Curie temperatures of titanomagnetite in ignimbrites: Effects of emplacement temperatures, cooling rates, exsolution, and cation ordering

Abstract

Pumices, ashes, and tuffs from Mt. St. Helens and from Novarupta contain two principal forms of titanomagnetite: homogeneous grains with Curie temperatures in the range 350-500 C and oxyexsolved grains with similar bulk composition, containing ilmenite lamellae and having Curie temperatures above 500 C. Thermomagnetic analyses and isothermal annealing experiments in combination with stratigraphic settings and thermal models show that emplacement temperatures and cooling history may have affected the relative proportions of homogeneous and exsolved grains and have clearly had a strong influence on the Curie temperature of the homogeneous phase. The exsolved grains are most common where emplacement temperatures exceeded 600 C, and in laboratory experiments, heating to over 600 C in air causes the homogeneous titanomagnetites to oxyexsolve rapidly. Where emplacement temperatures were lower, Curie temperatures of the homogeneous grains are systematically related to overburden thickness and cooling timescales, and thermomagnetic curves are generally irreversible, with lower Curie temperatures measured during cooling, but little or no change is observed in room temperature susceptibility. We interpret this irreversible behavior as reflecting variations in the degree of cation ordering in the titanomagnetites, although we cannot conclusively rule out an alternative interpretation involving fine-scale subsolvus unmixing. Shortrange ordering within the octahedral sites may play a key role in the observed phenomena. Changes in the Curie temperature have important implications for the acquisition, stabilization, and retention of natural remanence and may in some cases enable quantification of the emplacement temperatures or cooling rates of volcanic units containing homogeneous titanomagnetites.

Assessing New and Old Methods in Paleomagnetic Paleothermometry: A Test Case at Mt. St. Helens, USA

“…As demonstrated in Jackson and Bowles (2014) and discussed above, Curie temperatures at sites MSH12-08 and MSH14-08 systematically increase with depth in the deposit (Figure 3i) and this is shown to be linked to a thermally activated reordering phenomenon. As demonstrated in Jackson and Bowles (2014) and discussed above, Curie temperatures at sites MSH12-08 and MSH14-08 systematically increase with depth in the deposit (Figure 3i) and this is shown to be linked to a thermally activated reordering phenomenon.…”

“…It may be especially useful in lithic-poor deposits or to supplement data from thermal demagnetization of lithics. This involves isothermal annealing experiments, similar to those reported on in Jackson and Bowles (2014). The technique works best when the matrix has a relatively simple magnetic mineralogy, and one of the components must be homogeneous intermediate-composition titanomagnetite.…”

“…See Bowles et al (2013), Jackson and Bowles (2014), and Jackson and Bowles (2018) for a more detailed discussion of the reordering process.…”

“…The selection of this interval is admittedly somewhat subjective, but we take T dep-min as the last temperature that is clearly part of the low-T trend and T dep-max as the first temperature clearly part of the high-T trend. Ash data from site 8 (MSH12-08) are from Jackson and Bowles (2014). Where only one component is present and is not parallel to the 1980 field, we take T dep-max as the Figure 2.…”

Effects of titanomagnetite reordering processes on thermal demagnetization and paleointensity experiments

“…This phenomenon has now been documented in historical pyroclastic deposits from Mt. St. Helens (Washington State), Novarupta (Alaska), Lascar (Chile), and Soufrière Hills (Montserrat); historical extrusive basaltic lava flows from Fogo; ∼15 Ma Columbia River basalt feeder dikes; the 1100 Ma Duluth intrusive complex; and synthetic titanomagnetites [ Jackson and Bowles , ; Lappe et al ., ]. All these samples contain titanomagnetite with moderate amounts of titanium substitution ( x approximately 0.2 – 0.5) and frequently with small amounts of additional Mg and Al substitution.…”