Optimising Emerging PV Technologies Through LCOE Modelling

Contact: Professor Chris Groves

Collaborators: Conducted at Durham University with support from the Durham Energy Institute.

Project Overview

This research introduces a Levelized Cost of Energy (LCOE) model to evaluate perovskite (PVK) and organic (OPV) solar cells. Unlike traditional silicon PVs, these new technologies face early-life degradation ("burn-in"), impacting cost-effectiveness. By integrating real-world degradation data, the study identifies strategies to optimise PV performance. Reducing burn-in loss and long-term degradation is often as crucial as improving efficiency. Some state-of-the-art PVs are already approaching cost-competitiveness in wholesale electricity markets.

Key Insights

• LCOE modelling helps select the most viable emerging PVs.
• Efficiency alone isn't enough—longevity and cost also matter.
• Some PVK and OPV cells rival silicon PV in projected costs.

Click here to read more

Decarbonising Electrical Grids Using Photovoltaics with Enhanced Capacity Factors

Contact: Professor Chris Groves

Collaborators: Durham University, Newcastle University, University of Glasgow, and the Durham Energy Institute.

Project Overview

This research explores a novel High Capacity Factor Photovoltaic (CFPV) technology, designed to reduce variability in solar power generation. Unlike traditional silicon PVs, these devices improve efficiency at lower light levels, making them more effective at replacing fossil fuels in national grids. Through UK energy grid modelling, the study shows that CFPV devices can cut reliance on coal and gas more efficiently than conventional PVs. Experimental results confirm that dye-sensitised, perovskite, and organic PV technologies can be engineered for higher capacity factors, reducing the need for costly energy storage.

Key Insights

• CFPV devices outperform silicon PV in reducing carbon emissions.
• Higher capacity factors reduce variability, improving grid stability.
• Custom PV designs enable better alignment between solar generation and demand.

Click here to read more

Designing and Developing Sustainable Battery Materials with Artificial Intelligence

Contact: Dr Mehdi Hedayati

Collaborators: Conducted at Durham University in collaboration with WeLoop.

Project Overview

This research focuses on the design and development of eco-friendly rechargeable batteries using artificial intelligence. With the growing demand for energy storage in renewable systems, consumer electronics, and electric vehicles, there is an urgent need for sustainable alternatives to conventional lithium-ion batteries (LIBs). While LIBs offer high efficiency and long life cycles, they rely on scarce materials and pose significant environmental challenges in terms of cost and recycling.

This project leverages inverse engineering and deep learning to design next-generation battery materials using existing scientific and industrial data. By shifting away from traditional trial-and-error experimentation, the approach aims to accelerate the discovery of high-performance, low-impact battery systems that are both sustainable and scalable.

Key Insights

• AI-driven inverse design enables rapid discovery of new battery materials without trial-and-error experimentation.
• The approach supports the development of environmentally friendly alternatives to lithium-ion batteries.
• Data-driven design accelerates innovation while reducing cost and environmental impact.

Structure and function of electronic interfaces in two-dimensional materials 

Contact: Dr Iddo Amit

Collaborators: Andrew Gallant (Durham), Saverio Russo and Monica Craciun (Exeter), Manish Chhowalla, Yan Wang (Cambridge University), Oliver Rigby (Northumbria University).

Project Overview

Two‑dimensional semiconductors offer an exciting platform for emerging technologies in quantum sensing and on-chip computation. However, when these materials are produced at scale, they inevitably develop structural imperfections that can limit their performance. Our research views these features, such as grain boundaries, variations in layer thickness, and polymorphic regions, as potential opportunities. By studying and deliberately tailoring these unavoidable defects, we aim to deliver new functionalities that will enable novel applications.

Key Insights

We have identified a heterojunction‑like feature (which we termed a quasi‑heterojunction) embedded within a single material. Unlike conventional heterojunctions, which form at the interface between two different materials and are therefore vulnerable to contamination, this internal junction is intrinsically clean. Such a structurally pure interface has the potential to enhance device performance and open new possibilities in sensing, light emission, and other emerging technologies. (See: https://doi.org/10.1021/acsami.5c21803)

Stochastic computational approach to transport in disordered systems

Contact: Dr Iddo Amit

Collaborators: Members of the AMED Group (Durham)

Project Overview

Computational modelling of charge transport in electronic materials is a powerful approach that enables the design of next‑generation electronic devices. Yet the inherently random structure of polycrystalline thin films means that their complexity is difficult to capture accurately without relying on averaging methods, which obscure fine structural details that could be crucial for practical applications.

We are developing a hierarchical-stochastic modelling framework in which each structural feature within a thin film is simulated individually. These components are then combined to construct realistic films with fully randomised architectures. This approach aims to preserve essential microstructural details while enabling more accurate and predictive simulations of charge transport.

Key Insights

We have demonstrated this methodology on one‑dimensional chains and two‑dimensional films in which random dopant distributions influence the overall resistivity. In our case study, we found that conventional averaging approaches can underestimate the resistivity by as much as 18%, highlighting the importance of retaining microstructural detail in accurate transport simulations. (See: https://doi.org/10.1063/5.0231350)