«CURE KINETICS OF WOOD PHENOL-FORMALDEHYDE SYSTEMS By JINWU WANG A dissertation submitted in partial fulfillment of the requirements for the degree of ...»
Studying the cure of epoxy resins with torsional braids, Gillham (1979) found two peaks distinguished in the tan δ curve. The first peak was defined as gelation point and the second as vitrification. However, in most cases, only one peak appears on the tan δ curves and it has been interpreted either vitrification (Kim & Nieh1991) or gelation (Toffey and Glasser 1997). When only one peak is present, some researchers interpret the onset of increase in E’ as gelation (He and Yan 2005b) while others define the point where E’ levels as vitrification (Lopez et al. 2002).
For PF resins, some researchers find two peaks in the tan δ curve (Garcia and Pizzi 1998, Laborie 2002) while others only find a single peak (Kim & Nieh 1991;
Onica et al. 1998). Based on gelation and vitrification point at various isothermal temperatures, Time-Temperature-Transformation (TTT) diagrams can be constructed to characterize the curing behavior of thermosetting resins and composites (Simon and Gillham 1992). The TTT diagram can provide necessary information for process parameter optimization. For example, extended dwell time before gelation when the viscosity of the resin during curing process is a minimum allows good wetting of the wood fibers and consequently provides a good adhesion of the wood fibers in the final product. Garcia and Pizzi (1999) and Laborie (2002) have constructed partial TTT cure diagrams for the PF-wood bonded joint and PF impregnated strips respectively.
In summary, the data recorded with DMA enriches a lot of information about resin cure development.
OBJECTIVES Understanding cure kinetics and strength 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. To date, all cure kinetic analyses are exclusively used for modeling DSC data (Wang et al. 2006). In this perspective, the objectives of this
research are to:
3. Investigate the effectiveness of DMA in characterizing the gelation and vitrification events for PF resins,
4. Validate the application of model-fitting kinetics to DMA data.
EXPERIMENTALPF resins Two PF resole resins, tailored as adhesives for oriented strand boards, were obtained Georgia-Pacific Company, frozen and stored at -20°C until use. The low molecular weight resin (PF-low) had a weight-average molecular weight (Mw) of 621 g/mol and a polydispersity (Mw/Mn) of 1.41. The high molecular wieght resin (PF-hi) displayed an Mw = 6576 g/mol and Mw/Mn =1.72. The resin solid contents were 54.5% and 45.0% for PF-low and PF-high respectively. In addition, elemental analysis showed the presence of 3.9 and 3.7 wt % nitrogen for PF-low and PF-high respectively, indicating the presence of urea in both systems.
Specimen preparations Planed basswood strips (Midwest Products, Inc.) with nominal dimensions of 50x12x1 mm were oven-dried at 103 °C and stored in a desiccator over anhydrous calcium sulfate until use. Sandwich-type DMA specimens were produced from a layer of PF resin between two pieces of wood strips. Care was taken to match the grain, thickness, and weight of the two wood strips within the specimen to maintain a balanced composite design. The bonding surfaces were lightly hand sanded along the grain with 220-grit sandpaper and cleaned with a paper towel immediately prior to resin application. The resin was uniformly applied to the prepared surface of both wood strips using a small airbrush (BADGER Model 350). The amount of resin solid applied to each surface was set at ca. 50 g/m2, which equates to ca. 12% of dried wood mass.
Maintaining a consistent resin content was deemed important for obtaining reproducible repeated cure analysis. He & Yan (2005b) demonstrated that the degree of resin loading can influence the cure development. They concluded that this influenced occurred primarily through water absorption and evaporation during the DMA test. Therefore, other measures to maintain moisture content during the tests were investigated. These include (1) short open and closed assembly times in producing the specimens and (2) foil wrapping of the specimens for the DMA analysis.
The latter technique was only used while evaluating the cure kinetics to compare to DSC data.
DMA DMA measurements were conducted on the sandwich specimens in three point bending mode (span 25 mm) using a Tritec 2000 instrument (Triton Technology).
Scans were conducted using a fixed of 1 Hz under various isothermal conditions from 70 to 180 oC and thermal ramps from 2 to 5 °C/min. Low heating rates were selected to make sure that the effect of thermal lag was minimal. When conducting isothermal tests, the DMA oven was preheated to the predetermined isothermal temperature, upon which time the specimen was quickly installed. After the scan began, the oven was maintained at the cure temperature until both the E’ and tan δ approached a constant value signifying the completion of detectable cure. The specimen was then cooled to room temperature, and re-scan at 2 °C/min. Strain sweep tests were conducted to establish the linear viscoelastic ranges at each working temperature.
Typical strain settings ranged from 1-2•10-4.
RESULTS AND DISCUSSIONCharacterization of cure development The representative changes for E’ and tan δ with temperature is represented in Figure 7.1a for a typical aluminum foil-wrapped PF-high bonded wood joint cured at 2 °C/min. Three distinct zones were observed from E’ curve: thermal softening of un-cured wood-resin system, resin curing, and thermal softening of cured wood-resin system. Upon application of the liquid PF resin to the dry wood surface, water is absorbed by the wood causing the adhesive layer to become semi-solid at room temperature. In this state, the adhesive can transfer partial shear forces between two pieces of wood. With increasing temperature, the resin gradually softens and the E’ decreases reaching a minimum E’ plateau. This event appears on tan δ curve as the first peak centered at ca. 50 to 70 °C (Figure 7.1a). At this point, the E’ reaches a minimum plateau corresponding to a competitive relationship between the resin softening and curing during the heating process. With the subsequent increase in E’,
defined here as the onset of the mechanical cure (β = 0). Shortly after E’min, a second peak is evident in tan δ ( Figure 7.1a). This second peak is taken to be the gelation point, where the cross-links progressed to form an “infinitely’’ network (Gillham 1979;
Toffey and Glasser 1997). When only a single peak is evident in the tan δ curve, gelation was similarly defined by temperature corresponding to the onset of increase in E’ (He and Yan 2005b). Finally, the third peak in the tan δ curve was defined as the vitrification point, i.e. the attainment of a glass state where the glass transition temperature of the forming polymer exceeded or was equal to the oven cure temperature (Gillham 1979; Toffey and Glasser 1997). At higher temperatures E’ began to decrease slightly due to thermal softening of the cured resin and the attained ' maximum, Emax, may represent progressive degradation of wood substrate or differential expansion between resin and wood (Onicaa et al. 1998).
3 0.08 2 0.04 1 0.00
Figure 7.1 The DMA cure profiles of the aluminum foil-wrapped PF-high bonded wood joints: a) storage modulus development (E’) and three peaks in tan δ showing glass transition temperature of the uncured and dehydrated PF resin (Peak 1), gelation (Peak 2), and vitrification (Peak 3) events at 2 °C /min; b) gelation and vitrification temperature dependence on heating rates; c) modulus development and gelation (small shoulder 2) and vitrification (Peak 3) in tan δ curve at 120 °C; d) tan δ at different isothermal temperatures.
the time during isothermal cure were presented in Figure 7.1c & d. The E’ curve displayed a sigmoidal shape while the tan δ exhibited a peak followed by a decrease toward an asymptotic limit. Similar behavior may be observed as a function of decreasing temperature, and therefore increasing relaxation time, for a non-reacting system (Dillman & Seferis 1989). Polymer viscoelasticity or damping as quantified by the tan δ, is around from 0.1 to 1 or more while tan δ is 10-3 or less for structural metals such as steel, brass, and aluminum (Lakes and Quackenbush 1996). During the cure process, the magnitude of tan δ passed through a range typical of the rubbery state before vitrification and decreased to a minimum of around 0.05 after vitrification,, which is typical of cured polymers. The vitrification is an analog to the rubber to glass transition of the forming amorphous polymer. Under isothermal cure, the time required to reach full cure can be easily determined and can be subsequently used as a tool to establish an optimal pressing time during wood composites manufacturing. At 120 °C, it required around 20 min to complete the cure for foil-wrapped PF-high bonded joints. This state was confirmed by re-scanning the cured sample where no secondary curing has been detected.
Gelation and vitrification temperatures at different ramp rates are summarized in Table 7.1 for foil-wrapped PF-high bonded wood joints and vitrification times under isothermal cure regimes are summarized in Table 7.2. The gelation point under isothermal temperature appears only as a small shoulder (Figure 7.1c) and is unable to be reproducibly quantified in some samples. It was assumed that the sudden temperature increase from room temperature might account for this problem because gelation could occur while the instrument established equilibrium. For un-wrapped samples, the gelation peak was not evident both under linear heating and isothermal regimes. Typical tan δ traces for PF-low bonded wood joints are shown in Figure 7.2.
It was assumed that rapid moisture loss in the un-wrapped samples was responsible for unrecorded gelation peaks. Hence, only vitrification temperatures were summarized in Table 7.1under linear heating rates for un-wrapped wood joints and vitrification times in Table 7.2 under isothermal temperature.
0.40 0.35 2 0.30 3
Figure 7.2 DMA tan δ traces for PF-low bonded wood joints at different heating rates designated on the curves.
Only vitrification has been recorded at all heating rates.
Table 7.1 Characteristic temperature (°C) at different heating rates and activation energy Ea by the Kissinger equation.
PF-high Vitrification 114.5 123.2 127.5 132.2 61.8 PF-low Vitrification 121.4 127.5 132.6 137.0 72.4 CV is less than 1.4% for temperature, R2 0.99 for activation energy, Ea Table 7.2 Vitrification time (min) and activation energy Ea at different isothermal cure temperatures as determined using the peak time method.
CV is less than 5.8%, R2 0.99 for Ea Activation energy of gelation and vitrification Under the isothermal cure regimes, the time required to reach the peak tan δ (tipeak) at the resident temperature (Ti) was used to calculate the activation energy (Ea).
A linear regression of ln(tipeak) and 1/Ti across the isothermal temperatures yielded activation energy, i.e.
Under cure regimes using constant heating rates (ϕi), the Tipeak is defined as the temperature associated with the peak tan δ. A linear regression of ln (ϕi Tipeak ) versus 1/ Tipeak across several heating rates yields the activation energy with the Kissinger Eq.
The respective activation energy is then assigned to either the gelation or vitrification processes depending on which the peak is selected. Computed values for each are summarized in Table 7.1and Table 7.2.