Perovskite solar cells (PSCs) are solution processable, have high efficiencies and promise low cost renewable electricity. Unfortunately, the widespread application of PSCs is being held back by their poor long-term stability. Their established rivals, crystalline-silicon solar cells, offer a 25 year operational lifetime. However, high efficiency PSCs are operationally stable for less than 6 months. Perovskites have very low mechanical toughness due to the intrinsically low energy required to separate perovskite crystals. Solar cell operation lifetime increases with mechanical toughness and we aim to exploit this relationship to greatly enhance the stability of high efficiency PSCs. Taking inspiration from highly tough natural biomaterials (such as nacre) we will use synthetic analogues of adhesive proteins to glue the crystals together and increase perovskite mechanical toughness. Our new particles are ultra-deformable nanometre-sized gel particles (termed ultra-low crosslinked nanogels, ULC nanogels). Building on our earlier work where conventional nanogels improved lead-PSC stability, novel ULC nanogels will be prepared that conformally coat and interlink perovskite crystals. They will flatten to become ultra-thin and allow charges to move unhindered between crystals. We will also study lead-free, tin-perovskites and increase their operational stability by a combination of improvements in chemical stability and mechanical toughness. The link between the mechanical toughness and PSC stability will be investigated experimentally and using state-of-the-art modelling techniques. Modelling will also be used to study the energy changes involved in chemical degradation so as to establish materials design rules for PSCs with enhanced stability. A successful outcome to this project would provide improved fundamental understanding of the interplay between perovskite mechanical toughness and stability as well as a high efficiency demonstrator(s) with a projected operation lifetime of 8 years. Such a result would bring the large-scale deployment of perovskite photovoltaics for CO2-free electricity generation closer and increase energy security.

Notations are the formal manifestation of thought. Conventionally, the established idea of 'notation' is one dimensional, as a sequence of symbols; but the last decade has seen a very exciting development: formal two dimensional notations, based on graphs. "String diagrams" represent a mathematical connection between one-dimensional (term) and two-dimensional (graph) syntax, exploiting the existence of shared mathematical structures (called 'monoidal categories'). However, the one-dimensional manifestation of a string diagram, as it emerges from the mathematical theory, is not always the most practical or most convenient for people to manipulate. Graph languages have been developed independently in other contexts, for example digital circuit specifications, from rather different principles. The popularity of these notations suggests that in some sense they more appealing, yet these notations lack structure and are difficult to reason about mathematically. Graphical programming languages have been proposed numerous times and for a long time, but they consistently failed to reach the mainstream of programming. These recurring attempts testify to the fact that such languages are attractive, yet their failure shows that getting them right is not exactly easy. The mathematical structure of programming languages also has diagrammatic and graphical counterparts, but these have not been yet studied enough. These are "hierarchical" structures in which diagram graphs can nest or can even be used as labels of nodes in other graphs. In this proposal we aim to reconcile both kinds of notations, as emerging from the mathematics of string diagrams and from the practicalities of circuit and system design. The key idea that can bring these two together is the concept of 'name'. Names, used as labels, can greatly simplify certain term representations of graphs and make algorithmic processing of such terms easier. The mathematical theory of names ('nominal theory') is subtle, but by now quite well understood. We believe that nominal theory can bring together term and graph manipulation, via rewriting, in an elegant and effective way. And we further believe that this unify theory will lead to new insights into theoretical (and practical) models of digital circuits and other kind of systems. Zanasi is an expert in the theoretical aspects of string diagrams, Silva is an expert in nominal techniques, and Ghica is an expert in practical applications of string diagrams to language and circuit analysis.