. a: . a .«v. T #915: ”fitmfiflg . .w¢wm%£.,VwV. V ”3 ..~ ‘ . . . . . . A . . .. .32» ran“ «W. A. m. Wr .. V . V . . , ,V V . h . . 43..., . x , . . . _. a}... .;.,., ”Ami“? M» J I. . V . . . V . , . . _ V . . . .v ‘I. . L17»v.t‘. . . 4. 4...? 131‘; 33,... . , V . , . V . . . ufinf9+¥VmHUmmW hm” r» V V . tic/n 91 Law .. Mu , _ V .. VMWWufiALHVV‘f . \. 1 “92”}. mt» 1: flash». : . 31v mam? gm? 5"? 4), ‘ {7.3.3 .wl Dial t Dita”! . I“. 5.3V 1‘: . 1 .. .! ill... )1: "V 7%.»: . .. iv..." V :3... 1 2 unfitbnumm \ .flzfifahup... .v‘ «Cu. A. V '1 . '11.... ,awéwga ME aide?“ University This is to certify that the thesis entitled PROCESSING ‘kRAMETER-PPOPERTY RELATIONS OF SOY PROTEIN - COPOLYESTER BLENDS presented by LALITA HARISH WAIKULI has been accepted towards fulfillment of the requirements for S. CHEMICAL ENGINEERING degree in (We; We» \ Major professor 04 8? 177V 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution “3.892321% IIIIIIIIIII IIIIIII IIIIII I I III II‘" A ‘ fl..— - —‘, ‘ .._.-——_.——-" ~ Af~ PLACE IN RETURN BOX to remove this dweckout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 'JAN 2'1 2002 F“- 1!! Wu PROCESSING PARAMETER - PROPERTY RELATIONS OF SOY PROTEIN - COPOLYESTER BLENDS By Lalita Harish Waikul A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE DEPARTMENT OF CHEMICAL ENGINEERING 1998 ABSTRACT PROCESSING PARAMETER - PROPERTY RELATIONS OF SOY PROTEIN — COPOLYESTER BLENDS BY Lalita Harish Waikul A study of the effect of various process parameters on the properties Eastar Bio Copolyester blended with soy protein was conducted. Extrusion was the primary process used for all mixing purposes. Injection molding was used to prepare tensile test samples. The materials synthesized were characterized for mechanical properties , namely - tensile strength and percent elongation. Differential scanning calorimetry was used to obtain information on melting point depression to check for miscibility. Scanning electron microscopy of fractured tensile specimens was used to study morphology. Some of the critical process parameters varied were composition, screw configuration and thermal history. Film formability was also verified using a cast film technique. An improvement in mechanical properties was achieved. by suitably 'varying the jprocess parameters and introducing a plasticizer. ACKNOWLEDEGEMENTS I would like to thank my advisor, Dr. Ramani Narayan for his guidance and support. Many people in the Department of Chemical Engineering, Composite Materials and Structures Center and Michigan Biotechnology Institute have helped me during my work, in so many different ways. It is hard to name everyone individually without missing someone. I would like to thank them all. Special thanks to my colleagues Tina Kiley, Megan Rich and Dale Smith. iii TABLE OF CONTENTS Chapter Page LIST OF TABLES ........................................................... vii LIST OF FIGURES ......................................................... viii ABBREVIATIONS ......................................................... x 1. INTRODUCTION ........................................................... 1 2. LITERATURE REVIEW ................................................... 2 PROBLEM ................................................................... 6 PROBLEM SOLVING APPROACH ..................................... 6 3. RAW MATERIALS ........................................................ 10 3.A SOY PROTEIN .............................................................. 10 3B EASTAR BIO COPOLYESTER .......................................... ll 4. PROCESSING TECHNIQUES ............................................ 12 4.A SYNTHESIS .................................................................. 12 4.B INJECTION MOLDING ................................................... l7 4.C FILM CASTING ............................................................ 17 iv 5.A 5.B 5.C 6.A 6.A.1 6.A.2 6.A.3 6.A.4 6.A.5 6.A.6 6.A.7 6.A.8 6.B 6.B.l 6.B.2 6.B.3 6.B.4 CHARACTERIZATION TECHNIQUES ................................ 19 TEN SILE TESTING ........................................................ l9 ENVIRONMENTAL SCANNING ELECTRON MICROSCOPY.. 19 DIFFERENTIAL SCANNING CALORIIVIETRY ..................... 2O PROCESSING PARAMETER - PROPERTY RELATIONS ........ 23 MECHANICAL PROPERTIES ........................................... 26 EFFECT OF PLASTICIZATION ......................................... 26 EFFECT OF BLENDIN G WITH SOY-PROTEIN SOURCES ....... 27 EFFECT OF COMPOSITION ............................................. 28 EFFECT OF DIFFERENT PLASTICIZERS ............................ 29 EFFECT OF EXTRUDER SCREW CONFIGURATION ............. 30 EFFECT OF THERMAL HISTORY ..................................... 38 EFFECT OF DURATION OF COOLING DURING INJECTION MOLDING .................................................................... 40 EFFECT OF COMPOSITION ON FILMS .............................. 41 MORPHOLOGY ............................................................ 43 EFFECT OF PLASTICIZATION ON SOYCONCENTRATE ....... 46 EFFECT OF PLASTICIZATION ON SOY FLOUR .................. 48 EFFECT OF THERMAL HISTORY ..................................... 50 CONCLUSIONS ............................................................ 56 6.C 6.C.l 6.C.2 6.C.3 6.C.4 6.C.5 6.C.6 DIFFERENTIAL SCANNING CALORIMETRY ...................... DUAL MELTING BEHAVIOR ........................................... EFFECT OF ANNEALIN G TEMPERATURE ON MELTIN G POINTS ....................................................................... EFFECT OF ANNEALIN G TEMPERATURE ON CRYSTALLINITY .......................................................... EFFECT OF COMPOSITION ON THE MELTING POINT ......... EFFECT OF PROCESSING TEMPERATURE ON THE MELTING POINT ........................................................... CONCLUSION ............................................................... SCALE UP .................................................................... CONCLUSIONS AND RECOMMENDATIONS ...................... APPENDIX A - FILM CAST LINE ..................................... APPENDIX B - SOYPROTEIN ......................................... BIBLIOGRAPHY ............................................................ vi 58 59 59 6O 63 66 67 69 8O 85 86 89 LIST OF TABLES Table Page 1 Soy protein sources 10 2 EPE properties 11 3 Base screw configurations 31 4 Combinations of screw configuration used for 31 plasticization and compounding 5 Effect of varying composition on the heat of 61 fusion 5 Hoffman Weeks n parameter obtained by linear 54 regression of peak Tm vs. Tc data 7 Some of the scaling powers obtained from 76 different scaling strategies 8 Lab scale and production scale values of 76 some important parameters 9 Scale up assumption 77 10 Equipment cost break up 78 11 Investment break up 79 12 Estimation of annual operating costs for 80 plant capacity 2.5million lbs 13 Composition of soy flour and soy concentrate 85 14 Amino acid composition of soybean whole meal 86 vii LIST OF FIGURES Figure Page 1 Problem solving approach 10 2 Schematic of processing steps 16 3 Experimental Route 25 4 Effect of plasticization with glycerol on 26 EPE:SC::70:30 blends 5 Effect of using different sources of protein 28 before and after plasticization with glycerol on EPE:SC::70:30 and EPE:SF::70:3O blends 6 Effect of varying weight fraction of PSC on 29 EPE:PSC blends 7 Effect of different plasticizers on 30 EPE:PSC::70:30 blends 8 Low KB screw configuration 32 9 Moderate KB screw configuration 33 10 High HB screw configuration 34 11 Effect of screw configuration on 37 EPE:PSC::70:30 blends 12 Effect of extrusion temperature on 39 EPE:PSC::70:30 blends injection molded at 135C 13 Effect of extrusion temperature on 39 EPE:PSC::70:30 blends injection molded at 160C 14 Effect of duration of cooling on 41 EPE:PSC::50:50 blend 15 Effect of composition on films (D882) 42 16 SEM picture of Soy concentrate particles 43 viii 17 SEM picture of EPE:SC: :70:3O 45 18 SEM picture of EPE:PSC(G) : :70:3O El4OC, IlZOC 57 19 SEM picture of EPE:(UNCMP)PSF(G)::70:3O 49 20 SEM picture of EPE:PSC(G)::70:30 E135C, I135C 51 21 SEM picture of EPE:PSC(G)::70:30 E135C, I160C 52 22 SEM picture of EPE:PSC(G)::70:30 E170C, I135C 54 23 SEM picture of EPE:PSC(G)::70:30 E17OC, 1160C 55 24 Enthalpy versus Temperature of EPE 58 25 EPE melting point onset and peak versus 58 annealing temperature 26 Enthalpy versus annealing temperature for 60 four different compositions 27 Melting temperature vs. composition for 65 different annealing temperatures 28 Effect of processing temperatures on the 67 melting point depression of EPE:PSC::70:30 blends 29 Process flow diagram. 71 30 Some definitions of scaling powers for the 74 mass and momentum equations 31 Expression for relevant processing parameters 75 and dimensionless numbers in continuous mixers 32 Some scaling laws for throughput, mixing and 75 dimensionless numbers 33 Film cast line 84 34 Soy Protein Processing 87 ix ft; 3. D. Pv“ l V“ A AUNERENUHMPICWHB EPE Eastar bio copolyester SF Soy flour SC Soy concentrate SI Soy isolate GLY Glycerol EG Ethylene glycol PSC(EG) Soy concentrate plasticized with ethylene glycol in weight ratio 70:30 PSC(G)or PSC(GLY) Soy concentrate plasticized with glycerol in weight ratio 70:30 PSC Soy concentrate plasticized with glycerol in an extruder when no other details are indicated CMP PSC SC plasticized first by simple mechanical mixing and then in an extruder UNCMP PSC SC plasticized in only by simple mechanical mixing and not in an extruder A:B::X:Y X lbs. of A blended with Y lbs. of B ESEM Environmental Scanning Electron Microscopy DSC Differential Scanning Calorimetry TGA Thermo-Gravimetric Analysis Tm Melting point temperature Tg Glass transistion temperature TC Annealing temperature or crystallization temperature Guage length during tensile testing was reduced to 2" Sample did could not be tested to failure due elongation reaching jaw separation limits. D638 ASTM standard for mechanical testing of polymers D882 ASTM standard for mechanical testing of polymer films EXC IYC Extruded at X degrees C Injection molded at Y degrees C m 1 . INTRODUCTION Many agricultural materials contain proteins, which are essentially macromolecules. Hence the idea of exploiting these natural resources to make plastic articles has been around for a while. The primary reasons for trying to make plastics from renewable resources are (Narayan, 1991): 0 Biodegradability of plastics as a means for environment- friendly disposal. 0 An alternative to depleting non-renewable imported petroleum feedstock. 0 Abundance of agricultural feedstock that could benefit from.development of value added products. Soy’ proteins, which. are essentially polyamides obtained from soybeans, satisfy all the above criteria. Soy proteins and their derivatives can therefore be considered for partial or complete replacements for synthetic polymers. 2. LITERATURE REVIEW Mechanical properties and water absorption studies of plastics made by compression molding of cross—linked (Jane, 1997) and plasticized soy isolates (Jane, 1996) have been reported. Cross-linking resulted in materials having mechanical properties comparable to engineering plastics with tensile modulli of 4.4Gpa, much higher than commonly utilized petrochemical engineering plastics (Epoxy - 3.00GPa, Polycarbonate — 2.1GPa). Plastics made from soy protein alone were very rigid and brittle. Moreover, the data reported were based on use of compression molding, which is not suitable for large volume production or for the manufacture of complex shape articles. A severe problem in the form of an order magnitude drop in the modulus was reported in presence of moisture at room temperature. Plasticization with glycerol resulted in improved flexibility and an increase in percentage elongation at break. Tensile strengths of about 9MPa and a maximum %elongation at break of about 200% were attained using very high levels of plasticizer and using a plasticorder at very wa‘ hu 0 is e y.“ I C. A. 4 IEdC (Utr MI. 4 \TV De low rpms. Water absorption problems persisted. Most data reported were based on soy isolate, which is one of the more expensive sources of soy protein. Using other plasticizers like ethylene glycol as a plasticizer gave slightly higher percent elongation compared to use of glycerol. A lowering of the glass transition temperature from 150C to -50C on plasticization of soy isolate is reported based on dynamic mechanical spectroscopy. This lowering of the glass transition temperature induces thermoplastic behavior which makes it possible to melt soy protein before degradation commences (Jane, 1996). The other important concern that has been identified and somewhat addressed is the drastic reduction of stiffness in moist environments as compared to synthetic polymers. Silane coupled polyphosphate fillers were used to improve water resistance (Otaigbe, 1998) . The work was again based on usage of compression molding and does not report the effect on mechanical properties. There is considerable lack of information concerning the reactions between polyamide-polyester. Utracki's studies (Utracki and Pillion, 1984) on melt rheology of blends of PET and PA indicates a degree of association and/or co- crystallization involving the two polymers. Kamal has also reported a degree of compatibility in the molten state. Patent literature reveals that blending of PET/PA in the melt produces polyester-polyamide block polymers. Utracki and Pillion have used p-tolunesulphonic acid as a catalyst to achieve reaction between polyamides and polyesters requiring 2-4 minutes residence time. Most reactions tried by others (Firestone Tire & Rubber Co., 1969 and Toyo Rayon, 1969) even with catalyst required very long reaction times unsuitable for continuous processing Polyamides are strongly polar polymers while polyesters are neutral polymers. Thus they are not theoretically miscible from a thermodynamic point of view. But this is hardly a deterrent, since adequate compatabilization through use of compatibilizers and appropriate processing conditions is a method commonly used in the plastics industry to effect property improvement in a number of commercially successful blends. An extensive review by Kinsella (Kinsella, 1979) on the functional properties of soy proteins emphasizes the heterogeneity of soy proteins and gives details about the structural changes that occur under different temperatures s... all and chemical conditions. On heating, soy proteins undergo denaturing, which involves permanent unfolding of soy protein structure and eventually degrade above 140C. Presence of thiols, sodium sulphite causes cleavage of intermolecular disulphide bonds increasing solubility. Soy proteins have an isoelectric point of 4.5 The mechanical properties of cast film depend on the amount of orientation that can be applied and retained in the polymer film. An optimum melt temperature and chill roll temperature is necessary so that the melt strength is adequate to sustain the orientation stresses. Annealing may also be necessary to prevent loss of orientation in use (Park W. R. R., 1969 and Park H. C., 1990) THE PROBLEM: The main block to commercialization of soy based plastics is - attempts made heretofore have all solved only a particular aspect of the problem either processability or water absorption or mechanical properties. The problem can now be defined as the need for an integrated engineering solution to address simultaneously al 1 individual aspect s of the probl em, namely - processability, mechanical properties and water solubility in a single solution, with a keen eye on economics. PROBLEM SOLVING APPROACH The present work seeks to address the problem of processability, mechanical properties and moisture sensitivity in a single system. The properties and hence end use of any new plastic is inseparably linked with the choice of processing techniques and processing parameters. Hence property optimization involves optimal selection of materials, processing techniques and processing conditions. Process equipment selection: .Use of conventional processing equipment such as extruders is advocated to reduce cost of design from scratch and custom—built process equipment. The wide availability of this equipment also allows ease of installation of the process at any custom compounder’s facility. Considering all above arguments, it was decided that all-experimental data be obtained using a labscale extruder. Another advantage of this choice is the ease of scaling (Bisio and Kabel, 1985). Material selection: Traditional techniques like extrusion, injection molding, etc., require that materials to be processed be thermoplastic. Soy protein by itself in its natural form does not possess this property. On heating, soy protein undergoes permanent structural changes, called denaturing. Fortunately, thermoplasticity can be induced through additives like plasticizers or by blending with'a thermoplastic having good processability and reasonable mechanical properties. Since plasticization by itself has been tried and did not prove adequate, this thesis explores the possibility of using blending. The idea of blending of two or more polymers to obtain an acceptable combination of properties is a common method in the plastics industry. Additionally most synthetic thermoplastics have very good barrier properties, which can reduce the vulnerability of the blends to moist environments. Using a biodegradable thermoplastic can also aid in retaining the inherent biodegradability and future value of the soy protein product. Eastar bio copolyester 14766, a new film forming biodegradable polymer that was recently introduced by Eastman was found to satisfy all of the above criteria. It is intended to replace widely used non—biodegradable polyethylene films in disposable applications. Coupling of soy protein with this copolyester can lead to enormous a; 9» a. .m PK *5 value addition to abundantly available soy proteins. Copolyester films had 19MPa tensile strength, high percent elongations and low viscosity above the melting point. The other popular biodegradable polymer, PCL was excluded due to its lower melting point (Tm PCL = 60C, Tm EPE = 112C) making products susceptible to deformation during storage and transportation. The objective of this thesis can now be restated. It is to take a step towards the identification of optimal process design and parameters for blending of soy protein and EPE to obtain either an improvement or an acceptable reduction in mechanical properties, using traditional processing technique such as extrusion and injection molding. To achieve this objective, a set of experiments was designed to observe the effect of varying composition and a set of important process parameters on the commercially important mechanical properties. Also, morphology and melting point behavior were studied to aid the fundamental understanding of the underlying phenomena (Figure 1). nomoumnz mcfl>aom Emanoum I H munmfim 23.. 26.5 oan Banana—huh Began—3 3on 320m w:=3==< #5qu E:...— aazaacos all + commence “Eon was—22 Um: + one 235m 3on 380m 1 5333me 0503 2383809. mwmhfiaax 56.5de A :ouacswccouc 380m , Emma w=_e=a_E=sa:§m cosfincwE—oou E oan All x33 3 Swamp—2m e. A V BBMWQEoH fiwcobm 26:3. 5262 gas £8»... 5:8?— 3. RAW'MATERIALS 3.A SOY-PROTEIN Soy protein is available in three different forms — soy flour, soy concentrate and soy isolate. The lower the protein content, lower the cost. In order to maintain cost effectiveness of the end product only soy flour (SF) and soy concentrate(SC) were used. TABLE 1 - SOY PROTEIN SOURCES Protein source Cost %protein $IIB Soy flour 0.15 52 Soy concentrate 0.77 65 Soy isolate 1.5 90 3.A.1 SOYFLOUR Soyafluff 200W from Central Soya was used. It had a particle size (0.3. standard sieve) of 90Min% through #200. Soyflour is classified as per the heat treatment. The most soluble form of soy protein has the highest polydispersity index. 3.A.2 SOY-CONCENTRATE Procon 2000 was obtained from. Central Soya. Soy concentrates are classified depending on pH used to precipitate it. Procon 2000 had a pH of 4.5. It ‘was selected because the acidic pH would be conducive to its 10 interaction with glycerol. It also had a higher degradation temperature than Promax70. It had a particle size (U.S. Standard Sieve) 94Min% through #100 and a maximum moisture content of 10%. 3.8 EASTMAN COPOLYESTER 14766 A new biodegradable polymer EASTAR BIO Copolyester (EPE) introduced by Eastman was used. It is a random copolyester of adipic acid, terephthalic acid and 1,4-butanediol (Buchanan, 1995). TABLE 2 - EPE PROPERTIES Melting point 112C Glass transistion temperature -33C Mn 20 Kg/mole Mw 67Kg/mole M.W.D. 3.3 Theoretical heat of fusion for a 100% 91.96J/gm crystalline copolyester Film tensile strength of 2 mil film(D882) 19Mpa Max %elongation at break of 2 mil film(D882) 600% Density 1.27 g/cm3 Cost of EPE $2.15/Ib U 4 PROCESSING TECWIQUES TGA degradation studies of all basic raw materials were carried out to determine limits on processing temperature conditions. EPE underwent a sharp weight loss at 350C and procon2000 above 180C. Though the weight loss of soy protein powders commences after 180C, the reported degradation of soy-protein molecules starts at 140C suggesting an upper limit on extrusion temperatures. The melting point of the EPE at 112C set the lower limit. High temperatures are desirable to ensure flow-ability of EPE and hence good mixing. 4.A. SYNTHESIS Mixing was done using a co-rotating twin screw extruder, ZSK-30, from Werner & Plfeiderer. All materials in pellet form were metered using a gravimetric acrison feeder and powders using a calibrated K-tron volumetric feeder. Synthesis of blends was split into two steps. i. Plasticization — Mixing soy powder (SC/SF) with glycerol to obtain plasticized soy (PSC/PSF) 12 .m .nu -.L a . a. .3 I a. a l : .. .. .3 .. .. S .. .. v. .1 E . . . . . . . ._ .»~ 9: “a a: «a. VA“ ”he «flu Wu Q Q s 0 "iv E liillll ii. Compounding — Mixing PSC with EPE in a given ratio to obtain an EPE:PSC blend. The process is outlined here assuming the production of EPE and soy concentrate blend with glycerol as a plasticizer. A similar procedure was used when SF or ethylene glycol was used. While producing EPE:SC or EPE:SF blends the plasticization step was skipped. PLASTICIZATION: A fixed ratio of SC:GLY::70:30 was used based on the properties reported by Jane. Plasticization was done in two different ways to test different hypothesis about the mechanism of plasticization. 1. To obtain UNCMP PSC SC and GLY were mixed in a kitchen aid mixer for 30mins to obtain a wet SC powder that was termed UNCMP PSC. Thus UNCMP PSC was obtained by ordinary mixing and was not exposed to high shear and high temperatures. 2. To obtain CMP PSC SC and GLY were mixed in a kitchen aid mixer for 30mins and the 'wet powder obtained 'was extruded to obtain. a :more 13 ttttt M i 7} r: homogenized solid like brittle mass. Thus the CMP PSC was exposed to high shear and heat during plasticization. The idea behind trying these two methods was to compare properties of blends obtained using CMP PSC and UNCMP PSC. If UNCMP PSC resulted in an insignificant loss of properties then it would save the cost of the extrusion during the plasticization stage. It would also mean that high shear and high heat are not required to attain adequate plasticization. Whenever plasticization was carried out in an extruder, a flat temperature profile of 135C-140C was used. Screw speed was fixed. at 125rpmn Torque of about 65% ‘was observed during processing. A vent port was used in zone 5 (Figure 2) to get rid of volatiles generated. Different screw configurations ‘were tried. during this step. The Ibrittle solid mass of PSC obtained was ground in a BTP granulator. COMPOUNDING EPE pellets were mixed. with ground PSC in the desired ratio. This mixture was fed to the extruder and compounded. l4 .‘V amt: A flat temperature profile was used in all runs. Two different temperatures tried were - 1383 and 170C. Different screW' configurations *with. 'varying’ kneading capacities were examined. A fixed screw speed of 125rpm was used. Torque of about 55% was observed. The discharge in the form of strings was quenched in a water bath, dried using air wipes and then pelletized. The pellets were further dried before injection molding or film casting in a 58C room for atleast 24 hours. Hypothesized reactions: Plasticization: Acidic groups in protein + H0- groups in glycerol --~) PSC Compounding: Unreacted HO groups in the PSC + HOOC groups in EPE -) EPE:PSC blend 15 74251... N I mwn.~:_.:r..r»..C On «3G EAOCCIIMZE mnmwcm. mfluw Ha MEQHOHN’QHOZ u E Orr 5870: Evan >5 3:22.. a 3 m0 V won—Em N— Nu Nu Na nu Han .IIIIIvaav aer mango me me we I new: mafios anHcamH mwm mu QZU wmn ZEN mafia. N . OOEOQZUHZG mméwvmo B N... N. N. N. Na ca seams. wmrrmam En 2.: €w§ i :3: an. . we? SHEER. me we I «Syd mons mxnficmmfi wwncfim w I mofimsmnwo on firm UHoommmwbm mnmbm. l6 t 4.B INJECTION’MDLDING A Cincinnati Milacron injection-molding machine was used to prepare all D638 tensile test specimens from dried blend pellets. No mold temperature control was available. Hence, the mold temperature was at ambient. Mostly flat temperature profiles were used, with temperatures ranging from 27SF-360F in the screw barrel zones. The machine was operated in a single cycle mode because of occasional sprue sticking problems. A decrease in melt temperature or a 60-second cooling time was found to be adequate to avoid sprue sticking. A brief study was made of effect of reduction in cooling time on properties in an effort to reduce cycle times. 4.C FIEM CASTING Film casting was done using the Killion single screw extruder and film cast line (Appendix A). The EPE:PSC pellets were melted at about 300F, and a melt shaped by a film die was cast in the gap between the nip rolls. Positive cooling of the nip rolls to temperatures below ambient was necessary to make highly oriented, thin films. Good quenching of the films inhibits melt crystallization l7 m‘ '_ .28 re-a rolls c Increasi and retains amorphous domains, which are required for orientation and transparency. The relative speeds of the main roll vis—a-vis the winder rolls controlled the draw ratio and film thickness. Increasing the speed of the winder for a constant main roller speed increased tension, hence orientation and also reduced film. thickness. The film. thickness was further reduced by simultaneously increasing the roller speed and the winder speed. Typically a roller speed of 400 rpm and winder speed of 50-60rme was useful in obtaining good films. Ability to draw uniform films also depended on the film die gap. Increasing the extruder screw speed increased the discharge, allowing wider films and higher roller and winder speeds. Temperature increase decreases the viscosity, increases pliability and allows thinner films. 18 5 . CHARACTERIZATION TECHNIQUES 5.A TENSILE TESTING Universal Tensile Testing machine was used to conduct both the D638 and D882 tensile tests. Samples were conditioned in an ASTM room at 25% RHH. and 25C for atleast 40 hours. Bars ‘were tested as per the D638 standard. Films ‘were tested according to the D882 standard. For high elongation D638 samples a test rate 50% elongation per minute was used. Non standard conditions had to be used for samples with very high elongations to obtain failure within machine limits. For very high elongation samples the bars were cut to 2" guage length (G. L.) and a test rate of 2"/min was used. For even higher elongation materials, the bars were cut to a G. L. of l" and a test rate of 2"/min was tried. To test a film sample - dimensions of 1" by 4" and test speed of 100%elongation/min was used 5 . B . ENVIRONMENTAL SCANNING ELECTRON MICROSCOPY A Phillips ElectroScan 2020 Environmental Scanning Electron Microscope was used to determine the inter-phase homogenization, extent of adhesion and distribution of the 19 n». AA S.C I Trad is does soy— component in the matrix. X-ray scans were also used to obtain the elemental composition of a particular region to identify the two domains. Sample Preparation: All samples viewed in the ESEM were prepared by cutting with a blade a portion of the fracture zone from tensile specimens taken to failure. Samples were mounted on aluminum disc. Pictures for a given composition showing uniformity were analyzed in juxtaposition with those showing regions lacking in uniformity. Typical SEM Conditions used were — accelerating Voltage of 20 kV, GSED Gas sensitive electron detector, 2000 — 150 x magnification. Around 15-20 pictures, with resolution ranging from 10pm to 450m were taken of each sample to ensure coverage of a wide enough area. 5.C DIFFERENTIAL SCANING CALORIMETRY Traditionally, a shift in the glass transition temperature is taken as an indication of compatibility. But soy protein does not show a DSC-discernible glass transition. Only a single peak around 146C is observed, which has been attributed to denaturing. So this method could not be used to check for compatibilzation. 20 Z. I. C -O" .h. V“ -AR A study of melting point depression with increase in the PSC content was conducted to check for miscibility. A continuous and slight decrease in the melting point indicates miscibility of polymer blends. A number of blends have been analyzed in this way (Nazabal J ., 1995) . The polymer—polymer interaction parameter, x is also related to the melting point depression through the Nishi-Wang equation. The Hoffman — Weeks equation relates the melting point, Tm and the annealing temperature, Tc. T =17T +(1-17)TO mb C mb It can be used to compute the n parameter and the . . . . . o . equilibrium. melting p0int, Tmb. The n parameter is a morphological parameter. If the value of n tends to zero it indicates stable crystals while I] tending to 1 indicates unstable crystals. The EPE:PSC(GLY) blends in.tflua ratios 80:20, 70:30, 60:40 and EPE were analyzed. The experimental procedure used to determine the melting point depression and n parameter is: 21 cf OI 3.A. the 3 CC a for 1” Sample was heated to 150C and held isothermally for 10mins to ensure that all existing crystallites melt. The polymer was then rapidly cooled at 20C/min to the crystallization temperatures. Annealing temperature of 40C, 50C, 60C, 70C were tried for various runs. .2.Sample was left isothermally at the To for 60mins, then rapidly cooled to below —50C and then heated at 10C/min ramp rate and the melting curve was recorded. 23.The melting temperature Tm was plotted against the annealing temperature, Tc. A line was fit to this data. The slope of the line was equal to n 4. Area under the melting peak was obtained using the DSC analysis software to get a measure of the enthalpy of fusion. Another advantage of this study is the possibility of using the crystallization temperature which yields the highest enthalpy of nedting and hence the highest crystallinity as a good temperature for cooling during injection molding or for post production annealing. 22 6 . PROCESSING PARAMETER - PROPERTY DEPENDENCE Figure 3 shows the details of the route taken to obtain the dependence of properties on processing conditions. The EPE:SF and EPE:SC blends prepared in the early stages of experimentation had poor tensile properties. This was attributed to the lack of melt formation of the soy component. Similarity in 'viscosities of materials being blended is desirable (Potschke, 1997). Attaining a melt form of the soy component below its degradation temperature would make it easier to compound it with EP. Hence, later work was done using soy after plasticization to obtain a thermoplastic form. A few inherent constraints and general caveats on the interpretation of the data generated are warranted. a. Soy concentrate in the form used contains numerous other components (Appendix B) and higher sensitivity to heat than most synthetic plastics. It is not easy to characterize the more complex behavior of soy proteins using techniques developed for synthetic plastics. Though there is a profusion of literature about the food like properties of proteins there is inadequacy of information about soy proteins as polymers. 23 During tensile testing of some D638 samples, very high elongation was obtained. The jaw separation limit on the UTS prevented testing to failure. The data reported in such cases therefore do not reflect the ultimate properties, which are sure to be better. Film formability was tested under a small set of film casting conditions, since it in itself requires process optimization . Wherever not specified plasticization as well as compounding was done using hi KB screw configurations. When specific parameter comparison was made, a similarity in the rest of the parameters was maintained, unless otherwise stated. 24 INITIAL BROAD SET OF EPE:SF::70:30 60:40 EPE:SC ::70:30 60:40 EXPERIMENTS SELECTION OF SPECIFIC COMPOSITION PARAMETERS VARIED EPE:PSC(CMP):: 80:20 70:30 60:40 50:50 40:60 EPE:PSC::70:30 CMP PROTEIN SOURCE PROCESSING METHOD UNCMP ICMP PLASTICIZERS SCREW CONFIGURATION I EXTRUSION TEMPERATURE INJECTION MOLDING MELT INJECTION MOLDING COOL TIME FILM FORMABILITY PROPERTIES MEASURED PROCESSING AID - SILICONE F—D TENSILE STRENGTH & %ELONGATION MORPHOLOGY —> MELTING POINT DEPRESSION Figure 3 - Experimental Route 25 6.A1 6.A. ‘n~ .PL rt. as avo C “3"“. "‘."“*\r ‘ ‘4“VQ. ". ~Lr “L... on. r» oszm 362.“: 6.A MECHANICAL PROPERTIES 6.A.1 EFFECT OF PLASTICIZATION The presence of a plasticizer had a significant effect on the mechanical properties of the polymer (Figure 4). In fact, carrying out plasticization in an extruder was found to be important in promoting both tensile strength as well as %elongation at break. The difference between the CMP and UNCMP PSC is the duration and amount of exposure to heat and shear. Heat and shear denature proteins, exposing functional groups that become available for interaction. Processing under such conditions also increases the possibility of degradation of the protein, which may be favorable to a certain extent, if it were to increase the number of chain ends available for interaction. 2&0 lbOO 1400 a, 2000 5 I200 E 42.68 2 an :523 um “I . E — 800 1 91.10 000 3 (>00 30 g l 2 'mm Ii- ;5 1240 139-5 400 $0 470 564 200 0 . 0 SC ALONE UNCMP SC CMP SC sc ALCNE UNCNIP sc CMP sc Figure 4 — Effect of plasticization with glycerol on EPE:SC::70:30 blends. 26 6 .A.2 EFFECT OF BLENDING WITH SOY-PROTEIN SOURCES Blending of EPE with SC slightly increased both tensile strength as well as percent elongation at break compared to blending with SF (Figure 5). This increase was more appreciable after plasticization, especially with regards to tensile strength. The difference between the two sources is the percentage of protein. each. contains. This :means increased. presence of protein has a favorable effect on tensile properties of the blend. This can be explained if. an increase in protein content leads to an increase in the interaction between the protein and EPE or decrease the unknown effect due to the non—protein components. The increased difference in tensile strengths after plasticization suggests that presence of glycerol had a significant role in promoting the protein— polymer interaction. The percent elongation was not affected significantly by the use of either sources of protein. 27 BEFORE PLASTICIZAflON i . BEFORE PLASTICIZA'nON i I I400 I 5 “’0 ' . 35.23 ; I . 7622 I, i 58.74 30.62 - i ”f I i 5°° _—‘“ I I! m. I I P l l 400 I ,0, I 1164 1240 300 600 408 470 200 am 200 no 0 EPE:SF::70:30 EPE:SC::70:30 EPE :SF 1:70:30 EPE:SC::70:30 m AFTER PLAST'CIZATDN AFIER PLASTICZATION 1600 56.61 46.5 mm P— mo - g mm 1600 1w 1790 ‘ 1300 1509 1400 1500 1200 12m mo ‘°°° noo ”EPE:PSC::70:30 F 'EPE'PSFflD'SO mm2m °6Pepsr=zzmao ‘ ' Figure 5 - Effect of using different sources of protein before and after plasticization with GLY on EPE:SOY PROTEIN: :70:30. 6 .A. 3 EFFECT OF COMPOSITION A decrease in the PSC weight fraction led to an increase in both tensile strength and %elongation at break (Figure 6) . An increase in the PSC weight fraction above 40% led to a drastic degradation of both 28 tensile strength and %elongation to unacceptable levels. Compositions containing PSC below 40% weight were therefore studied in greater detail. It was due to better film properties obtained with the 70:30 composition that this composition was investigated. :more thoroughly' compared. to the 60:40 composition 2000 ' an 4an' __ g mu) *j5A1 um ll. E — 1650 ‘ 10M 0 an % 1200 5 E w %‘5" 2} an 10m) d n“ F: «m ”0 10 WIS 131 [-1‘ 0 4 . . 0 -—— 2'” . - 40:60 50:50 so;4o ”70:30 40:60 50:50 60:40 ”70:30 Isemxuwnnxmm I§EHKHWTW“W» Figure 6 — Effect of varying weight fraction of PSC on EPE:PSC blends 6.A.4 EFFECT OF DIFFERENT PLASTICIZERS Lower tensile strength and %elongation was obtained when ethylene glycol was used as a plasticizer instead of GLY (Figure 7). The differences in values are substantial. Processing difficulties were experienced during plasticization of soy concentrate with ethylene glycol as torque rose rapidly even at very low feed rates. This may 29 be due to the inability of ethylene glycol to effectively plasticize soy concentrate and hence improve flowability. IT 1 2000 I? K100 i . 236.94 ;: 210.89 ' 31...! ii j I . _..7. __ E 34.84 ‘i g 1200 _ I m 35.96 E 1592.2 t 800 III 800 3", 1234.4 ‘ 1211.17 2 . l‘.‘ 400 ‘ 400 849.00 'E j! 0 Y I swam. El'HYLBEG-LYmL i 0 - PLASTIC- ! QYC‘ERQ ETHYLENE QYCO. PLASTICIZEI Figure 7 - Effect of different plasticizers on EPE:PSC::70:30 blends 6 .A. 5 EFFECT OF EXTRUDER SCREW CONFIGURATIONS Three different screw configurations, shown in figures 8,9 and 10 were used. The screw configurations were classified based on the percentage of high shear kneading blocks present (Table 3). Four different combinations with different screw configurations used during the plasticization and the compounding steps were tried (Table 4). The location of the kneading blocks was fixed at approximately downstream of the feed zone. Optimization of the element positioning was not considered at this point. 30 TABLE 3 - BASE SCREW CONFIGURATIONS NOMENCLATURE KNEADING BLOCK AS %LENGTH OF TOTAL LENGTH HI KB 31.41 MOD KB 19.44 LO KB 10 . 86 TABLE 4 - COMBINATIONS OF SCREW CONFIGURATION USED FOR PLASTICIZATION AND COMPOUNDING CONFIGURATION DESCRIPTION P- PLASTICIZATION, C-COMPOUNDING C1 P- LO KB, C- LO KB C2 P- LO KB, C- HI KB C3 P- MOD KB, C- HI KB C4 P- HI KB, C- HI KB 31 screw length 160 900 am 750 10 ' 20 an namwwmflvm m u takeaway-mm) 20 I know-mun m Doorway-2021(1) I hum-409411) 450 Davey-3123(1) I Rum-awn ‘68 away-arms) nannmww» 21 Doorway-4221(1) 300 ncomaywdem 150 294 O screw UMNKB Figure 8 - Low KB screw configuration 32 1050 nmmwmmw n corwey-42/42(3) I knudngblock-405I1fl1) I knotdlng block-4051420) I knudlngbloek-405l14(3) D convey-W2) uwmwmfim n away-4242(4) I knudlngblock-mafl I knudlngbloek-AUOSIO‘I 4(3) uwmwmmm D WWW-2923(2) El away-4242(4) mo m 1% 7w u w) E m. 4% ‘ mm m a 1% 1m 0 80M MODKB Figure 9 — C2 Moderate KB screw configuration fl HIGH KB Figure 10 — C3 High KB screw configuration 34 Screws with lower kneading capability have a higher conveying capability, lower torque and power requirement. Hence they are preferred from the productivity point of View. Increasing the kneading blocks leads to better mixing and hence mechanical properties. However, it can also lead to degradation of temperature sensitive materials like SC. A compromise screw configuration is needed to solve the conflicting needs of higher productivity and energy requirements versus those for obtaining good mixing and hence better product quality. Tensile properties were found to be highly sensitive to the screw configurations used (Figure 11) . It was observed that both tensile strength and percent elongation improved with increase in the kneading blocks used. The importance of using a moderate to high shear screw during plasticization is especially clear when properties for the C2, C3 and C4 are compared. C2, C3 and C4, all use high KB screws for compounding and three different screws for plasticization. The differences in the properties can now be explained on the basis of the differences in screw configurations used during plasticization. An increase in the kneading capability of the screws used during 35 plasticization led to an improvement in both tensile strength and percent elongation. Thus both high shear and heat are necessary during plasticization in order to obtain good interaction between the two components. SC plasticized using low KB screws and then compounded with high KB screws had slightly lower tensile strength than that obtained by using low KB screws during compounding (c1, c2) . The difference however is within error bounds, suggesting that use of moderate to high KB screws during plasticization stage is necessary. Use of high KB screws during compounding does not make up for use of low KB screws during plasticization stage. Overall use of high KB screws during both plasticization and compounding resulted in the best tensile strength and %elongation. This emphasizes the need for thorough mixing of SC and plasticizer using both heat and shear, which can be achieved best in an extruder. Heat and shear therefore seem significant in property improvement. As stated elsewhere, heat and shear lead to denaturing of proteins, exposing functional group. Thus, if an increase in functional groups 36 is causing an improvement in tensile properties, it follows that there must be some positive interaction between the plasticized protein and EPE. :: 25m) 5 2000 —r-— 5 38.25 43.77 45.70 a:=> 15G) ”’ e. 1000 3 1253 1219 1575 25 500 z I! 0 , , fl C1 P-LO, C-LO C2 P-LO, C-Hl C3 P-NDD, C- C4 P-Hl,C-Hl HI SCREW CONFIGURATIONS :21600 o —. E £31200 0 116.17 52.50 -4 800 Ill 5 36.57 _ 748 1500 I“ 400 3% “5 55‘" l o I I I C1 P-LO, C-LO 02 P-LO, C-Hl C3 P-NDD, C-Hl C4 P-Hl,C-Hl SCREW CONFIGURATION Figure 11 — Effect of screw configuration on EPE:PSC::70:30 blends 37 6.A.6 EFFECT OF THERMAL HISTORY Compounding of EPE with PSC was carried out at two different temperatures - 135C and 170C. Each of these blends (3135C and E170) were injection molded at two different temperatures 135C and 160C (1135 and 1160). Among the blends injection molded at 135C, the E170 blend had higher tensile strength and %elongation. While among those injection molded at 170C, the E135 blend had higher tensile strength (Figure 12 and 13). The materials extruded at 135C when injection molded at 160C gave the highest tensile strengths. Blends extruded at 170C and injection molded at 135C gave the second highest. The lowest properties occurred when the material was both extruded and injection molded at 135C and the next lowest when the material was both extruded and injection molded at 160C. This could mean that exposure of the compound to higher temperatures during atleast one of the processing steps is required to obtain better properties. However exposure to high shear and heat during both steps is detrimental. Thus, there seems to be an optimum exposure to high temperatures. 38 Since during extrusion the material is subjected to severe shear stress compared injection nmlding, simultaneous imposition of high temperature may be leading to excessive degradation of the protein and hence loss of properties. I I ’l 2500 I 1600 I I 2m89 Fizz. I I 116.96 tn 2000 * g 236-94 F .200 g 1500 III E 1am 800 m 1000 1433 5'5 1592 1211 2 I! 500 400 o I '1350 '17oc 0 I mmmmm) "35° "7°C EXTIUSIONIEMPERAIURE Figure 12 - Effect of extrusion temperature on EPE:PSC::70:30 blends injection molded at 135C II 1000 1950 I 47.57 I I I . I 1 g 1850 I E I 1941 I 800 1500 m 18w I 1422 .1 a I z I IE 1750 m I 40° I | I I I 1n» . I I M1350 '1700 I. 0 I II ”135C 'I70C WONTBJPBATIEWI I sxnuslontsunnuua! I J .' Figure 13 - Effect of extrusion temperature on EPE:PSC::70:30 blend injection molded at 160C 39 The above discussion shows that the properties are an aggregate effect of the thermal history imposed by both the processing steps extrusion and injection molding. A material with lower thermal history during compounding resulted. in Ihigher tensile (properties 'when. subjected to higher temperatures during injection molding and vice- versa. However the decrease in tensile strength when the material was subjected to high temperature during both steps was not very high. Thus, higher temperatures lead to better properties than lower temperatures. The ideal combination. is the 'use of Lower temperatures (around 135C) during compounding and higher temperatures (around 160C) during injection molding 6mA.7 EFFECT OF DURATION OF COOLING DURING INUECTION MOLDING A decrease in cooling time during molding is important to reduce cycle times and increasing productivity during injection molding. Two cooling times were tried during the injection molding of EPE:PSC(GLY)::50:50. Halving mold cool time from 605 to 305 had negligible effect on the tensile strength as well percent elongation. A higher cooling time at an appropriate temperature is expected to allow better crystal growth thus improving tensile properties. Since the loss in mechanical properties is negligible a 308 cooling time seems adequate. cm :I no I 8.60 I 3 12.29 I I I i I w- I I I . a i I I I 60 I ; I I am II 78 76 880 I I 872 I I 4° ' I 825 I I I I 20 I I :I I em I 0 I 603 305 I I 605 30: I common»: m I I common»: m iL I Figure 14 - Effect of duration of cooling on EPE:PSC::50:50 blend. 6.A.8 EFFECT OF COMPOSITIION ON FILMS Film were cast from materials plasticized at .140C, compounded a t l 3 5C wi th a MOD KB screw during plasticization and HI KB screw during compounding. Films had a thickness of around 10 mils. Films obtained were translucent suggesting miscibility. An improvement in tensile strength was obtained when compared to the data reported by Eastman. However, work to check repeatability 41 could not be conducted. effect on the tensile strength of the films. Adding silicone had very little I t however improved the processbility of the EPE:PSC::60:40 blend. 6000 A 532.54 175 n. - 1:5 4000 (25 312.86 233.16 I‘- I 65.38 a) 4309 ":1 2000 1; 3098I 3178 2755 Ill 2163 '— 0 I T . , n9 :9 e“ 4" 9‘ 39° .3°' \00 \oo 4“? 0' 0' \v \v 9 9 '9 '9 9 Q) 4;? 39 .89" .43" 4,\ 93 <3 38' .88 <3 3° 9° 8 8 <2 2‘" Q7 <9 COMPOSITION Figure 15 - Effect of composition on film tensile strength (D882) 42 6 . B MORPHOLOGY ELI VII — pasce48: YIF Figure 16 - SEM pictures of soy concentrate particles. Top: A very wide particle size distribution can be seen. Bottom: Very irregularly shaped soy—particles 43 Soy concentrate particles shown in figure 16, have irregular shapes and a wide particle size distribution. Bigger particles appear to be aggregates of smaller ones. Particles have sharp edges. All blend pictures that follow show some common features. Since tensile fracture specimens were used the EPE matrix underwent ductile failure through yielding in the form of fibers. The differences were in the adhesion, particle sizes, particle size distribution and dispersion. EPE:SC::70:30 E135C resulted in a very distinctly phase separated blend. From figure 17, it can be easily appreciated that good mixing was not achieved, with particles showing practically no adhesion to the matrix. Failure occurred through yielding of the matrix alone in the form of fibers with no participation of the soy- concentrate particles. In fact, the almost free hanging particles with sharp edges could be acting as obstructions limiting the extension of EPE during the yielding process. This could lead to a reduction in the maximum elongation attainable and consequently the tensile strength. Also, a wide particle size distribution is seen. The largest particle detected was about 70m. ‘ I . ,‘I I" I“) 1 I “I; ’ “111‘. -2 Philip; E18. .rx. I 11 {c73100e TIF 9: c Figure 17 - SEM picture of EPE:SC::70:30. Very poor adhesion between matrix and particles, also very large 50—70pm particles exist 45 6.3.1 EFFECT OF PLASTICIZATION ON SOY CONCENTRATE Plasticized. soy-concentrate ‘when, compounded. ‘with. EPE (EPE:PSC::70:30 E14OC ) yielded the morphology in figure 18. On visual inspection the PSC particles appeared as dark brown, very brittle and glassy materials in the interior with a soft rough outer surface. In the blend, PSC particles show a better adhesion to the matrix and a finer dispersion compared to the EPE:SC::70:30. The particles are coated by the matrix better than in the case of the EPE:SC blend. Phase separation is still present, but the particle boundaries are slightly less distinct and. very slightly smeared. This may be due to the wetting or solvation of soy concentrate particles in the surface region leading to interaction with EPE. However, PSC does not seem to have melted or fused with the EPE matrix. Particle size distribution is improved vis-a-vis the EPE:SC blend but there is still noticeable variation and room. for improvement. The largest particles detected were in the range of lO-lSum. Also, the distribution of the particles in the matrix is not very uniform. 46 IDEAL-11?". TIF Figure 17 — SEM picture of EPE:PSC(30GLY) E140C, I120C. Tensile fracture zone. Slightly better adhesion. Distribution is uneven and there is room for improvement. 47 6.8.2 EFFECT OF PLASTICIZATION ON SOY FLOUR Compounding EPE with (UNCMP)PSF(GLY) was attempted to achieve cost reduction since a certain level of success had been. obtained. ‘using soy-concentrate. But inter—phase homogenization could not be achieved as can be seen in figure 19. The largest particle seen was about 25um. Almost no adhesion. was observed” MOst. particles 'were :not even coated. But this may be due to the use of PSF that had not been plasticized in an extruder. Lack of plasticization in the extruder seems to give ineffective mixing. A study EPE:(CMP)PSF should yield a better morphology. 48 Figure 19 — SEM picture of EPE:(UNCMP)PSF(G)::70:30. Very poor adhesion. No coated particles 49 Four compositions of EPE:PSC(G)::70:30 were prepared by extruding at El3SC and E17OC and each of these compositions was injection molded at two different temperatures 1135C and I160C. EPE:PSC(G)::70:30 E135C 1135C shown in figure 20, had a number of soy concentrate particles in the form of clusters leading to a very non—uniform dispersion. A number of particles seem to be coated but not smeared which would have meant some amount of dissolution. The particles appear to be of a finer size with most particles below 5pm. This blend showed the lowest tensile strength of the four compositions. EPE:PSC(G)::70:30 E135C 1160C shown in figure 21, has a very narrow particle size distribution as well as uniform dispersion within the matrix. A number of particles appear to be smeared and well coated with the matrix. Also the particle size is below Sum. Smearing suggests some amount of interaction. or dissolution of the PSC particles in the matrix; atleast at the surface. Sharp edges are hardly noticeable. This blend showed the highest tensile strength. 50 in) L U 2‘ r.- C'hx l 11,; F). r ‘y-.-.'¥v: . Figure 20 — SEM picture of EPE:PSC(G)::70:30 El35C I135C. Clusters or particles and a very non-uniform distribution of matrix are seen. 51 Figure 21 — SEM picture of EPE:PSC(G)::70:30 E135C I160C. A very uniform particle size distribution accompanied by reduction in particle size to less then Sum. Particles are well coated and smeared. 52 EPE:PSC(G)::70:30 E17OC I135C is shown in figure 22. PSC particles seem to be well coated though not particularly smeared with most particles below Sum. The adhesion between the particles and the matrix seems to be excellent since the matrix is stretching into thin fibers without disconnecting from the particles. However some variation in particle size and clustering of particles is noticeable. EPE:PSC(G)::70:BQ E17OC I160C shown in figure 23, has uniform particle size distribution. The particles are also very well coated and dispersed. A little smearing of particles leading to indistinctness of the particle boundaries is observable . Increase in temperature usually leads to a decrease in viscosity and hence an improvement in flowability, making good mixing possible. This should lead to an improvement in tensile properties. The improvement in morphology of the E170 I160 blend was not accompanied by a similar improvement in tensile properties. This can be explained if use of high temperature during both compounding and injection molding were to cause degradation of the protein molecules . 53 um“. 20 1" F'r '. ‘. I": Figure 22 - SEM picture of EPE:PSC(30GLY)::70:3O E170C 1135C. Particles are strongly adhering to the matrix even though the matrix itself is yielding. Non-uniform distribution of particles within the matrix is visible 54 Figure 23 - SEM picture of EPE:PSC(G)::70:30 E170C I160C. A uniform size distribution and a certain amount of smearing visible. Particles are well coated. 55 6.3.4 CONCLUSION Low temperatures during both compounding and injection molding leads to clusters and non—uniform dispersions. Simultaneous use of higher temperatures during both the steps leads to a reduction in particle size and uniform morphology, though it may also have been accompanied by degradation of soy protein. Use of low temperature during compounding and high temperature during injection molding led to increase in the smearing of the PSC particles in the blend, with fine particle size distribution. Low adhesion was observed in EPE:(UNCMP)PSF(G) blends. Passing PSF through the extruder during plasticization to form (CMP)PSF should yield better results. Above all the effect of how well plasticization was carried out during the formation of PSC and PSF can have a significant effect on the properties after compounding. It is not clear from these pictures whether what has been termed as PSC or PSF is indeed a thermoplastic material. Since the PSC seen in the images had a very particulate nature, it is likely that the material was cross-linked due to the —OH groups in glycerol. But the smearing seen in some of the blends and the persistence of the coating even 56 after material was stretched to failure seems to suggest a certain amount of dissolution or at least interaction between the three blend components, namely —SC, GLY and EPE 57 6.C DIFFERENTIAL SCANNING CALORIMETRY 400 0 50 100 150 Temperature C 700 g SOC 3 o 1:: SOC m v 400 T Figure 24 - Enthalpy versus Temperature of EPE -50 O 50 100 150 Temperature C 700 E 600 § 1:: 500 “'1 ' 4oc Figure 24 - Enthalpy versus Temperature of EPE 58 6.C.1 DUAL MELTING BEHAVIOR Figure 24 shows the dual-melting behavior of EPE. This phenomenon is observed in all its blends with PSC. The explanation for this behavior is usually the development of two different crystal species during crystallization. This is a very common phenomenon in case of copolymers (Kim, 1997) . 6.C.2 EFFECT OF ANNEALING TEMPERATURE ON MELTING POINT As can be seen from figure 25 the multiple melting characteristic is dependent on the annealing temperature, particularly, the melting temperature of the lower melting peak. As the annealing temperature increases the lower melting crystals rearrange to be part of the crystals with higher melting point. This is possible when the lower melting peak is caused by the crystallization of smaller, highly mobile molecules. Increase in melting point is usually a sign of increase in crystal size (Kim, 1997) . The increase in the melting point of the lower melting peak can thus be attributed to an increase in the crystal size, with eventual merging with the larger higher melting crystals. 59 10 l 10 I (A g a g g 8 J H I > ‘ A ; > I—‘+_—_'\- I 5 6 .— rr fl 3 6 E 4 I ' ' E 4 u 2 L A u 2 A A A Afi : v fi F v 7 fl 0 . . 0 . . . 30 40 50 60 70 80 l 30 40 50 a) 70 80 mm mm: +Hped<1+prt2 +TouH] [—0—peak1 H—l—pedQH—fi—tthqJezmcxm mo “meme 10 10 ENTHALPYJIG o to :- m a ENTHALPYJIG o M A a: on fink .f,,¥¢fi. ./+—o——o 304050007080 304050m7080 mm mm +pedt1H+ped<2H+mHmecpeczmfl [-0—pedt1H—I—ped2H—t—wrlqnecmczzmm Figure 26 — Enthalpy versus annealing temperature for four different compositions 6.C.3 EFFECT OF ANNEALING TEMPERATURE ON CRYSTALLINITY Figure 26 shows the enthalpy of fusion for the various compositions as a function of the annealing temperature. The enthalpy of fusion is a direct measure of crystallinity. Thus it can be seen that the total crystallinity. Thus it can be seen that the total crystallinity of the compositions studied, show a maxima in the temperature range of 40-7OC. The only exception was the 70:30 blend. The enthalpy of fusion for the lower melting peaks is much lower then the enthalpy of fusion of the higher melting peak for all the compositions. Thus, the higher melting peak is the major contributor towards the total crystallinity in the polymer. Each composition had a different temperature at which the maximum heat of fusion and hence crystallinity was obtained. As can be seen from figure 26, with increasing PSC content, a decrease in the annealing temperature increases the crystallinity. TABLE 5 - Effect of varying composition on the heat of fusion: Composition. Temp at max total B Total 8 J/g EPE 60C 6.231 EPE:PSC::80:20 60C 9.087 EPE:PSC::70:30 40C 8.062 EPE:PSC::60:4O 50C 6.946 The total enthalpy of all the blends considered was higher than that of EPE alone. Among the blends there 'was a 61 decrease in the highest enthalpy with increase in the PSC content. This can be explained if an increase in the PSC fraction may be hindering crystal growth. There are two possible explanations for higher heat of fusion of the blends over that of the EPE. The increase in the heat of fusion can be attributed to either denaturing of soy proteins, which escaped unaffected during processing or an increase in the crystallinity due to PSC fraction aiding crystallization. Denaturing would be likely if the heat of fusion increased with increasing soy concentrate fraction. But this does not happen. The other possibility is that the plasticization induced mobility of the soy protein molecules causes them to participate in the crystallization process in some kind of formation of co—crystalline domains. At the very least the PSC particles may be acting as nucleating agents. To summarize, increasing the annealing temperature increases the melting point of the lower melting peaks. However the crystallinity has a maxima at some intermediate temperature. This temperature can therefore be used for 62 injection mold cooling or for post forming annealing to improve crystallinity and hence mechanical properties. 6.C.4 EFFECT OF COMPOSITION ON THE MELTING POINT A depression in melting point of the higher melting peak with an increase in the PSC weight fraction is observable. However, there is no appreciable effect of composition on the melting point of the lower melting peak. .As the annealing temperature decreases, the depression in melting point of higher melting peak with increase in PSC content is :more impressive. There is no similar effect in the melting temperature of the lower melting peak even at different annealing temperatures. The 11 values obtained by fitting Hoffman-Weeks equation to the melting temperature versus annealing temperature plots for different compositions are listed in the table 6. According to theory, a value of n = 0 implies perfect crystals while n = 1 reflects inherently unstable crystals. Thus the lower melting peak is on account of unstable crystals while the higher melting peak is due to stable crystals. 63 TABLES - Hoffman Week's 1] parameters obtained by linear Jregression of Tm peak vs. Tc data Composition n for peak 1 n for peak 2 EPE 0.9957 -0.0197 Epe:psc::80:20 0.9829 0.0315 Epe:psc::70:30 0.9822 0.0661 Qpezpscusozzio 0.9895 0.0007 The significance of this discussion is that a melting point depression in the higher melting peak, arising from perfect crystals with increase in the amorphous phase can be taken as an indication of miscibility. i A1T0840C ES 40 3) 7O 90 T madame —O—Impajrlortd +Tmpeod +Tmpak2cnd -I—Tmp£o<2 AtTc=SOC —e—Tmpeonorse +Tmpck1 —e—Tmpeo(2orne +Impa'x2 ATTCIm 588 s 8258 8 8 5 0 "PE —e—1mpeo(lomd +Tmpeo¢l +Tmpeck2ond —II—Tmpeo(2 AtTc-OOC 1E?! —O—Tmpnlmd +Tmpai