Veneer and Plywood

The conversion of A. mangium into veneer and plywood is feasible with no specific processing requirements needed. In conventional rotary peeling studies by Wong et al. (1988), Chai (1989), and Salim (1992), the peeling process was rated as easy and the green veneers produced were tight, smooth, and of acceptable quality. They dried easily following the normal commercial species schedule to the desired moisture content. The green veneer recovery however, was found to range from 35-459?-. This fairly low return could be due to the facts that the logs were young (7-15 years old) and small in diameter with fluted boles. Wong et al. (1988) and Chai (1989) also found numerous knot marks, and some sapwood veneers marred with borer holes.

At present, not many plywood plants have the facilities to peel small-diameter logs. In Malaysia, for example, only six or seven companies have such equipment. But as the appropriate processing technologies become more available. the authors believe that better recovery will be obtained. Improved tree form and knot reduction through silvicultural practices will further improve the recovery from peeling.

On the production of decorative veneers from A. mangium , the slicing of green flitches was also rated as easy and the veneers were smooth even around

Table 113. Knife tests of urea and phenolic-bonded plywood from Acacia mangium (UF Board: pressed for 3 min at 100°C; PF Board: pressed for 6 min at 109°C).

















DRY= dry condition: INT = interior condition; BR = boil-resistant condition; MR = moisture resistant condition; WBP = weather-boil-proof. Source. Wong et al. (1988)

the solid knots (Wong et al. 1988). The thin veneers (0.26-0.28 mm) dried well to the desired moisture content.

Table 11.5 shows the results of the knife tests of plywood bonded with standard amount of urea formaldehyde (UF) and phenol formaldehyde (PF) resin adhesives (Wong et al. 1988). The evaluation, carried out in accordance with the British Standards, indicated that the UF plywood was poor when tested under the moisture-resistant (MR) condition. Under the less severe interior (INT) condition, the performance was just below the acceptable grade. The knife tests for those panels bonded with PF resulted in acceptable quality under the weather-boil-proof (WBP) grade and very good results were obtained when tested under the boil-resistant (BR) procedures. On the whole, these trials showed that the PF panels were of acceptable quality but the UF-bonded plywood were unsatisfactory. In order to improve the MR on INT grade panels, the UF resin adhesives need to be reformulated to suit the substrates, or the woods need to be physically reevaluated to identify a compatible environment. In a study on the wettability of A. mangium veneers, Chai (1989) indicated that PF resins would be more favorable with A. mangium. Similar results were obtained by Mohd. Hamami et al. (1991). who worked on wood lamination, and Salim (1992), who worked on laminated veneer lumber (LVL)(Table 11.6).

Wong et al. (1988) rated as attractive trial decorative or fancy plywood panels made using A. mangium sliced veneers as the faces. No difficulty was encountered in assembling the panels. Although the panels had more knots than sliced Tectona grandis and Palaquium xanthochymum decorative plywood, the plywood is nonetheless suitable for panelling, cabinets, and furniture manufacture. Sliced A. mangium veneers should also be suitable for veneer-wrapping works if compatible adhesives are employed.

Table 11.6. The average glue line shear strength and wood failure of different species combinations glued with phenol-resorcinol formaldehyde (PRF) and urea formaldehyde (UF) under dry condition 1.

Species combination

Shear strength (MPa) PRF UF

Reduction in shear (%) based upong PRF strength

A. mangium -A. mangium

Paraserianthes falcataria - P. falcataria

A. mangium -P. falcataria





28 b 84a

85 1

73 1

1 Each reading is an average of 15 replicates. Means in the same column followed by the same letter are not significantly different at 0.05 level of confidence using Duncan's New Multiple Range TcsiSource. Mohd. Hamami et al. (1991)

+1 0

Post a comment