Micrographic Evaluation of Water-in-Oil Emulsion Stability up to Critical Electric Field

Abstract Electrostatic demulsification is widely applied in the petroleum industry for the separation of water-in-oil emulsions; however, the relationship between microscopic droplet dynamics and the macroscopic electrical response remains insufficiently understood, particularly regarding the definition of the critical electric field (CEF). This work presents a quantitative investigation of electrocoalescence using optical microscopy coupled with electrical current monitoring under a direct current electric field. Water-in-oil (W/O) model emulsions were evaluated through a factorial experimental design comprising 81 experiments, in which the electrode gap (100–300 μm), water cut (5–15 wt %), and surfactant concentration (0.5–1.5 wt %) were systematically varied. Real-time observations enabled the identification of droplet polarization, chain formation, coalescence, and the development of conductive pathways. The CEF, ranging approximately from 0.7 to 4.0 kV/cm, was determined from the simultaneous occurrence of current spikes and the visual formation of coalesced droplets that establish conductive pathways within the emulsion. The results demonstrate that droplet alignment occurs at the initial stages of electric field application and does not correspond to a significant increase in current. Instead, the abrupt rise in current was associated with the growth of droplets that establish electrical conduction, indicating that the CEF corresponds to the final stage of the electrocoalescence process. Statistical analysis revealed a pronounced quadratic effect of water cut, while electrode spacing and surfactant concentration also exerted significant influences through linear, quadratic, and interaction effects. These findings provide mechanistic insight into electrocoalescence and support a revised interpretation of the critical electric field, contributing to the understanding and potential optimization of electrostatic separation processes.