We identify a dominant light-emitting center in ion-implanted for which the lattice damage has been completely healed, according to x-ray diffraction and Rutherford backscattering spectrometry measurements, by high-temperature, high-pressure annealing. This center is likely to be the isolated substitutional defect. It lacks a “subgap” excitation band and therefore has no state in the GaN band gap, shows threefold splitting of its level, with two sublevels nearly degenerate, and exhibits a long, single-exponential luminescence decay. Competing luminescent centers of GaN:Eu involve this prime center with intrinsic lattice defects, one of which may also be responsible for the GaN yellow band.
Optical detection of magnetic resonance is reported for the 0.97-eV luminescence in neutron-irradiated silicon. The resonance is of an excited triplet () state of the defect, which is not the radiative state, known to be a singlet (). The spectrum is unusual in that it is characteristic of a statically distorted defect (from to ), but with residual dynamic tunneling effects where random strain stabilizes mixtures of the static distortions. Vacancy-related models previously suggested for the defect are tentatively ruled out.
25 March 1991 in Journal of Physics: Condensed Matter
The structural properties of InGaN have attracted interest on account of the recent widespread use of the material in visible light-emitting devices. A key topic has been the indirect determination of the composition by x-ray diffraction (XRD). We examine critically the several levels of approximation involved in this procedure. It is shown by extended x-ray absorption fine structure (EXAFS) measurements that the local structure of InGaN is independent of the composition, in the range of InN fraction, from about 15 to 40%, that corresponds to blue to infrared light emission from this material. EXAFS-determined ratios of the numbers of indium and gallium atoms in the first metal co-ordination shell, M1, show very good agreement with the composition measured by established techniques, both structural and chemical, on similar samples. On the other hand, the atomic separations deviate markedly from values calculated using Vegard's law. In particular, the average radial separations, In-N1=2.11(2) Å and In-M1=3.28(3) Å, do not vary significantly with In/Ga ratio in the examined composition range. We conclude with some brief comments on the uncertain but challenging topic of InGaN nanostructure.
Europium was implanted into through a thick epitaxially grown layer that protects the surface during the implantation and also serves as a capping layer during the subsequent furnace annealing. Employing this layer prevents the formation of an amorphous surface layer during the implantation. Furthermore, no dissociation of the crystal was observed by Rutherford backscattering and channeling measurements for annealing temperatures up to . Remarkably, the intensity of the Eu related luminescence, as measured by cathodoluminescence at room temperature, increases by one order of magnitude within the studied annealing range between 1100 and .