Investigating the Influence of PbS Quantum Dot-Decorated TiO2 Photoanode Thickness on Photoelectrochemical Hydrogen Production Performance

To maximize the photoelectrochemical (PEC) hydrogen production performance of quantum dot (QD)-decorated photoelectrodes, it is crucial to prioritize the optimization of electrode’s structure, including thickness and porosity. In this study, we prepare PbS QD-decorated mesoporous TiO2 photoanodes for PEC hydrogen production, and systematically investigate the influence of the photoanode thickness on optical properties and PEC performances. As the thickness of photoanodes increases from 6.4 µm to 16.3 µm, the light absorption capability is enhanced across the entire visible and near-infrared (IR) spectrum due to the improved loading of PbS QDs. However, the photocurrent density is optimized for the 11.9 µm thick photoanode (15.19 mA/cm2), compared to the 6.4 µm thick (10.80 mA/cm2) and 16.3 µm thick photoanodes (11.93 mA/cm2). This optimization is attributed to the trade-off between the light absorption capability and the efficient mass transfer of the electrolyte as the photoanode thickness increases, which is confirmed by the lowest charge transfer resistance (Rct) evaluated from the electrochemical impedance data.


Introduction
To effectively address the rapidly increasing energy demand and the threat of global warming, the efficient utilization of clean and renewable energy is essential.Research aimed at maximizing the potential of solar energy is considered a key element in sustainable energy supply and climate change mitigation.One of the most promising approaches for the efficient utilization of solar energy and the generation of renewable and eco-friendly energy is photoelectrochemical (PEC) hydrogen production [1][2][3][4][5].
As light-absorbing materials utilized in solar energy devices, quantum dots (QDs) have garnered significant attention.QDs exhibit distinct electrical and optical properties because of the quantum confinement effect, differentiating them from bulk materials [6,7].The unique optical characteristics of QDs, coupled with advantages such as multiple-exciton generation (MEG), high absorption coefficients, and facile bandgap control, have led to substantial interest in the research fields of solar cells and PEC hydrogen pro-duction [8][9][10][11].
Moreover, introducing the QDs with a relatively small band-gap (E g < 1.5 eV) as photosensitizers allows for the effective utilization of incident light not only in the visible spectrum but even in the near-infrared range.In particular, PbS QDs can be effectively employed in PEC hydrogen production as the light-absorbing material because of their broad absorption spectrum and high quantum efficiency [12][13][14].Studies have reported the enhancement of the light absorption spectrum and outstanding PEC hydrogen production performance by decorating PbS QDs on wide band-gap semiconductors like TiO 2 and BiVO 4 [15][16][17][18].
To maximize the PEC hydrogen production performance of such QD-decorated photoelectrodes, it is crucial to prioritize the optimization of the electrode's structure, including thickness and porosity.As the thickness of the photoelectrode increases, it can absorb a larger quantity of photons per unit of active area.However, beyond a certain thickness, drawbacks such as increased electron recombination with the redox couple in the electrolyte and slowed mass transfer within the electrolyte may arise for photoelectrodes [19][20][21].Therefore, it is necessary to investigate the influence of the thickness of QD-decorated photoanodes on the performance of PEC hydrogen production.For instance, in the field of QD-sensitized solar cells, sufficient research has been conducted on the variation and optimization of photovoltaic performances according to the thickness of photoelectrodes [22,23].However, in the field of PEC hydrogen production, such research endeavors have been scarcely pursued, despite their fundamental necessity [24,25].
In this study, we coated PbS QDs onto the surface of mesoporous TiO 2 films through successive ionic-layer adsorption and reaction (SILAR) process and optimized the thickness of the photoelectrode to maximize PEC hydrogen production performance.The morphologies and uniformity, as well as the chemical states of the PbS QDs on the TiO 2 surface, were characterized.In addition, we systematically investigated the impact of the photoelectrode thickness on its optical properties and PEC hydrogen production characteristics.Additionally, we specifically explored the charge transfer properties depending on the photoelectrode thickness using electrochemical impedance analysis.Because of the superior optical properties of PbS QDs, including a wide absorption range as well as thickness optimization, the TiO 2 /PbS QD photoanode exhibited a remarkable photocurrent density of 15.19 mA/cm 2 at 0.6 V RHE under the condition of 11.9 µm in thickness.

Preparation of Mesoporous TiO 2 Film
Fluorine-doped tin oxide-coated (FTO) glass substrates (2.2 mm thickness, 8 Ω/sq, Pilkington, Tokyo, Japan) were used as the substrates for the photoelectrode.The FTO substrates were sequentially cleaned using acetone and ethanol in an ultrasonic cleaner (SD-B200H, MUJIGAE, Seoul, Republic of Korea).After cleaning, the substrates were treated with ultraviolet/ozone (UV/ozone cleaner, Yuil Ultra Violet System, YUILUV.Co., Ltd., Incheon, Republic of Korea) for 15 min to remove residual organic materials and moisture.On the cleaned FTO substrates, a solution of 7.5 wt% Ti(IV) bis(ethylacetoacetato)diiso-propoxide (Aldrich, St Louis, MO, USA) dissolved in n-butanol (Daejung, Siheung, Republic of Korea) was spin-coated, followed by heat treatment at 450 • C for 10 min in a box-type furnace.A semi-transparent TiO 2 paste (Ti-Nanoxide T/SP, SOLARONIX S.A, Aubonne, Switzerland) was coated on top of the pre-treated FTO substrate using a doctor blading method.The thickness of the TiO 2 film was controlled by varying the number of layers (1 to 3 layers) of scotch tape attached during the doctor blading.The coated TiO 2 film was then sintered by gradually increasing the temperature in the box-type furnace: 150 • C for 10 min, 250 • C for 10 min, 400 • C for 10 min, and finally 500 • C for 30 min in air.

Coating of PbS QDs on the Surface of Mesoporous TiO 2 Film
To coat PbS QDs on the surface of mesoporous TiO 2 film, the SILAR method was employed.Specifically, PbS QDs were coated by immersing the TiO 2 film on FTO substrate in a methanol solution of 0.02 M Pb(NO 3 ) 2 (Aldrich) for 60 s, followed by immersion in a solution of 0.02 M Na 2 S (Aldrich) in DI water/methanol (1:1, v/v).This process was repeated for a total of 4 cycles.Subsequently, a zinc sulfide (ZnS) passivation layer was formed by repeating the SILAR process three times using a solution of 0.05 M Zn(Ac) 2 (Aldrich) in ethanol (for 50 s), and a solution of 0.05 M Na 2 S in DI water/methanol (1:1, v/v) (for 50 s).

Characterization
The morphology of the PbS QD-coated TiO 2 nanoparticles was confirmed by highresolution transmission electron microscopy (HR-TEM; JEM-2010, JEOL, Tokyo, Japan).The cross-sectional structure of the fabricated photoelectrode was observed using scanning electron microscopy (SEM, S-4700, HITACHI, Tokyo, Japan).Elemental composition was analyzed using energy-dispersive X-ray spectroscopy (EDX) attached to an SEM (Bruker AXS Quantax 4010, HITACHI).The performance of the photoelectrode was measured using a three-electrode system consisting of the photoelectrode, Pt mesh, and Hg/HgO reference electrode in a quartz reactor containing an electrolyte composed of 0.25 M sodium sulfide pentahydrate (Daejung, Siheung, Republic of Korea) and 0.35 M sodium sulfite anhydrous (Daejung, Siheung, Republic of Korea) (pH~13).Impedance data, chronoamperometric curves, and photocurrent density-voltage (J-V) curves of the photoelectrode were analyzed using a potentiostat (M204, Autolab, Utrecht, The Netherlands).The light incident on the photoelectrode was generated using a 150 W xenon lamp (PEC-L01, Peccell, Yokohama, Japan) equipped with an AM 1.5 G filter.The light intensity was adjusted to 1 sun (100 mW/cm 2 ) using a certified silicon reference solar cell.UV-vis transmittance spectra and absorption spectra of the TiO 2 /PbS/ZnS films were achieved using UV-vis spectroscopy (OPTIZEN 2120 UV, Mecasys, Daejeon, Republic of Korea).X-ray photoelectron spectroscopy (XPS, K-alph+, Thermofisher Scientific, E. Grinstead, UK) was used to confirm the surface chemical states of the TiO 2 /PbS/ZnS films.X-ray diffraction (XRD) analyses were carried out using an X-ray diffractometer (SmartLab 9 kW system, Rigaku, Tokyo, Japan).The evolved hydrogen was analyzed by gas chromatography (GC) (Chrozen GC System, YoungIn Chromass, Anyang, Republic of Korea).

Results and Discussion
Figure 1a,b show the HR-TEM images of PbS QD-coated TiO 2 nanoparticles.After coating PbS QDs on the surface of the mesoporous TiO 2 film via the SILAR method, the film was detached from the substrate for TEM analysis.It was confirmed that TiO 2 nanoparticles, approximately 20 nm in size, were strongly and randomly bonded to each other through high-temperature annealing (Figure 1a).Particularly, the (200) lattice plane of galena PbS (fringe spacing~0.297nm) [26] was identified in a highly magnified image (Figure 1b). Figure S1 presents the HR-TEM images displaying the morphology of the bare TiO 2 nanoparticles.Figure S1a reveals a structural resemblance to that depicted in Figure 1a, suggesting the adsorption of PbS QDs onto the surface of TiO 2 nanoparticles.The (101) lattice plane of anatase TiO 2 (fringe spacing~0.353nm) [27] is identified in Figure S1b. Figure S2 represents the HR-TEM images of the PbS QD-coated TiO 2 nanoparticles.As shown in Figure S2, the average diameter of the PbS QDs is 4.81 ± 0.82 nm.The observed lattice fringe of 0.297 nm of PbS QDs that corresponds to the (200) plane, indicating the adsorption of PbS QDs over the TiO 2 nanoparticles.

Characterization
The morphology of the PbS QD-coated TiO2 nanoparticles was confirmed by highresolution transmission electron microscopy (HR-TEM; JEM-2010, JEOL, Tokyo, Japan).The cross-sectional structure of the fabricated photoelectrode was observed using scanning electron microscopy (SEM, S-4700, HITACHI, Tokyo, Japan).Elemental composition was analyzed using energy-dispersive X-ray spectroscopy (EDX) attached to an SEM (Bruker AXS Quantax 4010, HITACHI).The performance of the photoelectrode was measured using a three-electrode system consisting of the photoelectrode, Pt mesh, and Hg/HgO reference electrode in a quartz reactor containing an electrolyte composed of 0.25 M sodium sulfide pentahydrate (Daejung, Siheung, Republic of Korea) and 0.35 M sodium sulfite anhydrous (Daejung, Siheung, Republic of Korea) (pH~13).Impedance data, chronoamperometric curves, and photocurrent density-voltage (J-V) curves of the photoelectrode were analyzed using a potentiostat (M204, Autolab, Utrecht, The Netherlands).The light incident on the photoelectrode was generated using a 150 W xenon lamp (PEC-L01, Peccell, Yokohama, Japan) equipped with an AM 1.5 G filter.The light intensity was adjusted to 1 sun (100 mW/cm 2 ) using a certified silicon reference solar cell.UV-vis transmittance spectra and absorption spectra of the TiO2/PbS/ZnS films were achieved using UVvis spectroscopy (OPTIZEN 2120 UV, Mecasys, Daejeon, Republic of Korea).X-ray photoelectron spectroscopy (XPS, K-alph+, Thermofisher Scientific, E. Grinstead, UK) was used to confirm the surface chemical states of the TiO2/PbS/ZnS films.X-ray diffraction (XRD) analyses were carried out using an X-ray diffractometer (SmartLab 9 kW system, Rigaku, Tokyo, Japan).The evolved hydrogen was analyzed by gas chromatography (GC) (Chrozen GC System, YoungIn Chromass, Anyang, Republic of Korea).

Results and Discussion
Figure 1a,b show the HR-TEM images of PbS QD-coated TiO2 nanoparticles.After coating PbS QDs on the surface of the mesoporous TiO2 film via the SILAR method, the film was detached from the substrate for TEM analysis.It was confirmed that TiO2 nanoparticles, approximately 20 nm in size, were strongly and randomly bonded to each other through high-temperature annealing (Figure 1a).Particularly, the (200) lattice plane of galena PbS (fringe spacing~0.297nm) [26] was identified in a highly magnified image (Figure 1b). Figure S1 presents the HR-TEM images displaying the morphology of the bare TiO2 nanoparticles. Figure S1a reveals a structural resemblance to that depicted in Figure 1a, suggesting the adsorption of PbS QDs onto the surface of TiO2 nanoparticles.The (101) lattice plane of anatase TiO2 (fringe spacing~0.353nm) [27] is identified in Figure S1b.   Figure 2a-c shows SEM images and EDX spectra comparing TiO 2 /PbS/ZnS films with different thicknesses.As observed in the SEM images, the TiO 2 /PbS/ZnS films were coated with varying thicknesses between approximately 6 to 16 µm under each condition.The thickness of these films was controlled by varying the number of layers (one to three layers) of scotch tape attached during the doctor blading of the TiO 2 paste.The thickness of each film was found to be 6.4 µm (Figure 2a), 11.9 µm (Figure 2b), and 16.3 µm (Figure 2c), respectively.At thicknesses beyond this point, poor adhesion between the TiO 2 film and the FTO glass substrate led to the formation of cracks, making it difficult to fabricate a stable electrode.The bottom images and tables in Figure 2a-c represent the EDX spectra and the calculated atomic compositions of each electrode, implying the formation of PbS QDs and ZnS passivation layer on the surface of mesoporous TiO 2 film.From the mapping images in Figure 2d,e, it can be noted that PbS QDs dots and ZnS passivation layer were uniformly and conformally deposited along the cross-section regardless of the thickness of the TiO 2 film.These results indicate that the SILAR method we employed in this study is suitable for the coating of QDs and the passivation layer onto relatively thick mesoporous electrode surfaces.To investigate whether the coating of QDs influences the thickness of the TiO 2 electrode, we also measured the thickness of the bare TiO 2 electrode without QD coating (Figure S3).Depending on the number of layers, the thickness for each layer was determined to be 6.2 µm for a single layer (Figure S3a), 11.3 µm for two layers (Figure S3b), and 16.8 µm for three layers (Figure S3c).These values show only a minor difference within 1 µm compared to the electrode thickness after QD coating, which is considered to be within the expected margin of error associated with the Dr blading method for the TiO 2 film deposition.Furthermore, the mapping presented in Figure 2d-f indicates that PbS QDs are coated within the pores of the TiO 2 rather than forming an additional layer on the TiO 2 film.In conclusion, it can be stated that the coating of QDs does not significantly affect the thickness of the electrode.
Figure 2a-c shows SEM images and EDX spectra comparing TiO2/PbS/ZnS films with different thicknesses.As observed in the SEM images, the TiO2/PbS/ZnS films were coated with varying thicknesses between approximately 6 to 16 µm under each condition.The thickness of these films was controlled by varying the number of layers (one to three layers) of scotch tape attached during the doctor blading of the TiO2 paste.The thickness of each film was found to be 6.4 µm (Figure 2a), 11.9 µm (Figure 2b), and 16.3 µm (Figure 2c), respectively.At thicknesses beyond this point, poor adhesion between the TiO2 film and the FTO glass substrate led to the formation of cracks, making it difficult to fabricate a stable electrode.The bottom images and tables in Figure 2a-c represent the EDX spectra and the calculated atomic compositions of each electrode, implying the formation of PbS QDs and ZnS passivation layer on the surface of mesoporous TiO2 film.From the mapping images in Figure 2d,e, it can be noted that PbS QDs dots and ZnS passivation layer were uniformly and conformally deposited along the cross-section regardless of the thickness of the TiO2 film.These results indicate that the SILAR method we employed in this study is suitable for the coating of QDs and the passivation layer onto relatively thick mesoporous electrode surfaces.To investigate whether the coating of QDs influences the thickness of the TiO2 electrode, we also measured the thickness of the bare TiO2 electrode without QD coating (Figure S3).Depending on the number of layers, the thickness for each layer was determined to be 6.2 µm for a single layer (Figure S3a), 11.3 µm for two layers (Figure S3b), and 16.8 µm for three layers (Figure S3c).These values show only a minor difference within 1 µm compared to the electrode thickness after QD coating, which is considered to be within the expected margin of error associated with the Dr blading method for the TiO2 film deposition.Furthermore, the mapping presented in Figure 2d-f indicates that PbS QDs are coated within the pores of the TiO2 rather than forming an additional layer on the TiO2 film.In conclusion, it can be stated that the coating of QDs does not significantly affect the thickness of the electrode.Figure 3a is a graph analyzing XRD patterns to characterize the crystal structure and crystallinity of bare TiO 2 and TiO 2 /PbS/ZnS films.For the XRD analysis, the 11.9 µm thick film was used.Multiple diffraction peaks observed at 25.3, 37.9, 48.1, 54.0, 55.1, 63.8, 68.9, 70.3 and 74.3 2θ degrees in the XRD pattern correspond to the (101), (004), ( 200), ( 105), (211), ( 204), ( 116), ( 220) and (107) crystal planes of TiO 2 with an anatase phase [28][29][30][31][32][33].Additionally, two diffraction peaks at 30.1 and 43.0 2θ degrees for the TiO 2 /PbS/ZnS film represent the (200) and (220) planes of PbS in the galena phase [34][35][36][37][38].The Gaussian fitting for the (200) plane peak of PbS revealed a peak position at 30.2 2θ degrees and a full width at half maximum (FWHM) of approximately 1.75.The average crystallite size can be determined using Scherrer's equation [39], which is shown in Equation ( 1): where D represents the average crystallite size, λ is the X-ray wavelength (1.5406 Å), K is the dimensionless coefficient of particle shape (assumed to be 0.9), β is the FWHM in radians, and θ is the Bragg angle of diffraction.Based on this equation, the measured average crystallite size for PbS quantum dots deposited on the surface of mesoporous TiO 2 film was calculated to be approximately 4.71 nm.In our analysis, no impurity peaks, with the exception of TiO 2 and PbS, were detected, signifying the absence of other crystalline forms within the samples.The absence of the ZnS peak is ascribed to the amorphous nature of ZnS in our case [40].Figure 3a is a graph analyzing XRD patterns to characterize the crystal structure and crystallinity of bare TiO2 and TiO2/PbS/ZnS films.For the XRD analysis, the 11.9 µm thick film was used.Multiple diffraction peaks observed at 25.3, 37.9, 48.1, 54.0, 55.1, 63.8, 68.9, 70.3 and 74.3 2θ degrees in the XRD pattern correspond to the (101), (004), ( 200), ( 105), (211), ( 204), ( 116), ( 220) and (107) crystal planes of TiO2 with an anatase phase [28][29][30][31][32][33].Additionally, two diffraction peaks at 30.1 and 43.0 2θ degrees for the TiO2/PbS/ZnS film represent the (200) and (220) planes of PbS in the galena phase [34][35][36][37][38].The Gaussian fitting for the (200) plane peak of PbS revealed a peak position at 30.2 2θ degrees and a full width at half maximum (FWHM) of approximately 1.75.The average crystallite size can be determined using Scherrer's equation [39], which is shown in Equation ( 1): where D represents the average crystallite size, λ is the X-ray wavelength (1.5406 Å), K is the dimensionless coefficient of particle shape (assumed to be 0.9), β is the FWHM in radians, and θ is the Bragg angle of diffraction.Based on this equation, the measured average crystallite size for PbS quantum dots deposited on the surface of mesoporous TiO2 film was calculated to be approximately 4.71 nm.In our analysis, no impurity peaks, with the exception of TiO2 and PbS, were detected, signifying the absence of other crystalline forms within the samples.The absence of the ZnS peak is ascribed to the amorphous nature of ZnS in our case [40].are identified in the TiO 2 /PbS QD film.Figure 3c,d represents the high-resolution Pb 4f and S 2p spectra, respectively.The Pb 4f peaks exhibit binding energies (BEs) of 138.8 eV (Pb 4f 7/2 ) and 143.6 eV (Pb 4f 5/2 ), consistent with values for Pb-S bonds [41].Additionally, the S 2p peaks have BEs of 161.2 eV (S 2p 3/2 ) and 162.3 eV (S 2p 1/2 ), also corresponding to values for Pb-S bonds [42].The additional peaks at higher BEs at 166-171 eV exhibit the presence of oxidized sulfur groups, such as SO 4 2− [43] or SO 3 2− [44], indicating the partial oxidation of the PbS QD surface.This observed outcome is presumed to be a result of the SILAR process conducted in ambient air, which aligns with findings reported in previous studies [45].
Figure 4a,b shows the transmittance spectra and absorption spectra measured using UV-vis spectroscopy to evaluate the optical properties of the TiO 2 /PbS/ZnS films depending on the thicknesses.As shown in Figure 4a, the light transmittance gradually decreases with the increasing thickness of the transparent layer.Conversely, the absorbance of the film increases progressively in the wavelength range of 400 nm to 1100 nm as the thickness of the transparent layer becomes thicker, as shown in Figure 4b.These results arose from the absorption of incident light by the PbS QDs, and as the thickness of the TiO 2 /PbS/ZnS film increased, the loading of PbS QD was also enhanced, leading to a proportional increase in the absorbance.The improvement in absorbance across the entire visible and near-infrared (IR) spectrum is attributed to the narrow band-gap of PbS QDs.According to the previous literature [46], PbS exhibits an optical band-gap of approximately 0.41 eV in its bulk state; however, in the form of QDs with a size of approximately 4.8 nm, it demonstrates an optical band-gap of around 1.00 eV due to the quantum confinement effect.Given that the PbS QDs synthesized through the SILAR method in this study have a size of approximately 4.71 nm, it can be inferred that these QDs possess an optical band-gap slightly larger than 1.00 eV, absorbing photons of up to around 1200 nm wavelengths.This result indicates that as the film thickness increases, the loading of PbS QDs is enhanced, consequently improving the light absorption capability of the TiO 2 /PbS/ZnS film across the entire visible and near-IR spectrum.(Pb 4f7/2) and 143.6 eV (Pb 4f5/2), consistent with values for Pb-S bonds [41].Addition the S 2p peaks have BEs of 161.2 eV (S 2p3/2) and 162.3 eV (S 2p1/2), also correspondin values for Pb-S bonds [42].The additional peaks at higher BEs at 166-171 eV exhibi presence of oxidized sulfur groups, such as SO4 2− [43] or SO3 2− [44], indicating the pa oxidation of the PbS QD surface.This observed outcome is presumed to be a result o SILAR process conducted in ambient air, which aligns with findings reported in prev studies [45].
Figure 4a,b shows the transmittance spectra and absorption spectra measured u UV-vis spectroscopy to evaluate the optical properties of the TiO2/PbS/ZnS films dep ing on the thicknesses.As shown in Figure 4a, the light transmittance gradually decre with the increasing thickness of the transparent layer.Conversely, the absorbance o film increases progressively in the wavelength range of 400 nm to 1100 nm as the thick of the transparent layer becomes thicker, as shown in Figure 4b.These results arose f the absorption of incident light by the PbS QDs, and as the thickness of the TiO2/PbS film increased, the loading of PbS QD was also enhanced, leading to a proportiona crease in the absorbance.The improvement in absorbance across the entire visible near-infrared (IR) spectrum is attributed to the narrow band-gap of PbS QDs.Accor to the previous literature [46], PbS exhibits an optical band-gap of approximately 0.4 in its bulk state; however, in the form of QDs with a size of approximately 4.8 nm demonstrates an optical band-gap of around 1.00 eV due to the quantum confinem effect.Given that the PbS QDs synthesized through the SILAR method in this study a size of approximately 4.71 nm, it can be inferred that these QDs possess an optical b gap slightly larger than 1.00 eV, absorbing photons of up to around 1200 nm waveleng This result indicates that as the film thickness increases, the loading of PbS QDs is hanced, consequently improving the light absorption capability of the TiO2/PbS/ZnS across the entire visible and near-IR spectrum.Figure 5a shows the J-V characteristics of the TiO2/PbS/ZnS photoanodes depen on the thickness.The performance of each photoelectrode was measured under a 1 (100 mW/cm 2 ) condition using a three-electrode system within a quartz reactor.The Figure 5a shows the J-V characteristics of the TiO 2 /PbS/ZnS photoanodes depending on the thickness.The performance of each photoelectrode was measured under a 1-sun (100 mW/cm 2 ) condition using a three-electrode system within a quartz reactor.The 11.9 µm thick photoanode exhibited the highest photocurrent density, with a value of 15.19 mA/cm 2 at 0.6 V RHE .The other photoanodes showed photocurrent density values of 10.80 mA/cm 2 (6.4 µm thick) and 11.93 mA/cm 2 (16.3 µm thick), respectively, under the same condition.Compared to the 6.4 µm thick and 16.3 µm thick photoanodes, the photocurrent density was enhanced by 40.6% and 27.3%, respectively, for the optimized condition (11.9 µm thick), meaning that the photocatalytic activity of QD-decorated TiO 2 for the sulfite oxidation was greatly affected by the film thickness.The obtained photocurrent density of 15.19 mA/cm 2 from this study indicates a promising outcome for QD-decorated photoelectrodes when compared to previous reports as listed in Table S1 [22,[47][48][49][50][51][52][53].
µm thick photoanode exhibited the highest photocurrent density, with a value of mA/cm 2 at 0.6 VRHE.The other photoanodes showed photocurrent density values of mA/cm 2 (6.4 µm thick) and 11.93 mA/cm 2 (16.3 µm thick), respectively, under the condition.Compared to the 6.4 µm thick and 16.3 µm thick photoanodes, the photocu density was enhanced by 40.6% and 27.3%, respectively, for the optimized condition µm thick), meaning that the photocatalytic activity of QD-decorated TiO2 for the s oxidation was greatly affected by the film thickness.The obtained photocurrent de of 15.19 mA/cm 2 from this study indicates a promising outcome for QD-decorated p electrodes when compared to previous reports as listed in Table S1 [22,[47][48][49][50][51][52][53]. Figure 5b shows the applied bias photon-to-current efficiency (ABPE) curves c lated using the following Equation ( 2) [54,55]: where Jph represents the measured current density, Vapp is the applied external pot versus RHE and Pin is the power density of the incident light.The ABPE results ali with the observed patterns in photocurrent density.Specifically, the 11.9 µm thick toanode displayed the highest ABPE, reaching 16.17% at 0.153 V.In comparison, th µm thick photoanode exhibited an ABPE of 11.70% at 0.143 V, while the 16.3 µm photoanode demonstrated an ABPE of 13.43% at 0.109 V.
To assess photostability, chronoamperometry was conducted at 0.6 VRHE for one as shown in Figure 6a.The reduction in photocurrent density were observed as fol 20.45% for the 6.4 µm thick photoanode, and 11.19% and 11.85% for the 11.9 µm thic the 16.3 µm thick photoanode, respectively.As discussed above, as the thickness o photoanode increases, the loading of PbS QDs is improved, leading to the enhance sorbance and PEC performance of a photoanode.However, beyond a certain thresho is speculated that unfavorable factors such as increased electron recombination migh to decreased performance [19,20].Regardless of the thickness, it was observed tha photocurrent density decreased after 1 h in all three types of photoelectrodes.This lieved to be due to the decreased stability resulting from the facile oxidation of metal cogenides when used as electrodes in PEC hydrogen production [56,57].This insuffi stability observed under illumination can be improved by applying effective overl for the passivation [10,58].
As represented in Figure 6b-d, the theoretical hydrogen generation was evalu based on the chronoamperometry curves in Figure 6a, from Equation (3) [59]: Figure 5b shows the applied bias photon-to-current efficiency (ABPE) curves calculated using the following Equation (2) [54,55]: where J ph represents the measured current density, V app is the applied external potential versus RHE and P in is the power density of the incident light.The ABPE results aligned with the observed patterns in photocurrent density.Specifically, the 11.9 µm thick photoanode displayed the highest ABPE, reaching 16.17% at 0.153 V.In comparison, the 6.4 µm thick photoanode exhibited an ABPE of 11.70% at 0.143 V, while the 16.3 µm thick photoanode demonstrated an ABPE of 13.43% at 0.109 V.
To assess photostability, chronoamperometry was conducted at 0.6 V RHE for one hour, as shown in Figure 6a.The reduction in photocurrent density were observed as follows: 20.45% for the 6.4 µm thick photoanode, and 11.19% and 11.85% for the 11.9 µm thick and the 16.3 µm thick photoanode, respectively.As discussed above, as the thickness of the photoanode increases, the loading of PbS QDs is improved, leading to the enhanced absorbance and PEC performance of a photoanode.However, beyond a certain threshold, it is speculated that unfavorable factors such as increased electron recombination might lead to decreased performance [19,20].Regardless of the thickness, it was observed that the photocurrent density decreased after 1 h in all three types of photoelectrodes.This is believed to be due to the decreased stability resulting from the facile oxidation of metal chalcogenides when used as electrodes in PEC hydrogen production [56,57].This insufficient stability observed under illumination can be improved by applying effective overlayers for the passivation [10,58].
As represented in Figure 6b-d, the theoretical hydrogen generation was evaluated based on the chronoamperometry curves in Figure 6a, from Equation (3) [59]: where F represents the Faraday constant, I is the measured photocurrent, and t is the measured time.The theoretically calculated hydrogen productions for each photoanode over the one hour are as follows: 284 µmol/cm 2 (6.4 µm thick; Figure 6b), 382 µmol/cm 2 (11.9 µm thick; Figure 6c), and 298 µmol/cm 2 (16.3 µm thick; Figure 6d).Compared to the 6.4 µm thick and 16.3 µm thick photoanode, the hydrogen production was enhanced by 34.5%, 28.2%, respectively, for the optimized thickness condition (11.9 µm thick).The actual hydrogen production was measured at 0.6 V RHE for 1 h, as shown in Figure 6b-d.
For each photoanode, the faradaic efficiency was retained between 73~81%.The 6.4 µm thick photoanode produced 230 µmol/cm 2 over a period of 1 h (Figure 6b), the 11.9 µm thick photoanode produced 310 µmol/cm 2 (Figure 6c), and the 16.3 µm thick photoanode produced 233 µmol/cm 2 (Figure 6d), respectively.These results are matched well with the J-V curves for each photoanode, and indicate that the optimization of the photoanode thickness not only maximized the photocatalytic activity for the sulfite oxidation but also led to the actual maximum hydrogen production.

𝑀𝑜𝑙𝑒𝑠 𝑜𝑓 𝐻 2 𝐹
where F represents the Faraday constant, I is the measured photocurrent, and t i measured time.The theoretically calculated hydrogen productions for each photoa over the one hour are as follows: 284 µmol/cm 2 (6.4 µm thick; Figure 6b), 382 µmo (11.9 µm thick; Figure 6c), and 298 µmol/cm 2 (16.3 µm thick; Figure 6d).Compared t 6.4 µm thick and 16.3 µm thick photoanode, the hydrogen production was enhance 34.5%, 28.2%, respectively, for the optimized thickness condition (11.9 µm thick).Th tual hydrogen production was measured at 0.6 VRHE for 1 h, as shown in Figure 6b-d each photoanode, the faradaic efficiency was retained between 73~81%.The 6.4 µm photoanode produced 230 µmol/cm 2 over a period of 1 h (Figure 6b), the 11.9 µm photoanode produced 310 µmol/cm 2 (Figure 6c), and the 16.3 µm thick photoanode duced 233 µmol/cm 2 (Figure 6d), respectively.These results are matched well with t V curves for each photoanode, and indicate that the optimization of the photoanode t ness not only maximized the photocatalytic activity for the sulfite oxidation but als to the actual maximum hydrogen production.In addition, to thoroughly examine the influence of photoanode thickness on charge transfer property, we conducted impedance analysis, as shown in Figure 7 impedance spectra were obtained under the 1-sun condition (100 mW/cm 2 ) at 0.6 with a frequency range from 0.1 Hz to 100 kHz, and a sinusoidal perturbation of 10 Using the equivalent circuit shown in Figure 7, we performed fitting in the Z-view gram.The employed equivalent circuit consists of RS (solution resistance), Rct (ch transfer resistance), and CPE1 (constant phase element).The CPE represents the int cial capacitance, which generally replaces a capacitor in the equivalent circuit for trodes with high roughness [42].The CPE parameter includes two parts: CPE-T, w In addition, to thoroughly examine the influence of photoanode thickness on the charge transfer property, we conducted impedance analysis, as shown in Figure 7.The impedance spectra were obtained under the 1-sun condition (100 mW/cm 2 ) at 0.6 V RHE , with a frequency range from 0.1 Hz to 100 kHz, and a sinusoidal perturbation of 10 mV.Using the equivalent circuit shown in Figure 7, we performed fitting in the Z-view program.The employed equivalent circuit consists of R S (solution resistance), R ct (charge transfer resistance), and CPE1 (constant phase element).The CPE represents the interfacial capacitance, which generally replaces a capacitor in the equivalent circuit for electrodes with high roughness [42].The CPE parameter includes two parts: CPE-T, which represents the value of the capacitance, and CPE-P, known as the magnitude of the compression from an ideal semicircle in the Nyquist plot [60].The R ct and CPE1 are related to the charge transfer characteristics at the interface of photoanode/redox electrolyte [10,61].The values of each parameter corresponding to the thickness of the photoanode are listed in Table 1, and the value of the CPE parameters are listed in Table S2.The R S values were similar among the three photoanodes; however, the R ct value significantly decreased for the 11.9 µm thick photoanode.Compared to the 6.4 µm thick and 16.3 µm thick photoanode, it decreased by 37.1% and 30.0%, respectively, for the 11.9 µm thick photoanode, meaning that the charge transfer process occurs most efficiently for the 11.9 µm thick condition.This result Figure1a,b show the HR-TEM images of PbS QD-coated TiO2 nanoparticles.After coating PbS QDs on the surface of the mesoporous TiO2 film via the SILAR method, the film was detached from the substrate for TEM analysis.It was confirmed that TiO2 nanoparticles, approximately 20 nm in size, were strongly and randomly bonded to each other through high-temperature annealing (Figure1a).Particularly, the (200) lattice plane of galena PbS (fringe spacing~0.297nm)[26] was identified in a highly magnified image (Figure1b).FigureS1presents the HR-TEM images displaying the morphology of the bare TiO2 nanoparticles.FigureS1areveals a structural resemblance to that depicted in Figure1a, suggesting the adsorption of PbS QDs onto the surface of TiO2 nanoparticles.The (101) lattice plane of anatase TiO2 (fringe spacing~0.353nm)[27] is identified in FigureS1b.Figure S2 represents the HR-TEM images of the PbS QD-coated TiO2 nanoparticles.As shown in Figure S2, the average diameter of the PbS QDs is 4.81 ± 0.82 nm.The observed lattice fringe of 0.297 nm of PbS QDs that corresponds to the (200) plane, indicating the adsorption of PbS QDs over the TiO2 nanoparticles.

Figure 2 .
Figure 2. (a-c) SEM images and EDX spectra for TiO 2 /PbS films according to the thickness.(d-f) EDX mapping for each element (Ti, O, S, Zn, Pb) of TiO 2 /PbS films.

Figure 2 .
Figure 2. (a-c) SEM images and EDX spectra for TiO2/PbS films according to the thickness.(d-f) EDX mapping for each element (Ti, O, S, Zn, Pb) of TiO2/PbS films.

Figure
Figure 3b-d present the results of the XPS analysis conducted to verify the chemical states of the TiO 2 /PbS QD film.As seen in the survey scan of Figure 3b, Ti, O, Pb, S, and C

Figure
Figure 3b-d present the results of the XPS analysis conducted to verify the chem states of the TiO2/PbS QD film.As seen in the survey scan of Figure 3b, Ti, O, Pb, S, an are identified in the TiO2/PbS QD film.Figure 3c,d represents the high-resolution P and S 2p spectra, respectively.The Pb 4f peaks exhibit binding energies (BEs) of 138.(Pb 4f7/2) and 143.6 eV (Pb 4f5/2), consistent with values for Pb-S bonds[41].Addition the S 2p peaks have BEs of 161.2 eV (S 2p3/2) and 162.3 eV (S 2p1/2), also correspondin values for Pb-S bonds[42].The additional peaks at higher BEs at 166-171 eV exhibi presence of oxidized sulfur groups, such as SO4 2− [43] or SO3 2−[44], indicating the pa oxidation of the PbS QD surface.This observed outcome is presumed to be a result o SILAR process conducted in ambient air, which aligns with findings reported in prev studies[45].Figure4a,b shows the transmittance spectra and absorption spectra measured u UV-vis spectroscopy to evaluate the optical properties of the TiO2/PbS/ZnS films dep ing on the thicknesses.As shown in Figure4a, the light transmittance gradually decre with the increasing thickness of the transparent layer.Conversely, the absorbance o film increases progressively in the wavelength range of 400 nm to 1100 nm as the thick of the transparent layer becomes thicker, as shown in Figure4b.These results arose f the absorption of incident light by the PbS QDs, and as the thickness of the TiO2/PbS film increased, the loading of PbS QD was also enhanced, leading to a proportiona crease in the absorbance.The improvement in absorbance across the entire visible near-infrared (IR) spectrum is attributed to the narrow band-gap of PbS QDs.Accor to the previous literature[46], PbS exhibits an optical band-gap of approximately 0.4 in its bulk state; however, in the form of QDs with a size of approximately 4.8 nm demonstrates an optical band-gap of around 1.00 eV due to the quantum confinem effect.Given that the PbS QDs synthesized through the SILAR method in this study a size of approximately 4.71 nm, it can be inferred that these QDs possess an optical b gap slightly larger than 1.00 eV, absorbing photons of up to around 1200 nm waveleng This result indicates that as the film thickness increases, the loading of PbS QDs is hanced, consequently improving the light absorption capability of the TiO2/PbS/ZnS across the entire visible and near-IR spectrum.