0692-A2
Ong Lay Lee and Paridah Md. Tahir 1
The objective of this study was to develop structural-grade smooth surface oriented strand board (OSB) from rubberwood (heavea brasiliensis). The effects of fine layer content on strength and dimensional stability of the boards were studied. The surfaces (top and bottom) of the OSB were made up of fine particles (<0.1 mm) and the core was made up of 0.5 - 2.5mm X 75 mm sized strands. The proportion of fines content were 10%, 20% and 30% of total weight of the particles required for each board. PF resin was used at 10% and 15% w/w to bond the strands. All calculations were based on the ovendry weight of the wood particles. Three-layered OSBs were manufactured as control. The physical and mechanical properties of the boards were evaluated in accordance to the Japanese Industrial Standard - JIS A 5908-1994. The results show that the strength (MOR) and stiffness (MOE) of the OSB panels produced in this study were significantly (p < 0.1) affected by the amount of fines used; i.e. both the MOR and MOE decreased with an increase in fines content. Nonetheless, the values exceeded the minimum requirements for type 24 -10 board of the JIS. The study also shows that thickness swelling of the board was significantly improved with an increase in fines content. No significant difference (p > 0.1) was found in the internal bond strength among the levels of fines content used. Within the limits of this study, fines content of 20% and PF content of 15% resin are considered suitable for producing smooth surface OSB with acceptable strength properties.
Nowadays, wood-based panels have to satisfy various requirements in different market application to maintain its status quo. These requirements vary, not only for the product end uses but also the condition where it is going to be used. In order to compete with other panel products, OSB must have a better surface, much better strength properties and stability. Some OSB products, such as siding or concrete form works, require improved panel properties viz. surface roughness, thickness swelling at edges, edge checking, and structural integrity of the panels after being subjected to humidity and temperature variations encountered in normal use. Panels used for concrete shuttering are often laminated to give the concrete a smooth surface and to improve its resistance to severe exposure to moisture. Plywood is thus preferred due to its smoother surface, which eases the laminating process, and to the boards' lower sensitivity of edge swelling. It can be said that there is a huge potential market for applications requiring smooth surfaces OSB for furniture and residential applications, or for low thickness swelling board that is close to plywood for concrete shuttering. To enter this market, OSB needs to significantly refine its surfaces and improve swelling figures.
The rough surface of ordinary three-layered OSB has become a barrier to the marketability of this product. To obtain improved panel qualities, it will be necessary for OSB to improve surface smoothness i.e., by incorporating an additional layer of very fine materials on the board's surface; increasing the phenolic resin level in the fine layer, and overlaying one or both panel surfaces with resin-impregnated paper. This fine layer fills the voids between the strands and provides smoothness of OSB. The smooth surface also facilitates the laminating of OSB board in single-step pressing. Nevertheless, the presence of fine layers on the surface is believed to significantly reduce the strength and stiffness of the panel.
The extent of the effects of fine layer content on the panel strength, stiffness, and dimensional stability is thus needs to be established so that allowance can be made in design and infield inspection. So far little information about how much influence in strength properties and dimensional stability would be by distributing additional layer of fine particles onto ordinary three-layered OSB fabricated with flake length of 75mm. Hence, this study was conducted to assess the feasibility of producing smooth surface OSB of acceptable strength and dimensional stability. The objectives of the study were to evaluate the effects of different fine layer contents on strength and dimensional stability of smooth surface OSB and to determine the optimum fines content suitable for producing smooth surface OSB.
Rubberwood flakes of dimension 10mm X 75mm X 0.3 mm were generated using a disc flaker. After screening, flakes of sizes 5mm to 25mm were used to produce three-layered OSB panels. About 200 flakes were randomly selected and their dimension measured. The rest of the flakes were dried to a moisture content (MC) of 2 to 3 % using the laboratory oven at 80 o C. The MC of the flakes was determined.
The flakes were blended with 5% w/w PF resin in a rotary drum mixer. The fines were mixed with PF resin separately in a static mixer with resin dosage of 10% and 15% based on the oven dry weight of the fines. The amount of fines were at 10%, 20% and 30% of the total oven dry weight of the particles required. Three boards of 12 mm x 380 mm x 380 mm in size were manufactured for each fines content and resin level. The targeted board density was 700 kg/m3. A mechanical orienter with slits of 20 mm apart was employed in the mat forming. The boards were formed manually by laying the flakes in the former with each layer perpendicular to each other. The mat was cold pressed and the fines were distributed onto both surfaces of the pressed mat and were pressed again. The mat was then hot pressed at 175o C for 7 1/2 minutes and conditioned for several days at ambient temperature. The conditioned boards were trimmed and cut for static bending, internal bonding and thickness swelling tests in accordance to JIS A5908 - 1994.
The length of the flakes varied from 40.0 mm to 80.0 mm with an average of 70.2 mm. The width and thickness of the flakes range from 5.0 mm to 45.0 mm and 0.19 mm to 0.83 mm respectively. The average width was 16.2 mm and thickness being 0.42 mm. The distribution of flake dimensions is shown in Table 1. Generally the flakes are relatively none slender with an average slenderness ratio of 164.
Table 1: Average Dimensions of Rubberwood Flakes
Dimensions |
Length (mm) |
Width (mm) |
Thickness (mm) |
Slenderness ratio |
Mean |
70.2 |
16.2 |
0.417 |
164 |
Minimum |
40.0 |
5.0 |
0.185 |
61 |
Maximum |
80.0 |
45.0 |
0.831 |
400 |
Standard deviation |
7.77 |
7.58 |
0.14 |
57 |
Relatively high variation in the board densities which ranged from 712 kg/m3 to 818 kg/m3. The density mean was 769 kg/m3, about 10% lower than the targeted board density. The average MC of the board was 8.4%.
Effects of Fines Content on the Strength and Stiffness of Five-Layered OSB
The highest dry MOR (45 N/mm2) was obtained from panels having 10% fines content. The weakest panels were produced by using 30% fines content, the MOR being 27.7 N/mm2. Similar trend was observed for wet MOR (Figure 1). When subjected to analysis of variance, the MOR (dry and wet) were significantly (p < 0.01) affected by the amount of fines used. The strength of the OSB decreased following an increase in the amount of fines on board's surface. Many researchers have found that bending properties tend to increase directly with flake length (Post, 1958, Rackwitz, 1963, Ong, 1981). This may be attributed to the presence of large strand on the panel surfaces that have the tendency to retard moisture movement and, therefore, limits the heat transfer from outer to inner layers during pressing. Slower movement of moisture and heat towards the center of the mat enhances vertical density profile i.e., higher face layer density, hence increases the bending strength (Avramidis and Smith 1989).
The MOE (dry and wet) followed similar trend as the MOR. From Figure 1, the highest dry MOE among the five-layered OSB was obtained by boards having 10% fines content which were 5677 N/mm2 (15% RL) and 4883 N/mm2 (10% RL), the lowest being 3262N/mm2 produced by board of 30% fine content and 15% resin level. Increasing the amount of fines on the board's surfaces resulted in significant (p<0.01) reductions in both dry and wet MOE. Nonetheless, the MOE (dry and wet) values exceeded the requirements for type 24 -10 board as specified in JIS A 5908 - 1994 except for boards with 30% fines content.
Effects of Fines Content on the Internal Bond Strength and Thickness Swelling of Five-Layered OSB
The IB strength of the OSBs ranged from 0.16 N/mm2 to 0.34 N/mm2 (Figure 2). Three-layered OSB (control) had relatively higher IB than does five-layered smooth surface OSB. The analysis of variance using completely randomized design failed to show significant differences (P > 0.05) amongst the levels of fines content on the IB values. Particle alignment that has apparent effect on MOR and MOE has no effect on IB. This is in agreement with studies conducted by McNatt et al; 1992 and Lee et al.1996 who found showed that IB was not affected by the alignment of particle. All the smooth surface panels did not achieve the minimum requirement of JIS type 24 - 10-board for IB. Only the control three-layered boards and the smooth surface five-layered (having 10% fines content and bonded with 10% PF resin) boards had marginally passed this standard. One plausible cause for this is the slenderness ratio of the flakes used in this study was very much higher (160 -200) compared to the optimum slenderness ratio specified for good IB which is in between 25 to 50 (Canadido 1988). The resin coverage may not uniformly distributed to cover all surface area of the flakes that provide intimate bonding. Hence, this decreases the IB strength.
The range of TS for the boards was from 15.1% to 22.7% (Figure 2). The analysis of variance shows that the fine content levels have a very significant effect (p < 0.01) on the thickness stability of the board. Without the fine layer, the control three-layered OSB exhibited higher TS (22.7%) i.e., less thickness stability, when compare to smooth surface OSB (FC 10%, 20% and 30% of TS 17.0%, 16.2% and 15.1% respectively). Board comprises 10% and 20% fines content did not differ much in their TS. Board having 30% FC however, was the most stable (i.e. thickness). The fines content on the board surface could have acted as a protective layer preventing water absorption by the board. Another explanation being surface irregularities of the ordinary three-layered OSB which could cause surface degradation through higher water absorption. Biblis (1990) stated that smooth surface and paper overlay improve retention of board properties. Fine layers have smaller interparticle voids or passageways, and thus water absorption was slowed in a 24-hour test. Hence, the higher the fine layer thickness, the smoother the surface due to its gap-filling ability. These led to increasing thickness stability of the board. All the boards met the requirements (maximum 25.0%) specified for thickness swelling in JIS A 5908-1994.
Effect of Resin Content in Fine Layer on the Strength and Stiffness of Five-Layered OSB
Increasing resin level does not give significant effect (p > 0.1) on MOR (wet and dry). The dry MOR reductions was 2.1% and wet MOR reductions was 0.4% by the increase of resin level in the fine surface from 10% to 15%. Increase in resin level from 10% to 15% in the fine content layer only leads to slight increase in MOR (wet and dry) from 36.8 N/mm2 to 37.3 N/mm2 and 12.9 N/mm2 to 13.0 N/mm2. This result is in agreement with that of Manalo and Pulido (1983), who found that increasing resin content level at the face layers did not greatly influence the MOR and MOE of the boards. All boards treated in wet condition shows a decrease in wet MOR to that of dry condition. The reductions were shown in Figure 3. This is probably due to the reduction of interparticle bonding, as the particles begin to separate after absorbing water due to bond deterioration (swelling).
Resin levels do not shown significant effect (p > 0.1) on MOE (wet and dry) after subjected to statistical analysis. Increase in resin level from 10% to 15% in the fine content layer only leads to slight increase in MOE (dry) from 4346N/mm2 to 4358N/mm2. All boards treated in wet condition shows a reduction in wet MOR to that of dry condition.
Effect of Resin Content in Fine Layer on the Internal Bond Strength and Thickness Swelling of Five-Layered OSB
The IB strength of OSB were also not significantly (p > 0.05) affected by resin level in the surface fine layer used in this study i.e. 0.23 N/mm2 at 10% surface resin level and 0.20 N/mm2 at 15% surface resin level. As mentioned earlier, IB measures the board bonding efficiency especially at the mid-plane thickness of the board. Thus, increase resin level in surface does not contribute to IB strength. As stated by Stegmann and Durst (1964), increasing resin content throughout the board appears to increase IB strength in a linear fashion instead of in the surface layer.
Statistical comparisons of means showed no significant effect (p > 0.1) of resin level on TS. The TS for both resin levels, 10% and 15% were 16.6% and 15.5% respectively. Increased thickness stability and water resistance of the smooth surface OSB are noted as the addition of fine content on the board's surface increased with high quantity of resin. As stated by Xu and Winistorfer (1995b), TS at the surface regions of the board was higher than that in the center. Thus, the OSB surface with high resin content improved the stability at the surface region that slowed down the water absorption.
The surface smoothness was found to be improved with increase in resin level in the fine layer i.e. from 10% to 15%. This is in agreement with Moslemi (1974), who stated that a smoother surface appearance would result as the resin content of the surface layer is increased. In this study, due to the manner of which the flakes are being distributed, i.e. manually, the fine content of 10% was found not sufficient to obtain uniform distribution of particles to cover the voids on the board surface. Hence, there are some irregularities on the board surface. Boards having fine content of 20% to 30% of its total weight on their surfaces could produced acceptably smooth surface without markedly affecting its' strength.
Fines content was found to significantly affect the bending strength (MOR and MOE) and thickness stability except internal bonding strength. Increase of fine content decrease the retention of MOR and MOE in wet condition but, enhance the thickness stability. On the contrary, all MOR, MOE, IB and TS of produced smooth surfaced OSB were not significantly (p > 0.1) affected by resin level in the surface fine layer. The surface smoothness was found to be improved with increase in resin level in the fine layer i.e. from 10% to 15%.
All MOR (dry and wet) exceeded the standard requirement set for board type 24 - 10 in JIS A 5908 -1994. However, for MOE, only boards of 30% fines content failed to achieve the standard. The boards attained the thickness swelling specified in JIS standard for board type 24 - 10. However, the internal bonding strength has failed to achieve the requirement set by JIS.
The manufacture of smooth surface OSB appears to be technically feasible. Using 20% content and PF resin content of 15% on the surface produce acceptably good smooth surface OSB without the strength being markedly reduced.
Avramidis, S., and L. A. Smith. 1989. The effect of resin content and face-to-core ratio on some properties of oriented strand board. Holzforchung, 43 (2), pp. 131 - 133.
Biblis, E. J. 1990. Performance of Southern OSB Overlaid with Resin-Impregnated Paper. Forest Products Journal, 40 (4), pp. 55 - 62.
Canadido, L. S., F. Saito, and S. Suzuki. 1987. Effect of Particle Shape on the Orthotropic Properties of Oriented Strand Board. Mokuzai Gakkashi 34 (1), pp. 21 - 27.
Lee, A. W. C., X. Bai, and P. N. Peralta. 1996. Physical and Mechanical Properties of Strandboard Made from Moso Bamboo. Forest Product Journal 46 (11/12), pp.84 -88.
Manalo, N. N., and O. R. Pulido. 1983. Various Levels of Surface Resin Content: Their Effect on Properties of Phenol-Formaldehyde Bonded Kaatoan Bangkal [Anthocephalus chinensis (Lam.) Rich. Ex Walp.] Particleboards. FPRDI Journal 12 (3,4), pp. 35 - 40.
Mc Nett, J. D., L. Bach, and R. W. Wellwood. 1992. Contribution of Flake Alignment to Performance of Strandboard. Forest Products Journal 42 (3), pp. 45 - 50.
Moslemi, A. A. 1974. Particleboard Vol. 1 & 2. Southern Illinois Press, Carbondale and Edwardsville, Ill.
Ong, C. L. 1981. The Influence of Wood Density and Flake Dimensions on Particleboard Properties of Five Hardwood Species. The Malaysian Forester, 44 (4), pp. 508 - 515.
Post, P. W. 1958. Effect of Particle Geometry and Resin Content on Properties of Particleboard. Forest Products Journal, 8 (10), pp. 317 - 322.
Rackwitz, G. 1963. Influence of Chip Dimensions on Some Properties of Wood Particleboard. Holz als Roh und Werkstoff, 21 (6), pp. 380 - 387.
Stegmann, G., J. Durst. 1964. Beech pressboard. S. I. U. Cartographic Laboratory.
Xu, W. and P. M. Winistorfer. 1995a. A Procedure to Determine Thickness Swell Distribution in Wood Composite Panels. Wood and Fiber Science, 27 (2), pp. 119 -125.
Xu, W. and P. M. Winistorfer. 1995b. Layer Thickness Swell and Layer Internal Bond of Medium Density Fiberboard and Oriented Strand Board. Forest Products Journal, 45 (10), pp. 67 - 71.
1 Faculty of Forestry, Universiti Putra Malaysia
43400 UPM Serdang, Selangor, MALAYSIA