Wood fiber/polyolefin composites

Composites: Part A 35 (2004) 321–326 www.elsevier.com/locate/compositesa

Wood fiber/polyolefin composites Susan E. Selkea,*, Indrek Wichmanb a

School of Packaging, 130 Packaging Building, Michigan State University, East Lansing, MI 48824, USA b Department of Mechanical Engineering, Michigan State University, East Lansing, MI 48824, USA

Abstract Composites containing recycled plastics and wood fiber offer an interesting combination of properties, as well as lower cost than competitive materials, especially those based on synthetic fibers. By permitting use of moderately contaminated recycled plastics rather than requiring the use of virgin resin, these materials provide an additional market for recycled plastics, thereby helping to reduce waste disposal burdens. Composites can also be fabricated using recycled wood fiber, such as recovered paper fiber, providing an additional market outlet for recovered paper and thus further waste diversion benefits. Wood fiber/polyolefin composites are often unable to take full advantage of the potential of the fiber reinforcement, due to poor adhesion between the polymer matrix and the fiber. Use of additives to improve adhesion between the fibers and matrix can significantly improve performance. q 2003 Elsevier Ltd. All rights reserved. Keywords: A. Wood; A. Recycling; E. Extrusion; Polyolefin

1. Introduction Plastics account for an increasing fraction of municipal solid waste (MSW) in the United States, as well as around the world. In 2000, the amount of plastics in MSW reached 24.7 million tons, comprising 10.7% of the waste stream by weight, and an even larger percentage by volume [1]. Low density polyethylene (LDPE) plus linear low density polyethylene (LLDPE) is the largest component, followed by high density polyethylene (HDPE), polypropylene (PP), polyethylene terephthalate (PET), and polystyrene (PS), as shown in Fig. 1. This creates a substantial amount of polyolefins, especially packaging, that can potentially be recovered for recycling. However, contamination of these materials with paper such as from labels on containers, sales receipts left in plastic bags, etc., is common. While the tensile strength of the recycled plastics is usually good, the materials generally have fairly low stiffness and creep resistance, limiting their use in a number of applications. If these recycled plastics are combined with wood or other natural fibers in a composite structure, the stiffness and creep resistance can be improved substantially. In addition, the presence of ‘extra’ wood fibers from paper contaminants would not be expected to produce any significant change in properties for these materials, since they already contain similar fibers. * Corresponding author. Tel.: þ 1-517-355-9580; fax: þ1-517-353-8999. 1359-835X/$ - see front matter q 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.compositesa.2003.09.010

Thermoplastic/wood composites have been known for many years. Historically, most of these used wood flour to produce filled plastics. The wood flour decreased the cost, but was not usually intended to improve the performance in any substantial way. More recently, the use of natural fibers to provide a reinforcing mechanism in thermoplastics has been of substantial interest. In fact, several companies now manufacture wood fiber/thermoplastic materials for use as synthetic lumber in applications such as decking and window frames [2,3]. A major issue in achieving true reinforcement from the incorporation of natural fibers into wood is the inherent incompatibility between the hydrophilic fibers and the hydrophobic polymers, which results in poor adhesion and therefore in poor ability to transfer stress from the matrix to the reinforcing fiber. A number of investigators have explored the ability of additives to enhance adhesion and thereby improve properties such as tensile and flexural strength of these composite materials. This paper summarizes some of the research involving composites with wood and paper fibers and polyolefins that has been carried out at Michigan State University over the last several years. 2. Sample preparation All composite structures were produced using Baker – Perkins co-rotating twin screw extruders. Fibers were

322

S.E. Selke, I. Wichman / Composites: Part A 35 (2004) 321–326

Fig. 1. Plastics in the US municipal solid waste stream, 2000 [1].

conditioned for at least 40 h at 23 ^ 2 8C and 50 ^ 5% RH before extrusion. For determination of mechanical properties, the extrudate was cut into approximately 15.25 cm lengths, placed in a frame, and compression molded in a Carver laboratory press, model M25 ton. Details of the extrusion and compression molding varied. These are summarized below, and described in full elsewhere [4 – 10]. Compression molded plates were conditioned for at least 40 h at 23 ^ 2 8C and 50 ^ 5% RH before testing. For tensile testing, plate dimensions were 15 £ 15 £ 0.25 cm3. Samples were cut into strips 1.9 cm in width and 14 cm in length, and then machined into an ASTM Type I dumbbell configuration, using a Tensilkut Model 10-13 system. For impact testing, plate dimensions were 12.7 £ 12.7 £ 0.3175 cm3. Samples were cut into pieces 1.27 cm in width and 6.35 cm in length, and then notched using a TMI Notching Cutter, model 22-05. The notch angle was 22.5 ^ 0.58, and notch depth was 0.254 cm. All data reported here was for samples cut in the extrusion direction (machine direction). Mechanical testing was carried out in accordance with ASTM standard methods, D638 for tensile properties and D256 for impact resistance. Impact testing used a TMI 43-1 Izod Impact Tester. The majority of the tensile testing was done on a United Calibration Corp. Model SFM-20 tensile tester.

The bottles were blown, filled with milk, sent through the usual distribution cycle, and purchased at a retail outlet. After the milk was consumed, the bottles were stored for several days with some residual milk inside, to ensure maximum exposure to the product. The containers were then rinsed, washed, and ground into flakes, using a low-line granulator model 68-913 from Polymer Machinery Corp. Labels and adhesive were not included. Composites were produced using spruce fibers obtained from Masonite Corp., which were air-dried before use. A Baker –Perkins Model MPC/V-30 DE 38 mm, 13:1 corotating intermeshing twin-screw extruder was used to prepare the composites. Fibers were hand-fed into the second-stage feed port. The extruder and die were maintained at 150 8C, and extruder speed was set at 150 rpm. Compression molding was carried out at 150 8C, with pressure gradually increased to 30,000 lbs and then held for 10 min. Results for tensile strength and tensile modulus of the composites are compared in Fig. 2 [4]. As can be seen, performance of composites made from recycled HDPE was at least as good as that of composites made from virgin HDPE. In most cases, differences between the recycled and virgin matrix composites were not statistically significant. Comparisons of properties of the recycled and virgin HDPE used in production of the composites are shown in Table 1 [11]. The performance of composites made from mixed HDPE bottles obtained from a local recycling dropoff facility was also evaluated. While caps and fitments (e.g. detergent spouts) were removed, labels were not, so the ‘resin’ fraction contained labels and label adhesives in the proportions normally present in a mixed HDPE bottle stream. The bottles were rinsed with warm water, dried, and chipped, but not otherwise cleaned or processed. It should also be noted that the majority of these bottles

3. Comparison of virgin and recycled plastic as matrices As indicated, part of the motivation for making natural fiber/thermoplastic composites is to make useful materials from waste plastics. An obvious question that arises is whether use of recycled plastics rather than virgin plastics results in decreased performance in such composite materials. To evaluate this, the performance of HDPE/wood fiber composites made with virgin resin (Fortiflex A60-70-119, Soltex Polymer Corp.) was compared with composites made from the identical resin that had been used for production of milk bottles.

Fig. 2. Comparison of tensile strength and modulus of composites of virgin HDPE and recycled HDPE with wood fiber [4].

S.E. Selke, I. Wichman / Composites: Part A 35 (2004) 321–326

323

Table 1 Properties of virgin HDPE and recycled HDPE from milk bottles [11] Property

Melt index (g/10 min) Tensile strength (kPa) Tensile modulus (kPa) Elongation at yield (%) Elongation at break (%)

Virgin HDPE

Recycled milk bottles

Mean

Std. dev.

Mean

Std. dev.

0.727 33,300 5.87 £ 105 17 71.9

0.021 740 0.62 £ 105 1.1 17.0

0.715 34,400 6.60 £ 105 17 33.8

0.012 1360 0.85 £ 105 1.0 10.6

contained approximately 25% post-consumer recycled content [9]. Composites with aspen hardwood fibers (Abitibi Corp.) manufactured through thermomechanical pulping were produced in a Baker – Perkins Model ZSK-30, 30 mm, 26:1 co-rotating twin-screw extruder. The compounding speed was set at 120 rpm, and temperature maintained at 150 8C. Suitable lengths of extrudate were compression molded into test plates at 150 8C, as described above. While it was not possible to compare performance of the recycled material (a mix of different unidentified HDPE homopolymers and copolymers) directly with that of virgin material, the composites did exhibit useful tensile properties, as shown in Fig. 3. It should be noted that the standard deviation of the tensile modulus results was quite high, so they should be interpreted with caution. Because recycled polymers, containing as they do a mix of resins from different producers, are inherently more variable in properties than single-resin virgin polymers, most of our investigations have used virgin polymers as the matrix to facilitate comparisons. Incorporation of wood fibers tended to decrease elongation at break and impact strength of HDPE, as shown in Fig. 4 for composites made with virgin HDPE milk bottle resin [4]. Impact strength and elongation for composites made with mixed post-consumer HDPE bottles are shown in Fig. 5 [9]. Interestingly, incorporation of up to 40% wood fiber increased the impact strength of these materials, and impact strength exceeded that of comparable composites from virgin HDPE milk bottle resin. The reason for this effect is not fully understood, but may be related to

Fig. 3. Tensile strength and modulus of composites of mixed recycled HDPE bottles with wood fiber [9].

Fig. 4. Impact strength and elongation at break of composites of wood fiber and virgin HDPE milk bottle resin [4].

incompatibility between various HDPE resins, resulting in very poor impact strength for the mixed HDPE without wood fiber.

4. Composites with plastic blends When HDPE bottles are collected and recycled, they commonly are contaminated to some degree with other resins, especially polypropylene, which is present in caps and labels. Effective separation of HDPE and PP from each other is very difficult and generally not economical. Some end-users are forced to limit the amount of pigmented HDPE bottles they use in their product mix specifically because pigmented bottles tend to have larger amounts of PP than unpigmented bottles. With the increasing tendency to focus plastics recycling collection on ‘all plastic bottles’ rather than HDPE and PP alone, it is desirable to be able to use these other containers, rather than having to discard them. We have found that composites with wood fiber can tolerate contamination of HDPE with PP. Fig. 6 shows the machine direction tensile strength of HDPE/PP blends with 40% wood fiber. The PP (Pro-Fax 7823, Himont) was a low melt flow copolymer (0.5 g/ 10 min, density 0.897 g/cm3). The HDPE (AD 60-007, Paxon) was a homopolymer with a medium molecular

Fig. 5. Impact strength and elongation at break of composites of wood fiber and mixed postconsumer HDPE bottles [9].

324

S.E. Selke, I. Wichman / Composites: Part A 35 (2004) 321–326

Fig. 6. Tensile strength of composites of HDPE/PP blends with 40% wood fiber [7].

weight distribution, melt index 0.7 g/10 min, density 0.960 g/cm3. Composites with aspen hardwood fiber (Abitibi Corp.) were prepared in a Baker – Perkins Model ZSK-30, 30 mm, 26:1 co-rotating twin-screw extruder. Temperature was set at 155 8C in zones 3 to 6, and screw speed was set at 100 rpm. Zones 1 and 2 were set at either 150 or 180 8C, in order to compare the effects of processing above and below the PP melt temperature. Wood fiber was incorporated at port 3. Compression molding was carried out at 160 8C and 25,000 psi for 15 min. The tensile strength was affected by the relative concentrations of HDPE and PP, but remained acceptable across the entire range of concentrations. If concentrations of the PP were high, it was necessary to process at a higher temperature than was required for HDPE, as was expected. The higher temperature also appeared to be advantageous for the pure HDPE, although the differences were not statistically significant [7]. Composites made from blends of low-density polyethylene and polypropylene were also investigated. An extrusion and injection-molding grade PP homopolymer was used (Inspire H704-04, Dow Plastics), with a density of 0.9 g/cm3 and a melt index of 4 g/10 min. The LDPE (133A, Dow Plastics) was chosen to represent recycled material from grocery or merchandise bags, and had a density of 0.923 g/cm3 and a melt index of 0.22 g/10 min. Composites with aspen hardwood fiber (Abitibi Corp.) were prepared in a Baker – Perkins Model ZSK-30, 30 mm, 26:1 co-rotating twin-screw extruder. Temperature was set at 150 8C in zones 3 to 6, and screw speed was set at 100 rpm. Temperature in zones 1 and 2 was set at either 150 or 180 8C, as in the study described above, in order to compare the effects of processing above and below the presumed PP melt temperature. Wood fiber was incorporated at port 2, after the plastic resin was melted. Compression molding was carried out at 150 8C and 30,000 psi for 5 min. In contrast to the results with the lower flow PP and HDPE, successful composites with 40% wood fiber were produced across the range of blend compositions at both low and higher processing temperatures, as shown in Fig. 7 [10]. It is probable that this copolymer PP had a lower melting temperature than the one used previously, but the precise

Fig. 7. Tensile strength of composites of LDPE/PP blends with 40% wood fiber [10].

melting temperature was not determined. Differences in tensile strength between composites produced at the two sets of temperatures were not statistically significant, in most cases. Multilayer containers made with different resins pose a challenge for recycling. However, in the case of bottles used for ketchup and containing a random copolymer PP, ethylene vinyl alcohol (EVOH) and a proprietary tie layer (adhesive), using the mixed resin did not detract at all from performance, as shown in Fig. 8 [8]. These composites were manufactured from the pure PP used in the bottles (Fortilene 4104, Solvay) and from regrind material which contained 94.5% PP (by weight), 3.75% EVOH (EVAL Solarnol DC, EVALCA), and 1.75% adhesive (Admer, Mitsui Monoply MT38). Composites with aspen hardwood fibers (Lionite Hardboard, Phillips, WI) manufactured through thermomechanical pulping were produced in a Baker – Perkins Model ZSK30, 30 mm, 26:1 co-rotating twin-screw extruder. The compounding speed was set at 100 rpm, and temperature maintained at 185 8C. Suitable lengths of the extrudate were compression molded at 185 8C and 30,000 psi for 15 min.

5. Additives It is well established that compounding wood or other natural fibers with polyolefins results in poor adhesion

Fig. 8. Tensile strength of composites made from multilayer PP ketchup bottles and from pure PP resin [8].

S.E. Selke, I. Wichman / Composites: Part A 35 (2004) 321–326

325

Fig. 11. Tensile strength of composites of HDPE with 35% recovered paper fiber [6]. Fig. 9. Effect of MAPP on tensile strength of composites of HDPE with 30% wood fiber [5].

between the fibers and the matrix. A number of investigators have examined the potential for improving properties by improving the adhesion between the reinforcing fibers and the matrix. The most promising additives tend to incorporate polar groups in their structures. In fact, the composites made from multilayer ketchup bottles shown in Fig. 8 may have benefited from the polar groups present in the EVOH and in the adhesive tie layer resin. One material that can improve the performance of HDPE/wood fiber composites is maleic anhydride modified polypropylene (MAPP). Fig. 9 shows the tensile strength and Fig. 10 the tensile modulus of composites with 30% wood fiber containing 2 and 5% MAPP (Hercoprime, Himont). The HDPE used was obtained from mixed recycled milk bottles. Aspen hardwood fibers (Canfor Canadian Forest Products) were blended with the granulated mixed dairy bottles (20% unused bottles and 80% postconsumer bottles) in a Baker – Perkins Model MPC/V30 DE, 38 mm, 13:1 co-rotating twin screw extruder. The extruder and die were maintained at 150 8C, and speed was 150 rpm. Wood fibers were fed into zone 2 of the extruder. Test samples were compression molded at 150 8C and 30,000 psi for 10 min. The differences in tensile strength between the 100% recycled HDPE and composites containing 2 and 5% MAPP

Fig. 10. Effect of MAPP on tensile modulus of composites of HDPE with 30% wood fiber [5].

were not statistically significant, while differences between the 100% recycled HDPE and the composite with 5% MAPP were significantly different from the composite with no additive at an alpha level of 0.05. Differences in average tensile modulus between composites with and without MAPP were not statistically significant, but they did differ from the pure recycled HDPE [5].

6. Recycled fibers Composites can also be made with various plastics and other types of fiber. One obvious choice for a composite made from recycled plastic is recycled fiber. Composites of HDPE (AD60-007, Exxon Chemical Corp.) and recovered newspaper fiber (Interfibe) were prepared in a Baker – Perkins Model ZSK-30, 30 mm, 26:1 co-rotating twinscrew extruder. The temperature was set at 163 8C in zone 1, 164 8C in zone 2, 165 8C in zone 3, 172 8C in zone 4, 166 8C in zone 5, 165 8C in zone 6, and 183 8C in the die. Fibers were fed into zone 3. After extrusion, test samples were compression molded at 165 8C and 30,000 psi for 15 min. The tensile properties of composites made with HDPE and 35% recovered paper fiber are similar to those of composites made from virgin wood fiber, as shown in Fig. 11. Addition of maleic anhydride modified high-density polyethylene (MAHDPE) (Polybond 3009, Uniroyal) resulted in improved adhesion between the matrix and

Fig. 12. Impact strength of composites of HDPE with 35% recovered paper fiber [6].

326

S.E. Selke, I. Wichman / Composites: Part A 35 (2004) 321–326

fibers. Tensile strength of composites containing 2 and 6% MAHDPE was significantly higher than that of the composite without additive. Differences in impact properties, shown in Fig. 12, were not statistically significant at the 5% confidence level [6].

Acknowledgements The work described involved the efforts of these graduate students in the School of Packaging: Varsha Kalyankar, Kristine Nieman, Chate Pattanakul, Jonathan Ricciardi, Sunetra Rojanarungtawee, Rodney Simpson, Nualrahong Thepwiwatjit, and Kobdaj Vanichvarod.

References [1] U.S. Environmental Protection Agency, Municipal Solid Waste in The United States: 2000 Facts and Figures. 2002. Washington, DC.

[2] Doba J. Plast News 2001;Feb 26:4. [3] Bregar B. Plast News 2001;Feb 19:1. [4] Kalyankar V. Mechanical characteristics of composites made from recycled HDPE obtained from milk bottles. MS Thesis, Michigan State University; 1989. [5] Nieman KA. Mechanical property enhancement of recycled high density polyethylene and wood fiber composites due to the inclusion of additives. MS Thesis, Michigan State University; 1989. [6] Ricciaradi J. High density polyethylene/paper fiber composites: measuring the impact of additives on their physical and mechanical properties. MS Thesis, Michigan State University; 1999. [7] Rojanarungtawee S. Composite of wood fiber and mixed recycled thermoplastics. MS Thesis, Michigan State University; 1998. [8] Simpson R. Composite materials from recycled multi-layer polypropylene bottles and wood fibers. MS Thesis, Michigan State University; 1991. [9] Thepwiwatjit N. Composite of wood fiber and recycled HDPE bottles from household use. MS Thesis, Michigan State University; 2000. [10] Vanichvarod K. Composite material of mixed low density polyethylene/polypropylene and wood fiber. MS Thesis, Michigan State University; 2002. [11] Pattanakul C. Characteristic changes in recycled HDPE from milk bottles. MS Thesis, Michigan State University; 1987.