A novel method for determining crystal orientation using polarised Fourier-Transform infrared spectroscopy

Quantifying hydrogen incorporated via cation vacancies in nominally anhydrous minerals (NAMs) is essential for assessing its influence on mantle physical processes such as rheology. In addition, hydrogen partitioning between NAMs and hydrous melts has been proposed as a potential mechanism to explain the oxidised nature of arc magmas. Fourier-transform infrared (FTIR) spectroscopy is the most widely used technique for quantifying trace hydroxyl (OH) in NAMs due to its accessibility and cost-effectiveness. Moreover, FTIR spectroscopy can also constrain hydrogen-incorporation mechanisms and the orientation of the OH dipole within charge-balanced vacancy structures. These distinctions are critical for identifying defects that buffer or sense oxygen fugacity. However, because of the relatively low hydrogen solubility and the diversity of incorporation mechanisms, optimising analytical resolution, both in concentration and OH-dipole orientation, remains highly desirable but has traditionally been hampered in most minerals by the intrinsic anisotropy of the OH dipole. Most studies rely on unpolarised FTIR measurements of randomly oriented crystals to avoid pre-orientation biases, but this practice reduces both precision and accuracy because birefringent crystals show strong orientation-dependent absorbance. Polarised FTIR overcomes this limitation and provides access to OH speciation and bond orientation unavailable from unpolarised measurements. We present a method to recover the full crystallographic orientation of a crystal, expressed as Euler angles, from polarised FTIR spectra acquired at different angles relative to the polarisation direction by rotating the polariser. The method requires a reference standard and a single diagnostic wavelength within the silica overtone range, which allows selection of an optimal frequency for orientation recovery and makes the approach more flexible than spectrum-range methods (Asimov et al., 2006). We validate the method using numerical simulations and a comprehensive dataset of olivine crystals with known orientations based on EBSD measurements. Implemented in an open-source Python package (FTIRkit), the approach applies to any birefringent crystalline phase and can be extended to other polarised techniques, including Raman spectroscopy and optical microscopy.

References

Asimow, P.D., 2006. Quantitative polarized infrared analysis of trace OH in populations of randomly oriented mineral grains. American Mineralogist 91, 278–284. https://doi.org/10.2138/am.2006.1937

Funding: This project has been funded through the ERC project OZ (DOI: 10.3030/101088573).