Using linear regression analysis on several wood species as a tool for determining wood properties influencing decay resistance variations in wood

• Main Author: Jessica Mary Emily, anak Jem
• Format: Thesis
• Language: English
• Published: Universiti Malaysia Sarawak, (UNIMAS) 2014
• Subjects: GE Environmental Sciences
• Online Access: http://ir.unimas.my/id/eprint/9021/1/Jessica.pdf; http://ir.unimas.my/id/eprint/9021/

Several commercial timber species contributed in our economic development by supplying the world market mainly with logs as well as sawn timber and plywood. Since timber is a popular and useful material, it is important that enthusiasts and professionals be able to distinguish the wood quality of one species from another. Relationship between various wood properties namely decay resistance, methanol extractives content, wood density, hot water solubility, cold water solubility, pH of water extractives, fiber morphology as well as several other microscopic aspects of the combination of 36 timber species of Malaysian and temperate wood species were investigated. This research aims to determine those wood properties that perceivably govern decay resistance variations of numerous timbers species in order to predict decay resistance profile of commercial and potential timber species by using linear regression analysis. Laboratory test procedure according to ASTM D 2017 is chosen in this research by utilizing particular species of white rot (P. sanguineus), brown rot (G. trabeum) and soft rot decay test (C. globosum for unsterile soil burial test). A number of wood properties studied (extractives contents and anatomical features) are important factors which contribute to the decay resistance of wood species as differences in natural decay resistance of timber could be accounted for by changes in the extractives and fiber anatomy aspects. There were high percentages of mass loss caused by three decay fungi of P. sanguineus (43.9%), G. trabeum (38.2%) and C. globosum (50.2%) to rubberwood, Hevea brasiliensis a non- durable species. Conversely, belian (Eusiderozylon zwageri) a durable timber species recorded negligible decay heartwood with mean mass loss of only 0.7% when exposed to P. viii sanguineus. Mass loss by G. trabeum was the lowest when tested to senumpul (Hydnocarpus sp.) and selangan batu kuning (Shorea flava) both 0.4% mass loss whereas bindang (Agathis borneensis) yielded the lowest percentage weight loss of 0.7% due to the attack of C. globosum. The timber species were then classified according to the 4 natural durability classes of ASTM D 2017 and EN 350 where timber species were categorized in Class 1 (resistant) to Class 4 (resistant to non resistant). The variation of wood properties were summarized as follows: (i) methanol extractives content (0.7% to 36.5%), (ii) hot water soluble content (3.3% to 18.6%), (iii) cold water soluble content (0.4% to 9.9 %), (iv) density (0.40 g/cm3 and 0.97 g/cm3), (v) pH 3.3 to 6.6 for cold water extractables and (vi) pH 3.2 to 6.0 for hot water extractables. As for the fiber anatomical and dimension, the results are as follow: (i) fiber length (367.9μm- 4229.3μm), (ii) fiber diameter (13.0μm- 47.1μm), (iii) fiber lumen diameter (2.3 μm- 31.1μm), (vi) fiber wall thickness (0.7 μm- 5.9μm) and (v) Runkel ratio (0.1- 2.3). Moreover, the vessel diameters ranged from 24.2μm to 375.6μm for tangential and 16.1μm to 279.1μm for radial. Alternatively, the vessel density was in the range of 1.2 per mm2 to 6.9 per mm2 for overall wood species whereas the proportion of the rays was in the range of 6.7% to 34.6%. The relationship between decay resistance and wood properties were studied by correlation and regression analyses where the linear regression equation for estimating the natural durability has been established and perhaps can be used as a tool to predict the decay resistance and the factors associated with decay resistance. These findings would provide understanding and the comparative variation of the decay resistance among timber species will augment the existing information on wood quality classification of 36 timber species studied.