Multivariate Statistical Control of Emulsion and Nanoparticle Slurry Processes Based on Process Tomography, Dynamic Light Scattering, and Acoustic Sensor Data

This paper describes the use of multiple on-line sensors including electrical resistance tomography (ERT), dynamic light scattering (DLS) and ultrasound spectroscopy (USS) for real-time characterization of process operations processing emulsions and nanoparticle slurries. The focus is on making novel use of the spectroscopic data to develop multivariate statistical process control (MSPC) strategies. The ERT data at different normal operating conditions was processed using principal component analysis and used to derive two MSPC statistics, T-2 and SPE (squared prediction error) for detecting abnormal changes in mixing conditions. The corresponding particle size distribution was monitored using DLS and USS. Two case studies, a sunflower oil-water emulsion system and a silica suspension system, were examined.