Intrinsic and extrinsic doping of ZnO and ZnO alloys

In this article the doping of the oxidic compound semiconductor ZnO is reviewed with special emphasis on n-type doping. ZnO naturally exhibits n-type conductivity, which is used in the application of highly doped n-type ZnO as a transparent electrode, for instance in thin film solar cells. For prospective application of ZnO in other electronic devices (LEDs, UV photodetectors or power devices) p-type doping is required, which has been reported only minimally. Highly n-type doped ZnO can be prepared by doping with the group IIIB elements B, Al, Ga, and In, which act as shallow donors according to the simple hydrogen-like substitutional donor model of Bethe (1942 Theory of the Boundary Layer of Crystal Rectifiers (Boston, MA: MIT Rad Lab.)). Group IIIA elements (Sc, Y, La etc) are also known to act as shallow donors in ZnO, similarly explainable by the shallow donor model of Bethe. Some reports showed that even group IVA (Ti, Zr, Hf) and IVB (Si, Ge) elements can be used to prepare highly doped ZnO films-which, however, can no longer be explained by the simple hydrogen-like substitutional donor model. More probably, these elements form defect complexes that act as shallow donors in ZnO. On the other hand, group V elements on oxygen lattice sites (N, P, As, and Sb), which were viewed for a long time as typical shallow acceptors, behave instead as deep acceptors, preventing high hole concentrations in ZnO at room temperature. Also, 'self'-compensation, i.e. the formation of a large number of intrinsic donors at high acceptor concentrations seems to counteract the p-type doping of ZnO. At donor concentrations above about 1020 cm(-3), the electrical activation of the dopant elements is often less than 100%, especially in polycrystalline thin films. Reasons for the electrical deactivation of the dopant atoms are (i) the formation of dopant-defect complexes, (ii) the compensation of the electrons by acceptors (Oi, VZn) or (iii) the formation of secondary phases, for instance Al2O3, Ga2O3 etc. The strong influence of the different deposition methods and annealing conditions on the doping of ZnO is discussed. This review shows that, though it is one of the best-investigated oxide compound semiconductors over many decades, understanding of the details of the doping properties and mechanisms of zinc oxide is still in its infancy. Based on this review, prospective research opportunities are devised.