Fri 14 Nov 15:00: Bonded Interfaces: Worlds in-between
Adhesive bonding is a widely used joining technique across various industries and is particularly common in composite structures within the construction sector, especially for the repair and strengthening of existing infrastructure. Extensive research has been conducted to understand the behaviour of fibre-reinforced polymer (FRP) to concrete and FRP to steel bonded joints, leading to well-established design methodologies for such interfaces in civil engineering applications.
Previous studies have shown that bonded interfaces are subjected to complex stress states, and numerous theoretical models—with varying levels of simplification—have been developed to predict interfacial stresses. These models, along with numerical approaches, have been used to estimate key parameters governing the performance and failure of bonded joints, including debonding modes.
While there is broad consensus on several aspects such as bond strength, effective bond length, and bond–slip relationships, discrepancies remain regarding interfacial stress distributions, behaviour under cyclic loading, and response under elevated temperatures. Moreover, ongoing debate surrounds the suitability of homogenization approaches, such as traction–separation laws, in accurately capturing the behaviour of bonded interfaces.
In this talk, I will provide a critical review of existing research on FRP -to-concrete and FRP -to-steel bonded joints. The discussion will cover their behaviour under quasi-static monotonic loading, cyclic loading, and combined mechanical–thermal loading. Interfacial stress models and the appropriate use of homogenization techniques in modelling bonded interfaces will also be critically examined.
- Speaker: Dilum Fernando, University of Edinburgh
- Friday 14 November 2025, 15:00-16:00
- Venue: CivEng Seminar Room (1-33) (Civil Engineering Building).
- Series: Engineering Department Structures Research Seminars; organiser: Lowhikan.
HZO/HSO Superlattice ReFET Array Integrating Optical Sensing for Neuromorphic Vision Computing
A photonic ReFET array with an engineered HZO–HSO superlattice enables monolithic integration of optical sensing, nonvolatile memory, and synaptic computation. It exhibits forming-free, >8-bit analog states, >1010 endurance cycles, and light-tunable conductance. Integrated into a 20 × 20 array, it achieves 94.45% accuracy on Fashion-MNIST in an in-sensor vision transformer.
Abstract
Neuromorphic vision systems require artificial synapses that integrate sensing, memory, and computation with high precision and stability. Conventional memristors face limitations including forming requirements, few multilevel states, low endurance, and poor integration density, while ferroelectric and flash-based transistors suffer trade-offs among endurance, retention, and switching ratio, and generally lack intrinsic photonic sensitivity, constraining in-sensor computing. Here, a photonic resistive-gate field-effect transistor (ReFET) array is presented that combines optical sensing, forming-free multilevel memory, and analog computation in a single device. The ReFET employs a Hf0.5Zr0.5O2/Hf0.95Sr0.05O2 (HZO/HSO) superlattice gate and an amorphous InGaZnO (IGZO) channel, achieving 272 stable conductance states (>8-bit) in a 20 × 20 array, ON/OFF ratios >10⁶, endurance >1010 cycles, and retention >10⁶ s. The array functions as an in-sensor optical convolutional layer, performing multiply–accumulate (MAC) operations with 94.45% accuracy on Fashion-MNIST using 8-bit quantized weights, while delivering high energy efficiency. This platform enables scalable, high-precision, energy-efficient photonic neuromorphic computing, integrating sensing, memory, and computation in one architecture.
Fri 21 Nov 15:00: Turning Solid Wastes into Engineered Geopolymer Composites for Sustainable and Resilient Infrastructure
Civil infrastructure assets, such as buildings, bridges, tunnels, roads, and dams, are vital to societal well-being and economic development. Achieving Net Zero by 2050 remains a major challenge for the construction industry that is expected to deliver cost-effective, sustainable, and resilient infrastructure within a circular economy framework. This talk introduces a new type of low-carbon concrete recently developed at UCL , following a materials–microstructure–property–performance approach. The material combines a cement-free geopolymer, produced from industrial by-products, with recycled fibres to enhance sustainability and performance. When paired with lightweight fibre reinforced polymer (FRP) bars, it forms a novel reinforced concrete system that offers a viable alternative to conventional concrete. This innovation promises to extend infrastructure lifespan while reducing carbon emissions, as well as repair and maintenance costs, providing significant environmental and economic benefits.
- Speaker: Mingzhong Zhang, University College London
- Friday 21 November 2025, 15:00-16:00
- Venue: CivEng Seminar Room (1-33) (Civil Engineering Building).
- Series: Engineering Department Structures Research Seminars; organiser: Lowhikan.
Wed 22 Oct 15:30: Regulating the Weddell Sea polynya in a coupled model through ocean model parameter tuning - Tarkan Bilge
Exaggerated open-ocean deep convection in the Weddell Sea has been a prevailing issue in coupled climate models, having featured in both CMIP5 and CMIP6 simulations. These convection events vent subsurface heat and sustain a large hole in the sea-ice, known as the Weddell Sea polynya which can create large Southern Ocean biases in sea-ice, salinity and ocean heat content. In order to tackle this issue, we have run a reference ensemble to explore the mechanisms for preconditioning and convection in HadGEM3-GC5, a state-of-the-art coupled climate model developed by the UK Met Office. We then present a suite of parameter perturbation experiments in which we describe the influence of ocean model parameters on deep convection. By tuning parameters in the vertical mixing scheme we present simulations in which convection is reduced to more closely reflect observations.
- Speaker: Tarkan Bilge - British Antarctic Survey
- Wednesday 22 October 2025, 15:30-16:30
- Venue: BAS Seminar Room 12.
- Series: British Antarctic Survey - Polar Oceans seminar series; organiser: Katherine Turner.
Fri 24 Oct 14:30: Metallic Lattice Metamaterials: Extreme Manufacturing and Multifunctional Applications
Title: Metallic Lattice Metamaterials: Extreme Manufacturing and Multifunctional Applications
Abstract: Metallic lattice metamaterials, a class of architected materials characterized by their precisely engineered periodic micro-architectures, have emerged as a frontier in materials science and advanced manufacturing. We begin by discussing our developed Micro-Selective Laser Melting (µSLM), which enable the fabrication of metallic lattices with unprecedented geometric complexity and fine feature resolution. Then, building upon µSLM technology, we have developed an oxide dispersion strengthening strategy that enables high-precision (78 µm), cost-effective additive manufacturing of pure copper components using infrared lasers. Furthermore, through the implementation of dislocation-engineered 3D printing methodology, we have achieved integrated manufacturing across nano-to-macro scales. The presentation subsequently demonstrates the multifunctional applications of these advanced manufacturing technologies across various domains, including components for supersonic impact resistance (aerospace), high-precision metalens fabrication (terahertz communications), extreme heat environments (combustion chambers), and monolithic catalytic electrodes (sustainable development). Together, these advances promote the multi-scale development of lattice metamaterials and unlock their potential for next-generation engineering applications.
Bio: Dr. Liqiang is currently a Postdoctoral Research Associate in the Additive Microstructure Engineering (AddME) Lab at the University of Cambridge, under the supervision of Prof. Matteo Seita. He received his PhD from City University of Hong Kong in 2024. Following his Ph.D., he conducted postdoctoral research at The Chinese University of Hong Kong. His research specializes in the multi-scale design and additive manufacturing of metallic lattice metamaterials, with a focus on high-thermal-conductivity copper-based materials and high-strength/ductility high-entropy alloys for extreme service conditions. Dr. Liqiang’s work has resulted in high-impact publications in journals, covering topics such as extreme dynamic loading (Science Advances, 2025, featured as Cover Article), terahertz metalens manufacturing (Nature Communications, 2025), sustainable electrocatalyst design (Nature Communications, 2025), and extreme thermal management (Additive Manufacturing, 2024).
- Speaker: Dr Liqiang Wang, CUED
- Friday 24 October 2025, 14:30-15:00
- Venue: Oatley 1 Meeting Room, Department of Engineering.
- Series: Engineering - Mechanics and Materials Seminar Series; organiser: div-c.
Fri 24 Oct 14:00: Integrating thermodynamic and kinetic simulations within a CALPHAD framework for alloy design
The progress in alloy design has created a growing need for predictive models that connect thermodynamics, kinetics, and microstructure to understand the composition-process-structure-property (CPSP) relationships of alloys. This presentation introduces an integrated CALPHAD -based modelling framework that couples thermodynamics with kinetics to simulate microstructural evolution during phase transformations. The framework combines a Scheil model to describe non-equilibrium solidification and solute segregation with a mean-field precipitation model that captures the nucleation, growth, and coarsening of second phases. By coupling phase equilibria, diffusion, and interfacial thermodynamics, the model quantitatively predicts the evolution of precipitate size and distribution as a function of temperature and time, establishing a digital description of processes from solidification to heat treatment. This methodology provides a computational foundation for virtual process optimisation and digital alloy design. The framework can be readily applied to various processing routes, including rapid solidification, thermomechanical process, and additive manufacturing, supporting the development of high-performance structural materials. The established framework bridges computation and experiment, demonstrating how virtual alloy prototyping can support data-driven materials design and reduce experimental workload in intelligent manufacturing environments. ”
- Speaker: Dr Yanheng Xie, CUED
- Friday 24 October 2025, 14:00-14:30
- Venue: Oatley 1 Meeting Room, Department of Engineering.
- Series: Engineering - Mechanics and Materials Seminar Series; organiser: div-c.
Tue 21 Oct 15:15: Refined invariants in tropical geometry
Refined invariants were introduced into tropical geometry by Block and Gottsche in 2016. Since then various authors have investigated their properties and considered several generalizations of them. In this talk I will explore some of the approaches to investigating refined invariants and present a version of refined invariants in positive genus introduced jointly with Eugenii Shustin.
- Speaker: Uriel Sinichkin, Tel Aviv University
- Tuesday 21 October 2025, 15:15-16:15
- Venue: CMS MR14.
- Series: Algebraic Geometry Seminar; organiser: Dhruv Ranganathan.
Catalysts for Electrochemical Oxidation of Ammonia: A Comprehensive Review of Fundamentals and Optimization Strategies
Developing robust catalysts for ammonia electrochemical oxidation (AOR) is essential for advancing NH3 utilization technologies. This review summarizes recent progress in catalyst design and mechanistic understanding of AOR. In addition, it systematically investigates strategies to improve AOR performance for various types of catalysts. Finally, an outlook in multiple disciplines related to the future directions is presented.
Abstract
The realization of a hydrogen (H2) economy confronts formidable challenges in storage, transportation, and logistics. To address these challenges, H2 carriers have been proposed as alternative solutions. Ammonia (NH3), as one of the most promising H2 carriers, becomes a game-changer, offering higher volumetric H2 density than compressed H2, simplified logistics, and compatibility with existing infrastructure, which thereby reduces costs and supply chain complexities. However, fully realizing NH3’s potential requires overcoming downstream inefficiencies associated with its conversion either back into H2 or into energy. Downstream processes primarily include thermal cracking, NH3 electrolysis, and direct NH3 fuel cells, two of which are electrochemical systems leveraging the NH3 oxidation reaction (AOR). The efficiency of these electrochemical systems is significantly limited by severe surface poisoning and poor AOR catalytic activity, underscoring the urgent need for advanced catalyst design. Here, a comprehensive review of AOR electrocatalysis is provided, with a focus on mechanistic insights into activity-governing steps and surface poisoning pathways. Recent advances in catalyst design are summarized, and overlooked factors are highlighted for performance enhancement. Finally, perspectives on future research directions are presented for AOR catalyst development to accelerate the integration of NH3-based technologies into the hydrogen economy.
Liquid Jet‐Based Triboelectric Nanogenerator
A liquid jet triboelectric nanogenerator driven by Plateau–Rayleigh instability (PR-TENG) is devised. Compared with droplet and continuous water flow, the liquid jet exhibits enhanced charge transfer efficiency, delivering over two orders of magnitude higher current output. The mechanism originates from high-frequency contact, efficient electron capture, and droplet recontact-assisted charge release.
Abstract
Although the solid–liquid triboelectric nanogenerators have made great progress in energy harvesting, their efficiency is fundamentally limited by the continuous shielding effect of the solid–liquid electric double layer (EDL). Herein, a liquid-jet triboelectric nanogenerator that exploits Plateau-Rayleigh instability (PR-TENG) is demonstrated to overcome this barrier. PR-TENG converts low-frequency water flow into high-frequency droplets, avoiding shielding effects from continuous flow and achieving a contact-separation frequency of ≈50 Hz. At a flow rate of 320 mL min−1, the PR-TENG generates an output current of over 100 µA, ≈120 times the magnitude of conventional single-electrode droplet TENGs (D-TENGs) that rely on electrostatic induction. The operational mechanism of PR-TENG arises from droplet-fusion-induced triboelectrification, surface charge trapping, and charge redistribution, enabling energy harvesting from continuous water flow and water-level monitoring applications. This work significantly improves the electrical output performance of flow-based triboelectric nanogenerators and provides an effective solution for continuous flow-based energy harvesting.
Thu 04 Dec 15:00: Reliable and Sustainable AI: From Mathematical Foundations to Next Generation AI Computing
The current wave of artificial intelligence is transforming industry, society, and the sciences at an unprecedented pace. Yet, despite its remarkable progress, today’s AI still suffers from two major limitations: a lack of reliability and excessive energy consumption.
This lecture will begin with an overview of this dynamic field, focusing first on reliability. We will present recent theoretical advances in the areas of generalization and explainability—core aspects of trustworthy AI that also intersect with regulatory frameworks such as the EU AI Act. From there, we will explore fundamental limitations of existing AI systems, including challenges related to computability and the energy inefficiency of current digital hardware. These challenges highlight the pressing need to rethink the foundations of AI computing.
In the second part of the talk, we will turn to neuromorphic computing—a promising and rapidly evolving paradigm that emulates biological neural systems using analog hardware. We will introduce spiking neural networks, a key model in this area, and share some of our recent mathematical findings. These results point toward a new generation of AI systems that are not only provably reliable but also sustainable.
- Speaker: Gitta Kutyniok, Ludwig-Maximilians-Universität München
- Thursday 04 December 2025, 15:00-16:00
- Venue: Centre for Mathematical Sciences, MR14.
- Series: Applied and Computational Analysis; organiser: Matthew Colbrook.
Thu 11 Dec 15:00: Title to be confirmed
Abstract not available
- Speaker: Michael Woodley (University of Sussex)
- Thursday 11 December 2025, 15:00-16:00
- Venue: Centre for Mathematical Sciences, MR14.
- Series: Applied and Computational Analysis; organiser: Georg Maierhofer.
Thu 27 Nov 15:00: Title to be confirmed
Abstract not available
- Speaker: Marita Thomas (Freie Universität Berlin)
- Thursday 27 November 2025, 15:00-16:00
- Venue: Centre for Mathematical Sciences, MR14.
- Series: Applied and Computational Analysis; organiser: Georg Maierhofer.
Thu 20 Nov 15:00: Title to be confirmed
Abstract not available
- Speaker: Eleni D. Koronaki (Luxembourg Institute of Science and Technology)
- Thursday 20 November 2025, 15:00-16:00
- Venue: Centre for Mathematical Sciences, MR14.
- Series: Applied and Computational Analysis; organiser: Georg Maierhofer.
Thu 13 Nov 15:00: Title to be confirmed
Abstract not available
- Speaker: Felix Voigtlaender (KU Eichstätt-Ingolstadt)
- Thursday 13 November 2025, 15:00-16:00
- Venue: Centre for Mathematical Sciences, MR14.
- Series: Applied and Computational Analysis; organiser: Georg Maierhofer.
Thu 30 Oct 15:00: Title to be confirmed
Abstract not available
- Speaker: Pierre-Antoine Absil (University of Louvain)
- Thursday 30 October 2025, 15:00-16:00
- Venue: Centre for Mathematical Sciences, MR14.
- Series: Applied and Computational Analysis; organiser: Georg Maierhofer.
Making way for PEG alternatives
Nature Materials, Published online: 15 October 2025; doi:10.1038/s41563-025-02381-w
Replacing traditional PEG-lipids in lipid nanoparticle formulations with zwitterionic polymer–lipid and brush polymer–lipid conjugates offers enhanced intracellular delivery and reduced immunogenicity, making them promising alternatives to PEGylated nanoparticles.Designing around immune memory to counter PEG immunogenicity
Nature Materials, Published online: 15 October 2025; doi:10.1038/s41563-025-02383-8
The emergence of lipid nanoparticles as nucleic acid delivery vehicles has revolutionized medicine, with polyethylene glycol (PEG)-lipids playing a crucial role in particle formation and in vivo fate. However, PEG has been linked to immune responses that can provoke side effects and may prevent repeat dosing, and so PEG alternatives are now being developed. Here we argue that, rather than concentrating on PEG replacement, the field should prioritize designing around pre-existing immune memory.Grain boundary zirconia-modified garnet solid-state electrolyte
Nature Materials, Published online: 15 October 2025; doi:10.1038/s41563-025-02374-9
Garnet-type LLZO electrolytes are considered among the most promising solid-state electrolytes for all-solid-state batteries; however, numerous challenges need to be addressed before they are integrated into a cell. By precipitating amorphous zirconium oxide onto grain boundaries, increased ionic conductivity is observed and dendrite growth is suppressed.Rapid synthesis of subnanoscale high-entropy alloys with ultrahigh durability
Nature Materials, Published online: 15 October 2025; doi:10.1038/s41563-025-02358-9
A synthesis method mediated by laser-induced plasmon is developed to prepare subnanoscale high-entropy alloys, and a few such alloys display high stability for water splitting in proton exchange membrane electrolysers, operating at 2 A cm−2 for over 1,200 h.A formal Fe<sup>III/V</sup> redox couple in an intercalation electrode
Nature Materials, Published online: 15 October 2025; doi:10.1038/s41563-025-02356-x
An FeIII/V redox mechanism in Li4FeSbO6 on delithiation without FeIV or oxygen formation with resistance to aging, high operating potential and low voltage hysteresis is demonstrated, with implications for Fe-based high-voltage applications.