Recognizing trends in the impact of properties of the oil phase and emulsifiers on emulsion breakdown time with high-throughput image analysis and machine learning model.

Emulsion stability is a complex problem that has been studied for decades and is of interest to many industries. However, the impact of the oil properties on the overall emulsion stability needs to be better understood. We present a systematic study of the impact of properties of oil phases and emulsifiers on the stability of emulsions over time. A high-throughput image analysis method is utilized to profile the stability of emulsions across a wide range of materials and curate a dataset. Subsequently, machine learning (ML) methods are leveraged to discover the underlying relationships between the molecular properties of the formulation of an emulsion and its stability. An ML model is trained on the collected data to accurately predict emulsion stability from formulation and thus efficiently guide the emulsion design for a desired breakdown time.