This Is AuburnElectronic Theses and Dissertations

Defatted Soy Flour Substitution in Phenol Formaldehyde and Methylene Diphenyl Diisocyanate Wood Adhesives and their Curing Kinetic Behavior




Hand, William

Type of Degree

PhD Dissertation


Chemical Engineering


Composite wood products are manufactured utilizing reconstituted cellulosic material formed with thermosetting adhesives, primarily adhesives made from non-renewable materials. Natural wood adhesives, or bio-adhesives, had been used until the mid-20th century. Eventually bio-adhesives were replaced with synthetic adhesives because of superior qualities such as water resistance, durability, and eventually cost. Recent published work has shown the incorporation of bio-adhesives in composite wood products that required chemical modification and functionalization of the bio-adhesives. Defatted soy flour (SF) has shown the highest potential as a bio-adhesive due to its protein content and low cost. This work encompassed an investigation of the partial substitution of synthetic resins with SF for use in composite wood products. Other soy products investigated in this work included soy protein isolate (SPI) and denatured, defatted soy flour (soy-D). The synthetic adhesives studied were phenol formaldehyde (liquid and powder) and polymeric methylene diphenyl diisocyanate (pMDI) and are both used in the production of oriented strand board. The overall objective of this work was to advance the understanding of interactions between soy flour and synthetic wood adhesives in the use of composite wood products. An evaluation of the degree to which SF could be substituted into either phenol formaldehyde (PF) or pMDI in the formation of strand board was studied in the first phase of this work. The amount of SF substitution possible without compromising performance of the formed board was determined. The mechanism of adhesion enhanced through SF interaction with the PF or pMDI was studied in the second phase of this work. Curing kinetic data of neat PF, neat pMDI, and soy products substituted into PF or pMDI were analyzed. There were two possible primary mechanisms of adhesion influenced by the substitution of SF: SF influenced the cohesiveness in the bulk by chemically crosslinking with the PF or pMDI, or the SF affected the adhesiveness to the wood substrate by altering the wood/resin interface. This work was driven by the hypothesis that SF hydrolyzed during the pressing process to chemically crosslink with the synthetic resin in the bulk adhesive. This work differed from prior published works by focusing on the incorporation of SF during the board forming process. Other published works focused on SF modification or pretreatment which added cost. Some works incorporated the modified SF in the synthetic resin synthesis which decreased the pot life of the overall adhesive. The incorporation of SF during the board forming process required minimal equipment upgrades and used existing materials (PF and pMDI). Cost was the chief motivation for this work as SF was about half the cost of PF and less than half the cost of pMDI. Non-renewable resource elimination through substitution with a bio-adhesive was a secondary motivator. The board produced from the SF substituted adhesive also required equal or improved mechanical property data of the board produced with the synthetic adhesive. Mechanical properties measured were flexural strength and modulus, internal bond, thickness swell, and water absorption. Higher viscosity due to SF substitution in liquid PF inhibited penetration of the wood substrate and likely decreased resin spread. Less adhesive penetration benefited board strength with lesser amounts of SF substitution due to a decrease of overpenetration of the adhesive into the wood substrate. Larger amounts of SF substitution resulted in adhesive failure due to under-penetration. A decrease in the resin spread likely affected the wood/resin contact area which resulted in adhesive failure in higher density boards. SF substitution of 10-14% in liquid PF was possible without compromising performance. Resin spread was not an issue with SF substituted powder PF. SF substitution of 21-30% in powder PF was possible without compromising performance. There were two possible interactions affecting the SF substituted pMDI boards: urea formation due to SF substitution in pMDI positively affected wet properties of the formed board, or higher viscosity due to SF substitution in pMDI decreased overpenetration of pMDI into the substrate and contributed to the bulk. SF substitution of up to 30% in pMDI was possible without compromising performance. Thermogravimetric analysis (TGA) of the PF/soy product mixtures showed no new chemical structures formed in the degradation profile compared with TGA data of neat PF. Activation energies for curing reactions of PF/soy product mixtures showed no considerable difference to the curing reaction activation energies of neat PF. Fourier transform-infrared (FT-IR) spectral analysis of neat PF and PF/soy product samples subjected to differing heat treatments showed no significant difference of concentrations of methylene and ether bridge formation signifying no change in curing reaction with soy product interaction with PF. All curing kinetic analyses of PF and PF/soy product samples refuted the original hypothesis of the primary interaction being affected with soy product substitution in PF was a chemical crosslinking one. The higher viscosity of the soy flour substituted PF affected the resin spread and penetration during the blending and pressing process of strand board formation, therefore the primary mechanism of adhesion affected by the substitution of soy products in PF was adhesion affecting the wood/resin interface. TGA of the pMDI/soy product mixtures showed a new degradation profile of soft segments not observed in the neat pMDI TGA data indicating new chemical structures formed as a result of soy products interacting with pMDI. The formation of carbodiimide was identified through TGA for neat pMDI and pMDI/soy product mixtures. FT-IR spectral and activation energy analyses confirmed the formation of carbodiimide as two separate reaction pathways: polymerization reaction for neat pMDI and part of the urethane degradation reaction in the pMDI/soy product mixtures. Urethane and urea formation in the pMDI/soy product mixture was discovered as the primary polymerization reaction through FT-IR spectral analysis. All curing kinetic analyses of pMDI and pMDI/soy product samples supported the original hypothesis of the primary interaction being affected with soy product substitution in pMDI was a chemical crosslinking one.