Scholarly article on topic 'A strategy of engineering impurity distribution in metal oxide nanostructures for photoelectrochemical water splitting'

A strategy of engineering impurity distribution in metal oxide nanostructures for photoelectrochemical water splitting Academic research paper on "Nano-technology"

CC BY-NC-ND
0
0
Share paper
Academic journal
Journal of Materiomics
OECD Field of science
Keywords
{"Metal oxides" / "Photoelectrochemical water splitting" / "Visible light" / "Engineered impurity distribution" / Doping}

Abstract of research paper on Nano-technology, author of scientific article — Shaohua Shen, Jianan Chen, Li Cai, Feng Ren, Liejin Guo

Abstract Impurity doping has been evidenced as one of the most effective methods to activate wide band gap semiconductors in visible light. However, these doped metal oxides always encounter serious charge recombination at the impurity levels introduced by the foreign dopants, leading to relatively low solar water splitting performances. In this perspective, a strategy of engineering impurity distribution is presented for metal oxide nanostructures for improved photoelectrochemical water splitting under visible light. Particular attention is paid to those doped systems with optical absorption and electron transport decoupled by spatially engineering the associated impurity distribution. In the context of this discussion, some selected systems of inhomogeneously doped ZnO nanostructured photoelectrodes, in which the visible-light absorbing and charge conducting regions are of isostructural nature to avoid introducting large amounts of interface recombination centers, are briefly discussed. Later on, a concept of impurity distributed homojunction is demonstrated to enhance charge separation by implementing a gradient in the dopant concentration. An ion implantation method is used to inject foreign dopants into ZnO and TiO2 nanorod arrays to create an impurity distributed homojunction with enhanced optical absorption and photoelectrochemical water splitting performances in visible light. The electronic and physicochemical properties of these ion implanted samples are of great dependence on the implanted ions as well as the metal oxide substrates. In particular, the N ion implanted ZnO nanorod arrays show good photoelectrochemical performance in visible light, because the gradient distribution of N dopants in ZnO nanorods not only extends optical absorption to visible light region, but also introduces internal driving force for charge carrier separation by the gradient band bending in the N dopants distributed ZnO homojunction. It is then conclusive that by spatially engineering impurity distribution, metal oxide nanostructures could be activated in visible light, synergistically with promoted charge carrier separation, for enhanced photoelectrochemical water splitting performances in visible light.

Academic research paper on topic "A strategy of engineering impurity distribution in metal oxide nanostructures for photoelectrochemical water splitting"

Accepted Manuscript

A Strategy of Engineering Impurity Distribution in Metal Oxide Nanostructures for Photoelectrochemical Water Splitting

Shaohua Shen, Jianan Chen, Li Cai, Feng Ren, Liejin Guo

Materiomics

PII: S2352-8478(15)00030-1

DOI: 10.1016/j.jmat.2015.02.003

Reference: JMAT 16

To appear in: Journal of Materiomics

Received Date: 6 January 2015 Revised Date: 19 January 2015 Accepted Date: 20 February 2015

Please cite this article as: Shen S, Chen J, Cai L, Ren F, Guo L, A Strategy of Engineering Impurity Distribution in Metal Oxide Nanostructures for Photoelectrochemical Water Splitting, Journal of Materiomics (2015), doi: 10.1016/j.jmat.2015.02.003.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract

Visible light

We proposed a strategy of impurity distribution engineering in metal oxide nanostructures for enhanced photoelectrochemical water splitting performances in visible light.

Original Article

A Strategy of Engineering Impurity Distribution in Metal Oxide Nanostructures for Photoelectrochemical Water Splitting

Shaohua Shen*1, Jianan Chen1, Li Cai1, Feng Ren2, Liejin Guo1

1. International Research Centre for Renewable Energy, State Key Laboratory of Multiphase Flow in Power Engineering, Xi'an Jiaotong University, Shaanxi 710049, China. Email:

shshen_xjtu@mail.xjtu.edu.cn

2. School of Physics and Technology, Center for Ion Beam Application, Wuhan University,

Wuhan 430072, China

Received date:2015-01-06; Revised date:2015-01-19; Accepted date:2015-02-20

Abstract: Impurity doping has been evidenced as one of the most effective methods to activate wide band gap semiconductors in visible light. However, these doped metal oxides always encounter serious charge recombination at the impurity levels introduced by the foreign dopants, leading to relatively low solar water splitting performances. In this perspective, a new strategy of engineering impurity distribution has been proposed for metal oxide nanostructures for improved photoelectrochemical water splitting under visible light. Particular attention is paid to those doped systems with optical absorption and electron transport decoupled by spatially engineering the associated impurity distribution. In the context of this discussion, some selected systems of inhomogeneously doped ZnO nanostructured photoelectrodes, in which the visible-light absorbing and charge conducting regions are of isostructural nature to yield low concentrations of interface recombination centers, are introduced. Later on, a concept of impurity distributed homojunction is demonstrated to enhance charge separation by implementing a gradient in the dopant concentration. An ion implantation method is used to inject foreign dopants into ZnO and TiO2 nanorod arrays to create an impurity distributed homojunction with enhanced optical absorption and photoelectrochemical water splitting performances in visible light. The electronic and physicochemical properties of these ion implanted samples are of great dependence on the implanted ions as well as the metal oxide substrates. In particular, the N ion implanted ZnO nanorod arrays show good photoelectrochemical performance in visible light, because the gradient distribution of N dopants in ZnO nanorods not only extends optical absorption to visible light region, but also introduces internal driving force for charge carrier separation by the gradient band bending in the N dopants distributed ZnO homojunction. It is then conclusive that by

spatially engineering impurity distribution, metal oxide nanostructures could be activated in visible light, synergistically with promoted charge carrier separation, for enhanced photoelectrochemical water splitting performances in visible light.

Keywords: Metal oxides; Photoelectrochemical water splitting; Visible light; Engineered impurity distribution; Doping

1. Introduction

Solar-driven water splitting to produce H2 and O2 has been received much attention from its potential as a green and low cost energy solution.[1]-[4] Since the first discovery of photo-assisted electrochemical water oxidation on TiO2 in 1972,[5] photoelectrochemical (PEC) water splitting based on semiconductors has been studied extensively for solar-driven hydrogen conversion.[6][9] As the best representatives of metal oxide semiconductors, ZnO and TiO2 have been intensively investigated as photoanodes for solar water splitting, due to their excellent properties such as low cost, nontoxicity, chemical and thermal stability and superior electron mobility. However, the major drawback of a large band gap of ~3.26 eV for ZnO and ~3.10 eV for TiO2, which makes them only utilize ultraviolet (UV) light, limiting their application for solar water splitting. [10][11]

To utilize the large portion of visible and even infrared light in solar spectrum, many approaches in order to improve the light absorption ability of ZnO and TiO2 have been developed. Dye sensitization has been well established since Gratzel's pioneering work in 1991,[12] and different dyes have been used to sensitize the wide band gap semiconductors such as ZnO and TiO2 to utilize the very large portion of visible light in sunlight, due to sufficiently narrow "gap" between the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) of dyes.[13]-[15] Although impressive photocurrents have been achieved, photostability is still the bottleneck for dye sensitization. Inspired by dye sensitization, inorganic quantum dots (QDs) were also used in purpose of efficient sunlight utilization.[16]-[18] Moreover, the type II band alignment formed in the interface between wide band gap semiconductors and QDs is of great advantage for the electron-hole separation.[19] However, QDs, metal chalcogenides in most occasions, are of poor stability and high toxicity, which greatly limits the application of QDs sensitized systems for PEC water splitting. Ion doping as another effective approach to modify the band structure of semiconductor for visible light harvesting has been frequently reported by numerous groups.[1] The mechanisms for visible light response after ion doping are shown in Fig. 1, taking TiO2 as the example of wide band gap semiconductor. For metal ions doping, additional discrete levels, acting as either donor or acceptor levels in the forbidden band of TiO2, introduced by impurity dopants would induce visible light charge transition. For nonmetal ions doping, such as N and C doping, the p orbit of the incorporated ions will mix with the O 2p orbit to level up the valence band of TiO2 and thus the band gap is narrowed.

2. Doping to metal oxides for solar water splitting

There is considerable content in the literature that suggests introducing impurity dopants associated with visible-light-active electronic transition as one effective technique to sensitize metal oxides to visible light. So far, plenty of work has been conducted on metal ions doped TiO2,[20]-[25] and various elements including Si, N, Cr, V, Fe, Co, etc., have been successfully doped into TiO2 for solar water splitting.[26]-[29] For example, Vinogradov et al.[30] revealed that Fe doping effectively enhanced the optical absorption ability of TiO2, and the Fe-doped TiO2

presented a photocurrent density as high as 54 (jA/cm", which was 2.6 times higher than that of undoped TiCh. Xu el al[31] successfully incorporated Sn dopants into Ti02 nanowires by a one-pot hydrothermal synthesis. The obtained Sn-doped TiCb showed increased photocurrent due to the significantly increased density of n-type charge carrier by the Sn doping. However, the dopants introduced impurity levels will act as carrier recombination centers, resulting in loss of excited carriers and low charge mobility. Larsen et al [32] have even observed a decrease in the PEC performances for Cr-doped TiO? nanorod arrays used as photoanodes for solar water splitting. Then, co-doping has been reported to be an effective way to narrow the band gap with defect levels passivated. Hoang et a/. [33] reported an increased photocurrent density of hydrogenation and nitridation cotreated TiCh nanowire arrays, which reached 0.16 raA/cm" at 1.23 V us. reversible hydrogen electrode (RHE) under visible light illumination. They also found that N and Ta co-doped TiCh nanowire arrays synthesized via a solvothermal route followed by annealing in NH3 flow displayed significant enhancement in PEC performance with the photocurrent densities reaching 0.52 and 0.18 raA/cm" under solar and visible light illumination, respectively.[34] The significant performance enhancement of N and Ta co-doped TiCh film originated from suppression of the formation of amorphous layers on the nanowires during nitridation as well as the relatively few recombination centers originating from the charge compensation effects. Recently, Cho et al[35] prepared W and C co-doped TiCb by sequentially annealing W-precursor-coated TiO? nanowires in flame and carbon monoxide gas, which demonstrated double saturation photocurrent of undoped TiO? for PEC water splitting. Such significant performance enhancement was attributed to a greatly improved electrical conductivity and activity for oxygen-evolution reaction due to the synergistic effects of co-doping.

Various elements, such as C, Co, Cu, Fe, etc.,[36][40] have been successfully doped into ZnO for solar water splitting under visible light. For example, Yousefi et al [41] prepared Ce-doped ZnO films by sol-gel method at an annealing temperature of 500 °C, which displayed a maximum photocurrent of 1.2 [lA/mm" at 1.2 V us. Ag/AgCl under visible light irradiation. As a well-studied nonmetal ion dopant, N has been doped into ZnO for efficient PEC water splitting by different methods. Ahn et al [42] deposited ZnO:N films by reactive ratio frequency magnetron sputtering. Compared to pure ZnO films, N incorporation narrowed the bandgap of ZnO and shifted the optical absorption into the visible light regions. As a result, the ZnO:N films exhibited higher photocurrents than pure ZnO films under visible light illumination. In addition, the research of co-doped ZnO films have been reported as well.[43][44] In 2011, Shet et al[45] synthesized Ga, N co-doped ZnO thin film by co-sputtering in mixed N2 and O2 gas ambient at room temperature. Under visible light, Ga, N co-doped ZnO films deposited at 500 °C displayed a maximum photocurrent of about 100 (jA/cnr at 0.8 V us. Ag/AgCl.

3. A strategy to decouple optical absorption and electron transport by spatially engineering the associated impurity distribution in ZnO nanostructures: Discussion of selected recent systems

The previous section provides a summary of some doping systems of TiO2 and ZnO from the literatures that highlight the effectiveness of ion doping to visible light harvesting. However, the visible light activities of these doped metal oxides are always relatively low, mainly because of the generally poor transport of carriers associated with isolated impurity states. These observations suggest that the traditional doping of metal oxide photoelectrodes, in which the impurity introduced intra-bandgap energy levels might act as recombination centers for photoexcited charge carriers,[1] presents the unacceptable situation where many visible-light excited charge carriers recombine before reaching the back contact or the oxide-liquid interface for electrochemical reactions.

To accomplish the disparity between dopant-induced visible light absorption and charge recombination in metal oxide materials of interest for PEC application, a viable strategy to decouple the optical absorption and electronic conduction processes has been proposed in some rationally designed photoelectrode architectures. For example, in the architectures of QDs sensitized metal oxide photoelectrodes (e.g., CdSe QDs sensitized ZnO),[46][47] the operational purpose of QDs is to selectively absorb visible light to generate charges and ZnO to conduct these photogenerated electrons for collection in an external circuit. This configuration has been also applied to the metal oxide based photoelectrodes with doped metal oxide as the visible light sensitizer. In this section, some selected systems of inhomogeneously doped metal oxide photoelectrodes, in which the visible-light absorbing and charge conducting regions are of isostructural nature to yield low concentrations of interface recombination centers, will be demonstrated with ZnO nanostructures.

3.1 A ZnO:Al/ZnO:Ni core/shell nanorod structure system

With knowledge of the known deleterious effect of intra-bandgap impurities on charge transport properties of metal oxides, Kronawitter et al. introduced the concept of decoupling optical absorption and electron transport by spatially separating the associated impurity distributions in a ZnO:Al/ZnO:Ni core/shell nanorod structure.[48] In this unique nanostructured photoelectrode, ZnO nanorods doped in core regions with shallow Al donor levels allowed for the enhanced electronic conduction and in the near-surface volume with intragap Ni impurity states for the increased optical absorption in visible light. Fig. 2(a) shows the schematic of the proposed operating mechanisms within this doped ZnO structure. Ionized Al was identified as a suitable dopant to facilitate electronic conduction to the back contact during PEC operation, because Al dopant acting as a shallow donor in the ZnO crystal lattice was associated with large increase in electronic conductivity. Deposition of ZnO:Ni ultrathin layer onto ZnO:Al nanorods allowed for the optical absorption features beyond 400 nm associated with a change in sample color from transparent-white for ZnO:Al to green for ZnO:Al/ZnO:Ni, which should be related to the electronic transitions associated with Ni(II) with tetrahedral symmetry. Moreover, this nanostructured homojunction architecture gave rise to a dramatic increase in optical thickness at visible wavelengths, permitting the use of ZnO:Ni absorber layers with small physical thickness but large optical thickness. As presented in Fig. 2(b), approximately a

three-fold enhancement in conversion efficiencies for solar-abundant visible wavelengths was achieved over the ZnO:Al/ZnO:Ni nanostructure by tailoring the dopant profiles within the structures to maximize both the spectral overlap of optical absorption with the terrestrial solar flux and the quantity of photogenerated minority carriers reaching the oxide-water interface. The proposed operating mechanisms and band diagram within the core/shell structure were established in Fig. 2(c). Under visible light irradiation, electrons created in ZnO:Ni absorber layer by Ni-related charge transfer transition could be conducted by ZnO:Al nanorods to the external circuit to involve in hydrogen production on Pt counter electrode, with photoexcited holes transfer to the surface to oxidize water.

3.2 A ZnO/ZnO:Cr nanorods/nanosheets isostructural junction system As motivated by the conceptual framework for the design of visible-light active photoanodes based on the spatially inhomogeneous doping of metal oxide nanostructures, Shen et al. intentionally engineered Cr impurities in ZnO and fabricated a ZnO/ZnO:Cr isostructural nanojunction electrode via a two-step electrodeposition.[49] As schemed in Fig. 3(a), in this designed nanostructure, ZnO:Cr nanosheets with intragap Cr impurity states increased optical absorption; ZnO nanorods provided a direct pathway to transport photogenerated electrons for collection in the external circuit. It was also believed that the isostructural nature of the absorbing (ZnO:Cr nanosheets) and conducting (ZnO nanorods) regions had the potential to yield low concentrations of interface recombination centers, favoring efficient interface charge migration. As presented in Fig. 3(b), the isostructural nanojunction showed considerable performance for PEC water splitting under visible light (X > 510 nm), while its constituent components alone showed no visible light PEC activity. The band diagram in Fig. 3(c) shows a proposed operation mechanism of the ZnO/ZnO:Cr photoanode under visible light (X > 510 nm) illumination. In the ZnO:Cr component, impurity states derived from Cr dopants sensitized ZnO to visible light through the Cr 3d related charge transfer transitions. Upon visible light illumination, the photoexcited electrons via the visible charge transfer transition in ZnO:Cr nanosheets could fluently migrate through the ZnO:Cr/ZnO interface and then be directed by ZnO nanorods to external circuit, while photoexcited holes transferred to the surface to involve in the water oxidation reaction.

4. The gradient-doped ZnO system

4.1 Introduction to gradient doping to metal oxides for solar water splitting As discussed above, ion doping is effective and has been widely used to activate wide band gap semiconductors, especially metal oxides like ZnO and TiO2, etc., in visible light. However, these doped metal oxides always encounter serious charge recombination at the impurity levels introduced by the foreign dopants, leading to relatively low solar water splitting performances. Thus, good charge separation is pivotal for a semiconductor photoelectrode to achieve high solar water splitting performance. As illustrated in the previous section, an effective way to improve charge transfer and separation is to decouple optical absorption and electron transport

by spatially engineering the associated impurity distribution in the doped systems. Such doped system has been yet limited by available materials and synthesis techniques and thus few reported.[48][49] Another common way is to create heterojunctions those, unfortunately, always have interfaces with large amounts of defects acting as charge recombination centers. Thus, a different and rational strategy is needed to improve the charge separation in the doped systems.

Homojunction, a well-established concept in semiconductor physics by forming p-n homojunctions or homojunctions with gradient composition for enhancing bulk charge separation, has been recently proposed to design photocatalyst or photoelectrode systems for solar water splitting. Lin et al. [50] deposited a thin p-type layer of Mg-doped a-Fe2O3 over undoped n-type a-Fe2O3 to create an n-p homojunction that could drive charge separation and produce an additional photovoltage. This design of homojunction is analogous to that demonstrated previously by Boettcher et al. [51] for PEC hydrogen evolution on p-n+ Si microwire arrays, confirming the function of p-n homojunction in improving PEC water splitting performance. Recently, a strategy of homojunction engineered with gradient dopant distribution was proposed to improve the charge separation by bending conduction and/or valence bands over a larger region in the bulk of the sample. Liu et al. [52] reported a red anatase TiO2 microsphere with a bandgap gradient varying from 3.22 eV in its core to 1.94 eV on its surface by gradiently elevating the valence band edge, with a gradient B-N co-doping with high concentrations in the shell. Abdi et al.[53] demonstrated that the poor carrier-separation efficiency in BiVO4 photoanode can be overcome by introducing a gradient dopant concentration to create a distributed n+-n homojunction. As shown in Fig. 4(a)-4(b), an enhanced carrier separation could be expected, due to the additional band bending in the W:BiVO4 homonjunction. As comparison, in a reverse homojunction (Fig. 4(c)), the band bending is present in the opposite direction and should act as a barrier for carrier separation. This observation evidenced that the improvement in carrier-separation efficiency could be caused by the additional band bending at the W:BiVO4/BiVO4 n+-n homojunction interface. Furthermore, they synthesized a BiVO4 photoanode with a 10-step gradient in W doping, starting from 1% W at the interface with the back contact to 0% W at the semiconductor/electrolyte interface (Fig. 4(d)). This extended the presence of band bending over the entire thickness of the photoanode. As a result, the carrier-separation efficiency increased to ~60% at 1.23 V vs. RHE, compared with ~38% for homogeneously doped BiVO4, and hence an approximately 3 fold improvement of the AM1.5 photocurrent, up to 3 mA/cm at 1.23V vs. RHE. These results indicate that the concept of impurity distribution engineered homojunction could be generally applicable for improving charge separation in a doped photoelectrode system.

In this perspective, ion implantation as a facile physical doping method was introduced to dope various metal or nonmetal ions into metal oxide nanorod arrays. Fig. 5 provides the schematic diagram illustrating the ion implantation method, taking the process of implanting Cu ions into ZnO nanorod arrays as the example.[54] In the present case, all the ion implanted ZnO and TiO2 nanorod arrays showed obvious optical absorption in visible light, indicating that ion implantation is an effective

method to activate wide band gap semiconductor in visible light. However, the visible light PEC performances of these ion implanted samples were very different, closely depending on the physicochemical properties of the implanted ions as well as the metal oxide substrates, as described later in this report.

4.2 Methods

The fabrication procedure of doped ZnO and TiO2 nanorod arrays in this work consists of a combination of solution chemistry and physical ion implantation techniques. The ZnO and TiO2 nanorod arrays were first fabricated onto SnO2:F-coated glass (FTO; TEC-15, 15 Q/sp.) by a hydrothermal method described in detail in refs. [54] and [55], respectively. Dopant ions were implanted into ZnO or TiO2 nanorod arrays using a metal vapor vacuum arc (MEVVA) ion source implanter.[56] This synthesis procedure was carried out at room temperature with an accelerator voltage of 30 kV, and the nominal doses were 2 x 1016 ions/cm2 for Cu ions implantation, 5 x 1015 ions/cm2 for Fe and W ions implantation, and 2.5 x 1015 ions/cm2 for N ions implantation, respectively. Then the ion implanted samples were annealed at 450 °C for 1 h with a ramping rate of 5 °C/min. The doped ZnO and TiO2 nanorod arrays obtained via ion implantation were denoted as M-ZnO and M-TiO2 (M = Cu, Fe, W, N), if not specifically indicated. The optical absorption properties of all the doped films were determined with a Hitachi U-4100 UV-vis-near-IR spectrophotometer.

PEC measurements were carried out in a conventional three-electrode cell. The doped ZnO and TiO2 nanorod arrays mounted onto a special designed electrode holder were used as the working electrodes. The surface areas exposed to the electrolyte were fixed at 0.785 cm . An Ag/AgCl served as a reference electrode and a large area platinum plate was used as a counter electrode. A 0.5 M aqueous solution of Na2SO4 was used as the electrolyte. An electrochemical workstation (CHI 760D) and a 350 W Xe lamp solar simulator (100 mW/cm ) with adjustable power settings through an AM 1.5 G filter (Oriel) were used for amperometric photocurrent-potential (I-V) and photocurrent-time (I-t) measurements, with a 420 nm cut-off filter used to block UV light.

4.3 Metal ions implanted metal oxide nanorod arrays for PEC water splitting under visible light: ZnO vs. TiO2

As an effective doping method, ion implantation can inject dopant ions into the bulk of substrates (ZnO and TiO2 nanorod arrays in the present study) in a highly dispersed state, and then modify the electronic structures of the substrates without destroying their initial morphology. In a set of recent experiments in our group, different metal ions such as Cu, V, Fe and W have been doped into different substrates such as ZnO and TiO2 nanorod arrays by the ion implantation method. Both doped ZnO and TiO2 photoanodes exhibited extended optical absorption edges into visible light region and considerable PEC performances for water splitting under visible light (X > 420 nm). Detailed investigations have been demonstrated with Cu and V ions implanted ZnO nanorod arrays as photoanodes,[54][57] as briefly introduced in the following.

ZnO nanorod arrays were firstly fabricated by a hydrothermal method. After Cu or V ions implantation, ZnO nanorod arrays displayed obvious optical absorption in visible light region, with nanorod morphology kept almost unchanged. The absorption edges showed a gradual red shift as the implantation dose increased. This should be attributed to the impurity levels created by Cu or V dopants in the forbidden bad of ZnO, which narrowed the band gap of ZnO. Moreover, it was observed that the concentrations of Cu or V dopants gradually decreased with the increasing depth of ZnO nanorod films, indicating the gradient distribution of doping impurities in the ZnO nanorods. Fig. 6 shows the proposed microstructure and energy band diagram of Cu or V ions implanted ZnO nanorod arrays. It was clearly demonstrated that the impurity levels introduced by Cu or V ions doping, acting as electron donor and/or acceptor levels, in the forbidden band led to the visible light charge transition and hence the reduced band gap of ZnO. As shown in Fig. 6(a), when the Cu ions doped

ZnO nanorod arrays were illuminated by visible light, electrons in the valence band

can be first excited to the additional levels (Cu and Cu states) of the Cu dopants and then further excited to the conduction band of ZnO. For the V ions doped ZnO system (Fig. 6(b)), V4+ ions instead of V5+ ions created impurity donor level (3d1 level) above the valence band of ZnO, and hence in this additional level of V4+ dopant the electrons could be excited to the conduction band of ZnO under visible light irradiation.

Fig. 7 displays the amperometric I-t curves for Cu and V ions doped ZnO nanorod arrays under chopped visible light illumination (X > 420 nm). Compared with pure ZnO nanorod arrays with neglectable photocurrent under visible light, both Cu and V ions doped ZnO nanorods showed considerably improved and quite stable photocurrent densities, which should be mainly due to the visible light sensitization of ZnO by Cu and V ions implantation. By varying the dopant concentrations in ZnO nanorod arrays, the visible light PEC activities could be further optimized. Although either Cu or V ions doped ZnO nanorods showed relatively low PEC performances for water splitting under visible light, these observations may offer a new approach to developing novel photoelectrodes for solar water splitting. In our following systematic investigation, the ion implantation method was applied to doping different metal ions (e.g., Cu, Fe, and W) into different substrates (e.g., ZnO and TiO2 nanorod arrays) for PEC water splitting under visible light.

Fig. 8 shows the optical absorption spectra of undoped and metal ions doped ZnO and TiO2 nanorod arrays fabricated by the ion implantation method. The pure ZnO and TiO2 nanorod arrays could only utilize the ultraviolet light of X < 400 nm (Fig. 8(a)). After metal ion implantation, all the doped ZnO and TiO2 showed optical absorption obviously red shift to visible light region as compared to the undoped ZnO and TiO2 nanorod arrays (Fig. 8(b)-8(d)). This should be related to the successful doping of metal ions into ZnO and TiO2. It was believed that metal ion dopants always created discrete impurity levels in the forbidden bands of metal oxide semiconductors.[1][4] These additional levels could act as electron and/or acceptor levels for visible-light induced charge transition, making wide band gap semiconductors sensitive to visible light. As a result, the band gaps of ZnO and TiO2

were effectively narrowed by Cu, Fe, and W ions implantation.

Fig. 9 shows the I-V curves of pure and metal ions doped ZnO and T1O2 nanorod arrays under chopped visible light illumination (k > 420 nm). As shown in Fig. 9(a), the pure ZnO and T1O2 nanorod arrays exhibited neglectable photocurrents. Compared with the undoped films, the photocurrent densities of metal ions doped ZnO and T1O2 nanorod arrays were improved more or less under visible light irradiation (k > 420 nm), shown in Fig. 9(b)-9(d). However, the visible light PEC activities for all the doped ZnO and T1O2 nanorod arrays were still very low, mainly due to the deleterious effect of metal ion dopants. It is well known that metal ion dopants always created discrete impurity levels in the forbidden bands of ZnO and T1O2. These discrete impurity levels, though could narrow the band gaps of ZnO and Ti02 for visible light sensitization, would also act as charge carrier recombination centers, leading to serious electron-hole recombination. Moreover, the high ion flux implantation process could lead to damage in ZnO and Ti02 crystal lattice and produce vacancies, which could also serve as charge recombination centers.[54][57] Therefore, the low visible light PEC performances of these metal ions implanted ZnO and Ti02 nanorod arrays are reasonable. By further comparison between doped ZnO and Ti02, it is easily found that the visible light PEC activities of doped ZnO nanorod arrays are always higher than those of the Ti02 nanorod arrays doped with the same metal ions. This should be possibly related to their various electrical conductivities, regardless of the lattice defects introduced by ion implantation. It was reported that the typical electron mobility in ZnO is 10-100 folds higher than Ti02, leading to reduced electrical resistance and enhanced electron transfer ability in ZnO.[58] From the view of this point, ZnO should be the better substrate candidate for ion implantation to develop high efficiency photoanodes for PEC water splitting under visible light.

4.4 Nonmetal ions implanted ZnO nanorod arrays for PEC water splitting under visible light

To avoid the negative effects of ion implanted metal dopants acting as charge recombination centers, in our following study, N ions, as the alternative dopants of metal ions, were doped into ZnO nanorod arrays by ion implantation to activate ZnO in visible light for PEC water splitting. N ions have been evidenced to the effective dopants for sensitizing wide band gap metal oxides like Ti02, Ta20s, ZnO, and WO3 in visible light,[59][62] by leveling up the O 2p orbital composed valence band of metal oxides instead of forming intra-bandgap levels in the forbidden band. Fig. 10(a) shows the UV-Vis spectra of pure ZnO and N ions implanted ZnO nanorod arrays. The pure ZnO showed optical absorption only in ultraviolet region with wavelength of < 390 nm. After N ion implantation, the optical absorption edges of N doped ZnO nanorod arrays significantly red shift to ca. 600 nm. It is clear that N implantation effectively narrowed the band gap of ZnO. As shown in Fig. 10(b), the PEC activity of N ions implanted ZnO nanorod arrays was greatly enhanced, with photocurrent density achieving -160 (jA/cm" at 1.1 V us. Ag/AgCl, which was about 2 orders of magnitude higher than that of the bare ZnO nanorod arrays and more than 10 times higher than those of the metal ions doped ZnO nanorod arrays (Fig. 9). This indicates

that when compared to metal ions, N ions are the more effective dopants to activate ZnO in visible light for PEC water splitting.

It was previously revealed that different from metal ion dopants creating discrete impurity levels in the forbidden band, N ion dopants are more likely to shift the valence band edge upward, which narrows the band gap of ZnO. In addition, ZnO is an n-type semiconductor, and then N doping will lower the Fermi level of ZnO due to the p-type nature of N ion dopants. When N-doped and undoped ZnO are brought into contact, the Fermi energy levels equilibrate by electron transfer from the undoped part to the N-doped part of the material, resulting in the formation of a modest depletion layer at the interface between N-doped and undoped ZnO.[53] In the present study, N dopants were injected into ZnO nanorods, yielding a gradient distribution in dopant concentration along the ZnO nanrods (Fig. 11(a)). Such a gradient distribution of N dopants induces alignment of Fermi levels, obtaining a N-doped ZnO homojunction with gradient band structure (Fig. 11(b)), which is supposed to induce built-in electric field for promoted charge transfer. Under visible light irradiation, the visible light excited electrons in conduction band and holes in valence band can efficiently separate as driven by the built-in electric field and transfer to the back contact and the nanorod surface through the gradient band structure for electrochemical water splitting reactions. This means that the gradient distribution of N dopants in ZnO nanorods not only extends optical absorption to visible light region, but also introduces internal driving force for charge carrier transfer and separation by creating gradient band structure.[63] Thus, it is reasonable that the N ions implanted ZnO nanorod arrays show much higher visible light PEC performances than those metal ions implanted ZnO nanorod arrays. These observations suggest that ion implantation could be an effective method to developing rationally designed nanostructures for efficient solar water splitting.

5. Conclusions and perspective

A perspective of impurity distribution engineering in metal oxide nanostructures has been presented for photoelectrochemical water splitting under visible light. Performance enhancement could be associated with the synergetic effects of spatially inhomogeneous doping, extending visible light absorption and favoring charge carrier separation.

The unique strategy of decoupling optical absorption and electron transport has succeeded with some doped ZnO systems with visible-light absorbing and charge conducting regions are of isostructural nature yielding low concentrations of interface recombination centers, leading to considerable enhancement in visible light photoelectrochemical water splitting performances. It is, however, not always readily accessible to materials and dopants with suitable electronic and physicochemical properties as well as method for facile fabrication of such impurity engineered junctions of isostructural nature. The demonstration of the successful combination of large-scale wet-chemical and physical ion implantation techniques for the fabrication of rationally designed metal oxide nanostructures with a gradient in the dopant profile is of direct relevance to optimized and doped nanostructured oxide materials by

addressing the primary requirements for electrodes enabling efficient solar energy conversion: visible light activity, efficient charge carrier separation, and potential for low-cost fabrication. These results suggest that through impurity distribution engineering it is possible to tune optical absorption and charge carrier transport properties for photoelectrochemical applications of metal oxide nanostructures. This concept of impurity distribution engineering should be also applicable to the other metal oxide photoelectrodes, including a-Fe2O3, WO3, BiVO4, etc.

Based on these observations, the next steps toward a continuing understanding of metal oxide photoelectrode efficiency enhancement driven by engineered impurity distribution should be comprehensive evaluation of the electronic and interfacial structures. In particular, understanding the role of the distributed dopants in electrodes in band structure and charge transport should be emphasized in future work.[64][65] Novel in-situ synchrotron-based X-ray absorption spectroscopy measurements provide important new insights toward understanding of photoinduced electronic structure change and charge transition in such systems.[66][67] Multiple time-scale studies of carrier dynamics by laser transient spectroscopy will also provide additional intrinsic clues for performance optimization.[9][69] Finally, a facile and widely applicable method is required for large scale and low cost fabrication of nanostructured photoelectrodes for solar water splitting.

Acknowledgment

The authors gratefully acknowledge the financial support of the National Natural Science Foundation of China (No. 51323011, No. 51102194), the Natural Science Foundation of Shaanxi Province (No. 2014KW07-02), the Natural Science Foundation of Jiangsu Province (No. BK20141212) and the Nano Research Program of Suzhou City (No. ZXG201442, No. ZXG2013003). S. Shen is supported by the Foundation for the Author of National Excellent Doctoral Dissertation of P. R. China (No. 201335) and the "Fundamental Research Funds for the Central Universities".

References

[1] Chen X, Shen S, Guo L, Mao SS. Semiconductor-based photocatalytic hydrogen generation. Chem Rev 2010; 110: 6503-70.

[2] Mao SS, Shen S. Hydrogen production: catalysing artificial photosynthesis. Nature Photonics 2013; 7: 944-6.

[3] Mao SS, Shen S, Guo L. Nanomaterials for renewable hydrogen production, storage and utilization. Prog Nat Sci: Mater Int 2012; 22: 522-34.

[4] Shen S, Shi J, Guo P, Guo L. Visible-light-driven photocatalytic water splitting on nanostructured semiconducting materials. Int J Nanotechnol 2011; 8: 523-91.

[5] Fujishima A, Honda K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972; 238: 37-8.

[6] Shen S, Mao SS. Nanostructure designs for effective solar-to-hydrogen conversion. Nanophotonics 2012; 1: 31-50.

[7] Sun K, Shen S, Liang Y, Burrows PE, Mao SS, Wang D. Enabling silicon for solar-fuel production. Chem Rev 2014; 114: 8662-719.

[8] Shen S. Toward efficient solar water splitting over hematite photoelectrodes. J Mater Res 2014; 29: 29-46.

[9] Kronawitter CX, Vayssieres L, Shen S, Guo L, Wheeler DA, Zhang JZ, et al. A perspective on solar-driven water splitting with all-oxide hetero-nanostructures. Energy Environ Sci 2011; 4: 3889-99.

[10] Chen X, Mao SS. Titanium dioxide nanomaterials: synthesis, properties, modifications, and applications. Chem Rev 2007; 107: 2891-959.

[11] Djurisic AB, Chen X, Leung YH, Ng AMC. ZnO nanostructures: growth, properties and applications. J Mater Chem 2012; 22: 6526-35.

[12] Oregan B, Grätzel M. A low-cost, high-efficiency solar-cell based on dye-sensitized colloidal TiO2 films. Nature 1991; 353: 737-40.

[13] Yella A, Lee HW, Tsao HN, Yi CY, Chandiran AK, Nazeeruddin MK, et al. Porphyrin-sensitized solar cells with cobalt (II/III)-based redox electrolyte exceed 12 percent efficiency. Science 2011; 334: 629-34.

[14] Hagfeldt A, Grätzel M. Molecular photovoltaics. Acc Chem Res 2000; 33: 269-77.

[15] Grätzel M. Recent advances in sensitized mesoscopic solar cells. Acc Chem Res 2009; 42: 1788-98.

[16] Zhang JZ. Metal oxide nanomaterials for solar hydrogen generation from photoelectrochemical water splitting. MRS Bull 2011; 36: 48-55.

[17] Sheng WJ, Sun B, Shi TL, Tan XH, Peng ZC, Liao GL. Quantum dot-sensitized hierarchical micro/nanowire architecture for photoelectrochemical water splitting. ACS Nano 2014; 8: 7163-9.

[18] Cho S, Jang JW, Lee KH, Lee JS. Research update: Strategies for efficient photoelectrochemical water splitting using metal oxide photoanodes. APL Mater 2014; 2:010703.

[19] Vaneski A, Schneider J, Susha AS, Rogach AL. Colloidal hybrid heterostructures based on II-VI semiconductor nanocrystals for photocatalytic hydrogen generation. J Photochem Photobio C 2014; 19: 52-61.

[20] Xie Z, Zhang Y, Liu X, Wang W, Zhan P, Li Z, et al. Visible light photoelectrochemical properties of N-doped TiO2 nanorod arrays from TiN. J Nanomater 2013; 2013: 930950.

[21] Cheng C, Sun Y. Carbon doped TiO2 nanowire arrays with improved photoelectrochemical water splitting performance. Appl Surf Sci 2012; 263: 273-6.

[22] Wang Y, Zhang YY, Tang J, Wu H, Xu M, Peng Z, et al. Simultaneous etching and doping of TiO2 nanowire arrays for enhanced photoelectrochemical performance. ACS Nano 2013; 7: 9375-83.

[23] Shankar K, Tep KC, Mor GK, Grimes CA. An electrochemical strategy to incorporate nitrogen in nanostructured TiO2 thin films: modification of bandgap and photoelectrochemical properties. J Phys D 2006; 39: 2361.

[24] Park JH, Kim S, Bard AJ, Novel carbon-doped TiO2 nanotube arrays with high aspect ratios for efficient solar water splitting. Nano Lett 2006; 6: 24-8.

[25] Ruan C, Paulose M, Varghese OK, Grimes CA, Enhanced

photoelectrochemical-response in highly ordered TiO2 nanotube-arrays anodized in boric acid containing electrolyte. Sol Energy Mater Sol Cells 2006; 90: 1283-95.

[26] Bui DN, Kang SZ, Li X, Mu J. Effect of Si doping on the photocatalytic activity and photoelectrochemical property of TiO2 nanoparticles. Catal Commun 2011; 13: 14-7.

[27] Bae JY, Yun TK, Ahn KS, Kim JH. Visible-photoresponsive nitrogen-doped mesoporous TiO2 films for photoelectrochemical cells. Bull Korean Chem Soc 2010; 31: 925-8.

[28] Gracia F, Holgado J, Caballero A, Gonzalez-Elipe A. Structural, optical, and photoelectrochemical properties of Mn+-TiO2 model thin film photocatalysts. J Phys Chem B 2004; 108: 17466-76.

[29] Wang Y, Cheng H, Hao Y, Ma J, Li W, Cai S. Preparation, characterization and photoelectrochemical behaviors of Fe(III)-doped TiO2 nanoparticles. J Mater Sci 1999; 34: 3721-9.

[30] Vinogradov AV, Vinogradov VV, Agafonov AV. A simple preparation of highly photoactive Fe(III)-doped titania nanocrystals by annealing-free approach. J Alloys Compd 2013; 581: 675-8.

[31] Xu M, Da P, Wu H, Zhao D, Zheng G. Controlled Sn-doping in TiO2 nanowire photoanodes with enhanced photoelectrochemical conversion. Nano Lett 2012; 12: 1503-8.

[32] Larsen GK, Fitzmorris R, Zhang JZ, Zhao Y. Structural, optical, and photocatalytic properties of Cr:TiO2 nanorod array fabricated by oblique angle codeposition. J Phys Chem C 2011; 115: 16892-903.

[33] Hoang S, Berglund SP, Hahn NT, Bard AJ, Mullins CB. Enhancing visible light photo-oxidation of water with TiO2 nanowire arrays via cotreatment with H2 and NH3: synergistic effects between Ti3+ and N. J Am Chem Soc 2012; 134: 3659-62.

[34] Hoang S, Guo S, Mullins CB. Coincorporation of N and Ta into TiO2 nanowires for visible light driven photoelectrochemical water oxidation. J Phys Chem C 2012; 116: 23283-90.

[35] Cho IS, Lee CH, Feng Y, Logar M, Rao PM, Cai L, et al. Codoping titanium dioxide nanowires with tungsten and carbon for enhanced photoelectrochemical performance. Nature Commun 2013; 4: 1723.

[36] Lin YG, Hsu YK, Chen YC, Chen LC, Chen SY, Chen KH. Visible-light-driven photocatalytic carbon-doped porous ZnO nanoarchitectures for solar water-splitting. Nanoscale 2012; 4: 6515-9.

[37] Jaramillo TF, Baeck SH, Kleiman-Shwarsctein A, Choi KS, Stucky GD, McFarland EW. Automated electrochemical synthesis and photoelectrochemical characterization of Zn1-xCoxO thin films for solar hydrogen production. J Comb Chem 2005; 7: 264-71.

[38] Sharma V, Kumar P, Shrivastava J, Solanki A, Satsangi VR, Dass S, et al. Vertically aligned nanocrystalline Cu-ZnO thin films for photoelectrochemical splitting of water. J Mater Sci 2011; 46: 3792-801.

[39] Prajapati CS, Kushwaha A, Sahay PP. Experimental investigation of spray-deposited Fe-doped ZnO nanoparticle thin films: structural, microstructural, and optical properties. J Therm Spray Technol 2013; 22:1230-41.

[40] Xie W, Li Y, Sun W, Huang J, Xie H, Zhao X. Surface modification of ZnO with Ag improves its photocatalytic efficiency and photostability. J Photochem Photobio A 2010; 216: 149-55.

[41] Yousefi M, Amiri M, Azimirad R, Moshfegh AZ. Enhanced photoelectrochemical activity of Ce doped ZnO nanocomposite thin films under visible light. J Electroanal Chem 2011; 661: 106-12.

[42] Ahn KS, Yan Y, Lee SH, Deutsch T, Turner J, Tracy CE, et al. Photoelectrochemical properties of N-incorporated ZnO films deposited by reactive RF magnetron sputtering. J Electrochem Soc 2007; 154: B956-9.

[43] Shet S, Ahn KS, Deutsch T, Wang H, Nuggehalli R, Yan Y, et al. Influence of gas ambient on the synthesis of co-doped ZnO:(Al, N) films for photoelectrochemical water splitting. J Power Sources 2010; 195: 5801-5.

[44] Yogeeswaran G, Chenthamarakshan CR, de Tacconi NR, Rajeshwar K. Cadmium-and indium-doped zinc oxide by combustion synthesis using dopant chloride precursors. J Mater Res 2006; 21: 3234-41.

[45] Shet S, Ahn KS, Nuggehalli R, Yan Y, Turner J, Al-Jassim M. Phase separation in Ga and N co-incorporated ZnO films and its effects on photo-response in photoelectrochemical water splitting. Thin Solid Films 2011; 519: 5983-7.

[46] Chen HM, Chen CK, Liu RS, Zhang L, Zhang J, Wilkinson DP. Nano-architecture and material designs for water splitting photoelectrodes. Chem Soc Rev 2012; 41: 5654-71.

[47] Sheng W, Sun B, Shi T, Tan X, Peng Z, Liao G. Quantum dot-sensitized hierarchical micro/nanowire architecture for photoelectrochemical water splitting. ACS Nano 2014; 8: 7163-9.

[48] Kronawitter CX, Ma Z, Liu D, Mao SS, Antoun BR. Engineering impurity distributions in photoelectrodes for solar water oxidation. Adv Energy Mater 2012; 2: 52-7.

[49] Shen S, Kronawitter CX, Jiang J, Guo P, Guo L, Mao SS. A ZnO/ZnO:Cr isostructural nanojunction electrode for photoelectrochemical water splitting. Nano Energy 2013; 2: 958-65.

[50] Lin Y, Xu Y, Mayer MT, Simpson ZI, McMahon G, Zhou S, et al. Growth of p-type hematite by atomic layer deposition and its utilization for improved solar water splitting. J Am Chem Soc 2012; 134: 5508-11.

[51] Boettcher SW, Warren EL, Putnam MC, Santori EA, Turner-Evans D, Kelzenberg MD, et al. Photoelectrochemical hydrogen evolution using Si microwire arrays. J Am Chem Soc 2011; 133: 1216-9.

[52] Liu G Yin LC, Wang J, Niu P, Zhen C, Xie Y, et al. A red anatase TiO2 photocatalyst for solar energy conversion. Energy Environ Sci 2012; 5: 9603-10.

[53] Abdi FF, Han L, Smets AHM, Zeman M, Dam B, van de Krol R. Efficient solar water splitting by enhanced charge separation in a bismuth vanadate-silicon tandem photoelectrode. Nature Commun 2013; 4: 2195.

[54] Wang M, Ren F, Cai G, Liu Y, Shen S, Guo L. Activating ZnO nanorod photoanodes in visible light by Cu ion implantation. Nano Res 2014; 7: 353-64.

[55] Chen J, Yu M, Wang Y, Shen S, Wang M, Guo L. Au@SiO2 core/shell nanoparticle-decorated TiO2 nanorod arrays for enhanced photoelectrochemical water splitting. Chin Sci Bull 2014; 59: 2191-8.

[56] Ren F, Jiang C, Liu C, Wang J, Oku T. Controlling the morphology of Ag nanoclusters by ion implantation to different doses and subsequent annealing. Phys Rev Lett 2006; 97: 165501.

[57] Cai L, Ren F, Wang M, Cai G, Chen Y, Liu Y, et al. V ions implanted ZnO nanorod arrays for photoelectrochemical water splitting under visible light. Int J Hydrogen Energy 2015; 40: 1394-401.

[58] Yang X, Wolcott A, Wang G, Sobo A, Fitzmorris RC, Qian F, et al. Nitrogen-doped ZnO nanowire arrays for photoelectrochemical water splitting. Nano Lett 2009; 9: 2331-6.

[59] Asahi R, Morikawa T, Ohwaki T, Aoki K, Taga Y. Visible-light photocatalysis in nitrogen-doped titanium oxides. Science 2001; 293: 269-71.

[60] Wang Z, Hou J, Yang C, Jiao S, Huang K, Zhu H. Hierarchical metastable Y-TaON hollow structures for efficient visible-light water splitting. Energy Environ Sci 2013; 6: 2134-44.

[61] Ahn KS, Yan Y, Al-Jassim M. Band gap narrowing of ZnO:N films by varying rf sputtering power in O2/N2 mixtures. J Vac Sci Technol B 2007; 25: L23-6.

[62] Paluselli D, Marsen B, Miller EL, Rocheleau RE. Nitrogen doping of reactively sputtered tungsten oxide films semiconductor devices, materials, and processing. Electrochem Solid-State Lett 2005; 8: G301-3.

[63] Shen S. Efficient photoelectrochemical water splitting in visible light. SPIE Newsroom 2014; DOI: 10.1117/2.1201408.005579.

[64] Shen S, Kronawitter CX, Jiang J, Mao SS, Guo L. Surface tuning for promoted charge transfer in hematite nanorod arrays as water-splitting photoanodes. Nano Res 2012; 5: 327-36.

[65] Shen S, Jiang J, Guo P, Kronawitter CX, Mao SS, Guo L. Effect of Cr doping on the photoelectrochemical performance of hematite nanorod photoanodes. Nano Energy 2012; 1: 732-41.

[66] Shen S, Zhou J, Dong CL, Hu Y, Tseng EN, Guo P, et al. Surface engineered doping of hematite nanorod arrays for improved photoelectrochemical water splitting. Sci Rep 2014; 4: 6627.

[67] Kronawitter CX, Zegkinoglou I, Shen S, Liao P, Cho IS, Zandi O, et al. Titanium incorporation into hematite photoelectrodes: theoretical considerations and experimental observations. Energy Environ Sci 2014; 7: 3100-21.

[68] Shen S, Guo P, Wheeler DA, Jiang J, Lindley SA, Kronawitter CX, et al. Physical and photoelectrochemical properties of Zr-doped hematite nanorod arrays. Nanoscale 2013; 5: 9867-74.

[69] Shen S, Kronawitter CX, Wheeler DA, Guo P, Lindley SA, Jiang J, et al. Physical

and photoelectrochemical characterization of Ti-doped hematite photoanodes

prepared by solution growth. J Mater Chem A 2013; 1: 14498-506.

Dr. Shaohua Shen obtained his Ph.D. degree in Thermal Engineering from Xi'an liaotong University in 2010. During 2008-2009 and 2011-2012, he worked as a guest researcher at Lawrence Berkeley National Laboratory and a postdoctoral researcher at the University of California at Berkeley. He is currently a Professor at Xi'an liaotong University, China. His research interests include synthesis of nanomaterials and development of devices for photocatalytic and photoelectrochemical solar energy conversion.

Professor Shaohua SHEN, Xi'an Jiaotong University Email: shshen_xjtu@mail.xjtu.edu.cn

Figure caption page

Fig. 1. Band structure change induced by metal ions doping and nonmetal ions (taking N as the example of nonmetal dopant) doping in TiO2. CB: conduction band, VB: valence band. Fig. 2. (a) A schematic of idealized operating mechanisms overlayed onto the tip of an individual nanostructure, (b) Incident photon conversion effi ciency at visible wavelengths for ZnO:Al/ZnO:Ni homojunction array (blue squares), ZnO:Ni thin fi lm (red circles), and ZnO:Al nanorod array (black triangles), with + 1 V applied vs. a Pt counter electrode, (c) Idealized energetics of the functional homojunction nanostructure. Reprinted with permission from ref. [48]. Fig. 3. (a) Schematic of operating mechanisms of charge transfer within the ZnO/ZnO:Cr structures under visible light illumination, (b) Time-dependent photocurrents (I-t curve) of ZnO/ZnO:Cr structure at an applied voltage of +1.0 V vs. Pt counter electrode under simulated solar illumination (100 mW cm 2) with and without 510 nm cut-off filter, (c) Proposed operating mechanism of the ZnO/ZnO:Cr structure as a photoanode for PEC water splitting. Reprinted with permission from Ref. [49].

Fig. 4. Band diagram schematic of the BiVO4 samples. (a) 1% W-doped BiVO4 (W-BiVO4), (b) W:BiVO4 homojunction, (c) W:BiVO4 reverse homojunction and (d) gradient-doped W:BiVO4. In all cases, the light enters from the right-hand side (through the electrolyte), and the FTO back contact is situated on the left. The space charge region at the semiconductor/electrolyte interface is not depicted for clarity. Reprinted with permission from Ref. [53].

Fig. 5. Schematic diagram of ion implantation method (left) and the process of Cu ion implanted into ZnO substrate (right). Reprinted with permission from Ref. [54].

Fig. 6. The proposed microstructure and energy band diagram of (a) Cu ions and (b) V ions implanted ZnO nanorod arrays. Reprinted with permission from Refs. [54] and [57]. Fig. 7. Photocurrent plots of (a) Cu ion and (b) V ion doped ZnO nanorod arrays under visible light (X > 420 nm). The obtained Cu ion doped ZnO nanorod arrays with different implantation doses (3^10 , 5* 1015, and 2*10 ions/cm2) were named as Cu/ZnO-1, Cu/ZnO-2 and Cu/ZnO-3, respectively. The obtained V ions doped ZnO nanorod arrays with implantation fluence of 2.5*1015 ions/cm2 were named as V/ZnO-4. Reprinted with permission from Refs. [54] and [57]. Fig. 8. UV-Vis optical absorptance of (a) pure ZnO and TiO2, (b) Cu-, (c) Fe-, and (d) W-implanted ZnO and TiO2.

Fig. 9. Photocurrent-potential (I-V) curves of (a) pure ZnO and TiO2, (b) Cu-, (c) Fe-, and (d) W-implanted ZnO and TiO2.

Fig. 10. (a) UV-Vis absorption spectra and (b) photocurrent-potential (I-V) curves of pure and N ion implanted ZnO nanorod arrays.

Fig. 11. Schematics of (a) gradient distribution of N dopants in N implanted ZnO nanorod arrays and (b) charge transfer and separation processes in N gradient-doped ZnO nanorod arrays. Reprinted with permission from Ref. [63].

, Ti 3d

Vis f lN2p

□ VB

Figure 1

Photon Energy |eVj 3.3 3 2.7 2.4

ZnO:AI C^f \ '^p-'ZnOiNi To External Circuit

360 390 420 450 480 510 540 Wavelength [nmj

3.37 eV

H20 Pt

Figure 2

Externat circuit

Bandgap transition

VB ■

hv >1V

VisiNe change transfer transiïon

Cr3d (C

OH-(Cl )

ZnO:Cr

Figure 3

BiVO, BlV°4

200 nm

100 nm 100 nm

100 nm 100 nm

W-BiVO,

200 nm

Figure 4

Figure 5

• • •> : \'À

• ••

• • •

Conduction band

■ VIS

0 <6 ev I t/U

Valence band

Cu ioti implanted ZnO nanorod

-1 0 1 2 3

Figure 6

Conduction Band

-i—Zn,+

Valence Band

V ions implanted ZnO nanorod

Figure 7

Figure 8

Applied potential! V vs. Ag/AgCI Applied potential / V vs. Ag/AgCI

Applied potential / V vs. Ag/AgCI Applied potential / V vs. Ag/AgCI

Figure 9

Wavelength / nm

E 300 o

m a 0) T3

r h" r

-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2

Applied potential /V

Figure 10

N gradient distribution N gradient-doped ZnO

Figure 11