The substitution of isotopes is well established in photoluminescence (PL) spectroscopy as a means of unambiguously determining the chemical identity of defect impurities. This technique is limited to cases where stable isotopes of probable defect constituents are available and where the associated line shifts are greater than the available resolution. Using radioactive isotopes the appearance or disappearance of a spectrum at a rate consistent with the specific radioactive half-life of the isotope can provide information on the chemical identity of defect impurities. Minimum detectable defect populations in PL are frequently low making it suitable where low concentrations or small volumes are involved, as is desirable while using radioactive species. A range of examples of this technique are reported. Firstly, Si was implanted with the radioactive isotope 111In (τ1/2)=2.8d) and treated to produce the intense PL characteristic of Si:In (labelled PQR). The luminescence intensity was found to decay non-exponentially and at a higher rate than expected for the 111In half-life, whereas the neutral In acceptor bound exciton luminescence intensity decayed as expected. New spectral features were found to appear after several half-lives, and are attributed to Cd the daughter atom of 111In. Preliminary results obtained for Hg implanted silicon are also reported together with data from As implanted GaAs, which are shown to be consistent with Hall and resistivity measurements. These results confirm the feasibility of using radioactive isotopes in PL spectroscopy, highlight some of the limitations of the technique, and illustrate the power of the technique for defect analysis.