Reverse bias current-voltage measurements of ϳ100-m-diameter gold Schottky contacts deposited on as-received, n-type ZnO(0001) wafers and those exposed for 30 min to a remote 20% O 2 /80% He plasma at 525Ϯ20°C and cooled either in vacuum from 425°C or the unignited plasma gas have been determined. Plasma cleaning resulted in highly ordered, stoichiometric, and smooth surfaces. Contacts on as-received material showed A leakage currents and ideality factors Ͼ2. Contacts on plasma-cleaned wafers cooled in vacuum showed ϳ36Ϯ1 nA leakage current to Ϫ4 V, a barrier height of 0.67Ϯ0.05 eV, and an ideality factor of 1.86Ϯ0.05. Cooling in the unignited plasma gas coupled with a 30 s exposure to the plasma at room temperature resulted in decreases in these parameters to ϳ20 pA to Ϫ7 V, 0.60Ϯ0.05 eV, and 1.03Ϯ0.05, respectively. Differences in the measured and theoretical barrier heights indicate interface states. ͑0001͒ and (0001) are used in this letter to designate the polar zinc-and oxygen-terminated surfaces, respectively.

Current-voltage measurements of Au contacts deposited on ex situ cleaned, n-type ZnO͑0001͒ ͓͑0001͔͒ surfaces showed reverse bias leakage current densities of ϳ0.01 ͑ϳ0.1͒ A / cm 2 at 4.6 ͑3.75͒ V reverse bias and ideality factors Ͼ2 ͑both surfaces͒ before sharp, permanent breakdown ͑soft breakdown͒. This behavior was due primarily to the presence of ͑1.6-2.0͒ ± 0.1 ͓͑0.7-2.6͒ ± 0.1͔ monolayers ͑ML͒ of hydroxide, which forms an electron accumulation layer and increases the surface conductivity. In situ remote plasma cleaning of the ͑0001͒ ͓͑0001͔͒ surfaces using a 20 vol % O 2 / 80 vol % He mixture for the optimized temperatures, times, and pressure of 550± 20°C ͑525± 20°C͒, 60 ͑30͒ min, and 0.050 Torr reduced the thickness of the hydroxide layer to ϳ0.4± 0.1 ML and completely eliminated all detectable hydrocarbon contamination. Subsequent cooling of both surfaces in the plasma ambient resulted in the chemisorption of oxygen and a change from 0.2 eV of downward band bending for samples cooled in vacuum to 0.3 eV of upward band bending indicative of the formation of a depletion layer of lower surface conductivity. Cooling in either ambient produced stoichiometric ZnO͕0001͖ surfaces having an ordered crystallography as well as a step-and-terrace microstructure on the ͑0001͒ surface; the ͑0001͒ surface was without distinctive features. Sequentially deposited, unpatterned Au films, and presumably the rectifying gold contacts, initially grew on both surfaces cooled in the plasma ambient via the formation of islands that subsequently coalesced, as indicated by calculations from x-ray photoelectron spectroscopy data and confirmed by transmission electron microscopy. Calculations from the current-voltage data of the best contacts revealed barrier heights on the ͑0001͒ ͓͑0001͔͒ surfaces of 0.71± 0.05 ͑0.60± 0.05͒ eV, a saturation current density of ͑4 ± 0.5͒ ϫ 10 −6 A / cm 2 ͑2.0± 0.5ϫ 10 −4 A / cm 2 ͒, a lower value of n = 1.17± 0.05 ͑1.03± 0.05͒, a significantly lower leakage current density of ϳ1.0ϫ 10 −4 A / cm 2 ͑ϳ91ϫ 10 −9 A / cm 2 ͒ at 8.5 ͑7.0͒ V reverse bias prior to sharp, permanent breakdown ͑soft breakdown͒. All measured barrier heights were lower than the predicted Schottky-Mott value of 1.0 eV, indicating that the interface structure and the associated interface states affect the Schottky barrier. However, the constancy in the full width at half maximum of the core levels for Zn 2p ͑1.9± 0.1 eV͒ and O 1s ͑1.5± 0.1 eV͒, before and after sequential in situ Au depositions, indicated an abrupt, unreacted Au/ ZnO͑0001͒ interface. Transmission electron microscopy confirmed the abruptness of an epitaxial interface. Annealing the contacts on the ͑0001͒ surface to 80± 5 and 150± 5°C resulted in decreases in the ideality factors to 1.12± 0.05 and 1.09± 0.05 and increases in saturation current density to 9.05 and 4.34 A / cm 2 , the barrier height to 0.82± 0.5 and 0.79± 0.5 eV, and in the leakage current densities to ϳ2 ϫ 10 −3 A / cm 2 at 6 V and ϳ20ϫ 10 −3 A / cm 2 at 7 V, respectively.

A layer containing an average of 1.0 monolayer ͑ML͒ of adventitious carbon and averages of 1.5 ML and 1.9 ML of hydroxide was determined to be present on the respective O-terminated (0001 ) and Zn-terminated ͑0001͒ surfaces of ZnO. A diffuse low-energy electron diffraction pattern was obtained from both surfaces. In situ cleaning procedures were developed and their efficacy evaluated in terms of the concentrations of residual hydrocarbons and hydroxide and the crystallography, microstructure, and electronic structure of these surfaces. Annealing ZnO(0001 ) in pure oxygen at 600-650°CϮ20°C reduced but did not eliminate all of the detectable hydrocarbon contamination. Annealing for 15 min in pure O 2 at 700°C and 0.100Ϯ0.001 Torr caused desorption of both the hydrocarbons and the hydroxide constituents to concentrations below the detection limits (ϳ0.03 MLϭϳ0.3 at. %) of our x-ray photoelectron spectroscopy instrument. However, thermal decomposition degraded the surface microstructure. Exposure of the ZnO(0001 ) surface to a remote plasma having an optimized 20% O 2 /80% He mixture for the optimized time, temperature, and pressure of 30 min, 525°C, and 0.050 Torr, respectively, resulted in the desorption of all detectable hydrocarbon species. Approximately 0.4 ML of hydroxide remained. The plasma-cleaned surface possessed an ordered crystallography and a step-and-terrace microstructure and was stoichiometric with nearly flat electronic bands. A 0.5 eV change in band bending was attributed to the significant reduction in the thickness of an accumulation layer associated with the hydroxide. The hydroxide was more tightly bound to the ZnO͑0001͒ surface; this effect increased the optimal temperature and time of the plasma cleaning process for this surface to 550°C and 60 min, respectively, at 0.050 Torr. Similar changes were achieved in the structural, chemical, and electronic properties of this surface; however, the microstructure only increased slightly in roughness and was without distinctive features.