Interactions between wood and coatings with low organic solvent content

The aim of the work described in this thesis is to improve the knowledge on the fundamental interactions between low VOC-coatings and wood, in particular in relation to wood protection in exterior use. To avoid environmental damage and dangerous conditions in the workplace, low-VOC paints have gained increasing importance by the use of waterborne and so called high solids paints. These low-VOC coatings are more and more being used on wood species with: a reduced natural durability against biological decay, higher fluctuations in wood moisture content and reduced dimensional stability. To maintain a good competitive position for wood in the joinery market a longer product lifetime and lower maintenance demands are needed. This will also contribute to a reduction of the total environmental impact during the life cycle of finished wood. Knowledge on the interactions between wood and low-VOC coatings might contribute to the development of technologies that will enlarge the lifetime and protective capacities of wood coatings.

This research has focused on understanding three main aspects of the interaction between coatings and wood. The first one deals with the penetration and wetting of low-VOC coatings, in particular waterborne, coatings. The second aspect deals with the adhesion mechanism of a coating on wood. The third part of the research looks into the sorption of moisture and the related dimensional changes of coated wood. In this respect, special attention has been given to the processes occurring at the surface. The research described in chapter 2 to 7 was carried with model paint systems, based on commercially available raw materials. Both pigmented and unpigmented coatings based on acrylic dispersions, acrylic emulsions, alkyd emulsions, solventborne or high solid alkyd resins were used. The research in chapter 8 and 9 was done with commercially available waterborne and solventborne paints. The majority of the work was done on the wood species: spruce (Picea abies) , pine sapwood (Pinus sylvestris) and meranti (Shorea spp.).

Chapter 1 gives an introduction to this thesis and an overview of the background, the protective capacities and the performance of wood coatings.

Chapter 2 describes the microscopic investigations of the penetration of coatings in wood. The degree of penetration is determined by the possibility of the flow of a coating into the wood capillaries. This depends both on the structure of the wood and the properties of the coating. For softwoods, three potential penetration routes exist: flow in the axial direction from open ends of capillaries, flow into rays ending at the surface and flow from rays through cross-field pits into adjacent axial tracheids. The permeability of the wood only has an influence on the degree of penetration if the penetration depth exceeds the maximum length of a single cell element. In the case of meranti the penetration is limited to the vessels and rays.

In chapter 3 two quantitative techniques for the measurement of penetration are given. The first one measures the maximum depth of penetration into axial direction on a microscopic scale. The second one determines the penetration rate from the volumetric uptake from the axial surface. The maximum penetration depth can be partially related to the diameter of the axial tracheids but both negative and positive correlations exist. Large differences in the maximum penetration depth between coatings were observed. Generally, acrylic dispersion showed the lowest penetration followed by alkyd emulsions, solventborne and high solid alkyd resins. Addition of pigment has a negative influence on the penetration. The viscosity proved to be an important factor to explain differences between coatings. An increasing wood moisture content (2 to 28 %) caused a deeper penetration.

The dynamics of capillary flow into wood are further studied in chapter 4 using the Washburn-model. Accordingly, the penetration rate is proportional to the capillary radius, the liquid surface tension times the cosine of the contact angle and inversely proportional to the viscosity and the height of capillary rise. During the capillary uptake, water or solvent is selectively taken up into the wood. This causes an increasing solid matter content and hence also increasing viscosity and a decreasing capillary pressure. The waterborne coatings showed a very rapid increase of viscosity with increasing mass fraction of binder. At a mass fraction of about 0.45 to 0.55 the viscosity reaches extremely high values, in particular at low shear rates (< 1 s ^-1). With solventborne binders this increase is less pronounced. The rapid increase of the viscosity limits the capillary penetration in most cases. The increase in mass fraction of the binder during capillary uptake was calculated from the analysis of evaporation rates on glass and wood. These showed that at least half of the water or solvent was taken up into the wood on first instance. The addition of a thickener decreased the total capillary uptake and enhanced the selective uptake of water. The Washburn-equation can predict the penetration process fairly well in qualitative terms. To obtain quantitative predictions, a correction for the increase in viscosity is needed because in reality the viscosity increase happens faster than is predicted by the model calculations. Additional research showed that capillary uptake was reduced with decreasing surface tension as long as the wetting is complete, this is in agreement with the Washburn equation.

The surface free energy of wood was determined from the acid-base and Lifshitz-van der Waals components based on contact angle measurements with water, diiodomethane and formamide as described in chapter 5 . Different measuring techniques for the contact angle were compared. The obtained angle was shown to be highly dependent on the way the measurement was performed. This is due to the adsorption of liquids onto the wood, the surface roughness and the chemical heterogeneity of the wood. The surface free energy of wood ranged between 40 and 50 mJ m ^-2with a dominant Lifshitz-van der Waals component. The acid and base parameters were strongly dependent on the measuring conditions.

Chapter 6 describes a newly developed method to measure the adhesion quantitatively. The method is based on measuring the force need to peel the coating from the wood by means of a pressure sensitive tape. This method has the advantage of showing differences in adhesion between early- and latewood zones, corresponding to differences in penetration. A disadvantage is the limited maximum measurable force which restricts its application to measurement under wet conditions (wood moisture content over 25 %) where adhesion is lower. The lower adhesion at higher moisture content can partly be explained by the internal stress due the higher expansion of some coatings in comparison to wood. The interfacial of work of adhesion as it was calculated from the surface free energy of coating and wood was positive under both dry and wet conditions.

The influence of the penetrated coating on the moisture sorption and dimensional change of pine sapwood is described in chapter 7 . If compared to a coating present as a film on the surface, the penetrated coating had a very limited influence on the equilibrium moisture content, the water vapour diffusion rate and the dimensional changes. Only the capillary uptake of water was reduced depending on the type of coating. The small influence of the penetrated coating upon the wood moisture content, can partly be explained by the swelling of the coating itself and the relative low void filling due to shrinkage of the coating after drying.

How a coating can influence the moisture distribution in spruce is given in chapter 8 . During long term measurements of the uptake of water, the moisture distribution was determined experimentally and compared with calculated profiles from the apparent diffusioncoefficient. The most realistic prediction was found if the apparent diffusioncoefficient of unfinished wood was used, taking the impact of a coating into account by a reduced moisture content at the surface. Moisture diffuison is the dominant transport mechanism; it is only close to the surface that capillary water uptake is noticeable, in particular with unfinished wood.

Chapter 9 studies the adsorption and desorption of water and the dimensional change in spruce and meranti as a function of coating type, film thickness, temperature and relative humidity of the environment. The apparent moisture diffusion coefficients are dependent on all these factors and also on the initial and final wood moisture content. The dimensional change can be described as a function of time by a two-parameter regression model. The rate constant in this model is directly related to the apparent diffusion coefficient, as long as the influence of capillary water uptake is small.

The general conclusions of this research are given in chapter 10 . A certain amount of coating penetration into the wood improves adhesion and prevents rapid capillary water uptake. It was shown that coatings, like wood, can adsorb water and swell. Therefore it is recommended to consider the moisture-related aspects of wood coatings in terms of the differences between both materials. To achieve a good wet adhesion, the dimensional change of the coating and the wood should be in the same range. The surface free energy and the work of adhesion between a coating and wood can not explain the observed differences in adhesion between different coatings. The penetration is mainly controlled by the viscosity of the coating as a function of the solid matter content and the shear rate. In this respect the influence of the carrier medium (water or organic solvent) is more important than the chemical composition of the binder. A positive capillary pressure is, however, a necessity to enable the penetration of a coating into wood. If a coating is present as a film on the surface, the moisture sorption and dimensional change of wood is strongly reduced. The strongest fluctuations in wood moisture content take place directly under the surface. Both waterborne and solventborne coatings largely prevent the rapid capillary uptake of moisture. The moisture uptake into coated wood is a lot slower than in unfinished wood. Therefore in practice a permanent reduction of wood moisture content for both water- and solventborne can be expected.

Further information can be obtained from:
SHR Timber Research
Mari de Meijer
P.O.Box 497 NL 6700 AL Wageningen
The Netherlands
m.demeijer@shr.nl