The characterization of bulk as-grown and annealed ZnO by the Hall effect

Abstract

A fully automated Temperature Dependent Hall (TDH) measurement setup has been assembled for the purposes of this study. This TDH setup is capable of measuring samples in the 20 K to 370 K temperature range. Sample sizes of up to 20 mm × 20 mm can be accommodated by the custom designed and manufactured sample holder. Samples with a resistance in the 1Ω to 250 MΩ range can be measured with this setup provided that the mobility of the sample is greater than 1 cm²/Vs. The computer program controlling the automated measurement processwas written in LabView™ version 6.1. Single crystal Zinc Oxide (ZnO) was the material under investigation in this study. Bulk ZnO samples grown by three different methods, namely pressurized melt growth, seeded chemical vapor transport (SCVT) growth and hydrothermal growth, were measured in the 20 K to 370 K range. The effect of annealing in argon atmosphere in the 550 ºC to 930 ºC range was investigated on all three ZnO types. In addition, hydrogen-implanted layers on semi-insulating hydrothermally grown ZnO were studied. These samples were annealed in the 200 ºC to 400 ºC range and Hall measurements in the 20 K to 330 K range were performed. Programs were written to fit, wherever possible, the obtained temperature dependent carrier concentration and mobility profiles to suitable theoretical models. The carrier concentration data was fitted to a multi-donor single acceptor charge balance equation for the purpose of extracting donor concentrations and activation energies. Before fitting, the data was corrected for the Hall scattering factor and, where necessary, for two-layer effects particularly a degenerate surface conduction channel that developed through annealing on the SCVT-grown and hydrothermally grown samples. The acceptor concentrations of the samples were obtained by fitting the mobility data to a model based on D.L. Rode’s method of solving the Boltzmann transport equation. Scattering mechanisms included in the model were piezoelectric and deformation potential acoustic modes, polar optic modes and ionized impurity scattering. It was found that the mobility data did not fit the model very well without assigning questionable values to other parameters, in this case the deformation potential. Plausible values for the acceptor concentration were however obtained. The carrier concentration data fitted the model well, but due to the large number of parameters to be extracted (up to six parameters in the case of three donors) there was often not much certainty in the extracted values This study shows that TDH analysis is a valuable tool to assess the quality of semiconductors. Bulk and degenerate surface (or interfacial) conduction are separated with relative ease, and shallow defect concentrations as well as compensation level concentrations could be extracted. The generally observed uncertainty in values obtained in the multi-parameter regression of carrier concentration data indicates that supplementary techniques such as photoluminescence are needed to support results obtained by the TDH technique.