 Source (www.bygg.ntnu.no/batek/dr_ing/bt/bt-abs.htm)
Spruce together with pine are the main wooden building materials used in Norway. Due to the complex composition of the material and the many-sided usage, there still are many un-answered questions concerning moisture transport in wood in buildings. Wood and wooden materials in buildings are exposed to diurnal and annual climatic cycles. Uncertainties concerning transient wood-water relations in buildings and building components have demanded a more thorough investigation on basic moisture transport in wood and related properties.
This thesis is a result of a comprehensive study on hygroscopic moisture transport in wood in general and spruce (Picea abies) in particular. The themes focused in this thesis can be divided into three parts. In the first part, a thorough study of present knowledge of moisture sorption and moisture transport in wood have been conducted. Material properties have also been measured and investigated in this part. It has been shown that diffusion coefficients, Dp, determined from cup measurements are increasing with an increasing average relative humidity across a wooden specimen. It has also been confirmed that if diffusion coefficients, DC, determined from transient sorption measurements are to be converted to Dp , significant variations in diffusion coefficients can be obtained depending on the parameters included in the conversion factor. Absorption and desorption isotherms of spruce (Picea abies) measured in this work have been compared with corresponding measurements on spruce performed in other Nordic countries, and significant deviations have been found.
In the second and main part of this thesis, experimental investigations on wood exposed to cyclic step-changes in climatic conditions have been performed. Two different sets of experiments have been performed. In the main set of experiments an experimental apparatus has been designed and set up in the laboratory at the department. Well known principles for measurement of moisture sorption in wood have been employed. As there is a demand for less labour intensive and more efficient methods for measuring moisture sorption in wood, another set of experiments has been conducted in the laboratory at Department of Chemical and Process Engineering (CAPE), University of Canterbury, New Zealand. In these experiments another, more efficient, principle of measurement, originally meant for drying experiments, has been employed and assessed for measuring moisture sorption in wood.
The results from four different measurement series, conducted at the department, with diurnal and weekly cycles between two different levels have been reported. A two-step sorption process has been observed with the major change in moisture content in the first fast initial part. No significant phase lag has been observed for neither transverse nor longitudinal specimens up to 10 mm thickness. A repetitive pattern in moisture content change is found for both weekly and daily changes. The same level of moisture content is reached in both absorption and desorption every cycle. Although average moisture contents have been measured only, the fast response in moisture sorption indicates that significant moisture gradients are present immediately after a relatively large change in surrounding humidity.
The results from four different sorption experiments, conducted at CAPE, have indicated the presence of the same two-step sorption process, with the major change in moisture content within the first fast initial part of sorption. The specimens in these experiments were exposed to time gradients in both temperature and relative humidity. The overall impression from these experiments is that the apparent "equilibrium" moisture contents are rather high for all temperature and humidity levels. The reason for this has not been possible to fully explain. However, weaknesses in the principle of measurement have been revealed.
In the third part of this thesis, a one-dimensional transient model for moisture transport in wood is set up. The model is based on Fick's law with water vapour pressure and temperature as driving potentials. A model for hysteresis has been proposed and included in the model. Comparisons between the experimental results obtained in this work and calculations have been done. The level of the absorption and the desorption isotherm is the most important parameter in order to obtain good fit between measurements and calculations. A better agreement is found if hysteresis is taken into account, compared to calculations where only an average sorption isotherm is used. It is also found that the diffusion coefficient and its dependency on RH and the density of the wood have a certain influence on the sorption rate. The convection mass transfer coefficient on the other side has very little influence on the average moisture content.
As an approach to modelling moisture transport in wood exposed to cyclic environmental changes in the hygroscopic range, the model has proved promising. The shape of the measured and the calculated curves presented in this work correspond very well. This could imply that application of linear intermediate curves, based on empirical data, could be a suitable way of modelling hysteresis in wood. | 