What is Magma Ascent Dynamics?

Understanding geological processes within heterogeneous materials is frequently a multi-scale problem. Petrological studies are frequently correlative, connecting field observations to hand samples and textural and geochemical data from entire rock scale and micro-to-nano scale investigations.

Correlative microscopy workflows are an essential component of these investigations, connecting multi-modal and multi-scale data and directing the research to design the most effective and efficient processes to characterize the sample.

A volcanic eruption is an excellent example of a multi-scale process that endangers both local and global communities. Knowing what transpires within magma as it ascends and erupts is critical to effective hazard assessment in these areas, and this is one of volcanology’s major challenges.

Ascending magmas exsolve gases, develop crystals, and are heterogeneously strained during volcanic eruptions. The resulting mixture of bubbles, melt, and crystals impacts and is impacted by the melt’s rheology. Finally, micro-scale interactions between these phases regulate an ascending magma's capability to outgas, determining the eruption's explosivity.

The sample in this article presents a basalt-rhyolite mixed fall from Ascension Island in the South Atlantic. A hot basaltic melt infringed on the rhyolite magma, causing a fast rise and eruption (Chamberlain et al., 2020). Carl Zeiss looks at micro-scale variations in vesicle shape and composition in 2D and 3D to figure out how and why bubble-crystal-melt interactions developed differently for eruptions with various starting conditions.

Sample Selection and Methodology

Clasts were chosen for their representativeness of field observations and significant textural qualities.

Clasts between 16 and 32 mm in diameter were chosen as they are less likely to have undergone post-eruption textural maturation (Shea et al., 2012) and are easier to examine using X-Ray computed micro-tomography.

Clasts were scanned for non-destructive 3D analysis using a ZEISS 620 Versa X-Ray microscope (XRM) at the Carl Zeiss X-Ray Microscopy facility in Pleasanton, CA. Backscattered electron (BSE) pictures were collected for each sample at the ZEISS facility in Cambridge, UK, using a ZEISS Sigma 300 VP.

To integrate the textural and geochemical analyses, a workflow (Figure 1) used non-destructive 3D XRM analysis, which was used to categorize samples and direct the selection of locations for thin sections.

These 2D thin sections helped validate 3D observations, create new data, and provide polished surfaces for in situ microanalysis using secondary ion mass spectrometry (SIMS) and electron probe microanalysis (EPMA).

Figure 1. Full correlative sample workflow for the project. Initial XRM scans highlight key areas for higher resolution imaging and target locations for thin section orientation within the volume. Subsequent 2D analysis includes electron and light microscopy, leading to correlation with in situ microanalytical data. Image Credit: Carl Zeiss Raw Materials

X-Ray Microscopy (XRM) Study

Two scans were carried out to pinpoint critical areas inside the sample, beginning with an initial “scout” of the entire clast with a voxel size of ~27 µm. The clast’s key regions of interest were then scanned with a “zoom” at a ~1 µm voxel size.

This two-tiered Scout-and-Zoom strategy, which uses a lower-resolution scan of the sample followed by a higher-resolution scan, provides optimal workflow efficiencies while integrating the entire sample context with critical sample information.

The low-resolution scan was analyzed to identify large-scale textural and compositional variations and “problem” regions within clasts that would be inappropriate for high-resolution imaging. This included regions protected by large crystals, which could generate additional noise in the high-resolution data.

High-resolution scans were acquired from analytically acceptable, texturally representative regions, while also ensuring that any features of interest, such as the boundaries between basalt and rhyolite compositions in mingling clasts, were captured.

To facilitate the comparison of features acquired by each scan type, high and low-resolution volumes were connected using image processing software (Avizo). Carl Zeiss also looked at the orientation of finer-scale fabrics disclosed by high-resolution imaging, such as vesicle elongation.

Greyscale Scout-and-Zoom volumes were then processed to obtain a variety of textural information. Simple thresholding algorithms may easily recover high-density items such as big crystals or plutonic clast inclusions, and their 3D volume can be displayed to indicate their locations in each sample.

The glass walls of the bubbles are an important aspect to examine because they define their size and shape. Scoria (basalt) clasts have thicker glass walls and less thin melt films, necessitating fewer unresolved melt film adjustments.

On the other hand, Pumice clasts often feature a more heterogeneous bubble population and plentiful thin melt films. Therefore, noise in the data must be examined more carefully to recreate broken or unresolved films while assuring the bubble population is appropriately represented (Figure 2).

Figure 2. Careful segmentation of 3D datasets allow for comparison of textures in regions of different composition. Statistically relevant datasets can be obtained by extracting bubble geometry in 3 dimensions without damaging the sample. Image Credit: Carl Zeiss Raw Materials

An iterative procedure of manual and algorithm-based segmentation was used to acquire the best results. Once bubble volumes have been properly specified, volume, shape, connectivity, and orientation data for hundreds of thousands of bubbles may be recovered, making this technique perfect for obtaining statistically robust textural data in 3D.

This non-destructive approach creates a 3D record of textural and, to a lesser extent, compositional heterogeneity inside clasts.

Thin Section SEM Imaging and Analysis

2D thin sections must be created to investigate crystal textures and perform in situ geochemical analyses of volatile, major, and trace element concentrations. It can be challenging to ensure that a sectioned plane accurately represents the interior structure of the clast.

Carl Zeiss was able to confirm that the sections did the following by starting with the 3D study (Figure 1) and using the 3D scans of each clast:

1. Selected a texturally representative area to sample
2. Sectioned some larger crystals for future study
3. The section plane was properly orientated in relation to the interior fabrics (depending on the desired outcome)
4. Allow for the comparison of 3D and 2D textural observations and analyses

This method provided more control over which textures were taken and evaluated in the thin slice. It also enables later testing and comparison of 2D versus 3D textural quantification methodologies, as the intersected region in 2D should nearly match the imaged region in 3D.

Samples were captured at various scales, with images obtained to capture specific textural regions of interest, such as phenocryst textures and compositional boundaries. Whole-section maps were created on which higher magnification and correlated image locations were registered, which was critical for ensuring a good spread of image collection and for future reference.

These 2D images were divided to remove bubbles from glass and crystals. Thin films, in particular, are frequently seen in BSE imagery. However, some may have been fractured during sample preparation or badly captured by the segmentation process (Figure 3).

Figure 3. A combination of electron and light microscopy means that the most delicate structures can be manually identified, improving the process of segmentation and the resulting analysis. Image Credit: Carl Zeiss Raw Materials

The user can manually recreate films using the BSE image as a reference. After thoroughly segmenting and reconstructing the films, the user can extract measurements, such as bubble perimeters, long and short axes of the best fit ellipse, orientation, and area.

These values can subsequently be utilized to compute textural descriptors, including regularity, circularity, and roundness, and to create vesicle size distributions (VSDs) using 2D–3D stereological conversion (Sahagian and Proussevitch, 1998).

Further Geochemical Analysis

The 3D and 2D textural characterization of bubble populations gives critical information on magma rheology during ascent. Beginning with XRM, the correlative technique provides a textural platform on which further geochemical measurements can be performed and directly compared to the textural information.

Several techniques, including energy dispersive spectroscopy (EDS), which may immediately determine mineral and glass compositions and serve as a reference for more extensive investigations by EPMA and SIMS, can be applied to 2D thin slices to yield critical geochemical information.

Secondary Ion Mass Spectrometry (SIMS)

Each section’s reflected light (RL) imagery was acquired, and relevant areas for SIMS analysis were selected and mapped. Since this is the view observed when utilizing the ion probe, reflected light imagery was used.

Comparisons of BSE and RL imagery made it easier to identify acceptable locations, such as microlite-free and across a range of textural properties. Crosscorrelating BSE and RL images allowed SIMS analysis locations to be precisely determined in the sample.

The Cameca IMS 4f ion Probe was then used to perform matrix glass SIMS investigations for H₂O, CO₂, Cl, and F at the NERC Ion Micro-Probe facility in Edinburgh.

Figure 4. Correlation of XRM and EM data with reflected light images of polished samples allows for easy identification of targets for microanalysis in systems (e.g., large sector SIMS) that use reflected light cameras for targeting. Image Credit: Carl Zeiss Raw Materials

Electron Microprobe Analysis (EMPA)

High-precision major and trace element analysis of matrix glasses and crystals can now be applied to every sample in regions of varying texture using the BSE and RL imagery as a reference. This enables the user to track compositional changes during each eruption and determine if crucial textural aspects are linked to micro-scale compositional variations.

Compositional analyses can also be used to reconstruct temperature and volatile content before an eruption using Fe-Ti thermometry and crystal-melt hygrometers (Ghiorso and Evans, 2008; Mollo et al., 2015). These investigations also reveal which crystal populations are equilibrated with the host melt.

This is significant when comparing pre- vs. syn-eruptive crystallization. Constricting these parameters is critical because they greatly impact the viscosity and outgassing of the magma.

Outcomes

This Ascension Island mingled fall example demonstrates how the methodologies above provide critical insights into factors regulating ascension dynamics and eruptive behavior.

The differing behavior of these melts was underlined by differences in the processing of XRM data required for basaltic vs. rhyolitic regions within the mingled clasts. While this difference was expected, it was not as prominent when comparing scoriaceous and pumaceous end-member clasts from this deposit.

VSDs for the two regions are anticipated to be only slightly different, with consequences for equilibrium vs. non-equilibrium degassing, supersaturation, and permeability during the eruption's dominantly mixed phase. VSDs from the early pumaceous and late scoriaceous phases will show distinct gas-melt-crystal interactions.

Figure 5. Contrasting textures of both the bubble size and density, as well as bubble wall (glass) thickness is readily highlighted by the correlative XRM-EM-LM workflow. Allowing for large accurate datasets to be used to interrogate rheological factors in eruption dynamics. Image Credit: Carl Zeiss Raw Materials

Summary

Petrological investigations in geosciences frequently entail a variety of methodologies that must be connected to yield relevant results. The article details how non-destructive 3D analysis can guide the project's next step, where 2D thin slices must be created for thorough observations and in situ microanalyses.

This coupled, multimodal, multi-scale strategy yields the most efficient geoscience research procedures. It is also important to note that it allows such processes to be adaptive, with observations at each stage of the process influencing exactly how the workflow develops, resulting in the most full and relevant datasets.

The research was carried out as part of the ZEISS-GeolSoc Ph.D. student scholarship awarded to Bridie Verity Davies.

References

1. Shea, T., Houghton, B.F., Gurioli, L., Cashman, K.V., Hammer, J.E. and Hobden, B.J., 2010. Textural studies of vesicles in volcanic rocks: an integrated methodology. Journal of Volcanology and Geothermal Research, 190(3-4), pp.271-289.
2. Sahagian, D.L., Proussevitch, A.A., 1998. 3D particle size distributions from 2D observations: stereology for natural applications. J. Volcanol. Geother. Res. 84, 173–196
3. Humphreys, M.C., Kearns, S.L. and Blundy, J.D., 2006. SIMS investigation of electron-beam damage to hydrous, rhyolitic glasses: Implications for melt inclusion analysis. American Mineralogist, 91(4), pp.667-679.
4. Ghiorso, M.S. and Evans, B.W., 2008. Thermodynamics of rhombohedral oxide solid solutions and a revision of the Fe-Ti two-oxide geothermometer and oxygen-barometer. American Journal of Science, 308(9), pp.957-1039.
5. Mollo, S., Masotta, M., Forni, F., Bachmann, O., De Astis, G., Moore, G. and Scarlato, P., 2015. A K-feldspar–liquid hygrometer specific to alkaline differentiated magmas. Chemical Geology, 392, pp.1-8.
6. Chamberlain, K. J., Barclay, J., Preece, K., Brown, R., & McIntosh, I. 2020. Deep and disturbed: conditions for formation and eruption of a mingled rhyolite at Ascension Island, south Atlantic. Volcanica, 3(1), 139-153.

This information has been sourced, reviewed and adapted from materials provided by Carl Zeiss Raw Materials.

For more information on this source, please visit Carl Zeiss Raw Materials.