Holz Roh Werkst (2007) 65: 157–162 DOI 10.1007/s00107-006-0148-2
ORIGINALARBEITEN · ORIGINALS
Effect of radiata pine juvenile wood on the physical and mechanical properties of oriented strandboard Alain Cloutier · Rub´en A. Ananias · Aldo Ballerini · Robert Pecho
Published online: 28 September 2006 © Springer-Verlag 2006
Abstract This work analyzes the impact of radiata pine (Pinus radiata D. Don) juvenile wood on the physical and mechanical properties of oriented strandboards (OSB). Radiata pine logs were obtained from 10 trees of a 26-year old managed stand located in the 8th Region of Chile. The experimental design considered the proportion of juvenile wood and strand orientation as independent variables. OSB panels of 0.4 m × 0.4 m × 12 mm were produced and tested. The results show that the juvenile wood proportion has a significant impact on the physical and mechanical properties of OSB. Strands orientation had a significant impact on all the properties studied with the exception of the modulus of elasticity in bending. However, this impact was small in all cases and would not change panel grade with the exception of linear expansion. In this case, panels made from tangential strands showed a higher linear expansion. According to these results, radiata pine juvenile wood can be used for the manufacturing of OSB up to a proportion of 70% of the oven-dry wood weight without significant losses of the physical and mechanical properties if the juvenile wood strands are located in the surface layers.
A. Cloutier (u) Centre de recherche sur le bois, D´epartement des sciences du bois et de la forˆet, Facult´e de foresterie et de g´eomatique, Universit´e Laval, Qu´ebec, Qu´ebec G1K 7P4, Canada e-mail:
[email protected] R. A. Ananias · A. Ballerini Departamento de Ingenier´ıa en Maderas, Facultad de Ingenier´ıa, Universidad del B´ıo-B´ıo, Concepci´on, Chile R. Pecho Universidad Agraria de la Selva, Tingo Mar´ıa, Peru
Einfluss von juvenilem Holz der Radiatakiefer auf die physikalischen und mechanischen Eigenschaften von OSB Zusammenfassung In dieser Arbeit wird der Einfluss von juvenilem Holz der Radiatakiefer (Pinus radiata D. Don) auf die physikalischen und mechanischen Eigenschaften von OSB untersucht. Zur Verf¨ugung standen Stammabschnitte von 10 B¨aumen aus einem 26 Jahre alten, bewirtschafteten Bestand in der 8. Region von Chile. Im Versuchsdesign wurden der Anteil an juvenilem Holz und die Orientierung der Sp¨ane als unabh¨angige Gr¨oßen ber¨ucksichtigt. OSB-Platten mit den Abmessungen 0,4 m × 0,4 m × 12 mm wurden hergestellt und gepr¨uft. Die Ergebnisse zeigen, dass der Anteil an juvenilem Holz einen signifikanten Einfluss auf die physikalischen und mechanischen Eigenschaften hat. Die Orientierung der Sp¨ane hatte einen signifikanten Einfluss auf alle untersuchten Eigenschaften mit Ausnahme des Biege-E-Moduls. Jedoch waren diese Einfl¨usse in allen F¨allen gering und w¨urden mit Ausnahme der L¨angsquellung keine Auswirkung auf die Einstufung in G¨uteklassen haben. Platten mit tangentialen Sp¨anen wiesen eine gr¨oßere L¨angenausdehnung auf. Diesen Ergebnissen zufolge kann juveniles Holz der Radiatakiefer bis zu einem Anteil von 70% des Trockengewichts zur Herstellung von OSB ohne signifikante Einbußen der physikalischen und mechanischen Eigenschaften verwendet werden, wenn die Sp¨ane aus juvenilem Holz in die Deckschichten eingebracht werden.
1 Introduction The oriented strandboard (OSB) industry uses various wood species without clear requirements with respect to diameter or straightness of the logs. Moreover, many OSB plants
13
158
around the world are using material from short rotation forests. This represents an advantage compared to plywood which requires large diameter logs obtained from long rotation forests (FAO 1968, Roffael and Schneider 2003). Logs obtained from short rotation forests typically contain a higher proportion of juvenile wood but have lower costs than raw material obtained from natural forests. Juvenile wood becomes a determinant factor in OSB manufacturing when working on the basis of short rotations since it constitutes the main part of the wood volume (Delmastro et al. 1980, Zobel and Van Buijtenen 1989). Moreover, the trees harvested during the thinning treatments performed in short rotation plantations can be used for OSB manufacturing, increasing even more the need for information on the impact of juvenile wood on OSB performance. In Chile, radiata pine (Pinus radiata D. Don) has the potential to be used as prime material in OSB manufacturing. In this context, the impact of radiata pine juvenile wood on OSB properties needs to be addressed. Knowledge of the impact of juvenile wood on the physical and mechanical properties of different composite products is quite limited, particularly on OSB. Technical studies have demonstrated that the use of juvenile wood in composite materials can improve manufacturing variables such as strands geometry and strands compression (Geimer and Crist 1980, Dimitri et al. 1981, Stefaniak 1981, Pugel et al. 1990, Baill`eres et al. 1996, Larson et al. 2001, Pugel et al. 2004). Wasniewski (1989) found a 10 percent increase in the modulus of elasticity (MOE) and modulus of rupture (MOR) in static bending of randomly distributed strandboards made from Douglas fir (Pseudotsuga menziesii (Mirbel) Franco) strands obtained from wood of cambial age varying from one to 50 years. The same properties increased from 30 to 40 percent when using sawmill residues to produce strands. Betanzo and Salinas (1998) have shown that OSB panels made from mature radiata pine wood strands have better shear resistance. Wasniewski (1989) shows that the impact of wood cambial age is significant in board density, in particular in the vertical density profile. A decrease in the linear expansion of random strandboards associated with the increase in wood cambial age was attributed by the same author to the reduction of the microfibril angle in the secondary cell wall of mature wood. At the opposite, Betanzo and Salinas (1998) indicated that OSB manufactured from radiata pine juvenile wood strands was more stable dimensionally. Considering the results mentioned above, the objective of this project was to evaluate the impact of juvenile wood proportion and strands orientation on the physical (swelling and linear expansion) and mechanical (internal bond, strength and stiffness in bending) properties of oriented strandboard.
13
Holz Roh Werkst (2007) 65: 157–162
2 Material and methods 2.1 Material selection and preparation The trees used for the present study were obtained from a 26-year old, intensively managed radiata pine stand of 3.8 ha located in the plantation “El Patagual”, town of Coronel, 8th Region of Chile. Ten trees were randomly selected under the criteria of dominance (dominant and co-dominant), trunk straightness and health. The northern side of each selected tree was marked, the diameter was measured at breast height (DBH) with a caliper, and dominance was determined visually. After felling a tree, a 1.30 m long log was cut at breast height and its larger and smaller diameters were measured (Table 1). Then, the logs were brought to the Silvotechnological Laboratory of the Universidad de Concepci´on, Concepci´on, Chile where a 50 mm thick box-pith piece and two half logs (north and south) were obtained. Subsequently, the half logs were taken to the Wood Technology Laboratory of the Universidad del B´ıoB´ıo, Concepci´on, Chile where test pieces were prepared for wood anatomy characterization and strands preparation. 2.2 Wood anatomy characterization The measurement of growth rings from pith to bark was performed using a scanner and the WINDENDRO software. One measurement per tree was performed. The thickness of the cell walls was measured from microtomed sections in the transverse plane and by image analysis performed with the WINCELL software. For each tree, 20 microtomed sections were prepared following hot water treatment, safranin staining, and mounting in Canada balsam resin. A total of 3000 cell wall thickness measurements were performed for each combination of juvenile – mature wood and earlywood – latewood. The first nine growth rings were considered as juvenile wood and growth rings 16 and over were considTable 1 Sample trees status and logs characteristics Tabelle 1 Soziale Klasse, Abmessungen und Durchmesser des juvenilen Holzes der Stammabschnitte Tree #
1 2 3 4 5 6 7 8 9 10
Status
Log large Log small Log Juvenile diameter diameter length wood (m) (m) (m) diameter (m) Dominant 0.43 0.41 1.30 0.18 Dominant 0.49 0.46 1.30 0.24 Dominant 0.40 0.39 1.30 0.17 Dominant 0.47 0.45 1.30 0.20 Dominant 0.45 0.44 1.30 0.20 Dominant 0.48 0.46 1.30 0.22 Co-dominant 0.44 0.41 1.30 0.19 Dominant 0.44 0.42 1.30 0.20 Dominant 0.45 0.44 1.30 0.20 Dominant 0.46 0.44 1.30 0.22
Log volume (m3 ) 0.141 0.179 0.124 0.168 0.157 0.176 0.146 0.149 0.160 0.163
Holz Roh Werkst (2007) 65: 157–162 Table 2 Experimental design and distribution of the type of strands and strands orientation in the three panel layers (JW: juvenile wood; MW: mature wood) Tabelle 2 Versuchsdesign mit Anteil der Sp¨ane aus juvenilem Holz und Orientierung der Sp¨ane in den drei Plattenschichten (JW: juveniles Holz; MW: adultes Holz)
159
JW proportion (%) 0 30 70 100
Strands orientation
Number of replications
Radial Tangential Radial Tangential Radial Tangential Radial Tangential
4 4 4 4 4 4 4 4
Proportion of furnish by weight of dry wood (%) Face Core Face
35
30
Type of wood Face MW
Core MW
Face MW
MW
JW
MW
JW
MW
JW
JW
JW
JW
35
ered as mature wood on the basis of results obtained in a previous study (Salvo 1999).
were cooled to room temperature and edge-trimmed to the final dimensions of 0.4 m × 0.4 m.
2.3 Strands preparation and OSB manufacturing
2.4 Data analysis
Each one of the two half logs of each tree was sawn to obtain juvenile wood (JW) and mature wood (MW) samples. Radially oriented sticks (90 × 55 × 150 mm3 ; R × T × L) and tangentially oriented sticks (55 × 90 × 150 mm3 ) were prepared. The sticks dimensions were determined according to the size of the laboratory flaker used to produce the strands. From these sticks, radial and tangential strands of the following dimensions were obtained: 0.7 mm in thickness, 25 to 55 mm in width, and 90 mm in length (Fig. 1). Following a preliminary classification, the strands were dried to 4% moisture content at a temperature of 70 ◦ C for about 24 hours. Strand blending was performed using a compressed air sprayer located inside a rotary drum blender. The mats were formed using forming boxes. A panel of the size of the forming boxes was used to pre-press the strand mats. The face layer strands were oriented in the same direction while the core layer strands were cross-oriented. The experimental design and distribution of the type of strands and strands orientation in the three panel layers are summarised in Table 2. The manufacturing parameters are summarised in Table 3. Once the panels were made, they
The statistical analysis of the data obtained on the effect of wood type on ring width and cell wall thickness was performed using a completely random design. For the study of the physical and mechanical properties of the OSB panels, a multifactor analysis of variance was used considering two factors: the proportion of juvenile wood and strands orientation. The significance of the differences between averages was determined using the Tukey multiple range test at a probability level of 95 percent. 2.5 Determination of physical and mechanical properties of OSB The physical and mechanical properties of the OSB panels were determined according to the CSA O437-93 standard.
Table 3 OSB panel manufacturing parameters Tabelle 3 Parameter f¨ur die Herstellung der OSB-Platten Parameter Mat characteristics Resin type Resin solid content Resin viscosity Resin pH Resin content (by oven-dry weight of wood) Mat moisture content (by oven-dry weight of wood) Board thickness Board size Panel target density Mat total weight Mat face layers (2) weight Mat core layer weight
Value Liquid phenol-formaldehyde 50.5% 0.45 Pa · s 11.0 5% 8% 12 mm 0.5 m × 0.5 m 650 kg/m3 2.0 kg 1.4 kg 0.6 kg
Pressing parameters (from Freire 2002 and Pino 2002) Fig. 1 Orientation of the strands used for OSB panel manufacturing Abb. 1 Orientierung der Sp¨ane zur Herstellung der OSB-Platten
Platen temperature Maximum specific pressure Pressing time
195 ◦ C 4.0 MPa 230 s
13
160
Holz Roh Werkst (2007) 65: 157–162
3 Results and discussion
Table 4 Growth ring width and cell wall thickness of juvenile wood and mature wood (10 trees) (JW: juvenile wood; MW: mature wood; EW: earlywood; LW: latewood) Tabelle 4 Jahrringbreite und Zellwanddicke von juvenilem und adultem Holz (10 B¨aume) (JW: juveniles Holz; MW: adultes Holz; EW: Fr¨uhholz; LW: Sp¨atholz)
3.1 Growth ring width and cell wall thickness of juvenile wood Juvenile wood has growth rings significantly wider than those of MW (Table 4). With respect to cell wall thickness, the JW tracheids have thinner cell walls than MW. Also as expected, latewood has significantly thicker cell walls than earlywood. The relatively high standard deviations obtained for cell wall thickness measurements are most likely due to natural variation between the 10 trees studied. From the previously described characteristics of juvenile wood (wider growth rings and thinner cell walls), it can be inferred that during hot pressing the compaction of JW should be easier than that of MW. 3.2 Thickness swelling and linear expansion of OSB panels The thickness swelling values obtained varied between 24 and 32%, well beyond the maximum of 10% allowed by the CSA O437-93 standard for grade O–1 panels (Table 5). The absence of wax in the panel can explain a large part of the high TS obtained. Thickness swelling was significantly larger for panels made from radial strands. This can be explained by the tangential swelling of wood across strand thickness occurring for radial strands. The juvenile wood proportion had a significant impact on TS (Table 5). Tukey‘s multiple range test indicates that the panels made from 100 percent JW have a significantly higher thickness swelling than the panels made from a lower proportion of JW. The panels made from 70, 30 and 0% Table 5 Physical and mechanical properties of the OSB panels as a function of strands orientation and juvenile wood proportion. Corresponding values allowed by standard CSA O437-93 for Grade O-1 are also presented Tabelle 5 Physikalische und mechanische Eigenschaften der OSB-Platten in Abh¨angigkeit von der Orientierung der Sp¨ane und dem Anteil an juvenilem Holz. Zus¨atzlich sind nach der Norm CSA 0437-93 f¨ur Klasse O-1 zugelassene Werte aufgef¨uhrt.
Property Avg. thickness swelling (%) Std. dev. (%) n Tukey
Growth ring width (mm) JW MW
Average n Tukey Std. Dev. Min. Max.
9.7 90 A 1.3 7.7 12.2
4.4 110 B 0.7 3.6 5.5
Cell wall thickness (µm) JW MW EW LW EW LW 4.1 6.2 4.5 8.4 3000 3000 3000 3000 A B A B 1.0 2.5 1.8 3.9 2.7 4.3 2.9 3.5 5.4 13.1 7.3 15.0
JW did not present significantly different thickness swelling levels. The higher thickness swelling in panels made from 100% JW is consistent with the presence of larger ring width and lower cell wall thickness in JW, providing a larger water adsorption capacity. These results are in agreement with others reported in the literature (Geimer and Crist 1980, Pino 2002, Pugel et al. 1990). The linear expansion obtained for panels made from radial strands was lower than the maximum value of 0.20% allowed by the CSA O437-93 standard but it was higher for panels made from tangential strands. The LE was significantly higher for panels made from tangential strands than from radial strands. In this case, the tangential swelling of wood occurred across the width of the strands. Therefore, the swelling component of the strands along the panel length can explain the higher linear expansion of panels made from tangential strands.
Strands orientation Radial Tangential 29.3 25.5 2.9 5.4 8 8 A B
Juvenile wood proportion (%) CSA O437-93 100 70 30 0 Grade O-1 32.0 27.0 26.0 24.0 10 3.3 4.3 3.9 5.0 16 16 16 16 A B B B
Avg. linear expansion (%) Std. dev. (%) n Tukey
0.18 0.04 4 A
0.22 0.05 4 B
0.22 0.05 8 A
0.23 0.06 8 A
0.19 0.02 8 AB
0.17 0.04 8 B
0.20
Avg. internal bond (MPa) Std. dev. (MPa) n Tukey
0.666 0.15 24 A
0.532 0.10 24 B
0.46 0.11 48 C
0.65 0.13 48 A
0.58 0.13 48 B
0.70 0.13 48 A
0.345
Avg. MOR bending (MPa) Std. dev. (MPa) n Tukey
38.7 7.6 4 A
30.2 6.1 4 B
34.4 6.1 8 A
38.7 6.4 8 A
33.4 7.9 8 A
36.2 6.9 8 A
Avg. MOE bending (MPa) Std. dev. (MPa) n Tukey
13
Value
4522 821 4 A
4262 779 4 A
3271 668 8 B
4907 936 8 A
4232 834 8 AB
5159 766 8 A
23.4
4500
Holz Roh Werkst (2007) 65: 157–162
The proportion of JW had a significant impact on LE (Table 5). The results show that there was no significant difference between LE obtained with 100, 70 and 30% JW. However, the LE at 0% JW content was significantly lower than at the 70 and 100% levels. Only the LE values obtained for 30 and 0% JW met the CSA O437-93 standard for grade O-1 panels since they were under 0.20% . These results are in agreement with those reported in the literature (Geimer and Crist 1980, Pugel et al. 1990, Pugel et al. 2004, Wasniewski 1989) and can be explained by the higher longitudinal swelling of JW. 3.3 Mechanical properties of OSB panels It can be observed that values obtained for IB and MOR are well above the minimum values required by the CSA O43793 standard for O-1 panels (Table 5). The value obtained for MOE in bending of panels made from radial strands is just above the minimum value required by the standard while the MOE obtained for panels made from tangential strands does not meet the standard. The internal bond was significantly higher for panels made from radial strands than from tangential strands (Table 5). The IB of panels made from 0 and 70% JW is significantly higher than for panels made from 30 and 100% JW. Panels made from 100% JW resulted in the lower IB. The IBs of panels made from 0 and 70% JW were not significantly different. It is worthwhile to note that when JW was found in the core layer of the panels (100 and 30% JW, Table 2), the IB was lower. This can explain why internal bond was higher when 70% JW was used compared to 30% JW. Therefore, according to our results the presence of JW in the panel core reduces IB more significantly than a higher proportion of JW in the panel surface layers. The results obtained for the MOR and MOE in static bending as a function of strand orientation (Table 5) show that panels made from radial strands present a significantly higher MOR than those made from tangential strands. However, the MOE of panels made from radial strands was not significantly different than the MOE of panels made from tangential strands. No significant difference was found between the MOR values of panels made from different juvenile wood proportions (Table 5). The MOE of the panels made from 0, 30 and 70% JW were not significantly different. However, the MOE of panels made from 100% JW was significantly lower than for panels made from 70 and 0% JW which presented the highest MOE values. The panels made from 70% JW presents juvenile wood in the surface layers and mature wood in the core layer (Table 2). This structure of the panels may have resulted in a higher compaction level of the
161
surface layers of the 70% JW panels compared to the 30 percent JW panels for which the surface layers were made of mature wood. Although not significant statistically, the tendency observed between panels made from 30% and 70% JW can be explained by the location of the JW in the panel layers. According to these results, radiata pine juvenile wood can be used for the manufacturing of OSB up to a proportion of 70% without significant losses of the physical and mechanical properties if the juvenile wood strands are located in the surface layers. The strands orientation did have some impact on the properties studied but not to a point that would justify controlling this parameter in the manufacturing process. Therefore, it seems technically possible to use short rotation radiata pine trees for OSB manufacturing.
4 Conclusions The objective of this project was to evaluate the impact of juvenile wood proportion and strands orientation on the physical and mechanical properties of oriented strandboard. The results obtained have shown the following: 1. Juvenile wood proportion had a significant negative impact on thickness swelling, linear expansion, internal bond and modulus of elasticity in static bending. It had no significant effect on the modulus of rupture in static bending. 2. Strands orientation had a significant impact on all the properties studied to the exception of the modulus of elasticity in bending. However, this impact was small in all cases and would not change panel grade to the exception of linear expansion. In this case, panels made from tangential strands showed a higher linear expansion compared to radial strands. 3. According to these results, radiata pine juvenile wood can be used for the manufacturing of OSB up to a proportion of 70% of oven-dry wood weight without significant loss of the physical and mechanical properties if the juvenile wood strands are located in the surface layers. Therefore, it seems technically possible to use short rotation radiata pine for the manufacturing of OSB panels that meet the CSA O437-93 standard requirements. 4. An industrial trial is required to confirm the results mentioned above. Acknowledgement The authors wish to thank Dr. Luis Valenzuela, Universidad de Concepci´on, Chile, for his valuable advice and technical assistance. This work was supported by the Fundaci´on Andes and the Research Direction of the Universidad del B´ıo-B´ıo, Concepci´on, Chile, under project number 0311/R.
13
162
References Baill`eres H, Chanson B, Fournier-Djimbil M (1996) Plantaciones de a´ rboles de calidad y de r´apido crecimiento de productos forest ales en los tr´opicos. IUFRO WP S5(3):01–04, Tema 12, August 26–31 Betanzo K, Salinas F (1998) Estudio del comportamiento de elementos combinados de madera juvenil y madura de pino insigne (Pinus radiata D. Don) Seminario de Titulaci´on, Ingenier´ıa de Ejecuci´on en Maderas, Universidad del B´ıo B´ıo, Concepci´on, Chile CSA O437-93 (1993) Standards on OSB and Waferboard. Canadian Standards Association Delmastro R, Diaz-Vaz, J, Schlatter J (1980) Variabilidad de las caracter´ısticas t´ecnicas hereditarias del Pinus radiata (D. Don): Revisi´on bibliogr´afica, Investigaci´on y Desarrollo Forestal, Documento de Trabajo (34) Dimitri L, Bismarck C, Bottcher P, Schulze J (1981) Production and use of poplar small-wood particleboard manufacture. Holzzucht 35(1–2):1–7 FAO (1968) Tableros contrachapados y otros paneles a base de madera. Organizaci´on de las Naciones Unidas para la Agricultura y la Alimentaci´on, Roma Freire L (2002) Evaluaci´on de ciclos de prensados en la fabricaci´on de tableros OSB-02 de a´ lamo y pino. Seminario de Titulaci´on Ingenier´ıa de Ejecuci´on en Maderas. Universidad del B´ıo-B´ıo, Concepci´on, Chile Geimer R, Crist J (1980) Structural flakeboard from short-rotation intensively cultured hybrid Populus clones. For Prod J 36(6):42–48 Larson P, Kretchmann D, Clark A, Isebrands J (2001) Formation and properties of juvenile wood in southern pines: A Sinopsis. USDA For Prod Lab Madison, p 42
13
Holz Roh Werkst (2007) 65: 157–162 Pino L (2002) Evaluaci´on de tres especies: pino insigne (Pinus radiata D. Don), coig¨ue (Nothofagus dombeyi), roble (Nothofagus obliqua) en la fabricaci´on de tableros Oriented strand board tipo 02. Seminario de Titulaci´on Ingenier´ıa de Ejecuci´on en Maderas. Universidad del B´ıo-B´ıo, Concepci´on, Chile Pugel A, Price E, Hse C (1990) Composites from southern pine juvenile wood. Part I, Panel fabrication and initial properties. For Prod J 40(1):29–33 Pugel A, Price E, Hse C, Shupe T (2004) Composites from southern pine juvenile wood. Part 3, Juvenile and mature wood furnish mixtures. For Prod J 54(1):47–52 Roffael E, Schneider T (2003) Investigation on partial subtitution of strands in oriented strand boards (OSB) by different lignocellulosic raw materials. Institute for Wood Biology and Wood Technology, Georg August University of G¨ottingen Busgenweg Salvo L (1999) Caracter´ısticas macrosc´opicas de la anatom´ı a de la madera de Pinus radiata D. Don, provenientes de la zona de Arenales Bulnes-Mulch´en para 20, 25 y 30 a˜nos de edad. Proyecto de T´ıtulo Ingenier´ıa Civil en Industrias Forestales. Universidad del B´ıo-B´ıo, Concepci´on, Chile, p 72 Stefaniak J (1981) Use of juvenile wood in production of particleboard: Properties of particleboard produced from pine branch wood. Prace Komiisji Technol Drewna 10:95–116 Wasniewski J (1989) Evaluation of juvenile wood and its effect on Douglas fir structural composite panel. Proceedings of the 23rd Particleboard and Composite Materials Symposium, Washington State University, USA Zobel B, Van Buijtenen J (1989) Wood variation: Its causes and control. Springer Series in Wood Science. Springer, New York