Advances in Roll to Roll Slot Die Coating: Process Optimisation, Colloidal Crystallisation and Advective Assembly

Slot die coating is a precision thin film manufacturing technique that deposits a liquid layer onto a moving substrate through a narrow channel. Its excellent uniformity, high throughput and low material waste make it a key manufacturing process for Li-ion batteries, solar photovoltaics and medical diagnostics.

Despite its widespread use in manufacturing cutting-edge devices, slot die coating is typically optimised through trial-and-error experimentation, provides limited control over crystal structure, and usually deposits only one fluid at a time. To address these limitations, this thesis develops new strategies to advance roll to roll slot die coating.

First, a machine learning based optimisation method is developed to improve coating properties such as uniformity and thickness accuracy. This model is used to identify the process parameters that influence coating uniformity. The predictions are mapped onto established coating theory and used to suggest optimal parameter combinations. The predictions are experimentally validated, producing significant improvements in coating quality.

Next, the capability of slot die coating to produce colloidal crystals with tunable structures is demonstrated by varying process parameters, including pump rate and substrate velocity. A scaling analysis reveals how drying dynamics influence the final crystal morphology. In particular, the importance of colloidal particle size relative to drying rate in avoiding disruption of self-assembled layers is experimentally demonstrated.

Finally, devices integrating advective assembly with slot die coating enable one-step fabrication of structured multi-component coatings, including stripe and layered architectures. Optical imaging and synchrotron μCT confirm that 3D printed coating heads generate the required fluid patterns, and that these structures are preserved in the final thin film. Variation in pump rate controls the architecture.

Collectively, this thesis introduces machine-learning-guided process optimisation, reveals key parameters governing coating quality, and establishes slot die coating as a versatile platform capable of producing colloidal crystal structures and multi-phase architectures.