Phase B vanadium dioxide: characteristics, synthesis and applications

Yujing Zhang , Nan Chen , Yang Zhou *, Haojie Lai , Pengyi Liu and Weiguang Xie *
Guangdong Provincial Engineering Technology Research Center of Vacuum Coating Technologies and New Energy Materials, Jinan University, Guangzhou 510632, People's Republic of China. E-mail: yangzhou@email.jnu.edu.cn; wgxie@email.jnu.edu.cn

First published on 29th November 2021

---

---

Yujing Zhang

Yujing Zhang received his Bachelor's degree from Jinan University in 2019. Currently, he continues to study for a Master's degree at Jinan University under the supervision of Prof. Weiguang Xie. His current research focuses on vanadium dioxide prepared by the hydrothermal method and its applications in uncooled infrared detection and terahertz detection.

Nan Chen

Nan Chen is an undergraduate in the Department of Physics, Jinan University, Guangzhou, Guangdong, China. Her current research focuses on the growth mechanism and preparation of controlled vanadium oxide nanomaterials and their applications in optoelectronic properties.

Yang Zhou

Dr. Yang Zhou received his BSc degree in materials science from University of Science and Technology of China in 2011 and received his PhD degree from School of Materials Science and Engineering at Nanyang Technological University in 2016. After two-year postdoctoral research at Nanyang Technological University, he is currently working at Jinan University. His research interest is focused on optoelectrical and piezoelectric/ferroelectric materials.

Haojie Lai

Haojie Lai received his Master's degree from Jinan University in 2020. Currently, he continues to study for a doctoral degree at Jinan University under the supervision of Prof. Weiguang Xie. His current research focuses on the photoelectric detection of two-dimensional materials.

Pengyi Liu

Prof. Pengyi Liu received his PhD from Sun Yat-sen University in 2005. He is a professor of the Department of Physics, and Vice Dean of School of Science and Technology in Jinan University. He mainly engages in vacuum technology and thin film materials research, including functional thin film technology, photoelectric materials and devices, and solar cells. Prof. Liu has published over 50 research papers, and filed 4 patents. He has won the second prize of the Science and Technology Award of the China Association for Instrumental Analysis in 2020.

Weiguang Xie

Prof. Weiguang Xie received his PhD from Sun Yat-sen University in 2008. He is now a professor at Jinan University. His main research interests cover the optoelectrical properties and application of van der Waals materials and perovskite materials. Prof. Xie has published over 90 research papers, and filed 8 patents. He has won the first prize of the Science and Technology Award of the Chinese Association for Analysis and Testing in 2016, and the second prize of Science Technology Improvement Award at a higher institution in China in 2014.

---

1. Introduction

Vanadium (V), which has the ground state electronic configuration of [Ar]3d³4s², is the fifth most abundant transition metal on Earth. Its content accounts for about 0.02% of the earth's crust. V exists in numerous possible valence states; thus, it has many different oxides.¹ Vanadium oxides such as VO, V₂O₃, VO₂ and V₂O₅ exist in a single oxidation state, whereas many others, for instance, V₃O₅, V₄O₇, V₆O₁₁, V₆O₁₃, V₇O₁₃, and V₈O₁₅, remain in mixed (two) valence states. Among them, V₂O₃, VO₂ and V₂O₅ have been widely studied as optical, electrical, electrochemical, thermochromic and thermal-switching materials.² VO₂, in particular, has received great attention due to its multiple structure phases and unique optical and electrical properties.³ So far, several VO₂ crystal phases have been reported, including tetragonal VO₂(R) (P4₂/mnm),⁴ monoclinic VO₂(M) (P2₁/c),^5,6 tetragonal VO₂(A) (P4₂/nmc),⁷ single oblique VO₂(B) (C2/m)⁸ and triclinic VO₂(T) (P*(2)).⁹

Among the various phase states of VO₂, VO₂(M) and VO₂(R) have attracted much attention in applications such as smart windows and photoelectric conversion switches due to their reversible metal–insulator transition between the two phases. VO₂(A) is an unstable intermediate phase. It usually occurs between the phase transition process from VO₂(B) to VO₂(R).^7,10 VO₂(B) is a material with a metastable monoclinic structure. It has been applied in many fields, including energy conversion and storage,^11–13 uncooled infrared detectors,^14–16 as well as gas and humidity sensors.^17–19

In the current study of VO₂(B), the most popular and widely reported applications are metal ion batteries and uncooled infrared detectors. At room temperature, VO₂(B) is in the metallic phase, which is required for superior electrical transport. The tunnel structure of VO₂(B) can be utilized as a large diffusion channel for the rapid intercalation and deintercalation of metal ions,^20,21 enabling ions to intercalate and de-intercalate in a reversible metal ion battery.^22,23 For uncooled infrared detectors, VO₂(B) has no obvious phase change at room temperature, no sudden changes in electrical and optical parameters, and no thermal hysteresis. The 1/f noise is low and its TCR is high. These advantages are beneficial as a thermal sensitive material for uncooled infrared detectors.

In recent years, gratifying progress has been made in the research on VO₂(B). Several reviews^8,24,25 have mentioned VO₂(B) and recognized the achievements of VO₂(B) in the corresponding fields. However, these reviews just mentioned part of the features of VO₂(B) rather than introducing it systematically or in detail. Therefore, in this review, we aim at making a detailed summary of VO₂(B). The first chapter gives an overview on the structural and opto-electrical properties of VO₂(B). The second chapter will specifically introduce the major preparation methods of VO₂(B), mainly to control the structure and properties. The advantages and disadvantages of different methods will be discussed. The third chapter will outline the popular applications of VO₂(B) at present. The fourth chapter will summarize and look forward to the future of VO₂(B). We hope that this review can provide a reference for those who are working on the application of VO₂(B), and also make a little contribution to the fundamental research on oxide materials.

2. The structure and properties of VO₂(B)

VO₂(B) is a metastable monoclinic phase of VO₂ belonging to the space group C2/m, as shown in Fig. 1a.^26–28 In 1975, Théobald et al.²⁶ reported the structure of VO₂(B) powder for the first time. At room temperature, the crystallographic parameters of VO₂(B) are a = 12.0414 Å, b = 3.6892 Å, c = 6.4299 Å, and β = 106.95°. The specific structural parameters and atomic coordinates are shown in Table 1.

---

Similar to the case of VO₂(M) and VO₂(R), where there is a metal-insulation transition at 68 °C,³¹ VO₂(B) undergoes a structural transformation below room temperature, which is accompanied by abnormal magnetic susceptibility. Through X-ray powder diffraction studies, Oka et al.³² determined that the structural transformation of the compound was caused by V⁴⁺–V⁴⁺ dimerization. The structure of VO₂(B) at high temperature (300 K) and low temperature (50 K) measured by Oka et al. are shown in Fig. 1b and c, respectively.³³ During the high-temperature to low-temperature phase transition, the monoclinic space group does not change, mainly due to the formation of non-magnetic V–V pairs between the V(2) atoms. With the increase in the paramagnetic V atoms in all high temperature phases (HTP) and low temperature phases (LTP), the V(2)–V(2) bond is similar to that in rutile VO₂, and the covalent bond formation with the decrease in the temperature leads to V⁴⁺–V⁴⁺ pairing. Specific examples are shown in Tables 2 and 3. One of the manifestations is the change in the transmittance in this temperature range, as shown in Fig. 1d and e. It can be found that, compared with the HTP, the lattice along the x- and z-axis, which matches a-axis and c-axis, respectively, changes more obviously in the LTP. Analyzing the band structures of HTP and LTP (Fig. 1f and g), it could be found that the LTP of VO₂(B) has a larger d_∥ band split, and the π* band also moves to a relatively higher energy.³⁰ In the HTP, the energy bands coincide at the Fermi level, which is the reason that VO₂(B) exhibits a certain metal-like property at room temperature.

---

---

In addition to the temperature dependence of the phase change of VO₂(B)²⁹ (Fig. 1d and e), there is also a phase change controlled by the luminous flux at a certain temperature³⁰ (Fig. 2a and b). Applying a certain amount of light influence to VO₂(B) in the LTP until the critical luminous flux threshold (F_th) is exceeded, the insulating state will change to a metallic state. The F_th value, in accordance with the experimental data of luminous flux and temperature, can draw the boundary phase (Fig. 2c).³⁰ The red- and the blue-shaded regions represent the metallic state and the insulating states, respectively. The region (I) of the red and blue streaks represents the insulation-metal transition state with a longer thermal relaxation time. The reason VO₂(B) stays in the region (I) is that the pure laser energy or low-temperature heat cannot make the system reach region (II) of metallic state. Thus, when the laser energy and system temperature rise, VO₂(B) changes to the metallic state. The width of region (I) decreases with increasing temperature, which is a manifestation of this mechanism.

According to the literature,^34–36 the V_2p peak splits into V_2p1/2 and V_2p3/2. The binding energy of the V⁵⁺_2p3/2 peak is located in the range of 516.9–517.7 eV. The binding energy of the V⁴⁺_2p3/2 peak is located in the range of 515.7–516.2 eV. The binding energy of the V³⁺_2p1/2 peak is located near 523 eV. Fig. 3a shows the core-level spectra of V_2p of the VO₂(B) center at 515.56 eV, 517.1 eV and 522.8 eV, showing that all the V atoms are at +4 states. The valence spectrum of VO₂(B) is shown in Fig. 3b. There is a large band in the range of 2–10 eV, which is mainly attributed to the O-2p orbital. At 0–2 eV, Choi et al.¹⁵ shows the existence of the occupied band for VO₂(B) at room temperature, which is primarily dominated by the V-3d orbitals. Compared with V₂O₅, it is found that the band tail of V-3d of VO₂(B) and V₆O₁₃ crosses the Fermi level, which is consistent with the metallic properties of VO₂(B) at room temperature.³⁷

Fig. 3c shows the Fourier transform infrared (FTIR) spectrum of VO₂(B). The absorption belonging to the vibration of V–O is mainly concentrated in the range of 400–1100 cm⁻¹. Among the three absorption peaks of 540, 926 and 1047 cm⁻¹, the strong absorption peak near 540 cm⁻¹ corresponds to the bending of the V–O–V octahedron, and the absorption peak of 926 cm⁻¹ is caused by the coupling vibration of VO and V–O–V. Both of the above belong to the inherent properties of VO₂(B). The cause of the weak absorption peak near 1047 cm⁻¹ is more complicated. It is caused by the stretching vibration of the VO short bond, which involves vanadium ions in the oxidation state between V⁵⁺ and V⁴⁺. These vanadium ions easily cause partial interaction due to the large specific surface area of VO₂(B) and O₂ in the atmosphere.

3. Synthesis of VO₂(B)

Since Théobald firstly discovered a new phase in vanadium dioxide in 1976,²⁶ named VO₂(B), the preparation of VO₂(B) has always been an issue of great concern. The bulk crystal of VO₂(B) is easy to fracture and can be converted to other phases owing to temperature changes.^10,28 Moreover, different application scenarios require different VO₂(B) with different microstructures. Therefore, a lot of effort has been made in the past to prepare VO₂(B).^24,40 Although obtaining pure VO₂(B) is more challenging than the other phases of VO₂,^37,41,42 after decades of development, several methods for preparing VO₂(B) have been realized, such as chemical vapor deposition (CVD), hydrothermal method, pulsed laser deposition (PLD), sol–gel method and magnetron sputtering.

The above methods for preparing VO₂(B) have their own characteristics. Some methods can produce films with high crystallinity, some methods can control the morphology of material growth, some methods can accurately control the proportions of different components, and some methods can be adapted to commercial processes. The above methods for preparing VO₂(B) will be specifically introduced and compared to illustrate their respective advantages and disadvantages.

3.1. Chemical vapor deposition

CVD is widely studied and applied in industry because it can be used to prepare homogeneous films by precisely controlling the stoichiometric ratio, phase and morphology. As early as 1966, Takei et al.⁴⁷ grew single crystal VO₂ bulks by CVD. Generally speaking, CVD utilizes a gas precursor to react in a reaction chamber and deposit the reaction product on the substrate to form a thin film. The usual complete process of CVD is shown in Fig. 4a. In most cases, due to poor crystallinity and quality, the film deposited with a powder as a crystalline nucleus is not desirable.⁴⁸ In 2017, Guo et al.⁴⁹ of Shanghai University, China, utilized a simple CVD method, which does not require high vacuum to produce thermally sensitive VO₂(B) films with suitable crystallinity, high uniformity and high performance at low cost. The vanadium(III)-2,4-pentanedionate (V(C₅H₇O₂)₃, 97%) precursor was heated at 150 °C for 2 h in the beginning and the injected high-purity argon flow rate (30, 50, 80, 120 standard cm³ min⁻¹ (sccm)), substrate temperature (350, 375, 400, 450, 500 °C), and in situ annealing time (0, 1, 3, 5 h) were adjusted, and finally, the sample was taken out of the furnace below 50 °C. After the above process, VO₂(B) films with different crystallinity and thermal sensitivity were prepared. In the experiment with controlled variables, the VO₂(B) film at 375 °C (Fig. 5a) has the lowest vis-NIR transmittance, and the VO₂(B) film at 30 sccm (Fig. 5b) and 1 h (Fig. 5c) has excellent TCR. These experiments have proved that the thermal sensitivity of the film annealed for 1 h at an argon flow rate of 30 sccm and a deposition temperature of 375 °C is relatively good, with a resistance of 11.27 kΩ, TCR of −2.87% K⁻¹, suitable crystallinity and high uniformity, as shown in Fig. 5d.

In less than a year, the same team successfully utilized CVD to dope Mo or/and Al into the VO₂(B) film.⁵⁰ The method they chose was similar to that mentioned above. Molybdenum acetylacetonate ([CH₃COCHC(O)CH₃]₂MoO₂, 99%) or/and aluminum acetylacetonate (Al(C₅H₇O₂)₃, 99%) were employed so that Mo or/and Al was doped into the VO₂(B) film. The SEM images of the final sample are shown in Fig. 6a. In general, there is no synergy between high conductivity and great TCR.⁵¹ According to the simulation results of first-principle calculations, Guo et al. found that if the co-doped elements meet two conditions – the radius is close to V and there is charge compensation, the lattice deformation of the doped VO₂(B) film could be reduced. Fig. 6b shows the V L-edge and O K-edge X-ray absorption spectra (XAS) of the samples, and the results are tabulated in Fig. 6c. It can be found that the change about π* and σ* of the Mo–Al co-doped is the slightest in all the results, compared with undoped VO₂(B). Thus, its film performance should be the best. The experimental results in Fig. 6d prove that the Mo–Al co-doped film has the best TCR, reaching −3.6% K⁻¹. Therefore, the change in the lattice deformation in the VO₂(B) film is the smallest when Mo and Al are co-doped. The ability to prepare uniform VO₂(B) thin film and superior TCR make CVD a potential method for application in uncooled infrared detector.

3.2. Hydrothermal method

Broadly speaking, hydrothermal synthesis directly crystallizes materials from aqueous solutions by controlling the thermodynamic variables (temperature, pressure, composition, etc.).⁵² Nowadays, the hydrothermal method is closely linked with nanotechnology. Due to the highly controlled diffusivity in the strong solvent medium of the closed system, the hydrothermal method has special advantages for processing nanomaterials. Compared with traditional coarse-grained materials, the mechanical strength, diffusion coefficient, specific heat and resistivity of nanomaterials are higher.⁵³ The core components for preparing the VO₂(B) film using the hydrothermal method is the autoclave reactor, as shown in Fig. 4b.

Recent studies have shown that the hydrothermal method has the advantages of low temperature (180–200 °C) preparation, friendly reaction environment, and ease of morphology and size distribution control. The autoclave facilitates the synthesis of VO₂(B) nanostructures with a controllable shape.⁵⁴ Many nanostructures have been successfully synthesized, including nanowires,⁵⁵ nanobelts,⁵⁶ nanorods,⁵⁷ nanotubes,¹⁹ and 3D nanostructures.^58,59 As shown in Fig. 7a, nanowires with a diameter of 50–60 nm and thickness of 6–10 nm were prepared by the reaction of V₂O₅, water and ethylene glycol.⁵⁵ The ultra-thin wires deliver a higher capacity to store charge than the standard wires cycled under the same conditions (200 mAh g⁻¹). Fig. 7b shows VO₂(B) nanosheets with a thickness of several nanometers and width of tens of nanometer, which were obtained by the hydrothermal reaction between NH₄VO₃ and oxalic acid.⁵⁷ As the cathode, the VO₂(B) nanosheets exhibit better initial discharge capacity and cyclability than the VO₂(B) bulk because of the high ability in overcoming the aggregation and stress to induce structural collapse of the sheet-like morphology. The stacked VO₂(B) shown in Fig. 7c is a free-standing VO₂(B) film formed by the hydrothermal reaction of V₂O₅, H₂O₂ and isopropyl alcohol on the plane (001) superposed by nanosheets.⁶⁰Fig. 7d shows that using V₂O₅ as the vanadium source and glycerin as the reducing agent, VO₂(B) nanobelts with the length of 1 μm and width of 50 nm with uniform morphology can be synthesized.⁵⁶ It has been reported that hierarchical VO₂(B) nanostructures are assembled from lower-dimensional nanostructures. For example, novel flowerlike VO₂(B) micro-nanostructures assembled from single-crystalline nanosheets were prepared using PVP as a surfactant (Fig. 7e),⁵⁸ and sea urchin-like VO₂(B) nanostructures assembled from nanobelts were prepared by the hydrothermal method (Fig. 7f).⁵⁹ This three-dimensional assembly can be promoted by increasing the concentration of the reducing agent and the reaction temperature. However, the high concentrations of the reducing agent may prevent VO₂ from crystallizing, resulting in polycrystalline formation. In order to maintain the uniformity of the crystal form, the reaction temperature or reaction time should be increased. The ability to grow different nanostructures make the hydrothermal method a popular method in the field of VO₂(B) preparation.

Tian et al.⁶¹ of Chinese Academy of Inspection and Quarantine reported an ingenious preparation method based on the traditional hydrothermal method in 2019. The traditional hydrothermal method performs the oxidation–reduction reaction in a sealed and heated solution to generate VO₂(B), while Zhang et al. transfer the reaction from inside the solution to a glass placed right above the solution, as shown in Fig. 8a. The solution is evaporated and redeposited on the glass to form the final product, which is called the hydrothermal evaporation method. After the reaction, the black product covers the surface of the glass plate. As shown in Fig. 8b, the black product is layered, uniform nanosheet of VO₂(B), which is vertically distributed on the glass surface with a thickness in the range of 3–15 nm. The hydrothermal evaporation method overcomes the shortcoming of the hydrothermal method, i.e., difficulty of forming a uniform film with large area, thus greatly expanding its application range.

3.3. Pulsed laser deposition

Pulsed laser deposition is one of the most commonly used physical vapor deposition processes. The schematic diagram of the PLD experimental device is shown in Fig. 4c. PLD technology began to develop in the late 1980s. In 1993, Borek et al.⁶² deposited directional grains of VO₂ on a sapphire substrate for the first time. PLD, which is especially suitable for oxide growth, has developed rapidly in the field of film deposition in the recent years.

VO₂(B) thin film is usually prepared at low-temperature because the phase transition from VO₂(B) to VO₂(R) occurs after 400–450 °C.^26,63 High quality thin films prepared by PLD normally required high temperature; thus, it was not until 2014 that PLD was used to prepare metastable VO₂(B) thin films. Chen et al.⁶⁴ took V₂O₅ as the target material to grow VO₂(B) films on the SrTiO₃ (STO) (001) substrate under the optimal substrate temperature of 550 °C and oxygen pressure of 10 mTorr. They found that the lattice matching between the film and the substrate was the main driving force for the preservation of the metastable phase. There are four possibilities for the growth of VO₂(B) on SrTiO₃ (STO) (001), as shown in Fig. 9b. Fig. 9d–f reveal the simulated diffraction patterns from different zones of VO₂(B). The symmetric parallelograms (red solid and dashed lines) connect the diffraction patterns from [010] VO₂(B) and [00] VO₂(B) zones, respectively. The symmetric squares (green dashed lines) connect the diffraction patterns from [100] VO₂(B) or/and [00] VO₂(B) zone axes, respectively. The small blue square corresponds to the substrate diffraction pattern.

The PLD method shows advantages such as low film pollution, fast growth rate and uniform orientation. It is possible to accurately control the growth conditions to obtain the phase, which researchers need. In 2015, Srivastava's team, National University of Singapore, controlled the laser frequency of PLD to control the rate at which vanadium reaches. They have grown monoclinic VO₂(M), epitaxial tetragonal VO₂(A) film and monoclinic VO₂(B) film on the STO(100) substrate (Fig. 10a), as well as compared their structure and transportation. They investigated the effects of oxygen pressure and laser frequency on the stability of these crystal forms. The phase diagram of the different phases of VO₂ was established (Fig. 10b). Their formation may be related to the distance between the V–V atoms caused by different oxygen pressures. At low oxygen pressures, the VO₂(R) phase is preferred at the growth temperature, which subsequently return to the M phase (Fig. 10c) at room temperature due to the metal–insulator transition (MIT). When the oxygen pressure increases, the A phase with a longer average V–V distance (Fig. 10d) may be favored. Furthermore, by lowering the laser frequency to decrease the V arrival rate, the B phase with the longest V–V distance (Fig. 10e) is stabilized. Moreover, the symmetry of the phases could also be an important factor for their stabilization. The relatively more symmetric structure of VO₂(A) over VO₂(B) suggests VO₂(A) to be more stable over a wide range of laser frequency.

3.4. Sol–gel method

The sol–gel method refers to the material synthesis method that involves the preparation of a sol, then gelling it and removing the solvent. Fig. 4d shows the general sol–gel process. Turner summarized some advantages of the sol–gel method over the traditional process: good chemical homogeneity, large surface area, high chemical purity, and various forms of products.⁶⁶

In 2011, a team⁶⁷ from the National Institute of Advanced Industrial Science and Technology in Japan obtained VO₂(B) nanofibers (Fig. 11a) with high specific surface areas by simply annealing the aerogel in high vacuum, taking the aerogel of hydrated vanadium pentoxide (V₂O₅·xH₂O) as the precursor. Since VO₂(B) has the closest double-layer structure to the aerogel precursor, the conversion of the aerogel to VO₂(B) can proceed smoothly with the lowest structural rearrangement resistance. Therefore, VO₂(B) obtained by high vacuum annealing well inherits the unique structural characteristics of the aerogel precursor, such as high surface area, mesoporous network and nanofiber morphology. These unique structures leave more active sites and larger specific surface area in VO₂(B), which is suitable as a cathode material for ion batteries. Fig. 11b shows that the discharge curve undergoes a rapid decline to 2.6 V, then a well-defined flat voltage plateau at 2.5 V, and finally a less-resolved second plateau at about 1.9 V, down to 1.5 V. They respectively correspond to the two intensive reduction peaks from the derived dQ/dV curves in Fig. 11c. The reduction branch and oxidation branch of the dQ/dV curve are in good symmetry, showing good reversibility of the electrode reaction. However, there is a problem in the above preparation process – its preparation cycle is too long, and the complete process requires nearly two weeks.

The VO₂(B) film prepared by the sol–gel method by Monfort et al.⁶⁸ from Comenius University overcame the problem of excessively long preparation time. V₂O₅ powder obtained by calcining 1 g ammonium metavanadate was dissolved in 100 mL of 15% H₂O₂ drop by drop, and V₂O₅ aqueous solution was synthesized. After all the powders were completely dissolved, the transparent orange solution containing oxo- and peroxyvanadate was heated to 70 °C, and the solution turned into a deep red gel. The V₂O₅·xH₂O hydrogel was rotated at 60 revolutions per second (RPS) for 1 min and deposited on a clean Si/SiO₂ and sapphire substrate. After deposition, the hydrogel was annealed at 400 °C for 10 min to dehydrate the film. Finally, it was annealed in a tubular oven. The whole process took no more than 12 h. As shown in Fig. 11d and e, the grain size of the film annealed in vacuum is smaller and more uniform than the film annealed in H₂/Ar. It is because H₂/Ar reacts with O₂ in the film, resulting in more particles, more micro-cracks, and a thinner film. With these advantages, VO₂(B) exhibits considerable photocatalytic activity in both UVA and visible light, as shown in Fig. 11f and g. It is worth thinking about enhancing its photocatalytic properties by doping it with another element or by increasing its surface area. Moreover, the report from Monfort et al. provides a detailed reference for the preparation of VO₂(B) films by the sol–gel method, and the specific conditions and parameters can be designed according to the actual needs of VO₂(B).

3.5. Magnetron sputtering

Fig. 4d shows the DC magnetron sputtering process. Usually, high-purity vanadium or vanadium oxide is the target, and the sputtering target is annealed to form a dense and uniform film. The magnetron sputtering method can deposit thin films at a low temperature with a high deposition speed. It is one of the mature thin film deposition techniques in the industry. Ding et al.⁶⁹ proposed an effective strategy to modify the electronic properties of VO₂(B) by inducing elastic strain through the TiO₂(A) buffer layer in magnetron sputtering. Superior vis-NIR absorption (Fig. 12a) and high TCR (−3.48% K⁻¹, Fig. 12b) was achieved. Investigation on the microscopic mechanism in Fig. 12c shows that tensile strain contributes to a reduction in the overlapping of the V-3d orbitals and an increase in the carrier concentrations along the c-axis in VO₂(B). Fig. 12d shows that the larger the tensile stress applied to the sample, the more obvious increase in the conductivity was observed. Besides interfacial stress, the temperature rise causes additional tensile stress inside the material. Both improve the electrical conductivity and TCR values of VO₂(B).

Zhang et al.⁷⁰ successfully prepared a layered W-doped VO₂(B) film by magnetron sputtering, which has better crystallinity and texture than the deposited film. It can be seen from Fig. 12e and f that the two-layer configuration of this 20 nm-thick film is formed by 12/8 nm-thick W-poor and W-rich layers, respectively. The inconspicuous boundary implies that there is a good mutual diffusion between W and VO₂(B). Besides, 8 cat% W in VO₂(B) strikes a good balance between a sufficient number of 3d electrons and the negative effect on the crystallinity by W substitution. As shown in Fig. 12g, the TCR of VW10-II is higher than that of undoped VW10 and VW10-I with larger thickness (30 nm) and obvious boundary due to larger grains with less grain-boundaries to generate lower site density of the defects in the bandgap. The above result shows that the doping can tailor the multi-layered microstructure to optimize the performance of the VO₂(B) films, as well as affects the crystallinity of the VO₂(B) phase and its degree of texture.

3.6. Comparison of various preparation methods for VO₂(B)

The five mainstream methods for preparing VO₂(B) have been introduced above. It can be found that most of these methods convert V in the vanadium source into V of +4 via an oxidation–reduction reaction. The commonly used vanadium source is V₂O₅. During the formation process, VO₂(B) needs an oxygen-rich environment.⁷¹ It could be found that the main factors affecting the morphology, particle size, microscopic morphology and performance of the VO₂(B) products are concentrated in the reaction atmosphere, reducing agent concentration, reaction temperature and time.

Table 4 summarizes the above methods for preparing VO₂(B), listing the advantages and disadvantages of each method. At present, most of studies on VO₂(B) is carried out in the form of nanomaterials. Hydrothermal and sol–gel methods are simple and more suitable for growing nanomaterials with different types of nanostructures. They are more popular to prepare VO₂(B) with a finely adjustable nanostructure for application as metal ion battery electrode. CVD, PLD and magnetron sputtering are more suitable to prepare thin film samples with high quality. In particular, magnetron sputtering is one of the most widely applied thin film deposition technologies in the industrial field, which can obtain uniform thin films with a large area. Therefore, VO₂(B) prepared by magnetron sputtering could quickly integrate into the industrial production process. These methods are more suitable for application in semiconductor devices, for example, uncooled infrared detectors. Last but not least, researchers should choose a suitable method to prepare according to the actual situation so that the obtained VO₂(B) product could meet the final requirement.

---

4. Application of VO₂(B)

The most popular and widely reported applications of VO₂(B) are metal ion batteries and uncooled infrared detectors. VO₂(B) shows large interlayer space (8.2 Å along the b-axis²⁰), and its nanostructure with a large specific surface area can also provide more active sites as the diffusion channel for the rapid intercalation and de-intercalation of ions in reversible lithium-ion batteries.^22,23 In addition, VO₂(B) also has the advantages of proper electrode potential and excellent lattice shear resistance⁷⁴ during the charging cycles. They are exactly what metal ion batteries need. Therefore, VO₂(B) is a promising cathode material in lithium-ion batteries.⁷⁵

The material requirements for the new generation of uncooled infrared detectors mainly focus on two aspects: increasing the temperature coefficient of resistance (TCR) value and reducing the noise level. Compared with VO₂(M)⁷⁶ with insulator–metal phase transition characteristics, V₂O₅⁷⁷ with high resistivity and high noise level, and VO_x with TCR affected by vacancies, VO₂(B) shows no phase transition around the room temperature. It is thermal hysteresis and 1/f noise is low, while TCR is high with suitable resistance, which is suitable for thermal sensors in uncooled infrared detectors.

4.1. Cathode for metal ion batteries

Metal ion batteries have broad application prospects in large-scale energy storage systems.^78,79 Various metal ion batteries, including lithium-ion batteries,²³ aluminum-ion batteries,⁸⁰ zinc-ion batteries,¹¹ and magnesium-ion batteries,³⁹ have requirements for cathode materials focused on the following points: capacity, rate ability, energy density, power density, and cycling performance. The capacity, rate capability, and cycle performance are obtained by fitting the experimental data, while the energy density and power density are obtained by calculation. The calculation formulas are as follows.

	(1)

	(2)

where Q denotes the discharge capacity (Ah), m is the mass of the active material (kg), I is the discharge current (A), and U is the main operating voltage (V).

As we all know, VO₂(B) is constructed from irregular VO₆ octahedra, which share corners and edges to form a tunnel-like frame. The tunnel structure is conducive for the rapid intercalation and deintercalation of partial metal cations in VO₂(B). The diffusion time (t) of ions is proportional to the square of the diffusion length (L): t ≈ L²/D (D is the diffusion coefficient);⁸¹ thus, reducing the feature size of the material is beneficial for shortening the diffusion time of ions. Therefore, when used in metal ion battery cathode material, VO₂(B) usually appears in the form of nanomaterials.

The charge–discharge cycle in water-based Zn²⁺ battery: insertion of metal cations: as shown in Fig. 13a and b, when the Zn²⁺ battery is discharged, Zn²⁺ is gradually inserted into the interlayer tunnel, which is parallel to the b or c axes.²² At this point, the lattice spacing of the (−601), (020), and (−404) planes in VO₂(B) will increase slightly, with d₋₆₀₁ = 0.202 + 0.011 nm, d₀₂₀ = 0.185 + 0.008 nm, and d₋₄₀₄ = 0.156 + 0.003 nm, respectively. The extraction of metal cations: when the Zn²⁺ battery is charged, Zn²⁺ is gradually extracted from the interlayer tunnel, and the lattice spacing of the (−601), (020) and (−404) planes will reversibly restore to the initial position.

In addition to interlayer intercalation/deintercalation behavior, Li⁺ also follow diffusion dynamics and preferential pathways.¹³Fig. 13c and d show the distribution of Li⁺ (green symbols) at a small concentration of Li (x = 0.25 in Li_xVO₂(B)). Li⁺ are mainly found at the center of the cage, i.e., the C site, which is the most stable Li adsorption site. In addition, there is a non-negligible probability to find Li⁺ at A1 and A2. When the Li⁺ concentration increases up to the maximum theoretically modeled value (x = 1.25), more accumulation of Li⁺ at A sites, especially A1, is observed (Fig. 13e and f). This result suggests that the relative site adsorption energies could vary with the Li⁺ concentration. Overall, most Li⁺ ions are found to diffuse along the b-axis, as expected due to the large open channel space (Fig. 13f), while the diffusion along the a-axis also becomes significant when the Li concentration is increased, as shown in Fig. 13e. Fig. 13g–i proves the correctness of the simulation results. Though only the heavy atom V shows a bright contrast in the HAADF image (Fig. 13g), it is able to visualize other light atoms including O and Li atoms by ABF imaging (Fig. 13h). Even after 200 charging/discharging cycles, there is no structural distortion in the discharged VO₂(B).

The pure layered nanobelt structure (Fig. 14a) allows metal cations to intercalate between the layers. However, the traditional structure will produce unavoidable volume expansion when metal cations are intercalated, resulting in strains that cannot be released in time. At this point, nanobelts tend to accumulate during the circulation process, and accumulation will hinder the circulation. If the structure is converted to a hybrid VO₂(B) nanostructure (Fig. 14b), it could provide rapid strain relaxation for the expansion of metal cations during the intercalation–deintercalation process, increase the effective electrode–electrolyte contact area, and shorten the ion's diffusion path. The lattice strain can also be adjusted along the c-axis (the direction of the co-angular VO₆ octahedron).³² The electron sharing along the [001] direction (c-axis) between the vanadium atoms may lead to stronger correlation effects between the electrons, which affects the electronic properties of VO₂(B).⁶⁹ As shown in Fig. 14c and d, compared to pure nanobelt, the hybrid nanostructure VO₂(B) cathode has better charge/discharge and cycle performance.

In the case of lithium ions storage, VO₂(B)/graphene ribbons prepared by Yang et al.⁸³ exhibit a high diffusion coefficient of ≈10⁻¹⁰ cm² s⁻¹, leading to an ultrafast rate ability of 204 mAh g⁻¹ at 37.2 A g⁻¹. One reason why VO₂(B) prepared by Yang et al. through the hydrothermal method can maintain a good rate during rapid discharge is that VO₂(B) is a nanomaterial. Another reason is that the radius of Li⁺ is small (0.76 Å). Even when storing the large-sized sodium ions, VO₂(B) is still able to tolerate the high current rates up to 93 mAh g⁻¹ at 36 A g⁻¹.⁸⁴ Chao et al. designed VO₂(B)@GQD (graphene quantum dots), which is a cathode material for sodium ion battery, with extremely high cycle stability. After 1500 cycles at 18 A g⁻¹, its capacity remained at 88%.

As for magnesium ion, which has a slightly smaller ionic radius than sodium ion, the performance of VO₂(B) is not satisfactory. When the pure VO₂(B) nanorods prepared by Luo et al.³⁹ were used as the cathodes of magnesium ion batteries, although they have good capacity and rate capabilities, the cycle performance was indeed poor. This may be due to the fact that only part of the V³⁺ (belonging to V_2p1/2) of vanadium oxide participates in the re-oxidation after losing magnesium during Mg²⁺ intercalation and deintercalation in the VO₂(B) nanorods, resulting in the degradation of the VO₂(B) performance. Therefore, when Cai et al.⁸⁶ took VO₂(B) as the cathode material of aluminum-ion batteries, the solution environment was changed to acidic because H⁺ and Al³⁺ participated in the intercalation and deintercalation, which would reduce the volume change of VO₂(B) caused by intercalation and deintercalation.

Zinc ion batteries (ZIB) may have a larger ionic radius in metal ion batteries. Because VO₂(B) nanofibers possess unique tunnel transport pathways and little structural change on Zn²⁺ intercalation, the limitation from solid-state diffusion in the vanadium dioxide electrode is eliminated. Thus, it can still play an important role in ZIB. The nanofibers prepared by Ding et al.²² have a maximum capacity of 357 mAh g⁻¹, a rate capability of 171 mAh g⁻¹ at 51.2 A g⁻¹, and a capacity retention rate of 91.2% of the initial capacity (274.1 mAh g⁻¹) after 300 cycles. If VO₂(B) is combined with a suitable material to make a composite material, its performance can be further improved. Cui et al.¹¹ reported their results in 2020. As shown in Fig. 15, the combination of VO₂(B) and reduced graphene oxide (RGO) improved the performance to a higher level – the highest capacity of 456 mAh g⁻¹, a rate capability of 292 mAh g⁻¹ at 10 A g⁻¹, and a capacity retention rate of 94.6% after 600 cycles. They also listed the comparison of the energy density of the current zinc-ion battery cathode materials (Fig. 15b).

Table 5 summarizes the performance reports of VO₂(B) as metal ion battery cathode materials in the recent years. At present, the hydrothermal method has become the most popular preparation method by virtue of its advantages in controlling the structural details of VO₂(B). One can say that the research progress of VO₂(B) in this field in the past 10 years is encouraging, and the experimental results have gradually approached that of practical products. In particular, the VO₂(B) base composite is considered as a better choice to improve the performance, reduce the costs, and make up for the shortcomings. The change in the thinking is more suitable for promoting the research and development of materials.

---

4.2. Uncooled infrared detectors

Infrared detectors are widely applied in different fields including military, security, housing construction, transportation, and household appliances, with huge market demands and sufficient development potentials. At present, infrared detectors have gotten rid of the limitations of large-scale refrigeration equipment and are booming in uncooled infrared detectors.⁸⁷ The microbolometer is the key pixel detector in the uncooled thermal imaging system.⁸⁸ The main advantage of the microbolometer technology is that it can work at room temperature and its response band is wide enough from infrared to millimeter wavelengths.⁴⁹ At room temperature, VO₂(B) has the characteristics of low 1/f noise, suitable resistivity and high TCR, which is very suitable for microbolometers. As early as 1998, Wada et al.⁸⁹ used VO₂(B) film as the thermally sensitive material to produce a 256 × 256 bolometer-type uncooled infrared detector, and finally obtained thermal images with a noise equivalent temperature difference of 0.15 K.

The key parameter for a bolometer material is the TCR that quantifies the change in the resistance as a function of a given temperature change.

	(3)

where R₀ is the resistance of the microbolometer at temperature T₀. High TCR produces an ideal high voltage response rate R_v, where the voltage response rate is⁹⁰

	(4)

where I_b is the bias current of the microbolometer, η is the light absorption coefficient of the microbolometer, G is the thermal conductivity of the microbolometer, ω is the modulation frequency, and τ is the response time. For the definition of thermal response time:

	(5)

where C is the heat capacity of the sensitive area of the pixel, and G is the thermal conductivity of the support structure.

The preparation of vanadium oxide films with suitable sheet resistance (R₀) and high TCR is the basis for the preparation of high-performance microbolometers.⁹¹ Ding et al.⁶⁹ prepared a VO₂(B) film with a TCR of −2.13% K⁻¹ and a resistance of 94.0 kΩ by magnetron sputtering; Guo et al.⁴⁹ prepared a VO₂(B) film with a TCR of −2.89% K⁻¹ and a resistance of 37.3 kΩ by CVD; Gao et al.⁶⁰ prepared a VO₂(B) film with a TCR of −3.0% K⁻¹ by the hydrothermal method. The above reports show that the potential of VO₂(B) is sufficient, and the potential contained in VO₂(B) can be realized by improving the preparation process.

Table 6 is a summary of TCR and resistance of VO₂(B) thin film that shows potential in a bolometer. Through comparison, it can be clearly found that VO₂(B) doping (W doping,⁷⁰ Mo–Al co-doping⁵⁰) or making VO₂(B) have a synergistic effect with other materials (RGO,¹⁶ TiO₂(A),⁶⁹ graphene¹⁴) can well improve its thermal performance. The higher the TCR, the more finely VO₂(B) monitors the temperature changes and the suitable resistance can effectively suppress the 1/f noise and improve the detection rate.

---

Recently, VO₂(B) has been reported to be integrated into the pipeline process of uncooled infrared detectors. Lee et al.⁹³ have fabricated highly textured and thermally stable VO₂(B) thin films on an amorphous SrTiO₃ layer using a reactive sputtering technique. The schematic fabrication process, which conforms to the standard industrialization process, is shown in Fig. 16a. It is also worth mentioning that the fabrication process used is suitable for application to monolithic infrared detectors since all fabrication processes are carried out at temperatures lower than 350 °C. The sensor arrays consist of microbolometers with a single active pixel area of 24 × 26 μm² (Fig. 16b). Fig. 16c shows that the resistivity and TCR values at 100 °C were quite stable for a month at about 8.40 × 10⁻² Ω cm and 1.24% K⁻¹, respectively, and there was no significant change in the resistivity and TCR value at this temperature. This shows the high thermal stability of the VO₂(B) film and indicates the feasibility of long-term, stable, high-temperature operation of the microbolometer without a thermoelectric cooler, which are highly encouraging to those seeking cooler-free microbolometer image sensors operating at high temperature.

VO₂(B) also shows good response performance from ultraviolet (UV) to NIR (Fig. 5a).^{15,29,49,50,69} It is because of the excitation from the V 3d band to the conduction band.^15,28,30 The photoresponsive mechanism of VO₂(B) in UV-NIR region includes the generation of free carriers and their subsequent transport. The surface state from surface defects, vacancies and dangling bonds will affect the generation and transport of free carriers. It has been reported that VO₂(B) is susceptible to the influence of O₂ and humidity in the atmosphere, thereby partially changing its electrical properties. The interaction between VO₂(B) and O₂ can be divided into two steps: the adsorption and desorption of oxygen molecules, as shown in Fig. 17a.⁹²

4.3. Gas sensors

In terms of humidity sensors, few people regarded VO₂(B) as sensing materials before 2010. It was not until Yin et al.¹⁸ found that VO₂(B) had higher sensitivity than the other semiconductor oxide humidity sensors under low humidity; Zhu et al.⁷² found that VO₂(B) nanoparticles had good sensitivity to alcohols (ethanol, isopropanol, butanol) and acetone that VO₂(B) in humidity/gas sensors was gradually carried out. However, the gas sensor based on VO₂(B) is far from being systematically studied in terms of its mechanism, which makes its development lacking of theoretical guidance. Related reports on the gas response of VO₂(B) have gradually appeared in the recent years.

In addition to the effects of water molecule adsorption mentioned above (Fig. 17c), different gas molecules act in different ways. The most stable adsorption configurations of gases on Au–VO₂ are shown in Fig. 18a–d. The bonds lengths of H–H, C–H, C–O and C–O in H₂, CH₄, CO and CO₂ molecules are 0.74, 1.14, 1.51 and 1.51 Å, respectively. CH₄-adsorbed Au–VO₂ system shows the smallest adsorption distances, which is related to the strong sensing characteristics of Au–VO₂ toward CH₄ gas, and consists of experimental observation.⁹⁴ The above results were obtained by Joy et al.⁹⁵ through density functional theory and molecular dynamics studies. They speculated that H₂, CO, and CO₂ are related to physical adsorption, and CH₄ corresponds to chemical adsorption.

For this chemical adsorption mechanism, Dey et al.⁹⁶ extended their scope of action to liquefied petroleum gas (LPG). The entire mechanism is illustrated pictorially through Fig. 18e. When the flow of LPG is stopped for recovery, the oxygen molecules in air will be adsorbed on the film, and the capture of electrons through the processes indicated in eqn (6) and (7) will increase the sensor resistance and the initial conditions get restored and the cycling reaction continues. Because the sensing mechanism is based on the chemisorption reaction that takes place at the surface of the metal oxide, increasing the specific surface area of the sensitive material leads to more sites for the adsorption of gases.

In addition to the influence of the adsorption mechanism, the nanostructure of VO₂(B) also affects the sensitivity of the gas sensors. Liang et al.⁹⁷ prepared VO₂(B) ultrathin vertical nanosheet array using a hydrothermal method. They used density functional theory to analyze the adsorption of NO₂ on the surface of VO₂(B) and found that when the thickness of nanosheets is much larger than 2λ_D (λ_D is the Debye length of about 10 nm), a large part of the nanosheet interior cannot interact with the gas. As shown in Fig. 18f, when the thickness of the nanosheet is close to 2λ_D, the space charge layer completely occupies the conductive path. At this time, the resistance of the conductive path is almost completely controlled by the space charge layer, and the gas response reaches the highest. Moreover, when the VO₂(B) nanosheets contact each other, the energy band at the grain boundary bends to form a homojunction potential barrier. When the gas sensor is placed in NO₂, the homojunction barrier becomes higher, the inversion layer becomes wider, and the depletion layer becomes narrower (Fig. 18g).

To sum up, most of the research on the mechanism of gas sensors based on VO₂(B) currently belongs to the effect of a single gas element on it. From the common points extracted from them, the chemically adsorbed gas can obtain relatively good sensing properties, which may be due to their combination with electrons on the surface of VO₂(B) so as to change the thickness of its surface depletion layer. It leads to a reduction in the resistance, thereby improving its sensing performance. This kind of universal mechanism research is exactly what researchers are pursuing, which can guide the development of research in the future and promote the development of gas sensors based on VO₂(B).

5. Summary and outlook

Starting from the structure of VO₂(B), the characteristics of VO₂(B) have been introduced in detail, including the metal–insulator transition of VO₂(B) at 100–200 K and the inherent opto-electrical properties of VO₂(B). The function and performance of the materials usually depends on the microstructure, geometry, and preparation methods. It is such fascinating features that make VO₂(B) stand out among the many vanadium oxides as a popular choice for energy conversion and storage,^11–13,25 and semiconductor applications.^14–19

Then, we focus on the methods of preparing VO₂(B) materials with different characteristics including CVD, hydrothermal method, PLD, sol–gel method and magnetron sputtering. Most of these methods convert V in the vanadium source into V with +4 valence state through an oxidation–reduction reaction. The commonly chosen vanadium source is V₂O₅. During the formation process, VO₂(B) needs an oxygen-rich environment.⁹⁸ After summary and analysis, it was found that the main factors affecting the morphology, particle size, microscopic morphology and performance of the VO₂(B) products are concentrated in the reaction atmosphere, reducing agent concentration, reaction temperature and time. Through the comparison of these five mainstream methods, it can be seen that most of the current research on VO₂(B) is carried out in the form of nanomaterials. Due to the highly controlled diffusion in the solvent medium of a closed system and the special advantages for the preparation of nanomaterials, the hydrothermal method has become a more popular technique for preparing VO₂(B). Besides, PLD and magnetron sputtering could be promising methods in the future because they are mature methods in semiconductor industry. Since highly crystallized VO₂(B) thin film has been prepared using these methods, the application progress of VO₂(B) thin film in the semiconductor industry could be accelerated.

In the current research on VO₂(B), the fields of metal ion batteries and uncooled infrared detectors are popular and have a lot of reports. At present, the hydrothermal method has become the most popular preparation of VO₂(B) as the metal ion battery cathode materials by virtue of its advantages in controlling the structural details of VO₂(B). As for uncooled infrared detectors, the selection of preparations is more diverse, which involves CVD, hydrothermal method, sol–gel method and magnetron sputtering. In the early stage of research on VO₂(B), the researchers wished to control the preparation conditions to control the structural details of VO₂(B) in order to realize the own potential of VO₂(B). In recent years, researchers have begun to develop VO₂(B)-based composite materials, which can improve the performance, reduce costs, and make up for the shortcomings by combining other suitable materials. For example, the synergistic effect of VO₂(B) and RGO¹¹ greatly improve the capacity and cycle stability of zinc ion batteries; Mo–Al co-doped VO₂(B) film exhibits unusual thermal sensitivity,⁵⁰ which is beneficial to improve the detection efficiency of uncooled infrared detectors. As for gas sensors, VO₂(B) is still catching up, and more mechanisms need to be revealed to provide theoretical guidance for its development.

However, VO₂(B) is still far from commercial applications. As a metal ion battery cathode material, the first difficulty is to achieve the industrial production of VO₂(B) with nanostructures for high capacity and rate stability of intercalation/deintercalation. Secondly, the cyclic stability of VO₂(B) needs to be continuously improved because during deep discharge, the structure of the VO₂(B) cathode will inevitably deteriorate, resulting in serious capacity reduction. In response to these two problems, the composite material is an effective strategy to prevent the dissolution occurring at the surface of the VO₂(B) cathode and to stabilize the structure. 1) Compounding with other materials (e.g., carbon fiber cloth and RGO) to form a cathode improves the structural stability; 2) pre-introducing guest species (e.g., H₂O molecules and transition metal ions) into VO₂(B) could relieve the structural deterioration, enlarge the intercalation spacing for more ion occupation and improve the electrochemical performance; 3) composite surface coating is also a considerable method but it cannot reduce the rate of metal ion intercalation/deintercalation.

For VO₂(B) applied to uncooled infrared detector, the problems are still complicated, such as the absorption of infrared radiation and the limitation of the detection range. Commercial applications also face difficulties such as process adaptation, performance stability, and market competition with existing products. For the problem of infrared absorption, multilayer films (e.g., VO₂(B)/TiO₂ film and VO₂(B) film with 5-stack Ti/MgF₂) can be used to enhance the absorption. For the problem of the detection range, different types of detectors working together is a good solution. The production process requires a dense and uniform film with a large area, which should be well adapted to the preparation process of micro-electro-mechanical system (MEMS). At present, the promising means are deposition and sputtering, which are commercially mature production processes and also suitable for coating multilayer materials. We hope that this review can provide a valuable reference for future research on VO₂(B).

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Grants No. 62174072, 61934004), Guangdong Basic and Applied Basic Research Foundation (Grants No. 2019B151502049, 2017B090907019, 2020A1515111192), the Science and Technology Planning Project of Guangzhou (Grants No. 201605030008, 202002030142), and the Fundamental Research Funds for the Central Universities (Grant No. 21619416).

References

1. Y. F. Zhang, J. C. Zhang, X. Z. Zhang, Y. Deng, Y. L. Zhong, C. Huang, X. Liu, X. H. Liu and S. B. Mo, Ceram. Int., 2013, 39, 8363–8376 CrossRef CAS .
2. Q. Tang, F. Y. Li, Z. Zhou and Z. F. Chen, J. Phys. Chem. C, 2011, 115, 11983–11990 CrossRef CAS .
3. Y. F. Zhang, J. C. Zhang, X. Z. Zhang, S. B. Mo, W. B. Wu, F. Niu, Y. L. Zhong, X. Liu, C. Huang and X. H. Liu, J. Alloys Compd., 2013, 570, 104–113 CrossRef CAS .
4. M. Li, S. Magdassi, Y. Gao and Y. Long, Small, 2017, 13, 1701147 CrossRef PubMed .
5. Y. Gao, H. Luo, Z. Zhang, L. Kang, Z. Chen, J. Du, M. Kanehira and C. Cao, Nano Energy, 2012, 1, 221–246 CrossRef CAS .
6. M. M. Seyfouri and R. Binions, Sol. Energy Mater. Sol. Cells, 2017, 159, 52–65 CrossRef CAS .
7. S. Zhang, B. Shang, J. Yang, W. Yan, S. Wei and Y. Xie, Phys. Chem. Chem. Phys., 2011, 13, 15873–15881 RSC .
8. S. Wang, K. A. Owusu, L. Mai, Y. Ke, Y. Zhou, P. Hu, S. Magdassi and Y. Long, Appl. Energy, 2018, 211, 200–217 CrossRef CAS .
9. P. Shvets, K. Maksimova and A. Goikhman, J. Appl. Phys., 2021, 129, 055302 CrossRef CAS .
10. S. Lee, I. N. Ivanov, J. K. Keum and H. N. Lee, Sci. Rep., 2016, 6, 19621 CrossRef CAS PubMed .
11. F. Cui, J. Zhao, D. Zhang, Y. Fang, F. Hu and K. Zhu, Chem. Eng. J., 2020, 390, 124118 CrossRef CAS .
12. Y. Zhang, X. Jing, Y. Cheng, T. Hu and M. Changgong, Inorg. Chem. Front., 2018, 5, 2798–2810 RSC .
13. S. Lee, X.-G. Sun, A. A. Lubimtsev, X. Gao, P. Ganesh, T. Z. Ward, G. Eres, M. F. Chisholm, S. Dai and H. N. Lee, Nano Lett., 2017, 17, 2229–2233 CrossRef CAS PubMed .
14. Y. Lu, H. Zhang and D. Wan, J. Materiomics, 2020, 6, 280–285 CrossRef .
15. S. Choi, J. Kang, S. Ryu, H. Jeen, J. Son and S. Lee, APL Mater., 2020, 8, 041111 CrossRef CAS .
16. H. Zhang, S. Zhu, B. Guo, D. Wan, H. Luo and Y. Gao, Mater. Chem. Phys., 2018, 218, 154–159 CrossRef CAS .
17. S. E. Zaki, M. A. Basyooni, M. Shaban, M. Rabia, Y. R. Eker, G. F. Attia, M. Yilmaz and A. M. Ahmed, Sens. Actuators, A, 2019, 294, 17–24 CrossRef CAS .
18. H. Yin, K. Yu, Z. Zhang, M. Zeng, L. Lou and Z. Zhu, Electroanalysis, 2011, 23, 1752–1758 CrossRef CAS .
19. G. P. Evans, M. J. Powell, I. D. Johnson, D. P. Howard, D. Bauer, J. A. Darr and I. P. Parkin, Sens. Actuators, B, 2018, 255, 1119–1129 CrossRef CAS .
20. F. Wan, L. Zhang, X. Wang, S. Bi, Z. Niu and J. Chen, Adv. Funct. Mater., 2018, 28, 1804975 CrossRef .
21. L. Qiao and M. T. Swihart, Adv. Colloid Interface Sci., 2017, 244, 199–266 CrossRef CAS PubMed .
22. J. Ding, Z. Du, L. Gu, B. Li, L. Wang, S. Wang, Y. Gong and S. Yang, Adv. Mater., 2018, 30, e1800762 CrossRef PubMed .
23. N. Li, W. Huang, Q. Shi, Y. Zhang and L. Song, Ceram. Int., 2013, 39, 6199–6206 CrossRef CAS .
24. M. Liu, B. Su, Y. Tang, X. Jiang and A. Yu, Adv. Energy Mater., 2017, 7, 1700885 CrossRef .
25. H. T. Tan, X. Rui, W. Sun, Q. Yan and T. M. Lim, Nanoscale, 2015, 7, 14595–14607 RSC .
26. F. Théobald, J. Less-Common Met., 1977, 53, 55–71 CrossRef .
27. S. R. Popuri, A. Artemenko, R. Decourt, M. Josse, U. C. Chung, D. Michau, M. Maglione, A. Villesuzanne and M. Pollet, J. Phys. Chem. C, 2015, 119, 25085–25092 CrossRef CAS .
28. A. Srivastava, H. Rotella, S. Saha, B. Pal, G. Kalon, S. Mathew, M. Motapothula, M. Dykas, P. Yang, E. Okunishi, D. D. Sarma and T. Venkatesan, APL Mater., 2015, 3, 026101 CrossRef .
29. Y. Yang, S. Long, G. Sun, Z. Shao, P. Jin and X. Cao, Jpn. J. Appl. Phys., 2018, 57, 100303 CrossRef .
30. J. Lourembam, A. Srivastava, C. La-o-vorakiat, L. Cheng, T. Venkatesan and E. E. M. Chia, Sci. Rep., 2016, 6, 25538 CrossRef CAS PubMed .
31. W. Zeng, N. Chen and W. Xie, CrystEngComm, 2020, 22, 851–869 RSC .
32. Y. Oka, T. Yao, N. Yamamoto, Y. Ueda and A. Hayashi, J. Solid State Chem., 1993, 105, 271–278 CrossRef CAS .
33. S. A. Corr, M. Grossman, Y. Shi, K. R. Heier, G. D. Stucky and R. Seshadri, J. Mater. Chem., 2009, 19, 4362 RSC .
34. G. Silversmit, D. Depla, H. Poelman, G. B. Marin and R. De Gryse, J. Electron Spectrosc. Relat. Phenom., 2004, 135, 167–175 CrossRef CAS .
35. J. Li, Z. Su, B. Yang, S. Cai, Z. Dong, J. Ma and R. Li, J. Phys. Chem. Solids, 2010, 71, 407–411 CrossRef CAS .
36. L. Liu, T. Yao, X. Tan, Q. Liu, Z. Wang, D. Shen, Z. Sun, S. Wei and Y. Xie, Small, 2012, 8, 3752–3756 CrossRef CAS PubMed .
37. C. Lamsal and N. M. Ravindra, Semiconductors, 2019, ch. 4, pp. 127–218 Search PubMed .
38. D. Wan, P. Xiong, L. Chen, S. Shi, A. Ishaq, H. Luo and Y. Gao, Appl. Surf. Sci., 2017, 397, 30–39 CrossRef CAS .
39. T. Luo, Y. Liu, H. Su, R. Xiao, L. Huang, Q. Xiang, Y. Zhou and C. Chen, Electrochim. Acta, 2018, 260, 805–813 CrossRef CAS .
40. K. Liu, S. Lee, S. Yang, O. Delaire and J. Wu, Mater. Today, 2018, 21, 875–896 CrossRef CAS .
41. L. Zhong, K. Li, Y. Luo, M. Li, H. Wang and G. Li, Cryst. Growth Des., 2017, 17, 5927–5934 CrossRef CAS .
42. W. Yu, S. Li and C. Huang, RSC Adv., 2016, 6, 7113–7120 RSC .
43. P. Palanisamy, M. Chavali, E. M. Kumar and K. C. Etika, Nanofabrication for Smart Nanosensor Applications, 2020, ch. 10, pp. 253–280 Search PubMed .
44. Z. A. Umar, N. Ahmed, R. Ahmed, M. Arshad, M. Anwar-Ul-Haq, T. Hussain and M. A. Baig, Surf. Interface Anal., 2018, 50, 297–303 CrossRef CAS .
45. S. Dhere, Curr. Sci., 2018, 115, 436–449 CrossRef CAS .
46. M. Bellardita, A. Di Paola, S. Yurdakal and L. Palmisano, Heterogeneous Photocatalysis, 2019, ch. 2, pp. 25–56 Search PubMed .
47. H. Takei and S. Koide, J. Phys. Soc. Jpn., 1966, 21, 1010 CrossRef .
48. M. E. A. Warwick and R. Binions, J. Solid State Chem., 2014, 214, 53–66 CrossRef CAS .
49. B. Guo, D. Wan, A. Ishaq, H. Luo and Y. Gao, J. Alloys Compd., 2017, 715, 129–136 CrossRef CAS .
50. B. Guo, D. Wan, J. Wang, S. Zhu, H. Luo and Y. Gao, J. Alloys Compd., 2018, 745, 247–255 CrossRef CAS .
51. A. Krammer, A. Magrez, W. A. Vitale, P. Mocny, P. Jeanneret, E. Guibert, H. J. Whitlow, A. M. Ionescu and A. Schüler, J. Appl. Phys., 2017, 122, 045304 CrossRef .
52. M. Shandilya, R. Rai and J. Singh, Adv. Appl. Ceram., 2016, 115, 354–376 CrossRef CAS .
53. V. Yamakov, D. Wolf, S. R. Phillpot, A. K. Mukherjee and H. Gleiter, Nat. Mater., 2004, 3, 43–47 CrossRef CAS PubMed .
54. X. Rui, D. Sim, C. Xu, W. Liu, H. Tan, K. Wong, H. H. Hng, T. M. Lim and Q. Yan, RSC Adv., 2012, 2, 1174–1180 RSC .
55. G. Armstrong, J. Canales, A. R. Armstrong and P. G. Bruce, J. Power Sources, 2008, 178, 723–728 CrossRef CAS .
56. S. Song, Q. Huang and W. Zhu, J. Nanopart. Res., 2017, 19, 353 CrossRef .
57. Q. Wang, J. Pan, M. Li, Y. Luo, H. Wu, L. Zhong and G. Li, J. Mater. Sci. Technol., 2015, 31, 630–633 CrossRef CAS .
58. S. Zhang, Y. Li, C. Wu, F. Zheng and Y. Xie, J. Phys. Chem. C, 2009, 113, 15058–15067 CrossRef CAS .
59. G. C. Li, K. Chao, C. Q. Zhang, Q. S. Zhang, H. R. Peng and K. Z. Chen, Inorg. Chem., 2009, 48, 1168–1172 CrossRef CAS PubMed .
60. C. Gao, X. Xie, R. He, B. Yu, B. Wu, Z. Luo, P. Liu, K. Chen and W. Xie, Mater. Lett., 2018, 225, 29–32 CrossRef CAS .
61. Z. Tian, H. Bai, C. Chen, Y. Ye, Q. Kong, Y. Li, W. Fan, W. Yi and G. Xi, iScience, 2019, 19, 836–849 CrossRef CAS PubMed .
62. M. Borek, F. Qian, V. Nagabushnam and R. K. Singh, Appl. Phys. Lett., 1993, 63, 3288–3290 CrossRef CAS .
63. C. Leroux, G. Nihoul and G. Van Tendeloo, Phys. Rev. B: Condens. Matter Mater. Phys., 1998, 57, 5111 CrossRef CAS .
64. A. Chen, Z. Bi, W. Zhang, J. Jian, Q. Jia and H. Wang, Appl. Phys. Lett., 2014, 104, 071909 CrossRef .
65. Y. Zhang, W. Xiong, W. Chen and Y. Zheng, Nanomaterials, 2021, 11, 338 CrossRef CAS PubMed .
66. C. Turner, Am. Ceram. Soc. Bull., 1991, 70, 1487–1490 CAS .
67. H. Li, P. He, Y. Wang, E. Hosono and H. Zhou, J. Mater. Chem., 2011, 21, 10999 RSC .
68. O. Monfort, T. Roch, L. Satrapinskyy, M. Gregor, T. Plecenik, A. Plecenik and G. Plesch, Appl. Surf. Sci., 2014, 322, 21–27 CrossRef CAS .
69. Z. Ding, Y. Cui, D. Wan, H. Luo and Y. Gao, RSC Adv., 2017, 7, 29496–29504 RSC .
70. C. Zhang, D. Hu, H. Gu, J. Xing, P. Xiong, D. Wan and Y. Gao, J. Adv. Ceram., 2017, 6, 196–206 CrossRef CAS .
71. C.-W. Huang, S.-S. Kuo and C.-L. Hsin, CrystEngComm, 2018, 20, 6857–6860 RSC .
72. H. Zhu, Z. Zhang and X. Jiang, J. Nanosci. Nanotechnol., 2020, 20, 1946–1954 CrossRef CAS PubMed .
73. A. Rúa, R. D. Díaz, S. Lysenko and F. E. Fernández, J. Appl. Phys., 2015, 118, 125308 CrossRef .
74. C. Tsang and A. Manthiram, J. Electrochem. Soc., 1997, 144, 520 CrossRef CAS .
75. J. Kohler, H. Makihara, H. Uegaito, H. Inoue and M. Toki, Electrochim. Acta, 2000, 46, 59–65 CrossRef CAS .
76. Z. Wen, H. Lai, C. Tianyin, L. Yueheng, Z. Liang, T. Shi, C. Ke, P. Liu, F. Xie, J. Chen, J. B. Xu and C. Qiulan, CrystEngComm, 2019, 21, 5749–5756 RSC .
77. O. Celik and M. Duman, Thin Solid Films, 2019, 691, 137590 CrossRef CAS .
78. H. Kim, J. Hong, K.-Y. Park, H. Kim, S.-W. Kim and K. Kang, Chem. Rev., 2014, 114, 11788–11827 CrossRef CAS PubMed .
79. X. Xie, S. Liang, J. Gao, S. Guo, J. Guo, C. Wang, G. Xu, X. Wu, G. Chen and J. Zhou, Energy Environ. Sci., 2020, 13, 503–510 RSC .
80. M.-C. Lin, M. Gong, B. Lu, Y. Wu, D.-Y. Wang, M. Guan, M. Angell, C. Chen, J. Yang, B.-J. Hwang and H. Dai, Nature, 2015, 520, 324–328 CrossRef CAS PubMed .
81. P. G. Bruce, B. Scrosati and J. M. Tarascon, Angew. Chem., Int. Ed., 2008, 47, 2930–2946 CrossRef CAS PubMed .
82. L. Mai, Q. Wei, Q. An, X. Tian, Y. Zhao, X. Xu, L. Xu, L. Chang and Q. Zhang, Adv. Mater., 2013, 25, 2969–2973 CrossRef CAS PubMed .
83. S. Yang, Y. Gong, Z. Liu, L. Zhan, D. P. Hashim, L. Ma, R. Vajtai and P. M. Ajayan, Nano Lett., 2013, 13, 1596–1601 CrossRef CAS PubMed .
84. D. Chao, C. Zhu, X. Xia, J. Liu, X. Zhang, J. Wang, P. Liang, J. Lin, H. Zhang, Z. X. Shen and H. J. Fan, Nano Lett., 2015, 15, 565–573 CrossRef CAS PubMed .
85. S. Li, G. Liu, J. Liu, Y. Lu, Q. Yang, L.-Y. Yang, H.-R. Yang, S. Liu, M. Lei and M. Han, J. Mater. Chem. A, 2016, 4, 6426–6432 RSC .
86. Y. Cai, S. Kumar, R. Chua, V. Verma, D. Yuan, Z. Kou, H. Ren, H. Arora and M. Srinivasan, J. Mater. Chem. A, 2020, 8, 12716–12722 RSC .
87. P. Martyniuk, J. Antoszewski, M. Martyniuk, L. Faraone and A. Rogalski, Appl. Phys. Rev., 2014, 1, 041102 Search PubMed .
88. M. Abdel-Rahman, S. Ilahi, M. F. Zia, M. Alduraibi, N. Debbar, N. Yacoubi and B. Ilahi, Infrared Phys. Technol., 2015, 71, 127–130 CrossRef CAS .
89. H. Wada, M. Nagashima, N. Oda, T. Sasaki, A. Kawahara, M. Kanzaki, Y. Tsuruta, T. Mori, S. Matsumoto, T. Shima, M. Hijikawa, N. Tsukamoto and H. Gotoh, Infrared Detectors and Focal Plane Arrays V, 1998, vol. 3379, pp. 90–100 Search PubMed .
90. R. Bhan, R. Saxena, C. Jalwani and S. Lomash, Def. Sci. J., 2009, 59, 580–589 CrossRef CAS .
91. J. Dai, X. Wang, S. He, Y. Huang and X. Yi, Infrared Phys. Technol., 2008, 51, 287–291 CrossRef CAS .
92. J. Hou, B. Wang, Z. Ding, R. Dai, Z. Wang, Z. Zhang and J. Zhang, Appl. Phys. Lett., 2017, 111, 072107 CrossRef .
93. H. J. Lee, D. Wang, T. H. Kim, D.-H. Jung, T.-H. Kil, K.-S. Lee, H.-J. Choi, S.-H. Baek, E. Yoon, W. J. Choi and J. Min Baik, Appl. Surf. Sci., 2021, 547, 149142 CrossRef CAS .
94. J. Liang, W. Li, J. Liu and M. Hu, Mater. Lett., 2016, 184, 92–95 CrossRef CAS .
95. R. Joy, Z. Han, K. Xu, X. Pan, N. Liao and H. Zhou, Phys. Lett. A, 2020, 384, 126823 CrossRef CAS .
96. K. K. Dey, D. Bhatnagar, A. K. Srivastava, M. Wan, S. Singh, R. R. Yadav, B. C. Yadav and M. Deepa, Nanoscale, 2015, 7, 6159–6172 RSC .
97. J. Liang, Q. Lou, W. Wu, K. Wang and C. Xuan, ACS Appl. Mater. Interfaces, 2021, 13, 31968–31977 CrossRef CAS PubMed .
98. H. Liu, D. Wan, A. Ishaq, L. Chen, B. Guo, S. Shi, H. Luo and Y. Gao, ACS Appl. Mater. Interfaces, 2016, 8, 7884–7890 CrossRef CAS PubMed .

This journal is © The Royal Society of Chemistry 2022