Photoluminescence spectroscopy has been used to study single crystalline ZnO samples systematically annealed in inert, Zn-rich and O-rich atmospheres. A striking correlation is observed between the choice of annealing ambient and the position of the deep band emission (DBE) often detected in ZnO. In particular, annealing in O2 results in a DBE at 2.35±0.05eV, whereas annealing in the presence of metallic Zn results in DBE at 2.53±0.05eV. The authors attribute the former band to zinc vacancy (VZn) related defects and the latter to oxygen vacancy (VO) related defects. Additional confirmation for the VO and VZn peak identification comes from the observation that the effect is reversible when O- and Zn-rich annealing conditions are switched. After annealing in the presence of ZnO powder, there is no indication for the VZn- or VO-related bands, but the authors observe a low intensity yellow luminescence band peaking at 2.17eV, probably related to Li, a common impurity in hydrothermally grown ZnO.

The role of vacancy clustering and acceptor activation on resistivity evolution in N ion-implanted n-type hydrothermally grown bulk ZnO has been investigated by positron annihilation spectroscopy, resistivity measurements, and chemical profiling. Room temperature 220 keV N implantation using doses in the low 10 15 cm −2 range induces small and big vacancy clusters containing at least 2 and 3-4 Zn vacancies, respectively. The small clusters are present already in as-implanted samples and remain stable up to 1000°C with no significant effect on the resistivity evolution. In contrast, formation of the big clusters at 600°C is associated with a significant increase in the free electron concentration attributed to gettering of amphoteric Li impurities by these clusters. Further annealing at 800°C results in a dramatic decrease in the free electron concentration correlated with activation of 10 16 -10 17 cm −3 acceptors likely to be N and/or Li related. The samples remain n type, however, and further annealing at 1000°C results in passivation of the acceptor states while the big clusters dissociate.

No abstract

Schottky barrier contacts have been formed by electron beam evaporation of Pd on the (0001¯)-face of hydrothermally grown n-type single-crystalline ZnO. The contacts can be operated under reverse bias voltages up to −5V and in the temperature range between 130 and 350K. A barrier height of 0.83eV is deducted at room temperature, which is in reasonable agreement with the value predicted by fundamental theory. The ideality factor for the current-voltage characteristics varies between 1.01 and 1.03 in the voltage range of −5.0to+0.5V, indicating that thermionic emission is a dominant mechanism for charge-carrier transport. Thermal admittance measurements were performed between 15 and 350K, and two prominent levels were resolved at 0.05(1) and 0.33(2)eV below the conduction band edge, respectively. The origin of these levels is not known, but possible candidates are complexes involving hydrogen, Zn interstitials, and vacancy-oxygen centers.

Li is present in hydrothermally grown ZnO at high concentrations and is known to compensate both n-and p-type doping due to its amphoteric nature. However, Li can be manipulated by annealing and ion implantation in ZnO. Fast, 20 ms flash anneals in the 900-1400°C range result in vacancy cluster formation and, simultaneously, a low-resistive layer in the implanted part of the He-and Li-implanted ZnO. The vacancy clusters, involving 3-4 Zn vacancies, trap and deactivate Li, leaving other in-grown donors to determine the electrical properties. Such clusters are not present in sufficient concentrations after longer ͑1 h͒ anneals because of a relatively low dissociation barrier ϳ2.6± 0.3 eV, so ZnO remains compensated until Li diffuses out after 1250°C anneals.ZnO has great potential as a material for optoelectronic applications. 1,2 Hydrothermally ͑HT͒ grown ZnO material is of particular interest, as this growth method is scalable. 3 However, electronic doping issues in ZnO in general, and in HT ZnO in particular, are not fully controlled or understood. For example, the role of lithium needs to be addressed. HT ZnO is synthesized in a solution containing LiOH and is therefore abundant with Li. Lithium's lattice position decides whether it exhibits donor-or acceptorlike character in ZnO; occupying zinc sites ͑Li Zn ͒ it is an acceptor, occupying interstitial sites ͑Li i ͒ it is a donor. 4-6 This amphoteric behavior 7 explains why Li doping produces highly resistive or even semi-insulating 8 material. Interestingly, it has recently been reported that in sputtered thin ZnO films Li may act as a dominating p-type dopant. 9 However, the atomistic doping mechanism is not well understood, and the doping efficiency depends strongly on the sputtering and annealing conditions. 9 It is thus important to investigate if and how Li can be ͑i͒ stabilized as Li Zn or Li i , ͑ii͒ deactivated or gettered, or ͑iii͒ removed from the HT ZnO material. Either scenario would facilitate electronic doping, minimizing compensation by amphoteric Li.Ion implantation introduces intrinsic defects, and in some cases electronic states associated with the implanted impurity. However, activation of the implanted impurities by annealing results in limited modifications in the conductivity of the highly resistive HT ZnO, 10 presumably because of the amphoteric role of Li. The measurements in Ref. 10 were performed on highly Li-contaminated samples employing conventional anneals at temperatures Շ1000°C; these conditions were probably insufficient to remove Li from the samples, but sufficient to promote Li amphoteric behavior. In this work we report on how an extremely fast heat treatment ͑ϳ20 ms͒, so called flash annealing, influences the interaction between Li and the implantation-induced defects and how it affects the electrical properties of the ion-implanted ZnO. We have used two types of ions, He + and Li + . The former generates intrinsic defects only, whereas the latter, in addition, alters the Li concentration in the sample. Thus, we are able to ...

An overview of our recent results on characterization and modification of high-resistivity n-type bulk zinc oxide samples, grown by hydrothermal techniques, is given. Three specific topics are addressed; (i) the role of lithium (Li) as an electrically compensating impurity, (ii) extrinsic n-type doping by hydrogen implantation, and (iii) influence of annealing conditions on deep band emission. In (i), furnace annealing of as-grown samples at temperatures above ~800 °C is shown to cause out-diffusion of residual Li impurities and concurrently, the resistivity decreases. After annealing at 1400 °C, a resistivity close to 10 -1 Ω cm is obtained and the Li content is reduced from above 10 17 cm -3 to the mid 10 15 cm -3 range, providing evidence for the crucial role of Li as an electrically compensating impurity. For ion-implanted samples, vacancy clusters evolve during post-implant flash lamp annealing (20 ms duration) and these clusters appear to trap and deactivate Li with a resulting improvement of the n-type conductivity. However, these clusters have a limited stability and start to dissociate already after 1h at 900 °C, accompanied by a decrease in the conductivity. For topic (ii), n-type doping by hydrogen implantation is shown to enhance the conductivity by about 5 orders of magnitude already in the as-implanted state. Despite substantial loss of hydrogen, the conductivity remains stable, or even increases, after annealing up to ≥ 600 °C, and necessary conditions for doping by hydrogen are discussed. In (iii), the origin of the commonly observed deep band emission from monocrystalline zinc oxide is investigated using a concept of annealing as-grown samples in different atmospheres. A strong influence by the atmosphere and temperature is observed and the results can be interpreted in terms of dominant effects on the emission by vacancy-related defects.

No abstract

No abstract