This work included two separate studies; the common theme in each was the use of mode-I fracture testing to evaluate wood adhesion.

In the first study, mode-I fracture testing was used to compare the thermal stability of polyurethane (PUR) and resorcinol-formaldehyde (RF) wood adhesives. Bonded specimens for both adhesives were subjected to prolonged thermal exposure, and fracture testing was subsequently conducted after re-equilibration to standard test conditions. It was found that both PUR and RF suffered a significant fracture energy loss after heat treatment, and that RF was more thermally stable than PUR, as expected. However, both adhesives suffered significant thermal degradation, and fracture testing did not distinguish the RF system as being clearly superior to PUR. Dynamic mechanical analysis (DMA) was also used to analyze and compare the thermal softening of PUR and RF in terms of the decline in storage modulus. DMA results indicated that PUR specimens suffered greater stiffness loss due to simple thermal softening. Because fracture testing indicated that both adhesives suffered significant degradation, the DMA results suggested that the generally superior fire resistance of RF adhesives is born from greater high temperature stiffness; whereas the more compliant PUR suffers greater immediate softening during thermal exposure. In other words, both systems suffer from thermal degradation, but the more highly cross-linked RF system suffers less thermal softening and therefore maintains a greater load carrying capacity during fire exposure.

In the second study, mode-I fracture testing was used to test the effects of wood surface thermal deactivation (surface energy reduction) on the adhesion between southern pine wood (Pinus spp.) and polyethylene (PE). Pine specimens were progressively surface deactivated by 185°C heat treatments for periods of 5, 15, and 60 minutes. Control and deactivated pine laminae were subsequently hotpressed/bonded using PE film as the adhesive. Mode-I fracture testing was conducted under the assumption of linear elasticity, however load/displacement test curves suffered from a severe degree of nonlinearity believed to be caused by PE bridging behind the advancing crack tip. Instead of applying a nonlinear data analysis, a standard linear elastic analysis was conducted and deemed acceptable for comparative purposes within this study. Under dry conditions (unweathered specimens), 5 and 15 minute thermal treatments resulted in progressively worse adhesion (lower fracture energies) when compared to control surfaces; but the 60 minute heat treatment improved adhesion relative to 5 and 15 minute treatments, and showed a trend of improving adhesion as surface deactivation became more extreme. Simulated-weather resistance was also studied and it was determined that the highest degree of surface deactivation slightly improved weather durability in comparison to control surfaces. Overall, the findings here were similar to those in a previously published work- thermal deactivation of wood surfaces shows promise as a method to improve adhesion between wood and nonpolar polyolefins.