflectance Fourier Transform Infrared (ATR-FTIR) spectroscopy (Thermo Scientific Nicolet iS5) was used. The as-prepared NGQD was also analyzed by Confocal Raman Microscopy (WITEC alpha 300R Raman spectrometer with a 532 nm Nd-YAG laser with 1800-line mm⁻¹ grating) and the size of the as-prepared NGQD was determined by Atomic Force Microscope (AFM) analysis in tapping mode under ambient conditions using a BRUKER multimode 8 AFM. The powder X-ray diffraction (PXRD) studies were carried out using Cu Kα radiation (BRUKER, ecoD8 ADVANCE) to investigate the crystalline structure and phase of the deposited materials on the surface of p-SiNWs. To investigate the impregnation of NGQD on the surface of p-SiNWs–Ta₂O₅, X-ray Photoelectron Spectra (XPS) (EscaLab: 220i-XL) were obtained with a monochromatized Mg K_α energy source (1253 eV).

Fabrication of the photocathode for photoelectrochemical measurements

The as-prepared p-SiNWs–Ta₂O₅ sample was coated with NGQD by drop-casting a 400 μL solution of NGQD and Nafion (5 : 1) (each with 100 μL in a time interval of 40 minutes) followed by drying under an IR lamp for 2 hours. The as-prepared final samples p-SiNWs–Ta₂O₅–NGQD acted as the main working electrode for PEC measurement. The back side of the electrode, which was connected with the copper wire, was covered by the insulating polymer adhesive to avoid any shorting or electrocatalytic contribution from the backside of the Si-chip. The measurements were carried out in a three-electrode electrochemical workstation (Metrohm-Autolab PGSTAT302N). A double junction Ag/AgCl electrode saturated with 3 M KCl was used as a reference electrode (Metrohm) and Pt wire was taken as the counter electrode (Metrohm). Prior to the electrochemical measurements, the electrodes were rinsed several times with DI water and dried with compressed air. The distance between the working and reference electrodes was kept fixed (∼1 cm) for every electrochemical analysis. For the irradiation source of the PEC-HER reaction, a xenon lamp (AM 1.5 G filter) calibrated at a power of 100 mW cm⁻² was kept at a distance of 5 cm from the quartz cell reactor. The PEC-HER activity of the as-prepared photocathode was investigated via linear sweep voltammetry (LSV) with a scan rate of 5 mV s⁻¹ under both illuminated and dark conditions. Chronopotentiometric measurements were carried out at an applied current density of 10 mA cm⁻² for at least 10 hours in a 1 M HClO₄ electrolyte solution. The electrochemical impedance analysis (EIS) was performed at an applied voltage of −300 mV (vs. RHE) over a frequency range from 100 kHz to 0.01 Hz with an AC amplitude of 10 mV in a three-electrode system. All electrochemical analyses were carried out in 1 M HClO₄ electrolyte solution and applied potentials were converted in terms of the reversible hydrogen electrode (RHE) using the Nernst equation: E_RHE = 0.197 + 0.059 × pH + E_Ag/AgCl.

Results and discussion

The schematic representation of the fabrication of p-SiNWs via MAC-etching, followed by coating with Ta₂O₅ thin film and the design of the heterostructure with the metal-free catalyst of NGQD are shown in Fig. 1. A detailed description of each step is given in the Experimental section.

Fig. 1 (a–g) Schematic of the fabrication of vertically aligned p-SiNWs using Metal-Assisted Chemical Etching (MACE). (h) Deposition of Ta₂O₅ by PLD. (i) The decoration of NGQD on the p-SiNWs–Ta₂O₅ matrix and the formation of the p-SiNWs–Ta₂O₅–NGQD heterostructure.

The morphological and structural analysis of each step during the growth of the p-SiNWs–Ta₂O₅–NGQD heterostructure was probed by FE-SEM analysis and shown in Fig. 2. Fig. 2a shows the surface morphology of the as-dispersed PS beads, which confirms the presence of hexagonally oriented monolayer beads with a close packing geometry on the Si surface having a diameter of ∼450 nm (the inset shows a zoomed-in view). The as-dispersed monolayer of the PS beads was then etched with an optimized power of O₂ plasma to reduce the diameter from 450 nm to ∼230 nm, thereby creating a uniform gap between the PS beads as shown in Fig. 2b. After the size reduction, the sample was uniformly coated with a thin layer of ∼30 nm gold (Au), followed by the formation of a gold mesh-like architecture (Fig. S1a†) created by the dissolution of PS beads in toluene solvent. The Au mesh catalyzed the oxidation process of the Si underneath by converting it into SiO₂ in the presence of the oxidizing agent H₂O₂. HF solution was then used to dissolve the SiO₂, resulting in the formation of uniformly dispersed vertically aligned SiNWs on the Si substrate. Thereafter, the Au was removed by aqua regia treatment to get rid of any metal contamination from the Si nanowire matrix. Fig. 2c shows the top view of the as-fabricated Si nanowires having diameters in the range of ∼220–225 nm and the average gap between the nanowires was ∼100–170 nm; the corresponding high magnification image is shown in the inset. Here, the etching duration was optimized to get the nanowires with diameter and length below the desired range of ∼250 nm and 1.5 μm, respectively, which provided efficient light trapping and low exciton diffusion length for enhanced PEC-HER performance.^14,39 Subsequently, the as-grown p-SiNWs were coated with ∼40 nm Ta₂O₅ layer and the corresponding tilted image of the homogeneously patterned vertically aligned p-SiNWs–Ta₂O₅ arrays with an average length of ∼1.2 μm and diameter of ∼260 nm were observed as shown in Fig. 2d. Moreover, the thickness of the Ta₂O₅ film was optimized to maintain a conformal coating of the Si nanowires, avoiding any direct connection between the nanowires. Thus, the Ta₂O₅ layers with varying thicknesses of 15, 40, and 80 nm were grown and the respective FE-SEM images are shown in Fig. S1.† For the 15 nm coating (Fig. S1b†), only the top faces of the SiNWs were covered while the lower portions of the nanowires were left bare, which made them prone to oxidation under electrochemical analysis. On the other hand, 80 nm coating (Fig. S1d†) resulted in the blockage of gaps between the nanowires, resulting in the deterioration of effective light trapping, low catalyst loading and less accessibility of the electrolyte to the electrode surface. However, the optimized 40 nm coating of the Ta₂O₅ layer clearly represents a uniform decoration of Si nanowires with Ta₂O₅ while maintaining the equidistant gap between the nanowires, which is highly desired for the PEC-HER performance as shown in Fig. 2d and S1c.†

Fig. 2 FE-SEM images of (a) the as-deposited monolayer of PS nano-beads (∼diameter 450 nm) on the silicon surface (inset shows the respective high-magnification image); (b) size reduction to ∼230 nm using O₂ plasma etching; (c) top view of the as-prepared p-SiNWs; the inset shows a high-magnification image after MACE; (d) tilted view of the homogeneously patterned and vertically aligned p-SiNWs–Ta₂O₅ arrays where the diameter and height of the SiNWs were ∼260 nm and 1.2 μm, respectively.

To characterize the as-prepared NGQD, various techniques like UV-Vis absorption, photoluminescence, FTIR, and Raman spectroscopy have been adopted. The as-prepared solutions of GQD and NGQD showed the transparent and faint yellow colour under sunlight (inset of Fig. S2a† and 3a); however, upon irradiation with a UV lamp, both showed blue photoluminescence with a brighter emission from NGQD (inset of Fig. S2b† and 3b). The UV-Vis spectra of GQD (Fig. S2a†) confirmed the formation of the C=O bond by showing a prominent absorption peak at 362 nm corresponding to the n–π* transition, which is in agreement with the previous report.³⁸ In general, NGQD shows two characteristic peaks, one is at ∼230 nm, corresponding to the π–π* transition of the C=C bond, and another at ∼340 nm, which is attributed to the n–π transition in the C=O bond, (Fig. 3a).^38,40 Upon excitation with a 370 nm beam, the PL spectrum of GQD showed a prominent peak at 461 nm with a Stokes shift of 91 nm (Fig. S2b†).³⁸ However, the PL spectrum of the as-prepared NGQD, showed a strong peak at 423 nm for an excitation wavelength of 340 nm, confirming a Stoke's shift of 83 nm (Fig. 3b).³⁸ This blue shift of the NGQD emission spectrum upon irradiation under a UV lamp is attributed to the relatively strong electron affinity of N-functionalities in NGQD, which further confirmed the formation of NGQD from GQD.³⁸ The N-functionalities present in the NGQD were examined by the ATR-FTIR absorption spectrum, where bands at 3260–3132, 1625 cm⁻¹, and 1452, 1150 cm⁻¹ correspond to the formation of N–H, C=N, and C–N bonds, respectively, as shown in Fig. S3.†³⁸ Moreover, the Raman spectrum of the as-grown NGQD, shown in Fig. S4,† represents the peaks centered at ∼1365 and 1600 cm⁻¹ corresponding to the D and G bands, which are attributed to the in-plane A_1g and E_2g mode of vibrations, respectively.^41–43 Thus, the ratio of I_D/I_G (∼1.1), as well as the broadening of the D band in NGQD may be attributed to the incorporation of defect densities, the appearance of surface functional groups and/or the presence of non-carbonaceous elements in the graphitic matrix.⁴⁴ The broadening of the D band as compared to the G band is a further authentication of the intercalation of nitrogen atoms into the base carbon backbone creating disordered structures.⁴² The diameter of the as-grown NGQD was found to be in the range of 3–4 nm as measured in AFM analysis (Fig. S5†). The tiny NGQD particles uniformly decorated on the nanowire surface act as a catalyst in the photoelectrochemical process, as discussed in a later section. Further elucidation of the hybrid nanowire architecture and its crystallinity was achieved via in-depth TEM analysis.

Fig. 3 (a) Absorbance spectra and the inset shows the solution colour under sunlight. (b) PL spectra of NGQD at an excitation wavelength of 340 nm and the inset shows the solution colour under UV irradiation.

Fig. 4a shows the cross-sectional schematic of the designed p-SiNWs–Ta₂O₅–NGQD heterostructure where the Si nanowires having diameters in the range of ∼220 nm are uniformly decorated with 40 nm thick Ta₂O₅ (Fig. 4b and its corresponding inset). This analysis corroborates the previously observed macroscopic FE-SEM characterization (Fig. 2d). The high-resolution TEM (HR-TEM) analysis indicated the lattice fringes corresponding to Si nanowires in the (100) direction, confirming the vertically oriented growth and uniform coating with the Ta₂O₅ thin film.¹⁶ The corresponding d-spacing of ∼0.30 nm is responsible for the Si (100) plane, whereas the d-spacing of ∼0.23 nm refers to the formation of the Ta₂O₅ (111) plane corresponding to orthorhombic facets and confirms the decoration of the heterostructure interface as shown in Fig. 4c.^16,45,46 The selected area electron diffraction (SAED) pattern shows the planes corresponding to both Si and Ta₂O₅ and validates that the formation of the Si–Ta₂O₅ heterojunction is polycrystalline in nature as shown in Fig. 4d. From the SAED pattern, the corresponding lattice spacings have been calculated and are tabulated in Table S1.† However, the presence of all lattice fringes corresponding to SiNWs, Ta₂O₅, and NGQD in a single frame of HR-TEM analysis is hard to detect because the lattice fringes of NGQD (100) and Ta₂O₅ (111) fall in a nearly similar range of ∼0.21 nm and ∼0.23 nm, respectively.^16,47 Similarly, the lattice fringes of NGQD (002) and Si (100) also fall in the identical range of ∼0.34 nm and ∼0.30 nm, respectively.^16,47,48 Thus, Si (100) and NGQD (002) lattice fringes have been separately identified in a single platform (Fig. S6†) which indicates the presence of three-sub lattices in the designed heterostructure. The formation of Ta₂O₅ can also be verified by XRD analysis (Fig. S7†) where the peaks at 23.81 29.16, 36.52, and 55.52° correspond to the lattice plans of (001), (200), (201) and (1112), respectively, which come from the crystalline facets of Ta₂O₅ and the peak at 33.09° is responsible for the plane of Si (100).^49–52 The surface impregnation of NGQD decorated on top of p-SiNWs–Ta₂O₅ was probed by the XPS survey scan of the as-synthesized p-SiNWs–Ta₂O₅–NGQD photocathodes (Fig. S8†) where the presence of all the elements (Si, Ta, O, N, C) confirmed the formation of the desired heterostructure.^53–56 The deconvoluted high-resolution peaks of Si 2s and Si 2p at the binding energies of 151.18 and 100.36 eV reveal the crystalline chemical state of the Si atom which is from the p-SiNWs (Fig. S9†).^57,58 The uniform deposition of the Ta₂O₅ can be confirmed by the high-resolution spectra (Fig. S9†) of Ta 4d (230.25 eV) and O 1s (530.82 eV).^59–61 The Ta 4f spectrum was also analyzed, and it was further split into Ta 4f_7/2 (26.01 eV) and Ta 4f_5/2 (27.86 eV) due to strong spin–orbital coupling revealing the crystalline phases of Ta₂O₅.⁵⁹ The O 1s spectrum shows an additional peak at a slightly higher binding energy centered at 532.4 eV, indicating the presence of the –OH and –CO groups associated with the NGQD matrix.⁵⁹ The C 1s spectrum (Fig. S9f†) of NGQD has a major peak centered at 284.48 eV, which revealed the presence of C atoms in the C–C bonds, whereas the formation of C–OH and N–C=N in aromatic rings can be verified by the peak positions at 285.98 and 287.74 eV, respectively.^12,38,62 A small peak at 289.95 eV represents plasmon excitations of the carbon atom.¹² In the N 1s spectrum, the two primary peaks at 398.59 and 400.98 eV correspond to the pyridinic-N and graphitic-N functionalities in the NGQD matrix.^12,38,62 The higher binding energy peak at 403.86 eV indicated the presence of oxidative N–O(–C) species that may be responsible for creating the trap layer during the electron transfer process.¹²

Fig. 4 (a) Cross-sectional schematic of the designed p-SiNWs–Ta₂O₅–NGQD heterostructure. (b) TEM image of p-SiNWs decorated with a Ta₂O₅ layer on its surface, the inset reveals the zoomed-in view of the interface. (c) HR-TEM image depicting the interface of p-SiNWs (d ∼ 0.30 nm of silicon (100) planes) and Ta₂O₅ (d ∼ 0.24 nm corresponding to the (111) in-plane lattice spacing) depicting the lattice fringes of Si and Ta₂O₅, respectively. (d) Selective area electron diffraction (SAED) pattern confirming the planes corresponding to both Si and Ta₂O₅, validating the presence of the Si–Ta₂O₅ heterojunction.

The electrochemical performance was analyzed via linear sweep voltammetry (LSV) measurements in a conventional three-electrode set-up connected to a potentiostat in 1 M HClO₄ solution. The as-prepared p-SiNWs, p-SiNWs–NGQD, and p-SiNWs–Ta₂O₅–NGQD photocathodes have been tested under both dark and illuminated (100 mW cm⁻²) conditions as shown in Fig. S10† and 5, respectively. A rapid increase in the photocurrent density (J_ph) was observed for p-SiNWs–NGQD as compared to bare p-SiNWs and finally reached a maximum value of −49.5 mA cm⁻² at −0.78 V (vs. RHE) for the p-SiNWs–Ta₂O₅–NGQD photocathode as evidenced from Fig. 5a. The J_ph value of p-SiNWs–NGQD was found to be −8.5 mA cm⁻², which was ∼6 times lower as compared to the p-SiNWs–Ta₂O₅–NGQD photocathode at an applied potential of −0.78 V (vs. RHE). Moreover, it has been observed that under dark conditions, the current density (J_dark) value was reduced to almost one-fifth and was measured to be −10.05 mA cm⁻² for the p-SiNWs–Ta₂O₅–NGQD photocathode. To investigate the photoelectrochemical effect of p-SiNWs–Ta₂O₅–NGQD with the variation of the thickness of Ta₂O₅ as a passivation layer, three different samples with Ta₂O₅ thicknesses of 15 nm, 40 nm, and 80 nm along with NGQD coating were prepared and the respective LSV plots are shown in Fig. S11.† It is evident from the LSV plots that the photocathode having the 40 nm Ta₂O₅ passivation layer provided the maximum photocurrent density as compared to all other thicknesses. This is attributed to the uniform decoration of the Si nanowires by the 40 nm Ta₂O₅ layer, which provided a conformal coating and enabled efficient charge separation and transfer throughout the nanowire length, as well as the accessibility of the electrolyte to the effectively exposed nanowire surface. In contrast, a 15 nm coating of Ta₂O₅ was deposited only at the top portion of the SiNWs, however, 80 nm deposition entirely blocked the gaps between the nanowires and thereby hindered the uniform catalyst loading throughout the nanowire surface and the facile accessibility of the electrolyte, which in turn led to low photocurrent density as shown in Fig. S11.† The optimized p-SiNWs–Ta₂O₅–NGQD photocathode performance towards the PEC-HER was investigated in terms of low overpotentials of −449 mV @ 5 mA cm⁻² and −498 mV @ 10 mA cm⁻², as compared to all other prepared electrodes, which indicates the compatibility of Ta₂O₅ with Si as an effective electron transport layer along with NGQD, as shown in Table S2.† The significant enhancement of J_ph as compared to J_dark and the low value of the overpotential proved the superiority of Ta₂O₅, which not only behaves as a protective layer but also acts as a smooth charge transport matrix because of negligible lattice mismatch with Si.²⁷

Fig. 5 (a) LSV polarization curve (without iR correction) for p-SiNWs, p-SiNWs–NGQD, and p-SiNWs–Ta₂O₅–NGQD in 1 M HClO₄ solution under illumination. (b) Tafel plot of the p-SiNWs–Ta₂O₅–NGQD and p-SiNWs–NGQD heterostructure under illuminated conditions. (c) ABPE of p-SiNWs–Ta₂O₅–NGQD and p-SiNWs–NGQD heterostructures. (d) Transient photo-response of the p-SiNWs–Ta₂O₅–NGQD heterostructure under repeated ON/OFF cycles at an interval of 10 seconds.

For an in-depth analysis of the catalytic activity and the HER reaction kinetics of p-SiNWs–Ta₂O₅–NGQD and the p-SiNWs–NGQD photoelectrode, the corresponding Tafel slopes were calculated using the Tafel equation under illuminated conditions, as shown in Fig. 5b. The relatively low value of the Tafel slope of 78 mV dec⁻¹ for the p-SiNWs–Ta₂O₅–NGQD photoelectrode as compared to p-SiNWs–NGQD (96 mV dec⁻¹) indicates a faster electron transfer via the Volmer–Heyrovsky mechanism where the electrochemical desorption following the Heyrovsky pathway is referred to as the rate-limiting step for HER kinetics.^32,63 The detailed mechanism has been discussed in the ESI.† Moreover, the exchange current density was calculated (Fig. S12†) via extrapolating the Tafel plots to the X-axis¹² and its larger value of 118 μA cm⁻² for p-SiNWs–Ta₂O₅–NGQD as compared to p-SiNWs–NGQD photocathode (53 μA cm⁻²), indicates a faster electron transfer process at the electrode/electrolyte interface with a low kinetic barrier. The electrochemical impedance spectroscopy (EIS) measurements were carried out for the p-SiNWs–Ta₂O₅–NGQD photocathode under both illuminated and dark conditions at an applied voltage of −300 mV (vs. RHE) and the corresponding Nyquist plots have been analyzed (Fig. S13†). The charge transfer resistance (R_ct) value under illumination was found to be 306 ohms, which is nearly half of that of the dark condition (663 ohm). This significant reduction in R_ct at the photocathode/electrolyte interface under illuminated conditions confirmed the role of Ta₂O₅–NGQD decoration on SiNWs as an effective moderator to hasten the charge transfer kinetics via a low barrier interface and the corresponding equivalent circuit diagram has been incorporated as the inset of Fig. S13.† The low value of R_ct under illuminated conditions is attributed to the enhanced photoexcited carrier generation, and the faster and smoother electron transfer process. The applied bias photon-to-current conversion efficiency (ABPE) was calculated using the following equation:^11,12,16

ABPE = [J_ph × (V_reverse − V_applied)/P_in] × 100%

where J_ph is the photocurrent density under applied bias V_applied, V_reverse is the standard reversible potential, in the case of the water-splitting reaction V_reverse ∼1.23 V, and P_in is the incident light intensity under AM 1.5 G conditions (∼100 mW cm⁻²). The maximum ABPE of ∼21.1% at −0.78 V vs. RHE potential for the p-SiNWs–Ta₂O₅–NGQD photoelectrode was achieved (Fig. 5c), which is ∼6 times higher as compared to p-SiNWs–NGQD. This revealed the good interface quality of the p-SiNWs–Ta₂O₅–NGQD photocathode with a low concentration of interfacial traps and defects, which usually serves as a recombination center for photo-generated excitons.⁶⁴ Thus, the effectiveness of the Ta₂O₅ layer as an electron selective hetero-contact²⁷ owing to its low conduction band offset with Si, facilitates the transport of photo-generated carriers through its conduction band via NGQD to the electrolyte. Moreover, it was considered that the choice of NGQD is highly beneficial for hydrogen evolution due to the high charge mobility, lower hydrogen binding energies and higher number of active sites.^12,32 Here, the Ta₂O₅ and NGQD heterostructure with Si plays an important role in capturing high-intensity light (DRS UV-Vis spectra shown in Fig. S14†), followed by the facile generation and separation of photo-excitons at heterojunction interfaces leading to significantly high ABPE as compared with the state-of-the-art Si-based reported photocathodes, listed in Table S3.† The detailed pictorial representation of the charge transfer mechanism is further discussed in a later section. The transient photo-response of p-SiNWs–Ta₂O₅–NGQD (Fig. 5d) was tested under repeated ON–OFF cycles at an interval of 10 seconds at an applied potential of −0.7 V and confirmed an excellent photo-response. Under illumination, the initial photo-current density (J_in/on) and the average saturated photocurrent density (J_st/on) were found to be −37.17 and −33.81 μA cm⁻². The ratio of J_st/on/J_in/on was calculated to be ∼1, which indicates a significant reduction of recombination, as well as the surface charge trapping, resulting in enhanced charge transfer rates and thereby affecting the HER performance of p-SiNWs–Ta₂O₅–NGQD.³² Additionally, the relevant rise-time and fall-time were calculated from the transient photo-response curve and were found to be 0.51 and 0.85 seconds, respectively. These low values indicate the faster recovery of the current density under both ON/OFF conditions, thereby implying efficient charge transfer and the relatively suppressed recombination of the photo-generated carriers at sub-bandgap energy levels for their participation in the proton reduction process.

The applicability of the as-designed catalyst for prolonged hydrogen generation was verified under chronopotentiometric measurement (Fig. 6a) in 1 M HClO₄ solution for at least 10 hours at an applied current density of 10 mA cm⁻² and showed a minimum potential loss of ∼5%, indicating a moderately stable p-SiNWs–Ta₂O₅–NGQD photocathode. The stability of the photocathode may be attributed to the uniform coating of the protective Ta₂O₅ layer, followed by its decoration with the non-corrosive carbonaceous matrix, which diminished the chance of Si surface oxidation under operational conditions. The cyclic stability test was performed under repeated LSV at a 5 mV s⁻¹ scan rate (Fig. 6b) and there was high cycling stability of the robust photocathode with a minimum decrease of overpotential of 3% @ −20 mA cm⁻² after 1000 cycles.

Fig. 6 (a) Chronopotentiometry measurement of p-SiNWs–Ta₂O₅–NGQD at an applied current density of 10 mA cm⁻² under illumination. (b) LSV polarization curve of the 1^st and 1000^th cycle at 5 mV S⁻¹ under a rigorous cycling process.

The extent of band-bending for the overall PEC performance of the Ta₂O₅–NGQD-modified SiNWs was compared with p-SiNWs–NGQD, p-SiNWs–Ta₂O₅, and bare SiNWs under capacitance measurements, followed by the corresponding Mott–Schottky (M–S) plots, as shown in Fig. 7a and S15,† respectively. In semiconductor electrochemistry, the M–S plot describes the capacitance value versus the potential difference between the semiconductor and the electrolyte.⁶⁵ Therefore, from the M–S plot, the qualitative analysis of the conductivity types and the carrier concentration of semiconductors can be obtained. The p-type conductivity of the Si wafer was confirmed from the negative slope of the M–S plot⁶⁶ as shown in Fig. S15a.† However, coating with Ta₂O₅ and/or NGQD imparted the positive slope, indicating the n-type character. Different parameters like flat-band potential, carrier concentration, and barrier height have been calculated from the M–S plot (Table S4†) and the detailed calculation methods⁶⁷ have been described in the ESI.† The flat band potential (E_fb) of p-SiNWs–Ta₂O₅–NGQD was found to be −0.78 V, while that of p-SiNWs–Ta₂O₅, p-SiNWs–NGQD and p-SiNWs were calculated to be −0.29 V, 0.12 V, and 0.07 V, respectively. The higher value of E_fb in the case of p-SiNWs–Ta₂O₅–NGQD augments the extent of higher band bending at the depletion region of the heterostructure and thereby signifies faster excitonic charge separation and simultaneously suppresses recombination at the electrode–electrolyte interface, leading to significantly enhanced photo-current density.^65,67 For the p-SiNWs–Ta₂O₅–NGQD photocathode, Ta₂O₅ plays a crucial role due to its low lattice mismatch, less trap-state generation, and low conduction band offset, enabling a faster electron transfer process and thereby suppressing carrier recombination. In addition, the carrier concentrations were calculated from the slope of the M–S plot and were found to be 4.19 × 10¹⁷ cm⁻³ and 6.67 × 10¹⁶ cm⁻³ for p-SiNWs–Ta₂O₅–NGQD and p-SiNWs–NGQD, respectively, which can be attributed to the enhancement of the carrier generation upon the introduction of the Ta₂O₅ passivation layer. The barrier heights of the p-SiNWs–NGQD matrix with and without metal oxide (Ta₂O₅) were found to be 0.67 and 0.22 V, respectively. Such a large barrier height for p-SiNWs–Ta₂O₅–NGQD can be attributed to the generation of a strong inversion layer at the Si surface owing to the increased band bending and consequently, it enhanced the charge-carrier generation and separation within the semiconductor heterostructure. Thus, the Ta₂O₅ layer allows electron transfer (very small lattice mismatch and low conduction band offset) and prevents hole transfer (high valence band offset) from Si, which makes it an effective passivation material for the dangling ends of Si, promoting higher band-bending and large photocurrent density, thus endorsing improved PEC performance. The overall mechanism can be understood by the bi-functional role of Ta₂O₅–NGQD, which not only enhances the absorption but also plays a crucial role as an active passivation layer (Ta₂O₅) and catalyst (NGQD). Since the band alignment of Ta₂O₅ lies in a similar range of water redox potential, it helps in transferring hot electrons, thereby facilitating the directional superior electron transfer from Si to the electrolyte assisted by the energy levels of Ta₂O₅–NGQD. Further, the charge transport mechanism is explained with the help of the schematic energy band diagram as portrayed in Fig. 7b. The band edge positions of Ta₂O₅ and NGQD have been schematically illustrated as sourced from the literature reports.^27,30,68 Upon both solar irradiation and applied potential, the majority of the photo-generated electrons moved from Si to the NGQD via the Ta₂O₅ passivation layer, whereas holes (h⁺) returned from NGQD to Si.

Fig. 7 (a) Mott–Schottky analysis of p-SiNWs–Ta₂O₅–NGQD, p-SiNWs–NGQD, and p-SiNWs–Ta₂O₅ photocathodes. (b) A schematic illustration of the charge transfer mechanism of the p-SiNWs–Ta₂O₅–NGQD photocatalyst in the PEC-HER.

In summary, the outstanding performance of the as-designed p-SiNWs–Ta₂O₅–NGQD heterostructure can be attributed to the following: (1) as compared to planar Si, the nano-texturing of Si (i.e. SiNWs) facilitates superior light absorption through multiple scattering and enhancing photo-generated carriers. (2) The high surface-to-volume ratio of the one-dimensional Si nanowire arrays provides a high surface area for catalyst loading. (3) The introduction of Ta₂O₅ as a passivation layer with a negligible lattice mismatch provides a less resistive path via reducing the defects at the interface. (4) The excellent passivation, inhibition of the recombination of e⁻ and h⁺ under light irradiation, and facile electron transfer via Ta₂O₅ promote higher band-bending and large photocurrent density, which enhance the overall performance of the p-SiNWs–Ta₂O₅–NGQD electrode. (5) The incorporation of the metal-free catalyst NGQD enables absorption in the UV region, reduces the possibility of surface poisoning, and it is known to have sharp catalytic edges and high charge mobility. (6) Additionally, NGQD facilitates a low resistance path for electron transfer towards the electrolyte and also accelerates the photo-electrochemical HER kinetics on its surface. (7) The uniform coating of the Si surface with Ta₂O₅ and its decoration with NGQD contribute to the moderate catalytic stability, which may be attributed to the negligible chance of surface oxidation as well as the non-corrosive nature of NGQD under an acidic medium. In this context, the as-grown p-SiNWs–Ta₂O₅–NGQD heterostructure creates a new avenue for delivering low-cost, stable, highly efficient silicon-based PEC-HER technology.

Conclusion

The present report demonstrates the design of vertically aligned p-SiNWs uniformly coated with Ta₂O₅ and finally decorated with a uniform metal-free (NGQD) catalyst as an efficient photocathode for the photoelectrochemical HER. The superiority of the as-designed p-SiNWs–Ta₂O₅–NGQD heterostructure towards the PEC-HER has been examined in terms of its high photocurrent density of −49.5 mA cm⁻² and ABPE of ∼21.1% at a potential of −0.78 V (vs. RHE). Moreover, the lower values of the Tafel slope (78 mV dec⁻¹) and overpotential (−498 mV @ 10 mA cm⁻²) add to its effectiveness towards HER kinetics. This remarkably improved photo electrocatalytic performance is attributed to the optimal band alignment of the heterostructure, owing to the sequential decoration of Ta₂O₅ and NGQD on the silicon nanowire surface facilitating the broadband absorption and faster excitonic separation and transportation. The cost-effective and environment-friendly approach, compatible with the ongoing Si-based technology, unveils the innovation towards the development of renewable hydrogen fuel for industry-grade applications.

Author contributions

S. R. did all the electrochemical study, characterization, data analysis and manuscript writing/revision, J. S. was involved in manuscript writing and data analysis, S. A. S. and D. B. were associated with the fabrication and characterization of as prepared materials, S. K. and D. B. were involved in Mott–Schottky analysis, S. S. Y. and A. V. helped to perform plasma etching of PS beads and SEM characterization, S. G. and S. C. were involved in deposition of Ta₂O₅ using PLD system, K. G. was associated with writing, revision and overall supervision of the work.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

For the successful execution of this work, K. G. is thankful to Nanomission (Grant No-SR-/NM/NS-91/2016), Department of Science and Technology, Govt. of India for financial support and Institute of Nano Science and Technology, Mohali, India for the research facility.

References

1. Y. Hou, E. Aydin, M. De Bastiani, C. Xiao, F. H. Isikgor, D.-J. Xue, B. Chen, H. Chen, B. Bahrami, A. H. Chowdhury, A. Johnston, S.-W. Baek, Z. Huang, M. Wei, Y. Dong, J. Troughton, R. Jalmood, A. J. Mirabelli, T. G. Allen, E. Van Kerschaver, M. I. Saidaminov, D. Baran, Q. Qiao, K. Zhu, S. De Wolf and E. H. Sargent, Science, 2020, 367, 1135–1140.
2. A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253–278.
3. S. Chu and A. Majumdar, Nature, 2012, 488, 294–303.
4. N. S. Lewis and D. G. Nocera, Proc. Natl. Acad. Sci., 2006, 103, 15729–15735.
5. A. Angelis-Dimakis, M. Biberacher, J. Dominguez, G. Fiorese, S. Gadocha, E. Gnansounou, G. Guariso, A. Kartalidis, L. Panichelli, I. Pinedo and M. Robba, Renewable Sustainable Energy Rev., 2011, 15, 1182–1200.
6. M. Grätzel, Nature, 2001, 414, 338–344.
7. O. Khaselev, Science, 1998, 280, 425–427.
8. A. J. Bard and M. A. Fox, Acc. Chem. Res., 1995, 28, 141–145.
9. C. Tang, R. Zhang, W. Lu, Z. Wang, D. Liu, S. Hao, G. Du, A. M. Asiri and X. Sun, Angew. Chem., 2017, 129, 860–864.
10. Y. Lin, C. Battaglia, M. Boccard, M. Hettick, Z. Yu, C. Ballif, J. W. Ager and A. Javey, Nano Lett., 2013, 13, 5615–5618.
11. D. Ghosh, K. Roy, K. Sarkar, P. Devi and P. Kumar, ACS Appl. Mater. Interfaces, 2020, 12, 28792–28800.
12. S. Riyajuddin, S. Kumar, D. Badhwar, S. A. Siddiqui, J. Sultana and K. Ghosh, Sustainable Energy Fuels, 2021, 5(12), 3160–3171.
13. Y. Yang, M. Wang, P. Zhang, W. Wang, H. Han and L. Sun, ACS Appl. Mater. Interfaces, 2016, 8, 30143–30151.
14. S. Huang, H. Zhang, Z. Wu, D. Kong, D. Lin, Y. Fan, X. Yang, Z. Zhong, S. Huang, Z. Jiang and C. Cheng, ACS Appl. Mater. Interfaces, 2014, 6, 12111–12118.
15. I. Oh, J. Kye and S. Hwang, Nano Lett., 2012, 12, 298–302.
16. K. Roy, D. Ghosh, K. Sarkar, P. Devi and P. Kumar, ACS Appl. Mater. Interfaces, 2020, 12, 37218–37226.
17. B. Weng, F. Xu and J. Xu, Nanotechnology, 2014, 25, 455402.
18. P. D. Tran, S. S. Pramana, V. S. Kale, M. Nguyen, S. Y. Chiam, S. K. Batabyal, L. H. Wong, J. Barber and J. Loo, Chem.–Eur. J., 2012, 18, 13994–13999.
19. C. Lv, Z. Chen, Z. Chen, B. Zhang, Y. Qin, Z. Huang and C. Zhang, J. Mater. Chem. A, 2015, 3, 17669–17675.
20. D. Zhang, J. Shi, W. Zi, P. Wang and S. Liu, ChemSusChem, 2017, 10, 4324–4341.
21. L. Ji, M. D. McDaniel, S. Wang, A. B. Posadas, X. Li, H. Huang, J. C. Lee, A. A. Demkov, A. J. Bard, J. G. Ekerdt and E. T. Yu, Nat. Nanotechnol., 2015, 10, 84–90.
22. D. V. Esposito, I. Levin, T. P. Moffat and A. A. Talin, Nat. Mater., 2013, 12, 562–568.
23. S. M. Thalluri, L. Bai, C. Lv, Z. Huang, X. Hu and L. Liu, Adv. Sci., 2020, 7, 1902102.
24. Y. Wan, S. K. Karuturi, C. Samundsett, J. Bullock, M. Hettick, D. Yan, J. Peng, P. R. Narangari, S. Mokkapati, H. H. Tan, C. Jagadish, A. Javey and A. Cuevas, ACS Energy Lett., 2018, 3, 125–131.
25. C. Battaglia, S. M. de Nicolás, S. De Wolf, X. Yin, M. Zheng, C. Ballif and A. Javey, Appl. Phys. Lett., 2014, 104, 113902.
26. S. Avasthi, W. E. McClain, G. Man, A. Kahn, J. Schwartz and J. C. Sturm, Appl. Phys. Lett., 2013, 102, 203901.
27. T. Wang, S. Liu, H. Li, C. Li, Z. Luo and J. Gong, Ind. Eng. Chem. Res., 2019, 58, 5510–5515.
28. R. Ghosh and P. K. Giri, Nanotechnology, 2016, 28, 012001.
29. Y. Sim, J. John, J. Moon and U. Sim, Appl. Sci., 2018, 8, 2046.
30. U. Sim, J. Moon, J. Lee, J. An, H.-Y. Ahn, D. J. Kim, I. Jo, C. Jeon, S. Han, B. H. Hong and K. T. Nam, ACS Appl. Mater. Interfaces, 2017, 9, 3570–3580.
31. U. Sim, T.-Y. Yang, J. Moon, J. An, J. Hwang, J.-H. Seo, J. Lee, K. Y. Kim, J. Lee, S. Han, B. H. Hong and K. T. Nam, Energy Environ. Sci., 2013, 6, 3658–3664.
32. U. Sim, J. Moon, J. An, J. H. Kang, S. E. Jerng, J. Moon, S.-P. Cho, B. H. Hong and K. T. Nam, Energy Environ. Sci., 2015, 8, 1329–1338.
33. N. Ou, H. Li, B. Lyu, B. Gui, X. Sun, D. Qian, Y. Jia, X. Wang and J. Yang, Catalysts, 2019, 9, 345.
34. S. Riyajuddin, S. K. T. Aziz, S. Kumar, G. D. Nessim and K. Ghosh, ChemCatChem, 2020, 12, 1394–1402.
35. X.-Q. Bao, D. Y. Petrovykh, P. Alpuim, D. G. Stroppa, N. Guldris, H. Fonseca, M. Costa, J. Gaspar, C. Jin and L. Liu, Nano Energy, 2015, 16, 130–142.
36. S. M. Thalluri, J. Borme, K. Yu, J. Xu, I. Amorim, J. Gaspar, L. Qiao, P. Ferreira, P. Alpuim and L. Liu, Nano Res., 2018, 11, 4823–4835.
37. Z. Huang, N. Geyer, P. Werner, J. de Boor and U. Gösele, Adv. Mater., 2011, 23, 285–308.
38. Y.-N. Hao, H.-L. Guo, L. Tian and X. Kang, RSC Adv., 2015, 5, 43750–43755.
39. S. M. Thalluri, J. Borme, D. Xiong, J. Xu, W. Li, I. Amorim, P. Alpuim, J. Gaspar, H. Fonseca, L. Qiao and L. Liu, Sustainable Energy Fuels, 2018, 2, 978–982.
40. D. Qu, M. Zheng, L. Zhang, H. Zhao, Z. Xie, X. Jing, R. E. Haddad, H. Fan and Z. Sun, Sci. Rep., 2014, 4, 5294.
41. N.-J. Kuo, Y.-S. Chen, C.-W. Wu, C.-Y. Huang, Y.-H. Chan and I.-W. P. Chen, Sci. Rep., 2016, 6, 30426.
42. Y. Li, Y. Zhao, H. Cheng, Y. Hu, G. Shi, L. Dai and L. Qu, J. Am. Chem. Soc., 2012, 134, 15–18.
43. S. Riyajuddin, S. Kumar, K. Soni, S. P. Gaur, D. Badhwar and K. Ghosh, Nanotechnology, 2019, 30, 385702.
44. S. Riyajuddin, S. Kumar, S. P. Gaur, A. Sud, T. Maruyama, M. E. Ali and K. Ghosh, Nanotechnology, 2020, 31, 295501.
45. K. N. Manukumar, B. Kishore, R. Viswanatha and G. Nagaraju, J. Solid State Electrochem., 2020, 24, 1067–1074.
46. G. Nagaraju, K. Karthik and M. Shashank, Microchem. J., 2019, 147, 749–754.
47. Y. Dong, H. Pang, H. B. Yang, C. Guo, J. Shao, Y. Chi, C. M. Li and T. Yu, Angew. Chem., Int. Ed., 2013, 52, 7800–7804.
48. B. Zheng, Y. Chen, P. Li, Z. Wang, B. Cao, F. Qi, J. Liu, Z. Qiu and W. Zhang, Nanophotonics, 2017, 6, 259–267.
49. Y. C. Lam, H. Y. Zheng, R. T. Tjeung and X. Chen, J. Phys. D: Appl. Phys., 2009, 42, 042004.
50. K. Miura, T. Osawa, Y. Yokota, T. Suzuki and O. Hanaizumi, Results Phys., 2014, 4, 148–149.
51. S. Zhou, Z. Deng, Z. Wu, M. Xie, Y. Tian, Y. Wu, J. Liu, G. Li and Q. He, Nanomaterials, 2019, 9, 811.
52. R. R. Krishnan, K. G. Gopchandran, V. P. MahadevanPillai, V. Ganesan and V. Sathe, Appl. Surf. Sci., 2009, 255, 7126–7135.
53. L. F. Edge, D. G. Schlom, R. T. Brewer, Y. J. Chabal, J. R. Williams, S. A. Chambers, C. Hinkle, G. Lucovsky, Y. Yang, S. Stemmer, M. Copel, B. Holländer and J. Schubert, Appl. Phys. Lett., 2004, 84, 4629–4631.
54. R. Bywalez, H. Karacuban, H. Nienhaus, C. Schulz and H. Wiggers, Nanoscale Res. Lett., 2012, 7, 76.
55. M. Hellwig, A. Milanov, D. Barreca, J.-L. Deborde, R. Thomas, M. Winter, U. Kunze, R. A. Fischer and A. Devi, Chem. Mater., 2007, 19, 6077–6087.
56. D. Y. Osadchii, A. I. Olivos-Suarez, A. V. Bavykina and J. Gascon, Langmuir, 2017, 33, 14278–14285.
57. C. Albonetti, J. Martinez, N. S. Losilla, P. Greco, M. Cavallini, F. Borgatti, M. Montecchi, L. Pasquali, R. Garcia and F. Biscarini, Nanotechnology, 2008, 19, 435303.
58. S. C. Ray, D. K. Mishra, A. M. Strydom and P. Papakonstantinou, J. Appl. Phys., 2015, 118, 115302.
59. K. Kosiel, K. Pągowska, M. Kozubal, M. Guziewicz, K. Lawniczak-Jablonska, R. Jakieła, Y. Syryanyy, T. Gabler and M. Śmietana, J. Vac. Sci. Technol., A, 2018, 36, 031505.
60. R. V. Gonçalves, R. Wojcieszak, P. M. Uberman, S. R. Teixeira and L. M. Rossi, Phys. Chem. Chem. Phys., 2014, 16, 5755–5762.
61. H. Cho, K. W. Park, J. H. Ahn, C. H. Park, H. J. Cho, S. J. Yeom, K. Hong and N. J. Kwak, ECS Trans., 2014, 61, 21.
62. L. Tang, R. Ji, X. Li, G. Bai, C. P. Liu, J. Hao, J. Lin, H. Jiang, K. S. Teng, Z. Yang and S. P. Lau, ACS Nano, 2014, 8, 6312–6320.
63. S. Riyajuddin, K. Azmi, M. Pahuja, S. Kumar, T. Maruyama, C. Bera and K. Ghosh, ACS Nano, 2021, 15, 5586–5599.
64. P. Perez-Rodriguez, D. Cardenas-Morcoso, I. A. Digdaya, A. M. Raventos, P. Procel, O. Isabella, S. Gimenez, M. Zeman, W. A. Smith and A. H. M. Smets, ChemSusChem, 2018, 11, 1797–1804.
65. B. Liu, S. Feng, L. Yang, C. Li, Z. Luo, T. Wang and J. Gong, Energy Environ. Sci., 2020, 13, 221–228.
66. Z. Luo, T. Wang and J. Gong, Chem. Soc. Rev., 2019, 48, 2158–2181.
67. I. A. Digdaya, G. W. P. Adhyaksa, B. J. Trześniewski, E. C. Garnett and W. A. Smith, Nat. Commun., 2017, 8, 15968.
68. W.-J. Chun, A. Ishikawa, H. Fujisawa, T. Takata, J. N. Kondo, M. Hara, M. Kawai, Y. Matsumoto and K. Domen, J. Phys. Chem. B, 2003, 107, 1798–1803.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1se01280k

---