On the possible identification of defects using the autocorrelation function approach in double Doppler broadening of annihilation radiation spectroscopy

Abstract

The recent revived interest in the use of double-Doppler broadening of annihilation radiation (D-DBAR) spectroscopy, which employs two Ge detectors in back-to back geometry has stemmed mainly from its potential in defect identification as a result of its elemental sensitivity through core annihilations in atoms at the defect site. Emphasis has thus largely concentrated on the high momentum spectral range. In contrast the present work emphasizes the need to also focus attention on the low momentum region of the D-DBAr spectra. It is argued that the √2 improved resolving power of D-DBAR, in conjunction with spectral deconvolution, should give future ID (one dimensional) momentum data approaching in quality those obtainable using 1D-ACAR (angular correlation of annihilation radiation), thus formaing an alternative technique for observing the structure containing diffraction patterns that originate from annihilations with localized electron states at positron trapping defects. Rotation of the sample about a specified crystal axis, and the binning of events by angle, is suggested as a means of extending the technique to form a 2D- (two dimensional) DBAR counterpart to 2D-ACAR. The advantages of considering the real space positron electron wavefunction product AF (autocorrelation function), obtained by simple manipulation of the D-DBAR data in Fourier space, are outlined. In particular the possible visualization offered in real space of a defect's physical geometry, with the prospect of building up a library of contour patterns for future defect identification, is discussed, taking the silicon monovacancy in Si and the negative As vacancy in GaAs as examples.