Xiumin Ma*a,
Quantong Jianga,
Weimin Guob,
Meng Zhenga,
Weichen Xua,
Fubin Maa and
Baorong Houa
aKey Laboratory of Marine Environmental Corrosion and Bio-fouling, Institute of Oceanology, Chinese Academy of Science, 7 Nanhai Road, Qingdao 266071, China. E-mail: xma@qdio.ac.cn; Fax: +86 0532 82898731; Tel: +86 0532 82898731
bLuoyang Ship Materials Research Institute, State Key of Marine, Corrosion and Protection, Qingdao 266101, China
First published on 3rd March 2016
In this study, a sandwich-structured C3N4/Au/CdZnS photocatalyst with Au nanoparticles sandwiched between C3N4 and CdZnS was fabricated. The results showed that there was a 6.3 mmol g−1 increase of hydrogen production for C3N4/Au/CdZnS photocatalyst compared to C3N4/CdZnS in 4 h. The photoinduced I–t curve, electrochemical impedance spectrum (EIS), photogenerated electron lifetime and photoluminescence spectra showed that the photocatalytic ability of the composite was enhanced by anchoring Au as the electron transfer mediator between C3N4 and CdZnS through the Z-scheme charge-carrier transfer mechanism. Au functioned as an electron transition mediator, which could induce a combination between photogenerated electrons and holes in the CdZnS and C3N4, respectively. These photogenerated holes of CdZnS can be involved in the S2−/SO32− oxidation process and the electrons from the C3N4 surface might be involved in the water to hydrogen reduction reaction, thus enhancing the photocatalysis performance of C3N4/Au/CdZnS photocatalyst.
Bio-inspired by photosynthesis, Bard5 proposed the concept of artificial photosynthesis and designed a Z-scheme photocatalytic water splitting system, in which the energy required to drive each photocatalyst is reduced, allowing visible light to be utilized more efficiently than on conventional photocatalysis systems. Tada et al.6 proposed the concept of all-solid-state Z-scheme photocatalysts in a sandwich-structured CdS/Au/TiO2 photocatalyst, where Au nanoparticles (NPs) were anchored between CdS and TiO2. It was found that the photogenerated electrons in the conduction band (CB) of TiO2 could transfer through the Au layer and recombine with the photogenerated holes left in the valance band (VB) of CdS. The reducing property of the photogenerated electrons was enhanced and the strongly oxidizing photogenerated holes produced by TiO2 could be involved in the oxidation reaction of H2O, thus dramatically improving the efficiency of redox stability in the photocatalysis reaction. In Li et al.’s study,7 Au nanoparticles were deposited in the nanorod array and then CdS was in situ deposited on Au, which formed a sandwich-structured TiO2–Au–CdS photoanode. It was found that addition of Au nanoparticles increases the charge-transfer lifetime, reduces the trap-state Auger rate, suppresses the long-time scale back transfer, and partially compensates the negative effects of the surface trap states, thus improving the photoelectric conversion efficiency. Iwase8 took graphene oxide as an electron mediator and connected two semiconductors, Ru/SrTiO3:Rh and BiVO4, showing that graphene oxide could transfer photogenerated electrons as in a Z-scheme. Ye et al.9 studied Ag/AgX/BiOX photocatalytic activity and showed that Ag played a different role for Ag/AgX/BiOX photocatalysts. When Ag was present between AgX and BiOX, it functioned as a stronger photocatalyst than that on the AgX surface, contributing to a Z-scheme charge-carrier transfer mechanism of Ag sandwiched between AgX and BiOX.10
g-C3N4 is an analog of graphite polymer and studies have shown that g-C3N4 shows good photocatalytic property, which can be further improved by combination with other semiconductors11–15 or by impurity doping.16–23 The conduction band of g-C3N4 is more negative in potential while the valence band is negative, and the oxidation ability of the photogenerated holes is comparatively weak. If we fabricate the Z-scheme structure and induce photogenerated holes which can recombine with the electrons remaining in another semiconductor, the photocatalysis performance can be certainly improved. Based on this premise, the g-C3N4 photocatalyst material was fabricated and ultrasonically dispersed. Then Au nanoparticles were deposited on the surface of g-C3N4 as the electron transition mediator. Finally stable ZnCdS was applied on the surface of Ag to form a sandwich-structured Z-scheme g-C3N4–Au–CdZnS nanoparticles system.
A three-electrode system was used to measure the PEC performance. The photoelectrode, a large piece of platinum, and Ag/AgCl (saturated KCl) were used as the working, counter and reference electrodes, respectively. All tests were performed in 0.1 M Na2SO4 electrolyte by a CHI 660D electrochemical system (Shanghai Chenhua Instrument Co., Ltd., Shanghai, China). The photoinduced current density–time (I–t) curves was measured at a bias potential of 0 V (vs. Ag/AgCl). The electrochemical impedance spectroscopy measurements were carried out at a bias potential of 0 V and the frequency range was from 105 to 10−1 Hz.
![]() | ||
Fig. 1 XRD patterns of different materials (a) CN; (b) CN/Au; (c) CN/CdZnS; (d) CN/Au/CdZnS; (e) CdZnS. |
Fig. 2 shows SEM images of the composites. Fig. 2A is the microstructure of C3N4, from which we can observe lamellar structures of C3N4. Fig. 2B shows the microstructure of C3N4 after the CdZnS deposition. Fig. 2C is the microstructure of CN/Au which shows numerous large nanoparticles on the surface of the flake structure of C3N4, which is the result of Au deposition. Fig. 2D is the microstructure of CN/Au/CdZnS, which is similar to that of Fig. 2B, with numerous large nanoparticles scattered on the surface of the flake structure of C3N4. We can not identity the existence of Au or the microscopic distribution of CN/Au/CdZnS by the SEM result. Therefore EDS mapping and TEM are needed for further clarification.
The EDS mapping results of CN/Au/CdZnS are shown in Fig. 3. We tested the elements distribution of an selected area of this sample with 5000 times magnification. The image indicated that the sample was formed by the elements N, C, Cd, Zn, S and Au.
The CN/Au/CdZnS composite was characterized by X-ray photoelectron spectroscopy (XPS) in Fig. 4. Fig. 4A shows the XPS survey spectrum of the CN/Au/CdS composite, indicating the presence of N, C, Cd, Zn, S and Au in the composite, which is consistent with the SEM mapping results. As the XRD patterns (curve d) showed no diffraction peak of Au in CN/Au/CdZnS composite, we paid special attention to the high-resolution diffraction peak in the XPS survey. Fig. 4B shows the XPS spectrum of Au4f with the characteristic peak at 90.0 eV, confirming the existence of Au in the composite.
The morphologies and the distribution of CN/Au/CdZnS three-phase material were analyzed using TEM (see Fig. 5). Fig. 5A shows the TEM image of CN/Au, from this result we can find that some Au nanoparticles are distributed on the surface of the CN layer. Low-resolution TEM of CN/Au/CdZnS (Fig. 5B) shows that particles with less than 50 nm diameter were observed directly on the large surface of C3N4, with darker and smaller particles depositing on it. The high-resolution TEM image (Fig. 5C) reveals that the Au nanoparticles were loaded between C3N4 and CdZnS, indicating that C3N4, Au, CdZnS three phase composites were not randomly distributed but arranged in an orderly manner in the CN/Au/CdZnS sandwich nanostructure. Because of the strong electropositivity of the Au, the S2+ could be trapped by Au easily, so the CdZnS selectively grew on Au nanoparticles.6
![]() | ||
Fig. 5 TEM of (A) CN/Au; (B) low-resolution TEM of CN/Au/CdZnS; (C) high-resolution TEM of CN/Au/CdZnS. |
To evaluate the photocatalytic performance of the composite, its photocatalytic ability to reduce water to hydrogen under visible light illumination was measured. Fig. 6A shows the curves of photocatalytic hydrogen production yields of various composites over 4 h. Curve a of Fig. 6A represents that of C3N4, which indicates that the hydrogen production from pure g-CNS was very low and was only 2.0 mmol g−1. After the deposition of Au (curve b), the catalytic performance of the CN/Au was better than that of the pure C3N4, because of the photocatalytic ability of Au to reduce water to hydrogen.
Curve c of Fig. 6A is the photocatalytic hydrogen production of pure CdZnS, showing that the generated hydrogen is 14.9 mmol g−1, which was substantially improved compared to those of C3N4 because of the contribution of the expanded visible light absorption range of CdZnS. Curve d shows that the hydrogen yield using composite CN/CdZnS after 5 h light exposure reached 18.3 mmol g−1 and its catalytic ability was improved compared to that of pure CdZnS. This was resulting from the heterojunction electric field between CdZnS and C3N4, which could dramatically improve the efficiency of photoelectron generation and hole separation.
Curve e shows the photocatalytic hydrogen generation curve using the composite CN/Au/CdZnS, indicating that the hydrogen yield reached 24.6 mmol g−1 after 4 h exposure, higher than for composite CN/CdZnS. This demonstrates that when existing between the C3N4 and CdZnS, Au functioned as an electron transition system, which could induce a combination between photogenerated electrons and holes in CdZnS and C3N4, respectively. These photogenerated electrons of CdZnS can be involved in the S2−/SO32− reduction process and the holes from the C3N4 surface might be involved in the water to hydrogen oxidation reaction, thus enhancing the catalytic ability. Fig. 6B shows the photocatalytic stability measurement of CN/Au/CdZnS. Every measurement cycle was for 4 h. After five cycles, the hydrogen production yield shows no decrease, confirming that the CN/Au/CdZnS photocatalyst was quite stable in the S2−/SO32− system.
In order to study the role of Au between C3N4 and CdZnS, powders of composites CN/CdZnS and CN/Au/CdZnS were made into photoelectrodes and the photoinduced I–t and EIS data were tested in 0.1 M Na2SO4 solution. Fig. 7A shows the photoelectrode photoinduced I–t curves, which reflect the photoelectric conversion ability of the composites. Comparing the two I–t curves, we can see that when Au is anchored between C3N4 and CdZnS, its photogenerated current increased significantly, which shows that as the electron transfer mediator, Au can increase the separation efficiency of photo-induced electrons and holes as well as enhancing the transferability of those materials.
To some extent, EIS curves can reflect the electron migration resistance in thin films. Fig. 7B shows the thin-film photoelectrode EIS result of CN/CdZnS and CN/Au/CdZnS. Comparing these two curves we can easily see that when is Au anchored between C3N4 and CdZnS, the EIS arc resistance declined obviously, demonstrating that the existence of Au can lower the electron transferability in the composite thin film, thus optimizing the transferability of photogenerated electrons between C3N4 and CdZnS, to attain a Z-scheme charge-carrier transfer mechanism.
For most semiconductors, photogenerated electrons and holes can bind after the excitation by incident light, leading to the transfer of energy to fluorescence. Binding efficiency and photoelectron lifetime can be measured by monitoring the fluorescence intensity and lifetime. For photocatalyst CN/Au/CdZnS, if Au can act as carrier of photogenerated electrons, it will inhibit the binding efficiency between photogenerated holes and electrons, resulting in an increased lifetime of the photoelectrons. Therefore photoluminescence (PL) spectra and the excited state electron radioactive decay lifetime of CN/CdZnS and CN/Au/CdZnS were measured to characterize the effect of Au in the CN/Au/CdZnS photocatalyst. Fig. 8A shows the photoluminescence spectra of CN/CdZnS and CN/Au/CdZnS, indicating strong fluorescence peaks in the 400–750 nm range for both catalysts. The fluorescence peak intensity of CN/Au/CdZnS was much weaker than that of CN/CdZnS, because Au inhibited the binding efficiency between photogenerated holes and electrons. Fig. 8B shows the excited-state electron radioactive decay lifetime of CN/CdZnS and CN/Au/CdZnS. The excited-state electron radioactive decay lifetime of CN/CdZnS and CN/Au/CdZnS were 16.8 and 18.8 ns, respectively, indicating that the incorporation of Au increased the photoelectron lifetime of the CN/Au/CdZnS photocatalyst, because of the contribution of Au as the electron transfer mediator.
![]() | ||
Fig. 8 (A) PL spectra of CN/CdZnS and CN/Au/CdZnS powders; (B) excited-state electron radioactive decay of CN/CdZnS and CN/Au/CdZnS powders. |
Fig. 9 presents the UV-vis diffuse reflection results of the as prepared photocatalysts. The absorption threshold of CdZnS was nearly 528 nm, corresponding to a 2.3 eV bandgap.25 The absorption threshold of CN is approximately 460 nm, corresponding to a bandgap of 2.7 eV. However, the as-prepared CN/Au/CdZnS photocatalyst showed a similar absorption threshold as CdZnS, indicating that the CdZnS acts as the main light absorption material in the CN/Au/CdZnS composite.
Fig. 10 presents the possible electron transfer mechanism of the g-C3N4/Au/CdZnS composite photocatalyst. When this photocatalyst is illuminated by light, the photogenerated holes in the VB of g-C3N4 will shift to the Au and annihilate with the electrons generated by CdZnS. So the photogenerated holes on CdZnS with strong oxidizing power will participate in the oxidation process of S2+. Meanwhile, the photogenerated electrons in g-C3N4 with strong reducing power, will react with the H+ for hydrogen evolution. Through this Z-scheme electrons transfer process, the redox reaction ability of photogenerated electrons and holes could be increased, at the same time, the recombination process of the separated electrons and holes could be inhabited, so that photocatalytic performance of the g-C3N4/Au/CdZnS composite photocatalyst will be improved.
![]() | ||
Fig. 10 Proposed Z-scheme mechanism for electron transfer in g-C3N4/Au/CdZnS composite photocatalyst. |
This journal is © The Royal Society of Chemistry 2016 |