Printer Friendly

Evaluation of varnish coating performance for three surfacing methods on sugar maple wood.

Abstract

Sanding is the most common surfacing method preceding wood coating. Sanding, cross-grain helical-knife planing, and oblique-knife pressure-bar cutting were evaluated in terms of the surface quality of sugar maple wood for coating. Sanded surfaces presented the highest roughness and the best wetting properties. The oblique-knife pressure-bar cutting provided the smoothest surfaces, but produced slight tom or fuzzy grain. Sanding and cross-grain helical planing produced surfaces with no visible defects and yielded the best pull-off adhesion before an accelerated aging; however, these surfaces showed a loss of adhesion after aging. In contrast, the accelerated aging did not affect the adhesion in the oblique-knife pressure-bar cut surfaces. As a result, the pull-off adhesion measured after aging was similar for the three surfacing methods. Even so, the accelerated aging caused more coating deterioration in cross-grain helical-planed samples than in sanded samples. Both fixed-knife pressure-bar cutting and cross-grain helical planing could reduce the need for sanding. Thus, a single-stage 100-grit sanding should be enough to eliminate tom and fuzzy grain after fixed-knife pressure-bar cutting and to reduce coating deterioration caused by aging after cross-grain helical planing.

**********

The understanding of adhesion mechanisms on wood surfaces is essential for extending the service life of transparent film-forming coatings. Surface wettability and roughness analyses provide important information on surface quality for adhesion of coating films. Good wetting is fundamental to good adhesion, since it provides better mechanical interlocking, molecular-level interactions, and secondary force interactions between the coating film and the wood surface. For any type of coating, good wetting might contribute to good film performance (Wulf et al. 1997). If a coating cures prior to complete wetting, a weak boundary layer of air bubbles may form in the interface (Lewis and Forrestal 1969). An increase in surface roughness improves wettability by accelerating liquid spreading by capillarity (Garrett 1964, Lewis and Forrestal 1969). In addition, surface roughness provides a greater actual surface available for adhesion mechanisms if good wetting is achieved. The importance of wood surface wettability and roughness on adhesion of a polyurethane varnish has been verified in a previous study (de Moura and Hernandez 2005).

Sanding is the most common surfacing method preceding wood coating. Sanding produces defect-free uniform surfaces (Richter et al. 1995). Sanded wood is characterized by a layer of crushed cells at the surface and subsurface, lumens clogged by fine dust, scratches, and packets of microfibrils torn out from cell walls. Crushing and clogging of cells hinder penetration (de Meijer et al. 1998), while fibrillation and scratches accelerate spreading of liquid coatings on sanded surfaces. The benefits of fibrillation for mechanical adhesion of coating films have already been demonstrated for sanded wood surfaces (de Moura and Hernandez 2005).

Sanding is, however, one of the most skill-based, time-consuming, and expensive operations in the wood industry (Taylor et al. 1999). In attempts to reduce the need for sanding, the helical-knife peripheral planing and the fixed-oblique knife pressure-bar cutting have been proposed (Stewart and Lehmann 1974, Stewart 1989).

In true helical-knife planing, the knives are mounted onto the periphery of a cutterhead at an angle to the axis of rotation and form a continuous oblique cutting edge (Stewart 1971). Torn grain, raised grain, and chipmarks are reduced in helical planing, due to a gradual cutting action (Jones 1994). Helical planing performed across the grain appears to have a good potential to reduce dependence on sanding to improve surface adhesion properties and enhance performance of coatings. This surfacing method provides surfaces with improved wetting properties, good fibrillation, and virtually no cell crushing (de Moura and Hernandez 2006a).

In oblique cutting, the knife cutting edge forms an angle with a line perpendicular to the feed direction. This angle induces changes in tool geometry, cutting forces, and the quality of machined surfaces (Ozaki and Fukui 1985; Jin and Cai 1996, 1997). Stewart (1989) proposed an oblique-cutting system including a pressure bar, similar to that in rotary veneer cutting. The pressure bar modifies the field of stresses in the cutting zone, preventing the propagation of cracks and reducing the occurrence of torn grain. This surfacing method produces surfaces virtually free of cell crushing (Stewart 1989).

This work evaluates the effect of three surfacing methods on the surface quality of sugar maple wood for varnish coating. Surface quality is assessed by wetting analyses, topography measurements, and cell damage evaluation. The adhesion and performance of coating films are assessed by accelerated aging and pull-off tests. The feasibility of reducing the need for sanding is considered. The relationships among surface roughness, wettability, coating adhesion, and performance are also discussed.

Materials and methods

Testing materials

Sugar maple (Acer saccharum Marsh.), a diffuse-porous hardwood common in indoor applications, was selected for this study. Forty air-dried flat-sawn 2400-mm (L) boards were stored in a conditioning room at 20[degrees]C and 40 percent relative humidity (RH), until they reached 8 percent equilibrium moisture content (EMC). After conditioning, each board was crosscut into three matched 750-mm (L) sections. These sections were machined at 50 mm (T) width and 19.8 mm (R) thickness. Each section underwent a surfacing treatment and was resectioned to prepare samples for roughness (50 by 160 mm), wetting (50 by 20 mm), varnishing, accelerated aging, and pull-off adhesion tests (50 by 550 mm). The average and standard deviation (SD) of basic density of the boards were 598 and 28 kg/[m.sup.3], respectively.

Machining treatments

Three surfacing methods were tested: sanding, cross-grain helical-knife peripheral planing, and fixed-oblique knife pressure-bar cutting. These surfacing methods were studied preliminarily to select the best conditions for varnishing purposes, and the two best conditions of each surfacing method were tested. For each condition, 20 samples were machined. The sanding treatments were performed according to de Moura and Hernandez (2006b) with aluminum-oxide sanding belts and a 14 m/min feed speed. Two sanding programs were tested: 100-grit and 100-120-150-grit. The cross-grain helical-knife planing was carried out according to de Moura and Hernandez (2006a) at a 0.5 mm cutting depth and at two (8.5 m/min and 10.0 m/min) feed speeds. These feed speeds produced 17 and 14 knife marks per 25.4-mm length (or 1.53-mm and 1.80-mm wavelengths), respectively. The fixed-oblique knife pressure-bar cutting was performed according to de Moura and Hernandez (2006c) with a 25[degrees] nominal rake angle, 60[degrees] nominal knife angle, and a 0.25-mm cutting depth. Two oblique angles (30[degrees] and 50[degrees]) were tested.

Surface quality evaluation

Surface quality was assessed by wetting analyses, topography measurements, and cell damage evaluation. Surface wettability tests were performed as in de Moura and Hernandez (2006a). The initial contact angles ([[theta].sub.i]) of water and formamide, recorded immediately after droplet deposition, helped estimate the wood surface energy by the harmonic mean approach (Wu 1971). In order to quantify water spreading and penetration, the k-value proposed by Shi and Gardner (2001) was calculated. The time taken to complete surface wetting by water was also recorded. Roughness measurements were carried out as in de Moura and Hernandez (2006b). The roughness average ([R.sub.a]) and skewness coefficient ([R.sub.sk]) were determined according to ISO 4287-1 (ISO 1984). The surface profile was assessed parallel and perpendicular to the grain. Cell damage evaluation was performed by scanning electron microscopy (SEM) examination after coating, as described by de Moura and Hernandez (2005).

Coating and film evaluation

Samples were roll-coated with high solids ultraviolet (UV)-curable polyurethane within 24 hours after machining treatments. Freshly machined surfaces had been kept face against face and enveloped in black plastic bags to reduce exposure to air and light. The coating procedure was the same as in de Moura and Hernandez (2005). The average film thickness was 59 [micro]m. The coated samples were stored in a conditioning room at 20[degrees]C and 40 percent RH for 1 week prior to the accelerated aging and adhesion tests. Accelerated aging was similar to that by de Moura and Hernandez (2005). The end edges of samples were sealed with paraffin to prevent moisture penetration during aging. A thin wire mesh screen was installed in front of the water spray to avoid direct contact of water with coated samples. During dry periods, the chamber conditions were about 60[degrees]C and 45 percent RH. The conditions during humid periods were about 47[degrees]C and 93 percent RH. After aging, the samples were conditioned at 20[degrees]C and 40 percent RH for 1 month to reach their initial 8 percent EMC. All aged samples were ranked by the degree of deterioration of their coatings: grades varied from 1 (lowest coating deterioration) to 30 (highest coating deterioration). Aging resistance was expressed in terms of the global ranking. The adhesion of films was evaluated before and after aging by pull-off tests according to ASTM D 4541 (ASTM 1995). Small 20-mm-diameter dollies were glued onto the film surface. A cylindrical actuator connected to a 5-kN load cell was placed over the dolly head. A universal testing machine performed pulling at a 7 mm/min speed until separation of the dolly. The maximal normal pull strength at rupture was recorded.

Results and discussion

Surface topography

The values of [R.sub.a] and [R.sub.sk] measured along and across the grain for three surfacing methods, each of them performed at two machining conditions, are presented in Table 1. [R.sub.a] was significantly higher across than along the grain for all surfacing methods. The highest roughness average measured along the grain ([R.sub.a[parallel]]) was provided by cross-grain helical planing (4.1 jim), followed by sanding (3.0 [micro]m), and fixed-oblique knife pressure-bar cutting (2.1 [micro]m). Sanded surfaces presented the highest Ra measured across the grain ([R.sub.a[perpendicular]], 5.6 [micro]m), followed by the cross-grain helical planing (4.6 [micro]m), and fixed-oblique knife pressure-bar cutting (2.3 [micro]m, selected conditions pooled).

During wood sanding, the abrasive grains produce scratches and superficial cell fibrillation, which results in a microfuzzy texture (de Moura and Hernandez 2006b). In cross-grain helical planing, superficial cells are often pulled out and raised, leaving small longitudinal grooves and also a fuzzy texture (de Moura and Hernandez 2006a), In contrast, surfaces obtained by fixed-oblique knife pressure-bar cutting present plateau-like regions, with virtually no sign of fibrillation, which reduces roughness (de Moura and Hernandez 2006c).

As expected, the addition of higher grits significantly reduced roughness in sanded surfaces (Table 1). [R.sub.a[parallel]] and [R.sub.a[perpendicular to]] were reduced by 42 percent and 40 percent, respectively, when the sanding program shifted from 100-grit to 100-120-150-grit. In cross-grain helical planing, the increase in feed speed tended to increase roughness; however, this latter increase was only significant for [R.sub.a[parallel]] (Table 1). The fixed-oblique knife pressure-bar cutting showed no differences in roughness between the two oblique angles (Table 1).

As observed in a previous study (de Moura and Hernandez 2006b), sanded surfaces presented positive [R.sub.sk] in both directions of measurement (Table 1). These positive [R.sub.sk] values indicate that these surfaces presented more material near the base of the roughness profile, which resulted in a predominance of peaks. These peaks corresponded to packets of microfibrils tom out and raised from cell walls by the abrasive action. In contrast, the fixed-oblique knife pressure-bar cutting produced surfaces with higher concentrations of material near the top of the roughness profile, as indicated by the negative [R.sub.sk] values in both directions of measurement (Table 1). These values suggest a predominance of valleys, which corresponded to open lumens, ruptured rays, and longitudinal voids left by pulled-out groups of cells (de Moura and Hernandez 2006c). Cross-grain helical planing, in turn, presented negative [R.sub.sk[parallel]], similar to those observed in oblique-knife pressure-bar cutting, and positive [R.sub.sk[perpendicular to]], similar to those seen in sanded surfaces (Table 1).

In general, samples presented a good surface appearance before coating. Surfaces produced by sanding and cross-grain helical planing were free of visible machining defects, independent of grain orientation. The fixed-oblique knife pressure-bar cutting induced slight tom and fuzzy grain in some samples with severe grain deviations, however. These results confirm those presented in earlier studies (de Moura and Hernandez 2006a, 2006b, 2006c). A single-stage 100-grit sanding (0.3-mm removal depth) should be enough to eliminate tom and fuzzy grain after fixed-knife pressure-bar cutting. For these cutting conditions, the machining defects are rarely deeper than 0.25 mm (de Moura and Hernandez 2006c).

Surface wettability

The results of wetting tests for three surfacing methods, each of them performed at two selected conditions, are shown in Table 2.

The highest initial contact angle ([[theta].sub.i]) and lowest surface energy were observed on sanded surfaces (Table 2). In spite of this, sanded surfaces presented the fastest wetting, as indicated by the lowest wetting time (37 s) and highest k-value (0.899, selected conditions pooled). Thus, both [[theta].sub.i] and surface energy, which are affected by surface chemistry, were not correlated with wetting time and k-value (Table 3). This suggests that the capillary effect provided by roughness prevailed over the effect of chemical aspects on wetting. Accordingly, the wetting time and k-value were significantly correlated with roughness parameters ([R.sub.a[perpendicular to]] and [R.sub.sk[perpendicular to]], Table 3). This latter result agrees with those reported in previous studies (de Moura and Hernandez 2005, 2006a, 2006b, 2006c).

Sanded surfaces offered the best conditions for liquid spreading due to the scratches left by the abrasive grains, which accelerated conduction by capillarity. Such behavior has been noticed in earlier studies (Garrett 1964; de Moura and Hernandez 2005, 2006b). This capillary effect was low in cross-grain helical-planed surfaces and even lower in oblique-knife pressure-bar cut surfaces. The highest wetting time (211 s) and lowest k-value (0.592, selected conditions pooled) confirm that the oblique-knife pressure-bar cutting induced the lowest capillary effect. This lack of capillaries might be less important with oil finishes, which could be absorbed directly by the cell walls. Such a characteristic is, however, unfavorable to adhesion of film-forming coatings.

Interfaces and coating films

As expected, cell damage at the surface and subsurface was more pronounced in sanded samples. Fibers and vessels were often crushed near these surfaces (Fig. 1). This damage was not distributed uniformly and was not visible in some zones. in sanded wood, the superficial cell damage was attributed to the negative rake angles of abrasive grains, which induce high normal forces (Stewart and Crist 1982). The single-stage 100-grit sanding program provoked deeper cell crushing than did the 100-120-150-grit sanding program (Fig. 1). The average maximum damage depth was 69 [micro]m and 49 [micro]m for the single-stage and three-stages sanding programs, respectively. These values are similar to those reported in a previous study (de Moura and Hernandez 2006b).

[FIGURE 1 OMITTED]

In contrast, the cross-grain helical planing provided surfaces with slight lateral crushing and distortion, while the fixed-oblique knife pressure-bar cutting produced slight crushing only in the most outward cell layer (Figs. 2 and 3). When helical planing tangential surfaces across the grain, the knife attacks rays transversally with its edge virtually parallel to the height of rays, then facilitates ruptures by bending of rays below the cutting plane (Fig. 2). This also was noted previously by de Moura and Hernandez (2006a).

[FIGURES 2-3 OMITTED]

The cross-grain helical planing provides more paths for coating penetration into cells. For samples surfaced by this method, the polyurethane sealer could penetrate virtually all types of cells, including thick-walled fibers (Fig. 2). In planing across the grain, chip formation often takes place by pulling out cells or groups of cells from middle lamella (de Moura and Hernandez 2006a). Due to its particular mechanism of chip formation, this surfacing method further opens up and exposes cell lumens that might be paths for penetration, Moreover, in the absence of cell crushing, the sealer has more paths to penetrate and establish a mechanical anchorage at the surface itself.

In surfaces obtained by fixed-oblique knife pressure-bar cutting, the sealer often penetrated through vessels, while fibers filled with sealer were scarce (Fig. 3). In sanded surfaces, the presence of a crushed-cell layer and the lack of totally opened vessels hindered coating penetration into the wood capillaries. Thus, the penetration in sanded surfaces occurred only through vessels, but it was limited (Fig. 1). The importance of vessels as paths for coating penetration in wood has been reported previously (de Meijer et al. 1998). In contrast, the penetration via rays was not seen in this study. In maple woods, the ray cells are short and present thick pit membranes, which contribute to greater resistance to penetration (Wheeler 1982).

The coating film was significantly thicker in sanded than in cross-grain helical-planed surfaces. The average film thickness was 61 [micro]m, 58 [micro]m, and 56 [micro]m on sanded, oblique-knife cut, and cross-grain helical-planed surfaces, respectively. This suggests that the coating penetration had a significant effect on film thickness. Thus, surfaces offering more paths for penetration (Fig. 2) tended to have thinner films.

Machining defects were visible after coating in 20 percent of samples that had been surfaced by the fixed-oblique knife pressure-bar system. The film was not sufficiently thick to cover up the machining defects in these surfaces. We postulate that the machining defects observed could be reduced by decreasing the cutting depth even more. Even so, thinner cutting might strongly reduce roughness, thereby impairing the surface aptitude for adhesion of film-forming coatings.

Adhesion and aging resistance

The results of adhesion and aging tests for three surfacing methods, each of them performed at two selected conditions, are summarized in Table 4. The highest pull-off adhesion measured before aging was found on cross-grain helical-planed and sanded surfaces (5.7 MPa), while the fixed-oblique knife pressure-bar cutting provided a significantly lower pull-off adhesion before aging (5.0 MPa, selected conditions pooled).

No references are available nor direct evidence in this work indicate the presence of chemical linkages between wood and coatings. Thus, mechanical bonding was considered the major adhesion factor. The relatively smooth surfaces produced by the fixed-oblique knife pressure-bar cutting did not offer optimal conditions for mechanical anchorage. In these surfaces, mechanical bonding was associated with the coating penetration into opened cells, mainly vessels. This amount of penetration was sufficient to obtain a stronger bond than that reported for straight-knife peripheral-planed surfaces (de Moura and Hernandez 2005).

The strongest before-aging adhesion in sanding and cross-grain helical planing is attributed to the microfibrils and fibers torn-out from the surface (de Moura and Hernandez 2005, 2006a, 2006b). These elements diffuse into the liquid coating, adequately maintaining the film bond to the wood surface after curing. According to Lewis and Forrestal (1969), diffusion may be an important adhesion mechanism on rough and porous surfaces. When diffusion occurs, adhesion takes place in a three-dimensional interface and is directly proportional to the length and number of molecules crossing this interface. In this case, a composite of polymer and wood is observed as a transition between wood and coating (Backman and Lindberg 2002). Furthermore, the torn-out elements induce an increase in roughness and yield a greater actual surface available for other adhesion mechanisms. These facts are confirmed by the significant positive correlations detected between the surface roughness parameters and the before-aging pull-off adhesion of films (Table 3).

The pull-off adhesion measured after aging was statistically similar for all surfacing methods (Table 4). The adhesion of surfaces obtained by oblique-knife pressure-bar cutting was not significantly affected by the aging treatment. Even so, the accelerated aging caused a 10-percent loss of adhesion in sanded and helical-planed samples (four conditions pooled). This loss of adhesion was probably associated with weak boundary layers formed at the surface during machining. As shown in previous works (Stewart 1989, de Moura and Hernandez 2006c), the fixed-oblique knife pressure-bar cut produced surfaces with very low distortion of superficial wood tissues. In sanded surfaces, however, the layer of crushed cells at the surface and subsurface was certainly affected by aging cycles. In cross-grain helical planing, the high (30[degrees]) rake angle might have induced negative normal forces (upward) during chip formation, mainly when planing at a 1.80-mm wavelength. These negative normal forces might have distorted tissues near the cutting plane, weakening their structure and decreasing their aging resistance (Fig. 2). Thus, it appears that normal forces, either downward or upward, contributed to the formation of weak boundary layers during surfacing. The presence of weakness zones was manifest only after accelerated aging.

During the accelerated aging, coating failures tended to occur at the end and lateral edges of the samples. Failures also initiated near wood cracks where water infiltrated the interface and gradually removed the coating film from the surface (Lewis and Forrestal 1969). The aging resistance of films, as expressed by the global ranking, was not correlated with the pull-off adhesion (Table 3). Severely cracked samples occasionally showed relatively good adhesion values. The lowest average global ranking was found in sanded surfaces (5.8), followed by oblique-knife pressure-bar cut (8.8), and cross-grain helical-planed surfaces (9.9, selected conditions pooled). The aging resistance of coating films was, hence, significantly better in sanded than in cross-grain helical-planed samples. Therefore, a cross-grain helical-planed surface should be sanded prior to coating.

The surface parameters studied were correlated significantly with the pull-off adhesion measured before aging, excepting [[theta].sub.i] and surface energy. The roughness parameters showed the highest correlations with this adhesion (Table 3). In contrast, considering a 5 percent significance level, the after-aging pull-off adhesion was only correlated with [R.sub.a[perpendicular to]] and wetting time (Table 3). These results confirm the possibility of assessing the adhesion ability of a wood surface by means of roughness and wetting analyses, as reported by de Moura and Hernandez (2005).

Conclusions and recommendations

The three surfacing methods yielded marked differences in surface topography and wetting properties. Sanding and cross-grain helical planing produced surfaces with no visible machining defects. The fixed-oblique knife pressure-bar cutting occasionally caused slight torn or fuzzy grain; but the use of lower cutting depths could eliminate these machining defects. Thus, further research is needed to optimize cutting depth to eliminate torn and fuzzy grain with minimum loss of adhesion aptitude.

The highest roughness and best conditions for wetting were obtained with sanding. The sanded and cross-grain helical-planed surfaces presented the strongest adhesions before the aging treatment. These improved adhesions were attributed mainly to surface fibrillation, which enhanced mechanical bonding. Accelerated aging, however, caused a loss of adhesion in sanded and cross-grain helical-planed samples. The oblique-knife cut samples, in turn, presented no loss of adhesion after aging. As a result, the pull-off adhesion measured after aging was similar for all surfacing methods. Aging also caused more coating deterioration in cross-grain helical-planed samples than in sanded samples.

Both fixed-knife pressure-bar cutting and cross-grain helical planing could reduce the need for sanding, These two alternative surfacing methods could be followed by a single-stage 100-grit sanding, instead of the three stages common in the flooring industries. This slight sanding should be enough to eliminate torn and fuzzy grain after fixed-knife pressure-bar cutting, and to reduce coating deterioration caused by aging after cross-grain helical planing.

Further studies should consider optimizations of helical planers and cutterheads. For instance, rake angle and wear resistance of helical knives should be further optimized. Moreover, the economical aspects of either the oblique-knife pressure-bar cutting or the cross-grain helical planing should be studied, since they can potentially reduce costs in wood sanding operations.

Literature cited

American Society for Testing and Materials (ASTM). 1995. Standard test method for pull-off strength of coatings using portable adhesion testers. ASTM D 4541. West Conshohocken, PA.

Backman, A.C. and K.A.H. Lindberg. 2002. Interaction between wood and polyurethane-alkyd lacquer resulting in a decrease in the glass transition temperature. J. Appl. Polym. Sci. 85:595-605.

de Meijer, M., K. Thurich, and H. Militz. 1998. Comparative study On penetration characteristics of modern wood coatings. Wood Sci. and Tech. 32(5):347-365.

de Moura, L.F. and R.E. Hernandez. 2005. Evaluation of varnish coating performance for two surfacing methods on sugar maple wood. Wood and Fiber Sci. 37(2):355-366. --and --. 2006a. Characteristics of sugar maple wood surfaces produced by helical planing. Wood and Fiber Sci. 38(1): 166-178. --and --. 2006b. Effects of abrasive mineral, grit size and feed speed on the quality of sanded surfaces of sugar maple wood. Wood Sci. and Tech. 40(6):517-530. --and --. 2006c. Characteristics of sugar maple wood surfaces machined with the fixed-oblique knife pressure-bar cutting system. Wood Sci. and Tech. doi: 10.1007/s00226-006-0074-9.

Garrett, H.E. 1964. Contact angles and their significance for adhesion. In: Proc. Conferences of the Northampton College of Advanced Tech., EC1, England. p. 19.

Inter. Organization for Standardization (ISO). 1984. Surface roughness-Terminology--Part 1: Surface and its parameters. ISO 4287-1. Geneva, Switzerland.

Jin, W. and L. Cai. 1996. Study and analysis on cutting forces of oblique cutting of wood. Holz als Roh- und Werkstoff 54:283-286. --and --. 1997. Study on the normal component force in oblique cutting of wood. Holz als Roh- und Werkstoff55:118-120.

Jones, C.W. 1994. Cutterheads and Knives for Machining Wood. C.W. Jones, Seattle, WA, 138 pp.

Lewis, A.F. and L.J. Forrestal. 1969. Adhesion of coatings. In: Treatise on Coatings, vol II. Characterization of Coatings: Physical Techniques. Marcel Dekker, New York, pp. 57-98.

Ozaki, S. and H. Fukui. 1985. Studies on the oblique cutting of wood. III. Roughness of machined surface in 90[degrees]-0[degrees] and 0[degrees]-90[degrees] cutting situations. Mokuzai Gakkaishi 31 (5):354-360.

Richter, K., W.C. Feist, and M.T. Knaebe. 1995. The effect of surface roughness on the performance of finishes. Part 1. Roughness characterization and stain performance. Forest Prod. J. 45(7/8):91-97.

Shi, S.Q. and D.J. Gardner. 2001. Dynamic adhesive wettability of wood. Wood and Fiber Sci. 33(1):58-68.

Stewart, H.A. 1971. Chips produced with a helical cutter. Forest Prod. J. 21(5):44-45. --. 1989. Fixed-knife pressure-bar planing method reduces or eliminates subsurface damage. Forest Prod. J. 39(7/8):66-70. --and W.F. Lehmann. 1974. Cross-grain cutting with segmented helical cutters produces good surfaces and flakes. Forest Prod. J. 24(9):104-106. -- and J.B. Crist. 1982. SEM examination of subsurface damage of wood after abrasive and knife planing. Wood Sci. 14(3):106-109.

Taylor, J.B., A.L. Carrano, and R.L. Lemaster. 1999. Quantification of process parameters in a wood sanding operation. Forest Prod. J. 49(5): 41-46.

Wheeler, E.A. 1982. Ultrastructural characteristics of red maple (Acer rubrum L.) wood. Wood and Fiber 14(1):43-53.

Wu, S. 1971. Calculation of interfacial tension in polymer systems. J. Polym. Sci. C34:19-30.

Wulf, M., P. Netuschil, G. Hora, P. Schmich, and H.K. Cammenga. 1997. Investigation of the wetting characteristics of medium density fibreboards (MDF) by means of contact angle measurements. Holz als Roh- und Werkstoff 55:331-335.

Luiz Fernando de Moura Roger E. Hernandez *

The authors are, respectively, former Ph.D. Candidate and Professor, Centre de recherche sur le bois (CRB), Departement des sciences du bois et de la foret, Universite Laval, Quebec, Canada (luiz-fernando.de-moura.1@ulaval.ca, roger.hernandez@sbf.ulaval.ca). The authors thank Forintek Canada Corp. and Les Industries PG for technical support during testing. This research was supported by the National Scientific and Technological Development Council of Brazil and by the Canada Economic Development for Quebec Regions. This paper was received for publication in January 2006. Article No. 10148.

* Forest Products Society Member.

[c] Forest Products Society 2006. Forest Prod. J. 56(11/12): 130-136.
Table 1.--Roughness averages ([R.sub.a]) and skewness coefficients
([R.sub.sk]), measured along ([parallel]) and across
([perpendicular to]) the grain, obtained for three surfacing methods
and two selected conditions applied to sugar maple wood.

Surfacing method Surfacing method [R.sub.a
 [parallel]]

 ([micro]m)

Sanding (a) 100-grit 3.8 (b) (0.3)
 (c) BC (d)

 100-120-150-grit 2.2 (0.1) A

Cross-grain helical- 1.53-mm wavelength 3.7 (0.2) Bb
planing (e)
 1.80-mm wavelength 4.6 (0.3) C

Oblique-knife pressure-bar 30[degrees]-oblique angle 2.1 (0.1) A
cutting (f)
 50[degrees]-oblique angle 2.0 (0. l) A

Surfacing method [R.sub.a [R.sub.sk
 [perpendicular to]] [parallel]]

 ([micro]m) ([micro]m)

Sanding (a) 6.9 (0.2) C 0.1 (0.1) BC

 4.2 (0.1) B 0.5 (0.1) C

Cross-grain helical- 4.4 (0.3) B -0.4 (0.2) AB
planing (e)
 4.8 (0.2) B -0.6 (0.2) A

Oblique-knife pressure-bar 2.3 (0.1) A -0.6 (0.1) A
cutting (f)
 2.3 (0.1) A -0.4 (0.2) AB

Surfacing method [R.sub.sk
 [perpendicular to]]

 ([micro]m)

Sanding (a) 0.4 (0.1) C

 0.2 (0.1) BC

Cross-grain helical- 0.2 (0.2) BC
planing (e)
 0.2 (0.1) BC

Oblique-knife pressure-bar -0.8 (0.2) A
cutting (f)
 -0.4 (0.2) AB

(a) Removal depths were fixed to 0.3 mm, 0.2 mm, and 0.1 mm for
100-grit, 120-grit, and 150-grit stages, respectively.

(b) Means of 40 replicates.

(c) Standard error of the mean in parentheses.

(d) Means within a column followed by the same letter are not
significantly different at the 5% probability level.

(e) 30[degrees]-rake angle, 14[degrees]-helix angle, and 0.5-mm
cutting depth were used.

(f) 25[degrees]-rake angle, 0.25-mm cutting depth, and 65[degrees]
single-face pressure-bar were used.

Table 2.--Wetting parameters obtained for three surfacing
methods and two selected conditions applied to sugar maple wood.

 Contact angle
Surfacing method Surfacing method ([??]

Sanding (a) 100-grit 71.56 (b)
 (1.4) (c) C (d)

 100-120-150-grit 66.5 (0.7) BC

Cross-grain helical- 1.53-mm wavelength 56.5 (1.9) A
planing (e) 1.80-mm wavelength 56.2 (1.4) A

Oblique-knife pressure- 30[degrees]-oblique angle 60.6 (1.6) AB
bar cutting (f) 50[degrees]-oblique angle 63.0 (2.0) B

Surfacing method Wetting time k-value
 (s)

Sanding (a) 39 (6) AB 0.758 (0.085) ABC
 34 (3) A 1.039 (0.084) C

Cross-grain helical- 121 (25) ABC 0.907 (0.090) BC
planing (e) 129 (19) BCD 0.657 (0.055) AB

Oblique-knife pressure- 221 (33) D 0.512 (0.043) A
bar cutting (f) 201 (32) CD 0.671 (0.051) AB

Surfacing method Surface energy
 (mN/m)

Sanding (a) 36.4 (0.9) A
 39.4 (0.5) AB

Cross-grain helical- 46.2 (1.3) C
planing (e) 46.3 (0.9) C

Oblique-knife pressure- 43.4 (1.1) BC
bar cutting (f) 41.8 (1.3) B

(a) Removal depths were fixed to 0.3 mm, 0.2 mm, and 0.1 mm
for 100-grit, 120-grit, and 150-grit stages, respectively.

(b) Means of 20 replicates

(c) Standard error of the mean in parentheses.

(d) Means within a column followed by the same letter are
not significantly different at the 5% probability level.

(e) 30[degrees]-rake angle, 14[degrees]-helix angle, and
0.5-mm cutting depth were used.

(f) 25[degrees]-rake angle, 0.25-mm cutting depth, and
65[degrees]-single-face pressure-bar were used.

Table 3.--Statistical correlations obtained among surface
roughness parameters, wetting properties, coating film
adhesion, and performance for sugar maple wood coated
with high solids polyurethane after three surfacing methods
at two selected conditions.

 Wetting Surface
Parameter time k-value energy
 (s) (mN/m)

[R.sub.a [perpendeicular to] -0.457 (d) 0.242 n/a
([micro]m) 0.001 (e) 0.008

[R.sub.sk [perpendicular to] -0.493 0.266 n/a
([micro]m) 0.001 0.003

[[theta].sub.i] ([??]) -0.055 0.045 n/a
 0.555 0.631

Wetting time (s) -0.505 0.049
 0.001 0.599

k-value -0.045
 0.624

Surface energy (mN/m)

Pull-off before (a) (MPa)

Pull-off after (b) (MPa)

 Pull-off Pull-off Global
Parameter before (a) after (b) ranking
 (MPa) (MPa) (1 to 30) (c)

[R.sub.a [perpendeicular to] 0.400 0.269 -0.319
([micro]m) 0.002 0.038 0.014

[R.sub.sk [perpendicular to] 0.460 0.239 0.028
([micro]m) 0.001 0.066 0.835

[[theta].sub.i] ([??]) 0.209 0.223 -0.276
 0.109 0.087 0.035

Wetting time (s) -0.334 -0.282 0.022
 0.013 0.029 0.866

k-value 0.302 0.186 -0.043
 0.028 0.154 0.751

Surface energy (mN/m) -0.235 -0.219 0.278
 0.070 0.092 0.033

Pull-off before (a) (MPa) 0.355 -0.167
 0.007 0.223

Pull-off after (b) (MPa) 0.074
 0.575

(a) Pull-off adhesion strength measured before the accelerated aging.

(b) Pull-off adhesion strength measured after the accelerated aging.

(c) Degree of coating deterioration after aging (higher ranking values
indicate higher deterioration).

(d) Pearson correlation coefficient (r), for correlations between two
continous variables, and Spearman correlation
coefficient (r), for correlations between global ranking (ordinal
variable) and any other variable.

(e) Prob. > | r | under [H.sub.o: [rho] = 0, for n = 120 (roughness and
wetting properties) and n = 60 (pull-off adhesions and
global ranking).

Table 4.--Pull-off adhesion strength and global ranking for
a polyurethane coating applied on sugar maple wood machined by three
surfacing methods and two selected conditions.

 Pull-off
 adhesion
Surfacing method Surfacing method before aging

 (MPa)

Sanding (b) 100-grit 5.9 (c) (0.1)
 (d) B (e)

 100-120-150-grit 5.7 (0.2) AB

Cross-grain helical 1.53-mm wavelength 5.5 (0.2) AB
-planing (f)
 1.80-mm wavelength 5.8 (0.2) B

Oblique-knife pressure-bar 30[degrees]-oblique angle 4.8 (0.1) A
cutting (g)
 50[degrees]-oblique angle 5.1 (0.2) AB

Surfacing method Pull-off
 adhesion Global
 after aging ranking

 (MPa) (1 to 30)(a)

Sanding (b) 5.4 (0.3) A 4 (1) A

 5.1 (0.2) A 8 (2) AB

Cross-grain helical 5.2 (0.2) A 11 (1) B
-planing (f)
 5.0 (0.2) A 9 (1) B

Oblique-knife pressure-bar 4.8 (0.1) A 8 (1) AB
cutting (g)
 4.9 (0.3) A 9 (2) B

(a) Degree of coating deterioration after aging
(higher ranking values indicate higher deterioration).

(b) Removal depths were fixed to 0.3 mm, 0.2 mm and 0.1 mm for
100-grit, 120-grit and 150-grit stages, respectively.

(c) Means of 20 replicates (ten replicates for global ranking).

(d) Standard error of the mean in parentheses.

(e) Means within a column followed by the same letter are not
significantly different at the 5% probability level.

(f) 30 [degrees]-rake angle, 14[degrees]-helix angle and 0.5-mm
cutting depth were used.

(g) 25 [degrees]-rake angle, 0.25-mm cutting depth and
65[degrees]-single-face pressure-bar were used.
COPYRIGHT 2006 Forest Products Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:de Moura, Luiz Fernando; Hernandez, Roger E.
Publication:Forest Products Journal
Date:Nov 1, 2006
Words:5819
Previous Article:Degradation of a wood-plastic composite exposed under tropical conditions.
Next Article:Effect of openings on the racking strength of structural log walls.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters