No abstract

Transparent semiconducting β-Ga 2 O 3 single crystals were grown by the Czochralski method from an iridium crucible under a dynamic protective atmosphere to control partial pressures of volatile species of Ga 2 O 3 . Thermodynamic calculations on different atmospheres containing CO 2 , Ar and O 2 reveal that CO 2 growth atmosphere combined with overpressure significantly decreases evaporation of volatile Ga 2 O 3 species without any harm to iridium crucible. It has been found that CO 2 , besides providing high oxygen concentration at high temperatures, is also acting as a minor reducing agent for Ga 2 O 3 . Different coloration of obtained crystals as well as optical and electrical properties are directly correlated with growth conditions (atmosphere, pressure and temperature gradients), but not with residual impurities. Typical electrical properties of the n-type β-Ga 2 O 3 crystals at room temperature are: ρ = 0.1 -0.3 Ωcm, μ n,Hall = 110 -150 cm 2 V -1 s -1 , n Hall = 2 -6×10 17 cm -3 and E Ionisation = 30 -40 meV. A decrease of transmission in the IR-region is directly correlated with the free carrier concentration and can be effectively modulated by the dynamic growth atmosphere. Electron paramagnetic resonance (EPR) spectra exhibit an isotropic shallow donor level and anisotropic defect level. According to differential thermal analysis (DTA) measurements, there is substantially no mass change of β-Ga 2 O 3 crystals below 1200 °C (i.e. no decomposition) under oxidizing or neutral atmosphere, while the mass gradually decreases with temperature above 1200 °C. High resolution transmission electron microscopy (HRTEM) images at atomic resolution show the presence of vacancies, which can be attributed to Ga or O sites, and interstitials, which can likely be attributed to Ga atoms.

No abstract

We present a new approach for scaling-up the growth of β-Ga2O3 single crystals grown from the melt by the Czochralski method, which has also a direct application to other melt-growth techniques involving a noble metal crucible. Experimental and theoretical results point to melt thermodynamics as the crucial factor in increasing the volume of a growing crystal. In particular, the formation of metallic gallium in the liquid phase in large melt volumes causes problems with crystal growth and eutectic or intermetallic phase formation with the noble metal crucible. The larger crystals to be grown the higher oxygen concentration is required. The minimum oxygen concentration ranges from about 8 to 100 vol.% for 2 to 4 inch diameter cylindrical crystals, challenging the use of iridium crucibles in a combination with such high oxygen concentrations. A specific way of oxygen delivery to a growth furnace with the iridium crucible allows to minimize the formation of metallic gallium in the melt and thus obtaining large crystal volumes while decreasing the probability of the eutectic formation.

A systematic study of various, nominally undoped ZnO single crystals, either hydrothermally grown ͑HTG͒ or melt grown ͑MG͒, has been performed. The crystal quality has been assessed by x-ray diffraction, and a comprehensive estimation of the detailed impurity and hydrogen contents by inductively coupled plasma mass spectrometry and nuclear reaction analysis, respectively, has been made also. High precision positron lifetime experiments show that a single positron lifetime is observed in all crystals investigated, which clusters at 180-182 ps and 165-167 ps for HTG and MG crystals, respectively. Furthermore, hydrogen is detected in all crystals in a bound state with a high concentration ͑at least 0.3 at. %͒, whereas the concentrations of other impurities are very small. From ab initio calculations it is suggested that the existence of Zn-vacancyhydrogen complexes is the most natural explanation for the given experimental facts at present. Furthermore, the distribution of H at a metal/ZnO interface of a MG crystal, and the H content of a HTG crystal upon annealing and time afterward has been monitored, as this is most probably related to the properties of electrical contacts made at ZnO and the instability in p-type conductivity observed at ZnO nanorods in literature. All experimental findings and presented theoretical considerations support the conclusion that various types of Zn-vacancy-hydrogen complexes exist in ZnO and need to be taken into account in future studies, especially for HTG materials.