We perform first-principles calculations to investigate the electronic and vibrational spectra and the electron mobility of β-GaO. We calculate the electron-phonon scattering rate of the polar optical phonon modes using the Vogl model in conjunction with Fermi's golden rule; this enables us to fully take the anisotropic phonon spectra of the monoclinic lattice of β-GaO into account. We also examine the scattering rate due to ionized impurities or defects using a Yukawa-potential-based model. We consider scattering due to donor impurities, as well as the possibility of compensation by acceptors such as Ga vacancies. We then calculate the room-temperature mobility of β-GaO using the Boltzmann transport equation within the relaxation time approximation, for carrier densities in the range from 10 to 10 cm. We find that the electron-phonon interaction dominates the mobility for carrier densities of up to 10 cm. We also find that the intrinsic anisotropy in the mobility is small; experimental findings of large anisotropy must therefore be attributed to other factors. We attribute the experimentally observed reduction of the mobility with increasing carrier density to increasing levels of compensation, which significantly affect the mobility.

BaSnO3 (BSO) is a promising transparent conducting oxide (TCO) with reported roomtemperature (RT) Hall mobility exceeding 320 cm 2 V −1 s −1 . Among perovskite oxides, it has the highest RT mobility, about 30 times higher than that of the prototypical SrTiO3. Using firstprinciples calculations based on hybrid density functional theory, we elucidate the physical mechanisms that govern the mobility by studying the details of LO-phonon and ionized impurity scattering. A careful numerical analysis to obtain converged results within the relaxation-time approximation of Boltzmann transport theory is presented. The k dependence of the relaxation time is fully taken into account. We find that the high RT mobility in BSO originates not only from a small effective mass, but also from a significant reduction in the phonon scattering rate compared to other perovskite oxides; the origins of this reduction are identified. Ionized impurity scattering influences the total mobility even at RT for dopant densities larger than 5 × 10 18 cm −3 , and becomes comparable to LO-phonon scattering for 1 × 10 20 cm −3 doping, reducing the drift mobility from its intrinsic LO-phonon-limited value of ∼594 cm 2 V −1 s −1 to less than 310 cm 2 V −1 s −1 . We suggest pathways to avoid impurity scattering via modulation doping or polar discontinuity doping. We also explicitly calculate the Hall factor and Hall mobility, allowing a direct comparison to experimental reports for bulk and thin films and providing insights into the nature of the dominant mechanisms that limit mobility in state-of-the art samples.

BaSnO3 (BSO) is a transparent perovskite oxide with high room-temperature mobility, a property that is highly desirable for a channel material in transistors. However, its low density of states (DOS) makes it challenging to confine a high-density two-dimensional electron gas (2DEG). Using hybrid density functional theory, we calculate the band structure of BSO, its DOS, and its band offsets with candidate barrier materials, such as SrTiO3 (STO), LaInO3, and KTaO3. With the calculated material parameters as input, Schrödinger-Poisson simulations are then performed on BSO heterostructures to quantitatively address the issue of 2DEG confinement. The BSO/STO interface with a conduction-band offset of 1.14 eV limits the 2DEG density confined within BSO to 8×1013 cm−2. Strategies to improve the confinement via band-offset engineering are discussed.

No abstract

The perovskite stannates (ASnO3; A = Ba, Sr, Ca) are promising for oxide electronics, but control of n-type doping has proved challenging. Using first-principles hybrid density functional calculations, we investigate La dopants and explore the formation of compensating acceptor defects. We find that La on the A site always behaves as a shallow donor, but incorporation of La on the Sn site can lead to self-compensation. At low La concentrations and in O-poor conditions, oxygen vacancies form in BaSnO3. A-site cation vacancies are found to be dominant among the native compensating centers. Compared to BaSnO3, charge compensation is a larger problem for the wider-band-gap stannates, SrSnO3 and CaSnO3, a trend we can explain based on conduction-band alignments. The formation of compensating acceptor defects can be inhibited by choosing oxygen-poor (cation-rich) growth or annealing conditions, thus providing a pathway for improved n-type doping.