Success of Rapid Continuous Thermal Demagnetization When Conventional Methods Failed

Wednesday, 17 December 2014: 2:25 PM
Robert S Coe, University of California, Santa Cruz, CA, United States and Maxime Le Goff, Institut de Physique du Globe, Équipe de Paléomagnétisme, Paris, France
Conventional stepwise thermal demagnetization of samples spanning a basalt flow erupted during a polarity transition at Steens Mountain, Oregon yielded scattered directions of high-temperature remanence, whereas the results of continuous thermal demagnetization cluster convincingly among the characteristic directions of the next several flows below. The continuous demagnetization was performed using the Triaxe1, a 3-axis vibrating sample magnetometer in which the directions of ~1 cmsub-samples were measured repeatedly as temperature increased during heated from 20 to 500-550°C in only 12-13 minutes. The demagnetization trajectories suggest that normal-polarity secondary magnetization, acquired both at room temperature in today’s polarity chron and during modest reheating in a normal field during cooling of the overlying flow, was responsible for the failure of conventional thermal demagnetization. Our favored explanation is that alteration during ordinary thermal demagnetization raised the blocking temperature while preserving the direction of the overprint, thereby masking the primary component. The rapid heating (~40°C/min) during continuous demagnetization appears to have been fast enough to demagnetize the normal overprint before this masking could happen. Thermomagnetic cycles exhibit significant irreversibility starting around 300°C, both in air and in argon. Changes in room-temperature hysteresis parameters after heating in air to temperature T also start to change around T=300°C. Titanomagnetite of composition TM65-70, partially oxidized to titanomaghemite, plus a minor low-Ti, oxyexsolved phase are observed in thin section and inferred from thermomagnetic curves. Thus, inversion of secondary titanomaghemite that carries a normal overprint could be the masking mechanism. The failure of AF demagnetization, on the other hand, we attribute to overlapping coercivity spectra of primary and secondary magnetization.

1Le Goff and Gallet, 2004, Earth Planet. Sci. Lett. 229, 31–43.