Abstract. Diffusion and partition coefficients are two key parameters of importance for the characterization of building materials as sources and sinks of air pollutants indoors. The number of coefficients for each pair 'chemical compound -material surface' available in the literature are still scarce and with a high discrepancy in the results obtained by different methods, even for the same method [1]. The objectives of this study were (1) to develop a simple and rapid experimental method for directly measuring the diffusion coefficient (D), and (2) to perform determinations with both the new and the cup method comparing the results obtained. The new method is based on two existent methodologies, the passive sampling on Tenax TA and the dry cup method, proposed by Haghighat et al. [1] and also developed by this research team [2]. The experiments were carried for three different building materials (cork based material, gypsum board and PVC) and three different compounds (cyclohexane, n-octane and m-xylene). The discussion of the results obtained for D by the two methods is presented in this paper. Observed differences were not of the same order of magnitude for the pairs compound/material studied, varying from the order of 10 1 in the case of PVC to 10 2 in the gypsum and in the cork. However some facts were the same: the gypsum board presents the higher values of D and PVC the lower values of D for the three compounds studied in both methods. One of the causes of these marked differences could be high difference of the concentration of the compound in the air between the methods, saturation value in cup method (dozens g/m 3 ) and low values typical from indoor environments (few mg/m 3 ) in the proposed method. IntoductionThe quality of the indoor environment of non-industrial buildings has been receiving a growing interest from the scientific community due to the implications of the 'non quality' on the health and well being of the occupants. Many studies present construction and cladding materials as main sources of indoor air pollution, particularly of volatile organic compounds (VOC).In this context the development of mathematical models for the prevision of the concentration of pollutants in real situations requires the previous study of the behaviour of the materials vs pollutants to quantify the parameters involved, represented namely by the diffusion coefficients and partition coefficients [3]. The available data in the literature for those coefficients is still scarce and, in general, with highly dispersed values. The main objective of the present work was to develop a new method for determination of D. The values of D were also determined by the cup method for comparison. Research methodsDiffusion is the mode of mass transfer of a compound caused by a concentration gradient. Diffusion transport of gases in building materials involves a complex variety of processes including molecular, Knudsen and surface diffusion within the porous interstices [4]. Several experimental techniques have been used to esti...

Diffusion and partition coefficients are key parameters for the characterization of building materials as sources and sinks of air pollutants indoors. A simple and rapid experimental method for direct measurement of the diffusion coefficient (D) was presented in a previous article [1]: the modified dry cup method. This method was based upon two existent methodologies, the passive sampling on Tenax TA and the dry cup method, proposed by Haghighat et al. [2]. In the present article, some improvements were made in the experimental set-up in order to obtain more precise results. The previous cup was modified by increasing the number of Tenax tubes, connected to the bottom of the cup, from one to five. The results obtained with both cups were compared. Measurements with single compounds and with three compounds simultaneously were compared in order to evaluate the influence of competition between molecules of different compounds.