Green composites of polypropylene and eggshell: Effective biofiller ...
Composites Science and Technology 102 (2014) 152–160
Contents lists available at ScienceDirect
Composites Science and Technology journal homepage: www.elsevier.com/locate/compscitech
Green composites of polypropylene and eggshell: Effective biofiller size reduction and dispersion by single-step processing with solid-state shear pulverization Krishnan A. Iyer a, John M. Torkelson a,b,⇑ a b
Department of Chemical and Biological Engineering, Northwestern University, Evanston, IL 60208, United States Department of Materials Science and Engineering, Northwestern University, Evanston, IL 60208, United States
a r t i c l e
i n f o
Article history: Received 15 May 2014 Received in revised form 6 July 2014 Accepted 31 July 2014 Available online 8 August 2014 Keywords: A. Polymer–matrix composites (PMCs) A. Recycling A. Particle-reinforced composites B. Mechanical properties B. Thermal properties
a b s t r a c t Eggshell (ES), a waste byproduct from food processing and hatcheries, contains 95% calcium carbonate (CC), making it a potentially attractive, less expensive substitute for commercial CC. Past work used complex grinding–sieving and/or chemical modification steps to aid in dispersing ES in polymers such as polypropylene (PP). Both steps add to the cost and reduce the green aspect of the composite. Here, green composite materials of PP with 5–40 wt% unmodified ES shards of several centimeters in size are directly processed using continuous, single-step solid-state shear pulverization (SSSP). Electron microscopy and particle size analysis show very good dispersion with some ES particles near the nanoscale in the composite. Well-dispersed ES particles dramatically increase PP crystallization rates with a 5–7% increase in PP crystallinity. The very good dispersion leads to a major increase in Young’s modulus (87% increase relative to neat PP for 40 wt% ES) and a modest increase in hardness; composites exhibit reductions in yield strength, elongation at break, and impact properties. Mechanical and crystallization properties are equal to or better than the best literature data for PP/ES composites without chemical modification made by multi-step approaches involving melt processing. In addition, the composites exhibit high thermal degradation temperatures compared to neat PP, indicating the potential for ES to improve processing stability. Composites with 20–40 wt% ES exhibit solid-like rheological response with no crossover of shear storage and loss moduli. Nevertheless, PP/ES composites retain viscosities close to that of neat PP at shear rates experienced in melt processing. Overall, property enhancements resulting from superior dispersion of ES in PP achieved by SSSP reveal ES to be a promising green filler for thermoplastics. Ó 2014 Published by Elsevier Ltd.
1. Introduction Polymer composites with inorganic fillers such as glass fibers, silica nanoparticles, graphene, carbon nanotubes and nanofibers have been studied extensively over the past two decades [1–12]. Inorganic mineral fillers like calcium carbonate (CC), silica, and talc have gained interest as low cost fillers for thermoplastics [13–21]. These fillers offer significant enhancement in stiffness, crystallization rate and thermal stability [8,22–27]. There has been growing demand for green and renewable substitutes for inorganic fillers [28–33]. Such fillers offer major advantages such as low density as well as reduced cost and mechanical wear during processing. ⇑ Corresponding author at: Department of Chemical and Biological Engineering, Northwestern University, Evanston, IL 60208, United States. Tel.: +1 847 491 7449; fax: +1 847 491 3728. E-mail address: [email protected] (J.M. Torkelson). http://dx.doi.org/10.1016/j.compscitech.2014.07.029 0266-3538/Ó 2014 Published by Elsevier Ltd.
Widely investigated green fillers include natural fibers, rice husk, and wood flour [29–33]. As reported by CNN in 2012, there is an interest among food processing companies to find uses for chicken eggshells (ESs) [34]. Approximately 250,000 tons of ES are produced world-wide by food processing industries annually [35]. Most ES waste is disposed in landfills or turned into low value protein supplements for animal feed. Eggshells typically contain about 95% CC with the remainder being organic material [36–38]. This makes waste ES an excellent source of bio-mineral CC; moreover, the abundant supply makes ES an attractive candidate for replacing CC in thermoplastic composites. Despite its potential as a green alternative to CC, fewer than twenty studies have investigated the possibility of exploiting ES as a composite filler [39–54], about one-half of which focused on polyolefins as the matrix [39–46]. A patent for the preparation of PP/ES composites by melt extrusion was registered by Universidad
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
de Chile in 2006 [40]. That patent and subsequent papers by Toro et al. [41,42] describe a detailed grinding procedure using a flatmetal-blade crusher to produce particles which are sieved using an ASTM 100 mesh. The resulting ES powder was dried at 100 °C for 8 h and then ground further in a concentric-metal-ring mill to obtain particles that pass through an ASTM 400 mesh. The resulting ES powder contained particles of variable size, 50 vol% of which was less than 8.4 lm and 90 vol% of which was less than 27.5 lm. This material was melt mixed with PP to obtain composites with 40 wt% ES. The end product resulting after these multiple steps exhibited an 85% increase in Young’s modulus, a 12% decrease in yield strength, and a 47% decrease in impact strength. Ghabeer et al. [42] investigated the effect of chemical modifications on ground ES powder with an average particle size of 90 lm. The ES powder was treated with stearic acid in an attempt to improve its adhesion with PP matrix. Composites containing 40 wt% untreated ES powder exhibited no change within error in Young’s modulus or PP crystallinity. Treatment of ES with stearic acid increased the Young’s modulus of the composite (relative to neat PP) by 200% due in part to increased crystallinity of the sample. Kumar et al. [45] used isophthalic acid to chemically treat ES and observed 3–18% increases in tensile modulus in PP composites using the modified ES relative to those using unmodified ES or CC. In order to increase the composite impact strength, Lin et al. [44] examined the use of pimelic acid to convert CC in ES into a b-nucleating agent for PP crystallization and observed a 228% increase in impact strength. Supri et al. [43] showed a 130% enhancement in Young’s modulus for polyethylene (PE)/ES composites using PE grafted maleic anhydride as compatibilizer while Sutapun et al. [46] observed improvements in composite impact strength when using PE grafted maleic anhydride relative to unmodified PE. In spite of these property enhancements, ES has failed to become a major commercial filler due in part to the complex steps required to achieve good dispersion of ES. Batch grinding–sieving steps add to cost and chemical modifications used in attempts to improve ES dispersion impair the ‘‘green’’ aspect of ES composites (Fig. 1a).
153
Here, we investigate the potential of using a single-step, continuous, and industrially scalable process called solid-state shear pulverization (SSSP) [11,12,55–69] to produce well-dispersed, green composite materials of PP/ES with significant property enhancements (Fig. 1b). Although the current study employs a lab-scale/ pilot-plant apparatus, polyolefins have been processed at throughputs exceeding 150 kg/h using a commercial-scale SSSP apparatus at Northwestern University [66]. The SSSP apparatus is a modified twin-screw extruder in which materials are cooled rather than heated, so that the polymer is processed below its glass transition temperature if amorphous or below its melting temperature if semicrystalline. The use of near-ambient temperature aids in exposing polymers to higher forces and stresses in the solid state than normally encountered in conventional twin-screw melt extrusion. The absorption of mechanical energy during SSSP results in material fracture followed by random fusion of the materials, which is repeated many times during the average residence time of the material in the pulverizer. SSSP eliminates the common limitations of thermodynamics, viscosity, and degradation encountered in melt processing of polymers. Previously, SSSP has been used to produce well-dispersed polymer nanocomposites with fillers such as clay, graphite and carbon nanotubes [11,12,57–59]. In addition, SSSP has been used to produce compatibilized blends systems in which the dispersed-phase size approaches 100 nm [60,61]. The formation of trace levels of block copolymers at the polymer blend interface resulting from recombination of polymer radicals has been cited as the reason for immiscible blend compatibilization observed in SSSP [62–64]. Green composites of PP with starch and rice husk ash have been produced with SSSP [65,66]. Recently, SSSP was used to functionalize PP with maleic anhydride and ester moieties without significant reduction in PP molecular weight, which accompanies industrial melt-state processes employed to achieve such grafting [67,68]. This work explores the potential of using the single-step SSSP process in the novel production of green PP/ES composite materials. As shown in Fig. 1b, the current study utilizes ES shards that are several centimeters in size without any chemical modification.
Fig. 1. (a) Previous approach to produce PP/ES composites involves multiple grinding/sieving and chemical modification steps to produce ES particles which are further melt processed with PP to yield the final composite; (b) single-step solid-state SSSP with input of unmodified, several-centimeter ES shards produces sub-micron ES particles in the final PP/ES composite material.
154
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
The SSSP process circumvents the grinding–sieving processes and chemical modification previously thought to be necessary for good dispersion of ES (Fig. 1a). Particle size analysis demonstrates very good dispersion and size reduction of ES achieved via SSSP. Crystallization and mechanical properties are measured to characterize property enhancements produced by well-dispersed ES particles in the PP matrix. Improvements in thermal degradation behavior due to the presence of well-dispersed ES in PP are observed in air and nitrogen environments. Rheological measurements are performed to quantify the effect of ES particles on composite viscosity, which is important to the production of final composite products by extrusion or injection molding. Where appropriate, comparisons of the properties of SSSP-prepared PP/ES composites are drawn with literature data for PP/ES composites prepared by melt processing. 2. Experimental section 2.1. Preparation of PP/ES composites Polypropylene pellet with density of 0.905 g/cm3 and MFI of 9 g/10 min (ASTM D1238, 230 °C and 2.2 kg) was from Total Petrochemicals. Eggshell shards of several centimeters size were procured locally and used directly after washing briefly with water to remove dirt. The PP pellets and varying amounts of ES were dry blended and fed to a Berstorff ZE-25 pulverizer using a K-tron S-60 feeder. The pulverizer has a section with a 25-mm-diameter barrel containing spiral conveying and bilobe kneading elements (mixing), resulting in intimate mixing between PP and ES. Following this is a small section where the barrel diameter changes from 25 to 23 mm. (SSSP can be done with all sections having a 25 mm diameter barrel.) The 23 mm section contains trilobe shearing elements, in which the bulk of the pulverization and particle size reduction occurs. Depending on the motor running the apparatus, the pulverizer can be run with all bilobe elements. The SSSP barrels were cooled by a recirculating ethylene glycol/ water mix at 7 °C supplied by a Budzar Industries WC3 chiller. The screw was designed to impart high specific energy input to the material, with the mixing zone containing one reverse, two neutral, and three forward kneading elements and the pulverization zone containing three forward, two neutral, and two reverse shearing elements [55]. A screw speed of 200 RPM and a feed rate of 80 g/h were used to produce a fine powder. 2.2. Characterization of PP/ES composites Field emission-scanning electron microscopy (FE-SEM) samples were prepared by melt extrusion at 200 °C via a MiniMax molder. The morphologies of the cryofractured sections of PP/ES composites were obtained via a Hitachi SU 8030 FE-SEM after sputter coating (Cressington 208HR) with gold. The number-average particle size distribution was quantified based on a minimum of 150 particles using ImageJ. Uniaxial tensile test and hardness test samples were prepared by compression molding the SSSP powder in a PHI (model 0230C-X1) press at 200 °C for 5 min with a 5 ton ram force followed by cooling immediately in a cold press at 10 °C under 5 ton ram force. Tensile coupons were cut using a standard Dewes-Gumbs dogbone die. Samples were equilibrated at room temperature for 48 h and then tested (ASTM D 1708) using an MTS Sintech 20/G tensile tester equipped with a 100 kN load cell at a crosshead speed of 50 mm/min. Seven samples were tested for each composite; the results were averaged to obtain a mean value. Hardness tests were performed on sections with a Vickers diamond indenter, using a 250 mN load for 10 s. Results were
reported as the mean value of ten indentations. Unnotched impact test bars of 62 mm in length, 12.6 mm width and 3.5 mm width were prepared by using an injection molding machine (Morgan press) with barrel and nozzle temperatures of 200 and 210 °C. (The input feed to the injection molding machine were composite pellets made by single-screw melt extrusion of the SSSP composite powder followed by pelletization.) A clamp force of 10 tons and injection pressure of 4.5 105 psi were used to produce the impact test bars. A Tinius-Olsen IT504 pendulum tester was used in an unnotched Izod setup (ASTM D4812). Crystallization of PP in the composites and the neat state was characterized using a Mettler Toledo 822e differential scanning calorimeter (DSC). After heating to 200 °C, a 10 °C/min cooling rate was employed to determine the non-isothermal crystallization onset temperature of PP in the composite. In order to calculate the crystallinity of PP in the composite, the specific enthalpy derived from the area associated with the crystallization portion of the nonisothermal cooling curve was divided by the mass fraction of PP in the composite; this value was subsequently divided by the theoretical heat of fusion for neat PP of 207.1 J/g [70]. Isothermal crystallization half-times were also determined at 140 °C using DSC after cooling samples at a rate of 40 °C/min from 200 °C. Thermal degradation behavior was monitored using a thermogravimetric analyzer (Mettler Toledo 851e) calibrated with an indium/aluminum standard. Composite samples were heated to 700 °C using a 10 °C/min heating ramp in both N2 and O2 environment to determine thermal and thermoxidative degradation behavior of PP in the PP/ES composites. Final ash at 700 °C was used to establish ES content in the composites. Rheological behavior was characterized using a TA Instruments ARES rheometer with a 25 mm parallel-plate fixture at 200 °C. Test samples were prepared using a compression molding protocol similar to that used for producing mechanical test samples. Small amplitude oscillatory shear measurements were performed as a function of frequency from 0.01 to 100 rad/s. 3. Results and discussion 3.1. Morphology FE-SEM images of fractured surfaces of PP/ES composites are shown in Fig. 2. Samples prepared by SSSP show effective dispersion and size reduction of ES in PP even at 30 wt% filler loading. Very good dispersion without filler aggregation is critical in achieving excellent properties, good processability, and optimum appearance of final composite products [21,40]. Small particle size aids in uniform stress transfer between filler and matrix, thereby resulting in a composite with superior mechanical performance. The solidstate nature of the materials allows SSSP to impart large forces and stresses to the material, resulting in intimate mixing of both components with a reduction of several-centimeter ES shards to sub-micron dimensions. Fig. 2 shows ES particles embedded in PP with some particles as small as 100–200 nm in size. Particle size analysis is summarized in Table 1. The PP/ES composites exhibit small particle size, with 78–85% of the particles being submicron and only 5–8% larger than 5 lm. In short, SSSP is a simple, single-step process yielding well-dispersed PP/ES composite materials without need for grinding–sieving or chemical modification formerly thought to be crucial for achieving effective size reduction and dispersion of ES within polymers. 3.2. Crystallization behavior The surfaces of well-dispersed fillers and nanofillers serve as strong heterogeneous nucleation sites for PP crystallization. The
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
155
Fig. 2. FE-SEM images of ES particles dispersed in the PP matrix by SSSP processing: (a) 90/10 wt% PP/ES, (b) 80/20 wt% PP/ES, (c) 70/30 wt% PP/ES, (d) ES particle embedded in the PP matrix.
Table 1 Particle size distribution of ES filler in PP/ES composites produced by SSSP. Sample
5 lm (%)
90/10 wt% PP/ES 80/20 wt% PP/ES 70/30 wt% PP/ES
85 80 78
10 15 14
5 5 8
ability of filler to act as nucleating agent depends on particle size, surface area and dispersion [15,18,22,24]. Fig. 3 shows isothermal crystallization curves at 140 °C which demonstrate that the incorporation of well-dispersed ES filler reduces crystallization halftime dramatically from 28 min for neat PP (with no exposure to SSSP) to 3 and 2 min for composites made by SSSP with 30 and 40 wt% ES, respectively. Table 2 summarizes non-isothermal and isothermal crystallization half-time behavior of PP in PP/ES composites. The strong nucleating ability of ES can be seen from the significant increase in the nonisothermal crystallization onset temperature (Tc,onset) from DSC cooling curves. Relative to neat PP, Tc,onset increases 2–4 °C for 5–10 wt% ES and 11 °C for 40 wt% ES. Further evidence of synergistic property enhancements is demonstrated by the isothermal crystallization behavior. Size reduction to sub-micron dimensions and relatively homogenous dispersion of ES fillers in the composite contribute to the ability of ES to act as a strong nucleating agent. In composites prepared by melt processing, CC and ES particles exhibit good nucleating effects on PP crystallization: a 10–11 °C increase in Tc,onset for filler loadings of 10–40 wt% ES was reported by Ghabeer et al. [42]. However, the presence of particle agglomerates in melt-mixed samples impedes crystal growth and suppresses the increment in crystallization rate. As a result, due to limited dispersion of untreated ES from melt processing, the
Fig. 3. Isothermal crystallization (140 °C) curves of neat PP (with no processing by SSSP) and PP/ES composites produced by SSSP.
reduction in crystallization half-time reported by Ghabeer et al. [42] was modest compared to those of SSSP-produced composites. Relative to neat PP, composites prepared by SSSP exhibit an increase in PP crystallinity, from 49% for neat PP to 56% for the 70/30 wt% PP/ES composite using unmodified ES. Such an increase is likely due to the large amount of ES surface present in the welldispersed composites, which acts as a very effective nucleating site
156
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
Table 2 Crystallization behavior of PP/ES composites produced by SSSP.
a
Material
Tc,onset (°C)
PP crystallinity (%)
Isothermal PP crystallization half-time (min) at 140 °C
Neat PPa 95/5 wt% PP/ES 90/10 wt% PP/ES 80/20 wt% PP/ES 70/30 wt% PP/ES 60/40 wt% PP/ES
122 124 126 133 133 133
49 51 52 55 56 53
28 22 19 4 3 2
With no processing by SSSP.
for PP crystallization. In contrast, Ghabeer et al. [42] reported no increase in PP crystallinity with the incorporation of up to 40 wt% unmodified ES. In the same study, a 6% increase in crystallinity of PP was observed after surface treatment of ES with stearic acid. Supri et al. [43] showed a 6–7% increase in crystallinity after a series of grinding steps and modification of ES using NaOH–isophthalic acid. Notably, incorporation of unmodified ES by singlestep SSSP results in higher PP crystallinity and vastly increased crystallization rates. 3.3. Mechanical properties – tensile, impact and hardness Table 3 summarizes mechanical property enhancements provided by well-dispersed ES in PP. Introducing well-dispersed, rigid ES filler restricts the mobility and deformability of the PP matrix, thereby increasing Young’s modulus of the composite. Literature reports have indicated that good dispersion and reduction in ES particle size are necessary for achieving such increments in Young’s modulus [39,41,42]. Consistent with excellent dispersion of rigid ES particles and improvements in PP crystallinity, PP/ES composites produced by SSSP exhibit a monotonic increase in Young’s modulus with increasing filler loading. Relative to neat PP, 80/20 wt% and 60/40 wt% PP/ES composites exhibit 38% and 87% increases in Young’s modulus, respectively. In comparison, CC, a major component in ES and a common filler for PP, has been reported to yield 60–70% increases in Young’s modulus at 43 wt% loading in PP [17]. Thus, 60/40 wt% PP/ES composites produced by SSSP exhibit mechanical performance superior to that of PP/ CC composites with similar filler loadings produced by twin-screw extrusion. Fig. 4 compares percent increments in Young’s modulus for PP/ ES composites with untreated ES particles observed in this study with those reported in literature. Toro et al. [41] observed no increase in Young’s modulus within error for all filler loadings when ES particles were relatively large (50 vol% less than 90 lm in size). Further size reduction of ES fillers, with 50 vol% less than 8.4 lm in size, resulted in 40% and 85% increases in Young’s modulus for ES loadings of 20 and 40 wt%, respectively. Ghabeer et al. [42] observed no change within (very large) error bars in Young’s modulus over neat PP for up to 40 wt% ES filler loading with particle size ranging from 14 to 125 lm. In contrast, the current study demonstrates that single-step SSSP yields PP/ES composites with
Fig. 4. Comparison of percentage change in Young’s modulus of PP/ES composites produced by SSSP relative to neat PP (with no processing by SSSP). Data by Toro et al. [41] for 90 and 8.4 lm median ES particle sizes (for composites produced by grinding and sieving followed by melt processing). (Error bars are one standard deviation.) (Note: Ghabeer et al. [42] also reported Young’s modulus data for PP/ES composites. However, the error bars associated with their data were sufficiently large to indicate that there was no change in Young’s modulus within error.)
38% and 87% increases in Young’s modulus relative to neat PP for ES loadings of 20 and 40 wt%, respectively. The results obtained with SSSP-produced samples are equal or superior to the best reported values for PP composites with unmodified ES made by grinding–sieving and melt processing. As produced by single-step SSSP, PP/ES composites exhibit yield strength values that are nearly invariant with ES loadings up to 20 wt%. Relative to neat PP, PP/ES composites exhibit only a 12– 15% decrease in yield strength for 30–40 wt% ES loading. With melt processing, Ghabeer et al. [42] observed a 37% decrease in yield strength for 40 wt% loading of untreated ES particles. In our case, the effective particle size reduction and dispersion facilitate stress transfer between the PP matrix and ES. Additionally, the uniform stress transfer due to good filler dispersion prevents stress concentration around ES embedded in the PP matrix so that SSSP-processed PP/ES composites show good retention of yield strength compared to neat PP. Due to the increasing stiffness with increasing filler loading, the material exhibits a significant reduction in elongation at break. Relative to neat PP with a 700% elongation at break, the 95/5 wt% and 80/20 wt% PP/ES composites made by SSSP exhibit 60% and 20% elongation at break, respectively. Superior dispersion by SSSP enables efficient stress transfer between PP matrix and filler, resulting in retention of ductile behavior at these compositions. Reduced impact properties for composites with PP are often observed with the incorporation of micro-size mineral fillers such
Table 3 Mechanical properties of PP/ES composites produced by SSSP.
a
Material
Young’s modulus (MPa)
Yield strength (MPa)
Elongation at break (%)
Unnotiched Izod impact strength (J/m)
Neat PPa 95/5 wt% PP/ES 90/10 wt% PP/ES 80/20 wt% PP/ES 70/30 wt% PP/ES 60/40 wt% PP/ES
1030 ± 20 1230 ± 40 1320 ± 20 1420 ± 20 1600 ± 30 1900 ± 20
33.0 ± 0.2 33.0 ± 1.0 32.0 ± 0.3 31.0 ± 0.4 29.0 ± 0.5 28.0 ± 0.3
700 ± 40 60 ± 10 30 ± 5 20 ± 5 10 ± 2 7±1
34 ± 2 29 ± 2 29 ± 2 20 ± 1 18 ± 1 12 ± 1
With no processing by SSSP.
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
157
Fig. 5 shows the increment in Vickers hardness seen for PP/ES composites produced by SSSP. Calcium carbonate is soft compared to some other mineral fillers such as silica. Thus, PP/ES composites produced by SSSP show a moderate increase in Vicker’s hardness from 100 MPa for neat PP to 130 MPa for 20–40 wt% ES loadings. 3.4. Thermal degradation – nitrogen and air
Fig. 5. Vicker’s hardness values reported in units of MPa for neat PP (with no processing by SSSP) and PP/ES composites produced by SSSP. (The numbers inside the bars represent the wt% of ES in the PP/ES composite. One standard deviation errors are ±2 MPa for each sample.)
as CC and talc [17,21]. Despite excellent particle size reduction and dispersion, the SSSP-processed composites exhibit a decrease in impact strength with increasing ES content. The 10 wt% PP/ES composite shows unnotched Izod impact strength of 29 J/m compared to 34 J/m for PP. The unnotched Izod impact strength decreases further to 20 and 12 J/m for composites with 20 and 40 wt% ES loadings, respectively. Within error bars, literature studies show equivalent or greater reductions in impact properties relative to melt-mixed composites [40,42]. Related behavior has been observed with PP–graphite nanocomposites produced by SSSP [12]. It is important to note that at 20–40 wt% ES content, the biofiller accounts for 7.5–15 vol% of the composite, increasing the probability for microcrack formation accompanying impact. These microcracks can propagate through the matrix and cause fracture at the polymer–filler interface.
Thermal degradation behavior of PP in PP/ES composites under inert nitrogen atmosphere provides information regarding thermal stability and the true filler content in the composites. Fig. 6a and Table 4 shows the thermal degradation behavior of PP in PP/ES composites under nitrogen atmosphere. Here, the temperature corresponding to 5% mass loss is defined as the onset temperature for degradation (Tdeg) for neat PP and PP in PP/ES composites under nitrogen atmosphere. Neat PP exhibits almost no mass loss up to 370 °C and 5% mass loss at 407 °C. In comparison, PP in PP/ES composite with 5 wt% ES exhibits a 16 °C increase in its Tdeg relative to neat PP. A maximum increase in Tdeg of 20 °C is observed in composites containing 30 wt% ES. The presence of well-dispersed ES in PP matrix is expected to provide a barrier to the diffusion of degradation products, suppressing PP mass loss in the composite. Table 4 reports leftover ash content after thermal degradation under nitrogen. Due to its high CC content, ES displays only 8% mass loss up to 700 °C, consistent with a 92% CC content. In contrast, PP leaves behind no measurable residue at 700 °C. Residue in composites at 700 °C correlates well with the expected ES content after accounting for the minor ES mass loss. In addition, the reproducibility of the residual ash content is consistent with effective ES dispersion. As shown in Fig. 6b, thermo-oxidative stability of PP in PP/ES composite in air was characterized to determine composite stability during processing. Polypropylene in PP/ES composites shows a minor 2–3 °C decrease in temperature corresponding to 5% mass loss. Interestingly, the thermograms demonstrate enhanced thermal stabilities for temperatures corresponding to higher mass loss. In order to better quantify the enhancement in thermal stabilities, Table 5 summarizes the temperatures corresponding to 10% (T10%) and 20% (T20%) mass loss in addition to Tmax (maximum rate of mass loss obtained from the peak value determined from the
Fig. 6. Thermogravimetric analysis curves showing degradation behavior of eggshell, neat PP (with no processing by SSSP) and PP in PP/ES composites produced by SSSP: (a) in nitrogen and (b) in air. Heating rate is 10 °C/min. (Note: At 700 °C, all PP is degraded and volatilized, i.e., normalized PP wt% = 0.) (In the figures, PP40ES denotes 60/40 wt% PP/ES composite.)
158
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
Table 4 Thermal degradation behavior under nitrogen atmosphere of neat PP (with no processing by SSSP) and PP in PP/ES composites produced by SSSP.
a b c
Material
Tdega (°C)
Final ashb (%)
ES content in compositesc (%)
Neat PP 95/5 wt% PP/ES 90/10 wt% PP/ES 80/20 wt% PP/ES 70/30 wt% PP/ES 60/40 wt% PP/ES
407 423 419 425 427 421
0 5 8 16 26 35
0 5 9 17 28 38
sub-micron dimensions increases its effective surface area and further aids such adsorption [42].
3.5. Rheology
Temperature corresponding to 5% mass loss. Residue measured at 700 °C. Corrected for 8% organic content in EG not accounted for in final ash.
Table 5 Thermal degradation behavior under air atmosphere of neat PP (with no processing by SSSP) and PP in PP/ES composites produced by SSSP. Material
T10a (°C)
T20b (°C)
Tmaxc (°C)
Neat PP 95/5 wt% PP/ES 90/10 wt% PP/ES 80/20 wt% PP/ES 70/30 wt% PP/ES 60/40 wt% PP/ES
289 292 292 307 304 294
304 310 313 338 336 324
289 370 375 406 403 389
a
Temperature corresponding to 10% PP mass loss. Temperature corresponding to 20% PP mass loss. Temperature corresponding to maximum rate of PP mass loss from derivative thermogravimetric curve. b
c
derivative thermogram) for PP in PP/ES composites. Neat PP exhibits T10% and T20% values of 289 and 304 °C, respectively. The presence of ES enhances thermal stability of PP under oxidative conditions at all filler loadings examined. Composites prepared by SSSP with 20 and 30 wt% ES show 18 and 15 °C increases in T10% relative to that of neat PP. The composites show even greater enhancements in T20% and Tmax; e.g., 80/20 wt% PP/ES shows a 34 °C increase in T20% and a 117 °C increase in Tmax. The substantial enhancement in thermal stability suggests that well-dispersed, sub-micron sized ES particles provide a mass transport barrier to oxygen diffusion and molecular mobility. In addition, the ES filler offers sites for adsorption of volatile degradation products, thereby arresting the autocatalytic degradation process. In particular, particle size reduction of ES to
Fig. 7 shows the magnitude of complex viscosity (|g*|) and shear storage modulus (G0 ) as a function of frequency measured at 200 °C for neat PP and PP/ES composites. Based on the Cox-Merz rule [71], which equates the frequency in simple oscillatory shear experiments to shear rate in simple shear flow experiments, Fig. 7a is equivalent to a plot of shear viscosity as a function of shear rate. At all shear rates greater than 1 s1, composites with ES loadings up to 10 wt% have viscosities and G0 values that are similar to those of neat PP. At 20–40 wt% ES loadings, composites exhibit higher viscosities and G0 values, especially at low frequency. Notably, at low frequencies, G0 shows only a weak frequency dependence for 20–40 wt% filler loadings, consistent with a change in rheological response from liquid-like to more solid-like [72]. Such response is characteristic of a percolated network within the PP matrix [12,73,74]. Fig. 8 shows storage modulus G0 and loss modulus (G00 ) as a function of frequency in the linear viscoelastic regime. The 90/ 10 wt% PP/ES composite exhibits classical viscoelastic behavior, with a terminal regime at low frequencies (G00 x, G0 x2; x is frequency) where G0 < G00 , and a crossover at x = 42 s1 above which G0 > G00 . Incorporation of 20 wt% ES shifts the crossover to x = 16 s1. The 70/30 wt% PP/ES composite shows a solid-like response with no crossover for all frequencies probed in this study. These results are consistent with literature results, demonstrating that filler incorporation increases the solid-like response of polymer composites and shifts the crossover point to lower x [74]. Excellent dispersion of ES in the composites produced by SSSP results in shear flow behavior similar to that of neat PP for filler loadings up to 10 wt%. Even for higher ES loadings of 20 and 30 wt%, the viscosity values are only four times higher than that of neat PP at a shear rate of 100 s1. (Shear rates of 100 s1 and higher are typically employed in conventional melt processing techniques such as twin-screw extrusion and injection molding.) Thus, these viscosity data indicate that PP/ES composite materials made by SSSP can be easily processed into final products by highshear-rate processes twin-screw extrusion and injection molding. In fact, as the samples used here for impact strength tests were made by single-screw extrusion and pelletization of the SSSP
Fig. 7. (a) Magnitude of complex viscosity (|g*|) and (b) shear storage modulus (G0 ) at 200 °C as a function of angular frequency for neat PP with no processing by SSSP (plus) and 95/5 wt% PP/ES (diamond), 90/10 wt% PP/ES (triangle), 80/20 wt% PP/ES (cross), 70/30 wt% PP/ES (circle), and 60/40 wt% PP/ES (asterisk) composites made by SSSP.
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
159
Fig. 8. Magnitude of shear storage modulus (G0 ; diamond) and loss modulus (G00 ; square) at 200 °C as a function of angular frequency for (a) 90/10 wt% PP/ES, (b) 80/20 wt% PP/ES and (c) 70/30 wt% PP/ES composites made by SSSP.
powder output followed by injection molding of the pellets, the current study proves that such processing is easily achieved using PP/ES composites produced by SSSP.
Northwestern University (ISEN) and Northwestern University. This study made use of Central Facilities supported by the MRSEC program of the National Science Foundation at the Northwestern University Materials Research Science and Engineering Center.
4. Conclusion References Single-step, continuous SSSP was employed to produce green PP/ES composite materials with 5–40 wt%. filler. The SSSP process eliminates the need for complex multi-step processing that was previously thought to be crucial in achieving good dispersion of ES within a polymer matrix. Good dispersion and particle size reduction were demonstrated by electron microscopy with the vast majority of ES particles being sub-micron in size. Synergistic property enhancements associated with dispersion and particle size reduction were observed via dramatic improvement in crystallization rate together with a 5–7% increase in PP crystallinity. Relative to neat PP, PP/ES composites produced by SSSP show up to an 87% enhancement in Young’s modulus and a 30 MPa increase in hardness. The composites exhibit reasonable retention of yield strength while maintaining ductility at substantial ES loading. Remarkably, these properties equal or greatly exceed those previously reported in the literature for PP/ES composite materials using untreated ES and made by complex, multi-step processing including melt extrusion. The PP/ES composites made by SSSP exhibited exceptionally good thermal stabilities in nitrogen (maximum of 20 °C increase in Tdeg corresponding to 5% mass loss relative to that of neat PP). Additionally, ES particles dramatically improved the thermo-oxidative stability of neat PP (increase of 34 °C for 20% mass loss relative to that of neat PP). The excellent thermoxidative stability of the PP/ES composites suggests the adsorption of degradation products to ES in addition to the barrier effect being present at high temperature. Rheological measurements showed a strong solidlike response at low frequencies with no crossover of G0 and G00 for 30 wt% filler loading, indicative of a percolated network within the PP matrix. Nevertheless, the excellent dispersion of ES by SSSP resulted in only modest increases in viscosity of the PP/ES composites relative to that of neat PP at shear rates commonly employed in melt extrusion and injection molding. Such results indicate that PP/ES composite materials produced by single-step SSSP can result in not only excellent properties relative to PP/ES composites made by other methods, but also easy post-SSSP melt processing into final products. Acknowledgments We thank Total Petrochemicals for providing PP and acknowledge support from the Initiative for Sustainability and Energy at
[1] Tjong SC. Mater Sci Eng R 2006;53:73–197. [2] Stankovich S, Dikin DA, Dommett GH, Kohlhaas KM, Zimney EJ, Stach EA, et al. Nature 2006;442:282–6. [3] Paul DR, Robeson LM. Polymer 2008;49:3187–204. [4] Kim H, Abdala AA, Macosko CW. Macromolecules 2010;43:6515–30. [5] Compton OC, Kim S, Pierre C, Torkelson JM, Nguyen ST. Adv Mater 2010;22:4759–63. [6] Fu SY, Lauke B. Compos Sci Technol 1996;56:1179–90. [7] Tjong SC, Xu SA, Li RKY, Mai YW. Compos Sci Technol 2002;62:831–40. [8] Chan CM, Wu J, Li JX, Cheung YK. Polymer 2002;43:2981–92. [9] Giannelis EP. Adv Mater 1996;8:29–35. [10] Thostenson ET, Ren Z, Chou TW. Compos Sci Technol 2001;61:1899–912. [11] Wakabayashi K, Pierre C, Dikin DA, Ruoff RS, Ramanathan T, Brinson LC, et al. Macromolecules 2008;41:1905–8. [12] Wakabayashi K, Brunner PJ, Masuda J, Hewlett SA, Torkelson JM. Polymer 2010;51:5525–31. [13] Castrillo PD, Olmos D, Torkelson JM, González-Benito J. Polym Compos 2009;31:781–91. [14] Gutierrez-Diez S, Rodriguez-Perez MA, De Saja JA, Velasco JI. Polymer 1999;40:5345–53. [15] Jain S, Goossens H, van Duin M, Lemstra P. Polymer 2005;46:8805–18. [16] Leong YW, Abu Bakar MB, Mohd Ishak ZA, Ariffin A. Polym Degrad Stabil 2004;83:411–22. [17] Leong YW, Abu Bakar MB, Mohd Ishak ZA, Ariffin A, Pukanszky B. J Appl Polym Sci 2004;91:3315–26. [18] Rong MZ, Zhang MQ, Pan SL, Friedrich K. J Appl Polym Sci 2004;92:1771–81. [19] Rong MZ, Zhang MQ, Zheng YX, Zeng HM, Walter R, Friedrich K. Polymer 2001;42:167–83. [20] Wang WY, Zeng XF, Wang GQ, Chen JF. J Appl Polym Sci 2007;106:1932–8. [21] Zuiderduin WCJ, Westzaan C, Huetink J, Gaymans RJ. Polymer 2003;44:261–75. [22] Tjong SC, Xu SA. Polym Int 1997;44:95–103. [23] Mitsuishi K, Kodama S, Kawasaki H. Polym Eng Sci 1985;25:1069–73. [24] Huang YP, Chen GM, Yao Z, Li HW, Wu Y. Eur Polym J 2005;41:2753–60. [25] Jipa S, Zaharescu T, Supaphol P. Polym Bull 2009;64:783–90. [26] Yang K, Yang Q, Li G, Sun Y, Feng D. Polym Compos 2006;27:443–50. [27] Zaharescu T, Jipa S, Kappel W, Supaphol P. Macromol Symp 2006;242:319–24. [28] Toro P, Quijada R, Murillo O, Yazdani-Pedram M. Polym Int 2005;54:730–4. [29] Ichazo MN, Albano C, Gonzalez J, Perera R, Candal MV. Compos Struct 2001;54:207–14. [30] Premalal HGB, Ismail H, Baharin A. Polym Test 2002;21:833–9. [31] Joshi SV, Drzal LT, Mohanty AK, Arora S. Compos Part A 2004;35:371–6. [32] Yang HS, Kim HJ, Son J, Park HJ, Lee BJ, Hwang TS. Compos Struct 2004;63:305–12. [33] Fung KL, Xing XS, Li RKY, Tjong SC, Mai YW. Compos Sci Technol 2003;63:1255–8. [34] ‘‘Scientists hatch plan to recycle eggshells into plastics’’. CNN Tech. April 3; 2012. [35] Verma N, Kumar V, Bansal MC. Pol J Environ Stud 2012;21:491–7. [36] Arias JL, Fink DJ, Xiao SQ, Heuer AH, Caplan AI. Int Rev Cytol 1999;145:217–50. [37] Arias J, Fernandez MS. Mater Charact 2003;50:189–95. [38] Tsai WT, Yang JM, Hsu HC, Lin CM, Lin KY, Chiu CH. Microporous Mesoporous Mater 2008;111:379–86. [39] Bautista JLA, Abarca JRQ, Toro PF, Yazdani PM. US Patent 2008; 7459492.
160
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
[40] Toro P, Quijada R, Yazdani-Pedram M, Arias JL. Mater Lett 2007;61:4347–50. [41] Toro P, Quijada R, Arias JL, Pedram-Yazdani M. Macromol Mater Eng 2007;292:1027–34. [42] Ghabeer T, Dweiri R, Al-Khateeb S. J Reinf Plast Comp 2013;32:402–9. [43] Supri AG, Ismail H, Shuhadah S. Polym Plast Technol Eng 2010;49:347–53. [44] Lin Z, Zhang Z, Mai K. J Appl Polym Sci 2012;125:61–6. [45] Kumar R, Dhaliwal JS, Kapur GS, Shashikant. Polym Compos 2014;35:708–14. [46] Sutapun W, Pakdeechote P, Suppakam N, Ruksakulpiwat Y. Polym Plast Technol Eng 2013;52:1025–33. [47] Ashok B, Naresh S, Reddy KO, Madhukar K, Cai J, Zhang L, et al. Int J Polym Anal Charac 2014;19:245–55. [48] Hassan TA, Rangari VK, Jeelani S. ACS Sustain Chem Eng 2014;2:706–17. [49] Hassan TA, Rangari VK, Jeelani S. J Appl Polym Sci 2013;130:1442–52. [50] Bootklad M, Kaewtatip K. Carbohyd Polym 2013;97:315–20. [51] Saeb MR, Ramerzani-Dakhel H, Khondaker HA, Heinrich G, Wagenknecht U. J Appl Polym Sci 2013;127:4241–50. [52] Intharapat P, Kongnoo A, Kateungngan K. J Polym Environ 2013;21:245–58. [53] Kang DJ, Pal K, Park SJ, Bang DS, Kim JK. Mater Des 2010;31:2216–9. [54] Xu Y, Hanna M. Packag Technol Sci 2007;20:165–72. [55] Brunner PJ, Clark JT, Torkelson JM, Wakabayashi K. Polym Eng Sci 2012;52:1555–64. [56] Brunner PJ, Torkelson JM. Plast Eng 2012;68:28–34.
[57] Torkelson JM, Lebovitz AH, Kasimatis K, Khait K. US patent 2007;20080176980-A1. [58] Masuda J, Torkelson JM. Macromolecules 2008;41:5974–7. [59] Jiang X, Drzal LT. J Appl Polym Sci 2012;124:525–35. [60] Lebovitz AH, Khait K, Torkelson JM. Polymer 2003;44:199–206. [61] Tao Y, Kim J, Torkelson JM. Polymer 2006;47:6773–81. [62] Lebovitz AH, Khait K, Torkelson JM. Macromolecules 2002;35:8672–5. [63] Lebovitz AH, Khait K, Torkelson JM. Macromolecules 2002;35:9716–22. [64] Furgiuele N, Lebovitz AH, Khait K, Torkelson JM. Macromolecules 2000;33:225–8. [65] Walker AM, Tao Y, Torkelson JM. Polymer 2007;48:1066–74. [66] Iyer KA, Torkelson JM. Polym Compos 2013;34:1211–21. [67] Diop MF, Torkelson JM. Polymer 2013;54:4143–54. [68] Diop MF, Torkelson JM. Macromolecules 2013;46:7834–44. [69] Whittington AM, Brouse SM, Malusis MA, Wakabayashi K. Adv Polym Technol 2013;32:21334. [70] Bu HS, Cheng SZD, Wunderlich B. Makromol Chem, Rapid Commun 1988;9:75–7. [71] Cox WP, Merz EH. J Polym Sci 1958;23:619–22. [72] Krishnamoorti R, Giannelis EP. Macromolecules 1997;30:4097–102. [73] Galgali G, Ramesh C, Lele A. Macromolecules 2001;34:852–8. [74] Lele A, Mackley M, Galgali G, Ramesh C. J Rheology 2002;46:109–1110.
Contents lists available at ScienceDirect
Composites Science and Technology journal homepage: www.elsevier.com/locate/compscitech
Green composites of polypropylene and eggshell: Effective biofiller size reduction and dispersion by single-step processing with solid-state shear pulverization Krishnan A. Iyer a, John M. Torkelson a,b,⇑ a b
Department of Chemical and Biological Engineering, Northwestern University, Evanston, IL 60208, United States Department of Materials Science and Engineering, Northwestern University, Evanston, IL 60208, United States
a r t i c l e
i n f o
Article history: Received 15 May 2014 Received in revised form 6 July 2014 Accepted 31 July 2014 Available online 8 August 2014 Keywords: A. Polymer–matrix composites (PMCs) A. Recycling A. Particle-reinforced composites B. Mechanical properties B. Thermal properties
a b s t r a c t Eggshell (ES), a waste byproduct from food processing and hatcheries, contains 95% calcium carbonate (CC), making it a potentially attractive, less expensive substitute for commercial CC. Past work used complex grinding–sieving and/or chemical modification steps to aid in dispersing ES in polymers such as polypropylene (PP). Both steps add to the cost and reduce the green aspect of the composite. Here, green composite materials of PP with 5–40 wt% unmodified ES shards of several centimeters in size are directly processed using continuous, single-step solid-state shear pulverization (SSSP). Electron microscopy and particle size analysis show very good dispersion with some ES particles near the nanoscale in the composite. Well-dispersed ES particles dramatically increase PP crystallization rates with a 5–7% increase in PP crystallinity. The very good dispersion leads to a major increase in Young’s modulus (87% increase relative to neat PP for 40 wt% ES) and a modest increase in hardness; composites exhibit reductions in yield strength, elongation at break, and impact properties. Mechanical and crystallization properties are equal to or better than the best literature data for PP/ES composites without chemical modification made by multi-step approaches involving melt processing. In addition, the composites exhibit high thermal degradation temperatures compared to neat PP, indicating the potential for ES to improve processing stability. Composites with 20–40 wt% ES exhibit solid-like rheological response with no crossover of shear storage and loss moduli. Nevertheless, PP/ES composites retain viscosities close to that of neat PP at shear rates experienced in melt processing. Overall, property enhancements resulting from superior dispersion of ES in PP achieved by SSSP reveal ES to be a promising green filler for thermoplastics. Ó 2014 Published by Elsevier Ltd.
1. Introduction Polymer composites with inorganic fillers such as glass fibers, silica nanoparticles, graphene, carbon nanotubes and nanofibers have been studied extensively over the past two decades [1–12]. Inorganic mineral fillers like calcium carbonate (CC), silica, and talc have gained interest as low cost fillers for thermoplastics [13–21]. These fillers offer significant enhancement in stiffness, crystallization rate and thermal stability [8,22–27]. There has been growing demand for green and renewable substitutes for inorganic fillers [28–33]. Such fillers offer major advantages such as low density as well as reduced cost and mechanical wear during processing. ⇑ Corresponding author at: Department of Chemical and Biological Engineering, Northwestern University, Evanston, IL 60208, United States. Tel.: +1 847 491 7449; fax: +1 847 491 3728. E-mail address: [email protected] (J.M. Torkelson). http://dx.doi.org/10.1016/j.compscitech.2014.07.029 0266-3538/Ó 2014 Published by Elsevier Ltd.
Widely investigated green fillers include natural fibers, rice husk, and wood flour [29–33]. As reported by CNN in 2012, there is an interest among food processing companies to find uses for chicken eggshells (ESs) [34]. Approximately 250,000 tons of ES are produced world-wide by food processing industries annually [35]. Most ES waste is disposed in landfills or turned into low value protein supplements for animal feed. Eggshells typically contain about 95% CC with the remainder being organic material [36–38]. This makes waste ES an excellent source of bio-mineral CC; moreover, the abundant supply makes ES an attractive candidate for replacing CC in thermoplastic composites. Despite its potential as a green alternative to CC, fewer than twenty studies have investigated the possibility of exploiting ES as a composite filler [39–54], about one-half of which focused on polyolefins as the matrix [39–46]. A patent for the preparation of PP/ES composites by melt extrusion was registered by Universidad
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
de Chile in 2006 [40]. That patent and subsequent papers by Toro et al. [41,42] describe a detailed grinding procedure using a flatmetal-blade crusher to produce particles which are sieved using an ASTM 100 mesh. The resulting ES powder was dried at 100 °C for 8 h and then ground further in a concentric-metal-ring mill to obtain particles that pass through an ASTM 400 mesh. The resulting ES powder contained particles of variable size, 50 vol% of which was less than 8.4 lm and 90 vol% of which was less than 27.5 lm. This material was melt mixed with PP to obtain composites with 40 wt% ES. The end product resulting after these multiple steps exhibited an 85% increase in Young’s modulus, a 12% decrease in yield strength, and a 47% decrease in impact strength. Ghabeer et al. [42] investigated the effect of chemical modifications on ground ES powder with an average particle size of 90 lm. The ES powder was treated with stearic acid in an attempt to improve its adhesion with PP matrix. Composites containing 40 wt% untreated ES powder exhibited no change within error in Young’s modulus or PP crystallinity. Treatment of ES with stearic acid increased the Young’s modulus of the composite (relative to neat PP) by 200% due in part to increased crystallinity of the sample. Kumar et al. [45] used isophthalic acid to chemically treat ES and observed 3–18% increases in tensile modulus in PP composites using the modified ES relative to those using unmodified ES or CC. In order to increase the composite impact strength, Lin et al. [44] examined the use of pimelic acid to convert CC in ES into a b-nucleating agent for PP crystallization and observed a 228% increase in impact strength. Supri et al. [43] showed a 130% enhancement in Young’s modulus for polyethylene (PE)/ES composites using PE grafted maleic anhydride as compatibilizer while Sutapun et al. [46] observed improvements in composite impact strength when using PE grafted maleic anhydride relative to unmodified PE. In spite of these property enhancements, ES has failed to become a major commercial filler due in part to the complex steps required to achieve good dispersion of ES. Batch grinding–sieving steps add to cost and chemical modifications used in attempts to improve ES dispersion impair the ‘‘green’’ aspect of ES composites (Fig. 1a).
153
Here, we investigate the potential of using a single-step, continuous, and industrially scalable process called solid-state shear pulverization (SSSP) [11,12,55–69] to produce well-dispersed, green composite materials of PP/ES with significant property enhancements (Fig. 1b). Although the current study employs a lab-scale/ pilot-plant apparatus, polyolefins have been processed at throughputs exceeding 150 kg/h using a commercial-scale SSSP apparatus at Northwestern University [66]. The SSSP apparatus is a modified twin-screw extruder in which materials are cooled rather than heated, so that the polymer is processed below its glass transition temperature if amorphous or below its melting temperature if semicrystalline. The use of near-ambient temperature aids in exposing polymers to higher forces and stresses in the solid state than normally encountered in conventional twin-screw melt extrusion. The absorption of mechanical energy during SSSP results in material fracture followed by random fusion of the materials, which is repeated many times during the average residence time of the material in the pulverizer. SSSP eliminates the common limitations of thermodynamics, viscosity, and degradation encountered in melt processing of polymers. Previously, SSSP has been used to produce well-dispersed polymer nanocomposites with fillers such as clay, graphite and carbon nanotubes [11,12,57–59]. In addition, SSSP has been used to produce compatibilized blends systems in which the dispersed-phase size approaches 100 nm [60,61]. The formation of trace levels of block copolymers at the polymer blend interface resulting from recombination of polymer radicals has been cited as the reason for immiscible blend compatibilization observed in SSSP [62–64]. Green composites of PP with starch and rice husk ash have been produced with SSSP [65,66]. Recently, SSSP was used to functionalize PP with maleic anhydride and ester moieties without significant reduction in PP molecular weight, which accompanies industrial melt-state processes employed to achieve such grafting [67,68]. This work explores the potential of using the single-step SSSP process in the novel production of green PP/ES composite materials. As shown in Fig. 1b, the current study utilizes ES shards that are several centimeters in size without any chemical modification.
Fig. 1. (a) Previous approach to produce PP/ES composites involves multiple grinding/sieving and chemical modification steps to produce ES particles which are further melt processed with PP to yield the final composite; (b) single-step solid-state SSSP with input of unmodified, several-centimeter ES shards produces sub-micron ES particles in the final PP/ES composite material.
154
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
The SSSP process circumvents the grinding–sieving processes and chemical modification previously thought to be necessary for good dispersion of ES (Fig. 1a). Particle size analysis demonstrates very good dispersion and size reduction of ES achieved via SSSP. Crystallization and mechanical properties are measured to characterize property enhancements produced by well-dispersed ES particles in the PP matrix. Improvements in thermal degradation behavior due to the presence of well-dispersed ES in PP are observed in air and nitrogen environments. Rheological measurements are performed to quantify the effect of ES particles on composite viscosity, which is important to the production of final composite products by extrusion or injection molding. Where appropriate, comparisons of the properties of SSSP-prepared PP/ES composites are drawn with literature data for PP/ES composites prepared by melt processing. 2. Experimental section 2.1. Preparation of PP/ES composites Polypropylene pellet with density of 0.905 g/cm3 and MFI of 9 g/10 min (ASTM D1238, 230 °C and 2.2 kg) was from Total Petrochemicals. Eggshell shards of several centimeters size were procured locally and used directly after washing briefly with water to remove dirt. The PP pellets and varying amounts of ES were dry blended and fed to a Berstorff ZE-25 pulverizer using a K-tron S-60 feeder. The pulverizer has a section with a 25-mm-diameter barrel containing spiral conveying and bilobe kneading elements (mixing), resulting in intimate mixing between PP and ES. Following this is a small section where the barrel diameter changes from 25 to 23 mm. (SSSP can be done with all sections having a 25 mm diameter barrel.) The 23 mm section contains trilobe shearing elements, in which the bulk of the pulverization and particle size reduction occurs. Depending on the motor running the apparatus, the pulverizer can be run with all bilobe elements. The SSSP barrels were cooled by a recirculating ethylene glycol/ water mix at 7 °C supplied by a Budzar Industries WC3 chiller. The screw was designed to impart high specific energy input to the material, with the mixing zone containing one reverse, two neutral, and three forward kneading elements and the pulverization zone containing three forward, two neutral, and two reverse shearing elements [55]. A screw speed of 200 RPM and a feed rate of 80 g/h were used to produce a fine powder. 2.2. Characterization of PP/ES composites Field emission-scanning electron microscopy (FE-SEM) samples were prepared by melt extrusion at 200 °C via a MiniMax molder. The morphologies of the cryofractured sections of PP/ES composites were obtained via a Hitachi SU 8030 FE-SEM after sputter coating (Cressington 208HR) with gold. The number-average particle size distribution was quantified based on a minimum of 150 particles using ImageJ. Uniaxial tensile test and hardness test samples were prepared by compression molding the SSSP powder in a PHI (model 0230C-X1) press at 200 °C for 5 min with a 5 ton ram force followed by cooling immediately in a cold press at 10 °C under 5 ton ram force. Tensile coupons were cut using a standard Dewes-Gumbs dogbone die. Samples were equilibrated at room temperature for 48 h and then tested (ASTM D 1708) using an MTS Sintech 20/G tensile tester equipped with a 100 kN load cell at a crosshead speed of 50 mm/min. Seven samples were tested for each composite; the results were averaged to obtain a mean value. Hardness tests were performed on sections with a Vickers diamond indenter, using a 250 mN load for 10 s. Results were
reported as the mean value of ten indentations. Unnotched impact test bars of 62 mm in length, 12.6 mm width and 3.5 mm width were prepared by using an injection molding machine (Morgan press) with barrel and nozzle temperatures of 200 and 210 °C. (The input feed to the injection molding machine were composite pellets made by single-screw melt extrusion of the SSSP composite powder followed by pelletization.) A clamp force of 10 tons and injection pressure of 4.5 105 psi were used to produce the impact test bars. A Tinius-Olsen IT504 pendulum tester was used in an unnotched Izod setup (ASTM D4812). Crystallization of PP in the composites and the neat state was characterized using a Mettler Toledo 822e differential scanning calorimeter (DSC). After heating to 200 °C, a 10 °C/min cooling rate was employed to determine the non-isothermal crystallization onset temperature of PP in the composite. In order to calculate the crystallinity of PP in the composite, the specific enthalpy derived from the area associated with the crystallization portion of the nonisothermal cooling curve was divided by the mass fraction of PP in the composite; this value was subsequently divided by the theoretical heat of fusion for neat PP of 207.1 J/g [70]. Isothermal crystallization half-times were also determined at 140 °C using DSC after cooling samples at a rate of 40 °C/min from 200 °C. Thermal degradation behavior was monitored using a thermogravimetric analyzer (Mettler Toledo 851e) calibrated with an indium/aluminum standard. Composite samples were heated to 700 °C using a 10 °C/min heating ramp in both N2 and O2 environment to determine thermal and thermoxidative degradation behavior of PP in the PP/ES composites. Final ash at 700 °C was used to establish ES content in the composites. Rheological behavior was characterized using a TA Instruments ARES rheometer with a 25 mm parallel-plate fixture at 200 °C. Test samples were prepared using a compression molding protocol similar to that used for producing mechanical test samples. Small amplitude oscillatory shear measurements were performed as a function of frequency from 0.01 to 100 rad/s. 3. Results and discussion 3.1. Morphology FE-SEM images of fractured surfaces of PP/ES composites are shown in Fig. 2. Samples prepared by SSSP show effective dispersion and size reduction of ES in PP even at 30 wt% filler loading. Very good dispersion without filler aggregation is critical in achieving excellent properties, good processability, and optimum appearance of final composite products [21,40]. Small particle size aids in uniform stress transfer between filler and matrix, thereby resulting in a composite with superior mechanical performance. The solidstate nature of the materials allows SSSP to impart large forces and stresses to the material, resulting in intimate mixing of both components with a reduction of several-centimeter ES shards to sub-micron dimensions. Fig. 2 shows ES particles embedded in PP with some particles as small as 100–200 nm in size. Particle size analysis is summarized in Table 1. The PP/ES composites exhibit small particle size, with 78–85% of the particles being submicron and only 5–8% larger than 5 lm. In short, SSSP is a simple, single-step process yielding well-dispersed PP/ES composite materials without need for grinding–sieving or chemical modification formerly thought to be crucial for achieving effective size reduction and dispersion of ES within polymers. 3.2. Crystallization behavior The surfaces of well-dispersed fillers and nanofillers serve as strong heterogeneous nucleation sites for PP crystallization. The
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
155
Fig. 2. FE-SEM images of ES particles dispersed in the PP matrix by SSSP processing: (a) 90/10 wt% PP/ES, (b) 80/20 wt% PP/ES, (c) 70/30 wt% PP/ES, (d) ES particle embedded in the PP matrix.
Table 1 Particle size distribution of ES filler in PP/ES composites produced by SSSP. Sample
5 lm (%)
90/10 wt% PP/ES 80/20 wt% PP/ES 70/30 wt% PP/ES
85 80 78
10 15 14
5 5 8
ability of filler to act as nucleating agent depends on particle size, surface area and dispersion [15,18,22,24]. Fig. 3 shows isothermal crystallization curves at 140 °C which demonstrate that the incorporation of well-dispersed ES filler reduces crystallization halftime dramatically from 28 min for neat PP (with no exposure to SSSP) to 3 and 2 min for composites made by SSSP with 30 and 40 wt% ES, respectively. Table 2 summarizes non-isothermal and isothermal crystallization half-time behavior of PP in PP/ES composites. The strong nucleating ability of ES can be seen from the significant increase in the nonisothermal crystallization onset temperature (Tc,onset) from DSC cooling curves. Relative to neat PP, Tc,onset increases 2–4 °C for 5–10 wt% ES and 11 °C for 40 wt% ES. Further evidence of synergistic property enhancements is demonstrated by the isothermal crystallization behavior. Size reduction to sub-micron dimensions and relatively homogenous dispersion of ES fillers in the composite contribute to the ability of ES to act as a strong nucleating agent. In composites prepared by melt processing, CC and ES particles exhibit good nucleating effects on PP crystallization: a 10–11 °C increase in Tc,onset for filler loadings of 10–40 wt% ES was reported by Ghabeer et al. [42]. However, the presence of particle agglomerates in melt-mixed samples impedes crystal growth and suppresses the increment in crystallization rate. As a result, due to limited dispersion of untreated ES from melt processing, the
Fig. 3. Isothermal crystallization (140 °C) curves of neat PP (with no processing by SSSP) and PP/ES composites produced by SSSP.
reduction in crystallization half-time reported by Ghabeer et al. [42] was modest compared to those of SSSP-produced composites. Relative to neat PP, composites prepared by SSSP exhibit an increase in PP crystallinity, from 49% for neat PP to 56% for the 70/30 wt% PP/ES composite using unmodified ES. Such an increase is likely due to the large amount of ES surface present in the welldispersed composites, which acts as a very effective nucleating site
156
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
Table 2 Crystallization behavior of PP/ES composites produced by SSSP.
a
Material
Tc,onset (°C)
PP crystallinity (%)
Isothermal PP crystallization half-time (min) at 140 °C
Neat PPa 95/5 wt% PP/ES 90/10 wt% PP/ES 80/20 wt% PP/ES 70/30 wt% PP/ES 60/40 wt% PP/ES
122 124 126 133 133 133
49 51 52 55 56 53
28 22 19 4 3 2
With no processing by SSSP.
for PP crystallization. In contrast, Ghabeer et al. [42] reported no increase in PP crystallinity with the incorporation of up to 40 wt% unmodified ES. In the same study, a 6% increase in crystallinity of PP was observed after surface treatment of ES with stearic acid. Supri et al. [43] showed a 6–7% increase in crystallinity after a series of grinding steps and modification of ES using NaOH–isophthalic acid. Notably, incorporation of unmodified ES by singlestep SSSP results in higher PP crystallinity and vastly increased crystallization rates. 3.3. Mechanical properties – tensile, impact and hardness Table 3 summarizes mechanical property enhancements provided by well-dispersed ES in PP. Introducing well-dispersed, rigid ES filler restricts the mobility and deformability of the PP matrix, thereby increasing Young’s modulus of the composite. Literature reports have indicated that good dispersion and reduction in ES particle size are necessary for achieving such increments in Young’s modulus [39,41,42]. Consistent with excellent dispersion of rigid ES particles and improvements in PP crystallinity, PP/ES composites produced by SSSP exhibit a monotonic increase in Young’s modulus with increasing filler loading. Relative to neat PP, 80/20 wt% and 60/40 wt% PP/ES composites exhibit 38% and 87% increases in Young’s modulus, respectively. In comparison, CC, a major component in ES and a common filler for PP, has been reported to yield 60–70% increases in Young’s modulus at 43 wt% loading in PP [17]. Thus, 60/40 wt% PP/ES composites produced by SSSP exhibit mechanical performance superior to that of PP/ CC composites with similar filler loadings produced by twin-screw extrusion. Fig. 4 compares percent increments in Young’s modulus for PP/ ES composites with untreated ES particles observed in this study with those reported in literature. Toro et al. [41] observed no increase in Young’s modulus within error for all filler loadings when ES particles were relatively large (50 vol% less than 90 lm in size). Further size reduction of ES fillers, with 50 vol% less than 8.4 lm in size, resulted in 40% and 85% increases in Young’s modulus for ES loadings of 20 and 40 wt%, respectively. Ghabeer et al. [42] observed no change within (very large) error bars in Young’s modulus over neat PP for up to 40 wt% ES filler loading with particle size ranging from 14 to 125 lm. In contrast, the current study demonstrates that single-step SSSP yields PP/ES composites with
Fig. 4. Comparison of percentage change in Young’s modulus of PP/ES composites produced by SSSP relative to neat PP (with no processing by SSSP). Data by Toro et al. [41] for 90 and 8.4 lm median ES particle sizes (for composites produced by grinding and sieving followed by melt processing). (Error bars are one standard deviation.) (Note: Ghabeer et al. [42] also reported Young’s modulus data for PP/ES composites. However, the error bars associated with their data were sufficiently large to indicate that there was no change in Young’s modulus within error.)
38% and 87% increases in Young’s modulus relative to neat PP for ES loadings of 20 and 40 wt%, respectively. The results obtained with SSSP-produced samples are equal or superior to the best reported values for PP composites with unmodified ES made by grinding–sieving and melt processing. As produced by single-step SSSP, PP/ES composites exhibit yield strength values that are nearly invariant with ES loadings up to 20 wt%. Relative to neat PP, PP/ES composites exhibit only a 12– 15% decrease in yield strength for 30–40 wt% ES loading. With melt processing, Ghabeer et al. [42] observed a 37% decrease in yield strength for 40 wt% loading of untreated ES particles. In our case, the effective particle size reduction and dispersion facilitate stress transfer between the PP matrix and ES. Additionally, the uniform stress transfer due to good filler dispersion prevents stress concentration around ES embedded in the PP matrix so that SSSP-processed PP/ES composites show good retention of yield strength compared to neat PP. Due to the increasing stiffness with increasing filler loading, the material exhibits a significant reduction in elongation at break. Relative to neat PP with a 700% elongation at break, the 95/5 wt% and 80/20 wt% PP/ES composites made by SSSP exhibit 60% and 20% elongation at break, respectively. Superior dispersion by SSSP enables efficient stress transfer between PP matrix and filler, resulting in retention of ductile behavior at these compositions. Reduced impact properties for composites with PP are often observed with the incorporation of micro-size mineral fillers such
Table 3 Mechanical properties of PP/ES composites produced by SSSP.
a
Material
Young’s modulus (MPa)
Yield strength (MPa)
Elongation at break (%)
Unnotiched Izod impact strength (J/m)
Neat PPa 95/5 wt% PP/ES 90/10 wt% PP/ES 80/20 wt% PP/ES 70/30 wt% PP/ES 60/40 wt% PP/ES
1030 ± 20 1230 ± 40 1320 ± 20 1420 ± 20 1600 ± 30 1900 ± 20
33.0 ± 0.2 33.0 ± 1.0 32.0 ± 0.3 31.0 ± 0.4 29.0 ± 0.5 28.0 ± 0.3
700 ± 40 60 ± 10 30 ± 5 20 ± 5 10 ± 2 7±1
34 ± 2 29 ± 2 29 ± 2 20 ± 1 18 ± 1 12 ± 1
With no processing by SSSP.
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
157
Fig. 5 shows the increment in Vickers hardness seen for PP/ES composites produced by SSSP. Calcium carbonate is soft compared to some other mineral fillers such as silica. Thus, PP/ES composites produced by SSSP show a moderate increase in Vicker’s hardness from 100 MPa for neat PP to 130 MPa for 20–40 wt% ES loadings. 3.4. Thermal degradation – nitrogen and air
Fig. 5. Vicker’s hardness values reported in units of MPa for neat PP (with no processing by SSSP) and PP/ES composites produced by SSSP. (The numbers inside the bars represent the wt% of ES in the PP/ES composite. One standard deviation errors are ±2 MPa for each sample.)
as CC and talc [17,21]. Despite excellent particle size reduction and dispersion, the SSSP-processed composites exhibit a decrease in impact strength with increasing ES content. The 10 wt% PP/ES composite shows unnotched Izod impact strength of 29 J/m compared to 34 J/m for PP. The unnotched Izod impact strength decreases further to 20 and 12 J/m for composites with 20 and 40 wt% ES loadings, respectively. Within error bars, literature studies show equivalent or greater reductions in impact properties relative to melt-mixed composites [40,42]. Related behavior has been observed with PP–graphite nanocomposites produced by SSSP [12]. It is important to note that at 20–40 wt% ES content, the biofiller accounts for 7.5–15 vol% of the composite, increasing the probability for microcrack formation accompanying impact. These microcracks can propagate through the matrix and cause fracture at the polymer–filler interface.
Thermal degradation behavior of PP in PP/ES composites under inert nitrogen atmosphere provides information regarding thermal stability and the true filler content in the composites. Fig. 6a and Table 4 shows the thermal degradation behavior of PP in PP/ES composites under nitrogen atmosphere. Here, the temperature corresponding to 5% mass loss is defined as the onset temperature for degradation (Tdeg) for neat PP and PP in PP/ES composites under nitrogen atmosphere. Neat PP exhibits almost no mass loss up to 370 °C and 5% mass loss at 407 °C. In comparison, PP in PP/ES composite with 5 wt% ES exhibits a 16 °C increase in its Tdeg relative to neat PP. A maximum increase in Tdeg of 20 °C is observed in composites containing 30 wt% ES. The presence of well-dispersed ES in PP matrix is expected to provide a barrier to the diffusion of degradation products, suppressing PP mass loss in the composite. Table 4 reports leftover ash content after thermal degradation under nitrogen. Due to its high CC content, ES displays only 8% mass loss up to 700 °C, consistent with a 92% CC content. In contrast, PP leaves behind no measurable residue at 700 °C. Residue in composites at 700 °C correlates well with the expected ES content after accounting for the minor ES mass loss. In addition, the reproducibility of the residual ash content is consistent with effective ES dispersion. As shown in Fig. 6b, thermo-oxidative stability of PP in PP/ES composite in air was characterized to determine composite stability during processing. Polypropylene in PP/ES composites shows a minor 2–3 °C decrease in temperature corresponding to 5% mass loss. Interestingly, the thermograms demonstrate enhanced thermal stabilities for temperatures corresponding to higher mass loss. In order to better quantify the enhancement in thermal stabilities, Table 5 summarizes the temperatures corresponding to 10% (T10%) and 20% (T20%) mass loss in addition to Tmax (maximum rate of mass loss obtained from the peak value determined from the
Fig. 6. Thermogravimetric analysis curves showing degradation behavior of eggshell, neat PP (with no processing by SSSP) and PP in PP/ES composites produced by SSSP: (a) in nitrogen and (b) in air. Heating rate is 10 °C/min. (Note: At 700 °C, all PP is degraded and volatilized, i.e., normalized PP wt% = 0.) (In the figures, PP40ES denotes 60/40 wt% PP/ES composite.)
158
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
Table 4 Thermal degradation behavior under nitrogen atmosphere of neat PP (with no processing by SSSP) and PP in PP/ES composites produced by SSSP.
a b c
Material
Tdega (°C)
Final ashb (%)
ES content in compositesc (%)
Neat PP 95/5 wt% PP/ES 90/10 wt% PP/ES 80/20 wt% PP/ES 70/30 wt% PP/ES 60/40 wt% PP/ES
407 423 419 425 427 421
0 5 8 16 26 35
0 5 9 17 28 38
sub-micron dimensions increases its effective surface area and further aids such adsorption [42].
3.5. Rheology
Temperature corresponding to 5% mass loss. Residue measured at 700 °C. Corrected for 8% organic content in EG not accounted for in final ash.
Table 5 Thermal degradation behavior under air atmosphere of neat PP (with no processing by SSSP) and PP in PP/ES composites produced by SSSP. Material
T10a (°C)
T20b (°C)
Tmaxc (°C)
Neat PP 95/5 wt% PP/ES 90/10 wt% PP/ES 80/20 wt% PP/ES 70/30 wt% PP/ES 60/40 wt% PP/ES
289 292 292 307 304 294
304 310 313 338 336 324
289 370 375 406 403 389
a
Temperature corresponding to 10% PP mass loss. Temperature corresponding to 20% PP mass loss. Temperature corresponding to maximum rate of PP mass loss from derivative thermogravimetric curve. b
c
derivative thermogram) for PP in PP/ES composites. Neat PP exhibits T10% and T20% values of 289 and 304 °C, respectively. The presence of ES enhances thermal stability of PP under oxidative conditions at all filler loadings examined. Composites prepared by SSSP with 20 and 30 wt% ES show 18 and 15 °C increases in T10% relative to that of neat PP. The composites show even greater enhancements in T20% and Tmax; e.g., 80/20 wt% PP/ES shows a 34 °C increase in T20% and a 117 °C increase in Tmax. The substantial enhancement in thermal stability suggests that well-dispersed, sub-micron sized ES particles provide a mass transport barrier to oxygen diffusion and molecular mobility. In addition, the ES filler offers sites for adsorption of volatile degradation products, thereby arresting the autocatalytic degradation process. In particular, particle size reduction of ES to
Fig. 7 shows the magnitude of complex viscosity (|g*|) and shear storage modulus (G0 ) as a function of frequency measured at 200 °C for neat PP and PP/ES composites. Based on the Cox-Merz rule [71], which equates the frequency in simple oscillatory shear experiments to shear rate in simple shear flow experiments, Fig. 7a is equivalent to a plot of shear viscosity as a function of shear rate. At all shear rates greater than 1 s1, composites with ES loadings up to 10 wt% have viscosities and G0 values that are similar to those of neat PP. At 20–40 wt% ES loadings, composites exhibit higher viscosities and G0 values, especially at low frequency. Notably, at low frequencies, G0 shows only a weak frequency dependence for 20–40 wt% filler loadings, consistent with a change in rheological response from liquid-like to more solid-like [72]. Such response is characteristic of a percolated network within the PP matrix [12,73,74]. Fig. 8 shows storage modulus G0 and loss modulus (G00 ) as a function of frequency in the linear viscoelastic regime. The 90/ 10 wt% PP/ES composite exhibits classical viscoelastic behavior, with a terminal regime at low frequencies (G00 x, G0 x2; x is frequency) where G0 < G00 , and a crossover at x = 42 s1 above which G0 > G00 . Incorporation of 20 wt% ES shifts the crossover to x = 16 s1. The 70/30 wt% PP/ES composite shows a solid-like response with no crossover for all frequencies probed in this study. These results are consistent with literature results, demonstrating that filler incorporation increases the solid-like response of polymer composites and shifts the crossover point to lower x [74]. Excellent dispersion of ES in the composites produced by SSSP results in shear flow behavior similar to that of neat PP for filler loadings up to 10 wt%. Even for higher ES loadings of 20 and 30 wt%, the viscosity values are only four times higher than that of neat PP at a shear rate of 100 s1. (Shear rates of 100 s1 and higher are typically employed in conventional melt processing techniques such as twin-screw extrusion and injection molding.) Thus, these viscosity data indicate that PP/ES composite materials made by SSSP can be easily processed into final products by highshear-rate processes twin-screw extrusion and injection molding. In fact, as the samples used here for impact strength tests were made by single-screw extrusion and pelletization of the SSSP
Fig. 7. (a) Magnitude of complex viscosity (|g*|) and (b) shear storage modulus (G0 ) at 200 °C as a function of angular frequency for neat PP with no processing by SSSP (plus) and 95/5 wt% PP/ES (diamond), 90/10 wt% PP/ES (triangle), 80/20 wt% PP/ES (cross), 70/30 wt% PP/ES (circle), and 60/40 wt% PP/ES (asterisk) composites made by SSSP.
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
159
Fig. 8. Magnitude of shear storage modulus (G0 ; diamond) and loss modulus (G00 ; square) at 200 °C as a function of angular frequency for (a) 90/10 wt% PP/ES, (b) 80/20 wt% PP/ES and (c) 70/30 wt% PP/ES composites made by SSSP.
powder output followed by injection molding of the pellets, the current study proves that such processing is easily achieved using PP/ES composites produced by SSSP.
Northwestern University (ISEN) and Northwestern University. This study made use of Central Facilities supported by the MRSEC program of the National Science Foundation at the Northwestern University Materials Research Science and Engineering Center.
4. Conclusion References Single-step, continuous SSSP was employed to produce green PP/ES composite materials with 5–40 wt%. filler. The SSSP process eliminates the need for complex multi-step processing that was previously thought to be crucial in achieving good dispersion of ES within a polymer matrix. Good dispersion and particle size reduction were demonstrated by electron microscopy with the vast majority of ES particles being sub-micron in size. Synergistic property enhancements associated with dispersion and particle size reduction were observed via dramatic improvement in crystallization rate together with a 5–7% increase in PP crystallinity. Relative to neat PP, PP/ES composites produced by SSSP show up to an 87% enhancement in Young’s modulus and a 30 MPa increase in hardness. The composites exhibit reasonable retention of yield strength while maintaining ductility at substantial ES loading. Remarkably, these properties equal or greatly exceed those previously reported in the literature for PP/ES composite materials using untreated ES and made by complex, multi-step processing including melt extrusion. The PP/ES composites made by SSSP exhibited exceptionally good thermal stabilities in nitrogen (maximum of 20 °C increase in Tdeg corresponding to 5% mass loss relative to that of neat PP). Additionally, ES particles dramatically improved the thermo-oxidative stability of neat PP (increase of 34 °C for 20% mass loss relative to that of neat PP). The excellent thermoxidative stability of the PP/ES composites suggests the adsorption of degradation products to ES in addition to the barrier effect being present at high temperature. Rheological measurements showed a strong solidlike response at low frequencies with no crossover of G0 and G00 for 30 wt% filler loading, indicative of a percolated network within the PP matrix. Nevertheless, the excellent dispersion of ES by SSSP resulted in only modest increases in viscosity of the PP/ES composites relative to that of neat PP at shear rates commonly employed in melt extrusion and injection molding. Such results indicate that PP/ES composite materials produced by single-step SSSP can result in not only excellent properties relative to PP/ES composites made by other methods, but also easy post-SSSP melt processing into final products. Acknowledgments We thank Total Petrochemicals for providing PP and acknowledge support from the Initiative for Sustainability and Energy at
[1] Tjong SC. Mater Sci Eng R 2006;53:73–197. [2] Stankovich S, Dikin DA, Dommett GH, Kohlhaas KM, Zimney EJ, Stach EA, et al. Nature 2006;442:282–6. [3] Paul DR, Robeson LM. Polymer 2008;49:3187–204. [4] Kim H, Abdala AA, Macosko CW. Macromolecules 2010;43:6515–30. [5] Compton OC, Kim S, Pierre C, Torkelson JM, Nguyen ST. Adv Mater 2010;22:4759–63. [6] Fu SY, Lauke B. Compos Sci Technol 1996;56:1179–90. [7] Tjong SC, Xu SA, Li RKY, Mai YW. Compos Sci Technol 2002;62:831–40. [8] Chan CM, Wu J, Li JX, Cheung YK. Polymer 2002;43:2981–92. [9] Giannelis EP. Adv Mater 1996;8:29–35. [10] Thostenson ET, Ren Z, Chou TW. Compos Sci Technol 2001;61:1899–912. [11] Wakabayashi K, Pierre C, Dikin DA, Ruoff RS, Ramanathan T, Brinson LC, et al. Macromolecules 2008;41:1905–8. [12] Wakabayashi K, Brunner PJ, Masuda J, Hewlett SA, Torkelson JM. Polymer 2010;51:5525–31. [13] Castrillo PD, Olmos D, Torkelson JM, González-Benito J. Polym Compos 2009;31:781–91. [14] Gutierrez-Diez S, Rodriguez-Perez MA, De Saja JA, Velasco JI. Polymer 1999;40:5345–53. [15] Jain S, Goossens H, van Duin M, Lemstra P. Polymer 2005;46:8805–18. [16] Leong YW, Abu Bakar MB, Mohd Ishak ZA, Ariffin A. Polym Degrad Stabil 2004;83:411–22. [17] Leong YW, Abu Bakar MB, Mohd Ishak ZA, Ariffin A, Pukanszky B. J Appl Polym Sci 2004;91:3315–26. [18] Rong MZ, Zhang MQ, Pan SL, Friedrich K. J Appl Polym Sci 2004;92:1771–81. [19] Rong MZ, Zhang MQ, Zheng YX, Zeng HM, Walter R, Friedrich K. Polymer 2001;42:167–83. [20] Wang WY, Zeng XF, Wang GQ, Chen JF. J Appl Polym Sci 2007;106:1932–8. [21] Zuiderduin WCJ, Westzaan C, Huetink J, Gaymans RJ. Polymer 2003;44:261–75. [22] Tjong SC, Xu SA. Polym Int 1997;44:95–103. [23] Mitsuishi K, Kodama S, Kawasaki H. Polym Eng Sci 1985;25:1069–73. [24] Huang YP, Chen GM, Yao Z, Li HW, Wu Y. Eur Polym J 2005;41:2753–60. [25] Jipa S, Zaharescu T, Supaphol P. Polym Bull 2009;64:783–90. [26] Yang K, Yang Q, Li G, Sun Y, Feng D. Polym Compos 2006;27:443–50. [27] Zaharescu T, Jipa S, Kappel W, Supaphol P. Macromol Symp 2006;242:319–24. [28] Toro P, Quijada R, Murillo O, Yazdani-Pedram M. Polym Int 2005;54:730–4. [29] Ichazo MN, Albano C, Gonzalez J, Perera R, Candal MV. Compos Struct 2001;54:207–14. [30] Premalal HGB, Ismail H, Baharin A. Polym Test 2002;21:833–9. [31] Joshi SV, Drzal LT, Mohanty AK, Arora S. Compos Part A 2004;35:371–6. [32] Yang HS, Kim HJ, Son J, Park HJ, Lee BJ, Hwang TS. Compos Struct 2004;63:305–12. [33] Fung KL, Xing XS, Li RKY, Tjong SC, Mai YW. Compos Sci Technol 2003;63:1255–8. [34] ‘‘Scientists hatch plan to recycle eggshells into plastics’’. CNN Tech. April 3; 2012. [35] Verma N, Kumar V, Bansal MC. Pol J Environ Stud 2012;21:491–7. [36] Arias JL, Fink DJ, Xiao SQ, Heuer AH, Caplan AI. Int Rev Cytol 1999;145:217–50. [37] Arias J, Fernandez MS. Mater Charact 2003;50:189–95. [38] Tsai WT, Yang JM, Hsu HC, Lin CM, Lin KY, Chiu CH. Microporous Mesoporous Mater 2008;111:379–86. [39] Bautista JLA, Abarca JRQ, Toro PF, Yazdani PM. US Patent 2008; 7459492.
160
K.A. Iyer, J.M. Torkelson / Composites Science and Technology 102 (2014) 152–160
[40] Toro P, Quijada R, Yazdani-Pedram M, Arias JL. Mater Lett 2007;61:4347–50. [41] Toro P, Quijada R, Arias JL, Pedram-Yazdani M. Macromol Mater Eng 2007;292:1027–34. [42] Ghabeer T, Dweiri R, Al-Khateeb S. J Reinf Plast Comp 2013;32:402–9. [43] Supri AG, Ismail H, Shuhadah S. Polym Plast Technol Eng 2010;49:347–53. [44] Lin Z, Zhang Z, Mai K. J Appl Polym Sci 2012;125:61–6. [45] Kumar R, Dhaliwal JS, Kapur GS, Shashikant. Polym Compos 2014;35:708–14. [46] Sutapun W, Pakdeechote P, Suppakam N, Ruksakulpiwat Y. Polym Plast Technol Eng 2013;52:1025–33. [47] Ashok B, Naresh S, Reddy KO, Madhukar K, Cai J, Zhang L, et al. Int J Polym Anal Charac 2014;19:245–55. [48] Hassan TA, Rangari VK, Jeelani S. ACS Sustain Chem Eng 2014;2:706–17. [49] Hassan TA, Rangari VK, Jeelani S. J Appl Polym Sci 2013;130:1442–52. [50] Bootklad M, Kaewtatip K. Carbohyd Polym 2013;97:315–20. [51] Saeb MR, Ramerzani-Dakhel H, Khondaker HA, Heinrich G, Wagenknecht U. J Appl Polym Sci 2013;127:4241–50. [52] Intharapat P, Kongnoo A, Kateungngan K. J Polym Environ 2013;21:245–58. [53] Kang DJ, Pal K, Park SJ, Bang DS, Kim JK. Mater Des 2010;31:2216–9. [54] Xu Y, Hanna M. Packag Technol Sci 2007;20:165–72. [55] Brunner PJ, Clark JT, Torkelson JM, Wakabayashi K. Polym Eng Sci 2012;52:1555–64. [56] Brunner PJ, Torkelson JM. Plast Eng 2012;68:28–34.
[57] Torkelson JM, Lebovitz AH, Kasimatis K, Khait K. US patent 2007;20080176980-A1. [58] Masuda J, Torkelson JM. Macromolecules 2008;41:5974–7. [59] Jiang X, Drzal LT. J Appl Polym Sci 2012;124:525–35. [60] Lebovitz AH, Khait K, Torkelson JM. Polymer 2003;44:199–206. [61] Tao Y, Kim J, Torkelson JM. Polymer 2006;47:6773–81. [62] Lebovitz AH, Khait K, Torkelson JM. Macromolecules 2002;35:8672–5. [63] Lebovitz AH, Khait K, Torkelson JM. Macromolecules 2002;35:9716–22. [64] Furgiuele N, Lebovitz AH, Khait K, Torkelson JM. Macromolecules 2000;33:225–8. [65] Walker AM, Tao Y, Torkelson JM. Polymer 2007;48:1066–74. [66] Iyer KA, Torkelson JM. Polym Compos 2013;34:1211–21. [67] Diop MF, Torkelson JM. Polymer 2013;54:4143–54. [68] Diop MF, Torkelson JM. Macromolecules 2013;46:7834–44. [69] Whittington AM, Brouse SM, Malusis MA, Wakabayashi K. Adv Polym Technol 2013;32:21334. [70] Bu HS, Cheng SZD, Wunderlich B. Makromol Chem, Rapid Commun 1988;9:75–7. [71] Cox WP, Merz EH. J Polym Sci 1958;23:619–22. [72] Krishnamoorti R, Giannelis EP. Macromolecules 1997;30:4097–102. [73] Galgali G, Ramesh C, Lele A. Macromolecules 2001;34:852–8. [74] Lele A, Mackley M, Galgali G, Ramesh C. J Rheology 2002;46:109–1110.
