UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

High-yield production of graphene sheets by graphite electro-exfoliation for application in electrochemical… Taheri Najafabadi, Amin 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_november_taherinajafabadi_amin.pdf [ 6.75MB ]
Metadata
JSON: 24-1.0315348.json
JSON-LD: 24-1.0315348-ld.json
RDF/XML (Pretty): 24-1.0315348-rdf.xml
RDF/JSON: 24-1.0315348-rdf.json
Turtle: 24-1.0315348-turtle.txt
N-Triples: 24-1.0315348-rdf-ntriples.txt
Original Record: 24-1.0315348-source.json
Full Text
24-1.0315348-fulltext.txt
Citation
24-1.0315348.ris

Full Text

HIGH-YIELD PRODUCTION OF GRAPHENE SHEETS BY GRAPHITE ELECTRO-EXFOLIATION FOR APPLICATION IN ELECTROCHEMICAL POWER SOURCES by AMIN TAHERI NAJAFABADI   M.A.SC., THE UNIVERSITY OF BRITISH COLUMBIA, CANADA, 2012 B.SC., SHARIF UNIVERSITY OF TECHNOLOGY, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (CHEMICAL AND BIOLOGICAL ENGINEERING)  THE UNIVERSITY OF BRITISH COLUMBIA (VANCOUVER)  SEPTEMBER 2016  © AMIN TAHERI NAJAFABADI, 2016    ii Abstract This thesis first aims at developing an electrochemical approach for low temperature, simple, and cost-effective synthesis of graphene microsheets (GNs) using graphitic electrodes in ionic liquid (IL) medium. The second major focus involves the products application as cathode-modifying microporous layers (MPLs) in proton exchange membrane fuel cells (PEMFCs) as well as anode-modifying materials in microbial fuel cells (MFCs). For the electrochemical exfoliation, a novel IL/acetonitrile electrolyte is introduced, and investigated with low concentration of ionic liquids. Using iso-molded graphite rod as the anode, up to 86% of exfoliation was achieved with the majority of the products as graphene flakes in addition to smaller quantities of carbonaceous particles and rolled sheets. Moreover, the simultaneous anodic and cathodic GN production was developed here with a synergistic exfoliation effect. When graphitic anode and cathode were subjected to a constant cell potential, up to 3 times higher exfoliation yields were generated compared to single-electrode studies on each side (~6-fold improvement in total). Thorough materials characterization confirmed the production of ultrathin GNs (< 5 layers) on both electrodes, with cathodic sheets being relatively larger and less functionalized. On the application side, the successful integration of GNs in MPLs resulted in enhanced PEMFC performance over a wide range of operating conditions. GN-based MPLs improved performance in the kinetic and ohmic regions of the polarization curve, while the addition of carbon black (CB), particularly Vulcan XC72, to form a composite GN+CB MPL, further extended the improvement to the mass transport limiting region. This was reflected by an approximate 30% and 70% increase in peak power densities compared to CB and GN MPLs, respectively, at the relative humidity (RH) of 100%. Despite the presence of CB, GN+CB MPLs also retained their superior performance at a much lower RH of 20%, thereby widening the peak    iii power gap with CB MPLs to 80%. On the other side, the functionalized GN-modified carbon cloth anodes integrated within single-chamber MFCs generated an over four-fold improvement in peak power density compared to the plain carbon cloth (2.85 W m-2 vs 0.66 W m-2, respectively), exceeding the previously reported values with graphene anodes.    iv Preface The research undertaken in this dissertation including the identification of the research questions, design, and performing of the experiments, the analysis of the research data as well as the writing of the first draft of the dissertation and the four manuscripts arising from the work presented here are mainly done by the candidate. My research supervisor, Dr. Előd Gyenge, contributed by providing comments on the first draft, and thoroughly examining the content. Norvin Ng and Magrieta J Leeuwner were involved in some of the experiments (fuel cell performance benchmarking), thus placed as the second author each once in my last two publications. A list of published manuscripts for the work presented in this dissertation is given below:  A version of Chapter 3 has been published. Amin Taheri Najafabadi, Előd Gyenge, High-yield graphene production by electrochemical exfoliation of graphite: Novel ionic liquid (IL) - acetonitrile electrolyte with low IL content, Carbon 71 (2014), 58-69.  A version of Chapter 4 has been published. Amin Taheri Najafabadi, Előd Gyenge, Synergistic production of graphene microsheets by simultaneous anodic and cathodic electro-exfoliation of graphitic electrodes in aprotic ionic liquids, Carbon 84 (2015), 449-459.  A version of Chapter 5 has been published. Amin Taheri Najafabadi, Magrieta J Leeuwner, David P. Wilkinson, Előd Gyenge, Electrochemically produced graphene for microporous layers in fuel cells, ChemSusChem 9 (2016), 1689-1697.  A version of Chapter 6 has been published. Amin Taheri Najafabadi, Norvin Ng, Előd Gyenge, Electrochemically exfoliated graphene anodes with enhanced biocurrent production in single-chamber air-breathing microbial fuel cells, Biosensors and Bioelectronics 81 (2016), 103-110.  Chapters 1 and 2 also include some of the introductory materials along with the methods discussed in the aforementioned manuscripts.    v The following list summarizes the contributions arising from this work in various conferences:  Amin Taheri Najafabadi, Előd Gyenge, Paired electrosynthesis of graphene sheets in aprotic ionic liquids, Graphene Week, Manchester, United Kingdom (June 2015).  Miguel Garcia-Contreras, Pooya Hosseini Benhangi, Amin Taheri Najafabadi, Előd Gyenge, The Effect of Graphene As a Support for Non-PGM Bifunctional Oxygen Catalyst in Rechargeable Metal/Air Batteries, American Institute of Chemical Engineers (AIChE) Annual Meeting, Chicago, USA (November 2014).  Amin Taheri Najafabadi, Előd Gyenge, High throughput ionic-liquid-assisted electrosynthesis of graphene microsheets in aprotic media, Graphene 2014, Toulouse, France (May 2014).      vi Table of Contents Abstract ............................................................................................................................... ii Preface ................................................................................................................................. iv Table of Contents................................................................................................................ vi List of Tables ..................................................................................................................... ix List of Figures ......................................................................................................................x List of Symbols..................................................................................................................xix List of Abbreviations .........................................................................................................xxi Acknowledgements ......................................................................................................... xxiv Dedication ........................................................................................................................ xxv Chapter  1: Introduction ....................................................................................................... 1 1.1 Thesis motivation and novelty ................................................................................................ 4 1.2 Thesis objectives ....................................................................................................................... 6 1.3 Thesis approach ........................................................................................................................ 8 Chapter  2: Literature review ............................................................................................. 10 2.1 Theory and terminology ......................................................................................................... 10 2.2 Top-down synthesis techniques ............................................................................................ 11 2.3 Electrosynthesis approach ..................................................................................................... 14 2.4 Chemical functionalization .................................................................................................... 19 2.5 Characterization techniques .................................................................................................. 25 Chapter  3: Single-electrode exfoliation experiments ........................................................ 30 3.1 Introduction ............................................................................................................................. 30 3.2 Materials and methods ........................................................................................................... 31 3.3 Results and discussion ............................................................................................................ 35    vii 3.3.1 Electrosynthesis experiments ........................................................................................... 35 3.3.2 Product characterization .................................................................................................... 44 3.4 Summary and conclusions ..................................................................................................... 50 Chapter  4: Simultaneous exfoliation studies .................................................................... 52 4.1 Introduction ............................................................................................................................. 52 4.2 Materials and methods ........................................................................................................... 54 4.3 Results and discussion ............................................................................................................ 57 4.3.1 Electrosynthesis experiments ........................................................................................... 57 4.3.2 Product characterization .................................................................................................... 62 4.4 Summary and conclusions ..................................................................................................... 70 Chapter  5: Application I – Hydrogen fuel cells ................................................................ 74 5.1 Introduction ............................................................................................................................. 74 5.2 Theory and terminology ......................................................................................................... 77 5.2.1 Thermodynamics ................................................................................................................ 77 5.2.2 Energy losses ....................................................................................................................... 78 5.3 GDL and water management ................................................................................................ 81 5.4 Materials and methods ........................................................................................................... 84 5.4.1 Graphene synthesis and characterization ........................................................................ 84 5.4.2 PEMFC construction and characterization .................................................................... 84 5.5 Results and discussion ............................................................................................................ 88 5.6 Summary and conclusions .................................................................................................. 103 Chapter  6: Application II – Microbial fuel cells ............................................................. 105 6.1 Introduction .......................................................................................................................... 105 6.2 Anode studies ....................................................................................................................... 107    viii 6.3 Materials and methods ........................................................................................................ 110 6.4 Results and discussion ......................................................................................................... 113 6.5 Summary and conclusions .................................................................................................. 121 Chapter  7: Highlights, conclusions, and recommendations for future work ................ 122 7.1 Highlights and conclusions ................................................................................................. 122 7.2 Recommendations for future work ................................................................................... 124 References ........................................................................................................................ 127 Appendices ....................................................................................................................... 137 Appendix A Electrosynthesis protocols ......................................................................................... 137 A.1 Undivided exfoliation experiments ............................................................................... 137 A.2 Divided exfoliation experiments ................................................................................... 138 Appendix B ORR experiments ........................................................................................................ 139 B.1 Rotating disk electrode measurements ......................................................................... 139 B.2 ORR results ...................................................................................................................... 140 Appendix C MFC design & protocols ........................................................................................... 143     ix List of Tables Table 5-1: Mass transport characterization data of different MPLs. ................................................. 95 Table 6-1: The data comparison between various reports on the use of graphene as anode material for microbial fuel cells. The majority of the works have used reduced graphene oxide as the primary component. .............................................................................................................................. 114     x List of Figures Figure 1-1: Graphene as a building material for all other carbon dimensionalities with myriad of applications. It can be wrapped up into 0D buckyballs, rolled into 1D nanotubes or stacked into 3D graphite (right, Reprinted with permission from Ref.19, Copyright 2007, Nature Publishing Group). ......................................................................................................................................................... 2 Figure 1-2: Graphical illustration of two different approaches for synthesizing 2D graphene structures: bottom up (left – credits to Minot group) vs top down (right – credits to GLAB). ..... 3 Figure 1-3: Cost and quality comparison between various graphene synthesis techniques. Factors such as intrinsic conductivity, defect density, and structural disorder have been used as the general measures for the graphene quality. The illustration does not represent any quantitative scale (similar analogies have been made by Novoselov et al.46 and others) ................................................. 5 Figure 2-1: (A) Left: the electronic structure of the sp2 bonded graphene. Right: zoom-in of the Dirac point where valance and conduction bands intersect. Reprinted with permission from ref. 53, Copyright 2009, American Physical Society. (B) Ambipolar electric field effect in single-layer graphene (the position of the Dirac point and the Fermi energy (EF) are shown in the insets as a function of gate voltage). Reprinted with permission from ref. 19, Copyright 2007, Nature Publishing Group. (C) A schematic diagram for the Dirac point and the Fermi level positions as a function of doping; where the upper panel is n-type doped, pristine, and p-type doped free-standing graphene (a to c), and the lower section illustrates n-type doped, pristine and p-type doped epitaxial graphene grown on silicon carbide (SiC) (d to f). Reprinted with permission from ref. 56, Copyright 2008, American Chemical Society. ........................................................................... 11 Figure 2-2: Graphene production pathway via effective reduction of the mono-layered graphene oxide sheets. Reprinted with permission from ref. 44, Copyright 2011, Wiley-VCH. ..................... 13    xi Figure 2-3: Schematic of a sample electro-exfoliation process with graphite rod used as the anode WE, and platinum wire as the cathode CE to produce GN flakes. .................................................. 16 Figure 2-4: Graphical comparison between ionic solids with symmetric ions and ionic liquids with at least one asymmetric component resulting in dramatic drop in the melting point. .................... 18 Figure 2-5: As-prepared graphite oxide dispersions in water and 13 organic solvents by bath ultrasonication for 1 h. Top: Dispersions immediately after sonication. Bottom: Dispersions three weeks after sonication. The yellow color of the o-xylene sample is due to the solvent itself. Reprinted with permission from ref. 110, Copyright 2008, American Chemical Society. ................ 20 Figure 2-6: Variations in the proposed model structures for the GO sheets regarding the presence (left, Reprinted with permission from ref. 111, Copyright 1998, American Chemical Society) or absence (right, Reprinted with permission from ref. 112, Copyright 2009, Nature Publishing Group) of carboxylic acids on basal plane edges. ............................................................................................... 20 Figure 2-7: (a) Linear dependency of chemically induced charge transport, △n, to various concentration, C, of NO2. Lower inset: graphene characterization by the electric field effect. (b) Changes in resistivity, ρ, at zero-band-gap point of graphene induced by exposure to various diluted gases (1 ppm). The positive (negative) sign of changes indicate electron (hole) doping. (c) Constant mobility of charge carriers in graphene with increasing chemical doping. The parallel shift is caused by a negligible scattering effect of the charged impurities induced by chemical doping. Reprinted with permission from ref. 13, Copyright 2007, Nature Publishing Group. (d) Schematic of charge transfer at the F4-TCNQ/graphene interface. Reprinted with permission from ref. 165, Copyright 2007, American Chemical Society. ................................................................ 23 Figure 2-8: (a) Metal-free nitrogen-doped graphene as an efficient electrocatalyst for oxygen reduction reaction. Reprinted with permission from ref. 172, Copyright 2013, Royal Society of Chemistry. (b) Schematic illustration of various types of nitrogen-doped graphene (gray for the    xii carbon, blue for the nitrogen, and white for the hydrogen atom). A possible defect structure is shown in the middle of the ball-stick model. Reprinted with permission from ref. 154, Copyright 2010, American Chemical Society. ......................................................................................................... 25 Figure 2-9: (a) Raman spectra comparison between graphene and graphite measured at excitation wavelength of 514.5 nm. (b) 2D peaks comparison between graphene and graphite. (c) 2D peak evolution as a function of number of layers. Reprinted with permission from ref. 182, Copyright 2007, Elsevier. ............................................................................................................................................ 27 Figure 2-10: Raman shift trends for the graphene`s G band upon interaction with (a) 1 M solutions of monosubstituted benzenes, and (b) with various concentrations of aniline and nitrobenzene. Reprinted with permission from ref. 184, Copyright 2008, Royal Society of Chemistry. ................. 28 Figure 3-1: Two graphitic precursors chosen for the electro-exfoliation studies (Graphite Store) namely iso-molded graphite rod (left), and highly ordered pyrolytic graphite plate (right). .......... 32 Figure 3-2: Investigated ionic liquids and their respective electrochemical stability windows on platinum at 293 K versus normal hydrogen electrode (NHE). .......................................................... 33 Figure 3-3: Experimental setup used for graphite anodic exfoliation in the presence of ionic liquids. ..................................................................................................................................................................... 34 Figure 3-4: Radically different response to the intercalation of highly charged anions by two types of selected graphitic working electrodes with 4 cm of their length exposed to the electrolyte (0.1 M EMIM BF4 solution in acetonitrile at 7 V of applied potential). .................................................. 36 Figure 3-5: Electrolyte color changes during (a) electrochemical graphite exfoliation, and (b) control electrolysis experiment without graphite electrode in 0.1 M EMIM BF4/acetonitrile solution at 7 V and 293 K in the course of 4 hours (anode is on the left side in all images). ....... 38    xiii Figure 3-6: Fluorescence spectra comparison of the resulting electrolyte from (a) graphite electro-exfoliation experiment, and (b) control experiment both in 0.1 M EMIM BF4 /acetonitrile solution at 7 V and 293 K after 4 hours................................................................................................................ 39 Figure 3-7: Color changes and the exfoliation yield after 4 hours of ionic-liquid-assisted electrochemical graphite exfoliation at 7 V using 0.1 M IL/acetonitrile (~1:50 IL/solvent vol. ratio) at 293 K (the ± shows 95% confidence interval from 3 experiments). ................................. 41 Figure 3-8: (a) Progressive exfoliation of the graphite anode in the BMPyrr BTA during 4 hours with 0.1 M IL/acetonitrile at 7 V and 293 K. (b) Passing current profile versus time throughout the experiment. .......................................................................................................................................... 43 Figure 3-9: TEM and SEM images showing the variety of products generated by electrochemical exfoliation of iso-molded graphite in EMIM BF4/acetonitrile electrolyte at 7 V and 293 K: (a-d) crumpled/folded sheets, (e) semi-transparent carbonaceous particles, and (f) rolled sheets. ....... 46 Figure 3-10: AFM thickness profile analysis of products generated by electrochemical exfoliation of iso-molded graphite in EMIM BF4/acetonitrile electrolyte at 7 V and 293 K. .......................... 47 Figure 3-11: Changes in the Raman spectra from graphite rod to the final sonicated product obtained by electro-exfoliation in EMIM BF4/acetonitrile at 7 V and 293 K after 4 hours. ....... 48 Figure 3-12: XPS data for the exfoliation products obtained from EMIM BF4/acetonitrile electrolyte at 7 V and 293 K after 4 hours: survey scan, and N 1s narrow scan. ............................ 49 Figure 4-1: Experimental setup used for simultaneous electro-exfoliation of graphitic anodes and cathodes in acetonitrile containing ionic liquids (0.1 M). The highest standard electrochemical stability potential for each ion on platinum electrode is indicated at 293 K versus normal hydrogen electrode (NHE). ....................................................................................................................................... 56 Figure 4-2: (a) Simultaneous electrochemical exfoliation of iso-molded graphite rod in 0.1 M N1114 BF4 /ACN solution using the divided H-cell configuration at 15 V and 293 K in the course    xiv of 1 hour. The exfoliation patterns of the anode and cathode are also depicted on this right side. (b) Polarization curves in the undivided cell for three types of exfoliation experiments: simultaneous, anodic and cathodic. In the case of single-electrode exfoliation (anodic or cathodic), the counter-electrode was Pt. The numbers inside the box indicate the exfoliation percentages at 7 V after 2 hours for the three types of experiments........................................................................... 59 Figure 4-3: (a) Cyclic voltammetry of the ionic liquid solution (0.1 M N1114 BTA in ACN) on iso-molded graphitic electrodes compared to those on Pt electrodes. (b) Anionic deintercalation of BTA and BF4, compared to (c) cationic deintercalation of N1114 and BMPyrr. Experiments were conducted at the potential sweep rate of 100 mV s-1. .......................................................................... 61 Figure 4-4: Low and high resolution TEM images of the GNs generated by simultaneous anodic and cathodic electrochemical exfoliation of iso-molded graphite electrodes in N1114 BF4/ACN electrolyte at 15 V and 293 K using H-cell configuration: (a, b) anodic exfoliates resembling the folding fashion of ultrathin graphene layers, and (c, d) cathodic exfoliates exhibiting Moire patterns created via stacking ultrathin and almost intact honeycomb GNs with different rotation angles. ..................................................................................................................................................................... 63 Figure 4-5: Changes in the Raman spectra from graphite to the final products obtained by simultaneous electro-exfoliation of iso-molded graphite electrodes in N1114 BF4/ACN at 15 V and 293 K using the divided H-cell configuration. .............................................................................. 65 Figure 4-6: By-products detected amongst the anodic exfoliates generated by simultaneous anodic and cathodic exfoliation of iso-molded graphite with 0.1 M N1114 BF4 in ACN as electrolyte at 15 V and 293 K using the divided H-cell configuration: (a) nanoparticle clusters in TEM, (b) rolled sheet clusters in TEM, (c) larger tube agglomerates in FESEM, and (d) graphene flake bended half-way through along with a nanotube sharing similar characteristics observed in the same TEM grid. .............................................................................................................................................................. 66    xv Figure 4-7: XPS data for the anodic and cathodic products obtained from electro-exfoliation experiments in N1114 BF4/ACN electrolyte at 15 V and 293 K using H-cell configuration: (a) survey scan, and C 1s narrow scan of the (b) anodic, and (c) cathodic exfoliates. .......................... 68 Figure 4-8: Sample (a) XPS, and (b) TEM analysis of the anodic products obtained from electro-exfoliation experiments in N1114 B TA/ACN electrolyte at 15 V and 293 K using H-cell configuration. ............................................................................................................................................. 70 Figure 5-1: Schematic of the typical PEMFC components including flow field plates, gas diffusion layers, catalyst layers, and membrane. .................................................................................................... 75 Figure 5-2: A sample polarization curve of PEMFC highlighting three overpotential regions of kinetic, ohmic, and mass transport. ........................................................................................................ 79 Figure 5-3: Schematic representation of a PEMFC assembly and mass-transport processes through a cathode MPL. ......................................................................................................................................... 85 Figure 5-4: Process flow diagram of the PEMFC monitoring system used for the experimental section showing the jacket cooling/heating elements along with the humidifiers on the gas inlets to further hydrate the CCM. .................................................................................................................... 87 Figure 5-5: (a, and b) Polarization and power density curves for different MPLs at 100% RH. (c) Ohmic drop for different MPLs at 100% RH. (d) Cathode pressure drop as a function of air stoichiometry at a current density of 1000 mA cm-2 and 100% RH. ................................................. 90 Figure 5-6: (a) Cathodic sweep of the rotating disk electrode studies of the oxygen reduction reaction on the graphene compared to the platinum catalyst at 1600 rpm in 0.1 M H2SO4 solutions. (b) Electrochemical stability tests using cyclic voltammetry in nitrogen saturated solutions at 5 mV s-1 rate. ......................................................................................................................................................... 91 Figure 5-7: Characterization data for EGN obtained by cathodic electro-exfoliation of iso-molded graphite electrodes in N1114 BTA/ACN electrolyte at 7 V and 293 K. (a) XPS survey scan, and    xvi C 1s narrow scan data. (b) Raman spectra of the EGN compared with graphite, particularly for G/D peak ratios. (c) Raman spectra of the EGN compared with graphite, particularly for 2D peak transformation. .......................................................................................................................................... 93 Figure 5-8: FESEM images of the MPLs made from CB (left), EGN (middle), EGN+CB (right): (a) top view, and (b) cross sectional view. The inset in the bottom left is false colored to further highlight the distribution of CB through the EGN layers. ................................................................. 95 Figure 5-9: Image processing and schematic representation of structure for (a) EGN MPL, and (b) EGN+CB MPL. The darker blue color in the middle show the void areas in between the EGN sheets. .......................................................................................................................................................... 97 Figure 5-10: (a) Change in performance from 100 to 20% cathode RH (expressed relative to performance at 100% RH). (b) Power density curves at 20% RH. (c) Change in resistance/ohmic drop from 100 to 20% RH (expressed relative to resistance at 100% RH). (d) Longer term operation at constant current density of 1000 mA cm-2 and 20% cathode RH. ........................... 100 Figure 5-11: SEM images of (a) CB, (b) EGN and (c) EGN+CB MPL surfaces with total material load of 0.5 mg cm-2. ............................................................................................................................... 101 Figure 5-12: Performance results for different MPLs at a loading of 0.5 mg cm-2. (a, and b) Polarization and power density curves at 100% RH. (c) Change in performance from 100 to 20% RH (expressed relative to performance at 100% RH). (d) Longer term operation at constant voltage of 1000 mA cm-2 and 20% RH; no results are displayed for the CB MPL due to irreversible damage caused during polarization at 20% RH. ................................................................................ 102 Figure 6-1: A basic schematic for two-chamber microbial fuel cell with the bacterial anode operating in anaerobic mode, and oxygen reduction reaction occurring at the aerobic cathode. .................................................................................................................................................................. 106    xvii Figure 6-2: (a) Graphical representation of the single-chamber MFC showing biofilm growth on the graphene-based anode, and oxygen reduction on the air-cathode catalyzed by MnOx. The red arrows indicate the flow of electrons from the anode to the cathode through an external load. (b, and c) Images of the MFC units used in the present work. ............................................................. 109 Figure 6-3: Graphical representation of the graphene anode preparation and inoculation process. .................................................................................................................................................................. 112 Figure 6-4: (a) Polarization, and (b) power curves for anodic and cathodic GNs compared to carbon cloth normalized based on both electrode area and MFC’s empty bed volume with the step-sweep rate of 5 mV s-1 starting from the open circuit potential. ............................................ 114 Figure 6-5: Long-term performance data for MFCs with graphene-modifying anodes. The points are the average of 20-minute long tests with fresh nutrients at the peak power current during 3-day experiments. ..................................................................................................................................... 115 Figure 6-6: Characterization of anodic GNs compared with Vulcan XC-72. (a) Polarization curves normalized based on both electrode area and MFC’s empty bed volume obtained with a sweep rate of 5 mV s-1 starting from the open circuit potential. (b) EIS measurements at the open circuit potential in the frequency range of 104-10-3 Hz. (c) SEM images of the electrodes coated with graphene and carbon black compared with plain carbon cloth fibers. .......................................... 117 Figure 6-7: Electrochemical characterization of GN and CNT composites under the step-sweep rate of 5 mV s-1 starting from the open circuit potential. (a) Polarization, and (b) power curves for composites normalized based on both electrode area and MFC’s empty bed volume. (c) Polarization, and (d) power curves for CNT anodes in the first 24 hours based on both electrode area and MFC’s empty bed volume. (e) Suggested E.Coli growth mechanism on CNT-modified surfaces.301 ................................................................................................................................................ 120     xviii Figure A-1: Key dimensions and electrochemical conditions for the graphite exfoliation experiments in undivided cells (glassware specifications from BioLogic Science Instruments). 137 Figure A-2: Post-treatment protocols followed for further processing of the exfoliates obtained from the undivided setup. ..................................................................................................................... 137 Figure A-3: Key dimensions and electrochemical conditions for the graphite exfoliation experiments in divided H-cells (glassware specifications from Pine Research Instrumentation). .................................................................................................................................................................. 138 Figure A-4: Post-treatment protocols followed for further processing of the exfoliates obtained from the divided setup. ......................................................................................................................... 138 Figure B-1: Sample ORR analysis results for Vulcan XC-72 in 0.1 M KOH solutions with the sweep rate of 5 mV s-1 at 293 K. .......................................................................................................... 141 Figure B-2: Performance comparison between various graphene products in this study and the Vulcan-based samples in 0.1 M KOH solutions with the sweep rate of 5 mV s-1 at 293 K. ...... 142 Figure C-1: Single-chamber MFC design using Solidworks with exact specifications. ................ 143 Figure C-2: Bacteria growth protocol. ................................................................................................ 143 Figure C-3: Phosphate buffer solution preparation protocol. ......................................................... 144 Figure C-4: Tryticase soy agar/broth bacteria culture medium preparation protocol. ................ 144       xix List of Symbols Ea anode potential (V) CB bulk concentration (mol m-3) Ec cathode potential (V) ηc cathodic overpotential (V) ECell cell potential (V) Rc contact resistance (ohms m2) i current density (A m-2)  D diffusion coefficient (m2 s-1) δ diffusion distance (m) iL diffusion limiting current density (A m-2) ηD diffusion overpotential (V) Re electronic resistance (ohms m2) ΔH enthalpy change (J mol-1) ΔS entropy change (J K-1 mol-1) Eeq equilibrium potential (V) i0 exchange current density (A m-2) F Faraday’s constant (C mol-1) ΔG Gibbs free energy change (J mol-1) Ri ionic transport resistance (ohms m2) Vm molar volume (m3 mol-1)  n number of exchanged electrons k permeability coefficient (m2)    xx P pressure (Pa) ΔP pressure drop (Pa) Δx sample thickness (m) CS surface concentration (mol m-3) T temperature (K) RT total resistance (ohms m2)  α transfer coefficient R universal gas constant (J mol-1 K-1) v velocity (m s-1) µ viscosity (Pa s) ΔV voltage drop (V)      xxi List of Abbreviations BMPyrr 1-butyl-1-methylpyrrolidinium 1D 1-dimensional EMIM 1-ethyl-3-methylimidazolium 2D 2-dimensional 3D 3-dimensional ACN acetonitrile AC alternating current ARPES angle-resolved photoemission spectroscopy AFM atomic force microscopy BSL biosafety level BTA bis(trifluoromethylsulfonyl)imide BET Brunauer-Emmett-Teller N1114 butyltrimethylammonium CE/MS capillary electrophoresis/mass spectrometry CB carbon black CC carbon cloth CNS carbon nanostructures CCM catalyst coated membrane CVD chemical vapor deposition CE counter electrode CV cyclic voltammetry DC direct current     xxii EIS electrochemical impedance spectroscopy EGN electro-exfoliated graphene eV electron volt EDX energy-dispersive X-ray spectroscopy FFP flow field plate FCC fluid catalytic cracking GC gas chromatography GDL gas diffusion layer GO graphene oxide GIC graphite intercalation compound HOPG highly oriented pyrolytic graphite HPLC high-performance liquid chromatography IC ion chromatography MEA membrane electrode assembly MOE mercury/mercury oxide electrode MPL microporous layer MOSC mixed oxide super-conductor NHC N-heterocyclic carbenes NMP N-methyl-2-pyrrolidone  NHE normal hydrogen electrode NL normal liter OCP open circuit potential PES photoemission spectroscopy PTFE polytetrafluoroethylene    xxiii PC propylene carbonate PEMFC proton exchange membrane fuel cell REE red-edge effect rGO reduced graphene oxide RE reference electrode RDE rotating disk electrode SEM scanning electron microscope SHE standard hydrogen electrode BF4 tetrafluoroborate TEM transmission electron microscope DOE U.S. department of energy UV ultraviolet VIS visible WE working electrode XPS X-ray photoelectron spectroscopy      xxiv Acknowledgements I would like to express my appreciation to my supervisor, Dr. Előd Gyenge, who provided me with this opportunity to be a part of his research group. His support, patience, and constant guidance helped me overcome many of the challenges that I faced during my studies. I would like to acknowledge Dr. Kevin Smith, and Dr. John Madden for reviewing my thesis, and being a part of my research committee. This extends to Dr. Jim Lim, Dr. Edouard Asselin, and Dr. Peter Pintauro who kindly reviewed, and commented on my thesis as external examiners. I would also like to thank Bradford Ross and Derrick Horne from the Bioimaging Facility in the Department of Botany at the University of British Columbia (UBC) for FESEM and TEM training, respectively. Furthermore, I appreciate contributions by Drs. Ken Wong and Philip Wong, from the Interfacial Analysis and Reactivity Laboratory at UBC, to the XPS and Raman analysis, respectively. My gratitude extends to Dr. Louise Creagh, the advisor of undergraduate biotechnology laboratory, for her support throughout the bio-related operations, and proper disposal of the chemicals. I am very grateful to my family, particularly my parents and my brother, who have always been supportive and cheering in every aspect of my life, especially regarding continuing my education. The same goes to my dear friends and lab mates, Sayyed Soroush Nasseri, Amir Mehdi Dehkhoda, Pooya Hosseini, and Andrew Wang who always encouraged, and helped me whenever I was in need. Last but not least, I would like to thank the Almighty for all the failures and successes of my life as I have learnt a lot from failures, and gained a lot from successes! "Knock, And He'll open the door. Vanish, And He'll make you shine like the sun. Fall, And He'll raise you to the heavens. Become nothing, And He'll turn you into everything." –Rumi      xxv Dedication  TO MY BELOVED PARENTS AND BROTHER  1  Chapter  1: Introduction Graphene, the two-dimensional planar sheet of sp2-bonded carbon atoms, is among the most researched nanomaterials since it was isolated, and studied in freestanding form by Geim and Novoselov in 2004.1-2 Considering its numerous unique properties, graphene holds a great promise for many key research areas.3 It has a large specific surface area (2,630 m2 g−1),4 high intrinsic charge mobility (200,000 cm2 V−1 s−1),5 strong Young’s modulus (∼1.0 TPa),6 and excellent thermal conductivity (∼5,000 W m−1 K−1),7 while showing significant optical transmittance (∼97.7%) suitable for transparent conductive electrodes.8-9 Therefore, graphene can be used for diverse applications such as composite materials,10 supercapacitors,11 graphene-based electronics,12 molecular gas sensors,13 batteries,14 fuel cells,15 solar cells,16 transistors,17 and biosensors.18 Nonetheless, the commercial practical success of many graphene-based applications is still unproven, and among a number of factors, the need to cheaply produce large quantities of high-purity defect-free graphene, is a major impediment. The latter requirement is reflected by a large number of publications on various graphene production techniques. Moreover, successful integration of the synthesis products within the existing platforms has become another critical challenge, and there are not many compelling technologies coming through. Figure 1-1 shows graphene as a building material for all other carbon dimensionalities with myriad of applications.         2  In general, there are two fundamentally different approaches to prepare graphene sheets (GNs) as single-(or a few) layers of atoms, namely top-down and bottom-up (Figure 1-2).20-21 The top-down approach starts with macroscopic structures, breaking them down into smaller ones. In fact, Geim and Novoselov developed a top-down approach so-called “micromechanical cleavage” to extract single sheets of atoms from three dimensional graphitic crystals using scotch tape exfoliation.2 Graphite, as an earth-abundant starting material for the top-down preparation of graphene, offers a cost-efficient and environmentally-friendly alternative to bottom-up nano-carbon based synthesis. However, the key challenge here is to surmount the strong cohesive energy of the π-stacked layers in graphite (5.9 kJ mol−1 carbon).22-23 Other top-down methods include wet chemical or electrochemical synthesis from graphite intercalation compounds (GICs),10,24-25 direct liquid phase exfoliation,26 and solution-based chemical reduction of graphene oxide (GO).27-28 GNs produced by the top-down approach are usually mixtures of monolayers, bi-layers and multilayers (typically composed of three to ten monolayers), in the form of irregularly structured  Figure 1-1: Graphene as a building material for all other carbon dimensionalities with myriad of applications. It can be wrapped up into 0D buckyballs, rolled into 1D nanotubes or stacked into 3D graphite (right, Reprinted with permission from Ref.19, Copyright 2007, Nature Publishing Group).   3  flakes (flat or folded sheets).29 This morphology is well-suited for the majority of GNs applications in the energy sector, both for energy storage and conversion.3,30  Production of GNs with only a few layers and defect free is clearly most desirable. Furthermore, employing a straightforward top-down pathway is difficult to produce large pieces of GNs (in the size of mm) that would be needed for fabricating chips or other electronic devices. Therefore, bottom-up techniques have been intensely investigated for technological applications, in which single-layers of graphene are grown in a step-wise manner from carbon atoms. This is mainly achieved via epitaxial growth of GNs on a substrate by chemical vapour deposition (CVD),12,31 solvothermal synthesis,32-33 pyrolysis,34-35 and thermal decomposition of a SiC (silicon carbide) wafer under ultrahigh vacuum conditions.36-37 One of the challenges for the bottom-up approach is the difficulty of producing freestanding GNs since typically a substrate is required to support and direct the step-wise deposition. All the aforementioned processes have been studied in detail, and described in several review articles.38-40 Additionally, a high-throughput approach to determine the number of atomic planes in GN samples has been recently introduced via optical microscopy and subsequent image processing; which only applies to the flakes with relatively large lateral sizes.41-42 Here, the focus is  Figure 1-2: Graphical illustration of two different approaches for synthesizing 2D graphene structures: bottom up (left – credits to Minot group) vs top down (right – credits to GLAB).   4  on scalable top-down GNs production techniques considering the demanding industrial applications of graphene especially in energy and environment sector. 1.1 Thesis motivation and novelty With regards to graphene production methods, it is worth noting that the vast majority of samples utilized in the chemical engineering research are prepared via chemical reduction of graphene oxide (GO) using strong oxidizers/reducers, and excess of organic solvents, where applicable.43 Such methods are not only environmentally detrimental, but also seriously disrupt the sp2 lattice structure resulting in significant differences between the end product and pristine graphene.44 Consequently, electrochemical techniques are developed to address some of the aforementioned issues.24 They usually involve an ionic conductive solution (electrolyte) paired with a direct current (DC) power supply to render structural changes within the graphitic precursor (e.g., rod, plate, or wire used as the electrode) upon ionic intercalation. This offers several advantages such as ease of operation, tunability of the entire synthesis process, and the elimination of harsh chemicals with fast exfoliation rates. Such approaches also create fewer lattice defects, as opposed to the graphene films chemically reduced from GO derivatives.45  Figure 1-3 illustrates where electrochemical techniques stand amongst other approaches in terms of overall cost and quality. Being placed at the top left corner of the chart, they present one of the highest quality/cost ratios thus worth investing more time and effort to make further progress in this field. Factors such as intrinsic conductivity, defect density, and structural disorder have been used as the general measures for the graphene quality. For instance, although electrochemical methods bring the extra cost of using power sources, their superior yield compared to the shear-based synthesis result in lowering the overall cost of this technique. Further in-depth discussions along with the related references are presented in Chapter 2.   5   Figure 1-3: Cost and quality comparison between various graphene synthesis techniques. Factors such as intrinsic conductivity, defect density, and structural disorder have been used as the general measures for the graphene quality. The illustration does not represent any quantitative scale (similar analogies have been made by Novoselov et al.46 and others) Considering the limitation of water electrolysis in aqueous electrolytes, non-aqueous solvents are generally employed to provide a wide electrochemical window.25,45,47-48 Among those, ionic liquids have exhibited remarkable tendency to intercalate graphitic electrodes, and yield gram-scale quantities of carbon nanostructures (CNS) and GNs.24,45 Air- and moisture-stable room temperature ionic liquids (RTILs), salts of large organic cations with relatively bulky inorganic counter-ions, are molten salts with melting points close to room temperature.49-50 Owing to low vapor pressure, high chemical and thermal stability, solvating capability, non-flammability with potential recyclability, RTILs have received pronounced interest as ‘green’ solvents in organic synthetic processes to replace classic, toxic and volatile molecular solvents.51-52    6  The first published work on graphene electrosynthesis using ILs goes back to 2008 when Liu et al. reported production of GNs with 1:1 IL/water volume ratios.48 They applied 10-15 V to an electrochemical cell with graphitic anode and cathode; where exfoliation was observed at the anode. This line of research was later pursued by Lu et al., focusing mainly on the complementary role of water as cosolvent during the exfoliation process.45 However, none of the aforementioned studies discussed nor defined the yield for GN synthesis by electrolysis for the various types of ILs and different IL/water ratios employed. Thus, it is uncertain how practically feasible the electrosynthesis process could be for the production of specific nanostructures.  Another aspect that has been previously neglected is related to the possible parasitic reactions of the IL molecules on the electrode surface. The adsorption, intercalation and possible electrode reactions of the IL moieties can have a major impact both on the exfoliation process, and the overall process efficiency. The possible functionalization pathways to produce different classes of surface modified GNs for various applications are still remained unexplored. Additionally, use of cosolvents to modify the physicochemical properties of the electrolyte is another missing part in the literature. IL dilution in cosolvents such as acetonitrile and propylene carbonate could potentially enhance the exfoliation yields via co-intercalation while lowering the IL consumption. Also, the prospects of extending the exfoliation process to both electrodes thus increasing the yield is unexplored along with the successful integration of the products in various electrochemical power sources. Hence, the described potential research avenues became the major motivations, and consequently novelties for this dissertation.  1.2 Thesis objectives According to the thesis motivation, the major objectives are broken into four categories, and as explained in the next section, there are chapters dedicated to addressing each in detail.   7  i. Single electrode exfoliation: The very first objective of this research is to identify proper IL-based electrolytes for the graphene electrosynthesis process. The suitable candidates must possess some crucial properties including high ionic conductivity, good electrochemical stability, non-flammability, and very low vapor pressures. Another key factor along with fine-tuning the electrochemical settings (e.g., voltage, current, cell design) is the material used for the carbon-based electrode as a precursor which could vary dramatically in shape/structure, and respond quite differently to the intercalation process. ii. Simultaneous electrode exfoliation: Previous research focused mostly on either anode or cathode exfoliation. Consequently, in single graphite electrode studies, at the non-graphitic counter-electrode (e.g., platinum), unwanted electrode reactions such as gas evolution and electrolyte decomposition take place, leading to significant energy and chemical losses. Thus, identification of the electrolytes capable of exfoliating both electrodes is considered the second major objective in this thesis. iii. Product application challenge I: The microporous layer (MPL) has been established as a key component in proton exchange membrane fuel cells (PEMFCs) due to its beneficial influence on mass transport and the associated reduction of electrical contact resistance. The objective here is to explore GNs performance compared to the traditional carbon black materials, and investigate the potential improvements. iv. Product application challenge II: Use of the exfoliation products as anode-modifying materials in microbial fuel cells (MFCs). Currently, such devices suffer from low power density due to slow metabolism of the bacteria impregnated onto the anode. Graphene could potentially improve the situation due to its biocompatibility along with the high surface area. Thus, successful integration of GNs in MFCs is pursued as the last objective.   8  1.3 Thesis approach In accordance to the four major objectives elaborated above, the thesis approach is set in a way to cover each in the form of a chapter. They are all located after Chapter 2 which includes literature review. The following is a detailed description of how each of the objectives has been met with respect to their presentation in below chapters.  In the first step, Chapter 3 (addressing Objective i) focuses on single electrode exfoliation experiments, and further yield improvements. Various classes of ILs are tested for anodic exfoliation as an immediate follow up to the most recent literature. A cosolvent is added to the electrolyte to dramatically lower the IL loading, and further studies on the exfoliation mechanism are cross-checked with literature. This is coupled with proper characterization techniques to evaluate the products’ quality, and more comments are made on the role of various graphitic precursors on the exfoliation phenomenon. Alternative explanations are offered regarding the online exfoliation patterns observed in aprotic ionic liquids. Moreover, comments have been made on the role of operating conditions especially the applied voltage on the exfoliation process. Chapter 4 (addressing Objective ii) covers simultaneous electrode exfoliation experiments which were introduced for the first time. The rationale behind developing electrolytes capable of exfoliating both involved electrodes is discussed as well as side-by-side comparisons with single-electrode exfoliation experiments. More understanding is achieved in this chapter on the cyclic voltammetry behavior of the intercalating ions along with interesting observations over the exfoliation patterns on the electrodes. Synergistic effect of having exfoliation on both electrodes involved in the process is carefully measured, and documented.  Chapter 5 (addressing Objective iii) demonstrates the GN application as microporous layers for proton exchange membrane fuel cells. The gas diffusion layer particularly on the cathode   9  side is modified with graphene materials to investigate its impact on the fuel cell performance. Oxygen reduction on the cathode in hydrogen fuel cells generates appreciable amounts of water that needs to be managed properly; otherwise, it can adversely affect the overall performance as well as degrading the life time of the device. There is a brief introduction to the concept at the beginning of the chapter to further highlight the novelties of the presented work. The cells are subjected to a wide range of operating conditions, and their intriguing performance implications are discussed as the result. Finally, Chapter 6 (addressing Objective iv) presents the results of incorporating the exfoliated products as anode-modifying materials for microbial fuel cells. It starts with providing a proper background for the research status in the field, introducing novel cell designs in conjunction with the new electrode materials. In addition to the graphene electrodes, other carbon morphologies including nanotubes and spherical carbon black particles are tested, and compared with the exfoliated products. This reveals more about the morphology effects as well as the surface chemistry role for maximizing the performance of a particular material.         10  Chapter  2: Literature review  2.1 Theory and terminology From condensed matter standpoint, graphene is constructed by hybridization of s, px and py atomic orbitals forming sp2-bonded carbon atoms via three strong σ bonds with three adjacent atoms. The remaining pz orbital on each carbon center overlaps with those from the neighbouring atoms, establishing a filled band of π orbitals (valence band), and an empty band of π* orbitals (conduction band). Since the valence and conduction bands touch at the Brillouin zone corners, graphene emulates a zero-band-gap semiconductor.12 Figure 2-1A depicts the described energy distribution in graphene, and the corresponding inset for zero-band-gap at one of the Dirac points (i.e., the intersection of the valence and conduction bands).53 The abovementioned electronic structure can be readily manipulated via doping of graphene in the following ways: (i) electrical doping (by changing gate voltage),54-55 (ii) substrate-induced doping (by interactions with the support),56 and (iii) chemical doping (by insertion of external chemical species).19,57 Figure 2-1B illustrates the strong ambipolar field effect in pristine graphene (i.e., mismatch between Fermi level as the energy level of electrons and Dirac point) resulted from electrical doping. Figure 2-1C shows the Dirac point shift relative to the Fermi level prompted by chemical doping, and draws a simultaneous comparison with the substrate-induced doping. There will be further discussions throughout this chapter on graphene’s electronic properties, and chemical doping concept. Thus, the described theory and terminologies are helpful for further understanding of the upcoming sections.   11  2.2 Top-down synthesis techniques As presented in Chapter 1, chemical oxidation of graphite to graphite oxide followed by exfoliation to graphene oxide (GO) and then reduction of GO to graphene by chemical or thermal reduction  Figure 2-1: (A) Left: the electronic structure of the sp2 bonded graphene. Right: zoom-in of the Dirac point where valance and conduction bands intersect. Reprinted with permission from ref. 53, Copyright 2009, American Physical Society. (B) Ambipolar electric field effect in single-layer graphene (the position of the Dirac point and the Fermi energy (EF) are shown in the insets as a function of gate voltage). Reprinted with permission from ref. 19, Copyright 2007, Nature Publishing Group. (C) A schematic diagram for the Dirac point and the Fermi level positions as a function of doping; where the upper panel is n-type doped, pristine, and p-type doped free-standing graphene (a to c), and the lower section illustrates n-type doped, pristine and p-type doped epitaxial graphene grown on silicon carbide (SiC) (d to f). Reprinted with permission from ref. 56, Copyright 2008, American Chemical Society.   12  has emerged as a promising method due to its low-cost and mass production potential.10,38-39,58-59 Graphite oxide is usually synthesized by oxidizing graphite via concentrated sulphuric acid, nitric acid, and potassium permanganate based on the so-called Hummers method.60-61 It is important to note that although graphite oxide and GO share similar chemical properties (i.e., surface functional groups), their structures are different. GO is a monolayer material produced by the exfoliation of graphite oxide.38 Figure 2-2 depicts the graphene production pathway via effective reduction of the mono-layered graphene oxide sheets. Several reducing agents are proposed for chemical reduction of GO sheets including hydrazine,62-65 and sodium borohydrate.66-67 Hydrazine hydrate, unlike other strong reductants, is nonreactive with water, and is suggested to be the most effective in synthesizing ultrathin and fine graphene nanosheets.38 Throughout the reduction process, the brownish GO dispersion in water turns black and the reduced sheets begin to agglomerate and precipitate in the reaction vessel.59,68 Such trends in color changes and dispersibility of the reduced GO (rGO) in water are rendered by the removal of oxygen atoms from the GO sheets, making the final product more blackish and less hydrophilic.38   13  The restoration mechanism of the conjugated GN network through hydrazine-assisted GO reduction has been proposed by Stankovich et al. in Scheme 2-1.59 First, hydrazine partakes in a ring-opening reaction with epoxides and transforms the epoxides into hydrazino-alcohols.69 This derivative then reacts further, converting to an amino-aziridine moiety which goes through thermal elimination of di-imide to form a double bond.59 Importantly, without surfactant- or functionalization-assisted stabilization, rGOs tend to agglomerate in organic solvents due to their hydrophobicity.59,68   Figure 2-2: Graphene production pathway via effective reduction of the mono-layered graphene oxide sheets. Reprinted with permission from ref. 44, Copyright 2011, Wiley-VCH.   14   Scheme 2-1: Proposed reaction pathway for epoxide reduction with hydrazine. Reprinted with permission from ref. 59, Copyright 2007, Elsevier.  Sodium borohydride (NaBH4), on the other hand, is a more common reducing agent in the synthetic chemistry,70 and reported to remove GO’s oxygen-containing groups more effectively compared to hydrazine.66 Using borohydride solutions, Bourlinos et al. recorded lower sheet resistances of 59 kΩ m-2 (comparing with 780 kΩ m-2 for a hydrazine reduced sample under similar circumstances), and higher C:O ratios of 13.4:1 (compared to 6.2:1 for hydrazine).66 However, the tetrahedral BH4− anions are prone to the hydrolysis reaction in aqueous media, lowering their reducing ability against carbonyl groups on GO’s surface.38  GO reduction approaches are not limited to the abovementioned chemicals, and more comprehensive reviews are available in this regard.70-71 Use of gaseous hydrogen (after thermal expansion),72 hydroquinone,73 solvothermal methods,74 and strong alkaline solutions75 are among the alternatives; which only hydrogen-assisted reduction has produced comparable results to those of hydrazine and sodium borohydride.38 Nevertheless, they all leave a significant amount of oxygen impurities behind, resulting in numerous lattice defects.43-44 High temperature annealing is therefore required to lower the GN network defects,28,76 and further re-establish the sp2-bonded structure.77  2.3 Electrosynthesis approach It is important to note that harsh oxidizers (e.g., H2SO4/KMnO4), and organic solvents (e.g., dimethylformamide/tetrahydrofuran) used throughout the GO synthesis are not environmentally   15  benign.39,78 Additionally, the severe poisonous and explosive characteristics of hydrazine or sodium borohydride derivatives for the GO reduction dictate safety measures when large quantities are used, making the process challenging in real conditions.79 A handful of environmentally friendly processes are available to reduce GO to graphene either by chemical or electrochemical pathways;75,80 however, an integrated green approach to the synthesis of graphene has not been developed yet. To address the above mentioned issues, researchers have recently utilized electrochemical methods as a part of the GN fabrication process.24-25,45,47-48,81-83 In principle, GN electrosynthesis employs an ionically conductive solution (electrolyte) and a DC power source to drive the structural changes through the graphite precursor (e.g., rod, plate, or wire) placed as the working electrode (WE). This offers a number of potential advantages including ease of operation and control over the entire synthesis process, being more environmentally benign with elimination of harsh oxidizers/reducers, relatively fast fabrication rates, and high mass production potential at ambient pressure/temperature. Presumably, direct exfoliation of the graphene sheets from graphite would overcome the low electronic conductivity of graphene films chemically reduced from GO derivatives.45,59  The ionic electrochemical intercalation experiments were started in 1980s with intercalation of pure sulphuric acid into a graphite particulate system84 followed by the Li+ cations intercalation from butyrolactone solutions,85 and F- anions from aqueous86 and anhydrous HF electrolytes.87 Interestingly, there are some early reports about surface blister formation on highly graphitic crystals due to the intercalation,88 along with exfoliation of carbon fiber electrodes upon formation of electrochemical GICs.89 However, the controlled experiments aimed at high-yield   16  graphite exfoliation received the major attention after the groundbreaking studies of Geim and Novoselov on the exceptional physicochemical properties of graphene.2 The experiments designed for graphite exfoliation involve a graphitic working electrode (WE), a counter electrode (CE), which may vary in material selection, and different intercalants in the form of electrolyte solutions (aqueous, non-aqueous or their combination). Upon applying proper external voltage to the configuration, the target anions (or cations) begin intercalating within the graphite galleries of the anode (or cathode), and cause structural expansion that could result in releasing ultrathin graphene-like materials into the solution. Figure 2-3 shows the schematic of a sample electro-exfoliation process with graphite rod used as the anode WE, and platinum wire as the cathode CE to produce GN flakes. Depending on the nature of the study, a standard reference electrode can be also used to further monitor the surface voltage changes of the WE during the process.  Figure 2-3: Schematic of a sample electro-exfoliation process with graphite rod used as the anode WE, and platinum wire as the cathode CE to produce GN flakes.   17  Generally, the applied voltage to the graphite WE is in the form of a constant voltage, and it could be also cycled between some predetermined values based on cyclic voltammetry to improve the exfoliation yield. However, constant current experiments as well as alternating ones could be also used depending on the situation. Consequently, WE starts expanding and peeling off with black precipitates accumulating at the cell’s bottom while some other remaining stable in the electrolyte. Particularly in aqueous solutions, this is thought to be a combination of graphite oxidation (or reduction) thus accommodating the anions (or cations) accompanied by severe oxygen evolution (or hydrogen evolution) for subsequent structural expansions. Due to limitations imposed by water electrolysis in aqueous electrolytes, non-aqueous solvents or mixtures of aqueous and non-aqueous components are generally employed to provide a wider electrochemical window.25,45,47-48 This could significantly lower the energy consumption of the electrochemical cell with removing serious parasitic reactions at the electrodes (oxygen and hydrogen evolution), while providing a platform for much stronger intercalation of the ions.  Amongst anionic electro-exfoliations, ionic liquids have shown remarkable tendency to intercalate graphite electrodes, yielding gram-scale quantities of carbon nanostructures and GN flakes.24,45 Air- and moisture-stable room temperature ionic liquids (RTILs), salts of large organic cations with relatively bulky inorganic counter-ions, are molten salts with melting points close to room temperature.49-50 Owing to low vapor pressure, high chemical and thermal stability, solvating capability, non-flammability with potential recyclability, RTILs have received much interest as green solvents in organic synthetic processes to replace classic, toxic and volatile molecular solvents.51-52 Consequently, ILs are extensively used as a reaction medium for the fabrication of conducting polymers and nanoparticles.90-93 Figure 2-4 presents the graphical comparison between   18  ionic solids with symmetric ions and ionic liquids with at least one asymmetric component resulting in dramatic drop in the melting point. In cationic intercalation experiments, lithium co-intercalation with propylene carbonate (PC) has been proposed for production of few-layer graphene flakes (> 70%).47 The process resembles the destructive effects of PC as a molecular solvent on the graphitic electrodes in lithium-ion batteries.94-95 Intercalation of tetraalkylammonium cations from 1-methyl-2-pyrrolidone (NMP) solutions is also reported with significant exfoliation rates as a viable choice for GN synthesis with low energy consumption and ease of operation.96 In this research, the primary focus is on the ionic liquid solutions as target electrolytes for the electro-exfoliation of graphite. The preliminary results from IL-assisted graphene electrosynthesis revealed their high-throughput potentials for intercalation into the graphitic structures, and the subsequent electro-exfoliation. Additionally, their flexibility in physicochemical properties enables in-situ functionalization of GNs which is the most important way for tailoring the graphene applications in many key research areas.97-98  Figure 2-4: Graphical comparison between ionic solids with symmetric ions and ionic liquids with at least one asymmetric component resulting in dramatic drop in the melting point.   19  2.4 Chemical functionalization From chemistry viewpoint, functionalization and dispersion of graphene sheets are of crucial importance for their end applications. This enables graphene to be processed by solvent-assisted techniques such as layer-by-layer assembly, spin-coating, and filtration.97,99 It also avoids agglomeration of graphene monolayers throughout GO reduction process, and could preserve graphene’s intrinsic properties if necessary. Hence, surface modified graphene has been utilized for the fabrication of polymer nanocomposites,100 super-capacitor devices,101 drug delivery systems,102 solar cells,16 memory devices,103 transistor devices,104 biosensors,105 etc. GO is widely used as a precursor for the synthesis of processable graphene, owing to its highly oxygenated surface with abundant hydroxyl, epoxide, diol, ketone, and carboxyl functional groups.106-107 As shown in Figure 2-5, such oxygen-containing functionalities can result in a broad range of dispersibility in water and various organic solvents via alteration of the van der Waals interactions.108-112 Edge-positioned carbonyl and carboxyl groups afford strong hydrophilicity for the GO sheets, making them readily dispersed in water.113-114 According to various functionality scenarios, different model structures have been proposed for GO illustrated in Figure 2-6.112,115 As discussed in Section 2.2, chemical reduction of GO lead to a rapid irreversible precipitation due to the agglomeration of GNs, unless a stabilizer is used. Therefore, to address this issue, GO surface is modified prior to reduction, which is usually performed either by noncovalent or covalent functionalization.99,116 Serving an example, reduction of alkylamine-modified GO results in stable dispersion of functionalized graphene sheets in organic solvents. Impregnation of carboxylic or sulfonate groups on the basal planes of graphene is also reported to afford producing water-dispersible graphene sheets.117-119 On the other hand, a number of papers have discussed synthesis of functionalized graphene sheets directly from natural   20  graphite.43,120-121 For instance, ionic-liquid-assisted graphene electrosynthesis from graphitic anodes remains functional groups on the ultrathin exfoliates.48,122 Nonetheless, obtaining high-yield dispersions of non-functionalized GNs has sparked a broad wave of research.120,122  Figure 2-6: Variations in the proposed model structures for the GO sheets regarding the presence (left, Reprinted with permission from ref. 111, Copyright 1998, American Chemical Society) or absence (right, Reprinted with permission from ref. 112, Copyright 2009, Nature Publishing Group) of carboxylic acids on basal plane edges. Discussing the functionalization approaches, noncovalent graphene functionalization mainly involves physical surface adsorption of molecules/atoms via hydrophobic, van der Waals, and electrostatic forces (equivalent to surface transfer doping). This is a well-established concept for the surface modification of carbon-based materials, and it has been frequently employed to  Figure 2-5: As-prepared graphite oxide dispersions in water and 13 organic solvents by bath ultrasonication for 1 h. Top: Dispersions immediately after sonication. Bottom: Dispersions three weeks after sonication. The yellow color of the o-xylene sample is due to the solvent itself. Reprinted with permission from ref. 110, Copyright 2008, American Chemical Society.   21  alter the physicochemical properties of the sp2 networks in CNTs.123-128 Successful cases include adsorption of surfactants or small aromatic molecules, polymer wrapping, and biochemical interactions with DNA and peptides.129-132 Consequently, similar techniques can be borrowed for further surface modification of graphene via various types of organic compounds.  In contrast with noncovalent interactions, covalent functionalization involves rehybridization of sp2 carbon atoms through the graphene network into the sp3 configuration, associating with the loss of electronic conjugation (equivalent to substitutional doping).133 This is mainly due to the rigorous, and often irreversible, interactions of the dopant with the honeycomb graphene. Such strong surface modifications can be achieved in a variety of ways including nucleophilic substitution, electrophilic addition, condensation, and combinatory addition; which most of them use GO as the starting material.134-136  The mechanism of chemical doping in graphene resembles that of carbon nanotubes, however the latter is still debated.137-140 Surface transfer doping (i.e., noncovalent) occurs through charge transfer from the adsorbed dopant (or graphene) to graphene (or dopant).56,141 Charge transfer direction is determined by the relative position of density of states (DOS) at the highest occupied molecular orbital (HOMO), and the lowest unoccupied molecular orbital (LUMO) of the dopant compared to the Fermi level of graphene. If the dopant`s HOMO is above the graphene`s Fermi level, charge flows from dopant to graphene, and the dopant is considered donor; conversely, if the LUMO is below the Fermi level of graphene, charge transfers from graphene layer to dopant, and the dopant acts as an acceptor.53,142 Consequently, graphene can be p-type or n-type doped according to the described electron exchange patterns.142-143 p-Type doping lifts the graphene`s Dirac points above the Fermi level, while n-type doping pushes the Dirac points below the Fermi level. In general, surface-adsorbed molecules with electron-withdrawing   22  groups (electronegativity) induce p-type doping in electronic structure of graphene, and molecules with donating groups (electropositivity) lead to n-type doping. In contrast, the mechanism for substitutional doping (i.e., covalent) in pristine graphene still remains uncertain. For this category, p-type doping mostly takes place by adding atoms with fewer valence electrons than carbon (e.g., boron), while n-type doping is induced by adding atoms with more valence electrons than carbon (e.g., nitrogen). After years of research, it has become more evident that both covalent and noncovalent modifications significantly improve the processability of graphene.144-146 However, this adversely affects the electrical conductivity of surface modified GNs, and is observed to decrease the electron mobility noticeably compared to that of pristine graphene.107,147 In addition, both functionalization techniques severely decrease the surface area of the modified GNs due to the structural damages induced by graphite chemical oxidation, subsequent sonication, functionalization, and chemical reduction.97,99 To address the aforementioned issues, several studies have examined preparing functionalized GNs directly from graphite in a one-step process using, for instance, ionic liquids and surfactants via direct liquid- and electro-exfoliation methods.83,122 All the related studies have acknowledged the effectiveness of graphene functionalization in preventing GNs agglomeration in various solvents, and affording more physicochemical properties for a broad range of applications. As mentioned earlier, graphene is highly sensitive to surface transfer doping, and most of such doping processes are reversible. This two-dimensional material with its ultrahigh conductivity, which alters swiftly upon atomic/molecular absorbance, is therefore an excellent candidate for high-precision sensors. A linear conductivity response to various NO2 concentrations is reported (Figure 2-7a), which greatly facilitates the application of graphene-based   23  sensors to detect even individual NO2 molecules.13 This has become a strong incentive to develop high performance sensors based on graphene chemical manipulations.148-152 Figure 2-7 presents other encouraging cases of gas sensing and the related electronic effects on GNs. Graphene-based biosensors for detecting bacteria,153 glucose,154 pH and proteins155-156 have also been successfully fabricated. Additionally, DNA detection via GO is reported,157-158 while graphene is suggested to be an ideal material for DNA sequencing.159-164  Figure 2-7: (a) Linear dependency of chemically induced charge transport, △n, to various concentration, C, of NO2. Lower inset: graphene characterization by the electric field effect. (b) Changes in resistivity, ρ, at zero-band-gap point of graphene induced by exposure to various diluted gases (1 ppm). The positive (negative) sign of changes indicate electron (hole) doping. (c) Constant mobility of charge carriers in graphene with increasing chemical doping. The parallel shift is caused by a negligible scattering effect of the charged impurities induced by chemical doping. Reprinted with permission from ref. 13, Copyright 2007, Nature Publishing Group. (d) Schematic of charge transfer at the F4-TCNQ/graphene interface. Reprinted with permission from ref. 165, Copyright 2007, American Chemical Society.   24  It is important to achieve controllable, air-stable and high-performance n-type, p-type or even ambipolar field effect throughout the doping process. Pristine graphene and rGO often share similar p-type characteristics at ambient conditions due to the involuntary doping originated from chemical residues or oxygen molecules in the air, which is not fully explored yet. However, graphene embedded in insulating polymer matrices, like polystyrene (PS), is noticeably less affected by such ambient doping.166 Consequently, graphene as a buffer layer between the electrodes and the organic semiconductors of organic field effect transistors (OFETs) has exhibited competitive work functions similar to those for Ag and Cu.167-168 Moreover, the analogous molecular structure of graphene and organic semiconductors like pentacene affords strong π–π interactions amongst them, lowering the carrier injection barriers, thus improving the device overall performance (Figure 2-7d). Other applications are also reported including use of nitrogen-doped graphene as an efficient metal-free electrocatalyst for oxygen reduction reaction (ORR) in fuel cells (Figure 2-8),169-170 and graphene doping with thionyl chloride (SOCl2) for achieving high electrical conductivities.171 More practical applications are anticipated to emerge soon due to the rapid progress on graphene research.   25  2.5 Characterization techniques  There are various characterization techniques to analyze graphene materials along with monitoring the possible chemical doping fingerprints. This usually starts with advanced imaging techniques such as scanning electron microscopy (SEM) as well as transmission electron microscopy (TEM). Atomic force microscopy (AFM) is a complementary tool to estimate the thickness of the sheets via nanoscale surface profilometry.   SEM is usually the first and most convenient tool here due to the reasonable cost and variety of morphological information provided by secondary electrons reflected from graphene surface. The extra advantage of GNs in this case is their conductivity which eliminates the need for sample coating with other conductive materials; which is typically done to improve the image signal and reduce the thermal damage of the high energy beams.173 Using TEM, one could obtain further information about the size and thickness of the GN flakes via electron beam passing through the ultrathin sheets. It also provides imaging at higher resolutions compared to SEM allowing a better resolution for the fine structural features.174 Lastly, AFM could be coupled with the aforementioned imaging techniques to provide an apparent estimation for the flakes’ thickness.  Figure 2-8: (a) Metal-free nitrogen-doped graphene as an efficient electrocatalyst for oxygen reduction reaction. Reprinted with permission from ref. 172, Copyright 2013, Royal Society of Chemistry. (b) Schematic illustration of various types of nitrogen-doped graphene (gray for the carbon, blue for the nitrogen, and white for the hydrogen atom). A possible defect structure is shown in the middle of the ball-stick model. Reprinted with permission from ref. 154, Copyright 2010, American Chemical Society.   26  However, due to its sensitive nature, it requires more stringent sample preparations, and the GNs must be quite flat and defect-free to obtain accurate measurements. Consequently, the literature data for graphene thickness via AFM varies from 0.34 to 1.2 nm due to variations in the sample preparation, and working modes of the AFM probes.175  For further physicochemical analysis of the GNs’ surface and possible functionalization fingerprints, three other characterizations have proven useful, namely (i) photoemission spectroscopy (PES), especially X-ray photoelectron spectroscopy (XPS), and angle-resolved photoemission spectroscopy (ARPES), (ii) Raman spectroscopy, and (iii) charge mobility measurements.  XPS selectivity towards different surface elements, and its quantitative nature has made it widely popular in the field of surface analysis. It’s mostly used for solid analysis and capable of detecting all elements except hydrogen and helium. XPS has a typical analysis depth of less than 5 nm, and is the least destructive technique amongst all electron (or ion) spectroscopy techniques.176 From XPS spectrum, the dopant (or impurity) presence can be confirmed by the characteristic peak of the representative dopant`s element, and the corresponding peak area characterizes the doping level (i.e., survey scan analysis). Furthermore, the chemical and electronic states of the dopant can be obtained by analyzing the high-resolution spectra of the element (i.e., narrow scan analysis). Additionally, ARPES is employed as a strong probe to explore the band structure of graphene at Fermi level near the Dirac point in the Brillouin zone.177-179 This is achieved by relating the energy distribution of electrons to their momentum along the high-symmetry directions measured by ARPES. Comparing the Fermi level position (at zero bonding energy) with the Dirac point (near, below or above the Fermi level) determines whether the graphene is pristine or doped.   27  Raman spectroscopy is another non-destructive technique which is frequently used to study various carbon based materials.180-181 Under controlled conditions, parameters such as number of layers, and defect density along with crystal length and stacking order can be obtained from the Raman spectrum of graphene. It’s typically represented by three major peaks so-called D, G, and 2D bands; where the D band appearing at ~1350 cm-1 is suggested to be induced by defects (i.e., missing atoms or impurities). The remaining G and 2D bands occur at ~1580 cm-1 and ~2700 cm-1 where the G peak is the most common characteristic of all carbon materials. The 2D band that appears at almost double the frequency of the D band is produced by the two-photon resonance of the photons around the Dirac points. This creates a distinct shape when compared to the 2D peak of the graphite which appears with a shoulder shown in Figure 2-9. The transition of the 2D peak from graphite to graphene is often used as a measure to estimate the number of layers.182 For electrically doped graphene, it is well-known that the G band, the main characteristic band in graphene, sharpens and upshifts in both electron and hole doping.55,183 The full-width at half-maximum (FWHM) of G band also decreases for both kinds of doping. The corresponding  Figure 2-9: (a) Raman spectra comparison between graphene and graphite measured at excitation wavelength of 514.5 nm. (b) 2D peaks comparison between graphene and graphite. (c) 2D peak evolution as a function of number of layers. Reprinted with permission from ref. 182, Copyright 2007, Elsevier.   28  carbonaceous G peak in graphene is influenced by chemical doping too. In the case of surface transfer doping, there is an empirical correlation to identify the doping type: n-type doping downshifts and stiffens the G peak, while p-type doping upshifts and softens the representative G peak (Figure 2-10).184-186 Adsorption of aromatic groups is also reported to induce an asymmetry or splitting in the G band for graphene-containing samples (Figure 2-10).187 Nonetheless, the trend for substitutional doping is entirely different, where both nitrogen- and boron-doped graphene show upshifted G bands.188 Finally, the electrical conductivity (or resistance) of any graphene-containing substrate is measurable by field effect devices. The conductivity relation with the gate voltage of pristine graphene is a characteristic V-shaped curve (Figure 2-1). Typically, the minimum conductivity (or maximum resistivity) point (i.e., the Dirac point of pristine graphene) is  Figure 2-10: Raman shift trends for the graphene`s G band upon interaction with (a) 1 M solutions of monosubstituted benzenes, and (b) with various concentrations of aniline and nitrobenzene. Reprinted with permission from ref. 184, Copyright 2008, Royal Society of Chemistry.   29  observed at the zero gate voltage; while p-type (n-type) doping of graphene shifts the point toward positive (negative) gate voltages. In the case of band-gap opening, graphene-based transistors exhibit higher on/off current ratios, and theoretical/experimental studies have suggested many novel properties for such p–n junctions often shaped by local top gate control.189-193 Promising p–n junctions have been generated via selective chemical doping, and their properties have been investigated extensively.194     30  Chapter  3: Single-electrode exfoliation experiments1 3.1 Introduction As mentioned in Chapter 1, the first published work on graphene electrosynthesis using ILs goes back to 2008 when Liu et al. reported production of functionalized GNs with 1:1 IL/water volume ratios.48 They applied 10-15 V to an undivided electrochemical cell with graphitic anode and cathode; where the exfoliation was observed only at the anode. This line of research was then pursued by Lu et al., focusing mainly on the complementary role of water during the exfoliation process.45 The latter authors showed that replacing the graphitic cathode with a platinum wire does not affect the electrosynthesis since the main processes related to GN electrosynthesis occur solely at the anode. Furthermore, they attributed the wide range of produced nanostructured carbon materials to the fine tuning of the IL/water ratios. In addition, the fluorescence behavior of the product solution was assigned to the presence of carbon nanostructures. Ultimately, they suggested IL/water ratios of ~1:0.1 for the maximum production of sheet-like structures.45 Importantly, none of the aforementioned studies discussed nor defined the yield for GN electrosynthesis for the various types of employed ILs and different IL/water ratios. Thus, it is uncertain how practically feasible the electrosynthesis process could be for the production of specific nanostructures. Another aspect that has been previously neglected is related to the possible parasitic reactions of the IL molecules on the electrode surface. The adsorption, intercalation, and possible electrode reactions of the IL moieties can have a major impact both on the exfoliation process, and the overall efficiency.                                                   1 A version of this chapter has been published in Carbon:  A Taheri Najafabadi, E Gyenge, Carbon, 71 (2014), 58-69.   31  This chapter serves as an important follow up to the described studies, and aims at obtaining more understanding of the entire electrosynthesis process. A whole host of parameters are screened here in order to further increase the electrochemical exfoliation yield, and improve the quality of final products. From electrolyte selection to testing various types of graphitic precursors as the working electrode, the results here are a stepping stone for the upcoming chapter where simultaneous exfoliation of both electrodes are achieved for the first time. Consequently, a non-aqueous electrolyte composed of ILs and acetonitrile is introduced, and compared with previously reported water-based systems. Additionally, by carrying out pertinent control experiments, the background electrode processes are explored involving IL moieties that could interfere with the anodic exfoliation process. It’s also worth noting that ionic liquid is a generic term used for a wide range of ionic materials, and any generalization of the observations is cautiously carried out here unlike the previous studies.  3.2 Materials and methods Using well-characterized carbon precursors as the working electrode is an essential requirement for the robust studies on the electrosynthesis of graphene nanostructures. It is observed that in some previous studies, ill-characterized graphitic precursors are employed such as pencil lead.25 The latter is one of the most unreliable choices since a typical 9B lead consists of more than 20 compounds including clays and binders, along with undefined graphitic phases, that differ among various manufacturers.195 This significantly undermines the reproducibility of the results when pencil lead graphite is employed. Thus, two well-prepared electrode materials are chosen, and purchased from Graphite Store for the experiments namely iso-molded graphite rod, and highly ordered pyrolytic graphite plate shown in Figure 3-1. Iso-molded graphite is the cold iso-statically pressed graphite that guarantees uniform properties in all grain directions.196 This differs from   32  extruded graphite which is pushed through a die, and has varying properties in different grain directions. There is also no binder used in the manufacturing process providing more facile flake detachment upon effective expansion. HOPG (substrate nucleated) plate, on the other hand, is produced by the decomposition of a hydrocarbon gas at a very high temperature in a vacuum furnace. The material is grown onto a substrate giving it a layered composition. The result is an ultrapure product which is near theoretical density and extremely anisotropic,197 exhibiting directionally dependent properties. In the C plane (across its layers), it has low thermal conductivity and acts as an insulator; while in the A-B plane (along the layers) it exhibits very high thermal conductivity, acting as a conductor.  Figure 3-1: Two graphitic precursors chosen for the electro-exfoliation studies (Graphite Store) namely iso-molded graphite rod (left), and highly ordered pyrolytic graphite plate (right). Figure 3-2 shows four types of RTILs selected, and purchased from Iolitec Company for this research along with their electrochemical stability windows on platinum electrode at 293 K. These candidates reasonably meet the IL selection criteria stated in Chapter 1, affording the evaluation of the roles played by different anions and cations during electrosynthesis. Comparing to Figure 3-2, most of the previous research has examined only imidazolium-based ILs.45,48 Furthermore, to alter the ILs properties while lowering their load, diluted solutions of RTILs in   33  acetonitrile are employed more frequently. It is noteworthy that acetonitrile is commonly used as a nonreactive solvent for ionic liquids especially in supercapacitor applications as it assures good ionic transport and conductivity while preserving the electrochemical potential window of the RTIL.198-199  Figure 3-2: Investigated ionic liquids and their respective electrochemical stability windows on platinum at 293 K versus normal hydrogen electrode (NHE).  Figure 3-3 shows the schematic of the experimental setup used for the exfoliation behavior of the above-described graphitic working electrodes in the presence of ionic liquid electrolytes. The graphitic precursor with 4 cm effective length exposed to the electrolyte was used as the anode with a platinum wire as the counter-electrode (1.6 mm diameter and 4 cm effective length). Various IL solutions were tested (15 ml total volume in each trial), at different constant voltages (B&K Precision-9110 DC power supply), and room temperature (293 K). The extent of graphite anode exfoliation (if any) was evaluated by measuring the electrode weight changes. The rod was bath-sonicated for 3 minutes to detach the remaining exfoliates, and subsequently washed with water/acetone to remove any liquid remnant. The final weight was measured after 12 hours of drying in the vacuum oven at 90 °C. This variable was used primarily to quantify the exfoliation yield in the presence of different ILs. The exfoliation solid products were washed with copious amounts of water and ethanol followed by filtration and ultracentrifugation with rotational speed of 4000 rpm 293 K. Sample suspensions in N-methyl-2-pyrrolidone (NMP) were sonicated for 1   34  hour to reach stable suspensions using a VWR Scientific tabletop ultrasonic cleaner (B3500-MTH). All the chemicals were of the analytical grade, and double-distilled water was used during all of the preparation steps.  Figure 3-3: Experimental setup used for graphite anodic exfoliation in the presence of ionic liquids. As for product characterization, transmission electron microscopy (TEM) images were obtained with a Tecnai G2 microscope at acceleration voltages of 20-200 kV. TEM grids were covered with 5 microliters of the diluted products in water/isopropanol mixtures to provide more facile evaporation and product visibility in imaging. Field emission scanning electron microscopy (FESEM) was performed using a Hitachi S-4700 FESEM at 3-30 kV. X-ray photoelectron spectroscopy (XPS) was done by a Leybold Max200 for the surface elemental composition and binding energy analysis with five channeltrons, using an unmonochromated Mg Kα X-ray source (1253.6 eV). Deconvolution of the XPS narrow scan profiles was subsequently carried out using XPSPEAK 4.1 software. A Micromeritics surface area analyzer (ASAP2020) was utilized to determine the specific surface area of the electro-exfoliated flakes. The fluorescence spectra of the diluted electrosynthesis solution in water were collected using a Varian Eclipse fluorescence   35  spectrophotometer. The Raman spectra were recorded via a LabRAM ARAMIS microscope Raman spectrometer with an argon-ion laser at an excitation wavelength of 632 nm. The samples deposited on aluminum substrates were exposed to the laser beam (1 micron diameter at the focus) with the exposure time of 1 second. The data was collected from 3-5 different regions of each sample to ensure the consistency of the results. All the measurements were taken at room temperature without special mention.  3.3 Results and discussion 3.3.1 Electrosynthesis experiments For the initial screening of the graphitic precursors, the experiments were conducted using 0.1 M EMIM BF4 solutions in acetonitrile. The voltage was gradually ramped up until some signs of structural changes were observed in the graphite anode. Importantly, no visible changes occurred until the verge of potential window of the electrolyte presented in Figure 3-2 (i.e., equivalent of 5  V of applied potential to the cell). However, depending on the type of graphite used, the electrodes exhibited dramatically different patterns presented in Figure 3-4. HOPG did not exfoliate directly rather expanding in the form of carbon trees (Figure 3-4, right). Even the subsequent sonication step did not help much in order to release some significant portions of the ultrathin sheets to the NMP solutions. It is thought that the large grain size of HOPG (0.5-1 mm) along with its physical attachment to the power supply clips might have prevented the complete peel off. On the other hand, the iso-molded graphite rod was found to exfoliate directly without any need for sonication. This was started within the first minute of operation, and continued at a steady pace throughout the experiments. Generally, elevating the applied voltages further boosted the aforementioned phenomena, and more studies have been conducted to study its effect accordingly. Moreover, the observations here were similar for all of the selected electrolytes (either pure or diluted in   36  acetonitrile) only differing in the extent; which will be explored throughout this chapter. Importantly, for the sake of practicality of experiments, all of the subsequent studies were carried out using iso-molded graphite rods. Not only they exhibited direct exfoliation, but also their price point advantage which is vastly lower than HOPG, makes them a much more feasible choice as a graphite precursor for graphene synthesis.  Figure 3-4: Radically different response to the intercalation of highly charged anions by two types of selected graphitic working electrodes with 4 cm of their length exposed to the electrolyte (0.1 M EMIM BF4 solution in acetonitrile at 7 V of applied potential).  Moving forward with iso-molded graphite anodes and 0.1 M IL solutions in acetonitrile, at 7 V of applied voltages, the passed current was ~100 mA for all of the selected IL types with slight decrease in the course of experiments. As mentioned earlier, depending on the IL type, the exfoliation could happen at lower potentials as well (e.g., 5 V). However, the operating conditions were fixed at the previously mentioned levels for the sake of consistent comparison between different electrolytes during 4 hours of synthesis. According to Lu et al., the GNs synthesis process by ILs electro-intercalation can be monitored based on the color changes of the electrolyte.45 They used BF4- as the anion with an imidazolium-based cation similar to EMIM (Figure 3-2, left). Furthermore, these authors claimed that the specific solution colors can be assigned to different   37  categories of products generated including fluorescent carbon nanostructures and graphene sheets.45 In Figure 3-5, the color changes during different stages of graphite electro-exfoliation are compared using EMIM BF4 (Figure 3-5a) with control electrolysis experiments where the graphite electrode was replaced by platinum wire with a similar electrolyte (Figure 3-5b). Lu et al. proposed a three-stage color change classification,45 which is in good agreement with the images shown in Figure 3-5a. They speculated that in stage I, there is an induction period before visible signs of exfoliation can be detected. In stage II, the electrolyte color changes from colorless to yellow and then dark brown which is accompanied by a visible expansion of the graphite anode (Figure 3-5a). Finally, in stage III, the expanded flakes peel off from the anode and form the black slurry with the electrolyte.45 However, nearly identical color changes were observed in the control experiment, where the graphite anode was replaced with a platinum wire, and the electrolysis was repeated under exactly the same conditions (Figure 3-5b). Therefore, the color changes cannot be used as an indicator of various carbon nanostructure formation suggested by Lu and coworkers.45 Evidently, different color changes are solely originated from IL interaction with the electrode surface (Pt or graphite).  In fact, one of the key properties of the imidazolium-based RTILs, the most extensively studied class of ionic liquids, is their Brønsted acidity200 related to the C-2 hydrogen of the 1,3-dialkylimidazolium cation.201 The electrochemical reduction of the C-2 acidic proton initiates hydrogen evolution at the cathode presented in Scheme 3-1. It also leads to the formation of the electro-generated N-heterocyclic carbenes (NHCs) and their derivatives via subsequent chain and/or decomposition reactions.202-204 This was further confirmed by the broad spectrum of products observed on the positive channel of capillary electrophoresis/mass spectrometry (CE/MS) device. Therefore, these species are mainly responsible for the solution coloration   38  presented in Figure 3-5. NHCs are highly reactive, as base/nucleophile, and can also react with the graphite anode (see Section 3.3.2, XPS analysis results). It should be noted that the trends were similar but with much higher coloration intensity for the experiments conducted in pure ionic liquid solutions. Scheme 3-1: The electrochemical reduction of the C-2 acidic proton on imidazolium-based cation which initiates hydrogen evolution at the cathode.  In addition to the visible color changes of the electrolyte, another discussion has emerged with respect to the fluorescent nature of some of the carbon nanostructures (e.g., nanoparticles and bucky gels) produced during electrosynthesis.45 Lu et al. described the obtained spectra of the alleged fluorescent carbonaceous particles as generally broad and excitation-wavelength- Figure 3-5: Electrolyte color changes during (a) electrochemical graphite exfoliation, and (b) control electrolysis experiment without graphite electrode in 0.1 M EMIM BF4/acetonitrile solution at 7 V and 293 K in the course of 4 hours (anode is on the left side in all images).   39  dependent; attributed to the size heterogeneity and distribution of different emissive sites on the carbon nanoparticles.45 This presumption can also be objected by comparing the fluorescence spectra of the electrolyte solutions obtained from the two parallel electrolysis experiments, with and without graphite anode, respectively (Figure 3-6, a and b). When the final solutions from the two sets of experiments (a, and b) were analyzed in a fluorescence spectrophotometer, they both exhibited similar excitation and emission trends (Figure 3-6). As mentioned earlier, another complexity of studying RTILs, especially in the case of imidazolium-based cations, is their unusual optical properties in the ultraviolet (UV) and near visible regions.205-206 Imidazolium containing ILs exhibit significant UV absorption, which extends  Figure 3-6: Fluorescence spectra comparison of the resulting electrolyte from (a) graphite electro-exfoliation experiment, and (b) control experiment both in 0.1 M EMIM BF4 /acetonitrile solution at 7 V and 293 K after 4 hours.   40  into the early part of the visible region. Their fluorescence trend covers a large portion of the visible domain characterized by a significant excitation-wavelength-dependence. Such shifts of the fluorescence maximum peaks are attributed to the various associated structures present in the ionic liquids and the incongruous excitation energy transfer process among them.205-206 In general, wavelength-dependent fluorescence behavior of the organized assemblies such as membranes, proteins and RTILs is observed when their dipolar moieties get excited at the red edge of the absorption band, so-called red-edge effect (REE).207-208 The described optical abnormalities become more severe under electrosynthesis conditions where NHCs instigate a series of side reactions, and further affect the unusual excitation wavelength-dependent tendencies. Thus, the complex fluorescence behavior of the IL-containing solutions, especially under electrolysis conditions, undermines any direct connection between the fluorescence spectra and release of various carbon nanostructures to the solution during electrosynthesis. In the next step, as shown in Figure 3-7, parallel experiments using the four ILs illustrated in Figure 3-2 were carried out to assess the influence of the IL structure on the graphite electrochemical exfoliation. Firstly, it can be observed once again, that the solution coloration cannot be correlated with the graphite exfoliation yield. In other words, as shown in Figure 3-7, about 86% of the graphite electrode was exfoliated when BMPyrr BTA was used as IL, while the final solution remained virtually colorless. Conversely, when the exfoliation rate was at the lowest (29%), in the case of EMIM BF4, the solution coloration was the most severe, as discussed previously in relation to Figure 3-5. Furthermore, it is important to note that no color change of the solution does not imply that the IL remains unchanged during the electrosynthesis process. In the case of BMPyrr BTA, where over 85% of the graphite anode peeled off with minimum solution coloration, mild gas evolution at the Pt cathode was observed; mostly due to the cathodic   41  decomposition of BMPyrr into methylpyrrolidine, and a butyl radical (Scheme 3-2) with its subsequent disproportionation leading to gaseous product species formation.209 Scheme 3-2: Cathodic decomposition of BMPyrr into methylpyrrolidine, and a butyl radical leading to gaseous product species formation.    Figure 3-7: Color changes and the exfoliation yield after 4 hours of ionic-liquid-assisted electrochemical graphite exfoliation at 7 V using 0.1 M IL/acetonitrile (~1:50 IL/solvent vol. ratio) at 293 K (the ± shows 95% confidence interval from 3 experiments). The low exfoliation rate in the presence of EMIM BF4 can be explained by the strong adsorption of EMIM on the cathode surface followed by the electrode reaction leading to the formation of NHCs, and subsequent chain reaction products. These species formed at the cathode   42  dissolve back to the solution and can re-adsorb on the graphite anode, in competition with BF4-, hindering the BF4- adsorption and intercalation into the graphite layers. Thus, it’s inferred that the cathodic by-products are hampering the anodic graphite exfoliation. This hypothesis is further supported by the experiments using the same anion while changing the cations, where a dramatic improvement in exfoliation rate was obtained with minimum coloration in the case of pyrrolidinium-based cations (Figure 3-7).  Switching from BF4- to more oxygenated anions such as bis(trifluoromethylsulfonyl)imide (BTA) increases the extent of complexities, but also enhances the exfoliation yield. Figure 3-8a demonstrates how the initial brownish exfoliation of the graphite anode in the first 15 minutes is followed by an intense blackish flake precipitation during four hours of electrosynthesis. Due to the oxygenated nature of the BTA anions (Figure 3-2), there is a higher chance of reactivity with the vulnerable sites of graphite anode. Similar impact was reported during graphite exfoliation with SO42- intercalation.82 It was concluded again by control experiments utilizing only platinum electrodes that the mild coloration of the electrolyte observed in Figure 3-8a after 2 hours is originated from the products of the BMPyrr BTA electrode reactions such as cathodic reduction followed by chain reactions and re-adsorption of the products on the anode. Figure 3-8b presents the passing current profile versus time; which remained around 100 mA with slight decrease towards the end of 4 hours.   43   Figure 3-8: (a) Progressive exfoliation of the graphite anode in the BMPyrr BTA during 4 hours with 0.1 M IL/acetonitrile at 7 V and 293 K. (b) Passing current profile versus time throughout the experiment. In the previous studies, there has been an overemphasis on the critical role of water in water-based IL mixtures.45,48 Thus, it is necessary to make a few comments based on the experiments conducted here. First of all, no notable exfoliation was observed with equivalent IL/water ratios compared to that of the newly introduced IL/acetonitrile systems (~1:50 volume ratios). Thus, in order to induce any electro-exfoliation in the water-containing systems much higher IL content is necessary (e.g., 1:0.1-1 IL/water vol. ratios45,48), which substantially increases the cost of the entire process. Furthermore, high water contents significantly narrow the electrochemical stability window, thus reducing the exfoliation power of the anionic intercalation compounds. On the other hand, the water presence in RTILs is inevitable as most of the ionic liquids are hygroscopic.210-211 However, such water traces are rapidly removed by the electrode reactions during electrolysis in the IL/acetonitrile electrolyte, and will not have any long-lasting impact. Therefore, water presence was found unnecessary for the electrochemical graphite exfoliation. Instead, it was shown that other cosolvents such as acetonitrile that preserve the IL’s potential   44  window serve as much more advantageous alternatives by lowering the IL content with high rates of exfoliation. For instance, the ionic conductivity and diffusivity of 0.1 M EMIM BF4 in ACN is approximately 50, and 150 times higher than the pure EMIM BF4.212 This is further manifested by much higher exfoliation yields obtained in ACN solutions (~5-6 times more) than those in pure IL medium. Therefore, the majority of the experiments in this thesis were conducted using 0.1 M IL solutions in ACN. 3.3.2 Product characterization After four hours of electrosynthesis, the exfoliated precipitates were collected, and thoroughly washed using deionized water and ethanol. The solid products were then separated by filtration and ultracentrifugation followed by drying in a vacuum oven for 24 hours at 358 K. The dried products were subjected to one hour sonication in NMP. Figure 3-9 shows the corresponding transmission electron microscopy (TEM) and field-emission scanning electron microscopy (FESEM) images. With extensive sampling, a variety of structures were captured from ultrathin graphene flakes (Figure 3-9a-d) to semi-transparent carbonaceous particles (Figure 3-9e), and rolled sheets (Figure 3-9f). After taking 250 TEM and SEM images of more than 15 samples, it was inferred that the majority of the produced materials are of the ultrathin irregularly shaped sheets rather than the other described morphologies. Based on the statistical analysis of the TEM images, the average length of the flakes was in the order of 500 nm.  In terms of thickness, according to the critical letter published by Meyer et al. in Nature on the structure of suspended graphene sheets,213 the folding fashion of the edges is an important indication for the number of layers. Therefore, from the folded edges shown in Figure 3-9b-c, formation of monolayers and bilayers is inferable. Sample AFM analysis of the dried products in Figure 3-10 also suggests ultralow thicknesses corresponding to less than 5 layers. However, the   45  folded/crumpled nature of the flakes as well as their relatively small lateral dimensions (< 1 μm) undermine the reliability of this technique and the subsequent interpretations. Appearance of numerous bumps on the surface observed in the 3D image (Figure 3-10, right) is an indication of non-flat surface, where the AFM tip overly bounces along the surface creating such artifacts. Additionally, the conventional drying of the graphene suspensions also causes misleading impressions originated from the layered depositions. Even in case of flatly produced graphene flakes, significant anomalies are reported in AFM-based thickness measurements, that could cause 1 nm order errors which is quite significant in sub-nano levels.175 Thus, the combination of HRTEM images together with Raman analysis (explained later) are better representatives of the products’ characteristics in this case.   46   Figure 3-9: TEM and SEM images showing the variety of products generated by electrochemical exfoliation of iso-molded graphite in EMIM BF4/acetonitrile electrolyte at 7 V and 293 K: (a-d) crumpled/folded sheets, (e) semi-transparent carbonaceous particles, and (f) rolled sheets.   47   Figure 3-10: AFM thickness profile analysis of products generated by electrochemical exfoliation of iso-molded graphite in EMIM BF4/acetonitrile electrolyte at 7 V and 293 K. Nonetheless, it is oversimplification to assume that the rigorous electro-intercalation of energetic IL anions leads to planar morphologies only. Isolation of the particles and rolled sheets, even in trace amounts, which was successfully achieved here certainly sheds some light on the exfoliation mechanism. Presumably, the initial anionic intercalation/expansion poses a strong mechanical stress on the vertices and edges of the graphitic precursor which ultimately results in breakage of the vertices (i.e., particle formation), and bending the edges (i.e., rolled sheet formation). It should be noted that such structures were separated in trace amounts (less than 10% in imaging counts, and quite insignificant in terms of weight) from filtration and centrifuge disposals, which might be typically overlooked by the researchers. Thus, exploring them by itself is a significant step towards better understanding of the whole process. Importantly, no major deviation from the planar topographies to other described by-products was noticed when different IL structures were employed for the synthesis. Moreover, the Raman spectra presented in Figure 3-11, obtained from EMIM BF4/ACN electrosynthesis, displays the well-known graphitic G peak at 1581 cm−1 (from bond stretching of sp2 pairs in rings and chains), and a D band at 1335 cm−1 (corresponding to the breathing modes of sp2 atoms in rings).25,214 As mentioned in the previous chapter, according to Ferrari et al.,215 the   48  evolution of the 2D peak (second order of D peak) is a clear indication of structural transformation from graphite to graphene heterostructures after electro-exfoliation. The sonication impact is also noticeable by increasing the G/D peak ratios, which represent lower disorders therefore better graphene quality.216 In fact, there is a certain chance for the incomplete electro-exfoliation or aftermath agglomeration of the products which induces the disorder effect. NMP is consistently featured among the most promising solvents for the liquid-assisted graphite exfoliation,26,217 and its effectiveness was evident in improving the overall quality of the products. Brief sonication can also help to more rigorously remove any physically adsorbed IL remnant/contamination on the surface of graphene which can be another contribution to the quality enhancement. This implies the importance of implementing a systematic approach for isolating the effect of all the processing and treatments performed on the GNs in the course of synthesis. Similar trends were observed over the exfoliation products synthesized with the other investigated IL structures (Figure 3-2).  Figure 3-11: Changes in the Raman spectra from graphite rod to the final sonicated product obtained by electro-exfoliation in EMIM BF4/acetonitrile at 7 V and 293 K after 4 hours.   49  Lastly, X-ray photoelectron spectroscopy (XPS) analysis revealed nitrogen incorporation into the exfoliation products via NHCs’ parasitic adsorption on the anode during electro-exfoliation in EMIM BF4 electrolytes (Figure 3-12). The presence of the imidazolium moiety as well as the BF4− counter-ion is inferable considering the XPS survey scan presented in Figure 3-12. A well-defined peak at 400.2 eV can be assigned to the N 1s of the imidazolium ring (Figure 3-12, inset). 45 Additionally, B 1s peak at 193.2 eV and F 1s at 686.6 eV represent the BF4− anion remnants on the exfoliated flakes (< 2%). Graphene functionalization by imidazolium heterocycles has been perceived as the main promoter for the rapid dispersion of the GNs in NMP (1 mg ml-1).48 Noticeably, the resulting suspensions remain stable after serious centrifugation and sedimentation over a few months. Consequently, the results presented here provided a set of key information about of the electrochemical configuration for graphene electrosynthesis. This is further explored, and improved in the upcoming chapter with more electrochemical characterizations and substantial yield increase via simultaneous exfoliation of both electrodes.  Figure 3-12: XPS data for the exfoliation products obtained from EMIM BF4/acetonitrile electrolyte at 7 V and 293 K after 4 hours: survey scan, and N 1s narrow scan.   50  3.4 Summary and conclusions The room temperature electrochemical exfoliation of two structurally different graphite anodes, namely HOPG and iso-molded graphite rod, was studied in this chapter. A novel electrolyte composed of IL and acetonitrile (~1:50 vol. IL/solvent ratios) was mainly utilized with four different IL structures. Previous research was focused only on IL/water mixtures with much higher IL content (e.g., up to 1:0.1 IL/water vol. ratios). The approach here provided three main advantages of cost efficiency due to low IL content, extended electrochemical stability in the non-aqueous electrolytes, and higher exfoliation yields caused by the effective anionic intercalation within the graphitic layers. Thus, up to 86% exfoliation yield was achieved in 4 hours using BMPyrr BTA/acetonitrile (~1:50 vol. ratio). The anodic intercalation of the large oxygenated BTA anions was observed to be promising in surpassing the strong van der Waals cohesion among the graphitic layers. The major products of the graphite exfoliation, for all the four tested ILs, were crumpled and folded graphene sheets. In addition, carbonaceous particles and rolled sheets were isolated, but in much smaller quantities. Furthermore, the complementary role of sonication after electrosynthesis was demonstrated by Raman analysis suggesting improvements in the overall quality of the graphene-based products.  Previous research had also placed a great deal of emphasis on the electrolyte coloration during electrosynthesis as an indicator of various graphite exfoliation stages and nanostructure formations. However, this was rejected via control experiments (i.e., replacing graphite anode by a platinum wire). Instead, the reactions of the IL moieties with the electrodes along with their unusual fluorescence behavior was found to be responsible. The cathodic reduction of the cationic moieties such as imidazolium rings generates carbene species that are reactive with the graphite anode, and interfere with the anionic intercalation. On the other hand, it provides the opportunity   51  for in-situ functionalization of the graphene sheets via redox species produced during electrosynthesis process. The resulting functionalized GNs can be used for a variety of applications as a competitive material for energy storage/generation, environmental remediation, biomedical areas, and so forth. Finally, the results presented here paved the way for more in-depth understanding of the underlying exfoliation phenomena. Next chapter deals with this concept more extensively with further advances made to improve both quantity and quality of the exfoliation products.     52  Chapter  4: Simultaneous exfoliation studies2 4.1 Introduction All the results discussed in Chapter 3 along with other literature were focused on rendering structural changes in the graphite either at the cathode or the anode; while the counter-electrode was undergoing gas evolution and solution decomposition, leading to significant energy and chemical losses. Having graphite exfoliation taking place on both electrodes would considerably improve the energy efficiency of the electrochemical process for graphene production, and its feasibility for scale-up. There has been very little progress so far on this part including the work presented by Mao et al. where they attempted to simultaneously exfoliate graphitic flakes attached to the stainless steel mesh electrodes in protic ionic liquids.218 However, their long run exfoliation process (7 h) coupled with high shear liquid phase exfoliation (2.5 h) accomplished as little as 30 mg of exfoliation at best on both electrodes combined;218 while the actual impact of the electro-exfoliation was not isolated from the subsequent solution-based exfoliation of the flakes in their work. Additionally, using protic electrolytes, Mao et al. configuration was still suffering from electrolyte decomposition and gas evolution at the electrodes.218 As discussed in the previous chapter, when only the anode is involved in graphite exfoliation, the ILs’ reactivity at the counter-electrode could generate carbene species causing in-situ functionalization of GNs during electro-exfoliation; which depending on the application, might or might not be desired.97-98 It is also noted that ‘ionic liquid’ is a generic term for a wide range of ionic materials consisting of bulky and asymmetric ions which noticeably lowers their melting points.219 Thus, any generalization of their complex behavior requires precautions                                                  2 A version of this chapter has been published in Carbon:  A Taheri Najafabadi, E Gyenge, Carbon, 84 (2015), 449-459.    53  especially in this area where liquid-solid interactions at electrified interfaces play a critical role. Considering the tunability of ILs’ physicochemical properties, and continuing the previous line of research, here the simultaneous anodic and cathodic electro-exfoliation of graphite in aprotic electrolytes is demonstrated.  It’s worth noting that the main objective of this chapter was to increase the quantity and quality of the exfoliation products studied in Chapter 3. In the run up to the simultaneous exfoliation idea, several other attempts were made which did not seem to succeed. First, the in-situ sonication of the electrochemical cell was pursued, and the exfoliation yield dramatically dropped instead. Also, with the graphite counter electrode, continuous altering of the voltage rendered no visible exfoliation on neither of the electrodes. The initial impression was that any change in the electro-intercalation process which disturbs the ionic diffusion would significantly decrease the exfoliation yield. In the case of in-situ sonication, the ultrasonic waves create disturbance while altering the voltage retracts the ions before they can effectively diffuse within the graphitic galleries. Therefore, the extent of exfoliation was lowered by at least 100 times for the in-situ sonication, while no exfoliation was recorded when the voltage was switched from positive to negative at the sweep rates faster than 200 mV s-1. Considering the above, a third alternative was followed which was combining the most effective anionic and cationic exfoliation methods using a divided cell. The anodic compartment was established quite satisfactorily using ionic liquids in Chapter 3. However, on the cationic side, a few options remained to be investigated. First, the work conducted by Wang et al. which used lithium co-intercalation with propylene carbonate (PC) followed by 8 h sonication.47 Their process resembled the destructive effects of PC as a molecular solvent on the graphitic electrodes in lithium-ion batteries at relatively high applied potentials (−15±5 V).94-95 However, considering its   54  low yield, and lithium passivation phenomenon on the cathode,220 which prevents further intercalation, it did not emerge promising in the initial trials. Inevitably, organic cations such as those in tetraalkylammonium salts remained the only option with a few studies discussing the possibility of graphite exfoliation upon intercalation.96 In addition, the possibility of integrating those cations in ionic liquid electrolytes was a major advantage. Therefore, this concept became the basis of the studies presented here, which led to the successful development of a high-throughput graphene production method via simultaneous exfoliation of both involved electrodes. There are also more in-depth studies on the cyclic voltammetry behavior of the ions along with their effect on the quality of the final products. 4.2 Materials and methods Figure 4-1 shows the conceptual experimental setup for simultaneous graphite electro-exfoliation in the presence of IL electrolytes along with the investigated intercalating ions. The numbers below each ion indicate their standard potential stability versus the normal hydrogen electrode (NHE). For anodic exfoliation, tetrafluoroborate (denoted as BF4) and bis(trifluoromethylsulfonyl)imide (denoted as BTA), emerged as promising intercalating ions in the previous chapter’s single-electrode studies.122 Hence, they have been retained in the present work as well. The cationic intercalant selection was guided originally by the work of Simonet et al. describing graphitic electrode expansion driven by the intercalation of symmetric tetraalkylammonium cations.221-222 Here, after successful preliminary experiments, this idea was taken one step further by examining asymmetric quaternary ammonium cations for graphite electro-exfoliation as shown in Figure 3-3. They were selected such that to examine and verify the hypotheses regarding the roles played by various anions and cations during simultaneous graphene electrosynthesis. Acetonitrile was used again as cosolvent, and all of the main electrolytes were of 0.1 M concentration.   55  For the simultaneous exfoliation experiments, a high purity (> 99.95 %) iso-molded graphite rod (GraphiteStore – 6.35 mm diameter with 2.5 cm effective length exposed to the electrolyte) was used as both anode and cathode. In the case of controlled single-electrode exfoliations a platinum wire counter-electrode was used (1.6 mm diameter and 2.5 cm effective length). Two different electrochemical cell configurations were tested for the simultaneous exfoliation. First, depicted in Figure 4-1, a divided cell equipped with a fine porous ceramic frit (Pine Research Instrumentation – RRPG060) and an inter-electrode distance of 5 cm. The anode and cathode were separated, hence, the products from each electrode were individually collected and investigated. The second setup was an undivided cell with an inter-electrode distance of 1 cm between the two graphitic electrodes, which further decreased during electro-expansion. This simple configuration is advantageous from energy consumption, and electrochemical data analysis viewpoint, but does not allow separate characterization of the anodic and cathodic exfoliation products. The aforementioned cell configurations were employed in constant cell potential (i.e., potentiostatic) exfoliation experiments (B&K Precision – 9110 DC power supply) using the following applied voltage ranges: 10-20 V for the divided H-cell (25 ml on each side), and 0-12 V for the one-pot undivided cell (15 ml of total electrolyte). The current versus voltage curve (i.e., cell polarization curve) was recorded and related to the exfoliation rates. The macroscopic extent of graphite exfoliation was evaluated by measuring the electrode weight changes with the procedure explained in Section 3.2. This variable was used as a primary means to quantify the exfoliation yield in the presence of different electrolytes.    56  After 2 hours of potentiostatic experiments, the exfoliated solid products generated at the anode and cathode of the divided H-cell were collected, thoroughly washed with copious amounts of water/acetone, and dispersed in N-methylpyrolidone (NMP, 99.5 % from Sigma-Aldrich) for 1 hour using a tabletop ultrasonic cleaner (VWR Scientific – B3500). The suspension was filtered using an ultrafine ceramic filter after being centrifuged at 4000 rpm for 30 min. The porous ceramic filter (~100 nm pore size) containing the GNs was finally dried in vacuum oven at 383 K for 12 h, and the products were then taken for various characterization analyses. All the chemicals were of the analytical grade, and double-distilled water was used during all of the preparation steps. Further details regarding the electrosynthesis protocols are presented in the form of figures and flowcharts in Appendix A.  Furthermore, cyclic voltammetry experiments were performed mainly at a potential sweep rate of 100 mV s-1 with iso-molded graphite as the working and counter electrode (6.35 mm diameter and 2.5 cm effective length exposed to the electrolyte). Also, similar experiments were  Figure 4-1: Experimental setup used for simultaneous electro-exfoliation of graphitic anodes and cathodes in acetonitrile containing ionic liquids (0.1 M). The highest standard electrochemical stability potential for each ion on platinum electrode is indicated at 293 K versus normal hydrogen electrode (NHE).    57  carried out by replacing graphitic electrodes with platinum wires (1.6 mm diameter and 2.5 cm effective length exposed to the electrolyte). Ag/Ag+ reference electrode was used to measure the working electrode potential. The Ag wire was submerged in the reference electrode solution composed of 0.1 M AgNO3 and 0.1 M tetrabutylammonium perchlorate (TBAP) in acetonitrile. The reference solution was separated from the bulk electrolyte by a glass frit, and the reference electrode was calibrated against the normal hydrogen electrode (NHE) showing a stable +0.542 V potential difference. 4.3 Results and discussion  4.3.1 Electrosynthesis experiments Figure 4-2 shows the successful implementation of the proposed concept for simultaneous electro-exfoliation of the iso-molded graphitic electrodes in 0.1 M N1114 BF4/ACN solutions. Upon applying 15 V to the divided cell (Figure 4-2a, left), it took only a few seconds until exfoliation at both electrodes became noticeable, and the experiment was continued over a 2-hour period. The exfoliation products from both anode and cathode were collected and analyzed separately as explained in the next section. Additionally, a closer look into the apparent exfoliation patterns on the electrodes revealed that the anode was peeled off more irregularly than the cathode; which showed a more uniform exfoliation pattern (Figure 4-2a, right). In the undivided cell, the first signs of exfoliation were observed at potentials as low as 5 V, and the extent of exfoliation increased with the applied cell voltage as indicated by the increased current on the two graphite rod electrodes (Figure 4-2b). However, there is a trade-off between high current and quality of exfoliation products. Higher cell voltages (> 10 V) were found to cause more significant IL reactivity, and aggressive destruction of the graphitic rod therefore releasing unaffected chunks of material to the solution; whereas voltages lower than 5 V did not lead to   58  significant exfoliation yields during 2 hours of experiment. Thus, in the case of undivided cell configuration used here, an applied voltage of 7 V provided an optimal compromise between exfoliation yield and product quality throughout the experiments (this number was about 15 V for the divided cell). Moreover, based on the inset in Figure 4-2b, the exfoliation percentage at 7 V, calculated from the weight changes of the electrode, showed a remarkable synergistic effect. The exfoliation percentages for each of the electrodes were the highest when both were iso-molded graphite rod (i.e., simultaneous anodic and cathodic exfoliation). Thus, 634% exfoliation was achieved on the cathode, and 453% on the anode ( shows 95% confidence interval for 3 experiments). However, when only either cathodic or anodic exfoliation was performed, with a Pt counter-electrode, the yields were 24% and 15%, respectively. Therefore, due to the synergistic effect between the anodic and cathodic exfoliation processes, the exfoliation efficiency increased by about 3 times for each of the iso-molded graphite electrodes (Figure 4-2b).   59   Figure 4-2: (a) Simultaneous electrochemical exfoliation of iso-molded graphite rod in 0.1 M N1114 BF4 /ACN solution using the divided H-cell configuration at 15 V and 293 K in the course of 1 hour. The exfoliation patterns of the anode and cathode are also depicted on this right side. (b) Polarization curves in the undivided cell for three types of exfoliation experiments: simultaneous, anodic and cathodic. In the case of single-electrode exfoliation (anodic or cathodic), the counter-electrode was Pt. The numbers inside the box indicate the exfoliation percentages at 7 V after 2 hours for the three types of experiments. In the previous chapter, the impact of counter-electrode reactions on the exfoliation process was explored, and here is no exception.122 Employing graphite for both anode and cathode significantly enhances the charge accommodation capacity within the graphitic layers as reflected by the increase in the cell current. At 7 V, the cell current in undivided configuration was equal to   60  75 mA, which is approximately the sum of the currents in the single graphite electrode experiments (Figure 4-2b).  In order to further investigate the ionic interactions with the electrodes, cyclic voltammetry (CV) was conducted using the ionic-liquid-based aprotic electrolytes. Figure 4-3a shows the repeated CV scans recorded at 100 mV s-1 using graphite working and counter electrodes in 0.1 M N1114 BTA/ACN. For comparison, the voltammograms obtained with Pt electrodes are also presented, showing the large electrochemical stability window of the electrolyte. The onset potential for electrode reactions on Pt occurs at about –3 and +3 V (vs. Ag/Ag+), respectively.  On the iso-molded graphite electrode both anionic and cationic deintercalation waves are apparent for all the investigated ionic liquids, but the deintercalation potential ranges are different (Figure 4-3b, c). In the case of N1114 BTA, the charge transfer associated with N1114 intercalation on the cathodic scan commences at -2 V and the corresponding deintercalation wave occurs around -0.4 V on the reverse scan. The BTA anion intercalation-related charge transfer starts at about 1.5 V followed by a deintercalation peak at ~0.2 V decreasing to -0.4 V with increasing the scan numbers (Figure 4-3b, c). In the case of BF4 anion on the other hand, deintercalation occurs at more positive potentials with a deintercalation peak potential between about 1.1 to 0.7 V, as a function of the number of scans performed (Figure 4-3b). There are also differences in the deintercalation peak potentials of the cations N1114 and BMPyrr at -0.4 V and -1 V, respectively (Figure 3c).  More importantly, as the number of potential scans increased between –4 V and +4 V vs. Ag/Ag+, the anionic deintercalation peaks shifted much more significantly than those for cationic deintercalation (Figure 4-3, b compared to c). Thus, the electric potential required to deintercalate the anionic species from the graphite anode becomes more negative with cycling (e.g., for BTA   61  from +0.2 V for the 1st cycle to -0.4 V at the 5th cycle) indicating a strong interaction between the anion and the graphitic layers compared to the cation (e.g., in the case of N1114, the potential shifted only from -0.4 V to -0.3 V during five cycles). This could also explain the differences observed for macroscopic exfoliation patterns on the anode compared to that for the cathode, displayed in Figure 4-2a.  Figure 4-3: (a) Cyclic voltammetry of the ionic liquid solution (0.1 M N1114 BTA in ACN) on iso-molded graphitic electrodes compared to those on Pt electrodes. (b) Anionic deintercalation of BTA and BF4, compared to (c) cationic deintercalation of N1114 and BMPyrr. Experiments were conducted at the potential sweep rate of 100 mV s-1. Regarding the IL selection, it is noted that most of the previous research on the IL-assisted graphite electro-exfoliation had examined only imidazolium-based cations in aqueous solutions without observing any structural changes at the cathode.45,48 As discussed in Chapter 3, one of the key properties of the imidazolium-based RTILs, the most widely studied class of ionic liquids,200 is their Brønsted acidity related to the C-2 hydrogen of the 1,3-dialkylimidazolium cation.201 The electrochemical reduction of the C-2 acidic proton initiates intense hydrogen evolution at the cathode leading to the formation of the electro-generated N-heterocyclic carbenes (NHCs).202,204 NHCs are highly reactive, as base/nucleophile, and can also react with the graphite anode.122 More   62  controlled experiments were performed using imidazolium cations, resulting in no significant exfoliation on the cathode. Thus, the imidazolium-based ionic liquids were no longer pursued for further studies in this area due to electrolyte reactivity and instability of the electric current under constant voltage conditions. 4.3.2 Product characterization After 2 hours of potentiostatic experiments, the exfoliated solid products obtained at both electrodes of the divided cell were separately processed, and analyzed. Figure 4-4 shows the transmission electron microscopy (TEM) images representative for the majority products generated by simultaneous exfoliation in N1114 BF4 electrolyte at the anode and cathode, respectively. Over five hundred images were taken from more than fifteen trials in order to assure the repeatability and reproducibility of the findings. The anodic products mostly resembled the folding patterns of suspended graphene flakes suggested by Meyer et al.,213 where physical differences between layers overlaying on each other are explained via intensified visual effects at the edges (Figure 4-4a, b). On the other hand, Moire patterns were observed for the cathodic exfoliates created via stacking ultrathin and almost intact honeycomb GNs with different rotation angles similar to those studied by Miller et al. (Figure 4-4c, d).223 More irregularities were found generally including multiple structural curls and edge deformations on the anodic products while cathodic exfoliates showed more wrinkles, and appeared to be relatively larger (~800 nm compared to ~500 nm on the anodic side). Nonetheless, it is evident from TEM imaging (Figure 4-4) that both electrodes here were effectively employed to enhance the overall yield of GN production during electro-exfoliation. Moreover, no major deviation from the planar flakes to other morphologies was captured when various IL structures presented in Figure 4-1 were employed for the electrochemical exfoliation.    63  Next, Raman spectroscopy was conducted to further investigate the quality of the graphene products formed by simultaneous electrochemical exfoliation in the aprotic IL electrolytes. Figure 4-5 shows the full Raman spectra acquired from the iso-molded pristine graphite compared to the exfoliated products on each electrode. It can be seen that the 2D band  Figure 4-4: Low and high resolution TEM images of the GNs generated by simultaneous anodic and cathodic electrochemical exfoliation of iso-molded graphite electrodes in N1114 BF4/ACN electrolyte at 15 V and 293 K using H-cell configuration: (a, b) anodic exfoliates resembling the folding fashion of ultrathin graphene layers, and (c, d) cathodic exfoliates exhibiting Moire patterns created via stacking ultrathin and almost intact honeycomb GNs with different rotation angles.   64  has shifted from ca. 2685 cm-1 for the pristine graphite to ca. 2650 cm-1 with a more symmetrical shape for both anodic and cathodic products. Also a slightly negative shift in the G peak (approximately 20 cm-1) was detected over the exfoliation products compared to that of graphite. Both of these features of the Raman spectra suggest the presence of ultrathin graphene flakes (<5 layers) amongst the anodic and cathodic exfoliates.215  Furthermore, comparing the G/D peak intensity ratios, which is a well-known indicator of the disorder degree in graphene,224 revealed significant differences between the products on each electrode. The anodically produced GNs exhibited lower G/D peak ratios pointing to higher disorder levels. This can be either caused by the edge defects and physical distortions present on the flakes as shown also by the TEM images (Figure 4-4), or a loss of sp2 bonding network due to any possible functionalization of GNs. It is thought that both of the aforementioned routes could play a role in the lower G/D peak ratios observed over anodic products. Among the anodic exfoliates, in addition to the majority product of GNs shown in Figure 4-4, a variety of other morphologies were detected again as by-products in less than 10% of the total imaging counts. As shown in Figure 4-6, these include nanoparticles, rolled sheets and larger hollow tubes. The presence of larger tubes with a variety of cross section shapes and areas was further confirmed via field emission scanning electron microscopy (FESEM, Figure 4-6c). Importantly, such structures were separated in trace amounts (less than 10% in imaging counts, and quite insignificant in terms of weight) from filtration and centrifuge disposals. This might be typically overlooked by the researchers, but can suggest some explanations for the actual electro-exfoliation mechanism. Otherwise, there was no major challenge in collecting the majority product (i.e., graphene microsheets) here. On the cathode side, however, the exfoliated products were virtually totally composed of GNs as exhibited by Figure 4-4.   65  As mentioned above, the isolation of trace amounts of nanoparticles and rolled sheets among the anodic products could shed more light on the anodic exfoliation mechanism. It is perceived that intercalation of the structurally rigid BF4- anions into the iso-molded graphite galleries poses strong mechanical stresses on the vertices and edges, which ultimately results in breakage of the vertices (i.e., particle formation, Figure 4-6a), and complete bending of the edges (i.e., rolled sheet formation, Figure 4-6b, c). The detection of the graphene flakes like the one shown in Figure 4-6d with the left edge bended half-way through besides nanotubes with similar bending characteristics in the same TEM grid provides further evidence for the proposed anodic exfoliation mechanism. The intense interaction of the anions with the iso-molded graphite rod was also inferred earlier where the cyclic voltammetry analysis and exfoliation patterns were discussed. Hence, the imaging evidence presented in Figure 4-6 further substantiates the structural disorders within the anodic products offered initially by the lower G/D peak ratios in Raman analysis (Figure 4-5).  Figure 4-5: Changes in the Raman spectra from graphite to the final products obtained by simultaneous electro-exfoliation of iso-molded graphite electrodes in N1114 BF4/ACN at 15 V and 293 K using the divided H-cell configuration.   66   Figure 4-6: By-products detected amongst the anodic exfoliates generated by simultaneous anodic and cathodic exfoliation of iso-molded graphite with 0.1 M N1114 BF4 in ACN as electrolyte at 15 V and 293 K using the divided H-cell configuration: (a) nanoparticle clusters in TEM, (b) rolled sheet clusters in TEM, (c) larger tube agglomerates in FESEM, and (d) graphene flake bended half-way through along with a nanotube sharing similar characteristics observed in the same TEM grid. The absence of the above-described side products amongst the cathodic exfoliates begs a different explanation regarding the cationic exfoliation mechanism. This could be best described by the flexibility of the alkyl/aryl groups in the organic quaternary ammonium cations to flatten (to some degree) between the graphene sheets as put forward by Sirisaksoontorn et al. recently.225   67  By conducting extensive X-ray powder diffraction analysis (XRD), they concluded that the intercalation of tetrabutylammonium (TBA) cations into graphite yields a single-phase first-stage product, C44TBA, with a gallery expansion of 0.47 nm. Since the freestanding TBA cations have much larger ionic size (0.89 nm), the smaller reported inter-layer distances necessitate an anisotropic “flattened” cation conformation.225 Such ionic conformational flexibilities possibly apply less deformation stress to the graphitic layers at the edges and vertices, generating a more uniform exfoliation pattern on the iso-molded graphite rod, as shown macroscopically also by Figure 4-2a. Similar exfoliation trends were observed for the other investigated ionic liquid electrolytes in this work (Figure 4-1). In addition to the morphological differences between the anodic and cathodic exfoliation products, some surface chemical differences were identified as well; which could contribute to the lower G/D peak ratios for the anodic products in Raman analysis. Comparing the X-ray photoelectron spectroscopy (XPS) results from anodic and cathodic exfoliates (Figure 4-7) suggests notable differences in surface composition between the two sides. Particularly for N1114 BF4 electro-exfoliation, the survey scan revealed that the anodic products contain relatively higher oxygen contents, and also remnants of the intercalating ions represented by F 1s at 686.6 eV. A closer look at the deconvolution results of C1s peak (Figure 4-7b, c) with Lorentzian functions corresponding to five different carbon states further confirms a more oxidized state of the exfoliated products at the anode. Particularly, the two carbon–oxygen peaks (C–O and C=O) located at 286.5 eV and 287.9 eV, respectively,226 had a total contribution of 32% to the C1s peak at the anode while this number was significantly lower for the cathodic products (18%). It should be noted that there is error associated with calculations related to carbon-oxygen bonds due to the adsorption of airborne hydrocarbons, and the analysis is mostly valid on the comparison basis.   68  The extent of GNs functionalization can be either advantageous or undesired depending on the intended application. For instance, use of GNs as the support for metallic catalysts requires some functional groups as anchoring sites to enhance metal-support interactions.227 For this particular purpose, the anodic exfoliates would be preferable. They also form higher concentration dispersions (up to 1 mg ml-1) as opposed to the less functionalized cathodic products (0.3-0.4 mg  Figure 4-7: XPS data for the anodic and cathodic products obtained from electro-exfoliation experiments in N1114 BF4/ACN electrolyte at 15 V and 293 K using H-cell configuration: (a) survey scan, and C 1s narrow scan of the (b) anodic, and (c) cathodic exfoliates.   69  ml-1). Conversely, in electronics-related areas where virtually defect free structures are of outmost importance,82 the cathodic products are more suitable. Thus, the simultaneous anodic and cathodic exfoliation can produce, at the same time, different application-specific GNs enhancing thereby the cost effectiveness of the method. This is further highlighted in the upcoming chapters which focus on the application of the graphene products in various types of fuel cells.   Lastly, aside from the general trends observed for various intercalating ions here, there are undoubtedly some differences amongst them as well. For instance, although BTA anions exfoliate the anode in greater extent compared to BF4 counterparts, two differences between their products were observed. First, the functionalization in the case of larger and more oxygenated BTA anions was more pronounced (Figure 4-8a), while the flakes appeared to be relatively thicker (Figure 4-8b). Importantly, there is a trade-off for the ions’ size to obtain decent quality flakes. While smaller anions like Cl- are incapable of exfoliation, having the BTA ions (which are much larger than BF4) would not be necessarily a good choice either. This is mainly due to the fact that they would not be able to form a first-stage intercalation thus unable to break the structure into thinner entities. The same argument goes for the cationic side as well where N1114 is preferred over BMPyrr. Consequently, the materials used throughout the upcoming chapters are mainly synthesized using the preferred intercalants.   70   Figure 4-8: Sample (a) XPS, and (b) TEM analysis of the anodic products obtained from electro-exfoliation experiments in N1114 B TA/ACN electrolyte at 15 V and 293 K using H-cell configuration. 4.4 Summary and conclusions  A novel electrochemical method was presented to effectively exfoliate graphitic electrodes to ultrathin GNs on both anode and cathode in an aprotic IL/ACN electrolyte. Previous research was mostly capable of triggering exfoliation only at one electrode while at the counter-electrode undesired secondary reactions took place such as electrolyte decomposition and gas evolution, which translated into lower energy efficiencies. Taking advantage of ILs’ broad physicochemical properties, the most effective cationic and anionic intercalating agents were combined, and simultaneous anodic and cathodic exfoliation of the iso-molded graphite electrodes was carried out successfully. It was demonstrated that having exfoliation on both sides has a synergistic effect on the exfoliation yield, with up to 3-fold increase compared to the best single-electrode studies on each side (~6-fold improvement in total). GN products with higher quality and relatively larger   71  lateral sizes (~800 nm compared to ~500 nm of the anodic exfoliates) were observed in the cathodic compartment.  Considering the lack of a universal term for the exfoliation rate in the electrochemical systems, the following unit of ‘mg cm-2 h-1’ could be suggested as a basis for more reasonable comparisons with other experimental studies. It is determined by the amount of exfoliated products per initial area of the graphitic electrode exposed to the electrolyte over time (Equation 4-1). With that in mind, the combined values of up to 156 mg cm-2 h-1 were obtained here (65 mg cm-2 h-1 on the anode, and 91 mg cm-2 h-1 on the cathode); while this number was 10 mg cm-2 h-1 at best for the closest rival (i.e., Mao et al. presented in the introduction). This corresponds to the energy consumption of ~2.5 MJ per kg of the exfoliates (based on Equation 4-2) which is far more efficient than Mao et al.’s power-hungry experiments (~200 MJ per kg of the product). Therefore, energy efficiencies as high as 20% were achieved here using Equation 4-3; where the theoretical exfoliation energy is the strong cohesive energy of the π-stacked layers in graphite elaborated in Chapter 1 (~0.5 MJ per kg of carbon atoms). Importantly, this is the first attempt in the field to characterize the process parameters in the form of aforementioned equations. Although there might be errors associated with their calculations and the underlying assumptions, it’s still useful to establish some benchmarks in this research area. 𝐸𝑥𝑓𝑜𝑙𝑖𝑎𝑡𝑖𝑜𝑛 𝑅𝑎𝑡𝑒 =𝐴𝑚𝑜𝑢𝑛𝑡 𝑜𝑓 𝐸𝑥𝑓𝑜𝑙𝑖𝑎𝑡𝑖𝑜𝑛 (𝑚𝑔)𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝐸𝑙𝑒𝑐𝑡𝑟𝑜𝑑𝑒 𝐴𝑟𝑒𝑎 (𝑐𝑚2) ∗ 𝑇𝑖𝑚𝑒 (ℎ) Equation 4-1 𝐸𝑛𝑒𝑟𝑔𝑦 𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 =𝐼 ∗ 𝑉 ∗ 𝑡𝐴𝑚𝑜𝑢𝑛𝑡 𝑜𝑓 𝐸𝑥𝑓𝑜𝑙𝑖𝑎𝑡𝑖𝑜𝑛 Equation 4-2   72  𝐸𝑛𝑒𝑟𝑔𝑦 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 (%) =𝑇ℎ𝑒𝑜𝑟𝑒𝑡𝑖𝑐𝑎𝑙 𝐸𝑥𝑓𝑜𝑙𝑖𝑎𝑡𝑖𝑜𝑛 𝐸𝑛𝑒𝑟𝑔𝑦 𝐴𝑐𝑡𝑢𝑎𝑙 𝐸𝑛𝑒𝑟𝑔𝑦 𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛∗ 100 Equation 4-3  All the physicochemical and structural characterization studies pointed towards major differences in the exfoliation mechanism on each side. Raman analysis suggested higher disorders for the anodically obtained GNs as indicated by lower G/D peak intensity ratios. Two major physical and chemical causes were suggested for this observation: from physical standpoint, lower lateral sizes compared to that of Raman laser spot size and higher deformations observed at the edges of the anodic products intensify the D peak. Isolation of rolled sheets along with nanoparticles in small amounts amongst anodic exfoliates was linked to the exfoliation mechanism at the anode. On the other hand, from chemical disorder point of view, surface composition analysis by XPS, both survey and narrow scans, showed higher degrees of oxidation for the anodic products along with more ionic remnants on the produced GNs. Moreover, the better quality of the cathodic products were connected to the ability of the alkyl/aryl groups in quaternary ammonium cations to flatten between the graphene sheets throughout the exfoliation process. This certainly poses less stress upon the graphite galleries’ entrance and probably explains the absence of other morphologies that were observed in the case of anodic products. In any case, the aforementioned differences between the cathodic and anodic products can broaden the application range of the electrochemically produced graphene sheets using the proposed simultaneous exfoliation process. Lastly, the evolved format of electrochemical exfoliation using both electrodes, introduced here, has some major advantages over the previous electrochemical methods. Having done preliminary experiments using aqueous electrolytes (from sulfuric acid to potassium hydroxide), the power consumption is reduced at least by one order of magnitude while the process yield is   73  improved by two orders of magnitude. This is mainly due to the elimination of parasitic reactions in aqueous media which restrict the amount of applied voltage, and lower the exfoliation yield. Compared to the reduced graphene oxide pathway, the production time is significantly decreased (from days to hours) with much less defects, and better product quality especially when cathodically produced. Needless to mention that this technique is also vastly safer than working with highly concentrated oxidizers/reducers routinely used for rGO synthesis. As discussed in the upcoming chapters, the aforementioned advantages are reflected by superior performance of the exfoliated products in electrochemical power sources.    74  Chapter  5: Application I – Hydrogen fuel cells3 5.1 Introduction Fuel cells, as electrochemical devices that convert the chemical energy stored in fuels directly into electricity, have received great attention as clean power sources in recent decades.228 Proton exchange membrane fuel cells (PEMFCs), in particular, are viewed among the most viable alternatives to the internal combustion engines owing to their substantially reduced harmful emissions, high power density, low operating temperature, quick start-up, and low noise levels.229 In a hydrogen PEMFC, where a proton conductive membrane (e.g., Nafion) is used as an electrolyte, electricity is produced through a catalytic-assisted (e.g., platinum) electrochemical reaction between hydrogen as fuel, and oxygen as oxidant. The only resultant by-products are excess heat and water. The associated electrical conversion efficiency exceeds 60%, which can further be increased to ~80% with utilization of the cogenerated heat, while lowering pollution by over 90%.230 The corresponding half reactions on the anode and cathode along with the overall cell reaction are shown in Equation 5-1, Equation 5-2, and Equation 5-3, respectively. The cell voltages are reported at the standard condition of 298.15 K, 1 atm pressure for gaseous species, and 1 M concentration for both anodic and cathodic solutions. The half-cell voltage values are reported against the standard hydrogen electrode (SHE). Anode: 2𝐻2 → 4𝐻+ + 4𝑒− 𝐸𝑒𝑞0 = 0 𝑉 [𝑣𝑠. 𝑆𝐻𝐸] Equation 5-1 Cathode: 𝑂2 + 4𝐻+ + 4𝑒− → 2𝐻2𝑂 𝐸𝑒𝑞0 = 1.23 𝑉 [𝑣𝑠. 𝑆𝐻𝐸] Equation 5-2 Overall: 2𝐻2 + 𝑂2 → 2𝐻2𝑂 +𝐻𝑒𝑎𝑡 𝐸𝑒𝑞0 = 1.23 𝑉  Equation 5-3                                                  3 A version of this chapter has been published in ChemSusChem:  A Taheri Najafabadi, MJ Leeuwner, DP Wilkinson, E Gyenge, ChemSusChem 9 (2016), 1689-1697.   75  A PEM fuel cell is a sandwich of seven layers each with distinct roles. As shown in Figure 5-1, from left to right, it starts with the hydrogen (fuel) flow field plate (FFP), the anode gas diffusion layer (GDL), the anode catalyst layer, proton exchange membrane (PEM), the cathode catalyst layer, the cathode GDL, and finally the air (or oxygen) FFP. This is the most general form of a fuel cell design, and often referred to as membrane electrode assembly (MEA) when compressed together for final application. The gaseous reactants diffuse perpendicularly to the channel flow through the GDL, and reach the catalyst layer where the reactions occur. The protons generated at the anode migrate towards the cathode via membrane, and the respective electron flow moves from the cathodic side to the external load. Figure 5-1 represents the described electrochemical process along with highlighting the typical PEMFC components. Depending on further customizations, some of the aforementioned components are merged to improve the overall MEA performance. For instance, the catalyst coated membranes (CCMs) are more popular due to their enhanced catalytic activity compared to the freestanding catalyst layers. This also reduces the total layers in MEA to five with inclusion of the catalyst on the membrane.   Figure 5-1: Schematic of the typical PEMFC components including flow field plates, gas diffusion layers, catalyst layers, and membrane.   76  Despite PEMFCs’ favorable features, they have not found widespread commercial application yet, due to remaining cost and performance limitations, as well as infrastructure requirements. These include the use of expensive platinum catalysts and membranes, stringent fuel requirements (avoiding carbon monoxide in the hydrogen stream which poisons the catalyst) as well as water management issues.231 Extensive research has been performed to address the aforementioned problems, with significant breakthroughs also being made.232 Platinum group metal (PGM) catalyst loadings as low as 0.15 mg cm-2 have been tested successfully, approaching the target value of 0.125 mg cm-2 set by the US Department of Energy (DOE) for 2020. The corresponding cost for transportation applications has also reduced to $49 kW-1 with lifetimes reaching 50% of the 2020 target (5000 h).233 However, further progress is still required to improve the overall performance, and to ultimately increase the viability of these devices for large scale implementation. Amongst other things such as reduced platinum loading, the required advances relate to the expansion of the PEMFC operating range, and more specifically include the goal to achieve long-term operation at low relative humidities. This chapter focuses on the efforts towards more advanced GDLs using graphene-based composites to improve the overall fuel cell performance over a wide range of operating conditions. Some of the necessary theory and terminologies to understand the context of the upcoming experimental results are presented in the next section. The materials are separated from the literature review explained in Chapter 2 to further highlight the fuel cell fundamentals in the related place. It includes the underlying thermodynamics of fuel cells (temperature/pressure effect), and different sources of energy losses depending on the operating conditions.   77  5.2 Theory and terminology 5.2.1 Thermodynamics The total Gibbs free energy change for the overall cell reaction (Equation 5-3) can be simply found via Equation 5-4 where the Gibbs free energy of water formation (product) is subtracted from those of the reactants (oxygen and hydrogen) with their corresponding stoichiometric proportions. This gives the amount of -238 kJ mol-1 where the negative number points to the spontaneous nature of the reaction. ∆G = ∆G𝑓𝐻2𝑂 − ∆G𝑓𝐻2 −12∆G𝑓𝑂2 Equation 5-4 In electrochemical systems, the relation between the theoretical cell voltage and Gibbs free energy change can be described by the Equation 5-5 where n is the number of exchanged electrons per mole of products, and F is the Faraday’s constant (96,485 Coulombs electron-mol-1). Their product with the cell voltage equals the maximum possible change for the Gibbs free energy. ∆G = −𝑛𝐹𝐸𝑒𝑞 Equation 5-5 Thus, one could obtain the 1.23 V value reported for the hydrogen fuel cell at standard conditions (Equation 5-3) from the Gibbs free energy change of the overall reaction calculated in Equation 5-6. The n equals 2 based on the reaction stoichiometry here. 𝐸𝑒𝑞0 =−∆𝐺0𝑛𝐹=237,340 𝐽 𝑚𝑜𝑙−12 ∗ 96,485 𝐴 𝑠 𝑚𝑜𝑙−1= 1.23 𝑉  Equation 5-6 Gibbs free energy change itself is comprised of two terms namely enthalpy change (∆𝐻) and minus entropy change multiplied by temperature (−𝑇∆𝑆). Since both terms of enthalpy and entropy changes are negative here, an increase in cell temperature will cause a decline in the equilibrium potential. Further calculations show that moving from 25 °C to 100 °C corresponds to a 63 mV loss in the reversible cell potential (~5% decrease).   78  For pressure dependency of the Gibbs free energy thus the equilibrium potential, under isothermal conditions, Equation 5-7 is valid where Vm is the molar volume and P is the pressure. 𝑑𝐺 = 𝑉𝑚𝑑𝑃 Equation 5-7 The ideal gas assumption (PVm=RT) is the simplest way to perform the related integral operation needed to solve Equation 5-7, and find the pressure dependency of the Gibbs free energy as presented in Equation 5-8. 𝐺 = 𝐺0 − 𝑅𝑇𝑙𝑛 (𝑃𝑃0) Equation 5-8 Further expansion of the Equation 5-8 for all the species involved in fuel cell reaction leads to the Equation 5-9 relating the cell voltage to the partial pressure of the components. This is done via subtraction of Gibbs energy of the reactants from those of the products with their corresponding stoichiometric ratios. In a general reaction of aA+bB→cC+dD, ∆G equals dGD+cGC-bGB-aGA. The partial pressure of oxygen in air is 0.21; which reduces the theoretical cell voltage by 12 mV in Equation 5-9. 𝐸𝑒𝑞 = 𝐸𝑒𝑞0 +𝑅𝑇𝑛𝐹ln(          𝑃𝐻2𝑃𝑂212𝑃𝐻2𝑂 )           Equation 5-9 5.2.2 Energy losses In practice, the voltage produced at a fuel cell assembly is always below the theoretical value, and affected by several factors at any given operating condition. Even at open circuit mode where there is no current passing to the external system, the voltage is below the equilibrium cell potential (1.23 V at 25 °C and 1 atm). This is attributed to the fuel crossover phenomena where hydrogen molecules directly pass through the membrane, and react with oxygen on the cathode. Thus, the open circuit voltages in PEMFCs typically vary between 0.95 to 1 V. Upon drawing current from   79  the cell, three other losses appear, namely activation (or kinetics), ohmic (or resistance), and mass transport. A typical polarization curve (i.e., voltage vs current density) shown in Figure 5-2 is often divided into three regions, where one of the three aforementioned losses is dominant in each.   Figure 5-2: A sample polarization curve of PEMFC highlighting three overpotential regions of kinetic, ohmic, and mass transport. At low current densities where there is abundance of reactants on the catalyst surface, the kinetics of the reaction determines the overpotential. Simply, the better the kinetics the lower the overpotential. The kinetics here is often expressed in the form of Butler-Volmer Equation (BVE) presented in Equation 5-10; where i0 is the exchange current density at the catalyst surface, α is the transfer coefficient for either reduction or oxidation, and (E-Eeq) is the reaction overpotential. 𝑖 = 𝑖0 {      exp (       −𝛼𝑅𝑒𝑑𝐹(𝐸 − 𝐸𝑒𝑞)𝑅𝑇 )       − exp (       𝛼𝑂𝑥𝐹(𝐸 − 𝐸𝑒𝑞)𝑅𝑇 )       }       Equation 5-10 It should be noted that Equation 5-10 is locally valid since the reactant concentration at the catalyst surface changes throughout the fuel cell area. Also, higher exchange current density values imply better kinetics as a measure of electron exchange rate with the catalyst surface at   80  equilibrium. The transfer coefficient in the current form is dimensionless, and reflects the level of symmetry in activation energy barriers of forward and reverse reactions (typically 0.2-0.5). The BVE can be further simplified for the anode (or cathode) where oxidation (or reduction) is the dominant reaction. Since numerous studies have found that the cathodic reaction (oxygen reduction) in PEMFCs is far slower than the anodic one (i0a>>i0c), the BVE is often written for the cathode in the form of Equation 5-11. Here, the overpotential for the oxygen reduction reaction (ORR) makes for the most of voltage losses in the kinetics controlled region of the polarization curve. 𝑖𝑐 = 𝑖0,𝑐 exp (       −𝛼𝑅𝑒𝑑𝐹(𝐸𝑐 − 𝐸𝑒𝑞,𝑐)𝑅𝑇 )        Equation 5-11 Solving Equation 5-11 to find the overpotential leads to the Equation 5-12. This can be further simplified in the form of Tafel equation (Equation 5-13), which describes electrode overpotential as a function of current with two constants. The higher the Tafel slope the lower the performance becomes, and the slope itself is directly affected by the alpha values where lower ones increase the slope.  |η𝑐| = |𝐸𝑐 − 𝐸𝑒𝑞,𝑐| =𝑅𝑇𝛼𝑐𝐹ln (|𝑖|𝑖0,𝑐) Equation 5-12 |η𝑐| = 𝑎 + 𝑏. log(|𝑖|)  𝑤ℎ𝑒𝑟𝑒 𝑎 = −2.3𝑅𝑇𝛼𝐹log(𝑖0) , 𝑎𝑛𝑑 𝑏 = 2.3𝑅𝑇𝛼𝐹 Equation 5-13 Moving towards higher current densities, ohmic losses begin to dominate the potential drops. They are comprised of ionic transport-related membrane resistance (Ri), electronic resistance of fuel cell components against electrons (Re), and contact resistance of the current collectors (Rc). Here, Ohm’s law can be used to simply relate these resistances to the potential loss   81  via Equation 5-14. In most cases, electronic resistance is negligible compared to the ionic and contact resistances which are typically in the same order resulting in total resistance values (RT) of 0.1-0.2 ohms cm2. ∆𝑉𝑜ℎ𝑚 = 𝑖𝑅𝑇 = 𝑖 ∗ (𝑅𝑖 + 𝑅𝑒 + 𝑅𝑐) Equation 5-14 Moving to much higher current densities make the mass transport to the catalyst layer the rate limiting step, and create the highest overpotential. The resulting overpotential can be derived from that of Equation 5-8 in the form of Equation 5-15 in which CB is the bulk concentration and CS is the reactant concentration at the surface. This can reach a point that CS goes to zero, and the resulting current (iL) is determined by the flux of reactant to the surface. Therefore, Equation 5-15 can be rearranged in the form of the limiting current as presented in Equation 5-16; where D is the diffusion coefficient and δ is the diffusion distance. η𝐷 =𝑅𝑇𝑛𝐹ln (𝐶𝐵𝐶𝑠) Equation 5-15 η𝐷 =𝑅𝑇𝑛𝐹ln (𝑖𝐿𝑖𝐿 − 𝑖)  𝑤ℎ𝑒𝑟𝑒 𝑖𝐿 =𝑛𝐹𝐷𝐶𝐵𝛿 Equation 5-16 Ultimately, all the aforementioned voltage losses can be presented together as done in Equation 5-17 to obtain an estimation of the cell voltage at any operating condition. 𝐸𝐶𝑒𝑙𝑙 = 𝐸𝑒𝑞𝑇,𝑃−𝑅𝑇𝛼𝐹ln (𝑖𝑖0) − 𝑖𝑅𝑇 −𝑅𝑇𝑛𝐹ln (𝑖𝐿𝑖𝐿 − 𝑖) Equation 5-17 5.3 GDL and water management The performance and durability of a PEMFC is directly influenced by the water management particularly on the cathode side. For the optimal performance, it is necessary to maintain a fine balance between two competing phenomena involving water.234-235 Firstly, the PEM needs to be fully hydrated to ensure maximum proton conductivity. Consequently, the hydrogen and oxygen   82  streams are most often humidified to prevent membrane dehydration, and associated localized heating that can lead to pinhole formation. On the other hand, failure to remove excess water can block the active sites from the reactant gases, and result in severe performance losses. To assist in the management of these issues, it has become a common practice to employ a microporous layer (MPL) between the catalyst layer and the gas diffusion layer. MPLs typically consist of a layer of porous, nano-sized carbon particles in combination with polytetrafluoroethylene (PTFE) as a binder, deposited directly onto the GDL. This layer serves to not only improve mass transport, but to also help to decrease the internal cell resistance (through improved interfacial contact with the catalyst layers), and to contribute to the overall mechanical stability of the fuel cell.236-237 A recent study by Blanco and Wilkinson shows the effect and mechanism of anode and cathode MPLs on water management in a PEMFC.238 Various MPL materials have been investigated as part of the continuous efforts to enhance transport processes and overall performance of PEMFCs. They cover a wide range of morphologies, porous structures, and conductivities, such as different types of carbon blacks.239-240 Other nanostructured carbons have also been considered including carbon nanofibers/nanochains,241 and both single-walled,242 and multi-walled carbon nanotubes.243 Passalacqua et al. found that the higher pore volume and smaller pore size of acetylene black promotes oxygen diffusion and lowers water accumulation within the MPL.244 Lin and Chang also reported improvements in acetylene black performance with the addition of nanotubes.245 Other studies have investigated the carbon loading effect246 along with binder percentage247 for further MPL improvements. Among the advanced carbon materials, graphene-based MPLs could be a promising alternative due to all the versatilities discussed before. Particularly in PEMFCs, graphene has   83  shown promise mostly as an oxygen reduction catalyst support,248 generating improved electrochemical performance and thermochemical resistance.249 Other studies have also investigated doped graphene structures as alternative metal-free catalysts for the oxygen reduction reaction (ORR). This application, however, has proved to be more effective in alkaline media.250 To date there has been only a very limited focus on the incorporation of graphene into MPL structures. A study by Zamora et al.251 presents a preliminary performance evaluation of graphene-based flakes in phosphoric acid, in comparison to other carbon nanostructures.  Leeuwner et al. recently studied the interfacial interactions between a freestanding graphene foam as an MPL and the cathode catalyst layer in a hydrogen PEMFC.252 Various benefits were demonstrated, including decreased interfacial contact resistance compared to the conventional carbon black MPLs, superior adherence to the catalyst layer as well as polarization improvements in the kinetic and ohmic regions. The freestanding commercial graphene foam utilized in the aforementioned study was, however, not treated with PTFE to alter its hydrophobicity. It also formed a flaky irregular structure when subjected to compression in the fuel cell. On the other hand, for optimal MPL performance in fuel cells, engineering the surface properties and morphology of graphene-based MPLs is essential. To this end, the goal here is to produce nearly pristine graphene microsheets using the cathodic electrochemical exfoliation technique elaborated in Chapter 4, and to further evaluate their performance as MPLs. Potential synergistic effects with carbon blacks in composite MPL formulations are also explored in the present study. For the sake of brevity, the electro-exfoliated graphene products, carbon black, and their composites are referred to as ‘EGN’, ‘CB’, and ‘EGN+CB’ throughout the rest of this chapter.   84  5.4 Materials and methods 5.4.1 Graphene synthesis and characterization EGNs were produced via cathodic electrochemical exfoliation of graphitic electrodes in aprotic ionic liquids (ILs). Cathodic approach was chosen here to eliminate the complicated effects of functionalization (mostly caused by the anodic exfoliation) on the fuel cell performance, and solely investigate the structural impact. A high-purity iso-molded graphite rod (Graphite Store – 99.99% purity), with 6.35 mm diameter and 4 cm effective length, was exposed to the electrolyte and used as the cathode with a platinum wire as the counter-electrode (1.6 mm diameter and 4 cm effective length). The electrodes were placed 1 cm apart and immersed in the 0.1 M N1114-BTA/ACN solutions. A constant potential of 7 V was applied for 4 hours at 293 K (B&K Precision – 9110 DC power supply), and the external current was stable at approximately 30 mA for all the experiments. The solid products from exfoliation were then filtered and washed thoroughly with water, followed by ethanol. Samples were sonicated for 1 h to achieve stable dispersions in N-methyl-2-pyrrolidone (NMP) using a tabletop ultrasonic cleaner (VWR Scientific – B3500). This proved to be beneficial for further exfoliation of the expanded flakes.122 All the chemicals were of analytical grade, and double-distilled water was used during all of the preparation steps. Further details regarding the electrosynthesis protocols are provided in Appendix A. 5.4.2 PEMFC construction and characterization Three different MPLs were made: a CB (Vulcan XC-72R – Cabot Corporation) MPL, an EGN MPL and a composite EGN+CB MPL (with a 1:1 mass ratio). Slurries of the solid suspensions were prepared in isopropanol/water mixtures (80:20 volume ratio) containing 60 wt% PTFE (DISP 30 – Fuel Cell Earth). The suspensions were then sprayed onto Toray 060 carbon fiber paper GDLs, containing 5 wt% wet proofing (TGP-H-060 – Fuel Cell Earth) to reach two target   85  loading of 1.5 mg cm-2 (high), and 0.5 mg cm-2 (low). Each MPL was prepared to contain 20 wt% PTFE based on previously suggested optimum values for carbon black.253 The depositions were dried on a hotplate at 80 °C for 30 minutes, and subsequently sintered at 350 °C in a furnace for another 30 minutes to enhance the binding properties of PTFE. For fuel cell assembly, the prepared MPLs were incorporated into MEAs by placing them on the cathode side, facing the catalyst coated membrane (CCM). Gore Primea Series 5510 CCMs were employed with a Pt loading of 0.4 mg cm-2 on both the cathode and anode. A commercial GDL with combined carbon MPL, Sigracet 25BC (25BC – Fuel Cells Etc.) was used on the anode side for all assembled fuel cells. The components were then sandwiched together by a pneumatic piston at 827 kPa (120 psi) between the flow field plates (FFPs) as shown in Figure 5-3. Cathode and anode FFPs, both containing serpentine flow fields, were set up in a cross-flow configuration, and silicon gaskets were used to seal the entire body of construction.  Figure 5-3: Schematic representation of a PEMFC assembly and mass-transport processes through a cathode MPL.   86  For fuel cell testing, the MEAs were assembled in Tandem TP-5 fuel cell hardware from Tandem Technologies, and operated with an automated 2 kW Hydrogenics station (G100 – Greenlight Innovation). The station enabled precise control over the reactant temperatures, pressures, flow rates, humidities as well as the overall cell temperature, represented by the process flow diagram in Figure 5-4. The operating conditions on both sides were set as follows: temperature at 75 °C, inlet pressure at 202.6 kPa(g), and relative humidity (RH) at 100%. Tests involving a lower cathode RH of 20% were also performed through adjustment of the air dew point temperature to approximately 41 °C, while maintaining the cell temperature at 75 °C. The cells were leak tested at the actual operating pressures, and subsequently conditioned to ensure the complete hydration of Nafion membrane. During this procedure, air and hydrogen flow rates of 0.39 and 0.12 NL min-1 were used, respectively. The load on the cell was gradually increased until the voltage dropped to approximately 0.5 V. Operation was then continued at the same load until the voltage behavior plateaued (typically taking place after 12 hours of operation). To assess the polarization behavior, the fuel cells were operated in 1-dimensional (1D) mode on the cathode side to allow virtually uniform reactant distribution across the active area (with concentration differences between the inlet and outlet typically smaller than 10%). This helps to attain a more true representation of polarization behavior across the entire fuel cell area. In contrast, large concentration gradients lead to polarization curves that represent an average result, based on all the different concentration ranges that arise over the entire active area. The anode flow rate was maintained at 0.12 NL min-1 during the tests since the polarization behavior was not significantly influenced by an increase in the hydrogen flow rate (due to the reaction’s facile nature). Polarization tests were performed in galvanostatic mode in the forward-scan direction (from open circuit voltage to the limiting current density) with data collected for 60 s at each point.   87  Each fuel cell polarization curve was repeated three times, and reported as the average. The standard deviation was small especially in the kinetic and ohmic regions (< 10 mV), making the error bars quite invisible in the plots. To prevent any water build-up from previous tests, the cathode side was purged before the onset of every new test. High-frequency resistance measurements were also conducted at 1 kHz throughout all tests using an LCR-meter (GW-Instek LCR-819) to monitor the cell’s internal resistance.  Figure 5-4: Process flow diagram of the PEMFC monitoring system used for the experimental section showing the jacket cooling/heating elements along with the humidifiers on the gas inlets to further hydrate the CCM.  The through-plane gas permeability of the samples was measured by employing an experimental apparatus similar to that used by Gostick et al.,254 while water permeability measurements were performed by Porous Materials Inc (USA). In each case, the permeability coefficient was determined from measured flow rates and pressure drops via Darcy’s law shown in Equation 5-18 where 𝑘 is the permeability coefficient (m2), 𝑣𝑔 is the gas velocity (m/s), 𝜇𝑔 is the FC TC PT TC PTFC TC PT TC PTTC PT TC PTHydrogen InAir OutHeating Liquid OutHeating Liquid InAir InHydrogen OutLoadBankHumidifierHumidifierCooledCooledFuel Cell  88  gas viscosity (Pa.s), ∆𝑥 is the sample thickness (m), and ∆𝑃 is the pressure drop measured across the sample thickness (Pa). 𝑘 =𝑣𝑔𝜇𝑔∆𝑥∆𝑃 Equation 5-18 Oxygen was used for the gas permeability measurements. Following the example of Williams et al.,255 the apparent porosity of the MPLs were estimated based on a weight difference before and after immersing the samples in a wetting liquid (hexane was used in this case). Through-plane resistance and contact angle measurements were furthermore performed by employing the techniques previously described in Leeuwner et al.252 5.5 Results and discussion The polarization and power density curves obtained for the three different MPLs at 100% RH are illustrated in Figure 5-5a, and b, respectively. When comparing the EGN and CB MPLs, two major differences in polarization behavior are observed. The EGN MPL improves the polarization performance in both the kinetic and ohmic regions (by up to 30 and 50 mV respectively), but suffers from severe mass transport limitations. However, through combination of CB and EGN (EGN+CB MPL), a dramatic enhancement is also observed in the mass transport limiting region, compared to both the EGN and CB only MPLs (Figure 5-5a). The composite EGN+CB MPL also exhibits the same extent of performance improvement observed for the EGN MPL in the kinetic and ohmic regions, even though it contains a lower loading of graphene (0.75 mg cm-2 vs 1.5 mg cm-2). A maximum power density of 1188 mW cm-2 is obtained with the EGN+CB MPL, corresponding to an approximate 30% and 70% improvement compared to the CB and EGN only MPLs, respectively (Figure 5-5b).  To gain a better understanding of the various MPLs’ impact on polarization behavior, the cell resistance and cathodic pressure drop were also measured (Figure 5-5c, and d, respectively).   89  The CB MPL has the highest ohmic potential drop followed by the EGN+CB MPL, and finally the EGN MPL, with the latter showing a 30 mV decrease compared to the CB MPL at 1000 mA cm-2. The reduced resistances (or ohmic losses) for EGN-based MPLs can be attributed to improved interfacial contact between the MPLs and the catalyst layer.  Figure 5-5d shows the cathode pressure drops (between the cathode inlet and outlet) associated with each MPL as a function of air stoichiometry at a constant current density of 1000 mA cm-2. The EGN-based MPLs are generally associated with higher pressure drop regimes compared to the CB MPL (this trend becomes more evident at air stoichiometries above 10). The addition of CB to EGN, however, helps to lower the overall cathode pressure. Ex-situ permeability measurements show an opposite trend with the lowest through-plane gas permeability obtained for the composite EGN+CB MPL (2.12x10-15 m2 for EGN+CB vs. 3.58x10-15 and 2.44x10-12 m2 for EGN and CB, respectively). It should be noted, however, that through-plane gas permeability measurements (via forced convective flow), do not include/represent the effect of oxygen diffusion towards the catalyst layer. The measurements furthermore also do not include the effect of simultaneous water transport. Ultimately, it is therefore believed that the EGN+CB MPL’s mass transport improvement (and decreased cathode pressure drop) is due to the presence of CB which induces structural changes, and opens more pathways for water ejection, thereby also allowing more oxygen diffusion towards the catalyst layer. Since all three MPLs have similar surface contact angles (140°-150°), mass transport differences are also not attributed to significant surface wetting differences at the MPL/catalyst interface.   90   Figure 5-5: (a, and b) Polarization and power density curves for different MPLs at 100% RH. (c) Ohmic drop for different MPLs at 100% RH. (d) Cathode pressure drop as a function of air stoichiometry at a current density of 1000 mA cm-2 and 100% RH. To compare the performance in the kinetic region more rigorously, a Tafel analysis was performed on the collected fuel cell data. It was assumed that the anodic surface overpotential is negligible. The smaller Tafel slopes obtained for the EGN-based MPLs (95±3 mV dec-1 vs 110±3 mV dec-1) indicate more favorable kinetics, and larger transfer coefficients for the oxygen reduction reaction. It was, however, confirmed that the produced EGN itself does not show any catalytic activity for the ORR in acidic media (i.e., it has a high overpotential compared to the Pt   91  catalyst, as shown in Figure 5-6) using rotating disk electrode (RDE). This would suggest that the EGN-based MPLs create a more favorable chemical environment for the ORR, which leads to the improvements observed in the kinetic region. It should be noted that extensive ORR experiments were performed throughout this thesis to explore the application of the exfoliated products as standalone carbon-based catalysts. Although none of the samples showed promising activity in acidic media (manifested by Figure 5-6a), interesting observations were made in alkaline media which are elaborated in Appendix B. More details of the experimental conditions along with the underlying theory are also included in Appendix B. Additionally, the eletrochemical stability tests (Figure 5-6b) indicated the absence of any features such as the quinone-hydroquinone redox couple, and can likely be attributed to the very low oxygen content of the the EGN (see XPS results in Figure 5-7a).  Figure 5-6: (a) Cathodic sweep of the rotating disk electrode studies of the oxygen reduction reaction on the graphene compared to the platinum catalyst at 1600 rpm in 0.1 M H2SO4 solutions. (b) Electrochemical stability tests using cyclic voltammetry in nitrogen saturated solutions at 5 mV s-1 rate. Following fuel cell analyses, a series of additional characterizations were conducted on the EGN, as shown in Figure 5-7. An XPS survey scan of the EGN (Figure 5-7a) showed very low   92  oxygen content (~3%), pointing towards promising intrinsic conductivities of the material.182 For the C 1s narrow scan in particular, the two carbon–oxygen peaks (C–O and C=O), located at 286.5 eV and 287.9 eV, respectively,226 indicate a relatively small contribution of 11% to the total C 1s response (Figure 5-7a, inset). This is further supported by the Raman spectroscopy data which suggests the formation of fairly intact and defect-free 2D EGN sheets, demonstrated by the high G/D peak ratios in Figure 5-7b. The successful production of ultrathin EGNs from graphitic electrodes is also confirmed through the Raman 2D peak evolution (Figure 5-7c). For the EGN a distinct single peak is observed compared to the 2D peak of the graphite which, in turn, also displays a smaller adjacent peak.215 The high quality of the produced graphene is also reflected in its in-plane sheet resistance of ~0.5 Ω cm (at 20-30 µ thickness) which is quite low compared to other values reported for graphene coatings in the literature.59,82    93  The physicochemical differences between the examined MPLs are further highlighted by the cross-sectional and top view FESEM imaging of the layers (Figure 5-8). It should be noted that all of the prepared MPLs were compressed at 827 kPa (120 psi) prior to imaging to emulate the structural changes of the sprayed layer under actual fuel cell conditions. Surface imaging shows that the spherical CB particles cover the GDL’s carbonaceous fibers with a greater tendency to agglomerate (Figure 5-8a, left), while planar sheets of EGN evenly spread over the entire GDL surface (Figure 5-8a, middle). The addition of CB to EGN creates a relatively homogenous distribution of spherical particles amongst the graphene microsheets (Figure 5-8a, right).  The cross-sectional images (Figure 5-8b) provide further valuable insights into the different porous structures of the MPLs. In the CB MPL, the particles form a thick, dense layer  Figure 5-7: Characterization data for EGN obtained by cathodic electro-exfoliation of iso-molded graphite electrodes in N1114 BTA/ACN electrolyte at 7 V and 293 K. (a) XPS survey scan, and C 1s narrow scan data. (b) Raman spectra of the EGN compared with graphite, particularly for G/D peak ratios. (c) Raman spectra of the EGN compared with graphite, particularly for 2D peak transformation.   94  on top of the carbon fiber paper, while the macropores of the substrate remain mostly exposed and only partially filled by CB particles. The EGNs, on the other hand, assemble as highly compressed stacks of planar sheets over the entire substrate, displaying lower porosity and a tortuous porous network (Figure 5-8b, middle). This explains both the significant mass transport limitations, and larger pressure drop observed in case of the EGN MPL (Figure 5-5a, and d, respectively). However, when CB is added to EGN, restacking of the EGN sheets is avoided by the CB particles that act as fillers between the EGN layers, and are homogenously dispersed throughout the thickness of the layer (Figure 5-8b, right). This morphological structure accounts for the low gas and liquid permeabilities (Table 5-1) associated with significant mass transfer limitations, as discussed in the previous paragraphs.    95  The effect of the CB particles on the composite MPL’s porous structure is confirmed by image processing of the MPL cross-sections (Figure 5-9). For the EGN MPL, the void area between the graphene sheets is calculated to be about 10%, while the EGN+CB MPL shows a significant increase in voidage to 55%. Although this results in a lower macroscopic porosity for  Figure 5-8: FESEM images of the MPLs made from CB (left), EGN (middle), EGN+CB (right): (a) top view, and (b) cross sectional view. The inset in the bottom left is false colored to further highlight the distribution of CB through the EGN layers. Table 5-1: Mass transport characterization data of different MPLs. MPL Through-plane gas permeability [m2] Porosity [%] Through-plane water permeability [m2] CB 2.4 x 10-12 70 - 75 5.9 x 10-12 EGN 3.6 x 10-15 30 - 40 4.5 x 10-14 EGN+CB 2.1 x 10-15 20 - 30 6.5 x 10-14   96  the composite MPL compared to the EGN only version (Table 5-1), it also facilitates improved water removal, as illustrated by the water permeability measurements, mass transport improvement and the decrease in pressure drop (comparing EGN and EGN+CB in Figure 5-5a, and d). It is further believed that the created bimodal porosity specifically enhances mass transport by allowing sufficient gas transport through the larger pores between EGN sheets, while water wicking preferentially takes place through the micro-pores between CB particles. It’s worth noting that BET analyses of the prepared MPLs were attempted, but did not deem feasible. This is due to the low MPL loadings which do not allow significant adsorptions to be made, thereby compromising the accuracy of results. Hg porosimetry, as the other typical method to determine porosity, was also not an option due to the larger amounts of material required. In the absence of this information, the alternative hexane absorption method was employed to make inferences regarding structural aspects which may influence mass transport.   97  All the fuel cell performance results presented in Figure 5-5 were obtained at an RH of 100%, which assures that the Nafion membrane is well-hydrated.256 Operating at high humidity however, necessitates the inclusion of additional water management systems such as humidifiers which, in turn, may introduce parasitic losses and additional space and cost requirements. Consequently, there has been extensive research on alternative solutions to enable fuel cell operation at reduced humidity, particularly in the field of transportation.257 To assess how the EGN-based MPLs would respond to humidity, the MEAs were subjected to a low cathode RH of 20%. To emphasize the difference between the two humidity conditions, results are expressed as the percentage change in voltage (at a specific current density) between 100% and 20% RH, relative to the voltage at 100% RH (Figure 5-10a). A negative voltage difference implies performance degradation as the cathode humidity is decreased from 100% to 20%. Similarly, the  Figure 5-9: Image processing and schematic representation of structure for (a) EGN MPL, and (b) EGN+CB MPL. The darker blue color in the middle show the void areas in between the EGN sheets.   98  percentage change in internal cell resistance is plotted in Figure 5-10c, with a positive difference indicating a resistance increase from 100% to 20% RH. The MEA with the CB MPL shows a dramatic performance loss of up to 40% at a current density of 1600 mA cm-2 and a maximum power density of approximately 650 mW cm-2 (Figure 5-10a, and b). This is associated with a 30% to 50% increase in the overall cell resistance, indicating severe membrane dehydration (Figure 5-10c). In contrast, all the EGN-based MPLs were able to maintain their performance at 20% cathode RH, while also showing slight improvements in the mass transport limiting region, likely due to decreased flooding (Figure 5-10a, and b). For the most part, their resistance/ohmic drop values remained relatively unchanged (Figure 5-10c). The maximum power density at 20% cathode RH for the EGN+CB MPL is 1188 mW cm-2, indicating very little change from 1173 mW cm-2 at 100% RH (Figure 5-10b vs Figure 5-5b). This results in an even greater improvement in peak power for the EGN+CB MPL compared to the conventional CB MPL, which now exceeds 80% compared to an improvement of 30% at 100% RH. The above results suggest that the outstanding performance at low humidity of the EGN-based MPLs, unlike the CB MPL, is due to better water retention near the membrane, thereby ensuring that it remains sufficiently hydrated. This behavior can be attributed to the nature of the MPLs’ porous structure. The stacks of graphene sheets close to the membrane create more tortuous pathways that make it more difficult to lose water through the MPL at drier conditions. As previously discussed, in the case of EGN-only MPLs, this water retention capability can cause unfavorable flooding. However, for the EGN+CB MPL a balance is maintained between the water retention capabilities of the graphene sheets, and the CB which creates the additional pathways required for sufficient water removal. This further suggests that the EGN:CB ratio can   99  be tailored to balance the competing phenomena as required for specific applications and operating conditions, while also yielding improvements over more conventional CB MPLs.  Longer term tests were performed at low humidity and a constant current of 1000 mA cm-2 as shown in Figure 5-10d. Over the course of six hours, both EGN-based MPLs were able to maintain performance very well. The EGN MPL shows a small degradation rate of ~1 mV h-1. Within the first hour, the EGN+CB MPL displays a slight performance decrease, after which a new steady state is established with a degradation rate smaller than 3 mV h-1. In contrast, the CB MPL showed significant performance loss within the first hour (410 mV h-1) associated with drying and irreversible damage to the membrane.   100  Lastly, the surface SEM images of the 0.5 mg cm-2 MPLs indicate that the lower loads result in less complete coverage of the carbon fiber GDLs (Figure 5-11). This increased exposure of GDL macropores leads to significant mass transport improvements for all three MPLs (Figure 5-12a). This is especially true for the EGN MPL which displays an increase in the limiting current density from 1500 to 3600 mA cm-2 as the load is decreased from 1.5 to 0.5 mg cm-2. At the lower load, the EGN-based MPLs also result in improvements in the kinetic and ohmic  Figure 5-10: (a) Change in performance from 100 to 20% cathode RH (expressed relative to performance at 100% RH). (b) Power density curves at 20% RH. (c) Change in resistance/ohmic drop from 100 to 20% RH (expressed relative to resistance at 100% RH). (d) Longer term operation at constant current density of 1000 mA cm-2 and 20% cathode RH.   101  regions, compared to the CB MPL. However, these improvements are less drastic than the 1.5 mg cm-2 MPLs. For example, the voltages achieved in the ohmic region for the EGN+CB MPL with a 0.5 mg cm-2 load, are 30-40 mV less than those of the 1.5 mg cm-2 version. This is attributed to higher interfacial contact resistance associated with the rougher MPL surfaces. At these conditions, the EGN+CB MPL shows ~20% and ~15% improvement in the maximum power density compared to the CB and EGN only MPLs, respectively (Figure 5-12b).  Figure 5-11: SEM images of (a) CB, (b) EGN and (c) EGN+CB MPL surfaces with total material load of 0.5 mg cm-2. Similar to the 1.5 mg cm-2 MPLs, the EGN-based MPLs at 0.5 mg cm-2 maintain performance very well at a lower humidity during polarization (Figure 5-12c). The associated ohmic loss/drop at 20% RH however, is 20 and 25 mV higher than in the case for the 1.5 mg cm-2 load MPLs, indicating a greater degree of membrane dehydration. On the other side, the CB MPL shows performance decrease that is more severe than for its 1.5 mg cm-2 counterpart (80% performance decrease versus 40%), and is attributed to severe drying of the membrane (Figure 5-12c). This is corroborated by ohmic losses that are up to 70 mV higher than for the higher load version at similar conditions. Polarization tests at 20% RH furthermore caused irreversible damage to the MEA, rendering it unsuitable for further testing.    102   Figure 5-12: Performance results for different MPLs at a loading of 0.5 mg cm-2. (a, and b) Polarization and power density curves at 100% RH. (c) Change in performance from 100 to 20% RH (expressed relative to performance at 100% RH). (d) Longer term operation at constant voltage of 1000 mA cm-2 and 20% RH; no results are displayed for the CB MPL due to irreversible damage caused during polarization at 20% RH. Even though the EGN-based MPLs also show promise over the conventional CB MPLs at these lower loads, longer term tests indicate that it is less suited for low humidity applications compared to 1.5 mg cm-2 MPLs. This is illustrated by a more pronounced increase in the degradation rates to 15 and 25 mV h-1 for the EGN and EGN+CB MPLs, respectively   103  (Figure 5-12d). EGN-based MPLs, which partially cover the underlying GDLs here, serve as less effective water retention barriers due to increased water outflow through the GDL macropores.  5.6 Summary and conclusions The performance of ultrathin EGN sheets, synthesized electrochemically from iso-molded graphite rods, was investigated as MPLs for PEMFCs. Compared to the traditional CB (in particular Vulcan XC-72R), the EGN-based MPLs introduce distinct performance enhancements.  Firstly, significant improvements were observed in the kinetically controlled region (30 mV increase at < 100 mA cm-2) as well as the ohmic region (54 mV increase at 760 mA cm-2). However, at 100% RH, the EGN MPL suffered from mass transfer limitations due to flooding; which can be overcome by combining CB with EGN to form a composite MPL. This resulted in a beneficial synergistic effect on the power density with an approximate 30% and 70% increase compared to the CB and EGN only MPLs, respectively. The performance results were further supported by extensive characterization showing that the spherical CB particles disperse homogenously among the stacked graphene layers, thereby avoiding restacking of the EGN sheets. This creates a bimodal porosity which is believed to enhance gas diffusion in the larger pores between the EGN sheets and water wicking in the CB micro-pores.  Interestingly, EGN-based MPLs also showed high tolerance to a low humidity environment (e.g., 20% RH) while CB MPLs dried out within the first hour of operation. The performance improvement between EGN+CB and CB MPLs at 20% RH increased to over 80% in peak power. The MEA with the EGN+CB MPL achieved a power density of 1188 mW cm-2 at 20% RH, which is very similar to the performance obtained at 100% RH (1173 mW cm-2). Furthermore, the performance under low humidity conditions was stable during six hours of operation. This   104  promising observation can be attributed to the enhanced oxygen transport to the catalyst layer (increasing water generation) coupled with improved water retention in the MPL Overall, EGN-based MPLs do not only result in higher power densities at inlet 1D operating conditions, but their ability to maintain performance at much lower RHs has significant implications for durability, operational flexibility, and cost reduction in PEMFCs. These very promising and novel findings warrant further investigation aimed at optimizing the MPL composition in terms of the weight ratio of the two components (EGN and CB), PTFE content, and the total conductive material load.     105  Chapter  6: Application II – Microbial fuel cells4 6.1 Introduction Microbial fuel cells (MFCs) are devices capable of electricity generation by employing bacteria to decompose organic compounds through a series of electrochemical reactions.258 Perhaps, the most relevant and immediate implication of MFCs lies in the wastewater treatment for simultaneous power generation and degradation of organic matter.259 Studies have shown that wastewater in a modern treatment plant may contain as much as nine times the energy used for its treatment.260 Such significant amount of stored energy could potentially be harnessed by MFCs. Although their power output has not yet reached levels required for self-sustaining a wastewater treatment plant, they could theoretically be used to offset energy consumption throughout the treatment process;261 especially considering the energy-intense nature of the conventional wastewater treatment operations such as aeration.262 Under anaerobic conditions, a typical MFC process involves bacterial decomposition of organic matter (Equation 6-1), from organic-rich sources such as wastewater, which produces electrons that are passed on to the anode, and transferred through an external circuit towards an electron acceptor at the cathode, usually air or oxygen, where reduction occurs (Equation 6-2).263 Figure 6-1 illustrates a basic schematic for two-chamber microbial fuel cell with the bacterial anode operating in anaerobic mode and oxygen reduction reaction occurring at the aerobic cathode. Similar to PEMFCs, the two chambers are separated by a polymer electrolyte membrane (mostly Nafion) with high proton conductivity.                                                   4 A version of this chapter has been published in Biosensors and Bioelectronics:  A Taheri Najafabadi, N Ng, E Gyenge, Biosensors and Bioelectronics, 81 (2016), 103-110.    106  Anode: 𝑂𝑟𝑔𝑎𝑛𝑖𝑐 𝑀𝑎𝑡𝑡𝑒𝑟 + 𝐻2𝑂 → 𝑛𝐶𝑂2 + 𝑛𝐻+ + 𝑛𝑒− Equation 6-1 Cathode: 𝑂2 + 4𝐻+ + 4𝑒− → 2𝐻2𝑂 Equation 6-2   Figure 6-1: A basic schematic for two-chamber microbial fuel cell with the bacterial anode operating in anaerobic mode, and oxygen reduction reaction occurring at the aerobic cathode. The underlying fundamentals of MFCs have significant similarities to those explained for PEMFCs in the previous chapter. Thus, the same analogies can be implemented to explain the electrochemical observations for MFCs. The major difference here is that the anodic process involving bacteria is far slower than oxygen reduction reaction at the cathode. Thus, the main attention in MFCs is typically towards improving the bacteria-electrode interactions as well as the organic matter digestion kinetics under anaerobic conditions.264 This makes the whole discussion around the basic thermodynamics of the system, and the nature of voltage losses quite challenging. Also, since the bacteria metabolism varies depending on the type and organic matter mediums, a vast number of stable intermediates and end products are reported, making the analysis task even more complicated. To avoid such variations in this chapter, only one component (i.e., the anode) is modified with various graphene samples (i.e., from anodic and cathodic exfoliation), and other   107  carbon materials, while keeping the other variables untouched during the experimental investigations. This aims at isolating the effect of electrochemically exfoliated graphene on improving the anode performance.  6.2 Anode studies Enhancing the MFCs’ performance requires further fundamental research on their design, and substrate composition along with exploring novel electrode materials,265 and better understanding the underlying metabolism of the anode-bacteria interactions.266 Overall, the majority of research has been devoted to studying the anode at the expense of the cathode and other components. As mentioned earlier, this is mainly due to the fact that the organic matter oxidation on the anode is typically the main bottleneck, and any improvement on the anodic interactions translates into further increase in the overall efficiency.267-268 Among the advanced materials used as the anode for MFCs, carbon-based electrodes are the most promising due to their chemical inertness, high surface area, and low cost.269 Synthesis of unique carbon nanomaterials such as porous carbons, fullerenes, carbon nanotubes (CNT) and graphene has had a great impact on the research and development of high-performance materials all across energy storage and production area.270 Particularly, graphene with its fascinating physicochemical properties while being remarkably biocompatible271-272 could be an excellent choice. More importantly, chemical treatment of graphene, which tailors its electronic properties, significantly broadens its applications in cross-disciplinary areas with structural doping and surface functionalization.273-274 Thus, graphene can be a viable alternative to the traditional MFC anodes such as graphite rod, activated carbon, and carbon cloth.275    108  Despite the extensive research on MFC electrode materials, the development of graphene-based anodes is still in the early stages, and the reports on related studies are relatively sparse in literature. Liu et al. examined the electrochemical performance of a graphene-modified carbon cloth anode, and reported a 2.7-fold improvement in power density compared to a plain carbon cloth anode.276 Xiao et al. observed that crumpled graphene-modified anodes can produce twice as much power as activated carbon anodes.277 Zhang et al. attributed the increase in power generation to the high surface area of the graphene-based anode, and subsequent promotion of microbial loading on the anode compared to an unmodified stainless steel mesh.278 Hou et al. synthesized graphene composites with polyaniline (PANI) reporting a maximum power density of 1390 mW m−2, with ~40% improvement over reduced graphene oxide performance.279 Zhao et al. also modified graphene’s surface using ionic liquids, and reported beneficial effects with power densities exceeding 600 mW m-2.280 Finally, Tang and coworkers obtained 1.7-fold higher power density over graphite plate for graphene sheets produced via graphite exfoliation in ammonium sulfate solutions.281 Nonetheless, it is important to note that graphene’s physicochemical properties such as thickness, number and type of structural defects and oxygen content, are strongly dependent on the production technique;98 which can have a major impact on the material’s performance. Thus, the novel synthesis technique of this thesis provided some unique opportunities to examine the products as anode materials, and explore the potential performance differences in MFCs. Furthermore, sample-modified anodes combining GNs with other low-dimensional allotropes of carbon, CNTs, and carbon black in particular, are also investigated. This is due to synergistic effects observed in other areas of energy research,282 that might result in even stronger MFC anode performance owing to its unique combination of morphology and surface chemistry.   109  Lastly, in terms of MFC configuration, it is shown that removing the ion exchange membrane along with the use of air cathodes in a single chamber MFC design brings about substantial improvements (e.g., the power density increases by 88%) due to increased ionic conductivity, while eliminating the possibility of membrane fouling.283 Thus, the home-made graphene-based anodes are combined with a simple single-chamber MFC design, and the resulting performance is compared with other carbon-based materials. Additionally, nonprecious metal air cathode catalysts, namely manganese oxide, are used for the oxygen reduction reaction (ORR), which further contributes to the feasibility of the proposed system.284 Such state-of-the-art MFC constructions are the key elements to obtain higher performances, and improve scalability. Figure 6-2 depicts the schematic diagram of the MFC configuration used in the present work with biofilm growth on the graphene-based anode, and oxygen reduction on the air-cathode catalyzed by MnOx (Figure 6-2a) along with the images of the multiple constructed units (Figure 6-2b,c).  Figure 6-2: (a) Graphical representation of the single-chamber MFC showing biofilm growth on the graphene-based anode, and oxygen reduction on the air-cathode catalyzed by MnOx. The red arrows indicate the flow of electrons from the anode to the cathode through an external load. (b, and c) Images of the MFC units used in the present work.   110  6.3 Materials and methods GNs were synthesized via simultaneous electrochemical exfoliation procedure elaborated throughout Chapter 4 using 0.1 M N1114-BF4 ionic liquid solutions in acetonitrile. Figure 6-3 shows the anode preparation overview along with the inoculation process. All bacteria cells in this experiment were derived from a commercial agar slant containing pure Escherichia coli B (Merlan Scientific – 155070), an easily available exoelectrogenic bacterium especially for carbonaceous substrates with polytetrafluoroethylene-based binders.285-286 A small sample of cells was removed from the source using an inoculum loop, previously sterilized under a Bunsen burner flame, and streaked onto a petri dish containing tryptic soy agar nutrient (Sigma-Aldrich – 22091). The petri dish was sealed and incubated at 310 K for 24 hours. Further scale-up to a liquid culture was performed by transferring cells from the petri dish into a shake flask containing 50 mL of 30 g L-1 sterile tryptic soy broth (Sigma-Aldrich – 22092). The cells were agitated at 310 K and 150 rpm for 24 hours with consecutive dilutions via stock tryptic soy broth until an optical density reading of 0.8 at 600 nm (equivalent to approximately 6.2x108 cell mL-1) was achieved using a spectrophotometer. Subsequently, the assembled electrode inoculation brace was submerged into the solution, and incubated at 310 K for 18 hours. The inoculated sample-modified anode was then used for the MFC operations. As for the anode preparation itself, GNs were dispersed in isopropanol (IPA), then mixed with 5 wt% Nafion solution (Sigma-Aldrich – 274704) as binder, and sonicated for 1 h to form a homogenous solution using a tabletop ultrasonic cleaner (VWR Scientific – B3500). The dry Nafion to carbon weight ratio on the electrode was set at 15 wt%. The ink solutions were subsequently deposited on carbon cloth (Fuel Cell Earth – CCP40) with a machine sprayer, and dried in the oven at 333 K for 24 hours. For performance comparison, other MFC anodes with   111  porous carbon black (Cabot Corporation – Vulcan XC-72), and CNTs (Sigma-Aldrich – 724769) were prepared with the same loading of 2 mg cm-2 on carbon cloth. An unmodified carbon cloth anode was also used as a background experiment. All the chemicals were of the analytical grade, and double-distilled water was used during all of the preparation steps. After anode inoculation, the membrane-less MFC with a 20 mL empty bed volume (EBV) was assembled. The inoculated anode was secured on one end of the cell with the circular exposed surface of 2.5 cm diameter. On the other side, the manganese oxide air cathode (Gaskatel – 82010) was sandwiched between two plates with the same breathing window of 2.5 cm diameter (4 cm apart from the anode). Next, the nutrient solution was pipetted into the cell; which was prepared by mixing the following volumetric proportions of 50% tryptic soy broth (Sigma-Aldrich – 22092), 49.125% 0.1 M phosphate buffer solution with pH 7.4 (Sigma-Aldrich – P4417), and 0.875% vitamin/mineral solution mix (Sigma-Aldrich – K3129), along with powdered glucose added to reach a final concentration of 1 g L-1. The solution was purged with nitrogen for 15 minutes to ensure the complete oxygen removal, and the sampling ports were sealed with parafilm afterwards. More details on the experimental setup and protocols are included in Appendix C.  With the assembled MFCs, electrochemical characterizations were carried out using a commercial potentiostat/galvanostat (Princeton Applied Research – 263A) equipped with electrochemical impedance spectroscopy (EIS). Polarization curves were measured at the sweep rate of 5 mV s-1 starting from the open circuit potential (OCP) value. The power output was normalized to the electrode’s geometric working area (W m-2), and the MFC’s empty bed volume (W m-3), respectively. More polarization discharge experiments were performed at constant currents to evaluate the longer-term performance.   112  Furthermore, electrochemical impedance spectroscopy (EIS) was conducted at OCP in the frequency range of 104-10-3 Hz with anode as the working electrode, and the cathode serving as both the counter and reference electrode. The EIS data was then analyzed using EIS Spectrum Analyzer software with an equivalent circuit assuming that the anode reaction is affected by both reaction kinetics and diffusion with the symbol Rs for solution resistance and Rct for charge transfer resistance. A constant phase element (CPE) is used instead of a capacitor in order to model double layer capacitance when surface roughness or a distribution of reactions across the surface possibly affect overall kinetics.287 Thus, the Nyquist plot forms via superimposition of a preceding frequency-dependent semi-circle (high frequency region) and a subsequent straight line (low frequency region), with the former’s diameter representing the charge-transfer resistance.288     Figure 6-3: Graphical representation of the graphene anode preparation and inoculation process.   113  6.4 Results and discussion Figure 6-4 shows the polarization curves using the anodic and cathodic GNs compared to that of plain 3D carbon cloth (CC) electrode. The error bars indicate the performance reproducibility in the course of 3-day testing. The current and power outputs are normalized based on the electrode area as well as the empty bed volume of the MFC. In terms of OCP, the plain carbon cloth produced the highest voltage of 988 mV, while the cathodic and anodic GNs products gave 949 mV and 930 mV, respectively. However, both anodic and cathodic GN products showed superior polarization performance as anodes in the MFC compared to the conventional 3D CC electrode. In terms of maximum current, anodic products with 10.25 A m-2 came first while cathodic GNs produced 9.5 A m-2 at 50 mV. These numbers were 310% and 280% higher than the maximum current produced on CC at the same potential (2.5 A m-2 at 50 mV), respectively. As for the maximum power density, the anodic GN exfoliates emerged as the most promising candidate with 2.85 W m-2 at 530 mV compared to the cathodic GNs (2.3 W m-2 at 450 mV). The maximum power densities were 330% and 250% higher with the GN electrodes compared to the CC electrode (0.66 W m-2 at 506 mV), respectively. It should be noted that the observed power densities with the GN electrodes are significantly higher than the results presented in the introduction. For example, Liu et al. used a dual-chamber MFC construction separated by a Nafion membrane, and reported a maximum of 2.7-fold improvement in power density for rGO anodes compared to a plain carbon cloth anode.276 Similarly, while Hou et al. aimed at improving the rGO performance by adding PANI (~40% improvement), their maximum power density of 1390 mW m−2 in a two-chamber MFC is still less than half of what was achieved in this work.279 More comprehensive data comparison with literature is offered in Table 6-1. The MFC assemblies produced stable currents throughout the   114  three days of testing in 20-minute batch periods at the cell’s peak power current as shown in Figure 6-5. Table 6-1: The data comparison between various reports on the use of graphene as anode material for microbial fuel cells. The majority of the works have used reduced graphene oxide as the primary component. Anode Material Bacteria Type Feed Maximum Power Density (W m-2) Reference Carbon cloth + graphene P. aeruginosa Glucose 0.052 Liu et al. 2012276 Carbon paper + graphene Mixed Glucose 0.368 Guo et al. 2014289 Graphene/ionic liquid Shewanella Lactate 0.601 Zhao et al. 2013280 Graphite + graphene Mixed Acetate 0.670 Tang et al. 2015281 Graphene/polyaniline foam S. oneidensis Lactate 0.768 Yong et al. 2012290 Graphene/TiO2 S. oneidensis Lactate 1.060 Zhao et al. 2014291 Carbon cloth + polyaniline/graphene Mixed Acetate 1.390 Hou et al. 2013279 Carbon cloth + graphene/SnO2 E. coli Glucose 1.624 Mehdinia et al. 2014292 This work E. coli Glucose 2.850   Figure 6-4: (a) Polarization, and (b) power curves for anodic and cathodic GNs compared to carbon cloth normalized based on both electrode area and MFC’s empty bed volume with the step-sweep rate of 5 mV s-1 starting from the open circuit potential.   115   Figure 6-5: Long-term performance data for MFCs with graphene-modifying anodes. The points are the average of 20-minute long tests with fresh nutrients at the peak power current during 3-day experiments. It is important to note that the surface chemistry of the anode in MFCs influences the electron transfer rate and biofilm development.293 Inducing hydrophilic functional groups on the electrode surface has been shown to improve the MFC performance by enhancing bacteria attachment to the electrode, and further promoting electron transfer from bacteria to the surface.294 Recently, Pinto et al. also reported improved biocompatibility for graphene compounds with higher oxidation levels and smaller particle sizes.295 As explained in Chapter 4, the electro-exfoliated GN products exhibit distinctive levels of functionalization/disorder. Consequently, this has led the more functionalized and relatively smaller anodic products generating better MFC polarization performance compared to the cathodic exfoliates (Figure 6-4a). Next, to examine the performance of the GN-modified anodes against other conventional porous carbon materials, a CC anode was sprayed with Vulcan XC-72. Figure 6-6a compares the polarization curves for the anodic GNs compared to those of Vulcan-modified electrode, respectively; with similar loading and experimental conditions. As demonstrated by the curves,   116  anodic GNs outperform Vulcan XC72 with 40% higher OCPs, and 90% increase in peak power density. Importantly, Vulcan also falls short in the mass transfer controlled region (i.e., nutrient diffusion through the anodic biofilm) demonstrated by the lower slopes in current increase upon approaching voltages below 400 mV; which indicates that its porosity/morphology might not be well-suited for this particular application as opposed to the relatively flat and open-ended structure of anodic GNs. The MFC polarization performance was further supported by electrochemical impedance spectroscopy (EIS) presented in Figure 6-6b. By data fitting, the Nyquist plots reveal that the charge-transfer resistance (Rct) on the two electrodes was quite different. The Vulcan-modified anode generated almost three times higher Rct compared to the anodic GN exfoliate (i.e., 577 Ω vs. 183 Ω). The faster electron transfer for the GN anode is responsible for the higher power density observed in the fuel cell polarization experiments. The radical structural differences are further manifested by the SEM images of the materials coated on carbon cloth showing the planar nature of the GNs (with irregular shapes), and the fine spherical particles of Vulcan (Figure 6-6c). Thus, it is plausible to assume that the anode morphology could have an impact on the MFC polarization behavior affecting the nutrient diffusion and cell growth.   117   Figure 6-6: Characterization of anodic GNs compared with Vulcan XC-72. (a) Polarization curves normalized based on both electrode area and MFC’s empty bed volume obtained with a sweep rate of 5 mV s-1 starting from the open circuit potential. (b) EIS measurements at the open circuit potential in the frequency range of 104-10-3 Hz. (c) SEM images of the electrodes coated with graphene and carbon black compared with plain carbon cloth fibers. In the last step, the effect of adding carbon nanotubes to the GN modified anodes was explored (Figure 6-7). Thus, the best performing material in the previous step (i.e., anodic GNs) was combined with 50% weight ratios of CNTs. However, contrary to the synergistic effect observed in other areas of energy storage and production, a drastic drop in overall performance of the materials was witnessed. Although the GN-CNT composite produced appreciable amounts of current at 50 mV (10.7 A m-2, ~4% higher than anodic GNs at the same voltage), it showed far lower OCPs (360 mV compared to 930 mV for the anodic GNs). This was further translated into   118  dramatically lower maximum power densities of 1.51 W m-2 at 226 mV (~50% performance drop vs anodic GNs). The above-mentioned pattern of performance became worse when CNTs were tested alone as anode material in MFC (Figure 6-7a, b). Importantly, a major delay was observed regarding the CNT anodic behavior in the first 24-hour period after deaeration (Figure 6-7c, d). The peak power density with CNT electrodes increased from the initial value of 0.03 W m-2 at 106 mV, to 0.41 W m-2 at 185 mV after 6 hours of rest time, and finally it reached 0.94 W m-2 at 228 mV (~50% improvement over CC) after 24 hours, and remained stable afterwards. Similarly, the maximum current/OCPs were quite insignificant (0.4 A m-2/162 mV) at the beginning compared to even plain carbon cloth (2.5 A m-2/988 mV). However, after 6 hours these values increased to 3.4 A m-2/271 mV, and eventually stabilized after 24 hours at 7.3 A m-2/334 mV. Such time-dependent improvements were observed over GN-CNT composites with less extreme deviations (~60% jump in peak power during the first 24 hours), suggesting that the lag is mainly caused by CNT addition to the anode.  The performance-enhancing features of CNTs as composite materials have been reported in literature for various types of fuel cells including direct formic acid fuel cells (DFAFC),296 proton exchange membrane fuel cells (PEMFC),297 and to a lesser extent, MFCs.298 Presumably, in addition to its intrinsic high conductivity and easier charge transfer, CNTs possess a unique morphology which increases the electrochemically accessible surface areas for reactions to occur. In the context of MFCs, it has also been suggested this morphology offers a favorable nanostructure environment for bacterial growth and facilitates the external electron transfer from bacterial cells to the electrode.298 Xie et al. reported an anode fabricated from a CNT-textile composite which provided a 10-fold lower charge-transfer resistance compared to an unmodified   119  carbon cloth anode.299 According to the authors, the CNT layer promotes active surface interaction with the biofilm and facilitates electron transfer to the anodes, resulting in high power outputs. Qiao et al. described a CNT/polyaniline composite as anode material for MFCs which delivered a power output of 0.042 W m-2.300 The authors attributed the increased MFC performance to the addition of CNTs to polyaniline, suggesting that it increases the electrode specific area and enhances its charge transfer capabilities.  Nevertheless, to complement the electrochemical experiments, the biological effects of nanomaterials such as CNT must be also taken into account when discussing the MFC performance of such anodes. It was found that single-walled CNTs are, in fact, toxic to the development of E.coli biofilms, which appears to contradict the performance increase reported for MFCs using CNTs.301 The authors of this study proposed that direct contact of E.coli with CNTs damages the cell wall due to the rough morphology of CNTs, thereby releasing the cell’s intracellular materials. The dead cells then provide a protective barrier for live cells against CNTs and the intracellular materials released serve as nutrients that promote biofilm formation above the layer of dead cells (Figure 6-7e). This could perhaps explain the time-dependent improvement among the CNT-containing MFCs as opposed to the graphene anodes. However, more studies are required to find direct evidences for bacteria interactions with various carbon morphologies.   120   Figure 6-7: Electrochemical characterization of GN and CNT composites under the step-sweep rate of 5 mV s-1 starting from the open circuit potential. (a) Polarization, and (b) power curves for composites normalized based on both electrode area and MFC’s empty bed volume. (c) Polarization, and (d) power curves for CNT anodes in the first 24 hours based on both electrode area and MFC’s empty bed volume. (e) Suggested E.Coli growth mechanism on CNT-modified surfaces.301   121  6.5 Summary and conclusions The performance of ultrathin graphene microsheets (GNs), synthesized from the simultaneous anodic and cathodic electrochemical exfoliation process, was investigated as anode materials for microbial fuel cells. Using a membrane-less single-chamber configuration paired with an air-cathode utilizing manganese oxide as an oxygen reduction catalyst, the configuration complexity was reduced while increasing the power output.  Adding mildly functionalized GNs (i.e., anodic exfoliates) to the carbon cloth increased over four times the peak power density compared to the unmodified carbon cloth (2.85 W m-2 vs 0.66 W m-2). Less functionalized, cathodically synthesized GNs also displayed a 250% improvement over carbon cloth. This observation can be attributed to the desirable surface chemistry of mildly-functionalized graphene which enhances bacterial attachment to the electrode and facilitates bacteria-electrode electron transfer. The GNs also outperformed Vulcan XC72 anodes in peak power density by 90% with enhanced charge/mass transfer properties demonstrated by the electrochemical impedance spectroscopy. The CNT-GN hybrid electrodes were also studied to determine the effects of adding different carbon allotropes and morphologies to GN-modified anodes. However, contrary to the expectations and some previous reports, CNT-modified anodes caused dramatic drops in the MFC performance especially on the open circuit voltage (up to 60%) with significant performance lags in the first 24 hours. Based on literature, this was proposed to be the result of CNTs’ anti-bacterial properties toward E.coli biofilm development; which is mainly attributed to the rough morphology of CNTs, where the bacteria membrane can be readily pierced in case of direct contact. However, further biological investigations – out of the scope of this work – are required to explore the roots of the aforementioned observations.   122  Chapter  7: Highlights, conclusions, and recommendations for future work 7.1 Highlights and conclusions In this thesis, a greener electrochemical approach for low temperature, simple, and cost-effective synthesis of graphene microsheets was introduced using graphitic electrodes in ionic liquid medium. It’s greener compared to the pathways via graphene oxide with the elimination of harsh oxidizers/reducers from the process, and the prospects of recycling ionic liquids for repeated synthesis. It’s also more cost-effective with significantly lower power consumption compared to the other previously developed electrochemical methods. Subsequently, the products were successfully used for two target applications: (i) cathode-modifying microporous layers as gas diffusion enhancing materials in proton exchange membrane fuel cells, and (ii) anode-modifying structures in microbial fuel cells. First, a novel IL/acetonitrile electrolyte was developed with dramatically lower loads of ionic liquids (~1:50 IL/acetonitrile vol. ratio). It provided three main advantages of cost efficiency due to low IL content, extended electrochemical stability in a non-aqueous electrolyte, and high exfoliation yield by effective anionic intercalation within the graphitic layers. Using iso-molded graphite rod as the anode, up to 86% of exfoliation was achieved with the majority of the products as graphene flakes in addition to smaller quantities of carbonaceous particles and rolled sheets (<10% in imaging counts). Moreover, in contrast with the previous literature, it was found that the electrolyte coloration during electro-exfoliation in the IL media is related to the occurrence of diverse reactions involving the IL moieties, and cannot be associated with different stages of graphene formation. The cathodically generated species can also interfere with the anionic intercalation in the graphite anode.    123  Secondly, previous research focused mostly on either anode or cathode exfoliation due to restrictions imposed by the investigated intercalating ions, and insufficient consideration given to the design of the electrochemical cell. Consequently, in single graphite electrode studies, at the non-graphitic counter-electrode (e.g., platinum), unwanted electrode reactions such as gas evolution and electrolyte decomposition take place, leading to significant energy and chemical losses. Here, the simultaneous anodic and cathodic GN production was achieved in two types of electrochemical cells (undivided and divided) using aprotic electrolytes containing ionic liquids. A synergistic exfoliation effect was observed when the iso-molded graphite anode and cathode were subjected to a constant cell potential, generating up to 3 times higher exfoliation yields compared to the single-electrode studies on each side (~6-fold improvement in total). Thorough characterization of the products collected from both electrode compartments confirmed the production of ultrathin GNs (< 5 layers). The cathodic exfoliates were almost exclusively composed of GNs; whereas among the anodic products, in addition to the majority GNs, traces of other morphologies were detected such as nanoparticles, nanotubes, and larger rolled sheets. For the first application, the successful integration of cathodically exfoliated graphene (i.e., non-functionalized) in MPLs resulted in enhanced performance over a wide range of operating conditions. GN-based MPLs improved performance in the kinetic and ohmic regions of the polarization curve, while the addition of carbon black to form a composite GN+CB MPL, dramatically extended this improvement to the mass transport limiting region. This was reflected by an approximate 30% and 70% increase in peak power densities compared to CB and GN MPLs, respectively, at relative humidity (RH) of 100%. Despite the presence of CB, GN+CB MPLs also retained their superior performance at a much lower RH of 20%, thereby widening the peak power   124  gap with CB MPLs to 80%. This has major implications for cost reduction in PEMFCs by decreasing the need for humidifiers that are typically used to keep the Nafion membrane hydrated. For the second application, the anodically exfoliated GNs (i.e., more functionalized) sprayed on carbon cloth 3D electrodes generated an over four-fold improvement in peak power density compared with the plain carbon cloth (2.85 W m-2 vs 0.66 W m-2, respectively), exceeding the previously reported values for the graphene anodes. The GNs also outperformed carbon black and carbon nanotube anodes in peak power by an approximate 90% and 200%, respectively. The fuel cell polarization results were corroborated by the electrochemical impedance spectroscopy indicating lower charge transfer resistance for GNs (up to 3 times). It was demonstrated that significant improvements in MFC performance are possible by the combination of novel materials and cell design advancements. A single-chamber membrane-free MFC equipped with a passive manganese oxide air cathode was designed and manufactured in large quantities; which further attributed to the novelty and feasibility of the work. The safety considerations associated with various experimental modules of this thesis are explained in Appendix C. 7.2 Recommendations for future work Although there have been significant breakthroughs in this thesis towards high-yield graphene production techniques, there is always more work to be done and recommendations to be made. First of all, the need for more sophisticated characterization techniques is always present; especially regarding the interaction pathways of ionic electrolytes with graphitic electrodes at the electrified interface. Thus, one could think of advanced in-situ spectroscopic techniques such as in-situ Raman, UV-VIS or FTIR to further investigate the underlying processes behind the electro-exfoliation.    125  Moreover, despite achieving high yields, all the experiments are done in the lab scale to address the more important scientific priorities. However, for the commercial success, further scale up remains to be done for operating such electrochemical systems at larger scales. Strategies such as increasing the cell dimensions, using multiple modular units for synthesis, and changing the operational mode of the system (i.e., from batch to continuous) should be considered to grow the production capacity from gram scale to kilos and ultimately tons per day. Importantly, complementary feasibility studies are necessary to evaluate the financial aspects of the process (e.g., electricity costs per kg of the product, material and process handling fees, etc.). Fine-tuning of the functionalization process is something that could have pronounced implications towards broadening the application of the exfoliated products. In this regard, a wide range of ionic liquids combined with different cosolvents could be employed to customize graphene’s properties for various purposes. As mentioned in the introductory chapters, such materials could be used as oxygen reduction catalysts in fuel cells (e.g., nitrogen-doped graphene); which could replace their expensive platinum counterparts with significant cost reduction.   On the application side, emerging electrochemical power sources such as PEMFCs and MFCs have their own share of challenges and potential avenues for more improvements. In PEMFCs, application of graphene-based MPLs proved to be quite effective in this work. However, one could go one step further studying various composite ratios of graphene with carbon black and its impact on the water management. There are also several other variables that could be optimized when switching to radically new morphologies. This includes binder percentage, total MPL loading on the GDL, and the fuel cell’s operating conditions itself (e.g., gas flow rates, the relative humidity, temperature of the inlets, etc.).    126  As for the MFCs, low power densities along with their relatively high cost are among the persistent issues which were addressed to some degree here. Further enhancing the bacteria-graphene interactions on the anode via various functionalization techniques as well as incorporation of graphene in the cathodic compartment are among the possible recommendations for future work. Additional measurements of the organic matter removal as well as reaction intermediates would lead to better understanding the whole process, and the ways to improve the MFCs’ power output. Ultimately, operating such devices at much larger scales using wastewater while maintaining high efficiencies is key to find compelling real life applications for them.     127  References  (1) Novoselov, K. S. Rev. Mod. Phys. 2011, 83, 837. (2) Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.; Zhang, Y.; Dubonos, S. V.; Grigorieva, I. V.; Firsov, A. A. Science 2004, 306, 666. (3) Brownson, D. A. C.; Kampouris, D. K.; Banks, C. E. J. Power Sources 2011, 196, 4873. (4) Zhu, Y.; Murali, S.; Cai, W.; Li, X.; Suk, J. W.; Potts, J. R.; Ruoff, R. S. Adv. Mater. 2010, 22, 3906. (5) Bolotin, K. I.; Sikes, K. J.; Jiang, Z.; Klima, M.; Fudenberg, G.; Hone, J.; Kim, P.; Stormer, H. L. Solid State Commun. 2008, 146, 351. (6) Lee, C.; Wei, X. D.; Kysar, J. W.; Hone, J. Science 2008, 321, 385. (7) Balandin, A. A.; Ghosh, S.; Bao, W. Z.; Calizo, I.; Teweldebrhan, D.; Miao, F.; Lau, C. N. Nano Lett. 2008, 8, 902. (8) Cai, W.; Zhu, Y.; Li, X.; Piner, R. D.; Ruoff, R. S. Appl. Phys. Lett. 2009, 95, 123115. (9) Li, X.; Zhu, Y.; Cai, W.; Borysiak, M.; Han, B.; Chen, D.; Piner, R. D.; Colombo, L.; Ruoff, R. S. Nano Lett. 2009, 9, 4359. (10) Stankovich, S.; Dikin, D. A.; Dommett, G. H. B.; Kohlhaas, K. M.; Zimney, E. J.; Stach, E. A.; Piner, R. D.; Nguyen, S. T.; Ruoff, R. S. Nature 2006, 442, 282. (11) Stoller, M. D.; Park, S.; Zhu, Y.; An, J.; Ruoff, R. S. Nano Lett. 2008, 8, 3498. (12) Berger, C.; Song, Z.; Li, X.; Wu, X.; Brown, N.; Naud, C.; Mayou, D.; Li, T.; Hass, J.; Marchenkov, A. N.; Conrad, E. H.; First, P. N.; de Heer, W. A. Science 2006, 312, 1191. (13) Schedin, F.; Geim, A. K.; Morozov, S. V.; Hill, E. W.; Blake, P.; Katsnelson, M. I.; Novoselov, K. S. Nat. Mater. 2007, 6, 652. (14) Yoo, E.; Kim, J.; Hosono, E.; Zhou, H.-s.; Kudo, T.; Honma, I. Nano Lett. 2008, 8, 2277. (15) Seger, B.; Kamat, P. V. J. Phys. Chem. C 2009, 113, 7990. (16) Wu, J.; Becerril, H. A.; Bao, Z.; Liu, Z.; Chen, Y.; Peumans, P. Appl. Phys. Lett. 2008, 92, 263302. (17) Ponomarenko, L. A.; Schedin, F.; Katsnelson, M. I.; Yang, R.; Hill, E. W.; Novoselov, K. S.; Geim, A. K. Science 2008, 320, 356. (18) Shao, Y.; Wang, J.; Wu, H.; Liu, J.; Aksay, I. A.; Lin, Y. Electroanalysis 2010, 22, 1027. (19) Geim, A. K.; Novoselov, K. S. Nat. Mater. 2007, 6, 183. (20) Tour, J. M. Chem. Mater. 2014, 26, 163. (21) Biswas, A.; Bayer, I. S.; Biris, A. S.; Wang, T.; Dervishi, E.; Faupel, F. Adv. Colloid Interface Sci. 2012, 170, 2. (22) Chen, D.; Tang, L.; Li, J. Chem. Soc. Rev. 2010, 39, 3157. (23) Thrower, P. A.; Loader, R. T. Carbon 1969, 7, 467. (24) Low, C. T. J.; Walsh, F. C.; Chakrabarti, M. H.; Hashim, M. A.; Hussain, M. A. Carbon 2013, 54, 1. (25) Singh, V. V.; Gupta, G.; Batra, A.; Nigam, A. K.; Boopathi, M.; Gutch, P. K.; Tripathi, B. K.; Srivastava, A.; Samuel, M.; Agarwal, G. S.; Singh, B.; Vijayaraghavan, R. Adv. Funct. Mater. 2012, 22, 2352. (26) Hernandez, Y.; Nicolosi, V.; Lotya, M.; Blighe, F. M.; Sun, Z. Y.; De, S.; McGovern, I. T.; Holland, B.; Byrne, M.; Gun'ko, Y. K.; Boland, J. J.; Niraj, P.; Duesberg, G.; Krishnamurthy, S.; Goodhue, R.; Hutchison, J.; Scardaci, V.; Ferrari, A. C.; Coleman, J. N. Nat. Nanotechnol. 2008, 3, 563. (27) Zhu, C.; Guo, S.; Fang, Y.; Dong, S. ACS Nano 2010, 4, 2429.   128  (28) Li, X.; Zhang, G.; Bai, X.; Sun, X.; Wang, X.; Wang, E.; Dai, H. Nat. Nanotechnol. 2008, 3, 538. (29) Yan, L.; Zheng, Y. B.; Zhao, F.; Li, S.; Gao, X.; Xu, B.; Weiss, P. S.; Zhao, Y. Chem. Soc. Rev. 2012, 41, 97. (30) Liu, J.; Xue, Y. H.; Zhang, M.; Dai, L. M. MRS Bull. 2012, 37, 1265. (31) Reina, A.; Jia, X. T.; Ho, J.; Nezich, D.; Son, H. B.; Bulovic, V.; Dresselhaus, M. S.; Kong, J. Nano Lett. 2009, 9, 3087. (32) Choucair, M.; Thordarson, P.; Stride, J. A. Nat. Nanotechnol. 2009, 4, 30. (33) Zhou, D.; Cheng, Q. Y.; Han, B. H. Carbon 2011, 49, 3920. (34) Hong, N. N.; Yang, W.; Bao, C. L.; Jiang, S. H.; Song, L.; Hu, Y. Mater. Res. Bull. 2012, 47, 4082. (35) Cheng, I. F.; Xie, Y. Q.; Gonzales, R. A.; Brejna, P. R.; Sundararajan, J. P.; Kengne, B. A. F.; Aston, D. E.; McIlroy, D. N.; Foutch, J. D.; Griffiths, P. R. Carbon 2011, 49, 2852. (36) Rutter, G. M.; Crain, J. N.; Guisinger, N. P.; Li, T.; First, P. N.; Stroscio, J. A. Science 2007, 317, 219. (37) Rollings, E.; Gweon, G.-H.; Zhou, S.; Mun, B.; McChesney, J.; Hussain, B.; Fedorov, A.; First, P.; De Heer, W.; Lanzara, A. J. Phys. Chem. Solids 2006, 67, 2172. (38) Singh, V.; Joung, D.; Zhai, L.; Das, S.; Khondaker, S. I.; Seal, S. Prog. Mater Sci. 2011, 56, 1178. (39) Park, S.; Ruoff, R. S. Nat Nano 2009, 4, 217. (40) Allen, M. J.; Tung, V. C.; Kaner, R. B. Chem. Rev. 2010, 110, 132. (41) Nolen, C. M.; Denina, G.; Teweldebrhan, D.; Bhanu, B.; Balandin, A. A. Acs Nano 2011, 5, 914. (42) Wang, Y. Y.; Gao, R. X.; Ni, Z. H.; He, H.; Guo, S. P.; Yang, H. P.; Cong, C. X.; Yu, T. Nanotechnology 2012, 23. (43) Lotya, M.; Hernandez, Y.; King, P. J.; Smith, R. J.; Nicolosi, V.; Karlsson, L. S.; Blighe, F. M.; De, S.; Wang, Z. M.; McGovern, I. T.; Duesberg, G. S.; Coleman, J. N. J. Am. Chem. Soc. 2009, 131, 3611. (44) Bai, H.; Li, C.; Shi, G. Q. Adv. Mater. 2011, 23, 1089. (45) Lu, J.; Yang, J.-x.; Wang, J.; Lim, A.; Wang, S.; Loh, K. P. ACS Nano 2009, 3, 2367. (46) Novoselov, K. S.; Falko, V. I.; Colombo, L.; Gellert, P. R.; Schwab, M. G.; Kim, K. Nature 2012, 490, 192. (47) Wang, J.; Manga, K. K.; Bao, Q.; Loh, K. P. J. Am. Chem. Soc. 2011, 133, 8888. (48) Liu, N.; Luo, F.; Wu, H.; Liu, Y.; Zhang, C.; Chen, J. Adv. Funct. Mater. 2008, 18, 1518. (49) Imperato, G.; König, B.; Chiappe, C. Eur. J. Org. Chem. 2007, 2007, 1049. (50) Jain, N.; Kumar, A.; Chauhan, S.; Chauhan, S. M. S. Tetrahedron 2005, 61, 1015. (51) Liu, H. T.; Liu, Y.; Li, J. H. PCCP 2010, 12, 1685. (52) Rehman, A.; Zeng, X. Q. Acc. Chem. Res. 2012, 45, 1667. (53) Castro Neto, A. H.; Guinea, F.; Peres, N. M. R.; Novoselov, K. S.; Geim, A. K. Rev. Mod. Phys. 2009, 81, 109. (54) Zhang, Y. B.; Small, J. P.; Amori, M. E. S.; Kim, P. Phys. Rev. Lett. 2005, 94. (55) Das, A.; Pisana, S.; Chakraborty, B.; Piscanec, S.; Saha, S. K.; Waghmare, U. V.; Novoselov, K. S.; Krishnamurthy, H. R.; Geim, A. K.; Ferrari, A. C.; Sood, A. K. Nat. Nanotechnol. 2008, 3, 210. (56) Gierz, I.; Riedl, C.; Starke, U.; Ast, C. R.; Kern, K. Nano Lett. 2008, 8, 4603. (57) Hwang, E. H.; Adam, S.; Das Sarma, S. Phys. Rev. B 2007, 76.   129  (58) Park, S.; An, J. H.; Jung, I. W.; Piner, R. D.; An, S. J.; Li, X. S.; Velamakanni, A.; Ruoff, R. S. Nano Lett. 2009, 9, 1593. (59) Stankovich, S.; Dikin, D. A.; Piner, R. D.; Kohlhaas, K. A.; Kleinhammes, A.; Jia, Y.; Wu, Y.; Nguyen, S. T.; Ruoff, R. S. Carbon 2007, 45, 1558. (60) Hummers, W. S.; Offeman, R. E. J. Am. Chem. Soc. 1958, 80, 1339. (61) Marcano, D. C.; Kosynkin, D. V.; Berlin, J. M.; Sinitskii, A.; Sun, Z. Z.; Slesarev, A.; Alemany, L. B.; Lu, W.; Tour, J. M. Acs Nano 2010, 4, 4806. (62) Gómez-Navarro, C.; Weitz, R. T.; Bittner, A. M.; Scolari, M.; Mews, A.; Burghard, M.; Kern, K. Nano Lett. 2007, 7, 3499. (63) Becerril, H. A.; Mao, J.; Liu, Z.; Stoltenberg, R. M.; Bao, Z.; Chen, Y. ACS Nano 2008, 2, 463. (64) Eda, G.; Fanchini, G.; Chhowalla, M. Nat. Nanotechnol. 2008, 3, 270. (65) Lee, C.-G.; Park, S.; Ruoff, R. S.; Dodabalapur, A. Appl. Phys. Lett. 2009, 95, 023304. (66) Bourlinos, A. B.; Gournis, D.; Petridis, D.; Szabó, T.; Szeri, A.; Dékány, I. Langmuir 2003, 19, 6050. (67) Shin, H.-J.; Kim, K. K.; Benayad, A.; Yoon, S.-M.; Park, H. K.; Jung, I.-S.; Jin, M. H.; Jeong, H.-K.; Kim, J. M.; Choi, J.-Y.; Lee, Y. H. Adv. Funct. Mater. 2009, 19, 1987. (68) Stankovich, S.; Piner, R. D.; Chen, X.; Wu, N.; Nguyen, S. T.; Ruoff, R. S. J. Mater. Chem. 2006, 16, 155. (69) Zalan, Z.; Lazar, L.; Fulop, F. Curr. Org. Chem. 2005, 9, 357. (70) Chua, C. K.; Pumera, M. Chem. Soc. Rev. 2014, 43, 291. (71) Pei, S. F.; Cheng, H. M. Carbon 2012, 50, 3210. (72) Wu, Z.-S.; Ren, W.; Gao, L.; Liu, B.; Jiang, C.; Cheng, H.-M. Carbon 2009, 47, 493. (73) Wang, S.; Chia, P.-J.; Chua, L.-L.; Zhao, L.-H.; Png, R.-Q.; Sivaramakrishnan, S.; Zhou, M.; Goh, R. G. S.; Friend, R. H.; Wee, A. T. S.; Ho, P. K. H. Adv. Mater. 2008, 20, 3440. (74) Dubin, S.; Gilje, S.; Wang, K.; Tung, V. C.; Cha, K.; Hall, A. S.; Farrar, J.; Varshneya, R.; Yang, Y.; Kaner, R. B. ACS Nano 2010, 4, 3845. (75) Fan, X.; Peng, W.; Li, Y.; Li, X.; Wang, S.; Zhang, G.; Zhang, F. Adv. Mater. 2008, 20, 4490. (76) Jung, I.; Dikin, D.; Park, S.; Cai, W.; Mielke, S. L.; Ruoff, R. S. J. Phys. Chem. C 2008, 112, 20264. (77) Li, X.; Wang, H.; Robinson, J. T.; Sanchez, H.; Diankov, G.; Dai, H. J. Am. Chem. Soc. 2009, 131, 15939. (78) Vadukumpully, S.; Paul, J.; Valiyaveettil, S. Carbon 2009, 47, 3288. (79) Boukhvalov, D. W.; Katsnelson, M. I. J. Am. Chem. Soc. 2008, 130, 10697. (80) Guo, H.-L.; Wang, X.-F.; Qian, Q.-Y.; Wang, F.-B.; Xia, X.-H. ACS Nano 2009, 3, 2653. (81) Yang, J.; Gunasekaran, S. Carbon 2013, 51, 36. (82) Parvez, K.; Li, R.; Puniredd, S. R.; Hernandez, Y.; Hinkel, F.; Wang, S.; Feng, X.; Mullen, K. ACS Nano 2013, 7, 3598. (83) Alanyalıoğlu, M.; Segura, J. J.; Oró-Solè, J.; Casañ-Pastor, N. Carbon 2012, 50, 142. (84) Jnioui, A.; Metrot, A.; Storck, A. Electrochim. Acta 1982, 27, 1247. (85) Takada, Y.; Fujii, R. 炭素 1985, 1985, 110. (86) Noel, M.; Santhanam, R.; Flora, M. F. J. Appl. Electrochem. 1994, 24, 455. (87) Takenaka, H.; Kawaguchi, M.; Lerner, M.; Bartlett, N. Journal of the Chemical Society-Chemical Communications 1987, 1431.   130  (88) Hathcock, K. W.; Brumfield, J. C.; Goss, C. A.; Irene, E. A.; Murray, R. W. Anal. Chem. 1995, 67, 2201. (89) Toyoda, M.; Shimizu, A.; Iwata, H.; Inagaki, M. Carbon 2001, 39, 1697. (90) Snook, G. A.; Best, A. S. J. Mater. Chem. 2009, 19, 4248. (91) Safavi, A.; Maleki, N.; Tajabadi, F.; Farjami, E. Electrochem. Commun. 2007, 9, 1963. (92) Pringle, J. M.; Ngamna, O.; Lynam, C.; Wallace, G. G.; Forsyth, M.; MacFarlane, D. R. Macromolecules 2007, 40, 2702. (93) Xiao, F.; Zhao, F.; Li, J.; Liu, L.; Zeng, B. Electrochim. Acta 2008, 53, 7781. (94) Wagner, M. R.; Albering, J. H.; Moeller, K. C.; Besenhard, J. O.; Winter, M. Electrochem. Commun. 2005, 7, 947. (95) Abe, T.; Mizutani, Y.; Kawabata, N.; Inaba, M.; Ogumi, Z. Synth. Met. 2001, 125, 249. (96) Cooper, A. J.; Wilson, N. R.; Kinloch, I. A.; Dryfe, R. A. W. Carbon 2014, 66, 340. (97) Kuila, T.; Bose, S.; Mishra, A. K.; Khanra, P.; Kim, N. H.; Lee, J. H. Prog. Mater Sci. 2012, 57, 1061. (98) Georgakilas, V.; Otyepka, M.; Bourlinos, A. B.; Chandra, V.; Kim, N.; Kemp, K. C.; Hobza, P.; Zboril, R.; Kim, K. S. Chem. Rev. 2012, 112, 6156. (99) Hirsch, A.; Englert, J. M.; Hauke, F. Acc. Chem. Res. 2012, 46, 87. (100) Das, T. K.; Prusty, S. Polymer-Plastics Technology and Engineering 2013, 52, 319. (101) Wu, Q.; Xu, Y. X.; Yao, Z. Y.; Liu, A. R.; Shi, G. Q. Acs Nano 2010, 4, 1963. (102) Pan, Y. Z.; Bao, H. Q.; Sahoo, N. G.; Wu, T. F.; Li, L. Adv. Funct. Mater. 2011, 21, 2754. (103) Zhuang, X. D.; Chen, Y.; Liu, G.; Li, P. P.; Zhu, C. X.; Kang, E. T.; Neoh, K. G.; Zhang, B.; Zhu, J. H.; Li, Y. X. Adv. Mater. 2010, 22, 1731. (104) Xia, F. N.; Farmer, D. B.; Lin, Y. M.; Avouris, P. Nano Lett. 2010, 10, 715. (105) Liu, Y.; Yu, D. S.; Zeng, C.; Miao, Z. C.; Dai, L. M. Langmuir 2010, 26, 6158. (106) Kang, S. M.; Park, S.; Kim, D.; Park, S. Y.; Ruoff, R. S.; Lee, H. Adv. Funct. Mater. 2011, 21, 108. (107) Kuila, T.; Khanra, P.; Bose, S.; Kim, N. H.; Ku, B. C.; Moon, B.; Lee, J. H. Nanotechnology 2011, 22. (108) Dreyer, D. R.; Park, S.; Bielawski, C. W.; Ruoff, R. S. Chem. Soc. Rev. 2010, 39, 228. (109) Dikin, D. A.; Stankovich, S.; Zimney, E. J.; Piner, R. D.; Dommett, G. H. B.; Evmenenko, G.; Nguyen, S. T.; Ruoff, R. S. Nature 2007, 448, 457. (110) Paredes, J. I.; Villar-Rodil, S.; Martinez-Alonso, A.; Tascon, J. M. D. Langmuir 2008, 24, 10560. (111) Lerf, A.; He, H. Y.; Forster, M.; Klinowski, J. J. Phys. Chem. B 1998, 102, 4477. (112) Gao, W.; Alemany, L. B.; Ci, L. J.; Ajayan, P. M. Nature Chemistry 2009, 1, 403. (113) Nethravathi, C.; Rajamathi, J. T.; Ravishankar, N.; Shivakumara, C.; Rajamathi, M. Langmuir 2008, 24, 8240. (114) Szabo, T.; Szeri, A.; Dekany, I. Carbon 2005, 43, 87. (115) Szabo, T.; Berkesi, O.; Forgo, P.; Josepovits, K.; Sanakis, Y.; Petridis, D.; Dekany, I. Chem. Mater. 2006, 18, 2740. (116) Fang, M.; Wang, K. G.; Lu, H. B.; Yang, Y. L.; Nutt, S. J. Mater. Chem. 2010, 20, 1982. (117) Kuila, T.; Bose, S.; Hong, C. E.; Uddin, M. E.; Khanra, P.; Kim, N. H.; Lee, J. H. Carbon 2011, 49, 1033. (118) Si, Y.; Samulski, E. T. Nano Lett. 2008, 8, 1679. (119) Bourlinos, A. B.; Gournis, D.; Petridis, D.; Szabo, T.; Szeri, A.; Dekany, I. Langmuir 2003, 19, 6050.   131  (120) Hamilton, C. E.; Lomeda, J. R.; Sun, Z. Z.; Tour, J. M.; Barron, A. R. Nano Lett. 2009, 9, 3460. (121) Matsuo, Y.; Sakai, Y.; Fukutsuka, T.; Sugie, Y. Chem. Lett. 2007, 36, 1050. (122) Taheri Najafabadi, A.; Gyenge, E. Carbon 2014, 71, 58. (123) Nakayama-Ratchford, N.; Bangsaruntip, S.; Sun, X. M.; Welsher, K.; Dai, H. J. J. Am. Chem. Soc. 2007, 129, 2448. (124) Lu, F. S.; Wang, X.; Meziani, M. J.; Cao, L.; Tian, L. L.; Bloodgood, M. A.; Robinson, J.; Sun, Y. P. Langmuir 2010, 26, 7561. (125) Zhao, Y. L.; Stoddart, J. F. Acc. Chem. Res. 2009, 42, 1161. (126) Zheng, X. H.; Zhou, F. F. Journal of X-Ray Science and Technology 2011, 19, 275. (127) Karachevtsev, V. A.; Glamazda, A. Y.; Dettlaff-Weglikowska, U.; Leontiev, V. S.; Peschanskii, A. V.; Plokhotnichenko, A. M.; Stepanian, S. G.; Roth, S. In Spectroscopy of Emerging Materials, 2004; Vol. 165. (128) Chichak, K. S.; Ramirez, R. J. A.; Stoddart, J. F. Abstracts of Papers of the American Chemical Society 2004, 227, U214. (129) Bai, H.; Xu, Y. X.; Zhao, L.; Li, C.; Shi, G. Q. Chem. Commun. 2009, 1667. (130) Xu, Y. X.; Bai, H.; Lu, G. W.; Li, C.; Shi, G. Q. J. Am. Chem. Soc. 2008, 130, 5856. (131) Choi, E. Y.; Han, T. H.; Hong, J. H.; Kim, J. E.; Lee, S. H.; Kim, H. W.; Kim, S. O. J. Mater. Chem. 2010, 20, 1907. (132) Liu, J. Q.; Yang, W. R.; Tao, L.; Li, D.; Boyer, C.; Davis, T. P. "J. Polym. Sci., Part A: Polym. Chem. " 2010, 48, 425. (133) Park, M. J.; Lee, J. K.; Lee, B. S.; Lee, Y. W.; Choi, I. S.; Lee, S. G. Chem. Mater. 2006, 18, 1546. (134) Laaksonen, P.; Kainlauri, M.; Laaksonen, T.; Shchepetov, A.; Jiang, H.; Ahopelto, J.; Linder, M. B. Angew. Chem. Int. Ed. 2010, 49, 4946. (135) Cui, Y.; Kim, S. N.; Jones, S. E.; Wissler, L. L.; Naik, R. R.; McAlpine, M. C. Nano Lett. 2010, 10, 4559. (136) Xu, L. Q.; Yang, W. J.; Neoh, K. G.; Kang, E. T.; Fu, G. D. Macromolecules 2010, 43, 8336. (137) Kong, J.; Franklin, N. R.; Zhou, C. W.; Chapline, M. G.; Peng, S.; Cho, K. J.; Dai, H. J. Science 2000, 287, 622. (138) Chang, H.; Lee, J. D.; Lee, S. M.; Lee, Y. H. Appl. Phys. Lett. 2001, 79, 3863. (139) Bradley, K.; Gabriel, J. C. P.; Briman, M.; Star, A.; Gruner, G. Phys. Rev. Lett. 2003, 91. (140) Zhang, J.; Boyd, A.; Tselev, A.; Paranjape, M.; Barbara, P. Appl. Phys. Lett. 2006, 88. (141) Ristein, J. Science 2006, 313, 1057. (142) Leenaerts, O.; Partoens, B.; Peeters, F. M. Phys. Rev. B 2008, 77, 125416. (143) Lherbier, A.; Blase, X.; Niquet, Y. M.; Triozon, F.; Roche, S. Phys. Rev. Lett. 2008, 101. (144) Park, S.; Ruoff, R. S. Nat. Nanotechnol. 2009, 4, 217. (145) Wang, Z.; Zhou, X.; Zhang, J.; Boey, F.; Zhang, H. J. Phys. Chem. C 2009, 113, 14071. (146) Wang, Y.; Li, Z.; Wang, J.; Li, J.; Lin, Y. Trends Biotechnol. 2011, 29, 205. (147) Zhang, D. D.; Zu, S. Z.; Han, B. H. Carbon 2009, 47, 2993. (148) Robinson, J. T.; Perkins, F. K.; Snow, E. S.; Wei, Z. Q.; Sheehan, P. E. Nano Lett. 2008, 8, 3137. (149) Sundaram, R. S.; Gomez-Navarro, C.; Balasubramanian, K.; Burghard, M.; Kern, K. Adv. Mater. 2008, 20, 3050. (150) Fowler, J. D.; Allen, M. J.; Tung, V. C.; Yang, Y.; Kaner, R. B.; Weiller, B. H. Acs Nano 2009, 3, 301.   132  (151) Lu, G. H.; Ocola, L. E.; Chen, J. H. Appl. Phys. Lett. 2009, 94. (152) Lu, G. H.; Ocola, L. E.; Chen, J. H. Nanotechnology 2009, 20. (153) Mohanty, N.; Berry, V. Nano Lett. 2008, 8, 4469. (154) Wang, Y.; Shao, Y. Y.; Matson, D. W.; Li, J. H.; Lin, Y. H. Acs Nano 2010, 4, 1790. (155) Ang, P. K.; Chen, W.; Wee, A. T. S.; Loh, K. P. J. Am. Chem. Soc. 2008, 130, 14392. (156) Ohno, Y.; Maehashi, K.; Yamashiro, Y.; Matsumoto, K. Nano Lett. 2009, 9, 3318. (157) Lu, C. H.; Yang, H. H.; Zhu, C. L.; Chen, X.; Chen, G. N. Angew. Chem. Int. Ed. 2009, 48, 4785. (158) He, S. J.; Song, B.; Li, D.; Zhu, C. F.; Qi, W. P.; Wen, Y. Q.; Wang, L. H.; Song, S. P.; Fang, H. P.; Fan, C. H. Adv. Funct. Mater. 2010, 20, 453. (159) Xu, M. S.; Fujita, D.; Hanagata, N. Small 2009, 5, 2638. (160) Postma, H. W. C. Nano Lett. 2010, 10, 420. (161) Merchant, C. Biophys. J. 2011, 100, 521. (162) Merchant, C. A.; Healy, K.; Wanunu, M.; Ray, V.; Peterman, N.; Bartel, J.; Fischbein, M. D.; Venta, K.; Luo, Z. T.; Johnson, A. T. C.; Drndic, M. Nano Lett. 2010, 10, 2915. (163) Schneider, G. F.; Kowalczyk, S. W.; Calado, V. E.; Pandraud, G.; Zandbergen, H. W.; Vandersypen, L. M. K.; Dekker, C. Nano Lett. 2010, 10, 3163. (164) Garaj, S.; Hubbard, W.; Reina, A.; Kong, J.; Branton, D.; Golovchenko, J. A. Nature 2010, 467, 190. (165) Chen, W.; Chen, S.; Qi, D. C.; Gao, X. Y.; Wee, A. T. S. J. Am. Chem. Soc. 2007, 129, 10418. (166) Eda, G.; Chhowalla, M. Nano Lett. 2009, 9, 814. (167) Di, C. A.; Wei, D. C.; Yu, G.; Liu, Y. Q.; Guo, Y. L.; Zhu, D. B. Adv. Mater. 2008, 20, 3289. (168) Di, C. A.; Liu, Y. Q.; Yu, G.; Zhu, D. B. Acc. Chem. Res. 2009, 42, 1573. (169) Qu, L. T.; Liu, Y.; Baek, J. B.; Dai, L. M. Acs Nano 2010, 4, 1321. (170) Jeon, I. Y.; Choi, H. J.; Choi, M.; Seo, J. M.; Jung, S. M.; Kim, M. J.; Zhang, S.; Zhang, L. P.; Xia, Z. H.; Dai, L. M.; Park, N.; Baek, J. B. Sci. Rep. 2013, 3. (171) Wassei, J. K.; Cha, K. C.; Tung, V. C.; Yang, Y.; Kaner, R. B. J. Mater. Chem. 2011, 21, 3391. (172) Lu, Z. J.; Bao, S. J.; Gou, Y. T.; Cai, C. J.; Ji, C. C.; Xu, M. W.; Song, J.; Wang, R. Y. RSC Adv. 2013, 3, 3990. (173) Newbury, D. E. Advanced scanning electron microscopy and X-ray microanalysis; Plenum Press: New York, 1986. (174) Joy, D. C.; Romig, A. D.; Goldstein, J. Principles of analytical electron microscopy; Plenum Press: New York, 1986. (175) Nemes-Incze, P.; Osvath, Z.; Kamaras, K.; Biro, L. P. Carbon 2008, 46, 1435. (176) Brundle, C. R.; Evans, C. A.; Wilson, S. Encyclopedia of materials characterization : surfaces, interfaces, thin films; Butterworth-Heinemann ; Manning: Boston Greenwich, CT, 1992. (177) Ohta, T.; Bostwick, A.; Seyller, T.; Horn, K.; Rotenberg, E. Science 2006, 313, 951. (178) Zhou, S. Y.; Siegel, D. A.; Fedorov, A. V.; Lanzara, A. Phys. Rev. Lett. 2008, 101. (179) Bostwick, A.; McChesney, J.; Ohta, T.; Rotenberg, E.; Seyller, T.; Horn, K. Prog. Surf. Sci. 2009, 84, 380. (180) Malard, L.; Pimenta, M.; Dresselhaus, G.; Dresselhaus, M. Phys. Rep. 2009, 473, 51. (181) Jorio, A. Raman spectroscopy in graphene related systems; Wiley-VCH: Weinheim, Germany, 2011.   133  (182) Ferrari, A. C. Solid State Commun. 2007, 143, 47. (183) Pisana, S.; Lazzeri, M.; Casiraghi, C.; Novoselov, K. S.; Geim, A. K.; Ferrari, A. C.; Mauri, F. Nat. Mater. 2007, 6, 198. (184) Das, B.; Voggu, R.; Rout, C. S.; Rao, C. N. R. Chem. Commun. 2008, 5155. (185) Su, Q.; Pang, S. P.; Alijani, V.; Li, C.; Feng, X. L.; Mullen, K. Adv. Mater. 2009, 21, 3191. (186) Dong, X. C.; Fu, D. L.; Fang, W. J.; Shi, Y. M.; Chen, P.; Li, L. J. Small 2009, 5, 1422. (187) Dong, X. C.; Shi, Y. M.; Zhao, Y.; Chen, D. M.; Ye, J.; Yao, Y. G.; Gao, F.; Ni, Z. H.; Yu, T.; Shen, Z. X.; Huang, Y. X.; Chen, P.; Li, L. J. Phys. Rev. Lett. 2009, 102. (188) Panchokarla, L. S.; Subrahmanyam, K. S.; Saha, S. K.; Govindaraj, A.; Krishnamurthy, H. R.; Waghmare, U. V.; Rao, C. N. R. Adv. Mater. 2009, 21, 4726. (189) Cheianov, V. V.; Fal'ko, V. I. Phys. Rev. B 2006, 74. (190) Cheianov, V. V.; Fal'ko, V.; Altshuler, B. L. Science 2007, 315, 1252. (191) Williams, J. R.; DiCarlo, L.; Marcus, C. M. Science 2007, 317, 638. (192) Abanin, D. A.; Levitov, L. S. Science 2007, 317, 641. (193) Young, A. F.; Kim, P. Nat. Phys. 2009, 5, 222. (194) Lohmann, T.; von Klitzing, K.; Smet, J. H. Nano Lett. 2009, 9, 1973. (195) Cain, S.; Cantu, A.; Brunnelle, R.; Lyter, A.; Testing, A. S. f.; Materials; Manager, P.; Sciences, A. A. o. F. Journal of Forensic Sciences 1978, 23, 643. (196) Wight, S. A.; Holbrook, R. D. Journal of Microscopy-Oxford 2005, 220, 65. (197) Reina, A.; Jia, X.; Ho, J.; Nezich, D.; Son, H.; Bulovic, V.; Dresselhaus, M. S.; Kong, J. Nano Lett. 2008, 9, 30. (198) Simon, P.; Gogotsi, Y. Nat. Mater. 2008, 7, 845. (199) Fu, C. P.; Kuang, Y. F.; Huang, Z. Y.; Wang, X.; Yin, Y. F.; Chen, J. H.; Zhou, H. H. J. Solid State Electrochem. 2011, 15, 2581. (200) Li, S.; Lin, Y.; Xie, H.; Zhang, S.; Xu, J. Org. Lett. 2006, 8, 391. (201) Hapiot, P.; Lagrost, C. Chem. Rev. 2008, 108, 2238. (202) Feroci, M.; Chiarotto, I.; Orsini, M.; Sotgiu, G.; Inesi, A. Advanced Synthesis & Catalysis 2008, 350, 1355. (203) Barrosse-Antle, L. E.; Bond, A. M.; Compton, R. G.; O'Mahony, A. M.; Rogers, E. I.; Silvester, D. S. Chemistry – An Asian Journal 2010, 5, 202. (204) Ott, L. S.; Cline, M. L.; Deetlefs, M.; Seddon, K. R.; Finke, R. G. J. Am. Chem. Soc. 2005, 127, 5758. (205) Paul, A.; Mandal, P. K.; Samanta, A. J. Phys. Chem. B 2005, 109, 9148. (206) Paul, A.; Mandal, P. K.; Samanta, A. Chem. Phys. Lett. 2005, 402, 375. (207) Demchenko, A. P. Luminescence 2002, 17, 19. (208) Lakowicz, J. R. In Principles of fluorescence spectroscopy; Springer, 1999. (209) Kroon, M. C.; Buijs, W.; Peters, C. J.; Witkamp, G. J. Green Chem. 2006, 8, 241. (210) Maiti, A.; Kumar, A.; Rogers, R. D. PCCP 2012, 14, 5139. (211) Chen, Y.; Cao, Y.; Lu, X.; Zhao, C.; Yan, C.; Mu, T. New J. Chem. 2013. (212) Chaban, V. V.; Voroshylova, I. V.; Kalugin, O. N.; Prezhdo, O. V. The Journal of Physical Chemistry B 2012, 116, 7719. (213) Meyer, J. C.; Geim, A. K.; Katsnelson, M. I.; Novoselov, K. S.; Booth, T. J.; Roth, S. Nature 2007, 446, 60. (214) Cancado, L. G.; Takai, K.; Enoki, T.; Endo, M.; Kim, Y. A.; Mizusaki, H.; Speziali, N. L.; Jorio, A.; Pimenta, M. A. Carbon 2008, 46, 272.   134  (215) Ferrari, A. C.; Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; Piscanec, S.; Jiang, D.; Novoselov, K. S.; Roth, S.; Geim, A. K. Phys. Rev. Lett. 2006, 97. (216) Pimenta, M. A.; Dresselhaus, G.; Dresselhaus, M. S.; Cancado, L. G.; Jorio, A.; Saito, R. PCCP 2007, 9, 1276. (217) Ciesielski, A.; Samori, P. Chem. Soc. Rev. 2013. (218) Mao, M.; Wang, M.; Hu, J.; Lei, G.; Chen, S.; Liu, H. Chem. Commun. 2013, 49, 5301. (219) Olivier-Bourbigou, H.; Magna, L.; Morvan, D. "Appl. Catal., A " 2010, 373, 1. (220) Etacheri, V.; Marom, R.; Elazari, R.; Salitra, G.; Aurbach, D. Energy Environ. Sci. 2011, 4, 3243. (221) Bernard, G.; Simonet, J. J. Electroanal. Chem. 1979, 96, 249. (222) Simonet, J.; Lund, H. J. Electroanal. Chem. Interfacial Electrochem. 1977, 75, 719. (223) Miller, D. L.; Kubista, K. D.; Rutter, G. M.; Ruan, M.; de Heer, W. A.; First, P. N.; Stroscio, J. A. Phys. Rev. B 2010, 81, 125427. (224) Ferrari, A. C. Solid State Commun. 2007, 143, 47. (225) Sirisaksoontorn, W.; Adenuga, A. A.; Remcho, V. T.; Lerner, M. M. J. Am. Chem. Soc. 2011, 133, 12436. (226) Phaner-Goutorbe, M.; Sartre, A.; Porte, L. Microscopy Microanalysis Microstructures 1994, 5, 283. (227) Lee, J.-S.; Lee, T.; Song, H.-K.; Cho, J.; Kim, B.-S. Energy Environ. Sci. 2011, 4, 4148. (228) Carrette, L.; Friedrich, K.; Stimming, U. Fuel cells 2001, 1, 5. (229) Barbir, F. PEM fuel cells: theory and practice; Academic Press, 2012. (230) Papageorgopoulos, D. Department of Energy, Lakewood, Colorado 2010. (231) Mehta, V.; Cooper, J. S. J. Power Sources 2003, 114, 32. (232) Debe, M. K. Nature 2012, 486, 43. (233) EERE Fuel Cell Technologies Office Multi-Year Research, Development, and Demonstration Plan | Department of Energy; Department of Energy: Washington, US, 2012. (234) Ji, M. B.; Wei, Z. D. Energies 2009, 2, 1057. (235) Li, A.; Chan, S. H. Int. J. Hydrogen Energy 2013, 38, 11988. (236) Weber, A. Z.; Newman, J. J. Electrochem. Soc. 2005, 152, A677. (237) Cindrella, L.; Kannan, A. M. Fuel Cells 2010, 10, 563. (238) Blanco, M.; Wilkinson, D. P. Int. J. Hydrogen Energy 2014, 39, 16390. (239) Antolini, E.; Passos, R. R.; Ticianelli, E. A. J. Power Sources 2002, 109, 477. (240) Hunsom, M.; Piumsomboon, P.; Pruksathorn, K.; Tantavichet, N.; Endoo, S.; Charutavai, K.; Poochinda, K. Renewable Energy 2011, 36, 369. (241) Kannan, A. M.; Munukutla, L. J. Power Sources 2007, 167, 330. (242) Kannan, A. M.; Kanagala, P.; Veedu, V. J. Power Sources 2009, 192, 297. (243) Schweiss, R.; Steeb, M.; Wilde, P. M.; Schubert, T. J. Power Sources 2012, 220, 79. (244) Passalacqua, E.; Squadrito, G.; Lufrano, F.; Patti, A.; Giorgi, L. J. Appl. Electrochem. 2001, 31, 449. (245) Lin, S. Y.; Chang, M. H. Int. J. Hydrogen Energy 2015, 40, 7879. (246) Park, S.; Lee, J.-W.; Popov, B. N. J. Power Sources 2006, 163, 357. (247) Park, S.; Lee, J.-W.; Popov, B. N. J. Power Sources 2008, 177, 457. (248) Marinkas, A.; Arena, F.; Mitzel, J.; Prinz, G. M.; Heinzel, A.; Peinecke, V.; Natter, H. Carbon 2013, 58, 139. (249) Antolini, E. Appl Catal B-Environ 2012, 123, 52. (250) Zhou, X.; Qiao, J.; Yang, L.; Zhang, J. Adv. Energy Mater. 2014, 4.   135  (251) Zamora, H.; Cañizares, P.; Rodrigo, M. A.; Lobato, J. Fuel Cells 2015, 15, 375. (252) Leeuwner, M. J.; Wilkinson, D. P.; Gyenge, E. L. Fuel Cells 2015, 15, 790. (253) Lufrano, F.; Passalacqua, E.; Squadrito, G.; Patti, A.; Giorgi, L. J. Appl. Electrochem. 1999, 29, 445. (254) Gostick, J. T.; Fowler, M. W.; Pritzker, M. D.; Ioannidis, M. A.; Behra, L. M. J. Power Sources 2006, 162, 228. (255) Williams, M. V.; Begg, E.; Bonville, L.; Kunz, H. R.; Fenton, J. M. J. Electrochem. Soc. 2004, 151, A1173. (256) Hou, S.; Liao, S.; Xiong, Z.; Zou, H.; Dang, D.; Zheng, R.; Shu, T.; Liang, Z.; Li, X.; Li, Y. J. Power Sources 2015, 273, 168. (257) Park, C. H.; Lee, C. H.; Guiver, M. D.; Lee, Y. M. Prog. Polym. Sci. 2011, 36, 1443. (258) Logan, B. E.; Hamelers, B.; Rozendal, R.; Schröder, U.; Keller, J.; Freguia, S.; Aelterman, P.; Verstraete, W.; Rabaey, K. Environ. Sci. Technol. 2006, 40, 5181. (259) Zhuang, L.; Zheng, Y.; Zhou, S.; Yuan, Y.; Yuan, H.; Chen, Y. Bioresour. Technol. 2012, 106, 82. (260) Shizas, I.; Bagley, D. J. Energy Eng. 2004, 130, 45. (261) Wang, H. M.; Ren, Z. Y. J. Biotechnol. Adv. 2013, 31, 1796. (262) Rabaey, K.; Verstraete, W. Trends Biotechnol. 2005, 23, 291. (263) Liu, H.; Ramnarayanan, R.; Logan, B. E. Environ. Sci. Technol. 2004, 38, 2281. (264) Bullen, R. A.; Arnot, T. C.; Lakeman, J. B.; Walsh, F. C. Biosens Bioelectron 2006, 21, 2015. (265) Chen, X. F.; Cui, D.; Wang, X. J.; Wang, X. S.; Li, W. S. Biosens Bioelectron 2015, 69, 135. (266) Yates, M. D.; Kiely, P. D.; Call, D. F.; Rismani-Yazdi, H.; Bibby, K.; Peccia, J.; Regan, J. M.; Logan, B. E. Isme J 2012, 6, 2002. (267) Zhao, F.; Harnisch, F.; Schroder, U.; Scholz, F.; Bogdanoff, P.; Herrmann, I. Electrochem. Commun. 2005, 7, 1405. (268) Schroder, U. ChemSusChem 2012, 5, 959. (269) Minteer, S. D.; Atanassov, P.; Luckarift, H. R.; Johnson, G. R. Mater. Today 2012, 15, 166. (270) Candelaria, S. L.; Shao, Y.; Zhou, W.; Li, X.; Xiao, J.; Zhang, J.-G.; Wang, Y.; Liu, J.; Li, J.; Cao, G. Nano Energy 2012, 1, 195. (271) Pinto, A. M.; Goncalves, I. C.; Magalhaes, F. D. Colloid Surface B 2013, 111, 188. (272) Gurunathan, S.; Han, J. W.; Eppakayala, V.; Dayem, A. A.; Kwon, D.-N.; Kim, J.-H. Nanoscale Res. Lett. 2013, 8, 1. (273) Machado, B. F.; Serp, P. Catal. Sci. Technol. 2012, 2, 54. (274) Chen, D.; Zhang, H.; Liu, Y.; Li, J. H. Energy Environ. Sci. 2013, 6, 1362. (275) Wang, Y.; Chang, H. X.; Wu, H. K.; Liu, H. L. J Mater Chem B 2013, 1, 3521. (276) Liu, J.; Qiao, Y.; Guo, C. X.; Lim, S.; Song, H.; Li, C. M. Bioresour. Technol. 2012, 114, 275. (277) Xiao, L.; Damien, J.; Luo, J. Y.; Jang, H. D.; Huang, J. X.; He, Z. J. Power Sources 2012, 208, 187. (278) Zhang, Y. Z.; Mo, G. Q.; Li, X. W.; Zhang, W. D.; Zhang, J. Q.; Ye, J. S.; Huang, X. D.; Yu, C. Z. J. Power Sources 2011, 196, 5402. (279) Hou, J. X.; Liu, Z. L.; Zhang, P. Y. J. Power Sources 2013, 224, 139. (280) Zhao, C.; Wang, Y.; Shi, F.; Zhang, J.; Zhu, J.-J. Chem. Commun. 2013, 49, 6668. (281) Tang, J. H.; Chen, S. S.; Yuan, Y.; Cai, X. X.; Zhou, S. G. Biosens Bioelectron 2015, 71, 387. (282) Yu, D.; Goh, K.; Wang, H.; Wei, L.; Jiang, W.; Zhang, Q.; Dai, L.; Chen, Y. Nat Nano 2014, 9, 555.   136  (283) Rismani-Yazdi, H.; Carver, S. M.; Christy, A. D.; Tuovinen, I. H. J. Power Sources 2008, 180, 683. (284) Li, X.; Hu, B. X.; Suib, S.; Lei, Y.; Li, B. K. J. Power Sources 2010, 195, 2586. (285) Qiao, Y.; Li, C. M.; Bao, S.-J.; Lu, Z.; Hong, Y. Chem. Commun. 2008, 1290. (286) Zhang, T.; Zeng, Y.; Chen, S.; Ai, X.; Yang, H. Electrochem. Commun. 2007, 9, 349. (287) Zhang, F.; Merrill, M. D.; Tokash, J. C.; Saito, T.; Cheng, S.; Hickner, M. A.; Logan, B. E. J. Power Sources 2011, 196, 1097. (288) He, Z.; Mansfeld, F. Energy Environ. Sci. 2009, 2, 215. (289) Guo, W.; Cui, Y.; Song, H.; Sun, J. Bioprocess Biosyst Eng 2014, 37, 1749. (290) Yong, Y.-C.; Dong, X.-C.; Chan-Park, M. B.; Song, H.; Chen, P. ACS nano 2012, 6, 2394. (291) Zhao, C. e.; Wang, W. J.; Sun, D.; Wang, X.; Zhang, J. R.; Zhu, J. J. Chemistry–A European Journal 2014, 20, 7091. (292) Mehdinia, A.; Ziaei, E.; Jabbari, A. Int. J. Hydrogen Energy 2014, 39, 10724. (293) Santoro, C.; Babanova, S.; Artyushkova, K.; Cornejo, J. A.; Ista, L.; Bretschger, O.; Marsili, E.; Atanassov, P.; Schuler, A. J. Bioelectrochemistry 2015, 106, Part A, 141. (294) Cheng, S.; Logan, B. E. Electrochem. Commun. 2007, 9, 492. (295) Pinto, A. M.; Gonçalves, C.; Sousa, D.; Ferreira, A. R.; Moreira, J. A.; Gonçalves, I. C.; Magalhães, F. D. Carbon 2015. (296) Selvaraj, V.; Alagar, M.; Kumar, K. S. "Appl. Catal., B " 2007, 75, 129. (297) Andersen, S. M.; Borghei, M.; Lund, P.; Elina, Y.-R.; Pasanen, A.; Kauppinen, E.; Ruiz, V.; Kauranen, P.; Skou, E. M. Solid State Ionics 2013, 231, 94. (298) Park, I. H.; Christy, M.; Kim, P.; Nahm, K. S. Biosens. Bioelectron. 2014, 58, 75. (299) Xie, X.; Hu, L.; Pasta, M.; Wells, G. F.; Kong, D.; Criddle, C. S.; Cui, Y. Nano Lett. 2011, 11, 291. (300) Qiao, Y.; Li, C. M.; Bao, S. J.; Bao, Q. L. J. Power Sources 2007, 170, 79. (301) Rodrigues, D. F.; Elimelech, M. Environ. Sci. Technol. 2010, 44, 4583.      137  Appendices Appendix A  Electrosynthesis protocols A.1 Undivided exfoliation experiments  Figure A-1: Key dimensions and electrochemical conditions for the graphite exfoliation experiments in undivided cells (glassware specifications from BioLogic Science Instruments).  Figure A-2: Post-treatment protocols followed for further processing of the exfoliates obtained from the undivided setup.   138  A.2 Divided exfoliation experiments  Figure A-3: Key dimensions and electrochemical conditions for the graphite exfoliation experiments in divided H-cells (glassware specifications from Pine Research Instrumentation).  Figure A-4: Post-treatment protocols followed for further processing of the exfoliates obtained from the divided setup.     139  Appendix B  ORR experiments B.1 Rotating disk electrode measurements The appropriate amount of catalyst was sonicated in 280 μL of DI water, and 70 μL of isopropanol for 15 min, followed by the addition of 100 μL of 5 wt% Nafion solution (Alfa Aesar), and sonication for another 15 min to make a well-mixed suspension. 7 μL of the ink was then placed on the glassy carbon (GC) disk electrode, and air dried at room temperature to achieve a catalyst loading of 0.66 mg cm−2 (dry weight catalyst:Nafion ratio was 10:4.4). Catalyst here refers to the electrochemically synthesized GNs including both cathodic exfoliates (i.e., undoped or nonfunctionalized), and anodic products (i.e., doped or functionalized). Cyclic voltammetry (CV) measurements were performed using a potentiostat system (Princeton Applied Research) with a standard three-electrode setup at room temperature. A mercury/mercury oxide (MOE) electrode was employed as the reference electrode, with a platinum wire as the counter electrode. The catalyst-loaded glassy carbon was used as the rotating disk working electrode, and the measurements were mainly conducted in 0.1 M KOH. CVs were first performed in nitrogen saturated electrolytes to characterize the redox behavior and capacitance of the GN catalysts. Subsequent CVs were done in an O2-sparged electrolyte with an O2 blanket above the solution to maintain O2 saturation. The scan rate for all the experiments was maintained at 5 mV s-1, and the working electrode was rotated with angular velocities varying from 100 to 1600 rpm. Koutecky-Levich plots (i-1 vs ω-0.5) were constructed using the RDE results at different electrode potentials for each rotation speed based on Equation B-1; where i is the measured current density, iL and iK are the mass transfer and kinetic limiting current densities, respectively, and ω is the rotation speed in radians per second. The slope of the linear regression lines was used   140  to obtain the number of electrons transferred (n), based on the Equation B-2; where F is Faraday’s constant, D is the diffusion coefficient of O2, C is the bulk concentration of O2 in the electrolyte, and v is the kinematic viscosity of the electrolyte. The intercept of the Koutecky-Levich plot allows for the determination of the purely kinetic (Tafel) current density (iK). A semi-logarithmic Tafel plot was constructed by plotting iK against potential to obtain the Tafel slope. 1𝑖=1𝑖𝐿+1𝑖𝐾=1𝐵𝜔0.5+1𝑖𝐾 Equation B-1 𝐵 = 0.620𝑛𝐹𝐷2/3𝑣−1/6𝐶 Equation B-2 Oxygen reduction reaction in alkaline media is described by Equation B-3, and the Nernst equation was used to determine the thermodynamic equilibrium potential for ORR in 0.1 M KOH at room temperature. The value of 𝐸𝑒𝑞0  obtained from the Nernst equation with respect to SHE (standard hydrogen electrode) was converted to potential versus MOE using the correction term of +0.926 V. 𝑂2 + 2𝐻2𝑂 + 4𝑒−  → 4𝑂𝐻− Equation B-3 B.2 ORR results The electrochemical reduction of O2 is a multi-electron reaction with two main possible pathways; (i) the transfer of two electrons resulting in the formation of H2O2, and (ii) the four-electron pathway to produce water. To achieve maximum efficiency in fuel cells, it is highly desirable for the oxygen reduction reaction to follow the four-electron pathway. Typically, carbon materials can only reduce oxygen through the two-electron pathway. However, it has been shown that doped carbon materials can facilitate more electrons in the oxygen reduction process. As mentioned in Chapter 5, RDE studies showed no catalytic activity towards ORR in acidic media. However, some of the doped samples especially those containing nitrogen and sulfur   141  (Figure 4-8a) showed some improvements in the alkaline media (0.1 M KOH). Thus, their performance was compared to VXC-72 and 20 wt% Pt on Vulcan. To do so, Matlab codes were written in order to analyze the results in a more systematic way. Figure B-1 shows a sample analysis made for the Vulcan results starting with nitrogen response subtraction from ORR results all the way to the calculation of number of electrons over a voltage range, and Tafel slope investigation. Figure B-2 presents the side-by-side ORR behavior comparison between graphene samples, pure Vulcan, and Vulcan-supported Pt catalysts. Doped graphene materials containing 5% N and 4% S (shown in red) had significant improvement in overpotential, and the number of  Figure B-1: Sample ORR analysis results for Vulcan XC-72 in 0.1 M KOH solutions with the sweep rate of 5 mV s-1 at 293 K.   142  electrons compared to the nonfunctionalized cathodically synthesized graphene (shown in blue). Further studies could be done to lower the overpotential, and increase the number of electrons via fine-tuning of the anodic doping process.    Figure B-2: Performance comparison between various graphene products in this study and the Vulcan-based samples in 0.1 M KOH solutions with the sweep rate of 5 mV s-1 at 293 K.   143  Appendix C  MFC design & protocols   Figure C-1: Single-chamber MFC design using Solidworks with exact specifications.   Figure C-2: Bacteria growth protocol.   144    Figure C-3: Phosphate buffer solution preparation protocol.   Figure C-4: Tryticase soy agar/broth bacteria culture medium preparation protocol.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0315348/manifest

Comment

Related Items