Abstract Praseodymium-143 (T 1/2=13.57 d and E β(max)= 0.937 MeV) is the beta decay product of 143Ce (T 1/2=33.04 h) and would provide an ideal complement to the widely used 32P (T 1/2= 14.3 d, E β(max)=1.71 MeV), for some applications in radionuclide therapy (RNT), as it has a lower beta energy, but comparable half-life. For the production of 143Pr, 10-500 mg of two cerium compounds, ceric ammonium sulphate and ceric oxide, were irradiated in the reactor for 7 days at a neutron flux of ∼1×1013 n cm−2 s−1. After five days of post irradiation cooling, the samples were dissolved in appropriate reagents and aliquots drawn were subjected to high resolution gamma ray spectrometry for assay of activity and radionuclide purity. 7.4 MBq 143Ce per mg ceric oxide was produced under the aforesaid conditions along with other long-lived isotopes of cerium. An effort was made for the separation of no-carrier added (nca) 143Pr in order to study its utility for radionuclide therapy (RNT). A maximum yield of ∼450 MBq of 143Pr was obtained at calibration time defined as 7 d from end of irradiation (EOI). Hydroxyapatite (HA) was labeled with 143Pr at pH 5–6 and greater than 99.5% of 143Pr activity remained bound to HA even after 90 h post labeling. Despite the low cross-section of 142Ce(n,γ) reaction (0.1 barn) and isotopic abundance of 142Ce in natural targets (11.114%) being inherent disadvantages in large scale production of 143Pr, therapies needing low to medium doses would benefit from the use of ‘nca’ 143Pr.