Science Programmes in Mineral Sciences
The study of displacive phase transitions is one of the central themes of our research. These phase transitions involve small symmetry-breaking displacements of atoms, as seen in the example on the right. The associated change in energy is small, and is accompanied by small changes in the phonon frequencies that gives rise to a change in entropy. Many displacive phase transitions can be understood within the context of the 'soft mode' model.
Displacive phase transitions are studied using a broad range of experimental techniques. Diffraction (x-ray or neutron) gives information about changes in lattice parameters and hence spontaneous strains, and about changes in structure. Total scattering studies provide information about the relationship between long-range and short-range order. Spectroscopy provides direct information about changes in vibrational frequencies. Additional information can be provided by the others probes used in the group.
The starting point for theoretical understanding is the idea of the order parameter. This describes the set of symmetry-breaking distortions, and can be measured directly through measurements of the crystal structure, or indirectly through measurements of lattice parameters or vibrational frequencies. For many displacive phase transitions we have investigated, the equilibrium behaviour of the order parameter can be described using Landau theory. The concept of the order parameter allows a description of the phase transition that is applicable to any of the experimental techniques applied to the study.
A significant advance in our understanding of displacive phase transitions in framework structures, such as found in many silicates, came with the development of the 'Rigid Unit Mode' model. This enabled the basic structural instabilities associated with displacive phase transitions to be understood, and have given a new understanding of the nature of high-temperature phases.
Related to our study of displacive phase transitions is the issue of the structure of high-temperature phases. Frequently these can have considerable structural disorder, and the nature of this disorder is studied using a mixture of experimental and computational tools, combined with our theory of Rigid Unit Modes (RUMs).
One of the major experimental tools for the study of disordered phases is neutron total scattering, with the data analysed using the Reverse Monte Carlo method. This approach has enabled us to show that the high-temperature phases of a number of silicate materials contain structural disorder associated with the excitation of a large number of RUMs, giving the link back to the theoretical efforts to better understand the nature of high-temperature disordered phases.
The main computational tool is molecular dynamics simulation. This is ideal for disordered systems since it captures the dynamic behaviour without the restrictions of the harmonic (small-amplitude) approximation. We find that there is good agreement between these simulations and RMC models. The molecular dynamics simulations are often used to produce animations of the dynamic nature of the disorder.
|Configuration of the disordered phase of cristobalite obtained by Reverse Monte Carlo modelling of neutron total scattering data. Similar pictures are obtained using molecular dynamics simulations.||The two phases of quartz, showing the rotations of the SiO4 tetrahedra that lead to changes in symmetry without a large-scale rearrangement of the structure. Quartz serves as a prototype system for testing many ideas, such as our Rigid Unit Mode theory.|
The study of glasses has recently become a central theme in our research programme. There are two distinct aspects to this, which involve both experimental and computational work.
The first component of our glass work concerns the low-energy (0–5 meV) vibrational properties of silicate network glasses, which we study using computational methods and inelastic neutron scattering.
The first issue concerns the so-called Boson peak, which is a peak seen in the vibrational spectra of glasses with an energy of around 5 meV. Our inelastic scattering data have shown that there is also a Boson peak in related crystals, and that it is not a peculiar property of glasses. The second issue concerns the excitations in the energy range 0–5 meV. We have shown that these excitations are closely related to the Rigid Unit Modes of related crystal structures, and that the glasses have similar network flexibility to glasses. The third issue concerns very low-energy tunnelling states. We have observed candidate tunnelling events involving rotations of groups of SiO4 tetrahedra using molecular dynamics simulations. Many of these results have been reproduced in our experimental and simulation studies of a range of network silicate glasses.
We are now turning our attention to the effects of pressure on the low-energy excitations.
The second component to our work on glasses concerns the formation of silicate glasses.
The study of cation ordering processes is another of our central themes. Many minerals and ceramics contain different cations in similar site environments, and it is often found that the positions of these cations become disordered upon heating above a well-defined transition temperature. One common example is Al/Si ordering on tetrahedral sites in silicates, but we also study other types of ordering such as Mg/Ca ordering in garnets. In some cases, such as spinels, the cations order over qualitatively different sites (such as tetrahedral and octahedral as in spinel), and the disordering process does not give rise to a change in symmetry and hence no phase transition. However, in such cases, it is found that it is possible to obtain a good understanding of the ordering processes using tools developed to understand ordering phase transitions.
The main experimental techniques is diffraction, in particular structural studies to determine the state of long-range order at any temperature. Another useful approach is NMR spectroscopy, which provides a probe or short-range order. Experimental techniques are supported by computational studies. These use a mixture of empirical and quantum mechanical representations of the forces between atoms to determine the energies associated with ordering processes, and
|Crystal structure of garnet. The yellow spheres represent the atoms on the dodecahedral sites. We have studied solid solutions with chemical variation on these sites. The important interactions between these sites are in the horizontal (and symmetrically related) directions, and involve distortions of the intermediate SiO4 tetrahedra (darker blue polyhedra).||Needle domains in CaTiO3 perovskite. The origin of the needle shapes has been studied using theory and computational tools.|
When a material undergoes a phase transition, it breaks up into symmetrically-related domains. We have developed new theoretical and computational models to study the process of formation of domains, and new diffraction techniques to study the size of the boundaries between domains (domain walls) as a function of temperature. Our recent work has focused on the transport of cations along domain walls.
Minerals such as zircon and titanite are known to undergo amorphization as a consequence of the structural damage caused by alpha-decay of radioactive impurities (mostly uranium and thorium) that they contain. Understanding the self-radiation damage observed in these minerals is important, as they can provide fundamental data for the confirmation of extrapolated behaviour in nuclear waste forms, over long periods of time. We have been studying the impact of radiation damage on natural materials using a mixture of experimental and computational techniques.
Our experimental approach is to study the fundamental aspects of the amorphization process in natural sample, and characterise the structural changes involved during the radiation-induced transformation. By using techniques such as X-rays diffraction, Raman & IR spectroscopy, and NMR, we have been able to determine, for example, the number of atoms displaced per recoil nucleus and the volume of amorphous fraction as a function of the degree of damage.
Computational methods have been used to investigate the microscopic changes in the crystal structure the follow from one or a few recoil events. One of the most significant findings has been the observation of the polymerisation of the silicon and oxygen atoms in zircon following the original impact made by the recoiling atom.
It is hoped that a better understanding of the way that the structures of natural materials such as zircon and titanite will facilitate development of better ceramics for long-term storage of radioactive waste.
The study of surfaces is a growing area of activity within the group, with a wide range of projects.
Surface relaxation is an important component of the energy of a free surface. Because the forces acting on the surface layers are different from those within the bulk, they are able to lower their energy by being displaced away from their normal positions. The displacement patterns can be oscillatory, and will decay over a length scale of several unit cells below the surface.
We have started a project to study the adsorption of cations and organic molecules onto mineral surfaces using quantum mechanical calculations with localised basis sets. We have tested this approach by performing a series of calculations of the important bio-molecule adenine adsorbed onto graphite surfaces, because there is a good set of experimental data. We have computed an ordered arrangement of molecules on the surface that agrees with electron diffraction data, and the calculated energy of adsorption of single molecules is in agreement with thermal measurements.
We have been using this approach to study the adsorption of chlorinated organic molecules onto clay surfaces. The calculations have shown that the binding is stronger on the clay edges rather than on the flat layer surfaces. This work is leading to the calculation of binding of industrial waste molecules such as CFC's and dioxins.
These calculations are being extended to include water molecules. Similar calculations are now being carried out on carbonate surfaces.
For many geological applications it is important to understand the behaviour of materials at high pressures. Many minerals undergo phase transitions on increasing pressure, but the effects of pressure are quite distinct from those of temperature because an increase in pressure generates a much larger change in volume than does changes in temperature. We have developed facilities to study the behaviour of minerals under simultaneous high temperatures and pressures.
One of our major projects has been to develop neutron diffraction methods to determine the crystal structures of minerals at high T/P. We have developed an opposed-anvil press that is optimised for neutron diffraction at the ISIS spallation source. This is able to reach pressures of 6 GPa and temperatures in excess of 1500 K. Temperatures are measured using a novel radiographic method. This equipment is used to study the behaviour of hydrogenous minerals at temperatures and pressures of the inner earth, to study structural phase transitions, and to determine the pressure-dependence of cation-ordering processes.
We are also equipped with diamond anvil cells with heating capabilities for operation on laboratory X-ray sources.
|Adsorption of a molecule of carbon tetrachloride onto the later surface of pyrophyllite, as calculated using quantum mechanics methods.||Crystal structure of brucite, Mg(OH)2, which we have studied using neutron powder diffraction at high pressures and temperatures simultaneously (yellow represents Mg, red represents O and pink represents H). We have found that the best agreement with data is obtained when modelling the hydrogen atoms using a distribution of positions.|
Zeolites are silicates with pores built into their crystal structures. These pores usually contain water and extraneous cations, which are highly mobile. We have exploited this feature in developing the use of the common natural zeolite clinoptilolite as a bio-fertlizer.
When clinoptilolite is added to organic waste and composited, it absorbs ammonia ions. Then when clinoptilolite is introduced into soil, the ammonia ions are released. These act as a nutrient to nitrifying bacteria, which multiply at a very fast rate, supplying both nitrate and a source of free protons from the ensuing enzyme reactions. Plants growing in such an environment take direct advantage of the nitrate supply and the proton activity, in dissociating cations from the soil, provides additional plant nutrients.
Current laboratory studies are showing that enhanced plant growth in toxic soils polluted with heavy metals can be sustained by amending the substrate with the zeolite bio-fertilizer. This has a direct application in the restoration of toxic mine and metallurgical waste sites.
In the Landau theory of phase transitions, the size of the distortion of the structure is characterised by a quantity called the order parameter, symbol Q, and the free energy is expanded as a power series of Q. We have carried out detailed calculations in which we allow a coupling between the strain variables and Q.
The pressure-induced tetragonal–orthorhombic phase transition in stishovite, SiO2, displays characteristic features of a second order transition driven by a soft optic mode. We have shown from the Landau free energy that the transition will be accompanied by large anomalies in the elastic constants, which will be large enough to modify the velocity of sound waves in the earth’s mantle if stishovite is present in any significant quantity.
A similar analysis has been carried out for the displacive phase transition in quartz. Excellent agreement has been obtained with experimental data, thereby giving a complete account on the origin of the elastic anomalies associated with the phase transition.
|SEM image of a zeolitized volcanic glass shard. Well-formed crystals of clinoptilolite have grown from the clay coated rim of the shard inwards towards its centre.||(A) Calculated variation of the elastic constants of stishovite as a function of pressure. The large variation arises from the coupling with the order parameter. (B) Comparison of the calculated temperature-dependence of the elastic constants of quartz with experimental data. The theory provides an understanding of the elastic softening associated with the phase transition.|
Last updated on 08-Jul-10 16:48