Abstract
A test method was developed to evaluate the surface characteristics of veneers produced using the peeling process. Rotary cut Scots pine (Pinus sylvestris L.) veneers, which were produced from logs soaked at 65 °C for 42 h prior to peeling and dried at 125 °C, were used as the test specimens. The pine veneers were produced under industrial conditions. The density variation of the round wood was determined using X-ray computed tomography, the thickness variation of the pine veneers was measured using an optical microscope, and the structure of the veneers was examined using a scanning electron microscope. The pine wood was characterised by the density variation in the transverse section. At the same moisture level, broad-ring pith-adjacent juvenile sapwood had a lower density compared with that of perimeter-adjacent mature wood. The clear lack of density homogeneity was seen in the width of the subsequent annual growths. The roughness of the pine veneers demonstrated significant variation, depending on the tool angle (17° and 21°), wood area (sapwood and heartwood), and veneer side (tight and loose). The veneer thickness had a significant impact on the roughness values measured perpendicular to the grain.
Download PDF
Full Article
Surface Characteristics of Scots Pine Veneers Produced with a Peeling Process in Industrial Conditions
Agnieszka Laskowska,* Paweł Kozakiewicz, Marcin Zbieć, Patrycja Zatoń, Sylwia Oleńska, and Piotr Beer
A test method was developed to evaluate the surface characteristics of veneers produced using the peeling process. Rotary cut Scots pine (Pinus sylvestris L.) veneers, which were produced from logs soaked at 65 °C for 42 h prior to peeling and dried at 125 °C, were used as the test specimens. The pine veneers were produced under industrial conditions. The density variation of the round wood was determined using X-ray computed tomography, the thickness variation of the pine veneers was measured using an optical microscope, and the structure of the veneers was examined using a scanning electron microscope. The pine wood was characterised by the density variation in the transverse section. At the same moisture level, broad-ring pith-adjacent juvenile sapwood had a lower density compared with that of perimeter-adjacent mature wood. The clear lack of density homogeneity was seen in the width of the subsequent annual growths. The roughness of the pine veneers demonstrated significant variation, depending on the tool angle (17° and 21°), wood area (sapwood and heartwood), and veneer side (tight and loose). The veneer thickness had a significant impact on the roughness values measured perpendicular to the grain.
Keywords: Computed tomography; Peeling; Pine wood; Scanning electron microscopy; Surface characteristics; Veneer
Contact information: Faculty of Wood Technology, Warsaw University of Life Sciences – SGGW, 159 Nowoursynowska, 02-776 Warsaw, Poland; *Corresponding author: agnieszka_laskowska@sggw.pl
INTRODUCTION
The circumferential peeling of wood is one of the basic methods for obtaining veneers (Denaud et al. 2012). This method is also a type of chipless woodworking. Veneers obtained as a result of circumferential peeling are commonly used to manufacture plywood and laminated veneer lumber (LVL). These veneers may also be used as a surface-enhancing material in applications such as wood-based panels, i.e., particleboards and medium density fibreboards. Mainly softwood species (i.e., Scots pine) and, to a lesser degree, diffuse-porous hardwood species (i.e., beech, birch, and alder) are subjected to circumferential peeling in Central and Eastern Europe. The most serious difficulties arise when applying the peeling process to softwood. Softwood species are characterised by the greatest density variation in the width of single annual growths (Kollmann and Côté 1984). In Scots pine wood, the density range of earlywood in the absolutely-dry state ranges from 340 kg/m3 to 360 kg/m3, and the density of latewood ranges from 810 kg/m3 to 900 kg/m3; therefore, latewood is approximately three times as dense as earlywood. Unfortunately, the lack of radial homogeneity in pine wood has an adverse impact on circumferential peeling processes. The current resolution and speed of X-ray computed tomography (CT) scanners enable the complex internal structure of wood to be analysed in a non-destructive manner, with a full representation of the annual growth system and variation (lack of homogeneity) in the density (Wei et al. 2009). On a laboratory scale, it is necessary to convert the obtained CT images from the Hounsfield unit scale to the actual density (kg/m3).
The quality of the veneer is determined by many parameters, both material- and technology-related (Baldwin 1995; Bekhta et al. 2014; Li et al. 2018). Among the material-related factors, the species and area of the block from which the veneer is obtained (sapwood and heartwood) has a remarkable impact on the quality of the veneer. The veneer thickness is also important (Dundar et al. 2008a). It is generally accepted that better quality veneers are obtained from the perimeter-adjacent area, called sapwood. This is because sapwood has a lower density than heartwood. Additionally, sapwood demonstrates a greater susceptibility to softening when subjected to a hydrothermal treatment because of its structure and location next to the perimeter. This area has less knots, which determine the quality class of the veneer. The methods for the hydrothermal treatment (in water or vapour) of wood designated for circumferential peeling is well known, but the aspect that changes is the equipment. The treatment processes are optimised based on the wood species, moisture content, wood density, and dimensions of the products (Baldwin 1995; Aydin et al. 2006; Dundar et al. 2008a,b; Yamamoto et al. 2015). It has been demonstrated that drying also has an impact on the properties of the material subjected to circumferential peeling (Aydin and Çolakoğlu 2002).
The properties of the veneer surface are important factors when determining the wood wettability, glue adhesion, and adhesive strength (Neese et al. 2004; Budakçi et al. 2007; Aslan et al. 2008; Coelho et al. 2008; Rohumaa et al. 2014). The roughness is commonly used to measure the quality of the wood surface (Hiziroglu 1996; Aydin and Çolakoğlu 2003; Söğütlü 2010; Söğütlü and Togay 2011; Söğütlü et al. 2016; Söğütlü 2017; Li et al. 2018). However, it should be noted that the determined roughness parameters do not allow one to directly specify the destruction degree of the anatomical structure of wood elements or the depth of cracks. These aspects must be specified with the use of scanning electron microscopy (SEM) (Rohumaa et al. 2016). The longitudinal cracks are more intense when the area from which the veneer is obtained is closer to the core log, and the orientation deflection of the growth rings from the tangential direction in the transverse section of the veneer is greater. Veneer cracks may have a positive impact during the gluing process because they improve the penetration of glue into the veneer structure, which strengthens the adhesive bonds (Baldwin 1995; Kamke and Lee 2007; Kurowska et al. 2011). Stehr and Johansson (2000) demonstrated that the sawn specimens of Scots pine (Pinus sylvestris L.) had a joint that was approximately twice as strong as that of microtomed specimens. In this case, the adhesive was reinforced by the damaged fibres at the surface. Nevertheless, the presence of cracks in thick veneers, which can be used as a structural layer in multi-layer flooring materials, is an undesired characteristic because each crack leads to a decrease in the strength properties of veneers, and consequently, incurs a decrease in the product strength.
Veneers with a thickness ranging from 0.5 mm to 3.0 mm are usually obtained with circumferential peeling. Thicker veneers can be used, for example, as a balance material under the surface layer in multi-layer materials. There is currently no data available concerning the properties of this type of veneer. In view of the above discussion, the properties of 4-mm-, 6-mm-, and 8-mm-thick veneers were determined with the use of modern measurement systems and techniques. Because of the comprehensive nature of this study, the quality of the round wood used for manufacturing the veneers was also assessed.
EXPERIMENTAL
Materials
The Scots pine (P. sylvestris L.) wood was obtained from a forest in northwest Poland, which is located in Biała (52° 49′ 45″ N; 16° 18′ 35″ E) and managed by the State Forests National Forest Holding. The wood used had a circular cross-section, and knots with a diameter of 20 mm to 70 mm were visible on the side of the trunk. The study subject was pine wood that was 50 years old. The width of annual rings of wood was typical for planted Scots pine trees (narrower near the bark and wider near the core, from 0.6 mm to 4.5 mm). The density was changed cyclically in radial direction (lower in spring wood and higher in summer wood). The obtained raw material was subjected to preparatory procedures, cutting, and drying in a Polish plant. Round wood with a diameter of 200 mm to 360 mm and a length of 700 mm was treated hydrothermally at a temperature of 65 °C (low but applied temperature in energy-saving process for medium sized softwood logs). The treatment time was 42 h, while the pH of the water before the treatment was 8. After treatment, the wood was debarked. The moisture content of the wood prior to peeling was determined using a resistive hygrometer (WRD-100, TANEL Electronics & IT General Partnership, Gliwice, Poland). Measurements were performed at a depth of 6 mm. The time between the removal of the blocks from the hot water and peeling was 420 s. The average moisture content of the logs (taking into account the sapwood and heartwood) immediately before peeling was 76.8% (± 3.0%). The moisture content of the core log (heartwood) was only 44.4% (± 5.0%). The temperature of the wood before and after peeling was measured with a Voltcraft K101 digital thermometer (Wernberg-Köblitz, Germany). The temperature of wood after hydrothermal treatment (before peeling) was about 65 °C, but the average surface temperature of veneer (immediately after peeling – about 420 s) was only 27.6 °C (± 3.3 °C) and average temperature of the surface of core log was 31.6 °C (± 3.3 °C).
Wood-peeling Process
The wood-peeling process was done on a veneer lathe under industrial conditions. For peeling, flat knives with two tool angles (17° and 21°) were used. The pressure bar bevel was 55°. The blocks were peeled into 4-mm, 6-mm, and 8-mm veneers. The blocks for peeling were mechanically positioned. The ambient temperature of the machine hall was 16 °C and the air humidity was 60%. The veneers were dried in a tunnel kiln at 125 °C. The drying time was dependent upon the thickness of the veneers and was 14 min for the 4-mm-thick veneers, 21 min for the 6-mm-thick veneers, and 28 min for the 8-mm-thick veneers. The final moisture content of the veneers was measured using a capacitive method with a WIP-22D Hygrometer (TANEL Electronics & IT General Partnership) and was 6% ± 1%.
Methods
This study used an original methodology developed for the purposes of this work and was based on the most modern equipment. The pine wood was tested using X-ray CT, which was performed using a NeuViz 16 CT scanner (Neusoft Medical Systems Co., Ltd., Shenyang, China). This detected the density variability (diameter) and any defects. The study was performed on round wood immediately after felling. Because of the technical constraints of the X-ray CT scanner, testing was done on 700-mm-long blocks.
The test veneer samples were delivered to the plant in the form of sheets with the dimensions 700 mm (longitudinal) × 1200 mm (tangential). For this study, the sheets were cut into veneers with the dimensions 100 mm × 100 mm and were separately made from the sapwood and heartwood of the pine wood.
An analysis of the surface of the pine veneers was done using a scanning electron microscope (QUANTA 200, FEI Company, Hillsboro, OR, USA). The thickness variability of the pine veneers was examined using an optical microscope (SMZ 1500, Nikon, Tokyo, Japan) with image analysing software. The setting parameters of the devices are given in Table 1.
Table 1. Measuring Devices for the Test Equipment
The roughness of the pine veneers was examined in accordance with the requirements of ISO 4287 (1997). As part of the study, the arithmetic mean deviation of the assessed profile (Ra) and maximum height of the assessed profile (Rz), as defined in ISO 4287 (1997), were measured. The surface roughness was determined using the Surftest SJ-210 Series 178-Portable Surface Roughness Tester (Mitutoyo Corporation, Kawasaki, Japan). The parameters were measured 20 times each in the parallel and perpendicular to the grain directions for each veneer variant. The roughness parameters were marked separately for the pine sapwood and heartwood. The analysis was on the tight (“right”) and loose (“left”) sides of the veneers. The test factors and their variability levels are summarised in Table 2.
Table 2. Levels of the Tested Factors in Relation to the Pine Veneer Roughness
Statistical analysis was performed using STATISTICA software (version 12, StatSoft, Inc., Tulsa, OK, USA) and was performed at a significance level (p) of 0.05.
RESULTS AND DISCUSSION
Variation in the Round Wood Density
The results of the CT scanning test were recorded in the Hounsfield unit scale, and therefore the CT scanner that was used for this study was calibrated by examining wood samples with a known moisture content (ISO 13061-1 2014). The density was determined using the stereometric method (ISO 13061-2 2014). In this way, an equation with a linear correlation was obtained, which incorporated the relationship between the Hounsfield unit scale and actual density of the wood (Fig. 1). Therefore, the obtained CT images of the pine wood could then be used to take a direct density reading (kg/m3) (Figs. 2 and 3).
Fig. 1. Relationship between the Hounsfield density point scale and actual wood density used during CT scanning
Fig. 2. Typical CT image of the transverse section of the pine wood presented using the Hounsfield unit scale
There was a clear difference in the densities of the freshly cut round wood, which to a large extent may have been caused by the differences in the moisture content of the sapwood and heartwood. The heartwood, with a density ranging from 450 kg/m3 to 550 kg/m3, had a moisture content of 35% to 40%, whereas the wide sapwood area, with a density of approximately 1100 kg/m3, had a moisture content of 100% to 110%. When the entire log had a moisture content of 6%, as was recorded in the examined veneers (Kollmann and Côté 1984; ISO 13061-2 2014), the density was approximately 410 kg/m3 for the pine heartwood and reached up to 640 kg/m3 for the sapwood. In the case of flooring layered materials, which are produced by pressing, the wood density is an important property (Kurowska et al. 2010; Gaff and Gašparík 2015). Density plays an important role during pressing and influences the compatibility of veneers with one another (Aydin and Çolakoğlu 2005; Gaff and Gašparík 2015; Gaff et al. 2016).
Fig. 3. Typical variability of the density in round pine wood
Typical pine raw materials (round wood) are characterised by a clearly lower density in the rings adjacent to the pith, whose fibres tend to pill and tear during chip woodworking. The pith-adjacent area also includes juvenile wood with shorter structural elements and a different layout of fibrils in the cell walls (Fabisiak 2005), which contributes to increased shrinkage and cracking. Chipless woodworking, done by circumferential peeling, enables the separation of undesirable pith-adjacent wood (by encasing it in a core log) from full quality mature wood, which generally corresponds to the sapwood in pine wood with a diameter ranging from 200 mm to 360 mm. This ensures that only full quality raw material makes it to the final products. It should also be noted that the presence of low-density juvenile pine wood multi-layer flooring materials in a structural layer leads to a decreased resistance to dents. Chipless woodworking enables the elimination of undesirable wood areas from further processing.
Variation in the Pine Veneer Thickness
The veneers obtained from the heartwood were characterised by a greater variation in the thickness compared with those obtained from the sapwood (Fig. 4). The relative standard deviation (RSD) is defined as the ratio of the standard to the mean and was 2% and 5% for the 4-mm-thick veneers obtained from the sapwood and heartwood, respectively. The RSD for the 6-mm-thick veneers obtained from the sapwood and heartwood was 2% and 4%, respectively. Meanwhile, the RSD for the 8-mm-thick veneers obtained from the sapwood and heartwood was 1% and 2%, respectively.
The smaller variation in the thickness of the veneers obtained from sapwood was related to the better softening of the material and, maybe consequently led to more favourable temperature and moisture content conditions while peeling. The moisture content of the surface of the block after the hydrothermal treatment was higher than the moisture content of the surface of the core log. Furthermore, the heartwood had wider growth rings than the sapwood. Dupleix et al. (2013) showed that the thickness variation in spruce veneers decreased as the heating temperature increased (by soaking the wood), which demonstrated the positive effects of heating on the veneer thickness consistency. At a temperature of 50 °C, the positive effects of heating ensure an efficient peeling process.
Fig. 4. Variation in the pine veneer with a nominal thickness of 4 mm, 6 mm, and 8 mm obtained from the sapwood (S) and heartwood (H)
Roughness of the Pine Veneers
Table 3 shows the statistical evaluation of the specific influence of the factors on the roughness parameters of the pine veneers. The roughness of the pine wood demonstrated significant variation, depending on the tool angle (17° and 21°), wood area (sapwood and heartwood), and veneer side (tight and loose). The wood area had the greatest impact on the Ra and Rzparameters parallel to the grain (26% and 24%, respectively). In contrast, the veneer side had the greatest impact on the Ra and Rz determined perpendicular to the grain (20% and 19%, respectively). The wood area had a slightly lesser impact. The thickness of the veneers had a significant impact only in the case of the Ra and Rz measured perpendicular to the grain (p < 0.05). Among the interactions examined in this study, only the interaction between the veneer thickness and tool angle had a significant impact on the Ra and Rz measured parallel to the grain (14% and 16%, respectively).
The roughness (Ra and Rz) of the pine veneers varied significantly depending on the measurement direction (parallel and perpendicular to the grain). In general, it was concluded that the Ra values measured perpendicular to the grain of the pine veneers were twice as high as those measured parallel to the grain. A similar correlation was observed in the case of the Rz. Wood is a non-homogeneous material and its properties are different in different directions (Li et al. 2018). This is because of the anatomical structure of wood. The measurement perpendicular to the grain contains irregularities that are mainly caused by the size of the structural wood elements. In the case of softwood species, these elements are primarily tracheids. The diameter of tracheids in earlywood is approximately 38 μm, whereas in latewood it is approximately 18 μm (Wagenführ 2007).
Table 3. Statistical Evaluation of the Factors Influencing the Roughness Parameters of the Pine Veneers
Fig. 5. Influence of the tested factors on the Ra of the pine wood veneers: (a) thickness; (b) tool angle; (c) wood area; and (d) veneer side
The roughness of the pine veneers varied significantly depending on the examined factor (thickness, tool angle, wood area, and veneer side) and degree of variation (Figs. 5 and 6).
Fig. 6. Influence of the tested factors on the Rz of the pine wood veneers: (a) thickness; (b) tool angle; (c) wood area; and (d) veneer side
The 4-mm-thick veneers had the smallest roughness values. In general, it can be stated that the roughness values of the 6-mm-thick veneers were close to those of the 8-mm-thick veneers. Dundar et al. (2008a) showed that the surface roughness of a veneer increases with an increasing veneer thickness. Aydin and Çolakoğlu (2005) stated that a rough veneer is difficult to glue because of a lack of close contact between the veneer surfaces. Furthermore, a high roughness promotes conditions that can result in dryout and over-penetration. Higher soaking or peeling temperatures decrease the surface roughness (Aydin et al. 2006). Rohumaa et al. (2016) showed that soaking logs at 70 °C, rather than 20 °C, before peeling produced birch (Betula pendula Roth) veneers with a decreased surface roughness.
The roughness of the pine veneers was smaller after peeling with a knife with a tool angle of 21°. The Ra values (both parallel and perpendicular to the grain) for the veneers obtained using a knife with a tool angle of 21° were approximately 10% lower than those of the veneers obtained with a tool angle of 17°. Similar correlations were found for the Rz.
Lower roughness values were recorded for sapwood and the tight side of the veneer. The Ra values (both parallel and perpendicular to the grain) for veneers obtained from sapwood were approximately 20% lower than for the veneers obtained from heartwood. Similar correlations were found for the Rz. The sapwood being softened to a greater extent under the influence of hot water than the heartwood. As a result, the irregularities in the surface of the sapwood were smaller. Tanritanir et al. (2006) showed that the surface roughness of samples cut from the inner portion of beech logs (Fagus orientalis L.) had a significant dependence on the steaming time.
The tight side of the veneer shows lower roughness values than the loose side. The Ra and Rz values parallel to the grain for the tight side of the veneers were approximately 10% lower than those for the loose side of the veneers. The Ra and Rz values perpendicular to the grain for the tight side of the veneers were approximately 20% lower than those for the loose side of the veneers. The loose side was characterised by a greater roughness because more irregularities and cracks appeared on that side as a result of tensile stresses. Such stresses are an effect of veneer bending during the peeling process. This links to wood anatomy and parameters of machining processes (Mothe 1988; Thibaut et al. 2016). These factors also determine other types of wood machining, i.e. cutting (Söğütlü et al. 2016) or planing (Söğütlü 2010; Söğütlü and Togay 2011; Öhman et al. 2016). Rohumaa et al. (2016) demonstrated that the soaking temperature influenced the roughness of the loose side of birch veneers, but did not have a significant effect on the roughness of the tight side when measured by the stylus method.
Structure and Surface Conditions of the Pine Veneers
Figures 7 and 8 show selected images of the structure and quality of the pine veneer surfaces after the drying process. Typically, after the peeling processes and irrespective of the individual settings, the obtained veneers had a distinguishable tight side (smoother) and loose side (characterised by a greater roughness). The output quality of the raw material (presence of knots) did not have a significant impact on the quality of the circumferential peeling. Generally, it was concluded that, in the case of the studied veneers and irrespective of their thickness, the most visible longitudinal cracks were recorded from veneer sheets obtained from heartwood. When the tool angle was greater, i.e., 21°, the veneers were characterised with smaller roughness values, but the number of lathe checks increased.
The best quality, in terms of the coherence of the structure (the absence of cracks on the transverse section of the veneers) and irrespective of the applied tool angle, was found in the 4-mm-thick veneers obtained from sapwood (Figs. 7a, 7c, and 7e).