Development of crayfish protein-PCL biocomposite material processed by injection moulding

1 Development of crayfish protein-PCL biocomposite material 1 processed by injection moulding 2 M. Félix, A. Romero*, J.E. Martín-Alfonso, A. Guerrero 3 Departamento de Ingeniería Química, Universidad de Sevilla, Facultad de Química, 41012 4 Sevilla, Spain 5 Abstract 6 A combination of crayfish flour (CF, with 60% protein) and Polycaprolactone (PCL) was 7 successfully used to prepare biocomposites by a process that consists of two stages: 8 mixing with glycerol (GL) as plasticizer and injection moulding of CF/GL/PCL blends. 9 Mixing rheometry and Differential Scanning Calorimetry (DSC) measurements were 10 found to be useful to select suitable injection moulding conditions. A remarkable 11 enhancement in mechanical properties was found for PCL containing systems, even 12 when crystalline structure remains unaltered. PCL yields a dominant contribution to the 13 elastic response and confer a higher ability to absorb energy before rupture, but also 14 the protein/plasticiser ratio must be considered. 15 16 Keywords: A. Polymer-matrix composites (PMCs); B. Rheological properties 17 B. Thermomechanical; E. Injection moulding 18 _______________________ 19 *A. ROMERO 20 Departamento de Ingeniería Química, 21 Universidad de Sevilla, Facultad de Química, 22 41012 Sevilla (Spain) 23 E-mail: alrome[email protected] 24 Phone: +34 955420984; fax: +34 954556447. 25 2 1. Introduction 26 The exceptional properties of plastics, that have even lead to replace other 27 traditional materials such as metal and wood in many applications, have fostered a 28 continuous demand growth, following a trend which has been increasing since 1950s 29 [1]. However, petroleum production costs have also been increasing, subsequently 30 leading to a continuous rise in oil products. For these reasons, the development of eco- 31 friendly polymeric materials that could contribute to reduce petroleum dependence and 32 is becoming more urgently required. In fact, the development of biodegradable 33 materials with suitable properties has been a great research challenge to the scientific 34 community of polymer materials during recent years [2]. Nowadays, society has been 35 interested in the use of renewable biomass to manufacture high-quality, cost- 36 competitive and biodegradable consumer goods as a means to reduce the 37 consumption and dependence on petroleum, as well as, to diminish environmental 38 pollution [3, 4]. 39 For many years, proteins, lipids and polysaccharides have been selected as a 40 source of biopolymer materials [5-8]. Regarding proteins, most studies have used plant 41 proteins such as zein, wheat gluten or soybean in order to manufacture bioplastic 42 materials [9-11]. Moreover, animal proteins (milk proteins, collagen, gelatine, egg 43 albumen…) have also been used to obtain bioplastics [12-15]. 44 A further protein source consists of using those by-products, surpluses and wastes 45 that are produced in large quantities every year by the food industry. This alternative 46 also involves important environmental benefits derived from waste reduction. An 47 excellent example of this source is featured by the freshwater red-swamp crayfish 48 (Procambarus Clarkii), which has undergone a fast growth due to the initial lack of 49 predators, abundant food and favorable environmental conditions [16]. Previous results 50 have demonstrated the ability of proteins from low value-added crayfish byproducts in 51 3 the stabilization of emulsions [17] and the formation of thermal-induced gel products 52 [18]. 53 Protein coming from crayfish (CF) may be used to develop bioplastic materials with 54 significant advantages since they are derived from a sustainable bioresource and can 55 be processed in much the same way as conventional synthetic polymers [19]. 56 Traditionally, protein films are processed by casting method [20], however classical 57 polymer processing techniques (compression moulding or extrusion) are being 58 increasingly used in this field [21, 22]. Among them, injection moulding is a fairly 59 attractive operation that deserves particular attention. Typically in this process, proteins 60 are mixed with a plasticiser in order to reduce intermolecular forces among polymer 61 chains, increasing mobility and reducing the glass transition [7]. Thereafter, polymeric 62 blends are subjected to suitable thermal conditions, being injected at high pressure into 63 the mould cavity [12]. 64 The optimization of processing conditions is essential in order to achieve the 65 desirable properties of the final product. This fact is particularly relevant in protein- 66 based materials that require a proper thermoplastic mixing with a suitable plasticiser, 67 and may even come to show a predominant thermoset character upon injection 68 moulding. 69 There is not much information on the use of blends containing polycaprolactone 70 (PCL) and protein blends in order to produce bio-composite materials. PCL is classified 71 as a polyester from fossil source, which has been widely used as the polymer matrix in 72 the development of new materials. It is highly flexible, biodegradable, biocompatible 73 and easy to process [23, 24]. PCL can be blended with a variety of other polymers to 74 improve their properties [25]. In fact, blending a natural polymer with polyester is an 75 interesting way to reduce costs and to improve the biodegradability of the resulting 76 polymer blends [26]. 77 The overall objective of this work is to develop CF/PCL biocomposite plastic 78 materials plasticised with glycerol (GL) by means of two-stage processing that consists 79 4 of mixing and injection moulding, using different CF/GL/PCL ratios. Rheological and 80 differential scanning calorimetry (DSC) measurements of CF/GL/PCL blends have 81 been carried out in order to select suitable processing parameters (temperature and 82 residence time in the pre-injection mixing chamber as well as the temperature of the 83 mould). Properties of moulded specimens have been eventually assessed by dynamic 84 mechanical analysis, X-Ray diffraction and tensile strength measurements. 85 2. Material and methods 86 2.1. Materials 87 Crayfish flour (CF) was obtained from ALFOCAN S.A. (Isla Mayor, Seville, Spain). 88 The protein content was determined in quadruplicate as % N x 6.25 using a LECO 89 CHNS-932 nitrogen micro analyser (Leco Corporation, St. Joseph, MI, USA) [27] being 90 65 wt.%. Glycerol (GL) was purchased from Panreac Química, S.A. (Spain) and PCL 91 (CapaTM FB100) was supplied by Perstorp (Sweden). 92 2.2. Preparation of samples 93 Different CF/GL/PCL systems have been selected for this study using different 94 CF/GL ratios (1.3, 2.0, 2.3 and 2.6) and different PCL content (0, 10 and 30%). The 95 PCL-free system (denoted by 70/30/0) has been used as the reference. 96 CF/GL/PCL blends were manufactured by a thermomechanical procedure which 97 includes two stages: Initially, these blends were mixed in a two-blade counter-rotating 98 batch mixer Haake Polylab QC (ThermoHaake, Germany) at 25ºC and 50 rpm for 60 99 min., monitoring the torque and temperature during mixing. Secondly, the biocomposite 100 blends obtained after the mixing process were subsequently processed by injection 101 moulding using a MiniJet Piston Injection Moulding System II (ThermoHaake, 102 Germany) to obtain biocomposites specimens. Three stages are considered over this 103 process: the pre-injection stage, the injection itself and the after-injection packing 104 stage. Table 1 shows the processing parameters (temperature, pressure and time) 105 values for the injection moulding process used in this study, which is similar to those 106 5 used in previous papers [28, 29]. The values for the processing parameters at the pre- 107 injection cylinder are selected to ensure a blend viscosity low enough to facilitate its 108 injection into the mould. The residence time selected for the packing stage (right after 109 injection) has been 220 s since no further enhancement has been noticed by 110 increasing this period. In addition, exposition to high temperatures for a long time 111 typically leads to protein degradation (i.e. via Maillard-type reactions) [30]. 112 Two types of moulds were used to prepare CF/GL/PCL specimens: a 60×10×1 mm 113 rectangular shape mould for DMTA experiments and a Dumpbell type specimen 114 defined by ISO 527-2 [31] for Tensile Properties of Plastics. 115 2.3. Characterization of blends 116 2.3.1. Differential Scanning Calorimetry (DSC). DSC experiments were performed with 117 a Q20 (TA Instruments, USA), using 5 to 10 mg samples, in hermetic aluminium pans. 118 A heating rate of 10 ºC/min was selected. The sample was purged with a nitrogen flow 119 of 50 mL/min. 120 2.4. Characterization of biocomposites 121 2.4.1. Dynamic Mechanical Temperature Analysis (DMTA). DMTA tests were carried 122 out with a RSA3 (TA Instruments, New Castle, DE, USA), on rectangular specimens 123 using dual cantilever bending. All the experiments were carried out at constant 124 frequency (1 Hz) and strain (between 0.01 and 0.30%, within the linear viscoelastic 125 region). The selected heating rate was 3 ºC·min−1. All the samples were coated with 126 Dow Corning high vacuum grease to avoid water loss. 127 2.4.2. X-Ray diffraction (XRD). XRD studies of the specimens were carried out using a 128 D8 Discover (BRUKE, Massachusetts, USA) (45 kV, 100 mA) equipped with Cu Kα 129 radiation (λ = 0.1516 nm) in order to visualise different crystalline phases that may 130 indicate systems whose microstructure could be different. 131 2.4.3. Tensile strength measurements. Tensile tests were performed by using the 132 Insight 10 kN Electromechanical Testing System (MTS, Eden Prairie, MN, USA), 133 6 according to by ISO 527-2 [31] for Tensile Properties of Plastics. Tensile strength 134 parameters were evaluated from at least three duplicates for each product using type 135 IV probes and an extensional rate of 1 mm·min−1 at room temperature. 136 2.5. Statistical analysis 137 Statistical analysis was performed using t-test and one-way analysis of variance 138 (ANOVA, p0.05) by means of the statistical package SPSS 18. Standard deviations 139 from some selected parameters were calculated. 140 3. Results and discussion 141 3.1. Preparation of blends by thermoplastic mixing 142 Fig. 1 exhibits torque and temperature profiles as a function of mixing time for 143 different CF/GL/PCL blends. Fig, 1A, shows both profiles for blends maintaining the 144 same CF/GL ratio at ca. 2.3, using different PCL content, where the blend without PCL 145 is used as the reference system. On the other hand, those blends containing 10% PCL 146 at different CF/GL ratios are plotted in Fig. 1B. 147 These results put forward the relevant dependence of torque and temperature on 148 the CF/GL/PCL ratio. Thus, a rapid increase in torque up to a maximum value takes 149 place, followed by an asymptotic decrease towards a plateau value. Temperature 150 profiles generally follow an increase towards a plateau value. The time required to 151 reach the plateau values for both variables (torque and temperature), which is roughly 152 the same, is clearly dependent on the CF/GL/PCL ratio. This coincidence may be seen 153 as a consequence of the development of exothermic crosslinking reactions during the 154 mixing processes that involve both an increase in temperature and consistency 155 (reflected in torque). Both profiles show also an initial induction period, being more 156 evident for those blends displaying the slowest evolution. It is worth mentioning that a 157 torque peak appears when unmelted PCL is present in the blend. Thus, the PCL melt 158 point, which is reached at about 55ºC, is coincident with the maximum torque value. In 159 7 other words, the torque does not start to decrease until the PCL melting point is 160 exceeded. 161 As may be observed in Fig. 1A, an increase in PCL content leads to a faster torque 162 and temperature kinetics and, as a consequence, to an anticipation of both profiles. 163 The effect also gives rise to a general increase in torque and temperature values at any 164 time. Similar behaviour can be observed by increasing the CF/GL ratio (Fig. 1B). 165 3.2. Thermal characterization of blends 166 Heat flow patterns obtained from Differential Scanning Calorimetry (DSC) 167 measurements are shown in Fig. 2. Fig 2A shows the thermogram for CF flour and for 168 the reference system (CF/GL/PCL, 70/30/0), as well as the profiles corresponding to 169 CF/GL/PCL blends at constant CF/GL ratio, as a function of PCL content. On the other 170 and, Fig. 2B displays the DSC results for CF/GL/PCL blends containing 10% PCL as a 171 function of the CF/GL ratio. 172 CF flour displays a typical endotherm of a fairly denatured protein system. This 173 profile exhibits an endothermic first peak at 68ºC, a glass transition (Tg) at ca. 92ºC, as 174 well as a broad endothermic event between the Tg and 130ºC. The first thermal event 175 can be attributed to the physical aging effect, which was previously reported for this CF 176 flour [32]. Physical aging is a general phenomenon that occurs over time in glassy or 177 partial glassy polymers below their Tg and is a manifestation of the non-equilibrium 178 nature of the glassy state [33, 34]. The glass transition at around 90ºC is consistent 179 with previous results reported by Farahnaky et al. [32] for crayfish flour and Hashimoto 180 et al. [35] for fish muscle proteins. On the other hand, the broad endothermic event 181 may be related to the huge variety of protein fractions of different molecular weight that 182 constitute the CF flour a reported in a previous paper [36]. 183 With regard to the endotherms obtained for the reference system the two 184 endothermic events vanish. The total disappearance of the first endothermic peak, 185 most probably as a consequence of mixing, confirms its physical ageing-driven nature. 186 8 However, the glass transition remains roughly at the same temperature for all the 187 blends studied. 188 In addition, it is worth mentioning that all the systems containing PCL display an 189 apparent endothermic peak at ca. 55 ºC, which is attributed to the PCL melting point. 190 This peak becomes more pronounced with increasing PCL content, as may be 191 observed in Fig. 2. 192 The results obtained from DSC measurements also confirm the suitability of the 193 temperature selected for the cylinder and mould, since the former (60ºC) is higher than 194 the endothermic peak corresponding to the PCL melting whereas the later temperature 195 (100ºC) is higher than the glass transition in order to favour mobility and temperature- 196 induced protein crosslinking (in combination with pressure). 197 3.3. Mechanical characterization of biocomposites 198 3.3.1. Dynamic Mechanical Temperature Analysis 199 Fig. 3 shows the values of the elastic modulus, E’ (Fig. 3A), and the loss tangent, 200 tan  (Fig. 3B), as a function of temperature (from -30ºC to 140º) obtained from DMA 201 measurements. The reference system is compared with those specimens with the 202 same protein/plasticiser ratio (CF/GL/PCL: 63/27/10 and 49/21/30). 203 As may be observed in Fig. 3A, all the specimens studied show a similar profile for 204 E’ at low and medium temperature. In this way, an increase in temperature leads to a 205 decrease in elastic modulus (E’) that tends to reach a plateau value. It is remarkable, 206 however, that PCL containing samples exhibit higher elastic modulus provided that 207 temperature remains below the PCL melting point. At the melting point the three 208 samples show rather coincident values that undergo a decrease with increasing 209 temperature. In fact, there is a region where the PCL-free specimen shows slightly 210 higher E’ values. Eventually a temperature is reached above which the decrease in E’ 211 found for the reference specimen becomes faster whereas the PCL-containing 212 composites tend to reach a plateau region. 213 9 In addition, it is interesting to remark that the reference system shows a minimum in 214 E’ close to 100ºC, followed by an apparent enhancement of elastic properties increase, 215 which is characteristic of a thermosetting potential [37-39]. In other words, an increase 216 in the temperature of the packing stage would lead to a reinforcement of the 217 biopolymer matrix. In contrast, the addition of PCL leads to a reduction in this potential 218 than tends to disappear at the highest PCL content, where elastic properties remain 219 rather independent on temperature. These results suggest that the melting of PCL 220 results in a certain reduction in the free volume leading to an increase in physical 221 interactions. In any case, the response in the high temperature region seems to be 222 dominated by the presence of melted PCL. 223 Furthermore, all probes studied display similar loss tangent profiles (Fig. 3B) 224 showing one single peak, which is related to a glass-like transition of the plasticized 225 protein-based material. These unimodal profiles indicate a good compatibility between 226 compounds, obtained for all systems after the injection moulding process. However, 227 the increase in PCL content gives rise to a plasticising effect leading to a decrease in 228 the peak temperature (from ca. 100 to 50ºC). 229 Figure 4 shows the values of some parameters selected in order to compare the 230 behaviour of different biocomposite materials. The first one (shown in Fig. 4A) 231 corresponds to the value of the storage modulus at 1 Hz and 20ºC (E’1,20) and the 232 second is the value of the loss tangent at the peak temperature (tan δmax), whose 233 values (Tpeak) are also included in the plot (Fig. 4B). These results confirm the effect of 234 PCL-induced increase in elasticity that takes place also at a higher CF/GL ratio. 235 Moreover, at constant PCL content, an increase in CF/GL ratio also produces an 236 increase in E’ values. This effect is particularly noticeable at the highest PCL 237 concentration, which suggests occurrence of a synergistic effect between CF and PCL. 238 Similar results were reported by Aithani and Mohanty [40], which found an increase in 239 storage modulus for bioplastics containing PCL and corn gluten meal as the proportion 240 of PCL increase. 241 16 [16] Geiger W, Alcorlo P, Baltanas A, Montes C. Impact of an introduced Crustacean 402 on the trophic webs of Mediterranean wetlands. Biol Invasions. 2005;7(1):49-73. 403 [17] Romero A, Cordobes F, Puppo MC, Guerreroa A, Bengoechea C. Rheology 404 and droplet size distribution of emulsions stabilized by crayfish flour. Food 405 Hydrocolloids. 2008;22(6):1033-43. 406 [18] Romero A, Cordobes F, Guerrero A, Puppo MC. Crayfish protein isolated gels. 407 A study of pH influence. Food Hydrocolloids. 2011;25(6):1490-8. 408 [19] Sharma S, Luzinov I. Water Aided Fabrication of Whey and Albumin Plastics. 409 Journal of Polymers and the Environment. 2012;20(3):681-9. 410 [20] Genadios A. Proteins based films and coting. New York: CRC press; 2002. 411 [21] Jerez A, Partal P, Martinez I, Gallegos C, Guerrero A. Egg white-based 412 bioplastics developed by thermomechanical processing. Journal of Food Engineering. 413 2007;82(4):608-17. 414 [22] Gonzalez-Gutierrez J, Partal P, Garcia-Morales M, Gallegos C. Effect of 415 processing on the viscoelastic, tensile and optical properties of albumen/starch-based 416 bioplastics. Carbohydrate Polymers. 2011;84(1):308-15. 417 [23] Filipczak K, Wozniak M, Ulanski P, Olah L, Przybytniak G, Olkowski RM, et al. 418 Poly (epsilon-caprolactone) biomaterial sterilized by E-beam irradiation. 419 Macromolecular Bioscience. 2006;6(4):261-73. 420 [24] Wu CS. Physical properties and biodegradability of maleated- 421 polycaprolactone/starch composite. Polymer Degradation and Stability. 422 2003;80(1):127-34. 423 [25] Iannace S, Deluca N, Nicolais L, Carfagna C, Huang SJ. Physical 424 characterization of incompatible blends of Polymethylmethacrylate and 425 Polycaprolactone. Journal of Applied Polymer Science. 1990;41(11-12):2691-704. 426 [26] Corradini E, Mattoso LHC, Guedes CGF, Rosa DS. Mechanical, thermal and 427 morphological properties of poly(epsilon-caprolactone)/zein blends. Polymers for 428 Advanced Technologies. 2004;15(6):340-5. 429 17 [27] Etheridge RD, Pesti GM, Foster EH. A comparison of nitrogen values obtained 430 utilizing the Kjeldahl nitrogen and Dumas combustion methodologies (Leco CNS 2000) 431 on samples typical of an animal nutrition analytical laboratory. Animal Feed Science 432 and Technology. 1998;73(1-2):21-8. 433 [28] Felix M, Martin-Alfonso JE, Romero A, Guerrero A. Development of 434 albumen/soy biobased plastic materials processed by injection molding. Journal of 435 Food Engineering. 2014;125:7-16. 436 [29] Martin-Alfonso JE, Felix M, Romero A, Guerrero A. Development of new 437 albumen based biocomposites formulations by injection moulding using chitosan as 438 physicochemical modifier additive. Composites Part B-Engineering. 2014;61:275-81. 439 [30] Fayle SE, Gerrard JA, Chemistry RSo. The Maillard Reaction: Royal Society of 440 Chemistry; 2002. 441 [31] ISO U-E. Plastics - Determination of tensile properties - Part 2: Test conditions 442 for moulding and extrusion plastics Switzerland: International Organization for 443 Standardization; 2012. 444 [32] Farahnaky A, Guerrero A, Hill SE, Mitchell JR. Physical ageing of crayfish flour 445 at low moisture contents. Journal of Thermal Analysis and Calorimetry. 446 2008;93(2):595-8. 447 [33] Strink CE. Physical Aging in Amorphous Polymers and Other Materials: 448 Elsevier; 1978. 449 [34] Anon. Thermal Characterization of Polymeric Materials edited by Edith A. Turi. 450 Polym Test. 1997;16(5):523. 451 [35] Hashimoto T, Suzuki T, Hagiwara T, Takai R. Study on the glass transition for 452 several processed fish muscles and its protein fractions using differential scanning 453 calorimetry. Fisheries Science. 2004;70(6):1144-52. 454 [36] Romero A, Cordobes F, Guerrero A, Cecilia Puppo M. Crayfish protein isolated 455 gels. A study of pH influence. Food Hydrocolloids. 2011;25(6):1490-8. 456 18 [37] Zarate-Ramirez LS, Martinez I, Romero A, Partal P, Guerrero A. Wheat gluten- 457 based materials plasticised with glycerol and water by thermoplastic mixing and 458 thermomoulding. Journal of the Science of Food and Agriculture. 2011;91(4):625-33. 459 [38] Jerez A, Partal P, Martinez I, Gallegos C, Guerrero A. Protein-based 460 bioplastics: effect of thermo-mechanical processing. Rheologica Acta. 2007;46(5):711- 461 20. 462 [39] Romero A, Beaumal V, David-Briand E, Cordobes F, Anton M, Guerrero A. 463 Interfacial and emulsifying behaviour of crayfish protein isolate. Lwt-Food Science and 464 Technology. 2011;44(7):1603-10. 465 [40] Aithani D, Mohanty AK. Value-added new materials from byproduct of corn 466 based ethanol industries: Blends of plasticized corn gluten meal and poly(epsilon- 467 caprolactone). Industrial & Engineering Chemistry Research. 2006;45(18):6147-52. 468 [41] Vertuccio L, Gorrasi G, Sorrentino A, Vittoria V. Nano clay reinforced 469 PCL/starch blends obtained by high energy ball milling. Carbohydrate Polymers. 470 2009;75(1):172-9. 471 [42] Ungar T. Micro structural parameters from X-ray diffraction peak broadening. 472 Scripta Materialia. 2004;51(8):777-81. 473 474 Figure Captions 475 Fig. 1. Evolution of torque and temperature over the mixing process for crayfish 476 flour/glycerol/polycaprolactone (CF/GL/PCL) systems: (A) at constant CF/GL ratio: 477 70/30/0, 63/27/10 and 49/21/30; (B) at constant PCL concentration: 65/25/10, 63/27/10 478 and 60/30/10. 479 Fig. 2. DSC profiles for crayfish flour and CF/GL/PCL blends: (A) at constant CF/GL 480 ratio: 70/30/0, 63/27/10 and 49/21/30; (B) at constant PCL concentration: 65/25/10, 481 63/27/10 and 60/30/10. 482 19 Fig. 3. DMA temperature ramp measurements performed at constant frequency (1 Hz) 483 and heating rate (3ºC·min-1) for CF/GL/PCL injection moulded specimens at constant 484 CF/GL ratio: 70/30/0, 63/27/10, 49/21/30: (A) Storage modulus (E’); and (B) loss 485 tangent (tan δ). 486 Fig. 4. Linear viscoelastic functions from DMA measurements for all CF/GL/PCL 487 studied systems: 70/30/0, 65/25/10, 63/27/10, 60/30/10, 55/15/30, 49/21/30 and 488 40/30/30: (A) Storage modulus (E’); (B) loss tangent (tan δ). 489 Fig. 5. X-Ray Diffraction measurements (XRD) for CF/GL/PCL specimens: (A) 490 containing 10 wt.% PCL: 65/25/10, 63/27/10 and 60/30/10, as well as the reference 491 specimen 70/30/0; (B) containing 30 wt.% PCL: 55/15/30, 49/21/30 and 40/30/30, as 492 well as PCL powder. 493 Fig. 6. Results from uniaxial Tensile Strength measurements for the specimens studied 494 70/30/0, 65/25/10, 63/27/10, 60/30/10, 55/15/30, 49/21/30 and 40/30/30: (A) Tensile 495 stress-strain curves (only for specimens at constant CF/GL ratio: 70/30/0, 63/27/10 and 496 49/21/30); (B) Young’s modulus (E); (C) strain at break (ƐMax); (D) Toughness (T) 497 498 499 20 500 501 21 Figure 1 502 503 504 505 506 22 Figure 2 507 508 509 510 511 512 513 514 515 23 Figure 3 516 517 518 519 520 521 522 523 24 Figure 4 524 525 526 527 528 529 530 531 25 Figure 5 532 533 534 535 536 537 538