This project examined how neutron reflectometry experiments can be optimised to extract maximum structural information from thin films and buried interfaces while reducing experimental time. Neutron reflectometry is a powerful technique for studying soft-matter and biological systems at solid–liquid interfaces, but data analysis is often limited by parameter correlations and the inherent phase problem. The work addressed these challenges by focusing on substrate engineering and sensitivity-guided experimental design.
Polarised neutron reflectometry was combined with contrast variation and switchable magnetic reference layers to improve decoupling of structural and compositional parameters in reflectivity models. Substrate assemblies consisting of silicon, a ferromagnetic reference layer, and an inert capping layer were systematically designed and evaluated using sensitivity analysis based on Fisher information and parameter correlations. Measurements performed at the POLREF instrument demonstrated that carefully optimised layer thicknesses significantly enhanced sensitivity, allowing multiple reflectivity datasets to be obtained from a single sample under different magnetic and solvent conditions.
The results showed that optimised substrate designs can provide equivalent structural information with up to a fivefold reduction in measurement time. This work established a practical framework for more efficient neutron reflectometry experiments, improving access to neutron techniques and enabling more detailed studies of soft matter and biological interfaces under realistic experimental constraints.