Polyaniline Based Nanofibers for Energy Storage and Conversion Devices

A Thesis

Submitted to the Faculty

Drexel University

Silas K. Simotwo

in partial fulfillment of the

requirements for the degree

Doctor of Philosophy

June 2017

iii

DEDICATIONS

I dedicate this thesis to my family. My father and my mother, Mr. Simatwa Kitoo and Mrs. Difina Simatwa

Acknowledgements

I would like to express my sincere gratitude to my academic advisor, Professor Vibha Kalra for accepting me into her lab. None of this would have been possible without her positivity, unrelenting support, guidance, motivation and patience over the last 5 years of my doctoral program. I have learnt a lot from her regarding self-drive and diligence, and she is a great example for me to follow in my future endeavors. I would also like to thank all the members of my dissertation committee, Prof. Jason Baxter, Prof. Kenneth

Lau, Prof. Caroline Schauer and Prof. Christopher Li for their insightful feedbacks on my dissertation.

I would also like to thank our collaborators: Dr. Chinnam Parameswara and Prof. Stephanie Wunder

(Temple University), Dr. Ulises Martinez and Dr. Gautam Gupta, (Los Alamos National Lab), and Prof

Kenneth Lau, Yuriy Smolin and Xiaobo Li (Drexel University). It was truly a pleasure working with them.

I would like to acknowledge Ed Basgall, Dmitri Barbash, Katie Van Aken and Muhammad Boota for helping me in conducting various characterizations that helped me in conducting my research work. I would also like to acknowledge Prof. Giusseppe Palmese for allowing me to use equipment in his lab.

I am extremely grateful to all the former and current members of Kalra research group, Dr. Chau Tran, Dr.

Richa Singhal, Dr. Arvinder Singh, Caitlin Dillard, Rahul Pai, Chris Delre, Daniel Lawrence, Alda

Kapllani, Ayda Raffie and Amir Hegazy for their friendship, help, and insightful discussions. I am also thankful to the staff members, Andrea, Jennifer for being so nice, and helping me out whenever I needed them.

Most importantly, I wish to acknowledge my parents, my sisters and brothers, and all my friends for their patience, love, and endless support. I could not have done this without them.

TABLE OF CONTENTS

List of Tables ...... ix List of Figures ...... x Abstract ...... xiv vi

CHAPTER 1. INTRODUCTION...... 1

1.1 Energy Storage and Conversion Devices...... 1 1.2 Traditional Capacitors ...... 3 1.3 Supercapacitors ...... 4 1.3.1 Categories of supercapacitors ...... 5 1.3.1.1 Electrochemical double layer supercapacitors ...... 5 1.3.1.2 Pseudocapacitors ...... 7 1.3.2 Electrode materials for supercapacitor ...... 8 1.3.2.1 Carbon ...... 8 1.3.2.2 Pseudocapacitive/redox Materials ...... 8 1.3.3 Electrolyte for supercapacitors ...... 9 1.4 Fuel Cells...... 11 1.5 Electrospinning...... 13 1.6 Polyaniline ...... 14 1.6.1 Literature review on polyaniline ...... 15 1.6.1.1 PANI-based Composites as Supercapacitor Cathodes ...... 15 1.6.1.2 Pure PANI based Supercapacitor ...... 16 1.6.1.3 Electrolytes for PANI based supercapacitors ...... 17 1.7 Summary on polyaniline based electrochemical devices...... 18 1.8 Objectives ...... 19

CHAPTER 2. Electrospun Polyaniline for Supercapacitors ...... 20

2.1 Introduction ...... 20 2.2 Experimental Methods and Materials ...... 22 2.2.1 Fabrication of PANI and PANI-CNT nanofibers...... 22 2.2.2 Electrochemical testing and morphological characterization...... 23 2.3 Results and Discussion ...... 25 2.3.1 Nanofibers characterization...... 25 2.4 Conclusion ...... 36

CHAPTER 3. Asymmetric Supercapacitor Based on Polyaniline-Coated Carbon Nanofibers 37

3.1 Introduction ...... 37 vii

3.2 Experimental techniques ...... 40 3.2.1 Electrode preparations...... 40 3.2.2 Spectroscopic Characterization ...... 41 3.3 Results and Discussion ...... 42 3.3.1 Three-electrode tests: ...... 43 Full Cell: Asymmetric supercapacitor ...... 46 3.3.1.2 In Situ infrared spectroscopy: ...... 49 3.4 Conclusion...... 52

CHAPTER 4. Redox Mechanisms of Polyaniline in Protic Ionic Liquid ...... 53

4.1 Introduction ...... 53 4.2 Experiment Section...... 55 4.2.1 Synthesis of PILs...... 55 4.2.2 Synthesis of polyaniline...... 56 4.2.3 Characterization techniques...... 56 4.3 Results and Discussions...... 58 4.4 Conclusion...... 61

CHAPTER 5. Co-electrospun Nafion and Polyaniline for Fuel Cell Electrodes Application 62

5.1 Introduction ...... 62 5.2 Experimental Method ...... 64 5.2.1 Nanofiber Fabrication ...... 65 5.2.2 Nanofiber Characterization ...... 65 5.2.3 Water uptake and Conductivity ...... 66 5.3 Results and discussion ...... 67 5.3.1 Morphology of electrospun nanofibers ...... 67 5.3.2 XRD and FTIR characterization ...... 72 5.3.3 Impedance study ...... 75 5.4 Conclusion ...... 80

CHAPTER 6. Summary and Future Outlook ...... 81

6.1 Summary ...... 81 6.1 Future Outloook ...... 83 viii

LIST OF REFERENCES ...... 85 Appendix ...... 107 VITA ...... 112

List of Tables

Table 1-1. Polyaniline-based supercapacitors — Fabrication techniques, electroactivity (per gram of total electrode) at low and high current densities (CD), and cycle life...... 18

Table 2-1. Specific capacitance (SC) of PANI93/PANI-CNT electrodes at different scan rates. 30

Table 3-1. Performance comparison of A-PCNF negative electrodes (anode) to those in literature in neutral electrolytes. The A-PCNFs reported in this work provide competitive performance while enabling much higher electrode loading...... 45

Table 3-2 . The table below compares performance of PANI-PCNFs electrodes to those in literature. There is limited work reporting electroactivity of PANI based electrodes in neutral aqueous electrolytes. The PANI-PCNF reported herein shows competitive performance relative to PANI tested in both acidic and neutral media, particularly when the weight of binders is also considered...... 46

Table 3-3. The table below compares performance of A-PCNFs||Na2SO4||PANI-PCNFs supercapacitor to those in literature. The cell exhibit competitive electrochemical performance compared to similar work in literature with added advantage of lack of binders and eco-friendly electrolyte (compared to acid ones) ...... 50

Table 5-1 Conductivity and water uptake of electrospun nanofiber mats...... 78

List of Figures

Figure 1.1. Trend of energy consumption in the US in 2016. Source- U.S Energy Information

Administration, Monthly Energy Review, April 2017...... 2

Figure 1.2. Ragone plot showing the relationship between specific energy and power densities of various EEDs...... 3

Figure 1.3. Traditional capacitor composed of parallel plates and a dielectric material...... 4

Figure 1.4. Electrochemical double layer supercapacitor schematic...... 6

Figure 1.5 .Schematic showing structure of a pseudocapacitor. Copyright of J. Mater. Chem. A,

2014, 2, 10776-10787...... 7

Figure 1.6. Pourbaix diagram of water...... 10

Figure 1.7. Schematic showing structure of PEM fuel cell...... 12

Figure 1.8. Schematic of a typical electrospinning set up. Copyright @ J. Electrochem. Soc

162 (10) B275-B281 (2015)...... 14

Figure 1.9. Illustration showing the different oxidation states of polyaniline. Copyright@

Current Appl. Phys. 4 (2004) 389-393...... 15

Figure 2.1 SEM images of PANI93 (a) and PANI-CNT (b) electrospun fibers with an average nanofiber diameter of 678±54nm and 528±50nm, respectively, TEM images of PANI-CNT nanofiber showing distribution of CNTs (c,d )...... 26

Figure 2.2. XRD patterns (a) and FT-IR spectra (b) of electrospun nanofibers and pristine CNTs.

...... 27

Figure 2.3. Nyquist plots of electrospun PANI93 and PANI-CNT nanofiber mats (a) and CV curves (b) of PANI93 (red) and PANI-CNT (blue) nanofiber mats at 50 mVs-1...... 28 xi

Figure 2.4. Cyclic voltammetry of (a) PANI93 and (b) PANI-CNT at different scan rates, cyclic charge-discharge curves of PANI93 (c) and PANI-CNT (d) and CVs of symmetric PANI-CNT capacitor at different scan rates (e)...... 29

Figure 2.5. (a) Pore size distribution and (b) cumulative surface area as a function of pore size for PANI93 nanofibers...... 32

Figure 2.6. Plots showing electrochemical performance of symmetric PANI-CNT capacitor: CV curves at different scan rates (a) and specific capacitance as a function of charge−discharge rates

(b). The specific capacitance is based on the total mass of both electrodes...... 33

Figure 2.7. SEM micrographs showing post-mortem electrode morphologies of PANI93 (a) and

PANI-CNT (b); life cycle performance of PANI93, PANI-CNT, and symmetric PANI-CNT capacitors (c) and EIS plot of PANI-CNT before and after cycling for 1000 cycles (d). Scale bar is 2 μm, and inset scale bar is 200 nm for (a, b)...... 35

Figure 3.1 : SEM micrographs of PCNFs (a), PANI-coated PCNFs (b) and KOH-activated A-

PCNFs. Figure C shows the pores distribution of A-PCNFs...... 43

Figure 3.2: CVs for A-PCNFs (a) and PANI-coated PCNFs in a 3-electrode set up ...... 44

Figure 3.3: CV of ASC cell at different voltage windows (a), Energy density vs voltage window,

CVs obtained at 1.6V window at different scan rates (c), and CVs obtained at 1.8V window at different scan rates...... 47

Figure 3.4. Charge-discharge plot at 1 A/g (a) and energy density vs current density (b) plots for

ASC at voltage windows of 0—1.6 and 0—1.8V windows...... 48

Figure 3.5. Illustrations for the situ FTIR spectra of PANI-CNFs as function of voltage applied during oxidation...... 50

Figure 3.6. Plots showing stability of the ASC cell (a) and EIS before and after cycling (b) ..... 52 xii

Figure 4.1. Schematic showing the synthesis process for the PILs...... 56

Figure 4.2. Schematic showing three-electrode electrochemical cell...... 57

Figure 4.3. TGA (a) and CV (b) plots showing thermal and electrochemical stability of the PILs.

...... 58

Figure 4.4. CV showing the electrochemical activity of polyaniline in protic IL electrolyte at different scan rates and corresponding cycling performance over 500 cycles...... 59

Figure 4.5. The electrochemical mechanism of PANI deposited on PCFs was categorized similarly...... 61

Figure 5.1. Schematic showing the dual electrospinning set up...... 64

Figure 5.2. SEM micrograph of as electrospun Nafion-98 (a) and AN/C Nafion-98 (b) ...... 67

Figure 5.3. SEM micrograph of as electrospun PANI-90...... 69

Figure 5.4. SEM micrograph of as electrospun Nafion-PANI (a) and AN Nafion-PANI (b) AN/C

Nafion-PANI (c) and AN/C/T Nafion-PANI (d)...... 70

Figure 5.5. SEM image for dual Nafion-PANI nanofiber mat (top) and corresponding EDS mapping showing Nafion fluorine distribution (F) and polyaniline nitrogen distribution (N). Top

SEM images shows...... 71

Figure 5.6. XRD spectrum of Nafion-212 membrane and Nafion-98 nanofiber samples (a), de- convoluted peaks for Nafion-212 membrane (b) and AN Nafion-98 nanofibers (5 mins) (c).

Peaks were fitted using Pearson VII distribution function in OriginPro software...... 72

Figure 5.7. XRD PANI and dual Nafion-PANI nanofibers ...... 73

Figure 5.8. FTIR spectra of Nafion, PANI and dual Nafion-PANI nanofibers...... 74

Figure 5.9. Nyquist plot of co-electrospun Nafion-PANI and fitted data...... 76 xiii

Figure 5.10. Equivalent circuit used to decouple electrical conductivity from proton conductivity...... 79

xiv

Abstract

Polyaniline Based Nanofibers for Energy Storage and Conversion Devices

Silas K. Simotwo, Advisor: Prof. Vibha Kalra.

Electrochemical energy devices (EEDs) such as batteries, supercapacitors and fuel cells are promising energy alternative. However, commercialization of these devices is hampered by unsatisfactory performance and high production cost. Performance of the ES&C systems is typically quantified by usable energy density, power delivery and durability. The development of higher energy and power density EEDs, in turn, significantly depends upon the advancement in technology of materials used in these devices.

Organic materials such as conducting polymers are excellent candidates for application in EEDs because of their natural abundance and benign environmental effect.1 My dissertation focuses on preparation of polyaniline (conducting polymer) based nanofibers through electrospinning method, and establishing understanding of the property-structure-performance correlation of such nanofibers towards developing high performance electrochemical devices. Polyaniline is attractive for application in electrochemical devices due to its excellent electrical conductivity, high theoretical capacitance (750 F g-1)2 and tunable synthesis. The electrospinning is a simple and versatile fiber formation technique using a strong electric field to extrude polymer solution, forming ultrathin fibers with diameters in the range of 50–800 nm.3

Supercapacitors are generally characterized by fast charge/discharge kinetics, high power density and long life-cycle performance.4 However, large scale application of supercapacitors is limited by their low energy density (typically <10 W h kg-1 vs > 100 W h kg-1 for batteries). In this work I have sought to improve the energy density by: (i) developing self-standing, 3-D polyaniline/polyaniline-carbon nanofiber based electrodes and (ii) investigating electroactivity of the polyaniline electrodes in electrolyte with wider electrochemical voltage window. Electrochemical performance of freestanding high-purity electrospun polyaniline nanofibers was investigated in 0.5 M H2SO4. The nanofibers showed competitive xv electrochemical performance, including high gravimetric capacitance (385 F g-1) and good life cycle performance (~81% capacitance retention over 1000 cycles). The capacitance obtained represents ~40% increase relative to previously reported values for similarly prepared PANI electrodes.5 The promising electrochemical performance stems from its porous three-dimensional nonwoven nanofiber mat morphology. The inter-/intra-fiber porosity and the interconnected nanofiber network facilitates shuttling of charges to the electrode/electrolyte interface. The standard electrode assembly process—utilization of slurries—in most literature is likely to result in a lack of such an open, through-connected pore structure.

The voltage window of the acid aqueous electrolytes is thermodynamically limited to 1.0V due to water decomposition. To expand the voltage window, electrochemical performance of polyaniline in neutral aqueous electrolyte (Li2SO4) and protic ionic liquid was investigated. The neutral electrolyte is electrochemically stable than the acid based due to strong solvation energy involved for electrolyte species.

We developed an asymmetric supercapacitor (ASC) utilizing Li2SO4 based electrolyte, PANI-carbon cathode and activated carbon nanofibers as the anode. Due to the wide voltage window (1.8V) achieved in such system, a specific energy of 24 W h kg-1 was obtained with capacitance retention of ~75% after 4500 cycles. In Situ FTIR spectroelectrochemical study confirmed that polyaniline was undergoing redox process in Li2SO4 similar to that observed in acidic media. Redox activity of polyaniline was also investigated in protic ionic liquid with stable voltage window of ~2.7 V using both electrochemical and ex-situ FTIR spectroscopic techniques. Cyclic voltammetry plots (3-electrode set-up) showed that polyaniline was undergoing electrochemical mechanisms similar to those observed prior in acidic media. The promising electrochemical activity of polyaniline in neutral aqueous and protic liquids is crucial for development of advanced polyaniline based supercapacitors with high energy density.

A novel 3-D porous electrode architecture characterized by co-continuous pathways for electron- and proton-conductivity was developed for fuel cell application. This was possible through co-electrospinning of polyaniline (electron conductor) and Nafion (proton conductor). Conductivity of the dual nanofibers mat was investigated using electrochemical impedance spectroscopy and equivalent circuit was used to xvi decouple impedance from electron and proton transfer in the composite fiber mat. Such dual nanofiber mats with percolating phases of both electron- and proton-conductors can potentially serve as excellent catalyst supports for fuel cells.

xvii

CHAPTER 1. INTRODUCTION

1.1 Energy Storage and Conversion Devices.

Issues related to energy represent some of the more critical challenges facing humanity. The fossil based

fuels, which account for ~90% of the total energy consumed annually, face imminent exhaustion due high

demand caused by the ever increasing population and proliferation of electronic devices. Moreover, the

fossil based energy pose a great deal of environmental concern due to emissions of greenhouses gases such

as CO2. Therefore, developing sustainable energy reservoirs is of utmost urgency. Electrochemical energy

devices (EED) such as batteries, supercapacitors and fuel cells are promising sustainable energy sources.

These devices can be utilized in transportation to power electric vehicles and therefore minimize

overreliance on gasoline.6 Moreover, high energy density EEDs can be used for energy back-up and to

complement other renewable energy sources such as wind and solar whose power/energy production tend

be intermittent in nature. Large scale commercialization of EEDs is, however, greatly hampered by low

performance and high production cost. The performance of the EEDs systems is generally measured by

energy storage capacity/power density and durability,7 and therefore developing EEDs that are competitive

economically and performance-wise depends on advancement of these parameters. Figure 1.1 below shows

energy consumption trend in the US in the year 2016. The renewable energy sources only accounts for

~10% of the total energy consumed.

Figure 1.1. Trend of energy consumption in the US in 2016. Copyright@ U.S Energy Information Administration, Monthly Energy Review, January 2017.

Supercapacitors and fuel cells are promising alternative energy prospects which can complement each other. Supercapacitors possess power capabilities that far exceeds that of batteries but they energy density is low.8 If they are to make maximum impact, new generation supercapacitors characterized by high energy density are required. On the other hand, fuel cells are energy conversion devices which are characterized by high energy density (over 100 W h kg-1) but low power density and short life cycle. Fuel cells are characterized by higher energy efficiency compared to internal combustion engine (ICE) (50-60% vs

~25%). Fuel cells are also promising as clean and sustainable alternative to ICEs, however the high production cost limits their commercialization. Figure 1.2 below shows Ragone plot, which compares the energy density (range) and power density (power delivery) of various electrochemical devices.

Figure 1.2. Ragone plot showing the relationship between specific energy and power densities of various EEDs. Copyright@ J. Power Sources 91, 210 (2000).

1.2 Traditional Capacitors

Figure 1.3. Traditional capacitor composed of parallel plates and a dielectric material.

Conventional capacitors are made of two parallel metals electrodes separated by a dielectric materials as

illustrated in figure 1.3 above. The dielectric is made of an insulating material. Some of the materials

traditionally used as dielectrics include ceramics, polystyrene, mica and oxides of various metals.9 During

charging, excess and deficiency of charges occurs at the two electrode plates which are separated by the

polarized dielectric medium. The separation of charges leads to build up of electric field between the plates

which allow for the capacitor to store electric charge.10 The capacitance in such system is given by;

푄 퐶 = (1) 푉

Where V is the applied voltage and Q is the charge developed. The operating voltage of the dielectric

capacitor depends upon the strength of the dielectric materials. The capacitance, C, of the dielectric

capacitor depends upon geometric area (A) and the distance between the plates (d) as shown in equation 2

below.

ɛ 퐴 퐶 = (2) 푑

1.3 Supercapacitors

Energy can be stored in electrochemical devices via two main processes; through electrostatic

accumulations of charges at positively or negatively charged electrodes and through electrochemical

reactions at the electrode surface (also referred to as Faradaic process). Both of these modes of energy

storage are usually more efficient than fuel combustions which are limited by limited by the

thermodynamics of Carnot Engine.11 Conventional dielectric capacitors relies solely on accumulations of

charges on charged plates. The conventional capacitors are characterized by very fast charge-discharge

processes but very limited energy storage capacity (<1 W h kg-1). Batteries, on the hand, store energy

primarily through electrochemical reactions at the electrodes surface.12 Batteries are characterized by very

high energy density (50-150 W h kg-1) but possess very low power density and life-cycle (1000 cycles) 5 performance. Supercapacitors are intermediate between traditional capacitors and batteries in terms of energy storage mechanisms and energy density. Supercapacitors are characterized by high power densities

(>1000 W kg-1) and long life cycles (~100 000).13 However, relative to batteries, supercapacitors still possess low energy densities (1-10 W h kg-1).14 Ragone plot shown in figure. 2 above compares the energy/power relations of different electrochemical devices. The high power delivery capability of supercapacitors makes them suitable for applications such as accelerations and regenerative braking in automobiles, back-up power and in electronic devices.

1.3.1 Categories of supercapacitors

A supercapacitor is composed of two porous electrodes, a separator and a liquid or gel based electrolyte.15

Supercapacitors are broadly classified into two different groups depending on the mechanisms of energy storage: electrochemical double layer capacitors (EDLCs), and pseudocapacitors.16

1.3.1.1 Electrochemical double layer supercapacitors

For EDLCs, the charge storage mechanisms involve reversible electrostatic adsorption/desorption of electrolytes ions at the electrode/electrolyte interface. Figure 1.4 below represents a supercapacitor in a charged state 6

Figure 1.4. Electrochemical double layer supercapacitor schematic.

During the charging processes, anions from the electrolytes are drawn to the positively charged electrode while cations are drawn to the negatively charged plate. When a potential is applied between the supercapacitor electrodes, excess/deficiency of electrons develop at the electrodes. Consequently, counter ions in the electrolytes diffuse to the respective electrolytes. This leads to the formation of an electric double layer (EDL) in each electrode/electrolyte interface.10 The overall EDL thickness depends on the concentration of electrolyte and is typically less than 0.10 nm. The EDL is classified into two: Helmholtz and diffuse layer. During discharge process, ions are desorbed from the electrodes and the energy dissipated is given by:

1 퐸 = 퐶푉2 (3) 2 7

Where E is the energy stored and V is the applied voltage. V is dependent on the electrolyte used (see section 1.4.2) where C is depend on total surface area of the electrodes.

1.3.1.2 Pseudocapacitors

The charge storage mechanisms in a pseudocapacitor is dominated by reversible Faradic reactions

(intercalation/de-intercalation or doping/de-doping mechanisms) involving the electrode materials and electrolyte species/ions.17 When a negative potential is applied to the electrode, the energy of an electron is elevated and the electron is transferred into the unoccupied. Consequently, an electron flow from the electrode to the electrolyte. Similarly, the energy of an electron is lowered when a positive potential is applied to the electrode. Once any occupied state of the ionic species fall into the range of unoccupied states of the electrode, electron transfer may take place from the solution to electrode. The capacitance due to such oxidation and reduction reactions is termed as ‘pseudocapacitance’ or ‘redox capacitance’. Figure 1.5 illustrates a schematic for a pseudocapacitor.

Figure 1.5 .Schematic showing structure of a pseudocapacitor. Copyright of J. Mater. Chem. A, 2014, 2, 10776-10787. 8

1.3.2 Electrode materials for supercapacitor

1.3.2.1 Carbon

Various carbon based nanomaterials have been extensively studied recent for applications in EDLCs.

Carbon nanomaterials are highly favored due to their high mechanical resiliency, excellent electrical conductivity, high thermal/electrochemical stability and benign environmental effect.4, 18 Promising electrochemical performance have been reported for activated carbon , activated/porous graphene 19-20, carbon nanotubes 21-22, carbon nanotubes 23, carbon aerogel24, and activated carbon nanofibers 25-26.

Moreover, carbon nanomaterials characterized by hierarchical porosity have also been developed from sources such as biomass materials.27-29 The hierarchical porous architecture of such nanomaterials provide excellent interfacial transport and distribution of active sites at different length of scale of pores. The micropores acts as ions reservoirs contributing to high energy storage capacity whereas the mesopores provide excellent pathways for charge transport leading to excellent power handling capability.

1.3.2.2 Pseudocapacitive/redox Materials

There has been a concerted effort recently to develop supercapacitor electrodes based on redox materials.30 The interest is driven by the direct correlation between the energy stored by a supercapacitor and its capacitance as illustrated previously in equation 3 below. Capacitance typically obtained in high surface area supercapacitors is typically less than 300 F/g when appropriate electrode mass loading is used.15 On the other hand, capacitance originating from the redox-active materials is much larger (5-10 times) than the double layer capacitance. Redox active materials not only store charges in electrical double layer but also stimulate capacitance owing to the fast and reversible redox reactions with the electrolyte ions at the electrode/electrolyte interface.31 Among the most promising redox electrode materials are the conducting polymers and electroactive transition metal oxides. Conducting polymers (CPs) have received significant attention owing to their large capacitance, eco-friendliness, good electrical conductivity, low cost and ease of synthesis.32 During the oxidation process, electrolyte ions are doped/ transferred to the 9 polymer while, during reduction the ions are de-doped/released back into the electrolyte solution from the polymer backbone. The most commonly used conducting polymers include polyaniline, polypyrrole (PPy), polyy[3,4-ethylenedioxythiophene] (PEDOT). 33-34 (polypyrrole, ref) However, large volume change (i.e., contraction/expansion) during cycling of such polymers leads to mechanical instability and capacitance loss.35-36 Another group that has been heavily studied recently is the transitional metal compounds

(TMCs).37 TMCs are characterized by large theoretical capacitance attributed to reversible redox

38 39 40 41 reactions. Some of the widely investigated transition metal oxides include RuO2, MnO2, , MoS2, ,

42 37 V2O5, and Fe based oxides. Despite the excellent theoretical capacitance of the transitional metal compounds, their applications in commercial supercapacitors is hampered by their low electronic conductivity, high cost and structural instability.37 The low electrical conductivity leads to poor rate capability, and therefore poor power delivery performance. The repeated volume changes during cycling cause degradation of nanostructures and agglomeration of nanoparticles thus impeding electrolyte penetration and ion transport. This results in large capacitance fade during cycling.

To alleviate limitations of the redox materials, namely poor rate capability and life-cycle performance, composites of redox materials with porous carbons have been studied. A synergistic combination of excellent electrical conductivity, high surface area/pore volume and mechanical resilience of carbonaceous backbone and high pseudocapacitance of redox material is realized in such hybrid electrodes.43

1.3.3 Electrolyte for supercapacitors

Electrolytes are very critical as far as the energy density of the supercapacitor is concerned. The ideal characteristics of an electrolyte include high ionic conductivity, low viscosity, low toxicity and low cost. 44

Electrolytes commonly employed in supercapacitors are generally categorized into two: aqueous and non- aqueous.4 In aqueous media, the practical working voltage window is typically ~1.0V. In many cases, aqueous electrolytes are usually studied for use in supercapacitors to take advantages of their low cost, high

45 ionic conductivity, low environmental concern and better safety. Generally H2SO4 and KOH are the 10 commonly used aqueous electrolytes.46-47 However, commercial application of the aqueous electrolytes is typically limited by the moderate voltage window. The voltage range of supercapacitors operating with aqueous electrolytes is typically limited thermodynamically to ~1.0V due to water decomposition at 1.23V.

Figure 5 below illustrates the pourbaix diagram of water. Water is decomposed into hydrogen (at lower potential) and oxygen (at higher potential). Neutral aqueous electrolyte such as Na2SO4 and Li2SO4 which has been previously shown (in carbon based supercapacitor) to possess high electrochemical stability.42, 44

The large stable electrochemical window neutral aqueous electrolyte is attributed to strong solvation energy of ionic species which leads to high overpotentials for H2/O2 evolution.

Figure 1.6. Pourbaix diagram of water.

Non-aqueous electrolytes (organic & ionic liquids) are promising for development of advanced, high energy supercapacitors. Organic electrolytes are characterized by voltage range of 2.5-2.7V.4 However, they are environmentally unfriendly, and possess low ionic conductivity which leads to low capacitance values of

~100 F/g.4 Ionic liquids are attractive due to their wider electrochemical window (2.5-4V), low volatility, absence of solvents and high thermal stability.48-49 Ionic liquids are classified into aprotic and protic ionic liquids.50-51 Aprotic ionic liquids have been widely studied for supercapacitors applications while protic has mostly been studied for potential application of in fuel cells. Protic ionic liquids are promising for 11

applications in redox systems that requires protons for their reactions. We have previously demonstrated

applications of ionic liquids in our lab which showed gravimetric capacitance of 80 W h kg-1 (comparable

to that of lead acid batteries) and high rate performance.52 Although performance of ionic liquids and

corresponding gel based electrolytes has been promising, their widespread application is limited by high

cost and high viscosity which limits high rate performance in electrodes that lack well defined pathways

for charge transfer (such as activated carbon which is commercially used).53

1.4 Fuel Cells

A proton exchange membrane fuel cell (PEMFC) is structurally composed of current collectors (anode and

cathode), connecting wires and membrane electrode assembly (MEA).The five layer MEA is at the heart

of the PEMFC and is consists of a proton exchange membrane, two catalyst layers, and two gas diffusion

layers. The most commonly used electrolyte material the proton conductive perfluorinated sulfonic acid

membrane—such as Nafion.54 The hydrogen that is supplied is oxidized at the anode while oxygen fed is

reduced at the cathode. The protons from reduction of hydrogen are conducted to the anode via the polymer

electrolyte. Since the solid electrolyte is insulating to electrons flow, the electrons bypass the electrolyte

via a wire connecting the electrodes. A longitudinal schematic of a fuel cell is represented herein in figure

1.7 below. 12

Figure 1.7. Schematic showing structure of PEM fuel cell.

Anode electrochemical reactions:

+ - H2  H + 2 e 4

Cathode electrochemical reactions:

+ - O2 + 4H + 4e  2 H2O 5

Fuel cells exhibit remarkable efficiency, clean emissions and high power density, which makes them a pertinent alternative power source for applications such as electric vehicles, small portable electronics and distributed power generators.55 Fuel cells, however, face major setbacks due to the high cost of platinum

(Pt) catalyst employed for electrochemical reactions.56 Consequently, to increase cost competitiveness of fuel cells, studies have focused on reduction of Pt loading, development of Pt alloys and replacement of Pt with non-precious metal catalysts.24, 57 To effectively reduce platinum loading, one of the key strategies to reduce Pt cost, there is a need to develop electrode architectures that will facilitate efficient transport of protons, electrons, and reactant gases to the catalyst active sites for enhanced catalyst (Pt) utilization. 13

1.5 Electrospinning

Electrospinning is a polymer extrusion technique using electric field forces The technique allows

fabrication of continuous fibers characterized by porous, now-woven interconnected nanofibers.58 The

method can be used to generate nanofibers of polymers and polymer composites with additives such as

metal precursors, carbon nanoparticles, ceramaics or active agents.59-60 The method can be used to generate

nanofibers with various architectures such as core-shell or hollow nanofibers. The electrospinning setup

consists of a syringe pump, spinneret (metallic), high voltage source, and a grounded substrate collector

in a humidity-controlled chamber.59 In a typical electrospinning experiment, a polymer solution is pumped

through a thin nozzle and, simultaneously, a potential is applied between the nozzle and the substrate

collector. Due to the applied voltage, the polymer solution/melt is deformed into a cone-shaped drop

(referred to as Taylor cone) in the direction of the substrate collector.3 When further voltage is applied, a

jet is formed which is drawn towards the current collector. The polymer jet undergoes several bending

instabilities whereby its diameter decreases and a major portion of the solvent evaporates as it travels toward

the collector.61 The quality and size of the electrospun nanofiber is controlled by the polymer solution

properties (viscosity, type of solvent, polymer molecular weight) and the processing parameters (applied

voltage, distance between spinneret and collector, humidity and temperature). The collected nanofiber mat

is typically freestanding and can be integrated into devices without further processing. For production of

carbon based nanomaterials, the electrospun precursors (polyacrylonitrile, cellulose) are subjected to post-

electrospinning heat treatment in both air (<300◦C) and inert (>500◦C). Figure 1.8 below represents a typical

electrospinning set up.

Figure 1.8. Schematic of a typical electrospinning set up. Copyright @ J. Electrochem. Soc 162 (10) B275-B281 (2015).

1.6 Polyaniline

Polyaniline (PANI) is a readily available p-type conducting polymer which has commanded considerable

attention for application in electrodes of most energy devices either as a sole electrode active material or as

a supplement material to enhance performance. PANI is commonly synthesized through chemical oxidative

polymerization of aniline using a strong oxidant. Electrochemical synthesis of polyaniline has also been

widely applied. PANI can exist in three different oxidation states: leucoemaraldine, emaraldine base is the

partially-oxidized and pernigraniline which are the fully reduced, partially oxidized and fully oxidized

states, respectively.62 Synthesis of PANI is largely carried out in acidic consditions, which result in a

protonated/doped emeraldine salt. External doping or de-doping of the emeraldine salt has also been

achieved using an acid or base, respectively.63 Conductivity of PANI is strongly dependent on the degree

of oxidation and the doping. Doped emeraldine base is the most conductive state, with conductivity ranging

from 0.1 to over 100 S/cm.64 Un-doped emaraldine and the fully reduced/oxidized states show conductivity

in the order of 10-10 S/cm. PANI is typically prepared either via a template-assisted method where an

external soluble or insoluble substrate with pre-formed nanoscale features such as a metal oxide is employed 15 or template-free technique where material’s ability to self-assemble into various morphologies depending on the synthesis conditions (pH, temperature, oxidant type, nature of dopants, etc.), is exploited. 65 66. Figure

1.9 below represents the different oxidation state of polyaniline which can be accessed via either chemical or electrochemical process.67

Figure 1.9. Illustration showing the different oxidation states of polyaniline.

1.6.1 Literature review on polyaniline

1.6.1.1 PANI-based Composites as Supercapacitor Cathodes

Bulk of the PANI supercapacitor research so far has focused on nanocomposites of PANI with carbon and/or inorganic compounds to reinforce its electrical conductivity and/or mechanical stability to improve rate capability and cycle life performance. PANI/carbon composites have particularly been extensively studied since carbon is a readily available material with excellent electrical conductivity and mechanical properties.4 A synergistic combination of excellent electrical conductivity, high surface area/pore volume 16 and mechanical resilience of carbonaceous backbone and high pseudocapacitance of PANI is realized in such hybrid electrodes.43 The resultant graphene/PANI nanocomposite electrodes have demonstrated high capacitance, long cycle life and impressive flexibility in supercapacitors. 68 Recently, our lab developed binder-free freestanding PANI-coated porous carbon nanofiber electrodes which achieved SC of 409 F g-1 at 0.1 A g-1, 80% of which was retained at a high current density of 10 A g-1.69 Control experiments using

PANI-coated, non-porous carbon nanofibers delivered much lower SC at the same current densities indicating that the double layer capacitance of the underlying carbon was retained and true integration of double layer and pseudocapacitance was achieved. Overall, the PANI/carbon composite electrodes have demonstrated good initial specific capacitance (SC) and reasonable capacitance retention (>80 %) after long cycling (1000-5000 cycles) as illustrated in table 1.70-72.

Composites of PANI with transition metal oxides have also been studied owing to the inherently high

73 pseudocapacitance of these materials. MnO2 is considered as one of the promising members of the metal oxide group for supercapacitor applications due to its low cost and eco-friendliness in addition to excellent theoretical pseudocapacitance (1370 F g-1).74 Its composite with PANI has shown promising electroactivity where PANI acts not only as nanostructured scaffold for MnO2 but also as a conductive medium.

Composites of PANI with other transition metal oxides such as nanostructured V2O5, TiO2, RuO2 and

75-77 hollow MoO3 (h-MoO3) have also been investigated with good reported electrochemical performance.

Unfortunately, high cost, low conductivity and environmental unfriendliness of some of these oxides hinder the development of their composites with PANI. Table 1 summarizes the literature on polyaniline based supercapacitors.

1.6.1.2 Pure PANI based Supercapacitor

Supercapacitor cathodes based on neat PANI (without use of additives) can potentially provide high electrode-level specific energy density. To offset PANI’s mechanical instability and to maximize its utilization, interesting PANI architectures based on 1-D and 3-D interconnected nanostructures have been developed and studied for supercapacitors applications.78-79 The nanostructured morphology is imperative 17 since most of the PANI redox reactions occur at or near the electrode/electrolyte interface. The nanostructures help to truncate charge transfer pathways, relief strains within the electrode due to electroactivity and in some cases, enable new reactions not possible within bulk materials.80 Synthesis techniques and application of various 1-D PANI nanomaterials have been studied for several years now.

The excellence in electroactivity of these 1-D nanostructures emanate from their morphology marked by high aspect ratio, porous surface and/or tubular morphology, which enables good electrode/electrolyte interaction, leading to efficient charge transfer to the reaction interface. Unfortunately, most of the supercapacitor electrodes based on 1-D PANI nanostructures still display undesirable loss in capacitance as shown in table 1 below. In comparison, recent works based on porous 3-D interconnected PANI nanostructures have shown promising capacitance retention capability at both higher current densities and after over 1000 cycles. Such improvement stems from the 3-D intertwined nanostructured network, which enables continuous electron conductivity as well as enhanced mechanical stability, and the hierarchical porosity, which facilitates faster ion diffusion/higher rate capability. Moreover, the macropores within such

3-D structures possibly relieve the stresses associated with volume changes during the doping/de-doping of

PANI leading to enhanced cycling stability.81-82

1.6.1.3 Electrolytes for PANI based supercapacitors

Acidic based aqueous electrolytes have primarily been studies for polyaniline based supercapacitors.83 This is because a proton is required for protonation of amine group during the doping/de-doping process in the polymer chain.63 This makes it hard to use other non-aqueous electrolytes such as aprotic ionic liquid which lack protons but possess high electrochemical stability. Few studies have reported polyaniline based supercapacitors operating in ionic liquids and neutral aqueous electrolytes but the redox mechanisms in such systems is not well understood.67, 84 Moreover, the cyclic stability of the polymer in ionic liquid is typically low. Since the voltage window of acidic aqueous electrolytes is limited to 1.0V, there is need to investigated activity of polyaniline in other electrolyte systems with wider electrochemical window in order to develop advanced supercapacitors. 18

Table 1-1. Polyaniline-based supercapacitors — Fabrication techniques, electroactivity (per gram of total electrode) at low and high current densities (CD), and cycle life.

Low CD rate High CD rate % Retention SC in F g-1 SC in F g-1 (# cycles) (Current Density) (Current Density) PANI hybrids PANI wt% Graphene, 71 40 448 (1 A g-1) 334 (10 A g-1) 81 (5000) CNFs, 69 48 409 (0.1 A g-1) 327 (10 A g-1) 80 (1000) MWCNTs, 85 95 530 (1 A g-1) 500 (3 A g-1) 90 (1000) 86 -1 -1 MnO2, 90 525 (2 A g ) 330 (10 A g ) 77 (1000) 75 -2 -1 V2O5, 50* 443 (0.5 mA cm ) 281 (5 mA cm ) 92 (5000) Pure PANI Synthesis Morphology Method Nanotubes(a), 87 self-assembly 625 (1 A g-1) 481 (10 A g-1) 77 (500) (a) 88 -1 -1 Nanofiber , template, MnO2 404 (1 A g ) 385 (10 A g ) 61 (5000) Nanorods (a), 79 self-assembly 455 (1 mV s-1) — 65 (1300) Hydrogel (b), 82 soft template 450 (0.5 A g-1) 420 (5 A g-1) 83 (10 000) CPANI (b),33 self-assembly 350 (1 A g-1) 315 (40 A g-1) 99 (500) CPANI(b), 89 Self-assembly 274 (1 A g-1) — 89 (1300) Redox-Electrolyte 90 -2 H2SO4+HQ-BQ, 524 (12.5 mA cm ) — >99.5(50000 ) 91 -1 H2SO4 + HQ, 584 (1 A g ) — 68 (5000)

a1-D nanostructures, b 3-D nanostructures, CPANI—crosslinked PANI chains. * Calculated from data

provided.

1.7 Summary on polyaniline based electrochemical devices.

Challenges: Despite the availability of diverse routes for fabrication of PANI nanomaterials, there are still

many challenges facing these synthesis techniques and the corresponding electrode assembly processes.

These challenges include low or negligible capacitance of the substrate for the case of hybrid electrodes, a

possibility of over-estimating the capacitance normalized per weight of PANI due to only a small amount

of polymer being deposited on the substrates and multi-step fabrication processes leading to additional cost,

time, and resources. Moreover, prior to electrochemical testing, the supercapacitor electrodes are almost

universally prepared by mixing the PANI nanostructures with binders such as PVDF and conductive

additive. 92 The preparation of such paste is not only likely to cause distortion of the PANI nanostructures 19

but also introduces electrochemically inert materials in the form of binders into the electrode. A

straightforward, single-step technique for fabricating self-supporting nanostructured PANI electrodes is

therefore highly desirable. Moreover, PANI has only been studied for application in supercapacitors using

acidic electrolytes which is thermodynamically limited to ~1.0V. Although notably higher capacitance is

achieved relative to high surface area carbon nanomaterials, the overall energy density for PANI based

supercapacitor is still low. There is need to investigate and develop PANI based supercapacitors that operate

in electrolytes with wider electrochemical windows.

1.8 Objectives

My dissertation seeks to address some of the aforementioned challenges facing synthesis and application

of PANI nanomaterials in electrochemical devices. By utilizing electrospinning technique, I sought to

develop binder-free 3-D polyaniline based nanofibers. The 3-D interconnected electrode morphology

allows efficient shuttling of charges to the reaction interface, therefore enabling higher electrode utilization

and rate capability. The electrospinning technique will also truncate the PANI nanomaterials preparation

process, therefore saving on time and cost. I have also sought to investigate the redox activity of polyaniline

in electrolytes that possess wide electrochemical window (Li2SO4 and protic ionic liquid). This will enable

development of a supercapacitor system that possess both high capacitance (PANI pseudocapacitance) and

wider electrochemical window. Through both electrochemical and analytical techniques (spectroscopic,

microscopic and scattering), I have sought to establish property-structure-performance correlation of PANI

based nanofibers toward developing efficient electrochemical devices. As far as fuel cell is concerned, I

have co-electrospun polyaniline with Nafion and investigated, via AC impedance, the feasibility of utilizing

such dual mat as a catalyst support system in PEM electrodes. This will enable simultaneous transport of

electrons and protons to active sites, and therefore maximize utilization of the catalysts.

CHAPTER 2. Electrospun Polyaniline for Supercapacitors

2.1 Introduction

Polyaniline (PANI), an electrically conducting polymer, has shown enormous promise as a pseudocapacitor

material owing to its ease of synthesis, good environmental stability, low cost and excellent

electroactivity.93 Bulk PANI, however, is not suitable for application in energy storage device electrodes

due to its low accessible surface area. Therefore, nanostructured PANI materials with large surface-area-

to-volume ratios – and hence shorter ion diffusion paths – have garnered interest as suitable electrode

materials for supercapacitors.89, 94 Versatile hierarchical PANI nanostructures and nanocomposites with

promising electrochemical performance in supercapacitors electrodes have been developed and studied.32,

95-96 PANI nanocomposites have typically been prepared by employing a carbon template. Recently, Luo

and co-workers synthesized a self-assembled hierarchical graphene@polyaniline nanocomposite via a

simple polymerization route. An electrochemical capacitor based on this nanocomposite showed specific

capacitance of 488 F g-1, 79% of which was retained after 1000 cycles.97 Sacrificial templates have also

been employed to fabricate high surface area hollow or tubular PANI nanostructures with high

electroactivity for use as supercapacitor electrodes.84, 98 A hollow PANI nanofiber-based supercapacitor

electrode with specific capacitance of 601 F g-1 was recently fabricated by polymerizing aniline monomers

on pre-electrospun poly (amic acid) (PAA) nanofibers, followed by selective removal of the PAA nanofiber

template. The use of a sacrificial template to fabricate PANI nanostructures is perceived as less efficient

due to potential structural degradation during the template removal process. Techniques that don’t rely on

templates have therefore been employed to directly synthesize various PANI nanoarchitectures.99-101

Despite the availability of the aforementioned diverse routes for PANI fabrication, there are still many

challenges facing these synthesis techniques and the corresponding electrode assembly processes. These

challenges include low or negligible capacitance of the substrate for the case of hybrid electrodes, a multi-

step fabrication process leading to additional cost, time, and resources, and a possibility of over-estimating

the capacitance normalized per weight of PANI due to only a small amount of polymer being deposited on 21 the substrates. Moreover, prior to electrochemical testing, the supercapacitor electrodes are almost universally prepared by mixing the PANI nanostructures with binders such as PVDF and conductive additive. 92 The preparation of such paste is not only likely to cause distortion of the PANI nanostructures but also introduces electrochemically inert materials in the form of binders into the electrode. A straightforward, single-step technique for fabricating self-supporting nanostructured PANI electrodes is therefore highly desirable.

Electrospinning offers such an alternative technique. Although electrospinning has widely been adopted in literature to fabricate eclectic polymer nanostructures due to the simplicity of its setup and the ease with which the electrospun nanofiber morphologies can be modulated (ref), PANI is an extremely difficult polymer to process via electrospinning. The aromaticity in the polymer’s backbone makes individual PANI chains extremely rigid, preventing PANI solutions from experiencing sufficient chain entanglements to achieve the minimum solution viscosity required for successful electrospinning. 102-103

Additionally, PANI possesses limited solubility and dispersion in most organic solvents, making it difficult to obtain a PANI solution with sufficient viscosity just by increasing the polymer concentration. To circumvent these issues, an electrospinnable carrier polymer can be added to PANI to form an electrospinnable polymer blend. Unfortunately, efforts to electrospin nanofbers with high concentrations of

PANI have been hindered by the necessity to include a substantial amount of carrier polymer in order to successfully electrospin continuous and smooth nanofibers.104 The carrier polymer is typically electrically insulating; thus, high concentrations of carrier polymer severely reduce the electrical properties of the electrospun PANI nanofibers. Frontera et al electrospun PANI/PEO with w/w percent ratio as high as 80/20; however, the fiber mats of the resulting nanofibers exhibited a substantial amount of beads and fiber breakage.105 Such beads and discontinuity among the fibers are expected to inhibit optimal electrode performance when applied in a pseudocapacitor device. Chaudhari et al developed supercapacitor electrodes using electrospun PANI/PEO nanofibers with a w/w percent ratio of 50/50. They reported a maximum specific capacitance of 267 F g-1 at a current density of 0.35 A g-1 5. To our knowledge, this is 22

the only work in the literature that has reported the capacitance of electrospun PANI nanofibers. In addition

to the high fraction of electrochemically inert carrier polymer used, the nanofiber electrodes fabricated in

that work were not freestanding – the conventional slurry technique was employed to prepare the

supercapacitor electrodes with powdered nanofibers and insulating binders.

Our work demonstrates a facile, single-step fabrication of freestanding, high-purity PANI and

PANI/CNT nanofiber mats via electrospinning. We used high molecular weight PEO (8000 kDa) as a

carrier polymer, which enabled successful electrospinning of PANI/PEO and PANI/CNT/PEO nanofibers

with a low PEO composition of only 7wt%. In addition to the use of high MW PEO, successful

electrospinning of high purity PANI nanofibers was enabled by doping PANI with a salt, which enhanced

its solubility in chloroform.106 The addition of CNTs enhanced the electrical conductivity and in turn the

electrochemical performance of the PANI/PEO nanofibers. Specific capacitance of 308 F g-1 and 385 F g-1

was achieved for PANI/PEO and PANI/CNT/PEO, respectively, at a current density of 0.50 A g-1. Both

nanofiber mats showed high rate capability and good stability after 1000 cycles. To the best of our

knowledge, no previous work has reported the preparation of PANI/CNT supercapacitor electrodes via the

electrospinning technique. The majority of the PANI/CNT research in the literature reports electrochemical

or chemical deposition of polyaniline on a CNT template, often resulting in a high CNT fraction and dead

weight in the final electrode.

2.2 Experimental Methods and Materials

2.2.1 Fabrication of PANI and PANI-CNT nanofibers.

Materials. Polyaniline emeraldine base (PANI, Mw = 100,000 g/mole), polyethylene oxide (PEO, Mw =

8,000,000 g/mole), 10-camphorsulfonic acid (HCSA), and chloroform (>99.5%) were purchased from

Sigma Aldrich. Carbon nanotubes (CNTs) (multi-walled, 8-15 nm diameter) were purchased from NanoLab

Inc. 23

Chloroform, PANI, and HCSA were mixed simultaneously and stirred at approximately 300 rpm for more than 6 hours. The PANI concentration in chloroform was 0.67 wt%. A weight ratio of 1:1.29 PANI:HCSA was employed to achieve PANI doping. After stirring for at least 6 hours, the solution was sonicated for 15 minutes. PEO was added immediately after ultrasonication in a 93:7 PANI:PEO weight ratio and the resulting solution was further stirred for 6 hours. For the PANI-CNT solutions, CNTs were sonicated in chloroform for 2 hours prior to mixing. The PANI:CNT:PEO weight ratio was kept fixed at 81:12:7.

Electrospinning was carried out in a controlled environment to ensure that the relative humidity never exceeded 20%. The NE-4000 model from New Era Pump Systems, Inc. and a BD 5 mL syringe with a luer- lock tip were utilized as the syringe pump and syringe, respectively. The electrospinning spinneret was a

22 gauge needle from Hamilton. Identical electrospinning parameters were used for the PANI93 and PANI-

CNT solutions. The collector plate was placed approximately 25 cm away from the needle tip. The solutions were pumped at a flow rate of 0.6 mL hr-1 and a 5 kV DC voltage was applied to the system.

For ease of reference, the electrospun PANI/PEO (93/7) and PANI/CNT/PEO (81/12/7) electrospun nanofiber mats will be referred to as PANI93 and PANI-CNT, respectively. The composition of the nanofiber electrodes was confirmed using thermogravimetric analysis.

2.2.2 Electrochemical testing and morphological characterization.

Scanning electron microscopy (SEM, ZEIS SUPRA 50VP) was employed to characterize the surface morphology of the electrospun fibers. ImageJ software was utilized to measure the average diameter of the electrospun fibers from the SEM images. The software was used to measure diameters of 100 nanofibers for both PANI93 and PANI-CNT. The percolation of CNTs within PANI nanofibers was observed using transmission electron microscopy (TEM). Fourier transform infrared spectroscopy (FTIR) (Varian

Excalibur FTS-3000, range of 4000-800 cm-1), X-ray diffraction (XRD) (Rigaku SmartLab, X-ray diffractometer, Cu Kα, scanning range 5-50° and step size of 0.02°) were used to probe the chemical structure, composition and crystallinity of the PANI based nanofibers. The surface area of the electrospun nanofibers was characterized using nitrogen sorption isotherms at 77 K (Autosorb-1, Quantachrome). Prior 24 to the adsorption–desorption measurement, all samples were degassed at 60°C under vacuum for 24 h to remove impurities.

The as-made electrospun PANI93 and PANI-CNT nanofiber mats were punched into freestanding electrodes with a 0.50 inch diameter and tested electrochemically in a three electrode, T-type Swagelok set up, as shown in the supplementary figure S1. To illustrate the electrodes’ freestanding morphology, digital photographs are shown in figure S2. A 0.5 inch diameter graphite rod, Ag/AgCl and platinum mesh were used as the working, reference and counter electrodes, respectively. All electrochemical testing was carried

-2 out in a 1M H2SO4 electrolyte. The electrode mass loading used for the reported data was 2.60 mg cm .

Cyclic voltammetry (CV) experiments were conducted in the voltage range of -0.20 to 0.65 V. Specific capacitance was calculated from CV data using the expression below:

1 ∫ 퐼 푑푉 퐶 = (6) 푠 2 푣∗푚∗푉

-1 where Cs, I, m and V represent the specific capacitance (F g ), current (A), mass of the electrode (g), and the voltage window (V). Galvanostatic charge-discharge (GCD) measurements were carried out at varying current densities with a potential range of -0.20 V to 0.60 V vs. Ag/AgCl. Specific capacitance was calculated from the galvanostatic charge-discharge curves using the expression below:

훥푡∗퐼 퐶 = (7) 푠 훥푉∗푚

where Cs is the specific capacitance, I is the constant discharge current, m is the mass of the electrode, Δt is the discharge time, and ΔV is the potential window. Electrochemical impedance spectroscopy (EIS) in the range of 100 kHz to 10 mHz was used to study the impedance behavior of the PANI93 and PANI-CNT electrodes in a 3-electrode setup. A Reference 3000 instrument from Gamry Instruments was used to measure the CV, GCD and EIS. 25

Schematic 1. Illustration of the freestanding polyaniline nanofiber based electrodes.

2.3 Results and Discussion

2.3.1 Nanofibers characterization.

As shown in the SEM image in figure 2.1 (a), electrospinning the PANI93 solution results in smooth,

continuous, nonwoven nanofibers. Well defined inter-fiber porosity percolates throughout the nanofiber

mats. The addition of CNTs did not appear to have any negative consequences on the electrospinnability of

the solution: smooth PANI-CNT fiber mats were generated as shown in figure 2.2 (b). TEM images (figure

2.1 c,d) show that CNTs are distributed within the nanofibers with no major aggregation. The CNTs are

also shown to be well aligned within the nanofibers over a length of more than 1 µm. The PANI-CNT fiber

sample showed a smaller average diameter of 528±50nm compared to 678±54nm for PANI93. Introduction

of CNTs increases the conductivity of the PANI-CNT solution, which leads to higher charge density in the 26 polymer jet. The higher charge density increases the extent of jet elongation, resulting in a smaller average diameter.61

Figure 2.2 (a) shows the XRD pattern of pristine CNTs, PANI93 and PANI-CNT nanofibers. XRD pattern of pristine CNTs displayed a prominent diffraction peak at 2θ =26.3 corresponding to the graphite-like structure.107 The XRD spectra shows prominent PANI peaks at scattering angle of 2θ= ~15° and ~25° attributed to the periodicity of the repeat unit of PANI chain and periodicity parallel to the polymer chain backbone indicating higher degree of crystallinity within the nanofibers, possibly induced by extensional forces during electrospinning.108-109 Moreover, a weak peak at 2θ= ~20.9° ascribed to the periodicity perpendicular to the PANI chain was also observed.98 The PANI-CNT nanocomposite XRD pattern exhibited similar peak to that of PANI93. However, the diffraction peak at ~2θ= ~25° was much sharper for the PANI-CNT indicating a more pronounced graphitic structure within the nanofibers due to presence 27 of CNTs. Incorporation of CNTs within the PANI does not seem to introduce any additional crystalline order.

Figure 2.2. XRD patterns (a) and FT-IR spectra (b) of electrospun nanofibers and pristine CNTs.

Figure 2.2 (b) shows the FTIR spectra of the PANI93 and PANI-CNT nanofibers. The main bands at 1583 and 1492 ascribed to the C=C stretching vibration of the quinoid ring and benzonoid ring, respectively, in the PANI chain are observed for both PANI93 and PANI-CNT, indicating the emeraldine salt state of PANI.

85 The bands at 1307 and 1244 are ascribed to the C-N stretching of the benzonoid ring while the band at

~1138 is attributed to N-Q-N (where Q is quinoid) and this band is attributed to the degree of electron delocalization. The bands at 800 and 1735 are assigned to the out of plane bending vibration of –CH in the benzonoid ring and C=O stretching, arising possibly due to presence of PEO in the nanofibers. The position of these peaks are not altered in the FTIR spectra of PANI-CNT composite, indicating non-existence or weak chemical interaction between PANI and CNT. It should be noted that, the interaction between PANI has been primarily observed for cases where aniline is polymerized in situ on CNTs. Cochet et al showed that site-selective interaction between CNTs and quinoid ring of PANI existence when the PANI-CNT composite is synthesized via in situ polymerization.110 Such pronounced interaction has not been reported 28 for ex-situ synthesis. Nevertheless, enhanced electrical conductivity has been observed for PANI-CNT composites fabricated ex situ.60

Electrochemical Impedance Spectroscopy (EIS). Prior to three electrode testing, EIS was employed to establish the through plane electrical conductivity of electrospun PANI93 and PANI-CNT. Graphite current collectors with a diameter of 0.50 inches were used in a two way Swagelok cell. PANI93 and PANI-CNT exhibited electrical conductivity of 0.114 and 0.154 S/cm, respectively, indicating that the addition of CNTs enhanced the electrical conductivity of the nanofibers. EIS was further used to investigate the impedance in a three electrode set up. The corresponding EIS Nyquists plots are shown in figure 2.3 (a). The Nyquist plots are characterized by the distorted semi-circle located in the high frequency region and a straight line in the low frequency region. The high frequency x-axis intercepts give equivalent series resistance (ESR) values, which were observed to be identical for both electrodes. The intercepts gave an approximate ESR of 0.5 ohms. PANI-CNT electrodes exhibited lower charge transfer resistance compared to PANI 93 electrodes, as suggested from the respective diameters of the semi-circles. Additionally, PANI-CNT electrodes possessed faster ion diffusion based on the slope of the inclined straight line in the low frequency regime.

Figure 2.3. Nyquist plots of electrospun PANI93 and PANI-CNT nanofiber mats (a) and CV curves (b) of PANI93 (red) and PANI-CNT (blue) nanofiber mats at 50 mVs-1.

The decreased electrical resistance of the PANI-CNT mat compared to the PANI93 mat could be attributed to the combined effect of high electrical conductivity due to the presence of CNTs and the smaller diameter of the PANI-CNT nanofibers, which facilitates enhanced electron transfer and improved ion penetration and diffusion into the nanofiber electrode.

CV study of PANI93 and PANI-CNT nanofiber mats. Pseudocapacitive performance of the freestanding high purity electrospun PANI93 and PANI-CNT composite electrodes was investigated using cyclic voltammetry (CV) measurements at room temperature with a voltage window of 0.85V. Both PANI93 and

PANI-CNT exhibited a well-defined reversible pair of redox peaks in the CV curves in the voltage range of 0.10 V to 0.30 V vs. Ag/AgCl. The peaks are represented as A/A’ in figure 2.3 (b) below. These peaks are attributed to the redox transformation of polyaniline between the polaronic emaraldine state (ES) and the leucoemaraldine state (LS) 111. Another pair of peaks, denoted as B/B’, was observed for both systems and is believed to represent the PANI degradation by-products 112. PANI-CNT electrodes showed higher redox peaks than the corresponding PANI93 electrodes at all CV scan rates. During both the forward and 30 reverse sweeps, the conductivity of the PANI93 electrode varies as the polymer transitions into different oxidation states. The polymer is at its least conductive state when in the leucoemaraldine state. This variation in conductivity of PANI may inhibit the redox reactions. Conversely, the presence of CNTs in the composite system is expected to stabilize the electrode conductivity and therefore facilitate the redox reactions, which explains the larger redox peaks observed for the PANI-CNT electrode. The PANI-CNT redox peak is marginally shifted to the right for the forward scan and to the left for the reverse scan. We hypothesize that this phenomenon is a result of greater conversion of the ES to LS state in the presence of

CNTs. Specific capacitances of 324 F g-1 and 281 F g-1 were observed for PANI-CNT and PANI93 at 20 mVs-1, respectively based on the CV measurements. Figure 2.4 (a,b) shows the CV curves for PANI93 and

PANI-CNT at various scan rates from 5 to 100 mVs-1 and table 1 reveals the specific capacitances calculated from these curves. The cathodic and anodic peaks exhibit marginal shifts to the right and left, respectively, with increasing scan rate. The small shift suggests only a slight increase in internal resistance with increasing scan rate 113. 67% of the specific capacitance was retained at 100 mV s-1 for PANI93 electrodes.

For PANI-CNT electrodes, the specific capacitance remained relatively steady (80% retention at 100 mV s-1) with increasing scan rate, indicating excellent kinetics and rate capability due to the presence of CNTs facilitating fast electron transport. A linear correlation was observed for the wave current as a function of the square root of scan rate (Figure S4) for both PANI93 and PANI-CNT electrodes indicating that the kinetics during the redox reactions in both systems are diffusion controlled.

Table 2-1. Specific capacitance (SC) of PANI93/PANI-CNT electrodes at different scan rates.

Scan rate (mV/s) SC (F g-1) PANI93 SC (F g-1) PANI-CNTs

5 281 324

20 240 302

50 213 285

100 189 259 31

Galvanostatic charge-discharge (GCD) study of PANI93 and PANI-CNT nanofiber mats. Figure 2.4 (c,d) illustrates the charge-discharge curves in the potential range of -0.2V to 0.6V. These charge-discharge curves are quasi symmetric due to the redox reactions associated with the transition of PANI to various oxidation states 84.

The rate capability of the fabricated systems was investigated by varying the current density from a range of 0.5 A g-1 to 10 A g-1. Figure 2.4 e illustrates the variation of specific capacitance as a function of current density for both PANI93 and PANI-CNT. A maximum specific capacitance value of 308 F g-1 was obtained for PANI93 at a current density of 0.5 A g-1 and an electrode mass loading of 2.60 mg cm-2. Additionally, when the current density was increased to 5 A g-1, 58% of the maximum specific capacitance value was retained. An area based capacitance of 1.15 F cm-2 with a negligible loss in gravimetric capacitance was obtained when the electrode mass loading was increased to 3.80 mg cm-2. As previously indicated, this is the first time electrospinning has been employed directly to fabricate a binder-free high purity, freestanding

PANI nanofiber network for supercapacitors. Our freestanding PANI-based electrode material offers several benefits over PANI based electrodes that employ slurries, such as the aforementioned system reported by Chaudhari et al 5. Our electrode exhibited higher effective specific capacitance and enhanced rate capability owing to the binder-free 3D fiber mat morphology and the low concentration of the electrochemically inert PEO. Additionally, the high purity PANI nanofibers developed in our work are expected to be significantly more stable in aqueous electrolyte than those with higher concentrations of

PEO, a water-soluble polymer.

On comparing the PANI93 performance to other nearly-pure PANI electrodes – that is, those without any carbon substrates – reported in the literature, we find that PANI93 exhibits higher specific capacitance than many reported materials, including PANI powder (208-216 F g-1) 1146, PANI nanorods (273 F g -1) 115, PANI nanofibers (298 F g-1) 116 and compares favorably to those of PANI nanoworms (301 F g-1) 97, PANI nanotubes (422 F g-1) 92, 114, and 3-D PANI hydrogels (450 F g-1) 82. The promising electrochemical 32 performance of our as-electrospun freestanding PANI93 electrodes is thought to stem from its three- dimensional non-woven fiber mat morphology, which enables quick ion transport and diffusion via inter- fiber macrospores to PANI active sites. The standard electrode assembly process – utilization of slurries – in most literature works cited above is likely to result in a lack of such an open, through-connected pore structure. Moreover, PANI93 nanofibers display some microporosity on their surface as observed in the

BET data shown in figure 2.5. The microporosity is possibly the result of evaporative cooling of the polymer jet due to the presence of some humidity during electrospinning. As the surface of the jet cools, water vapor from the ambient air condenses to form tiny water droplets on the fiber surface that eventually evaporate to form pores 3. We believe that this porosity on the surface of the fibers further enhances PANI accessibility and utilization for redox reactions.

Figure 2.5. (a) Pore size distribution and (b) cumulative surface area as a function of pore size for PANI93 nanofibers.

CNTs were added to PANI nanofibers to address the well documented PANI instability issues 117 as well as to enhance the electrical conductivity of the PANI composite, particularly when PANI transitions into the less conductive leucoemaraldine state. Specific capacitance of 385 F g-1 at 0.50 A g-1 was obtained, with

72% retention of capacitance at 5 A g-1 and ~66% retention at 10 A g-1. Areal capacitance of 1.37 F cm-2 was observed when electrode loading was increased from 2.60 to 3.80 mg cm-2 with negligible loss in 33 gravimetric capacitance. As mentioned earlier, no prior work in the literature has reported the investigation of the pseudocapacitive behavior of electrospun PANI-CNT nanofibers. The bulk of the PANI/CNT composite electrode literature utilizes CNT as a substrate on which to deposit PANI ex-situ. Nevertheless, the capacitance value of 385 F g-1 obtained for our electrospun PANI-CNT is competitive compared to many of the recently reported PANI/CNT-based supercapacitors, which are mainly fabricated via aniline polymerization on CNTs templates ; PANI/CNT ( 320 F g-1, 20 wt% CNT) 117, PANI/SWCNTs (485 F g-1,

27wt% SWCNTs) 118, PANI/MWCNTs (400.2 F g-1, 25 wt% CNT) 119 , PANI/CNT nanocomposite film

(535 F g-1, 5 wt% CNT) 85. Compared to these other composites, the electrospun PANI-CNT nanofiber composite reported herein possesses the added advantages of low mass loading of the electrochemically inactive CNTs and a single step fabrication process, which eliminates the necessity of binders, reduces cost and saves time. Similar to PANI93, the specific capacitance and hence PANI utilization of our electrospun

PANI-CNT electrodes can be further improved by reducing the nanofiber diameter to decrease the ion diffusion path length.

Figure 2.6. Plots showing electrochemical performance of symmetric PANI-CNT capacitor: CV curves at different scan rates (a) and specific capacitance as a function of charge−discharge rates (b). The specific capacitance is based on the total mass of both electrodes.

Full cell capacitor setup. A full cell capacitor based on symmetric PANI-CNT electrodes was fabricated and tested in a two-way Swagelok set up. The device was characterized using both CV and galvanostatic 34 charge-discharge with a voltage window of 0.8V (0-0.8 V). Figure 2.6 (a) shows the CV plots for the full cell at various scan rates ranging from 20 to 100 mV s-1. The CVs display the characteristic pair of peaks observed in the three-electrode test, which are associated with the leucoemeraldine-emaraldine transformation (A/A’) and the formation of by-products (B/B’). Figure 2.6 (b) shows the rate capability of the symmetric electrodes using the galvanostatic charge-discharge process. Maximum specific capacitance of 320 F g-1 was obtained at 0.5 A g-1 using equation 3 below. 83% of the maximum capacitance was retained after 1000 charge-discharge cycles at a 2 A g-1 current density.

4훥푡∗퐼 퐶 = (6) 푠 훥푉∗푚

Energy density and power density values for the full device were also calculated according to equation 8 and 9 below, respectively. Ct is the specific capacitance based on total weight of both electrodes, V is the voltage window and Δt is the discharge time.

1 2 퐸 = ∗ 퐶 ∗ 푉 (8) 2 푡

퐸 푃 = (9) 훥푡

A maximum energy density of 7.11 w h kg-1 (corresponding to 28.4 w h kg-1 per weight of a single electrode) was obtained for the full cell at the power density of 201 w kg-1. The low energy density per total weight of both electrodes is due to the limited electrochemical window (0.8V). A Ragone plot is shown in figure 7.1S below.

Figure 2.7. SEM micrographs showing post-mortem electrode morphologies of PANI93 (a) and PANI- CNT (b); life cycle performance of PANI93, PANI-CNT, and symmetric PANI-CNT capacitors (c) and EIS plot of PANI-CNT before and after cycling for 1000 cycles (d). Scale bar is 2 μm, and inset scale bar is 200 nm for (a, b).

Life cycle performance. The cyclic stability of PANI93 and PANI-CNT electrodes was investigated through continuous charge—discharge cycling at 2 A g-1. The resulting data is shown in figure 2.7 and figure 2.7c (for symmetric PANI-CNT capacitor). The electropsun PANI93 supercapacitor electrode showed 70% retention of its initial specific capacitance after 1000 cycles while PANI-CNT electrodes showed a retention of 81% of its initial value. Furthermore, a small decrease in the intensity of peak A/A’ was observed in the CV plots of the PANI-CNT electrode after 1000 cycles at 20 mV s-1 (figure S6), attributted to the increase in internal resistance with cycling as shown in the EIS curves (figure 11c). PANI is known to undergo volumetric changes during the doping/de-doping process as a result of repeated insertion and de-insertion of ions, which results in the deterioration of its performance over the course of 36

operation 120. The presence of CNTs enhanced the mechanical stability of these electrodes, consistent with

previously reported literature 108. Addition of CNTs boosted the electrochemical stability of our nanofibers

by approximately 11%. From post-mortem SEM characterization of the electrodes shown in figure 7, we

observe that the electrospun nanofiber mats largely retain their porous interconnected network, indicating

that any PANI degradation might be occurring at molecular level. The relatively lower capacitance retention

of PANI93 is attributed to the larger PANI domain size corresponding to the larger average nanofiber

diameter. Efforts to reduce the nanofiber diameter led to the need for higher composition of the

electrochemically-inert carrier polymer. Future work on generating additional porous within high purity

PANI nanofibers to enhance accessibility, stability and rate capability of PANI is underway.

2.4 Conclusion

We have demonstrated a facile methodology to fabricate binder-free PANI and PANI-CNT nanofiber

electrodes. The electrodes exhibit a specific capacitance of up to 385 F g-1. The strong electrochemical

performance of the fabricated electrodes is enabled by the combination of low electrochemical impedance

and good inter and intra-fiber porosity, which facilitates ion diffusion to the PANI active sites. The

freestanding format of the electrodes eliminates the need for electrochemically inert binders, thereby

allowing us to not only minimize cell assembly procedures, but also to preserve the 3D porous structure of

the nanofiber mats throughout the lifespan of the capacitor.

CHAPTER 3. Asymmetric Supercapacitor Based on Polyaniline-Coated Carbon Nanofibers

3.1 Introduction

Supercapacitors can either store electric energy through electrostatic accumulation of charges at the

electrode/electrolyte interface (referred to as electrochemical double layer capacitor—EDLC) or via highly

reversible redox reactions occurring at the surface of the electrode materials (referred to as

pseudocapacitor).4, 32 The supercapacitors are widely desirable for application in portable electronics,

memory back-up systems and automobiles owing to their high power delivery capabilities and long cycle-

life. However, compared to batteries, the energy density of the supercapacitors is low (< 10 W h kg-1),

which has resulted in considerable attention being directed towards developing high energy supercapacitors.

Since the energy density of a supercapacitor is given by E= ½ CV2, improvements in energy (E) can be

achieved by enhancing the specific capacitance (C) and/or the cell voltage (V). Higher capacitance can be

attained through utilization of high surface area carbon and pseudo-capacitive materials such as conducting

polymers and transition metal compounds121-122 whereas high operating voltage (OV) can be enhanced

through adoption of organic- and ionic liquid-based electrolytes.52 The organic and ionic liquid electrolytes

are relatively expensive, eco-unfriendly and poor conductors (low power density) compared to aqueous

electrolytes.4 Developing asymmetric supercapacitor (ASC) based on aqueous electrolyte (ASC) is

therefore an attractive methodology to prepare a high energy supercapacitor while retaining its core feature

i.e. fast charge-discharge. Aqueous electrolytes generally possess moderate electrochemical stability

relative to organic- and ionic liquid-based electrolytes due to water decomposition (1.23V). By adopting an

asymmetric cell configuration, materials with high over-potential for hydrogen and oxygen evolution can

be utilized in separate potential ranges to extend the overall cell voltage window beyond its thermodynamic

limit.

Combination of electrochemical double layer based materials (anodes) and battery-type materials

(cathodes) are typically employed to fabricate ASCs. In such device set up, high power delivery is obtained 38 from high surface area carbon nanomaterials whereas high energy capacity is dervied from the redox based nanostructures. For development of high energy, eco-friendly and economically competitive ASCs, adoption of carbon based anodes and conducting polymer (CPs) based cathodes have been employed. Both carbon and conducting polymers occur in natural abundance and possess benign environmental effects.123

Activated carbons have primarily been used to develop anodes for ASCs. However, the activated carbons often lack well defined porous network, which results in poor electrode wettability and low capacitance values.124 Therefore, considerable effort is being made to develop anodes based on hierarchically porous carbon nanostructures to enable high energy density and power delivery.27, 47, 125 However, the synthesis procedures for such carbons are generally extensive and complex. For the cathode, polyaniline has drawn immense attention due to its excellent theoretical pseudocapacitance (750 F g-1), good electrical conductivity relative to other pseudocapacitive materials and processability. To minimize cycling losses resulting from repeated doping/dedoping, polyaniline is often integrated with carbon nanostructures at the cathode.35 The carbon backbone provides mechanical support and electrical conductivity. Luo and co- workers synthesized a self-assembled hierarchical graphene/polyaniline nanocomposite, which showed a gravimetric capacitance of 488 F g-1 and capacitance retention of 79% after 1000 cycles.97 The promising electrochemical performance is a result of the synergistic effect between the carbon core and polyaniline shell. Although commendable progress has been made in improving the performance of both anode and cathodes in ASCs, there is still need to enhance the electrode microstructure, truncate the electrode processing steps and eliminate the need for electrochemically inert binders (typically 10-15%) when integrating nanomaterials into electrochemical capacitors.

Moreover, it should be noted that the ASCs based on carbon||polyaniline have largely been studied in acidic aqueous electrolytes. The need for H+ ions for doping of PANI during redox processes has largely dictated the choice of the acid electrolyte. However, presence of high concentration of protons also lead to low potential range on the anode side due to hydrogen evolution. Moreover, acid electrolytes present possible corrosion and safety challenges. Neutral aqueous electrolytes are promising alternatives to acidic/basic 39 electrolyte due to high electrochemical stability and non-corrosiveness.44 It has been previously

44 demonstrated that the electrochemical windows of up to 2.2V is attainable for L2SO4-based electrolytes.

The large stable electrochemical window of Li2SO4 based aqueous electrolyte is attributed to strong

+ 2- solvation energy (~160-220 KJ/mol) of Li and SO4 ions, which leads to high overpotentials for H2/O2 evolution. Only a handful of studies have demonstrated PANI-based ASCs with neutral aqueous electrolytes and the electrochemical mechanisms of polyaniline is not properly delineated in such studies.75,

126-127 Operation of PANI in neutral electrolyte will not only enable wider electrochemical window but will also allow pairing of PANI with other anodes such as V2O5 which are chemically unstable in acidic media.128

In this work, we report asymmetric supercapacitor based on activated porous carbon nanofibers (A-PCNFs) as anode and PANI-coated porous carbon nanofibers (PANI-PCNFs) as cathode in a neutral aqueous electrolyte. The electrodes are characterized by interconnected porous networks which ensure short diffusion paths for ions and fast transport of electrons to the reaction interface. The electrodes are also freestanding and free of binders unlike most literature works that employ powder based electrode materials mixed with binders that are not only insulating but also add dead weight to the device. We obtain a specific capacitance of 170 and 320 F g-1 for A-PCNFs (anode) and PANI-PCNFs (cathode), respectively, in a 3- electrode set-up. Asymmetric cell based on these electrodes exhibits an energy density of 29 W h kg-1 at an operating voltage window of 1.8V. It exhibits a capacitance retention of 75% over 4500 cycles at 2 A g-1 with the capacitance remaining largely constant after the first 500 cycles. Such high capacitance in a neutral electrolyte is attributed to the excellent pore distribution in the A-PCNF network (surface area of 2218 m2 g-1) and conformal, thin coating of PANI on porous carbon nanofibers in PANI-PCNFs, which facilitate effective charge transport within the polymer and therefore higher utilization. To further develop understanding of the redox mechanisms of polyaniline in neutral aqueous electrolyte, in operando infrared spectroelectrochemical study was conducted. We showed that PANI exhibits similar redox transitions and doping/de-doping mechanism in Li2SO4 aqueous electrolyte as typically observed in acidic electrolytes. 40

The IR band at 1566 cm-1 typically associated with quinoid structure of the polymer chain is shown to

increase in intensity as potential is applied, indicating oxidation of the polymer. This characteristic band is

typically observed around ~1572 cm-1 in acidic media. The slight red shift in neutral electrolyte could be

attributed to the larger solvated Li ions (relative protons) which may affect the stretching vibrations of

adjacent molecules. Doping of the polymer chain was also observed by increase in the intensity of the IR

band at 1038 cm-1 associated with electron delocalization. In acidic media, protons interact with the amine

group (a strong Bronsted base group) during the doping of the polymer leading to the electron

delocalization. Due to absence of protons in the electrolyte, such doping phenomenon is attributed to Li

ions.

Schematic 1: illustration of the preparation of binder-free A-PCNFs and PANI-PCNFs electrodes.

3.2 Experimental techniques

3.2.1 Electrode preparations.

Detailed preparation of the activated porous carbon nanofibers is reported here.52 Briefly, a 30/70 w/w blend

of polyacrylonitrile (MW = 150 000 g mol−1, Sigma-Aldrich) and Nafion powder (obtained by drying 41

LIQUION 1115) was dissolved in DMF under gentle heating and stirring for 4hrs followed by electrospinning. The polymer blend comprises 21 wt% of the DMF based solution. The resultant nanofiber mat was then stabilized at 280°C in air for 5hrs followed by carbonization at 1000°C for 60 minutes under nitrogen atmosphere. During the carbonization process, Nafion is thermally decomposed out to generate predominantly mesoporous nanofibers. Chemical activation of the electrospun PCNFs was achieved by soaking the nanofibers mat in 30 wt% KOH solution overnight and blotting it with lint-free paper upon removal. The sample was then heat treated to and kept at 800°C for 30 mins under nitrogen flow. The activated PCNFs (a-PCNFs) were utilized as the anode. PANI coated PCNFs were generated via electrochemical polymerization of aniline on carbon nanofibers at 8mA. The loading of polyaniline on the porous carbon nanofiber was ~40 wt% as previously optimized.

3.2.2 Spectroscopic Characterization

In situ FTIR: To develop understanding of the dynamics of PANI in neutral aqueous electrolytes, we conducted in situ infrared spectroscopy studies while obtaining CV. A Fourier transform infrared (FTIR) spectrometer (Nicolet iS-50s, Thermo Electron Corporation) was used to collect all spectra using a diamond

ATR. Infrared spectra were collected using a kBR detector at 3 scans per spectrum and at a resolution of 4 cm−1 resulting in a spectrum collected every 3.8 s. All spectra were corrected with a background subtraction of the ATR crystal spectrum. Gamry Reference 3000 potentiostat was used to perform electrochemical measurement (CVs). Asymmetric cell, A-PCNFs||PANI-PCNFs, was clamped onto the ATR crystal in a two-electrode cell. The bottom electrode of the supercapacitor, PANI-PCNFs, is in intimate contact with the ATR crystal and extend beyond the ATR crystal surface. Aluminum leads were used as current collectors. Schematic 2 below illustrates the in-situ measurement set-up.

Schematic 2: Illustration for the in situ set up used to investigate electrochemical performance of polyaniline in 1M Li2SO4 electrolyte.

3.3 Results and Discussion

We have previously reported detailed preparation of the porous carbon nanofibers.129 In brief, blend of

polyacrylonitrile (carbon precursor) and Nafion (sacrificial polymer) were electrospun and subjected to heat

treatment at 1000C for 1 hr in nitrogen atmosphere. During the pyrolysis process, the Nafion is thermally

decomposed out to create self-standing porous carbon nanofibers (PCNFs). The nanopores generated from

Nafion removal create pathways for quick transport of electrolyte ions to the double layer interface. To

fabricate the ASC anode, the PCNFs were chemically activated by soaking in KOH overnight and

subjecting it to thermal treatment at 800◦C for 30 minutes. The activation mechanisms of PCNFs with KOH

is followed by 6 KOH + 2C 2K + 3 H2 +2 K2CO3, then decomposition of K2CO3 and reaction of K/K2CO3

with carbon. This activation process primarily generates micro-pores on nanofibers which has been shown

to cause increase in energy storage capacity of carbon based nanomaterials.14 The PCNFs were also used

as a scaffold to deposit PANI. Aniline was polymerized on PCNFs via galvanostatic deposition at 8 mA

69 from a 0.5M Aniline + 1M H2SO4 solution. The loading of PANI on the PCNFs was kept at ~40 wt%,

calculating from the total charge observed during deposition. The morphologies of PCNFs, PANI-PCNFs

and A-PCNFs are illustrated in figure 3.1 (a), (b) and (c), respectively. It should be noted that the nanofiber

mats of all three samples retain the 3-D non-woven nanofiber morphology which is essential for electrolyte

permeation. The A-PCNFs exhibit large surface area of 2218 m2 g-1 and is characterized by both macro- 43 and micro-pores as observed in the porosimetry studies using Brunner-Emmett-Teller (BET) studies (see figure 3.1(d). Polyaniline-coated PCNFs exhibit rough surface due to conformal growth of PANI on the nanofiber surface. Detailed discussion on deposition parameters of polyaniline is available in our previous studies.69

Figure 3.1 : SEM micrographs of PCNFs (a), PANI-coated PCNFs (b) and KOH-activated A-PCNFs. Figure C shows the pores distribution of A-PCNFs.

3.3.1 Three-electrode tests:

Electrochemical performance of A-PCNFs and PANI-PCNFs were characterized individually in a three electrode system using cyclic voltammetry (CV). Freestanding electrodes of A-PCNFs/PANI-APCNFs, Pt mesh and Ag/AgCl (1M KCl) were used as working, counter and reference electrode, respectively. The

CVs at 20, 50 and 100 mV/s were obtained for both electrodes. 44

Negative electrode (Anode): The CVs for A-PCNF show nearly rectangular shape, indicating nearly ideal double layer behavior as depicted in figure 3.2 a below. The shape of the CV remains relatively unchanged when the scan rate was increased from 20 to 100 mV/s, showing excellent rate capability. Capacitance of

-1 A-PCNFs in 1M Li2SO4 was 170 F g at 20 mV/s, is very competitive compared to performance of other carbon based nanomaterials in neutral aqueous electrolyte, while allowing significantly higher electrode loading as shown in Table 3-1.

Figure 3.2: CVs for A-PCNFs (a) and PANI-coated PCNFs in a 3-electrode set up

The promising electrochemical performance of A-PCNFs is attributed to the freestanding 3-D interconnected morphology which facilitates easy accessibility of the charge storage sites. The interconnected macro-, meso- and micro-pore network within the nanofiber mat ensures effective shuttling of charges to the reaction interface and high charge storage in these nanofibers. Most of the carbon nanostructures prepared in literature are not self-supporting and require binders.47 The slurry preparation process leads to supercapacitor electrodes which lack open, through-connected macro-pore structure. The

A-PCNFs also show high overpotential for H2 evolution, with potential window extending from 0 to -1.0V compared to that of 0 to -0.6V observed for the same material in the 0.5M H2SO4 .The large stable electrochemical window of Li2SO4 based aqueous electrolyte is attributed to strong solvation energy (~160- 45

+ 2- 44 220 KJ/mol) of Li and SO4 ions. The high solvation energy requires the application of higher voltage for water decomposition, hence results in higher overpotentials for H2/O2 evolution.

Carbon based anodes Electrolytes Capacitance (F/g) Electrode loading (mg cm-2)

-1 A-PCNFs (this work) Li2SO4 170 (20 mV s ) 3.0

130 Activated carbon K2SO4 150 (5 mV/s) 2.0

124 Carbon Nanofiber Na2SO4 211 (10 mV/s) 1.0

126 HPC* Na2SO4 128 (5 mV/s) 2.0

131 Graphene hydrogel Na2SO4 158 (1 A/g) 1.0-2.0

28 HPC Li2SO4 204 ( 2 mV/s) N/A

HPC*— hierarchical porous carbon.HPC*— hierarchical porous carbon.

Positive Electrode: PANI-PCNFs electrode was also tested in a similar three-electrode set-up. A pair of redox peaks shown in figure 3.2 b is analogous to that observed in acidic aqueous electrolyte. The peak potentially represents transition from leucoemeraldine to emeraldine. The PANI-PCNFs showed a capacitance of 320 F g-1 at 20 mV/s, which compares very favorably with capacitance of 366 F g-1 obtained in acidic electrolytes.69 Note this capacitance is per unit weight of the complete electrode including both

PANI and underlying carbon nanofibers. Moreover, the PANI-PCNF exhibited high rate capability when scan rate was increased to 100 mV/s with 80% capacitance retention from 20 to 100 mV s-1. The excellent electrochemical performance is attributed to the conformal coating of polyaniline on porous carbon nanofibers, possibly due to π-π interaction between PANI chains and carbon. The intimate interaction leads to excellent charge-transfer at the PANI/carbon interface. The thinner PANI shell (~20 nm) leads to short ion-diffusion paths, and therefore maximum utilization of the polymer. The underlying porous carbon 46 nanofibers with high surface area actively contribute to energy storage via double layer mechanism eliminating dead weight from the electrodes. The porous interconnected network of the PANI-PCNFs electrodes ensures high electron conductivity as well as high accessibility of the PANI chain by the electrolyte.

PANI-PCNF reported herein shows competitive performance relative to PANI tested in both acidic and neutral media, particularly when the weight of binders is also considered.

PANI based electrodes Electrolytes Capacitance (F/g) Active electrode mass (wt %)

This work Li SO 320 (20 mV/s) 100 2 4 128 -1 PANI nanofibers KCl 176 (1 A g ) 80 PANI/sGNS/CNT20 H SO -1 85 2 4 495 (1 A g ) PANI/RGO 132 H SO -1 85 2 4 400 (0.3 A g ) 16 -1 PANI/MnO2/Graphene Na SO 85 2 4 276 (1 A g ) PANI/HPC 126 Na SO 422 (20 mV/s) 80 2 4

Full Cell: Asymmetric supercapacitor We developed a full, two-electrode cell based on A-PCNF as the anode and PANI-coated CNFs as the cathode. To maximize the available operating voltage window (OVW), the charge stored by the negative electrode was balanced by the charge stored by the positive electrode, and the mass ratio of the electrodes was calculated precisely according to the equations below.

푄 = 푀 퐶 훥푉 (10)

Since 푄+ = 푄−, 푀+ 퐶+ 훥+ = 푀−퐶− 훥푉− (11)

푀+ 퐶−훥푉− = (12) 푀− 퐶+훥푉+ 47

Where Q is the charge stored in the electrode, C is the gravimetric capacitance of the electrode based on 3- electrode testing, ΔV is the potential and M is the total mass of the electrodes. Mass ratio of m+/m-=0.67 was observed at 20 mV/s. The stable electrochemical window for A-PCNFs was -1.0 to 0 whereas PANI-

PCNFs is stable between -0.2 to 0.8. Therefore a stable electrochemical window of 1.8V was expected as shown in figure 3.3 a below.

Figure 3.3: CV of ASC cell at different voltage windows (a), Energy density vs voltage window, CVs obtained at 1.6V window at different scan rates (c), and CVs obtained at 1.8V window at different scan rates.

The CV of the asymmetric cell, A-PCNFs || PANI-PCNFs, was obtained at 50 mV/s at different voltage windows as shown in figure 3.3 b above. The ASCs exhibit good capacitive behavior with pseudo- rectangular CV plots observed over voltage window of 2.0V with no noticeable oxygen/hydrogen evolution peaks observed. The capacitance of the cell remains largely constant at different voltage window at ~51±5

F/g resulting in an energy density of 14 W h kg-1 at 1.2V to 27 W h kg-1 at 2.0V. Based on the stability windows of the PANI-PCNFs and A-PCNFs electrodes, a voltage window of 0-1.8V was chosen for further study of the ASC configuration. As shown in figure 3.3 c above, the CVs of the ASC were obtained as a function of scan rates. The CVs remain relatively rectangular at different scan rates indicating fast charge- discharge characteristics. From the CV at 1.8V, capacitance of 65 F g-1 was obtained at 20 mV/s which corresponds to energy density of 29 W h kg-1. The cell exhibit excellent rate capability, achieving capacitance of 50 F g-1 at 100 mV/s which corresponds to energy density of 23 W h kg-1 and power density 48 of 5 067 W kg-1. When electrode mass loading was increased from 3.0 to 4.6 mg cm-2, only a slight drop in energy density was observed (to 23 W h kg-1 at 20 mV/s), indicating presence of well-defined intra-/inter- fiber porosity that allows excellent permeation of electrolytes and transport within both electrodes even at higher mass loading.

Figure 3.4. Charge-discharge plot at 1 A/g (a) and energy density vs current density (b) plots for ASC at voltage windows of 0—1.6 and 0—1.8V windows.

Electrochemical performance of the ASC was also evaluated using galvanostatic charge-discharge as shown in figure 3.4. It was established that the charge-discharge curves preserves a good symmetry at voltage window of 0-1.8V indicating excellent capacitive performance. At current density of 0.5 A/g energy density of 24 W h kg-1 was achieved at 1.8V window. The ASC exhibit excellent rate capability whereby energy density of 20 W h kg-1 at a 10-fold current density of 5 A/g is reported. It should be noted that such performance was obtained at total electrode loading of 4.64 mg cm-2. When the electrode loading was lowered to ~3.0 mg which is in the range of electrode loading typically employed in literature, energy density of 29.3 W h kg-1 for the cell was obtained at 1.8V window.

3.3.1.2 In Situ infrared spectroscopy:

For the spectroelectrochemical studies, the voltage of the ASC cell (PCNF || PANI-CNFs) was varied from

-0.2 to 1.6V at a scan rate of 10 mV/s while the FTIR spectra and cyclic voltammetry were acquired simultaneously as shown in figure 5. Note that the PANI-PCNF electrode was kept in intimate contact with the diamond crystal, as shown in schematic 2 and no additional layer/film of the current collector was used to maximize FTIR signal from the electrode of interest. Prior to in-situ IR spectroelectrochemical studies, a negative potential of -0.2V was applied on the working electrode (PANI-PCNFs) to fully reduce PANI and a spectrum was acquired at this potential. Subsequent spectra during oxidation or reduction process were automatically referenced to this spectrum. Therefore the IR spectra in figure 3.5 b show only changes during the electrochemical process compared to the fully reduced state at -0.2V. A series of spectra were obtained as a function of voltage at an interval of ~38 mV (resolution of 4 cm-1 and 3 scans). When a positive potential was applied, an increase in the absorption intensity at 1566, 1305 1138, and 833 cm-1.

The increase in intensities of the vibrational absorptions at 1566, 1305, and 1138 cm-1 is attributed to the semi quinoid ring vibrations suggesting that the benzenoid rings are changed to more quinoid ring distortion

-1 structures. The band at 833 cm associated with CHout-of-plane vibrations confirms presence of semiquinoid polaron lattice structure. The band at ~1138cm-1 was attributed by Ping et al to the presence of =N+H- structure.133

Typically for acidic media, the band at ~1138 is attributed to vibrational mode of B-N+H=Q structure which is formed during the protonation process (B and Q represent benzonoid and quinoid rings, respectively).133

Increase in the intensity of this peak indicate increased level of doping. A similar phenomena is observed

+ in our Li2SO4 based electrolyte, and it is likely that Li ions are contributing to doping of the PANI chain i.e formation of B-N+Li=Q structure. When the applied potential reached 1.4V, characteristic vibration band at 1625 cm-1 associated with fully formed quinoid ring starts to appear, indicating transition to pernigraniline phase. When the applied potential is reversed, the trend reverses as well. This indicate that polyaniline undergoes the reaction mechanism in Li2SO4 similar to that in acidic media. 50

Figure 3.5. Illustrations for the situ FTIR spectra of PANI-CNFs as function of voltage applied during oxidation.

Asymmetric Capacitors Capacitance Energy density Cycling (F/g) ( W h kg-1) (Capacitance

retention, %)

-1 This work 65 (1 A g ) 29.0 ( 1.8 V) 74 (4500) RGO-RuO ||H SO ||RGO-PANI 132 -1 26.3 (1.4V) 70 (2500) 2 2 4 98 (0.3 A g ) HPC||Na SO ||HPC-PANI 126 -1 60.4 (1.8V) 91 (1000) 2 4 134 (1 A g ) GNS*||H SO ||GNS-PANI-CNTs20 -1 38.0 (1.6V) 91 (5000) 2 4 107 (1 A g ) V O nanofibers||KCl||PANI 128 -1 21.7 ( 2.0V) 73 (2000) 2 5 58 (0.2 A g )

134 -1 AC fiber||H2SO4||AC fiber-PANI 61 (0.5 A g ) 20.0 (1.6V) 90 (1000) GO||Na SO || PANI@MnO2@CC135 61 (0.2 A cm-2) 33.9 (2.0 V) 70 (5000) 2 4

GNS*—graphene nanosheet, RGO—reduced graphene oxide, CC- carbon cloth 51

The cycling performance of the A-PCNFs||PANI-PCNFs is evaluated by galvanostatic charge-discharge at a current density of 2 A/g using at a voltage window of 1.8V for 4500 cycles as demonstrated in figure 3.6.

A capacitance retention of 74% was observed for the cell with the capacitance remaining largely constant after the first 500 cycles. Such retention is superior or comparable to those reported for ASC systems based

128 132 on V2O5 nanofibers/PANI (73%) , rGO/rGO-PANI (70%) and GO/PANI@MnO2@CC system

(70%).135 The device also achieved energy density of 29 W h kg-1, which is higher than most of the values reported for acid and neutral electrolytes. Yu et al reported impressive energy density of ~60 W h kg-1 at

1.8V window for a HPC/HPC-PANI system which they attribute to hierarchical porosity of the HPC and conformal coating of PANI on HPC due to nitrogen groups.126 Their fabrication process involves extensive steps and the electrode preparation process require the use of binders. Generally, the electrochemical performance of the ASC cell designed in this work is very competitive in terms of initial performance, rate capability and durability compared to those reported in literature as demonstrated in table 3-3 above. The promising performance is thought to stem from the conformal coating of PANI nanofibers on PCNFs due to the π−π interaction which greatly enhances the utilization of PANI. From in operando infrared studies, it can be concluded that polyaniline is undergoing redox mechanisms analogous to those observed in acidic media, which is contributing to the electrochemical performance herein. The unique electrodes structure whereby both electrodes are characterized by porous 3-D architecture further increases the accessible contact area for rapid charge transport at the electrolyte/electrode interface. The hierarchal porosity on A-

PCNFs electrode ensures both high storage of the anode and high power capability of the device. Moreover, absence of binders in the system ensures that transport of charges/electrons in not interrupted. 52

Figure 3.6. Plots showing stability of the ASC cell (a) and EIS before and after cycling (b)

3.4 Conclusion.

We have demonstrated electrochemical activity of polyaniline in a neutral aqueous electrolyte characterized

by a wider voltage window than acidic electrolyte typically used for PANI. Through in operando infrared

spectroscopic studies, we have shown that the polymer is doped by Li+ in the absence of protons. To further

expand the electrochemical window of the neutral aqueous electrolyte, an asymmetric cell based on

activated porous carbon nanofiber (A-PCNF) anode and polyaniline coated porous carbon nanofiber

(PCNF) cathode was designed. The cell achieved a maximum charging voltage of 1.8V and exhibited

highest energy density of 29 W h kg-1 at electrode mass loading of 3.0 mg cm-2. Furthermore, the cell

retained 74% of its capacitance after 4500 galvanostatic cycling at 2 A g-1. The performance of polyaniline

in neutral aqueous electrolyte is promising for pairing of PANI with other electrode material with poor

performance in acidic media. Moreover, neutral aqueous electrolytes are environmentally friendly and more

conductive compared to organic electrolytes.

CHAPTER 4. Redox Mechanisms of Polyaniline in Protic Ionic Liquid

4.1 Introduction

Supercapacitors are energy storage devices which are characterized by high power density and fast charging

process but low energy density compared to rechargeable batteries. Over the last decade, there has been a

concerted effort to develop advanced supercapacitors characterized by both high specific energy and

power.32 Ionic liquids (ILs) are promising electrolyte materials for development of high energy

supercapacitors, and therefore bridging the energy gap between the conventional capacitors and

rechargeable batteries. The ILs are defined as fused organic salts with melting point typically less than

100◦C. ILs are characterized by exceptionally thermal and electrochemical stability, non-volatility, and

negligible flammability.51 The absence of solvents enable ILs to operate over a wide potential window

spanning 2 to 5 V.136 Such a wide potential window not only improves the energy storage capability but

also aids in levelling the voltage range between supercapacitors and batteries in applications where these

devices complement each other. Ionic liquids are classified into two categories: aprotic ionic liquids (AILs)

and protic ionic liquids (PILs). The key distinction between the two is that a proton transfer takes place

during the synthesis of the PILs, which leads to creation of a proton-donor site which can be used to build

a hydrogen-bonded network.50

AILs have been widely investigated for application in supercapacitors. Electrochemical performance of

carbon based nanomaterials have particularly been widely investigated in AILs, where voltage window of

2.5-3.5 is typically reported.53, 137-138 Yang and co-workers developed high energy density supercapacitor

based on graphene doped carbon electrodes and 1-ethyl-3-methylimidazolium tetrafluoroborate

139 (EMIMBF4). The supercapacitor showed outstanding electrochemical performance with reported

gravimetric capacitance of 190 F/g at 3.5 V window and energy density of 76 W h kg-1-1. Our lab has also

developed advanced supercapacitor based on AILs and binder-free porous carbon nanofibers.52 The 54 capacitor system achieved capacitance of 180 F g-1 at voltage window of 3.5V, and demonstrated energy density of 80 W h kg-1, which is comparable to that of lead acid batteries. Despite the widespread study of

AILs in carbon based systems, very few studies have investigated the performance of redox materials such as transition metal oxides (TMOs) and conducting polymers (CPs) in ionic liquids. Most pseudocapacitance studies of TMOs and CPs based supercapacitors have been limited to aqueous solutions because of the need for protons in the electrolyte to achieve the faradaic reactions leading to pseudocapacitance.63 However aqueous electrolytes can only be used on a relatively narrow potential window which limits the maximum energy stored by the device. One strategy to overcome these limitations could be to replace the aqueous electrolytes with protic ionic liquids, which possess mobile protons and can be stable on a large variety of potential windows.

Polyaniline, a CP, is promising electrode material for such study. Polyaniline is characterized by high capacitance (> 750 F/g in acidic aqueous electrolyte), good electrical conductivity, low cost, tunable synthesis and benign environmental effect.83, 140 Few studies exists where the redox activity of polyaniline has been investigated. Innis and co-workers investigated the electroactivity of polyaniline in 1-Butyl-3- methyl-imidazolium hexafluorophosphate (BMI PF6) and 1-ethyl-3-methylimidazolium bis(brifluromethanesulfon) imide (EMI TFSI).67 They observed that polyaniline became electrochemically inactive after 150 and 400 cycles in BMI PF6 and EMI TFSI, respectively. The rapid decline in electroactivity is likely due to absence of protons in the AILs systems.50

In this study, we have sought to demonstrate the feasibility of using polyaniline in protic ionic liquid and to establish if electrochemical mechanism analogous to those observed in acid aqueous electrolytes could be obtained. Electrochemical polymerization of aniline on substrate(s) was done via galvanostatic process as previously demonstrated.69 Electrochemical performance of polyaniline-coated on ITO glass and freestanding highly porous carbon nanofibers (PCFs) in 2-methlypyridine: tetrafluoroacetic acid were investigated using cyclic voltammetry. Ex situ infrared spectroscopy was further employed to elucidate the polyaniline Faradaic reaction in the novel electrolyte. The PILs showed electrochemical stability of ~2.7V 55

(-1.0 to 1.7) using a bare ITO glass. For PANI coated on ITO glass/P-CNFs, cyclic voltammetry was

obtained within a voltage range of -0.2 to 1.0V. Two pair of redox peaks analogous to those typically

observed in acidic based medium (attributed to leucoemeraldine/emeraldine and emeraldine/pernigraniline

transitions) were observed. To further corroborate redox activity observed using CV, ex-situ FTIR was

employed to ascertain oxidation state of PANI at a given potential (CV experiment was stopped before and

after each redox peak and FTIR spectra Ire collected and compared at these potentials). Analysis of the

FTIR spectra confirmed that PANI was indeed oxidized from leucomeraldine to emeraldine when potential

was varied from -0.2 to 0.3 V. However, the reaction from emeraldine to pernigraniline when potential was

varied from 0.3 to 1.0V was inconclusive from ex-situ FTIR. This is potentially due to partial reduction of

the pernigraniline before collection of the ex situ spectrum. PANI-coated on CNFs showed capacitance of

180 F/g at scan rate of 10 mV/s. However, due to high viscosity of the PILs, the electrode showed

unsatisfactory rate performance. An asymmetric cell in protic ionic liquid consisting of activated porous

carbon, A-PCFs, as anode and PANI-coated CNFs cathode was assembled. The asymmetric cell

configuration achieved a voltage window of 1.8V, a significant improvement over the 1.0V typically

observed in acid media.

4.2 Experiment Section.

4.2.1 Synthesis of PILs.

The protic ionic liquid was prepared by gradual addition trifluoroacetic acid (TFA) into 2-methylpyridine

(a-picoline) to a desired molar composition. The Bronsted acid to base reaction was carried at room

temperature under vigorous stirring. The resulting PTFA ionic liquid was then dried under vacuum at 70◦C

for 48 h to remove water, yielding a clear to lightly brown colored liquid. Formation of ionic species is

evidenced by the complete lack of smell of the liquid due to strong electrostatic interactions between the

51 protonated a-picoline and the [(CF3CO2)2H] dimer as proposed by Angell and co-workers. The protic ionic

liquid was completely sealed using paraffin film between uses. Figure 4.1 below illustrate the synthesis

process for the PILs. 56

Figure 4.1. Schematic showing the synthesis process for the PILs.

4.2.2 Synthesis of polyaniline.

Aniline from Sigma-Aldrige was dissolved in 1 M H2SO4 to obtain a solution with 0.5M aniline concentration. An ITO glass and freestanding PCFs (obtained as reported previously)129 with ~ 1 cm2 area were used as substrates for PANI deposition. The electrochemical polymerization of PANI on these substrates was carried out using a three-electrode cell with Ag/AgCl and platinum mesh as the reference and counter electrodes, respectively. For the PCFs electrode, a T-type Swagelok cell was employed for deposition and while for the ITO glass, the set up shown in figure XX was used. Electropolymerization of the polyaniline was done using galvanostatic technique at 8 mA. Desired mass loading/thickness was obtained by varying deposition time. Equation 1 below shows the formula that could be used to estimate the mass deposited.

퐼∗푡∗푀 푀푎푠푠 = 푤 (13) 푛∗퐹

Where I, t, Mw, n and F represents the deposition current (A), time (s), aniline molar mass (g/mol), consumed electrons and Faraday constant (s A/mol), respectively.

4.2.3 Characterization techniques. Characterization of the electrolyte. To establish complete proton transfer during the reaction, the NMR of the PIL was obtained. Clean spectra (figure x1) were obtained in which all peaks were attributed to the protonated base and trifluoroacetate for the freshly prepared samples. Drying was kept to a minimal time to avoid changes in the IL composition. After 24 h of thermal treatment, new peaks appeared suggesting that some protons were transferred back to the conjugated base (TFA) due to the temperature. The 1H NMR spectra showed no signals or shifts that could be attributed to water, suggesting that a significant amount 57 of water was efficiently removed by the vacuum-drying procedure. Thermal properties for the electrolyte were obtained using thermogravimetric analysis (2950 HR V5.3C TGA).

Electrochemical and spectroscopic characterization. The electrochemical stability of the electrolyte was first obtained using a three-electrode set up as shown in figure 4.2 below. Ag wire was used as a pseudo- reference electrode while Pt mesh was utilized as the counter electrode. Electrochemical activity of the polyaniline deposited on ITO glass was obtained using the same set up. For polyaniline deposited on PCFs, a T-type Swagelok set up was used for three-electrode testing while two-way Swagelok was used for a full cell set up. Fourier Transform Infrared Spectroscopy (Nicolet iS50, Thermo Fischer) was used to characterize the oxidation state of polyaniline at different oxidation state. To establish complete proton transfer during the reaction, the NMR of the PIL was obtained. Drying was kept to a minimal time to avoid changes in the IL composition. After 24 h of thermal treatment, new peaks appeared suggesting that some protons were transferred back to the conjugated base (TFA) due to the temperature. The 1H NMR spectra showed no signals or shifts that could be attributed to water, suggesting that a significant amount of water was efficiently removed by the vacuum-drying procedure.

Figure 4.2. Schematic showing three-electrode electrochemical cell. 58

4.3 Results and Discussions.

Characterization of the electrolyte. Thermal and physical properties of the protic ionic liquid were first

characterized. TG figure 4.3a shows that the polymer possess thermal stability >100◦C. The PILs show

viscosity of 9.7 mPa.s which is comparable to values previously reported for PILs and similar to those

obtained for some AILs.141-142 The electrochemical behavior of the P-TFA protic ionic liquid was firstly

studied by cyclic voltammetry bare ITO glass and graphite rod using figure 1 shown above. To minimize

contamination of the electrolyte by water contained in the air, the electrochemical process was carried out

in a chamber with continuous nitrogen flow. Figure 4.3 b shows the electrochemical stability of the protic

ionic liquid in the PILs. Voltage window of 2.7 and 3.0V was obtained using bare ITO glass and graphite

rod, respectively. The voltage window obtained for the protic ionic liquids is wider than that typically

reported for aqueous electrolytes.45 When a negative potential is applied, both ITO and graphite rod

electrodes exhibit reduction current which is likely to be due reduction of labile H+. The overpotential for

hydrogen evolution is much lower with graphite rod, possibly due rougher surface leads to higher

adsorption of the protons. Formation of bubbles was observed with increase in cathodic current. The

reaction occurring when a positive potential is applied is possibly due to oxidation of TFA and/or α-picoline

species.

Figure 4.3. TGA (a) and CV (b) plots showing thermal and electrochemical stability of the PILs. 59

Characterization of the electrolyte. Electrochemical behavior of polyaniline deposited on ITO glass was first characterized using cyclic voltammetry. The thickness of polyaniline coated on ITO glass was estimated from equation 1 above to be ~530nm. Ag/Ag+ wire was used as the quasi-reference electrode while the mesh was employed as the counter electrode. Schematic of the three-electrode set up used for this test is shown in figure 4.2 above. CV was used to investigate the electroactivity of polyaniline at room temperature with a voltage range of -0.2 to 1.0V. Two pairs of reversible redox peaks were observed as shown in figure 4.4 (a) below. The first pair of peaks was observed 0 to 0.3V at 20 mV/s, while the second peaks occurs at 0.6 to 0.8V at the same scan rate. These pair of redox peaks is analogous to those observed in acidic aqueous electrolytes although it is slightly shifted to the right, possibly due to the higher viscosity of the PILs electrolyte.114, 121 The first pair of redox peak is likely to be due to leucoemaraldine to emeraldine transition while the second pair of redox peak is attributed to the emeraldine to pernigraniline electrolyte.

Figure 4.4. CV showing the electrochemical activity of polyaniline in protic IL electrolyte at different scan rates and corresponding cycling performance over 500 cycles.

When the scan rate was increased to 100 mV/s, the cathodic and anodic peaks were only shifted slightly to the right and left, respectively, indicating excellent charge transfer within the polyaniline film. When the

PANI was cycled in the PILs, minimal fade in capacitance/electrochemical performance was observed after

500 cycles. The ability of proton in protic IL to participate in electrochemical reaction has previously been investigated by Rochefort and co-workers who demonstrated that the RuO2 demonstrated pseudocapacitive

143 behavior in protic IL similar to that in H2SO4. Prior work of polyaniline in ionic liquid have not clearly demonstrated polyaniline redox mechanisms.

To further understand the redox mechanism of polyaniline in protic ILs, IR spectroscopy was employed.

Ex-situ FTIR was used to ascertain oxidation state of PANI at a given potential (CV experiment was stopped before and after each redox peak and FTIR spectra were collected and compared at these potentials). Figure

4.5 (a) below shows different potential where electrochemical reaction was stopped and corresponding

FTIR spectra in figure 4.5(b). All of the spectra have three bands at 1566 cm-1, 1480 cm-1 and 1300 cm-1 which are attributed to quinoid-ring stretching, benzene-ring stretching and C–N stretching, respectively.

The fully reduced form of the polymer, is the simplest form since mostly phenyl rings in the benzenoid form are present. We observed that for the spectra obtained at -0.2 vs Ag/Ag+, the peak at 1480 cm-1 corresponding to C=C stretching in the benzonoid ring is dominant. When the potential was increased to -

0.3V vs Ag/Ag+, the intensity of the peak at 1480 cm-1 decreases greatly, indicating oxidation of the

-1 -1 polymer. The ratio of the peaks at 1566 cm to 1480 cm (I1566/I1480) increases to 0.88 relative to that of

0.70 observed at -0.2V, further confirming the oxidation of the polymer. When the voltage was further

+ increased to 1.0V vs Ag/Ag , there was minimal change in the intensity of the peak at 1480. The I1566/I1480 ratio was 0.98. The transition to the pernigraniline phase is inconclusive based on the IR data. There is a likelihood that pernigraniline phase was reduced by moisture in air when the sample was transferred to the

IR set up.

Figure 4.5. The electrochemical mechanism of PANI deposited on PCFs was categorized similarly.

4.4 Conclusion.

We have demonstrated electrochemical activity of polyaniline in a protic ionic liquid. The protic IL shows

a wider electrochemically stable window compared to aqueous acid electrolyte which is thermodynamically

limited to 1.0V. The presence of mobile protons in the PILs enables polyaniline to undergo the same redox

mechanisms whereby transition through the three different oxidation states is observed. Spectroscopic

characterization using FTIR confirmed clearly the transition of leucoermaldine state to emeraldine state.

However, the transition from the emeraldine to perningraniline state was not conclusive from, likely due to

reduction of the pernigraniline state in air. Polyaniline coated on carbon nanofibers show a capacitance of

180 F/g at 10 mV/s. The rate performance of the cell is unsatisfactory due to the high viscosity of the

electrolyte. Future work to lower the viscosity of the electrolyte may include addition of additives. The

promising performance of polyaniline in ionic liquid is critical in development of advanced supercapacitors

characterized by both high power delivery and energy storage capability.

CHAPTER 5. Co-electrospun Nafion and Polyaniline for Fuel Cell Electrodes Application

5.1 Introduction

Fuel cells exhibit remarkable efficiency, clean emissions and high power density, which makes them a

pertinent alternative power source for applications such as electric vehicles, small portable electronics and

distributed power generators 144. Fuel cells, however, face major setbacks due to the high cost of platinum

(Pt) catalyst 145 employed for electrochemical reactions. Consequently, to increase cost competitiveness of

fuel cells, studies have focused on reduction of Pt loading, development of Pt alloys 57, 146 and replacement

of Pt with non-precious metal catalysts 147.

To effectively reduce platinum loading, one of the key strategies to reduce Pt cost, there is a need to develop

electrode architectures that will facilitate efficient transport of protons, electrons, and reactant gases to the

catalyst active sites for enhanced catalyst (Pt) utilization. In conventional fuel cells, carbon blacks (Vulcan

XC-72, black pearl, etc.) are typically employed as Pt support materials due to their large surface area, and

high electrical conductivity 148. The electrode fabrication process entails the preparation of Pt/carbon black

(Pt/C) slurry wherein Nafion is added as both a binder and proton conductor. The slurry is then painted onto

a carbon paper via various techniques 149. Electrode preparation via slurry technique, however, is likely to

stunt catalyst accessibility due to agglomeration of carbon black nanoparticles as well as blockage of

electron conduction pathways by the ionomer. Alternative nanostructured carbon materials such as carbon

nanotubes 144, 150, and carbon nanofibers 151 have also been explored in place of carbon blacks. Although

these high aspect ratio, nanostructured carbon support materials have demonstrated slightly enhanced

electrochemical performance, their electrode preparation processes and overall electrode morphologies are

akin to those of conventional Pt/C and thus similar shortcomings are expected. Recently, effort has been

made to improve electrode fabrication process in order to further enhance triple phase transport (electron,

protons and gases) to catalyst active sites. Some examples of these novel designs include nanofiber-based

cathodes developed by Zhang and Pintauro152 via electrospinning of a blend of Pt/C and Nafion, 63 nanostructured thin films prepared by Debe et al 153 via Pt coating of organic whiskers and nanowire electrodes prepared by Alia et al 154 wherein Pt was coated onto nickel nanowires.

Electrospinning has been employed in this work to develop nanostructured electrodes that consist of well- defined co-percolating electron- and proton-conducting phases as well as excellent porosity for gas transport using Nafion, a conventional ionomer used for proton transport in fuel cells, as the proton conductor and PANI as the electron conducting material. The electrospinning technique is well documented in the literature 59, 155 and has been widely employed to fabricate polymeric nanofibers for application in the battery electrodes 156 and supercapacitor electrodes129. Both Nafion and PANI solutions have been independently electrospun in the literature. Nafion has been electrospun previously with a small fraction of carrier polymers such as PEO 157 109, 158, poly(vinyl) alcohol 58 and PVA 159. Membranes based on electrospun Nafion have demonstrated better electrochemical performance than solution cast membranes

160. PANI, another conducting polymer, has previously demonstrated good applicability as a catalyst support material 161-163 due to its high electrical conductivity, large surface area and favorable interaction with Pt

164. PANI has also been electrospun with aid of carrier polymers for application in supercapacitors 5, sensors/actuators 165 and thermos-electrics 108. However, these works on PANI electrospinning use a large fraction of the insulating carrier polymer (>20 wt %).

In this paper, we report the fabrication of nanostructured Nafion/PANI composite architecture via a single- step, simultaneous electrospinning of Nafion/PEO and PANI/PEO blend solutions. PEO was added as carrier polymer to both mixtures to enhance electrospinnability of the solutions. Both Nafion and PANI solutions are not electrospinnable by themselves due to poor dispersion of these polymers in most solvents

5, 109. The fabricated dual nanofiber mat possesses a three-dimensional, porous, interconnected non-woven network of both electron- and proton-conducting nanofibers making it suitable for application in the fuel cells electrodes where facile accessibility of the triple phase boundary is critical. Prior to preparation of the nanofiber composite mat, preliminary electrospinning and characterization studies were conducted on individual Nafion and PANI systems to understand electrospinning conditions and properties of individual 64

mats. Sole Nafion and PANI nanofibers mat were prepared with PEO content of 2 and 10 wt%, respectively.

These solutions were then electrospun separately and simultaneous onto a single rotating collector to give

dual Nafion/PANI nanofiber mats. Fabricated nanofiber mats were subjected to annealing, densification

and treatment in boiling H2SO4 and DI water. Electrochemical impedance spectroscopy (EIS) showed

presence of independent charge transfer channels in the composite fiber mat as illustrated by multiple arcs

in the Nyquist plot spectra. Nafion/PANI nanofibers mat subjected to annealing, densification and acid

treatment showed electrical and proton conductivity of 0.014 and 0.023 S/cm, respectively.

Figure 5.1. Schematic showing the dual electrospinning set up.

5.2 Experimental Method

Materials. Nafion powder was prepared from LIQUION 1115 solution purchased from Ion Power Inc. by

evaporating the water/alcohol solvents in the fume hood for 72 hours and then drying the polymer

precipitate in a convection oven at 60°C for 45-60 minutes. Polyethylene oxide (2000kDa), polyaniline

(100kDa), camphorsulphonic acid (98%), and 2-propanol (99.5%) were obtained from Sigma-Aldrich.

Chloroform (ACS grade, 99.8%) was obtained from Alfa-Aesar. 65

5.2.1 Nanofiber Fabrication

For preparation of electrospinnable Nafion solutions, Nafion powder and PEO were dissolved in 2- propanol/water mixture (2:1 wt/wt) under gentle stirring for 4-6 hours. Nafion concentration in the solution was ~42 mg/ml. For preparation of polyaniline (PANI) solutions, first PANI and camphorsulfonic acid

(CSA) (1:1.29 wt/wt ratio) were dissolved in chloroform and ultrasonicated at room temperature for 15 minutes. The ultrasonicated solution was then stirred overnight also at room temperature. PEO was then added to CSA-doped PANI and the solution was further stirred for 6 hours. The solution was then left overnight without stirring. PANI concentration in chloroform was ~10 mg/ml. For all electrospinning experiments, the distance between the tip of the needle (22 gauge needle spinneret from Hamilton

Company) and the grounded collector was fixed at 5-6”, the applied voltage was kept at 5-7 kV and the relative humidity was maintained below 20% to obtain a stable Taylor cone. The solution flow rates for

Nafion and PANI streams were ~0.30 and 1.25 ml/hr, respectively. While the individual Nafion and PANI nanofiber mats were collected on a planar substrate, a cylindrical drum collector, rotating at 60 rpm, was employed for the dual system. A schematic of the dual electrospinning set up is shown in figure 1. For ease of reference in the subsequent discussions, as electrospun Nafion/PEO (98/2 wt %), PANI/PEO (90/10 wt%) and Nafion/PANI/PEO (47.5/47.5/5) nanofiber mats will be referred to as Nafion-98, PANI-90 and

Nafion-PANI, respectively.

The fabricated samples were subjected to post electrospinning vacuum annealing at 150°C for 5 minutes, followed by compression at ~500 psi and finally treatment in boiling acid (1M H2SO4) for 1 h and in DI water for another 1 h. For ease of reference, prefix AN, C and T will be added to the above defined sample nomenclature to indicate annealed, compressed, and acid/water treated samples. Combination such as AN/C will refer to samples both annealed and compressed.

5.2.2 Nanofiber Characterization

Scanning electron microscopy (SEM) (Zeiss Supra 50VP) was used to study the external morphology of as-electrospun and processed Nafion, PANI and Nafion/PANI nanofiber mats. Fourier transform infrared 66 spectroscopy (FTIR) (Varian Excalibur FTS-3000, range of 4000-800 cm-1), X-ray Diffraction (XRD)

(Rigaku SmartLab, X-ray diffractometer, Cu Kα, scanning range 5-40° and step size of 0.02°) and thermogravimetric analysis (TGA Q50 analyzer, range 20-800°C with step size of 10°C in argon atmosphere) were used to probe the chemical structure, crystallinity and composition/thermal stability of the nanofiber samples. XRD Peaks were fitted using Pearson VII distribution function in OriginPro software.

5.2.3 Water uptake and Conductivity

Weights of dry nanofiber mats were first measured using mass balance (Ohaus DV215CD). The dry samples were then soaked in milli-Q water (18.2 MΩ.cm) for 48hrs. Samples were removed from water bath, quickly blotted with tissue paper to remove surface water and weighed again. Mass swelling was then calculated as follow;

푊 −푊 Water uptake= 푤푒푡 푑푟푦 ∗ 100% (14) 푊푑푟푦

where w denotes the sample weight. In-plane conductivity (σ=) of the samples was studied via electrochemical impedance spectroscopy (EIS) where a four-point probe conductivity cell made of a Teflon block with four platinum wires mounted on it was used. Prior to impedance studies, the samples (with exception of PANI-90) were soaked in distilled water at room temperature for 48 hours. Gamry 3000 reference potentiostat was used to generate impedance spectra. Expression 1 below was employed to compute sample conductivity.

퐿 휎 = (14) 푅∗푤∗푡 where σ is the sample conductivity (S/cm), L is the distance between the inner probes of a 4 probe cell configuration, w and t represents the sample width and thickness respectively. Through plane conductivity

(σ┴) of the annealed, compressed and treated nanofibers mat was also established using stainless steel 67

current collectors (diameter 1.26cm). For through plane, L in expression 1 above represent thickness of the

nanofiber mat while (w*t) is the area of the current collector.

5.3 Results and discussion

5.3.1 Morphology of electrospun nanofibers

SEM was used to probe the morphology of electrospun Nafion-98, PANI-90 and Nafion-PANI nanofiber

mats. Figure 5.2 (a) shows the SEM micrograph of as electrospun Nafion-98 fiber mat. Nanofibers exhibited

smooth morphologies without any beads. The bead-free morphology is attributed to the presence of PEO

which is thought to impart the necessary entanglement in the polymer blend solution and therefore enhance

its viscoelasticity for better electrospinnability 158. The diameter of the electrospun nanofibers was in the

range of 200-400nm. Nafion-98 nanofiber mats were subjected to post electrospinning annealing at 150°C

under vacuum for 5 minutes and densified at 500 psi as shown in figure 5.2(b). Annealing was observed to

cause welding of intersecting Nafion nanofibers, a phenomena attributed to Nafion plasticization above its

glass transition temperature of ~107°C 166.

Figure 5.2. SEM micrograph of as electrospun Nafion-98 (a) and AN/C Nafion-98 (b)

PANI is a promising conducting polymer for applications in electrodes for energy devices. PANI is an extremely difficult polymer to process due to its poor solubility/dispersion in most organic solvents.

Moreover, PANI solutions lack the requisite viscosity for electrospinning attributed to its shorter and rigid polymer chains. Some work has been reported on electrospinning of PANI by blending it with a carrier polymer such as PEO which alleviates the lack of chain entanglement issue. MacDiarmid et al 104 reported fabrication of PANI/PEO blend with w/w composition of 50/50. Most of the PANI nanofibers reported, however, have high content of the non-conductive carrier polymers (>20 wt%) 103, 167.

In this work, we report fabrication of PANI nanofibers with low PEO content of only 10 wt% by use of a relatively higher molecular weight of PEO (2000 kDa) . PANI purchased in the emaraldine-base form was first doped with camphorsulfonic acid to protonate the imine groups. Emaraldine PANI exists in a semi- reduced/oxidized state and its protonation with aqueous acids have been shown to increase its conductivity remarkably due to increased π-delocalization of the polymer backbone.62-63, 168 Even though dispersion of

PANI in chloroform as high as 2 wt% has been reported 104, we observed that it was difficult to electrospin

PANI solution with more than 1 wt% in chloroform. As a result, chloroform solution with 0.67 wt% PANI concentration was used for electrospinning in this work. SEM micrograph in figure 5.3 shows the PANI-

90 nanofibers morphology characterized by nanofibers that are free of beads. The nanofibers diameter were in 506±115 nm (calculated using imageJ software for 100 nanofibers). The electrospun PANI-90 nanofibers were also subjected to annealing at 150°C under vacuum for 5 mins. From SEM micrographs, there were no observable morphological changes on PANI-90 nanofibers after annealing. 69

Figure 5.3. SEM micrograph of as electrospun PANI-90.

Fabrication of the dual Nafion/PANI system entailed co-electrospinning of Nafion/PEO (98/2 wt%) and

PANI/PEO (90/10wt %). For proper intertwining of Nafion and PANI nanofibers, a rotating drum substrate collector was used. Low rotational speed of the motor (60 rpm) was necessary since fibers have been known to align at higher speeds 108, 169. For the targeted application, it is critical to achieve a random orientation of nanofibers to maintain a three-dimensional non-woven structure to facilitate gas diffusion and for formation of triple phase surfaces. SEM micrograph in figure 5.4 (a) shows the nanostructured fiber morphology of the Nafion-PANI system. The nanofibers exhibit an excellent smooth morphology. SEM EDS shown in figure 5.5 below was employed to map the distribution of fluorine and nitrogen, elements unique to Nafion and PANI respectively. 70

Figure 5.4. SEM micrograph of as electrospun Nafion-PANI (a) and AN Nafion-PANI (b) AN/C Nafion-

PANI (c) and AN/C/T Nafion-PANI (d).

The images show even distribution of both phases. The nanofiber mats were also subjected to post electrospinning annealing at 150°C under vacuum. Figure 5.4(b) shows the morphology of annealed dual fiber mat. Nanofibers are observed to be intersecting/welding, an aforementioned phenomena encountered when annealing Nafion-98 nanofibers. Figure 5.4 (c) shows annealed Nafion-PANI nanofiber mat further densified at 500 psi. Densification of the nanofibers was aimed at increasing inter-fiber contact as well as well as reducing the void volume. Moreover, membrane electrode assembly fabrication process entails the application of such compression conditions144, 170 and therefore this treatment was necessary to gauge the ability of the fabricated nanofiber mats to withstand such conditions. Figure 5.4 (d) shows the Nafion-PANI nanofibers both annealed, compressed and treated in sulfuric acid as discussed in section 2. Solution cast

Nafion membranes are typically subjected to pretreatment for 1 hr in boiling H2SO4 and 1 hr in boiling DI water. The pretreatment process is aimed at enhancing protonation and hydration of Nafion’s sulfonic groups 171. The electrospun Nafion-PANI fiber mat was subjected to similar treatment after its annealing.

This treatment process additionally aids in selective dissolution of PEO 172. The Nafion-PANI nanofiber 71 morphology is retained, indicating presence of continuous phases of both Nafion and PANI in their respective blends with PEO.

Figure 5.5. SEM image for dual Nafion-PANI nanofiber mat (top) and corresponding EDS mapping showing Nafion fluorine distribution (F) and polyaniline nitrogen distribution (N). Top SEM images shows.

5.3.2 XRD and FTIR characterization

Figure 5.6. XRD spectrum of Nafion-212 membrane and Nafion-98 nanofiber samples (a), de-convoluted peaks for Nafion-212 membrane (b) and AN Nafion-98 nanofibers (5 mins) (c). Peaks were fitted using Pearson VII distribution function in OriginPro software.

Figure 5.6 shows the XRD spectra for as-electrospun and annealed Nafion-98 while figure 5.7 shows XRD pattern for PANI-90 and Nafion-PANI. The XRD spectrum of commercial Nafion-212 is provided for reference purposes. Crystallinity of Nafion is essential in controlling properties such as density, hardness/brittleness, and coefficient of expansion 173. 73

Figure 5.7. XRD PANI and dual Nafion-PANI nanofibers

A correlation between Nafion crystallinity and its conductivity has also been observed 169. For Nafion, the

X-ray beam is diffracted by the crystalline region according to Bragg’s law and diffuse scattered by the amorphous region. While the Nafion crystalline phase is represented by an XRD peak at an angle of ~18°, the amorphous phase is observed as a broad halo at a diffraction angle of ~16.5° due to diffuse scattering of the incident X-ray beam 166, 173. As shown in figure 5.8, the XRD pattern of Nafion-212 membrane shows a crystalline peak at a diffraction angle of ~17.5 whereas for the as-electrospun Nafion-98 sample, we largely observe a broad peak around ~17° where the peak due to crystallinity isn’t as clear as that observed for Nafion-212 possibly due to large interference from the scattering of x-ray beam by the amorphous region. Upon annealing for 5 minutes, we see a slight growth of the peak close to 17.5° which we believe indicates increase in crystallinity of Nafion nanofibers. Further annealing of the electrospun nanofibers for

2h at the same temperature led to further growth of the peak and an XRD spectrum much similar to that of

Nafion-212. The XRD peaks have been fitted using Pearson VII distribution function in OriginPro software. 74

Ballengee et al 166 observed that longer annealing (up to 2hrs) of electrospun Nafion nanofibers led to further growth in crystalline peak. However, they observed a drop in conductivity for an annealing time greater than 15 mins. PANI-90 XRD peaks were at 2θ= ~15°, 20° and 25°. Characteristic PANI peaks at

2θ= ~15° and 2θ= ~25° are attributed to the periodicity of the repeat unit of PANI chain and periodicity parallel to the polymer chain backbone 108. An overlap is observed between PANI peak at 2θ= ~15° and

Nafion crystallinity peak at 16-18°, thus the extent of Nafion crystallinity in the dual mat is difficulty to decipher without any ambiguity.

Figure 5.8. FTIR spectra of Nafion, PANI and dual Nafion-PANI nanofibers.

FTIR absorbance spectra shown in figure 5.8 were recorded in the 800 to 4000 cm-1 range at 25°C to investigate the chemical structure of electrospun nanofibers. Figure 5.8 shows the FTIR spectra for dry

Nafion-98 (a), PANI-90 (b) and Nafion-PANI (b) nanofibers. FTIR spectra of Nafion-212 was also obtained for reference. Characteristic Nafion peaks at; 1145 and 1200 cm-1 associated with symmetric and asymmetric stretching of –CF2 groups, respectively, 980 (stretching mode of C-O-C group), 1060 (vibration

-1 174-176 of –SO3 group) and 1410 cm (vibration of S=O) were observed for both Nafion-212 and the Nafion nanofiber mats. Weak water bands around 3460 and 1600-1700 cm-1 174 were also observed in all systems indicating absorption of moisture from the atmosphere. No perceptible shift was observed in the location 75 of typical Nafion bands for both as electrospun and annealed samples. The presence of PEO in a Nafion matrix is thought to lead to interaction between the ether oxygen of PEO and H atom of the Nafion sulfonic group via hydrogen bonding 177. The IR signal from such interaction is expected to overlap with the water bands at around 3500 cm-1. To better understand this, we compared the FTIR peak intensities (at 3460 cm-

1 ) of Nafion-98 with those of commercial Nafion-212. FTIR intensity ratio, I/Io, where I is the intensity of

Nafion-98 and Io is the intensity of Nafion-212 was calculated to be ~0.26, which can be corrected to 1.13 to account for the low Nafion fraction (20-25%) in fiber mats compared to films. I/Io ratio of greater than 1 could thus be potentially due to PEO interaction with Nafion sulfonic, group which might impede Nafion conductivity. FTIR spectra of dual Nafion-PANI nanofiber mat shows characteristics PANI and Nafion

-1 peaks. The peak at 1060 cm associated with symmetric stretching mode of -SO3 group is slightly shifted to 1050 cm-1. The shift could be due to the induced polarization of the S-O dipole by the electrostatic field of the adjacent protonated imine group of PANI 178.

5.3.3 Impedance study

EIS was employed to characterize the impedance of the nanofiber mats. Prior to obtaining the EIS data, fiber mats were equilibrated in distilled water for at least 48 h. The proton conductivity of Nafion occurs through dissociation of protons from –SO3H ionic clusters. These protons are then transported across water

− 159 and –SO3 ionic ﬁxed groups and thus the necessity of hydration of the fiber mats. 76

Figure 5.9. Nyquist plot of co-electrospun Nafion-PANI and fitted data.

Table 5.1 shows the conductivity summary of the prepared nanofiber mats. Impedance data for

Nafion-98 and PANI-90 mats was obtained from Nyquist plots. As electrospun Nafion-98 showed a proton conductivity of 0.021 S/cm. Thus the corrected conductivity of Nafion-98 is approximately 0.091 S/cm.

This value compares favorably with that of pristine Nafion film which has proton conductivity of 0.10 S/cm

171. PANI-90 showed an electrical conductivity of 0.078 S/cm from EIS spectra. This value was confirmed through dc polarization experiments where a small dc potential was applied and the corresponding steady current was recorded. The conductivity value recorded is comparable to that reported by MacDiarmid et al

104 of 0.10 S/cm, where they electrospun a PANI-PEO at 50/50wt%, but obtained fibers with diameters in the micron scale. Figure 5.9 shows the experimental and fitted impedance data for Nafion-PANI dual nanofiber mats. The Nyquist spectrum is characterized by two arcs, representing the parallel combination of ionic impedance and geometric capacitance at higher frequency, and electronic impedance and geometric capacitance at lower frequency 179. Equivalent circuits (discussed below) were used to model the EIS experimental data to decouple the protonic and electronic impedance values. As electrospun Nafion-PANI 77 nanofibers showed proton and electron conductivity of 0.0078 and 0.0046 S/cm, respectively. Both electrical and proton conductivity are reduced in the dual nanofiber system compared to that of individual

PANI-90 and Nafion-98 systems due to the lower fraction of both the PANI and Nafion phase in the dual fiber mat.

Annealing was observed to improve the conductivity of all samples with the exception of PANI-90 nanofiber mat (see table 5.1). Conductivity of annealed Nafion-98 was 0.026 S/cm, which represents a

~20% increase from that of as-electrospun mat. Annealing of Nafion-PANI also resulted in a similar enhancement of proton conductivity and about 9% increase in electrical conductivity. It was experimentally observed that annealing of nanofibers resulted in increase in the density of Nafion-based nanofibers implying increased compactness of the nanofibers (table S-1). Nafion nanofibers were also seen to weld together (also see AN Nafion-98 SEM micrographs) from SEM images upon annealing, which possibly resulted in improved fiber-fiber contact and higher charge transport. Indeed densification of nanofibers at

500 psi led to increase in both electrical and proton conductivity. Ballengee et al 166 observed similar effect on proton conductivity upon annealing of Nafion nanofibers for less than 15 mins at 150°C which they attributed to increased water uptake of the nanofibers. However, the water uptake for our annealed electrospun nanofibers is less than that of as-electrospun nanofibers. Annealing was also observed to increase the extent of crystallinity especially for Nafion-98 as shown in figure 5 (a). Increase in ordered structure of annealed materials could also contribute to the elevation of proton conductivity. Annealed and compressed Nafion-PANI nanofibers were subjected to treatment in boiling 1M H2SO4 for 1h and DI water for 1h. The treatment was intended to enhance hydration and protonation of the sulfonic group of Nafion.

The treatment process has also been to shown to aid PEO removal from Nafion nanofibers 172. Presence of insulating PEO in Nafion matrix is expected to suppress proton conductivity. Furthermore, PEO is chemically unstable in the presence of peroxides/hydroxides. It was observed that the annealing, compression and treatment of dual Nafion nanofibers in acidic/water media led to a significant increase in both proton (three-fold) and electrical conductivity (two-fold). 78

Since through plane conductivity set up gives a true representation of the dominant pathway travelled by the electrons and protons during fuel cell operation, we measured the through plane conductivity of the final dual nanofiber mat. For AN/C/T Nafion-PANI nanofiber mat, through plane proton and electron conductivities were measured as 0.018 and 0.010 S/cm resulting in a σ=/σ┴ ratio of 1.4 and 1.27, respectively. This anisotropy in the conductivity of our electrospun nanofibers can be hypothesized to be due to polymers’ molecular/ionic domain aligning along the fiber axis as a result of extensional strain during electrospinning process 109. Thus in-plane conductivity is slightly favored. The slight increase in through plane resistance might also be attributed to interference from mat/current collector resistance which is known to be more pronounced for through plane due to a large membrane/current collector contact area 180

Table 5-1 Conductivity and water uptake of electrospun nanofiber mats.

Sample Proton conductivity Electron Conductivity Water Uptake

(S/cm) (S/cm) (Δwt %)

PANI-90 0.0780 83±1.2

AN PANI-90 0.0760 81±2.3

Nafion-98 0.0210 40±3.2

AN Nafion-98 0.0260 36±2.7

Nafion-PANI 0.0078 0.0043 65±2.5

AN Nafion-PANI 0.0094 0.0050 56±2.1

AN/C/T Nafion-PANI 0.0230 0.0140 52±2.8

For AN/C/T, through plane set up gave conductivity of 0.018 and 0.010 S/cm for proton and electrons, respectively. 79

Figure 5.10. Equivalent circuit used to decouple electrical conductivity from proton conductivity.

Figure 5.10 shows the equivalent circuit used to decouple the proton and electron conductivity values provided above from the measured impedance data. The circuit was built using a Gamry 3000 reference model builder. The dual nanofiber mats are expected to contain co-percolating channels for electron and proton flow, thus a parallel combination of electronic (Ze) and protonic (represented as Zp) impedance units is proposed, both in parallel with the overall geometric capacitance, Cb. The capacitor elements are modeled as non-ideal capacitors 181. Since the electrodes used (Pt electrodes) are expected to

180, 182 be blocking to ionic flow at the sample/Pt electrode contact interface , a capacitive element, Cint , is added in series to the ionic impedance element. Any impedance to electron flow at the sample/Pt electrode contact is treated as negligible. There is also an initial resistance observed for the nanofiber mats. The resistance is represented as, Rs, and can be attributed to the resistance due to connecting wires. 80

5.4 Conclusion

Composite Nafion/polyaniline nanofiber mats were fabricated and studied with potential application as

catalyst supports in fuel cells electrodes. The electrospun dual nanofiber mats were subjected to post

electrospinning processing— annealing, compression and treatment in acidic/DI water media—in order to

improve its conductivity, mechanical strength and chemical stability. Electrochemical impedance

spectroscopy (EIS) was used to study conductivity properties of the dual fiber mats and an equivalent circuit

was used to decouple the proton and electron conductivities. EIS spectra was characterized by two semi-

circles indicating that the composite fiber mat possessed percolating channels for both electron and proton

conduction, a requisite property for fuel cell catalyst supports. The annealed/compressed/treated dual fiber

mat exhibited a proton and electron conductivity of up to 0.023 and 0.014 S/cm respectively. The fabricated

fiber mat with the presence of percolating channels for electron and proton conduction and three-

dimensional inter-fiber macrospores (for gas and water transport) can serve as an efficient freestanding

catalyst support for PEMFCs. Future work on incorporation of CNTs within the PANI nanofibers to bolster

its mechanical stability/ electrical conductivity as well introduction of platinum/catalyst nanoparticles via

electrodeposition on a freestanding nanofiber mat is underway.

CHAPTER 6. Summary and Future Outlook

6.1 Summary

This work demonstrate synthesis of polyaniline nanomaterials and subsequent characterization via various

analytical/ectrochemical techniques for development of high performance energy storage and conversion

devices (supercapacitors and fuel cells). Polyaniline based nanofiber electrodes fabricated in this work are

free-standing with a continuous interconnected network providing fast ion diffusion as well as fast transport

of electrons within the network, a characteristic essential for efficient energy storage. This unique

characteristic of the work distinguishes it from other studies which are mostly based on powder-based

materials as well as some binder-free materials (synthesized by techniques such as vacuum filtration) which

do not have an interconnected network and utilize high loading to establish good contact between the active

material surfaces. The key conclusions of this dissertation are summarized below:

 Electrospinning technique was employed to fabricate binder-free, porous 3-D polyaniline nanofibers

characterized by unprecedented high PANI loading in the fibers (93 wt%). The electrospun PANI based

electrodes show competitive electrochemical performance, including high initial capacitance (385 F g-1),

impressive rate capability and good life cycle performance (~81% capacitance retention over 1000 cycles).

The capacitance obtained represents ~40% increase relative to previously reported values for similarly

prepared PANI electrodes.5 The promising electrochemical performance of our as-electrospun freestanding

electrodes stems from its porous three-dimensional nonwoven nanofiber mat morphology. The inter-fiber

porosity enables quick ion transport and diffusion within the electrodes whereas the interconnected

nanofiber network facilitates uninterrupted electron conductivity. The standard electrode assembly

process—utilization of slurries—in most literature is likely to result in a lack of such an open, through-

connected pore structure. 82

 Adopted neutral aqueous electrolytes and designed asymmetric polyaniline based supercapacitor. We have

demonstrated that polyaniline display electrochemical process in Li2SO4 based electrolyte similar to that

observed in acidic electrolyte. The neutral aqueous electrolyte is much stable relative to acidic based

electrolytes due to strong solvation energy involved in the former. We have demonstrated that a voltage

window of 1.8V could be achieved in such cell with energy storage capacity of 29 W h kg-1. Such energy

density is twice that reported for conventional supercapacitors. The higher energy density is enabled by

pseudo-doping of polyaniline by Li+.

 Demonstrated feasibility of using protic ionic liquid for polyaniline based supercapacitors. Acidic based

electrolytes have primarily been used due to the need of doping of PANI. However, acidic electrolyte are

thermodynamically limited to ~1.0V. We have shown that PIL based on α-picoline and trifluoroacetic acid

possesses labile protons which can be involved in the redox mechanisms. Such PIL show electrochemical

stability of up to 3.0V. Asymmetric cell based on polyaniline cathode and other redox materials anodes can

therefore be explored to take advantage of the large electrochemical window.

 Potential catalyst support system based on co-electrospun polyaniline and Nafion has also been developed.

Such system is characterized by co-percolating network of electron and proton conducting pathways which

enable efficient shuttling of charges to the catalyst site. The high inter-fiber porosity exhibited by the co-

electrospun mat will enable unhindered transport of gaseous reactants to catalytic sites.

6.1 Future Outloook

As demonstrated in this work, the use of electrospinning technique is a promising route to prepare

polyaniline nanomaterials with optimized architecture and devoid of electrochemically inactive additives

(binders). Such architecture allow study of polyaniline in wide array of electrolyte systems. However, to

further advance the work reported herein, there are a few challenges that need to be addressed to meet the

goals of developing low cost, high energy electrochemical devices.

 Recent work by Vonlanthen 90 et al demonstrated that addition of redox additives such as quinone to

aqueous electrolytes could boost electrochemical performance and stability of polyaniline nanomaterials.

Therefore studies of electrospun polyaniline with optimal architecture could be explored in such media. In

particular, redox active additives could be introduced in the neutral aqueous electrolyte which has

demonstrated higher electrochemical window than acid electrolyte employed by Vonlanthen et al.

 Performance of polyaniline in neutral aqueous electrolytes is significance since it opens up opportunities to

pair polyaniline with other redox materials that normally do not work well in acidic media. Asymmetric

supercapacitor based on transition metal oxides anode and polyaniline cathode in neutral electrolytes are

promising for development of high energy density supercapacitor.

 To further develop good understanding of the redox activity of polyaniline in protic ionic liquid, there is

need to conduct in situ studies using techniques such as IR and Raman spectroscopy. This work was initiated

but faced obstacles due to acid intolerance of the stainless steel material used on FTIR puck. In situ studies

will further elucidate clearly mechanisms such as intercalations of cations and consumptions of anions, and

how such processes may affect the polymer structure/stability. 84

 The co-electrospun polyaniline/Nafion network has shown promise as catalyst system by preliminary

impedance studies. Further work will involve deposition of catalysts/Pt on such support system via

processes such as electrodeposition and evaluating electrochemical performance in a fuel cell stack.

 The electrospun polyaniline nanofibers could also find application in other field such as developing

sensors, membrane and actuators.

LIST OF REFERENCES

1. Wang, K.; Wu, H.; Meng, Y.; Wei, Z., Conducting polymer nanowire arrays for high performance supercapacitors. Small 2014, 10, 14-31.

2. Li, H.; Wang, J.; Chu, Q.; Wang, Z.; Zhang, F.; Wang, S., Theoretical and experimental specific capacitance of polyaniline in sulfuric acid. J Power Sources 2009, 190, 578-586.

3. Greiner, A.; Wendorff, J. H., Electrospinning: A Fascinating Method for the Preparation of Ultrathin Fibers. Angew. Chem. Int. Ed. 2007, 46, 5670 – 5703.

4. Béguin, F.; Presser, V.; Balducci, A.; Frackowiak, E., Carbons and electrolytes for advanced supercapacitors. Adv. Mater. 2014, 26, 2219-2251.

5. Chaudhari, S.; Sharma, Y.; Archana, P. S.; Rajan Jose; Ramakrishna, S.; Mhaisalkar, S.;

Srinivasan, M., Electrospun Polyaniline Nanofibers Web Electrodes for Supercapacitors. Journal of Applied Polymer Science 2013, 129, 1660-1667.

6. Bruce, P. G.; Freunberger, S. A.; Hardwick, L. J.; Tarascon, J. M., Li-O2 and Li-S batteries with high energy storage. Nat Mater 2012, 11, 19-29.

7. Cao, Y.; Xiao, L.; Sushko, M. L.; Wang, W.; Schwenzer, B.; Xiao, J.; Nie, Z.; Saraf, L.

V.; Yang, Z.; Liu, J., Sodium ion insertion in hollow carbon nanowires for battery applications.

Nano letters 2012, 12, 3783-7.

8. Choi, N. S.; Chen, Z.; Freunberger, S. A.; Ji, X.; Sun, Y. K.; Amine, K.; Yushin, G.;

Nazar, L. F.; Cho, J.; Bruce, P. G., Challenges facing lithium batteries and electrical double-layer capacitors. Angewandte Chemie 2012, 51, 9994-10024.

9. Sharma, P.; Bhatti, T. S., A review on electrochemical double-layer capacitors. Energy

Conversion and Management 2010, 51, 2901-2912. 86

10. Conway, B. E.; Pell, W. G., Double-layer and pseudocapacitance types of electrochemical capacitors and their applications to the development of hybrid devices. Journal of Solid State Electrochemistry 2003, 7, 637-644.

11. Conway, B. E.; Birss, V.; Wojtowicz, J., The role and utilization of pseudocapacitance for energy storage by supercapacitors. J. Power Sources 1997, 66, 1-14.

12. Reddy, M. V.; Subba Rao, G. V.; Chowdari, B. V., Metal oxides and oxysalts as anode materials for Li ion batteries. Chem Rev 2013, 113, 5364-457.

13. Santino, L. M.; Lu, Y.; Acharya, S.; Bloom, L.; Cotton, D.; Wayne, A.; D'Arcy, J. M.,

Enhancing Cycling Stability of Aqueous Polyaniline Electrochemical Capacitors. ACS applied materials & interfaces 2016, 8, 29452-29460.

14. Simon, P.; Gogotsi., Y., Material Electrochemistry for Supercapacitors. Nat. Materials.

2008, 7, 845.

15. Zheng, X.; Luo, J.; Lv, W.; Wang, D. W.; Yang, Q. H., Two-Dimensional Porous

Carbon: Synthesis and Ion-Transport Properties. Adv Mater 2015, 27, 5388-95.

16. Huang, W.; Yao, T.; Hao, Q.; Wang, W.; Xia, X.; Wang, X., Synthesis and electrochemical properties of graphene oxide/manganese oxide/polyaniline and its reduced composites. RSC Adv. 2014, 4, 56615-56624.

17. Sumboja, A.; Foo, C. Y.; Yan, J.; Yan, C.; Gupta, R. K.; Lee, P. S., Significant electrochemical stability of manganese dioxide/polyaniline coaxial nanowires by self-terminated double surfactant polymerization for pseudocapacitor electrode. Journal of Materials Chemistry

2012, 22, 23921.

18. Meng, F.; Zhao, J.; Ye, Y.; Zhang, X.; Li, Q., Carbon nanotube fibers for electrochemical applications: effect of enhanced interfaces by an acid treatment. Nanoscale 2012, 4, 7464-8. 87

19. Meng, Y.; Wang, K.; Zhang, Y.; Wei, Z., Hierarchical porous graphene/polyaniline composite film with superior rate performance for flexible supercapacitors. Adv Mater 2013, 25,

6985-90.

20. Shen, J.; Yang, C.; Li, X.; Wang, G., High-performance asymmetric supercapacitor based on nanoarchitectured polyaniline/graphene/carbon nanotube and activated graphene electrodes.

ACS applied materials & interfaces 2013, 5, 8467-76.

21. Kang, Y. J.; Chung, H.; Han, C.-H.; Kim, W., All-solid-state flexible supercapacitors based on papers coated with carbon nanotubes and ionic-liquid-based gel electrolytes.

Nanotechnology 2012, 23, 1-5.

22. Le Bideau, J.; Viau, L.; Vioux, A., Ionogels, ionic liquid based hybrid materials. Chem.

Soc. Rev. 2011, 40, 907-925.

23. Kumar, Y.; Pandey, G. P.; Hashmi, S. A., Gel Polymer Electrolyte Based Electrical

Double Layer Capacitors: Comparative Study with Multiwalled Carbon Nanotubes and

Activated Carbon Electrodes. The Journal of Physical Chemistry C 2012, 116, 26118-26127.

24. Zhang, J.; Zhao, Z.; Xia, Z.; Dai, L., A metal-free bifunctional electrocatalyst for oxygen reduction and oxygen evolution reactions. Nature nanotechnology 2015, 10, 444-52.

25. Liu, Z.; Fu, D.; Liu, F.; Han, G.; Liu, C.; Chang, Y.; Xiao, Y.; Li, M.; Li, S., Mesoporous carbon nanofibers with large cage-like pores activated by tin dioxide and their use in supercapacitor and catalyst support. Carbon 2014, 70, 295-307.

26. Su, D. S.; Chen, X.; Weinberg, G.; Klein-Hofmann, A.; Timpe, O.; Abd Hamid, S. B.;

Schlogl, R., Hierarchically structured carbon: synthesis of carbon nanofibers nested inside or immobilized onto modified activated carbon. Angewandte Chemie 2005, 44, 5488-92. 88

27. Dutta, S.; Bhaumik, A.; Wu, K. C. W., Hierarchically porous carbon derived from polymers and biomass: effect of interconnected pores on energy applications. Energy Environ.

Sci. 2014, 7, 3574-3592.

28. Hao, P.; Zhao, Z.; Tian, J.; Li, H.; Sang, Y.; Yu, G.; Cai, H.; Liu, H.; Wong, C. P.; Umar,

A., Hierarchical porous carbon aerogel derived from bagasse for high performance supercapacitor electrode. Nanoscale 2014, 6, 12120-9.

29. Cheng, P.; Li, T.; Yu, H.; Zhi, L.; Liu, Z.; Lei, Z., Biomass-Derived Carbon Fiber

Aerogel as a Binder-Free Electrode for High-Rate Supercapacitors. The Journal of Physical

Chemistry C 2016, 120, 2079-2086.

30. Zhao, X.; Sanchez, B. M.; Dobson, P. J.; Grant, P. S., The role of nanomaterials in redox- based supercapacitors for next generation energy storage devices. Nanoscale 2011, 3, 839-55.

31. Kuila, B. K.; Nandan, B.; Bohme, M.; Janke, A.; Stamm, M., Vertically oriented arrays of polyaniline nanorods and their super electrochemical properties. Chem Commun (Camb)

2009, 5749-51.

32. Ramya, R.; Sivasubramanian, R.; Sangaranarayanan, M. V., Conducting polymers-based electrochemical supercapacitors—Progress and prospects. Electrochimica Acta 2013, 101, 109-

129.

33. Gawli, Y.; Banerjee, A.; Dhakras, D.; Deo, M.; Bulani, D.; Wadgaonkar, P.; Shelke, M.;

Ogale, S., 3D Polyaniline Architecture by Concurrent Inorganic and Organic Acid Doping for

Superior and Robust High Rate Supercapacitor Performance. Sci Reports 2016, 6, 21002.

34. Liu, T.; Finn, L.; Yu, M.; Wang, H.; Zhai, T.; Lu, X.; Tong, Y.; Li, Y., Polyaniline and polypyrrole pseudocapacitor electrodes with excellent cycling stability. Nano letters 2014, 14,

2522-7. 89

35. Simotwo, S. K.; Kalra, V., Polyaniline-based electrodes: recent application in supercapacitors and next generation rechargeable batteries. Current Opinion in Chemical

Engineering 2016, 13, 150-160.

36. Pieta, P.; Obraztsov, I.; D'Souza, F.; Kutner, W., Composites of Conducting Polymers and Various Carbon Nanostructures for Electrochemical Supercapacitors. ECS J Solid State Sci

Techn 2013, 2, M3120-M3134.

37. Lang, X.; Hirata, A.; Fujita, T.; Chen, M., Nanoporous metal/oxide hybrid electrodes for electrochemical supercapacitors. Nature nanotechnology 2011, 6, 232-6.

38. Huang, X.; Sun, B.; Chen, S.; Wang, G., Self-assembling synthesis of free-standing nanoporous graphene-transition-metal oxide flexible electrodes for high-performance lithium-ion batteries and supercapacitors. Chemistry, an Asian journal 2014, 9, 206-11.

39. Chi-Chang Hu, K.-H. C., Ming-Champ Lin, Yung-Tai Wu, Design and Tailoring of the

Nanotubular Arrayed Architecture of Hydrous RuO2 for Next Generation Supercapacitors. ACS

Nano Lett. 2006, 6, 2690-2695.

40. Jiang, H.; Li, C.; Sun, T.; Ma, J., A green and high energy density asymmetric supercapacitor based on ultrathin MnO2 nanostructures and functional mesoporous carbon nanotube electrodes. Nanoscale 2012, 4, 807-12.

41. Hu, L.; Ren, Y.; Yang, H.; Xu, Q., Fabrication of 3D hierarchical MoS(2)/polyaniline and

MoS(2)/C architectures for lithium-ion battery applications. ACS applied materials & interfaces

2014, 6, 14644-52.

42. Singh, A.; Chandra, A., Significant Performance Enhancement in Asymmetric

Supercapacitors based on Metal Oxides, Carbon nanotubes and Neutral Aqueous Electrolyte. Sci.

Rep. 2015, 5, 1-5. 90

43. Zhou, Z.; Wu, X.-F.; Hou, H., Electrospun carbon nanofibers surface-grown with carbon nanotubes and polyaniline for use as high-performance electrode materials of supercapacitors.

RSC Advances 2014, 4, 23622.

44. Fic, K.; Lota, G.; Meller, M.; Frackowiak, E., Novel insight into neutral medium as electrolyte for high-voltage supercapacitors. Energy Environ. Sci. 2012, 5, 5842-5850.

45. Dai, Z.; Peng, C.; Chae, J. H.; Ng, K. C.; Chen, G. Z., Cell voltage versus electrode potential range in aqueous supercapacitors. Scientific reports 2015, 5, 9854.

46. Xu, D.; Xu, Q.; Wang, K.; Chen, J.; Chen, Z., Fabrication of free-standing hierarchical carbon nanofiber/graphene oxide/polyaniline films for supercapacitors. ACS applied materials & interfaces 2014, 6, 200-9.

47. Zhang, L. L.; Zhao, X.; Stoller, M. D.; Zhu, Y.; Ji, H.; Murali, S.; Wu, Y.; Perales, S.;

Clevenger, B.; Ruoff, R. S., Highly conductive and porous activated reduced graphene oxide films for high-power supercapacitors. Nano letters 2012, 12, 1806-12.

48. Van Aken, K. L.; Beidaghi, M.; Gogotsi, Y., Formulation of Ionic-Liquid Electrolyte To

Expand the Voltage Window of Supercapacitors. Angew. Chem. Int. Ed. 2015, 54, 4806-4809.

49. Galiński, M.; Lewandowski, A.; Stępniak, I., Ionic liquids as electrolytes. Electrochimica

Acta 2006, 51, 5567-5580.

50. <2008_Greaves_Protic Ionic Liquid.pdf>.

51. MacFarlane, D. R.; Tachikawa, N.; Forsyth, M.; Pringle, J. M.; Howlett, P. C.; Elliott, G.

D.; Davis, J. H.; Watanabe, M.; Simon, P.; Angell, C. A., Energy applications of ionic liquids.

Energy Environ. Sci. 2014, 7, 232-250. 91

52. Tran, C.; Lawrence, D.; Richey, F. W.; Dillard, C.; Elabd, Y. A.; Kalra, V., Binder-free three-dimensional high energy density electrodes for ionic-liquid supercapacitors. Chem.

Commun. 2015, 51, 13760-13763.

53. Lawrence, D. W.; Tran, C.; Mallajoysula, A. T.; Doorn, S. K.; Mohite, A.; Gupta, G.;

Kalra, V., High-energy density nanofiber-based solid-state supercapacitors. J. Mater. Chem. A

2016, 4, 160-166.

54. Du, C. Y.; Shi, P. F.; Cheng, X. Q.; Yin, G. P., Effective protonic and electronic conductivity of the catalyst layers in proton exchange membrane fuel cells. Electrochemistry

Communications 2004, 6, 435-440.

55. Zhao, Z.; Li, M.; Zhang, L.; Dai, L.; Xia, Z., Design Principles for Heteroatom-Doped

Carbon Nanomaterials as Highly Efficient Catalysts for Fuel Cells and Metal-Air Batteries. Adv

Mater 2015, 27, 6834-40.

56. Andersen, S. M.; Borghei, M.; Lund, P.; Elina, Y.-R.; Pasanen, A.; Kauppinen, E.; Ruiz,

V.; Kauranen, P.; Skou, E. M., Durability of carbon nanofiber (CNF) & carbon nanotube (CNT) as catalyst support for Proton Exchange Membrane Fuel Cells. Solid State Ionics 2013, 231, 94-

101.

57. Oezaslan, M.; Hasche, F.; Strasser, P., Oxygen Electroreduction on PtCo3, PtCo and

Pt3Co Alloy Nanoparticles for Alkaline and Acidic PEM Fuel Cells. Journal of The

Electrochemical Society 2012, 159, B394-B405.

58. Chen, H.; Snyder, J. D.; Elabd, Y. A., Electrospinning and Solution Properties of Nafion and Poly(acrylicacid). Macromolecules 2008, 41, 128-135. 92

59. Huang, Z.; Zhang, Y. Z.; Kotaki, M.; Ramakrishna, S., A review on polymer nanofibers by electrospinning and their applications in nanocomposites. Composites Science and

Technology 2003, 63, 2223-2253.

60. Shin, M.; Kim, Y.; Kim, S.; Kim, S.; Lee, H.; Spinks, G.; Kim, S., Enhanced conductivity of aligned PANi/PEO/MWNT nanofibers by electrospinning. Sensors and Actuators B:

Chemical 2008, 134, 122-126.

61. Tan, S.; Inai, R.; Kotaki, M.; Ramakrishn, S., Systematic parameter study for ultra-fine fiber fabrication via electrospinning process. Polymer. 2005, 46, 6128-6134.

62. MacDiarmid, A. G., Polyanilines: A Novel Class of Conducting Polymers. Farad Disc

Chem Soc 1989, 88, 317-332.

63. Chiang, J.; Macdiarmid, A. G., Polyaniline: Protonic acid doping of the emearaldine form to the metalic regime. Synthetic Metals 1986, 13, 193-205.

64. Xia, Y.; MacDiarmid, A. G., Camphorsulfonic Acid Fully Doped Polyaniline Emeraldine

Salt: In Situ Observation of Electronic and Conformational Changes Induced by Organic Vapors by an Ultraviolet/ Visible/Near-Infrared Spectroscopic Method. Macromolecules 1994, 27, 7212-

7214.

65. Long, Y.-Z.; Li, M.-M.; Gu, C.; Wan, M.; Duvail, J.-L.; Liu, Z.; Fan, Z., Recent advances in synthesis, physical properties and applications of conducting polymer nanotubes and nanofibers. Progress in Polymer Science 2011, 36, 1415-1442.

66. Tran, H. D.; D'Arcy, J. M.; Wang, Y.; Beltramo, P. J.; Strong, V. A.; Kaner, R. B., The oxidation of aniline to produce “polyaniline”: a process yielding many different nanoscale structures. J. Mater. Chem. 2011, 21, 3534-3550. 93

67. Innis, P. C.; Mazurkiewicz, J.; Nguyen, T.; Wallace, G. G.; MacFarlane, D., Enhanced electrochemical stability of polyaniline in ionic liquids. Current Applied Physics 2004, 4, 389-

393.

68. Jingjing Xu; Kai Wang; Sheng-Zhen Zu; Bao-Hang Han; Zhixiang Wei, Hierarchical

Nanocomposites of Polyaniline Nanowire Arrays on Graphene Oxide Sheets with Synergistic

Effect for Energy Storage. 2010, 4, 5019-5026.

69. Tran, C.; Singhal, R.; Lawrence, D.; Kalra, V., Polyaniline-coated freestanding porous carbon nanofibers as efficient hybrid electrodes for supercapacitors. J Power Sources 2015, 293,

373-379.

70. Li, L.; Raji, A. R.; Fei, H.; Yang, Y.; Samuel, E. L.; Tour, J. M., Nanocomposite of polyaniline nanorods grown on graphene nanoribbons for highly capacitive pseudocapacitors.

ACS applied materials & interfaces 2013, 5, 6622-7.

71. Hassan, M.; Reddy, K. R.; Haque, E.; Faisal, S. N.; Ghasemi, S.; I. Minett, A.; Gomes, V.

G., Hierarchical assembly of graphene/polyaniline nanostructures to synthesize free-standing supercapacitor electrode. Composites Sci Technol 2014, 98, 1-8.

72. Wei, J.; Zhang, J.; Liu, Y.; Xu, G.; Chen, Z.; Xu, Q., Controlled growth of whisker-like polyaniline on carbon nanofibers and their long cycle life for supercapacitors. RSC Advances

2013, 3, 3957.

73. Xie, K.; Li, J.; Lai, Y.; Zhang, Z.; Liu, Y.; Zhang, G.; Huang, H., Polyaniline nanowire array encapsulated in titania nanotubes as a superior electrode for supercapacitors. Nanoscale

2011, 3, 2202-7. 94

74. Chen, L.; Song, Z.; Liu, G.; Qiu, J.; Yu, C.; Qin, J.; Ma, L.; Tian, F.; Liu, W., Synthesis and electrochemical performance of polyaniline–MnO2 nanowire composites for supercapacitors. J Phys Chem Solids 2013, 74, 360-365.

75. Bai, M.-H.; Liu, T.-Y.; Luan, F.; Li, Y.; Liu, X.-X., Electrodeposition of vanadium oxide–polyaniline composite nanowire electrodes for high energy density supercapacitors. J

Mater Chem A 2014, 2, 10882.

76. Zhang, P.; Liu, Z.; Liu, Y.; Fan, H.; Jiao, Y.; Chen, B., Titanium Dioxide@Polyaniline

Core-Shell Nanowires as High-Performance and Stable Electrodes for Flexible Solid-State

Supercapacitors. Electrochimica Acta 2015, 184, 1-7.

77. Kumar, V.; Lee, P. S., Redox Active Polyaniline-h-MoO3Hollow Nanorods for Improved

Pseudocapacitive Performance. The J Physl Chem C 2015, 119, 9041-9049.

78. Xu, H.; Li, X.; Wang, G., Polyaniline nanofibers with a high specific surface area and an improved pore structure for supercapacitors. J Power Sources 2015, 294, 16-21.

79. Wang, X.; Deng, J.; Duan, X.; Liu, D.; Guo, J.; Liu, P., Crosslinked polyaniline nanorods with improved electrochemical performance as electrode material for supercapacitors. J Mater

Chem A 2014, 2, 12323.

80. Shi, Y.; Peng, L.; Ding, Y.; Zhao, Y.; Yu, G., Nanostructured conductive polymers for advanced energy storage. Chemical Society reviews 2015, 44, 6684-96.

81. Shi, Y.; Yu, G., Designing Hierarchically Nanostructured Conductive Polymer Gels for

Electrochemical Energy Storage and Conversion. Chemistry of Materials 2016, 28, 2466-2477.

82. Pan, L.; Yu, G.; Zhai, D.; Lee, H. R.; Zhao, W.; Liu, N.; Wang, H.; Tee, B. C. K.; Shi,

Y.; Cui, Y.; Bao, Z., Hierarchical nanostructured conducting polymer hydrogel with high electrochemical activity. PNAs 2012, 109, 9287-9292. 95

83. Wang, K.; Huang, J.; Wei, Z., Conducting Polyaniline Nanowire Arrays for High

Performance Supercapacitors. J. Phys. Chem. C. 2010, 114, 8062–8067.

84. Chen, W.; Rakhi, R. B.; Alshareef, H. N., Facile synthesis of polyaniline nanotubes using reactive oxide templates for high energy density pseudocapacitors. J. Mater. Chem. A. 2013, 1,

3315.

85. Chang, C.; Weng, C.; Chien, C.; Chuang, T.; TingYin Lee, J. Y.; Weib, Y.,

Polyaniline/carbon nanotube nanocomposite electrodes with biomimetic hierarchical structure for supercapacitors. J. Mater. Chem. A. 2013, 1, 14719-14728.

86. Ansari, S. A.; Parveen, N.; Han, T. H.; Ansari, M. O.; Cho, M. H., Fibrous polyaniline@manganese oxide nanocomposites as supercapacitor electrode materials and cathode catalysts for improved power production in microbial fuel cells. Physical chemistry chemical physics : PCCP 2016, 18, 9053-60.

87. Mu, J.; Ma, G.; Peng, H.; Li, J.; Sun, K.; Lei, Z., Facile fabrication of self-assembled polyaniline nanotubes doped with d-tartaric acid for high-performance supercapacitors. J Power

Sources 2013, 242, 797-802.

88. Chen, W.; Rakhi, R. B.; Alshareef, H. N., Morphology-Dependent Enhancement of the

Pseudocapacitance of Template-Guided Tunable Polyaniline Nanostructures. The J Phys Chem C

2013, 117, 15009-15019.

89. Wang, X.; Liu, D.; Deng, J.; Duan, X.; Guo, J.; Liu, P., Improving cyclic stability of polyaniline by thermal crosslinking as electrode material for supercapacitors. RSC Adv. 2015, 5,

78545-78552.

90. Vonlanthen, D.; Lazarev, P.; See, K. A.; Wudl, F.; Heeger, A. J., A stable polyaniline- benzoquinone-hydroquinone supercapacitor. Adv Mater 2014, 26, 5095-100. 96

91. Xie, H.; Zhu, Y.; Wu, Y.; Wu, Z.; Liu, E., The effect of hydroquinone as an electrolyte additive on electrochemical performance of the polyaniline supercapacitor. Mater Res Bulletin

2014, 50, 303-306.

92. Ran, F.; Tan, Y.-t.; Liu, J.; zhao, L.; Kong, L.-b.; Luo, Y.-c.; Kang, L., Preparation of hierarchical polyaniline nanotubes based on self-assembly and its electrochemical capacitance.

Polyme. Adv. Technologies. 2012, 23, 1297-1301.

93. Snook, G. A.; Kao, P.; Best, A. S., Conducting-polymer-based supercapacitor devices and electrodes. Journal of Power Sources 2011, 196, 1-12.

94. Cho, S.; Shin, K. H.; Jang, J., Enhanced electrochemical performance of highly porous supercapacitor electrodes based on solution processed polyaniline thin films. ACS applied materials & interfaces 2013, 5, 9186-93.

95. Kwon, S. R.; Jeon, J.-W.; Lutkenhaus, J. L., Sprayable, paintable layer-by-layer polyaniline nanofiber/graphene electrodes. RSC Adv. 2015, 5, 14994-15001.

96. Wang, L.; Lu, X.; Lei, S.; Song, Y., Graphene-based polyaniline nanocomposites: preparation, properties and applications. J Mater Chem A 2014, 2, 4491-4509.

97. Luo, Y.; Kong, D.; Jia, Y.; Luo, J.; Lu, Y.; Zhang, D.; Qiu, K.; Li, C. M.; Yu, T., Self- assembled graphene@PANI nanoworm composites with enhanced supercapacitor performance.

RSC Advances 2013, 3, 5851.

98. Miao, Y. E.; Fan, W.; Chen, D.; Liu, T., High-performance supercapacitors based on hollow polyaniline nanofibers by electrospinning. ACS. Appl. Mater. Interfaces. 2013, 5, 4423-8.

99. Sk, M. M.; Yue, C. Y., Synthesis of polyaniline nanotubes using the self-assembly behavior of vitamin C: a mechanistic study and application in electrochemical supercapacitors.

Journal of Materials Chemistry A 2014, 2, 2830. 97

100. Yang, W.; Gao, Z.; Song, N.; Zhang, Y.; Yang, Y.; Wang, J., Synthesis of hollow polyaniline nano-capsules and their supercapacitor application. Journal of Power Sources 2014,

272, 915-921.

101. Wu, W.; Pan, D.; Li, Y.; Zhao, G.; Jing, L.; Chen, S., Facile fabrication of polyaniline nanotubes using the self-assembly behavior based on the hydrogen bonding: a mechanistic study and application in high-performance electrochemical supercapacitor electrode. Electrochim Acta

2015, 152, 126-134.

102. Luzio, A.; Canesi, E. V.; Bertarelli, C.; Caironi, M., Electrospun Polym. Fibers for

Electronic Applications. Materials. 2014, 7, 906-947.

103. Zhang, Y.; Rutledge, G. C., Electrical Conductivity of Electrospun Polyaniline and

Polyaniline-Blend Fibers and Mats. Macromolecules. 2012, 45, 4238-4246.

104. Norris, I. D.; Shaker, M. M.; Ko, F. K.; MacDiarmid, A. G., Electrostatic fabrication of ultrafine conducting fibers: polyanilinerpolyethylene oxide blends. Synthetic Metals 2000, 114,

109-114.

105. Frontera, P.; Busacca, C.; Trocino, S.; Antonucci, P.; Faro, M. L.; Falletta, E.; Pina, C.

D.; Rossi, M., Electrospinning of Polyaniline: Effect of Different Raw Sources. J. Nanoscie. and

Nanotechnology 2013, 13, 4744-4751.

106. Cao, Y.; Qiu, h.; Smith, P., Effect of Solvents and Co-solvents on the Processibility of

Polyaniliie :I. Solubility and Conductivity studies. S. Metals. 1995, 69, 187-190.

107. Wu, T.-M.; Lin, Y.-W., Doped polyaniline/multi-walled carbon nanotube composites:

Preparation, characterization and properties. Polymer 2006, 47, 3576-3582. 98

108. Wang, Q.; Yao, Q.; Chang, J.; Chen, L., Enhanced thermoelectric properties of

CNT/PANI composite nanofibers by highly orienting the arrangement of polymer chains. J.

Mater. Chemistry A. 2012, 22, 17612.

109. Dong, B.; Gwee, L.; Salas-de la Cruz, D.; Winey, K. I.; Elabd, Y. A., Super proton conductive high-purity nafion nanofibers. Nano letters 2010, 10, 3785-90.

110. Cochet, M.; Maser, W. K.; Benito, A. M.; Callejas, M. A.; Martínez, M. T.; Benoit, J.-

M.; Schreiber, J.; Chauvet, O., Synthesis of a new polyaniline/nanotube composite: “in-situ” polymerisation and charge transfer through site-selective interaction. Chemical Communications

2001, 1450-1451.

111. Zhang, K.; Zhang, L. L.; Zhao, X. S.; Wu, J., Graphene/Polyaniline Nanofiber

Composites as Supercapacitor Electrodes. Chem. Materials. 2010, 22, 1392-1401.

112. Duic, L.; Mandic, Z.; Kovac, S., Polymer-Dimer distribution in the electrochemical synthesis of polyaniline. Electrochemical acta 1995, 40, 1681-1688.

113. Fan, W.; Zhang, C.; Tjiu, W. W.; Pramoda, K. P.; He, C.; Liu, T., Graphene-wrapped polyaniline hollow spheres as novel hybrid electrode materials for supercapacitor applications.

ACS applied materials & interfaces 2013, 5, 3382-91.

114. Miao, Y.-E.; Yan, J.; Huang, Y.; Fan, W.; Liu, T., Electrospun polymer nanofiber membrane electrodes and an electrolyte for highly flexible and foldable all-solid-state supercapacitors. RSC Adv. 2015, 5, 26189-26196.

115. Zhu, Z.-z.; Wang, G.-c.; Sun, M.-q.; Li, X.-w.; Li, C.-z., Fabrication and electrochemical characterization of polyaniline nanorods modified with sulfonated carbon nanotubes for supercapacitor applications. Electrochemical acta. 2011, 56, 1366-1372

116. Subramania, A.; Devi, S. L., Polyaniline nanofibers by surfactant‐assisted dilute polymerization for supercapacitor applications. Polym. Adv. Technologies. 2008, 19, 725-727.

117. Frackowiaka, E.; Khomenkob, V.; Jurewicza, K.; Lota, K.; Beguin, F., Supercapacitors based on conducting polymers/nanotubes composites. J. Power Sources. 2006, 153, 413-418.

118. Gupta, V.; Miura, N., Polyaniline/single-wall carbon nanotube (PANI/SWCNT) composites for high performance supercapacitors. Electrochimica Acta 2006, 52, 1721-1726.

119. Zhou, S.; Zhang, H.; Wang, X.; Li, J.; Wang, F., Sandwich nanocomposites of polyaniline embedded between graphene layers and multi-walled carbon nanotubes for cycle- stable electrode materials of organic supercapacitors. RSC Adv. 2013, 3, 1797-1807.

120. Li, J.; Xie, H.; Li, Y.; Liu, J.; Li, Z., Electrochemical properties of graphene nanosheets/polyaniline nanofibers composites as electrode for supercapacitors. Journal of Power

Sources 2011, 196, 10775-10781.

121. Simotwo, S. K.; DelRe, C.; Kalra, V., Supercapacitor Electrodes Based on High-Purity

Electrospun Polyaniline and Polyaniline-Carbon Nanotube Nanofibers. ACS applied materials & interfaces 2016.

122. Han, G.; Liu, Y.; Zhang, L.; Kan, E.; Zhang, S.; Tang, J.; Tang, W., MnO2 nanorods intercalating graphene oxide/polyaniline ternary composites for robust high-performance supercapacitors. Scientific reports 2014, 4, 4824.

123. Wang, G.; Zhang, L.; Zhang, J., A review of electrode materials for electrochemical supercapacitors. Chem. Soc. Rev. 2012, 41, 797-828.

124. Fan, Z.; Yan, J.; Wei, T.; Zhi, L.; Ning, G.; Li, T.; Wei, F., Asymmetric Supercapacitors

Based on Graphene/MnO2 and Activated Carbon Nanofiber Electrodes with High Power and

Energy Density. Adv. Funct. Mater. 2011, 21, 2366-2375. 100

125. Xu, Y.; Lin, Z.; Huang, X.; Yuan Liu, Y. H.; Duan, X., Flexible Solid-State

Supercapacitors Based on Three-Dimensional Graphene Hydrogel Films. ACS nano 2013, 7,

4042–4049.

126. Yu, P.; Zhang, Z.; Zheng, L.; Teng, F.; Hu, L.; Fang, X., A Novel Sustainable Flour

Derived Hierarchical Nitrogen-Doped Porous Carbon/Polyaniline Electrode for Advanced

Asymmetric Supercapacitors. Advanced Energy Materials 2016, 6, 1601111.

127. Cheng, Q.; Tang, J.; Shinya, N.; Qin, L.-C., Polyaniline modified graphene and carbon nanotube composite electrode for asymmetric supercapacitors of high energy density. Journal of

Power Sources 2013, 241, 423-428.

128. Mak, W. F.; Wee, G.; Aravindan, V.; Gupta, N.; Mhaisalkar, S. G.; Madhavi, S., High-

Energy Density Asymmetric Supercapacitor Based on Electrospun Vanadium Pentoxide and

Polyaniline Nanofibers in Aqueous Electrolyte. Journal of the Electrochemical Society 2012,

159, A1481-A1488.

129. Tran, C.; Kalra, V., Fabrication of porous carbon nanofibers with adjustable pore sizes as electrodes for supercapacitors. J. Power Sources 2013, 235, 289-296.

130. Li, X.; Shen, J.; Sun, W.; Hong, X.; Wang, R.; Zhao, X.; Yan, X., A super-high energy density asymmetric supercapacitor based on 3D core–shell structured NiCo-layered double hydroxide@carbon nanotube and activated polyaniline-derived carbon electrodes with commercial level mass loading. J. Mater. Chem. A 2015, 3, 13244-13253.

131. Gao, H.; Xiao, F.; Ching, C. B.; Duan, H., High-performance asymmetric supercapacitor based on graphene hydrogel and nanostructured MnO2. ACS applied materials & interfaces

2012, 4, 2801-10. 101

132. Zhang, J.; Jiang, J.; Li, H.; Zhao, X. S., A high-performance asymmetric supercapacitor fabricated with graphene-based electrodes. Energy & Environmental Science 2011, 4, 4009.

133. Zhao Ping, G. E. N., Helmut Neugebauer, Johannes Theiner and Adolf Neckel,

Redox of PANI_In situ IR.pdf>.

134. Salinas-Torres, D.; Sieben, J. M.; Lozano-Castelló, D.; Cazorla-Amorós, D.; Morallon,

E., Asymmetric hybrid capacitors based on activated carbon and activated carbon fibre–PANI electrodes. Electrochimica Acta 2013, 89, 326-333.

135. Zhao, X.; Chen, C.; Huang, Z.; Jin, L.; Zhang, J.; Li, Y.; Zhang, L.; Zhang, Q., Rational design of polyaniline/MnO 2/carbon cloth ternary hybrids as electrodes for supercapacitors. RSC

Advances 2015, 5, 66311-66317.

136. Armand, M.; Endres, F.; MacFarlane, D. R.; Ohno, H.; Scrosati, B., Ionic-liquid materials for the electrochemical challenges of the future. Nat. Mater. 2009, 8, 621-629.

137. Richey, F. W.; Tran, C.; Kalra, V.; Elabd, Y. A., Ionic Liquid Dynamics in Nanoporous

Carbon Nanofibers in Supercapacitors Measured within OperandoInfrared

Spectroelectrochemistry. J. Phys. Chem. C. 2014, 118, 21846-21855.

138. Ahn, Y.-k.; Kim, B.; Ko, J.; You, D.-J.; Yin, Z.; Kim, H.; Shin, D.; Cho, S.; Yoo, J.;

Kim, Y. S., All solid state flexible supercapacitors operating at 4 V with a cross-linked polymer– ionic liquid electrolyte. J. Mater. Chem. A 2016, 4, 4386-4391.

139. Yang, X.; Zhang, L.; Zhang, F.; Zhang, T.; Huang, Y.; Chen, Y., A high-performance all- solid-state supercapacitor with graphene-doped carbon material electrodes and a graphene oxide- doped ion gel electrolyte. Carbon 2014, 72, 381-386.

140. Ashok Kumar, N.; Baek, J. B., Electrochemical supercapacitors from conducting polyaniline-graphene platforms. Chem Commun (Camb) 2014, 50, 6298-308. 102

141. Anouti, M.; Timperman, L., A pyrrolidinium nitrate protic ionic liquid-based electrolyte for very low-temperature electrical double-layer capacitors. Physical chemistry chemical physics

: PCCP 2013, 15, 6539-48.

142. Timperman, L.; Skowron, P.; Boisset, A.; Galiano, H.; Lemordant, D.; Frackowiak, E.;

Beguin, F.; Anouti, M., Triethylammonium bis(tetrafluoromethylsulfonyl)amide protic ionic liquid as an electrolyte for electrical double-layer capacitors. Physical chemistry chemical physics : PCCP 2012, 14, 8199-207.

143. Rochefort, D.; Pont, A.-L., Pseudocapacitive behaviour of RuO2 in a proton exchange ionic liquid. Electrochemistry Communications 2006, 8, 1539-1543.

144. Fang, B.; Kim, M.; Kim, J. H.; Song, M. Y.; Wang, Y.; Wang, H.; Wilkinson, D. P.; Yu,

J.-S., High Pt loading on functionalized multiwall carbon nanotubes as a highly efficient cathode electrocatalyst for proton exchange membrane fuel cells. Journal of Materials Chemistry 2011,

21, 8066-8073.

145. Wu, G.; More, K. L.; Johnston, C. M.; Zelenay, P., High-Performance Electrocatalysts for Oxygen Reduction Derived from Polyaniline, Iron, and Cobalt. Science 2011, 332, 443-447.

146. Marcu, A.; Toth, G.; Srivastava, R.; Strasser, P., Preparation, characterization and degradation mechanisms of PtCu alloy nanoparticles for automotive fuel cells. Journal of Power

Sources 2012, 208, 288-295.

147. Othman, R.; Dicks, A. L.; Zhu, Z., Non precious metal catalysts for the PEM fuel cell cathode. International Journal of Hydrogen Energy 2012, 37, 357-372.

148. Hezarjaribi, M.; Jahanshahi, M.; Rahimpour, A.; Yaldagard, M., Gas diffusion electrode based on electrospun Pani/CNF nanofibers hybrid for proton exchange membrane fuel cells

(PEMFC) applications. Applied Surface Science 2014, 295, 144-149. 103

149. Litster, S.; McLean, G., PEM fuel cell electrodes. Journal of Power Sources 2004, 130,

61-76.

150. Tang, J. M.; Jensen, K.; Waje, M.; Li, W.; Larsen, P.; Pauley, K.; Chen, Z.; Ramesh, P.;

Itkis, M. E.; Yan, Y.; Haddon, R. C., High Performance Hydrogen Fuel Cells with Ultralow Pt

Loading Carbon Nanotube Thin Film Catalysts. J. Phys. Chem 2007, 111, 17901-17904.

151. Park, J.; Ju, Y.; Park, S.; Jung, H.; Yang, K.; Lee, W., Effects of electrospun polyacrylonitrile-based carbon nanofibers as catalyst support in PEMFC. Journal of Applied

Electrochemistry 2009, 39, 1229-1236.

152. Brodt, M.; Wycisk, R.; Pintauro, P. N., Nanofiber Electrodes with Low Platinum Loading for High Power Hydrogen/Air PEM Fuel Cells. Journal of the Electrochemical Society 2013,

160, F744-F749.

153. Debe, M. K.; Schmoeckel, A. K.; Vernstrom, G. D.; Atanasoski, R., High voltage stability of nanostructured thin film catalysts for PEM fuel cells. Journal of Power Sources 2006,

161, 1002-1011.

154. Alia, S. M.; Larsen, B. A.; Pylypenko, S.; Cullen, D. A.; Diercks, D. R.; Neyerlin, K. C.;

Kocha, S. S.; Pivovar, B. S., Platinum-Coated Nickel Nanowires as Oxygen-Reducing

Electrocatalysts. ACS Catalysis 2014, 4, 1114-1119.

155. Baji, A.; Maia, Y.; Wong, S.-C.; Abtahi, M.; Chen, P., Electrospinning of polymer nanofibers: Effects on oriented morphology, structures and tensile properties. Composites

Science and Technology 2010, 70, 703-710.

156. Hwang, K.; Kwon, B.; Byun, H., Preparation of PVdF nanofiber membranes by electrospinning and their use as secondary battery separators. Journal of Membrane Science

2011, 378, 111-116. 104

157. Laforgue, A.; Robitaille, L.; Mokrini, A.; Ajji, A., Fabrication and Characterization of

Ionic Conducting Nanofiber. 2007, 292, 1229–1236.

158. Ballengee, J. B.; Pintauro, P. N., Morphological Control of Electrospun Nafion Nanofiber

Mats. Journal of The Electrochemical Society 2011, 158, B568.

159. Mollá, S.; Compañ, V., Polyvinyl alcohol nanofiber reinforced Nafion membranes for fuel cell applications. Journal of Membrane Science 2011, 372, 191-200.

160. Choi, J.; Lee, K. M.; Wycisk, R.; Pintauro, P. N.; Mather, P. T., Nanofiber Network Ion-

Exchange Membranes. Macromolecules, 2008, 41, 4569-4572.

161. Choi, J.; Kim, Y.; Lee, J.; Cho, K.; Jung, H.; Park, J.; Park, I.; Sung, Y., A polyaniline supported PtRu nanocomposite anode and a Pd-impregnated nanocomposite Nafion membrane for DMFCs. Solid State Ionics 2005, 176, 3031-3034.

162. Chen, Z.; Xu, L.; Li, W.; Waje, M.; Yan, Y., Polyaniline nanofibre supported platinum nanoelectrocatalysts for direct methanol fuel cells. Nanotechnology 2006, 17, 5254-5259.

163. Rajesh, B.; Thampi, K. R.; Bonard, J. M.; Mathieu, H. J.; Xanthopoulos, N.;

Viswanathan, B., Nanostructured Conducting Polyaniline Tubules as Catalyst Support for Pt

Particles for Possible Fuel Cell Applications. Electrochemical and Solid-State Letters 2004, 7,

A404.

164. Chen, S.; Wei, Z.; Qi, X.; Dong, L.; Guo, Y. G.; Wan, L.; Shao, Z.; Li, L.,

Nanostructured polyaniline-decorated Pt/C@PANI core-shell catalyst with enhanced durability and activity. Journal of the American Chemical Society 2012, 134, 13252-5.

165. Pinto, N. J.; Ramos, I.; Rojas, R.; Wang, P.-C.; Johnson, A. T., Electric response of isolated electrospun polyaniline nanofibers to vapors of aliphatic alcohols. Sensors and Actuators

B: Chemical 2008, 129, 621-627. 105

166. Ballengee, J. B.; Pintauro, P. N., Preparation of nanofiber composite proton-exchange membranes from dual fiber electrospun mats. Journal of Membrane Science 2013, 442, 187-195.

167. Sundaray, B.; Choi, A.; Park, Y. W., Highly conducting electrospun polyaniline- polyethylene oxide nanofibrous membranes filled with single-walled carbon nanotubes. Synthetic

Metals 2010, 160, 984-988.

168. MacDiarmid, A., Synthetic Metals: A Novel Role for Organic Polymers Angewandte

Chemie 2001, 40, 2581-2590.

169. Ballengee, J. B.; Pintauro, P. N., Composite Fuel Cell Membranes from Dual-Nanofiber

Electrospun Mats. Macromolecules 2011, 44, 7307-7314.

170. Li, G.; Pickup, P. G., Ionic Conductivity of PEMFC Electrodes. Journal of The

Electrochemical Society 2003, 150, C745-C752.

171. Zawodzinski, T. A.; Derouin, C.; Radzinski, S.; Ruth J. Sherman; Smith, V. T.; Springer,

T. E.; Gottesfeld, S., Water Uptake by and Transport Through Nafion-117 Membranes. J.

Electrochemical Soc 1993, 140.

172. Choi, J.; Lee, K. M.; Wycisk, R.; Pintauro, P. N.; Mather, P. T., Nanofiber composite membranes with low equivalent weight perfluorosulfonic acid polymers. Journal of Materials

Chemistry 2010, 20, 6282.

173. Scigala, R.; Wlochowicz, A., Crystallinity of poly(tetrafluoroethylene). Acta polymerica

1989, 40, 15-18.

174. Grosmaire, L.; Castagnoni, S.; Huguet, P.; Sistat, P.; Boucher, M.; Bouchard, P.; Bebin,

P.; Deabate, S., Probing proton dissociation in ionic polymers by means of in situ ATR-FTIR spectroscopy. Physical chemistry chemical physics : PCCP 2008, 10, 1577-83. 106

175. Gruger, A.; Regis, A.; Schmatko, T.; Colomban, P., Nanostructure of Nafion membranes at different states of hydration an IR and Raman. Vibrational Spectroscopy 26, 215-225.

176. Laporta, M.; Pegoraro, M.; Zanderighi, L., PerÑuorosulfonated membrane (Nafion): FT-

IR study of the state of water with increasing humidity. Phys. Chem. Chem. Phys. 1999, 1 4619-

4628.

177. Zhang, F.; Zhang, Z.; Liu, Y.; Leng, J., Shape memory properties of electrospun nafion nanofibers. Fibers and Polymers 2014, 15, 534-539.

178. Choi, B. G.; Park, H.; Im, H. S.; Kim, Y. J.; Hong, W. H., Influence of oxidation state of polyaniline on physicochemical and transport properties of Nafion/polyaniline composite membrane for DMFC. Journal of Membrane Science 2008, 324, 102-110.

179. Huggins, R. A., Simple Method to Determine Electronic and Ionic components of the conductivity in mixed conductors. Ionics 2002, 8, 301-313.

180. Soboleva, T.; Xie, Z.; Shi, Z.; Tsang, E.; Navessin, T.; Holdcroft, S., Investigation of the through-plane impedance technique for evaluation of anisotropy of proton conducting polymer membranes. Journal of Electroanalytical Chemistry 2008, 622, 145-152.

181. Javier, A. E.; Patel, S. N.; Hallinan, D. T., Jr.; Srinivasan, V.; Balsara, N. P.,

Simultaneous electronic and ionic conduction in a block copolymer: application in lithium battery electrodes. Angewandte Chemie 2011, 50, 9848-51.

182. Xie, Z.; Song, C.; Andreaus, B.; Titichai Navessin; Shi, Z.; Zhang, J.; Holdcrof, S.,

Discrepancies in the Measurement of Ionic Conductivity of PEMs Using Two- and Four-Probe

AC Impedance Spectroscopy. Journal of The Electrochemical Society, 2006, 153 (10), E173-

E178.

107

Appendix

Figure S1: Three electrode set up used for electrochemical testing. Graphite rod, Pt Mesh and Ag/AgCl were used as working, counter and reference electrodes, respectively.

Figure S2: Digital photos of freestanding nanofiber electrodes. The electrodes are used in the working electrode without any binder.

108

Figure S3: Autostepwise isothermal thermogravimetric analysis obtained at 10°C /min for pure PEO, CNTS, PANI93 and PANI-CNT in nitrogen atmosphere. PEO composition in both PANI93 and PANI- CNT was ~7 wt% (PEO decomposes at ~200°C. Presence of CNTs seems to enhance the thermal stability of PANI-CNT composite.

Figure S4: TEM image showing distribution of CNTs within PANI-CNTs nanofibers. The fibers have diameters of 598 and 394 nm.

109

Figure S5: A linear relationship for the wave current as a function of the square root of scan rate for PANI93 (a) and PANI-CNT samples (b). 110

Figure S6: Differential scanning calorimetry (DSC) of PANI93 nanofibers done in argon atmosphere at 10°C/min.

Figure S7: CV plots showing life performance of PANI-CNT electrode over 1000 continuous charge- discharge cycles.

111

Figure S8: Ragone plot showing the energy densities of the symmetric PANI-CNT capacitor at different power densities.

112

VITA

SILAS K. SIMOTWO

E-mail: [email protected]

EDUCATION

Doctor of Philosophy, Chemical & Biological Engineering Sept. 2012-Sept. 2017

Drexel University, Philadelphia, PA

Bachelor of Science, Chemical Engineering Aug. 2008- May. 2012

Lehigh University, Bethlehem, PA

PUBLICATIONS

1. Simotwo, S.; Delre, C.; Kalra, V; Supercapacitor Electrodes Based on High-Purity Electrospun

Polyaniline and Polyaniline-Carbon Nanotube Nanofibers, ACS Appl. Mater. Interfaces 8 (2016).

2. Simotwo, S.; Kalra, V; Study of Co-Electrospun Nafion and Polyaniline Nanofibers as Potential

Catalyst Support for Fuel Cell Electrodes, Electrochemical Acta, 198 (2016).

3. Simotwo, S.; Kalra, V; Polyaniline-based Electrodes: Recent Application in Supercapacitors and

Next Generation Rechargeable Batteries, Current Opinion in Chemical Engineering 13 (2016)

4. Simotwo, S , Chinnam, P.R, Wunder, S.L and Kalra, V. Highly Durable, Self-Standing Solid-

State Supercapacitor Based on Ionic Liquid-Rich Ionogel and Porous Carbon Nanofibers

Electrodes, in press, ACS Appl. Mater. Interfaces (2017).

5. Li, D.; Simotwo, S.; Nyman, M. and Liu, T. Evolution of Actinyl Peroxide Clusters U28 in

Dilute Electrolyte Solution: Exploring the Transition from Simple Ions to Macroionic

Assemblies, Chemistry-A European Journal 20 (2014)

6. Li, D.; Song, J.; Yin, P.; Simotwo, S.; Bassler, A.; Aung, Y.; Robersts J.E.; Hill, L. and Liu, T.

Inorganic-Organic Vesicle Hybrid Vesicles with Counterion and pH-Controlled Fluorescent

Properties, J. Am. Chem. Society (2011). 113

7. Simotwo, S.; and Kalra, V. Fabrication and In-situ Spectroelectrochemical Study of Polyaniline-

Carbon based Binder-Free Asymmetric Supercapacitor in Neutral Aqueous Electrolyte, to be

submitted (2017)

8. Pai, R.; Singh, A.; Simotwo, S.; and Kalra, V. Iron Carbide/Carbon Nano fibers based Free-

standing Electrodes as Anode in Aqueous Supercapacitors submitted (2017).

9. Simotwo, S.; and Kalra, V. Redox Mechanisms of Polyaniline in Protic Ionic Liquid, to be submitted

(2017).

10. Simotwo, S.; Martinez, U.; Dumont, JH.; Mohite, A.; Kalra, V; Gupta, G.; Metal-free Tunable

Nanostructured Carbon Nanofibers for Selective Oxygen Reduction Reaction, submitted (2017).

AWARDS/HONORS o William-Casey Fellowship, Chemical and Biological Engineering, Drexel University, for

excellence in research work—2017. o Outstanding Ph.D. Student, Chemical and Biological Engineering, Drexel University for

excellence in academic work—2016. o All Patriots League academic honor roll for excellence in academic and sports activities—2010 o Dean’s List—2009 o KenSAP scholarship for excellence in academic work—2009.