2023, Vol. 34 
b State Key Laboratory of Materials Processing and Die & Mould Technology, School of Materials Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, China;
c School of Materials Science and Engineering, Zhengzhou University, Zhengzhou 450001, China
Recently, g-C3N4 nanosheets (NSs) have attracted much attention to hydrogen production due to excellent chemical stability, suitable band structure, simple syntheses, low cost and special two-dimensional (2D) layered structure [1-3]. Besides, the special nitrogen-rich polymeric structure of 2D layered g-C3N4 NSs could provide numerous active sites for HER [4,5]. Nevertheless, the photocatalytic hydrogen evolution performance of pure 2D layered g-C3N4 NSs is unsatisfactory because of fast electron-holes pairs recombination and insufficient absorption of light [6-8]. Hence, it is essential to exploit novel methods to enhance the H2 generation performance of 2D layered g-C3N4 NSs. The construction of an internal electric field (IEF) is an effective strategy to enhance photocatalytic hydrogen evolution performance because of its key role in photo-induced carrier separation [9-11]. Precious metals act as cocatalysts is a common strategy to construct IEF [12,13]. Yet, the widespread application of precious metals is limited by high cost [14]. Hence, the development of an inexpensive and efficient cocatalyst is crucial for enhancing the photocatalytic activity of 2D layered g-C3N4 NSs.
Among various cocatalysts, transition-metal chalcogenides, such as MoS2, receive widespread attention ascribed to the superior 2D layered structure [15-19]. MoS2 has both semiconductor 2H phase and metallic 1T phase. Among them, metallic 1T phase MoS2 NSs with octahedral coordination can improve the transfer and capture of photogenerated carriers to boost the photocatalytic H2 evolution activity of 2D layered g-C3N4 NSs, attributing to excellent conductivity [20-22]. In addition, the Gibbs free energy of 1T phase MoS2 NSs for H+ absorption is near-zero, which is suitable for HER [6,23]. However, the basal plane of MoS2 NSs is inert, which limits the photocatalytic hydrogen evolution reaction [6]. Recent researches show that MoS2′s basal plane can be activated through doping atoms to design the active sites of MoS2, attributed to the local electronic structure modulation [6,24-26]. Thus, boron (B) ions are incorporated into the lattice of 1T phase MoS2 could activate the basal plane, which can improve photocatalytic H2 production performance. Hence, 2D-layered g-C3N4 NSs modified by doping B into 1T phase MoS2 NSs (B-MoS2 NSs) could exhibit better photocatalytic performance.
In this work, we propose B-MoS2@g-C3N4 composites for photocatalytic H2 evolution, in which triethanolamine acts as the sacrificial agent. The B into 1T phase MoS2 NSs are powerfully connected with 2D layered g-C3N4 NSs through an easy hydrothermal method. The synthesized B-MoS2@g-C3N4 composites with 15 wt% B-MoS2 (B-MoS2@g-C3N4–15) display an efficient rate of hydrogen evolution (1612.75 µmol h−1 g−1), which is 52.33 times as much as pure g-C3N4 NSs (30.82 µmol h−1 g−1). In addition, the apparent quantum efficiency (AQE) of pure g-C3N4 NSs and B-MoS2@g-C3N4–15 composites are 0.41 and 5.54% under the light at λ = 370 nm. The loading of B-MoS2 NSs improves the light absorption of g-C3N4 to stimulate more photogenerated carriers. In addition, the incorporation of B ions into the lattice of 1T phase MoS2 NSs can provide more active sites and speed up the photocatalytic hydrogen evolution reaction. Thus, B-MoS2@g-C3N4 composites display excellent photocatalytic H2 production activity.
The fabrication of B-MoS2@g-C3N4 composites is displayed in Scheme S1 (Supporting information). Firstly, pure g-C3N4 NSs are synthesized via a direct thermal polymerization way of urea. During heating, urea first reacts to form bulk g-C3N4 when the muffle furnace temperature is kept at 550 ℃, and then g-C3N4 NSs with 2D layered structure are formed at the muffle furnace temperature of 500 ℃. Afterward, g-C3N4 NSs, ammonium tetrathiomolybdate and boric acid are added to the 70 mL N, N-dimethylformamide solution and evenly dispersed. Then, B-MoS2 NSs are grown on the surface of g-C3N4 NSs by hydrothermal method to obtain B-MoS2@g-C3N4 composites.
For pure g-C3N4 NSs and B-MoS2@g-C3N4 composites, two characteristic peaks are located at 13.1° and 27.4° (green, pink, blue, yellow and purple curves in Fig. 1a), attributing to (100) and (002) planes of g-C3N4 (JCPDS No. 87–1526) [27-29]. Simultaneously, the peaks at 13.1° (100) and 27.4° (002) of g-C3N4 in B-MoS2@g-C3N4 composites become weaker after B-MoS2 NSs loading on g-C3N4 NSs, attributing to the fact that the order degree of g-C3N4 NSs is decreased by B-MoS2 NSs incorporation [6]. As shown in black curve in Fig. 1a, there are two peaks at 10.2° and 32.5° indexed to (002) and (100) plane of MoS2. Compared with pure MoS2 (black curve in Fig. 1a), the peak at 10.2° of MoS2 shifts to a lower degree of 8.8° in contrast to the 10.2° (002) peak of MoS2 due to the B ions incorporating into the lattice of MoS2 cause the distortion of MoS2 crystal (red curve in Fig. 1a) [6]. In addition, the peak of B-MoS2@g-C3N4 composites at 8.8° is similar to B-MoS2, indicating that B-MoS2 and g-C3N4 coexist.
|
Download:
|
| Fig. 1. (a) X-ray diffraction patterns of pure g-C3N4 NSs and B-MoS2@g-C3N4 composites. (b) C 1s, (c) N 1s, (d) Mo 3d, (e) S 2p and (f) B 1s XPS spectra of B-MoS2@g-C3N4–15 composites. | |
To further analyze the chemical bonding state, the XPS spectra of B-MoS2@g-C3N4 composites are tested. As shown in Fig. S1 (Supporting information), the survey XPS spectrum indicates that B-MoS2@g-C3N4 composites consist of C, N, Mo, S and B. As shown in Fig. 1b, the C 1s XPS spectrum is deconvolved into two peaks at 284.8 and 288.4 eV, corresponding to C=C and C-(N)3 bonds, respectively [30,31]. Fig. 1c indicates that the N 1s spectrum is deconvolved into three peaks at 398.2, 399.1 and 400.6 eV, attributing to C=N—C, N-(C)3 and C—NHx, respectively [32,33]. As shown in Fig. 1d, there are two green peaks at 227.8 and 230.9 eV, assigning to Mo 3d5/2 and Mo 3d3/2 of 1T phase [34]. The two pink peaks at 228.7 and 232.2 eV are attributed to Mo 3d5/2 and Mo 3d3/2 of 2H phase [34]. In addition, the peaks at 225.2 and 234.9 eV are assigned to S 2s and oxidation of Mo [6]. As for the S 2p XPS spectrum (Fig. 1e), two green peaks at 160.7 and 162.1 eV are attributed to S 2p3/2 and S 2p1/2 of 1T phase [35]. Simultaneously, there are two pink peaks at 161.5 and 163.4 eV, assigning to S 2p3/2 and S 2p1/2 of 2H phase [35]. Besides, according to Mo 3d and S 2p spectra, the proportion of 1T phase MoS2 is about 71.2%, indicating that 1T phase MoS2 in B-MoS2@g-C3N4 composites is the main phase. Fig. 1f exhibits a visible peak of B element at 184.6 eV, illustrating the successful doping of B ions in MoS2 [6]. Besides, the Raman spectrum of B-MoS2 (Fig. S2 in Supporting information) is measured to determine the MoS2 phase. There are three peaks at 147, 237, and 335 cm−1, respectively, which correspond to the J1, J2, and J3 modes of 1T phase MoS2 [16].
As shown in Fig. 2a, g-C3N4 NSs display a special 2D layered structure, and the lamellae of g-C3N4 NSs present irregular wrinkled sheet morphologies. Pure B-MoS2 displays flower-like assemblies composed of numerous small nanosheets (Fig. 2b). In Fig. 2c, B-MoS2@g-C3N4 composites still keep typical 2D sheet-shaped morphology. As shown in Fig. 2d, B-MoS2 NSs are assembled on the g-C3N4 NSs, and the lattice space distance (0.98 nm) is attributed to the (002) plane of MoS2 [6]. The close connection between g-C3N4 NSs and B-MoS2 NSs facilitates the fast transfer of photogenerated electrons from g-C3N4 NSs to B-MoS2 NSs, which can effectively inhibit electron-hole pairs recombination. Fig. S3 (Supporting information) shows the EDX mapping images of B-MoS2@g-C3N4–15 composites, which display even distribution of C, N, Mo, S and B, indicating the coexistence of g-C3N4 and B-MoS2.
|
Download:
|
| Fig. 2. SEM images of (a) pure g-C3N4 NSs and (b) B-MoS2 NSs. (c) TEM and (d) HR-TEM images of B-MoS2@g-C3N4–15 composites. | |
As shown in Fig. 3a, an obvious absorption of g-C3N4 NSs and B-MoS2@g-C3N4 composites is observed, and the absorption edge is ca. 430 nm. The band gap energy (Eg) of pure g-C3N4 NSs is obtained through the formula (αhν)1/2 ∝ hν − Eg, and the Eg of pure g-C3N4 NSs is calculated and the value is 2.61 eV (Fig. S4 in Supporting information). Significantly, the optical absorption of B-MoS2@g-C3N4 composites is stronger than that of pure g-C3N4 NSs, which indicates that B-MoS2 NSs loading onto g-C3N4 NSs can effectively enhance the light absorption ability of the catalyst. Among these, the light absorption of B-MoS2@g-C3N4–15 composites is the strongest, which can boost the production of photogenerated carriers. To further research the role of B ions doping, the UV–vis DRS absorption spectra of MoS2@g-C3N4–15 and B-MoS2@g-C3N4–15 composites are shown in Fig. 3b. After B ions are doped in 1T-MoS2 NSs, the light absorption of B-MoS2@g-C3N4–15 composites is improved (yellow curve in Fig. 3b), indicating that doping B into MoS2 can boost the utilization of light.
|
Download:
|
| Fig. 3. (a) UV–vis DRS absorption spectra of g-C3N4 NSs and B-MoS2@g-C3N4 composites. (b) UV–vis DRS absorption spectra of B-MoS2@g-C3N4–15 and B-MoS2@g-C3N4–15 composites. (c) Transient photocurrent responses. (d) EIS, (e) cumulated evolution and (f) photocatalytic H2 production rates of the samples. | |
To study the photogenerated charge separation and transfer properties of pure g-C3N4 NSs, MoS2@g-C3N4–15 and B-MoS2@g-C3N4–15 composites, the photoelectrochemical (PEC) analysis is performed (Figs. 3c and d). Fig. 3c displays that pure g-C3N4 NSs, MoS2@g-C3N4–15 and B-MoS2@g-C3N4–15 composites present the photocurrent responses on each illumination [6]. In addition, the photocurrent values of MoS2@g-C3N4–15 and B-MoS2@g-C3N4–15 composites are higher than that of pure g-C3N4 NSs, which indicates that 1T-MoS2 assembled on the g-C3N4 NSs can effectively improve the generation and separation of photogenerated carriers. Notably, the photocurrent value of B-MoS2@g-C3N4–15 composites is superior to MoS2@g-C3N4–15 composites, indicating that B ions doping could inhibit the recombination of electron-hole pairs. The charge transfer activity of pure g-C3N4 NSs, MoS2@g-C3N4–15 and B-MoS2@g-C3N4–15 composites is further explored through EIS measurement (Fig. 3d). MoS2@g-C3N4–15 and B-MoS2@g-C3N4–15 composites present a smaller arc radius than that of pure g-C3N4 NSs, indicating 1T-MoS2 can accelerate carrier separation (Fig. 3d). Besides, B-MoS2@g-C3N4–15 composites display the smallest arc radius, indicating the separation of photoexcited carriers of B-MoS2@g-C3N4–15 composites is most effective. Hence, B ions doped into 1T-MoS2 and MoS2 NSs assembled on the g-C3N4 NSs can synergistically promote photocatalytic activity.
To explore the possibility of hydrogen production, Mott-Schottky plots are tested to estimate the conduction band (CB) potential of pure g-C3N4 NSs. As shown in Fig. S5 (Supporting information), g-C3N4 is identified as an n-type semiconductor because of the positive slope of the curves in the Mott-Schottky plots. Through extrapolation to the x-intercept in Mott-Schottky plots, the flat band potential (EFB) of g-C3N4 is obtained (-0.42 eV vs. Ag/AgCl). The obtained EFB is converted to a potential vs. standard hydrogen electrode (NHE), and then the value is subtracted by 0.2 eV to obtain an ECB vs. NHE of the sample. Therefore, the ECB of g-C3N4 is -0.4 eV vs. NHE. By the valence band potential (EVB) = Eg + ECB and Eg results (Fig. S4 in Supporting information), the EVB of g-C3N4 is 2.21 eV vs. NHE (Fig. S6 in Supporting information). Based on the above research, the ECB of as-prepared g-C3N4 NSs is lower than 0 eV, which indicates that prepared g-C3N4 NSs can conduct photocatalytic hydrogen production.
Fig. 3e indicates that the photocatalytic hydrogen production of all photocatalysts is linear with time, indicating that the catalyst has stable photocatalytic H2 evolution performance. As shown in Fig. 3f, bare g-C3N4 shows an unacceptable photocatalytic hydrogen evolution performance (30.82 µmol h−1 g−1), which is attributed to the fast recombination of carriers and low light utilization. Yet, B-MoS2@g-C3N4–15 composites present excellent photocatalytic hydrogen production and H2 evolution rate, indicating that adding B-MoS2 NSs as cocatalysts can promote the photocatalytic activity of catalysts. The H2 evolution rate of B-MoS2@g-C3N4–15 composites (1612.75 µmol h−1 g−1) is 52.33, 1.3, 1.15 and 1.31 times of bare g-C3N4 NSs (30.82 µmol h−1 g−1), B-MoS2@g-C3N4–5 (1236.04 µmol h−1 g−1), B-MoS2@g-C3N4–10 (1405.12 µmol h−1 g−1), B-MoS2@g-C3N4–20 composites (1238.47 µmol h−1 g−1), respectively. The improved photocatalytic HER performance of B-MoS2@g-C3N4 composites indicates that B-MoS2 cocatalyst loading onto g-C3N4 improves the utilization of light to stimulate more photogenerated electrons, accelerates carrier separation, and inhibits electron-hole pairs recombination. However, increasing B-MoS2 NSs content from 15% to 20%, a decrease in photocatalytic performance is detected, owing to the overmuch B-MoS2 NSs loading on g-C3N4 NSs impediment the photo-absorption of g-C3N4 NSs. As shown in Fig. S7 (Supporting information), the photocurrent values of B-MoS2@g-C3N4–15 composites is higher than that of B-MoS2@g-C3N4–20 composites, which indicates that the overmuch B-MoS2 NSs loading on g-C3N4 NSs in B-MoS2@g-C3N4–20 is adverse for the photocatalytic performance of photocatalyst. To study the effect of B ions doping, we measure the hydrogen production of MoS2@g-C3N4–15 composites. As shown in Fig. S8 (Supporting information), the photocatalytic hydrogen production amount and rate of B-MoS2@g-C3N4–15 composites (1612.75 µmol h−1 g−1) is higher than that of MoS2@g-C3N4–15 composites (1370.68 µmol h−1 g−1), attributing that B ions doping into MoS2 can activate the base planes of MoS2 and offer more active sites. To further evaluate the H2 production cycle property of B-MoS2@g-C3N4–15 composites, the photocatalytic H2 evolution performance is tested for 15 h (Fig. S9 in Supporting information). Almost 92.4% of the incipient property is kept, and the micromorphology of B-MoS2@g-C3N4–15 composites after the cycle test (Fig. S10 in Supporting information) does not change significantly, indicating the good stability of B-MoS2@g-C3N4–15 composites. We further determine the apparent quantum efficiency (AQE) of pure g-C3N4 NSs and B-MoS2@g-C3N4–15 composites under light at λ = 370 nm irradiation. As shown in Fig. S11 (Supporting information), the AQE value of B-MoS2@g-C3N4–15 composites (5.54%) is higher than that of bare g-C3N4 NSs (0.41%), which illustrates the optical utilization of B-MoS2@g-C3N4–15 composites is higher than that of pure g-C3N4 NSs.
According to the above studies, we propose a possible photocatalytic hydrogen evolution mechanism to explicate the reason for the improved photocatalytic hydrogen evolution performance of B-MoS2@g-C3N4 composites (Scheme 1). The loading of B-MoS2 NSs improves the light absorption of g-C3N4 to stimulate more photogenerated carriers and offer more active sites for HER. In addition, B-MoS2 as a cocatalyst can capture photoinduced electrons, accelerate electron transfer and inhibit recombination of electron-hole pairs. B ions are doped into the lattice of 1T-MoS2, which can activate more base planes of 1T-MoS2 and is suitable for the HER to activate H+. The photoexcited electrons produced under the sunlight through g-C3N4 moved to B-MoS2, and then reduced water to hydrogen. Concurrently, TEOA as a sacrificial agent consumed the holes. Hence, the close combination of B-MoS2 and g-C3N4 boosts the photocatalytic hydrogen evolution activity of photocatalysts.
|
Download:
|
| Scheme 1. Schematic illustration of the photocatalytic H2 evolution mechanism of B-MoS2@g-C3N4 composites. | |
In conclusion, we have successfully synthesized an efficient B-MoS2@g-C3N4 composite for photocatalytic H2 evolution via cocatalyst and doping strategy. The as-prepared B-MoS2@g-C3N4 composites with 15 wt% B-MoS2 (B-MoS2@g-C3N4–15) present an extremely improved photocatalytic H2 evolution rate of 1612.75 µmol h−1 g−1, which is 52.33 times of bare g-C3N4 NSs (30.82 µmol h−1 g−1). The above experimental results confirm that the enhanced photocatalytic activity of B-MoS2@g-C3N4–15 composites may be assigned to the following factors: (1) As a cocatalyst, B doped 1T phase MoS2 NSs greatly improves the light utilization of photocatalyst and stimulates more photogenerated carriers; (2) B-MoS2 NSs with excellent conductivity are closely connected onto g-C3N4 NSs, which can accelerate electron transfer and inhibit carrier recombination; (3) The base planes are activated through doping B ions into the lattice of MoS2, which can induce the distortion of MoS2 crystal and provide more active sites for HER. This easy assembly strategy offers guidance for rationally constructed photocatalysts based on B-doped 1T phase MoS2 as a cocatalyst for H2 production.
Declaration of competing interestThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
AcknowledgmentsThe authors are thankful for fundings from the National Natural Science Foundation of China (No. 51872173), Taishan Scholars Program of Shandong Province (No. tsqn201812068), Natural Science Foundation of Shandong Province (No. ZR2022JQ21), and Higher School Youth Innovation Team of Shandong Province (No. 2019KJA013). The authors would like to thank Shiyanjia Lab (www. Shiyanjia. Com) for the XPS analysis.
Supplementary materialsSupplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2023.108246.
| [1] |
Y. Zhang, Z. Huang, Y. Dong, et al., Chem. Eng. J. 431 (2022) 134101. DOI:10.1016/j.cej.2021.134101 |
| [2] |
X. Wang, J. Gong, Y. Dong, et al., Mater. Today Phys. 27 (2022) 100806. DOI:10.1016/j.mtphys.2022.100806 |
| [3] |
W. Zhao, H. Niu, Y. Yang, et al., J. Environ. Sci. 2 (2022) 14. |
| [4] |
P. Qiu, R. Hu, X. Wang, et al., Mater. Chem. Front. 6 (2022) 718. DOI:10.1039/d1qm01471d |
| [5] |
J. Huang, J. Mei, J. Han, et al., J. Colloid Interface Sci. 556 (2019) 224-231. DOI:10.1016/j.jcis.2019.08.043 |
| [6] |
Z. Zhang, Y. Xue, X. Wang, et al., Chem. Eng. J. 421 (2021) 130016. DOI:10.1016/j.cej.2021.130016 |
| [7] |
J. Fu, Q. Xu, J. Low, et al., Appl. Catal. B: Environ. 243 (2019) 556-565. DOI:10.1016/j.apcatb.2018.11.011 |
| [8] |
Q. Gong, S. Cao, Y. Zhou, et al., Int. J. Hydrog. Energy 46 (2021) 21442-21453. DOI:10.1016/j.ijhydene.2021.03.227 |
| [9] |
Y. Li, S. Zhu, X. Kong, et al., Environ. Res. 197 (2021) 111002. DOI:10.1016/j.envres.2021.111002 |
| [10] |
Y. Zhou, Y. Zhang, J. Xue, et al., Chem. Eng. J. 420 (2021) 129574. DOI:10.1016/j.cej.2021.129574 |
| [11] |
H. Zhou, R. Chen, C. Han, et al., J. Colloid Interface Sci. 197 (2021) 111002. |
| [12] |
W. Qureshi, X. Hong, X. He, et al., Chemosphere 291 (2022) 132987. DOI:10.1016/j.chemosphere.2021.132987 |
| [13] |
L. Wang, Y. Qi, H. Li, et al., Sci. China Mater. 65 (2022) 974-984. DOI:10.1007/s40843-021-1833-1 |
| [14] |
B. Sun, P. Qiu, Z. Liang, et al., Chem. Eng. J. 406 (2021) 127177. DOI:10.1016/j.cej.2020.127177 |
| [15] |
Z. Liang, Y. Xue, X. Wang, et al., Mater. Today Nano 18 (2022) 100204. DOI:10.1016/j.mtnano.2022.100204 |
| [16] |
C. Sun, L. Wang, W. Zhao, et al., Adv. Funct. Mater. 32 (2022) 2206163. DOI:10.1002/adfm.202206163 |
| [17] |
H. Zhao, Z. Jiang, K. Xiao, et al., Appl. Catal. B: Environ. 280 (2020) 119456. |
| [18] |
H. Yuan, F. Fang, J. Dong, et al., Colloid. Surfaces A 641 (2022) 128575. DOI:10.1016/j.colsurfa.2022.128575 |
| [19] |
C. Vennapoosa, S. Gonuguntla, S. Saddam, et al., ACS Appl. Nano Mater. 5 (2022) 4848-4859. DOI:10.1021/acsanm.1c04419 |
| [20] |
X. Chen, S. Zhang, X. Qian, et al., Appl. Catal. B: Environ. 310 (2022) 121277. DOI:10.1016/j.apcatb.2022.121277 |
| [21] |
X. Chen, C. Ma, Z. Tan, et al., Chem. Eng. J. 433 (2022) 134504. DOI:10.1016/j.cej.2022.134504 |
| [22] |
H. Xu, J. Yi, X. She, et al., Appl. Catal. B: Environ. 220 (2018) 379-385. DOI:10.1016/j.apcatb.2017.08.035 |
| [23] |
Y. Jia, Z. Wang, L. Wang, et al., ChemSusChem 12 (2019) 3336-3342. DOI:10.1002/cssc.201900936 |
| [24] |
M. Liu, H. Li, S. Liu, et al., Nano. Res. 15 (2022) 5946-5952. DOI:10.1007/s12274-022-4267-9 |
| [25] |
X. Cheng, L. Wang, L. Xie, et al., Chem. Eng. J. 439 (2022) 135757. DOI:10.1016/j.cej.2022.135757 |
| [26] |
J. Deng, H. Li, J. Xiao, et al., Energy Environ. Sci. 8 (2015) 1594-1601. DOI:10.1039/C5EE00751H |
| [27] |
P. Qiu, Z. Liang, X. Liu, et al., J. Colloid Interface Sci. 571 (2020) 318-325. DOI:10.1016/j.jcis.2020.03.062 |
| [28] |
X. Zhang, S. He, S. Jiang, Carbon 156 (2020) 488-498. DOI:10.3390/w12020488 |
| [29] |
J. Sun, S. Yang, Z. Liang, et al., J. Colloid Interface Sci. 567 (2020) 300-307. DOI:10.1016/j.jcis.2020.02.029 |
| [30] |
Y. Chang, Z. Liu, X. Shen, et al., J. Hazard. Mater. 344 (2018) 1188-1197. DOI:10.1016/j.jhazmat.2017.10.040 |
| [31] |
F. Shaik, R. Milan, L. Amirav, ACS Appl. Mater. Interfaces 14 (2022) 21340-21347. DOI:10.1021/acsami.2c01906 |
| [32] |
X. Wang, Y. Xue, Z. Liang, et al., Mater. Today Adv. 15 (2022) 100257. DOI:10.1016/j.mtadv.2022.100257 |
| [33] |
D. Wei, Y. Liu, X. Shao, et al., J. Environ. Chem. Eng. 10 (2022) 107216. DOI:10.1016/j.jece.2022.107216 |
| [34] |
Q. Liu, Q. Fang, W. Chu, et al., Chem. Mater. 29 (2017) 4738-4744. DOI:10.1021/acs.chemmater.7b00446 |
| [35] |
X. Li, X. Lv, N. Li, et al., Appl. Catal. B: Environ. 243 (2019) 76-85. DOI:10.1016/j.apcatb.2018.10.033 |