Join us

Dr Jianglin Huang

Supervisor

This project is part of cohort 3 of the EPSRC CDT in Developing National Capability for Materials 4.0, with the Henry Royce Institute. It will be a collaboration among the University of Strathclyde, the Advanced Forming Research Centre (AFRC), the National Manufacturing Institute Scotland (NMIS) and the Design, Manufacturing & Engineering Management (DMEM).

Titanium alloys are essential to aerospace applications but remain extremely challenging to forge due to their narrow processing windows, strong sensitivity to temperature gradients, and tendency to develop defects. Cogging is a critical hot-working process used to refine ingot microstructures and prepare billets of different sizes and shapes for downstream processing. The design of cogging schedules is currently highly empirical, relying on expert know-how and time-consuming trial-and-error. With the sector accelerating toward digital transformation, forging processes must evolve beyond traditional empirical practice. Digital twin technologies—integrating physics-based simulation, in-process sensing, and AI models—provide a transformative platform for understanding and optimising titanium cogging, enabling faster development cycles, improved quality, and data-driven manufacturing.

This PhD will tackle the challenge of digital transformation in forging by developing AI-driven surrogate models for titanium cogging and embedding them into the FutureForge digital twin at the National Manufacturing Institute Scotland (NMIS). FutureForge is one of the world’s most advanced and largest hot-forging research and innovation facilities, powered by digital data science. It offers a data-rich environment where the forging industry can de-risk the development of new processes, products, and technologies for faster industrial adoption.

The research will focus on:

• Developing and validating physics-based models for titanium cogging, generating datasets to understand how key process parameters and schedule designs influence deformation and microstructure.
• Training surrogate machine learning models to rapidly predict process outcomes such as temperature distribution, strain evolution and microstructural changes.
• Integrating these surrogate models into a digital twin framework to enable high-speed prediction, quick exploration of forging schedules, and optimisation under physical and operational constraints.
• Validating AI predictions against experimental titanium forging trials to ensure accuracy, robustness, and industrial relevance.
• Creating an automated decision-support and optimisation system that enables engineers to design and refine cogging schedules more efficiently and with greater confidence.

The vision is to establish an intelligent digital twin system through which engineers can evaluate, optimise, and adapt cogging schedules to accelerate process design, reduce development costs, and support the digital transformation of advanced metals processing. By creating an AI-empowered digital twin framework for cogging, this PhD will deliver both new scientific insights and transformative digital manufacturing tools for the aerospace metals industry.

Dr James McHugh

Supervisor

About the Project

Prof Ping Xiao

Supervisor

This PhD project will accelerate the development of novel environmental barrier coatings required for ceramic matrix composite components for the next generation of higher temperature aero-engines in collaboration with Rolls-Royce through machine learning. Silicon carbide-based ceramic matrix composites (SiC-SiC CMCs) are being developed to transform aeroengine design by replacing nickel-based superalloys in the hot section of an aero-engine, due to their superior high-temperature mechanical properties and lower density e.g. density of CMCs is about one third density of superalloys. The use of the lighter weight CMCs would allow engine operating at higher temperature, using less cooling air, and therefore fuel. However, environmental barrier coatings (EBCs) are required to protect SiC-SiC CMCs from steam corrosion in engine environment. Without EBCs, the underlying substrates rapidly degrade in the hostile gas-turbine environment, so the failure of the EBCs is life limiting for the entire component. To introduce EBCs into aero-engine service, extensive research on manufacture, characterization and testing of EBCs has been carried out over decades. A range of materials have been investigated, but the current state-of-the art EBCs are based on ytterbium disilicate deposited via the air plasma spraying (APS) process.

The PhD student will examine electrophoretic deposition (EPD) which is an alternative low cost and simple method alternative allowing easy control of the coating chemistry and microstructure. It can be applied to various substrate types of different sizes including rough surfaces such as SiC-SiC, or complex geometries, covering sharp corners or concave/convex surfaces. EPD of ytterbium disilicates has been developed by Prof Xiao’s group at University of Manchester to manufacture EBCs on SiC-SiC CMCs with better performance than APS produced coatings demonstrated. To optimise performance of the EBCs produced with use of EPD coatings, it is essential to establish relationships between processing parameters, microstructural attributes and properties/performance of the EBCs.

The PhD student will develop high throughput make-test and characterisation to explore various compositions and various powders supplied by different suppliers. This will open the way to a data centric approach. Fast characterisation and mechanical testing methods will be developed to enable our high throughput approach.

In tandem, the student will develop a digital twinning approach and use machine learning to establish the relation between EPD condition-microstructure-mechanical behaviour- performance of the coatings produced using EPD, and to point the way to better formulations and optimised processing at rates currently unachievable. In this way the student will accelerate the development of novel EBCs for this application and showcase how high throughput methods can be used to develop coatings across a wide range of coatings applications.

The successful candidate will collaborate closely with Rolls-Royce (the industrial sponsor) and will join a large research team based in the Henry Royce Institute, University of Manchester.

Prof Michael Shaver

Supervisor

This project is part of cohort 3 of the EPSRC CDT in Developing National Capability for Materials 4.0, with the Henry Royce Institute.

The studentship will develop machine learning methods to co-optimise environmental (life cycle assessment) and economic (return on investment) metrology to accelerate the circular economy for plastics and their multi-materials. While LCA & ROI are often used in materials innovation, laborious calculations prevent iteration. This studentship will develop AI-led multiparameter optimisations to apply these Materials 5.0 principles to both optimise waste fates (reuse, recycling, H₂) for plastics and predict waste composition for plastic, glass and aluminium. With potential impacts for our industry partner, Resource Futures, and in shaping recycling investment and policy interventions from local and national government, this project has potential to reshape circularity of materials.

We seek an enthusiastic candidate who wants to join an interdisciplinary team in the Sustainable Materials Innovation Hub. The project would be suitable for a range of academic backgrounds in either machine learning and computer science, sustainability metrics and life cycle assessment, or materials engineering. Purpose-driven researchers keen to learn and collaborate on global challenges as change-makers will benefit from this unique opportunity within the Materials 4.0 CDT.

Dr Alec Davis

Supervisor

Ti-6Al-4V (Ti64) jet-engine fan blades and discs fail through crack initiation and propagation during long-duration and cyclical loading. The fracture surfaces of such components often exhibit ‘facets’ that are indicative of localised rapid cleavage of the material, and important aspects of their formation are under current investigation. In particular, the influence of local microstructure and texture, and fatigue-loading frequency and stress on fracture-surface morphology are not understood well enough to enable rapid and straightforward failure investigations. It is therefore of key importance to develop tools for easy correlation between fracture morphologies, microstructure features, and specific loading conditions. This will enhance the efficiency of failure investigation lead time, and reduce aerospace-industry safety and risk-mitigation costs.

Facets and other such common surface morphologies induced by fatigue (e.g. surface striations) are readily recognisable in scanning electron microscopy (SEM) images, and even in optical microscopy images and topography scans should the features be large enough. However, due the complexity of fatigue plane geometry in 3-dimensional space and dissimilar length scales under consideration, it is often difficult to associate the features seen in these images to the precise time frame of bulk-material loading changes that occur during application, and even in lab-scale tests. Correlation of these aspects therefore requires much time and effort before the data can be applied to productive causation investigations. However, there is an opportunity to develop digital tools to correlate fatigue loading changes to fracture morphologies using SEM-image-based data input. In particular, a tool that can recognise facet morphologies in datasets consisting of many SEM images taken over a specific fracture surface area, and then automatically correlate these to known influences of heterogeneous, textured microstructures and mechanical loading changes would significantly accelerate the speed of failure analyses. In addition, the development of such tools would intrinsically mature the knowledge base of fatigue-loading condition and microstructure influence on fracture-surface morphology, and improve the fidelity of the causation investigations.

Dr Shijing Sun

Supervisor

In this project, the student will develop an autonomous experimental platform that integrates robotic synthesis, automated spin-coating, and AI-driven optimisation to discover and control next-generation metal halide perovskite semiconductors. Positioned at the interface of materials science, robotics, and machine learning, the project addresses a key bottleneck in AI-accelerated materials discovery: bridging the gap between AI-generated materials predictions and experimentally realisable, high-quality thin films for clean energy and optoelectronic applications.

Metal halide perovskites are leading candidates for future solar cells, light-emitting diodes, displays, lasers, and photodetectors due to their exceptional optoelectronic properties and compatibility with low-cost, solution-based processing. At the same time, advances in artificial intelligence have enabled the rapid generation of thousands of hypothetical perovskite and perovskite-inspired crystal structures with targeted stability or optoelectronic performance. However, many of these designs remain unrealised experimentally. This disconnect is particularly acute for solution-processed thin films fabricated via spin-coating, where small variations in processing parameters can lead to substantial changes in crystal phase, morphology, defect density, and functional properties.

This project provides training in autonomous laboratory methodologies designed to overcome these challenges. Building on developments in the Autonomous Materials Group at the University of Cambridge, the student will develop automated spin-coating workflows and integrate them into a high-throughput robotic platform for fully automated thin-film fabrication and characterisation. Optical and structural techniques, including photoluminescence (PL) spectroscopy and X-ray diffraction (XRD), will be incorporated to generate high-quality experimental data. These data will inform AI process models that learn processing–structure–property relationships and guide optimisation. The student will receive interdisciplinary training across materials science, automation, and machine learning, making this project well suited to candidates interested in combining experimental research with data-driven methods.

Dr Oriol Gavalda Diaz

Supervisor

The long-term performance of energy generation (e.g., nuclear plants) and transport applications (e.g., aviation) is underpinned by ceramic-based coatings and composites that can withstand demanding environments. For instance, thermal barrier coatings (TBCs), environmental barrier coatings (EBCs), and ceramic matrix composites (CMCs) are examples of ceramics which can continue to improve product performance and promote sustainability in the aero-engine industry. Given the harsh conditions, these materials may experience degradation through-life driven by complex combinations of thermal, mechanical, and chemical processes, often accelerated by contaminants such as CMAS (calcium–magnesium–alumino–silicates).

Rolls-Royce has identified Raman spectroscopy as a promising technology for non-destructive and potentially on-wing (without dismounting the aero-engine) inspection of such materials. However, the Raman spectra collected in-service are highly complex, and their interpretation requires new approaches that integrate databases of lab-based characterisation data and data-driven analysis protocols. The key challenge is linking spectral features to specific degradation mechanisms and relate these to the environmental degradation of the component. Moreover, a holistic view of the degradation of the material will require to link these degradation mechanisms to other sources of inspection data (e.g. advanced microscopy techniques or visual inspection).

This project will create a data-rich experimental and analytical framework to overcome these challenges. It will combine the following:

• Controlled laboratory experiments that simulate degradation mechanisms in ceramics, generating reference datasets of Raman spectra across different conditions (temperature, stresses, and contaminant exposure).
• Complementary nanoscale characterisation data (e.g., electron microscopy) to validate the key spectral signatures.
• Advanced data-analysis protocols to link lab data to real applications. Methods such as spectral decomposition, machine learning, and database-driven comparison will be applied to extract characteristic features from the Raman datasets. The goal is to create a systematic approach for data-informed inspection, in which spectra are not simply recorded but actively interpreted against a material database.

Rolls-Royce will provide real aero-engine samples to validate our laboratory-generated databases and analysis protocols. Hence, the outputs of the project will directly support the implementation of in-situ/on-wing inspection tools by Rolls-Royce, reducing reliance on destructive testing, enabling predictive maintenance strategies and extending the lifetime of components.

Dr Abigail Ackerman

Supervisor

​Low temperature electrolysis is an emerging technology that has the potential to revolutionise the ironmaking industry. Iron and steel production currently accounts for around 8% of global CO₂ emissions, with approximately 70% of these emissions resulting from the reduction of iron ore (naturally found as Fe₂O₃). The dominant reduction method—carbon-based blast furnace processing at ~2100 °C—releases oxygen from the ore, which bonds with carbon to form CO₂. On average, producing one tonne of iron this way emits about 1.8 tonnes of CO₂.

​​Low-temperature electrolysis offers a promising alternative. Over the past decade, this experimental method has shown potential to significantly lower both carbon emissions and energy demand compared with other green technologies such as Direct Reduced Iron (DRI), which uses hydrogen as a reducing agent. Unlike many existing approaches, this method produces only oxygen as a by-product and does not require high-purity ore. This makes it particularly suitable for processing lower-grade iron ores and recovering iron from waste streams such as mine water and tailings.

​An important yet underexplored aspect is the application of digital twins to remote sensing of electrolyser cells. In multi-electrolyser systems, digital twins are critical for balancing power input across cells and for predicting component degradation under demanding operating conditions. This project will build on existing digital twin blueprints for electrolysers by developing a robust sensor system that feeds real-time data into a digital twin, designed and implemented by the student. This capability will be central to ongoing upscaling efforts within both the supervisor’s research group and the project’s industrial sponsor.

Dr Jingwen Weng

Supervisor

Battery material performance, lifetime and safety are strongly influenced by gas evolution arising from electrolyte decomposition and interfacial reactions. Gas accumulation can trigger swelling, venting and thermally driven material failure. However, the mechanistic links between gas evolution, thermal effects and battery material degradation remain unclear due to the lack of analytical platforms capable of capturing gas release with high temporal resolution. Recent progress in on-chip electrochemical mass spectrometry (EC-MS) enables operando quantification of gas evolution with picomole-per-second sensitivity, providing a powerful route to uncover how gas signatures reflect interfacial reactions and degradation pathways. Yet this opportunity has not been translated into a generalisable Materials 4.0 capability integrating multi-modal sensing, real-time interpretation and structured data for mechanistic discovery.

This PhD will develop a new on-chip multi-modal operando gas analytics capability to understand how gas evolution triggers degradation in battery materials under thermal effects. EC-MS will be adapted to controlled elevated-temperature environments to quantify gas evolution during electrochemical cycling and thermal stress. By integrating gas, electrochemical and thermal measurements into a real-time analytical framework, the project will provide high-resolution mechanistic insight into temperature-driven degradation and thermally induced failure in battery materials. A curated Gas Evolution Database will document gas fingerprints, thermal conditions and cycling parameters across representative electrode–electrolyte chemistries.

Work packages/Timeline

Year 1: Platform establishment and operando capability development. The student will build the on-chip EC-MS platform for battery materials under controlled thermal conditions, calibrate synchronised gas–electrochemical–thermal measurements and validate data quality on selected chemistries.

Year 2: Multi-Material Screening & Data-Centric Interpretation. The capability will be applied across electrolyte formulations, coatings and additives using thermal perturbation to map gas evolution behaviours. Selected end-of-cycle samples will undergo complementary ex situ characterisation to capture chemical or microstructural signatures of temperature-driven degradation.

Year 3: Mechanistic discovery and ML-enabled insight. Machine-learning interpretation will be used to correlate gas evolution with interfacial reactivity, property decay and precursors to thermally driven failure. Ex situ techniques such as isotopic SIMS and cryogenic microscopy will validate reaction pathways inferred from operando analytics. A complex industrially relevant materials case will be studied in collaboration with SpectroInlets.

Year 4: Deployment, Transfer & Open Materials 4.0 Knowledge. The workflow will be extended beyond battery materials to demonstrate transferability to other reactive systems where gas evolution governs degradation. Final work will include data analysis, thesis writing and preparation of databases, protocols and models for open access.

The student will gain interdisciplinary expertise in electrochemical and thermal characterisation, chip-based sensing, real-time signal processing, database construction and machine learning. The project aligns strongly with the Materials 4.0 vision by combining smart characterisation, cyber-physical sensing and data-enabled modelling. Outcomes will be: (a) new scientific understanding of how gas evolution triggers thermally driven degradation in battery materials, and (b) a deployable Materials 4.0 capability for the wider research community.