Modeling the evaporation from a thin liquid surface beneath a turbulent boundary layer

The release and dispersion of toxic chemicals can cause a threat to military personnel and the population at large. In order to develop and implement appropriate protective capabilities and plan mitigating measures, modeling, simulation and assessments of hypothetical scenarios and historical incidents is a valuable and widely used methodology. This requires reliable CBRN modeling and simulation capabilities to model how toxic chemicals are released and dispersed in air. A physical and mathematical model of an event involving the dispersion of chemicals can roughly be divided into three parts: source modeling, transport modeling and effect modeling. This study focuses on source modeling. As part of the recent NATO-study SAS-061, research groups in the U.S., The Netherlands and FFI assessed the same scenario which involved release of the chemical warfare agent (CWA) sarin. The groups used an evaporation rate for sarin which varied by a factor of 10. This leads to correspondingly large variations of the calculations of the consequences and the extent of damage. This introduces an unacceptable uncertainty in consequence assessments and clearly demonstrates the need to improve our fundamental knowledge of evaporation processes. The current work is a continuation of work done previously at FFI. Odd Busmundrud developed a model for evaporation from surfaces and droplets. His model is in essence based on molecular diffusion through an assumed wind-free diffusion layer above the surface, and only indirectly takes account for fluid dynamical aspects of the evaporation process. The present study considers the evaporation of a non-buoyant contaminant from a thin liquid surface beneath a turbulent boundary layer. The analysis is based on near-surface asymptotics of turbulence velocity and scalar fluctuations. The objective of the study is to derive and verify an algebraic evaporation model sensitized to boundary layer turbulence. The dependence on the friction velocity is shown to be naturally included in the analysis and the model does not depend on any a priori assumption of the existence of an equilibrium logarithmic boundary layer region. The near-surface asymptotics are fairly universal and thus valid for a wide range of external flow conditions. The model is validated using recent experimental wind tunnel data. This work will be continued by including the model in CFD software. It is crucial to have access to good quality experimental data. In order to isolate the dependence on different aspects on the evaporation process (thermal effects, turbulence, the size of the liquid surface etc), measurements where the different parameters are systematically varied are necessary. It would be of great value to perform such measurements ourselves. Especially for toxic chemicals and chemical warfare agents (CWA) such data are hard to get.