December 18, 2007
Effects of Recycled Materials on the Properties of Wood Fiber- Polyethylene Composites-Part 2
By Hwang, Chin-yin Hse, Chung-yun; Shupe, Todd F
Abstract This study examined the effects of a compatibilizer on the wettability of birch plywood and polyolefins. The compatabilizer was a low molecular weight emulsion type maleated polypropylene (MAPP), Epolene E-43. The polyolefins investigated included low- density polyethylene (LDPE), linear low-density polyethylene (LLDPE), and polypropylene (PP). For E-43 treated wood surfaces, contact angles among four wetting liquids were in the order of phenol formaldehyde (PF) > urea formaldehyde (UF) > isocyanate (ISO) > distilled water. Contact angles increased upon addition of E-43, then decreased as concentrations further increased for UF, PF, and ISO. Among the three polyolefin polymers, the contact angles of PP, either wetting with various concentration of E-43 or with various wetting liquids, showed the highest value but was less sensitive to changes in E-43 concentration. For all four wetting liquids, their contact angle ratings for different plastic types all followed the sequences of PP > LLDPE > LDPE and distilled water > UF > PF > ISO.
Wood flour and fibers are excellent fillers for thermoplastics because of their low density, low cost, high strength and stiffness, desirable fiber aspect ratio, flexibility during processing, and biodegradability (Felix and Gatenholm 1991, Collier et al. 1995, Hwang 1998). However, satisfactory dispersion of wood fillers in the matrices of thermoplastics has always been a problem caused by the hydrophilic nature of wood and the hydrophobic nature of plastic (Felix et al. 1994). The surface characteristics of cellulosic fibers prohibit the formation of a durable interface in the plastic composites and cause failure in stress transfer from one phase to another. Therefore, to enhance the affinity between these two components, the surface properties of one or another must be modified. Maleated polypropylene (MAPP) is a promising compatibilizer to treat both wood fibers and polyolefin. Matching the surface tension is not sufficient to ensure good adhesion between PVC and wood veneers (Matuana et al. 1998). Hence, to understand the effect of maleated polypropylene addition on the chemical nature of the surfaces and the adhesion between wood and plastic, it is necessary to characterize the interrelationship in terms of wettability of liquid ingredients in wood-plastic composites.
This study is the second in a series of reports to examine the effects of recycled materials on the properties of wood fiber- polyethylene composites. The initial report examined the effect of recycled fiber (Hwang et al. 2005). The objective of this study was to examine the effect of MAPP compatabilizer on the interfacial properties of wood and polyolefins. This study is divided into two parts. In the first part, the effect of MAPP treatment on wettability of birch plywood was investigated. In the second part, the wettability of polyolefin polymers by several wetting liquids and MAPP was evaluated.
Materials and methods
A sheet of nominal 1/4-inch-thick 4- by 8-ft. birch plywood sheet was obtained from a local store. Birch was used because this species was used in an earlier study by Hwang (1998) in which birch dowells were embedded in molten plastic and tested for interfacial shear strength. Birch also provides a uniform and smooth surface that is favorable for bonding. The compatabilizer was a low molecular weight emulsion type MAPP, Epolene E-43 from the Eastman Chemical Co.
Four wetting liquids, namely, distilled water, urea formaldehyde (UF), phenol formaldehyde (PF), and isocyanate (ISO), were used to measure the contact angle. The resins were included because an earlier study had shown these different resins provided very different mechanical properties for wood-plastic composites (Wu et al. 1994). Both urea formaldehyde (Casco resin TD-33C) and phenol formaldehyde (Cascophen 1770 to 3) resins were supplied by Hexion Specialty Chemicals, Inc. The typical viscosity at 25 [degrees]C is 175 cps for UF and 130 cps for PF, and the solids content for UF and PF are 65 and 53 percent, respectively. The isocyanate used in this study was ISOBOND, a product of Dow Plastics, which has 30.9 percent NCO content and 185 cps viscosity (25 [degrees]C).
The three types of polyolefin polymers used in this study were supplied by the Exxon Chemical Co. They were polypropylene (PP), linear low-density polyethylene (LLDPE), and low-density polyethylene (LDPE). Specifications for these plastics are shown in Table 1.
Contact angle measurement for birch plywood
The birch plywood was cut into 0.5- by 2.0-in. specimens and polished with #600 sandpaper and vacuum cleaned to remove debris before subjecting to different drying conditions. Half of the specimens were air-dried in desiccators over anhydrous CaSO^sub 4^ for 2 days, and the other half were ovendried at 103+-2 [degrees]C for 24 hours. All specimens were brush-coated with E-43 emulsion of various concentrations onto the surface, and then conditioned in desiccators over anhydrous CaSO^sub 4^ for 2 days. The brush coat of E-43 was controlled by weight gain after coating at a spread rate approximately at 0.1 g of E-43 solution per 1 inch square of wood surface.
The variables investigated for measuring contact angle on birch plywood included drying condition, concentration of E-43, and liquid type. The drying method was either ovendry or air-dry. Coupling agent concentration was 0 (control), 1.25, 2.5, 5, 10, 20, 30, and 40 percent. Replications for concentration levels 0, 10, 20, 30, and 40 percent were 25, and for concentration levels 1.25, 2.5, and 5 percent were 10. The liquid type was distilled water, urea formaldehyde (UF), phenol formaldehyde, and isocyanate (ISO). A 2 x 8 x 4 factorial treatment and split plot design was employed to analyze the factors of interest.
The apparent contact angles of various liquids on the surface of treated birch plywood specimens were measured by the direct optical method using a Kernco (model G-1) contact angle meter at ambient room conditions. The specimen was placed on the stage, and a 0.06- ml droplet was applied with a pipette to the surface of the specimen. The contact angle was measured by rotating the goniometer eyepiece so that the hairline passed through the point of contact between droplet and veneer and was tangent to the droplet at that point. All measurements were made parallel to the grain and 5 seconds after the resin had been dropped. To control the biological variation of the birch plywood, contact angles of the four wetting liquids were measured in every specimen.
Contact angle measurement for plastics
Plastic pellets were placed in TEFLON(R) containers and melted in an oven for 2 hours. The temperature settings for LDPE, LLDPE, and PP were 140, 160, and 200 [degrees]C, respectively. After the molten plastics were set and conditioned in room temperature for 24 hours, contact angles of the four wetting liquids were measured using the same methodology described in the previous section. A 3 x 4 factorial treatment arrangement was used to investigate the effect of wetting liquid on the plastic type. Each treatment combination contained 24 replicates. Contact angles of four wetting liquids, UF, PF, ISO, and water, on polypropylene surfaces treated with 40 percent E-43 emulsion were also investigated in a separate study of completely randomized design. Each treatment had 10 replicates.
Results and discussion
Wettability of MAPP-Treated Birch Plywood
The average contact angles of the four wetting liquids against birch plywood treated with various concentrations of E-43 compatibilizer are shown in Table 2. In both air-dry and ovendry treatments, contact angles of ISO, PF, and UF increased at lower E- 43 concentration levels, then dropped progressively as E-43 concentration increased. Water was most sensitive to E-43 treatment; and, in concentration level above 5percent, there was complete wetting. The control groups (0% E-43) also indicate that contact angles of water were greatly affected by drying method, whereas ISO was least affected.
ANOVA was performed using SAS General Linear Model on square root transformed data, since data sets were unbalanced and did not pass Hartley F-max tests. However, results of ANOVA based on the original data coincided with the transformed data; therefore, for simplicity, the original data were used for further discussion. All of the interaction terms were highly significant (p < 0.01), and as such the ANOVA table is not shown. However, because the F-values in drying method, E-43 concentration, and wetting liquid type were very large, as compared with the three-way interaction of drying method, E-43 concentration, and wetting liquid type, the main effects still provided useful information.
The effect of drying method was highly significant (p = 0.0001). Ovendried birch plywood specimens had a higher mean contact angle (50.4) than air-dried (46.6) when all other factors were combined, implying that samples that were subjected to different drying methods before application of E-43 affected their wettabilities. Surface inactivation caused by drying may be responsible for the loss of wettability in ovendried wood (Christiansen 1990). When all wetting liquids and drying methods were combined, initial increase in E-43 concentration resulted in an increase in contact angle, however, as concentration further increased the contact angle decreased appreciably (Table 2). This is reasonable since MAPP can form hydrogen bonding with the hydroxyl groups on the surface of cellulosic fiber, and also with the hydroxyl group-rich wetting liquids. ISO, however, cannot form hydrogen bonds. Its NCO functional groups are reactive with the hydroxyl groups of wood and the carboxylic groups of MAPP. Therefore, at lower concentrations of E-43, the carboxylic groups of MAPP tend to orient toward wood fiber, leaving long-chain PP backbones facing outward. Low surface free energy of PP induces higher contact angle. However, when E-43 concentration is too high the orientation is interrupted, the carboxylic groups and backbones of MAPP become randomly oriented. This results in greater possibility for PF, UF and ISO to interact with carboxylic groups of E-43, yielding lower angles. One of the keys that Gauthier et al. (1999) reported to reach the expected performance of composites with a polyolefin matrix and natural lignocellulsoic fibers as reinforcement is the surface energies of the two components have to attain a similar level in order to assume good compatibilization.
The four wetting liquids had dramatically different wettabilities on E-43 treated surface (Table 2). In decreasing order of contact angle, were PF, UF, ISO, and water. Contact angle between a liquid and a solid substrate is affected by many factors, such as surface tension and viscosity of the liquid, surface molecular packing, chemical constitution, and critical surface tension of the solid, and the interaction between liquid and solid. In comparing wettability of various wetting liquids, the solid substrates that were subjected to the same treatment should exhibit the same surface properties. Thus, any difference in wettability is due to liquid type and the interaction between liquid and solid. The viscosity and surface tension of these four wetting liquids are different. Since they have different chemical constitutions, their interactions with the solid substrate are also differ chemically. These differences resulted in the different average values of contact angle for the four wetting liquids. The polarity of water and reactivity of ISO may be responsible for their higher wettabilities on E-43 treated wood surface.
The wettability of birch wood has not been widely reported in the literature. Shupe et al. (2001) investigated the wettability of 22 domestic hardwood species that did not include birch. The contact angles reported in Table 2 are much greater than the 5 diffuse species from Shupe et al. (2001) at low E-43 concentrations and much lower at high E-43 concentrations. It is also well known that surface roughness and surface chemistry greatly affects contact angle values. It is likely that with increasing MAPP concentrations, the wood became coated with a smooth MAPP layer which decreased contact angles for all wetting agents (Table 2).
The dependence of contact angle on E-43 concentration for various wetting liquids is shown for air-dried and ovendried plywood in Table 2. The contact angles of UF and PF increased slightly at lower levels of E-43 concentration, and decreased drastically with changes in E-43 concentration from 10 percent to 20 percent, then decreased slightly with further increase in E-43 concentration. The reduction in contact angle from low to high concentrations was most pronounced in UF. This phenomenon implied that the effect of MAPP treatment in improving wettability is more effective in UF for both air-dry and ovendry conditions. The decrease in contact angle from 10 to 40 percent E-43 concentration was 75 percent for UF and 62 percent for PF.
The contact angles of ISO were lower than UF, PF, and water for both air-dry and ovendry methods, but they increased profoundly at lower levels of E-43; as E-43 concentration further increased, contact angles decreased gradually. Among the three adhesives, ISO was most sensitive to the addition of E-43, the increase in contact angle from 0 to 1.25 percent E-43 concentration was 54 percent and 50 percent for air-dry and ovendry samples, respectively. However, the reduction of contact angles at higher concentration levels was less dramatic in ISO as compared to UF and PF, since the total contact angle reduction of ISO from 10 percent to 40 percent E-43 concentration was approximately 39 percent and 31 percent for air- dry and ovendry samples, respectively.
Water reacted totally different from the other wetting liquids and was the major reason for the three-way interaction. The mean contact angle of water decreased drastically upon the initial addition of E-43, and then reached zero, i.e., complete wetting, and remained zero as E-43 concentration exceeded 10 percent. This phenomenon indicates that water can change the equilibrium of MAPP from an anhydride form to a carboxylic form. As explained previously, UF, PF and ISO showed upward trends between 0 to 5 percent E-43 due to the outward orientation of MAPP backbone that is less wettable. However, distilled water, being a polar and nonpolymeric material, is able to react with MAPP to form carboxylic groups, regardless the orientation of MAPP. Hence, its contact angle decreased progressively until the point (10%) when the wood surface was totally covered by E-43 particles. Therefore, the contact angle of water was more sensitive to the application of low concentration E-43 than the other wetting liquids.
The relationship of contact angle with actual E-43 solids content on birch plywood surface was also investigated. Results of polynomial regression analyses, as shown in Table 3, indicate that 69 to 80 percent, 76 to 85 percent of the variations in contact angles of PF and UF could be explained by E-43 solids content.
Contact angle measurement for polyolefin polymers
The average contact angles of the four wetting liquids ranged from 49[degrees] to 92[degrees] for LDPE, 56[degrees] to 94[degrees] for LLDPE, and 65[degrees] to 101 [degrees] for PP (Table 4). The ANOVA (not shown) for the effect of plastic type and wetting liquid on contact angle of plastics found that the main effects of plastic type and wetting liquid) and their interaction were all highly significant (p = 0.0001). The relationship of contact angles with plastic types was PP > LLDEP > LDPE, indicating that PP may have lower surface free energy than LDPE and LLDPE and hence is less wettable. The effect of wetting liquid type (Table 4) shows that ISO was most capable of wetting polyolefin polymers, and water was the least capable. PF resin showed more affinity to polyolefin polymers, and water showed the least affinity. PF resin showed more affinity to polyolefin polymers than did UF resin.
Among the three adhesives, ISO exhibited relatively strong wettability to both wood (0 = 27[degrees]~49[degrees]) and plastics (theta = 49[degrees]~65[degrees]). ISO is a reactive wood adhesive because it is mostly comprised of low molecular weight dimers, diphenylmethane diisocyanate (MDI), which actively react with the hydroxyl groups present in water and cell wall components. However, the dipole-dipole interaction is weaker in ISO, and its non-aqueous carrier has higher affinity to nonpolar polyolefin polymers. Furthermore, ISO has an outstanding adherence capability to virtually any solid surface (Johns 1982). This may explain the strong affinity of ISO to both wood and plastics. Compared with PF, UF had higher affinity to higher concentration E-43 treated wood, and a lower affinity to untreated wood, lower concentration E-43 treated wood, and polyolefin surfaces. This may be due to less hydroxyl groups in UF than in PF. However, the difference in wettability between PF and UF was less dramatic. The responses of wetting liquid to different plastic types were consistent in the order of PP > LLDPE > LDPE and distilled water > UF > PF > ISO. The liquid and plastic type interaction is presumably due to the inconsistency in the magnitude of difference.
The contact angles of the four wetting liquids on E-43 coated PP surfaces are presented in Figure 1. The Tukey's test indicates that PF had the highest average contact angle, followed by UF, ISO, and water. This trend was similar to that of E-43 treated birch plywood (Table 2), but was notably different from that of untreated polyolefin polymers (Table 4). This demonstrates that E-43 can dramatically change substrate surface properties, conforming to the theory that a contact angle is affected only by the outermost constitution of the surface (Zisman 1964, 1976).
The wettabilities of wood and plastics by different wetting liquids suggest that if wettability is the major consideration of adhesive selection, ISO may be better than PF and UF, assuming that other factors such as cost, processibility, bond strength, etc. are not considered. No MAPP is needed with ISO as the binder. If UF and PF are used, MAPP may enhance the surface properties. However, in the design and manufacture of composites, bond strength is always a major concern. The next report of this series will summarize our findings on interfacial bond adhesion.
For E-43 treated wood surfaces, contact angles among the four wetting liquids were in the order of PF > UF > ISO > distilled water. Contact angles increased upon addition of E-43, then decreased as concentrations further increased for UF, PF, and ISO. Experimental results also showed that the E-43 treated surface could be completely wetted by water, and partially wetted by three other wetting liquids. The reduction in contact angle was most abrupt between 0 percent and 10 percent E-43 concentration for water, and between 10 percent and 20 percent for UF and PF. ISO was most sensitive to the addition of E-43, and least sensitive to the increase in E-43 concentration.Among the three polyolefin polymers, the contact angles of PP showed the highest value but was less sensitive to changes in E-43 concentration. However, contact angles of LDPE and LLDPE gradually increased with an increase in E-43 concentration. For all four wetting liquids, their contact angle ratings for different plastic types all followed the sequences of PP > LLDPE > LDPE and distilled water > UF > PF > ISO. The objective of this study was to examine the effect of MAPP compatabilizer on the interfacial properties of wood and polyolefins. These results should help grow the rapidly developing wood-plastic industry and provide the basis for additional research. Part 3 of this work will report on the effect of compatibilizers and adhesives on the interfacial adhesion of wood/plastic composites.
American Soc. for Testing and Materials (ASTM). 2000. D792-00 Standard test methods for density and specific gravity (relative density) of plastics by displacement. Annual book of standards. Vol. 08.01. West Conshohocken, Pennsylvania.
_____. 2004a. D1238-04c Standard test method for melt flow rates of thermoplastics by extrusion plastometer. Annual book of standards. Vol. 08.01. West Conshohocken, Pennsylvania.
_____. 2004b. Standard test method for tensile properties of plastics D638-03. Annual book of standards. Vol. 08.01. West Conshohocken, Pennsylvania.
Christiansen, A.W. 1990. How ovendrying wood reduces its bonding to phenol-formaldehyde adhesives: A critical review of the literature. Part I: Physical responses. Wood and Fiber Sci. 22(4):441-459.
Collier, J.R., M. Lu, M. Fahrurrozi, and B.J. Collier. 1995. Reactive composite systems. In: D. F. Caulfield, R. M. Rowell, and J. A. Youngquist (eds.). Woodfiber-plastic composites: Virgin and recycled wood fiber and polymers for composites. Forest Prod. Soc. Madison, Wisconsin. pp. 67-73.
Felix, J.M. and P. Gatenholm. 1991. The nature of adhesion in composites of modified cellulose fibers and polypropylene. J. Appl. Polym. Sci. 42(3):609-620.
_____, _____, and H.P. Schreiber. 1994. Plasma modification of cellulose fibers: Effects on some polymer composite properties. J. Appl. Polym. Sci. 51(2):285-295.
Gauthier, R., H. Gauthier, and C. Joly. 1999. Compatibilization between lignocellulosic fibers and a polyolefin matrix. In: Fifth Inter. Conf. on Woodfiber-Plastic Composites. Forest Prod. Soc. Madison, Wisconsin, pp. 153-164.
Hwang, C.Y. 1998. Effect of recycled fiber, compatibilizer and preformed fiber handsheet on the performance of wood-polyolefin composites. Ph.D. Diss. Louisiana State Univ. Baton Rouge, Louisiana. 194 pp.
_____, C.Y. Hse. and T.F. Shupe. 2005. Effects of recycled materials on the properties of wood fiber-polyethylene composites - Part 1: Effect of recycled fiber. Forest Prod. J. 55( 11 ):61-64.
Johns, W.E. 1982. Isocyanates as wood binders - a review. J. Adhesion. 15:59-67.
Matuana, L.M., J.J. Balatinecz, and C.B. Park. 1998. Effect of surface properties on the adhesion between PVC and wood veneer laminates. Polym. Eng. Sci. 38(5):765-773.
Shupe, T.F., C.Y. Hse, and W.H. Wang. 2001. An investigation of selected factors that influence hardwood wettability. Holzforschung 55(5):541-548.
Wu, Y., C.Y. Hse, E.T. Choong, and C.Y. Hwang. 1994. Effect of resin variables and plastic components on wood-plastic composites. In: 2nd Pacific rim bio-based composites symposium. Wood Sci. Dept., Univ. of British Columbia. Vancouver. B.C. Canada, pp. 64-71.
Zisman, W.A. 1964. Relation of the equilibrium contact angle to liquid and solid constitution. In: Contact angle, wettability and adhesion. Advances in chemistry series No. 43. American Chemical Soc. Washington, pp. 1-51.
_____, 1976. Influence of constitution on adhesion. In: I. Skeist (ed.). Handbook of adhesives. 2nd ed. Van Nostrant Reinhold Co. New York. pp. 33-71.
Todd F. Shupe*
The authors are, respectively, Senior Scientist and Head, Forest Chemistry Division, Taiwan Forestry Research Inst., Taipei, Taiwan ([email protected]); Principal Wood Scientist, Southern Research Sta., USDA Forest Serv., Pineville, Louisiana ([email protected]); and Professor, Louisiana Forest Products Development Center, School of Renewable Natural Resources, Louisiana State Univ. Agri. Center, Baton Rouge, Louisiana ([email protected]). This paper (No. 06-40-0136) is published with the approval of the Director of the Louisiana Agri. Expt. Sta. This paper was received for publication in January 2007. Article No. 10305.
*Forest Products Society Member.
(c)Forest Products Society 2007.
Forest Prod. J. 57(11):80-84.
Copyright Forest Products Society Nov 2007
(c) 2007 Forest Products Journal. Provided by ProQuest Information and Learning. All rights Reserved.