Diffusion effects on oxygen isotope temperatures of slowly cooled igneous and metamorphic rocks

Abstract

Recently obtained data on oxygen diffusion in feldspars, quartz, and hornblende permit the prediction of the apparent 18O 16O temperatures that would be measured in a rock that consisted only of those three minerals, and cooled slowly from high temperature. The computed temperatures would be based on the differences in the 18O 16O ratios between coexisting pairs of minerals. The present calculation takes into account the diffusion rates for oxygen as a function of temperature, the cooling rate of the rock, the mineral grain sizes, and the mode of the rock. For mineral grains 1 mm in radius, and a cooling rate of 10°C/m.y., the minimum difference in apparent temperature between quartz-feldspar and feldspar-hornblende pairs will be 115°C, despite the assumption of a normal, uneventful, slow cooling history to room temperature. Further, the apparent quartz-hornblende temperature will range over 30°C (590–620°C) depending on the mode of the rock. For a cooling rate of 1000°C/m.y., the apparent difference in temperature can be as much as 400°C. Consequently, consistency in temperatures obtained by oxygen isotope analysis should not be expected in most high-grade metamorphic rocks or igneous rocks which are cooled slowly. Departures from the pattern of temperatures obtained in this model would imply a very rapid quench from high temperature, or a complex history for the rock. For some minerals, including hornblende, the relation between temperature and the equilibrium fractionation of oxygen isotopes between coexisting phases has been derived from observed relations in natural specimens. The choice of the specimens used for such calibrations needs to be re-evaluated in light of these findings. This may result in a change in the equilibrium equation constants. An example from the literature, the San Jose tonalite, Baja California, Mexico, was modelled and yields δ 18O concentrations in the minerals that correspond closely with the measured values. This suggests that the model used is appropriate, that the rock has had a simple thermal history, and that it cooled at 100–200°C/m.y. over the temperature range 800–500°C. The set of paleotemperatures obtained for a rock will, in general, yield neither the mineral closure temperatures nor the formation or crystallization temperatures. On the other hand, the cooling rate of the rock may be derived from the data. This, in turn, may have important tectonic implications with regard to denudation and uplift rates.