Self-modification of titanium dioxide materials by Ti³⁺ and/or oxygen vacancies: new insights into defect chemistry of metal oxides

Abstract

The defect chemistry of metal oxides is a very important research aspect of inorganic solid-state materials. This is because (i) a certain amount of defects or imperfections are always present in metal oxide materials; (ii) the presence of defects affects, and even sometimes determines, the physical and chemical properties of the materials; and (iii) more importantly, defects do not necessarily have adverse effects on the properties of materials, and judicious “defect engineering” can bring about improved properties desired in material systems, and even some new useful functionalities that are not available to the “perfect” material. In this review, we specially highlight the recent research efforts toward understanding the defect chemistry of titanium dioxide (also known as titania, TiO₂), a widely-studied multifunctional metal oxide. In the discussion, particular attention is paid to the synthesis of Ti³⁺/oxygen vacancy self-modified TiO₂ materials and the favorable effects of these defects on the materials' properties and applications. This review, focusing on a representative metal oxide (i.e., TiO₂), is anticipated to provide some new insights into the general defect chemistry of metal oxides, and to give impetus to the development of the “defect engineering” of metal oxide materials.

---

Juan Su

Juan Su completed each of her academic degrees at the Department of Chemistry of Jilin University from 2004 to 2013, receiving her PhD in 2013 with Professor Jie-Sheng Chen on porous functional semiconductor materials. In 2013, she joined Prof. Kai-Xue Wang group as a postdoctor at the school of chemistry and chemical engineering of Shanghai Jiao Tong University in Shanghai. Her current scientific interest is mainly focused on the controllable preparation of porous reduced TiO2 materials and the microstructure-regulating of their functions for energy and environmental science.

Xiaoxin Zou

Xiaoxin Zou was awarded a PhD in Inorganic Chemistry from Jilin University (China) in 06/2011, and then moved to the University of California, Riverside and Rutgers, The State University of New Jersey as a Postdoctoral Scholar from 07/2011 to 10/2013. He is currently an associate professor at Jilin University. His research interests focus on the design and synthesis of noble metal-free, nanostructured and/or nanoporous materials for water splitting and renewable energy applications.

Jie-Sheng Chen

Jie-Sheng Chen received his BSc (1983) and MSc (1986) degrees from Sun Yat-sen University in Guangzhou and his PhD degree (1989) from Jilin University. He worked briefly from 1989 to 1990 as a lecturer at Jilin University until moving to the Royal Institution of Great Britain in London as a postdoctor. In 1994 he returned to Jilin University and was promoted to a full professor later on. In 2008, he joined the faculty of School of Chemistry and Chemical Engineering at Shanghai Jiao Tong University. His research interests are focused on the interfacial interactions of material components and the electrochemical and catalytic performances of solid materials.

---

1. Introduction

Titanium (Ti) is the ninth most abundant element (0.63%) in the Earth's crust, and it is only exceeded in abundance by O, Si, Al, Fe, Ca, Na, K and Mg.¹ As the most important oxide of titanium, titanium dioxide (also known as titania, TiO₂), which mainly exists in three polymorphs (i.e., anatase, rutile and brookite), is a multifunctional metal oxide material.^2–8 Since the early twentieth century, TiO₂ has been commercially used as a white pigment, sunscreen additive etc. These conventional applications of TiO₂ mainly benefit from its basic physical and chemical properties, such as a high refractive index, strong UV-light absorbing capabilities, excellent chemical stability and abundance.^2–8

In 1972, Fujishima and Honda discovered that the water splitting took place on a TiO₂ electrode under ultraviolet (UV) light irradiation.⁹ This pioneering work immediately evoked great interest among chemical researchers, and simultaneously enormous effort was devoted to the research of TiO₂ materials.^1–8 This has resulted in many promising TiO₂-based applications, ranging from photovoltaics, photocatalysis and self-cleaning techniques to sensors and photo-/electrochromics.^1–15 TiO₂ is generally the core component in these applications, and the properties of TiO₂ mainly determine the efficiency of these applications, and the operating circumstances that we finally use. Thus, judiciously tuning the structure of TiO₂ to optimize its properties/functions and further understanding the structure–property/function correlations has been actively pursued.

The main structure parameters of TiO₂, which are strongly related to its properties/functions, typically include the crystal phase, crystallinity, shape, size, surface structure and defects.^16–20 While many strategies have been developed to optimize the properties/functions of TiO₂ by tuning its crystal phase, crystallinity, shape, size, and/or surface structure, the defect tuning of TiO₂ remained elusive for a long time. However, breakthroughs have recently been made in the synthesis and applications of TiO₂ materials with a large amount of defects (i.e., Ti³⁺ and oxygen vacancies), and specifically the favorable effects of these defects on materials' properties has attracted wide attention.

In this review, the defects in TiO₂ are particularly designated as Ti³⁺ ions and oxygen vacancies, and the methods that are used to create these defects in TiO₂ are described as “self-modification”. The most important characteristic of the self-modification method is that the improvement of the properties of TiO₂ originates from the deliberately-introduced defects in it. In other words, through modifying the atomic structures of TiO₂ by deliberately introducing defects (e.g., removing or rearranging titanium/oxygen atoms), the properties of TiO₂ can be enhanced to a great extent. Herein, we summarize recent developments in the synthesis of self-modified TiO₂ materials with Ti³⁺ ions and/or oxygen vacancies, and the favorable effects of these defects on the materials' properties and applications.

2. Synthetic strategies for self-modified TiO₂

The synthetic strategies to self-modify TiO₂ can be roughly divided into the categories “partial reduction method” and “partial oxidation method”. The former is more frequently used to prepare self-modified TiO₂ materials.

X-ray photoelectron spectroscopy (XPS) and electron paramagnetic resonance spectroscopy (EPR) are generally applied to study the Ti³⁺ ions and oxygen vacancies in TiO₂. XPS is an effective technique to determine the presence and amount of Ti³⁺ ions in TiO₂, but it does not give any information about oxygen vacancies. In addition, because the detection depth of the XPS measurements is <10 nm, they can only provide the structural information of Ti³⁺ species at the surface and subsurface of the TiO₂ particles. As a complementary technique of XPS, EPR can give some valuable information about Ti³⁺ species and single-electron-trapped oxygen vacancies (V_o˙), and can quantitatively analyse Ti³⁺ and V_o˙ using 2-diphenyl-1-picrylhydrazyl (DPPH) and a manganese (Mn) marker as the standards.

It is worth noting that the relationships between Ti³⁺ and oxygen vacancies are very complex in a solid TiO₂ material. There are mainly three situations: (1) when the electric charges of Ti³⁺ species in TiO₂ can be totally balanced by oxygen vacancies, and Ti³⁺ and oxygen vacancies should appear/disappear simultaneously; (2) besides Ti³⁺ and oxygen vacancies, a certain amount of structural defects (mainly referring to the local structural rearrangements in TiO₂) are present in the TiO₂ materials. This will lead to inequalities between Ti³⁺ and oxygen vacancies in the material; (3) when the electric charges of Ti³⁺ species in TiO₂ are balanced by protons, Ti³⁺ has no direct connection with the oxygen vacancies.

2.1. Partial reduction method

The partial reduction method starts from a Ti(IV)-containing precursor (including TiO₂), which is partially reduced by a suitable reductant under certain conditions to finally create a self-modified (or oxygen-deficient) TiO₂ material. The specific means that were used in previously-reported partial reduction methods are summarized as below:

(i). High-temperature hydrogenation. Typically, TiO₂ samples are treated by H₂ at elevated temperatures to form self-modified TiO₂ materials.^21–26 During the hydrogenation process, H₂ will react with the lattice oxygen, leading to the formation of oxygen vacancies in TiO₂, and simultaneously one oxygen vacancy leaves behind two excess electrons. These electrons can locate at titanium positions (i.e., forming Ti³⁺), and can also remain at the positions of oxygen vacancies (i.e., forming electron-containing oxygen vacancies). It should be pointed out that the defect types are very sensitive to their formation conditions, and different defect types might exist and coexist in the same sample. For example, H. Liu et al. showed that the H₂-treated samples obtained at < 450 °C possessed only single-electron-trapped oxygen vacancies (V_o˙), whereas the samples obtained at >450 °C possessed both V_o˙ and Ti³⁺.²¹ In another example, X. Yu et al. showed that the defect types and their distribution between the surface and bulk strongly depended on the hydrogenation temperature and time.²² Longer hydrogenation at 600–700 °C induced the attenuation of Ti³⁺, and the increase of O⁻ species (Fig. 1). The reason behind this phenomenon proposed by the authors is that bulk Ti³⁺ defects might diffuse to the surface and react with surface oxygen vacancies and absorbed oxygen molecules during longer hydrogenation, finally resulting in the formation of O⁻ species.

Fig. 1 Digital images and the electron paramagnetic resonance (EPR) spectra of H₂-treated TiO₂ samples with different reaction times.²² Reprinted with permission from ref. 22. Copyright 2013 American Chemical Society.

(ii). Reduction of Ti⁴⁺ by other reductants. In addition to H₂, other reductants, including metallic zinc, Al, diethylene glycol (DEG), NaBH₄ and CO, were also employed to produce self-modified TiO₂ materials.^27–32 Using the redox reaction between Zn and Ti⁴⁺ (Zn + Ti⁴⁺ → Zn²⁺ + Ti³⁺), Zheng et al. prepared stable Ti³⁺ self-doped TiO₂ with tunable phase composition.²⁷ In addition, Sayed and co-workers reported that TiO₂ was refluxed at 220 °C in DEG (a reducing agent and solvent) to form Ti³⁺-contianing TiO₂ materials.²⁹ Kang et al. demonstrated that TiO₂ nanotube arrays were reduced by NaBH₄, a strong reducing agent.³⁰ The reaction was proposed as follows:

NaBH₄ + 8OH⁻ → NaBO₂ +8e⁻ + 6H₂O

Ti⁴⁺ + e⁻ → Ti³⁺

In another study, a combustion method was developed by Feng’s group to synthesize self-modified TiO₂ materials using in situ generated reducing gases (CO and NO) as the reductant. For instance, when a mixture containing ethanol, hydrochloric acid, titanium(IV) isopropoxide and ethylimidazole was introduced into a preheated oven (500–600 °C), bulk Ti³⁺ self-doped TiO₂ was directly obtained.³¹ Using a similar combustion method, but replacing titanium(IV) isopropoxide with porous amorphous TiO₂, the same group successfully prepared stable titania with a large number of V_o˙ defects (Fig. 2).³²

Fig. 2 Schematic representation of the synthesis of V_o˙–TiO₂ from porous amorphous TiO₂ in the presence of imidazole and HCl at 450 °C. The combustion of imidazole in the presence of HCl can release reducing gases such as CO and NO.³² Copyright 2013 WILEY-VCH Verlag GmbH & Co. KGaA Weinheim.

(iii). Hydro(solvo)thermal synthesis. The hydro(solvo)thermal technique was reported to synthesize self-modified TiO₂ materials by different groups.^33,34 Cheng’s group synthesized Ti³⁺-doped anatase TiO₂ with {001} active facets by using TiB₂ as the precursor in a HF-containing hydrothermal system (Fig. 3),³³ whereas Zhao’s group synthesized Ti³⁺-doped anatase mesocrystals with {001} and {101} active facets by using titanium isopropoxide as the titanium source and formic acid as the solvent.³⁴ Although the formation mechanism is still not clear in the above reaction systems, the hydro(solvo)thermal technique has proved to be very advantageous in combining defect engineering and facet control.

Fig. 3 Optical photo (A), typical SEM and TEM images (B and C), SAED patterns (D) and high-resolution TEM image (E) of oxygen-deficient anatase TiO₂ sheets.³³ Reprinted with permission from ref. 33. Copyright 2009 American Chemical Society.

(iv). Plasma treatment, vacuum activation and e-beam irradiation. In low-temperature plasma, high-energy electrons and atoms are available to reduce the surface/subsurface layer of TiO₂ materials, finally leading to the formation of self-modified TiO₂ materials.^35–38 This method has been employed by some Japanese researchers, and TiO₂ materials with V_o˙ defects, for example, were obtained by them.^35,36 By combining plasma and electrolysis, Zhang et al. synthesized oxygen-deficient TiO₂ microspheres that exhibited optical absorption covering the range from ultraviolet to infra-red.³⁷ In addition, vacuum treatment of TiO₂ at elevated temperatures can also produce oxygen-deficient TiO₂ materials due to the release of lattice oxygen atoms under vacuum conditions (TiO₂ → TiO_2−x + O₂);³⁹ and the e-beam irradiation method was reported to produce Ti³⁺–TiO₂ because of the surface reduction by electrons.⁴⁰

(v). Photochemical synthesis. The photocatalysis phenomenon was discovered in 1972.⁹ However, only in recent years has the photochemical technique been employed to synthesize Ti³⁺ self-modified TiO₂ materials.^41–46 This is because (1) in common TiO₂ materials the amount of photogenerated Ti³⁺ species is rather limited, and (2) the photogenerated Ti³⁺ was usually considered as an intermediate state in a photocatalytic system, and thereby its modification effects on the properties of TiO₂ were often ignored. To date, there are three main types of TiO₂ materials that can be efficiently modified by photogenerated Ti³⁺. They are porous amorphous TiO₂, TiO₂ gels and amorphous TiO₂ nanoparticles. Let us take the porous amorphous TiO₂ material as an example.^41,42 This porous material possesses an ultra-large BET surface area of ∼530 m² g⁻¹. With UV irradiation, the color of the porous TiO₂ turned from white to intense blue under the protection of inert gas (Fig. 4A), which is indicative of the presence of a large number of Ti³⁺ species in the porous TiO₂. The presence of Ti³⁺ in the porous TiO₂ was confirmed by EPR spectroscopy (Fig. 4B).⁴¹ No paramagnetic signals were observed for porous amorphous TiO₂ before UV irradiation, whereas after UV-irradiation, an intense signal at g = 1.925 was displayed. This EPR signal is ascribed to surface Ti³⁺. In addition, the amount of photogenerated Ti³⁺ can be further increased by the introduction of a dopant (e.g. V⁴⁺) into porous TiO₂.

Fig. 4 Digital images (A) and EPR spectra (B) of porous amorphous TiO₂ before and after UV irradiation; (C) schematic representation of the photochemical synthesis (step 1 and step 2) of Ti³⁺-containing TiO₂, and (step 3) the oxidation of Ti³⁺ in TiO₂ by an oxidant, such as O₂. TiO_s and ROH denote the surface oxygen atoms of TiO₂ and the hole scavenger, respectively. In this scheme, photogenerated electrons are present in the form of Ti³⁺, and Ti³⁺ storage and oxidation are proton-coupled processes.⁴¹ Reproduced by permission of The Royal Society of Chemistry.

The photogenerated Ti³⁺ in TiO₂ is fundamentally different from those formed through other routes. The differences mainly include: (1) the photogenerated Ti³⁺ ions only exist on the TiO₂ surface; (2) they are sensitive to oxidants such as O₂; (3) their electric charges are balanced by the surface protons, rather than the oxygen vacancies; and (4) their generation and consumption are proton-coupled processes. Because of these differences, the photochemically-synthesized Ti³⁺–TiO₂ showed many unique properties and functions (see below).

Fig. 4C shows a schematic representation of the photochemical synthesis (step 1 and step 2) of Ti³⁺–TiO₂; and (step 3) the oxidation of Ti³⁺ in TiO₂ by an oxidant such as O₂. When incident light possessing energy greater than the band gap of TiO₂ (i.e., UV light) hits the material, electrons in the valence band are excited to the conduction band, concomitantly leaving holes in the valance band. The photo-generated electrons and holes will allow reduction and oxidation reactions, respectively, to occur on the TiO₂ surface. In a typical photochemical system for the synthesis of Ti³⁺–TiO₂, a proper hole scavenger (e.g., ethanol) but no electron scavenger is needed in the reaction system. In this case, the photogenerated holes on the valence band react with the hole scavenger, and the electrons are comfortably stored in the form of Ti³⁺ (step 1 and 2). When the photogenerated Ti³⁺ contacts with a proper oxidant, such as O₂, it will be consumed rapidly (step 3). It should be emphasized that the Ti³⁺ storage and oxidation are proton-coupled processes.

2.2. Partial oxidation method

Although the +4 oxidation state dominates titanium chemistry, compounds in the +2 and +3 oxidation states are also common. The Ti-based compounds with low oxidation states include TiH₂, TiO, Ti₂O₃ and TiCl₃. Considering that it is possible to obtain Ti³⁺ self-doped TiO₂ materials by the partial oxidation of these compounds under appropriate conditions, researchers have been exploring synthetic approaches for translating this idea into reality.^47–51

Recently, Grabstanowicz et al. reported the synthesis of Ti³⁺ self-doped rutile TiO₂ using TiH₂ as the precursor.⁴⁷ As shown in Fig. 5, they first used a 30% aqueous solution of H₂O₂ to oxidize grey TiH₂ powder, yielding a yellow or green gel-like product.⁴⁷ H₂O₂ was selected as the oxidation agent because (1) TiH₂ is stable in air and water, but very reactive with H₂O₂ at room temperature, and (2) the only byproduct of H₂O₂ is H₂O, avoiding any unnecessary contamination. Finally, the dried gel was thermally treated at 630 °C in Ar to form Ti³⁺ self-doped rutile TiO₂. Furthermore, Liu et al. showed that rice-shaped Ti³⁺ self-doped anatase nanoparticles were synthesized by the direct hydrothermal treatment of TiH₂ in H₂O₂ aqueous solution (Fig. 6).⁴⁸ From the above results, it can be concluded that even when using the same precursor (i.e., TiH₂) and oxidation agent (H₂O₂), different thermal treatment methods can lead to different products (e.g., rutile phase obtained by the former method, but anatase phase obtained by the latter method).

Fig. 5 Scheme of the synthesis of Ti³⁺–TiO₂ from TiH₂. Grey TiH₂ powder was first converted to a yellowish gel after reaction with H₂O₂. The gel was further heated to yield the black Ti³⁺–TiO₂ powder at high temperatures.⁴⁷ Reprinted with permission from ref. 47. Copyright 2013 American Chemical Society.

Fig. 6 (A) Schematic of the formation mechanisms for the Ti³⁺ self-doped anatase TiO₂ nanoparticles, and (B and C) the interface diffusion–redox diagram. The green arrows indicate ion diffusion.⁴⁸ Reproduced by permission of The Royal Society of Chemistry.

In addition to TiH₂, TiO, Ti₂O₃ and Ti were also proved to be efficient precursors for the synthesis of self-modified TiO₂ materials.^49–51 Pei et al. demonstrated the synthesis of a self-doped TiO₂ material with Ti³⁺ and V_o˙ defects by the hydrothermal treatment of TiO in HCl solution.⁴⁹ They claimed that in the present reaction system, Ti³⁺ species can be generated via the reaction between TiO and HCl (TiO + HCl → Ti³⁺ + H₂ + H₂O). Liu et al. showed that the direct thermal oxidation of a Ti₂O₃ and TiO₂ mixture in air at 900 °C could also produce the Ti³⁺ self-doped TiO₂ material.⁵⁰ In another study, Zuo et al. synthesized Ti³⁺-doped rutile TiO₂ with {110} active facets by using titanium powder as the precursor in a HCl-containing hydrothermal system.⁵¹

3. Properties and functions of self-modified TiO₂

Defects in TiO₂ have often been considered to have only negative influence on the properties of materials. For example, the structural defects in TiO₂ were frequently referred by researchers to function as combination sites of photogenerated changes that lead to a low photocatalytic activity. However, recent studies have shown that proper defects can improve the properties of materials in some cases, and sometimes even result into some new useful functionalities that are not available to the “perfect” material. Herein, we summarize the favorable effects of these defects on materials' properties and applications.

3.1. Conductivity

In 2000, Diebold’s group investigated the influence of Ti³⁺-related bulk defects on the properties of TiO₂ (110) single crystals.⁵² In their study, TiO₂ (110) cubes were cut from the same crystal and were heated in an ultrahigh vacuum to produce Ti³⁺-related bulk defects in TiO₂. As shown in Fig. 7, five rutile crystals with different colors were prepared. The lighter crystal possessed a lower amount of bulk defects, and the amount of bulk defects increased in the order of cube 2 < cube 5 < cube 1 < cube 4 < cube 3. Table 1 shows the resistivity of the five cubes at 300 K. The results show that the darker crystal with a larger amount of defects has a lower resistivity than the lighter crystal with a smaller amount of defects. For example, the resistivity of cube 2 (the lightest cube) is 1835.0 Ω cm⁻¹, whereas the resistivity of cube 3 (the darkest cube) is only 8.94 Ω cm⁻¹. The resistivity of the former is about 205 times as high as that of the latter. In another study, in 2013 Chen’s group reported a porous crystalline titania material with heavily self-doped Ti³⁺ species.⁵³ In the material obtained, 29% of the titanium species in TiO₂ are present in the form of Ti³⁺, so this material is heavily self-doped by Ti³⁺. They further measured the room temperature electrical conductivity, and the results showed that the material exhibited a conductivity of 2.7 × 10⁻³ S cm⁻¹ whereas the conductivity value of the Ti³⁺-free sample was 9.7 × 10⁻⁸ S cm⁻¹. In other words, the conductivity of Ti³⁺–TiO₂ increased by ∼5 orders of magnitude in comparison with that of the Ti³⁺-free TiO₂ sample. The increase in conductivity (or the decrease in resistivity) of TiO₂ by Ti³⁺ self-modification was considered to be because Ti³⁺ species in TiO₂ can function as efficient donors, and the electrons of which could hop to the conduction band (or adjacent Ti⁴⁺ sites).^53–57

Fig. 7 Digital images of five rutile crystals with increased amounts of bulk defects in the order of cube 2 < cube 5 < cube 1 < cube 4 < cube 3. The lighter crystal possesses a lower amount of bulk defects.⁵² Reprinted with permission from ref. 52. Copyright 2001 American Chemical Society.

Table 1 Sample resistivity (Ω cm⁻¹)⁵²

	Cube 2	Cube 5	Cube 1	Cube 4	Cube 3
Resistivity	1835.0	108.24	46.76	24.06	8.94

---

3.2. Magnetism

Dilute magnetic semiconductors (DMSs), in particular the ferromagnetic oxides, have received great attention because of their potential applications in spintronics.⁵⁸ The study of DMSs was significantly accelerated by the discovery of room temperature ferromagnetism in Co-doped TiO₂ film.⁵⁸ With the development of DMS studies, researchers surprisingly found that some defect-containing undoped TiO₂ materials (without introducing magnetic ions in the materials) also exhibited room temperature ferromagnetism although TiO₂ itself is intrinsically non-magnetic.

In 2006, Yoon et al. observed that oxygen-deficient anatase TiO₂ film showed high-temperature ferromagnetism with a Curie temperature up to 880 K.⁵⁹ In this study, the oxygen defects were located at the interface between the anatase TiO₂ film and the (100) oriented lanthanum aluminate (LaAlO₃) substrates because of the presence of an atomic mismatch at the interface. The authors also showed that the interfacial oxygen defects in the anatase TiO₂ film on a (100) LaAlO₃ substrate are the origin of the observed magnetism. Furthermore, the authors proposed that the oxygen vacancy-related Ti³⁺ and Ti²⁺ species might provide magnetic moments because of their 3d¹ and 3d² electronic configurations respectively, although no evidence was provided to support the presence of Ti³⁺ and Ti²⁺ in the material. However, in a theoretical study by Kim et al., the ferromagnetism in undoped TiO₂ was thought be the result of the charge redistribution owing to the oxygen vacancy-induced lattice distortion.⁶⁰

More recently, several different groups experimentally showed that Ti³⁺ ions with one unpaired 3d electron (3d¹) provided the local magnetic moments in the undoped TiO₂ materials. Zhou’s group prepared Ti³⁺ self-doped rutile TiO₂ single crystals by the oxygen ion irradiation method, and observed ferromagnetism in the obtained material.⁶¹ In addition, Chen’s group prepared porous amorphous TiO₂ with a large amount of Ti³⁺ by a photochemical method.⁴¹ The investigation into the magnetic properties of Ti³⁺–TiO₂ revealed the co-existence of ferromagnetism and paramagnetism in the material. When the Ti³⁺ in TiO₂ was completely consumed by air, the as-obtained Ti³⁺-free TiO₂ turned diamagnetic. This observation strongly confirmed that the magnetism of Ti³⁺–TiO₂ was an intrinsic feature associated with the Ti³⁺ species. Furthermore, Santara and co-workers reported on the oxygen vacancy-induced ferromagnetism above room temperature in undoped TiO₂ nanoporous nanoribbons (Fig. 8).⁶² The observed magnetism was explained by the s–d interaction between the 1s¹ electron spin in V_o˙ and the 3d¹ electron spin of the Ti³⁺ ions. However, it should be noted that the exact mechanism of the magnetism for undoped TiO₂ is still not very clear at present. Thus, further experimental and theoretical studies on this unusual magnetic phenomenon are strongly encouraged.

Fig. 8 Temperature-dependent magnetization (M–T) curve of oxygen-deficient TiO₂ nanobelts showing a ferromagnetic to paramagnetic transition at ∼800 K.⁶² Reproduced by permission of The Royal Society of Chemistry.

3.3. Sensing application

TiO₂, a wide band gap semiconductor, is a well-established oxide material that has been employed as a promising candidate in sensing applications. Previously, surface modification with noble metal particles or other metal oxides was often attempted to improve the sensing properties of TiO₂.³ Recently, Chen’s group showed the importance of Ti³⁺ defects in TiO₂-based sensors for the first time, providing a new strategy to optimize the sensing properties of TiO₂. In their study, thermally stable, Ti³⁺ self-modified TiO₂ nanomaterials were demonstrated to exhibit enhanced sensitivity and ultrafast response/recovery (<3 s) for the detection of various organic vapors, such as ethanol, methanol and acetone.⁶³ The enhanced gas-sensing performance of Ti³⁺ self-modified TiO₂ materials could be related to the Ti³⁺-induced enhancement of oxygen absorption on the TiO₂ surface. Because the Ti³⁺ species in these nanomaterials were deeply buried in the bulk of the TiO₂ particles, the Ti³⁺ effect on increasing the surface reaction activity of TiO₂ is limited. This thereby led to the requirement for a high operating temperature (300 °C) in sensing applications. In view of this, the same group reported another porous crystalline titania material with heavily self-doped Ti³⁺ species.⁵³ This Ti³⁺ self-doped TiO₂ material contained a considerable proportion of Ti³⁺ in the subsurface region of the titania particles, and thus served as an efficient room temperature gas-sensing material for specific CO detection with fast response/recovery. The self-dopant (Ti³⁺) in the titania material was proved to possess dual functions: (1) decreasing the resistance of TiO₂ significantly, and (2) increasing the chemisorbed oxygen species substantially.

As shown in Fig. 9, the sensing mechanism of Ti³⁺ self-doped TiO₂ was discussed by Chen et al.,⁵³ based on the interaction of surface chemisorbed oxygen and CO gas at room temperature. When the TiO₂ sensor is exposed to air, the oxygen molecules in air are chemisorbed on the TiO₂ surface and then extract electrons from the TiO₂. This will result in a potential barrier and thus a high-resistance state. On the other hand, when the sensor is exposed to CO gas, the latter reacts with the surface oxygen species, reducing the amount of surface adsorbed oxygen. This finally leads to a decreased potential barrier and thus a low-resistance state. For such a surface reaction, the increase in the amount of surface-adsorbed oxygen should be beneficial to the reaction activity/kinetics for oxidation of the CO gas on the TiO₂ surface. In addition, the decrease of the electrical resistance not only makes it feasible to detect the resistance (and its variation) for a room temperature sensor but also enables rapid electron transport between the Au electrodes of the sensor. Therefore, the simultaneous accomplishment of high oxygen adsorption and low electrical resistance by Ti³⁺ self-doping played a crucial role in the room temperature sensing performance of the Ti³⁺–TiO₂ material.

Fig. 9 Schematic representation of the sensing mechanism of Ti³⁺-containing TiO₂. Resistance ↓ means resistance decrease. Percolation path describes the electron transport pathway between the Au electrodes of the sensor.⁵³ Reprinted with permission from ref. 53. Copyright 2013 American Chemical Society.

In another study, Pan et al. showed that Ti³⁺ self-doped rutile was used for the direct and ultrafast detection of H₂O₂.⁶⁴ This detection was based on the fast color change from blue to yellow when Ti³⁺ self-doped rutile was exposed to H₂O₂. The proposed chemical mechanism is shown by the equation:

2Ti³⁺ (blue) + 3H₂O₂ + 2H₂O → 2H₂TiO₄ (yellow) + 6H⁺

3.4. Photocatalytic hydrogen evolution and degradation of organic compounds

In 1972, Fujishima and Honda discovered the photoinduced water splitting at TiO₂ electrodes.⁹ Since then, photocatalytic H₂ production over TiO₂ materials has long been considered as a promising approach for solar energy exploitation, especially because H₂ is attractive as a clean fuel. However, the widespread use of TiO₂ is limited by its wide band gap energy, which causes the catalyst to exploit only a very small proportion (about 3–5%) of solar radiation. Recently, the self-modification strategy was demonstrated to be one of effective methods to shift the photo-responsive range of TiO₂ to the visible spectral region.

The first attempt at visible light-driven photocatalytic H₂ evolution over self-modified TiO₂ was reported by Feng’s group.³¹ Through a facile combustion method, Feng et al. prepared Ti³⁺ self-doped TiO₂, which possessed high stability in air and water under irradiation, and visible light activity for H₂ evolution from a methanol–water mixture (Fig. 10). Because the combustion process was very harsh, this method only yielded irregularly-shaped products. To overcome this shortcoming of the combustion method, the same group developed a hydrothermal method to grow Ti³⁺ self-doped rutile TiO₂ crystals with highly active facets.⁵¹ This material showed enhanced photocatalytic activity relative to the irregular nanoparticles prepared by a combustion method, which was attributed to the exposure of active {110} facets on the material. In another study, this group also reported a dopant-free, visible light-responsive TiO₂ photocatalyst with V_o˙ defects. This material exhibited not only satisfactory thermal and photo-stability, but also superior photocatalytic activity for H₂ evolution (115 μmol h⁻¹ g⁻¹) from water with methanol as a sacrificial reagent under visible light (λ > 400 nm) irradiation.³² After Feng's pioneering work, there were several reports on new methods to synthesize the self-modified TiO₂ materials (e.g., vacuum activity, and selective oxidation by TiH₂ or TiO) and their visible light activity for H₂ evolution.^{28,39,48,49,65}

Fig. 10 Time courses of evolved H₂ under visible light (>400 nm) illumination over Ti³⁺ self-doped TiO₂ synthesized by a combustion method.³¹ Reprinted with permission from ref. 31. Copyright 2010 American Chemical Society.

In addition to photocatalytic H₂ evolution, the photocatalytic degradation of organic compounds by self-modified TiO₂ materials was also reported for potential environmental remediation using visible light.^{21,22,25,27,28,38,39,47,48,50,65,66} In these studies, a range of organic compounds, including benzoic acid, 2-propanol, methylene blue (MA) and methyl orange (MO), were decomposed into CO₂ and H₂O under visible light. For example, Liu et al. demonstrated that the Ti³⁺ self-doped TiO₂ becomes an efficient visible light active photocatalyst for the degradation of 2-propanol by the grafting of Cu(II) oxide amorphous nanoclusters.⁵⁰ The results showed that the presence of Ti³⁺ did not significantly narrow the band gap, but led to the formation of isolated states between the forbidden gap. These isolated states have various electric levels from 0.3 to 0.8 eV below the conduction band minimum, resulting in the broad visible light absorption of the Ti³⁺ self-doped TiO₂ materials. On the other hand, Cu(II) oxide nanoclusters functioned as co-catalysts to suppress the recombination of photogenerated charges.

3.5. Other photocatalytic reactions

Zhao and co-workers studied the photocatalytic reduction and oxidation of nitrosobenzene over Ti³⁺ self-doped TiO₂ mesocrystals.³⁴ They found that the synergetic effect between Ti³⁺ and the (101)/(001) facet ratio are responsible for the higher activity of the TiO₂ mesocrystals for both the oxidation and reduction of nitrosobenzene. The authors claimed that the (101) facet was intrinsically more active than the (001) facet, and the existence of Ti³⁺ defects could shift the valence band maximum downwards and facilitate the generation of strongly reductive electrons. Note that in their study, no visible light photocatalytic activity was observed.

However, in another study, Mul et al. showed that Ti³⁺ self-doped titania with a band gap of 2.93 eV served as a photocatalyst for the liquid phase selective oxidation of methylcyclohexane under both UV and visible light irradiation.⁶⁷ The activity of Ti³⁺ self-doped titania surpassed those of the commercial titania catalysts, such as P25 TiO₂. In addition, Nakamura et al. demonstrated that self-modified TiO₂ with V_o˙ defects possessed photocatalytic activity for NO oxidation in the visible light region up to 600 nm.^35,36 The oxidation of NO under visible light proceeded consecutively as NO → NO₂⁻ → NO₃⁻. The visible light activity of V_o˙-modified TiO₂ was attributed to the presence of an oxygen vacancy state between the valence and conduction bands in the TiO₂ band structure.

3.6. Photoelectrochemical water splitting

In 2011, Li and co-workers demonstrated for the first time that hydrogen treatment on TiO₂ nanowires and nanotubes created a high density of oxygen vacancies, and thus significantly improved the performance of TiO₂ materials for photoelectrochemical water splitting (Fig. 11).²⁶ In comparison to pristine TiO₂ nanowires, the TiO₂ samples after hydrogen treatment showed a significantly enhanced photocurrent in the entire potential window, and low photocurrent saturation potentials of −0.6 V vs. Ag/AgCl (0.4 V vs. RHE). The optimized TiO₂ nanowire sample gave a photocurrent density of ∼1.97 mA cm⁻² at −0.6 V vs. Ag/AgCl under the illumination of simulated solar light (100 mW cm⁻² from a 150 W xenon lamp), and the solar-to-hydrogen efficiency was calculated to be around 1.1%. Furthermore, the incident photon-to-current conversion efficiency (IPCE) spectra showed that the photocurrent enhancement by hydrogen treatment was mainly due to the improved photoactivity of TiO₂ in the UV region (Fig. 11).

Fig. 11 IPCE spectra of pristine TiO₂ and several TiO₂ nanowires prepared by hydrogen treatment at 350, 400 and 450 °C.²⁶ Reprinted with permission from ref. 26. Copyright 2011 American Chemical Society.

To enhance the visible light water splitting performance of TiO₂ nanowires, C. B. Mullins and coworkers showed hydrogenation and nitridation cotreatment of TiO₂ nanowires to be a simple and effective method. This cotreatment led to the codoping of nitrogen and Ti³⁺ in TiO₂ nanowires.²³ The synergistic effect between the N-dopant and Ti³⁺ was believed to be the key to the extension of the active spectrum and the superior visible light water splitting activity. In other studies,^30,68 Ti³⁺ self-doped TiO₂ nanotubes were prepared by either chemical reduction or electrochemical reduction, and they all exhibited enhanced water splitting performance, regardless of the method used.

3.7. Proton-coupled electron transfer agents

Although it is well-known that Ti³⁺ self-modified TiO₂ materials (Ti³⁺–TiO₂) can be synthesized by a photochemical method (as mentioned before), it is little-known that the Ti³⁺–TiO₂ materials obtained by such a method can serve as proton-coupled electron transfer agents. In 2010, Zou et al. developed a photochemical approach for the preparation of porous TiO₂ with a large number of Ti³⁺ species.⁴¹ When studying the interaction of the Ti³⁺ in TiO₂ with nitrobenzene, they found that the TiO₂ with Ti³⁺ species supplied not only electrons (i.e., Ti³⁺), but also protons (H⁺) for the transformation of nitrobenzene into aniline. They realized that when an electron (in Ti³⁺) was released to an electron acceptor such as nitrobenzene, a proton (H⁺) on the surface of TiO₂ was also removed at the same time. However, in their paper,⁴¹ the concept “proton-coupled electron transfer (PCET)” was not proposed. In 2012, Schrauben et al. showed that photochemically reduced (or Ti³⁺ self-modified) TiO₂ nanoparticles in solution transferred an electron and a proton to phenoxyl and nitroxyl radicals, indicating that e⁻ and H⁺ were coupled in this interfacial reaction.⁴⁴ The concept of PCET was suggested by the authors, and correspondingly the Ti³⁺–TiO₂ materials synthesized by a photochemical method could serve as proton-coupled electron transfer agents. The above observations had important implications for the understanding and development of chemical energy technologies, such as photocatalysis, that were usually previously considered as electron transfer processes, rather than proton-coupled electron transfer processes.

In a recent study, Su et al. reported vanadium-doped porous TiO₂ with highly active hydrogen (V–TiO₂(H*)) prepared by a photochemical method.⁴² The vanadium doping could increase the amount of active hydrogen in the obtained TiO₂ material. The active hydrogen (H*) in this material is not molecular hydrogen, but a proton with an electron (i.e., Ti³⁺) in its vicinity. Furthermore, the authors used the as-obtained TiO₂ materials as proton-coupled electron transfer agents for the chemoselective hydrogenation of nitroarenes. The results showed that the TiO₂ materials can instantly (<10 s) and selectively reduce nitroarenes to aminoarenes under ambient conditions. During the reduction process, the V–TiO₂(H*) provided rich H* species (i.e., coupled e⁻ and H⁺), which could be regenerated through UV light irradiation of the reactant in methanol after the consumption of the H* species (Fig. 12).

Fig. 12 Schematic diagram for the selective hydrogenation of nitroarenes to aminoarenes by V–TiO₂(H*), and the regeneration process of V–TiO₂(H*). The active hydrogen (H*) in our material is not molecular hydrogen but a proton with an electron in its vicinity.⁴² Reproduced by permission of The Royal Society of Chemistry.

4. Conclusions

In this article, we reviewed recent developments in the synthesis of self-modified TiO₂ materials with Ti³⁺ ions and/or oxygen vacancies, and some favorable effects of these defects on the properties and applications of TiO₂ materials. Despite the enormous strides and many achievements made in the field of TiO₂ defect chemistry, as outlined in this review, there are still a lot of intractable difficulties ahead, such as the identification of the nature and local microstructure of the Ti³⁺/oxygen defects. In addition, exploring new applications related to Ti³⁺/oxygen defects and understanding the functions of these defects still remain challenging. This therefore calls for more research efforts in the general defect chemistry of metal oxides, a burgeoning and fascinating area of chemistry.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (91022019, 21331004) and the National Basic Research Program of China (2011CB808703, 2013CB934102).

Notes and references

1. O. Carp, C. L. Huisman and A. Reller, Prog. Solid State Chem., 2004, 32, 33.
2. H. Chen, C. E. Nanayakkara and V. H. Grassian, Chem. Rev., 2012, 112, 5919.
3. X. Chen and S. S. Mao, Chem. Rev., 2007, 107, 2891.
4. T. L. Thompson and J. T. Yates Jr, Chem. Rev., 2006, 106, 4428.
5. A. L. Linsebigler, G. Lu and J. T. Yates Jr, Chem. Rev., 1995, 95, 735.
6. S. G. Kumar and L. G. Devi, J. Phys. Chem. A, 2011, 115, 13211.
7. T. Tachikawa, M. Fujitsuka and T. Majima, J. Phys. Chem. C, 2007, 111, 5259.
8. M. Ni, M. K. H. Leung, D. Y. C. Leung and K. Sumathy, Renewable Sustainable Energy Rev., 2007, 11, 401.
9. A. Fujishima and K. Honda, Nature, 1972, 238, 37.
10. C. D. Valentin, G. Pacchioni and A. Selloni, J. Phys. Chem. C, 2009, 113, 20543; X.-X. Zou, G.-D. Li, J. Zhao, J. Su, X. Wei, K.-X. Wang, Y.-N. Wang and J.-S. Chen, Int. J. Photoenergy, 2012, 720183; C. S. Chen, T. C. Chen, C. C. Chen, Y. T. Lai, J. H. You, T. M. Chou, C. H. Chen and J. F. Lee, Langmuir, 2012, 28, 9996; K. Suriye, P. Praserthdam and B. Jongsomjit, Ind. Eng. Chem. Res., 2005, 44, 6599; L. B. Xiong, J. L. Li, B. Yang and Y. Yu, J. Nanomater., 2012, 831524; Y. B. He, O. Dulub, H. Z. Cheng, A. Sellini and U. Diebold, Phys. Rev. Lett., 2009, 102, 106105; M. Aizawa, Y. Morikawa, Y. Namai, H. Morikawa and Y. Iwasawa, J. Phys. Chem. B, 2005, 109, 18831.
11. J. Chen, L. B. Lin and F. Q. Jing, J. Phys. Chem. Solids, 2001, 62, 1257; E. Finazzi, C. D. Valentin, G. Pacchioni and A. Selloni, J. Chem. Phys., 2008, 129, 154113; J. Zhao, X. X. Zou, J. Su, P. P. Wang, L. J. Zhou and G. D. Li, Dalton Trans., 2013, 42, 4365; N. A. Deskins, R. Rousseau and M. Dupuis, J. Phys. Chem. C, 2011, 115, 7562; Z. Lin, A. Orlov, R. M. Lambert and M. C. Payne, J. Phys. Chem. B, 2005, 109, 20948; X. Zou, Z. Tao and T. Asefa, Curr. Org. Chem., 2013, 17, 1274; E. Lira, S. Wendt, P. Huo, J. Hansen, R. Streber, S. Porsgaard, Y. Wei, R. Bechstein, E. Lægsgaad and F. Besenbacher, J. Am. Chem. Soc., 2011, 133, 6529; N. G. Petrik, Z. Zhang, Y. Du, Z. Dohnálek, I. Lyubinetsky and G. A. Kimmel, J. Phys. Chem. C, 2009, 113, 12407; R. Schaub, E. Wahlström, A. Rønnau, E. Lægsgaard, I. Stensgaard and F. Besenbacher, Science, 2003, 299, 377; M. Setvin, U. Aschauer, P. Scheiber, Y. F. Li, W. Hou, M. Schmid, A. Selloni and U. Diebold, Science, 2013, 341, 988; S. Livraghi, S. Maurelli, M. C. Paganini, M. Chiesa and E. Giamello, Angew. Chem., Int. Ed., 2011, 50, 8038.
12. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269.
13. D. V. Bavykin, J. M. Friedrich and F. C. Walsh, Adv. Mater., 2006, 18, 2807.
14. S. In, A. Orlov, R. Berg, F. García, S. Pedrosa-Jimenez, M. S. Tikhov, D. S. Wright and R. M. Lambert, J. Am. Chem. Soc., 2007, 129, 13790.
15. R. Niishiro, R. Konta, H. Kato, W.-J. Chun, K. Asakura and A. Kudo, J. Phys. Chem. C, 2007, 111, 17420.
16. J. Su, X. Zou, G.-D. Li, Y.-M. Jiang, Y. Cao, J. Zhao and J.-S. Chen, Chem. Commun., 2013, 49, 8217; X.-X. Zou, G.-D. Li, M.-Y. Guo, X.-H. Li, D.-P. Liu, J. Su and J.-S. Chen, Chem. – Eur. J., 2008, 14, 11123.
17. A. Testino, I. R. Bellobono, V. Buscaglia, C. Canevali, M. D'Arienzo, S. Polizzi, R. Scotti and F. Morazzoni, J. Am. Chem. Soc., 2007, 129, 3564.
18. J. Pan, G. Liu, G. Q. Lu and H. M. Cheng, Angew. Chem., Int. Ed., 2011, 50, 2133.
19. H. G. Yang, C. H. Sun, S. Z. Qiao, J. Zou, G. Liu, S. C. Smith, H. M. Cheng and G. Q. Lu, Nature, 2008, 453, 638.
20. X. Zou, R. Silva, X. Huang, J. F. Al-Sharab and T. Asefa, Chem. Commun., 2013, 49, 382.
21. H. Liu, H. T. Ma, X. Z. Li, M. Wu and X. H. Bao, Chemosphere, 2003, 50, 39.
22. X. Yu, B. Kim and Y. K. Kim, ACS Catal., 2013, 3, 2479.
23. S. Hoang, S. P. Berglund, N. T. Hahn, A. J. Bard and C. B. Mullins, J. Am. Chem. Soc., 2012, 134, 3659.
24. A. Naldoni, M. Allieta, S. Santangelo, M. Marelli, F. Fabbri, S. Cappelli, C. L. Bianchi, R. Psaro and V. D. Santo, J. Am. Chem. Soc., 2012, 134, 7600.
25. W. Wang, C. Lu, Y. Ni, M. Su and Z. Xu, Appl. Catal., B, 2012, 127, 28.
26. G. Wang, H. Wang, Y. Ling, Y. Tang, X. Yang, R. C. Fitzmorris, C. Wang, J. Z. Zhang and Y. Li, Nano Lett., 2011, 11, 3026.
27. Z. Zheng, B. Huang, X. Meng, J. Wang, S. Wang, Z. Lou, Z. Wang, X. Qin, X. Zhang and Y. Dai, Chem. Commun., 2013, 49, 868.
28. C. Yang, Z. Wang, T. Lin, H. Yin, X. Lu, D. Wang, T. Xu, C. Zheng, J. Lin, F. Huang, X. Xie and M. Jiang, J. Am. Chem. Soc., 2013, 135, 17831.
29. F. N. Sayed, O. D. Jayakumar, R. M. Kadam, S. R. Bharadwaj, L. Kienle, U. Schürmann, S. Kaps, R. Adelung, J. P. Mittal and A. K. Tyagi, J. Phys. Chem. C, 2012, 116, 12462.
30. Q. Kang, J. Cao, Y. Zhang, L. Liu, H. Xu and J. Ye, J. Mater. Chem. A, 2013, 1, 5766.
31. F. Zuo, L. Wang, T. Wu, Z. Zhang, D. Borchardt and P. Feng, J. Am. Chem. Soc., 2010, 132, 11856.
32. X. Zou, J. Liu, J. Su, F. Zuo, J. Chen and P. Feng, Chem. – Eur. J., 2013, 19, 2866.
33. G. Liu, H. G. Yang, X. Wang, L. Cheng, H. Lu, L. Wang, G. Q. Lu and H.-M. Cheng, J. Phys. Chem. C, 2009, 113, 21784.
34. Q. Chen, W. Ma, C. Chen, H. Ji and J. Zhao, Chem. – Eur. J., 2012, 18, 12584.
35. I. Nakamura, N. Negishi, S. Kutsuna, T. Ihara, S. Sugihara and K. Takeuchi, J. Mol. Catal. A: Chem., 2000, 161, 205.
36. I. Nakamura, S. Sugihara and K. Takeuchi, Chem. Lett., 2000, 29, 1276.
37. Z. K. Zhang, M. L. Bai, D. Z. Guo, S. M. Hou and G. M. Zhang, Chem. Commun., 2011, 47, 8439.
38. T. Ihara and M. Miyoshi, J. Mater. Sci., 2001, 36, 4201.
39. M. Xing, J. Zhang, F. Chen and B. Tian, Chem. Commun., 2011, 47, 4947.
40. S. Kim, W. J. Jo, D. Lee, S. H. Baeck, J. H. Shin and B. C. Lee, Bull. Korean Chem. Soc., 2013, 34, 1397.
41. X.-X. Zou, G.-D. Li, K.-X. Wang, L. Li, J. Su and J.-S. Chen, Chem. Commun., 2010, 46, 2112.
42. J. Su, X.-X. Zou, G.-D. Li, L. Li, J. Zhao and J.-S. Chen, Chem. Commun., 2012, 48, 9032.
43. X.-X. Zou, G.-D. Li, Y.-N. Wang, J. Zhao, C. Yan, M.-L. Guo, L. Li and J.-S. Chen, Chem. Commun., 2011, 47, 1066.
44. J. N. Schrauben, R. Hayoun, C. N. Valdez, M. Braten, L. Fradley and J. M. Mayer, Science, 2012, 336, 1298.
45. A. I. Kuznetsov, O. Kameneva, A. Alexandrov, N. Bityurin, K. Chhor and A. Kanaev, J. Phys. Chem. B, 2006, 110, 435.
46. A. I. Kuznetsov, O. Kameneva, L. Rozes, C. Sanchez, N. Bityurin and A. Kanaev, Chem. Phys. Lett., 2006, 429, 523.
47. L. R. Grabstanowicz, S. Gao, T. Li, R. M. Rickard, T. Rajh, D.-J. Liu and T. Xu, Inorg. Chem., 2013, 52, 3884.
48. X. Liu, S. Gao, H. Xu, Z. Lou, W. Wang, B. Huang and Y. Dai, Nanoscale, 2013, 5, 1870.
49. Z. Pei, L. Ding, H. Lin, S. Weng, Z. Zheng, Y. Hou and P. Liu, J. Mater. Chem. A, 2013, 1, 10099.
50. M. Liu, X. Qiu, M. Miyauchi and K. Hashimoto, Chem. Mater., 2011, 23, 5282.
51. F. Zuo, K. Bozhilov, R. J. Dillon, L. Wang, P. Smith, X. Zhao, C. Bardeen and P. Feng, Angew. Chem., Int. Ed., 2012, 51, 6223.
52. M. Li, W. Hebenstreit, U. Diebold, A. M. Tyryshkin, M. K. Bowman, G. G. Dunham and M. A. Henderson, J. Phys. Chem. B, 2000, 104, 4944.
53. J. Su, X.-X. Zou, Y.-C. Zou, G.-D. Li, P.-P. Wang and J.-S. Chen, Inorg. Chem., 2013, 52, 5924.
54. E. Yagi, R. R. Hasiguti and M. Aono, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 7945.
55. N. A. Deskins and M. Dupuis, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 75, 195212.
56. J. Tang, W. Wang, G.-L. Zhao and Q. Li, J. Phys.: Condens. Matter, 2009, 21, 205703.
57. V. M. Khomenko, K. Langer, H. Rager and A. Fett, Phys. Chem. Miner., 1998, 25, 338.
58. Y. Matsumoto, M. Murakami, T. Shono, T. Hasegawa, T. Fukumura, M. Kawasaki, P. Ahmet, T. Chikyow, S. Koshihara and H. Koinuma, Science, 2001, 291, 854.
59. S. D. Yoon, Y. Chen, A. Yang, T. L. Goodrich, X. Zuo, D. A. Arena, K. Ziemer, C. Vittoria and V. G. Harris, J. Phys.: Condens. Matter, 2006, 18, L355.
60. D. Kim, J. Hong, Y. R. Park and K. J. Kim, J. Phys.: Condens. Matter, 2009, 21, 195405.
61. S. Zhou, E. Čižmár, K. Potzger, M. Krause, G. Talut, M. Helm, J. Fassbender, S. A. Zvyagin, J. Wosnitza and H. Schmidt, Phys. Rev. B: Condens. Matter Mater. Phys., 2009, 79, 113201.
62. B. Santara, P. K. Giri, K. Imakita and M. Fujii, Nanoscale, 2013, 5, 5476.
63. X. Zou, G. Li, Y. Zou, P. Wang, J. Su, J. Zhao, Y. Wang and J. Chen, Acta Chim. Sin., 2012, 70, 1477.
64. S. S. Pan, W. Lu, Y. Zhao, W. Tong, M. Li, L. Jin, J. Y. Choi, F. Qi, S. Chen, L. Fei and S.-F. Yu, ACS Appl. Mater. Interfaces, 2013, 5, 12784.
65. J. Wang, P. Zhang, X. Li, J. Zhu and H. Li, Appl. Catal., B, 2013, 134–135, 198.
66. I. Justicia, P. Ordejón, G. Canto, J. L. Mozos, J. Fraxedas, G. A. Battiston, R. Gerbasi and A. Figueras, Adv. Mater., 2002, 14, 1399.
67. M. S. Hamdy, R. Amrollahi and G. Mul, ACS Catal., 2012, 2, 2641.
68. Z. Zhang, M. N. Hedhili, H. Zhu and P. Wang, Phys. Chem. Chem. Phys., 2013, 15, 15637.

---