Abstract:
We present a simple new background modeling method particularly for the determination of trace concentrations of Pb, U and Th to date zircons chemically by electron microprobe. This method reduces the acquisition time considerably, and at the same time increases the precision of a single Pb, U and Th analysis. Depending on the analytical conditions, theoretical detection limits down to 45 ppm (99% confidence level) for U, Th and Pb could be obtained. The final precision on age for a single measurement was between +/-8 and +/-35% (1σ) at Pb concentrations between 100 and 200 ppm in Proterozoic zircons. Precise U, Th and Pb analyses, however, do not necessarily facilitate a straightforward chemical age calculation. U-Th-total Pb data of zircons from four granites could not be objectively interpreted because of a severe disturbance of the U-Th-Pb system. However, it is shown that chemical ages derived only from zircon areas containing less than ~0.2 wt.% CaO agree very well with published conventional radiometric ages of these rocks. We propose a model which explains the Ca gain and Pb loss of zircons which have not undergone any metamorphism after emplacement to be the result of a hydration of radiation damaged zircon domains through low temperature aqueous solutions. It is suggested that the Ca content can be used independently from U, Th and Pb analyses as a chemical criterion to discriminate between altered zircon domains, having suffered a low temperature fluid-induced Pb-loss, and non-altered zircon domains for which there is a high probability that they will give concordant U-Th-Pb ages.