Novel method cuts isotope production waste

Tc-99m has a short half-life so is produced from the radioactive decay of Mo-99 in so-called technetium generators which are supplied to the hospitals where the isotope is to be used. Most of the world's Mo-99 - which itself has a half-life of only 66 hours - is produced by irradiating uranium targets with a high-level neutron flux in research reactors.

The high-enriched uranium (HEU) traditionally used to fuel these research reactors, and used for the irradiation targets for Mo-99 production, is seen as a proliferation risk. Globally, efforts are being taken to minimise the use of HEU, with research reactors being converted to use low-enriched uranium (LEU) fuel and a drive towards the use of LEU targets. However, the lower the enrichment in the uranium target plate, the lower the specific yield of Mo-99 from irradiation, said Tobiaz Chemnitz of the MEDAPP medical irradiation facility at TUM's FRM II. This means that as least twice as many low-enriched uranium plates must be used to meet world demand for Mo-99, with a correspondingly higher volume of radioactive waste produced from processing the plates to recover the isotope.

The standard methods used to recover Mo-99 from irradiated plates involve treatment with an acid or alkali - typically sodium hydroxide - and result in large volumes of aqueous, intermediate-level radioactive waste. Since the replacement of HEU targets with LEU, this has led to the production of some 15,000 litres of aqueous wastes per year worldwide, which, once processed into a form suitable for final disposal, represent a final volume of 375,000 litres per year, according to TUM.

The new method involves reacting uranium-molybdenum test plates with nitrogen trifluoride in a plasma, before using a light-controlled reaction to separate excess uranium from molybdenum. This, say the researchers, is as efficient as the sodium hydroxide treatment in the first step of the conventional process, but without producing aqueous waste.

Chemnitz carried out the work for his doctoral thesis, working with colleague Riane Stene and in collaboration with fluorine chemists at Philipps University of Marburg. TUM has submitted a patent application for the process.