VIS-NIR and Raman analyses of acid altered volcanic rocks

Introduction

Alteration of mafic and ultramafic rocks on Mars surface and shallow subsurface has been postulated to have occurred throughout its history by means of different mechanisms, among which acid groundwater circulation (mainly sulfuric) is one of the most important [1,2,3,4]. This hypothesis is based on the high Fe and S concentration observed at various landing sites and strengthened by remote-sensing infrared detection of Fe/Al-bearing sulfates such as alunite, jarosite and Fe³⁺SO₄(OH) [4,5]. These sulfates typically form as alteration of K-rich volcanic rocks and Fe-sulfides in low-pH environments. Additionally the occurrence of perchlorates could be related to the action of other acids (perchloric) [6]. The investigation in laboratory of acidic alteration of volcanic rocks is thus an essential step in order to provide constraints in the interpretation of remote-sensing and in-situ data from Mars missions of the next future (Mars-2020, ExoMars-2022). Several works recently have explored the processes of acid alteration of minerals and rocks both in the field and in laboratory by using different spectroscopic and microscopy techniques [7,8].

Methods

In our work we studied by Visible-Near Infrared reflectance and Raman spectroscopy the acid alteration of two volcanic samples, a basalt from Aeolian Islands (FCD1) and a rhyodacite from Alps (RDO). These two samples have been chosen with the aim of investigating the action of acids on both a mafic and a felsic rock. Four acids have been used for the treatment of samples, namely hydrochloric (HCl 37 vol%), nitric (HNO₃ 65 vol%), sulfuric (H₂SO₄ 96 vol%) and phosphoric (H₃PO₄ 85 vol%), in order to contemplate a diversity of acid environments. Samples were acid-treated both in the form of fine powders (d<50 μm) and slabs. For each of the two volcanic samples four aliquots of powders were produced, with the aim of allowing different durations of alteration, and kept in solution for 1 h, 24 h, 7 days and 6 months respectively. At the end of each time slot the aliquot was centrifuged at 9500 rpm for 5 minutes, the excess liquid removed, and dried at 70°C in desiccator prior of measurements. The rocks slabs were instead kept in contact with acid for a few days and subsequently dried.

VIS-NIR reflectance spectroscopy measurements were carried on with ASD FieldSpec-4 spectro-photometer coupled with a QTH lamp and optical fibers (i=30°, e=0°), in the 0.35-2.5-μm range, using LabSphere Spectralon as reference. The spectral sampling was 3-8 nm, and the spatial resolution about 6 mm.

Raman spectroscopy data were acquired with a Bruker-Senterra II spectrometer, with a 532 nm excitation laser, a power of 50 mW, 4 cm^-1 spectral resolution, and about 5 mm of spatial resolution. Raman data were acquired on pre- and post-treatment slab samples.

Results

VIS-NIR. We present preliminary results of our experiments. Example of VIS-NIR spectra are shown in fig.1 for the basalt FCD1. The spectrum of pre-treated basalt (black curve, fig.1) is characterized by iron absorption features at 0.52 μm and 1 μm (faint), with a blue slope below 1.5 μm and a red slope above 1.5 μm. Although the effect due to some hydration is unavoidably present in the treatment, nevertheless the different acids produce different results. The overall spectral shape does not change much after treatment with HCl and HNO₃, although new absorption features appear or increase in intensity. In the HCl-treated sample the 1-μm band is notably more marked than the pre-existing one, and new features appear at 1.44, 1.60, 1.75, 1.93, 2.15 and 2.24 μm. Although especially the ~1.4 and ~1.9-μm bands are related to hydration in the sample or to some new hydrated phase, nevertheless these new bands indicate the appearance of new mineralogical phases. For example some of these bands are reported to occur in hydrated Na-perchlorates [9]. Curiously in the HCl-treated sample the ~0.5-μm feature seems to disappear. In the H₂SO₄-treated sample the overall spectral slope is changed from red to blue, and only two hydration bands appear. In the HNO₃-treated sample a new weak band appeared around 2.23 μm.

Raman. The Raman spectra of an alkali-feldspar phenocryst of the rhyodacite sample (RDO) collected pre- and post-treatment with HNO₃ are displayed in fig.2 (black and red, respectively). The spectrum of the untreated sample is characterized by the typical Raman peaks of the alkali-feldspar. On the contrary, in the spectrum of the acid-treated sample, the appearance of the peaks at 1053 and 715 cm^-1 indicate the formation of niter (NaNO3) in the same areas of the sample where the alkali-feldspar is present.

 

         

  

Conclusions

Visible-infrared and Raman analyses of both mafic and felsic volcanic rocks treated with different acids have shown the appearance of new absorption bands and changes in spectral shapes, indicating the formation of new mineralogical phases. While Raman data clearly indicate the presence of precise minerals, the interpretation of VIS-NIR spectra is not straightforward. Further analyses will be conducted both on the powders and slabs in order to recognize the new mineralogical phases, to discriminate between the alteration action of acid solution and water alone, and to disentangle the effects of the different acids over time.

Acknowledgements

This work is funded by ASI. We thank A.Pisello from University of Perugia for providing the rhyodacite sample.

References.

[1] Burns R.G., Journal of Geophysical Research, vol.92, n.B4, pp.E570-E574, 1987

[2] Burns R.G., 19th Lunar and Planetary Science Conference, 46-47, 1988

[3] Burns R.G. and Fisher D.S., Journal of Geophysical Research, vol.95, n.B9, pp.14,415-14,421, 1990

[4] Hurowitz J.A. and McLennan S.M., Earth and Planetary Science Letters, 260, 432, 443, 2007

[5] Ehlmann B.L. and Edwards C.S., Annu. Rev. Earth Planet. Sci., 42, 291-315, 2014

[6] Wilson E.H., Journal of Geophysical Research:Planets, 121, pp.1472-1487, 2016

[7] Gurgurewicz J. et al., Planetary and Space Science, 119, 137-154, 2015

[8] Smith R. et al., Journal of Geophysical Research:Planets, 122, pp.203-227, 2017

[9] Hanley J. et al., Journal of Geophysical Research:Planets, 119, pp.2370-2377, 2014