Characterization of SnO2-Ga2O3 alloys grown by spray pyrolysis and hot wire MOCVD for wide gap transparent conductors
Résumé
Ga2O3 is a wide band gap transparent conductive oxide (TCO) currently investigated for Ultra-Violet optroelectronic and photonic devices. The alloying of Ga2O3 with SnO2 is expected to provide n–type conductivity1. Previous investigation concerned single crystal growth and physical growth method but the use of chemical deposition presents great interest for large scale applications. Ga2O3 - SnO2 layers were grown on (0001) sapphire either by spray pyrolysis2 under atmospheric pressure and a substrate heated at 600°C or by hot wire CVD3 under a pressure of 10 Torr and a substrate heated at 700°C. In both cases, thicknesses were between 150 and 200 nm. Compositions were determined by EDX.
X-Ray diffraction patterns of layers in θ-2θ mode are depicted in Fig. 1 after growth. The undoped samples show a preferential growth direction with the family of lattice planes {-201}. This relation is kept for tin concentration up 15% while the interatomic distance decreases with an increase in SnO2 concentration. Amorphous patterns are observed beyond 21 mol%. Two remaining peaks in the XRD pattern of the heavily Sn-doped Ga2O3 film may probably attributed to (200) and (400) peaks of SnO2 phase, and may be related with a partial segreagation of SnO2 crystallites. Subsequent thermal annealing’s were performed for 15 min at temperature up to 1000°C either at atmospheric pressure or at 0.01 Torr showing a very weak crystallization of gallium oxide.
The optical properties of the annealed samples present a decrease in the band gap from 5 eV down to 4.6 eV to 21% SnO2 quite independently from the chosen growth process or annealing method (Fig 3). On the other hand, transmission measurements deviate from a direct absorption gap. The additional tails can originate from indirect transition or Urbach mechanism associated with Sn incorporation (Fig. 2).
However, the resistivity remains very high in all samples (>108Ω/cm). These results indicate that doping is not simply responsible for the conductivity. Vacuum annealing used to change oxygen stoichiometry did not provide any improvements. Transmission electron microscopy (TEM) analyses were carried out on annealed samples. They present a high density of nano-crystalline grains. These phases have particular orientations which are related to the twinning of Ga2O3. The nano-twinning of Ga2O3 can explain the reduction of carrier’-s’ mobility and provide some explanation about the high resistivity in our films. Others annealing and plasma treatments5 will be presented to corroborate these results.