Alumino-silicate melts represent the majority of the Earth’s molten rocks, and are fundamental starting materials for glass products. Their mechanical properties underlie volcanic eruption and magma flow [1,2], while simultaneously delivering advanced materials such as the “gorilla glass” used in mobile phone technology . This ambitious, interdisciplinary project fuses key challenges in condensed matter physics such as crystal nucleation and the glass transition  with geological problems of understanding the behaviour of magma . Here we are particularly interested in the effect of crystallisation in magma: solid magma doesn’t flow. Even “mush”, magma with solid particulates, can undergo jamming of the particulates with spectacular consequences for volcanic eruptions, and may shed light on fundamental mysteries like violent volcanic eruptions after long periods of dormancy .
Molecular dynamics computer simulation where the forces between atoms are used to evolve the atomic positions is an ideal way to tackle such high pressures and temperatures as are present in volcanoes. Models of the forcefield between atoms reasonably reproduce the bulk properties of magmas based on for example Al, O, Si. P and N, in which we have experience . However conventional computer simulation struggles to access the large timescales relevant for viscous, glassy materials, such as cooling magma. A liquid is said to be a (solid) glass when the relaxation time of its constituent atoms reaches 100s, so the atoms move 14 orders of magnitude more slowly than in a normal liquid, where the relaxation time is around 1ps. Alas conventional state-of-the-art computer simulation can only reach around 10ns: a staggering 10,000,000,000 times less than that required for a system equilibrated required 100s.
We have recently broken this “glass ceiling” through a novel simulation technique based on trajectory sampling . Our work has shown that when a simulation of around 1ns is run very many times, some configurations turn out to be indistinguishable from those that would usually be reached after 100s. The question is how to select those configurations that correspond to 100s? The key is that atoms in viscous, glassy liquids organize into geometric motifs known as locally favoured structures (LFS). One example of such an LFS is the icosahedron, effectively a 3d pentagon. The slower, i.e. the more viscous, a liquid becomes, the more LFS it exhibits [4,6].
So by selecting sets of configurations of atoms which have many such locally favoured structures, we can effectively obtain states which have been equilibrated for many orders of magnitude more than is possible with conventional methods.
We have developed the method for simple atomistic models . Here we adapt the methodology to models more appropriate to magma . We shall identify the locally favoured structures in the model magma using our topological cluster classification methodology [3,5]. Once we have magma configurations which have been suitably equilibrated, we shall test its mechanical properties, such as the shear modulus, and yielding behaviour.
We will then consider the crystallisation of the magma. We expect that a subset of the magma mixture may crystallise, and we shall test our new “universal” mechanism for crystallisation in mixtures . We will combine high-temperature brute-force simulations with rare-event sampling methods to characterise the crystallisation behaviour, before moving on to consider the mechanical properties of the crystals. We expect that the crystalline material will have a very much higher shear modulus than that of the amorphous glassy magma, but how much so remains to be seen, and is of course all but impossible to determine in-situ experimentally.
This interdisciplinary project breaks new ground in many ways, in bringing together ideas from condensed matter physics with earth science, which has rarely been done to date. Over the course of the project, we expect many new and exciting avenues of enquiry. One intriguing question is the role of water in cooling magma and what effect it may have on the crystallisation mechanisms.
How to Apply
Please make an online application for this project at http://www.bris.ac.uk/pg-howtoapply
. Please select Physics PhD on the Programme Choice page. You will be prompted to enter details of this specific project in the ‘Research Details’ section of the form.
Anticipated start date: September 2019
A first degree in physics or a related subject, normally at a level equivalent to at least UK upper second-class honours, or a relevant postgraduate master's qualification.
See international equivalent qualifications on the International Office website.
 Cashman, K. V.; Sparks, R. S. J. & Blundy, J. D. “Vertically extensive and unstable magmatic systems: A unified view of igneous processes” Science 335 1280 (2017).
 G.N. Kilgour, G. N.; Mader, H. M.; Blundy, J. D. & Brooker, R. A. “Rheological controls on the eruption potential and style of an andesite volcano: A case study from Mt. Ruapehu, New Zealand” J. Volcan. Geothhermal Res. 327 273-287 (2016).
 Mauro, J. C.; Philip, C. S.; Vaughn, D. J. & Pambianchi, M. S. “Glass Science in the United States: Current Status and Future Directions” Int. J. Appl. Glass Sci. 5 2-15 (2014).
 Royall, C. P. & Williams, S. R. “The role of local structure in dynamical arrest” Phys. Rep. 560 1-75 (2015).
 Skinner, L. B.; Barnes, A. C.; Salmon, P. S.; Fischer, H. E.; Drewitt, J. W. E. & Honkimäki “Structure and triclustering in Ba-Al-O glass” Phys. Rev. B 85 064201 (2012).
 Turci, F.; Royall, C. P. & Speck, T. “Non-Equilibrium Phase Transition in an Atomistic Glassformer: the Connection to Thermodynamics” Phys. Rev. X 7 031028 (2017).
 Ingebrigtsen, T. S.; Dyre, J. C.; Schroder, T. B. & Royall, C. P. “Crystallisation Instability in Glassforming Mixtures” ArXiV 1804.01378 (2018).