Surfactant-induced flow compromises determination of air-water interfacial areas by surfactant miscible-displacement

- Surfactant-induced drainage compromises air-water interfacial area measurement.
- Internal redistribution of water compromises interfacial area measurement.
- Physical interpretation of absolute interfacial area estimates is challenging.

Surfactant miscible-displacement (SMD) column experiments are used to measure air-water interfacial area (AI) in unsaturated porous media, a property that influences solute transport and phase-partitioning. The conventional SMD experiment results in surface tension gradients that can cause water redistribution and/or net drainage of water from the system (“surfactant-induced flow”), violating theoretical foundations of the method. Nevertheless, the SMD technique is still used, and some suggest that experimental observations of surfactant-induced flow represent an artifact of improper control of boundary conditions. In this work, we used numerical modeling, for which boundary conditions can be perfectly controlled, to evaluate this suggestion. We also examined the magnitude of surfactant-induced flow and its impact on AI measurement during multiple SMD flow scenarios. Simulations of the conventional SMD experiment showed substantial surfactant-induced flow and consequent drainage of water from the column (e.g., from 75% to 55% SW) and increases in actual AI of up to 43%. Neither horizontal column orientation nor alternative boundary conditions resolved surfactant-induced flow issues. Even for simulated flow scenarios that avoided surfactant-induced drainage of the column, substantial surfactant-induced internal water redistribution occurred and was sufficient to alter surfactant transport, resulting in up to 23% overestimation of AI. Depending on the specific simulated flow scenario and data analysis assumptions used, estimated AI varied by nearly 40% and deviated up to 36% from the system's initial AI. We recommend methods for AI determination that avoid generation of surface-tension gradients and urge caution when relying on absolute AI values measured via SMD.

174