Tunable Bandgap Engineering of Zn
x
Cd1−x
Se Solid Solution with Controlled Ratio via a Facile One‐Pot Synthesis for Visible‐Light Photocatalytic H2 Production

Metal selenide semiconductors in photocatalysis are limited, owing to their low activity and poor stability. Herein, a facile one‐pot solution approach is developed to prepare particulate Zn x Cd1−x Se solid solutions with tunable energy band structures. X‐ray diffractometer (XRD) patterns demonstrate that the crystal structure of the samples are not changed. The analysis of UV–vis and the photoluminescence spectra exhibits that the bandgap of Zn x Cd1−x Se photocatalysts utilizing oleylamine as an organic template can be accurately controlled, which gradually becomes wider from 1.60 to 2.70 eV with increasing Zn/Cd molar ratio. Under visible‐light irradiation, the optimal Zn0.5Cd0.5Se without any cocatalyst exhibits a superior photocatalytic H2 generation rate (438.3 μmol h−1 g−1), exceeding that of pristine CdSe and ZnSe by more than 12 and 17 times, an apparent quantum yield of 1.7% at 420 nm and excellent stability. The results are on account of the balance between the bandgap width and the conduction band (CB) potential of Zn0.5Cd0.5Se, implying the excitation of more photogenerated electrons and faster charge carrier separation efficiency, which could be substantiated by the transient photocurrent response and electrochemical impedance spectroscopy. Therefore, this work provides a straightforward strategy to synthesize metal selenide for diverse photocatalytic applications.


Introduction
Photocatalytic H 2 generation proffers a promising way to address energy and environmental issues. [1] The rapid development of visible-light-responsive, high-efficiency, stable, earth-abundant, and low-cost semiconductors is crucial for the practical photocatalytic H 2 evolution application. [2] Among the class of metal dichalcogenide, metal selenides have emerged as potential photocatalysts for H 2 production from water splitting owing to their narrow bandgap as compared to the corresponding metal oxides/sulfides, but they exhibit sluggish activity and serious photocorrosion. [3] Several strategies such as controlling morphologies and phases, [4] loading cocatalysts, [5] and constructing heterojunctions [6] have been demonstrated to improve the catalytic activity of metal selenide and other nanostructured semiconductors. Moreover, the formation of metal selenidesbased solid solutions can enhance light absorption and charge transfer. [7] Notably, the more negative conduction band (CB) position of semiconductors always leads to a higher reduction power, while the wider bandgap often connotes to weaker absorption ability. [8] In this respect, constructing semiconductor solid-solution nanostructures has been proposed to modulate electrical resistivity, expand the light responses, and provide a suitable CB edge position for the improved photocatalytic H 2 evolution. [8,9] Dan et al. constructed a series of novel Cd x In 1Àx S solid solutions through a mild hydrothermal method, which presented an orderly visible-light response range from 550 Metal selenide semiconductors in photocatalysis are limited, owing to their low activity and poor stability. Herein, a facile one-pot solution approach is developed to prepare particulate Zn x Cd 1Àx Se solid solutions with tunable energy band structures. X-ray diffractometer (XRD) patterns demonstrate that the crystal structure of the samples are not changed. The analysis of UV-vis and the photoluminescence spectra exhibits that the bandgap of Zn x Cd 1Àx Se photocatalysts utilizing oleylamine as an organic template can be accurately controlled, which gradually becomes wider from 1.60 to 2.70 eV with increasing Zn/Cd molar ratio. Under visible-light irradiation, the optimal Zn 0.5 Cd 0.5 Se without any cocatalyst exhibits a superior photocatalytic H 2 generation rate (438.3 μmol h À1 g À1 ), exceeding that of pristine CdSe and ZnSe by more than 12 and 17 times, an apparent quantum yield of 1.7% at 420 nm and excellent stability. The results are on account of the balance between the bandgap width and the conduction band (CB) potential of Zn 0.5 Cd 0.5 Se, implying the excitation of more photogenerated electrons and faster charge carrier separation efficiency, which could be substantiated by the transient photocurrent response and electrochemical impedance spectroscopy. Therefore, this work provides a straightforward strategy to synthesize metal selenide for diverse photocatalytic applications.
to 600 nm and remarkable separation performance of charge carriers. [10] In analogy with the traditional Zn x Cd 1Àx S solid solutions, [11] the band structures of the corresponding metal selenides (Zn x Cd 1Àx Se) can be adjusted by tuning the ratio of Zn/Cd precursors. [6c,12] However, unlike Zn x Cd 1Àx S, the Zn x Cd 1Àx Se nanostructures were generally prepared by various synthetic routes, such as solid-state reaction, [6c] ion-exchange methods, [13] diethylenetriamine template-assisted, [14] solvothermal, [15] and a multiple-step process. [16] Nevertheless, most of the reported approaches involving Zn x Cd 1Àx Se nanostructures suffer from irregular structures and complicated preparation conditions, hindering their practical application. For instance, Yosuke et al. synthesized Zn x Cd 1Àx Se particles via a solid-state reaction in a sealed quartz ampoule. [6c] At present, there have been several reports on the relatively facile preparation and photocatalytic applications of nanoscale Zn x Cd 1Àx Se. [8,17] Additionally, the Zn x Cd 1Àx Se photocatalyst exhibits trace H 2 production due to its rapid charge carriers recombination and sluggish surface reactions. [6c] To the best of our knowledge, diethylenetriamine (DETA) generally serving as an organic template is used in complex methods to synthesize Zn x Cd 1Àx Se solid solutions. [18] Therefore, developing a simple method to synthesize uniform Zn x Cd 1Àx Se nanocrystals with tunable bandgap engineering for highly efficient photocatalytic H 2 production performance remains a considerable challenge.
In this study, a series of Zn x Cd 1Àx Se solid solutions with tunable band structures and well-distributed nanoparticle morphology with a size of 5-11 nm were prepared via a facile one-pot method, which utilized oleylamine as an organic template. The absorption edges of Zn x Cd 1Àx Se vary in the range of 460-775 nm by adjusting the Zn/Cd ratio. Notably, the Zn 0.5 Cd 0.5 Se sample presents the optimal H 2 generation rate of 438.3 μmol h À1 g À1 in the absence of noble-metal cocatalysts under visible light (λ > 420 nm), along with superior stability during cycled photocatalytic reactions. Based on various characterizations, the optimal Zn 0.5 Cd 0.5 Se solid solution possesses a suitable bandgap of 1.82 eV and the CB level (À0.82 V vs RHE), which is favorable for optimizing the electronic structure and improving carrier dynamics. This work highlights the advantages of controllable band structures in the preparation of photocatalysts with enhanced H 2 evolution activity, paving the way for further exploration of the novel ternary selenide solid-solution-based photocatalysts with superior photocatalytic activity.

Results and Discussion
The uniform Zn x Cd 1Àx Se solid solutions with different Zn/Cd ratios were prepared through a simple one-pot solution-phase approach ( Figure 1). Typically, zinc acetate dihydrate (Zn (OAc) 2 ·2H 2 O) was used as the Zn ions source and cadmium acetate dihydrate (Cd (OAc) 2 ·2H 2 O) was regarded as the source of Cd ions. The samples were synthesized through a three-necked glass flask by means of oleylamine as a template with a certain amount of selenium (Se) powder as the Se ions source, which is equal to the total molar of Zn (OAc) 2 ·2H 2 O and Cd (OAc) 2 ·2H 2 O.
As illustrated by the X-ray diffractometer (XRD) patterns in Figure 2, it can be seen that all the diffraction peaks of all the samples are less sharp and broader, which indicates that the Zn x Cd 1Àx Se solid solution has small crystal sizes. [19] Furthermore, several peaks at 25.4 , 42.1 , and 53.7 can be well-indexed with the (111), (220), and (311) planes of cubic phase of CdSe (JCPDS no. 88-2346), respectively. For ZnSe, three  respectively. The shifts of the XRD peaks toward a higher degree are ascribed to the replacement of the larger Cd 2þ (0.97 Å) with the smaller Zn 2þ (0.74 Å), which makes the crystal cell shrink, illustrating the formation of Zn x Cd 1Àx Se solid solutions. In addition, a pronounced shift toward a higher angle and close to the corresponding diffraction peaks of pure ZnSe is observed when increasing the Zn contents. This indicates that the direction of the crystal plane of the samples has not changed, and the crystal structure remains constant. [20] Figure 3a shows the UV-vis diffuse reflectance spectra of the Zn x Cd 1Àx Se samples. An absorption edge at around 775 and 460 nm can be observed in pristine CdSe and ZnSe, respectively. It exhibits a continuous shift toward shorter wavelengths in the absorption edges of Zn x Cd 1Àx Se with increasing Zn concentrations, demonstrating that the synthesized samples are Zn x Cd 1Àx Se solid solutions. [21] Moreover, it illustrates that the band structures of Zn x Cd 1Àx Se photocatalysts can be precisely controlled with adjusted Zn concentrations. [22] The photoluminescence (PL) spectra of ZnSe, CdSe, and Zn 0.5 Cd 0.5 Se were displayed in Figure S1, Supporting Information, to explore the bandgap of these photocatalysts. It presents three fluorescence peaks for these samples at %470, 720, 800 nm, which is consistent with the result of UV-vis diffuse reflectance spectra. To determine the direct band gaps of the Zn x Cd 1Àx Se samples, the Tauc plots are delineated according to the KubelkaÀMunk (KM) method. As seen from Figure 3b, the bandgap of the Zn x Cd 1Àx Se gradually becomes wider from 1.60 to 2.70 eV with increasing the Zn/Cd molar ratio, suggesting that Zn x Cd 1Àx Se solid solution possesses the controllable band structures. The narrow bandgap is beneficial to absorb more photons, which is conducive to the photoinduced electron transfer from the valence band (VB) to CB. In contrast, the wider bandgap can reduce the recombination probability of photogenerated electrons and holes. Overall, the optimal band structure of Zn x Cd 1Àx Se incurs a better transfer efficiency of photogenerated electrons and reduction capacity toward H 2 production. [17] To explore the effect of Zn concentration on the morphology, the Zn x Cd 1Àx Se solid solutions were analyzed by transmission electron microscope (TEM). It can be seen from Figure 4 that the morphology of the samples prepared by changing the Zn/ Cd molar ratio is consistent with the morphology of pure CdSe ( Figure S2a, Supporting Information) and ZnSe ( Figure S3a, Supporting Information), exemplifying a large number of uniformly dispersed nanospheres. The diameters of the aggregated nanoparticles of Zn x Cd 1Àx Se are mainly distributed in the narrow range of 5-11 nm. As depicted in Table S1, Supporting Information, the average crystallite sizes of the samples were calculated with the full width at half of maximum intensity (FWHM). The calculation data was in accord with the results of TEM. This indicates that the morphology of samples does not change in Zn x Cd 1Àx Se solid solutions. As depicted in Figure 4a,b, one group of lattice fringes can be observed in the HRTEM image of Zn 0.5 Cd 0.5 Se. The lattice spacing of 0.338 nm is between the homologous lattice spacing of pure CdSe (0.350 nm in Figure S2b, Supporting Information) and ZnSe (0.327 nm in Figure S3b, Supporting Information), which corresponds to the (111) plane. With the addition of Zn concentration in the Zn x Cd 1Àx Se solid, the fringe lattice distance is reduced due to the larger radius of Cd 2þ (0.97 Å) than that of the Zn 2þ (0.74 Å). [23] As such, it further confirms the result obtained  The surface area and porous structures of pristine CdSe, ZnSe, and Zn 0.5 Cd 0.5 Se samples were explored using nitrogen adsorption-desorption and corresponding pore size distribution curves. As displayed in Figure 5, all three photocatalysts clearly exhibited type-IV adsorption-desorption isotherms according to the Brunauer-Deming-Deming-Teller classification with a distinct IUPAC-H2 type hysteresis loop, suggesting that the well-ordered pores with interconnecting channels present in CdSe, ZnSe, and Zn 0.5 Cd 0.5 Se. [24] The inset of Figure 5 illustrates the intensive peaks that occurred at %2-40 nm in the pore size distribution curves of the three samples, which further testified the presence of mesopores. [25] Additionally, Table S2, Supporting Information, exhibited the BET surface area, average pore diameter, and pore volume of these photocatalysts. The consistent average pore diameters between 6 and 7 nm of all samples demonstrate CdSe, ZnSe, and Zn 0.5 Cd 0.5 Se possess mesoporous structures. It can be observed that only slight differences exist in BET surface area (57.3371 m 2 g À1 of CdSe, 59.1672 m 2 g À1 of Zn 0.5 Cd 0.5 Se, 63.7806 m 2 g À1 of ZnSe) and pore volume (0.1188 cm 3 g À1 of CdSe, 0.1160 cm 3 g À1 of Zn 0.5 Cd 0.5 Se, 0.1245 cm 3 g À1 of ZnSe), which are attributed to the similar morphology, structure and particle distribution. Therefore, these results implied that the BET surface area should not be crucial factors boosting the photocatalytic H 2 evolution efficiency.
To further investigate the electronic interactions in Zn x Cd 1Àx Se catalysts, the surface chemical state of the elements was obtained ( Figure S4-S6, Supporting Information). Figure S4-S5, Supporting Information, presented that the peaks at 411.8 and 405.0 eV can be attributed to Cd 3d 3/2 and Cd 3d 5/2 of Cd 2þ in CdSe, while the peaks at 1044.8 and 1021.7 eV identify Zn 2p 1/2 and Zn 2p 3/2 of Zn 2þ in ZnSe, respectively. Moreover, the binding energy of the two peaks recorded in the Se 3d region shows a similar location at 54.5 and 53.7 eV, which can be ascribed to Se 3d 3/2 and Se 3d 5/2 . [26] As demonstrated in Figure 6a,b, the high-resolution X-ray photoelectron spectroscopy (XPS) spectrum of Zn 2p and Cd 3d in the Zn 0.5 Cd 0.5 Se   matches well with pristine ZnSe and CdSe, except a slight shift toward lower energy in Cd 3d and higher energy in Zn 2p. Thus, this infers that the Zn concentration can change the Cd core level electrons due to the chemical interaction. Additionally, the apparent shifts in Zn 2p and Cd 3d signified the redistribution of charges between Zn and Cd atoms. [27] Figure 6c exhibits that the XPS spectrum of Se 3d of Zn x Cd 1Àx Se, in which the 54.5 and 53.7 eV peak positions belong to Se 3d 3/2 and Se 3d 5/2 , is consistent with the Se 3d of Zn 0.5 Cd 0.5 Se. As illustrated in Figure 6d, the VB spectra of Zn 0.5 Cd 0.5 Se sample is %1.00 eV, and based on Figure 3, the corresponding conduction band position could be calculated. Apart from Zn 0.5 Cd 0.5 Se, the high-resolution XPS spectra of other Zn x Cd 1Àx Se nanocrystals are presented in Figure S6, Supporting Information. Overall, the aforementioned results indicate that Zn x Cd 1Àx Se solid solution has been successfully synthesized, which is in agreement with the XRD and HRTEM analyses. A comparison of the hydrogen generation of the Zn x Cd 1Àx Se (x ¼ 0, 0.1, 0.3, 0.5, 0.7, 0.9, and 1.0) samples is depicted in Figure 7a. Under simulated visible light, the pristine CdSe displays a photocatalytic H 2 evolution rate of 33.9 μmol g À1 h À1 . As the Zn concentration increases, the H 2 evolution of Zn x Cd 1Àx Se becomes higher and up to 438.3 μmol g À1 h À1 for the Zn 0.5 Cd 0.5 Se (x ¼ 0.5) sample with an apparent quantum yield (AQY) of 1.7% at 420 nm, exceeding that of pure CdSe by more than 12 times. Table S3, Supporting Information, depicts the comparison of our present H 2 evolution reaction results compared to other published literature on the metal selenide photocatalysts, which signifies the robustness of our as-developed Zn 0.5 Cd 0.5 Se photocatalysts. Nevertheless, the photocatalytic performance gradually deteriorates with a further increase in Zn concentration in the Zn x Cd 1Àx Se solid solution (x ¼ 0.7, 0.9, and 1). Surprisingly, the pure ZnSe shows the lowest activity with an H 2 production rate of 25.4 μmol g À1 h À1 . Typically, the pristine CdSe exhibits negligible properties due to its narrow bandgap, which leads to the fast recombination of photogenerated carriers. Contrarily, the pure ZnSe sample cannot absorb more photons stemming from its large bandgap even if it possesses a prominent ability of H þ reduction. Unlike the aforementioned two binary sulfide samples, the Zn 0.5 Cd 0.5 Se solid solution demonstrates a balance between an excellent light-absorption capacity and an appropriate band gap that leads to the ameliorated H 2 production efficiency. As depicted in Figure 7b, no significant reduction in H 2 evolution was detected through four experimental cycles, suggesting that Zn 0.5 Cd 0.5 Se solid solution possesses high photocatalytic stability. Furthermore, the crystal structure of the recovered Zn 0.5 Cd 0.5 Se sample was studied. Figure S7 and S8, Supporting Information, demonstrate no peak position change after irradiation as evidenced in the XRD patterns and XPS spectra, testifying that there is no distinct structural alteration and chemical state change appearing at the Zn 0.5 Cd 0.5 Se solid solution. The TEM characterization ( Figure S9, Supporting Information) further confirms that Zn 0.5 Cd 0.5 Se possesses excellent stability in morphology.
Transient photocurrent response is commonly used to explore the charge separation efficiency upon visible light irradiation. As depicted in Figure 8a, pure CdSe and ZnSe show relatively low  photocurrent intensity, revealing the inferior performance of effective electron transfer. Moreover, for the Zn x Cd 1Àx Se solid solutions, an increased photocurrent intensity illustrates that the appropriate ratio of Zn/Cd dictates a predominant factor to facilitate the migration of electrons. Among all, Zn 0.5 Cd 0.5 Se exhibits a higher photocurrent density, which demonstrates that appropriate band structure of Zn 0.5 Cd 0.5 Se could accelerate the charge carriers transfer and inhibit the recombination of photoinduced electron-hole pairs to the most extent. Electrochemical impedance spectroscopy (EIS) was performed to analyze the transfer of electrons in Zn x Cd 1Àx Se solid solutions. Zn 0.5 Cd 0.5 Se sample presents a quenched arc radius compared with pristine CdSe and ZnSe (Figure 8b), highlighting superior electron transfer efficiency. [28] Figure 9 illustrates the photocatalytic H 2 generation activity at different monochromatic light. For Zn 0.5 Cd 0.5 Se solid solution, H 2 evolution rate decreases with increasing wavelength, implying that the wavelengthdependent photocatalytic H 2 production performance variation trend is approximately identical to the UV-vis absorbance spectrum. The aforementioned results revealed that the proper band structures effectively boost the photo-response and the transformation of the photo-induced electron.
To unravel the band structures, the VBs of the as-synthesized Zn x Cd 1Àx Se solid solutions were observed by XPS valence band spectra in Figure S10, Supporting Information. The bandgap and CB of Zn x Cd 1Àx Se can be adjusted by changing Zn concentrations. As shown in Figure 10, the CB for Zn x Cd 1Àx Se shifts more negative potential when the Zn/Cd molar ratio increases. Generally, the higher positions of CB in Zn x Cd 1Àx Se imply   www.advancedsciencenews.com www.advenergysustres.com the faster transfer efficiency of electrons and excellent performance toward the H 2 evolution. Despite the fact that Zn x Cd 1Àx Se (x ¼ 0.7, 0.9, and 1.0) presents more negative CB levels than Zn 0.5 Cd 0.5 Se, the photocatalytic H 2 production activity is lower. Notably, the Zn 0.5 Cd 0.5 Se sample possesses the balance between optical absorption capacity and recombination of photogenerated carriers, which exhibits a significant promotion in the H 2 generation. [10] Hence, a suitable bandgap and auspicious CB position result in the absorption of more photons and the efficient transmission of photogenerated electrons for reducing water to H 2 ( Figure 10).

Conclusion
To sum up, a simple method has been explored to design active Zn x Cd 1Àx Se solid solutions for noble-metal-free photocatalytic H 2 evolution. When the Zn/Cd molar ratio is 1:1, the Zn 0.5 Cd 0.5 Se sample without any cocatalysts exhibits the optimal H 2 production rate of 320.6 μmol g À1 h À1 at 420 nm (AQY ¼ 1.7%), which is 9 times and 12 times higher than those of pristine CdSe and pristine ZnSe, respectively. The aforementioned result is attributed to the balance between the bandgap width and the CB edge position of the Zn 0.5 Cd 0.5 Se sample, hence revealing the excitation of more photogenerated electrons and the faster separation efficiency of charge carriers. It is further evidenced by the analyses of electrochemical impedance spectroscopy and transient photocurrent characterization. As a whole, this research paves a new direction towards the synthesis of Zn x Cd 1Àx Se nanostructured materials with controlled ratio and tunable bandgap engineering toward improved H 2 generation. This investigation offers new insights into unveiling the mechanism of visible light absorption and boosted photoactivity for Zn x Cd 1Àx Se solid solutions.

Experimental Section
Materials Synthesis: In a typical procedure, 2 mmol of Se and a total of 2 mmol of Zn (OAc) 2 ·2H 2 O and Cd (OAc) 2 ·2H 2 O were added into a three-necked glass flask, heated, and stirred with a magnetic stirrer until it was completely dissolved. The mixture was heated to 280 C for 1 h, and nitrogen was introduced into the whole reaction process. At the end of the procedure, the samples were naturally cooled to room temperature. The precipitates were collected by centrifugal tube after cleaning with hexane 3 times. The precipitates were dried in an oven at 60 C for 8 h. The dried sample was ground into fine powder. To remove the oleylamine on the surface of the sample, 10 mL toluene and 10 mL 10% 3-mercaptopropionic acid (MPA) solution were added and stirred with the fine powder for 4 h. Through centrifugal cleaning, drying, and grinding, the target sample was successfully obtained. The Zn x Cd 1Àx Se solid solutions were prepared by changing the molar ratio of Zn/Cd, where x was 0.1, 0.3, 0.5 0.7, and 0.9. Besides, CdSe and ZnSe were synthesized according to the above method without adding Zn (OAc) 2 ·2H 2 O or Cd (OAc) 2 ·2H 2 O, respectively.
Characterizations: The crystal phases of the photocatalysts were determined in the range of 10-80 (2θ) by using a powder XRD, Smartlab-3KW. The crystallite size was estimated according to the Scherrer formula ¼ 0.89λ β cos θ , in which λ is the wavelength of incident X-ray radiation, β is the FWHM and θ is Bragg's diffraction angle. UV-vis spectra of the samples were recorded by a PerkinElmer, Lambda 750. Transmission electron microscope (TEM) images were conducted with a TECNAI F-30 microscope. Specific surface area and distribution of pore diameter were carried out on a Micromeritics TriStar II 3020 apparatus. The samples were degassed at 200 C for 3 h before the BET-BJH measurements. The instrument employed for XPS analysis and valence band XPS spectra were carried out on the ESCALAB 250Xi spectrometer with Al Kα (hυ ¼ 1486.6 eV) as an excitation source. The difference between the measured value and the reference value (C 1s level at 284.8 eV) was taken as the charge correction value to correct the binding energy of other elements in the spectrum. The photoluminescence (PL) spectra were performed using an FL3C-111 (HORIBA Instruments Inc.).
Photocatalytic Measurements: The amount of H 2 was measured using gas chromatography (GC-7920). The photocatalytic H 2 evolution experiments were prepared through a closed-circulation apparatus. www.advancedsciencenews.com www.advenergysustres.com First, 40 mg of the as-prepared photocatalyst was dispersed in 60 mL of 0.2 M Na 2 S/0.35 M Na 2 SO 3 aqueous solution. Afterward, the mixture was vertically irradiated with a 300 W Xe lamp (CEL-HXF300, Beijing Aulight Co., Ltd., λ > 420 nm) under magnetic conditions. The (AQY), which is defined in the following formula, was measured under the same conditions, but utilized a 420 nm bandpass filter with an FWHM of 10 nm.
AQY ¼ 2 Â the number of the evolved H 2 molecules The number of emitted photons Â 100% (1) Photoelectrochemical (PEC) Measurements: For the photoelectrochemical (PEC) measurements, transient photocurrent responses and electrochemical impedance spectroscopy (EIS) were measured with a Bio-Logic VSP-300 potentiostat in a 0.1 M Na 2 SO 4 solution under visible light irradiation through a 300 W Xe lamp (CEL-HXUV300, Beijing Aulight Co., Ltd.). The working electrode was prepared as below: 5 mg of photocatalyst powder was dispersed in 0.5 mL of ethanol and 10 μL of nafion solution followed by sonication for 4 h. After that, the slurry was distributed uniformly on the 1 Â 2 cm ITO glass, which served as the working electrode after being vacuum-dried at 150 C for 2 h. Meanwhile, Pt sheet and Ag/AgCl were used as the counter electrode and reference electrode, respectively. The transient photocurrent response was collected by implementing several light on-off cycles under a constant potential of 0.4 V. EIS was characterized by applying 0.4 V bias versus Ag/AgCl over a frequency range of 10 À1 -10 5 Hz.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.