Radiolysis reduction in liquid solution targets for the production of 89Zr.

Increased interest in radiometals for nuclear medicine and imaging can be hampered by radionuclide supply. 89Zr for example, is a PET imaging nuclide for which no radionuclide generator exists. One method to produce 89Zr involves irradiating aqueous solutions of yttrium nitrate salt on small medical cyclotrons. However, in irradiating these solutions the radiolysis of water can cause significant H2 and O2 gas buildup, which can eventually rupture a sealed target vessel. We examine the role of nitrate and nitrite in radiolysis. Here, we find that using copper-coated cadmium pellets to chemically reduce nitrate to nitrite in solution prior to irradiation can reduce in-target radiolysis by approximately 60% as compared to other published methods of radiolysis reduction, but only in acidic solutions. We hypothesize that during irradiation, nitrate is converted to nitrite, consuming free radicals which would otherwise be available to eliminate molecular gas species. Performing this conversion before irradiation may limit the consumption of these beneficial free radicals.

Understanding radionuclide production in gas target systems: the effect of adsorption on the target body.

Gas target systems have been used for decades on cyclotrons to produce radionuclides for medical imaging. However, the activity recovered from such targets is often lower than its theoretically predicted value. Past research has suggested that nuclide interactions with the walls of the target body may play a key role in the loss of recoverable radionuclide activity. Here, we consider gas targets and modify the standard radionuclide production equation by adding a loss term representing radionuclides depositing on the walls of the target. We derive the form of the deposition term based on a simple adsorption model which is then linearized by solving for leading order terms. The resulting production equation uses one fitting parameter to give an estimate of the recoverable activity produced in a target system, taking adsorption into account. The model is then fit to six data series, taken in-house and reported in the literature and a parity plot compares model predictions to experimental data. The model is able to better track the data than any previous models, and points towards a phenomenological understanding of adsorption in target systems.