Fluorescence microscopy for detecting incipient decay and estimating residual strength of wood

Graduation date: 1982The objectives of this study were to 1) develop a relatively quick staining procedure that would detect incipient decay by color differentiation under fluorescence, 2) quantitatively assess the amount of decay present with the use of an image analyzer, and 3) re/ate loss in strength at various decay levels to staining for the presence of decay. The fluorochrome stain acridine orange was used to stain nondecayed and decayed wood sections from end-grain wafers and small beams of Southern yellow pine (Pinus spp.) and Douglas-fir [Pseudotsuga menziesii (Mirb.) Franco]. A .02% solution of acridine orange pH 6.0 was best for Southern yellow pine; a pH 8.0 was best for Douglas-fir. Staining the sections for two hours and washing them for 24 hours provided the best color differentiation between decayed and non-decayed areas. In wood with little or no weight loss (less than 3%), the springwood cells appeared bright green under fluorescence with traces of yellow-orange. As weight loss increased, more yellow and orange became evident. The latewood and ray cells fluoresced a yellow-orange most of the time regardless of percent weight loss. The computer image analyzer detected the green and orange as different gray levels, the green registering a lighter shade than the orange. However, due to inherent anatomical features of wood and the dulling of fluorescence when a slide has been exposed to the illumination source for an extended period of time, this method is not practical. Modulus of elasticity (MOE), modulus of rupture (MOR) and fiber stress at the proportional limit (FSPL) decreased significantly with increased weight loss. MOR was most affected sustaining a strength loss greater than 50% with a 3% weight loss in pine. Linear regression analyses of MOE and MOR vs. specific gravity indicated a relationship between these two variables and also a significant difference in the strength property means of the nondecayed and decayed samples. Percent strength loss (MOE and MOR) vs. percent weight loss was also regressed. The coefficient of determination (r²) for these regressions were low, with less than 50% of the variability being accounted for by percent weight loss. Alkali solubility of pieces from end-matched beams was correlated with percent strength loss but less so than the correlations between strength and weight loss. Radial sections from the broken beams stained the bright green typical of non-decayed wood possibly due to irregularities in the brown rot decay pattern, but more likely because the areas of the beam sampled may not have contained decay. Cross sections near the failure have responded better