scope:

INTRODUCTION

Many building materials emit volatile organic compounds (VOCs), which tend to cause sick building syndrome (SBS) (EPA US, 1990; Little et al., 1994; Meininghaus et al., 1999; Haghighat et al., 2005; Xiong et al., 2008). Knowing the emission characteristics of VOCs from building materials is necessary to effectively estimate and control indoor air quality. It is found that initial VOC concentration, C₀, partition coefficient, K, diffusion coefficient, D, are the three key parameters of building materials controlling the emission process. In order to obtain such parameters precisely and conveniently, a lot of fascinating work has been done. Haghighat et al. (2002) reviewed researches on the measurement of diffusion coefficients of VOCs for building materials: The cup method (Hansson and Stymne, 2000; Kirchner et al., 1999), twin chamber method (Bodalal et al., 2000; Meininghaus et al., 2000) and porosity test method (Blondeau et al., 2003) were comparised and analyzed. It is found that there can be a difference of up to 700% in the reported data for a given technique.

As far as the twin chamber method is concerned, Bodalal et al. (2000, 2001) used static twin chamber method (i.e. the chamber is sealed and the air flow is in steady state) to measure the diffusion coefficient and partition coefficient. They did not consider the impact of convective mass transfer resistance on the coefficients. Meininghaus et al. (2000) also ignored the convective mass transfer effect on the coefficients by using dynamic twin chamber method. Considering those, Haghighat et al. (2002) proposed to apply mass exchanger method to simulate the VOC transfer of the building material between two CLIMPAQ. In their model, VOC concentration gradients were assumed to be similar as that in the heat exchanger that was unrealistic for many cases due to well mixing of VOC in the chamber. Meininghaus et al. (2000) also pointed out that the flow rate of supply air into the chamber was small compared with the internal recirculation in CLIMPAQ and consequently the concentration gradients within the chamber were negligible. So we assume that the chamber air is fully mixed and just consider the convective mass transfer resistance of the boundary layer outside the material in the chamber.

Based on the above-mentioned analysis, an improvement for dynamic twin chamber method to measure VOC diffusion coefficient and partition coefficient is proposed. A formula describing the influence of the improvement on the measured results is derived based on mass transfer mechanism subsequently. Then the relative error of the traditional dynamic twin chamber method is analyzed by using the formula. In addition, a dimensionless analysis is introduced to further consider the improvement.