RISK CDT - Optimized sampling for advanced materials characterization

PLEASE APPLY ONLINE TO THE SCHOOL OF ENGINEERING, PROVIDING THE PROJECT TITLE, NAME OF THE PRIMARY SUPERVISOR AND SELECT THE PROGRAMME CODE "EGPR" (PHD - SCHOOL OF ENGINEERING)

This is a project within the multi-disciplinary EPSRC and ESRC Centre for Doctoral Training (CDT) on Quantification and Management of Risk & Uncertainty in Complex Systems & Environments, within the Institute for Risk and Uncertainty. The studentship is granted for 4 years and includes, in the first year, a Master in Decision Making under Risk & Uncertainty. The project includes extensive collaboration with prime industry to build an optimal basis for employability.

The development and application of new materials for clean energy technologies often takes 20+ years to move from the initial discovery to a first working commercial product. This delay is caused by a painstaking optimization process that requires different structures and compositions to be synthesized and characterized using state-of-the-art imaging and spectroscopic methods. Many of these methods involve the use of multi-million pound instrumentation where access is limited by overwhelming demand to only a few days per year. Complicating matters, the results of these short experiments using widely varying characterization tools can produce contradictory results as there are subtle factors (some defined by the expert operators) that are difficult to ascertain. To accelerate the time from discovery to applications it is therefore important to derive an optimal sampling strategy to maximize the amount of useful data obtained from each experiment and to define the metadata that allows comparison between the experimental methods.

The goal of this project is to develop this optimized approach to state-of-the-art materials characterization by obtaining images that contain precisely the minimum amount of data necessary to characterize a materials performance. This goal will be facilitated in this demonstration project by using sub-sampling, compressive sensing, and inpainting coupled with machine learning to understand the atomic scale properties of catalyst materials used to control emissions from combustion engines in cars. Typically aberration corrected Scanning Transmission Electron Microscopy (STEM) is used to obtain atomic resolution images of the Zeolite supported noble metal catalysts. From these images, the materials properties are derived from observing and then understanding how the structure (i.e. atom locations) change during the operation of the catalyst. To make observations at the atomic scale, these images by definition only sample a small area of the catalyst, making extrapolation to the larger scale difficult to do. Additionally, the observations themselves can change the structure if care is not taken to minimize the amount of electron dose used to form the images.

Experimental images will be obtained for this project using state-of-the-art scanning transmission electron microscopes at the University of Liverpool. Recent upgrades to the JEOL 21000 aberration corrected STEM in the NiCaL facility have permitted rapid sub-sampled images to be obtained that significantly lower the electron dose. Furthermore, these sub-sampled images allow the mathematics of image reconstruction and extension to larger scale variables in the sample to be determined directly. This experimental facility is currently unique in the UK. However, as part of this project the capabilities to obtain images of this kind will be applied to a set of samples provided by Johnson Matthey. After determining the reconstruction parameters and applicability of the reconstruction methods, the project will involve the PhD student spending extensive time at the Johnson-Matthey facility in the Diamond light source. The transition from a demonstration project to practical applications will allow Johnson-Matthey to apply these methods to their experimental microscopy analysis, and to the other characterization methods they use – leading to broad applicability of the capabilities being developed in this project to industrial developments.

Participants in the project will receive first hand training in the methods used for imaging by Johnson Matthey to characterize catalysts (and other materials) and develop a range of mathematical tools that will accelerate materials development processes across a wide range of practical engineering disciplines. There is a strong potential for collaboration with Johnson-Matthey after completion of the project, and opportunities to work with similar national/international research facilities around the world during the course of this research.