Unmixing complex mineralogies / domain-states

We use first-order reversal curve (FORC) diagrams to understand mixtures of different minerals and different domain states present in a sample. This is done both with numerical models and experimentally on ocean sediments to understand how different minerals vary in response to environmental factors.

Recently, I found through simulations of mixtures of magnetic minerals (Berndt et al., 2017) that NRM vector demagnetization plots of mixtures can appear much more complex than one would intuitively expect.
Simulations of AF demagnetization plots of a mixture of magnetite and titanomagnetite. 
Simulations of NRM behaviour of mixtures of magnetite and titanomagnetite showed spurious components at intermediate temperatures or AF fields. Not being aware that this effect is purely due to the mineralogical mixture, the demagnetization plots would be incorrectly interpreted as having more components than they actually do.
The presence of different domain-states (single-domain, multi-domain, PSD), adds another dimension of complexity to the problem that we are currently working on simulating.
We have found that FORC diagrams can reveal not only the presence of different domain-states but also of different minerals. We aim to use this fact to quantitatively model FORC diagrams and use these to un-mix the signals of real samples.
Simulated FORC diagrams of a mixture of magnetite and titanomagnetite. Changing the timescale of FORC measurement (right) allows to unmix the different signals. 
In a later stage, the unmixed FORC diagrams will be used to predict remanence behaviour of samples. This way NRM demagnetization plots can be simulated for a sample and the various magnetic components can be uniquely attributed to their remanance carriers. This will hopefully provide a new and better way to interpret the different components visible in NRM demagnetization plots.
  • Berndt, T., Ramalho, R. S., Valdez-Grijalva, M. A., & Muxworthy, A. R. (2017). Paleomagnetic field reconstruction from mixtures of titanomagnetites. Earth and Planetary Science Letters, 465, 70–81. https://doi.org/10.1016/j.epsl.2017.02.033
%d bloggers like this: