The hygroscopic properties of thermally modified wood have been studied in terms of adsorption and desorption processes. Poplar (Populus spp.) and European beech (Fagus sylvatica L.) were in focus. The obtained isotherms were parameterized with the models of Hailwood-Horrobin, Guggenheim-Anderson-deBoer, generalized D’Arcy and Watt, and Yanniotis and Blahovec. The changes in equilibrium moisture content (EMC) were quantified, and the accessibility of water vapor to the sorption sites was determined. The monolayer and multilayer sorption was studied and the sorption isotherms were classified. All sorption isotherms were type II, and the type was not changed after the modification. The monolayer sorption was found to be responsible for the reduction in EMC after thermal modification. The observed increase in the hysteresis coefficient was explained by the reorganization of the wood ultrastructure.

The methods so far applied to determine the bound water diffusion coefficient in wood do not provide credible results on this coefficient as well as on the boundary condition. An alternative approach based on the concept of solving the inverse transfer problems was recently applied. Two European species were investigated in the present study. A series of sorption experiments was performed and followed by the numerical identification of the coefficients. Several case studies were carried out for the constant and bound water content dependent diffusion coefficients. The obtained results were validated by comparison to a set of experimental data.

Beech wood, due to its properties, is one of the most versatile and successfully used construction materials. The wood properties could even be improved, and among different wood modification processes, the thermal modification approach is usually considered as an environment-friendly technology based only on the heat and water application during wood treatment. Changes in material properties resulting from the thermal treatment of wood increase applicability of this material, but on the other hand, detailed knowledge of the modified properties is definitely necessary for the proper application of such materials to construction engineering. Unfortunately, credible data on thermal characteristics of thermally modified wood are usually provided in a very limited way, and there is no information on specific heat in particular. An original calorimetric method was used to determine the specific heat of untreated and thermally modified European beech wood (Fagus sylvaticaL.). The inverse modeling was implemented to estimate the anisotropic thermal conductivity, and significant differences were found for the radial and tangential directions. The thermal modification highly influenced the increase in the thermal conductivity in the longitudinal direction. The validation procedure showed credibility of the applied methods, and it is clear that modeling of heat transfer in thermally modified wood leads to erroneous results when using thermal properties determined merely for untreated wood.

Ensuring reliability of data on thermal properties of wood-based panels is important for manufacturing processes, especially when it is recommended to shorten the cooling phase and stack the panels in hot conditions. Prediction of the heat transfer during cooling phase and normal or hot stacking based on accurate data is essentially important for attaining panels of required properties. The thermal properties are also required when designing houses, especially low-energy or passive ones. Therefore, a water calorimeter was adequately designed and constructed to ensure improvement in the accuracy of the specific heat measurements. The calorimeter was used to determine the specific heat. The attained accuracy estimated by the relative error was significantly increased, and the error values were less than 2 % for all types of the investigated particleboard and OSB. In case of low-density fiberboard (LDF), the maximum value of the relative error did not exceed 4 %. It was also shown that high accuracy required for the specific heat measurements was achieved for experiments in which high-mass samples were used, in contrast to measurements for such samples in traditional DSC systems. The results for the specific heat were within the range from 1420 to 1450 J/kg K for LDF and all types of particleboard. The effect of the investigated material density on the specific heat was not found. The only exception was in case of OSB for which the specific heat was ca. 1550 J/kg K, and it was approximately 100 J/kg K higher when compared to other panels.