Adhesion and weathering performance of waterborne coatings applied to different temperate and tropical wood species.

Wood is a renewable material that has been recognized for a long time as a valuable material for construction and engineering. Sound, solid wood can last for centuries. Nevertheless, wood is susceptible to biological and physical degradation. Therefore it needs to be protected properly, all the more when using less durable species. A variety of finishes can be applied. These coatings must meet certain minimum requirements, among which are sufficient penetration, resistance to physical and biological discoloration, and good adhesion. The last is considered the utmost important factor for the overall performance of a coating. Without an adequate anchoring to the wood, the coating will peel off prematurely and, thus, affect the lifetime of the substrate. In this article, the impact of the type of coating used, the wood species, and preweathering of the wood on adhesion are disclosed. In addition, the effect of weathering on the wood-coating system as a whole is looked at as well. The adhesion measurements were obtained with a mechanical torque device developed at VTT in Finland.

Keywords: Acrylics, alkyds, adhesion, solvent-based, water-based, wood

**********

Wood is still one of the major materials in the building market. However, there is a keen competition between synthetic materials and aluminum in regards to their use in exterior joinery. Wood may need maintenance, which can be seen as a drawback, however it is highly appreciated for its aesthetic value. It is a renewable resource and, when properly treated, the life expectancy is satisfying. (1) Therefore, coatings are important to reinforce the competitive position of wood. A wide range of coatings is available on the market, from opaque systems that completely mask the substrate, to transparent systems which accentuate the splendor of the wood. For many decades solventborne opaque and transparent primers have been used as the preeminent coatings for wood protection and have become common practice. (2) Nevertheless, the environmental profile of these coatings is below current standards due to their high solvent evaporation during processing as well as application. (3) Moreover, the European VOC regulations force the coatings industry to develop new, less emitting systems, as VOC emissions have to be reduced substantially in comparison with the emission level of 1990. Development of such systems started at the end of the 1980s, yet extra effort is required to improve their durability. However, enhancing the performance of a coating is useless without good application. Misunderstandings in the use and application of the new coatings hindered a smooth public acceptance and their overall breakthrough. (4,5) In addition to a code of good practice, the performance is strongly influenced by the characteristics of the substrate and the coating itself. Adhesion especially is one of the most complex parameters that determine the (long-term) performance of a coating system, all the more since the characteristics of the wood as well as those of the coating must be considered. (6,7)

Adhesion can be defined as the reversible thermodynamic process of separating the unit area of two phases that originally had a common interface. (8) The theoretical adhesion strength is the result of all the interfacial and intermolecular forces. The practical adhesion is the force needed to detach the coating from the surface and thus takes into account the combined action of mechanical interlocking and chemical bonding. (9,10) Still, a consensus has not yet been reached concerning the relation between coating penetration and adhesion. Although several researchers came to the conclusion that there is no relation between penetration and adhesion, (11,12) it was pointed out that varying degrees of substrate penetration correspond to differences in adhesion between early- and latewood zones. (13-15) Nevertheless, practical adhesion measurements will never reach the theoretical calculations due to flaws like blisters and cracks.

It is recognized that the adhesion of a coating to wood is affected by numerous aspects like the wood species and its structure and characteristics (early- or latewood, sap- or heartwood, extractives, surface roughness, etc.), moisture content, chemical and biological variation in the wood, chemical affinity between the coating and substrate, and aging of the coating.

The goal of this work was to examine the adhesion of different weathered and unweathered coatings on seven wood species, preweathered or not preweathered. The comparison of adhesion data with visual appreciation of the weathered coatings was not a focus objective. Techniques frequently used to measure the adhesion of a coating on wood are: the axial pull-off test with a dolly glued on the coating (ASTM D4541, ISO 4624, BS EN 24624), shear measurements in torque mode, block-shear tests, the semi-quantitative cross cut tests, e.g., ASTM D3002 (AFNOR NF T 58-112), ISO 2409, (16) and the AS/NZS 1580.405.1:1994. (17) In this work a torque device was used for the measurements. Within-board variation is expected to be considerable due to a relative small measurement area and earlywood-latewood differences, although repeated measurements will average this effect. In this experiment, between-board variation should be low, due to a proper selection of the material and the fact that test boards originate from the same wood panel. Therefore, every wood-coating combination is treated as an individual in a repeated measures model. The torque test is related to the pull-off test, but the direction of the detaching force is different. The applied shear force in the torque test is parallel to the paint surface, while the force is perpendicular in the pull-off test. (14) However, the limitations of all adhesion test methods should be recognized as the strength of several interphases is tested simultaneously. Therefore, in evaluating adhesion, all factors--regardless of how remotely they might be related--should be taken into account. Since there is no universal test method that can accurately and quantitatively determine adhesion values, conservative evaluations are recommended. Therefore, the available methods are most useful in providing relative ratings for a series of coatings exhibiting significant differences in adhesion. (18)

[FIGURE 1 OMITTED]

EXPERIMENTAL

Two different experiments were performed. At first, the influence of a weathered wooden substrate prior to coating was examined (experiment 1) and, secondly, the weathering of the coatings was evaluated (experiment 2). In both cases the same wood species were used: Pinus sylvestris L. (Scots pine [sapwood]), Pseudotsuga menziesii Mirb. Franco (Douglas fir), Fagus sylvatica L. (beech), Distemonanthus benthamianus Baill. (movingui), Baillonella toxisperma Pierre (moabi), Shorea spp. (meranti), and Pterocarpus soyauxii Taub. (padauk). The coating types tested are listed in Table 1 and consist of waterborne acrylic, alkyd, and hybrid resin opaque primers (O) as well as transparent primers (T) (hybrids have a mixture of two binder types). For the purpose of comparison, a solvent-borne alkyd and a high-solid alkyd paint were added. Modified coatings aim at improved performance, i.e., better penetration, and for adhesion due to lower binder quantity or different binder type.

Preparation of Test Samples

ADHESION ON PREWEATHERED WOOD (EXPERIMENT 1): The planed wood panels cut from straight grained material were 0.5 m in length and had a cross section of 40 mm x 10 mm. The tangent of the annual rings of the specimens was close to an angle of approximately 45[degrees] with the horizontal surface. The samples, two panels for each paint-wood combination, were divided into three groups upon conditioning for three weeks at 20[degrees]C and 65% RH. A first set of specimens was not preweathered and was used as a reference. A second set was aged with a weathering cycle (W1) prior to coating, which was comprised of six days of exposure in the Atlas UVCON weathering device with limited water spraying, i.e., continuous light and repeating cycles of 102 min without water spraying followed by 18 min of water spraying per day and finally, after six days, one day of storage in a deepfreeze. A third set of samples was subjected to a weathering cycle (W2) prior to painting, including six days in the UVCON with a high regime of water spraying, i.e., 23 hr of light, 1 hr of darkness, and alternating exposures of 4 hr of water spraying, 2 hr dry, 10 hr water spraying, 2 hr dry, and 6 hr of water spraying per day ending after six days with one day of storage in a refrigerator. A series of W1 and W2 cycles causes swelling and shrinkage of the wood, stressing the coating system severely. After the artificial preweathering, the test samples were reconditioned at 65% RH and 20[degrees]C for approximately seven days. Finally, the coatings, as listed in Table 1, were applied by brush (two layers). The specimens then remained in the conditioning room for at least three weeks.

[FIGURE 2 OMITTED]

ADHESION OF WEATHERED PAINT SYSTEMS (EXPERIMENT 2): Planed wood samples with an initial length of 50 cm along the grain and a cross-section of approximately 4 x 1 [cm.sup.2] were conditioned at 20[degrees]C and 65% RH. Solely, the standard opaque and transparent primers from Table 1 were applied here in a two-layered system (without modified coatings). The samples were cut to a length of 32 cm. All coated wood specimens were first subjected to a W2 cycle of 500 hr in the Atlas UVCON weathering device. The artificial weathering continued for another 500 hr (W1-W2-W1 cycle) for TW3 and 2 x 500 hr (W1-W2-W1 and W2-W1-W2 cycles) of exposure for all opaque primer formulations. The same reference series as in experiment 1 was used.

ADHESION MEASUREMENTS: Adhesion was measured with a mechanical torque device as described and developed by Holloway and Parker, (19) improved by Ahola, (14,20) and used by Bardage and Bjurman. (21) For the preparation of the adhesion measurements, steel studs had to be glued onto the roughened surface (sanding paper, P400) with a two-component epoxy glue. After curing overnight, a slot was drilled in the coating film around the studs. The studs with adhered paint were disconnected from the wood while measuring the moment of detachment from which the shear stress could be calculated using the following equation:

T = 16I/[[pi][d.sup.3]] (1)

With: T = shear stress (N/[m.sup.2])

I = maximum moment (Nm)

d = diameter stud (m) = 0.017 m

Besides the shear stress measurements, the failure types and their percentages were observed. The failure types included: glue failure (failure of the epoxide at the pin/paint interphase), cohesive coating failure, cohesive wood failure, failure between the coating layers, and failure of the wood-coating interphase (Figure 1). Replicates in which the glue failure exceeded 15% of the area of the underside of the steel stud/pin were discarded, as it was assumed to be attributed to incorrect preparation of pin or sample. By all means, glue bonding strength should be fairly larger than coating cohesion or adhesion of the coating-wood interphase. A minimum of five replicates was taken for each mean adhesion value, as each paint-wood combination was treated as one single subject.

PENETRATION MEASUREMENTS: Penetration was measured using the technique described by Van den Bulcke et al. (22) Coated wooden cubes of approximately 1 x 1 x 1 [cm.sup.3] were impregnated with a safranin solution. A confocal laser microscope focuses a laser beam onto the microtomed transverse section of the blocks and excites the chromophores in the coating and the wood. The safranin makes it possible to separate the coating from the wood by proper selection of the wavelength filters. Subsequently, the image is processed to obtain the penetration of the coating in the earlywood.

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

RESULTS AND DISCUSSION

Adhesion to Unweathered Wood

In Figure 2 the results of the adhesion test over several unweathered wood species are plotted for the various types of wood coatings under test.

To retrieve significant differences, statistical analysis of the data set was carried out. A general linear model (GLM) with repeated measures clearly showed the significant influence of coating (F = 219.66, p < 0.05), wood species (F = 22.29, p < 0.05), and wood-coating interaction (F = 4.76, p < 0.05); coating influence was strongest, as expected from the data. The results of a Tukey test for the coatings showed that only OW1 and TW3 had a similar adhesion on different wood species. The order (low to high adhesion) was as follows:

OW2 OS2 OW3 TW3 OW1 OS1

When conducting the same test for wood species, the adhesion performance ranking was:

[ILLUSTRATION OMITTED]

Meranti had the lowest adhesion values and its measurements were generally comparable with those of softwood. On the contrary, beech showed high shear stress values for almost all types of coatings applied, which might be related to its more homogeneous and fine texture with numerous small vessels. The adhesion of the waterborne alkyd (OW2) and the high-solid alkyd primer (OS2) was rather poor for most wood species tested. The finer textured species like beech, moabi, and movingui were generally performing better again, here.

Relation Between Penetration and Adhesion

When comparing the results of adhesion of pine and beech with penetration in the early-wood, there was an upward trend visible (Figure 3). Both parametric (Spearman) and nonparametric (Kendall's-tau b) tests confirm the correlation between penetration and adhesion (significant on the 5% level).

This upward trend was also found by Rijckaert et al., (23) who studied similar types of coatings on pine sapwood. While the effect of penetration on adhesion is quite distinct for both temperate wood species, pine sapwood and beech (Figure 3), it is rather doubtful for the tropical species under test. The microscopic analysis revealed that for tropical wood species the coating penetration data are largely influenced by the vessel penetration, while other anatomical structures (e.g. fibers) were not penetrated. (22)

Adhesion to Preweathered Wood (Experiment 1)

In Figures 4 and 5 the adhesion data for coatings applied on preweathered wood surfaces exposed, respectively, to a relative dry (W1) and moist (W2) aging prior to coating, are presented.

When running the GLM test for the data of W1 weathered test samples, the influence of coating (F = 20.79, p < 0.05), wood species (F = 20.95, p < 0.05), and the coating-wood interaction (F = 1.63, p = 0.031) was significant again. Grouping and ranking of the samples according to coating and wood species gave the following results:

[ILLUSTRATION OMITTED]

Weathering exposure to a W2 cycle, coating (F = 9.96, p = 0.05), wood species (F = 8.57, p = 0.05), and coating-wood interaction (F = 2.48, p < 0.05) all had a significant impact on weathering. Grouping and ranking for coatings and for wood species resulted in:

[ILLUSTRATION OMITTED]

The shear stress values obtained for detaching the coating films from the preweathered specimens was compared with the data of the unweathered ones. It seems that aging of the wood in the UVCON had a negative effect on the adhesion of the coating, which is mainly the case for softwoods. With the exception of the high-solid primer OS2 and the alkyd emulsion OW2 over the wood preweathered with a W2 cycle, it can be claimed that aging of bare wood specimens results in a reduction of the shear stress of all coatings. Arnold et al. (24) found for oil-based non-film forming stains that preweathering has only a negative effect on the cost, i.e., preweathered surfaces are rougher and more coating is applied as compensation for the surface degradation of the wood. Although there is a weaker coating-substrate interphase, the same level of protection is reached. Similar conclusions were obtained by Williams and Feist (15,25) for wood that was preweathered under natural conditions. On the contrary, for opaque or film-forming systems, the situation is different, as Williams et al. (25) found that preweathering led to decreased finish life. Duration of preweathering also seems to influence service life. (26) Williams et al. (6) and Thay and Evans (27) concluded that preweathering influenced the early- and latewood adhesion in a different way. Normally, earlywood adhesion is highest; but when preweathered wood is used, latewood adhesion increases as earlywood adhesion decreases, which, according to the researchers is ascribed to a higher penetration in the latewood due to the weathering and a more rapid degradation of earlywood. The high percentage of cohesive wood failure for pine sapwood and Douglas fir can be explained in that way. A specific pattern over the detached quarter sawn surface can be observed. Parallel to the fiber direction, the cohesive wood failure mainly occurs over the earlywood, whereas over the latewood the dominating failure types are wood-coating interphase failure and cohesive failure between the coating layers. Similar observations for pine sapwood were made by Rijckaert et al. (23) and Williams et al. (6) The average shear stress value showed the largest decrease with the reference solventborne alkyd OS1.

[FIGURE 7 OMITTED]

Upon comparison of the data of the samples preweathered with the W1 and W2 UVCON cycles, it seems that the loss in average shear stress value was quite similar, with the exception of the standard OW1 on pine and beech, OW3 on pine and padauk, OS1 on beech and moabi, OS2 on Douglas, and TW3 on meranti and movingui. The data recorded for pine sapwood exhibiting higher average shear stress values for preweathered specimens under a W2 UVCON regime confirmed earlier results reported by Rijckaert et al. (23) Padauk behaved similary; other wood species, however, like beech and meranti, yielded higher adhesion forces after a W1 preweathering cycle. This led to the overall conclusion that for both preweathering cycles applied, the adhesion data were strongly influenced by the wood species and the type of coating.

[FIGURE 8 OMITTED]

[ILLUSTRATION OMITTED]

[ILLUSTRATION OMITTED]

The highest shear values after preweathering of the wood surface were found for moabi and padauk. This is in contrast with the unweathered specimens where the highest average adhesive strength was noted for beech. It is supposed that the removal of extractive components during preweathering, in the case of tropical wood species, might affect the wood-coating interphase strength. When looking into the values of the different paint systems, the largest differences were found for the waterborne emulsion type OW2, where the lowest average shear stress values were found for the W2 preweathering cycle.

The waterborne acrylic OW1, the alkyd emulsion OW2, the alkyd-acrylic hybrid OW3, and to a lesser extent also their respective modifications OW1modl, OW2modl, and OW3modl (see Figures 6 and 7), showed a high percentage of cohesive failure between the coating layers.

It was noticeable that in some cases this seemed to improve after preweathering of the wood, probably due to a decrease of the wood-coating interphase strength and, hence, an increase of the wood-coating interphase failure. Such increase was observed for almost all the coating systems applied on the different wood species after preweathering.

The modified coatings in general showed better wood-coating adhesion, although not significant when t-tested, than the standard coatings after a W1 preweathering cycle, except for TW3mod1. This was confirmed by a Tukey test, ranking the standard and modified systems (Scheme 1). In general, comparing the means of the standard (unmodified) and modified coatings with a t-test, only OW2mod1 and TW3mod1 differed significantly from their standards. However, after a W2 cycle, OW2modl showed an overall better performance, probably due to the longer oil length (Scheme 2).

Weathering Performance of the Coating Systems

Figure 8 summarizes the results of the shear test for the weathered coated samples, i.e., weathering was done on the painted panels. It is worthwhile noting that all measurements failed on padauk for the OW1 coating, presumably because of an interaction with extractive components.

Upon comparison with the adhesion data of Figure 2 for unweathered wood species, it is remarkable that the adhesive strength of nearly all types of coating after weathering proved to be noticeably higher irrespective of the wood species. It is assumed that for short weathering exposure (< 500 hr), as applied for the transparent waterborne TW3, the solidity of the polymer network structure of the coatings increased due to further curing catalyzed by heat and UV radiation and the removal (through washing out) of weakening products. Nevertheless, statistical significant differences (p < 0.05) before and after weathering were found in only a few cases (Table 2). For example, adhesion of OS2 on all wood species, except for padauk, differed significantly and significant differences were found for OW2, OW3, and OS1 on meranti.

In general, a GLM with Tukey test on all the data (unweathered, W1, and W2 preweathering of the wood and 500 hr of weathering of the coated substrates) resulted in significant influence of weathering, coating, wood, and all the interaction terms. Ranking them for adhesion according to weathering, coating, and wood species, respectively, produced:

[ILLUSTRATION OMITTED]

The shear stress necessary to detach the coating film of the waterborne acrylic primer OW1 was high. The dominating types of failure noted here were cohesive coating and wood-coating interphase failure. The waterborne alkyd emulsion OW2 revealed a low average shear stress value for the weathered samples, as was the case too for the preweathered samples. For the majority of the tested wood species, the principal failure type was adhesion failure in the wood-coating interphase. In contrast with the unweathered wood species coated with OW3, the phenomenon of failure of the coating layer was not present with the weathered samples. The wood specimens revealed a moderate average shear stress value. The coating with the highest adhesive strength was the reference solventborne alkyd paint OS1. The hybrid transparent primer TW3 apparently had quite good adhesion to wood, and the difference in shear stress value between the weathered and unweathered samples was, as with the solventborne alkyd OS1, insignificant.

CONCLUSIONS

Several standard and modified coatings were applied on seven wood species. The adhesion of the unweathered samples was measured and the results indicated better performance of the finer textured species like beech, moabi, and movingui, as supported by statistical analysis. Next to adhesion, penetration of the coating in the substrate was also measured with confocal microscopy. An upward relationship was found between adhesion and penetration. The influence of weathering on adhesion was studied as well. Wood samples were subjected to two different weathering regimes (W1 and W2). As was for the unweathered samples, the effect of preweathering was significantly influenced by wood species, type of coating, and the wood-coating interaction. Hardwood species showed better results than softwoods. The modified coatings did not adhere better than their standard equivalents. Further on, coated samples were weathered for 500 hours. Softwoods gave less adherence of coatings than the hardwood species. Preweathering of the substrate in general had the most severe negative impact on adhesion. On the contrary, short-term weathering of coated specimens yielded better adhesion.

ACKNOWLEDGMENTS

The authors appreciate the support given by Pirjo Ahola of VTT, Espoo, Finland, for the adhesion measurements. The authors are also grateful for the various coating formulations made available by Akzo Nobel Decorative Coatings, Vilvoorde, Belgium.

References

(1) Chang, S.T., Hon, D.N.S., and Feist, W.C., "Photodegradation and Photoprotection of Wood Surfaces," Wood and Fiber, 14, No. 2, 104-117 (1982).

(2) Probst, F., Laborie, M.P., Pizzi, A., Merlin, A., and Deglise, X., "Molecular Mechanics Experimental Methods Applied to Varnish/Primer/Wood Interactions," Holzforschung, 51, No. 5, 459-466 (1997).

(3) Watson, I., "Polymers and the Art of High Solid Coatings," Polym. Paint Coat. J., 11, No. 29-30 (1995).

(4) Bodner, J., "Biologische Lacke," Holzkurier, 46, No. 15, 18 (1990).

(5) Martin, C.S., "Exterior Wood Stains and Lasures for the Future," Pigment & Resin Technology, 25, No. 5, 29-33 (1996).

(6) Williams, R.S., Winandy, J.E., and Feist, W.C., "Paint Adhesion to Weathered Wood," J. COAT. TECHNOL., 59, No. 749, 43 (1987).

(7) Williams, R.S., Plantinga, P.L., and Feist, W.C., "Photodegradation of Wood Affects Paint Adhesion," Forest Products Journal, 40, No. 1, 45-49 (1990).

(8) Myers, D., Surfaces, Interfaces and Colloids. Principles and Applications, Wiley-VCH, New York, NY, 1999.

(9) Allen, K.W., "A Review of Contemporary Views of Theories of Adhesion," J. Adh., 21, No. 261-277 (1969).

(10) Pizzi, A., Advanced Wood Adhesives Technology, Marcel Dekker Inc., New York, NY, 1994.

(11) Gray, R.V., "The Wetting, Adhesion and Penetration of Surface Coatings on Wood," JOCCA, 44, No. 11, 756-780 (1961).

(12) Van Loon, J., "The Interaction Between Paint and Substrate," JOCCA, 49, No. 844-867 (1966).

(13) de Meijer, M., "Interactions Between Wood and Coatings with Low Organic Solvent Content," Ph.D. Thesis, University of Wageningen, Wageningen, The Netherlands, 177, 1999.

(14) Ahola, P., "Adhesion Between Paint and Wood Substrate," JOCCA-Surf. Coat. Int., 74, No. 5, 173-176 (1991).

(15) Williams, R.S. and Feist, W.C., "Durability of Paint or Solid-Color Stain Applied to Preweathered Wood," Forest Products Journal, 43, No. 1, 8-14 (1993).

(16) de Meijer, M. and Militz, H., "Wet Adhesion of Low-VOC Coatings on Wood--A Quantitative Analysis," Prog. Org. Coat., 38, No. 3-4, 223-240 (2000).

(17) Dawson, B.S.W., Kroese, H.W., and Hong, S.O., "Pre-Service Adhesion of Coating Systems Based on Alkyd Primer Applied to Light Organic Solvent Preservative (LOSP) Treated Radiata Pine Sapwood Boards," Holz Als Roh-Und Werkstoff, 59, No. 6, 422-430 (2001).

(18) Baghdachi, J.A., "Adhesion Aspects of Polymeric Coatings," FSCT Series on Coatings Technology, FSCT, Blue Bell, PA, 1996.

(19) Holloway, M.W. and Parker, P.A., "A Method for Measurement of Adhesion of Surface Coating Under Service Conditions," JOCCA, 47, No. 812-823 (1964).

(20) Ahola, P., "Adhesion Between Paints and Wooden Substrates--Effects of Pretreatments and Weathering of Wood," Materials and Structures, 28, No. 180, 350-356 (1995).

(21) Bardage, S.L. and Bjurman, J., "Adhesion of Waterborne Paints to Wood," J. COAT. TECHNOL., 70, No. 878, 39 (1998).

(22) Van den Bulcke, J., Rijckaert, V., Van Acker, J., and Stevens, M., "Quantitative Measurement of the Penetration of Waterborne Coatings in Wood with Confocal Lasermicroscopy and Image Analysis," Holz Als Roh-Und Werkstoff, 61, No. 4, 304-310 (2003).

(23) Rijckaert, V., Stevens, M., Van Acker, J., de Meijer, M., and Militz, H., "Quantitative Assessment of the Penetration of Waterborne and Solventborne Wood Coatings in Scots Pine Sapwood," Holz Als Roh-Und Werkstoff, 59, No. 4, 278-287 (2001).

(24) Arnold, M., Lemaster, R.L., and Dost, W.A., "Surface Characterization of Weathered Wood Using a Laser Scanning System," Wood and Fiber Science, 24, No. 3, 287-293 (1992).

(25) Williams, R.S. and Feist, W.C., "Effect of Preweathering, Surface-Roughness, and Wood Species on the Performance of Paint and Stains," J. COAT. TECHNOL., 66, No. 828, 109 (1994).

(26) Williams, R.S. and Feist, W.C., "Duration of Wood Preweathering: Effect on the Service Life of Subsequently Applied Paint," J. COAT. TECHNOL., 73, No. 920, 65 (2001).

(27) Thay, P.D. and Evans, P.D., "The Adhesion of an Acrylic Primer to Weathered Radiata Pine Surfaces," Wood and Fiber Science, 30, No. 2, 198-204 (1998).

Jan Van den Bulcke, Veerle Rijckaert, Joris Van Acker, and Marc Stevens ([dagger])--Ghent University*

* Laboratory of Wood Technology, Faculty of Agricultural and Applied Biological Sciences, Coupure Links 653, 9000 Ghent, Belgium.

([dagger]) Author to whom correspondence should be addressed. Email: [e-mail].

Table 1 -- Composition of the Opaque and Transparent Primers

                                                    Solids  PVC   Water
Coating Type      Code (a)  Resins                  wt%     %     %

Waterborne        OW1       Primal MV-23 LO         52.4    34.3  46.6
  acrylic         OW1mod1   Primal MV-23 LO         47.5    25.1  51.3
Waterborne alkyd  OW2       Uradil XP 554 AZ        58.5    35.6  40.6
                            (O.L. 40%) (b)
                  OW2mod1   Uradil XP 516 AZ        58.5    35.6  40.6
                            (O.L. 63%) (b)
Waterborne        OW3       NeoCryl XK84            49.1    22.7  43.8
  acrylic-alkyd   OW3mod1   65% Mowilith LDM 777    48.7    22.8  43.1
                            35% Reshydrol AY 586 W
Solventborne      OS1       Setal 270 WS 70         69.9    16.3  46.1
  alkyd
High-solid alkyd  OS2       Akzo Nobel Resins EPL   60.0    32.1  25.5
Waterborne        TW3       46% Mowilith LDM 777,   35.74         61.55
  acrylic-alkyd   TW3mod1   30% Reshydrol AY 586 W
                            unknown (commerical
                            coating)

(a) O = opaque; T = transparent. (b) O.L. = oil length.

Table 2 -- Results of t-Tests Comparing Coating Adhesion Before and
After 500 hr of Weathering

          OW1    OW2    OW3    OS1    OS2    TW3

Pine      0.643  0.645  0.562  0.227  0.003  0.061
Douglas   0.424  0.060  0.061  0.303  0.007  0.186
Beech     0.431  0.191  0.025  0.125  0.020  0.023
Movingui  0.131  0.299  0.003  0.017  0.002  0.002
Moabi     0.045  0.017  0.09   0.804  0.001  0.498 (a)
Padauk    --     0.023  0.053  0.317  0.080  0.067
Meranti   0.082  0.009  0.006  0.007  0.031  0.453

(a) Assumption of normality not fulfilled; nonparametric Wilcoxon test.