- Email: [email protected]
Reprocessing of wood fibre reinforced polypropylene composites. Part I: Effects on physical and mechanical properties
Composites: Part A 39 (2008) 1091–1100
Contents lists available at ScienceDirect
Composites: Part A journal homepage: www.elsevier.com/locate/compositesa
Reprocessing of wood fibre reinforced polypropylene composites. Part I: Effects on physical and mechanical properties M.D.H. Beg, K.L. Pickering * Department of Engineering, University of Waikato, Private Bag 3105, Hamilton, New Zealand
a r t i c l e
i n f o
Article history: Received 9 May 2007 Received in revised form 21 April 2008 Accepted 22 April 2008
Keywords: A. Natural fibre composites E. Reprocessing B. Fibre length A. Coupling agent
a b s t r a c t This paper describes the effects of reprocessing on the physical and mechanical properties of composites based on radiata pine (Pinus radiata) fibre in a polypropylene (PP) matrix. Composites, containing either 40 wt% or 50 wt% fibre with 4 wt% maleated polypropylene (MAPP) as a coupling agent, were reprocessed up to eight times. For composites with 40 wt% fibre, tensile strength (TS) and Young’s modulus (YM) were found to decrease with increased reprocessing by up to 25% for TS and 17% for YM (after reprocessed eight times). Flexural tests were also carried out for 40 wt% fibre composites and flexural strength and modulus were found to decrease with increased reprocessing. Although, TS was lower for virgin composites with 50 wt% fibre than for those with 40 wt% fibre, this initially increased with reprocessing by up to 14% (after reprocessed two times), but then decreased with further reprocessing, and an overall 11% reduction of TS was found after reprocessing eight times compared to the virgin composites. YM was higher for virgin composites with 50 wt% than those with for 40 wt% fibre and also initially increased with reprocessing but decreased upon further reprocessing. Reprocessing was found to increase thermal stability. The TS of composites made by combining reprocessed with virgin materials was also assessed. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction In recent decades, growing environmental awareness has resulted in renewed interest in the use of natural materials for different applications. Increasingly more stringent environmental policies have forced industries such as the automotive, packaging and construction industries to search for new materials that can substitute traditional composite materials consisting of a plastic matrix and inorganic reinforcement [1]. Lignocellulosic fibres have many advantages, including that they are biodegradable and renewable, with acceptable specific properties compared to glass fibres. They also reduce dermal and respiratory irritation during handling, as well as tool wear. The main disadvantage of natural fibres is their hydrophilic nature, which results in incompatibility with hydrophobic polymeric matrices leading to poor composite mechanical properties [2]. However, coupling or compatibilizing agents can be used in order to improve adhesion between cellulosic fibres and polypropylene (PP) to improve mechanical properties [3]. In addition, the generation of a stronger interface between matrix and reinforcement material could reduce the hygroscopicity of lignocellulosic-based materials by reducing access to hydroxyl groups and therefore improve moisture resistance [4]. * Corresponding author. Tel.: +64 7 838 4672; fax: +64 7 838 4835. E-mail address: [email protected] (K.L. Pickering). 1359-835X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.compositesa.2008.04.013
Concern for the environment, both in terms of limiting the use of finite resources and the need to manage waste disposal, has led to increasing pressure to recycle materials at the end of their useful life. In the metals industries, for instance, materials recycling operations are already well established and are driven by economics [5]. Polymers are generally more difficult to recycle and the economic incentives to recycle have been less favourable, particularly when waste disposal in landfills is relatively cheap. However, waste management is now a high priority within the European Union. As a consequence, it is already illegal to landfill composites waste in many EU countries. The ‘‘End-of-Life Vehicle Directive” (Directive 2000/53/EC) regulates the disposal of vehicles and the requirements include that from 2015, 85% of the weight of all ‘‘End-of-Life” vehicles must be re-used or recycled, a further 10% may be subject to energy recovery with a maximum of only 5% of the vehicle allowed to be disposed of in landfill. As vehicles have a life expectancy of more than 10 years, vehicles currently being manufactured must meet the 2015 requirements [5]. As a consequence of increasing legislation, there is a need for recycling routes to be established [5]. Although there are governmental regulations in countries such as Germany, recycled materials are avoided, not only due to their physical properties, but mainly because of their surface appearance. Indeed, many designers are reluctant to use them as they can be rejected by the market. However, this is an attitude that can change [6], as demonstrated in Brazil where
1092
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
around 15% of all rigid plastics and films consumed, are recycled and returned to industry [7]. Some states in the US are also concerned with recycling. For example, in Michigan, the recycling rate is close to 100% and proves the potential for recycling plastic waste as well as changes in market attitude [8]. Plastic waste management can be carried out using three different approaches namely, thermo-mechanical recycling, energy recovery and biological recycling. Thermo-mechanical recycling first involves mechanical recycling where the thermoplastics are granulated, followed by techniques as extrusion or thermoforming. Energy recovery can be performed in two distinct ways. One is incineration where the hydrocarbon polymers replace fossil fuels. The second approach is pyrolysis or by hydrogenation to low molecular weight hydrocarbons for use either as portable fuels or as polymer feedstock. Biological recycling takes advantage of polymer biodegradation, which is highly dependent on the polymer type and environmental conditions. However, this type of recycling most often involves not only high costs and complex procedures but also potential damage to the environment [9]. Of these three techniques, thermo-mechanical would be expected to involve the least energy in terms of the product ‘‘life cycle”. It may be considered that, to be economical, a thermo-mechanical recycling process must be designed in such a manner that the energy to recover the post-consumer materials plus the energy to reprocessing must be equal to the amount of energy needed to produce the virgin material plus the energy required to dispose of the material. This balance does not, however, take into consideration environmental benefits. When those environmental gains are considered, higher energy consumption could be allowed during thermo-mechanical recycling for which the increase in cost could be compensated by indirect costs due to reduction in landfill, for example. Several authors have investigated the reprocessing of polypropylene. Guerrica-Echevarria et al. [10] have studied the evolution of rheological and mechanical properties of PP after repetitive injection moulding cycles. The main effect of reprocessing was found to be a decrease in molecular weight of the PP. In addition the melt viscosity and failure strain also decreased, however, in this work the chemical structure appeared unchanged. Xiang [11], however, has shown an important change in chemical structure and rheological values after seven cycles of injection moulding. Another study [12] showed an increase in crystallinity and decrease in tensile strength and elongation at break after seven extrusion cycles for virgin PP. Agro-based fibres are less brittle and softer than glass fibres and are likely to be easier to reprocess or recycle than mineral based fibres. Although no post-consumer based recycling studies have been done on agro-based fibres [13], a study on the effects of reprocessing had been conducted by Walz et al. [14] where, 50% Kenaf fibre reinforced PP composites were reprocessed for nine times. Both tensile and flexural properties were found to decrease with increase number of times the materials were reprocessed. Arbelaiz et al. discussed about reprocessing of flax fibre bundle/ polypropylene composites. After passing four times through injection moulding, tensile properties only showed a small decrease. A similar trend was shown by Joseph et al. [15] for 20 wt% sisal fibre/ LDPE matrix composites. Johan et al. [16] studied the properties of second-generation composites made from recycled materials (60% wood fibre, 30% HDPE, 5% PET and 5% tackifier). In general, the mechanical properties, water resistance, and dimensional stability of second-generation panels were equivalent to or better than properties obtained from first-generation panels. Bourmaud and Baley [17] studied the reprocessing of hemp and sisal fibre reinforced polypropylene composites. They found reduction of both TS and YM during reprocessing causing the reduction of fibre length and degradation of coupling agent.
In this study, thermo-mechanical processing was used for reprocessing of composites. The purpose of this investigation is to determine the extent to which wood fibre/PP composite materials are reprocessable. Optimization of fibre and coupling agent content was carried out for PP composites in a previous study [18]; 40 wt% fibre was found to provide a maximum TS and 50 wt% fibre was found to provide a maximum YM in composites with 4 wt% coupling agent (maleated polypropylene – MAPP). Using 4 wt% MAPP with 40 wt% fibre, for example, was found to give increases in TS and YM of 78% and 26%, respectively, compared with uncoupled composites. Based on the previous study, both 40 and 50 wt% wood fibre with 4 wt% MAPP were used to make composites in this study.
2. Experimental 2.1. Materials Radiata pine (Pinus radiata) bleached wood fibre (Kraft) was supplied by Tasman Pulp & Paper Co. Ltd., New Zealand. The average fibre length was 2.36 mm and the fibre diameter was 25 lm. The matrix polymer was a standard polypropylene (PP) powder with a density of 0.9 g/cm3 supplied by the Aldrich Chemical Company Inc. and MAPP (AC 950P) with a saponification value of 35– 40 mg KOH/g, a density of 0.93 g/cm3 and free maleic anhydride content of less than 0.5% was supplied by Honeywell International Inc., USA. 2.2. Methods 2.2.1. Composite fabrication Composites were fabricated with 40 wt% or 50 wt% fibre with 4 wt% MAPP and PP using a TSE-16-TC twin-screw extruder with a 15.6 mm screw blade diameter at 180 °C (maintaining five different temperature zones 100, 130, 160, 180 and 175 °C from feed zone to exit die) and a screw speed setting of 100 rpm. Prior to extrusion, wood fibre, PP and coupling agent were dried in an oven at 80 °C for a minimum of 48 h. Following extrusion, the material was pelletised into lengths of less than 5 mm and injection moulded into specimens for tensile and four-point bend testing using a BOY 15-S injection moulder. Specimens for tensile, four-point bend testing and impact testing were randomly selected from approximately 150 of each type to evaluate the mechanical properties. Impact test specimens were produced from standard four-point bend test samples by polishing to give a suitable sample width. The remaining specimens were granulated and injection moulded. Again specimens were randomly selected from these reprocessed materials for physical and mechanical property evaluation. The procedure of injection moulding and granulation was repeated for a total of eight times. 2.2.2. Tensile testing Tensile testing was carried out according to the ASTM 638-03: Standard Test Method for Tensile Properties of Plastics. Test specimens were placed in a conditioning chamber at 23 ± 3 °C and 50 ± 5% relative humidity for 40 h. The specimens were then tested using an Instron-4204 tensile testing machine fitted with a 5 kN load cell operated at a cross-head speed of 5 mm/min. An Instron 2630-112 extensometer was used to measure the strain. Six specimens were tested for each batch with a gauge length of 50 mm. 2.2.3. Four-point bend testing Four-point bend testing was carried out according to the ASTM D6272-02: Standard Test Method for Flexural Properties of Unreinforced and Reinforced Plastics and Electrical Insulating Materials
1093
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
using a Lloyd LR 100 K tensile tester fitted with a 50 kN load cell using a cross-head speed of 15 mm/min. Six specimens were tested for each batch.
2.2.13. Extraction of fibre from composites Fibre was extracted by dissolving the matrix in hot xylene at 110 °C followed by soxhlet extraction in xylene for 72 h.
2.2.4. Impact testing Charpy impact testing was carried out based on ISO 179: Plastics-Determination of Charpy Impact Strength. Dimensions of the samples were 80 mm 8 mm 3.2 mm with a 0.25 mm single notch (type A). Testing was carried out using a Polytest advanced universal pendulum impact tester with an impact velocity of 2.9 m/s at 21 °C.
2.2.14. Fibre length measurement The length and fibre distribution of virgin and extracted fibre from the composites was measured using a Kajaani FS-200 electronic sequential fibre analyzer. Fibre count of about 15,000– 20,000 was used to determine the fibre length distribution.
2.2.5. Hardness testing The surface hardness was measured using a LM 700 microhardness tester with a test load of 25 g and dwell time of 20 s.
3.1. Effects of reprocessing on mechanical properties
2.2.6. Melt flow index The melt flow index was obtained using a Dynisco MFI2 melt flow indexer following ASTM D1238: Standard Test Method for Melt Flow Rates of Thermoplastics by Extrusion Plastometer, using an applied load of 3.80 kg at 230 °C. Three independent tests were carried out on each sample. 2.2.7. Density Composite density was determined following the ASTM D 79200: Standard Test Methods for Density and Specific Gravity (Relative Density) of Plastics by Displacement, using the following formula:
Specific gravity ¼
apparent mass of specimen in air apparent mass of specimen in water
ð1Þ
2.2.8. Digital microscopy Following mounting, grinding and polishing, the fibre distribution and alignment of the fibre in the composites was observed using an Olympus B X 60 microscope. 2.2.9. Scanning electron microscopy (SEM) The fractured surfaces of the tensile test specimens were examined using a Hitachi S-4000 field emission scanning electron microscope, operated at 5 kV. Samples were mounted with carbon tape on aluminium stubs and then sputter coated with platinum and paladium to make them conductive prior to SEM observation. 2.2.10. Thermogravimetric analysis (TGA) Thermogravimetric analysis was carried out using a TA Instruments SDT 2960 Simultaneous DTA–TGA in air (flow rate 150 ml/ min) with a heating rate of 10 °C/min. The temperature range scanned was from 25 to 600 °C. The weights of all specimens were approximately 10 mg. Alumina open crucible was used for sample holding.
3. Results and discussion
Very little change on TS and YM was found for PP during reprocessing (Figs. 1 and 2). The TS and YM of the 40 wt% fibre composites decreased with increased number of times the materials were reprocessed in a linear fashion (Figs. 1 and 2). The virgin composites showed an average TS of 41 MPa and YM of 4553 MPa which reduced after being reprocessed eight times to 31 and 3800 MPa, respectively. TS and YM for the 50 wt% fibre composites are presented in Figs. 3 and 4. Based on a Rule of Mixtures type model, the most basic form being given below, increasing the fibre content in composites, would be expected to increase the TS and YM [19]:
Pcomposite ¼ ðPm vm þ Pf vf Þ
where P is the property (TS or YM), v is the volume fraction, and the suffixes m and f refer to matrix and fibre, respectively. However, TS was initially lower for 50 wt% fibre than for 40 wt% fibre content composites (Figs. 1 and 3) which is likely to be due to the limited dispersion of fibre in composites at higher fibre content due to the increase in viscosity as indicated by the reduction of melt flow index (Table 1). However, TS of 50 wt% fibre composites increased with increased number of times the materials were reprocessed up to two times from 37 MPa for virgin composites to 42 MPa for the composites reprocessed two times (Fig. 3), which was considered to be due to improved fibre dispersion due to the reduction of melt viscosity [10], but then decreased with further reprocessing and overall 11% reduction of TS was found for the composites after reprocessed eight times, compared to the virgin composites. As expected by the Rule of Mixtures, YM was higher for 50 wt% fibre content than 40 wt% fibre content composites (Figs. 2 and 4) and also increased with increased number of times the materials were reprocessed up to two times, and then decreased upon further reprocessing. During reprocessing, the occurrence of improvement of fibre dispersion and fibre damage occur simultaneously, but at lower levels of reprocessing the former seems to be predominant, although at
45 y = -1.36x + 41.16
Tensile strength (MPa)
40 2.2.11. Differential scanning calorimetry (DSC) DSC was carried out using a TA Instruments DSC 2920 differential scanning calorimeter in air (flow rate 50 ml/min) with a heating rate of 10 °C/min. The temperature range scanned was from 25 to 200 °C. The weights of all specimens were approximately 10 mg. Aluminium open crucible was used for sample holding.
ð2Þ
Composite
PP
35 30 25 20 15 10 5
2.2.12. X-ray diffraction (XRD) X-ray diffraction was carried out using a Philip’s X-ray diffractometer with a current of 15 mA and a voltage of 54 kV using Cu Ka radiation and a graphite monochromator with the range of 2h = 6–60°.
0 Virgin
1
2 3 4 5 6 Number of times reprocessed
7
8
Fig. 1. Tensile strength of virgin and reprocessed composites (40 wt% fibre) and PP.
1094
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
rN ¼ r0 þ aN
Young's modulus (MPa)
6000 y = -78.08x + 4553
5000
Composite
PP
4000 3000 2000 1000 0 Virgin
1
2 3 4 5 6 Number of times reprocessed
7
8
Tensile strength (MPa)
Fig. 2. Young’s modulus of virgin and reprocessed composites (40 wt% fibre) and PP.
50 45 40 35 30 25 20 15 10 5 0 Virgin
1
2 3 4 5 6 Number of times reprocessed
7
where rN is the tensile strength of the composites reprocessed N times and a is the slope of the TS versus number of times reprocessed graph. A 25% reduction in TS and 16% reduction in YM was observed for the composites after they were reprocessed eight times (Fig. 5). The higher reduction of TS than YM during reprocessing can be explained by the fact that considering simply load transfer alone, YM is a parameter measured at low strain, however, load transfer into fibre is more limited by fibre length at higher strain where failure will occur and therefore, TS could be expected to be more affected by fibre length than YM. However, this is a simple approach considering the number of processes that could be occurring at failure versus at low strain. Failure strain (FS) of PP was found to be greater than 500% (the maximum range of the tensile tester), but was found to be reduced drastically by the addition of fibre to about 2% for 40 wt% fibre composites and 1.3% for 50 wt% fibre composites (Figs. 6 and 7). The FS of both 40 and 50 wt% fibre composites increased exponentially with increased number of times the materials were reprocessed. Flexural tests, impact tests and hardness tests were also carried out for 40 wt% fibre composites. Flexural strength and flexural modulus were found to decrease with increased number of times the materials were reprocessed (Figs. 8 and 9). The virgin 40 wt% fibre composites showed an average flexural strength and flexural modulus of 57 and 7200 MPa, respectively, which reduced after being reprocessed eight times to 40 and 5700 MPa, respectively (Figs. 8 and 9). Impact strength was also found to decrease for both PP and composites containing 40 wt% fibre during reprocessing from 10.5 and 6.2 kJ/m2, respectively, for virgin PP and composites
8
30 Reduction of property (%)
Fig. 3. Tensile strength of virgin and reprocessed composites (50 wt% fibre).
8000
Young's modulus (MPa)
ð3Þ
7000 6000 5000 4000 3000
25
TS
YM
20 15 10 5 0
2000
1
2
1000 0 Virgin
1
2 3 4 5 6 Number of times reprocessed
7
8
3 4 5 6 Number of times reprocessed
7
8
Fig. 5. Reduction in TS and YM of composites during reprocessing (40 wt% fibre).
Fig. 4. Young’s modulus of virgin and reprocessed composites (50 wt% fibre).
6 Table 1 Melt flow index of composites and PP Type
Melt flow index (g/10 min)
PP Composite with 40 wt% fibre Composite with 50 wt% fibre
29.40 0.90 0.14
Failure strain (%)
5 4
y = 1.6701e0.1211x R2 = 0.9761
3 2 1
higher levels of reprocessing the latter appears to have an overriding influence. The trend of TS for the 40 wt% composites versus the number of times the materials were reprocessed (Fig. 1) was found to be closely represented by the empirical equation:
0 Virgin
1
2 3 4 5 6 Number of times reprocessed
7
8
Fig. 6. Failure strain of virgin and reprocessed composites (40 wt% fibre).
1095
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
12
4.5
Composite
Impact strength (kJ/m 2)
4 Failure strain (%)
3.5 y = 1.051e0.1366x R2 = 0.9701
3 2.5 2 1.5
PP
10 8 6 4 2
1 0
0.5
Virgin
2 4 6 Number of times reprocessed
0 Virgin
1
2
3 4 5 6 Number of times recycled
7
8
8 Fig. 10. Impact strength of virgin and reprocessed composites and PP (40 wt% fibre).
Fig. 7. Failure strain of virgin and reprocessed composites (50 wt% fibre).
12
Vicker's hardness number
Flexural strength (MPa)
70 60 50 40 30 20 10
10 8 6 4 2 0 Virgin
0 Virgin
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
Number of times reprocessed
Number of times reprocessed Fig. 11. Vicker’s hardness of virgin and reprocessed composites (40 wt% fibre). Fig. 8. Flexural strength of virgin and reprocessed composites (40 wt% fibre).
Flexural modulus (MPa)
9000 8000 7000 6000 5000 4000 3000 2000 1000 0 Virgin
1
2
3
4
5
6
7
8
Number of times reprocessed Fig. 9. Flexural modulus of virgin and reprocessed composites (40 wt% fibre).
to 6.2 and 3.2 kJ/m2, respectively, after the materials were reprocessed eight times (Fig. 10). The hardness of the composites increased with reprocessing (Fig. 11). Generally, fibre was observed to be well dispersed in virgin composites (Fig. 12). With increased reprocessing, interfacial bonding was found to improve (Fig. 13) with less fibre pull out observed for eight times reprocessed composites than the virgin composites fracture surface. Although better interfacial bonding would be expected to lead to better mechanical properties after reprocessing, TS, YM, flexural strength and flexural modulus were found
to decrease. A major reason for the reduction of these properties is likely to be due to the fact that reprocessing incurred some fibre damage. The average fibre length was found to decrease from 2.36 mm for virgin fibre to 0.37 mm for the fibre extracted from the 40 wt% fibre composites reprocessed eight times (Fig. 14). In addition, the amount of fibre fines (fibre length less than 0.20 mm) was found to increase (Fig. 15) and the length distribution of fibres became narrower and reduced to shorter fibre lengths (Fig. 16) with increased number of times the materials were reprocessed. The shorter fibre lengths and the increased fines percentage with reprocessing were also observable by optical light microscopy (Fig. 17). The reductions of fibre length would be expected to reduce reinforcing efficiency, leading to the observed reduction in TS, YM, flexural strength and flexural modulus [20]. The increase in FS was likely to be due to reduced constraint from the shorter fibre and the reduction of micro-voids as evaluated by the increased density of composites (Fig. 18). The reduction of impact strength could be due to the reduction of molecular weight of PP during reprocessing. The increase in hardness may have resulted from the reduction of micro-voids and the increase in composite density with increased reprocessing. The change in average fibre length can be correlated with the number of times the composite materials were reprocessed by the following empirical equation:
lN ¼ l0 ebN
ð4Þ
where lN is the average fibre length at any reprocessed composites, l0 is the length of virgin fibre, and b is the slope of fibre length versus number of times reprocessed graph.
1096
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
Fig. 12. Optical micrographs of 40 and 50 wt% composites surface showing fibre distribution in composites (magnification 200).
Fig. 13. SEM of composite fracture surface showing less pull out for reprocessed composites (magnification for a = 3.03k and b = 3.5k).
1
rg
8
in Vi rg
Virgin 2 4 6 composite Number of times reprocessed
Vi
Virgin fibre
co m
in
0
po
fib
si te
re
0.5
8
1.5
y = 3.1406e0.5785x R2 = 0.9787
6
y = 3.4564e-0.3545x R2 = 0.9802
2
4
Fines (%)
Length (mm)
2.5
120 100 80 60 40 20 0
2
3
Number of times reprocessed
Fig. 14. Weighted average fibre length of virgin fibre and the fibre extracted from composites.
Fig. 15. Fines (fibres with lengths less than 0.20 mm) in the virgin fibre and the fibre extracted from composites.
Rearranging Eq. (4) gives:
l0 ¼ bN ln lN
ð5Þ
and combining with Eq. (3) gives TS as a function of the ratio of the original fibre length to the reduced length that has occurred with reprocessing as follows:
rN
a l0 ¼ r0 þ ln b lN
ð6Þ
The theoretical TS calculated by using Eq. (6) is compared with experimental values in Fig. 19 and shows good agreement. 3.2. Effects of reprocessing on thermal stability and crystallinity The DSC curves for PP and 40 wt% fibre composites are shown in Figs. 20 and 21 and demonstrate the reduction of melting temperature (Tm) with reprocessing. The percentage crystallinity was obtained from the relation [21]:
1097
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
Virgin composite Population (%)
Population (%)
Virgin fibre 15 10 5 0
20 15 10 5 0
0.2 1 1.8 2.6 3.4 4.2 5 5.6 Length (mm)
0.2 1 1.8 2.6 3.4 4.2 5 5.6 Length (mm) 8 times reprocessed composite Population (%)
Population (%)
4 times reprocessed composite 30 20 10 0
60 40 20 0
0.2 1 1.8 2.6 3.4 4.2 5 5.6 Length (mm)
0.2 1 1.8 2.6 3.4 4.2 5 5.6 Length (mm)
Fig. 16. Fibre length distribution of virgin fibre and the fibre extracted from composites.
Fig. 17. Optical micrographs of 40 wt% composites surface showing fibre distribution in composites (magnification 200).
45
1.4
40
Tensile strength (MPa)
Density (g/cm3)
1.2 1 0.8 0.6 0.4
35 30 25 20 15 10
Theoretical
5
0.2
Experimental
0
0 Virgin
2
4
6
0
8
6
8
10
Fig. 19. Theoretical and experimental value of TS (verification of Eq. (6)) for mixture of virgin and reprocessed composites.
Fig. 18. Density of virgin and reprocessed composites.
DH 100 W DH m
4
Number of times reprocessed
Number of times reprocessed
% Crystallinity ¼
2
ð7Þ
where DH and DHm are the heats of fusion of PP and 100% crystalline PP, respectively, and W is the fraction of PP in the composite. Taking DHm = 148 J/g for PP [22], the crystallinity of virgin and reprocessed PP and that of the PP in the composites was calculated and is presented in Table 2. The crystallinity of PP only samples and PP in the composites increased with increased number of times the materials were reprocessed (Table 2) which is likely to be as a result
of molecular weight reduction [23]. The increase in crystallinity is also supported by the XRD traces (Figs. 22 and 23). The Tm for PP alone showed very little change during reprocessing (Table 2). The Tm for PP was found to increase slightly from 169.2 °C for PP only samples to 170.5 °C for PP in the virgin composites, which is likely to be due to the fibres acting as nucleating sites and increasing the crystallinity [24], however, the Tm for PP in composites decreased slightly with increased number of times the materials were reprocessed which could be due to the reduction of molecular weight [23].
1098
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
-2
Heat flow (mW)
160
170
180
190
-4 -5 -6 Virgin PP
-7 -8
8 times reprocessed PP
-9
Number of counts
600
-3150
Virgin composite
500 400
8 times reprocessed composite
300 200 100 0 10
-10
15
20 Angle (2θ)
Temperature (oC)
Heat flow (mW)
Fig. 20. DSC curves of virgin and reprocessed PP.
-1 -1.5120 -2
130
140
150
160
170
180
190
-3 -3.5 -4.5 -5
8 times reprocessed composites Temperature (oC)
Fig. 21. DSC curves of virgin and reprocessed 40 wt% fibre composites.
Table 2 Melting point and crystallinity (calculated from DSC analysis) of PP only samples and the PP in composites Melting point Tm (°C)
Crystallinity (%) of PP
Virgin PP Four times reprocessed PP Eight times reprocessed PP
169.2 168.8 168.7
38.8 39.0 40.1
Virgin composite Four times reprocessed composite Eight times reprocessed composite
170.5 167.7 167.0
43.7 44.9 45.2
800 Virgin PP
Number of counts
8 times reprocessed PP
600
ð8Þ
where y is the fraction of nonvolatilized material, not yet decomposed, Tmax is the temperature of maximum reaction rate, b is the heating rate, Z is the frequency factor and Ea is the activation energy. Initially, plots of ln ln(1/y) versus 1/T for various stages of decomposition were drawn such as in Fig. 28 for the first stage of PP decomposition and found to be linear, suggesting good agreement with the Broido equation. The activation energies, Ea determined from the slopes of these plots are given in Table 3. Tmax
Temp. diff. / weight (oC/mg)
Sample
700
weight loss for oxidative decomposition and finally slow decomposition corresponding to formation of char as the temperature increased [25]. Kinetic parameters for the various stages of thermal degradation were determined from the TGA graphs using the following equation, given by Broido [26]:
1 Ea RZ 2 ln ln ¼ þ ln Tm y Ea b RT
Virgin composites
-5.5
30
Fig. 23. XRD traces of virgin and reprocessed 40 wt% fibre composites showing higher intensity for reprocessed composites.
-2.5
-4
25
6 Virgin PP
5
8 times reprocessed PP
4 3 2 1 0 50
150
250
350
450
550
Temperature (oC)
500 400
Fig. 24. DTA traces for virgin and reprocessed PP.
300 200 100 7
15
20 Angle (2θ)
25
30
Fig. 22. XRD traces of virgin and reprocessed PP showing higher intensity for reprocessed PP.
Typical TGA and DTA traces for virgin PP as well as reprocessed PP and the composites are shown in Figs. 24–27. Three stages of decomposition were observed for all composites and PP samples (Figs. 24 and 25) starting with dehydration and decomposition of volatile components at low temperatures, followed by rapid
Temp. diff./weight ( oC/mg)
0 10
Virgin composites
6
8 times reprocessed composites
5 4 3 2 1 0 50
150
250 350 Temperature (oC)
450
550
Fig. 25. DTA traces for virgin and reprocessed 40 wt% fibre composites.
1099
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
40 20 0 0
100
200
300
400
500
600
o
Temperature ( C) Fig. 26. TGA curves of virgin and reprocessed PP.
Fig. 29. Tensile strength of composites produced from the mixture of virgin and reprocessed composites (40 wt% fibre).
Virgin composites
100
50%Virgin+50% 8 times reprocessed
60
50% Virgin+50% 6 times reprocessed
8 times reprocessed PP
50%Virgin+50% 4 times reprocessed
80
Experimental Theoretical
50%Virgin+50% 2 times reprocessed
Weight (%)
100
50 45 40 35 30 25 20 15 10 5 0 100% Virgin
Vrigin PP
Tensile strength (MPa)
120
Weight (%)
80 8 times reprocessed composites
60 40 20 0 0
100
200 300 400 Temperature (oC)
500
600
Fig. 27. TGA curves of virgin and reprocessed 40 wt% fibre composites.
3.3. Mixing virgin and reprocessed composites
0 0.0016 -1
0.0017
0.0018
0.0019
0.002
0.0021
The TS of composites prepared with a mixture of 50 wt% virgin and 50 wt% (two to eight times) reprocessed composites are shown in Fig. 29, and it can be seen that TS decreased from 41 MPa for virgin composites to 37.5 MPa for 50 wt% virgin + 50 wt% eight times reprocessed composites. TS of the mixture would be expected to follow the Rule of Mixtures equation Eq. (2) rewritten for clarity as follows:
lnln(1/y)
-2 -3 -4
and Ea for PP and composites were generally found to increase with increased number of times the materials were reprocessed (Table 3). The positions of weight loss on the TGA traces both for PP and composites shifted to higher temperatures with increased number of times the materials were reprocessed (Figs. 26 and 27) suggesting increased thermal stability. The increase in thermal stability is likely to be due to the increase in crystallinity of PP [27] resulting from molecular weight reduction [23] and better interfacial bonding during reprocessing supported by observations of fracture surfaces (Fig. 13).
y = -12399x + 19.838 R2 = 0.993
rmix ¼ ðr0 v0 þ rN vN Þ
-5 -6 1/T (K) Fig. 28. 1/T versus ln ln(1/y) for the first decomposition stage of PP.
ð9Þ
where r0 and rN are the TS of virgin and reprocessed composites, v0 and vN are the volume fraction of virgin and reprocessed composites, and N is the number of times the materials were reprocessed. Good agreement with the Rule of Mixtures equation was obtained (Fig. 29).
Table 3 Thermal properties of PP and composites Sample
Stage
Weight loss (%)
Temperature range (°C)
Tmax (°C)
Activation energy Ea (kJ/mol)
Virgin PP
First Second Third
26 70 3
224–320 320–445 445–534
293 375 504
104 78 43
Eight times reprocessed PP
First Second Third
18 79 2
224–328 320–445 445–534
285 390 506
115 84 50
Virgin composite
First Second Third
61 31 7
226–351 351–436 436–508
285 371 455
85 68 60
Eight times reprocessed composite
First Second Third
30 52 15
230–340 340–447 470–512
289 412 470
87 71 81
1100
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
4. Conclusions For 40 wt% fibre reinforced composites reprocessed eight times, a 25% reduction in TS and 16% reduction in YM was found. Also for these composites, impact strength, flexural strength and flexural modulus were found to decrease with increased number of times the materials were reprocessed. Reprocessing of 50 wt% fibre composites brought about an increase in both TS and YM from 37 and 4830 MPa, respectively, for virgin composites to 42 and 6421 MPa for two times reprocessed composites followed by decreases in TS and YM upon further reprocessing. The reduction of TS, YM, flexural strength, flexural modulus and impact strength was considered to be due to fibre damage that occurred during reprocessing as evaluated by the associated reduction of the average fibre length from 2.36 mm for virgin fibre to 0.37 mm for the fibre extracted from the composites reprocessed eight times. The FS was found to increase with increased number of times the materials were reprocessed due less fibre constrain and the reduction of voids as evaluated by the increase in composite density. Composite thermal stability was found to increase with increased reprocessing believed to be due to the improvement of interfacial bonding and increased crystallinity of PP resulting from the reduction of its molecular weight. TS of composites made with mixtures of virgin and reprocessed material were in close agreement with the Rule of Mixtures. However, reprocessing of composites subjected to actual weathering would give better understanding of the real-world performance. Acknowledgements The authors would like thanks to the Foundation for Research Science and Technology, New Zealand for their financial support of this project. References [1] Espert A, Vilaplana F, Karlsson S. Comparison of water absorption in natural cellulosic fibres from wood and one-year crops in polypropylene composites and its influence on their mechanical properties. Compos Pt A: Appl Sci Manuf 2004;35(11):1267–76. [2] Peijs T, Melick V, Garkhail SK, Pott GT, Baillie CA. Natural-fibre-mat reinforced thermoplastics based on upgraded flax fibres for improved moisture resistance. In: ECCM-8 conference; 1998. p. 119–26. [3] Tajvidi M, Ebrahimi G. Water uptake and mechanical characteristics of natural filler–polypropylene composites. J Appl Polym Sci 2003;88:941–6. [4] Avella M, Bozzi C, Dell’Erba R, Focher B, Marzetti A, Martuscelli E. Steam exploded wheat straw fibres as reinforcing material for polypropylene-based composites. Angew Makromol Chem 1995;233:149–66.
[5] Pickering SJ. Recycling technologies for thermoset composite materials current status. Compos Pt A 2006;37(8):1206–15. [6] Jones R, Baumann MH. Recycling of engineering plastics. In: ANTEC conference proceedings, vol. 4; 1997. p. 3066–74. [7] Martins MH, Paoli A. Polypropylene compounding with recycled material. I: Statistical response surface. Polym Degrad Stabil 2001;71(2):93–8. [8] Selke SE. Plastics recycling. Handbook of plastics, elastomers and composites. 4th ed. New York: McGraw- Hill; 2002. p. 693–757. [9] Scott G. Green polymers. Polym Degrad Stabil 2000;68:1–7. [10] Guerrica-Echevarria G, Eguiazabal JI, Nazabal J. Effects of reprocessing conditions on the properties of unfilled and talc-filled polypropylene. Polym Degrad Stabil 1996;53(1):1–8. [11] Xiang Q, Xanthos M, Mitra S, Patel SH, Guo J. Effects of melt reprocessing on volatile emissions and structural/rheological changes of unstabilized polypropylene. Polym Degrad Stabil 2002;7(1):93–102. [12] Martins MH, De Paoli M-A. Polypropylene compounding with post-consumer material. II: Reprocessing. Polym Degrad Stabil 2002;78(3):491–5. [13] Rowell RM, Young RA, Rowell JK. Paper and composites from agro-based resources. Lewis Publishers; 1996. [14] Walz K, Jacobson R, Sanadi AR. Effect of reprocessing/recycling on the mechanical properties of kenaf-PP composites. Internal report, University of Wisconsin-Madison and Forest Product Laboratory; 1994. [15] Joseph K, Thomas S, Pavithran C, Brahmakumar M. Tensile properties of short sisal fibre-reinforced polyethylene composites. J Appl Polym Sci 1993;47:1731–9. [16] Youngquist JA, Myers GE, Muehl JH, Krzysik AM, Clemens CM. Composites from recycled wood and plastics. US Environmental Protection Agency, Cincinnati, OH 45268.. [17] Bourmaud A, Baley C. Investigations on the recycling of hemp and sisal fibre reinforced polypropylene composites. Polym Degrad Stabil 2007;92:1034–45. [18] Beg MDH, Pickering KL. The improvement of interfacial bonding, weathering and recycling of wood fibre reinforced polypropylene composites. PhD thesis, University of Waikato, New Zealand; 2007. [19] Sanadi AR, Young RA, Clemons C, Rowell RM. Recycled newspaper fibres as reinforcing fillers in thermoplastics. Part I: Analysis of tensile and impact properties in polypropylene. J Reinf Plast Compos 1994;13:54–67. [20] Harper LT, Turner TA, Warrior NA, Rudd CD. Characterisation of random carbon fibre composites from a directed fibre preforming process: the effect of fibre length. Compos Pt A: Appl Sci Manuf 2006;37(11):1863–78. [21] Mathew AP, Oksman K, Sain M. The effect of morphology and chemical characteristics of cellulose reinforcements on the crystallinity of polylactic acid. J Appl Polym Sci 2006;101:300–10. [22] Monasse B, Haudin JM. Growth transition and morphology change in polypropylene. Colloid Polym Sci 1985;263(10):822–31. [23] La Mantia FP, Gardette JL. Improvement of the mechanical properties of photooxidized film after recycling. Polym Degrad Stabil 2002;75:1–7. [24] Lou N, Netravali AN. Mechanical and thermal properties of environmentfriendly ‘‘green” composites made from pineapple leaf fibres and poly(hydroxyl butyrate-co-valerate) resin. Polym Compos 1999;20(3): 367–78. [25] Gao M, Zhu K, Sun YJ. Thermal degradation of wood treated with amino resins and amino resins modified with phosphate in nitrogen. J Fire Sci 2004;22:505–15. [26] Broido A. A simple, sensitive graphical method of treating thermogravimetric analysis data. J Polym Sci Pt A-2 1969;7:1761. [27] Gaymans RJ. Toughening of semicrystalline thermoplastics. Polymer blends: performance. New York: John Wiley & Sons; 2000. p. 178–219.
Contents lists available at ScienceDirect
Composites: Part A journal homepage: www.elsevier.com/locate/compositesa
Reprocessing of wood fibre reinforced polypropylene composites. Part I: Effects on physical and mechanical properties M.D.H. Beg, K.L. Pickering * Department of Engineering, University of Waikato, Private Bag 3105, Hamilton, New Zealand
a r t i c l e
i n f o
Article history: Received 9 May 2007 Received in revised form 21 April 2008 Accepted 22 April 2008
Keywords: A. Natural fibre composites E. Reprocessing B. Fibre length A. Coupling agent
a b s t r a c t This paper describes the effects of reprocessing on the physical and mechanical properties of composites based on radiata pine (Pinus radiata) fibre in a polypropylene (PP) matrix. Composites, containing either 40 wt% or 50 wt% fibre with 4 wt% maleated polypropylene (MAPP) as a coupling agent, were reprocessed up to eight times. For composites with 40 wt% fibre, tensile strength (TS) and Young’s modulus (YM) were found to decrease with increased reprocessing by up to 25% for TS and 17% for YM (after reprocessed eight times). Flexural tests were also carried out for 40 wt% fibre composites and flexural strength and modulus were found to decrease with increased reprocessing. Although, TS was lower for virgin composites with 50 wt% fibre than for those with 40 wt% fibre, this initially increased with reprocessing by up to 14% (after reprocessed two times), but then decreased with further reprocessing, and an overall 11% reduction of TS was found after reprocessing eight times compared to the virgin composites. YM was higher for virgin composites with 50 wt% than those with for 40 wt% fibre and also initially increased with reprocessing but decreased upon further reprocessing. Reprocessing was found to increase thermal stability. The TS of composites made by combining reprocessed with virgin materials was also assessed. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction In recent decades, growing environmental awareness has resulted in renewed interest in the use of natural materials for different applications. Increasingly more stringent environmental policies have forced industries such as the automotive, packaging and construction industries to search for new materials that can substitute traditional composite materials consisting of a plastic matrix and inorganic reinforcement [1]. Lignocellulosic fibres have many advantages, including that they are biodegradable and renewable, with acceptable specific properties compared to glass fibres. They also reduce dermal and respiratory irritation during handling, as well as tool wear. The main disadvantage of natural fibres is their hydrophilic nature, which results in incompatibility with hydrophobic polymeric matrices leading to poor composite mechanical properties [2]. However, coupling or compatibilizing agents can be used in order to improve adhesion between cellulosic fibres and polypropylene (PP) to improve mechanical properties [3]. In addition, the generation of a stronger interface between matrix and reinforcement material could reduce the hygroscopicity of lignocellulosic-based materials by reducing access to hydroxyl groups and therefore improve moisture resistance [4]. * Corresponding author. Tel.: +64 7 838 4672; fax: +64 7 838 4835. E-mail address: [email protected] (K.L. Pickering). 1359-835X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.compositesa.2008.04.013
Concern for the environment, both in terms of limiting the use of finite resources and the need to manage waste disposal, has led to increasing pressure to recycle materials at the end of their useful life. In the metals industries, for instance, materials recycling operations are already well established and are driven by economics [5]. Polymers are generally more difficult to recycle and the economic incentives to recycle have been less favourable, particularly when waste disposal in landfills is relatively cheap. However, waste management is now a high priority within the European Union. As a consequence, it is already illegal to landfill composites waste in many EU countries. The ‘‘End-of-Life Vehicle Directive” (Directive 2000/53/EC) regulates the disposal of vehicles and the requirements include that from 2015, 85% of the weight of all ‘‘End-of-Life” vehicles must be re-used or recycled, a further 10% may be subject to energy recovery with a maximum of only 5% of the vehicle allowed to be disposed of in landfill. As vehicles have a life expectancy of more than 10 years, vehicles currently being manufactured must meet the 2015 requirements [5]. As a consequence of increasing legislation, there is a need for recycling routes to be established [5]. Although there are governmental regulations in countries such as Germany, recycled materials are avoided, not only due to their physical properties, but mainly because of their surface appearance. Indeed, many designers are reluctant to use them as they can be rejected by the market. However, this is an attitude that can change [6], as demonstrated in Brazil where
1092
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
around 15% of all rigid plastics and films consumed, are recycled and returned to industry [7]. Some states in the US are also concerned with recycling. For example, in Michigan, the recycling rate is close to 100% and proves the potential for recycling plastic waste as well as changes in market attitude [8]. Plastic waste management can be carried out using three different approaches namely, thermo-mechanical recycling, energy recovery and biological recycling. Thermo-mechanical recycling first involves mechanical recycling where the thermoplastics are granulated, followed by techniques as extrusion or thermoforming. Energy recovery can be performed in two distinct ways. One is incineration where the hydrocarbon polymers replace fossil fuels. The second approach is pyrolysis or by hydrogenation to low molecular weight hydrocarbons for use either as portable fuels or as polymer feedstock. Biological recycling takes advantage of polymer biodegradation, which is highly dependent on the polymer type and environmental conditions. However, this type of recycling most often involves not only high costs and complex procedures but also potential damage to the environment [9]. Of these three techniques, thermo-mechanical would be expected to involve the least energy in terms of the product ‘‘life cycle”. It may be considered that, to be economical, a thermo-mechanical recycling process must be designed in such a manner that the energy to recover the post-consumer materials plus the energy to reprocessing must be equal to the amount of energy needed to produce the virgin material plus the energy required to dispose of the material. This balance does not, however, take into consideration environmental benefits. When those environmental gains are considered, higher energy consumption could be allowed during thermo-mechanical recycling for which the increase in cost could be compensated by indirect costs due to reduction in landfill, for example. Several authors have investigated the reprocessing of polypropylene. Guerrica-Echevarria et al. [10] have studied the evolution of rheological and mechanical properties of PP after repetitive injection moulding cycles. The main effect of reprocessing was found to be a decrease in molecular weight of the PP. In addition the melt viscosity and failure strain also decreased, however, in this work the chemical structure appeared unchanged. Xiang [11], however, has shown an important change in chemical structure and rheological values after seven cycles of injection moulding. Another study [12] showed an increase in crystallinity and decrease in tensile strength and elongation at break after seven extrusion cycles for virgin PP. Agro-based fibres are less brittle and softer than glass fibres and are likely to be easier to reprocess or recycle than mineral based fibres. Although no post-consumer based recycling studies have been done on agro-based fibres [13], a study on the effects of reprocessing had been conducted by Walz et al. [14] where, 50% Kenaf fibre reinforced PP composites were reprocessed for nine times. Both tensile and flexural properties were found to decrease with increase number of times the materials were reprocessed. Arbelaiz et al. discussed about reprocessing of flax fibre bundle/ polypropylene composites. After passing four times through injection moulding, tensile properties only showed a small decrease. A similar trend was shown by Joseph et al. [15] for 20 wt% sisal fibre/ LDPE matrix composites. Johan et al. [16] studied the properties of second-generation composites made from recycled materials (60% wood fibre, 30% HDPE, 5% PET and 5% tackifier). In general, the mechanical properties, water resistance, and dimensional stability of second-generation panels were equivalent to or better than properties obtained from first-generation panels. Bourmaud and Baley [17] studied the reprocessing of hemp and sisal fibre reinforced polypropylene composites. They found reduction of both TS and YM during reprocessing causing the reduction of fibre length and degradation of coupling agent.
In this study, thermo-mechanical processing was used for reprocessing of composites. The purpose of this investigation is to determine the extent to which wood fibre/PP composite materials are reprocessable. Optimization of fibre and coupling agent content was carried out for PP composites in a previous study [18]; 40 wt% fibre was found to provide a maximum TS and 50 wt% fibre was found to provide a maximum YM in composites with 4 wt% coupling agent (maleated polypropylene – MAPP). Using 4 wt% MAPP with 40 wt% fibre, for example, was found to give increases in TS and YM of 78% and 26%, respectively, compared with uncoupled composites. Based on the previous study, both 40 and 50 wt% wood fibre with 4 wt% MAPP were used to make composites in this study.
2. Experimental 2.1. Materials Radiata pine (Pinus radiata) bleached wood fibre (Kraft) was supplied by Tasman Pulp & Paper Co. Ltd., New Zealand. The average fibre length was 2.36 mm and the fibre diameter was 25 lm. The matrix polymer was a standard polypropylene (PP) powder with a density of 0.9 g/cm3 supplied by the Aldrich Chemical Company Inc. and MAPP (AC 950P) with a saponification value of 35– 40 mg KOH/g, a density of 0.93 g/cm3 and free maleic anhydride content of less than 0.5% was supplied by Honeywell International Inc., USA. 2.2. Methods 2.2.1. Composite fabrication Composites were fabricated with 40 wt% or 50 wt% fibre with 4 wt% MAPP and PP using a TSE-16-TC twin-screw extruder with a 15.6 mm screw blade diameter at 180 °C (maintaining five different temperature zones 100, 130, 160, 180 and 175 °C from feed zone to exit die) and a screw speed setting of 100 rpm. Prior to extrusion, wood fibre, PP and coupling agent were dried in an oven at 80 °C for a minimum of 48 h. Following extrusion, the material was pelletised into lengths of less than 5 mm and injection moulded into specimens for tensile and four-point bend testing using a BOY 15-S injection moulder. Specimens for tensile, four-point bend testing and impact testing were randomly selected from approximately 150 of each type to evaluate the mechanical properties. Impact test specimens were produced from standard four-point bend test samples by polishing to give a suitable sample width. The remaining specimens were granulated and injection moulded. Again specimens were randomly selected from these reprocessed materials for physical and mechanical property evaluation. The procedure of injection moulding and granulation was repeated for a total of eight times. 2.2.2. Tensile testing Tensile testing was carried out according to the ASTM 638-03: Standard Test Method for Tensile Properties of Plastics. Test specimens were placed in a conditioning chamber at 23 ± 3 °C and 50 ± 5% relative humidity for 40 h. The specimens were then tested using an Instron-4204 tensile testing machine fitted with a 5 kN load cell operated at a cross-head speed of 5 mm/min. An Instron 2630-112 extensometer was used to measure the strain. Six specimens were tested for each batch with a gauge length of 50 mm. 2.2.3. Four-point bend testing Four-point bend testing was carried out according to the ASTM D6272-02: Standard Test Method for Flexural Properties of Unreinforced and Reinforced Plastics and Electrical Insulating Materials
1093
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
using a Lloyd LR 100 K tensile tester fitted with a 50 kN load cell using a cross-head speed of 15 mm/min. Six specimens were tested for each batch.
2.2.13. Extraction of fibre from composites Fibre was extracted by dissolving the matrix in hot xylene at 110 °C followed by soxhlet extraction in xylene for 72 h.
2.2.4. Impact testing Charpy impact testing was carried out based on ISO 179: Plastics-Determination of Charpy Impact Strength. Dimensions of the samples were 80 mm 8 mm 3.2 mm with a 0.25 mm single notch (type A). Testing was carried out using a Polytest advanced universal pendulum impact tester with an impact velocity of 2.9 m/s at 21 °C.
2.2.14. Fibre length measurement The length and fibre distribution of virgin and extracted fibre from the composites was measured using a Kajaani FS-200 electronic sequential fibre analyzer. Fibre count of about 15,000– 20,000 was used to determine the fibre length distribution.
2.2.5. Hardness testing The surface hardness was measured using a LM 700 microhardness tester with a test load of 25 g and dwell time of 20 s.
3.1. Effects of reprocessing on mechanical properties
2.2.6. Melt flow index The melt flow index was obtained using a Dynisco MFI2 melt flow indexer following ASTM D1238: Standard Test Method for Melt Flow Rates of Thermoplastics by Extrusion Plastometer, using an applied load of 3.80 kg at 230 °C. Three independent tests were carried out on each sample. 2.2.7. Density Composite density was determined following the ASTM D 79200: Standard Test Methods for Density and Specific Gravity (Relative Density) of Plastics by Displacement, using the following formula:
Specific gravity ¼
apparent mass of specimen in air apparent mass of specimen in water
ð1Þ
2.2.8. Digital microscopy Following mounting, grinding and polishing, the fibre distribution and alignment of the fibre in the composites was observed using an Olympus B X 60 microscope. 2.2.9. Scanning electron microscopy (SEM) The fractured surfaces of the tensile test specimens were examined using a Hitachi S-4000 field emission scanning electron microscope, operated at 5 kV. Samples were mounted with carbon tape on aluminium stubs and then sputter coated with platinum and paladium to make them conductive prior to SEM observation. 2.2.10. Thermogravimetric analysis (TGA) Thermogravimetric analysis was carried out using a TA Instruments SDT 2960 Simultaneous DTA–TGA in air (flow rate 150 ml/ min) with a heating rate of 10 °C/min. The temperature range scanned was from 25 to 600 °C. The weights of all specimens were approximately 10 mg. Alumina open crucible was used for sample holding.
3. Results and discussion
Very little change on TS and YM was found for PP during reprocessing (Figs. 1 and 2). The TS and YM of the 40 wt% fibre composites decreased with increased number of times the materials were reprocessed in a linear fashion (Figs. 1 and 2). The virgin composites showed an average TS of 41 MPa and YM of 4553 MPa which reduced after being reprocessed eight times to 31 and 3800 MPa, respectively. TS and YM for the 50 wt% fibre composites are presented in Figs. 3 and 4. Based on a Rule of Mixtures type model, the most basic form being given below, increasing the fibre content in composites, would be expected to increase the TS and YM [19]:
Pcomposite ¼ ðPm vm þ Pf vf Þ
where P is the property (TS or YM), v is the volume fraction, and the suffixes m and f refer to matrix and fibre, respectively. However, TS was initially lower for 50 wt% fibre than for 40 wt% fibre content composites (Figs. 1 and 3) which is likely to be due to the limited dispersion of fibre in composites at higher fibre content due to the increase in viscosity as indicated by the reduction of melt flow index (Table 1). However, TS of 50 wt% fibre composites increased with increased number of times the materials were reprocessed up to two times from 37 MPa for virgin composites to 42 MPa for the composites reprocessed two times (Fig. 3), which was considered to be due to improved fibre dispersion due to the reduction of melt viscosity [10], but then decreased with further reprocessing and overall 11% reduction of TS was found for the composites after reprocessed eight times, compared to the virgin composites. As expected by the Rule of Mixtures, YM was higher for 50 wt% fibre content than 40 wt% fibre content composites (Figs. 2 and 4) and also increased with increased number of times the materials were reprocessed up to two times, and then decreased upon further reprocessing. During reprocessing, the occurrence of improvement of fibre dispersion and fibre damage occur simultaneously, but at lower levels of reprocessing the former seems to be predominant, although at
45 y = -1.36x + 41.16
Tensile strength (MPa)
40 2.2.11. Differential scanning calorimetry (DSC) DSC was carried out using a TA Instruments DSC 2920 differential scanning calorimeter in air (flow rate 50 ml/min) with a heating rate of 10 °C/min. The temperature range scanned was from 25 to 200 °C. The weights of all specimens were approximately 10 mg. Aluminium open crucible was used for sample holding.
ð2Þ
Composite
PP
35 30 25 20 15 10 5
2.2.12. X-ray diffraction (XRD) X-ray diffraction was carried out using a Philip’s X-ray diffractometer with a current of 15 mA and a voltage of 54 kV using Cu Ka radiation and a graphite monochromator with the range of 2h = 6–60°.
0 Virgin
1
2 3 4 5 6 Number of times reprocessed
7
8
Fig. 1. Tensile strength of virgin and reprocessed composites (40 wt% fibre) and PP.
1094
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
rN ¼ r0 þ aN
Young's modulus (MPa)
6000 y = -78.08x + 4553
5000
Composite
PP
4000 3000 2000 1000 0 Virgin
1
2 3 4 5 6 Number of times reprocessed
7
8
Tensile strength (MPa)
Fig. 2. Young’s modulus of virgin and reprocessed composites (40 wt% fibre) and PP.
50 45 40 35 30 25 20 15 10 5 0 Virgin
1
2 3 4 5 6 Number of times reprocessed
7
where rN is the tensile strength of the composites reprocessed N times and a is the slope of the TS versus number of times reprocessed graph. A 25% reduction in TS and 16% reduction in YM was observed for the composites after they were reprocessed eight times (Fig. 5). The higher reduction of TS than YM during reprocessing can be explained by the fact that considering simply load transfer alone, YM is a parameter measured at low strain, however, load transfer into fibre is more limited by fibre length at higher strain where failure will occur and therefore, TS could be expected to be more affected by fibre length than YM. However, this is a simple approach considering the number of processes that could be occurring at failure versus at low strain. Failure strain (FS) of PP was found to be greater than 500% (the maximum range of the tensile tester), but was found to be reduced drastically by the addition of fibre to about 2% for 40 wt% fibre composites and 1.3% for 50 wt% fibre composites (Figs. 6 and 7). The FS of both 40 and 50 wt% fibre composites increased exponentially with increased number of times the materials were reprocessed. Flexural tests, impact tests and hardness tests were also carried out for 40 wt% fibre composites. Flexural strength and flexural modulus were found to decrease with increased number of times the materials were reprocessed (Figs. 8 and 9). The virgin 40 wt% fibre composites showed an average flexural strength and flexural modulus of 57 and 7200 MPa, respectively, which reduced after being reprocessed eight times to 40 and 5700 MPa, respectively (Figs. 8 and 9). Impact strength was also found to decrease for both PP and composites containing 40 wt% fibre during reprocessing from 10.5 and 6.2 kJ/m2, respectively, for virgin PP and composites
8
30 Reduction of property (%)
Fig. 3. Tensile strength of virgin and reprocessed composites (50 wt% fibre).
8000
Young's modulus (MPa)
ð3Þ
7000 6000 5000 4000 3000
25
TS
YM
20 15 10 5 0
2000
1
2
1000 0 Virgin
1
2 3 4 5 6 Number of times reprocessed
7
8
3 4 5 6 Number of times reprocessed
7
8
Fig. 5. Reduction in TS and YM of composites during reprocessing (40 wt% fibre).
Fig. 4. Young’s modulus of virgin and reprocessed composites (50 wt% fibre).
6 Table 1 Melt flow index of composites and PP Type
Melt flow index (g/10 min)
PP Composite with 40 wt% fibre Composite with 50 wt% fibre
29.40 0.90 0.14
Failure strain (%)
5 4
y = 1.6701e0.1211x R2 = 0.9761
3 2 1
higher levels of reprocessing the latter appears to have an overriding influence. The trend of TS for the 40 wt% composites versus the number of times the materials were reprocessed (Fig. 1) was found to be closely represented by the empirical equation:
0 Virgin
1
2 3 4 5 6 Number of times reprocessed
7
8
Fig. 6. Failure strain of virgin and reprocessed composites (40 wt% fibre).
1095
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
12
4.5
Composite
Impact strength (kJ/m 2)
4 Failure strain (%)
3.5 y = 1.051e0.1366x R2 = 0.9701
3 2.5 2 1.5
PP
10 8 6 4 2
1 0
0.5
Virgin
2 4 6 Number of times reprocessed
0 Virgin
1
2
3 4 5 6 Number of times recycled
7
8
8 Fig. 10. Impact strength of virgin and reprocessed composites and PP (40 wt% fibre).
Fig. 7. Failure strain of virgin and reprocessed composites (50 wt% fibre).
12
Vicker's hardness number
Flexural strength (MPa)
70 60 50 40 30 20 10
10 8 6 4 2 0 Virgin
0 Virgin
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
Number of times reprocessed
Number of times reprocessed Fig. 11. Vicker’s hardness of virgin and reprocessed composites (40 wt% fibre). Fig. 8. Flexural strength of virgin and reprocessed composites (40 wt% fibre).
Flexural modulus (MPa)
9000 8000 7000 6000 5000 4000 3000 2000 1000 0 Virgin
1
2
3
4
5
6
7
8
Number of times reprocessed Fig. 9. Flexural modulus of virgin and reprocessed composites (40 wt% fibre).
to 6.2 and 3.2 kJ/m2, respectively, after the materials were reprocessed eight times (Fig. 10). The hardness of the composites increased with reprocessing (Fig. 11). Generally, fibre was observed to be well dispersed in virgin composites (Fig. 12). With increased reprocessing, interfacial bonding was found to improve (Fig. 13) with less fibre pull out observed for eight times reprocessed composites than the virgin composites fracture surface. Although better interfacial bonding would be expected to lead to better mechanical properties after reprocessing, TS, YM, flexural strength and flexural modulus were found
to decrease. A major reason for the reduction of these properties is likely to be due to the fact that reprocessing incurred some fibre damage. The average fibre length was found to decrease from 2.36 mm for virgin fibre to 0.37 mm for the fibre extracted from the 40 wt% fibre composites reprocessed eight times (Fig. 14). In addition, the amount of fibre fines (fibre length less than 0.20 mm) was found to increase (Fig. 15) and the length distribution of fibres became narrower and reduced to shorter fibre lengths (Fig. 16) with increased number of times the materials were reprocessed. The shorter fibre lengths and the increased fines percentage with reprocessing were also observable by optical light microscopy (Fig. 17). The reductions of fibre length would be expected to reduce reinforcing efficiency, leading to the observed reduction in TS, YM, flexural strength and flexural modulus [20]. The increase in FS was likely to be due to reduced constraint from the shorter fibre and the reduction of micro-voids as evaluated by the increased density of composites (Fig. 18). The reduction of impact strength could be due to the reduction of molecular weight of PP during reprocessing. The increase in hardness may have resulted from the reduction of micro-voids and the increase in composite density with increased reprocessing. The change in average fibre length can be correlated with the number of times the composite materials were reprocessed by the following empirical equation:
lN ¼ l0 ebN
ð4Þ
where lN is the average fibre length at any reprocessed composites, l0 is the length of virgin fibre, and b is the slope of fibre length versus number of times reprocessed graph.
1096
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
Fig. 12. Optical micrographs of 40 and 50 wt% composites surface showing fibre distribution in composites (magnification 200).
Fig. 13. SEM of composite fracture surface showing less pull out for reprocessed composites (magnification for a = 3.03k and b = 3.5k).
1
rg
8
in Vi rg
Virgin 2 4 6 composite Number of times reprocessed
Vi
Virgin fibre
co m
in
0
po
fib
si te
re
0.5
8
1.5
y = 3.1406e0.5785x R2 = 0.9787
6
y = 3.4564e-0.3545x R2 = 0.9802
2
4
Fines (%)
Length (mm)
2.5
120 100 80 60 40 20 0
2
3
Number of times reprocessed
Fig. 14. Weighted average fibre length of virgin fibre and the fibre extracted from composites.
Fig. 15. Fines (fibres with lengths less than 0.20 mm) in the virgin fibre and the fibre extracted from composites.
Rearranging Eq. (4) gives:
l0 ¼ bN ln lN
ð5Þ
and combining with Eq. (3) gives TS as a function of the ratio of the original fibre length to the reduced length that has occurred with reprocessing as follows:
rN
a l0 ¼ r0 þ ln b lN
ð6Þ
The theoretical TS calculated by using Eq. (6) is compared with experimental values in Fig. 19 and shows good agreement. 3.2. Effects of reprocessing on thermal stability and crystallinity The DSC curves for PP and 40 wt% fibre composites are shown in Figs. 20 and 21 and demonstrate the reduction of melting temperature (Tm) with reprocessing. The percentage crystallinity was obtained from the relation [21]:
1097
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
Virgin composite Population (%)
Population (%)
Virgin fibre 15 10 5 0
20 15 10 5 0
0.2 1 1.8 2.6 3.4 4.2 5 5.6 Length (mm)
0.2 1 1.8 2.6 3.4 4.2 5 5.6 Length (mm) 8 times reprocessed composite Population (%)
Population (%)
4 times reprocessed composite 30 20 10 0
60 40 20 0
0.2 1 1.8 2.6 3.4 4.2 5 5.6 Length (mm)
0.2 1 1.8 2.6 3.4 4.2 5 5.6 Length (mm)
Fig. 16. Fibre length distribution of virgin fibre and the fibre extracted from composites.
Fig. 17. Optical micrographs of 40 wt% composites surface showing fibre distribution in composites (magnification 200).
45
1.4
40
Tensile strength (MPa)
Density (g/cm3)
1.2 1 0.8 0.6 0.4
35 30 25 20 15 10
Theoretical
5
0.2
Experimental
0
0 Virgin
2
4
6
0
8
6
8
10
Fig. 19. Theoretical and experimental value of TS (verification of Eq. (6)) for mixture of virgin and reprocessed composites.
Fig. 18. Density of virgin and reprocessed composites.
DH 100 W DH m
4
Number of times reprocessed
Number of times reprocessed
% Crystallinity ¼
2
ð7Þ
where DH and DHm are the heats of fusion of PP and 100% crystalline PP, respectively, and W is the fraction of PP in the composite. Taking DHm = 148 J/g for PP [22], the crystallinity of virgin and reprocessed PP and that of the PP in the composites was calculated and is presented in Table 2. The crystallinity of PP only samples and PP in the composites increased with increased number of times the materials were reprocessed (Table 2) which is likely to be as a result
of molecular weight reduction [23]. The increase in crystallinity is also supported by the XRD traces (Figs. 22 and 23). The Tm for PP alone showed very little change during reprocessing (Table 2). The Tm for PP was found to increase slightly from 169.2 °C for PP only samples to 170.5 °C for PP in the virgin composites, which is likely to be due to the fibres acting as nucleating sites and increasing the crystallinity [24], however, the Tm for PP in composites decreased slightly with increased number of times the materials were reprocessed which could be due to the reduction of molecular weight [23].
1098
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
-2
Heat flow (mW)
160
170
180
190
-4 -5 -6 Virgin PP
-7 -8
8 times reprocessed PP
-9
Number of counts
600
-3150
Virgin composite
500 400
8 times reprocessed composite
300 200 100 0 10
-10
15
20 Angle (2θ)
Temperature (oC)
Heat flow (mW)
Fig. 20. DSC curves of virgin and reprocessed PP.
-1 -1.5120 -2
130
140
150
160
170
180
190
-3 -3.5 -4.5 -5
8 times reprocessed composites Temperature (oC)
Fig. 21. DSC curves of virgin and reprocessed 40 wt% fibre composites.
Table 2 Melting point and crystallinity (calculated from DSC analysis) of PP only samples and the PP in composites Melting point Tm (°C)
Crystallinity (%) of PP
Virgin PP Four times reprocessed PP Eight times reprocessed PP
169.2 168.8 168.7
38.8 39.0 40.1
Virgin composite Four times reprocessed composite Eight times reprocessed composite
170.5 167.7 167.0
43.7 44.9 45.2
800 Virgin PP
Number of counts
8 times reprocessed PP
600
ð8Þ
where y is the fraction of nonvolatilized material, not yet decomposed, Tmax is the temperature of maximum reaction rate, b is the heating rate, Z is the frequency factor and Ea is the activation energy. Initially, plots of ln ln(1/y) versus 1/T for various stages of decomposition were drawn such as in Fig. 28 for the first stage of PP decomposition and found to be linear, suggesting good agreement with the Broido equation. The activation energies, Ea determined from the slopes of these plots are given in Table 3. Tmax
Temp. diff. / weight (oC/mg)
Sample
700
weight loss for oxidative decomposition and finally slow decomposition corresponding to formation of char as the temperature increased [25]. Kinetic parameters for the various stages of thermal degradation were determined from the TGA graphs using the following equation, given by Broido [26]:
1 Ea RZ 2 ln ln ¼ þ ln Tm y Ea b RT
Virgin composites
-5.5
30
Fig. 23. XRD traces of virgin and reprocessed 40 wt% fibre composites showing higher intensity for reprocessed composites.
-2.5
-4
25
6 Virgin PP
5
8 times reprocessed PP
4 3 2 1 0 50
150
250
350
450
550
Temperature (oC)
500 400
Fig. 24. DTA traces for virgin and reprocessed PP.
300 200 100 7
15
20 Angle (2θ)
25
30
Fig. 22. XRD traces of virgin and reprocessed PP showing higher intensity for reprocessed PP.
Typical TGA and DTA traces for virgin PP as well as reprocessed PP and the composites are shown in Figs. 24–27. Three stages of decomposition were observed for all composites and PP samples (Figs. 24 and 25) starting with dehydration and decomposition of volatile components at low temperatures, followed by rapid
Temp. diff./weight ( oC/mg)
0 10
Virgin composites
6
8 times reprocessed composites
5 4 3 2 1 0 50
150
250 350 Temperature (oC)
450
550
Fig. 25. DTA traces for virgin and reprocessed 40 wt% fibre composites.
1099
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
40 20 0 0
100
200
300
400
500
600
o
Temperature ( C) Fig. 26. TGA curves of virgin and reprocessed PP.
Fig. 29. Tensile strength of composites produced from the mixture of virgin and reprocessed composites (40 wt% fibre).
Virgin composites
100
50%Virgin+50% 8 times reprocessed
60
50% Virgin+50% 6 times reprocessed
8 times reprocessed PP
50%Virgin+50% 4 times reprocessed
80
Experimental Theoretical
50%Virgin+50% 2 times reprocessed
Weight (%)
100
50 45 40 35 30 25 20 15 10 5 0 100% Virgin
Vrigin PP
Tensile strength (MPa)
120
Weight (%)
80 8 times reprocessed composites
60 40 20 0 0
100
200 300 400 Temperature (oC)
500
600
Fig. 27. TGA curves of virgin and reprocessed 40 wt% fibre composites.
3.3. Mixing virgin and reprocessed composites
0 0.0016 -1
0.0017
0.0018
0.0019
0.002
0.0021
The TS of composites prepared with a mixture of 50 wt% virgin and 50 wt% (two to eight times) reprocessed composites are shown in Fig. 29, and it can be seen that TS decreased from 41 MPa for virgin composites to 37.5 MPa for 50 wt% virgin + 50 wt% eight times reprocessed composites. TS of the mixture would be expected to follow the Rule of Mixtures equation Eq. (2) rewritten for clarity as follows:
lnln(1/y)
-2 -3 -4
and Ea for PP and composites were generally found to increase with increased number of times the materials were reprocessed (Table 3). The positions of weight loss on the TGA traces both for PP and composites shifted to higher temperatures with increased number of times the materials were reprocessed (Figs. 26 and 27) suggesting increased thermal stability. The increase in thermal stability is likely to be due to the increase in crystallinity of PP [27] resulting from molecular weight reduction [23] and better interfacial bonding during reprocessing supported by observations of fracture surfaces (Fig. 13).
y = -12399x + 19.838 R2 = 0.993
rmix ¼ ðr0 v0 þ rN vN Þ
-5 -6 1/T (K) Fig. 28. 1/T versus ln ln(1/y) for the first decomposition stage of PP.
ð9Þ
where r0 and rN are the TS of virgin and reprocessed composites, v0 and vN are the volume fraction of virgin and reprocessed composites, and N is the number of times the materials were reprocessed. Good agreement with the Rule of Mixtures equation was obtained (Fig. 29).
Table 3 Thermal properties of PP and composites Sample
Stage
Weight loss (%)
Temperature range (°C)
Tmax (°C)
Activation energy Ea (kJ/mol)
Virgin PP
First Second Third
26 70 3
224–320 320–445 445–534
293 375 504
104 78 43
Eight times reprocessed PP
First Second Third
18 79 2
224–328 320–445 445–534
285 390 506
115 84 50
Virgin composite
First Second Third
61 31 7
226–351 351–436 436–508
285 371 455
85 68 60
Eight times reprocessed composite
First Second Third
30 52 15
230–340 340–447 470–512
289 412 470
87 71 81
1100
M.D.H. Beg, K.L. Pickering / Composites: Part A 39 (2008) 1091–1100
4. Conclusions For 40 wt% fibre reinforced composites reprocessed eight times, a 25% reduction in TS and 16% reduction in YM was found. Also for these composites, impact strength, flexural strength and flexural modulus were found to decrease with increased number of times the materials were reprocessed. Reprocessing of 50 wt% fibre composites brought about an increase in both TS and YM from 37 and 4830 MPa, respectively, for virgin composites to 42 and 6421 MPa for two times reprocessed composites followed by decreases in TS and YM upon further reprocessing. The reduction of TS, YM, flexural strength, flexural modulus and impact strength was considered to be due to fibre damage that occurred during reprocessing as evaluated by the associated reduction of the average fibre length from 2.36 mm for virgin fibre to 0.37 mm for the fibre extracted from the composites reprocessed eight times. The FS was found to increase with increased number of times the materials were reprocessed due less fibre constrain and the reduction of voids as evaluated by the increase in composite density. Composite thermal stability was found to increase with increased reprocessing believed to be due to the improvement of interfacial bonding and increased crystallinity of PP resulting from the reduction of its molecular weight. TS of composites made with mixtures of virgin and reprocessed material were in close agreement with the Rule of Mixtures. However, reprocessing of composites subjected to actual weathering would give better understanding of the real-world performance. Acknowledgements The authors would like thanks to the Foundation for Research Science and Technology, New Zealand for their financial support of this project. References [1] Espert A, Vilaplana F, Karlsson S. Comparison of water absorption in natural cellulosic fibres from wood and one-year crops in polypropylene composites and its influence on their mechanical properties. Compos Pt A: Appl Sci Manuf 2004;35(11):1267–76. [2] Peijs T, Melick V, Garkhail SK, Pott GT, Baillie CA. Natural-fibre-mat reinforced thermoplastics based on upgraded flax fibres for improved moisture resistance. In: ECCM-8 conference; 1998. p. 119–26. [3] Tajvidi M, Ebrahimi G. Water uptake and mechanical characteristics of natural filler–polypropylene composites. J Appl Polym Sci 2003;88:941–6. [4] Avella M, Bozzi C, Dell’Erba R, Focher B, Marzetti A, Martuscelli E. Steam exploded wheat straw fibres as reinforcing material for polypropylene-based composites. Angew Makromol Chem 1995;233:149–66.
[5] Pickering SJ. Recycling technologies for thermoset composite materials current status. Compos Pt A 2006;37(8):1206–15. [6] Jones R, Baumann MH. Recycling of engineering plastics. In: ANTEC conference proceedings, vol. 4; 1997. p. 3066–74. [7] Martins MH, Paoli A. Polypropylene compounding with recycled material. I: Statistical response surface. Polym Degrad Stabil 2001;71(2):93–8. [8] Selke SE. Plastics recycling. Handbook of plastics, elastomers and composites. 4th ed. New York: McGraw- Hill; 2002. p. 693–757. [9] Scott G. Green polymers. Polym Degrad Stabil 2000;68:1–7. [10] Guerrica-Echevarria G, Eguiazabal JI, Nazabal J. Effects of reprocessing conditions on the properties of unfilled and talc-filled polypropylene. Polym Degrad Stabil 1996;53(1):1–8. [11] Xiang Q, Xanthos M, Mitra S, Patel SH, Guo J. Effects of melt reprocessing on volatile emissions and structural/rheological changes of unstabilized polypropylene. Polym Degrad Stabil 2002;7(1):93–102. [12] Martins MH, De Paoli M-A. Polypropylene compounding with post-consumer material. II: Reprocessing. Polym Degrad Stabil 2002;78(3):491–5. [13] Rowell RM, Young RA, Rowell JK. Paper and composites from agro-based resources. Lewis Publishers; 1996. [14] Walz K, Jacobson R, Sanadi AR. Effect of reprocessing/recycling on the mechanical properties of kenaf-PP composites. Internal report, University of Wisconsin-Madison and Forest Product Laboratory; 1994. [15] Joseph K, Thomas S, Pavithran C, Brahmakumar M. Tensile properties of short sisal fibre-reinforced polyethylene composites. J Appl Polym Sci 1993;47:1731–9. [16] Youngquist JA, Myers GE, Muehl JH, Krzysik AM, Clemens CM. Composites from recycled wood and plastics. US Environmental Protection Agency, Cincinnati, OH 45268.