Ab initio study of electronic and optical properties of $\mathit{Fe}$ doped anatase ${\mathit{TiO}}_2$ (1 0 1) surface

Highlights

• The calculations performed by the SIESTA density functional theory code.

• All calculations were spin-polarized.

• The band gap decreases as the concentration of the dopant increases.

• Enhanced optical absorption is clearly observed for Fe doped anatase ${\mathit{TiO}}_2$.

• The optical absorption is found to increase with the increase $\mathit{Fe}$ concentration.

Abstract

In this paper we investigated the effects of $\mathit{Fe}$ -doping of the anatase ${\mathit{TiO}}_2$ (1 0 1) surface on the crystal structure, electronic and optical properties, and impurity formation energy by means of density functional theory (DFT). The calculations were performed by the SIESTA DFT code and were carried out by using Troullier-Martins pseudopotentials for the 12-electron valence configuration ($3s^{2}3p^{6}3d^{2}4s^2$) of $\mathit{Ti}$ atom, 6-electron valence configuration ($2s^{2}2p^4$) of $O$ atom and 8-electron valence configuration ($3d^{6}4s^2$) of $\mathit{Fe}$ atom. We used a double- $\zeta$ basis set including polarization functions. All calculations were spin-polarized. The mechanism of narrowing the band gap and increasing the photocatalytic activity in the visible light region, of the doped ${\mathit{TiO}}_2$ is discussed by investigating the density of state. The band gap decreases as the concentration of the dopant increases. The Partial Density of States (PDOS) is not the same in the case of spin-up state or spin-down state. Enhanced optical absorption, for light polarized in the $z$ direction (parallel to the surface normal) is clearly observed for $\mathit{Fe}$ doped as compared to the pure anatase ${\mathit{TiO}}_2$ and the optical absorption is found to increase with the increase in the $\mathit{Fe}$ concentration. The DFT results indicate that the source of the increasing photocatalytic activity in the visible light region of the $\mathit{Fe}$ doped material is due to the introduction of additional electronic states within the band gap.

Since the $\mathit{Fe}$ atoms are more stable in $\mathit{Ti}$ substitutional lattice positions for the entire range of Fermi energy $E_F$ over the band gap, only this substitutional position is considered.

We hope that our results will highlight a route to improved electronic and optical properties of anatase ${\mathit{TiO}}_2$ for industrial applications.

Introduction

Titanium dioxide (${\mathit{TiO}}_2$) is one of the most studied transition metal oxides. It occurs naturally in three most common polymorphic structures: rutile, anatase and brookite. The brookite structure is very rare and difficult to prepare by experimental methods [1], [2] so this phase is of lesser interest in applications. The bulk single crystal anatase is less stable than bulk rutile (rutile is denser than anatase), but is more efficient and more widely used in photocatalysis and photoelectrochemistry [3], [4], [5], owing to higher levels of surface area and thus higher activity [6]. In addition to these three phases, Zhu and Gao [7] and Zhi-Gang Mei and co-authors [8] investigated the properties of nine different ${\mathit{TiO}}_2$ polymorphs (rutile, anatase, brookite, columbite, baddeleyite, cotunnite, pyrite, fluorite and tridymite).

Because of its high efficiency and photostability ${\mathit{TiO}}_2$ is widely applied to the production of low cost solar cells [9], [10], [11], hydrogen and environment protection (water and air detoxification) [12], [13], [14], [15], [16], [17], [18]. The advantages of ${\mathit{TiO}}_2$ as a semiconductor material are long-term stability, excellent functionality, non-toxic environmental acceptability and low cost availability [2]. ${\mathit{TiO}}_2$ is a wide-gap semiconductor; $\sim 3.2\mathit{eV}$ and $\sim 3.0\mathit{eV}$ for anatase and rutile, respectively. This wide band gap is its major disadvantage, because it means that it is mainly activated by ultraviolent (UV) light and, consequently, it is inefficient as an active solar cell material. The gap between the valence and conduction bands and the optical absortion properties are the most important quantities for industrial applications. Many attempts have been made to narrow the band gap, and also, to increase the photoreactivity in the visible light region. Generally, dopants will change the ${\mathit{TiO}}_2$ physical and chemical properties, in whatever phase; rutile or anatase.

The electronic structure and optical properties of anatase and rutile have been investigated both experimentally [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29] and theoretically [30], [31], [32], [33], [34], [35].

The investigation of the fundamental properties of ${\mathit{TiO}}_2$ are still very important because of their important role to effectively utilize solar energy [7]. Reducing the band gap has attracted much interest of researchers. A good method for this and to extend the spectral response of the ${\mathit{TiO}}_2$ to the visible light region is impurity doping. It has been noticed that doping ${\mathit{TiO}}_2$ with different metal and non-metal can improve the properties of material; may that doping lead to better synergistic effect, decrease the band gap and the recombination rate of the photo-generated electron-hole pairs [36], [37], [38], enhance the visible light absortion efficiently [39], [40]. There are number of papers regarding the narrowing the band gap by doping different elements in order to improve the photocatalytic activity of ${\mathit{TiO}}_2$ (see, for example, Refs. [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66]).

The complications with doping ${\mathit{TiO}}_2$ in order to reduce the band gap while maintaining the beneficial photocatalytic properties is noticed by Phattalung and coworkers [3]. To solve these problems it is necessary to understand, in detail, the changes resulting from the impurity. It is therefore desirable to study the properties of impurities.

In the last decade the research interest is mainly focused on adsorption of the ${\mathit{TiO}}_2$ surfaces [67], [68], [69], [70], [71], [72], [73], [74], [75]. In general, the anatase (1 0 1) surface is regarded to be most relevant to catalytic reactivity of ${\mathit{TiO}}_2$ [68]. The (1 0 1) surface is the predominant face that is exposed on anatase minerals, and theoretical calculations also show that it is thermodynamically the lowest-energy surface [76], [77]. Investigation of interface structures between ${\mathit{TiO}}_2$ surfaces and water [78] are performed by researchers, focusing on the adsorption manner of water molecules. The research of interface structures between ${\mathit{TiO}}_2$ surfaces and gas (for example $\mathit{NO}$ and ${\mathit{NO}}_2$ - mainly consistent of motor vehicle exhaust gas) are then continues by researchers [79], [80], [81], focusing on the adsorption manner of gas molecules. Researcher have found that light catalytic oxidation technology can effectively remove pollutants. Photocatalytic reactions at the ${\mathit{TiO}}_2$ surface have been attracting much attention of their applications to environmental cleaning; water and air purification. Because of strong oxidation activity and superhydrophilicity [82] ${\mathit{TiO}}_2$ can be used as antibacterial agent.

Doping of the surface layers may induce new features in the electronic structure and geometry of ${\mathit{TiO}}_2$ which are not present in the bulk.

In this paper we investigate the effects of $\mathit{Fe}$ -doping at the surface of ${\mathit{TiO}}_2$ on the electronic structure. It has been shown [69] that the presence of dopants at or near the surface is relevant for the activation of the redox-reaction processes. In general, the anatase (1 0 1) surface is regarded to be most relevant for catalytic reactivity of ${\mathit{TiO}}_2$ [68] and it plays a relevant role in the activation of photochemical processes [69].

Since the $\mathit{Fe}$ atoms are more stable in $\mathit{Ti}$ substitutional lattice positions (see Ref. [83]) for the entire range of Fermi energy $E_F$ over the band gap, only this substitutional position is considered.

One of our main motivations is to gain further insight into the electronic properties of anatase ${\mathit{TiO}}_2$ (1 0 1) surface by introducing $\mathit{Fe}$ dopants. This investigation has been done by means of density-functional theory (DFT) plane-wave pseudopotential method.

There are a number of papers regarding the properties of $\mathit{Fe}$ doped ${\mathit{TiO}}_2$; see, for example, Refs. [84], [85], [86], [87], but all these calculations have been done for the bulk structure.

The efficiency of photocatalytic activity of metal doped ${\mathit{TiO}}_2$ under visible light strongly depend on the dopant nature and concentration, as well as on the preparation methods and the thermal and reductive treatments [88], [89], [90], [91], [92], [93].

Yalcin and co-authors [85] investigated by DFT the influence of ${\mathit{Fe}}^{3+}$ doping on electronic and structural properties of ${\mathit{TiO}}_2$ and found that the additional electronic state within the band gap is formed, and consequently, decrease the band gap. First-principles calculations were conduct in the case of $\mathit{Fe}/N$ co-doped ${\mathit{TiO}}_2$ [84], [85], [86]. Hsuan-Chung Wu and co-authors [87] performed first-principles calculations using GGA+Hubbard U (GGA+U) approach, but all these calculations considered the bulk structure. Yuan and co-authors [94] performed first-principle calculations using LSDA + U approach but for rutile ${\mathit{TiO}}_2$ (1 1 0) and (0 1 1) surfaces.