Composites of wood in a thermoplastic matrix (wood-plastic composites) are considered a low maintenance solution to using wood in outdoor applications. Knowledge of moisture uptake and transport properties would be useful in estimating moisture-related effects such as fungal attack and loss of mechanical strength. Our objectives were to determine how material parameters and their interactions affect the moisture uptake and transport properties of injection-molded composites of wood-flour and polypropylene and to compare two different methods of measuring moisture uptake and transport. A two-level, full-factorial design was used to investigate the effects and interactions of wood-flour content, wood-flour particle size, coupling agent, and surface removal on moisture uptake and transport of the composites. Sorption and diffusion experiments were performed at 208C and 65 or 85% relative humidity as well as in water, and diffusion coefficients were determined. The wood-flour content had the largest influence of all parameters on moisture uptake and transport properties. Many significant interactions between the variables were also found. The interaction between wood-flour content and surface treatment was often the largest. The diffusion coefficients derived from the diffusion experiments were different from those derived from the sorption experiments, suggesting that different mechanisms occur.

The gas analysis method according to EN 717 part 2 is a well-known technique to measure the formaldehyde emission of plywood and coated wood-based materials for quality control purposes. There is an increasing interest to apply this method also for uncoated wood-based panels compared to the perforator method (EN 120: 1992) owing to many reasons. As there is still limited experience in testing particleboard with this method, it is the aim of this study to evaluate the effect of some parameters to ensure correct and precise testing using the gas analysis method. Furthermore, the study aimed to show how the m.c. of panels before testing affects the Gm, and to calculate an m.c. correction factor from these results for particleboard with modern bonding systems at a low emission level. Gas analysis value and m.c. show a very good correlation. Based on this correlation an m.c. correction factor can be determined, which facilitates an m.c. correction of Gm data. This may help to improve the quality of Gm data for quality control and data comparison (i.e. for round robin tests). Combining the sampling for hour 3 and 4 as well as 2, 3 and 4 has no significant effect on the level of formaldehyde emission tested. Also the accuracy of testing was not reduced. Consequently, the change in sampling procedure may lead to a reduction on the required number of water analyses needing to be performed in the laboratory, and therefore to a reduction in consumption of chemical consumables and expenditure of human labour. The reduction of overpressure during gas analysis did not have a major effect on the values as such, but increased the coefficient of variation. Also, an increase in air flowrate did not affect the Gm at lower emission levels.