Biodegradable Polymer Blends and Their Biocomposites: Compatibilization and Performance Evaluation

Transcription

1 i Biodegradable Polymer Blends and Their Biocomposites: Compatibilization and Performance Evaluation By Rajendran Muthuraj A Thesis Presented to The University of Guelph In partial fulfilment of requirements for the degree of Doctor of Philosophy in Engineering Guelph, Ontario, Canada Rajendran Muthuraj, November, 2015

2 ii ABSTRACT BIODEGRADABLE POLYMER BLENDS AND THEIR BIOCOMPOSITES: COMPATIBILIZATION AND PERFORMANCE EVALUATION Rajendran Muthuraj University of Guelph, 2015 Advisor: Dr. Amar K. Mohanty Co-advisor: Dr. Manjusri Misra Non-biodegradable polymers, polymer blends and composites are known to persist in the environment over a long time. The use of certain biodegradable polymers is limited as they often fail to match some of the non-biodegradable counterpart perfromances. Blends of biodegradable polymers and composites with complementary attributes can provide materials that strike a balance between cost and performance. This research was focused on the fabrication and performance evaluation of biodegradable polymer blends and composites, as potential alternatives to non-biodegradable polymeric materials. Industrially viable melt processing techniques like extrusion and injection molding were adopted to fabricate and evaluate the structure-property-relationship of biodegradable polymer blends and composites. In this research work, two different types of commercially available biodegradable polyesters, namely poly(butylene adipate-co-terephthalate) (PBAT) and poly(butylene succinate) (PBS) were used to fabricate binary blends and composites. Melt blending these two polymers yielded synergistic properties, which are not present in the respective polymers. The optimal PBS/PBAT blend was selected based on its overall performance and it was used as the standard biocomposite matrix. Miscanthus is a purpose grown energy crop and has not been explored much for polymer

3 iii composite applications. This fiber was used as reinforcing agents in the PBS, PBAT and blend of PBS/PBAT matrix to compare the effects of matrix properties upon the performance of the resulting composites. One important aspect of this study was reactive compatibilization to improve the interfacial adhesion between the miscanthus fibers and polymer matrix. Maleic anhydride grafted polyesters were synthesized as a compatibilizer, which was used to improve the compatibility with the miscanthus fibers and polymer matrices. The improved fiber-to-polymer matrix adhesion exhibited in better mechanical performances of the resulting composites compared to that of uncompatibilized counterparts. The influence of major processing parameters such as processing temperature, screw speed, fiber length, and holding pressure on the mechanical performance were statistically analyzed by factorial design of experiment. The impact strength of the PBS/PBAT/miscanthus fiber composites was significantly dependent on the fiber length. The durability of the biodegradable polymer (PBS, PBAT and PBS/PBAT blend) was investigated after being exposed to elevated temperature (50 o C) and humidity (90%) for 30 days. It was found that the mechanical properties of the samples were heavily affected under the selected environmental conditions and exposure time. An optimum biocomposite formulation was successfully extruded and injection molded for continuous prototype manufacturing in pilot-scale production facilities.

4 iv Dedicated to my family

5 v Acknowledgements It would not have been possible to accomplish this PhD thesis without my advisors, Dr. Amar Kumar Mohanty and Dr. Manjusri Misra. I would like to thank them for their consistent support, encouragement and guidance during the course of this project. I am also grateful to my advisory committee members, Dr. Animesh Dutta and Dr. Loong-Tak Lim for their fruitful comments towards my research work and support for my research work. Dr. Fantahun Defersha is gratefully acknowledged for his support in statistical analysis. I would like to express my sincere thanks to my parents Mr. P. Muthuraj, Mrs. M. Karuppathal, and my brother Mr. M. Venkatachalam and my sisters Miss. M. Bhaghyalakshmi and Mrs. M. Manoranjitham who provided never-ending support and kindness throughout my life. Also, I am very much thankful to the Bioproducts Discovery and Development Center (BDDC) colleagues and my friends who have directly and indirectly helped me throughout this project. I would like to express my gratitude to funding agencies that provided financial support to carry out this research: Ontario Ministry of Agriculture, Food and Rural Affairs (OMAFRA),, Ontario Ministry of Economic Development and Innovation (MEDI), Ontario Research Fund - Research Excellence Round 4 program (ORF-RE04), Natural Sciences and Engineering Research Council (NSERC) Discovery Grant, National Centre of Excellence (NCE) AUTO21 Network and our private sector partner - New Energy Farms for providing miscanthus fiber samples.

6 vi Patent List of publications Mohanty. A.K., Misra. M., Zarrinbakhsh. N., Muthuraj. R., Wang. T., U-Rodriguez. A., Vivekanandhan. S., Biodegradable polymer-based composites with tailored properties and method of making those, US provisional application, Application number , filed on March Peer-reviewed journal publications Muthuraj. R., Misra. M., Mohanty. A.K., 2014, Biodegradable Poly(butylene succinate) and Poly(butylene adipate-co-terephthalate) Blends: Reactive Extrusion and Performance Evaluation, Journal of Polymers and the Environment, 22, Muthuraj. R., Misra. M., Mohanty. A.K., 2015, Hydrolytic degradation of biodegradable polyesters under simulated environmental conditions. Journal of Applied Polymer Science, 132, Muthuraj. R., Misra. M., Mohanty. A.K., Injection Molded Sustainable Biocomposites From Poly(butylene succinate) Bioplastic and Perennial Grass, ACS sustainable chemistry & engineering, 2015, 3, Muthuraj. R., Misra. M., Defersha.F., Mohanty. A.K., Influence of Processing Parameters on the Impact Strength of miscanthus composites: A Statistical Approach, Composites Part A: Applied Science and Manufacturing, 2015, DOI: /j.compositesa Book chapter Muthuraj. R., Misra. M., Mohanty. A.K., 2015, Chapter 5: Studies on mechanical, thermal, and morphological characteristics of biocomposites from biodegradable polymer blends and natural fibers. In: Misra. M, Pandey. J. K, Mohanty. A.K., (Eds.) Biocomposites: Design and Mechanical Performance, Woodhead Publishing Limited, Cambridge, UK, pp

7 vii Peer-reviewed conference publications Muthuraj. R., Misra. M., Mohanty. A.K., 2015, Durability Studies of Biodegradable Polymers under Accelerated Weathering Conditions, Society of Plastic Engineering (ANTEC), March 23-25, Orlando, Florida, USA. Muthuraj. R., Misra. M., Mohanty. A.K., 2015, Binary Blends of poly(butylene adipateco-terephthalate) and poly(butylene succinate): A new matrix for biocomposites applications, AIP Conf. Proc. 1664, Muthuraj. R., Misra. M., Mohanty. A.K., 2013, Plasticization of Co-products from Bioethanol Industries: Potential Uses in Biocomposites, The 19 th International Conference on Composite Materials (ICCM), pp

8 xxi Table of Contents Acknowledgements... v List of publications... vi List of Tables... xxix List of Figures... xxxii List of abbreviations and defined terms... xlii Chapter 1: Introduction... 1 Abstract Research problems Objectives and Hypotheses Thesis organization... 4 Chapter 2: Studies on Mechanical, Thermal and Morphological Characteristics of Biocomposites from Biodegradable Polymer Blends and Natural Fibers* Introduction Biodegradable and compostable polymeric materials Renewable resource based biodegradable polymers: Some examples Poly(lactic acid), PLA Microbial polyesters-polyhydroxyalkanoates (PHAs) Fossil fuel based biodegradable polymers: Some examples Poly(butylene succinate), PBS Poly(butylene adipate-co-terephthalate), PBAT Poly(caprolactone), PCL Recyclability of biodegradable polymers Durability of biodegradable polymers Polymer blends: Some examples Miscible biodegradable polymer blends Immiscible biodegradable polymer blends Compatibilization of polymer blends Non-reactive compatibilization of biodegradable polymer blends Reactive compatibilization of biodegradable polymer blends: Few specific examples... 35

9 xxii Reactive compatibilization of PLA/PBAT blends: Reactive compatibilization of PLA/PBS blends: Reactive compatibilization of PLA/PHB and PHBV/PBS blends: Reactive compatibilization of PLA/PCL blends: Natural fibers Classification of natural fibers Natural fibers: nature and behavior Advantages and challenges in using natural fibers Biocomposites Advantageous of natural fiber composites Attributes of natural fiber composites Biocomposites based on biodegradable blends as matrix material: Some specific examples Biocomposites based on PHBV blends Biocomposites based on PLA blends Natural fiber composites market and their applications Conclusions References Chapter 3: Fully Biodegradable Poly (butylene succinate) and Poly (butylene adipate-co-terephthalate) Blends: Reactive Extrusion and Performance Evaluation* Abstract Introduction Experimental section Materials Fourier transform infrared spectroscopy (FTIR) Mechanical properties Differential scanning calorimetry (DSC) Dynamic mechanical analysis (DMA) Heat deflection temperature (HDT) Thermogravimetric analysis (TGA) Rheological studies... 94

10 xxiii Polarizing optical microscopy (POM) Scanning electron microscopy (SEM) Results and Discussion Fourier transform infrared spectroscopy (FTIR) Mechanical properties Melt flow index Differential scanning calorimetry Dynamic mechanical analysis Heat deflection temperature Thermogravimetric analysis Rheological properties Polarizing optical microscopy Scanning electron microscopy Conclusions References Chapter 4: Preparation and Characterization of Maleic Anhydride Grafted Biodegradable Polyesters Abstract Introduction Materials and Methods Synthesis of MAH grafted PBS, PBAT and their blend Grafting mechanism Purification of MAH grafted samples Determination of grafting percentage Gel percentage measurement Fourier transform infrared (FTIR) spectroscopy Thermogravimetric analysis (TGA) Differential scanning calorimetry (DSC) Results and Discussion Infrared spectroscopy MAH grafting percentage calculation

11 xxiv Gel content measurement Thermogravimetric analysis Differential scanning calorimetry Conclusions References Chapter 5: Enhanced Mechanical Performances of Fully Biodegradable Miscanthus Fibers Reinforced Poly (butylene succinate) Composites* Abstract Introduction Materials and Methods Materials Thermal property Biocomposite preparation Mechanical testing Statistical analysis Dynamic mechanical analysis Melt flow index (MFI) Differential scanning calorimetry (DSC) Scanning electron microscopy (SEM) Results and discussion Thermogravimetric analysis Efficiency of compatibilizer Mechanical properties versus fiber loading Dynamic mechanical analysis Adhesion factor calculation Heat deflection temperature Melt flow analysis Differential scanning calorimetry Morphological analysis Conclusions References

12 xxv Chapter 6: Mechanical Performances of Biocomposites Made From Miscanthus Fibers and Poly(butylene adipate-co-terephthalate) Matrix Abstract Introduction Materials Biocomposite fabrication method Characterization methods Results and Discussion Mechanical properties Melt flow index and Heat deflection temperature Scanning electron microscopy Conclusions References Chapter 7: Biocomposites Consisting of Miscanthus Fibres in a Biodegradable Binary Blend Matrix: Preparation and Performance Evaluation* Abstract Introduction Materials Processing of polymer blend and their composites Mechanical properties Density Dynamic mechanical analysis (DMA) Differential scanning calorimetry (DSC) Thermogravimetric analysis (TGA) Morphological analysis Rheological property Results and Discussion Mechanical properties Theoretical approximation of Young s modulus of the PBS/PBAT biocomposites Dynamic mechanical properties Density

13 xxvi Heat deflection temperature Thermogravimetric analysis Differential scanning calorimetry Measurements of fiber diameter, length, and aspect ratio Morphology of composites Rheological property Conclusions References Chapter 8: Influence of Processing Parameters on the Impact Strength of Biocomposites: A Statistical Approach* Abstract Introduction Full factorial design methodology Materials Experimental procedure Samples preparation Characterization methods Fiber dimension measurement Mechanical testing and scanning electron microscopy (SEM) Results and Discussion Mechanical properties Analysis of variance (ANOVA) for impact strength Effect of processing parameters on the impact strength Fiber length distribution Scanning electron microscopy Mathematical model development Diagnostic verification of the developed model Conclusions References Chapter 9: Hydrolytic Degradation of Biodegradable Polyesters under Simulated Environmental Conditions*

14 xxvii Abstract Introduction Materials and Methodology Materials Sample preparation and conditioning Moisture absorption Fourier transform infrared spectroscopy (FTIR) Mechanical properties Differential scanning calorimetry (DSC) Dynamic mechanical analysis (DMA) Rheological properties Polarizing optical microscopy (POM) Morphological analysis Results and Discussion Moisture absorption Hydrolytic degradation mechanism of PBS and PBAT Changes in mechanical properties Differential scanning calorimetry Dynamic mechanical analysis Rheological properties Polarizing optical microscopy Morphological analysis Conclusions References Chapter 10: Conclusions, Contributions, and Recommendations for Future Work Abstract Overview Conclusions Significant contributions Recommendations for future works

15 xxviii Appendix I: Binary Blends of Poly(Butylene Succinate) and Poly(Butylene Adipateco-Terephthalate): A New Matrix for Biocomposites Applications* Abstract A-I.1.Introduction A-I.2. Materials and Methods A-I.3. Results and Discussion A-I.4. Conclusions References Appendix II: Durability Studies of Biodegradable Polymers under Accelerated Weathering Conditions* Abstract A-II.1. Introduction A-II.2. Materials and Methods A-II.3. Results and Discussion A-II.4. Conclusions

16 xxix List of Tables Table 2.1. Classification and molecular structure of some biodegradable polymers Table 2.2. Properties of some biodegradable polymers in comparison to nonbiodegradable polymers Table 2.3. Ecoflex based masterbatches for different applications [66,67] Table 2.4. List of companies engaged in the production of some biodegradable polymer blends (the table was modified after referene [9,94,95]) Table 2.5. Properties of some natural and synthetic fibers Table 2.6. Recently developed biodegradable polymer blend matrix based biocomposites56 Table 3.1. Melt flow index (MFI) of the neat polymers and their blends Table 3.2. Solubility parameter values for polymers Table 3.3. Heat deflection temperatures of the neat polymers and their blends Table 4.1. Properties of the neat PBS, PBAT and PBS/PBAT blend Table 4.2. Proposed formulation for producing maleation of PBS, PBAT, and PBS/PBAT blend Table 4.3. Maleic anhydride grafting percentage of the PBS, PBAT and PBS/PBAT blend Table 4.4. Detailed DSC results of the maleated PBS and PBS/PBAT blend (MAH grafted samples were prepared with 1 phr DCP and 5 phr MAH) Table 5.1. Mechanical properties of PBS composites with two different MAH grafting levels of compatibilizer Table 5.2. A comparison of mechanical properties of injection molded PBS/natural fiber composites (Note: the reported percentage differences were calculated based on the neat PBS matrix properties)

17 xxx Table 5.3. Heat deflection temperature (HDT) and melt flow index (MFI) of neat PBS and its composites Table 5.4. Summary of differential scanning calorimetry traces of neat PBS and its composites Table 6.1. Melt flow index (MFI) and Heat deflection temperature (HDT) measurement204 Table 7.1. Notched Izod impact strength of PBS/PBAT blend and its compatibilized and uncompatibilized composites Table 7.2. Heat deflection temperature (HDT) and density of PBS/PBAT blend and its compatibilized and uncompatibilized composites Table 7.3. Thermogravimetric data of miscanthus, PBS/PBAT blend and their composites Table 7.4. Detailed differential scanning calorimetry analysis of the PBS/PBAT blend and their composites Table 7.5. Average fiber length (L), average fiber diameter (D), and aspect ratio (L/D) of the miscanthus fiber before and after compounding Table 8.1. Selected processing parameters and their respected levels Table 8.2. Physical and mechanical properties of the PBS/PBAT blend and miscanthus fibers Table 8.3. The 16 investigated experimental conditions Table 8.4. A complete summary of all the experiments and the related mechanical properties of PBS/PBAT/miscanthus composites Table 8.5. Analysis of Variance (ANOVA) for notched Izod impact strength Table 8.6. Average fiber length (L), average fiber diameter (D), and aspect ratio (L/D) of the miscanthus fiber before and after compounding Table 9.1. Notched Izod impact strength of the samples before and after conditioned at 50 o C with 90% relative humidity

18 xxxi Table 9.2. DSC results for PBS, PBAT and their blend before and after 30 days conditioned at 50 o C with 90% relative humidity Table 9.3. Relative molecular weight (M 1 /M 2 ) of the PBS, PBAT and PBS/PBAT blend before and after 6 days conditioned at 50 o C with 90% relative humidity Table A-II.1. General properties of PBS, PBAT, PBS/PBAT (60/40 wt%) and PP. ( a obtained from material data sheet, b and PBS/PBAT (60/40 wt%) data were measured in the lab) Table A-II.2. Notched Izod impact strength of PBS, PBAT, PBS/PBAT (60/40 wt%) and PP before and after 18 days conditioning at 50 o C with 90% RH 340

19 xxxii List of Figures Figure 2.1. (a) Schematic representation of the evolution of morphology in a binary immiscible blend, (b) matrix/dispersed morphology, and (c) co-continuous morphology (adapted with kind permission from Ravati and Favis, Polymer, 2010, 51: , Copyright 2015, Elesevier, Licence number [113]).. 30 Figure 2.2. Adding a compatibilizing agent, such as a diblock copolymer, to a polymer blend can improve its stability, but is more likely to result in a dispersed morphology rather than a co-continuous morphology. a) A two-dimensional slice of a compatibilized blend with dispersed phase morphology, represented by minority dark blue phase and a majority turquoise phase. b) A molecular schematic showing how the diblock copolymers are segregated at the interface between the two phases (adapted with kind permission from Ryan, Nature Materials, 2002, 1: Copyright 2015, nature publishing group, License number [142]) Figure 3.1. Evaluation of the normalized FTIR spectra of the carbonyl region ( cm -1 ) of PBS, PBAT and their blends Figure 3.2. Tensile stress-strain curves of PBS, PBAT, and their blends Figure 3.3. Tensile strength and elongation at break of PBS, PBAT, and their blends: (A) PBS; (B) PBS/PBAT(70/30 wt%); (C) PBS/PBAT (60/40 wt%); (D) PBS/PBAT (50/50 wt%); (E) PBAT Figure 3.4. Second heating DSC thermograms of PBS, PBAT, and their blends after cooling at 5 C/min Figure 3.5. Theoretical and experimental values of T g for PBS/ PBAT blends Figure 3.6. Storage moduli of PBS, PBAT, and their blends Figure 3.7. Tan curves of PBS, PBAT, and their blends Figure 3.8. TGA curves of PBS, PBAT, and their blends

20 xxxiii Figure 3.9. DTG curves of PBS, PBAT, and their blends Figure Complex viscosity of PBS, PBAT, and their blends with different weight fractions of PBAT at 140 o C Figure Loss modulus versus frequency for PBS, PBAT, and their blends with different weight fractions of PBAT at 140 o C Figure Storage modulus versus frequency for PBS, PBAT, and their blends with different weight fractions of PBAT at 140 o C Figure Cole Cole plot of the PBS/PBAT blends at 140 o C Figure (a) Photograph of the film annealed at 80 o C: (i) PBS; (ii) PBS/PBAT (70/30 wt%); (iii) PBS/PBAT (60/40 wt%) and (iv) PBS/PBAT (50/50 wt%). Figure (b) Photograph of the film annealed at 90 o C: (i) PBS; (ii) PBS/PBAT (70/30 wt%); (iii) PBS/PBAT (60/40 wt%) and (iv) PBS/PBAT (50/50 wt%) Figure (a) SEM images of PBS and PBAT blends (left hand side) (i) PBS/PBAT (70/30 wt%); (ii) PBS/PBAT (60/40 wt%) and (iii) PBS/PBAT (50/50 wt%). (b) SEM images of PBS and PBAT blends surface after et al.,hing with THF (right hand side): (i) PBS/PBAT (70/30 wt%); (ii) PBS/PBAT (60/40 wt%) and (iii) PBS/PBAT (50/50 wt%) Figure 4.1. Proposed reaction mechanism of maleic anhydride grafted PBS (MAH-g- PBS) Figure 4.2. Proposed reaction mechanism of maleic anhydride grafted PBAT (MAH-g- PBAT) Figure 4.3. FTIR spectra of MAH, neat PBS and MAH-g-PBS with 1 phr DCP and 5 phr MAH Figure 4.4. FTIR spectra of the MAH, neat PBAT and MAH-g-PBAT with 1 phr DCP and 5 phr MAH

21 xxxiv Figure 4.5. FTIR spectra of MAH, neat PBS/PBAT blend and MAH-g-PBS/PBAT blend with 1 phr DCP and 5 phr MAH Figure 4.6. Gel content of maleic anhydride grafted PBS, PBAT, and PBS/PBAT samples with 5 phr MAH and different concentration of DCP Figure 4.7. TGA thermograms of neat and maleated PBS, PBAT and PBS/PBAT blend (the maleated samples were obtained with 1 phr DCP and 5 phr MAH) Figure 4.8. DSC second heating curves of neat PBS, PBS/PBAT (60/40 wt%) and their maleated samples with 1 phr DCP and 5 phr MAH Figure 4.9. DSC first cooling curves of neat PBS, PBS/PBAT (60/40 wt%) and their maleated samples with 1phr DCP and 5 phr MAH Figure 5.1.SEM micrograph of as received miscanhtus fibers Figure 5.2. Thermogravimetric analysis of miscanthus fiber under different environment Figure 5.3. Tensile properties of PBS and PBS/miscanthus composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5wt%) Figure 5.4. Reduced tensile strength of uncompatibilized and compatibilized PBS/miscanthus composites plotted against volume fraction of fibers according to equation Figure 5.5. Flexural properties of PBS and PBS/miscanthus composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%).. 172

22 xxxv Figure 5.6. Nothced Izod impact strength of PBS and PBS/miscanthus composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) Figure 5.7. Dynamic mechanical analysis of PBS and PBS/miscanthus composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) Figure 5.8. Tan δ curves of PBS and its composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) Figure 5.9. Adhesion factor of PBS/miscanthus composites: A) PBS/miscanthus (70/30 wt%), B) PBS/miscanthus/compatibilizer (65/30/5 wt%), C) PBS/miscanthus (60/40 wt%), D) PBS/miscanthus/compatibilizer (55/40/5 wt%), E) PBS/miscanthus (50/50 wt%), and F) PBS/miscanthus/compatibilizer (45/50/5 wt%) Figure DSC second heating thermograms of PBS and its composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) Figure DSC first cooling thermograms of PBS and its composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) Figure SEM micrograph of tensile fractured surface of uncompatibilized PBS composites with low (150x) and high (500x) magnification; (a) and (b) are PBS/miscanthus (70/30 wt%) composites; (c) and (d) are PBS/miscanthus (60/40 wt%) composites; (e) and (f) are PBS/miscanthus (50/50 wt%) composites

23 xxxvi Figure SEM micrograph of tensile fractured surface of compatibilized PBS composites with low (150x) and high (500x) magnification; (a) and (b) are PBS/miscanthus (70/30 wt%) composites; (c) and (d) are PBS/miscanthus (60/40 wt%) composites; (e) and (f) are PBS/miscanthus (50/50 wt%) composites Figure 6.1. Tensile strength and tensile modulus of PBAT and its composites; (A) neat PBAT, (B) PBAT/miscanthus fibers (70/30 wt%), (C) PBAT/miscanthus fibers/mah-g- PBAT (65/30/5 wt%), (D) PBAT/miscanthus fibers (60/40 wt%), and (E) PBAT/miscanthus fibers/mah-g-pbat (65/30/5 wt%) Figure 6.2. Flexural properties of PBAT and its compatibilized and uncompatibilized composites: (A) neat PBAT, (B) PBAT/miscanthus fibers (70/30 wt%), (C) PBAT/miscanthus fibers/mah-g-pbat (65/30/5 wt%), (D) PBAT/miscanthus fibers (60/40 wt%), and (E) PBAT/miscanthus fibers/mah-g-pbat (65/30/5 wt%) Figure 6.3. Notched Izod impact strength of PBAT and its compatibilized and uncompatibilized composites: (A) neat PBAT, (B) PBAT/miscanthus fibers (70/30 wt%), (C) PBAT/miscanthus fibers/mah-g-pbat (65/30/5 wt%), (D) PBAT/miscanthus fibers (60/40 wt%), and (E) PBAT/miscanthus fibers/mah-g-pbat (65/30/5 wt%) Figure 6.4. SEM micrographs of uncompatibilized PBAT/misanthus fibers (60/40 wt%) composites (A) and compatibilized PBAT/misanthus fibers/mah-g-pbat (55/40/5 wt%) composites (B) Figure 7.1. Tensile properties of PBS/PBAT blend and their composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (65/30/5 wt%), (D) PBS/PBAT + miscanthus fibers (60/40 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), (F) PBS/PBAT + miscanthus fibers (50/50 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%) Figure 7.2. Stress-strain curves of PBS/PBAT blend and their composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%)

24 xxxvii Figure 7.3. Flexural properties of PBS/PBAT blend and their composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (65/30/5 wt%), (D) PBS/PBAT + miscanthus fibers (60/40 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), (F) PBS/PBAT + miscanthus fibers (50/50 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%) Figure 7.4. Expected reaction between the miscanthus and the compatibilizer Figure 7.5. Variation of experimental and theoretical values of Young s modulus as a function of fiber loading Figure 7.6. Storage moduli of PBS/PBAT blend and their composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT+ miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%) Figure 7.7. Tan δ of PBS/PBAT blend and their composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT+ miscanthus fibers + MAHg-PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (45/50/5 wt%) Figure 7.8. Thermogravimetric traces for miscanthus, PBS/PBAT blend and its composites Figure 7.9. DSC second heating thermograms: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT+ miscanthus fibers + MAH-g- PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (45/50/5 wt%)

25 xxxviii Figure DSC first cooling thermograms: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT+ miscanthus fibers + MAH-g- PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (45/50/5 wt%) Figure Fiber length distribution before and after compounding: (A) as received miscanthus fibers distribution, (B) fibers distribution in 30 wt% composites, (C) fibers distribution in 40 wt% composites, and (D) fibers distribution in 50 wt% composites Figure SEM micrographs of uncompatibilized PBS/PBAT blend composites with different fiber loads: (A) PBS/PBAT + miscanthus fibers (70/30 wt%), (B) PBS/PBAT + miscanthus fibers (60/40 wt%), and (C) PBS/PBAT + miscanthus fibers (50/50 wt%). 238 Figure SEM micrographs of compatibilized PBS/PBAT blend composites with different amount of fiber loads: (A) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (65/30/5 wt%), (B) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), and (C) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%) Figure Complex viscosity of PBS/PBAT blend and its composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%) Figure 8.1. Main effect plot for the impact strength Figure 8.2. Plot of interaction effects for the impact strength of biocomposites Figure 8.3. Half Normal probability plot of the standardized effects for impact strength of the PBS/PBAT/miscanthus composites

26 xxxix Figure 8.4. Pareto chart of the standardized effects for the impact strength of the PBS/PBAT/miscanthus biocomposites Figure 8.5. Tensile stress-strain curves of PBS/PBAT/miscanthus composites with changing fiber length 4 mm (A) and 2 mm (B) Figure 8.6. Histograms of miscanthus fiber length distribution before and after compounding in a twin screw extruder Figure 8.7. Represents the SEM micrographs of the PBS/PBAT/miscanthus composites; (a) PBS/PBAT composites with 2 mm miscanthus (b) PBS/PBAT composites with 4 mm miscanthus Figure 8.8. Normal probability plot of the residuals for impact strength Figure 8.9. Residual plots versus fitted values for impact strength Figure Variation of the residuals with observed order values of the impact strength of the PBS/PBAT/miscanthus composites Figure 9.1. Moisture absorption curves as a function of conditioning time Figure 9.2. Hydrolysis reaction of PBS Figure 9.3. Hydrolysis reaction of PBAT Figure 9.4. FTIR spectra of PBS, PBAT and PBS/PBAT before and after 30 days exposed to 50 o C with 90% relative humidity Figure 9.5. Tensile strength of PP, PBS, PBAT and PBS/PBAT as a function of exposure time at 50 o C with 90% relative humidity Figure 9.6. Flexural strength of PP, PBS, PBAT and PBS/PBAT as a function of exposure time at 50 o C with 90% relative humidity

27 xl Figure 9.7. Testing failure mode of PBS, PBAT, PBS/PBAT and PP after 30 days exposed to 50 o C with 90% relative humidity Figure 9.8. Percentage elongation of PP, PBS, PBAT and PBS/PBAT as a function of exposure time at 50 o C with 90% relative humidity Figure 9.9. Tensile modulus of PP, PBS, PBAT and PBS/PBAT as a function of exposure time at 50 o C with 90% relative humidity Figure Flexural modulus of PP, PBS, PBAT and PBS/PBAT as a function of exposure time at 50 o C with 90% relative humidity Figure DSC heating cycles for PBS, PBAT and PBS/PBAT before and after exposed to 50 o C with 90% relative humidity for 30 days Figure DSC cooling curves for PBS, PBAT and PBS/PBAT before and after exposed to 50 o C with 90% relative humidity for 30 days Figure Storage modulus of PBS, PBAT and PBS/PBAT before and after exposed to 50 o C with 90% relative humidity for 30 days Figure Loss factor peak (tan δ) of PBS, PBAT and PBS/PBAT before and after 30 days exposed to 50 o C with a relative humidity of 90% Figure Shear viscosity curves for PBS, PBAT and PBS/PBAT before and after 6 days exposed to 50 o C with a relative humidity of 90% Figure Polarized optical micrographs of PBS, PBAT and PBS/PBAT before and after 30 days conditioned at 50 o C and 90% relative humidity Figure SEM micrographs of PBS, PBAT and PBS/PBAT before and after 30 days conditioned at 50 o C and 90% relative humidity

28 xli Figure Prototypes were made from biodegradable polymers/miscanthus fiber by extrusion and injection molding method Figure A-I. 1. Tensile properties of (A) PBAT, (B) PBS, (C) PBAT/PBS (60/40 wt%) and (D) PBAT/PBS (70/30 wt%) blend Figure A-I. 2. HDT and MFI values of (A) PBAT, (B) PBS, (C) PBAT/PBS (60/40 wt%) and (D) PBAT/PBS (70/30 wt%) blend Figure A-I. 3. SEM image of the cryofractured PBAT/PBS (60/40 wt%) blend Figure A-I. 4. POM image of the (i) PBS and (ii) PBAT/PBS (60/40 wt%) blend Figure A-II. 1. Tensile strength of PBS, PBAT, PBS/PBAT (60/40 wt%) and PP before and after 18 days exposed to 50 o C with 90% RH Figure A-II. 2. Flexural strength of PBS, PBAT, PBS/PBAT (60/40 wt%) and PP before and after 18 days exposed to 50 o C with 90% RH Figure A-II. 3. SEM micrographs of PBS, PBAT, and PBS/PBAT (60/40 wt%) before (A, B and C) and after (D, E and F) 18 days exposed to 50 o C with 90% RH

29 xlii List of abbreviations and defined terms ABS Acrylonitrile butadiene styrene ANOVA APTMS ASTM ATR BDO DCP DDGS DMA DSC DTG EBA-GMA EGMA EMAA-Zn EMA-GMA ENR EPA EPDM FDA FRCs FTIR GMA HB HDPE Analysis of Variance (3-aminopropyl) trimethoxysilane American society for testing and materials Attenuated total reflectance 1,4-butanediol Dicumyl peroxide Distiller s dried grains with solubles Dynamic mechanical analysis Differential scanning calorimetry Derivative thermogram Ethyl-butyl acrylate and glycidyl methacrylate copolymer Ethylene-glycidyl methacrylate Ethylene methacrylic acid zinc ionomer Ethylene methyl acrylate-glycidyl methacrylate terpolymer Epoxidized natural rubber Environmental protection agency Ethylene-propylene-diene terpolymer Food and drug administration Fiber reinforced plastic composites Fourier Transform Infrared Spectroscopy Glycidyl methacrylate Hydroxybutyrate High-density polyethylene

30 xliii HDT HSD HV ISO KOH LTI LVE MA-g-PP MAH Heat deflection temperature Honestly significant difference Hydroxyvalerate International organization for standardization Potassium hydroxide Lysine triisocynate Linear viscoelastic Maleic anhydride-grafted-polypropylene Maleic anhydride MAH-g-PBS/PBAT Maleic anhydride-grafted-poly(butylene succinate)/poly(butylene adipate-co-terephthalate) MFI NaOH PBAT PBS PBSA PBT PC PCL PDI PDLA PE PEBA PET PHAs Melt flow index Sodium hydroxide Poly(butylene adipate-co-terephthalate) Poly(butylene succinate) Poly(butylene succinate-co-adipate) Poly(butylene terephthalate) Polycarbonate Polycaprolactone Poly dispersity index Poly(D-lactic acid) Polyethylene Poly(ether-b-amide) copolymer Poly(ethylene terephthalate) Polyhydroxyalkanoates

31 xliv PHB PHB-g-PBS PHB-g-PLA PHBV PHBV-g-PBS PLA PLLA PMDI PMMA PP PPC PS PVA PVAc PVDF RH ROP RWF SAN SEM TFC TGA T-GMA THF TMPTA Poly(3-hydroxybutyrate) Poly(hydroxybutyrate)-grafted-poly(butylene succinate) Poly(hydroxybutyrate)-grafted-poly(lactic acid) Poly(3-hydroxybutyrate-co-3-hydroxyvalerate) Poly(3-hydroxybutyrate-co-3-valerate)-grafted-poly(butylene succinate) Poly(lactic acid) Poly(L-lactic acid) Polymeric methylene diphenylene diisocyanate Poly(methyl methacrylate) Polypropylene Poly(propylene carbonate) Polystyrene Poly(vinyl alcohol) Poly(vinyl acetate) Poly(vinylidene fluoride) Relative humidity Ring opening polymerization Recycled wood fiber Styrene acrylonitrile Scanning electron microscopy Twice functionalized nanoclay Thermogravimetric analysis Random terpolymer of ethylene, acrylic ester and glycidyl methacrylate Tetrahydrofuran Trimethylolpropane triacrylate

32 xlv TPP Triphenyl phosphite Nomenclature % Percentage Crystallization enthalpy Melting enthalpy Theoretical melting enthalpy of 100% crystalline polymers Young s modulus of the fibers Young s modulus of the matrix Volume fraction of fiber Volume fractions of matrix G m H m S c m S e m DF Gibbs free energy Heat of mixing Mixing of combinatorial entropy Mixing of excess entropy Degree of freedom E Storage modulus E L E T F G'' G' g/10min g/cm 3 GPa Longitudinal modulus Transverse modulus F-value (statistics) Loss modulus at melt state Storage modulus at melt state Grams per 10 minutes Gram per cubic centimeter Giga Pascal

33 xlvi h J/g kn kv l/d M 1 /M 2 ma M c mol% MPa MS Mw NA P Pa.s phr R R 2 R 2 adj SS T Tan δ T c T g Hours Joule per gram kilonewton kilovolts length/diameter of the reinforcement Relative molecular weight milli-ampere Young s modulus of the composites mole percentage Mega pascal Mean square Weight average molecular weight Not applicable P-value (in ANOVA table) Pascal second Parts per hundred Universal gas constant R-squared Adjusted R-squared Sum of square Absolute temperature Loss modulus to storage modulus ratio Crystallization temperature Glass transition temperature

34 xlvii T m T max T onset wt% Melting temperature Maximum degradation temperature Onset degradation temperature Weight percentage Greek symbols Volume fractions of the fiber Volume fractions of the matrix Densities of the composites Densities of the fiber Densities of the matrix T B ƞ o x α δ Load-bearing capacity of the dispersed component Zero shear viscosity Stress transfer constant Alpha level (statistics) Solubility parameters η* Complex viscosity λ σ χ χ Ratio of length measured before (L0) and after (L) tensile test Tensile strength of composites Crystallinity Flory-Huggins interaction parameter Group molar attraction constant of the polymer Reinforcement geometry

35 Chapter 1: Introduction Abstract: This chapter briefly describes the research problems associated to this research work. In order to address these research problems, this chapter provides a hypothesis followed by objectives of the present work. The main objectives to be accomplished in this research work are presented. This chapter also discusses the thesis organization and a short summary of each following chapters. 1.1 Research problems After use, petroleum-based non-biodegradable polymers persist in the soil for hundreds of years, creating environmental concerns. Carbons from these polymers are not renewable and their supply is quickly depleting. In addition, non-biodegradable polymers are not satisfactory materials for short-lifespan applications such as packaging, catering, textile, agricultural, household uses, and surgery, as incineration of many non-biodegradable polymeric material wastes can produce non eco-friendly emissions. Recycling these products may require highenergy consumption. Therefore, biodegradable/compostable polymers are an alternative for these non-biodegradable/non-compostable polymers. The insufficient performances of the biodegradable polymers are limiting their certain applications. This can be overcome by melt blending two or more dissimilar biodegradable polymers while preserving biodegradability of the parent components. Furthermore, the cost of biodegradable polymer is quite expensive compared to that of commonly used polymers. Biodegradable polymers and their blends are often used to produce biocomposites to modify overall properties. The use of biodegradable polymers and their blends as a matrix for producing biocomposites will result in products with enhanced performances and lower cost. The relatively hydrophobic nature of biodegradable polymer matrix is less compatible with hydrophilic natural fibers due to the polarity difference 1

36 between them. The incompatibility between the polymers and fibers leads to weak stress transfer from one phase to another phase. This weak stress transfer causes inferior mechanical properties of the resulting composites. This problem can be effectively addressed by compatibilization strategy. 1.2 Objectives and Hypotheses The optimal blend can be selected based on its overall performance and it can be used as the standard biocomposite matrix. Increase in the cost of biocomposites could be mitigated by adding environmentally friendly low cost fibers/reinforcement such as miscanthus fibers [1]. The selected fiber crop does not currently have good value added applications in composites sectors [2]. Finding value-added applications for this material would improve the economics for the respective producers. Compatibilization chemistry has a predominant role in the performances of yielded biocomposites. Maleic anhydride (MAH) is a well-known functional monomer for grafting onto the polymer backbone because of its chemical reactivity, lower efficiency of self polymerization, reactive compatibilizer, and low toxicity. The MAH grafted copolymer can significantly improve interfacial adhesion and thereby result in good mechanical properties of the composites. Carlson et al., [3] observed the enhanced interfacial adhesion/interaction between the polylactide (PLA)/starch composites with the addition of MAH grafted PLA compatibilizer. Some other researchers have reported improved interfacial adhesion in the PLA composites with the addition of MAH grafted PLA as a reactive compatibilizer [4, 5]. Similarly, poly(butylene adipate-co-terephthalate) (PBAT) and poly(butylene succinate) (PBS) based composites were compatibilized by the use of MAH grafted PBAT [6] and MAH grafted PBS [7], respectively. To our knowledge, there was no literature available on the compatibilized PBS/miscanthus 2

37 composites, PBAT/miscanthus composites and PBS/PBAT/miscanthus composites with MAH grafted reactive compatibilizer. The purpose of this study was to use reactive compatibilization strategy to increase the compatibility between the PBS/miscanthus fiber, PBAT/miscanthus fiber and PBS/PBAT/miscanthus fibers. In this study, MAH grafted biopolymers were synthesized as a compatibilizer, which was anticipated to have a high compatibility with the hydroxyl functionality of the reinforcements (e.g., miscanthus). Hence, a MAH grafted biopolymer was used for the target compatibilization of this work. Improved fibers-polymer matrix adhesion was hypothesized to result in better mechanical properties of the resulting biocomposites. Moreover, fibers-matrix adhesion was evaluated through theoretical methods, microscopical analysis, and the performances of the resulting biocomposites. The objectives of this project were as follows: Objective 1 was to generate optimum property for biodegradable binary polymer blends of poly(butylene adipate-co-terephthalate), PBAT and poly(butylene succinate), PBS, which can be used as a matrix for biocomposite fabrication. In addition, the durability of PBS, PBAT and their blend was investigated after being exposed to elevated humidity (90%) and temperature (50 o C). Objective 2 was to functionalize PBS, PBAT and their blend by the melt free radical grafting of maleic anhydride (MAH) onto its backbone. These MAH functionalized polymers were use as a compatibilizing agent in subsequent steps. The first part of Objective 3 was to investigate the compatabilization effect of MAH grafted PBS on performances of the PBS/miscanthus composites. Second part of this objective 3

38 was to understand and investigate how the addition of miscanthus fibers and MAH gafted PBAT to PBAT influences the melt processes as well as their mechanical properties. Objective 4 was to fabricate and to evaluate performance of miscanthus fiber reinforced PBS/PBAT blend matrix based biocomposites. The effect of compatibilizer (MAH-grafted- PBS/PBAT blend) on the resulting biocomposites was investigated by means of mechanical, thermo-mechanical, morphological, and rheological properties. A statistical analysis was used to map the relations between the mechanical properties of the PBS/PBAT/miscanthus biocomposites and the processing parameters. 1.3 Thesis organization This thesis is separated into several chapters. Chapter 1 explains outline of the thesis, including hypothesis, and objectives of this project. Chapter 2 reviews the literature on mechanical, thermal and morphological characteristics of biodegradable polymer blends and their natural fiber reinforced composites (biocomposites) made from biodegradable polymer blends and natural fibers. Structure-propertyperformance of biodegradable polymer blends and their biocomposites fabricated using melt processing methods is the prime focus of this chapter. In addition to the economical advantages, natural fibers also offer several other benefits and challenges in fabrication of biocomposite materials, which are all reviewed in detail in different sections of this chapter. Strategies like reactive compatibilization have been widely studied to overcome the compatibility issue between the matrix and fiber phases. This chapter has also attempted to correlate the enhanced fiber/matrix adhesion with the resulting biocomposite performances. Chapter 3 discusses the preparation and performance evaluation of biodegradable PBS and PBAT binary blends. The mechanical properties of the resulting blends were correlated with 4

39 compatibility of the blended components and infrared spectroscopy. The processability and phase morphology of the prepared PBS/PBAT blends were investigated by rheological properties and microscopic analysis, respectively. Furthermore, the performance of the PBS/PBAT blends including thermal and thermo-mechanical properties were investigated and reported. Chapter 4 describes experiments carried out in accomplishment of an objective 3. The melt state graft copolymerization of biodegradable polymers with maleic anhydride (MAH). The MAH grafting efficiency on the PBS, PBAT and PBS/PBAT blend was investigated while varying concentrations of the dicumyl peroxide (DCP) free radical initiator. The percentage of maleic anhydride grafting on the polyesters backbone was quantified based on acid base titration method. In addition, the graft copolymerization was confirmed by FTIR spectroscopic analysis. Differential scanning calorimetry (DSC) and thermogravimertic analysis (TGA) were used to study the effect of MAH grafting on the polyesters backbone. Chapter 5 discusses the effects of compatibilizer concentration on the mechanical performances of the PBS matrix based biocomposites. In order to achieve maximum performance of the PBS/miscantus composites, the composites were prepared as a function of compatibilizer concentrations and fiber loadings. Two different compatibilizers, i.e., higher and lower degree of maleic anhydride grafted PBS (MAH-g-PBS), were used to explore the influence of compatibilizer on mechanical performance of resulting composites. It was observed that the composites with 5 wt% compatibilizer exhibited optimum mechanical performances. The loadbearing capacity/reinforcing effect of the miscanthus fiber composite was analyzed with the help of theoretical methods. In addition, the degree of interaction between the components was demonstrated from the height of tan δ peak values as well as SEM analysis. The melt 5

40 processability of the PBS composites with fiber loading up to 50 wt% was assessed by melt flow index (MFI) measurement. Chapter 6 provides manufacture and characterization of biocomposites made from miscanthus fibers and PBAT matrix. In this chapter, PBAT biocomposites were produced at various loadings of miscanthus fibers with and without maleic anhydride grafted PBAT (MAHg-PBAT) compatibilizer. The effect of fiber loading on melt flow and mechanical properties was also studied. In Chapter 7, preparation and performance evaluation of biocomposites consisting of miscanthus fibers in a biodegradable binary blend matrix were conducted. The mechanical performance of the composites was investigated with different weight percentage of fiber loadings. The elastic modulus of the composites was evaluated by parallel, series, Hrisch and Halpin-Tsai theoretical models and the values were compared with the experimental values. Maleic anhydride functionalized PBS/PBAT (MAH-g-PBS/PBAT) blend used as a compatibilizer improved interfacial bonding between the phases in the resulting composites. Thermal, thermo-mechanical, rheological and physical properties were studied for resulting compatibilized and uncompatibilized composites. SEM analysis was employed to demonstrate the interfacial bonding between matrix and fibers in the compatibilized composites. In Chapter 8, effects of processing parameter on the impact performance of biodegradable polymer blend matrix based biocomposites were studied. A biocomposite consisting of miscanthus fibers and a biodegradable PBS/PBAT blend matrix was produced by extrusion and injection molding method. A full factorial experimental design was used to predict the statistically significant variables on the impact strength of the PBS/PBAT biocomposites. The main and interaction effects of the variables were studied using analysis of variance 6

41 (ANOVA) at 95% confidence level. The accuracy of the developed model was examined by residuals plots and coefficients. Among the selected independent processing parameters, the most and least significant processing parameters on the impact strength were fiber length and holding pressure, respectively. Chapter 9 investigated the durability of PBS, PBAT and their blend was assessed by exposure to elevated temperature and humidity. The influence of moisture and temperature on the mechanical performances was examined as a function of exposure time. The mechanical properties of the PBS, PBAT and PBS/PBAT blend were heavily affected after being exposed to 90% humidity and 50 o C. The change in crystallinity of the exposed samples was correlated with observed modulus. At last, Chapter 10 provides an overall conclusion and recommendations for future work in this research. References [1] (Accessed on Jaunary 2015) [2] R. M. Johnson, N. Tucker, S. Barnes, Impact performance of Miscanthus/Novamont Mater- Bi biocomposites, Polymer Testing, 2003, 22 (2): [3] D. Carlson, L. Nie, R. Narayan, P. Dubois, Maleation of polylactide (PLA) by reactive extrusion, Journal of Applied Polymer Science, 1999, 72 (4): [4] L. Petersson, K. Oksman, A. P. Mathew, Using maleic anhydride grafted poly(lactic acid) as a compatibilizer in poly(lactic acid)/layered-silicate nanocomposites, Journal of Applied Polymer Science, 2006, 102 (2): [5] D. Plackett, Maleated Polylactide as an Interfacial Compatibilizer in Biocomposites, Journal of Polymers and the Environment, 2004, 12 (3):

42 [6] H.-S. Kim, B.-H. Lee, S. Lee, H.-J. Kim, J. Dorgan, Enhanced interfacial adhesion, mechanical, and thermal properties of natural flour-filled biodegradable polymer bio-composites, Journal of Thermal Analysis and Calorimetry, 2011, 104 (1): [7] Y. Nabar, J. M. Raquez, P. Dubois, R. Narayan, Production of Starch Foams by Twin-Screw Extrusion: Effect of Maleated Poly(butylene adipate-co-terephthalate) as a Compatibilizer, Biomacromolecules, 2005, 6 (2):

43 Chapter 2: Studies on Mechanical, Thermal and Morphological Characteristics of Biocomposites from Biodegradable Polymer Blends and Natural Fibers* *A version of this chapter has been published in: R. Muthuraj, M. Misra, A. K. Mohanty, Chapter: 5 Studies on mechanical, thermal, and morphological characteristics of biocomposites from biodegradable polymer blends and natural fibers. In: M. Misra, J.K. Pandey, A. Mohanty (Eds.) Biocomposites: Design and Mechanical Performance, Woodhead Publishing, 2015, pp Introduction Bioplastics can contribute significantly to sustainable development in terms of the environment and ecological systems. Over the past two decades, the utilization of biobased and/or biodegradable polymeric materials has gained significant research attention from both academia and industry. Bioplastics are a group of plastics with wide range of properties and applications. According to European Bioplastics, a bioplastic can be biobased, biodegradable or a combination of both biobased and biodegradable. In 2010, the worldwide annual production of plastics was reported to be 265 Mt [1], reached at 299 Mt in 2013 [2]. Meanwhile, bioplastics production was estimated to be 1.6 Mt in 2013 [3]. European bioplastics market report has predicted the annual bioplastic production to be ~6.7 Mt in 2018 [3], which is a 22% hike compared to bioplastics production in Currently, bioplastics contribute 10-15% of the total plastic market share and by 2020 the market share is expected to be around 25-30%, mainly driven by the gradual increase in the number of bioplastics processing companies [4]. Based on overwhelming chemical and material demand, the US Department of energy aims to reach 10% renewable resource based chemical production by 2020 and 50% by 2050 [5]. Main reasons for biopolymers being on prime focus are reduction of CO 2 emissions, reduced waste problems, and establishment of sustainable alternatives to conventional materials. 9

44 Petroleum based non-biodegradable polymers can remain intact in landfills for decades, causing grave environmental concerns. Such non-biodegradable polymers are not suited for applications requiring short lifespan. Incineration of many non-biodegradable plastics waste after their end use can produce harmful emissions. The US Environmental Protection Agency (EPA) has reported that most of the synthetic plastic waste enters landfills, oceans, and lakes [6]. This problem can be overcome by using biodegradable polymers that can offer effective solution to this scenario as they are degradable in the presence of naturally occurring microorganisms. Biodegradable polymers, whether obtained from renewable resources or petroleum resources, are potential substitutes for some non-biodegradable polymers Biodegradable and compostable polymeric materials All biodegradable plastics fall under the bioplastic category and for it to be biodegradable, it should possess the ability to break down into smaller molecules through the action of naturally occurring microbes [7]. In order to assess the biodegradability of polymeric substances, standard experimental procedures are established by national and international organizations. These standards are used to investigate the biodegradability of polymeric materials in specific environmental conditions. The ASTM D6400 standard defines compostable polymeric materials as follows: a plastic that undergoes biological degradation during composting to yield CO 2, H 2 O, inorganic compounds and biomass at a rate consistent with other known compostable materials and leaves no visually distinguishable or toxic residues [7]. This (ASTM D6400) test method is similar to that given in the ISO Detailed biodegradability and compostability mechanisms of polymeric materials have been reviewed by Muniyasamy et al., [8]. In 1980s, synthetic biodegradable polymers were first introduced and were considered potential substitutes for non-biodegradable polymers [9]. Currently, there are many 10

45 biodegradable polymers produced from petrochemical resources and renewable resources [10]. Biodegradability of the polymeric materials generally depends on the molecular formula and does not depend on the monomer origin [11,12]. Therefore, biodegradable polymers can be derived from non-renewable resources and/or renewable resources. We can sort these biodegradable polymers based on the source of the feedstock used to make them (Table 2.1). Table Classification and molecular structure of some biodegradable polymers 2.3. Renewable resource based biodegradable polymers: Some examples Renewable resource based biodegradable polymers have their origin either from plant or biological resources. On the other hand, non-renewable resources based biodegradable polymers are generally made from fossil fuel based monomers. Few examples of renewable resource based biodegradable polymers are discussed below. 11

46 2.3.1 Poly(lactic acid), PLA PLA was first synthesized by DuPont scientist Wallace Hume Carothers in 1931 as a typical linear aliphatic thermoplastic polyester, derived from agricultural feedstock like corn, sugar cane, and sugar beet [13,14]. PLA is synthesized from lactic acid (2-hydroxypropanoic acid) monomers through application of heat under vacuum conditions. PLA can be produced by various methods, like direct condensation polymerization, ring-opening polymerization, and azeotropic dehydration [15]. Low molecular weight PLA is not acceptable for some of the applications due to poor mechanical performance. High molecular weight PLA preparation by direct dehydration condensation is not feasible due to the equilibrium not favoring a high molecular weight polymer. Therefore, Cargill Dow LLC developed a new technology (ringopening polymerization) to produce high molecular weight PLA through an economic route [16,17]. This process involves three separate steps [15,16]. First step involves a polycondensation reaction, which converts lactic acid into PLA oligomers ( Daltons). In the second step, PLA oligomers are catalytically converted into a mixture of lactide stereoisomers. After vacuum distillation of lactide stereoisomers, ring-opening polymerization (ROP) is typically preferred to produce high molecular weight PLA due to its high degree of controllability. One of the biggest advantages of the ROP method is the elimination of water formation step, allowing to attain high molecular weight structures to be attained [18]. This was the first attempt to synthesize renewable resources based economically and commercially viable biodegradable polymer (PLA) by melt stage rather than in solution [19]. Stereochemistry of PLA plays a vital role in the synthesis of stereoisomers such as L (+), D (-), and meso (L, D) lactic acid by using an appropriate microorganism. Four different types of PLA can be produced using different kinds of lactide monomers: L-PLA (PLLA), D-PLA (PDLA), mixtures of L and D, and meso-pla. The L-PLA and D-PLA are optically active, as opposed to mixtures of L and D, and 12

47 meso-pla, which are optically inactive. These steriochemical structures determine the crystallinity of PLA. For instance, PLA with >93% L-lactic acid is semicrystalline in nature whereas PLA containing 50-93% L-lactic acid becomes amorphous in nature [15]. Due to amorphous nature, PDLA does not show crystallization or melting enthalpies under differential scanning calorimetry tests, while PLLA have melt crystallization, cold crystallization, and melting enthalpies due to the formation and melting of crystals. Other parameters affecting PLA properties include polymerization conditions, thermal history, purity, and molecular weight. In terms of tensile strength, tensile modulus (Table 2.2) [20] and gas (O 2 and CO 2 ) barrier properties [21], PLA is more superior than polypropylene, however, toughness and impact strength are lower than commodity polymers [16]. PLA can be processed in conventional processing equipments such as extrusion, injection, stret al.,h blow molding, extrusion blow film, thermoforming, foaming, melt spinning, cast film and sheet [16]. Therefore, PLA has potential applications ranging from food packaging, compostable bags, household items, and textiles to medical supplies. Due to these attractive properties of PLA, the global production of PLA is expected to reach $5.2 billion by 2020 with growth rate of 19.5% during 2013 to 2020 [22]. Unfortunately, PLA has some inherent drawbacks including low heat deflection temperature [23], poor melt strength in comparison with polyolefins [24], slow crystallization rate [19], and too brittle. Moreover, PLA is not able to biodegrade at room temperature i.e., it is not home compostable [23]. These properties prevent PLA from being applied in varied applications. Several attempts have been made to address these issues using plasticization, rubber toughening, blending with other polymers, and reinforcement with natural fibers [25-27]. Recently, the toughens modification of PLA has been reviewed exhaustively by Kfoury et al., [26] 13

48 2.3.2 Microbial polyesters-polyhydroxyalkanoates (PHAs) Microbial polyester synthesis began in the year 1901 and detailed investigations were conducted in 1925 [13]. In the late 1920s, Maurice Lemoigne first discovered polyhydroxyalkanoates (PHAs) using the Bacillus megaterium bacteria [28]. Two different types of bacterial polyesters along with their chemical structure are shown in Table 2.1. PHAs are a family of linear thermoplastic polyesters, which can degrade by the microorganisms [29]. PHAs can be found as homopolymers or copolymers. The performance of PHAs is dependent on their chemical composition. After the 1970s, researcher s curiosity and an impending energy crisis initiated the commercialization of poly(3-hydroxybutyrate), PHB. PHB is the first microbial polyester synthesized through fermentation of sugar and starch with the help of a microorganism. PHB is a saturated linear homopolymer, which behaves like traditional thermoplastics. It can be processed in traditional processing equipments such as extrusion, injection and compression molding. PHB has properties that are very similar to those of a commodity polymer like polypropylene (PP) [30,31]. However, poor thermal stability [29,32], embrittlement due to high crystallinity [13,32] and narrow processing window [33] of PHB hinder its wide range of applications. The brittleness/impact strength of PHB can be modified by copolymerization of hydroxybutyrate (HB) and hydroxyvalerate (HV) [13] or blending with tough polymers [31]. For instance, poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV) is more ductile and flexible compared to PHB [32]. In 1990, poly(3-hydroxybutyrate-co-3-valerate) (PHBV), was introduced into the market by Imperial Chemical Industries Inc. (ICI) [13]. Currently, the major PHAs producers are Tianan (Enmat TM ), Kaneka (Nodax TM ), PHB Industrial (Biocycle ), Mitsubishi Gas Chemicals (Biogreen TM ), Biomer biopolyester, and Metabolix (Mirel ). PHAs have desirable properties like higher melting point (Table 2.2) and higher crystallinity, comparable to commodity polymer (PP), with added advantageous like good biodegradability, compostability 14

49 and biocompatibility. PHAs have potential applications in medical fields, like wound management, vascular system, drug delivery, orthopedic, and urology [34]. However, the thermal degradation temperature range of PHBV is o C, which is very close to their melting temperature. In addition, the main drawback of PHB and low valerate content PHBV is their poor impact strength and low elongation at break caused by low nucleation density. All the above factors combined with high cost hinder the application of PHBV in a wide range of fields [35]. Increasing the valerate unit up to 34 mol% can help in decreasing the brittleness of PHBV [30]. However, increase in the HV content can decrease the melting temperature, tensile strength, modulus of elasticity and glass transition temperature [30]. Most of the commercially produced PHBV contains HV content below 15 mol% [32]. The melting temperature (T m ) of such PHBV is still high (e.g. HV = 11 mol%, T m = 157 C) and thermal degradation may occur during melt processing [30]. In addition, it possesses very low ductility, very similar to PHB. The synthesis of PHBV with higher valerate content requires a lot of energy consumption for production, sterilization, and purification. Therefore, intensive research needs to be directed towards reducing the production cost of PHBV, and in bringing the cost to be on par with traditional plastics like PP [36]. Extensive research has been conducted on polymers from PHA family to overcome the above discussed drawbacks [30,32,33,37]. Different approaches adopted to address the challenges of PHAs are discussed in the following blend and composites sections. In addition to PHBV based polymer blends, natural fibers are potential candidate to reduce the cost of PHBV based materials while enhancing some of their mechanical properties [27]. 15

50 Table Properties of some biodegradable polymers in comparison to non-biodegradable polymers Polymers Density Melting point Tensile strength Tensile modulus Elongation at References (g/cm 3 ) ( o C) (MPa) (MPa) break (%) PLA [36, 38] PLLA [36, 38] PDLA Amorphous [38] PHB [36, 38] PHBV [36, 39] PBS [40] PCL [38, 41] PBAT [39, 42] HDPE [40-41] PP [36] 16

51 2.4 Fossil fuel based biodegradable polymers: Some examples Biodegradability of a polymer is basically based on chemical structure unlike from the source from which it is obtained. Thus, biodegradable polymers besides renewable based sources (as discussed above) can also be made from fossil fuel sources. Few examples of such polymer are reviewed below Poly(butylene succinate), PBS PBS is a promising aliphatic biodegradable thermoplastic polyester (chemical structure is shown in Table 2.1). It is synthesized by condensation polymerization of 1,4-butanediol (BDO) and succinic acid in the presence of a catalyst [40]. Low molecular weight PBS is weak and brittle in nature. Therefore, Showa Highpolymer Co. Ltd in Japan, produced relatively high molecular weight PBS under the trade name Bionolle TM [40]. High molecular weight PBS was achieved through melt condensation polymerization followed by coupling reaction with a chain extender (e.g. hexamethylene diisocynate). The condensation polymerization can be performed using either direct melt polymerization or solution polymerization. Direct melt polycondensation is the most promising method for PBS synthesis as it can produce the high molecular weight PBS suitable for food contact applications [43]. Currently some of the companies producing PBS include Xinfu pharmaceutical (China), Hexing (China), Ire Chemical (Korea). Nippon Shokubai (Japan), Mitsubishi Chemical (Japan), Showa Highpolymer (Japan), and Jinfa Tech (China) [44,45]. Interestingly, PBS can also be produced from either biobased, fossil fuel based monomers or a combination of both bio and fossil fuel based monomers. There are several companies around the world producing pilot scale biobased succinic acid, including BioAmber and Myriant. Mitsubishi Chemical has developed a new technology to produce partially biobased PBS having 17

52 54% biobased content derived from biobased succinic acid [46]. It can be used in applications requiring food contact and applications besides non-food contact uses while meeting global biodegradability standards. Traditionally, 1,4-BDO is produced from fossil fuel resources. Alternatively, the biobased 1,4-BDO is produced through a fermentation process using glucose monomers. Therefore, there is a possibility to produce PBS from 100% renewable resource based monomers [47]. PBS is a biodegradable polyester with balanced thermo-mechanical properties compared to other polymers such as polypropylene, polyethylene, PHB and PLA [48]. Mechanical and thermal properties of PBS depends mainly on crystallinity, crystal structure and molecular weight [49]. Good thermal stability, processing properties, semi crystalline nature and an ideal melting point range, makes PBS a promising biodegradable polymer [50-52]. Notably, the strength of PBS is comparable to those of low-density polyethylene (LDPE) and polypropylene (PP), as shown in Table 2.2. PBS based products are widely used for many applications, such as packaging film, agriculture mulch film, sheets, laminates, bags, hygiene products, non-woven, split yarn, monofilament and multifilament [53]. However, the stiffness, water vapor barrier, melt viscosity and impact strength of PBS are often not sufficient for many applications. These drawbacks can be effectively addressed by blending with other polymers [54], fillers [55,56] and additives [51]. The stiffness of the PBS has been improved by the addition of natural fibers [55] and blending with brittle polymers like PHB [57]. The melt viscosity and gas barrier properties of the PBS were improved with the help of reactive agents [51] and nanofillers [56], respectively. Poly(butylene succinate-co-adipate), PBSA, is a random copolyester, made from 1,4- butanediol, succinic acid, and adipic acid by polycondensation reaction. It has excellent impact strength, elongation and processability. Co-monomer content dictates the physical properties of 18

53 the copolymer of PBS like PBSA. For instance, the adipate content must be lower than 15 mol% in order to maintain the melting point of PBSA above 100 o C [58]. In addition, adipate comonomer content has a greater influence on the biodegradability and mechanical properties of PBSA [40]. PBSA has higher impact strength and elongation as compared to PBS, whereas the tensile strength and melting point of the PBSA are lower than that of PBS Poly(butylene adipate-co-terephthalate), PBAT Polymers that are solely aromatic polyesters such as poly(ethylene terephthalate) (PET) and poly(butylene terephthalate) (PBT) are resistant to microbial attack [59]. However, copolymerization of aliphatic monomers with aromatic monomers results in a biodegradable polymer, like PBAT. PBAT, a biodegradable aliphatic-aromatic copolyester, is derived from fossil fuel based adipic acid, terephthalic acid, and butane diol by a polycondensation reaction [60]. The chemical structure of PBAT is shown in Table 2.1. Due to its mixed chemical structure, PBAT can be enzymatically hydrolyzed in the presence of microorganisms followed by mineralization like aliphatic polyesters [11]. The concentration of aromatic functionality plays a vital role in the performance of the PBAT. PBAT shows higher thermal and mechanical properties with a terephthalic acid concentration above 35 mol% but good biodegradability was achieved with an aromatic moiety concentration lower than 55 mol% [33]. Therefore, PBAT with a range of about 35 to 55 mol% of terephthalic acid can offer an optimal performance [61]. In 1998, PBAT with 22.2 mol% of aromatic moiety was commercialized under the trade name, Ecoflex, by BASF [11]. BASF studied the biodegradability of PBAT under composting conditions. More than 90% of PBAT was observed to be metabolized within three months [62]. Witt and coworkers [11] studied the biodegradability of Ecoflex under composting conditions according to DIN-V standard and concluded that there is no environmental risk after 19

54 composting Ecoflex. Mechanical properties of PBAT are comparable to polyethylene (Table 2.2) [63], but with water vapor permeability of PBAT (240 gm -2 d -1 ) being higher than LDPE (3 gm -2 d -1 ) [62]. In order to overcome this shortcoming, BASF is producing a wide range of biodegradable master-batched Ecoflex (mixed with wax, talc, carbon black, chalk or silica et al.,) to fulfill the requirements set for different applications. Details of the masterbatches are shown in Table 2.3 [64]. Among them, wax based masterbatch film has potential to reduce the water vapor permeability up to 75% compared to virgin PBAT [62]. PBAT possesses excellent toughness, biodegradability and processability. Therefore, PBAT is widely used for compostable organic waste bags, agricultural mulch films, as well as lamination/coatings for starch-based products [65]. Moreover, it can be used to tailor the properties of starch, PLA, and PHAs, thereby opening up new applications for PBAT based materials. Currently, there are few companies producing PBAT, including BASF (Germany), Ire Chemical (Korea), Shandong Fuwin New Material (China), and Xinfu Pharmaceutical (China). Table Ecoflex based masterbatches for different applications [66,67] Masterbatches Filler/additive types Filler content Proposed applications Ecoflex Batch AB1 Ecoflex Batch SL 05 Ecoflex Batch C Black Ecoflex Batch C White Ecoflex Batch SL 2 Chalk 60% Packaging films, compost bags and agricultural films Erucamide ESA (lubricant) 5% Packaging films, compost bags and agricultural films Carbon black 35% Packaging films, compost bags and agricultural films Titanium 60% Packaging films, compost bags and dioxide (TiO 2 ) agricultural films Lubricant (Wax) 5% Packaging films, compost bags and agricultural films 20

55 2.4.3 Poly(caprolactone), PCL PCL is a linear aliphatic biodegradable polyester derived from caprolactone by ringopening polymerization. PCL can also be produced by anionic, cationic, coordination, radical and enzymatic catalyzed polymerization [68,69]. PCL is a typical semi-crystalline synthetic thermoplastic with crystallinity ~50%, melting temperature of o C (depending on the crystallinity), and glass transition temperature of -60 o C [38]. PCL has been most widely used in drug delivery, tissue engineering, and dentistry as it is approved by the US Food and Drug Administration (FDA). PCL is commercially available under different trade names: Capa (Perstorp, UK), Placcel TM (Daicel Chemical Indus, Japan) and Tone (Dow, USA). A wellknown PCL/starch biodegradable blend is commercially available in the market under the brand name Mater-Bi Z grades, and is produced by Novamount, Italy [70]. This blend is mainly used for biodegradable and/or compostable films and sheets. Inherent toughness and biocompatibility of PCL are desired attributes that makes it an ideal choice of blending partner. In this view, PCL has been blended with PLA, PHAs and PBS for different biomedical applications [71-75] Recyclability of biodegradable polymers Recyclability of plastics is a process that extends their service lives before discarding. Plastic waste can be recycled in two different ways: mechanical and chemical recycling. Mechanical recycling involves a polymer being re-melted and re-processed into desired products through different processing techniques [76]. Chemical recycling is a process (depolymerization) that converts polymers into their monomers, which are then used as a feedstock for further polymerization process [9]. Among these recycling methods, mechanical recycling is most favored method for bioplastics recycling. Many studies have been investigated on the mechanical recyclability of biodegradable polymers, like PLA [77-80], PCL [81], PHBV [80, 82], PLA/PHBV blend [80] and PBS [83]. Multiple extrusion and injection molding of the PLA 21

56 (3002D NatureWorks, USA) has been studied by Żenkiewicz [77]. After re-extruding PLA for 10 times, the tensile strength, tensile strain and impact strength were reduced by 5.2, 2.4, and 20.2%, respectively. On the other hand, melt flow rate, oxygen and water vapor permeability rates were significantly increased. After the 10-time extrusion cycle of PLA, the oxygen and water vapor permeability rates were increased by 39 and 18%, respectively. Thermal properties, such as thermal stability, crystallization, and melting point were slightly reduced while there was no change in glass transition temperature. Another study investigated the recyclability of PLLA (L9000, Biomer, Germany) up to seven-injection molding [78]. The authors found that the tensile modulus was not heavily affected, whereas rheological properties, molecular weight, hardness, tensile strength and strain at break were reduced considerably. A recent study also proved that eight times mechanical recycling of PLLA (L9000, Biomer, Germany) leads to considerable reduction in its mechanical properties as well as thermal stability [79]. Meanwhile, the melt flow index (MFI) of the eight times recycled PLLA showed 15 times higher value than unprocessed PLLA. The reduced mechanical properties were attributed to thermo-mechanical degradation of PLLA during melt re-processing. Consequently, the molecular weight, thermal stability and viscosity of the PLLA decreased. Such reduction in properties can be overcome by using thermal stabilizers such as quinine and tropolone [78]. However, the study concluded that PLLA (L9000, Biomer) can be recycled up to five times without a significant change in the tensile properties and flexural modulus [79]. Zaverl and coworkers [82] investigated thermal and mechanical properties of PHBV being recycled for seven times. Mechanical properties such as impact, tensile, and flexural properties were not affected after recycling up to four times; however, there was slight decrease after a fourth cycle. According to this study [82], there was no change observed in the molecular weight of PHBV after two cycles. A considerable 22

57 molecular weight reduction was observed after the third (8.7%), fourth (13.5%) and fifth cycles (16.6%). Even after fifth round of recycling, melting temperature and thermal stability of the PHBV were not affected significantly. The chemical structure of the PHBV was not changed during melt reprocessing, which was confirmed by Fourier transform infrared (FTIR) spectroscopy. On the contrary, PHBV is more susceptible to thermo-mechanical degradation compared to PLA [80]. Therefore, blending of PHBV with PLA can minimize the degradation of PHBV during recycling. Recyclability of PCL has been investigated by Moraczewski [81]. After eight cycles, there were no changes observed in tensile strength and elongation at break. However, MFI, thermal stability, and Charpy impact strength of PCL were dependent on the number of cycles. This study suggests that the recycled PCL can be blended with virgin PCL. According to Kanemura et al., [83], PBS (Bionolle 1020) can be recycled without affecting its bending strength and modulus when reprocessed up to three times at 140 C using a compression molding method. In addition, the molecular weight of the PBS was not heavily affected after three times of reprocessing at 140 o C. Therefore, this study concluded that PBS is a suitable candidate for material recycling Durability of biodegradable polymers Generally, durability is very important for polymers as it increases their suitability for a wide range of applications. The durability of polymeric materials can be investigated by exposing to simulated environmental variables such as temperature, and humidity. Only a few studies have examined the durability of biodegradable polymeric materials [84-88]. These literature sources report that biodegradable polymers are very sensitive to elevated temperature and humidity. Recently, Harries and Lee [86,87] assessed the durability of PLA at elevated temperature and humidity with respect to time. Their experimental findings suggest that after 28 23

58 days of exposure at 70 o C temperature with 90% relative humidity, PLA materials experienced severe loss in mechanical properties. Therefore, commercial PLA cannot fulfill long-term durability requirements set by the automotive industry. This drawback can be addressed either by reducing the hydrolysis reaction through addition of scavengers/sacrificial compounds or by blending/alloying with other durable polymers [86]. The degradation of PHBV has been examined in distilled water and marine environments at different temperatures [89,90]. The mechanical property performance of the PHBV gradually deteriorated with increasing immersion time. This effect is attributed to the hydrolytic degradation of ester bonds in the PHBV backbone. Usually, the durability of the polyesters depends on the chemical structure. Among PLA, PBS, PBAT, and PBSA, PLA and PBAT are more durable compared to PBS and PBSA [84]. However, the durability of PBS and its composites has been noticed to improve with the addition of trimethylolpropane triacrylate (TMPTA) as an anti-hydrolysis agent [84]. The durability of PBS, PBAT and their blend has been investigated by exposure to 50 o C and 90% relative humidity for a duration of up to 30 days [91]. This study concluded that the biodegradable polymers (PBS, PBAT and PBS/PBAT) could readily undergo hydrolytic degradation in the presence of elevated temperature and humidity. The durability of biodegradable polymers need to be improved under high humidity and temperature for expanding their applications Polymer blends: Some examples Neat polymers for commercial applications usually cannot fulfill all product requirements. One of the alternatives for modifying the properties of polymers are to blend with other polymers (one or more), which have complementary properties. Blending two or more structurally different polymers might improve or tailor some of the properties and its cost 24

59 performance, making a blend that is able to meet specific end-use requirements [26]. Furthermore, polymer blending can be performed in standard processing equipment such as twin-screw extruders. Generally, polymer blends possess unique physico-mechanical properties, which are usually not present in their individual components. Polymer blends have a wide range of applications in fields such as automotive, household items, electrical and electronics. Some of the commercially available biodegradable polymer blends are listed in Table 2.4. A blend of polylactic acid/lignin/fatty acid/wax shows mechanical properties similar to acrylonitrile butadiene styrene (ABS). It has strong thermal insulation properties as well as excellent chemical resistance against polar substances [92]. The properties of STARCLA (20 µm film) are comparable to polyethylene [93]. BASF is producing biodegradable PBAT/PLA (55/45 wt%) blend under the trade name of Ecovio. The mechanical performance of Ecovio is similar to the properties of high-density polyethylene [62]. Table List of companies engaged in the production of some biodegradable polymer blends (the table was modified after referene [9, 94,95]). Blends Applications Trade name Manufacturer PLA/PBAT Blown films Ecovio BASF, Germany PLA/copolymer Injection molding: Cereplast Cereplast Inc., USA housewares, tackle sustainable PCL based blend Films and injection molded Mater-Bi Novamount, Italy items PLA/co-polyester Blown film and coextrusion Bio-Flex FKuR, Germany blend PLA based blend Extrusion and blown film Starcla TM Showa Denko Europe Biodegradable copolyester Blown film and injection Terraloy TM Teknor Apex blend molded parts PLA/PBS or PBSA Food service ware Ingeo TM AW NatureWorks, USA 240D PBAT/PHA Fiber and non-woven BioTuf TM 970 Heritage Plastics Inc. USA 25

60 As discussed previously, high cost and inferior mechanical properties relative to several commodity plastics are the downside of most of the biodegradable polymers and therefore wide scale adaptability of these polymers have been significantly hampered. In order to effectively overcome these drawbacks, melt blending is a fast, economical and a convenient approach compared to developing a novel polymeric material through synthetic polymerization technique [96]. For example, slow crystallization behavior and poor impact resistance of PLA hinders its wide range of industrial applications. Researchers have found that crystallization and cold crystallization were enhanced by incorporating PBS into PLA systems [97]. Yokohar and Yamaguchi [98] have conducted studies on the structure-properties correlation of PLA/PBS blend. They found that the cold-crystallization temperature of PLA was reduced in the blend compared to neat PLA. This result indicated that PBS acted as a nucleating agent in the PLA matrix. Blends of biodegradable polymers are being used for short-term applications, such as absorbable medical implantations, compostable food packaging containers, and agricultural mulch films. Kim and coworkers [85, 88] have reported the durability of talc filled biodegradable PBS/PBAT (55/45 wt%) blends under marine environments. Better elastic properties were observed in biodegradable materials compared to that of the recycled PE Miscible biodegradable polymer blends At thermodynamic equilibrium, mixtures of polymers that exist in a single phase at the molecular scale are referred to as miscible (soluble) blends [99]. Miscibility of polymers in the blend system can be detected through glass transition temperature (T g ) and phase morphology. Miscible polymer blends show single glass transition temperature. In most cases, the miscibility of two polymers depends on the molecular weight, chemical structure of the polymer, and crystallinity [100]. In the polymer blend system, some of the properties are determined by 26

61 miscibility, interaction and phase morphology. For example, the mechanical properties of the miscible blends follow the rule of mixture [101]. A blend is considered partially miscible if there exists phase separation but each polymer rich phase contains a sufficient amount of the other polymer to alter the properties of that phase (e.g., the glass transition temperature) [101]. Normally, polymers are miscible with each other up to a certain limit. The Flory-Huggins theory is most frequently used to calculate mutual solubility of polymer blends using an interaction parameter. The Flory-Huggins interaction parameter (χ) can be obtained by the following equation 1.1 [102]: = ( ) (1.1) The term V r is reference volume, T is absolute temperature, δ 1 and δ 2 are the solubility parameters of the components and R is the gas constant. The solubility parameters can be calculated using group contribution methods [103]. If the interaction parameter (χ) value is less than 0.5, it can be considered as miscible blend [104]. The Flory-Huggins theory always yields negative interaction parameter values for miscible blends. For instance, a PHB/ poly(epichlorohydrin-co-ethylene oxide), PEEO blend has an interaction parameter value of [105] while a PLA/PBS blend has an interaction parameter value of [106]. Another study showed limited miscibility of PHB with 20% PBS blend [107]. This miscibility was evidenced by shift in glass transition temperature and depression of the PHB melting temperature. Bhatia et al., [108] found that a PLA/PBS blend system has partial miscibility when the PBS composition was lower than 20 wt%. This result has been confirmed by rheological properties. However, tensile strength and modulus of the PLA/PBS blends are lower compared to neat PLA. The brittleness of the PLA is reduced with the addition of PBS, thus making it a 27

62 suitable material for packaging applications [108]. Park and Im [106] studied blends of two typical biopolyesters (PLA and PBS) by melt blending through an extruder. In this polymer blend system, a single glass transition temperature was observed, which was attributed to enhanced miscibility, between the polymer phases. Spherulites morphology of the blends was observed through optical polarizing microscopy and they concluded that crystallization induced phase separation occurred when the PBS volume increased by more than 40 wt% in the blend system. Moreover, PLA/poly(ethylene glycol) [ ], PLA/poly(ethylene oxide) [104, 111], and PHB/poly(ethylene oxide) [112] blends were able to form miscible blend system Immiscible biodegradable polymer blends If mixtures of two polymers exhibit separate phase morphology (i.e., matrix-droplet and co-continuous morphology), they are referred to as immiscible blends [113], as shown in Figure 2.1a. In the binary blends, the matrix-droplet phase morphology (Figure. 2.1b) changes to cocontinuous morphology (Figure. 2.1c) with an increase in the volume fraction of the dispersed phase. Immiscible polymer blends exhibits different glass transition temperatures characteristic of their pure blend components and core-shell morphology. The core-shell morphology of the immiscible blends depends on the viscosity of the individual blend components, processing aspects, composition, and compatibility between the components. Inherently some biodegradable polymers (PBS, PBSA, PBAT, certain PHAs and PCL) have good toughness and elongation. Therefore, they are considered as a promising candidate for modifying brittle biodegradable polymers while preserving the biodegradability. However, a large amount of literature shows that most of the polymer blends are immiscible due to poor dispersion of the inclusion phase in the continuous phase, weak interfacial adhesion, and instability between the components [26]. For instance, biodegradable polymer blends of PHBV/PCL 28

63 [114], PHBV/PBAT [115], PHBV/PBS [116], PHBV/PLA [117], PLA/PBS [98], PLA/PBAT [118,119], PLLA/PCL [120], PLA/PHB [121], PCL/PBS [122], and PBS/PBAT [54] are all immiscible blends. In order to improve the interaction between the pure polymers in the blend system, compatibilization has proven to be an effective strategy for the immiscible polymer blend systems. Jiang et al., [119] studied the toughening of PLA through melt blending with PBAT. Dynamic mechanical analysis (DMA) results showed two independent glass transition temperature (T g ) peaks corresponding to their individual components, suggesting that the PLA/PBAT blends were immiscible. Mechanical properties of the PLA blends showed the greatest improvement in elongation and toughness with incorporation of PBAT from 5 to 20 wt%. These improvements were correlated with de-bonding-induced by shear yield and weak interfacial adhesion between the components in the blends. Wu et al., [123] studied the PBS/PLA blend morphology, rheological properties, and interfacial tension between the polymers. They observed that PBS/PLA blends were thermodynamically not miscible and possessed co-continuous phase morphology in the 50/50 wt% blend. DSC analysis reveled that PBS has no significant effect on melting point and crystallization behavior of PLA, but on the other hand, PLA had little effect on melting point and cold crystallization of PBS. 29

64 Figure 2.1. (a) Schematic representation of the evolution of morphology in a binary immiscible blend, (b) matrix/dispersed morphology, and (c) co-continuous morphology (adapted with kind permission from Ravati and Favis, Polymer, 2010, 51: , Copyright 2015, Elesevier, Licence number [113]). Furthermore, Zhou et al., [124] studied PBS/PLA blends for biomedical applications. They also made similar observations i.e., PBS/PLA blends were immiscible and their mechanical properties followed the rule of mixtures. The hydrolysis behavior of PBS, PLA and their blends was investigated under simulated body fluid for 16 months at 37 o C. The PBS/PLA blends lost their tensile properties earlier than the neat polymers due to faster hydrolysis reaction facilitated by the interface present between the blend components. Throughout the hydrolysis time, the molecular weight reduction of the blends was faster than that of parent polymers. A blend of PHBV/PCL has been studied by Jenkins et al., [125] and they suggested that the blend is not miscible.melt-blended immiscible PHBV/PLA blend was studied by Nanda et al., [117]. The mechanical performances of the blends were dependent on processing parameters and composition of the constituents. The 30

65 elongation at break observed in PHBV/PLA blend [117] increased slightly compared to neat PLA and PHBV. Similar trends were observed in compostable PLA/PHBV blend by Ma et al., [126]. It was demonstrated that matrix yielding and interfacial debonding were responsible for the increased toughness in PLA/PHBV blends. According to Qiu et al., [116], a PHBV/PBS blend is not miscible due to the independent T g of two phases. Some of the immiscible blends are considered compatible blends when they show fine phase morphology and satisfactory performance in their mechanical properties [96]. These types of blends show their characteristic T g but most often the T g values are shifted towards other components in the blends indicating improved compatibility. The performance of the compatible, immiscible polymer blends depends on many factors including their interaction, component properties, composition, and compatibility between the polymers phases. Among all the parameters, interaction is a factor of greater importance because it determines the thickness of the interface between the components. John et al., [122] have investigated the compatibility and miscibility of PBS, PBAT, and PCL blends. From DSC analysis, the blends of PCL/PBAT and PCL/PBS are seen to have two melting points, which correspond to their individual melting points. This result clearly indicates that these blends are not completely miscible. However, from DMA analysis, a single T g was observed for the PBS/PBAT blend. This is because the T g values of both the neat PBS and PBAT are very close to each other. Moreover, these blends showed an interesting behavior when tested for tensile strength. The tensile strength of PBS/PBAT (70/30 wt%) and PCL/PBS (30/70 wt%) blends was higher than that of neat PBS, PBAT and PCL. This improvement was attributed to the interaction between the blended components as well as a change in crystallinity. Recently, Muthuraj et al., [54] have reported a similar type of synergistic effect in the PBS/PBAT blends. 31

66 2.7.3 Compatibilization of polymer blends Compatibilization is an interfacial phenomenon in the heterogeneous polymer blends. This interfacial activity can be modified by reactive and non-reactive processing strategies. Compatibilization is a process by which the blend properties are enhanced while increasing adhesion between the phases, reducing the interfacial tension, and stabilizing morphology [127]. To achieve these goals, there are several strategies to the method of compatibilization. The degree of compatibility between the components in the blends can be enhanced through the addition of compatibilizers [60-62]. The resulting mechanical properties of the immiscible, compatible blend will have a balance of their parent polymer properties or show a synergistic improvement [99]. Many commercially available polymer blends on the market are mostly compatibilized using a compatibilizer [1]. Polymers with polar groups can interact between the components in the blend through non-bonded interactions such as dipole-dipole interactions and hydrogen bonding [1]. Therefore, biopolymers have more mutual miscibility than polyolefins. Compatibilization of polymer blends was reviewed in detail by Koning et al., [96], and Imre and Pukanszky [1]. According to these sources, the process of adding a premade compatibilizer (block-copolymers) that has a strong affinity towards the blend components is referred to as physical compatibilization. Introducing a compatibilizer that can promote chemical reaction and/or specific interaction between the blend components is referred to as reactive or in-situ compatibilization. Varieties of reactive compatibilizers have been identified in the earlier publications [ ]. Next section of this chapter discusses these compatibilization methods. Generally, premade graft or block copolymers are mostly used to modify the heterogeneous polymer blends interface by a non-reactive compatibilization method. In order to make effective compatibilization of the blends by block-copolymers, the copolymer must have a 32

67 maximum solubility with components in the blend. This leads to strong interfacial adhesion while lowering the interfacial tension between the components. The molecular weight and concentration of compatibilizer (block copolymer) should be slightly higher than that of critical entanglement and critical micelle concentration, respectively [137]. In addition, compatibilized blends exhibit dispersed particle size at a sub-micron level which prevents coalescence during subsequent processing [96]. Non-reactive biopolymer blends with block-copolymers have been widely studied in the literature [ ]. However, non-reactive compatibilization is less efficient than reactive compatibilization [96]. Reactive extrusion generates reactive sites, which interact with the other polymer component and localize themselves at interface between the components in the blends. Such localization at an interface enhances the mechanical properties and decreases the domain size of the inclusion phase. Reactive extrusion of the polymer blends can be performed by one step (in-situ) or two-step extrusion. All of the components are introduced simultaneously in the one-step reactive extrusion [141]. In the two-step reactive extrusion, the first step involves functionalization of polymer with reactive agents followed by the second step in which the functionalized polymers are blended with other components through extrusion [142] Non-reactive compatibilization of biodegradable polymer blends Historically, immiscible polymer blends can be compatibilized through the addition of a graft or block copolymers [127]. The block-copolymer acts as a compatibilizer between the components in the immiscible polymer blend systems. As a consequence, the block copolymer can reduce the interfacial tension between the components [142], as shown in Figure

68 Figure 2.2. Adding a compatibilizing agent, such as a diblock copolymer, to a polymer blend can improve its stability, but is more likely to result in a dispersed morphology rather than a cocontinuous morphology. a) A two-dimensional slice of a compatibilized blend with dispersed phase morphology, represented by minority dark blue phase and a majority turquoise phase. b) A molecular schematic showing how the diblock copolymers are segregated at the interface between the two phases (adapted with kind permission from Ryan, Nature Materials, 2002, 1: Copyright 2015, nature publishing group, License number [142]). There are many commodity polymer blends compatibilized with a block-copolymer. For example, styrene acrylonitrile, SAN, and styrene butadiene styrene, SBS blends, have been compatibilized by the addition of a diblock-copolymer [143]. These compatibilized blends exhibit good mechanical properties compared to uncompatibilized blends. A similar type of compatibilization effect has been reported for polystyrene/low and high density PE blends via addition of butadiene-styrene block-copolymers [144]. In this blend system, the compatibility between the components has been found to be dependent on processing parameters. Recently, similar attempts have been made for bioplastic blends [120, 138, 140, ]. LDPE-PLA block copolymers have been used as a compatibilizer in the PLA/LDPE blends in order to improve the properties of resulting blends [149]. These blends showed considerable increase in deformability with tensile strength lower than that of neat PLA. Another study also showed a remarkable increase in ductility with the addition of a block-copolymer, PHB-block-poly(methyl methacrylate), PMMA in the PHB/PMMA blend [148]. PLLA/PCL [120] and PBS/PCL [150] 34

69 blends were compatibilized with triblock copolymer, (polyethylene oxide polypropylene oxide polyethylene oxide). These two studies concluded that the blends with compatibilizer achieved better compatibility and toughness compared to those of the uncompatibilized blends. The enhanced compatibility between the components was further confirmed by a T g shift and decrease in PCL domain size in the matrix Reactive compatibilization of biodegradable polymer blends: Few specific examples Poor compatibility between the phases is a major drawback in multi phase immiscible polymer blends. Immiscible polymer blends can be compatibilized by introducing a compatibilizer. Extensive work has been conducted on improving the compatibility between the phases in heterogeneous biodegradable polymer blend systems Reactive compatibilization of PLA/PBAT blends: Recently, significant research has been directed towards producing compatible PLA/PBAT blend through reactive processing. For instance, Sirisinha and Somboon [128] investigated the reactive PLA/PBAT blends by adding 0 to 0.5 wt% of dicumyl peroxide (DCP) for blown-film processing. This study was completed using fixed PLA/PBAT ratio (70/30 wt%) and varied the DCP loading from 0 to 0.5 wt%. Rheological, melt flow index, and melt strength properties were measured to investigate the synergistic effect of DCP addition into the blend system. The addition of DCP into a PLA/PBAT blend was found to increase the elasticity and viscosity. In addition, good melt strength of the blend allowed for easy processing when blowing the plastic into a film. According to the literature [129], solubility parameter values of PLA and PBAT are 10.1 and (cal/cm 3 ) 1/2, respectively. The difference in solubility parameter leads to the formation of immiscible blends as well as a reduction in their mechanical performance. These drawbacks are addressed by using glycidyl methacrylate (GMA) as a reactive compatibilizer in melt blended 35

70 PLA/PBAT. Kumar and coworkers [129] studied the effect of mechanical, thermal as well as morphological properties of PLA/PBAT blends with GMA. They incorporated GMA (3 and 5 wt%) into the optimum PLA/PBAT blend system in order to improve compatibility between the components. They found that the impact strength of the PLA/PBAT/GMA (70/25/5) blends increased (72%) as compared to virgin PLA. The increment in properties was attributed to the formation of ethylene acrylic ester at the interface with the addition of GMA. Morphological examinations confirmed that compatibility between the PLA and PBAT phases was in fact improved in the presence of GMA. Thermo-mechanical analysis (DMA) further corroborated the improvement noticed in compatibility between the PLA and PBAT and heterogeneous phase morphology of the PLA/PBAT blends Furthermore, with the addition of GMA within PLA/PBAT blends, reduction in crystallinity and inward shift in melting point were observed by DSC analysis, supporting the enhanced compatibility between the two polymers. Zhang et al., [130] investigated reactive blending of PLA/PBAT with a reactive processing agent, a random terpolymer of ethylene, acrylic ester and glycidyl methacrylate (abbreviated as T-GMA in their work). They prepared reactive blends of PLA/PBAT with T- GMA at a varying concentration of 1 to 10 wt% by a melt blending technique. They found that the tensile toughness and impact strength of the blends were improved with 1 wt% of T-GMA incorporation, but no changes occurred with increasing T-GMA concentrations. The toughness and impact strength improvement were attributed to increased miscibility between PLA and PBAT phase, which was confirmed by SEM analysis. From the DSC analysis, they found two glass transition temperatures from the blends, indicating that the blends were not homogenous after adding T-GMA. During melt blending, the PLA/PBAT blend can be compatibilized by a transesterification product with the help of terbutyl titanate (TBT) [131]. The mechanical and 36

71 thermo-mechanical properties of this compatibilized blend were improved significantly compared to the non-compatibilized PLA/PBAT blend. Best elongation at break (298%), impact strength (9 kj/m 2 ), and tensile strength (45 MPa) of PLA/PBAT blend were obtained with 0.5% TBT concentration. The compatibility between PLA and PBAT has been improved through transesterification reaction products, which was demonstrated by SEM micrographs Reactive compatibilization of PLA/PBS blends: PLA/PBS blends have been compatibilized using lysine triisocynate (LTI) as a reactive processing agent [135]. Significantly higher level of improvement was observed in the impact strength (2-4 fold increase compared to PLA), and elongation at break with a decrease in melt flow rate. This improvement could possibly be due to the grafting or crosslinking of LTI with PBS or PLA in the blends. However, this reactive compatibilization was detrimental to tensile strength of PLA/PBS blends. A similar trend was observed in the PLLA/PBS blend with a dicumyl peroxide (DCP) reactive agent [51]. These blends achieved stronger interfacial adhesion and fine dispersed phase morphology owing to the increased compatibility between the blended components. Consequently, the PLLA/PBS blend with 0.1 parts per hundred (phr) of DCP showed 12-fold increment in the Izod impact strength compared to virgin PLLA. This improvement was attributed to debonding through shear yielding. However, both flexural strength and flexural modulus continuously decreased with increasing DCP concentration. Yet, DCP is not an appropriate choice when trying to improve elongation at break of PLLA/PBS blends. Unlike PLLA/PBS blends, PLA/PBAT blends with phr DCP showed significant improvements in elongation at break as well as increased impact toughness [141]. Chen et al., [151] investigated the compatibility between the PLA and PBS blends with twice functionalized nanoclay (TFC). SEM revealed that the PBS domain size gradually decreased when 2 wt% TFC was incorporated into PLA/PBS blends, due to increased 37

72 compatibility between the polymer phase in the blend system. Elongation at break was increased with the addition of TFC into PLA/PBS blends. Flory-Huggins theory suggested that TFC has a high compatibility with the PLLA phase when compared to the PBS phase. In addition, there was no improvement in the elongation at break when nanoclay without functionalization was incorporated, which indicates that no compatibility existed between the PLA/PBS blends. Ojijo et al.,[152] have improved the toughness of PLA/PBSA blends by in-situ compatibilization with triphenyl phosphite (TPP). Comparing the impact strength and elongation at break for neat PLA and PLA/PBSA blend containing 10% PBSA and 2% TPP, the impact strength and elongation at break of compatibilized PLA/PBSA blend showed 166% and 516% higher value compared to neat PLA. These improvements were attributed to shear yielding of the matrix, which was initiated by the debonding between the blended components. In addition, there was no significant change in the tensile strength and thermal stability of the compatibilized blends. The authors also suggested that the barrier properties, thermal properties and mechanical properties of this toughened PLA/PBSA blend system could be improved by the addition of clay into such a system Reactive compatibilization of PLA/PHB and PHBV/PBS blends: Reactive compatibilized PLA/20 wt% PHB (Mirel TM M4300) copolymers showed 1.3 J tensile toughness and 200% elongation at break [121]. These values are significantly higher than neat PLA and non-reactive blends of PLA/20 wt% PHB. It is important to note that this amorphous PHB is thermodynamically immiscible with PLA. However, reactively compatibilized PLA/PHB blend shown that the notched Izod impact strength of immiscible PLA/20 wt% PHB copolymer blend is 20 times greater than that of virgin PLA (from 0.4 ft.lb/inch to ~8.2 ft.lb/inch) [121]. The same 38

73 blend displayed a tensile modulus of ~2 GPa and tensile strength of ~50 MPa. Therefore, the stiffness-toughens performance of this blend is superior compared to PBT and polypropylene. PBS is another biodegradable thermoplastic polyester with tensile strength comparable to PP and stiffness similar to LDPE with fair processability using processing equipments of conventional polyolefin thermoplastics [153]. PBS has a glass transition temperature of around - 30 o C, which gives it a high toughness compared to the polymer at room temperature. Also, the high elongation of PBS (more than 200%), makes it an ideal candidate to blend with a brittle biopolymer [116]. Therefore, PBS seems to be one of the biodegradable polymers that can be effective in improving the toughness of PHBV. Ma et al., [57] investigated the toughening mechanism of PHB/PBS blend and PHBV/PBS blend by using an in-situ compatibilizer DCP through melt blending. They found that the PBS domain size was considerably reduced by incorporating DCP into those two blends. The domain size reduction was directly related to the interfacial adhesion between the components in the blend systems. During melt blending, addition of DCP into the blends could generate a grafting reaction between the components (PHBV, PHB, and PBS) in the blends such as PHBV-g-PBS and PHB-g-PBS. The PHBV-g-PBS and PHB-g-PBS acted as a compatibilizer in the blend system which in turn improved the interfacial adhesion between the polymer phases. Consequently, the mechanical properties of the PHB/PBS and PHBV/PBS blends were considerably increased. For example, after incorporation of DCP (0.5 wt%), the elongation at break of PHBV/PBS (80/20) blend was increased by 39- fold, in comparison to the corresponding uncompatibilized blend. In addition, a significant improvement was observed in un-notched Izod impact toughness of the PHB/PBS blends with 0.5 wt% DCP. Similarly, the poor compatibility between the PHB and PLA has been improved by using DCP as reactive agent [154]. The strength and impact toughness of the compatibilized 39

74 blends were improved in comparison to uncompatibilized PHB/PLA blends. The improved compatibility was attributed to the formation of crosslink network and/or copolymer (PHB-g- PLA) at the interfaces. Consequently, the the inclusion phase (PLA) domain size was observed to have reduced significantly Reactive compatibilization of PLA/PCL blends: Shin and Han [132] showed melt blending of PLA/PCL resulted in immiscible blend without compatibilizer and that the compatibility was improved by adding GMA into the blend. The improved compatibility has a direct relationship with mechanical properties and fine morphology. In addition, they have observed that GMA acted as a reactive monomer between the PLA/PCL blend interfaces. Liu et al., [133] have studied super toughened ternary blends with PLA/ethyl-butyl acrylate and glycidyl methacrylate copolymer (EBA-GMA)/ethylene methacrylic acid zinc ionomer (EMAA- Zn). This reactive blend exhibited remarkable improvements in notched Izod impact strength (from 38 to 770 J/m) and elongation at break (from 5 to 229%). These improvements were attributed to the compatibilization that occurred between the components in the resulting blends. In another study, reactive blending of PLA and ethylene-glycidyl methacrylate (EGMA) showed a dramatic improvement in impact strength (70 kj/m 2 ) and elongation at break (>200%) compared to virgin PLA [134]. Zhang et al., [155] have studied a super-toughened multiphase PLA blends with series of renewable poly(ether-b-amide) (PEBA) copolymer and ethylene methyl acrylate-glycidyl methacrylate (EMA-GMA) terpolymer. This reactive multiphase blend (PLA/EMA-GMA/PEBA (70/20/10 wt%) system showed a massive improvement in the impact toughness (~500 J/m) and elongation at break (~75%). The authors concluded that the improvements are attributed to the combination of strong interfacial adhesion between the phases and massive shear yielding phenomena. Both the miscibility and impact energy of the PLA/PCL 40

75 blends were dramatically improved with the addition of lysine triisocynate (LTI) [136]. This was due to reduction in size of the PCL phase in the PLA matrix; this behavior reduced the localized stress concentration triggered by interfacial failure. A blend of PLA/PCL (85/15) was melt processed with and without LTI (1 wt%) as a reactive additive [136]. The average fracture energy of PLA/PCL/LTI was measured to be five times higher than that of PLA/PCL blend. Similar to other studies, addition of LTI was found to increase the miscibility of PLA with PCL. The improved compatibility between the PLA and PCL contributed to the reduction of PCL domain size, which facilitated increased energy absorption during the fracture process. Wang et al., [156] have improved the toughness of PLA/PCL blends by in-situ compatibilization with triphenyl phosphite (TPP). The percentage elongation of the reactively compatibilized PLA/PCL/TPP (80/20/2 or 20/80/2 by weight) blends improved considerably when compared to neat PLA and corresponding uncompatibilized blend. This result suggested that a synergism existed for certain compositions of PLA/PCL blends. Moreover, the compatibilized blends showed higher susceptibility to enzymatic degradation when compared to neat PLA, PCL and PLA/PCL blend without reactive agent. 2.8 Natural fibers Classification of natural fibers Natural fibers are are classified based on their origin as it can be obtained from either plant or animal sources. The major component of animal fibers (e.g., wool, feather, angora fiber, and silk fiber) is protein, whereas the major component of plant fibers is cellulose [157]. Plant fibers (cellulosic/lignocellulosic fibers) can be classified based on their origin: seed/fruit fibers (e.g., cotton, coir, et al.,), leaf fibers (e.g., sisal, pineapple), bast fibers (e.g., jute, keneaf, flax et al.,), agriculture residues (e.g., corn straw, wheat straw, corn stover, et al.,) and grass/reed fibers 41

76 (switch grass, miscanthus, bamboo, et al.,). Plant fibers (excluding cotton) are mainly composed of water-soluble organic components, lignin, waxes, cellulose, hemicellulose, and pectin. Based on the type of the biomass, the chemical compositions of fibers can vary widely [158] Natural fibers: nature and behavior The performance of the plant fiber depends on its internal structure, architecture, chemical composition, crystallinity and micro-fibril orientation angle [159] (Table 2.5). Among the natural fiber constituents, cellulose has higher crystallinity. In contrast, lignin is amorphous in nature with a highly complex aromatic structure. Because of its stiffness, lignin acts as a protective barrier for cellulose in the plant. Hemicellulose is a highly branched polymer with a degree of polymerization times lower than cellulose [160]. Pectin is usually found in primary cell wall of most plant fibers. Pectin is more hydrophilic component in plant fiber due to presence of carboxylic acid group. Generally, native cellulose has strength > 2GPa and a stiffness of 138 GPa [161]. The stiffness of natural fibers depends on the microfibril angle. Therefore, plant fibers with high cellulose content (~60-80%) and low microfibril angles have a strong reinforcing effect in composites. Both pectin and hemicelluloses play important roles in the water absorption, swelling, elasticity, wet strength and fiber bundle integration of the fibers [41] Advantages and challenges in using natural fibers Unlike synthetic fibers, natural fibers are lightweight and inexpensive. They possess high toughness, reduced tool wear, good insulation properties, reduced human health hazard, enhanced energy recovery, recyclability, and biodegradability [153]. The specific modulus and strength of the natural fibers are comparable to glass fibers [46, 160, 162], as shown in Table 2.5. Furthermore, the natural abundance, renewability, CO 2 neutrality and lower density of natural 42

77 fibers have caused an increasing demand for these fibers in material applications. However, there are some major drawbacks with respect to plant fibers, including complex supply chain, geographical availability, moisture absorption, and inconsistency. The inconsistency of the plant fibers is related to many factors, such as harvesting time, species type, soil quality, fertilization, moisture content of the fibers, location of the plant growth, as well as climatic conditions during plant growth [157]. Moisture absorption of the natural fiber can vary greatly depending on the fiber types. This moisture absorption creates performance problems for products made out of such natural fiber composites. Due to low thermal stability, natural fibers are recommended for process/use below 200 o C [36, 157]. These issues can be effectively addressed by appropriate material selection/screening techniques [162, 163] Biocomposites The term composites originated from the Latin word, compositus. In general, composites are materials made from two or more distinct constituent having significantly different physical and/or chemical properties. At least two components are required in order to prepare composites material: matrix or continuous phase and reinforcement or discontinuous phase. The continuous phase acts as a binder while the discontinuous phase acts as reinforcement. Often both continuous and discontinuous phases are easily distinguishable in the resulting composite materials. Generally, the mechanical performances of the reinforcements are higher than those of the matrix phases. Composite materials properties can be tailored to suit specific application requirements. Composites are generally classified based on their matrix sources like metal, ceramic and polymer (thermoplastic and thermoset). Based on their reinforcements, the composites can be further categorized as fibre (short or long) composites and particulate (powder, flacks and spherical) composites, fabric based composites and long fiber reinforced 43

78 thermoplastic composites. Most common fibers, such as aramid, carbon and glass are used as a reinforcement in conventional polymer matrices, such as PP, epoxy resins, unsaturated polyesters and polyurethane [164], while E-glass fibers dominate in the fiber reinforced plastic composites (FRCs) market [161]. Biocomposites offer a combination of high mechanical properties, lightweight, moldability and design flexibility. Thermoplastic polymer matrices usually dominate the class of natural fiber composites. FRCs have diversified applications in automotive interior parts, packaging, construction, furniture, consumer goods, electrical, musical instruments, wind turbine rotor blade and decking industries [161]. A combination of flax and carbon fiber reinforced composites are used in sporting equipment (snowboards, tennis racket and bicycle frames) because of its superior vibration damping properties and mechanical performance [157]. In 2002, fiber reinforced composites share of the global market was 771 million kg and this has continued to grow extensively [41]. The amount increased to 8.7 million tons in 2011 [161]. In 2002, fiber composites had a maximum share of 31% in automotive applications followed by construction applications (26%) [10]. In Europe, this trend increased significantly because of mounting environmental concerns. Increasing interest in biodegradable and/or biologically derived materials has fueled vivid research in biocomposites [46]. Either a composite produced from the combination of non-biodegradable polymer with a natural fiber/filler or a bioplastics with synthetic fibers/fillers can be called as biocomposites Advantageous of natural fiber composites Natural fiber composites (NFCs) have numerous benefits compared to synthetic fibers such as abundant availability, lightweight, eco-friendly, nonabrasive, inexpensive, good fatigue and specific strength [165]. Biocomposites made from plant based fiber and renewable resource 44

79 based polymers can be termed eco-composites or all green composites [46]. Any biocomposites that are compostable, recyclable, environmentally friendly and commercially adequate can help in waste management [162]. Biocomposites are in huge demand for automotive uses, as they reduce both manufacturing costs and fuel consumption while meeting the consumer pull for greener products [36]. Because of global warming, environmental concerns, and an increased interest in greener products, biocomposites based on bioplastics have been developed Attributes of natural fiber composites Properties of NFCs are strongly influenced by many factors, including stress transfer between the fiber-matrix, fiber orientation, fiber-matrix interaction, fiber volume, fiber length and aspect ratio. There are some drawbacks associated with natural fiber composites, like incompatibility with some polymer system, poor wetting, moisture sensitive, non-destructive behavior, and lower thermal stability that could affect the composites performance [36, 164]. Natural fibers are mainly hydrophilic in nature making them incompatible with the hydrophobic polymer matrix and their addition can lead to a reduction in the mechanical performance of the resulting products [153]. The polarity difference between the thermoplastics matrix and the natural fibers may result in poor fiber-matrix adhesion in the resulting composites. A lack of interaction/compatibility between reinforcement and matirx reduces the overall performance of the resulting materials [41]. The compatibility between the reinforcement and matrix can be improved by chemical surface modification (sodium hydroxide, permanganate, peroxide, maleic anhydride, organosilanes, and isocynates) and physical surface modification (cold plasma treatment, corona treatment) of the fibers [41]. 45

80 Table 2.5. Properties of some natural and synthetic fibers Fibers Density (g/cm 3 ) Specific tensile strength (MPa) Specific tensile modulus (GPa) Elongation at break (%) Cellulose content (wt%) Cellulose Crystallinity (%) Microfibril angle (Degree) Moisture (wt%) Price (US$/kg) References Flax [157, 159, 161] Hemp Jute [153, 157, 159, 161] [153, 157, 159, 161] Kenaf [157, 159] Ramie [153, 157, 159] Sisal [157, 159, 161] Bamboo [157] Coir [159, 161] Cotton [159, 161] Abaca [36, 159] Pineapple [153, 161, 165] Banana [159, 161] Miscanthus [166, 167] Switchgrass [166, 168] Softwood [159] E-glass [36, 159, 165] 46

81 Another strategy is to improve the interfacial adhesion through the addition of a compatibilizing/coupling agent into the composites during processing [36]. Generally, the compatibilizer/coupling agents are used to modify the interface between the matrix and reinforcement in the composite structures. The interfacial modification may be in the form of a physical and/or chemical interaction between the components in the composites. Enhanced compatibility between the phases of the composites helps in improving fiber-matrix adhesion, resulting in improved performance of the resulting composites. Natural fiber composites with compatibilizers are reviewed in detail in subsequent sections Biocomposites based on biodegradable blends as matrix material: Some specific examples There have been many reviews on natural fiber composites [157, 159, 163, 169], [161], [41, 170, 171], [10, 36, 46, 172]. Most of these publications reviewed single polymer matrix based biocomposites. Reviews of biocomposites fabricated using biodegradable blends as matrix material are very limited. Recently, biocomposite fabrication using blends of polymer matrices is part of the growing trend because blend matrices could provide an optimum stiffness-toughness balance, or tailored properties for resulting biocomposites. Currently, few biodegradable preblends are commercially available in the market, for instance PHBV/PBAT blend, available under the name of ENMAT TM from Tianan Biologic Materials Company. Ltd., China [173]. BASF commercialized a biodegradable polymer blend under the trade name Ecovia. It is a blend of PLA/PBAT (45/55 wt%) with 45 wt% biobased content [174]. In addition, FKuR is producing PLA-Copolyester blends under the trade name Bio-Flex. All of these blends are successful matrix material for the incorporation of natural fibers. Zhang et al., [175] fabricated a biodegradable and renewable ternary blends from PHBV, PLA and PBS. The ternary blends (PLA/PHBV/PBS) showed a unique stiffness-toughness balanced mechanical properties 47

82 compared to their neat components. The authors concluded that the ternary blend (PLA/PHBV/PBS) system is very promising for the biocomposites application. Similarly, Muthuraj et al., [54] have also developed a binary blend of PBS and PBAT with good thermal, mechanical and thermo-mechanical properties to be used as a matrix material for composite applications. These research efforts are indicative of the fact that the multiphase polymer blend provides a simple and an effective way to develop a new polymer matrix system for natural fiber composites Biocomposites based on PHBV blends As stated previously PHBV is a brittle and biodegradable polymer. By blending PHBV with tough, biodegradable polymers, their mechanical and thermal properties can be tailored without sacrificing the biodegradability [117]. Therefore, many researchers have extensively studied the toughened PHBV blend biocomposites (Table 2.6). For example, Javadi et al., [173] investigated the preparation and performance of solid and microcellular pre-blend of PHBV/PBAT and its recycled wood fiber (RWF) based composites with and without silane treatments. There were no significant changes in the SEM morphology of the PHBV/PBAT/silane-treated-RWF composites and PHBV/PBAT/untreated-RWF composites, suggesting that silane treatment was not worthy for improving thermal properties and interfacial bonding between the fiber-matrix. However, addition of 10% RWF into PHBV/PBAT blends increased the crystallinity, storage modulus, specific tensile strength, tensile modulus and decreased strain at break and toughness compared to the neat PHBV/PBAT blend. Overall, the authors concluded that silane-treated-rwf reinforced PHBV/PBAT composites did not induce any significant changes in morphology, mechanical and thermal properties of PHBV/PBAT/RWF untreated-rwf composites. 48

83 Nagarajan et al., [168] studied the performance of biocomposites fabricated from a preblend of PHBV/PBAT (45/55%, ENMAT TM ) as matrix and switchgrass fiber (20-40 wt%) as reinforcement. An obvious increase was observed in the modulus of the composites because of the reinforcing effect of the switchgrass fibers. However, the resulting composites showed poor interfacial adhesion between the fiber-matrix. Therefore, the interfacial adhesion was improved by adding poly (diphenylmethane diisocynate), (PMDI) as a compatibilizer (0.5, 0.75 and 1 phr). SEM analysis indicated the occurrence of a better stress transformation between the fibers and matrix through improved interfacial adhesion, which lead to an increase in the mechanical performance of the resulting composites. The critical concentration of compatibilizer was found to be 0.75 phr. This critical concentration yielded 80, 40 and 56% improvements in notched Izod impact strength, flexural strength and tensile strength, respectively. Furthermore, the composites with and without compatibilizer showed a considerable improvements in storage modulus and heat resistance properties compared to the matrix. The melt flow of both compatibilized and uncompatibilized composites sharply reduced in comparison to the PHBV/PBAT blend matrix. For example, with addition of 30 wt% switchgrass fibers into PHBV/PBAT matrix, melt flow index (MFI) reduced to 7 g/10min. It could be due to the restriction of polymer chain mobility in the presence of stiff fibers. The MFI of the composites with 30 wt% switchgrass fibers and 0.75 phr compatibilizer is around 5 g/10 min. Authors suggest that this biocomposites can be injection molded with the help of some processing additives. Biocomposites was prepared from a PHBV/PBAT(45/55%) blend and distiller s dried grains with solubles (DDGS) using melt processing technique by Zarrinbakhsh et al., [176]. This study explored the influence of compatibilizer (0.5 and 1% PMDI) and lubricating agent (3 and 6% corn oil) on the performance of the resulting biocomposites. Use of PMDI as a 49

84 compatibilizer was observed to increase mechanical properties significantly when compared to uncompatibilized composites. The composites with only 0.5% PMDI showed a maximum improvement in tensile strength (39%) and modulus (24%). There was no significant change of these mechanical properties while increasing PMDI content from 0.5 to 1%. Similar trends were observed in flexural properties. These improvements may be due to the good interfacial adhesion that occurred through cross-linking reactions between the components in the composites. This reaction was confirmed by MFI reduction, as well as by the increased melt viscosity of the composites. However, the addition of only corn oil reduced both the stiffness and strength of the composites. Interestingly, combination of both corn oil and PMDI displayed a synergistic effect in impact strength and elongation at break. Authors attributed this improvement to the lubrication effect of corn oil, which enhanced the chemical reaction between the components. Water absorption rate of all the biocomposites was higher than PHBV/PBAT blend matrix. However, the initial water absorption rate of the compatibilized composites was lower than that of the uncompatibilized composites. Reduction of free hydroxyl groups in the DDGS particles after chemically reacting with the matrix was believed to be the reason. The interfacial interaction between matrix and DDGS was justified by scanning electron microscopy analysis. In few other studies, the biodegradability of DDGS/PBAT [177], DDGS/PLA [178], and DDGS/PHA [179] composites has been investigated under different environmental conditions. These DDGS based biocomposites exhibited superior biodegradation rate compared to their biodegradable polymer matrix. In a different study, Nagarajan et al., [158] studied the influence of adding 30 wt% of various natural fibers (miscanthus, switchgrass, wheat straw, corn stalk, and soy stalk) into PHBV/PBAT (45/55 wt%) blend by extrusion and injection molding technique. They have 50

85 evaluated the water uptake behavior, mechanical, thermal and thermo-mechanical properties. Among biocomposites with different types of fibers, miscanthus based composites exhibited higher mechanical properties, heat deflection temperature, higher thermal stability, and comparatively low water absorption. The observed property differences between the composites were due to difference in fiber composition, fiber length distribution, and the interaction between the fibers-matrix. Moreover, SEM micrographs showed debonding and fiber pullout from the matrix, which was attributed to the inefficient stress transfer occurring between the components. Similar to this study, various types of agricultural residues reinforced PLA composites have also been reviewed [180]. The biodegradability rate of these nature fiber reinforced composites can be expected to be much faster in comparison to their parent components [181]. The influences of man-made cellulose, jute and abaca fiber on the mechanical properties of PHBV/PBAT (70/27.6% and 2.6% processing additive) blend were also investigated by Bledzki and Jaszkiewicz [182]. The composites samples were produced in extrusion and injection molding process. The fillers had positive effects on tensile properties and notched Charpy impact strength. Among the abaca, jute fibers and man-made cellulose, abaca fiber has a high tensile strength (980 MPa) and a high modulus (27-32 GPa) [182]. However, abaca fiber based composites underperformed compared to man-made cellulose and jute fiber composites. The composites with man-made cellulose showed a massive improvement (> 500%) in notched Charpy impact strength and a 50% improvement in tensile strength over the PHBV/PBAT matrix. SEM micrograph of PHBV/PBAT composites showed a poor adhesion between the fiber-matrix. This was attributed to the partial degradation of PHBV/PBAT blend during melt processing and post-molding shrinkage of the PHBV/PBAT blend matrix. Similar to a previous study, a massive increase in the tensile modulus was observed for the PHBV/PBAT/(70/27.6% 51

86 and 2.6% processing additive) blend with jute fiber composites [183]. The observed results were attributed to chemical composition, aspect ratio and the geometry of the fibers. SEM analysis revealed that the compatibility between the fiber and the matrix was poor due to the shrinkage nature of PHBV in the matrix. Even still, PHBV/PBAT composite properties are comparable with PP biocomposites. The effects of sodium hydroxide (NaOH) treated and pectinase retted kenaf fibers on PHBV/PBAT composite properties were studied using compression molding [184]. This study investigated the dynamic mechanical analysis and crystallization behavior of composites with two different fiber lengths (5 and 10 mm). The dynamic mechanical properties of the composites showed that the storage modulus of the composites increased with the addition of fiber compared to the PHBV/PBAT matrix. The storage modulus of pectinase retted fiber reinforced composites was higher compared to corresponding composites with NaOH retted fiber. The composites with 5 mm retted fibers showed a higher storage modulus than the 10 mm fiber and their hybrid (1:1) composites. In the presence of both pectinase and NaOH retted fibers, the glass transition temperature of PBAT increased and the damping factor of PHBV phase reduced. The fiber retting had influence on crystallization, melting and spherulite morphology of the PHBV/PBAT matrix, which was confirmed by polarizing optical microscopy (POM) analysis. POM was used to study the nucleation density and spherulite growth of PHBV/PBAT blends and their biocomposites. The nucleation density of the PHBV/PBAT blend increased with the addition of fibers and it lead to a change in the crystal morphology of the resulting composites. Consequently, the melting point of the composites was reduced as compared to the PHBV/PBAT matrix with increase in the melting enthalpy. This suggested that the pectinase retted kenaf fibers have more surface activity and thus increased the crystallinitythan the unmodified kenaf fiber. 52

87 The complex viscosity of the composites was higher than the matrix because of the restricted polymer chain mobility in the presence of rigid fibers. NaOH retted fiber composites showed less nucleation density and larger spherulites as compared to pectinase retted fiber composites. This was attributed to the enhanced roughness of the NaOH retted kenaf, which tend to form tiny spherulites. Zarrinbakhsh et al., [185] investigated green composites made from distiller s dried grains with solubles (DDGS) and an optimized blend of PHBV/PBS (70/30 wt%) matrix. Mechanical, thermal, morphological, and thermo-mechanical properties of the injection molded composites were studied. Interestingly, after water washing of DDGS, the thermal stability of DDGS was improved because of water solube moleculaes were washed away. The improved thermal stability allowed for higher processing temperatures without any thermal degradation of DDGS up to 180 o C. The water-washed DDGS composites showed an improved tensile and flexural strength compared to non-washed DDGS composites. When poly(diphenylmethane diisocynate) (PMDI) was introduced as a compatibilizer into water-washed DDGS composite, there were significant improvements in flexural strength and tensile strength. The flexural and tensile modulus of the DDGS composites has an increasing trend in the following order: PHBV/PBS blend matrix<non-washed DDGS composites<water-washed DDGS composites<water-washed/compatibilized DDGS composites. The improved tensile and flexural properties were attributed to the enhanced interfacial adhesion between the fillers and matrix, which was confirmed through SEM. The impact strength of water-washed DDGS composites and water-washed/compatibilized DDGS composites was lower compared to non-washed DDGS composites. It was hypothesized that the poor interfacial bonding between the phases needs more 53

88 energy to break the fillers because the crack propagation occurred around the fillers rather than braking fillers. Recently, Zhang et al., [186] studied preparation and performance evaluation of sustainable green composites from ternary PHBV/PBAT/epoxidized natural rubber (50/35/15 wt%) blend matrix and miscanthus fibers. The selected ternary blend system has non-break behavior under notched Izod impact test. However, the addition of 10 wt% miscanthus fiber into this ternary blend matrix, the impact strength of the resulting composites was observed around 273 J/m. A similar trend was observed in the elongation at break and tensile strength. This reduction was mainly due to weak interfacial interaction between the fiber-matrix as well as poor load transfer capability between the fiber-matrix. Interestingly, this drawback was overcome by adding 0.3 wt% DCP as a reactive agent in the composites. In the presence of DCP, the polymers formed graft copolymers and partially crosslink network between the polymers, which improved the compatibility between the blend components. The authors also investigated the effect of 0.3 wt% DCP while increasing fiber content up to 20 wt% in the resulting composites. The impact energy of the composite with 20 wt% fiber and 0.3 wt% DCP still remained at J/m. The flexural properties and storage modulus were found to increase with increasing fiber content up to 20 wt%, which is a fairly common observation in natural fiber reinforced polymer composites Biocomposites based on PLA blends PBAT, PBS, and PCL are the best candidates for increased toughness of PLA in their respective blends. Blending PLA with tough polymers (PBAT, PBS, and PCL) resulted in a higher toughness but lower stiffness in the blends in comparison to virgin PLA. The low stiffness can be addressed by adding fibers/filler to PLA blends. Therefore, many researchers have extensively studied toughened PLA based biocomposites, as shown in Table 2.6. Commercially 54

89 available pre-blends of PLA/PBAT, 45/55 wt%, (BASF-Ecovio ) with chemically (alkali treatment) treated curaua fiber reinforced biocomposites were studied by Harnnecker et al., [174]. There was a slight (4 o C) increase in T g of the composites compare to the PLA/PBAT blend. This fact was attributed to the intermolecular (hydrogen bonds) interaction between the fibers and polymer matrix. Moreover, DSC thermograms showed an endothermic peak during heating cycle for all the composites whereas PLA/PBAT blend did not show any endothermic peak during heating cycle. This was believed to be due to the nucleation effect of curaua fiber in the composites. Performance of these composite was studied with and without the addition of compatibilizer (2 wt%) i.e., maleic anhydride-grafted-polypropylene (MA-g-PP). Tensile, impact and flexural strength of the compatibilized composites increased with increased fiber content from 5 to 20 wt%. For example, the compatibilized composites with 20 wt% curaua fibers showed 56% increase in flexural strength and 75% increase in the tensile strength. 55

90 Table 2.6. Recently developed biodegradable polymer blend matrix based biocomposites Blend matrix Fiber/filler Compatibilizer/Coupling agent Manufacturing Type Content Name Chemical structure process PHBV/PBAT Recycled wood fiber 10% Gammamethacryloxypropylt rimethoxysilane PHBV/PBAT Switchgrass 20-40% Poly diphenylmethane diisocyanate (PMDI) CH 2 =C(CH 3 ) CO 2 CH 2 CH 2 CH 2 Si (OCH 3 ) 3 Injection molding Injection molding References [173] [168] PHBV/PBAT Distiller s dried grains with solubles (DDGS) 20% Poly diphenylmethane diisocyanate (PMDI) Injection molding [176] PHBV/PBAT Miscanthus, switchgrass, wheat straw, corn stalk, and soy stalk Jute, abaca and manmade cellulose Cellulose, Jute, abaca and flax 30% NA NA Injection molding PHBV/PBAT 30% NA NA Injection molding PHBV/PBAT 30% NA NA Injection molding PHBV/PBAT Kenaf 5 and NA NA Compression 10% molding PHBV/PBS Distiller s dried grains Injection with solubles (DDGS) molding 30% Poly diphenylmethane diisocyanate (PMDI) PHBV/PLA Soy hull, switchgrass, and miscanthus 30% NA NA Injection molding PHBV/PLA Soy fiber 30 and 2,5-Bis(tert- Injection 50% butylperoxy)-2,5- molding dimethylhexane [158] [182] [183] [184] [185] [166] [187] 56

91 PLA/PBAT Kenaf 10-50% (3-aminopropyl) trimethoxysilane (APTMS) Compression molding [188] PLA/PBAT Curaua 5-20% Maleic anhydride grafted polypropylene (MAg-PP) Compression molding PLA/PBAT Softwood flour 30-50% NA NA Injection molding PLA/PBAT Wood flour 20-30% NA NA Compression molding PLA/PBAT Ramie 10% NA NA Compression molding PLA/PCL Alkali treated palm fiber 10-25% Dicumyl peroxide Injection (DCP) molding PLA/PCL Silane treated Jute and Untreated Jute 50% Trimethoxy (methyl) silane CH 3 Si (OCH 3 ) 3 Compression molding PLLA/PBS Flax 25.5% NA NA Injection molding [174] [189] [190] [191] [192] [193] [194] 57

92 The increased mechanical properties were attributed to the improvement of interfacial adhesion between the fibers and matrix, which was confirmed by cross-sectional fracture analysis. In addition, water absorption of the compatibilized composite was almost 100% lower compared to the uncompatibilized composite. Sis et al., [188] examined the impact, flexural, and tensile properties of the PLA/PBAT/kenaf fiber composites with and without coupling agent ((3- aminopropyl) trimethoxysilane), APTMS, by melt blending. The tensile, flexural and impact strength of PLA/PBAT/kenaf composites decreased with increasing fiber loading. This was because of inadequate fiber-matrix interaction and poor dispersion of fiber in the matrix. The composites were modified with different amount (from 1 to 5 wt%) of APTMS coupling agent. The optimum properties were obtained for PLA/PBAT/kenaf composites with the addition of 2 wt% coupling agent. This composite showed 22, 42 and 63% improvement in Izod impact strength, tensile strength, and flexural strength, respectively. In addition, the tensile and flexural moduli of the composites with coupling agent were considerably higher than composites without coupling agent. The storage modulus of the composites with and without coupling agent exhibits similar trends, like the tensile and flexural modulus. This behavior was ascribed to good adhesion between the phases. Unlike PLA/PBAT/kenaf and curaua fiber composites, PLA/PBAT/wood flour composites exhibited an increase in both tensile and flexural properties compare to the neat PLA/PBAT (45/55 wt%) blend (Ecovio ) [189]. With increasing fiber load from 30 to 50 wt%, a 48.2% reduction was observed in the Charpy impact strength of the resulting composites compared to PLA/PBAT blend matrix. The loss in impact strength was attributed to low fiber aspect ratio of wood flour [189]. The water resistance property of the composites was reduced because of hygroscopic nature of the wood flour. However, the composites with an addition of 58

93 50 wt% wood flour showed significant improvements in tensile strength (27.2%), modulus (174.3%), and flexural strength (20.6%) as compared to neat PLA/PBAT blend. These improvements were related to good fiber-matrix adhesion and better distribution of fibers in the matrix. This trend is in contradiction to the finding of Georgiopoulos et al., [190]. Another study observed similar results for PLA/PBAT(70/30 wt%) biocomposites [195]. More recently, a comparison study between PLA/30% ramie composites and PLA/PBAT(95/5, 90/10 and 85/15%) with 30 wt% ramie composites has been investigated by Yu and Li [191]. In general, strong interfacial bonding between fiber and the matrix decreases the water absorption of the natural fiber composites. The Vicat softening temperature, flexural and tensile strength of the PLA/PBAT/ramie (85/15%) composites were lower than PLA based composites, which was due to the soft elastomeric nature of the PBAT phase. Yet, the notched Izod impact strength of the PLA/PBAT/ramie composites was superior compared to PLA/ramie composites. This suggests that the PBAT phase induced stress concentration under impact load. Consequently, the PBAT phase absorbed more energy during failure of the PLA/PBAT composites. In order to maintain energy dissipation source in the PLA/PBAT composite system, concentration of PBAT was considered an important factor. Furthermore, thermal stability of the PLA/PBAT based composites was studied by thermogravimetric analysis. It was found that the thermal stability of PLA/PBAT based ramie composites was higher than PLA based composites because the PBAT phase enhanced the charring process for the composites [191]. Ibrahim et al., [192] studied the performance of uncompatibilized and compatibilized PLA/PCL (10/90 wt%)/alkali treated palm fiber composites with 10 to 25 wt% fiber loading. With the addition of 0.01 phr DCP, the flexural modulus, impact strength, tensile strength, and flexural strength of the composites were superior when compared to uncompatibilized 59

94 composites. These improvements were attributed to strong fiber and matrix adhesion and good fiber distribution in the matrix. Fiber and matrix adhesion was confirmed by damping behavior as well as by SEM analysis. DCP increased the viscosity and crosslinking while reducing the brittleness of the composites. Consequently, the tensile modulus of compatibilized composites was lower than uncompatibilized composites in the entire fiber composition. A similar trend was noticed in storage modulus of the compatibilized composites. The storage modulus of the composites increased up to 15 wt% fiber loading and then decreased with increasing fiber up to 25 wt%. The increased storage modulus was due to the reinforcing effect of the fibers. The decreased storage modulus with increasing fiber loading was a result of free volumes that were created due to crosslinking. Furthermore, the thermal stability of the composites was higher than that of the matrix. Overall, the composites with 15 wt% fiber load showed balanced mechanical, thermal and thermo-mechanical properties. Recently, PLLA/PBS blends with 25.5% flax fiber composites showed well balanced mechanical properties without any compatibilizer [194]. The consequence of incorporating 30 wt% miscanthus, switchgrass, and soy hull their hybrids as a reinforcement in the PHBV/PLA (70/30 wt%) blend matrix has been investigated by Nanda et al., [166]. The performance of the green composites was examined by means of their thermo-mechanical properties. Due to lack of interfacial adhesion between the components, the tensile strength of all the composites was reduced significantly when compared to PHBV/PLA matrix. Yet, the tensile and flexural moduli of all the composites were higher compared to PHBV/PLA blend. There was no significant difference observed in the notched Izod impact strength of all the composites compared to the PHBV/PLA matrix. Lower aspect ratio, and smaller cellulose and lignin content of soy hull in comparison to miscanthus and switchgrass resulted in soy hull composites displaying inferior mechanical properties. All of the composites 60

95 showed superior heat resistance properties compared to their matrix. Soy hull based composites showed lower thermal stability compared to miscanthus and switchgrass based composites. This could probably be due to the less thermally stable compounds, such as pectin and proteins, present in the soy hull [187]. The melting temperature and melting enthalpy of the PLA and PHBV were not affected in the presence of all the fibers. Soy hull based composites showed more water absorption compared to other composites. This was due to the smaller amount of lignin present in the soy hull [187]. This trend was changed in compatibilized PHBV/PLA/soy fiber composites Natural fiber composites market and their applications Natural fiber reinforced polymeric composites were produced early in 1908 [46]. Since then, natural fiber composites (NFCs) have become attractive in both automotive and building industries because of its low cost, low CO 2 emissions and because it is lightweight. The specific mechanical properties of NFCs are superior to synthetic fiber (glass, carbon and kevlar) composites [160]. Currently, PP and polyurethane based biocomposites are being used in automotive interior and exterior parts [10, 36]. In 2010, million pounds of natural fiber composites were produced and its worth was million US$. This is expected to reach million US$ by 2016, which means an 11% growth rate over the next 5 years [196]. Automotive companies have set a goal and are working towards the possible replacement of some nonbiodegradable polymer products by using biocomposites and bioplastics. An average vehicle contains 113 kg of plastic parts and approximately 50% of automotive interior parts are made up of plastics [197]. Annually million vehicles are discarded, and 25% of their parts cannot be recycled or reused because they are made from foams, rubber, and synthetic fibers [197]. 61

96 Biocomposites from PLA/kenaf and PBS/bamboo are used for making automotive spare tire cover and tailgate trim, respectively [36, 198]. There exists the possibility of producing durable car parts entirely from biodegradable materials. More specifically, a vehicle (Agri-Car) with 90% biodegradable materials has been jointly designed by the University of Akron and Ohio State University [46]. In 2009, the University of Warwick developed a prototype World First Formula 3 racing car by using green composites [46]. In addition, the mechanical and heat resistance properties of PLA/kenaf biocomposites were almost equal to those of conventional polycarbonate/glass fiber composites. PLA based biocomposites are used to produce prototype kayaks, wind turbine rotors, laptop computers and cellular phones [46, ]. According to this resource [159], the NFC market is expected to grow by up to 20% and more than 50% in automotive and building applications, respectively. The global NFC market is expected to grow to million US$ by 2016 [196]. This forecast is significantly higher than the 2010 world annual production (US$ million) of the NFCs market. The European biocomposites market growth rate is predicted at 21% per year. In 2008, biocomposite production totalled 129,000 tons in Europe and this growth is expected to increase to 427,000 tons in Parallel to this, hemp and flax fibers are major players in the natural fiber composites market share in the Europe [163]. In 2000, the North American natural fiber market was worth 155 million US$ and it is expected to reach US$ 1.38 billion by 2025 [157] Conclusions Recently, the utilization of biodegradable polymers has been accelerated due to the environmental concerns associated with disposal of non-biodegradable plastics in landfills. Higher cost, thermal instability, low melt strength and brittleness of biodegradable polymers currently limit their extended application. Great efforts have been taken to overcome these 62

97 drawbacks by blending technique as well as incorporation of natural fibers. However, most of the polymer blends are thermodynamically immiscible due to a difference in solubility. Generally, the immiscible polymer blends lead to a weak interface between the blend partners. The poor interfacial adhesion between the blend components can be modified using various additive, compatibilizer and compatibilization strategies. The compatibilization effect and its influence on the thermal, thermo-mechanical, morphological and mechanical properties of the blends have been reviewed in this chapter. With the increasing cost of fossil fuels, more research attention is being focused on developing renewable resource-based materials for various industrial applications. Natural fiber reinforced biodegradable polymers provide a sustainable alternative to non-biodegradable polymeric composites. It was found that biocomposites could be fabricated from a stiffness-toughness balanced blend matrix with various natural fibers. The difference in strength and modulus values between the natural fibers is attributed to the difference in individual fiber compositions. Recently, multiple binary blend matrix based biocomposites have been studied and they all exhibit superior performance compared to single polymer based biocomposites. Due to uncompatibility between the matrix and fiber, several approaches have been explored to modify both matrix and fiber to enhance the compatibility between them. The mechanical properties of all the compatibilized composites are significantly higher in comparison to uncompatibilized one. However, there are only a limited number of applications for biodegradable polymers and their biocomposites. It is believed that significant research into existing processing technology will help to widen the application potential of biodegradable polymer-based blends and their biocomposites. 63

98 References [1] B. Imre, B. Pukánszky, Compatibilization in bio-based and biodegradable polymer blends, European Polymer Journal, 2013, 49 (6): [2] (Last accessed in January, 2015). [3] (Last accessed in January, 2015) [4] (Last accessed in January, 2015). [5] T. R. Plant, The technology roadmap for plant/crop-based renewable resources 2020, (Last accessed in January, 2015). [6] B. E. DiGregorio, Biobased Performance Bioplastic: Mirel, Chemistry & Biology, 2009, 16 (1): 1-2. [7] ASTM D , Standard Specification for Labeling of Plastics Designed to be Aerobically Composted in Municipal or Industrial Facilities, ASTM International, West Conshohocken, PA, 2012, [8] S. Muniyasamy, A. Anstey, M. M. Reddy, M. Misra, A. Mohanty, Biodegradability and Compostability of Lignocellulosic Based Composite Materials, Journal of Renewable Materials, 2013, 1 (4): [9] A. Soroudi, I. Jakubowicz, Recycling of bioplastics, their blends and biocomposites: A review, European Polymer Journal, 2013, 49 (10): [10] M. J. John, S. Thomas, Biofibres and biocomposites, Carbohydrate Polymers, 2008, 71 (3): [11] U. Witt, T. Einig, M. Yamamoto, I. Kleeberg, W. D. Deckwer, R. J. Müller, Biodegradation of aliphatic aromatic copolyesters: evaluation of the final biodegradability and ecotoxicological impact of degradation intermediates, Chemosphere, 2001, 44 (2):

99 [12] U. Witt, M. Yamamoto, U. Seeliger, R.-J. Müller, V. Warzelhan, Biodegradable Polymeric Materials-Not the Origin but the Chemical Structure Determines Biodegradability, Angewandte Chemie International Edition, 1999, 38 (10): [13] L. Yu, K. Dean, L. Li, Polymer blends and composites from renewable resources, Progress in Polymer Science, 2006, 31 (6): [14] W. Groot, J. V. Krieken, O. Sliekersl, A. D. Vos, Chapter 1: production and purifica tion of lactic acid and lactide. In: Auras, R.A., Lim, L.-T., Selke, S.E., Tsuji, H. (Eds.), Poly(Lactic Acid): Synthesis, Structures, Properties, Processing, and Applications. John Wiley & Sons inc., Hoboken, New Jersey, 2011, pp. 3. [15] R. Auras, B. Harte, S. Selke, An Overview of Polylactides as Packaging Materials, Macromolecular Bioscience, 2004, 4 (9): [16] L. T. Lim, R. Auras, M. Rubino, Processing technologies for poly(lactic acid), Progress in Polymer Science, 2008, 33 (8): [17] D. E. Henton, P. Gruber. J. Lunt, J. Randall, Chapter 16: polylactic acid technology. In: Mohanty, A.K., Misra, M., Drzal, L.T. (Eds.), Natural Fibers, Biopolymers, and Biocomposites. CRC Press LLC, Boca Raton, FL, 2005, pp [18] D. Garlotta, A Literature Review of Poly(Lactic Acid), Journal of Polymers and the Environment, 2001, 9 (2): [19] K. Madhavan Nampoothiri, N. R. Nair, R. P. John, An overview of the recent developments in polylactide (PLA) research, Bioresource Technology, 2010, 101 (22): [20] M. S. Huda, L. T. Drzal, M. Misra, A. K. Mohanty, Wood-fiber-reinforced poly(lactic acid) composites: Evaluation of the physicomechanical and morphological properties, Journal of Applied Polymer Science, 2006, 102 (5):

100 [21] J. Vandewijngaarden, M. Murariu, P. Dubois, R. Carleer, J. Yperman, P. Adriaensens, S. Schreurs, N. Lepot, R. Peeters, M. Buntinx, Gas Permeability Properties of Poly(3- hydroxybutyrate-co-3-hydroxyhexanoate), Journal of Polymers and the Environment, 2014, 22 (4): [22] (Last accessed in January, 2015). [23] M. Zhang, N. L. Thomas, Blending polylactic acid with polyhydroxybutyrate: The effect on thermal, mechanical, and biodegradation properties, Advances in Polymer Technology, 2011, 30 (2): [24] J. Lunt, Large-scale production, properties and commercial applications of polylactic acid polymers, Polymer Degradation and Stability, 1998, 59 (1 3): [25] J. Ren, (ed.) Biodegradable Poly(Lactic Acid), Synthesis, Modification, Processing and Applications, Tsinghua University Press, Beijing (Springer, Heidelberg), 2010, pp [26] G. Kfoury, J.-M. Raquez, F. Hassouna, J. Odent, V. Toniazzo, D. Ruch, P. Dubois, Recent advances in high performance poly(lactide): From green plasticization to super-tough materials via (reactive) compounding, Frontiers in Chemistry, 2013, 1,1. [27] Z. N. Terzopoulou, G. Z. Papageorgiou, E. Papadopoulou, E. Athanassiadou, E. Alexopoulou, D. N. Bikiaris, Green composites prepared from aliphatic polyesters and bast fibers, Industrial Crops and Products, 2014, (0): (doi: /j.indcrop ). [28] R. A. J. Verlinden, D. J. Hill, M. A. Kenward, C. D. Williams, I. Radecka, Bacterial synthesis of biodegradable polyhydroxyalkanoates, Journal of Applied Microbiology, 2007, 102 (6):

101 [29] D. d. S. Rosa, D. M. Lenz, Chapter 16: biocomposites: Influence of matrix nature and additives on the properties and biodegradation behaviour, In: Chamy. R., Rosenkranz. F., (Eds.), Biodegradation-Engineering and Technology, InTech, Croatia, 2013, pp [30] M. Avella, E. Martuscelli, M. Raimo, Review Properties of blends and composites based on poly(3-hydroxy)butyrate (PHB) and poly(3-hydroxybutyrate-hydroxyvalerate) (PHBV) copolymers, Journal of Materials Science, 2000, 35 (3): [31] Y. Parulekar, A. K. Mohanty, Biodegradable toughened polymers from renewable resources: blends of polyhydroxybutyrate with epoxidized natural rubber and maleated polybutadiene, Green Chemistry, 2006, 8 (2): [32] Q.-S. Liu, M.-F. Zhu, W.-H. Wu, Z.-Y. Qin, Reducing the formation of six-membered ring ester during thermal degradation of biodegradable PHBV to enhance its thermal stability, Polymer Degradation and Stability, 2009, 94 (1): [33] I. Vroman, L. Tighzert, Biodegradable Polymers, Materials, 2009, 2 (2): [34] Q. Wu, Y. Wang, G.-Q. Chen, Medical application of microbial biopolyesters polyhydroxyalkanoates. Artificial Cells, Blood Substitutes, and Biotechnology. 37 (1), [35] N. Grassie, E. J. Murray, P. A. Holmes, The thermal degradation of poly(-(d)-βhydroxybutyric acid): Part 3 The reaction mechanism, Polymer Degradation and Stability, 1984, 6 (3): [36] G. Koronis, A. Silva, M. Fontul, Green composites: A review of adequate materials for automotive applications, Composites Part B: Engineering, 2013, 44 (1): [37] M. G. A. Vieira, M. A. da Silva, L. O. dos Santos, M. M. Beppu, Natural-based plasticizers and biopolymer films: A review, European Polymer Journal, 2011, 47 (3):

102 [38] K. Van de Velde, P. Kiekens, Biopolymers: overview of several properties and consequences on their applications, Polymer Testing, 2002, 21 (4): [39] A. K. Mohanty, M. Misra. T. L. Drzal, S. E. Selke, B. R. Harte, G. Hinrichsen, Chapter:1 Natural fibers, and biocomposites: an introduction, (Eds) A. K. Mohanty, M. Misra, T. L. Drzal, Natural fibers, biopolymers, and biocomposites, CRC Press-Taylor & Francis Group, Boca Raton (2005), pp [40] T. Fujimaki, Processability and properties of aliphatic polyesters, BIONOLLE, synthesized by polycondensation reaction, Polymer Degradation and Stability, 1998, 59 (1 3): [41] G. Bogoeva-Gaceva, M. Avella, M. Malinconico, A. Buzarovska, A. Grozdanov, G. Gentile, M. E. Errico, Natural fiber eco-composites, Polymer Composites, 2007, 28 (1): [42] J.-L. Audic, B. Chaufer, Properties of biodegradable poly(butylene adipate-co-terephtalate) and sodium caseinate blends, Journal of Applied Polymer Science, 2012, 125 (S2): E459-E467. [43] J. Xu, B.-H. Guo. Microbial Succinic Acid, Its Polymer Poly(butylene succinate), and Applications: G.-Q. Chen, (Editor), Plastics from Bacteria: Springer Berlin Heidelberg; p [44] S. Kabasci, I. Bretz, Succinic acid: synthesis of biobased polymers from renewable resources, in Renewable Polymers (Ed. V. Mittal), Scrivener Publishing LLC, ISBN: S, 2012, pp [45] R. Babu, K. O'Connor, R. Seeram, Current progress on bio-based polymers and their future trends, Progress in Biomaterials, 2013, 2 (1): 8. [46] E. Zini, M. Scandola, Green composites: An overview, Polymer Composites, 2011, 32 (12):

103 [47] (Last accessed in January, 2015). [48] P. Ma, Z. Ma, W. Dong, Y. Zhang, P. J. Lemstra, Structure/Property Relationships of Partially Crosslinked Poly(butylene succinate), Macromolecular Materials and Engineering, 2013, 298 (8): [49] N. Jacquel, F. Freyermouth, F. Fenouillot, A. Rousseau, J. P. Pascault, P. Fuertes, R. Saint- Loup, Synthesis and properties of poly(butylene succinate): Efficiency of different transesterification catalysts, Journal of Polymer Science Part A: Polymer Chemistry, 2011, 49 (24): [50] E. S. Yoo, S. S. Im, Melting behavior of poly(butylene succinate) during heating scan by DSC, Journal of Polymer Science Part B: Polymer Physics, 1999, 37 (13): [51] R. Wang, S. Wang, Y. Zhang, C. Wan, P. Ma, Toughening modification of PLLA/PBS blends via in situ compatibilization, Polymer Engineering and Science, 2009, 49 (1): [52] Y. J. Kim, O. O. Park, Miscibility and Biodegradability of Poly(Butylene Succinate)/Poly(Butylene Terephthalate) Blends, Journal of Polymers and the Environment, 1999, 7 (1): [53] S. Sinha Ray, K. Okamoto, M. Okamoto, Structure Property Relationship in Biodegradable Poly(butylene succinate)/layered Silicate Nanocomposites, Macromolecules, 2003, 36 (7): [54] R. Muthuraj, M. Misra, A. Mohanty, Biodegradable Poly(butylene succinate) and Poly(butylene adipate-co-terephthalate) Blends: Reactive Extrusion and Performance Evaluation, Journal of Polymers and the Environment, 2014, 22 (3):

104 [55] B. Calabia, F. Ninomiya, H. Yagi, A. Oishi, K. Taguchi, M. Kunioka, M. Funabashi, Biodegradable Poly(butylene succinate) Composites Reinforced by Cotton Fiber with Silane Coupling Agent, Polymers, 2013, 5 (1): [56] S. Risse, L. Tighzert, F. Berzin, B. Vergnes, Microstructure, rheological behavior, and properties of poly(lactic acid)/poly(butylene succinate)/organoclay nanocomposites. Journal of Applied Polymer Science, 2014, 131 (12), [57] P. Ma, D. G. Hristova-Bogaerds, P. J. Lemstra, Y. Zhang, S. Wang, Toughening of PHBV/PBS and PHB/PBS Blends via In situ Compatibilization Using Dicumyl Peroxide as a Free-Radical Grafting Initiator, Macromolecular Materials and Engineering, 2012, 297 (5): [58] V. Tserki, P. Matzinos, E. Pavlidou, D. Vachliotis, C. Panayiotou, Biodegradable aliphatic polyesters. Part I. Properties and biodegradation of poly(butylene succinate-co-butylene adipate), Polymer Degradation and Stability, 2006, 91 (2): [59] E. Marten, R.-J. Müller, W.-D. Deckwer, Studies on the enzymatic hydrolysis of polyesters. II. Aliphatic aromatic copolyesters, Polymer Degradation and Stability, 2005, 88 (3): [60] S.-Y. Gu, K. Zhang, R. Jie, Z. Hui, Melt rheology of polylactide/poly(butylene adipate-coterephthalate) blends, Carbohydrate Polymers, 2008, 74 (1): [61] U. Witt, R.-J. Müller, W.-D. Deckwer, Studies on sequence distribution of aliphatic/aromatic copolyesters by high-resolution 13C nuclear magnetic resonance spectroscopy for evaluation of biodegradability, Macromolecular Chemistry and Physics, 1996, 197 (4): [62] K. O. Siegenthaler, A. Künkel, G. Skupin, M. Yamamoto. Ecoflex and Ecovio : Biodegradable, Performance-Enabling Plastics. In: Rieger, Künkel, Coates, Reichardt, Dinjus, 70

105 Zevaco, editors. Synthetic Biodegradable Polymers: Springer Berlin Heidelberg; p [63] M. Breulmann, A. Künkel, S. Philipp, V. Reimer, K. O. Siegenthaler, G. Skupin, M. Yamamoto. Polymers, biodegradable, In: Ullmann's Encyclopedia of Industrial Chemistry. Wiley-VCH, Berlin [64] dable_plastics/ecoflex_product_literature, (Last accessed in January 2015). [65] A. Javadi, A. J. Kramschuster, S. Pilla, J. Lee, S. Gong, L.-S. Turng, Processing and characterization of microcellular PHBV/PBAT blends, Polymer Engineering and Science, 2010, 50 (7): [66] dable_plastics/ecoflex_product_literature, (Last accessed in January 2015). [67] (Last accessed in January 2015). [68] K. M. Stridsberg, M. Ryner, A. C. Albertsson, Controlled Ring Opening Polymerization: Polymers with controlled Architecture. In Albertsson, A. C., (Ed.), Advances in Polymers Science; SpringerVerlag, Berlin, 2002, 157, p [69] H. Uyama, S. Kobayashi, Enzyme-catalyzed polymerization to functional polymers, Journal of Molecular Catalysis B: Enzymatic, 2002, (0): [70] C. Bastioli, Properties and applications of Mater-Bi starch-based materials, Polymer Degradation and Stability, 1998, 59 (1 3):

106 [71] P. Sarazin, X. Roy, B. D. Favis, Controlled preparation and properties of porous poly(llactide) obtained from a co-continuous blend of two biodegradable polymers, Biomaterials, 2004, 25 (28): [72] C. Hinüber, L. Häussler, R. Vogel, H. Brünig, G. Heinrich, C. Werner, Hollow fibers made from a poly(3-hydroxybutyrate)/poly (caprolactone) blend, Express polymer Letters, 2011, 5 (7): [73] S. A. Casarin, S. M. Malmonge, M. Kobayashi, J. A. M. Agnelli, Study on In-Vitro Degradation of Bioabsorbable Polymers Poly (hydroxybutyrate-co-valerate)-(phbv) and Poly (caprolactone)-(pcl), Journal of Biomaterials and Nanobiotechnology, 2011, 2 (3): [74] C. Del Gaudio, L. Fioravanzo, M. Folin, F. Marchi, E. Ercolani, A. Bianco, Electrospun tubular scaffolds: On the effectiveness of blending poly (ε caprolactone) with poly (3 hydroxybutyrate co 3 hydroxyvalerate), Journal of Biomedical Materials Research Part B: Applied Biomaterials, 2012, 100 (7): [75] E. Can, S. Bucak, E. Kınacı, A. Çalıkoğlu, T. Köse, Polybutylene Succinate (PBS)- Polycaprolactone (PCL) Blends Compatibilized with Poly (ethylene oxide)-block-poly (propylene oxide)-block-poly (ethylene oxide)(peo-ppo-peo) Copolymer for Biomaterial Applications, Polymer-Plastics Technology and Engineering, 2014, 53 (11): [76] (Last accessed in January, 2015). [77] M. Żenkiewicz, J. Richert, P. Rytlewski, K. Moraczewski, M. Stepczyńska, T. Karasiewicz, Characterisation of multi-extruded poly(lactic acid), Polymer Testing, 2009, 28 (4): [78] I. Pillin, N. Montrelay, A. Bourmaud, Y. Grohens, Effect of thermo-mechanical cycles on the physico-chemical properties of poly(lactic acid), Polymer Degradation and Stability, 2008, 93 (2):

107 [79] J. P. Lopez, J. Girones, J. A. Mendez, J. Puig, M. A. Pelach, Recycling Ability of Biodegradable Matrices and Their Cellulose-Reinforced Composites in a Plastic Recycling Stream, Journal of Polymers and the Environment, 2012, 20 (1): [80] I. Zembouai, S. Bruzaud, M. Kaci, A. Benhamida, Y.-M. Corre, Y. Grohens, Mechanical Recycling of Poly(3-Hydroxybutyrate-co-3-Hydroxyvalerate)/Polylactide Based Blends, Journal of Polymers and the Environment, 2014, 22 (4): [81] K. Moraczewski, Characterization of multi-injected poly(ε-caprolactone), Polymer Testing, 2014, 33 (0): [82] M. Zaverl, M. Ö. Seydibeyoğlu, M. Misra, A. Mohanty, Studies on recyclability of polyhydroxybutyrate-co-valerate bioplastic: Multiple melt processing and performance evaluations, Journal of Applied Polymer Science, 2012, 125 (S2): E324-E331. [83] C. Kanemura, S. Nakashima, A. Hotta, Mechanical properties and chemical structures of biodegradable poly(butylene-succinate) for material reprocessing, Polymer Degradation and Stability, 2012, 97 (6): [84] H.-S. Kim, H.-J. Kim, Enhanced hydrolysis resistance of biodegradable polymers and biocomposites, Polymer Degradation and Stability, 2008, 93 (8): [85] S. Kim, S.-w. Park, K. Lee, Fishing performance of environmentally friendly tubular pots made of biodegradable resin (PBS/PBAT) for catching the conger eel Conger myriaster, Fisheries Science, 2014, 80 (5): [86] A. M. Harris, E. C. Lee, Heat and humidity performance of injection molded PLA for durable applications, Journal of Applied Polymer Science, 2010, 115 (3): [87] A. M. Harris, E. C. Lee, Durability of polylactide-based polymer blends for injectionmolded applications, Journal of Applied Polymer Science, 2013, 128 (3):

108 [88] S. Kim, S. Park, K. Lee, Fishing Performance of an Octopus minor Net Pot Made of Biodegradable Twines, Turkish Journal of Fisheries and Aquatic Sciences, 2014, 14: [89] M. Deroiné, A. Le Duigou, Y.-M. Corre, P.-Y. Le Gac, P. Davies, G. César, S. Bruzaud, Accelerated ageing and lifetime prediction of poly(3-hydroxybutyrate-co-3-hydroxyvalerate) in distilled water, Polymer Testing, 2014, 39 (0): [90] M. Deroiné, A. Le Duigou, Y.-M. Corre, P.-Y. Le Gac, P. Davies, G. César, S. Bruzaud, Seawater accelerated ageing of poly(3-hydroxybutyrate-co-3-hydroxyvalerate), Polymer Degradation and Stability, 2014, 105 (0): [91] R. Muthuraj, M. Misra, A. K. Mohanty, Hydrolytic degradation of biodegradable polyesters under simulated environmental conditions. Journal of Applied Polymer Science. 2015, 132 (27), [92] (Last accessed in January 2015). [93] (Last accessed in January 2015). [94] (Last accessed in Janueary 2015) [95] (Last accessed in January 2015) [96] C. Koning, M. Van Duin, C. Pagnoulle, R. Jerome, Strategies for compatibilization of polymer blends, Progress in Polymer Science, 1998, 23 (4): [97] M. Shibata, Y. Inoue, M. Miyoshi, Mechanical properties, morphology, and crystallization behavior of blends of poly(l-lactide) with poly(butylene succinate-co-l-lactate) and poly(butylene succinate), Polymer, 2006, 47 (10):

109 [98] T. Yokohara, M. Yamaguchi, Structure and properties for biomass-based polyester blends of PLA and PBS, European Polymer Journal, 2008, 44 (3): [99] E. A. gelos Manias, L. A. Utracki, Chapter: 2 Thermodynamics of polymer blends. in Polymer blends handbook, (Eds): Utracki LA, Wilkie CA, Springer, Dordrecht, 2014, 1 (2nd ed), [100] E. Fekete, E. Földes, B. Pukánszky, Effect of molecular interactions on the miscibility and structure of polymer blends, European Polymer Journal, 2005, 41 (4): [101] Robeson LM. Polymer Blends: A Comprehensive Review. Munich: Carl Hanser Verlag, 2007, 1-7. [102] P. Szabó, E. Epacher, E. Földes, B. Pukánszky, Miscibility, structure and properties of PP/PIB blends, Materials Science and Engineering: A, 2004, 383 (2): [103] P. A. Small, Some factors affecting the solubility of polymers, Journal of Applied Chemistry, 1953, 3 (2): [104] I. Pillin, N. Montrelay, Y. Grohens, Thermo-mechanical characterization of plasticized PLA: Is the miscibility the only significant factor?, Polymer, 2006, 47 (13): [105] L. L. Zhang, S. H. Goh, S. Y. Lee, G. R. Hee, Miscibility, melting and crystallization behavior of two bacterial polyester/poly(epichlorohydrin-co-ethylene oxide) blend systems, Polymer, 2000, 41 (4): [106] J. W. Park, S. S. Im, Phase behavior and morphology in blends of poly(l-lactic acid) and poly(butylene succinate), Journal of Applied Polymer Science, 2002, 86 (3): [107] Z. Qiu, T. Ikehara, T. Nishi, Poly(hydroxybutyrate)/poly(butylene succinate) blends: miscibility and nonisothermal crystallization, Polymer, 2003, 44 (8):

110 [108] A. Bhatia, R. K. Gupta, S. N. Bhattacharya, H. Choi, Compatibility of biodegradable poly (lactic acid)(pla) and poly (butylene succinate)(pbs) blends for packaging application, Korea- Australia Rheology Journal, 2007, 19 (3): [109] M. Sheth, R. A. Kumar, V. Davé, R. A. Gross, S. P. McCarthy, Biodegradable polymer blends of poly(lactic acid) and poly(ethylene glycol), Journal of Applied Polymer Science, 1997, 66 (8): [110] K. Sungsanit, N. Kao, S. N. Bhattacharya, Properties of linear poly(lactic acid)/polyethylene glycol blends, Polymer Engineering and Science, 2012, 52 (1): [111] A. J. Nijenhuis, E. Colstee, D. W. Grijpma, A. J. Pennings, High molecular weight poly(llactide) and poly(ethylene oxide) blends: thermal characterization and physical properties, Polymer, 1996, 37 (26): [112] M. Avella, E. Martuscelli, Poly-d-( )(3-hydroxybutyrate)/poly(ethylene oxide) blends: phase diagram, thermal and crystallization behaviour, Polymer, 1988, 29 (10): [113] S. Ravati, B. D. Favis, Morphological states for a ternary polymer blend demonstrating complete wetting, Polymer, 2010, 51 (20): [114] Z. Qiu, W. Yang, T. Ikehara, T. Nishi, Miscibility and crystallization behavior of biodegradable blends of two aliphatic polyesters. Poly(3-hydroxybutyrate-co-hydroxyvalerate) and poly(ε-caprolactone), Polymer, 2005, 46 (25): [115] A. Javadi, A. J. Kramschuster, S. Pilla, J. Lee, S. Gong, L.-S. Turng, Processing and characterization of microcellular PHBV/PBAT blends, Polymer Engineering and Science, 2010, 50 (7):

111 [116] Z. Qiu, T. Ikehara, T. Nishi, Miscibility and crystallization behaviour of biodegradable blends of two aliphatic polyesters. Poly(3-hydroxybutyrate-co-hydroxyvalerate) and poly(butylene succinate) blends, Polymer, 2003, 44 (24): [117] M. R. Nanda, M. Misra, A. K. Mohanty, The Effects of Process Engineering on the Performance of PLA and PHBV Blends, Macromolecular Materials and Engineering, 2011, 296 (8): [118] F. Signori, M.-B. Coltelli, S. Bronco, Thermal degradation of poly(lactic acid) (PLA) and poly(butylene adipate-co-terephthalate) (PBAT) and their blends upon melt processing, Polymer Degradation and Stability, 2009, 94 (1): [119] L. Jiang, M. P. Wolcott, J. Zhang, Study of Biodegradable Polylactide/Poly(butylene adipate-co-terephthalate) Blends, Biomacromolecules, 2005, 7 (1): [120] V. Vilay, M. Mariatti, Z. Ahmad, K. Pasomsouk, M. Todo, Improvement of microstructures and properties of biodegradable PLLA and PCL blends compatibilized with a triblock copolymer, Materials Science and Engineering: A, 2010, 527 (26): [121] (Last accessed in January, 2015). [122] J. John, R. Mani, M. Bhattacharya, Evaluation of compatibility and properties of biodegradable polyester blends, Journal of Polymer Science Part A: Polymer Chemistry, 2002, 40 (12): [123] D. Wu, L. Yuan, E. Laredo, M. Zhang, W. Zhou, Interfacial Properties, Viscoelasticity, and Thermal Behaviors of Poly(butylene succinate)/polylactide Blend, Industrial and Engineering Chemistry Research, 2012, 51 (5):

112 [124] J. Zhou, X. Wang, K. Hua, C. e. Duan, W. Zhang, J. Ji, X. Yang, Enhanced mechanical properties and degradability of poly(butylene succinate) and poly(lactic acid) blends, Iranian Polymer Journal, 2013, 22 (4): [125] M. J. Jenkins, Y. Cao, L. Howell, G. A. Leeke, Miscibility in blends of poly(3- hydroxybutyrate-co-3-hydroxyvalerate) and poly(ɛ-caprolactone) induced by melt blending in the presence of supercritical CO2, Polymer, 2007, 48 (21): [126] P. Ma, A. B. Spoelstra, P. Schmit, P. J. Lemstra, Toughening of poly (lactic acid) by poly (β-hydroxybutyrate-co-β-hydroxyvalerate) with high β-hydroxyvalerate content, European Polymer Journal, 2013, 49 (6): [127] L. A. Utracki, Compatibilization of Polymer Blends, The Canadian Journal of Chemical Engineering, 2002, 80 (6): [128] K. Sirisinha, W. Somboon, Melt characteristics, mechanical, and thermal properties of blown film from modified blends of poly(butylene adipate-co-terephthalate) and poly(lactide), Journal of Applied Polymer Science, 2012, 124 (6): [129] M. Kumar, S. Mohanty, S. K. Nayak, M. Rahail Parvaiz, Effect of glycidyl methacrylate (GMA) on the thermal, mechanical and morphological property of biodegradable PLA/PBAT blend and its nanocomposites, Bioresource Technology, 2010, 101 (21): [130] N. Zhang, Q. Wang, J. Ren, L. Wang, Preparation and properties of biodegradable poly(lactic acid)/poly(butylene adipate-co-terephthalate) blend with glycidyl methacrylate as reactive processing agent, Journal of Materials Science, 2009, 44 (1): [131] S. Lin, W. Guo, C. Chen, J. Ma, B. Wang, Mechanical properties and morphology of biodegradable poly(lactic acid)/poly(butylene adipate-co-terephthalate) blends compatibilized by transesterification, Materials and Design, 2012, 36 (0):

113 [132] B. Y. Shin, D. H. Han, Compatibilization of immiscible poly(lactic acid)/poly(εcaprolactone) blend through electron-beam irradiation with the addition of a compatibilizing agent, Radiation Physics and Chemistry, 2013, 83 (0): [133] H. Liu, F. Chen, B. Liu, G. Estep, J. Zhang, Super Toughened Poly(lactic acid) Ternary Blends by Simultaneous Dynamic Vulcanization and Interfacial Compatibilization, Macromolecules, 2010, 43 (14): [134] H. T. Oyama, Super-tough poly(lactic acid) materials: Reactive blending with ethylene copolymer, Polymer, 2009, 50 (3): [135] M. Harada, T. Ohya, K. Iida, H. Hayashi, K. Hirano, H. Fukuda, Increased impact strength of biodegradable poly(lactic acid)/poly(butylene succinate) blend composites by using isocyanate as a reactive processing agent, Journal of Applied Polymer Science, 2007, 106 (3): [136] T. Takayama, M. Todo, Improvement of impact fracture properties of PLA/PCL polymer blend due to LTI addition, Journal of Materials Science, 2006, 41 (15): [137] N. C. Liu, W. E. Baker, Reactive polymers for blend compatibilization, Advances in Polymer Technology, 1992, 11 (4): [138] Y.-H. Na, Y. He, X. Shuai, Y. Kikkawa, Y. Doi, Y. Inoue, Compatibilization Effect of Poly(ε-caprolactone)-b-poly(ethylene glycol) Block Copolymers and Phase Morphology Analysis in Immiscible Poly(lactide)/Poly(ε-caprolactone) Blends, Biomacromolecules, 2002, 3 (6): [139] K. Chavalitpanya, S. Phattanarudee, Poly(Lactic Acid)/Polycaprolactone Blends Compatibilized with Block Copolymer, Energy Procedia, 2013, 34 (0):

114 [140] C.-H. Kim, K. Y. Cho, E.-J. Choi, J.-K. Park, Effect of P(lLA-co-εCL) on the compatibility and crystallization behavior of PCL/PLLA blends, Journal of Applied Polymer Science, 2000, 77 (1): [141] P. Ma, X. Cai, Y. Zhang, S. Wang, W. Dong, M. Chen, P. J. Lemstra, In-situ compatibilization of poly(lactic acid) and poly(butylene adipate-co-terephthalate) blends by using dicumyl peroxide as a free-radical initiator, Polymer Degradation and Stability, 2014, 102 (0): [142] A. J. Ryan, Polymer science: Designer polymer blends, Nature Material, 2002, 1 (1): [143] R. Fayt, P. Teyssie, Molecular design of multicomponent polymer systems. 8. New acrylonitrile-butadiene-styrene-like resins from styrene-acrylonitrile-based mechanical blends, Macromolecules, 1986, 19 (7): [144] R. Fayt, R. Jérôme, P. Teyssié, Molecular design of multicomponent polymer systems. III. Comparative behavior of pure and tapered block copolymers in emulsification of blends of lowdensity polyethylene and polystyrene, Journal of Polymer Science: Polymer Physics Edition, 1982, 20 (12): [145] D. Wu, Y. Zhang, L. Yuan, M. Zhang, W. Zhou, Viscoelastic interfacial properties of compatibilized poly(ε-caprolactone)/polylactide blend, Journal of Polymer Science Part B: Polymer Physics, 2010, 48 (7): [146] H. Tsuji, T. Yamada, M. Suzuki, S. Itsuno, Part 7. Effects of poly(l-lactide-co-ϵcaprolactone) on morphology, structure, crystallization, and physical properties of blends of poly(l-lactide) and poly(ϵ-caprolactone), Polymer International, 2003, 52 (2):

115 [147] M. Avella, M. E. Errico, P. Laurienzo, E. Martuscelli, M. Raimo, R. Rimedio, Preparation and characterisation of compatibilised polycaprolactone/starch composites, Polymer, 2000, 41 (10): [148] G. Ceccorulli, M. Scandola, G. Adamus, Compatibilizing effect of a graft copolymer on bacterial poly(3-hydroxybutyrate)/poly(methyl methacrylate) blends, Journal of Polymer Science Part B: Polymer Physics, 2002, 40 (13): [149] Y. Wang, M. A. Hillmyer, Polyethylene-poly(L-lactide) diblock copolymers: Synthesis and compatibilization of poly(l-lactide)/polyethylene blends, Journal of Polymer Science Part A: Polymer Chemistry, 2001, 39 (16): [150] E. Can, S. Bucak, E. Kınacı, A. C. Çalıkoğlu, G. T. Köse, Polybutylene Succinate (PBS) Polycaprolactone (PCL) Blends Compatibilized with Poly(ethylene oxide)-block-poly(propylene oxide)-block-poly(ethylene oxide) (PEO-PPO-PEO) Copolymer for Biomaterial Applications, Polymer-Plastics Technology and Engineering, 2014, 53 (11): [151] G.-X. Chen, H.-S. Kim, E.-S. Kim, J.-S. Yoon, Compatibilization-like effect of reactive organoclay on the poly(l-lactide)/poly(butylene succinate) blends, Polymer, 2005, 46 (25): [152] V. Ojijo, S. Sinha Ray, R. Sadiku, Toughening of Biodegradable Polylactide/Poly(butylene succinate-co-adipate) Blends via in Situ Reactive Compatibilization, ACS Applied Materials & Interfaces, 2013, 5 (10): [153] A. K. Mohanty, M. Misra, G. Hinrichsen, Biofibres, biodegradable polymers and biocomposites: An overview, Macromolecular Materials and Engineering, 2000, (1):

116 [154] W. Dong, P. Ma, S. Wang, M. Chen, X. Cai, Y. Zhang, Effect of partial crosslinking on morphology and properties of the poly(β-hydroxybutyrate)/poly(d,l-lactic acid) blends, Polymer Degradation and Stability, 2013, 98 (9): [155] K. Zhang, V. Nagarajan, M. Misra, A. K. Mohanty, Supertoughened Renewable PLA Reactive Multiphase Blends System: Phase Morphology and Performance, ACS Applied Materials and Interfaces, 2014, 6 (15): [156] L. Wang, W. Ma, R. A. Gross, S. P. McCarthy, Reactive compatibilization of biodegradable blends of poly(lactic acid) and poly(ε-caprolactone), Polymer Degradation and Stability, 1998, 59 (1 3): [157] M. P. M. Dicker, P. F. Duckworth, A. B. Baker, G. Francois, M. K. Hazzard, P. M. Weaver, Green composites: A review of material attributes and complementary applications, Composites Part A: Applied Science and Manufacturing, 2014, 56 (0): [158] V. Nagarajan, A. K. Mohanty, M. Misra, Sustainable Green Composites: Value Addition to Agricultural Residues and Perennial Grasses, ACS Sustainable Chemistry and Engineering, 2013, 1 (3): [159] F. Ahmad, H. S. Choi, M. K. Park, A Review: Natural Fiber Composites Selection in View of Mechanical, Light Weight, and Economic Properties, Macromolecular Materials and Engineering, 2015, 300 (1): [160] M.A. Al-Maadeed, S. Labidi, Chapter 4: Recycled polymers in natural fiber-reinforced polymer composites. In: Hodzic, A., Shanks, R. (Eds.), Natural Fibre Composites: Materials, Processes and Properties. Woodhead Publishing Limited, Cambridge, UK, 2014, pp [161] D. Shah, Developing plant fibre composites for structural applications by optimising composite parameters: a critical review, Journal of Materials Science, 2013, 48 (18):

117 [162] A. K. Mohanty, M. Misra, L. T. Drzal, Sustainable Bio-Composites from Renewable Resources: Opportunities and Challenges in the Green Materials World, Journal of Polymers and the Environment, 2002, 10 (1-2): [163] D. U. Shah, Natural fibre composites: Comprehensive Ashby-type materials selection charts, Materials and Design, 2014, 62 (0): [164] P. Wambua, J. Ivens, I. Verpoest, Natural fibres: can they replace glass in fibre reinforced plastics?, Composites Science and Technology, 2003, 63 (9): [165] H. M. Akil, M. F. Omar, A. A. M. Mazuki, S. Safiee, Z. A. M. Ishak, A. Abu Bakar, Kenaf fiber reinforced composites: A review, Materials and Design, 2011, 32 (8 9): [166] M. R. Nanda, M. Misra, A. K. Mohanty, Performance Evaluation of Biofibers and Their Hybrids as Reinforcements in Bioplastic Composites, Macromolecular Materials and Engineering, 2013, 298 (7): [167] A. Bourmaud, S. Pimbert, Investigations on mechanical properties of poly(propylene) and poly(lactic acid) reinforced by miscanthus fibers, Composites Part A: Applied Science and Manufacturing, 2008, 39 (9): [168] V. Nagarajan, M. Misra, A. K. Mohanty, New engineered biocomposites from poly(3- hydroxybutyrate-co-3-hydroxyvalerate) (PHBV)/poly(butylene adipate-co-terephthalate) (PBAT) blends and switchgrass: Fabrication and performance evaluation, Industrial Crops and Products, 2013, 42 (0): [169] F. M. Al-Oqla, S. M. Sapuan, Natural fiber reinforced polymer composites in industrial applications: feasibility of date palm fibers for sustainable automotive industry, Journal of Cleaner Production, 2014, 66 (0):

118 [170] M. Avella, A. Buzarovska, M. Errico, G. Gentile, A. Grozdanov, Eco-Challenges of Bio- Based Polymer Composites, Materials, 2009, 2 (3): [171] M. U. Wahit, N. I. Akos, W. A. Laftah, Influence of natural fibers on the mechanical properties and biodegradation of poly(lactic acid) and poly(ε-caprolactone) composites: A review, Polymer Composites, 2012, 33 (7): [172] K. G. Satyanarayana, G. G. C. Arizaga, F. Wypych, Biodegradable composites based on lignocellulosic fibers-an overview, Progress in Polymer Science, 2009, 34 (9): [173] A. Javadi, Y. Srithep, J. Lee, S. Pilla, C. Clemons, S. Gong, L.-S. Turng, Processing and characterization of solid and microcellular PHBV/PBAT blend and its RWF/nanoclay composites, B: Applied Science and Manufacturing, 2010, 41 (8): [174] F. Harnnecker, D. dos Santos Rosa, D. Lenz, Biodegradable Polyester-Based Blend Reinforced with Curauá Fiber: Thermal, Mechanical and Biodegradation Behaviour, Journal of Polymers and the Environment, 2012, 20 (1): [175] K. Zhang, A. K. Mohanty, M. Misra, Fully Biodegradable and Biorenewable Ternary Blends from Polylactide, Poly(3-hydroxybutyrate-co-hydroxyvalerate) and Poly(butylene succinate) with Balanced Properties, ACS Applied Materials and Interfaces, 2012, 4 (6): [176] N. Zarrinbakhsh, A. Mohanty, M. Misra, Improving the interfacial adhesion in a new renewable resource-based biocomposites from biofuel coproduct and biodegradable plastic, Journal of Materials Science, 2013, 48 (17): [177] S. Muniyasamy, M. M. Reddy, M. Misra, A. Mohanty, Biodegradable green composites from bioethanol co-product and poly(butylene adipate-co-terephthalate), Industrial Crops and Products, 2013, 43 (0):

119 [178] H. Lu, S. A. Madbouly, J. A. Schrader, G. Srinivasan, K. G. McCabe, D. Grewell, M. R. Kessler, W. R. Graves, Biodegradation Behavior of Poly(lactic acid) (PLA)/Distiller s Dried Grains with Solubles (DDGS) Composites, ACS Sustainable Chemistry and Engineering, 2014, 2 (12): [179] S. A. Madbouly, J. A. Schrader, G. Srinivasan, K. Liu, K. G. McCabe, D. Grewell, W. R. Graves, M. R. Kessler, Biodegradation behavior of bacterial-based polyhydroxyalkanoate (PHA) and DDGS composites, Green Chemistry, 2014, 16 (4): [180] C. Nyambo, V. Nagarajan, A. K. Mohanty, M. Misra, Chapter 8: natural fiber composites from agricultural by-products: an overview. In: A. N. Netravali, C.M. Pastore, (Eds.), Sustainable Composites: Fibers, Resins and Applications. DEStech Publications, Inc, Lancaster, Pennsylvania, 2014, pp [181] R. Pradhan, M. Misra, L. Erickson, A. Mohanty, Compostability and biodegradation study of PLA wheat straw and PLA soy straw based green composites in simulated composting bioreactor, Bioresource Technology, 2010, 101 (21): [182] A. K. Bledzki, A. Jaszkiewicz, Mechanical performance of biocomposites based on PLA and PHBV reinforced with natural fibres A comparative study to PP, Composites Science and Technology, 2010, 70 (12): [183] J. Adam, B. A. Korneliusz, M. Agnieszka, Dynamic mechanical thermal analysis of biocomposites based on PLA and PHBV-A comparative study to PP counterparts, Journal of Applied Polymer Science, 2013, 130 (5): [184] B. Yang, M. Nar, D. K. Visi, M. Allen, B. Ayre, C. L. Webber Iii, H. Lu, N. A. D Souza, Effects of chemical versus enzymatic processing of kenaf fibers on poly(hydroxybutyrate-co- 85

120 valerate)/poly(butylene adipate-co-terephthalate) composite properties, Composites Part B: Engineering, 2014, 56 (0): [185] N. Zarrinbakhsh, M. Misra, A. K. Mohanty, Biodegradable Green Composites from Distiller's Dried Grains with Solubles (DDGS) and a Polyhydroxy(butyrate-co-valerate) (PHBV)-Based Bioplastic, Macromolecular Materials and Engineering, 2011, 296 (11): [186] K. Zhang, M. Misra, A. K. Mohanty, Toughened Sustainable Green Composites from Poly(3-hydroxybutyrate-co-3-hydroxyvalerate) Based Ternary Blends and Miscanthus Biofiber, ACS Sustainable Chemistry and Engineering, 2014, 2 (10): [187] M. R. Nanda, M. Misra, A. K. Mohanty, Mechanical Performance of Soy-Hull-Reinforced Bioplastic Green Composites: A Comparison with Polypropylene Composites, Macromolecular Materials and Engineering, 2012, 297 (2): [188] A. Sis, N. Ibrahim, W. Yunus, Effect of (3-aminopropyl)trimethoxysilane on mechanical properties of PLA/PBAT blend reinforced kenaf fiber, Iranian Polymer Journal, 2013, 22 (2): [189] E. Sykacek, W. Schlager, N. Mundigler, Compatibility of softwood flour and commercial biopolymers in injection molding, Polymer Composites, 2010, 31 (3): [190] P. Georgiopoulos, E. Kontou, M. Niaounakis, Thermomechanical properties and rheological behavior of biodegradable composites, Polymer Composites, 2014, 35 (6): [191] T. Yu, Y. Li, Influence of poly(butylenes adipate-co-terephthalate) on the properties of the biodegradable composites based on ramie/poly(lactic acid), Composites Part A: Applied Science and Manufacturing, 2014, 58 (0):

121 [192] A. Ibrahim, M. Wahit, A. Yussuf, Effect of fiber reinforcement on mechanical and thermal properties of poly(ɛ-caprolactone)/poly(lactic acid) blend composites, Fibers and Polymers, 2014, 15 (3): [193] B. K. Goriparthi, K. N. S. Suman, M. R. Nalluri, Processing and characterization of jute fiber reinforced hybrid biocomposites based on polylactide/polycaprolactone blends, Polymer Composites, 2012, 33 (2): [194] A. Bourmaud, Y.-M. Corre, C. Baley, Fully biodegradable composites: Use of poly- (butylene-succinate) as a matrix and to plasticize l-poly-(lactide)-flax blends, Industrial Crops and Products, 2015, 64, [195] C. N. Ludvik, G. M. Glenn, A. P. Klamczynski, D. F. Wood, Cellulose Fiber/Bentonite Clay/Biodegradable Thermoplastic Composites, Journal of Polymers and the Environment, 2007, 15 (4): [196] O. Faruk, A. K. Bledzki, H.-P. Fink, M. Sain, Progress Report on Natural Fiber Reinforced Composites, Macromolecular Materials and Engineering, 2014, 299 (1): [197] (Last accessed in January, 2015). [198] (Last accessed in January, 2015). [199] M. Darder, P. Aranda, E. Ruiz-Hitzky, Bionanocomposites: A New Concept of Ecological, Bioinspired, and Functional Hybrid Materials, Advanced Materials, 2007, 19 (10): [200] (Last accessed in January, 2015). 87

122 Chapter 3: Fully Biodegradable Poly (butylene succinate) and Poly (butylene adipate-coterephthalate) Blends: Reactive Extrusion and Performance Evaluation* *A version of this chapter has been published in: R. Muthuraj, M. Misra, A. K. Mohanty, Biodegradable poly (butylene succinate) and poly (butylene adipate-co-terephthalate) blends: Reactive extrusion and performance evaluation, Journal of Polymers and the Environment, 2014, 22: (adapted with kind permission from Springer, Jul 09, 2015, License number ). Abstract Two biodegradable polyesters, poly(butylene succinate) (PBS), and poly(butylene adipate-co-terephthalate) (PBAT) were melt-compounded in a twin screw extruder to fabricate a novel PBS/PBAT blends. The compatibility of the blends was attributed to the transesterification reaction that was confirmed by Fourier transform infrared spectroscopy. The Gibbs free energy equation was applied to demonstrate the miscibility of resulting blends. Dynamic mechanical analysis of the blends exhibits an intermediate tan δ peak compared to the individual components which suggests that the blend achieved compatibility. One of the key findings is that the tensile strength of the optimized blends is higher compared to blended partner. Rheological properties revealed a strong shear-thinning tendency of the blend by the addition of PBAT into PBS. The phase morphologies of the blends were examined through scanning electron microscopy, which revealed that phase separation occurred in the blends. The spherulite growth in the blends was highly influenced by the crystallization temperature and composition. In addition, the presence of a dispersed amorphous phase was found to be a hindrance to the spherulite growth, which was confirmed by polarizing optical microscopy. Furthermore, the increased crystallization ability of PBAT in the blend systems gives the blend a balanced thermal resistance property. 88

123 3.1 Introduction The development of biodegradable material as a potential substitute for non-biodegradable material is an emerging field of research and development. In recent years, different types of biodegradable polymers have received an immense amount of attention for developing various new materials and to reduce environmental concerns [1]. Some biodegradable polymers are commercially available in the market, such as poly (propylene carbonate) (PPC), poly (butylene adipate-co-terephthalate) (PBAT), polyhydroxyalkanoates (PHAs), polycaprolactone (PCL), poly (lactic acid) (PLA), poly (butylene succinate) (PBS), and thermoplastic starch [2-5]. Biodegradable polymers are not currently widely used due to some limitations such as their cost, mechanical properties, and thermal stability. Researchers have been trying to address these issues by utilizing blending techniques to obtain biodegradable blends with tailored properties. Compared to other methods, melt blending is a cost effective and less time-consuming process for the development of new material with balanced properties [6,7]. During melt blending, dipole interactions, hydrogen bonding, or a combination of these occur naturally, and such interactions can enhance the overall performance of the resulting products [8,9]. The modification can also be made during processing to further improve the strength of the material. Well established literature is available for modification studies of polymer blends using techniques such as in-situ compatibilization [10], graft copolymerization [11], copolymerization [12], and transreactions [8, 13]. Transreactions include an alcoholysis, acidolysis, and ester interchange reaction. These three reactions are generally referred to as transesterification. The transesterfication reaction is an exchange mechanism which can help to form a new type of ester linkage between the components in the blends [14]. The resulting transesterification products very often play important roles in the miscibility, compatibility, crystallinity, and mechanical properties of the blends. During the past several 89

124 years of research, many studies have been done on the transesterification of polyester blends such as PBS/PCL [6], poly(ethylene terephthalate)/poly(ether imide) [15], poly(triethylene terephthalte)/polycarbonate [16], PHB/PLA [17], and polycarbonate/poly(trimethylene terephthalate) [18]. Among the biodegradable polyesters, both PBS and PBAT have been widely studied because of their commercial availability and inherent biodegradability [1, 19]. PBS is an aliphatic polyester which is synthesized from the polycondensation reaction of petroleum based aliphatic dicarboxylic acid (succinic acid) and 1,4-butane diol [20-22]. The biodegradability of PBS is similar to cellulose and bacterial polyesters like poly(hydroxybutyrate-co-valerate) (PHBV) [23]. Only limited biodegradable thermoplastics are produced from renewable resources. Recently, the PBS is produced from biobased monomer i.e., bio-succinic acid [24]. Therefore, exploring PBS as a matrix system for polymer blends and composites could diversify the renewable resource based material applications. PBS is a good candidate for making biodegradable products as well as having some unique physical properties such as semicrystalline nature, thermal stability, good processing properties, good gas barrier properties, and a lower melting point [25-28]. The main drawback of the PBS is low impact strength, which is limiting its applications. PBAT is commercially synthesized from petroleum based adipic acid, 1,4-butane diol, and terephthalic acid, which is a good biodegradable polymer in the presence of naturally occurring microorganisms [29-31]. Furthermore, it has excellent toughness and is mostly used for film extrusion and coatings [32]. PBAT is a promising material to improve the toughness of polymer blends which contain brittle polymers like poly(lactic acid) [33], polycarbonate [34], and poly(hydroxybutyrate-co-valerate) [35]. As noted above, PBS and PBAT are the most 90

125 promising candidates for future biodegradable materials in various potential applications. Many studies have reported the blending of either PBS or PBAT with other biodegradable polymers. For instance, PBS has been incorporated with many polymers such as poly(triethylene succinate) [21], poly(ethylene oxide) [36], poly(propylene carbonate) [37], poly(butylene terephthalate) [28], copolyesters [30, 38], polyhydroxybutyrate [27], and polycaprolactone (PCL) [6]. Although many studies have reported stiffness-toughness balanced biodegradable binary blends, to the best of our knowledge no literature is available for PBS/PBAT binary blends. As two typical thermoplastic biodegradable polyesters, the blend of PBS and PBAT are of great interest due to their unique properties, which can extend their applications in diversified areas. Hence, the present work focused on the fabrication of a novel high performance PBS/PBAT blend. The binary blend was fabricated by an extrusion-injection molding technique. The resulting binary blend was characterized by different analytical techniques. 3.2 Experimental section Materials PBS pellets (Bionolle 1020) with a weight average molecular weight (M w ) of g/mol and PDI of 1.82, manufactured by Showa Highpolymer, Japan, and were used. The commercially available PBAT (Biocosafe 2003F) was purchased from Xinfu Pharmaceutical, China. The molecular structures of the neat PBS and PBAT are shown in Scheme

126 Scheme 3.1. Molecular structure of PBS and PBAT Blend preparation Prior to blending, PBS and PBAT pellets were vacuum dried at 80 o C for 12 h. Samples with different compositions of PBS/PBAT were prepared in a DSM Xplore micro-compounder. The micro-extruder was equipped with a co-rotating twin screw and had a barrel volume of 15 cm 3. A twin-screw aspect ratio of 18 and length of 150 mm was used in the melt blending process. The process temperature, cycle time, and screw speed were kept constant at 140 o C, 2 min, and 100 rpm, respectively for different compositions of the PBS/PBAT blend. The molten polymer was collected and injected into the mold at 30 o C using a 12 cm 3 micro-injection molder (DSM Xplore ) kept at 140 o C. The molded test specimens were used for further characterization Fourier transform infrared spectroscopy (FTIR) FTIR spectra of PBS, PBAT and PBS/PBAT blends were obtained by using a Thermo Scientific Nicolet TM 6700 at ambient temperature. All the scans were performed from 400 to 4000 cm -1 with a resolution of 4 cm -1. For each sample of the spectrum, 32 accumulated scans were produced and the absorbance was recorded as a function of wavenumbers Mechanical properties Tensile properties of the neat PBS, PBAT and PBS/PBAT blends were obtained by Instron 3382, using a constant crosshead speed of 50 mm/min at room temperature. The tensile properties were measured according to ASTM D638 test method using dumbbell shaped 92

127 samples. Data was collected by Blue hill software. All the reported values are an average of at least five samples for each formulation Melt flow index (MFI) MFI measurement of the neat polymers and their blends was carried out according to the ASTM D1238 standard using a Qualitest (model 2000A) Melt Flow Indexer at 190 C with a standard weight of 2.16 kg. The presented results are an average of five replicates of each formulation Differential scanning calorimetry (DSC) DSC analysis was carried out in a thermal analysis instrument (TA Q-200), and the analysis was performed under the nitrogen atmosphere. The sample was first scanned from 25 to 150 o C with a rate of heating 10 o C min -1 and subsequently cooled down from 150 to -50 o C at a cooling rate of 5 o C min -1. A second heating scan of the samples was performed from -50 to 150 o C at a heating rate of 10 o C min -1. The first heating cycle was used for the removal of thermal history and the reported results are from the second heating cycle. The data were analyzed using TA Instrument software Dynamic mechanical analysis (DMA) The storage modulus and tan delta of PBS, PBAT and PBS/PBAT blends were measured by a DMA Q800 from TA Instruments. The analysis was performed between -50 to 100 o C at a heating rate of 3 o C/min. The experiment was carried out in a dual cantilever clamp with 1 Hz frequency and 15 µm oscillating amplitude. 93

128 3.2.8 Heat deflection temperature (HDT) HDT measurement was performed based on the ASTM D648 standard at a constant load MPa in the same DMA Q800 thermal instrument. The analysis was performed at a heating rate of 2 o C/min from ambient temperature to 100 o C in a three point bending mode Thermogravimetric analysis (TGA) Thermal stability of the PBS, PBAT and PBS/PBAT blends was measured using a TA Q500 Instrument. The experiment was carried out under the nitrogen environment from 25 to 600 o C with a rate of heating 20 o C min -1. The maximum rate of degradation was observed from the derivative thermogram (DTG) Rheological studies A strain-controlled rheometer (Anton Paar Modular Compact Rheometer MCR 302) was used to observe the rheological properties of neat polymers and their blends. Injection molded samples were placed between the parallel-plates (diameter of the parallel plate is 25 mm), and the experiment was performed at 140 o C. The plates subsequently compressed the samples, and the distance between the parallel plates was adjusted to 1 mm. Dynamic properties were determined by a dynamic frequency sweep test. During the test, the range of frequency and strain used was 500 to 0.01 rad/s and 3 %, respectively. These limits were fixed based on the polymer torque sensitivity and their thermal stability Polarizing optical microscopy (POM) Polarizing optical microscopy was performed on a Nikon, Universal Design Microscope. The microscope was equipped with a Linkam LTS 420 hot stage, which is used to control the temperature. A DS-2Mv (with DS-U 2 ) color video camera with the capture NIS element imaging software was used for POM observations. Samples were sandwiched between two glass slides. 94

129 Subsequently, samples were annealed at the crystallization temperature. The spherulities growth was observed at two different crystallization temperatures of 80 and 90 o C Scanning electron microscopy (SEM) Morphology of the PBS/PBAT blends was captured by an Inspect S50-FEI Company. The cryofractured samples were used to observe the phase morphology of the blends. A selective dissolution of polyester in tetrahydrofuran (THF) was used to distinguish the polymer phases. All the samples were dried and sputtered with gold prior to imaging in order to make them conductive. 3.3 Results and Discussion Fourier transform infrared spectroscopy (FTIR) The FTIR analysis was performed to identify the physical and chemical interaction between the neat polymers during melt blending. The FTIR spectra of PBS, PBAT and PBS/PBAT blends are shown in Figure 3.1. The carbonyl group frequency of neat polymers and their blends was detected at 1712 cm -1 and 1716 cm -1, respectively. The peak of the carbonyl group was shifted towards higher wavenumbers for the PBS/PBAT blends compared to each of the neat polymers, which clearly indicates that a strong chemical interaction occurred during the melt process at 140 o C. Kwei [39] has reported that the shift (from 1722 to about 1705 cm -1 ) of the carbonyl group peak in the blends occurred as a result of chemical interactions between the parent polymers. John et al., [6] also observed a similar type of transesterification reaction in the PBS/PCL and PCL/EASTAR blends. These results can be explained by the formation of copolyester of PBS and PBAT by an ester-ester interchange reaction. The resulting copolyester which is compatible with the homopolymer of the unreacted PBS and PBAT may act as a compatibilizer in the blend system. In the present study, no external transesterification catalyst 95

130 was added into the blends. Even though a transesterification reaction was observed during melt blending, the amount of reaction gradually decreased with increasing PBAT wt%. This resultant ester exchange reaction was due to the residual catalysts existing in the homopolymer synthesis, which was supported by Wang et al., [40-42]. The transesterification product can further enhance the mechanical performance of the resultant blends. Scheme 3.2 shows the expected chemical structure of the transesterification product in the PBS/PBAT blends. Figure 3.1. Evaluation of the normalized FTIR spectra of the carbonyl region ( cm -1 ) of PBS, PBAT and their blends 96

131 Scheme 3.2. Expected transesterification product of PBS/PBAT blend Mechanical properties Figure 3.2 shows the stress-strain curves of PBS, PBAT, and their blends. Neat PBS showed a higher tensile yield strength but lower elongation compared to neat PBAT. For PBS, no apparent strain hardening was observed during the tensile test. On the other hand, PBAT showed excellent elongation and obvious strain hardening regions in the stress-strain curves, while its tensile and yield strength was poor. As for the blends, all the samples presented three clear regions such as elastic, plastic deformation, and strain hardening. The first region showed linear stretching with recoverable deformation, followed by the second region which revealed plastic deformation which is a non recoverable deformation of the samples. The second region indicated cold drawing behavior after the neck forming occurred in the samples. The third region showed strain hardening, the tensile stress gradually increasing until the samples broke. Crystalline slippage was also observed. Interestingly, after the strain increased to 150%, each composition of the blend showed clear evidence of cold drawing which affected the polymer chain alignment and resulted in strain hardening. Generally, amorphous and semi crystalline 97

132 polymer chain entanglement can lead to strain hardening. Strain hardening behavior is of great importance in polymer processing such as film blowing, thermoforming, and blow molding due to its good resistance against stretching of polymer segments. Also, strain hardening behavior can make the process easier and lead to higher quality products [43]. Figure 3.3 shows tensile strength and percentage elongation data. A significant property improvement was observed in the blends. The tensile strength of the PBS/PBAT (70/30 wt%) blend increased by 30% and 148% over the neat PBS and PBAT, respectively. The percent elongation of PBS/PBAT (70/30 wt%) blend was 150% higher than neat PBS. Tensile strength improvement is directly related to the intermolecular forces, crystallinity, miscibility, compatibility, and molecular orientation of the polymers in the blend [6]. In the present study, tensile strength improvement was directly related to the amount of PBS present in the blend. Figure 3.2. Tensile stress-strain curves of PBS, PBAT, and their blends 98

133 Tensile strength gradually reduced with increasing PBAT concentration in the blend system, which reveals that the blend transesterification ability was reduced. Furthermore, the increased PBAT content in the blend system may cause phase separation because of decreased compatibility in the blend which can lead to the reduction of tensile strength. These results were also observed from morphological analysis of the blends. John et al., [6] observed that a similar phase separation occurs when increasing one component in a PBS/EASTAR and PCL/EASTAR blend leading to reduced tensile strength in the blends. Tensile properties of the PBS/PBAT blends were sharply increased compare to their parent polymers. Apparently, the PBS/PBAT blends mechanical properties are comparable with literature polyethylene mechanical properties [44]. Therefore, we believe that the PBS/PBAT blends can be potential substitute for nobiodegradable polymers in the packaging applications. Figure 3.3. Tensile strength and elongation at break of PBS, PBAT, and their blends: (A) PBS; (B) PBS/PBAT(70/30 wt%); (C) PBS/PBAT (60/40 wt%); (D) PBS/PBAT (50/50 wt%); (E) PBAT. 99

134 3.3.3 Melt flow index MFI measurement is a common technique for studying the flow behavior of the polymers [45]. Table 3.1 shows the melt flow index values of neat polymers and their blends. The MFI values of extruded PBAT and PBS were 9 g/10min and 25 g/10min, respectively. After blending both polymers, MFI of all the blends increased compared to the neat polymers. The reduction of molecular weight, and changing thermal properties during the melt blending may be responsible for this [3]. PBS/PBAT (70/30 wt%) blend had the highest melt flow rate compared to neat polymers and other PBS/PBAT blends. The observed MFI improvement of the blends is probably attributed to the residual catalyst in the PBS system which is accelerating the PBAT degradation during MFI measurement. Possible degradation can be occurring by chain scission of the polymers, depolymerization, oxidative degradation, and transesterification reactions. In addition, reactive end groups, residual catalyst, unreacted starting monomers in the polymers, and other impurities can accelerate the thermal degradation of the polymers. A similar observation has been observed in the literature [46]. Increased MFI of the blends indicate better flow behavior compared to neat polymers. Therefore, this blend system is suitable to use as a new matrix system for polymer composites. Table 3.1. Melt flow index (MFI) of the neat polymers and their blends Samples MFI (g/10min) Neat PBS 25.3±2.4 PBS/PBAT (70/30 wt%) 41.5±3.2 PBS/PBAT (60/40 wt%) 33.3±2.8 PBS/PBAT (50/50 wt%) 33.6±1.6 Neat PBAT 9.4±

135 3.3.4 Differential scanning calorimetry The thermal behaviors of the neat PBS, PBAT, and their blends were measured through non-isothermal DSC analysis. Non-isothermal DSC results of PBS, PBAT, and PBS/PBAT blends are shown in Figure 3.4. The PBS and PBAT showed melting points at 114 and 116 o C, respectively. Interestingly, the second heating cycle showed a double melting peak for neat PBS and their blends due to the melt re-crystallization of the polymers. The second heating cycle DSC result is given in Figure 3.4, the imperfect crystals melt at lower temperatures but the more structurally perfect crystals melt at higher temperature [3]. Another possible reason may be that the molecular weight distribution could also affect the melt of the polymers [47]. The PBS/PBAT blend showed a similar melting behavior to that of PBS. The enthalpy of fusion for neat PBS and PBAT was found to be 32 and 10 J/g, respectively. The blends showed a high enthalpy of fusion compared to the neat polymers. The PBAT phase may act as a nucleating agent for the PBS phase, which will improve the crystallization of PBS in the blend. Another reason is a change in the regular structure of the interchange reaction product, which may lead to the production of thicker lamellar crystals that melt with a higher enthalpy of fusion [6]. In all the blends, single glass transition temperature (T g ) was observed because the T g values of both the neat polymers are very close to each other. Therefore, the values may be overlapping. The T g value of the blends shifted to lower temperatures compared with that of the neat polymer. A similar observation was found through DMA analysis. This variation in T g could be the cause of an interchange reaction which occurs during the melt blending process and it is also evidence for compatibility of the polymer in the blends. John et al., [6] have observed similar synergistic effects in PBS/PCL blends. Miscibility of the binary polymer blends can be predicted from Gordon- Taylor (G -T) equation (3.1) [48-49] 101

136 Figure 3.4. Second heating DSC thermograms of PBS, PBAT, and their blends after cooling at 5 C/min T g = (3.1) where W 1 and T g1 are the weight fraction and glass transition temperature of PBAT, respectively. The W 2 and T g2 are the weight fraction and glass transition temperature of PBS, respectively, and the parameter k is the fitting constant. The T g values are observed from the DSC analysis. If k=1, the Gordon and Taylor theory represents a good interaction between two blended components. If the k value is lower or higher than 1, it indicates poor interaction between the components in the blend. Figure 3.5 shows the T g value obtained from the Gordon-Taylor equation. From the DSC, the T g values observed for all the blends were -35 o C, close to the G -T curve. This indicates that the theoretical and experiment T g values are close to each other. In our present study, the k value was found to be 0.98; this semi-quantitatively measured value can further support the interaction, which occurred between the polymers in the blends. The k value 102

137 of the diglycidyl ether of bisphenol-a/poly(ethylene terephthalate) blend was 0.10; the small value of k suggests that only weak interactions exist between the components in the blend [8, 50]. Richard et al., [50] observed a k value of 0.18 for the PLA/PHBV blend indicating poor miscibility, which was confirmed through SEM analysis. These results are supportive of our present work: that the components have a good compatibility in the blend system. Miscibility and compatibility of the blends can be explained by Gibbs free energy. Thermodynamically compatible PBS/PBAT blends were analyzed according to the Gibbs free energy equation (3.2) [51]: G m = H m -T ( S c m+ S e m) (3.2) where G m is Gibbs free energy, T is absolute temperature, H m is heat of mixing, S c m is mixing of combinatorial entropy and S e m is the mixing of excess entropy. The molar volume of the components inversely depends on the combinatorial entropy. Hence, molecular weight of the polymers is directly related to combinatorial entropy. If the polymers have higher molecular weight, the S c m becomes zero. Therefore, the system is spontaneous; it could lead to a G m which is less than zero while H m is less than zero. In practical fields, this is rarely possible and so can be ignored. The H m is calculated using expression (3.3) [51]: H m = ( ) 2 (3.3) where and are the solubility parameter values of PBS and PBAT; and are the volume fraction of PBS and PBAT. The solubility parameter ( ) value of the PBS and PBAT was calculated according to the expression (3.4) [51]: = (3.4) 103

138 Where M, G and are the monomer molecular weight, group molar attraction constant of the polymer, and density of the polymer, respectively. The group molar attraction constant was calculated by Mark [52]. The Gibbs free energy and solubility parameter values for PBS and PBAT were calculated by equation 3.3 and 3.4, and the values are given in Table 3.2. Gibbs free energy values for PBS/PBAT blends are low and very close to each other, indicating that some extent of compatibility was achieved in the blend system. Previous studies have reported similar observations for some of the biopolymer blends such as PLA/PCL, PLA/PHBV, and PHBV/PCL blends, and they too have reported slight miscibility was achieved in their melt blend process [7, 51, 53]. Table 3.2. Solubility parameter values for polymers Samples Group Molar attraction Solubility Parameter Gibbs free energy constant G G m (J 1/2 cm 3/2 mol -1 ) (J 1/2 cm 3/2 ) (J g -1 m -3 ) PBS PBAT PBS/PBAT (70/30 wt%) PBS/PBAT (60/40 wt%) PBS/PBAT (50/50 wt%)

139 Figure 3.5. Theoretical and experimental values of T g for PBS/ PBAT blends Dynamic mechanical analysis Figure 3.6 shows the storage modulus of neat PBS, PBAT, and their blends. The storage modulus value of the PBS and PBAT at room temperature was found to be 0.6 and 0.1 GPa, respectively. PBS had a higher storage modulus compared to PBAT at all temperatures, and their blends had values in between the PBS and PBAT. A similar trend was observed in the tensile and flexural modulus values. Reduction in modulus while increasing temperature is attributed to increasing polymer chain mobility. Generally, above the alpha transition temperature, molecular motion increases and polymer segments move from glassy to a rubbery state, which is accompanied by an increase in the molecular relaxation in the polymers [54]. 105

140 Figure 3.6. Storage moduli of PBS, PBAT, and their blends Figure 3.7 depicts the Tan curves of PBS, PBAT, and their blends. It shows the primary and secondary transition peaks in neat PBAT at -20 and 62 o C, respectively. The primary transition peak corresponds to the poly(butylene adipate) segment mobility, and the secondary transition peak corresponds to the terephthalate unit mobility [55]. The T g value of the PBS, PBAT and PBS/PBAT blends was calculated from the maximum height of the Tan peak. Generally, an incompatible blend shows two transition peaks which correspond to the T g of individual components in the system [56]. For a highly compatible and partially compatible blend, a single transition peak can be seen lying between the transition temperature of individual components with an increased broadness in the transition peak [9, 57]. In our present study, all the blends were observed to have a single transition peak. The T g values of the PBS/PBAT blends were shifted towards lower temperatures compared to the neat PBS, which is a dilution effect with the addition of PBAT into PBS. Another possible reason is that for partially or completely compatible blends, the T g shifts towards lower or higher temperatures as a function 106

141 of composition [56]. Moreover, the small variation in the T g value shows further evidence of an interchange reaction occurring between the neat homopolymers. The T g shift was observed by the influence of a transesterification reaction when polycarbonate was incorporated into the poly(trimethylene terephthalate) [56]. Figure 3.7. Tan curves of PBS, PBAT, and their blends Heat deflection temperature HDT represents the maximum working temperature of materials and is defined as the temperature at which a material will be deformed by 250 µm under a constant load of MPa [58]. The HDT value of the neat polymers and their blends is shown in Table 3.3. The HDT value of the neat PBS and PBAT is 88 and 46 o C, respectively. In general, the HDT of amorphous polymers is low, and close to their glass transition temperature. In the crystalline polymers, the HDT is close to its melting point [7, 59]. In the present study, a balanced HDT value of PBS/PBAT blends was observed due to the PBAT having a lower crystallinity and thermal resistance compared to PBS. 107

142 Table 3.3. Heat deflection temperatures of the neat polymers and their blends Samples HDT ( o C) Neat PBS 88.06±0.4 PBS/PBAT (70/30 wt%) 73.66±1.2 PBS/PBAT (60/40 wt%) 70.27±2.1 PBS/PBAT (50/50 wt%) 65.88±2.3 Neat PBAT 46.12± Thermogravimetric analysis Figure 3.8 shows the thermal stability of PBS, PBAT, and their blends as a function of temperature. PBS undergoes cyclic degradation mechanism and some of the predominant byproducts are anhydrides, olefins, carbon dioxide, and esters [60]. PBAT degradation takes place by the breakdown of the ester groups and chain scission of C-O and C-C bonds on the polymer backbone. The onset degradation temperature (T onset ) of PBAT was 377 o C and PBS was 372 o C. This suggests that PBAT has slightly more thermal stability compared to PBS. TGA results reveal that PBS and PBAT present a relatively good thermal stability up to 300 o C. These data are similar to published literature results [55, 61]. The maximum degradation temperature (T max ) was observed at 405, 413, 408, 410, and 413 o C for extruded PBS, PBAT, PBS/PBAT (70:30 wt%), PBS/PBAT (60:40 wt%), and PBS/PBAT (50:50 wt%), respectively. The T onset and T max of the blends were quite similar to those estimated for PBS and PBAT homopolymer. All the blends showed single step degradation because the neat polymer degradation temperatures were close to each other, which was made clear through a derivative thermogram (Figure 3.9). According to the melt flow rate results, all the blends showed a molecular weight reduction, but the thermal stability of the blends was gradually increased compared to neat PBS. This may be improved compatibility between the polymer phases. While 108

143 increasing the PBAT content, thermal stability increased because PBAT is more thermally stable compared to PBS. The results indicate that a compatible blend was achieved and supports the DSC results. Figure 3.8. TGA curves of PBS, PBAT, and their blends Figure 3.9. DTG curves of PBS, PBAT, and their blends 109

144 3.3.8 Rheological properties Rheological properties were investigated to identify the interaction between polymer phases in the blends. Figure 3.10 shows the complex viscosity (η*) of the neat polymers and their blends as a function of frequency at 140 o C. Apparently, a higher complex viscosity was observed in the lower frequency range compared to its higher frequency region. This rheological behavior of the blends indicates that the blend is a pseudo plastic liquid. Furthermore, the neat PBS and PBAT exhibited almost Newtonian behavior at below 1 rad/s frequency, and a strong shear thinning behavior was observed beyond 1 rad/s frequency range. Figure Complex viscosity of PBS, PBAT, and their blends with different weight fractions of PBAT at 140 o C The PBS/PBAT blends had a higher complex viscosity compared to the neat polymers at lower oscillation frequencies. At higher frequencies, the complex viscosity of the blends was between that of the neat polymers. This increased viscosity may have occurred because transesterification can form pseudo structures [62] that can withstand shear forces. In addition, 110

145 this is due to phase morphology of the blends and compatibility between the phases. The higher compatibility between the two phases leads to good dispersion of the discrete phase in the blend system. As we can see in the SEM image, when PBAT content increases from 30 to 50 wt% in the blend, the discrete phase (PBAT) morphology is changed form spherical droplet to cocontinuous morphology. This indicates that, the PBS/PBAT 70/30 wt% blend was more compatible than PBS/PBAT 60/40 and 50/50 wt% blends. Consequently, the PBS/PBAT 70/30 wt% blend had higher melt viscosity than PBS/PBAT 60/40 and 50/50 wt% blend. A similar behavior of the PLA/PBS and PLA/PBAT blends was reported in literature [63, 64]. In addition, the transesterification reaction also plays a predominant role in viscosity improvement of the blends. At lower frequency, the transesterification product acts as a solid-like particle in all the PBS/PBAT blends and leads to a higher viscosity compared to the parent polymers. However, the FTIR results showed transesterification product gradually decreased with increasing PBAT content from 30 to 40 and 50 wt%. The higher content of PBAT reduces the transesterification reaction in the blend system and thus reduces the viscosity compared to PBS/PBAT 70/30 wt% blend. Li et a.l [64] have reported similar behavior for PLA/PBAT blends at lower frequencies and also that the interaction between the polymers can increase the melt viscosity. Figure 3.11 and 3.12 show the dynamic loss modulus and the storage modulus of PBS, PBAT, and PBS/PBAT blends. Generally, dynamic loss modulus and storage modulus represent the amount of energy dissipated in the viscous portion and the ability of a material to store energy during deformation, respectively. Figure 3.11 shows that the storage modulus (G') of each sample increased with increase in frequency. It was also observed that with the blending of PBAT into PBS, there were no changes in the storage modulus (G') at higher frequencies. All the blends showed a higher storage modulus at lower frequencies compared to the neat polymers. 111

146 The 70:30 wt% of PBS/PBAT blend had a higher loss modulus (G'') than other blends (Figure 3.12). Figure Loss modulus versus frequency for PBS, PBAT, and their blends with different weight fractions of PBAT at 140 o C When the PBAT phase was finely dispersed in the blend, the fine dispersal could be reason of interaction existing between the two phases. Stronger interactions were observed in the PBS/PBAT (70:30 wt%) blend, which were exhibited as a higher loss modulus. In addition, the increased storage modulus is attributed to the PBAT molecular chain entanglement with PBS molecular chain mediated by transesterification product acting as a compatibilizer. The higher entanglement density of the blends would store more recoverable energy. Generally, the entanglement density of the blend is depends on the existing compatibility between the two phases. The PBS/PBAT 70/30 wt% blend had more compatibility than PBS/PBAT 60/40 and 50/50 wt% blends as shown in SEM. The higher entanglement density of the PBS/PBAT 70/30 wt% blend leads to higher storage modulus. However, the storage modulus gradually decreased 112

147 at lower frequency while increasing PBAT content in the blends. A similar trend was observed in the complex viscosity. This is consistent with higher trasesterification product present in 70/30 wt% blend. The reduced storage modulus of the blends is due to the morphology changes [64] and entanglement density due to lower lower transesterification product present in the blend. Figure Storage modulus versus frequency for PBS, PBAT, and their blends with different weight fractions of PBAT at 140 o C The Cole Cole plot was used to explain the phase structure of PBS/PBAT blends and the plot was performed between the real and imaginary viscosity components of the blends. If the blend gives a single arc curve, it can suggest phase homogeneity at the melt stage [65]. Furthermore, if any deviations from a single arc are observed, they are evidence for inhomogeneous morphology and phase separation occurring in the blends due to a second relaxation mechanism occurring in the samples. The Cole-Cole plot for PBS/PBAT blend at 140 o C is depicted in Figure

148 Figure Cole Cole plot of the PBS/PBAT blends at 140 o C A second circular arc was observed on the right-hand side of the curve, and it is clear evidence for a second relaxation mechanism happening for all PBS/PBAT blends. Nevertheless, when the PBAT reaches 40 and 50 wt% in the blends, the blends showed a tail on the right hand side of the plot. This is probably due to the phase inversion occurred in the blends. This result shows that co-existing phase morphology was formed in the entire blend system and that the formed morphology may be the droplet-matrix or co-continuous phase morphology. Consequently, the Cole Cole plot shows an inhomogeneous morphology formed in PBS/PBAT blend systems. There have been reports of PLA/PBAT blend systems with similar observations of phase behavior when the PBAT concentration is more than 30 wt% in the matrix [64]. The Cole-Cole plot clearly shows that the PBS/PBAT 70/30 wt% blend is more heterogeneous compared to the 50/50 wt% blend. 114

149 3.3.9 Polarizing optical microscopy The spherulite morphology of the blends was investigated by optical microscopy. The dark and light regions represent the amorphous and crystalline phases, respectively. Figure 3.14 (a) shows that PBS/PBAT blends were annealed at a crystallization temperature of 80 o C for 30 min. When the PBAT concentration was 50 wt% in the blend, an increased number of PBS spherulites were observed with decreasing spherulite size. This decrease in size suggests that the PBS chain mobility was disrupted and PBAT acted as a nucleating site to promote the formation of crystalline nuclei. A PBS/polyvinylidene fluoride (PVDF) blend exhibited a similar phenomenon when PVDF was the predominant species in the blend compared to the PBS [66]. The PBS chain mobility was slowed in the presence of highly viscous PBAT, thereby decreasing spherulite growth. With increasing PBAT content from 30 to 50 wt% in the blends, the texture of PBS spherulites became coarse. Figure 3.14 (b) shows the PBS/PBAT blends after being annealed at a crystallization temperature of 90 o C for 30 min. The increasing PBAT composition in the blends caused a reduced number of spherulites because PBAT has less crystallinity and the PBS chain mobility increases at the crystallization temperature of 90 o C compared to 80 o C. With increasing PBAT content in the blends, the spherulites became rougher. For 30 wt% PBAT content in the blend, the PBS spherulites were uniformly distributed with uniform dimensions after being annealed for a given time. Consequently, our present results conclude that the spherulite size controls the mechanical properties and morphology of the blends. For example, the large spherulite size leads to the fracture along the spherulite boundaries and only occasionally through the spherulites. Therefore, small spherulite size can yield better tensile stress as well as strain [67]. 115

150 Figure (a) Photograph of the film annealed at 80 o C: (i) PBS; (ii) PBS/PBAT (70/30 wt%); (iii) PBS/PBAT (60/40 wt%) and (iv) PBS/PBAT (50/50 wt%). Figure (b) Photograph of the film annealed at 90 o C: (i) PBS; (ii) PBS/PBAT (70/30 wt%); (iii) PBS/PBAT (60/40 wt%) and (iv) PBS/PBAT (50/50 wt%) 116

151 Scanning electron microscopy Phase morphology of the blends depends on the second components, processing parameters, molecular weight of the virgin polymers, and compatibility between the polymers. If the blending components have a similar melt viscosity, the resulting morphology will be very fine and both polymers will be uniformly distributed throughout the blend whether it is the minor or major phase. The same is true if the blend consists of similar melt viscosity components. If the minor phase has a lower or higher viscosity compared to the major phase, it leads to the spherical domains of finely or coarsely dispersed morphology in the matrix. As shown in Figure 3.15, phase morphology in the blend was identified by the solvent et al.,hing method. Being a good solvent for PBAT while unable to dissolve PBS, THF was used as the et al.,hing solvent. The observed morphology of PBAT phase selectively removed from the blends without disturbing the PBS matrix is shown in Figure 3.15 (b). The PBAT phase was completely extracted under these conditions. The corresponding unextracted samples are shown in Figure 3.15 (a). This indicates that the holes are the extracted PBAT phase by THF. The surface morphology of the blend reveals that spherical PBAT particles were uniformly distributed throughout the matrix. Finer dispersions were observed in lower PBAT composition in the blend. When the PBAT composition was increased in the blend, the domain shape and size gradually changed due to the coalescence phenomenon. This may also act to decrease the tensile strength of the blends. 117

152 Figure (a) SEM images of PBS and PBAT blends (left hand side) (i) PBS/PBAT (70/30 wt%); (ii) PBS/PBAT (60/40 wt%) and (iii) PBS/PBAT (50/50 wt%). (b) SEM images of PBS and PBAT blends surface after et al.,hing with THF (right hand side): (i) PBS/PBAT (70/30 wt%); (ii) PBS/PBAT (60/40 wt%) and (iii) PBS/PBAT (50/50 wt%) 118

153 3.4 Conclusions We have succeeded in fabricating a high performance and biodegradable PBS/PBAT blend through the melt blending technique. There is a significant improvement in tensile strength and elongation by the incorporation of PBAT into PBS, indicating that a good level of compatibility is achieved between the polymers. The observed compatibility is caused by the formation of copolyester due to transesterification between the neat polymers, which was confirmed by FTIR analysis. DSC and DMA analysis suggested that the blends show compatibility between the PBS and PBAT. The rheological properties of blends such as the complex viscosity, loss modulus, and storage modulus were increased with the addition of PBAT into PBS. As the PBAT composition was increased, phase morphology changes occurred in the blends, leading to decreased values of complex viscosity, loss modulus, and storage modulus. The phase morphology of the PBS/PBAT blends shows a two phase structure in which PBAT is the minor phase. Furthermore, polarizing optical microscopy analysis revealed that the PBAT has disturbed the spherulite growth of the matrix. The prepared biodegradable PBS/PBAT blend is a potential substitute for non-biodegradable packaging films, blow molding bottles and flexible tubes. References [1] Wu D, Yuan L, Laredo E, Zhang M, Zhou W (2012) Ind Eng Chem Res 51: [2] Tokiwa Y, Calabia BP (2007) J Polym Environ 15: [3] Zhang K, Mohanty AK, Misra M (2012) ACS Appl Mater Interfaces 4: [4] Ouyang W, Huang Y, Luo H, Wang D (2012) J Polym Environ 20:1-9. [5] Yu T, Luo F, Zhao Y, Wang D, Wang F (2011) J Appl Polym Sci 120: [6] John J, Mani R, Bhattacharya M (2002) J Polym Sci A Polym Chem 40: [7] Nanda MR, Misra M, Mohanty AK (2011) Macromol Mater Eng 296:

154 [8] Huang P, Zhong Z, Zheng S, Zhu W, Guo Q(1999) J Appl Polym Sci 73: [9] Varughese KT, Nando GB, De PP, De SK (1988) J Mater Sci 23: [10] Nesarikar AR, Carr SH, Khait K, Mirabella FM (1997) J Appl Polym Sci 63: [11] Singh D, Malhotra VP, Vats JL (1999) J Appl Polym Sci 71: [12] Tang W, Murthy NS, Mares F, McDonnell ME, Curran SA (1999) J Appl Polym Sci 74: [13] Kotliar AM (1981) Journal of Polymer Science: Macromolecular Reviews 16: [14] Jayakannan M, Anilkumar P (2004) J Polym Sci A Polym Chem 42: [15] Chen HL (1995) Macromolecules 28: [16] Aravind I, Ahn KH, Ranganathaiah C, Thomas S (2009) Ind Eng Chem Res 48: [17] Focarete ML, Scandola M, Dobrzynski P, Kowalczuk M (2002) Macromolecules 35:8472. [18] Aravind I, Eichhorn KJ, Komber H, Jehnichen D, Zafeiropoulos NE, Ahn KH, Grohens Y, Stamm M, Thomas S (2009) J Phys Chem B 113: [19] Liu B, Bhaladhare S, Zhan P, Jiang L, Zhang J, Liu L, Hotchkiss AT (2011) Ind Eng Chem Res 50: [20] Fujimaki T (1998) Polym Degrad Stab 59: [21] Soccio M, Lotti N, Gigli M, Finelli L, Gazzano M, Munari A (2012) Polym Int 61: [22] Huang CL, Jiao L, Zhang JJ, Zeng JB, Yang KK, Wang YZ (2012) Polym Chem 3: [23] Kim SW, Lim JC, Kim DJ, Seo KH (2004) J Appl Polym Sci 92: [24] Myriant Technologies websites Available at: (Accessed on Febraury 2013). [25] Yoo ES, Im SS (1999) J Polym Sci B Polym Phys 37: [26] Wang J, Zheng L, Li C, Zhu W, Zhang D, Xiao Y, Guan G (2012) Polym Test 31:

155 [27] Qiu Z, Ikehara T, Nishi T (2003) Polymer 44: [28] Kim YJ, Park OO (1999) J Polym Environ 7: [29] Gu SY, Zhang K, Ren J, Zhan H (2008) Carbohydr Polym 74: [30] Gan Z, Abe H, Kurokawa H, Doi Y (2001) Biomacromolecules 2: [31] Sykacek E, Hrabalova M, Frech H, Mundigler N (2009) Compos Part A Appl Sci Manuf 40: [32] Javadi A, Kramschuster AJ, Pilla S, Lee J, Gong S, Turng LS (2010) Polym Eng Sci 50: [33] Jiang L, Liu B, Zhang J (2009) Ind Eng Chem Res 48: [34] Jang MO, Kim SB, Nam B-U (2012) Polym Bull 68: [35] Javadi A, Srithep Y, Lee J, Pilla S, Clemons C, Gong S, Turng LS (2010) Compos Part A Appl Sci Manuf 41: [36] Qiu Z, Ikehara T, Nishi T (2003) Polymer 44: [37] Li Y, Shimizu H (2009) ACS Appl Mater Interfaces 1: [38] Huang X, Li C, Zheng L, Zhang D, Guan G, Xiao Y (2009) Polym Int 58: [39] Kwei TK (1984) J Polym Sci B 22: [40] Wang LH, Huang Z, Hong T, Porter RS (1990) J Macromol Sci-Phys B 29: [41] Takiyama.E, Fujimaki.T, Seki.S, Hokari.T, Hatano.Y (1994) US Patent No [42] Takiyama.E, Hatano.Y, Fujimaki.T, Seki.S, Hokari.T, Hosogane.T, Harigai.N (1995) US Patent No [43] Mittal V. Functional polymer blends: Synthesis, Properties, and Performance (CRC press, New York 2012, p.235). [44] Joseph K, Thomas S, Pavithran C (1996) Polymer 37:

156 [45] Dobkowski Z (1986) Rheol Acta 25: [46] Carrasco F, Pagès P, Gámez-Pérez J, Santana O-O, Maspoch ML (2010) Polymer Polym Degrad Stab 95: [47] Corre YM, Bruzaud S, Audic JL, Grohens Y (2012) Polym Test 31: [48]Siciliano A, Seves A, Maro TD, Cimmino S, Martuscelli E, Silvestre C (1995) Macromolecules 28: [49] Liu AS, Liau WB, Chiu WY (1998) Macromolecules 31: [50] Richards E, Rizvi R, Chow A, Naguib H (2008) J Polym Environ 16: [51] Parulekar Y, Mohanty AK (2007) Macromol Mater Eng 292: [52] Mark JE. Physical Properties of Polymers Handbook (2nd edition, Spring Science, New York, p.293). [53] Jain S, Redy MM, Mohanty AK, Misra M, Ghosh AK (2010) Macromol Mater Eng 295: [54] Menard KP. Dynamic Mechanical Analysis: A Practical introduction (2nd edition, CRC Press, New York 1999, p. 85). [55] Mohanty S, Nayak SK (2012) J Polym Environ 20: [56] Aravind I, Boumod A, Grohens Y, Thomas S (2010) Ind Eng Chem Res 49: [57] Thomas S, Gupta BR, De SK (1987) J Vinyl Technol 9: [58] ASTM D Standard Test Method for Deflection Temperature of Plastics Under Flexural Load in Edgewise Position. [59] Kawamoto N, Saki A, Horikoshi T, Urushihara T, Tobita E (2007) J Appl Polym Sci 103: [60] Lu SF, Chen M, Chen CH (2012) J Appl Polym Sci 123:

157 [61] Chrissafis K, Paraskevopoulos KM, Bikiaris DN (2005) Thermochim Acta 435: [62] Di Y, Iannace S, Maio ED, Nicolais L (2005) Macromol Mater Eng 290: [63] Bhatia A, Gupta, Rahul K., Bhattacharya, S.N. and Choi, H.J (2007) Korea-Aust Rheol J 19: [64] Li K, Peng J, Turng LS, Huang HX (2011) Adv Polym Technol 30: [65] Wang L, Jing X, Cheng H, Hu X, Yang L, Huang Y (2012) Ind Eng Chem Res 51: [66] Wang T, Li H, Wang F, Schultz JM, Yan S (2011) Polym Chem 2:1688. [67] V. I. Pavlov, Soviet materials science : a transl of Fiziko-khimicheskaya mekhanika materialov / Academy of Sciences of the Ukrainian SSR, 1971, 4 (5):

158 Chapter 4: Preparation and Characterization of Maleic Anhydride Grafted Biodegradable Polyesters Abstract Maleic anhydride (MAH) grafting was quantified onto poly(butylene adipate-coterephthalate) (PBAT), poly(butylene succinate) (PBS), and their blend in the presence of free radical initiator (dicumyl peroxide, DCP) by reactive melt processing in an internal batch mixer. Fourier transform infrared (FTIR) spectroscopy confirmed the successful grafting of MAH on the polyesters backbone. The dependence of grafting yield on initiator and MAH concentration was calculated by back-titration. Compared to the MAH grafted PBS, the MAH grafted PBAT and MAH grafted PBS/PBAT blend had slightly lower grafting yield over the entire range of DCP concentrations. This difference in grafting yield could be due to the structural difference and proton abstraction capability of the polymer backbone. Then, MAH grafting efficiency of batch and continuous process was compared. The result shows that the batch processed samples had slightly higher grafting yield than continuous processed samples. The slightly higher grafting yield as observed in the internal batch process may be attributed to higher residence time and air contact of the reaction medium. The crystallization and melting temperatures of the MAH grafted samples were significantly decreased compared to their counterparts. This is possibly due to the MAH group preventing the lamella growth and nucleation of MAH grafted samples, thus leading to the formation of imperfect crystal structure compared to the virgin polymer. The onset temperatures for thermal degradation of the MAH grafted samples were reduced compared to neat polymers while their maximum and final decomposition temperature remained unchanged. 124

159 4.1 Introduction Recently, biodegradable polymeric materials have gained great interest because of certain environmental concerns in using non-biodegradable polymers. In order to reduce the problems regarding the non-biodegradable polymer disposal in the environment, some of the biodegradable polymers have been commercialized. However, many biodegradable polymers are expensive compared to the commonly used polymers. Therefore, the use of biodegradable polymers as a matrix for producing biocomposites will result in products with higher costs. Increase in the cost of biocomposites could be mitigated by adding low-cost natural fillers/fibers to make them cost competitive with traditional non-biodegradable composites. Nevertheless, there are some technological challenges in successfully developing a natural fiber/bioplastic composite. The hydrophilic character of the natural fibers/fillers hinders its compatibility with the hydrophobic polymer matrix due to its lack of interfacial adhesion. The incompatibility between the reinforcement and polymer leads to weak stress transfer from one phase to another phase. This weak stress transfer causes inferior mechanical properties of the resulting composites. Furthermore, very often polymer blends and composites are thermodynamically immiscible. In most cases, immiscibility of the components of the composites and blends is due to solubility parameter difference. Often such blends and composites are not able to target specific applications. The immiscibility and poor compatibility of the composites and blends can be overcome through a compatibilizer. A compatibilizer helps to improve the better stress transfer between the components through interfacial adhesion, uniform dispersed phase and reduced surface tension between the two phases [1]. Generally, the compatibilization of blends/composites can be performed in two ways i.e., non-reactive compatibilization and reactive compatibilization. Usually block copolymers are 125

160 used as a non-reactive compatibilizer because one constitutive block is miscible with one blend component while the second block is miscible with the other component. This type of compatibilizers is not economically viable for industrial applications and it is not used extensively as a compatibilizer [2]. A reactive compatibilizer is a good choice for producing compatibilized composites/blends. The reactive compatibilization involves forming the block or graft copolymer in-situ during blend preparation via interfacial reaction of added functionalized polymeric components. The maleic anhydride (MAH) grafted polymer is a well-known reactive compatibilizer. For example, maleated polypropylene is extensively studied as a compatibilizer in polypropylene based blends [3] as well as composites [4]. The polymer grafting reaction has a predominant role in preparing reactive compatibilizer because it can introduce new reactive functional groups on the polymer backbone. The grafting can be performed in different ways such as melt state, solution state, solid state, irradiation, suspension in aqueous/organic solvents [5, 6] and ball milling [7]. Among them, melt state reactive extrusion is a very common and economical way to create the grafted polymers. A variety of functional monomers has been used as grafting agent on the polymer backbones [8-10]. MAH is a very common functional monomer for grafting because of its unique free radical reactive double bonds, reactive anhydride groups, and poor homopolymerization tendency during free radical graft polymerization [11]. The poor homopolymerization of MAH enhances the grafting efficiency of the resulting grafted polymers. In addition, MAH functional groups have high reactivity with specific functional groups such as carboxyl, amine, and hydroxyl [12]. The MAH grafting onto molten polypropylene (PP) [13], polyethylene (PE) [14], and ethylenepropylene-diene (EPDM) terpolymer [15] has been investigated in the literature. In addition, the MAH grafted polyolefins are commercially available in the market and act as a good 126

161 compatibilizer for PP and PE based blends/composites [16, 17]. Thus, MAH reactive functional group on the polymer backbone significantly reduces the dispersed domain size and surface tension, and increases the interfacial adhesion in the composites/blends. Recently, this research has expanded to the biodegradable/biocompostable polymers like polylactide (PLA) [18], poly(hydroxybutyrate-co-valerate) (PHBV) [19], poly(butylene adipateco-terephthalate) (PBAT) [20], poly(butylene succinate) (PBS) [21], and polycaprolactone (PCL) [22]. MAH grafted copolymers have been show to be a good compatibility enhancer in the biodegradable polymer based blends [23] and composites [24]. Research findings confirmed that a small quantity of maleated compatibilizer increased the compatibility between two thermodynamically immiscible materials. Few researchers have done detailed research on maleic anhydride grafting onto PLA backbone by melt process [25, 26]. The use of MAH grafted PLA as a compatibilizer has been studied while investigating the performance of composites materials [18, 27, 28]. This literature has clearly shown that the compatibility was improved between the matrices, and reinforcement through MA grafted PLA meanwhile improved their mechanical performances. Similar observations were found in some other biodegradable polymer composites with MAH grafted compatibilizers [29-31]. Many researchers have performed maleation of biodegradable and non-biodegradable polymers but none of the literature was reported maleation of PBS/PBAT blend. This study synthesized and characterized MAH grafted biodegradable polymers. The MAH grafted PBS, PBAT and PBS/PBAT blend were synthesized by varying initiator concentration from 0.5 to 1 phr while keeping MAH concentration constant at 5 phr. The grafting percentage was determined by back-titration method. In addition, the maleated samples 127

162 were characterized by thermal analysis, and FTIR spectroscopy. In addition, melt viscosity and elasticity of the MAH grafted samples was studied by rheological analysis. 4.2 Materials and Methods Commercially available PBAT (Biocosafe 2003) and PBS (Biocosafe 1903) pellets were procured from Xinfu Pharmaceutical Co., Ltd, China. PBS, PBAT and their blend (60/40 wt%) were processed by micro compounder (DSM Xplore, The Netherland) using 140 o C processing temperature. General properties of the virgin PBS, PBAT and blend of PBS/PBAT (60/40 wt%) were measured and are reported in Table 4.1. Organic free radical initiator, dicumyl peroxide (DCP-99% purity) was purchased from Acros Organics with half-life time about 9.2 s at 150 C and maleic anhydride (MAH) was obtained from Sigma-Aldrich, USA. All the materilas were used as received. Table Properties of the neat PBS, PBAT and PBS/PBAT blend Polymers Density a (g/cm 3 ) Melting Point b ( o C) Glass Transition Temperature b ( o C) Melt Flow Index c (g/10 min) PBAT ±1.8 PBS ±2.4 PBS/PBAT (60/40 wt%) ±2.8 a: Measured by Archimedes method, b: Measured by DSC with heating rate 10 o C/min, c: 190 o C with 2.16 kg Synthesis of MAH grafted PBS, PBAT and their blend Before processing the polymer pellets were dried at 80 o C for 12 h. Based on our previous research results, PBS/PBAT (60/40 wt%) blend was chosen as an optimum blend for further study [33]. Hereafter, PBS/PBAT (60/40 wt%) blend will be referred to as PBS/PBAT blend. Maleation of PBS (MAH-g-PBS), PBAT (MAH-g-PBAT) and PBS/PBAT blend (MAH-g- PBS/PBAT) were prepared by varying the initiator concentration. The MAH-g-PBS, MAH-g- 128

163 PBAT, and MAH-g-PBS/PBAT samples were prepared in a laboratory-scale internal batch mixer (Torque Rheometer, Haake PolyLab QC, Thermo scientific). The formulations are shown in Table 4.2. The batch mixer barrel was divided into three heating zones. The temperatures of these three heating zones were kept constant at 160 o C with a screw speed of 60 rpm and reaction time of 6 min for synthesizing all the samples. Reaction time of 6 min was divided into three steps. In the first step, polymers were melted at 160 o C for 2 min and in the second step a free radical initiator was introduced into the molten polymers to make reactive sites for 2 min. Finally, MAH was added into the reaction medium and reaction was carried out for another 2 min to obtain desired amount of grafting content onto polymers. It s known that homopolymerization of MAH is difficult with molten polymers because of the 1,2-disubstituted double bond present in the MAH. In order to diminish the homopolymerization of the MAH in the reaction medium, the reaction was carried out above the ceiling temperature (150 o C). During the reaction, the reaction mixture torque was monitored over time showing a significant increment. The increased torque was likely due to grafting and gel formation occurred after addition of MAH and DCP into polymers. The MAH grafted samples were taken out and ground into small pieces for further analysis. In addition, the MAH grafting of PBS/PBAT was performed in a continuous process using a Leistritz extruder Micro 27, USA. The extruder had three kneading disc block regions and self-wiping co-rotating twin screws. The extruder was consisting of 11 heating zones with electric heating and a water cooling system. Appropriate amount of monomer, initiator and polymers were manually premixed and fed into the hopper with a feed rate of 5 kg/h. In order to compare batch process with continuous process, the continuous process was performed with similar batch processing parameters such as processing temperature and screw speed. In 129

164 continuous process, 11 heating zones were set at a constant temperature of 160 o C and screw speed of 60 rpm. The MAH grafted molten materials were pumped through the extruder die, quenched and pelletized. Table 4.2. Proposed formulation for producing maleation of PBS, PBAT, and PBS/PBAT blend Samples DCP (phr) MAH (phr) MAH-g-PBS MAH-g-PBS MAH-g-PBS MAH-g-PBAT MAH-g-PBAT MAH-g-PBAT MAH-g-PBS/PBAT blend MAH-g-PBS/PBAT blend MAH-g-PBS/PBAT blend Grafting mechanism Melt state free radical grafting mechanism is a widely accepted method in academia and industry. In the presence of peroxide free radical initiator, the possible MAH grafting reaction of PBS and PBAT is shown in Figure 4.1 and 4.2 [20, 34]. Basically, the free radical mechanism occurs in three steps of initiation, propagation and termination. The initiation step is well established, which is the homolytic rupture of the initiator and produces free radicals at higher temperature (peroxide decomposition temperature). Then the second step, propagation, is to form macroradicals through abstraction of hydrogen atoms from the polymer α-carbon atom. 130

165 Figure 4.1. Proposed reaction mechanism of MAH grafted PBS (MAH-g-PBS) [34] 131

166 Figure 4.2. Proposed reaction mechanism of MAH grafted PBAT (MAH-g-PBAT) [20] 132

167 Once the radicals are formed, they can directly react with MAH onto the polymer backbone. Finally, grafting process is terminated by recombination of radicals. This termination reaction leads to the formation of succinic type anhydride structure on the polymer chains that further undergo β-scission reaction on the grafted copolymers (Figure 4.1 and 4.2). There are other possibility to form chain scission reaction either through backbiting or thermo hydrolysis reactions [26]. Thermodynamically, when the maleation reaction takes place above ceiling temperature, the homopolymerization of MAH (poly-mah) grafting will not occur on the polymer backbone [35]. In addition, if the grafting reaction takes place above ceiling temperature the formed homopolymers (poly-mah) decompose, unlike other functional monomers [36] Purification of MAH grafted samples The purification of the MAH grafted samples was performed according to a modified procedure [20]. The unreacted MAH was removed by vacuum drying at 80 o C for 24 h. Vacuum dried MAH grafted sample was dissolved in chloroform at room temperature overnight. After dissolution of MAH grafted sample, they were selectively precipitated in methanol and filtered. The filtered sample was repeatedly washed with excess methanol to remove residual MAH and DCP, followed by drying at 80 o C under vacuum for 24 h. These dried samples were used for further analysis Determination of grafting percentage The MAH grafting percentage was determined by back-titration, which was modified from Nabar et al., [20]. The purified MAH grafted sample was dissolved in 100 ml of chloroform at ambient temperature for 2 h. Subsequently, the MAH groups were hydrolyzed into carboxylic acid by adding a few drops of 1 N hydrochloric acid. Immediately, the solutions were titrated against N alcoholic KOH in the presence of few drops of phenolphthalein 133

168 indicator. Under this condition, maleated samples were completely soluble in chloroform and did not precipitate during titration against alcoholic KOH. The grafting percentage was calculated as follows: Grafting Percentage (%) = x98.06x100 (4.1) where is the weight (g) of the maleated sample, is the volume of the KOH in liters, (g/mol). is the normality of KOH dissolved in methanol, and MAH molecular weight is Gel percentage measurement Soxhlet extraction was used to measure the gel percentage of the samples. Both grafted and ungrafted PBS, PBAT and their blend are completely soluble in chloroform but cross-linked PBS, PBAT, and blend of PBS/PBAT were found to be insoluble in chloroform. Based on the PBS, PBAT and their blend solubility, the extraction was perfromed by refluxing chloroform for 24 h. After 24 h extraction, the obtained insoluble fractions on the filter paper were dried at 60 o C for 12 h to remove the residual chloroform. The dried filter papers were weighed, and gel (crosslinked) content was calculated as follow: Gel percentage = X100 (4.2) where W i and W f are the initial weight of sample in the filter paper and the final weight of sample left in the filter paper after extraction, respectively. The gel content measurement was repeated at least three times. 134

169 4.2.6 Fourier transform infrared (FTIR) spectroscopy Structural differences between the samples were evaluated by FTIR spectroscopy (Nicolet 6700-Thermo scientific) at room temperature with 36 consecutive scans. The analysis was performed by measuring transmittance versus wavenumbers in the range of cm -1 at a resolution of 4 cm Thermogravimetric analysis (TGA) Thermal stability of the ungrafted and MAH grafted PBS, PBAT and their blend was measured using a TA Q500 Instruments. The experiment was performed in the N 2 atmosphere at a flow rate of 60 ml min -1 with a heating rate of 20 o C/min Differential scanning calorimetry (DSC) The melting temperature (T m ), crystallization temperature (T c ) and crystallization enthalpy ( H c ) of MAH grafted and ungrafted samples were analysised by TA Q200 DSC Instruments. Accurately weighed samples were encapsulated into the aluminum pan and placed into the DSC machine. The samples were heated in the presence of N 2 atmosphere with a flow rate of 50 ml min -1. The reported melting temperature and crystallization temperature were collected from the second heating and first cooling scans, respectively. 4.3 Results and Discussion Infrared spectroscopy The PBS, PBAT, PBS/PBAT blend and their respective MAH grafted samples were analyzed trough FTIR spectroscopy (Figures ). The FTIR spectra of MAH, PBS and maleated PBS samples are shown in Figure 4.3. The broad and medium intensity peak at 955 cm - 1 is attributed to C OH bending of carboxylic groups in PBS. The stretching vibration of ester carbonyl (>C=O) group was observed at 1718 cm -1. Most of the saturated hydrocarbons contain 135

170 methyl groups. These methyl groups show a symmetric stretching band at 2962 cm -1 and an asymmetric stretching band at 2872 cm -1. In PBS, methyl and methylene C-H stretching bands occur at 2945 and 2854 cm -1, respectively. Two new small peaks (1859 and 1788 cm -1 ) were formed in MAH grafted PBS, which corresponds to the saturated cyclic anhydride carbonyl ring (succinic anhydride group) [37, 38]. These characteristic peaks confirm that the MAH moieties were successfully grafted on to the PBS backbone. The characteristic functional group of alkyl ether is C-O-C. A strong symmetric C-O-C stretching peak can be seen at 1151 cm -1 in PBS, which confirms that the PBS contains an alkyl ether group. In Figure 4.3, observed peaks at 955, 806, and 654 cm -1 were attributed to C-O stretching, CH 2 in OC(CH 2 ) 2 CO in-plane bending, and COO bending bands of the PBS and MAH grafted PBS. These peaks are clear evidence to differentiate PBS from other polymers. Figure 4.3. FTIR spectra of MAH, neat PBS and MAH-g-PBS with 1 phr DCP and 5 phr MAH FTIR spectra of MAH, PBAT and maleated PBAT samples are presented in Figure 4.4. The neat PBAT had peaks at 2951, 2864, 1716, 1463, 1402, 1259, 1157, 1111, 729 and 875 cm

171 The maleated PBAT sample shows additional peaks at 1786 and 1855 cm -1 compared to neat PBAT. These additional peaks correspond to symmetric and an asymmetric stretching for the carbonyl functionality of MAH [20]. The observed peaks at 2951 and 2864 cm -1 were assigned to -CH 3 and -CH 2 - stretching vibrations. The band at the cm 1 results from the C O C group in the PBAT ether linkage. The peak at 1716 cm 1 was attributed to the C=O stretching of ester groups in PBAT. Peaks at 875 and 729 cm -1 were due to out-plane bending vibration of =C H in benzene ring. Generally, out-plane bending vibration of =C H group in benzene ring should occur at 830 cm -1. In PBAT, conjugated C=O groups were influenced and the benzene ring out-plane bending of =C H groups appeared at 729 cm -1. This is a characteristic peak of PBAT in the FTIR spectra. Furthermore, PBAT has characteristic peaks at 729, 1111 and 1157 cm -1 and they can be used as identification of PBAT structure [39]. Figure 4.4. FTIR spectra of the MAH, neat PBAT and MAH-g-PBAT with 1 phr DCP and 5 phr MAH 137

172 FTIR spectra of MAH, PBS/PBAT blend and MAH grafted PBS/PBAT blend were scanned separately and are shown in Figure 4.5. Peaks at 2956 and 2860 cm -1 were assigned to - CH 2 - and -CH 3 stretching vibration in the PBS/PBAT blend. More specifically, the peak at 1716 cm -1 was due to the C=O stretching vibration in the ester groups. The bands at 1259 and 1157 cm -1 attributed to the stretching vibration of ether groups in the PBAT and PBS, respectively. The out plane and in plane bending vibration of terephthalic acid (benzene =C-H) unit in the PBAT was observed peaks at 731 and 1036 cm -1. The succinic acid -COO bending mode peak appeared at 648 cm -1 and the succinic acid -CH 2 - group peak was observed at 806 cm -1. However, MAH grafted PBS/PBAT blend exhibited two distinguish peaks at 1857 and 1782 cm - 1, which were assigned to symmetric and asymmetric stretching of C=O groups in the grafted MAH functionality [5]. This literature suggests that the FTIR absorption peak near 1850 cm -1 (symmetric stretching) and a peak near 1780 cm -1 (asymmetric stretching) are due to vibration of anhydride. As a result, our FTIR spectra confirmed that the MAH groups are grafted on the PBS/PBAT polyester backbone. The characteristic peaks of MAH-g-PBS/PBAT (1857 cm -1 ) were between that of MAHg-PBS (1859 cm -1 ) and MAH-g-PBAT (1855 cm -1 ). This suggests that the MAH was successfully grafted on both the PBS and PBAT backbone in the PBS/PBAT blend. A similar trend was observed by Li et al., [40] in the MAH grafted polyolefin (PE/PP) blend. In addition, elimination of unreacted MAH in the grafted samples can be confirmed by disappearance of the MAH C=C characteristic peak at 684 cm -1 in the FTIR spectra [13]. The FTIR peak at 1590 cm -1 belongs to the MAH C=C stretching. This cyclic C=C peak was not found in the MAH grafted PBS, PBAT or PBS/PBAT blend, which suggests that the unreacted MAH was not present in the MAH grafted samples. 138

173 Figure 4.5. FTIR spectra of MAH, neat PBS/PBAT blend and MAH-g- PBS/PBAT blend with 1 phr DCP and 5 phr MAH MAH grafting percentage calculation The MAH grafting percentage can be controlled by many parameters including reaction temperature, rotor speed, resistance time, and monomer to initiator ratio [5, 22]. In this literature, initiator concentration was changed upto 1% while MAH concentration, screw speed, and processing temperature were kept constant. The main reason for keeping MAH content constant was based on literature suggesting a higher influence on the grafting efficiency of the concentration of initiator (DCP) than concentration of MAH [5]. Mani et al., [5] studied the MAH grafting of PBS, poly(butylene succinate adipate) (PBSA), PLA, and Eastar co-polyester with various initiator and MAH concentrations at different temperatures. They found that there is more possibility to form cross-linking when the initiator concentration higher than 1%. Therefore, in the current study, initiator concentration was kept below 1 phr for all the 139

174 experiments. Moreover, the MAH grafting reaction was conducted at only one temperature (160 o C) because previous researchers have reported that the temperature does not have a significant influence on the MAH grafting efficiency of polyesters backbone [20, 26]. The MAH grafting was performed in a twin-screw extruder as well as through an internal batch mixer. The MAH grafting percentage of the internal batch processed samples is shown in Table 4.3. The grafting yields of the MAH-g-PBS, MAH-g-PBAT and MAH-g-PBS/PBAT blend were gradually increased with increasing DCP concentration from 0.5 to 1 phr. However, the MAH-g-PBS in the presence of 0.7 and 1 phr DCP initiator had a grafting yield of 2.45 and 2.56%, respectively. This observed small difference in the grafting yield may be due to the faster termination reaction of free radicals at higher free radical concentrations in the reaction medium [40]. Table 4.3. MAH grafting percentage of the PBS, PBAT and PBS/PBAT blend Samples DCP (phr) MAH (phr) Grafting Percentage (%) MAH-g-PBS MAH-g-PBS MAH-g-PBS MAH-g-PBAT MAH-g-PBAT MAH-g-PBAT MAH-g-PBS/PBAT blend MAH-g-PBS/PBAT blend MAH-g-PBS/PBAT blend

175 The MAH grafting onto the PBAT was performed in the presence of 0.5 phr DCP and 5 phr MAH, and the grafting yield observed was 1.34%. In a previous study, Chen and Zhang [29] performed the MAH grafting onto PBAT with initiator and MAH concentrations of 0.5 and 5 phr, respectively. They also found that the MAH grafting yield was 1.34%.This result suggested that the grafting yield variation in the PBS and PBAT found in the current study is reasonable. When comparing grafting efficiency of PBS and PBAT with 1 phr DCP and 5 phr MAH, PBS showed higher MAH grafting yield than PBAT (Table 4.3). This observed difference in MAH grafting yield could be due to the structural difference and proton abstraction capability of the polymer backbone. The same trend was observed in MAH grafted PBS/PBAT blends. In PBAT, the observed lower MAH grafting yield compared to PBS may be attributed to a lower number of free radical formations on its backbone than the PBS backbone. This lower number of radicals leads to lower grafting efficiency [5]. In addition, the grafting yield difference may be due to the viscosity difference between the polymers. The comparatively lower viscosity of PBS compared to PBAT might allow the MAH to better disperse in the reaction medium and it can enhance their grafting yield. This phenomenon may be limited in the high viscosity PBAT. An optimal concentration of initiator and monomer are required to achieve a better grafting efficiency. The present study shows that the optimal concentration of initiator and grafting monomer were 1 and 5 phr, respectively. There is literature available for comparison of MAH-g-PBS and MAH-g-PBAT in an internal mixer and continuous mixer [20, 21, 29, 30, 34, 41]. There is no literature available for comparison of MAH-g-PBS/PBAT blend in both internal mixer and continuous mixer. Therefore, in the present study investigated that MAH-g-PBS/PBAT blend in batch as well as in continoues process. In both these processes, 1 phr DCP and 5 phr MAH were used to graft MAH 141

176 onto PBS/PBAT blend. The batch process and continuous process grafting yields were 2.05 and 1.65%, respectively. The grafting efficiency was significantly lower in continuous processed MAH-g-PBS/PBAT blend compared to in batch process. This significantly lower grafting yield in continuous process may be due to lower residence time compared to batch process. Generally, the total residence time in a twin-screw extruder is lower when compared to batch mixture [42]. The slightly higher grafting yield as observed in internal batch process may be attributed to higher residence time and air contact of the reaction medium [37]. Similarly, the internal batch process had higher grafting efficiency than intermeshing co-rotating twin-screw extrusion process [43]. Moreover, the grafting yields for MAH-g-PBS, MAH-g-PBAT and MAH-g- PBS/PBAT blends were higher than that the commercially available MAH-g-PP (grafting yield between %) [4]. Therefore, we believe that MAH grafted PBAT, PBS, and their blend samples might act as an effective compatibilizer in their composites Gel content measurement Generally, in the presence of DCP initiator, PBAT and PBS can form complex reactions such as copolymers of PBAT and PBS (e.g. PBAT-g-PBS) as well as cross-linked/branched PBAT and PBS [44]. The gel formation (cross-linking) was investigated by a solvent extraction method and the results are shown in Figure 4.6. The gel content of the MAH grafted PBAT, PBS and their blend steadily increased with increasing DCP content. This is reasonable because increasing DCP content increases the free radical formation. Consequently, the formed excess radicals can easily react with each other and leads to a cross-linking reaction. It was observed that the PBS had more gel content than PBS/PBAT blend and PBAT in three different concentrations of DCP content with the same MAH content (5 phr). The gel formation may be reduced by increasing MAH concentration in the reaction medium. Increasing MAH content in 142

177 the reaction medium can react with excess free radicals on the polymer backbone. This may increase the grafting percentage while reducing the gel formation. Ma et al., [44] examined cross-linking behavior of PBS, poly(hydroxybutyrate-co-valerate) (PHBV), and PBS/PHBV blend using DCP and they found that PBS forms more cross-linking than PHBV and PBS/PHBV blend. A similar type of cross-linking reaction was observed in MAH grafting on PLA in the presence of DCP [25, 45]. They found that the cross-linking of the PLA increases with increasing DCP content. Figure 4.6. Gel content of MAH grafted PBS, PBAT, and PBS/PBAT samples with 5 phr MAH and different concentration of DCP A study [46] reported that the gel formation could be controlled by adding styrene as a co-monomer in the free radical initiated MAH grafting reaction. This co-monomer not only reduces the gel formation in the reaction but also accelerates the grafting yield and grafting reaction rate in the melt state. Nevertheless, using styrene as a co-monomer is not recommended for certain applications. Therefore, it needs to be replaced by some other co-monomers [47]. A 143

178 few researchers have tried to prevent the gel formation during MAH grafting of polyolefin by using electron donor additives [48]. Although, they found a gel-free MAH grafted polyolefin product, the grafting yield was reduced Thermogravimetric analysis Figure 4.7 shows the thermogravimetric result of neat and maleated PBS, PBAT and their blends with respect to the temperature. Thermal stability of PBAT was higher than PBS and PBS/PBAT blend. The onset degradation temperatures of the samples were detected from the deviation of the baseline thermogravimetric curve by tangent plots. MAH grafted PBS, PBAT, and blend of PBS/PBAT showed lower onset thermal degradation compared to neat PBS, PBAT and PBS/PBAT. This onset of thermal degradation reduction may be attributed to the polymer chain scission occurs during maleation reaction. Similar thermal stability reduction was found in maleated PLA with peroxide initiators [25, 26]. However, maximum and final degradation temperatures were not heavily affected in the current study. Figure 4.7. TGA thermograms of neat and maleated PBS, PBAT and PBS/PBAT blend (the maleated samples were obtained with 1 phr DCP and 5 phr MAH) 144

179 4.3.5 Differential scanning calorimetry In polymer processing, non-isothermal DSC analysis is more practical interest than isothermal DSC analysis. The heating and cooling non-isothermal DSC thermograms of the neat and maleated samples are shown in Figures 4.8 and 4.9. Table 4.4 shows the T c, H c, and T m of neat and maleated PBS and PBS/PBAT blend, and the reported values are obtained from second heating and first cooling cycles. Both PBS and PBS/PBAT blends had almost the same melting points because both PBS and PBAT melting points are very close to each other. Therefore, the melting peaks may be overlapping. John et al., [49] also observed a similar phenomenon in PBS/aliphatic aromatic copolyester blend. The major melting points of virgin PBS, PBAT and blend of PBS/PBAT are 114, 115 and 114 o C, respectively. Figure 4.8. DSC second heating curves of neat PBS, PBS/PBAT (60/40 wt%) and their maleated samples with 1 phr DCP and 5 phr MAH. 145

180 Figure 4.9. DSC first cooling curves of neat PBS, PBS/PBAT (60/40 wt%) and their maleated samples with 1phr DCP and 5 phr MAH Interestingly, the PBS and blend of PBS/PBAT showed two distinct endothermic peaks at 108 (T m1 ) and 114 o C (T m2 ), which may be attributed to two different lamellar thicknesses presented in the PBS and PBS/PBAT blend. Usually double melting peaks are observed for semicrystalline polymers and they can be explained by a melt re-crystallization mechanism [50]. These two melting points had a significantly lower temperature shift after MAH grafting on both PBS and PBS/PBAT blend. This is possibly due to the fact that the MAH group may prevent the lamella growth and nucleation of MAH grafted samples, thus leading to imperfect crystal structure compared to their parent polymers [51]. However, all the MAH grafted samples had a sharp melting point and a weak shoulder melting point. The weak shoulder melting peak and sharp melting peak are attributed to lamella with more imperfect crystals and lamella with perfect crystals, respectively. In addition, after MAH grafting, the PBS and blend of PBS/PBAT showed one additional exothermic peak prior to melting points. This additional exothermic peak 146

181 resulted from recrystallization of PBS during heating [21]. The crystallization temperatures of PBS/PBAT and PBS blend were considerably affected with the addition of MAH group on the polyester backbone. An obvious crystallization temperature reduction was observed in PBS and PBS/PBAT blend from 93 to 63 o C and from 94 to 55 o C, respectively. The significant reduction in crystallization temperature was attributed to the decrease in polymer chain regularity which hinders crystal growth, leading to lower crystallization temperatures [25]. The crystallization enthalpies (Table 4.4) of fusion for MAH-g-PBS and MAH-g-PBS/PBAT blend were found to be 65 and 35 J/g, respectively. Both MAH-g-PBS and MAH-g-PBS/PBAT blend crystallization enthalpies were lower than those of ungrafted PBS and PBS/PBAT blend. An apparent decrease in crystallization enthalpy of MAH grafted samples was due to the reduction of polymer chain regularity. A similar result for solid state MAH grafting onto PP was reported [52]. Table 4.4. Detailed DSC results of the maleated PBS and PBS/PBAT blend (MAH grafted samples were prepared with 1 phr DCP and 5 phr MAH) Samples T m1 ( o C) T m2 ( o C) T c ( o C) H c (J/g) PBS MAH-g-PBS PBS/PBAT blend MAH-g-PBS/PBAT blend Conclusions The radical grafting reaction of MAH onto the PBAT, PBS, and blend of PBS/PBAT was investigated succefully. The characteristic peaks of MAH in the MAH grafted PBAT, PBS and blend of PBS/PBAT were observed by FTIR spectroscopy. The characteristic peaks of MAH-g- PBS/PBAT blend were located between that of MAH-g-PBS and MAH-g-PBAT. This revealed 147

182 that MAH was successfully grafted on both PBS and PBAT backbone in the MAH-g-PBS/PBAT blend. The maximum MAH grafting yield was reached at MAH (5 phr) and DCP (1phr) concentration in PBS, PBAT and PBS/PBAT blend. However, a higher grafting yield on the PBS backbone was observed (2.56%) compared to MAH grafted PBAT (1.90%) and PBS/PBAT blend (2.05%). The MAH grafting yield was compared in the batch and continuous process. The batch processed sample had a slightly higher yield than the continuous processed sample. The slightly higher grafting yield as observed in internal batch process may be attributed to higher residence time. It was also observed that both grafting reaction and cross-linking can occur during MAH grafting in the presence of DCP initiator. The crystallization temperature was significantly reduced after MAH grafting on the polymer backbone. This crystallization temperature reduction is attributed to chain branching that occurred on the polymer backbones. Thermal stability of all the maleated samples (MAH-g-PBS, MAH-g-PBAT and MAH-g- PBS/PBAT blend) were found to be reduced compared to their counterparts. This reduction can be correlated with molecular chain scission that occurred in the maleated samples. These MAH grafted polymers are expected to be a good adhesion promoter for natural fiber reinforced composites. References 1. J.W. Barlow, D.R. Paul, Mechanical compatibilization of immiscible blends. Polym. Eng. Sci., (8): p D.R. Paul, C.B. Bucknall, Polymer blends: formulation and performance. New york, Wiley:

183 3. J. Roeder, R.V.B. Oliveira, M.C. Gonçalves, V. Soldi, A.T.N. Pires., Polypropylene/polyamide-6 blends: influence of compatibilizing agent on interface domains. Polym Test, (7): p H.S. Kim, B.H. Lee, S.W. Choi, S. Kim, H.J. Kim, The effect of types of maleic anhydride-grafted polypropylene (MAPP) on the interfacial adhesion properties of bioflour-filled polypropylene composites. Composites Part A., (6): p R. Mani, M. Bhattacharya, J. Tang, Functionalization of polyesters with maleic anhydride by reactive extrusion. J. Polym. Sci. Part A: Polym. Chem., (11): p E. Cataría, C. Albano, A. Karam, R. Perera, P. Silva, J. González, Grafting of a LLDPE using gamma irradiation. Nucl. Instrum. Meth B., (1 4): p W. Qiu, T. Endo, T. Hirotsu, A novel technique for preparing of maleic anhydride grafted polyolefins. Eur. Polym. J., (9): p M. Xanthos, M.W. Young, J.A. Biesenberger, Polypropylene/polyethylene terephthalate blends compatibilized through functionalization. Polym. Eng. Sci., (6): p M. Hert, Tough thermoplastic polyesters by reactive extrusion with epoxy-containing copolymers. Angew Makromol Chem., (1): p S. Cartasegna, Silane-Grafted/Moisture-Curable Ethylene-Propylene Elastomers for the Cable Industry. Rubber. Chem.Technol., (5): p M.R. Thompson, C. Tzoganakis, G.L. Rempel, Terminal functionalization of polypropylene via the Alder Ene reaction. Polymer, (2): p

184 12. S.J. Kim, S.J. Kima, B.S. Shina, J.L. Honga, W.J. Chob, C.S. Hab, Reactive compatibilization of the PBT/EVA blend by maleic anhydride. Polymer, (9): p B. De Roover, M. Sclavons, V. Carlier, J. Devaux, R. Legras, A. Momtaz, Molecular characterization of maleic anhydride-functionalized polypropylene. J. Polym. Sci. Part A: Polym. Chem., (5): p Ganzeveld, K.J., L.P.B.M. Janssen, The grafting of maleic anhydride on high density polyethylene in an extruder. Polym. Eng. Sci., (7): p O.P. Grigoryeva, J. Karger-Kocsis, Melt grafting of maleic anhydride onto an ethylene propylene diene terpolymer (EPDM). Eur. Polym. J., (7): p P. Choudhary, S. Mohanty, S. K. Nayak, L. Unnikrishnan, Poly(L-lactide)/polypropylene blends: Evaluation of mechanical, thermal, and morphological characteristics. J. Appl. Polym. Sci., (6): p Y.F. Kim, C.N. Choi, Y.D. Kim, K.Y. Lee, M.S. Lee, Compatibilization of immiscible poly(l-lactide) and low density polyethylene blends. Fiber Polym, (4): p M. Avella, G. Bogoeva-Gaceva, A. Buzarovska, M. E. Errico, G. Gentile, A. Grozdanov, Poly(lactic acid)-based biocomposites reinforced with kenaf fibers. J. Appl. Polym. Sci., (6): p M. Avella, G. Bogoeva-Gaceva, A. Buzarovska, M. E. Errico, G. Gentile, A. Grozdanov, Poly(3-hydroxybutyrate-co-3-hydroxyvalerate)-based biocomposites reinforced with kenaf fibers. J. Appl. Polym. Sci., (5): p

185 20. Y. Nabar, J.M. Raquez, P. Dubois, R. Narayan, Production of Starch Foams by Twin- Screw Extrusion: Effect of Maleated Poly(butylene adipate-co-terephthalate) as a Compatibilizer. Biomacromolecules, (2): p Y.J. Phua, N.S. Lau, K. Sudesh, W.S. Chow, Z.A. Mohd Ishak, Biodegradability studies of poly(butylene succinate)/organo-montmorillonite nanocomposites under controlled compost soil conditions: Effects of clay loading and compatibiliser. Polym. Degrad. Stabil., (8): p J. John, J. Tang, Z. Yang, M. Bhattacharya, Synthesis and characterization of anhydridefunctional polycaprolactone. J. Polym. Sci. Part A: Polym. Chem., (6): p M.A. Huneault, H. Li, Morphology and properties of compatibilized polylactide/thermoplastic starch blends. Polymer., (1): p V. Tserki, P. Matzinos, C. Panayiotou, Effect of compatibilization on the performance of biodegradable composites using cotton fiber waste as filler. J. Appl. Polym. Sci., (7): p S.W. Hwang, S.B. Lee, C.K. Lee, J.Y. Lee, J.K. Shim,S.E.M. Selke, H. Soto-Valdez, L. Matuana, M. Rubino, R. Auras, Grafting of maleic anhydride on poly(l-lactic acid). Effects on physical and mechanical properties. Polym Test., (2): p D. Carlson, L. Nie, R. Narayan, P. Dubois, Maleation of polylactide (PLA) by reactive extrusion. J. Appl. Polym. Sci., (4): p W.S. Chow, W.L. Tham, P.C. Seow, Effects of maleated-pla compatibilizer on the properties of poly(lactic acid)/halloysite clay composites. J. Thermoplast. Compos., 2013; 26(10): pp

186 28. V.H. Orozco, W. Brostow, W. Chonkaew, B.L. López, Preparation and Characterization of Poly(Lactic Acid)-g-Maleic Anhydride + Starch Blends. Macromol. Symp., (1): p F. Chen, J. Zhang, In-situ poly(butylene adipate-co-terephthalate)/soy protein concentrate composites: Effects of compatibilization and composition on properties. Polymer., (8): p H.S. Kim, B.H. Lee, S. Lee, H.J. Kim, J.R. Dorgan, Enhanced interfacial adhesion, mechanical, and thermal properties of natural flour-filled biodegradable polymer biocomposites. J. Therm. Anal. Calorim., (1): p C.S. Wu, Physical properties and biodegradability of maleated-polycaprolactone/starch composite. Polym. Degrad. Stabil., (1): p G. Moad, The synthesis of polyolefin graft copolymers by reactive extrusion. Prog. Polym. Sci., (1): p R. Muthuraj, M. Misra, A.K. Mohanty, Fully Biodegradable poly (butylene succinate) and poly (butylene adipate-co-terephthalate) Blends: Reactive Extrusion and Performance Evaluation. J. Polym. Environ., (3): Y. J. Phua, W.S.Chow., Z. A. Mohd Ishak, Reactive processing of maleic anhydridegrafted poly(butylene succinate) and the compatibilizing effect on poly(butylene succinate) nanocomposites. Express Polymer Letters., (4): p K.E. Russell, Grafting of maleic anhydride to hydrocarbons below the ceiling temperature. J. Polym. Sci. Part A: Polym. Chem., (3): p N.G. Gaylord, R. Mehta, D.R. Mohan, V. Kumar, Maleation of linear low-density polyethylene by reactive processing. J. Appl. Polym. Sci., (11): p

187 37. S. Detyothin, S.E.M. Selke, R. Narayan, M. Rubino, R. Auras, Reactive functionalization of poly(lactic acid), PLA: Effects of the reactive modifier, initiator and processing conditions on the final grafted maleic anhydride content and molecular weight of PLA. Polym. Degrad. Stabil., (12): p G. Xiaohong, C.Q. Yang, FTIR Spectroscopy Study of the Formation of Cyclic Anhydride Intermediates of Polycarboxylic Acids Catalyzed by Sodium Hypophosphite. Textile Research Journal, (1): p Y. Cai, J. Lv, J. Feng, Spectral Characterization of Four Kinds of Biodegradable Plastics: Poly (Lactic Acid), Poly (Butylenes Adipate-Co-Terephthalate), Poly (Hydroxybutyrate- Co-Hydroxyvalerate) and Poly (Butylenes Succinate) with FTIR and Raman Spectroscopy. J. Polym. Environ., (1): p C. Li, Y. Zhang, Y. Zhang, Melt grafting of maleic anhydride onto low-density polyethylene/polypropylene blends. Polym Test., (2): p M. Z. Ahmad Thirmizir, Z. A. Mohd Ishak, R. Mat Taib, S. Rahim, S. Mohamad Jani, Natural Weathering of Kenaf Bast Fibre-Filled Poly(Butylene Succinate) Composites: Effect of Fibre Loading and Compatibiliser Addition. J. Polym. Environ., (1): p G. Moad, Chemical modification of starch by reactive extrusion. Prog. Polym. Sci., (2): p J. Cha, J.L. White, Maleic anhydride modification of polyolefin in an internal mixer and a twin-screw extruder: Experiment and kinetic model. Polym. Eng. Sci., (7): p

188 44. P. Ma, D.G. Hristova-Bogaerds, P.J. Lemstra, Y. Zhang, S. Wang, Toughening of PHBV/PBS and PHB/PBS Blends via In situ Compatibilization Using Dicumyl Peroxide as a Free-Radical Grafting Initiator. Macromol. Mater. Eng., (5): p M. Takamura, T. Nakamura, T. Takahashi, K. Koyama, Effect of type of peroxide on cross-linking of poly(l-lactide). Polym. Degrad. Stabil., (10): p H.R.Z. Sheshkali, H. Assempour, H. Nazockdast, Parameters affecting the grafting reaction and side reactions involved in the free-radical melt grafting of maleic anhydride onto high-density polyethylene. J. Appl. Polym. Sci., (4): p H. Cartier, G.H. Hu, Styrene-assisted melt free radical grafting of glycidyl methacrylate onto polypropylene. J. Polym. Sci. Part A: Polym. Chem., (7): p N.G. Gaylord, R. Mehta, V. Kumar, M. Tazi, High density polyethylene-g-maleic anhydride preparation in presence of electron donors. J. Appl. Polym. Sci., (2): p J. John, R. Mani, M. Bhattacharya, Evaluation of compatibility and properties of biodegradable polyester blends. J. Polym. Sci. Part A: Polym. Chem., (12): p P. M. Ma, R. Y. Wang, S. F. Wang, Y. Zhang, Y. X. Zhang, D. Hristova, Effects of fumed silica on the crystallization behavior and thermal properties of poly(hydroxybutyrate-co-hydroxyvalerate). J. Appl. Polym. Sci., (3): p B. Song, Y. Wang, H. Bai, L. Liu, Y. Li, J. Zhang, Z. Zhou, Crystallization and melting behaviors of maleic anhydride grafted poly(propylene) nucleated by an aryl amide derivative. J. Therm. Anal. Calorim., (2): p

189 52. T. Liu, G-H. Hu, G-S Tong, L. Zhao, G-P. Cao, W-K. Yuan, Supercritical Carbon Dioxide Assisted Solid-State Grafting Process of Maleic Anhydride onto Polypropylene. Ind. Eng. Chem. Res., (12): p

190 Chapter 5: Enhanced Mechanical Performances of Fully Biodegradable Miscanthus Fibers Reinforced Poly (butylene succinate) Composites* *A part of this chapter has been published in: R. Muthuraj, M. Misra, A. K. Mohanty, Injection Molded Sustainable Biocomposites From Poly(butylene succinate) Bioplastic and Perennial Grass, ACS Sustainable Chem. Eng., 2015, 3, (adapted with kind permission from American Chemical Society, Nov 01, 2015). Abstract Miscanthus fiber reinforced poly (butylene succinate) (PBS) composites were fabricated at various concentrations of fiber loadings. Two different compatibilizers, i.e., higher (2.5%) and lower (1.6%) degree of maleic anhydride grafted PBS (MAH-g-PBS), were used to investigate the influence of compatibilizer on mechanical performance of resulting composites. The composites compatibilized with a higher degree of MAH-g-PBS showed superior mechanical properties compared to neat PBS, uncompatibilized composites, and compatibilized composites with a lower degree of MAH-g-PBS compatibilizer. The improved mechanical properties were attributed to the enhanced interfacial adhesion that occurred between the fibers and the matrix. The optimum coupling efficiency of the compatibilizers was determined by examining 3, 5 and 10 wt% compatibilizer while fiber content was kept constant at 30 wt%. It was found that the PBS composites with 5 wt% MAH-g-PBS exhibit optimal performance. Consequently, the PBS composites were prepared with fiber loading up to 50 wt% while the higher degree (2.5%) of MAH-g-PBS compatibilizer was kept constant at 5 wt%. All the compatibilized composites had greater thermo-mechanical and mechanical properties as compared to their corresponding uncompatiblized composites as well as the neat PBS matrix. These improvements were attributed to the enhanced interfacial bonding between the components with the addition of compatabilizer. This phenomenon was confirmed by scanning electron microscopy (SEM) and theoretical 156

191 adhesion parameter values. Over all, this study provides an option for preparing a sustainable biocomposite with superior mechanical and thermo-mechanical properties. 5.1 Introduction Increasing environmental pollution, global warming and waste accumulation issues are impetus for developing sustainable and environmentally friendly biodegradable materials to replace non-biodegradable materials. One category of biodegradable materials is green composites, which can be produced from biodegradable polymer matrices with natural fibers as reinforcement. These green composite materials have been finding increased favor across packaging, horticultural, automotive and biomedical applications [1-4]. There are several biodegradable polymers commercially available in the market that are being used for green composite fabrication. Among them, poly(butylene succinate), PBS, is one of the promising candidates for green composite fabrication because it has good melt processability, relatively higher heat deflection temperature than other biodegradable polymers, good thermo-mechanical properties and biodegradability under composting environments [5]. PBS is typically produced from petroleum based monomers but also can be produced from renewable-resource-based succinic acid with bio-based content of ~54% [4]. These attractive properties may be responsible for the increase in the use of PBS based green composites for various applications [3]. The cost of PBS is more expensive than conventional non-biodegradable polymers [6]. It is well known that this shortcoming can be overcome by blending with inexpensive natural fibers/fillers while maintaining or enhancing the matrix performance. Therefore, a variety of natural fibers are used to fabricate composites with PBS matrix [4]. In addition to being inexpensive, natural fibers are renewable, sustainable, biodegradable, abundant and have good specific properties and low density compared to synthetic fibers such as 157

192 glass fiber [7]. Among the natural fibers, miscanthus fiber is mainly used for green energy production, soil preservation and composite application [8]. It is an attractive fiber for composite fabrication because of suitable fiber properties [7, 8], higher yield per hectare, lower production cost [8] and thus realistic price [9]. Similar to many other natural fibers, miscanthus fiber is thermally stable up to 200 o C without any major thermal decomposition [7, 10]. It can be processed with low melting (< 200 o C) temperature polymers in conventional processing equipment. According to Bourmaud and Pimbert [11], the miscanthus fibers have modulus of 9.49 GPa which is between hemp (12.14 GPa) and sisal (8.52 GPa) fibers. Kirwan et al., [9] reported that miscanthus fibers have mechanical properties similar to commodity thermoplastics. As a result, it can be expected that miscanthus fibers could offer good reinforcing effect in the resulting composites. Due to these inherent properties, recently, miscanthus fiber is widely used as reinforcement in biodegradable polymer matrix such as polylactide (PLA) [7], Mater-Bi [12], poly(vinyl alcohol) (PVA) [9], PLA/poly(hydroxybutyrate-co-valerate) (PHBV) blend [8] and pre-blend of poly(butylene adipate-co-terephthalate) (PBAT)/PHBV [13] matrices. From these studies, the properties of PLA/miscanthus fiber composites are comparable to some other fiber reinforced composites. Interestingly, the impact load of Mater-Bi is increased up to 30% with the addition of miscanthus fibers. Furthermore, the flexural modulus of PVA/miscanthus fiber composites is significantly higher when compared to PVA matrix. A similar occurrence has been observed in the PHBV/PLA/miscanthus and PHBV/PBAT/miscanthus composites. In general, compatibility between the natural fibers and biopolymer matrix is not adequate because of their polarity differences. Because of incompatibility between the matrixfibers, the resulting composites have yielded an inferior mechanical performance than their neat counterparts. For instance, the flexural strength and tensile strength of PHBV/PLA/miscanthus 158

193 composites are considerably reduced when compared to PHBV/PLA blend matrix [8]. Another study by Nagarajan et al., [13] showed a drastic reduction in impact strength of PHBV/PBAT pre-blend after incorporation of 30 wt% miscanthus fiber. Therefore, it is well documented that the compatibility/interfacial bonding between the fiber-matrix can be enhanced by two ways. The first method is surface modification of fibers through chemical, physical and biological process [14]. In second method, a reactive compatibilizer/coupling agent can be introduced into the composites system in order to improve the matrix-fiber interfacial adhesion [15]. In the natural fiber/filler composites, most commonly used method is reactive compatibilizer i.e., maleic anhydride (MAH) grafted functional polymers. It has tendency to form a chemical bond with surface free hydroxyl groups of the natural fibers in the resulting composites [15, 16]. Until now several types of natural fiber/filler reinforced PBS composites have been investigated in the literature [5,17-20]. There is no data available for miscanthus fibers reinforced PBS composites with improved performance. In order to overcome insufficient impact strength and stiffness of PBS, the present study was aimed to develop PBS/miscanthus fiber biocomposite with and without a reactive compatibilizer i.e., MAH grafted PBS (MAH-g-PBS). Besides, the effect of compatibilizer concentrations, grafting level of MAH and fiber loading up to 50 wt% upon the performance of resultant biocomposites were also evaluated by means of mechanical and thermo-mechanical properties. 5.2 Materials and Methods Materials Injection molding grade poly(butylene succinate) (PBS, Biocosafe 1903F) pellets were purchased from Xinfu Pharmaceutical Co., Ltd, China. The PBS pellets had melt flow index (at 190 o C with 2.16 kg load) of 22.23±1.78 g/10min, density of 1.26 g/cm 3 and melting point of 159

194 115 o C. Miscanthus fibers with average length of 4.65±2.5 mm were kindly supplied by New Energy Farms, Ontario, Canada. The chemical composition of miscanthus fiber is reported in our group earlier publications [8, 13]. The surface morphology of as received miscanthus fibers is shown in Figure 5.1. From the SEM analysis, it can be seen that the miscanthus fibers are in the form of bundules. Two different grafting level of maleic anhydride grafted PBS (MAH-g-PBS) compatibilizer (C1 and C2) is synthesized according to procedure reported in our previous Chapter (Chapter 4). The grafting level of compatibilizer C1 and C2 is 2.5 and 1.6%, respectively. Figure 5.1. SEM micrograph of as received miscanhtus fibers Thermal property Thermogravimetric analysis was performed in a TGA Q500 (TA Instruments, USA) at a heating rate of 20 ºC/min. In order to compare the thermal degradation of the samples under different atmosphere, the experiments were performed both inert (nitrogen) and air atmosphere with a flow rate of 50 ml/min. The experiment results were analyzed in TA instruments software. 160

195 5.2.3 Biocomposite preparation Both PBS and miscanthus fibers were dried for 12 h at 80 o C. After drying, the remaining moisture content of both polymers and fibers was measured in an electrical moisture analyzer (Denver Instrument) and the percentage of moisture content was 0.1% for PBS and 2.5% for miscanthus fibers. The biocomposites fabrication was perfromed in an extrusion followed by injection molding devices (DSM Xplore 15 cc microcompounder, Netherlands). The extruder consists of three collectively control heating zones, twin screws with an length and aspect ratio of 150 mm and 18, respectively. Compounding was achieved by using following processing parameters: the screw speed was set at 100 rpm; the processing temperature was set at 140 o C; and the processing/dwell time of the material inside the barrel was 2 minutes. All the test samples were molded with mould temperature of 30 o C, injection pressure of 10 bar and injection time 8 s. The composites were prepared with three different concentrations (3, 5 and 10 wt%) of compatibilizer and two different amount of functionality (C1 and C2) compatibilizer to study the influence of compatibilizer concentration and functionality upon the performance of the PBS/miscanthus composites. Furthermore, the PBS/miscanthus composites were prepared with fiber loading up to 50 wt% in the presence and absence of optimum compatibilizer (C1) concentration (5 wt%) Mechanical testing Before performing the mechanical test, test samples were conditioned at room temperature for at least 40 h. The tensile, flexural and impact test were performed according to ASTM D 638, ASTM D790 and ASTM D256, respectively. Tensile and flexural properties of the test specimens were measured in an Instron-3382 at room temperature. The tensile properties of the prepared samples were measured at a strain rate of 50 mm/min for neat PBS and 5mm/min for all the composite samples. The flexural properties were measured with a cross-head speed of 161

196 14 mm/min and a support span length of 52 mm in a three point bending mode. Notched Izod impact strength measurement was carried out in a TMI digital impact testing machine with 5 ftlb pendulum Statistical analysis Statistical significant differences of mechanical properties were identified by Minitab 17 software at 95% confidence level (P< 0.05). In order to determine significant difference among mean values of mechanical properties, one-way analysis of variance (ANOVA) with Tukey s honestly significant difference (HSD) tests was conducted with a sample size of 5 for each group Dynamic mechanical analysis The temperature dependent storage modulus (E ) of PBS and its composites is examined in a dynamic mechanical analyzer (DMA, Q800). The experiments were performed from -50 to 100 o C with a heating ramp of 3 o C/min as per ASTM D4065 standard. The storage modulus measurements were investigated in a dual cantilever mode using strain of 1Hz frequency and 15µm oscillating amplitude. Heat deflection temperature (HDT) analyses were also performed using a same DMA machine according to ASTM D648. The HDT measurements were carried out from 25 to 115 o C with a heating rate of 2 o C/min in a three point bending mode. According to ASTM D648, the reported HDT values are obtained from the temperature at which sample deflection occurred at 250 µm with a constant load of MPa Melt flow index (MFI) MFI values of the samples were determined according to ASTM D1238 standard. All the MFI measurements were performed in a melt flow indexer (Qualitest model 2000A) at 190 o C 162

197 with a constant load of 2.16 kg. Five measurements were made for each batch and their average values are reported here with standard deviations Differential scanning calorimetry (DSC) DSC anlysis was employed to study the thermal properties of PBS and its composites. The experiments were carried out in the presence of nitrogen atmosphere with a flow rate of 50 ml/min. Exactly weighed samples were heated from -50 to 160 o C with a heating rate of 10 o C/min subsequently the samples were cooling until -50 o C with a rate of 5 o C/min. After end of cooling cycle, a second heating cycle was performed from -50 to 160 o C with a heating ramp of 10 o C/min. A first heating cycle was used to erase the thermal history of the samples. A second heating and first cooling cycles are considered for the analysis. The percentage of crystallinity was calculated as follows [21]: χ = X 100% χ = X 100% (5.1) The parameter H c is crystallization enthalpy, respectively and H o m is theoretical melting enthalpy of one-hundred percentage crystalline PBS taken to be J/g [21]. The term w f is weight fraction of the fibers in the composite samples Scanning electron microscopy (SEM) To examine the fracture, surface morphology of the fractured sample was analyzed using an Inspect S50-FEI Company SEM at an accelerating voltage of 20 kv. Before observing sample morphology, all the samples were gold coated with a final thickness of 20 nm with 20 ma in order to make them electrically conductive. 163

198 5.3 Results and discussion Thermogravimetric analysis Thermal stability of the miscanthus fiber under air and nitrogen atmosphere is shown in Figure 5.2. Figure 5.2. Thermogravimetric analysis of miscanthus fiber under different environment The miscanthus fiber showed only a 2% weight loss by 105 o C in both nitrogen and air atmosphere; this 2% weight loss is attributed to the residual moisture. There was no significant further weight loss observed by 200 o C under air or nitrogen atmospheric conditions, suggesting that the miscanthus fiber can be compounded with polymers up to 200 o C without severe thermal degradation. Furthermore, the thermal stability of the miscanthus fiber was found to be similar to other natural fibers [12]. Beyond 200 o C, the degradation of miscanthus fiber was more pronounced under air atmosphere compared to nitrogen atmosphere, perhaps due to the effect of oxygen on degradation. Unlike under the nitrogen atmosphere, the weight of the char residue of the miscanthus fiber under oxygen atmosphere was almost negligible at 790 o C, this lack of char 164

199 residue may be due to the oxidation of char residue in air [22]. There were two and three major derivative weight loss peaks observed for the miscanthus fiber under N 2 and air atmosphere, respectively. The oxidation of char residue under air atmosphere can be confirmed by the observed extra derivative peak around 460 o C (Figure 5.2) Efficiency of compatibilizer The efficiency of the two different MAH grafting level compatibilizers (C1 and C2) was accessed by means of PBS composites mechanical properties. For this series of experiments, the mechanical properties of the PBS/miscanthus composites are presented in Table 5.1 with (3, 5 and 10 wt%) and without MAH-g-PBS compatibilizers. All the mechanical properties were increased except for tensile strength with addition of 30 wt% miscanthus fiber into the PBS matrix. The observed flexural strength and modulus improvement of the composites can be attributed to the reinforcing capability of miscanthus fibers [9]. The impact strength of shortfiber composites can be influenced by several mechanisms, namely matrix shear yielding, fiber pullout, fiber-matrix debonding, and fracture of fiber and matrix [23-25]. Among the impact fracture mechanisms, the fiber pullout and strong interfacial adhesion are the most effective mechanisms to improve the fracture toughness of the composites [24,26]. The impact strength of PBS was increased upon addition of 30 wt% miscanthus fibers. This increased strength could be due to the energy dissipation that occurres when matrix shear yielding and fiber pull-out terminate the unwanted growth of crazes [25]. The tensile and flexural strength values of the composites higher with addition of both compatibilizers (C1 and C2) as compared the strength of uncompatibilized composites. This increased strength suggests that the MAH grafting on the PBS backbone was successful and it acts as a compatibilizing agent between the matrix-fibers [27]. Similar tensile strength 165

200 improvement was observed in PLA/wood and PLA/wheat straw composites in the presence of MAH grafted PLA compatibilizer [27, 28]. It can be seen (Table 5.1) that the composites with C1 compatibilizer showed significantly higher flexural and tensile strength than did the composites with C2 compatibilizer. This strength improvement is likely due to the difference in molecular weight between C1 and C2 as well as number of MAH groups of the compatibilizer [29]. An appropriate molecular weight and number of MAH groups of the compatibilizer could provide better entanglement and stronger interfacial bonding between the phases [29, 30]. The formation of chemical interaction between the fiber and MAH grafted compatibilizer has been found to enhance the interfacial adhesion [31]. Since all the composites had the same amount of fiber content (30 wt%), there was no considerable change observed in flexural and tensile modulus of the composites with and without compatibilizer. The elongation at break of the PBS was substantially reduced after incorporation of miscanthus fibers; this reduction has good agreement with observed modulus of the PBS composites. Both compatibilized and uncompatiblized PBS composites showed an insignificant difference in impact strength. This similarities in the impact strength could be due to the same amount of energy being absorpted during impact fracture, no matter whether the composites were compatibilized (fiber break along with matrix) or not (fibers pull-out from matrix) [25]. Overall, it can be concluded that the composites with C1 compatibilizer (5 wt%) showed an optimum mechanical performance in comparison to uncompatibilized and compatibilized composites with C2 compatibilizer. Consequently, the PBS composites were prepared with C1 compatibilizer (5 wt%) increasing as a function of miscanthus fiber loading.the performances of the prepared composites are discussed in the next section of this research work. 166

201 Table 5.1. Mechanical properties of PBS composites with two different MAH grafting levels of compatibilizer Samples Neat PBS PBS 70 wt%+30 wt% Miscanthus PBS 67 wt%+3 wt% MAH-g-PBS +30 wt% Miscanthus PBS 65 wt%+5 wt% MAH-g-PBS +30 wt% Miscanthus PBS 60 wt%+10 wt% MAH-g-PBS +30 wt% Miscanthus PBS 67 wt%+3 wt% MAH-g-PBS +30 wt% Miscanthus PBS 65 wt%+5 wt% MAH-g-PBS +30 wt% Miscanthus PBS 60 wt%+10 wt% MAH-g-PBS +30 wt% Miscanthus Tensile strength (MPa) 39.4±0.7 (B, C) 33.2±1.7 (D) Elongation at break (%) 246±10.5 (A) 3.76±0.6 (B) Tensile Modulus (MPa) 642±12.7 (B) 2210±84.9 (A) Flexural Strength (MPa) 33.18±1.0 (E) 55.04±1.3 (D) Biocomposites prepared with C1 compatibilizer 41.3±1.5 (B, C) 44.4±2.0 (A) 45.3±1.6 (A) 3.67±0.2 (B) 3.66±0.3 (B) 4.02±0.3 (B) 2180±121 (A) 2300±91 (A) 2210±47 (A) 73.59±1.8 (A, B) 75.32±2.4 (A) 75.51±2.0 (A) Biocomposites prepared with C2 compatibilizer 37.9±1.1 (C) 38.4±0.6 (C) 41.3±0.6 (B) 3.41±0.2 (B) 3.61±0.2 (B) 3.61±0.2 (B) 2240±39 (A) 2220±69 (A) 2200±15 (A) 65.28±2.0 (C) 63.49±1.6 (C) 70.79±1.4 (B) Flexural Modulus (MPa) 759±29 (D) 2379±54 (A, B, C) 2277±104 (C) 2346±98 (B, C) 2347±211 (B, C) 2556±60 (A) 2503±56 (A, B) 2565±83 (A) Means that do not share a letter in parentheses are significantly different at p<0.05 level. Notched Izod impact strength (J/m) 28.40±0.91 (C) 46.42±1.4 (A, B) 47.48±4.8 (B) 53.23±3.5 (A) 52.38±1.2 (A) 45.65±1.0 (B) 46.22±0.98 (B) 47.01±1.70 (B) Mechanical properties versus fiber loading The tensile properties of the natural fiber reinforced composites are dependent on both the matrix and fiber characters. The tensile strength of natural fiber composite is mainly influenced by three factors such as interfacial interaction between the components fiber orientation, and stress concentration [23]. On the other hand, wettability of fibers by the matrix, 167

202 volume fraction of fibers, and aspect ratio of fibers determine tensile modulus of the natural fiber composite [23]. Figure 5.3 shows the variation of tensile strength and tensile modulus of PBS composites with different fiber loadings with and without compatibilizer. The neat PBS had tensile strength of 39 MPa and tensile modulus of 0.66 GPa. Due to lack of interfacial interaction between the phases and incompatibility between the miscanthus fiber and the PBS, the tensile strength decreased significantly after inclusion of miscanthus fibers into the PBS. Particularly, the addition of the 50 wt% miscanthus fiber into the PBS matrix yielded a 22% reduction in tensile strength. As expected, the tensile modulus of the composites can be influenced by the fiber content in the composite matrix. Consequently, the tensile modulus of the PBS composites was steeply increased with increasing miscanthus fiber content up to 50 wt%. For instance, the PBS composite with 50 wt% miscanthus fiber showed a maximum tensile modulus value of 3.88 GPa, which is 475% higher than that of neat PBS. This improvement was due to the reinforcing effect of miscanthus fiber in the PBS matrix. Among the tensile modulus and tensile strength of the composites, tensile strength was influenced by the addition of compatibilizer. The tensile strength of all the compatibilized composites was significantly higher than that of the uncompatibilized composites as well as of PBS matrix. The observed improvement in the tensile strength of the compatibilized composites is possibly due to an enhanced interfacial adhesion between the components. In the presence of compatibilizer, the tensile strength improvement of the PBS/miscanthus composite leveled off beyond 30 wt% fiber loading. Furthermore, the tensile modulus of the compatibilized composites did not show any substantial difference compared to that of the corresponding uncompatibilized composites. This observation is consistent with PLA/wood composites [27]. 168

203 Figure 5.3. Tensile properties of PBS and PBS/miscanthus composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) The load-bearing capacity/reinforcing effect of the particulate or short-fiber-filled composites can be expressed quantitatively with the help of a simple model developed earlier (equation 5.2) [32-36]. This equation can be used to describe the composition-dependent tensile properties of the biocomposites [27, 33]. σ = exp (Bφ) (5.2) The term is tensile strength of neat matrix; σ is tensile strength of composites; λ is ratio (L/L0) of length measured before (L0) and after (L) the tensile test; n is the strain hardening parameter; B is the load-bearing capacity of the dispersed component, which depends on interfacial adhesion/interaction; φ is the volume fraction of fibers in the composites. From equation (5.2), n can be neglected when strain hardening tendency of the matrix is low [32]. Then, equation (5.2) can be written as 169

204 σ = exp (Bφ) (5.3) Equation 5.4 is obtained from simplified linear form of equation 5.3 [32, 36] ln σ red = ln σ = ln + Bφ (5.4) If we plot the natural logarithm of reduced tensile strength as a function of fiber content, this reduced tensile strength must give a straight line with a slope of B, which corresponds to load-bearing capacity [34,36]. In Figure 5.4, the tensile strength of both compatibilized and uncompatibilized composites is plotted against fiber content in equation 5.4. It was observed that a linear correlation occurred for compatibilized and uncompatibilized composites with dissimilar slope (B) value. The B value was proportional to the load-bearing capacity/reinforcing effect of the resulting composites. The compatibilized composites showed superior load-bearing capacity with B value of 4.85 while uncompatibilized composites yielded least reinforcing effect with B value of The enhanced load-bearing capacity of the compatibilized composites was consistent with observed tensile modulus and strength of the PBS/miscanthus composite. The B value (4.85) of the compatibilized PBS/composites found in this study was much higher than compatibilized PLA/wood composites [27, 32, 35], suggesting that the load-bearing capacity of compatibilized PBS/miscanthus composites is greater than compatibilized PLA/wood composites. 170

205 Figure 5.4. Reduced tensile strength of uncompatibilized and compatibilized PBS/miscanthus composites plotted against volume fraction of fibers according to equation 5.4. Flexural properties of PBS and its compatibilized and uncompatibilized composites are given in Figure 5.5. Unlike tensile strength, the flexural strength of all the uncompatibilized composites is remarkably higher as compared to neat PBS. This increase can be attributed to the enhanced stiffness of the PBS matrix after incorporation of the miscanthus fibers. However, there is a slight reduction in flexural strength of uncompatibilized composites with increasing fiber loadings from 30 to 50 wt%. For instance, the flexural strength of uncompatibilized composites with 50 wt% miscanthus fiber is 12% lower than that of uncompatibilized composites with 30 wt% fiber. This reduction is attributed to the agglomeration of fibers and thus leads to uneven dispersion of the fibers in the matrix. In contrast, flexural modulus of the composites gradually increased with increasing fiber content up to 50 wt%. This trend has good agreement with tensile modulus observation in this current study. 171

206 Figure 5.5. Flexural properties of PBS and PBS/miscanthus composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) Regarding flexural modulus, there was insignificant difference observed between the compatibilized and uncompatibilized composites. In contrary, the compatibilized composites exhibit superior flexural strength compared to uncompatibilized composites. Similar to tensile strength of compatibilized PBS composites; the flexural strength of compatibilized PBS composites levels off when fiber loading is increased from 30 to 50 wt%. Compared to neat PBS, the flexural strength and flexural modulus of compatibilized PBS composites with 50 wt% miscanthus fibers were found to increase 139 and 515%, respectively. The observed improvement in the compatibilized composites could be due to the strong reinforcing effect of miscanthus fibers and reduced fiber agglomeration in the matrix. The notched Izod impact strength of PBS and its composites with and without compatibilizer is shown in Figure 5.6. The impact strength of neat PBS is around 28 J/m, which 172

207 can be comparable to homopolypropylene [37]. Unlike impact strength of compression molded PBS/30 wt% kenaf composites [38] and injection molded PLA/30 wt% wheat straw composites [28], there is a trend for impact strength of injection molded PBS composite to increase with miscanthus fibers loading up to 40 wt%. In contrast, the PBS composites with 50 wt% miscanthus fibers did not show any significant impact strength improvement when compared to neat PBS. This could be due to the uneven fiber dispersion in the matrix when composites contain more fibers i.e., 50 wt%. The impact strength improvement was more pronounced in the compatibilized composites in comparison to their corresponding uncompatibilized composites. When impact strength of compatibilized and uncompatibilized PBS/miscanthus composites was compared, a maximum improvement (47%) was noticed in compatibilized composites with 50 wt% fiber loading. Avella et al., [39, 40] have observed a similar trend in compatibilized PHBV/kenaf composites and compatibilized PLA/kenaf fiber composites. According to these studies, the observed improvement was due to the better adhesion between the components, less fiber pullout under impact load, and enhanced uniform fiber dispersion in the matrix. However, the impact strength of compatibilized composites gradually declined with increasing fiber loading from 30 to 50 wt%, which follows a similar trend to uncompatibilized composites. This could be due to the fiber-fiber contact that was increased with increasing fiber loadings. When the fiber content is increased in the polymer matrix system, the fibers tend to form an agglomeration in the matrix due to strong hydrophilic fiber to fiber interaction. The agglomeration creates strong stress concentration regions that require less energy to elongate the crack propagation under selected impact testing condition. Therefore, the composites with fiber agglomeration lead to less impact strength [41]. 173

208 In order to compare the effectiveness of this work, mechanical properties of injection molded PBS biocomposites as described in the literature are presented in Table 5.2. It can be seen that bamboo, ramie and flax fiber reinforced PBS composites tensile strength is greater than that of PBS/miscanthus composites. In contrast, miscanthus fiber reinforced PBS composites enhanced tensile modulus by 473% which is significantly higher than other natural fiber (bamboo, flax, hemp, wood and waste silk) reinforced PBS composites [5, 42, 43]. This tensile modulus increase indicates that the reinforcing effect of miscanthus fibers is obvious. Even though chopped short miscanthus fibers were used in the current study, the flexural properties of the PBS/miscanthus composite are quite comparable with continuous basalt fiber roving/pbs composites [42]. In addition, the PBS/miscanthus fiber composites exhibit superior flexural properties compared to PBS composites with ramie, kenaf and waste silk fibers. The notched Izod impact strength of PBS/miscanthus fibers composite was about 47% higher in comparison to neat PBS. When kenaf (30 wt% ) short fibers were incorporated into PBS, the impact strength of neat PBS was reduced by ~79% (from 40 to 11 J/m 2 ) [38]. In the present study, the impact strength was improved about 47% in comparison to neat PBS. From the above, it clearly suggests that the improved mechanical performances of the PBS/miscanthus composites can be achieved by the simple method. 174

209 Figure 5.6. Nothced Izod impact strength of PBS and PBS/miscanthus composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) 175

210 Table 5.2. A comparison of mechanical properties of injection molded PBS/natural fiber composites (Note: the reported percentage differences were calculated based on the neat PBS matrix properties) Fiber type Optimum fiber (wt%) Type of compatibilizer used Change in tensile properties (%) Change in flexural properties (%) Strength Modulus Strength Modulus Change in Izod impact strength (%) References Bamboo 15 poly(ethylene glycol) ~76 ~52 NA NA NA [51] methacrylate Ramie 30 Alkali treatment ~55 ~431 ~86 ~267 NA [5] Kenaf 30 Maleic anhydride grafted PBS NA NA ~23 ~399 ~(-79) [38] Flax 30 Maleic anhydride grafted PBS ~28 ~358 NA NA NA [43] Wood 30 Maleic anhydride grafted PBS ~5 ~329 NA NA NA [43] Hemp 30 Maleic anhydride grafted PBS ~13 ~353 NA NA NA [43] Bamboo 30 Maleic anhydride grafted PBS ~(-16) NA NA NA NA [52] Wood 30 Maleic anhydride grafted PBS ~(-40) NA NA NA NA [52] Waste silk 40 NA ~17 ~140 ~26 ~54 NA [53] Miscanthus 50 Maleic anhydride grafted PBS ~22 ~473 ~139 ~515 ~47 Present study 176

211 5.3.4 Dynamic mechanical analysis DMA was used to investigate the viscoelastic properties of PBS and its composites with respect to temperatures. The interfacial adhesion between the matrix and the fibers can be estimated by loss factor (tan δ) and storage modulus (E ) properties. For example, the E values of the natural fiber composites are sensitive to the dispersion of fibers in the matrix as well as interfacial bonding between the phases. Figure 5.7 illustrates the storage modulus of PBS and its composites over the wide temperature range. It can be remarked that the E value of composites monotonically increased with the addition of miscanthus fibers up to 50 wt%. This can be attributed to the reinforcing effect of the miscanthus fibers in the PBS matrix. In addition, the E of compatibilized composites is slightly higher with respect to corresponding uncompatibilized composites. For instance, the storage modulus of compatibilized composites with 50 wt% miscanthus fibers showed 6.8 GPa at -60 o C, which is higher than corresponding uncompatibilized composites (6.52 GPa) as well as neat PBS (2.98 GPa) at the glassy region (- 60 o C). The same trend has been observed across the entire investigated range of temperature. This suggests that compatibilized composites had uniform fiber dispersion and the high degree of interaction between the phases. However, all the samples showed a drastic reduction of E at - 16 o C, which is attributed to the glass transition temperature (tan δ) of PBS. Moreover, the E of all the samples were gradually decreased with increasing temperature up to 100 o C. This decrease in E is an expected consequence of the molecular motion/relaxation while increasing temperatures. 177

212 Figure 5.7. Dynamic mechanical analysis of PBS and PBS/miscanthus composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) Figure 5.8 shows the tan δ curves of PBS and its composites with different weight percentage fiber loadings. The glass transition temperature (position of tan δ peak maximum) of the PBS is not affected considerably as content of miscanthus fiber increases up to 50 wt%. In contrast, the height of the loss factor (tan δ) of the PBS was reduced by incorporation of miscanthus fibers. This reduction corresponds to the stiffness improvement of the PBS in the presence of fibers and is evidenced by increase of the E. The height of the tan δ value was reduced in the compatibilized composites compared to their corresponding uncompatibilized counterparts. These reductions were due to the improved interaction between the phases with addition of compatibilizer. This can be further confirmed by evaluating the interface adhesion factor (A) between the fiber and matrix. 178

213 Figure 5.8. Tan δ curves of PBS and its composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) Adhesion factor calculation From the height of tan δ peak values, Kubar et al., [44] described a methodology to evaluate degree of interaction between the composite components. When there is strong interfacial interaction/adhesion between the phases and reduction of macromolecular mobility around the reinforcement surface, the value of adhesion factor (A) decreases. Thus, lower values of adhesion factor (A) are evidence of a high degree of interactions between the fibers and the matrix. According to this methodology, the adhesion/interaction between the fiber and the matrix can be estimated by adhesion factor (A), which was calculated in the following equation [44]: A = 1 (5.5) where tan δ M (T) and tan δ C (T) represent the relative damping values of the neat matrix and the composite at a given temperature, respectively, and V f is the volume fraction of fiber. 179

214 Figure 5.9. Adhesion factor of PBS/miscanthus composites: A) PBS/miscanthus (70/30 wt%), B) PBS/miscanthus/compatibilizer (65/30/5 wt%), C) PBS/miscanthus (60/40 wt%), D) PBS/miscanthus/compatibilizer (55/40/5 wt%), E) PBS/miscanthus (50/50 wt%), and F) PBS/miscanthus/compatibilizer (45/50/5 wt%) Figure 5.9 shows the adhesion factor values of the composites with respect to temperatures. Unlike composites with higher fiber content (40 and 50 wt%), the composites with lower fiber content (30 wt%) showed lower adhesion factor values. Better interfacial interaction between components in the composites with 30 wt% fibers in contrast to composites with 50 wt% fibers content indicates the lower adhesion factor value. Moreover, the adhesion factor value of all the compatibilized composites was found to be lower in comparison to that of corresponding uncompatibilized composites. The lower adhesion factor value may be result of further enhancement of the interfacial adhesion between the fibers and the matrix in the presence of compatibilizer Heat deflection temperature HDT value is used to determine physical deformation of a polymeric material at elevated temperatures with a set of testing conditions. The HDT values of neat PBS and its compatibilized 180

215 and uncompatibilized composites are presented in Table 5.3. The HDT value of neat PBS was around 90 o C, and PBS shows significant increase in HDT with increasing miscanthus fiber loading from 30 to 50 wt%. For instance, the PBS with 40 wt% miscanthus fiber composites showed a 29% improvement in comparison to neat PBS. The observed HDT improvements of PBS composites were attributed to the enhanced stiffness/reinforcement effect of resulting composites, as reports in literature [8]. The enhanced stiffness of all the PBS composites has good agreement with the observed tensile and flexural modulus. Similarly, the HDT value of PBS/basalt fiber (85/15 wt%) composites [42] and PBS/switchgrass (50/50 wt%) composites [19] was increased by 40 and 36%, respectively, in comparison to neat PBS. The HDT values of compatibilized and uncompatibilized composites were not significantly different in the present study. This can be ascribed to the HDT values of all the composites being very close to their melting temperature i.e., ~114 o C. Therefore, it can be concluded that an optimum HDT value of PBS composite is 115 o C, which was able to be achieved in the PBS composites with 40 wt% miscanthus fiber loading. Table 5.3. Heat deflection temperature (HDT) and melt flow index (MFI) of neat PBS and its composites Samples HDT ( o C) MFI (g/10 min) Neat PBS 89.59± ± 1.78 PBS/miscanthus fibers (70/30 wt%) ± ± 1.11 PBS/miscanthus fibers/mah-g-pbs (65/30/5 wt%) ± ± 0.55 PBS/miscanthus fibers (60/40 wt%) ± ± 0.16 PBS/miscanthus fibers/mah-g-pbs (55/40/5 wt%) ± ± 0.46 PBS/miscanthus fibers (50/50 wt%) ± ± 0.15 PBS/miscanthus fibers/mah-g-pbs (45/50/5 wt%) ± ±

216 5.3.7 Melt flow analysis The melt processability of the PBS and its composites was assessed by MFI analysis. Table 5.3 represents the effect of fiber content and compatibilizer on the MFI value of PBS composites. It was found that the MFI value of neat PBS is around 22 g/10 min. Unlike neat PBS, the MFI value of PBS composites was drastically decreased with increasing fiber content from 30 to 50 wt%. For example, the lowest MFI value (~1 g/10min) was observed for the uncompatibilized PBS with 50 wt% miscanthus fiber composites. In general, the ability of polymer melt flow is hampered in the presence of rigid fibers and fillers. Recently, this type of MFI reduction was observed in PLA/50 wt% jute fiber composites and PLA/50 wt% hemp fiber composites by Gunning et al., [45]. However, it is worth noting that all the compatibilized composites showed a marginal improvement in MFI value when compared to that of corresponding uncompatibilized composites. This increased MFI value of the compatibilized composites is attributed to the good fiber dispersion in the matrix, which can enhance the flowability of composite materials [45]. Over all, the observed MFI value of the PBS composites is appropriate for injection molding applications [46]. Table 5.4. Summary of differential scanning calorimetry traces of neat PBS and its composites Samples T c ( o C) H c (J/g) T m ( o C) H m (J/g) Crystallinity (%) Neat PBS PBS/miscanthus fibers (70/30 wt%) PBS/miscanthus fibers/mah-g-pbs (65/30/5 wt%) PBS/miscanthus fibers (60/40 wt%) PBS/miscanthus fibers/mah-g-pbs (55/40/5 wt%) PBS/miscanthus fibers (50/50 wt%) PBS/miscanthus fibers/mah-g-pbs (45/50/5 wt%)

217 5.3.8 Differential scanning calorimetry The effect of the incorporated miscanthus fibers on the melting temperature (T m ), melting enthalpy ( H m ), crystalline temperature (T c ), crystalline enthalpy ( H c ) and percentage crystallinity of PBS was analysed by DSC. For PBS and its composites, the DSC second heating and first cooling cycle results are presented in Figures 5.10 and 5.11, respectively. Table 5.4 shows the melting enthalpy, crystallization enthalpy and percentage crystallinity of PBS before and after incorporation of miscanthus fibers. The melting temperature and crystallization temperature of neat PBS are found to be 113 and 93 o C, respectively. The melting temperature of PBS and its composites is very similar, which suggests that the PBS melting temperature is not affected in the presence of miscanthus fibers. Figure DSC second heating thermograms of PBS and its composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) A similar trend was observed in crystallization temperature and percentage of crystallinity of PBS after addition of miscanthus fibers. During the DSC heating cycle, a broad bimodal melting peak for neat PBS was observed and can be attributed to the 183

218 meltrecrystallization phenomena. This bimodal melting peak of PBS became more distinct after addition of miscanthus fibers. PBS crystallinity percentage was not heavily affected after incorporation of miscanthus fibers; consequently, it can be expected that the biodegradability of the PBS composites will be similar to that of neat PBS [47]. Figure DSC first cooling thermograms of PBS and its composites: A) neat PBS, B) PBS/miscanthus (70/30 wt%), C) PBS/miscanthus/compatibilizer (65/30/5 wt%), D) PBS/miscanthus (60/40 wt%), E) PBS/miscanthus/compatibilizer (55/40/5 wt%), F) PBS/miscanthus (50/50 wt%), and G) PBS/miscanthus/compatibilizer (45/50/5 wt%) Morphological analysis SEM was employed to investigate the interfacial bonding between the component in the composites and the degree of fiber dispersion in the matrix. The tensile fractured surface morphology of uncompatibililized and compatibilized PBS composites with different fiber loadings is presented in Figures 5.11 and SEM micrographs of uncompatibilized composites (Figure 5.12) reveal interfacial gaps between the fiber and the matrix, poor fiber wetting, fibers pullout traces, and poor dispersion of fibers in the matrix. Such phenomena can clearly be observed in higher (500x) magnification SEM images (Figure b, d and f). A 184

219 similar observation has been reported in PLA/kenaf composites [40], PHBV/kenaf composites [39], PHBV/PLA/miscanthus composites [8], PBS/bamboo composites [48], PBS/kenaf fiber composites [38] and PP/bio-flour composites [29]. In addition, increased fiber bundles/aggregates are evident with increasing fiber content up to 50 wt%. This is possibly due to an increase fiber-fiber interaction with increasing fiber contents in the matrix, which could hinder the interaction between the phases in the composites [49]. This contributes to a reduction in the tensile strength of resulting PBS composites with increasing fiber content from 30 to 50 wt%. It is well documented that most of the natural fibers are not compatible with a hydrophobic polymer matrix, this lack of compatibility is responsible for fiber debonding from the matrix during tensile fracture [29, 50]. Figure 5.12 shows SEM micrographs of compatibilized PBS composites at lower (150x) and higher (500x) magnification. All the compatibilized composites clearly showed paucity of fiber pullout traces from the matrix, good fiber dispersion in the matrix, and fibers wholly embedded into the PBS matrix due to strong adhesion at the interfacial regions compared to that of uncompatibilized composites. These observations are more pronounced in higher (500x) magnification SEM micrographs (Figure b, d and f). In the compatibilized composites, it can be seen (Figure 5.13) that many fibers are coated with PBS matrix without the interfacial gap formation. This result was attributed to the enhanced fiber/matrix adhesion with the help of a reactive compatibilizer [29]. Consequently, all the compatibilized composites enhanced mechanical performances as compared to their corresponding uncompatibilized composites, which is also consistent with the observed lower adhesion factor A values of compatibilized composites. Similar findings have been recently reported for the compatibilized PBS/kenaf fiber composites [38]. 185

220 Figure SEM micrographs of tensile fractured surface of uncompatibilized PBS composites with low (150x) and high (500x) magnification; (a) and (b) are PBS/miscanthus (70/30 wt%) composites; (c) and (d) are PBS/miscanthus (60/40 wt%) composites; (e) and (f) are PBS/miscanthus (50/50 wt%) composites. 186

221 Figure SEM micrographs of tensile fractured surface of compatibilized PBS composites with low (150x) and high (500x) magnification; (a) and (b) are PBS/miscanthus (70/30 wt%) composites; (c) and (d) are PBS/miscanthus (60/40 wt%) composites; (e) and (f) are PBS/miscanthus (50/50 wt%) composites. 187

222 5.4 Conclusions Biocomposites from PBS and miscanthus fibers were successfully prepared by extrusion and injection molding methods with different fiber loadings. The performances of the resulting composites were evaluated based on mechanical, thermo-mechanical, and morphological properties. The strong reinforcing effect of micanthus fibers led to an increase in the stiffness/modulus of resulting PBS composites. The tensile strength of uncompatibilized PBS/miscanthus composites was much lower compared to that of neat PBS. Unlike tensile strength, the flexural and impact strengths were significantly enhanced after incorporation of miscanthus fibers into the PBS matrix. The enhanced flexural strength was attributed to the reinforcing effect of miscathus fibers. The fiber pullout mechanism is likely responsible for the observed impact strength improvement. Addition of 5 wt% MAH-g-PBS into PBS composites resulted a significant improvement in tensile and flexural strength compared to the corresponding uncompatibilized composites and neat matrix. For example, the PBS composites with 50 wt% miscanthus and 5 wt% MAH-g-PBS resulted in 22, 139 and 47% improvements in tensile, flexural and impact strength compared to neat PBS. These improvements were attributed to the enhanced interfacial interaction between the components, as confirmed by adhesion parameter values and by surface morphological analysis. Furthermore, the enhanced thermo-mechanical properties were consistent with tensile and flexural modulus. Although, the MFI value of the PBS was considerably reduced after incorporation of miscanthus fibers, the observed MFI value of the PBS composites is still appropriate for the injection molding process. Because the PBS crystallinity was not affected by incorporation of miscanthus fibers, it can be expected that the biodegradability of the PBS composites will be similar to that of neat PBS. 188

223 References [1] M. P. M. Dicker, P. F. Duckworth, A. B. Baker, G. Francois, M. K. Hazzard, P. M. Weaver, Green composites: A review of material attributes and complementary applications, Composites Part A: Applied Science and Manufacturing, 2014, 56 (0): [2] F. P. La Mantia, M. Morreale, Green composites: A brief review, Composites Part A: Applied Science and Manufacturing, 2011, 42 (6): [3] G. Koronis, A. Silva, M. Fontul, Green composites: A review of adequate materials for automotive applications, Composites Part B: Engineering, 2013, 44 (1): [4] E. Zini, M. Scandola, Green composites: An overview, Polymer Composites, 2011, 32 (12): [5] M. Zhou, J. Yan, Y. Li, C. Geng, C. He, K. Wang, Q. Fu, Interfacial strength and mechanical properties of biocomposites based on ramie fibers and poly(butylene succinate), RSC Advances, 2013, 3 (48): [6] A. Hamma, M. Kaci, Z. A. Mohd Ishak, A. Pegoretti, Starch-grafted-polypropylene/kenaf fibres composites. Part 1: Mechanical performances and viscoelastic behaviour, Composites Part A: Applied Science and Manufacturing, 2014, 56 (0): [7] G. Gamon, P. Evon, L. Rigal, Twin-screw extrusion impact on natural fibre morphology and material properties in poly(lactic acid) based biocomposites, Industrial Crops and Products, 2013, 46 (0): [8] M. R. Nanda, M. Misra, A. K. Mohanty, Performance Evaluation of Biofibers and Their Hybrids as Reinforcements in Bioplastic Composites, Macromolecular Materials and Engineering, 2013, 298 (7):

224 [9] K. Kirwan, R. M. Johnson, D. K. Jacobs, G. F. Smith, L. Shepherd, N. Tucker, Enhancing properties of dissolution compounded Miscanthus giganteus reinforced polymer composite systems: Part 1. Improving flexural rigidity, Industrial Crops and Products, 2007, 26 (1): [10] M. Johnson, N. Tucker, S. Barnes, K. Kirwan, Improvement of the impact performance of a starch based biopolymer via the incorporation of Miscanthus giganteus fibres, Industrial Crops and Products, 2005, 22 (3): [11] A. Bourmaud, S. Pimbert, Investigations on mechanical properties of poly(propylene) and poly(lactic acid) reinforced by miscanthus fibers, Composites Part A: Applied Science and Manufacturing, 2008, 39 (9): [12] R. M. Johnson, N. Tucker, S. Barnes, Impact performance of Miscanthus/Novamont Mater-Bi biocomposites, Polymer Testing, 2003, 22 (2): [13] V. Nagarajan, A. K. Mohanty, M. Misra, Sustainable Green Composites: Value Addition to Agricultural Residues and Perennial Grasses, ACS Sustainable Chemistry & Engineering, 2013, 1 (3): [14] O. Faruk, A. K. Bledzki, H.-P. Fink, M. Sain, Progress Report on Natural Fiber Reinforced Composites, Macromolecular Materials and Engineering, 2014, 299 (1): [15] T. Yu, N. Jiang, Y. Li, Study on short ramie fiber/poly(lactic acid) composites compatibilized by maleic anhydride, Composites Part A: Applied Science and Manufacturing, 2014, 64 (0): [16] M. J. A. van den Oever, H. W. Elbersen, E. R. P. Keijsers, R. J. A. Gosselink, B. de Klerk-Engels, Switchgrass (Panicum virgatuml.) as a reinforcing fibre in polypropylene composites, Journal of Materials Science, 2003, 38 (18):

225 [17] G. Dorez, A. Taguet, L. Ferry, J. M. Lopez-Cuesta, Thermal and fire behavior of natural fibers/pbs biocomposites, Polymer Degradation and Stability, 2013, 98 (1): [18] S. M. Lee, D. Cho, W. H. Park, S. G. Lee, S. O. Han, L. T. Drzal, Novel silk/poly(butylene succinate) biocomposites: the effect of short fibre content on their mechanical and thermal properties, Composites Science and Technology, 2005, 65 (3 4): [19] S. Sahoo, M. Misra, A. K. Mohanty, Effect of compatibilizer and fillers on the properties of injection molded lignin-based hybrid green composites, Journal of Applied Polymer Science, 2013, 127 (5): [20] H.-S. Kim, H.-J. Kim, J.-W. Lee, I.-G. Choi, Biodegradability of bio-flour filled biodegradable poly(butylene succinate) bio-composites in natural and compost soil, Polymer Degradation and Stability, 2006, 91 (5): [21] Y. J. Phua, W. S. Chow, Z. A. Mohd Ishak, The hydrolytic effect of moisture and hygrothermal aging on poly(butylene succinate)/organo-montmorillonite nanocomposites, Polymer Degradation and Stability, 2011, 96 (7): [22] F. Sliwa, N.-e. El Bounia, G. Marin, F. Charrier, F. Malet, A new generation of wood polymer composite with improved thermal stability, Polymer Degradation and Stability, 2012, 97 (4): [23] D. N. Saheb, J. P. Jog, Natural fiber polymer composites: A review, Advances in Polymer Technology, 1999, 18 (4): [24] F. Ramsteiner, R. Theysohn, Tensile and impact strengths of unidirectional, short fibrereinforced thermoplastics, Composites, 1979, 10 (2):

226 [25] J. Wu, D. Yu, C.-M. Chan, J. Kim, Y.-W. Mai, Effect of fiber pretreatment condition on the interfacial strength and mechanical properties of wood fiber/pp composites, Journal of Applied Polymer Science, 2000, 76 (7): [26] D. A. Norman, R. E. Robertson, The effect of fiber orientation on the toughening of short fiber-reinforced polymers, Journal of Applied Polymer Science, 2003, 90 (10): [27] Á. Csikós, G. Faludi, A. Domján, K. Renner, J. Móczó, B. Pukánszky, Modification Of Interfacial Adhesion With A Functionalized Polymer In Pla/Wood Composites, European Polymer Journal, 2015, 68 (0): [28] C. Nyambo, A. K. Mohanty, M. Misra, Effect of Maleated Compatibilizer on Performance of PLA/Wheat Straw-Based Green Composites, Macromolecular Materials and Engineering, 2011, 296 (8): [29] H.-S. Kim, B.-H. Lee, S.-W. Choi, S. Kim, H.-J. Kim, The effect of types of maleic anhydride-grafted polypropylene (MAPP) on the interfacial adhesion properties of bioflour-filled polypropylene composites, Composites Part A: Applied Science and Manufacturing, 2007, 38 (6): [30] T. J. Keener, R. K. Stuart, T. K. Brown, Maleated coupling agents for natural fibre composites, Composites Part A: Applied Science and Manufacturing, 2004, 35 (3): [31] A. K. Mohanty, M. Misra, L. T. Drzal, Surface modifications of natural fibers and performance of the resulting biocomposites: An overview, Composite Interfaces, 2001, 8 (5):

227 [32] R. Csizmadia, G. Faludi, K. Renner, J. Móczó, B. Pukánszky, PLA/wood biocomposites: Improving composite strength by chemical treatment of the fibers, Composites Part A: Applied Science and Manufacturing, 2013, 53 (0): [33] P. Müller, K. Renner, J. Móczó, E. Fekete, B. Pukánszky, Thermoplastic starch/wood composites: Interfacial interactions and functional properties, Carbohydrate Polymers, 2014, 102 (0): [34] G. Faludi, J. Hári, K. Renner, J. Móczó, B. Pukánszky, Fiber association and network formation in PLA/lignocellulosic fiber composites, Composites Science and Technology, 2013, 77 (0): [35] G. Faludi, G. Dora, K. Renner, J. Móczó, B. Pukánszky, Improving interfacial adhesion in pla/wood biocomposites, Composites Science and Technology, 2013, 89 (0): [36] G. Faludi, G. Dora, B. Imre, K. Renner, J. Móczó, B. Pukánszky, PLA/lignocellulosic fiber composites: Particle characteristics, interfacial adhesion, and failure mechanism, Journal of Applied Polymer Science, 2014, 131 (4): [37] R. Muthuraj, M. Misra, A. K. Mohanty, Hydrolytic degradation of biodegradable polyesters under simulated environmental conditions, Journal of Applied Polymer Science, 2015, 132, [38] M. Z. A. Thirmizir, Z. A. M. Ishak, R. M. Taib, S. Rahim, S. M. Jani, Kenaf-bast-fiberfilled biodegradable poly(butylene succinate) composites: Effects of fiber loading, fiber length, and maleated poly(butylene succinate) on the flexural and impact properties, Journal of Applied Polymer Science, 2011, 122 (5): [39] M. Avella, G. Bogoeva-Gaceva, A. Buzõarovska, M. Emanuela Errico, G. Gentile, A. Grozdanov, Poly(3-hydroxybutyrate-co-3-hydroxyvalerate)-based biocomposites 193

228 reinforced with kenaf fibers, Journal of Applied Polymer Science, 2007, 104 (5): [40] M. Avella, G. Bogoeva-Gaceva, A. Bužarovska, M. E. Errico, G. Gentile, A. Grozdanov, Poly(lactic acid)-based biocomposites reinforced with kenaf fibers, Journal of Applied Polymer Science, 2008, 108 (6): [41] M.S. Huda, L.T. Drazel, D. Ray, A.K. Mohanty, M. Misra, Natural-fiber composites in the automotive sector: In, Properties and Perfromance of Natural-Fiber Composites (Edt:K Pickering), Woodhead Publishing Limited, England, 248, [42] Y. Zhang, C. Yu, P. K. Chu, F. Lv, C. Zhang, J. Ji, R. Zhang, H. Wang, Mechanical and thermal properties of basalt fiber reinforced poly(butylene succinate) composites, Materials Chemistry and Physics, 2012, 133 (2 3): [43] V. Tserki, P. Matzinos, N. E. Zafeiropoulos, C. Panayiotou, Development of biodegradable composites with treated and compatibilized lignocellulosic fibers, Journal of Applied Polymer Science, 2006, 100 (6): [44] J. Kubát, M. Rigdahl, M. Welander, Characterization of interfacial interactions in high density polyethylene filled with glass spheres using dynamic-mechanical analysis, Journal of Applied Polymer Science, 1990, 39 (7): [45] M. A. Gunning, L. M. Geever, J. A. Killion, J. G. Lyons, C. L. Higginbotham, Improvement in mechanical properties of grafted polylactic acid composite fibers via hot melt extrusion, Polymer Composites, 2014, 35 (9): [46] ng/kr aft_papers/brochures/fibromer_broschuere_low_res.pdf. 194

229 [47] Y. Du, T. Wu, N. Yan, M. T. Kortschot, R. Farnood, Fabrication and characterization of fully biodegradable natural fiber-reinforced poly(lactic acid) composites, Composites Part B: Engineering, 2014, 56 (0): [48] S. Lee, S. Wang, Biodegradable polymers/bamboo fiber composite with bio-based coupling agent, Composites Part A: Applied Science and Manufacturing, 2006, A [49] M. N. Islam, M. R. Rahman, M. M. Haque, M. M. Huque, Physico-mechanical properties of chemically treated coir reinforced polypropylene composites, Composites Part A: Applied Science and Manufacturing, 2010, 41 (2): [50] H. Demir, U. Atikler, D. Balköse, F. Tıhmınlıoğlu, The effect of fiber surface treatments on the tensile and water sorption properties of polypropylene luffa fiber composites, Composites Part A: Applied Science and Manufacturing, 2006, 37 (3): [51] L. Bao, Y. Chen, W. Zhou, Y. Wu, Y. Huang, Bamboo poly(ethylene glycol)- reinforced poly(butylene succinate) biocomposites, Journal of Applied Polymer Science, 2011, 122 (4): [52] H.-S. Kim, B.-H. Lee, S. Lee, H.-J. Kim, J. Dorgan, Enhanced interfacial adhesion, mechanical, and thermal properties of natural flour-filled biodegradable polymer biocomposites, Journal of Thermal Analysis and Calorimetry, 2011, 104 (1): [53] S. Ok Han, S. Muk Lee, W. Ho Park, D. Cho, Mechanical and thermal properties of waste silk fiber-reinforced poly(butylene succinate) biocomposites, Journal of Applied Polymer Science, 2006, 100 (6):

230 Chapter 6: Mechanical Performances of Biocomposites Made From Miscanthus Fibers and Poly(butylene adipate-co-terephthalate) Matrix Abstract Miscanthus fibers reinforced biodegradable poly(butylene adipate-co-terephthalate) (PBAT) based biocomposite was successfully produced by traditional melt processing methods. The material properties of the produced PBAT/miscanthus composites were evaluated by means of mechanical, thermal and morphological properties. Compared to neat PBAT, the flexural properties, and tensile modulus were increased after the incorporation of miscanthus fibers into the PBAT matrix. These improvements were attributed to the strong reinforcing effect of miscanthus fibers. Relatively high hydrophilic nature of the miscanthus fibers and the relatively high hydrophobic character of PBAT matrix lead to weak interfacial interaction between the components in the resulting PBAT/miscanthus composites. This weak interface was evidenced in the impact and tensile strength of the uncompatibilized PBAT composite. The interfacial bonding between the miscanthus fibers and PBAT was modified with 5 wt% of maleic anhydride grafted PBAT (MAH-g-PBAT) as a reactive compatibilizer. In the compatibilized PBAT composites, the improved interfacial interaction between the PBAT and the miscanthus fiber was corroborated with mechanical, thermal and morphological properties. The compatibilized PBAT composite with 40 wt% miscanthus fibers exhibited an average heat deflection temperature of 81 o C, notched Izod impact strength of 184 J/m, tensile strength of 19.4 MPa, and flexural strength of 22 MPa. From the scanning electron microscopy analysis, a better interfacial interaction between the components can be observed in the compaitibilized PBAT composites, which contribute to enhanced mechanical properties. 196

231 6.1 Introduction Natural fiber composites (biocomposites) are an attractive material for many applications because natural fibers are renewable, sustainable, low cost, lightweight, environmental friendly nature and easy processing. Recently, biodegradable polymer matrix based biocomposites are widely studied because of some environmental impacts such as waste disposal of nonbiodegradable plastic based materials. Among the biodegradable polymers, poly(butylene adipate-co-terephthalate) (PBAT) is an aliphatic-aromatic copolyester, which is typically synthesized from fossil fuel based monomers. PBAT has properties similar to non-biodegradable polymers like polyethylene. However, the high cost, low heat resistance property, inferior stiffness and strength are major shortcomings of the PBAT to extend its applications. These shortcomings can be overcome by the addition of natural fibers/fillers into PBAT matrix. In order to reduce the cost of PBAT, Muniyasamy et al., [1] and Torres et al., [2] prepared biocomposite from PBAT and biofuel industry co-products. Due to the polarity difference between the natural fibers/fillers and polymer matrices, most of the biocomposites are not compatible between the phases. As a result, chemically modified (sulphuric acid hydrolysis, silanization and acetylation) curua fibers reinforced PBAT biocomposites were investigated by Marques et al., [3]. The authors suggest that the chemically modified curua fibers were enhanced the performance of the resulting biocomposites owing to improve the fiber-matrix interaction/adhesion. Similarly, PBAT/peanut husk composites were prepared with and without maleic anhydride grafted PBAT (MAH-g-PBAT) compatibilizer [4]. The compatibilized PBAT/peanut husk composites exhibited superior mechanical properties than that of uncompatibilized composites and neat PBAT. Furthermore, the degradability of PBAT, compatibilized PBAT/peanut husk composites and uncompatibilized PBAT/peanut husk composites were investigated under biological environments [4]. It has been found that the 197

232 compatibilized PBAT/peanut husk composites showed slightly lower degradation rate than corresponding uncompatibilized composites. Interestingly, the rate of PBAT biodegradation was enhanced after incorporation of natural fillers into PBAT [4]. This observation had good agreement with another study investigated by Muniyasamy et al., [1]. Perennial grasses are good candidate for polymer matrix based composite applications because of its good fiber properties [5] and reducing food Versus fuel competition. Very limited literature is available for perennial grass (switchgrass, miscanthus, indiangrass, napiergrass, Arundo donax L, big bluestem, little bluestem, et al.,,) reinforced polymer composites. Miscanthus (elephant grass) is a typical lignocellulosic C4 grass, which grows in North America, Europe and Asia. Miscanthus is mostly used for biofuel production and animal bedding uses. Furthermore, miscanthus fiber based composites are developed for packaging application under the name of Sunatura TM [6]. Only few researchers have been investigated the performance of miscanthus fiber reinforced biocomposites. Consequently, the present study was aimed to produce a biocomposite from PBAT and miscanthus fibers. In addition, this study was also employed to improve the compatibility between the PBAT matrix and the miscanthus fibers by a reactive compatibilization strategy. 6.2 Materials Commercially available Oyster white PBAT pellets (Biocosafe 2003) were purchased from Xinfu Pharmaceutical Co., Ltd, China. According to the manufacturer, the density and melting temperature of PBAT are 1.26 g/cm 3 and o C, respectively. New Energy Farms, Ontario, Canada, kindly provided miscanthus fibers with an average length of 4.65±2.5 mm. It was used as received without further purification and/or modification. In this work, a maleic anhydride grafted PBAT (MAH-g-PBAT) was used as a compatibilizer. A detailed MAH-g- 198

233 PBAT (MAH grafting level 1.9%) was explained in Chapter 4. Melt flow index (MFI) value of the MAH-g-PBAT was measured at 190 o C with 2.16 kg load according of ASTM D1238. The MFI value of the MAH-g-PBAT was 44.46±9.96 g/10 min. 6.3 Biocomposite fabrication method Both PBAT granules and miscanthus fibers were dried at 80 o C for 10 h before composites fabrication. The composite preparation was performed in a lab-scale extrusion and injection molding technique. The extrusion process was carried out in a co-rotating twin-screw extruder (DSM Xplore, The Netherlands) with a barrel volume of 15cc. In order to produce desired test specimens, the molten extrudate was immediately transferred into 12cc injection molding device (DSM Xplore, The Netherland). The extrusion processing was conducted at 140 o C for 2 min while screw speed kept constant at 100 rpm and injection mould temperature of 30 o C. All the samples including neat PBAT, compatibilized and uncompatibilized composites were prepared using above said processing conditions. In the present study, the compatibilizer (MAH-g-PBAT) concentration was fixed at 5 wt% for all the compatibilized composites preparation. The chosen compatiblizer concentration (5 wt%) was based on optimal performances of the composites in our previous work (Chapter 5). 6.4 Characterization methods Tensile, flexural and impact properties of the PBAT and its composites were measured according to the procedure reported in Chapter 5 (Section 5.2.4). The heat deflection temperature (HDT), melt flow index (MFI) and fracture surface morphology of the samples were analyzed based on the procedure reported in Chapter 5 (Section 5.2.6, and 5.2.9). 199

234 6.5 Results and Discussion Mechanical properties The tensile properties of the neat PBAT and its composites are presented in Figure 6.1. Neat PBAT had an average tensile strength of 20 MPa. There was a significant reduction in tensile strength of PBAT after incorporation of miscanthus fibers; this reduction, suggesting that the compatibility between the PBAT and miscanthus was not sufficient [4]. Similar trend has been noticed in the PBAT/peanut husk composites [4] and PBAT/sisal fiber composites [7]. In the presence of compatibilizer (MAH-g-PBAT), the PBAT/miscanthus fiber composites showed a substantial increase in tensile strength as compared to uncompatibilized PBAT/misccanthus composites. The observed increased in tensile strength for the compatibilized composite was attributed to the enhanced compatibility between the components. This observation agreed with other research works reported elsewhere for PBAT composites [4, 7]. The tensile modulus of the PBAT was increased with the addition of miscanthus fibers, which can be attributed to the strong reinforcing capability of miscanthus fibers in the PBAT matrix. The modulus of compatibilized and uncompatibilized composites was not different. A maximum tensile modulus (~700 MPa) was observed in the PBAT composite with 40 wt% fiber loading. The observed tensile strength and modulus values of PBAT/miscnthus fiber composites are comparable with low-density polyethylene (LDPE) [8]. 200

235 Figure 6.1. Tensile strength and tensile modulus of PBAT and its composites; (A) neat PBAT, (B) PBAT/miscanthus fibers (70/30 wt%), (C) PBAT/miscanthus fibers/mah-g-pbat (65/30/5 wt%), (D) PBAT/miscanthus fibers (60/40 wt%), and (E) PBAT/miscanthus fibers/mah-g- PBAT (65/30/5 wt%) The flexural properties of the PBAT and its composites are depicted in Figure 6.2. Both flexural modulus and flexural strength of PBAT were increased with increasing miscanthus fiber content up to 40 wt%. Zhang et al., [9] also observed such a type of flexural properties improvement in the miscanthus fiber reinforced toughened green composites. In another study by Kirwan et al., [10] found that the flexural properties of the composites with miscanthus were superior compared to their matrix of the composite. They concluded that the enhanced flexural properties were attributed to the strong reinforcing effect of miscanthus fibers in the matrix. In the present study, the flexural strength of the compatibilized composites was higher compared to their uncompatibilized counterparts and neat PBAT. The enhanced flexural strength of the compatibilized composite has same trend like tensile strength of compatibilized composites. 201

236 Figure 6.2. Flexural properties of PBAT and its compatibilized and uncompatibilized composites: (A) neat PBAT, (B) PBAT/miscanthus fibers (70/30 wt%), (C) PBAT/miscanthus fibers/mah-g-pbat (65/30/5 wt%), (D) PBAT/miscanthus fibers (60/40 wt%), and (E) PBAT/miscanthus fibers/mah-g-pbat (65/30/5 wt%). Notched Izod impact test result of the neat PBAT and its composites are demonstrated in Figure 6.3. Neat PBAT showed non-break impact strength while PBAT composites showed hinge break impact strength under selected test conditions. There was a significant reduction in impact strength after incorporation of 30 wt% miscanthus fibers into the PBAT matrix. This impact strength reduction was more pronounced in the PBAT composite with the addition of 40 wt% miscanthus fiber. A similar trend was observed in the composites produced from miscanthus fiber and toughen polymer matrix [9]. In this study, the compatibilized composites yielded better impact strength compared to their corresponding uncompatibilized composites. In particular, the impact strength of PBAT/miscanthus/MAH-g-PBAT (55/40/5 wt%) composites showed 76% improvement in comparison to uncompatibilized PBAT/miscanthus (60/40 wt%) composites. The observed effect of compatibilization upon the impact strength of resulting 202

237 PBAT/miscanthus composites is consistent with compatibilized miscanthus fiber reinforced toughen biocomposites impact strength [9]. Figure 6.3. Notched Izod impact strength of PBAT and its compatibilized and uncompatibilized composites: (A) neat PBAT, (B) PBAT/miscanthus fibers (70/30 wt%), (C) PBAT/miscanthus fibers/mah-g-pbat (65/30/5 wt%), (D) PBAT/miscanthus fibers (60/40 wt%), and (E) PBAT/miscanthus fibers/mah-g-pbat (65/30/5 wt%) Melt flow index and Heat deflection temperature Table 6.1 summarises the melt flow index (MFI) and heat deflection temperatures (HDT) of the neat PBAT, as well as uncompatibilized and compatibilized composites. MFI is one of the important properties for composite materials because it can determine the possible processing methods for different application. The MFI of the PBAT composites was lower than that of neat PBAT. The observed MFI value reduction of the PBAT composites was attributed to the restriction of polymer chain mobility in the presence of fibers. Due to the high MFI value of the MAH-g-PBAT compatibilizer (44.46±9.96 g/10 min), the compatibilized PBAT/miscanthus composites showed slightly high MFI value compared to their corresponding uncompatibilized composites. The observed MFI values of the both compatibilized and uncompatibilized PBAT composites are not high enough for mass production of household injection moulding articles 203

238 [12]. The low MFI value of PBAT composites could be modified by the addition of flow additives. Table 6.1 depicts the HDT values of neat PBAT and its composites with different concentration of miscanthus fibers. It can be seen that the HDT value of PBAT was around 46 o C. The HDT value of PBAT composites was considerably higher compared to that of neat PBAT. This increased HDT value of PBAT composites was attributed to the increased stiffness of the resulting composites. The HDT value of the composite was more pronounced in the presence of compatibilizer. This may be due to the enhanced compatibility between components as discussed earlier. Such HDT improvement has been observed in the compatibilized perennial grass reinforced biocomposites [9, 11]. Table 6.1. Melt flow index (MFI) and heat deflection temperature (HDT) measurement Samples MFI (g/ o C with 2.16 kg HDT ( o C) Neat PBAT 9.4± ± 1.5 PBAT/miscanthus (70/30 wt%) 1.28± ±2.05 PBAT/miscanthus/MAH-g-PBAT 1.47± ±2.75 (65/30/5 wt%) PBAT/miscanthus (60/40 wt%) 1.19± ±1.11 PBAT/miscanthus/MAH-g-PBAT (55/40/5 wt%) 1.59± ± Scanning electron microscopy Figure 6.4 shows the SEM micrographs of tensile fractured compatibilized and uncompatibilized PBAT composites. In the uncompatibilized composites (Figure 6.4 A), there is a clear evidence for lack of adhesion between the fibers and matrix phase and fiber pullout. Furthermore, the fibers were not completely coated with PBAT matrix in the resulting uncompatibilized composite. This suggests that the affinity between the matrix and fibers is very 204

239 poor. The observed insufficient interaction between the components is responsible for the inferior mechanical performance of uncompatibilized PBAT composites. On the other hand, the compatibilized PBAT/miscanthus fiber composites morphology (Figure 6.4 B) showed better interfacial adhesion and the miscanthus fibers were well embedded with PBAT matrix. Unlike uncompatibilized composites, most of the fibers were covered by PBAT matrix in the composites with compatibilizer, indicating that the compatibilizer (MAH-g-PBAT) played a crucial role to enhance the interfacial bonding between the phases. Consequently, the compatibilized composites can lead to stronger stress transfer between the phase compared the PBAT composites without compatibilizer. The enhanced compatibility between the PBAT and miscanthus fibers was consistant with its mecahcnial properties. Figure 6.4. SEM micrographs of uncompatibilized PBAT/misanthus fibers (60/40 wt%) composites (A) and compatibilized PBAT/misanthus fibers/mah-g-pbat (55/40/5 wt%) composites (B) 6.6 Conclusions The effect of miscanthus fibers on the performance of PBAT/miscanthus fiber composite was examined. Due to strong reinforcing effect of miscanthus fibers in the PBAT matrix, the stiffness of the resulting PBAT/miscanthus composites is higher than neat PBAT. At the same 205

240 time, the tensile strength and impact strength had negative effect on the addition of miscanthus fibers into PBAT matrix. This detrimental effect of the uncompatibilized composites tensile strength and impact strength was considerably improved by compatibilization strategy. The compatibilized PBAT/miscanthus composites exhibited superior performances more than that of uncompatibilized composites. The enhanced compatibility was confirmed by microscopic analysis. Overall, the compatibilized PBAT composite with 40 wt% miscanthus fibers showed a maxima improvement in mechanical properties. As a result, the prepared sustainable PBAT/miscanthus composite is possible alternative for non-biodegradable composite materials. References [1] S. Muniyasamy, M. M. Reddy, M. Misra, A. Mohanty, Biodegradable green composites from bioethanol co-product and poly(butylene adipate-co-terephthalate), Industrial Crops and Products, 2013, 43 (0): [2] S. Torres, R. Navia, R. Campbell Murdy, P. Cooke, M. Misra, A. K. Mohanty, Green Composites from Residual Microalgae Biomass and Poly(butylene adipate-coterephthalate): Processing and Plasticization, ACS Sustainable Chemistry and Engineering, 2015, 3 (4): [3] M. V. Marques, J. Lunz, V. Aguiar, I. Grafova, M. Kemell, F. Visentin, A. Sartori, A. Grafov, Thermal and Mechanical Properties of Sustainable Composites Reinforced with Natural Fibers, Journal of Polymers and the Environment, 2014, [4] C.-S. Wu, Utilization of peanut husks as a filler in aliphatic aromatic polyesters: Preparation, characterization, and biodegradability, Polymer Degradation and Stability, 2012, 97 (11):

241 [5] W. Liu, A. K. Mohanty, P. Askeland, L. T. Drzal, M. Misra, Influence of fiber surface treatment on properties of Indian grass fiber reinforced soy protein based biocomposites, Polymer, 2004, 45 (22): [6] (accessed on November, 2015) [7] C.-S. Wu, Process, Characterization and Biodegradability of Aliphatic Aromatic Polyester/Sisal Fiber Composites, Journal of Polymers and the Environment, 2011, 19 (3): [8] J. Morawiec, A. Pawlak, M. Slouf, A. Galeski, E. Piorkowska, N. Krasnikowa, Preparation and properties of compatibilized LDPE/organo-modified montmorillonite nanocomposites, European Polymer Journal, 2005, 41 (5): [9] K. Zhang, M. Misra, A. K. Mohanty, Toughened Sustainable Green Composites from Poly(3-hydroxybutyrate-co-3-hydroxyvalerate) Based Ternary Blends and Miscanthus Biofiber, ACS Sustainable Chemistry and Engineering, 2014, 2 (10): [10] K. Kirwan, R. M. Johnson, D. K. Jacobs, G. F. Smith, L. Shepherd, N. Tucker, Enhancing properties of dissolution compounded Miscanthus giganteus reinforced polymer composite systems: Part 1. Improving flexural rigidity, Industrial Crops and Products, 2007, 26 (1): [11] V. Nagarajan, M. Misra, A. K. Mohanty, New engineered biocomposites from poly(3- hydroxybutyrate-co-3-hydroxyvalerate) (PHBV)/poly(butylene adipate-co-terephthalate) (PBAT) blends and switchgrass: Fabrication and performance evaluation, Industrial Crops and Products, 2013, 42 (0): [12] A. V. Shenoy, D. R. Saini, Melt flow index: More than just a quality control rheological parameter. Part I, Advances in Polymer Technology, 1986, 6 (1):

242 Chapter 7: Biocomposites Consisting of Miscanthus Fibres in a Biodegradable Binary Blend Matrix: Preparation and Performance Evaluation* *A part of this chapter has been filed US provisional patent application: A. K. Mohanty, M. Misra, N. Zarrinbakhsh, R. Muthuraj, T. Wang, A. U-Rodriguez, and S.Vivekanandhan, Biodegradable polymer-based composites with tailored properties and method of making those, US provisional patent application, Application number , Abstract Biocomposites were fabricated from miscanthus fibers and a blend composed of poly(butylene succinate) (PBS)/poly(butylene adipate-co-terephthalate) (PBAT) matrix by extrusion and injection molding process. The performance of the composites was investigated with different weight percentages of fiber loading. Due to the strong reinforcing ability of the miscanthus fibers, the elastic modulus increased dramatically to 2.1 GPa when fiber loading increased to 50 wt%. The elastic modulus of the composites was evaluated by parallel, series, Hrisch and Halpin-Tsai theoretical models and the values were compared to experimental values. Addition of miscanthus fibers into PBS/PBAT blend; it sharply reduced both tensile and impact strength. These reductions were attributed to the incompatibility between the miscanthus fibers and PBS/PBAT blend resin. As a result, maleic anhydride functionalized PBS/PBAT blend was prepared and used as a compatibilizer to improve interfacial interaction between the different phases in the resulting composites. With the addition of 5 wt% compatibilizer to the composites a significant improvement in the mechanical properties as compared to corresponding uncompatibilized counterparts was found. SEM analysis demonstrated strong interface between fiber and matrix in the compatibilized composites. The aspect ratio of the fibers was drastically reduced after compounding because of fibers underwent attrition during processing. The shear thinning behavior was found to increase with increasing fiber content. It could be due to the 208

243 reduced entanglement of the polymer chains in the composites. The density of the resulting biocomposites was lower than glass fibers. Over all, the addition of miscanthus fibers into a PBS/PBAT blend matrix to form composites can offer a significant benefit in terms of economic competitiveness and functional performance. 7.1 Introduction Due to global environmental concerns, recent academic research and literature in the field of polymer has been focused on research into biobased and/or biodegradable polymeric materials. The US Environmental Protection Agency has reported that most non-durable plastic waste goes to landfills because recycling of these plastics is still a challenge. Because of this, material scientists are interested as this can be reduced through the use of biodegradable polymeric materials. Biodegradable polymers are easily degraded in the presence of naturally occurring microbes to CO 2 and H 2 O under aerobic conditions or CH 4 and H 2 O under anaerobic conditions [1]. Commercially available aliphatic and aliphatic-aromatic biodegradable polyesters such as poly(butylene adipate-co-terephthalate) (PBAT) and poly(butylene succinate) (PBS) are promising for use in natural fiber composites. Both PBS and PBAT are produced from fossil fuel based monomers. PBS can be produced from renewable resource based succinic acid with ~54% biobased content [2], however, these polymers are currently have poor cost-performance as compared to traditional polymers. This poor cost performance is due to limited production of these polymers [3]. Interest into this kind of polymers has increased recently because of the reduced carbon footprint and greenhouse gas emissions associated with their use. Despite this interest and the potential environmental benefits, these polymers cannot yet be utilized on their own for some commercial applications because they do not satisfy many product requirements. These 209

244 shortcomings can be overcome by reinforcing some inexpensive natural fibers into polymer matrices [4]. As such, in 2010 natural fiber composites production amounted to million pounds in the United States with an associated worth amounting to million USD [5]. As a commodity this market is estimated to reach million US$ by 2016, with a short-term growth rate of 11% in the next 5 years. Effective use of renewable resource derived fibers provides environmental benefits with respect to ultimate disposability. Natural fibers have many economic advantages compared to synthetic fibers in addition to being biodegradable, renewable, and lightweight. Reduce reliance on petroleum oil, reduced tool wear, good specific strength and minimized hazardous materials emission (noxious gases or solid residue) during combustion are also beneficial [6, 7]. Miscanthus is a typical lignocellulosic biomass and is a promising fast growing non-food crop. If this biomass could be successfully reinforced into a polymer matrix, it could increase revenue for miscanthus growers. Miscanthus fibers are advantageous because it is low cost, high yield, has low input conditions and a low maturation time, has potential for soil remediation, can help balance carbon dioxide in the environment, and is able to sequester carbon underground. Due to these benefits, the best strategy is to combine biodegradable polymers with miscanthus fibers in order to create cheap sustainable biocomposites with good reinforcement properties. Bourmaud and Pimbert [7] measured the modulus and hardness value of miscanthus fibers using the nanoindentation method. They found average modulus and hardness values are 9.49 and 0.34 GPa, respectively. These properties are quite comparable with other agro fibers such as hemp and sisal [7]. Green composite materials can be derived from natural/bio fibers and biodegradable polymers and have been developed for many industrial applications [8]. The performance of the 210

245 resulting composite mainly depends on orientation in the matrix, volume fraction of fibers, fiber distribution, fiber-matrix interaction and ability of stress transfer between the components [8]. Natural fibers are inherently hydrophilic in nature [9] and are as such incompatible the hydrophobic polymer matrix. This leads to inferior mechanical performance of the resulting biocomposites. Additionally, during processing, natural fibers have a tendency to disperse poorly in the matrices due to agglomeration [7]. Therefore, various strategies have been developed in order to overcome these drawbacks in biocomposites [10]. Some of these strategies include using a compatibilizing agent/interfacial modifier to improve interfacial bonding between the phases in the composites [11-15]. Keener et al., [16] used a commercially available and economically produced maleic anhydride (MAH) grafted polymer as a compatibilizing agent for fabricating biocomposites. This study has proven that the relatively polar nature of maleic anhydride groups may have a tendency to form covalent bonds, secondary bonds and mechanical interlocking in the biocomposites. Another study reported by Tserki et al., [11] showed the effect of cotton fibers on the mechanical properties of poly(butylene succinate-co-butylene adipate) (PBSA) with and without MAH grafted PBSA (MAH-g-PBSA) as a compatibilizing agent. This study showed significant improvement in all mechanical properties as compared to uncompatibilized composites. A similar trend has been reported in the poly(3-hydroxy-co-3-hydroxyvalerate) (PHBV)/kenaf fiber biocomposites [13]. Kim et al., [12] used MAH grafted polymers as a compatibilizer for bamboo and wood flour filled PBS and poly(lactic acid), PLA biocomposites. In these composites, MAH grafted PBS (MAH-g-PBS) and MAH grafted PLA (MAH-g-PLA) showed improvements in mechanical and thermal properties when compared to other maleated compatibilizers. However, only a few researchers have investigated miscanthus fiber reinforced biodegradable polymer composites. Johnson et al., [17] have studied Mater-Bi /miscanthus fiber 211

246 composites and found that the impact performances were significantly increased as compared to neat polymer. Nanda et al., [18] examined a composite made from agro fibers (miscanthus, switchgrass and soy hull) and a binary blend matrix of PLA/PHBV. The authors concluded that the miscanthus fiber based biocomposites have superior mechanical and thermo-mechanical properties compared to many other agro fibers (switchgrass and soy hull) based PLA/PHBV blend matrix composites. However, un-to-date no literature has been reported on fabrication of composites from miscanthus fiber with a PBS/PBAT blend matrix. The present study aims to explore the fabrication and performance evaluation of miscanthus fiber reinforced PBS/PBAT based biocomposites. The effect of compatibilizer (MAH-grafted-PBS/PBAT blend) on the resulting biocomposites was investigated by means of mechanical, thermo-mechanical, morphological, and rheological properties. 7.2 Materials and Methodology Materials Miscanthus fibers were kindly supplied by New Energy Farms, Ontario, Canada. The miscanthus fibers were used as received without any further purification and/or modification. The predetermined fiber lengths and diameters were 4.65±2.5 and 0.074±0.024 mm, respectively. Commercially available PBAT (Biocosafe 2003F) and PBS (Biocosafe 1903) granules were procured from Xinfu Pharmaceutical Co., Ltd (China). Maleic anhydride grafted PBS/PBAT blend (MAH-g-PBS/PBAT) was prepared in Haake PolyLab at 160 o C with 5 phr of MAH and 1 phr of dicumyl peroxide (DCP) according to the procedure reported in Chapter 4 and it was used as a compatibilizer in this chapter. The grafting percentage of the compatibilizer was determined by acid-base titration and the grafting percentage was 2.05%. The melt flow index 212

247 (MFI) of the compatibilizer was 81±24 g/10 min (MFI 190 o C, 2.16 kg according to ASTM D1238) Processing of polymer blend and their composites Before melt processing, both polymers and fibers were dried at 80 o C for at least 10 h in an oven. According to our previous research findings, PBS/PBAT (60/40 wt%) blend was taken as a matrix for composites fabrication in the present study [20]. Here after PBS/PBAT (60/40 wt%) blend will be referred as PBS/PBAT blend. This blend matrix based composites were prepared by incorporating 30, 40 and 50 wt% of miscanthus fibers. For comparison, these composites were also prepared with the addition of 5 wt% MAH-g-PBS/PBAT. All the compounding and injection molding were carried out in a DSM Xplore, The Netherlands. The capacity of DSM extruder and injection molder was 15 and 12 cm 3, respectively. Compounding of the formulation was done with co-rotating twin-screw extruder with a length of 150 mm and a L/D of 18. The desired composite test specimens were produced at 140 o C with residence time of 2 min, mould temperature at 30 o C, and screw speed of 100 rpm. Injection molding was done with injection pressure of 10 bar for 8 s, and holding and packing pressures of 10 bar for 10 s each Mechanical properties The tensile, flexural and impact properties of the PBS/PBAT blend and its composites were measured according to the procedure reported in Chapter 5 (Section 5.2.4) Density An electronic densimeter (Alfa Mirage, model MD-300S) was used to measure the density of the PBS/PBAT blend and their composites. The density measurement was performed 213

248 by Archimedes principle. The density of the miscanthus fiber was calculated by rule of mixture, i.e., = + (7.1) where, and, are the densities and volume fractions of the fiber and matrix, respectively Dynamic mechanical analysis (DMA) Dynamic mechanical properties were measured according to ASTM D4065 standard in a DMA Q800, TA instruments Inc, USA. The storage modulus and tan delta results were obtained as a function of temperature with a temperature ramp of 3 o C/min. The tests were carried out in a dual cantilever clam with a 15 µm oscillating amplitude and 1 Hz vibrating frequency. Heat deflection temperatures (HDT) of the samples were measured using same DMA machine as per ASTM D648 standard. The experiments were carried out in a three point bending clamp at a heating ramp of 2 o C/min from room temperature to desire temperature. The reported HDT values of the samples are an average of two measurements Fiber length measurements In order to measure the fiber length and diameter after processing, the composite samples were dissolved in chloroform and then fibers were isolated by filtering. The isolated fibers were rinsed thoroughly with the same solvent and dried at 70 o C for 24 h. The processed and unprocessed fibers were photographed through a digital camera and the fiber dimensions were measured by Image J software (at least 85 individual fibers were measured). The measured fibers length and diameter were inserted in Minitab 17 statistical software to get fiber distribution histogram. 214

249 7.2.7 Differential scanning calorimetry (DSC) DSC analyses were performed under controlled nitrogen environment (50 ml/min) by using TA instruments DSC Q200. The data were recorded for heat flow with respect to temperatures using standard aluminum pan. The heating ramp rate was kept at 10 C/min while cooling ramp rate was kept at 5 C/min for conventional DSC. The experiment was carried out from -50 to 180 o C, followed by cooling to -50 o C. Then they were reheated from -50 to 180 o C. The first heating scans were used to remove thermal history of the sample. The results were collected from the second heating and first cooling scans Thermogravimetric analysis (TGA) The thermal degradation behavior of the samples was evaluated using TA instruments TGA Q500. The experiments were performed in a controlled nitrogen flow rate of 60 ml/min with a heating rate of 20 o C/min Morphological analysis In order to enhance the electrical conductivity of the specimen surface, the samples were gold coated prior to observe the composites samples morphology. Morphology of the impact fractured specimens was observed by Inspect S50-FEI Company scanning electron microscopy Rheological property The rheological property was measured at 140 o C using Anton Parr Rheometer MCR 301. The experiments were carried out in a parallel-plate with 25 mm diameter and 1 mm gap between the plates. The shear viscosity of the samples was measured from 629 to 0.1 rad/s by frequency sweep test. Disc shape injection molded samples were used to study the rheological behaviour of PBS/PBAT blend and their composites. 215

250 7.3 Results and Discussion Mechanical properties The tensile properties of PBS/PBAT blend and their composites are presented in Figure 7.1. Mechanical properties of the natural fiber reinforced composites are significantly dependent upon many factors including fiber type, aspect ratio of fibers, orientation of fibers, fiber-matrix adhesion, fiber dispersion in the matrix, and chemical composition of the fibers [21]. Apparently, all the composites showed higher tensile modulus value than the PBS/PBAT blend matrix. The tensile modulus of the composites gradually increased with increasing fiber loading from 30 to 50 wt%. Generally the addition of stiff fibers into polymer matrices restricts polymer chain mobility and thus increases the stiffness of the resulting composites [22]. However, a reduction in tensile strength was observed with an increase in fiber content up to 50 wt%. Incorporation of hygroscopic natural fibers into the hydrophobic polymer matrix has a negative effect on tensile strength of the composite. A number of researchers have reported a similar effect with natural fiber composites [18, 22]. This reduction is associated with uncompatibility between the fibermatrix as well as lack of sufficient fiber dispersion in the matrix [23]. These drawbacks can be overcome by modifying composite matrices or by reinforcement. In general, commercially available MAH grafted polyolefins are more widely used in composite industries as a compatibilizer [14]. Many studies have documented that the compatibilizers could connect the fibers and the matrix through chemical bonds like covalent bonds or hydrogen bonds [15]. Therefore, MAH grafted compatibilizers have great potential to surpass the performance of composites through strong interfacial adhesion between the different phases. In this study, MAH grafted PBS/PBAT blend was used as a compatibilizer in order to improve compatibility between the PBS/PBAT blend matrix and the miscanthus fibers. One of 216

251 the main advantages of modifying PBS/PBAT matrix as a compatibilizer for PBS/PBAT blend based composites is that the chemical structures are identical and, consequently, compatible. All the compatibilized composites showed a distinct enhancement in tensile properties in comparison to their corresponding uncompatibilized composites. These improvements were likely due to the enhanced interfacial bonding between the fiber-matrix. For instance, incorporation of compatibilizer (5 wt%) into 50 wt% miscanthus fiber composite showed 69% tensile strength and 7% tensile modulus improvement as compared to their corresponding uncompatibilized composites. Similar improvements in the compatibilized biocomposites have been shown in the literature [23,24]. The improved tensile strength of the compatibilized composites is likely due to the physical entanglements, which can occur between the compatibilizer and matrix. Due to the entanglement of polymer chains, the applied stress on the matrix can be effectively transformed to the fibers [25]. Figure 7.1. Tensile properties of PBS/PBAT blend and their composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (65/30/5 wt%), (D) PBS/PBAT + miscanthus fibers (60/40 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), (F) PBS/PBAT + miscanthus fibers (50/50 wt%) and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%). 217

252 Figure 7.2 shows the typical stress-strain curves of the PBS/PBAT blend and their composites. It was noted that the stress-strain curve of the PBS/PBAT blend clearly shows three types of behavior i.e., elastic deformation, plastic deformation and strain hardening. However, the stress-strain curves of all the composites samples shows elastic and plastic deformation after maximum stress has reached. There was no strain hardening observed in the composites samples. These observations suggest that composites are less ductile than the PBS/PBAT blend. This is not surprising because the short fibers reinforced composite does not necessarily increase the ductility of the composites. Sahoo et al., [26] found a similar trend in PBS based biocomposites. This ductility reduction is also evidence for the improved stiffness of the composites when compared to PBS/PBAT blend. A drastic reduction was observed in elongation at break after incorporation of miscanthus fibers into the PBS/PBAT blend matrix. This is common observation for almost all the fiber reinforced composites [11]. All the compatibilized composites exhibited slightly higher elongation at break compared to their corresponding uncompatibilized composites. This implies that the interfacial adhesion was improved between the phases [27]. Our findings agreed well with the previous works by Wang et al., [27] and Zhang et al., [28]. However, the percentage elongation values of their compatibilized composites were much lower than the PBS/PBAT blend. 218

253 Figure 7.2. Representative stress-strain curves of PBS/PBAT blend and their composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (45/50/5 wt%). Figure 7.3 illustrates the influence the flexural properties of PBS/PBAT blend matrix based composites with and without compatibilizer. The flexural modulus and strength of the PBS/PBAT blend matrix were 380 and 17 MPa, respectively. Similarly, to the tensile modulus, the addition of miscanthus fibers into PBS/PBAT blends lead to a remarkable improvement in the flexural modulus of the composites. Similar occurrence has been reported in flexural modulus with increasing filler content in different thermoplastic matrix systems [29, 30]. Moreover, the addition of miscanthus fibers into poly(vinyl acetate) (PVAc)/poly(vinyl alcohol) (PVA) blend resulted in superior flexural properties than PVAc/PVA blend [31]. Normally the polymer chain mobility is hindered by presence of fibers, consequently the composites become stiffer and have a higher modulus as compared to their matrix [15]. The compatibilized 219

254 composites showed slight improvement in flexural modulus compared with their corresponding uncompatibilized composites. This indicates that the compatibilizer may reduce the fiber agglomeration in the resulting biocomposites. Additionally, the flexural strength of the uncompatiblized composites reduced after incorporation of 50 wt% fibers. This could be associated with weak fiber-matrix interaction, insufficient wetting between the fiber-matrix, and agglomeration (fiber-fiber interaction) of fibers in the matrix [29]. Interestingly, the flexural strength of the uncompatibilized composites is still superior in comparison to PBS/PBAT blend matrix. This phenomenon once again proves that the miscanthus fibers have good reinforcing capabilities in the PBS/PBAT blend matrix. Overall, flexural properties of the compatibilized composites were superior to their uncompatibilized counterparts. Similar trends have been reported elsewhere [15]. The compatibilized composite with 50 wt% miscanthus fiber exhibited the highest flexural strength (192%) and modulus (520%) compared with PBS/PBAT matrix. The enhanced fiber-matrix adhesion can be hypothesized as follows: the grafted maleic anhydride groups of MAH-g-PBSPBAT are able to interact with surface hydroxyl groups of the miscanthus fibers, while the PBS/PBAT segments are miscible with the bulk PBS/PBAT blend matrix phase through cocrystallization [32]. The expected reaction mechanism of MAH-g- PBSPBAT with miscanthus fibers is shown in Figure

255 Figure 7.3. Flexural properties of PBS/PBAT blend and their composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (65/30/5 wt%), (D) PBS/PBAT + miscanthus fibers (60/40 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), (F) PBS/PBAT + miscanthus fibers (50/50 wt%) and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%). Figure 7.4. Expected reaction between the miscanthus fiber and the compatibilizer 221

256 The results in Table 7.1 represent the Izod impact strength of uncompatibilized and compatibilized composites as a function of miscanthus fibers. The PBS/PBAT blend showed non-break impact strength. Contarory, with the addition of miscanthus fibers into PBS/PBAT blend matrix resulted in a significant reduction in impact strength. This is attributed to the failure mode of PBS/PBAT blend changed from ductile to brittle failure in the presence of fibers. As explained before, the lignocellulosic fibers are restricting the polymer chain mobility in the resulting composites, and thus reduce capability to absorb energy during impact fracture [33]. In addition, the weak interface between the components is also important factor for deteriorated impact strength of the composites. This can also be confirmed by area reduction under the tensile stress-strain curves. However, a considerable amount of impact strength was increased after incorporation of compatibilizer into the composites. The compatibilized composites with 30, 40 and 50 wt% fibers showed 59, 62, and 36% improvement in impact strength when compared to their corresponding uncompatibilized composites, respectively. These improvements could be due to enhanced interfacial bonding between the matrix-fiber. Good interfacial adhesion helps to promote stress transfer between the phase through covalent and/or hydrogen bonds [34]. Furthermore, there is a possibility of mechanical interlocking in the compatibilized composites, which may occur between the compatibilizer-fiber and/or between the compatibilizer-matrix. It can be noted that the compatibilized composite with 50 wt% fiber showed lower impact strength compared to compatibilized composites with fiber content of 30 and 40 wt%. This could be due to the high fiber content lead to agglomeration and thus weaken the resulting composites [18]. Overall, the observed mechanical properties of the compatibilized PBS/PBAT matrix based composites suggest that the compatibility between the fiber-matrix has been greatly improved. 222

257 Table 7.1. Notched Izod impact strength of PBS/PBAT blend and its compatibilized and uncompatibilized composites Samples Notched Izod Impact strength (J/m) PBS/PBAT Non-break PBS/PBAT + miscanthus fibers (70/30 wt%) ± 3.59 PBS/PBAT + miscanthus fibers (60/40 wt%) ± 2.94 PBS/PBAT + miscanthus fibers (50/50 wt%) ± 1.64 PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT ± 4.36 (65/30/5 wt%) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT ± 2.59 (55/40/5 wt%) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT ± 3.12 (45/50/5 wt%) Theoretical approximation of Young s modulus of the PBS/PBAT biocomposites There are several mathematical models that have been proposed to predict composite properties [35]. Some models such as parallel, series, Hirsch s, and Halpin-Tsai models are very often used to determine the randomly oriented rigid short fiber composite behavior. In parallel and series models, Young s modulus can be calculated by using following equations, Parallel model M c = + (7.2) Series model M c = (7.3) In both models, M c, M m and M f represent the Young s modulus of the composites, matrix, and fibers while V f and V m represent the volume fractions of fiber and matrix, respectively. 223

258 Hirsch s model This model is combination of both parallel and series models. In a fiber composite with random fiber orientation, the elastic modulus can be predicted by Hirsch s model. The model is shown below, M c = x ( + ) + (1-x) (7.4) where χ is a value (0 to 1), which describes the stress transfer between the fiber and matrix. Halpin-Tsai and Tsai Pagano model Theoretical moduli of the aligned discontinuous fiber composites can be determined according to the Halpin-Tsai (H-T) equation. According to the H-T model, the longitudinal (E L ) and transverse modulus (E T ) can be calculated using the following equations (7.5 and 7.6): E L = E m (7.5) E T = E m (7.6) where = (7.7) = (7.8) where is the measure of reinforcement geometry and it can be defined as: = 2( (7.9) 224

259 Aspect ratio of the reinforcement is l/d. In a fiber composite with random (E random ) fiber orientation, the elastic modulus can be determined using the Tsai Pagano model and it is written as: E random = (7.10) Figure 7.5 compares the theoretical and experimental Young s modulus of the composites with different volume fraction of fibers. Both the theoretical and experimental Young s moduli of the composites were increased with increasing fiber content from 30 to 50 wt%. However, the predicated modulus by parallel and series models are significantly deviated from experimental modulus values. These deviations could be due to the model operating under the assumption that there is no interaction between the fiber-matrix in the composites. According to this literature [36], there is possibility for interaction between the composite components. It was observed that the experimental moduli of the composites had good agreement with theoretically calculated moduli by the Hirsch model. In the Hirsch model, the parameter x determines the stress transfer between the fiber and the matrix. In order to determine a linear-fit value with experimental values, the x values varied from 0 to 1 in the equation 7.4. A best fit was found between theoretical and experimental values when the x value 0.33 in equation 7.4. The obtained x value is slightly higher than the carbon fiber reinforced PLA and PHBV composites [37,38]. This suggests that the PBS/PBAT/miscanthus composites have a more effective stress transfer than carbon fiber reinforced PLA and PHBV composites. It was found that the experimental and predicted modulus values of the composites are in good agreement with 30 wt% fiber loading. When increasing fiber content from 30 to 50 wt% the modulus values are slightly deviating from the experimental values. This deviation probably attributed to the agglomeration of fibers in the matrix and thereby the applied stress is not able to transfer uniformly between aggregated fibers 225

260 to dispersed fibers [35]. It is clear that the predicated moduli using the H-T model are slightly deviated from the experimental modulus values. This behavior could be attributed to the modulus difference in the nodes, internodes, stem and leaves of the miscanthus fiber [7]. A similar trend has been observed in the switchgrass reinforced PBAT/PHBV blend matrix composites [39]. Figure 7.5. Variation of experimental and theoretical values of Young s modulus as a function of fiber loading Dynamic mechanical properties As shown in Figure 7.6, the storage modulus for PBS/PBAT blend and their composites varies with respect to temperature. The storage modulus (E ) values of PBS/PBAT blend and their composites were higher below the glass transition temperature (T g ) which is -19 o C. However, the E values of all the samples observed an abrupt decrease at the T g of PBS/PBAT blend. This is due to the fact that the polymer chain mobility increases above the T g of PBS/PBAT blend. It was found that the storage modulus values of compatibilized and uncompatibilized composites were significantly higher as compared to the PBS/PBAT blends. 226

261 Therefore, it was concluded that the miscanthus fibers had a strong influence on the storage modulus improvement of the resulting composites. The improved storage modulus was in accordance with the flexural and tensile moduli values of the PBS/PBAT blend composites. Nanda et al., [18] reported that the stiffness of the miscanthus based PHBV/PLA composites was higher than PHBV/PLA matrix due to reinforcing effect of miscanthus fibers. The E values of the compatibilized composites were slightly higher as compared to the corresponding uncompatibilized composites. A similar observation has been reported in the compatibilized PLA and PBS composites [12]. Figure 7.6. Storage moduli of PBS/PBAT blend and their composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT+ miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%). The molecular transitions and energy dissipation of polymeric materials can be determined by damping factor peaks (tan δ). Figure 7.7 shows the tan δ peaks of PBS/PBAT blend and their composites. There is no significant shift observed in tan δ peak after the addition 227

262 of miscanthus fiber into PBS/PBAT blend. This observation is consistent with elsewhere [40]. Incorporation of fibers into PBS/PBAT blends reduces the tan δ peak intensity, which is attributed with the restriction of the polymer chains mobility in the presence of fibers. In the neat polymer systems, the polymer chain segments are free from restraints, thus have high intensity tan δ peaks when compared to the composites. In the present study, compatibilized composites had a slightly higher damping effect than uncompatibilized composites. This indicates that the fiber-matrix adhesion was improved with the addition of compatibilizer. The molecular motion at the interfacial regions contributes to the damping of the materials [31]. As seen in Figure 7.7, the broadness of the tan δ peak is increased with increasing fiber from 30 to 50 wt%. This is due to the increased heterogeneity in the composite system, which was supported by SEM analysis. Figure 7.7. Tan δ of PBS/PBAT blend and their composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT+ miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%). 228

263 7.3.4 Density It is well known that the density of natural fibers is lower than glass fibers. The low density of natural fibers enables a reduction in weight of the composite material for many applications including automotive industries. Density of the PBS/PBAT blend and their composites is shown in Table 7.2. The densities of the miscanthus fibers reinforced composites are above 1.3 g/cm 3, which is comparatively higher than PBS/PBAT blend. Moreover, the composites density was gradually increased with increasing fiber content from 30 to 50 wt%. This is mainly due to the higher density of the miscanthus fiber (1.412 g/cm 3 ) when compared to neat PBS/PBAT blend (1.252 g/cm 3 ). The calculated density of miscanthus fiber is g/cm 3, which can be compared to some other natural fibers that are reported in the literature [41, 42]. There is no significant difference observed in the compatibilized and corresponding uncompatibilized composites densities. The observed densities of the both compatibilized and uncompatibilized PBS/PBAT blend composites are lower than glass fibers [43]. Table 7.2. Heat deflection temperature (HDT) and density of PBS/PBAT blend and its compatibilized and uncompatibilized composites Samples HDT ( o C) Density (g/cm 3 ) PBS/PBAT ± ± PBS/PBAT + miscanthus fibers (70/30 wt%) ± ± PBS/PBAT + miscanthus fibers (60/40 wt%) ± ± PBS/PBAT + miscanthus fibers (50/50 wt%) ± ± PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (65/30/5 wt%) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (55/40/5 wt%) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (45/50/5 wt%) ± ± ± ± ± ±

264 7.3.5 Heat deflection temperature According to ASTM D648, the temperature at which the test specimen deforms by 250 µm under a bending stress of MPa is called heat deflection temperature (HDT) of polymeric materials. HDT plays a vital role in selecting polymeric materials for specific applications because it represents the maximum working limit temperature of materials. Table 7.2 shows the HDT value of the PBS/PBAT blend and its composites. All the composites have higher HDT than PBS/PBAT blend. The HDT of the 30, 40 and 50 wt% fiber reinforced composites was around 100, 100 and 105 o C, respectively. According to Nanda et al., [18], miscanthus fiber reinforced PHBV/PLA composites have higher HDT values than switchgrass based PHBV/PLA composites. It has been suggested that this is due to the higher stiffness and reinforcing effect of miscanthus fibers. A marginal improvement was observed while increasing fiber content from 30 to 50 wt%. This could be because the HDT value of the composites is very close to the melting point (115 o C) of PBS/PBAT blend. The HDT values of the compatibilized composites were not significantly different as compared to their corresponding uncompatibilized composites. From these observations, we can conclude that the PBS/PBAT composite with 30 wt% miscanthus fiber reached an optimum HDT value Thermogravimetric analysis Figure 7.8 shows the thermal stability of miscanthus fibers, PBS/PBAT blend and their composites as a function of temperature. Table 7.3 summarizes the temperature at 5 (T 5 ), 25 (T 25 ), and 50 (T 50 ) percent weight loss of all the samples. It can be seen that the miscanthus fibers had a lower onset thermal stability (256 o C) as compared to PBS/PBAT blend and its composites. This could be attributed to the less thermally stable components such as pectin, proteins and residual moisture. Despite this temperature is being much higher than the processing 230

265 temperature (140 o C) of PBS/PBAT blend based composites. As a result, there is no possibility for thermal degradation of the miscanthus fibers in the prepared composites. A two-step thermal degradation was observed in all the composites and miscanthus fibers whereas PBS/PBAT had only one-step degradation. The two-step degradation of miscanthus is due to the degradation of cellulose (294 o C) and lignin (353 o C). Figure 7.8. Thermogravimetric traces for miscanthus, PBS/PBAT blend and its composites The thermal decomposition mechanism of the PBS/PBAT blend has been explained in our previous publication [20]. The first and second step degradation of the composites corresponds to the miscanthus and PBS/PBAT blend matrix, respectively. Apparently, the thermal stability of the composites monotonically decreased with increasing fiber loading up to 50 wt%. All the compatibilized composites showed slightly lower thermal stability compared to the corresponding uncompatibilized composites. This behavior may be due to the less thermal stable compatibilizer counterbalancing the PBS/PBAT blend content in the composites. Residue 231

266 remains even after heating to 500 o C under the nitrogenous atmosphere due to the carbonaceous residue and inorganic matter in the materials [44]. In miscanthus, about 20% char residue was obtained at 600 o C. Consequently, the char residues of the composites were increased with increasing fiber loading from 30 to 50 wt%. Table 7.3. Thermogravimetric data of miscanthus, PBS/PBAT blend and their composites Sample T 5 ( o C) T 25 ( o C) T 50 ( o C) Char residue at 600 o C (%) Miscanthus PBS/PBAT MAHgPBS/PBAT PBS/PBAT + miscanthus fiber (70/30 wt%) PBS/PBAT + miscanthus fiber (60/40 wt%) PBS/PBAT + miscanthus fiber (50/50 wt%) PBS/PBAT + miscanthus fibers + MAH-g PBS/PBAT (65/30/5 wt%) PBS/PBAT + miscanthus fibers + MAH-g PBS/PBAT (55/40/5 wt%) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (45/50/5 wt%) Differential scanning calorimetry Thermal properties of composites were investigated by non-isothermal DSC analysis. Figures 7.9 and 7.10 show the melting temperatures (T m ) and crystallization temperatures (T c ) of the PBS/PBAT blend and their composites with and without compatibilizer. The detailed second heating and first cooling cycles are summarized in Table 7.4. All the composites and blend matrix showed bimodal endoderm peaks, which correspond to two different types of crystal lamella formed during cooling. In the presence of miscanthus fibers, the melting point (114 C) 232

267 of PBS/PBAT blend is not affected significantly, as shown in Figure 7.9. However, the melting enthalpy ( H m) of the composites decreased considerably (Table 7.4). This can be attributed to the reduced volume fraction of the polymer in the resulting composites. The H m values of both compatibilized and uncompatibilized composites were significantly reduced in comparison to blend of PBS/PBAT. This result suggests that the composites require less energy to melt in comparison to the blend of PBS/PBAT. The melting enthalpy of the composites is directly related to the amount of polymer present in the composites. As such, less energy is required to melt a smaller amount of polymer in the composites than in the neat polymers. A double melting peak of PBS/PBAT blend has clearly separated in the composites. This can be attributed to the enhanced heterogeneous crystal formation with the addition of miscanthus fiber into PBS/PBAT blend. The crystallization temperature of PBS/PBAT blend was observed at o C (Figure 7.10). Figure 7.9. DSC second heating thermograms: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT+ miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%). 233

268 The crystallization rate is slightly decreased after the addition of miscanthus fiber into PBS/PBAT blend matrix. This is attributed to the miscanthus fibers restricting the polymer chain mobility and diffusion to the surface of the nuclei [45]. This result is in accordance with the PHBV/bamboo fiber composites [45], PHBV/pineapple leaf fiber composites [46] and PHBV/kneaf fiber composites [13]. Lee and Wang [47], however, reported that the PBS crystallization rate can be enhanced with the addition of natural fibers. Their finding contradicts results from the present study. This discrepancy could be due to the differences in the nature of the fibers [46]. Both compatibilized and uncompatibilized composites had a lower H c value than PBS/PBAT blend (Table 7.4). This can be attributed to the amount of PBS/PBAT blend present in the composites. Figure DSC first cooling thermograms: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT+ miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%). 234

269 Table 7.4. Detailed differential scanning calorimetry analysis of the PBS/PBAT blend and their composites Samples T c ( o C) H c (J/g) Tm ( o C) H m (J/g) PBS/PBAT PBS/PBAT + miscanthus fibers (70/30 wt%) PBS/PBAT + miscanthus fibers (60/40 wt%) PBS/PBAT + miscanthus fibers (50/50 wt%) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (65/30/5 wt%) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (55/40/5 wt%) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (45/50/5 wt%) Measurements of fiber diameter, length, and aspect ratio The average length and the diameter of miscanthus fibers before and after processing is summarized in Table 7.5. Before processing, the average length of miscanthus fibers was 4.65±2.5 mm with aspect ratio (L/D) of 63. After processing, however, the fiber length, diameter and aspect ratio were significantly reduced. The measured fiber lengths were within the range of mm. The reduced fiber length is attributed to the strong shear force, which developed during the compounding process. During the course of processing, the fibers can undergo defibrillation. This defibrillation leads to reduce the diameter of the fibers in the resulting composites. Figure 7.11 shows the fiber length distribution before and after compounding. Before compounding the fiber lengths were observed in the wide range of 1.5 to 15.5 mm. However, after compounding the fiber length distribution became narrower (range of 0.1 to 2.5 mm). 235

270 Before extrusion, the diameter of the fibers was 0.74±0.024 mm. After being subjected to an extrusion process a drastic reduction was observed in the fibers diameter. This is because of the fiber breakage during extrusion in a twin screw extruder [15]. Moreover, the individualization of the fiber bundles during high mechanical shear produced in the compounding chamber is perhaps responsible for this observation. Similar trends were observed in the sisal fiber reinforced PBS composites [48] as well as kenaf fiber reinforced starch grafted PP composites [49]. Table Average fiber length (L), average fiber diameter (D), and aspect ratio (L/D) of the miscanthus fiber before and after compounding Samples Miscanthus fibers before compounding PBS/PBAT+miscanthus fibers (70/30 wt%) PBS/PBAT+miscanthus fibers (60/40 wt%) PBS/PBAT+miscanthus fibers (50/50 wt%) Number of fibers Average length (L) (mm) Average diameter (D) (mm) Aspect ratio (L/D) ± ± ± ± ± ± ± ±

271 Figure Fiber length distribution before and after compounding: (A) as received miscanthus fibers distribution, (B) fibers distribution in 30 wt% composites, (C) fibers distribution in 40 wt% composites, and (D) fibers distribution in 50 wt% composites Morphology of composites The performance of the composite materials mainly depends on the interfacial interaction between the phases and dispersion of all components in a given matrix system. Generally, efficiently reinforced natural fibers are withstanding from the fiber pullout during fracture phenomenon. This type of fracture can be observed only in composites with good interfacial bonding between the phases. A composite with dissimilar polarity constituents resulted in weak interfacial adhesion between the constituents. The fracture surface SEM micrographs of uncompatibilized and compatibilized composites with different weight percentage of fiber load are shown in Figures 7.12 and As can be clearly seen, all the uncompatibilized composites had worse fiber dispersion, fiber debonding, fiber pullout, and fiber aggregation. This gives 237

272 evidence of poor interfacial bonding between the fiber-matrix and thus reduced performance of the composites [14]. Figure SEM micrographs of uncompatibilized PBS/PBAT blend composites with different fiber loads: (A) PBS/PBAT+miscanthus fibers (70/30 wt%), (B) PBS/PBAT+miscanthus fibers (60/40 wt%) and (C) PBS/PBAT+miscanthus fibers (50/50 wt%) By contrast, the morphology of the compatibilized composites (Figure 7.13) showed less fiber pullout and better interfacial bonding between fibers-matrix compared to the uncompatibilized composites. Similar types of interactions have been reported for PBS and PBAT based composites with MAH grafted compatibilizer [12, 24]. The improved fiber-matrix adhesion is consistent with enhanced mechanical properties of the compatibilized composites. Overall, morphological analysis concludes that the fiber-matrix adhesion has improved with the help of compatibilizer. 238

273 Figure SEM micrographs of compatibilized PBS/PBAT blend composites with different amount of fiber loads: (A) PBS/PBAT+miscanthus fibers+mah-g-pbs/pbat (65/30/5 wt%), (B) PBS/PBAT+miscanthus fibers+mah-g-pbs/pbat (55/40/5 wt%), and (C) PBS/PBAT+miscanthus fibers+mah-g-pbs/pbat (45/50/5 wt%) Rheological property Rheological properties can offer a detailed structural-property relationship of polymer composites. Therefore, the influences of fiber content and compatibilizer on the complex viscosity of the composites were investigated (Figure 7.14). The PBS/PBAT blend showed Newtonian flow behavior at lower frequency whereas a slight shear thinning behavior was observed at higher frequency. This behavior was commonly found in polymer melts because polymer chain entanglement density drastically reduced with increasing frequency. Additionally, the average end-to-end distance of the polymer chains increased at higher frequency range. Complex viscosity of the composites is higher at lower frequency as compared to the PBS/PBAT blend. Theoretically, the addition of fillers/fibers into the thermoplastic polymer matrix will lead 239

274 to an increased viscosity of the melt. This is possibly due to the rigidity of the fibers which restricts the polymer chain mobility in the melt state, thus causing viscosity improvement in the composites [50]. Figure Complex viscosity of PBS/PBAT blend and its composites: (A) PBS/PBAT, (B) PBS/PBAT + miscanthus fibers (70/30 wt%), (C) PBS/PBAT + miscanthus fibers (60/40 wt%), (D) PBS/PBAT + miscanthus fibers (50/50 wt%), (E) PBS/PBAT + miscanthus fibers + MAH-g- PBS/PBAT (65/30/5 wt%), (F) PBS/PBAT+ miscanthus fibers + MAH-g-PBS/PBAT (55/40/5 wt%), and (G) PBS/PBAT + miscanthus fibers + MAH-g-PBS/PBAT (45/50/5 wt%). In addition, an increase in viscosity can be observed with an increase in the fiber loading. This may be due to agglomeration of fibers and the increased fiber-fiber interaction in the composites. A similar effect is commonly observed in all the melt state composites. A steady viscosity reduction was observed in all the composite samples with increasing frequency. The complex viscosity of the compatibilized composites was slightly lower compared to the uncompatibilized counterparts in the tested frequency range. This could be due to the lower molecular weight of the compatibilizer, following the mixture rule [51]. The MAH grafted 240

275 PBS/PBAT compatibilizer showed much higher MFI value (81 g/10min at 190 o C with 2.16kg) than the neat PBS/PBAT blend MFI value (33 g/10min at 190 o C with 2.16kg). The MFI improvement of the maleic anhydride PBS/PPBAT compatibilizer is attributed to the molecular weight reduction in the presence of free radical initiator. Maleated compatibilizer viscosity should, thereby, be considered when analyzing compatibilized composites melt viscosities. The same trend has been already reported in the compatibilized PP blends and their composites [52]. 7.4 Conclusions A biocomposite was prepared from PBS/PBAT blend matrix and miscanthus fibers by a melt process. The moduli of the prepared composites were increased remarkably with the incorporation of miscanthus fibers into blend of PBS/PBAT matrix. These improvements suggest that the miscanthus fiber acts as reinforcement in the PBS/PBAT matrix system. The tensile and impact strength of the composites deteriorated with the addition of miscanthus fibers to PBS/PBAT blend matrix. This was due to the lack of interfacial bonding between the phases. In order to improve this issue in the resulting composites, a reactive compatibilizer (MAH-g- PBS/PBAT) was introduced into the composites. It was found that the mechanical properties of the compatibilized composites were noticeably increased. This is mainly because of the improved interfacial bonding between the components, which were demonstrated by SEM. The DSC analysis revealed that the crystallization temperatures of the composites were lower compared to the PBS/PBAT blend matrix. This can be attributed to the miscanthus fiber restricting the mobility and diffusion of polymer chains to the surface of the nuclei. The length and diameter of the fibers was reduced considerably after compounding. This is due to the fibrillation, which occurred during compounding. It was also found that the density of all the composites was slightly higher when compared to the PBS/PBAT blend. From this study, it can 241

276 be concluded that the prepared biodegradable polymer based composites are a possible candidate to replace non-biodegradable composites in applications where biodegradability is essential but extreme thermal and humidity exposure is not required. References [1] E. Chiellini, R. Solaro, Biodegradable Polymeric Materials, Advanced Materials, 1996, 8 (4): [2] E. Zini, M. Scandola, Green composites: An overview, Polymer Composites, 2011, 32 (12): [3] P. Bordes, E. Pollet, L. Avérous, Nano-biocomposites: Biodegradable polyester/nanoclay systems, Progress in Polymer Science, 2009, 34 (2): [4] M. Avella, E. Martuscelli, M. Raimo, Review Properties of blends and composites based on poly(3-hydroxy)butyrate (PHB) and poly(3-hydroxybutyrate-hydroxyvalerate) (PHBV) copolymers, Journal of Materials Science, 2000, 35 (3): [5] O. Faruk, A. K. Bledzki, H.-P. Fink, M. Sain, Progress Report on Natural Fiber Reinforced Composites, Macromolecular Materials and Engineering, 2013, n/a-n/a. [6] K. G. Satyanarayana, G. G. C. Arizaga, F. Wypych, Biodegradable composites based on lignocellulosic fibers-an overview, Progress in Polymer Science, 2009, 34 (9): [7] A. Bourmaud, S. Pimbert, Investigations on mechanical properties of poly(propylene) and poly(lactic acid) reinforced by miscanthus fibers, Composites Part A: Applied Science and Manufacturing, 2008, 39 (9): [8] Z. N. Terzopoulou, G. Z. Papageorgiou, E. Papadopoulou, E. Athanassiadou, E. Alexopoulou, D. N. Bikiaris, Green composites prepared from aliphatic polyesters and 242

277 bast fibers, Industrial Crops and Products, 2014, (0): (doi: /j.indcrop ). [9] A. K. Mohanty, M. Misra, G. Hinrichsen, Biofibres, biodegradable polymers and biocomposites: An overview, Macromolecular Materials and Engineering, 2000, (1): [10] H.-S. Kim, S. Kim, H.-J. Kim, H.-S. Yang, Thermal properties of bio-flour-filled polyolefin composites with different compatibilizing agent type and content, Thermochimica Acta, 2006, 451 (1 2): [11] V. Tserki, P. Matzinos, C. Panayiotou, Effect of compatibilization on the performance of biodegradable composites using cotton fiber waste as filler, Journal of Applied Polymer Science, 2003, 88 (7): [12] H.-S. Kim, B.-H. Lee, S. Lee, H.-J. Kim, J. Dorgan, Enhanced interfacial adhesion, mechanical, and thermal properties of natural flour-filled biodegradable polymer biocomposites, Journal of Thermal Analysis and Calorimetry, 2011, 104 (1): [13] M. Avella, G. Bogoeva-Gaceva, A. Buzõarovska, M. Emanuela Errico, G. Gentile, A. Grozdanov, Poly(3-hydroxybutyrate-co-3-hydroxyvalerate)-based biocomposites reinforced with kenaf fibers, Journal of Applied Polymer Science, 2007, 104 (5): [14] M. Z. A. Thirmizir, Z. A. M. Ishak, R. M. Taib, S. Rahim, S. M. Jani, Kenaf-bast-fiberfilled biodegradable poly(butylene succinate) composites: Effects of fiber loading, fiber length, and maleated poly(butylene succinate) on the flexural and impact properties, Journal of Applied Polymer Science, 2011, 122 (5):

278 [15] J. M. Lee, Z. A. Mohd Ishak, R. Mat Taib, T. T. Law, M. Z. Ahmad Thirmizir, Mechanical, Thermal and Water Absorption Properties of Kenaf-Fiber-Based Polypropylene and Poly(Butylene Succinate) Composites, Journal of Polymers and the Environment, 2013, 21 (1): [16] T. J. Keener, R. K. Stuart, T. K. Brown, Maleated coupling agents for natural fibre composites, Composites Part A: Applied Science and Manufacturing, 2004, 35 (3): [17] R. M. Johnson, N. Tucker, S. Barnes, Impact performance of Miscanthus/Novamont Mater-Bi biocomposites, Polymer Testing, 2003, 22 (2): [18] M. R. Nanda, M. Misra, A. K. Mohanty, Performance Evaluation of Biofibers and Their Hybrids as Reinforcements in Bioplastic Composites, Macromolecular Materials and Engineering, 2013, 298 (7): [19] G. Moad, The synthesis of polyolefin graft copolymers by reactive extrusion, Progress in Polymer Science, 1999, 24 (1): [20] R. Muthuraj, M. Misra, A. Mohanty, Biodegradable Poly(butylene succinate) and Poly(butylene adipate-co-terephthalate) Blends: Reactive Extrusion and Performance Evaluation, Journal of Polymers and the Environment, 2014, 22 (3): [21] A. K. Mohanty, M.Misr, L.T. Drzal. Natural fibers, biopolymers, and biocomposites; CRC Press-Taylor & Francis Group, Boca Raton (2005), pp , [22] C. Nyambo, A. K. Mohanty, M. Misra, Polylactide-Based Renewable Green Composites from Agricultural Residues and Their Hybrids, Biomacromolecules, 2010, 11 (6):

279 [23] S. Sahoo, M. Misra, A. K. Mohanty, Biocomposites From Switchgrass and Lignin Hybrid and Poly(butylene succinate) Bioplastic: Studies on Reactive Compatibilization and Performance Evaluation, Macromolecular Materials and Engineering, 2013, n/a-n/a. [24] C.-S. Wu, Utilization of peanut husks as a filler in aliphatic aromatic polyesters: Preparation, characterization, and biodegradability, Polymer Degradation and Stability, 2012, 97 (11): [25] A. Arbelaiz, B. Fernández, J. A. Ramos, A. Retegi, R. Llano-Ponte, I. Mondragon, Mechanical properties of short flax fibre bundle/polypropylene composites: Influence of matrix/fibre modification, fibre content, water uptake and recycling, Composites Science and Technology, 2005, 65 (10): [26] R. Bhardwaj, A. K. Mohanty, L. T. Drzal, F. Pourboghrat, M. Misra, Renewable Resource-Based Green Composites from Recycled Cellulose Fiber and Poly(3- hydroxybutyrate-co-3-hydroxyvalerate) Bioplastic, Biomacromolecules, 2006, 7 (6): [27] H. Wang, X. Sun, P. Seib, Strengthening blends of poly(lactic acid) and starch with methylenediphenyl diisocyanate, Journal of Applied Polymer Science, 2001, 82 (7): [28] J.-F. Zhang, X. Sun, Mechanical Properties of Poly(lactic acid)/starch Composites Compatibilized by Maleic Anhydride, Biomacromolecules, 2004, 5 (4): [29] Z. A. M. Ishak, B. N. Yow, B. L. Ng, H. P. S. A. Khalil, H. D. Rozman, Hygrothermal aging and tensile behavior of injection-molded rice husk-filled polypropylene composites, Journal of Applied Polymer Science, 2001, 81 (3):

280 [30] Y. C. Shinichi Shibata, Isao Fukumoto, Press forming of short natural fiber-reinforced biodegradable resin: Effects of fiber volume and length on flexural properties, Polymer Testing, 2005, 24 (8): [31] K. Kirwan, R. M. Johnson, D. K. Jacobs, G. F. Smith, L. Shepherd, N. Tucker, Enhancing properties of dissolution compounded Miscanthus giganteus reinforced polymer composite systems: Part 1. Improving flexural rigidity, Industrial Crops and Products, 2007, 26 (1): [32] E. Igarza, S. G. Pardo, M. J. Abad, J. Cano, M. J. Galante, V. Pettarin, C. Bernal, Structure fracture properties relationship for Polypropylene reinforced with fly ash with and without maleic anhydride functionalized isotactic Polypropylene as coupling agent, Materials & Design, 2014, 55 (0): [33] H.-S. Kim, H.-S. Yang, H.-J. Kim, Biodegradability and mechanical properties of agroflour filled polybutylene succinate biocomposites, Journal of Applied Polymer Science, 2005, 97 (4): [34] C. Varga, N. Miskolczi, L. Bartha, G. Lipóczi, Improving the mechanical properties of glass-fibre-reinforced polyester composites by modification of fibre surface, Materials & Design, 2010, 31 (1): [35] G. Kalaprasad, K. Joseph, S. Thomas, C. Pavithran, Theoretical modelling of tensile properties of short sisal fibre-reinforced low-density polyethylene composites, Journal of Materials Science, 1997, 32 (16): [36] S. Joseph, M. S. Sreekala, Z. Oommen, P. Koshy, S. Thomas, A comparison of the mechanical properties of phenol formaldehyde composites reinforced with banana fibres and glass fibres, Composites Science and Technology, 2002, 62 (14):

281 [37] R. Chen, M. Misra, A. K. Mohanty, Injection-moulded biocomposites from polylactic acid (PLA) and recycled carbon fibre: Evaluation of mechanical and thermal properties, Journal of Thermoplastic Composite Materials, 2013, [38] J. P. Reddy, S. Ahankari, M. Misra, A. Mohanty, A New Class of Injection Moulded Structural Biocomposites from PHBV Bioplastic and Carbon Fibre, Macromolecular Materials and Engineering, 2013, 298 (7): [39] V. Nagarajan, M. Misra, A. K. Mohanty, New engineered biocomposites from poly(3- hydroxybutyrate-co-3-hydroxyvalerate) (PHBV)/poly(butylene adipate-co-terephthalate) (PBAT) blends and switchgrass: Fabrication and performance evaluation, Industrial Crops and Products, 2013, 42 (0): [40] C. Nyambo, A. K. Mohanty, M. Misra, Effect of Maleated Compatibilizer on Performance of PLA/Wheat Straw-Based Green Composites, Macromolecular Materials and Engineering, 2011, 296 (8): [41] A. K. Mohanty, M. Misra, L. T. Drzal, Surface modifications of natural fibers and performance of the resulting biocomposites: An overview, Composite Interfaces, 2001, 8 (5): [42] H. Ku, H. Wang, N. Pattarachaiyakoop, M. Trada, A review on the tensile properties of natural fiber reinforced polymer composites, Composites Part B: Engineering, 2011, 42 (4): [43] S. Sahoo, M. Misra, A. K. Mohanty, Biocomposites From Switchgrass and Lignin Hybrid and Poly(butylene succinate) Bioplastic: Studies on Reactive Compatibilization and Performance Evaluation, Macromolecular Materials and Engineering, 2014, 299 (2):

282 [44] P. R. Hornsby, E. Hinrichsen, K. Tarverdi, Preparation and properties of polypropylene composites reinforced with wheat and flax straw fibres: Part I Fibre characterization, Journal of Materials Science, 1997, 32 (2): [45] S. Singh, A. K. Mohanty, T. Sugie, Y. Takai, H. Hamada, Renewable resource based biocomposites from natural fiber and polyhydroxybutyrate-co-valerate (PHBV) bioplastic, Composites Part A: Applied Science and Manufacturing, 2008, 39 (5): [46] S. Luo, A. N. Netravali, Mechanical and thermal properties of environment-friendly green composites made from pineapple leaf fibers and poly(hydroxybutyrate-covalerate) resin, Polymer Composites, 1999, 20 (3): [47] S.-H. Lee, S. Wang, Biodegradable polymers/bamboo fiber biocomposite with bio-based coupling agent, Composites Part A: Applied Science and Manufacturing, 2006, 37 (1): [48] Y.-H. Feng, D.-W. Zhang, J.-P. Qu, H.-Z. He, B.-P. Xu, Rheological properties of sisal fiber/poly(butylene succinate) composites, Polymer Testing, 2011, 30 (1): [49] A. Hamma, M. Kaci, Z. A. Mohd Ishak, A. Pegoretti, Starch-grafted-polypropylene/kenaf fibres composites. Part 1: Mechanical performances and viscoelastic behaviour, Composites Part A: Applied Science and Manufacturing, 2014, 56 (0): [50] N. Marcovich, M. Reboredo, J. Kenny, M. Aranguren, Rheology of particle suspensions in viscoelastic media. Wood flour-polypropylene melt, Rheologica Acta, 2004, 43 (3):

283 [51] F. Martí-Ferrer, F. Vilaplana, A. Ribes-Greus, A. Benedito-Borrás, C. Sanz-Box, Flour rice husk as filler in block copolymer polypropylene: Effect of different coupling agents, Journal of Applied Polymer Science, 2006, 99 (4): [52] N. Le Moigne, M. van den Oever, T. Budtova, Dynamic and capillary shear rheology of natural fiber-reinforced composites, Polymer Engineering and Science, 2013, 53 (12):

284 Chapter 8: Influence of Processing Parameters on the Impact Strength of Biocomposites: A Statistical Approach* *A version of this chapter has been published in: R. Muthuraj, M. Misra, A. K. Mohanty, Influence of Processing Parameters on the Impact Strength of Biocomposites: A Statistical Approach, Composites Part A: Applied Science and Manufacturing, 2015, DOI: /j.compositesa Abstract A biocomposite consisting of miscanthus fibers and a biodegradable polymer blend (poly(butylene succinate) (PBS) and poly(butylene adipate-co-terephthalate) (PBAT)) matrix was produced. The flexural strength of the PBS/PBAT/miscanhtus composites was increased considerably (~127%) when compared to the neat PBS/PBAT blend. This increase was attributed to the strong reinforcing effect of miscanthus fiber. The tensile strength of the PBS/PBAT/miscanthus composite was inferior to the PBS/PBAT blend matrix. The impact performance of the composites was optimized by manipulating the processing parameters such as processing temperature, fiber length, the holding pressure, and screw speed. A full factorial experimental design was used to predict the statistically significant variables on the impact strength of the PBS/PBAT biocomposites. Furthermore, a regression model was developed to study the impact strength of the composites. The main effects and interaction effects of the variables were studied using analysis of variance (ANOVA) and factorial plots at 95% confidence level. The accuracy of the developed model was examined by using residuals plots and coefficients. Among the selected independent processing parameters, fiber length has a most significant effect on the impact strength of the PBS/PBAT/miscanthus composites. The least significant effect on the impact strength was attributed to the holding pressures. A poor interfacial interaction between the miscanthus fibers and the PBS/PBAT blend was observed by means of impact fractured surface morphological analysis. From scanning electron microscopy 250

285 (SEM) analysis, it can be observed that the composites prepared with 2.07 mm miscanthus fibers had more fiber pullouts than 4.65 mm counterparts. This phenomenon could be responsible for the observed high impact strength of the composites with 2.07 mm fibers compared to composites prepared with 4.65 mm fibers. 8.1 Introduction The increasing environmental pollution throughout the world has placed great emphasis on eco-friendly sustainable material development. Consequently, more attention has been focused on a sustainable material development by using bio-based and/or biodegradable materials instead of petroleum based non-biodegradable materials. Governments in many countries are supporting the usage of green products, and a reduction of dependence on petroleum because of the associated environmental benefits [1]. Currently there are many bioplastics (biodegradable and/or bio-based polymers) available in the market. Among them, poly(butylene succinate) PBS, and poly(butylene adipate-co-terephthalate) PBAT are promising biodegradable polyesters. The impact toughness/strength of the PBS is insufficient for a wide range of applications [2]. Blending PBS with PBAT can enhance the impact and tensile toughness of the PBS [3]. However, these polymers still cannot be used for wide range of applications on their own because they cannot fulfill some of the product requirements [4]. These issues can be overcome by blending, reinforcing, and forming composites with inexpensive natural fibers in the polymer matrices [1]. Natural fiber (Kenaf, flax, hemp and jute) reinforced composites have been used for many applications including those in the automotive, electronic, horticultural, packaging, consumer goods and construction sectors [5]. Miscanthus is an alternative fiber for viable biocomposite applications. Miscanthus is a typical lignocellulosic perennial grass and is a 251

286 promising non-food crop, which grows rapidly compared to some other crops. There are many advantages of utilizing miscanthus as reinforcement in composites such as high yield [6], low moisture content at harvest [6], low input conditions [7], soil remediation potential [7], good fiber properties (tensile strength, hardness, and modulus) [8], and thermal stability up to 200 o C [9]. Currently, miscanthus fibers have only limited applications though these could be diversified by developing viable biocomposites. The key strategy is the combination of bioplastics with miscanthus fibers, which could create an eco-friendly sustainable biocomposite. Recently, the performance of miscanthus fibers reinforced in a biodegradable polymer matrix has been investigated by few researchers [6-8, 10-13]. In order to compare the effect of miscanthus fibers on the resulting composites, Nagarajan et al., [12] investigated the performance of five different lignocellulosic fibers (miscanthus, switchgrass, wheat straw, soy stalk and corn stalk) reinforced poly(hydroxybutyrate-co-valerate) PHBV/PBAT (45/55 wt%) composites. This study revealed that the miscanhtus fiber reinforced PHBV/PBAT composites exhibited superior properties compared to other fiber reinforced PHBV/PBAT composites. Similar observations were made in the miscanthus fibers reinforced PHBV/polylactide (PLA) (60/40 wt%) composites [7]. A recent study by Zhang et al., [11] reported that the toughened multiphase green composite can be obtained from miscanthus fiber reinforced PHBV/PBAT/epoxidized natural rubber (ENR) matrix. In a multi-phase material, processing parameters and variables play a vital role in the performance of the resulting material. Recently, many researchers conducted experimental studies to investigate the performance of heterogeneous composite materials. Johnson et al., [13] used a two-level factorial design to investigate the influence of processing parameters such as temperature, screw speed, filler content, and size on the impact performance of Mater- 252

287 Bi /miscanthus composites. In another study [6] a Box-Cox transformation method was used to examine the influence of processing parameters on the performance of the Mater-Bi /miscanthus fiber composites. From these studies, it was noted that processing temperature has more influence on the performance of the composites. The significant influences of processing parameters (processing temperature, screw speed, humidity, filler content, and the aspect ratio of filler) on the elastic modulus, heat deflection temperature and impact strength of the Mater- Bi /wood flour composites has been studied by Morreale et al., [14]. The selected processing variables are based on specific industrial target applications including automotive indoor furnishing and panels. The filler aspect ratio had more influence on the impact strength while filler content exhibited more influence on the heat deflection temperature as well as the elastic modulus. Another study by Kirwan et al., [5] studied the influence of processing parameters on the flexural properties of poly(vinyl alcohol), PVA/poly(vinyl acetate), PVAc/miscanthus composites. The authors found that the processing temperature to be the most influencing factor on the flexural properties followed by the washing of fibers. Most studies in current literature investigate the composites (short fiber reinforced biodegradable polymer blend matrix based composites) characteristics with fixed processing parameters. The aim of this work was to fabricate biocomposites using a bioplastic blend of PBS/PBAT (60/40 wt%) as the matrix and miscanthus fibres as the reinforcement with several independent processing variables (processing temperature, screw speed, holding pressure, and fiber length). The influence of the processing variables on the performance of the biocomposites was investigated by using a statistical approach, i.e., full factorial design and analysis of variance (ANOVA). 253

288 8.2 Full factorial design methodology More than one factor can affect an experimental result. In general, a factorial experiment studies the simultaneous effects of two or more factors on experimental results [15]. In the literature, it was suggested that there are many independently controllable processing parameters/factors (processing temperature, mixing speed, pressure, reinforcement amount, and size) which influence the performance of the resulting biocomposite [6, 13, 14]. As such, a detailed investigation was conducted to fix the upper and lower limits of the independent processing parameters for PBS/PBAT/miscanthus composites fabrication. The miscanthus fibers can be melted and compounded with polymers at up to 200 o C without exhibiting any major thermal decomposition [6]. As a result the miscanthus fiber reinforced composites preparation should be performed below said temperature. Shear force occurs during the extrusion process and can damage the fiber geometry. The mixing speed or screw speed of the composites can play a vital role in maintaining sufficient fiber geometry (aspect ratio). For instance, high screw speed may cause more fiber breakage while low screw speed can lead to less homogeneity of the components in the composites. In this study miscanthus is used as a short fiber which should have a wide range of aspect ratio distributions. This work aims to study the effect of two different fiber lengths (2.07 and 4.65 mm) on the mechanical performance of the composites. One of the hypotheses that were tested in this work is that high holding pressure will lead to a composite with higher performance. High holding pressure may also help to reduce the shrinkage of the resulting composites. The processing temperature, fiber length, holding pressure and screw speed were selected to be the variables in the factorial design. Two levels were assigned for each of these parameters for the composite fabrication as shown in Table

289 Table 8.1. Selected processing parameters and their respected levels S.No Factor Notation Lower level Higher level 1 Temperature ( o C) A Holding pressure (bar) B Screw speed (rpm) C Fiber length (mm) D 2 4 These parameter levels were selected after a series of screening experiments had been conducted. In order to fully understand the interaction between the parameters, a full factorial design was selected in a 2 k experimental design. In this study, randomization was carried out to increase the precision of the experimental results by reducing the sampling variability. It was assumed that the experimental results between the two levels are linear. The experimental design was performed in statistical software Minitab 17 and, the same software was used to analyze the results by means of statistical plots (main effect plot, interaction effect plot, normal probability plot, residual plots, and Pareto plot) at a 95% confidence level. 8.3 Materials Poly(butylene succinate), PBS and poly(butylene adipate-co-terephthalate), PBAT were obtained from Xinfu pharmaceuticals Co. Ltd, China. Both PBS and PBAT are semi-crystalline grade produced from fossil fuel based monomers. The PBS and PBAT have onset thermal degradation temperatures of 372 and 377 o C, respectively [16]. Two different lengths (2.08±0.95 and 4.65±2.45 mm) of miscanthus fiber were kindly supplied by New Energy Farms, Ontario, Canada. Hereafter, these two fiber lengths (2.08±0.95 and 4.65±2.45 mm) will be referred as 2 and 4 mm. All the materials were used without any further purification. 255

290 Table 8.2. Physical and mechanical properties of the PBS/PBAT blend and miscanthus fibers Properties Values Melt flow index of PBS/PBAT (60/40 wt%) 33±3 g/10min (190 o C with 2.16 kg) Notched Izod impact strength of PBS/PBAT blend Non-break [3] Onset thermal degradation of miscanthus fiber* ~260 o C Density of the miscanthus fibers* 1.41 g/cm 3 Modulus of the miscanthus fibers 9.49 GPa [8] *Thermal stability and density of miscanthus fibers were measured in our previous study (Chapter 7) 8.4 Experimental procedure Samples preparation Based on our previous study [16], we have selected a PBS/PBAT (60/40 wt%) blend as an optimum composition for composites fabrication. Hereafter, the PBS/PBAT (60/40 wt%) blend will be referred as PBS/PBAT blend. Table 8.2 shows some of the general properties of miscanthus fiber and PBS/PBAT blend. A 30 wt% miscanthus fiber is selected to fabricate biocomposites with a blend of biodegradable polymer (PBS/PBAT) matrix. The choice of using miscanthus fiber in this present work was because of its good fiber properties and the strong potential supply. Prior to melt compounding, both polymers and miscanthus fibers were dried at 80 C at least 12 h. Appropriate amounts of dried polymers and fibers were manually pre-mixed at the solid state and the composites fabrication was performed by changing four processing variables, as shown in Table 8.3. The PBS/PBAT/miscanthus fiber composites were prepared in a lab-scale extrusion and injection molding process. The lab-scale co-rotating twin-screw extruder (DSM explore, Netherlands) and injection molding machine (DSM explore, Netherlands) had volume of 15 and 12 cm 3, respectively. The composite samples were molded 256

291 with a mold temperature of 30 o C and the residence time of the materials inside the extrusion barrel was 2 min. Table 8.3. The 16 investigated experimental conditions Experiment Temperature ( o C) Screw speed (rpm) Holding pressure (bar) Fiber length (mm) Characterization methods Fiber dimension measurement In order to measure the fiber length after processing, the composite samples were dissolved in chloroform and then fibers were isolated by filtering. The isolated fibers were rinsed thoroughly with the same solvent and dried at 70 o C for 24 h. The processed and unprocessed fibers were photographed through a digital camera (Nikon AF-S DX) and the fiber dimensions were measured by Image J software (at least 85 individual fibers were measured). The measured fibers length was inserted in Minitab 17 statistical software to get fiber length distribution histogram. 257

292 8.5.2 Mechanical testing and scanning electron microscopy (SEM) All the prepared test specimens were conditioned at room temperature for at least 40 h before evaluating mechanical performances. Universal Testing Machine (Instron, Model-3382) was use to measure the flexural and tensile properties of the test samples in accordance with ASTM D790 and ASTM D638, respectively. The flexural testing was performed with a crosshead speed of 14 mm/min. The tensile properties of neat PBS/PBAT blend and all the composites were measured with a cross-head movement of 50 and 5 mm/min, respectively. Notched Izod impact testing was performed according to ASTM D256 in a TMI monitor impact testing machine using a 5 ft.lb pendulum. The reported tensile and flexural properties are averages of five samples for each formulation. Minimum six test samples were tested for each formulation and the average values are reported for impact strength. The morphologies of the fracture surface were observed by using a SEM. Prior to observation of the sample morphology; the samples were sputter-coated with a thin layer of gold. The analysis has been performed in Inspect S50-FEI SEM. 8.6 Results and Discussion Mechanical properties Table 8.4 represents the mechanical properties of biocomposites as well as the factors that are used for each experiment. Due to the reinforcing effect of miscanthus fibers, the flexural strengths of all the composites were higher than that of the neat PBS/PBAT blend (denoted as control). A similar trend has been observed in the PHBV/PBAT/miscanthus composites and rubber toughened PHBV/PBAT/miscanthus fiber composites [11, 12]. Kirwan et al., [5] have found to improve flexural properties of miscanthus fiber reinforced PVA/PVAc blend. Due to strong reinforcing capability of miscanthus fibers, both tensile and flexural modulus of the composites increased (data not shown) compared to neat PBS/PBAT blend. On the other hand, 258

293 all the composites showed inferior tensile strength as compared to control sample (matrix). The observed reduction is attributed to the incompatibility between the miscanthus fiber and the PBS/PBAT blend matrix. Such a reduction is very often observed in natural fiber reinforced thermoplastic composites [7]. The miscanthus fibers have higher wax and silica content compared to wood fibers which may be responsible for the incompatibility between the PBS/PBAT matrix and the miscanthus fibers [13]. The tensile and flexural strengths of the PBS/PBAT/miscanthus composites are not significantly affected with varying processing parameters. Similarly, tensile and flexural modulus of the PBS/PBAT/miscanthus composites was not affected significantly with varying processing parameters (data not shown). The Izod impact strength of the control sample (matrix) showed non-break behavior under tested impact conditions. Contrastingly, all composite samples showed hinge break behavior under the selected test conditions. This phenomenon was attributed to the incorporation of stiff fibers into a ductile polymer matrix. Taking this into consideration, the observed impact strength of the PBS/PBAT/miscanthus composites was still superior to carbon fiber reinforced composites such as PLA/carbon fiber (70/30 wt%) [17], PHBV/carbon fiber (70/30 wt%) [18], poly(trimethylene terephthalate), PTT/carbon fiber composites (70/30 wt%) [19]. It can be noted that the processing parameters have more influenced on the impact strength of PBS/PBAT/miscanthus composites. 259

294 Table 8.4. A complete summary of all the experiments and the related mechanical properties of PBS/PBAT/miscanthus composites Experiment Temperature ( o C) Screw speed (rpm) Holding pressure (bar) Fiber length (mm) Tensile strength (MPa) Flexural strength (MPa) Impact strength (J/m) ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±2.95 Control ± ±0.27 non-break Analysis of variance (ANOVA) for impact strength ANOVA is a statistical model, which can be used to investigate the significant main and interaction effects of factors with respect to response. The model had 15 degrees of freedom with four factors and two levels. To estimate the individual and interaction factors upon the impact strength, sum of square (SS), mean square (MS), F-test statistics and p-values are presented in the ANOVA Table

295 Table 8.5. Analysis of Variance (ANOVA) for notched Izod impact strength Source DF Sum of squares (SS) Mean squares (MS) F P Temperature Screw speed Holding pressure Fiber length Temperature*Screw speed Temperature*Holding pressure Temperature*Fiber length Screw speed*holding pressure Screw speed*fiber length Holding pressure*fiber length Error Total S = R-Sq = 87.78% R-Sq(adj) = 63.35% In this study we have used an alpha level (α) = 0.05 for each F test to analyze the factorial design experiment. Usually, the higher value of F-ratio suggests more influence of that factor on the experiment response. According to an F-test, F=25.61 has a p-value of Since the p- value is less than 0.05, we then have sufficient evidence to conclude that the mean impact strength of the biocomposites was significantly influenced by fiber length. At a 95% confidence interval (P<0.05), it should be noted that the screw speed and holding pressure do not show significant effects on the impact strength of the composites. The interaction effects did not significantly influence the impact strength of the composites. The square correlation coefficient (R 2 ) was used to judge the adequacy of the developed model fit. The R 2 value can be interpreted 261

296 as the percentage reduction in the total variation in the experiment obtained by using the developed model. The typical R 2 value is 0 R 2 1. The value of R 2 (87.78%) and R 2 adj (63.35%) is substantial and hence the developed model fits the experimental results very well Effect of processing parameters on the impact strength Among the mechanical properties, impact strength was more affected by the processing factors than other mechanical properties. The impact strength of the short fiber composites mainly influenced by many factors including matrix intrinsic properties, optimum fiber-matrix interaction, fiber concentration, fiber geometry, fiber-matrix stress transfer efficiency, fiber orientation, and fiber dispersion and distribution [20]. At the same time, the fiber bridging, fiber pull-outs, crack propagation and matrix deformation mechanisms contribute a vital role in the impact rupture of short fiber reinforced composites [21]. Many of these mechanisms contribute simultaneously during impact tests, which make it complicated to determine the impact strength of the composites. Therefore, it is important to investigate the statistically significant factors upon the impact performance. Morreale et al., [14] have studied the impact performance of the composites with varying processing parameters. In addition, John et al., [6, 13] have performed a systematic study of impact performance of Mater-Bi /miscanthus composites by using factorial design. In the present study, the effect of processing parameters/factors on the notched Izod impact strength was statistically analyzed. More specifically, the statistical analysis was mainly focused on determining which factors and interactions parameters had more influence on the Izod impact energy of the biocomposites. Generally, the plot which provides a response with respect to the changes in the levels of the factors is called the main effect plot. Figure 8.1 shows the influence of the investigated factors on the impact strength of the resulting biocomposites. 262

297 Figure 8.1. Main effect plot for the impact strength Factors with steeper slopes have larger effects and thus a greater influence on the results. From the main effect plots, it can be observed that the temperature and fiber length factor levels have more significant effect, which is evidenced with a strong line slope. On the other hand, holding pressure has almost no effect on the response when varying its levels. The composites prepared with low temperature processing (140 o C), low screw speed (100 rpm), and small fiber length (2 mm) have higher impact strength compared to those produced with high processing temperature (180 o C), high screw speed (150 rpm) and high fiber length (4 mm). The holding pressure did not have a great effect on the impact strength upon changing levels such as 6 and 10 bar. Figure 8.2 represents the statistically significant binary interaction between the selected variables. The joint effects of two factors such as fiber length/holding pressure, fiber length/screw speed, holding pressure/screw speed, fiber length/temperature, holding pressure/temperature, and screw speed/temperature were investigated. If there was no interaction between the selected variables, the lines on the display should have been approximately parallel. 263

298 When the response of two factors was not parallel, this indicates a possible interaction between the selected factors. Figure 8.2. Plot of interaction effects for the impact strength of biocomposites Among the selected variable combinations, it can be noted that the most significant interaction variables are holding pressure/temperature, fiber length/temperature, screw speed/holding pressure, and fiber length/holding pressure. There is no significant interaction between the temperature/screw speed and screw speed/holding pressure on the impact strength of the resulting biocomposites. It can be concluded that the selected variable combinations (temperature/screw speed and screw speed/fiber length) behave separately, which are not dependent on each other. Out of the selected four variables the fiber length had the most significant effect on impact strength while screw speed had the least significant effect. 264

299 Figure 8.3 shows the significant factors influencing the impact strength of the PBS/PBAT/miscanthus composites with a confidence level (α) of According to the half normal probability plot the points which are farther away from the fitted line represent the significant effect on the impact strength. The points which appear close to the straight line indicate insignificant effects on the impact strength. In the Figure 8.3, it can be seen that all the significant factors (temperature and fiber length) are represented as square symbols while those not significant factors are presented as circle symbols. Figure 8.3. Half Normal probability plot of the standardized effects for impact strength of the PBS/PBAT/miscanthus composites The individual and interaction factors for the impact strength of the biocomposites can be investigated using a horizontal Pareto chart and the results are shown in Figure 8.4. A Pareto chart is a bar chart that orders the bars from largest to smallest along with a vertical line. This chart is often used to analyze the statistical significant difference of the individual and interaction effects on the response. The vertical line in the Pareto chart indicates the significant factors on 265

300 the response. For example, the bars extended to the right hand side of the vertical line are significant. Figure 8.4. Pareto chart of the standardized effects for the impact strength of the PBS/PBAT/miscanthus biocomposites In the present study, a Student s t-test was performed in a Pareto chart with 15 degrees of freedom at a 95% confidence interval. The t-value (vertical line in the chart) was found to be 2.57 which determine the significant factors and/or interactions on the impact strength of the composites. The fiber length (D) had significant effect upon the impact strength of the composites because the standardized effect value is higher than vertical line standardized effect value 2.57 (t-value). The processing temperature exhibited significant effect on the impact performance of the Mater-Bi /miscanthus composites [6, 13]. Contrary to our present result particle size did not significantly influenced the impact performance of the biopolymer/miscanthus composites [6, 13]. This could be due to the morphological difference between the materials due to changing the processing variables. 266

301 Generally, the tensile toughness of the composites can be calculated from area under the stress-strain curve. Figure 8.5 shows the stress-strain curves of PBS/PBAT composites with 2 mm miscanthus fibers (B) and 4 mm miscanthus fibers (A). These two composites are prepared with same processing conditions whilst changing fiber lengths. It can be noticed that the composites prepared with 2 mm fibers showed better tensile toughness compared to composites prepared with 4 mm fibers. This result has good agreement with observed impact strength of the composites with 2 mm fibers. Figure 8.5. Tensile stress-strain curves of PBS/PBAT/miscanthus composites with changing fiber length 4 mm (A) and 2 mm (B) Fiber length distribution Before and after processing, the length distribution of miscanthus fibers is shown in Figure 8.6. After processing, the fiber length distribution is broader compared to before processing. At the same time, the fiber length is drastically reduced. It can be observed that the length of 4.65 and 2.07 mm miscanthus fibers is not significantly different after processing. For instance, after processing, the miscanthus fibers with average length reduced from 4.65±2.5 to 1.07±0.34 mm and from 2.07±0.94 to 0.80±0.39 mm. After processing, the length distribution of 267

302 Number of fibers Number of fibers Number of Fibers Number of fibers 4.65 mm fibers is varied from 0.45 to 1.9 mm while 2.07 mm fibers varied from 0.2 to 1.9 mm. This is because of the fiber breakage during extrusion in a twin-screw extruder [22] Long fibers before processing Mean StDev N Long fibers after processing Mean StDev N Fiber Length (mm) Fiber length (mm) Short fibers before processing Mean StDev N Short fibers after processing Mean StDev N Fiber length (mm) Fiber length (mm) Figure 8.6. Histograms of miscanthus fiber length distribution before and after compounding in a twin screw extruder Moreover, the individualization of the fiber bundles during high mechanical shear produced in the compounding chamber is perhaps responsible for this observation. Similar trends were observed in the sisal fiber reinforced PBS composites [23] as well as kenaf fiber reinforced starch grafted PP composites [24]. The fiber length has a strong influence on mechanical performances [25]. It can be noted from Figure 8.6, most of the fibers distributed with >0.9 mm length in the composites fabricated with 4.65 mm fibers. On the other hand, the composites 268

303 processed with 2.07mm fiber composites showed most of the fibers distributed <0.9 mm length in the resulting composites. Based on the fiber distribution after processing, more number of fiber ends can be observed in the composites prepared with 2.07 mm miscanthus fibers compared to 4.65 mm miscanthus fibers. Consequently, more number of fiber pullouts can be expected from the composites prepared with 2.07 mm miscanthus fibers compared to 4.65 mm miscanthus fibers counterpart. The occurrence of more fiber pullout may be responsible for the observed high impact strength in the composites prepared with 2.07 mm miscanthus fibers. Table 8.6. Average fiber length (L), average fiber diameter (D), and aspect ratio (L/D) of the miscanthus fiber before and after compounding Samples Number of fibers Average length (L) (mm) Average diameter (D) (mm) Aspect ratio(l/d) Long fibers before ± ± compounding Short fibers before ± ± compounding Long fibers after compounding ± ± Short fibers after compounding ± ± In general, the composites with a higher aspect ratio fiber should provide superior impact strength than the composites with lower aspect ratio fiber. For a given composites system, the recovered fiber length and aspect ratio were examined (Table 8.6) which revealed that the longer fibers had a higher aspect ratio (3.8) than short fibers (3.2) after processing. The composite with lower aspect ratio showed higher impact strength while the composite with higher aspect ratio had lower impact strength. This could be due to the difference in the fiber orientation during sample preparation [5, 26, 27]. The high aspect ratio fibers can align across the samples and thus fail to effectively transfer stress between the fiber and matrix. During impact fracture the crack 269

304 initiation and propagation are mainly influenced by matrix behavior and morphology of the sample, respectively [28]. This phenomenon could play a vital role on the impact strength of PBS/PBAT/miscanthus composites when changing the fiber lengths Scanning electron microscopy The impact strength of the short fiber reinforced composites is influenced by many parameters including fiber pull-out and degree of adhesion [29]. In order to study the impact fracture mechanism of PBS/PBAT/miscanthus composites, the surface morphology of the impact fractured samples was investigated by SEM analysis. In the short fiber composites the fibers with subcritical aspect ratio lead to fiber pullout during fracture [30]. Figure 8.7 (a) and (b) represent the SEM morphology of the 2 mm miscanthus fiber reinforced PBS/PBAT composites and 4 mm miscanthus fiber reinforced PBS/PBAT composites, respectively. Figure 8.7. Represents the SEM micrographs of the PBS/PBAT/miscanthus composites; (a) PBS/PBAT composites with 2 mm miscanthus (b) PBS/PBAT composites with 4 mm miscanthus The SEM micrographs of both composites indicate that the fiber pullout mechanism and poor interfacial bonded regions played eminent role during impact fracture of the composites. There was no clear morphological difference witnessed in the composites with 2 and 4 mm 270

305 fibers. However, the observed impact strength difference between the 2 and 4 mm fiber composites may be due to the combined effects of pullout, energy dissipation mechanism and fiber orientation [30, 31]. Further work could be performed to find out which mechanism is responsible to determine the impact strength of miscanthus fibers reinforced PBS/PBAT composites Mathematical model development The predicated response of the composites is Y and it can be represented by equation (8.1) as a function of independent factors: Y=f(A,B,C,D) (8.1) The polynomial equation was used to explain the main and interaction effect of all the independent variables [15]. The polynomial equation can be expressed as follows, Y = X0 + X1 (A) + X2 (B) + X3 (C) + X4 (D) + X5 (AB) + X6 (AC) + X7 (AD) + X8 (BC) + X9 (BD) +X10 (CD) + X11 (ABC) +X12 (ABD) +X13 (ACD) + X14 (BCD) + X15 (ABCD) (8.2) The term X0 represents average response (impact strength) value, X1, X2,.X15 is the regression coefficient of main and interaction effects, A is processing temperature, B is screw speed, C is holding pressure and D is fiber length. In the equation (8.2), three and four factor interactions are not considered due to their insignificance [32]. The equation (8.2) can thereby be modified as; Y = X0 + X1 (A) + X2 (B) + X3 (C) + X4 (D) + X5 (AB) + X6 (AC) + X7 (AD) + X8 (BC) + X9 (BD) + X10 (CD) (8.3) The regression coefficients were calculated using MINITAB 17 statistical software for impact strength. Substituting significant factor coefficients into Equation 8.3, it can be rewritten as follows: 271

306 Y (impact strength) = (temperature) (fiber length) (8.4) When compared to longer and shorter miscanthus fiber reinforced PBS/PBAT composites, the composites with a shorter fiber showed higher impact strength than the composites with a longer fiber. The observed impact strength difference could be mainly due to the difference in fiber distribution in the matrix. Generally, the longer fibers can lead to agglomeration and thus favoring for crack initiation and poor stress transfer during impact test Diagnostic verification of the developed model The assumption underlying the analysis of variance for each experimental design is similar to those required for a regression analysis. Assumptions for a completely randomized design are that the data for the treatment have normal probability distribution with equal variances. The assumptions can be checked with the residual plots. The normal probability plot/normal plot for the notched Izod impact strength of the biocomposites is shown in Figure 8.8. To meet the normality assumption points should fall close to straight line on the normal plot. The normal plot of the impact strength data is dispersed along a straight line, which indicates that the assumption of normal distribution is valid. The plot of residuals versus fit can be used to ensure the linear model adequacy. 272

307 Figure 8.8. Normal probability plot of the residuals for impact strength Figure 8.9 shows the plot of the residual versus fit for impact strength of the PBS/PBAT/miscanthus composites. Figure 8.9 shows the variation of impact strength from 1.5 to 1.5 J/m in between fitted and observed values. From the residual versus fit plot, the random scatter of residuals around the horizontal line can be seen which indicates that the model is adequate for impact strength data. Figure 8.9. Residual plots versus fitted values for impact strength 273

308 The typical residual plot in Figure 8.10 represents the residual versus observation order of the impact strength of PBS/PBAT/miscanthus composites. There is no distinct pattern observed in the residuals plot. Both positive and negative residuals are evenly distributed along the observation order in Figure This observation suggests that the impact strength of PBS/PBAT/miscanthus composites is distributed normally. Figure Variation of the residuals with observed order values of the impact strength of the PBS/PBAT/miscanthus composites. 8.7 Conclusions In conclusion the miscanthus fibers can be used as a reinforcing agent for tough biodegradable polymers. The stiffness and flexural strength of the PBS/PBAT (60/40 wt%) blends is improved with addition of miscanthus fibers. This is a common observation in natural fiber reinforced composites. The impact strength of the PBS/PBAT blend was considerably reduced after incorporation of miscanthus fiber into PBS/PBAT blend matrix. This could be due to the phase separation of the components in the multiphase material. However, the composites with 2 mm fiber showed superior impact resistance than 4 mm fiber reinforced composites. This impact strength variation could be due to the difference in fiber pullout mechanism during impact test. The influence of independent processing variables on the impact strength of 274

309 PBS/PBAT/miscanthus composites has been investigated by 2 4 full factorial design of experiment. Using student s t-test and F-test, the statistically significant main and interaction variables were analyzed at a 95% confidence level. According to main effect plot, Pareto plot, and half-normal plot, the fiber length plays an important role on the impact strength of the composites as does processing temperature. From the normality plot, it was observed that the data are normally distributed along the straight line with R 2 value of 87.78%. Further work could be performed by maximizing more number of variables as well as levels. References [1] G. Bogoeva-Gaceva, M. Avella, M. Malinconico, A. Buzarovska, A. Grozdanov, G. Gentile, M. E. Errico, Natural fiber eco-composites, Polymer Composites, 2007, 28 (1): [2] J. Wang, L. Zheng, C. Li, W. Zhu, D. Zhang, Y. Xiao, G. Guan, Fully biodegradable blends of poly(butylene succinate) and poly(butylene carbonate): Miscibility, thermal properties, crystallization behavior and mechanical properties, Polymer Testing, 2012, 31 (1): [3] R. Muthuraj, M. Misra, A. K. Mohanty, Hydrolytic degradation of biodegradable polyesters under simulated environmental conditions, Journal of Applied Polymer Science, 2015, 132, [4] H.-S. Kim, H.-J. Kim, J.-W. Lee, I.-G. Choi, Biodegradability of bio-flour filled biodegradable poly(butylene succinate) bio-composites in natural and compost soil, Polymer Degradation and Stability, 2006, 91 (5): [5] K. Kirwan, R. M. Johnson, D. K. Jacobs, G. F. Smith, L. Shepherd, N. Tucker, Enhancing properties of dissolution compounded Miscanthus giganteus reinforced polymer composite systems: Part 1. Improving flexural rigidity, Industrial Crops and Products, 2007, 26 (1):

310 [6] M. Johnson, N. Tucker, S. Barnes, K. Kirwan, Improvement of the impact performance of a starch based biopolymer via the incorporation of Miscanthus giganteus fibres, Industrial Crops and Products, 2005, 22 (3): [7] M. R. Nanda, M. Misra, A. K. Mohanty, Performance Evaluation of Biofibers and Their Hybrids as Reinforcements in Bioplastic Composites, Macromolecular Materials and Engineering, 2013, 298 (7): [8] A. Bourmaud, S. Pimbert, Investigations on mechanical properties of poly(propylene) and poly(lactic acid) reinforced by miscanthus fibers, Composites Part A: Applied Science and Manufacturing, 2008, 39 (9): [9] P. Szabó, G. Várhegyi, F. Till, O. Faix, Thermogravimetric/mass spectrometric characterization of two energy crops, Arundo donax and Miscanthus sinensis, Journal of Analytical and Applied Pyrolysis, 1996, 36 (2): [10] K. Zhang, V. Nagarajan, N. Zarrinbakhsh, A. K. Mohanty, M. Misra, Co-injection molded new green composites from biodegradable polyesters and miscanthus fibers, Macromolecular Materials and Engineering, 2014, 299 (4): [11] K. Zhang, M. Misra, A. K. Mohanty, Toughened Sustainable Green Composites from Poly(3-hydroxybutyrate-co-3-hydroxyvalerate) Based Ternary Blends and Miscanthus Biofiber, ACS Sustainable Chemistry and Engineering, 2014, 2 (10): [12] V. Nagarajan, A. K. Mohanty, M. Misra, Sustainable Green Composites: Value Addition to Agricultural Residues and Perennial Grasses, ACS Sustainable Chemistry and Engineering, 2013, 1 (3): [13] R. M. Johnson, N. Tucker, S. Barnes, Impact performance of Miscanthus/Novamont Mater- Bi biocomposites, Polymer Testing, 2003, 22 (2):

311 [14] M. Morreale, R. Scaffaro, A. Maio, F. P. La Mantia, Mechanical behaviour of Mater- Bi /wood flour composites: A statistical approach, Composites Part A: Applied Science and Manufacturing, 2008, 39 (9): [15] D.C. Montgomery Design and Analysis of Experiments (4th ed.) Wiley, New York (1997). [16] R. Muthuraj, M. Misra, A. Mohanty, Biodegradable Poly(butylene succinate) and Poly(butylene adipate-co-terephthalate) Blends: Reactive Extrusion and Performance Evaluation, Journal of Polymers and the Environment, 2014, 22 (3): [17] R. Chen, M. Misra, A. K. Mohanty, Injection-moulded biocomposites from polylactic acid (PLA) and recycled carbon fibre Evaluation of mechanical and thermal properties, Journal of Thermoplastic Composite Materials, 2014, 27 (9): [18] J. P. Reddy, S. Ahankari, M. Misra, A. Mohanty, A New Class of Injection Moulded Structural Biocomposites from PHBV Bioplastic and Carbon Fibre, Macromolecular Materials and Engineering, 2013, 298 (7): [19] S. Vivekanandhan, M. Misra, A. K. Mohanty, Thermal, mechanical, and morphological investigation of injection molded poly(trimethylene terephthalate)/carbon fiber composites, Polymer Composites, 2012, 33 (11): [20] F. Ramsteiner, R. Theysohn, Tensile and impact strengths of unidirectional, short fibrereinforced thermoplastics, Composites, 1979, 10 (2): [21] S. C. Tjong, S. A. Xu, Y. W. Mai, Impact fracture toughness of short glass fiber-reinforced polyamide 6,6 hybrid composites containing elastomer particles using essential work of fracture concept, Materials Science and Engineering: A, 2003, 347 (1 2):

312 [22] J. M. Lee, Z. A. Mohd Ishak, R. Mat Taib, T. T. Law, M. Z. Ahmad Thirmizir, Mechanical, thermal and water absorption properties of kenaf-fiber-based polypropylene and poly(butylene succinate) composites, Journal of Polymers and the Environment, 2013, 21 (1): [23] Y.-H. Feng, D.-W. Zhang, J.-P. Qu, H.-Z. He, B.-P. Xu, Rheological properties of sisal fiber/poly(butylene succinate) composites, Polymer Testing, 2011, 30 (1): [24] A. Hamma, M. Kaci, Z. A. Mohd Ishak, A. Pegoretti, Starch-grafted-polypropylene/kenaf fibres composites. Part 1: Mechanical performances and viscoelastic behaviour, Composites Part A: Applied Science and Manufacturing, 2014, 56 (0): [25] W. Liu, L. T. Drzal, A. K. Mohanty, M. Misra, Influence of processing methods and fiber length on physical properties of kenaf fiber reinforced soy based biocomposites, Composites Part B: Engineering, 2007, 38 (3): [26] A. N. Oumer, O. Mamat, A Review of Effects of Molding Methods, Mold Thickness and Other Processing Parameters on Fiber Orientation in Polymer Composites, Asian Journal of Scientific Research, 2013, 6 (3): [27] B.-D. Park, J. J. Balatinecz, Mechanical properties of wood-fiber/toughened isotactic polypropylene composites, Polymer Composites, 1997, 18 (1): [28] V. N. Hristov, R. Lach, W. Grellmann, Impact fracture behavior of modified polypropylene/wood fiber composites, Polymer Testing, 2004, 23 (5): [29] D. N. Saheb, J. P. Jog, Natural fiber polymer composites: A review, Advances in Polymer Technology, 1999, 18 (4): [30] V. Alvarez, A. Vazquez, C. Bernal, Effect of microstructure on the tensile and fracture properties of sisal fiber/starch-based composites, Journal of composite materials, 2006, 40 (1):

313 [31] K. Senthil Kumar, I. Siva, P. Jeyaraj, J. T. Winowlin Jappes, S. C. Amico, N. Rajini, Synergy of fiber length and content on free vibration and damping behavior of natural fiber reinforced polyester composite beams, Materials and Design, 2014, 56 (0): [32] K. Palanikumar, J. Paulo Davim, Mathematical model to predict tool wear on the machining of glass fibre reinforced plastic composites, Materials and Design, 2007, 28 (7):

314 Chapter 9: Hydrolytic Degradation of Biodegradable Polyesters under Simulated Environmental Conditions* *A version of this chapter has been published in: R. Muthuraj, M. Misra, A. K. Mohanty, Hydrolytic degradation of biodegradable polyesters under simulated environmental conditions, Journal of Applied Polymer Science, 2015, 132, (adapted with kind permission from John Wiley and Sons, Jul 09, 2015, License number ). Abstract In this study, the durability of poly(butylene succinate) (PBS), poly(butylene adipate-coterephthalate) (PBAT) and PBS/PBAT blend was assessed by exposure to 50 o C and 90% relative humidity for a duration of up to 30 days. Due to the easy hydrolysis of esters, the mechanical properties of PBS and PBAT are significantly affected with increasing conditioning time. The PBS, PBAT and PBS/PBAT showed an increase in modulus as well as a decrease in tensile strength and elongation at break with increased exposure time. Furthermore, the impact strength of PBAT remains unaffected up to 30 days of exposure. However, it was clearly observed that the fracture mode of PBS/PBAT changed from ductile to brittle after being exposed to high heat and humid conditions. This may be attributed to the hydrolysis products of PBS accelerating the degradation of PBAT in the PBS/PBAT blend. The differential scanning calorimetry results suggested that the crystallinity of the samples increased after being exposed to elevated temperature and humidity. This phenomenon was attributed to the induced crystallization from low molecular weight polymer chains that occurred during hydrolysis. Therefore, low molecular weight polymer chains are often favored to the crystallinity enhancement. The increase in crystallinity eventually increased the modulus of the conditioned samples. The enhanced crystallinity was further confirmed by polarizing optical microscopy 280

315 (POM) analysis. Moreover, the hydrolysis of the polyesters was evaluated by scanning electron microscopy (SEM), rheology, and Fourier transform infrared (FTIR) spectroscopy analysis. 9.1 Introduction During the past decade, biodegradable polymers and their blends have gained great attention in wide range of applications due to their low environmental footprint. Among the biodegradable polymers, poly(butylene succinate) (PBS) is a promising aliphatic polyester, made from fossil fuel based 1,4-butanediol and succinic acid precursors, which can also be derived from biobased succinic acid. PBS has many desirable properties including good toughness and melt processability. The mechanical properties of the PBS fall between polyolefins with a wide processing window [1, 2]. In addition, the mechanical and thermal properties of the PBS depend on the degree of crystallinity and the spherulite morphology [3]. Degradability of the PBS has been widely studied under different environmental conditions [4-8].These studies claimed that the PBS is susceptible to hydrolysis in the presence of moisture/water. The main route of hydrolytic degradation occurs through cleavage of ester linkages and leads to lower molecular weight compounds. Solely aromatic polyesters are resistant to biological degradation. Therefore, an attempt has been made to introduce aliphatic moieties into aromatic polyesters in order to enhance the hydrolytic degradation [9]. For example, poly(butylene adipate-co-terephthalate) (PBAT) is a commercialized biodegradable aliphatic-aromatic copolyester [10]. The PBAT exhibits good thermal and mechanical properties with a terephthalic acid concentration above 35 mol% [11]. At the same time, PBAT possesses good biodegradability with an aromatic moiety concentration below 55 mol%. The properties of PBAT can be compared to that of low-density polyethylene with regards to its tensile properties. Nowadays, PBS and PBAT are widely used for many 281

316 applications because of their inherent properties in addition to biodegradability. The only shortcomings of PBS are its insufficient impact strength and gas barrier properties for certain applications. This could be overcome by physical blending with a highly flexible PBAT while maintaining biodegradability. The application of the polymeric materials depends on their durability and performance under different environments. The durability of the polymers and composites is strongly related to the degradation mechanism. The degradation mechanism is a key factor for the lifetime prediction of polymeric materials [12, 13]. If the polymeric materials maintain their required mechanical performance at least 60 weeks at elevated temperature (50 o C) and humidity (90%) it may be used for 10 year durable applications [14]. The biodegradable polymers are sensitive to hydrolysis under high temperature and humidity and thus limit their durability as well as longterm performance under these conditions. In order to incorporate more widespread semicrystalline biodegradable polymers in durable applications, including automotives and electronics, the the performance of the polymers must be maintained throughout their lifetime. It is well known that the amorphous regions are more susceptible to degradation than crystalline regions in a semicrystalline polymer. This can be explained by the rate of moisture penetration being higher in the amorphous regions than in the crystalline regions [15]. These drawbacks could be overcome by blending or alloying polymers while tailoring the material s overall performance and cost [14]. With this regard, we have extensively studied the PBS/PBAT blend in our previous research [16]. However, it is very important to understand the durability behaviors of the PBS, PBAT and PBS/PBAT blend in order to diversify as well as in predicting their applications. Such understanding will help to find out new areas in improving the required durability of these polymeric systems in different applications. 282

317 Only limited research works have been reported on the long-term durability behaviours of biodegradable polymers under simulated environmental conditions [13,14,17-19]. For instance, the long-term durability of polylactide (PLA) samples has been studied by few researchers [13, 14, 20]. These studies showed that the mechanical performance of the PLA was significantly affected after exposure to elevated temperature and moisture levels. Therefore, PLA is still an underperforming biopolymer for long-term durable applications such as automotive parts. In addition, Harris and Lee [13] have studied the hydrolytic degradation of PLA and a PLA/polycarbonate (PC) blend exposed to elevated temperature and humidity for 28 days. They have noticed a reduction in the mechanical performance of PLA and PLA/PC blend with increasing conditioning time. Therefore, the author concludes that the PLA accelerated the degradation of PC in the PLA/PC blend under these conditions. However, PLA/PC blend exhibits superior flexural strength than neat PLA during the entire conditioning period. Another study by Kim and Kim [17] showed that polypropylene (PP) has a more hydrolytic resistant behaviour than biodegradable polymers (PBS, PBAT and PLA) because of its inherent nonbiodegradability character. To the best of our knowledge, there have not been many studies available in literature on the durability of PBS, PBAT and their blends at elevated temperature and humidity. Considering the above, in the present study, our attention was to investigating the durability of PBS, PBAT and PBS/PBAT blend at an elevated temperature and humidity level. In this sense, the present study was aimed to investigate the effect of mechanical and physico-mechanical properties of PBS, PBAT and PBS/PBAT blend at 50 o C with 90% relative humidity for duration of up to 30 days. The samples were evaluated before conditioning and after 6, 12, 24 and 30 days of 283

318 continuous conditioning. The hydrolytic degradation of the polyesters was examined by using various analytical techniques. 9.2 Materials and Methodology Materials For this study, commercially available PBAT pellets were supplied by Xinfu Pharmaceutical Co., Ltd, China, under the trade name of Biocosafe 2003F with a melting point of 117 o C. PBS pellets were supplied by the same company under the trade name of Biocosafe 1903F with a melting point of 115 o C. PP-1350N homopolymer was procured from Pinnacle Polymers (Garyville, LA). According to manufacturer information, the density and melt flow index of the PP-1350N are 0.9 g/cm 3 and 55 g/10min, respectively. Neat PBS and PBAT were dried in an oven for six h at 80 o C to remove the moisture prior to melt processing Sample preparation and conditioning Neat PP, PBS, PBAT and blend of PBS/PBAT (60/40 wt%) were extruded in a Leistritz extruder with a screw speed of 100 rpm. The extruder was equipped with co-rotating twin-screws with a screw diameter of 27 mm and a L/D ratio of 48. Prior to the injection molding, the extrudates were pelletized and dried in an oven at 80 o C for 12 h. The dried extruded pellets were injection molded in an ARBURG allrounder 370C (Model No: 370 S /70, Germany) injection molding machine to obtain desired test specimens. The injection-moulding machine had a maximum injection pressure of 2000 bar and a screw diameter of 35 mm. The extrusion and injection molding process was carried out with a processing temperature of 140 o C for PBS, PBAT and PBS/PBAT and 180 o C for PP. In the literature, the durability of polymers, polymer blends and their composites was studied at different accelerated environmental conditions, [21-23] in vehicle and in-field 284

319 conditions [13,14]. Furthermore, long-term durability of the polymeric material has been studied in the presence of Xenon light, UV light, metal halide and carbon arc lamps by many researchers [24, 25]. However, in order to model the PBS, PBAT and PBS/PBAT blend for automotive interior applications; all the moulded samples were conditioned under simulated temperature (50 o C) and relative humidity (90%) [18]. These conditions were simulated using an environmental chamber, Envirotronics Endurance C340. The samples were tested initially before and after 6, 12, 24, and 30 days continuous conditioning at 50 o C and 90% relative humidity (RH). Except moisture absorption analysis, all other characterizations were performed after drying the test samples at 80 o C for 24 h in order to avoid plasticization effect of excess moisture in the specimens Moisture absorption Before performing moisture absorption test, all the samples were dried at 80 o C till a constant weight is reached. The moisture absorption of the samples was calculated by taking out the samples at required time interval for the set environmental exposure conditions (50 o C and 90% RH). The percentage of moisture uptake was calculated by using the equation: Moisture uptake (%) = x 100 (9.1) where W a and W b are weight of the samples after and before moisture exposure. The reported moisture absorption values are an average of three samples Fourier transform infrared spectroscopy (FTIR) FTIR analysis was performed in a Thermo Scientific Nicolet TM 6700 at room temperature with a Smart Orbit attachment. FTIR spectrum was recorded in the range of cm -1 with a resolution of 4 cm -1 and averaged over 36 readings. 285

320 9.2.5 Mechanical properties Tensile and flexural tests were performed in an Instron Universal Testing Machine (Model 3382) according to ASTM D638 and D790, respectively. The crosshead movement speeds of 14 mm/min for flexural test and 50 mm/min for tensile test were used as recommended by the respective standards. The tensile tests were performed until the conditioned samples broke at the grip region as a consequence of embrittlement. Therefore, the experiment was conducted only up to 30 days. Notched Izod impact strength was assessed with an impact test machine from TMI 43-02, USA, complying with ASTM D256.The results are reported as an average of five samples for each formulation Differential scanning calorimetry (DSC) The DSC analysis was performed in a TA-Q200 instrument with a heating and cooling rate of 10 and 5 o C/min, respectively. The samples were heated under a nitrogen flow rate of 50 ml/min. The melting enthalpy was calculated by measuring area under the curves using TA analysis software. The first heating cycle was considered in order to measure sample crystallinity before and after conditioning. The percentage crystallinity of the PBS and PBAT was calculated by using the following formula: % Crystallinity (χ c ) = x 100% (9.2) where H m100 is the theoretical enthalpy of melting for 100% crystalline PBS (110.3 J/g) [8] and PBAT (114 J/g) [26]. H m is the measured enthalpy of melting. The PBS cystallinity in the PBS/PBAT blend was calculated as follows: χ c = x 100% (9.3) 286

321 where w f is the weight fraction of the PBAT in the PBS/PBAT blend Dynamic mechanical analysis (DMA) DMA analysis was performed using TA Instrument (DMA Q800), USA. The experiments were conducted from -60 to 100 o C. The selected temperature range was based on the glass transition temperature and melting temperature of the samples. The scans were performed at a constant rate of heating (3 o Cmin -1 ) with oscillating amplitude of 15 µm and a frequency of 1 Hz in a dual cantilever clamp mode Rheological properties Rheological properties were obtained in an Anton Paar Rheometer MCR302. The experiments were carried out in parallel plates with a gap of 1 mm and a diameter of 25 mm. In order to avoid degradation of the samples during the experiments, all the samples were vacuum dried at 80 o C for 4 h before performing the experiments. The shear viscosity values of the samples both before and after conditioning were measured at 140 o C from 300 to 0.01 rad/s Polarizing optical microscopy (POM) Spherulite morphology of the samples was observed by using optical polarizing microscope (Nikon Eclipse LV100) equipped with a Linkam LTS 420 hot stage. Thin films of the samples were made by heating the sample between two transparent microscope glass slides. All the samples were heated to 150⁰C for 60 s followed by the samples being quickly transferred to 90⁰C in the microscope hot stage. Subsequently, the samples were kept at close to crystallization temperature (90 o C) and the spherulite growth was recorded using a Nikon camera Morphological analysis The specimens were prepared by sputtering gold particles in order to avoid electrical charging during examination. A scanning electron microscope, Inspect S 50, FEI Netherlands, 287

322 was utilized to examine the fracture surface morphology of the specimens. The surface morphology of the specimens was examined at an accelerating voltage of 20 kv. 9.3 Results and Discussion Moisture absorption Moisture absorption of all the samples was investigated as a function of exposure time. Figure 9.1 shows the moisture absorption curves in percent of the PP, PBS, PBAT and PBS/PBAT blend up to 34 days. Generally, more or less; all the polymers tend to absorb moisture in a humid atmosphere. Usually, polymers with strong polar functionality such as carbonyl (>C=O) groups and amine groups are able to absorb moisture by hydrogen bonds [27]. Therefore, it is expected that the polyesters can absorb more moisture than the relatively nonpolar polymers such as PP. It can be seen that the PP absorbed a very small amount (0.011±0.004%) of moisture and the moisture absorption curve has reached a typical Fickian behavior. It has been reported that the PP is resistant to moisture absorption even at elevated temperatures [28]. On the contrary, moisture absorption of PBS, PBAT and their blend was monotonically increased with increasing exposure time up to 34 days. After 34 days exposure, the PBS showed a maximum moisture absorption (1.11±0.002%) followed by PBS/PBAT (1.05±0.01%) and PBAT (0.99±0.003%). The observed moisture absorption difference between the PBS and PBAT may be due to polarity differences between the polymers [29]. Due to the moisture absorption, it can be expected that the PBS and PBAT can undergo hydrolytic degradation at elevated temperature and humidity. Normally, higher moisture absorption of polyesters causes undesirable losses in mechanical performances [13, 30]. 288

323 Figure 9.1. Moisture absorption curves as a function of conditioning time Hydrolytic degradation mechanism of PBS and PBAT It is known that the ester linkages of PBS and PBAT are more sensitive to elevated temperature and moisture [17, 19]. Therefore, in the presence of moisture, the PBS and PBAT primarily can undergo hydrolytic degradation through cleavage of ester linkages on the polymer backbone. In addition, the hydrolysis reaction may occur in the form of depolymerization process and random chain scission mechanism [30]. The possible hydrolytic degradation of PBS and PBAT under elevated humidity and temperature is depicted in Figure 9.2 and 9.3, respectively. 289

324 Figure 9.2. Hydrolysis reaction of PBS The chain scission is frequently terminated by carboxylic acid end groups [13,30,31] and hydroxyl end groups [32]. A similar type of hydrolytic degradation mechanism was proposed for PLA [13], PBS [8, 33] and poly(ethylene terephthalate) (PET) [30]. When PBS is exposed to high temperature and humidity environment, the surrounding moisture can interact with ester functionality of PBS and thus can create the low molecular weight PBS through hydrolytic degradation mechanism [33]. Figure 9.3. Hydrolysis reaction of PBAT 290

325 The hydrolytic degradation of PBS, PBAT and their blend was further confirmed by FTIR analysis. Figure 9.4 shows FTIR spectrum of PBS, PBAT and PBS/PBAT before and after 30 days of being exposed to elevated humidity and temperature. In PBS, the band at 917 cm -1 was due to the C-O-C- groups in the ester linkage of PBS [8]. The band at 1325 cm -1 resulted from the asymmetric stretching of the -CH 2 - group in the PBS backbone. The peak at 1045 cm -1 was attributed to the -O-C-C- stretching vibration and the peak in the range 1151 cm -1. The band at 1712 cm -1 resulted from the C=O stretching vibration of the ester group in PBS [17]. After 30 days hydrolysis of PBS, a remarkable decrease of C-O-C- and C=O absorption intensity was observed. These reductions in absorption intensity were due to lowering of the molecular weight and deterioration of the chemical structure by hydrolysis after being exposed to moisture and heat [8, 18, 34]. The characteristic peaks of the PBAT can be described as follows: a sharp peak at 1710 cm -1 represents the C=O functionality of the ester linkage; the band at around 1267 cm -1 assigned to the C-O group in the ester linkage; the peak at 727 cm -1 resulted from four or more adjacent -CH 2 - groups in the PBAT backbone. The peaks in the range of cm -1 were attributed to benzene substitutes [35]. After 30 days of exposure to moisture and heat, there is no significant change observed in the FTIR spectra of PBAT. This is possibly due to the partial aromatic structure of PBAT. On the contrary, the FTIR spectra of PBS/PBAT showed a remarkable decrease in the characteristic peak intensity. This phenomenon may be attributed to the hydrolysis product of PBS accelerating the degradation of PBAT in the PBS/PBAT blend [31]. 291

326 Figure 9.4. FTIR spectra of PBS, PBAT and PBS/PBAT before and after 30 days exposed to 50 o C with 90% relative humidity Changes in mechanical properties Mechanical properties are the main indicators in order to evaluate the durability of the polymeric materials. The influence of moisture and heat on the mechanical properties was measured by tensile and flexural properties as well as impact strength. The mechanical properties of neat PBS, PBAT, PBS/PBAT blend and PP are provided in our previous study [19]. Figure 9.5 shows the tensile strength of PBS, PBAT, PBS/PBAT and PP before and after exposure at 50 C with 90% RH up to 30 days. In general, the mechanical properties of semicrystalline polymers are dependent on their molecular weight, crystal size and percentage of crystallinity [36]. The tensile strength of PBS and PP showed slight enhancement after 6 days of exposure to 50 o C and 90% RH. This can be attributed to the post crystallization of the samples after being exposed to elevated humidity and temperature. A similar result has been found for PLA [13], 292

327 poly(hydroxybutyrate-co-valerate) (PHBV) [37], and homo polypropylene [38] specimens when exposed to different environmental conditions. However, after 6 days of exposure; PBAT as well as PBS/PBAT blend showed a slight reduction in tensile strength. This could be due to the plasticization effect of hydrolytically degraded amorphous region in the PBAT. It can be observed that the tensile strength of PBS, PBAT and PBS/PBAT blend decreased significantly with increasing hydrolysis time. For example, after 12 days exposure time, the tensile strength of PBS, PBAT, and PBS/PBAT blend was reduced by 40, 39 and 11%, respectively. The reduced tensile strength may be attributed to the combined effect of hydrolytic degradation and molecular weight reduction after being exposed to the raised humidity and temperature [33]. Generally, the hydrolytic degradation of the biodegradable polymers is higher in the amorphous regions than crystalline regions under high humidity [39]. A similar type of observation has been made in PBS, PBAT, PBS/PBAT and PP after being exposure to 18 days of elevated humidity and heat [19]. However, after 30 days of conditioning, the tensile strength of the PBS and PBS/PBAT blend exhibited extreme degradation in contrast to PBAT. This is possibly due to the accelerated degradation of PBS with the increased time at elevated temperature and humidity [15]. Our finding had good agreement with the recent study by Kim and Kim [17]. Usually, the hydrolytic degradation and biodegradability of the polymers mainly depend on the easily hydrolysable ester functionality in the polymer backbone. In the present study, PP did not show any significant reduction in the tensile strength up to 30 days conditioning, which is due to the non-polar as well as its hydrophobicity type of characteristic. 293

328 Figure 9.5. Tensile strength of PP, PBS, PBAT and PBS/PBAT as a function of exposure time at 50 o C with 90% relative humidity Figure 9.6 demonstrates the flexural strength of the PBS, PBAT, PBS/PBAT and PP after and before exposure to elevated temperature and humidity. After 6 days of conditioning, the PBAT did not show any significant improvement in the flexural strength, which may be due to PBAT possessing a high entanglement density. Interestingly, the flexural strength of PBS, PBS/PBAT blend and PP were increased 13, 15 and 15% respectively with increasing exposure time up to 6 days. After 18 days of continuous conditioning at 50 o C with 90% RH, the flexural strength of PBS, PBS/PBAT and PP samples was found to increase slightly [19]. The increased flexural strength is probably due to the increased crystallinity of the samples after being exposed to elevated temperature [40]. 294

329 Figure 9.6. Flexural strength of PP, PBS, PBAT and PBS/PBAT as a function of exposure time at 50 o C with 90% relative humidity However, it is important to note that the PBS and PBS/PBAT blend samples became more brittle after 30 days conditioning and leading to premature failure during flexural test, as shown in Figure 9.7. Harris and Lee [13] found that the PLA and PLA/PC blend underwent severe flexural strength reduction because of the hydrolytic degradation under the exposed elevated temperature (70 o C) and humidity (90% RH). On the other hand, they have noticed that the PC/ABS blend did not show any significant changes in the flexural strength up to 30 days conditioning because of the resistance to the hydrolysis. 295

330 Figure 9.7. Testing failure mode of PBS, PBAT, PBS/PBAT and PP after 30 days exposed to 50 o C with 90% relative humidity Figure 9.8 represents the elongation at break of the samples with respect to the exposure time. Except PBAT, all the samples showed drastic reduction in the elongation at break from early exposure time. The PP, PBS and PBS/PBAT blend showed a drastic decrease in the percent elongation after 6 days conditioning. Therefore, it is clear that the toughness is more sensitive than the strength after being exposed to raised humidity and temperature. A similar trend has been reported in the literature for PP [41], high-density polyethylene (HDPE) [42], and PHBV [37]. After conditioning, PBS showed lower elongation than PP during the entire exposed time. This implies that the PBS is more moisture sensitive than the PBAT and the PP. Toughness of the polymer is mainly dependent on the tie molecules and entanglement of the polymer chains [42-44]. When, the entanglement density decreased in the polymers it led to a reduction in the toughness of the resultant materials. Apparently, PBAT is more ductile and less crystalline than PBS due to the higher degree of chain entanglements. Therefore, PBAT maintains its elongation 296

331 up to 12 days conditioning even after extensive chain scission occurred. Interestingly, the PBS/PBAT blend has higher elongation than PBS and PP up to 12 days due to the PBAT chain entanglements. After 12 days of conditioning, the PBS/PBAT blend experienced severe loss in elongation because of heavy chain scission of PBS leading to hydrolysis of the PBAT phase in the blend system [31]. Figure 9.8. Percentage elongation of PP, PBS, PBAT and PBS/PBAT as a function of exposure time at 50 o C with 90% relative humidity Figure 9.9 and 9.10 shows the tensile and flexural modulus of the PP, PBS, PBAT and PBS/PBAT as a function of conditioning time. Both tensile and flexural modulus of the PP, PBS, and PBS/PBAT gradually increased with increasing conditioning time, whereas PBAT remains constant throughout the entire conditioning period. More specifically, the tensile and flexural modulus of PP and PBS were improved by around 200 MPa after 30 days conditioning period. This could be related to the increased crystallinity and subsequently increase in modulus [45, 46]. 297

332 Figure 9.9. Tensile modulus of PP, PBS, PBAT and PBS/PBAT as a function of exposure time at 50 o C with 90% relative humidity A number of researchers have observed a similar tendency in the modulus after exposure to different weathering conditions [12, 37, 42]. These studies were concluded that the modulus improvement of the conditioned samples is associated with structural relaxation, increase in crystallinity, crystal perfection and increase of lamella thickness. In addition, the brittleness of the PBS and PBS/PBAT blend sample was also improved with increasing conditioning time up to 30 days, accounting for the reduction in impact toughness. 298

333 Figure Flexural modulus of PP, PBS, PBAT and PBS/PBAT as a function of exposure time at 50 o C with 90% relative humidity Among the mechanical properties, impact energy is more sensitive to the environmental exposure. Table 9.1 shows the impact strength of PBS, PBAT, PBS/PBAT and PP after and before conditioning at 50 o C and 90% RH. Before conditioning, the PBAT showed non-break impact strength of 211 J/m while PBS and PP showed complete break with impact strength of 25 and 30 J/m, respectively. The impact energy of the PBS and PP decreased for the first 6 days of conditioning. This reduction is probably due to the inadequate degree of entanglement between amorphous and crystalline phase after exposed to 50 o C and 90% RH [45]. The notched Izod impact strength of PBS, PBAT, PBS/PBAT blend, and PP samples after conditioning for 18 days is explained elsewhere [19]. With increased exposure time (from 6 to 30 days), the impact energies of both PBS and PP were not significantly affected. In contrast, the impact energy of the PBAT remains unchanged up to 30 days of conditioning at elevated temperature and humidity. This may be due to the PBAT having sufficient molecular weight to form a significant degree of 299

334 entanglement up to 30 days of hydrolysis environments [14, 31]. Furthermore, the impact strength of PBS/PBAT changed from a ductile to brittle fracture with increasing conditioning time, as shown in Figure 9.7. This could be due to the accelerated degradation of PBS with the increased exposure time. Moreover, except PBAT, all the samples exhibit brittle failure with increasing conditioning time. This observation corroborated with the modulus improvements. Table 9.1.Notched Izod impact strength (J/m) of the samples before and after conditioned at 50 o C with 90% relative humidity Samples Before conditioning PBS ± 6.55 PBAT Non-break ( ± 10.37) PBS/PBAT Non-break ( ± 43.95) PP ± 7.08 After 6 days conditioning ± 6.20 Non-break ( ± 46.54) ± ± 0.79 After 12 days conditioning ± 1.91 Non-break ( ± 27.01) ± ± 2.33 After 24 days conditioning ± Non-break ( ± 25.22) ± ± 0.79 After 30 days conditioning ± 1.07 Non-break ( ± 35.47) ± ± Differential scanning calorimetry After exposing the polymers to raised humidity and temperature, it is expected that the spherulitic growth rates, lamellar thickness, and crystal interphase be modified due to the free energy changes in the crystals formation. DSC traces of the samples before and after conditioning are shown in Figure 9.11 and The thermal properties of PBS, PBAT and PBS/PBAT blends are summarized in Table 9.2. Melting enthalpy ( H m ) of the sample was 300

335 calculated by measuring the area under the melting peak while crystallization enthalpy ( H c ) was the measured area under the crystallization peak. Figure DSC heating cycles for PBS, PBAT and PBS/PBAT before and after exposed to 50 o C with 90% relative humidity for 30 days Before conditioning, all the samples showed a single melting temperature (T m ). In the heating cycles (Figure 9.11), a small exothermic peak was also observed for PBS and PBS/PBAT prior to melting peak. This resulted from the melt-recrystallization of PBS while heating [47]. However, after 30 days exposure, PBS and PBS/PBAT samples displayed a bimodal melting peak, as shown in Figure The hydrolytic degradation of the polymers leads to wide range of molecular weight distribution. The shorter polymer chains are having tendency to form less thick crystal lamella than the high molecular weight polymer chains. Due to the difference in the lamella thickness, the hydrolytic degraded samples were melting at two different temperatures. These observed double endothermic peaks are attributed to the different crystal lamella thickness formation [8]. In addition, the T m of PBAT shifted to low temperature after 30 days conditioning. 301

336 Either the change in amorphous-crystal surface energy or a decreased in the lamellar thickness was responsible for the T m decrease of a polyester after exposure to elevated temperature and humidity [30]. For both PBS and PBS/PBAT, no change was observed in the melting temperature (~115 o C) after 30 days of exposure time. Figure DSC cooling curves for PBS, PBAT and PBS/PBAT before and after exposed to 50 o C with 90% relative humidity for 30 days In a semi-crystalline polymer, initially the amorphous regions are more susceptible for hydrolysis [6]. From the DSC analysis, it was clearly observed that the H m and H c of the PBS, PBAT and PBS/PBAT increased after 30 days of conditioning, indicating that degradation mainly occurred in the amorphous regions. In addition, this phenomenon may be due to induced crystallization from low molecular weight polymer chains that occurs during conditioning [13]. Therefore, low molecular weight polymer chains are often favored to the crystallinity enhancement. Our findings have good agreement with previous studies [14]. According to these studies, the chain scission leads to reduced entanglement density and tie molecules of the semi 302

337 crystalline polymers. The small molecular chains have potential to rearrange into the crystalline region, which is called chemi-crystallization. This behavior has been observed in most of the semicrystalline aliphatic biodegradable polymers including PBS [46]. The increased crystallinity (χ c ) further accounts for the enhanced modulus as well as stiffness. The crystallization temperature (T c ) of the PBS and PBS/PBAT significantly reduced after 30 days of conditioning. This is attributed to the low molecular weight polymer chains leading to slow crystallization. A similar crystallization behavior for PBS has been reported after exposure to raised humidity and temperature [8]. Contrary, the crystallization temperature of PBAT was shifted to higher temperature. This is probably due to the nucleation effect which is caused by oligomers [48]. Table 9.2. DSC results for PBS, PBAT and their blend before and after 30 days conditioned at 50 o C with 90% relative humidity Samples T m ( o C) H m (J/g) T c ( o C) χ c (%) T g * ( o C) PBS before PBS after PBAT before PBAT after PBS/PBAT before PBS/PBAT after *T g obtained from tan δ peaks Dynamic mechanical analysis Figure 9.13 shows the temperature dependence dynamic modulus of PBS, PBAT, and their blend. It can be seen that the PBS had higher storage modulus than PBAT and PBS/PBAT. Similar occurrence has been observed in the tensile and flexural modulus. However, the storage 303

338 modulus of all the samples gradually decreased with increasing temperature. This is attributed to the enhanced polymer chain mobility with increasing temperatures [16]. As reported by Van der wal et al., [49] above the glass transition temperature, storage modulus is dependent to the degree of crystallinity. Below glass transition temperature, the modulus of crystalline as well as amorphous phase is almost identical. Interestingly, after 30 days of conditioning, the storage modulus of PBS, PBAT and their blend samples was found to increase slightly. This is because the samples become stiffer, as evidenced by the increase in crystallinity after conditioning at elevated temperature and humidity. Figure Storage modulus of PBS, PBAT and PBS/PBAT before and after exposure to 50 o C with 90% relative humidity for 30 days Figure 9.14 shows the tan δ (loss factor) curves with respect to temperatures. In fact, the peak temperature of the tan δ represents the glass transition temperature (T g ). The T g values of PBS, PBAT and PBS/PBAT blends are summarised in Table 9.2. The position of each tan δ peak 304

339 is affected slightly after conditioning at 50 o C with 90% RH. In general, T g value of the amorphous phase in semicrystalline polymers depends on the degree of crystallinity [46]. Initially, the PBS and PBAT had T g values of -17 and -20 o C, respectively. In PBS/PBAT blend, a single T g (-19 o C) was observed. This is due to the fact that T g values of both neat PBS and PBAT were very close to each other and thus T g may be overlapping in the PBS/PBAT blend [16]. After 30 days conditioning, T g of the PBS increased from to o C. This slight change can be attributed to the enhanced crystallinity, as corroborated by DSC result. Figure Loss factor peak (tan δ) of PBS, PBAT and PBS/PBAT before and after 30 days exposed to 50 o C with a relative humidity of 90% As reported in Table 9.2, the crystallinity of the PBS increased from to 73.43% after 30 days exposed to 50 o C with 90% RH. Similar observations have been reported by Harris and Lee for PLA [14]. However, after 30 days conditioning, the T g values of the PBAT and PBS/PBAT reduced marginally with slightly increased in crystallinity. This can be related to the plasticization effect by the diffused moisture, which induces an increase in the amorphous chain 305

340 mobility [30]. A similar type of negative T g shift was observed in the PLA films [20] and poly(ethylene terephthalate) (PET) composites [30] after exposure to elevated temperature and humidity Rheological properties Figure 9.15 represents the shear viscosity of the samples before and after 6 days of conditioning at the elevated temperature and humidity. It was observed that all the samples showed Newtonian and non-newtonian flow behavior at lower and higher frequencies, respectively. The 6 days conditioned samples exhibit a slight decrease in the shear viscosity compared to the before conditioned samples. As expected, this behavior should be due to the molecular weight reduction by random chain scission after being exposed to elevated temperature and humidity. The molecular weight changes can be correlated with shear viscosity of the sample at low shear rate. According to the literature [50], the weight average molecular weight (M) is directly proportional to the viscosity of the polymer melt at a zero shear rate. However, molecular weight distribution is independent to zero shear viscosity (ƞ o ). Generally, the ƞ o is obtained from extrapolation of the shear viscosity at lower shear rate (Newtonian region), which considered as weight average molecular weight [51]. This relationship can be explained as follows[50]: ƞ o = KM 3.4 (9.4) where K and M are the material constant and molecular weight respectively. In the present study, relative molecular weight (M 1 /M 2 ) of the samples before and after conditioning can be calculated by using following equation: log ( ) = 3.4log ( ) (9.5) 306

341 where ƞ 1 and ƞ 2 are the Zero shear viscosity of the samples before and after conditioning. The molecular weight reduction is permanent damage caused by hydrolysis of the ester functionalities on the polyesters backbone. Phua and coworkers [8] have studied the molecular weight of hydrolytically degraded PBS samples. The authors found that the molecular weight reduction was higher with an increasing conditioning period. Table 9.3 reports the zero shear viscosity, viscosity ratio and relative molecular weight (calculated from equation 9.5) of the samples before and after 6 days of conditioning. After 6 days conditioning, a significant reduction in molecular weight and viscosity were observed for all the samples. As mentioned before, PBS and PBAT are susceptible to the moisture. Therefore, it can be expected that the moisture can easily hydrolyze the PBS and PBAT at 90% RH and it leads to a decrease in the molecular weight as well as viscosity [13]. The molecular weight of the PBS/PBAT blend was 1.70 times lower after being subjected to hydrolytic degradation. This is relatively high compared to PBS and PBAT. This may be due to the hydrolysis product of PBS or PBAT accelerating the molecular weight reduction of the PBS/PBAT blend. After 6 days of exposure to heat (50 o C) and humidity (90%), the molecular weight reduction occurred in the following order PBS/PBAT>PBAT>PBS, as shown in Table 9.3. This result agrees with the observed mechanical properties of the conditioned samples as studied. 307

342 Figure Shear viscosity curves for PBS, PBAT and PBS/PBAT before and after 6 days exposed to 50 o C with a relative humidity of 90% Table Relative molecular weight (M 1 /M 2 ) of the PBS, PBAT and PBS/PBAT blend before and after 6 days conditioned at 50 o C with 90% relative humidity Samples Zero shear viscosity (Pa.s) Viscosity ratio Before (ƞ 1 ) After (ƞ 2 ) (ƞ 1 /ƞ 2 ) Relative molecular weight (M 1 /M 2 ) PBS PBAT PBS/PBAT Polarizing optical microscopy Figure 9.16 shows the spherulite morphology of PBS, PBAT and PBS/PBAT before and after 30 days of exposure to raised humidity and temperature. The spherulite morphology of the samples was analyzed at close to crystallization temperature (90 o C). Before being exposed to hydrolysis conditions, it was difficult to notice clear spherulite morphology at 90 o C for all the samples. However, after 30 days the conditioned samples exhibited an obvious spherulite 308

343 structure. It is generally agreed that the amorphous region is more susceptible for hydrolysis than crystalline regions in semicrystalline polymers. Therefore, these findings have good agreement with the improved percentage of crystallinity, which was observed by DSC. This type of phenomenon is commonly found in the polymers when exposed to a degradation environment [52, 53]. Interestingly, the amount of spherulite formation was higher in the samples with a lower percentage of crystallinity. For instance, the PBS and PBS/PBAT blend showed less number of spherulites than PBAT after 30 days exposed to hydrolysis. This is attributed to the nucleation density difference in the samples. A similar trend has been observed in the degraded polypropylene sample [53]. In addition, the PBS had a less amount of nucleation sites than PBAT because of the severe molecular weight reduction by hydrolytic degradation. This phenomenon was consistent with observed crystallization temperature by the DSC analysis. Furthermore, the spherulite morphology of the 30 days exposed PBS and PBS/PBAT exhibits clear ring-banded spherulites, which can be attributed to their reduced molecular weight. This finding has good agreement with a previous study [52]. According to Kfoury et al., [54] the percentage of crystallinity, size of crystallites and spherulite morphology have great influence on the impact strength. The stress concentration ability of crystallites has been increased with an enhanced percentage of crystallinity. Consequently, this could lead to a reduction in the impact strength. In the present study, observed impact strength had good agreement with the spherulite morphology and crystallinity. 309

344 Figure Polarized optical micrographs of PBS, PBAT and PBS/PBAT before and after 30 days conditioned at 50 o C and 90% relative humidity 310

345 9.3.8 Morphological analysis To investigate the hydrolysis caused by moisture and temperature, SEM analysis was carried out before and after 30 days conditioned samples. SEM micrographs of the PBS, PBAT and their blend are depicted in Figure Before exposure to hydrolysis environment, a smooth surface morphology was observed for all the samples. On the other hand, after 30 days of hydrolysis test, the PBS, PBAT and PBS/PBAT blend showed deep holes, cavities as well as eroded regions. This observation indicates that the biodegradable polyesters (PBS, PBAT and PBS/PBAT blend) can readily undergo severe degradation after being exposed to elevated humidity and temperature. A similar type of physical damage in the hydrolytically degraded PBS and PLA samples has been observed by Kanemura et al., [33] and Deroiné et al., [55] These studies suggest that the formed irregular surface morphology is ascribed to the dissolution of the oligomers during hydrolysis process [55]. It can be seen that the SEM image (Figure 9.17) of the PBS showed significant erosion pits and large eroded regions compared to PBAT and PBS/PBAT. This is corresponding to the higher rate of hydrolytic degradation of PBS after conditioning for 30 days under the simulated environment [8]. In addition, an irregular fractured surface was observed in the 30 days conditioned PBS sample. This is possibly due to the increased crystallinity after being exposed to the hydrolysis environment. 311

346 Figure SEM micrographs of PBS, PBAT and PBS/PBAT before and after 30 days conditioned at 50 o C and 90% relative humidity 312

347 9.4. Conclusions The hydrolytic degradation of PP, PBS, PBAT and PBS/PBAT samples was examined after exposure to elevated temperature and humidity. Because of chain scission through the hydrolysis mechanism, the elongation at break and tensile strength of the PBS, PBAT and PBS/PBAT were significantly affected after conditioning. However, the flexural and tensile modulus of the PP, PBS and PBS/PBAT were slightly improved after exposure to heat and humidity. This could be due to the improved crystallinity by molecular weight reduction during the exposure time. The increased crystallinity was consistent with observed spherulite morphology. The zero shear viscosity of the 6 days exposed samples was lower compared to corresponding unexposed samples. This suggests that the molecular weight of the exposed sample is reduced because of hydrolytic degradation. Interestingly, it was found that the impact strength of the PBAT was not affected significantly over the entire exposure time, whereas for PP, PBS and PBS/PBAT impact strength decreased up to 6 days of conditioning. Over the hydrolysis time, the samples had rough surfaces and corrosive holes in the SEM micrographs. This result agrees with the considerable reduction in the mechanical properties of the samples after being exposed to elevated temperature and humidity. Our findings allow us to conclude that the hydrolytic degradation of biodegradable polyesters needs to be reduced under high humidity and temperature for diversifying their applications. 313

348 References 1. Fujimaki, T., Polym Degrad Stab, 1998, 59, Gerngross, T.; Slater, S.; Gross, R. A., Science, 2003, 299, Babu, R.P.; O Connor, K.; Seeram, R., Prog Biomater, 2013, 2, Phua, Y.J.; Lau, N. S.; Sudesh, K.; Chow, W. S.; Mohd Ishak, Z. A., Polym Degrad Stab, 2012, 97, Anstey, A.; Muniyasamy, S.; Reddy, M.M; Misra, M.; Mohanty, A.K, J Polym Environ, 2014, 22, Kim, H.-S.; Kim, H.-J.; Lee, J.-W.; Choi, I.-G., Polym Degrad Stab, 2006, 91, Tan, L.; Hu, J.; Ye, S.; Wei, J.; Chen, Y., Thermochim Acta, 2014, 575, Phua, Y. J.; Chow, W. S.; Mohd Ishak, Z. A., Polym Degrad Stab, 2011, 96, Gan, Z.; Kuwabara, K.; Yamamoto, M.; Abe, H.; Doi, Y., Polym Degrad Stab, 2004, 83, Milinkovic, A.; Miroslav, P.; Gregorova, A., Thermochimi Acta 590, Vroman, I.; Tighzert, L., Materials, 2009, 2, Deroiné, M.; Le Duigou, A.; Corre, Y.-M.; Le Gac, P.-Y.; Davies, P.; César, G.; Bruzaud, S., Polym Test, 2014, 39, Harris, A. M.; Lee, E. C., J Appl Polym Sci, 2013, 128, Harris, A. M.; Lee, E. C., J Appl Polym Sci, 2010, 115, Deng, M.; Zhou, J.; Chen, G.; Burkley, D.; Xu, Y.; Jamiolkowski, D.; Barbolt, T., Biomaterials, 2005, 26, Muthuraj, R.; Misra, M.; Mohanty, A., J Polym Environ, 2014, 22, Kim, H.-S.; Kim, H.-J., Polym Degrad Stab, 2008, 93,

349 18. Berthé, V.; Ferry, L.; Bénézet, J. C.; Bergeret, A., Polym Degrad Stab, 2010, 95, Muthuraj, R.; Misra, M.; Mohanty, A., Society of Plastic Engineering (SPE), 2015, Orlando, Florida, USA. 20. Copinet, A.; Bertrand, C.; Govindin, S.; Coma, V.; Couturier, Y., Chemosphere, 2004, 55, Saha, S. K.; Tsuji, H., Polym Degrad Stab, 2006, 91, Proikakis, C. S.; Mamouzelos, N. J.; Tarantili, P. A.; Andreopoulos, A. G., Polym Degrad Stab, 2006, 91, Li, X.; Liu, K. L.; Wang, M.; Wong, S. Y.; Tjiu, W. C.; He, C. B.; Goh, S. H.; Li, J., Acta Biomater, 2009, 5, Marek, A.; Kaprálková, L.; Schmidt, P.; Pfleger, J.; Humlíček, J.; Pospíšil, J.; Pilař, J., Polym Degrad Stab, 2006, 91, Azuma, Y.; Takeda, H.; Watanabe, S.; Nakatani, H., Polym Degrad Stab, 209, 94, Chivrac, F.; Kadlecová, Z.; Pollet, E.; Avérous, L., J Polym Environ, 2006, 14, Baschek, G.; Hartwig, G.; Zahradnik, F., Polymer, 1999, 40, Baghaei, B.; Skrifvars, M.; Rissanen, M.; Ramamoorthy, S. K., J Appl Polym Sci, 2014, 131, Ravati, S.; Favis, B. D., Polymer, 2013, 54, Foulc, M. P.; Bergeret, A.; Ferry, L.; Ienny, P.; Crespy, A., Polym Degrad Stab, 2005, 89,

350 31. Lyu. S.; Schley, J.; Loy, B.; Lind, D.; Hobot, C.; Sparer, R.; Untereker, D., Biomacromolecules, 2007, 8, Dong, W.; Zou, B.; Yan, Y.; Ma, P.; Chen, M., Int. J. Mol. Sci, 2013, 14, Kanemura, C.; Nakashima, S.; Hotta, A., Polym Degrad Stab, 2012, 97, Fanconi, B., Annu Rev Phys Chem, 1980, 31, Al-Itry, R.; Lamnawar, K.; Maazouz, A., Polym Degrad Stab, 2012, 97, Tordjeman, P.; Robert, C.; Marin, G.; Gerard, P., Eur. Phys. J. E 4, 2001, Deroiné, M.; Le Duigou, A.; Corre, Y.-M.; Le Gac, P.-Y.; Davies, P.; César, G.; Bruzaud, S., Polym Degrad Stab, 2014, 105, Fiebig, J.; Gahleitner, M.; Paulik, C.; Wolfschwenger, J., Polym Test, 1999, 18, Tsuji, H.; Echizen, Y.; Nishimura, Y., Polym Degrad Stab, 2006, 91, Harris, A. M.; Lee, E. C., J Appl Polym Sci, 2008, 107, Song, D.; Gao, J.; Li, X.; Lu, L., Polym Test, 2014, 33, Mendes, L. C.; Rufino, E. S.; de Paula, F. O. C.; Torres Jr, A. C., Polym Degrad Stab, 2003, 79, Tidjani, A., J Appl Polym Sci, 1997, 64, Lyons, J. S., Polym Test, 1998, 17, Ferrer-Balas, D.; Maspoch, M. L.; Martinez, A. B.; Santana, O. O., Polymer, 2001, 42, Pegoretti, A.; Penati, A., Polymer, 2004, 45, Qiu, Z.; Komura, M.; Ikehara, T.; Nishi, T., Polymer, 2003, 44, Xu, H.; Teng, C.; Yu, M., Polymer, 2006, 47, van der Wal, A.; Mulder, J. J.; Gaymans, R. J., Polymer, 1998, 39,

351 50. Pang, M.-M.; Pun, M.-Y.; Ishak, Z. A. M., J Appl Polym Sci, 2013, 129, Jordens, K.; Wilkes, G. L.; Janzen, J.; Rohlfing, D. C.; Welch, M. B., Polymer, 2000, 41, Sabino, M. A.; Feijoo, J. L.; Müller, A. J., Polym Degrad Stab, 2001, 73, Rabello, M. S.; White, J. R., Polymer 38, Kfoury, G.; Raquez, J.-M.; Hassouna, F.; Odent, J.; Toniazzo, V.; Ruch, D.; Dubois, P., Frontiers in Chemistry2013, Deroiné, M.; Le Duigou, A.; Corre, Y.-M.; Le Gac, P.-Y.; Davies, P.; César, G.; Bruzaud, S., Polym Degrad Stab, 2014, 108,

352 Chapter 10: Conclusions, Contributions, and Recommendations for Future Work Abstract This chapter reflects on the conclusions of the research study carried out in previous chapters, and contribution to the knowledge made during this investigation. Furthermore, this chapter presents recommendations/suggestions for future directions Overview The objective of this project was to create a sustainable biocomposite with stiffnesstoughness balanced composites from biodegradable polymer matrix and miscanthus short fibers by melt processing. Chapters 3 and 4 described the experiments performed on the preparation and characterization of biodegradable polymer blends and compatibilizer for composite applications. The effects of the compatibilizer synthesized in Chapter 4 were investigated in Chapters 5, 6 and 7. Chapters 5 and 6 presented the performances of the single polymer matrix based biocomposites with and without compatibilizer. In chapter 7, the composites were produced with a binary blend matrix that showed optimum stiffness-toughness balanced properties described in Chapter 3. In Chapter 8, a statistical approach was adopted to identify the processing parameters that had the most significant effect on the performance of the uncompatibilized composite prepared in Chapter 7. Finally, the durability of the selected biodegradable polymer blends and their parent polymers were investigated in Chapter 9. Individual conclusions were presented for the research described in each chapter. The following section summarizes the interrelationship between results and objectives for this thesis Conclusions Blending of polymers is an effective and economical way to obtain new materials with desired properties. In order to design a biodegradable polymer blend with balanced mechanical 318

353 properties for biocomposite matrix application, commercially available poly(butylene succinate) (PBS) and poly(butylene adipate-co-terephthalate) (PBAT) were melt blended through extrusion and injection molding. Due to compatibility between the PBS and PBAT, the tensile toughness and tensile strength of the PBS were remarkably enhanced after the inclusion of PBAT. The observed compatibility results from the formation of copolyester due to the transesterification reaction between the parent polymers, which was observed using an infrared spectroscopy. Furthermore, the enhanced compatibility between the blended components was corroborated with DSC and DMA analysis. The PBS/PBAT blend mechanical properties appeared to be comparable with polyethylene mechanical properties. The rheological properties indicate that the PBS/PBAT blend has good processability, which can allow higher fiber loading for the composite fabrication. The surface morphological analysis of the PBS/PBAT blends provided evidence that the PBS and PBAT are not miscible at a molecular level. The prepared biodegradable PBS/PBAT (60/40 wt%) blend can be considered as a potential candidate for biocomposite applications. As reported in the literature, the compatibility between the hydrophilic natural fibers and hydrophobic polymer matrix is very poor. Therefore, this research aimed to synthesize compatibilizer as the next step of this thesis work. Solvent free and economically viable maleic anhydride grafted PBS, PBAT and PBS/PBAT blend were prepared in the presence of dicumyl peroxide (DCP) as a free radical initiator. The FTIR analysis was used to confirm the structural changes in MAH grafting samples. The MAH grafting yield was calculated by titration method. Among the MAH grafted PBS, PBAT and PBS/PBAT blend samples, a higher MAH grafting yield was observed on the PBS backbone. In addition, the MAH grafting efficiency was compared in the batch and continuous process. The batch processed sample had a slightly higher 319

354 yield than the continuous processed sample. During a MAH grafting reaction, both grafting and cross-linking phenomena can occur in the presence of initiator. Thermogravimetic analysis revealed that thermal stability of all the MAH grafted samples were found to be slightly reduced compared to their counterparts. Miscanthus fibers could be used as a reinforcement to produce PBS based biocomposites while reducing cost and retaining PBS mechanical properties. Therefore, biocomposites were produced from PBS and miscanthus fibers with and without compatibilizer by melt processing. A strong reinforcing effect of micanthus fibers led to an increase in the tensile and flexural modulus of the resulting PBS biocomposites. It was found that the tensile strength of uncompatibilized PBS/miscanthus composites was reduced compared to neat PBS. At the same time, the melt flow of the PBS composites was much lower compared to neat PBS. Indeed, the observed MFI value of the PBS biocomposites is appropriate for some injection molding processes. The compatibilizing efficiency of the MAH grafted PBS was investigated in the PBS biocomposites by means of mechanical performances. It was observed that the compatibilized PBS biocomposites showed superior impact strength, tensile strength, and flexural compared to their uncompatibilized one and neat PBS. However, the shortcoming of the PBS/miscnathus biocomposites was insufficient impact strength/toughness, which could limit its range of applications. The impact strength of the polymeric materials is one of the most important properties, which relates to the service life of the products. As a result, a higher impact strength biocomposite was produced from PBAT and miscanthus fibers while lacking stiffness, flexural strength and tensile strength of the resulting PBAT/miscanthus fiber composites. Balanced stiffness, toughness and thermal properties of the biocomposites were produced from PBS/PBAT blend matrix and miscanthus fibers. The tensile and impact strength of the 320

355 PBS/PBAT blend were decreased with the addition of miscanthus fibers. In order to overcome this issue in the resulting formulations, a reactive compatibilizer (MAH-g-PBS/PBAT) was introduced into the composite system. It was found that the mechanical properties of the compatibilized composites were noticeably increased as compared to the uncompatibilized one. The density of all the composites was found to be lower than synthetic fibers like glass fibers. The comptibilized PBS/PBAT/miscanthus fiber composites showed balanced performance compared to their individual compatibilized PBS/miscanthus composites and PBAT/miscanthus composites. From this study, it can be concluded that the prepared biodegradable polymer blend matrix based composites are a possible candidate to replace non-biodegradable composites in applications where biodegradability is essential after use but extreme thermal and humidity exposure are avoided. The influence of independent processing parameters upon the impact strength of PBS/PBAT/miscanthus composites was investigated by full factorial design of experiment. The statistically significant variables were analyzed at a 95% confidence level. It was found that the fiber length plays an important role to predict the impact strength of PBS/PBAT/miscanthus composites. The durability of PBS, PBAT and blend of PBS/PBAT was examined after being exposed to 50 o C and 90% humidity with comparison to polypropylene. Due to hydrolytic degradation of polyesters, the performances of the PBS, PBAT and their blend are not stable like PP under selected environmental conditions. The hydrolytic degradation of the polyesters was confirmed by FTIR, DSC, rheological, and morphological analysis Significant contributions Most of the commercially available biodegradable polymers are not satisfying their application requirements because of their cost and insufficient mechanical performances. In this 321

356 research work, the developed biodegradable PBS/PBAT blends have good mechanical, thermal and rheological properties, which are not commercially available in the market. This biodegradable polymer blend can be used for different industrial applications including composite fabrication, blow molding, films and packaging. The cost of the biodegradable polymers and their blends can also be reduced by composite fabrication with miscanthus fibers. Utilizing these biodegradable polymers as a matrix and miscanthus fibers as a reinforcement for biocomposite fabrication can reduce green house gas emissions while meeting consumer s short term application requirement. In the present study, the observed mechanical performances of the blends and composites were superior compared to the composites made with PP and miscanthus fiber composites. The developed biodegradable green composites are possible substitutes for a class of 100% petroleum-based non-biodegradable plastics and composites, which are currently used in different applications. The prepared fully biodegradable material is a promising candidate for environmental policies and public awareness, which can increase environmentally friendly material usage. Providing value addition for these environmentally friendly materials could increase the revenues for miscanthus growers. Dissemination of the new knowledge discoveries of the present study could result in the implementation of a wide range of bio-based materials. The final optimized biocomposite formulations were extruded and injection molded to produce prototypes in an industrial trial to further the commercialization prospects of the new technology, as shown in Figure The produced biocomposites could be optimized to meet the requirements for use in automotive applications, construction panels, and consumer products. 322

357 Figure Prototypes were made from biodegradable polymers/miscanthus fiber by extrusion and injection molding method Recommendations for future works Various electron-rich co-monomers (e.g., styrene) could be used to investigate the MAH grafting yield while reducing side reactions of the electron deficient monomers. The developed biocomposites should be compatibilized by modification of miscanthus fibers with sizing agents and using commercially available compatibilizers to compare to MAH grafted compatibilizer. The influence of fiber length on the performance of the resulting biocomposites should be further explored by varying fiber length. Further work could be performed by maximizing the number of processing variables as well as levels to predict the most significant processing parameters on the resulting biocmposites. During processing, the odor of the composite fabrication process should be eliminated to make industrial processing smooth and effective. 323

358 If these developed materials are required for long term applications, the durability of the biodegradable polymers and their composites needs to be improved by antihydrolysis agents. Future studies can be conducted on these developed materials to evaluate and certify them as a compostable blend and biocomposite. In order to compare the cost of the prepared composites with commodity plastics, economic analysis of the prepared composites should be studied. Life cycle analysis is necessary to prove that the developed biocomposites are an environmentally superior alternative to synthetic fiber reinforced composites. 324

359 Appendix I: Binary Blends of Poly(Butylene Succinate) and Poly(Butylene Adipate-co- Terephthalate): A New Matrix for Biocomposites Applications* *A version of this appendix has been published in: R. Muthuraj, M. Misra, and A. K. Mohanty, Binary blends of poly (butylene adipate-co-terephthalate) and poly (butylene succinate): A new matrix for biocomposites applications. PROCEEDINGS OF PPS-30: The 30 th International Conference of the Polymer Processing Society Conference Papers. Vol AIP Publishing, (adapted with kind permission from AIP Publishing LLC, Jul 09, 2015, License number ). Abstract In this study, biodegradable poly(butylene adipate-co-terephthalate) (PBAT) and poly(butylene succinate) (PBS) binary blends were melt compounded. The mechanical, thermal and morphological properties of the PBAT/PBS blends were investigated. The melt compounded binary PBAT/PBS blends showed balanced mechanical properties (especially in tensile strength and elongation) compared to neat components. The obtained melt flow index (MFI) value of the blends is much higher than PBAT. This may be attributed to PBS phase residual catalyst because it pronouncing thermal degradation of the polymers at higher temperature. The toughness of the PBAT is not significantly affected with addition of 40 wt% PBS in the PBAT/PBS blend. This could be the reason of good compatibility achieved between the PBAT and PBS phase in the blends. The phase morphology and spherulite morphology were also correlated with compatibility between the PBAT and PBS in the blends. A-I.1.Introduction In recent years, a new trend has arisen where polymer blends are being selected over individual biopolymers as matrices for composite applications. This is because the process of polymer blending is one of the most promising techniques to create materials with specific desired properties by combining the different advantageous qualities of two or more neat polymers. The resulting mechanical properties are typically a compromise between the parent 325

360 polymers. In cases of very successful compatibility, performance can be an overall improvement; however, blending typically results in a general decrease in mechanical performance. The most frequently used solution to overcome this problem is to compatibilize the blends. Block copolymers or grafted polymers are widely used as compatibilizers leading to finer phase morphologies and better interfacial properties [1]. However, the development of stiffness and toughness balanced polymer blends for biocomposite matrix applications is in high demand. PBS and PBAT are promising biodegradable polymers, with many inherent advantages. They present good biodegradability, excellent toughness, good thermal stability, and commercial availability in market. In addition, PBS and PBAT have been extensively studied in blends with other brittle biodegradable polymers. PBS is obtained from the petroleum-based succinic acid and 1,4- butanediol monomers. Interestingly, new methods have recently been developed such that these two monomers can also be synthesized from renewable resources [2]. Therefore, PBS has the newfound potential to be a biopolymer that is both biodegradable and bio-based. This new development may further diversify the PBS utilization in different fields of applications. Blending PBAT and PBS is of great interest because it returns desirable, unique properties while retaining the biodegradable nature of the base materials [3]. Although many studies have reported on balancing the stiffness-toughness in biodegradable binary blends. We report here the binary blends of PBAT and PBS which represents good examples of super tough blends from the biodegradable aliphatic-aromatic (PBAT) and aliphatic (PBS) polyesters. The resulting binary blends exhibit balanced mechanical and thermal properties. A-I.2. Materials and Methods Injection grade PBS, trade name Bionolle 1020, was procured from Showa Highpolymer Co. Ltd, Japan, with a molecular weight (Mw) of g/mol and PDI of The PBAT 326

361 (Biocosafe 2003F) was purchased from Xinfu Pharmaceutical Co., Ltd, China. The chemical structures of the neat PBS and PBAT are shown in Scheme A-I.1. Scheme A-I.1. Chemical structures of PBS and PBAT Prior to melt processing, the polymers were dried in conventional oven for 6 h at 80 o C to remove moisture and prevent degradation. Both PBAT and PBS were mixed and processed using a lab-scale twin-screw extruder, and injected into moulds (DSM Xplore 15 cc microcompounder). For all samples prepared, the operating temperature was 140 o C, the screw speed was 100 rpm, and the processing or dwell time of the materials inside the barrel was 2 min. Tensile properties of the PBAT/PBS blends were measured according to ASTM D638 in a Universal Testing Machine (Instron-3382) at a strain rate of 50 mm/min at room temperature; results are reported as average values of five replicates for each experiment. Heat deflection temperature (HDT) analyses were performed using a Dynamic Mechanical Analysis Q800 from TA Instruments, according to ASTM D648. Melt flow index (MFI) was measured according to ASTM D1238 at 190 o C temperature using a 2.16 kg load. Cryofractured sample morphology was examined using a Inspect S 50-FEI Netherlands scanning electron microscope (SEM) at an accelerating voltage of 20 kv. Before observing sample morphology, the samples were gold 327

362 coated with a final thickness of 20 nm with 20 ma. The spherulite morphology of the PBAT/PBS blend was observed using a Nikon polarizing optical microscope (POM) with hot stage; these micrographs were taken to observe crystallization. Samples were sandwiched between two microscope glass slides and heated to 150 o C for 5 min before quickly transferring the slide to the 150 o C microscope hotplate. Subsequently, samples were annealed at the crystallization temperature with a heating rate of 10 o C/min. The spherulite growth was recorded at crystallization temperatures of 85 o C using a Nikon camera. A-I.3. Results and Discussion The tensile properties of the PBAT, PBS and PBAT/PBS blends are shown in the Figure A-I.1. The tensile strength of the PBAT/PBS blends was higher than that of the neat PBAT. Specifically, the tensile strength of the PBAT/PBS (60/40 wt%) blend increased by 84% compared to neat PBAT. At the same time, the percentage elongation of the PBAT/PBS (60/40 wt%) is similar to the PBAT. The tensile strength improvement is directly related to the intermolecular forces, compatibility, and molecular orientation of the polymers in the blend [4]. In our previous study (Chapter 3) [3], it was concluding that the PBS/PBAT transesterification product acts as a compatibilizer in the PBS/PBAT blends and leads to the improvement of their tensile properties. In that specific work, the transesterification reaction was confirmed via normalized FTIR spectra. Particularly in that research, the carbonyl peak of the neat PBS, PBAT and PBS/PBAT blend was focused to observe its variation between the materials. The neat PBS and PBAT carbonyl group frequency was observed at 1712 cm -1. The carbonyl peak of the PBS/PBAT blend shows a slight shift towards higher wavenumbers (1716 cm -1 ). This shift could be the reason a transesterification product formed during melt processing in the presence of existing residual catalyst in the neat PBS and PBAT. In general, the resulting transesterification 328

363 product is more compatible with the homopolymers of the unreacted PBS and PBAT in the PBS/PBAT blends. A similar observation is reported in the literature for PBS/PCL blends [4]. The mechanical properties of PBAT/PBS blends are comparable with commercially available polyethylene (PE) reported previously on literature [5]. As such, we believe this indicates that PBAT/PBS blends could be a potential substitute for non-biodegradable polymers in packaging applications. Figure A-I.1. Tensile properties of (A) PBAT, (B) PBS, (C) PBAT/PBS (60/40 wt%) and (D) PBAT/PBS (70/30 wt%) blend MFI values of the PBAT, PBS and PBAT/PBS blends are presented in Figure A-I.2. The MFI of the PBAT/PBS blends (both 30 and 40 wt% PBS) were higher than neat PBAT, with tests resulting in blend values similar to neat PBS. This suggested that the viscosity of the blends is highly influenced by the incorporation of PBS into PBAT matrix. Alternatively, the low observed viscosities of the polymer blends may be due to thermal degradation resulting from exposure to high test temperatures i.e., 190 o C. Figure A-I.2 also presents the HDT of the PBAT, 329

364 PBS and PBAT/PBS blends. The HDT value of the PBAT (46 o C) is lower than PBS (90 o C). Both PBAT/PBS (60/40 wt% and 70/30 wt%) blends displayed HDT values intermediate of the neat PBAT and PBS. Figure A-I.2. HDT and MFI values of (A) PBAT, (B) PBS, (C) PBAT/PBS (60/40 wt%) and (D) PBAT/PBS (70/30 wt%) blend To confirm the previous discussion regarding phase morphology, cryofractured PBAT/PBS blend samples were observed by SEM (Figure A-I.3). The surface morphology of the blend reveals that spherical PBS particles were uniformly distributed throughout the matrix. This uniform dispersion of the PBS phase is attributed to the enhancement of compatibility in the blends [6]. 330

365 Figure A-I.3. SEM image of the cryofractured PBAT/PBS (60/40 wt%) blend The spherulite morphology of PBS and the PBAT/PBS (60/40 wt%) blend are shown in Figure A-I.4. Generally, PBAT has poor capability to form perfect spherulite morphology when compared to PBS. Therefore, the PBAT spherulite morphology is not shown in this work. The crystallization temperatures of PBAT, PBS, and the PBAT/PBS blend are 53, 83 and 85 o C, respectively. The PBS and PBAT/PBS blend spherulite morphologies were observed at 85 o C for 30 min. The selected temperature was based on the crystallization temperature of PBS and PBAT/PBS blends. High levels of spherulite growth were observed in both PBS and the PBAT/PBS blend. However, the higher number of nucleations in the PBAT/PBS blend interferes with crystal growth and lead to distorted lamellae. The fine dispersion of PBS phase in the PBAT matrix reduces the lamellae thickness of the blend compared to neat PBS lamellae. 331

366 Figure A-I.4. POM image of the (i) PBS and (ii) PBAT/PBS (60/40 wt%) blend A-I.4. Conclusions The PBAT/PBS blends were produced on a lab-scale injection-molding machine. After blending PBAT and PBS, the blends showed balanced stiffness-toughness properties compared to neat samples. These balanced properties were attributed to the excellent compatibility between the PBS and PBAT in the blends. The compatibility was caused by a transesterification product formed during melt blending which acts as a compatibilizer in the blend. SEM image of the PBAT/PBS blend confirms that there is a good compatibility achieved in the blends evidenced by a uniform dispersion of the PBS phase in the PBAT matrix. Optical polarizing microscopy results imply that the PBAT/PBS blend spherulite morphology was affected by the addition of PBS into the PBAT matrix; this may be due to the poor crystalline nature of PBAT. The MFI value of the PBAT/PBS blend is quite high compared to PBAT. This improved viscosity will help to facilitate the production of high fiber-content composites. The PBAT/PBS blend-based composites fabrication and performance evaluation are under investigation. 332

367 References 1. B. Imre, B. Pukánszky, Compatibilization in bio-based and biodegradable polymer blends. Eur.Polym.J, 49, (2013) (accessed on Febraury 06, 2014). 3. R. Muthuraj, M. Misra, A.K. Mohanty, Biodegradable poly (butylene succinate) and poly (butylene adipate-co-terephthalate) blends: Reactive extrusion and performance evaluation. J.Polym.Environ, 22, (2014). 4. J. John, R. Mani, M. Bhattacharya, Evaluation of compatibility and properties of biodegradable polyester blends. J.Poly.Sci.Part A: Poly.Chem, 40, (2002). 5. K. Joseph, S. Thomas, and C. Pavithran, Effect of chemical treatment on the tensile properties of short sisal fibre-reinforced polyethylene composites. Polymer, 37, (1996). 6. J.M. Willis, B.D. Favis, Processing-morphology relationships of compatibilized polyolefin/polyamide blends. Part I: The effect of an lonomer compatibilizer on blend morphology. Polym.Eng.Sci, 28, (1988). 333

368 Appendix II: Durability Studies of Biodegradable Polymers under Accelerated Weathering Conditions* *A version of this appendix has been published in: R. Muthuraj, M. Misra, and A. K. Mohanty, Durability Studies of Biodegradable Polymers under Accelerated Weathering Conditions, Society of Plastic Engineering (SPE, ANTEC), 2015, Orlando, Florida. Abstract Poly(butylene adipate-co-terephthalate), (PBAT) and poly(butylene succinate), (PBS) are promising biodegradable polyesters whose blends have gained great attention in wide range of applications. However, there are some drawbacks to the use of these biodegradable polymer blends in durable applications. The main disadvantage of these materials is hydrolytic degradation at elevated temperature and humidity. In this study, we have assessed the durability of PBAT, PBS and PBS/PBAT blends at 50 o C with 90% relative humidity (RH) for duration of up to 18 days. The mechanical properties of these polyesters were evaluated before and after 18 days of conditioning at 50 o C with 90% RH. The mechanical properties of the polyesters were affected with increasing conditioning time. This can be attributed to the susceptibility of ester bonds to hydrolytic degradation at elevated temperature and humidity. The hydrolytic degradation was further confirmed by scanning electron microscopy. A-II.1. Introduction PBAT, a biodegradable aliphatic-aromatic copolyester, is derived from adipic acid, terephthalic acid, and butane diol by polycondensation reaction [1]. Solely aromatic polyesters are insensitive to the microbial attack [2]. However, this insensitivity can be modified through copolymerization of aliphatic monomers with aromatic monomers, which resulted a biodegradable polymer i.e., PBAT. The mechanical properties of PBAT are comparable with low density polyethylene (LDPE) [3]. PBAT is widely used for compostable organic waste bags, agricultural mulch films as well as lamination/coatings for starch-based products [4]. However, 334

369 there are some aspects that limit the use this biodegradable polymer in large-scale applications. PBAT possesses excellent toughness, biodegradability and processability, allowing it to be used to tailor the properties of some biopolymers, thereby opening up new applications for PBAT based materials. PBS is an aliphatic polyester which is traditionally synthesized from fossil fuel based 1, 4- butanediol (BDO) and succinic acid by polycondensation reaction [1]. Recently, renewable resource based succinic acid and BDO can be produced by fermentation process. These biobased monomers allow for a biobased PBS production. PBS has potential as a commercial product as it shows wide variety of commercial applications because of its good processability, thermomechanical properties, relatively high heat deflection temperature and biodegradability. Also, PBS has properties close to commercial commodity polymers such as polyethylene (PE) and polypropylene (PP) [5]. However, insufficient impact strength of the PBS is limits its extensive applications. Polymer blending is an effective approach to creating a material with some desired properties by combining the different advantageous of two or more polymers. The resulting mechanical properties are typically a compromise between those of the parent polymers. Therefore, blending of PBS and PBAT is of great interest because it returns desirable properties while retaining the biodegradable nature of the parent polymers in the resulting blends. The resulting blend exhibits good mechanical, thermal and rheological properties [1]. Generally, durability is very important for the polymers in order to increase their suitability for a wider range of applications. However, only few studies have been examined the durability of biodegradable polymer based materials [6-8]. These literature sources report that 335

370 the biodegradable polymers are very sensitive to the elevated temperature and humidity. Additionally, Kim and coworkers [9,10] have reported the durability of talc filled biodegradable PBS/PBAT blends under marine environment. The authors suggest that the biodegradable material showed better elastic properties than that of the commercial non-biodegradable material. To the best of our knowledge, the durability study of PBS/PBAT blend has not yet been reported in the literature. We report here the durability of PBS/PBAT blend under simulated temperature (50 o C) and relative humidity (90%). The present study was aimed to investigate the durability of PBS, PBAT, PBS/PBAT and PP under simulated environmental conditions. The durability was analyzed by means of mechanical properties such as tensile, flexural and impact strength. A-II.2. Materials and Methods Commercially available PBS (Biocosafe 1903F) and PBAT (Biocosafe 2003F) pellets were procured from Xinfu Pharmaceutical Co., Ltd, China. Polypropylene (PP-1350N) was obtained from Pinnacle Polymers (Garyville, LA). The general properties of the PBS, PBAT, PBS/PBAT (60/40 wt%) and PP are shown in Table A-II.1. All the polymers were dried in oven for at least 8 h at 80 o C to remove the moisture prior to melt processing. Prior to injection molding, all the samples were extruded in a Leistritz twin-screw extruder with a screw speed of 100 rpm. The extruder had an L/D ratio of 48 and a screw diameter of 27 mm. The extrudates were pelletized and dried at 80 o C for 12 h prior to the injection molding. The dried extruded pellets were then injection molded in an ARBURG injection molding machine to obtain desired test specimens. The injection molding machine had a maximum injection pressure of 2000 bar and a screw diameter of 35 mm. The extrusion and injection molding process were carried out with a processing temperature of 140 o C for PBS, PBAT, PBS/PBAT (60/40 wt%) and 180 o C for PP. 336

371 Table A-II.1. General properties of PBS, PBAT, PBS/PBAT (60/40 wt%) and PP. (a obtained from material data sheet, b and PBS/PBAT (60/40 wt%) data were measured in the lab) Properties PBS PBAT PBS/PBAT (60/40 wt%) PP Melt flow index (g/10min) a Melting point ( o C) a Density (g/cm 3 ) a Moisture absorption after ± ± ± ± days at 50 o C with 90% RH b In order to evaluate the durability, the injection molded samples were exposed to 50 o C with 90% RH in an environmental chamber (Endurance C340, Envirotronics, Inc). The tensile and flexural properties of the samples before and after conditioning were measured according to ASTM standards in a Universal testing machine (Instron-3382) with a 50 kn load cell at room temperature. The crosshead speed for tensile and flexural test was 50 mm/min and 14 mm/min, respectively. Notched Izod impact strength was measured as per ASTM D256 in a TMI impact testing machine. The results are reported an average values of five replicates for each set of samples. Cryofractured sample morphology was examined using scanning electron microscopy (Inspect S50-FEI Company) at an accelerating voltage of 20 kv. Prior to observing sample morphology, the samples were gold coated with a final thickness of 20 nm with 20 ma. A-II.3. Results and Discussion Durability of PBS, PBAT, PBS/PBAT (60/40 wt%) and PP samples were examined in terms of mechanical properties. The results obtained from the samples before and after conditioning are depicted in Figures A-II.1 and A-II.2 and Table A-II.2. Figure A-II.1 shows the 337

372 tensile strength of the samples before and after 18 days conditioning. The tensile strength of PBS, PBAT and PBS/PBAT significantly reduced after 18 days exposed to 50 o C with 90% RH. Generally, biodegradable polymers undergo hydrolytic degradation in the presence of moisture and heat [6]. After 14 days exposed to 50 o C with 90% RH, biodegradable polymers showed greater moisture absorption than PP (Table A-II.1). This is attributed to the relatively high polarity of the biodegradable polymers studied here. The higher moisture absorption of PBS, PBAT and their blend leads to a decrease in molecular weight and tensile strength via hydrolysis of the ester bonds. Due to the non-polar nature of PP, this plastic samples maintains its tensile strength after 18 days exposed to 50 o C with 90% RH. Our findings have good agreement with literature [8]. Figure A-II.1. Tensile strength of PBS, PBAT, PBS/PBAT (60/40 wt%) and PP before and after 18 days exposed to 50 o C with 90% RH. Flexural strength of the PBS, PBAT, PBS/PBAT and PP samples before and after 18 days conditioning is shown in Figure A-II.2. The flexural strength of PBS and PP is higher than PBAT and PBS/PBAT blend. Except PBAT, all the samples showed a slight increase in the 338

373 flexural strength after 18 days conditioning. This increase can be ascribed to post-crystallization phenomena which occurred during conditioning [11]. Figure A-II.2. Flexural strength of PBS, PBAT, PBS/PBAT (60/40 wt%) and PP before and after 18 days exposed to 50 o C with 90% RH. The notched Izod impact strength of the samples is shown in Table A-II.2. After 18 days conditioning, there is no significant change observed in the impact strength of PBAT. This could be due to the PBAT maintaining its sufficient molecular weight in order to form a significant degree of entanglement [12]. In the present study, impact strength of PBS was improved by adding PBAT. It can be seen that the impact strength of PBS, PBS/PBAT decreased considerably after 18 days exposed. In addition, it was clearly observed that the fracture mode of PBS/PBAT changed from ductile to brittle after exposure to heat and humidity. This may be attributed to the hydrolysis products of PBS accelerated the degradation of PBAT in the PBS/PBAT blend system. As mentioned before, PP is resistant to hydrolytic degradation due to its non-polar nature. Therefore, as expected, mechanical properties are not affected significantly after 18 days exposure to the hydrolysis environment. 339

374 Table A-II.2. Notched Izod impact strength of PBS, PBAT, PBS/PBAT (60/40 wt%) and PP before and after 18 days conditioning at 50 o C with 90% RH. Samples PBS PBAT PBS/PBAT (60/40 wt%) PP Before conditioning (J/m) ± 6.55 Non-break Non-break 30.4 ± 7.08 After 18 days conditioning (J/m) ± 1.21 Non-break ± ± 5.45 Figure A-II.3 shows the SEM micrographs of the samples before and after 18 days of conditioning at 50 o C with 90% RH. The PBS, PBAT and PBS/PBAT exhibits relatively very smooth and clear surface morphology before exposed to elevated temperature and humidity. It is clear from the SEM images that after 18 days of conditioning, the samples had gained rough surface, deep holes as well as cavities. This result agrees with the considerable reduction observed in the mechanical properties of the samples after 18 days exposure [8]. Furthermore, ongoing degradation can be seen (Figure A-II.3) in the form of slightly eroded regions. This shows that the PBS, PBAT and PBS/PBAT underwent degradation with increasing conditioning time. The formed holes and cavities are attributed to the dissolution of the oligomers during hydrolysis mechanism. All these observations suggest that the PBS, PBAT and PBS/PBAT can readily undergo hydrolytic degradation in the presence of high humidity and temperature. 340

375 Figure A-II.3. SEM micrographs of PBS, PBAT, and PBS/PBAT (60/40 wt%) before (A, B and C) and after (D, E and F) 18 days exposed to 50 o C with 90% RH. 341