INVERSE-EMULSION STABILITY: QUANTIFICATION WITH AN ARTIFICIAL NEURAL NETWOR

ABSTRACT

Due to the multiplicity of parameters governing emulsion stability, existing theories generally cannot quantitatively predict the phase separation in oil/water systems. In this work, an artificial neural network, which is known to have a strong nonlinear mapping ability, was used to “learif” the correlation between the factors influencing emulsion stability (phase ratio, surfactant concentration and comonomer concentrations) and the magnitude of phase separation. This has been applied to the water-in-oil copolymerizations of acrylamide with quaternary ammonium cationic monomers. It was found that the ANN can accurately predict the subset of the stability state (stable, metastable, unstable), along with the extent of oil separation for metastable systems.

Keywords:

• inverse-emulsion
• backpropagation (13P) network
• stability
• acrylamide
• cationic monomer

Additional information

Notes on contributors

David Hunkeler

Author to whom correspondence should be addressed at Laboratory of Polymers and Biomaterials, Department of Chemistry, Swiss Federal Institute of Technology, CH-1015 Lausanne, Switzerland