Cu2O nanoparticles sensitize TiO2/CdS nanowire arrays to prolong charge carrier lifetime and highly enhance unassisted photoelectrochemical hydrogen generation with 4.3% efficiency

Kangbo Jianga, Wenzhong Wang*a, Jun Wangb, Tianyu Zhua, Lizhen Yaoa, Ying Chenga, Yue Wanga, Yujie Lianga and Junli Fua
aSchool of Science, Minzu University of China, Beijing, 100081, P. R. China. E-mail: wzhwang@muc.edu.cn
bCollege of Physics and Electronic Information Engineering, Fujian Provincial Key Laboratory of Functional Marine Sensing Materials, Minjiang University, Fuzhou 350108, P. R. China

Received 6th May 2020 , Accepted 10th June 2020

First published on 10th June 2020


Effective separation of charge carriers and substantial prolongation of charge carrier lifetime are two vital issues for a photoanode to achieve highly efficient photoelectrochemical (PEC) hydrogen generation. Herein, a Cu2O-nanoparticle-sensitized TiO2/CdS (TiO2/CdS/Cu2O) nanowire array photoanode is fabricated via subtly combining successive ionic layer adsorption and reaction with a chemical bath deposition method. Both the type-II band alignment with a stair-like structure and a p–n junction are integrated into this ternary photoanode. The fabricated photoanode shows a significant enhancement in the PEC H2 production with 4.3% efficiency. The measurements of light absorption, photoluminescence and electrochemical spectra undoubtedly demonstrate that the enhanced PEC H2 generation is ascribed to the remarkable enhancement of visible-light absorbing ability, efficient space charge separation and substantial prolongation of charge carrier lifetime, which are achieved by the synergetic effects of the type-II band alignment with a stair-like structure and the p–n junction. The enhanced PEC H2 generation demonstrates the potential of the TiO2/CdS/Cu2O nanowire arrays as a photoanode to efficiently convert solar energy into chemical fuels.


1. Introduction

It is well known that hydrogen (H2) with a high combustion calorific value burns in air only to produce water vapour. Thus, it is totally nonpolluting and possibly the cleanest fuel in the world. These merits make H2 one of the most promising energy sources for addressing the energy shortage and the problems of environmental contamination in the future.1–3 Therefore, it is highly important to develop techniques to efficiently produce H2. Recently, numerous studies have demonstrated that solar-light-driven photoelectrochemical (PEC) water splitting of semiconductor photoelectrodes is considered a promising technique for the efficient production of H2, because PEC water splitting can directly split water into H2 at room temperature, and the solar energy existing widely in the world is renewable clean energy.3–5 To date, various semiconductors have been studied extensively in the search for desirable water splitting photoelectrode materials.

As an n-type semiconducting metal oxide, TiO2 possesses several merits such as abundant availability, superior physical and chemical properties, excellent photocorrosion resistance and nontoxicity.6–8 Most importantly, TiO2 possesses appropriate band positions to reduce and oxidize water into H2 and oxygen (O2).6–8 Therefore, many TiO2 nanostructures with various morphologies have been intensively used as photoanode materials for PEC water splitting.9,10 Among these nanostructures, the TiO2 nanowire array has been considered an ideal photoanode material because its large surface area can facilitate the transport and separation of charge carriers favorably. In addition, the large surface area of the TiO2 nanowire array can provide more reactive sites for surface reaction.9–14 However, as a wide band gap semiconductor (3.2 eV), TiO2 can only absorb UV light, resulting in the low utilization of solar energy.15–23 In addition, the low carrier separation efficiency and short charge carrier lifetime of TiO2 seriously limit its application in PEC water splitting for H2 generation.9 Therefore, it is still a great challenge to improve the visible-light absorbing ability, separate the charge carriers effectively and prolong the charge carrier lifetime of TiO2 photoanode materials.

To date, many methods have been used to solve the above mentioned problems of TiO2 photoelectrode materials.6,9–14 As studied previously, an effective strategy is to sensitize TiO2 photoelectrode materials with the nanostructures of narrow-bandgap semiconductors.24–30 Particularly, narrow-bandgap semiconductors that can establish a type-II band alignment with TiO2 are the most ideal sensitization materials.24–30 In this TiO2-based heterojunction, not only the visible-light harvesting ability can be effectively enhanced, but also the charge transfer and separation can be greatly facilitated. Indeed, many TiO2-based heterojunction photoelectrodes sensitized with narrow-bandgap semiconductors exhibited a remarkable enhancement in PEC water splitting activity.22–28

As reported in previous studies, CdS is one of narrow-bandgap semiconductors and its band gap is about 2.4 eV.30,31 In addition, CdS possesses suitable band positions to split water into H2 and O2.32–34 The narrow-bandgap and suitable band structure enable CdS to strongly harvest visible-light and exhibit superior PEC water splitting performance. However, the CdS nanostructures easily suffer from photocorrosion during the PEC reaction process. So many methods have been employed to inhibit the photocorrosion of CdS nanostructures. One effective approach is to separate the photogenerated charge carriers of CdS via coupling it with other photocatalytic materials.35–38 In this study, when CdS is used to sensitize TiO2, a type-II band alignment will be established between the two constituents.24,25,30 This band alignment can effectively move the photoproduced electrons from CdS to TiO2 and the holes from TiO2 to CdS, realizing the effective separation of photogenerated electron–hole pairs. Therefore, CdS is a very suitable candidate that can be used to sensitize TiO2 to achieve outstanding PEC water splitting performance. On the other hand, Cu2O, a p-type semiconducting metal oxide, possesses superior ability to harvest visible-light due to its small band gap (2.0 eV).23,26,31,39 Therefore, Cu2O is also usually used to sensitize other photoelectrode materials to improve the PEC water splitting activity.26,29,31,40 Particularly, when Cu2O is used to sensitize CdS, not only a type-II band alignment is built between the two semiconductors, but also a p–n junction is established at the interface of Cu2O and CdS. This band structure is highly advantageous for the transport and separation of charge carriers between Cu2O and CdS, leading to remarkably enhanced PEC performance.31 Thus, when Cu2O is used to sensitize TiO2/CdS to construct the TiO2/CdS/Cu2O heterojunction structure, both the type-II band alignment with a stair-like structure and the p–n junction are integrated into the constructed heterojunction structure. Therefore, the photoanode constructed by sensitizing TiO2/CdS with Cu2O is highly expected to achieve a great enhancement in PEC H2 generation.

In this study, a Cu2O-nanoparticle-sensitized TiO2/CdS (TiO2/CdS/Cu2O) nanowire array photoanode has been successfully fabricated via rational design and subtly combining the successive ionic layer adsorption and reaction (SILAR) with a chemical bath deposition (CBD) method, as illustrated in Fig. 1. Moreover, not only the stair-like type-II band alignment but also the p–n junction are integrated into the constructed novel ternary photoanode. The fabricated photoanode exhibits significantly enhanced PEC water splitting H2 production and apparently prolonged charge carrier lifetime. The measurements of light absorption, photoluminescence (PL) and electrochemical spectra undoubtedly demonstrate that the enhanced PEC H2 generation is contributed by the remarkable enhancement of visible-light harvesting ability and the efficient space charge separation as well as the substantial prolongation of charge carrier lifetime.


image file: d0dt01643h-f1.tif
Fig. 1 Fabrication process of the TiO2/CdS/Cu2O heterojunction NWAs.

2. Experimental section

2.1. Preparation of TiO2 nanowire arrays (NWAs)

A modified hydrothermal method was employed to prepare TiO2 NWAs.12 Firstly, the purchased fluorine-doped tin oxide (FTO) glasses were cleaned by ultrasonic cleaning with deionized water, ethanol (EtOH), isopropanol, acetone, EtOH and deionized water for 25 min, respectively, followed by drying in a constant-temperature drying box at 60 °C for 30 min. Two pieces of the cleaned FTO glass with the conductive surface downward were obliquely placed in the inner tank of a 50 mL Teflon-liner autoclave. Secondly, 15 mL of concentrated hydrochloric acid (36%–38%) and 15 mL of deionized water were mixed, and the mixed solution was stirred vigorously for 5 min, followed by dropping 530 μL of titanium butoxide into the solution and stirring for 30 min vigorously. Thirdly, the freshly prepared solution was transferred into the tank of the Teflon-liner autoclave with the FTO glasses, followed by heating at 150 °C for 3.5 h. After that, the samples were cooled down to room temperature (RT), followed by washing with deionized water, and annealing at 450 °C for 30 min in a tubular furnace.

2.2. Growth of the CdS nanoparticle shell

To manipulate the PEC H2 production of the TiO2/CdS/Cu2O nanowire array photoanode via controlling the CdS nanoparticle shell, the CdS nanoparticle shell was grown on the surface of TiO2 NWAs through an improved SILAR process.24 Briefly, the FTO glass grown with TiO2 nanowires was soaked in 0.02 M Cd(NO3)2 EtOH solution for 1 min and washed with EtOH, followed by soaking in 0.02 M Na2S solution (1[thin space (1/6-em)]:[thin space (1/6-em)]1 volume ratio of methanol (MeOH) and water) for 1 min. After the reaction, the sample was washed with MeOH and then dried at 80 °C for 10 min. The above adsorption and reaction process was called one SILAR cycle. The samples prepared with 4, 8, 12 and 16 SILAR cycles were referenced as TiO2/4CdS, TiO2/8CdS, TiO2/12CdS, and TiO2/16CdS NWAs, respectively. The sample which exhibited the best PEC water splitting activity was further used to construct the TiO2/CdS/Cu2O heterojunction structure.

2.3. Growth of Cu2O nanoparticles (NPs) on TiO2/CdS NWAs

A modified CBD method was employed to grow Cu2O NPs on TiO2/CdS NWAs.3,6 Firstly, a mixed solution was prepared by mixing 0.02 M Cu2SO4 and Na2S2O3 solutions at a volume ratio of 1[thin space (1/6-em)]:[thin space (1/6-em)]4. Secondly, the TiO2 NWAs grown with CdS NPs were soaked in the mixed solution for 2 seconds, and then rinsed with deionized water, followed by soaking in 0.5 M NaOH aqueous solution for 2 seconds at 60 °C. The above CBD process for the growth of Cu2O NPs was repeated two times. The above growth process of Cu2O NPs was called 2 CBD cycles. Finally, the prepared sample was washed with deionized water five times and dried in air. The fabricated sample was referenced as TiO2/CdS/2Cu2O NWAs.

2.4. Characterization

The crystal structure and composition of the samples were characterized by using an X-ray diffractometer (XRD, Japan) employing Cu Ka radiation (λ = 1.5406 Å) at 40 kV and 40 mA. The morphological features and size were evaluated by scanning electron microscopy (SEM, S-4800 HITACHI, Japan). X-ray photoelectron spectroscopy (XPS, ESCALAR 250XI, USA) measurements were performed to determine and analyze the valence states of elements in the sample. The absorption spectra were recorded by using a UV-vis spectrophotometer (Lambda 950 PerkinElmer, USA). RT photoluminescence (PL) was evaluated by using a fluorescence FLS920 spectrometer (Edinburgh) under excitation with 360 nm light.

2.5. PEC and H2 production measurements

The PEC tests were conducted by using a three-electrode system (CHI760E). In this system, the prepared sample was used as the working electrode; a saturated calomel electrode (Ag/AgCl) and Pt wire were employed as the reference and counter electrode, respectively. A mixed solution of 0.35 M Na2S and 0.25 M Na2SO3 was used as an electrolyte. The pH value of the electrolyte is 13. Before the PEC tests, the air in the solution was removed by injecting N2 into the electrolyte for 20 min. A 300 W Xe lamp (AM 1.5 G) was used as the light source during the PEC tests. The intensity of the light illuminated on the sample was controlled at 100 mW cm−2. The linear sweep voltammograms (LSVs) were obtained by using a CHI760E electrochemical workstation. The electrochemical impedance spectra (EIS) were recorded by using the same workstation at the open circuit potential over the frequency ranging from 10−1 to 105 Hz. The incident photon to current efficiency (IPCE) was measured by using a QE/IPCE system (Newport, USA) with a monochromator. The area of excitation light irradiated on the sample is 1 cm2. The data were collected at a wavelength interval of 10 nm. The experiments for H2 evaluation were conducted with the same electrochemical workstation using the electrolyte of 0.35 M Na2S and 0.25 M Na2SO3 without applying any external bias referenced to the Ag/AgCl electrode. An improved displacement method was used to collect H2 in the experiments.

3. Results and discussion

The XRD plots of the TiO2 NWAs sensitized with different loadings of CdS NPs and pure TiO2 NWAs are presented in Fig. 2a. As shown in Fig. 2a, the TiO2 NWAs exhibit obvious characteristic peaks at 36.1°, 41.2°, 62.8° and 69.8°, respectively. These characteristic diffraction peaks of the pure TiO2 NWAs are similar to those observed in the previous literature,30 and are attributed to the (101), (111), (002) and (112) planes of rutile phase TiO2 (JCPDS NO. 76-0318), respectively. The XRD plots demonstrate that no obvious characteristic diffraction peaks of CdS NPs are observed for the TiO2 NWAs with different loadings of CdS NPs. Possible reasons for this phenomenon are that the loaded CdS NPs are small and the CdS NPs are highly dispersed onto the surface of the TiO2 NWAs. However, with an increase of the CdS nanoparticle loading upon increasing the SILAR cycle times, the intensities of the characteristic diffraction peaks of the TiO2 NWAs decreased obviously, as shown in Fig. 2a. Thus the results indicate that the surface of the TiO2 NWAs is covered with the CdS NPs.
image file: d0dt01643h-f2.tif
Fig. 2 (a) XRD plots of the TiO2 NWAs and TiO2/CdS NWAs prepared with different SILAR cycles. XPS spectra of the TiO2/12CdS NWAs: (b) survey scan, (c) Ti 2p, (d) Cd 3d, (e) S 2p and (f) O 1s.

The surface valence states of elements in the TiO2/CdS NWAs prepared with 12 SILAR cycles (TiO2/12CdS) were investigated by XPS, as presented in Fig. 2b–f. As shown in Fig. 2b, the elements of Ti, Cd, S and O are obviously observed in the survey XPS spectrum, showing that the TiO2/CdS NWAs are successfully prepared via a simplified SILAR process. In the high-resolution spectrum of Ti 2p (Fig. 2c), two peaks at 458.2 and 464.3 eV correspond to Ti 2p3/2 and Ti 2p1/2,12,16,27 respectively. In the spectrum of Cd 3d (Fig. 2d), two peaks located at 405.1 and 411.8 eV can be assigned to Cd 3d3/2 and Cd 3d5/2,21,31 respectively. The distance between the two characteristic peaks is 6.7 eV, further confirming the existence of Cd2+ in the sample. In the high-resolution spectrum of S 2p (Fig. 2e), two characteristic peaks at 161.4 and 162.6 eV correspond to S 2p3/2 and S 2p1/2,20,31 respectively. The spectrum of O 1s can be divided into two characteristic peaks at 529.7 and 531.4 eV, as shown in Fig. 2f. The peak at 529.7 eV is attributed to Ti–O of TiO2,12,15,41,42 and the peak located at 531.4 eV is ascribed to Ti–OH of the sample.27,41,42

The SEM images show the morphological features of the prepared samples. Fig. 3a and b display the images of the pure TiO2 sample at low and high magnifications, respectively. It can be found that the sample shows the nanowire array structure. The tips of the nanowires show the morphological feature of a quadrilateral prism. The diameter of the TiO2 nanowires is about 110 nm. The surface of the TiO2 nanowires is smooth, as shown in the high magnification SEM images in Fig. 3b. Fig. 3c–f show the SEM images of the samples prepared by growing CdS NPs on TiO2 NWAs via 4, 8, 12 and 16 SILAR cycles, respectively. Also for the sample prepared via 4 SILAR cycles, the surface of TiO2 is almost smooth as displayed in Fig. 3c. The inset of Fig. 3c shows the enlarged SEM image of the sample, indicating that there are a few CdS NPs on the surface of the TiO2 nanowire. However, with an increase of SILAR cycle times, the surface of the TiO2 nanowire becomes coarse as shown in Fig. 3d and e. The insets of Fig. 3d and e show the enlarged SEM images of the corresponding samples, respectively, clearly demonstrating that the surface of the TiO2 nanowire is covered by a layer of uniform and dense small particles. Interestingly, the SEM images also show that the CdS NPs are uniformly distributed onto the whole surface of the TiO2 nanowire. When the SILAR cycle is further increased to 16 times, the surface of the TiO2 nanowires becomes rather rough, as shown in Fig. 3f. The enlarged SEM image shows that many CdS NPs grow on the surface of the TiO2 nanowire. The SEM images demonstrate that the CdS NPs can be successfully deposited onto the surface of the TiO2 nanowires via a facile SILAR process.


image file: d0dt01643h-f3.tif
Fig. 3 SEM images at different magnifications: (a) low and (b) high magnification images of pure TiO2 NWAs. SEM images of the (c) TiO2/4CdS, (d) TiO2/8CdS, (e) TiO2/12CdS and (f) TiO2/16CdS NWAs. The insets show the enlarged images of the corresponding samples.

The light harvesting capability of the samples was studied by UV-vis absorption spectroscopy. As displayed in Fig. 4, the pure TiO2 NWAs can only absorb ultraviolet light (λ < 400 nm). After growing CdS NPs, the TiO2/CdS heterojunction NWAs exhibit obvious light harvesting ability in the visible-light region (from 420 to 575 nm). Moreover, the visible-light absorption of the TiO2/CdS heterojunction NWAs is enhanced apparently with an increase of the CdS nanoparticle loading. The visible-light absorption of the TiO2/CdS heterojunction NWAs results from the ability of the CdS NPs to harvest visible light. In addition, the results show that the light harvesting ability of the TiO2/CdS heterojunction NWAs can be easily manipulated by adjusting the SILAR cycle times to control the loadings of the CdS NPs.


image file: d0dt01643h-f4.tif
Fig. 4 UV-vis light absorption spectra of the TiO2 NWAs and TiO2/CdS NWAs prepared with different SILAR cycles.

Fig. 5a shows the current density-time (Jt) plots of the samples under simulated solar light irradiation without applying any external bias vs. reference electrode. The electrolyte was a mixed solution of Na2S (0.35 M) and Na2SO3 (0.25 M), and the reference electrode was a saturated Ag/AgCl electrode. As presented in Fig. 5a, all samples show very rapid light response and low dark current. The nearly vertical rising and falling of the photocurrent densities mean that the charge carriers in the electrodes transport quickly. Apparently, all TiO2/CdS heterojunction nanowire array photoanodes exhibit enhanced PEC performance, compared with the pure TiO2 nanowire array photoanode. The pristine TiO2 nanowire array photoanode produces a low photocurrent density of 0.58 mA cm−2. After growing CdS NPs on the surface of the TiO2 NWAs, the photocurrent density of the TiO2/CdS heterojunction nanowire array photoanodes increases significantly with a gradual increase of the CdS nanoparticle loading. The experimental results show that the heterogeneous TiO2/12CdS nanowire array photoanode fabricated via 12 SILAR cycles obtains the highest photocurrent density. The photocurrent density of this heterojunction photoanode reaches up to 5.3 mA cm−2, 9 times that of the pristine TiO2 nanowire array photoanode.


image file: d0dt01643h-f5.tif
Fig. 5 (a) Photocurrent density–time (Jt) plots of the TiO2 NWAs and TiO2/CdS NWAs prepared with different SILAR cycles under simulated solar light irradiation with no bias vs. Ag/AgCl. (b) LSV plots of the samples.

Fig. 5b shows the LSV curves of the samples. The results also demonstrate that all TiO2/CdS heterojunction nanowire array photoanodes exhibit greatly enhanced photocurrent densities with an increase of bias, compared with the pure TiO2 nanowire array photoanode. In addition, the as-constructed TiO2/12CdS heterojunction nanowire array photoanode shows the best PEC water splitting activity among all TiO2/CdS heterojunction nanowire array photoanodes. Thus, in the following studies, the TiO2/12CdS heterojunction NWAs are further used to construct a ternary heterojunction photoanode by sensitizing Cu2O NPs for the evaluation of H2 generation via PEC water splitting.

Fig. 6a displays the XRD plots of the pure TiO2, TiO2/12CdS and TiO2/12CdS/2Cu2O NWAs. As discussed above (Fig. 2a), the pure TiO2 NWAs are composed of rutile phase TiO2. The diffraction peaks of CdS are not observed obviously in the XRD pattern due to its low loading amount. Similarly, the peaks of Cu2O NPs are not observed apparently in the XRD pattern of the TiO2/12CdS/2Cu2O samples. The possible reason is that the loading amount of the Cu2O NPs is low in the sample.


image file: d0dt01643h-f6.tif
Fig. 6 (a) XRD plots of the TiO2, TiO2/12CdS and TiO2/12CdS/2Cu2O NWAs. XPS spectra of the TiO2/12CdS/2Cu2O NWAs: (b) survey scan, (c) Cu 2p, and (d) O 1s.

The surface valence states of elements in the TiO2/12CdS/2Cu2O NWAs were investigated by XPS. As shown in the XPS spectrum in Fig. 6b, the elements of Ti, Cd, S, Cu and O are obviously observed, showing that Cu2O is successfully grown on the TiO2/12CdS NWAs by a simplified CBD method. Fig. 6c displays the XPS spectrum of Cu 2p; two characteristic peaks centered at 932.3 and 952.1 eV are ascribed to Cu 2p3/2 and Cu 2p1/2,43–45 respectively. Moreover, no shoulder peaks are observed in the characteristic peaks, confirming that only Cu+ exists in the sample. The XPS spectrum of O 1s is presented in Fig. 6d. The spectrum can be divided into four characteristic peaks at 529.8, 530.9 531.4 and 532.5 eV, respectively. The peak at 529.8 eV is attributed to Ti–O of TiO2,12,15,41,42 the peak centered at 530.9 eV originates from Cu–O of Cu2O,31,46 the peak centered at 531.4 eV is attributed to Ti-OH of the sample,27,41,42 and the final peak at 532.5 eV results from the oxygen of water absorbed on the sample.46

After growing Cu2O NPs on the TiO2/12CdS NWAs, the morphological features of the prepared samples were evaluated by SEM. Fig. 7 shows the SEM images of the as-constructed TiO2/12CdS and TiO2/12CdS/2Cu2O NWAs. Fig. 7a and b show the SEM images of the TiO2/12CdS NWAs at low and high magnifications, respectively. The SEM images show that many NPs are deposited on the surface of the nanowires. However, after the growth of the Cu2O NPs by an easy CBD method, the size of these NPs becomes larger, as clearly displayed in Fig. 7c and d at low and high magnifications, respectively. Thus, the results of XPS measurements and SEM evaluations reveal that the Cu2O NPs are successfully grown on the surface of the TiO2/12CdS NWAs.


image file: d0dt01643h-f7.tif
Fig. 7 SEM images at (a) low and (b) high magnification of the TiO2/12CdS NWAs, respectively, (c) low and (d) high magnification of the TiO2/12CdS/2Cu2O NWAs, respectively.

The light absorption of the as-fabricated TiO2/12CdS/2Cu2O NWAs was investigated by UV-vis absorption spectroscopy, as shown in Fig. 8. For comparison, the light absorption spectrum of the prepared TiO2/12CdS NWAs is also displayed in Fig. 8. The light absorption measurements show that the visible-light absorption of the as-constructed TiO2/12CdS/Cu2O NWAs is further enhanced, as compared with that of the TiO2/12CdS NWAs. Moreover, the TiO2/12CdS/2Cu2O NWAs exhibit a widen visible-light absorption ranging from 420 to 650 nm, which is contributed by the superior visible-light harvesting capability of the Cu2O NPs because of the narrow band gap. Therefore, the sensitization of the Cu2O NPs not only further enhances the visible-light harvesting but also extends the visible-light harvesting range of the TiO2/12CdS NWAs.


image file: d0dt01643h-f8.tif
Fig. 8 UV-vis light absorption spectra of the TiO2, TiO2/12CdS and TiO2/12CdS/2Cu2O NWAs.

The PEC water splitting property was evaluated by using a three-electrode electrochemical testing system. To evaluate the effect of the Cu2O nanoparticle loading on the PEC water splitting capability of the TiO2/12CdS NWAs, the PEC water splitting activities of the samples prepared via 1 (TiO2/12CdS/1Cu2O NWAs) and 4 (TiO2/12CdS/4Cu2O NWAs) CBD cycles were also studied. Fig. 9a shows the Jt plots of the TiO2/12CdS NWAs with different Cu2O nanoparticle loadings. For comparison, the Jt plot of the heterogeneous TiO2/12CdS NWAs is also presented in Fig. 9a. Apparently, the TiO2/12CdS/2Cu2O NWAs exhibit enhanced PEC water splitting ability and achieve a photocurrent density of 6.4 mA cm−2, 1.2 times that (5.3 mA cm−2) of TiO2/12CdS at 0 V vs. the reference electrode. In addition, the experimental results demonstrate that the excess loading of Cu2O NPs is not favorable for the transfer and separation of the photogenerated charge carriers in the TiO2/12CdS/4Cu2O NWAs, resulting in a decrease of the photocurrent density. The results show that the loading of Cu2O NPs exhibits an obvious effect on the PEC water splitting performance of the TiO2/12CdS NWAs. In the following studies, we just analyse the PEC water splitting performances of the TiO2/12CdS NWAs with the optimized Cu2O NPs (TiO2/12CdS/2Cu2O). The LSV measurements also demonstrate that the TiO2/12CdS/2Cu2O NWAs exhibit enhanced PEC water splitting ability, as shown in Fig. 9b.


image file: d0dt01643h-f9.tif
Fig. 9 (a) Jt plots of TiO2/12CdS and TiO2/12CdS with different loadings of Cu2O NPs, (b) LSV curves, (c) STH and (d) IPCE of the TiO2, TiO2/12CdS and TiO2/12CdS/2Cu2O NWAs at 0 V under simulated solar light illumination.

The PEC water splitting capability for H2 production of all samples was evaluated via the calculation of solar-to-hydrogen (STH) energy conversion efficiency (η). According to the previous studies, the following equation is used to calculate the STH efficiency of the samples:10,47

 
image file: d0dt01643h-t1.tif(1)

In this equation, Jp and VRHE are the photocurrent density (mA cm−2) corresponding to the measured potential and the potential applied to the reversible hydrogen electrode (RHE), respectively. Plight is the light intensity (100 mW cm−2). The measured electrode potentials vs. reference electrode (Ag/AgCl) are converted to the RHE using the Nernst equation: VRHE = VAg/AgCl + VAg/AgCl0 + 0.059 pH. In this equation, VAg/AgCl0 is the standard electrode potential of Ag/AgCl and its value is 0.197 V at 25 °C. The pH value of the electrolyte (0.35 M Na2S and 0.25 M Na2SO3) is 13. As presented in Fig. 9c, the maximum efficiency of the pure TiO2 nanowire array electrode is 0.3% at 0.55 V vs. RHE. However, the TiO2/12CdS/2Cu2O nanowire array electrode achieves the highest efficiency of 4.3% at 0.30 V vs. RHE, while the highest efficiency of the TiO2/12CdS nanowire array electrode is 3.9% at 0.22 V vs. RHE. Even at the same potential (0.22 V vs. RHE), the efficiency of the TiO2/12CdS/2Cu2O electrode nanowire array electrode is as high as 4.1%, approximately 1.1 times that of the TiO2/12 CdS nanowire array electrode.

As proved in previous studies, the IPCE is another important property used to evaluate the PEC water splitting capability of a photoanode. According to the previous studies, the IPCE of a photoanode can be calculated using the following equation:25,47

 
image file: d0dt01643h-t2.tif(2)
Where, Jph is the photocurrent density (mA cm−2), λ is the wavelength of monochromatic light, and P is the power of incident light. Fig. 9d shows the IPCE plots of the pure TiO2, TiO2/12CdS and TiO2/12CdS/2Cu2O NWAs, demonstrating that the deposition of CdS and Cu2O NPs significantly extends the light response region of the TiO2 NWAs and optimizes the conversion efficiency of light energy. In addition, the IPCE plots clearly demonstrate that the TiO2/12CdS and TiO2/12CdS/2Cu2O NWAs exhibit a broadened visible-light response. The light response range of the pure TiO2 nanowire array photoelectrode is from 320 to 400 nm and its maximum IPCE value is about 26.7% at 350 nm. The light response of the TiO2/12CdS nanowire array photoelectrode extends to 550 nm, and its IPCE value is as high as 29.7% at 370 nm. When the Cu2O NPs are deposited on the surface of the TiO2/12CdS NWAs, the photoresponse of the TiO2/12CdS/2Cu2O nanowire array photoelectrode is further extended to 610 nm. At 390 nm, the maximum IPCE value of the TiO2/12CdS/2Cu2O nanowire array photoelectrode is 32.3%. Moreover, even at 420 nm, the TiO2/12CdS/2Cu2O nanowire array photoelectrode obtains an IPCE value of 27.5%. The improvement of the IPCE shows that the sensitization of the Cu2O NPs not only improves the visible-light harvesting capability but also promotes the efficient transport and separation of the photogenerated charge carriers in the TiO2/12CdS/2Cu2O heterojunction NWAs.

The PEC water splitting for H2 production was evaluated under simulated solar light irradiation with the same electrolyte used in the above PEC performance analysis without applying any external bias referenced to the Ag/AgCl electrode. The area of the electrode used to evaluate the H2 generation is 3 cm2 (1.5 × 2 cm2). Fig. 10a shows the amount of H2 recorded every 15 min over a period of 2 h. During the measured time of 2 h, the amount of H2 generated by the TiO2/12CdS/2Cu2O nanowire array electrode reaches 428.6 μmol, which is approximately 1.8 times that (241.7 μmol) of the TiO2/12CdS and 8.2 times that (52.1 μmol) of pure TiO2 nanowire array electrodes, respectively. In addition, a nearly linear increase of the H2 amount shows that all electrodes display good stability during the water splitting reaction. What is more, the overall H2 production rate is stable. The average H2 production rate is presented in Fig. 10b. The results show that the average production rates of the pure TiO2 and TiO2/12CdS nanowire array electrodes are 26.0 and 120.8 μmol h−1, respectively. However, the average H2 production rate of the TiO2/12CdS/2Cu2O nanowire array electrode reaches up to 214.3 μmol h−1, 8.2 and 1.8 times those of the pristine TiO2 and heterogeneous TiO2/12CdS nanowire array electrodes, respectively.


image file: d0dt01643h-f10.tif
Fig. 10 (a) H2 volume vs. time graph and (b) average rate of the TiO2, TiO2/12CdS and TiO2/12CdS/2Cu2O nanowire array photoanodes at 0 V under simulated solar light illumination.

As proved in previous studies, the transport and separation of photoinduced electrons and holes in a photoelectrode greatly determines its PEC property. To investigate the transfer and separation property, the EIS measurements of the samples were conducted under AM 1.5 G light irradiation. Fig. 11a shows the EIS plots of the samples. Generally, it is believed that the smaller semicircle radius in the EIS Nyquist curve indicates the effective separation of photogenerated electrons and holes in a photoelectrode.31,40 As displayed in Fig. 11a, the semicircle radius of the TiO2/12CdS/2Cu2O nanowire array electrode is the smallest among all electrodes, demonstrating that the photoexcited charge carriers in the TiO2/12CdS/2Cu2O nanowire array electrode are transferred and separated effectively. The efficient separation of the photogenerated charge carriers in the TiO2/12CdS/2Cu2O nanowire array electrode is well supported by RT PL studies. As presented in Fig. 11b, the pure TiO2 NWAs exhibit a wide RT PL emission from 390 to 550 nm. The strong emission centered at 410 nm may originate from the intrinsic emission of the rutile phase TiO2 NWAs. The visible-light PL emission is induced by the defects of the TiO2 NWAs. A similar PL emission feature was also reported in the previous studies.41,48,49 However, after the sensitization with CdS and Cu2O NPs, the RT PL emission intensities of the TiO2 NWAs decrease significantly. Particularly, the RT PL emission intensity of the TiO2/12CdS/2Cu2O NWAs decreases drastically, indicating that the recombination of the photoexcited electro-hole pairs is suppressed greatly.


image file: d0dt01643h-f11.tif
Fig. 11 (a) EIS plots and (b) PL spectra of the TiO2, TiO2/12CdS and TiO2/12CdS/2Cu2O NWAs. (c) Transient open-circuit voltage decay (OCVD) of the TiO2 TiO2/12CdS and TiO2/12CdS/2Cu2O nanowire array photoanodes upon exposure to UV-vis light excitation. (d) Average electron lifetimes (τn) within the photoanodes.

Aside from the effective separation of the charge carriers, the PEC water splitting capability of a photoelectrode is also greatly determined by its charge carrier lifetime. The previous studies have demonstrated that the charge carrier lifetime of a photoelectrode can be evaluated via measuring the transient open-circuit voltage decay (OCVD).50,51 Fig. 11c displays the OCVD plots of the TiO2, TiO2/12CdS and TiO2/12CdS/2Cu2O NWAs. Based on the OCVD measurements, the average charge carrier lifetime (τn) of the photoelectrode can be determined according to the following equation:50,51

 
image file: d0dt01643h-t3.tif(3)

In this equation, the definitions of variables are given in the previous studies.44,45 As displayed in Fig. 11d, the average charge carrier lifetime within the TiO2/12CdS/2Cu2O nanowire array photoelectrode is substantially prolonged, compared to those of the bare TiO2 and heterogeneous TiO2/12CdS nanowire array photoelectrodes. The prolonged charge carrier lifetime within the TiO2/12CdS/2Cu2O nanowire array photoelectrode is in good agreement with the experimental data of the RT PL emission (Fig. 11b). The experimental measurements apparently show that the sensitization of the Cu2O NPs not only greatly enhances the visible-light harvesting ability but also substantially prolongs the charge carrier lifetime of the TiO2/12CdS/2Cu2O nanowire array photoelectrode, enabling the sensitized heterojunction photoelectrode to exhibit a remarkable enhancement in H2 production.

Compared with the TiO2/12CdS NWAs, the Cu2O nanoparticle-sensitized TiO2/12CdS NWAs exhibit greatly enhanced PEC water splitting capability. Thus it is highly desirable to gain an insight into the enhanced mechanism of the PEC water splitting activity of the TiO2/12CdS/Cu2O NWAs. As demonstrated in the previous studies,52,53 the CB position (−0.7 eV) and VB position (1.7 eV) of CdS are more negative than those of TiO2, because the CB and VB energies of TiO2 are −0.2 and 3.0 eV,23 respectively. In addition, as shown in our previous work31 and other published reports,39,46 the CB position (−1.5 eV) and VB position (0.5 eV) of Cu2O are also more negative than those of CdS. Therefore, a type-II band alignment with a stair-like structure is established within the TiO2/CdS/Cu2O heterojunction structure. On the other hand, a p–n junction is built at the interface of Cu2O and CdS, because Cu2O and CdS are p-type and n-type semiconductors, respectively. In the TiO2/CdS/Cu2O heterojunction structure, TiO2, CdS and Cu2O can be excited with simulated solar light to generate electrons and holes. The type-II band alignment with a stair-like structure not only enables the photoexcited electrons on the CB of CdS to move to that of TiO2, but also enables the photoproduced electrons on the CB of Cu2O to move to that of TiO2 via the CB of CdS. Finally, the photogenerated electrons of Cu2O, CdS and TiO2 arrive at the Pt electrode to produce H2 via taking part in the water reduction reaction. Meanwhile, the type-II band alignment with a stair-like structure enables the photoproduced holes on the VB of TiO2 and CdS to move to that of Cu2O. The photoproduced holes existing together on the VB of Cu2O are consumed by sacrificial agents (S2− and SO32−) in the electrolyte. Moreover, the inner electric field built in the p–n junction can significantly facilitate the transport and separation of the photogenerated charge carriers between Cu2O and CdS. Therefore, the remarkably enhanced PEC water splitting capability of the TiO2/CdS/Cu2O heterojunction structure is attributed to the following contributions: (1) a remarkable enhancement of the visible-light harvesting ability achieved by narrow-bandgap Cu2O and CdS and (2) efficient space charge transport and separation achieved by the synergetic effects of the type-II band alignment with a stair-like structure and the p–n junction. The charge transfer process and enhanced mechanism for the PEC H2 generation of the TiO2/CdS/Cu2O heterojunction are illustrated in Fig. 12.


image file: d0dt01643h-f12.tif
Fig. 12 The charge transfer process and enhanced mechanism for the PEC H2 generation of the TiO2/12CdS/2Cu2O heterojunction.

4. Conclusions

A novel ternary photoanode constructed by sensitizing TiO2/CdS core/shell NWAs with Cu2O NPs was designed and fabricated by combining the facile successive ionic layer adsorption and reaction with a chemical bath deposition process. This novel photoanode not only possesses a type-II band alignment with a stair-like structure but also possesses a p–n junction. The type-II band alignment with a stair-like structure and the p–n junction enable the photoanode to separate the charge carriers effectively and to prolong the charge carrier lifetime substantially. After optimizing the loading of Cu2O NPs, the PEC measurements demonstrate that the TiO2/CdS/2Cu2O core/shell heterojunction nanowire array photoanode displays remarkably improved PEC water splitting capability for H2 production. The maximum solar-to-hydrogen efficiency is as high as 4.3%. The average H2 production rate of the TiO2/12CdS/2Cu2O nanowire array photoanode is 214.3 μmol h−1, 8.2 and 1.8 times those of the pristine TiO2 and heterogeneous TiO2/CdS nanowire array photoanodes, respectively. The experimental results of light absorption, photoluminescence and electrochemical spectra show that the improved PEC H2 generation is ascribed to the significant enhancement of the visible-light harvesting ability, effective space charge separation and substantial prolongation of the charge carrier lifetime, which are achieved by the synergetic effects of the type-II band alignment with a stair-like structure and a p–n junction. The outstanding PEC H2 production demonstrates that the TiO2/CdS/Cu2O nanowire arrays have the potential as a photoanode for the efficient conversion of solar energy into chemical fuels.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China under Grant No. 61575225, 11074312, 11374377, 11474174, and 11404414, the Fujian Key Laboratory of Functional Marine Sensing Materials, Minjiang University under Grant No. MJUKF-FMSM201912, and the Undergraduate Research Training Program of Minzu University of China under Grant No. URTP2018110045, URTP2018110046 and URTP2018110028.

References

  1. X. Li, J. G. Yu, J. X. Low, Y. P. Fang, J. Xiao and X. B. Chen, J. Mater. Chem. A, 2015, 3, 2485–2534 RSC.
  2. R. Van de Krol, Y. Liang and J. Schoonman, J. Mater. Chem., 2008, 18, 2311–2320 RSC.
  3. M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A. Santori and N. S. Lewis, Chem. Rev., 2010, 110, 6446–6473 CrossRef CAS PubMed.
  4. D. Kang, T. W. Kim, S. R. Kubota, A. C. Cardiel, H. G. Cha and K.-S. Choi, Chem. Rev., 2015, 115, 12839–12887 CrossRef CAS PubMed.
  5. S. C. Wang, G. Liu and L. Z. Wang, Chem. Rev., 2019, 119, 5192–5247 CrossRef CAS PubMed.
  6. X. B. Chen and S. S. Mao, Chem. Rev., 2007, 107, 2891–2959 CrossRef CAS PubMed.
  7. M. D. Hernández-Alonso, F. Fresno, S. Suárez and J. M. Coronado, Energy Environ. Sci., 2009, 2, 1231–1257 RSC.
  8. M. Ni, M. K. H. Leung, D. Y. C. Leung and K. Sumathy, Renewable Sustainable Energy Rev., 2007, 11, 401–425 CrossRef CAS.
  9. T. Wu, C. Chen, Y. Wei, R. Lu, L. Wang and X. Jiang, Dalton Trans., 2019, 48, 12096–12104 RSC.
  10. G. M. Wang, H. Y. Wang, Y. C. Ling, Y. C. Tang, X. Y. Yang, R. C. Fitzmorris, C. C. Wang, J. Z. Zhang and Y. Li, Nano Lett., 2011, 11, 3026–3033 CrossRef CAS PubMed.
  11. Z. Li, L. Shi, D. Franklin, S. Koula, A. Kushima and Y. Yang, Nano Energy, 2018, 51, 400–407 CrossRef CAS.
  12. L. Z. Yao, W. Z. Wang, L. J. Wang, Y. J. Liang, J. L. Fu and H. L. Shi, Int. J. Hydrogen Energy, 2018, 43, 15907–15917 CrossRef CAS.
  13. W. Y. Yuan, J. Yuan, J. L. Xie and C. M. Li, ACS Appl. Mater. Interfaces, 2016, 8(9), 6082–6092 CrossRef CAS PubMed.
  14. J. Yuan, C. M. Li, T. H. Li, M. X. Jing, W. Y. Yuan and C. M. Li, Sol. Energy Mater. Sol. Cells, 2020, 207, 110349 CrossRef CAS.
  15. J. Wang, D. N. Tafen, J. P. Lewis, Z. L. Hong, A. Manivannan, M. J. Zhi, M. Li and N. Q. Wu, J. Am. Chem. Soc., 2009, 131, 12290–12297 CrossRef CAS PubMed.
  16. K. Lalitha, G. Sadanandam, V. D. Kumari, M. Subrahmanyam, B. Sreedhar and N. Y. Hebalkar, J. Phys. Chem. C, 2010, 114, 22181–22189 CrossRef CAS.
  17. Q. J. Xiang, J. G. Yu and M. Jaroniec, J. Am. Chem. Soc., 2012, 134, 6575–6578 CrossRef CAS PubMed.
  18. J. Jia, P. Xue, X. Y. Hu, Y. S. Wang, E. Z. Liu and J. Fan, J. Catal., 2019, 375, 81–94 CrossRef CAS.
  19. M. G. Wang, H. Zhang, H. L. Zu, Z. R. Zhang and J. Han, Appl. Surf. Sci., 2018, 455, 729–735 CrossRef CAS.
  20. J. M. Wang, Z. J. Wang, P. Qu, Q. H. Xu, J. F. Zheng, S. P. Jia, J. Z. Chen and Z. P. Zhu, Int. J. Hydrogen Energy, 2018, 43, 7388–7396 CrossRef CAS.
  21. F. X. Xiao, J. W. Miao, H. Y. Wang and B. Liu, J. Mater. Chem. A, 2013, 1, 12229–12238 RSC.
  22. Y. Lu, X. Cheng, G. Tian, H. Zhao, L. He, J. Hu, S. M. Wu, Y. Dong, G. G. Chang, S. Lenaerts, S. Siffert, G. V. Tendeloo, Z. F. Li, L. L. Xu, X. Y. Yang and B. L. Su, Nano Energy, 2018, 47, 8–17 CrossRef CAS.
  23. H. M. Yuan, J. L. Liu, J. Li, Y. P. Li, X. P. Wang, Y. Q. Zhang, J. B. Jiang, S. Y. Chen, C. Zhao and D. Qian, J. Colloid Interface Sci., 2015, 444, 58–66 CrossRef CAS PubMed.
  24. W. T. Sun, Y. Yu, H. Y. Pan, X. F. Gao, Q. Chen and L. M. Peng, J. Am. Chem. Soc., 2008, 130, 1124–1125 CrossRef CAS PubMed.
  25. D. R. Baker and P. V. Kamat, Adv. Funct. Mater., 2009, 19, 805–811 CrossRef CAS.
  26. Z. Zolfaghari-Isavandi and Z. Shariatinia, J. Alloys Compd., 2018, 737, 99–112 CrossRef CAS.
  27. C. C. Hao, W. Z. Wang, R. Zhang, B. Zou and H. L. Shi, Sol. Energy Mater. Sol. Cells, 2018, 174, 132–139 CrossRef CAS.
  28. J. F. Lin, Y. Liu, Y. P. Liu, C. Huang, W. H. Liu, X. H. Mi, D. Y. Fan, F. T. Fan, H. D. Lu and X. B. Chen, ChemSusChem, 2019, 12, 961–967 CrossRef CAS PubMed.
  29. P. K. Dubey, R. Kumar, R. S. Tiwari, O. N. Srivastava, A. C. Pandey and P. Singh, Int. J. Hydrogen Energy, 2018, 43, 6867–6878 CrossRef CAS.
  30. Y. X. Hu, B. Y. Wang, J. Q. Zhang, T. Wang, R. Liu, J. Zhang, X. N. Wang and H. Wang, Nanoscale Res. Lett., 2013, 8, 222 CrossRef PubMed.
  31. L. J. Wang, W. Z. Wang, Y. L. Chen, L. Z. Yao, X. Zhao, H. L. Shi, M. S. Cao and Y. J. Liang, ACS Appl. Mater. Interfaces, 2018, 10, 11652–11662 CrossRef CAS PubMed.
  32. Q. Li, B. D. Guo, J. G. Yu, J. R. Ran, B. H. Zhang, H. J. Yan and J. R. Gong, J. Am. Chem. Soc., 2011, 133, 10878–10884 CrossRef CAS PubMed.
  33. K. Zhang, W. J. Kim, M. Ma, X. J. Shi and J. H. Park, J. Mater. Chem. A, 2015, 3, 4803–4810 RSC.
  34. C. M. Wolff, P. D. Frischmann, M. Schulze, B. J. Bohn, R. Wein, P. Livadas, M. T. Carlson, F. Jäckel, J. Feldmann, F. Würthner and J. K. Stolarczyk, Nat. Energy, 2018, 3, 862–869 CrossRef CAS.
  35. Y. Chen, J.-F. Li, P.-Y. Liao, Y.-S. Zeng, Z. Wang and Z.-Q. Liu, Chin. Chem. Lett. DOI:10.1016/j.cclet.2019.12.013.
  36. N.-C. Zheng, T. Ouyang, Y. Chen, Z. Wang, D.-Y. Chen and Z.-Q. Liu, Catal. Sci. Technol., 2019, 9, 1357–1364 RSC.
  37. R.-B. Wei, Z.-L. Huang, G.-H. Gu, Z. Wang, L. Zeng, Y. Chen and Z.-Q. Liu, Appl. Catal., B, 2018, 231, 101–107 CrossRef CAS.
  38. R.-B. Wei, P.-Y. Kuang, H. Cheng, Y.-B. Chen, J.-Y. Long, M.-Y. Zhang and Z.-Q. Liu, ACS Sustainable Chem. Eng., 2017, 5, 4249–4257 CrossRef CAS.
  39. Y. C. Pu, H. Y. Chou, W. S. Kuo, K. H. Wei and Y. J. Hsu, Appl. Catal., B, 2017, 204, 21–32 CrossRef CAS.
  40. P. Y. Kuang, X. J. Zheng, J. Lin, X. B. Huang, N. Li, X. Li and Z. Q. Liu, ACS Omega, 2017, 2, 852–863 CrossRef CAS.
  41. J. G. Yu, H. G. Yu, B. Cheng, X. J. Zhao, J. C. Yu and W. K. Ho, J. Phys. Chem. B, 2003, 107, 13871–13879 CrossRef CAS.
  42. J. C. Yu, L. Z. Zhang, Z. Zheng and J. C. Zhao, Chem. Mater., 2003, 15, 2280–2286 CrossRef CAS.
  43. L. Yu, X. Ba, M. Qiu, Y. F. Li, L. Shuai, W. Zhang, Z. F. Ren and Y. Yu, Nano Energy, 2019, 60, 576–582 CrossRef CAS.
  44. T. Ghodselahi, M. A. Vesaghi, A. Shafiekhani, A. Baghizadeh and M. Lameii, Appl. Surf. Sci., 2008, 255, 2730–2734 CrossRef CAS.
  45. D. Sharma, S. Upadhyay, V. R. Satsangi, R. Shrivastav, U. V. Waghmare and S. Dass, J. Phys. Chem. C, 2014, 118, 25320–25329 CrossRef CAS.
  46. Y. G. Zhang, L. L. Ma, J. L. Li and Y. Yu, Environ. Sci. Technol., 2007, 41, 6264–6269 CrossRef CAS PubMed.
  47. S. S. Liu, Z. B. Luo, L. L. Li, H. M. Li, M. X. Chen, T. Wang and J. L. Gong, Nano Energy, 2018, 53, 125–129 CrossRef CAS.
  48. Y. Lei, L. D. Zhang, G. W. Meng, G. H. Li, X. Y. Zhang, C. H. Liang, W. Chen and S. X. Wang, Appl. Phys. Lett., 2001, 78, 1125–1127 CrossRef CAS.
  49. F. Y. Ning, M. F. Shao, S. M. Xu, Y. Fu, R. K. Zhang, M. Wei, D. G. Evans and X. Duan, Energy Environ. Sci., 2016, 9, 2633–2643 RSC.
  50. J. Bisquert, A. Zaban, M. Greenshtein and I. Mora-Seró, J. Am. Chem. Soc., 2004, 126, 13550–13559 CrossRef CAS PubMed.
  51. J. S. DuChene, B. C. Sweeny, A. C. Johnston-Peck, D. Su, E. A. Stach and W. D. Wei, Angew. Chem., Int. Ed., 2014, 53, 7887–7891 CrossRef CAS PubMed.
  52. X. Y. Guo, C. F. Chen, W. Y. Song, X. Wang, W. H. Di and W. P. Qin, J. Mol. Catal. A: Chem., 2014, 387, 1–6 CrossRef CAS.
  53. J. R. Ran, J. G. Yu and M. Jaroniec, Green Chem., 2011, 13, 2708–2713 RSC.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/D0DT01643H

This journal is © The Royal Society of Chemistry 2020