ABSTRACTPhotoluminescence (PL) spectroscopy, cathodoluminescence (CL) spectroscopy and imaging, transmission electron microscopy (TEM), and preferential defect etching and optical microscopy have been used to characterise dislocations in plastically deformed Si and in Si1−x Gex alloys. The PL spectra from plastically deformed Si produced under clean conditions, contained no dislocation related luminescence features. However, after deliberated Cu contamination and annealing, the PL spectra were dominated by the four main D-bands (D1-D4). CL spectroscopy and imaging revealed that the D3 and D4 bands originate on the slip lines, whereas D1 and D2 are dominant between the slip lines. Pseudomorphic Si1−x Gex layers grown by molecular beam epitaxy (MBE) and forced to relax by Cu contamination and annealing have also been examined. Both the PL and CL spectra were dominated by the D1-D4 bands, and monochromatic CL imaging showed that the D3 and D4 bands originate at the misfit dislocations with the location of the Dl and D2 bands being much less well defined. PL spectra from Si1−x Gex capping layers, grown by MBE on linearly compositionally graded buffer layers, were dominated by bound exciton luminescence features, identical with those observed in bulk grown alloys, and showed relatively weak D-band features. PL spectra obtained after chemical step etching, and CL spectra obtained with different beam energies, showed that the D-bands originated in the buffer layers which contained high densities of misfit dislocations. Well defined monochromatic CL images of the misfit dislocations could only be obtained using a narrow band pass filter centred on the D4 band. No D-band luminescence was observed from Si1−x Gex layers grown by high temperature chemical vapour deposition (CVD). However, after deliberate contamination with Ni, the D-bands became the dominant features and CL imaging clearly showed that the D4 band originates at the misfit dislocations and the D2 band predominantly between the misfit dislocations.