Zircon is an important mineral for understanding the evolution of the continental crust through time. However, advancement in zircon-melt partitioning studies has been restricted by the inability to grow zircons of sufficient size for analysis by LA-ICP-MS, the analytical method required to measure 1 ppb – 1000 ppm trace element levels. Multiple methods have been adopted to increase crystal size but with limited success. Despite this shortcoming, it is still possible to determine trace element levels in zircons < 20 μm in size by LA-ICP-MS. Partitioning experiments contain two end-member components, namely zircon and a silicate melt. If the melt end-member can be accurately determined, then the zircon end-member can be revealed by regression using analyses, which contain a mixture of both melt and zircon.
Here we present this regression method, taking the effects of spot size, ablation rate and sector zoning in zircon into account, with the aim to predict trace element concentrations in zircons from a series of experiments to determine the effect of melt composition, temperature, pressure and fO2 on partitioning.