Thermal Conductivity Enhancement of Polymer Based

Home , Polymer

THERMAL CONDUCTIVITY ENHANCEMENT OF POLYMER BASED

MATERIALS

A Dissertation

Submitted to

The Graduate Faculty of The University of Akron

In Partial Fulfillment

Of the Requirements for the Degree

Doctor of Philosophy THERMAL CONDUCTIVITY ENHANCEMENT OF POLYMER BASED

MATERIALS

Marjanalsadat Kashfipour

Dissertation

Accepted: Approved:

Adviser Department Chair Dr. Jiahua Zhu Dr. H Michael Cheung

Interim Dean of College Committee Member Dr. Craig Menzemer Dr. Shiva Sastry

Committee Member Dean of the Graduate School Dr. Rajeev Gupta Dr. Chand Midha

Committee Member Date Dr. Tianbo Liu

Committee Member Dr. Qixin Zhou

ii DEDICATION

My beloved parents and my dearest brother for their endless love, support and

encouragements. Without them none of my success would be possible.

iii ACKNOWLEDGEMENTS

I would like to express the deepest appreciation to my advisor Dr. Jiahua Zhu for his leadership, and encouragement throughout the entirety of this research. Without his guidance and persistent help this dissertation would not have been possible.

I would like to thank my committee members, Dr. Rajeev Gupta, Dr. Qixin Zhou,

Dr. Tianbo Liu, and Dr. Shiva Sastry for their valuable comments and suggestions that enhance the quality of the work. I am especially indebted to Dr. Shiva Sastry, without him and his kind support during my graduate studies, I would have never been able to stand where I am. As my teacher and mentor, he has taught me more than I could ever give him credit for here. I also have to extend my appreciation to Dr. H Michael Cheung, Chairman of the Department of Chemical and Biomolecular Engineering, and Dr. Craig Menzemer,

Interim Dean of College of Engineering, for their invaluable support.

I am grateful to all of those with whom I have had the pleasure to work during my graduate studies in University of Akron. I’d like to recognize the assistance that I received from Russell. S. Dent, a smart undergraduate student, whose help cannot be underestimated.

I am also grateful to Ali Eghtesadi, my best friend and great companion, who always supported me with his unwavering patience during this process.

Last but not least, I would like to express my deepest gratitude to my parents,

Ashraf Salami and Ali Kashfipour, and my beloved brother, Dr. Mojtaba Kashfipour, for their unconditional trust, timely encouragement, and endless patience. I am indebted to my parents for making the necessary sacrifices to ensure that I received the best education and iv opportunities possible. I wouldn't be where I am today without the strength and courage they have cultivated in me. I cannot thank my brother enough for his influences upon my strengths, his unbelievable support, and never giving up on me and always remaining by my side.

v ABSTRACT

Enhancing the thermal conductivity (TC) of polymeric materials for thermal management applications has attracted attentions because of their beneficiary features such as light weight, anti-corrosive, low cost, flexibility and controllable electrical conductivity.

Since phonons are the dominant heat carriers in insulating materials, creating pathways for better phonon transfer and decreasing the phonon scattering inside the matrix are the major strategies for TC enhancement. TC of bulk polymers is much less than their single chains because of the chains entanglement that increases the phonon scattering. Therefore, any approaches that decreases the entanglement of chains or enhances their alignment can be used for TC improvement. Traditionally, TC of insulating materials have been enhanced by incorporation of thermally conductive fillers. Formation of a continuous network of these fillers and their alignment can enhance TC even further in the desired direction. The network of fillers can be achieved at high content of fillers that is accompanied with sacrificing other properties such as mechanical properties and results in high cost of final products. As a result, alternative approaches that can form such a network at low content of fillers have attracted attentions.

Here, we present three different approaches, which were utilized for TC enhancement of different systems. First, induced co-continuous morphology of an immiscible polymer blend, blend of high-density polyethylene (HDPE) and poly (methyl methacrylate) (PMMA) was used for localization of carbon nanofibers (CNFs). The co- continuous morphology of immiscible polymer blends has been previously used for vi formation of continuous network of electrically conductive fillers and electrical

conductivity (EC) enhancement. This method, known as double percolation method, requires both the composition of polymers and fillers reach to percolation threshold above which they form co-continuous phase morphology and continuous network of fillers, respectively. Being inspired by this method and considering the involved parameters in tuning the morphology and distribution of fillers, we could show that processing temperature that affect the viscosity ratio of components and distribution of fillers is a key role for EC and TC enhancements. To investigate the effect of temperature on the morphology and distribution of fillers, two different temperatures of 150 and 230 °C were used for processing the blend with different contents of CNFs. The samples that were processed at 230 °C showed finer morphology and higher EC than the processed samples at 150 °C for all the content of CNFs. While, the samples processed at 150 °C showed

higher TC and coarser morphology. The difference in the trends of TC and EC is because

of their different mechanisms. The finer morphology and better distribution of fillers at

230 °C is accompanied with formation of more interfaces that increases the interfacial

phonon scattering and decreases TC consequently. To our knowledge, this is the first study investigating the effect of processing temperature on the location of fillers in an immiscible polymer blends and its effect on TC and EC of the composite obtained.

Secondly, a series of isotropic thermally conductive composites were fabricated by incorporating xylitol crystals into aligned boron nitride (BN) aerogel (BNA) that was

formed by ice-template method. As it was mentioned earlier, both the continuous network

vii of fillers and their alignment in the desired direction can further increase the efficiency of

TC enhancement. BN is a two dimensional (2D) ceramic filler that is electrically insulating and has high TC. Hence, it is a good candidate for fabricating the composite which should be exclusively thermally conductive to be used in electronic devices. Similar to the other

2D fillers, BN has anisotropic TC with high TC in the in-plane direction. Therefore, TC will be efficiently increased by aligning the in-plane direction of BN parallel to the heat flow direction. There are numerous methods for aligning 2D fillers in the horizontal direction, whereas their alignment in the vertical direction is not that easy. Ice-template method is a simple and low-cost method for vertical alignment of fillers along the direction of ice crystal growth. Therefore, by employing this method, BNA with vertically aligned

BN walls was fabricated. Besides the filler network, the filling agent has a key role in the

TC of the final composite as well. Sugar alcohols (SAs) have relatively high TC compared to polymers and this is because of their high crystallinity. In this study the fabricated BNA went through carbonization (CBNA) first to decrease its hydrophilicity and increase its structure integrity and then molten xylitol was infiltrated into it. Xylitol crystals were solidified perpendicular to the BN walls, creating crystal packs between BN walls. As a result the xylitol crystal packs offset the anisotropic TC of the scaffold. These results offer new insights into isotropic thermally conductive composites that can be used for next generation of heat dissipating materials. Similarly, this scaffold was filled with erythritol, another SA, and the effect of scaffold on the phase change properties and TC of obtained composite was investigated. In addition to TC enhancement and gaining isotropic TC, BN

viii

scaffold improved the subcooling effect, shape and thermal stability, and the ability of the

erythritol for releasing heat during crystallization process.

At last, a filler-free approach was used for TC enhancement of a polymer-based

system. Here, the effects of intermolecular interactions and engineering such interactions

on mechanical flexibility, optical transparency and TC of the system were investigated. For

this purpose, sodium carboxymethyl cellulose (SCMC), a well-known hydrophilic biopolymer, and xylitol, were selected as the matrix and the filler, respectively. Increasing the content of xylitol up to 50 wt% resulted in enhanced TC of up to 1.75 times of neat

SCMC. Besides, the mechanical flexibility and optical transparency were also improved.

The achieved enhancements are attributed to the newly formed hydrogen bonding that is due to presence of numerous hydrophilic functional groups in the both components.

Formation of new hydrogen bonding between SCMC and xylitol was accompanied with formation of homogenously distributed thermal network throughout SCMC. Therefore, engineering the interchain interactions can be an alternative filler-free approach for enhancing TC of polymeric-based materials.

TABLE OF CONTENTS

LIST OF FIGURES ...... xiii

LIST OF TABLES ...... xviii

CHAPTER

INTRODUCTION ...... 1

BACKGROUND ...... 3

2.1. Heat Conduction in Non-metals ...... 3

2.1.1. Importance of MFP ...... 5

2.2. Thermally Conductive Composites ...... 6

2.2.1. TC of the Matrix and Involved Parameters ...... 7

2.2.2. Thermally Conductive Fillers ...... 11

2.2.3. Improvement of Filler/Matrix Interactions ...... 14

2.3. Copyright Notice ...... 16

THE EFFECT OF PROCESSING TEMPERATURE ON THE DISTRIBUTION OF

CONDUCTIVE FILLERS IN CO-CONTINUOUS POLYMER BLENDS AND THEIR

ELECTRICAL AND THERMAL CONDUCTIVITY ...... 17

3.1. Outline ...... 17

3.2. Introduction ...... 17

3.3. Experimental Procedures...... 21

3.3.1. Materials ...... 21

3.3.2. Sample Preparations...... 21

3.3.3. Characterization ...... 22

3.4. Results and Discussion ...... 23

3.5. Conclusion ...... 41

3.6. Copyright Notice ...... 42

DIRECTIONAL XYLITOL CRYSTAL PROPAGATION IN ORIENTED MICRO-

CHANNELS OF BORON NITRIDE AEROGEL FOR NON-DIRECTIONAL HEAT

CONDUCTION ...... 43

4.1. Outline ...... 43

4.2. Introduction ...... 44

4.3. Experimental Procedures...... 46

4.3.1. Materials ...... 46

4.3.2. Sample Preparations...... 46

4.3.3. Characterization ...... 48

4.4. Results and Discussion ...... 49

4.5. Conclusion ...... 66

4.6. Copyright Notice ...... 66

REGULATING INTERMOLECULAR CHAIN INTERACTION OF BIOPOLYMER

WITH NATURAL POLYOL FOR FLEXIBLE, OPTICALLY TRANSPARENT AND

THERMALLY CONDUCTIVE HYBRIDS ...... 68

5.1. Outline ...... 68

5.2. Introduction ...... 68

5.3. Experimental Procedures...... 71

5.3.1. Materials and Sample Preparation ...... 71

5.3.2. Characterization Methods ...... 72

5.4. Result and Discussion ...... 72

5.5. Conclusions ...... 86

5.6. Copyright Notice ...... 87

COMBINATION OF ERYTHRITOL CRYSTAL PROPAGATION IN ORIENTED

BORON NITRIDE AEROGEL FOR NON-DIRECTIONAL HEAT CONDUCTION

AND PHASE CHANGE PROPERTIES ...... 88

6.1. Outline ...... 88

6.2. Introduction ...... 89

6.3. Experimental Procedures...... 90

6.3.1. Materials ...... 90

6.3.2. Sample Preparations...... 90

6.3.3. Characterization ...... 91

6.4. Results and Discussion ...... 92

6.5. Conclusions ...... 111

REFERENCES ...... 112

APPENDIX A: COPYRIGHT PERMISSIONS ...... 133

xii

LIST OF FIGURES

Figure 1. Generation of phonon by lattice vibration.4 ...... 4

Figure 2. Different types of phonon scattering: (a) phonon-phonon scattering, (b) phonon- impurity scattering, (c) phonon-boundary scattering.4 ...... 6

Figure 3. Parameters that should be considered for designing thermally conductive composites...... 7

Figure 4. Thermal conductivity of polymers with different molecular structure. The polymers with π-conjugated bonds are shown with solid lines and circle markers, whereas others are shown with dashed lines and triangle markers.10 ...... 9

Figure 5. Heat transfer mechanism as a function of size and aspect ratio of fillers in a polymer composite.3 ...... 13

Figure 6. Schematic drawing of heat conduction at the interface of a thermally conductive filler and a polymer matrix, over time.3 ...... 15

Figure 7. Schematic S-shaped curve of log conductivity versus electrically conductive filler (carbon black, here) concentration.36 ...... 18

Figure 8. (a) Change of EC in a blend of polylactide (PLA), poly(ε-caprolactone) (PCL) and acid-oxidized multiwalled carbon nanotubes (A-MWCNTs) (PLA/PCL/A-MWCNT) as a function of PCL composition, (b) schematic drawing for the morphological changes of part (a) where blue regions represent PLA phase, green regions represent PCL phase, and black lines represent A-MWCNTs.41 ...... 20

Figure 9. SEM micrographs of PMMA-extracted blends with the PMMA/ HDPE wt% of: 30/70 (a), 40/60 (b), 50/50 (c), 60/40 (d) and 70/30 (e). All samples demonstrate co- continuous morphology...... 24

xiii

Figure 10. Shear viscosity as a function of shear rate for neat PMMA and HDPE at two temperatures (150 and 230°C). The viscosity ratios are calculated based on the viscosity at processing shear rate range of 60 s-1 to 200 s-1 (between green dash lines)...... 25

Figure 11. SEM micrographs of cryo-microtomed surface of neat blends processed at 150 °C (a) and 230 °C (b) after PMMA-extraction. 230-PMMA/HDPE sample illustrates a finer morphology...... 26

Figure 12. TC of HDPE, PMMA, 230-PMMA/HDPE/X and 150-PMMA/HDPE/X filled with X= 2, 4, 8 and 16 wt% of CNF...... 27

Figure 13. Volume EC of HDPE, PMMA, 230-PMMA/HDPE/X and 150- PMMA/HDPE/X filled with X= 2, 4, 8 and 16 wt% of CNF...... 28

Figure 14. SEM micrographs of cryo-microtomed surface of 150-PMMA/HDPE/16 (a) and 230-PMMA/HDPE/16 (b) after PMMA-extraction. Note that the co-continuous morphology was maintained in presence of 16 wt% CNF...... 29

Figure 15. AFM phase images of 230-PMMA/HDPE/16 (a&a’) and 150-PMMA/HDPE/16 (b&b’)...... 30

Figure 16. (a) Cumulative area ratio as a function of agglomerate area for 150- PMMA/HDPE/16 and 230-PMMA/HDPE/16, (b) Storage modulus and loss modulus as a function of angular frequency for 150-PMMA/HDPE/16 and 230-PMMA/HDPE/16. ... 34

Figure 17. Storage (G’) and loss modulus (G”) as a function of angular frequency for the (a) PMMA/CNF and (b) HDPE/CNF composites...... 35

Figure 18. Storage modulus and loss modulus as a function of angular frequency for the 150-PMMA/HDPE/X (a) and 230-PMMA/HDPE/X (b) composites with different CNF contents...... 36

Figure 19. Schematic drawing of 150-PMMA/HDPE/X (a) and 230-PMMA/HDPE/X (b) composites showing their phase morphology and fillers distribution...... 37

Figure 20. Comparing EC (a) and TC (b) of epoxy composites filled with different types of CNTs.99 ...... 38

xiv

Figure 21. (a) Tensile modulus of neat samples and corresponding calculated modulus by parallel and series model, (b) tensile modulus of filled samples and the predicted values from the models...... 40

Figure 22. Tensile strength of individual components, neat blends and composites with 16 wt% CNF content...... 41

Figure 23. Schematic illustration of fabrication process of Xy-CBNA composites...... 48 Figure 24. SEM cross-section of (a) 1BNA, (b) 2BNA, (c) 3BNA and (d) 4BNA. Their respective higher magnification is presented in (e to h)...... 49

Figure 25. XRD patterns of powdered CBNA and CBNA scaffold...... 50

Figure 26. (a) BNA supports 500 g weight, (b-e) CBNAs support 500 g weight, (f) and (g) compare the size of BNA and CBNA, (h) FT-IR spectra of BNA and CBNA, showing the presence and the absence of O-H stretching peak of BNA and CBNA, respectively, (i) filling BNA with xylitol that led to cracked scaffold, (j) Xy-xCBNA composite that remained intact after xylitol infiltration...... 52

Figure 27. TGA thermogram of xylitol and Xy-xCBNA composites with x= 1 to 4...... 53

Figure 28. Cross-section SEM visualization of Xy-1CBNA (a), Xy-2CBNA (b), Xy- 3CBNA (c), Xy-4CBNA (d). The crystals are perpendicular to the ex-BN walls in all the samples...... 55

Figure 29. The PLLA crystal growth in the confined spaces between GONs. 135 ...... 55

Figure 30. XRD patterns of xylitol, CBNA and Xy-xCBNA with x= 1 to 4...... 56

Figure 31. The through-plane TC of Xy-xCBNA with x = 1 to 4, the in-plane TC of Xy- 1CBNA and Xy-3CBNA and the through-plane TC of Ep-4CBNA...... 59

Figure 32. FLIR images of the through-plane direction of EP-4CBNA (a, a’ & a”), EP- 4CBNA (b, b’ & b”) and Xy-4CBNA (c, c’ & c”), and the in-plane direction of Xy-4CBNA (d, d’ & d”) at different times. (e) The temperature-time profile of the through-plane direction of EP-4CBNA, EP-4BNA and Xy-4CBNA, and the in-plane direction of Xy- 4CBNA...... 60

Figure 33. Micro-CT scan slices of xylitol filled CBNA at different times that shows the presence of empty hollows inside the composite. These images confirmed that presence of this much of air inside the scaffold has a minor effect on the bulk TC of the composites...... 64

Figure 34. Predicted through-plane TC of Xy-xCBNA with x= 1 to 4 with different theoretical models...... 65

Figure 35. (a) Schematic drawing of how the relatively short and rigid polymer A penetrates within the gyration radius of a longer polymer B and the homogenous distribution of polymer A within polymer B results in formation of percolating thermal pathways, (b) measured thermal conductivities of spin-cast PAP:PAA films at various 17 monomer mole fractions of PAP (ϕPAP)...... 70

Figure 36. The H-bonds involved in SCMC are illustrated; intrachain H-bonding are shown in blue and interchain H-bonding are shown in green. The H-bonds between the molecules of SCMC can be replaced by new H-bonds between xylitol and SCMC molecules. The shown configuration is only one of the possibilities in SCMC-xylitol system...... 73

Figure 37. (A) FT-IR spectra of SCMC-xylitol films with different xylitol contents (SX0- SX5 respectively), (B) higher magnification of hydroxyl stretching peaks which is highlighted in part (A)...... 75

Figure 38.Stress-strain curves of SCMC-xylitol films SX0- SX5 respectively...... 78

Figure 39.(A) Summary of the effect of xylitol content on the modulus of elasticity (ME) and (B) maximum stress (TS) of SCMC-xylitol films...... 78

Figure 40. (A) Before and (A’) after applying force on SX0, SX1 and SX2 samples. These samples could not be bent and were broken immediately after bending. (B) Sample SX4 could be bent but broke when rolled 6(B’), and the SX5 sample was flexible enough to be rolled (C)...... 79

Figure 41.Thermal conductivity of SCMC-xylitol film increases with the content of xylitol in 5.0 g SCMC. Notice the improvement from 0.43 W·m-1·K-1 (neat SCMC) to 0.75 W·m- 1·K-1 when the mixture 50 wt% xylitol...... 81

Figure 42. Formation of H-bonds between the hydroxyl-hydroxyl groups or hydroxyl- carboxyl groups in SCMC-xylitol system that results in formation of a thermally conductive network within SCMC...... 82

xvi

Figure 43. (A) TGA and (B) Derivative TGA (DTGA) thermograms of SX0-SX5 samples and xylitol (Xyl)...... 83

Figure 44. UV-Vis and optical images of all the films. The above figures illustrate transparency of SX0, SX1, SX2, SX4 and SX5 films. The film containing 50 wt% of xylitol has the highest transmittance of ~ 93%...... 85

Figure 45. Inner structure of the scaffold before (a) and after (b) carbonization visualized with SEM. Comparing the size of the scaffold before (a’) and after (b’) carbonization, and both the scaffolds could support 500 g weight before and after carbonization (c&d) and (e&f), respectively...... 95

Figure 46. Internal microstructure of 1CBNA (a) and 2CBNA (b), their corresponding higher magnification images (a'&b'), and internal structure of filled scaffolds, 1CBNA (a”) and 2CBNA (b”)...... 97

Figure 47. XRD diffractograms of xylitol, CBNA, Ery-1CBNA and Ery-2CBNA...... 98

Figure 48. TGA thermogram of neat erythritol, Ery-1CBNA and Ery-2CBNA composites (a), though-plane TC of erythritol, Ery-1CBNA and Ery-2CBNA as a function of BN loading and in-plane TC of Ery-1CBNA (b), cross-section SEM visualization of composites showing the perpendicular direction of xylitol crystals versus BN walls (c&d), FLIR images of the through-plane direction of Ery-1CBNA (c, c' & c") and Ery-2CBNA (d, d' & d") at different times, and their corresponding temperature-time profile (e) .... 100

Figure 49. (a) DSC heating and (b) cooling curves of erythritol, Ery-1CBNA and Ery- 2CBNA obtained at scan rate of 2 °C/min and their corresponding trend of melting (c) and solidification (d), and subcooling (e) points...... 107

Figure 50. DSC thermograms of Erythritol (a), Ery-1CBNA (b) and Ery-2CBNA (c) after every 10 heating-cooling cycles. Shape stability test of the composites was investigating by placing the composite (d) on a hot plate set at 140 °C, the top (d’) and side-view (d”) of the composite after 45 mins...... 110

xvii

LIST OF TABLES

Table 1. Efficiency of PMMA extraction from the extruded blend composites ...... 24

Table 2.Summary of polar, dispersive and total interfacial tensions at 150 and 230 °C .. 32

Table 3. Calculated surface tensions between pairs of components and the wettability parameter of the tertiary system...... 32

Table 4. Characteristic domain size (ξ) of 150-PMMA/HDPE/16 and 230- PMMA/HDPE/16 ...... 33

Table 5. The relation between the ex-BN content, the wall distance and the crystal size of xylitol in Xy-xCBNA with x = 1 to 4...... 57

Table 6. Summarized literatures of the fabricated composites with either isotropic or anisotropic TC...... 61

Table 7. Tdmax, T50% and weight loss (%) of samples at 300 and 500 °C is shown. Increasing the content of xylitol resulted in faster thermal degradation and higher weight loss...... 84

Table 9. The loading of BN in Ery-1CBNA and Ery-2CBNA composites and their corresponding Tdecomposition-onset versus neat erythritol...... 101

Table 10. Phase change temperatures and corresponding enthalpies of Ery-1CBNA and Ery-2CBNA composites...... 108

Table 11. Phase change temperatures and enthalpies of erythritol and the obtained composites extracted from DCS cyclic thermograms...... 110

xviii

CHAPTER I

INTRODUCTION

Thermal conductivity and thermal management have been critical problems in a

broad spectrum of applications for several decades. Unmanaged heat affects the behavior

of the devices and increases the overall cost of the system. Materials with higher thermal

conductivity (TC) offer pathways for faster heat transfer and thus prevent unanticipated

malfunctions. In particular, polymeric materials are attractive for such applications because

of their versatility, light weight and non-corrosive properties. Polymers are, however, intrinsically thermally and electrically insulating materials and, hence for thermal applications it is necessary to improve their thermal conductivity. As a result, TC

enhancement of polymers and also other insulating materials has attracted attentions.

Recent efforts have focused on using polymer nanocomposite materials containing

thermally conductive nano-fillers because their attractive mechanical and thermal properties

can be adjusted to suit specific application demands. Typically, continuous conductive network

pathways should be created to enable phonon transfer, however, such an approach requires a

high content of fillers and is difficult to manufacture. The processing of composites containing

fillers at this level is difficult, and the product costs will also increase. Therefore, alternative

approaches that facilitate formation of continuous thermally conductive network with lower

content of fillers have attracted great attentions.

Recently, it was shown that by engineering the interchain interactions between polymer

chains a thermal network was formed along the polymeric matrix. This filler-free approach has

raised attentions because it sheds light on the fundamental aspects of TC in bulk polymers and

also it does not suffer from wide variety of problems, which polymer composites have.

Here, we will focus on selectively enhancing TC of polymeric and other insulating materials with either formation of a continuous network of thermally conductive fillers or using filler-free approaches.

CHAPTER II

BACKGROUND

2.1. Heat Conduction in Non-metals

It is known that heat transfer occurs through conduction, convection and radiation.1

Generally, heat conduction is the main mode for heat transfer inside solid materials. Heat

conduction for a steady state and in one dimension can be expressed by Fourier’s equation

as follow:

= . . (1) 𝛥𝛥𝛥𝛥 𝑞𝑞 𝑘𝑘 𝐴𝐴 𝐿𝐿 Where

q is the rate of heat conduction (W)

k is the TC parameter (W·m-1·K-1)

A is the cross section area (m2)

ΔT is the temperature difference (°C)

L is the length of conduction path (m).2

Basically, TC of a material is evaluated based on its TC parameter (k), which shows the rate of heat diffusion in the material.3 In solid materials, heat conduction can be carried

out by either electrons or phonons, which are the quantized form of energy produced by

lattice vibration, as shown in Figure 1.2-3

Figure 1. Generation of phonon by lattice vibration.4

In non-metallic materials such as polymers, the contribution of phonons is

dominant over electrons and TC can be obtained from the following equation5-6:

( ) = ( ) ( ) (2) 1 𝐾𝐾 𝑇𝑇 3 ∑𝑖𝑖 ∫ 𝐶𝐶𝑖𝑖 𝜔𝜔 𝜐𝜐𝑖𝑖𝑙𝑙𝑖𝑖 𝜔𝜔 𝑑𝑑𝑑𝑑 Where Ci ( ) d is the heat capacity contribution of phonons with polarization of

i and frequency of 𝜔𝜔 , 𝜔𝜔i is the velocity of phonons and li is the mean free pathway of

phonons. This equat𝜔𝜔ion𝝊𝝊 was simplified in Debye equation where k is simply defined as:

= (3) 𝐶𝐶𝑝𝑝𝑣𝑣 𝑙𝑙 𝑘𝑘 3 In this equation, Cp is the specific heat capacity per volume, v is the phonon velocity and l is the mean free pathway (MFP)of phonons.2-3 The velocity of a phonon is a function of the bonding strength in the molecules’ structure and MFP is the average distance that a phonon travels between two collisions.2

2.1.1. Importance of MFP

Although polymers are known as thermal insulator materials (TC of 0.1- 0.5 W·m-

1·K-1), TC of their single chains is significantly higher than their bulk values. For instance, an extended single chain of polyethylene (PE) with chain length of above 100 nm showed

TC of 350 W·m-1·K-1.7 Similarly, single chain of poly (dimethyl siloxane) PDMS showed

TC of ~7 W·m-1·K-1 contrary to its bulk value that is significantly smaller due to the phonon

scattering, which is induced by conformational disorder of chains.2, 8

Considering Debye’s equation, Cp and v values are the same for both the bulk and single chain of a polymer. Hence, l, MFP, should be responsible for different TC values.

In general, more phonon scattering lead to shorter MFP, which results in lower TC. The

inherent imperfections in non-metals such as voids, impurities, polymer chain

entanglements and chain ends cause the phonon scattering and result in decreased bulk TC

compared to the single chain value.2, 9 Therefore, for bulk polymers, improving the

orientation of chains, decreasing the chain entanglement, increasing the amount of ordered

and oriented region, and crystallinity can improve the MFP and consequently TC.2

There are three types of phonon scattering: (i) phonon-phonon scattering, (ii)

phonon-impurity scattering and (iii) phonon-boundary scattering, shown in Figure 2.4

Figure 2. Different types of phonon scattering: (a) phonon-phonon scattering, (b) phonon- impurity scattering, (c) phonon-boundary scattering.4

Harmonic vibration of all the atoms in a lattice with the same frequency is an ideal case. In reality, when atoms vibrate without harmony it results in phonons with different frequencies that causes the phonon-phonon (Umklapp) scattering. Impurities and defects of any kind act as a barrier for phonon transfer and cause phonon-impurity scattering.

Phonon-boundary scattering is the major type of phonon scattering in composites that occurs because of the acoustic mismatch at the filler-matrix interface. This phenomena results in interfacial thermal resistance. Therefore, incorporation of thermally conductive fillers can enhance TC of the matrix only if the phonon transfer is dominant over the induced phonon scattering.

2.2. Thermally Conductive Composites

Usually, TC of non-metallic materials such as polymers and sugar alcohols (SAs) has been improved with incorporation of thermally conductive fillers. Depending on the final applications of the fabricated composite, either electrically conductive or exclusively

thermally conductive fillers can be incorporated for TC improvement of the matrix. Carbon

based and metallic fillers can enhance both electrical conductivity (EC) and TC of the matrix, whereas ceramic fillers can only enhance TC. Since composites are composed of

both matrix and fillers, the TC of final composite is a function of the properties of both the

components. Therefore, constructing a composite with relatively high TC requires considering the properties of its components that affect their individual TCs and that of the

final product. These properties are summarized in Figure 3.

Figure 3. Parameters that should be considered for designing thermally conductive

composites.

2.2.1. TC of the Matrix and Involved Parameters

Since polymers and SAs have been used as the matrix and filling agent for the

scaffolds of thermally conductive fillers in this thesis, these types of materials are the main

focus of this part.

As it was discussed earlier, TC of polymers in chain level is really high, which is

due to a good phonon transfer in the absence of disordered structure and entanglements

that can induce the phonon scattering. In this scale, orientation of polymer chains and their

ordered structure results in TC improvement. Besides, the strength of bonding and

molecular weight are the other parameters that can influence TC of a polymer. Hence, a

comprehensive understanding of the role of individual parameters on TC of polymer chains

and also TC of bulk polymer is of great interest.4

2.2.1.1. Effect of Molecular Structure

The effect of molecular structure of polymers (including their composition and

conformation) on TC has been investigated with simulation. A recent study, using molecular dynamic simulation investigated the effect of various molecular structure on TC.

The results are demonstrated in Figure 4. Based on these results, polymers with more rigid backbone showed higher TC that was due to larger phonon velocity and suppression of segmental rotation in presence of strong bonds.10 In addition, other parameters such as the

comprising atoms and the functional groups should be considered as well. Presence of

heavier atoms such as oxygen lower the phonon velocity10 whereas the functional groups

can cause mass disorder and consequent localized vibrational modes 11.

Figure 4. Thermal conductivity of polymers with different molecular structure. The

polymers with π-conjugated bonds are shown with solid lines and circle markers, whereas

others are shown with dashed lines and triangle markers.10

2.2.1.2.Effect of Crystallinity

Crystallinity can enhance TC of bulk polymers due to enhanced intrinsic ordering

of chains and consequent decreasing of the phonon scattering, which is caused by random

conformation of amorphous polymer chains. As a result, most of polymers with higher degree of crystallinity show higher TC than their amorphous counterparts.2 In addition to

the intrinsic crystallinity of a polymer, the processing condition can impact the crystallinity.

For instance, Yu et al. studied TC of low-density (LDPE), high-density (HDPE) and ultra-

high density PE (UHPE). The authors also carried out high pressure and temperature

treatments on HDPE and UHPE to evaluate their effects on TC of these polymers. UHPE

demonstrated higher TC improvement, which was explained by increased thickness of its

lamellae from ~20 nm to the range of 100-150 nm. As a result, the extended crystal

structure and increased lamellae thickness can effectively increase TC.12

2.2.1.3.Effect of Bonding Strength

It is known that in single polymer chains the stronger covalent bonds in the

backbones can transfer phonons more efficiently. Additionally, in bulk polymers the

interactions between the chains should be considered for bulk TC values. This interactions

can be either covalent (in crosslinked polymers) or non-covalent bonding.

Similar to the covalent bonding in the backbone of polymer chains, presence of

inter-chain covalent bonding offers a more efficient phonon transfer across the matrix.13-14

Chains can be covalently bonded if they are crosslinked. However, it should be considered

that crosslinking has a dual effect of creating both the inter-chain covalent bonds and

branching that lead to decreased linearity of polymer chains and TC. To get benefit from

crosslinking for TC enhancement, the effect of branching and decreased linearity of

polymer chains should cancel. Different studies showed that higher degree of crosslinking

can enhance TC more.15 Incorporation of the linkers that can cause strong interactions

between the chains leads to creation of better phonon pathways. Besides, it has been

demonstrated that the shorter linkers can create strong interactions that is accompanied

with confining the structural disorder of polymer chains.14 Whereas, bulkier linkers

decrease TC due to more phonon scattering and segmental rotations.10

The non-covalent interactions in polymers are mainly categorized as van der Waals

(vdW) and hydrogen bonding (H-bonding) interactions. Generally, it is known that H- bonding is stronger than vdW interactions.16 Therefore, they can be more efficient for

development of phonon transfer pathways in between of polymer chains. In this regard, the

strength and density of H-bonding, size of H-bonding functional groups and their

distribution in the matrix should be considered as well.4 Recently, Kim et al. showed that

a well-defined thermal network can be formed within a hydrophilic matrix with

incorporation of hydrophilic linkers and increased H-bonding along the matrix. For this

purpose, the homogenous distribution of H-bonding is important that needs good

miscibility between the components of the polymer blend.17 Their results show that in

addition to the traditional methods of incorporation of solid thermally conductive fillers

into polymers, the TC of polymers can be enhanced by engineering inter-chain interaction

as well.

2.2.2. Thermally Conductive Fillers

Depending on the final application of a composite, corresponding types of fillers

should be incorporated into the matrix. Metallic and carbon-based fillers are both thermally

and electrically conductive. Besides, electrons transfer heat more efficiently because they

are more resistant to the scattering and also have higher speed than phonons. Ceramic fillers

have been studied for the applications that need electrically insulating and thermally

conductive materials. Since, in ceramic fillers the heat transfer is predominantly through

phonons their TC is generally lower than the other two types of fillers.2

In addition to the types of fillers and their intrinsic TC, the TC of the ultimate

composite is a function of shape18-19 and size of fillers20-23, the adhesion and the thermal resistance at the interface of the components.

It should be considered that shape of fillers can also affect the viscosity of the

matrix, which is an essential factor in processing composites. Viscosity is related to the

filler-filler and filler-matrix frictions. The irregular shapes of fillers can increase the

viscosity and reduce the processability. However, the fillers with less effect on the viscosity may not be very effective on TC. Therefore, it is necessary to select the filler shape that can cause optimum properties.2

Spherical fillers have high viscosity percolation threshold. Therefore, high loading

of them can be incorporated into polymeric materials without significant change of

viscosity.24 On the other hand, fillers with high aspect ratio that can be either one- or two-

dimensional have lower filler percolation threshold to create thermally conductive

network.25 The incorporation of these fillers into a polymeric matrix is accompanied with

change of the viscosity but with lower percolation threshold for creation of an effective network along their aspect ratio direction. Since they usually have anisotropic TC along this direction, their alignment leads to anisotropic TC in a matrix.2

As for the effect of particle size, it should be considered that it is hard to study the

size of fillers as an independent factor affecting TC. In general, the smaller fillers create

larger interfacial area that leads to increased phonon scattering and decreased TC.20-21, 26-27

Figure 5 shows the heat transfer mechanism as a function of the size and the aspect ratio

of fillers.3

Figure 5. Heat transfer mechanism as a function of size and aspect ratio of fillers in a

polymer composite.3

However, there are some reports about better impact of smaller fillers on TC of

composites. The better effect of smaller fillers (nano-sized vs. micron-sized) was explained

due to their better distribution23 that resulted in easier formation of thermally conductive

network.28 Although nano-sized fillers suffer from their high interfacial area and increased phonon scattering they can have a positive role in TC improvement when they are mixed

with micron-sized fillers. In this case they act as thermal bridges between the micron-sized

fillers and result in further improvement of TC.23, 27

2.2.3. Improvement of Filler/Matrix Interactions

As it was mentioned earlier, the interfacial thermal resistance between the components of a composite is one of the main reason for the low TC values. The interfacial thermal resistance is mainly due to I) thermal contact resistance and II) thermal boundary resistance. Thermal contact resistance is due to the mismatches of the surface conditions

of the components, whereas the thermal boundary resistance originates from variation of

vibrational mode of heat carriers (e.g., electrons or phonons) at the interface.29 Figure 6

shows the schematic drawing of heat conduction at the interface of a thermally conductive

filler and a polymer matrix, over time.3

Figure 6. Schematic drawing of heat conduction at the interface of a thermally conductive

filler and a polymer matrix, over time.3

The usual strategy for decreasing the interfacial thermal resistance is to

functionalize the fillers and improve the filler-matrix interactions.2 However, it should be noted that this approach can only be useful if it does not cause surface defects on fillers.

Presence of defects on the surface of fillers increases the chance of phonon scattering.30

There are numerous studies on the type of surface modifiers and the optimum conditions

for TC improvement, which is beyond the scope of this dissertation.31-35

Considering the above parameters, we designed innovative composites with

selectively enhanced TC and great potential for heat management applications.

2.3. Copyright Notice

Significant part of this chapter was adapted with permission from following publications:

“Polymer Nanofibers with Outstanding Thermal Conductivity and Thermal Stability:

Fundamental Linkage between Molecular Characteristics and Macroscopic Thermal

American Chemical Society.

CHAPTER III

THE EFFECT OF PROCESSING TEMPERATURE ON THE DISTRIBUTION OF CONDUCTIVE FILLERS IN CO-CONTINUOUS POLYMER BLENDS AND THEIR ELECTRICAL AND THERMAL CONDUCTIVITY

3.1. Outline

In this study, we aimed to construct a continuous thermally conductive network of

fillers by getting benefit from the induced co-continuous morphology of an immiscible polymer blend. This approach, known as double percolation method, has been successfully

used for enhancing EC of polymeric based materials. Although this method has been applied for TC enhancement of polymeric based materials but different aspects of materials and process parameters need more careful adjustment to get better results. In this study we showed the capability of this approach for selective enhancement of TC and EC.

Additionally, the effect of processing temperature, as one of the involved processing parameter, on the behavior of EC and TC of the blend was investigated.

3.2. Introduction

It is known that formation of a continuous electrically conductive network of fillers

inside insulating materials can enhance their conductivity more effective than the random

distribution of fillers.2 Formation of such a network requires reaching to the critical

concentration of fillers (percolation threshold) to create continuous pathways within the

17 matrix. Figure 7 shows the trend of EC as a function of the content of conductive fillers and it illustrates the zone that is known as percolation threshold.36

Figure 7. Schematic S-shaped curve of log conductivity versus electrically conductive filler

(carbon black, here) concentration.36

It should be considered that formation of such a network with incorporation of high loading of fillers is accompanied with sacrificing mechanical properties and low cost of final product.2 Therefore, other strategies, which can create the conductive network with relatively lower content of fillers are of great importance. It has been shown that the induced co-continuous morphology of immiscible blends can be utilized as an alternative

method for this purpose. Co-continuous morphology is defined as the presence of three-

dimensional spatial network of each polymeric component in their immiscible blend.37

Selective localization of fillers in either one of the polymeric components38 or at their interface39-40 can decrease the percolation threshold of fillers. Consequently, the conductivity will be a function of the concentration of conductive fillers in the containing phase and the continuity of this phase, as shown in Figure 8.41 This approach is known as

double percolation method since both the composition of polymers and fillers should reach

to the percolation threshold above which they form co-continuous morphology and the

continuous conductive network, respectively.39, 42 As a result, as long as fillers are

homogenously distributed within the matrix and are close enough that electrons can

transfer with tunneling effect, EC will be enhanced.43-44

Similar to EC, significant enhancement of TC can be achieved with presence of a

continuous network of fillers within the targeted matrix. However, for TC, there is no

tunneling effect and phonons scatter at the interfacial areas and disconnections.2, 4, 45-49

Hence, the phonon pathways should be formed in a way that phonon transfer dominates

the interfacial phonon scattering. Therefore, using double percolation method for TC

enhancement requires more accurate adjustment of involved parameters than EC.

Figure 8. (a) Change of EC in a blend of polylactide (PLA), poly(ε-caprolactone) (PCL)

and acid-oxidized multiwalled carbon nanotubes (A-MWCNTs) (PLA/PCL/A-MWCNT) as a function of PCL composition, (b) schematic drawing for the morphological changes of part (a) where blue regions represent PLA phase, green regions represent PCL phase, and black lines represent A-MWCNTs.41

Although double percolation method has been successfully reported in some recent

studies for TC enhancement50-53, different aspects of the materials and processing

parameters that can effectively improve TC should be further investigated. These parameters such as blend composition54-56, mixing time37, 57-58, processing temperature59-63,

sequence of addition58, viscosity64-68 and elasticity ratio69-72 and interfacial tensions73-75 can

alter the phase morphology, the corresponding composition range for co-continuous

morphology and the location of fillers.

In this study, the effect of processing temperature on the viscosities of polymeric

components, location of fillers (carbon nanofibers (CNFs)) and their corresponding effects

on TC and EC of the immiscible blend of high density polyethylene (HDPE) and poly

(methyl methacrylate) (PMMA) was investigated. The results of this study can be helpful

as a strong foundation for the design of novel materials with high EC and TC that also have

desirable mechanical properties. Furthermore, this investigation can be helpful in obtaining

the optimum processing conditions and understanding of the involved mechanisms for EC

and TC of polymer.

3.3. Experimental Procedures

3.3.1. Materials

Bimodal HDPE, mixture of 20k and 400k molecular weight with the ratio of 1:1,

MFI of 0.35 g/10 min and density of 0.95 g/cm3 was provided by Chevron Phillips

Company. PMMA with a density of 1.18 g/cm3 and MFI of 15 g/10min was provided by

Atlugas International of Arkema Inc. CNF, PR-24-XT-HHT with average diameter of ~

100 nm, was purchased from Pyrograf Products Inc.

3.3.2. Sample Preparations

HDPE and PMMA were dried at 80 °C in a vacuum oven for 12 h prior to melt compounding. Desired amount of HDPE and PMMA powders were simultaneously mixed with specific amount of CNF powder at room temperature in a brabender volumetric feeder with a constant rate of 1kg/h. The mixture went through melt mixing in Thermoscientific

Process 11 Parallel co-rotating intermeshing twin screw extruder with a 40:1 L/D ratio.

The screw rotation speed was kept constant as 300 rpm. Material was extruded through a strand die with a hole diameter of 2.5 mm and collected through a Thermoscientific water bath and pelletizer. Two different melting temperature of 150 and 230 °C were selected for melt mixing of the composite. The obtained composites are referred to as T-

PMMA/HDPE/X where T and X are the adjusted temperature of the extruder and wt% of

CNF, respectively. After extrusion, pelletized beads were compression-molded at 24.1

MPa and processing temperatures for 3 min in the compression molding machine (Carver).

The obtained disks were used for TC tests as well as rheology tests, while the electrical

conductivity characterizations were performed on rectangular compression-molded

samples

3.3.3. Characterization

Capillary rheometer test:

Capillary rheometer RH10 from Malvern Company equipped with a 32mm/1mm die and a zero-length die were used to perform the viscosity tests. The viscosity of neat

PMMA and HDPE were measured in a shear rate range from 20 s-1 to 3000 s-1 at two

temperatures of 150 °C and 230 °C, respectively. The viscosity ratios were calculated based

on the viscosities at shear rate of processing window from 30 s-1 to 200 s-1.

Oscillatory rheometer test:

Small amplitude oscillatory shear measurements were performed using a TA

Instruments ARES G2 rotational rheometer equipped with a 25mm parallel plate geometry enclosed in an environmental chamber for temperature control. Amplitude sweeps were conducted at a frequency of 1 rad/s to find out the limit of the linear viscoelastic region.

Frequency sweeps were performed at angular frequencies ranging from 0.1 rad/s to 100 rad/s for each test. All tests were conducted at processing temperatures.

Mechanical properties measurements:

Tensile measurements were performed in accordance with MTS Insight

Mechanical Testing System. ASTM 638 was selected for all of the mechanical tests with

5mm/min rate at room temperature. The Young’s modulus was determined from the initial

slope of the stress–strain curve, while the tensile strength was measured at the break point.

SEM morphology visualization:

The Field Emission Scanning Electron Microscopy (FESEM) visualization was

carried out using a JEOL instrument (JSM-6510LV) under 20kV voltage. For morphology

visualization, samples were cryo-fractured in liquid nitrogen. To extract PMMA, the cryo-

fractured samples were submerged into acetone for 24 h to selectively dissolve and extract

PMMA.

AFM morphology visualization:

AFM measurements were carried out on XE7 manufactured by Park System. Prior

to AFM visualizations, samples went through cryo-sectioning at -80 °C using a Diatome

cryo 25° diamond knife mounted on Leica UC7/FC7 cryo-ultramicrotome. Then,

topographic and phase maps were taken with AFM tapping mode using a PPP-NCHR tip.

The scan rate and the other mapping parameters were optimized for the mode and the

environment of the experiment.

Electrical and Thermal Conductivity measurements:

The TC and EC of compression-molded samples were measured using C-Therm

TCi Thermal Conductivity Analyzer and Prostat PRS-801 resistance system, respectively.

These measurements were carried out at room temperature.

3.4. Results and Discussion

The SEM images of cryo-fractures surfaces of PMMA-extracted blends is shown in Figure 9. Samples were prepared with PMMA/ HDPE weight ratios of: 30/70, 40/60,

50/50, 60/40 and 70/30, respectively. As illustrated in these images, all the compositions

show the co-continuous morphology. The blend with the weight ratio of 40/60 was selected

for the further studies due to the complete extraction of PMMA (Table 1).

Figure 9. SEM micrographs of PMMA-extracted blends with the PMMA/ HDPE wt% of:

30/70 (a), 40/60 (b), 50/50 (c), 60/40 (d) and 70/30 (e). All samples demonstrate co- continuous morphology.

Table 1. Efficiency of PMMA extraction from the extruded blend composites

Composition PMMA/ HDPE wt%

30/70 40/60 50/50 60/40 70/30 PMMA 97.4 99.6 97.8 97.0 95.8 Dissolved%

Samples were prepared at two temperatures of 150 and 230 °C. Shear viscosities of neat PMMA and HDPE at these temperatures are shown as a function of shear rate in Figure

10. Based on these results, both materials had shear-thinning characteristic. On the other hand, the viscosity of PMMA was decreased more than the HDPE’s due to higher

activation energy of PMMA. As a result, there is a big difference between the viscosity

ratios in these two temperatures and consequent filler distribution in the systems.76

100000 PMMA 150oC HDPE 150oC PMMA 230oC o 10000 HDPE 230 C Viscosity Ratios (PMMA/HDPE) 150oC :3.5 230oC :0.9 1000

100 Shear Viscosity (Pa.s)

10 10 100 1000 10000 Shear rate (1/s)

Figure 10. Shear viscosity as a function of shear rate for neat PMMA and HDPE at two

temperatures (150 and 230°C). The viscosity ratios are calculated based on the viscosity at processing shear rate range of 60 s-1 to 200 s-1 (between green dash lines).

Figure 11 (a&b) show the SEM micrographs of cryo-microtomed surface of neat

PMMA/HDPE 40/60 blend at 150 and 230 °C (150-PMMA/HDPE and 230-PMMA/HDPE, respectively) after PMMA extraction. Although both samples showed co-continuous morphology, 230-PMMA/HDPE sample demonstrated a finer morphology which was

25 likely because of better mixing of the blend with viscosity ratio close to 1.0. This phenomenon was due to easier breakage of lower viscous droplets by shear stress.77

Figure 11. SEM micrographs of cryo-microtomed surface of neat blends processed at

150 °C (a) and 230 °C (b) after PMMA-extraction. 230-PMMA/HDPE sample illustrates a finer morphology.

In the next step, 2, 4, 8 and 16 wt% CNF was mixed and extruded directly with

HDPE and PMMA powders at 150 and 230 °C.

0.8

) PMMA/CNF HDPE/CNF 0.7 150-PMMA/HDPE/X 230-PMMA/HDPE/X W/(m.K) ( 0.6

0.5

0.4

0.3 Thermal Conductivity 0.2 0% 2% 4% 8% 16%

CNF (wt%)

Figure 12. TC of HDPE, PMMA, 230-PMMA/HDPE/X and 150-PMMA/HDPE/X filled with X= 2, 4, 8 and 16 wt% of CNF.

Figure 12 shows the TC of HDPE, PMMA, 230-PMMA/HDPE/X and 150-

PMMA/HDPE/X filled with different content of CNF. The TC of neat blends 150-

PMMA/HDPE and 230-PMMA/PE were measured as 0.39 and 0.37 W·m-1·K-1, respectively. The TC of CNF has been reported to be > 600 W·m-1·K-1.78 The large TC difference between the matrix and CNF causes high interfacial thermal resistance between the components.50 As a result, TC of the blend was not affected significantly with low content of CNF. With increasing the content of CNF up to 16 wt% the TC was increased to 0.65 W·m-1·K-1 for 150-PMMA/HDPE/16 and 0.55 W·m-1·K-1 for 230-

PMMA/HDPE/16. These results demonstrate the effect of processing temperature on TC of the final composites. In addition, contrary to all the other composites, the TC of 150-

PMMA/HDPE/16 composite was higher than the TC of HDPE containing the same content of CNF (HDPE/16).

Figure 13. Volume EC of HDPE, PMMA, 230-PMMA/HDPE/X and 150-

PMMA/HDPE/X filled with X= 2, 4, 8 and 16 wt% of CNF.

As demonstrated in Figure 13, the EC of all samples as a function of filler content is S-shaped. This confirms that the electrical percolation behavior resulted in a sudden change in the EC.36 HDPE, PMMA and their neat blend have EC of 10-15 79,1.82× 10-13 79 and 10-13 80, respectively and increasing the CNF content results in a stepwise increment of

EC and consequently a significant change with CNF above 8 wt%. Interestingly, the EC of

230-PMMA/HDPE/16 composite was higher than its counterpart, 150-PMMA/HDPE/16,

which was in contrast with the TC results. Therefore, although the processing temperature affected both TC and EC of the polymer blend composites with same content of CNF, the composite with higher TC had lower EC.

The distributions of CNF in 150-PMMA/HDPE/16 and 230-PMMA/HDPE/16

composites were visualized with FESEM and AFM techniques. In Figure 14, the FESEM

micrograph of these two composites after PMMA extraction is shown. It is demonstrated

that the co-continuous morphology of both composites was maintained at 16 wt%

concentration of CNF. Similar to the neat blends, the composite processed at 150 °C had

coarser morphology.

Figure 14. SEM micrographs of cryo-microtomed surface of 150-PMMA/HDPE/16 (a) and

230-PMMA/HDPE/16 (b) after PMMA-extraction. Note that the co-continuous morphology was maintained in presence of 16 wt% CNF.

The AFM phase images of 230-PMMA/HDPE/16 and 150-PMMA/HDPE/16 are displayed in Figure 15. The CNF is shown in red, PMMA in blue and PE in green. These images confirmed the finer morphology of 230-PMMA/HDPE/16 as well.

Figure 15. AFM phase images of 230-PMMA/HDPE/16 (a&a’) and 150-PMMA/HDPE/16

(b&b’).

In both samples, most of CNF was present in PMMA phase which was consistent with the thermal dynamic theory.81 Based on this theory, the location of fillers can be predicted by the value of wettability parameter. Based on Young’s equation, the wettability parameter is as follow 58:

= cos = (4) 𝛾𝛾𝑠𝑠−1−𝛾𝛾𝑠𝑠−2 𝜔𝜔12 𝜃𝜃 𝛾𝛾12 Where s-1 and s-2 are the surface tensions of polymer 1 and 2 at the filler’s surface

58 and 12 is the su𝛾𝛾rface tension𝛾𝛾 between polymers. If the ω12 < -1 fillers will be in polymer

1, if 𝛾𝛾ω12 > 1 fillers will be located in polymer 2 and if -1 <ω12 < 1 then the fillers will be

located at the interfaces of the blend. s-1, s-2 and 12 can be calculated with help of

different proposed models82-84 which has𝛾𝛾 been𝛾𝛾 used in 𝛾𝛾many studies.39, 85-87 Here, Owens–

Wendt model88 was used:

= + 2 2 (5) 𝑑𝑑 𝑑𝑑 𝑝𝑝 𝑝𝑝 𝛾𝛾12 𝛾𝛾1 𝛾𝛾2 − �𝛾𝛾1 𝛾𝛾2 − �𝛾𝛾1 𝛾𝛾2 It should be noted that the interfacial tensions at the processing temperature should be used for this equation. For this purpose, if γp/γ = xp then the surface tensions of polymer

at different temperatures can be calculated with these assumptions:

-dγ/dT= 0.06

-dxp,d/dT = 0

and the surface tension variation of CNF with temperature is neglected.89 The calculated

surface tensions at 150 and 230 °C are summarized in Table 2Error! Reference source

not found. which can be used for calculating the wettability parameters with Young’s

equation assuming HDPE as polymer 1 and PMMA as polymer 2. The calculated surface

tensions and consequent wettability parameters are summarized in

Table 3.

Table 2.Summary of polar, dispersive and total interfacial tensions at 150 and 230 °C

Components 20 °C (mN·m-1) 150 °C (mN·m-1) 230 °C (mN·m-1)

γ γd γp γ γd γp γ γd γp

PMMA87 41.1 33.2 7.9 33.3 26.9 6.4 28.5 23.0 5.5

HDPE 90 34.3 28.6 5.6 26.5 22.1 4.3 21.7 18.1 3.6

CNF89 40.3 18.4 21.8 40.3 18.4 21.8 40.3 18.4 21.8

Table 3. Calculated surface tensions between pairs of components and the wettability parameter of the tertiary system.

γ12 γ12 1 2 (150 °C) (230 °C)

HDPE PMMA 0.43 0.50

PMMA CNF 5.48 5.78

HDPE CNF 6.94 7.83

ω12 3.38 4.10

The wettability parameters in both temperatures were above 1.0 meaning that the

CNF will be localized in PMMA phase. However, CNF could be found in the PE phase as

well; as a result of the effect of kinetic force favoring adsorption of fillers into less viscous

phase. The polymer with lower viscosity infiltrates into the filler agglomerates more easily

and results in better wetting of fillers.91

Figure 15 (b) demonstrates the bigger CNF agglomerates in 150-PMMA/HDPE/16 compared to the ones in 230-PMMA/HDPE/16. The degree of phase coarsening was

determined with the characteristic domain size ( ) using the following equation92:

=ξ (6) 𝐴𝐴 𝐿𝐿 Where A is the total area of the image ξand L is the total length of the interfaces. The characteristic domain sizes are summarized in Table 4. The characteristic domain size of

150-PMMA/HDPE/16 was larger than 230-PMMA/HDPE/16 which means that 150-

PMMA/HDPE/16 has a coarser morphology and more stretching but less breakage of the droplets during the processing.

Table 4. Characteristic domain size (ξ) of 150-PMMA/HDPE/16 and 230-

PMMA/HDPE/16

Polymer composites

150-PMMA/HDPE/16 230-PMMA/HDPE/16

Characteristic 0.48μm 0.39μm domain size ( )

Figure 16 (a) illustrates cumulative area ratio as a function of agglomerate area.

The cumulative area ratio distribution was calculated by = 𝑗𝑗 ∑𝑖𝑖=1 𝐴𝐴𝑖𝑖 𝑛𝑛 𝐹𝐹𝑗𝑗 �∑𝑖𝑖=1 𝐴𝐴𝑖𝑖� 93:

= 𝑗𝑗 (7) ∑𝑖𝑖=1 𝐴𝐴𝑖𝑖 𝑛𝑛 𝐹𝐹𝑗𝑗 �∑𝑖𝑖=1 𝐴𝐴𝑖𝑖� Where Fj is cumulative area ratio of Aj, which is the summation of agglomerate

areas from the smallest to the jth agglomerate ( ) divided by the total agglomerate 𝑗𝑗 𝑖𝑖=1 𝑖𝑖 area ( ). In this figure, the slope value of 230∑ -PMM𝐴𝐴 A/HDPE/16 was higher than its 𝑛𝑛 𝑖𝑖=1 𝑖𝑖 value ∑for 150𝐴𝐴 -PMMA/HDPE/16 sample, demonstrating a better dispersed system. In addition, almost all the agglomerates were smaller than 0.5 μm2 while in 150°C composites,

more than 30% of the CNF agglomerates were larger than 0.5 μm2 and the biggest one was

~ 1.5 μm2. These numbers were consistent with the AFM images, confirming a better dispersive mixing of CNF when the temperature was set at 230°C. The improved dispersion establishes stronger CNF network which was reflected in higher modulus shown in Figure

16 (b).

Figure 16. (a) Cumulative area ratio as a function of agglomerate area for 150-

PMMA/HDPE/16 and 230-PMMA/HDPE/16, (b) Storage modulus and loss modulus as a function of angular frequency for 150-PMMA/HDPE/16 and 230-PMMA/HDPE/16.

The polymer-filler and filler-filler interactions were further investigated with

rheological measurements. For this purpose, the oscillatory shear rheology behavior of

samples at low frequency was investigated.94 Figure 17 (a) and Figure 17 (b) show the storage (G’) and loss modulus (G”) of PMMA and HDPE composites as a function of angular frequency, respectively.

Figure 17. Storage (G’) and loss modulus (G”) as a function of angular frequency for the

(a) PMMA/CNF and (b) HDPE/CNF composites.

The neat polymers behaved like a Newtonian liquid with the viscoelastic properties exhibiting liquid like features95 such as the storage and loss moduli being a functions of

frequency in the terminal regions as follow 96:

• G ′ 2 • G" ∝ 𝜔𝜔

The neat PMMA performed the prototypical∝ 𝜔𝜔 fluid characteristics at very low frequency, while the G’ went through a plateau at 8 wt% CNF concentration. This plateau

was caused by a solid percolated network of the CNF, illustrating a transition from liquid-

like to a solid-like behavior.42 This concentration is known as rheological percolation

threshold and it was consistent with the electrical percolation threshold. Appearance of a

plateau in the storage modulus as a proof for creation of a solid network has been reported

in the other filled polymer systems as well.97-98 Since the used HDPE was a mixture of low

and high molecular weight polymers (bimodal), it did not show a typical behavior of G’ at

low frequency.

Figure 18. Storage modulus and loss modulus as a function of angular frequency for the

150-PMMA/HDPE/X (a) and 230-PMMA/HDPE/X (b) composites with different CNF

contents.

Figure 18 (a&b) demonstrate the G’ and G” of filled blends as a function of angular

frequency, processed at 150°C and 230°C, respectively. A plateau in G’ was observed at

low frequency for neat PMMA/HDPE blends which was due to the existence of lamellar

PMMA phase in the co-continuous blends42 and the broad molecular weight distribution

HDPE. With any particular frequency, G’ and G” systematically increased with increasing

CNF loading because the stiff agglomerates enhanced the modulus of the systems. In both systems processed at 150 and 230 °C, materials started to display a solid like behavior (G’ >

G”) when the filler concentration reached 8 wt%. Therefore, a better CNF network began

to form with high CNF loadings which led to the establishment of better pathways of

electrons and phonons.

Based on all the summarized results, the finer morphology of 230-PMMA/PE/16

with more homogenous distribution of CNF was responsible for its higher electrical

conductivity. The schematic drawing of the phase morphology and distribution of fillers in these two composites are represented in Figure 19.

Figure 19. Schematic drawing of 150-PMMA/HDPE/X (a) and 230-PMMA/HDPE/X (b) composites showing their phase morphology and fillers distribution.

Although fillers were not fully connected in the fabricated composites, their homogenous distribution through the matrix with help of tunneling effect of electrons significantly enhanced the EC. However, for TC, the finer morphology led to presence of

more interfaces followed by more phonon scattering.99 Therefore, due to different mechanisms of electron and phonon transfer, sample with higher EC does not show

necessarily higher TC. This results are consistent with Gonjy et al. study, which

investigated the influence of different types of carbon nanotubes (i.e. SWNT, DWNT and

MWNT) on EC and TC of epoxy based composites. They observed that incorporation of

fillers with lower specific surface area (SSA), MWNT, could enhance TC better than

SWNT and DWNT that was ascribed to less interfacial phonon scattering of MWNT.

However, composites with higher SSA (SWNT, DWNT) fillers had higher EC. TC and EC values of the epoxy composites as a function of the filler content are presented in Figure

20.99

Figure 20. Comparing EC (a) and TC (b) of epoxy composites filled with different types

of CNTs.99

The contribution of electrons in TC can be calculated based on Wiedermann–Franz

law100:

= × × (8)

𝑘𝑘𝑒𝑒 𝐿𝐿0 𝜎𝜎𝑒𝑒 𝑇𝑇

Where ke is the TC of electrons, L0 is the Lorentz constant and is equal to 2.44 ×

-8 -2 3 10 ΩWK , σe is the EC and T is the temperature. Substituting the obtained EC values in

-7 -1 -1 this formula resulted in ke of < 8 × 10 W·m ·K which demonstrated the negligible effect of electrons in TC of these composites. This is consistent with the reported negligible role of electron as energy carriers in semiconductors or insulating materials.101 Moreover, based

on AFM images, CNFs were mostly located in PMMA phase in 150-PMMA/HDPE/16

composite as bigger agglomerates that resulted in decreased distance between fillers, increased network density and improved TC.

The effect of co-continuous morphology and incorporation of CNF on the mechanical properties was evaluated as well. The young’s moduli of individual polymers, the neat blends, and 150-PMMA/HDPE/16 and 230-PMMA/HDPE/16 composites are illustrated in Figure 21 (a&b). It is demonstrated that with incorporation of 16 wt% of CNF, the modulus was increased for both the neat polymers and also the blends. Parallel model

(E = + ) and series model (E = / + / ) are two commonly used −1 1 1 2 2 1 1 2 2 models𝜙𝜙 to𝐸𝐸 predict𝜙𝜙 𝐸𝐸 the young’s modulus of blends𝜙𝜙 or𝐸𝐸 composites𝜙𝜙 𝐸𝐸 based on the properties of

neat materials. Here, and represent the modulus and volume fraction of individual

𝑖𝑖 𝑖𝑖 components, respectively.𝐸𝐸 Both𝜙𝜙 the neat and filled blends illustrated higher modulus than

the values predicted by these two models, representing a more effective stress transfer in

presence of co-continuous morphology.102

It was concluded that the co-continuous structure had better mechanical properties than the matrix-dispersion morphology. In addition, there was no substantial difference between samples compounded in two temperatures, so the location of CNF in the composites did not affect the modulus significantly. Regarding to the tensile strength,

PMMA had the highest strength (Figure 22), but there was not much change by adding 16% of CNF fillers. Similarly, there was not much difference between neat blends, but 230-

PMMA/HDPE/16 composites displayed higher tensile strength than 150 °C’s due to the enhanced CNF dispersion.

1000 1400 a b 1200 800 1000

600 800

400 600

Modulus (MPa) 400 Modulus (MPa) 200 200

0 0 PMMA HDPE B-150 B-230 B-parallel B-series PMMA/ HDPE/ B-150/ B-230/ B-parallel/ B-series/ CNF CNF CNF CNF CNF CNF

Figure 21. (a) Tensile modulus of neat samples and corresponding calculated modulus by parallel and series model, (b) tensile modulus of filled samples and the predicted values from the models.

50 PMMA PMMA/CNF HDPE 40 HDPE/CNF B-150 B-CNF-150 B-230 30 B-CNF-230

20 Strength (MPa) 10

Figure 22. Tensile strength of individual components, neat blends and composites with 16

wt% CNF content.

3.5. Conclusion

In this study, the processing temperature as one of the factors affecting the phase

morphology and location of fillers in the polymer blend composites was investigated. For

this purpose, components of a tertiary blend (PMMA, HDPE and CNFs) were extruded at

150 and 230 °C. The results of EC and TC measurements demonstrated that the composite

containing 16 wt% of CNF that was extruded at 150 °C (150-PMMA/HDPE/16) had a

higher TC and lower EC compared with its counterpart extruded at 230 °C (230-

PMMA/HDPE/16). Moreover, 150-PMMA/HDPE/16 composite was the only sample that

illustrated higher TC than HDPE containing same content of CNF (HDPE/16), whereas,

the percolation threshold for EC occurred at 8 wt% of CNF. The different EC and TC

behavior in these composites was due to different mechanisms of EC and TC. Tunneling

effect of electrons, as the governing mechanism for EC, requires close distance between

the fillers so well distribution of fillers in 230 °C could create a network for electron

tunneling and EC enhancement. While, for TC, well distribution of fillers increased the

interfacial phonon scattering and decreased the TC. As a result, based on AFM and SEM

images, it was confirmed that the processing temperature plays a key role on the

morphology of the blend and distribution of the fillers which can cause different behavior

of TC and EC of the final composite.

3.6. Copyright Notice

In this chapter adapted figures with permission from following publications were used:

“Enhancement of Electrical Conductivity by Changing Phase Morphology for Composites

Consisting of Polylactide and Poly (ε-caprolactone) Filled with Acid-Oxidized Multiwalled

Chemical Society.

CHAPTER IV

DIRECTIONAL XYLITOL CRYSTAL PROPAGATION IN ORIENTED MICRO- CHANNELS OF BORON NITRIDE AEROGEL FOR NON-DIRECTIONAL HEAT CONDUCTION

4.1. Outline

In this study the continuous network of thermally conductive fillers was constructed by utilizing a self-assembly process, known as ice-template method that resulted in formation of aerogels of fillers. Boron nitride (BN) that has a platelet-like shape with high aspect ratio and significant high TC in the in-plane direction was used as the thermally conductive filler. We present results of TC obtained by incorporating xylitol crystals into aligned BN aerogel (BNA), prepared by ice-template method, followed by carbonization.

Carbonization of the BNA was necessary for strengthening of the scaffold. The crystallization of ice facilitates the formation of aligned BN walls in the aerogel and the subsequent filling of xylitol in the microchannels of the BN aerogel pushes the crystallization of xylitol in transverse direction. Combination of aligned crystal packs and

BN walls in these composites results in high TC in both horizontal and vertical directions.

TC of the composites increases with increasing BN content and the TC reaches as high as

4.53 W·m-1·K-1 at BN loading of 18.2 wt%. These new results offer an alternative strategy

to fabricate isotropic thermally conductive composites that can be used for the next

generation of heat dissipating materials.103

4.2. Introduction

Self-assembly process is one of the known method for formation of the thermally

conductive network in a matrix with relatively lower content of fillers comparing to the

traditional methods. Aerogels as a light weight porous network are often prepared with a

self-assembly process104 followed by substituting the solvent with air.105 They have

demonstrated exceptional TC enhancement as a continuous network of fillers for the

fabrication of thermally conductive composites.106-107

Among the thermally conductive fillers, platelet-like fillers are preferred due to

their high aspect ratio and low percolation threshold. The ordering of such fillers directly

impacts the formation of filler network and phonon pathways.108-109 These fillers are anisotropic in nature with higher TC in the in-plane direction, so it is better to align the fillers along the direction of heat flow to maximize TC.110-111 Methods such as doctor blade

casting112, electrospinning113, magnetic field alignment110, injection molding114 etc., have

been used for aligning fillers mostly in the horizontal direction. Ice-template method,

known as ice segregation induced self-assembly104, is a simple and low-cost approach that has attracted attention for vertically orienting the fillers along the direction of ice crystal growth.115-116 As a result, utilizing this approach results in formation of aligned walls of fillers within the aerogel.

For applications such as the thermal management in the electronic devices, ceramic fillers are usually preferred because of their high TC and electrical insulation.111-

112, 117 Hexagonal boron nitride (h-BN) that is analogous to graphite and known as the white graphite; is a typical platelet-like thermally conductive ceramic filler.118 Due to the high

intrinsic TC and electrical insulating properties, h-BN has been used for TC enhancement

of electrically insulating polymer composites. It has been demonstrated that by

incorporation of exfoliated BN (ex-BN) from h-BN powder, TC is enhanced more effectively. Decreasing the number of BN layers decreases the interlayer phonon-phonon scattering (the dominant phonon scattering mechanism in h-BN) which in turn results in higher TC.[119] Hence, the formation of aligned walls of ex-BN in an aerogel can further

enhance TC of the scaffold and ultimately the corresponding composites.

In addition to the properties of the thermally conductive fillers and their alignment,

the properties of the filling material that is incorporated into the scaffold plays a significant

role in the TC of the composite.2 For semi-crystalline polymers, it has been demonstrated that because of the alignment of polymer chains in crystalline region, phonons can be transferred with less scattering.4, 120 In general, a higher degree of crystallinity leads to higher achievable TC.12 Also, the presence of heterogeneous nucleating agents can facilitate the crystal nucleation and affects the orientation of the crystals.121-122 Sugar alcohols (SAs) have high crystallinity123-124 and consequently higher inherent TC than most

of polymers. Xylitol is one the most usual SA, commonly found in food and pharmaceutical

industries. Further, xylitol has a relatively higher TC among the other SAs and at the same

time provides a good handle to manipulate its crystalline structure which can significantly

affect phonon transport.125 Recently, the ex-BN aerogel composites were fabricated with

combining ice-template and vacuum infiltration methods.

The majority of existing studies focus on the preparation of composites of BN

aerogel filled with polymers such as epoxy109, poly (ethylene glycol) (PEG)126 and

paraffin127, and relatively high TC of 4.42, 2.36 and 1.33 W·m-1·K-1 were achieved. Further,

it should be considered that the crystallinity of filling materials could be affected especially

within the confined channels of the aerogel. This leads to two fundamental questions: (1)

what is the role of the aerogel microchannels on the crystalline behavior of filled materials?

(2) How does the crystallization of the filling material impact the overall thermal

conduction? With these two questions, xylitol was selected as the filling material in the ex-

BN scaffold and a series of composites were prepared. The effects of xylitol crystal packs

on TC and TC anisotropy of ex-BN aerogels (BNA)/xylitol were investigated and

characterized. TC results of BNA/xylitol composites were compared with their counterpart

epoxy-filled composites. These results offer a new perspective into thermally conductive

composites that are likely to be useful for designing the next generation of heat dissipating

materials.

4.3. Experimental Procedures

4.3.1. Materials

High viscosity grade SCMC and xylitol from Sigma-Aldrich were used. The epoxy resin was a mixture of Epon resin 828 and Jeffamine T403 (as the curing agent) from

Hexion Inc. and Huntsman, respectively. The BN (PCTF5) used was provided by Saint

Gobain Ceramics. All the materials were used as received. Deionized (DI) water with minimum resistivity of 18.2 MΩ was used throughout the experiment.

4.3.2. Sample Preparations

To prepare the BN aerogel, BN powder was dispersed, exfoliated and then assembled by ice-templating method. Specifically, a certain amount of BN powder (16-40 g) was firstly added to 100 mL of 1 wt% SCMC aqueous solution. The dispersion was mechanically mixed for 10 minutes and ball milled using Emax ball milling (Retsch,

Germany) for 6 h at 600 rpm. The obtained viscous slurries were poured into a small beaker

and placed on a copper rod that was partially immersed in liquid nitrogen. During this

process, the growth of the ice crystals from bottom to top resulted in a scaffold with aligned

ex-BN walls. The frozen slurries were then freeze-dried (Labcono, FreeZone 4.5, under -

44 °C and 0.036 Torr) for 48 h and scaffolds with different contents of ex-BN were obtained. The presence of SCMC in this work was necessary to bind ex-BNs and enhance

the mechanical strength of the aerogels. The aerogels prepared with BN loading of 16, 20,

32 and 40 g were named xBNA (x=1, 2, 3 and 4, respectively). These samples were then

carbonized under an inert environment at 800 °C for 2 h with a heating rate of 5 °C min-1.

The carbonized samples are referred to as xCBNA (x=1, 2, 3 and 4, respectively) in this

work.

To fill the scaffolds with xylitol, the neat scaffolds were submerged into molten

xylitol at 110 °C followed by applying vacuum. Similarly, the neat 4BNA and 4CBNA

scaffolds were submerged in a mixture of epoxy and hardener (2.3:1.0) and went through the vacuum infiltration process. Samples were then removed from the vacuum oven and solidified at room temperature. The detailed process for fabrication of these composites is

depicted in Figure 23. The composites of xylitol and xCBNAs are referred to as Xy-

xCBNA (x= 1 to 4) while the epoxy filled 4BNA and 4CBNA are referred to as Ep-4BNA

and Ep-4CBNA respectively.

Figure 23. Schematic illustration of fabrication process of Xy-CBNA composites.

4.3.3. Characterization

The cross-section morphology of the samples was characterized by SEM, Hitachi

tabletop electron microscope TM3030Plus operating at 10 kV. The X-ray powder

diffraction (XRD) analysis was carried out with a Bruker AXS D8 Discover diffractometer with GADDS (General Area Detector Diffraction System) operating with a Cu-K α radiation source filtered with a graphite monochromator (λ = 1.541 Å). Fourier-transform infrared spectroscopy (FT-IR) characterization was performed using Perkin Elmer ATR

FT-IR. The content of ex-BN in each sample was determined by Thermogravimetric

Analysis TGA (TGA Q50, TA instrument) with air atmosphere. Thermal conductivity was measured using C-Therm TCi Thermal Conductivity Analyzer. For this purpose, samples with 2.5 cm diameter were placed on TCi sensor. FLIR thermal camera E40 was used to record the thermal images during heating of the composites. For this purpose, samples were placed on a hot plate that was pre-set at 70 °C followed by recording thermal images every

10 seconds. The micro-CT scanning was carried out with a Skyscan1172 (80 kV X-ray

energy, image rotation of 0.53, and Al filter) on a rectangular cuboid sample with 2.0 mm

thickness.

4.4. Results and Discussion

The cross-sectional microstructures of the BN aerogels are shown in Figure 24. As

can be seen, ex-BN walls are aligned along the ice crystal growth direction. As shown, ex-

BN walls were aligned along the ice crystal growth direction and increasing the ex-BN

loading resulted in a honeycomb-like structure, Figure 24 (c&g). Meanwhile, the interlayer

wall distance continuously decreased. The average distance between the walls was

measured as 21.04 μm for 1BNA and decreased to 20.51, 12.49 and 9.70 μm for 2BNA,

3BNA and 4BNA, respectively. Structural transformation from aligned ex-BN walls to

honeycomb-like structure was reported in previous literatures.128 At relatively low content of ex-BN, fillers are repelled by the ice crystals and aligned in the interspaces of the crystals.

By increasing the filler content, some ex-BN cannot be localized in the crystal interspaces.

This is due to freezing of the dispersion before localization of all ex-BNs in the crystal interspaces and hence, they will be trapped within the ice crystal instead. As a result, horizontal ex-BN bridges across the vertical walls start to appear, leading to honey-comb

structures.

Figure 24. SEM cross-section of (a) 1BNA, (b) 2BNA, (c) 3BNA and (d) 4BNA. Their

respective higher magnification is presented in (e to h).

The vertical alignment of ex-BN walls was further confirmed by XRD results in which the (002) and (100) lattice planes were distinguished by 26.8 and 41.6° peaks in

109 Figure 25. The intensity ratio of these peaks (I002/I100) can be used as an index to compare

the structural alignment. For this purpose, XRD was conducted on a CBNA scaffold and

CBNA powder, which was prepared by grinding the scaffold with silicon carbon paper

(grits 120). This I002/I100 ratio decreased from 42.42 (CBNA powder) to 1.43 (CBNA

scaffold), confirming the increased alignment of ex-BNs along the growth direction of ice

crystals. Such vertical alignments are favorable for driving phonon transport, which

ultimately leads to higher TC.

Figure 25. XRD patterns of powdered CBNA and CBNA scaffold.

It is known that BN is susceptible to hydrolysis under milling/sonication

processes.129-130 By adding SCMC in the milling solution, a good binding between

hydrolyzed ex-BN sheets and SCMC could be expected and thus a stable mixture after

milling. This is critical to control the uniformity of the aerogel. However, the structural

integrity of ex-BN/SCMC is not strong enough to sustain compressive forces when

exposed to infiltration media. The mechanical strength of the scaffold becomes even more

critical for highly viscous infiltration media. Therefore, the aerogels were annealed in an

inert environment to convert SCMC into carbon with the aim to enhance the bonding

between ex-BN sheets. Figure 26 (a-e) shows the aerogels could support 500 g weight

before and after annealing. In addition, the annealing process did not change the dimension

of the aerogels, as shown in Figure 26 (f&g). The annealing process eliminated the

hydroxyl groups from the aerogel, which was confirmed with FT-IR. As shown in Figure

26 (h), the O-H stretching peak at 3337 cm-1 of for BNA was disappeared after

carbonization. When subjected to xylitol infiltration into the scaffold, BNA was collapsed,

whereas CBNAs remained intact, demonstrating the strengthened scaffold structure by the

annealing process.

Figure 26. (a) BNA supports 500 g weight, (b-e) CBNAs support 500 g weight, (f) and (g)

compare the size of BNA and CBNA, (h) FT-IR spectra of BNA and CBNA, showing the

presence and the absence of O-H stretching peak of BNA and CBNA, respectively, (i)

filling BNA with xylitol that led to cracked scaffold, (j) Xy-xCBNA composite that

remained intact after xylitol infiltration.

After filling the CBNAs with xylitol, the content of ex-BN in each filled scaffold

was determined by TGA, Figure 27. Since xylitol mostly decomposed at 500 °C, the

percentage of final residues is ex-BN in each composite. The carbon residue in the

composites can be ignored due to its very low weight ratio (< 1.0 wt%) in the scaffold,

especially after annealing. The ex-BN content in Xy-1CBNA, Xy-2CBNA, Xy-3CBNA and Xy-4CBNA was determined to be 6.0, 10.1, 17.2 and 18.2 wt%, respectively.

100 Xy-4CBNA 80 Xy-3CBNA Xy-2CBNA Xy-1CBNA 60 Xylitol

Weight percentage (%) percentage Weight 20

0 100 200 300 400 500 600 700 Temperature (oC)

Figure 27. TGA thermogram of xylitol and Xy-xCBNA composites with x= 1 to 4.

The inner structures of these composites after infiltrating xylitol were investigated

with SEM. It is observed in Figure 28 (a-d) that the scaffolds were all filled with the aligned crystals of xylitol. It is known that SAs such as mannitol131-132 and sorbitol132 crystallize in

the form of spherulites from the melt.123 Their spherulites are composed of fibril needles

that grow in radial directions.131, 133 Further, it is known that impurities or surface defects

play a key role in the formation of crystals and spherulites.133 For instance, some studies

have shown that the fillers in polymer composites can act as heterogeneous nucleating

agents and facilitate the crystal nucleation 121, 134-135 due to their large specific surface areas

and high aspect ratio.136-137 It has also been demonstrated that fillers can affect the growth

53 direction of crystals and cause both radial spherulitic and transcrystalline growth of polymers that are perpendicular to the surface of fillers.121, 134 Huang et al. have studied the crystallization of poly (L-lactic acid) (PLLA) in the confined spaces between the graphene oxide nanosheets (GONs). They have reported that in a dense GONs network (> 4 wt%), the distance between the GONs layers significantly decreased and resulted in heavily confined growth of PLLA lamella in a two-dimensional growth mode (Figure 29).135

Similar in this work, the ex-BN walls acted as heterogeneous nucleating agents and facilitated the growth of xylitol crystals. As shown in Figure 28, the crystal packs are aligned perpendicularly to the aerogel walls. Therefore, the achieved composites comprise both the aligned ex-BN walls as well as crystal packs of xylitol.

Figure 28. Cross-section SEM visualization of Xy-1CBNA (a), Xy-2CBNA (b), Xy-

3CBNA (c), Xy-4CBNA (d). The crystals are perpendicular to the ex-BN walls in all the samples.

Figure 29. The PLLA crystal growth in the confined spaces between GONs. 135

The crystal structure of xylitol in each composite was examined further by XRD.

Error! Reference source not found. shows the XRD patterns of xylitol, CBNA and all

the composites. The characteristic diffraction peaks of CBNA were detected at 2θ = 26.9

and 41.9o and the characteristic peaks of xylitol were detected at 2θ = 17.8, 20, 22.6, 24.8,

31.7, 35.6, 38 and 38.6o. The XRD patterns of the composites include all the major peaks

of xylitol and CBNA without the appearance of any new peaks. This indicates that the

components are only in physical contact and are in equilibrium.138 Furthermore, due to the absence of peak shifts, it can be concluded that the crystal forms of neat xylitol and xylitol filling the microchannels are the same.139

Figure 30. XRD patterns of xylitol, CBNA and Xy-xCBNA with x= 1 to 4.

Further, the crystal size of xylitol inside the scaffold was calculated using Debye

Scherrer’s equation 9:

. = . (9) 𝜅𝜅 𝜆𝜆 𝐿𝐿 𝛽𝛽 𝑐𝑐𝑐𝑐𝑐𝑐 𝜃𝜃 Where k is the Scherrer’s factor (~ 0.9), λ is the Cu K α beam (1.54 °A) and β is the measured full width at half maxima (FWHM) at the Bragg angle 2θ.140 The crystal sizes of

xylitol in each composite, which are summarized in Table 5, are the average of calculated

sizes from the major peaks mentioned earlier. Size of xylitol crystal was reduced

continuously with decreasing the distance between the ex-BN walls. The smallest crystal

size was obtained in Xy-4CBNA that had the smallest wall distance. This phenomenon can

be explained by the confinement effect of the walls. The reduction of inter-wall distance prevented the xylitol crystals from growing further, which resulted in smaller crystals.

Table 5. The relation between the ex-BN content, the wall distance and the crystal size of

xylitol in Xy-xCBNA with x = 1 to 4.

Xylitol crystal size ex-BN walls Sample BN (wt%) (nm) distance (μm)

Xylitol - 36.01 -

Xy-1CBNA 6.01 36.20 21.04

Xy-2CBNA 10.07 34.91 20.51

Xy-3CBNA 17.15 33.56 12.49

Xy-4CBNA 18.16 29.99 9.70

Figure 31 shows the through-plane TC of Xy-xCBNA composites and Ep-4CBNA,

as well as the in-plane TC of Xy-1CBNA and Xy-3CBNA. The through-plane TC of Xy- xCBNA increased with an increase of ex-BN content and TC reached as high as 4.53 W·m-

1·K-1 at 18.2 wt% ex-BN loading. The ex-BN walls inside the composites acted as phonon

pathways and, hence, increasing the content of fillers improved phonon transfer by creating

a more effective phonon transfer network. To investigate the role of xylitol crystal packs

on the TC of the composites, 4CBNA was filled with epoxy and its through-plane TC was

measured as 1.73 W·m-1·K-1. It is apparent that the incorporation of xylitol in 4CBNA

scaffold enhances TC by 2.6 times compared to the one filled by epoxy. Moreover, the in-

plane TC of Xy-1CBNA and Xy-3CBNA were very close to their through-plane values showing that xylitol filled aerogels had isotropic TC property, which could not be achieved from traditional approaches. It is suggested that the orientational crystallization of xylitol within the scaffold counterpart the preferential heat transfer along the CBNA scaffold. The vertically packed xylitol crystals perpendicular to the BN sheet orientation provided transverse phonon transport pathways together with the BN framework. Therefore, the heat transfer from both parallel and vertical directions was simultaneously improved, benefiting from the directional pattern of the CBNA scaffold as well as the orientational growth of xylitol crystals.

5.0 In-plane ) 4.5 -1 Through-plane K ⋅

-1 4.0 m ⋅ 3.5 W ( 3.0

2.5 2.0

1.5 Ep-4CBNA 1.0

Thermal conductivity/ 0.5 0 5 10 15 20 ex-BN content/wt% Figure 31. The through-plane TC of Xy-xCBNA with x = 1 to 4, the in-plane TC of Xy-

1CBNA and Xy-3CBNA and the through-plane TC of Ep-4CBNA.

Heat dissipation performance of xylitol-filled and epoxy-filled scaffolds was further characterized by thermal camera. The time dependent heat propagation was monitored in the through-plane direction of Ep-4CBNA, Ep-4BNA, and Xy-4CBNA. For

comparison, heat propagation along in-plane direction of Xy-4CBNA composites was

characterized as well. The thermal images, which were recorded every 10 seconds, of these

composites were analyzed afterwards, Figure 32 (a-d), (a’-d’) and (a”-d”). Figure 32(e) summarizes the results by plotting the temperature vs. time profile. As can be inferred from the plot, Xy-4CBNA had a steeper slope, indicating a faster heat transfer process. It also attained a higher equilibrium temperature compared to both Ep-4CBNA and Ep-4BNA, which is in good agreement with the TC results in Figure 31. Moreover, it is observed that

Ep-4CBNA had a steeper profile and higher equilibrium temperature than Ep-4BNA, 59

which further suggested that the annealing process positively contributed to the thermal

conduction in the composites.

Figure 32. FLIR images of the through-plane direction of EP-4CBNA (a, a’ & a”), EP-

4CBNA (b, b’ & b”) and Xy-4CBNA (c, c’ & c”), and the in-plane direction of Xy-4CBNA

(d, d’ & d”) at different times. (e) The temperature-time profile of the through-plane

direction of EP-4CBNA, EP-4BNA and Xy-4CBNA, and the in-plane direction of Xy-

4CBNA.

It is known that platelet-like fillers such as boron nitride nanosheets (BNNS) have

anisotropic TC property.2 In addition to the inherent anisotropic TC of fillers, their

alignment and conformational plane-plane contact would effectively decrease the

interfacial thermal resistance and generate efficient heat transfer along the direction of

alignment.141 Therefore, anisotropic TC properties can be expected. In earlier studies,

fillers have been aligned parallel to the flow direction (Table 6) leading to high TC along

in-plane direction. Hence, these composites are more focused on spreading heat in the in-

plane direction than the through-plane heat extraction. Therefore, these composites are not

ideal candidates for applications that requires isotropic heat dissipation.142. The obtained

composites in this study have demonstrated unique features when compared with the

previously reported composites: Firstly, they had similar or higher TC with much lower

content of fillers; secondly, they had high TC in both in-plane and through-plane directions.

The isotropic TC values in epoxy-filled BNAs with high content of ex-BN was reported previously and it was explained by the formation of honeycomb-liked structure of BNAs.128

In this work, similar in-plane and through-plane TC properties was achieved with much

lower ex-BN content of 6.01 wt% in Xy-1CBNA. At such a low ex-BN content, aligned wall structure was formed rather than honeycomb-like structure, as shown in Figure 24

(a&e). Crystal packs of xylitol were likely responsible for the isotropic TC property.

Therefore, xylitol crystal packs offered dual advantages of channeling the efficient phonon transport for enhanced TC as well as the isotropic TC property in the composites.

Table 6. Summarized literatures of the fabricated composites with either isotropic or anisotropic TC.

TC (W·m- TC TC (W·m- Filler Alignment 1·K-1) ┴ Re Matrix Fillers measureme 1·K-1)║ to content technique to f nt technique alignment alignment Epoxy Ice-template Laser flash BNNS 9.29 vol% 2.85 2.4 128 resin method technique Poly (vinyl Thermowave BN nanotube alcohol) 10 wt% Electrospinning 0.54 0.27 143 (BNNT) analyzer (PVA) Magnetically Silicone Magnetic field Laser flash responsive 9.14 vol% 0.354 0.154 144 gel alignment method BN

Laser-flash Vacuum PVA BN 13 vol% diffusivity 8.28 0.63 filtration 145 instrument

Cellulose Vacuum- Steady-state nanofibers BNNT 25 wt% assisted 21.39 4.71 146 method (CNF) filtration Hydrogen Tape casting Laser-flash PVA peroxide 5.9 vol% with doctor diffusivity 3.92 0.44 147 treated BN (10 wt% ) blade instrument Reduced Vacuum- LFA 447 CNF graphene 30 wt% assisted self- 6.168 < 0.072 141 Nanoflash oxide (rGO) assembly Multiwall Epoxy carbon Chemical vapor Infrared (IR) 16.7 vol% 4.87 < 2.41 148 resin nanotube deposition microscope (MWCNT) LW-9389 TIM Tester Epoxy Graphite 11 wt% Lateral flow using steady- 4.5 0.9 149 resin nanoplateles state heat flow technique P-phenylene Transient laser Polystyren diamine 10 wt% Hot press flash 0.244 ~0.02 150 e (PS) (PPD)- technique graphene

NH2–PEG– Polyamide- NH2 Mold 5 wt% Nanoflash 9.71 0.406 151 6 (PA6) functionalize compression d rGO Self- polymerizatio n of tannic acid (TA) and pyrrole monomer Vacuum (PG)on Laser flash PVA 10 wt% filtration 7.81 0.23 152 cellulose technique Self-assembly nanofiber (CNF)- anchored graphene nanosheets (GNs) Ultrahigh molecular weight Mold Laser-flash polyethyle BN platelets 50 vol% 12.42 ~ 1.13 153 compression conductometer ne (UHMWP E) Poly(dially l dimethyl Vacuum- Steady-state ammonium BNNS 90 wt% assisted 200 1.0 154 method chloride) filtration (PDDA) Vacuum- Laser flash CNF BNNS-OH 25 wt% assisted 22.67 1.08 155 technique filtration

Carbothermal reduction chemical vapor deposition Poly (methyl (CR-CVD) methacrylate substitution Temperature- ) BN 56 wt% reaction and wave-analysis 6.75 6.14 156 (PMMA) conversion method (CR-CVD-SR- C) using B2O3 vapor and nitrogen Layer by layer Nanoflash NFC GO 1 wt% 12.6 0.042 157 assembly (LbL) technique High Multistage density Light Flash BN 40 wt% stretching 3.57 0.74 158 polyethyle Apparatus extrusion ne (HDPE) Ball milling, high-pressure Polyimide compression Nano-Flash (PI) BN 30 wt% 2.81 0.73 159 and low- apparatus

temperature sintering Laser flash PI BNNS 7 wt% Solution casting 2.95 0.44 160 apparatus 17.15 Ice-template This Xylitol ex-BN TCi 3.67 3.28 wt% method work

Investigation of the interior structure of xylitol-filled composites with micro-CT

scan slices revealed that the scaffold is mostly filled, without obvious empty voids. Still, a

few micro-cracks were present inside the CBNA, which might be the closed micro-domains

formed during aerogel formation, Figure 33. It is well known that airgaps should be

avoided to achieve high thermal conduction. The measured high TC values of these

composites indirectly proved the efficient filling of the scaffolds. In other words, the

presence of traces of voids in these composites had a minor effect on the overall thermal

conduction.

Figure 33. Micro-CT scan slices of xylitol filled CBNA at different times that shows the

presence of empty hollows inside the composite. These images confirmed that presence of

this much of air inside the scaffold has a minor effect on the bulk TC of the composites.

There are numerous proposed theoretical models available for predicting TC of

composites.2, 161-166 Generally, the theoretical models should consider the intrinsic TC of

the matrix, the loading of fillers, filler properties including particle size, size distribution,

shape and topology, the distribution of fillers162 and the interfacial thermal resistance167-168.

In this work, the structural factors of the filler and filler arrangement must be considered

for better accuracy of the predicting model. To be able to compare the measured values with those predicted using theoretical models, the weight percentage of ex-BN was 64

converted to volume fraction (vol%).The measured TC of xylitol composites were compared with the predicted values from different models (by considering TC of BN = 600

W·m-1·K-1)144 and the results are summarized in Figure 34. Among all these models, the

one proposed by Reinecke et al.169 fitted the best with the experimental results. This model

is a combination of Bruggeman170 and effective medium approximation (EMA)171 models,

which was originally proposed for the systems containing magnetic field induced columnar

microstructure of iron particles. The TC of particle chains (kchain) was calculated with the

Bruggeman model that takes the particle-particle thermal interaction into account. Then, the bulk TC was calculated by substituting the kchain in Nan’s EMA model which

incorporates interfacial thermal resistance. Therefore, in this model the loading of fillers,

the filler alignment, thermal interactions between the fillers, and the filler/matrix interfacial thermal resistance were all considered despite not being typically considered in other models.

6.5 Hatta-Taya )

-1 6.0 Maxwell K ⋅ Agari-Uno

-1 5.5 Lewis-Nielsen m ⋅ 5.0 Bruggeman W

( 4.5 EMA 4.0 Bruggeman-EMA Experimental data 3.5

3.0 2.5 2.0 1.5 1.0 Thermal conductivity/ 0.5 0 2 4 6 ex-BN content/vol%

Figure 34. Predicted through-plane TC of Xy-xCBNA with x= 1 to 4 with different theoretical models.

4.5. Conclusion

In this study, we presented thermally conductive composites which were fabricated

by incorporating xylitol crystals into aligned BNAs that were formed by using the ice- template method. SCMC was introduced to stabilize the ex-BN sheets during exfoliation and scaffold formation, which was then carbonized to strengthen the scaffold and reduce thermal resistance within the scaffold. The heterogeneous nucleation and growth of xylitol crystals within the aerogel micro-channels led to the alignment of xylitol crystal packs perpendicular to the wall alignment. The presence of such patterned xylitol crystal packs provided additional thermal transport pathways in the perpendicular direction of the BN walls and offset the anisotropic TC of CBNA. As a result, high TC values were achieved in both horizontal and vertical directions. In general, the more condensed ex-BN scaffold is beneficial for TC enhancement. Specifically at an ex-BN loading of 18.2 wt%, the TC value reached up to 4.53 W·m-1·K-1. To conclude, directional crystal growth and patterning

in an oriented scaffold provides an effective approach for the development of isotropic heat

dissipation materials, having great potential in the heat management field.

4.6. Copyright Notice

In this chapter adapted figures with permission from following publications were

used:

“Poly (l-lactic acid) Crystallization in a Confined Space Containing Graphene Oxide

American Chemical Society.

A a significant part of this chapter also was adapted with permission from following

publications:

“Directional xylitol crystal propagation in oriented micro-channels of boron nitride aerogel for isotropic heat conduction” originally published in Composite Science and Technology.

CHAPTER V

REGULATING INTERMOLECULAR CHAIN INTERACTION OF BIOPOLYMER WITH NATURAL POLYOL FOR FLEXIBLE, OPTICALLY TRANSPARENT AND THERMALLY CONDUCTIVE HYBRIDS

5.1. Outline

In this work well-defined thermal bridges were formed by tuning the molecular

interaction. For this purpose, a sustainable hybrid comprising of SCMC) and xylitol was

used. With optimized ratio of SCMC and xylitol, enhanced TC of up to 1.75 times of neat

SCMC was achieved in addition to excellent flexibility and optical transparency. These enhancements are attributed to the formation of new H-bonds between SCMC and xylitol molecules resulting in formation of homogenously distributed thermal bridges throughout the polymer matrix. The intermolecular interaction in SCMC-xylitol composites elucidates some of the fundamental factors (e.g. H-bond intensity) responsible for promoted phonon transfer in polymeric materials, and at the same time sheds light on the capability of employing biopolymer-polyol based materials for thermal management applications.

5.2. Introduction

Despite the attractive properties of petroleum-based polymers as foundation of innovative material for thermal management applications, the impact of plastics on environment and eco-system has aroused huge concern nowadays. Therefore, it is of

paramount importance to find green alternatives with proper approaches for their TC

enhancement.

As it was mentioned in the other chapters, the traditional methods for TC

enhancement of polymeric based materials is to incorporate thermally conductive fillers.

However, in terms of scale-up manufacturing, it is often challenging to inject a large

amount of fillers to ensure the formation of such continuous networks because it increases

the cost of material and also the difficulty in processing.17 In addition, the introduction of

new polymer-filler interfaces would increase phonon scattering and adversely affect TC.120,

168, 172 Recently, Kim, et al. reported a significant enhancement of TC in miscible blend of poly (N-acryloyl piperidine) (PAP) and poly (acrylic acid) (PAA) without using any fillers.

Interestingly, the formation of strong H-bonds between PAP and PAA resulted in the

formation of continuous H-bond network inside the matrix and therefore significantly

increased TC (Figure 35).17 It is claimed that such a network can be obtained by fine tuning

the interchain interactions between the polymer chains.173

Using the same strategy, Mehra et al. enhanced the thermal conductivity of PVA with small molecules including water174 and short organic molecules4, 172. It has been

demonstrated that incorporation of small molecules which can form H-bonds with PVA

facilitated the formation of a continuous thermal network inside the matrix followed by

boosting the phonon transfer and thermal conductivity.

17 monomer mole fractions of PAP (ϕPAP).

Inspired by the aforementioned approach and considering the alternative biopolymers instead of petroleum-based polymers, we investigated the effects of interchain interactions between a biopolymeric host and a naturally occurring polyol on the thermal,

mechanical and optical properties of the resulting hybrid film.175 The selected host is

SCMC and the additive is xylitol. SCMC is an anionic linear polysaccharide, which is an

esterified derivative of cellulose176 that is widely used in industries such as food 177,

agriculture and forestry178, textile and paper179, drug and cosmetics180, ceramics, paints and

lacquers, adhesives etc.181 Its wide applicability is largely because of its high viscosity176,

182-183, low cost, non-toxic, non-allergenic184-186, biodegradability183 and its oil and lipid

barrier properties.183, 187 However, the presence of strong H-bonds in SCMC leads to its

poor thermoplasticity, which can be improved by reducing the inter-molecular

interactions.188 Plasticizers are often added to improve the processability and flexibility of

such biopolymers.189 In this work, xylitol as a natural polyol was selected as plasticizer due

to its rich hydroxyl groups190 and relatively high thermal conductivity125. The insertion of such small plasticizers into SCMC could increase the distance between polymer chains and thus their mobility.191 Moreover, it is expected that new H-bonds can be formed between

SCMC and xylitol and thus new thermally conductive networks inside the polymeric

matrix. Here, the effect of xylitol concentration on the mechanical, optical and thermal

properties of SCMC/xylitol hybrids was investigated by various characterization

techniques. The complete green nature of such hybrids with desired multi-functionality

offers a new perspective of utilizing sustainable products for advanced materials design

and applications.

5.3. Experimental Procedures

5.3.1. Materials and Sample Preparation

High SCMC and xylitol were purchased from Sigma-Aldrich. These products were used as received. Deionized (DI) water (Millipore) with minimum resistivity of 18.2 MΩ

71 was used as solvent. To prepare the films, 5.0 g of SCMC was gradually added to 150 mL

DI water while stirring mechanically. After obtaining a transparent viscous solution, the required amount of xylitol was added to the solution and mixed for 30 minutes. Blends with different weight ratios of SCMC: xylitol (5: 0 (SX0), 5: 1 (SX1), 5: 2 (SX2), 5: 4

(SX4) and 5: 5 (SX5)) were prepared. The transparent solutions were poured into glass petri dishes and dried at 55 °C for 5 days to obtain freestanding films. These films were used for all the other measurements.

5.3.2. Characterization Methods

FT-IR characterizations were carried out using Perkin Elmer ATR FT-IR.

Mechanical tests were conducted using ADMET 500 universal testing machine (MTEST

Quattro, USA). Thermal conductivity measurements were carried out using C-Therm TCi

Thermal Conductivity Analyzer. Thermal behavior of the samples was characterized by

TGA (Q50, TA instrument). The transmittance characterization of films was carried out using UV-1800 Shimadzu spectrophotometer in the range of 200-800 nm.

5.4. Result and Discussion

SCMC is a cellulose derivative with sodium carboxymethyl (CH2COONa) groups.176 The large number of hydroxyl and carboxyl groups present in SCMC, as illustrated in Figure 36, is responsible for its hydrophilic nature. Since SCMC and xylitol are both hydrophilic, they are soluble and miscible in water. Xylitol molecules can easily diffuse into the intermolecular space between SCMC chains. When the water is evaporated, the mixture results in a transparent film.

Figure 36. The H-bonds involved in SCMC are illustrated; intrachain H-bonding are shown

in blue and interchain H-bonding are shown in green. The H-bonds between the molecules of SCMC can be replaced by new H-bonds between xylitol and SCMC molecules. The shown configuration is only one of the possibilities in SCMC-xylitol system.

To distinguish the molecular interactions by using FT-IR, the wavelength shift of

a certain functional group should be monitored.192,193 Since SCMC and xylitol are

hydrophilic components with numerous hydroxyl groups, their O-H band shifts are a

reliable indicator for monitoring the changes in interactions of the system. Figure 37 shows

the characteristic peaks associated with the main functional groups in SCMC. The peak at

2961.93 cm-1 is associated with the stretching of C-H bonds while peaks at 1588.12 and

1413.13 cm-1 are ascribed to asymmetry and symmetry stretching of carboxylate groups,

respectively. The band at 1024.51 cm-1 is assigned to the asymmetry stretching frequency

of C-O-C groups. The strong and broad peak at 3318 cm-1 is associated with the stretching

frequency of O-H bonding as well as interchain and intrachain H-bonding.194-195 These H-

bonds are formed between SCMC molecules due to the presence of large number of

hydroxyl and carboxyl groups.173 The O-H bond stretching of xylitol appears at 3431, 3364

and 3290 cm-1 wavelengths.196

Figure 37 (A) shows the FT-IR spectra of the neat SCMC and the SCMC-xylitol

hybrid films with different contents of xylitol. The peaks representing the O-H stretching of SCMC-xylitol system, shown in the highlighted wavelength range 3000-3550 cm-1

appears smooth for samples with xylitol up to 44.4 wt% (SX4). In contrast, for SX5 sample,

there are multiple peaks, which appear at wavelength of 3425, 3360 and 3293 cm-1. As

discussed earlier, these peaks are the finger prints of xylitol. Therefore, it seems that this

concentration of xylitol exceeds the compatibility limit of SCMC-xylitol system.197

Figure 37 (B) shows the O-H stretching peaks at higher magnification. The hydroxyl stretching peak shifted from 3335 to 3268, 3276 and 3288 cm-1 for SX1, SX2 and

SX4 confirming the enhanced strength of hydrogen bonding interactions198. However, this

is in contrast with the plasticizing role of xylitol. Plasticizing action of a plasticizer is

accompanied with weakening of the attraction forces between the polymer chains leading

to flexibility enhancement of the polymer.199 Thus, xylitol is acting as an antiplasticizer in

these concentrations. Antiplasticizing effect or antiplasticization is the effect of low

concentration of plasticizer in which they increase the stiffness of polymers.200 Previously,

Chaudhary has demonstrated the antiplasticizing effect of polyols on the glass transition

(Tg) of starch films. It was found that incorporation of xylitol and glycerol in low

concentration led to increased Tg. This phenomenon was explained by increased

interactions followed by enhanced networking and molecular entanglements.201

Figure 37. (A) FT-IR spectra of SCMC-xylitol films with different xylitol contents (SX0-

SX5 respectively), (B) higher magnification of hydroxyl stretching peaks which is highlighted in part (A).

Finally, the OH peak of the SX5 sample appeared with OH peaks of xylitol at higher wavelength than the corresponding wavelength in pure SCMC. The higher the wavelength, the weaker is the interaction.202-203 This means that the strength of intermolecular hydrogen bonding of SCMC is decreased. Formation of H-bonds between SCMC and xylitol can affect the intra-molecular interaction of individual component191, 201 and at the same time leads to the formation of thermal bridges, which facilitates phonon transfer across the polymer chain.17, 172, 204 The former increases the mobility and flexibility of the macromolecules187, while latter enhances the thermal conductivity of the resulted blend films.17, 172, 174, 204

SCMC, like most of other biopolymers, suffers from poor mechanical properties such as fragility and brittleness that limit its application in several industries. Fragility can be addressed by using plasticizers. The low molecular weight of plasticizers allows them to diffuse into the intermolecular spaces and reduce intermolecular forces. As a result, the free volume is increased resulting in enhanced molecular mobility, and hence making the biopolymer more flexible. When biodegradable plasticizers are used for biopolymers, the hybrids retain their biodegradability. Polyols (e.g. sorbitol, mannitol, xylitol, etc.) are hydrophilic natural plasticizers that can enhance the mechanical properties of the compatible hydrophilic biopolymers such as SCMC.187

Mechanical test was performed to evaluate the effect of xylitol content on mechanical properties of SCMC namely elongation at break (%elongation), tensile strength

(TS) and modulus of elasticity (ME). As shown in Figure 38, introduction of xylitol up to

4.0 g into SCMC does not significantly affect the %elongation of the hybrids, while for

SX5, %elongation is about 2 times of its value for pure SCMC. This phenomenon can be explained by the plasticizing effect of xylitol and weakened intermolecular bonds of SCMC with newly formed H-bonds between xylitol and SCMC. As a result, the polymer chains are less restricted leading to better flexibility.191, 197 The maximum stress and modulus of elasticity of all samples is summarized in Figure 39. It is observed that lower loading of xylitol into SCMC (SX1 and SX2) results in more rigid films which is reflected in both stress and modulus of elasticity. It is well-known that the mechanical properties are mainly affected by inter- and intramolecular interactions between the molecules.205 For SX1, the tensile strength improves from 123 to 153 MPa and its modulus increases from 10.21 to

10.88 GPa. Further addition of xylitol (SX2) decreased the tensile strength to 127 MPa and

the modulus to10.58 GPa which is still higher than strength and modulus of the pure SCMC

film. The increased strength and modulus of SX1 and SX2 films can be explained by the

antiplasticizing effect of xylitol. For SCMC-xylitol films with low content of xylitol, the

hydroxyl groups of xylitol can increase the SCMC-xylitol interaction, networking and molecular entanglement that result in higher strength and modulus.201, 206 On the other hand,

further increasing xylitol content in SX4 and SX5 samples, decreases both the strength and

modulus of the films which means xylitol is softening the films at these concentrations.

Therefore, there is a transition zone for xylitol in the range of 28.5 to 44.4 wt% in which it

switches from antiplasticizer to plasticizer for SCMC.

Figure 38.Stress-strain curves of SCMC-xylitol films SX0- SX5 respectively.

Figure 39.(A) Summary of the effect of xylitol content on the modulus of elasticity (ME) and (B) maximum stress (TS) of SCMC-xylitol films

Figure 40. (A) Before and (A’) after applying force on SX0, SX1 and SX2 samples. These

samples could not be bent and were broken immediately after bending. (B) Sample SX4

could be bent but broke when rolled 6(B’), and the SX5 sample was flexible enough to be

rolled (C).

Figure 40 shows the flexibility of the SCMC-xylitol films. Figure 40 (A&A’) as the before and after applying bending, demonstrate the fragility of SX0, SX1 and SX2 films.

These films could not bend and break immediately when bending force was applied.

Similar phenomenon can be observed in SX4, Figure 40 (B&B’). In contrast, SX5 can be

rolled easily as illustrated in Figure 40 (C).

Figure 41 shows the TC values of all the samples. The TC of neat SCMC is 0.43

W·m-1·K-1 and it improved to 0.75 W·m-1·K-1 (1.75 times) when the xylitol content was increased to 50 wt% (SX5). The formation of thermal conducting pathways between polymer chains that arise because of intermolecular interactions, i.e. H-bonding, among dissimilar molecules has been reported in the literatures.17, 172, 174, 207 For instance,

incorporation of water molecule in PVA matrix resulted in TC enhancement of over 200%.

TC improvement was based on the newly formed hydrogen bonding between the PVA

chains and water molecules.174

Improving TC of polymeric systems with this approach requires a homogenous

distribution of interchain interactions above the percolation threshold.17 Therefore, miscibility of the polymeric host with the additive is a key factor for enhancing TC. If the components are immiscible, the mixture will phase separate leading to agglomeration of polymer chains with increased phonon scattering, and thus lower TC.172 Here, due to

hydrophilic nature of both SCMC and xylitol, H-bond can be formed uniformly within the

mixture and thus improved TC was achieved. .

As it was mentioned earlier, due to presence of hydroxyl and carboxyl groups in

SCMC, abundant intra- and interchain hydrogen bonding is present in SCMC. These

interactions lead to random self-association and rigidity of SCMC. The addition of xylitol

disassociates such intrachain bonds (in SCMC) and forms new H-bonds with SCMC. The

five hydroxyl groups in each xylitol molecule has great capability to interact with SCMC

via H-bonds those are formed between the hydroxyl-hydroxyl groups or hydroxyl-carboxyl groups (Figure 42). As a result, a network of H-bonds through the matrix can be formed.

Meanwhile, antiplasticizer effect of xylitol in low concentration increases the strength of

present interactions followed by increased networking and molecular entanglements.

Although plasticizer effect of xylitol in high concentration is accompanied with weakening

of the interactions, but due to increased density of thermal bridges between the polymer

chains a well-defined thermal network will be formed through the matrix. Therefore, the

loading of thermal bridges through the matrix is a dominant factor over the strength of

interactions, which results in enhancing TC up to 0.75 W·m-1·K-1 with 50 wt% loading of

xylitol.

Figure 41.Thermal conductivity of SCMC-xylitol film increases with the content of xylitol in 5.0 g SCMC. Notice the improvement from 0.43 W·m-1·K-1 (neat SCMC) to 0.75 W·m-

1·K-1 when the mixture 50 wt% xylitol.

Figure 42. Formation of H-bonds between the hydroxyl-hydroxyl groups or hydroxyl- carboxyl groups in SCMC-xylitol system that results in formation of a thermally conductive network within SCMC.

Thermal stability of the hybrids was evaluated by TGA. Figure 43 (A) shows the

thermal degradation of samples under nitrogen environment as a function of temperature.

Decomposition of all the samples in a single step confirms the presence of a strong

interaction between SCMC and xylitol.208-209 The initial decrease in weight of the samples

that can be noticed in the Figure 43 (A) is due to the presence of moisture in the samples.

In the temperature range of 250 to 323 °C, the decrease in weight of SCMC can be attributed to decarboxylation.176, 210 Finally in the temperature range of 373 to 511 °C, the decrease in weight is because the remaining material in the film is converted into carbon residues.210 Since xylitol can be almost fully decomposed at ~500 °C, only SCMC residues remain at higher temperatures. The temperature, Tdmax, at which maximum decrease of weight occurs, is illustrated in the DTGA curves presented in Figure 43 (B). The temperature at which the sample weight reduced by 50% was recorded as T50%. The weight of the samples at two specific temperatures, 300 °C and 500 °C were also recorded as shown in Table 7.

Figure 43. (A) TGA and (B) Derivative TGA (DTGA) thermograms of SX0-SX5 samples and xylitol (Xyl).

Table 7. Tdmax, T50% and weight loss (%) of samples at 300 and 500 °C is shown. Increasing the content of xylitol resulted in faster thermal degradation and higher weight loss.

Weight loss (%) o o Sample Tdmax ( C) T50% ( C) 350 °C 500 °C

SX0 288.0 327 51.4 60.7

SX1 285.0 303 58.1 67.4

SX2 286.0 303 61.8 70.7

SX4 283.0 289 69.9 76.2

SX5 281.0 284 75.1 80.1

Xylitol 275.1 271.6 8.4 6.0

As shown in Table 7, presence of xylitol in the SCMC-xylitol mixtures resulted in reduced T50% meaning that the degradation rate becomes higher. The weight percentage

(%) of the samples at 350 and 500 °C also confirmed that samples with higher content of

xylitol lost more weight than other samples.

It is known that enhanced interaction between the polymer chains decreases the

chains mobility and decomposition rate and enhances the thermal stability.211 However, it

should be noted that the TC of reinforcing agents can also affect the thermal stability.

Higher TC of additives can facilitate the diffusion of heat into the polymeric matrix.212-213

As a result, in low content of xylitol the impact of enhanced heat diffusion is dominant

over the enhanced interactions between the SCMC chains for the thermal stability.

Similarly, increasing the content of xylitol is accompanied with enhancing the heat

84 diffusion and also improving the chains mobility. However the summarized results in Table

7 demonstrate that the maximum decomposition temperature of the SCMC-xylitol film was only reduced by 7 °C.

In spite of varying the xylitol content, all the films that were obtained in this study were optically transparent. No phase separation was observed even when the sample has

50 wt% of xylitol as illustrated in Figure 44. In a polymeric system, phase separation can result in turbid slurries172, exudated drops187 or in xylitol containing systems it can result in crystallization of xylitol191. The absence of these phenomena in our system demonstrates sufficient compatibility between SCMC and xylitol even when the content of xylitol is high.

Figure 44 shows that the increase of xylitol content can enhance the film

transparency up to ~ 93% in SX5 for the entire range of wavelength. In addition to the light

barrier properties of the plasticizers214, they can affect the polymer chain compaction and

help to pass the light through polymeric films.215 In this work, the increased transparency

of the SX5 film could be attributed to less compactness of SCMC chains.

5.5. Conclusions

Here, the thermal conductivity, optical transparency and flexibility of SCMC,

which is a known biopolymer was successfully enhanced by engineering its interchain

interactions. These enhanced properties were achieved with incorporation of xylitol, a

natural polyol with 5 hydroxyl groups. As demonstrated, by increasing the content of xylitol thermal conductivity increased from 0.43 to 0.75 W·m-1·K-1, which represents a

1.75 times improvement compared to neat SCMC. TC enhancement at low concentration

of xylitol was due to both strong H-bonding interactions and the newly formed thermal bridges between the polymer chains. At higher concentrations of xylitol, though H-bonding interactions were weakened, but resulted in increased number of thermal bridges. As a result, a well-defined thermal network was formed through the matrix. In addition to the thermal conductivity, xylitol affects the mechanical property of SCMC in two different ways. At low concentration, xylitol increases the brittleness of the SCMC films due to its antiplasticizing effect, reflecting strong H-bonding interactions between the components while at higher concentrations, it acts as a plasticizer and enhance the flexibility of the films. Even though xylitol increased flexibility, the maximum decomposition temperature of the SCMC-xylitol film was only reduced by 7 °C. In summary, incorporation of xylitol effectively enhances the thermal conductivity and flexibility of SCMC with a negligible

change in the maximum decomposition temperature. These materials provide a promising

approach for the development of biodegradable materials for flexible electronics in thermal

management applications.

5.6. Copyright Notice

In this chapter adapted figures with permission from following publications were used:

Also a significant part of this chapter was adapted with permission from following publications:

“Regulating Intermolecular Chain Interaction of Biopolymer with Natural Polyol for

Flexible, Optically Transparent and Thermally Conductive Hybrids” originally published in

CHAPTER VI

COMBINATION OF ERYTHRITOL CRYSTAL PROPAGATION IN ORIENTED BORON NITRIDE AEROGEL FOR NON-DIRECTIONAL HEAT CONDUCTION AND PHASE CHANGE PROPERTIES

6.1. Outline

Being inspired by the investigation discussed in the chapter IV, erythritol, as another sugar alcohol with high crystallinity and relative high TC was selected to fill the aligned BN aerogels. Since erythritol has high solidification rate that resulted in releasing the stored heat in a reasonable time it can be a potential phase change material (PCM) as well. Here, we present the results of TC of the obtained composites comprising erythritol and carbonized boron nitride aerogel (CBNA), and their PCM properties. Similar to xylitol, filling the microchannel of CBNAs with erythritol resulted in crystallization of erythritol in the perpendicular direction of the BN walls and consequent isotropic TC in low content of BN. TC was increased as a function of BN loading and at loading of 15.5 wt% TC reaches 2.77 W·m-1·K-1 value. In addition to high TC, heterogeneous nucleation effect of

BN walls resulted in releasing more heat comparing to the neat erythritol. This phenomenon is the result of crystallization of more amorphous phase of erythritol in presence of heterogeneous nucleates. The results of this study are useful for fabricating

PCM composites with high TC and capability of storing and releasing large amount of energy.

6.2. Introduction

Renewable resources of energy are of great importance because of the high demand

of energy and negative environmental effects and limited supply of nonrenewable sources.

Among different types of natural energies, thermal energy is extensively distributed in solar

radiation, geothermal energy, etc. Although abundant thermal energy is available during

day time but most of this energy is wasted. Therefore, storing the generated heat energy

and releasing it when it is demanded has attracted great attentions.216 Among different

types of thermal energy storage, latent heat plays a key role due to the high storage density

and small difference between storing and releasing heat temperatures.217 Phase change

materials (PCMs) are the kind of materials that are used for storing latent heat, which is

accompanied with a phase change at a constant temperature.218 Using these materials could improve the efficiency of storage and reusage of energy.219 The application of PCMs such

as construction, air conditioning, textile, refrigeration and waste hear recovery is a function

of their phase change temperatures.220-223 Among the organic solid-liquid PCMs, natural

polyols have gained attentions because of their high phase change enthalpy values, low

cost, non-toxicity224, and their melting temperatures that are suitable for medium

temperature applications.225 Polyols have been used as PCMs for the first time in the patent

of Guex et al.226 and Hormansdorfer227.125 Recently, erythritol with latent heat of more than

300 kJ/kg and melting point of ~ 119 °C has been studied for the applications such as heat

energy recovery of industrial heat waste.228-230 However, organic PCMs including

erythritol suffer from low TC that affects their overall thermal performance. Selective TC

enhancement of PCMs improves the charging and discharging rate of heat. Therefore,

selective TC enhancement is one of the main focuses in the field of PCMs.

The importance of 2D fillers for creation of continuous network of filler has been

discussed in detail in previous chapters. Besides, as it was mentioned earlier, due to intrinsic high TC and electronic insulation of BN it is preferred to be used in the applications, which require exclusively TC e.g. thermal management of electronic devices.

Similar to the previous study, here CBNAs with two different content of BN were fabricated. First, BN was probe-sonicated, then ice-template method was applied and finally carbonization of BNAs was carried out.

Erythritol which is a high crystalline material with relatively high TC and better

PCM properties than polymers, was selected to fill the fabricated CBNAs. The effect of

BN scaffold on the thermal performance of erythritol including thermal stability, TC, heat storage and release were investigated and characterized. Results shed light on the role of heterogeneous nucleating agents and the alignment of thermal conductive fillers on the overall performance of obtained PCM composites.

6.3. Experimental Procedures

6.3.1. Materials

High viscosity grade SCMC and erythritol from Sigma-Aldrich and Jedwards

International, Inc, respectively were used. The BN (PCTF5) used was provided by Saint

Gobain Ceramics. All the materials were used as received. Deionized (DI) water with minimum resistivity of 18.2 MΩ was used throughout the experiment.

6.3.2. Sample Preparations

To prepare the BN aerogel, BN powder was dispersed, exfoliated with probe sonication and then assembled by ice-templating method. Specifically, a certain amount of

BN powder (24 and 32 g) was firstly added to 100 mL of DI water. The dispersion was 90

mechanically mixed for 10 minutes and probe sonicated using 300W Ultrasonic Processor

(MSK-USP-3N-LD model) from MTI Corporation for 1 h at 50% amplitude. Then 1.0 g

of SCMC was added to the dispersion and mixed mechanically for 2 hours. Then the

obtained viscous slurries were poured into a small beaker and placed on a copper rod that

was partially immersed in liquid nitrogen, which resulted in scaffolds with aligned ex-BN

walls. The frozen slurries were then freeze-dried (Labcono, FreeZone 4.5, under -44 °C and 0.036 Torr) for 48 h and scaffolds with different contents of BN were obtained (BNAs).

These samples were then carbonized under an inert environment at 800 °C for 2 h with a

heating rate of 5 °C/min. The carbonized samples that were prepared with initial BN

content of 24 and 32 g are referred to as 1CBNA and 2CBNA, respectively in this work.

To fill the scaffolds with erythritol, the neat scaffolds were submerged into molten

erythritol at 130 °C followed by applying vacuum. Samples were then kept inside the oven

and solidified gradually to the room temperature. The composites of erythritol and CBNAs

are referred to as Ery-1CBNA and Ery-2CBNA.

6.3.3. Characterization

The cross-section morphology of the samples was characterized by SEM, Hitachi

tabletop electron microscope TM3030Plus operating at 10 kV. The X-ray powder diffraction (XRD) analysis was carried out with a Bruker AXS D8 Discover diffractometer with GADDS (General Area Detector Diffraction System) operating with a Cu-K α radiation source filtered with a graphite monochromator (λ = 1.541 Å). The content of BN in each sample was determined by Thermogravimetric Analysis TGA (TGA Q50, TA instrument) under air atmosphere with 20 °C/min rate. TC was measured using C-Therm

TCi Thermal Conductivity Analyzer. For this purpose, samples with 2.5 cm diameter were

placed on TCi sensor. FLIR thermal camera E40 was used to record the thermal images

during heating of the composites. For this purpose, samples were placed on a hot plate that was pre-set at 140 °C followed by recording thermal images every 10 seconds. The melting and solidification points, and the latent and solidification heats of these samples were

measured using a Differential Scanning Calorimeter (DSC Q200, TA instrument). The

heating-cooling rate was 2 °C/min. All experiments were performed under nitrogen with a

flow rate of 50 mL/min. The enthalpies of each phase change were determined by the

integration of the phase transition peaks in Trios software. To investigate the chemical

stability of samples under nitrogen, 20 heating-cooling cycles with 20 °C/min rate was

carried out on samples and then their DSC curves were recorded with 5°C/min ramping

rate.

6.4. Results and Discussion

Since samples went through carbonization with the aim of enhancing the

hydrophobicity of the scaffold and enhancing its tolerance for a hydrophilic matrix, it was

essential to investigate the effect of carbonization process on the internal structure,

92 mechanical tolerance and size of scaffold.

Figure 45 (a&b) show the internal structure of a scaffold before and after carbonization, respectively. Both images confirm the alignment of BN particles and there is no significant change in their arrangement after carbonization. Moreover, comparing

the size of scaffolds, BNAs and CBNAs

Figure 45 (a’&b’), shows no change, and both the scaffold could tolerate 500 g weight before and after carbonization,

Figure 45 (c&d, and e&f), meaning that after carbonization the scaffold could still tolerate ~0.007 MPa pressure.

Figure 46 (a&b) demonstrates the cross-sectional microstructure of 1CBNA and

2CBNA, respectively. These figures and their corresponding larger magnification images,

Figure 46 (a’&b’), demonstrates the alignment of BN walls along the ice crystal growth direction and also increasing the thickness of BN walls with increasing the BN loading. Additionally,

Figure 46 (a’&b’) prove the presence of connection between the parallel walls at some points. This phenomenon can be attributed to the freezing of dispersion before localization all the fillers in the ice crystals interspaces. The BN walls remained intact after erythritol infiltration as is shown in

Figure 46 (a’’&b’’). Therefore, it can be concluded that the scaffold are sufficiently strong to tolerate the infiltration and crystallization of erythritol.

Figure 46. Internal microstructure of 1CBNA (a) and 2CBNA (b), their corresponding higher magnification images (a'&b'), and internal structure of filled scaffolds, 1CBNA (a”) and 2CBNA (b”).

Figure 47 shows the XRD patterns of erythritol, the scaffold, and the obtained composites. The fabricated scaffolds showed characteristic peaks of BN at 2θ= 26.8 and

41.6° and erythritol showed numerous peaks with the intense ones at 2θ= 14.8, 19.7, 20.3,

24.6, 27.9, 28.4, 29.7, 31.7, 32.8, 34.4, 37.5, 39.9, 41.2 and 42.7°. The obtained results for

both the scaffold103 and erythritol231 are consistent with previously reported values. The

XRD diffractograms of the composites contains the characteristic peaks of both BN and erythritol with neither peak shift nor new peak formation, which confirms the physical interaction between these components.231

Ery-2CBNA Ery-1CBNA CBNA Erythritol

10 20 30 40 50 2θ (deg) Figure 47. XRD diffractograms of xylitol, CBNA, Ery-1CBNA and Ery-2CBNA.

The content of BN in each composites were measured with TGA (Figure 48 (a) and

Table 8) revealing that Ery-1CBNA and Ery-2CBNA contain 13.8 and 15.5 wt% BN,

respectively. Besides, the Tdecomposition-onset was decreased by increasing the BN loading due to higher TC of BN that leads to facilitated heat diffusion inside the composite.212, 232 Hence,

it can be concluded that increasing the content of BN is accompanied with TC enhancement

and decreasing the decomposition initiation temperatures.

The TC values and trend (through-plane) of the composites as a function of BN loading, and the in-plane TC value for Ery-1CBNA is shown in Figure 48 (b). Through- plane TC values of Ery-1CBNA and Ery-2CBNA composites with 13.8 and 15.5 wt% of

BN were measured as 1.93 and 2.77 W·m-1·K-1, which are 2.24 and 3.22 times of erythritol

TC, respectively. These results confirm the significant role of BN scaffold as the phonon transport pathways on TC enhancement of the obtained composites. Besides, the in-plane

TC value of Ery-1CBNA composite is very close to its through-plane value confirming the isotropic TC of this composite even at low content of fillers. This result is consistent with the results of chapter IV, which showed the beneficial effect of xylitol crystal packs on TC enhancement and gaining isotropic TC in xylitol/CBNA composites.103 Similarly,

erythritol creates crystals packs (Figure 48 (c&d) in perpendicular direction of the BN walls that caused the isotropic and high TC values. The heat dissipation performance of the Ery-

1CBNA and Ery-2CBNA composites was further investigated with thermal camera shown

in Figure 48 (e-g). The time dependent heat propagation in through-plane direction of the

samples was recorded and the temperature-time profile was plotted for each composite,

Figure 48 (g). Ery-2CBNA was found to have higher temperature than Ery-1CBNA during

entire time of measurement and subsequently reached higher equilibrium temperature. This

confirms faster heat transfer in Ery-2CBNA, which in turns means higher TC than Ery-

1CBNA.This result is in good agreement with the TC values reported in Figure 48 (b),

signifying the beneficial effect of higher BN loading on TC enhancement.

Figure 48. TGA thermogram of neat erythritol, Ery-1CBNA and Ery-2CBNA composites

(a), though-plane TC of erythritol, Ery-1CBNA and Ery-2CBNA as a function of BN

loading and in-plane TC of Ery-1CBNA (b), cross-section SEM visualization of

composites showing the perpendicular direction of xylitol crystals versus BN walls (c&d),

FLIR images of the through-plane direction of Ery-1CBNA (e, e' & e") and Ery-2CBNA

(f, f' & f") at different times, and their corresponding temperature-time profile (e)

100

Table 8. The loading of BN in Ery-1CBNA and Ery-2CBNA composites and their corresponding Tdecomposition-onset versus neat erythritol.

Sample BN loading (%) Tdecompositon-onset (°C)

Erythritol 0 266.14

Ery-1CBNA 13.8 240.92

Ery-2CBNA 15.5 238.04

Figure 49 (a) show the DSC curves of eythritol, Ery-1CBNA and Ery-2CBNA composites for the heating process. Based on these results (summarized in Table 9) erythritol has melting point of 120.24 °C with 346.06 J/g latent heat. Ery-1CBNA

101

showed melting point of 118.99 °C with 264.86 J/g latent heat and Ery-2CBNA had melting point and latent heat of 118.72 °C and 258.92 J/g, respectively. Therefore, increasing the BN content in the composites led to both lower latent heat and melting point. Decreasing the latent heat of the composites with increasing the BN is due to decreasing the erythritol content and consequently less heat storage ability. However, high TC of BN that facilitates the heat diffusion within the composites is responsible for the decrement of melting temperatures and Ery-2CBNA with higher content of BN and higher TC showed lower melting point,

102

Figure 49 (c). On the other hand, all the heating curves of

Figure 49 (a) showed exothermic peaks (shown by arrows) that are attributed to the

cold crystallization (CC) phenomenon.233 CC originates from the rearrangement of amorphous phase into crystalline phase234 and it suggests that erythritol has a slow melt- crystallization behavior. However, selecting the scan rate of 2 °C/min decreases the incomplete melt crystallization.235 Since CC is a function of segmental mobility, it is reasonable that in presence of fillers and consequently restricted mobility of the amorphous phase, ΔH of CC and corresponding temperature were decreased comparing to the neat erythritol.236

103

During the solidification process shown in

Figure 49 (b) pure erythritol showed three distinct solidification peaks at 29.75, 24.89 and 22.29 °C, which are attributed to presence of different polymorphs.237-238 Polymorphism arises from existence of several crystalline forms as a function of processing conditions.239 Since only one peak was detected during the melting process for erythritol, it can be concluded that it has three distinct polymorphs with different solidification temperature and similar melting point. For the fabricated composites first, only one single peak was appeared during solidification process, and second, the solidification temperature was shifted to higher temperatures as 65.45 and 65.54 ºC

104

(

Figure 49 (d)) with higher or about the same values of solidification enthalpy for Ery-1CBNA and Ery-2CBNA, respectively. The higher or similar solidification enthalpy for lower content of erythritol shows conversion of more amorphous phase of eryhtritol to crystals. Besides, pure erythritol showed a temperature difference of ~94.6 °C between melting and solidification points (average value of the mentioned points) that is known as subcooling240. However, this temperature difference was decreased for ~ 41.4

105

ºC for the composites,

Figure 49 (e). Hence, BN walls within CBNAs caused dual beneficial effects of improving the crystallization process with production of more stable polymorph and also reducing the subcooling effects. These effects are due to the heterogeneous nucleation role of the BN walls and facilitating the crystallization of amorphous erythritol.

106

Figure 49. (a) DSC heating and (b) cooling curves of erythritol, Ery-1CBNA and Ery-

2CBNA obtained at scan rate of 2 °C/min and their corresponding trend of melting (c) and solidification (d), and subcooling (e) points.

107

Table 9. Phase change temperatures and corresponding enthalpies of Ery-1CBNA and Ery-

2CBNA composites.

ΔHmelting TMelting ΔHCC TCC ΔHCrystallization TCrystallization Sample (J/g) (°C) (J/g) (°C) (J/g) (°C) 22.29, Erythritol 346.04 120.24 20.88 67.16 188.75 24.89, 29.75 Ery-1CBNA 264.86 118.99 17.72 67.93 197.96 65.45 Ery-2CBNA 258.92 118.72 16.89 68.04 188.39 65.54

The cyclic charging-discharging capability of the erythritol and the obtained

composites were investigated using DSC. For this purpose, their thermal behavior was

recorded after every 10 cycles and the obtained results are shown in Figure 50 (a-c) and

Table 10. Based on these results, for the composites the melting point and its enthalpy

remained almost constant even after 20 cycles. However, the solidification temperature and

enthalpy were changed slightly (< 3%) compared to the cycle-0. This negligible difference can be explained by agglomeration and changed distribution of part of fillers during the heating-cooling cycles241 or also due to the effect of processing temperature on the present

polymorphs and their different behavior242. For pure erythritol, both the crystallization and

melting enthalpies were changed significantly over the cycles. Besides, the polymorphs’

crystallization temperatures and their peaks behavior were not constant. Hence, it can be

concluded that the fabricated scaffolds significantly improved the thermal stability of

erythritol, which led to thermally stable PCM composites.

The shape stability is one of other important features of PCM composites because

it prevents the leakage of the used PCM. To investigate this property of the obtained

108 composites, they were placed on a hot plate set at 140 °C for 45 minutes. As shown in

Figure 50 (d-d”), the shape of composite was maintained with negligible leakage at the bottom, thus the porous scaffold can efficiently hold erythritol and improves its shape stability. Therefore, the fabricated scaffold could improve simultaneously the heat releasing enthalpy, subcooling, TC and shape stability of erythritol and consequently make the Ery-CBNAs composites as a proper potential PCM composites for medium temperature environment.

109

Figure 50. DSC thermograms of Erythritol (a), Ery-1CBNA (b) and Ery-2CBNA (c) after

every 10 heating-cooling cycles. Shape stability test of the composites was investigating by placing the composite (d) on a hot plate set at 140 °C, the top (d’) and side-view (d”) of the composite after 45 mins.

Table 10. Phase change temperatures and enthalpies of erythritol and the obtained composites extracted from DCS cyclic thermograms.

ΔHmelting TMelting ΔHCrystallization Sample Cycle TCrystallization (°C) (J/g) (°C) (J/g) 0 361.39 120.40 155.96 Sharpest peak 20.92 Erythritol 1 349.10 120.05 189.09 Sharpest peak 13.64 2 349.14 120.14 153.77 Sharpest peak 11.02 0 273.68 119.49 192.46 61.05 Ery- 1 273.29 119.45 195.76 62.05 1CBNA 2 274.76 119.43 190.22 59.42 0 266.17 119.16 192.70 63.43 Ery- 1 264.05 119.08 190.02 60.99 2CBNA 2 260.52 119.04 188.37 60.89

110

6.5. Conclusions

Here, the PCM properties of erythritol incorporated into CBNA scaffolds were

investigated. The CBNA scaffolds were fabricated with similar procedure as explained in

chapter IV. Similar to xylitol-CBNA composites, Ery-CBNA composites also showed high

TC of 2.77 W·m-1·K-1 at 15.5 wt% loading of BN. TC improvement caused lower melting point and higher solidification temperature and consequently the infiltrated erythritol

showed lower subcooling. Additionally, TC enhancement reduces the required time for

charging and discharging heat of the fabricated PCM. In addition to TC enhancement, BN

has improved the PCM properties and shape stability of erythritol significantly. The

heterogeneous nucleation effect of BN enhanced the solidification of erythritol from the

melt that lead to higher solidification enthalpy of composites comparing to neat erythritol.

The obtained composites showed thermal cycling stability as well. As a result, combination

of erythritol and aligned CBNA scaffold resulted in shape and thermal cycling stable PCM

composites with high TC up to 2.77 W·m-1·K-1 at ~ 15.5 wt% of BN.

111

REFERENCES

(1) Lee, J.-H.; Lee, S.-H.; Choi, C.; Jang, S.; Choi, S. A review of thermal conductivity data, mechanisms and models for nanofluids. International Journal of Micro-Nano Scale Transport 2011.

(2) Chen, H.; Ginzburg, V. V.; Yang, J.; Yang, Y.; Liu, W.; Huang, Y.; Du, L.; Chen, B. Thermal conductivity of polymer-based composites: Fundamentals and applications. Progress in Polymer Science 2016, 59, 41-85.

(3) Burger, N.; Laachachi, A.; Ferriol, M.; Lutz, M.; Toniazzo, V.; Ruch, D. Review of thermal conductivity in composites: mechanisms, parameters and theory. Progress in Polymer Science 2016, 61, 1-28.

(4) Mehra, N.; Mu, L.; Ji, T.; Yang, X.; Kong, J.; Gu, J.; Zhu, J. Thermal transport in polymeric materials and across composite interfaces. Applied Materials Today 2018, 12, 92-130.

(5) Choy, C. Thermal conductivity of polymers. Polymer 1977, 18 (10), 984-1004.

(6) Klemens, P. Thermal conductivity and lattice vibrational modes. In Solid state physics; Elsevier: 1958; pp 1-98.

(7) Henry, A.; Chen, G. High thermal conductivity of single polyethylene chains using molecular dynamics simulations. Physical review letters 2008, 101 (23), 235502.

(8) Luo, T.; Esfarjani, K.; Shiomi, J.; Henry, A.; Chen, G. Molecular dynamics simulation of thermal energy transport in polydimethylsiloxane. Journal of Applied Physics 2011, 109 (7), 074321.

(9) Shen, S.; Henry, A.; Tong, J.; Zheng, R.; Chen, G. Polyethylene nanofibres with very high thermal conductivities. Nature nanotechnology 2010, 5 (4), 251.

(10) Zhang, T.; Wu, X.; Luo, T. Polymer nanofibers with outstanding thermal conductivity and thermal stability: fundamental linkage between molecular characteristics and macroscopic thermal properties. The Journal of Physical Chemistry C 2014, 118 (36), 21148-21159. (11) Liu, J.; Yang, R. Length-dependent thermal conductivity of single extended polymer chains. Physical Review B 2012, 86 (10), 104307.

(12) Yu, J.; Sundqvist, B.; Tonpheng, B.; Andersson, O. Thermal conductivity of highly crystallized polyethylene. Polymer 2014, 55 (1), 195-200.

112

(13) Ni, B.; Watanabe, T.; Phillpot, S. R. Thermal transport in polyethylene and at polyethylene–diamond interfaces investigated using molecular dynamics simulation. Journal of Physics: Condensed Matter 2009, 21 (8), 084219.

(14) Rashidi, V.; Coyle, E. J.; Sebeck, K.; Kieffer, J.; Pipe, K. P. Thermal conductance in cross-linked polymers: Effects of non-bonding interactions. The Journal of Physical Chemistry B 2017, 121 (17), 4600-4609

(15) Kikugawa, G.; Desai, T. G.; Keblinski, P.; Ohara, T. Effect of crosslink formation on heat conduction in amorphous polymers. Journal of Applied Physics 2013, 114 (3), 034302.

(16) Steiner, T. The hydrogen bond in the solid state. Angewandte Chemie International Edition 2002, 41 (1), 48-76.

(17) Kim, G.-H.; Lee, D.; Shanker, A.; Shao, L.; Kwon, M. S.; Gidley, D.; Kim, J.; Pipe, K. P. High thermal conductivity in amorphous polymer blends by engineered interchain interactions. Nature materials 2015, 14 (3), 295.

(18) Jiajun, W.; Xiao-Su, Y. Effects of interfacial thermal barrier resistance and particle shape and size on the thermal conductivity of AlN/PI composites. Composites Science and Technology 2004, 64 (10-11), 1623-1628.

(19) Yuan, F.-Y.; Zhang, H.-B.; Li, X.; Li, X.-Z.; Yu, Z.-Z. Synergistic effect of boron nitride flakes and tetrapod-shaped ZnO whiskers on the thermal conductivity of electrically insulating phenol formaldehyde composites. Composites Part A: Applied Science and Manufacturing 2013, 53, 137-144.

(20) Tsutsumi, N.; Takeuchi, N.; Kiyotsukuri, T. Measurement of thermal diffusivity of filler‐polymide composites by flash radiometry. Journal of Polymer Science Part B: Polymer Physics 1991, 29 (9), 1085-1093.

(21) Wu, H.; Drzal, L. T. High thermally conductive graphite nanoplatelet/polyetherimide composite by precoating: effect of percolation and particle size. Polymer Composites 2013, 34 (12), 2148-2153.

(22) Kemaloglu, S.; Ozkoc, G.; Aytac, A. Thermally conductive boron nitride/SEBS/EVA ternary composites:“processing and characterization”. Polymer Composites 2010, 31 (8), 1398-1408.

(23) Fu, J.; Shi, L.; Zhang, D.; Zhong, Q.; Chen, Y. Effect of nanoparticles on the performance of thermally conductive epoxy adhesives. Polymer Engineering & Science 2010, 50 (9), 1809-1819.

(24) Bicerano, J.; Douglas, J. F.; Brune, D. A. Model for the viscosity of particle dispersions. 1999.

113

(25) Tao, Y.; Yang, Z.; Lu, X.; Tao, G.; Xia, Y.; Wu, H. Influence of filler morphology on percolation threshold of isotropical conductive adhesives (ICA). Science China Technological Sciences 2012, 55 (1), 28-33.

(26) Zhou, W.; Qi, S.; Tu, C.; Zhao, H.; Wang, C.; Kou, J. Effect of the particle size of Al2O3 on the properties of filled heat‐conductive silicone rubber. Journal of Applied Polymer Science 2007, 104 (2), 1312-1318.

(27) Li, T.-L.; Hsu, S. L.-C. Enhanced thermal conductivity of polyimide films via a hybrid of micro-and nano-sized boron nitride. The Journal of Physical Chemistry B 2010, 114 (20), 6825-6829.

(28) Pashayi, K.; Fard, H. R.; Lai, F.; Iruvanti, S.; Plawsky, J.; Borca-Tasciuc, T. High thermal conductivity epoxy-silver composites based on self-constructed nanostructured metallic networks. Journal of Applied Physics 2012, 111 (10), 104310.

(29) Warzoha, R. J.; Fleischer, A. S. Heat flow at nanoparticle interfaces. Nano Energy 2014, 6, 137-158.

(30) Gulotty, R.; Castellino, M.; Jagdale, P.; Tagliaferro, A.; Balandin, A. A. Effects of functionalization on thermal properties of single-wall and multi-wall carbon nanotube– polymer nanocomposites. ACS nano 2013, 7 (6), 5114-5121.

(31) Wattanakul, K.; Manuspiya, H.; Yanumet, N. The adsorption of cationic surfactants on BN surface: Its effects on the thermal conductivity and mechanical properties of BN- epoxy composite. Colloids and Surfaces A: Physicochemical and Engineering Aspects 2010, 369 (1-3), 203-210.

(32) Huang, X.; Iizuka, T.; Jiang, P.; Ohki, Y.; Tanaka, T. Role of interface on the thermal conductivity of highly filled dielectric epoxy/AlN composites. The Journal of Physical Chemistry C 2012, 116 (25), 13629-13639.

(33) Peng, W.; Huang, X.; Yu, J.; Jiang, P.; Liu, W. Electrical and thermophysical properties of epoxy/aluminum nitride nanocomposites: Effects of nanoparticle surface modification. Composites Part A: Applied Science and Manufacturing 2010, 41 (9), 1201- 1209.

(34) Ganguli, S.; Roy, A. K.; Anderson, D. P. Improved thermal conductivity for chemically functionalized exfoliated graphite/epoxy composites. Carbon 2008, 46 (5), 806-817.

(35) Xu, Y.; Chung, D. Increasing the thermal conductivity of boron nitride and aluminum nitride particle epoxy-matrix composites by particle surface treatments. Composite Interfaces 2000, 7 (4), 243-256.

114

(36) Brigandi, P. J.; Cogen, J. M.; Pearson, R. A. Electrically conductive multiphase polymer blend carbon‐based composites. Polymer Engineering & Science 2014, 54 (1), 1-16.

(37) Pötschke, P.; Paul, D. Formation of co-continuous structures in melt-mixed immiscible polymer blends. Journal of Macromolecular Science, Part C: Polymer Reviews 2003, 43 (1), 87-141.

(38) Feng, J.; Chan, C. M. Carbon black–filled immiscible blends of poly (vinylidene fluoride) and high density polyethylene: Electrical properties and morphology. Polymer Engineering & Science 1998, 38 (10), 1649-1657.

(39) Sumita, M.; Sakata, K.; Asai, S.; Miyasaka, K.; Nakagawa, H. Dispersion of fillers and the electrical conductivity of polymer blends filled with carbon black. Polymer bulletin 1991, 25 (2), 265-271.

(40) Gubbels, F.; Jérôme, R.; Teyssie, P.; Vanlathem, E.; Deltour, R.; Calderone, A.; Parente, V.; Brédas, J.-L. Selective localization of carbon black in immiscible polymer blends: a useful tool to design electrical conductive composites. Macromolecules 1994, 27 (7), 1972-1974.

(41) Xu, Z.; Zhang, Y.; Wang, Z.; Sun, N.; Li, H. Enhancement of electrical conductivity by changing phase morphology for composites consisting of polylactide and poly (ε- caprolactone) filled with acid-oxidized multiwalled carbon nanotubes. ACS applied materials & interfaces 2011, 3 (12), 4858-4864.

(42) Chen, J.; Cui, X.; Zhu, Y.; Jiang, W.; Sui, K. Design of superior conductive polymer composite with precisely controlling carbon nanotubes at the interface of a co-continuous polymer blend via a balance of π-π interactions and dipole-dipole interactions. Carbon 2017, 114, 441-448.

(43) Hu, N.; Karube, Y.; Yan, C.; Masuda, Z.; Fukunaga, H. Tunneling effect in a polymer/carbon nanotube nanocomposite strain sensor. Acta Materialia 2008, 56 (13), 2929-2936.

(44) Ounaies, Z.; Park, C.; Wise, K.; Siochi, E.; Harrison, J. Electrical properties of single wall carbon nanotube reinforced polyimide composites. Composites science and technology 2003, 63 (11), 1637-1646.

(45) Singh, V.; Bougher, T. L.; Weathers, A.; Cai, Y.; Bi, K.; Pettes, M. T.; McMenamin, S. A.; Lv, W.; Resler, D. P.; Gattuso, T. R. High thermal conductivity of chain-oriented amorphous polythiophene. Nature nanotechnology 2014, 9 (5), 384-390.

(46) Henry, A. Thermal transport in polymers. Annu. Rev. Heat Transfer 2013, 17, 485- 520.

115

(47) Li, Y.; Mehra, N.; Ji, T.; Yang, X.; Mu, L.; Gu, J.; Zhu, J. The stiffness–thermal conduction relationship at the composite interface: the effect of particle alignment on the long-range confinement of polymer chains monitored by scanning thermal microscopy. Nanoscale 2018, 10 (4), 1695-1703.

(48) Mehra, N.; Li, Y.; Zhu, J. Small Organic Linkers with Hybrid Terminal Groups Drive Efficient Phonon Transport in Polymers. The Journal of Physical Chemistry C 2018, 122 (19), 10327-10333.

(49) Mehra, N.; Kashfipour, M. A.; Zhu, J. Filler free technology for enhanced thermally conductive optically transparent polymeric materials using low thermally conductive organic linkers. Applied Materials Today 2018, 13, 207-216.

(50) Cao, J.-P.; Zhao, J.; Zhao, X.; You, F.; Yu, H.; Hu, G.-H.; Dang, Z.-M. High thermal conductivity and high electrical resistivity of poly (vinylidene fluoride)/polystyrene blends by controlling the localization of hybrid fillers. Composites Science and Technology 2013, 89, 142-148.

(51) Cao, J.-P.; Zhao, X.; Zhao, J.; Zha, J.-W.; Hu, G.-H.; Dang, Z.-M. Improved thermal conductivity and flame retardancy in polystyrene/poly (vinylidene fluoride) blends by controlling selective localization and surface modification of SiC nanoparticles. ACS applied materials & interfaces 2013, 5 (15), 6915-6924.

(52) Droval, G.; Feller, J.-F.; Salagnac, P.; Glouannec, P. Conductive polymer composites with double percolated architecture of carbon nanoparticles and ceramic microparticles for high heat dissipation and sharp PTC switching. Smart Materials and Structures 2008, 17 (2), 025011.

(53) Huang, J.; Zhu, Y.; Xu, L.; Chen, J.; Jiang, W.; Nie, X. Massive enhancement in the thermal conductivity of polymer composites by trapping graphene at the interface of a polymer blend. Composites Science and Technology 2016, 129, 160-165.

(54) Hammani, S.; Moulai-Mostefa, N.; Benyahia, L.; Tassin, J.-F. Effects of composition and extrusion parameters on the morphological development and rheological properties of PP/PC blends. Co-continuity investigation. Journal of Polymer Research 2012, 19 (8), 9940.

(55) Steinmann, S.; Gronski, W.; Friedrich, C. Influence of selective filling on rheological properties and phase inversion of two-phase polymer blends. Polymer 2002, 43 (16), 4467- 4477.

(56) Zhang, b.; Yi, X.; Yui, H.; Asai, S.; Sumita, M. Selective location and double percolation of short carbon fiber filled polymer blends: high-density polyethylene/isotactic polypropylene. Materials Letters 1998, 36 (1-4), 186-190.

116

(57) Harrats, C.; Groeninckx, G.; Thomas, S. Micro-and nanostructured multiphase polymer blend systems: phase morphology and interfaces, CRC press: 2005.

(58) Fenouillot, F.; Cassagnau, P.; Majesté, J.-C. Uneven distribution of nanoparticles in immiscible fluids: Morphology development in polymer blends. Polymer 2009, 50 (6), 1333-1350.

(59) Moreira, A. C. F.; Cario Jr, F. O.; Soares, B. G. Cocontinuous morphologies in polystyrene/ethylene–vinyl acetate blends: The influence of the processing temperature. Journal of applied polymer science 2003, 89 (2), 386-398.

(60) Lee, J. K.; Han, C. D. Evolution of polymer blend morphology during compounding in a twin-screw extruder. Polymer 2000, 41 (5), 1799-1815.

(61) Lee, J. K.; Han, C. D. Evolution of polymer blend morphology during compounding in an internal mixer. Polymer 1999, 40 (23), 6277-6296.

(62) Li, W. Effect of silica nanoparticles on the morphology of polymer blends. Technische Universiteit Eindhoven 2011, 1-159.

(63) de Luna, M. S.; Filippone, G. Effects of nanoparticles on the morphology of immiscible polymer blends–challenges and opportunities. European Polymer Journal 2016, 79, 198-218.

(64) Favis, B. D. In Processing/morphology/rheology relationships in polymer blends, Makromolekulare Chemie. Macromolecular Symposia, Wiley Online Library: 1992; pp 143-150.

(65) Favis, B. D. Polymer alloys and blends: Recent advances. The Canadian Journal of Chemical Engineering 1991, 69 (3), 619-625.

(66) Avgeropoulos, G.; Weissert, F.; Biddison, P.; Bohm, G. Heterogeneous blends of polymers. Rheology and morphology. Rubber Chemistry and Technology 1976, 49 (1), 93- 104.

(67) Favis, B. D.; Chalifoux, J.-P. The effect of viscosity ratio on the morphology of polypropylene/polycarbonate blends during processing. Polymer Engineering & Science 1987, 27 (21), 1591-1600.

(68) Han, C. D. Rheological properties of calcium carbonate‐filled polypropylene melts. Journal of Applied Polymer Science 1974, 18 (3), 821-829.

(69) Reignier, J.; Favis, B. D.; Heuzey, M.-C. Factors influencing encapsulation behavior in composite droplet-type polymer blends. Polymer 2003, 44 (1), 49-59.

117

(70) Van Oene, H. Rheology of polymer blends and dispersions. In Polymer Blends, Volume 1; Elsevier: 1978; pp 295-352.

(71) Van Oene, H. Colloid Interface Sci. 1972, 40, 448. CrossRef| CAS| Web of Science® Times Cited 1972, 248.

(72) Bourry, D.; Favis, B. Cocontinuity and phase inversion in HDPE/PS blends: influence of interfacial modification and elasticity. Journal of polymer science part B: polymer physics 1998, 36 (11), 1889-1899.

(73) Hong, J. S.; Kim, Y. K.; Ahn, K. H.; Lee, S. J.; Kim, C. Interfacial tension reduction in PBT/PE/clay nanocomposite. Rheologica Acta 2007, 46 (4), 469-478.

(74) Jafari, S.; Pötschke, P.; Stephan, M.; Warth, H.; Alberts, H. Multicomponent blends based on polyamide 6 and styrenic polymers: morphology and melt rheology. Polymer 2002, 43 (25), 6985-6992.

(75) Verhoogt, H.; Van Dam, J.; Posthuma de Boer, A. Morphology-Processing Relationship in Interpenetrating Polymer Blends. Styrene-Ethylene/Butylene-Styrene Block Copolymer and Poly (Ether Ester). Interpenetrating Polymer Networks 1991, 333- 351.

(76) Manas-Zloczower, I. Mixing and compounding of polymers: theory and practice, Carl Hanser Verlag GmbH Co KG: 2012.

(77) Grace, H. P. Dispersion phenomena in high viscosity immiscible fluid systems and application of static mixers as dispersion devices in such systems. Chemical Engineering Communications 1982, 14 (3-6), 225-277.

(78) Agarwal, S.; Khan, M. M. K.; Gupta, R. K. Thermal conductivity of polymer nanocomposites made with carbon nanofibers. Polymer Engineering & Science 2008, 48 (12), 2474-2481.

(79) Malekie, S.; Ziaie, F. A two-dimensional simulation to predict the electrical behavior of carbon nanotube/polymer composites. Journal of Polymer Engineering 2017, 37 (2), 205-210.

(80) Patra, R.; Suin, S.; Mandal, D.; Khatua, B. Sequential mixing as effective method in the reduction of percolation threshold of multiwall carbon nanotube in poly (methyl methacrylate)/high ‐ density poly (ethylene)/MWCNT nanocomposites. Journal of Applied Polymer Science 2014, 131 (10).

(81) Sumita, a.; Sakata, K.; Hayakawa, Y.; Asai, S.; Miyasaka, K.; Tanemura, M. Double percolation effect on the electrical conductivity of conductive particles filled polymer blends. Colloid and Polymer Science 1992, 270 (2), 134-139.

118

(82) Gast, A. P.; Adamson, A. W., Physical chemistry of surfaces. John Wiley and Sons, Inc: New York: 1997.

(83) Ross, S.; Morrison, E. Colloidal systems and interfaces. 1988.

(84) Wu, S. Interfacial and surface tensions of polymers. Journal of Macromolecular Science—Reviews in Macromolecular Chemistry 1974, 10 (1), 1-73.

(85) Asai, S.; Sakata, K.; Sumita, M.; Miyasaka, K. Effect of interfacial free energy on the heterogeneous distribution of oxidized carbon black in polymer blends. Polymer journal 1992, 24 (5), 415-420.

(86) Katada, A.; Buys, Y. F.; Tominaga, Y.; Asai, S.; Sumita, M. Relationship between electrical resistivity and particle dispersion state for carbon black filled poly (ethylene-co- vinyl acetate)/poly (L-lactic acid) blend. Colloid and Polymer Science 2005, 284 (2), 134- 141.

(87) Pan, Y.; Liu, X.; Hao, X.; Starý, Z.; Schubert, D. W. Enhancing the electrical conductivity of carbon black-filled immiscible polymer blends by tuning the morphology. European Polymer Journal 2016, 78, 106-115.

(88) Owens, D. K.; Wendt, R. Estimation of the surface free energy of polymers. Journal of applied polymer science 1969, 13 (8), 1741-1747.

(89) Nuriel, S.; Liu, L.; Barber, A.; Wagner, H. Direct measurement of multiwall nanotube surface tension. Chemical Physics Letters 2005, 404 (4-6), 263-266.

(90) Li, L.-p.; Yin, B.; Zhou, Y.; Gong, L.; Yang, M.-b.; Xie, B.-h.; Chen, C. Characterization of PA6/EPDM-g-MA/HDPE ternary blends: The role of core-shell structure. Polymer 2012, 53 (14), 3043-3051.

(91) Liebscher, M.; Tzounis, L.; Pötschke, P.; Heinrich, G. Influence of the viscosity ratio in PC/SAN blends filled with MWCNTs on the morphological, electrical, and melt rheological properties. Polymer 2013, 54 (25), 6801-6808.

(92) Bai, L.; Sharma, R.; Cheng, X.; Macosko, C. W. Kinetic Control of Graphene Localization in Co-continuous Polymer Blends via Melt Compounding. Langmuir 2017, 34 (3), 1073-1083.

(93) Jamali, S.; Paiva, M. C.; Covas, J. A. Dispersion and re-agglomeration phenomena during melt mixing of polypropylene with multi-wall carbon nanotubes. Polymer Testing 2013, 32 (4), 701-707.

(94) Kim, J. A.; Seong, D. G.; Kang, T. J.; Youn, J. R. Effects of surface modification on rheological and mechanical properties of CNT/epoxy composites. Carbon 2006, 44 (10), 1898-1905.

119

(95) Mitchell, C. A.; Bahr, J. L.; Arepalli, S.; Tour, J. M.; Krishnamoorti, R. Dispersion of functionalized carbon nanotubes in polystyrene. Macromolecules 2002, 35 (23), 8825- 8830.

(96) Larson, R. G. The structure and rheology of complex fluids (topics in chemical engineering). Oxford University Press, New York• Oxford 1999, 86, 108.

(97) Solomon, M. J.; Almusallam, A. S.; Seefeldt, K. F.; Somwangthanaroj, A.; Varadan, P. Rheology of polypropylene/clay hybrid materials. Macromolecules 2001, 34 (6), 1864- 1872.

(98) Krishnamoorti, R.; Yurekli, K. Rheology of polymer layered silicate nanocomposites. Current Opinion in Colloid & Interface Science 2001, 6 (5-6), 464-470.

(99) Gojny, F. H.; Wichmann, M. H.; Fiedler, B.; Kinloch, I. A.; Bauhofer, W.; Windle, A. H.; Schulte, K. Evaluation and identification of electrical and thermal conduction mechanisms in carbon nanotube/epoxy composites. Polymer 2006, 47 (6), 2036-2045.

(100) Liu, J.; Wang, X.; Li, D.; Coates, N. E.; Segalman, R. A.; Cahill, D. G. Thermal conductivity and elastic constants of PEDOT: PSS with high electrical conductivity. Macromolecules 2015, 48 (3), 585-591.

(101) Poehler, T. O.; Katz, H. E. Prospects for polymer-based thermoelectrics: state of the art and theoretical analysis. Energy & Environmental Science 2012, 5 (8), 8110-8115.

(102) Veenstra, H.; Verkooijen, P. C.; van Lent, B. J.; van Dam, J.; de Boer, A. P.; Nijhof, A. P. H. On the mechanical properties of co-continuous polymer blends: experimental and modelling. Polymer 2000, 41 (5), 1817-1826.

(103) Kashfipour, M. A.; Dent, R. S.; Mehra, N.; Yang, X.; Gu, J.; Zhu, J. Directional xylitol crystal propagation in oriented micro-channels of boron nitride aerogel for isotropic heat conduction. Composites Science and Technology 2019, 107715.

(104) Wang, Y.; Kong, D.; Shi, W.; Liu, B.; Sim, G. J.; Ge, Q.; Yang, H. Y. Ice Templated Free ‐ Standing Hierarchically WS2/CNT ‐ rGO Aerogel for High ‐ Performance Rechargeable Lithium and Sodium Ion Batteries. Advanced Energy Materials 2016, 6 (21), 1601057.

(105) Tingaut, P.; Zimmermann, T.; Sèbe, G. Cellulose nanocrystals and microfibrillated cellulose as building blocks for the design of hierarchical functional materials. Journal of Materials Chemistry 2012, 22 (38), 20105-20111.

(106) Yang, J.; Zhang, E.; Li, X.; Zhang, Y.; Qu, J.; Yu, Z.-Z. Cellulose/graphene aerogel supported phase change composites with high thermal conductivity and good shape stability for thermal energy storage. Carbon 2016, 98, 50-57.

120

(107) Yang, J.; Qi, G.-Q.; Liu, Y.; Bao, R.-Y.; Liu, Z.-Y.; Yang, W.; Xie, B.-H.; Yang, M.-B. Hybrid graphene aerogels/phase change material composites: thermal conductivity, shape-stabilization and light-to-thermal energy storage. Carbon 2016, 100, 693-702.

(108) Zhi, C.; Bando, Y.; Tang, C.; Kuwahara, H.; Golberg, D. Large‐scale fabrication of boron nitride nanosheets and their utilization in polymeric composites with improved thermal and mechanical properties. Advanced Materials 2009, 21 (28), 2889-2893.

(109) Hu, J.; Huang, Y.; Yao, Y.; Pan, G.; Sun, J.; Zeng, X.; Sun, R.; Xu, J.-B.; Song, B.; Wong, C.-P. Polymer composite with improved thermal conductivity by constructing a hierarchically ordered three-dimensional interconnected network of BN. ACS applied materials & interfaces 2017, 9 (15), 13544-13553.

(110) Lin, Z.; Liu, Y.; Raghavan, S.; Moon, K.-s.; Sitaraman, S. K.; Wong, C.-p. Magnetic alignment of hexagonal boron nitride platelets in polymer matrix: toward high performance anisotropic polymer composites for electronic encapsulation. ACS applied materials & interfaces 2013, 5 (15), 7633-7640.

(111) Cho, H.-B.; Nakayama, T.; Suzuki, T.; Tanaka, S.; Jiang, W.; Suematsu, H.; Niihara, K. Linear assembles of BN nanosheets, fabricated in polymer/BN nanosheet composite film. Journal of Nanomaterials 2011, 2011, 25.

(112) Shen, H.; Guo, J.; Wang, H.; Zhao, N.; Xu, J. Bioinspired modification of h-BN for high thermal conductive composite films with aligned structure. ACS applied materials & interfaces 2015, 7 (10), 5701-5708.

(113) Datsyuk, V.; Trotsenko, S.; Reich, S. Carbon-nanotube–polymer nanofibers with high thermal conductivity. Carbon 2013, 52, 605-608.

(114) Wieme, T.; Tang, D.; Delva, L.; D'hooge, D. R.; Cardon, L. The relevance of material and processing parameters on the thermal conductivity of thermoplastic composites. Polymer Engineering & Science 2018, 58 (4), 466-474.

(115) Du, Y.; Hedayat, N.; Panthi, D.; Ilkhani, H.; Emley, B. J.; Woodson, T. Freeze- casting for the fabrication of solid oxide fuel cells: A review. Materialia 2018.

(116) Hedayat, N.; Du, Y.; Ilkhani, H. Review on fabrication techniques for porous electrodes of solid oxide fuel cells by sacrificial template methods. Renewable and Sustainable Energy Reviews 2017, 77, 1221-1239.

(117) Pezzotti, G.; Kamada, I.; Miki, S. Thermal conductivity of AlN/polystyrene interpenetrating networks. Journal of the European Ceramic Society 2000, 20 (8), 1197- 1203.

121

(118) Wang, Y.; Shi, Z.; Yin, J. Boron nitride nanosheets: large-scale exfoliation in methanesulfonic acid and their composites with polybenzimidazole. Journal of Materials Chemistry 2011, 21 (30), 11371-11377.

(119) Pakdel, A.; Bando, Y.; Golberg, D. Nano boron nitride flatland. Chemical Society Reviews 2014, 43 (3), 934-959.

(120) Mu, L.; Ji, T.; Chen, L.; Mehra, N.; Shi, Y.; Zhu, J. Paving the thermal highway with self-organized nanocrystals in transparent polymer composites. ACS applied materials & interfaces 2016, 8 (42), 29080-29087.

(121) Beuguel, Q.; Boyer, S. A.; Settipani, D.; Monge, G.; Haudin, J. M.; Vergnes, B.; Peuvrel‐Disdier, E. Crystallization behavior of polypropylene/graphene nanoplatelets composites. Polymer Crystallization 2018, 1 (3), e10024.

(122) Li, L.; Li, C. Y.; Ni, C. Polymer crystallization-driven, periodic patterning on carbon nanotubes. Journal of the American Chemical Society 2006, 128 (5), 1692-1699.

(123) Duquesne, M.; Godin, A.; del Barrio, E. P.; Achchaq, F. Crystal growth kinetics of sugar alcohols as phase change materials for thermal energy storage. Energy Procedia 2017, 139, 315-321.

(124) Palomäki, E.; Ahvenainen, P.; Ehlers, H.; Svedström, K.; Huotari, S.; Yliruusi, J. Monitoring the recrystallisation of amorphous xylitol using Raman spectroscopy and wide- angle X-ray scattering. International journal of pharmaceutics 2016, 508 (1-2), 71-82.

(125) del Barrio, E. P.; Godin, A.; Duquesne, M.; Daranlot, J.; Jolly, J.; Alshaer, W.; Kouadio, T.; Sommier, A. Characterization of different sugar alcohols as phase change materials for thermal energy storage applications. Solar Energy Materials and Solar Cells 2017, 159, 560-569.

(126) Yang, J.; Yu, P.; Tang, L.-S.; Bao, R.-Y.; Liu, Z.-Y.; Yang, M.-B.; Yang, W. Hierarchically interconnected porous scaffolds for phase change materials with improved thermal conductivity and efficient solar-to-electric energy conversion. Nanoscale 2017, 9 (45), 17704-17709.

(127) Qian, Z.; Shen, H.; Fang, X.; Fan, L.; Zhao, N.; Xu, J. Phase change materials of paraffin in h-BN porous scaffolds with enhanced thermal conductivity and form stability. Energy and Buildings 2018, 158, 1184-1188.

(128) Zeng, X.; Yao, Y.; Gong, Z.; Wang, F.; Sun, R.; Xu, J.; Wong, C. P. Ice‐Templated Assembly Strategy to Construct 3D Boron Nitride Nanosheet Networks in Polymer Composites for Thermal Conductivity Improvement. Small 2015, 11 (46), 6205-6213.

122

(129) Streletskii, A.; Permenov, D.; Bokhonov, B.; Kolbanev, I.; Leonov, A.; Berestetskaya, I.; Streletzky, K. A. Destruction, amorphization and reactivity of nano-BN under ball milling. Journal of Alloys and Compounds 2009, 483 (1-2), 313-316.

(130) Lin, Y.; Williams, T. V.; Xu, T.-B.; Cao, W.; Elsayed-Ali, H. E.; Connell, J. W. Aqueous dispersions of few-layered and monolayered hexagonal boron nitride nanosheets from sonication-assisted hydrolysis: critical role of water. The Journal of Physical Chemistry C 2011, 115 (6), 2679-2685.

(131) Shtukenberg, A. G.; Cui, X.; Freudenthal, J.; Gunn, E.; Camp, E.; Kahr, B. Twisted mannitol crystals establish homologous growth mechanisms for high-polymer and small- molecule ring-banded spherulites. Journal of the American Chemical Society 2012, 134 (14), 6354-6364.

(132) Yu, L. Nucleation of one polymorph by another. Journal of the American Chemical Society 2003, 125 (21), 6380-6381.

(133) Shtukenberg, A. G.; Punin, Y. O.; Gunn, E.; Kahr, B. Spherulites. Chemical reviews 2011, 112 (3), 1805-1838.

(134) Zhang, S.; Minus, M. L.; Zhu, L.; Wong, C.-P.; Kumar, S. Polymer transcrystallinity induced by carbon nanotubes. Polymer 2008, 49 (5), 1356-1364.

(135) Huang, H.-D.; Xu, J.-Z.; Fan, Y.; Xu, L.; Li, Z.-M. Poly (L-lactic acid) crystallization in a confined space containing graphene oxide nanosheets. The Journal of Physical Chemistry B 2013, 117 (36), 10641-10651.

(136) Moniruzzaman, M.; Winey, K. I. Polymer nanocomposites containing carbon nanotubes. Macromolecules 2006, 39 (16), 5194-5205.

(137) Lu, K.; Grossiord, N.; Koning, C. E.; Miltner, H. E.; Mele, B. v.; Loos, J. Carbon nanotube/isotactic polypropylene composites prepared by latex technology: morphology analysis of CNT-induced nucleation. Macromolecules 2008, 41 (21), 8081-8085.

(138) Gunasekara, S. N.; Chiu, J. N.; Martin, V.; Hedström, P. The experimental phase diagram study of the binary polyols system erythritol-xylitol. Solar Energy Materials and Solar Cells 2018, 174, 248-262.

(139) Wang, Z.; Zhang, X.; Chen, B.; Hou, M.; Liu, T. The Controllable Preparation, Properties and Structural Characteristics of Xylitol/Menthol Co-crystals. International Journal of Food Engineering 2017, 13 (8).

(140) Alexander, L. E. X-ray diffraction methods in polymer science. 1969.

123

(141) Song, N.; Jiao, D.; Ding, P.; Cui, S.; Tang, S.; Shi, L. Anisotropic thermally conductive flexible films based on nanofibrillated cellulose and aligned graphene nanosheets. Journal of Materials Chemistry C 2016, 4 (2), 305-314.

(142) Su, Z.; Wang, H.; He, J.; Guo, Y.; Qu, Q.; Tian, X. Fabrication of Thermal Conductivity Enhanced Polymer Composites by Constructing an Oriented Three- Dimensional Staggered Interconnected Network of Boron Nitride Platelets and Carbon Nanotubes. ACS applied materials & interfaces 2018, 10 (42), 36342-36351.

(143) Terao, T.; Zhi, C.; Bando, Y.; Mitome, M.; Tang, C.; Golberg, D. Alignment of boron nitride nanotubes in polymeric composite films for thermal conductivity improvement. The Journal of Physical Chemistry C 2010, 114 (10), 4340-4344.

(144) Yuan, C.; Duan, B.; Li, L.; Xie, B.; Huang, M.; Luo, X. Thermal conductivity of polymer-based composites with magnetic aligned hexagonal boron nitride platelets. ACS applied materials & interfaces 2015, 7 (23), 13000-13006.

(145) Zhang, J.; Wang, X.; Yu, C.; Li, Q.; Li, Z.; Li, C.; Lu, H.; Zhang, Q.; Zhao, J.; Hu, M. A facile method to prepare flexible boron nitride/poly (vinyl alcohol) composites with enhanced thermal conductivity. Composites Science and Technology 2017, 149, 41-47.

(146) Zeng, X.; Sun, J.; Yao, Y.; Sun, R.; Xu, J.-B.; Wong, C.-P. A combination of boron nitride nanotubes and cellulose nanofibers for the preparation of a nanocomposite with high thermal conductivity. ACS nano 2017, 11 (5), 5167-5178.

(147) Xie, B.-H.; Huang, X.; Zhang, G.-J. High thermal conductive polyvinyl alcohol composites with hexagonal boron nitride microplatelets as fillers. Composites Science and Technology 2013, 85, 98-103.

(148) Marconnet, A. M.; Yamamoto, N.; Panzer, M. A.; Wardle, B. L.; Goodson, K. E. Thermal conduction in aligned carbon nanotube–polymer nanocomposites with high packing density. ACS nano 2011, 5 (6), 4818-4825.

(149) Tian, X.; Itkis, M. E.; Bekyarova, E. B.; Haddon, R. C. Anisotropic thermal and electrical properties of thin thermal interface layers of graphite nanoplatelet-based composites. Scientific reports 2013, 3, 1710.

(150) Ding, P.; Zhang, J.; Song, N.; Tang, S.; Liu, Y.; Shi, L. Anisotropic thermal conductive properties of hot-pressed polystyrene/graphene composites in the through- plane and in-plane directions. Composites Science and Technology 2015, 109, 25-31.

(151) Song, N.; Yang, J.; Ding, P.; Tang, S.; Shi, L. Effect of polymer modifier chain length on thermal conductive property of polyamide 6/graphene nanocomposites. Composites Part A: Applied Science and Manufacturing 2015, 73, 232-241.

124

(152) Wang, Z.; Mo, L.; Zhao, S.; Li, J.; Zhang, S.; Huang, A. Mechanically robust nacre- mimetic framework constructed polypyrrole-doped graphene/nanofiber nanocomposites with improved thermal electrical properties. Materials & Design 2018.

(153) Huang, Y.-F.; Wang, Z.-G.; Yin, H.-M.; Xu, J.-Z.; Chen, Y.; Lei, J.; Zhu, L.; Gong, F.; Li, Z.-M. Highly Anisotropic, Thermally Conductive, and Mechanically Strong Polymer Composites with Nacre-like Structure for Thermal Management Applications. ACS Applied Nano Materials 2018, 1 (7), 3312-3320.

(154) Wu, Y.; Xue, Y.; Qin, S.; Liu, D.; Wang, X.; Hu, X.; Li, J.; Wang, X.; Bando, Y.; Golberg, D. BN nanosheet/polymer films with highly anisotropic thermal conductivity for thermal management applications. ACS applied materials & interfaces 2017, 9 (49), 43163-43170.

(155) Hu, Z.; Wang, S.; Chen, G.; Wu, K.; Shi, J.; Liang, L.; Lu, M. An aqueous-only, green route to exfoliate boron nitride for preparation of high thermal conductive boron nitride nanosheet/cellulose nanofiber flexible film. Composites Science and Technology 2018, 168, 287-295.

(156) Xue, Y.; Zhou, X.; Zhan, T.; Jiang, B.; Guo, Q.; Fu, X.; Shimamura, K.; Xu, Y.; Mori, T.; Dai, P. Densely Interconnected Porous BN Frameworks for Multifunctional and Isotropically Thermoconductive Polymer Composites. Advanced Functional Materials 2018, 1801205.

(157) Song, N.; Jiao, D.; Cui, S.; Hou, X.; Ding, P.; Shi, L. Highly anisotropic thermal conductivity of layer-by-layer assembled nanofibrillated cellulose/graphene nanosheets hybrid films for thermal management. ACS applied materials & interfaces 2017, 9 (3), 2924-2932.

(158) Zhang, X.; Zhang, J.; Xia, L.; Li, C.; Wang, J.; Xu, F.; Zhang, X.; Wu, H.; Guo, S. Simple and Consecutive Melt Extrusion Method to Fabricate Thermally Conductive Composites with Highly Oriented Boron Nitrides. ACS applied materials & interfaces 2017, 9 (27), 22977-22984.

(159) Wang, H.; Ding, D.; Liu, Q.; Chen, Y.; Zhang, Q. Highly anisotropic thermally conductive polyimide composites via the alignment of boron nitride platelets. Composites Part B: Engineering 2019, 158, 311-318.

(160) Wang, T.; Wang, M.; Fu, L.; Duan, Z.; Chen, Y.; Hou, X.; Wu, Y.; Li, S.; Guo, L.; Kang, R. Enhanced Thermal Conductivity of Polyimide Composites with Boron Nitride Nanosheets. Scientific reports 2018, 8 (1), 1557.

(161) Nielsen, L. E. Thermal conductivity of particulate‐filled polymers. Journal of applied polymer science 1973, 17 (12), 3819-3820.

125

(162) Mittal, V. Modeling and prediction of polymer nanocomposite properties, John Wiley & Sons: 2012.

(163) Nielsen, L. E. The thermal and electrical conductivity of two-phase systems. Industrial & Engineering chemistry fundamentals 1974, 13 (1), 17-20.

(164) Progelhof, R.; Throne, J.; Ruetsch, R. Methods for predicting the thermal conductivity of composite systems: a review. Polymer Engineering & Science 1976, 16 (9), 615-625.

(165) Hiroshi, H.; Minoru, T. Equivalent inclusion method for steady state heat conduction in composites. International Journal of Engineering Science 1986, 24 (7), 1159-1172.

(166) Agari, Y.; Uno, T. Thermal conductivity of polymer filled with carbon materials: effect of conductive particle chains on thermal conductivity. Journal of applied polymer science 1985, 30 (5), 2225-2235.

(167) Pietrak, K.; Wiśniewski, T. S. A review of models for effective thermal conductivity of composite materials. Journal of Power Technologies 2014, 95 (1), 14-24.

(168) Kashfipour, M. A.; Mehra, N.; Zhu, J. A review on the role of interface in mechanical, thermal, and electrical properties of polymer composites. Advanced Composites and Hybrid Materials 2018, 1-25.

(169) Reinecke, B. N.; Shan, J. W.; Suabedissen, K. K.; Cherkasova, A. S. On the anisotropic thermal conductivity of magnetorheological suspensions. Journal of Applied Physics 2008, 104 (2), 023507.

(170) Bruggeman, D. DAG Bruggeman, Ann. Phys.(Leipzig) 24, 636 (1935). Ann. Phys.(Leipzig) 1935, 24, 636.

(171) Nan, C.-W.; Birringer, R.; Clarke, D. R.; Gleiter, H. Effective thermal conductivity of particulate composites with interfacial thermal resistance. Journal of Applied Physics 1997, 81 (10), 6692-6699.

(172) Mehra, N.; Mu, L.; Zhu, J. Developing heat conduction pathways through short polymer chains in a hydrogen bonded polymer system. Composites Science and Technology 2017, 148, 97-105.

(173) Li, W.; Sun, B.; Wu, P. Study on hydrogen bonds of carboxymethyl cellulose sodium film with two-dimensional correlation infrared spectroscopy. Carbohydrate polymers 2009, 78 (3), 454-461.

(174) Mehra, N.; Mu, L.; Ji, T.; Li, Y.; Zhu, J. Moisture driven thermal conduction in polymer and polymer blends. Composites Science and Technology 2017, 151, 115-123.

126

(175) Marjan A. Kashfipour, N. M., Russell S. Dent, Jiahua Zhu. Regulating Intermolecular Chain Interaction of Biopolymer with Natural Polyol for Flexible, Optically Transparent and Thermally Conductive Hybrids. Engineered Science 2019, 7, DOI: 10.30919/es8d508.

(176) Biswal, D.; Singh, R. Characterisation of carboxymethyl cellulose and polyacrylamide graft copolymer. Carbohydrate polymers 2004, 57 (4), 379-387.

(177) Bifani, V.; Ramírez, C.; Ihl, M.; Rubilar, M.; García, A.; Zaritzky, N. Effects of murta (Ugni molinae Turcz) extract on gas and water vapor permeability of carboxymethylcellulose-based edible films. LWT-Food Science and Technology 2007, 40 (8), 1473-1481.

(178) Nie, H.; Liu, M.; Zhan, F.; Guo, M. Factors on the preparation of carboxymethylcellulose hydrogel and its degradation behavior in soil. Carbohydrate Polymers 2004, 58 (2), 185-189.

(179) Chen, J.; Wang, J.; Zhang, X.; Jin, Y. Microwave-assisted green synthesis of silver nanoparticles by carboxymethyl cellulose sodium and silver nitrate. Materials chemistry and physics 2008, 108 (2-3), 421-424.

(180) Adeyeye, M. C.; Jain, A. C.; Ghorab, M. K.; Reilly, W. J. Viscoelastic evaluation of topical creams containing microcrystalline cellulose/sodium carboxymethyl cellulose as stabilizer. AAPS PharmSciTech 2002, 3 (2), 16-25.

(181) Hollabaugh, C.; Burt, L. H.; Walsh, A. P. Carboxymethylcellulose. Uses and applications. Industrial & Engineering Chemistry 1945, 37 (10), 943-947.

(182) Guo, J.-H.; Skinner, G.; Harcum, W.; Barnum, P. Pharmaceutical applications of naturally occurring water-soluble polymers. Pharmaceutical science & technology today 1998, 1 (6), 254-261.

(183) Perioli, L.; Dorigato, A.; Pagano, C.; Leoni, M.; Pegoretti, A. Thermo‐mechanical and adhesive properties of polymeric films based on ZnAl‐hydrotalcite composites for active wound dressings. Polymer Engineering & Science.

(184) Obeid, K. A.-A.; Al-Bermany, A.-K. J.; Habeeb, M. A. Enhancement of Some Mechanical Properties of Polyethylene Glycol by Adding Carboxymethyl Cellulose as a Blends and Applied in Wood Glue. World Scientific News 2015, 21, 12-23.

(185) GVRTEN, A. Adsorption behavior and inhibition corrosion effect of sodium carboxymethyl cellulose on mild steel in acidic medium. Acta Physico-Chimica Sinica 2008, 24 (12), 2236-2242.

(186) Choi, Y.; Simonsen, J. Cellulose nanocrystal-filled carboxymethyl cellulose nanocomposites. Journal of nanoscience and nanotechnology 2006, 6 (3), 633-639.

127

(187) Vieira, M. G. A.; da Silva, M. A.; dos Santos, L. O.; Beppu, M. M. Natural-based plasticizers and biopolymer films: A review. European Polymer Journal 2011, 47 (3), 254- 263.

(188) Lin, X.; Li, Y.; Chen, Z.; Zhang, C.; Luo, X.; Du, X.; Huang, Y. Synthesis, characterization and electrospinning of new thermoplastic carboxymethyl cellulose (TCMC). Chemical engineering journal 2013, 215, 709-720.

(189) Choi, J. S.; Park, W. H. Effect of biodegradable plasticizers on thermal and mechanical properties of poly (3-hydroxybutyrate). Polymer testing 2004, 23 (4), 455-460.

(190) Qiao, X.; Tang, Z.; Sun, K. Plasticization of corn starch by polyol mixtures. Carbohydrate Polymers 2011, 83 (2), 659-664.

(191) Talja, R. A.; Helén, H.; Roos, Y. H.; Jouppila, K. Effect of various polyols and polyol contents on physical and mechanical properties of potato starch-based films. Carbohydrate polymers 2007, 67 (3), 288-295.

(192) Yin, Y. J.; Yao, K. D.; Cheng, G. X.; Ma, J. B. Properties of polyelectrolyte complex films of chitosan and gelatin. Polymer international 1999, 48 (6), 429-432.

(193) Liu, H.; Adhikari, R.; Guo, Q.; Adhikari, B. Preparation and characterization of glycerol plasticized (high-amylose) starch–chitosan films. Journal of Food Engineering 2013, 116 (2), 588-597.

(194) Lin, R.; Li, A.; Lu, L.; Cao, Y. Preparation of bulk sodium carboxymethyl cellulose aerogels with tunable morphology. Carbohydrate polymers 2015, 118, 126-132.

(195) Yu, M.; Li, J.; Wang, L. KOH-activated carbon aerogels derived from sodium carboxymethyl cellulose for high-performance supercapacitors and dye adsorption. Chemical engineering journal 2017, 310, 300-306.

(196) Salaün, F.; Bedek, G.; Devaux, E.; Dupont, D.; Gengembre, L. Microencapsulation of a cooling agent by interfacial polymerization: Influence of the parameters of encapsulation on poly (urethane–urea) microparticles characteristics. Journal of membrane science 2011, 370 (1-2), 23-33.

(197) Muscat, D.; Adhikari, B.; Adhikari, R.; Chaudhary, D. Comparative study of film forming behaviour of low and high amylose starches using glycerol and xylitol as plasticizers. Journal of Food Engineering 2012, 109 (2), 189-201.

(198) Pawlak, A.; Mucha, M. Thermogravimetric and FTIR studies of chitosan blends. Thermochimica acta 2003, 396 (1-2), 153-166.

(199) Soman, V. V.; Kelkar, D. S. In FTIR Studies of Doped PMMA‐PVC Blend System, Macromolecular symposia, Wiley Online Library: 2009; pp 152-161.

128

(200) Snejdrova, E.; Dittrich, M. Pharmaceutically used plasticizers. In Recent advances in plasticizers; InTech: 2012.

(201) Chaudhary, D. S. Competitive plasticization in ternary plasticized starch biopolymer system. Journal of applied polymer science 2010, 118 (1), 486-495.

(202) Moskala, E. J.; Howe, S. E.; Painter, P. C.; Coleman, M. M. On the role of intermolecular hydrogen bonding in miscible polymer blends. Macromolecules 1984, 17 (9), 1671-1678.

(203) Coleman, M.; Moskala, E. FTi. r. Studies of polymer blends containing the poly (hydroxy ether of bisphenol A) and poly (ε-caprolactone). Polymer 1983, 24 (3), 251-257.

(204) Mehra, N.; Li, Y.; Zhu, J. Small Organic Linkers with Hybrid Terminal Groups Drive Efficient Phonon Transport in Polymers. The Journal of Physical Chemistry C 2018.

(205) Shahbazi, M.; Ahmadi, S. J.; Seif, A.; Rajabzadeh, G. Carboxymethyl cellulose film modification through surface photo-crosslinking and chemical crosslinking for food packaging applications. Food Hydrocolloids 2016, 61, 378-389.

(206) Bader, H. G.; Göritz, D. Investigations on high amylose corn starch films. Part 2: water vapor sorption. Starch‐Stärke 1994, 46 (7), 249-252.

(207) Mu, L.; He, J.; Li, Y.; Ji, T.; Mehra, N.; Shi, Y.; Zhu, J. Molecular origin of efficient phonon transfer in modulated polymer blends: Effect of hydrogen bonding on polymer coil size and assembled microstructure. The Journal of Physical Chemistry C 2017, 121 (26), 14204-14212.

(208) Nykänen, V. P. S.; Härkönen, O.; Nykänen, A.; Hiekkataipale, P.; Ruokolainen, J.; Ikkala, O. An efficient and stable star-shaped plasticizer for starch: cyclic phosphazene with hydrogen bonding aminoethoxy ethanol side chains. Green Chemistry 2014, 16 (9), 4339-4350.

(209) Su, J.-F.; Huang, Z.; Yuan, X.-Y.; Wang, X.-Y.; Li, M. Structure and properties of carboxymethyl cellulose/soy protein isolate blend edible films crosslinked by Maillard reactions. Carbohydrate Polymers 2010, 79 (1), 145-153.

(210) Rani, M. S. A.; Rudhziah, S.; Ahmad, A.; Mohamed, N. S. Biopolymer electrolyte based on derivatives of cellulose from kenaf bast fiber. Polymers 2014, 6 (9), 2371-2385.

(211) Mu, L.; Shi, Y.; Wang, H.; Zhu, J. Lignin in ethylene glycol and poly (ethylene glycol): Fortified lubricants with internal hydrogen bonding. ACS Sustainable Chemistry & Engineering 2016, 4 (3), 1840-1849.

129

(212) Bao, C.; Song, L.; Wilkie, C. A.; Yuan, B.; Guo, Y.; Hu, Y.; Gong, X. Graphite oxide, graphene, and metal-loaded graphene for fire safety applications of polystyrene. Journal of Materials Chemistry 2012, 22 (32), 16399-16406.

(213) Bao, C.; Song, L.; Xing, W.; Yuan, B.; Wilkie, C. A.; Huang, J.; Guo, Y.; Hu, Y. Preparation of graphene by pressurized oxidation and multiplex reduction and its polymer nanocomposites by masterbatch-based melt blending. Journal of Materials Chemistry 2012, 22 (13), 6088-6096.

(214) Zhang, Y.; Han, J. Plasticization of pea starch films with monosaccharides and polyols. Journal of Food Science 2006, 71 (6), E253-E261.

(215) Ortega-Toro, R.; Jiménez, A.; Talens, P.; Chiralt, A. Properties of starch– hydroxypropyl methylcellulose based films obtained by compression molding. Carbohydrate polymers 2014, 109, 155-165.

(216) Lin, Y.; Jia, Y.; Alva, G.; Fang, G. Review on thermal conductivity enhancement, thermal properties and applications of phase change materials in thermal energy storage. Renewable and sustainable energy reviews 2018, 82, 2730-2742.

(217) Milian, Y. E.; Gutierrez, A.; Grageda, M.; Ushak, S. A review on encapsulation techniques for inorganic phase change materials and the influence on their thermophysical properties. Renewable and Sustainable Energy Reviews 2017, 73, 983-999.

(218) Jamekhorshid, A.; Sadrameli, S.; Farid, M. A review of microencapsulation methods of phase change materials (PCMs) as a thermal energy storage (TES) medium. Renewable and Sustainable Energy Reviews 2014, 31, 531-542.

(219) Pielichowska, K.; Pielichowski, K. Phase change materials for thermal energy storage. Progress in materials science 2014, 65, 67-123.

(220) Farid, M. M.; Khudhair, A. M.; Razack, S. A. K.; Al-Hallaj, S. A review on phase change energy storage: materials and applications. Energy conversion and management 2004, 45 (9-10), 1597-1615.

(221) Rathod, M. K.; Banerjee, J. Thermal stability of phase change materials used in latent heat energy storage systems: a review. Renewable and sustainable energy reviews 2013, 18, 246-258.

(222) Sharma, R.; Ganesan, P.; Tyagi, V.; Metselaar, H.; Sandaran, S. Developments in organic solid–liquid phase change materials and their applications in thermal energy storage. Energy Conversion and Management 2015, 95, 193-228.

(223) Mishra, A.; Shukla, A.; Sharma, A. Latent heat storage through phase change materials. Resonance 2015, 20 (6), 532-541.

130

(224) Hérault, D.; Rodembusch, F.; Campo, L.; Gingras, M.; Cerveau, G.; Corriu, R. J. Valorization of by-products of the sugar industry: New nanostructured hybrid materials containing sugar derived structures. Comptes Rendus Chimie 2010, 13 (5), 566-574.

(225) Solé, A.; Neumann, H.; Niedermaier, S.; Martorell, I.; Schossig, P.; Cabeza, L. F. Stability of sugar alcohols as PCM for thermal energy storage. Solar Energy Materials and Solar Cells 2014, 126, 125-134.

(226) Guex, W.; Klaeui, H.; Pauling, H.; Voirol, F., Reusable heat devices containing xylitol as the heat-storage material. Google Patents: 1981.

(227) Hormansdorfer, G., Latent heat storage material and use thereof. Google Patents: 1989.

(228) Nomura, T.; Tsubota, M.; Oya, T.; Okinaka, N.; Akiyama, T. Heat storage in direct- contact heat exchanger with phase change material. Applied thermal engineering 2013, 50 (1), 26-34.

(229) Ona, E. P.; Zhang, X.; Kyaw, K.; Watanabe, F.; Matsuda, H.; Kakiuchi, H.; Yabe, M.; Chihara, S. Relaxation of supercooling of erythritol for latent heat storage. Journal of chemical engineering of Japan 2001, 34 (3), 376-382.

(230) Nomura, T.; Tsubota, M.; Oya, T.; Okinaka, N.; Akiyama, T. Heat release performance of direct-contact heat exchanger with erythritol as phase change material. Applied Thermal Engineering 2013, 61 (2), 28-35.

(231) Yuan, M.; Ren, Y.; Xu, C.; Ye, F.; Du, X. Characterization and stability study of a form-stable erythritol/expanded graphite composite phase change material for thermal energy storage. Renewable energy 2019, 136, 211-222.

(232) Yang, Y.; Pang, Y.; Liu, Y.; Guo, H. Preparation and thermal properties of polyethylene glycol/expanded graphite as novel form-stable phase change material for indoor energy saving. Materials Letters 2018, 216, 220-223.

(233) Pei, A.; Zhou, Q.; Berglund, L. A. Functionalized cellulose nanocrystals as biobased nucleation agents in poly (l-lactide)(PLLA)–Crystallization and mechanical property effects. Composites Science and Technology 2010, 70 (5), 815-821.

(234) Wellen, R.; Rabello, M. The kinetics of isothermal cold crystallization and tensile properties of poly (ethylene terephthalate). Journal of materials science 2005, 40 (23), 6099-6104.

(235) Tang, Z.; Zhang, C.; Liu, X.; Zhu, J. The crystallization behavior and mechanical properties of polylactic acid in the presence of a crystal nucleating agent. Journal of Applied Polymer Science 2012, 125 (2), 1108-1115.

131

(236) Mano, J. F.; Wang, Y.; Viana, J. C.; Denchev, Z.; Oliveira, M. J. Cold crystallization of PLLA studied by simultaneous SAXS and WAXS. Macromolecular Materials and Engineering 2004, 289 (10), 910-915.

(237) Diarce, G.; Quant, L.; Campos-Celador, Á.; Sala, J.; García-Romero, A. Determination of the phase diagram and main thermophysical properties of the erythritol– urea eutectic mixture for its use as a phase change material. Solar Energy Materials and Solar Cells 2016, 157, 894-906.

(238) Jesus, A. L.; Nunes, S. C.; Silva, M. R.; Beja, A. M.; Redinha, J. Erythritol: Crystal growth from the melt. International journal of pharmaceutics 2010, 388 (1-2), 129-135.

(239) Wiese, M. DSC Detection of polymorphism in pharmaceutical anhydrous dexamethasone acetate. TA Instruments 2009, 302, 1-2.

(240) Karthik, M.; Faik, A.; Blanco-Rodríguez, P.; Rodríguez-Aseguinolaza, J.; D’Aguanno, B. Preparation of erythritol–graphite foam phase change composite with enhanced thermal conductivity for thermal energy storage applications. Carbon 2015, 94, 266-276.

(241) Shen, S.; Tan, S.; Wu, S.; Guo, C.; Liang, J.; Yang, Q.; Xu, G.; Deng, J. The effects of modified carbon nanotubes on the thermal properties of erythritol as phase change materials. Energy conversion and management 2018, 157, 41-48.

(242) Telford, R.; Seaton, C. C.; Clout, A.; Buanz, A.; Gaisford, S.; Williams, G. R.; Prior, T. J.; Okoye, C. H.; Munshi, T.; Scowen, I. J. Stabilisation of metastable polymorphs: the case of paracetamol form III. Chemical Communications 2016, 52 (81), 12028-12031.

132

APPENDIX A

133

134

135

136

137

138

139

140