Turbulent eddy diffusion models in exposure assessment - Determination of the eddy diffusion coefficient

ABSTRACT

The use of the turbulent eddy diffusion model and its variants in exposure assessment is limited due to the lack of knowledge regarding the isotropic eddy diffusion coefficient, D_T. But some studies have suggested a possible relationship between D_T and the air changes per hour (ACH) through a room. The main goal of this study was to accurately estimate D_T for a range of ACH values by minimizing the difference between the concentrations measured and predicted by eddy diffusion model. We constructed an experimental chamber with a spatial concentration gradient away from the contaminant source, and conducted 27 3-hr long experiments using toluene and acetone under different air flow conditions (0.43–2.89 ACHs). An eddy diffusion model accounting for chamber boundary, general ventilation, and advection was developed. A mathematical expression for the slope based on the geometrical parameters of the ventilation system was also derived. There is a strong linear relationship between D_T and ACH, providing a surrogate parameter for estimating D_T in real-life settings. For the first time, a mathematical expression for the relationship between D_T and ACH has been derived that also corrects for non-ideal conditions, and the calculated value of the slope between these two parameters is very close to the experimentally determined value. The values of D_T obtained from the experiments are generally consistent with values reported in the literature. They are also independent of averaging time of measurements, allowing for comparison of values obtained from different measurement settings. These findings make the use of turbulent eddy diffusion models for exposure assessment in workplace/indoor environments more practical.

KEYWORDS:

• Diffusion model
• eddy diffusivity
• exposure assessment
• indoor air pollution
• turbulent eddy diffusion coefficient (D_T)

Acknowledgment

The authors thank Yue Shao of Rice University for providing assistance during the experiments.

Funding

This research was made possible by funding under NIOSH 1R010H010093-01A2.

Disclaimer

The authors declare that there are no conflicts of interest relating to the material in relation to this article.