A Thermal Infrared Emission Spectral Morphology Study of Lizardite

Research into compositions of small bodies and planetary surfaces, such as asteroids, is key to understanding the origin of water and organics on Earth [1], as well as placing constraints on planetary dynamics and migration models [2] that can help understand how planetary systems around other stars may form and evolve. Compositional estimates can be found with thermal infrared (TIR; 5-25μm) spectroscopy, as the TIR region is rich in diagnostic information and can be used in remote sensing observations and laboratory measurements. However, TIR spectra of the same material may appear differently depending on several factors, such as particle size, surface roughness, porosity etc. This work quantifies the changes in spectral morphology (i.e., shapes and depths of spectral features) as particle size transitions from fine (<60μm) to coarse (≥60μm). 

The motivation for this work stems from TIR spectral modelling. The linear mixing model is most commonly used, and describes a mixture spectrum as being the linear sum of its individual component spectra, weighted by their abundances [3,4]. This works under the assumption that a single photon reaching the detector has interacted with a single particle. Whilst this is accurate for whole rock and coarse particulate samples, the assumption begins to break down as the particle size of the material decreases, approaching the photon wavelength, leading to an increase in multiple/volume scattering. This switch in scattering regime as particle size decreases has well-known effects on the morphology of TIR spectra, and is shown in Figure 1. Feature positions remain largely the same with changing particle size, making the effects seen attributable to the non-linear scattering behaviour. The main differences observed are in the Christiansen Feature (CF) “roll-off” region, on the shorter wavelength side of the CF, and the presence of the Transparency Feature (TF), also illustrated in Figure 1. 

Previous work into particle size effects in TIR spectra have largely been qualitative, or quantified different aspects, without a focus on improving spectral mixing models [5-8]. The work presented here will quantify the observed changes in the CF roll-off and TF, with the overarching aim to model these changes, using this to parameterise the linear mixing model, or derive an alternative method altogether. 

Presented will be the results for a serpentinite sample consisting primarily of lizardite (>90%), at several size fractions, aimed to be <5μm, 5-10μm, 10-15μm, 15-20μm, 20-25μm, 25-30μm, 30-35μm, 35-40μm, 40-45μm, 46-51μm, 51-55μm, 55-62μm, 62-73μm, 73-120μm, 120-200μm, 200-250μm, 250-350μm, and 350-500μm. These size fractions were separated out from the same bulk material, with those <45μm being separated via centrifugation and filtration, and those >46μm being separated via manual sieving, followed by rinsing. Each size fraction will also have detailed particle size distributions obtained through scanning electron microscopy (SEM), where the particles in the SEM images have been traced manually using the ImageJ software. The thermal infrared emission spectra were collected using the PASCALE instrument at the University of Oxford’s Planetary Spectroscopy Laboratory. 

With a clearer understanding of how TIR spectra evolve with changing particle size, the hope is that it will be possible to derive a model that more accurately estimates compositions of coarse and fine particulate materials (e.g., planetary regolith) from their TIR spectra. Having more accurate estimations of planetary surface compositions is vital for future planetary exploration and understanding the history of our solar system. 

References: 

[1] Morbidelli, A., et al. (2000). Meteoritics & Planetary Science 35.6, pp. 1309–1320. 

[2] Walsh, K. J., et al. (2012). Meteoritics & Planetary Science 47.12, pp. 1941–1947. 

[3] Lyon, R. J. P. (1964). NASA Contractor Report CR-100. 

[4] Ramsey, M. S. and Christensen, P. R. (1998). Journal of Geophysical Research: Solid Earth 103.B1, pp. 577–596. 

[5] Salisbury, J. W., Walter, L. S., and Vergo, N. (1987). Mid-Infrared (2.1-25 μm) Spectra of Minerals: First Edition. 

[6] Mustard, J. F. and Hays, J. E. (1997). Icarus 125.1, pp. 145–163 

[7] Ito, G., Arnold, J. A., and Glotch, T. D. (2017). Journal of Geophysical Research: Planets 122.5, pp. 822–838. 

[8] Shirley, K. and Glotch, T. (2019). Journal of Geophysical Research: Planets 124.4, pp. 970–988.