NC State
BioResources
Hong, H., Liu, H., Zhang, H., He, H., Liu, T., and Jia, D. (2015). "Blossom morphology and correlative performance improvement of recycled polyethylene/wood flour composites with steam-activated interfaces," BioRes. 10(4), 7817-7833

Abstract

Interfacial compatibility plays a key role in the performances of natural fiber-reinforced composites. The measures commonly used to improve the interfacial compatibility focus more on the addition of various compatibilizers than on the structural modification of the natural fiber. In this paper, an attempt was made to enlarge the interfacial interaction areas of the recycled polyethylene (rPE)/wood flour (WF) composites by steaming the WF. Multi-monomer graft copolymers of polyethylene (GPE) were used as compatibilizers for the composites. How the enlarged interfaces affected the morphology, mechanical properties, water resistance, thermal stability, and dynamic rheological properties of the rPE/WF composites was investigated. The steaming process was able to enlarge the voids of the WF and therefore activate more interfaces for interactions. It was found that the interfacial morphology of the composites was affected by the degree of interfacial compatibility of the composites and so was characterized by various distinctive blossom shapes having a variation of compositions. With the help of GPE, the steaming process was able to significantly improve the interfacial compatibility of the composites and therefore improve the mechanical properties, water resistance, thermal stability, and dynamic rheological properties of the composites.


Download PDF

Full Article

Blossom Morphology and Correlative Performance Improvement of Recycled Polyethylene/Wood Flour Composites with Steam-Activated Interfaces

Haoqun Hong,a,b Hao Liu,a Haiyan Zhang,a,b,* Hui He,c Tao Liu,d and Demin Jia c

Interfacial compatibility plays a key role in the performances of natural fiber-reinforced composites. The measures commonly used to improve the interfacial compatibility focus more on the addition of various compatibilizers than on the structural modification of the natural fiber. In this paper, an attempt was made to enlarge the interfacial interaction areas of the recycled polyethylene (rPE)/wood flour (WF) composites by steaming the WF. Multi-monomer graft copolymers of polyethylene (GPE) were used as compatibilizers for the composites. How the enlarged interfaces affected the morphology, mechanical properties, water resistance, thermal stability, and dynamic rheological properties of the rPE/WF composites was investigated. The steaming process was able to enlarge the voids of the WF and therefore activate more interfaces for interactions. It was found that the interfacial morphology of the composites was affected by the degree of interfacial compatibility of the composites and so was characterized by various distinctive blossom shapes having a variation of compositions. With the help of GPE, the steaming process was able to significantly improve the interfacial compatibility of the composites and therefore improve the mechanical properties, water resistance, thermal stability, and dynamic rheological properties of the composites.

Keywords: Polymer-matrix composites; Recycling; Polymers; Wood; Interface; Graft copolymers

Contact information: a: School of Materials and Energy, Guangdong University of Technology, Guangzhou 510006, China; b: Guangdong Provincial Key Laboratory of Functional Soft Condensed Matter, Guangdong University of Technology, Guangzhou 510006, China; c: College of Materials Science and Engineering, South China University of Technology, Guangzhou 510640, China; d: Product Research & Development Center, Guangzhou LESCO-KINGFA WPC Technology Co., Ltd, Guangzhou 510520, China; *Corresponding author: hyzhang@gdut.edu.cn

INTRODUCTION

With the ongoing energy crisis and environmental deterioration, the recycling and reuse of wastes and by-products have become important measures for conserving resources and protecting the environment (Thompson et al. 2009; Najafi 2013). Preparing specific, useful products is one of the most effective means of recycling and reusing these wastes and by-products. Among these useful products, wood-plastic composites (WPC) are one of the most attractive because they provide a promising use for both recycled plastics and forestry and agricultural by-products in the preparation of valuable composites by means of conventional polymer processing and fabricating machines such as the extruder (Guo and Wang 2008; Wu et al. 2014) and intensive mixer (Fang et al. 2013). The polymers most frequently used to prepare WPC are thermoplastics such as polypropylene (PP) (Gao and Wang 2008), polyethylene (PE) (Ou et al. 2010; Xu et al. 2010; Gao et al. 2012; Wu et al. 2014; Zhou et al. 2014), polyvinyl chloride (PVC) (Xu et al. 2010; Fang et al. 2013), and polystyrene (PS) (Nair and Thomas 2003; Mansour et al. 2006). The by-products of preparing WPC are natural cellulosic fillers such as wood flour (WF), bamboo flour, oil palm fiber, sisal fiber, rice chaff, and bagasse (Ashori 2008). Thus, WPC has the advantages of ease of manufacture, environmental safety, biodegradability, low cost, and high strength. It has been widely applied in the construction, automotive, packaging, and other industries (Ashori 2008; Thompson et al. 2009; Najafi 2013).

However, most polymers used for WPC are hydrophobic, and natural fillers are hydrophilic. The hydrophobic polymers are weakly compatible with the hydrophilic natural fillers at their interfaces (Bledzki and Gassan 1999; Guo et al. 2013). These weakly compatible interfaces result in the low performance of WPC. To successfully prepare a high-performance WPC, the interfacial compatibility between the polymers and natural fillers has to be addressed. The interfacial compatibility of WPC has been improved by modifying natural fillers with acetylation (Zafeiropoulos et al. 2002; Mbarek et al.2013), alkali (Aziz and Ansell 2004), organosilane (Demir et al. 2006), or using compatibilizers (Kazayawoko et al. 1999; Artzi et al. 2003; Manchado et al. 2003; Bengtsson et al. 2007; Danyadi et al. 2007; Guo and Wang 2008; Ashori and Nourbakhsh 2009; Lashgari et al. 2011; Gao et al. 2012; Hoang et al. 2013; Zhou et al. 2014).

Of those methods, using compatibilizers such as graft copolymers, which chemically and physically enhance the interfacial interactions between the polymers and natural fibers, has been shown to be an effective method for improving the interfacial compatibility (Bledzki and Gassan 1999; Li et al. 2007; Guo and Wang 2008; Lashgari et al. 2011; Gao et al. 2012). Moreover, using graft copolymers such as the compatibilizers is a convenient and attractive method of preparing WPC in that it can facilitate the WPC preparation procedure by directly blending the natural fibers with the polymer matrix and compatibilizers. The most popular graft copolymers for use with WPC are those such as maleated polyolefins (Kazayawoko et al. 1999; Manchado et al. 2003; Demir et al. 2006; Gao et al. 2012; Zhou et al. 2014), because these graft copolymers can facilitate strong interfacial interactions between the polymers and natural fibers by means of the esterification reaction between maleic anhydride (MAH) and the hydroxyl groups of natural fibers (Guo et al. 2013). However, most WPC is filled with more than 50 wt.% natural fibers in order to reduce the cost of WPC. Higher filler contents tend to worsen the performance of WPC and challenge the compatibilizing capability of compatibilizers. The graft degrees of maleated polyolefins are not high enough to confer efficient interfacial interactions to a highly filled composite because MAH is difficult to homo-polymerize at higher temperature (Li et al.2001). To maintain acceptable performance levels for WPC, attempts have been made to improve the compatibilizing capability of compatibilizers, by such methods as the in situ reactive compatibilization (Li et al. 2012), synergistic compatibilization (Gao et al. 2012; Yeh et al. 2013), and use of a multi-monomer graft copolymer (Jia et al. 2000; Ding et al. 2005; Hong et al. 2009; Hong et al. 2011). The multi-monomer graft copolymer is one of the most efficient compatibilizers, and is prepared by grafting MAH with one or two co-monomers onto the melting or solid state polyolefin so as to increase the graft degree of MAH and thereby improve the compatibilizing capability of the graft copolymer (Ding et al. 2005; Hong et al. 2009; Hong et al. 2011) through chemical or physical interaction at the interfaces (Bledzki and Gassan 1999).

However, the interfacial compatibility and performance of WPC are associated not only with the use of compatibilizers but also with the structure of the natural fibers (Liu et al. 2014; Ou et al. 2014a,b). Ou et al. (2014b) studied the influence of the removal of hemicelluloses and lignin (together or respectively) from wood flours on the mechanical properties, dimensional stability, and rheological properties of wood particle/high density polyethylene composites. They found that these performances were influenced by the cell wall composition of the wood particles (Ou et al. 2014a,b). Moreover, it was found that high filler content usually causes the aggregation of natural fibers, such as wood particles, in the WPC (Danyadi et al. 2007). The aggregated natural fillers usually have gaps among them, in addition to the cavities and capillaries on their coarse surfaces, which originate from their grinded cellular structure. These voids cannot be easily accessed by the polymer chains without any modification to the natural fibers (Ou et al. 2014b). WPC can be improved not only by improvements in the interfacial interactions at the coarse surfaces of the natural fillers, but also by improvements in the interfacial interactions at these voids. This means that the interfacial interactions of WPC could either be enhanced by improving the infiltration of compatibilizers into these voids or by enlarging the voids for the fillers. In order to enhance the interfacial interactions at these minimally accessible locations, alkali and glycerin solutions have been used to pre-treat and modify natural fibers. This pre-treatment method has improved the interfacial compatibility of WPC (Aziz and Ansell 2004; Luo et al.2014). Recently, the compounds of two complementary multi-monomer graft copolymers with improved infiltrating capability were used to compatibilize the recycled polyethylene (rPE)/WF composites and achieve satisfactory improvements in performance (Hong et al. 2014). However, whether it is possible to enhance the interfacial interactions by manipulating the interfacial interaction areas of the natural fibers without changing the infiltrating capability of the compatibilizers, such that the compatibilizers can easily infiltrate into the voids of natural fibers, remains unchallenged in the literature.

In this paper, an attempt was made to enlarge the interfacial interaction areas of WPC by steaming the wood flours. This enlarged the voids of the steamed wood flours (SWF), into which the compatibilizer could easily infiltrate. The infiltrated voids significantly enhanced the interfacial interactions. How the enhanced interfacial interactions affected the morphology, mechanical properties, water resistance, thermal stability, and rheological properties of rPE/WF composites was investigated.

EXPERIMENTAL

Materials

Recycled polyethylene (rPE), with a melting flow rate (MFR) of 2.12 g/min, was supplied by Dongguan Dada Co., Ltd and used as received. The WF was obtained from the aspen (Populus tremuloides) wood that was provided by Jiangmen Weihua Co., Ltd. The steamed wood flour (SWF) was prepared by steaming WF in a boiling pot for 1 h. The ternary graft copolymers of polyethylene (GPE) were used as the compatibilizers for rPE/WF composites. GPE with a 10.5% graft degree was synthesized via solid phase grafting of MAH, MMA, and BA (the mass ratio of the MAH/MMA/BA was 6/4/2) onto powdered PE initiated by benzoyl peroxide (BPO) at 120 °C (Jia et al. 2000).

Methods

Preparation of rPE/WF composites

The WF-filled rPE composites were prepared on a two-roll mill whose roll diameter was 16 cm. While the rolls were being heated, rPE was first added onto the running rolls and plasticized for 3 min. GPE and WF were then added successively onto the melted rPE and mixed for 5 min. The loadings of GPE and the fillers were 7 phr (parts per hundred rPE in weight) and 50 wt.%, respectively. The melted composites were compressed into 250 250 4 mm plates on a 25-ton flat vulcanizer. The plates were cooled at room temperature for 24 h before tailoring and testing.

Measurements

The tensile and flexural tests were carried out on a universal testing machine (DCS-5000, Shimadzu, Japan) at room temperature according to ISO527-2: 1993 (with dumbbell sample dimensions of 170 10 4 mm) and ISO178: 2001 (with sample dimensions of 80 10 4 mm). A non-notched impact test was performed according to ISO179-1eA, with the sample dimensions of 80 10 4 mm. The particle size distribution of WF was characterized on a Beckman-Coulter N4 Plus particle size analyzer (USA). After the steaming process and after the drying, the SWF was fully crumbed, with the aim of retaining the original shape of the SWF as much as possible. The WF and SWF were characterized in a dried state. Fourier transform infrared spectroscopy (FTIR) was carried out on a Nicolet 710 FTIR Spectrometer. For purifying purposes, the GPE were extracted using acetone in a Soxhlet extractor for 24 h and dried at 60 °C until they had reached constant weight. Before testing, the specimens were compression-molded into tablets using KBr. In order to study the fracture surface morphology of the impacted specimens, the surface was sputtered with a thin layer of gold, and then scanning electron microscopy (SEM) was carried out on an S-550 SEM instrument made by Hitachi Co., Ltd., Japan. Differential Scanning Calorimetry (DSC) was carried out on a TA DSC 2910 by heating the sample to 200 °C at calefactive rates of 10 °C/min. The endothermic flows were recorded as a function of temperature. Thermogravimetric analysis (TGA) was carried out on a TA TGA 2050 instrument by heating the sample to 600 °C at calefactive rates of 20 °C/min. The weight loss percentage was recorded as a function of temperature. Vicat Softening Temperature (VST) analysis was carried out on RW-3 VST testing equipment made by Hebei Chengde Testing Machinery Inc., China. The water resistance was estimated by the water absorption (WA) as a function of time. Before testing the WA, the samples were dried at 60 °C for 24 h, transferred to a dried ambience for cooling, and weighed to obtain their original weights (W0). Then, the samples were immersed into water at ambient temperature. After a certain immersion time, the samples were weighed, and the immersed weights of the samples (Wt) were obtained. WA was determined by WA = (WtW0)/W0×100%. The WA testing lasted for 60 days. A dynamic rheological test was carried out on a RDA III Rheometmer (Rheometric Science Inc., USA). The dynamic rheological behavior was tested on a pair of circular plates with a diameter of 25 mm at 210 °C. The gap between the two plates was set at 2.0 mm. The scanning frequency was 0.1 to 500 rad/s.

RESULTS AND DISCUSSION

Characterization of Wood Flours

Figure 1 and Table 1 demonstrate the effect of the steaming process on the particle sizes of WF. 93.5% of the WF particles were smaller than 100 m, while only 62.2% of the SWF particles were smaller than 100 m. 78.6% of the WF particles and 44.3% of the SWF particles were smaller than 50 m. Note that all the WF particles were smaller than 400 m, while there were some SWF particles larger than 400 m but smaller than 1000 m. The particle size characterization made it clear that the WF particles had been significantly enlarged by the steaming process.

Table 1. Particle Size Distribution and Weight Change in Wood Flour after Processing

Because the particle size of the WF had been enlarged by the steaming process, the question to be further investigated was whether the steaming process caused any distinct change to the structure of the WF. The FTIR characterization helped to disclose these results, as shown in Fig. 2a. There were no obvious spectra differences between WF and SWF, except for the spectra in the range from 1800 cm-1to 1000 cm-1, which were attributed to primary compositions such as cellulose, hemicelluloses, and lignin (Ou et al. 2014b). In the Ou et al. (2014b) report, wood particles were treated to remove cellulose, hemicelluloses, and lignin, respectively. FTIR was used to characterize the variation in wood compositions. The wood particles extracted with individual composition showed peaks varying from 1800 cm-1 to 1000 cm-1. In this research, the peaks for the SWF ranging from 1800 cm-1 to 1000 cm-1, as shown in Fig. 2b, were wider than those of WF, as shown in Fig. 2a. Meanwhile, the peaks of SWF tended to shift to lower frequencies than those of WF. It is suggested that the primary compositions of WF were changed by the steaming process. The possible reason of the broadening and shifting of IR spectra is because of the increased hydrogen bonding in WF that was caused by water absorption and the changed structure of WF being opened to facilitate the interactions of polymers. However, the question to be further investigated was whether the steaming process really promoted the movement of primary components within the SWF. Figure 2b compares the DSC curves for WF and SWF. This curve was attributed to the glass transition enthalpy of the primary compositions in WF (Backman and Lindberg 2001). In the reports of Backman (Backman and Lindberg 2001) and Kelley (Kelley et al.1987), both the DSC and dynamic mechanical thermal analysis (DMTA) results showed that wood has an α-peak from 30 °C to 60 °C, which represents the glass transition peak of primary compositions. These reports helped to identify the glass transition peak of the primary compositions that appeared at 36.2 °C in WF under natural state, as indicated in Fig. 2b. As depicted in Fig. 2b, the endothermic enthalpy of SWF (9.41 J/g) was higher than that of WF (7.25 J/g). The results made it clear that the chain segments of primary compositions in SWF were able to move more easily than those in WF. The steaming process is a popular method for drying wood (Ishikawa et al. 2004). Thus, the steaming process should be also an effective method for drying WF. This can be seen from Table 1. Table 1 compares the steaming process with the baking process in terms of their effects on the weight change of WF before and after these drying processes. The weight change for the WF following the baking process was only 0.0671 g, while the weight change for the SWF following the steaming process was as high as 1.9433 g. The baking process hardly changed the weight of WF, while the steaming process noticeably changed the weight of SWF. The reason for this difference was that the baking process is unable to release the water trapped inside the WF cell walls, whereas the steaming process can release the water trapped inside the SWF cell walls (van Meel et al. 2011). The above results made it clear that the steaming process enlarged the voids in the SWF, activated the primary compositions in the SWF, and left the SWF drier than it left the WF, as is favorable for enhancing the interfacial interactions of rPE/SWF composites.

Fig. 1. The particle size change of wood flour (a) and steamed wood flour (b)

Morphology of rPE/WF Composites

Once the structural differences between WF and SWF had been clarified, a visualization of the correlative morphology of rPE/WF composites, manifesting as three lotus flowers at different florescence stages, was undertaken. Without compatibilizers, it was difficult for the polymer matrix to interact with the WF; thus, the rPE/WF composites were weakly compatible. The weakly compatible rPE/WF composites showed obvious interface fracture, and the voids could be seen on the WF surface. This weakly compatible interface of the composites resembled a fully blooming lotus, as shown in Fig. 3a. The WF particles corresponded to the seedpod of the lotus. The polymer matrix corresponded to the leaves of the lotus.

Fig. 2. The FTIR (a) and DSC (b) of wood flour

The seedpod-like WF particles broke away from the leaf-like polymer matrix. When GPE was used, GPE promoted the interaction of the polymer matrix with the WF. As shown in Fig. 3b, it was easier for the polymers to infiltrate into the voids and enwrap the surface of WF with the help of GPE. The interface of the composites did not show any distinct voids. It appeared that GPE had improved the interfacial interactions of the composites and therefore improved the interfacial compatibility. This compatible interface resembled a half-blooming lotus. The seedpod-like WF particles firmly interacted with the leaf-like polymer matrix.

When the WF was steamed, the voids of WF were enlarged. As shown in Fig. 3c, it was easier for the polymers to infiltrate into these voids and enwrap the SWF particles tightly. Therefore, the interfacial interactions were further enhanced. The interface of the composites resembled a budding lotus. The seedpod-like WF particle interacted with the leaf-like polymer matrix more firmly. These results revealed that rPE/SWF/GPE composites had better interfacial compatibility than did the other tested rPE/WF composites.

合辑-1-2015-09-10-可用