Combining microfluidic-SANS with Deep Learning for the Rapid Mapping of Micellar Structure

Surfactants are extensively used in industrial formulations impacting the personal care, food, coatings, oil/lubricant, and agri-chemical sectors, corresponding to an annual turnover of around £200bn in the UK alone. Formulations are often complex multi-component mixtures, comprising of surfactant solutions and various active ingredients, each expected to play a specific role. Microfluidics-SANS will be combined with deep learning analysis to rapidly determine the micellar structure of surfactant systems. The 2D SANS images will be used directly and fed into a convolutional neural network which will then predict the micellar structure. The micellar structures that will be examined are: spherical, ellipsoidal, rod-like and worm-like.