| ZnO has long been touted as an excellent material for UV light-emitting diodes (LEDs) and lasers. The main reasons are a bandgap of 3.37 eV, and a free exciton with a 60 meV binding energy, which permits excitonic emission at room temperature and above. Furthermore, (1) ZnO can be grown by most types of epitaxy, or in ingot form; (2) it has a native substrate; (3) wet chemical processing is possible; and (4) it is more resistant to radiation damage. For these advantages, ZnO is promised to replace III- V semiconductors, such as GaN, and GaAs, and most widely used in the field of photoelectricity devices, in future. The naturally occurring ZnO has n-type conductivity which can be enhanced by doping with group-Ill elements (Al, Ga, etc.). However, ZnO has largely failed to live up to its potential, because it has proven too difficult to produce high-conductivity p-type ZnO due to (i) a too low p-type dopant solubility, (ii) a too deep dopant energy level, and (iii) self-compensation by intrinsic defects. According to theoretical prediction, among all possible acceptors, nitrogen is the best candidate to produce a shallow acceptor level in ZnO, and several groups have achieved p-type ZnO with N as a dopant. But it either had low N concentration and high resistivity, or was not stable. The reason is that N doping increases the Madelung energy of ZnO, resulting in the instability of the ionnic charge distributions. However, band structure calculations showed that co-doping N and Ga in the ratio of 2 : 1 will decrease the Madelung energy, and it can greatly enhance the incorporation of N, thus stable p-type ZnO with high hole concentration can be obtained using co-doping technique. According to this idea, p-type ZnO was prepared in N2O - O2 atmospheres by Al - N co-doping method using DC reactive magnetron sputtering in this paper. Effects of growth parameters, such as substrate temperatures and Al contents in sputtering targets, on the properties of as-grown co-doped ZnO were discussed. Moreover, ZnO p - n homojunctions have been fabricated, and the mechanism of co-doping was also investigated in detail. The main results are as follows:1. Al-N co-doped p-type ZnO has been firstly prepared by magnetron sputtering, with N2O as an acceptor source and Al as a donor source. Co-doped ZnO films with a thickness of about 270 nm exhibit polycrystalline structure with the preferential orientation of (002) plane, smooth dense surface, and good grain size uniformity. Thehole concentration of Al-N co-doped ZnO is two orders higher, with one order lower for the resistivity, compared with those of N doping alone. The highest carrier concentration of the as-grown ZnO is 1.3 x 10 18 cm 3, with the lowest resistivity of 54 cm, respectively, and the growth condition is: substrate temperature of 500 , 0.4 at% Al in target, and in pure N2O.2. With In-Zn alloys as electrodes, ZnO p - n homojunctions were fabricated successfully on quartz substrates by depositing Al doped n-type ZnO layer (5.4 x 10 3 cm) on Al-N co-doped p-type ZnO layer (54 cm), with a turn-on voltage of about 3 V and a low leakage current. The electrical properties are better than those of ZnO p - n homojunctions with a p-type ZnO layer using N as the acceptor source reported until now.3. According to the results of us and other groups, several new viewpoints on the co-doping technique were proposed:(a) For co-doping technique (with N2O as an acceptor and Al as a donor source), a nitrogen atom from NO (from the dissociation of N2O) and an Al - N pair form the complex [Al, 2N]; the formation of Al - N pairs is necessary to achieve stable p-type ZnO with a high hole concentration.(b) The low carrier mobility of the co-doped ZnO is likely due to the existence of Al-N pairs and the poor crystal 'quality of ZnO.(c) Once the crystal quality of ZnO improves, the carrier mobility of the co-doped ZnO will increase enormously, thereby p-type ZnO with low resistivity can be achieved using co-doping technique.
|
PDF Full Text Request