University of Central Florida STARS

Electronic Theses and Dissertations, 2004-2019

2018

Polyelectrolyte Complexes Based on Poly(acrylic acid): Mechanics and Applications

Xiaoyan Lu University of Central Florida

Part of the Chemistry Commons Find similar works at: https://stars.library.ucf.edu/etd University of Central Florida Libraries http://library.ucf.edu

This Doctoral Dissertation (Open Access) is brought to you for free and open access by STARS. It has been accepted for inclusion in Electronic Theses and Dissertations, 2004-2019 by an authorized administrator of STARS. For more information, please contact [email protected].

STARS Citation Lu, Xiaoyan, "Polyelectrolyte Complexes Based on Poly(acrylic acid): Mechanics and Applications" (2018). Electronic Theses and Dissertations, 2004-2019. 5769. https://stars.library.ucf.edu/etd/5769 POLYELECTROLYTE COMPLEX BASED ON POLY(ACRYLIC ACID): MECHANICS AND APPLICATIONS

by

XIAOYAN LU M.S. Beijing Institute of Technology, 2012 B.S. Taishan University, 2009

A dissertation submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy in the Department of Chemistry in the College of Science at the University of Central Florida Orlando, Florida

Spring Term 2018

Major Professor: Lei Zhai c 2018Xiaoyan Lu

ii ABSTRACT

Poly(acrylic acid) (PAA) is a weak polyelectrolyte presenting negative charge at basic condition when the carboxylic group loses a proton. These carboxylate group can interact with polycations and metal ions to form stable polyelectrolyte complexes (PECs), leading to tunable properties and multifunctional nanoscale structures through chemical reactions. This research focuses on nanofiber and nanoparticle fabricated by PAA-based PECs. We demonstrated the effect of ferric ion concentration on the mechanical properties of PAA-based single naonofiber by using dark field microscopy imaging and persistence length analysis. The application of PAA-based nanofiber

mats loaded with MnO2 for supercapacitors was also explored. As a free-standing and flexible supercapacitor electrode, the nanofiber mat exhibited outstanding properties including high specific capacitance, excellent reversible redox reactions, and fast charge/discharge ability. Since PAA is a biocompatible polymer, PAA-based PEC was applied as a drug-carrier in a drug delivery system. In this project, core-shell nanoparticles were fabricated with chitosan as the core and PAA as the shell to incorporate with the drug gemcitabine. Several parameters were investigated to obtain the optimal nanoparticle size. The as-prepared drug delivery system shows prolonged releasing profile.

iii ACKNOWLEDGMENTS

I am extremely thankful to Dr. Lei Zhai for being my dissertation advisor. I do not have enough words to describe my respect and gratitude for him. His constant support towards my research and academics in general made me more professional. I feel extremely thankful to to have him as my mentor and supervisor. He is not only an advisor, but also a good friend.

I am also very grateful to Dr. Shengli Zou, Dr. Demitry Kolpashchikov, Dr. Karin Chumbimuni- Torres, and Dr. Yajie Dong for being part of my dissertation committee and for their encouraging and valuable comments and remarks. Each one of them is my role model for study and research.

I gratefully acknowledge the support provided by the department of Chemistry, Nanoscience and Technology Center (NSTC), and Advanced Materials Processing and Analysis Center (AMPAC).I am also thankful to Ernie, Ted, Kirk, Matt, Mikhail and Karen for their help in the usages of instrumentations.

I would also like to thank all my lab-mates Astha Malhotra, Jean Calderon, Matthew McInnis, Chen Shen, Nilab Azim, Zeyang Zhang, Elizabeth Barrios; my undergraduate researchers Ruginn Catarata and Olivia McClellan and the colleagues, like Angie Diaz from collaborating labs. It was a pleasure working with them.

I am also immensely thankful to my friends who helped my stay here more enjoyable and my friends in China and around whose encouragement has really been a great support to me here in a foreign land.

Words or expressions are not enough to thank my husband and parents, back home for their pa- tience, support, understanding and sacrifices. I have to thank my sisters for being there for me and encouraging me. My baby Daniel highly motivates me to acheive the study accomplishement.

iv I would also like to express my thanks and sincere regards to everyone else whose help, encour- agement and prayers must have strengthened me and helped me achieve things in life.

Finally but most importantly, I have to thank GOD for his continuous love and amazing grace. Without him, I couldn’t do anything.

v TABLE OF CONTENTS

LIST OF FIGURES ...... xi

LIST OF TABLES ...... xx

CHAPTER 1: INTRODUCTION ...... 1

Polyelectrolyte ...... 1

Polyelectrolyte complex ...... 4

Nanostructure of polyelectrolyte and PEC ...... 8

PEC nanoparticle ...... 8

PEC nanofiber ...... 10

PEC membrane ...... 14

PEC hydrogel ...... 16

Application of polyelectrolyte complex ...... 19

Biotechnology ...... 19

Tissue engineering ...... 19

Drug delivery ...... 21

Gene delivery ...... 23

vi Wound self-healing ...... 25

Actuator ...... 27

Energy storage ...... 28

Theme of the research ...... 29

CHAPTER 2: TUNABLE EFFECT OF METAL IONS ON POLYELECTROLYTE NANOFIBER MECHANICAL PERFORMANCE ...... 31

Abstract ...... 31

Introduction ...... 31

Materials and Methods ...... 35

Nanofiber synthesis by elecrospinning ...... 35

Scanning electron microscopy and transmission electron microscopy ...... 35

Fourier Transform Infrared Spectroscopy (FTIR) ...... 36

Confocal microscopy imaging and bending mechanics data analysis ...... 36

Results and Discussion ...... 37

Ferric ions modulate the mechanical properties and average length of polyacrylic acid nanofibers...... 37

Chitosan helps ferric ions modulate the mechanical properties of PAA nanofibers. . 43

Electrospun PAA:CS nanofibers could exist in hydrated states...... 45

vii Characterization sheds light on how ferric metal ions affect individual PAA:CS nanofiber structure...... 47

Conclusion ...... 51

CHAPTER 3: FLEXIBLE CORE-SHELL COMPOSITE FIBERS FOR HIGH PERFOR- MANCE SUPERCAPACITOR ELECTRODES ...... 52

Abstract ...... 52

Introduction ...... 53

Experimental Procedure ...... 55

Materials ...... 55

Materials synthesis ...... 56

Fabrication of Nanofibers ...... 56

Creating MnO2 Nanoparticles on PAA Nanofibers ...... 56

Depositing the Polypyrrole Shell on the Surface of MnO2@PAA Nanofibers ...... 57

Characterization ...... 57

Results and Discussion ...... 59

Preparation of supercapacitor electrodes ...... 59

Electrochemical Characterization ...... 77

viii Conclusion ...... 83

CHAPTER 4: GEMCITABINE LOADED POLY(ACRYLIC ACID)-CHITOSAN NANOPAR- TICLES FOR TARGETED THERAPY OF PANCREATIC CANCER . . . . 84

Abstract ...... 84

Introduction ...... 84

Experimental Procedure ...... 87

Materials ...... 87

Fabrication of Nanofibers ...... 88

Characterization of the nanoparticles ...... 89

Incubation and encapsulation ...... 90

In vitro drug releasing ...... 90

Results and Discussion ...... 91

Effect of CS:TPP ratio on CS/TPP nanoparticle properties ...... 91

Effect of PAA:CS ratio on CS/TPP nanoparticle properties ...... 93

Effect of CS:EDC ratio on CS/TPP-PAA nanoparticle properties ...... 95

Effect of pH value on CS/TPP-PAA nanoparticle properties ...... 97

In vitro drug releasing study ...... 99

ix Conclusion ...... 100

CHAPTER 5: SUMMARY ...... 101

APPENDIX A: LIST OF PUBLICATIONS ...... 103

APPENDIX B: PERMISSIONS FOR COPYRIGHTED MATERIALS ...... 105

Permission Request Policies from ACS Publications ...... 106

LIST OF REFERENCES ...... 107

x LIST OF FIGURES

Figure 1.1: Molecular structures of common polyelectrolytes: (a) PSS, (b) PAA, (c) CS, (d) PDADMAC, (e) PAH, and (f) PPy...... 3

Figure 1.2: Schematic illustration of three ways in which the addition of a polyelec- trolyte may affect the colloidal stability of negatively charged particles in an aqueous suspension [1]...... 4

Figure 1.3: PEC formation by direct method (a) and indirect method (b)...... 5

Figure 1.4: Mechanism of PEC formation [2]...... 6

Figure 1.5: Different dimentions of PEC materials ...... 8

Figure 1.6: Schematic illustration of PEC nanoparticles of polyanion and polycation at different charge ratios [3]...... 10

Figure 1.7: Setup of electrospinning [4]...... 11

Figure 1.8: Different poly(L-lactide) (PLA) nanofiber forms at different condition: (A) porous fibers with beads; (B)uniform fibers with high porousity; (C) belt- shaped solid fibers; and (D) aligned uniform solid fibers with a circular cross section. [5]...... 13

Figure 1.9: Schematic illustrations of the nanoparticles formation on nanofibers from metal ions with different diffusion profile. [6] ...... 14

Figure 1.10: Illustration of layer-by-layer assembly...... 16

xi Figure 1.11: Deswell and swelling of PEC hydrogels...... 18

Figure 1.12: Basic principles of tissue engineering [7]...... 20

Figure 1.13: The various uptake pathways of macromolecules and nanoparticles into cells according to size [8]...... 22

Figure 1.14: The intracellular journey of gene delivery vectors [9]...... 24

Figure 1.15: Schematic mechanism of PEC multilayers for self-healing...... 27

Figure 1.16: Schematic mechanism of PEC hydrogel actuator...... 28

Figure 2.1: (A) Schematic diagrams of polyacrylic acid (PAA) nanofiber fabrication process by electrospinning. Functional groups with negative charges on PAA and ferric ions associate through ionic interactions. Chitosan (CS) and ferric metal ions (Fe3+) provide positive charges to create crosslinks in the system, therefore making it more resistant to environmental changes. (B) Scanning electron microscopy image of a polyacrylic acid nanofiber network. Scale bar represents 500 nm. (C) Transmission electron mi- croscopy image of a single PAA/CS nanofiber complex. Scale bar repre- sents 200 nm...... 34

Figure 2.2: Formation of PAA:CS:Fe3+ nanofibers at different concentrations after contrast enhancement. Representative darkfield images of nanofibers at different concentrations of ferric metal ions show qualitative changes in length and stiffness. Scale bar represents 300 µm...... 39

Figure 2.3: Illustration of θ(s) and θ(0)...... 40

xii Figure 2.4: (A) Changes in ferric concentration affect angular correlation of nanofibers. The stiffer the fiber, the closer the correlation is to a value of 1. (B) Ferric ion modulation of bending stiffness in PAA:CS nanofibers is determined by persistence length analysis revealing that interaction between ferric and PAA:CS raises stiffness of the nanofibers in a non-linear manner. (C) Av- erage length distribution of all nanofiber samples. The shortest nanofibers at the lowest possible concentration are created at 0.10% ferric metal ions, with ion effects also revealing a non-linear pattern of effect. (N=630-2200). 41

Figure 2.5: Schematic representation of the Fe-PAA (a) and Fe-CS (b) [10] complex. . 42

xiii Figure 2.6: Comparison analysis of length (A) and bending stiffness(C) of nanofibers electrospun with and without chitosan at initial concentrations 0%, 0.04%, 0.07%, and 0.10% of ferric metal ions. (A) Nanofibers with no chitosan (circles) show a higher average nanofiber length that falls dramatically as the concentration of ferric rises. (B) A point for control nanofibers contain- ing no CS or Fe3+ ions could not be accurately measured due to length sur- passing imaging limits of magnification. Scale bar represents 250 µm. (C) Nanofibers with chitosan (triangles) did have measurable lengths that be- gan well below estimations for PAA only nanofibers, and therefore showed more overall stability in length changes. Nanofibers with no chitosan (cir- cles) at 0%, 0.04%, and 0.07% concentrations of ferric have lower persis- tence length and therefore stiffless than the average nanofiber with chitosan (triangles). At concentration 0.07% and 0.10% of ferric, both nanofibers with and without chitosan converge at similar persistence length averages, as well as highest impact concentration of ferric on overall stiffness of the nanofibers. Bending stiffness measurements were acquired on fragments of nanofibers that could be captured in span of image...... 43

xiv Figure 2.7: Comparison analysis of nanofiber length (A) and bending stiffness(B) of nanofibers electrospun with and without chitosan at concentrations 0%, 0.04%, 0.07%, and 0.10% of ferric metal ions in box plots. (A) Length distribution comparison. The box range represents quartiles of 10% and 90% and reveals the effect of ferric metal ions and chitosan on each set. Nanofibers with no chitosan (black) show a wider variation of lengths than nanofibers with chitosan. A box was not created for nanofibers with no chi- tosan due to their lengths surpassing imaging limitations. (B) Nanofibers with no chitosan (black) mostly have lower persistence length and there- fore stiffless than the average nanofiber with chitosan (red). At concentra- tion 0.10% of ferric, both nanofibers with and without chitosan converge at similar persistence length averages, as well as highest impact concen- tration of ferric on overall stiffness of the nanofibers. Bending stiffness of nanofibers without chitosan or ferric were obtained on filament fragments that fit within the imaging window...... 46

Figure 2.8: Mean diameter is represented by the yellow columns, and young’s mod- ulus is represented by the blue columns in the kPa range for individual nanofibers in their hydrated state and show optimal estimates at 0.10% ferric metal ion concentration...... 47

xv Figure 2.9: FTIR and conductivity analysis of nanofibers with varying ferric concen- trations. (A) The PAA:CS solutions with differing metal ion amounts were dried in vacuum for 3 days at 60◦C. Ferric metal ions influence the vibra- tion of bonds in the system, which is reflected in the shifting and strength- ening of certain peaks. (B) Measurement of conductivity difference cre- ated by Fe3+ and HCl used in nanofiber preparation. Polyacrylic acid and chitosan have low conductivity that rises sharply with addition of ferric metal ions and hydrochloric acid. There was no real conclusive pattern to conductivity for further differentiation...... 48

Figure 3.1: The scheme of the fabrication of the MnO2@PAA/PPy core-shell nanofibers from PAA electrospun fibers loaded with Mn2+ ions...... 55

Figure 3.2: TEM images of (a) α-MnO2 and (b) β-MnO2nanoparticles...... 60

Figure 3.3: (a) TEM and (b) SEM image for PAA/MnO2 nanofibers fabricated by di- rect method...... 61

Figure 3.4: TEM image of a MnO2@PAA nanofiber produced by a post-loading ap- proach...... 62

Figure 3.5: SEM images of PAA/Mn2+ (insert is TEM image of PAA/Mn2+, and the scale bar is 50 nm) (a) and PAA fibers (b)...... 63

2+ Figure 3.6: SEM images of PAA/Mn fibers cross-linked by 0.2 M FeCl3 (inset is a higher resolution SEM image of the fiber) (a), the cross-linked fibers after

18 h immersion in water (b), PAA@MnO2 (c), and PAA@MnO2/PPy (d) fibers after 18 month immersion in water...... 64

xvi 2+ Figure 3.7: SEM images of crosslinked PAA/Mn fibers ((a) and (d)), MnO2@PAA

fibers ((b) and (e)), and MnO2@PAA/PPy fibers ((c) and (f)) immersed in water for 2 days and 5 days...... 65

Figure 3.8: TEM image of images of PAA/MnO2 fibers after one-time KMnO4 oxida-

tion(a).High magnification image (b) shows that MnO2 particle size is less

than 5 nm. (c)TEM image of MnO2@PAA fibers after four-time KMnO4 oxidation...... 67

Figure 3.9: SEM images of PAA/MnO2 fibers produced through one (a), two (b), three

(c), and four (d) times oxidation using KMnO4. Insets are the digital pic- tures of fibers showing a darker color with multiple oxidations...... 68

2+ Figure 3.10: XPS spectrum of PAA/Mn fibers (a) and fibers before KMnO4 oxidation (b)...... 71

Figure 3.11: (a) Deconvoluted high-resolution XPS spectrum of the as-spun PAA/MnAc2

nanofibers. (b) Deconvoluted high-resolution XPS spectrum of PAA/MnAc2 fibers stabilized by Fe3+. (c) High-resolution XPS spectra of fibers after

different times of KMnO4 oxidation (14 represents the oxidation times). . . 72

Figure 3.12: XPS spectra of fibers after (a) one, (b) two, (c) three, and (d) four times of

KMnO4 oxidation...... 73

Figure 3.13: XPS spectrum of Mn 3s (a), and XRD image (b) of composites after

KMnO4 Oxidation...... 74

xvii Figure 3.14: (a) SEM image of the PAA@MnO2/PPy fiber produced by 1 min polymer- ization. The inset shows that the fibers are flexible. (b) HRTEM image of

the fiber showing about a 50 nm thick PPy shell and small MnO2 particles inside the fiber...... 76

Figure 3.15: An SEM image of PAA@MnO2/PPy fiber produced by 5 minutes (a) and 10 minutes (b) polymerization...... 77

Figure 3.16: CV measurements of (a) PAA@MnO2, (b) PAA@MnO2/PPy at different

3+ 3+ scan rates, and (c) PAA/Fe , PAA/Fe /PPy, PAA@MnO2, and PAA@MnO2/PPy at a scan rate of 10 mV/s...... 78

Figure 3.17: GCD measurements of PAA@MnO2 (a) and PAA@MnO2/PPy (b). The insets are specific capacitance at different current densities...... 80

Figure 3.18: Cycling performance of PAA@MnO2 and PAA@MnO2/PPy at a scan rate of 100 mV/s...... 81

Figure 3.19: SEM images of PAA@MnO2 (a) and PAA@MnO2/PPy (b) after 5000 cy- cles at a scan rate of 100 mV/s...... 82

Figure 3.20: EIS of PAA@MnO2 and PAA@MnO2/PPy...... 82

Figure 4.1: Schematic representations of Gemcitabine loaded PAA/CS/TPP nanoparticle 89

Figure 4.2: Schematic representations of Gemcitabine loaded PAA/CS/TPP nanoparticle. 92

Figure 4.3: TEM image of CS/TPP core nanoparticles (a) and GEM loaded CS/TPP/PAA nanoparticles (b)...... 93

xviii Figure 4.4: Effect of PAA:CS ratio on PAA-CS/TPP nanoparticle size (black line) and zeta potential (red line)...... 95

Figure 4.5: Effect of CS:EDC ratio on PAA-CS/TPP nanoparticle size (black line) and zeta potential (red line)...... 96

Figure 4.6: Effect of pH value on PAA-CS/TPP nanoparticle size (black line) and zeta potential (red line)...... 98

Figure 4.7: In-vitro release profile of GEM in nanopure water. Experiments were car- ried out at 37 ◦C. Values are the average of three different experiments standard deviation...... 100

xix LIST OF TABLES

Table 3.1: Summary of weight% and atom% of Mn from EDS equipped on SEM . . . 69

Table 3.2: Summary of Elemental Compositions from XPS ...... 75

3+ 3+ Table 3.3: Specific capacitance of PAA/Fe , PAA/Fe /PPy, PAA@MnO2 and PAA@MnO2/PPy at a scan rate of 10 mV/s...... 79

xx CHAPTER 1: INTRODUCTION

Polyelectrolyte

Polyelectrolytes are polymers which are with ionic groups in the structures. The basic architecture of a polyelectrolyte contains a polymeric backbone and repetitive units of charge bearing func- tional group(s) on it [11]. These functional groups undergo partially or completely dissociate in polar solvents like water to be either positively or negatively charged. This property makes poly- electrolyte adopt similar properties of both electrolyte and polymer. For example, polyelectrolyte solution can be both electrically conductive and viscous. Further more, while neutral polymers em- ploy coil conformation in solution, polyelectrolytes stretch out more due to the repulsion among the charges. The extended structure leads to higher viscosity in the solution because the polymer molecules need more space and resist the stream of the solvent [12].

Polyelectrolytes can be devided as strong and weak types. Like strong and weak electrolytes, polyelectrolytes can completely or partially dissociate. Strong polyelectrolytes are those which bear functional groups thatare dissociate extensively and not dependent on pH, like sulfonic group and quaternary ammine group. These polyelectrolytes keep highly charged in most conditions and their surface charge is not easily altered. Poly(diallyldimethyl ammonium chloride) (PDAD- MAC) bearing quaternary amine as functional group, is a good example for strong polyelectrolyte. It retains its positive charge over a wide range of pH. Similarly, poly(allylamine hydrochloride) (PAH) is a strong anionic polyelectrolyte, on which the sulfate group can completely dissociate. In contrast, weak polyelectrolytes have functional groups that are pH dependent and partially disso- ciate, such as polyelectrolytes with primary or secondary amine (Chitosan), and carboxylic group (Poly(acrylic acid), PAA) (Figure 1.1).

1 The properties of weak polyelectrolyte, such as the degree of dissociation and polymer chain con- formation, can be changed by modifying the pH, counter-ion concentration, and ionic strength of the solution. For instance, CS is a weak polyelectrolyte containing primary amine which is posi- tively charged only at acidic condition by accepting a proton. PAA is also a weak polyelectrolyte that is negatively charged only at basic condition when the carboxylic group loses a proton. In a so- lution without salt, strong coulombic repulsion between charges on the polyelectrolyte backbone stretches and elongates the polymer chains. There are many counter-ions surrounding the poly- mer blobs, which makes the solution maintain charge neutrality. However, if extra counter-ions, such as alkali metal salt, are added to the solution, the electrostatic repulsion between polyelec- trolyte backbones will be screened, therefore, the polymer blob will become smaller and more compact [13]. The viscosity of polyelectrolyte solution decreases when the polymer concentra- tion is dilute and semidilute entangled, while increases when the concentration is higher than the critical concentration due to the local charge inversion, which causes chain expansion among the entanglements [14]. Normally, natural polyelectrolytes, such as proteins and DNA, and natural de- rived polyelectrolytes, such as cellulose, chitosan (CS), alginin, heparin, are weak, while synthetic polyelectrolytes (e.g. PAH, PAA, poly styrenesulfonate (PSS), PDADMAC, etc.) contain both weak and strong ones. These polymers can be synthesized by solution polymerization [15, 16], emulsion polymerization [19, 20], and precipitation polymerization [17, 18].

According to the nature of the charges on the polymer backbone at neutral pH, polyelectrolyte also can be divided into polyanion, polycation and polyampholyte. Polycation are usually heteroatom based (charges on nitrogen, phosphorus ,etc.) with polymers containing ammonium, phosphonium, orimidazolium groups being particularly common [9]. For example, PAH, CS and PDADMAC are cationic polyelectrolytes. and PAA and PSS are anionic due to their negative charged carboxyl and sulfate groups. Polyelectrolytes containing both positive and negative charges are polyampholytes, which are common, such as proteins. A special type of polyampholytes that carry both positive

2 and negative terminals in the each repetitive unit are termed as polybetaines, which are rather rare and usually sized in nature [21]. The coexistence of positive and negative charge on the backbone leads to more complicated physical behavior. These polyelectrolytes only dissolve in the solutions with enough salt, for the ions from the salt can screen the interaction between the opposite charges. Figure 1.1 shows the chemical molecular structures of some common polyelectrolytes.

Figure 1.1: Molecular structures of common polyelectrolytes: (a) PSS, (b) PAA, (c) CS, (d) PDADMAC, (e) PAH, and (f) PPy.

The charges and following properties on the polyelectrolytes results in many new and more ap- plications than that of neutral polymers. Polyelectrolytes can be used in many fields in engineer- ing and technology for nanomaterial thickening and binding [22, 23], flocculation [24–26], oil recovery [27, 28], etc. Colloids and dispersions with particle size 1 nm to 10 µm are used in many processes, such as biotechnology, agriculture and new energy [24, 29, 30]. With the charge on the surface, polyelectrolyte can absorb the particles in colloids and dispersions to stabilize or destabilize them [31]. Nanomaterial thickening and binding is the example of stabilization, while floccuation and oil recovery are the examples of destabilization. Figure 1.2 is the illustration of the effects of polyelectrolyte on colloidal stability of particles with negative charge in aqueous solution [1]. D.A. Sievers et al. [24] empolyed polyamide as polyelectrolyte flocculant to purify the corn stover by removing the solids from the sugar streams after enzymatic hydrolysis of corn stover. After the treatment of polyelectrolyte flocculant, the filtration capacity was enhanced up to 40 folds compared with untreated, while the used of cellulosic filter-aids for pressure fultration

3 was only slightly improvement [32]. Also, with this method, the cost of the fuel was decreased by $1.35 US per gallon gasline equivalent.

Figure 1.2: Schematic illustration of three ways in which the addition of a polyelectrolyte may affect the colloidal stability of negatively charged particles in an aqueous suspension [1].

Polyelectrolyte complex

When two polyelectrolytes carrying opposite charges encounter, polyelectrolyte complex (PEC) is formed through the electrostatic interaction between the opposite charges [33, 34], as shown in Figure 1.3. There are two pathways to fabricate PECs: direct and indirect method. Direct method is that just mix the polyelectrolytes with different charges, and the PECs form by self-assmbly [35– 37] (Figure 1.3(a)), while indirect method starts with one polyelectrolyte and monomers carrying the other kind of charge, and then undergoes template polymerization [38–41] (Figure 1.3(b)). The PECs chains may go furthe self-assmbly after polymerization. For direct method, positively

4 charged polyelectrolytes and negatively charged polyelectrolytes cooperate with each other ran- domly, while PECs formed by indirect method are with better arrangement. When the opposite charges on the two polymers are stoichiometric (every charge on one polymer is connected to the opposite charge on the other polymer), the PECs are more rigid than those when the charges are non-stoichiometric(one polymer is excess compared with the other one by charge), which is more hydrophilic and soluble [42]. The formation of PECs involves not only electrostatic interaction between the polymers, but also the interactions between the polymer and other molecules in the solution, such as hydrogen bonding, van der waals interaction, hydrophobic interaction, and dipole interactions [34].

Figure 1.3: PEC formation by direct method (a) and indirect method (b).

The self-assembly of the PECs also depends on chain conformation of the polyelectrolytes. The process of PECs formation contains several molecular interactions levels (Figure 1.4). In the polyelectrolyte solution, there are polymer chains with charges and the counter ions with low molecular weight (they are released during the dissociation process). At the beginning of mix-

5 ing, polymers are absorbed to each other by electrostatic interaction, which results in polymer arrangement. Polyions starts to partially connected to each other due to the originally connected to counter ions, and the connection will be elongated with longer time. At the same time, PECs reconfiguration comes up on both connected and unconnected part. This leads to the secondary and steric structure of PECs, and it takes time to achieve dynamic equilibrium [2, 34, 43, 44]. The absorption/connection process and reconfiguration process can be controlled by many parameters, such as charge ratio, polyelectrolyte concentration, pH value, etc. [34]. Different PEC structures and properties can be obtained at different stage [2], as shown in Figure 1.4.

Figure 1.4: Mechanism of PEC formation [2].

6 The electrostatic attraction, which is non-covalent bond, brings two main advantages: 1) it takes place without the presence of catalyst or crosslinker, therefore, PEC can be non-toxic and simply fabricated with low cost as long as both of the polymers are non-toxic; 2) it makes PECs more flexible and more sensitive to the pH, temperature and charge concentration of the solution than that of complexes formed by covalent bonding [45, 46]. This interaction results a more stable complex than that of only one polyelectrolyte in aqueous soluton. For example, PAA are sensitive to ionic environmentscan and can immediately dissolve in water solution, and chitosan has fast gastric acid dissolution. However, when these two polymer solution are mixed, even at low concentration (0.02% and 1% for chitosan and PAA, respectivly), PAA-CS nanoparticles are formed and no longer dissolve in water [47].

Interactions between metal ions, especially heavy metal, and polyelectrolytes, especially biopoly- electrolytes plays an important role in building PEC networks, which is also widely employed in nature, such as mussel byssal threads, nereis jaw and insect mandibles [6, 48–50]. In this interac- tion, polyelectrolytes act as the ligand of metal ions, because the interaction usually occurs between metal ions and the charged functional groups on polyelectrolyte by metal-ligand coordination in- teraction, which is stronger than the electrostatic attraction and is more stable in water [51]. There- fore, this interaction provides structural stability of PEC and results in adaptive and responsive structures [6]. The strength of the interaction between metal ions and sodium alginate was inves- tigated by Maturana etal. [52]. Their result was Fe(II)>Fe(III)>Co(II) >Ni(II)>Pb(II)>Hg(II). Rivas et al. obtained the flollowing affinity order of polyelectrolyte-metal ion interaction: tri- valent >di-valent >mono-valent ion at pH 3, 5, and 7 and the highest interaction happened at pH 7 after studied the many metal ions, such as Ag+ , Ni2+, Cd2+, etc. [53] The polyelectrolyte they used was a water-soluble polymer containing hydrophilic groups in every monomeric unit synthe- sized by radical polymerization. Their result indicated that the electron-rich groups on the polymer paticipated the interaction with metal ions. The polymer-metal ion interaction depends on pH due

7 to the protonated and deprotonated equilibrium of the charged groups on the backbone [54].

Nanostructure of polyelectrolyte and PEC

The advancement of nanotechnology and its application is a significant revolution in the past sev- eral decades. Nanotechnology is about the materials and structures within nanoscale, which brings large surface area and much more active points than normal bulk materials. Therefore, these mate- rials can be used in a broad range of fields [55]. PECs are good materials that has been fabricated into 0D (nanoparticle), 1D (nanofiber), 2D (membrane) and 3D (hydrogel) forms (Figure 1.5).

Figure 1.5: Different dimentions of PEC materials

PEC nanoparticle

PEC nanoparticles are usually formed by adding one diluted polyelectrolyte solution dropwise to the other diluted solution that contains the oppositely charged polyelectrolyte under stirring [56, 57]. This is called co-precipitation or polyelectrolyte titration. The stability and size of the col-

8 loid are affected by many parameters, such as mixing procedure, charge ratio of polyelectrolyte, pH. Ankerfors et al. [58] demonstrated the effect of the mixing procedure of PAA and PAH on the PEC properties compared with two complexation techniques, polyelectrolyte titration and jet mixing. They observed that low-moluclar-weight, low concentration, and short mixing time pro- duced smaller size nanoparticles. The order of mixing is more important to the case when the titrant solution is added dropwise to the starting solution than that of one-shot addition. For the slow process, the insufficient polyelectrolyte should be added to the excess one to avoid aggrega- tion [59]. Muller et al. [60] reported that, for the nanoparticles fabricated by polyvinyl alcohol and polyethyleneimine (PEI) and PAA, when the insufficient polyelectrolyte was added to the excess one, the particle is more compact and smaller. The size of the particles also drop from ∼400 nm to ∼160 nm with decreasing of pH value (10, 8.5, 7, and 4). The zeta potential of PEC nanoparticles is affected by the charge ratio [3]. As shown in Figure 1.6, the zeta potential is kept as nearly zero when the charge ratio is 1, and it becomes negative or positive when polyanion or polycation is excess.

9 Figure 1.6: Schematic illustration of PEC nanoparticles of polyanion and polycation at different charge ratios [3].

PEC nanofiber

For nanofiber fabrication, there are several techniques: electrospinning, phase separation, chem- ical vapor deposition, and self-assembly. Among these technologies, electrospinning is the most widely used due to its simple setup (Firgure 1.7) and excellent performance [61–63] to conve- niently prepare non-woven fibrous materials with fine diameters ranging from submicron to sev- eral nanometers [21]. Briefly, the precursor polymeric solution is loaded into the syringe with a stainless steel needle. Then the fibers are produced by applying an electric field to the precursor solution between the needle and the collector. During the electrospinning process, the solution drop at the tip of the needle is stretched out by the electric field and stretched back by the surface tension. When the two forces get a balance, Taylor Cone is formed, which is the threshold to pro- duce nanofibers [4]. The fiberous structure and morphology can be affected by the design of the

10 needle and the collection setup [63, 64].

Figure 1.7: Setup of electrospinning [4].

With the properties, such as high and large surface area, the non-waven mat of electrospun nanofibers can be applied as scaffold and matrix for various applications [65]. In the past years, both poly- electrolyte itself and PECs attracted growing interest to prepare nanofibers. The charge and elec- trostatic interaction in polyelectrolyte and PEC have profound influence on the electrospinning process and the morphology of the fibers thereof[66]. Polyelectrolyte can increases the viscosity of the solution, therefore produces thicker nanofibers and lower beads productivity [66]. 3 wt% chitosan solution in trifluoroacetic acid was used for electrospinning to fabricate nanofibers [21].

11 The thickness of as-prepared fibers distributes from 100 nm to 1 µm containing belt-shape fibers. Hyaluronic acid/chitosan PEC-based core-shell nanofibers prepared by electrospinning was re- ported previously [67]. The slectrospinning setup employed a built-in electric field to form a core-shell structrue by CS and hyaluronic phase separation before fiber formation. Dr. Xia’s group reported that poly(L-lactide) (PLA) can be electrospun to produce beaded porous fiber, uniform fibers with high porousity , belt-shaped solid fibers and aligned uniform solid fibers at different feeding rate, polymer concentration, in different solvent, or salt concentration, as shown in Fig- ure 1.8. [5] PAA/CS polyelectrolyte complex with different metal ions (Ca2+, Ce3+, Fe3+) was adapted to fabricate nanofibers impregnating metallic nanoparticles [6]. It was shown that the in- teraction between polyelectrolyte and metal ions affected the morphology of the fiber because the conductivity and viscosity change with different concentration and valent state of metal ions. The position and shape of the metallic nanoparticles formed on the fibers were also influented by the interaction and reaction rate, as shown in Figure 1.9. The swelling of PEC nanofibers is a property that is different with normal polymer fibers due to the intermoluclar interaction, such as hydrogen bonding. Nanofibers also can be folded and designed to fabricate 2D and 3D bulk structure by the collection setup [5].

12 Figure 1.8: Different poly(L-lactide) (PLA) nanofiber forms at different condition: (A) porous fibers with beads; (B)uniform fibers with high porousity; (C) belt-shaped solid fibers; and (D) aligned sylinder-shaped uniform solid fibers. [5].

13 Figure 1.9: Schematic illustrations of the nanoparticles formation on nanofibers from metal ions with different diffusion profile. [6]

PEC membrane

Memberanes with various types and compositions have been investigated for decades to be applied in environmental engineering, biotechnology and medical devices [68–71]. Among the membrane fabrication technologies, layer-by-layer (LbL) assembly is a facile, efficient and versatile approach to fabricate tunable and ultra thin nanostructured multilayer polyelectrolyte membranes [72]. Poly- electrolytes are perfect materials for LbL assembly because of the charges on the backbone and electrostatic attraction is the dominating driving force of LbL assembly, which results in highly ef- ficient process and very stable structue [73, 74]. Therefore, polyelectrolyte films can be fabricated by the continuous adsorption of oppositely charged species on a charged substrate [72], as shown

14 in Figure 1.10. Depostion strategies has significant affects on weak polyelectrolyte LbL films at layering, stability, and chain mobility [75]. Shear forces was observed as the main reason for that becaue the forces arise during spin-assisted assembly, which leads to smaller adsorption amount, causing a more orderly polymer chain, and dramatically improve the stability of the films, than that of the dip-assisted assembly. The high degree of molecular order also increases the ionization degree of the weak polyelectrolytes, bringing lower chain mobility and stronger binding between layers. Electric field controlled LbL depositon is another approach for weak polyelectrolyte as- sembly at a high deposition rate. However, the ionization conditions of the multilayer is highly influenced by the solution conditions and electric field strength [76]. In situ crosslinking is another method to fabricate multilayer membrane. Recently, polyvinyl alcohol and PEI were crosslinked by glutaricdialdehyde to form a stable bipolar membrane [77].

Joseph et al. [72] reported an ultrathin selective layer , including only one single bilayer of PDAD- MAC and sulfonated poly(aryleneoxindole) with excellent separation properties by a simple two- step dip-coating procedure. Polyelectrolyte multilayer films containing weak and strong poly- electrolyte blocks has been reported [73]. They took poly(4-styrenesulfonic acid-co-maleic acid) (PSSMA) as the sample. The strong charged group styrenesulfonic acid formed electrostatic link- age to the substrate with PAH, while the weak charged group maleic acid can alter the properties because they are responsive to pH value of the environment. The thickness of the assembled film decreases with the higher pH. This is because the polycation chanis have lower charge density at low pH, and a more coiled confirmation is adapted. Yin et al. [78] investigated the LbL film build- ing up by dipping approach at vibration conditions. It was observed that the density of multilayer smoothness increases with the vibration frequency. This is becuase of the influence of vibration on the configuration of the polymer chain.

15 Figure 1.10: Illustration of layer-by-layer assembly.

PEC hydrogel

Polyelectrolyte and PEC 3D networks can be obtained by polymerization [79–81] or mixing poly- electrolytes carring opposite charge at high concentration and proper pH and temperature [82, 83]. PEC hydrogels can be formed by both covalent bond and non-permanent and reversible ionic bond [84]. Covalent linkage leads to higher immobilization and better mechanical properties, be- cause it is permanent and irreversible bonding. While ionic crosslinking is generally considered as biocompatible and well-tolerated, because it’s non-permanent and reversible. Therefore, ionically crosslinked hydrogels are more sensitive to the stimuli, such as pH, temperature, and electrical stimulus when in electrolytes [84–87]. The response can be swelling-deswelling process or me- chanical motion at different conditions [80] by absorbing or releasing water, but still remaining at solid state, as shown in Figure 1.12. The hydrogels response to pH due to the weak acid and amine groups on the polymer backbones. These groups and interactions between them are sensitive to pH changes especially around their pKa. For instance, hydrogel contraction of polyelectrolytes such as poly(sodium acrylate) (PAANa), depends on the repulsion of COO− groups, which is adjusted by the exchange of H+ ions. Since the electrostatic repulsion of COO− groups significantly depends

16 on the concentration of COO− groups, H+ exchange is the most important parameter to determine the number of COOH groups on the polymer backbone and therefore initiates polymer swelling and deswelling. Thus, the concentraiton H+ (pH) becomes a critical determinant for swelling and deswelling of the hydrogels [88]. For temperature change, it is known that increasing of temper- ature weakens the hydrogen bond between water molecules and acid/amine groups, which causes deswelling of the hydrogels. The thermal behavior of PEC hydrogels are studied previously. It was observed that Poly(N-isopropylacrylamide) (PNIPA)-based hydrogels can shrink and expand in volume on heating and cooling, respectively [85, 89]. Recently, Kim et al. [90] synthesized a PNIPA-based hydrogel containing titanate(IV) nanosheets (TiNSs) which reveals thermal motions in contrast with conventional PNIPA-based hydrogles. Namely, on heating and cooling, the hydro- gel elongates and shortens, respectively, as the result of the expansion and contraction of the co- facial TiNS distance. Electronic field causes ionic concentration asymmetrically to create osmotic pressure differences, which consequently leads to swelling/deswelling and bending of the hydro- gels [90]. The bending direction of the hydrogels is determined by the sign of the fixed charges on the polyelectrolyte network [79]. Solvent also can be a stimulus to trigger the shape change of PEC hydrogels. Zhao et al. [82] modified PAA with catechols and CS with bis(trifluoromethane-

− sulphonyl)imide (Tf2N ) to get a hydrogel that can be stable for solvent exchange (H2O/DMSO) with fast wet adhesion.

17 Figure 1.11: Deswell and swelling of PEC hydrogels.

Except for the tranditional methods to create PEC hydrogels by polymerization [79] or simply mix- ing and diffusion process, many new approaches have been utilized. Crosslinkers are introduced into the fabrication of hydrogel to stabilize the product and shorter the formation time [80, 81, 91]. Since greater microporosity leads to fast and high flexibility of PEC hydrogels [83], because high porousity generates large deformation which is a result of enhanced deswelling mechanisms [88]. 3D printing technique was reported to be a method to increase the porosity of the hydrogel, con- sequently increase the bending performance and sensitivity of the chitosan-based hydrogel [81]. Emulsion polymerization was adopted to create porous structures into PAANa hydrogel [88]. This kind of hydrogels is more sensitive and flexible because the porousity decrases the pH gradient across the hydrogel to protonate carboxylate groups and initiate bending. When it’s hard to fab- ricate hydrogels from one polyelectrolyte only, PEC shows advantages. For example, Polyaniline (PANI), as one the most promising conductive polymers, can’t form hydrogels with a facile method because of its high hydrophobicity. However, PANI-cellulose hydrogel was fabricated from a transparent solution dissolved in a NaOH-urea aqueous system by cross-linking with epichloro- hydrin [92]. In this PEC hydrogel, PANI-cellulose crosslinked networks entangled together by noncovalend bonding, resulting in a homogeneous macroposrous structure, good miscibility and

18 excellent mechanical strength.

The morphology of polyelectrolyte and PEC nanostructure can be characterized by transmission electron microscopy (TEM) and scanning electron microscopy (SEM). X-ray photoelectron spec- troscopy (XPS) and infrared spectroscopy (IR) are usually employed to determine the composition of the material. [66] Atomic force microscopy (AFM) can characerize the surface morphologies of the materials. [73]

Application of polyelectrolyte complex

Biotechnology

Tissue engineering

The field of tissue engineering uses the combination of cells and scaffolds to repair or replace por- tions of or whole biological tissues (Figure 1.12). Over the past few years, the definition of the tissue engineering building blocks has changed significantly from a material that functionalized only as an inert structural support for cell attachment, to serve as a more complex and dynamic environment for tissue development [93]. The scaffold for tissue engineering should be stable enough to permit cell infiltration and support tissue growth, however, it can’t stay too long to interfere with the connection of the new tissue and old tissue [94]. Furthermore, the surface of the implantable material has to be suitable to promote cell adhesion and regulate the inflamma- tion response [95, 96]. The mechanical property can be improved by adding “stiff” composition, modulating pH during assembly, carrying out cross-linking, or incorporating nanoparticles. The strategy for biochemical functionalization includes modification with growth factors and hormones or types of specific interactions, such as the interaction with receptor [97]. The ability to manipu-

19 late the chemical, physical, and surface properties of PEC structures by simply changing pH, ionic strength, thickness, and postassembly modifications make them highly suitable to detect the effects of environmental stimuli on cellular response and to be applied on tissue engineering [98].

Figure 1.12: Basic principles of tissue engineering [7].

PEC multilayer films made by layer-by-layer assemble is a promising technique to get a bioactive biomimetic surface to trigger a specific cell response and form a new tissue [97]. The properties of PEC multilayer films present highly suitability to immobilize the biomolecules. For example. PSS/PAH film is the most frequently studied synthetic PEC multilayer film, because it is a good substrate for cell adhesion and proliferation, which may be attributed partially to the presence of sulfonate groups [99]. This linaerly growing and dense film can also improve the mechanical prop- erties of the cryopreserved vessel. Natural polyelectrolytes, such as proteins, polysaccharides, and other bioactive molecules such as growth factors, can be used in multilayer construction to enhance cell adhesion and proliferation, because the tissues are originally composed by these molecules. For instance, when gelatin was associated with CS and deposited onto titanium films, the modified titanium substrates presents better proliferation and viability of osteoblast cells than the control surfaces after 1 day and 7 days of culture in vitro [100]. HA/CS core-shell nanofibers membrane

20 presented high biocompatibility for mouse fibroblast cell line [67]. Cells on the surface began to exhibit pseudopodia-like morphology with pseudopodial protrusions after one-day cultivation, and the pseudopod-like cells had widely proliferated across the surface of the fiber membrane after three-day cultivation.

Hydrogels, with structural and compositional similarities to biological tissues and their extensive framework for cell survival, have attracted considerable attention for tissue engineering [101, 102]. Comparing with conventional hydrogels, polyelectrolyte gels show unique responses to environ- mental stimuli, such as electrically induced chemomechanical contraction similar to biological re- sponses [103, 104]. Furthermore, the natural polyelectrolytes, such as polysaccharide and charged filamentous proteins, are highly biocompatible. If neutral polymers are introduced into polyelec- trolyte networks, the gel can have better mechanical properties close to native cartilage [105, 106]. In the double network gel composed of PAMPS and PDMAAm (poly(2-acrylamide-2-methyl- propane sulfonic acid)/poly(N, N-dimethyl acrylamide) DN gel), the final stress and the tangent modulus were significantly enhanced from 3.10 and 0.20 MPa to 5.40 and 0.37 MPa, respectively, while the water content was significant reduced after implantation (94 to 91%) [107]. This signifi- cantly solved the problem of the limitation by weak mechanical behavior for artificial cartilage.

Drug delivery

Drug delivery is a technology of transporting a pharmaceutical compound into the confocal tissue to safely achieve its desired therapeutic effect (Figure 1.13). PECs have attracted high interest in the field of drug delivery due to their sensitivity to specific conditions, such as pH, temperature, magnetic and electric fields. Also, the charge on polyelectrolyte brings additional benefits for drug delivery, such as penetration enhancement and pH-specific disintegration. Certain pH can facilitate the polymer distingration, therefore triger the targeted drug release [108]. Both natural

21 and synthetic polyelectrolytes can be used for drug delivery. The advantages of natural polymers are biocompatible, biodegradable, low-cost, and prevalently existing in nature, while synthetic polymers are more thermally stable with better repeatability [93].

Figure 1.13: The various uptake pathways of macromolecules and nanoparticles into cells accord- ing to size [8].

For polyelectrolyte hydrogel, covalent linkage promotes absorption of water and/or bioactive com- pounds without dissolution and allows drug release by diffusion, while ionically crosslinked hy- drogels are generally considered as biocompatible and well-tolerated, also exhibit higher swelling sensitivity to pH changes in comparison to crosslinked hydrogels by covalent bonding [84]. The swelling-deswelling property of PEC network plays an important role in drug releasing process. When PECs swell, more charges are available to interact with water and the drug can diffuse out from the vector matrix. Drug release may be triggered by pH-sensitive shrinking, as well. The

22 releasing behaviors of dextran sulfate-chitosan complex gel were studied previously [109]. It was shown that the release of dextran was enhanced by shrinking of the complex gel, and the complex gel released dextran more rapidly at a greater degree of shrinking.

Sometimes the polyelectrolytes for drug delivery are modified to achieve an optimized result, such as enhancing stability and loading efficiency, controlled release, and targeting. For example, 1,1,3,3-tetramethoxypropane (TMP) and PEG were introduced to enhance the hydrophilicity of chitosan hydrogel network [110, 111], and therefore the penetration of the material into the tissues can be improved. A core-shell nanofiber structure was fabricated from chitosan and hyaluronic acid as the vector of anionic drug combretastatin A-4 phosphate [67]. Drug release profile showed that these core-shell nano- fibers have a continuous drug release profile and the burst release was reduced when compared with HA/CS hybrid nanofiber, because the CS shell layer hindered the release of CA4P. Histamine was conjugated to poly(L-glutamate) (PLG) to a form a copolymer, which exhibited a response to environmental pH from 5 to 7 on complex formation [112]. This can be used for controlled release because tumor tissues usually present lower pH than normal tissue. Folic acid (FA) is a popular targeting moiety for drug delivery because tumor cells overexpress folic acid receptor. Therefore, FA is often conjugated to polyelectrolyte building blocks through EDC/NHS coupling in the drug delivery system to improve the targeting property and reduce the influence on normal tissue [113].

Gene delivery

Gene delivery is a process to transport nucleic acids into cell by a delivery agent for therapeutic ef- fect. Polyplexes comprising DNA and a delivery vector must overcome a broad array of obstacles and barriers both outside andinside the cell, which have been described in detail, to yield efficient cellular transfection (Figure 1.14) [9]. To get an ideal gene delivery vehicle, polycations are em-

23 ployed as the building block to balance hydrophobicity and charge density for non-viral vectors because they pose less safety risk and offer a reliably tune macromolecular structure [114, 115]. Liu et al [116] reported that using a water soluble chitosan with molecular weight of 5000 and deacetylated degree of 99% to interact with DNA could get rid of the effect of hydrophobicity of N-acetyl group. Their study revealed that the binding attraction of chitosan with DNA is affected by pH of the media. Highly charged chitosan and strong binding affinity with DNA occurs at pH 5.5, whereas weak interaction presents at pH 12.0.

Figure 1.14: The intracellular journey of gene delivery vectors [9].

However, simply conjugation of polyelectrolyte and DNA showed rapid removal from the blood- stream combined with nonspecific accumulation in organs and tissues [117]. This challenge can be addressed by crosslinking and surface modification to make sure that the vector is stable enough for the several barriers during transportation. For example, poly-L-lysine serves as the delivery vector

24 for DNA by multicovalent crosslinking of primary amines with a crosslinking agent that can be cleaved by reduction later [118]. It was found that the concentration of crosslinked polyplexes was maintained more than 7 times higher than that of non-crosslinked vector. As surface-modification, Oupicky [117] reported that after multivalent coated with N-(2-hydroxypropyl)methacrylamide (PHPMA), the α-half-life for bloodstream clearance of polycation/DNA polyplexes extended from less than 5 minutes to more than 90 minutes.

Wound self-healing

In general, polymers are promising materials for self-healing due to their mobility of the chains at physiological conditions [119]. PECs exhibits reversible properties due to the reversible interaction between the oppositely charged polymers. PEC multilayers and hydrogels are stimuli-responsive to external environment when the surrounding conditions change, such as pH, humidity, temperature, and conductivity. Therefore, they can prove active feedback for regeneration of wounded tissue. Furthermore, their ability to react with a lot of organic/inorganic molecules through electrostatic absorptions (complexation) and hydrogen bonding allows the loading of healing agents [120]. The mechanism of PEC multilayers or hydrogels working as self-healing materials is that the external stimuli affect ionization of the function groups of the polyelectrolytes, resulting in lower electro- static interaction and increasing repulsion between the polymers, whereas counter ions penetrate the matrix to compensate the charges [121–123]. This leads to higher osmotic pressure inside com- pared to the surrounding solution, which causes the penetration of water into the PEC multilayers. Thus, the material starts to swell, and self-healing (Figure 1.15).

It was proved that the weak-weak PECs are better material for self-healing than strong-strong PECs. For example, the PEI/PAA multilayer system self-healed in 6-8 h, however, polyelectrolyte multilayers formed by PDADMAC and PSS did not exhibit healing properties [124]. This is be-

25 cause the high charge density on strong polyelectrolyte renders too strong interaction to be altered by external stimuli. This problem can be overcome by modification of the polymers. For example, 2-hydroxypropyltrimethyl ammonium chloride chitosan (HACC), instead of CS, was used to form a self-healing hydrogel with alginate (ALG) [125]. The as-prepared hydrogel presents reversible PEC properties. PAA/PAH system is another type of popular PEC multilayer assembly. This type of film can be modified by post-processing in acidic solution to form nanoporous or submicrop- orous structures [126]. This film was used for wound healing in cornea [127]. Porous surfaces with 100 nm (nanoporous) and 600 nm (submicron) pore diameters both supported corneal cell ab- sorption, however, the corneal epithelial cellular response was significantly enhanced by nanoscale porosity, and thenanoporous topographies significantly improved the cell proliferation and migra- tion speeds. The PEC hydrogel formed by polyelectrolyte and metal ions also can be reversible and applied for self-healing. Wang et al [128] studied Zinc-induced polyelectrolyte hydrogel (a poly(acrylic acid) backbone functionalized with 30% catechol appendants) for self-healing. Zn2+ was used because it not only results in the injectable adhesive with superb adhesion after the coac- ervation formation compared to the one chelated by a stronger metallic interaction (e.g. Fe3+), but also generates improved mechanical performance as a hydrogel for self-healing application after the oxidation of catechol groups with a pH trigger.

Figure 1.15: Schematic mechanism of PEC multilayers for self-healing.

26 Actuator

Actuators are materials that show large deformation in the presence of external stimuli, such as pH [129, 130], ionic strength [129, 131], temperature [130], electric fields [81], and solvent [82] (Figure 1.16). Polyelectrolyte hydrogels, which are formed by the ionic bonding, produce me- chanical motion in response to those stimuli, making them good candidates as soft actuators. Liu et al [129] reported a versatile strategy to fabricate a bilayer hydrogel actuator by assembling two polymers with opposite charges through electrostatic attraction. The hydrogel was made from the copolymerization of acrylamide and 2-acrylamido-2-methylpropanesulfonic acid, followed by crosslinking with dimethylaminoethyl methacrylate methylchloride. The as-prepared hydrogel ac- tuator was responsive to ionic strength and pH. The direct and degree of bending of PEC hydrogel actuator can be tuned by adjusting the composition of hydrogel [130]. For example, pNIPAmAAc (15%)pDADMAC hydrogel shows bidirectional bending, while the hydrogel without pDADMAC can only bend on direction, when the solution pH or temperature is changed. This is because the introducing of pDADMAC enhances the osmotic pressure. The form of deflection rate and scale of PEC hydrogel actuators can be improved by increasing the surface to volume ration. Recently, a chitosan hydrogel with rectilinear cavities was fabricated by 3D printing[81]. Comparing with solid link hydrogels, this 3D printed hydrogel shows faster bending rate and larger bending angle, due to the higher surface area.

27 Figure 1.16: Schematic mechanism of PEC hydrogel actuator.

Energy storage

Polyelectrolytes usually act as the supportive materials on the substrates, such as carbon nan- otube and graphene, for the active materials, such as metal oxide nanoparticles, through self- assembly [132–134]. The introducing of polyelectrolytes into the system takes advantage of the interface engineering of multilayer thin film structures, leading to significantly improved elec- tronic conductivity and high electrochemical performance of energy storage devices [132]. Leem et al. [135] firstly reported that Layer-by-layer polyelectrolyte self-assembly films are applied to con- struct chromophore-catalyst assemblies for light-driven water splitting to produce hydrogen and oxygen. They took advantage of PAA as the oppositely charged layer of the cationic polystyrene- based Ru polychromophore to build up multilayer as the cell electrode. Pappa et al. [136] combined the advantages of well-established polyelectrolyte multilayers with a high-performance electronic transducer. They took poly(3,4-ethylenedioxythiophene) polystyrenesulfonate (PEDOT:PSS) film as the electrically active layer to study the impact of biofunctionalization on organic electrochemi- cal transistor operation. The system can be applied to the design of bioelectronic platforms for drug delivery, tissue engineering, and medical diagnostics. Furthermore, polyelectrolytes can be used in

28 self-healable and stretchable energy storage devices [137], such as electronic skin [138, 139] and smart storage clothes [140, 141]. Huang et al. [137] reported a self-healable and highly stretchable supercapacitor based on the dual crosslinked of PAA by hydrogen bonding and vinyl hybrid silica nanoparticles (VSNPs-PAA), which can be easily self-repaired at room temperature and can be stretched over 600% without any crack. Meanwhile, the scan rate applied to this supercapacitance electrode can be reached up to 100mV/s, which is much higher than the rates of most PPy-based electrodes measured even those in aqueous electrolytes [142, 143]. Polyelectrolytes also have al- ready been used for hydrogen storage by enhancing the thermal stability of sodium borohydrides (SBH) [144], which is the most attractive hydrogen storage material. The interaction between sodium borohydrides can be either electrostatic adsorption or chemical bond, preventing SBH from high reactivity with moisture and difficult handling at ambient atmosphere, consequently reduce hydrogen penetration.

Theme of the research

This research is focused on the fundamental and application of PAA and PAA/CS based PECs.

In Chapter 2, we investigated how ferric ions (Fe3+) affect the bending stiffness and size of indi- vidual electrospun PAA:CS nanofibers measured by confocal microscopy imaging analysis. The results indicate the effect of ferric ions on the length and strength of the fibers. Understanding the metal ion effects on individual nanofiber mechanics could optimize a method of fine-tuning for de- sired stiffness, length, and Young’s modulus of nanofibers, offering further control over hydrogel mechanics.

Chapter 3 focuses on the application of MnO2@PAA/PPy core-shell nanofibers as supercapacitor electrode, which exhibited outstanding properties including high specific capacitance, excellent

29 reversible redox reactions, and fast charge/discharge ability.

In Chapter 4, PAA-CS/TPP core-shell nanoparticles were fabricated to load gemcitabine as drug delivery system for pancreatic ductal adenocarcinoma (PDAC). The nanoparticles encapsulate the drug gemcitabine at high efficiency and have controlled-release of the drug.

30 CHAPTER 2: TUNABLE EFFECT OF METAL IONS ON

POLYELECTROLYTE NANOFIBER MECHANICAL PERFORMANCE

Abstract

Polyelectrolyte-based hydrogel fibers can mimic extracellular network and have various biomedi- cal applications such as drug delivery, wound dressing and tissue scaffolding. Metal ions play an essential role in hydrogel fiber stability through electrostatic interactions. However, the knowledge of how ions modulate the mechanical performance of individual polyelectrolyte nanofiber is still lacking. In this study, individual electrospun nanofiber of polyacrylic acid (PAA) with chitosan is used as a model system to evaluate the effect of ferric ions on the nanofiber mechanical prop- erties. Using dark field microscopy imaging and persistence length analysis, we demonstrate that ferric ions affect nanofiber characteristics optimally at lower concentrations (0.1%), allowing for shortest and stiffest nanofibers in the ferric ion concentration range of 0 ∼ 2% (molar percentage to PAA monomer). Young’s modulus of the nanofibers is estimated at values of a few kilopascals, indicating comparable strength for medical applications. Furthermore, Fourier-transform infrared (FTIR) spectra show the chelation type of ferric ions with carboxylate group on polyacrylic acid polymer chain. Our results suggest that metal ions can regulate single polyelectrolyte fiber stiffness by bidentate bridging, thereby providing designs to fabricate hydrogels in a tunable fashion.

Introduction

Polyelectrolyte nanofiber hydrogels demonstrate exceptional mechanical properties and structural stability that can be utilized in various biomedical applications including drug delivery, tissue scaffolding and wound healing [145–147]. Such nanofibers generate crosslinked networks through

31 thermal or chemical interactions for mechanical support in differing environments (e.g., pH, tem- perature, ionic strength) [6, 148–150]. Metal ions provide nanofiber networks with structural sta- bility through electrostatic attractions, allowing for tunability of nanofiber hydrogels’ mechanical properties [6, 148–150].

Electrospun nanofibers of polyacrylic acid (PAA) and chitosan (CS) with metal ions use ionic and covalent interactions for tunability of the hydrogel (Figure 2.1A). Polyacrylic acid (PAA) is an anionic polyelectrolyte with carboxylate functional groups on top of a hydrocarbon backbone [151, 152]. The flexible cationic polysaccharide chitosan (CS) interacts with PAA to form hydrogels through electrostatic attraction for improved stability in varying pH ranges [153]. Mechanical properties such as tensile strength of these hydrogels as a bulk material have been investigated [154, 155], but fabrication processes have been somewhat complicated. Hydrogel nanofibers of PAA and CS with metal ions can be fabricated by electrospinning (Figure 2.1A), which is quick and streamlined. During electrospinning process, the length, amount, and mechanics can be tuned via adding metal ions to the nanofiber precursors. The mechanical properties (e.g., tensile strength and Young’s modulus) of hydrogels have been characterized utilizing various techniques including atomic force microscopy [156, 157], scanning electron microscopy [158], and rheometers [158, 159]. However, to our best knowledge, mechanical properties of PAA/CS single nanofiber have not been well established. Control of individual nanofibers before introduction into a hydrogel system optimizes them for biomedical applications requiring opposing parameters such as high elasticity (artificial muscles) or higher stiffness (cell scaffold) [160–162] while offering regulation past hydrogel mechanics.

Although the effects of metal ions on hydrogel stability have been shown [148, 163], how metal ions modulate the mechanical properties of individual polyelectrolyte nanofibers is still unclear. In this study, we investigated how ferric ions (Fe3+) affect the bending stiffness and size of individ- ual electrospun PAA:CS nanofibers measured by confocal microscopy imaging analysis. Further

32 characterization was carried out through Fourier transform infrared spectroscopy (FTIR), scan- ning electron microscopy (SEM), and transmission electron microscopy (TEM). Understanding the metal ion effects on individual nanofiber mechanics could optimize a method of fine-tuning for desired stiffness, length, and Young’s modulus (E) of nanofibers, offering further control over hydrogel mechanics.

33 Figure 2.1: (A) Schematic diagrams of polyacrylic acid (PAA) nanofiber fabrication process by electrospinning. Functional groups with negative charges on PAA and ferric ions associate through ionic interactions. Chitosan (CS) and ferric metal ions (Fe3+) provide positive charges to create crosslinks in the system, therefore making it more resistant to environmental changes. (B) Scan- ning electron microscopy image of a polyacrylic acid nanofiber network. Scale bar represents 500 nm. (C) Transmission electron microscopy image of a single PAA/CS nanofiber complex. Scale bar represents 200 nm.

34 Materials and Methods

Nanofiber synthesis by elecrospinning

0.1 M FeCl3 solution and 6 M HCl was added to 0.568 g 4.4% CS solution in 25% acetic acid and stirred for 30 min to get a homogeneous solution. Then, 1.43 g 35% PAA (Mw ∼240,000, Sigma- Aldrich) was added to the above solution to make PAA:CS mass ratio 20:1. Different amount of

3+ 0.1 M FeCl3 was added to make of Fe 0.04%, 0.07%, 0.1%, 0.25%, 0.5%, 1% and 2% by molar percentage of the carboxylate group on PAA. 6 M HCl was added to keep the pH of the solution at 1 to avoid gelation between Fe3+ and PAA. Different amount of DI water was added to the solution to make the total polymer concentration at 20% wt.

The above solution was then loaded into a plastic syringe equipped with a 16 mm gauge needle made of stainless steel. The needle was connected to a high-voltage power supply. The solution was pumped continuously supplied by a syringe pump. A piece of aluminum foil as the collector was placed at 27 cm distance of the tip of needle. The electrospinning process was conducted in air at room temperature. The actual voltages ranged between 10 and 12 kV. To obtain continu- ous and homogeneous nanofiber, the voltages applied were adjusted during the process whenever necessary. The electrospinning process usually took 1∼2 minutes to obtain individual filaments without overlap and circle. The asspun nanofibers were dried at 40◦C under vacuum and stored in a desiccator for later use.

Scanning electron microscopy and transmission electron microscopy

The morphology of the fibers was characterized by transmission electron microscopy (TEM, JEOL 1011) and scanning electron microscopy (SEM, ZEISS ULTRA 55). A Formvar/Carbon 400 mesh

35 Copper grid (Ted Pella 01754-F) was put on the collector during electrospinning to collect fibers for TEM imaging at 100kV. SEM images were obtained from nanofibers electrospun on glass slides. The samples were coated with a thin layer of Au/Pb (Emitech k550). Working distance for SEM was between 6 ∼ 10 mm and required 5 kV for imaging. The diameter of the fibers was measured through the scale bar on the images.

Fourier Transform Infrared Spectroscopy (FTIR)

The infrared spectra of the nanofibers were characterized by Fourier Transform Infrared Spec- troscopy using a Perkin Elmer Spectrum 100 series FT-IR spectrometer in the range of 600 to 4000 cm−1 with a resolution of 4 cm−1 and scan number of 4 to simulate the fibers. Single nanofibers were too small for detection, and therefore vacuum-dried membranes of sample solution were used. None of the components were lost during the electrospinning process (simulated using vac- uum drying on the solution).

Confocal microscopy imaging and bending mechanics data analysis

Electrospun nanofibers were visualized with an Olympus BX51M confocal dark field microscope equipped with a digital camera (Progres Gryphax model from Jenoptik) with a 20× objective. Dark field microscope images were processed for enhanced contrast and skeletonized using ImageJ soft- ware (NIH) to enhance the prominence of filament shapes to be efficiently tracked [164]. Contrast enhancement improves the signal-to-noise ratio and normalizes the signal for processing and vi- sualization. Skeletonization converts the image to a binary format and pares down the filament shapes to a single pixel width to remove the noise and enhance the tracking efficiency. Persistence software [164] was used to measure the average length (Lavg) and bending persistence length (Lp) of PAA nanofibers. Lp values were calculated from best fits to angular correlation analyses [164]

36 of nanofibers (n =210 ∼ 2200) using OriginPro 8 software. Young’s modulus (E) was calculated by equation 2.1:

κ L κ T E = = p B (2.1) I I

where κ and I denote flexural rigidity and moment of inertia respectively, Lp represents persistence length and κBT is thermal energy (κB is the Boltzmann constant, and T is temperature). The thermal energy is essential at fluctuated temperature because it can affect the bending of a filament. For a rigid rod (stiff nanofiber), changes in temperature will affect it less than a flexible nanofiber, which is more likely to respond to temperature with bending. But since we kept everything at room temperature, our measurements do not have to take this factor into account, and we therefore used the constant. Moment of inertia (I) of PAA nanofibers was calculated based on radius (r) measurement from SEM and TEM images (Figure 2.1B and C) through equation 2.2 [165–167].

π I = r4 (2.2) 4

Results and Discussion

Ferric ions modulate the mechanical properties and average length of polyacrylic acid nanofibers.

Since the software may give erroneous results at filament reconstruction if selected filaments are too close together [164], the electrospinning time can not be too long to make the filaments indi- vidually separated. Besides, circles are not allowed for the software, so the length of the filaments has to be controlled by the voltage. To determine the bending stiffness of nanofibers, we directly visualize the electrospun PAA:CS nanofibers at varying ferric ion concentrations using confocal

37 (dark field) microscopy (Figure 2.2). The bending persistence length (Lp) is a physical quantity that relates to the bending stiffness of a polymer [164, 165], which is the fundamental parameter of the modeling. Assuming the fiber as a bending cylinder, over a length scale Lp, the persis- tence length, the cylinder can be approximated as a rigid rod, while over larger length scales it bends [164, 168]. Based on nanofiber microscopic images, we determine the nanofiber bending

Lp from two-dimensional angular correlation analysis that has successfully captured semi-flexible biopolymer mechanics by equation 2.3 [164, 165].

C(s) = cos[θ(s) − θ(0)] = e−s/2Lp (2.3) where C(s) is the two-dimensional cosine correlation of length s, θ is the tangent angle along a segment of length s (Figure 2.3). The stiffer the fiber, the less difference of θ(s) and θ(0), the closer the correlation C to a value of 1, and the longer of Lp.

38 Figure 2.2: Formation of PAA:CS:Fe3+ nanofibers at different concentrations after contrast en- hancement. Representative darkfield images of nanofibers at different concentrations of ferric metal ions show qualitative changes in length and stiffness. Scale bar represents 300 µm.

39 Figure 2.3: Illustration of θ(s) and θ(0).

As shown in Figure 2.4A, the curve for Fe concentration 0.1% is closest to 1, suggesting that fibers at this ferric concentration are the stiffest. This is content to the results shown in Figure 2.4B, ferric ions modulate the bending stiffness and average lengths of PAA:CS nanofibers and allow for optimization of stiffness at lower concentrations (0.1%). Between 0% to 0.04% of ferric ion concentrations, the bending Lp (and therefore stiffness of fibers) rises from an average of 67 µm to 81 µm, only to come back down at the next concentration (0.07%) to levels similar to those of nanofibers without ferric ions(67 µm). Though this suggests a threshold which must be overcome at low concentrations of ferric ions, it also implies potential competition between ferric metal ions with pre-existing crosslinking of polyacrylic acid and chitosan before a final settling into optimal conformations. At 0.10% concentration, a rise to the highest average stiffness of about 122 µm is indicated. This could be the optimal condition at which all steric and ionic effects are balanced and take advantage of crosslinking between PAA and CS, while still having the addition of ferric metal ions in a way that does not compete with existing charges adversely. Persistence length decreases at

0.25% ferric to the lowest Lp (about 62 µm), but then rises (though not to the extent of the plateau pre-0.10% concentration) at 0.25% to reach a threshold that holds throughout the rest of rising amounts of ferric studied at a range of 79 to 87 µm, indicating that a dynamic equilibrium has been obtained. Higher concentrations could create too much of a struggle between rising number

40 of ferric ions vying to better interact with the PAA:CS complex in nanofibers and therefore lead to the gelation of the system, as was seen during creation of the solution.

Figure 2.4: (A) Changes in ferric concentration affect angular correlation of nanofibers. The stiffer the fiber, the closer the correlation is to a value of 1. (B) Ferric ion modulation of bending stiff- ness in PAA:CS nanofibers is determined by persistence length analysis revealing that interaction between ferric and PAA:CS raises stiffness of the nanofibers in a non-linear manner. (C) Average length distribution of all nanofiber samples. The shortest nanofibers at the lowest possible concen- tration are created at 0.10% ferric metal ions, with ion effects also revealing a non-linear pattern of effect. (N=630-2200).

The interaction between the positive charges of ferric and PAA:CS complex is strong enough even at a low concentration to affect nanofiber length (Figure 2.4C). For instance, at 95 µm nanofibers with no ferric have about 55% decrease in length when compared to nanofibers with 0.04% fer- ric. Previous studies show that ferric could successfully bind with both chitosan and polyacrylic acid through electron-rich ligands on the polymer chain (carboxylate group, amine group, and hy- droxyl group) [10, 150, 163, 169, 170](Figure 2.5), consistent with the ferric ion’s effect on PAA

41 nanofibers in this study. Nonetheless, the solution creation at a lower pH affects chitosan through protonation that keeps interaction between it and the other components to a negligible level, mak- ing most of the effect a focus between PAA and ferric metal ions [6, 169]. Between 0.04% and 0.10%, change is not drastic but reaches shortest average length before rising at the following con- centration of 0.25% to a length that still does not compare to that of the PAA/CS complex without ferric. A further shortening of nanofibers from 0.25% concentration of ferric until 1.00% in direct correlation with rising concentration of ferric then follows. From 1.00% to 2.00%, a plateau is evident, though length values are in range of 0.10% ferric concentration samples. This suggests that at their shortest, PAA/CS/Fe3+ nanofibers reach a threshold range through either a controlled addition of ferric metal ions (0.10%) or through further saturation.

Figure 2.5: Schematic representation of the Fe-PAA (a) and Fe-CS (b) [10] complex.

42 Chitosan helps ferric ions modulate the mechanical properties of PAA nanofibers.

Figure 2.6: Comparison analysis of length (A) and bending stiffness(C) of nanofibers electrospun with and without chitosan at initial concentrations 0%, 0.04%, 0.07%, and 0.10% of ferric metal ions. (A) Nanofibers with no chitosan (circles) show a higher average nanofiber length that falls dramatically as the concentration of ferric rises. (B) A point for control nanofibers containing no CS or Fe3+ ions could not be accurately measured due to length surpassing imaging limits of magnification. Scale bar represents 250 µm. (C) Nanofibers with chitosan (triangles) did have measurable lengths that began well below estimations for PAA only nanofibers, and therefore showed more overall stability in length changes. Nanofibers with no chitosan (circles) at 0%, 0.04%, and 0.07% concentrations of ferric have lower persistence length and therefore stiffless than the average nanofiber with chitosan (triangles). At concentration 0.07% and 0.10% of ferric, both nanofibers with and without chitosan converge at similar persistence length averages, as well as highest impact concentration of ferric on overall stiffness of the nanofibers. Bending stiffness measurements were acquired on fragments of nanofibers that could be captured in span of image.

43 Nanofiber fabrication at acidic pH leads to negligible electrostatic interaction between chitosan with PAA or ferric metal ions. Chitosan does not fully block ferric interaction with PAA but could rather help reinforce nanofibers physically. To identify the roles of chitosan in PAA nanofiber mechanics, we imaged PAA nanofibers in the presence or absence of chitosan and measured nanofibers’ lengths and bending persistence lengths. The average nanofiber length without chitosan was longer than those with chitosan (Figure 2.6A and 2.6B). Since a point for control nanofibers containing no CS or Fe3+ ions could not be accurately measured due to length surpassing imaging limits of magnification (Figure 2.6B)), the study of nanofibers without CS starts at ferric concen- tration 0.04%. Nanofibers with no chitosan (circles) show a higher average nanofiber length that falls dramatically as the concentration of ferric rises, while the average length of nanofibers with ∼5% CS shows a slightly shortening (Figure 2.6A). According to length distribution comparison, nanofibers with no chitosan (black) show a wider variation of lengths than nanofibers with chitosan (Figure 2.7A).

The chitosan interaction gives the complex a better framework by immobilizing PAA polymer chains through minimal interaction with the PAA chains and ferric metal ions. Nanofibers with chitosan (triangles) did have measurable lengths that began well below estimations for PAA only nanofibers, and therefore showed more overall stability in length changes (Figure 2.6B). Nanofibers with no chitosan at 0% and 0.04% concentrations of ferric have lower persistence length and there- fore stiffless than the average nanofiber with chitosan(Figure 2.6C). At concentration 0.07% and 0.10% of ferric, both nanofibers with and without chitosan converge at similar persistence length averages, as well as highest impact concentration of ferric on overall stiffness of the nanofibers. This boost could be due to the bidentate bridging created by the ferric ions with multiple chains to make up for the lack of chitosan in the complex, as well as lack of competition for interactions with PAA [150, 163, 169].

44 Electrospun PAA:CS nanofibers could exist in hydrated states.

While various studies done for the mechanics of hydrogels created using different methods yielded a wide range of Young’s modulus (6 ∼ 100,000 kPa) [171–173], the knowledge of Young’s mod- ulus for individual PAA nanofibers is lacking. We estimated the Young’s modulus (see Methods for details) of the nanofibers with and without CS based on average persistence length analysis (Figure 2.8A) and diameter measured from SEM images (Figure 2.8B). The diameter of the fibers increased along with time, and the newer samples have closer diameter points in cluster. The nanofibers with the highest strength are at a ferric concentration of 0.10% for both nanofiber samples with CS and controls without CS (Figure 2.8B). They have similar modulus results with 0.10% (with CS) calculated at 2.41 kPa. Without CS, the strength changes to 2.43 kPa as well as similar bottommost diameter measurements for new samples. At lowest ferric value (0.04% with no CS) nanofiber E is estimated at 0.599 kPa, which could be due to lack of optimization in complex formation. With not enough ferric to interact with the PAA:CS complex, the structure is not as scaffolded as a nanofiber with more ferric to create more bonds for stability. Our estimated Young’s modulus of PAA nanofibers has a comparable stiffness to that of PAA hydrogel created through micellar copolymerization as well as commercially available basement mimicry ECM ma- terial [171, 174], suggesting that electrospun nanofibers may be in their hydrated states lead to functional hydrogel strength.

45 Figure 2.7: Comparison analysis of nanofiber length (A) and bending stiffness(B) of nanofibers electrospun with and without chitosan at concentrations 0%, 0.04%, 0.07%, and 0.10% of ferric metal ions in box plots. (A) Length distribution comparison. The box range represents quartiles of 10% and 90% and reveals the effect of ferric metal ions and chitosan on each set. Nanofibers with no chitosan (black) show a wider variation of lengths than nanofibers with chitosan. A box was not created for nanofibers with no chitosan due to their lengths surpassing imaging limitations. (B) Nanofibers with no chitosan (black) mostly have lower persistence length and therefore stiffless than the average nanofiber with chitosan (red). At concentration 0.10% of ferric, both nanofibers with and without chitosan converge at similar persistence length averages, as well as highest im- pact concentration of ferric on overall stiffness of the nanofibers. Bending stiffness of nanofibers without chitosan or ferric were obtained on filament fragments that fit within the imaging window.

46 Figure 2.8: Mean diameter is represented by the yellow columns, and young’s modulus is repre- sented by the blue columns in the kPa range for individual nanofibers in their hydrated state and show optimal estimates at 0.10% ferric metal ion concentration.

Characterization sheds light on how ferric metal ions affect individual PAA:CS nanofiber structure.

Metal ion interaction at varying concentrations influences bonds of polyacrylic acid in individual nanofibers. According to FTIR spectra, free carboxyl groups of PAA have a sharp band at 1702 cm−1, indicating a strong C=O stretching vibration and the low concentration of Fe3+ to attach to these groups (Figure 2.9A). The C=O stretching vibration for monomer carboxylic acid is at 1770 ∼ 1750 cm−1, while for PAA, this peak moves to lower wave number because of hydrogen bonding among the carbonyl groups of the polymer [175–177]. This peak further shifts to lower wave number a little bit, due to the reconfiguration of the polymer chain after CS and Fe3+ was added [175, 176].

47 Figure 2.9: FTIR and conductivity analysis of nanofibers with varying ferric concentrations. (A) The PAA:CS solutions with differing metal ion amounts were dried in vacuum for 3 days at 60◦C. Ferric metal ions influence the vibration of bonds in the system, which is reflected in the shifting and strengthening of certain peaks. (B) Measurement of conductivity difference created by Fe3+ and HCl used in nanofiber preparation. Polyacrylic acid and chitosan have low conductivity that rises sharply with addition of ferric metal ions and hydrochloric acid. There was no real conclusive pattern to conductivity for further differentiation.

Interestingly, a peak at 1591 cm−1 only appeared for samples with a ferric concentration at 1% showing carboxyl C=O asymmetric stretching verified through AFM. When the ferric ion concen- tration is lower than that, this peak couldn’t be detected due to the small amount of interaction between PAA and Fe3+. Since symmetric stretching of the carboxyl groups on PAA are at 1412 cm−1, the difference between this peak and the asymmetric one in 1% ferric metal ion samples is

48 179 cm−1, which is approximately equal to the difference of the peaks for sodium salt of PAA (154 cm−1) [178], indicating that the type of bonds between PAA and Fe3+ are due to bidentate bridging in that concentration of ferric specifically [6, 172]. The 1% concentration also displayed a sharper, slightly larger peak of 1412 cm−1 owing to more symmetric carboxyl group vibrations than in the other samples. At this concentration in both length and bending stiffness, it can be noted that the nanofibers reach a sort of saturation point in ferric concentration. The growth of these peaks could reflect the steric effect of chitosan sterically affecting sites of interaction as well as repellant elec- trostatic forces both between CS and Fe3+ affecting PAA functional groups. It could also be due to metal ions competing amongst themselves as they saturate the system. Preliminary data from AFM has indicated the creation of ferric clusters in the nanofiber complex which could also be part of the effect on the bonds seen.

The rise of a peak at 1452 cm−1 represents C-H deformation with rising Fe3+ concentrations and therefore rearrangement of the system. Further C=O stretching coupled with O-H in-plane bending is reflected in the peak at 1233 cm−1 but does not see major change across concentrations. Nonetheless, this peak does suggest that PAA is a polymer with syndiotactic configuration in which repeating units alternate stereochemical configuration in a regulated manner. Had it been atactic (lacking regular stereochemical configurations), there would have been a peak at 1250 cm−1, as well as a shoulder at 1300 cm−1 [179]. The peak at 1169 cm−1, according to Liew et al. indicates C-O stretching of carboxyl groups [180]. It remains clearly defined across all samples, though it does weaken slightly with the rising concentration of ferric.

Previous work states that the chitosan interaction with the ferric metal ions and PAA is minimal due to the pH of the solution being low. This leads to protonation of CS chains and does not allow for further interaction with the system [181]. Protonation effect is verified by the PAA-CS- Fe3+ FTIR data, which is similar to past characterization of PAA-Fe3+ systems with no chitosan characteristic peaks at 1154 or 893 cm−1 present [182]. Comparison of PAA:CS spectra with that

49 of PAA alone also shows the lack of interaction since they are mostly identical. Direct verification of Fe-O interaction bands would have to be on FTIR with a lower range, seeing as they appear below 500 cm−1 and therefore outside the scope of our instrument [169].

Finally, there are peaks at about 800 cm−1 indicative of C-H twisting on PAA as well as C-COOH stretching [179]. Though clear across all samples, the effect is strongest at 1% ferric concentration, further confirming the higher reconfiguration effect occurring within the system when compared with other concentrations.

According to the FTIR rspectrum, both free and crosslinked carboxylate groups showed up. This reveals that only a fraction of PAA molecules has interacted with ferric ions and CS. The rest are uninteracted polymer chains as loops and tails. The fraction of crosslinked PAA is very little due to the low concentration of ferric ions and CS.

The conductivity of the nanofiber system was also studied. PAA and CS individually show a very low level of conductivity (Figure 2.9B). Addition of HCl raised the measurement to 83.2 mS/cm, but the value barely changed with the addition of 0.04% ferric metal ions. Ferric at a concentration of 0.1M showed the highest conductivity by itself with values of the mixed nanofiber solutions never rising as high. Nonetheless, the solution with the highest conductivity of the mixed samples was at 0.10% and could explain how the nanofiber diameters have been smaller than in other concentrations. When the same concentration of ferric was used without chitosan, the conductivity dropped slightly. The final concentration tested (1%) did not show significant change from the conductivity of 0.10% ferric metal ions. Overall no real conductivity differences could be used definitively to differentiate one concentration from another, but the process did show the difference that the HCl and ferric metal ions could make in nanofiber conductivity, and therefore diameter.

50 Conclusion

The effect of metal ion (Fe3+) on mechanical properties of PAA/CS complex nanofiber was demon- strated by Persistence software via two-dimensional angular correlation analysis. Various ferric ion concentrations were studied. The result indicates that the stiffest fiber presents at ferric con- centration 0.1% molar ratio to PAA monomer with relatively shortest average length and longest persistence length (Lp), therefore the highest young’s modulus. Interestingly, the Lp and young’s modulus are pretty similar for PAA-Fe3+ complex with or without CS. This study firstly provides a facile method to study the mechanical properties of single fiber at solid status without solvent and can be further applied to other fibrous systems. The results offer the opportunity to choose fibers with suitable mechanical properties for various applications, such as tissue engineering and wearable device.

51 CHAPTER 3: FLEXIBLE CORE-SHELL COMPOSITE FIBERS FOR

HIGH PERFORMANCE SUPERCAPACITOR ELECTRODES

Xiaoyan Lu, Chen Shen, Zeyang Zhang, Elizabeth Barrios, and Lei Zhai, “CoreShell Compos- ite Fibers for High-Performance Flexible Supercapacitor Electrodes”, ACS Applied Materials & Interfaces 2018, 10(4), 4041-4049. DOI: 10.1021/acsami.7b12997 Copyright 2018 ACS Publica- tions

Abstract

Coreshell nanofibers containing poly(acrylic acid) (PAA) and manganese oxide nanoparticles as the core and polypyrrole (PPy) as the shell were fabricated through electrospinning the solution of PAA and manganese ions (PAA/Mn2+). The obtained nanofibers were stabilized by Fe3+ through

3+ the interaction between Fe ions and carboxylate groups. Subsequent oxidation of Mn by KMnO4 produced uniform manganese dioxide (MnO2) nanoparticles in the fibers. A PPy shell was created on the fibers by immersing the fibers in a pyrrole solution where the Fe3+ ions in the fiber polymer- ized the pyrrole on the fiber surfaces. In the MnO2@PAA/PPy coreshell composite fibers, MnO2 nanoparticles function as high-capacity materials, while the PPy shell prevents the loss of MnO2 during the charge/discharge process. Such a unique structure makes the composite fibers efficient electrode materials for supercapacitors. The gravimetric specific capacity of the MnO2@PAA/PPy coreshell composite fibers was 564 F/g based on cyclic voltammetry curves at 10 mV/s and 580

F/g based on galvanostatic charge/discharge studies at 5 A/g. The MnO2@PAA/PPy coreshell composite fibers also present stable cycling performance with 100 capacitance retention after 5000 cycles.

52 Introduction

Supercapacitors (SCs) are an emerging technology that fills a crucial gap in today’s rapidly evolv- ing energy storage requirements such as high energy density, high power density, lightweight and flexibility. With desirable properties of high power density, fast charge-discharge rate, and excel- lent cycle stability [183–187], SCs can be used in a wide range of applications such as hybrid vehicles, portable electronics, pacemakers, etc. [188, 188–190] SCs are categorized into electro- chemical double layer capacitors (EDLCs) and pseudo-capacitors depending on the charge storage mechanisms. Ion absorption of the electrical double layer at the electrode and electrolyte interface is the mechanism for charge storage in EDLCs which are usually built from carbon-based active materials with high surface area. They have good charge/discharge performance but low specific capacitance (10 ∼200 F/g) [184–186, 191–196]. Pseudo-capacitors store energy through fast and reversible redox reactions of transition metal oxides and electrical conductive polymers [197–207].

Manganese dioxide (MnO2) is a promising material for electrochemical supercapacitors because of its high specific capacitance (theoretical value of 1400 F/g), fast charge-discharge process, low

cost, and environmentally benign nature [187, 208, 209]. The MnO2 pseudocapacitive behavior is attributed to a Mn4+/Mn3+ redox system involving a single-electron transfer [184]. In aque- ous electrolytes, the general charge/discharge mechanism in MnOO−-based supercapacitors can

+ − + be described by: MnO2 + M + e → MnOOM where M represents hydrated protons (H3O )

+ + + and/or alkali cations such as K , Na , and Li [184]. Energy extracted from a MnO2 composite electrode strongly depends on crystal and morphological structure of MnO2 as well as the quality of MnO2/current collector and MnO2/electrolyte interfaces. Textural and morphological effects at interface determine the MnO2 capacitance while the material bulk structures affect electrolyte accessibility [199, 208, 210–212].

One-dimensional (1D) MnO2 composite fibers have been manufactured through electrospinning

53 and extensively investigated as supercapacitor electrode materials because they can provide large electrolyte/MnO2 interfacial surface area and good control of MnO2 morphology [202, 209, 213–

216]. In these studies, MnO2 nanoparticles were produced in carbon nanofiber matrices through an electronspin/treatment process. The electrospinning of a solution of carbon precursor (e.g., polyvinylpyrrolidone and polyacrylonitrile) and a Mn salt (or MnOx nanoparticles) produced 1D

fibers which were converted to carbon/MnO2 composite fibers by a controlled heat treatment.

Phase separation of the polymers produced porous structures to control MnO2 morphology and facilitate accessibility of electrolyte. While these 1D composite fibers have efficient use of MnO2 that leads to high specific capacitance, the charge/discharge mechanism of MnO2 indicated that water soluble MnOO− could diffuse away into solution in the process, imposing a serious stability issue that limits the practical applications for MnO2 [200, 217].

Here, we report the fabrication of freestanding MnO2@poly(acrylic acid) (PAA)/polypyrrole (PPy) coreshell nanofibers from electrospun PAA fibers loaded with manganese ions (PAA/Mn2+) (Fig- ure 3.1). This work is based on our recent discovery that well-defined nanoparticles can be pro- duced from metal ion loaded polyelectrolyte fibers [6]. PAA nanofibers with manganese ions, as the core of the composite fiber, provided a robust and porous nanostructure as a freestand-

3+ ing and flexible supporting scaffold for the production of MnO2 nanoparticles. Fe ions were used to cross-link PAA fibers to improve their stability and initiate the polymerization of pyr- role to generate a PPy shell on the PAA fibers. The PPy shell inhibited the loss of MnO2 dur- ing the charge/discharge process. As a result, the coreshell structured fibers exhibited outstand- ing properties including high specific capacitance, excellent reversible redox reactions, and fast charge/discharge ability, making them effective materials for supercapacitor electrodes.

54 Figure 3.1: The scheme of the fabrication of the MnO2@PAA/PPy core-shell nanofibers from PAA electrospun fibers loaded with Mn2+ ions.

Experimental Procedure

Materials

25% poly(acrylic acid) (PAA, MW = ∼240,000 g/mol) solution, was purchased from Sigma-

Aldrich Chemical Co. (USA). Manganese acetate (Mn(CH3COO)2), potassium permanganate

(KMnO4), ferric (III) chloride (FeCl3), and pyrrole were purchased from Fisher Scientific (USA). All chemicals were of analytical grade, and they were used as received. All solutions were prepared in deionized water. Samples were synthesized by the following procedures.

55 Materials synthesis

Fabrication of Nanofibers

Four grams of 25% PAA (Mw ∼240,000, Sigma-Aldrich) solution was mixed with 700 µL 1 M manganese acetate (Mn(CH3COO)2), which was the precursor solution. The solution was loaded into a plastic syringe equipped with a 16-gauge needle made of stainless steel. The needle was connected to a high-voltage power supply. The solution was pumped continuously with a syringe pump. A piece of aluminum foil as the collector was placed at a distance of 27 cm from the tip of the needle. The electrospinning process was conducted in air at room temperature. The actual voltages ranged between 10 and 12 kV. To obtain continuous nanofibers with consistent thickness, the voltages applied were adjusted during the process whenever necessary. For example, a lower voltage was applied when the concentration and viscosity of the drop at the needle tip increased during the electrospinning. The as-spun nanofibers were simply peered off from the collector as a freestanding fiber mat after electrospinning and dried at 40◦C under vacuum.

Creating MnO2 Nanoparticles on PAA Nanofibers

A solution contained 0.2 M FeCl3 and 0.6 M Mn(CH3COO)2) was dropped onto the as-spun

nanofibers to wet them fully, and nanofibers were then dried in air. FeCl3 served as the cross-linker

2+ of PAA to make fibers insoluble in water. Mn(CH3COO)2 provides an Mn environment to pre-

2+ vent the diffusion of Mn from the nanofibers. After that, 0.01 M KMnO4 solution was brushed

onto nanofibers by using a small brush with a tip size of 4.1 mm 2.3 mm, where KMnO4 reacted

with Mn(CH3COO)2 to form MnO2 nanoparticles on the nanofibers. As a result, MnO2@PAA nanofibers were formed. The fiber mats were brushed second, third, and fourth time with the 0.01

M KMnO4 solution after the KMnO4 solution from the previous brush was completely dried in air

56 to obtain different amounts of MnO2 nanoparticles on PAA nanofibers. The remaining FeCl3 and unreacted KMnO4 and Mn(CH3COO)2) on the dry nanofibers were washed away by de-ionized water.

Depositing the Polypyrrole Shell on the Surface of MnO2@PAA Nanofibers

The MnO2@PAA nanofiber mat was immersed into a mixed solution of 2 mL pyrrole and 10 mL ethanol for 1 min for polymerization. After being washed with ethanol and water, the polypyrrole coated nanofiber was dried at 40◦C under vacuum. The freestanding nanofiber mat was pasted onto a piece of graphite electrode (1 cm × 3 cm) by silver paint, which is the working electrode for the three-electrode electrochemical workstation. The loading mass was around 0.2 mg for each electrode.

Characterization

The morphologies were characterized by scanning electron microscopy (SEM, ZEISS ULTRA 55) and transmission electron microscopy (TEM, JEOL 1011). The size and distribution of MnO2 nanoparticles were determined by high resolution transmission electron microscopy (HRTEM, Tecnai F30) equipped with energy-dispersive X-ray spectroscopy (EDS). The chemical compo- sitions of nanofibers were analyzed by X-ray photoelectron spectroscopy (XPS, PHI 5400). X-ray 152 diffraction (XRD) spectra were obtained using a PANalytical Empyrean X-ray diffractome- ter. To compare the electrochemical performance of different electrode materials, a three-electrode system consisting of a working electrode, a platinum counter electrode, and a saturated calomel electrode (SCE) as reference electrode was used. The representative cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), and galvanostatic charge/discharge (GCD) were performed on this three-electrode configuration in 1 M Na2SO4 solution at room temperature. CV

57 was carried out at different scan rates of 2, 10, 20, 50, and 100 mV/s. GCD was measured at 0.5, 5, 10, 25, and 50 A/g. EIS measurements were performed in a frequency range from 1 Hz to 1000

2+ kHz. To measure the electrical conductivity, PAA/Mn , MnO2@PAA, and MnO2@PAA/PPy fiber mats were cut into 1 cm × 1 cm mats (the typical thickness of mats was 1.5 m.). Silver pastes were coated on the two opposite sides of the fiber mats to get sheet resistance. The electrical conductivity was calculated using eq 3.1

l σ = (3.1) RA where σ is the electrical conductivity, l is the length of the piece of the material, A is the cross- sectional area of the specimen, and R is the electrical resistance of the specimen.

The specific capacitances (Csp, F/g) are calculated from the CV study according to eq 3.2

R V2 I(V )dV V1 Csp = (3.2) υ(V2 − V1)m where I(V ) is the chargedischarge current of the CV loop, υ is the scan rate, m is the mass of the active materials (g), and V1,V2 are the switching potential in the CV curve [218].

The specific capacitance (Csp, F/g) of the electrode can also be calculated from the constant current discharging curve of GCD experiment using eq 3.3

I × ∆t C = (3.3) sp ∆V × m where I is the constant current, ∆t is the discharge time, ∆V is the discharge potential range and

58 m is the mass of the active materials (MnO2 in this study).

Results and Discussion

Preparation of supercapacitor electrodes

To load nanoparticles into nanoparticles, there are two fabrication methods. One is direct fabrica- tion during electrospinning process, and the other one is indirect fabrication of NPs-electrospun fibers after electrospinning process [219]. Direct fabrication during electrospinning process is the most straightforward strategy to fabricate composite fibers and has been widely used. In this strategy, NPs can be directly added into the polymer solution. In this way, the composite fibers could be synthesized in on-pot by electrospinning the mixed solution as long as the NP-precursor can be uniformly distributed into the electrospun solution. Up to now, there are many metallic

NPs (Au, Ag, Cu, etc.) [220–223], and metal oxide (TiO2, Fe3O4, MgO, ZnO, etc.) [224] are di- rectly fabricated during electrospinning contains. Indirect fabrication, such as surface treatment, hydrothermal sputtering etching, or gas-solid reaction, is usually applied to the NPs that are not well-dispersed in the electrospun solution, or can’t be directly fabricated by large scale.

In the beginning, we tried direct fabrication which is more straight forward and facile. With this method, we fabricate MnO2 nanoparticles by oxidation reaction of Mn(CH3COO)2 and KMnO4 solution. After centrifugation, wash, and dry, TEM was carried out to show the morphology of the MnO2 nanoparticles. According to Figure 3.2, both of α-MnO2 and β-MnO2 were obtained, and the size of the nanoparticles was around 20 nm. The MnO2 nanoparticles powder was then dispersed into 25% PAA solution by horn sonication with a concentration of 10% based on the mass of PAA. After sonication, MnO2 was homogeneously mixed with PAA solution. The solution became dark brown and kept homogeneous even after one year without any precipitation. This

59 homogeneous mixture was then used as the precursor of electrospinning. The electrospinning setup was completely the same as previously mentioned in the experimental section. The as-

prepared fiber mat was light brown, indicating that there were MnO2 particles dispersed in the

fibers. However, TEM image (Figure 3.3a) shows that the MnO2 nanoparticles were not completely even-dispersed. There were still some large particle clusters inside of the fibers. The clusters also made thinner nanofibers as shown in Figure 3.3b. As mentioned previously, the morphology and contact area of MnO2 determines its electrochemical properties. Consequently, these clusters would reduce the efficiency of the particles and impact the electrochemical properties.

Figure 3.2: TEM images of (a) α-MnO2 and (b) β-MnO2nanoparticles.

To produce small and uniform MnO2 nanoparticles dispersed in the fiber structure, we tried in-

direct method. In this method, the precursor manganese acetate (Mn(CH3COO)2) was added

to PAA solution before electrospinning and was converted to MnO2 through an oxidation reac-

tion by KMnO4. Figure 3.1 illustrates the fabrication procedure of MnO2@PAA/PPy core-shell nanofibers. The advantage of using electrospun poly(acrylic acid) (PAA) fibers loaded with Mn2+

ions to fabricate MnO2@PAA/PPy core-shell nanofibers, which can be used as supercapacitor elec-

60 trodes are following: small and well-dispersed MnO2 nanoparticles can be produced in PAA matrix upon oxidation due to the interaction between carboxylate groups and Mn2+ ions. PAA nanofibers

not only provides the support of MnO2 nanoparticles but also functions as an electrolyte media to facilitate cation diffusion during the charge/discharge process. Such structure increases the con-

tacts area between MnO2 and the electrolytes and significantly improves the utilization efficiency

3+ of MnO2. In addition, the loaded Fe will stabilize the PAA fibers and function as the oxidant of pyrrole polymerization to produce a polypyrrole (PPy) shell on PAA fibers. The PPy shell can improve the electrical conductivity as a conductive polymer, increase the capacitance of the elec- trode as a pseudo capacitive material —citewang2012rational and hinder the diffusion of MnOO− to improve the cyclic stability of the electrode.

Figure 3.3: (a) TEM and (b) SEM image for PAA/MnO2 nanofibers fabricated by direct method.

As mentioned previously, the morphology of MnO2 determines its electrochemical properties.To produce small and uniform MnO2 nanoparticles dispersed in the fiber structure, the precursor man- ganese acetate (Mn(CH3COO)2) was added to PAA solution before electrospinning and converted to MnO2 through an oxidation reaction by KMnO4. The rational of preloading manganese acetate into PAA solutions is based on our previous observation that loading metal ions into polyelectrolyte fibers generate metal ion concentration gradient and nanoparticle aggregates upon subsequent re-

61 action (Figure 3.4) [225]. While the interaction between Mn2+ and carboxylate group on PAA is bidentate, the Mn2+ interaction with the second ligand is much weaker than its interaction with the first one. Therefore, Mn2+ does not affect the viscosity of the PAA solution and electrospinning process. As shown in Figure 3.5a, PAA/Mn2+ nanofibers have an average diameter of 200 ∼ 250 nm and similar morphology to pure PAA nanofiber (Figure 3.5b).

Figure 3.4: TEM image of a MnO2@PAA nanofiber produced by a post-loading approach.

62 Figure 3.5: SEM images of PAA/Mn2+ (insert is TEM image of PAA/Mn2+, and the scale bar is 50 nm) (a) and PAA fibers (b).

On the other hand, the weak Mn2+ interactions with PAA make the fiber soluble in water. To stabilize the PAA nanofiber, the as-prepared fiber mat was wetted by a FeCl3 solution, utilizing the strong interactions between PAA and ferric ions [6]. It was observed that the nanofibers were

3+ dissolved or fused together when exposed to the FeCl3 solution with a concentration of Fe lower than 0.2 M. In contrast, the PAA nanofiber mat shrank and became stiff and fragile when exposed

3+ to the FeCl3 solution with a concentration of Fe higher than 0.5 M. PAA fibers are undergoing two processes when they are exposed to a ferric solution: dissolving of PAA and cross-linking of PAA by ferric ions. In the solutions of low ferric concentrations, not enough Fe3+ ions are available to crosslink PAA and prevent the fiber from dissolving, whereas at higher concentrations, too many Fe3+ ions bonded with carboxylate groups and made the fiber brittle and fragile. To keep the morphology and softness of nanofibers, 0.2 M FeCl3 solution was used to stabilize the PAA nanofiber (Figure 3.6a). After being cross-linked with Fe3+, PAA nanofiber mats were stable in water, and the nanostructure was maintained after being immersed in water for 5 days (Figure 3.7). Our previous study shows that the nanostructure of the fiber mat remained constant after 15 days when immersed in phosphate-buffered saline solution after being cross-linked with Fe3+ ions [6].

63 2+ Figure 3.6: SEM images of PAA/Mn fibers cross-linked by 0.2 M FeCl3 (inset is a higher res- olution SEM image of the fiber) (a), the cross-linked fibers after 18 h immersion in water (b),

PAA@MnO2 (c), and PAA@MnO2/PPy (d) fibers after 18 month immersion in water.

64 65

2+ Figure 3.7: SEM images of crosslinked PAA/Mn fibers ((a) and (d)), MnO2@PAA fibers ((b) and (e)), and MnO2@PAA/PPy fibers ((c) and (f)) immersed in water for 2 days and 5 days. MnO2 nanoparticles were produced by the oxidation of Mn(CH3COO)2 by KMnO4. The amount of MnO2 in PAA fiber can be tuned by multiple oxidations by KMnO4. With one-time oxidation, the fiber mat was still flexible and no obvious size change was observed when compared with PAA/Mn2+ fibers (Figure 3.9a). However, the color of the fiber mat changed from light orange to brown after the reaction, indicating the formation of MnO2 nanoparticles on the fibers. According to TEM, the size of the MnO2 nanoparticles for one-time KMnO4 oxidation is smaller than 5 nm

(Figure 3.8a, b). The formation of small MnO2 nanoparticles is due to the interaction between Mn2+ ions and carboxylates in polyelectrolyte fibers. Our previous study [6] revealed that the for- mation of nanostructures on nanofibers was affected by two factors: the diffusion of the embedded ions to the surface and the reaction rate of the metal ions with other chemicals. As the diffusion of Mn2+ in PAA nanofibers is slow because of the interaction between Mn2+ and the carboxy-

2+ late group, the fast oxidation of Mn to MnO2 by KMnO4 leads to small MnO2 nanoparticles.

Multiple KMnO4 oxidations made the fiber mat darker (brown) and more fragile, suggesting the production of more MnO2 nanoparticles upon each oxidation (Figure 3.9b-d). The TEM image of

PAA@MnO2 fibers after the fourth oxidation also shows the formation of larger amounts of MnO2 nanoparticles on the fibers (Figure 3.8c). The weight % and atom % of Mn on the PAA@MnO2 fiber surface after the first oxidation and the fourth oxidation were examined by electron diffraction spectroscopy in SEM, showing that the values increased from 26.04 and 9.31 to 39.63 and 18.10%, respectively (Table 3.1).

66 Figure 3.8: TEM image of images of PAA/MnO2 fibers after one-time KMnO4 oxidation(a).High magnification image (b) shows that MnO2 particle size is less than 5 nm. (c)TEM image of

MnO2@PAA fibers after four-time KMnO4 oxidation.

67 Figure 3.9: SEM images of PAA/MnO2 fibers produced through one (a), two (b), three (c), and four (d) times oxidation using KMnO4. Insets are the digital pictures of fibers showing a darker color with multiple oxidations.

68 Table 3.1: Summary of weight% and atom% of Mn from EDS equipped on SEM

One-time KMnO4 Two-time KMnO4 Three-time KMnO4 Four-time KMnO4 Sample oxidation oxidation oxidation oxidation

Weight % of Mn 26.04 30.08 33.23 39.63 Atom % of Mn 9.31 11.38 12.71 18.10 69 2+ 3+ The composition of PAA/Mn fibers stabilized by Fe and PAA@MnO2 fibers was also in-

vestigated using XPS. XPS results show that the as-spun PAA/MnAc2 fibers have 1.4% atomic percentage of Mn (Figure 3.10a). Additionally, the high-resolution XPS spectrum (Figure 3.11a) of PAA/Mn2+ fibers was used to investigate the chemical state of Mn in PAA/Mn2+ fibers. Char-

acteristic XPS peaks of Mn 2P1/2 (652.8 eV), Mn 2P3/2 (641.2 eV), and the satellite peak (645.5 eV) indicate the existence of Mn2+ in the as-spun fibers [226–229]. The successful loading of

2+ FeCl3 and additional Mn by exposing the as-spun fibers to the FeCl3 and Mn(CH3COO)2 mixed solution is verified by XPS. Mn (5.7%) and Fe (2.7%) in atomic percentage clearly show the exis- tence of Fe and Mn ions in fibers(Figure 3.10b). Similar to the as-spun fibers, signature peaks of

Mn 2P1/2 (652.8 eV) and Mn 2P3/2 (641.1 eV) and the satellite peak (645.2 eV) in Figure 3.11b

indicates the presence of Mn2+ in fibers before KMnO4 oxidation. The XPS survey results of

fibers after KMnO4 oxidation show a higher Mn atomic percentage compared to the fibers before

KMnO4 oxidation (Figure 3.12). The atomic percentage of Mn increased with increased times

of KMnO4 treatment. Additionally, the atomic percentage of Mn after three and four times of

oxidation is almost identical, which indicates a saturated Mn loading after three times of KMnO4 treatment. The chemical state of Mn after oxidation was examined by high-resolution XPS, and the results are shown in Figure 3.11c. The XPS spectra of the samples after different times of

KMnO4 oxidation are almost identical. They all have the Mn 2P1/2 peak at around 654 eV and

Mn 2P3/2 peak at around 642 eV. These two characteristic peaks indicate the formation of MnO2

among all the samples after KMnO4 oxidation [204, 228–230]. In the XPS investigation of Mn 3s peak, Mn 3s duplets at 83.1 and 87.8 eV are observed. The energy separation between these two peaks is 4.7 eV (Figure 3.11a). The position of duplets and the separation between them verify

the formation of MnO2 nanocrystals in the composite fibers. The XRD spectra of the composite

fibers clearly show the signature XRD peaks of α-MnO2 (Figure 3.11b). The lattice constants were calculated to be a = 9.375(5) A˚ and c = 2.908(2) A˚ , which are in good agreement with the standard values (JCPDS 44-0141). Furthermore, the atomic percentage of Mn after two or more

70 2+ times of KMnO4 brushing is higher than the theoretical value (Table 3.2). If 5.7% of Mn in

2+ 7+ 4+ atomic percentage was within the fibers before KMnO4 brushing and 3Mn + 2Mn = 5Mn was the only reaction that happened, the highest Mn atomic percentage that could be reached is 8.6%. The high Mn atomic percentage (i.e., >20%) is probably because of the reaction between

KMnO4 and acrylic acid residues in PAA to produce MnO2.

2+ Figure 3.10: XPS spectrum of PAA/Mn fibers (a) and fibers before KMnO4 oxidation (b).

71 Figure 3.11: (a) Deconvoluted high-resolution XPS spectrum of the as-spun PAA/MnAc2 nanofibers. (b) Deconvoluted high-resolution XPS spectrum of PAA/MnAc2 fibers stabilized by

3+ Fe . (c) High-resolution XPS spectra of fibers after different times of KMnO4 oxidation (14 represents the oxidation times).

72 Figure 3.12: XPS spectra of fibers after (a) one, (b) two, (c) three, and (d) four times of KMnO4 oxidation.

73 Figure 3.13: XPS spectrum of Mn 3s (a), and XRD image (b) of composites after KMnO4 Oxida- tion.

74 Table 3.2: Summary of Elemental Compositions from XPS

Atomic Concentration (%) Sample Mn C O Fe

PAA/Mn2+ fibers 1.4 60.9 37.7 N/A Fe3+ stabilized PAA/Mn2+ fibers 5.7 61.8 29.8 2.7

Fibers after 1 time of KMnO4 oxidation 6.9 62.2 30.8 N/A

Fibers after 2 time of KMnO4 oxidation 15.5 37.6 46.9 N/A

Fibers after 3 time of KMnO4 oxidation 22.2 29.2 48.6 N/A

Fibers after 4 time of KMnO4 oxidation 22.5 28.2 49.3 N/A

− To prevent the degradation of MnO2 through the diffusion of MnOO during the charge/discharge process, a positively charged conductive polymer layer was deposited as a shell on the surface of the PAA nanofiber, which also can improve the conductivity of the electrode. Polyaniline, PPy and Poly(3,4-ethylenedioxythiophene) are the most common conductive polymers used in SCs [191, 196, 197, 199, 203, 205–207, 210, 211, 231–236]. PPy was chosen as the shell material in this study because of its low cost and neutral polymerization condition [197, 199, 203, 235, 236]. In our studies, the PAA@MnO2 fibers were immersed in pyrrole solutions for 1 min, where the Fe3+ ions in PAA fibers polymerized pyrrole and produced 50 nm PPy shells on PAA fibers (Fig- ure 3.14). Compared with the reported approach of producing the PPy shell using oxidants in solutions [237, 238], our method produced a uniform PPy shell on the fiber surface because the polymerization was controlled by the diffusion of Fe3+ ions in PAA fibers. Longer time polymer- ization would produce a thick PPy shell and granule that would significantly reduce the porosity and surface area of the fiber mat and reduce the accessibility of the electrolyte, imposing a negative effect on the electrochemical performance (Figure 3.15). On the basis of the mass ratio and atomic

75 percentage of each element obtained from the EDS in HRTEM, the weight percentage of MnO2 in the electrode materials is 14.43 and 13.19% for one-time KMnO4 oxidation on PAA@MnO2 and PAA@MnO2/PPy, respectively. This was taken as the active material mass for electrochem- istry study because the EDS in HRTEM reveals the composition of the whole fiber [239]. The as-prepared PAA@MnO2 and PAA@MnO2/PPy fiber mats were stable and the fibrous structure remained constant even after 18 month immersion in water (Figure 3.6c,d). Electrical conductivi-

2+ −4 −4 ties of PAA/Mn , PAA@MnO2, and PAA@MnO2/PPy fibers are 3.22 × 10 , 2.61 × 10 , and 1.19 × 10−3 S/m, respectively. The data show that the PPy shell increases the electrical conduc- tivity of PAA/MnO2 fibers.

Figure 3.14: (a) SEM image of the PAA@MnO2/PPy fiber produced by 1 min polymerization. The inset shows that the fibers are flexible. (b) HRTEM image of the fiber showing about a 50 nm thick

PPy shell and small MnO2 particles inside the fiber.

76 Figure 3.15: An SEM image of PAA@MnO2/PPy fiber produced by 5 minutes (a) and 10 minutes (b) polymerization.

Electrochemical Characterization

To study the electrochemical performance of the as-prepared freestanding PAA@MnO2/PPy fibers, CV measurements at different scan rates were performed by a three-electrode configuration in

1 M Na2SO4 aqueous solution. Figure 3.16b shows that the CV curves of PAA@MnO2/PPy

fibers, which are more rectangular than those of PAA@MnO2 fibers (Figure 3.16a), indicating

that PAA@ MnO2/PPy fibers have lower charge transfer resistance, leading to an improved capac- itive behavior. The higher current density in PAA@MnO2/PPy fibers compared to other composite

fibers indicates more efficient MnO2 utilization and electron transportation, as PPy is used as a

protecting shell. The CV measurement of PAA@MnO2/PPy at different scan rates ranging from 2 to 100 mV/s (Figure 3.16b) shows rectangular consistency, suggesting its outstanding capacitive

behavior. This superior performance is attributed to several facts: (1) small MnO2 nanoparticles significantly increased the number of electrochemically active sites for the redox reaction, con-

sequently increasing the MnO2 utilization efficiency; (2) although the porous structure of PAA fibers cannot contribute to electron conductivity as other carbon materials [240], it improves elec-

77 trolyte penetration and promotes the ion diffusion within the electrode materials; (3) the PAA/Fe3+ cross-linking network provides effective pathways for charge transport [148]; and (4) PPy, as the conductive shell, also enormously improved the electron transfer at the interface.

Figure 3.16: CV measurements of (a) PAA@MnO2, (b) PAA@MnO2/PPy at different scan rates,

3+ 3+ and (c) PAA/Fe , PAA/Fe /PPy, PAA@MnO2, and PAA@MnO2/PPy at a scan rate of 10 mV/s.

To study the contribution of Fe3+ and PPy to the specific capacitance, CV measurements of

3+ 3+ PAA/Fe and PAA/Fe / PPy without MnO2 were carried out. The results reveal that the contri- bution of Fe3+ and PPy is not significant (Figure 3.16c). The specific capacitance of PAA/Fe3+ and PAA/Fe3+/PPy was less than 5 and 10%, respectively, of the capacitance obtained from

PAA@MnO2/PPy (Table 3.3). The gravimetric specific capacitance of PAA@MnO2/PPy fibers reaches up to 564 F/g at a scan rate of 10 mV/s, which is about twice that of PAA@MnO2 fibers (288 F/g). The specific capacitance calculation is based on the subtraction of the capacitance which

3+ is attributed to Fe and PPy from the total capacitance. PAA@MnO2/PPy fibers showed a capac- itance about twice that of PAA@MnO2 fibers at different scan rates. The loading mass of MnO2 can be controlled by varying the KMnO4 oxidation time. However, higher loading mass of MnO2

78 reduced the gravimetric specific capacitance. For example, the specific capacitance of the sample with two-time oxidation by KMnO4 was 320 F/g and the one with 3-time loading was only 203 F/g with 0.2 mg loading mass for both the electrodes. According to the fiber morphology, multiple oxidations by KMnO4 increased the size of the MnO2 particle, thereby reducing its utilization and lowering the conductivity of the electrode.

3+ 3+ Table 3.3: Specific capacitance of PAA/Fe , PAA/Fe /PPy, PAA@MnO2 and PAA@MnO2/PPy at a scan rate of 10 mV/s.

3+ 3+ Material PAA/Fe PAA/Fe /PPy PAA@MnO2 PAA@MnO2/PPy Specific Capacitance (F/g) 26 55 288 546

To study the charge storage capacity of different composite fibers, we performed GCD measure- ments from 0 to 0.8 V at current density ranging from 5 to 50 A/g, as shown in Figure 3.17. The charging and discharging curves were symmetric, indicating that the electrochemical behavior of the composite fibers was reversible. While PAA@MnO2/PPy electrodes have high specific capac- itance (692 F/g) at 0.5 A/g, the charge/discharge curve is asymmetric. It is believed that the long discharging time at 0.5 A/g led to water splitting during the charge/discharge process and caused the asymmetric charging and discharging curves [241, 242]. At low current density, the charge/ discharge process is slow, which gives it enough time for water to be electrolyzed, whereas at high current density, the process is much faster and the contribution of water splitting is insignificant. Therefore, the specific capacitance obtained at 5A/g reflects a more realistic performance. In our study, PAA@MnO2/PPy fibers have higher specific capacitance than PAA@MnO2 fibers because of the PPy shell. The specific capacitance of PAA@MnO2/PPy fibers is higher than most of the re- ported systems with MnO2 on carbon nanofibers [201, 209, 243, 244]. MnO2 on carbon nanofibers and MnO2/PPy on carbon nanofibers have been reported to have high specific capacitance (e.g.,

79 855 [215] and 700 F/g [203]); however, the performance reduced to 87% after 5000 cycles and

91% after 2000 cycles, respectively. In comparison, the specific capacitance of PAA@MnO2/PPy in our study remains 100% of its initial value after 5000 cycles at a scan rate of 100 mV/s (Figure

3.18). The increase of the capacitance of PAA@MnO2/PPy before 1000 cycles results from the improved usage of MnO2, the wetting of the electrode surface by the electrolyte, and the swelling of the fiber during the charge/discharge process [245]. In the beginning of the charge/ discharge

− process, MnOO diffuses from the core and deposits as MnO2 at the PAA/PPy interface, lead-

− ing to an improved utilization of MnO2. At the same time, the PPy shell prevents MnOO from diffusing into the electrolyte solution. The capacitance value stops increasing when the diffu- sion of MnOO− species reaches equilibrium. In contrast, the step-like decrease of capacitance of

PAA@MnO2 in Figure 3.18 was attributed to the loss of MnO2 during the charge/discharge pro- cess [200]. MnOO− species diffuse from the fiber to the electrolyte solution without the protection of the PPy shell, leading to decreased capacitance.

Figure 3.17: GCD measurements of PAA@MnO2 (a) and PAA@MnO2/PPy (b). The insets are specific capacitance at different current densities.

80 Figure 3.18: Cycling performance of PAA@MnO2 and PAA@MnO2/PPy at a scan rate of 100 mV/s.

The morphology and structure of PAA@MnO2 and PAA@ MnO2/PPy fibers after the long-term cycle were determined by SEM (Figure 3.19). The PAA@MnO2 fiber structure has been com- pletely disintegrated after the long-term cycle, whereas the PAA@MnO2/PPy fiber structure was still preserved. According to EDS in SEM, the atomic percentage of Mn was 21 and 2.6% for PAA@MnO2 460 fiber and PAA@MnO2/PPy fiber, respectively. The preserved morphol- ogy and the superior electrochemical performance of PAA@MnO2/PPy fibers suggest that the

− PPy shell greatly improved the cycling stability of MnO2 nanoparticles by preventing MnOO from emigrating out of the fibers during the charge/discharge process. As shown in Figure 3.18,

PAA@MnO2/PPy fibers maintained full capacitance, whereas PAA@MnO2 fibers retained 77% of their initial value after 5000 cycles. The loss of MnO2 was also reflected by the color change of the electrolyte solutions after 5000 cycles, where the electrolyte solution for PAA@MnO2 fibers became yellow, whereas that for PAA@MnO2/PPy remained colorless. In addition, the Nyquist plots of PAA@MnO2 and PAA@MnO2/PPy fibers before the cycle test (Figure 3.20) show that the

81 solution resistance of PAA@MnO2 and PAA@MnO2/PPy fibers are 3.42 and 1.48 Ω, respectively, indicating that PPy significantly enhanced the conductivity of the electrode material.

Figure 3.19: SEM images of PAA@MnO2 (a) and PAA@MnO2/PPy (b) after 5000 cycles at a scan rate of 100 mV/s.

Figure 3.20: EIS of PAA@MnO2 and PAA@MnO2/PPy.

82 Conclusion

In summary, PAA@MnO2/PPy core-shell freestanding and flexible fiber mats were fabricated from electrospun PAA fibers loaded with Mn2+ ions. The fibers were stabilized by Fe3+ ions which were subsequently used as the oxidants to polymerize pyrrole. Small and uniform MnO2 nanoparticles were produced in PAA fibers through the oxidation of KMnO4. The obtained flexible nanostruc- tured mats were further investigated as supercapacitor electrodes, showing remarkable enhance- ments in specific capacitance, charge/discharge ability, as well as cyclic stability. The reported method provides a facile and effective approach to produce flexible supercapacitor electrodes.

83 CHAPTER 4: GEMCITABINE LOADED POLY(ACRYLIC

ACID)-CHITOSAN NANOPARTICLES FOR TARGETED THERAPY OF

PANCREATIC CANCER

Abstract

CS/TPP-PAA core-shell nanoparticle was fabricated by a facile method to post-load drug gemc- itabine through incubation the nanoparticles with high concentration gemcitabine. The dense core of CS/TPP nanoparticle provides a stable structure for PAA conjugation, while the shell PAA pro- vides large space for drug loading and negative charge to make itself stable in tissue environment and prevent the adsorption of protein. To obtain the optimum size and zeta potential, the ratios between the ingredient were investigated. The chosen ratios were CS:TPP 2:1 (w:w), PAA:CS 1:1 (w:w), and CS:EDC 1:1 (w:w). The pH for fabrication was 6, at where the nanoparticles were under most stable condition. TEM images confirmed the core-shell structure at around 300 nm. The encapsulation efficiency (EE) and loading capacity (LC) are 3.64% and 5.04%, respectively. In vitro releasing profile show that this drug delivery system has a prolonged controlled release of gemcitabine (36.74% gemcitabine released in 12 days), thus is a potential system for cell and clinical study.

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is the 4th leading cause of cancer-associated death in the US and has less than a 5% 5-year survival rate [246]. PDAC is often diagnosed at a late stage and has a very complex and unique tumor microenvironment (TME) with elevated interstitial

84 fluid pressure (IFP), which narrows the tumor vasculature and restricts drug delivery [247–251]. Gemcitabine (GEM, 2’-deoxy-2’, 2’-difluorocytidine), a b-difluoronucleoside, has been used in the treatment of various cancers, such as lung, head, neck and colon cancers. The antitumor action of GEM involves the inhibition of DNA synthesis by inhibition of ribonucleotide reductase and by competition with dCTP for incorporation into DNA [252]. Although GEM presents anticancer ac- tivity, its properties such as rapid enzymatic degradation, poor cellular uptake, and fast metabolism into inactive uracil derivatives [253–255]. This leads to the using of high drug dose, which causes severe side-effect to normal tissue.

These conundrums make it urgent to develop an efficient delivery system of gemcitabine. GEM with Abraxane (nanoparticle albuminbound paclitaxel) [256] or FOLFIRINOX (a combination of leucovorin, 5-fluorouracil, irinotecan, and oxaliplatin) [257] is the current frontline treatments for PDAC by blocking DNA replication and inducing S-phase cell cycle arrest [258], but have minimal survival benefits [256, 257]. Many drug delivery carriers, such as liposome and micelles can’t improve GEM plasma stability due to their low drug loading efficiency and fast release [258– 262]. Conjugation of GEM onto lipid and polymeric carriers improves the loading efficiency, but has low aqueous solubility [263, 264]. So far, most of these drugs are not tumor selective, which limits the delivery of an effective dose to the tumor without systemic toxicity. In addition, the properties of the tumor tissue, such as high hydrostatic pressure, disorganized vasculature and lacking of functional lymphatics, slowdown the movement of drug molecules towards to the tumor cell, consequently lower the drug efficiency [254]. Hence, there is an unmet need to develop a multimodal platform for effective drug delivery with improved therapeutic outcomes, meaning that get a maximum therapeutic effect with minimum toxicity.

The development of nanotechnology has made possibilities that could help to enhance antitu- mor efficacy by increasing drug penetration due to the small size. Biodegradable hydrogel sys- tems allow sustained release of chemotherapeutics locally [265–275]. Furthermore, combina-

85 tions of nanoparticles and hydrogel have shown its promise for carrying cytotoxic agents, in- cluding chemotherapeutic drugs, to effectively impede the growth of cancer in preclinical set- tings [273, 276–282]. Comparing with liposome, polymer carriers cause an extra intracellular trafficking pathway, which prolongs drug release, leading to less drug dose usage. Also, polymer nanovector delivers higher payload to the cells than that of liposomal vectors, due to less diffusion of drugs through the platform [255]. Chitosan(CS), a non-toxic, biocompatible and biodegrad- able polysaccharide with high charge density, has gained considerable attention and been explored as potential drug carrier and delivery systems [283, 284]. Its susceptibility to lysozyme makes CS biodegradable to be an ideal material for controlled release of drug or genes [254]. Repro- ducible chitosan hydrogel nanoparticles can be fabricated via chemical cross-linking mediated by polyanions. A commonly studied polyanion is tripolyphosphate (TPP) due to its biocompatibil- ity [284]. CS/TPP nanoparticles are attractive carrier and delivery systems due to their formation under mild conditions, monodispersity, adjustable size and positive surface charge [285]. However, recent studies have demonstrated that positively charged nanoparticles strongly interact with serum components. These interactions lead to aggregation and consequently, clearance from circulation, limiting in vivo applications [286, 287]. A previous study by Y. Wu and others developed an ap- proach toward the formation of stable CS/TPP nanoparticles in biological media by hydrophilic modification on chitosan particle surface to enhance its solubility. This was done by fabricating CS/TPP nanoparticles by standard ionic crosslinking technique. Next, poly(acrylic acid) PAA was electrostatically attracted to positively charged CS/TPP, followed by covalently bonded to surface of CS/TPP using carbodiimide (EDC) [288].

In this study, we replicate the aforementioned nanoparticles and attempt to load with GEM to study in vitro and in vivo long-term release. This core-shell nanoparticle system provides several advan- tages: (a) GEM can be loaded not only in core CS/TPP nanoparticles due to formation of hydrogen bonds with CS, but also in PAA shell by electrostatic interactions between amine groups of GEM

86 and carboxyl groups of PAA, leading to a high drug loading amount; (b) the negatively charged surface and hydrophilicity resulting from the PAA shell improves the resistivity to protein adsorp- tion and electrostatic repulsion with cellular membrane, which prevent cellular internalization in normal tissue [288]; (c) the swelling-deswelling property of this core-shell hydrogel nanoparticle can be triggered in a tumor environment, i.e., low pH environment [288]. Thus, this nanoparti- cle system can provide the ideal carrier and delivery system of GEM to PDAC tumor cells. This study focuses on the optimization of the fabrication of these nanoparticles and studies drug load- ing, release, and biocompatibility testing of GEM after loading. Since epidermal growth factor receptor (EGFR) is overexpressed on the surface of cancer cells, we selected anti-EGFR antibody GE11 (sequence YHWYGYTPQNVI) as the targeting moiety to delivery the drug delivery system to pancreatic cancer cells. Future work aims to functionalize with targeting peptide GE11 and perform animal studies, and we expect endocytotic mechanism when the drug delivery system is uptaken by the cells.

Experimental Procedure

Materials

25% poly(acrylic acid) (PAA, MW = ∼250,000 g/mol) solution, low molecular weight chitosan (CS, MW = 50,000 ∼ 190,000 g/mol) and tripolyphosphate (TPP, 85%), were purchased from Sigma-Aldrich Chemical Co. (USA). 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochlo- ride (EDC) was purchased from Thermo Scientific (Japan). Gemcitabine (GEM, 98%) was pur- chased from Ark Pharm, Inc (USA). All chemicals were of analytical grade, and they were used as received. Nanopure water was obtained by passing house deionized water through a Barnstead NANOpure Diamond UV ultrapure water system.

87 Fabrication of Nanofibers

Based on previously reported method [284], chitosan-tripolyphosphate (CS/TPP) nanoparticle was achieved by ionic gelation between positively charged CS amine groups and negatively charged sodium tripolyphosphate (TPP). Briefly, solid low molecular weight chitosan was dissolved in 1 wt% acetic acid to achieve a 0.5 mg/ml stock solution. The pH of the chitosan solution was then adjusted to 6 by 1 M sodium hydroxide (NaOH). TPP was dissolved in nanopure water to make a 0.5 mg/ml stock solution. The TPP solution was then added dropwise (1 drop/second) into 10 mL chitosan solution at ambient temperature under intensive stirring. The nanoparticles were produced at selected chitosan to TPP weight ratios of 1:1, 2:1, 3:1, 4:1 and 5:1. The CS/TPP solution was then stirred for another 10 minutes before being stored at 10◦C for overnight.

After fabrication of the CS/TPP “template” nanoparticles, poly(acrylic acid) (PAA) was introduced to conjugate onto the surface of the CS/TPP nanoparticle as a shell by electrostatic interaction be- tween chitosan and PAA following layer-by-layer mechanism. 1% (w/v) PAA solution (pH 6) was quickly added to the CS/TPP nanoparticle suspension while stirring. Different PAA:CS mass ratios of 1:1, 5:1, 10:1, 15:1 and 20:1 were investigated. The PAA/CS/TPP solution is then allowed to stir for 30 minutes before the addition of 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochlo- ride (EDC) for crosslinking between the amine groups on CS and carboxylic acid groups on PAA to form amide bonds. Various amount of EDC solution (0.5 mg/mL) was added to PAA/CS/TPP nanoparticle suspension based on the ratio with CS to study its effect on particle properties. The suspension was then stirred for overnight and filtered through a 0.45 m filter the following day. The supernatant was collected for later use. The structure of GEM loaded core-shell PAA-CS/TPP nanoparticle is showed in Figure 4.1.

For pH study, different amount of 1 M HCl or 1 M NaOH was slowly added to the nanoparticle suspension to tune the pH value. The pH was determined by pH meter after being calibrated at 25

88 ◦C.

Figure 4.1: Schematic representations of Gemcitabine loaded PAA/CS/TPP nanoparticle

Characterization of the nanoparticles

The hydrodynamic diameter and zeta potential of the nanoparticles were determined by dynamic light scattering (DLS, Nano-ZS90, Malvern Instruments Ltd, Worcestershire, UK) equipped with 4 mW He Ne laser at a scattering angle of 90◦ at 532 nm at 25 ◦C. Polydispersity index (PDI) was also measured to determine particle size distribution. The supernatant after filtered by 0.45 µm filter was directly used for DLS analysis. All the measurements were repeated three times, and an average value was reported. The stability of nanoparticles under simulated physiological conditions was monitored by size measurements at different time points.

The morphology of the nanoparticles was characterized by transmission electron microscopy (TEM, JEOL 1011). To prepare the TEM samples, the sample solution was dropped onto a carbon-coated copper grid, and dried in air. The acceleration voltage was 100 kV.

89 Incubation and encapsulation

Since GEM can be conjugated to CS and PAA molecules by hydrogen bonding and electrostatic interaction between amine group and carboxylic group, respectively, the PAA-CS/TPP nanoparti- cle is an excellent vehicle for GEM delivery system. The as-prepared PAA-CS/TPP nanoparticles were then incubated with high concentration GEM (15 mg/mL) for 48 hours. After incubation, nanoparticles were separated from the suspension by ultra-centrifuged at 60,000 rpm for 30 min- utes at 4 ◦C in an ultra centrifuge (Sorvall MTX150 Micro Ultracentrifuge, Thermo Scientific). The supernatant was collected to determine the encapsulation efficiency (EE) and loading capacity (LC) of GEM via UV-visible spectroscopy at the characteristic wavelength = 268 nm. The pellet was redispersed in nanopure water and frozen for lyophilization to obtain the mass of nanoparti- cles. Encapsulation efficiency (EE) and loading capacity (LC) of GEM were calculated as follows:

EE = (mass of encapsulated Gemcitabine)/(mass of starting material Gemcitabine)×100

(4.1)

LC = (mass of encapsulated Gemcitabine)/(total mass of nanoparticle) × 100 (4.2)

In vitro drug releasing

1 mL of redispersed nanoparticle suspension (0.5 mg/mL) was sealed in a dialysis membrane bag (spectra pro molecular porous membrane tubing, MWCO = 12 ∼ 14 kD) and immersed in 50 mL of nanopure water. The system was incubated at room temperature (25 ◦C) under continuous gentle

90 stirring, and the dialysis was kept for up to 12 days. 3 mL of the releasing phase was extracted from the incubation medium at predetermined time intervals, and was analyzed via UV-visible spectroscopy at the characteristic wavelength of λ = 268 nm. After withdrawing each aliquot, the incubation medium was replenished with 3 mL of fresh nanopure water.

Results and Discussion

The final product GEM loaded nanoparticle GemPAA-CS/TPP core-shell structure is demonstrated in Figure 4.1. As mentioned before, CS/TPP nanoparticles were formed as the core through ionic crosslinking, where TPP acts as a crosslinker. Then, PAA is conjugated onto CS at the surface of CS/TPP core as shell by covalent bond between carboxylic group on PAA and amine group on CS to form PAA-CS/TPP core-shell nanoparticle. Finally, GEM is loaded by incubation of PAA- CS/TPP nanoparticles in GEM solution mainly through the interaction of GME and PAA. Several parameters were determined to obtain an optimal result.

Effect of CS:TPP ratio on CS/TPP nanoparticle properties

A few CS:TPP ratios (w:w) were tested to determine optimal composition of core nanoparticle. The size and zeta potential of the nanoparticle were characterized by DLS and TEM. The effect of the mass ratio of CS and TPP on the particle size, stability, PDI and zeta potential is summarized in Figure 4.2. Except for CS:TPP ratio at 1:1, Figure 4.2a shows that the CS/TPP particle size was under 150 nm, and higher CS:TPP ratio results in larger diameter, while lower ratios results in smaller diameter, due to TPP entangled with more CS chains, leading to dense particles, which is consistent with previous report [289]. Higher CS:TPP ratio leads to higher zeta potential (Figure 4.2c), resulting in more stable nanoparticle suspension due to the stronger repulsion between the

91 nanoparticles. With higher CS:TPP ratio at range of 2:1 to 5:1, the size change gets smaller during the storage period. It’s 31 nm and 12 nm for CS:TPP rate at 2:1 and 5:1, respectively. When CS:TPP ratio is 1:1, the particle size is much larger than the other ratios, and the particles tended to aggregate during storage period. This is caused by the large amount of free TPP connecting the existing loose nanoparticles and forming aggregation. Even though less size change occurred at higher CS:TPP ratio, the ratio of CS:TPP of 2:1 was chosen as the optimal ratio due to small size and relatively low PDI (Figure 4.2b).

Figure 4.2: Schematic representations of Gemcitabine loaded PAA/CS/TPP nanoparticle.

From the TEM image in Figure 4.3a, the dark contrast shows that CS/TPP nanoparticles are dense, as expected, due to the gelation of CS molecules upon interaction with TPP. TEM image indicates a size of approximately 30 ∼ 40 nm for CS/TPP compared to DLS measured size of approximately 100 nm, because dense and hydrophobic particles were dry and shrunk during the preparation of TEM sample, while DLS reveals the hydrodynamic diameter and the particles swell in aqueous solution.

92 Figure 4.3: TEM image of CS/TPP core nanoparticles (a) and GEM loaded CS/TPP/PAA nanopar- ticles (b).

Effect of PAA:CS ratio on CS/TPP nanoparticle properties

It is essential for the nanoparticle to be stable in the tissue environment as a nanovector for drug delivery system. Since the isoelectric point of most proteins is under 6 and the pH of body fluid is higher than 7, the proteins are usually negatively charged in biological environment.

In this study, PAA was introduced to the system for surface modification of CS/TPP nanoparticles to make negatively charged surface to prevent protein absorption by repulsion of negative charge on protein. In addition, the hydrophilic property of PAA can improve nanoparticle stability in biological environments. The effect of PAA:CS ratio on particle size and zeta potential were mea- sured by DLS, and the result is shown in Figure 4.4. As expected, the size of the nanoparticles

93 increased, and the zeta potential became more negative, as more PAA was added. When PAA was added, the positive charge on the particle surface was neutralized by the negative charges on PAA, and the corona formed by PAA molecules rendered large particle size than CS/TPP nanoparticles and led to a cloudy suspension. The corona becomes thicker when more PAA is added. Mean- while, the zeta potential of the nanoparticle suspension decrease from -29 mV to -45 mV when the PAA:CS ratio increases from 1:1 to 20:1, due to more PAA molecules on the particle surface increases the negative charge density. Although the highly negative zeta potential implies repulsive forces for a stable suspension of nanoparticles, excess PAA brings up large size nanoparticles and unstable colloid due to the low interaction rate between PAA and CS, and the unpaired chain of PAA is easy to conjugate onto other CS/TPP nanoparticles, leading to flocculation. Consequently, PAA:CS ratio of 1:1 was fixed for remaining characterization and drug loading due to desirable size and zeta potential results. The nanoparticle size and zeta potential at this ratio is 268 nm and -29mV, respectively. The zeta potential is around -30 mV, which causes a strong enough repulsion for a table suspension [289].

94 Figure 4.4: Effect of PAA:CS ratio on PAA-CS/TPP nanoparticle size (black line) and zeta poten- tial (red line).

Effect of CS:EDC ratio on CS/TPP-PAA nanoparticle properties

Since both PAA and CS are weak polyelectrolyte, the electrostatic interaction between them can be easily broken at tissue environment, and leads to fast drug release. In this study, EDC is intro- duced into the system to covalently crosslink PAA and CS by the formation of amide bond [290]. As previous reported [288], when CS:EDC molar ratio is higher than 1:1, the suspension is un- stable, because there are free CS molecules on the nanoparticle surface that are not covered by PAA molecules. In our study, CS:EDC ratios of 1:1, 1:2, 1:5 and 1:10 were investigated to deter- mine the effect of the amount of EDC on the size and zeta potential of CS/TPP-PAA nanoparticle (Figure 4.5). Comparing to CS/TPP/PAA nanoparticle at PAA:CS 1:1, the size of the nanoparti- cles decreased from 268 nm to 230 nm, at various CS:EDC ratios. This indicates that the EDC crosslinking endows the nanoparticles denser through covalent bonding. EDC and PAA are always

95 excess at other CS:EDC ratios, where all CS molecules are involved into crosslinking, and no ob- vious size change when more EDC is added. The zeta potential decreases from -29 mV to -31mV after EDC is added at CS:EDC ratio 1:1, because the excess and unbounded PAA would leave the nanoparticle surface. The surface charge of the nanoparticles slightly increased with higher ratio of EDC, due to the present of intermediate product after the interaction of PAA and EDC.

Figure 4.5: Effect of CS:EDC ratio on PAA-CS/TPP nanoparticle size (black line) and zeta poten- tial (red line).

Figure 4.3b provides a TEM image of PAA conjugated CS/TPP nanoparticles, which is a core-shell structure. The size of the core is approximately 50 nm, and the total diameter of the nanoparticles is approximately 300 nm. This is larger than the measured size via DLS (230 nm). This is because during the dry process of TEM sample preparation, the soft PAA shell of the nanoparticle flattened out and formed “pancake” structure due to its high hydrophilicity, which causes a larger diameter observed for the CS/TPP/PAA nanoparticles. Figure 4.3b confirms the core-shell structure and demonstrates a dense core and soft shell, implying that majority of GEM loading would occur in the shell.

96 Effect of pH value on CS/TPP-PAA nanoparticle properties

It’s important to note that the pH of the solution plays a critical role during fabrication of PAA- CS/TPP nanoparticles because both CS and PAA are weak polyelectrolyte. As previously reported, the extracellular pH of normal tissues and blood is 7.4, the tumor extracellular pH is ∼6.8, the en- dosomal pH is 5.0-6.0 and the lysosomal pH is 4.0-5.0 [291]. The environmental pH and pKa of CS (6.5) and PAA (4.2) determine the global charge densities, which in turn affect the electrostatic interaction between the hydroxyl group of PAA and amine group of CS [288]. When the pH of the solution is below 4.2, both CS and PAA are completely protonated, therefore no interaction between the two polymers. When the pH is higher than 6.5, CS is completely deprotonated and PAA is completely ionized, resulting in no bonding for the polymers. Electrostatic attraction be- tween CS and PAA occurs at pH 4.2 ∼ 6.5, when CS and PAA are both partially protonated. From experimentation, it was determined that addition of PAA at pH lower than 5 and higher than 6.5 resulted in precipitation. Thus, fabrication needs to be performed within the pH range 5 ∼ 6.5 to achieve maximal electrostatic interaction.

Nanoparticles composed of CS:TPP 2:1 and PAA:CS 1:1 were fabricated and exposed to vari- ous pH environments and then characterized. Figure 4.6 demonstrates the size and zeta potential change in response to environmental pH value after EDC was added. Before EDC was added, there is only electrostatic interaction between PAA and CS, which is pretty weak and can be easily impacted by pH. CS/TPP/PAA nanoparticle stability was investigated at pH 2, 5, 7, 9, and 11 after fabricated at pH 6 and determined by DLS. The size of nanoparticles become larger at low and high pH (larger than 1000 nm). Further more, the values of PDI are beyond 0.4 except for pH 7 (size was 280 nm, and PDI was 0.204), whereas the PDI for original samples at pH 6 was 0.140, meaning that CS/TPP/PAA nanoparticles are only stable at pH 6 ∼ 7. As mentioned previously, there’s no interaction between PAA and CS due to full protonation at low pH, leading to nanopar-

97 ticle swell, or even be destroyed. The nanoparticle suspension even became clear at pH 2, and nanoparticles almost couldn’t be detected by DLS. PAA and CS are completely deprotonated at high pH, resulting in inaccessible interaction, and the nanoparticles are not stable. The nanoparti- cles became much more stable after EDC was added. Even though the particle size became larger at pH 2, the PDI was still pretty low (0.253), indicating that the nanoparticles are just swelled but neither destroyed nor conjugated. The size and PDI are both kept at good condition at pH 5 and 7 compared to pH 6. At higher pH 9 and 11, DLS peaks divided into several ones at large size, and PDI were larger than 0.6, meaning the nanoparticles are unstable, and aggregation may occur. pH value affects zeta potential. At pH 2, zeta potential is as high as 21 mV, which is caused by the protonated CS. When pH is higher (from pH 5), zeta potential approximately keeps the same at -20 ∼ -30 mV.

Figure 4.6: Effect of pH value on PAA-CS/TPP nanoparticle size (black line) and zeta potential (red line).

98 In vitro drug releasing study

Release study of GEM loaded NP was performed for the sample with CS:TPP 2:1, PAA:CS 1:1, and CS:EDC 1:1. Figure 4.7 shows the in vitro releasing behavior of GEM loaded PAA-CS/TPP nanoparticles. The encapsulation efficiency (EE) is 3.64% with a loading capacity (LC) of 5.04%. The initial burst release of the GEM was 29.5% in the first 24 hours of releasing, which occurred in many drug delivery systems [254, 255, 292], due to the hydrophilicity of gemcitabine. The loaded nanoparticles continued to release in 12 days, with a maximum release of GEM at 36.74%. Comparing to other nanovectors [254, 255, 292], our system PAA-CS/TPP nanoparticles manifests prolonged releasing profile. This could be because of the thick PAA shell, which can interact with GEM through electrostatic attraction. This property makes it a potential drug delivery system with low drug injection frequency, while keeping high plasma drug concentration, therefore reduce the drug dosage. Further in vitro releasing study will be performed in human pancreatic carcinoma cell line (PANC1), which exhibits GEM resistance, by Mayo Clinic.

99 Figure 4.7: In-vitro release profile of GEM in nanopure water. Experiments were carried out at 37 ◦C. Values are the average of three different experiments standard deviation.

Conclusion

In conclusion, we were able to successfully fabricate core-shell nanoparticles composed of core CS/TPP crosslinked with PAA using EDC as the shell and load GEM. Upon characterization of nanoparticles, it was found that the nanoparticles were of desirable size (230 nm), stability and loading capacity of GEM. Since there is a thick PAA shell, large amount of GEM and be loaded by incubating PAA-CS/TPP nanoparticles with high concentration GEM solution. The encapsulation efficiency (EE) and loading capacity (LC) are 3.64% and 5.04%, respectively. In addition, the nanoparticle presents a prolonged releasing profile, which would lead to good clinical performance. More experimentation is currently being performed to fabricate nanoparticles functionalized with GE11 targeting protein and test out the performance in cell lines.

100 CHAPTER 5: SUMMARY

In this dissertation, mechanical properties and applications in energy storage and biotechnology of PAA-based PECs are investigated. Electrospinning was adapted to produce nanofibers for mechan- ical properties study and fabricate nanofiber mat for supercapacitor electrode, while the nanopar- ticles for drug delivery system was fabricated in aqueous solution. Both methods are facile under ambient conditions. The major achievements are summarized as follows:

1. Mechanical properties of single PAA/CS nanofiber was first demonstrated by dark field mi- croscopy imaging and persistence length analysis at various ferric ion concentration. The persis- tence length and Young’s modulus were carried out to determine the stiffness of the fibers. FTIR was also tested out to obtain the chelating style of PAA and ferric ion. As a result, the nanofiber became stiffest at 0.1% ferric ion concentration (molar percentage to PAA monomer), where the nanofibers have the longest persistence length and highest Young’s modulus, during the concen- tration range 0% ∼ 2%. At this point, there’s almost no difference of stiffness for the PAA/Fe3+ with or without CS. According to the analysis of FTIR, the chelating style between PAA and Fe3+ is bidentate bridging.

2. Core-shell nanofibers containing poly(acrylic acid) (PAA) and manganese oxide nanoparticles as the core and polypyrrole (PPy) as the shell were fabricated. PAA solution with Mn2+ was elec- trospun to fabricate nanofiber mat. The nanofiber mat was then stabilized by ferric ion through its interaction between carboxylate group on PAA, so that the fiber mat became insoluble in water.

2+ Mn in the nanofibers was then oxidated by KMnO4 to form MnO2 nanoparticles as the active ma- terial for the supercapacitor. The results show small size and well-dispersed MnO2 nanoparticles due to the in-situ fabrication in PAA nanofibers. Fe3+ served as the catalyst for pyrrole polymer- ization when the shell was coated. The positively charged PPy shell benefits the lifetime of MnO2

101 during charge/discharge process, by preventing the diffusion out of MnOO− species. As a result, the coreshell structured fibers exhibited outstanding properties including high specific capacitance, excellent reversible redox reactions, fast charge/discharge ability, and long cycle life.

3. Core-shell nanoparticles with CS/TPP core and PAA shell was fabricated as a drug delivery system to load drug gemcitabine. The ratios of the ingredients were optimized to obtain the op- timum size and dispersity of the nanoparticles. Since the nanoparticles were made of weak poly- electrolytes, they are sensitive to pH. The best pH for the nanoparticles is 5 ∼ 7. Post loading by incubating nanoparticles with GEM solution was applied. The encapsulation efficiency (EE) is 3.64% with a loading capacity (LC) of 5.04%. This drug delivery system has a prolonged releasing profile, in which there was 36.74% of the total loaded GEM was released in 12 days. Future work will focus on the validation of the results in vivo, and conjugation of targeting moiety (GE11) to accomplish the targeted release-controlled drug delivery system.

102 APPENDIX A: LIST OF PUBLICATIONS

103 1. Lu, Xiaoyan, Chen Shen, Zeyang Zhang, Elizabeth Barrios, and Lei Zhai.“Core-Shell Com- posite Fibers for High Performance Flexible Supercapacitor Electrodes” ACS applied materials & interfaces (2018), 10(4), 4041-4049.

2. Deng, Jinan, Xiaoyan Lu, Colin Constant, Aristide Dogariu, and Jiyu Fang.“Design of β- CDsurfactant complex-coated liquid crystal droplets for the detection of cholic acid via competitive hostguest recognition” Chemical Communications 2015, 51(43), 8912-8915.

104 APPENDIX B: PERMISSIONS FOR COPYRIGHTED MATERIALS

105 Permission Request Policies from ACS Publications

The following wording is from the ACS Thesis/Dissertation Policy and the ACS Journal Publishing Agreement:

Reuse/Republication of the Entire Work in Theses or Collections: Authors may reuse all or part of the Submitted, Accepted or Published Work in a thesis or dissertation that the author writes and is required to submit to satisfy the criteria of degree-granting institutions. Such reuse is permitted subject to the ACS Ethical Guidelines to Publication of Chemical Research; the author should secure written confirmation (via letter or email) from the respective ACS journal editor(s) to avoid potential conflicts with journal prior publication*/embargo policies. Appropriate citation of the Published Work must be made**. If the thesis or dissertation to be published is in electronic format, a direct link to the Published Work must also be included using the ACS Articles on Request author-directed link.

106 LIST OF REFERENCES

[1] Martin Hubbe, Anna Sundberg, Paulina Mocchiutti, Yonghao Ni, and Robert Pelton. Dissolved and colloidal substances (dcs) and the charge demand of papermaking pro- cess waters and suspensions: A review. BioResources, 7(4), 2012. ISSN 1930-2126. URL http://ojs.cnr.ncsu.edu/index.php/BioRes/article/view/ BioRes_07_4_6109_Hubbe_Dissolved_Colloidal_Substances_Review.

[2] Eishun Tsuchida. Formation of polyelectrolyte complexes and their structures. Journal of Macromolecular SciencePure and Applied Chemistry, 31(1):1–15, 1994.

[3] Heidi Vogt Sæther, Hilde K Holme, Gjertrud Maurstad, Olav Smidsrød, and Bjørn T Stokke. Polyelectrolyte complex formation using alginate and chitosan. Carbohydrate Polymers, 74 (4):813–821, 2008.

[4] Fengyu Li, Yong Zhao, and Yanlin Song. Core-shell nanofibers: nano channel and capsule by coaxial electrospinning. In Nanofibers. InTech, 2010.

[5] Jingwei Xie, Xiaoran Li, and Younan Xia. Putting electrospun nanofibers to work for biomedical research. Macromolecular Rapid Communications, 29(22):1775–1792, 2008. ISSN 1521-3927. doi: 10.1002/marc.200800381. URL http://dx.doi.org/10. 1002/marc.200800381.

[6] Astha Malhotra, Tanmay Bera, and Lei Zhai. Bioinspired metal ion coordinated polyelec- trolyte fibrous nanoreactors. Advanced Materials Interfaces, 3(22):1600692–n/a, 2016. ISSN 2196-7350. doi: 10.1002/admi.201600692. URL http://dx.doi.org/10. 1002/admi.201600692. 1600692.

[7] Domokos Bartis and Judit Pongracz.´ Three dimensional tissue cultures and tissue engi-

107 neering. Teach. Mater. Med. Biotechnol. Masters Program. Univ. Pecs´ Univ. Debrecen, 15, 2011.

[8] Nazila Kamaly and Andrew D Miller. Paramagnetic liposome nanoparticles for cellular and tumour imaging. International journal of molecular sciences, 11(4):1759–1776, 2010.

[9] Jessica L. Freyer, Spencer D. Brucks, and Luis M. Campos. Fully charged: Maximizing the potential of cationic polyelectrolytes in applications ranging from membranes to gene delivery through rational design. Journal of Polymer Science Part A: Polymer Chemistry, 55(19):3167–3174, 2017. ISSN 1099-0518. doi: 10.1002/pola.28711. URL http://dx. doi.org/10.1002/pola.28711.

[10] Subhash C Bhatia and Natarajan Ravi. A magnetic study of an fe- chitosan complex and its relevance to other biomolecules. Biomacromolecules, 1(3):413–417, 2000.

[11] Satish Nayak and L. Andrew Lyon. Soft nanotechnology with soft nanoparticles. Ange- wandte Chemie International Edition, 44(47):7686–7708, 2005. ISSN 1521-3773. doi: 10. 1002/anie.200501321. URL http://dx.doi.org/10.1002/anie.200501321.

[12] Kevin C. Taylor and Hisham A. Nasr-El-Din. Water-soluble hydrophobically associ- ating polymers for improved oil recovery: A literature review. Journal of Petroleum Science and Engineering, 19(3):265 – 280, 1998. ISSN 0920-4105. doi: https://doi. org/10.1016/S0920-4105(97)00048-X. URL http://www.sciencedirect.com/ science/article/pii/S092041059700048X.

[13] Nicholas B. Wyatt and Matthew W. Liberatore. Rheology and viscosity scaling of the poly- electrolyte xanthan gum. Journal of Applied Polymer Science, 114(6):4076–4084, 2009. ISSN 1097-4628. doi: 10.1002/app.31093. URL http://dx.doi.org/10.1002/ app.31093.

108 [14] Nicholas B. Wyatt and Matthew W. Liberatore. The effect of counterion size and valency on the increase in viscosity in polyelectrolyte solutions. Soft Matter, 6:3346–3352, 2010. doi: 10.1039/C000423E. URL http://dx.doi.org/10.1039/C000423E.

[15] Erdener Karadag,˘ Zehra Deniz Kasim, Semiha Kundakcı, and Omer¨ Barıs¸ Uz¨ um.¨ Acry- lamide/potassium 3-sulfopropyl methacrylate/sodium alginate/bentonite hybrid hydrogels: Synthesis, characterization and its application in lauths violet removal from aqueous so- lutions. Fibers and Polymers, 18(1):9–21, Jan 2017. ISSN 1875-0052. doi: 10.1007/ s12221-017-5910-z. URL https://doi.org/10.1007/s12221-017-5910-z.

[16] Gang Li, Guoping Zhang, Rong Sun, and Ching-Ping Wong. Dually ph-responsive poly- electrolyte complex hydrogel composed of polyacrylic acid and poly (2-(dimthylamino) ethyl methacrylate). Polymer, 107(Supplement C):332 – 340, 2016. ISSN 0032-3861. doi: https://doi.org/10.1016/j.polymer.2016.11.037. URL http://www.sciencedirect. com/science/article/pii/S0032386116310382. Self-Assembly.

[17] Huaiping Zhang, Kai Xu, Hao Ai, Dehong Chen, Linli Xv, and Mingcai Chen. Synthe- sis, characterization and solution properties of hydrophobically modified polyelectrolyte poly(aa-co-tmspma). Journal of Solution Chemistry, 37(8):1137, May 2008. ISSN 1572-8927. doi: 10.1007/s10953-008-9300-4. URL https://doi.org/10.1007/ s10953-008-9300-4.

[18] Huijuan Bai, Haoqin Zhang, Yakun He, Jindun Liu, Bing Zhang, and Jingtao Wang. En- hanced proton conduction of chitosan membrane enabled by halloysite nanotubes bear- ing sulfonate polyelectrolyte brushes. Journal of Membrane Science, 454(Supplement C):220 – 232, 2014. ISSN 0376-7388. doi: https://doi.org/10.1016/j.memsci.2013. 12.005. URL http://www.sciencedirect.com/science/article/pii/ S0376738813009587.

109 [19] Fei Zhao, Li Li, Yu-chuan Tian, Jian-jia Liu, Jian-jun Wang, Zhi-ming Zhou, Chun-xiang Lv, and Xu-hong Guo. Preparation of au/ag multilayers via layer-by-layer self-assembly in spherical polyelectrolyte brushes and their catalytic activity. Chinese Journal of Polymer Science, 33(10):1421–1430, Oct 2015. ISSN 1439-6203. doi: 10.1007/s10118-015-1700-5. URL https://doi.org/10.1007/s10118-015-1700-5.

[20] Hossein Adelnia and Saeed Pourmahdian. Soap-free emulsion polymerization of poly (methyl methacrylate-co-butyl acrylate): effects of anionic comonomers and methanol on the different characteristics of the latexes. Colloid and Polymer Science, 292(1):197– 205, Jan 2014. ISSN 1435-1536. doi: 10.1007/s00396-013-3043-5. URL https: //doi.org/10.1007/s00396-013-3043-5.

[21] Sajjad Haider and Soo-Young Park. Preparation of the electrospun chitosan nanofibers and their applications to the adsorption of cu(ii) and pb(ii) ions from an aqueous solu- tion. Journal of Membrane Science, 328(1):90 – 96, 2009. ISSN 0376-7388. doi: https://doi.org/10.1016/j.memsci.2008.11.046. URL http://www.sciencedirect. com/science/article/pii/S0376738808010156.

[22] Paul Rouster, Marko Pavlovic, Endre Horvth, Lszl Forr, Sandwip K. Dey, and Ist- van Szilagyi. Influence of protamine functionalization on the colloidal stability of 1d and 2d titanium oxide nanostructures. Langmuir, 0(0):null, 0. doi: 10.1021/ acs.langmuir.7b01815. URL http://dx.doi.org/10.1021/acs.langmuir. 7b01815. PMID: 28829607.

[23] Paul Rouster, Marko Pavlovic, and Istvan Szilagyi. Improving the stability of titania nanosheets by functionalization with polyelectrolytes. RSC Adv., 6:97322–97330, 2016. doi: 10.1039/C6RA23707J. URL http://dx.doi.org/10.1039/C6RA23707J.

[24] David A. Sievers, James J. Lischeske, Mary J. Biddy, and Jonathan J. Stickel. A

110 low-cost solidliquid separation process for enzymatically hydrolyzed corn stover slurries. Bioresource Technology, 187:37 – 42, 2015. ISSN 0960-8524. doi: https://doi.org/10. 1016/j.biortech.2015.03.087. URL http://www.sciencedirect.com/science/ article/pii/S0960852415004198.

[25] Beata Bien´ and Jurand D Bien.´ Coagulant and polyelectrolyte application performance testing in sonicated sewage sludge dewatering. Desalination and Water Treatment, 57(3): 1154–1162, 2016.

[26] Yasine Sakhawoth, Laurent J. Michot, Pierre Levitz, and Natalie Malikova. Flocculation of clay colloids induced by model polyelectrolytes: Effects of relative charge density and size. ChemPhysChem, 18(19):2756–2765, 2017. ISSN 1439-7641. doi: 10.1002/cphc. 201700430. URL http://dx.doi.org/10.1002/cphc.201700430.

[27] Nishkriya Kalyanaraman, Cameron Arnold, Aadish Gupta, Jyun Syung Tsau, and Reza Barati Ghahfarokhi. Stability improvement of co2 foam for enhanced oil-recovery applications using polyelectrolytes and polyelectrolyte complex nanoparticles. Journal of Applied Polymer Science, 134(6):n/a–n/a, 2017. ISSN 1097-4628. doi: 10.1002/app.44491. URL http://dx.doi.org/10.1002/app.44491.

[28] Mahesh Budhathoki, Sai Hari Ram Barnee, Bor-Jier Shiau, and Jeffrey H Harwell. Improved oil recovery by reducing surfactant adsorption with polyelectrolyte in high saline brine. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 498:66–73, 2016.

[29] Armin Mooranian, Rebecca Negrulj, Ryu Takechi, Emma Jamieson, Grant Morahan, and Hani Al-Salami. Electrokinetic potential-stabilization by bile acid-microencapsulating for- mulation of pancreatic -cells cultured in high ratio poly-l-ornithine-gel hydrogel colloidal dispersion: applications in cell-biomaterials, tissue engineering and biotechnological ap- plications. Artificial Cells, Nanomedicine, and Biotechnology, 0(0):1–7, 2017. doi: 10.

111 1080/21691401.2017.1362416. URL http://dx.doi.org/10.1080/21691401. 2017.1362416. PMID: 28776395.

[30] KGID Kumari, Per Moldrup, Marcos Paradelo, Lars Elsgaard, and Lis W de Jonge. Ef- fects of biochar on dispersibility of colloids in agricultural soils. Journal of Environmental Quality, 46(1):143–152, 2017.

[31] A. Rabiee. Acrylamide-based anionic polyelectrolytes and their applications: A survey. Journal of Vinyl and Additive Technology, 16(2):111–119, 2010. ISSN 1548-0585. doi: 10.1002/vnl.20229. URL http://dx.doi.org/10.1002/vnl.20229.

[32] Teemu Kinnarinen, Mohammad Golmaei, and Antti Hkkinen. Use of filter aids to improve the filterability of enzymatically hydrolyzed biomass suspensions. Industrial & Engineering Chemistry Research, 52(42):14955–14964, 2013. doi: 10.1021/ie4021057. URL http: //dx.doi.org/10.1021/ie4021057.

[33] Danjun Wu and Thierry Delair. Stabilization of chitosan/hyaluronan colloidal polyelec- trolyte complexes in physiological conditions. Carbohydrate polymers, 119:149–158, 2015.

[34] Abhijeet D. Kulkarni, Yogesh H. Vanjari, Karan H. Sancheti, Harun M. Patel, Veena S. Belgamwar, Sanjay J. Surana, and Chandrakantsing V. Pardeshi. Polyelectrolyte complexes: mechanisms, critical experimental aspects, and applications. Artificial Cells, Nanomedicine, and Biotechnology, 44(7):1615–1625, 2016. doi: 10.3109/21691401.2015.1129624. URL http://dx.doi.org/10.3109/21691401.2015.1129624. PMID: 26757773.

[35] Yuehua Sun, Chaoqun Peng, Xiaofeng Wang, Richu Wang, Junliang Yang, and Dou Zhang. Rheological behavior of al2o3 suspensions containing polyelectrolyte complexes for direct ink writing. Powder Technology, 2017.

112 [36] Xuehong Wang, Jie Wei, Jianding Chen, and Songchao Tang. Tuning of the surface bi- ological behavior of poly(l-lactide)-based composites by the incorporation of polyelec- trolyte complexes for bone regeneration. Journal of Biomaterials Science, Polymer Edi- tion, 28(15):1713–1727, 2017. doi: 10.1080/09205063.2017.1348926. URL http: //dx.doi.org/10.1080/09205063.2017.1348926. PMID: 28657452.

[37] Jing-Kun Yan, Wen-Yi Qiu, Yao-Yao Wang, and Jian-Yong Wu. Biocompatible polyelec- trolyte complex nanoparticles from lactoferrin and pectin as potential vehicles for antiox- idative curcumin. Journal of Agricultural and Food Chemistry, 65(28):5720–5730, 2017. doi: 10.1021/acs.jafc.7b01848. URL http://dx.doi.org/10.1021/acs.jafc. 7b01848. PMID: 28657749.

[38] Zhanna A. Boeva and Vladimir Sergeyev. Soluble and highly ionically conducting in- terpolyelectrolyte complexes prepared via chemical template polymerization of aniline in the presence of perfluorinated polysulfonic acid. RSC Adv., 4:56677–56681, 2014. doi: 10.1039/C4RA11217B. URL http://dx.doi.org/10.1039/C4RA11217B.

[39] Zhiqiang Feng, Zhipeng Wang, Changyou Gao, and Jiacong Shen. Hollow microcap- sules with a complex polyelectrolyte shell structure fabricated by polymerization of 4- vinylpyridine in the presence of poly (sodium 4-styrenesulfonate) and silica particles. Ma- terials Letters, 61(11):2560–2564, 2007.

[40] Yulia V Shulevich, Gudrun Petzold, Alexander V Navrotskii, and Ivan A Novakov. Prop- erties of polyelectrolyte–surfactant complexes obtained by polymerization of an ionic monomer in a solution of an oppositely charged surfactant. Colloids and Surfaces A: Physic- ochemical and Engineering Aspects, 415:148–152, 2012.

[41] C. L. de Vasconcelos, P. M. Bezerril, T. N. C. Dantas, M. R. Pereira, and J. L. C. Fonseca. Adsorption of bovine serum albumin on template-polymerized chitosan/poly(methacrylic

113 acid) complexes. Langmuir, 23(14):7687–7694, 2007. doi: 10.1021/la700537t. URL http://dx.doi.org/10.1021/la700537t. PMID: 17547429.

[42] Simona De Robertis, Maria Cristina Bonferoni, Lisa Elviri, Giuseppina Sandri, Carla Caramella, and Ruggero Bettini. Advances in oral controlled drug delivery: the role of drug–polymer and interpolymer non-covalent interactions. Expert opinion on drug deliv- ery, 12(3):441–453, 2015.

[43] Carolina De las Heras Alarcon,´ Sivanand Pennadam, and Cameron Alexander. Stimuli responsive polymers for biomedical applications. Chemical Society Reviews, 34(3):276– 285, 2005.

[44] Cesar´ Marquez-Beltr´ an,´ L. Castaneda,˜ M. Enciso-Aguilar, Gerardo Paredes-Quijada, Herib- erto Acuna-Campa,˜ Amir Maldonado-Arce, and Jean-Franc¸ois Argillier. Structure and mechanism formation of polyelectrolyte complex obtained from pss/pah system: effect of molar mixing ratio, base–acid conditions, and ionic strength. Colloid and Polymer Science, 291(3):683–690, Mar 2013. doi: 10.1007/s00396-012-2775-y. URL https: //doi.org/10.1007/s00396-012-2775-y.

[45] William B. Liechty, David R. Kryscio, Brandon V. Slaughter, and Nicholas A. Pep- pas. Polymers for drug delivery systems. Annual Review of Chemical and Biomolecu- lar Engineering, 1(1):149–173, 2010. doi: 10.1146/annurev-chembioeng-073009-100847. URL https://doi.org/10.1146/annurev-chembioeng-073009-100847. PMID: 22432577.

[46] Vladimir Torchilin. Multifunctional and stimuli-sensitive pharmaceutical nanocarriers. Eu- ropean Journal of Pharmaceutics and Biopharmaceutics, 71(3):431–444, 2009.

[47] Yong Hu, Xiqun Jiang, Yin Ding, Haixiong Ge, Yuyan Yuan, and Changzheng Yang. Syn-

114 thesis and characterization of chitosan–poly (acrylic acid) nanoparticles. Biomaterials, 23 (15):3193–3201, 2002.

[48] Manfred Schliwa and Gnther Woehlke. Molecular motors. Nature, 422:759–765, 2003.

[49] Abdalin E Asinas and Robert M Weis. Competitive and cooperative interactions in receptor signaling complexes. Journal of Biological Chemistry, 281(41):30512–30523, 2006.

[50] Elia Beniash. Biomineralshierarchical nanocomposites: the example of bone. Wiley Inter- disciplinary Reviews: Nanomedicine and Nanobiotechnology, 3(1):47–69, 2011.

[51] Hongyan Zhen, Tingting Wang, Rui Jia, Baowei Su, and Congjie Gao. Preparation and performance of antibacterial layer-by-layer polyelectrolyte nanofiltration membranes based on metal-ligand coordination interactions. RSC Adv., 5:86784–86794, 2015. doi: 10.1039/ C5RA15427H. URL http://dx.doi.org/10.1039/C5RA15427H.

[52] Hernan´ A. Maturana, Ivan´ M. Peric,ˇ Bernabe´ L. Rivas, and S. Amalia Pooley. Interac- tion of heavy metal ions with an ion exchange resin obtained from a natural polyelec- trolyte. Polymer Bulletin, 67(4):669–676, Aug 2011. ISSN 1436-2449. doi: 10.1007/ s00289-011-0454-7. URL https://doi.org/10.1007/s00289-011-0454-7.

[53] Bernab L. Rivas, Antonio E. Maureira, and Mara A. Mondaca. Aminodiacetic water-soluble polymermetal ion interactions. European Polymer Journal, 44(7):2330 – 2338, 2008. ISSN 0014-3057. doi: https://doi.org/10.1016/j.eurpolymj.2008.05.001. URL http://www. sciencedirect.com/science/article/pii/S0014305708002164.

[54] Bernab L. Rivas and Eduardo D. Pereira. Synthesis of copolymers containing opposite charged comonomers and their interactions with metal ions. Journal of Applied Polymer Science, 89(6):1715–1721, 2003. ISSN 1097-4628. doi: 10.1002/app.12237. URL http: //dx.doi.org/10.1002/app.12237.

115 [55] AP Nikalje. Nanotechnology and its applications in medicine. Med chem, 5(2):081–089, 2015.

[56] Herbert Dautzenberg. Polyelectrolyte complex formation in highly aggregating systems. 1. effect of salt: polyelectrolyte complex formation in the presence of nacl. Macromolecules, 30(25):7810–7815, 1997.

[57] Linda Gardlund,¨ Lars Wagberg,˚ and Renate Gernandt. Polyelectrolyte complexes for sur- face modification of wood fibres: Ii. influence of complexes on wet and dry strength of paper. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 218(1):137– 149, 2003.

[58] Caroline Ankerfors, Sedat Ondaral, Lars Wagberg,˚ and Lars Odberg.¨ Using jet mixing to prepare polyelectrolyte complexes: Complex properties and their interaction with silicon oxide surfaces. Journal of colloid and interface science, 351(1):88–95, 2010.

[59] Christophe Schatz, Jean-Michel Lucas, Christophe Viton, Alain Domard, Christian Pichot, and Thierry Delair. Formation and properties of positively charged colloids based on poly- electrolyte complexes of biopolymers. Langmuir, 20(18):7766–7778, 2004.

[60] Martin Muller,¨ Bernd Keßler, Johanna Frohlich,¨ Sebastian Poeschla, and Bernhard Torger. Polyelectrolyte complex nanoparticles of poly (ethyleneimine) and poly (acrylic acid): preparation and applications. Polymers, 3(2):762–778, 2011.

[61] G Che, BB Lakshmi, CR Martin, ER Fisher, and Rodney S Ruoff. Chemical vapor deposi- tion based synthesis of carbon nanotubes and nanofibers using a template method. Chemistry of Materials, 10(1):260–267, 1998.

[62] Hong-tao Cui, O Zhou, and Brian R Stoner. Deposition of aligned bamboo-like carbon

116 nanotubes via microwave plasma enhanced chemical vapor deposition. Journal of Applied Physics, 88(10):6072–6074, 2000.

[63] Esmaeil Biazar. Application of polymeric nanofibers in medical designs, part i: Skin and eye. International Journal of Polymeric Materials and Polymeric Biomaterials, 66(10): 521–531, 2017. doi: 10.1080/00914037.2016.1276062. URL http://dx.doi.org/ 10.1080/00914037.2016.1276062.

[64] Rajkishore Nayak, Rajiv Padhye, Illias Louis Kyratzis, Yen Bach Truong, and Lyndon Arnold. Recent advances in nanofibre fabrication techniques. Textile Research Journal, 82 (2):129–147, 2012. doi: 10.1177/0040517511424524. URL https://doi.org/10. 1177/0040517511424524.

[65] Jiajia Xue, Jingwei Xie, Wenying Liu, and Younan Xia. Electrospun nanofibers: New concepts, materials, and applications. Accounts of Chemical Research, 50(8):1976–1987, 2017. doi: 10.1021/acs.accounts.7b00218. URL http://dx.doi.org/10.1021/ acs.accounts.7b00218. PMID: 28777535.

[66] Elmira Hadipour-Goudarzi, Majid Montazer, Masoud Latifi, and Ali Akbar Ghare Aghaji. Electrospinning of chitosan/sericin/pva nanofibers incorporated with in situ synthesis of nano silver. Carbohydrate polymers, 113:231–239, 2014.

[67] Haoqin Ma, Guangkai Chen, Jingnan Zhang, Yong Liu, Jun Nie, and Guiping Ma. Facile fabrication of core-shell polyelectrolyte complexes nanofibers based on electric field in- duced phase separation. Polymer, 110:80–86, 2017.

[68] Jay R Werber, Chinedum O Osuji, and Menachem Elimelech. Materials for next-generation desalination and water purification membranes. Nature Reviews Materials, 1:16018, 2016.

117 [69] Jovan Kamcev and Benny D. Freeman. Charged polymer membranes for envi- ronmental/energy applications. Annual Review of Chemical and Biomolecular En- gineering, 7(1):111–133, 2016. doi: 10.1146/annurev-chembioeng-080615-033533. URL https://doi.org/10.1146/annurev-chembioeng-080615-033533. PMID: 26979410.

[70] Paul A Beales, Sanobar Khan, Stephen P Muench, and Lars JC Jeuken. Durable vesicles for reconstitution of membrane proteins in biotechnology. Biochemical Society Transactions, 45(1):15–26, 2017.

[71] S. Breiter. 1 - membranes for oxygenators and plasma filters. In Michael Lysaght and Thomas J. Webster, editors, Biomaterials for Artificial Organs, Woodhead Publish- ing Series in Biomaterials, pages 3 – 33. Woodhead Publishing, 2011. ISBN 978-1- 84569-653-5. doi: https://doi.org/10.1533/9780857090843.1.3. URL http://www. sciencedirect.com/science/article/pii/B9781845696535500017.

[72] Nithya Joseph, Joice Thomas, Pejman Ahmadiannamini, Hans Van Gorp, Roy Bernstein, Steven De Feyter, Mario Smet, Wim Dehaen, Richard Hoogenboom, and Ivo F. J. Vank- elecom. Ultrathin single bilayer separation membranes based on hyperbranched sulfonated poly(aryleneoxindole). Advanced Functional Materials, 27(9):1605068–n/a, 2017. ISSN 1616-3028. doi: 10.1002/adfm.201605068. URL http://dx.doi.org/10.1002/ adfm.201605068. 1605068.

[73] Elvira Tjipto, John F. Quinn, and Frank Caruso. Assembly of multilayer films from polyelectrolytes containing weak and strong acid moieties. Langmuir, 21(19):8785–8792, 2005. doi: 10.1021/la051197h. URL http://dx.doi.org/10.1021/la051197h. PMID: 16142961.

[74] Feng Wu, Jian Li, Yuefeng Su, Jing Wang, Wen Yang, Ning Li, Lai Chen, Shi

118 Chen, Renjie Chen, and Liying Bao. Layer-by-layer assembled architecture of poly- electrolyte multilayers and graphene sheets on hollow carbon spheres/sulfur compos- ite for high-performance lithiumsulfur batteries. Nano Letters, 16(9):5488–5494, 2016. doi: 10.1021/acs.nanolett.6b01981. URL http://dx.doi.org/10.1021/acs. nanolett.6b01981. PMID: 27479273.

[75] Aliaksandr Zhuk, Victor Selin, Iryna Zhuk, Benjamin Belov, John F. Ankner, and Svet- lana A. Sukhishvili. Chain conformation and dynamics in spin-assisted weak polyelectrolyte multilayers. Langmuir, 31(13):3889–3896, 2015. doi: 10.1021/acs.langmuir.5b00401. URL http://dx.doi.org/10.1021/acs.langmuir.5b00401. PMID: 25768113.

[76] Young Hoon Ko, Young Hun Kim, Juhyun Park, Ki Tae Nam, Jong Hyeok Park, and Pil J. Yoo. Electric-field-assisted layer-by-layer assembly of weakly charged polyelectrolyte multilayers. Macromolecules, 44(8):2866–2872, 2011. doi: 10.1021/ma102112a. URL http://dx.doi.org/10.1021/ma102112a.

[77] Wen Zhu, Haizhi Wang, Guoshan Jing, Qianqian Liu, Chunxi Li, and Hong Meng. Rapid spray-crosslinked assembly of a stable high-performance polyelectrolyte bipolar membrane. RSC Adv., 7:36313–36318, 2017. doi: 10.1039/C7RA05367C. URL http://dx.doi. org/10.1039/C7RA05367C.

[78] Mingjie Yin, Jinwen Qian, Quanfu An, Qiang Zhao, Zhangliang Gui, and Jing Li. Poly- electrolyte layer-by-layer self-assembly at vibration condition and the pervaporation perfor- mance of assembly multilayer films in dehydration of isopropanol. Journal of Membrane Science, 358(1):43–50, 2010.

[79] Daniel Morales, Etienne Palleau, Michael D. Dickey, and Orlin D. Velev. Electro-actuated hydrogel walkers with dual responsive legs. Soft Matter, 10:1337–1348, 2014. doi: 10. 1039/C3SM51921J. URL http://dx.doi.org/10.1039/C3SM51921J.

119 [80] Mario Casolaro and Ilaria Casolaro. Polyelectrolyte hydrogel platforms for the delivery of antidepressant drugs. Gels, 2(4):24, 2016. ISSN 2310-2861. doi: 10.3390/gels2040024. URL http://www.mdpi.com/2310-2861/2/4/24.

[81] Ali Zolfagharian, Abbas Z Kouzani, Sui Yang Khoo, Bijan Nasri-Nasrabadi, and Akif Kay- nak. Development and analysis of a 3d printed hydrogel soft actuator. Sensors and Actuators A: Physical, 265:94–101, 2017.

[82] Qiang Zhao, Dong Woog Lee, B Kollbe Ahn, Sungbaek Seo, Yair Kaufman, Jacob N Is- raelachvili, and J Herbert Waite. Underwater contact adhesion and microarchitecture in polyelectrolyte complexes actuated by solvent exchange. Nature materials, 15(4):407, 2016.

[83] Wei Long Ng, Wai Yee Yeong, and May Win Naing. Development of polyelectrolyte chitosan-gelatin hydrogels for skin bioprinting. Procedia CIRP, 49:105–112, 2016.

[84] J. Berger, M. Reist, J.M. Mayer, O. Felt, N.A. Peppas, and R. Gurny. Structure and interac- tions in covalently and ionically crosslinked chitosan hydrogels for biomedical applications. European Journal of Pharmaceutics and Biopharmaceutics, 57(1):19 – 34, 2004. ISSN 0939-6411. doi: https://doi.org/10.1016/S0939-6411(03)00161-9. URL http://www. sciencedirect.com/science/article/pii/S0939641103001619. Chi- tosan.

[85] Howard G Schild. Poly (n-isopropylacrylamide): experiment, theory and application. Progress in polymer science, 17(2):163–249, 1992.

[86] Yasushi Maeda, Tomomi Higuchi, and Isao Ikeda. Change in hydration state during the coil- globule transition of aqueous solutions of poly (n-isopropylacrylamide) as evidenced by ftir spectroscopy. Langmuir, 16(19):7503–7509, 2000.

[87] Tohru Shiga, Yoshiharu Hirose, Akane Okada, and Toshio Kurauchi. Bending of poly (vinyl

120 alcohol)–poly (sodium acrylate) composite hydrogel in electric fields. Journal of Applied Polymer Science, 44(2):249–253, 1992.

[88] Megan L. OGrady, Po-ling Kuo, and Kevin Kit Parker. Optimization of electroactive hydrogel actuators. ACS Applied Materials & Interfaces, 2(2):343–346, 2010. doi: 10.1021/am900755w. URL http://dx.doi.org/10.1021/am900755w. PMID: 20356178.

[89] Robert Pelton. Temperature-sensitive aqueous microgels. Advances in colloid and interface science, 85(1):1–33, 2000.

[90] Youn Soo Kim, Mingjie Liu, Yasuhiro Ishida, Yasuo Ebina, Minoru Osada, Takayoshi Sasaki, Takaaki Hikima, Masaki Takata, and Takuzo Aida. Thermoresponsive actuation enabled by permittivity switching in an electrostatically anisotropic hydrogel. Nature mate- rials, 14(10):1002, 2015.

[91] Oleksandra Zavgorodnya, Veronika Kozlovskaya, and Eugenia Kharlampieva. Nanos- tructured highly-swollen hydrogels: Complexation with amino acids through copper (ii) ions. Polymer, 74:94 – 107, 2015. ISSN 0032-3861. doi: https://doi.org/10. 1016/j.polymer.2015.08.010. URL http://www.sciencedirect.com/science/ article/pii/S003238611530149X.

[92] Xingwei Shi, Yanli Hu, Kai Tu, Lina Zhang, Hao Wang, Jian Xu, Hongming Zhang, Ji Li, Xianhong Wang, and Min Xu. Electromechanical polyaniline-cellulose hydrogels with high compressive strength. Soft Matter, 9:10129–10134, 2013. doi: 10.1039/C3SM51490K. URL http://dx.doi.org/10.1039/C3SM51490K.

[93] Jiyoung M Dang and Kam W Leong. Natural polymers for gene delivery and tissue engi- neering. Advanced drug delivery reviews, 58(4):487–499, 2006.

121 [94] Gloria AA Saracino, Daniela Cigognini, Diego Silva, Andrea Caprini, and Fabrizio Gelain. Nanomaterials design and tests for neural tissue engineering. Chemical Society Reviews, 42 (1):225–262, 2013.

[95] Kefeng Ren, Thomas Crouzier, Christian Roy, and Catherine Picart. Polyelectrolyte mul- tilayer films of controlled stiffness modulate myoblast cell differentiation. Advanced func- tional materials, 18(9):1378–1389, 2008.

[96] Benjamin Thierry, Franc¸oise M Winnik, Yahye Merhi, and Maryam Tabrizian. Nanocoat- ings onto arteries via layer-by-layer deposition: toward the in vivo repair of damaged blood vessels. Journal of the American Chemical Society, 125(25):7494–7495, 2003.

[97] Varvara Gribova, Rachel Auzely-Velty, and Catherine Picart. Polyelectrolyte multilayer assemblies on materials surfaces: From cell adhesion to tissue engineering. Chemistry of Materials, 24(5):854–869, 2012. doi: 10.1021/cm2032459. URL http://dx.doi. org/10.1021/cm2032459. PMID: 25076811.

[98] Christopher J Detzel, Adam L Larkin, and Padmavathy Rajagopalan. Polyelectrolyte multi- layers in tissue engineering. Tissue Engineering Part B: Reviews, 17(2):101–113, 2011.

[99] Halima Kerdjoudj, Cedric´ Boura, Vanessa Moby, Karine Montagne, Pierre Schaaf, J-C Voegel, J-F Stoltz, and Patrick Menu. Re-endothelialization of human umbilical arteries treated with polyelectrolyte multilayers: A tool for damaged vessel replacement. Advanced Functional Materials, 17(15):2667–2673, 2007.

[100] Kaiyong Cai, Annett Rechtenbach, Jianyuan Hao, Jrg Bossert, and Klaus D. Jandt. Polysaccharide-protein surface modification of titanium via a layer-by-layer technique: Characterization and cell behaviour aspects. Biomaterials, 26(30):5960 – 5971, 2005. ISSN 0142-9612. doi: https://doi.org/10.1016/j.biomaterials.2005.03.020. URL http://www. sciencedirect.com/science/article/pii/S0142961205002541.

122 [101] Nicholas A Peppas, Robert Langer, et al. New challenges in biomaterials. Science-AAAS- Weekly Paper Edition-including Guide to Scientific Information, 263(5154):1715–1719, 1994.

[102] Nicholas A Peppas. Hydrogels in medicine and pharmacy: properties and applications, volume 3. CRC PressI Llc, 1987.

[103] Yoshihito Osada and Jianping Gong. Stimuli-responsive polymer gels and their application to chemomechanical systems. Progress in Polymer Science, 18(2):187–226, 1993.

[104] Ryo Yoshida. Self-oscillating gels driven by the belousov–zhabotinsky reaction as novel smart materials. Advanced Materials, 22(31):3463–3483, 2010.

[105] J.P. Gong, Y. Katsuyama, T. Kurokawa, and Y. Osada. Double-network hydrogels with extremely high mechanical strength. Advanced Materials, 15(14):1155–1158, 2003. ISSN 1521-4095. doi: 10.1002/adma.200304907. URL http://dx.doi.org/10.1002/ adma.200304907.

[106] Kazunori Yasuda, Jian Ping Gong, Yoshinori Katsuyama, Atsushi Nakayama, Yoshie Tan- abe, Eiji Kondo, Masaru Ueno, and Yoshihito Osada. Biomechanical properties of high- toughness double network hydrogels. Biomaterials, 26(21):4468–4475, 2005.

[107] Chinatsu Azuma, Kazunori Yasuda, Yoshie Tanabe, Hiroko Taniguro, Fuminori Kanaya, Atsushi Nakayama, Yong Mei Chen, Jian Ping Gong, and Yoshihito Osada. Biodegradation of high-toughness double network hydrogels as potential materials for artificial cartilage. Journal of Biomedical Materials Research Part A, 81(2):373–380, 2007.

[108] Margaret Siyawamwaya, Yahya E. Choonara, Divya Bijukumar, Pradeep Kumar, Lisa C. Du Toit, and Viness Pillay. A review: Overview of novel polyelectrolyte complexes as prospec- tive drug bioavailability enhancers. International Journal of Polymeric Materials and Poly-

123 meric Biomaterials, 64(18):955–968, 2015. doi: 10.1080/00914037.2015.1038816. URL http://dx.doi.org/10.1080/00914037.2015.1038816.

[109] Takaharu Sakiyama, Hajime Takata, Takehiro Toga, and Kazuhiro Nakanishi. ph-sensitive shrinking of a dextran sulfate/chitosan complex gel and its promotion effect on the release of polymeric substances. Journal of applied polymer science, 81(3):667–674, 2001.

[110] D Capitani, A.A De Angelis, V Crescenzi, G Masci, and A.L Segre. Nmr study of a novel chitosan-based hydrogel. Carbohydrate Polymers, 45(3):245 – 252, 2001. ISSN 0144-8617. doi: https://doi.org/10.1016/S0144-8617(00)00255-1. URL http://www. sciencedirect.com/science/article/pii/S0144861700002551.

[111] Seong Soo Kim, Young Moo Lee, and Chong Soo Cho. Synthesis and properties of semi-interpenetrating polymer networks composed of β-chitin and poly (ethylene glycol) macromer. Polymer, 36(23):4497–4501, 1995.

[112] Wei-Chiang Shen. Acid-sensitive dissociation between poly(lysine) and histamine-modified poly(glutamate) as a model for drug-releasing from carriers in endosomes. Biochim- ica et Biophysica Acta (BBA) - General Subjects, 1034(1):122 – 124, 1990. ISSN 0304-4165. doi: https://doi.org/10.1016/0304-4165(90)90163-Q. URL http://www. sciencedirect.com/science/article/pii/030441659090163Q.

[113] T.S. Anirudhan, J. Christa, and Binusreejayan. ph and magnetic field sensitive folic acid conjugated proteinpolyelectrolyte complex for the controlled and targeted delivery of 5- fluorouracil. Journal of Industrial and Engineering Chemistry, 2017. ISSN 1226-086X. doi: https://doi.org/10.1016/j.jiec.2017.08.024. URL http://www.sciencedirect. com/science/article/pii/S1226086X17304434.

[114] John M. Layman, Sean M. Ramirez, Matthew D. Green, and Timothy E. Long. Influence of polycation molecular weight on poly(2-dimethylaminoethyl methacrylate)-mediated dna

124 delivery in vitro. Biomacromolecules, 10(5):1244–1252, 2009. doi: 10.1021/bm9000124. URL http://dx.doi.org/10.1021/bm9000124. PMID: 19331402.

[115] Margreet A. Wolfert, Philip R. Dash, Ola Nazarova, David Oupicky, Leonard W. Seymour, Sarah Smart, Jiri Strohalm, and Karel Ulbrich. Polyelectrolyte vectors for gene delivery: Influence of cationic polymer on biophysical properties of complexes formed with dna. Bioconjugate Chemistry, 10(6):993–1004, 1999. doi: 10.1021/bc990025r. URL http: //dx.doi.org/10.1021/bc990025r. PMID: 10563768.

[116] Wenguang Liu, Shujun Sun, Zhiqiang Cao, Xin Zhang, Kangde Yao, William W. Lu, and K.D.K. Luk. An investigation on the physicochemical properties of chitosan/dna polyelec- trolyte complexes. Biomaterials, 26(15):2705 – 2711, 2005. ISSN 0142-9612. doi: https: //doi.org/10.1016/j.biomaterials.2004.07.038. URL http://www.sciencedirect. com/science/article/pii/S0142961204006908.

[117] David Oupicky, Manfred Ogris, Kenneth A. Howard, Philip R. Dash, Karel Ulbrich, and Leonard W. Seymour. Importance of lateral and steric stabilization of polyelectrolyte gene delivery vectors for extended systemic circulation. Molecular Therapy, 5(4):463 – 472, 2002. ISSN 1525-0016. doi: https://doi.org/10.1006/mthe.2002.0568. URL http:// www.sciencedirect.com/science/article/pii/S152500160290568X.

[118] KD Fisher, K Ulbrich, V Subr, Ch M Ward, V Mautner, D Blakey, and LW Seymour. A versatile system for receptor-mediated gene delivery permits increased entry of dna into target cells, enhanced delivery to the nucleus and elevated rates of transgene expression. Gene therapy, 7(15):1337, 2000.

[119] Stefano Burattini, Barnaby W Greenland, David Chappell, Howard M Colquhoun, and Wayne Hayes. Healable polymeric materials: a tutorial review. Chemical Society Reviews, 39(6):1973–1985, 2010.

125 [120] Nedal Y. Abu-Thabit and Abdel Salam Hamdy. Stimuli-responsive polyelectrolyte mul- tilayers for fabrication of self-healing coatings a review. Surface and Coatings Tech- nology, 303(Part B):406 – 424, 2016. ISSN 0257-8972. doi: https://doi.org/10. 1016/j.surfcoat.2015.11.020. URL http://www.sciencedirect.com/science/ article/pii/S0257897215303947. Smart Functional Coatings for corrosion pro- tection.

[121] Stephan T Dubas and Joseph B Schlenoff. Factors controlling the growth of polyelectrolyte multilayers. Macromolecules, 32(24):8153–8160, 1999.

[122] Tarek R Farhat and Joseph B Schlenoff. Ion transport and equilibria in polyelectrolyte multilayers. Langmuir, 17(4):1184–1192, 2001.

[123] Dean M DeLongchamp and Paula T Hammond. Fast ion conduction in layer-by-layer poly- mer films. Chemistry of materials, 15(5):1165–1173, 2003.

[124] Daria V. Andreeva, Ekaterina V. Skorb, and Dmitry G. Shchukin. Layer-by-layer polyelec- trolyte/inhibitor nanostructures for metal corrosion protection. ACS Applied Materials & Interfaces, 2(7):1954–1962, 2010. doi: 10.1021/am1002712. URL http://dx.doi. org/10.1021/am1002712.

[125] Ying Ren, Ruyun Lou, Xiaocen Liu, Meng Gao, Huizhen Zheng, Ting Yang, Hongguo Xie, Weiting Yu, and Xiaojun Ma. A self-healing hydrogel formation strategy via exploiting endothermic interactions between polyelectrolytes. Chem. Commun., 52:6273–6276, 2016. doi: 10.1039/C6CC02472F. URL http://dx.doi.org/10.1039/C6CC02472F.

[126] JD Mendelsohn, Ch J Barrett, VV Chan, AJ Pal, AM Mayes, and MF Rubner. Fabrication of microporous thin films from polyelectrolyte multilayers. Langmuir, 16(11):5017–5023, 2000.

126 [127] Christina S Hajicharalambous, Jenny Lichter, William T Hix, Magdalena Swierczewska, Michael F Rubner, and Padmavathy Rajagopalan. Nano-and sub-micron porous polyelec- trolyte multilayer assemblies: biomimetic surfaces for human corneal epithelial cells. Bio- materials, 30(23):4029–4036, 2009.

[128] Weina Wang, Yisheng Xu, Ang Li, Tao Li, Miaomiao Liu, Regine von Klitzing, Christo- pher K. Ober, A. Basak Kayitmazer, Li Li, and Xuhong Guo. Zinc induced polyelectrolyte coacervate bioadhesive and its transition to a self-healing hydrogel. RSC Adv., 5:66871– 66878, 2015. doi: 10.1039/C5RA11915D. URL http://dx.doi.org/10.1039/ C5RA11915D.

[129] Shuhui Liu, Guorong Gao, Ying Xiao, and Jun Fu. Tough and responsive oppositely charged nanocomposite hydrogels for use as bilayer actuators assembled through interfacial electro- static attraction. Journal of Materials Chemistry B, 4(19):3239–3246, 2016.

[130] Xue Li, Xiangbin Cai, Yongfeng Gao, and Michael J Serpe. Reversible bidirectional bending of hydrogel-based bilayer actuators. Journal of Materials Chemistry B, 5(15):2804–2812, 2017.

[131] Shengwei Xiao, Yin Yang, Mingqiang Zhong, Hong Chen, Yanxian Zhang, Jintao Yang, and Jie Zheng. Salt-responsive bilayer hydrogels with pseudo-double-network structure actuated by polyelectrolyte and antipolyelectrolyte effects. ACS applied materials & interfaces, 9 (24):20843–20851, 2017.

[132] Yan Xiang, Shanfu Lu, and San Ping Jiang. Layer-by-layer self-assembly in the develop- ment of electrochemical energy conversion and storage devices from fuel cells to superca- pacitors. Chemical Society Reviews, 41(21):7291–7321, 2012.

[133] Xiong Zhang, Wensheng Yang, and David G Evans. Layer-by-layer self-assembly of man-

127 ganese oxide nanosheets/polyethylenimine multilayer films as electrodes for supercapaci- tors. Journal of Power Sources, 184(2):695–700, 2008.

[134] Huajun Zheng, Fengqiu Tang, Yi Jia, Lianzhou Wang, Yuchun Chen, Melvin Lim, Lei Zhang, and Gaoqing Max Lu. Layer-by-layer assembly and electrochemical properties of sandwiched film of manganese oxide nanosheet and carbon nanotube. Carbon, 47(6): 1534–1542, 2009.

[135] Gyu Leem, Benjamin D. Sherman, Alex J. Burnett, Zachary A. Morseth, Kyung-Ryang Wee, John M. Papanikolas, Thomas J. Meyer, and Kirk S. Schanze. Light-driven wa- ter oxidation using polyelectrolyte layer-by-layer chromophorecatalyst assemblies. ACS Energy Letters, 1(2):339–343, 2016. doi: 10.1021/acsenergylett.6b00171. URL http: //dx.doi.org/10.1021/acsenergylett.6b00171.

[136] Anna-Maria Pappa, Sahika Inal, Kirsty Roy, Yi Zhang, Charalampos Pitsalidis, Adel Hama, Jolien Pas, George G. Malliaras, and Roisin M. Owens. Polyelectrolyte layer-by-layer as- sembly on organic electrochemical transistors. ACS Applied Materials & Interfaces, 9(12): 10427–10434, 2017. doi: 10.1021/acsami.6b15522. URL http://dx.doi.org/10. 1021/acsami.6b15522. PMID: 28263552.

[137] Yan Huang, Ming Zhong, Yang Huang, Minshen Zhu, Zengxia Pei, Zifeng Wang, Qi Xue, Xuming Xie, and Chunyi Zhi. A self-healable and highly stretchable supercapacitor based on a dual crosslinked polyelectrolyte. Nature communications, 6:10310, 2015.

[138] Dae-Hyeong Kim, Nanshu Lu, Rui Ma, Yun-Soung Kim, Rak-Hwan Kim, Shuodao Wang, Jian Wu, Sang Min Won, Hu Tao, Ahmad Islam, et al. Epidermal electronics. science, 333 (6044):838–843, 2011.

[139] Jeong-Yun Sun, Christoph Keplinger, George M Whitesides, and Zhigang Suo. Ionic skin. Advanced Materials, 26(45):7608–7614, 2014.

128 [140] Sungmee Park and Sundaresan Jayaraman. Smart textiles: Wearable electronic systems. MRS bulletin, 28(8):585–591, 2003.

[141] Liangbing Hu, Mauro Pasta, Fabio La Mantia, LiFeng Cui, Sangmoo Jeong, Heather Dawn Deshazer, Jang Wook Choi, Seung Min Han, and Yi Cui. Stretchable, porous, and conduc- tive energy textiles. Nano letters, 10(2):708–714, 2010.

[142] Tianyu Liu, Lauren Finn, Minghao Yu, Hanyu Wang, Teng Zhai, Xihong Lu, Yexiang Tong, and Yat Li. Polyaniline and polypyrrole pseudocapacitor electrodes with excellent cycling stability. Nano letters, 14(5):2522–2527, 2014.

[143] Zi-Long Wang, Rui Guo, Liang-Xin Ding, Ye-Xiang Tong, and Gao-Ren Li. Controllable template-assisted electrodeposition of single-and multi-walled nanotube arrays for electro- chemical energy storage. Scientific Reports, 3:1204, 2013.

[144] Georgia Sourkouni, Florian Voigts, Jan C Namyslo, Sebastian Dahle, Wolfgang Maus- Friedrichs, and Christos Argirusis. Interaction mechanism of hydrogen storage materials with layer-by-layer applied protective polyelectrolyte coatings. international journal of hy- drogen energy, 39(27):14834–14842, 2014.

[145] Linghui Meng, Wimonwan Klinkajon, Shannon Harkin, Pitt Supaphol, Gary E Wnek, et al. Electrospun crosslinked poly (acrylic acid) fiber constructs: towards a synthetic model of the cortical layer of nerve. Polymer International, 64(1):42–48, 2015.

[146] R Jayakumar, M Prabaharan, SV Nair, and H Tamura. Novel chitin and chitosan nanofibers in biomedical applications. Biotechnology advances, 28(1):142–150, 2010.

[147] Yiming Wang, Jie Wang, Zhenyu Yuan, Haoya Han, Tao Li, Li Li, and Xuhong Guo. Chi- tosan cross-linked poly (acrylic acid) hydrogels: Drug release control and mechanism. Col- loids and Surfaces B: Biointerfaces, 152:252–259, 2017.

129 [148] Percy Calvo-Marzal, Mark P Delaney, Jeffrey T Auletta, Tianqi Pan, Nicholas M Perri, Lisa M Weiland, David H Waldeck, William W Clark, and Tara Y Meyer. Manipulating mechanical properties with electricity: electroplastic elastomer hydrogels. ACS Macro Let- ters, 1(1):204–208, 2011.

[149] Ming Zhong, Yi-Tao Liu, and Xu-Ming Xie. Self-healable, super tough graphene oxide– poly (acrylic acid) nanocomposite hydrogels facilitated by dual cross-linking effects through dynamic ionic interactions. Journal of Materials Chemistry B, 3(19):4001–4008, 2015.

[150] Zengjiang Wei, Jie He, Tony Liang, Hyuntaek Oh, Jasmin Athas, Zhen Tong, Chaoyang Wang, and Zhihong Nie. Autonomous self-healing of poly (acrylic acid) hydrogels induced by the migration of ferric ions. Polymer Chemistry, 4(17):4601–4605, 2013.

[151] Lei Li and You-Lo Hsieh. Ultra-fine polyelectrolyte fibers from electrospinning of poly (acrylic acid). Polymer, 46(14):5133–5139, 2005.

[152] Ching-Yi Chen, Jian-Wen Wang, and Ming-Hsiun Hon. Polyion complex nanofibrous struc- ture formed by self-assembly of chitosan and poly (acrylic acid). Macromolecular Materials and Engineering, 291(2):123–127, 2006.

[153] Visith Chavasit, Carlos Kienzle-Sterzer, and J Antonio Torres. Formation and characteriza- tion of an insoluble polyelectrolyte complex: chitosan-polyacrylic acid. Polymer Bulletin, 19(3):223–230, 1988.

[154] Jun You, Shuyi Xie, Jinfeng Cao, Hao Ge, Min Xu, Lina Zhang, and Jinping Zhou. Quat- ernized chitosan/poly (acrylic acid) polyelectrolyte complex hydrogels with tough, self- recovery, and tunable mechanical properties. Macromolecules, 49(3):1049–1059, 2016.

[155] Sang Yong Nam and Young Moo Lee. Pervaporation and properties of chitosan-poly (acrylic acid) complex membranes. Journal of membrane science, 135(2):161–171, 1997.

130 [156] Chun Yang, Frank W DelRio, Hao Ma, Anouk R Killaars, Lena P Basta, Kyle A Kyburz, and Kristi S Anseth. Spatially patterned matrix elasticity directs stem cell fate. Proceedings of the National Academy of Sciences, 113(31):E4439–E4445, 2016.

[157] James P Best, Martin P Neubauer, Sameen Javed, Henk H Dam, Andreas Fery, and Frank Caruso. Mechanics of ph-responsive hydrogel capsules. Langmuir, 29(31):9814–9823, 2013.

[158] Hongchen Liu, Liduo Rong, Bijia Wang, Ruyi Xie, Xiaofeng Sui, Hong Xu, Linping Zhang, Yi Zhong, and Zhiping Mao. Facile fabrication of redox/ph dual stimuli responsive cellulose hydrogel. Carbohydrate polymers, 176:299–306, 2017.

[159] Xin Wang, Shiyao Jing, Xiumin Qiu, Shile Zhao, Yusheng Liu, and Yebang Tan. Novel bile acid sequestrant: A biodegradable hydrogel based on amphiphilic allylamine copolymer. Chemical Engineering Journal, 304:493–502, 2016.

[160] Nasim Annabi, Su Ryon Shin, Ali Tamayol, Mario Miscuglio, Mohsen Afshar Bakooshli, Alexander Assmann, Pooria Mostafalu, Jeong-Yun Sun, Suzanne Mithieux, Louis Cheung, et al. Highly elastic and conductive human-based protein hybrid hydrogels. Advanced Materials, 28(1):40–49, 2016.

[161] Marina Koulikovska, Mehrdad Rafat, Goran Petrovski, Zoltan´ Vereb,´ Saeed Akhtar, Per Fagerholm, and Neil Lagali. Enhanced regeneration of corneal tissue via a bioengineered collagen construct implanted by a nondisruptive surgical technique. Tissue Engineering Part A, 21(5-6):1116–1130, 2015.

[162] Christopher A Durst, Michael P Cuchiara, Elizabeth G Mansfield, Jennifer L West, and K Jane Grande-Allen. Flexural characterization of cell encapsulated pegda hydrogels with applications for tissue engineered heart valves. Acta biomaterialia, 7(6):2467–2476, 2011.

131 [163] Ming Zhong, Yi-Tao Liu, Xiao-Ying Liu, Fu-Kuan Shi, Li-Qin Zhang, Mei-Fang Zhu, and Xu-Ming Xie. Dually cross-linked single network poly (acrylic acid) hydrogels with supe- rior mechanical properties and water absorbency. Soft Matter, 12(24):5420–5428, 2016.

[164] John S Graham, Brannon R McCullough, Hyeran Kang, W Austin Elam, Wenxiang Cao, and M Enrique. Multi-platform compatible software for analysis of polymer bending me- chanics. PloS one, 9(4):e94766, 2014.

[165] Brannon R McCullough, Laurent Blanchoin, Jean-Louis Martiel, and M Enrique. Cofilin increases the bending flexibility of actin filaments: implications for severing and cell me- chanics. Journal of molecular biology, 381(3):550–558, 2008.

[166] Paweł Nakielski, Sylwia Pawłowska, Filippo Pierini, Wioletta Liwinska,´ Patryk Hejduk, Krzysztof Zembrzycki, Ewelina Zabost, and Tomasz A Kowalewski. Hydrogel nanofila- ments via core-shell electrospinning. PloS one, 10(6):e0129816, 2015.

[167] A Ott, M Magnasco, A Simon, and A Libchaber. Measurement of the persistence length of polymerized actin using fluorescence microscopy. Physical Review E, 48(3):R1642, 1993.

[168] Hsiao-Ping Hsu, Wolfgang Paul, and Kurt Binder. Understanding the multiple length scales describing the structure of bottle-brush polymers by monte carlo simulation meth- ods. Macromolecular Theory and Simulations, 20(7):510–525, 2011.

[169] Roberto Baigorri, Jose´ Mar´ıa Garc´ıa-Mina, and Gustavo Gonzalez-Gaitano.´ Supramolecular association induced by fe (iii) in low molecular weight sodium polyacrylate. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 292(2-3):212–216, 2007.

[170] Jingyi Nie, Zhengke Wang, and Qiaoling Hu. Chitosan hydrogel structure modulated by metal ions. Scientific reports, 6:36005, 2016.

132 [171] Umit Gulyuz and Oguz Okay. Self-healing polyacrylic acid hydrogels. Soft Matter, 9(43): 10287–10293, 2013.

[172] Umit Gulyuz and Oguz Okay. Self-healing poly (acrylic acid) hydrogels with shape memory behavior of high mechanical strength. Macromolecules, 47(19):6889–6899, 2014.

[173] Jenny A Lichter, M Todd Thompson, Maricela Delgadillo, Takehiro Nishikawa, Michael F Rubner, and Krystyn J Van Vliet. Substrata mechanical stiffness can regulate adhesion of viable bacteria. Biomacromolecules, 9(6):1571–1578, 2008.

[174] Shauheen S Soofi, Julie A Last, Sara J Liliensiek, Paul F Nealey, and Christopher J Murphy. The elastic modulus of matrigel as determined by atomic force microscopy. Journal of structural biology, 167(3):216–219, 2009.

[175] M Hoerter, A Oprea, N Barsan,ˆ and U Weimar. Chemical interaction of gaseous ammonia and water vapour with polyacrylic acid layers. Sensors and Actuators B: Chemical, 134(2): 743–749, 2008.

[176] Tae Hee Kim, Jae Soon Ahn, Hoo Kyun Choi, Yun Jaie Choi, and Chong Su Cho. A novel mucoadhesive polymer film composed of carbopol, poloxamer and hydroxypropylmethyl- cellulose. Archives of pharmacal research, 30(3):381–386, 2007.

[177] M Todica, R Stefan, CV Pop, and L Olar. Ir and raman investigation of some poly (acrylic) acid gels in aqueous and neutralized state. Acta Physica Polonica A, 128(1):128–135, 2015.

[178] Luke J Kirwan, Phillip D Fawell, and Wilhelm van Bronswijk. In situ ftir-atr examination of poly (acrylic acid) adsorbed onto hematite at low ph. Langmuir, 19(14):5802–5807, 2003.

[179] Jian Dong, Yukihiro Ozaki, and Kenichi Nakashima. Infrared, raman, and near-infrared spectroscopic evidence for the coexistence of various hydrogen-bond forms in poly (acrylic acid). Macromolecules, 30(4):1111–1117, 1997.

133 [180] Chiam-Wen Liew, HM Ng, Arshid Numan, and S Ramesh. Poly (acrylic acid)–based hybrid inorganic–organic electrolytes membrane for electrical double layer capacitors application. Polymers, 8(5):179, 2016.

[181] Eric Guibal, Thierry Vincent, and Ricardo Navarro. Metal ion biosorption on chitosan for the synthesis of advanced materials. Journal of Materials Science, 49(16):5505–5518, 2014.

[182] Peng Su, Changjun Wang, Xianyan Yang, Xiaoyi Chen, Changyou Gao, Xin-Xing Feng, Jian-Yong Chen, Juan Ye, and Zhongru Gou. Electrospinning of chitosan nanofibers: The favorable effect of metal ions. Carbohydrate Polymers, 84(1):239–246, 2011.

[183] Patrice Simon and Yury Gogotsi. Materials for electrochemical capacitors. Nature materi- als, 7(11):845–854, 2008.

[184] Martin Winter and Ralph J Brodd. What are batteries, fuel cells, and supercapacitors?, 2004.

[185] Guoping Wang, Lei Zhang, and Jiujun Zhang. A review of electrode materials for electro- chemical supercapacitors. Chemical Society Reviews, 41(2):797–828, 2012.

[186] ZHANG LI LI. Carbon-Based Materials as Supercapacitor Electrodes. PhD thesis, NA- TIONAL UNIVERSITY OF SINGAPORE, 2010.

[187] Zenan Yu, Binh Duong, Danielle Abbitt, and Jayan Thomas. Highly ordered mno2 nanopil- lars for enhanced supercapacitor performance. Advanced Materials, 25(24):3302–3306, 2013.

[188] John R Miller and Patrice Simon. Electrochemical capacitors for energy management. Sci- ence Magazine, 321(5889):651–652, 2008.

[189] Ye Hou, Yingwen Cheng, Tyler Hobson, and Jie Liu. Design and synthesis of hierarchical mno2 nanospheres/carbon nanotubes/conducting polymer ternary composite for high per- formance electrochemical electrodes. Nano letters, 10(7):2727–2733, 2010.

134 [190] Zhibin Yang, Jue Deng, Xuli Chen, Jing Ren, and Huisheng Peng. A highly stretchable, fiber-shaped supercapacitor. Angewandte Chemie International Edition, 52(50):13453– 13457, 2013.

[191] SOUMYADEB Ghosh and O Inganas.¨ Conducting polymer hydrogels as 3d electrodes: applications for supercapacitors. Advanced Materials, 11(14):1214–1218, 1999.

[192] Chang Liu, Feng Li, Lai-Peng Ma, and Hui-Ming Cheng. Advanced materials for energy storage. Advanced materials, 22(8), 2010.

[193] Jung Joon Yoo, Kaushik Balakrishnan, Jingsong Huang, Vincent Meunier, Bobby G Sumpter, Anchal Srivastava, Michelle Conway, Arava Leela Mohana Reddy, Jin Yu, Robert Vajtai, et al. Ultrathin planar graphene supercapacitors. Nano letters, 11(4):1423–1427, 2011.

[194] Yi Huang, Jiajie Liang, and Yongsheng Chen. An overview of the applications of graphene- based materials in supercapacitors. Small, 8(12):1805–1834, 2012.

[195] Long Zhang, Fan Zhang, Xi Yang, Guankui Long, Yingpeng Wu, Tengfei Zhang, Kai Leng, Yi Huang, Yanfeng Ma, Ao Yu, et al. Porous 3d graphene-based bulk materials with excep- tional high surface area and excellent conductivity for supercapacitors. Scientific reports, 3, 2013.

[196] Zenan Yu, Laurene Tetard, Lei Zhai, and Jayan Thomas. Supercapacitor electrode materials: nanostructures from 0 to 3 dimensions. Energy & Environmental Science, 8(3):702–730, 2015.

[197] Jianfeng Zang, Shu-Juan Bao, Chang Ming Li, Haijiao Bian, Xiaoqiang Cui, Qiaoliang Bao, Chang Q Sun, Jun Guo, and Keryn Lian. Well-aligned cone-shaped nanostructure of

135 polypyrrole/ruo2 and its electrochemical supercapacitor. The Journal of Physical Chemistry C, 112(38):14843–14847, 2008.

[198] Qiong Wu, Yuxi Xu, Zhiyi Yao, Anran Liu, and Gaoquan Shi. Supercapacitors based on flexible graphene/polyaniline nanofiber composite films. ACS nano, 4(4):1963–1970, 2010.

[199] Raj Kishore Sharma, Ajay Karakoti, Sudipta Seal, and Lei Zhai. Multiwall carbon nanotube- poly (4-styrenesulfonic acid) supported polypyrrole/manganese oxide nano-composites for high performance electrochemical electrodes. Journal of Power Sources, 195(4):1256– 1262, 2010.

[200] Qiang Li, Jordan M Anderson, Yiqing Chen, and Lei Zhai. Structural evolution of multi- walled carbon nanotube/mno 2 composites as supercapacitor electrodes. Electrochimica Acta, 59:548–557, 2012.

[201] Jian-Gan Wang, Ying Yang, Zheng-Hong Huang, and Feiyu Kang. Synthesis and elec- trochemical performance of mno 2/cnts–embedded carbon nanofibers nanocomposites for supercapacitors. Electrochimica Acta, 75:213–219, 2012.

[202] Mingjia Zhi, Ayyakkannu Manivannan, Fanke Meng, and Nianqiang Wu. Highly conductive electrospun carbon nanofiber/mno 2 coaxial nano-cables for high energy and power density supercapacitors. Journal of Power Sources, 208:345–353, 2012.

[203] Jian-Gan Wang, Ying Yang, Zheng-Hong Huang, and Feiyu Kang. Rational synthe- sis of mno 2/conducting polypyrrole@ carbon nanofiber triaxial nano-cables for high- performance supercapacitors. Journal of Materials Chemistry, 22(33):16943–16949, 2012.

[204] Qingxin Chu, Juan Du, Wenbo Lu, Guohui Chang, Zhicai Xing, Haiyan Li, Chen- jiao Ge, Lei Wang, Yonglan Luo, Abdullah M Asiri, et al. Synthesis of a mno2

136 nanosheet/graphene flake composite and its application as a supercapacitor having high rate capability. ChemPlusChem, 77(10):872–876, 2012.

[205] Peng Lv, Yi Y Feng, Yu Li, and Wei Feng. Carbon fabric-aligned carbon nanotube/mno 2/conducting polymers ternary composite electrodes with high utilization and mass loading of mno 2 for super-capacitors. Journal of power sources, 220:160–168, 2012.

[206] Yanfang Yan, Qilin Cheng, Vladimir Pavlinek, Petr Saha, and Chunzhong Li. Fabrication of polyaniline/mesoporous carbon/mno 2 ternary nanocomposites and their enhanced elec- trochemical performance for supercapacitors. Electrochimica Acta, 71:27–32, 2012.

[207] Haosen Fan, Ning Zhao, Hao Wang, Jian Xu, and Feng Pan. 3d conductive network-based free-standing pani–rgo–mwnts hybrid film for high-performance flexible supercapacitor. Journal of Materials Chemistry A, 2(31):12340–12347, 2014.

[208] Teng Zhai, Fuxin Wang, Minghao Yu, Shilei Xie, Chaolun Liang, Cheng Li, Fangming Xiao, Renheng Tang, Qixiu Wu, Xihong Lu, et al. 3d mno 2–graphene composites with large areal capacitance for high-performance asymmetric supercapacitors. Nanoscale, 5 (15):6790–6796, 2013.

[209] Lei Zhao, Jie Yu, Wenjun Li, Shuguang Wang, Chenglong Dai, Junwei Wu, Xuedong Bai, and Chunyi Zhi. Honeycomb porous mno 2 nanofibers assembled from radially grown nanosheets for aqueous supercapacitors with high working voltage and energy density. Nano Energy, 4:39–48, 2014.

[210] Raj Kishore Sharma and Lei Zhai. Multiwall carbon nanotube supported poly (3, 4- ethylenedioxythiophene)/manganese oxide nano-composite electrode for super-capacitors. Electrochimica Acta, 54(27):7148–7155, 2009.

137 [211] Qiang Li, Jianhua Liu, Jianhua Zou, Anindarupa Chunder, Yiqing Chen, and Lei Zhai. Syn- thesis and electrochemical performance of multi-walled carbon nanotube/polyaniline/mno 2 ternary coaxial nanostructures for supercapacitors. Journal of Power Sources, 196(1): 565–572, 2011.

[212] Zenan Yu, Matthew McInnis, Jean Calderon, Sudipta Seal, Lei Zhai, and Jayan Thomas. Functionalized graphene aerogel composites for high-performance asymmetric supercapac- itors. Nano Energy, 11:611–620, 2015.

[213] Xiaofeng Lu, Ce Wang, Fred´ eric´ Favier, and Nicola Pinna. Electrospun nanomaterials for supercapacitor electrodes: Designed architectures and electrochemical performance. Ad- vanced Energy Materials, 2016.

[214] Oh Seok Kwon, Taejoon Kim, Jun Seop Lee, Seon Joo Park, Hyun-Woo Park, Minjeong Kang, Ji Eun Lee, Jyongsik Jang, and Hyeonseok Yoon. Fabrication of graphene sheets intercalated with manganese oxide/carbon nanofibers: Toward high-capacity energy storage. Small, 9(2):248–254, 2013.

[215] Jaemoon Jun, Jun Seop Lee, Dong Hoon Shin, Sung Gun Kim, and Jyongsik Jang. Multidi- mensional mno 2 nanohair-decorated hybrid multichannel carbon nanofiber as an electrode material for high-performance supercapacitors. Nanoscale, 7(38):16026–16033, 2015.

[216] Dan Zhou, Huiming Lin, Feng Zhang, Hao Niu, Liru Cui, Qian Wang, and Fengyu Qu. Freestanding mno 2 nanoflakes/porous carbon nanofibers for high-performance flexible su- percapacitor electrodes. Electrochimica Acta, 161:427–435, 2015.

[217] Xinhua Li, Guangyong Wang, Xiaowei Wang, Xiaoping Li, and Junhui Ji. Flexible superca- pacitor based on mno 2 nanoparticles via electrospinning. Journal of Materials Chemistry A, 1(35):10103–10106, 2013.

138 [218] Y-G Wang, H-Q Li, and Y-Y Xia. Ordered whiskerlike polyaniline grown on the surface of mesoporous carbon and its electrochemical capacitance performance. Advanced Materials, 18(19):2619–2623, 2006.

[219] Chuan-Ling Zhang and Shu-Hong Yu. Nanoparticles meet electrospinning: recent advances and future prospects. Chemical Society Reviews, 43(13):4423–4448, 2014.

[220] KD McKeon-Fischer and JW Freeman. Characterization of electrospun poly (l-lactide) and gold nanoparticle composite scaffolds for skeletal muscle tissue engineering. Journal of tissue engineering and regenerative medicine, 5(7):560–568, 2011.

[221] Dian He, Bo Hu, Qiao-Feng Yao, Kan Wang, and Shu-Hong Yu. Large-scale synthesis of flexible free-standing sers substrates with high sensitivity: electrospun pva nanofibers embedded with controlled alignment of silver nanoparticles. Acs Nano, 3(12):3993–4002, 2009.

[222] Nasser AM Barakat, AM Hamza, Salem S Al-Deyab, Ahsanulhaq Qurashi, and Hak Yong Kim. Titanium-based polymeric electrospun nanofiber mats as a novel organic semiconduc- tor. Materials Science and Engineering: B, 177(1):34–42, 2012.

[223] Tong Xu, Chunping Li, Hongqiang Li, Jie Bai, Haili Qin, and Weiyan Sun. Synthesis of well-dispersed copper nanoparticles in electrospun polyacrylonitrile nanofibres. Micro & Nano Letters, 8(12):849–852, 2013.

[224] Sung Kyu Choi, Soonhyun Kim, Sang Kyoo Lim, and Hyunwoong Park. Photocatalytic comparison of tio2 nanoparticles and electrospun tio2 nanofibers: effects of mesoporosity and interparticle charge transfer. The Journal of Physical Chemistry C, 114(39):16475– 16480, 2010.

[225] Hui Chen, Anindarupa Chunder, Xiong Liu, Feroz Haque, Jianhua Zou, Lauren Austin,

139 Genevieve Knowles, Lei Zhai, and Qun Huo. A multifunctional gold nanoparti- cle/polyelectrolyte fibrous nanocomposite prepared from electrospinning process. Materials Express, 1(2):154–159, 2011.

[226] Yongming Sun, Xianluo Hu, Wei Luo, and Yunhui Huang. Porous carbon-modified mno disks prepared by a microwave-polyol process and their superior lithium-ion storage prop- erties. Journal of Materials Chemistry, 22(36):19190–19195, 2012.

[227] Bin Liu, Xianluo Hu, Henghui Xu, Wei Luo, Yongming Sun, and Yunhui Huang. Encap- sulation of mno nanocrystals in electrospun carbon nanofibers as high-performance anode materials for lithium-ion batteries. Scientific reports, 4:4229, 2014.

[228] HW Nesbitt and D Banerjee. Interpretation of xps mn (2p) spectra of mn oxyhydroxides and constraints on the mechanism of mno 2 precipitation. American Mineralogist, 83(3-4): 305–315, 1998.

[229] Mark C Biesinger, Brad P Payne, Andrew P Grosvenor, Leo WM Lau, Andrea R Gerson, and Roger St C Smart. Resolving surface chemical states in xps analysis of first row transi- tion metals, oxides and hydroxides: Cr, mn, fe, co and ni. Applied Surface Science, 257(7): 2717–2730, 2011.

[230] Mathieu Toupin, Thierry Brousse, and Daniel Belanger.´ Charge storage mechanism of mno2 electrode used in aqueous electrochemical capacitor. Chemistry of Materials, 16(16):3184– 3190, 2004.

[231] Xiaoxia Bai, Xiujie Hu, Shuyun Zhou, Jun Yan, Chenghua Sun, Ping Chen, and Laifeng Li. 3d flowerlike poly (3, 4-ethylenedioxythiophene) for high electrochemical capacitive energy storage. Electrochimica Acta, 106:219–225, 2013.

[232] Narendra Kurra, Ruiqi Wang, and Husam N Alshareef. All conducting polymer electrodes

140 for asymmetric solid-state supercapacitors. Journal of Materials Chemistry A, 3(14):7368– 7374, 2015.

[233] Yajie Yang, Wentao Yuan, Shibin Li, Xiaojie Yang, Jianhua Xu, and Yadong Jiang. Man- ganese dioxide nanoparticle enrichment in porous conducting polymer as high performance supercapacitor electrode materials. Electrochimica Acta, 165:323–329, 2015.

[234] Graeme A Snook, Pon Kao, and Adam S Best. Conducting-polymer-based supercapacitor devices and electrodes. Journal of Power Sources, 196(1):1–12, 2011.

[235] Cheng Zhou, Yangwei Zhang, Yuanyuan Li, and Jinping Liu. Construction of high- capacitance 3d coo@ polypyrrole nanowire array electrode for aqueous asymmetric super- capacitor. Nano letters, 13(5):2078–2085, 2013.

[236] Fengmei Wang, Xueying Zhan, Zhongzhou Cheng, Zhenxing Wang, Qisheng Wang, Kai Xu, Muhammad Safdar, and Jun He. Tungsten oxide@ polypyrrole core–shell nanowire arrays as novel negative electrodes for asymmetric supercapacitors. Small, 11(6):749–755, 2015.

[237] JC Dias, Daniela M Correia, Gabriela Botelho, Senentxu Lanceros-Mendez,´ and Vitor Sen- cadas. Electrical properties of intrinsically conductive core–shell polypyrrole/poly (vinyli- dene fluoride) electrospun fibers. Synthetic Metals, 197:198–203, 2014.

[238] Wei Wang, Zhenyu Li, Tingting Jiang, Zhiwei Zhao, Ye Li, Zhaojie Wang, and Ce Wang. Sulfonated poly (ether ether ketone)/polypyrrole core–shell nanofibers: a novel polymeric adsorbent/conducting polymer nanostructures for ultrasensitive gas sensors. ACS applied materials & interfaces, 4(11):6080–6084, 2012.

[239] S Eswara, C Mitterbauer, T Wirtz, S Kujawa, and JM Howe. An in situ correlative stem-

141 eds and hrtem based nanoscale chemical characterization of solid–liquid interfaces in an aluminium alloy. Journal of microscopy, 264(1):64–70, 2016.

[240] Weijia Zhou, Kai Zhou, Xiaojun Liu, Renzong Hu, Hong Liu, and Shaowei Chen. Flexible wire-like all-carbon supercapacitors based on porous core–shell carbon fibers. Journal of Materials Chemistry A, 2(20):7250–7255, 2014.

[241] Guanhua Cheng, Tianyi Kou, Jie Zhang, Conghui Si, Hui Gao, and Zhonghua Zhang. O22-/o-functionalized oxygen-deficient co3o4 nanorods as high performance supercapaci- tor electrodes and electrocatalysts towards water splitting. Nano energy, 38:155–166, 2017.

[242] Muhammad-Sadeeq Balogun, Yongchao Huang, Weitao Qiu, Hao Yang, Hongbing Ji, and Yexiang Tong. Updates on the development of nanostructured transition metal nitrides for electrochemical energy storage and water splitting. Materials Today, 2017.

[243] Seungki Hong, Sangkyu Lee, and Ungyu Paik. Core-shell tubular nanostructured electrode of hollow carbon nanofiber/manganese oxide for electrochemical capacitors. Electrochimica Acta, 141:39–44, 2014.

[244] Tao Wang, Dengfei Song, Hao Zhao, Jiayi Chen, Changhui Zhao, Lulu Chen, Wanjun Chen, Jinyuan Zhou, and Erqing Xie. Facilitated transport channels in carbon nanotube/carbon nanofiber hierarchical composites decorated with manganese dioxide for flexible superca- pacitors. Journal of Power Sources, 274:709–717, 2015.

[245] Jun Yan, Zhuangjun Fan, Wei Sun, Guoqing Ning, Tong Wei, Qiang Zhang, Rufan Zhang, Linjie Zhi, and Fei Wei. Advanced asymmetric supercapacitors based on ni (oh) 2/graphene and porous graphene electrodes with high energy density. Advanced Functional Materials, 22(12):2632–2641, 2012.

142 [246] Michael B Ujiki and Mark S Talamonti. Guidelines for the surgical management of pancre- atic adenocarcinoma. In Seminars in oncology, pages 311–320. Elsevier, 2007.

[247] Lola Rahib, Benjamin D Smith, Rhonda Aizenberg, Allison B Rosenzweig, Julie M Flesh- man, and Lynn M Matrisian. Projecting cancer incidence and deaths to 2030: the unexpected burden of thyroid, liver, and pancreas cancers in the united states. Cancer research, 74(11): 2913–2921, 2014.

[248] Joshua A McCarroll, Stephanie Naim, George Sharbeen, Julia Lee, Maria Kavallaris, David Goldstein, and Phoebe A Phillips. Role of pancreatic stellate cells in chemoresistance in pancreatic cancer. Frontiers in physiology, 5, 2014.

[249] Marvin Schober, Ralf Jesenofsky, Ralf Faissner, Cornelius Weidenauer, Wolfgang Hag- mann, Patrick Michl, Rainer L Heuchel, Stephan L Haas, J Lohr,¨ et al. Desmoplasia and chemoresistance in pancreatic cancer. Cancers, 6(4):2137–2154, 2014.

[250] Shin Hamada, Atsushi Masamune, and Tooru Shimosegawa. Novel therapeutic strategies targeting tumor-stromal interactions in pancreatic cancer. Frontiers in physiology, 4, 2013.

[251] Clifford J Whatcott, Richard G Posner, Daniel D Von Hoff, and Haiyong Han. Desmoplasia and chemoresistance in pancreatic cancer. Transworld Research Network, 2012.

[252] RL Merriman, LW Hertel, RM Schultz, Peter J Houghton, JA Houghton, PG Rutherford, LR Tanzer, GB Boder, and GB Grindey. Comparison of the antitumor activity of gemc- itabine and ara-c in a panel of human breast, colon, lung and pancreatic xenograft models. Investigational new drugs, 14(3):243–247, 1996.

[253] Patrick J Jansen, Michael J Akers, Robert M Amos, Steven W Baertschi, Gary G Cooke, Douglas E Dorman, Craig AJ Kemp, Steven R Maple, and Karen A McCune. The degra- dation of the antitumor agent gemcitabine hydrochloride in an acidic aqueous solution at ph

143 3.2 and identification of degradation products. Journal of pharmaceutical sciences, 89(7): 885–891, 2000.

[254] Jose L Arias, L Harivardhan Reddy, and Patrick Couvreur. Superior preclinical efficacy of gemcitabine developed as chitosan nanoparticulate system. Biomacromolecules, 12(1): 97–104, 2010.

[255] Anne-Laure Papa, Sudipta Basu, Poulomi Sengupta, Deboshri Banerjee, Shiladitya Sen- gupta, and Rania Harfouche. Mechanistic studies of gemcitabine-loaded nanoplatforms in resistant pancreatic cancer cells. BMC cancer, 12(1):419, 2012.

[256] David Goldstein, Robert Hassan El-Maraghi, Pascal Hammel, Volker Heinemann, Volker Kunzmann, Javier Sastre, Werner Scheithauer, Salvatore Siena, Josep Tabernero, Luis Teix- eira, et al. nab-paclitaxel plus gemcitabine for metastatic pancreatic cancer: long-term survival from a phase iii trial. Journal of the National Cancer Institute, 107(2):dju413, 2015.

[257] Dionysios Papadatos-Pastos, Kiruthikah Thillai, Roy Rabbie, Paul Ross, and Debashis Sarker. Folfirinox–a new paradigm in the treatment of pancreatic cancer. Expert review of anticancer therapy, 14(10):1115–1125, 2014.

[258] Goutam Mondal, Virender Kumar, Surendra K Shukla, Pankaj K Singh, and Ram I Mahato. Egfr-targeted polymeric mixed micelles carrying gemcitabine for treating pancreatic cancer. Biomacromolecules, 17(1):301–313, 2015.

[259] Donato Cosco, Alessandra Bulotta, Monica Ventura, Christian Celia, Teresa Calimeri, Gino Perri, Donatella Paolino, Nicola Costa, Paola Neri, Pierosandro Tagliaferri, et al. In vivo ac- tivity of gemcitabine-loaded pegylated small unilamellar liposomes against pancreatic can- cer. Cancer chemotherapy and pharmacology, 64(5):1009–1020, 2009.

144 [260] Andrei Maksimenko, Julie Mougin, Simona Mura, Eric Sliwinski, Elise Lepeltier, Claudie Bourgaux, Sinda Lepetre,ˆ Fatima Zouhiri, Didier Desmaele,¨ and Patrick Couvreur. Polyiso- prenoyl gemcitabine conjugates self assemble as nanoparticles, useful for cancer therapy. Cancer letters, 334(2):346–353, 2013.

[261] Christian Celia, Natalia Malara, Rosa Terracciano, Donato Cosco, Donatella Paolino, Massimo Fresta, and Rocco Savino. Liposomal delivery improves the growth-inhibitory and apoptotic activity of low doses of gemcitabine in multiple myeloma cancer cells. Nanomedicine: Nanotechnology, Biology and Medicine, 4(2):155–166, 2008.

[262] R Moog, A Burger, M Brandl, J Schuler,¨ R Schubert, C Unger, H Fiebig, and U Massing. Change in pharmacokinetic and pharmacodynamic behavior of gemcitabine in human tu- mor xenografts upon entrapment in vesicular phospholipid gels. Cancer chemotherapy and pharmacology, 49(5):356–366, 2002.

[263] Mingliang Fan, Xiaofei Liang, Zonghai Li, Hongyang Wang, Danbo Yang, and Bizhi Shi. Chlorambucil gemcitabine conjugate nanomedicine for cancer therapy. European Journal of Pharmaceutical Sciences, 79:20–26, 2015.

[264] Xiao-Mei Tao, Jian-cheng Wang, Jia-bao Wang, Qiang Feng, Shan-yun Gao, Liang-Ren Zhang, and Qiang Zhang. Enhanced anticancer activity of gemcitabine coupling with con- jugated linoleic acid against human breast cancer in vitro and in vivo. European Journal of Pharmaceutics and Biopharmaceutics, 82(2):401–409, 2012.

[265] Lin Yu, Guang Tao Chang, Huan Zhang, and Jian Dong Ding. Injectable block copolymer hydrogels for sustained release of a pegylated drug. International journal of pharmaceutics, 348(1):95–106, 2008.

[266] Hend Mohamed Abdel-Bar, Amal Youssef Abdel-Reheem, Rihab Osman, Gehanne AS Awad, and Nahed Mortada. Defining cisplatin incorporation properties in thermosensitive

145 injectable biodegradable hydrogel for sustained delivery and enhanced cytotoxicity. Inter- national journal of pharmaceutics, 477(1):623–630, 2014.

[267] Jung-Kyo Cho, Ji Min Hong, Taesu Han, Han-Kwang Yang, and Soo-Chang Song. In- jectable and biodegradable poly (organophosphazene) hydrogel as a delivery system of do- cetaxel for cancer treatment. Journal of drug targeting, 21(6):564–573, 2013.

[268] ChangYang Gong, QinJie Wu, YuJun Wang, DouDou Zhang, Feng Luo, Xia Zhao, YuQuan Wei, and ZhiYong Qian. A biodegradable hydrogel system containing curcumin encapsu- lated in micelles for cutaneous wound healing. Biomaterials, 34(27):6377–6387, 2013.

[269] S Mitra, U Gaur, PC Ghosh, and AN Maitra. Tumour targeted delivery of encapsulated dextran–doxorubicin conjugate using chitosan nanoparticles as carrier. Journal of Con- trolled Release, 74(1):317–323, 2001.

[270] Hua-shan Shi, Chang-yang Gong, Hai-long Zhang, Yong-sheng Wang, Jing Zhang, Zi-chao Luo, Zhi-yong Qian, Yu-quan Wei, and Li Yang. Novel vaccine adjuvant lps-hydrogel for truncated basic fibroblast growth factor to induce antitumor immunity. Carbohydrate polymers, 89(4):1101–1109, 2012.

[271] Nava Shpaisman, Larisa Sheihet, Jared Bushman, James Winters, and Joachim Kohn. One- step synthesis of biodegradable curcumin-derived hydrogels as potential soft tissue fillers after breast cancer surgery. Biomacromolecules, 13(8):2279–2286, 2012.

[272] Yongsheng Wang, Changyang Gong, Li Yang, Qinjie Wu, Shuai Shi, Huashan Shi, Zhiyong Qian, and Yuquan Wei. 5-fu-hydrogel inhibits colorectal peritoneal carcinomatosis and tumor growth in mice. BMC cancer, 10(1):402, 2010.

[273] Qinjie Wu, Guoyou Li, Senyi Deng, Liang Ouyang, Ling Li, Lei Liu, Na Luo, Xiangrong Song, Gu He, Changyang Gong, et al. Enhanced antitumor activity and mechanism of

146 biodegradable polymeric micelles-encapsulated chetomin in both transgenic zebrafish and mouse models. Nanoscale, 6(20):11940–11952, 2014.

[274] Yourui Xu, Yan Shen, Ammar Ouahab, Chang Li, Yerong Xiong, and Jiasheng Tu. Antitu- mor activity of tnf-α after intratumoral injection using an in situ thermosensitive hydrogel. Drug development and industrial pharmacy, 41(3):369–374, 2015.

[275] Bing Yang, ChangYang Gong, ZhiYong Qian, Xia Zhao, ZhengYu Li, XiaoRong Qi, Sheng- Tao Zhou, Qian Zhong, Feng Luo, and YuQuan Wei. Prevention of post-surgical abdominal adhesions by a novel biodegradable thermosensitive pece hydrogel. BMC biotechnology, 10 (1):65, 2010.

[276] Hai-Qiu Wu and Chang-Chun Wang. Biodegradable smart nanogels: a new platform for targeting drug delivery and biomedical diagnostics. Langmuir, 32(25):6211–6225, 2016.

[277] Yuling Mao, Xue Li, Ge Chen, and Shujun Wang. Thermosensitive hydrogel system with paclitaxel liposomes used in localized drug delivery system for in situ treatment of tumor: better antitumor efficacy and lower toxicity. Journal of pharmaceutical sciences, 105(1): 194–204, 2016.

[278] Jiangni Yun, Dan Guo, Yuanyuan Chen, and Zhiyong Zhang. The effect of la doping con- centration on optical and electrical properties of la x sr 1-x tio 3 thin film fabricated by sol–gel process. Thin Solid Films, 600:1–5, 2016.

[279] Xiali Zhu, Huijuan Zhang, Heqing Huang, Yingjie Zhang, Lin Hou, and Zhenzhong Zhang. Functionalized graphene oxide-based thermosensitive hydrogel for magnetic hyperthermia therapy on tumors. Nanotechnology, 26(36):365103, 2015.

[280] Nathaly Segovia, Maria Pont, Nuria Oliva, Victor Ramos, Salvador Borros,´ and Natalie

147 Artzi. Hydrogel doped with nanoparticles for local sustained release of sirna in breast can- cer. Advanced healthcare materials, 4(2):271–280, 2015.

[281] Shouyan Wang, Gwangseong Kim, Yong-Eun Koo Lee, Hoe Jin Hah, Manivannan Ethi- rajan, Ravindra K Pandey, and Raoul Kopelman. Multifunctional biodegradable polyacry- lamide nanocarriers for cancer theranostics a see and treat strategy. ACS nano, 6(8):6843– 6851, 2012.

[282] Samantha A Meenach, Chinedu G Otu, Kimberly W Anderson, and J Zach Hilt. Controlled synergistic delivery of paclitaxel and heat from poly (β-amino ester)/iron oxide-based hy- drogel nanocomposites. International journal of pharmaceutics, 427(2):177–184, 2012.

[283] Hui Liu and Changyou Gao. Preparation and properties of ionically cross-linked chitosan nanoparticles. Polymers for Advanced Technologies, 20(7):613–619, 2009.

[284] Quan Gan, Tao Wang, Colette Cochrane, and Paul McCarron. Modulation of surface charge, particle size and morphological properties of chitosan–tpp nanoparticles intended for gene delivery. Colloids and Surfaces B: Biointerfaces, 44(2):65–73, 2005.

[285] XZ Shu and KJ Zhu. A novel approach to prepare tripolyphosphate/chitosan complex beads for controlled release drug delivery. International Journal of Pharmaceutics, 201(1):51–58, 2000.

[286] Jin-Zhi Du, Tian-Meng Sun, Wen-Jing Song, Juan Wu, and Jun Wang. A tumor-acidity- activated charge-conversional nanogel as an intelligent vehicle for promoted tumoral-cell uptake and drug delivery. Angewandte Chemie, 122(21):3703–3708, 2010.

[287] David Oupicky, Manfred Ogris, Kenneth A Howard, Philip R Dash, Karel Ulbrich, and Leonard W Seymour. Importance of lateral and steric stabilization of polyelectrolyte gene delivery vectors for extended systemic circulation. Molecular Therapy, 5(4):463–472, 2002.

148 [288] Yukun Wu, Jing Wu, Jing Cao, Yajie Zhang, Zhe Xu, Xiuyi Qin, Wei Wang, and Zhi Yuan. Facile fabrication of poly (acrylic acid) coated chitosan nanoparticles with improved stabil- ity in biological environments. European Journal of Pharmaceutics and Biopharmaceutics, 112:148–154, 2017.

[289] Wei Xiong, Li Li, Yingying Wang, Yibin Yu, Shenxia Wang, Yunyun Gao, Yanyao Liang, Guogang Zhang, Weisan Pan, and Xinggang Yang. Design and evaluation of a novel poten- tial carrier for a hydrophilic antitumor drug: Auricularia auricular polysaccharide-chitosan nanoparticles as a delivery system for doxorubicin hydrochloride. International journal of pharmaceutics, 511(1):267–275, 2016.

[290] Ji Hyun Ryu, Seongyeon Jo, Mi-Young Koh, and Haeshin Lee. Bio-inspired, water-soluble to insoluble self-conversion for flexible, biocompatible, transparent, catecholamine polysac- charide thin films. Advanced Functional Materials, 24(48):7709–7716, 2014.

[291] Eun Seong Lee, Zhonggao Gao, and You Han Bae. Recent progress in tumor ph targeting nanotechnology. Journal of Controlled Release, 132(3):164–170, 2008.

[292] Donato Cosco, Donatella Paolino, Francesco De Angelis, Felisa Cilurzo, Christian Celia, Luisa Di Marzio, Diego Russo, Nicolas Tsapis, Elias Fattal, and Massimo Fresta. Aqueous- core peg-coated pla nanocapsules for an efficient entrapment of water soluble anticancer drugs and a smart therapeutic response. European Journal of Pharmaceutics and Biophar- maceutics, 89:30–39, 2015.

149