Resin modification using a bio-based resin derived from cashew nut shell liquid (CNSL) was undertaken on the sapwood of three species of tropical hardwood timber: Obeche (Triplochiton scleroxylon), Gmelina (Gmelina arborea) and Alstonia (Alstonia scholaris). The species had been selected on the basis of low density to enable uptake of resin, however level of uptake varied, in relation to presence of tyloses within the vessels. Dynamic mechanical analysis (DMA) was used to observe the secondary relaxations of the timbers and the glass transition (Tg) of the resin. A thermal scan from -150 to +150°C was used in three point bending, with multiple frequencies (0.1 Hz, 1.0 Hz and 10 Hz). The glass transition temperature of cured CNSL resin was observed between 60 and 75°C, with the value increasing with frequency used for testing.

Each species was impregnated with three concentrations of CNSL resin solution, giving three different weight percent gain (WPG) levels. Within each species, the Tg of resin was seen to increase as the WPG of resin increased. However, between species this relationship did not hold. Alstonia scholaris, which had the highest WPG values (17.4, 31.4 and 42.5%) recorded only intermediate values of Tg, whereas Gmelina, which had lowest WPG values (4.6, 4.8 and 10.8%), showed the highest range of Tg values. It is proposed that differences in wood anatomy and cell structure influence the distribution of resin, and size of droplets, plugs or layers of resin formed within the modified wood. In samples with higher WPG (Alstonia and Obeche), the activation energy of the resin Tg event was 738 and 767 kJ/mol respectively.