Chapter IV Flexural Properties, Internal Bond Strength, and Dimensional Stability of Medium Density Fiberboard Panels Made from Hybrid Poplar Clones

Table des matières

•  Résumé
• Abstract
• 4.1 Introduction
• 4.2 Material and Methods
  • 4.2.1 Material Collection and Preparation
  • 4.2.2 Panel Manufacturing
  • 4.2.3 Panel Property Testing
  • 4.2.4 Fiber Size Analysis
  • 4.2.5 Data Analyses and Statistical Methods
• 4.3 Results and Discussion
  • 4.3.1 Fiber Size Analysis
  • 4.3.2 Wood Density, Compaction Ratio, and Density Profile
  • 4.3.3 Flexural Properties
  • 4.3.4 Internal Bond Strength
  • 4.3.5 Dimensional Stability
  • 4.3.6 Panel Density versus MOR/MOE
• 4.4 Conclusions
• 4.5 Literature Cited
• 4.6 List of Tables
• 4.7 List of Figures

Chapter IVFlexural Properties, Internal Bond Strength, and Dimensional Stability of Medium Density Fiberboard Panels Made from Hybrid Poplar Clones
Authors: Jun Li Shi, S. Y. Zhang, and Bernard Riedl

Résumé

Les propriétés en flexion, de cohésion interne, et stabilité dimensionnelle des panneaux de fibres de densité moyenne faits à partir de bois des trois clones du peuplier ( Populus . spp.), dont les codes sont 915303, 915311, et 915313 ont été étudiées. L’analyse de variance (ANOVA) et l’analyse de covariance (ANCOVA) ont été faites dans cette étude pour examiner les différences de module de rupture (MOR) et de module d'élasticité (MOE) des panneaux de fibres de densité moyenne faits à partir des trois hybrides de peuplier. Les résultats indiquent que les MOR des panneaux faits à partir du clone 915311 étaient sensiblement plus élevés que ceux des panneaux faits à partir des clones 915303 ou 915313; cependant, il n’y avait aucune différence significative entre les MOR de panneaux faits à partir des clones 915303 ou 915313. Les MOE des panneaux de fibres de densité moyenne faits à partir du clone 915311 ont été les plus élevés, et sensiblement différents de ceux des panneaux faits à partir de clones 915303 ou 915313 alors que les MOE des panneaux faits à partir de fibres tirées du clone 915303 étaient les plus bas et sensiblement inférieurs à ceux des panneaux faits de fibres tirées du clone 915313. Les panneaux de fibres de densité moyenne faits à partir des deux clones 915303 et 915311 étaient supérieurs aux panneaux faits à partir du clone 915313 en ce qui concerne la cohésion interne; mais il n’y avait aucune différence significative entre la cohésion interne (IB) des panneaux faits à partir des clones 915303 ou 915311. La stabilité dimensionnelle des panneaux de fibres de densité moyenne a été évaluée via l’expansion linéaire (LE), le gonflement en épaisseur (TS), et l’absorption d’eau, et aucune différence significative n'a été trouvée parmi les trois types de panneaux. Cette étude montre un effet significatif de variation clonale de peuplier hybride sur les propriétés en flexion et de cohésion interne des panneaux de fibres de densité moyenne, et elle suggère que des améliorations des panneaux MDF dans les propriétés en flexion et en cohésion interne peuvent être obtenues en choisissant un ou plusieurs hybrides à des fins de sélection reproductive. De plus, la densité des panneaux étant un facteur considérable influençant les MOR et MOE des panneaux de fibres de densité moyenne, des relations linéaires significatives entre les MOR, MOE et la densité des panneaux ont été déterminées.

Abstract

Flexural properties, internal bond strength, and dimensional stability of medium density fiberboard (MDF) panels made from three hybrid poplar ( Populus spp.) clones with codes 915303, 915311, and 915313 were studied. Analysis of variance (ANOVA) and analysis of covariance (ANCOVA) were both performed in this study to test the differences in modulus of rupture (MOR) and modulus of elasticity (MOE) of MDF panels made from the three poplar hybrids. Results indicate that MOR of MDF panels made from clone 915311 was significantly higher than those of panels made from clones 915303 or 915313; however, there was no significant difference in MOR between panels made from clones 915303 or 915313. MOE of MDF panels made from clone 915311 was the highest value, which was significantly different from those of panels made from either clone 915303 or 915313; MOE of panels made from clone 915303 was the smallest and significantly lower than those of panels from clone 915313.MDF panels made from both clones 915303 and 915311 were superior to those panels made from clone 915313 in internal bond (IB) strength; but there was no significant difference in IB between panels made from clones 915303 or 915311. Dimensional stability of MDF panels was evaluated by linear expansion (LE), thickness swell (TS), and water absorption, and no significant differences were found among the three types of panels. This study shows a significant effect of hybrid poplar clonal variation on flexural properties and internal bond strength. This suggests that improvements in MDF panel flexural properties and internal bond strength may be made through tree breeding. Additionally, panel density was a factor influencing MDF panel MOR and MOE considerably; as significant linear relationships between MOR, MOE and panel density were determined. 

4.1 Introduction

The materials for paper making and composite panel manufacturing have been primarily and conventionally coming from softwood species such as spruce, pine, fir, etc. However, a shortfall in softwood supply is currently being faced by these industries. Recently, forest products companies in Canada and the United States have become interested in fast-growing species as potential replacements for softwood chips to sustain raw material supply.
Hybrid poplar is a hardwood species with a high growth rate and short rotation, which can be expected to produce a promising wood fiber source with high yields (Dix et al. 1999; Cisneroset al. 2000). In Alberta, five major pulping projects have been built recently using aspen resources (Cisneroset al. 2000). Hybrid poplar has also been studied for its potential as a raw material substitute for composite panel products. In fact, the advantages of using hybrid poplar as softwood substitution for composite panel making include not only in its high wood fiber yield, but also good performance of the end products. Geimer (1986) studied the properties of structural flakeboard panels manufactured from poplar, tamarack, and pine. Results indicate that modulus of rupture (MOR), modulus of elasticity (MOE), thickness swell (TS), and linear expansion (LE) of the flakeboards made from tamarack and pine were inferior to those from poplar. Short rotation and intensively cultured hybrid poplar was also investigated as a possible raw material source for hardboard (Myers and Crist 1986). Hardboard panels were tested for strength properties and dimensional stability, and results indicate that hybrid poplar is a suitable raw material for hardboard manufacturing. Moreover, high bending and internal bond (IB) strength and low TS in water of MDF panels made from 19-year-old poplar wood bonded with either a melamine-reinforced urea-formaldehyde (UF) resin, a tannin-formaldehyde resin, or a polymeric diisocyanate (PMDI) resin were also reported (Roffael and Dix 1994). The better performance of composite panels made from poplar wood may be due to its wood and fiber characteristics. As we know, poplar wood is low in density, thus, causing relatively high compaction ratio that has been reported to lead to superior panel strength properties (Maloney 1993; Hsu 1997; Peter et al. 2002; Shi et al. 2005). The thin-walled fiber can be packed better during pressing resulting in more gluelines per unit panel thickness.
The requirements of wood fiber characteristics for various end uses are diverse (Zhang et al. 1997). For some applications such as lumber, construction and plywood, low density wood is not preferred because of the low strength of such wood. On the contrary, some studies showed the advantages of using light wood species for fiberboard making (Nelson 1973; Woodson 1976). In addition, it has been known that wood and fiber properties (e.g. wood density and fiber cell wall thickness) can be affected by genetic control on hybrid poplar trees (Ivkovich, 1996; Law et al. 1997; Xing 2000; Cisneros et al. 2000; Savita 2001). Nevertheless, very little attention has been paid to genetic manipulation and selection of poplar trees for specific end uses. Only a few studies on the utilization of wood and fiber produced from genetically manipulated poplar trees as raw material for composite panels manufacturing were found in the literature. Geimer and Crist (1980) investigated the properties of structural flakeboard panels made from five hybrid poplar clones. It was found that the clonal variation had an effect on structural flakeboard panel properties; some panels performed better than others depending on furnish origins. Peter et al. (2002) studied the flexural properties, IB, density, water absorption, and TS of OSB panels made from eleven hybrid poplar clones. The flexural properties of OSB panels made from some clones were superior to those of panels made from others. It was concluded that these Populus hybrids showed great promise for use in structural panel products because of superior flexural and IB properties.
In this study, laboratory MDF panels were manufactured from three hybrid poplar clones and these panels were evaluated for flexural properties, internal bond strength and dimensional stability. We intend to examine the flexural properties, internal bond strength, and dimensional stability of MDF panels made from these three clones. The information derived from this study is essential to assist in poplar genetic and tree breeding program.

4.2 Material and Methods

4.2.1 Material Collection and Preparation

The material for this study came from a hybrid poplar clonal trial established by the Forest Research Branch of the Québec Ministry of Natural Resources in St-Ours, Southern Québec, Canada in 1993. The poplar trees were planted at 1.5×3.5 m spacing on this site, which is a part of the Champlain marine deposit with a rich salty-clay soil (40 % clay). Systematic thinning was carried out in the spring of 1996 and the spacing after the thinning was 2.5×3.0 m (Zhang et al. 2003). Four trees from three clones with codes 915303, 915311 and 915313, coming from a hybrid family of P. maximowiczii and P. balsamifera , were harvested from this site in December 2002 at an age of nine years. Butt logs of the three poplar hybrids were selected, debarked, and subsequently chipped using a portable chipper. Wood chips were collected, pooled by poplar clone, and then refined in a pressurized disc refiner located at the pilot plant of Forintek Canada Corp. Refining was processed without wax and resin injection. Moisture contents of the wood chips in the pre-steaming bin for clones 915303, 915311 and 915313 were 39.0 %, 44.0 % and 40.0 %, respectively. Specific refining energy for clones 915303, 915311, and 915313 were 134 KWh/t, 121 KWh/t, 125 KWh/t, respectively. To prevent fibers converted from different clones from mixing, 30 minutes of transition time was allowed between the three types of clonal wood materials while refining, and the fibers produced in the transition time period were discarded. 

4.2.2 Panel Manufacturing

Fibers generated from the three poplar clones were dried in a laboratory-scale dryer until the moisture content reached 2-3 %. Then, fibers were passed through a hammer mill to fully separate the fibers from each other. Since almost all panel properties can be improved by means of increasing resin content (Maloney 1993), in this study resin was blended into the fibers at a relatively low level so as not to conceal any effect of clonal variation on MDF panel properties. Thus, 10 % (by weight of dry fiber) Borden 302 urea formaldehyde (UF) resin (65 % solid content, no catalyst added) was first diluted, and then slowly sprayed onto the fibers using a laboratory-scale blade blender. At the same time, 0.5 % wax was blended together with the fibers. No other additives were added into furnish. Resin- and wax-blended fibers were passed through the hammer mill once again to disperse fiber balls. The average moisture contents of the fibers after wax and resin blending were 10.7 %, 11.7 % and 12.3 % for the clone order listed previously, respectively. Mats were hand-formed immediately in a wood frame of 610 by 610 mm. All panels were manufactured at a target density of 740 kg/m³ at the pilot plant of Forintek Canada Corp. Target panel thickness was 12 mm. Panels were pressed under the same schedule which consisted of a closing period of 160 s, maintained under pressure for another 160 s, and gradual relieved of pressure over 40 s. The temperature of the two platens was set at 135 ^oC. Due to the fact that panel density profile through thickness is one of the most important factors in determining properties of composite panels (Suchsland and Woodson 1974; Kelly 1977; Harless et al. 1987; Winistorfer et al. 1996; Wang et al. 2001; Wang et al. 2004), initially, we designed the press schedule mentioned previously in order to realize flatter panel density profiles, which may result in reduction of variations in panel properties caused by different density profiles. Three replicate panels were made for each poplar clone. Panels were trimmed immediately after pressing.

4.2.3 Panel Property Testing

All MDF panels were kept for conditioning in a chamber at 22 ^oC and 65 % relative humidity (RH) for at least four weeks until the panels reached a constant equilibrium moisture content. Testing of modulus of rupture (MOR), modulus of elasticity (MOE), internal bond (IB), liner expansion (LE), thickness swell (TS) and water absorption was carried out in accordance with the standard methods of ASTM D 1037-99 (2001) for evaluating properties of wood-based fiber and particle panel materials and ANSI A 208.2-2002 (2002) for evaluating MDF for interior application. Flexural properties were determined on specimens of 338 by 75 mm. Three specimens were cut from each panel for flexural properties testing, producing nine specimens in total for each clone. Ten IB specimens of 50 by 50 mm were taken from each panel, resulting in thirty specimens for each clone. LE was measured from two specimens of 305 by 76 mm taken from each panel, for a total of six per clone. Specimens of 152 by 152 mm were used for both TS and water absorption assessment; two were cut from each panel, making six specimens in total for each clone. LE specimens were first stored in a conditioning chamber at 22 ^oC and 50 % RH for four weeks until their weight became constant, and then kept in another chamber at 22 ^oC and 80 % RH for another four weeks until the moisture content of specimens was checked to be at a constant equilibrium moisture content. LE was obtained by the variation in length caused by the change of relative humidity from 50 to 80 % divided by the length measured at 50 % RH. TS was calculated as the increase in thickness after a 24 h water immersion divided by the thickness measured on the specimens that had reached constant equilibrium moisture content under a condition of 22 ^oC and 65 % RH. Water absorption was calculated as the percentage of water absorbed in the specimens after 24 h water submersion based on the weight of specimens that had reached constant equilibrium moisture content at 22 ^oC and 65 % RH. About 1.5 mm was sanded from both surfaces of all IB specimens before they were glued to blocks. The weight and dimension of all specimens for panel flexural properties, internal bond strength and dimensional stability were measured before testing and densities of these specimens were calculated for the purpose of analysis of covariance (ANCOVA). Panel density refers to the density of specimens calculated from the weight and volume obtained from the specimens conditioned at 22 ^oC and 65 % RH for four weeks. Moisture contents of the three types of panels were measured on the equilibrated moisture content MOR/MOE specimens after four weeks of conditioning while following the methods described in ASTM D 1037-99 (2001). Panel density profiles were determined from the ten IB specimens before they were sanded and glued to blocks, and the collected data were averaged for three types of panels. The surface and bottom of each IB specimen were marked clearly to ensure that the directionality of the surfaces was maintained during density profile testing. 

4.2.4 Fiber Size Analysis

A Bauer-McNett Classifier was used to determine the fraction distribution of fibers generated from the three poplar clones. The procedures described in TAPPI T 233 cm-95 (1995) were followed for the testing.

4.2.5 Data Analyses and Statistical Methods

Statistical Analysis System (SAS) (1990) software package was used for data analysis in this study. Analysis of variance (ANOVA) was employed to compare the mean property values of MDF panels made from three poplar clones. Since the influence of panel density on panel properties, especially on flexural properties, has been reported in several studies (Maloney 1993; Olson 1996; Peter et al. 2002; Shi et al. 2005), ANCOVA was also performed using panel density as a covariate for the purpose of adjusting the mean MOR and MOE values, which were partly attributed to panel density. The removal of panel density influence on panel properties may improve the precision of the analysis and meanwhile reduce the error. Before performing ANCOVA, assumptions such as homogeneity and linearity of within-group regression, randomization, statistical independence of covariate and treatment, etc, must be tested to be sure that they are met; otherwise, the analysis will lead to misleading results (Huitema 1980). All pairwise multiple comparison procedures were conducted to compare the adjusted means using the PDIFF option in SAS with an α level of 0.05.
Simple linear regression analysis was also performed as a consequent statistical technique to develop equations describing the relationships between MOR, MOE and panel density. The coincidence of the three regression lines was tested by introducing dummy variables into the analysis of variance. 

4.3 Results and Discussion

4.3.1 Fiber Size Analysis

It is known that fibers with different size distribution can have an impact on MDF panel properties (Groom et al. 1999). The size distributions for the three types of fibers are shown in Table 4-1. Since the percentages of fibers distributed in mesh size ranges of >3.240 mm², 0.828-3.240 mm², 0.281-0.828 mm², 0.017-0.281 mm², and smaller than 0.017 mm² for the three clones were similar, we assumed that the refining process produced fibers with similar size for the three clones. Therefore, it can be assumed that the influence of fiber size on properties of the three types of panels was identical. 

4.3.2 Wood Density, Compaction Ratio, and Density Profile

Panel average densities and compaction ratios are presented in Table 4-2. Generally, wood density is believed to be the most important wood characteristic in determining final panel properties; the basic requirement of raw materials for making panels with acceptable properties is a relatively low wood density (Maloney 1993; Hsu 1997; Woodson 1976). Basically, low density wood is easier to consolidate into target thickness. As a matter of fact, for the same panel density, compaction ratio of panels made from low density wood is always higher. Compaction ratio is of importance since well-bonded panels are primarily associated with high compaction ratios (Maloney 1993; Hsu 1997). Shi et al. (2005) reported that properties of MDF panels made from black spruce juvenile wood were significantly superior to those of panels made from mature wood. They explained that it was due to the low density of juvenile wood. However, in this study, these relationships were not apparent because only a very narrow range of wood density was involved in the experiment; nevertheless, the adjusted MOR and MOE of the three groups appeared statistically different with the use of ANCOVA since the precision of the analysis was improved.
Slight differences existed among the average density profiles of MDF panels made from the three poplar clones as shown in Figure 4-1 even though we made these panels under the same press schedule. Small difference in mat moisture content (10.7 %, 11.7 % and 12.3 %) seemed not to be a cause of different panel properties. So the slight difference in density profile might result from fiber origin while panels were compressed under heat and pressure. Since the density profiles of the three types of panels were nearly comparable, it is assumed that differences in panel properties were not the cause of density profile. However, the flatter density profile with large face layer thickness may lower the flexural properties of all types of panels. 

4.3.3 Flexural Properties

1) Modulus of Rupture (MOR)

The unadjusted mean MOR of MDF panels made from the three poplar clones is shown in Table 4-2. No significant difference was found in MOR among the panels from the three clones by Duncan’s multiple-range test at the 0.05 of significance level if the effect of panel density was not considered in the analysis. However, as mentioned previously, panel density is a factor in determining panel properties (Maloney 1993; Olson 1996; Peter et al. 2002; Shi et al. 2005). Taking panel density into consideration, panel density was then analyzed as a covariate using ANCOVA in order to eliminate the effect on MOR caused by different panel density and and to adjust mean MOR values. First, the assumptions of ANCOVA such as linearity and homogeneity of within-group regressions were tested to ensure the data fits the ANCOVA model. The assumption testing showed that all the assumptions were met. A statistical comparison of ANOVA and ANCOVA for MOR is presented in Table 4-3. F=0.78 (P=0.4702) was obtained from ANOVA, showing that there were no significant differences in panel MOR among the three types of panels. With the use of ANCOVA, F=4.50 (P=0.0236) was acquired, which was to say that here, the clonal variation effect on panel MOR was significant (α=0.05). The results from regression analysis indicate a significant linear relationship between MOR and panel density for all three types of panels. The effect of panel density on MOR was indicated by a significant linear regression test, where F=96.99 (P<.0001) was calculated (Table 4-3, Figure 4-2). The adjusted mean MOR for the three types of poplar panels is listed in Table 4-2. Adjusted mean MOR of MDF panels made from clone 915311 was significantly higher than that of panels from clones 915303 and 915313, there was no significant difference in MOR between panels made from clones 915303 and 915313. The difference in MOR between the panels made from three poplar hybrids indicates a significant effect of raw material on panel performance. 

2) Modulus of Elasticity (MOE)

The same analytical procedures were employed for MOE data. According to ANOVA, no significant differences were found between panels made from the three hybrids by Duncan’s test (Table 4-2). Consequently, ANCOVA was performed in order to eliminate the effect of panel density on MOE. ANCOVA assumptions were tested by verifying linearity and homogeneity of within-group regressions. A linear regression test was significant, and the slopes of within-group regression were all equal, meaning that there were no interactions existing among the three groups (Table 4-3, Figure 4-3). The adjusted mean MOE was calculated by using the same regression slope for the three types of panels, and the results are presented in Table 4-2. MOE of MDF panels made from clone 915311 was significantly higher than that of panels made from clones 915303 and 915313, while MOE of panels made from clone 915303 was the smallest and significantly lower than that of panels from clone 915313 (Table 4-2). The precision of testing was greatly improved and error was reduced by using ANCOVA. Significant difference in MOE between three types of panels was observed meaning that there was an impact of clonal variation on the performance of panels made from hybrid poplar. 

4.3.4 Internal Bond Strength

The mean IB of MDF panels made from clones 915303 and 915311 was significantly higher than that of panels from clone 915313; there was no significant difference in IB strength between the panels made from clones 915303 and 915311 by Duncan’s multiple-range test (Table 4-2). The results show that there was no impact of clones 915303 and 915311 on panel IB strength; however, compared to the panels made from those two clones, clone 915313 had a significant effect on panel IB. 

4.3.5 Dimensional Stability

LE, TS, and water absorption of MDF panels made from the three poplar clones were not significantly different at a significance level of 0.05 (Table 4-2). Therefore, the clonal variation effect on MDF panel dimensional stability was not significant. 

4.3.6 Panel Density versus MOR/MOE

Panel density affected considerably panel flexural properties. Significant linear relationships between panel MOR and MOE and panel density for all three types of panels were found as mentioned previously. Equations describing the relationships (Figures 4-2 and 4-3) were developed using simple linear regression from the model Y = b₀ + b₁x, where Y represents panel MOR or MOE, and x is panel equilibrium density. The relationships between MOR, MOE and panel density indicate the dependence of the two variables on panel density, that is, MOR and MOE of panels made from the three poplar hybrids can be improved by increasing panel density. This is due to more fiber being used in heavier MDF panels resulting in more resistance against mechanical loads (Maloney 1993).
However, a further question is whether there is a difference between the three regression lines, or, they are coincident.
When dummy variables are used in regression analysis, it is important to ensure that all groups can be distinguished in the analysis. Therefore, two dummy variables must be defined in order to index three groups. Thus, variables z₁ and z₂ need to be defined for the data to distinguish the three different clones, which are shown in the following.
z₁ = 1if the panels were made from clone 915303
0otherwise
z₂ = 1if the panels were made from clone 915311
0otherwise
Three clones can be obtained by combining z₁ and z₂.
z₁=1 z₂=0: panels made from clone 915303
z₁=0 z₂=1: panels made from clone 915311
z₁=0 z₂=0: panels made from clone 915313
A model for the relationship between MOR and panel density is given by
Y = b₀ + b₁x + b₂z₁ + b₃z₂ + b₄xz₁ + b₅xz₂ + ε
where Y represents MOR of panels;
x is panel density; z₁, z₂ are clones with codes 915303 and 915311;
ε is the error term.
The model for the panels made from the three poplar clones can be written as:
model for clone 915303: Y = (b₀ + b₂) + (b₁ + b₄)x + ε
model for clone 915311: Y = (b₀ + b₃)+ (b₁ + b₅)x + ε
model for clone 915313: Y = b₀ + b₁x + ε
The hypothesis of the coincidence of the three regression lines is the one that the slopes and intercepts are the same for all three types of panels, which can be written as follows:
H₀: b₂ = b₃ = b₄ = b₅ = 0
Thus, we can test the hypothesis of coincidence by testing significance of the terms z₁, z₂, xz₁, and xz₂. The three regression lines shown in Figure 4-2 were tested (Table 4-4), and the terms z₁, z₂, xz₁, and xz₂ were not significant indicating that the hypothesis that the three regression lines were coincident was accepted. Thus, we can fit a single overall regression line to the three lines. A model can be given as follows to describe the relationship of MOR to panel density for the three clones.
Y = - 51.0521 + 0.0967 x (R² = 0.8954)
where Y is MOR of panels made from the three hybrid poplar clones;
x is panel density.
Through the same analysis method, the coincidence of the regression lines for MOE was tested subsequently (Table 4-4). The hypothesis of coincidence was accepted as well. Therefore, only one overall equation can be used to describe the relationship between MOE and panel density for all three clones. The equation is shown below.
Y = - 4214.7052 + 7.9633 x (R² = 0.8677)
where Y is MOE of panels made from the three hybrid poplar clones;
x is panel density.

4.4 Conclusions

The following conclusions can be made from this study:

1. MOR of MDF panels made from clone 915311 was significantly higher than those of panels from clones 915303 and 915313; however, there was no significant difference in MOR between panels made from clones 915303 and 915313. 
2. MOE of MDF panels made from clone 915311 was the highest and was significantly different from those of panels from clone 915303 and 915313; MOE of panels from clone 915303 was the smallest and was significantly lower than that of panels from clone 915313.
3. MDF panels made from either clone 915303 or 915311 were superior to panels from clone 915313 in IB strength; there was no significant difference in IB between panels made from clone 915303 and 915311.
4. No significant differences were found in LE, TS, and water absorption among panels made from the three poplar hybrids; the effect of clonal variation on the dimensional stability of hybrid poplar MDF panels was not appreciable.

This study shows that clonal variation had a significant effect on the flexural properties and internal bond strength of hybrid poplar MDF panels; however, its effect on panel dimensional stability was not significant. The difference in flexural properties and internal bond of panels may result from different wood fiber characteristics of the three poplar hybrids. MDF panel properties in relation to wood fiber characteristics will be studied and reported in Chapter VI. Knowing the relationships between panel properties and wood fiber characteristics, it is likely that improvement in panel flexural properties and internal bond strength can be achieved through tree breeding. 

4.5 Literature Cited

American National Standards Institute (ANSI). Medium Density Fiberboard (MDF) for Interior Application . ANSI A 208.2-2002. Composite Panel Association, Gaithersburg. MD. 83(7): 60; 2002.
American Society of Testing and Materials (ASTM). Evaluating properties of wood-based fiber and particle panel materials . ASTM D 1037-99. Vol. 04.10. ASTM, Philadelphia. PA. pp. 141-170; 2001.
Cisneros, H. A., Belanger, L., Gee, W. Y., Watson, P. A. and Hatton, J. V. Wood and fiber properties of hybrid poplars from southern British Columbia . TAPPI Journal. 2000.
Dix, B., Thole, V. and Marutzky, R. Poplar and eucalyptus wood as raw material for wood-based panels . Eurowood Technical Workshop Proceedings: Industrial End-uses of Fast-grown Species. CNR/IRL and CNR/ITL. Florence. Pp93-102; 1999.
Geimer, R. L. Properties of structural flakeboard manufactured from 7-year-old intensively cultured poplar, tamarack, and pine . Forest Prod. J. 36(4): 42-46; 1986.
Geimer, R. L. and Crist, J. B. Structural flakeboard from short-ratation, intensively cultured hybrid populus clones . Forest Prod. J. 30(6): 42-48; 1980.
Groom, L. H., Mott, L. and Shaler, S. M. Relationship between fiber furnish properties and the structural performance of MDF . 33^rdInternational Particleboard/Composite Materials Symposium Proceedings. April 13-15, Washington State Univ., Pullman, Washington, USA. 1999.
Harless, T. E. G., Wagner, F. G., Short, R. D., Seale, R. D., Mitchell, P. H. and Ladd, D. S. A model to predict the density profile of particleboard . Wood Fiber Sci. 19(1): 81-92; 1987.
Hsu, W. E. Wood quality requirements for panel products . Proceeding of CTIA/IUFRO International Wood Quality Workshop, Timber Management toward Wood Quality and End-product Value. Forintek Canada Corp. Quebec City, Canada. I-7/10; 1997.
Huitema, B. E. The Analysis of Covariance and Alternatives . Jone Wiley & Sons, New York. 1980.
Ivkovich, M. Genetic variation of wood properties in Balsam poplar (Populus balsamifera L.) . Silvae Genet. 45(2/3): 119-124; 1996.
Kelly, M. W. Critical literature review of relationship between processing parameters and physical properties of particleboard . Gen. Tech. Rep. FPL-10. USDA Forest Serv., Forest Prod. Lab., Madison. WI. 1977.
Law, K., Valade, J., Rioux, S. and Lanouette, R. Wood and paper properties of five short-rotation poplar clones grown in Québec . October 19, 1997 pulping conference: proceedings (TAPPI): 847-856; 1997.
Maloney, T. M. Modern Particleboard and Dry-Process Fiberboard Manufacturing . Miller Freeman Publications. San Francisco, CA. 1993.
Myers, G. C. and Crist, J. B. Feasibility of manufacturing hardboard from short-rotation intensively cultured Populus . Forest Prod. J. 36(1): 37-44; 1986.
Nelson, N. D. Effects of wood and pulp properties on medium-density, dry-formed hardboard . Forest Prod. J. 23(9): 72-80; 1973.
Olson, B. D. Developing wood composites using small diameter timber resources from dense, stagnant stands . M.S. thesis. Washington State Univ., Pullman, WA; 1996.
Peter, J. J., Bender, D. A., Wolcott, M. P. and Johnson, J. D. Selected properties of hybrid poplar clear wood and composite panels . Forest Prod. J. 52(5): 45-54; 2002.
Roffael, E. and Dix, B. Mechanical-technological properties of MDF from poplar wood . Holz Roh Werkst. 52(1): 48; 1994.
SAS Institute, Inc. SAS/STAT User’s guide . Cary, N.C. 1990.
Savita, G. Wood and paper properties of poplar clones . IPPTA. 13(4):35-37; 2001.
Shi, J. L., Zhang, S.Y. and Riedl, B. Effect of juvenile wood on strength properties and dimensional stability of black spruce medium density fiberboard panels . Holzforschung. 59(1): 1-9; 2004.
Suchland, O. and Woodson, G. Effect of press cycle variables on density gradient of medium density fiberboard . In proceedings 8^thparticleboard symp. Pp 375-396; 1974.
Tappi Standard. Fiber length of pulp by classification . T 233cm-95. 1995.
Wang, S., Winistorfer, T. M., Young, T. M. and Helton, C. Step-closing pressing of medium density fiberboard; Part I: Influences on the vertical density profile . Holz-als-Roh-und-Werkstoff. 59(1-2): 19-26; 2001.
Wang, S., Winistorfer, P. and Young, T. Fundamentals of vertical density profile formation in wood composites. Part III. MDF density formation during hot-pressing . Wood Fiber Sci. 36(1): 17-25; 2004.
Winistorfer, P. M., Young, T. M. and Walker, E. Modeling and comparing vertical density profiles . Wood Fiber Sci. 28(1): 133-141; 1996.
Woodson, G. E. Properties of medium-density fiberboard related to hardwood specific gravity . Proceeding 10^th Washington State University Particleboard Symposium. Pullman, WA. 1976.
Xing, X. Genetic variation in wood density of triploid clones of Populus tomentosa . Beijing Linye Daxue Xuebao. 22(6): 16-20; 2000.
Zhang, S. Y., Gosselin, R. and Chauret, G. Wood quality: its definition, impact and implication for value-added timber management and end uses . Timber management toward wood quality and end product value. Proceeding of CTIA/IUFRO international wood quality workshop, Québec City, Canada. 1997.
Zhang, S.Y., Yu, Q., Chauret, G. and Koubaa, A. Selection for both growth and wood properties in hybrid poplar clones . Forest Sci. 49(6): 1-8; 2003.

4.6 List of Tables

Table 4-1 Size distribution of the fibers produced from the three hybrid poplar clones.
Table 4-2 Mean modulus of rupture (MOR), modulus of elasticity (MOE), internal bond (IB), linear expansion (LE), thickness swell (TS), and water absorption of MDF panels made from the three hybrid poplar clones.
Table 4-3 Comparison of analysis of variance (ANOVA) and analysis of covariance (ANCOVA) for panel modulus of rupture (MOR) and modulus of elasticity (MOE).
Table 4-4 Test for coincidence of the relationships between MOR/MOE and panel density.

Table 4-1 Size distribution of the fibers produced from the three hybrid poplar clones.

Note: Numbers in columns ‘>3.240’, ‘0.828-3,240’, ‘0.281-0.828’, and ‘0.017-0.281’ were the percentages of fibers retained on the screens with mesh sizes of 3.240 mm², 0.828 mm², 0.281 mm², and 0.017 mm². Numbers in columns ‘<0.017’ were the percentages of fibers passed through the screen with mesh size of 0.017 mm².
Methods described in Tappi 233 cm-95 were followed.

Table 4-2 Mean modulus of rupture (MOR), modulus of elasticity (MOE), internal bond (IB), linear expansion (LE), thickness swell (TS), and water absorption of MDF panels made from the three hybrid poplar clones.

Note: Unadjusted means with the same small letter were not significantly different by Duncan’s multiple-range test ( p =0.05). Adjusted means with the same capital letter were not significantly different by multiple comparison procedures for all pairwise comparisons using the PDIFF option in SAS ( P =0.05).
Methods for panel density and moisture content determination were in accordance with ASTM D 1037-99. Compaction ratios were based on panel equilibrium density and density of wood chips.
S represents standard deviation. 

Table 4-3 Comparison of analysis of variance (ANOVA) and analysis of covariance (ANCOVA) for panel modulus of rupture (MOR) and modulus of elasticity (MOE).

Note: SS is sum of square. DF is degree of freedom. F values were significant at p =0.05. 

Table 4-4 Test for coincidence of the relationships between MOR/MOE and panel density.

4.7 List of Figures

Figure 4-1 Average density profiles of MDF panels made from the three hybrid poplar clones.
Figure 4-2 Relationship between modulus of rupture (MOR) and panel density.
Figure 4-3 Relationship between modulus of elasticity (MOE) and panel density.

Figure 4-1 Average density profiles of MDF panels made from the three hybrid poplar clones.

Figure 4-2 Relationship between modulus of rupture (MOR) and panel density.

Note: y₁, y₂, y₃ correspond to the modulus of rupture (MOR) of MDF panels made from hybrid poplar clone 915303, 915311, and 915313, respectively; x refers to panel density.

Figure 4-3 Relationship between modulus of elasticity (MOE) and panel density.

Note: y₁, y₂, y₃ correspond to the modulus of elasticity (MOE) of MDF panels made from hybrid poplar clone 915303, 915311, and 915313, respectively; x refers to panel density.

For further details log on website :
http://archimede.bibl.ulaval.ca/archimede/fichiers/23592/ch01.html