First principle study of CuN, Cu and N-doped anatase TiO2
Introduction
TiO2 has received more and more attention as a promising material for catalyst carrier, dye-sensitized solar cells and photochemical applications [1], [2], [3], [4], [5]. TiO2 phase have three distinct structures at the nature state: rutile, anatase and brookite, of which rutile is the most stable one. Although anatase is less stable than rutile, it is more efficient and more widely used in catalysis and photoelectrochemistry at lower temperature about 873 K [6], [7]. For example, some researchers find that metals/TiO2-anatase show higher activities than metals/TiO2–rutile catalysts [4], [5], [8].
As for the photochemical applications of pure anatase, the band gap energy is about 3.2 eV, which is mainly activated by ultraviolet (UV) lights. The result shows that the application of UV radiation alone for photoelectrochemistry is not an economically feasible option, because this part of the spectrum accounts for only 3–4% of solar radiation [9]. Therefore, in order to harvest the energy from the visible (V) region of sunlight, the absorption edge of anatase must be shifted to this region. Many studies have been carried out in the attempt to shift the absorption edge to the V region by implanting an adequate amount of cation or anion such as Co, Cu, Pd, V, N, C and F [10], [11], [12], [13], [14], [15]. The band gap of TiO2 for the photo-excitation is reduced by implanting an adequate amount of cation or anion. More recently, the simultaneous doping of two kinds of atoms into TiO2 has attracted considerable interest, since it can result in a higher photocatalytic activity and peculiar characteristics compared with single element doping into TiO2, such as CoN, CeN and NTa co-doped TiO2 [16], [17], [18], [19]. Therefore, in order to further improve the photocatalytic activity, simultaneous doping of two kinds of atoms in TiO2 is necessary.
The Cu- or N-doped anatase was studied by many experiments, which lead to significant enhancement of photocatalytic activity compared with pure anatase. In order to explain the reason of Cu- or N-doped anatase samples, the Cu- or N-doped anatase were studied by theorists [11], [14]. They found that the band gap decreases due to N or Cu-doped, thereby, the photocatalytic efficiency of anatase increases. The calculation results are inconsistent with experiment. To the best of our knowledge, two kinds of ions doped in anatase lattice have been studied by some experimenters and theorists, such as F and Zr, N and F, N and Zr [12], [20], [21], but the co-doped of N and Cu in anatase lattice is lacking. To understand the microscopic mechanisms of CuN co-doped and give advice for the experiments and enhance anatase's photocatalysis in the future, Cu and N co-doped in anatase is studied in the paper.
Section snippets
Theoretical methods
Cambridge Serial Total Energy Package (CASTEP) has been used in our calculations, which is performed within DFT plane-wave pseudopotential method [22]. The general gradient approximation (GGA) with Perdew–Burke–Ernzerhof (PBE) [23] functional and ultra-soft pseudo-potentials was used to describe the exchange correlation effects and electronion interactions, respectively. A (2×2×1) anatase supercell (Ti16O32) was studied, in which 3×3×3 meshes of k points was used to sample the Brillouin zone. A
Result and discussion
The equilibrium lattice constants of anatase are a=3.806 Å, c=9. 874 Å after optimization, compared to an experimental value of a=3.786 Å, c=9.514 Å [24]. The calculated band gap is 1.91 eV, which is smaller than that of the experimental value (3.20 eV) [25]. This difference mainly arises from the use of GGA for the exchange-correlation functional, which generally underestimates the energy gap [11], [26]. Therefore, DFT+U (Hubbard coefficient) approach should be used to compensate for the
Conclusion
In summary, the stability of Cu interstitial system is higher than that of Cu substitutional system, thereby N and Cu co-doped anatase is prefer to formation of N substitution O and Cu interstitial system. An isolated energy state about 1.6 eV occurs due to Ti 3d and N 2p hybridization after CuN doped anatase, the VB maximum and CB minimum shifts to high energy and low energy respectively due to O 2p and Cu 3d hybridization. The photocatalytic activity shows the order CuN co-doped⪢Cu-doped⪢N
Acknowledgments
The authors gratefully acknowledge the financial support of this study by the National Natural Science Foundation of China (Grant nos. 51175363, 51274149) and the Program for Chang jiang Scholar and Innovative Research Team in University (IRT0972).
References (28)
Surf. Sci. Rep.
(2001)- et al.
J. Mol. Catal. A: Chem.
(2000) - et al.
J. Nat. Gas Chem.
(2009) - et al.
J. Catal.
(2005) - et al.
Surf. Sci.
(2007) - et al.
J. Electron Spectrosc.
(2009) - et al.
Energy
(2009) - et al.
J. Mol. Struct.
(2009) - et al.
Solid State Commun.
(2005) - et al.
Solid State Commun.
(2009)