Effect of temperature on the dynamic mechanical properties of resin film and wood.
The objective of this research was to investigate the heat resistant performance of structural adhesive as compared with wood samples. Phenol-resorcinol-formaldehyde (PRF) and two wood species, southern pine (Pinus palustris) and Douglas-fir (Pseudotsuga menziesii), were investigated. Uniform thin cured resin films and wood samples were carefully prepared. Dynamic mechanical analysis (DMA) was employed to evaluate the fixed frequency oscillation properties of these samples over the temperature range of 60 to 250[degrees]C. Shear storage moduli, shear loss moduli, and tan [delta] values of the prepared samples were obtained through the DMA temperature scan and isothermal testing. DMA temperature scan tests showed that the storage moduli of the tested materials decreased as the temperature increased. At the same temperature level, PRF resin films had higher storage moduli than that of the wood samples. No obvious difference was found between the two wood species in their dynamic performances. The DMA isothermal tests results indicated that the storage moduli of cured rein films increased with prolonged time at the four different temperature levels (90, 150, 200, and 250[degrees]C), while the loss moduli and tan g values decreased with prolonged time. The moduli and tan g values of wood samples decreased dramatically with prolonged time at 250[degrees]C, while changed slightly at temperatures of 90, 150, and 200[degrees]C.
Wood has a long history of use in building construction. Because the availability of large dimension solid timber has been decreasing in the past decades, the forest products industry has developed a new generation of glued engineered wood products (EWPs), including I-joists, glued laminated timber (Glulam), structural composite lumber (SCL), and others. These EWPs are processed using small-diameter trees as raw materials and exterior-type adhesives. The adhesive plays an important role since it bonds the wood components (dimensional lumbers, veneers, strands, fibers, and others.) together, transfers and distributes stresses, and provides the required strength and structural integrity when the EWPs are subjected to moisture conditions, long-term weathering, and long-term loading. In addition, the adhesive bonding should provide adequate structural performance under high-temperature conditions, such as those reached in the early phases of a structure fire.
The failure of EWPs' bondline suffering heating may be caused by two reasons: 1) insufficient heat durability of the adhesive bondline itself; and 2) the heat-related dimensional change of the adhesive bondline being so different from that of the wood substrate that de-bonding occurs due to the differential expansions. To better understand the principle of delamination, it is necessary to investigate how the adhesive and wood behave at elevated temperatures. In the recent years, since more adhesives have been used in the EWPs, concerns have been raised that whether those adhesives are able to hold up the wood elements when the members are subjected to high temperature, for instance, caused by fire. In residential and commercial building codes, fire durability of adhesives used in structural load carrying members need to be established. Wood, mainly consisting of cellulose, lignin and hemicellulose, can be considered as crosslinked polymers (Tarkow and Feist 1968, 1969). The behavior of wood under elevated temperatures has been studied for many years. Wood is a thermally degradable and combustible material. The thermal degradation of wood depends on the heating rate as well as the temperature. According to White and Dietenberger (2001), wood has a permanent strength reduction at temperatures higher than 65[degrees]C because of the depolymerization reactions without any significant carbohydrate weight loss; and chemical bonds begin to break at temperatures higher than 100[degrees]C. Between 100 and 200[degrees]C, wood is dehydrated to generate C[O.sub.2], formic acid, acetic acid, and [H.sub.2]O. From 200 to 300[degrees]C, hemicelluloses begin to undergo significant pyrolysis to produce additional C[O.sub.2] and high-boiling-point tar. Cellulose begins a significant depolymerization in the range 300 to 350[degrees]C. Lignin is pyrolyzed in the range 225 to 450[degrees]C. The carbon-carbon bonds between lignin structural units are cleaved from 370 to 400[degrees]C. When the temperature is higher than 450[degrees]C, the remaining wood residue is charcoal.
For EWPs, the behaviors of cured resin film at elevated temperature play an important role in EWPs when exposed to high temperature. But little work had been done to investigate the performance of cured resin film at elevated temperatures. Both wood and cured resin film are complex polymeric systems, exhibiting viscoelastic properties. Viscoelasticity is one of the important properties related to polymeric material. It exhibits behaviors that simultaneously possess both solid-like as well as liquid-like characteristics. Every substance will show this nature to some extent. The degree to which the materials exhibit more solid-like or liquid-like properties is dependent upon both temperature and time. Viscoelasticity is commonly characterized by the viscoelasticity parameter, which is usually determined by dynamic mechanical analysis (DMA). DMA is a powerful technique to characterize the behavior of polymeric material. When performing the DMA, an oscillating force, causing a sinusoidal stress is applied to the test specimen, which generates a sinusoidal strain. The following three viscoelastic properties can be obtained: 1) storage modulus, G', which is a measure of the material stiffness; 2) loss modulus, G", which reflects the amount of energy that has been dissipated by the sample; and 3) damping term, tan value, is the ratio of the loss modulus to the storage modulus (tan [delta] = G"/G'). As a ratio of the viscous and elastic components, tan [delta] is a useful index of material viscoelasticity. The changes in the moduli of the material as a function of temperature and time can be monitored from the DMA testing and thus allow the determination of the effects of these changes on the material.
Many studies have been reported on the DMA analysis of wood (Timothy and Glasser 1984, Birkinshaw et al. 1986, 1989). Birkinshaw et al. (1986) investigated the dynamic mechanical properties of 10 species of wood between -100 and +150[degrees]C. The results from this study indicated that the shear storage moduli of all the wood samples decreased steadily with increasing temperature and there was a low temperature transition around -50[degrees]C. They also found a progressive increase in relative damping when the temperature is 50[degrees]C and above. Birkinshaw et al. (1989) also investigated the thermomechanical behavior of solid wood, fiberboard, and wood laminates in a temperature range of -100 to +150[degrees]C. The results showed that the thermo-mechanical performance of solid wood is consistent with the previous study (Birkinshaw et al. 1986) and the adhesives have a significant influence on the damping properties of the fiberboard and wood laminates. Currently, no study has been reported on how the temperature affects the thermo-mechanical properties of cured resins. It is the objective of this study to apply the DMA technique to evaluate the viscoelastic response of cured adhesive at elevated temperatures as well as prolonged time at a constant temperature.
[FIGURE 1 OMITTED]
Wood sample preparation: Two wood species, southern pine (Pinus palustris) and Douglas-fir (Pseudotsuga menziesii), were tested. DMA test samples were milled from the bulk material using a HITACHI 14 to 1/2 inches band saw to give the final dimensions of 40 mm (length) by 10 mm (width) by 1 mm (thickness). The long dimension of the specimen was parallel to the grain whereas the width was in the direction of radial growth. All the prepared samples were conditioned in a standard conditioning room of a target relative humidity of 65 percent and room temperature (20 [+ or -] 3[degrees]C) for at least 2 days, and then put into zipper bags. All the samples were sanded with a fine 3M aluminum oxide sandpaper to obtain a smooth surface before the DMA testing.
Resin film sample preparation: phenol-resorcinol-formaldehyde (PRF) obtained from Dynea, was used to make the resin film samples. The cured PRF resin films were carefully prepared with the assistance of glass plates. The actual curing parameters recommended by the manufacturer during application to a wood component were used (Dynea 2003). Liquid PRF resin was poured on one glass plate (6 inch by 6 inch by 1/4 inch) with four 0.3-mm-thickness gauges placed on the edge (see Fig. 1). Another glass plate with same size was carefully applied on the PRF resin without forming bubbles in the PRF resin. Thirty-six-pound weight was uniformly applied on the glass plate to make PRF resin cured for 24 hours at room temperature. Pieces of cured PRF resin films were cut into a dimension of 40 mm (length) by 10 mm (width) by 0.1 to 0.5 mm (thickness). Care was made during the sample preparation process to keep the specimens consistent.
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
DMA tests were performed using the TA Instruments Model DMA 983, which offers a rapid and sensitive means to simultaneously obtain an elastic modulus (stiffness) and a mechanical damping (toughness) for materials. The instrument has the capability not only for the flexural bending deformation mode of strain, but also for other deformations, such as shear. There are four modes of operation for the DMA 983: resonant frequency, fixed frequency, stress relaxation, and creep. In this study, the fixed frequency mode was used to do the testing.
A single cantilever holder was used to test the specimens in a shear mode. The distance between the two sample clamps was 15 mm. A torque wrench was used to tighten the clamp locking screws with clamping torque of 1.1 N-m (10 N-lb) on the wood samples and cured resin films. The wood samples were clamped on the tangential surfaces so that shear occurred in the radial direction to produce the shear deformation perpendicular to the grain.
DMA temperature scans were performed in a temperature range 60 to 250[degrees]C. A heating rate of 5[degrees]C [min.sup.-1] with a 1 Hz frequency and an oscillation amplitude of 0.2 mm were used for all the DMA testing to obtain the shear storage modulus, shear loss modulus and loss tangent of the samples. Under the same frequency and oscillation amplitude, DMA isothermal tests were performed at four temperature levels (90, 150, 200, and 250[degrees]C) for 30 minutes. A minimum of five replicates were tested for the DMA test (see Table 1). An effort was made to keep the curing conditions of the replicate resin samples consistent with each other. Although there is variation in absolute values, the general trends of the DMA response among the replicates are similar to each other.
Results and discussions
DMA temperature scan tests
The DMA temperature scan results for cured PRF films, southern pine, and Douglas-fir samples were shown in Figures 2 through 4, respectively. Dynamic storage modulus is a structure dependent property. During the DMA tests, as the materials are heated up, they will expand, the free volume of the chain segment of the polymer increases, and its ability to move in various directions also increases. As the temperature continues to increase to some extent, the localized bond and side chain movements can occur, resulting in a decrease in storage modulus. As the temperature continues to go high, the segment, whole side chains or whole main chains in the amorphous regions of the polymer begin to move, resulting in a further decreasing modulus. Therefore, the storage moduli of all the tested materials decreased with an increasing temperature. As shown in Figure 2, the storage modulus of cured PRF film decreased steadily from the 650 GPa at 60[degrees]C to 200 GPa at 250[degrees]C. For Douglas-fir and southern pine samples, the storage moduli decreased from initial storage moduli of about 60 GPa at 60[degrees]C to 28 GPa and 35 GPa at 250[degrees]C, respectively (Figs. 3 and 4). In comparing the storage modulus-temperature curves of the three materials in Figures 2 through 4, it can be seen that, at the same temperature level, PRF film has the higher storage modulus than that of the wood samples.
[FIGURE 4 OMITTED]
The higher moduli of PRF film maybe caused by two reasons: (1) a high degree of cross-linking of the resin; and (2) a high initial molecular weight (Nielsen 1974). The slope of the storage modulus curve can be represented by tan ([pi]-[alpha]) (tan equals to the ratio of the decreased storage modulus to the increased temperature (see Figs. 2 to 4). The storage modulus curve of cured PRF film is almost linear, resulting in a constant tan [alpha] value with increasing temperature. It can be seen from Figure 2 that the storage moduli of the cured PRF films decreased more drastically with increasing temperature compared with wood samples (the values of tan [alpha] for resin film curve are larger than that for wood samples), indicating that the cured resin films are more sensitive to the elevated temperatures. It can also be seen from Figures 3 and 4 that there is no obvious difference on the storage modulus curves between the two wood species. In addition, the tan [alpha] values of the wood samples increased as the temperature increased, indicating that the molecular motion in wood is greater at high temperatures than that at low temperatures.
Loss modulus-temperature curves were also shown in Figures 2 to 4. Loss modulus is related to the energy absorbed by a sample. It is directly proportional to the amount of energy that has been dissipated in form of heat by the sample. It was seen from Figures 2 to 4 that the performance of loss modulus of the cured resin films is totally different from those of wood samples under the same temperature range. The loss moduli of cured PRF films decreased as the temperature increased. At a low temperature (60[degrees]C), the free volume within the polymer is small and the molecules in the polymer samples are tightly compressed. As the temperature increases, the free volume increases and the molecular motion, including the motion of whole side chain and main chain in amorphous region within the material, have more space to move. The greater the free volume, the less heat is produced which caused by the friction of the molecular motion. Therefore, the loss modulus of PRF resin films decreased as the temperature increased. For the two different wood samples, both the loss moduli increased as the temperature increased. Small difference was observed between the two different wood species, possibly caused by the small difference in the structure of the two wood species. There are also two peaks in the loss modulus-temperature curves for wood samples: 160 and 240[degrees]C for Douglas-fir; 140 and 240[degrees]C for southern pine. The different performance between the two wood species and the cured resin films is possibly caused by the different structure of different material and the water molecules in the wood samples. Early study (Furuta et al. 2001) has shown that dynamic mechanical properties are strongly related to the MC of the wood samples. The initial MC of wood samples in this study was 12 percent. As the temperature increases during the DMA testing, the MCs of wood samples are decreased. According to Beall and Eickner (1970), the last traces of water in wood are removed at about 140[degrees]C. As the wood samples are warmed, the molecular motion in wood samples may be associated with water and --OH-units attached to the polymeric material in wood until the temperature reaches 140[degrees]C. Therefore, as the temperature increases, the energy dissipates in form of heat by the wood samples also increases with the removing of water. When the temperature reached 160[degrees]C for Douglas-fir and 140[degrees]C for southern pine, the water in the wood samples was completely removed and the dissipated heat by removing of water reached a peak. When temperature reached about 230[degrees]C for Douglas-fir and about 245[degrees]C for southern pine, the energy absorbed by the wood samples was high enough to soften the main polymeric components (hemicellulose, lignin, and cellulose), and the loss modulus peaks were shown (Nanjian et al. 2007). In this process, the heat released by the friction between the motions of the three main components became highest. It can also be seen from Figures 2 to 4 that the loss modulus of PRF films changed in the range of 20 to 50 GPa, while the loss moduli of wood samples changed in the range of 1.5 to 3.5 GPa. These results indicate that the energy needed to cause the molecular motions in cured PRF films is less than that in wood samples. It also indicated that the motions of molecules in the cured PRF resin films are more drastic than that in wood samples at the same temperature level. Therefore, the storage moduli and loss moduli of cured resin films changed more dramatically than that of wood samples at the same temperature change range. For wood samples, high temperature can soften the lignin so that the storage moduli and loss moduli of wood samples change dramatically at high temperature level (see Figs. 3 and 4).
Tan [delta] is a damping term which is related to the degree of molecular mobility of the material. The higher the peak tan value, the greater the degree of molecular mobility and the better the damping characteristics. From the curves shown in Figures 2 through 4, it can be seen for cured PRF film, the tan value increased continuously as the temperature increased with no peak was observed. For Douglas-fir sample, the tan value increased with the increasing temperature from 60 to 250[degrees]C. But in the range of 160 to 200[degrees]C, the increase rate of the tan [delta] value became smaller and almost constant. For southern pine sample, the same thing occurred. The tan 8 value was nearly constant from 150 to about 200[degrees]C. The reason for the constant tan S value for the two wood samples may be associated with the removal of water in the samples. Among the tested materials in Figures 2 to 4, the tan [delta] values are in the range of 0.06 to 0.10 for cured PRF resin films, 0.02 to 0.11 for Douglas-fir, and 0.02 to 0.08 for southern pine. The higher the peak tan [delta] value, the greater degree of molecular mobility and, hence, the better damping characteristics. In comparing the tan [delta] curves in Figures 2 through 4, it was seen that the molecular motions in cured PRY films is greater than that in wood as the temperature increases. The results are consistent with the results obtained from the loss modulus curves shown in Figures 2, 3, and 4.
[FIGURE 5 OMITTED]
The dynamic response curves (Figs. 2 to 4) show that different materials had different dynamic mechanical properties. The results for wood samples obtained in this research are consistent with the results of Birkinshaw et al. (1986). No obvious difference was found in terms of the temperature scan test between Douglas-fir and southern pine.
DMA isothermal tests
The results of DMA isothermal tests at four temperatures, 90, 150,200, and 250[degrees]C were shown in Figures 5, 6, 7, and 8, respectively. DMA isothermal tests give the storage modulus, loss modulus and tan [delta] as a function of prolonged time at a constant temperature level. It can be seen from the DMA isothermal curves that the shear storage moduli of cured PRF films increased as function of time for all four temperature levels while the shear loss modulus and tan [delta] values decreased with the prolonged time. The increased storage moduli of cured PRF films may caused by postcuring. The postcuring leads to a greater degree of cross-linking from which the molecular motions become more restricted resulting in a decreased tan [delta] value as shown in the Figures 5 through 8. In addition, from the Figures 5 through 8, it can be seen that the postcuring effect is a function of both the temperature and the time.
[FIGURE 6 OMITTED]
For both the Douglas-fir and southern pine, the performances of wood samples in the isothermal tests were almost identical. At 90[degrees]C in the isothermal testing, the storage moduli were almost the same during the 30-minute testing while the loss moduli and tan [delta] values increased slightly. At 150[degrees]C, the storage moduli were also the same while the loss moduli and tan [delta] values decreased slightly with the prolonged time. At 200[degrees]C, all the moduli and tan [delta] values decreased slightly as a function of time. At 250[degrees]C, the moduli and tan values decreased dramatically during the 30-minute testing. Since there are still water molecules in the wood samples at 90[degrees]C, the energy absorbed by the samples may cause the motion of the water molecules in the wood and cannot cause the molecular motion of wood components, shown no change to the storage moduli of the samples at this temperature. At 150[degrees]C, the energy absorbed by the samples may not be high enough to cause the molecular motion of wood components. Therefore, the test results showed the almost same storage moduli during the test time. At 200[degrees]C, as the time prolonged, a slight decrease in the storage moduli was observed. This might be due to the movement of some small groups attached to the main polymeric components, including hemicellulose, lignin and cellulose. At 250[degrees]C in the isothermal testing, a dramatic decrease in the storage moduli from 33 GPa to 26 GPa for Douglas-fir sample and from about 29 GPa to about 19 Gpa for southern pine samples was observed. This might be because of the breakdown of certain polymeric structures of the wood samples.
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
DMA is a useful technique for characterizing viscoelastic behavior of cured resin film and wood. From the three dynamic mechanical parameters obtained from the DMA tests: shear storage modulus, shear loss modulus, and tan [delta] value, the heat dynamic performance of the materials is interpreted. PRF resin film had a higher modulus than that of the wood samples at the same temperature level. Tan [delta] values of PRF films were a little greater than those of wood samples. PRF resin films also had much higher stiffness and better damping capacity than wood samples. In the DMA isothermal tests, the resin films' moduli increased with prolonged time probably because of the postcuring taking place during the tests. At the same time, postcuring caused the decreased damping characteristics of resin films. For the wood samples, no obvious difference was observed in the heat dynamic performances between the two wood species. The moduli and tan [delta] values of wood samples decreased dramatically at high temperature (i.e., 250[degrees]C) while not changing much at the other three temperatures (90, 150, and 200[degrees]C) during the isothermal tests.
Beall, F.C. and H.W. Eickner. 1970. Thermal degradation of wood components: A review of the literature. RP-FPL-130. USDA Forest Serv., Forest Products Lab., Madison, Wisconsin.
Birkinshaw, C., M. Buggy, and A. Carew. 1989. Thermo-mechanical behavior of wood and wood products. J. Mater. Sci. 24(1):359-362.
--, --, and G.G. Henn. 1986. Dynamic mechanical analysis of wood. J. Mater. Sci. Lett. 5(9):898-900.
Dynea. 2003. Tech. data sheet for Prefere[TM]4001/Prefere[TM] 5830S, U.S. version. Dynea USA, Inc. Springfield, Oregon. pp. 1-7.
Furuta, Y., Y. Obata, and K. Kanayama. 2001. Thermal-softening properties of water-swollen wood: The relaxation process due to water soluble polysaccharides. J. Mater. Sci. 36:887-890.
Nanjian, S., S. Das, and C.E. Frazier. 2007. Dynamic mechanical analysis of dry wood: Linear viscoelastic response region and effects of minor moisture changes. Holzforschung 61:28-33.
Nielsen, L.E. 1974. Mechanical Properties of Polymers and Composites. Marcel Dekker, New York.
Tarkow, H. and W.C. Feist. 1968. The super-swollen state of wood. Tappi 51:80-83.
-- and --. 1969. A mechanism for improving the digestibility of lignocellulosic materials with dilute alkali and liquid ammonia. Adv. Chem. Ser. 95:197-217.
Timothy, G.R. and W.G. Glasser. 1984. Characterizing wood components as network polymers by dynamic mechanical analysis. Wood and Fiber Sci. 16(4):537-542.
White, R.H. and M.A. Dietenberger. 2001. Wood products: Thermal degradation and fire. Encyclopedia of Materials: Sci. and Tech. 26: 9712-9716.
Yucheng Peng *
Sheldon Q. Shi *
Moon G. Kim *
The authors are, respectively, Graduate Research Assistant, Assistant Professor, and Professor, Forest Products Dept., Forest and Wildlife Research Center, Mississippi State Univ., Mississippi State, Mississippi (firstname.lastname@example.org, email@example.com, firstname.lastname@example.org). FWRC manuscript #FP450. This research is supported by the USDA Wood Utilization Research Program. Special thanks are given to Dynea for providing the adhesive samples. This paper was received for publication in March 2008. Article No. 10470.
Table 1.--Experiment design for the DMA test. Samples Temperature scan test Replicates (temperature range[degrees]C) Cured PRF resin film 60-250 5 Douglas-fir samples 60-250 5 Southern pine samples 60-250 5 Samples Isothermal test Replicates (temperature[degrees]C) Cured PRF resin film 90 5 150 5 200 5 250 5 Douglas-fir samples 90 5 150 5 200 5 250 5 Southern pine samples 90 5 150 5 200 5 250 5
|Printer friendly Cite/link Email Feedback|
|Author:||Peng, Yucheng; Shi, Sheldon Q.; Kim, Moon G.|
|Publication:||Forest Products Journal|
|Date:||Dec 1, 2008|
|Previous Article:||Transverse mechanical behavior of wood under high temperature and pressurized steam.|
|Next Article:||Strategy to control the effectiveness of microwave treatment of wood in the framework of the implementation of ISPM 15.|