Protection of wood using oxy-aluminum compounds Fabiano A. Ximenes Philip D. Evans Abstract The dimensional stability of wood and its resistance to decay and fire can be improved by depositing inorganic materials in the wood matrix and forming wood-inorganic composites. In this paper, four novel wood-inorganic (aluminum) composites were developed. The composites are referred to according to the oxy-aluminum compound deposited in the wood: sodium aluminate, aluminum hydroxide, magnesium aluminate, and aluminum borate. The aim of the research was to assess whether the dimensional stability and resistance of the oxy-aluminum composites to microbial attack and fire was comparable to that of wood treated with a hydrophobic chromated copper arsenate preservative (CCA-wax). Fourier transform infrared spectroscopy of the oxy-aluminum composites suggested some modification of the wood matrix, especially for the sodium aluminate or aluminum hydroxide treatments. Magnesium aluminate-treated samples had the highest weight gain (28.1%) and anti-swelling efficiency (105.4%). The sodium aluminate, aluminum hydroxide, and aluminum borate treatments had lower anti-swelling efficiencies of 70.8, 68.9, and 55.0 percent, respectively. The anti-swelling efficiency of the magnesium aluminate-treated wood decreased significantly to 58.6 percent after a severe leaching procedure. The anti-swelling efficiencies of the other treatments, however, particularly the aluminum borate treatment, were less affected by leaching, and the dimensional stability of the composites after leaching exceeded that of unleached CCA-wax-treated wood. The weight losses of the wood-inorganic composites after 30 weeks of exposure in soil varied from 2.4 percent (aluminum hydroxide) to 4.9 percent (magnesium aluminate) compared to 36.6 percent for untreated controls. None of the treatments, however, were as effective as CCA-wax at preventing decay. Thermo- gravimetric analysis and differential thermal analysis indicated that the oxy-aluminum composites had enhanced resistance to thermal degradation, in contrast to CCA-treated wood which showed no such resistance. Further research on a larger scale is needed to optimize and further evaluate the oxy-aluminum composites including their mechanical properties and effects on fixings. There is a strong need to develop wood protection sys- tems that are both effective and of low toxicity (Preston 2000). Numerous studies have demonstrated that chemical modifica- tion can increase the dimensional stability, as well as the de- cay and fire resistance of wood (Rowell 1983), but commer- cial wood protection systems that rely on chemical modifica- tion of wood have been slow to evolve in North America due in part to their high cost and the sophisticated plants required to modify the wood (Preston 2000). Accordingly, there is a need to develop simpler and less costly methods of modifying wood to improve its dimensional stability and resistance to decay and fire. One possible route to achieving this is to treat the wood sequentially with two different solutions of inor- ganic compounds that react together to generate insoluble compounds within the cell walls (Saka 1998). In principle such treatments could be performed using simple technology as past studies have used double diffusion treatments to treat wood successfully (Markstrom et al. 1970, Mayer et al. 1995). Aluminum is one of the most abundant elements in the soil, making up approximately 7.1 percent of the solid matter in an average soil sample and it has low mammalian toxicity (Sposito 1989). Only a few studies, however, have tested alu- minum-based compounds in wood protection systems. Fu- runo and coworkers combined aluminum sulfate with sodium silicate to deposit aluminum silicate in wood (Furuno et al. 1991, Furuno 1992, Furuno and Imamura 1998). The resulting wood-inorganic composite showed high weight gain and had enhanced decay and fire resistance (Furuno et al. 1991), but it The authors are, respectively, Research Officer, Forests R&D Di- vision, New South Wales Dept. of Primary Industries, Beecroft, NSW, Australia (fabianox@sf.nsw.gov.au) and Professor, Centre for Advanced Wood Processing, Univ. of British Columbia, Van- couver, Canada (phil.evans@ubc.ca). F. Ximenes acknowledges a scholarship provided by The Australian National Univ., and both au- thors acknowledge the assistance of Denes Bogsanyi and Dr. John Broomhead of The Australian National Univ. during the initial stages of the work. This paper was received for publication in June 2005. Article No. 10075. ©Forest Products Society 2006. Forest Prod. J. 56(11/12):116-122. 116 NOVEMBER/DECEMBER 2006