Effect of low-temperature demagnetization on Thellier paleointensity determinations

Abstract:

We investigated the effect of low-temperature demagnetization (LTD) on paleointensity determinations using the double-heating Thellier method. Model experiments were conducted on synthetic samples containing single-domain (SD), pseudosingle-domain (PSD), or multidomain (MD) magnetite grains, as well as mixture of SD and MD magnetite. In order to model a natural remanent magnetization (NRM) before the experiments, a thermal remanent magnetization (TRM) was imparted in the samples in a 50 μT magnetic field. LTD was performed after each Thellier heating by three successive cycles of the samples through the Verwey transition temperature (~120 K) in a field-free environment. The magnetic remanence was measured both before and after each LTD. The effect of LTD on paleointensity determinations from SD samples was negligible; both pre-LTD and post-LTD data defined linear high-temperature (>300°C) segments on the Arai plots that yielded paleofield values statistically indistinguishable from the field of original TRM acquisition. The PSD and SD/MD mixture samples yielded concave-up Arai plots underestimating the laboratory field by 10-20 percent. The LTD treatment resulted in steeper and more linear high-temperature (>350°C) segments on the Arai plots giving accurate estimates of the TRM acquisition field. The Arai plots measured from MD samples were substantially non-linear and failed to provide acceptable paleointensity results both before and after LTD. Our experimental results indicate that the integration of LTD with the Thellier experimental protocol improves the accuracy and quality of paleointensity determinations from samples containing SD and PSD magnetite, but is ineffective for very large, MD grains.