A} .5... .5 r 1:: I . . 1......12 I: 3. x. ’3‘: .e .539! I 3.3:... :r. . $3.? I ‘ 5 1.HI: ~ . ‘9 9 n2 1...}.? «all! 3pr u , x05» (vistfiflrflnflnhl . .wv ..m.uxw‘u.m§..fl .aiillixs w w .39; { Inmate: :L 2min LIBRARY Michigan State +1 University This is to certify that the thesis entitled THE DESIGN AND ENGINEERING OF STARCH:POLYESTER BIODEGRADABLE PLASTICS USING REACTIVE EXTRUSION presented by Dale Michael Smith has been accepted towards fulfillment of the requirements for M. 3. degree in Chemical Engineering ll [Kt/w W. A (TL/L377 Go». I Major professor (J Date {Mg} :27, me 0-7639 MS U i: an Affirmative Action/Equal Opportunity Institution PLACE IN RETURN BOX to remove this checkout 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 SEP13 2007 052102 an fin to; 9m LUUL 2/05 p:/ClRC/Date0ue.indd~p.1 THE DESIGN AND ENGINEERING OF STARCHzPOLYESTER BIODEGRADABLE PLASTICS USING REACTIVE EXTRUSION by Dale Michael Smith 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 THE DESIGN AND ENGINEERING OF STARCHzPOLYESTER BIODEGRADABLE PLASTICS USING REACTIVE EXTRUSION By Dale Michael Smith Environmental and life cycle considerations have excited recent interest in biodegradable polymeric materials. Properties of many of the commercially available biodegradable materials are not competitive with traditional non-biodegradable materials. In this study reactive extrusion was used to compatibilize a commercially available biodegradable polyester. This polyester was then blended with thermoplastic starch. The properties of these blends were examined and a plant design and cost analysis was performed to determine the feasibility of producing such blends on a large scale. Copyright by Dale Michael Smith 1 998 To my parents who made this all possible, and To Tammy who never let me lose sight of this dream. iv ACKNOWLEDGEMENTS I would like to thank Dr. Ramani Narayan for providing his uniquely styled guidance and support. In addition I would like to thank the Department of Chemical Engineering for providing me with an enjoyable learning environment, and the Composite Materials Structure Center; MBI International and the School of Packaging for use of excellent facilities. I would also like to thank several others that provided me with help and/or inspiration in my life: -Gary Scherger, thank you for instilling in me my love for the sciences. -Myron Cline, thank you for showing me that life truly is about the journey. -The University of Dayton Department of Chemistry, thank you for providing an excellent educational base that made my transition to chemical engineering possible -Bob Lehmann, thank you for teaching me how to truly write a good manuscript and how to keep good laboratory notebooks. Jose Sanchez , thank you for providing me inspiration and wisdom in so many ways- Finally, I thank my Lord who has guided my down this path of life bringing me only true happiness when I follow him willingly! TABLE OF CONTENTS Chapter List of Tables List of Figures 1 .Introduction Goals and Objectives Rationale for Biodegradable Plastics Biodegradable Polymer Systems Polymer Blend Terminology Polymer Blend Compatibility Compatibilization of Polymer Blends by Reactive Extrusion 2. Background Polymer Blends Basic Thermodynamics Blend Morphology Blend Preparation Miscibility determination Compatibilization vi Page xi ix IO 11 12 I3 15 16 18 18 20 Viscosity of Dilute Polymer Systems 3. Design of a Compatibilizing Agent Synthesis of Maleated Soybean Oil Grafting of Maleated Soybean Oil onto Eastar Bio Copolyester Characterization of Maleated Soybean Oil and the Grafted Copolyester Products MSO Characterization Extraction Intrinsic Viscosity 4. Preparation of Materials Preparation of Thermoplastic Starch Reactive Blending of Thermoplastic Starch with the Grafted Copolyester Injection Molding of Test Bars 5. Properties of Biodegradable EBC-g-MSO Blends Examination of Morphology by Environmental Scanning Electron Microscopy Tensile Testing Biodegradability Testing Under Composting Conditions 6. Preliminary Process Design Process Description Process Flow Chart Extruder Sizing vii 21 25 28 34 34 35 36 38 42 44 53 58 61 62 63 Material and Energy Balances 7. Economic Analysis Process Economics 8. Conclusions and Recommendations Conclusions Recommendations for further work Bibliography viii 65 69 79 81 84 Figure 1.1 1.2 1.3 1.4 2.1 2.2 2.3 3.1 3.2 3.3 3.4 3.5 3.6 4.1 LIST OF FIGURES Page Experimental Protocol 3 Composition of municipal solid waste in the USA, 1993 5 Polymer blend classification scheme 10 Location of block or graft copolymers at the interface in blend systems Property relationship in polymer blends Dispersion of a polymer (dark regions in the matrix of an immiscible polymer: a) spherical droplets b) platelets c) fibrils Fluid flow behavior A) Variation of laminar flow with respect to the distance r from the center of a tube. B) Sphere suspended in a flowing fluid Experimental apparatus for the maleation of soybean oil Molecular Structure of Dicumyl Peroxide Structure of Eastar Bio Copolyester Grafting of maleated soybean oil onto Eastar Bio Copolyester Screw configuration for reactive extrusion Intrinsic Viscosity Determination for Eastar Polyesters Screw configuration for starch plasticization ix 14 17 22 27 28 28 3O 32 34 39 4.2 4.3 5.1 5.2 5.3. 5.4 5.5 5.6 5.7 5.8 5.9 5.10 5.11 5.12 5.13 6.1 7.1 7.2 7.3 ESEM of Starch Granules ESEM of Thermoplastic Starch ESEM of Eastar Bio Copolyester EBC:TPS 70:30 blend, note the small starch particles clearly visible EBC:TPS; 70:30 blend, note large TPS particles (lighter color) laying on the surface EBC-g-MSOI OGzTPS; 70:30 blend, note thermoplastic starch particles are partially integrated into the grafted polyester matrix EBC:EBC-g-MSOIOGzTPS; 65:5:30 blend, note no thermoplastic starch particles are visible EBC:TPS; 70:30 blend, note no adhesion of thermoplastic starch particles to matrix EBC-g-MSOIOGzTPS; 70:30 blend, note some adhesion of thermoplastic starch particles to the matrix EBC:EBC-g-MSOIOGzTPS; 65:5:30 blend, note thermoplastic starch particles are no longer readily distinguishable from polyester matrix Tensile test results for modified copolyesters Tensile strength at break for polyester blends Tensile testing of modified polyester: thermoplastic starch blends Biodegradability Testing Apparatus Biodegradability Testing Results Process Flow Diagram Determination of Internal Rate of Return by a discounted cash flow for the proposed process Sensitivity analysis for the proposed process Effects of varying graft copolyester amounts on material costs 40 41 45 47 48 49 50 51 52 54 56 57 59 60 62 76 77 78 List of Tables Table Page 3.1 Soybean Oil Composition 25 3.2 Selected Properties of Eastar Bio C0polyester 28 3.3 Temperature profile for reactive extrusion 33 3.4 Grafting efficiency of reactive extrustion 35 4.1 Extruder temperature profile for starch plasticization 38 4.2 Temperature profile for compounding with thermoplastic starch 42 4.3 Conditions for injection molding of Specimens 43 6.1 Material and Energy Balances for Production of Biodegradable Resins 65 7.1 Assumptions for Economic Evaluation 70 7.2 Equipment List for Resin Production 71 7.3 Capital Investment Breakdown 72 7.4 Manufacturing cost breakdown for 5 million pound per year plant 73 7.5 Selling price based on discounted cash flow 75 xi CHAPTER I INTRODUCTION Goals and Objectives The goal of this work are twofold: 1) the design and engineering of biodegradable starchzcopolyester blend. The copolyester used in this blend is Eastar Bio Copolyester (EBC), a biodegradable copolyester containing Adipic acid, Terepthalic acid, and l ,4-Butanediol. 2) the evaluation of maleated soybean oil (MSO) as a compatibilizing agent. Specifically the objectives are as follows: 1) Design of a compatibilizing agent a) Maleation of soybean oil b) Grafting of maleated soybean oil onto the Eastar Bio Copolyester c) Characterization of the maleated soybean oil and the grafted copolyester products 2) Preparation of Materials a) Preparation of Thermoplastic Starch a) Reactive blending of thermoplastic starch with the grafted copolyester 3) Determination of properties of the biodegradable EBC-g-MSO: TPS blends a) Examination of morphology by Environmental Scanning Electron microscopy. b) Tensile testing of materials c) Biodegradability testing of the EBC-g-MSO: TPS blends 4) Engineering of a large scale process a) description of a 5 million pound per year plant b) calculation of material and energy balances c) sizing of process equipment 5) Analysis of process economics a) Determination of the Discounted Cash Flow selling price b) Analysis of sensitivity of selling price to cost fluctuation Chapter 2 provides the literature background for this work. Objective 1 is discussed in Chapter 3. Objective 2 is described in Chapter 4. Objective 3 is outlined in Chapter 5. The process engineering and economics are described in Chapters 6 and 7. Conclusions and further work are finally discussed in Chapter 8. An outline of this work is shown in Figure 1.1. Eastar Bio Copolyester Maleated Soybean OII Dicumyl Peroxide 1)Reactive Extrusion to produce grafted copolyester 2)Pel|etizing and drying Materials Process Eastar Bio Copolyester- g-Maleated Soybean Oil (EBC-g-MSO1OG) Product Starch Glycerol Polycaprolactone 1) Extrusion to produce thermoplastic starch 2) Pelletizing and drying Thermoplastic Starch (T PS) V 1) Reactive extrusion blending 2) Pelletizing and drying Viscosity Measurements Scanning Electron Extraction Microsopy Biodegradability Testing 1) Injection Molding 2) Tensile Testing Characterization Figure 1.1 Experimental Protocol Rationale for Biodegradable Plastics New environmental regulations, societal concerns, and a growing environmental awareness throughout the world have triggered the search for new products and processes that are compatible with the environment. Thus, new products have to be designed and engineered from cradle to grave incorporating a holistic "life cycle thinking" approach. The impact of raw material resources used in the manufacture of a product and the ultimate fate (disposal) of the product when it enters the waste stream have to be factored into the design of the product. The use of annually renewable resources and the biodegradability or recyclability of the product are becoming important design criteria. This has opened up new market opportunities for developing biodegradable products.(Narayan, 1998). Currently, most products are designed with limited consideration of their ultimate disposability. Of particular concern are plastics used in single-use disposable packaging. Designing these materials to be biodegradable and ensuring that they end up in an appropriate disposal system is environmentally and ecologically sound. For example, by composting our biodegradable plastic and paper waste along with other "organic" compostable materials like yard, food, and agricultural wastes, we can generate much- needed carbon-rich compost (humic material). Compost amended soil has beneficial effects by increasing soil organic carbon, increasing water and nutrient retention, reducing chemical inputs, and suppressing plant disease. Composting infrastructures, so important for the use and disposal of biodegradable plastics, are growing in the US. and are in part being regulatory driven on the state level. Polymers have been designed in the past to resist degradation. The challenge is to design polymers that have the necessary functionality during use, but destruct under the stimulus of an environmental trigger after use. The trigger could be microbial, hydrolytic or oxidation susceptible linkages built into the backbone of the polymer, or additives that catalyze breakdown of the polymer chains in specific environments. More importantly, the breakdown products should not be toxic or persist in the environment, and should be completely utilized by soil microorganisms. In order to ensure market acceptance of biodegradable products, the ultimate biodegradability of these materials in the appropriate waste management infrastructures needs to be demonstrated beyond doubt. The US generates 207 million tons of municipal solid waste each year. (Franklin and Associates, 1994) Plastics account for 9.3 % of this waste stream as seen in Figure 1.2. The traditional waste disposal altematives-recycling, incineration, and landfilling account for 22%, 16% and 62% of the total Municipal solid waste disposed of today. Other Plastics 13% 9%Metals 8% Yard Waste Food Waste 16% 7% Rubber/Leather Glass 7% 3% Paper Products 37% Figure 1.2: Composition of municipal solid waste in the USA, 1993 Mounting societal concerns about our environment have led to an increased interest in making polymeric materials from renewable resources. The primary reasons for this interest are (Narayan 1991): 0 Environmental concerns 0 Providing alternate feedstocks to non-renewable imported petroleum feedstocks o The need to utilize the nation’s abundant agricultural feedstocks - Legislation and public opinion The incorporation of renewable resources into polymeric systems can often result in polymers that are partially or fully biodegradable. Bags for collection of yard trimmings, food residuals, and other organics, cutlery, plates, and agricultural film are among the items made from such polymer systems that are commercially available.(Riggle, 1998) These biodegradable products in a composting infrastructure will can be converted into CO2 and water by microbial activity thus helping reduce the solid waste problem in this country. Biodegradable Polymer Systems ASTM, European (CEN), and ISO (International Standards Organization) Standards have been developed or are under development to evaluate biodegradability under different environmental/disposal conditions like composting, soil, marine, wastewater treatment facility, and anaerobic digestors. Armed with an understanding of the rationale for biodegradable plastics, and with standards in place to evaluate biodegradability, technologies are under development that meet biodegradability and/or compostability criteria. The technologies can be classified in the following manner: 1. Aliphatic polyester 0 petrochemical feedstock o agricultural feedstock o microbial synthesis 2. Natural polymer 0 starch & starch derivatives (starch esters) o cellulose and cellulose esters - proteins, polysaccharides and amino acids 3. Blends, alloys and graft copolymers of natural polymers and polyesters Aliphatic polyesters are excellent candidates for biodegradable plastic products. The basic composition involves a polyester prepared by using diols like ethylene glycol, 1,4- butanediol, 1,6-hexanediol, 1,4-cyclohexanedimethanol and dicarboxylic acids like succinic, sebacic, and adipic acid. In some cases, a few mole percent terephthalic acid has been used along with adipic acid to generate copolyesters to obtain suitable properties for plastics applications. Polyesteramide copolyesters have also been introduced into the marketplace as fully biodegradable resins for plastic applications. Poly(lactic acid) polymers (PLA) are derived from agricultural feedstocks. There is considerable commercial and R&D activity in this area. The basic chemistry involves step polymerization (condensation polymerization) of lactic acid (or-hydroxy acid). High molecular weight poly(lactic acid) polymer has been prepared by the direct condensation route. This has been achieved by carrying out the reaction in a high boiling solvent using Sn compounds or protoic acids as catalysts. The conventional route to high molecular weight PLA is through the dilactone of lactic acid by ring opening polymerization. Polyhydroxyalkanoates (PHBV, poly(3-hydroxybutyrate-co-3-hydroxyvalerate)) are novel thermoplastic polyesters that are prepared by microbial synthesis using a variety of feed stocks including glucose and acetic acid . Starch formulated with addditives & plasticizers has found application as loose- fill packaging material. They offer a biodegradable, water soluble, anti-static, environmentally friendly alternative to expanded polystyrene (peanut packaging). Water functions as a plasticizer and blowing agent. Starch based foams have captured 15-20% of this market. However, because of starch’s extreme water sensitivity, it cannot be used in the majority of plastics applications. Starch esters and blends of starch esters with biodegradable aliphatic polyesters provide better water resistance and processability, and are being evaluated for various molded products and film applications. Cellulose esters with a degree of substitution of around 2 are also potential biodegradable resins. Blends of starch and poly(caprolactone) (PCL) have been successfully used as compost bags. However, because PCL has a low melting point it is unsuitable for many other film applications. Polymers based on renewable resources such as starch, tend to be more expensive and have inferior properties compared to petroleum based polymers. To offset these problems these polymers are often blended with other synthetic biodgradable polymers.(Karnani, 1996). Blending of polymers offers several advantages (Utracki, 1989): o Extending engineering resin performance by diluting it with a low cost polymer 0 Developing materials with a full set of desired properties 0 Forming a high performance blend from synergistically interacting polymers - Adjusting the composition of the blend to customer specifications 0 Recycling industrial and/or municipal plastics scrap Polymer Blend Terminology In the study of polymer blends, some terminology must be introduced. The relationship between the various terms used in the polymer blending field can be visualized as shown in Figure 1.2. The terminology described below (Utracki, 1989) will be used throughout this thesis: a) Polymer Blend: a mixture of at least two polymers or copolymers. b) Miscible Polymer Blend: a polymer blend homogenous down to the molecular level, associated with the negative value of the free energy of mixing (1) AGmEAHmSO c) Immiscible Polymer Blend: any polymer blend whose GmsAHm>0 (1) Compatible mlymer blend: a utilitarian term indicating a commercially attractive polymer mixture, normally homogenous to the eye, frequently with enhanced physical properties over the constituent polymers e) Polvmer Allov: an immiscible PB having a modified interface and/or morphology f) Compatibilization: a process of modification of interfacial properties of an immiscible polymer blend, leading the creation of a polymer alloy. Polymers Copolymers Polymer Blends Miscible Blends Immiscible Blends Compatilglization Polymer Alloys Figure 1.3 Polymer blend classification scheme 10 Polymer Blend Compatibility Most polymer blends are not miscible, (Paul, 1978) but miscibility is neither required nor desired in all polymer blends. Immiscible but compatibilized blends- polymer alloys, often exhibit synergistic properties that are characteristic of both components in the blend. For this synergism to occur however there must be some adhesion between the two phases in the system. Mixing natural polymers with synthetic ones results often in two-phase morphologies with high interfacial tension and poor adhesion between the phases. An incompatible blend with poor properties is the result. Two ways to improve adhesion between the phases are by using a compatibilizing agent or by modification of the blend components. (Xanthos, 1988) Block and graft A-B copolymers located at the interface between the two phases can reduce the surface tension between the A-rich and B-rich phases thereby serving well as compatibilizers. O :A Unit C : B Unit Phase A Interface Phase B Graft Block Copolymer Copolymer Figure 1.4: Location of block or graft copolymers at the interface in blend systems 11 Compatibilization of Polymer Blends by Reactive Extrusion Extruders can melt, mix, pump compound, and devolatilize high viscosity polymers. These characteristics are essential characteristics for a chemical reactor. Reactive extrusion is a process by which polymers undergo a chemical modification during the extrusion process. The advantages to performing a reaction in an extruder as opposed to in a solvent or diluent include (Xanthos, 1992): eliminate the energy of recovery of the diluent if no solvent or diluent is used, there will be no emissions from it. most of the plant equipment and the space it occupies can be saved controlled residence time and temperatures improved surface/volume ratio Reactive extrusion is beneficial in production of polymer blends because the blending and compatibilization of the two components can be performed in one step. This in-line compounding can reduce the machinery investment and energy cost for processors by more than 50%. (Kreisher, 1990) 12 CHAPTER 2 Background Polymer Blends Over the last several decades the realization that new molecules are not necessary for the development of new materials has driven the commercial and scientific progress of the field of polymer blends. . Blending can usually be implemented more rapidly and economically than the development of new materials} It is estimated that successful commercialization of a new polymer requires 15-20 years while development of an alloy or blend based on existing polymers may take half as long. (Kienzle, 1988) Blending already consumes 30 wt% of all polymers, and is growing at a rate of around 9%/year. Successful polymer blending requires a sound scientific basis. Two component polymer mixtures can be described often by the following equation: (Kienzle, 1988) P = P,C,+P2C2+IP,P2 (1) P is a property value of the blend. P1 and P2 are the property values of the components. C , and C2 are the concentrations of the components, and I is an interaction coefficient. The property relationship of polymer blends is illustrated below in figure 2.1. The polymer properties of the blend are a weighted arithmetic average of the constituents properties if I equals zero. If the blend is incompatible, usually because of poor interfacial adhesion, I has a negative value. Polymer compatibilization results in polymer l3 alloys that possess positive values of I, i.e. the properties of the polymer combination are better than the weighted arithmetic average of the constituents properties. AHOY (|>0) Actitive R&sdt (|=0) Property Imon‘patible Herr! (|<0) 0% 50% ConcentrationofPolymer1 Figure 2.1: Property relationship in polymer blends l4 Basic Thermodynamics The free energy of mixing governs the equilibrium phase behavior of mixtures as illustrated by the expression: AGmix = AHmi,‘ — TASmi,‘ (2) where AG is the free energy of mixing, AH,nix the enthalpy of mixing, T the mix temperature, and AS the entropy of mixing. The Flory-Huggins equation can be used to calculate the entropy of mixing. Asmix = '15 miln¢r+Nzln¢2) (3) where k is Boltzrnan’s constant, ii), is the volume fraction of component I, and N1 is the number of moles of component i. As a result of the large molecular weight of each polymer in a typical blend, the number of moles of each polymer and hence the ASmix is very small. It therefore does not contribute substantially to the free energy of mixing. The enthalpy of mixing then becomes key to determining the miscibility of a polymer blend. The enthalpy of mixing expressed as a function of solubility parameters 6, volume fractions o, and total volume V is always positive. AHmix = V¢I¢2(51'82)2 (4) This results in a positive AGM since ASM is very small. The entropy of mixing can at best approach zero. The enthalpy of mixing can be negative however if specific interactions such as hydrogen bonding or dipole-dipole interactions occur. This negative enthalpy of mixing in turn can result in a negative free energy of mixing and the accompanying blend miscibility. Blend Morphology Most combinations of polymers are immiscible. Phase separated blends are often preferred for achieving useful properties. In principle such polymer-polymer composites yield materials whose stiffness could be adjusted to any value between those of the component polymers. To blend materials in such a fashion, control over the morphology of the phases must be exercised. The morphology generated during mixing of immiscible blends depends on three factors: interfacial tension between the phases, elasticity, and viscosity. (Han, 1981) Several general rules can be applied to determine which polymer is the continuous phase in the blend and which is the dispersed phase. Generally, three types of common morphology of immiscible blends include; the spherical droplet, the platelet, and the fibril. Fibrils are formed from droplets by uniaxial elongational flow, eg extrusion through a die. Bi-axial stretching of droplets forms platelets by processes such as blow molding. Examples of such morphologies can be seen in Figure 2.2 16 Figure 2.2: Dispersion of a polymer (dark regions in the matrix of an immiscible polymer: a) spherical droplets b) platelets c) fibrils In these blends, the continuous phase is often the component occupying the most space. The lower viscosity component often will encapsulate the more viscous component reducing the rate of energy dissipation. These two factors can combine to yield a polymer that has a distinct continuous phase and dispersed phase or a blend with co-continuous phases of both polymers. What must be considered at all times is that morphology created during processing is a dynamic structure that is subject to change as a result of further processing. 17 Blend Preparation Solution casting and melt mixing are the most common techniques used for preparing polymer blends. For immiscible pairs the details of the mixing process determine the composite morphology The economics of melt mixing make it the most common method for preparing polymer blends. Disadvantages of melt mixing such as temperature induced degradation, difficulties in mixing materials with large differences in melt viscosity, and cleaning and the machinery, and equipment cost are often offset by the advantages of melt mixing. The primary advantages being Speed, no solvents-environmentally friendly, and simplicity. Solution casting is often used for small quantities of for polymers that do not process well by melt mixing methods. Care must be exercised when preparing a solution cast material if the true miscibility is to be determined. For example two immiscible polymers form a single phase solution when diluted enough in solvent. Rapid solvent removal can trap the polymer in a non-equilibrium homogenous state. (Schultz, 1976, 1980). In any processing situation, if miscibility is to be determined care must be exercised to determine if in fact physical equilibrium has been achieved. Miscibility Determination Visual inspection of an immiscible blend reveals a material of limited transparency that is caused by the light scattering due to phase separation. Domain sizes that are small relative to the wavelength of light or are of similar refractive indices limit light scattering, making visual inspection unreliable. Another problem associated with polymer visual inspection occurs if a blend with a miscible amorphous phase also contains a crystalline phase. This phase will scatter light as described and reduce transparency (Paul, 1988). Examination of the glass transition is a simple and often reliable method for determining the miscibility of a blend. Miscible blends exhibit a single transition, while two-phase blends exhibit two Tg’s, characteristic of the separate components. Several relationships have been proposed to describe the composition dependence of the glass transition. One of the simplest of these is the Fox relationship, (Fox, 1956): _1_ = IL + )1: (5) T8 7:21 T82 For accurate results however, the glass transitions must be separated by approximately 20 °C, and the concentration of the lesser material should be greater that 10%. Dynamic mechanical analysis is more sensitive than DSC. Less than 1% of the non-miscible component can be detected. However it requires relatively more complex equipment. A third analytical technique that is extensively used for determining miscibility is electron microscopy.(Shaw, 1985) Appearance of two phases in electron microscopy would Show that the components are not miscible. Poor phase contrast and expensive equipment, though. disadvantages of this technique have not outweighed the advantages associated with it.(Folkes, 1993) 19 Compatibilization Most polymers are thermodynamically immiscible and do not result in homogenous blends upon mixing. Typically, high interfacial tension and poor adhesion is seen at the interface. The high viscosity and inherent difficulty of imparting the desired degree of dispersion to random mixtures caused by the interfacial tension leads to lack of stability of the blend upon further processing (Paul, 1978b). The brittle mechanical behavior observed in dispersed blends is caused by the poor adhesion. (Paul, 1978b) The two general routes for achieving compatibility are by a) adding a third component capable of interactions with the blend constituents, block or graft copolymer are examples of this or b) blending functionalized polymers capable of enhanced specific interactions, and/or chemical reactions.(Xanthos, 1988) Systems compatibilized by graft or block copolymers include polystyrene- polyethylene blends (Barentsen et a1, 1974) and cellulose acetate-polyacrylonitrile blends.(Paul, 1978b) Functionalized polymers can be prepared via reactive extrusion. (Tucker, 1987). Examples of such functionalized polymers include anhydride-modified polypropylene used in nylon 6/polypropylene blends. Styrene-glycidyl methacyrlate- reacts with poly ethylene terephthalate- to form an in situ compatibilizer in PET/PS blends (MAA, 1993). In natural fiber systems, biomass fibers have been compatibilized with hydrophobic materials (N arayan, 1991) 20 The addition of compatibilizer can in principle affect the blend system in three ways: 1. Reduce interfacial tension leading to finer dispersions during mixing 2. Increase adhesion at the phase boundaries thereby facilitating stress transfer. 3. Stabilize the dispersed phase against growth. These factors usually lead to an increase in mechanical properties such as impact strength, and tensile strength. Consequently, these macroscopic properties can be used as an indirect means of measuring compatibilization. Viscosity of Dilute Polymer Systems The viscosity of dilute polymer solutions is considerably higher than that of pure solvent. This increase in viscosity depends on temperature, the nature of the solvent and polymer, the polymer concentration, and the sizes of the polymer molecules. Since viscosity is dependent on sizes of molecules, an average molecular weight can be determined from the solution viscosity of such a system. If no Slip boundary conditions at the tube wall are imposed on a fluid flowing through a tube, the resulting velocity profile can be visualized as in Figure 2.3.A. A particle suspended in such a flowing fluid would rotate since it impinges on fluid particles flowing at different rates as shown in Figure 2.3.3. 21 Radial Position, r ‘_—___ Tube Wall > .5 i3 a) I > IV 11% V A B Figure 2.3: Fluid flow behavior A) Variation of laminar flow with respect to the distance r from the center of a tube. B) Sphere suspended in a flowing fluid Rotation of this suspended particle, since it is wetted by the liquid, brings adhering liquid from a region with one velocity into a volume element which is flowing at a different speed. The readjustrnents of momenta because of this wetting cause an expenditure of energy, which is greater than that which, would be required to keep the same volume of fluid moving with a particular velocity gradient. Consequently, the suspension has a higher viscosity than the suspending medium. It can be shown that the intrinsic viscosity is related to Viscosity Average Molecular Weight by the Mark-Houwink-Sakurada (MHS) relation (Rudin, 1982). 22 [II] = KM; (6) Given K, a, and [n], the viscosity average molecular weight can be determined. K and a are constants that are characteristics of particular solute-solvent systems. Intrinsic Viscosity can be found by measuring the suspension viscosity at different polymer concentrations. An extrapolation method using the Kraemer equation (7): __1_ 3. [n] — chino) (7) can be used to determine intrinsic viscosity. 1] and no are the suspension and solute viscosities respectively while c is the concentration of the polymer in the solution. For dilute polymer solutions 11h]0 = t/t0 where t and to are the times require for viscometer flow times for the suspension and solvent respectively. By extrapolating equation (7) to zero concentration the intrinsic viscosity of a particular polymer can be obtained. The other constants in the MHS relation can be found by can be found by first fractionating a polymer into nearly mono-disperse molecular weight fractions. For mono- disperse polymer fractions [17,, = Mn= any average molecular weight of the sample. The intrinsic viscosities of a number of such mono-disperse fractions are then fit to the equation ln[n]=an+aln(A7v) (8) 23 to yield the MHS constants for a particular polymer/solvent system. With these parameters then the Viscosity average molecular weight of two samples of the same polymer can be compared by simple viscosity measurements using readily available equipment such as a Ubbelohde viscometer. 24 CHAPTER 3 Design of a Compatibilizing Agent Synthesis of Maleated Soybean Oil Soybean oil was purchased from Spectrum chemical. This oil had a molecular weight of approximately 872. The typical composition of soybean oil is shown in Table 3.1 (Kirk-Othmer, 1983). Carboxylic acid Chemical Acronym“ Wt% Soybean Oila Fatty Acid Saturated fatty acid Palmitic acid 16:0 10.7 Stearic acid 18:0 3.87 Unsaturated fatty acid Oleic acid 18:1 22.8 Linoleic acid 18:2 50.8 Linolenic acid 18:3 6.76 *Chain length, number of double bonds, and functional groups, if any Table 3.1: Soybean Oil Composition Maleation of soybean oil was performed using 99% pure reagent grade maleic anhydride, and Lupersol 101 as a catalyst. The reaction chemistry postulated by Root (Root, 1945) suggests that soybean oil in the presence of a catalyst like peroxide will form conjugated double bonds from non-conjugated double bonds. These double bonds can then undergo a 1-4 Diels-Alder type of a reaction, resulting in the addition of maleic anhydride to the soybean oil as shown below. 25 13 12 11 10 9 CH3(CH2)4CH=CHCH2CH=CH(CH2)7COOR l peroxide C H3(CH2)4CH2CH=CHCH=CH(CH2)7COOR .M. l CH3(CH2)4CH20H=CHEH\-CH(CH2)7COOR o ' o CH=CH \ CH3(CH2)4CHZCH H(CH2)7COOR o 0 0 Synthesis of maleated soy oil was performed in a 2L Parr reactor. Approximately 975g of soybean oil were combined with 125g maleic anhydride and 10ml of Lupersol 101 in the reactor. Lupersol 101 a mixture of 2,5-Dimethyl-2,5-di-tert-butylperoxy- hexane, Di-tert-butyl peroxide, 3,3,6,6-tetramethyl-l,2-dioxacyclohexane, and 2,2,5,5- tetramethyl tetrahydrofuran, was obtained from Vanderbilt under the industrial name Varox BPDH. This organic peroxide used as a catalyst in the maleation of soy oil possesses a half-life of 1 minute at 180°C and 13 seconds at 200°C. Reaction time was 30 minutes. Reaction temperature was 150°C. The reaction times and temperatures and 26 stoichiometries chosen were optimums based on previous work. (Seybold, 1997) The above reaction was performed under a nitrogen atmosphere. Vacuum was pulled on the reaction vessel at the end of the reaction time to remove any excess maleic anhydride and catalyst. A schematic of the reactor system used is shown in Figure 3.1. l:||:ll:| O 0 Temp (C) Motor Heating RPM Mantle (C) Process Controller Process — — — - — Electrical Control Figure 3.1: Experimental apparatus for the maleation of soybean oil 27 Grafting of Maleated Soybean Oil onto Eastar Bio Copolyester Maleated Soybean Oil was reactively grafted to Eastar Bio Copolyester using the ZSK 3O twin screw extruder, and Dicumyl Peroxide as a catalyst. (DCP) Dicumyl Peroxide is a dialkyl peroxide that is often used as an initiator in polymerization and crosslinking reactions. (Kirk Othmer, 1983b) The material used in the grafting reactions was obtained from Aldrich. The melting point of this material is 78°C and the 10hr half-life temperature is 115°C. The structure of Dicumyl peroxide is shown below. CT3 C H. C H3 CH3 Figure 3.2: Molecular Structure of Dicumyl Peroxide Eastar Bio Copolyester (BBC) is a random copolymer of Adipic acid, Terepthalic acid, and 1,4-butanediol. (Buchanan et al, 1995) The repeat structure of BBC is shown below. 0 {—w (CH2),—O —(cH,),—— iHo— (CH,I,— °—°—‘©—l°|+... Figure 3.3: Structure of Eastar Bio Copolyester In the structure, m and n are dependent on degree of polymerization. Weight average is 67,000, while number average molecular weight is 20,000. (Wright, 1998) Melt temperature at peak melt is approximately 110°C. Properties of Eastar are listed in Table 28 3.2. Biodegradability testing of Eastar as measured by ASTM D-5338 standards indicates the polymer reaches full degradation within 60—90 days. Properties ASTM Method Value Units Film Thickness D 374 1.0 mils Density 1 .27 g/cm3 Crystalline T. 110 °C T, D 3418 -33 °C Melt Index D 1238 at 190°C 25-31 g/10 min at 125°C 0.7-2.6 g/ 10min Tensile Strength at Break D 882 18.9 MPa Elongation at Break D 882 425 % Tensile Modulus of Elasticity D 882 86 MPa Dart Impact D 1709 100 g Table 3.2: Selected Properties of Eastar Bio Copolyester The reaction proceeds via a peroxide catalyzed attack at the carbon nearest the ester bond in the copolyester. This attack creates a radical that can then attack at double bond in the M80. The result of this reaction is the grafted copolyester structure shown in Figure 3.4. H-[0*(C|-5).-0—9- (cuegHo—tcrrh—o—E —©—C},53H Dicumyl Peroxide o o o ”{o—(cucho—g4crwiHo—(crpio—Ji—©—ii /CI-i=c{ic CH3(CH2)4CH2c%—§ H(CH,),cooa d o b CH3(CH2)4CH H H(CH 2),cooR M44 0 o o H—[o—(CI-igci-I—o—iHm-prfllm—[vo—(crpro—ii—Q—b Figure 3.4: Grafting of maleated soybean oil onto Eastar Bio Copolyester 30 Several parameters were examined to determine the most efficient procedure for performing this process. The first procedure involved simply mixing the M80 with the catalyst, DCP and the copolyester and then feeding this mix to the extruder. This process was unsuccessful because the high viscosity of the M80 inhibited the free flow of the mixture. The next process involved mixing the copolyester with the DCP and then adding the M80 into the extruder through a liquid feed port. This process was partly successful but was not totally successful because it was difficult to pump the viscous MSO with the available peristaltic pump. The final refinement to this process involved heating the M80 and then pumping the less viscous warm MSO into the extruder liquid feed port. Two screw configurations were tested to determine which configuration would be most effective for this process. The first configuration tested was the configuration used in the starch plasticization process (see Chapter 4). The process did not successfully operate in this configuration. A large zone of kneading blocks nearest the feed throat prevented forward conveying of the M80. The second configuration eliminated much of this kneading zone and replaced those kneading elements with conveying elements as shown in Figure 3.5. 31 Discharge 020/020 (4) 042/042 (3) KB 405/014 (1) KB 405/042 (1) KB 405/014 (3) 020/020 (2) 028/028 (2) 0421042 (4) KB 905/028 (1) KB 405/0l4 (3) 020/020 (3) 028/028 (2) 042/042 (4) Feed Throat Figure 3.5: Screw configuration for reactive extrusion The numbers associated with the conveying elements (clear) in Figure 3.5 a/b (c) are a code in which a = screw pitch (mm) b = element length (mm) c = number of elements. The code for the kneading elements (gray) is aab/bcc where aa = paddle offset in degrees b/b = first b is number of paddles on element, second b =0 cc = element length in mm 32 A final problem that needed to be solved to ensure the success of this process was an equipment problem associated with the extruder. MSO served as a lubricant for the extruder and the extruder would shut down automatically due to extruder under-load. To solve this problem, the temperature profile chosen was lower than the normal processing temperatures for this polymer system, thereby increasing viscosity of the melt and load on the extruder eliminating the extruder under-load problem. The profile used in shown below. Zones 1 2 3 4 5 6 Feed Throat Vent Temperature (°C) 70 175 175 175 120 115 Table 3.3: Temperature profile for reactive extrusion The entire process was conducted under a blanket of nitrogen gas to prevent any unwanted side reactions. Extruder RPM was 100. Five materials were prepared reactively for further analysis. All mixtures contained 0.3% Dicumyl peroxide. The first material was simply the copolyester and DCP. The second, third and fourth mixes consisted of the copolyester, DCP and 5%, 10%, and 20% MSO. The fifth mix consisted of copolyester and 10% M80 without DCP. These mixes were extruded as described above, run through a water bath and pelletized. The ensuing material was then dried in a vacuum oven for 15 hours and further blended and characterized. 33 Characterization of maleated soybean oil and the grafted copolyester products. MSO characterization Characterization of the MSO included measurement of acid numbers, and Iodine values for reaction performed. To measure acid number, small amount of M80 was dissolved in l-octanol. This solution was titrated against a standard KOH in octanol solution to a phenopthalein endpoint. The average acid number for the MSO was 82. Based on this acid number the number of moles of maleic anhydride added per mole of soybean oil was determined to be 1.14. The Iodine value is a measure of unsaturation of fats and oils. To measure Iodine value, the sample material is first dissolved in chloroform. Wij’s solution (Iodine monochloride in acetic acid) is added to the flask and incubated for 30 minutes. After incubation, potassium iodide solution is added to the mixture. A nominal amount of water is then added to the solution and the contents are titrated with sodium thiosulfate solution until the yellow color almost disappears. At this point, a small amount of starch indicator solution is added creating a blue mixture. This final mix is titrated to a clear endpoint. The amount of thiosulfate solution can then be used to determine the Iodine number. Iodine numbers were approximately 100 for the M80. This Iodine value corresponds to about 3.3 double bonds/mole of soybean oil. Further characterization of this product should include GC-MS, and detailed NMR and FTIR spectra of the product to verify the actual structure of the product. 34 Extraction Approximately 5 g of each material were extracted to determine the grafting efficiency. To do this each material was refluxed in xylene for 4-5 hours. After refluxing the solutions were precipitated into methanol to eliminate any free maleic anhydride. The resulting solids were then washed with warm chloroform to remove any fiee MSO. The final product was titrated against ethanolic KOH to determine the amount of grafted MSO. Given the amount of M80 added to the extruder, grafting efficiency was determined by dividing the weight percent grafted by the weight percent added. Grafiing efficiency for selected materials is shown in Table 3.4. Weight % MSO added 5% 10% 10%(no DCP) 20% Grafting Efficiency 58% 38% 23% 30% Table 3.4: Grafting efficiency of reactive extrusion Based on the results in Table 3.4 grafting efficiency is a nearly a linear function of theoretical MSO concentration. The results from the 10% M80 blend without DCP show that DCP increases grafting efficiency but is not necessary to achieve grafting of the M80 to the copolyester. The 10% material was chosen for blending with starch for further study because of ease of preparation and middle range grafiing efficiency. Further understanding of this grafting reaction and verification of the structure of the actual product is needed. Studies of the effects of DCP concentration on efficiency need to be performed. Quantitative solution FTIR of the extracted materials could be performed to verify the grafting efficiency. Finally, GC MS, NMR and FTIR spectra of the product should be obtained to verify the product structure. 35 Intrinsic Viscosity The intrinsic viscosity measurements were carried out at 30°C in a constant temperature bath using a Ubbelohde viscometer. The pure and grafted samples were dissolved in methylene chloride and diluted to the required concentrations. These intrinsic viscosity measurements were performed for four materials: BBC, EBC-g- MSOIOB, EBC-g-MSOIOG, BBC-DCP. The intrinsic viscosity was determined by using the Kraemer equation: -1 .1. [ml—c1411] o Plots of the right had side of the equation versus viscosity yield a y intercept equal to the intrinsic viscosity. These plots were constructed for the four test materials and are shown in Figure 3.4. It can be seen that the intrinsic viscosity goes up for the MSO grafted material (EBC-g-MSOIOG) when compared to the 10% M80 blended material (EBC-g-MSOIOG) or the copolyester itself (EBC). This increase in intrinsic viscosity and in turn molecular weight may be attributed in part to the presence of Dicumyl peroxide as evidenced by the great increase in intrinsic viscosity for the BBC-DCP material. These increases in intrinsic viscosity can be related directly to increases in molecular weight. The materials, then, in order of increasing molecular weight would be: EBC _ I 5 it E! Electrons: ' ".-’ EEzTF—‘IEHIM TIF Figure 5.3: EBC:TPS; 70:30 blend, note large TPS particles (lighter color) laying on the surface. 47 s. :3 s3 .. [:5 Electro? Figure 5.4: EBC-g-MSOIOGzTPS; 70:30 blend, note thermoplastic starch particles are partially integrated into the grafted polyester matrix 48 1* . . v , >- Y I .— . > a.‘ V b. I UU t' I” — ' h i . u. z - - EEiTF4UGS TIF Figure 5.5: EBC:EBC-g-MSOIOG:TPS; 65:5:30 blend, note no thermoplastic starch particles are visible. 49 Close up photographs of each of these three 70:30 blends reveals that in the case of the EBC:TPS blend, Figure 5.6, there is no adhesion between the starch particles and the polyester. In the EBC-g-MSOIOG:TPS blend, Figure 5.7, though, adhesion is seen between the thermoplastic starch particles and the grafted copolyester. In contrast, in the EBC:EBC-g-MSOIOG:TPS blend, Figure 5.8 the thermoplastic starch particles are barely distinguishable from the polyester matrix, indicating a higher level of adhesion and particle integration. This morphology being exhibited by the last blend indicates that compatibilization is occurring for this mixture. Physical property testing to be discussed later also supports this hypothesis. Figure 5.6: EBC:TPS;70:30 blend, note no adhesion of thermoplastic starch particles to matrix 50 EE’TF'Z‘UIC‘ TlF Figure 5.7: EBC-g-MSOloG:TPS;70:30 blend, note some adhesion of thermoplastic starch particles to the matrix Figure 5.8: EBC:EBC-g-MSOIOG:TPS; 65:5:30 blend, note thermoplastic starch particles are no longer readily distinguishable from the polyester matrix 52 Tensile Testing Tensile tests were conducted using a UTS testing machine to determine stress, strain, and Young’s Modulus for all samples. The blends lacking starch were tested at 1 in/min or 0.5 in/min for the modulus measurement. These samples exhibited high elongation, therefore, stress at break was not measured. Stress at high elongation was measured for comparison purposes. To prevent bar slippage during the test, the UTS clamps were tightened frequently. The starch containing blends were generally tested at 2 in/min for breaking and l or 0.5 in/min for modulus measurement. Conditioning of all samples was performed at 50% relative humidity and 75°F for at least 40 hrs prior to testing. At least six specimens of each material were tested The first group of materials tested were the polyesters and modified polyesters: EBC, EBC-g-MSOSG, BBC-g- MSOlOG, EBC-g-MSOIOB, and EBC:DCP. The second set of materials tested consisted of nine blends and BBC. The nine blends are: 70:30 - EBC:TPS and EBC-g-MSOIOGzTPS 60:40 - EBC:TPS and EBC-g-MSOIOGzTPS 50:50 - EBC:TPS and BBC-g-MSOIOGzTPS 65:5:30 - EBC:EBC-g-MSOIOGzTPS 48:32:20 - EBC:TPS:Fiber and EBC-g-MSOIOG:TPS:Fiber Of the materials in the first set the EBC:DCP exhibited the highest tensile strength at break (Figure 4.9). This is not surprising since this material also exhibited the highest intrinsic viscosity and therefore the highest molecular weight of this group of materials. This material was also the only material to break during the tensile test. The other materials stretched to high elongation without breaking. This high elongation caused 53 problems since the test bars tended to slip in the clamps. This slippage was fixed by tightening the clamps in mid-test. This tightening caused the spikes seen in Figure 5.9. 3500 3000 2500 -- g 5 2000 ii I, J & 'lllflm" v ' l” 1 g . . ‘ ll fir‘ I (.5 1500-. |,‘“ g! . I gig: «- 1000 __ / —EBC' WWI " EBC:DCP 500 ' __EBC-g-MSOSG ' _EBC-g-MSO10 ' ' _EBC-g-MSO10 0 . . 0 100 200 300 400 500 600 700 800 Stratum Figure 5.9: Tensile test results for modified copolyesters The fact that physical properties correlate well with intrinsic viscosity measurements indicates that the grafting reaction was successful. This additional strength could also be a function though of crosslinking caused by the presence of Dicumyl peroxide. Intrinsic viscosity data support this theory. Further characterization of this group of materials by NMR and FTIR techniques in attempts to elucidate the structures will provide great help in determining the actual graft structure and provide another check for the grafting efficiency. 54 Strength at Break or High elongation was examined for 70:30, 60:40, and the 50:50 EBC:TPS, and EBC-g-MSO:TPS blends. As seen in Figure 5.10, the blends containing EBC-g-MSOIOG outperforms the blends containing BBC in the same ratios All blends broke except for the EBC:EBC-g-MSO:TPS; 65:5:30 blend and BBC itself, therefore ultimate tensile strengths of these materials can’t be compared to the other materials. Since these materials are already stronger than there counterparts are at break there break strength is also stronger than the other materials break strengths. Compatibilization of the EBC-g-MSO:TPS blends occur but are not as effective as the EBC:EBC-g-MSOzTPS blend. This indicates that a small amount of the grafled copolyester is a better compatibilizer than large amounts. This small amount of modified copolyester is most likely at the interface between the two phases serving in the same fashion as a graft copolymer does to compatibilize this system. 55 3.3—m 3.82.5 he i.e.... «a inseam 9:25;. afim charm a p 093 H 00—. , . 82-3mm , . 8.8 l ovnom omm _ omnmmnmv H wah_oo.o.>m¥§mo>_oa anon u. wahOo F092- 0mm v.35 yo: Be .556... omumumo u..a.n_._.“Ow—2.90mw”0mm atom H. .F canon wmhuomm W :35 Ho: Eu 3538 0mm o com S n... a u 83 .m. u. e 1.. m. m 89 n. my. 3 ooom comm 56 Error bars indicate one standard deviation. This fact coupled with the ESEM pictures would indicates compatibilization is occurring with the addition of EBC-g-MSOIOG. The observation also that the EBC:EBC-g-MSOIOGzTPS; 65:5;30 blend was even stronger than the EBC-g-MSOzTPS; 70:30 blend is also supported again by the ESEM pictures. The only situation in which the BBC outperforms the EBC-g-MSOIOG is when fiber is added. The tensile strength at break as a function of concentration is shown in Figure 5.11. Increasing levels of starch, though helpful for biodegradability are detrimental to the material properties. fl 1500 2 + EBC:TPS Blend g + EBC-g-MSO1OGITPS Blend I: 1000 «— ‘5', f." g (D 2 E; a"; 500 . 2 '7: C ,9 o 0% 20% 40% 60% 80% 100% BBC Concentration (%) Figure 5.11 Tensile testing of modified polyester: thermoplastic starch blends 57 Biodegradability Testing Under Composting Conditions Four materials were tested for biodegradability under composting conditions. These materials were EBC, EBC-g-MSOIOG, EBC-g-MSOIOG:TPS;70:30, and, EBC-g- MSOlOG:TPS:Fiber; 48:32:20. Carbon content of each sample was determined using a Perkine Elmer CHN analyzer. Thirty grams dry weight sample was added to 180g dry weight compost. Water was then added to all samples as needed to reach the centrifuge moisture equivalent described in ASTM D-424-79. These mixtures were then placed into two liter glass bottles for testing. Each sample was tested in triplicate and subjected to the conditions described below. The duration of the test was 65 days. Test Apparatus The experimental test apparatus (Figure 5.12) employed a Siemens Ultramat 22P carbon dioxide analyzer (Siemens AG, Germany) to measure CO2 production from the samples. The analyzer was calibrated using certified 3.83% CO2 from AGA Industries. Airflow was regulated by Mott Metallurgical (Farmington, CT) flow restrictors to 60 mL/min. Test Conditions Biodegradation of polymer blends under composting conditions was measured by the ASTM D-5338-92 method. This method exposes samples to a mature compost in a controlled temperature chamber. The temperature set points for this experiment were 35° C for the first 24 hours and 58° for until test conclusion. This test method measures the percent of test material converted to CO2 by microbial activity by determining the difference in CO2 production between the test vessels and a set of vessels containing no test material (blanks). 58 (45 sample vessels . valve In parallel) manifold carbon dioxide analyzer g I©J flow ‘ reetrietor . ....... i —-I-— E 5 l humidifier sampleveeee i i i v i controlled temperature incubator [3 =H computer IIO board Figure 5.12: Biodegradability Testing Apparatus The results of the biodegradability testing are displayed in Figure 5.13. ASTM standards dictate that a material reach 80% biodegradability in approximately 180 days. Conversion to CO2 is increased 10% compared to BBC alone, by grafting MSO to BBC. Blending EBC-g-MSOIOG with thermoplastic starch at 30% increases conversion by 25%. The addition of fiber increases conversion by 48%.. 59 100 _-___.,_. __- . ..-._._..---.-____ WW- ms..- _ -. '*_."*'Ea'c“"'” 9° _._ EBC-g-M some 80 + EBC-g-M SO1OG:T PS. 70:30 1 —e— EBC-g-M SOtOG:TPS:Kayocel; 48:32:20 2 8 70 '* —Kraft Paper ! 0 — ————_— —--~~~--— --— ~-——~~—- E 3 60 g '8 . l .5 . , 50 - u: r: g 40 I o 3: 30 20 10 .- / 0 ,. o 10 20 30 40 so 60 7o Tlme(Days) Figure 5.13: Biodegradability Testing Results The first derivative of each biodegradation curve was determined and using a polynomial based on this curve predictions were made concerning the biodegradability of each material. Eighty percent conversion of sample to carbon dioxide was predicted to occur at 330 days for BBC, 215 days for EBC-g-MSOIOG, 120 days for EBC-g- MSOlOG:TPS; 70:30, and 78 days for EBC-g-MSOIOG:TPS:Fiber; 48:32:20. The polyester and grafted copolyester do not meet international standards, but the addition of thermoplastic starch and fiber improve biodegradability of these blends to the point where international standards are met. 60 CHAPTER 6 Preliminary Process Design Process Description A flow diagram (Figure 6.1) and material and energy balances (Table 6.1) of a reactive extrusion and compounding process for two difl‘erent resins was determined from experimental data. The basis for the design was: - 5 million pound per year of total resin. - 2.5 million pound per year of resin containing 70% EBC-g-MSOIOG and 30% TPS - 2.5 million pound per year of resin containing 48% EBC-g-MSOIOG, 32% TPS and 20% Fiber. -Stream factor of 0.9 This plant will contain two extrusion lines, E1 andE2. The two final products will be produced on line E2. Thermoplastic starch and EBC-g-MSOIOG will also be produced on line, E1. Hourly production rates for this design were based on equal usage of the extruder for each product produced in it- production time for each product was 3942 hours. The EBC-g-MSO line will be maintained under a nitrogen blanket to prevent unwanted side reactions during extrusion. Storage will be provided for two weeks inventory of all required materials and products produced, Sl-SIO. Kinetic studies of the grafiing reaction have not been performed, so it was assumed that extruder residence time and temperature affects scaled up with equal effect. All reactants will be stored in carbon steel storage tanks. Pelletized and powdered materials will be conveyed to gravimetric feeders for feeding into the extruder by 61 EEqu 33h 33:...— 36 953..— pneumatic conveying systems. Feeders have been omitted from the process flow diagram for simplicity. Flow of liquid reagents, streams 2 and 6, pumped to the extruders will be controlled by feedback loops containing the appropriate pump-PU] and PU2, and a liquid flow meter. maleated soybean oil, stream 2, will be heated to 70°C to increase the ease of pumping. Polycaprolactone, stream 6, and starch, stream 7 will be mixed in a tumble mixer prior to addition to the extruder in making thermoplastic starch. In all production lines except for the thermoplastic starch line, extruded material will be cooled by passing through a water bath, BI and B2. Cooled material will be dried using a twelve inch air wipe, and then pelletized, Pl-PZ. Thermoplastic starch will be passed through a stream of chilled air, prior to pelletizing. Pelletized material may need firrther drying, however estimates concerning the extent of drying have not yet been made. Several assumptions were made in order to close the energy balance. Enthalpy of the materials was considered constant. The mechanical energy exerted on the materials in the extruder by the screws was de-coupled from the heating of the materials. In the balances, this energy was entirely dissipated by water flow through the extruders. Material heating efliciency was assumed 100%. Heat of reaction for the grafting reaction, since it is unknown, was neglected. Extruder Sizing Extruders used in this process were all twin screw co-rotating intermeshing machines. The extruder used for grafting and thermoplastic starch production, E1, was a 70mm-screw diameter machine while the machine used for compounding, E2, was also a 70mm machine. These machine sizes were based on manufacturer recommendations 63 (Krupp Werner & Pfleiderer Corporation). Machine size can be determined, however using appropriate scale up equations. capacity of b = capacity of a x scale up factor scale up factor = (screw d of b3/screw d of a3) The maximum throughput for the grafting reaction is not known, however assuming it to be 50 lb/hr in a 30mm extruder, one could calculate the size of extruder necessary for the needed capacity of 748 lb/hr. Using the above equations, the scale factor would be 14.96, while extruder diameter would be 74 mm. This calculation provides good agreement with the 70mm recommendation by the manufacturer. To truly determine appropriate extruder size though, studies must be done in which the absolute maximum capacity of the 30mm is determined. Studies conceming kinetics of the grafting reaction, and residence time distribution of the material in the extruder also need to be done. Completion of these studies would enable an engineer to do pilot plant studies of this process and then further refining of equipment choice for a large scale plant. 64 3.9 n. 58 33. 3.3» 389. 4.9.2. 4.3 .38 02.3: Aisha 32a 3.55 o. a how 2 m um 32 tom 2 2 mam 22393355 a; med to no no to to to no $835530 mm mm mm mm m: 2 mm mm ..... .3. 693 23888..» Gas: 6mm. 30% on: on. __o 52:8 .6 52.3» 58:88 208205820“. :25 ES .2820 8266mm 825.03 333: 3:56 38.22 2m 58m 5398538 a a o m 4 m an m . .oz 58am 00:23 >a_a£cm 2: $8 Q: ~68 «4.8 32 RN 3! 3mm .88 ad ad ad o.o ad ad 3o 3 o.o in; o.o ad od 9o 3. ed 8d ad 3. Sun: song-xaahoaseomu o.o o.o ad ad ad ad cod o.o o.o encumato»??omm Qo o.o od 38 «.08 ad 8d ad ad 32-...03 ad ad ad ad od 0... 8d ad ad :28» 83.852: ad ad ad ad ad ad 8d ed o.o 3.8.3. 2: ad ad ad ad ad 8d ad ad 888.2833 ed 2. a: 2. od o.o 8d ad ad .2850 o.o 3% ad ad ad o.o cod .3 o.o seam Eco S. o... S. o.o ad ad RN o.o o.o 3x58 o.o ad ad ad ed 33 cod 3: 9o .6 5238 3.2.3: od o.o ad ad ad ad cod ed 3% 53:88 2m 58m 62): 6mm. 30% on: 02 __o 52:8 .6 533m 5828.8 805305828 :23 58 .2820 3.25.08 owsaomm 8.83.2 328.5 38...: 3 firm. resecocoasoo m a m m 4 m «a a F .oz 525 £3. £3: cote—6o:— EE :8 r: :2 a: :8 co. oazadmm :3. :3. 880mm 02 82.0.8”.— ....2. :2. 8983 cam 2.7.3— ozaeuauoaefi he notes—3...— ..8 82.2.5 Rheum 33 3.832 3. 035. 65 0.088 3 .8 0.88 8.88 .83. 8.8.... 8.88 0.88... ESE. 8.. 3.2.5. 08.8. 8.2 m.- m.- «.8 «.8 v.8 .8: 355.3355 08... 3... .3 m... ~m.o 80 8... 8... Coon-85580 at 8 m~ m~ m~ m~ m~ 8. .080. 238850: on: $8 02 0.: 8:. .280 ems-90$ so: .8: 00.20-08 00.28.08. 0.: 8885.2: 28.88880 8 t 2 2 ~. 2 .oz 588 00:23 3.053 :5 8.8. 8.8.. 38 m8. m.~o~ ~88 ~88 .80: o... o... o... o... o... o... c... o... 5...... o... o... o... o... o... O... 8... .8 8.8.8 ”.8oaaxnmatoms-o-0mm I; o o o... o o o o o... o... o... 8.2.6.:602-008 o o o o 83. v .8 o o o... o... 8... 002.003 o o o o o... o o m 8. m.~o~ ~88 ~88 5:80. 2.8.852: o o w 8. o... o o o o o... o... o... .3093. o o o o o... o o o o o... o... o... .eoBo.E..8...oa o o o o o... o o o o o... o... o... .2850 o o o o o... o o o o o... o... o... 5.0.0 :30 o o o o o... o o o o o... o... o... .9830 o o o o 3. o o o o o... o... o... ..o 53.30 8.8.»... o o o o o... o o o o o... o... o... 3.8880 so .800 8: $8 0.: o... 8:. 8.80 00.28.08. 38 .2... 00.28-03 00.2008. 8: 88.882: ..§_..coc880 m. t .2 2 ~. 3 .oz 588 ”~wa 3303 :23ng .2... 8... m: :8 m: :8 o. ems-8-0mm :8 ..8. 8888 8. 82-0-08. :8 .8. 8o.o8.~ eam Ceeovnemmom «Suzanne—.05 .3 5.2033...— ..e._ 30:20: 3.2—m new .0283). 3. 03:. 66 88.8.0 0.080.. 0880.0 0.00% .. 8600.0 0.00%.. 0.k~0~0 0.080.. 0.00. 0.00.0. 06.... 00.2.00 00 0. 00 0. 00 0. 00 0. 3~.- 300.00.22.00 . . . . . . . . 3.00.0 .000300580 00 0. 00 0. 00 0. 00 0. .- 0~ -- .0000. 232.05.. E000. .0260 .00.“. $00 53.0 0.0000 00. $~0 02,0 02,0 02.0 030 0.50 02.0 502, 0......0 .802. 02.5 00.2.0000 $00 50.058980 00 .0 00 00 00 80 ~0 .0 .~ .oz .530 800.3 00.055 .8080 .8000 .880 .880 .803 .890 .890 .890 ..~.8 .50. .880 .8080 .890 .880 .880 .880 .880 .880 0.0 in; 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 .8... 888.8 .=o&£.00..00:-0.00m 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 086.60.620.80 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 002-0000 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0280 003.025... 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.808. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8203200903.. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 .2930 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0280 500 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 .3800 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 .6 8.0000 022...... 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 38.08.00 0.0 53m .030. .026. .00.“. 0.8 53.0 .0090 00. $~0 02,0 02,0 02.0 0.50 0.50 02.0 50.5 8.....0 52$ 0......0 00.2.0000 $0.. 0.000.082.500 00 .0 00 00 ..0 00 ~0 .0 .~ .02 502.0 .00. e3: 80802. 5...“. :8 00. :8 00.. :08 00. 0020-000 0...... 5.0. 000003 8. 002.0000 :0. :0. 000.0088 00.0 3:00. 0500: 0300009005 .0 5302.0...— 08 80:20: .8025 95 3.502 —.e 030,—. 67 - cage 088. 388 388 39: .38 «.882 2va 2.5983358 - 9. 2 9. R 9 R 9 R 3555355 - F F F P P P P F £935.58 - mu 2 mm 2 mm a ma 9 692 osfioefi 60.. so .325 28 so: 2833 35:35 60; so: so; 60: so: 1592.. 28 3382 2525828 B 2 N. m. 2 m. u. : .oz 588 85.8 326 $8 88 23$ 22$ 2% SE 2%: v.8: 18» o 38 88 $3. 22$ 2% Sam v.8: v.8: .325 o 3 9o 3 9o 3 3 od 2. 8&3. "gugfim‘fig o od 3 3 od o.o 0.0 9o 3 8623269868 0 od 3 2. od 8 o.o 3 od 82.3mm o 8 ed 2 8 2. od 2. 2o :25 8832.2» o 3 3 ad 2. od 3 2. ed 2895. o o.o 3 ad ad Qo o.o cc 8 8239185“. o 3 2. 9o 8 8 od 3 3 3.85 o 3 3 3 od 2. 8 ad ad $.56 E8 0 9o 3 o.o od 3 od 3 ed 138 o 8 od od 2. 8 od 2. o... =0 52:8 383.. o 3 9o 3 od 8 od 0.0 3 3% 2m Sam 60.. so .222 28 co: 2 3.2,) 326mm 60: so: so; 60: so; 3532.. 28 32mg... 2.5.220888 E 2 2 m. 2 m. N. 2 .oz 585 g 282 52?: B: 2.2 Exam $22.8 8. 82.3% 2% 22 8802 8. 82.0.08 2.? 52 o8.8m.~ 3m 352 223. «33.29.55 3 522.25 .2 .853. 998 2a 3332 S 03.2.. 68 CHAPTER 7 Economic Analysis Process Economics Table 7.] lists the assumptions that were used in the economic evaluation of the proposed plant. Several of the assumptions made for non-manufacturing labor and utilities can from Peters and Timmerhaus (Peters and Timmerhaus, 1991). The plant will produce 5 million total pounds of resin per year. One half of that resin will be 70% BBC- g-MSO 10G: 30% TPS, while the other half of that resin will be 48% EBC-g-MSO: 32% TPS: 20% Fiber. The plant will operate for 10 years at which time salvage value will be considered 10% of the initial fixed capital investment. The depreciation will be straight- line over the lO-year period for 90% of the fixed capital investment. Labor will be paid $15 per hour. Three operators will be required per shift. Supervision, maintenance, local taxes and insurance will be 8%, 7% and 2% of the fixed capital investment, respectively. Future R & D, marketing, and general sales and administration expenses will be 5%, 5%, and 4% of the total manufacturing cost respectively. Utility costs will be purchased ate $0.07 per KWl-l for electrical power, $0.15 per 1000 gallons industrial cold water, $0.25 per 1000 gallons chilled water, and $0.015 per gallon for sewer. Purchased chemical will cost $2.15 per lb. for Eastar Bio Copolyester, $0.60 per lb. for Maleated soybean oil, $3.00 per lb. for Dicumyl Peroxide, $0.145 per lb. for starch, $1.00 per lb. for glycerol, $1.25 per lb. for Polycaprolactone, and $0.10 per 69 lb. for fiber. Chemical costs were obtained directly from suppliers when possible and estimated for maleated soybean oil using estimates based on Seybold (Seybold, 1997). Plant Capacity Factor Resin 1: 70% EBC-g-MSO 10G 30% Thermoplastic Starch Resin 2: 48% EBC-g-MSO 106 32% Thermoplastic Starch 20% Fiber Depreciation Total Federal Taxes Working Capital Plant Lifetime Base Rate Operating Labor Supervision Maintenance Local Taxes and Insurance R & D Marketing General, Sales, Admin Expense Utility Costs Purchased Electric Power Industrial Cold Water Chilled Water Sewer Purchased Chemicals Eastar Bio Copolyester Maleated Soybean Oil Catalyst Com Starch Glycerol Polycaprolactone Fiber 7884 hours per year 2,500,000 lb./year 2,500,000 lb./year 9 percent, straight line 34 percent 15 percent of Total Capital 10 years $15 per hour 8 percent fixed capital investment 7 percent fixed capital investment 2 percent fixed capital investment 5 percent total manufacturing costs 5 percent total manufacturing costs 4 percent total manufacturing costs 0.07 $ per KWH 0.15 $ per 1000 gallons 0.25 $ per 1000 gallons 0.015 $ per gallon 2.15 $ per pound 0.60 $ per pound 3.00 $ per pound 0.15 $ per pound 1.00 $ per pound 1.25 $ per pound 0.10 $ per pound Table 7.1: Assumptions for Economic Evaluation 70 The necessary equipment is described in Table 7.2. Equipment costs were taken from Peters and Timmerhaus (Peters and Timmerhaus, 1991), when possible. Plastics processing equipment costs were obtained directly from vendors. Extruder cost estimates specifically were obtained from Krupp Werner & Pfleiderer Corporation and are quotes based on purchase of ZSK series extruders. Equipment costs when necessary were adjusted to 1998 dollars using a Marshall and Swift Index (Chemical Engineering, May l 998). Item Description Size Purchase Cost(1998 5) E1 Twin Screw Extruder 70mm S 410,000 E2 Twin Screw Extruder 70mm S 410,000 PU 1 Pump 9.25 gthr S 2,746 PU2 Pump 7.13ga1/hr S 2,347 P1 Pelletizer S 5,258 P2 Pelletizer S 5,258 $1 Storage Tank 9618 gallons S 23,473 Carbon Steel 82 Storage Tank 1399 gallons S 8,216 Carbon Steel 83 Storage Tank 3566 gallons S 11,854 Carbon Steel S4 Storage Tank 261 gallons S 2,347 Carbon Steel SS Storage Tank 1077 gallons S 5,868 Carbon Steel 86 Storage Tank 3586 gallons S 11,854 Carbon Steel 87 Storage Tank 4890 gallons S 17,605 Carbon Steel S8 Storage Tank 10945 gallons S 25,821 Carbon Steel S9 Storage Tank 8671 gallons S 22,300 Carbon Steel $10 Storage Tank 9549 gallons S 23,473 Carbon Steel Bl-BZ Cooling Baths 2011 S 2,347 Gas blanketing system - S 3,228 Feeding Systgms S 58,684 Air Wipe 2- 12 inch lip units S 4,695 Total Purchased Equipment Cost $1,077,375 Table 7.2: Equipment List for Resin Production 71 The total capital investment for construction of the 5 million pound per year resin facility (2.5 million pound per year of two different resins) is shown in Table 7.3. All costs are based on industrial averages for new plant fabrication from the work of Peters and Timmerhaus. The total capital investment is calculated using the equipment cost as a basis and industrial averages for needed accessories. The actual costs and percentages of purchased equipment cost each is shown below in 7.3. Plant Capacity Factor 7884 hours per year Resin 1: 2,500,000 lb./year 70% EBC-g-MSO 10G 30% Thermoplastic Starch Resin 2: 2,500,000 1b./year 48% EBC-g-MSO 10G 32% Thermoplastic Starch 20% Fiber Direct Costs As a % of Purchased Equipment Cost Purchased Equipment 100 S 1,077,375 Purchase Equipment Installation 39 S 484,819 Instrumentation and Controls 13 S 96,964 Piping (Installed) 31 S 3.12.438 Electrical (Installed) 10 $ 172,380 Buildings 29 S 107,737 Yard Improvements 10 S 269,344 Service Facilities 55 S 140,059 Land (if purchase is required) 6 S 64,642 Total Direct Plant Cost S 2,725,758 Indirect Costs Engineering and Supervision 32 S 355,534 Construction Expenses 34 S 420,176 Total Direct and Indirect Costs S 3,501,468 Contractors Fee 5 S 1 83,1 54 Contingency 10 S 366,307 Fixed Capital Investment S 4,050,929 Working Capital S 732,615 Total Capital Investment S 4,783,543 Table 7.3 Capital Investment Breakdown. 72 Table 7.4 shows the manufacturing cost breakdown for the proposed production facility. The variable and material costs determined as part of the material and energy balances are included in this table. Variable Costs Industrial Water Chilled Water Sewer Electricity Material Cost Eastar Bio Copolyester Maleated Soybean Oil Catalyst Corn Starch Glycerol Polycaprolactone Kayocell Fixed Costs Labor Supervision Maintenance Depreciation Local Taxes and Insurance Total Fixed Costs Total Manufacturing Costs R&D Marketing General, Sales, Administration Expense Total Operating Costs S/Unit Units Cost S/year 0.15 1000 gallon S 291 0.25 1000 gallon S 3013 0.015 gallon S 29,064 0.07 KWH S 158,41 Total Variable Costs $191,009 2.15 lbs $5,708,250 0.6 lbs S 177,000 3 lbs $ 26,550 0.145 lbs S 171,238 1.00 lbs S 295,238 1.25 lbs S 92,262 0.1 lbs $ 50,000 Total Material Costs $6,520,538 3 operators @SlSlhr S 236,520 8% of PC] $ 324,074 7% of FCI S 283,565 10% Salvage, 9% FCI S 364,584 2% F CI 3 81 .019 S 1 ,289,761 S 8,001,308 5% of TMC S 400,065 5% of TMC $ 400,065 4% of TMC $ 320,052 5 9,121 .491 Table 7.4 Manufacturing cost breakdown for 5 million pound per year plant Fixed costs consist of three operators per shift at $15/hour, and Supervision, maintenance and local taxes and insurance at 8%, 7%, 10% and 2% of the fixed capital investment. 73 Other costs include R&D, Marketing, and General, Sales and Administration expenses at 5%, 5%, and 4% of the total manufacturing cost respectively. Examination of the table reveals that manufacturing cost is most sensitive to the cost of raw materials. Energy and labor costs are insignificant compared to the cost of raw materials. There are several methods for determining the selling price for a product. One method, Return on Investment-ROI can be calculated by adding the total manufacturing cost to the required return and dividing this number by the total production. For the situation for which minimum return on investment is 15%, the selling prices are $2.01/lb. for the resin without fiber and 1.47/lb. for the resin with fiber. Another method for determining selling price for a product is discounted cash flow. This method takes into account the time value of money by establishing a rate of return that can be applied to the yearly cash flow so that the original investment is reduced to zero during the project life. This rate of return is equivalent to the maximum interest rate, after taxes, at which money could be borrowed to finance the project over the project’s life under conditions where the net cash flow would be just sufficient to pay the principal and interest accumulated on the outstanding principal. At the end of 11 years the cash flow compounded based on end of year income will be: F(1+i)“"+F(1+i)"'2+.....F(1+i) = S (1) where F = yearly cash flow i = interest rate S = future worth of the project The future worth of the initial investment would be: 74 S’ = P(1+i)" (2) where P = principal investment Given an interest rate and principle the equation: S-S’ = 0 (3) can be solved by a trial and error method. Other terms that can be included in equation (3) include the working capital, and the project salvage value. These terms were not included in the simple economic analysis performed below. The selling price can be varied until a solution to this equation is found. Using this method for the 5 million pound per year plant yields the selling prices at various rates as shown in Table 7.5. The differences in selling price between the two resins is based on different material costs. Rate of Return Selling Price ($llb.): Selling Price ($llb.): 70% EBC-g—MSO 10G 48% EBC—g—MSO 30% TPS 32% TPS 20% Fiber 10% S 2.23 S 1.63 20% S 2.34 S 1.70 25% S 2.40 S 1.74 Table 7.5: Selling price based on discounted cash flow Assuming an average selling price for both products of $2.05/lb., the net present worth, the right hand side of equation (3) was calculated for various interest rates. Figure 7.1 shows several discounted rates and the net present value for the proposed process. The discounted rate, the internal rate of return-IRR, at which the net present value equals zero is 23%. Using this method various alternative manufacturing methods could be compared. The ‘best’ manufacturing method would have the highest IR. 75 O D g $10,000 0 2 $5,000 ~ g *5 9" ‘ ' Q) g $(5,000) - 5.; $00,000) 2 0% 10% 20% 30% Discount Rate Figure 7.1: Determination of Internal Rate of Return by a discounted cash flow for the proposed process Sensitivity analysis A sensitivity analysis was performed for the proposed process. In this analysis the effect of fluctuations of Eastar Bio Copolyester cost, glycerol cost and capital investment on the two product average selling price were examined. The assumptions made for this analysis were: - Production capacity of 2.5 million pounds per year of each resin - Return on investment of 15% The results of this sensitivity analysis are shown in Figure 7.2. The proposed process is highly sensitive to fluctuations in capital investment and Eastar Bio Copolyester costs, and relatively insensitive to all other factors. 76 2.2 l + “ East—ar‘B‘ib‘Copolyestfer‘ 2-1 —Glycerol —Capital Investment A 2 m_-_..._.._._-_,, -1“.— 9.. 8 s t 1.9 n. a .5 g 1.8 m 1.7 1.6 1.5 -20% -15% -10% -5% 0% 5% 10% 15% 20% Percent Change in Cost Figure 7.2: Sensitivity analysis for the proposed process Conclusions The selling price for 10% discounted cash factor for the 70:30 EBC-g- MSOlngTPS resin is $2.23/lb while the selling price for the fiber containing blend is $1.63/lb. The average of these two resin costs is $1.93/lb. This average cost is $0.22/1b below the purchase cost of the copolyester itself. Since blend properties are inferior to the polyester properties itself, the small difference in selling price is not great enough to encourage product sales, unless increased rates of biodegradation are desired. Further optimization of product production to greatly improve properties and reduce production cost is necessary if these blends are to be attractive to perspective customers. 77 Examination of the effects of variation of the amount of grafted c0polyester on the product material cost for the 70:30;EBC-g-MSO:TPS blend reveals that not only does the grafting result in a product with superior properties, but it also reduces the material cost and in turn the selling cost as shown in Figure 7.3 $1.65 E a; $1.60 “2-2-2-- ,_- 3' 8 0 $1.55 - R '5 ,, $1.50 - _ - S $1.45 - o o o o o o o o o o If) C "7' 3’- ” 9‘! .15 as o ° ° 0 «3 N 00 <1- I!) lend Ratio(EBC:EBC-g-MSO1OG:TPS) Figure 7.3: Effects of varying graft copolyester amounts on material costs Though the lowest material costs are for the blend containing entirely graft copolyester, the best properties probably lie somewhere between the extremes. The overall improvement in properties coupled with a reduction in material costs associated with grafting provide a strong impetus for further examination of grafted copolyester as a compatibilizing agent for EBC:TPS blends. 78 CHAPTER 8 CONCLUSIONS AND RECOMMENDATIONS Conclusions The two goals outlined at the start of this work were: 1) the design and engineering of biodegradable starchzcopolyester blend. The copolyester used in this blend is Eastar Bio Copolyester (EBC), a biodegradable copolyester containing Adipic acid, Terepthalic acid, and 1 ,4-Butanediol. 2) the evaluation of maleated soybean oil (M80) as a compatibilizing agent. With the above goals in mind and based on the work accomplished the following conclusions can be drawn: 1) Soybean oil can be maleated cheaply and effectively using existing technology. 2) Maleated soybean oil can be grafted to Eastar Bio Copolyester using Dicumyl Peroxide as a catalyst, or with less efficiency no catalyst. 3) Grafting of M80 to BBC is verifiable using common extraction techniques. 4) Thermoplastic starch is easily prepared and reactively blended with EBC-g-MSOIOG. 79 5) Environmental Scanning Electron Microscopy and Tensile testing of the EBC-g-MSOIOGzTPS blends reveals direct evidence of blend compatibilization. 6) A process to produce 2.5 million pounds/year of two blends: EBC-g- MSOlOG:TPS; 70:30 and EBC-g-MSOIOGzTPS: Fiber; 48:32:20 was shown to produce these two resins at an average selling price of $1.93/lb. The addition of grafted copolyester to blends containing copolyester and thermoplastic starch results in a material possessing superior blend properties at a reduced material cost. This fact provides a driving impetus for further study of such blends. 80 Recommendation for further work The conclusions above do demonstrate achievement of the outlined goals. The goal of all engineers should be to take an idea and convert it to an economically profitable product. To achieve that end with the product idea examined in this thesis further work must be done. The suggest work will help achieve this goal of converting an idea into a marketable reality. 1) Further characterization maleated soybean oil and optimization of the maleation reaction. : Determination of the actual structure of MSO using GC-MS, FTIR and NMR techniques will provide insight as to the reactivity of this material in regards to increasing grafting efficiency. Optimization of this reaction will reduce production costs thereby decreasing its price. 2) Examination of catalyst efficiency and non catalytic grafting: Grafting of M80 to BBC occurs both in the presence and absence of catalyst. Only the 0.3% Dicumyl Peroxide concentration was investigated as a potential catalyst system. Other concentrations of catalyst and other catalysts should be investigated. In addition the mechanism of non- catalytic grafting must be further investigated. Better understanding of the graft reaction will lead to increased efficiency in product production. 3) Verification of grafting and characterization of the copolyester graft structure. Extraction of the grafted copolyester was performed to verify grafting. Further characterization of this extracted product using appropriate 81 analytical techniques to identify and verify the grafted copolyester structure. With this structure in hand the reaction can be better tailored to produce a better compatibilizing agent. 4) Production of different blends and examination of effects of varied processing conditions. Blending of thermoplastic starch with grafted copolyester should be done for different levels of grafting. Only the 10% by weight grafted copolyester was blended with TPS. Blending of the 5%, 10% non- catalytic, and the 20% by weight materials with thermoplastic starch should be performed to determine which one provides the best compatibilization. A 65:5:30; EBC:EBC-g-MSOIOGzTPS blend was found to provide excellent properties. Further blends of this nature should be made to determine if there is an optimum amount of grafted copolyester that should added to a copolyesterzthermoplastic starch blend to yield the best properties. In addition processing parameters such as throughput and extruder zone temperatures should be examined to determine if there is an optimum condition at which reactive blending should be performed. 5) Further Characterization of blends Tools such as ESEM, and property testing, and others such as Dynamic Mechanical Analysis-DMA, should be used to determine the extent of compatibilization. Biodegradability testing should be used to screen new materials and as a guide to which application the biodegradable materials are best suited for. 82 6) Pilot plant construction and determination of economic viability. The construction of a pilot plant is necessary to determine what if any problems there will be associated with scale up. These scale up problems should be addressed prior to completion of accurate assessment of economic viability. Completion of the further work will create a more accurate picture of the economic viability of the desired product. With this picture more clearly shown, a final decision can be made as to whether the capital investment necessary for large-scale plant construction will return a profit in the desired amount of time. 83 Bibliography Argyropoulos, D., MS. Thesis, Michigan State University, 1992. Bates, F.S., Science, 251, 898-905 (1991). Barentsen, M.M., Heikens, D., and Piet P., Polymer, 15, 119 (1974). Brown, SB. and Orlando, C.M., in Encyclopedia of Polymer Science and Engineering, John Wiley & Sons, 1988, 169-189Vol. 14. Buchanan, C.M., etal., 1995, US. Patent # 5446079 Callais, P.A., and Kazrnierzak, R.T., ANTEC 90, 1921-1923, 1990. Callais, P.A., Compalloy 89 (1989). Cheremisinoff, N.P., Guidebook to Extrusion Technology, PTR Prentice Hall, 1993. Cheung, P., Suwanda, D., and Balke, S.T., Polym. Eng. Sci., 30, 1063-1072 (1990). Dalvag, H., Klason, C., and Stromvall, H.E., Intern. J. Polymeric Mater., 1 1, 9-38, (1985). Eise, K., Curry, .1., and Nangeroni, F.F., Polym. Eng. Sci., 23, 642-643 (1983). F olkes, M.J., and Hope, P.S., eds., Polymer Blends and Alloys, Blackie Academic & Professional, 1993. Fox, T.G., Bull Amer, Phys. Soc., 1, 123 (1956). Franklin Associates Limited, Characterization of Municipal Said Waste in the United States, 1994 Update, Report No. EPA 530-5-94-042, Nov. 1994. Gupta, A., MS. Thesis, Michigan State University, 1994. Gaylord, N.G., and Mehta, M., J.Polym. Sci. Polym. Let. Ed, 20, 481 (1982). 84 Han, C .D., Multiphase Flow in Polymer Processing, Academic Press, Inc., 1981. John, J. et al., J. Polym. Sci. A: Polym Chem, 35, 1139-1148. Karnani, R., Krishnan, M., and R. Narayan, Polym. Eng. Sci., 37: 476-483, (1996). Kienzle, S.Y., in Polymer Blends and Alloys: Guidebook to Commercial Products, Technomic Publishing, 1988, 1-6. Kirk, - Othmer, “Soybeans and Other Oilseeds”, Encyclopedia of Chemical Technology, 422 (1 983a). Kirk-Othmer, Peroxides and Peroxy Compounds, Organic, Enxyclopedia of Chemical Technology, 13, 362 (1983b). Kreisher, K.R., Mod. Plastics, July 1990, 38-40. Krishnan, M., Doctoral Dissertation, Michigan State University, 1998. Levy, 8., Plastics Extrusion Technology Handbook, Industrial Press Inc., 1981. MAA, C., and Chang, F., Polym. Eng. Sci., 49, 913-924 (1993). Manson, J .A., Sperling, L.H., Polymer Blends and Composites, Elsevier,(l976). Michaeli, W., and Grefenstein, A., Polym. Eng. Sci., 35, 1485-1504 (1995). Narayan, R, in ACS Symposium Series 476, 1 (1991). Narayan, R., Materials Research Society, Symposium Proceedings, 266: 93 (1992). Noshay, A., McGrath, J .E., Block Copolymers Overview and Critical Survey, Academic Press, 1977. Paul, D.R., Polymer Blends, Vol 1, Academic Press, 1978a. Paul, D.R., Polymer Blends, Vol 2, Academic Press, 1978b. Paul, D.R., Barlow, J .W., and Keskkula, H.., in Encyclopedia of Polymer Science and Engineering, John Wiley & Sons, 1988, 399-461, Vol. 12. Peters, M.S., and Timmerhaus, K.D., Plant Design and Economics for Chemical Engineers, Fourth Edition, McGraw-Hill, Inc., 1991. 85 Plochocki, A.P., Polym. Eng. Sci., 23, 618-626 (1983). Rader, C.P., et al, eds., Plastics Rubber and Paper Recycling A Pragmatic Approach, ACS Symposium Series 609, 1995. Riggle, D. Biocycle, 39, 64-70 (1998). Root, RB, 1945, US. Patent # 2374381. Rudin, A., J ., Macromol. Sci. —Rev. Macromol. Chem, C19, 267-292 (1980). Rudin, A.J., The Elements of Polymer Science and Engineering, Academic Press, (1982). Sathe, S.N., Rao, G.S.S., and Devi, S., J. Appl, Polym. Sci., 53, 239 (1994). Shaw, M.T., in Polymer Blends and Mixtures, Rartinus J ijhoff Publishers, 37 (1985). Shultz, A.R., and Mankin, G.I., J. Polym Sci. Polym. Symp., 54, 341 (1976). Shultz, A.R., and Young, A.’L., Macromolecules, 13, 663 (1980). Seybold, K.R., MS. Thesis, Michigan State University, 1997. Sneller, J .A., Mod. Plastics, July1985, 56-60. Stinson, S.C., Chem. & Eng. News, 20-22, January 15, 1996. Tucker, C.S., abd Nichols, R.J., S.P.E. ANTEC Tech. Papers, 33, 117 (1987). Utracki, L.A., Polymer Blends Alloys and Blends, Hanser Publishers, 1989. Utracki, L.A., Polym. Eng. Sci., 35, 2-17 (1995). Vaidya, RR, and Bhattacharya, M., J.Appl. Polym. Sci., 52, 617-628, (1994). Vermesch, 1., Groeninck, G., J. Appl. Poluym. Sci., 53, 1365 (1994). Wright, R. personal communication, e-mail 1998. Xanthos, M., ed., Reactive Extrusion Principles and Practice, Hanser Publishers, 1992. Xanthos, M., Polym. Eng. Sci., 28, 1392-1400 (1988). 86 I"Illlllllllll