QUEX PhD Scholarship

Project team

UQ - Dr Mickael Mounaix

Exeter - Associate Professor Jacopo Bertolotti

Project description

Despite being often referred to as "random", light scattering from a disordered medium is a deterministic process, completely determined by the incident light profile and the position and characteristics of the scattering elements (also known as the "scattering potential"). The process can be visualised as the superposition of all the possible paths light can take bouncing from scatterer to scatterer. By interpreting the scattering centres as nodes and the propagation between them as edges, multiple scattering of light can be mapped on a fully connected network. The relation between the input and the output light fields is complex but linear, preventing it from being used for machine learning. However, the relationship between the scattering potential and the output light is not [7]. This allows the use of a controllable scattering medium (by means of a spatial light modulator) as a neural network, offloading part of the calculation work to light propagation itself, hence using only a minimal amount of energy.

The goal of this project is to demonstrate a compact working device able to be trained to perform data classification and extrapolation, similarly to a more conventional neural network architecture, which relies solely on the control of optical waves. One of the shortcomings of the proof of principles demonstrated so far was the limited fraction of the scattering potential that could be controlled, which limits the degree of control one has on the network and thus how much the network can be trained. We aim at using a combination of spatial light modulators to generate a small cavity-like region where the scattering from the walls can be controlled at will.

The student will commence the project at the University of Exeter, in the laboratory of Prof Bertolotti, where they will learn the theoretical framework and the necessary optics techniques, and develop a first prototype. During this time we envision the student to visit for a period Prof Gigan at the Kastler–Brossel Laboratory (Paris, Fr), a world-leading pioneer on optical computing through complex photonic structures. Both CIs have a long history of collaborating with Prof Gigan: CI Bertolotti was a postdoc in Gigan’s group, and CI Mounaix graduated from his PhD in Gigan’s group. The student will then spend the second half of the PhD at the University of Queensland, where they will use their fabrication facilities to build and demonstrate the final device, in the Photonics Lab of the School of Electrical Engineering and Computer Science.

The project plan is summarised below:

1.  Goal 1 (0-9 months): Theoretical work on machine learning with optical waves (CI Bertolotti) at the University of Exeter
2. Goal 2 (10-18 months) : Develop the first prototype device using a spatial light modulator at the University of Exeter
3. Goal 3  (19-36 month) : Develop an advanced second prototype that uses two spatial light modulator to form a controllable photonic cavity (CI Mounaix, University of Queensland)

[7] M. Yildirim et al. "Nonlinear processing with linear optics" arxiv.org/abs/2307.08533 (2023)

Contact

Questions about this project should be directed to Dr Mickael Mounaix m.mounaix@uq.edu.au

Project team

UQ - Professor Guido Zuccon

Exeter - Dr Oleksandr Kyriienko

Project description

The goal of this PhD project is to investigate the use of quantum computing to improve Large Language Modelling technologies for information access. Specifically, the project aims to (1) identify new problem formulations to efficiently and effectively employ quantum computing to improve the efficiency of search engine algorithms based on LLMs, (2) evaluate and compare the performance of quantum computing approaches with respect to their non-quantum counterparts using traditional hardware, (3) explore the potential of quantum computing to enhance the effectiveness and efficiency of LLMs for information access.

This project squarely fits into the QUEX theme Digital Worlds and Disruptive Technologies in that it focuses on two largely disruptive technologies, Large Language Models and Quantum Computing, that are poised to radically change information access and computing products. The project brings together experts from these two key technologies. The project will also touch upon issues related to the QUEX theme of Environmental Sustainability, as it recognises the increasing environmental impact of the latest Artificial Intelligence technologies and it attempts to reduce these by leveraging Quantum Computing.

This project builds upon recent efforts in adapting feature selection algorithms for learning to rank to Quantum Annealers [1]. These previous efforts have shown the feasibility of mapping the feature selection problem into a Quadratic Unconstrained Binary Optimization (QUBO) which can be solved on Quantum Annealers. However, these efforts were limited in that (1) considered search technology (learning to rank) that is not the current state of the art, and has limited computational and energy requirements, (2) it showed that current generation quantum computers are able to provide only a limited speedup for these search algorithms.

Our project takes a novel, impactful direction by examining the feasibility of applying quantum computing when using LLMs for information access. These models, in fact, represent the current state of the art for information access technologies (across both search engines and recommendation systems), but are characterized by high computational costs, latency, energy consumption and environmental impact: all aspects that will be solved if we are able to crack open the problem of using quantum computing for these LLMs approaches.

Research in this PhD project will be first directed to examine current Quantum Annealers and their applicability for performing a high-quality model reduction for LLMs. Model reduction allows building lightweight LLMs that require less computational resources and thus are faster and consume less energy to run. This however often results in a trade-off between computational requirements and model effectiveness. We posit that the enhanced ability of solving complex optimization problems offered by QUBO formulations coupled with their execution on quantum annealers will deliver both more effective models and result in less energy and time being spent performing the model reduction process. We have access to a Quantum Annealers platform through our participation in this year’s Quantum CLEF initiative and its future iterations [2].

The project will then examine how gate-based quantum computers can be exploited to execute the computations involved in performing LLM-based search algorithms. Differently from the quantum annealer architecture, gate-based quantum computing can be used to speed up data processing sub-routines, potentially offering qualitatively new quantum IR protocols. However, as current gate-based quantum computers remain noisy, the development of suitable algorithms requires adaptive strategies and needs to operate at a limited depth of operations. We will approach this challenge by developing quantum machine learning (QML) protocols [3, 4]. Specifically, for quantum IR algorithms we will concentrate on the feature extraction stage and will design suitable embeddings that can be used for LLM-based search algorithms.

Throughout the project, the focus will be on identifying ways in which quantum hardware can be used to speed up and reduce consumption and environmental impact of Large Language Models for information access. The project will use standard Information Retrieval datasets (currently available to UQ researchers) and evaluation practices to collect results and demonstrate the benefit of the developed approaches.


[1] Ferrari Dacrema, M., Moroni, F., Nembrini, R., Ferro, N., Faggioli, G., & Cremonesi, P. (2022). Towards feature selection for ranking and classification exploiting quantum annealers. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 2814-2824).
[2] Pasin, A., Dacrema, M. F., Cremonesi, P., & Ferro, N. (2024). QuantumCLEF-Quantum Computing at CLEF. In European Conference on Information Retrieval (pp. 482-489).
[3] Zhang, Y., & Ni, Q. (2020). Recent advances in quantum machine learning. Quantum Engineering, 2(1), e34.
[4] Kyriienko, O., & Magnusson, E. B. (2022). Unsupervised quantum machine learning for fraud detection. arXiv preprint arXiv:2208.01203.

Contact

Questions about this project should be directed to Professor Guido Zuccon g.zuccon@uq.edu.au

Project team

UQ - Professor Ali Cheshmehzangi

Exeter - Dr Alicia Hayashi Lazzarini 

Project description

Urban centres are grappling with an unprecedented housing crisis, characterized by soaring prices, inadequate housing supply, and growing homelessness. This crisis is exacerbated by factors such as rapid urbanization, population growth, income inequality, and the impacts of climate change. Traditional top-down approaches to housing policy and urban planning have often failed to adequately address the needs of vulnerable communities, underscoring the importance of community-driven solutions and resilience planning.
The project aims to tackle several key challenges associated with the housing crisis:

1. Affordability: Many households struggle to afford housing costs, leading to housing instability, overcrowding, and homelessness.
2. Vulnerability: Vulnerable communities, including low-income households, minorities, and marginalized groups, are disproportionately affected by the housing crisis and face heightened risks of displacement and homelessness.
3. Resilience: Urban areas are increasingly vulnerable to climate-related hazards, such as extreme weather events and sea-level rise, which threaten the resilience and stability of housing infrastructure and communities.

Aims and Objectives:

1. Develop meso-scale community resilient planning frameworks that integrate housing affordability, resilience, and social equity considerations.
2. Identify and prioritize community-driven interventions and policy recommendations to address housing affordability and resilience challenges.
3. Engage with diverse stakeholders, including community organizations, local governments, housing advocates, and residents, to ensure inclusive and participatory decision-making processes.
4. Provide capacity-building support and technical assistance to empower communities to implement resilient housing solutions and advocate for policy change.

Deliverables:

1. Meso-scale community resilient planning frameworks tailored to the unique needs and characteristics of diverse communities.
2. Policy briefs and recommendations for local governments and policymakers to support affordable housing and resilience-building initiatives.
3. Community engagement reports documenting input and feedback from stakeholders and residents.
4. Training workshops and resource materials to build community capacity in resilient housing planning and advocacy.

Envisaged Approaches and Methods:

1. Community Engagement: Engaging with diverse stakeholders through community meetings, workshops, surveys, and focus groups to identify priorities, concerns, and aspirations related to housing affordability and resilience.
2. Data Analysis: Analysing housing market data, demographic trends, climate projections, and vulnerability assessments to inform resilience planning and decision-making.
3. Participatory Planning: Co-designing resilient housing solutions with communities, integrating local knowledge, preferences, and cultural values into planning processes.
4. Policy Advocacy: Collaborating with advocacy groups, policymakers, and government agencies to advocate for policy changes and investments that support affordable housing and community resilience.

 
Through collaborative and participatory approaches, the project seeks to empower communities to take ownership of their housing futures and build resilience in the face of ongoing challenges. By integrating housing affordability, resilience, and social equity considerations into meso-scale planning frameworks, the project aims to create more inclusive, sustainable, and resilient communities where all residents have access to safe, affordable, and dignified housing options.

Contact

Questions about this project should be directed to Professor Ali Cheshmehzangi a.chesh@uq.edu.au

Project team

UQ - Professor Paul Meehan

Exeter - Professor Philipp Thies

Project description

Vortices shed in the wake of a subsea power cable, apply fluctuating hydrodynamic forces, causing the cable to vibrate. If the vortex shedding frequency approaches the natural frequency of the cable, high amplitude vibration known as ‘Vortex-Induced Vibration’ or VIV occurs. VIV induces bending of the cable and sliding between the layers of the cable structure, often resulting in fatigue failure. While the research body on the modelling of the dynamic response and fatigue behaviour of power cables is mature, simultaneous modelling of these phenomena is still in its infancy.

The project aims to develop an analytical, simulation and experimental framework that can model the multibody dynamic response and resulting fatigue damage accumulation in a unified manner. The analytical approach supported by more complex simulation and experimentation is preferred for the modelling of complex nonlinear phenomena.

A multibody simulation model will be developed using FE and multibody simulation models in conjunction with a widely used global hydrodynamic model, Orcaflex. The analytical reduced order model, known as the wake oscillator model, will be utilised to predict VIV conditions efficiently in a wider range investigation. In this modelling approach, the fluctuating forces generated by vortex shedding are idealised by a nonlinear oscillator with a limit cycle. The structural motion interacts with the wake oscillator through a forcing term, forming a coupled system. The team at UQ have used a similar approach for prediction of Aeolian vibration in power lines, wind turbine flutter, brake squeal and railway wheel squeal. The developed modelling approach will advance previous studies by considering the nonlinear bending response of the helically wound power cable armour and conductors, which may improve the accuracy of fatigue damage calculations. A range of cable configurations will be considered, aimed at developing preventative guidelines against premature fatigue failures.

Prof Meehan leads the nonlinear mechanics group at UQ, which has a long history of solving engineering vibration and fatigue phenomena in a wide range of practical fi elds i.e. railway mechanics, advanced manufacturing, and electrical power transmission. This includes prediction and control of blade flutter vibrations in wind turbines and vortex induced vibrations in long span powerline cables. The project will also have the expertise of Dr Aditya Khanna who is an expert in theoretical and experimental fatigue analysis. Their expertise is supported by state-of-the-art experimental facilities in vibrations and tribology including contact mechanics, dynamic wear phenomena, fatigue analysis and microscopy. The modelling and simulation work as well as the experimental fatigue analysis will be progressed at UQ for the majority of the project.

Prof Thies leads the University of Exeter research in the reliability engineering of renewable energy technologies, with a focus on offshore energy. A focus of his work has been on critical components such as dynamic submarine cables and mooring for floating offshore wind. He is currently involved in a cable project [EP/W015102/1] that investigates the dynamic loading, motion response and fatigue failure of subsea power cables subjected to combined 3-dimensional waves, currents, and turbulence. His industry collaborative work is supported by world-leading ISO9001 accredited experimental facilities of the Dynamic Marine Component test facility (DMAC). The experimental validation of the modelling and simulation will be performed at this test facility at the University of Exeter.

Contact

Questions about this project should be directed to Professor Paul Meehan meehan@uq.edu.au

Project team

UQ - Professor Ali Cheshmehzangi

Exeter - Alicia Hayashi Lazzarini

Project description

Urban areas are on the front lines of climate change, facing a myriad of challenges that threaten the health and safety of residents. Heatwaves, flooding, air pollution, and vector-borne diseases are among the many consequences of a changing climate that disproportionately affect vulnerable populations in cities. This project seeks to address these interconnected challenges by developing innovative solutions that integrate climate resilience and public health considerations into urban planning and development.
Identification of the Problem: The project aims to tackle several key challenges faced by cities:

1. Climate Vulnerability: Cities are increasingly vulnerable to the impacts of climate change, including extreme weather events, heat stress, and air pollution, which pose significant risks to public health.
2. Siloed Approaches: Traditional urban planning and public health strategies often operate in silos, overlooking the interconnected nature of climate resilience and public health.
3. Health Inequities: Climate change exacerbates existing health inequities, disproportionately affecting marginalized communities with limited access to resources and healthcare services.

Aims and Objectives:

1. Develop a climate-resilient framework that integrates public health considerations into urban planning and development strategies.
2. Identify and prioritize climate adaptation and mitigation measures that promote public health co-benefits.
3. Engage with stakeholders from diverse sectors, including government agencies, healthcare providers, community organizations, and academic institutions, to ensure the relevance and effectiveness of the framework.
4. Provide evidence-based recommendations and guidelines for policymakers, urban planners, and public health practitioners to implement climate-resilient strategies in cities.

Deliverables:

1. Climate-resilient framework for healthy cities, incorporating climate adaptation, mitigation, and public health considerations.
2. Policy briefs and guidelines for integrating climate resilience and public health into urban planning and development processes.
3. Community engagement reports outlining feedback and input from local stakeholders.
4. Capacity-building workshops and training programs for city officials and practitioners on climate-resilient urban development and public health strategies.

Envisaged Approaches and Methods:

1. Literature Review: Reviewing existing research and best practices in climate resilience, public health, and urban planning to inform the development of the framework.
2. Stakeholder Engagement: Engaging with diverse stakeholders through workshops, focus groups, and interviews to gather insights and feedback on climate-resilient strategies.
3. Data Analysis: Analysing climate data, health indicators, and demographic information to identify vulnerable populations and prioritize interventions.
4. Policy Analysis: Evaluating existing policies and regulations to identify barriers and opportunities for integrating climate resilience and public health considerations into urban planning processes.
    

Contact

Questions about this project should be directed to Professor Ali Cheshmehzangi a.chesh@uq.edu.au

Project team

Exeter - Professor Gavin Tabor

UQ - Associate Professor Marcus Gallagher

Project description

CFD is a critical component of modern engineering, and has also found many applications in areas of science and medicine. CFD codes typically use the Finite Volume (FV) method, in which the equations to be solved are discretised on a mesh comprising numerous polyhedral cells. This approach gives enormous flexibility to suit the mesh to the local flow conditions, however the quality of the solution can be critically dependent on the quality of the mesh. Meshing is commonly regarded as the single most important, challenging and human-time consuming task in the CFD workflow. A lot of meshing activity relies on simple mesh quality metrics, rules of thumb, experience, and repetition of the process in a human centered optimisation process. This has proved challenging to improve on through conventional approaches, but of course these are characteristics of problems that are susceptible to the modern tools of AI and Machine Learning!

Overall there are two core aims of the project. The first is to apply Machine Learning optimisation techniques to meshing, to develop automated tools that could be used to iteratively improve mesh quality, “learn” strategies to mesh key types of geometry, and ultimately take over the whole meshing proves. The second core aim is to use Large Language Models (LLMs) such as ChatGPT to revolutionise the process of mesh development and case setup. Proximate objectives of the project include conducting a survey of meshing methodologies across a range of users and disciplines to identify common challenges and solutions that might be duplicated by AI.

Mesh quality can be optimised through at least two AI-inspired methodologies. The first is to treat it as an optimisation problem; commonly accepted metrics can be used to rank different meshes, and techniques such as Bayesian Optimisation can be used to iteratively improve mesh quality. Another approach is to use Artificial Neural Networks (ANNs) to “learn” how to build a good mesh. With the case setup; many mesh input formats use plain text, similar to computer code; input files for OpenFOAM in particular were included in the training sets for ChatGPT and so that tool can already generate input files for the code. We aim to extend this through retraining LLMs to ensure robust and correct outputs. OpenFOAM will be used here as it is an open source CFD code with an estimated 50,000 users; it is non-proprietary, and can be easily modified to integrate with optimisation and ANN tools.

The team brings together complementary skills covering all aspects of the work. Prof Tabor is an international expert in CFD and OpenFOAM, and has extensive connections in industry, which will be leveraged to support the project. He has worked closely with Prof Fieldsend, whose research interests include multiobjective optimisation and application of machine learning techniques to computational modelling. In Queensland Prof Gallagher brings in research interests in AI, Optimisation and Machine Learning; including cross-disciplinary collaborations and real-world applications of AI techniques.
 

Contact

Questions about this project should be directed to Professor Gavin Tabor G.R.Tabor@exeter.ac.uk

More information and how to apply

Project team

Exeter - Dr Benno Simmons

UQ - Dr Tatsuya Amano

Project description

Businesses, NGOs and governments are increasingly using AI to monitor biodiversity at scale. Generally, AI is combined with a passive sensor technology, such as camera traps, satellites, bioacoustics, or eDNA. The most widely used of these is camera traps, where AI is used to identify species in images and videos in order to produce insights about ecosystem health. AI holds great promise, but also can cause harms if not developed responsibly. Camera trap AI is nascent and often highly inaccurate yet is being deployed widely today. This is concerning given that we currently have no understanding of how accurate these systems need to be, or the consequences if they are not. Bad AI systems could miss species declines, resulting in bad conservation outcomes and misallocation of funds.
This PhD will develop the first research into responsible AI for biodiversity monitoring, allowing the full potential of this revolutionary technology to be harnessed.

Aims, methods and deliverables:

1. Recommendations for AI accuracy. We need to understand how accurate species identification AIs must be for reliable population trends. Using camera trap data, population trends will be calculated using unmarked models at varying levels of artificial labelling errors. The minimum labelling accuracy needed to avoid erroneous conclusions (e.g. trend in the wrong direction) will be determined across space and taxa.
2. Developing a safety standard. The first safety standard for species identification AIs will be developed, incorporating the kinds of images that AIs most often get wrong, but which are important to classify correctly from a conservation perspective. The performance of leading species identification AIs will be evaluated against the benchmark and the benchmark will be publicly released to be adopted as a standard.
3. A new evaluation metric Standard. AI evaluation metrics are not suitable for conservation as they treat all mistakes equally e.g. misclassifying an endangered species is equal to a non-threatened species. If the field has, as its north star, an evaluation metric that ignores ecological reality, then AIs will be chasing the wrong goal, with potentially perverse outcomes. This project will therefore develop a new evaluation metric, to be adopted as a standard, that weights misclassifications by phylogenetic distance and trait dissimilarities, like conservation status.
4. Develop transparent, explainable Ais. AI models used in conservation and funding decisions must be explainable and accountable. Explainable AI techniques like Testing with Concept Activation Vectors (TCAV) will make AI decisions interpretable by relating them to human-understandable concepts. A Bayesian modelling framework will also be developed that ensures uncertainty from species identification is propagated through to the final insights like population trend.

Benno Simmons (primary) specialises in using AI and technology for biodiversity conservation. Tatsuya Amano is a conservation scientist with expertise in using statistical modelling approaches to overcome information gaps. The student will be primarily Exeter-based, using the JADE II HPC for computation, and will spend 12 months at UQ to work with Dr Amano to develop the novel statistical methods for O1 and O4.

Contact

Questions about this project should be directed to Dr Benno Simmons b.simmons2@exeter.ac.uk 

More information and how to apply

Project team

Exeter - Dr Diego Panici

UQ - Professor Rod Fensham 

Project description

Nature-based Solutions (NbS) in hydrological sciences manage water in the landscape to reduce flood risk, enhance drought resilience, restore ecosystem functioning and improve water quality. Over 90% of NbS research concentrates in temperate regions like Europe, North America, and China, leaving many drylands under-explored and under-researched (Alikhanova and Bull, 2023; Dunlop et al., 2024). Drylands rely on shallow groundwater aquifers, have distinct hydrogeological processes, and unique features (e.g., wadis and oases) making them vulnerable to water scarcity, habitat loss, and biodiversity shifts (Fensham et al., 2023). Archaeological evidence shows substantial climate-change driven water depletion (Wang et al., 2021), compared with relative water stability in the pre-Anthropocene era. Key questions are ‘where’ and ‘what’ NbS can be used (Alikhanova and Bull, 2023), since conventional, engineering solutions may be unsuitable due to costs, ephemeral flow regimes, and the intensity/duration/frequency of hydrological cycles. 

This PhD aims to develop a framework for identifying, evaluating, and implementing effective hydrological NbS for shallow groundwater aquifers in arid and hyper-arid environments.

Objectives are:

1. Systematic review of the scientific literature on NbS for water resource management. The PhD will evaluate the transferability to dryland, considering effectiveness of groundwater recharge of theoretical (e.g., Budyko curves) and empirical (e.g., field monitoring) frameworks.
2. Identify and map optimal NbS deployment areas using remote sensing (e.g., topography, Land-use-land-change, multi-spectral) on an open-source cloud-based platform (e.g., Google Earth Engine). Analyse satellite-derived indicators (e.g., elevation, slope, soil characteristics, NDVI) utilising machine learning approaches to pinpoint areas suitable for NbS deployment in dryland globally. Deliverable will be interactive open-source web application for global NbS feasibility.
3. Create a novel numerical model framework, testing case studies (identified in O2 and informed by fieldwork at Queensland). Open-source/open-access models (e.g., MODFLOW and HEC-RAS) will be used while pioneering modelling analogues to simulate NbS hydrology. Simulations will inform NbS selection for effective groundwater recharge. Deliverable will include development of a MODFLOW extension for dryland NbS.
4. Formulate a decision-support approach evaluating NbS efficacy and informing deployment in arid areas that will be applied to case studies to validate the approach. Deliverables will include a handbook and an open-access webinar.

The University of Exeter and CREWW, global leader for NbS in hydrology, have shaped environmental policies, including land management and hydrological restoration. Collaborating with a plethora of stakeholders, CREWW boasts state-of-the-art facilities, and over £41m in funding, directing world-leading NbS-based projects such as Upstream Thinking and South West Peatland Partnership. The University of Queensland brings expertise in hydrogeology and ecology conservation in arid environments, focusing on dryland water management. The combined strengths of CREWW and the Queensland School of Environment offer a robust foundation for this project.

At Exeter, the student will explore NbS (O1) and hydrological modelling (O3). Both universities will support dryland mapping techniques (O2) and policy change (O4). Work in Queensland will enrich the student’s ability to characterise eco-hydro-morphological features for case-studies (O3 and O4).

Contact

Questions about this project should be directed to Dr Diego Panici  D.Panici@exeter.ac.uk

More information and how to apply

Project team

Exeter - Professor Stu Bearhop

UQ -Professor Rich Fuller

Project description

Migratory birds act as bioindicators, signalling environmental health and deliver important ecosystem services including pollination, seed dispersal, and pest control. Their trans-hemispheric movements inspire awe and connections among different regions of the world that they link influencing art, literature, and traditions. Thus, there is considerable interest among both scientists and the public about the impact of our changing environment on migratory birds. However, although we know that numerous populations are in decline, our understanding of the drivers is restricted and measuring the way species can respond has proved difficult. The challenge is that populations can face threats at any point in their migration. This means that the place where a decline is detected isn’t necessarily where the threat is occurring. To overcome this, potential threats and migratory routes need to be mapped across the species’ full distribution. While there are comprehensive data sets on global change in habitats, climate, and other human pressures, it has proven difficult to model migration routes for more than a few species. This means the causes of decline, and therefore appropriate mitigation measures, remain unclear for many migratory species.

This multidisciplinary PhD would address these gaps, using the huge citizen science data sets that are now available, providing a comprehensive understanding of population change, its associated drivers, and potential routes to mitigation. The project would map threats to migratory birds globally, and then derive strategies for their conservation. It would achieve this by (1) discovering which threats are operating where in the annual cycle, (2) determining which species are most impacted, (3) measuring the extent to which full migration pathways of birds lie in the global protected area estate, and (4) building a plan for expansion of protected areas and other conservation measures to safeguard migratory birds globally.

The project will use status maps, summarising the distribution of migratory bird species at weekly intervals across the year (n»1,000). Spatial data (land use change, climate change, etc) will be overlain on these to determine where and when in the annual cycle threats/change happen (Aim 1). Then, data on ecological and behavioural traits of species will be correlated with threats to understand why some species are more at risk than others (Aim 2). The first ever global summary of progress in migratory bird conservation (Aim 3) will be achieved by overlaying protected areas onto seasonal distributions of each species to assess protection throughout the year. Finally, mathematical optimisation tools will determine cost-effective locations of new protected areas to achieve joined-up migratory bird conservation globally.

This multidisciplinary project will combine the expertise in population ecology and behaviour of Stuart Bearhop in UoE with the global spatial skills of Richard Fuller at the UQ and the conservation planning skills of Zhijun Ma at Fudan University. Alison Johnston at the University of St Andrews will bring world-leading big data expertise, and Tom Auer from Cornell University will support geospatial data analysis. This globally diverse team is well placed to support the student to conduct the global analyses necessary for the project’s success.

Contact

Questions about this project should be directed to Professor Stu Bearhop s.bearhop@exeter.ac.uk

More information and how to apply

Project team

Exeter - Professor Frank Vollmer

UQ - Professor Irina Vetter

Project description

The target of many venom-derived peptides that interfere with pain signals are voltage gated ion channels. These proteins consist primarily of transmembrane helices with small extra and intracellular loops. The action of ligands of these receptors are therefore either via the membrane or involves components of the membrane. Untangling the kinetic and thermodynamic contributions of ligand binding in this complex tripartite system is critical for the rational design of novel analgesic drugs.
Currently, researchers use techniques which involve modification with fluorescent labels on molecules or using bulk sensors to study how molecules interact with membranes as well as transmembrane ion channels and receptors critical to function of pain-sensing nerves.  However, these methods have limits in their ability to discern the molecular mechanisms governing interactions between ligands, the plasma membrane, and the therapeutic targets located within the membrane environment. 
In particular, these techniques fall short of capturing the transient kinetics and conformational changes inherent in peptide-membrane interactions. To address this gap, our project aims to apply a cutting-edge single-molecule sensor developed at the University of Exeter to study the dynamics of peptide interactions with reconstituted supported lipid membranes to understand how peptide sequence and structure, membrane composition the ionic environment control long lasting peptide-membrane interactions.

The project incorporates three aims: 

1. Understanding Molecular Interactions: In collaboration with Prof Vollmer at Exeter, you will first learn to use the single molecule sensor. This involves learning to coat the sensor with lipid bilayers, analysing signals to discern transient and binding interactions, and extracting rate constants to understand the dynamics of peptide-membrane interactions. You will establish this technique for the first time for to study molecular interactions at membranes.
2. Synthesising and Characterising Peptides: In collaboration with Profs Vetter and Mobli at the University of Queensland, you will synthesise peptides and evaluate their biological activity using pharmacological methods including patch-clamp electrophysiology. These studies will be complemented by structural studies using nuclear magnetic resonance spectroscopy to understand the static structures of peptides interacting with membranes.
3. Integration and Analysis: Returning to Exeter, you will apply the single molecule sensing methods to the peptides characterised in Queensland. This involves quantifying how changes in sequence, membrane composition and ionic strength affect interaction kinetics, integrating sensor signals with analysis from Queensland, and studying the role of 2D and 3D diffusion in peptide-membrane interactions.

 
Throughout the project, you will achieve several milestones:

• Learning to operate the single molecule sensor and analysing signals
• Obtaining information on the kinetics and conformational changes in the sensor signals.
• Establishing methodologies to study 2D and 3D diffusion of peptides near membranes.
• Applying advanced sensor measurements and signal analysis to study changes in membrane force and pressure.

Ultimately, your contributions will lead to significant scientific advancements, with results expected to be published in high-impact journals and presented at international conferences and QUEX workshops. Additionally, you will participate in seminars and workshops to develop transferable skills, all within an environment that prioritises equality, diversity, and inclusion, open research practices, and collaborative teamwork.

Contact

Questions about this project should be directed to Professor Frank Vollmer f.vollmer@exeter.ac.uk

More information and how to apply

Project team

Exeter - Dr David Richards

UQ - Dr Mel White

Project description

This project will study how the neural tube develops in a quail animal model, with applications to understanding developmental defects that account for some of the most common and severe human birth defects (including anencephaly and spina bifida that globally affect up to 1 in 500 births). The overall aim is to develop a novel mathematical model, driven by time-lapse imaging, of the spatiotemporal dynamics of neural tube development. This model will then be used to suggest potential avenues for tackling related birth defects. Key questions that will be addressed include how biophysical forces and cell fate specification interact to generate the spatiotemporal pattern, the role of the actin cytoskeleton, and how system robustness is achieved. One of the chief advantages of this project is its multidisciplinary nature, intimately combining wet-lab work and theoretical work. This interdisciplinary combination is increasingly being used in both biological and biomedical research, and will give the student an excellent, highly sought-after skillset that will place them in a strong position with a broad range of future career options.

The project will involve the following three objectives:

1. High-resolution long-term time-lapse imaging (based on expertise in Queensland). The student will perform confocal imaging of the developing neural tube in transgenic quail embryos expressing various markers of the nucleus, cell membrane, actin cytoskeleton and cell fate markers. Quail embryos will be cultured on agar-albumin in glass-bottomed imaging dishes for 48 hours from the start of neural tube formation using a protocol optimised in the White group. Cellular dynamics of the developing neural tube will be visualised by capturing confocal images every 7 minutes at 40x magnification for ~16 hours.
2. Automatic image analysis (based on expertise in Exeter). Building on existing custom-built software within both the Richards and White groups, software will be developed that automatically segments and tracks cells in the developing neural tube. This will involve a combination of blob detection, edge detection, thresholding and Hough transforms. Tracking between frames will use a custom-built Hungarian algorithm. Cell division will be identified as in the recent work by Katie McDole et al. and others.
3. Mathematical modelling (based on expertise in Exeter). A mathematical model/computer simulation will be designed based on a vertex model that the Richards group have already developed to describe an earlier stage of development. Cells will be described by marker points that move due to vertex-vertex forces. Vertices will be connected by springs with a tension force governed by a viscoelastic extension of Hooke’s law. Other forces will operate, including a curvature force (given by the Helfrich energy), an intracellular actin force and a volume conservation force. Cell-to-cell interaction and signalling will be included at adjacent cells.

Project timeline:

• Year 1 (Queensland) – time-lapse imaging; initial image analysis development; planning of public involvement event. Deliverables: new imaging data sets, first paper (experimental).
• Year 2 (Exeter) – continued image analysis development; creation of the mathematical model; public involvement event. Deliverables: bespoke image analysis algorithm, second paper (theoretical).
• Year 3 (Exeter and Queensland) – model validation and parameter fitting; experimental testing of model predictions. Deliverables: novel mathematical model and potential avenues for tackling birth defects, third paper (combined experimental and theoretical).

Contact

Questions about this project should be directed to Dr David Richards david.richards@exeter.ac.uk

More information and how to apply

Project team

Exeter - Dr Zhenyu Zhang

UQ - Professor Lianzhou Wang

Project description

Electrochemical lithium extraction offers an appealing method for extracting lithium from low-concentration brine lake water or seawater, without the need for the time-consuming evaporation process. The basic principle of this system involves two main approaches: (1) driving lithium ions with an external electric current either into an electrode or (2) through a lithium-selective membrane. For example, TiO2-coated LiFePO4 electrode material combined with a pulse electrochemical method was used to adsorb lithium ions from seawater (Joule 2020, 4, 1459.). However, the adsorption suffers from slow kinetics and material degradation by competing ions. The alternative method employs a lithium-ion selective membrane, allowing only lithium ions to be electrically mobilized across it, while other cations are blocked due to crystal mismatch, remaining in the original electrolyte. Typically, the membrane consists of a lithium ion-conducting material, similar to those used as solid electrolytes in all-solid-state lithium-ion batteries. For example, NASICON-type (Li1+xAlyGe2-y(PO4)3, LAGP) (Joule 2018, 2, 1648.) and glass-type (Li0.33La0.56TiO3, LLTO) (Energy Environ. Sci. 2021, 14, 3152.) lithium-ion conductors have been used in this system. As shown in the figure, by continuously introducing seawater or brine water on the anode side, lithium-ion concentration can be increased on the cathode side. Lithium salt precipitation in the cathode side aqueous electrolyte or lithium metal deposition on cathode with organic electrolyte can be achieved. Simultaneously, byproduct such as Cl2/O2 and H2 gases are generated on the electrodes. Although this design has been demonstrated in several publications, significant challenges persist, such as the low selectivity ratio of lithium ions, slow kinetics for the lithium transport, low lithium generation rate and low energy efficiency.

In this project, the main objectives include:

1. Develop novel lithium selective membrane, based on the advanced solid state electrolyte materials for all-solid-state lithium-ion battery. Using surface coating materials to protect and improve the ion selectivity, conductivity, and long-term stability in different electrolyte.
2. Design new electrolyte systems for the electrochemical process, such as catalyst materials on the anode, using of anion exchange membrane for the anode protection electrolyte, and electrolyte of cathode side, to optimize the efficiency of lithium production.
3. Integrate the system with renewable energy sources to achieve sustainable lithium extraction. Demonstrate the practical application of the according to the industry needs.

From the Exeter side, Dr Zhenyu Zhang is an expert in the material degradation study of electrochemical systems, especially in lithium-ion battery and solid-state electrolytes; Prof. Xiaohong Li's team provides specialised knowledge in energy conversion systems such as hydrogen production from sea water and membrane-free flow batteries. From the Queensland side, Prof. Lianzhou Wang and Dr. Haijiao Lu offer expertise in materials synthesis and characterisation and developing catalysts electrochemical reactions. The objectives 1 are expected to be progressed at Queensland with their abundant resources for catalyst synthesis and performance evaluation, and the objectives 2-3 will be achieved at Exeter with the excellent facilities and infrastructures for system development and large-scale demonstration. The collaboration between the teams will leverage the combined knowledge and facilities to ELITE.

Contact

Questions about this project should be directed to Dr Zhenyu Zhang zhenyu.zhang@exeter.ac.uk

More information and how to apply

Project team

Exeter - Dr Prakash Kripakaran

UQ - Professor Daniel Franks

Project description

Ordinary Portland cement (OPC), which remains the predominant building material in the construction sector, is responsible for over 8% of the emissions. The construction sector has therefore been exploring a range of alternatives for OPC. An important recent development is the limestone calcined clay cements (LC3), pioneered by the work of Scrivener et al. (2018) [1] . LC3 has been demonstrated to have a carbon footprint that is 40% less than OPC and offers potential to significantly reduce CO2 emissions in the cement sector. However, current methods of LC3 production rely on clays with very high levels of kaolinite, which is naturally available only in limited locations geographically and therefore unsuitable for scale-up. The use of low grade clays, which are available abundantly in the UK and often generated as a waste in the mining sector, for LC3 remains an open research question. This research aims to address this challenge. Further to that, it will also examine the use of biogenic calcium carbonate (from sea shells) as an alternative to limestone as filler in concrete or as a raw ingredient for clinker production – which can further reduce embodied carbon and directly benefit coastal economies that significantly rely on seafood and produce localised amount of shells. The aim of this project is hence to investigate the potential for using low grade clays, such as those generated by the mining sector in Cornwall, and biogenic calcium carbonate sources for LC3 and its positive benefits for sustainability and circular economy.

The specific project objectives are as follows:

1. Characterise the compositional properties of waste clays sourced from Cornwall through materials analysis.
2. Investigate processing pathways for calcination and develop the optimal parameters for processing to achieve the desired chemical and compositional properties.
3. Undertake microstructural characterisation (FTIR, SEM, XRD) of cement pastes produced with calcined clays and biogenic calcium carbonate.
4. Assess the mechanical properties (compressive and flexural strength) of mortar and concrete samples produced with the waste-based LC3 cement.
5. Assess the availability of the identified secondary raw materials to determine the viability of LC3 cement production in the targeted areas.
6. Perform lifecycle analysis (LCA) to assess the energy and emission impacts of material flows in the preparation of calcined clay cements and formulate approaches for scaling up developed processing methods.

The PhD will have two main work strands, both leveraging expertise of the academics at the two institutions.

1. Materials processing and characterisation: This work will use the experimental facilities (XRD, TGA FTIR, SEM) at Exeter and Queensland for understanding the chemical and mineralogical composition of the clays and processed raw materials, as well as on reaction products later in the study. The kiln at Exeter will be used to investigate the calcination step, assessing the impact of temperatures and duration of the process on the reactivity of the materials through XRD, FTIR and TGA. The student will then undertake early age shrinkage and macro properties (compression strength etc) at Queensland drawing on their expertise in this area.
2. LCA and circular economy: The student will use SimaPro or similar tool for LCA. The energy and material flows will be modelled according to the process pathways identified in the first work strand. Industry input will be sought on scaling up of the processes. [1] Scrivener, K., Martirena, F., Bishnoi, S. and Maity, S., 2018. Calcined clay limestone cements (LC3). Cement and concrete research, 114, pp.49-56.

Contact

Questions about this project should be directed to Dr Prakash Kripakaran p.kripakaran@exeter.ac.uk

More information and how to apply