Synthesis, Structural and Optical Properties of ZrBi2Se6 Nanoflowers: A Next-Generation Semiconductor Alloy Material for Optoelectronic Applications.

ZrBi2Se6 nanoflower-like morphology was successfully prepared using a solvothermal method, followed by a quenching process for photoelectrochemical water splitting applications. The formation of ZrBi2Se6 was confirmed by field emission scanning electron microscopy (FE-SEM), X-ray diffraction (XRD), Raman spectroscopy, and X-ray photoelectron spectroscopy (XPS). The estimated value of work function and band gap were found to be 5.5 and 2.26 eV measured using diffuse reflection spectroscopy and ultraviolet photoelectron spectroscopy, suggesting the potential candidate for water splitting. The highest current density of 9.7 μA/cm2 has been observed for the ZrBi2Se6 photoanode for the applied potential of 0.5 V vs SCE. The flat-band potential value was -0.46 V, and the 1.85 nm width of the depletion region is estimated from the Mott-Schottky (MS) analysis. It also reveals that the charge carrier density for the ZrBi2Se6 nanoflowers is 4.8 × 1015 cm-3. The negative slope of the MS plot indicates that ZrBi2Se6 is a p-type semiconductor. It was observed that ZrBi2Se6 nanoflowers had a high charge transfer resistance of ∼730 kΩ and equivalent capacitance of ∼40 nF calculated using electrochemical impedance spectroscopy (EIS) measurements. Using chronoamperometry, the estimated rise time and decay time were 50 ms and 0.25 s, respectively, which reveals the fast photocurrent response and excellent PEC performance of the ZrBi2Se6 photoanode. Furthermore, an attempt has been made to explain the PEC activity of ZrBi2Se6 nanoflowers using an energy band diagram. Thus, the initial results on ZrBi2Se6 nanoflowers appear promising for the PEC activity toward water splitting.

Introduction

The production of hydrogen is a renewable, clean, and carbon-free process. It shows a remarkable energy density of (140 MJ/kg),[1] which will solve the current energy crisis and environmental deterioration due to fast industrialization and population enlargement. Hydrogen is an environmentally friendly source of energy. It has been applied widely in fuel-cell electric vehicles refining petroleum, producing fertilizers, treating metals, and processing foods.[2−4] On the other hand, traditional fossil fuels release toxic and greenhouse gases and will soon exhaust in the upcoming few decades. Thus, it is high time to search for a way to generate energy with minimum damage to mother nature, so hydrogen is one of the most promising candidates. Many efforts have been made to produce hydrogen as fuel throughout the world. Cheng et al.(5) obtained solar to hydrogen (STH) efficiencies of 18.5 and 19.3% in neutral and acidic electrolytes, respectively. Karuturi et al.[6] have developed self-driven solar water splitting with 17.6% STH efficiency. Varadhan et al.(7) fabricated a monolithically integrated photoelectrochemical (PEC) cell. They achieved a remarkable STH efficiency of ∼9% and better stability of ∼150 h with the photoanode through epitaxial lift-off and transferring of the grown InGaP/GaAs into a robust Ni substrate.

Although steam methane reforming at a low temperature produces extensive hydrogen, it liberates a large amount of greenhouse gas.[8] Many green technologies such as electrocatalytic,[9,10] photocatalytic,[11−13] and the thermo-photo hybrid catalytic water splitting[14−16] have been attempted in the last decade for efficient hydrogen production from water. The PEC water splitting is highly promising and utilizes abundant solar and water resources for hydrogen production.[17,18] The search for photoelectrode material and PEC for hydrogen production became a research hotspot after the discovery of TiO2 as a photoelectrode to produce hydrogen using UV irradiation treatment by Fujishima and Honda in 1972.[19] Therefore, the growth and exploration of active and effective semiconductor materials as photoelectrodes/photocatalysts for the production of H2 are sought. However, one of the most challenging parts is identifying suitable semiconductor electrodes with high conversion efficiency, good electrical conductivity, and optical conductivity with a low electron–hole recombination rate to realize smooth charge mobility. Many bulk semiconductor nanomaterials show limited photoanode application. Therefore, they strongly need further modifications or search for new semiconductor nanomaterials, alloys/composites with the appropriate optoelectronic properties, optimum band gap, high absorption coefficient, band edge positions, etc.

Even with suitable properties, devices show poor performance because the main issue of most semiconductors is the recombination rate of photogenerated electron–hole pairs, which dominates the electron transfer rate. The new class of materials, such as topological insulators and 2D layered nanomaterials, has drawn significant interest because of the high specific surface area, fast electron transfer rate, and good light harvesting property of metallic surfaces.[20] The ultrathin geometrical structure of such layered materials shows many exciting features in the electronic and optical properties domain due to their mixed ionic-covalent characteristics.[21] The hybrid ionic-covalent characteristics are favorable for efficient charge carrier transport. These fundamental characteristics help to enhance electronic and dielectric properties because such states with mixed ionic-covalent characters have dispersive conduction and valence bands with a large static dielectric constant.[22] Additionally, it is advantageous because it reduces carrier scattering and trapping because of the strong screening of impurities and charged defects.[21,23,24] These materials show excellent surface properties and satisfy two essential PEC properties for water splitting to generate hydrogen. The first is the ideal optical band gap (∼2.0–3.00 eV). The second is that the valence band maximum energy must be more positive than that of the oxidation potential of O2/H2O (1.23 V vs NHE), and the conduction band minimum energy must be more negative than that of the reduction potential of H+/H2 (0 V vs NHE). Zirconium triselenide (ZrSe3) and bismuth selenide (Bi2Se3) are layered semiconductors and are predicted to be topological insulators.[25−28] Their fundamental band gap is indirect. However, the direct gap presents strong excitonic effects and a mixed covalent and ionic character inside the layer, resulting in excellent conduction properties through the layered structures.[29,30] Therefore, such semiconductors have been extensively studied for device fabrication. Wu and his group[31] proposed the use of ZrSe3 nanoflakes for optical limiters. Based on the results obtained by Xiong et al.,[29] ZrSe3 nanobelts were used in photodetectors under a range of visible light conditions. Likewise, Bi2Se3 thin films showed good rectifying properties suitable for photodetectors in the UV–visible–NIR region.[32] Desai et al.(33) employed it in solar cells and obtained the highest power conversion efficiency of 0.14%. Recently, nanoflowers of Bi2Se3 have been used with TiO2 as photoanodes for photoelectrochemical water splitting.[33]

There is a great interest in investigating the ZrSe3-Bi2Se3 system by exploring the Zr–Bi–Se mixed alloy system, which provides an opportunity for a tailorable and highly customizable structure with expanded material choices and structural property variations. However, there is no report in the literature on such a Zr–Bi–Se mixed alloy system to date. With this motivation, an attempt has been made to synthesize a novel semiconductor alloy, zirconium bismuth selenide (ZrBi2Se6), using the solvothermal method. The formation of ZrBi2Se6 was confirmed by XRD, Raman spectroscopy, and XPS analysis. In addition, FE-SEM has established the nanoflower morphology of ZrBi2Se6. Finally, the PEC properties of ZrBi2Se6 photoanodes were explored for water splitting under the illumination of white light. We have found that ZrBi2Se6 can be a good candidate for PEC water splitting over traditional photoanodes.

Experimental Section

Chemicals and Materials

Zirconium oxychloride octahydrate [ZrOCl2.8H2O] and bismuth acetate [(CH3CO2)3Bi], oleic acid, 1-octadecane, selenourea, and ethanol were used directly as purchased from Sigma Aldrich.

Synthesis of ZrBi2Se6 Nanoflowers and Thin Films

The two-step strategy has been used to synthesize zirconium bismuth selenide (ZrBi2Se6) nanoflowers. In the typical synthesis procedure, zirconium oxychloride octahydrate and bismuth acetate are added to 16 mL of oleic acid. The molar ratio of Zr:Bi was maintained at about 1:2. Zr and Bi salts were dissolved at 100 °C in the presence of Argon gas in a round bottom flask of 100 mL. To make the solution homogeneous, the mixture was heated for 1 h with continuous stirring at 800 rpm. In this homogeneous solution, 0.73 mmol selenourea and 25 mL of 1-octadecene were added as selenium sources, and then finally, the solution was transferred to a 100 mL Teflon-lined stainless steel autoclave. The autoclave was maintained at 180 °C for 24 h, followed by quenching in an ice bath. The resultant residue was washed with absolute ethanol several times. The black-brownish precipitate dried in the tube furnace at 200 °C for 12 h under argon gas. After drying, black powder of ZrBi2Se6 was obtained and used for further characterization. The thin films of ZrBi2Se6 have been prepared by simple electrophoretic deposition using two-electrode systems (Pt and FTO) at 25 mV potential for a 10 min duration.

Material Characterization

In the present work, various complementary techniques and spectroscopies have been carried out for structural, optical, and band structure analyses of as-synthesized ZrBi2Se6 nanoflowers. The XRD pattern of the prepared ZrBi2Se6 nanoflowers was performed using a Bruker D8 Advance X-ray diffractometer (Germany, CuKα radiation of 1.54056 Å). A JASCO, V-670 UV–visible near-infrared (UV–Vis–NIR, Japan) spectrophotometer was recorded for the diffuse reflection spectra (DRS) in the 200–800 nm range. The optical absorption and band gap of ZrBi2Se6 were estimated from the DRS spectra. The surface morphology of ZrBi2Se6 nanoflowers was studied using a scanning electron microscope (FEI Nova NanoSem 450 FE-SEM). XPS analysis of ZrBi2Se6 nanoflowers with Al Kα (1486.6 eV) radiation was recorded using Thermo Scientific, Kα, UK machine with a resolution of 0.1 eV. The binding energy was corrected for specimen charging by referencing C 1s to 284.6 eV. The ultraviolet photoemission spectroscopy (UPS) spectra were recorded using a comprehensive facility for AIPES measurements at the INDUS-2 synchrotron source at Raja Ramanna Centre for Advanced Technology (RRCAT), Indore (India). The I–V characteristics of the deposited film were recorded using a potentiostat (Metrohm Autolab-PGSTAT302N) and a solar simulator (PEC-L01) of 100 mW/cm2 intensity with AM 1.5. Electrochemical impedance spectroscopy has been studied under illumination using the same potentiostat, between 0.1 and 100 kHz. The Mott–Schottky curves were performed at the frequency of 100 Hz under dark conditions. To study the PEC activity of ZrBi2Se6 nanoflower photoanodes, linear sweep voltammetry (LSV) was used. The three-electrode system was used for the measurements at the potential region of −1.0 to 1.0 V vs SCE in 0.5 M Na2SO4 (pH = 7) electrolyte under illumination and dark conditions.

Results and Discussion

X-ray Diffraction Analysis

In the X-ray diffraction (XRD) pattern, the peak position and intensity provide the crystal structure and unit cell information. Furthermore, the atom’s position can also be predicted from the intensities of the peaks. Figure a depicts the XRD analysis of ZrBi2Se6 powder obtained using the solvothermal method. For comparison, the JCPDS XRD patterns of Bi2Se3 [Figure b] and ZrSe3 [Figure c] are also included in Figure . The recorded XRD pattern of ZrBi2Se6 reveals the contribution of major peaks from Bi2Se3 and Zr2Se3. It has been reported that if the electronegativity difference between the two bonded atoms is 0.5–2.1, then the bonds formed between them are polar covalent. The electronegativity difference (ΔEN) values between Bi–Se, Zr–Se, and Zr–Bi are 0.53, 1.22, and 0.69, respectively, and the ionic radii of the Bi, Zr, and Se atoms are 230, 230, and 190 nm, respectively. It results in the formation of polar covalent bonds between Bi–Se, Zr–Se, and Zr–Bi. Bi2Se3 and ZrSe3 have a hexagonal crystal structure, and both materials belong to layered structured families. Therefore, the atomic layer polar covalent interaction in the plane and van der Waals interaction out of the plane promotes the epitaxial growth of Zr2Se3 on Bi2Se3. Furthermore, due to Bi’s dominant valency and electronegativity, the formation of the Bi2Se3-like structure is more prominent than any other possible structure. These results show the formation of ZrBi2Se6 by the solvothermal method.

Figure 1

XRD pattern of (a) ZrBi2Se6 powder obtained using the solvothermal method, (b) JCPDS XRD pattern of Bi2Se3, and (c) JCPDS XRD pattern of ZrSe3.

The average crystallite size and lattice strain of ZrBi2Se6 were measured using the Rietveld refinement method. Figure shows the Rietveld refinement for the ZrBi2Se6 powder. For Rietveld refinement, various parameters, viz., lattice parameters, background points, zero-point parameter, scale parameter, overall thermal parameter, and half-width parameters, were varied during refinement. The analysis confirms the formation of hexagonal ZrBi2Se6 with the R3m space group. The interplanar distance for hexagonal ZrBi2Se6 is given by

Figure 2

Rietveld refinement of ZrBi2Se6 powder obtained using the solvothermal method.

The estimated values of the lattice constants for the ZrBi2Se6 hexagonal structure are a = b = 4.14 Å, c = 28.66 Å and V = 491.2 (Å)3. The Bi2Se3 single crystals belong to the family of hexagonal systems with lattice parameters a = 4.14 Å, c = 28.62 Å, and V = 424.6 (Å)3,[34] whereas ZrSe3 has the monoclinic structure with lattice parameters of a = 5·13 Å, b = 3·61 Å, and c = 9·01 Å and V = 193·5 (Å)3.[35] In the comparison, the lattice of the ZeBi2Se6 has been expanded.

The values of full width at half-maximum (FWHM) with Bragg’s angle derived from Rietveld refinement were used for Williamson–Hall (W-H) analysis to estimate lattice strain (ε) and crystalline size (D) using[36]where λ is the wavelength in Å, K is the shape factor (0.89), θB is the Bragg angle, and FWHM is represented by β in radians. The average crystallite size and lattice strain are estimated by plotting βcosθ vs 4 sinθ.

Figure shows the Williamson–Hall plot for ZrBi2Se6, which reveals the case of pure strain broadening for the present ZrBi2Se6 nanoflowers. The lattice strain and average crystallite size calculated from the slope and the fitted line intercepts are 0.0030 and 35.97 nm, respectively.

Figure 3

Williamson–Hall (W-H) plots for ZrBi2Se6 powder obtained using the solvothermal method.

Raman Spectroscopy

Raman spectroscopy is an effective technique to detect the composition, polytypism, number of layers, strain, and material defects.[37] Furthermore, it provides vital information such as the symmetry–asymmetry nature of the bond vibrations, molecular orientations, isotropic and ordered phases, and crystalline and amorphous phases.[38,39] Furthermore, the formation of ZrBi2Se6 was confirmed through Raman spectroscopy. Figure shows the Raman spectra for ZrBi2Se6 synthesized using the solvothermal method.

Figure 4

Raman spectra of the ZrBi2Se6 nanoflowers synthesized using the solvothermal method.

The unpolarized Raman spectrum shows peaks at ∼71.6, 130.8, 170.8, and 254.2 cm–1 indexed to the A1g1, Eg2, and A1g2 optical phonon modes of α-Bi2Se3 in the as-synthesized ZrBi2Se6 powder. These results are consistent with the previously reported Raman spectra for α-Bi2Se3.[40] The peak ∼254.2 cm–1 is associated with a-Se, indicating that the synthesized ZrBi2Se6 is selenium-rich.[41,42] The non-existence of low-frequency modes (<50 cm–1) in the Raman spectra may be due to the high Rayleigh background and instrument limitation. These results follow the previously reported literature for ZrSe2[37] and Bi2Se3.[43]

X-ray Photoelectron Spectroscopy (XPS)

The electronic structure and chemical properties of the ZrBi2Se6 nanoflowers were analyzed qualitatively using high-resolution X-ray photoelectron spectroscopy (XPS). Figure a shows the survey XPS spectra of the ZrBi2Se6 sample. As seen, the peaks corresponding to zirconium [Zr(3d)], bismuth [Bi(4f)], selenium [Se(3d)], carbon [C(1s)], and oxygen [O(1s)] appear in the XPS spectra.

Figure 5

XPS spectra of the ZrBi2Se6 sample: (a) Survey scan from 0 to 800 eV, (b) narrow scan for Zr 3d in the range 171–186 eV, (c) narrow scan for Bi 4f in the range 152–168 eV, and (d) narrow scan for Se 3d in the range 48–58 eV.

Figure b–d shows the narrow scan XPS spectra for Zr 3d, Bi 4f, and Se 3d elements. In Figure b, two peaks were observed for the narrow XPS spectra of Zr 3d. The peak at ∼176.09 eV is Zr 3d5/2, and the peak at ∼181.51 eV is due to Zr 3d3/2 with spin–orbit splitting separation at 5.32 eV. It indicates the existence of Zr4+ in the as-prepared ZrBi2Se6 sample.[44,45] Sometimes, another doublet set with Zr4+ can coexist at high-energy parts. The broadening in the core-level Zr 3d5/2 peak is mainly due to the suboxide A peak found at the lower energy side and the ZrO2 and suboxide B peaks found at the higher energy side, which is realized after deconvolution of the spectra. The Bi 4f spectrum contains two peaks at ∼156.20 and ∼161.25 eV accomplice with Bi 4f7/2 and Ag 4f5/2, respectively [Figure c]. These two peaks were disjointed with an energy value of ∼5.05 eV. The splitting in Bi 4f peaks can be attributed to the surface polarization of Se and Zr/Se in ZrBi2Se6. It confirms that the Bi exists with a +3 oxidation state. Two satellite peaks at ∼157.55 and ∼162.68 eV are found at lower binding energy sides. These satellite peaks are the secondary order XPS peaks with low intensity, close to the intense parent peak. These low-energy satellite peaks can appear due to the increased localized relaxed orbital occupation probability because of the charge transferred from a ligand orbital and an abrupt change in Coulombic potential as the photo rejected electron passes through the valence band.

On the other hand, the observed splitting and suppressed binding energy of the Se 3d peak is due to various chemical environments. The chemical shift in the Se core electron depends on the neighboring atom’s nature and the element’s oxidation state. Electronegativity also plays a vital role in the observed spin–orbit splitting (∼0.7 eV) for 3d5/2 and 3d3/2. Two suboxide peaks (SeO) are observed at the higher energy side near the parent Se 3d3/2 peak. These results confirm that the Se is present in their −2 oxidation state in the ZrBi2Se6 sample. Furthermore, these results show that ZrBi2Se6 is stable and is not easily oxidized under atmospheric conditions, favoring its practical applications.

Field Emission-Scanning Electron Microscopy (FE-SEM) Analysis

The surface morphology of solvothermally grown ZrBi2Se6 powder was determined using FE-SEM. Figure a,b shows FE-SEM images of ZrBi2Se6 powder at ×10000 and ×100000 resolutions. It was observed that the formation is well-organized and with flower-like nanostructures with diameters of ∼0.6–0.8 μm.

Figure 6

FESEM images of as-synthesized ZrBi2Se6 powder at different magnifications: (a) ×10,000 and (b) ×100,000.

Furthermore, after careful observation of the FE-SEM images, it is observed that ZrBi2Se6 evolved vertically oriented sharp nano pedals with porous microcavities inside it that are assembled into nanoflowers with increased surface area.

Ultraviolet Photoelectron Spectroscopy (UPS) Analysis

It is necessary to characterize the ZrBi2Se6 nanoflower–metal interface to integrate bulk nanomaterial into effective devices. For this purpose, it is essential to know the surface parameters such as work function, ionization potential, electron affinity, Fermi level, etc. Furthermore, band bending mainly influences the interface property caused by a difference in the work functions between the semiconductor and metal. Hence, it plays a crucial role in device operation. Thus, ultraviolet photoelectron spectroscopy (UPS) was used to calculate the work function (ϕ) and predict the valence band structure of ZrBi2Se6 nanoflowers. The typical UPS spectra of ZrBi2Se6 nanoflowers measured at an incident photon energy ∼39.10 eV is shown in Figure a.

Figure 7

(a) Typical UPS spectra of ZrBi2Se6 nanoflowers obtained at an incident photon energy of 40 eV. (b) Deconvoluted UPS spectra. (c) Magnified view of the valence band (VB) region to estimate the valence band edge (VBEdge). (d) Magnified view of the cutoff region to estimate cut off energy (ECutoff).

Figure b represents the deconvoluted UPS spectra of the ZrBi2Se6 nanoflowers. Figure c,d represents the magnified view of the valence band maxima (VBM) and the cut-off energy regions of the UPS spectra, respectively. The deconvoluted UPS spectra of the ZrBi2Se6 nanoflowers consist of six Gaussian peaks at ∼9.00, ∼19.41, ∼26.67, ∼29.56, ∼31.32, and ∼32.45 eV. Peak A is mainly attributed to the Se p-orbital. Peaks B, C, and D are contributed from the Bi p-orbital, Zr d-orbital, and Bi p-orbital. The last E and F peaks may result from strong hybridized Bi and Se states contributing to the p-orbital.[25,46,47]

The work function can be calculated using the equation[48]

where EPhoton = hν = incident photon energy (39.1 eV), E is the Fermi energy (0.0 eV, calibrated by using the Au standard sample), and ECut-off is the cut-off energy (33.6 eV) calculated from the UPS spectra. Thus, the value of the work function of ZrBi2Se6 nanoflowers was ∼5.56 eV. The estimated work function, ionization potential, and electron affinity of the ZrBi2Se6 nanoflowers are summarized in Table .

Table 1

Estimated Work Function (ϕ), Ionization Potential (IP), and Electron Affinity (χe) of the ZrBi2Se6 Nanoflowers

material	VBM edge (eV)	Ip (eV)	χe (eV)	ϕ (eV)
ZrBi2Se6	0.45	5.96	3.7	5.5

Diffuse Reflectance Spectroscopy (DRS)

DRS is generally applied to highly light scattering materials and absorbing particles in a matrix. Therefore, the photon energy-dependent optical properties of ZrBi2Se6 nanoflowers were studied using diffuse reflectance spectroscopy. Figure a shows the typical reflectance and absorbance plot of ZrBi2Se6 nanoflowers as a function of wavelength.

Figure 8

(a) Absorption and reflection plot as a function of wavelength of ZrBi2Se6 nanoflowers. (b) Tauc plot used to evaluate the band gap of ZrBi2Se6 using the Kubelka–Munk transformation.

If t and A are the thickness and absorbance of the film, the absorption coefficient (α) was measured by using Beer–Lambert’s law [[49]],

It has been observed that the ZrBi2Se6 nanoflowers have more than a >104 cm–1 absorption coefficient in the visible spectra region, indicating the high probability of direct transition.

To calculate the optical band gap of ZrBi2Se6 nanoflowers, the values of diffuse reflectance R have been changed to equivalent emission extinction coefficients, [F(R)], by using the Kubelka–Munk transformation,[49]

Using DRS, the analogous Tauc plots may be derived from the Kubelka–Munk function, F(R), according to equation,[50]where B is a proportionality constant. The optical band gap is measured by plotting the tangential line to the axis of the photon energy (E = hυ) in the plot of [F(R) × hυ]1/2 as a function of hυ (Tauc plot). Figure b represents the Tauc plot for the ZrBi2Se6 sample. The evaluated band gap value for ZrBi2Se6 is ∼2.26 eV, which lies in the ideal band gap range for catalytic water splitting.[51]

Photoelectrochemical (PEC) Analysis

The ZrBi2Se6 nanoflower photoelectrode was systematically investigated using LSV for its PEC activity. The measurements were conducted with a three-electrode system at the potential window between −0.4 and 1.0 V vs SCE in a 0.5 M Na2SO4 (pH = 7) electrolyte. The graph of the photocurrent density (J) vs applied potential (V) is shown in Figure a. Due to a nonfaradaic reaction, a dark current was observed for the ZrBi2Se6 nanoflower photoanode while measuring the current density. However, under illumination, the photocurrent density increases with increasing applied potential. The highest photocurrent density of 9.7 μA/cm2 was observed for the ZrBi2Se6 nanoflower photoanode at 0.5 V of applied potential. It is worth noting that no photocurrent saturation was observed for a positive bias, indicating better charge separation upon illumination. The enhancement in photocurrent density can be attributed to the high surface area of ZrBi2Se6 nanoflowers in the vicinity of the electrolyte, which helps harvest many photons. This results in an increase in the charge carriers and, hence, in photocurrent density. The Mott–Schottky (MS) and EIS analyses further support this.

Figure 9

Photoelectrochemical (PEC) activity. (a) Variation of photocurrent density as a function of applied potential. (b) Mott–Schottky plot. (c) Nyquist plot. (d) Bode plot. (e) Chronoamperometric light switching behavior. (f) Stability curve of the ZrBi2Se6 nanoflower photoelectrode.

To study the semiconductor material for photoelectrochemical (PEC) water splitting application, the most important parameters are flat-band potential (Vfb) and charge carrier density (Nd). One of the simplest and most reliable methods for estimating Vfb and Nd values is the Mott–Schottky analysis. According to this theory, the value of capacitance C at the electrode–electrolyte interface at different potentials (V) is given by[52]where εo indicates the permittivity of free space, εs is the dielectric constant, A is an effective surface area of the semiconductor electrode, q is the electronic charge, Nd is the free charge carrier density, KB represents Boltzmann’s constant, and T is the temperature in Kelvin. The flat band potential (Vfb) and charge carrier density (Nd) can be determined from the slope of the tangent of the MS plot. Using the values of Vfb and Nd in eqs and 10, the depletion region width (w) can be determined.[53]

Figure b shows the MS curve of ZrBi2Se6 nanoflower photoanodes prepared using the solvothermal method. The estimated Vfb, Nd, and w values are −0.46 V, 4.8 × 1015 cm–3, and 1.85 nm, respectively. Furthermore, the high carrier density and electrical conductivity increase the PEC performance of ZrBi2Se6 nanoflower photoanodes. In addition, the significantly low depletion width (1.85 nm) of ZrBi2Se6 nanoflowers suggests simple diffusion of photogenerated charge carriers, which is promising for PEC activity.

The suitability of the charge transfer process across the ZrBi2Se6 nanoflower photoanodes and electrolyte interface was further revealed through EIS measurements. Figure c shows the Nyquist plots for ZrBi2Se6 nanoflower samples under visible light. The Randles equivalent electrical circuit was obtained from impedance data extracted by Nyquist plot as shown in the inset of Figure c. The fitted model includes the solution resistance Rs as a series and Rp inter-facial charge transfer resistance. Cp indicates the capacitance, which offers a constant phase element (CPE) as the non-ideal capacitor in the equivalent circuit model. Cp can result from the diffusion process related to the charge transfer reaction over the interface and the specific adsorption process of the different species on the electrode surface. In addition, the semicircle radius is correlated with the charge transfer capability occurring at the semiconductor electrode/electrolyte.[54] The estimated charge transfer resistance values across the electrode–electrolyte interface and equivalent capacitance were found to be ∼730 kΩ and ∼40 nF, respectively. Therefore, for the suitability of a PEC application, the material must have low charge transfer resistance and a high equivalent capacitance value. However, a high equivalent capacitance value may also be responsible for the recombination of the charge carriers caused by the absorption of photon energy.

Variant impedance and phase-shift as a function of frequency can be revealed from the plot depicted in Figure d, which gives rise to the Bode plot. The Bode plot has prominent advantages over the Nyquist plot. The Bode plot indicates the complete transition from capacitive to resistive behavior. In the resistive behavior of the Randles-cell, the phase angle and impedance do not change with the frequency. In contrast, the imaginary component is minimal for low and high-frequency limits. Nevertheless, the cell behaves like having more capacitance for the intermediate frequencies, and impedance becomes frequency-dependent because the capacitor’s impedance begins to affect it. Therefore, due to significant imaginary components, the phase angle reaches 90°, and the cell is likely to behave as an inductor. The charge carrier (electron) lifetime (τe) was calculated according to the equation[55,56]

The calculated value of the lifetime of the charge carrier is found to be ∼0.25 ms.

Figure e indicates the time-dependent chronoamperometry photocurrent of ZrBi2Se6 nanoflowers measured under light ON and OFF at a time interval of 20 s. As seen, the photocurrent of ZrBi2Se6 nanoflower films abruptly changes under light ON and OFF conditions, showing effective photogenerated charge carrier separation. The nature of the time-dependent photocurrent curve is correlated with the light absorbing capacity and the ratio of surface to volume of ZrBi2Se6 nanoflowers. The estimated rise time (∼50 ms) and decay time (∼0.25 s) reveal the fast photocurrent response and excellent PEC performance of the ZrBi2Se6 nanoflower photoanode.

For applying ZrBi2Se6 nanoflowers as a photoanode, photoelectrochemical current stability is a critical parameter. Figure f shows the photocurrent versus time plot of the ZrBi2Se6 nanoflower photoanode recorded at 0.5 V of applied potential. It demonstrates the excellent stability of the ZrBi2Se6 nanoflower photoanode under the prevailing experimental conditions. These results further elucidate the potential candidature of ZrBi2Se6 nanoflowers as a photoanode for PEC activities.

Table depicts the comparative analysis of ZrBi2Se6 nanoflowers as a photoanode with some binary material, heterojunction, and ternary material systems.

Table 2

Comparative Analysis of ZrBi2Se6 Nanoflowers as a Photoanode with some Binary Material, Heterojunction, and Ternary Material Systems

material and morphology	synthesis method	J (μA/cm2)	Vfb (V)	Nd (cm–3)	τ	C	reference
CdS nanospheres	hot Injection	12.0					(57)
CdSe nanocrystals	hot Injection	0.08	–0.12	1.30 × 1016	29.3 μs	34.5 nF	(58)
ZnS nanospheres	hot Injection	0.29	–0.39	6.54 × 1016	4.42 μs	0.12 nF	(59)
SnSe2 nanosheets	CVD	12.0	–0.42	2.25 × 1020	126 ms		(60)
ZnWO4 1D rods	hydrothermal	32.0	+0.17				(61)
NiFe2O4 nanoplates	hydrothermal	42.0	+0.81				(61)
ZrBi2Se6 nanosheets	solvothermal	9.70	–0.46	4.80 × 1015	0.25 ms	40 nF	this work

Although the photocurrent density observed for the ZrBi2Se6 nanoflower photoanode is about 9.7 μA/cm2, the material still has a lot of scope for improvement and needs to be explored further. Nevertheless, after comparing, we found that the ZrBi2Se6 shows appreciable performance than some binary material, heterojunction, and ternary material systems. Thus, we believe that ZrBi2Se6 can be a promising material for photocatalytic water splitting applications.

Figure represents the phenomenon of p-type ZrBi2Se6 nanoflower semiconductor/Na2SO4 electrolyte band-bending and the space charge region. Figure a–c illustrates the band energy schematic between the p-type ZrBi2Se6 nanoflower semiconductor/Na2SO4 electrolyte before, after, and quasi-static equilibrium conditions under constant illumination, respectively. When a p-type semiconductor photoanode is placed in an electrolyte, it contains a redox couple like H2O/O2, and an electron transfer occurs between the solution and the photoanode though equilibrium condition is obtained. After the equilibrium condition, the p-type semiconducting electrode has an extra positive charge, and the solution has an excess negative charge. These positive charges are distributed on the depletion layer with width (w), whereas the negative charges spread over a small region between the electrolyte and p-type semiconducting electrode known as the Helmholtz layer, as shown in Figure a,b. The steady-state illumination produces a nonequilibrium electron–hole pair, expressed by the quasi-Fermi energy level. The change in the quasi-Fermi level establishes an electric field near the p-type ZrBi2Se6 nanoflower semiconductor surface and gives rise to voltage and current. The photo-voltage (VPh) or open circuit voltage (Voc) is the voltage produced by the built-in electric field of the semiconductor. Experimentally, it can be calculated by computing the potential difference between the hole and electron quasi-Fermi energy levels at no current flow.[62,63] The maximum current generated in the built-in electric field is the short circuit current (Jsc).

Figure 10

The band energy schematics of the p-type semiconductor (ZrBi2Se6 nanoflowers) and electrolyte (Na2SO4) interface indicates the relation between the electrolyte work function (ϕe), electrolyte redox couple (H2O/O2 and H2/H+), semiconductor work function (ϕs), Helmholtz layer potential drop (VH), and open circuit voltage (Voc) in three cases: (a) before equilibrium (b) after equilibration under dark conditions, and (c) in quasi-static equilibrium conditions under constant illumination.

According to Jiang et al.,[63] the water-splitting reaction requires a minimum Gibbs free energy of ∼237 kJ/mol. During the complete water splitting process, three major physiochemical processes occur. The first is the absorption of light by the semiconducting photoelectrode. A pair of charge carriers are created when a semiconductor (p- or n-type) absorbs photons. Therefore, the potential of the valence band for water oxidation should be more positive than that of the O2/H2O redox potential (1.23 V vs NHE, pH = 0), and the potential of the conduction band must be more negative than that of the H+/H2 redox potential (0 V vs NHE). In addition, the overpotential is also required to compensate for the energy losses related to the transportation of the photogenerated holes via the space charge region and electrons via the external circuit to the counter electrode. The second process is the efficient separation and transport of photogenerated electron–hole pairs with high mobility to avoid recombination of charge carriers in bulk or at the surface. The third and last process is the surface redox reaction for efficient water splitting. The potential of the charge carriers and the suitable kinetic reactions are necessary for efficient water splitting.