Effects of Thermomechanical Refining on Douglas fir Wood

Medium density fiberboard (MDF) production uses thermomechanically refined fiber processed under shear with high pressure steam. The industry evaluates fiber quality with visual and tactile inspection, emphasizing fiber dimensions, morphology, and bulk density. Considering wood reactivity, the hypothesis is that a variety of chemical and physical changes must occur that are not apparent in visual/tactile inspection. An industry/university cooperation, this work studies effects of refining energy (adjusted by refiner-plate gap) on fiber: size, porosity, surface area, surface and bulk chemistry, fiber crystallinity and rheology, and fiber interaction with amino resins. The intention is to reveal novel aspects of fiber quality that might impact MDF properties or process control efficiency, specific to a single industrial facility.

In cooperation with a North American MDF Douglas fir plant, two refining energies were used to produce resin and additive-free fibers. Refining reduced fiber dimensions and increased bulk density, more so at the highest energy. Thermoporosimetry showed increases in sub-micron scale porosity, greatest at the highest energy. Mercury intrusion porosimetry (MIP) revealed porosity changes on a higher dimensional scale. Brunauer-Emmett-Teller gas adsorption and MIP showed that refining increased specific surface area, more so at the highest energy. Inverse gas chromatography showed that the lowest refining energy produced surfaces dominated by lignin and/or extractives. The highest energy produced more fiber damage, revealing higher energy active sites. A novel rheological method was devised to study fiber compaction and densification; it did not distinguish fiber types, but valuable aspects of mechano-sorption and densification were observed.

Refining caused substantial polysaccharide degradation, and other degradative effects that sometimes correlated with higher refining energy. Lignin acidolysis was detected using nitrobenzene oxidation, conductometric titration of free phenols, and formaldehyde determination. Formaldehyde was generated via the C2 lignin acidolysis pathway, but C3 cleavage was the dominant lignin reaction. Observations suggested that in-line formaldehyde monitoring might be useful for process control during biomass processing. According to rheological and thermogravimetric analysis, lignin acidolysis was not accompanied by repolymerization and crosslinking. Lignin repolymerization must have been prevented by the reaction of benzyl cations with non-lignin nucleophiles. This raises consideration of additives that compete for lignin benzyl cations, perhaps to promote lignin crosslinking and/or augment the lignin network with structures that impart useful properties.

Fiber/amino resin interactions were studied with differential scanning calorimetry (DSC) and X-ray diffraction (XRD). All fiber types, refined and unrefined, caused only a slight increase in melamine-urea-formaldehyde (MUF) resin reactivity. Generally, all fiber types decreased the enthalpy of MUF cure, suggesting fiber absorption of small reactive species. But DSC did not reveal any dependency on fiber refining energy. According to XRD, all fiber types reduced crystallinity in cured MUF, more so with refined fiber, but independent of refining energy. The crystallinity in cured urea-formaldehyde resin was studied with one fiber type (highest refining energy); it caused a crystallinity decrease that was cure temperature dependent. This suggests that resin crystallinity could vary through the thickness of an MDF panel.