Development of a procedure for estimating the parameters of mechanistic VOC emission source models from chamber testing data

In order to evaluate the impacts of volatile organic compounds (VOCs) emissions from building materials on the indoor air quality beyond the standard chamber test conditions and test period, mechanistic emission source models have been developed in the past. However, very limited data are available for the required model parameters including the initial concentration (C_m0), in-material diffusion coefficient (D_m), partition coefficient (K_ma), and convective mass transfer coefficient (k_m). In this study, a procedure was developed for estimating the model parameters by using VOC emission data from standard small chamber tests. In the procedure, initial values of the model parameters were refined by multivariate regression analysis of the measured emission data. To verify the procedure and estimate its uncertainty, simulated chamber test data were generated by adding 10% experimental uncertainties on the theoretical curve from the analytical solution to a mechanistic emission model. Then the procedure was applied to the generated data to estimate the model parameters. Results indicated that estimates converged to the original parameter values used for the data generation and the error of estimated parameters D_m, C_m0 and K_ma were within ±10%, ±23%, and ±25% of the true values, respectively. The procedure was further demonstrated by applying it to estimate the model parameters from real chamber test data. Wide application of the procedure would result in a database of mechanistic source model parameters for assessing the impact of VOC emissions on indoor pollution load, which are essential input data for evaluating the effectiveness of various indoor air quality (IAQ) design and control strategies as well as the energy required for meeting given IAQ requirements.