And Acrylonitrile Butadiene Styrene by R

Home , Adipate, Engineering plastic, Ingeo, Polylactic acid, Rubber toughening

Designing and Engineering a Sustainable Polymer Material by Blending Poly(lactic acid) and Acrylonitrile Butadiene Styrene

Ryan Vadori

A Thesis Presented to The University of Guelph

In partial fulfillment of requirements for the degree of Doctor of Philosophy in Engineering

Guelph, Ontario, Canada © Ryan Vadori, December, 2016

ABSTRACT

DESIGNING AND ENGINEERING A SUSTAINABLE POLYMER MATERIAL BY BLENDING POLY(LACTIC ACID) AND ACRYLONITRILE BUTADIENE STYRENE

Ryan Richard Vadori Advisor: University of Guelph, 2017 Professor M.Misra

Polymer blends containing poly(lactic acid) (PLA) and acrylonitrile butadiene styrene (ABS) were made by extrusion and injection molding. Due to the immiscible nature of the two, coupling was required. Additives were found that increased adhesion between the two, leading to increased performance. This work outlines the development and detailed characterization of the blends. There are several areas of focus in this work. The processing of each polymer was assessed and the effects of particular process conditions were studied. Notably, ABS had a propensity to thermo-oxidative degradation, which is outline in the first original research chapter.

The effects of the additives on the blend properties were investigated on the blends, showing the development of high performance blends. The effect of increasing the range of PLA contents was also studied. Statistical optimization was done to find the best blend formulation. Finally, the structure development of the blends was studied.

Dedicated to my wife, Alexandra

iii

Acknowledgements

The road to a Ph.D. dissertation is paved with the help of friends, colleagues, and mentors.

The help and guidance of my co-advisors, Dr. Manju Misra and Dr. Amar Mohanty cannot be overstated. I wish to sincerely thank them for providing me the opportunity to work as their student. Their presence, guidance, and support, throughout my studies cannot be expressed in words. I would like to thank them for their trust and belief in me, their encouragement has instilled a self-confidence that I will carry with me into my future.

I am also grateful to my committee advisors, Dr. Fantahun Defersha and Dr. Alfons

Weersink who has brought perspective to me work. They took an interest not only in the path of my research, but also in my personal welfare throughout my studies.

I am thankful to my Bioproducts Discovery and Development Centre (BDDC) colleagues, especially the ones that were available for guidance. Dr. Nima Zarrinbakhsh was always available to offer very knowledgable advice. Ehsan Behazin, my fellow Ph.D. candidate, was my brother in the lab and the coffee shop.

It is because of the support and encouragement of my family, especially my Mother and

Father that I was able to complete. This doctorate too is due to them.

The final but certainly not the least of my acknowledgments is to my wife. Her patience and understanding while I complete my studies are truly the attributes more commonly seen from members of sainthood. Her unending support is the reason that I am attaining this degree. This work is dedicated to her.

The financial support from the Ontario Ministry of Agriculture and Food Rural Affairs

(OMAFRA)/University of Guelph - Bioeconomy for Industrial Uses Research Program (Project

#200245); the Natural Sciences and Engineering Research Council (NSERC, Canada Discovery grants (Project #400322) and NSERC- AUTO21 NCE (Project #400372 & 400373); Ontario

Research Fund, Research Excellence Program; Round-4 (ORF-RE04) from the Ontario Ministry of Economic Development and Innovation (MEDI) (Project #050289) and the Ontario Research

Fund, Research Excellence Program; Round-7 (ORF-RE07) from the Ontario Ministry of

Research and Innovation (MRI), currently known as the Ontario Ministry of Research,

Innovation and Science (MRIS) (Project # 052644 and # 052665) to carry out this research is gratefully acknowledged. I greatly acknowledge the financial support from the Canada

Foundation for Innovation (CFI), Canada and the FedDev, Ontario for the BDDC infrastructure facility to carry out this research.

Table of Contents

Acknowledgements ...... iv List of Tables ...... ix List of Figures ...... x List of Abbreviations and Defined Terms ...... xiii List of Peer reviewed journal Publications ...... xv Chapter 1: Introduction ...... 1 1.1 Introduction ...... 2 1.2 Structure Of The Thesis and Overall Connections ...... 7 1.3 Research Problem ...... 9 1.5 Significance ...... 11 Chapter 2: A Criterion for Toughening PLA through Blending: Special Focus on Blending With ABS ...... 12 2.1 Introduction ...... 13 2.2 PLA ...... 16 2.2.1 Structure & Synthesis ...... 16 2.2.2 Crystallinity ...... 19 2.2.3 Properties & Processing Techniques ...... 20 2.3 ABS ...... 21 2.3.1 Structure and Synthesis ...... 21 2.3.2 Properties...... 23 2.3.3 ABS Toughening ...... 24 2.4 General Blending/Toughening Theory ...... 26 2.5 Failure Mechanisms ...... 28 2.5.1 Poly(lactic acid) ...... 28 2.5.2 Acrylonitrile Butadiene Styrene ...... 30 2.6 Toughening Criterion – Critical Matrix Ligament Thickness Theory ...... 30 2.7 Strategies For Toughening ...... 33 2.7.1 Reactive Compatibilization ...... 34 2.7.2 Reactive Polymer Synthesis ...... 40 2.7.3 Processing Steps ...... 44 2.7.4 Strategy Outlook ...... 44 2.8 Philosophy of Experimentation ...... 46 2.9 Conclusion ...... 46 2.10 References ...... 47 Chapter 3: Studies on the Reaction of Acrylonitrile Butadiene Styrene to Melt Processing Conditions ...... 57 3.1 Introduction ...... 58 3.2 Experimental ...... 59 3.2.1 Materials ...... 59 3.2.2 Sample Preparation ...... 60 3.2.3 Characterization ...... 60 vi

3.3 Results And Discussion ...... 60 3.3.1 Neat ABS ...... 61 3.3.2 Increasing Retention Time ...... 65 3.3.3 Temperature Versus Shear ...... 67 3.4 Conclusion ...... 74 3.5 Acknowledgements ...... 75 3.6 References ...... 75 Chapter 4: Sustainable Biobased Blends from the Reactive Extrusion of Polylactide (PLA) and Acrylonitrile Butadiene Styrene ...... 78 4.1 Introduction ...... 79 4.2 Experimental ...... 81 4.2.1 Materials ...... 81 4.2.2 Preparation of the Blends ...... 82 4.2.3 Mechanical Testing ...... 82 4.2.4 Dynamic Mechanical Analysis...... 83 4.2.5 Differential Scanning Calorimetry ...... 83 4.2.6 Rheology ...... 83 4.2.7 Fourier Transform Infrared Spectroscopy (FT-IR) ...... 84 4.3 Results And Discussion ...... 84 4.3.1 Tensile and flexural properties ...... 84 4.3.2 Impact strength ...... 88 4.3.3 Phase Morphology (SEM) ...... 90 4.3.4 Thermal Crystallization (DSC) ...... 92 4.3.5 Rheology ...... 95 4.3.6 Atomic Force Microscopy (AFM) ...... 98 4.3.7 Chemical Reaction Mechanism ...... 100 4.4 Conclusions ...... 107 4.5 Acknowledgements ...... 108 4.6 References ...... 109 Chapter 5: Bio-Based Acrylonitrile Butadiene Styrene (ABS) Polymer Compositions And Methods Of Making And Using Thereof ...... 114 5.1 Field Of The Invention ...... 115 5.2 Background of the Invention ...... 115 5.3 Summary of the Invention ...... 119 5.4 Brief Description of the Drawings ...... 120 5.5 Detailed Description Of The Invention ...... 121 5.5.1 Definitions ...... 121 5.5.2 Acrylonitrile butadiene styrene (ABS)/polylactic acid (PLA)-based blends ...... 123 5.5.3 Composites ...... 128 5.5.4 Method of manufacturing blends and composites containing the blends...... 130 5.6 Examples ...... 134 5.6.1 Materials and Methods ...... 134 Example 1. Preparation of acrylonitrile butadiene styrene (ABS)/polylactic acid (PLA) blend containing an acrylic copolymer-based lubricant and chain extender using lab scale equipment ... 135 Example 2. Preparation of acrylonitrile butadiene styrene (ABS)/polylactic acid (PLA) blend containing an acrylic copolymer-based lubricant and chain extender using pilot scale equipment . 139 Example 4. Preparation of acrylonitrile butadiene styrene (ABS)/polylactic acid (PLA) blend processed at previous art conditions ...... 142 vii

Example 5. Preparation of acrylonitrile butadiene styrene (ABS)/polylactic acid (PLA) blend processed with alternate additives ...... 143 Example 6. Morphology changes for high performance blends...... 144 5.7 Claims ...... 145 5.8 Discussion ...... 150 Chapter 6: Statistical Optimization of Compatibilized Blends of Poly(lactic acid) and Acrylonitrile Butadiene Styrene ...... 155 6.1 Introduction ...... 156 6.2 Experimental ...... 158 6.2.1 Materials ...... 158 6.2.2 Material Processing ...... 159 6.2.3 Material Characterization ...... 159 6.2.4 Statistical Design and Analysis ...... 159 6.3 Results And Discussion ...... 161 6.3.1 Mechanical Properties ...... 161 6.3.2 Model Development ...... 164 6.3.3 Model Adequacy ...... 165 6.3.4 Fitted Models ...... 169 6.3.5 Optimization ...... 176 6.4 Conclusions ...... 194 6.5 Acknowledgements ...... 195 6.6 References ...... 195 Chapter 7: Overall Conclusions and Future Work ...... 198 7.1 Preliminary Economic Analysis of BioABS Blends ...... 199 7.2 Key Outcomes ...... 201 7.3 Overall Conclusions of The Thesis ...... 202 7.4 Future Work ...... 203 7.5 References: ...... 204

viii

List of Tables

Table 2.1 – Thermal Properties of PLA –L and –D copolymers

Table 2.2 – Mechanical properties of PLA and ABS

Table 2.3 – Mechanical Properties of different ABS grades

Table 2.4 – Notable PLA/ABS blends in the literature

Table 3.1 - Impact strength values for various retention times.

Table 3.2 - Impact strength values for various shear speeds at 220°C and 240°C.

Table 4.1 – Mechanical properties of BioABS molded at different temperatures

Table 5.1 - Mechanical properties of neat ABS and ABS/PLA blend

Table 5.2 - Mechanical properties of ABS/PLA blend using pilot scale equipment

Table 5.3 - Mechanical properties of composites of ABS/PLA blends

Table 5.4 - Mechanical properties of ABS/PLA blend with altered processing conditions

Table 5.5 - Mechanical properties of ABS/PLA blends processed with alternate additives

Table 6.1 - Mechanical properties of 13 formulations

Table 6.2 - ANOVA results for models of impact strength, elongation at break, tensile strength,

and tensile modulus

Table 6.3 - Regression models and statistics values

List of Figures

Figure 2.1 – Ring opening polymerization of PLA

Figure 2.2 – Chirality of PLA

Figure 2.3 – Fracture toughness coefficient as a function of crystallinity of PLA

Figure 2.4 – Impact strength as a function of rubber particle diameter in Nylon/Rubber blends

Figure 2.5 – Convergence of toughening at the critical ligament size in Nylon/Rubber blends

Figure 2.6 – Morphology of PLA/ABS blends with SAN-GMA compatibilizer by Li et al.

Figure 2.7 – Morphology of PLA/ABS blends from Choe et al.

Figure 2.8 – Morphology of blends of PLA/ABS with a synthesized compatibilizer

Figure 2.9 – Morphology of PLA/ABS blends compatibilized by a reactive comb polymer by

Dong et al.

Figure 3.1 - Viscosity evolution over time of ABS at 180°C, 200°C, 220°C and 240°C.

Figure 3.2 - Viscosity evolution over time of ABS at 180°C, 200°C, 220°C and 240°C.

Figure 3.3 - Frequency sweep of ABS at 220°C, after processing at 0.5, 1, 1.5,2,3,5 min.

Figure 3.4 - Frequency sweep of ABS processed at 240°C and 100, 50 and 0 RPM.

Figure 3.5 - Frequency sweep of ABS processed at 220°C and 100, 50 and 0 RPM.

Figure 3.6 - Frequency sweep of ABS processed at 100RPM: 220°C versus 240°C.

Figure 3.7 - Frequency sweep of ABS processed at 50RPM: 220°C versus 240°C.

Figure 3.8 - Frequency sweep of ABS processed at 0RPM: 220°C versus 240°C.

Figure 4.1 – Tensile properties of neat polymers and blends: (A) Neat PLA (B) PLA/ABS (C)

PLA/ABS/Acrylic copolymer (D) PLA/ABS/Chain extender (E) PLA/ABS/Chain

extender/Acrylic copolymer (F) Neat ABS

Figure 4.2 – Elongation at break and impact strength of neat polymers and blends: (A) Neat PLA

(B) PLA/ABS (C) PLA/ABS/Acrylic copolymer (D) PLA/ABS/Chain extender (E)

PLA/ABS/Chain extender/Acrylic copolymer (F) Neat ABS

Figure 4.3 – Scanning electron microscopy of impact fractured surfaces of A) PLA/ABS B)

PLA/ABS/Acrylic copolymer C) PLA/ABS/Chain extender D) PLA/ABS/Acrylic

copolymer/Chain extender

Figure 4.4 – Differential scanning calorimetry thermogram of first heating cycle

Figure 4.5 – Storage modulus of PLA/ABS blends

Figure 4.6 – Complex viscosity curve of PLA/ABS blends

Figure 4.7 – AFM modulus mapping of PLA/ABS blends: (A) PLA/ABS (B) PLA/ABS/Acrylic

copolymer (C) PLA/ABS/Chain extender (D) PLA/ABS/Chain extender/Acrylic

copolymer

Figure 4.8 – Fourier transform infrared spectra of blend constituents compared to full final

prepared blend

Figure 4.9 – Proposed reaction between PLA and the chain extender

Figure 4.10 – DSC heating and cooling curves of PLA/ABS and the full blend

Figure 4.11 – DSC heating curves of the full blend annealed at different temperatures

Figure 4.12 – AFM modulus mapping of the full blend microstructure

Figure 5.1 – SEM of impact fracture surface of blend 6A

Figure 5.2 – SEM of impact fracture surface of Blend 1B

Figure 5.3 – AFM Deformation plot of blend 6B

Figure 5.4– AFM Deformation plot of blend 1B

Figure 6.1 - Design of extreme vertices mixture experiment schematic

Figure 6.2(A) - Residual plots for Impact Strength

Figure 6.2(B) - Residual plots for Elongation at Break

Figure 6.2(C) - Residual plots for Tensile Strength

Figure 6.2(D) - Residual plots for Tensile Modulus

Figure 6.3(A) - Cox response trace plot of impact strength

Figure 6.3(B) - Cox response trace plot of elongation at break

Figure 6.3(C) - Cox response trace plot of tensile strength

Figure 6.3(D) - Cox response trace plot of tensile modulus

Figure 6.4(A) - Optimization plot of impact strength

Figure 6.4(B) - Optimization plot of elongation at break

Figure 6.4(C) - Optimization plot of tensile strength

Figure 6.4(D) - Optimization plot of tensile modulus

Figure 6.4(E) - Composite optimization plot of all responses

Figure 6.5(A) - Mixture contour plot of impact strength

Figure 6.5(B) - Mixture contour plot of elongation at break

Figure 6.5(C) - Mixture contour plot of tensile strength

Figure 6.5(D) - Mixture contour plot of tensile modulus

Figure 6.6(A) - Overlaid contour plot all four responses

xii

List of Abbreviations and Defined Terms

ABS Acrylonitrile Butadiene Styrene (standard grade)

Adj MS Adjusted mean squares

Adj SS Adjusted sums of squares

ANOVA Analysis of variance

ASTM American Society for Testing and Materials

DMA Dynamic mechanical analysis

DOE Design of experiments

DSC Differential scanning calorimetry

E' Storage modulus

E'' Loss modulus

F F-value (statistics)

FM Flexural modulus

FTIR Fourier transform infrared spectroscopy

HDT Heat deflection temperature

IS Impact strength

J/m Joules per meter

LDPE Low Density Polyethylene

MPa Mega Pascal

MWD Molecular weight distribution

P P-value (in ANOVA table)

PBT Polybutylene terephthalate

PBAT Poly(butylene adipate-co-terephthalate)

xiii

PCL Poly(caprolactone)

PE Polyethylene

PHBV Poly(hydroxybutyrate-co-valerate)

PLA Poly(lactic acid)

POM Polarizing optical microscopy

PP Polypropylene

PS Polystyrene

PVC Poly(vinyl chloride)

R2 R-squared

R2adj Adjusted R-squared

R2pred Predicted R-squared

RPM Revolutions per minute

SEM Scanning electron microscopy

Seq SS Sequential sums of squares

TGA Thermogravimetric analysis

Tg Glass transition temperature

Tm Melting temperature

Tc Crystallization temperature

TS Tensile strength wt Weight

ε Error term (in regression model)

ΔHf Heat of fusion

xiv

List of Peer Reviewed Journal Publications

1. Vadori, Ryan, Misra, Manjusri and. Mohanty, Amar K. "Studies on the Reaction of Acrylonitrile Butadiene Styrene to Melt Processing Conditions. “Macromolecular Materials and Engineering 300.7 (2015): 750-757. Found in thesis as Chapter 3.

2. Ma, Zhaohui, Weersink, Alfons, Vadori, Ryan, and Misra, Manjusri "Financial Cost Comparison of Acrylonitrile Butadiene Styrene (ABS) and BioABS." Journal of Biobased Materials and Bioenergy 9.2 (2015): 244-251.

3. Vadori, Ryan, Misra, Manjusri, and Mohanty, Amar K. "Sustainable biobased blends from the reactive extrusion of polylactide and acrylonitrile butadiene styrene." Journal of Applied Polymer Science (2016). Found in thesis as Chapter 4.

4. Misra, Manjusri, Vadori, Ryan and Mohanty, Amar K. "Bio-Based Acrylonitrile Butadiene Styrene (Abs) Polymer Compositions And Methods Of Making And Using Thereof." U.S. Patent No. 20,160,009,913. 14 Jan. 2016. Found in thesis as Chapter 5.

5. Vadori, Ryan, Misra, Manjusri, and Mohanty, Amar K. "Statistical Optimization of Compatibilized Blends of Poly(lactic acid) and Acrylonitrile Butadiene Styrene." Journal of Applied Polymer Science 134.9 (2017). Found in thesis as Chapter 6.

List of Conference Proceeding Papers:

1. Vadori, Ryan, Misra, Manjusri, and. Mohanty, Amar K.. "Design and Engineering of BioABS." International Symposium on Bioplastics, Biocomposites, and Biorefining. 2016.

2. Vadori, Ryan, Misra, Manjusri, and Mohanty, Amar K.. "Design and Engineering of BioABS" Society of Plastics Engineers Annual Technical Conference (SPE-ANTEC). 2015.

3. Vadori, Ryan, Misra, Manjusri, and Mohanty, Amar K. "Design and Engineering of BioABS" Biobased New Industrial Bioproducts (BioNIBS). 2015.

4. Vadori, Ryan, Misra, Manjusri, and Mohanty, Amar K.. "Bioplastics for High Performance Durable Automotive Applications" AUTO21 Annual Meeting (Poster). 2015.

5. Vadori, Ryan, Misra, Manjusri, and Mohanty, Amar K.. "Studies on the Reactive Blending of Poly(lactic acid) and Acrylonitrile Butadiene Styrene Rubber." International Symposium on Bioplastics, Biocomposites, and Biorefining. 2014.

xvi

Chapter 1: Introduction

______

Abstract

In this chapter, the topic of the work is introduced, along with the structure of the thesis. A short description of each chapter is given, before detailing the research aim, objective, hypothesis, and significance of the work.

______

1.1 Introduction

Amidst ever-growing evidence of the negative effects of the use of petroleum, many are becoming increasingly eco-conscious. In addition to increased focus on alternative fuels, society is interested in sustainable materials as alternatives to traditional petroleum based plastics

(polymers). Petroleum based plastics have emerged in the 20th century as an extremely economic material with exceptional overall performance. The strength to weight ratio of many plastics is very high, lending to their widespread use across many industries. Due to their unique long-chain molecular structure, they can be very flexible and be deformed a great amount without permanent damage. Another very attractive attribute of plastics has been in technological developments that have taken place to process and mold them into final form. Injection molding, a method of heating the plastic into a molten state then injecting it into a clamped mold, is by far the fastest, easiest and most economical method of material molding. The advent of this process has allowed the making of shapes previously thought impossible, all with relative ease.

Similarly, blow molding, cast film molding, thermoforming, compression molding and resin transfer molding have allowed relatively fast processing of unique geometries, all at a low cost.

Thus, when product development requires material selection, it seldom overlooks plastic as the best material for the given application.

It is perhaps the wide variability structure, performance, and cost that is the most impressive of plastic as a material. Plastics can be broken down into the categories of commodity and engineering. As can be ascertained from their names, commodity plastics are lower performing, cheaper, and more abundant, while engineering plastics may be used for higher performance applications, but also have a higher cost. Not only do the qualities of plastics vary from each type of polymer, but a given polymer can be changed to have longer chains, higher density, or

differently spatially arranged chains, which can cause a drastic change in the performance. It is certainly not surprising why plastics have become extremely prevalent in most industries.

However, as previously stated, most plastics are petroleum based, causing an alarming rate of use of petroleum for everyday products. This is the motivation for many to focus on the development and use of bioplastics.

Therefore, there is great interest in developing sustainable bioplastic and biocomposite materials.

Bioplastics can refer to two differing qualities of a material. First, a bioplastic is any plastic that is derived from renewable resources such as cellulose, protein, sugars, etc. These bio-based plastics offset the use of petroleum as a derivative, and their advantages as eco-friendly materials are obvious. The second quality that is referred to in the term bioplastic is the biodegradability of the material. This stems from the problem of landfilling, and its inherent risks such as groundwater pollution, convenience, and space considerations. The biodegradability of a plastic material refers to the rate at which it degrades back into (non-toxic) constituent molecules, and absorbed into soil. These rates can be tested compared to a standard to obtain the biodegradable designation. Biodegradability and biobased content are not mutually inclusive, such that a biodegradable plastic may be biobased or petroleum based. Thus, there is mainly one of two separate but related problems that is being addressed with the development and use of a bioplastic.

Biocomposites refer to the blending polymers with natural fibers to create a totally new material.

Natural fibers are derived from renewable resources, namely agricultural or forest resources.

Generally, these fibres are co-products or by-products of an agricultural stream. This means that the natural fibres are very often inexpensive and plentiful. The use of natural fibres in polymer biocomposites can have advantages over traditional composite materials in cost, weight,

sustainability, and good specific strength. Natural fibres can be classified based on the derivation. Leaf fibres are obtained from the leaves of plants (pineapple, banana, sisal, etc.), bast fibres are obtain from the stalk of the plant (jute, hemp, kenaf, flax, etc.), fruit fibres (coconut, etc.), grass fibres (Miscanthus, switchgrass, etc.) and agricultural residues (corn straw, wheat straw, corn stover, soy stalk, and rice stalk etc.). Natural fibres have three main constituents in cellulose, hemicellulose, and lignin. Based on which biomass the natural fibre is obtained, the chemical composition can vary widely. One main drawback of using natural fibres is consistency across each fibre based on growing region and conditions, creating biocomposites that may have a varying performance.

The challenges that must be faced for wider use of bioplastics are not trivial. Most bioplastics have performance issues that prevent their immediate use. Generally, bioplastics may be very tough (ductile) and weak, or they may be strong but brittle. A balance of properties, which is attractive, is not currently known in the ordinary bioplastics. In many cases, the cost is somewhat higher than the traditional plastics that currently hold industry relevance. Of these attributes, one bioplastic emerges has having the highest potential for viable use in several industries.

Poly(lactic acid) (PLA) is a polyester material that is derived from the sugars of common crops such as corn and beets. In addition to being biobased, it is compostable according to the standards set forth by the American Standard of Testing Materials (ASTM). Due to several technological advancements in its synthesis and due to economies of scaling up, PLA has decreased in cost since its discovery to a point that it is competitive with petroleum based commodity plastics. It is for this reason that it has seen focus as a potential material for many low performance applications. The few performance limitations of PLA is its true barrier to entry to wider applications. It is very strong, with a high stiffness, but is very brittle (low toughness).

In addition, it has poor thermal stability at relatively low temperatures, barring it from use for many common applications. Thus, it is currently only used for very niche applications, and likely only for the reason that it is biobased and biodegradable. If PLA is to become a more widely used bioplastic, research must focus on increasing its toughness, or raising its thermal stability, or both.

In order to serve as an alternative for a petroleum based plastic, the biobased plastic must exhibit similar properties. Thus, it is prudent to choose a petroleum based polymer to replace. In this work, acrylonitrile butadiene styrene (ABS) has been chosen due to its extensive use in several industries, including automotive and electronics. In fact, ABS is the third most used polymer in the automotive industry, with over 1 million tons to be used in 2016. This is due to its relatively low cost, but good performance, placing it in the engineering polymer category of traditional polymers. ABS has a good balance of properties including strength and stiffness, but it is most known for its high toughness, specifically impact strength.

One method for finding an alternative to an entirely petroleum based polymer is to blend it with a biobased polymer. Although this does not offset all petroleum use, it increases the likelihood of developing a polymer that meets the performance requirements. Additionally, by choosing the constituent polymers to be complementary, a balance in properties can be found. However, as mentioned, ABS target properties have a good balance of strength, stiffness, and toughness.

Thus, choosing ABS as a blending partner to PLA increases the likelihood of retaining the strength and stiffness, while improving toughness of the PLA. Choosing a polymer with a very high toughness, but lacks in strength and stiffness may not allow PLA to retain the strength and stiffness after blending.

In this work, PLA will be blended with ABS. This increases the likelihood of a successful blend in terms of properties. Depending on several factors, such as thermodynamics of blending, compatibility, initial properties, etc., the final blend may be a balance of properties between the two constituent polymers. The challenge is to increase the compatibility of the two polymers, as generally, good compatibility is rarely seen between neat polymers alone. In fact, the overwhelming majority of industrially used polymer blends are two phase, not alloys. There are, again, several approaches to increasing the compatibility between two plastics. The most popular involves adding a third (or more) element to the blend. The goal here can be for the third component to act as an intermediary, reducing tension, or, it can be to react, bonding the two polymers together. Obviously, this is a much applied approach, and deep chemical knowledge of the constituent polymers and the compatibilizer is needed to try to predict the outcome.

There have been several studies that have blended PLA and ABS, which are outlined in subsequent chapters. However, as the next chapter will explicitly outline, a PLA/ABS blend has not yet been developed that is economically viable as a commercial material. Thus, one of the aims of this work is to develop a material to fill this need. All components used in this research are commercially available, easily obtainable, and economically inexpensive. Additionally, the methods used are traditional industry standards, so that this material stands as a drop-in replacement for ABS. The properties of ABS compliment PLA well, as it is a very tough polymer. In addition, its structure, comprised of three polymers, allows for a wide range of resulting properties as the ratios are adjusted. This can also be done to increase the thermodynamic likelihood of creating a well-dispersed polymer after blending. Thus, the specific grades of polymers used become important to increase their complimentary nature. Considering the low toughness of PLA, the ABS grade should be high impact. This gives a high likelihood

that once compatibility is achieved between the polymers, the high toughness of the ABS will increase the overall toughness of the blend. .

1.2 Structure Of The Thesis and Overall Connections

This thesis is composed of several chapters, all of which present ideas or data on a topic related to the blending and processing of PLA/ABS blends. Presented together, these sections represent a unified research project that details the development and characterization of high performance

PLA/ABS blends. The present chapter, Chapter 1, outlines the background problem, the research question, and the aims, hypothesis, and objectives.

Chapter 2 is a literature review of the studies done on PLA, ABS, and their blends. Here, the authors give details on the synthesis, structure, mechanical properties, and thermal properties of each polymer before focusing on a specific criterion for creating a tough PLA/ABS blend. It is the contention of the authors that the morphology development is the single most important aspect of creating tough PLA/ABS blends. The methods for creating such morphology through traditional blending methods are given, and the most promising identified.

The introduction into the materials and mechanisms from Chapter 2 leads to a processing study of ABS in Chapter 3. Chapter 3 is an original research article that deals with the fundamental aspect of blending ABS polymer. Through extensive studies, the authors have found that it is possible for ABS to undergo thermo-oxidative degradation under melt compounding conditions at high retention times. Thus, the processing parameter chosen are of vital importance for retaining the high performance of ABS while blending.

After investigating the effects of processing on the individual polymers, blending formulation was then possible. Chapter 4 is an original research article that presents the results of the development of a high performance PLA/ABS blend. The article further characterizes the blended material, proving the effect of the additives on the properties. Additionally, a few ideas are given on how the blend properties are developed in terms of reaction and thermal properties.

The authors believe that this blend is commercially viable for its high performance, low cost, and ease of processing. Finally, an analysis of the structure of the blend is given along with an investigation of the toughening mechanism present in the blends. This analysis is built upon by adjusting the failure mode of the PLA, creating an increase in the impact strength.

The development of the material in Chapter 4 led the authors to seek a patent on this novel and commercially viable material. Chapter 5 is a patent application written by the authors on the blend developed and described in Chapter 3. In this, a plethora of experimental formulations of the blend are described, with PLA contents from 25 – 70 wt. %. Additionally, the patent presents data proving the ability to scale up the production of the material by blending on pilot scale equipment. A full scientific discussion (not found in the patent application) is amended explaining the properties and the effects of the additives at these differing formulation and process conditions.

Chapters 4 and 5 describe the developed material, but leave a need to investigate the optimization of such a blend. Chapter 6 is an original research article that shows the statistical optimization of the blend containing 50 wt. % PLA. Through this analysis and optimization, it was found that the blending characteristics and resulting material properties are highly sensitive to one of the additives, which presents a local maximum in the range of contents tested.

Additionally, it was found that most properties could be raised or lowered collectively, meaning

that a balance of strength to toughness is not necessary, rather both are optimized in the best blends.

Chapter 7 is a summary of the conclusions reached through these research projects, a brief economic analysis, and a reflection on the future work required.

1.3 Research Problem

In addition to leading to the creation of greenhouse gases, petroleum use is depleting fossil fuel reserves. Thus, materials that are derived (or partially derived) from plant based plastics are needed to replace petroleum based traditional plastics. The aim of this work is to develop a polymer material with increased bio-content to replace traditional acrylonitrile butadiene styrene.

A further goal of this work is to develop a durable polymer blend, for use in application that have relatively long life. Thus compostability is not as much a concern as biobased content in the blend. For a durable good, the sustainability of the material is influenced more by its derivatives

(petroleum versus plant) than for a compostable, single use product. Specifically, the goal of this work is to incorporate renewable resource polymers into a blend with ABS. The potential applications for such a material are for those that ABS is currently used, such as in the automotive and electronics industries. Thus, the property targets for this material are equivalent or better to standard grades of ABS. The bulk of the work to be done will be on increasing performance of PLA/ABS blends, as it has been universally determined that these two polymers are incompatible with each other. Thus, this project is focused on finding an additive that can efficiently increase the performance of the blends by increasing adhesion between the two polymer phases.

1.4 Objectives and HypothesesIt has been well established in previous works that PLA and ABS have a poor affinity for each other during blend compounding. Thus, this research is focused on finding a formulation between PLA and ABS that incorporates additives to achieve better compatibility, and therefore better performance. Specifically, this research sets out to:

 Investigate the fundamental process parameters of the neat polymers. Processing

windows may be found that will foster better blending between the two polymers.

Conversely, the reaction of these polymers to processing limits will give insight into

reducing the degradation of the polymers during melt compounding.

 Prepare blends of ABS and PLA with different coupling agents through traditional

methods, namely twin screw compounding and injection molding. These screening

experiments will set out to find a working formulation between PLA and ABS. The

impact strength of these formulations will serve as an indication of effectiveness of

the coupling agent(s) used. However, as all mechanical properties are important, the

blends with high impact strength values will also be tested for tensile and flexural

strength and stiffness in addition to elongation at break.

 Once the performance is increased, the mechanism by which the blends achieve better

performance must be studied in terms of chemical reaction mechanisms and

mechanical interaction mechanisms. If these mechanisms are pinpointed, further

development of the material may be diagnosed.

 A range of formulations, including the amount of PLA (and therefore biobased

content), must be investigated for the mechanical performance. This will give insight

into the nature of the formulation and its upper and lower bounds in terms of biobased

content.

 Additionally, statistical methods may be used to optimize a formulation for

mechanical properties. This will also give insight into the sensitivity of the

formulation and the true effects of each component.

1.5 Significance

The aforementioned sections culminate in the development of polymeric materials that have notable advantages in several key areas:

 The PLA/ABS blends, with proper compatibilization, will serve as an alternative to a

vastly used petroleum based polymer, thus offsetting an amount of petroleum as a

material derivative.

 As PLA has a lower cost than ABS, the resulting blends have the potential to become

cheaper than currently utilized ABS, and therefore be very attractive in industry as a

result.

 With the high strength and stiffness of PLA, the blends will find some improvement in

these areas, which is where ABS is lacking.

This is the potential impact of developing a well performing blend in this field of research.

Obviously, the economic advantages are very attractive, as many are in search of solutions to many of the world’s similar environmental and sustainable problems. With the proper attention and focus of this project, in addition to utilizing and building from previous works, this project has a high likelihood of success.

Chapter 2: A Criterion for Toughening PLA through Blending: Special Focus on Blending With ABS

______

Abstract

Poly(lactic acid) (PLA) has been hailed as a sustainable material because it is derived from renewable resources. However, it has several performance drawbacks such as low toughness that limit its use. For this reason, many researchers are attempting to blend PLA with traditional polymers to balance resources and properties. Acrylonitrile butadiene styrene (ABS) presents itself as a prime candidate for blending due to complimentary properties to PLA. However, to this point, researchers have yet to develop a commercially viable blend of PLA and ABS.

Problems include high amount of additives required or extra processing steps needed. In this paper, a single criterion for creating a tough PLA/ABS blend is identified from the literature.

Additionally, an analysis of PLA/ABS blending work is given where several strategies are elucidated to compatibilize PLA/ABS blends. Of these strategies, one method is identified as having a high potential to generate a commercially viable blend.

______

2.1 Introduction

With ever increasing evidence of climate change, environmental concerns continue to grow.1–5

The use of petroleum for energy, chemicals, and materials is considered an unsustainable practice, as petroleum reserves are not infinite. Additionally, the use of petroleum has led to an increase in atmospheric greenhouse gas levels.1,2 Due to these concerns, a need arises to find alternatives to petroleum based materials. In regards to addressing environmental problems, there are two main areas of focus. First, the aforementioned use of petroleum as a constituent in plastic materials. To address this problem, biobased plastics are used.6–9 The second problem is the landfilling, or disposal of the plastic material. For this, biodegradable plastics can be used.10–

12 Plastics may be biobased, biodegradable, or both. All of which may be referred to as bioplastics. The focus of this work is the use of biobased plastics.

Currently, there are several plastic materials that are derived from renewable resources such as plants.6–8,13 These biobased plastics have increased sustainability over their petroleum based counterparts because of their carbon source. Since the carbon in the plant material originated in the atmosphere, the release of this carbon does not create a net increase in the atmosphere.

Additionally, plants are easily grown in relatively short timeframes, reducing the concern over depletion of resources. Of all biobased plastics, poly(lactic acid) (PLA) is at the forefront commercially, due to (1) price, (2) strength and stiffness, (3) biobased content, (4) biodegradability, and recyclability.14–26 However, PLA is very brittle, which is discerned through measurements of impact strength and elongation at break. Additionally, the heat deflection temperature (HDT), which is the temperature that it begins to distort under mild loading, is very low14,16–18,20,21,25. These drawbacks have rendered PLA useful for niche or specific applications only.

Attempting to use biobased plastics for many applications, such as durable, would involve a vast jump in the research progress of increasing their performance. Instead, a better approach is to develop incremental increases in the biobased content, while keeping properties high and balanced. This increases the likelihood of developing a material with good properties, yet still offsets some of the petroleum use. This can be done through blending of a biobased plastic with a petroleum based plastic. One main target petroleum based material, which is used widely in several fields, such as electronic and automotive, is acrylonitrile butadiene styrene (ABS).27–30

This is due to the fact that ABS divides the line between commodity and engineering plastics. It has strength and toughness that allow it to be used for semi-structural applications, but has a cost that is much less than many other plastics that can be used for these applications.27–30

Additionally, because of its terpolymer structure, the ratio of acrylonitrile, butadiene, and styrene can be altered to tailor properties to specific applications.28 The toughness of ABS originates largely from the butadiene rubber content.28

Blends of PLA and ABS have been investigated in several studies.22,23,31–34 The two plastics create an immiscible polymer blend with poor dispersion, poor adhesion between phases, and unstable morphology. This indicates that the tension between them is high, creating unfavourable thermodynamics of blending. This results in a blend with very poor performance, most notably impact strength. Thus, much of the research done on blending PLA and ABS has been on trying to use additives to compatibilize the plastics together. Generally, the properties of interest in these works are the impact strength and elongation at break, since they are severely decreased in blends that are not compatible. Thus, the aim of many works is to toughen the blend through the use of toughening agents, such as rubbers. There have been several attempts at toughening PLA with ABS in the literature, with varying degrees of success. In all cases, PLA/ABS blends devoid

of additives are not miscible, and produce very poor performing blends. Thus, the aim of these works are to toughen the blends using blending chemistry, or through a third mutually compatible component. Li et al. incorporated reactive styrene/acrylonitrile/glycidyl methacrylate copolymer (SAN-GMA) alongside ethyltriphenyl phosphonium bromide (ETPB) as a catalyst.32

They found a modest increase in film impact strength in their best blend. They found that the use of SAN-GMA and ETPB caused the ABS phase domain size to decrease, and for the size distribution to become narrower, creating a compatibilized blend. Sun et al. prepared ABS grafted GMA (ABS-GMA) particles to blend with PLA, which produced very high toughness with high butadiene content.34 Jo et al. used SAN-GMA at 20 wt. % loading along with a heat stabilizer to increase the impact strength to 158 J/m. They concluded that the high rubber content

ABS and the heat stabilizer enhanced the dispersion to increase the impact strength.31 Dong et al. synthesized a reactive comb composed of polymethyl methacrylate with epoxy group dispersed throughout the structure.23 When used as a compatibilizer, the reactive comb polymer increased performance of the blend by improving interfacial adhesion.

A crucial step in creating a blend with improved performance is to understand the failure mechanisms that lead to rupture. During an impact strength test, PLA may undergo one of two types of failure, depending on its crystallinity.35,36 Amorphous PLA tends to undergo multiple craze formation, while crystalline PLA tends to undergoes brittle crack initiation without stress relaxation. Understanding these changes in failure mechanisms may have practical value in blending PLA to create a tough blend with ABS.37,38

In attempting to toughen plastics with rubbers, a generalized toughening criterion has been proposed by Wu.39 This work gives the evidence that the most important element in creating a tough plastic through rubber blending is the ligament thickness of the matrix phase, rather than

the rubber particles. This may be applicable to PLA/ABS blends in which ABS acts as the rubber, while PLA as the matrix. From deep analysis of the literature, it can be found that there are only a few strategies of creating a tough blend, many of which satisfy the critical matrix ligament thickness theory of Wu.39

2.2 PLA

2.2.1 Structure & Synthesis

The structure of poly(lactic acid) (PLA) is that of an aliphatic polyester.40–44 The building block of PLA is lactic acid, which is an alpha hydroxyl acid, with both a hydroxyl and carbonyl group.40,41 Lactic acid has many applications in the food industry, pharmaceuticals, cosmetics, and as a polymer precursor.41,45–50 In the food industry, which accounts for a majority of lactic acid applications, uses include buffering agents, acidic flavouring agents, acidulants, and bacterial inhibitors. Lactic acid was first refined from sour milk, but is found in other biological systems, including in the human body. There are two enantiomers, L- and D- lactic acid. The L- lactic acid monomer is produced in the body, while D-lactic acid is made in bacterial biological processes.40 Commercially, lactic acid is manufactured through a fermentation process of carbohydrates. This fermentation process uses lactobacillus as the bacterial agent, which is efficient in the fermentation of glucose, maltose, or dextrose.50 To manufacture PLA, lactic acid is synthesized through the fermentation of dextrose, which is obtained from corn, making PLA an entirely biobased polymer.

Figure 2.1 – Ring opening polymerization of PLA reprinted with permission from reference [40]

There are two different routes for making PLA, condensation or ring-opening polymerization, shown in Figure 2.1.40 Lactic acid can be polymerized via condensation into relatively low molecular weight polymer. The presence of water and other impurities in addition to the ‘back biting’ equilibrium reaction limit the molecular weight that can be achieved.40 The second route for PLA synthesis is through the purification and ring-opening polymerization of six-member ring lactide. PLA can undergo ring-opening polymerization with cationic initiators, initiated by cleavage of alkyl group oxygen bond. This was method was developed by Dupont® in 1954.40,51–

54 The PLA chains can then have hydroxyl or carbonyl end groups. In addition, there are three chiral structures that result, poly(L-lactic acid), poly(D-lactic acid), and poly(meso-lactic acid).40,51–54 . These chiral structures are given in Figure 2.2. More attention on these structures will be paid in proceeding sections.

Figure 2.2 – Chirality of PLA reprinted with permission from reference [40]

Hydrolytic degradation of PLA commences at the ester bond.26,55,56 Under compostable conditions, the PLA can degrade within certain timeframes, making it classified as a biodegradable polymer.56 PLA is also susceptible to thermal degradation.55,56 This process is dependent on the time, temperature, and presence of low molecular weight impurities and catalysts. Obviously, this is a concern for processing PLA as thermal degradation can lead to decrease in the performance of the polymer, and premature failure.

Due to the presence of two chiral carbon centers, the lactide ring precursor to PLA can manifest as one of three isomers (L-, D-, or meso-lactide). The type of poly(lactic acid) that is produced is controlled by polymerization with one or more of these isomers. The stereochemical architecture affects the many performance aspects of the material, such as speed and degree of crystallinity, mechanical properties, and processing temperatures. Commercial grades of poly(lactic acid)

generally have melting points around 170°C to 180°C. This is due to slight racemization and impurities found in the commercial grades, as the melting point of pure poly(L-lactide) or poly(D-lactide) is around 207°C. It has been found that a 1:1 blend of pure poly(L-lactide) with pure poly(D-lactide) results in stereocomplexation, which can exhibit a melting point of 230°C and improved mechanical properties over either pure polymer.

2.2.2 Crystallinity

The crystallization of PLA is important, as it can lead to increased mechanical and thermal properties. Poly(lactic acid) displays a property known as polymorphism, which means that it can display several crystal structures.20,44,57,58 There are two categories of polymorphism, both found in PLA. The first is differing chain conformation, while the second shows the same chain conformation with different special packing arrangements.44,46 Under any given set of conditions, there is only one stable crystal form, while the others are metastable, which means that given enough time, eventually they will transform into the stable arrangement. There are five different forms of PLA crystal structures: α, α’, β, γ, sc (stereocomplex).20,44 The α was the first discovered, but was later found to be one of two α structures. The α’ structure was found to form at temperatures under 120°C. Both the α and the α’ structures are 103 helical. The β was found in hot drawn PLA fibers. The γ is an epitaxial crystal structure, while the sc structure refers to stereocomplexed crystal structures.20,44

The crystallization of PLA has been studied extensively. PLA crystallizes very slowly compared to some commodity plastics, which can lead to increases in production time. PLA crystallizes the fastest between 110-130°C. A study by Vadori et al. investigated the molding conditions of a high-impact grade of PLA available from NatureWorks, LLC.59 They found that with changes in the mold temperature from 30-90°C, there were dramatic changes seen not only in the

crystallinity, but also in the mechanical properties. It was found that the modulus and strength increase with increasing crystallinity. The modulus of PLA is well modelled by the Tsai-Halpin equation. The thermal resistivity of amorphous PLA is lowest. This has been found by several researchers, where increases in crystallinity have been correlated to increases in the heat deflection temperature (HDT).60 Increasing the crystallinity of PLA also has a positive effect on the barrier properties of the polymer. The crystallinity can be controlled through annealing processes, nucleating agents, flow field, and plasticizers.35,36,61

2.2.3 Properties & Processing Techniques

Commercial grades of PLA have excellent strength and stiffness compared to commodity polymers.46,62 However, the main drawback of using PLA for most applications is its poor toughness, manifested in impact strength and elongation at break properties. The β transition temperature of PLA is higher than room temperature (approximately 45°C), meaning that under normal conditions, PLA is a glassy polymer and is very brittle. PLA has low melt strength and viscosity, making difficult for some processing techniques.21,63

PLA can be extruded and injection molded under normal conditions, which is ideal, as these traditional manufacturing techniques are the simplest, fastest, and most economical means to produce a plastic part. The thermal properties of PLA for different –L and –D ratios are given in

Table 2.2. Generally, PLA is better used for melt processing over other processing techniques such as thermoforming, although these can also be achieved.21,63

Table 2.2 – Thermal Properties of PLA –L and –D copolymers reprinted with permission from reference [21]

Glass Poly(L-lactide) to Transition Melting Poly(D-lactide) Temperature Temperature Copolymer Ratio (°C) (°C) 100/0 63 178 95/5 59 164 90/10 56 150 85/15 56 140 80/20 56 125

2.3 ABS

2.3.1 Structure and Synthesis

Few thermoplastic polymers have as much commercial success as acrylonitrile butadiene styrene

(ABS). It has been used for an assortment of applications, largely because of its performance, cost, and compositional flexibility.64 ABS divides the line between engineering polymers and commodity polymer in both price and performance. Thus, when an application calls for semi- structural rigidity, ABS is a prime candidate over commodity plastics for performance, and over engineering plastics for cost. In addition, there is flexibility in the composition and synthesis that allows tailoring of ABS to certain applications.64 In general terms, the acrylonitrile is responsible for chemical resistance and weathering, the butadiene adds rubber toughness, and the styrene increases gloss, processability, and is an economic ingredient. ABS can be tailored for applications that require high and low gloss, high and low viscosity, flame-retardants, high heat, or glass filled.64 This allows customers to tailor ABS to suit the needs of one of a plethora of applications. There are several main types of ABS grades: flame retardant, high heat, chemical resistant, static dissipative, extrusion, among others. The use of ABS is relegated mainly to the

appliance, automotive, and business machines industries, although it is used widely for many applications.64 Table 2.3 shows mechanical properties for an ABS grade compared to a PLA grade.

Table 2.3 – Mechanical properties of PLA and ABS

Tensile Tensile Elongation Flexural Flexural Impact Polymer Strength Modulus at Break Strength Modulus Strength [MPa] [GPa] [%] [MPa] [GPa] [J/m] ABS Magnum 1150 EM 35.5 ± 2.87 1.90 ± 0.01 34.2 ± 3.9 57.7 ± 1.08 1.99 ± 0.02 506 ± 24 PLA (3801X) 30.4 ± 0.56 2.69 ± 0.03 131 ± 98 42.3 ± 0.5 2.95 ± 0.06 135 ± 7

The structure of ABS is a terpolymer, but is comprised of polybutadiene (PB) particles grafted to styrene-acrylonitrile (SAN) and dispersed inside a SAN matrix .65 Generally, ABS is considered a two-phase polymer. ABS can be prepared by either emulsion polymerization, or mass polymerization. The resulting ABS blend can vary in several ways, from the size and distribution of the PB particles, the amount of grafted SAN, the styrene to acrylonitrile ratio, molecular weight, and crosslink density of the rubber, etc.

Emulsion polymerization of ABS begins with an emulsion of monomer, polymer droplets, and a surfactant.64 A free radical initiator is added to the emulsion, and the heat is raised to begin the polymerization process. This can be done in a batch, semi-batch, or continuous process.64 The

PB is first prepared, then monomers of styrene and acrylonitrile are polymerized in the presence of PB. This results in the grafting of SAN to the PB. Emulsion polymerization allows the control of the copolymer composition and the particle morphology.64

In mass polymerization, the uncrosslinked PB is dissolved in styrene and acrylonitrile monomers.64 A peroxide initiator is used to begin the polymerization process. The advantage of mass polymerization is that it is a much cleaner process in terms of monomer and solvent recycling.64 Large PB particles can be produced by mass polymerization. These types of ABS are used more for low gloss, food contact (no residual monomer remains after the process). Due to the need to dissolve the PB phase, there is a limit on the amount of PB that can be in the resulting

ABS, between 15-18 wt. %.64

2.3.2 Properties

The properties of ABS are considered to be better than commodity polymers.64 However, when compared with engineering plastics, there are a few performance issues. For these reasons, ABS is planted firmly between the two, both in terms of properties and in terms of price. Often, ABS is blended with engineering plastics to provide specific properties, such as low temperature impact strength, or to reduce cost. The mechanical properties of ABS can change by a relatively large amount, as synthesis, composition and structure changes (Table 2.4). Generally, ABS has a tensile strength in the range of 30-40 MPa, while stiffness is in the range of 1.5-2.0 GPa, impact strength can vary drastically from approximately 100 J/m to over 500 J/m for high impact strength grades.64,66–68 The heat deflection temperature can range from 80°C to over 110°C for high applications.

Table 2.4 – Mechanical Properties of different ABS grades

Tensile Tensile Elongation Flex Flex Impact Formulation Strength Modulus at Break Strength Modulus Strength (MPa) (GPa) (%) (MPa) (GPa) (J/m) ABS Magnum 35.5 ± 2.9 1.90 ± 0.01 34.2 ± 3.9 57.7 ± 1.08 1.99 ± 0.02 506 ± 24 ABS INJ Starex 43.6 ± 3.4 2.14 ± 0.03 22.6 ± 8.7 54.6 ± 1.43 2.19 ± 0.01 451 ± 22 0167 ABS EXT 34.7 ± 1.2 1.94 ± 0.02 18.6 ± 9.2 53.8 ± 0.86 2.13 ± 0.02 255 ± 4.2 Lustran 752

2.3.3 ABS Toughening

2.3.3.1 Polycarbonate

Polycarbonate (PC) is an engineering plastic that is strong and tough. However, it has been noted

to be susceptible to notch impacts and low-temperature impacts. Additionally, it is more costly

than ABS, and for these reasons, PC and ABS can be blended together to form a polymer with a

balance of properties and cost. On the other hand, blending the two can be viewed as increasing

the performance of ABS for certain applications.

Due to the commercial success of their blends, ABS and PC have been blended extensively in

the literature. Keitz et al. explained the anomaly of compatibility of PC/ABS blends by citing

that the partial miscibility achieves a maximum as a function of the acrylonitrile content in the

ABS.69 Mechanical properties reflected this result, and were best at the same acrylonitrile

content. A binary interaction model was used to explain this behavior of the blends.

2.3.3.2 Poly(butylene terephthalate)

Blends of ABS and poly(butylene terephthalate) have gained commercial interest not only

because of their mechanical properties such as impact strength and modulus, but also due to the

heat, chemical, and abrasion resistance. However, it was shown that good properties can only be

achieved through very narrow processing ranges, due to the fact that the dispersion of ABS in the

PBT coarsens under low-shear conditions. Hale et al. addressed this issue through the use of methyl methacrylate-glycidyl methacrylate-ethyl acrylate terpolymer compatiblizers.70,71 The compatibilizers improved dispersion, even in conditions of low shear.

2.3.3.3 Nylon

Although nylon is an attractive engineering polymer, there are several issues associated with its use for certain applications. Nylon has been known to be brittle, have high moisture sorption, low dimensional stability, and only mid-range heat deflection temperature. Therefore, there has been some interest in blending with ABS to increase these properties. Initial blending experiments revealed a poorly performing blend unless compatibilized. Triacca et al. used styrene-maleic anhydride SMA, with MA content of 25% to reactively compatibilize the blends and vastly improve toughness.72

2.3.3.4 Poly(vinyl chloride)

Blends of ABS with poly(vinyl chloride) (PVC) were prepared and studied by Maiti et al.73 The authors melt blended PVC with ABS of different butadiene contents to study the efficiency of toughening of the butadiene phase. ABS with 36.5% rubber was used, and SAN was added to dilute the amount of rubber in the final blend down to 30% and 25% in addition to the original

36.5%. The authors found that the addition of ABS in PVC results in a large improvement in the impact properties of the material. In fact, they concluded that ABS can be used to toughen the

PVC without considerable detriment to strength and stiffness properties. Additionally, the thermal stability was found to increase in the blends as compared to neat PVC. Finally, the polybutadiene amount in the ABS was found to be highly correlated with the impact properties of the blends.

2.3.3.5 Poly(ethylene terephthalate)

ABS was used as a toughening agent for poly(ethylene terephthalate) by Cook et al.74,75 The authors showed that blends consisting of four discrete phases were developed via melt blending.

However, the mechanical properties were still attractive, and the authors compared the impact toughness of the blends with that of commercially available PC/ABS blends.

2.4 General Blending/Toughening Theory

Polymer blending provides a means for creating an economic material with a balance of properties. However, due to compatibility issues, many cases of simple blending results in a poor performing material.76–81 These incompatible blends are termed immiscible, meaning that the thermodynamic conditions of blending do not promote the morphology or adhesion required.

Even in blends that can satisfy these conditions, there are generally only ‘windows’ of miscibility in which several factors allow proper mixing.79,80 This being said, there have been over 1000 cases of polymer miscibility found, but they are the exceptions rather than the rule itself.

There is some significant interpretation that is involved with the detection of miscibility.

Although many adhere to searching for changes in the glass transition temperature (Tg), it has been shown that the Tg is sensitive to the degree of dispersion, not thermodynamic miscibility.79,80 This is a rather indirect means of testing for miscibility, and in fact it is only a test of correlation to miscibility (dispersion).79 There are many benefits to polymer blending, (1) engineering properties of plastics economically (2) extending performance into other domains

(e.g. increasing notch impact) (3) improving processability (4) quick formulation changes (5) manufacturing flexibility on the plant scale.79 The performance of a blend is dependent on the

ingredients, their concentration, and the morphology. This means that the morphology must be stable (or at least have predictable changes) during forming. The blending of two (or more) polymers involves the mechanical mixing and compatibilization.79

Attempts to compatibilize may be more difficult than expected. Three goals must be achieved in order for compatibilization to be successful (1) reduce the interfacial tension (2) stabilize the morphology, and (3) provide interphasial adhesion in the solid state. It is important to remember that these all need to be satisfied simultaneously. It is a simple process to add a surfactant to decrease tension between the phases (increasing dispersion), but this may result in an even more unstable morphology.80 Some approaches are either to use one compatibilizer that can do all three, or a combination of additives that can achieve the same result. There is some risk-reward associated with using a combination of additives, as their effects could be synergistic, or undermine each other.

There are several strategies to approach compatibilization. Reactive compounding is an often used approach due to potential efficiency in compatibilizing. The modification of one segment can lead to localized miscibility regions. The final approach is to use mechano-chemical blending to create a finer dispersion and compatibilize. Obviously, some of these approaches are set up to be more efficient and more effective than others. For example, a copolymer that migrates to the interface between the two phases would not be expected to create a material with properties far outside the original polymers. However, a reactive species may create entirely new segments, which can dramatically change the properties at a far lower amount.

Helfand and Tagami theory predicts that in binary blends with high molecular weight, the

76–78,82 interfacial thickness, Δl is inversely proportional to the interfacial tension, v∞. For many, it was assumed that the compatibilizer would migrate to the surface, increasing Δl. However,

reports now state that the thickness of the interface is rarely increased. This method also facilitates the tendency for a copolymer to migrate to several different locations.76–78,82 To counteract this, the copolymer should have maximum miscibility with components, molecular weight of each block only slightly higher than the entanglement value and concentration just above ‘critical miscelle concentration’ value (CMC).

2.5 Failure Mechanisms

In order to better approach toughening a polymer blend, a full understanding of the failure mechanisms that each polymer undergoes is necessary. Clearly, the way in which a polymer fails determines not only the extent to which it can be toughened, but also set out a preferential method towards toughening. In the following, the failure mechanisms of both PLA and ABS are discussed.

2.5.1 Poly(lactic acid)

Poly(lactic acid), being a semi-crystalline, has a fairly complex structure in which both the crystalline and amorphous sections play roles in the toughness of the polymer. This is a major complication of studying polymers with a variety of microstructures. What’s more, the arrangement of these structures is dependent on not only inherent polymer characteristics, but also on variables that are unstable and under constant change, such as the strain and thermal history.

The different mechanical properties are affected differently by the changing microstructure of crystallinity.35,36,83 Spherulite size of the crystalline portion is an important factor in changing the mechanical properties. As spherulite size increases, the fracture toughness of semi-crystalline polymers tends to decrease.35 Moderate spherulite sizes tend to accommodate high or peak values of elastic modulus and the yield strength.

Figure 2.3 – Fracture toughness coefficient as a function of crystallinity of PLA reprinted with permission after reference [35]

The critical stress transfer coefficient, KIC, which is a measure of the fracture toughness of a

35 polymer, was measured for PLA by Park et al. They measured changes in the value of KIC as a function of the crystallinity in PLA (Figure 2.3). It was found that as the crystallinity increases up to about Xc=12%, the KIC value decreases, and stays constant up to Xc=50%, then KIC experiences a very drastic drop. They observed multiple crazes at low crystallinity, similar to the behavior of an amporphous polymer.35 However, as the crystallinity increases, the craze density decreases, which would be expected to bring a decrease in the KIC value. However, this does not occur, and the KIC value remains constant. The authors postulated that this could be due to the spherulites in the crack growth region acting to reduce the local stresses, counterbalancing the decrease in the craze density. At Xc=56%, a single straight crack without craze formation was observed under loading. This corresponds to a typical brittle failure, as seen in many semi- crystalline polymers.35 These studies indicate that PLA tends to craze when amorphous, but not when crystalline. Crystalline materials such as PLA tend to deform by crystal-mediated deformation, along with cavitation and fibrillated shearing. Differences in the length scales of the

spherulite structures have very little effect on the mechanism of deformation; however, differences were seen in the bulk properties.35

2.5.2 Acrylonitrile Butadiene Styrene

Donald and Kramer investigated the deformation mechanisms in ABS.84 They found that films of

ABS containing 0.1μm particle size of polybutadiene showed little tendency for crazing. Instead, there was cavitation of rubber particles and localized shear deformation, and little toughening effect. When large particles were added along with the small particles, they provided a nucleation site for crazing to begin, while the smaller particles cavitate without impeding craze growth. Commercial ABS, which has both large and smaller particles, both crazing and shear deformation can be expected to the contribute to the toughenss of the material. The authors admit that there is not clear understanding of why an optimum particle size exists or why it would change from system to sytem.

2.6 Toughening Criterion – Critical Matrix Ligament Thickness Theory

In a 1988 paper, Wu set out a generalized criterion for the toughening of a polymer through blending with a rubber.39 Before that point, the major criterion associated with efficient rubber toughening was thought to be a function of the rubber particles themselves, either their size or dispersion. Wu argued that the focus for toughening is on the matrix, not the particles, and defined a ligament as the region of matrix between two rubber particles. In the study, the impact strength was plotted as a function of the average particle diameter in nylon 6,6/rubber blends for three levels of rubber in the blend, 10%, 15%, and 25% (Figure 2.5).39 With each rubber content, there is an obvious brittle to ductile transition that takes place at different particle sizes. For the

25% rubber content, the brittle to ductile transition occurs at approximately 1.5μm, while at 15% and 10%, the transition occurs at 0.7μm, and 0.4μm, respectively. However, when taking the

same data, and plotting it as impact strength versus the average matrix ligament thickness, all three volume contents converge to show that the brittle to ductile transition occurs at a matrix ligament thickness of 0.2μm, regardless of the amount of rubber in the system (Figure 2.6).

Instead, the amount of rubber only controls the final impact strength, increasing the rubber content only increases the potential for toughening, not the transition to a ductile polymer.39

Figure 2.5 – Impact strength as a function of rubber particle diameter in Nylon/Rubber blends reprinted with permission after reference [39]

Figure 2.6 – Convergence of toughening at the critical ligament size in Nylon/Rubber blends reprinted with permission after reference [39]

Going further, Wu pointed out that there are mainly two types of polymer failure. Pseudoductile failure occurs in polymers that tend to shear yield (e.g. polycarbonate), while brittle polymers tend to craze (e.g. polystyrene). The mechanism for rubber toughening in a pseudoductile polymer occurs in a minimal spacing between particles, below the critical matrix ligament thickness.39 Due to the shear yielding process of pseudoductile polymers, as the blend undergoes impact fracture, the rubber particles tend to cavitate to relieve stresses in the matrix ligament. If the matrix ligament thickness is smaller than the critical thickness, a transition from a plane stress to a plane strain can occur, allowing the ligament to shear yield.39 Thus, for a pseudoductile polymer, the optimal way for rubber toughening to occur would be for the ligament thickness to be minimal. The study noted the effects of polydisperity of the rubber particle sizes, indicating that the average ligament thickness is important. Furthermore, this

theory accounts for the poor performance of poorly dispersed, or flocculated rubber particles.

With higher flocculation, the ligament thickness is increased, making toughening difficult.39

In a brittle polymer, there is a small change to the toughening criterion. As noted by Gilbert and

Donald, large particles act as nucleating sites for crazing.85 As the crazing occurs, if the ligament thickness is above critical, secondary crazes begin to form, and the polymer fails in a brittle fashion. When the ligament thickness is below critical, secondary crazes are not found to occur, and the ligament will yield, making the blend tough.39 Thus a second criterion is installed for a brittle matrix, the ligament thickness must be below critical, and the particle sizes must be above the minimum diameter for primary craze initiation. At constant rubber content, these are competing criterion, as the particles reduce in size, the ligament also reduces, but the ability to initiate crazing is reduced. Thus, there is an optimal particle size for maximizing the toughness of the blend.39

2.7 Strategies For Toughening

The following is a review of the studies published on blending PLA and ABS. These studies are heretofore analyzed critically on their ability to increase the performance of the blends. A well performing blend is judged so based on a balance of the mechanical and thermal properties, but in this case, special attention is given to the toughness of the blends, for two reasons. Firstly, the reason most often cited for blending PLA and ABS is to increase the toughness of PLA, as this is the major lacking quality of neat PLA. Secondly, of the mechanical properties toughness (impact and elongation) is the most sensitive to the definition of compatibility given in previous sections of this paper. In other words, if the interfacial tension is reduces, the morphology stable, and the adhesion between the phases increased, the toughness of the blend will see the largest increase, as this is the most disparate property between PLA and ABS.

In blends of PLA and ABS, it is the contention of the authors that to increase performance in an efficient manner, the morphology would have to substantially manipulated. In reference to the

Wu ligament theory, the PLA phase must allow ABS to disperse to a high degree allowing the

PLA to form thin boundaries around the ABS. With this in mind, the following section has a special focus on the morphology of the blends, commenting on the correlation between it and the mechanical properties of the blends.

2.7.1 Reactive Compatibilization

There are several routes that can be chosen to increase the performance of an incompatible blend.

Outlined in previous sections of this work, the details will not be discussed here. However, they may be summed into one of three categories: 1) passive compatibilization 2) high impact modifier content 3) reactive compatibilization.79,80 Of the three, reactive compatiblization is the best method by which to increase performance of a drastically incompatible blend such as

PLA/ABS. This will be further shown in the following section with a discussion of previous work in the literature.

Li et al. attempted the compatibilization of PLA and ABS in an effort to increase the toughness of PLA, while not sacrificing the strength and stiffness.32 To do this, they used SAN-GMA as a reactive compatibilizer to act as an intermediary between the PLA and ABS phases. The idea is that SAN is compatible with ABS, while the GMA can react with the PLA, increasing the adhesion of the phases. Ethyltriphosphonium Bromide was used as a catalyst in the reaction. In total, the majority of blends used approximately 5 wt. % additives. The reaction, which was shown to occur, brought about only moderate changes in the mechanical properties. The best blend, containing a 1:1 ratio of PLA and ABS with 5 wt. % SAN-GMA results in an impact

strength that is still well below the rule of mixtures values expected with a well compatibilized blend.

Figure 2.6 – Morphology of PLA/ABS blends with SAN-GMA compatibilizer by Li et al reprinted with permission after reference [32]

The SAN-GMA is used as the compatibilization agent where the epoxide groups of the GMA are likely to react with the COOH and OH (carbonyl and hydroxyl) groups at the end of the PLA chains, forming an in-situ compatibilizer. The FT-IR results show that this epoxide reaction does

in fact occur, meaning the SAN-GMA works as intended. The morphology development, as seen through the SEM and TEM images (Figure 2.6), shows the improvement in dispersion with the addition of the compatiblizer. In the blends of PLA and ABS alone, the immiscibility of the blend is evident. ABS domains in the neat blends range from 1-10µm. In the 1:1 PLA:ABS blend, the ABS domain undergoes agglomeration in the poor mixing regime of PLA/ABS. Due to the viscosity differences of PLA and ABS, the PLA remains the matrix and the ABS a dispersion at a 1:1 mixing level. This effect of viscosity difference also appears to be affecting the blending. Once the compatibilizer is present, the morphology shows a finer dispersion. The addition of 5 wt. % of the SAN-GMA reduces the ABS domain size.32 However, the modest improvement in the mechanical properties suggests that a transition from brittle to ductile did not occur. The compatibilizer was not efficient enough at the loading level used to reduce the PLA ligament size enough to undergo this transition and improve performance. However, if the reaction was more efficient in creating an optimal and stable dispersion, this would be an effective approach.

The work of Li et al. is an example of the reactive compatibilization approach. In this case, however, the additives are not very effective at the loading levels used. The SAN-GMA, although does increase the compatibility of the polymers, does so inefficiently, and for a high toughening effect, a high content of SAN-GMA would be necessary.

Jo et al. also used SAN-GMA as a compatibilizer between ABS and PLA.31 However, the authors identified the issue of low PLA thermal stability at temperatures required for blending.

To solve this issue, they used Songsorb 3270 from Songwon Industries as a heat stabilizer.

Although the impact strength did change by a high amount, the final value was still far below the expected value via rule-of-mixture analysis. The authors suggest that the heightened

compatibility is evident from the lack of definitive boundary between the phases under SEM study. Much of the credit was given to the addition of the heat stabilizer for decreasing the degradation of the PLA during blending. There isn’t evidence that the brittle to ductile transition has occurred, because although the impact strength has increased, the amount of PLA has fairly low in these blends (under 50 wt. %) and the impact strength value is still well below rule-of- mixture expectations.

Choe et al. made blends of PLA and ABS for use as a matrix in nanoclay/cellulose composites.22

At most, the blends contained 30 wt. % PLA, while the additive content was relatively high. In some cases, coupling agents such as g-ABS and SAN-g-ABS were used in amounts above 40 wt.

%. The properties of the blends were relatively good, with high toughness, but the strength was sacrificed, and the stiffness remained below the authors target value of 2 GPa for all unfilled blends. SEM studies showed a finer dispersion, causing a brittle to ductile transition to occur, shown in Figure 2.7. However, the toughness comes at the cost of strength and stiffness.

Figure 2.7 – Morphology of PLA/ABS blends from Choe et al. reprinted with permission after reference [22]

Recently, Vadori et al. have used reactive compatibilization to increase the properties of

PLA/ABS blends containing at least 50 wt. % PLA.67 In this study, two commercial additives were used in very low loading levels. The additives used were a highly reactive chain extender under the tradename Joncryl ADR 4368 from BASF and a acrylic copolymer under the trade name Biostrength 900 from Arkema. The content of additives in the blend was restricted to lower than 4 wt. % with the idea of making blends with increased commercial viability. The authors showed that each additive had only a modest effect on both the mechanical properties and the morphology of the blends. However, when used together, the additives worked synergistically to completely change the morphology. The authors describe the phenomenon as both additives working in opposite ways to achieve the same goal. Through both AFM and SEM images, it is shown that there is in fact a complete change in the dispersion, and the morphology becomes a nanostructured assembly of extremely thin ligaments of PLA surrounding a fine dispersion of

ABS. In fact, the PLA phase was measured to be as thin as 14 nm in some areas. This change in morphology has doubtlessly led to the vast improvement in the toughness of the blends. At the same time, the strength and stiffness are not sacrificed, and are well above values found for neat

ABS grades, while the impact strength and elongation at break increase drastically to values above the rule of mixture values.

2.7.2 Reactive Polymer Synthesis

Sun et al. prepared ABS-g-GMA particles, where GMA was successfully grafted onto the butadiene phase of the ABS.34 The analysis indicated that the epoxide groups of the GMA reacted with COOH and OH of the PLA end chains. The reaction was effective as a compatibilizer between the PLA and ABS, where the blends underwent the transition from brittle to ductile, showing vastly improved toughness values. This is confirmed through SEM, where the morphology shows a fine dispersion of ABS in the PLA, shown in Figure 2.9.

Although the work by Sun et al. was successful in increasing the toughness of the blends, this may not be a commercially viable method of creating tough PLA/ABS blends for a number of reasons. The authors synthesized their own compatibilizer by grafting the GMA onto the butadiene phase of the ABS. This is not a commercially viable method, as this compatibilizer is not already commercially available. Additionally, incorporation of a high rubber content in the blends may be damaging to the tensile strength and modulus properties of the blends. These values are not reported in this work, making it difficult to draw conclusions on the effect of the

ABS-g-GMA particles on the strength and stiffness of the blends.

Figure 2.9 – Morphology of blends of PLA/ABS with a synthesized compatibilizer reprinted with permission after reference [34]

Recently, Dong et al. have synthesized reactive comb polymers from PMMA and GMA to create a backbone of PMMA with two side PMMA chains, and epoxide groups from the GMA distributed randomly along the backbone.23 When blended with PLA and ABS, the epoxide groups were expected to react with the carboxyl groups on the end of the PLA chain, grafting

PLA chains on the PMMA backbone to create an in-situ compatibilizer. The reactive comb polymer was used at a 3 wt. % loading level. In these blends, a brittle to ductile transition seemed to have occurred, with dispersion becoming extremely fine (Figure 2.10), with excellent mixing between the PLA and ABS. This resulted in drastic improvements in the toughness values of the blends. Impact strength values were not reported.

Figure 2.10 – Morphology of PLA/ABS blends compatibilized by a reactive comb polymer by

Dong et al. reprinted with permission after reference [23]

2.7.3 Processing Steps

Oyama et al. blended PLA with ethylene-glycidyl methacrylate copolymer (EGMA) to increase toughness.33 The blends only contained these two components, but the authors changed the type of PLA used from low to high molecular weight in addition to the loading level. Both 5 wt. % and 20 wt. % of the EGMA copolymer were used with both high and low molecular weight PLA.

The authors found that blending with a screw speed at 200 RPM was much more effective at creating a finer dispersion than 30 RPM, noting that the added shear stress of the increased speed aided in blending the two phases. The morphology was studied through SEM. The authors attempted to anneal the blends to increase the crystalline content. Prior to annealing, there is only a modest increase in the toughness of the blends, despite the aforementioned finer dispersion.

However, after annealing, the impact strength increased dramatically. This aligns with the Wu ligament theory very well. Once crystallized, the PLA undergoes a shear-yield type failure

(although brittle), while amorphous PLA tends to craze. It is possible that the dispersion was too fine to allow fast nucleation of crazes under the impact loading scheme, causing the PLA to remain fairly brittle despite the improvement in dispersion. However, because a shear-yield type polymer requires minimum size dispersion, the blend exhibits a very tough behavior once crystallized.

2.7.4 Strategy Outlook

As previously mentioned, the goal of this work is to extricate the methods for making a commercially viable blend from PLA and ABS. The idea of a commercially viable blend is defined by the characteristics that define other commercial blends. First and foremost, the cost of the blend must be similar to the cost of similarly performing materials. The idea of blending the two, for the purposes of the authors, is to create an alternative for ABS. Thus, the blend must not

only perform as similar or better than normal ABS grades, but must have a similar cost. For this

reason, pre-compounding chemistry such as grafting, synthesis or surface chemistry tends to

reduce the viability of creating a commercial blend. This is because the constituents can be

costly. Additionally, the preparation steps involved increases the processing steps in a

commercial setting, and therefore increases the cost of the material. For the same reason, extra

processing steps such as annealing, or even increasing cycle times can dramatically reduce the

attractiveness of the blends. The impact strength values of notable PLA/ABS blends in the

literature are given in Table 2.5.

Table 2.4 – Notable PLA/ABS blends in the literature

Additive Impact Impact PLA ABS Additive(s) content test Strength Film Li et al.32 47.5 47.5 SAN-GMA & ETPB ~5 162 Impact Notched Jo et al.31 40 40 SAN-GMA & Songsorb ~20 158 Izod

g-ABS, SAN-GMA & Notched Choe et al.22 30 26.7 43.3 508 DCP Izod

Vadori et Acrylic Copolymer & Notched 50 48 2 230 al.67 Chain Extender Izod

Notched Sun et al.34 70 / ABS-g-GMA 30 550 Izod

In the methods for creating a tough blend of PLA and ABS details in the previous sections, one

stands above as having the highest likelihood of producing a commercially viable blend. This is

the reactive compatibilization method. If the constituent additives are already commercially

available (at a low cost), used in low amounts, and are efficient enough, the reactive

compatibilization method is the one that is best for creating a commercially viable blend.

2.8 Philosophy of Experimentation

The preceding sections offer a knowledge gap in the area of PLA/ABS blends. To this point, the development of a PLA/ABS blend that is economically and commercially viable has not been demonstrated. Additionally, a route to creating a blend that satisfies this criterion has been identified. Looking at the works that have succesfully improved the properties of the blend, many share a similar thread. The use of a functional polymer with an epoxide group, such as that found on glycidyl methacrylate (GMA) has been shown to increase dispersion, giving better morphology, and ultimately better performance. Thus, the strategy for this work is to identify similar commercial functional additives that contain this group, preferably in high densities. It is hypothesized that the higher the density of epoxide groups in the functional polymer, the higher the effectiveness of improving performance. It is possible that further increases in the effectiveness of the additives are required, thus additional additives may be used to facilitate this.

Additionally, the grades of the polymers are important. For this study, a high impact injection grade of ABS and a general injection grade of PLA have been chosen. This increases commercial viability for the blend. The low melt flow index (MFI) of ABS offsets the high MFI value of the

PLA used. The PLA is a common injection grade, which also was chosen for commercial viability of the final blend.

2.9 Conclusion

PLA is a versatile polymer with the advantages of being biodegradable and biobased. It is useful to note that PLA can be processed on traditional polymer processing equipment. It has received growing attention recently as its price has reduced drastically in the last decade. The inherent disadvantages of PLA are brittle mechanical properties and low thermal stability. To reconcile this, many researchers are attempting to blend PLA with tough polymers as a cost effective

means of increasing the toughness of PLA. ABS is considered a prime candidate for blending with PLA as it is a heavily used plastic in many application fields, and its flexibility in structure allows very tough grades available, which complements PLA well in a blending scenario. To date, most researchers have attempted compatibilization through reactive blending. This has shown promise as the method with the most ability to change a PLA/ABS blend from brittle to ductile by reactively changing morphology to decrease ligament size. However, no reactive compatibilizer has been discovered that can efficiently accomplish this at a low additive loading or without the need for costly extra processing or synthesis. The exception to this is Vadori et al. who have used less than 4 wt. % commercially available additives. These additives work synergistically to drastically reduce the PLA ligament to an extremely small size. This has drastically increased the toughness of the blend, while keeping the strength and stiffness higher than that of neat ABS grades. The result is a commercially viable blend in terms of cost and processing.

2.10 References

1. Schuur, E.A.G., McGuire, A.D., Schädel, C., Grosse, G., Harden, J.W., Hayes, D.J.,

Hugelius, G., Koven, C.D., Kuhry, P., Lawrence, D.M. and Natali, S.M. Climate change and the permafrost carbon feedback. Nature. 2015;520(7546):171-179.

2. Schuur, E.A., Bockheim, J., Canadell, J.G., Euskirchen, E., Field, C.B., Goryachkin,

S.V., Hagemann, S., Kuhry, P., Lafleur, P.M., Lee, H. and Mazhitova, G. Vulnerability of

Permafrost Carbon to Climate Change: Implications for the Global Carbon Cycle. Bioscience.

2008;58(8):701.

3. Whiteman G, Hope C, Wadhams P. Climate science: Vast costs of Arctic change. Nature.

2013 499(7459):401-403.

4. O’Hare M, Plevin RJ, Martin JI, et al. Proper accounting for time increases crop-based biofuels’ greenhouse gas deficit versus petroleum. Environ Res Lett. 2009;4(2):024001.

5. Khan SA, Rashmi, Hussain MZ, Prasad S, Banerjee UC. Prospects of biodiesel production from microalgae in India. Renew Sustain Energy Rev. 2009;13(9):2361-2372.

6. Street W, Lieberman N, Muhlbauer WK. Handbook of Bioplastics and Biocomposites

Engineering Applications.

7. Pilla S. Engineering Applications of Bioplastics and Biocomposites - An Overview. :1-

15. Handbook of Bioplastics and Biocomposites Engineering Applications;

8. Kumar MN, Yaakob Z. Biobased Materials in Food Packaging Applications. Handb

Bioplastics Biocomposites Eng Appl. 2011:121-159.

9. Akil HM, Mazuki AAM. Sustainable biocomposites based for construction applications.

Handb Bioplastics Biocomposites Eng Appl. 2011:285-316.

10. Xiao H, Lu W, Yeh J-T. Crystallization behavior of fully biodegradable poly(lactic acid)/poly(butylene adipate- co -terephthalate) blends. J Appl Polym Sci. 2009;112(6):3754-

3763.

11. Siracusa V, Rocculi P, Romani S, Rosa MD. Biodegradable polymers for food packaging: a review. Trends Food Sci Technol. 2008;19(12):634-643.

12. Zarrinbakhsh N, Misra M, Mohanty AK. Biodegradable Green Composites from

Distiller’s Dried Grains with Solubles (DDGS) and a Polyhydroxy(butyrate-co-valerate)

(PHBV)-Based Bioplastic. Macromol Mater Eng. 2011;296(11):1035-1045.

13. Kumar MNS, Yaakob Z. Biobased Materials in Food Packaging Applications. Handbook of Bioplastics and Biocomposites Engineering Applications (2011): 121-159.

14. Saeidlou S, Huneault MA, Li H, Park CB. Poly(lactic acid) crystallization. Prog Polym

Sci. 2012;37(12):1657-1677.

15. Choe I-J, Lee JH, Yu JH, Yoon J-S. Mechanical properties of acrylonitrile-butadiene- styrene copolymer/poly(l-lactic acid) blends and their composites. J Appl Polym Sci.

2014;131(11)

16. Vadori R. Studies On The Reactive Blending Of Poly(Lactic Acid) And Acrylonitrile

Butadiene Styrene Rubber. M.A.Sc. Thesis. December 2012.

17. Wu C-P, Wang C-C, Chen C-Y. Enhancing the PLA Crystallization Rate and Mechanical

Properties by Melt Blending with Poly(styrene-butadiene-styrene) Copolymer. Polym Plast

Technol Eng. 2015;54(10):1043-1050.

18. Vadori R, Mohanty AK, Misra M. The Effect of Mold Temperature on the Performance of Injection Molded Poly (lactic acid) ‐ Based Bioplastic. Macromol Mater Eng.

2013;298(9):981-990.

19. Lasprilla AJR, Martinez GAR, Lunelli BH, Jardini AL, Filho RM. Poly-lactic acid synthesis for application in biomedical devices - a review. Biotechnol Adv. 2012;30(1):321-328. doi:10.1016/j.biotechadv.2011.06.019.

20. Liu G, Zhang X, Wang D. Tailoring crystallization: towards high-performance poly(lactic acid). Adv Mater. 2014;26(40):6905-6911.

21. Lim L-T, Auras R, Rubino M. Processing technologies for poly(lactic acid). Prog Polym

Sci. 2008;33(8):820-852. doi:10.1016/j.progpolymsci.2008.05.004.

22. Choe I, Lee J, Yu J, Yoon J. Mechanical properties of acrylonitrile–butadiene–styrene copolymer/poly (l ‐ lactic acid) blends and their composites. J Appl Polym. 2014. http://onlinelibrary.wiley.com/doi/10.1002/app.40329/full. Accessed August 16, 2016.

23. Dong, Wenyong, Meifeng He, Hengti Wang, Fanglu Ren, Junqing Zhang, Xuewen Zhao, and Yongjin Li. PLLA/ABS Blends Compatibilized by Reactive Comb Polymers: Double Tg

Depression and Significantly Improved Toughness. ACS Sustain Chem Eng. 2015;3(10):2542-

2550.

24. Al-Itry R, Lamnawar K, Maazouz A. Improvement of thermal stability, rheological and mechanical properties of PLA, PBAT and their blends by reactive extrusion with functionalized epoxy. Polym Degrad Stab. 2012;97(10):1898-1914.

25. Kfoury, G., Raquez, J. M., Hassouna, F., Odent, J., Toniazzo, V., Ruch, D., & Dubois, P.

Recent advances in high performance poly(lactide): from “green” plasticization to super-tough materials via (reactive) compounding. Front Chem. 2013;1:32.

26. Rasselet D, Ruellan A, Guinault A, Miquelard-Garnier G, Sollogoub C, Fayolle B.

Oxidative degradation of polylactide (PLA) and its effects on physical and mechanical properties. Eur Polym J. 2014;50(1):109-116.

27. Bair HE, Boyle DJ, Kelleher PG. The effects of light and heat on the rubber content and impact strength of acrylonitrile‐butadiene‐styrene. Polym Eng Sci. 1980;20(15):995-1001.

28. Kulich, D. M., Gaggar, S. K., Lowry, V., & Stepien, R. Acrylonitrile-Butadiene-Styrene

Polymers. In: Encyclopedia of Polymer Science and Technology. Vol Hoboken, NJ, USA: John

Wiley & Sons, Inc.; 2001.

29. Choi J-H, Ryu J-H, Kim SY. Linear viscoelastic behavior of acrylonitrile-butadiene- styrene (ABS) polymers in the melt: Interpretation of data with a linear viscoelastic model of matrix/core-shell modifier polymer blends. Korea-Australia Rheol J. 2000;12(2):135-141.

30. Bai X, Isaac DH, Smith K. Reprocessing acrylonitrile–butadiene–styrene plastics:

Structure–property relationships. Polym Eng Sci. 2007;47(2):120-130.

31. Jo M, Ryu Y, Ko J, Yoon J. Effects of compatibilizers on the mechanical properties of

ABS/PLA composites. J Appl Polym. 2012. 125(S2).

32. Li Y, Shimizu H. Improvement in toughness of poly(l-lactide) (PLLA) through reactive blending with acrylonitrile–butadiene–styrene copolymer (ABS): Morphology and properties.

Eur Polym J. 2009;45(3):738-746.

33. Oyama HT. Super-tough poly(lactic acid) materials: Reactive blending with ethylene copolymer. Polymer. 2009;50(3):747-751.

34. Sun S, Zhang M, Zhang H. Polylactide toughening with epoxy‐functionalized grafted acrylonitrile–butadiene–styrene particles. J Appl Polym. 2011;122(5):2992-9.

35. Park SD, Todo M, Arakawa K, Koganemaru M.

Effects of isothermal crystallization on fracture toughness and crack growth behavior of poly

(lactic acid) Journal of materials science, 40(4), 1055-1058.

36. Renouf-Glauser AC, Rose J, Farrar DF, Cameron RE. The effect of crystallinity on the deformation mechanism and bulk mechanical properties of PLLA. Biomaterials.

2005;26(29):5771-5782. doi:10.1016/j.biomaterials.2005.03.002.

37. Kausch HH, ed. Crazing in Polymers. Vol 52-53. Berlin/Heidelberg: Springer-Verlag;

1983. doi:10.1007/BFb0024054.

38. Bubeck R, Jr DB, Kramer E. Modes of deformation in rubber-modified thermoplastics during tensile impact. J Mater. 1991;26(23):6249-59.

39. Wu S. A generalized criterion for rubber toughening: The critical matrix ligament thickness. J Appl Polym Sci. 1988;35(2):549-561.

40. Garlotta D. A Literature Review of Poly(Lactic Acid). J Polym Environ. 2001;9(2):63-

84.

41. Groot W, van Krieken J, Sliekersl O, de Vos S. Production and Purification of Lactic

Acid and Lactide. In: Poly(Lactic Acid). Vol Hoboken, NJ, USA: John Wiley & Sons, Inc.;

2010;1-18.

42. Lim L-T, Auras R, Rubino M. Processing technologies for poly(lactic acid). Prog Polym

Sci. 2008;33(8):820-852.

43. Rasal RM, Janorkar A V., Hirt DE. Poly(lactic acid) modifications. Prog Polym Sci.

2010;35(3):338-356.

44. Saeidlou S, Huneault MA, Li H, Park CB. Poly(lactic acid) crystallization. Prog Polym

Sci. 2012;37(12):1657-1677.

45. Alsaheb, R. A. A., Aladdin, A., Othman, N. Z., Malek, R. A., Leng, O. M., Aziz, R., & El

Enshasy, H. A. Recent applications of polylactic acid in pharmaceutical and medical industries. J

Chem Pharm Res. 2015;7(12):51-63.

46. Gupta a, Kumar V. New emerging trends in synthetic biodegradable polymers –

Polylactide: A critique. Eur Polym J. 2007;43(10):4053-4074.

47. Li S, McCarthy S. Influence of Crystallinity and Stereochemistry on the Enzymatic

Degradation of Poly(lactide)s. Macromolecules. 1999;32(13):4454-4456.

48. Tsuji H, Tezuka Y. Stereocomplex Formation between Enantiomeric Poly(lactic acid)s.

12. Spherulite Growth of Low-Molecular-Weight Poly(lactic acid)s from the Melt.

Biomacromolecules. 2004;5(4):1181-1186.

49. Dusselier, M., Van Wouwe, P., Dewaele, A., Makshina, E., & Sels, B. F. Lactic acid as a platform chemical in the biobased economy: the role of chemocatalysis. Energy Environ Sci.

2013;6(5):1415.

50. John RP, Nampoothiri KM, Pandey A. Fermentative production of lactic acid from biomass: an overview on process developments and future perspectives. Appl Microbiol

Biotechnol. 2007;74(3):524-534.

51. Auras R, Lim L, Selke S, Tsuji H. Poly (Lactic Acid): Synthesis, Structures, Properties,

Processing, and Applications.; Vol. 10. John Wiley & Sons, 2011.

52. Avérous L. Polylactic Acid: Synthesis, Properties and Applications.; Elsevier: Oxford,

UK, 2008.

53. Mehta R, Kumar V, Bhunia H, Upadhyay SN. Synthesis of poly (lactic acid): a review.

Journal of Macromolecular Science, Part C: Polymer Reviews. 2005;45(4):325-49.

54. Auras R, Lim L, Selke S, Tsuji H. Poly (Lactic Acid): Synthesis, Structures, Properties,

Processing, and Applications.; Vol. 10. John Wiley & Sons, 2011.

55. Proikakis CS, Mamouzelos NJ, Tarantili PA, Andreopoulos AG. Swelling and hydrolytic degradation of poly(d,l-lactic acid) in aqueous solutions. Polym Degrad Stab. 2006;91(3):614-

619.

56. Iwata T, Abe H, Kikkawa Y. Enzymatic Degradation. In: Poly(Lactic Acid). Vol

Hoboken, NJ, USA: John Wiley & Sons, Inc.; 2010:383-399.

57. Fambri L, Migliaresi C. Crystallization and Thermal Properties. In: Poly(Lactic Acid).

Vol Hoboken, NJ, USA: John Wiley & Sons, Inc.; 2010:113-124.

58. Sarasua, J. R., Prud'Homme, R. E., Wisniewski, M., Le Borgne, A., & Spassky, N.

Crystallization and Melting Behavior of Poly ( l -lactic Acid). Macromolecules.

2007;40(26):9463-9469.

59. Vadori R, Mohanty AK, Misra M. The Effect of Mold Temperature on the Performance of Injection Molded Poly (lactic acid) ‐ Based Bioplastic. Macromol Mater Eng.

2013;298(9):981-990.

60. Jain S, Reddy MM, Mohanty AK, Misra M, Ghosh AK. A New Biodegradable Flexible

Composite Sheet from Poly(lactic acid)/Poly(ε-caprolactone) Blends and Micro-Talc. Macromol

Mater Eng. 2010;295(8):750-762.

61. Park SD, Todo M, Arakawa K. Effect of Annealing on Fracture Mechanism of

Biodegradable Poly(lactic acid). Key Eng Mater. 2004;261-263:105-110.

62. Nanda MR, Misra M, Mohanty AK. The Effects of Process Engineering on the

Performance of PLA and PHBV Blends. Macromol Mater Eng. March 2011:n/a - n/a.

63. Auras RA, Lim L-T, Selke SEM, Tsuji H. Poly (Lactic Acid): Synthesis, Structures,

Properties, Processing, and Applications. Vol 10. John Wiley & Sons; 2011.

64. Kulich DM, Gaggar SK, Lowry V, Stepien R. Acrylonitrile–Butadiene–Styrene

Polymers. Encyclopedia Of Polymer Science and Technology. 1985.

65. Hale WR, Pessan LA, Keskkula H, Paul DR. Effect of compatibilization and ABS type on properties of PBT / ABS blends. Chem Eng. 1999;40:4237-4250.

66. Vadori R. Studies On The Reactive Blending Of Poly(Lactic Acid) And Acrylonitrile

Butadiene Styrene Rubber. M.A.Sc. Thesis. December 2012.

67. Vadori R, Misra M, Mohanty A. Sustainable biobased blends from the reactive extrusion of polylactide and acrylonitrile butadiene styrene. J Appl Polym. 2016.

68. Vadori R, Mohanty A, Misra M. Blending Of Poly (Lactic Acid) And Acrylonitrile

Butadiene Styrene For Use As Biocomposite Matrix. 2012

69. Keitz J, Barlow J, Paul D. Polycarbonate blends with styrene/acrylonitrile copolymers. J

Appl Polym. 1984;29(10):3131-45.

70. Hale W, Pessan L, Keskkula H, Paul D. Effect of compatibilization and ABS type on properties of PBT/ABS blends. Polymer. 1999;40(15):4237-50.

71. Hale W. Effect of PBT melt viscosity on the morphology and mechanical properties of compatibilized and uncompatibilized blends with ABS. Polymer. 40(13):3621-3629.

72. Triacca V, Ziaee S, Barlow J, Keskkula H, Paul D. Reactive compatibilization of blends of nylon 6 and ABS materials. Polymer. 1991;32(8):1401-13.

73. Maiti SN, Saroop UK, Misra A. Studies on polyblends of poly(vinyl chloride) and acrylonitrile-butadiene-styrene terpolymer. Polym Eng Sci. 1992;32(1):27-35.

74. Cook W, Zhang T, Moad G. Morphology–property relationships in ABS/PET blends. I.

Compositional effects. J Appl. 1996; 62(10):1699-708.

75. Cook W, Moad G, Fox B, Deipen G Van. Morphology–property relationships in

ABS/PET blends. II. Influence of processing conditions on structure and properties. J Appl.

1996; 62(10):1709-14.

76. Helfand E. Theory of inhomogeneous polymers: lattice model for polymer–polymer interfaces. J Chem Phys. 1975; 63(5):2192-8.

77. Helfand E, Tagami Y. Theory of the interface between immiscible polymers. J Polym Sci

Part B. 1971; 9(10):741-6.

78. Helfand E, Tagami Y. Theory of the interface between immiscible polymers. II. J Chem

Phys. 1972; 56(7):3592-601.

79. Utracki LA. Polymer Blends Handbook.; 2002. Dordrecht, The Netherlands: Kluwer

Academic Publishers

80. Utracki, Leszek A. Compatibilization of polymer blends. The Canadian Journal of

Chemical Engineering 80, 2002;1008-1016.

81. Utracki, L. A. Introduction to polymer blends. Polymer blends handbook. Springer

Netherlands, 2003. 1-122.

82. Helfand E, Sapse A. Theory of unsymmetric polymer–polymer interfaces. J Chem Phys.

1975; 62(4):1327-31.

83. Arakawa K, Mada T, Park S-D, Todo M. Tensile fracture behavior of a biodegradable polymer, poly(lactic acid). Polym Test. 2006; 25(5):628-634.

84. Donald A, Kramer E. Plastic deformation mechanisms in poly (acrylonitrile-butadiene styrene). J Mater Sci. 1982; 17(6):1765-72.

85. Gilbert D, Donald A. Toughening mechanisms in high impact polystyrene. J Mater Sci.

1986; 21(5):1819-1823.

Chapter 3: Studies on the Reaction of Acrylonitrile Butadiene Styrene to Melt Processing Conditions

______

Abstract

Acrylonitrile butadiene styrene (ABS) was investigated for its reaction to melt processing.

Studies were done in a lab scale micro extruder and injection molder. It was found that during typical processing times, the ABS begins to undergo a hydrogen abstraction reaction in the presence of oxygen in its polybutadiene (PB) phase. This leads to a crosslinking of the PB chains, which in turn can have an effect on the performance of the polymer. The ABS also was shown to undergo chain scission, opposing the crosslinking effect on the viscosity of the polymer. With higher temperatures, it was found that the crosslinking occurred earlier and at a faster rate. Crosslinking preceded chain scission such that the higher temperatures had the highest peak viscosities. Ultimately, chain scission became the dominant mechanism, decreasing the viscosity. Similar results were seen with changing retention time. Higher retention times in the melt increased the viscosity to a point before decreasing again. During processing, it was found that temperature and shearing both have an effect in progressing the reaction.

______

3.1 Introduction

Acrylonitrile Butadiene Styrene (ABS) is used extensively in industry for various applications including in automotive and electronic sectors [1-3]. This is due to a number of factors, not the least of which is its performance to cost ratio. ABS divides the line between commodity and engineering polymers, and has a cost to match. Thus, in applications that require increased performance to commodity polymers, especially in the semi-structural applications found in the electronics and automotive sectors, ABS is a good option. In addition, ABS, which is a terpolymer comprised of acrylonitrile butadiene and styrene, has a high flexibility in its composition by altering the ratios of its three constituents [4, 5]. Thus, ABS can be tailored to specific applications such as high heat resistance, increased toughening, or improved mold ability.

Polymers that are heavily used in industry must be studied intensely in order to predict their performance suitability for each application. In many cases, the product lifetime must be taken into account to investigate how the material will hold up to the environmental stresses that it will be subjected to over time. For instance, materials for the automotive industry must be able to withstand the ultraviolet energy from the sun that will be encountered.

Degradation studies are vitally important in assessing the stability of polymers. Understanding the mechanisms by which polymers degrade, including apparent changes in performance can aid in finding suitable applications or suggesting a possible path for stability improvement [6-13].

Rubber modified polymers are crucially susceptible to such degradation through thermal- oxidative reaction mechanisms. In the literature, acrylonitrile butadiene styrene (ABS) has been identified as having a high susceptibility to this degradation scheme.

However, a major portion of these studies concerned themselves with the behavior of ABS after molding, during the apparent lifetime of the product. Tiganis et al. [14] found that ABS undergoes degradation under accelerated thermal aging. It was concluded that thermal oxidation occurs through hydrogen abstraction in the polybutadiene phase. This led to deterioration of toughness, key marked by a substantial decrease in impact strength. Again, this was a study of the aging of

ABS, and did not seek the effect of melt processing on ABS. Ghaemy and Scott [15] attempted to correlate the decrease in rubber damping peak (tan δ-80) due to crosslinking during processing.

They reported loss in impact strength, but did not go in depth on the effect of processing on the material, nor the reactions that take place.

This has been the extent of studies on the effect of processing conditions on ABS. It is of interest to investigate the behavior of ABS under processing conditions in order to gain a perspective on which processing parameters are important in forming the product. Many polymer processing techniques subject the material to high temperatures and strains, all in an oxygen environment.

Due to this, the material may undergo some degradation. Therefore, in this study, the effect of melt processing on the ABS is investigated, including molecular changes and mechanical properties. It is the aim of this work to outline which process parameters are critical during the melt processing of ABS.

3.2 Experimental

3.2.1 Materials ABS was obtained from Styron under the trade name Magnum 1150 EM. It is a high impact, and high rubber ABS, although exact composition was not disclosed. Undoubtedly, this commercial grade of ABS contains some additives to curb unwanted reactions, which allows an industry relevant study of the melt processing effect on ABS.

3.2.2 Sample Preparation The ABS was processed on both a lab scale and pilot scale extruder. For samples made with the lab scale extruder, a lab scale injection molder was also used. For each individual process used, it will be specified in the results section as they pertain greatly to results. The lab scale microcompounder and injection molder are from DSM and are 15 and 12cc respectively. The pilot scale processing was done on a Leistritz twin screw co-rotating extruder.

3.2.3 Characterization

3.2.3.1 Parallel Plate Rheology A rotational rheometer from Anton Paar was used to measure the flow properties of ABS in the melt after several pre-conditioning treatments. The parameters used will be outlined in the results.

3.2.3.1 Impact Strength The impact strength was studied used a TMI mechanical impact tester, according to the notched

Izod method of impact testing, in accordance with ASTM D256.

3.3 Results And Discussion

The ABS has been subjected to several processing trials, which include different conditions to view the effects of these conditions on the ABS. For a baseline study of the overall behavior of

ABS during a parallel plate test of degradation, the viscosity of neat ABS was measured for the duration of 1 hour with a shear rate of 1 /sec, which has been tested to be inside the linear viscoelastic range of the neat ABS.

3.3.1 Neat ABS

Normally, we would expect an increase in temperature to lead to a decrease in viscosity in a viscoelastic fluid, as the polymer chains have more energy to be lent to their mobility [15]. We have taken neat ABS and subjected it to four one-hour studies, as seen in Figure 3.1. As the moisture of the sample can affect the behavior of the material in such a test, it was dried overnight and tested to contain no more than 0.1% moisture by weight. Viscosity was measured while shear rate, frequency were held constant, at temperatures of 180°C, 200°C, 220°C and

240°C.

Table 3.1 - Impact strength values for various retention times.

1 min 3 min 5 min

Impact Strength (J/m) 514.6 430.9 430.0

Standard Deviation 44.2 50.5 27.5

The initial increase in viscosity can be seen as the rearrangement of the chains in the presence of the rubber particles, which can add friction and thus a resistance to the mobility of the chains.

Aoki et. al [16] found that there was an increase in the storage modulus of molten ABS over time due to the breakup of agglomerated rubber particles. This occurs within the first 30 seconds of the test, and the rest of the changes can be interpreted as chain-reaction functions due to crosslinking and chain-scission of the polymer.

Table 3.2 - Impact strength values for various shear speeds at 220°C and 240°C.

0 RPM 50 RPM 100 RPM 0 RPM 50 RPM 100 RPM

220°C 220°C 220°C 240°C 240°C 240°C

Impact 486.5 472.9 460.9 483.9 481.7 476.2 Strength (J/m)

Standard 34.7 25.5 27.0 32.5 18.4 20.5 Deviation

The difference in slopes of the plots is the effect that temperature has on the initialization mobility of the polymer chains. Thus, 180°C inhibits the mobility of chains to a very high degree causing the apparent increase in the slope of the viscosity curve. The 180°C test has five times the slope than the 220°C, while 200°C has twice the slope than 220°C. This agrees with conventional thought, as the chains with less mobility have a higher resistance to the imposed shearing of the test, and thus a higher viscosity is seen.

1,000,000

100,000 180C

200C Viscosity (Pa.s) Viscosity 220C 240C

10,000 1 10 100 1000 10000

Time (s)

Figure 3.1 - Viscosity evolution over time of ABS at 180°C, 200°C, 220°C and 240°C.

The general trend of the three plots is similar, although the kinetics are changed due to the temperature difference, and the available energy. After initial chain alignment and initialization of mobility, the 200°C and 180°C plots appear to begin to plateau, which they would be expected to. Instead, they begin to crosslink and react, causing the second jump in viscosity. This second jump in viscosity is attributed to a molecular reaction that takes place, and since it is an increase that is seen in viscosity instead of a decrease, it is likely that it is a crosslinking reaction. Several studies [13, 14] have found ABS to be able to undergo this reaction scheme when subjected to

conditions that promote the reaction. This is of interest because it is the first evidence that the material may be crosslinking within the timeframe of a processing cycle. For the purposes of this study, the assumed timeframe for a processing cycle is 2 min. This includes retention time in any extrusion process and molding process.

The 220°C curve does not exhibit this shoulder, but instead begins this rapid increase. This indicates that before the material is able to begin to find a more favorable energy state under this new shear-loading regime it begins the crosslinking reaction.

The inflection point, as located by the 2nd derivative of viscosity with respect to time, signifies

(for the purposes of this study) the transition from a mechanically dominant change in viscosity to a reaction-dominant change in the viscosity. Therefore, the crosslinking reaction has become an important parameter in terms of molecular mechanisms occurring in the material. According to the kinetics from this viscosity curve, under the conditions created in this experiment, the crosslinking reaction occurs within one minute at 220°C, falling within normal processing conditions of the ABS. This is a very low lag time before reaction, considering the very low shearing that the ABS is subjected to. It is possible that more shearing opposes the reaction, however it is very likely that increased shearing increases the energy of the system at the same time as increasing the mobility of the chains, increasing the likelihood of reaction. This hypothesis is tested in other sections.

The 200°C sample begins the transition into the crosslinking region around 110 seconds, still within the pre-defined processing cycle time. However, the 180°C, being far below the recommended processing temperature, and even slightly lower than melting, does not undergo this until 1200 seconds.

From these results, we can interpolate an exponential relationship between onset of crosslinking and temperature as shown by Figure 3.1. The crosslinking reaction kinetics are also greatly affected by the temperature, as shown by the crosslink dominated region of the graphs. The

220°C sample has more than double the slope of the 200°C, which is more than 20 times the slope of the 180°C sample. There is an approximately linear relation with the intercept at just below 180°C, which makes sense, as the polymer would not be yet molten in far below 180°C.

As far as peak viscosity, this is a view into the dynamics of crosslinking versus chain scission, which is occurring. With the higher temperature, the peak is higher, showing that crosslinking dominates the process early. The higher the peak, the further the crosslinking reaction has commences before the chain scission begins, reducing the viscosity.

The molecular weight distribution (MWD) of a polymer can be estimated through crossover points of storage and loss modulus. As shown in Figure 3.2, the crossover point increases with increasing processing, showing a narrowing of MWD. As the ABS is subjected to longer retention times, the chains begin to crosslink. The narrowing in the MWD is seen through the crosslinking of the smaller chain lengths first, increasing their molecular weight closer to those of higher molecular weight, narrowing the MWD.

3.3.2 Increasing Retention Time

Dynamic flow properties have been investigated in parallel plate rheometer (Figure 3.3). The test was preceded by a test for linear viscoelasticity, which confirmed that the viscoelastic range for

ABS was below about 5 rad/s. Thus, all tests were conducted which a shear rate of 1 rad/s, well within this range. The tests of complex modulus were done with frequency sweeps from 100 down to 1 Hz. For investigation of the effect of increasing the retention time during processing,

keeping all other process variables constant, the samples for rheological experiment were made in the DSM microextruder at 220°C and 100 RPM at 0.5, 1, 1.5, 2, 3 and 5 minutes.

Regarding the general viscoelastic properties of the ABS, as the frequency decreases, the complex viscosity increases, meaning that the polymer displays more resistance to a slower shearing and to a high shearing. This is a general trend for viscoelastic materials, however, there is an increase in the slope of the curve below about 0.25 Hz, showing that there is a high friction opposing the movement of chains between each other, which appears at shearing rates that are lower than those that stretch individual chains rather than push chains between each other [18].

There is a divergence that is seen between the each successive retention of the samples. Although they are roughly similar viscosities to start the test, their pre-conditioned processing affects their reaction during the approximately 11 minutes of testing. The trend that is immediately seen looking at the final value complex viscosity, where there is a gradual decrease in viscosity from each sample with increasing retention time during processing.

During the pre-conditioning of the samples, when they are subjected to shear in the melt, they undergo some combination of crosslinking and chain scission. It is possible that the presence of a substantial amount of shear (above that which it is subjected to in the rheometer) causes the chain scission to occur earlier in the process. Because of these opposite reactions (increasing and decreasing molecular weight of crosslinking and chain scission), there is no real difference seen in the viscosities as the test begins. However, as the test commences, the divergence is seen because of the absence of considerable shearing – the crosslinking may now be the dominant reaction. Because pre-conditioning saw more crosslinking with an increase in retention time, during the rheological test, there is less drive (a fewer proportion of molecules that have remained unreacted) and there is less increase in the complex viscosity.

The effects of these reactions during processing can pervade into the apparent mechanical properties of the ABS. Table 3.1 summarizes the impact strength as a result of increasing the retention time in the microcompounder at 240°C and 100 RPM. These parameters were chosen as a representation of typical conditions seen in application of the material. Immediately, the effects of increasing retention time are seen, as the sample retained for only 1 min has a 514 J/m impact strength. The samples retained for 3 and 5 min both had roughly the same impact strength value at 430 J/m. This exemplifies the degradation that was seen through the rheological studies, showing that the longer retention times led to a decrease in toughness of the material.

Interestingly, both the 3 and 5 min samples had the same impact strength, suggesting that the degradation may come to an end or plateau before this time, and that longer time scale mechanisms of degradation have not yet begun. Furthermore, looking at the standard deviation of the samples also gives some insight into the process of degradation. The 3 min retention time sample had a higher standard deviation than the 5 min sample, which can be an indication that although the strength value showed that the degradation was done, it may not have been completed in its entirety. The 5 min sample may have fully reached stability, an equilibrium of sorts, which is why the standard deviation was lowered.

3.3.3 Temperature Versus Shear

A very important parameter in a material for application is its reaction to processing. Since it is now known that the ABS undergoes reaction during processing, it is under interest to find which processing conditions increase this reaction. For this experiment, ABS was processing in a lab scale microcompounder. To investigate the effect of shear and temperature on the reaction process, the parameters of the extruder were adjusted by temperature and screw speed.

Temperature was changed from 220°C to 240°C while three levels of screw speed were used:

100, 50 and 0 RPM. For all screw speeds, the material was fed and extruded at 100 RPM, only during retention was the RPM different. The retention for all tests was three minutes.

200 180

160

140 120 G' 4min 100 G" 4min 80 G' 2min

Modulus(KPa) 60 40 G" 2min 20 0 0 20 40 60 80 100

Angular Frequency (1/s)

Figure 3.2 - Viscosity evolution over time of ABS at 180°C, 200°C, 220°C and 240°C.

Figure 3.2 shows the graph of the frequency sweep of all three screw speeds at 240°C. The sample exposed to the least shear falls between the others in terms of complex viscosity. The 50

RPM sample is the highest viscosity, while the 100 RPM sample has the lowest viscosity.

Although it may seem an odd result, this falls in line with the expected outcome. At 240°C, there is plenty of energy available for reaction and at 100 RPM, which subjects the material to the highest shear. There is enough energy to quickly speed the kinetics of the reaction, along with increasing mobility of material. The 100 RPM sample has already reached the chain scission dominant portion of the reaction, lowering its viscosity compared to 0 and 50 RPM.

Since the 0 RPM sample experiences the least shear and is therefore closest to neat ABS, it can be taken as baseline, a starting point for reaction with which to compare other samples. From

there, viscosity increases to the 50 RPM sample, which is the highest of the three. This is evidence that at this temperature, speed and retention time, more crosslinking has taken place and chain scission, increasing viscosity. At 50 RPM and 240°C, there is enough energy for oxidative reaction and sufficient mobilization to facilitate it. In contrast at 240°C and 100 RPM, possibly too much energy is present, or perhaps the ratio of thermal (from heating bands) to kinetic (shearing) may be too low.

0.5 Min 1 Min 1.5 Min 2 Min 3 Min

1.80E+04 5 Min

Viscosity Viscosity (Pa.s)

1.80E+03 0.1 1 10 100 Angular Frequency (1/s)

Figure 3.3 - Frequency sweep of ABS at 220°C, after processing at 0.5,1,1.5,2,3,5 min.

The presence of excessive shear may initiate chain scission early, as there now exists an extra amount of mechanical elongational stress pulling on the chains. Thus, at 100 RPM, we see a marked decrease in viscosity. This may be due to shear pulling, increased energy, or likely a combination of the two. At 220°C, the effects of changing shear are far less prominent, as seen in

Figure 3.3. In fact, the difference between them is so minute, the only reliable thing that can be

said is the viscosity of 100 RPM is higher than that of 50 and 0 RPM. In contrast to 240°C, here, the 100 RPM crosslinks more than undergoes chain scission. This means that 220°C is sufficiently less energy applied to the system as to not facilitate either early chain scission or fast enough kinetics to continue past crosslink dominant stage.

100000

240C, 100 RPM

240C, 50 RPM 240C, 0 RPM

10000 Complex Viscosity Complex Viscosity (Pa·s)

1000 0.1 1 10 100 Angular Frequency (1/s)

Figure 3.4 - Frequency sweep of ABS processed at 240°C and 100, 50 and 0 RPM.

Looking at the difference in temperature, it becomes less effective at generating changes in the material as the shearing decreases. At 100 RPM and 3 min (Figure 3.4), the effect is that 240°C has crossed the transition from crosslinking dominant to chain scission dominant regions.

Therefore, it shows a marked decrease in viscosity from 220°C, as in addition to pre-defined baseline, whereas 220°C has commenced through crosslinking just enough to be discernable by rheometry. This is the only shearing level which 240°C has a lower complex viscosity than

220°C. Therefore it is between these two levels of shear (50 and 100 RPM) that the difference between 220°C and 240°C is to cause a decrease in apparent molecular weight. At 50 RPM

(Figure 3.5), 240°C has increased viscosity over 220°C, but is perhaps just beginning to cross the crosslink-chain scission boundary, as the end of the test shows a decrease in viscosity, as compared to 220°C. At 0 RPM (Figure 3.6), no significant change can be discerned. This suggests that the presence of some shearing rapidly increases reaction rate as mobility is increased along with energy.

100000

220C, 100 RPM

220C, 50 RPM 220C, 0 RPM

10000 Complex Viscosity Complex Viscosity (Pa·s)

1000 0.1 1 10 100 Angular Frequency (1/s)

Figure 3.5 - Frequency sweep of ABS processed at 220°C and 100, 50 and 0 RPM.

100000

220C, 100 RPM

240C, 100 RPM

10000 Complex Viscosity Complex Viscosity (Pa·s)

1000 0.1 1 10 100 Angular Frequency (1/s)

Figure 3.6 - Frequency sweep of ABS processed at 100RPM: 220°C versus 240°C.

100000

220C, 50 RPM 240C, 50 RPM

10000 Complex Viscosity Complex Viscosity (Pa·s)

1000 0.1 1 10 100 Angular Frequency (1/s)

Figure 3.7 - Frequency sweep of ABS processed at 50RPM: 220°C versus 240°C.

100000

220C, 0 RPM 240C, 0 RPM

10000 Complex Viscosity Complex Viscosity (Pa·s)

1000 0.1 1 10 100 Angular Frequency (1/s)

Figure 3.8 - Frequency sweep of ABS processed at 0RPM: 220°C versus 240°C.

The impact strength was also measured according to this approximate processing scheme. The samples were made at the same temperatures and shear speeds, but were only held for a retention time of 2 minutes. The impact strength values are shown in Table 3.2, showing for the six samples (3 shear speeds, 2 temperature profiles). Here, unlike with the samples of increasing retention time, there is only a very small trend. In fact, the difference of the impact strength of these samples is not significant. This being said, a slight trend is seen that with the increase of shearing, there is a decrease in the impact strength. With larger changes in the shearing, temperature, or retention time, it is expected that these differences in impact strength would eventually become significant. However, the shearing and temperature ranges are fairly wide and representative of general processing conditions. It is for this reason that bigger differences in the shear and temperature were not selected, as the aim of this work is to outline the applicable

processing effect on the ABS. Therefore, changes in shear speed and temperature do not bring significant changes in the mechanical properties of the ABS.

3.4 Conclusion

Degradation mechanisms of ABS during melt processing have been investigated, in which it was found that ABS undergoes a thermal oxidative reaction in the polybutadiene phase. This proceeds through hydrogen extraction and crosslinking bonds being formed. As expected, the kinetics are increased as a result of an increase in temperature. It appears that at the temperatures required for processing, ABS undergoes crosslinking, followed by chain scission. This was viewed through the lens of the rheological characteristics, which saw an increase in the viscosity over time, followed by a constant phase, and finally a drop in viscosity over the 1 hour time period of which studies commenced.

Increasing the retention time of the ABS in the melt processing equipment did not lead to a decrease in the viscosity of the samples as it would for other polymers. Instead, the viscosity was increased and decreased roughly according to the findings of the 1 hour rheology studies.

However, the sample impact values were decreased with increasing retention time, showing the reaction causes a decrease in the toughness of the material. In contrast, with increasing temperature and shear speed, no large difference was seen in the impact strength. It was found that the type of energy present has an effect on the degradation reaction of the ABS. High shear energies were found to decrease the viscosity of the sample quickly, possibly causing the onset of chain scission to occur earlier than those at lower shear rates.

3.5 Acknowledgements

The financial support from the Ontario Ministry of Agriculture and Food Rural Affairs

(OMAFRA)/University of Guelph - Bioeconomy for Industrial Uses Research Program (Project

#200245); Natural Sciences and Engineering Research Council (NSERC) AUTO21 NCE project

(Project #400372 & 400373) to carry out this research is gratefully acknowledged.

3.6 References

1. Keitz, J. D., J. W. Barlow, and D. R. Paul. "Polycarbonate blends with styrene/acrylonitrile copolymers." Journal of applied polymer science 29.10 (1984): 3131-3145.

2. Tjong, S. C., and Y. Z. Meng. "Effect of reactive compatibilizers on the mechanical properties of polycarbonate/poly (acrylonitrile-butadiene-styrene) blends." European polymer journal 36.1 (2000): 123-129.

3. Yang, Kumin, Shi ‐ Ho Lee, and Jong ‐ Man Oh. "Effects of viscosity ratio and compatibilizers on the morphology and mechanical properties of polycarbonate/acrylonitrile‐ butadiene‐styrene blends." Polymer Engineering & Science 39.9 (1999): 1667-1677.

4. H.F. Mark, N.M. Bikales, C.G. Overberger, G. Menges, J.I. Kroschwitz (Eds.),

Encyclopedia of polymer science and engineering, Wiley, New York (1996), p. 388

5. Wu, Jiann-Shing, Shu-Chen Shen, and Feng-Chih Chang. "Effect of rubber content in acrylonitrile–butadiene–styrene and additional rubber on the polymer blends of polycarbonate and acrylonitrile–butadiene–styrene." Polymer journal 26.1 (1994): 33-42.

6. Zhong, Hanfang, et al. "Thermal degradation behaviors and flame retardancy of PC/ABS with novel silicon‐containing flame retardant." Fire and Materials 31.6 (2007): 411-423.

7. Kumar, Annamalai Pratheep, et al. "Nanoscale particles for polymer degradation and stabilization—trends and future perspectives." Progress in Polymer Science 34.6 (2009): 479-

515.

8. Wang, Shaofeng, et al. "Preparation and thermal properties of ABS/montmorillonite nanocomposite." Polymer degradation and stability 77.3 (2002): 423-426.

9. Bokria, Jayesh G., and Shulamith Schlick. "Spatial effects in the photodegradation of poly (acrylonitrile–butadiene–styrene): a study by ATR-FTIR." Polymer 43.11 (2002): 3239-

3246.

10. Motyakin, Mikhail V., and Shulamith Schlick. "Thermal degradation at 393 K of poly

(acrylonitrile-butadiene-styrene)(ABS) containing a hindered amine stabilizer: a study by 1D and

2D electron spin resonance imaging (ESRI) and ATR–FTIR." Polymer degradation and stability 76.1 (2002): 25-36.

11. Bair, H. E., D. J. Boyle, and P. G. Kelleher. "The effects of light and heat on the rubber content and impact strength of acrylonitrile‐ butadiene‐styrene." Polymer Engineering &

Science 20.15 (1980): 995-1001.

12. Suzuki, Masanori, and Charles A. Wilkie. "The thermal degradation of acrylonitrile- butadiene-styrene terpolymei as studied by TGA/FTIR." Polymer degradation and stability 47.2

(1995): 217-221.

13. Adam, Claudie, Jacques Lacoste, and Jacques Lemaire. "Photo-oxidation of elastomeric materials: Part 3—Photo-oxidation of acrylonitrile-butadiene copolymer." Polymer degradation and stability 27.1 (1990): 85-97.

14. Tiganis, B. E., et al. "Thermal degradation of acrylonitrile–butadiene–styrene (ABS) blends." Polymer degradation and stability 76.3 (2002): 425-434.

15. Ghaemy, M., and G. Scott. "Photo-and thermal oxidation of ABS: Correlation of loss of impact strength with degradation of the rubber component." Polymer Degradation and

Stability 3.3 (1981): 233-242.

16. Aoki, Yuji, et al. "Nonlinear stress relaxation of ABS polymers in the molten state." Macromolecules 34.9 (2001): 3100-3107.

17. Shimada, Junichi, and Kimiaki Kabuki. "The mechanism of oxidative degradation of

ABS resin. Part I. The mechanism of thermooxidative degradation." Journal of Applied Polymer

Science 12.4 (1968): 655-669.

18. Ferry, John D. Viscoelastic properties of polymers. John Wiley & Sons, 1980.

Chapter 4: Sustainable Biobased Blends from the Reactive Extrusion of Polylactide (PLA) and Acrylonitrile Butadiene Styrene

______

Abstract

Polymer blends containing poly(lactic acid) (PLA) and acrylonitrile butadiene styrene (ABS) with high biobased content (50%) were made by extrusion and injection molding. Two additives, one acrylic copolymer and one chain extender were used separately and in combination to increase mechanical properties. Interestingly, the combination of both the acrylic copolymer and chain extender worked to synergistically increase the impact strength by almost 600%. This was attributed to the complementary additive toughening effects which allowed increased energy dissipation of the blend at high speed testing, such as in the impact test. Morphology and rheology investigation showed that the two additives worked together to vastly change the dispersion and phase sizes, suggesting a decreased tension between the PLA and ABS. Finally,

Fourier transform infrared spectroscopy supported the evidence that the epoxy groups of the chain extender undergo ring opening to react with the functional groups of the PLA.

______

4.1 Introduction

As concerns continue to increase over environmental detriment due to the use of petroleum, there is a need for increased sustainability in polymer materials. This has caused considerable interest in biobased plastics, which are made from sustainable resources. However, many have one or more deficiencies in properties that restrict their widespread use. Most notable among biobased plastics is poly(lactic acid) (PLA) because of its affordable cost, strength and biodegradability.

However, inherent deficiencies of PLA have been shown to be barriers to its widespread use.

The low toughness and heat deflection temperature have been known to be the major barriers from its use in many applications such as packaging and automotive, among others.

PLA is an aliphatic polyester thermoplastic polymer and the leading most bioplastic in terms of production.1–7 Until recently, PLA was only used for niche applications such as in the biomedical industry.6 However, increased production economies in addition to more efficient synthesis reaction have brought the cost of PLA down substantially. The thermal stability of PLA, measured by heat deflection temperature (HDT), is very low along with its toughness, which indicates the target areas for improvement. Several methods have been attempted to improve the performance of PLA, including blending and plasticization.8–10

Traditional petroleum based polymers have positioned themselves in the market due to their low cost, good mechanical properties, and ease of processing.11–13 Still to this point, a vast majority of plastics are petroleum based. Acrylonitrile butadiene styrene (ABS) is an excellent example, and is used heavily in the electronics and automotive industries due to its tailorable properties and a relatively low cost considering its mechanical properties. The proportions of its constituents can be adjusted, creating a material with the desired balance of properties and cost.

This is an extremely advantageous trait as an ABS may be developed for specific applications

such as high impact. However, the fact that it is petroleum derived has caused interest in a more sustainable replacement. For this reason, the industry is looking for replacement materials for

ABS that are more environmentally friendly and sustainable. This is becoming an increasingly important topic as the climate change movement gains momentum, and many show great concern for the future. Thus, if ABS was blended with a biobased polymer, such as PLA, while keeping the performance similar to neat ABS, then the petroleum use of ABS would be reduced, increasing the sustainability of the material.

ABS has been blended with a range of polymers, serving to a large degree as a toughening agent.

Hale et al. found that ABS could be used to increase the impact strength of polybutylene terephthalate (PBT).14 They found that 30% by weight illicited vast improvements in the impact strength over neat PBT. Kudva et al. blended ABS with nylon 6 to decrease the notch sensitivity.15 Again, they saw dramatic improvement in the notched Izod impact strength with a variety of ABS types and amounts. Zhang et al. found that ABS can be blended with PC with exceptional results, and when compatibilized with ABS grafted maleic anhydride (ABS-g-

MAH), the performance of the blend further increased.16 Recently, Dong et al. used a reactive comb polymer made from methyl methacrylate (MMA), glycidyl methacrylate (GMA), and

MMA macromer to compatibilize PLA and ABS.17 They concluded that the reactive comb polymer drastically increased the interfacial adhesion between the PLA and ABS, which led to a very large increase in the toughness of the blend.

The disadvantages of both PLA and ABS indicate that they may be blended to create a partially bio-based polymer with balanced properties. In addition, a well performing blend that incorporates ABS will have mechanical properties that closely resemble ABS. However, the literature describes a poor interaction between the two polymers, resulting in poor adhesion

between phases, poor dispersion, and unstable morphology.3,18–21 Sun et al. found that there was an the interaction between molecules of the PLA/ABS blend was unfavorable, producing poor interfacial adhesion and poor dispersion.21 Additionally, Li et al. observed the ABS phases tended to connect in irregular shapes with large phase sizes and a weak interface, which were the cause of poor mechanical properties.20 This suggests that compatibilization is required to increase the adhesion and facilitate stable morphology between the two22–26. Several attempts have been made to perform such compatibilization, with varied results.

In this work, ABS was blended with PLA to create a multiphase polymer. The ABS used was a toughened grade, with high butadiene content. The higher rubber content is used to increase the impact strength and toughness of the blends. This approach gives higher likelihood of resulting in a blend that demonstrates high impact strength. PLA and ABS are blended together along with two additives that work synergistically to drastically and efficiently improve the mechanical properties of the blends. The goal is to create a good performing, low cost, easy to process material. The main performance criteria that are the focus are impact strength and elongation at break, as the material is meant to be used in applications that require high toughness.

4.2 Experimental

4.2.1 Materials

The PLA used in this study (Ingeo 3052D) is an injection grade acquired from Natureworks

LLC, USA. The ABS used in this work (Magnum 1150 EM) is a low flow injection grade (MFI

= 0.90 g/10min at 3.8kg 240°C) with high toughness and was acquired from Trinseo, USA.

These specific polymer grades were chosen to create a blend with high toughness and impact strength. Biostrength 900 is an acrylic copolymer and was obtained from Arkema (USA) in powder form. Biostrength is a processability enhancer developed for PLA. As its structure is

proprietary, it is not well known by the authors. Joncryl ADR-4368C is a styrene-acrylic oligomer with epoxy functional groups. The structure of Joncryl can be found in the literature, as it is used in other work as a chain extender for PLA.27 It was obtained from BASF (Germany).

4.2.2 Preparation of the Blends

The neat polymers were dried in a convection oven at 80°C for a minimum of six hours prior to compounding. A batch-style microcompounder, DSM Xplore (DSM, Netherlands) was used to create blends with 50% PLA. This creates the potential for a polymer blend with a high biobased content, but can still adopt material properties from ABS. The length to diameter ratio, L/D is 18 and the screw length is 150 mm. The polymer pellets were weighed according to composition of the blend, then fed into the compounder in a single batch. The compounding conditions were

240°C barrel temperature, 100 RPM screw speed, and a retention time of 2 minutes. The compositions were injected into tensile, flexural, and impact samples for testing. The dimensional specifications of the samples are outlined in the pertinent ASTM standard.

4.2.3 Mechanical Testing

Tensile tests were conducted using an Instron mechanical testing machine according to ASTM

D638. The crosshead speed was set to 5 mm/min. Tensile strength, Young’s modulus, and elongation at break were analyzed and reported using Bluehill software. The flexural properties of the blends were measured according to ASTM D790 at a crosshead speed of 14 mm/min.

Flexural strength and modulus were analyzed and reported. The notched Izod impact strength of the compositions was measured with a TMI inc pendulum impact tester according to ASTM

D256.

4.2.4 Dynamic Mechanical Analysis

The solid-state dynamic properties of the materials were tested using a TA Instruments DMA, model Q800. Under a dual-cantilever clamp, the frequency was constant at 1 Hz, the amplitude set to 15 mm. The tests were carried out from -100°C to 150°C with a heating rate of 3°C/min.

Storage modulus and tan delta were analyzed and reported.

4.2.5 Differential Scanning Calorimetry

A DSC Q200 from TA Instruments was used on heat/cool/heat mode. From room temperature

(approximately 20°C), the samples were heated at 10°C/min to 250°C. The samples were then cooled to 0°C at a rate of 20°C min. This created a uniform thermal history among specimen.

Finally, the samples were heated back to 250°C at a heating rate of 10°C/min. The melting temperature (Tm), the glass transition temperature (Tg), and the crystallization were studied.

4.2.6 Rheology

Rheological experiments were done on a parallel plate rheometer at 220°C under nitrogen atmosphere. Frequency sweeps were completed from 0.01 s-1 to 100 s-1 at an amplitude of 0.1%.

Complex viscosity and storage modulus were measured as a function of temperature.Atomic

Force Microscopy

Samples were microtomed in a Leica Ultramicrotome to create a flat surface for scanning. A

Bruker Multimode 8 atomic force microscope was used in tapping mode to map the morphology and nanomechanical properties of the surface. The instrument was used on quantum nanomechanical mapping mode to create micrographs.

4.2.7 Fourier Transform Infrared Spectroscopy (FT-IR)

Specimen were prepared from premade samples such as tensile, flexural, and impact. The reaction between polymer and additives were discerned on a Thermo Scientific FT-IR. A typical spectrum consisted of 64 individual runs, to provide desired resolution.

4.3 Results And Discussion

4.3.1 Tensile and flexural properties

The tensile strength and tensile modulus of neat ABS, neat PLA, and their blends are provided in

Figure 4.1. The strength of all blends is approximately equal, which is reasonable considering that the strength value is that of low deformation, and thus, the polymer chains are still largely held in relative position, as the intermolecular bonds have just begun to break, resulting in plastic deformation.28 Since both polymers have inherently good ability to absorb energy in this manner, the tensile strength shows a high value regardless of compatibility between the polymers.29

Instead, compatibility of the blend would be much easier to discern at higher deformations. The tensile strength does not often indicate the compatibility of the polymers, especially at low strain rates. These approximately equal strengths are not indicative of the overall performance of the blend.

80 2.9 Tensile Strength Tensile Modulus 70

2.7

60 2.5 50 2.3 40 2.1 30

Tensile Strength (MPa) Strength Tensile 1.9

20 Tensile Modulus (GPa) Modulus Tensile

10 1.7

0 1.5 A B C D E F

Figure 4.1 – Tensile properties of neat polymers and blends: (A) Neat PLA (B) PLA/ABS (C)

PLA/ABS/Acrylic copolymer (D) PLA/ABS/Chain extender (E) PLA/ABS/Chain extender/Acrylic copolymer (F) Neat ABS

Similarly, the modulus values are a representation of the ability of the polymers and the blends to resist strain at low strain rates, below yield. There is a slight increase seen in these values, which is likely the increased compatibility from the acrylic copolymer, chain extender, and to a greater extent, the effect of both the acrylic copolymer and chain extender together. Li and Shimizu30 found that an increase in the compatibility of the ABS/PLA blend through the use of SAN-GMA along with a catalyst also produced an increase in the modulus of the blend at a 50/50 ratio of

ABS/PLA.

160 600 Elongation at Break 140 Impact Strength

500

120

400 100

80 300

200

Elongation at Break (%) Break at Elongation Impact Strength (J/m) Strength Impact 40

100 20

0 0 A B C D E F

Figure 4.2 – Elongation at break and impact strength of neat polymers and blends: (A) Neat PLA

(B) PLA/ABS (C) PLA/ABS/Acrylic copolymer (D) PLA/ABS/Chain extender (E)

PLA/ABS/Chain extender/Acrylic copolymer (F) Neat ABS

The elongation at break and impact strengths of neat PLA, neat ABS, and their blends are given in Figure 4.2. The toughness of the blends, which in this case is approximated by elongation at break, shows large differences between blends due to the effect of the acrylic copolymer and chain extender. Firstly, the elongation at break better shows the compatibility of the constituent polymers than the strength and modulus because it is a high deformation test.30 Li and Shimizu again have shown that poor compatibility will have a greater negative effect on elongation at break than on tensile strength or modulus.30 Here, the elongation at break experienced a drastic

increase compared to the modest increase in the tensile strength and static modulus values.

Instead of a measurement value at or before yield, where the polymer chains have not yet undergone plastic deformation, the elongation at break value is representative of the weak portions of the blend.30 This is due to the mechanism of failure under a tensile load. The site of failure initiation, which in the case of a poorly compatible blend is the interface, will allow propagation to occur, and proportionally large amount of stress is then placed on the individual constituent polymers, creating failure at a lower strain rate than a compatible blend.31 Yang et al. found that high speed tensile testing and impact testing shared a brittle to ductile mechanism.31

The PLA/ABS blend, which has very low compatibility as cited in the literature, also has a very low elongation at break.18,19,21,30 This will likely also lead to low impact strength, which will be analyzed in the next section, as there is insufficient energy absorption with even low loading of the blend. At a low strain rate such as 5 mm/min, there is increased time for the Brownian motion of the polymer chains to disperse energy as loading increases, as compared to high strain rates. Thus, the elongation at break is increased. The addition of the chain extender increases the elongation at break in a different mechanism. Here, the increased distribution of bonding creates a loose network of PLA chains, which allows the uptake of energy, especially at lower strain rates. There is decreased tension between the PLA and the ABS. These two processes account for the increase in the elongation at break in this case. Finally, the blend with both the acrylic copolymer and chain extender has an elongation at break between each of those blends. This is likely due to the compromising of the two mechanisms of failure described for each additive.

The increased bonding due to chain extension is offset slightly by the increased mobility of the chains. Since there is more mobility, the intermolecular bonds require less energy to break, thus the chain extender has somewhat less of an effect when used with the acrylic copolymer as it

would by itself.28 This is for the mechanism of failure for elongation at break, where strain rates are relatively low.

4.3.2 Impact strength

The impact strength, as with the elongation at break, is able to show relative compatibility of blends. The speed at which the material is tested at lends to this ability, as both the inter- and intra- molecular bonds are unable to disperse energy quickly enough to conceal the true performance of the polymer. The movement of the chains is an important factor with a slower test such as tensile, where the natural Brownian motion that the chains undergo can work to disperse stresses. However, an instantaneous test does not allow this dispersion.32 With such a test, the Brownian motion of the chains is negligible and therefore failure is usually dependent on the weakest link in the energy transfer chain. Undoubtedly, with immiscible polymer blends, the weakest link mainly comes in the form of the interface between the polymer phases.23

Neat PLA has a very low impact strength value, due to its very brittle nature. This is in contrast to low impact strength of non-brittle polymers. In non-brittle polymer that still exhibit low impact strength, the low impact is generally associated with a weak intra-molecular bond, such as with polyethylene.12 In contrast, PLA has a very high intra-molecular strength, as is shown in its tensile and flexural strength and modulus. The insufficiency in PLA that is shown in its low impact strength is the inability of chains to mobilize relative to each other, and therefore inability to absorb energy.29 Failure of inter-molecular bonds allows easy energy propagation through the polymer network, and therefore impact strength is low. The additives used in this study are intended to address this limitation.

The impact strengths of the ABS/PLA blends are shown in Figure 4.2. The blend without additives has a very low impact strength value. This is expected, as the poor compatibility of this

blend is covered in the literature.19,21,33 In fact, many works have started from a point of compatibilization, knowing that the neat blend has very poor performance. These works sought to increase the compatibility of PLA/ABS blends through the use of compatibilizers. Even with

50 % wt. of ABS, the impact strength is very close to the value of neat PLA, as the high toughness of ABS is unable to overcome the brittleness of PLA and the low adhesion of the interface between the polymer phases.

The addition of the acrylic copolymer allows an increase in the impact strength of the blend.

However, the value is still far below that which rule of mixtures may suggest. With an increase of mobility of the chains that the acrylic copolymer provides, there is increased ability of the

PLA to absorb energy through relative deformation before total failure occurs. The poor interface still limits the toughness of the blend, however, keeping the impact strength value below 100

J/m. The chain extender increases the impact strength by chain extension of the PLA phase. Al-

Itry et al. also found the chain extender was able to increase the apparent molecular weight of

PLA blends.3 In this case, increased bonding of the PLA chains increases the overall energy required for fracture (as more bonds are required to be broken). This is in agreement with Najafi et al., who found an increased molecular weight with use of chain extender.34 The combination of the chain extender and acrylic copolymer vastly increase the impact strength by working synergistically through two distinct mechanisms. The acrylic copolymer allows increased mobility of chains, while the chain extender increases the bonding of the PLA phase. This sort of increase has been seen in similar PLA based systems such as with Li et al. or with Liu et al.30,35

Li et al. saw an increase of 133% over neat PLA in the film impact test. Likewise, Liu found an increase of over 3000% in notched Izod impact strength over neat PLA with an 80% loading of

PLA. However, these systems tend to create toughness through blending PLA with a high

amount of additives or by using a post process such as annealing, both of which induce additional costs to the material. The aim of our work was to increase thoughness of PLA/ABS without incurring much cost. By incorporating only 2 wt. % of additives, and using minimal processing techniques, the costs remain low.

4.3.3 Phase Morphology (SEM)

Scanning electron micrographs were taken of impact-fractured surfaces of samples and are shown in Figure 4.3. These images agree with the results of Li and Shimizu, which found that

ABS was dispersed in a PLA phase, and that the domain size was in the range of 1 -10 µm.30 In addition, they found the domain size distribution to be very large due to the immiscibility between the constituent polymers. The PLA/ABS blend without additives shows a very poor morphology with obvious debonding between phases. The size and dispersion of each phase is extremely heterogeneous, showing complete incompatibility, as evidenced in the impact strength. Since PLA has a much lower viscosity than the ABS, it seems to form a quasi- continuous phase, while the ABS forms a quasi-dispersed phase. This arrangement further impairs the ability of the blend to absorb energy through impact, as the PLA domains are further- reaching than the ABS, limiting the ABS exposure to fracture energy propagation. The voids, space between phases, and poor dispersion are all evidence of high interfacial tension and low interfacial adhesion. The addition of acrylic copolymer in the blend does not drastically change the morphology. In fact, there is very high similarity between micrographs of PLA/ABS and

PLA/ABS/acrylic copolymer. The likely effect of the acrylic copolymer is not manifested through a change in the morphology. Instead, it is evidenced through increased mobility of chains during shearing, as shown through rheology measurements. The addition of the chain

extender has an effect on the morphology, as seen in the SEM micrograph. There is still obvious evidence of poor compatibility, as in the PLA/ABS blend, and with the acrylic copolymer.

However, the dispersion of ABS in the PLA is less random, and as a result, the ABS is able to assume more loading. This causes the morphology change that is seen, the pullout of ABS fibrils.

There is more smoothness seen with the chain extender, suggesting a lower tension between the phases. The blend containing both the acrylic copolymer and the chain extender show the largest change in morphology. There is a drastic change in the size of the structures present, suggesting much decreased tension between phases, allowing the morphology to become much more refined. This reduction in size and distribution upon compatibilization was also cited by Li and

Shimizu.30 This aids in the ability of the material to absorb impact energy propagation, increasing impact strength.

A B

20 μm 20 μm C D

20 μm 20 μm

Figure 4.3 – Scanning electron microscopy of impact fractured surfaces of A) PLA/ABS B)

PLA/ABS/Acrylic copolymer C) PLA/ABS/Chain extender D) PLA/ABS/Acrylic copolymer/Chain extender

4.3.4 Thermal Crystallization (DSC)

Differential scanning calorimetry was used to study the glass transition, melting, and crystallization behavior of the blends. The second heating scan of samples is shown in Figure

4.4. This removed the thermal history of the samples to show the intrinsic behavior of the blends.

The PLA/ABS sample exhibits behavior expected of PLA. Due to the amorphous nature of ABS, there is little evidence pertaining to the ABS phase in these results. The glass transition of PLA occurs around 57-60°C, which is followed by a crystallization peak around 115°C. Immediately, the polymer begins melting, the peak occurring at approximately 150°C. A double peak is seen as the PLA undergoes a melt/crystallize/melt behavior, creating a beta crystalline phase.4,36,37

The blend with the acrylic copolymer behaves in the same manner as the neat blend.

Interestingly enough, the inclusion of the chain extender causes a vast change in the thermal behavior of the blend. Crystallization of the chains shows a much smaller and broader peak, suggesting that the chain extender is inhibiting the ability of the chain to crystallize.4,36,37 This is inherently so as the molecular weight of a polymer is inversely tied to the crystallization of the chains. Following the crystallization, the melting peak of the sample with chain extender exhibits only one peak, as opposed to the double peaks seen in the PLA/ABS and PLA/ABS/acrylic copolymer blends. The blend with both the acrylic copolymer and chain extender exhibits similar thermal properties to the blend with the chain extender alone. The inclusion of the chain extender affects the properties of the blend in a dominant way, such that similar properties are seen regardless of whether the acrylic copolymer is added or not. In other words, the acrylic copolymer does not impede the chain extender from its effect on the blend.

The crystallinity of the blends were also studied according to Equation 1:

∆퐻푚+∆퐻푐 (1) 푋푐(%) = 푥 100 푓∆퐻푓

Where ∆퐻푚 and ∆퐻푐 are the enthalpies of melting and cold crystallization, respectively. The weight fraction of the PLA in the blends (which is of main concern) is denoted by 푓. The hea of

-1 fusion of 100% crystalline PLA, ∆퐻푓, was taken as 93.7 Jg , as found in the literature.

A PLA/ABS B PLA/ABS/Mold Release Agent C PLA/ABS/Chain Extender D PLA/ABS/Mold Release Agent/Chain Extender

D A

B Heat Heat Flow(W/g)

0 50 100 150 200 250 300 Temperature (°C)

Figure 4.4 – Differential scanning calorimetry thermogram of first heating cycle

The crystallinity of PLA changed as it was blended with the different components. The blend of

PLA/ABS without additives created a crystalline content of only 8.6% in the PLA fraction. The addition of acrylic copolymer increased the crystallinity to 11.8%, which is expected due to the increased mobility of the PLA chains from the addition of acrylic copolymer. On the other hand, the chain extender decreased the crystalline percentage to only 3.1%. Again, this was an expected result, as the chain extender caused increased difficulty for the chains to gather into highly ordered crystalline form. The blend with both additives had the highest crystalline percentage of all the blends, which was surprising as it included the chain extender. Again, the

synergy of both the acrylic copolymer and the chain extender are seen, causing the highest crystalline content.

4.3.5 Rheology

Rheological experiments were done on a parallel plate rheometer at 220°C under nitrogen atmosphere. This temperature allows melting of all components, but is not so high that the viscosity of PLA is negligible. Frequency sweeps were completed from 0.01 s-1 to 100 s-1 at an amplitude of 0.1%. Complex viscosity and storage modulus were measured as a function of temperature. In the storage modulus curve (Figure 4.5), all blends exhibit a shoulder at lower frequencies, as expected due to the presence of the interfacial layer. At lower frequencies, the polymer blend is being strained at a slower rate, thus the chains exhibit less stretching compared to higher frequencies. The slow strain rates allow chains to slide past each other rather than stretch, which in general are a relatively higher measure of the interface as there is little elasticity, especially in interfaces with higher tension. This increases tension at these low frequencies, causing an inflection in the storage modulus curve, which can be qualitatively noted.

3,24,38 The PLA/ABS blend exhibits a high amount of tension, and has the highest increase after the shoulder curve. With the addition of the acrylic copolymer, the increased mobility of the chains is seen, in addition to what can be perceived as a decrease in the amount of interfacial tension from only a slight increase in the storage modulus at low frequencies. The addition of the chain extender without the acrylic copolymer showed much less tension, and only a slight shoulder appeared, and an increase in the storage modulus is only seen at frequencies around

0.01 s-1. The storage modulus value for this blend is higher throughout the frequency range as expected with increased chain lengths. 39 Finally, the blend containing both the acrylic copolymer and chain extender showed the least interfacial tension. In fact, although a slight

inflection was seen, the storage modulus value did not increase throughout the entire range of testing frequencies. Gu et al3 also found that the polymer blends showed a departure from thermo-rheological simplicity. This was ascribed to the presence of immiscible structures in the polymer blends.

1000000

100000

10000

D 1000 C A

100 B

Storage Modulus (Pa)Modulus Storage

10 A PLA/ABS B PLA/ABS/Acrylic Copolymer C PLA/ABS/Chain Extender D PLA/ABS/Acrylic Copolymer/Chain Extender 1 0.01 0.1 1 10 100 Frequency (1/s)

Figure 4.5 – Storage modulus of PLA/ABS blends

1000000

100000

10000 C A

1000 B

100 Complex Viscosity (Pa·s) Viscosity Complex

PLA/ABS 10 A B PLA/ABS/Acrylic Copolymer C PLA/ABS/Chain Extender D PLA/ABS/Acrylic Copolymer/Chain Extender 1 0.01 0.1 1 10 100 Frequency (1/s)

Figure 4.6 – Complex viscosity curve of PLA/ABS blends

The complex viscosity provides another view of the effects of the components to the blend

(Figure 4.6). 24,25,38,40 The addition of acrylic copolymer causes a large effect, decreasing the complex viscosity especially at low frequencies. The addition of chain extender has an opposing effect, increasing the complex viscosity over the entire curve. Interestingly, as seen in the solid dynamic properties, the inclusion of acrylic copolymer and chain extender does not show a large deviation from the curve of the chain extender alone. Instead, its effect of increasing the chain mobility has a slight effect, and is only seen at lower frequencies, suggesting that it aids in decreasing the interfacial tension somewhat. This is seen in the impact strength of the polymer, increasing value to a highly toughened state.

4.3.6 Atomic Force Microscopy (AFM)

The PLA/ABS blend shows a great amount of stress, as seen in abnormal shapes of phases, wide phase size distribution, and a clear difference between phases. The evidence of tension between phases is present in these micrographs, showing the inability of the material to absorb energy.

The addition of the acrylic copolymer again vastly changes the phase structures to appear much more consistent, heterogeneous, and with less tension. However, there is still a distinct separation between phases, indicating that tension is still quite high. The addition of chain extender shows a similar micrograph, although the ABS structures are smaller still than those of the acrylic copolymer sample. The tension between samples is becoming successively lower, which is shown in the elongation and impact strength of the blends. Finally, the blend with both the acrylic copolymer and chain extender shows an effective decrease in tension, as the structures are much smaller, and there is much less space between structures. In fact, from a mechanical standpoint, the ABS and PLA phases appear to have become somewhat interlocked, as there is much less clear distinction between the phases. Davies et al41 also performed AFM imaging on PLA blends. They found clear phase separation using the AFM, which led to the conclusion that increased PLA caused a surface enrichment effect.

A B

C D

Figure 4.7 – AFM modulus mapping of PLA/ABS blends: (A) PLA/ABS (B) PLA/ABS/Acrylic copolymer (C) PLA/ABS/Chain extender (D) PLA/ABS/Chain extender/Acrylic copolymer

4.3.7 Chemical Reaction Mechanism

The FT-IR spectrum is given in Figure 4.8. The chain extender has two peaks at 907 and 842 cm-

1 which correspond to the ring deformation vibrations of the epoxide groups. These are the functional groups with high likelihood of reacting with the acid group of the PLA. When we compare these two peaks to the blended sample, they appear to be gone, suggesting that the PLA has reacted with the chain extender to open the epoxide group causing chain extension in the

PLA. The proposed reaction scheme is given in Figure 4.9. Ojijo and Ray saw the same reaction when blending PLA and this chain extender.27 They noted that it is possible for this reaction to give rise to a long-chain branched structure.27 In addition, the main peaks in the ABS spectra are also found in the full blend spectra, suggesting that the ABS functionality remains unchanged from neat to the final blend. This confirms the evidence that ABS does not play a significant role in the reaction, and instead the interaction between PLA and ABS is largely mechanical.

100

Figure 4.8 – Fourier transform infrared spectra of blend constituents compared to full final prepared blend

101

PLA Chain Extender

Product

Figure 4.9 – Proposed reaction between PLA and the chain extender

Understanding blends of biobased and petroleum based polymers is of great importance as environmental concerns call for the development of sustainable materials. Due to current shortcomings of biobased plastics, the need to blend with petroleum-based polymers is obvious.

Therefore, these blends must be developed into materials with attractive properties and cost. In this work, the authors outline the mechanism by which poly(lactic acid) (PLA) and acrylonitrile butadiene styrene (ABS) may be blended into a high performance material by manipulation of the phases into a nanostructured arrangement.

The DSC of the first and second heating scan of the full BioABS blend are given in Figure 4.10.

A melting peak was not observed for the neat PLA. The cooling curve of PLA, offered in the supporting information, did not have a crystallization peak, and thus the PLA was completely

102

amorphous during the subsequent heating scan. However, the expected cold crystallization peak was also not observed. In comparison, the blended systems, even the PLA/ABS blend without additives showed cold crystallization and subsequent melting peak under the same conditions, indicating that the ABS acts as crystal formation agent, either in the form of improved nucleation, or chain mobility. Interestingly, the blend showed two distinct melting peaks. Likely due to the polymorphic nature of PLA, the different peaks were each pronounced during the different heating cycles. The peak at 150˚C was pronounced during the first heating scan, while the peak at 154˚C was pronounced during the second heating scan. These interesting results lead the researchers to investigate the crystallization temperatures of the blend and the resulting crystal structures.

Figure 4.11 shows the subsequent heating scans after thermal treatment of melting, then quenching to each denoted crystallization temperature from 90˚C-150˚C. These results showed that the crystal formation could be controlled through the crystallization temperature. The literature states that for neat PLA, the alpha crystals will develop above 120˚C, alpha prime crystals will develop below 90˚C, and a mixture will develop at intermediate temperatures. These temperature ranges do not hold for the blend, however. It seems that in the blend, both crystal types form below 110˚C, while only the alpha develops at 120˚C and above.

The formation of this same crystal structure was attempted by controlling mold temperature and time. Through the DSC studies, it was found that the crystal formation was stopped after approximately 25 minutes at 90˚C, thus the samples were molded at 90˚C and kept in the mold for 30 minutes to ensure full crystallization. DSC study of the sample after molding confirmed that full crystallization was achieved. Furthermore, it showed a very similar crystal structure to the sample quenched at 90˚C.

103

The mechanical properties of the blend containing this crystal structure were studied and compared to an identical blend molded at 30˚C (Table 4.1). Annealing had a drastic effect on the mechanical properties, increasing all mechanical properties except for the elongation at break.

The impact strength increased by almost 200 J/m, while the strength and modulus also increased.

It is unusual for the strength, modulus and impact resistance to increase simultaneously, especially in the case of annealing. However, this is due to the very specific morphology of the blend. Because the PLA has formed very thin fibrils, or ligaments around the ABS, there is a great amount of energy dissipation that is allowed. In this case, by changing the PLA from near amorphous to a crystalline structure, the failure mechanism was also changed.42 Amorphous PLA undergoes crazing during failure, which in a blend requires an optimal size of dispersion. This optimal size is such that the dispersed phase is large enough to nucleate crazes, but small enough to effectively dissipate energy. However, under shear yield failure, the dispersion should be minimized to increase the toughness. In this blend, the dispersed phase of ABS is small enough to be below the optimum size for crazing, and thus when the failure mechanism is switched, a toughening effect is achieved. Wu, studying at rubber filled systems, explains this effect in the work.43 Additionally, it is seen in a PLA system with EGMA as the filler in the work by Oyama et al.44 The heat deflection temperature, which is notably low for PLA based polymers, did undergo a dramatic change after annealing. In fact, the annealed value is higher than either neat polymer.

104

Temperature (°C)

0 50 100 150 200 250 300

First

Heating Scan

Second Heating

Scan Heat Flow (Exo up) (Exo Flow Heat

Figure 4.10 – DSC first and second heating curves of the full BioABS blend

F E D C B 20 70 120 170 A A – Quench to 150°C B – Quench to 140°C

C – Quench to 130°C Heat Flow (Exo up) (Exo Flow Heat D – Quench to 120°C E – Quench to 110°C F – Quench to 100°C G – Quench to 90°C Temperature (°C)

Figure 4.11 – DSC heating curves of the full blend annealed at different temperatures

105

Table 4.1 – Mechanical properties of BioABS molded at different temperatures

Mold Tensile Tensile Elongation Impact HDT

Temperature Strength Modulus at Break Strength

˚C MPa GPa % J/m ˚C

30 58.8±0.45 2.71±0.04 36.2±4.69 300±13 74.2±0.5

90 55.1±1.66 2.61±0.03 14.7±8.23 488±42 96.4±0.7

The atomic force microscopy (AFM) images of the blends are given in Figure 4.12. Utilizing the nano-mechanical mapping feature of the AFM, the different phases could be clearly separated.

These particular images are mapping of modulus of the surface as measured by the AFM tip.

Areas of low stiffness (darker) and areas of high stiffness (lighter) are an effective way to define the phases of PLA and ABS. From these images, the nano-structured morphology of the blends are easily seen. The sum of the effects of the additives and processing conditions on the blend morphology create encapsulated ABS phases with nano-sized shell of PLA. In some areas, the thickness of PLA was measured to be under 20nm. This is the foundation of the high performance properties of the blends. Due to the very small thickness of PLA, a plane-strain to plane-stress transition occurs, allowing the energy to be dispersed with very high effectiveness.

106

ABS Phase PLA Phase

10-15nm

Figure 4.12 – AFM modulus mapping of the PLA/ABS/Acrylic copolymer/Chain extender microstructure

4.4 Conclusions

The uncompatibilized blends, as expected, had very poor performance values, especially impact strength and elongation at break. The very weak interface was not able to absorb stress energy applied and distribute it throughout the structure. The addition of the acrylic copolymer increased the toughness in terms of elongation at break, and impact strength. This was done through increasing the mobility of the chains. With increased mobility, the chains are more able to slide past each other, allowing more absorbing of energy with subjected to an applied stress. This was

107

shown through rheology where the complex viscosity was decreased with the addition of the acrylic copolymer. Additionally, the morphological studies indicated a better dispersion and decreased tension of the blends. The addition of the chain extender allowed the PLA chains to form bonds between chains, increasing toughness. This resulted in an increase in the complex viscocity over the frequency span investigated. Again, the morphology showed a better dispersion, which increased the performance, most notably the impact strength and toughness.

The incorporation of both acrylic copolymer and chain extender allowed the PLA/ABS blend to be more thermodynamically stable, creating a morphology with decreased tension, better dispersion, and vastly improved mechanical properties. The additives worked synergistically to create this blend, which resulted in vastly improved performance over any other blend system.

These blends exhibited an impact strength of over 200 j/m and an elongation at break of over

100%. Due to the effectiveness of both additives, their low quantities allow the material to be a viable option for many applications. In this system, because the additive content are very low (≤2 wt. %), and there is no post processing or advanced processing steps, the cost of producing such a material are very low, giving it an advantage over many similar pla based blend materials with high toughness.

4.5 Acknowledgements

The financial support from the Ontario Ministry of Agriculture and Food Rural Affairs

(OMAFRA)/University of Guelph - Bioeconomy for Industrial Uses Research Program (Project

#200245); the Natural Sciences and Engineering Research Council (NSERC, Canada Discovery grants (Project #400322) and NSERC- AUTO21 NCE (Project #400372 & 400373); and Ontario

Research Fund, Research Excellence Program; Round-4 (ORF-RE04) from the Ontario Ministry of Economic Development and Innovation (MEDI) (Project #050289) to carry out this research

108

is gratefully acknowledged. The authors would like to thank Dr. Jean-Mathieu Pin of BDDC,

University of Guelph for his valuable suggestions during the preparation of this manuscript.

4.6 References

1. Poly(lactic acid): Synthesis, Structures, Properties, Processing, and Applications. John

Wiley and Sons, 2011

2. Rasal, R. M., Janorkar, A. V. and Hirt, D. E. Poly(lactic acid) modifications. Prog.

Polym. Sci. 35, 338–356.

3. Al-Itry, R., Lamnawar, K. and Maazouz, A. Improvement of thermal stability, rheological and mechanical properties of PLA, PBAT and their blends by reactive extrusion with functionalized epoxy. Polym. Degrad. Stab. 97, 1898–1914 (2012).

4. Saeidlou, S., Huneault, M. A., Li, H. and Park, C. B. Poly(lactic acid) crystallization.

Prog. Polym. Sci. 37, 1657–1677, 2012.

5. Lim, L.-T., Auras, R. and Rubino, M. Processing technologies for poly(lactic acid). Prog.

Polym. Sci. 33, 820–852, 2008.

6. Lasprilla, A. J. R., Martinez, G. A. R., Lunelli, B. H., Jardini, A. L. and Filho, R. M.

Poly-lactic acid synthesis for application in biomedical devices - a review. Biotechnol. Adv. 30,

321–8 (2012).

7. Garlotta, D. A Literature Review of Poly(Lactic Acid). J. Polym. Environ. 9, 63–84

8. Liu, H. and Zhang, J. Research progress in toughening modification of poly(lactic acid).

J. Polym. Sci. Part B Polym. Phys. 49, 1051–1083, 2011.

9. Anderson, K. S., Schreck, K. M. and Hillmyer, M. A. Toughening Polylactide. Polym.

Rev. (2008).

109

10. Kfoury, G., Raquez, J-M., Hassouna, F., Odent, J., Toniazzo, V., Ruch, D., and Dubois,

P. Recent advances in high performance poly(lactide): from ‘green’ plasticization to super-tough materials via (reactive) compounding. Front. Chem. 1, 32 (2013).

11. Barentsen, W. M. and Heikens, D. Mechanical properties of polystyrene/low density polyethylene blends. Polymer. 14, 579–583 (1973).

12. Lovinger, A. J. and Williams, M. L. Tensile properties and morphology of blends of polyethylene and polypropylene. J. Appl. Polym. Sci. 25, 1703–1713 (1980).

13. Mansfield, M. and Boyd, R. H. Molecular motions, the α relaxation, and chain transport in polyethylene crystals. J. Polym. Sci. Polym. Phys. Ed. 16, 1227–1252 (1978).

14. Hale, W. R., Pessan, L. A., Keskkula, H. and Paul, D. R. Effect of compatibilization and

ABS type on properties of PBT / ABS blends. Chem. Eng. 40, 4237–4250 (1999).

15. Kudva, R. A. Properties of compatibilized nylon 6/ABS blends. Polymer (Guildf). 41,

225 – 237

16. Zhang, X. et al. Effects of ABS-g-MAH on mechanical properties and compatibility of

ABS/PC alloy. J. Appl. Polym. Sci. 81, 831–836 (2001).

17. Dong, W.; He, M.; Wang, H.; Ren, F.; Zhang, J.; Zhao, X.; Li, Y. ACS Sustain. Chem.

2015, 3, 2542–2550.

18. Vadori, R. Studies On The Reactive Blending Of Poly(Lactic Acid) And Acrylonitrile

Butadiene Styrene Rubber. (2012).

19. Choe, I.-J., Lee, J. H., Yu, J. H. and Yoon, J.-S. Mechanical properties of acrylonitrile- butadiene-styrene copolymer/poly( l -lactic acid) blends and their composites. J. Appl. Polym.

Sci. 131, (2014).

110

20. Li, Y. and Shimizu, H. Toughening of polylactide by melt blending with a biodegradable poly(ether)urethane elastomer. Macromol. Biosci. 7, 921–8 (2007).

21. Sun, S., Zhang, M., Zhang, H. and Zhang, X. Polylactide toughening with epoxy- functionalized grafted acrylonitrile-butadiene-styrene particles. J. Appl. Polym. Sci. 122, 2992–

2999 (2011).

22. Utracki, Leszek A. Polymer blends handbook. Ed. Charles A. Wilkie. Springer

Reference, 2014.

23. Utracki, L. A. Compatibilization of Polymer Blends. 80, 1008–1016 (2002).

24. Al-Itry, R., Lamnawar, K. and Maazouz, A. Rheological, morphological, and interfacial properties of compatibilized PLA/PBAT blends. Rheol. Acta 53, 501–517 (2014).

25. Wu, S. Formation of dispersed phase in incompatible polymer blends: Interfacial and rheological effects. Polym. Eng. Sci. 27, 335–343 (1987).

26. Harada, M., Iida, K., Okamoto, K., Hayashi, H. and Hirano, K. Reactive compatibilization of biodegradable poly(lactic acid)/poly(ɛ-caprolactone) blends with reactive processing agents. Polym. Eng. Sci. 48, 1359–1368 (2008).

27. Ojijo, V.; Ray, S. Polymer. 2015, 80, 1–17

28. Doi, Masao. Introduction to polymer physics. Oxford university press, 1996.

29. Gedde, U. Polymer Physics; Springer Science & Business Media, 1995.

30. Li, Y. and Shimizu, H. Improvement in toughness of poly(l-lactide) (PLLA) through reactive blending with acrylonitrile–butadiene–styrene copolymer (ABS): Morphology and properties. Eur. Polym. J. 45, 738–746 (2009).

31. Yang, J., Zhang, Y. and Zhang, Y. Brittle–ductile transition of PP/POE blends in both impact and high speed tensile tests. Polymer. 44, 5047–5052 (2003).

111

32. Doi, M. and Edwards, S. F. Dynamics of concentrated polymer systems. Part 1. Brownian motion in the equilibrium state. J. Chem. Soc. Faraday Trans. 2 74, 1789 (1978).

33. Wu, C.-P., Wang, C.-C. and Chen, C.-Y. Enhancing the PLA Crystallization Rate and

Mechanical Properties by Melt Blending with Poly(styrene-butadiene-styrene) Copolymer.

Polym. Plast. Technol. Eng. 54, 1043–1050 (2015).

34. Najafi, N., Heuzey, M. C., Carreau, P. J. and Wood-Adams, P. M. Control of thermal degradation of polylactide (PLA)-clay nanocomposites using chain extenders. Polym. Degrad.

Stab. 97, 554–565 (2012).

35. Liu, H.; Chen, F.; Liu, B.; Estep, G.;Zhang, J. 2010 Macromolecules 43, 6058–6066.

36. Li, H. and Huneault, M. A. Effect of nucleation and plasticization on the crystallization of poly(lactic acid). Polymer (Guildf). 48, 6855–6866 (2007).

37. Zhang, L., Xiong, C. and Deng, X. Miscibility, crystallization and morphology of poly(β- hydroxybutyrate)/poly(d,l-lactide) blends. Polymer. 37, 235–241 (1996).

38. Gu, S.-Y., Zhang, K., Ren, J. and Zhan, H. Melt rheology of polylactide/poly(butylene adipate-co-terephthalate) blends. Carbohydr. Polym. 74, 79–85 (2008).

39. Ferry, John D. Viscoelastic properties of polymers. John Wiley & Sons, 1980.

40. Dealy, J. M.; Wissbrun, K. Melt Rheology and Its Role in Plastics Processing: Theory and Applications. 27, (Springer Science and Business Media, 2013).

41. Davies, M. C., Shakesheff, K. M., Shard, A. G., Domb, A., Roberts, C. J., Tendler, S. J.

B., & Williams, P. M. Surface Analysis of Biodegradable Polymer Blends of Poly(sebacic anhydride) and Poly( dl -lactic acid). Macromolecules 29, 2205–2212 (1996).

42. Park, S. D., Todo, M. & Arakawa, K. Effect of Annealing on Fracture Mechanism of

Biodegradable Poly(lactic acid). Key Eng. Mater. 261-263, 105–110 (2004).

112

43. Wu, S. A generalized criterion for rubber toughening: The critical matrix ligament thickness. J. Appl. Polym. Sci. 35, 549–561 (1988).

44. Oyama, H. T. Super-tough poly(lactic acid) materials: Reactive blending with ethylene copolymer. Polymer 50, (2009).

113

Chapter 5: Bio-Based Acrylonitrile Butadiene Styrene (ABS) Polymer Compositions and Methods Of Making And Using Thereof

______

Abstract

The formulation that was discovered in chapter 4 is developed into a full range of material formulations with differing mechanical properties. This development culminated into a patent application, which is presented in the current chapter. The invention refers to the formulation incorporating PLA, ABS, Biostrength acrylic copolymer and Joncryl chain extender. A full range of forumulations are characterized for mechanical properties, from 25 to 70 wt. % PLA.

Additionally, the scale-up feasibility is investigated through melt blending in pilot scale equipment, as opposed to lab scale.

______

114

5.1 Field Of The Invention

The present invention is in the field of acrylonitrile butadiene styrene (ABS) blends that incorporate a significant portion of renewable content through blending with poly (lactic acid)

(PLA), and methods of making and using thereof.

5.2 Background of the Invention

Acrylonitrile Butadiene Styrene (ABS) was first discovered during World War II when its basis, Styrene Butadiene Rubber (SBR), was used for alternatives to rubber. Commercially

ABS polymers first became available in the early 1950s in an attempt to obtain the best properties of both polystyrene and styrene acrylonitrile.

ABS, however, has several disadvantages. ABS is (1) sensitive to thick sections in articles of manufacture which may cause voids, bubbles or sink; (2) susceptible to attack by hydrocarbons and other organic solvents; (3) exhibits low heat resistance and limited weather resistance; (4) relatively expensive to manufacture; and (5) is petroleum resource based.

There exists a need for materials, such as composite materials, that exhibit similar or improved mechanical properties, such as impact resistance and/or heat resistance, compared to neat ABS while being less expensive to manufacture, and from a resource base that is sustainable.

Numerous efforts have been made to increase the sustainability of traditional petroleum based polymers including blending with renewable resource based polymers. The preliminary research on ABS/PLA blends has showed poor performance due to incompatibility between the two. Several attempts have been made to increase the compatibility through the use of additives.

115

United States Patent Application No US/20120252977 describes a PLA composition in which PLA is modified with ABS and at least one compatibilizer selected from selected from the group consisting of: a poly (styrene-ethylene-butadiene-styrene) copolymer grafted with maleic anhydrides, an acrylonitrile-butadiene-styrene copolymer grafted with maleic anhydrides, a polystyrene grafted with maleic anhydrides, and an ethylene-ethyl acrylate-glycidyl methacrylate. In this application the PLA based blends include from about 34% to 47% by weight (wt.%) PLA, while the ABS can be anywhere from 15 wt.% to 70 wt.%. The highest impact strength alleged is 101 J/m with this system. The blends do not include a lubricant, including an acrylic copolymer based lubricant.

European Patent Application No EP/2706090 describes a blended polymer system that contains polycarbonate (PC), ABS and PLA, and additives such as flame retardants and modified talc. Unfortunately, mechanical properties are not given for these blends, only crystallinity, heat distortion, and flame retardancy are investigated. The blends do not include a lubricant, including an acrylic copolymer based lubricant.

European Patent Application No EP/2700678 describes a biodegradable polymer composite material comprising a biodegradable resin (PLA, polyhydroxybutarate, or polycaprolactone), ABS, and a reactive compatibilizer, such as glycidyl methacrylate, or maleic anhydride. Some improvement was seen relative to PLA properties, but only when high ABS amounts were used. The blends do not include a lubricant, including an acrylic copolymer based lubricant.

United States Patent Application No US/20120220711 describes heat resistant PLA/ABS blends. The blends of this document include an oligomeric chain extender to increase adhesion, thereby increasing the heat distortion temperature. The blends may optionally include a

116

stabilizer, an impact modifier and a filler. The blends do not include a lubricant, including an acrylic copolymer based lubricant. The best blend outlined in this document has an impact strength of 102 J/m with a heat distortion temperature of 98.1°C. According to this application, the PLA should be the “significant component”, meaning that PLA is present in at least thirty weight percent (30 wt.%) or more of the composition.

United States Patent Application No US 2012/0184672 describes the use of additives to increase the impact resistance and/or heat deflection temperature. The use of an oligomeric chain extender in addition to several other additives, such as plasticizer, fillers, impact modifier, and lubricant are employed. The PLA content in these blends is about 50 wt.% or more. The blends do not include ABS, or a lubricant, including an acrylic copolymer based lubricant.

Li et al.1, European Polymer Journal 45(3) 738-746 (2009) describes the use of styrene- acrylonitrile (SAN) grafted with glycidyl methacrylate (GMA), which is denoted SAN-GMA to act as an intermediary between the two phases. This paper describes a decrease in the size of the dispersed phase (ABS) upon reactive blending, which led to an increase in the toughness of the blend. This increase, however, was modest for the amount of additive that was used. The blends do not include a lubricant, including an acrylic copolymer based lubricant.

Jo et al.2, Journal of Applied Polymer Science 125 (S2) 231-238 (2012) again attempts to use additives during blending to improve the performance of PLA/ABS blends. Although many different additives were attempted, the researchers showed that the best performance was attained with SAN-GMA in addition to a small amount of thermal stabilizer. However, a great deal of additive was used to achieve a modest improvement, their best blend being over 20 wt.% additive The blends do not include a lubricant, including an acrylic copolymer based lubricant.

117

Sun et al.3, Journal of Applied Polymer Science 122(5) 2992-2999 (2011) explains grafting GMA onto ABS before using it to toughen PLA. The improvement in the impact resistance achieved through this method was substantial. However, the method used to graft the

GMA onto the ABS involved polymerizing the ABS from constituent monomers. This adds complexity and cost to the processing of such materials, decreasing their value and potential for use in commercial markets. The blends do not include a lubricant, including an acrylic copolymer based lubricant.

Chevali et al.4, Polymer-Plastics Technology and Engineering 54 375-382 (2015) describes blending ABS with two natural fibers to form hybrid lignocellulosic biocomposites.

With the inclusion of 20 wt.% fiber into the ABS matrix, impact resistance was reduced while the modulus was increased compared to neat ABS. The blends do not include a lubricant, including an acrylic copolymer based lubricant.

Yeh et al.5, Composites Science and Technology 69 2225-2230 (2009) describes blending

ABS with wood flour to form ABS based wood-plastic composites. The composites contained 50 wt.% wood fiber. With the use of coupling agents between the fiber and matrix, they were able to increase the properties of the composites compared to without coupling agent. The blends do not include a lubricant, including an acrylic copolymer based lubricant.

The art described above details blending of ABS with PLA and other bio-based plastics for increased sustainability. However, blends achieving performance similar to ABS has remained difficult to achieve.

There is a need for ABS based materials that incorporate some amount of bio-content to increase the sustainability of the material, and to exhibit performance values that are similar to

ABS materials, and methods of making thereof.

118

5.3 Summary of the Invention

Acrylonitrile butadiene styrene (ABS)-based blends containing polylactic acid (PLA) and methods of making and using thereof are described herein. Composites containing the blend are also described herein.

In one embodiment, the blends of the present invention contain ABS in an amount from about 50 wt.% to about 85 wt.% of the entire blend and any range there in between, for example from about 60 wt.% to about 80 wt.%.

In another embodiment, the blends contain up to about 30 wt.% PLA. In other embodiments, the blends contain more than 30 wt.% PLA.

In another embodiment, the blends of the present invention contain one or more additives. In some embodiments, the blends contain a lubricant in combination with one or more chain extenders.

In one embodiment, the concentration of the lubricant is from about 1 wt.% to about 10 wt.% and any range there in between, such as from about 2 wt.% to about 5 wt. %.

In some embodiments, the chain extender is an epoxy-functionalized styrene-acrylic oligomers/polymers, such as those available under the tradename JONCRYL. In some embodiments, the chain extender contains glycidyl methacrylate (GMA). Exemplary chain extenders of this type include, but are not limited to, JONCRYL ADR-4368-C/CS. In some embodiments, the concentration is from about 0.1 wt.% to about 2 wt.% and any range there in between, for example from about 0.5 wt.% to about 1.5 wt.%, or about 1 wt.%.

In one embodiment, the blends of the present invention exhibit a Notched Izod Impact Strength of at least about 140 J/m. In another embodiment, the blends of the present invention exhibit a

Notched Izod Impact Strength ranging from 140 J/m to 600 J/m.

119

In one embodiment, the blends of the present invention exhibit a tensile strength of at least about

40 MPa. In another embodiment, the blends of the present invention exhibit a tensile strength ranging from about 40 MPa to about 50 MPa.

In one embodiment, the blends of the present invention exhibit a flexural strength of at least about 60 MPa. In another embodiment, the blends of the present invention exhibit a tensile strength ranging from about 60 MPa to about 78 MPa

In another embodiment the present invention provides for composites including the blends of the previous embodiments and one or more additives, such as fillers. Classes of fillers include natural fibers, mineral fillers, and combinations thereof. In particular embodiments, the composite contains one or more natural fibers.

The blends and/or composites of the present invention can be used to form a variety of articles of manufacture, such as injection molded articles, such as car parts, toys, consumer products, building materials, etc.

In another embodiment, the present invention provides for a method of making a high performance acrylonitrile butadiene styrene (ABS). The method includes: (a) forming a mixture of ABS, polylactic acid (PLA), an acrylic copolymer based lubricant and a polymeric chain extender polymer; and (b) extruding the mixture at a temperature of 230oC to 250oC to form the high performance ABS polymer blend.

5.4 Brief Description of the Drawings

The invention will be better understood and objects of the invention will become apparent when consideration is given to the following detailed description thereof. Such description makes reference to the annexed drawings wherein:

120

FIG. 1 is a scanning electron microscopy (SEM) image of the impact fracture surface of example blend 6A (see Example 6).

FIG. 2 is a scanning electron microscopy (SEM) image of the impact fracture surface of example blend 1B (see Example 1).

FIG. 3 is a Bruker Multimode atomic force microscope (AFM) image showing the structure of the morphology of example blend 6B (see Example 6).

FIG. 4 is a Bruker Multimode atomic force microscope (AFM) image showing the structure of the morphology of example blend 1B (see Example 1).

5.5 Detailed Description Of The Invention

5.5.1 Definitions

In this specification and in the claims that follow, reference will be made to a number of terms that shall be defined to have the meanings below. All numerical designations, e.g., temperatures, concentrations, dimensions and weight, including ranges, are approximations that typically may be varied ( + ) or ( - ) by increments of 0.1, 1.0, or 10.0, as appropriate. All numerical designations may be understood as preceded by the term “about”.

The singular form “a”, “an”, and “the” includes plural references unless the context clearly dictates otherwise.

The term “comprising” means any recited elements are necessarily included and other elements may optionally be included. “Consisting essentially of” means any recited elements are necessarily included, elements that would materially affect the basic and novel characteristics of the listed elements are excluded, and other elements may optionally be included. “Consisting of”

121

means that all elements other than those listed are excluded. Embodiments defined by each of these terms are within the scope of this invention.

“Composite”, as used herein, generally means a combination of two or more distinct materials, each of which retains its own distinctive properties, to create a new material with properties that cannot be achieved by any of the components acting alone.

The prefix “bio-” as used herein refers to a material that has been derived or partially derived from a renewable resource.

The term “renewable resource”, as used herein, refers to a resource that is produced by a natural process at a rate comparable to its rate of consumption (e.g., within a 100 year time frame). The resource can be replenished naturally or via agricultural techniques.

The term “bio-based content”, as used herein, refers to the amount of bio-plastic in a material as a percent of the weight (mass) of the total weight (mass) in the product.

“Recyclable”, as used herein, refers to a product or material that can be reprocessed into another, similar or often different products.

“Blend”, as used herein, means a macro-homogeneous mixture of two or more different polymers. The resultant blend may contain distinct phases of its components.

The terms “heat deflection temperature” or “heat distortion temperature” (HDT) are used interchangeably and refer to the temperature at which a polymer or plastic sample deforms under a specified load. The heat distortion temperature is determined by the following test procedure outlined in ASTM D648. The test specimen is loaded in three-point bending in the edgewise direction. The two most common loads are 0.455 MPa or 1.82 MPa and the temperature is increased at 2°C/min until the specimen deflects 0.25 mm.

122

“Impact strength”, as used herein, refers to the capability of a material to withstand a suddenly applied load and is expressed in terms of energy. Impact strength is typically measured with the Izod impact strength test or Charpy impact test, both of which measure the impact energy required to fracture a sample. Izod impact testing is an ASTM standard method of determining the impact resistance of materials. An arm held at a specific height (constant potential energy) is released. The arm hits the sample and breaks it. From the energy absorbed by the sample, its impact energy is determined. A notched sample is generally used to determine impact energy and notch sensitivity.

The terms “super tough” and “non-breakable” are used interchangeably and refer to a polymer blend which shows a no break notched Izod impact behavior, as determined according to ASTM Standard D256.

The term “non-break”, as used herein, refers to an incomplete break where the fracture extends less than 90% of the distance between the vertex of the notch and the opposite side as per ASTM D256. Results obtained from the non-break specimens shall not be reported as per

ASTM D256.

5.5.2 Acrylonitrile butadiene styrene (ABS)/polylactic acid (PLA)-based blends

5.5.2.1 Acrylonitrile butadiene styrene (ABS)

Acrylonitrile butadiene styrene (ABS) is a common thermoplastic. Its glass transition temperature is approximately 105°C (221°F). ABS is amorphous and therefore has no true melting point.

Acrylonitrile-butadiene-styrene has the formula of (C8H8)x(C4H6)y(C3H3N)z), wherein x is a number resulting in the ABS having from about 40 to about 60 weight percent of styrene content, y is a number to resulting in the ABS having from about 5 wt.% to about 45 wt.% of

123

butadiene content, and z is a number resulting in the ABS having from about 15 wt.% to about

35 wt.% of acrylonitrile content. ABS can be recycled, an important property considering its use with PLA.

ABS is a terpolymer made by polymerizing styrene and acrylonitrile in the presence of polybutadiene. The result is a long chain of polybutadiene criss-crossed with shorter chains of poly(styrene-co-acrylonitrile). The nitrile groups from neighboring chains, being polar, attract each other and bind the chains together, making ABS stronger than pure polystyrene. The styrene gives the plastic a shiny, impervious surface. The polybutadiene, a rubbery substance, provides resilience even at low temperatures.

ABS can be functional through a temperature range of -40C to 90C.

For the majority of applications, ABS can be used between −20°C and 80°C (−4 and 176 °F) as its mechanical properties vary with temperature. The properties are created by rubber toughening, where fine particles of elastomer are distributed throughout the rigid matrix.

ABS is commercially available from a number of companies including Dow Chemical

Co., LG Chemical Company, Sabic Innovative Plastics, and BASF. These commercially available ABS polymers are not entirely pure resins. As a part of the manufacturing process, particularly the emulsion polymerization process, there are surfactants and other minor ingredients used to facilitate polymerization of the ABS. Because these trace amounts of surfactants remain a part of the polymer resin when sold commercially, their presence can have a positive or negative effect on the mixing of such resins with PLA. Commercially available ABS that may be used in the blend of the present invention include those sold under the trademark

MAGNUM®, including the high impact ABS MAGNUM® grade 1150 EM.

124

The concentration of ABS in the blend of the present invention may be from about 20 wt.% to about 80 wt.% of the blend, preferably from about 70 wt.% to about 80 wt.% of the blend.

5.5.2.2 Polylactic acid (PLA)

Polylactic acid (PLA) is a renewable polymer derived from naturally sourced monomers and derivatives thereof. PLA is a commercially-available polyester-based resin made using lactic acid. The lactic acid may be obtained, for example, by decomposing biomass, such as corn starch, to obtain the monomer. In some embodiments, the PLA may be a homopolymers of lactic acid, including poly(L-lactic acid) (PLLA), in which the monomer unit is L-lactic acid, poly(D- lactic acid) (PDLA), in which the monomer unit is D-lactic acid, and poly(D,L-lactic acid) in which the monomer structure units are D,L-lactic acid, that is, a mixture in various proportions

(e.g., a racemic mixture) of D-lactic acid and L-lactic acid monomer units. In other embodiments, the PLA is a stereocomplex PLLA and PDLA. In other embodiments, polylactic acid resins which are crosslinked may be used.

In other embodiments, the PLA is a copolymer of lactic acid containing at least about 50,

60, 70, 80, or 90 wt.% lactic acid comonomer content based on the weight of the copolymer and containing one or more comonomers other than lactic acid comonomer in amounts of less than

50, 40, 30, 20, or 10 wt.%, of the copolymer. Exemplary comonomers include hydroxycarboxylic acids other than lactic acid, for example, one or more of any of the following hydroxycarboxylic acids: glycolic acid, hydroxybutyrate (e.g., 3-hydroxybutyric acid, 4- hydroxybutyric acid), hydroxyvaleric acid (e.g., 4-hydroxyvaleric acid, 5-hydroxyvaleric acid) and hydroxycaproic acid (e.g., 6-hydroxycaproic acid).

125

In one embodiment, PLA may be virgin PLA. In some embodiments, the PLA has high optical purity. Using PLA of high optical purity may improve the HDT of the composites prepared from PLA. The weight average molecular weight of the PLA can vary. However, in some embodiments, the average molecular weight of the PLA is from about 10,000 and 500,000

Dalton, preferably from about 10,000 to about 300,000 Daltons.

In one embodiment, the concentration of PLA in the blend may range from about 10 wt.% to about 70 wt.% of the blend. More than 70 wt.% of blend may also be used. In another embodiment, the PLA may be provided as a minor component in the blend compared to ABS.

“Minor component” as used herein means that the concentration of PLA is less than about 30 wt.% of the blend, i.e. no more than about 30 wt.%, but not including 30 wt.%. When provided as a minor component of the blend, the concentration of polylactic acid is from about 10 wt.% to about 29 wt.% of the composition and any concentration in between, such as from about 15 wt.% to about 20 wt.% of the composition.

In some embodiments, PLA may be generated as post-consumer and post industrial waste, which can be used in place of virgin PLA or in combination with virgin PLA, may also be used in the blends and composites described herein. In those embodiments where recycled PLA is used, the recycled PLA has a relatively high weight average molecular weight, such as at least about 50,000, 60,000, 70,000, 75,000, 85,000, 90,000, 95,000, or 100,000 Daltons. In some embodiments, the weight average molecular weight of PLA is from about 5,000 Daltons to about

100,000 Daltons. In other embodiments, the weight average molecular weight is from about

70,000 to about 100,000 Daltons. In particular embodiments, the PLA is semi-crystalline and has the molecular weight described above. In those embodiments wherein recycled PLA is used in combination with virgin PLA, the concentration of recycled PLA is from about 10 wt.% to

126

about 30 wt.% of the combination of recycled PLA and virgin PLA. Virgin or neat PLA refers to formulations containing only PLA. The impact strength of virgin or neat PLA is 31 J/m and its HDT (under a 455 kPa flexural load) is 55C.

5.5.2.3 Lubricant

In some embodiments, the blend of the present invention includes one or more lubricants.

“Lubricant”, as used herein, refers to an additive introduced to improve the metal release and enhance processability of PLA. In particular embodiments, the lubricant is an acrylic copolymer based lubricant such as that available under the trade name BIOSTRENGTH900. The concentration of the lubricant may vary. In some embodiments, the concentration of the lubricant may be from about 1 wt.% to about 10 wt.% of the composition and any range in between, for example from about 2 wt.% to about 5 wt.% of the composition.

BIOSTRENGTH900 is described as being effective in minimizing sticking of PLA to metal during demanding calendaring and injection molding processes. However,

BIOSTRENGTH900 has not been described to change mechanical properties of a blend such as impact strength, tensile strength and flexural strength.

Polymeric chain extender

In some embodiments, the blend includes a lubricant, as described above, in combination with a chain extender. “Chain extender”, as used herein, refers to a molecule that increases the molecular weight of one or more polymers in the blend. In some embodiments, the chain extender is a polymeric chain extender that increases the molecular weight in the blend.

Increasing the molecular weight of the PLA increases the melt viscosity, melt strength, and intrinsic viscosity of the polymer.

127

Suitable polymeric chain extenders include, but are not limited to, epoxy-functionalized styrene- acrylic oligomers/polymers available under the tradename JONCRYL. In some embodiments, the JONCRYL chain extender contains glycidyl methacrylate (GMA). Exemplary

JONCRYL materials of this type include, but are not limited to, JONCRYL ADR-4368-C/

CS.

The concentration of the chain extender in the blends of the present invention may vary. In some embodiments, the concentration of the chain extender may be from about 0.1 wt.% to about 2 wt.% of the composition and any range in between, like from about 0.5 wt.% to about 1.5 wt.% of the composition, more preferably about 1 wt.% of the composition.

5.5.2.4 Properties of the blends

Some blend compositions described herein exhibit similar or improved mechanical properties compared to neat or unmodified ABS of two different grades. For example, a summary of various mechanical properties for unmodified ABS and exemplary blends of the present invention is shown in Example 1. Some of the present invention exhibit improved Notched Izod

Impact Strength compared to unmodified ABS, as well as similar or improved tensile strength and improved flexural strength.

5.5.3 Composites

The blends described herein can be used to prepare ABS-based composites. The composites may be prepared by combining the blends described above with one or more additives selected from fillers, such as natural fibers and/or mineral fillers to form the composites.

128

5.5.3.1 Natural fibers/biofillers

The composites can contain one or more natural fibers. Exemplary natural fibers include, but are not limited to, bast fibers, leaf fibers, grass fibers (perennial grasses), straw fibers (agricultural residues), and seed/fruit fibers. In addition, pyrolyzed biomass based chars may be used as biofillers. These chars are herein referred to collectively as BioChar.

The composites may also include one or more residues (biofillers) from industrial processes, such as those in the food industry. Exemplary residues include, but are not limited to, coffee chaff, coffee grind, food or leftover residues.

Perennial grasses include, but are not limited to, switchgrass and miscanthus. Agricultural residues include, but are not limited to, soy stalk, wheat straw, corn stover, soy hull, and oat hull.

Perennial grasses and agricultural residues include, but are not limited to, lignocellulosic fibers having about 35% cellulose and other constituents such as hemicellulose, lignin, pectin, protein and ash.

The natural fibers can have an average length from about 2 to about 10 mm, preferably from about 2 to about 6 mm, particularly for injection molding processes. However, fibers less than 2 mm and greater than 6 mm or 10 mm may also be used. Natural fibers may also be in a particulate form, particles may be as small as 50 μm or as large as 1 mm.

Natural fibers can be added to the ABS-based blend system directly with or without any surface treatment (e.g., devoid of surface treatment, such as chemical treatment) to achieve the desired performance. A mixture of two more fibers can also be used. This may especially be important in the case of fiber supply chain issues that can arise while using one particular type of fiber.

The content of the natural fiber(s) in the composite may be from about 0 wt.% to about 35 wt.% and any range in between, such as from about 0 wt.% to about 30 wt.% of the composite, more

129

preferably from about 0 wt.% to about 25 wt.%, most preferably from about 0 wt.% to about 15 wt.%. When the natural fibers are present, the concentration can be from about 10 wt.% to about

30 wt.% of the composite, preferably from about 10 wt.% to about 25 wt.% of the composite, more preferably from about 10 wt.% to about 20 wt.% of the composite.

The use of natural fiber also reduces the cost of the final formulation, as up to 30 wt.% of the composite can be replaced with these fibers as per the property requirements of the end product.

5.5.3.2 Mineral fillers

The composite can also contain one or more mineral fillers. Suitable mineral fillers include those known to be useful in the compounding of polymers. Exemplary mineral fillers include, but are not limited to, talc, calcium carbonate, calcium sulphate, mica, magnesium oxysulphate, silica, kaolin and combinations thereof.

The content of the mineral filler(s) in the PLA-based composite may be from about 0 wt.% to about 25 wt.% of the composite. When mineral filler is present, it can be present in an amount from about 5 wt.% to about 25 wt% of the composite, preferably from about 5 wt.% to about 20 wt. % of the composite, more preferably from about 5 wt.% to about 15 wt.% of the composite.

5.5.4 Method of manufacturing blends and composites containing the blends

5.5.4.1 Polylactic acid

Poly (lactic acid) can be made using a variety of techniques known in the art, such as polycondensation. In the polycondensation method, L-lactic acid, D-lactic acid, or a mixture of these, or lactic acid and one or more other hydroxycarboxylic acids, may be directly subjected to dehydropolycondensation to obtain a polylactic acid of a specific composition. For example, in the direct dehydration polycondensation process the lactic acid or other hydroxycarboxylic acids

130

may be subjected to azeotropic dehydration condensation in the presence of an organic solvent, such as a diphenyl ether-based solvent. Such polymerization reaction, for example, may progress by removing water from the azeotropically distilled solvent and returning substantially anhydrous solvent to the reaction system.

Polylactic acid may also be made by ring-opening polymerization methods. In the ring- opening polymerization method, lactide (i.e., cyclic dimer of lactic acid) is polymerized via a polymerization adjusting agent and a catalyst to obtain polylactic acid. Lactide includes L-lactide

(i.e., dimer of L-lactic acid), D-lactide (i.e., dimer of D-lactic acid), DL-lactide (i.e., mixture of

L- and D-lactides), and meso-lactide (i.e., cyclic dimer of D- and L-lactic acids). These isomers can be mixed and polymerized to obtain polylactic acid having a desired composition and crystallinity. Any of these isomers may also be copolymerized by ring-opening polymerization with other cyclic dimers (e.g., glycolide, a cyclic dimer of glycolic acid) and/or with cyclic esters such as caprolactone, propiolactone, butyrolactone, and valerolactone.

5.5.4.2 Blends

The blends can be prepared using techniques known in the art. In some embodiments, prior to the processing, all the components were dried, for example, at 60°C - 80°C for at least 4 h. In some embodiments, the blends can be prepared by extrusion. In some embodiments, the components of the blend were extruded at a temperature from about 230°C to about 250°C and any temperature in between, preferably from about 230°C to about 240°C. In one embodiment, the composites are prepared by co-extruding (a) acrylonitrile butadiene styrene (ABS); (b) poly

(lactic acid) (PLA), (c) one or more acrylic copolymer-based lubricants, and (d) one or more chain extenders.

131

The injection molding conditions can vary as well. In some embodiments, the injection molding conditions were as follows: melt temperature from about 220°C to about 240°C, mold temperature from about 30°C to 60°C and cooling time from about 30 to about 60 seconds.

In some embodiments, lab scale extrusions and injection moldings were performed on a micro twin-screw extruder and micro injection molder (DSM Research, Netherlands). The screw configuration in the extruder was co-rotating and was operated at a RPM of 100. Pilot scale extrusion can be carried out in a co-rotating twin-screw extruder (Leistritz, Germany) with a screw diameter of 27mm. Injection molding machine (Arburg, Germany) can be used for the pilot scale injection molding.

5.5.4.3 Composites

The composites can be prepared using techniques known in the art. The blend can be dried prior to extrusion to form the composite. In some embodiments, the blend is dried at a temperature from about 60°C to 80°C for a period of time from about 4 to about 6 hours.

In one embodiment, the composites are prepared by co-extruding a blend containing, with at least one filler (e.g., natural fibers and/or mineral fillers), nucleating agent, and/or chain extender. In those embodiments where the filler is or contains one or more natural fibers, the fiber may be added to the PLA-based blend directly without any surface treatment to achieve the desired performance. After extrusion, the extruded pellets can be dried, for example at 80° C for at least 42 hours.

In some embodiments, the PLA-based blend forms a matrix or continuous phase of the composite and the fillers and/or other additives form a dispersed phase.

The method may further include injection molding of the extrudate. The PLA-based composites may be used for manufacturing a molded article or product having enhanced impact

132

strength and enhanced HDT relative to neat or virgin PLA. The method of manufacturing a molded product may include a step of molding the above-described composites by injection molding, extrusion molding, blow molding, vacuum molding, compression molding, and so forth. The injection molding conditions may vary. However, in some embodiments, the injection molding conditions may be as follows: melt temperature from about 180°C to about

240°C, mold temperature from about 30°C to about 90°C, and cooling time from about 30 seconds to about 90 seconds.

Lab scale extrusion and injection molding can be done using a variety of equipment known in the art. In some embodiments, lab scale extrusions and injection moldings were performed on a micro twin-screw extruder and micro injection molder (DSM Research,

Netherlands). The screw configuration in the extruder was co-rotating and was operated at a

RPM of 100. Pilot scale extrusion can be carried out in a co-rotating twin-screw extruder

(Leistritz, US) with a screw diameter of 27mm. Two component injection molding machine

(Arburg, Germany) can be used for the pilot scale injection molding.

5.5.5 Methods of using the composites

The composites described herein can be used to prepare an article of manufacture that is made from plastics and or plastic/synthetic fillers and fibers. Examples include but are not limited to, injection molded articles, such as car parts, toys, consumer products, building materials, etc.

In one embodiment, the composites may contain up to 90 wt.% bio-based content. The

ABS-based composites may be derived from a combination of a renewable (e.g., derived from a renewable resource) material along with a recycled material, a regrind material, or mixtures

133

thereof. In some embodiments, the composite contains one or more fillers, such as natural fibers and/or mineral fillers.

5.6 Examples

5.6.1 Materials and Methods

The following materials and methods were, unless indicated otherwise, used for the

Examples.

Tensile and flexural properties of the neat ABS properties were measured using Instron

Instrument Model 3382 following ASTM standards D638 and D790, respectively. Tensile tests were carried out at room temperature with a gauge length of 50 mm and at a crosshead speed of

5 mm/min. A span length of 52 mm and crosshead speed of 14 mm/min were used for flexural tests. All blends were measured under the same conditions for comparison.

Notched Izod impact tests as per ASTM D256 at room temperature were accomplished using TMI 43-02 Monitor Impact Tester with a 5 ft-lb pendulum. Heat deflection temperature

(HDT) was evaluated using a dynamic mechanical analyzer (DMA Q800) supplied by TA

Instruments in three-point bending mode at a constant applied load of 0.455 MPa. The samples were heated from room temperature to the desired temperature at a ramp rate of 2°C · min−1.

HDT was reported as the temperature at which a deflection of 0.25 mm occurred. The results reported herein are average values obtained after testing at least 5 samples for tensile and flexural properties, 6 samples for impact strength and 2 samples for HDT.

134

Example 1. Preparation of acrylonitrile butadiene styrene (ABS)/polylactic acid (PLA) blend containing an acrylic copolymer-based lubricant and chain extender using lab scale equipment

Example blend 1A (Blend 1) was prepared containing 71.9 wt.% ABS (Styron-Magnum

1150EM), 26 wt.% PLA (PLA Ingeo3052D, Nature Works), 2 wt.% lubricant

(BIOSTRENGTH900), and 0.1 wt.% chain extender (JONCRYL ADR 4368 (BASF)). The blend was extruded in a DSM Micro 15 cc twin screw compounder, with a 240°C extrusion temperature, a 100 RPM screw speed, and a 2 minute retention time. The material was molded using a DSM micro 12 cc mini injection molder. Both injection and holding were set to 6 bar pressure, injection took 4 seconds, while holding took 8 seconds.

Example blend 1B (Blend 2) was prepared containing 71.9 wt.% ABS (Styron-Magnum

1150EM), 25 wt.% PLA (PLA Ingeo 3052D, Nature Works), 3 wt.% lubricant

(BIOSTRENGTH 900), and 0.1 wt.% chain extender (JONCRYL ADR 4368 BASF). The blend was extruded in a DSM Micro 15 cc twin screw compounder, with a 240°C extrusion temperature, a 100 RPM screw speed, and a 2 minute retention time. The material was molded using a DSM micro 12 cc mini injection molder. Both injection and holding were set to 6 bar pressure, injection took 4 seconds, while holding took 8 seconds.

Example blend 1C (Blend 3) was prepared containing 69.9 wt.% ABS (Styron-Magnum

1150EM), 28 wt.% PLA (PLA Ingeo 3052D, Nature Works), 2 wt.% lubricant

135

Example blend 1D (Blend 4) was prepared containing 67.9% ABS (Styron-Magnum 1150EM),

30 wt.% PLA (PLA Ingeo 3052D, Nature Works), 2 wt.% lubricant (BIOSTRENGTH 900), and 0.1 wt.% chain extender (JONCRYL ADR 4368 BASF). The blend was extruded in a

DSM Micro 15 cc twin screw compounder, with a 240°C extrusion temperature, a 100 RPM screw speed, and a 2 minute retention time. The material was molded using a DSM micro 12 cc mini injection molder. Both injection and holding were set to 6 bar pressure, injection took 4 seconds, while holding took 8 seconds.

Example blend 1E (Blend 5) was prepared containing 62.9 wt.% ABS (Styron-Magnum

1150EM), 35 wt.% PLA (PLA Ingeo 3052D, Nature Works), 2 wt.% lubricant

Example blend 1F (Blend 6) was prepared containing 57.9 wt.% ABS (Styron-Magnum

1150EM), 40 wt.% PLA (PLA Ingeo 3052D, Nature Works), 2 wt.% lubricant

Example blend 1G (Blend 7) was prepared containing 48 wt.% ABS (Styron-Magnum 1150EM),

50 wt.% PLA (PLA Ingeo 3052D, Nature Works), 1.5 wt.% lubricant (BIOSTRENGTH 900),

136

and 0.5 wt.% chain extender (JONCRYL ADR 4368 BASF). The blend was extruded in a

Example blend 1H (Blend 8) was prepared containing 28 wt.% ABS (Styron-Magnum 1150EM),

70 wt.% PLA (PLA Ingeo 3052D, Nature Works), 1.5 wt.% lubricant (BIOSTRENGTH 900), and 0.5 wt.% chain extender (JONCRYL ADR 4368 BASF). The blend was extruded in a

A comparison of the mechanical properties of neat ABS and the blend formed in

Example 1 is shown in Table 5.1:

137

Table 5.1 - Mechanical properties of neat ABS and ABS/PLA blend

Example Tensile Tensile Elongation at Flexural Flexural Impact Melt Flow Heat

Strength Modulus Break (%) Strength Modulus Strength Index Deflection

(MPa) (GPa) (MPa) (GPa) (J/m) (g/10min) Temperature

(°C)

ABS Lustran 39.9 ±3.68 1.58± 0.01 39.5± 5.4 65.3 ± 1.66 1.92± 0.03 414± 20 1.5 ± 0.3 89.8 ± 1.9

ABS Magnum 35.3 ± 2.87 1.90 ± 0.01 34.2 ± 3.9 57.7 ± 1.08 1.99 ± 0.02 507 ± 24 1.1 ± 0.2 93.5 ± 0.8

Blend 1 41.5 ± 0.92 1.70 ± 0.05 90.9 ± 27.9 62.6 ± 2.57 1.88 ± 0.10 573 ± 28 3.02 ± 1.4 90.9 ± 6.3

Blend 2 44.6 ± 0.91 1.65 ± 0.05 91.4 ± 5.86 68.5 ± 0.72 2.20 ± 0.03 580 ± 26 7.52 ± 1.8 88.9 ± 2.3

Blend 3 42.9 ± 0.34 1.73 ± 0.02 114 ± 5.65 66.2 ± 1.99 1.99 ± 0.07 449 ± 13 9.42 ± 0.4 84.5 ± 3.9

Blend 4 41.5 ± 0.24 2.21 ± 0.03 94.5 ± 3.75 67.7 ± 4.29 2.20 ± 0.10 458 ± 51 12.4 ± 1.1 88.6 ± 1.4

Blend 5 44.3 ± 0.17 2.37 ± 0.01 104 ± 0.53 74.3 ± 2.17 2.58 ± 0.1 367 ± 27 13.3 ± 1.2 86.7 ± 2.2

Blend 6 41 ± 0.41 1.73 ± 0.02 137 ± 6.28 68.4 ± 0.98 2.31 ± 0.01 151 ± 9.9 23.7 ± 2.0 82.8 ± 1.8

Blend 7 48.3 ± 0.49 1.82 ± 0.08 174 ± 4.66 73.7 ± 4.3 2.48 ± 0.12 360 ± 24 2.4 ± 0.1 73.7 ± 4.6

Blend 8 53.1 ± 2.26 1.84 ± 0.5 205 ± 25 84.92 ± 3.4 2.85 ± 0.09 167 ± 35 4.8 ± 0.2 58.8 ± 3.0

138

In the Tables the following abbreviations will be used: EG: Extrusion Grade, HIG: High Impact

Grade, TS: Tensile strength, TM: Tensile Modulus, EAT: Elongation At Break, FS: Flex strength, FM: Flex Modulus, IS: Impact Strength, MFI: melt flow index, and HDT: heat deflection temperature” or “heat distortion temperature.

Example 2. Preparation of acrylonitrile butadiene styrene (ABS)/polylactic acid (PLA) blend containing an acrylic copolymer-based lubricant and chain extender using pilot scale equipment

Two blends were prepared: example blend 2A (Blend 9) containing 72.9 wt.% ABS

(Styron-Magnum 1150EM), 25 wt.% PLA (PLA 3052D, Nature Works), 2 wt.% lubricant

(BIOSTRENGTH 900), and 0.1 wt.% chain extender (JONCRYL ADR (BASF)), and example blend 2B (Blend 10) containing 47.75 wt.% ABS (Styron-Magnum 1150EM), 50 wt.%

PLA (PLA 3052D, Nature Works), 2 wt.% BIOSTRENGTH 900, and 0.25 wt.% JONCRYL

ADR (BASF).

All formulations including blends and composites were extruded in a Leistritz twin screw compounder with 100 RPM screw speed and 180°C - 240°C temperature profile.

The material was molded using an Arburg Allrounder injection molder. The material was molded with 15 mm/sec injection time, and 20 second cooling time.

A comparison of the mechanical properties of neat ABS and the blend formed in

Example 2 is shown in Table 5.2:

139

Table 5.2 - Mechanical properties of ABS/PLA blend using pilot scale equipment

Example Tensile Tensile Elongation Flexural Flexural Impact Melt Flow

Strength Modulus at break Strength Modulus Strength Index

(MPa) (GPa) (%) (MPa) (GPa) (J/m) g/10min

ABS

Magnum 35.3 ± 2.87 1.90 ± 0.01 34.2 ± 3.9 57.7 ± 1.08 1.99 ± 0.02 507 ± 24 1.1 ± 0.2

Blend 9 41.0 ± 0.79 2.25 ± 0.03 43.4 ± 5.16 63.8 ± 0.62 2.10 ± 0.04 359 ± 64 9.64 ± 1.18

Blend 10 47.7 ± 0.55 2.62 ± 0.05 116 ± 27 72.0 ± 0.54 2.40 ± 0.04 433 ± 14 13.82 ± 1.0

Example 3. Preparation of acrylonitrile butadiene styrene (ABS)/polylactic acid (PLA) blend containing an acrylic copolymer based lubricant and chain extender with fillers

Four examples were prepared:

Example 3A (composite 1) was prepared containing 68.3 wt.% ABS (Styron-Magnum

1150EM), 24.7 wt.% PLA (PLA 3052D, Nature Works), 1.9 wt.% lubricant (BIOSTRENGTH

900), 0.1 wt.% chain extender (JONCRYL ADR (BASF)), and 5 wt.% BioChar.

Example 3B (composite 2) was prepared containing 75 wt.% ABS (Styron-Magnum

1150EM), 14.25 wt.% PLA (PLA 3052D, Nature Works), 4.75 wt.% BIOSTRENGTH 900, 1 wt.% JONCRYL ADR (BASF) and 5 wt.% Miscanthus fiber.

Example 3C (composite 3) was prepared containing 63 wt.% ABS (Styron-Magnum

1150EM), 12 wt.% PLA (PLA 3052d, Nature Works), 4 wt.% BIOSTRENGTH 900, 1 wt.%

JONCRYL ADR (BASF) and 20 wt.% Talc.

140

Example 3D (composite 4) was prepared containing 75 wt.% ABS (Styron-Magnum

1150EM), 14.25 wt.% PLA (PLA 3052D, Nature Works), 4.75 wt.% BIOSTRENGTH 900, 1 wt.% JONCRYL ADR (BASF) and 5 wt.% Coffee Chaff.

All composites were prepared in a DSM Micro 15 cc twin screw compounder, with a

220°C extrusion temperature, a 100 RPM screw speed, and a 2 minute retention time. The material was molded using a DSM micro 12 cc mini injection molder. Both injection and holding were set to 6 bar pressure, injection took 4 seconds, while holding took 8 seconds

A comparison of the mechanical properties of composites of ABS/PLA blends is shown in Table 5.3:

141

Table 5.3 - Mechanical properties of composites of ABS/PLA blends

Example Impact Strength

(J/m)

Example 3A 64.2 ± 2.9 (Composite 1)

Example 3B 79.7 ± 8.1 (Composite 2)

Example 3C 59.0 ± 4.8 (Composite 3)

Example 3D 135 ± 8.2 (Composite 4)

Example 4. Preparation of acrylonitrile butadiene styrene (ABS)/polylactic acid (PLA) blend processed at previous art conditions

Example blend 4A (Blend 11) was prepared using the same formulation as example blend 1B, with adjusted processing conditions. The method for making this blend involved changing the processing temperature to 206°C, to conform with processing conditions outlined in aforementioned US Patent Application No US/20120220711. Under these processing conditions, the toughness of the blend is severely decreased.

The mechanical properties of the blend formed in Example 4 is shown in Table 5.4:

142

Table 5.4 - Mechanical properties of ABS/PLA blend with altered processing conditions

Example Tensile Tensile Elongation Flexural Flexural Impact

Strength Modulus at Break Strength Modulus Strength

(MPa) (GPa) (%) (MPa) (GPa) (J/m)

Blend 11 41.6 ± 0.88 1.73 ± 0.3 15.1 ± 4.94 70.7 ± 0.94 2.33 ± 0.03 181±15.8

Example 5. Preparation of acrylonitrile butadiene styrene (ABS)/polylactic acid (PLA) blend processed with alternate additives

Example blend 5A (Blend 12) was prepared using the same formulation as blend 1B, with a substitute for Biostrength 900. Instead, 3 wt.% of Elvaloy PTW (DuPont) was used. The blend was made with identical processing conditions as blend 1B. However, the inclusion of this impact modifier showed decreased performance to blend 2.

The mechanical properties of the blend formed in Example 5 is shown in Table 5.5:

Table 5.5 - Mechanical properties of ABS/PLA blends processed with alternate additives

Example Tensile Tensile Elongation Flexural Flexural Impact

Strength Modulus at Break (%) Strength Modulus Strength

(MPa) (GPa) (MPa) (GPa) (J/m)

Blend 12 37.9 ± 0.65 1.72 ± 0.1 8.05 ± 2.03 62.1 ± 0.59 2.09 ± 0.02 60.9±1.42

143

Example 6. Morphology changes for high performance blends.

Example blend 6A and 6B were prepared to provide evidence to the changing morphological structure of materials with both a lubricant and chain extender. Blend 6A contains

75 wt.% ABS (Styron-Magnum 1150EM), 25 wt.% PLA (PLA 3052D, Nature Works); blend 6B contains 70 wt.% ABS (Styron-Magnum 1150EM), 25 wt.% PLA (PLA 3052D, Nature Works) and 5 wt.% lubricant (BIOSTRENGTH 900).

Figures 5.1 – 5.4 are high resolution images showing the structure of the morphology of the blends. Figures 5.1 and 5.2 are scanning electron microscopy (SEM) images of the impact fracture surface of blend 6A and blend 1B. Blend 1B shows a more refined structure than a blend without additives, with less separation between phases, which allows absorption of more energy during impact testing.

Figures 5.3 and 5.4 were taken on a Bruker Multimode atomic force microscope (AFM), and shows the deformation of the tip into surface, with a constant force. Figure 5.3 shows blend

6B structure, and Figure 5.4 shows blend 1B structure. One can see that without the chain extender and the lubricant together, the phases are much more defined in terms of separation, and do not show as much compatibility. Deformation plots show a better absorption of energy for blend 1B than with only the lubricant.

The blend in Example 1 exhibit a high Notched Izod Impact Strength compared to neat

ABS. Example 2 shows the scale up ability of the material, processed under pilot scale equipment. Although impact strength and heat deflection temperature are the properties of focus, all stated mechanical properties must be similar to unmodified ABS. For some embodiments, which cost is sufficiently low, lesser properties may be acceptable. Table 5.3 suggests that adding natural fiber/biofiller requires some optimization for better performance, and that smaller

144

fiber sizes produce better performing impact resistance in the material. Examples 4, 5, and 6 establish that processing conditions and precise ingredients are required for these formulations to create high performing materials.

Unless defined otherwise, all technical and scientific terms used herein have the same meanings as commonly understood by one of skill in the art to which the disclosed invention belongs.

Publications cited herein and the materials for which they are cited are specifically incorporated by reference.

Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the following claims.

5.7 Claims

We claim:

1. A high performance acrylonitrile butadiene styrene (ABS) polymer blend comprising

ABS, polylactic acid (PLA), an acrylic copolymer based lubricant and a polymeric chain extender.

2. The blend of claim 1, wherein the PLA concentration is less than 30 wt.% of the blend.

3. The blend of claim 1, wherein the PLA concentration is more than 30 wt.% of the blend.

4. The blend of claim 1, wherein the PLA comprises virgin PLA, recycled PLA, or combinations thereof.

5. The blend of claim 1, wherein the ABS is a high impact ABS.

6. The blend of claim 1, wherein the concentration of the ABS is from about 60 wt.% to about 85 wt.% of the blend.

145

7. The blend of claim 1, wherein the concentration of the acrylic copolymer based lubricant is from about 1 wt.% to about 6 wt.% of the blend.

8. The blend of claim 1, wherein the chain extender is an epoxy-functionalized styrene-acrylic polymer comprising glycidyl methacrylate (GMA).

9. The blend of claim 8, wherein the concentration of the epoxy-functionalized styrene-acrylic polymer is from about 0.1 wt.% to about 1.5 wt.% of the blend.

10. The blend of claim 1, wherein the blend exhibits a Notched Izod Impact Strength greater than

151 J/m when measured in accordance with ASTM D256.

11. The blend of claim 1, wherein the blend exhibits a Notched Izod Impact Strength greater than

414 J/m when measured in accordance with ASTM D256.

12. The blend of claim 1, wherein the blend exhibits both a tensile strength and flexural strength greater than the tensile strength and flexural strength of neat ABS when measured in accordance with ASTM D638 and D790, respectively.

13. The blend of claim 1, wherein the blend exhibits a tensile strength greater than 40 MPa when measured in accordance with ASTM D638.

14. The blend of claim 1, wherein the blend a flexural strength greater than 60 MPa when measured in accordance with ASTM D790.

15. An article comprising the blend of claim 1.

16. A high performance acrylonitrile butadiene styrene (ABS) polymer blend made by a process comprising:

(a) forming a mixture of ABS, polylactic acid (PLA), an acrylic copolymer based lubricant and a polymeric chain extender polymer; and

(b) extruding the mixture at a temperature of 230oC to 250oC to form the high performance ABS polymer blend.

Figure 5.1 – SEM of impact fracture surface of blend 6A

Figure 5.2 – SEM of impact fracture surface of Blend 1B

Figure 5.3 – AFM Deformation plot of blend 6B

Figure 5.4– AFM Deformation plot of blend 1B

5.8 Discussion

This patent work was done as there was considerable interest in the commercial viability of the blends. The constituent materials are cost effective, they can be processed on traditional equipment and the materials are commercialized and easily obtainable. Secondly, the mechanical performance of the blends is very high.

Example 1 shows the mechanical properties of several blends ranging in PLA content from 25 wt. % to 70 wt. % (Table 5.1). These blends were made on a laboratory scale microextruder and injection molder. This was done to increase the repeatability of the samples. As every individual test bar is a new blend, the repeated performance of the blend is tested through this method. This is a much better statistical demonstration of the resulting performance than if processed in one batch on a larger extruder and injection molded; only truly testing one blend. Additionally, the nature of the lab scale machine allowed the testing of many different formulations in a relatively small time frame, and to reduce the waste of material. In general, the method of making these blends is a proof-of-concept that these materials can be made with the performance attributes that are attractive for such a material.

Looking at the mechanical properties of the blends, there are several properties that only undergo relatively small changes through the range of PLA content, while others change drastically. The tensile strength and modulus undergo small changes, increasing with increasing PLA content.

This is expected as PLA is the stronger and stiffer of the two main components of the blend.

Similar remarks can be made for the flexural strength and modulus, although they undergo larger increases with the increased PLA. The most interesting is the fact that as these increase, one would expect that the toughness, as demonstrated by the elongation at break and impact strength, would be drastically reduced as PLA is very brittle. However, the elongation at break remains

very high through the whole range of PLA contents. In fact, the elongation at break remains around three times that of ABS, which is the higher between PLA and ABS. Blends 7 and 8 show that very high elongation at break values can be found even on blends with a majority of

PLA. Similarly, the impact strength does reduce generally as the PLA content is increased, but it remains very high even in the 50 wt. % and 70 wt. % blends. The variation in impact strength as the PLA is increased, such as the 50 wt. % blend having a higher impact strength than 40 wt. % can be explained by considering that these blends have not been optimized for the additive content and ratios, in addition to changes in processing to aid reaction. The MFI remains higher than neat ABS grades through the whole range of PLA contents, indicating an increase in the ease of processing for all blends over that of ABS. Finally, the HDT shows to remain fairly high

(above 80˚C) through blends containing up to 40 wt. % PLA.

Through preliminary experimentation, it was found that reducing additives from 5 wt. % and 1 wt. % down to 1-3 wt. % and 0-1 wt. % of Biostrength and Joncryl, respectively, resulted in better performance and less property variation. It was also found that as PLA increased, the requirement for additives also increased. This is likely due to the increased amount of COOH and OH end groups in the PLA end chains in addition to the increased surface area between the

ABS and PLA, making the requirement for in-situ compatibilizer higher. This indicates that not only does there exist an optimally efficient concentration of additives for each PLA loading, but also that deviations from this optimal (including higher amounts) results in a blend with reduced performance. ample 2 of the patent, the blends are produced on pilot scale equipment. The reason for this is to understand the ability of the blends to be scaled up to production level equipment. Additionally, because the conditions are different, including a higher shear rate, it is expected that the larger

scale pilot equipment will affect the blends mechanical properties. For this, two blends were prepared, one with 25 wt. % PLA, the other with 50 wt. % PLA. The mechanical properties of these are given in Table 5.2. It is obvious from these properties in general that scale up is indeed possible, and in fact looks to produce even higher results in some cases. Comparing the 25 wt. %

PLA loading at lab scale to the same loading at pilot scale, the strength values are lower for the pilot scale blend, while the modulus values are much higher. Interestingly, the impact strength is reduced in the pilot machine. It was expected that the higher shearing would aid in blending, creasing a better dispersion. Oyama et al. fournd that by increasing the screw speed in the extruder from 30 to 200 RPM, tougher blends of PLA/EGMA were made.1 This worked by creating a finer dispersion. In the case of the patent blends, it could simply be that the loading levels of additives were not optimal for these processing conditions. However, in reference to the

Wu ligament theory, when blended in the lab scale extruder, the ABS particles are slightly larger than when processed in pilot scale.2 As PLA is slow-crystallizing and highly amorphous, its main failure mode is through crazing. Therefore, there exists an ABS particle size that is optimum for toughening: large enough particles to nucleate crazes, while small enough to reduce ligament size of the PLA below critical to allow for a brittle to ductile transition to occur. When the increased shearing of the pilot scale extruder reduces the ABS particle size, it is no longer optimum, but below optimum.

Regarding the 50 wt. % PLA blends at lab scale compared to pilot scale, the strengths are similar, while the tensile modulus is drastically higher in the pilot scale blend. Interestingly, this increased stiffness in tensile loading does not translate to flexural loading. The most intesting change in properties between the lab and pilot scale blends at 50 wt. % PLA is the impact strength, which is 360 J/m at lab scale, but jumps to 433 J/m at pilot scale. Again, the same

explanation can be used. At higher PLA loading, the increased shearing of pilot scale equipment is needed to bring the ABS particle size closer to optimal.

Four composites were made with the ABS/PLA blends as the matrix in Example 3 (Table 5.3).

This work was done to show the ability of this material to be used as a matrix in composites – most notably natural fiber composites. The reinforcing natural fibers used were biochar, miscanthus fiber, talc, and coffee chaff. This is a proof of concept, thus, only impact strength was measured. As is expected with any rigid reinforcement agent, the impact strength was heavily reduced. However, the composite with coffee chaff retained an impact strength value in the range of neat ABS grades, showing the potential feasibility of these blends as a matrix in natural fiber composites.

Example 4 shows that the development of this blend is in part due to process engineering (Table

5.4). The blend was processed at the condition used by the previous art. The notable change here is the extrusion temperature is reduced to 206˚C. The resulting properties show very low toughness compared to the same formulation processed at the higher temperature (181 compared to 507 J/m).

Example 5 (Table 5.5) is a proof of concept that this compatiblization could not have been accomplished with alternate additives. This is just one of many blends that was attempted by the authors during preliminary experimentation. Here, the Elvalor PTW is used as a replacement for

Biostrength. Properties are extremely poor in general, showing that the specific additive used are needed.

Finally, Example 6 (Figures 5.1-5.4) shows the morphology evolution from blending. Further discussion is given in Chapter 4 of this work; Figures 5.1-5.4 show the refined structure, better dispersion, and smaller ligament sizes that allow for the high toughness properties of the blends.

1. Li, Y. and Shimizu, H., European Polymer Journal 45(3) 738-746 (2009)

2. Jo, M.Y., Ryu, Y.J., Ko, J.H. and Yoon, J.S.,, Journal of Applied Polymer Science 125 (S2) 231-238 (2012)

3. Sun, S., Zhang, M., Zhang, H. and Zhang, X., Journal of Applied Polymer Science 122(5) 2992-2999 (2011)

4. Chevali, V.S., Nerenz, B.A., Ulven, C.A. and Kandare, E., Polymer-Plastics Technology and Engineering 54 375-382 (2015)

5. Yeh, S.K., Agarwal, S. and Gupta, R.K., Composites Science and Technology 69 2225- 2230 (2009)

6. Oyama, H. T. Super-tough poly(lactic acid) materials: Reactive blending with ethylene

copolymer. Polymer 50, (2009).

7. Wu, S. A generalized criterion for rubber toughening: The critical matrix ligament

thickness. J. Appl. Polym. Sci. 35, 549–561 (1988).

Chapter 6: Statistical Optimization of Compatibilized Blends of Poly(lactic acid) and Acrylonitrile Butadiene Styrene

______

Abstract

A mixture design of experiment and subsequent regression analysis was used to study the effects of two additives on blends of poly(lactic acid) (PLA) and acrylonitrile butadiene styrene (ABS).

The statistical analysis was used to find a blend with a balance of high toughness, strength, and stiffness, which were measured through impact strength, elongation at break, tensile strength, and tensile stiffness. The blends were prepared by lab scale reactive extrusion and injection molding. Least-square regression models of statistically significant effects were built by analysis of variance (ANOVA). Using these models, optimization studies were used to study the predicted maximum values of each measurement criteria. Very large increases were seen in the measured responses with the addition of both additives at their optimized amounts. The impact strength was increased by over 600% from no additives to optimized amounts, while the elongation at break was increased by over 1000%, the tensile strength increased by 11%, and the tensile modulus increased by over 7%. Surprisingly, the composite optimization, which included all measured criteria, occurred at a point that allowed all four criteria values to remain very close to their individual maximums. The result is a partially biobased blend that does not sacrifice strength or stiffness to achieve very high toughness.

______

6.1 Introduction

Continued use of petroleum has raised many environmental concerns including the creation of greenhouse gases.1 Additionally, there are concerns over the depletion of such a highly used resource. It seems increasingly obvious that we must adopt sustainable practices to counter these effects. This is especially true in the manufacturing sector, where billions of pounds of petroleum are used to produce low cost polymer materials.2 The alternative to petroleum based polymers, certain biopolymers, have poor performance or very high cost. Thus, there is a need to develop low cost, partially biobased blends of petroleum and biobased polymers that have a balance of cost and performance. This is no easy feat, as blending thermodynamics generally does not favor blending two polymers, however, many blends still have some level of technical compatibility.

Poly(lactic acid) (PLA) is a biobased polymer that makes a promising candidate for blending with its high strength, stiffness, and low cost.3–7 However, there are several performance drawbacks that affect its use for durable goods. It has very poor toughness, which is reflected in its elongation and impact strength values. Additionally, the thermal stability of PLA, revealed in its heat deflection temperature (HDT), is very low for most applications. These performance attributes indicate that blending PLA with a tough polymer may bring the desired performance, while offsetting a portion of the petroleum use of a similar material. However, very few PLA blend systems have desirable performance without the presence of some additives to enhance adhesion and dispersion.8 The authors have developed such a blend system, which uses two additives in small amounts to efficiently and synergistically increase the toughness of high biobased content (>50%) systems.9 This system uses acrylonitrile butadiene styrene (ABS) as a tough blending companion, and a chain extender in conjunction with an acrylic copolymer to achieve better performance.

Originally, ABS was chosen as a blending partner to PLA because of target performance criteria and its ability to toughen other polymers.10–12 A high impact grade of ABS was chosen because it created a higher likelihood of producing a very tough blend. There have been several attempts at toughening PLA with ABS in the literature, with varying degrees of success. In all cases,

PLA/ABS blends without the use of additives are not miscible, and produce very poor performing blends. Thus, the aim of these works are to compatibilize the blends using blending chemistry, or through a third mutually compatible component. Li et al incorporated reactive styrene/acrylonitrile/glycidyl methacrylate copolymer (SAN-GMA) alongside ethyltriphenyl phosphonium bromide (ETPB) as a catalyst.8 They found a modest increase in film impact strength in their best blend. They found that the use of SAN-GMA and ETPB caused the ABS phase domain size to decrease, and for the size distribution to become narrower, creating a compatibilized blend. Sun et al. prepared ABS grafted GMA (ABS-GMA) particles to blend with

PLA, which produced very high toughness with high butadiene content.13 Jo et al. used SAN-

GMA at 20 wt. % loading along with a heat stabilizer to increase the toughness to 158 J/m. They concluded that the high rubber content ABS and the heat stabilizer enhanced the dispersion to increase the impact strength.14 Dong et al synthesized a reactive comb composed of polymethyl methacrylate with epoxy group dispersed throughout the structure.15 When used as a compatibilizer, the reactive comb polymer increased the interfacial adhesion between the ABS and PLA phases.

Statistical analysis is a very useful tool for investigating causal relationships to discern true effects of variables.16 This can be applied to our PLA/ABS blend system, in which the use of two additives confounds the true effects of each, especially considering interaction between them.

The effect of the ratios of acrylic copolymer and chain extender, which are effective at very low

concentrations, has not yet been studied extensively. In order to optimize the blend for a given criterion, the individual and synergistic effect of the additives must be studied. Zarrinbakhsh et al. used a full factorial design of experiments to find the effect of two additives in the composite blends, allowing the optimization, and subsequent testing, of better composites.17

To find the effect of the additives, we developed a statistical design of mixtures experiment using

Minitab® software. The PLA content was locked at 50%, while the ABS changed to facilitate the change in the additives. Through 13 formulation experiments, the tensile and impact properties were investigated to find the effects of the additives on elongation at break, impact strength, tensile strength, and tensile modulus. In this paper, we described the influence of each additive on the performance, their interaction, and found an optimal ratio for the responses chosen.

6.2 Experimental

6.2.1 Materials

The blend systems were made of four components. The PLA was injection molding grade

Ingeo® 3052D acquired from NatureWorks, LLC. High impact grade ABS was attained from

Styron under the grade name Magnum® 1150 EM. The acrylic copolymer has the tradename

Biostrength® 900 (Arkema, USA). Finally, the chain extender is Joncryl® ADR 4368 and was acquired from BASF. Since the objective of this work is to create a polymer blend with high biobased content (50% or greater), the amount of PLA in the blends was fixed to 50 wt. %. The acrylic copolymer amount ranged from 0 – 5 wt. % of the final weight of the blend. The chain extender ranged from 0 – 2 wt. % of the final weight of the blend. The content of ABS was changed to allow the additives to change, from 43 – 50 wt. %. The exact formulation contents of the experiments are given in Table 1 along with the mechanical properties of each formulation.

6.2.2 Material Processing

Before processing, materials were dried to an acceptable moisture level, less than 0.1 wt. %. All materials were blended and molded in a micro compounder and micro injection unit (DSM

Xplore®). The processing conditions used were kept constant for all formulations, and were:

240°C barrel temperature, 100 RPM screw speed, and 2 minute mixing time. The materials were weighed to 10g batches, according to the formulation, and all added simultaneously. Samples were injected from melt to a mold at 30°C over a 12 second period at 6 bars pressure into tensile and impact testing bars. The bars adhered to ASTM standards.

6.2.3 Material Characterization

The formulations were characterized via tensile and impact testing methods according to the

ASTM standards for testing materials. The tensile properties were measured on an Instron 3382 mechanical testing unit according to ASTM D638. The crosshead speed was set to 50 mm/min for all samples. The notched Izod impact strength was measured on a TMI Impact Tester, according to ASTM D256. Impact tests were completed with six samples, while tensile tests used five.

6.2.4 Statistical Design and Analysis

The intention of the experimental design was to examine the effects of each additive on the four measured responses: elongation at break, impact strength, tensile strength (at yield), and tensile modulus. The variable factors considered were the changing formulation constituents. Since PLA was held constant for all formulations (50%), its effect was not studied. The major effects investigated were generated through changes in the chain extender and the acrylic copolymer, while ABS was studied as it changed to allow changes in the additives quantities. The aim of this work is to develop a real-world durable polymer material with target properties similar to that of

ABS. Thus, the additive amounts were limited to keep the cost as low as possible. The acrylic copolymer was limited to the range of 0-5 wt. % of the formulation, while the chain extender was limited to 0-2 wt.%. These ranges were selected through preliminary experiments. 9

The experiment was a 3rd degree extreme vertices mixture experimental design, which allowed investigation of regression models up to a cubic in addition to the interaction between the additives. The schematic diagram of the model design is shown in Figure 6.1. This design also allowed for optimization of the mixture with respect to individual and groups of response variables to take place. To analyze the significance of the effects of the variables, Minitab® software was implemented. This analysis consisted of generating a regression model for each response, checking model adequacy, and optimize the variables for given responses and response groups.

1.5

0.5

Chain Extender Amount Chain Extender Amount (%) 0 1 2 3 4 5

Acrylic Copolymer Amount (%)

Figure 6.1 - Design of extreme vertices mixture experiment schematic

6.3 Results And Discussion

6.3.1 Mechanical Properties

The mechanical properties of the formulations are given in Table 6.1. The impact strength and elongation at break, both indications of the toughness of a material, were used as criteria to measure the effectiveness of the additives. Generally, however, as toughness increases, the strength and stiffness are likely to decrease, thus the tensile strength and tensile modulus were measured alongside impact and elongation. The strength and modulus values again show very little change performance with changes in the additive amounts.

The impact strength peaks at 340 J/m for the formulation with 2.5 wt. % acrylic copolymer and 1 wt. % chain extender. With only 3.5 wt. % additive (total), this is a dramatic increase from the blend without additives, at 34 J/m. Most blends fall between 200 and 300 J/m, with several below 200 J/m. This indicates that there is a sudden and significant drop in impact strength as the additive deviate from their maximum effective amounts. The elongation at break values follow a similar pattern, however, the effect is less dramatic. The highest value is 36.5 %, which is interestingly the same formulation as for the highest impact strength value, 2.5 wt. % acrylic copolymer and 1 wt. % chain extender. The increase in tensile strength and modulus is much less dramatic than that of impact strength and elongation. This is due to the fact that the impact strength and elongation at break are much more sensitive to the compatibility of the components.

With addition of acrylic copolymer and chain extender, the compatibility is increased, and as a result, the morphology is much more stable. Therefore, the transfer of stress is more efficient, allowing larger increases in the impact strength and elongation at break values, which are more sensitive to stress transfer than tensile strength and modulus. The highest value for tensile strength is 60.4 MPa, while the lowest is 54.3 MPa. Similarly, the highest modulus value is 2.81

GPa, while the lowest is 2.58 GPa. Interestingly, three of the four highest values are from the same formulation (2.5 wt. % acrylic copolymer and 1 wt. % chain extender), while all four lowest values are from the blend without additives. This indicates that an compatibility is roughly correlated to the performance values being tested, causing performance to be increased for multiple responses at a higher level of compatibility.

Table 6.1 - Mechanical properties of 13 formulations

Tensile Tensile Elongation at

Run Formulation Strength Modulus Break Impact Strength

GPa Mpa % J/m

1 50/46.5/2.5/1 60.4 ± 0.5 2.77 ± 0.04 36.5 ± 3.6 340 ± 13

2 50/43/5/2 60.0 ± 1.5 2.81 ± 0.03 28.4 ± 7.3 228 ± 13

3 50/48.25/1.25/0.5 59.5 ± 0.3 2.76 ± 0.03 28.4 ± 4.9 301 ± 20

4 50/45.75/3.75/0.5 59.6 ± 0.5 2.76 ± 0.05 27.2 ± 5.6 287 ± 16

5 50/50/0/0 54.3 ± 0.9 2.58 ± 0.06 3.1 ± 0.5 34.2 ± 0.9

6 50/49/0/1 59.6 ± 0.7 2.74 ± 0.01 29.8 ± 11.0 159 ± 3

7 50/48/0/2 59.5 ± 0.5 2.75 ± 0.02 29.0 ± 4.84 136 ± 7

8 50/45/5/1 56.2 ± 0.4 2.71 ± 0.01 13.4 ± 4.0 190 ± 5

9 50/47.5/2.5/0 55.5 ± 0.8 2.66 ± 0.04 8.47 ± 4.15 123 ± 3

10 50/47.25/1.25/1.5 60.1 ± 0.4 2.78 ± 0.02 26.6 ± 6.2 259 ± 14

11 50/45.5/2.5/2 58.7 ± 0.8 2.74 ± 0.05 27.9 ± 5.2 272 ± 18

12 50/44/5/1 58.8 ± 0.5 2.74 ± 0.02 33.7 ± 16.1 248 ± 12

13 50/44.75/3.75/1.5 59.6 ± 0.6 2.76 ± 0.03 29.4 ± 11.4 259 ± 14

The effect of acrylic copolymer seems to depend on the amount of chain extender, indicating an

interaction between the two additives. This is supported through the model. By itself, increasing

the acrylic copolymer increases the impact strength and the elongation in a somewhat linear

manner. However, as the chain extender is increased, the acrylic copolymer seems to hit a max

effect in its mid-range, with the best impact strength and elongation occurring at 2.5 wt. % loading of acrylic copolymer. For the tensile strength and modulus, the effect of the acylic copolymer appears the same as for the impact strength and elongation, except at high loadings, in which increases in acrylic copolymer causes increases in the performance.

6.3.2 Model Development

Analysis of variance (ANOVA) is a powerful method of determining the significance of a variable, in addition to investigating the interactions between more than one factor. In this study a level three model allowed the investigation of models up to cubic with interaction terms. The regression equation follows the general format:

푅 = 푎0 + 푎1퐴퐵푆 + 푎2퐴퐶 + 푎3퐶퐸 + 휀

(1)

Where ABS, AC, and CE represent the weight amounts of ABS, acrylic copolymer, and chain extender respectively, while 푎푖 are the regression coefficient terms that are determined by least square regression, and ε is the error term of the model.

The ANOVA results of the full regression are given in Table 6.2. For each model tested, the P value is also given. In this study, the chosen significance level is the typical 5%, which indicates a significance threshold of 0.05 for P values. Table 6.2 shows the ANOVA results for the models tested. The special cubic model is not significant for impact strength. For elongation, the quadratic ABS-acrylic copolymer and the special cubic model are not significant. Both tensile strength and modulus regressions include all model terms as significant.

6.3.3 Model Adequacy

The regression models developed need to be tested for the validity of the assumptions upon which they are based. There are three aspects to test, (1) the validity of the assumption of normality of the data, (2) lack-of-fit tests, and (3) the proportion of variance in the data that is explained by the model, measured through the R2 statistic. The validity of the assumption of normality can be confirmed through investigation of the normal probability plots. Figure 6.2 shows that the residuals plotted follow a straight line. In addition, the data fairly closely adheres to the normal line, with very little deviation. More importance is placed near the middle of the plot. Small deviations from the line near the plot limits are of little importance.

The lack-of-fit test can be used to check when too much error in prediction is due to lack of model fit to the data. Lack-of-fit may mean may indicate that higher order models can better describe the data. It is recommended that the lack-of-fit statistic and the R2 statistic be studied simultaneously. Table 6.2 shows the models and the regression statistics associated with each.

2 2 Looking at the R and the R adj statistic, the models seem to fit quite well. In addition, the PRESS values are a measure of the model’s ability to predict values generated from an entirely new experiment. These values indicate that the models are able to describe the data generated reasonably well.

165

Figure 6.2(A) - Residual plots for Impact Strength

166

Figure 6.2(B) - Residual plots for Elongation at Break

167

Figure 6.2(C) - Residual plots for Tensile Strength

168

Figure 6.2(D) - Residual plots for Tensile Modulus

6.3.4 Fitted Models

6.3.4.1 Impact Strength

The impact regression model is given in Table 6.3, the Cox trace plot is Figure 6.3. The model implies that all interactions are significant factors affecting impact strength between the changing components of the blend. In this model, it is difficult to make conclusions on the direct influence of each component, since the data is best described with interaction terms, which confound the meaning of the coefficients. In this case however, the magnitude of the coefficients gives a

169

glimpse of the impact of the variable on the response variable. Acrylic copolymer and chain extender can both change the impact strength dramatically with very little changes in their component amounts. In other words, the impact strength is extremely sensitive to small changes in the additive amounts: chain extender by almost an order of magnitude more than acrylic copolymer. The sensitivity of the chain extender is due to the high reactivity of the additive.

Having several functional epoxy groups in its structure makes it very efficient at very low loading levels. This is intuitive considering the amounts at which these additives are most affective, below 3 wt. %. Thus, at very small loadings, these additives have great influence on the impact strength of the blend. The fact that they both have negative coefficients implies that as they near zero, there is a very sudden sharp drop in the impact strength. This is indicative of a very sharp optimization point. The impact strength is very high at the optimized value of the amount of additive, but suddenly and dramatically reduces as the loading varies away from the optimization point. The interaction terms make up for the negative nature of the individual term coefficients, creating a response curve that has a relatively sharp maximum at the optimized amount of chain extender and acrylic copolymer. The interaction term that has the highest effect on the impact strength is the ABS*Chain extender which is surprising, since small changes in

ABS are not expected to elicit large changes in the impact strength. However, it does show that the effect of chain extender is increased due to a higher portion of ABS present in the blend. The foundation of ABS is where the increase of impact strength is expected.9

170

6.3.4.2 Elongation at Break

The elongation at break regression model is given in Table 6.3, the Cox trace plot is Figure 6.3.

Again, all terms are significant factors affecting the elongation at break. As with the impact strength, the elongation at break has large negative effects of chain extender and acrylic copolymer, with chain extender having approximately 1 order of magnitude higher effect on the elongation at break than acrylic copolymer. At relatively small contents of acrylic copolymer and chain extender, there are large effects on the elongation at break. The optimal point for both comes at relatively low amounts of each. Additionally, the optimal point value for elongation at break is much higher than the zero-additive amount. Thus, the terms need to have large coefficients for the model to accurately exhibit such a large increase in the value of the elongation at break with relatively small changes in the component amounts of chain extender and acrylic copolymer. The interaction term that has the largest effect on the elongation at break is acrylic copolymer chain extender interaction. Both of these additives work to increase the elongation at break, however in different ways. The chain extender increases the molecular weight of the PLA, while the acrylic copolymer increases the chain mobility. When both of these mechanisms work together, the result is a vastly improved elongation at break value where the chains are able to disperse much more of the energy in a tensile test before failure. 9

6.3.4.3 Tensile Strength

The tensile strength regression model is given in Table 6.3, the Cox trace plot is Figure 6.3. The tensile strength is much less affected by changes in the component amounts of the additives.

However, all factors considered are statistically significant variables in the outcome of tensile

171

strength. It appears that chain extender has the highest effect on the tensile strength of the blend, which is an expected as the chain extender increases the molecular weight of the PLA and therefore increases the strength with longer chains.18 With longer chains, there is more bonding able to occur between two given chains, making them relatively more stationary.19 This, in turn, increases the strength of the polymer. The interaction that plays the largest factor on the tensile strength is the bio*chain extender interaction. As seen with other response variables, the interaction between these two additives has a large effect on the tensile strength of the blend. It seems that including only the acrylic copolymer, there is much more subtle increase. When the chain extender is added to the blend alone, there is a large and significant jump in the tensile strength. However, the chain extender can only increase the tensile strength up to a point, roughly 59.5 MPa. To achieve a tensile strength beyond this, the acrylic copolymer is needed to provide a subtle increase.

Tensile Modulus

The tensile modulus regression model is given in Table 6.3, the Cox trace plot is Figure 6.3.

Again, each of the additives and interaction terms create a similar effect on the outcome of the model to previous iterations of the models for the other response variables. The fact that the models for the tensile modulus and tensile strength are very similar is a somewhat expected result. This implies that the additives work in the same way to increase strength and stiffness.

The chain extender seems to have the largest effect on the tensile modulus, which has been true for all response variables. The chain extender works to increase the molecular weight. With higher molecular weight, and therefore more chain entanglement, there is an expected increase in

172

the force required to move the chains relative to each other, and break the secondary bonds holding them in relative position. The interaction term that carries the most weight in determining the change in modulus due to the additives is again the Bio*chain extender interaction. In the same way as the tensile strength, the chain extender and the acrylic copolymer are both able to increase the tensile modulus alone. However, when used in tandem, the overall effect is one of greater stiffness than if each were used alone.

173

Analysis of Variance for Impact Strength Analysis of Variance for Elongation at Break Source DF Seq SS Adj SS Adj MS F P DF Seq SS Adj SS Adj MS F P Regression 6 351412 351412 58568.7 37.08 0.000 6 5791.82 5791.82 965.3 18.12 0.000 Linear 2 132417 98024 49012.0 31.03 0.000 2 3054.57 2401.98 1201.0 22.54 0.000 Quadratic 3 218821 218128 72709.3 46.04 0.000 3 2697.59 2653.24 884.4 16.60 0.000 ABSxAC 1 107706 30197 30197.0 19.12 0.000 1 94.07 9.83 9.8 0.18 0.669 ABSxCE 1 367 85192 85192.0 53.94 0.000 1 540.98 2431.86 2431.9 45.65 0.000 ACxCE 1 110748 624 624.0 0.40 0.532 1 2062.54 131.13 131.1 2.46 0.123 Special Cubic 1 174 174 174.0 0.11 0.741 1 39.66 39.66 39.7 0.74 0.392 ABSxACxCE 1 174 174 174.0 0.11 0.741 1 39.66 39.66 39.7 0.74 0.392 Residual Error 54 85286 85286 1579.4 54 2876.78 2876.78 53.3 Lack-of-Fit 6 79128 79128 13188.0 102.81 0.000 6 625.54 625.54 104.3 2.22 0.057 Pure Error 48 6157 6157 128.3 48 2251.24 2251.24 46.9 Total 60 436697 60 8668.6 Analysis of Variance for Tensile Strength Analysis of Variance for Tensile Modulus Source DF Seq SS Adj SS Adj MS F P DF Seq SS Adj SS Adj MS F P Regression 6 203.998 203.998 34.000 48.27 0.000 6 0.17394 0.17394 0.028990 17.67 0.000 Linear 2 107.172 86.038 43.019 61.08 0.000 2 0.10471 0.05749 0.028745 17.53 0.000 Quadratic 3 87.387 94.503 31.501 44.73 0.000 3 0.0562 0.06188 0.020627 12.58 0.000 ABSxAC 1 10.14 9.759 9.759 13.86 0.000 1 0.01039 0.01204 0.012040 7.34 0.009 ABSxCE 1 11.058 86.368 86.368 122.63 0.000 1 0.01382 0.0555 0.055500 33.84 0.000 AcxCE 1 66.189 15.861 15.861 22.52 0.532 1 0.03199 0.01792 0.017920 10.93 0.002 Special Cubic 1 9.439 9.439 9.439 13.40 0.001 1 0.01303 0.01303 0.013030 7.94 0.007 ABSxACxCE 1 9.439 9.439 9.439 13.40 0.001 1 0.01303 0.01303 0.013030 7.94 0.007 Residual Error 54 38.033 38.033 0.704 54 0.08857 0.08857 0.001640 Lack-of-Fit 6 16.725 16.725 2.788 3.96 6 0.02231 0.02231 0.003718 2.69 0.025 Pure Error 48 21.309 21.309 0.444 48 0.06626 0.06626 0.001380 Total 60 242.031 60 0.26251 Table 6.2 - ANOVA results for models of impact strength, elongation at break, tensile strength, and tensile modulus

174

Table 6.3 - Regression models and statistics values

Response Regression Model R2 Statistic

2 2 Impact Strength (J/m) IS = 0.801659ABS - 546.310(AC) - R = 80.47%; R adj =

4487.26(CE) + 12.7239(ABS)(AC) + 78.30%;

94.5241(ABS)(CE) + 67.2135(AC)(CE) PRESS=105275;

2 +0.763055(ABS)(AC)(CE) R pred = 75.89%

2 2 Elongation at Break (%) EL = 0.0669447ABS - 8.04225(AC) - R = 66.81%; R adj =

754.083(CE) + 0.229553(ABS)(AC) + 63.13%;

15.9702(ABS)(CE) + 30.8213(AC)(CE) - PRESS=3541.25;

2 0.364639(ABS)(AC)(CE) R pred = 59.15%

2 2 Tensile Strength (MPa) TS = 1.08621ABS - 8.88907(AC) - R = 84.29%; R adj =

140.790(CE) + 0.228740(ABS)(AC) + 82.54%;

3.00968(ABS)(CE) + 10.7195(AC)(CE) - PRESS=52.0236;

2 0.177879(ABS)(AC)(CE) R pred = 78.51%

2 2 Tensile Modulus (GPa) TM = 0.0517152ABS - 0.289033(AC) - R = 66.26%; R adj =

3.52337(CE) + 0.00803409(ABS)(AC) + 62.51%;

0.0762904(ABS)(CE) + 0.360305(AC)(CE) PRESS=0.114981;

2 - 0.00660958(ABS)(AC)(CE) R pred = 56.20%

175

6.3.5 Optimization

The Minitab 17® software was used to generate optimization plots for the component amount in relation to the response variables. Each response was optimized individually, and a fifth optimization was run for all together. The contour plots for each response are given in Figure

6.4. The impact strength was optimized using the model generated in the software. The optimization output of the model shows a maximum with the range of component amounts for each additive. As the models implied, there were fairly sharp peaks for the optimization area of the components. The optimal component amounts to create an impact strength maximum are

3.16 wt. % acrylic copolymer and 1.19 wt. % chain extender. The acrylic copolymer can change the impact strength from under 190 J/m to over 322 J/m at the maximum point. This is a fairly large change for under 3 wt. % change in the additive. Likewise the chain extender can change the maximum from 194 J/m to 322 J/m just in the range from 0 wt. % to 1.19 wt. %. This has described the individual impacts of the additives on the impact strength. However, with this model, with both additives set to zero, the impact strength drops to 41 J/m. These values agree fairly well with expected values. Thus, according to the model, the additives can change the impact strength from 41 J/m to 322 J/m with only inclusion of 4.35 wt. % additives. The optimized impact strength value of 322 J/m is above expected values for ABS/PLA blends with

50 wt % PLA.

176

Figure 6.3(A) - Cox response trace plot of impact strength

177

Figure 6.3(B) - Cox response trace plot of elongation at break

178

Figure 6.3(C) - Cox response trace plot of tensile strength

179

Figure 6.3(D) - Cox response trace plot of tensile modulus

The elongation at break was also optimized via the statistical model provided in Minitab ®.

Unlike the optimization of the impact strength, which showed a maximum with each additive range, the maximum for the acrylic copolymer occurs at its max range, implying that the elongation at break could be optimized further by increasing the acrylic copolymer past 5 wt. %.

However, the range was chosen to create an economically viable and industrially applicable blend. Therefore, above 5 wt. % is considered too high for this component amount. It also implies that according to the model, the acrylic copolymer has a broad maximum, with increasing positive effect throughout a wide range of amounts for the component. The optimized point creates an elongation at break value of 37 %. This occurs at 5 wt. % acrylic copolymer and

180

1.26 wt. % chain extender. Again, this is higher than expected for an ABS/PLA blend with 50 wt. % PLA. In fact, this value is higher than the elongation at break value of the neat ABS. The value generated with both additives set to 0 wt. % in the model is only 3.3%, so the additives together create a ten-fold increase in the elongation at break in the blend. The elongation at break value when the acrylic copolymer is set to zero is 34%, meaning that the chain extender alone can increase the elongation at break value of the blend from 3.3% to 34%, which is over 90% of the increase up to the optimized point. This indicates that the chain extender is the cause of a large majority of the increase in elongation at break. With the chain extender set to zero, the acrylic copolymer increases the elongation at break up to 14%. Thus with the addition of both the additives, the acrylic copolymer simply boosts the elongation at break from 34% to the optimal 37%. The broad peak in the acrylic copolymer graph suggests that the addition of this additive increases the elongation at break over a fairly large component amount range. This is an expected result as the acrylic copolymer acts to increase the molecular mobility of the polymer chains, thus the elongation should see increases over a wide range of acrylic copolymer amount values.20 In contrast, the chain extender, which works to increase the molecular weight of the polymer chains, sees a sharp peak after which there is a drop off in the elongation at break.

Again, this is expected as increases in the intermolecular bonding will increase the ability of the polymer to absorb tensile loading at the speed tested.19 However, at a certain point, too much bonding, or chain entanglement increases the stiffness of the polymer, increasing the force in the polymer at a given deformation level. This will act to decrease the elongation at break value above this point.19

181

Figure 6.4(A) - Optimization plot of impact strength

182

Figure 6.4(B) - Optimization plot of elongation at break

183

Figure 6.4(C) - Optimization plot of tensile strength

184

Figure 6.4(D) - Optimization plot of tensile modulus

185

Figure 6.4(E) - Composite optimization plot of all responses

The optimization of tensile strength shows similar characteristics to that of the elongation at break. The optimized tensile strength is 60 MPa, which occurs at 0 wt. % acrylic copolymer and

1.43 wt. % chain extender. The acrylic copolymer does not have a local maximum in the range of amounts chosen for these experiments. Instead, there is a decrease in the tensile strength through the entire range of amounts for the acrylic copolymer. Near the low end of the range of the acrylic copolymer, there is a fairly linear decrease in the tensile strength. However, as the amount of acrylic copolymer increases past approximately 2.5 wt. %, the decrease becomes exponential. The amount of chain extender at the optimal point is 1.43 wt. %, which, when

186

compared to the optimal amount for elongation at break and impact strength, is fairly similar, suggesting that the point at which the chain extender is most effective increases all responses similarly. The predicted tensile strength with both additives set to zero wt. % is 54 MPa. The increase in tensile strength from this point to the optimal point is only approximately 11%. Thus, the additives have less effect on the tensile strength than the elongation at break or impact strength.

Figure 6.5(A) - Mixture contour plot of impact strength

187

Figure 6.5(B) - Mixture contour plot of elongation at break

188

Figure 6.5(C) - Mixture contour plot of tensile strength

189

Figure 6.5(D) - Mixture contour plot of tensile modulus

190

Figure 6.6(A) - Overlaid contour plot all four responses

191

Figure 6.6(A) - Overlaid contour plot all four responses

The final response to optimize is the tensile modulus, which is optimized when both additives are at their highest value in the range, 5 wt. % acrylic copolymer and 2 wt. % chain extender. This produces a tensile modulus of 2.78 GPa. At its lowest, with both additives set to zero wt. %, the tensile modulus is 2.58 GPa. This represents an increase of approximately 8%. Since the lowest tensile modulus value of the blend is higher than that of ABS, all the values are acceptable and the tensile modulus is the least critical response variable of the four tested.

Taking into account all four responses simultaneously, the optimized response can be generated.

The overlaid contour plot along with the selected parameters is given in Figure 6.5. The optimization was run with both the impact strength and the elongation at break importance and weight factors set to 1.0, while the tensile strength and the tensile modulus importance and

192

weight factors set to 0.5. This is due to the fact that the strength and modulus values are above the target values in throughout a wide range of additive amounts, and thus are acceptable for most formulations. However, the impact strength and elongation have fairly narrow localized maximums associated with the additive amounts, which drop below target values within the experimental range of the amounts of the additives. The optimized formulation for all four responses calls for acrylic copolymer content of 3.59 wt. % and a chain extender amount of 1.21 wt. %. According to the model, this will yield a blend that has an impact strength of 320 J/m, which is very close to value that was optimized for impact strength alone, only 0.6 % from the maximum value. The elongation at break is also very high, 35%, which is only 5% difference from its maximum value. The tensile value in this optimization is unchanged (60 MPa) from the maximum value according to the model. Finally, the tensile modulus value is also unchanged from its maximum value, at 2.78 GPa. Looking at the composite desirability, it is clear that there are sharp maximum peaks in the additive amounts that result in a very highly performing blend.

This is a very well balanced result, and is an indication that the additives work on the molecular scale to compatibilize the PLA and ABS together, stabilizing the morphology. This was demonstrated in previous work.9

The outcome of this work is best demonstrated by comparing the properties of the optimized blend with that of neat ABS from Table 6.1. In the tensile strength, tensile modulus, and elongation at break, the blend results are drastically higher than neat ABS. The material is much stronger and more ductile than ABS. Even considering the impact strength, of which the ABS is

193

a high impact strength grade, the blend is approaching the value, and certainly is inside the range found for other grades of ABS.

6.4 Conclusions

A mixture design of experiments with 13 experiment formulations of a blend was prepared and the mechanical properties were tested. The statistical analysis allowed models to be built that describe the impact strength, elongation at break, tensile strength, and tensile modulus. These models were used to analyze the influence of the additive amounts on each of the four response variables. It was shown that both the acrylic copolymer and the chain extender have local maximums for impact strength inside their amount ranges. The local maximums, especially for the chain extender, are quite narrow, and the impact strength can decrease drastically as the amount values begin to deviate from the optimized point. The elongation at break has a local maximum for the chain extender, but the acrylic copolymer increases the elongation through the entire range of amounts. However, this is not a steep increase, instead is part of a larger broad peak that continues beyond the amounts tested. The acrylic copolymer has a very broad peak for the tensile strength, indicating that it does not have as great an influence as the chain extender.

Finally, the tensile modulus shows similar result to that of elongation, a very broad peak with increasing values as the acrylic copolymer increases, and a narrower peak for the chain extender.

Finally, the optimized formulation was calculated taking into account all four responses. When compared to the blend without additives, there are vast improvements in all responses. The impact strength was increased by over 600%, the elongation at break was increased by over

1000%, the tensile strength increased by 11%, while the tensile modulus increased by over 7%.

194

The result, is a partially biobased polymer with high toughness without sacrificing strength or stiffness. In fact, the impact strength and elongation at break are comparable to commercially available ABS grades, but the tensile strength and modulus are much higher.

6.5 Acknowledgements

The financial support from the Ontario Ministry of Agriculture and Food Rural Affairs

(OMAFRA)/University of Guelph - Bioeconomy for Industrial Uses Research Program (Project

#200245); the Natural Sciences and Engineering Research Council (NSERC, Canada Discovery grants (Project #400322) and NSERC- AUTO21 NCE (Project #400372 & 400373); and Ontario

Research Fund, Research Excellence Program; Round-4 (ORF-RE04) from the Ontario Ministry of Economic Development and Innovation (MEDI) (Project #050289) to carry out this research is gratefully acknowledged.

6.6 References

1. Balat, M. & Balat, H. Recent trends in global production and utilization of bio-ethanol fuel. Appl. Energy 86, 2273–2282 (2009).

2. Global Plastic Production Rises, Recycling Lags | Worldwatch Institute.

3. Auras, R. A., Lim, L. T., Selke, S. E. & Tsuji, H. Poly(lactic acid): Synthesis, Structures,

Properties, Processing, and Applications. (John Wiley & Sons, 2011).

4. Vadori, R., Mohanty, A. K. & Misra, M. The Effect of Mold Temperature on the

Performance of Injection Molded Poly(lactic acid)-Based Bioplastic. Macromol. Mater. Eng.

298, 981–990 (2013).

195

5. Garlotta, D. A Literature Review of Poly(Lactic Acid). J. Polym. Environ. 9, 63–84

6. Liu, H., Chen, F., Liu, B., Estep, G. & Zhang, J. Super Toughened Poly(lactic acid)

Ternary Blends by Simultaneous Dynamic Vulcanization and Interfacial Compatibilization.

Macromolecules 43, 6058–6066 (2010).

7. Liu, H. & Zhang, J. Research progress in toughening modification of poly(lactic acid). J.

Polym. Sci. Part B Polym. Phys. 49, 1051–1083 (2011).

8. Li, Y. & Shimizu, H. Improvement in toughness of poly(l-lactide) (PLLA) through reactive blending with acrylonitrile–butadiene–styrene copolymer (ABS): Morphology and properties. Eur. Polym. J. 45, 738–746 (2009).

9. Vadori, Ryan, Manjusri Misra, and Amar K. Mohanty. Sustainable biobased blends from the reactive extrusion of polylactide and acrylonitrile butadiene styrene. Journal of Applied

Polymer Science (2016).

10. Cook, Wayne D., et al. "Morphology–property relationships in ABS/PET blends. I.

Compositional effects." Journal of applied polymer science 62.10 (1996): 1699-1708.

11. Kudva, R. A., H. Keskkula, and D. R. Paul. "Properties of compatibilized nylon 6/ABS blends: Part I. Effect of ABS type." Polymer 41.1 (2000): 225-237.

12. Mantovani, G. L., Bresciani Canto, L., Hage Junior, E. and Pessan, L. A. Toughening of

PBT by ABS, SBS and HIPS systems and the effects of reactive functionalised copolymers.

Macromolecular Symposia. (2001) Vol. 176. No. 1.

196

13. Sun, S., Zhang, M., Zhang, H. & Zhang, X. Polylactide toughening with epoxy- functionalized grafted acrylonitrile-butadiene-styrene particles. J. Appl. Polym. Sci. 122, 2992–

2999 (2011).

14. Jo, M. Y., Ryu, Y. J., Ko, J. H. & Yoon, J.-S. Effects of compatibilizers on the mechanical properties of ABS/PLA composites. J. Appl. Polym. Sci. 125, E231–E238 (2012).

15. Dong, W. et al. PLLA/ABS Blends Compatibilized by Reactive Comb Polymers: Double

T g Depression and Significantly Improved Toughness. ACS Sustain. Chem. Eng. 3, 2542–2550

(2015).

16. Montgomery, D. C. Design and Analysis of Experiments. (John Wiley & Sons, 2008).

17. Zarrinbakhsh, N., Defersha, F. M., Mohanty, A. K. & Misra, M. A statistical approach to engineer a biocomposite formulation from biofuel coproduct with balanced properties. J. Appl.

Polym. Sci. 131, n/a–n/a (2014).18. Fornes, T. D., Yoon, P. J., Keskkula, H. and Paul, D. R.

Polymer. 42, 09929–09940 2001.

19. Gedde, U. Polymer Physics. (Springer Science & Business Media, 1995).

20. Zarrinbakhsh, N., Misra, M. & Mohanty, A. K. Biodegradable Green Composites from

Distiller’s Dried Grains with Solubles (DDGS) and a Polyhydroxy(butyrate-co-valerate)

(PHBV)-Based Bioplastic. Macromol. Mater. Eng. 296, 1035–1045 (2011).

197

Chapter 7: Overall Conclusions and Future Work

______

Abstract

This chapter begins with a preliminary economic analysis of the material. Key outcomes are briefly presented, along with recommendations for further study.

______

198

7.1 Preliminary Economic Analysis of BioABS Blends

As the developed BioABS material is meant to serve as a sustainable alternative to traditional

ABS, it is intended for use in the durable goods sector. The uses include, but are not limited to, the automotive and electronics industries. The largest use of ABS by sector is household appliances, at 38% of ABS production, while electronics and automotive sectors account for

24% and 14% of ABS production; respectively.1 According to market reports, there is an expected growth of automotive plastics by 8.5% from 2011 to 2016. This would put production of automotive plastics at over 10 million tons in 2016. The highest manufactured plastic for automotive uses is polypropylene for its extremely low cost (36% of automotive plastics).

However, ABS is the third most used plastic at 12%, meaning that over 1 million tons of ABS plastic is predicted for use in the automotive industry in 20162. Additionally, the outlook for

ABS production remains positive, as production of ABS has grown over the last ten years.

As stated throughout this document, the goal of this work was to create a commercially viable material. This was kept in mind from the outset of work, and methods were chosen to complement this goal. Obviously, using the inherent properties of the polymers instead of relying heavily on modification through additives provides a superior method toward development of a satisfactory material. Thus, the additives employed were already commercial products, and used in relatively low quantities. Additionally, production methods are a vitally important piece of polymer materials. A heavy majority of durable polymer goods use the same processing techniques, extrusion and injection molding. These two techniques are at the center of polymer

199

production, as they provide a method of compounding and molding the materials at an astoundingly high speed and low cost. Thus, these methods were the lone methods employed in processing the developed blends. With the material now developed, and the processing methods set, the impact of production and inputs must now be assessed to determine commercial viability.

To assess commercial viability, several factors are considered. The production process for traditional ABS materials in comparison to the biobased blend is a crucial factor. The inputs for both materials are also important. Pricing of inputs is needed. Finally, the break-even point for each material can be used to find if BioABS is financially feasible. Ma et al. carried out a financial cost comparison on a very similar material, also referred to as BioABS.3 In this material, the inputs are ABS, PLA, and an engineered soy hull, creating a composite material. As

ABS and PLA are the major inputs in both BioABS, it is worth the comparison between each.

Ma et al. used a technical cost modeling method to carry out the comparison between their version of BioABS and traditional ABS.3 Using this activity based approach, they first identified the production process for traditional ABS versus BioABS. They found that BioABS is only roughly 4% more costly than ABS. The authors surmised that with a small market premium on biobased materials, BioABS can compete with traditional ABS. Through the analysis, they found that the raw materials accounted for the major portion of cost, more than 96%. As the raw material cost is a significant factor in the production cost, they found that when the price of PLA is reduced by 5.69%, the cost of ABS and BioABS are the same. However, as the raw materials

200

of ABS are petroleum based, increases in styrene, propylene, and C4 can cause the cost of ABS to rise at an increased rate compared to BioABS.

7.2 Key Outcomes

The aim of this project was to find a method of improving PLA/ABS blends in a commercially viable manner. Thus, as mentioned throughout this treatise, there were several limitations inherent in the work. The use of commercial additives must be cost effective, and thus either kept to very low quantities or inexpensive on the scale of the constituent polymers themselves.

Additionally, as traditional polymer processing methods are extremely efficient and refined, these are the methods to be kept. Thus, the processing methods were limited to twin-screw extrusion and injection molding.

In the first part of the work, the effects of processing were investigated on both materials.

Interestingly, it was found that ABS can undergo a thermo-oxidative degradation during processing at higher temperatures.

In the second of the work, potential compatibilizer agents were found and used in a plethora of preliminary experiments. These additives were used both standalone and in conjunction with each other at low contents (below about 7 wt. %) until a suitable formulation was found to effectively increase the performance of the blend. These additives were found to be an acrylic copolymer and a chain extender.

201

The effects of these additives were investigated in the third part of the work, including the mechanism by which they were able to increase the performance of the blends. This was done through a series of polymer characterization techniques.

Finally, an optimal formulation was found through statistical analysis. In this work, the PLA content was held constant, while various formulations were made.

7.3 Overall Conclusions of The Thesis

The conclusion of this work represents a deeper understanding of PLA, ABS, and their blends.

The knowledge collected through the detailed investigations can be used in several related areas.

The ability to highly toughen a brittle polymer by blending with a complementary tough polymer is described and outlined in detail in this work. Additionally, this work serves to understand the development of partially biobased blends and to study their ability to be used as an alternative to petroleum based polymers. It is the hope of the authors that this serves as a fundamental work in the future development of sustainable polymers.

Chapter 1 presented the background information, the problem statement, and the objectives of the work. This information led to a further investigation of previous studies, which was presented in Chapter 2. Through this literature research, a definitive knowledge gap was identified, which is a commercially viable compatiblized blend of PLA/ABS. This research also pinpointed methods which represented the best approach to achieving a high performing polymer blend, namely certain functional groups and toughening mechanisms.

202

Chapter 3, which was an original research article, investigated the effect of processing conditions on the performance of ABS. The knowledge acquired through this study allowed better choices of processing conditions during the blending studies of Chapter 4, 5, and 6. Chapter 4 demonstrated the better compatibility that can be found through the efficient use of certain additives. The mechanism of compatibilization is studied, and reasons for the high toughening effect are given. In the work of Chapter 5, the formulation effectiveness is demonstrated through a wide range of PLA contents. Chapter 6 further investigates the blend by finding a theoretical optimized point at 50 wt. % PLA. Finally, the thesis is concluded through a preliminary economic analysis, and outlining key outcomes in addition to considerations of future work.

7.4 Future Work

There are several key aspects of this work that fall beyond the scope of the studies detailed above. This includes, but is not limited to the following:

 Although the effects of the additives were studied in detail, the effect of processing conditions was only superficially investigated. Adjusting conditions such as extrusion temperature, retention time, screw speed, and mold temperature could have very beneficial effects on the performance of the material.

 The preceding point could be further used to create a framework for tailoring specific properties of each formulation. For example, high mold temperatures could be used to increase the crystallinity if modulus and heat deflection temperature are of importance in the blend.

203

 The developed materials should be investigated for their viability in natural fiber composites. This could increase biobased content, decrease cost, and if engineered correctly, could increase certain performance attributes.

 Complete life cycle studies would be valuable to ascertain the environmental benefits of the material over petroleum based ABS. This includes the effect of a possible reduction in the ability of the material to be recycled, as compared to ABS.

 Finally, full cost analysis, taking into account the production of the material at different scaling sizes would be beneficial.

7.5 References:

1. Acrylonitrile Butadiene Styrene (ABS) Global Market to 2020, http://www.prlog.org/12068741-acrylonitrile-butadiene-styrene-abs-global-market-to-2020.html,

(Accessed 2013).

2. Automotive Plastics Market for Passenger Cars by Type, http://www.marketsandmarkets.com/Market-Reports/automotive-plastics-market-passenger-cars-

506.html, (2013).

3. Ma, Z., Weersink, A., Vadori, R., & Misra, M. (2015). Financial Cost Comparison of

Acrylonitrile Butadiene Styrene (ABS) and BioABS. Journal of Biobased Materials and

Bioenergy, 9(2), 244-251.

204