Assembly of g-C₃N₄-based type II and Z-scheme heterojunction anodes with improved charge separation for photoelectrojunction water oxidation

Abstract

Graphitic carbon nitride (g-CN) has been widely studied as a metal-free photocatalyst, leading to some excellent results; however, the rapid recombination of photogenerated charge carriers substantially limits its performance. Here, we establish two types of g-CN-based heterojunction (type II and nonmediator assisted Z-scheme) photoanodes on a transparent conducting substrate via coupling with rod-like and nanoparticulate WO, respectively. In these composites, g-CN film grown by electrophoretic deposition of exfoliated g-CN serves as the host or guest material. The optimized type II WO/g-CN composite exhibits an enhanced photocurrent of 0.82 mA cm at 1.23 V vs. RHE and an incident photo-to-current conversion efficiency (IPCE) of 33% as compared with pure WO nanorods (0.22 mA cm for photocurrent and 15% for IPCE). Relative to pure g-CN film (with a photocurrent of several microampere and an IPCE of 2%), a largely improved photocurrent of 0.22 mA cm and an IPCE of 20% were acquired for the Z-scheme g-CN/WO composite. The enhancement can be attributed to accelerated charge separation in the heterointerface because of the suitably aligned band gap between WO and g-CN, as confirmed by optical spectroscopy and ultraviolet photoelectron spectroscopy (UPS) analysis. The photocatalytic process and mechanism of the g-CN-based heterojunctions are proposed herein, which potentially explain the origin of the enhanced photoelectrochemical performance. This achievement and the fundamental information supplied here indicate the importance of rationally designing heterojunction photoelectrodes to improve the performance of semiconductors. This is particularly important for materials such as pure g-CN and WO, as their photoactivities are strongly restricted by high recombination rates.

An Efficient Strategy for Boosting Photogenerated Charge Separation by Using Porphyrins as Interfacial Charge Mediators

“…To further understand the effect of the CoPy layer in the assembled photoanodes on interface modulation, IMPS measurements developed by Peter and co‐workers were carried out. IMPS was conducted on different photoanodes to compare the relative rates of charge recombination ( k rc ) and transfer ( k ct ) occurring at the S–E interface (Figure a) . Pure BV showed a ringlike curve (Figure b), meaning k ct ≪ k rc , that is, surface recombination plays a substantial role, namely, the bulk BV exhibits poor PEC performance, which is consistent with a previous report .…”

“…According to the our previous work, the transit time ( τ d ) values of different samples can be estimated from τ d =(2π f max ) −1 , in which f max is the characteristic frequency at the lowest point of the IMPS plot and shows an increasing trend: R‐BV/CoPy/FN‐H<BV/CoPy/FN‐H< BV/FN‐H<BV/CoPy<BV (Figure d; see also Table S5). Typically, the calculated τ d values for BV/CoPy/FN‐H and BV/FN‐H are 0.92 and 1.00 ms, respectively, which indicates that the BV/CoPy/FN‐H photoanodes show a higher charge‐transfer rate than that of BV/FN‐H because the “setter” (e.g., CoPy) accelerates the hole transfer from BV to the OEC surface.…”

Tungsten Trioxide Nanostructures for Photoelectrochemical Water Splitting: Material Engineering and Charge Carrier Dynamic Manipulation

“…Thus the PEC performance of the composite photoanode with Z‐scheme heterojunction is improved. Similar work has also been done by Liu's team . Based on ultraviolet photoelectron spectroscopy (UPS) spectra and UV–vis absorption spectra, they investigated the band structure of WO 3 and g‐C 3 N 4 and found that both the CB and VB of g‐C 3 N 4 are more negative than those of WO 3 .…”

An Efficient Strategy for Boosting Photogenerated Charge Separation by Using Porphyrins as Interfacial Charge Mediators

“…According to the our previous work, [37] the transit time (t d )v alues of different samples can be estimated from t d = (2pf max ) À1 ,i nw hich f max is the characteristic frequencya tt he lowest point of the IMPS plot and shows an increasing trend: R-BV/CoPy/FN-H < BV/CoPy/FN-H < BV/FN-H < BV/ CoPy < BV (Figure 3d;s ee also Table S5). Ty pically,t he calculated t d values for BV/CoPy/FN-H and BV/FN-H are 0.92 and 1.00 ms,r espectively,w hich indicates that the BV/ CoPy/FN-H photoanodes show ah igher charge-transfer rate than that of BV/FN-H because the "setter" (e.g., CoPy) accelerates the hole transfer from BV to the OEC surface.…”

“…IMPS was conducted on different photoanodes to compare the relative rates of charge recombination (k rc )a nd transfer (k ct )o ccurring at the S-E interface (Figure 3a). [36,37] Pure BV showed ar inglike curve (Figure 3b), meaning k ct ! k rc ,t hat is,s urface recombination plays as ubstantial role, namely,the bulk BV exhibits poor PEC performance,which is consistent with ap revious report.…”