Wood flour is a renewable cellulosic material that has environmental and economical advantages as filler in polymers (Son et al. 2001; Balasuriya et al. 2002; Bledzki and Faruk 2005; Kim et al. 2005; Arbelaiz et al. 2006; Kumar et al. 2011). The performance of vegetable reinforcing fillers is widely investigated as such materials are biodegradable and non toxic, have a low-cost on volume basis, lower specific gravity of plastics relative to mineral fillers, and cause less abrasion during processing (Premalal et al. 2002; Yang et al. 2004). The most often used wood species for production of wood plastic composites (WPCs) are pine, oak and maple (Wolcott and Englund 1999; Ashori and Nourbakhsh 2010). It was demonstrated that hardwood flours provide higher flexural strength and better tensile properties than softwood flours in WPCs (Berger and Stark 1997). In the present work, maritime pine wood flour will be in focus. Polypropylene (PP) is one of the most important commercial thermoplastics. It has good mechanical and thermal properties, and low density and low melting temperature (Son et al. 2001; Premalal et al. 2002; Yang et al. 2004; Bledzki and Faruk 2005; Arbelaiz et al. 2006; Ashori and Nourbakhsh 2010; Lee et al. 2012). However, there is a phase incompatibility between the hydrophobic PP matrix (low surface energy) and the hydrophilic wood flour (high surface energy) (Maldas and Kokta 1993; Kazayawoko et al. 1999; Dominkovics et al. 2007), with weak interfacial adhesion between them. The strong fiber-fiber interactions, which are leading to agglomeration, limit the dispersion of the fibers in the matrix. The hydroxyl groups in the wood flour interact with water molecules and cause swelling and this has a detrimental effect on the WPC durability. The unsatisfactory compatibility of wood fiber and PP leads to poor mechanical properties (Lu et al. 2000; Mohanty et al. 2002). The performance of fiber-matrix interfacial adhesion can be improved by pre-treatment of the fibers, which can reduce the moisture absorption of the latter (Joseph et al. 1996; Saiful Islam et al. 2012). Compatibility in the blends may also contribute to a better the fiber-matrix compatibility (Gardea-Hernandez et al. 2008). Treatment of fibers with sodium hydroxide (Mwaikambo and Ansell 1999), the coupling with functional silanes (Nachtigall et al. 2007), and functionalization of PP with maleic anhydride (MA) in the presence of an organic peroxide as an initiator, were investigated (Kazayawoko et al. 1999; Keener et al. 2004). The MA was grafted onto the PP first, and the grafted product (MA-PP) was used as a compatibilizer with a dosage around 3%. This approach improved tensile strength, dynamic modulus, and the dispersion of fibers in the matrix (Kazayawoko et al. 1999; Keener et al. 2004). It is believed that the succinic anhydride moieties of MA-PP form ester linkages and hydrogen bondings with wood hydroxyl groups. Important parameters in this context are the chemical structure and the amount of the compatibiliser and the addition method. Nachtigall et al. (2007) studied the properties of WPC with PP as polymer matrix. Here, PP was functionalized with organosilane or with MA. In the first case, the interfacial adhesion between PP and wood flour was better than in the case of MA-PP. Ichazo et al. (2001) treated the wood flour with NaOH and silane and applied MA-PP as the matrix. All treatments improved the dispersion of the wood flour particles, and WPCs with silane modification combined with MA-PP addition had a lower degree of water absorption. The alkaline treatments did not show improvements, probably because of the increased presence of hydrophylic OH groups. Successful pretreatment of wood flour with hot water extraction (Pelaez-Samaniego et al. 2013), extraction and delignification (Chen et al. 2014), combination the flour with vitamin E (Peng et al. 2014), chemical modification with styrene acrylonitril co-polymer and nanoclay (Devi et al. 2012), are mentionable.
Some of those above discussed parameters of WPC production which should be studied in relation to the present paper. Maritime pine wood flour should be modified by thermal treatment, called “retification” and carried out at 280°C under inert atmosphere (Argon) (Guyonnet and Bourgois 1988). Retification is a mild pyrolysis (T<280°C), mainly used to crack hemicelluloses and slightly modify lignin. By-products of hemicelluloses condense and polymerize on lignin chains (reticulation). The word “retification” is a composite of “reticulation” and “torrefaction”. These reactions lead to a modified lignin, which is more hydrophobic and rigid than the original one (Guyonnet and Bourgois 1988). The hydrophobic character of the retified wood is elevated and the amount of volatile components is lowered (Jaziri et al. 1987; Bourgois and Guyonnet 1988; Bourgois et al. 1989). The expectation is that this pre-treatment will improve the interfacial compatibility of wood and PP. The composite blends will be prepared by means of a co-rotating twin-screw extruder with two different screw speeds. The effects of production parameters and retification will be evaluated based on the physical properties of the WPCs obtained.
Materials and methods
The polypropylene (PP-Borcom™ WH107AE, manufactured by Borealis) is commercially available. It contains 13% fillers (carbon black, talc); it has a density of 985 kg m-3 and a Melt Flow Index (MFI) of 40 g 10 min-1 (230°C/2.16 kg). This polymer was chosen because of its high MFI value, which provides a good processibility. The wood fillers were various maritime pine flours.
Retification was carried out in a batch pilot-scale reactor, which is electrically heated and the heat transfer is enforced by convection. After heating, the wood was cooled down rapidly to ambient temperature via an air driven double shell. The average diameter of wood flour is 400 μm.
Preparation of the composites: The WPC was prepared at 200°C under an inert atmosphere (N2) in a co-rotating twin-screw high shear extruder Leistritz ZSE 18HP with a screw diameter of 18 mm and a length-to-diameter ratio L/D=60. Screw speeds of 500 and 1200 rpm were applied. The mass ratio of the wood flour to polymer (W/P) was 3:7 (w:w) for all the blends. The screw configuration is illustrated in Figure 1. The extruded materials were cooled and granulated. The samples were prepared at 200°C with an injection molding machine Battenfeld.
Thermo gravimetric analysis (TGA) were carried out with a TA Instruments TGA Q 5000 at heating rates of 10°C min-1 under N2 atmosphere from 20°C to 600°C.
For scanning electron microscopy (SEM) the samples were coated with a thin layer of gold. Instrument: Zeiss SUPRA 55VP with the improved GEMINI column. The micrographs were taken at magnifications of 40 and 500.
Contact angle (CA) measurements were performed with a CAM unit (KSV instruments) equipped with a camera having telecentric optics and 55 mm focus length. Distilled water droplets were placed on the composite surfaces. The CA values obtained are the average of at least 5 measurements of each composite at room temperature.
Shear dynamic measurements were carried out with an ARES rheometer (Rheometric Scientific), which is a rotational rheometer manufactured by TA instruments. Parallel plate geometry with a diameter of 25 mm was used for frequency sweeps. The test specimens are discs that have been prepared by compression molding at 200°C. The range of frequency sweeps was from 0.1 to 100 rad s-1, and a strain of 5% was used to ensure that all measurements were conducted within the linear viscoelastic region. The shear dynamic measurements were carried out at a constant temperature of 200°C and a gap height of 1 mm. From the data, storage modulus G′ and complex shear viscosity were calculated as a function of the frequencies.
The tensile strength properties were evaluated according to ASTM D63 by means of an Instron mechanical machine MTS 2/M at a crosshead speed of 20 mm min-1. The specimens dimension: 25 mm length, 5 mm width and 2 mm thick. Tests were carried out at room temperature (r.t.) and at least 5 specimens were tested for each composite blend and the results were averaged to a mean value.
Impact testing of the composites was carried out with an instrumented Charpy impact tester with 4 J capacity at the maximum pendulum height. Specimens of thickness 4 mm, width 9 mm and length 77 mm were tested. The samples were V-notched and were placed into a special holder with the notch oriented vertically and toward the origin of impact. The specimen was struck by a hammer tup attached to a swinging pendulum with a wedge-shaped nose. The specimen broke at its notched cross-section upon impact, and the upward swing of the pendulum was the measure of the energy absorbed in the process. All the reported values were the average of at least 8 experimental results at r.t.
Results and discussion
The TGA weight loss curves of treated and untreated wood (Wtr and W, respectively) are presented in Figure 2 as a function of temperature under N2 atmosphere. The fibers were thermally stable. As seen, there was a first stage of mass loss (ML) ranged from 40°C to 105°C and the ML of W is around 4.8% and that of Wtr is ca. 3.1%. These were caused by the loss of free water by splitting off free OH groups, which led to elevated hydrophobicity. The second stage of ML started at 293°C for W, and Wtr began to degrade at 309°C. Similarly, the ML of W was 50% at 365°C, whereas W shows the same ML at 374°C. Expectedly, the thermal stability of Wtr was higher than that of W. It was obvious that Wtr is better suited for WPC manufacturing.
Lignin starts to decompose earlier than cellulose, but its decomposition rate is slower than that of cellulose (Gasparovic et al. 2010). The aromatic ether linkages in lignin have a high thermal stability. The crystalline cellulose with its intrachain hydrogen bonds has a higher thermal stability compared with hemicelluloses (Mihai and Cornelia 2009). Hemicelluloses are the least stable polymers of wood. The thermal stability of wood flour is increased with thermal treatment, which mainly affects hemicelluloses. Ether and ester linkages were splitting during the treatment and the modified molecules stabilized by molecular rearrangement and condensation, which lead to higher cross-linking densities. These reactions make the lignin more hydrophobic and more rigid than the original lignin (Duchez and Guyonnet 1998).
The particles of Wtr were shrunk and darkened as a result of partial degradation of some lignocellulosic components. The fibers of Wtr become brittle. Thus the mechanical effects of the process lead to milling, and the particle size was diminished.
SEM analysis was performed to observe the effects of thermal treatment (Figure 3). The anisotropic geometry of the particles can be observed and their average aspect ratio clearly differs from 1? Their size covered a wide range from several μm to mm. The fibers surface was rougher after treatment (Figure 3a,b). No changes in the tracheids were visible at 500× magnification, and the pits also seemed to be intact (Figure 3c,d) but the pore sizes was elevated, which favored the penetration of polymers into wood.
The contact angles (CA) of PP, WPC, and Wtr PC extruded at 500 rpm are displayed in Table 1. The CA of PP seems to be lowered from 100° to 96° in the case of WPC and to 98° in the case of Wtr PC, but these data are statistically not reliable. It is possible that a PP layer is formed on the WPC surfaces which masks the effect of fibre treatment in deeper layers (Hosseinaei et al. 2012). To avoid this effect, CA was also observed on wood flour surfaces. Expectedly, untreated wood flour has a lower CA than the Wtr (Table 1). It is well known that the reduced hydrophylicity of heat treated wood is the main reason for its higher outdoor resistance (Commereuc et al. 2010).
The viscoelastic behavior was very sensitive to interactions between the polymer components and the filler. However, before the rheological tests, the effect of the screw speed for WPC production was studied as it is known that the viscosity is affected by the shear rate, and the residence time in the extruder (Sombatsompop and Panapoy 2000), because of the lower residence time of PP in the extruder at higher screw speeds. It is important to work under inert atmosphere and in the presence of stabilizers to limit the β-scission degradation of the PP matrix in the extruder. Figure 4 shows the evolution of the elastic modulus G′ (Figure 4a) and complex viscosity η* (Figure 4b) for the commercial PP at screw speeds of 500 and 1200 rpm. No important impact of the screw speed was detectable on the rheological behavior of PP, G′ and η* as the corresponding curves are almost overlapped.
Figure 4c shows the effect of the screw speeds on the WPCs. The sol-gel transition due the formation of gel-like structure by solid particles inside the PP was characterized in this figure by the yield stress behavior of the viscosity. There was no real differences, i.e., the yield stress behavior was nearly the same for both composites extruded at 500 and 1200 rpm.
The same was true for Wtr PC as presented in Figure 4d. It is noted that the dispersion state of the blend seemed not to be affected. The viscosities of all WPCs at the 500 rpm level were higher than that of PP matrix (Figure 4e). The incorporation of Wtr into the composite exhibits a higher viscosity compared with that of W. The elastic modulus follows the same pattern (Figure 4f). As expected, the inter-particle interactions were clearly visible in the blends and wood particles contributed considerably to the apparent hardening of the composites. In all cases, the effects of Wtr were beneficial. The behavior of the WPCs extruded at 1200 rpm is similar.
The data of tensile modulus and elongation at break are presented in Table 2. Also here, the data decrease slightly by increasing the screw speed from 500 to 1200 rpm. The PP was subjected to the β-scission degradation which detracted from the mechanical properties. This effect was not visible by means of rheological analysis. The tensile modulus of PP was improved in the case of WPC by 15% from 1600 to 1900 and 2990 MPa (500 rpm and 1200 rpm, respectively). The best results were obtained in the case of Wtr PC, namely 3700 MPa (125% improvement against PP). Remarkably, the tensile modulus of Wtr PC deteriorated at 1200 rpm. Probably, the high shear forces of the twin screw extruder caused chains scission of the PP and the screw speed parameters had to be optimized separately for each application. It is believed that the improved interfacial bonding between the filler and the polymer matrix in case of Wtr PC resulted in better stress propagation and improved the tensile modulus.
Conversely, the elongation at break decreased steadily with the wood-fiber content. There was no significant difference in elongation at break for WPC and Wtr PC. In presence of the wood filler the elongation at break decreased rapidly because of wood fibers with their low elongation at break and restricted the flow of polymer molecules. This behavior is typical of reinforced thermoplastics in general and has been reported by other researchers (Felix and Gatenholm 1991).
On the other hand, the notched impact strength of the composites decreased (Table 3). This is consistent with the results of Bledzki and Faruk (2004) and Nygard et al. (2008). The wood–fibers in the PP matrix provide points of stress concentrations and were sources of crack initiation. The thermal degradation of the fibers due to high shear forces in the kneading section of the twin screw extruder could be another reason for the decrease in impact strength.
Generally, wood in WPCs increases stiffness, and at the same time reduces their toughness. The brittleness of the composites increased significantly when the wood fiber content was high (in our case 30%). The optimization of the wood flour particle size could be a possible way of improving the toughness of the composites. In spite of this disadvantage, there is a wide range of WPC application.
The TGA in inert atmosphere reveals that the thermal stability of the treated wood flour (Wtr) was better than that of untreated wood (W) and its moisture content is lower. The SEM micrographs show that the thermal treatment did not damage the physical structure of the wood. The viscoelastic behavior and the mechanical properties demonstrate that the fibers act as effective reinforcing agents for PP. The W and Wtr provide rigidity to the composites; this effect was more pronounced in the case of Wtr PC. At the same time, a decrease of the elongation at break and impact energy was observed as compared with the polymer matrix. Finally, the high screw speed was efficient only for the WPC prepared with untreated fibers, which have the lowest affinity to the hydrophobic matrix.
Arbelaiz, A., Fernandez, B., Ramos, J.A., Mondragon, I. (2006) Thermal and crystallization studies of short flax fibre reinforced polypropylene matrix composites: Effect of treatments. Thermochim. Acta. 440:111–121.Google Scholar
Berger, M.J., Stark, N.M. (1997) Investigations of species effects in an injection-molding-grade, wood-filled polypropylene. The fourth international conference on woodfiber-plastic composites. pp. 19–25.Google Scholar
Bledzki, A.K., Faruk, O. (2004) Creep and impact properties of wood fibre-polypropylene composites: influence of temperature and moisture content. Compos. Sci. Technol. 64:693–700.CrossrefGoogle Scholar
Bledzki, A.K., Faruk, O. (2005) Effects of the chemical foaming agents, injection parameters, and melt-flow index on the microstructure and mechanical properties of microcellular injection-molded wood-fiber/polypropylene composites. J. Appl. Polym. Sci. 97:1090–1096.CrossrefGoogle Scholar
Chen, Y., Stark, N.M., Tshabalala, M.A., Gao, J., Fan, Y. (2014) Properties of wood-plastic composites (WPCs) reinforced with extracted and delignified wood flour. Holzforschung 68:933–940.Web of ScienceGoogle Scholar
Commereuc, S., Askanian, H., Verney, V., Celli, A., Marchese, P. (2010) About Durability of Biodegradable Polymers: Structure/Degradability Relationships. Macromol. Symp. 296:378–387.Google Scholar
Devi, R.R., Mandal, M., Maji, T.K. (2012) Physical properties of simul (red-silk cotton) wood (Bombax ceiba L.) chemically modified with styrene acrylonitrile co-polymer and nanoclay. Holzforschung 66:365–371.Web of ScienceGoogle Scholar
Dominkovics, Z., Danyadi, L., Pukanszky, B. (2007) Surface modification of wood flour and its effect on the properties of PP/wood composites. Composites Part A. 38:1893–1901.Google Scholar
Gardea-Hernandez, G., Ibarra-Gomez, R., Flores-Gallardo, S.G., Hernandez-Escobar, C.A., Perez-Romo, P., Zaragoza-Contreras, E.A. (2008) Fast wood fiber esterification. I. Reaction with oxalic acid and cetyl alcohol. Carbohydr. Polym. 71:1–8.CrossrefWeb of ScienceGoogle Scholar
Gasparovic, L., Korenova, Z., Jelemensky, L. (2010) Kinetic study of wood chips decomposition by TGA. Chem. Pap. 64:174–181.Google Scholar
Guyonnet, R., Bourgois, J. (1988) Procédé de fabrication d’un matériau ligno-cellulosique par traitement thermique et matériau obtenu par ce procédé. FR Patent 2 604 942 (A1).Google Scholar
Hosseinaei, O., Wang, S., Enayati, A.A., Rials, T.G. (2012) Effects of hemicellulose extraction on properties of wood flour and wood-plastic composites. Composites Part A. 43:686–694.Google Scholar
Jaziri, M., May, J.F., Bourgois, J., Guyonnet, R. (1987) Matériaux Composites à Base de Bois Rétifié: Contribution à l’Etude des Propriétés Rhéologiques à l’Etat Fondu en Régime Stationnaire. Angew. Makromol. Chem. 155:67–81.Google Scholar
Joseph, K., Thomas, S., Pavithran, C. (1996) Effect of chemical treatment on the tensile properties of short sisal fibre-reinforced polyethylene composites. Polymer. 37:5139–5149.CrossrefGoogle Scholar
Keener, T.J., Stuart, R.K., Brown, T.K. (2004) Maleated coupling agents for natural fibre composites. Composites Part A. 35:357–362.Google Scholar
Kim, H.S., Yang, H.S., Kim, H.J., Lee, B.J., Hwang, T.S. (2005) Thermal properties of agro-flour-filled biodegradable polymer bio-composites. J. Therm. Anal. Calorim. 81:299–306.CrossrefGoogle Scholar
Kumar, V., Tyagi, L., Sinha, S. (2011) Wood flour–reinforced plastic composites: a review. Reviews in Chem. Eng. 27:253–264.Google Scholar
Lee, C.-H., Hung, K.-C., Chen, Y.-L., Wu, T.-L., Chien, Y.-C., Wu, J.-H. (2012) Effects of polymeric matrix on accelerated UV weathering properties of wood-plastic composites. Holzforschung 66:981–987.Web of ScienceGoogle Scholar
Lu, J., Wu, Q., McNabb, H. (2000) Chemical coupling in wood fiber and polymer composites: a review of coupling agents and treatments. Wood Fiber Sci. 32:88–104.Google Scholar
Mihai, B., Cornelia, V. (2009) Thermal degradation of lignin-A review. Cellulose Chem. Technol. 44:353–363.Google Scholar
Mohanty, A.K., Drzal, L.T., Misra, M. (2002) Novel hybrid coupling agent as an adhesion promoter in natural fiber reinforced powder polypropylene composites. J. Mater. Sci. Lett. 21:1885–1888.CrossrefGoogle Scholar
Mwaikambo, L.Y., Ansell, M.P. (1999) The effect of chemical treatment on the properties of hemp, sisal, jute and kapok for composite reinforcement. Angew. Makromol. Chem. 272:108–116.Google Scholar
Nygard, P., Tanem, B.S., Karlsen, T., Brachet, P., Leinsvang, B. (2008) Extrusion-based wood fibre-PP composites: Wood powder and pelletized wood fibres-a comparative study. Compos. Sci. Technol. 68:3418–3424.CrossrefWeb of ScienceGoogle Scholar
Pelaez-Samaniego, M.R., Yadama, V., Lowell, E. Amidon, T.E., Chaffee, T.L. (2013) Hot water extracted wood fiber for production of wood plastic composites (WPCs). Holzforschung 67:193–200.Web of ScienceGoogle Scholar
Peng, Y., Liu, R., Cao, J., Guo, X. (2014) Effects of vitamin E combined with antioxidants on wood flour/polypropylene composites during accelerated weathering. Holzforschung 69:113–120.Web of ScienceGoogle Scholar
Premalal, H.G.B., Ismail, H., Baharin, A. (2002) Comparison of the mechanical properties of rice husk powder filled polypropylene composites with talc filled polypropylene composites. Polym. Test. 21:833–839.CrossrefGoogle Scholar
Saiful Islam, M., Hamdan, S., Jusoh, I., Rezaur Rahman, M., Ahmed, A.S. (2012) The effect of alkali pretreatment on mechanical and morphological properties of tropical wood polymer composites. Mater. Des. 33:419–424.Google Scholar
Son, J., Kim, H.J., Lee, P.W. (2001) Role of paper sludge particle size and extrusion temperature on performance of paper sludge–thermoplastic polymer composites. J. Appl. Polym. Sci. 82:2709–2718.CrossrefGoogle Scholar
Wolcott, M.P., Englund, K. (1999) A technology review of wood-plastic composites. Proceeding 33th Int Particleboard Composites Material Symposium. pp. 13–5.Google Scholar
Yang, H.-S., Kim, H.-J, Son, J., Park, H-J., Lee, B.-J., Hwang, T-S. Rice-husk flour filled polypropylene composites; mechanical and morphological study. Compos. Struct. 2004;63:305–12.Google Scholar