【Domestic Papers】Hybrid functional study of acceptor levels in β-Ga₂O₃ with stress engineering

      Researchers from the Harbin Institute of Technology have published a dissertation titled "Hybrid functional study of acceptor levels in β-Ga₂O₃ with stress engineering" in Journal of Applied Physics.

Project Support

      This work was supported by the National Natural Science Foundation of China (NSFC) (No. 12305300) and the Innovation Project of China Electronics Technology Group Corporation 46th Research Institute (No. WDZC202446008).

Background

      With the increasing demand for high-power semiconductors, β-Ga₂O₃ has attracted much attention due to its ultrawide bandgap (∼4.9 eV) and extremely high breakdown field(∼8 MV/cm), which is much greater than that of GaN (3.3 MV/cm) and 4H-SiC(∼2.5 MV/cm). β-Ga₂O₃ has the potential to construct low loss devices to achieve a green economy, in response to the development needs of fields such as electric vehicles, high-voltage rectifiers, and photovoltaic solar cells. Nevertheless, the challenge of p-type doping is a huge obstacle to the application of β-Ga₂O₃.

      Gallium oxide has five configurations, namely, α, β, γ, δ, and ɛ phase. Among them, β-Ga₂O₃ is the most stable and widely studied. Similar to other wide-bandgap semiconductors, its n-type doping has considerable flexibility; on the contrary, there is currently no workable solution for p-type conductivity. This is primarily due to the fact that (1) the valence band maximum(VBM) of β-Ga₂O₃ is deep below the vacuum level, making it difficult for holes to be excited to form free carriers. (2) There are natural compensating defects like O vacancy in β-Ga₂O₃. (3) The VBM of β-Ga₂O₃ is mainly composed of the p-orbitals of oxygen atoms with small dispersion leading to a large effective mass of holes. Even with a certain concentration of holes, their conduction is hindered. (4) The holes in β-Ga₂O₃ will be spontaneously captured by O atoms in the form of self-trapped holes (STHs), and polarons will be formed through strong lattice distortion. In addition, compensating defects can be controlled by adjusting the growth conditions, and other problems are difficult to overcome.

Abstract

      β-Ga₂O₃ has emerged as a wide-bandgap semiconductor material, which can be widely used in high-power electronic devices and solar blind ultraviolet detectors. Nevertheless, the challenge of p-type doping is a huge obstacle to the application of β-Ga₂O₃. Numerous studies have been devoted to finding p-type dopants for β-Ga₂O₃. However, the outcomes have been less than satisfactory due to its valence bands with small dispersions and self-trapped hole properties. Herein, using the hybrid functional method, 27 defect configurations are built to reveal the possibility of p-type doping, containing Mg_Ga, N_O, P_O, transition metal substitutions (TM_Ga), and complex defects. First, unlike conventional defect calculations that employ generalized gradient approximation (GGA) or meta-GGA for structural optimization, we demonstrate that only hybrid functional can capture the large structural distortions caused by polarons in β-Ga₂O₃. Notably, defects previously classified as shallow-level centers (e.g., P_o and Mg_Ga–N_o complexes) are revealed to form deep-level states at our theoretical level of treatment. Second, for TM_Ga defects, only the TM atoms from VIII to IIB exhibit acceptor level. Furthermore, the strain can significantly regulate the acceptor levels of Ni_Ga and Cu_Ga. However, the acceptor levels of Mg_Ga and Zn_Ga exhibit high tolerance to strain. Our results provide a reference for the study of p-type doping of β-Ga₂O₃, emphasizing the criteria of hybrid functional optimization and the difficulty in forming shallow acceptor levels using conventional substitutional defects.

Conclusion

      In summary, our work provided a guideline for first-principle research on acceptor defects of β-Ga₂O₃, stating that hybrid functional must be employed to optimize the structure, whereas GGA and meta-GGA functionals fail to capture the significant distortions caused by polarons in the β-Ga₂O₃ system. We have investigated defects in β-Ga₂O₃ for potential p-type doping but found all defects indicating deep acceptor levels. We believed that the previous first-principle calculations that suggested that N_O, P_O, and Mg_Ga–N_O complex defects could achieve p-type doping, which are considered deep-level under our theoretical level, are due to the absence of hybrid functional optimization. TM substitutes from group IIIB to VIIB do not possess ɛ(0/−1) transition levels, whereas substitutes from VIII to IIB exhibit acceptor levels. Finally, strain can significantly change the bandgap but minimally alter the shapes of the band edge. Mg_Ga and Zn_Ga exhibit tolerance to strain, while the acceptor level of Cu_Ga and Ni_Ga significantly decreases under tensile strain and increases under compression. Our results can be utilized as a valuable reference database. Strain can effectively regulate the acceptor levels of TM substitutes, but it does not effectively facilitate p-type conductivity. Future researchers are recommended to use alloys or heavy doping for achieving p-type β-Ga₂O₃.