Table des matières
Authors:
Jun Li Shi, Bernard Riedl, and S. Y. Zhang
Strength properties and dimensional stability of medium density fiberboard (MDF) panels made from black spruce ( Picea mariana(Mill.) BSP.) top, mid, and butt logs were studied. An analysis of covariance (ANCOVA) was conducted to examine the effect of log position in the tree on panel modulus of rupture (MOR) and modulus of elasticity (MOE) using panel density as a covariate. The results indicate that MOE and internal bond (IB) strength of MDF panels made from top and mid logs were significantly superior to those of panels made from butt logs; however, there was no significant difference in MOE and IB between panels made from top and mid logs. Water absorptions of the panels made from top and mid logs were significantly lower than that of panels from butt logs, and the difference in water absorption between panels made from top and mid logs was not significant. Thickness swell (TS) of the panels made from top logs was significantly smaller than that of panels from mid and butt logs. The panels made from butt logs yielded the highest TS, which was significantly different from those of panels from top and mid logs. The differences in linear expansion (LE) among the panels fabricated from top, mid, and butt logs were not significant. The comparison of MOR was dependant upon panel density due to the interactions in the three groups. Panel density affected both panel MOR and MOE considerably; however, its impact on IB, LE, TS, and water absorption was not significant in this study. Linear relationships between MOR, MOE and panel density were found and the equations describing the relationships were developed.
Black spruce ( Picea mariana (Mill.) BSP.) is one of the most important commercial species in Eastern Canada. Black spruce wood is highly valued for different end uses including pulpwood, lumber, panels and many others (Mullins and McKnight 1981). Different black spruce logs from a stem are traditionally allocated to sawmills for lumber conversion regardless of log position in the tree. As the forest industry is shifting toward value-added end uses, there is an increasing need to develop an optimal log allocation strategy, where logs are assigned to the best uses to maximize the value of the resource and to ensure the quality of the final products. This approach has been emphasized and practiced in New Zealand and Sweden in recent years, and has been beneficial (Corson 1997; Albert et al. 2002). For example, high grade logs are used for decorative purpose or construction, while low-quality logs such as tops and core wood are being used for fiber-based products such as pulp and medium density fiberboard (MDF). To develop an optimal log allocation strategy, it is necessary to understand the product quality and value recovery in relation to log position in the tree.
Logs coming from different positions in a tree possess different wood and fiber characteristics; butt log wood usually has on average higher density and the fiber cell wall is thicker, whereas, wood density is lower and the fiber cell wall is thinner at the top of a tree since top logs contain a higher proportion of juvenile wood (Panshin and DeZeeuw 1980; Zobel and Sprague 1998). So far, limited studies (Zhang and Chauret 2001; Zhang et al. 2002) have examined lumber quality as well as value recovery in relation to log position in black spruce. In the literature little study has investigated the properties of MDF panels from different log positions in this species. Some studies focused on the properties of composite panels made from different types of wood, such as juvenile and mature wood. For example, Wasniewski (1989) examined the strength properties, dimensional stability, and durability of Douglas fir (Pseudotsuga menziesii var. menziesii [Mirb.] Franco) flakeboard panels from five age classes, which were pith-7, 8-14, 15-21, 22-28, 29-bark, and found that panel properties decreased with age class increase. A study by Pugel et al. (1989; 1990) indicates that the properties of flakeboard, particleboard, and fiberboard panels made from four different types of loblolly pine ( Pinus taeda L.) juvenile wood, namely fast-grown trees, the inner core of older trees, branches, and tops, were comparable to or better than those of mature wood composite panels, and that the differences in panel properties between juvenile and mature wood panels were at a minimum in fiberboard. Additionally, MDF panels made from black spruce ( Picea mariana (Mill.) BSP.) juvenile wood performed better in modulus of rupture (MOR), internal bond (IB), and water absorption than the panels fabricated from mature wood; but linear expansion (LE) of the panels from juvenile wood was significantly higher than that of mature wood panels (Shi et al. 2005). Therefore, in most cases, composite panels made from juvenile wood are superior to the panels made from mature wood. This is in accordance with the fact that most panel properties improve with increasing the compaction ratio, and that adding high density wood to the panels made from low density species decreases board quality, for instance, by adding white birch strands to aspen strands in OSB panels.
In this study, MDF panels were manufactured from black spruce top, mid, and butt logs, and the strength properties and dimensional stability of the three types of panels were examined. The objective of this study was to determine the effect of log position in the tree on strength properties and dimensional stability of MDF panels made from black spruce wood.
Logs coming from different positions in a tree possess different wood and fiber characteristics; butt log wood usually has on average higher density and the fiber cell wall is thicker, whereas, wood density is lower and the fiber cell wall is thinner at the top of a tree since top logs contain a higher proportion of juvenile wood (Panshin and DeZeeuw 1980; Zobel and Sprague 1998). So far, limited studies (Zhang and Chauret 2001; Zhang et al. 2002) have examined lumber quality as well as value recovery in relation to log position in black spruce. In the literature little study has investigated the properties of MDF panels from different log positions in this species. Some studies focused on the properties of composite panels made from different types of wood, such as juvenile and mature wood. For example, Wasniewski (1989) examined the strength properties, dimensional stability, and durability of Douglas fir (Pseudotsuga menziesii var. menziesii [Mirb.] Franco) flakeboard panels from five age classes, which were pith-7, 8-14, 15-21, 22-28, 29-bark, and found that panel properties decreased with age class increase. A study by Pugel et al. (1989; 1990) indicates that the properties of flakeboard, particleboard, and fiberboard panels made from four different types of loblolly pine ( Pinus taeda L.) juvenile wood, namely fast-grown trees, the inner core of older trees, branches, and tops, were comparable to or better than those of mature wood composite panels, and that the differences in panel properties between juvenile and mature wood panels were at a minimum in fiberboard. Additionally, MDF panels made from black spruce ( Picea mariana (Mill.) BSP.) juvenile wood performed better in modulus of rupture (MOR), internal bond (IB), and water absorption than the panels fabricated from mature wood; but linear expansion (LE) of the panels from juvenile wood was significantly higher than that of mature wood panels (Shi et al. 2005). Therefore, in most cases, composite panels made from juvenile wood are superior to the panels made from mature wood. This is in accordance with the fact that most panel properties improve with increasing the compaction ratio, and that adding high density wood to the panels made from low density species decreases board quality, for instance, by adding white birch strands to aspen strands in OSB panels.
In this study, MDF panels were manufactured from black spruce top, mid, and butt logs, and the strength properties and dimensional stability of the three types of panels were examined. The objective of this study was to determine the effect of log position in the tree on strength properties and dimensional stability of MDF panels made from black spruce wood.
Ten mature black spruce ( Picea mariana (Mill.) BSP.) trees were harvested in July 1999 from a second growth natural stand at Réserve Ashuapmushuan located 400 kilometers north of Québec City, Canada. The trees were cut from the bottom into 8-foot long butt, mid and top logs. The logs collected from the ten trees were first debarked by hand, and then chipped using a portable chipper. The wood chips were fed into a pressurized disc refiner for conversion of fibers at the Forintek pilot plant in Québec City without resin or wax injection. The moisture content of the wood chips in the pre-steaming bin for top, mid, and butt log chips were 36.5 %, 27.1 %, and 22.4 % respectively. The refining speed, cooker pressure and retention time for the three groups were 2500 rpm, 7.5 bar, and 3 min. The specific refining energy for top, mid, and butt logs were 266 KWh/t, 95 KWh/t, and 160 KWh/t respectively. The clearance between the two plates for top, mid, and butt logs were adjusted to be 0.007 mm, 0.001mm, and 0.009 mm. A transition time of 30 minutes was allowed to avoid the fibers that were produced from two different groups from mixing, and the fibers obtained during the transition time were discarded.
Fibers were dried in a laboratory-scale dryer until the moisture content was about 2-3 %. Before blended with wax and resin, the fibers were passed through a hammer mill to fully separate the fibers from each other, which would result in a more uniform resin distribution. Because high resin level is always associated with good composite panel properties (Maloney 1993), a relatively low resin content was applied to the dry fibers so as to avoid hiding the effect of different log position in the tree on panel properties. Therefore, 10 % (by weight of dry fiber) Borden 302 urea-formaldehyde (UF) resin (65 % solid content, without catalyst added) and 0.5 % wax were slowly sprayed onto the fibers using a laboratory-scale blade blender. After resin and wax blending, the fibers were passed through the hammer mill once again to disperse fiber balls. All panels were made using the same pressing program at the pilot plant of Forintek Canada Corp. with a target density of 740 kg/m3. Panel size was 610×610×12 mm. Mats were hand formed in a forming frame without fiber orientation. Three panels were replicated for each group. The moisture content of top, mid, and butt log furnishes after resin and wax blending were 14 %, 13.3 %, and 11 %. The platens were slowly closed over 160 s and the mats were kept under pressure for another 160 s. Thereafter, the platens were gradually relieved of pressure over a 40 s time span. The temperature of the platens was set at 135 oC. A relative long closing time and low platen temperature were adopted here since we attempted to achieve homogeneous panel vertical density profiles that would reduce the differences in panel properties due to different density gradient along the thickness.
All panels were placed in a conditioning chamber at 22 oC and 65 % relative humidity (RH) for four weeks until the moisture content of the panels reached equilibrium before cutting into testing specimens. Three static bending specimens with a dimension of 338 by 75 mm were cut from each panel, nine in total for each group; the panels were cut down the size 50 by 50 mm for IB testing, a total of thirty specimens were made for each group; two specimens with a dimension of 305 by 76 mm were cut from each panel for linear expansion (LE) measurement, producing six for each group; TS and water absorption were determined from the same two specimens (152×152 mm) that were cut from each panel, thus six specimens were taken for each group. The specimens for LE measurement were conditioned consequently at a RH cycle of 50 % to 80 % until they reached equilibrium moisture content. LE was calculated by the percentage change in length based on the length measured at 50 % RH. TS and water absorption were obtained respectively by calculating the increase in thickness and weight following a 24 h water submersion divided by the thickness and weight measured at 22 oC and 65 % RH equilibrium condition. The procedures and methods described in ASTM D 1037-99 (2001) and ANSI A 208.2-2002 (2002) were followed for MOE, MOR, IB, LE, TS, and water absorption testing. About 1.5 mm was sanded from the two surfaces of the IB specimens before they were glued with the blocks. The density profiles were measured by X-ray densitometry from the IB specimens before the specimens were sanded. For the analysis of covariance (ANCOVA) purpose, Specimen densities were determined prior to MOR, MOE, IB, LE, TS, and water absorption testing.
Data were analyzed using the statistical software Statistical Analysis System (SAS) (1990). The analysis of variance (ANOVA) and ANCOVA were both performed to examine the differences in panel strength properties and dimensional stability for the three groups. The previous studies indicate that panel density can have a significant influence on composite panel properties (Maloney 1993; Olson 1996; Peter et al. 2002; Shi et al. 2005). ANCOVA was conducted in this study to adjust the mean values of MDF panel properties that were partly attributed to panel density (covariate). Before ANCOVA was performed, the assumptions must be tested and ensure that they are met. If the assumption that the regression slopes are equal is not met, consequently, Johnson-Neyman technique (Huitema 1980) was employed to calculate the limits of significance and nonsignificance regions on the covariate. Additionally, simple linear regression analysis was used to develop the equations describing the relationships of MOR and MOE to panel density. The dummy variables that represent different log positions in the tree were created. Using the dummy variables, the three regression lines describing the relationships of either MOR or MOE and density were compared for coincidence.
Fiber size distribution was determined using Bauer McNett classifier and Tappi T 233 cm-95 (1995) methods were followed. The results listed in Table 3-1 indicate that the percentage of fibers distributed in each mesh size range differed slightly for the group 1, 2, and 3. In this case, an assumption was made that the fibers with similar size distribution were converted from top, mid, and butt log wood chips during refining process.
The average density of top log wood chips was the smallest and increased with lowering tree height of the samples (Table 3-2). This is due to a higher proportion of juvenile wood contained in top logs. The panel density for the three groups was different although the panels were targeted at the same density level. Generally, the density of the specimens taken from the same panel may remain different because in-plane density variation exists in almost all cases. For this reason, ANCOVA was utilized to remove that part of effect on panel properties, ultimately, to compare the panel properties on the same density basis.
Compaction ratio refers to the ratio of panel density to wood density (Hsu 1997). The compaction ratio of the panels made from top logs was the largest and the smallest was due to the panels made from butt logs (Table 3-2). Wood density is one of the most important wood characteristics which determines final composite panel properties (Maloney 1993; Hsu 1997). Low-density wood produces higher compaction ratio if panel density is maintained at the same level. It was found that good mechanical properties of composite panels were associated with high compaction ratios (Pugel et al. 1989; Shupe et al. 1999). Compaction ratio was also found to be positively linearly related to panel MOR and MOE in hardboards that were made from fourteen hardwood species (Woodson 1976). MDF panels made of black spruce juvenile wood were superior to the panels made from mature wood, which was believed due to the higher compaction ratio of the former (Shi et al. 2005).
The average vertical density profiles of the panels made from top, mid, and butt logs are shown in Figure 3-1. These density profiles do not represent those obtained from the panel mills because a different pressing program was adopted in this study. The low surface density can result in reduction in bending properties. However, since the density profiles were nearly homogeneous for the three types of panels, this allowed comparison of panel properties between the three groups. The slight difference in vertical density profile between the three groups can be explained by a number of reasons, for example, different density of top, mid, and butt logs, different compaction ratio, in-plane variation in mat density, mat moisture content variations, etc.
Compaction ratio refers to the ratio of panel density to wood density (Hsu 1997). The compaction ratio of the panels made from top logs was the largest and the smallest was due to the panels made from butt logs (Table 3-2). Wood density is one of the most important wood characteristics which determines final composite panel properties (Maloney 1993; Hsu 1997). Low-density wood produces higher compaction ratio if panel density is maintained at the same level. It was found that good mechanical properties of composite panels were associated with high compaction ratios (Pugel et al. 1989; Shupe et al. 1999). Compaction ratio was also found to be positively linearly related to panel MOR and MOE in hardboards that were made from fourteen hardwood species (Woodson 1976). MDF panels made of black spruce juvenile wood were superior to the panels made from mature wood, which was believed due to the higher compaction ratio of the former (Shi et al. 2005).
The average vertical density profiles of the panels made from top, mid, and butt logs are shown in Figure 3-1. These density profiles do not represent those obtained from the panel mills because a different pressing program was adopted in this study. The low surface density can result in reduction in bending properties. However, since the density profiles were nearly homogeneous for the three types of panels, this allowed comparison of panel properties between the three groups. The slight difference in vertical density profile between the three groups can be explained by a number of reasons, for example, different density of top, mid, and butt logs, different compaction ratio, in-plane variation in mat density, mat moisture content variations, etc.
According to ANSI (2002) standard, MOR of the panels made from top, mid, and butt logs met the requirement of Grade 120. To compare the MOR mean values of the three types of panels, ANOVA was first carried out, and the results show a significant difference in the three groups (Table 3-2). The unadjusted means were computed using the ANOVA method, which were not adjusted based on the linear relationship between the dependent variable (MOR) and the covariate (panel density). The unadjusted mean MOR of MDF panels made from top and mid logs were significantly higher than that of panels made from butt logs, whereas, the difference between those panels made from top and mid logs was not significant. The linear relationship between MOR and panel density was significant (Table 3-3), meaning that panel density accounted for a proportion of MOR. As seen in Table 3-3, the precision of the analysis was highly improved by performing the ANCOVA. The adjusted means were calculated by the ANCOVA method, which the linear relationship between the dependent variable (MOR) and the covariate (panel density) was incorporated in the computation. In SAS, this can be done with the syntax ‘LSMEANS MOR/PDIFF’. ‘LSMEANS’ option requests adjusted means (least squares means) for the variable MOR, while ‘PDIFF’ option gives significance probabilities for all pairwise comparisons of these adjusted means. Table 3-3 also indicates significant interactions within the three groups, which means that the regression slopes were not equal. A consequent homogeneity test was performed and the results indicate that the regression slopes of group 1 and 3 were heterogeneous, but the slopes of group 1 and 2 and group 2 and 3 were homogeneous (Table 3-4). For the pairs of group1 and 2, and group 2 and 3, means were adjusted based on the same slopes. Thus, it is shown that MOR of panels made from top logs was significantly higher than MOR of panels made from mid and butt logs, and MOR of panels made from mid logs was significantly higher than that of panels from butt logs (Table 3-2). For the comparison of groups 1 and 3 that had heterogeneous slopes, Johnson-Neyman technique (Huitema 1980) was performed to compute the limits of significance and nonsignificance regions on panel density. The results are shown in the following.
X L1 = 519 kg/m3
X L3 = 710 kg/m3
Therefore, Johnson-Neyman suggests that there is no significant difference in MOR between panels made from top and butt logs if panel density is in the region of 519-710 kg/m3. However, if panel density is lower than 519 kg/m3, MOR of panels made from butt logs is significantly higher than that of panels made from top logs. When panel density is higher than 710 kg/m3, panels made from top logs are significantly superior to the panels made of butt log wood in MOR.
Through simple linear regression analysis, the equations describing the relationships between MOR and panel density for the three groups were obtained as shown in Figure 3-2. Two dummy variables were created to identify the three log positions in the tree. Dummy variables z1 and z2 were defined as follows.
z1 = 1if the panels were made from top logs
0otherwise
z2 = 1if the panels were made from mid logs
0otherwise
The three different log positions can be identified by combining z1 and z2.
z1=1 z2=0: panels made from top logs
z1=0 z2=1: panels made from mid logs
z1=0 z2=0: panels made from butt logs
A model for the relationship between MOR and panel density is given by:
Y = b0 + b1x + b2z1 + b3z2 + b4xz1 + b5xz2 + ε
where Y represents panel MOR;
x is panel density, z1, z2 are dummy variables representing top and mid log positions;
ε is the error term.
The models for the relationships between MOR and panel density for the three types of panels can be given by:
for the panels made from top logs: Y = (b0 + b2) + (b1 + b4)x + ε
for the panels made from mid logs: Y = (b0 + b3)+ (b1 + b5)x + ε
for the panels made from butt logs: Y = b0 + b1x + ε
The hypothesis of the coincidence of the three regression lines is that both the slopes and intercepts are equal for the three types of panels, which can be written as follows:
H0: b2 = b3 = b4 = b5 = 0
The hypothesis of the coincidence can be tested by checking the significance of the terms z1, z2, xz1, and xz2. The results indicate that the terms z1, z2, xz1, and xz2 were significant as shown in Table 3-5, which means that the hypothesis of the coincidence was rejected. Tests for parallelism and equality of the intercept of each two of the three regression lines were then followed. Table 3-6 indicates that the relationships between MOR and panel density for the three types of panels must be described using three individual equations, which are presented in Figure 3-2.
X L1 = 519 kg/m3
X L3 = 710 kg/m3
Therefore, Johnson-Neyman suggests that there is no significant difference in MOR between panels made from top and butt logs if panel density is in the region of 519-710 kg/m3. However, if panel density is lower than 519 kg/m3, MOR of panels made from butt logs is significantly higher than that of panels made from top logs. When panel density is higher than 710 kg/m3, panels made from top logs are significantly superior to the panels made of butt log wood in MOR.
Through simple linear regression analysis, the equations describing the relationships between MOR and panel density for the three groups were obtained as shown in Figure 3-2. Two dummy variables were created to identify the three log positions in the tree. Dummy variables z1 and z2 were defined as follows.
z1 = 1if the panels were made from top logs
0otherwise
z2 = 1if the panels were made from mid logs
0otherwise
The three different log positions can be identified by combining z1 and z2.
z1=1 z2=0: panels made from top logs
z1=0 z2=1: panels made from mid logs
z1=0 z2=0: panels made from butt logs
A model for the relationship between MOR and panel density is given by:
Y = b0 + b1x + b2z1 + b3z2 + b4xz1 + b5xz2 + ε
where Y represents panel MOR;
x is panel density, z1, z2 are dummy variables representing top and mid log positions;
ε is the error term.
The models for the relationships between MOR and panel density for the three types of panels can be given by:
for the panels made from top logs: Y = (b0 + b2) + (b1 + b4)x + ε
for the panels made from mid logs: Y = (b0 + b3)+ (b1 + b5)x + ε
for the panels made from butt logs: Y = b0 + b1x + ε
The hypothesis of the coincidence of the three regression lines is that both the slopes and intercepts are equal for the three types of panels, which can be written as follows:
H0: b2 = b3 = b4 = b5 = 0
The hypothesis of the coincidence can be tested by checking the significance of the terms z1, z2, xz1, and xz2. The results indicate that the terms z1, z2, xz1, and xz2 were significant as shown in Table 3-5, which means that the hypothesis of the coincidence was rejected. Tests for parallelism and equality of the intercept of each two of the three regression lines were then followed. Table 3-6 indicates that the relationships between MOR and panel density for the three types of panels must be described using three individual equations, which are presented in Figure 3-2.
MOE of panels made from top, mid, and butt logs met the requirement of ANSI grade 120. ANOVA was applied to compare MOE of panels made from top, mid, and butt logs. The results indicate a significant difference in MOE between panels made from top and mid logs, and MOE of both types of panels were significantly higher than that of panels made from butt logs (Table 3-2). The test of within group linear relationship between MOE and panel density indicates a significant interaction, which means that panel density had an effect on MOE; however, the interactions between the three groups were not significant (Table 3-3). Thus, the adjusted mean MOE of the three groups can be computed based on the equal regression slope. Table 3-2 shows that MOE of panels made from top and mid logs was significantly higher than that of panels from butt logs; but the difference in MOE between panels made from top and mid logs was not significant.
The regression functions in Figure 3-3 were developed separately using three datasets measured from top, mid, and butt log panels. Using the same method mentioned previously, the coincidence of the three regression lines for the relationship between MOE and panel density was tested as well (Table 3-5). The hypothesis of the coincidence was accepted meaning that a single regression line can be fit to the relationships for top, mid, and butt logs. Thus, a model to describe the relationship between MOE and panel density is given by:
Y = -5296.8224 + 9.0574 x (R2 = 0.7906)
where: Y is MOE of the panels made from top, mid, and butt logs;
x is panel density;
for the three types of panels.
The regression functions in Figure 3-3 were developed separately using three datasets measured from top, mid, and butt log panels. Using the same method mentioned previously, the coincidence of the three regression lines for the relationship between MOE and panel density was tested as well (Table 3-5). The hypothesis of the coincidence was accepted meaning that a single regression line can be fit to the relationships for top, mid, and butt logs. Thus, a model to describe the relationship between MOE and panel density is given by:
Y = -5296.8224 + 9.0574 x (R2 = 0.7906)
where: Y is MOE of the panels made from top, mid, and butt logs;
x is panel density;
for the three types of panels.
Linear relationships between IB and panel density for the three groups were not significant, so the assumption for ANCOVA was not met. Thus, ANOVA was performed to test the group mean differences. The lack of linear relation between IB and panel density might be due to a relatively narrow density range drawn into the analysis. ANOVA shows that IB of panels made from top and mid logs were not significantly different, but both were significantly higher than that of panels made from butt logs (Table 3-2).
The better performance of top log panels in static bending properties and internal bond strength indicate a significant effect of fiber origin on MDF panels. Since top logs contain higher proportion of juvenile wood when compared with butt logs, more furnish is needed to make the panels from juvenile wood at the same panel density level. In this case, panels can be bonded better using the thin-walled juvenile wood fibers as raw material, which leads to a greater mechanical load resistance as discussed in Chapter II. The observation here tends to the same conclusion as it is in Chapter II.
The better performance of top log panels in static bending properties and internal bond strength indicate a significant effect of fiber origin on MDF panels. Since top logs contain higher proportion of juvenile wood when compared with butt logs, more furnish is needed to make the panels from juvenile wood at the same panel density level. In this case, panels can be bonded better using the thin-walled juvenile wood fibers as raw material, which leads to a greater mechanical load resistance as discussed in Chapter II. The observation here tends to the same conclusion as it is in Chapter II.
The dimensional stability of MDF panels was evaluated by LE, TS, and water absorption. Statistical analysis was performed to examine the relationships between LE, TS, and water absorption and panel density. The results indicate that the linear relationships were not significant for the variables LE, TS, and water absorption. Therefore, the comparisons of these group means were based on the unadjusted values. As shown in Table 3-2, LE of panels made from top, mid, and butt logs were not significantly different. TS of panels made from top logs was significantly lower than that of panels made from mid and butt logs according to ANOVA, and TS of panels made from butt logs was significantly higher than that of panels made from mid logs (Table 3-2). For water absorption, panels made from butt logs absorbed more water after a 24 h submersion than the panels from top and mid logs, but there was no significant difference between panels from top and mid logs (Table 3-2).
The bending properties of MDF panels made from black spruce top, mid, and butt logs were affected by low panel surface density. The low surface density caused a reduction in panel MOR and MOE. Nevertheless, the bending properties of the panels made from top, mid, and butt logs were still comparable since the differences in vertical density profile were fairly slight. MOR and MOE of MDF panels made from top, mid, and butt logs met the requirement of ANSI grade 120.
MOE and IB of MDF panels made from top and mid logs were significantly higher than those of panels made from butt logs; however, there were no significant differences in MOE and IB between panels made from top and mid logs. Water absorption of MDF panels made from top and mid logs was significantly lower than that of panels from butt logs; again, the difference in water absorption between panels made from top and mid logs was not significant.
Following a 24 h water-soaking period, the panels made from butt logs swelled the most and the panels made from top logs swelled least. There was no significant difference in LE among those panels made from top log, mid log and butt logs.
MDF panels made from top logs performed the best in MOR, while MOR of the panels made from butt logs was the smallest. Johnson-Neyman suggests that there is no significant difference in MOR between panels made from top and butt logs when panel density is between 519 kg/m3and 710 kg/m3. It also suggests that MOR of panels made from top logs is significantly higher than that of panels from butt logs if panel is heavier than 710 kg/m3. If panel density is lower than 519 kg/m3, the reverse is true.
Panel density had a considerable effect on panel MOR and MOE. The relationships between MOR, MOE and panel density were linear. But the linear relationships between IB, LE, TS, or water absorption and panel density were not significant.
It was concluded that most of the properties of MDF panels made from top logs were superior to those of panels made from butt logs. This is due to the high proportion of juvenile wood contained in top logs. Fibers generated from low-density juvenile wood can be easily packed and lead to high performance. This could be helpful for sorting black spruce logs into various uses. Butt logs can be sorted to lumber manufacturers because of the higher density, strength properties and product recovery compared to mid and top logs. Top logs can be allocated to composite mills for fiberboard manufacturing since the performance of the products is better. Thus, values of black spruce forest resource are added and the performance of the products is maximized.
MOE and IB of MDF panels made from top and mid logs were significantly higher than those of panels made from butt logs; however, there were no significant differences in MOE and IB between panels made from top and mid logs. Water absorption of MDF panels made from top and mid logs was significantly lower than that of panels from butt logs; again, the difference in water absorption between panels made from top and mid logs was not significant.
Following a 24 h water-soaking period, the panels made from butt logs swelled the most and the panels made from top logs swelled least. There was no significant difference in LE among those panels made from top log, mid log and butt logs.
MDF panels made from top logs performed the best in MOR, while MOR of the panels made from butt logs was the smallest. Johnson-Neyman suggests that there is no significant difference in MOR between panels made from top and butt logs when panel density is between 519 kg/m3and 710 kg/m3. It also suggests that MOR of panels made from top logs is significantly higher than that of panels from butt logs if panel is heavier than 710 kg/m3. If panel density is lower than 519 kg/m3, the reverse is true.
Panel density had a considerable effect on panel MOR and MOE. The relationships between MOR, MOE and panel density were linear. But the linear relationships between IB, LE, TS, or water absorption and panel density were not significant.
It was concluded that most of the properties of MDF panels made from top logs were superior to those of panels made from butt logs. This is due to the high proportion of juvenile wood contained in top logs. Fibers generated from low-density juvenile wood can be easily packed and lead to high performance. This could be helpful for sorting black spruce logs into various uses. Butt logs can be sorted to lumber manufacturers because of the higher density, strength properties and product recovery compared to mid and top logs. Top logs can be allocated to composite mills for fiberboard manufacturing since the performance of the products is better. Thus, values of black spruce forest resource are added and the performance of the products is maximized.
Albert, D.J., Clark, T.A., Dickson, R.L. and Walker J.C.F. Using acoustics to sort radiata pine pulp logs according to fiber characteristics and paper properties . International Forest Review. 4(1): 12-19; 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.
American National Standards Institute (ANSI). Medium Density Fiberboard (MDF) for Interior Application . ANSI A 208.2-2002. Composite Panel Association, Gaithersburg. MD; 2002.
Corson, S.R. Tree and fiber selection for optimal TMP quality . Proceeding of the International Mechanical Pulping Conference, Stockholm. 1997.
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. Québec City, Canada. I-7/10; 1997.
Huitema, B. E. The Analysis of Covariance and Alternatives . Jone Wiley & Sons, New York. 1980.
Maloney, T. M. Modern Particleboard and Dry-Process Fiberboard Manufacturing . Miller Freeman Publications. San Francisco, CA. 1993.
Mullins, E. J. and McKnight, T. S. Canadian Woods: Their Properties and Uses . Univ. of Toronto, Toronto, Ontario. 1981.
Olson, B. D. Developing wood composites using small diameter timber resources from dense, stagnant stands . M.S. thesis. Washington State Univ., Pullman, WA. 1996.
Panshin, A. J. and DeZeeuw, C. Textbook of Wood Technology . 4th ED. McGraw-Hill Book Co. New York, NY. 1980.
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.
Pugel, A. D., Price, E. W. and Hse, C. Y. Composites from southern pine juvenile wood. Part 1. Panel fabrication and initial properties . Forest Prod. J. 40(1): 29-33; 1989.
Pugel, A. D., Price, E. W. and Hse, C. Y. Composites from southern pine juvenile wood . Part 2. Durability and dimensional stability . Forest Prod. J. 40(3): 57-61; 1990.
SAS Institute, Inc. SAS/STAT User’s guide . Cary, N.C. 1990.
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; 2005.
Shupe, T. F., Hsu, C. Y., Choong, E. T. and Groom, L. H. Effect of silvicultural practice and wood type on loblolly pine particleboard and medium density fibreboard properties . Holzforschung. 53(2): 215-222; 1999.
Tappi Standard. Fiber length of pulp by classification . T 233cm-95. 1995.
Wasniewski, J. L. Evaluation of juvenile wood and its effect on Douglas fir structural composite panels . Proceedings of the twenty-third Washington State University International Particleboard-Composite Material Symposium. Pullman, WA. 161-175; 1989.
Woodson, G. E. Properties of medium-density fiberboard related to hardwood specific gravity . Proc. 10th Washington State University particleboard symposium. Pullman, WA. pp175-192; 1976.
Zhang, S. Y., Chauret, G., Ren, H. Q. and Desjardins, R. Impact of initial spacing on plantation black spruce lumber grade yield, bending properties, and MSR yield . Wood Fiber Sci. 34(3): 460-475; 2002.
Zhang, S.Y. and G. Chauret. Impact of initial spacing on tree and wood characteristics, product quality and value recovery in black spruce (Picea mariana) . Canadian forest Service Report. No. 35, Forintek Canada Corp., Sainte-Foy, Québec; 2001.
Zobel, B. J. and Sprague J. R. Juvenile Wood in Forest Trees . Springer-Verlag; 1998.
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.
American National Standards Institute (ANSI). Medium Density Fiberboard (MDF) for Interior Application . ANSI A 208.2-2002. Composite Panel Association, Gaithersburg. MD; 2002.
Corson, S.R. Tree and fiber selection for optimal TMP quality . Proceeding of the International Mechanical Pulping Conference, Stockholm. 1997.
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. Québec City, Canada. I-7/10; 1997.
Huitema, B. E. The Analysis of Covariance and Alternatives . Jone Wiley & Sons, New York. 1980.
Maloney, T. M. Modern Particleboard and Dry-Process Fiberboard Manufacturing . Miller Freeman Publications. San Francisco, CA. 1993.
Mullins, E. J. and McKnight, T. S. Canadian Woods: Their Properties and Uses . Univ. of Toronto, Toronto, Ontario. 1981.
Olson, B. D. Developing wood composites using small diameter timber resources from dense, stagnant stands . M.S. thesis. Washington State Univ., Pullman, WA. 1996.
Panshin, A. J. and DeZeeuw, C. Textbook of Wood Technology . 4th ED. McGraw-Hill Book Co. New York, NY. 1980.
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.
Pugel, A. D., Price, E. W. and Hse, C. Y. Composites from southern pine juvenile wood. Part 1. Panel fabrication and initial properties . Forest Prod. J. 40(1): 29-33; 1989.
Pugel, A. D., Price, E. W. and Hse, C. Y. Composites from southern pine juvenile wood . Part 2. Durability and dimensional stability . Forest Prod. J. 40(3): 57-61; 1990.
SAS Institute, Inc. SAS/STAT User’s guide . Cary, N.C. 1990.
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; 2005.
Shupe, T. F., Hsu, C. Y., Choong, E. T. and Groom, L. H. Effect of silvicultural practice and wood type on loblolly pine particleboard and medium density fibreboard properties . Holzforschung. 53(2): 215-222; 1999.
Tappi Standard. Fiber length of pulp by classification . T 233cm-95. 1995.
Wasniewski, J. L. Evaluation of juvenile wood and its effect on Douglas fir structural composite panels . Proceedings of the twenty-third Washington State University International Particleboard-Composite Material Symposium. Pullman, WA. 161-175; 1989.
Woodson, G. E. Properties of medium-density fiberboard related to hardwood specific gravity . Proc. 10th Washington State University particleboard symposium. Pullman, WA. pp175-192; 1976.
Zhang, S. Y., Chauret, G., Ren, H. Q. and Desjardins, R. Impact of initial spacing on plantation black spruce lumber grade yield, bending properties, and MSR yield . Wood Fiber Sci. 34(3): 460-475; 2002.
Zhang, S.Y. and G. Chauret. Impact of initial spacing on tree and wood characteristics, product quality and value recovery in black spruce (Picea mariana) . Canadian forest Service Report. No. 35, Forintek Canada Corp., Sainte-Foy, Québec; 2001.
Zobel, B. J. and Sprague J. R. Juvenile Wood in Forest Trees . Springer-Verlag; 1998.
Table 3-1 Size distribution of the fibers produced from black spruce top, mid, and butt logs.
Table 3-2 Black spruce MDF panel moisture content, average density, compaction ratio, and mean modulus of rupture (MOR), modulus of elasticity (MOE), internal bond (IB), linear expansion (LE), thickness swell (TS), and water absorption.
Table 3-3 Comparison of analysis of variance (ANOVA) and analysis of covariance (ANCOVA) for panel modulus of rupture (MOR) and modulus of elasticity (MOE).
Table 3-4 Homogeneity of regression test and statistics summary of panel modulus of rupture (MOR) where interactions occurred in the analysis of covariance (ANCOVA).
Table 3-5 Test for coincidence of the relationships between MOR/MOE and panel density.
Table 3-6 Tests for parallelism and equality of intercept of the linear relationships between MOR/MOE and panel density. *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, respectively. Numbers in column ‘<0.017’ were the percentages of fibers passed through the screen with mesh size of 0.017 mm2.
Methods described in Tappi T 233 cm-95 were followed.
*Note: Unadjusted means were computed without consideration of panel density effect on panel properties. Adjusted means were computed using the regression equations of MOR, MOE, and TS to panel density by means of LSMEANS option in SAS, and compared pairwisely using the PDIFF option in SAS.
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 all pairwise comparisons ( P =0.05).
Methods for panel density and moisture content determination were in accordance with ASTM D 1037-99.
Compaction ratios were calculated based on panel equilibrated density at 22 oC and 65 % RH and density of wood chips.
The adjusted mean MOR with ‘①’ was calculated using the slope of group 1 and group 2. The adjusted mean MOR with ‘②’ was calculated using the slope of group 2 and group 3.
S is standard deviation.
*Note: SS is sum of square. DF is degree of freedom. F values were significant at 0.05 of probability.
Table 3-2 Black spruce MDF panel moisture content, average density, compaction ratio, and mean modulus of rupture (MOR), modulus of elasticity (MOE), internal bond (IB), linear expansion (LE), thickness swell (TS), and water absorption.
Table 3-3 Comparison of analysis of variance (ANOVA) and analysis of covariance (ANCOVA) for panel modulus of rupture (MOR) and modulus of elasticity (MOE).
Table 3-4 Homogeneity of regression test and statistics summary of panel modulus of rupture (MOR) where interactions occurred in the analysis of covariance (ANCOVA).
Table 3-5 Test for coincidence of the relationships between MOR/MOE and panel density.
Table 3-6 Tests for parallelism and equality of intercept of the linear relationships between MOR/MOE and panel density. *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, respectively. Numbers in column ‘<0.017’ were the percentages of fibers passed through the screen with mesh size of 0.017 mm2.
Methods described in Tappi T 233 cm-95 were followed.
Table 3-2 Black spruce MDF panel moisture content, average density, compaction ratio, and mean modulus of rupture (MOR), modulus of elasticity (MOE), internal bond (IB), linear expansion (LE), thickness swell (TS), and water absorption.
 |
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 all pairwise comparisons ( P =0.05).
Methods for panel density and moisture content determination were in accordance with ASTM D 1037-99.
Compaction ratios were calculated based on panel equilibrated density at 22 oC and 65 % RH and density of wood chips.
The adjusted mean MOR with ‘①’ was calculated using the slope of group 1 and group 2. The adjusted mean MOR with ‘②’ was calculated using the slope of group 2 and group 3.
S is standard deviation.
Table 3-3 Comparison of analysis of variance (ANOVA) and analysis of covariance (ANCOVA) for panel modulus of rupture (MOR) and modulus of elasticity (MOE).
 |
Table 3-4 Homogeneity of regression test and statistics summary of panel modulus of rupture (MOR) where interactions occurred in the analysis of covariance (ANCOVA).
 |
Table 3-6 Tests for parallelism and equality of the intercept of the linear relationships between MOR/MOE and panel density.
 |
Figure 3-1 Average density profiles of MDF panels made from black spruce top, mid, and butt logs.
Figure 3-2 Relationship between modulus of rupture (MOR) and panel density.
Figure 3-3 Relationship between modulus of elasticity (MOE) and panel density. Note: y1, y2, y3 correspond to modulus of rupture (MOR) of MDF panels made from top, mid, and butt logs; x refers to panel density. Note: y1, y2, y3 correspond to modulus of elasticity (MOE) of MDF panels made from top, mid, and butt logs; x refers to panel density.
Figure 3-2 Relationship between modulus of rupture (MOR) and panel density.
Figure 3-3 Relationship between modulus of elasticity (MOE) and panel density. Note: y1, y2, y3 correspond to modulus of rupture (MOR) of MDF panels made from top, mid, and butt logs; x refers to panel density. Note: y1, y2, y3 correspond to modulus of elasticity (MOE) of MDF panels made from top, mid, and butt logs; x refers to panel density.
For further details log on website :
http://archimede.bibl.ulaval.ca/archimede/fichiers/23592/ch01.html
No comments:
Post a Comment