«CURE KINETICS OF WOOD PHENOL-FORMALDEHYDE SYSTEMS By JINWU WANG A dissertation submitted in partial fulfillment of the requirements for the degree of ...»
E: activation energy; n: reaction order; R2: coefficient of determination.
CONCLUSION DSC analysis showed that the curing behavior of the PF resin did not change significantly when the wood content was below 20%. When the wood content was over 35%, wood catalyzed one reaction and moved it to the lower temperature. The cure behavior differed significantly from that of pure resin in terms of the curve shape;
however, it appeared that wood substrates’ effects on each individual reaction were different. An activation energy analysis for specific peaks demonstrated that the activation energy of one peak did not change with the addition of wood; thus, the cure rate could be controlled by this reaction and the cure process reached completion at the same end temperature as PF alone for various wood/PF mixtures. If conversion above 70% is of particular interest for practical application, the kinetic parameters obtained from pure resin can be used for prediction of PF/wood mixture. The wood effects were mainly in low temperature, yet when the mixtures were suddenly subjected to a high isothermal temperature, the overall effects of wood addition on cure behavior would be minor. When kinetic parameters from various mixtures by Vyazovkin method were used to predict isothermal cure behaviors, they all obtained similar predictions. The activation energy by nth-BD method decreased with wood addition and with increasing wood content.
Further investigation with the Vyazovkin method indicated that activation energy decreased only at lower conversion, and followed similar trends with PF alone at higher conversion. Moreover, the effects of wood on the curing behavior of the resins among the wood and its extracted counterparts were similar. Additionally, the paper cellulose and hemicelluloses did not change the cure behavior of PF resin at the current studied content level, but the heat of reaction was reduced due to physical separation of PF resin molecules, while lignin and SYP extractives changed the cure development significantly. A small peak appeared in the thermograms of wood/PF and lignin/PF mixtures, supporting the idea that lignin may be the main contributor of the catalytic effect of PF curing.
REFERENCES ASTM E2041-03 Standard Method for Estimating Kinetic Parameters by Differential Scanning Calorimeter Using the Borchardt and Daniels Method.
Barry, A. O.; Peng, W.; Riedl. B. The effect of lignin content on the cure properties of phenol formaldehyde resin as determined by differential scanning calorimetry.
Holzforschung (1993), 47(3): 247-252.
Brent, R. P. Algorithms for minimization without derivatives. Englewood Cliff, N.J., Prentice-Hall, 1973.
Chow, S.-Z. A kinetic study of the polymerization of phenol-formaldehyde resin in the presence of cellulosic materials. Wood Science (1969), 1(4), 215-221.
Detlefsen W. D. In: M. Chaudhury, A.V. Pocius (Eds.), Surfaces, chemistry and applications. Elsevier: Amsterdan, (2002) pp. 869-945.
He, G., &B. Riedl (2004). Curing kinetics of phenol formaldehyde resin and wood-resin interactions in the presence of wood substrates. Wood Science Technology 38(1): 1432-5225.
He, G.; Yan, N. Effect of wood on the curing behavior of commercial phenolic resin systems. Journal of Applied Polymer Science (2005), 95(2): 185-192.
Jones, T. T. Some preliminary investigations of the phenol-formaldehyde reaction.
Transaction, Journal of Society of Chemical Industry, London (1946), 65, 264-275.
Kissinger, H. E. Reaction kinetics in differential thermal analysis. Analytical Chemistry (1957), 29, 1702-1706.
Kottes-Andrews, B.A.; Reinhardt, R. M.; Frick, J. G. Jr.; Bertoniere N. R. In Formaldehyde Release from Wood Products, American Chemical Society Symposium Series No. 316, (1986), Washington, DC.
Laborie, M.-P. G.; Frazier, C. E. 13C CP/MAS NMR study of a wood/phenol-formaldehyde resin bondline. Journal of Materials Science (2006), 41(18), 6001-6005.
Mizumachi, H.; Morita, H. Activation energy of the curing reaction of phenolic resin in the presence of woods. Wood Science (1975), 7 (3): 256-260.
Pizzi, A., Mtsweni, B.; Parsons, W. Wood-induced catalytic activation of PF adhesives.
Auto-polymerization vs. PF/wood covalent bonding. Journal of Applied Polymer Science (1994), 52 (13), 1847-56.
Prime R. B. Thermosets, In: Turi E. A., (Ed.). Thermal characterization of polymeric materials. New York: Academic Press (1997), 435–569.
Wang, J.; Laborie, M.-P. G.; Wolcott, M. P. Comparison of model-free kinetic methods for modeling the cure kinetics of commercial phenol–formaldehyde resins.
Thermochimica Acta (2005), 439: 68–73.
Wang, J.; Laborie, M.-P. G.; Wolcott, M. P. Comparison of model-fitting kinetic methods for modeling the cure kinetics of commercial phenol–formaldehyde resins.
Journal of Applied Polymer Science (2006), accepted.
Vyazovkin, S. Modification of the integral isoconversional method to account for variation in the activation energy. Journal of Computational Chemistry (2001), 22 (2), 178-183.
Vyazovkin, S.; Sbirrazzuoli, N. Isoconversional kinetic analysis of thermally stimulated processes in polymers. Macromolecular Rapid Communications (2006), 27(18), 1515-1532.
Chapter 6 Dynamic Mechanical Analyses of Phenol-formaldehyde
ABSTRACT Modeling and optimizing of wood-based composite manufacture is playing a larger role in the design of processes and manufacturing equipment. In these models, internal temperature and moisture conditions are predicted with an aim towards predicting when polymeric cure is sufficient to avoid delamination. However, most cure kinetics models are focused on predicting the chemical state of the resin rather than the resulting mechanical properties. The objective of this research is to examine the feasibility of obtaining kinetic cure data using dynamic mechanical analysis (DMA). Dynamic three-point bending tests were conducted on a sandwich specimen of two wood adherends bonded with an adhesive layer. The specimen was cured using various isothermal and linear heating regimes. In addition, two commercial PF resins of different molecular weights distributions (labeled as PF-high and -low respectively) were evaluated under different experimental conditions influencing moisture loss. The
parameter to evaluate bond development because it eliminates the variation in adherend modulus. However, this parameter was found to be sensitive to variables such as adhesive thickness and changes in the adherend modulus due to moisture loss
as well as a lower tan δ after curing than the PF-high joints suggesting superior interphase development. The shear modulus and flexural storage modulus of the adhesive were calculated by an analytical solution. The values are in general agreement with the results obtained by parallel-plate rheometry. Overall, the sandwich beam was deemed to be simple in both sample preparation and measurement procedure for obtaining PF resin cure transitions and modulus development.
Key words: Dynamical mechanical analysis (DMA); phenol formaldehyde resins;
shear modulus; storage modulus.
INTRODUCTIONThe curing of thermoset resins is most typically characterized using differential scanning calorimetry (DSC) where the measurement of the energy release rate provides information for modeling the reaction kinetics. However, DSC does not provide information about the structural changes at the molecular level leading towards mechanical property development. In contrast, dynamic mechanical analysis (DMA) offers a quantitative view of the adhesion mechanics from which the glass transition, gelation, and vitrification points may be inferred (Prime 1997).
Thermosetting resins used as adhesives are usually in a fluid state at ambient temperatures. Different approaches have been used to obtain the solid specimens suitable for DMA. First, resin samples may be cured beyond gelation prior to the DMA testing. Then the cure development and characteristic transitions of the partially cured materials can be measured (Prime 1997). However, to fully characterize the resin cure, from the fluid to solid states, necessitates the use of a support. For phenol formaldehyde (PF) resins, both an impregnated, multifilament glass braid (Steiner & Warren 1981) and a glass cloth (Kim and Nieh 1991, Follensbee et al. 1993)) have been investigated. A glass cloth impregnated with the PF resol was used to evaluate the effects of pre-treatment and in situ cure conditions on cure development (Follensbee et al. 1993). These supports are inert to the PF resins (Follensbee et al.
1993) and successfully provided information regarding the cure development for the neat resins. However, it is well established that for PF resins, cure kinetics are influenced by the presence of wood (Chow 1969, Pizzi et al. 1994, Wang et al. 2007).
In addition, the adhesive/glass cloth interface is very different from the interphase formed between PF and wood (Ebewele 1995). Therefore, it is important to characterize PF cure in situ when seeking a realistic view of adhesive bond development. Towards this end, impregnated poplar strips (Laborie 2002) and sandwich specimens, composed of two wood strips separated by an adhesive layer (Garcia and Pizzi1998; He and Yan 2005a) have been used to study the PF cure development. A sandwich structure was favored in that the compliant adhesive layer was subjected to the maximum shear force within the specimen and thus enhancing any phase transitions (Carlier et al. 2001). More importantly, the sandwich geometry more closely resembles the practical application of the adhesives when compared to impregnated specimens. As compared with a fiber glass support, wood is a hygroscopic and viscoelastic material that can display significant variation in storage modulus (E’) resulting from moisture loss and thermal softening during the DMA scans that may reach 200 °C. This condition presents a challenge to interpreting the resulting DMA spectra (Follensbee et al. 1993).
The difference in the storage modulus (∆E’) before and after cure has been used as a criterion to evaluate the effects of resin synthesis parameters (He and Yan 2005a), bio-scavengers (Kim et al. 2006), and catalysts (Onica et al. 1998) on the rigidity of PF/wood strip sandwich joints. He and Yan (2005b) realized that ∆E’ was affected by the wood substrate and recommended that ∆E’ should be normalized by minimum E’. Onica et al. (1998) recommended using the difference between
degradation of wood-adhesive joints.
It is also worthwhile to mention that the DMA signal may be dominated by the substrate, rather the resin layer depending on the ratio of the adherend (h) to adhesive (t) thicknesses. When h/t is high, the sensitivity of the beam stiffness to the presence of adhesive layer is low. Therefore, some researchers (Starkweather & Giri 1982) have recommended a thick adhesive layer to enhance the behavior of the resin when polymer properties, not interphase behavior, are of interest. The characteristic properties associated with pure polymers may not be observed with a sandwich structure. It is important to remember that the DMA spectra should only be interpreted as the behavior of the total joint rather than that of the polymeric resin alone (Onica et al. 1998).
Considerable interest exists in measuring the in situ shear properties of an adhesive when it is used in a bonded joint. Such measurements allow one to assess the state of the adhesive as a function of time and temperature. For example, during hot-pressing of wood-based composites, the wood-adhesive system experiences the thermodynamic and viscoelastic process of consolidation. The pressing time should be minimized to reduce energy use and production time while avoiding defects such as delamination during press opening. The in situ shear or flexural modulus of adhesive during cure is one of basic material parameters needed to construct a useful hot-pressing model. Adams and Weinstein (1975) provided an analytical expression to calculate the shear modulus of the adhesive in a sandwich beam. In this solution, the adherends were assumed to be thin enough that the induced axial stress can be approximated as constant along the cross section. Moussiaux et al. (1987) provided another analytical solution to deduce the shear modulus for similar geometry. This analysis assumed that the adhesive is constrained to a thin layer at the core of a thick, bonded structure. He et al. (2001) confirmed Adams and Weinstein’s solution using a finite element analysis and concluded that it provided a better estimate of shear modulus than the more simplistic Moussiaux’s solution. The analytical results produced significantly different values of shear modulus compared to those obtained from rheometry of neat resins evaluated in the glassy region. However, the results were in good agreements in rubbery region (He et al. 2001). Miyagi et al. (1999) found that the behavior of sandwich beam was liner viscoelastic. Consequently, extension of the method to viscoelastic characterization of the adhesive might be possible.
OBJECTIVES Understanding cure kinetics and property development in wood/adhesive systems is important for evaluating adhesive performance, formulating new resins, and optimizing process parameters. DMA is a commonly used analytical technique for evaluating cure development of polymer systems but has not been standardized in
wood adhesion research. Therefore, the objectives of this research are to:
1. Explore improved techniques for directly evaluating wood-adhesive systems
2. Investigate the potential to use an analytical expression of sandwich specimens to estimate the in situ shear modulus development of the adhesive layer during