Ion Dynamics at Carbon-Grafted-Polypyrrole Electrode–Electrolyte

Jan 5, 2018 - Figure S4 shows several snap shots of PPy structure taken from the software used in these calculations (0 corresponds to initial time). ...
0 downloads 8 Views 4MB Size
Subscriber access provided by READING UNIV

Article

Ion Dynamics at Carbon-Grafted-Polypyrrole Electrode – Electrolyte Interfaces: Study on Charge Carrier Mobility and Ion Co-adsorption in Liquid and Hydrogel Electrolytes by Electrochemical, Gravimetric and Computational Methods Mariusz Radtke, Markus Rettenmayr, and Anna Ignaszak J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.7b06112 • Publication Date (Web): 05 Jan 2018 Downloaded from http://pubs.acs.org on January 10, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Ion Dynamics at Carbon-Grafted-Polypyrrole Electrode – Electrolyte Interfaces: Study on Charge Carrier Mobility and Ion Co-adsorption in Liquid and Hydrogel Electrolytes by Electrochemical, Gravimetric and Computational Methods Mariusz Radtke1, Markus Rettenmayr2, Anna Ignaszak1, a 1

University of New Brunswick, Department of Chemistry, 30 Dineen Drive, E3B5A3,

Fredericton, NB, Canada 2

Friedrich-Schiller-Universität Jena, Otto Schott Institute of Materials Research, Löbdergraben

32, 07743 Jena, Germany a

Corresponding author: Anna Ignaszak, E-mail: [email protected]

Abstract We analyzed ionic processes within polypyrrole (PPy) covalently attached to graphene, multiwalled carbon nanotubes (MWCNTs), and single-walled carbon nanohorns (SWCNHs), operating in aqueous and hydrogel electrolytes that contain KCl, NaCl, LiCl, KBr, NaBr and LiBr salts. Computational studies were carried out for a five-membered PPy chain with a single charged pyrrole using molecular mechanics (MMFFaq) and density functional theory (DFT B3LYP/ 6-31G). The analysis revealed that cation co-adsorption (and corresponding doping of anions) occurs during oxidation of the polymer for potassium and sodium halides, but not for lithium salts. Electro-gravimetric analysis carried out by an electrochemical quartz crystal microbalance confirmed the cation co-adsorption on the oxidized polymer in KCl. The highest ion mobility measured for several carbon-grafted PPy composites was found for MWCNTgrafted-PPy in liquid electrolytes. The ionic mobilities for the same materials embedded in a hydrogel electrolyte were comparable, except for graphene-based electrodes (one order of 1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 54

magnitude lower than that of MWCNT- and SWCNH-grafted-PPy). Ionic mobility is a potential dependent entity. The highest values were observed for KCl in the range of redox activity of PPy, i.e. from 0.522 to 0.922 V vs. standard hydrogen electrode (SHE). It significantly dropped at higher potentials due to saturation of PPy (maximum doping). All bromides showed considerably higher mobility as compared to chlorides at polarizations above 0.922 V vs. SHE. This phenomenon can be related to additional electrochemical activity of bromides, leading to the formation of Br3- under positive potentials. Cyclic voltammetry revealed the highest gravimetric capacitance for the MWCNT-grafted-PPy incorporated in hydrogels and synthesized via oxidative radical (516.86 F g-1) or electrochemically aided atom transfer radical polymerization, e-ATRP (456.31 F g-1). The capacitances of hydrogel-embedded composites were significantly higher than those of the same electrodes measured in liquid electrolytes (350.49 F g-1 after oxidative radical polymerization, and 338.43 F g-1 for e-ATRP). Introduction Intervals of overproduction of energy, especially generated by renewable technologies (solar, hydro or wind power) become reality in many countries. The implementation of clean energy in power grids increases continuously, and efficient storage of energy (or the momentary excess of energy) becomes increasingly important.1 The phenomenon called the “California Duck” power production/demand imbalance2 demonstrates the importance of energy storage systems in energy power plants and its impact on efficient energy management. Energy disproportion describes the time intervals when power production exceeds its consumption or demand, and the energy consumption cannot be deferred or delayed. In order to provide sufficient electricity for temporary needs, a viable solution is to incorporate energy storage systems (e.g., capacitors) in a power grid or in a smart grid communication infrastructure that allows demand response. The 2 ACS Paragon Plus Environment

Page 3 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

California Independent System Operation (CAISO) has predicted that energy imbalance will be one of the most challenging tasks for current smart grids to overcome in upcoming years.3 Super- and ultra-capacitors that allow rapid uptake/release of considerable amounts of energy (in milliseconds) are promising devices that can be incorporated in general power grids.4 The type of materials, their structure and the processing technology used to fabricate these devices are vital to the success of capacitor technology. Not only fast charge uptake and release, but also longterm stability (both chemical and electrochemical), and an enlarged electrochemical surface are required features for capacitive materials.5,6 The improvement of gravimetric capacitance is a subject of many studies. It deals with optimization of morphology (expansion of specific surface area of a double-layer capacitive electrode) or the molecular structure that allows redox processes in pseudo-capacitors.7,8 Since the energy storage capacity for both pseudo-capacitors and double-layer capacitors relies explicitly on the rate of ion exchange, the connection between electrode performance and chemistry of an operating electrolyte, irrespectively whether it is a solid or liquid, is still under optimization.9,10 An effective electrode/electrolyte interface and a high reactivity of electrode components (especially for pseudo-capacitors) are necessary features for an active electrode material. Not only a long-term stability, but also the toxicity of both electrode and electrolyte are a challenge. As an example, electrodes made of heavy metals11 are often replaced with so-called synthetic metals based on conjugated polymers (e.g., polypyrrole, polyaniline) that have shown quite promising specific capacitance.12 However, their electrochemical stability is still below of that of their metallic counterparts. Nanostructured carbons are not only used in capacitors, but also in other energy storage and conversion devices. Multi-walled carbon nanotubes (MWCNTs), graphene, activated carbons, carbon blacks, or recently single-walled carbon nanohorns (SWCNHs) have been extensively 3 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 54

studied, and their roles in energy systems are quite well understood.13,14 For example, an important feature is their ability for fast ion exchange on large and easily polarized π-extended surfaces (especially at highly graphitized surfaces in MWCNTs, graphene, SWCNHs). This is essential for rapid charge/discharge, which is a key feature for electrochemical double layer capacitors (EDLCs).15 Pairing EDLCs with various redox active materials such as oxides (MnO2 or RuO2) results in additional improvement of charge accumulation,16 leading to a specific capacitance as high as 1000 F g-1.17 Although the initial capacitance is remarkable, excessive reduction and oxidation cycles lead to degradation and etching of these materials, which is a serious drawback for possible industrial applications.18 Another way of improving the EDLCs’ performance is to combine them with redox active conjugated polymers such as poly(3,4ethylenedioxythiophene), PPy or polyaniline. Even though these combined electrode materials initially show high specific capacitances, repeated and extensive charge/discharge generates volumetric changes in the organic fraction, resulting in linearization of the polymer chain, and considerable mechanical and electrochemical instabilities of the electrode.19,20 One way of improving its stability is to introduce chemical bonds between the nanostructured carbon and conjugated polymer. Rigid, conductive and chemically stable linkers can generate strong electronic effects (e. g., reverse electron donor/acceptor interactions), leading to more durable and active electrodes as compared to physical mixtures of the same carbon and polymer or metal oxide.21,22 The electrochemical double-layer and pseudo-capacitance in carbons chemically grafted with conjugated polymers is strongly influenced by the type of electrolyte used in these devices. The effective electrode/electrolyte interface is vital for generating a maximum power density in the capacitor.23 All aspects of chemical and physical properties as well as the molecular structure of

4 ACS Paragon Plus Environment

Page 5 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

electrode and electrolyte materials influence the ion diffusion in the system. Thus, the ion mobility is critical for efficient ion doping in the redox active fraction (pseudo-capacitance of conjugated polymer), and in creating the electrochemical double-layer at the carbon/electrolyte junction. The functions of liquid electrolytes in the electrode/electrolyte interactions are relatively well understood. For semi-liquid hydrogels that recently have become popular as ionic matrix for wearable, stretchable and leakage-free capacitors, the ion dynamics at the electrode/electrolyte interface is still unexplored. It has been already proven using both computational and experimental methods that the mass transport in hydrogel electrolytes proceeded via a vehicular mechanism due to the presence of water percolation channels, which are responsible for the mobility of solvated ions.24 In this work we analyse the ion mobility for various common electrolytes in aqueous solution and in hydrogels, and correlate the mobility with the gravimetric capacitance of carbon-graftedPPy electrodes. The main objective of this analysis is to understand the impact of ionic radius on the rate of charge exchange. This is correlated with gravimetric capacitance, especially for new formulations of a hydrogel electrolyte consisting of a polyacrylate matrix modified with sulfonated tetrafluoroethylene (Nafion). Electrochemical impedance spectroscopy combined with an electrochemical quartz crystal microbalance are employed to study the rate of diffusion (defined as the transition time needed for ions to reach the pseudo-capacitive active centers). The electronic states of the conjugated polymer (and therefore its pseudo-capacitance) strongly depend on the polarization.25 Thus, potential-dependent impedance measurements will allow us to analyze polymer electronic transitions, and related to this, ion doping.26 The quantity of adsorbed ions at the polarized electrodes using the electrochemical quartz crystal microbalance is correlated with their specific capacitance. The experimental results are further corroborated by

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 54

qualitative computations using two theoretical methods based on energy minimization in aqueous phase. The first is a molecular mechanics model MMFFaq that is suitable for the conformational analysis of large and flexible molecules. This analysis includes aqueous solvent correction to energy data. The applied method provides a preliminary estimation of the minimized energy after ion adsorption in PPy, and serves as a starting parameter in computations, resulting in further minimized energy values, carried out using the DFT B3LYP 6-31 G/PCM model. The synthesis procedures and material characterization of the carbon-grafted-PPy studied here are presented in our previous work.22,27 Briefly, methods of pairing the carbon (graphene, MWCNT, SWCNH) with the polymer via chemical bonds were optimized in order to supress homopolymerization of pyrrole and to create porous composites. Based on previous studies, the best performing capacitor materials were obtained by oxidative radical polymerization, electrochemically aided atom transfer radical polymerization (e-ATRP) and reversible additionfragmentation chain transfer polymerization (RAFT). The structures of capacitors are shown in Figure 1C and S1 (supporting information). Materials and experimental methods 2-propanol for molecular biology, BioReagent, C99.5%, acrylamide (AAm) >99 % HPLC purity, lithium bromide, ReagentPlus®, ≥99%, anhydrous free-flowing lithium chloride, Redi-Dri™, ACS reagent, ≥99%, multi-walled carbon nanotubes, O.D.9L 6-9 nm, 95% carbon BET 220 m2 g-1, N,N-methylenebisacrylamide (MBA) powder for electrophoresis, ≥99.5%, Nafion 117®, ~5% in a mixture of lower aliphatic alcohols and water, poly(vinylidene difluoride) with an average Mw = 180.000, potassium bromide FT-IR grade, ≥99% trace metals basis, potassium chloride

6 ACS Paragon Plus Environment

Page 7 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

analytical grade, p.a. 99.5%, single walled carbon nanohorns, as-grown, BET 400 m2 g-1, sodium bromide, BioXtra, ≥99.0%, sodium chloride, BioXtra, ≥99.5%, were obtained from Sigma Aldrich, Canada. Hydrochloric acid (37%) was supplied by VWR Canada and carbon black, Super P Conductive, 99%, was purchased from Alfa Aesar. Transmission electron microscopy (TEM) imaging was performed using a Jeol 2100s operating at 200 kV. All electrochemical experiments were carried out on a CH Instruments electrochemical workstation model C760 and with an electrochemical cell containing a platinum wire counter electrode (CE), an Ag/AgCl reference electrode (RE), and an ink-deposited glassy carbon rotating disk electrode (RDE) with 0.1963 cm2 area used as a working electrode (WE). The WE was mounted on a Pine AFMSRCE rotating disk electrode station. All potentials in the 3-electrode system are quoted vs. Ag/AgCl (222 mV vs. SHE). All electrochemical measurements were carried out in 0.1 M LiCl, LiBr, NaCl, NaBr, KCl, and KBr electrolytes, respectively, purged with N2 for 30 min prior to experiments (N2 gas blanket was kept during the measurement). The glassy carbon disk (PINE Instrument Company, USA) was mechanically polished with 0.05 µm Al2O3 slurry (Cypress Systems Inc., USA), rinsed in double-distilled water, sonicated for 5 min, finally rinsed in ultrapure isopropanol and ethanol and dried in air stream. The ink was prepared by dissolving PVDF (2 mg) in isopropanol (2 mL), sonicated for 0.5 h together with carbon black (2 mg) and 8.5 mg of active material (carbon-grafted-PPy). The resulting ink was dropped (17 µL overall with 8.5 µL for each cast) onto the polished glassy carbon WE and dried for 2 min under a 60 W lamp. The CV scans were conducted in the potential range from 1.022 to -0.578 V vs. SHE. Chronocoulometry was carried out at 0.522 V, at the onset potential of polypyrrole oxidation. Electrochemical Quartz Crystal Microbalance (EQCM) analysis was performed using the CHI Instruments 400C series time-resolved EQCM

7 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 54

with a customized glassy carbon resonator electrode, Ag/AgCl reference and Pt wire as a counter electrode. Electrochemical impedance spectroscopy was conducted in the frequency range from 106 to 0.3 Hz and a potential amplitude of 0.005 V, using electrode polarization ranging from 0.522 to 1.222 V. The impedance spectra were analyzed and fitted with the software ZView (Scribner Associates Inc., US). The same electrochemical tests were done for the hydrogel-embedded active material (carbongrafted-PPy) in a two-electrode system. For this, a laminate (20 mm square of 1.3 cm thick film) was placed between two stainless steel square plates (20 mm) that were connected to the electrochemical station. The two-electrode cells were enclosed in order to prevent evaporation of water from the hydrogel electrolyte (Figure S2). The same electrochemical techniques and settings were applied to the three- and two-electrode cells. Synthesis All electrode materials were synthesized and characterized according to procedures reported previously.21,22 Phase identification, surface analysis, morphology characterization, elemental mapping

and

characterization

of

thermal

properties

were

carried

out

by

FTIR/Raman/TEM/XPS/EDX and TGA, respectively, and are described in previous work. The hydrogel electrolyte was fabricated as follows: 50 mL of deionized water was purged in a Schlenk line with a dry nitrogen for 30 minutes, 0.0256 g of ammonium persulfate (APS) was dissolved in 1 mL of deionized water in a separate flask. Furthermore, 1.56 g AAm, 1.18 g potassium chloride, 0.5 mL of 5wt% Nafion 117® solution, 0.05 mL tetramethylethylenediamine (TEMED), and 0.01g of active material were mixed with 9 ml of deionized and deoxygenated water. All procedures were carried out in a N2-filled glove bag. The mixture was sonicated for 30 minutes to ensure dissolution of KCl. Then it was combined with the ammonium persulfate 8 ACS Paragon Plus Environment

Page 9 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

solution under vigorous stirring, and poured onto the Petri dish and placed in the UV cross-linker operating at 254 nm wavelength with irradiation of 300.000 µJ cm-2 for 20 min. Theoretical methods The minimum energy of adsorption of ions in the electrolyte was computed only for the polymer capacitor, since ionic adsorption on the carbon double-layer capacitor has been studied previously.28 The base for the molecular mechanistic force field energy minimization model (MMFFaq) was a simplified 5-membered PPy chain, carrying an immonium radical cation active center on third pyrrole unit. The evaluation of non-bonding, van der Waals and Coulombic interactions using MMFFaq provides primary conformational preferences of ion adsorption on the charged polymer. In next step, the initial energy minimization calculated by molecular mechanistics was used as a starting parameter in the DFT model. For this, a hybrid exchange energy correlation was applied to the grid-based density functional theory model, DFT B3LYP (Becke 3-parameters Lee-Yang-Parr were applied to the local density approximation). Also, 631G with a restricted open shell wave function and a POPN31 exponent were included in the base set of approximations. Quantum annealing was chosen as an algorithm to find the global minimum of an objective function. Spin multiplicity was assumed to be unity, and the net charge (due to the presence of unbalanced alkali metal cations from the electrolyte) to be +1. A maximum self-consistent field iteration was set to 1000 with a Hartree/Bohr convergence tolerance of 0.001000. Also, a Polarizable Continuum Model (PCM) without compensation, and three isotropic dielectrics in integral equations were applied to model solvation effects. The Hückel guess and C1 symmetry were used for all computations. Orbitals were optimized by a direct second order self-consistent field calculation. These models were applied to six salts used in the electrolyte, particularly LiCl, LiBr, NaCl, NaBr, KCl, and KBr. The initial minimum

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 54

energy computed using molecular mechanics applied to the DFT model were analyzed using the ORCA 3_0_3 and GAMESS-64-2016-intel-linux-mkl software. Furthermore, the energy level of the Highest Occupied Molecular Orbital (HOMO), corresponding to the immonium radical cation/halide center, and interaction between the PPy chain and alkali metal (attraction or repulsion) were projected with the Avogadro 1.1.1 software. The results were finally exported in a format compatible with graphical software and presented in Figure 2. The effect of solvent concentration was not included in these computations. Results and Discussion Pseudo- and double-layer capacitance in carbon-grafted-polypyrrole The role of electrolyte in the redox activity of conjugated polymers is critical for the effective pseudo-capacitance and for the rate of charge/discharge.29 Stripping of electrons from the pyrrole lone pair during electrochemical oxidation generates a positively charged radical cation that is neutralized by doping with counter anions from the operating electrolyte (Figure 1A). Upon reduction, the positively charged radical cation returns to its neutral state, and the doping ion is expelled to solution, resulting in electrode discharge. Both doping (charge) and un-doping (discharge) are very fast. This characteristic of pseudo-capacitors makes them highly sought in capacitor technologies.30 The rate of doping varies significantly for different electrolytes and strongly depends on the type of doping anion.31 The trend is fairly straightforward, preferentially the doping anion with the smallest ionic radius charges fastest. The function of cation in combined capacitor electrodes is primarily to contribute to the charging of carbon.32 Doublelayer formation with a positive charge from the electrolyte is particularly favoured in nanostructured carbons consisting of a large π-extended surface.33 Figure 1B shows a cyclic voltammogram of carbon-grafted-PPy, with features that are typical for the double-layer 10 ACS Paragon Plus Environment

Page 11 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

capacitance (cation adsorption on carbon, and negative polarization or anion adsorption on positively polarized carbon). This feature is demonstrated by a square shape of voltammogram. In addition, the pseudo-capacitance due to the redox activity of polymer and associated anion doping is represented by corresponding peaks in the potential range between -0.178 and 0.422 V (anodic scan). The total capacitance of such composite electrodes is usually higher than that of carbon or PPy alone.46 It is believed that one of the reasons for this improvement are electronic effects between the extended π system of the graphitized carbon and π electrons at the conjugated double-bonds in PPy. The stacking of orbitals can potentially increase the charge density at the carbon/polymer junctions, resulting in stronger attraction of the counter-ion from electrolyte. Although the improvement of capacitance in combined carbon-PPy electrodes is acknowledged, the origin of this phenomenon is still under investigations. In the following section, we compute the possible energy states of the PPy with respect to its interaction with doping ions and the associated co-adsorption phenomena.

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 54

Figure 1. Pseudo-capacitance generated by the redox activity of polypyrrole (A), cyclic voltammogram of the MWCNT-grafted-PPy composite (prepared by oxidative radical polymerization) in 0.1 M KCl solution (B), structures of carbon, molecular linker, and polymer (C). Adsorption of specific ions on PPy: theoretical predictions The computational methods are employed to predict the minimum energy of adsorption for various ions in electrolyte, taking into account the size of ions and their interaction with the simplified five-membered PPy chain. The charged state of oxidized PPy (due to electron stripping from a nitrogen lone pair) was mimicked by generating and immonium radical cation on third pyrrole unit in the five-membered fragment. This polymer structure was assumed based

12 ACS Paragon Plus Environment

Page 13 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

on previous theoretical studies, which revealed that the oxidation of PPy results in the formation of a radical cation at every third monomer unit.34 Regarding the choice of electrolyte, for this computation we selected the most common electrolytes in capacitor devices, which are alkali metal salts. Initially, the energy was minimized with the molecular mechanics force fieldaq approximation that allows us to establish an initial conformational structure of the charged PPy, taking into account its van der Waals and Coulombic interactions with the alkali metal halide. The approximate minimized energy obtained by MMFFaq was the starting value for more sophisticated DFT calculations using the B3LYP 6-31G method and the PCM water solvent model, resulting in more accurate estimate of minimized energy. Based on calculated energies, the highest occupied molecular orbital and interactions between the immonium radical cation and doping counter anions (chloride or bromide) are plotted in Figure 2 (values of minimized energies are also shown in Figure 2). The computations revealed that upon charging, leading to the formation of delocalized polaron in PPy, the electronic structure of the first adjacent pyrrole unit changes considerably. The modification of this monomer leads to a stronger local electric field around the polaron. It is attracted to positive charges, leading to the co-adsorption of cation on the PPy surface (Figure 2E). Among all analyzed salts, the strongest cation co-adsorption was found for KCl (Figure 2E), and the opposite effect, i.e. the repulsion between the lithium cation and pyrrole unit, was observed for all Li salts (Figure 2A, 2B). Doping with an anion associated with the formation of immonium radical cation during oxidation of PPy was not identified for lithium bromide. This indicates that neither pseudo-capacitance nor metal co-adsorption (doublelayer capacitance) occurred in the polymer (Figure 2B). From a practical point of view, capacitance of the carbon-grafted-PPy operating in LiBr solution is mostly generated by an ion double-layer at the carbon surface. This is further corroborated by the cyclic voltammetry tests

13 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 54

that revealed the absence of signals related to the redox activity of PPy (Figure 4, green and black scans). For both sodium chloride and bromide, the co-adsorption of alkali metals on the pyrrole unit adjacent to the charged ring was observed, as it was for potassium chloride and potassium bromide. Although preliminary calculations revealed a lack of interaction between the polymer and ions in lithium bromide, the same HOMO orbital (HOMO-5) was applied to visualize the minimum energy for co-adsorption, and was used for comparison with other electrolytes. As expected, the 5-membered PPy chain involving a polaron doped with LiBr showed the highest HOMO energy. Contrarily, the lowest HOMO energy is obtained in the presence of KCl. Based on computational results, the following trend is found: salts with the smallest anion (chloride) and the largest cation (potassium) generate the most stable and energetically favored structures. This effect may be associated with the formation of a delocalized 3-center and 4-electron bond between the immonium radical cation, halide and the metal cation. Likely, these three atoms may share four electrons. Thus, cations from alkali earth metals are preferred, since their valence electrons are present on outer orbitals and are only loosely bonded to the atom core compared to much smaller lithium cation. On the other hand, the electronegativity of chloride (according to the Pauling scale) is higher (3.16) than that of bromide (2.96), which may favour paring between the PPy immonium radical, halide and the metal cation with a doping anion of smaller size (thus chloride is preferred over bromide). The choice of aqueous electrolytes with alkali metal salts, especially sodium chloride or potassium chloride, is dictated by the need of preserving the environmentally benign character of the proposed materials. The use of salts like lithium chloride or lithium bromide to achieve pseudo-capacitance within the conjugate polymer has little or no potential application. However, they are included in this work in order to observe the effect of supporting electrolyte on

14 ACS Paragon Plus Environment

Page 15 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

capacitance. Salts such as LiF or CsBr are not at all used as electrolyte in capacitors, therefore they were not considered here.

Figure 2. The distribution of HOMO orbitals and interactions between an immonium radical cation in pyrrole, halide anion and metal cation in LiCl (A), LiBr (B), NaCl (C), NaBr (D), KCl (E) and KBr (F) electrolytes.

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 54

Another important conclusion made based on computations was the folding of PPy molecule upon oxidation (charge), and its linearization upon reduction (discharge). Figure S4 shows several snap shots of PPy structure taken from the software used in these calculations (0 corresponds to initial time). The formation of polarons, and associated with this the breaking of aromaticity of pyrrole rings during charge and discharge cycles, leads to significant volumetric changes in the PPy chain. This limits the lifetime of all conjugated polymers and is a major disadvantage of these materials, restricting their usage in commercial pseudo-capacitors. In order to minimize this effect, nanostructured polymers are recommended rather than bulky materials, since polymer thin films or nanoparticles are more flexible, allowing to accommodate volumetric changes. Also, the ion doping associated with formation of polarons upon charging is more efficient in nanostructured polymers, since their length scale (thickness, particle size) is close to the effective diffusion length (~ 40 nm for PPy). This results in utilization of the entire material during a single charge/discharge cycle. Similar effects were observed for polyaniline and confirmed by computations. These studies also demonstrated how electronic effects associated with electrochemical oxidation (and with the modification of the polymer chain due to folding and linearization) lead to energy consumption.35,36 Regarding instability of the polymer, our previous work has revealed that physical mixtures of PPy and carbon are only slightly more stable than pure PPy due to electronic effects (π orbital stacking between the graphitized carbon and the polymer37). Stability was significantly improved when we introduced a covalent coupling between the PPy and carbon.21 In physical blends, carbons are usually utilized as electron acceptors and the conjugated polymers as electron donors. The covalent bond generates a reverse donor/acceptor system. After doping of PPy with halides from the electrolyte, the polymer becomes highly conductive and behaves like an electron

16 ACS Paragon Plus Environment

Page 17 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

acceptor, whereby the function of carbon is reversed: it becomes an electron donor, resulting in the composite that is electrochemically and mechanically more stable.21 The co-adsorption of cations on the charged PPy is a new finding, and possibly contributes to the improvement of total capacitance of the carbon-grafted-PPy. This phenomenon generates an additional double-layer capacitance at the polymer/electrolyte interface, similar to the process occurring at the polarized carbon. The cation co-adsorption on PPy predicted by computational studies are further corroborated by experimental results carried out using the electrochemical quartz crystal microbalance. Co-adsorption of specific ions on PPy analyzed by EQCM The electrochemical quartz crystal microbalance measures the difference in vibrational frequency of a quartz crystal upon the formation or release of molecules at the resonator surface. The change of resonator frequency is correlated with the mass change of electrode coated onto the quartz crystal according to the Sauerbrey equation (Equation 1): ∆ =

−2 ∆ 

(1)

where f0 is the resonance frequency of quartz (7.995 MHz), A is area of the carbon-coated quartz resonator (0.1963 cm2), ρ is a crystal density (2.684 g cm-3) and µ represents the shear modulus of quartz (2.947 × 10-11 g cm-1 s-2). Figure 3B represents the EQCM scans of the MWCNT-grafted-PPy in KCl (red line), LiCl (black line) and pristine MWCNT in KBr (Figure 3B, green line). An increase in ∆m corresponds to ion adsorption and a decrease to desorption. For the carbon-grafted-PPy (red line), a peak at +0.872 V represents the uptake of chloride associated with the formation of immonium radical 17 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 54

cation in the PPy chain. The desorption of chloride occurs at +0.862 V in the reverse scan. Also, during the cathodic scan (from higher to lower potential), the reduction of PPy begins and the electrode mass decreases due to the expelling of doping anions. A small mass increase (relative to the trend) appears at +0.442 V, which corresponds to potassium uptake in the PPy. This is not observed for the bare MWCNTs (green scan). It indicates the cation co-adsorption at the oxidized polymer. Co-adsorption is reversible as soon as the potential reaches 0.122 V (Figure 3B, red line). For pristine MWCNTs in KCl, the mass increase is not observed at the potential of PPy oxidation (chloride uptake), only cation adsorption at -0.068 V and its expelling at 0.032 V (Figure 3B, green line). The black scan in Figure 3B corresponds to the MWCNT-grafted-PPy in 0.1 M LiCl solution, and the trend of mass change is much weaker as compared to the test in KCl. This indicates that ion exchange occurs only on the carbon fraction (without mass changes at the potentials of redox activity of PPy). The mass increase due to cation uptake on PPy was calculated after subtracting the vague signal from resonator, ink binder and carbon, as shown in Table 1 and in Figure 3A. Overall, we observe the trend that electrolytes with bigger cationic and smaller anionic radii are favourable for ion exchange at the carbon-grafted-PPy electrode, which is in agreement with the theoretical predictions. Regarding the ionic radius, which for K+ is 138 pm, for Na+ is 102 pm and for Li+ is 76 pm, their hydrated radii are 232 pm, 276 pm and 340 pm, respectively. It is well know that the ionic mobility decreases with increasing size of the hydrated radii (i.e., for K+ it is 64.5, for Na+ is 43.5 and for Li+ it is 33.5 ohm-1 cm2 mol-1). This is apparently a major reason for the rate of cation co-adsorption, which in the present case favours K+ not Li+. It also influences the interaction between the polaron center in the oxidized PPy chain and hydrated cations, depending on electronegativity of hydrated cations. For group IA ions, the electronegativity exhibits characteristic periodic oscillations, which generally decrease from top

18 ACS Paragon Plus Environment

Page 19 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

to bottom in a group and increase from left to right along a period. For example, for cations investigated in this work, values are in order: Li+ (0.984) < Na+ (1.063) > K+ (0.924). The aqueous electron affinities (As) of cations of group IA are another indicator of the tendency for cations in aqueous solution to take up electrons, and their As factors follow the same order as their solution phase electronegativity values. In summary, since the electronegativities of hydrate cations are similar for K+ and Li+, this parameter cannot justify the difference in co-adsorption due to additional electronic effects (repulsion or interaction) between cation and oxidized monomer. The only possible explanation of the preferential co-adsorption for potassium and sodium will be their smaller hydrated ionic radius and thus their higher mobility as compared to Li+. The surface coverage (number of molecular monolayers deposited on the electrode) was calculated using Equation 2:38 ∆ =  ∙  ∙

∆ (2) 

∆ represents the number of monolayers of molecules (NaCl, NaBr, KCl, KBr, LiCl and LiBr, respectively), ∆m is the mass change, MW the molecular weight of the adsorbed species, NA is Avogadro number and  =   corresponds to the surface area occupied by a molecule. The van der Waals geometry approximation was used to calculate the molecular radius according to Equations 3-4:38 $/& 3!  = # (3) 4

19 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 54

The van der Waals volume VvdW is estimated assuming a spherical shape of atoms (in Å molecule-1), n is the number of atoms in the molecule and rj is the atomic van der Waals radius of each element in Å (K = 3.00, Na = 2.57 Li = 2.24 Cl = 1.89 and Br = 2.03): 

! 4 & = ' ( − 5.92(, − 1) (4) 3 (.$

Table 1. Mass change (∆m) on the quartz resonator coated with a MWCNT-grafted-PPy synthesized by oxidative radical polymerization. Electrolytea Cation uptakeb (ng) Anion uptakeb (ng) Number of molecular monolayers (ML)c LiCl 0 0 0 LiBr 0 0 0 NaCl 5.05 0.87 3.04 NaBr 1.44 0.32 0.20 KCl 5.56 2.28 3.22 KBr 1.57 0.60 0.30 a all salts are in 0.1 M aqueous solution b

Calculated according to Equation (1)

c

Calculated for molecular species according to Equations (2-4)

20 ACS Paragon Plus Environment

Page 21 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

21 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 54

Figure 3. Mass uptake determined by EQCM and related to cationic radii (A). EQCM scans for MWCNT-grafted-PPy in 0.1 M KCl (blue) and in LiCl (black), and for bare MWCNTs (green) (B). The highest number of monolayers of adsorbed ions (counted as sum of anions and cations) was observed for potassium and sodium chloride. This also indicates preferential cation co-adsorption and anion doping of species with large ionic and small hydrated radii, and small anions at the carbon-grafted-PPy (Table 1). The co-adsorption of multilayers is also an indication for a synergistic electronic effect related to the increase of electron density at the carbon/polymer interface. This effect was reported for carbon/PPy or carbon/metal oxide, resulting in an improvement of their copacitance.39 The higher concentration of negative charge at the carbon/polymer (or metal oxide) interface results in the attraction of corresponding counter-ions from the electrolyte. This in total results in an enhancement of the capacitance of combined electrodes as compared to their individual components. For the coated resonator, a dissipation factor was taken into account in Sauerbrey's equation, as well as appropriate background subtraction, as proposed by others.45 Effect of supporting electrolyte on the gravimetric capacitance of carbon-grafted-PPy electrodes Cyclic voltammetry carried out for several types of carbon-grafted-PPy electrodes was used to determine the specific gravimetric capacitance in liquid and hydrogel electrolytes (the concentration of salt was the same for all tests), and their values are given in Table 2. Briefly, the charge calculated from the area under CV curve (Figure 3) was standardized to the mass of the

22 ACS Paragon Plus Environment

Page 23 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

capacitor active component (m in g), measured in the potential window (E2-E1 in V), and at the scan rate (ν in V sec-1) according to Equation 5: 6

2 3(4)54 /(0/1) = 6$ (5) 27(4 − 4$ ) The mass of active material (m in Equation 5) in the 3-electrode experiments carried out in liquid electrolyte was 7.225 × 10-5 g. For the hydrogel-embedded electrodes (2-electrode tests) it was 1.443 10-4 g for all combinations of the carbon-grafted-PPy.

23 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 54

Figure 4. CV scans for the SWCNH-grafted-PPy synthesized by oxidative radical polymerization, carried out in the 3-electrode system in various electrolytes. The effect of electrolyte chemical composition on the gravimetric capacitance was tested in solutions (Table 2). All electrodes showed the highest specific capacitance in KCl, followed by NaCl. All composites revealed the smallest capacitance in Li salts. This agrees with the electrogravimetric analysis carried out by EQCM and with theoretical predictions. With respect to the different carbon modifications, results are similar to the ones reported in our previous study,22 demonstrating that MWCNTs is the most suitable carbon for grafting with the organic linker and for pairing with the polymer, followed by SWCNH and graphene. The detailed electrochemical and morphological analysis of these composites was carried out in our previous work. Briefly, MWCNT are straightforward to functionalize due to the size of carbon particles that are suitable for the attachment of PPy structures. The length scale in SWCNHs, in particular the size of carbon particles is too small as compared to the polymer. In all proposed synthesis methods and using various organic linkers, the SWCNH-based materials showed that carbon is fully covered by PPy. Thus, the measured capacitance was mainly from the PPy, since the carbon surface is not accessible for the electrolyte. The electrochemical performance of the graphene-based composites depends on their fabrication. For instance, the product synthesized via e-ATRP showed significantly better capacitance as compared to RAFT or oxidative polymerization. This is related to the size of linker that is larger for the composite made by e-ATRP, resulting in a good separation of the carbon and polymer fractions. The product is more porous, favouring the electrolyte transport to the redox active centers of PPy. We also observed good separation of the graphene sheets in this composite, which indicated that the carbon surface is easily accessible for ions, resulting in good utilization of carbon and relatively high double-layer capacitance. In

24 ACS Paragon Plus Environment

Page 25 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

contrary, both the RAFT and oxidative radical polymerizations led to a material composed of agglomerated large PPy particles (decrease in pseudo-capacitance) and to graphene layers that were prone to stacking, resulting in reduced double-layer capacitance. After testing various composites in liquid electrolytes, KCl was chosen for the hydrogel electrolyte and subjected to the electrochemical analysis. A poly(acrylamide)/poly (N, Nmethylenebisacrylamide) blend was synthesized by UV cross-linking from an aqueous suspension of AAm and N, N-methylenebisacrylamide (MBA) in the presence of ammonium persulfate (APS) as polymerization initiator. Nafion 117® was used to modify the surface tension, which improved the dispersion of composite (0.1 wt%) in the hydrogel. These laminates were tested in a two-electrode cell with a configuration similar to commercial setups (results presented in Table 2, samples 10-18). The capacitance measured for laminates follows the same trend as for corresponding composites analysed in liquid electrolytes, with the MWCNT-based electrodes demonstrating the highest capacitance as compared to graphene- and SWCNHgrafted-PPy. For the laminate no. 10, the gravimetric capacitance exceeded that of the one measured in liquid electrolyte. The specific capacitance summarized in Table 2 for all electrodes was also determined using impedance spectroscopy as complementary method. Figure S6 shows an example of log(susceptance) = f(log(ω)) that was used to calculate the specific capacitance according to the equations shown in Figure S6. The specific capacitances are in the same order of magnitude as compared to values found by the CV method. Also, overall trend in capacitance for all types of electrodes is similar in voltammetry and AC impedance experiments. This indicates that both electrochemical methods are suitable for quantifying the charge capacity for electrodes proposed in this study.

25 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 54

Not only had the hydrogel-containing KCl good ionic conductivity (close to that of 0.1 M KCl), but the overall dispersion of the capacitor material within the hydrogel matrix was significantly better than the same material tested onto the glassy carbon disk. Details regarding the optimization of manufacturing procedures and characteristics related to stretchable laminates are investigated in separate work.40 Table 2. Comparison of specific gravimetric capacitances, C, (F g-1) for the active materials cast on the glassy carbon and embedded in the hydrogel electrolyte. Material a 1b 2b 3b 4c 5c 6c 7d 8d 9d 10e 11e 12e 13f 14f 15f 16g 17g 18g

LiCl 283.42 7.5700 32.090 112.35 92.41 11.39 37.58 55.86 9.280 -

LiBr 250.27 10.090 29.700 101.23 35.42 7.41 29.62 52.17 8.700 -

NaCl 334.75 16.860 79.130 298.75 210.14 85.61 87.88 97.75 38.52 -

NaBr 260.96 8.4700 43.040 205.41 184.32 15.76 52.08 66.95 16.28 -

KBr 311.22 22.750 88.389 299.14 217.32 87.88 87.48 104.81 17.21 -

KCl 350.49 43.610 33.200 338.43 231.88 98.540 456.86 254.01 113.94 516.86 102.18 27.230 456.31 375.99 358.13 298.61 212.30 197.32

a

(1,4,7) MWCNT-grafted-PPy, (2,5,8) Graphene-grafted-PPy, (3,6,9) SWCNH-grafted-PPy measured in aqueous media, (10,13,16) MWCNT- grafted -PPy, (11,14,17) Graphene-graftedPPy, (12,15,18) SWCNH- grafted-PPy b prepared by oxidative radical polymerization, tested in liquid electrolyte c prepared by e-ATRP polymerization, tested in liquid electrolyte d prepared by RAFT polymerization, tested in liquid electrolyte e active material prepared by oxidative radical polymerization embedded in hydrogel f active material prepared by e-ATRP embedded in hydrogel g active material prepared by RAFT embedded in hydrogel

26 ACS Paragon Plus Environment

Page 27 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Ionic mobility in carbon-grafted-PPy electrodes analysed by electrochemical impedance spectroscopy: effect of the type of electrolyte

Electrochemical impedance spectroscopy (EIS) allowed to further determine mass transfer parameters, such as diffusion coefficient and ion mobility for the carbon-grafted-PPy electrodes in liquid and hydrogel electrolytes. Since the ion mobility depends on the applied potential, EIS tests were carried out under various electrode polarizations. The electrical equivalent circuit was further used for data fitting, allowing to calculate the transition time (τD) of the ion transport from the electrolyte to the active centers of electrode (in case of pseudo-capacitance), and to determine the capacitive double-layer at the polarized carbon, according to Equation (6): 89 = :; / (6) Rs is the solution resistance, C is the total capacitance (C = C1 + C2, where C1 and C2 are estimated from the fitting of impedance spectra and the electrical equivalent circuit, Figures 6 and 7). τD was further used to calculate the diffusion coefficient (Dn) using Equation (7): > =- = (7) 89 L is an effective diffusion length that was estimated using an atomic force microscope (AFM) operating in a contact mode. The parameter L is influenced by the material's porosity and surface roughness, and an example of AFM topography image for the MWCNT-grafted-PPy is shown in Figure 5.

27 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 54

Figure 5. AFM topography image of MWCNT-grafted-PPy (prepared by oxidative radical polymerization) coated on a glassy carbon disk. The statistical distribution of the film thickness was used for all calculations, and for the hydrogel-embedded capacitors, elements of the electrical equivalent circuit were also normalized to the thickness and size of laminates. Furthermore, the Einstein-Nernst model (Equation 8) was applied to compute the ionic mobility:

- =

@= (8) AB C

kB is the Boltzmann constant, T is temperature (kbT is the thermal energy), and µn stands for the ionic mobility. The parameters Rs and C were estimated from fitting the electrical equivalent circuits shown as inserts in Figure 6 and Figure 7. Depending on the applied polarization, the 28 ACS Paragon Plus Environment

Page 29 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

impedance spectra are composed of two semi-circles at higher frequencies, followed by a straight line with the slope of about 45o and, more or less developed, the vertical line at the lowest frequency (right side of Nyquist plots). The left side intercept of the first arc represents the solution resistance, Rs, R1 and R2 are charge transfer resistances at the electrode/electrolyte interface and within the material, respectively. C1 is the capacitance at the electrode/electrolyte interface, and C2 represents the capacitance within the electrode; Wo is the Warburg element associated with the total mass transfer. Table 3 lists transition time (τd), resistance of solution (Rs), and ionic mobility (µn) calculated from Equations 6-8, and based on fittings of Nyquist plots at polarization of 0.872 V (where the maximum of the pseudo-capacitance of PPy and double-layer capacitance was observed). Table 3. Transition time, τd (ms), resistance of solution Rs and ionic mobility, µ (cm2 V-1 s-1), for carbon-grafted-PPy electrodes in liquid and hydrogel electrolytes at a polarization of 0.872 V. The concentration of all electrolytes is 0.1 M. Samplea τd (ms) × 10-4 1b 29.0 b 2 3.10 3b 3.65 c 4 2.95 5c 3.64 c 6 4.65 7d 34.5 d 8 6.80 d 9 4.68 10e 0.137 e 11 6270 12e 4044 f 13 3301 14f 1540 f 15 3450 g 16 3360

Rs (Ω) µ × 10-4 KCl 18.6 34.3 16.8 32.1 89.1 27.3 14.1 3.34 13.7 2.74 1.51 2.14 7.84 6.86 50.0 1.46 3.12 2.13 0.78 0.073 0.88 0.0047 1.42 0.0384 1.65 0.0179 0.62 0.0402 1.29 0.0391 1.26 0.0185

µ × 10-4 NaCl 2.87 26.6 13.0 2.83 3.11 0.252 1.46 0.494 1.03 -

µ × 10-4 NaBr 3.32 2.38 3.71 3.22 3.06 0.327 1.23 0.376 0.768 -

µ × 10-4 LiCl 2.35 2.23 9.52 1.98 3.10 66.2 0.114 0.205 0.502 -

µ × 10-4 LiBr 2.53 1.80 3.26 2.27 2.43 0.348 0.609 0.503 0.607 -

µ × 10-4 KBr 2.52 2.38 5.01 3.69 2.27 0.280 0.969 0.502 0.753 29

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

17g 18g

1590 1580

1.23 3.91

0.0184 0.0036

-

-

-

Page 30 of 54

-

-

a

(1,4,7) MWCNT-grafted-PPy, (2,5,8) Graphene-grafted-PPy, (3,6,9) SWCNH-grafted-PPy measured in aqueous media, (10,13,16) MWCNT-grafted -PPy, (11,14,17) Graphene-graftedPPy, (12,15,18) SWCNH-grafted-PPy b prepared by oxidative radical polymerization, tested in liquid electrolyte c prepared by e-ATRP polymerization, tested in liquid electrolyte d prepared by RAFT polymerization, tested in liquid electrolyte e active material prepared by oxidative radical polymerization embedded in hydrogel f active material prepared by e-ATRP embedded in hydrogel g active material prepared by RAFT embedded in hydrogel

Overall, the MWCNT-grafted-PPy composites prepared by various methods featured the highest ionic mobilities in KCl solution as compared to the same electrodes measured in other liquid electrolytes. The MWCNT-based electrodes also showed the best performance among all tested electrodes. Mobilities in graphene- and SWCNH-grafted-PPy were comparable, but their electrochemical performances were significantly worse than that of MWCNT-grafted-PPy electrodes. There is a substantial difference between electrodes made from MWCNTs using different methods of grafting and polymerization; e.g., MWCNT-grafted-PPy synthesized by oxidative radical polymerization is significantly better than MWCNT-based electrodes prepared via living radical polymerization (sample 1b, 4c and 7d in Table 3). For graphene- and SWCNHgrafted-PPy, the effect of polymerization method and the type of covalent linker connecting carbon and PPy is not that strong as for the MnWCNT-grafted-PPy. These observations agree with theoretical predictions, EQCM and capacitance analysis. The conclusions is that the KCl solution (i.e. a combination of ions with large cationic (but small hydrated), and small anionic radii is the most suitable electrolyte for all electrodes. Opposed to that, an electrolyte composed of small cations such as Li+ (but having a large hydrated radius) and large anions (Br-) is not the best choice for the proposed composite electrodes. Similar trends were observed for the same

30 ACS Paragon Plus Environment

Page 31 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

materials embedded in hydrogels containing KCl as ion source, except that the ionic mobility was on average two to three orders of magnitude lower than for the same electrodes in liquids. Although Rs was smaller in hydrogels than in liquid electrolytes, the transition times were significantly longer, resulting in smaller Dn and µ for hydrogels. Also, their total resistance (R2, which is sum of Rs and R1) was higher due to the contribution of charge transfer resistance at the grain boundaries of the polyacrylamide phase. This acts as a barrier for mobile ions, leading to longer transition times (τD, Table 3). The total capacitance (parameter C in Equation 6) of the hydrogel-embedded electrodes is much higher than that in liquid, which is also consistent with the capacitance analysed by cyclic voltammetry (Table 2), where the highest gravimetric specific capacitance was recorded for solid electrodes. Regardless of lower ion mobility in hydrogel electrolytes, the overall performance of the hydrogel-based electrodes is excellent. In fact, it is comparable to that of liquid systems, as summarized in Table 2. Interestingly, mobilities for all electrodes embedded in hydrogel are similar. There is some improvement for composites made by oxidative radical polymerization as compared to the living polymerization method, but this trend is much weaker than electrodes tested in liquid electrolytes. Thus, the hydrogel-embedded capacitor electrodes are an excellent solution for leakage-free, flexible, even printable capacitors or components of other energy harvesting systems that require lightweight materials.

31 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 54

Figure 6. Nyquist plots recorded at various potential biases (from 0.522 to 1.222 V) for MWCNT-grafted-PPy prepared by oxidative radical polymerization and coated on a glassy carbon disk electrode in aqueous 0.1 M KCl liquid electrolyte. All presented spectra were taken at the same electrode polarization (0.872 V). Figure 7 shows the Nyquist plots recorded for liquid electrolytes taking as example (A) graphene-grafted-PPy synthesized by e-ATRP, (B) various materials embedded into hydrogel 32 ACS Paragon Plus Environment

Page 33 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

containing KCl, and (C) the impedance spectra of all electrodes in liquid KCl. The insert in Figure 7A shows the equivalent circuit consisting of a series of resistors and capacitors as described earlier in this section. The C2 element that corresponds to the capacitive behavior of examined specimens is represented by the vertical line at the lowest frequency (right side of the spectra), and a weak Warburg element shown as the line with a slope of 45° (Figure S5 shows the middle range of frequencies with Warburg impedance).41 For some electrodes, Dn values calculated from the Warburg element were very similar to those estimated using Equation 7 (data not shown), however the fitting of a Wo resistance showed high errors due to the small number of experimental points in the part of spectra corresponding to the Warburg element (short line with the slope of 45o in the middle of the Nyquist plot, Figure 7). The estimation of the Warburg resistance was even more problematic for electrodes showing significant potential-dependent variations of the R2 resistance (charge transfer within the bulk electrode material). For comparison, the Rs resistance was considered to be the most relevant in order to show the overall trend in the charge mobility. Figure 7A clearly demonstrates that the lowest resistances of the solution (Rs) and the charge transfer at the electrode/electrolyte interface (R1) are observed for KCl (green), NaCl (yellow) and NaBr (blue), and the highest resistances for lithium salts. For the hydrogel/KCl -embedded electrodes made with different carbons (Figure 7B), the Rs values (left-side intercept of arc on the Nyquist plot) are similar for all types of electrodes, and the MWCNT-based electrodes showed some advantage in terms of the total charge transfer resistance, as compared to the SWCNH- and graphene-based systems. The lower Rs value observed in hydrogels containing active materials (Figure 7B) are probably due to the presence of Nafion 117®, which is known to have positive effect, leading to improvement of the ionic conductivity of similar systems. Nafion

33 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 54

generates percolated water channels (an arrangement of Nafion hydrophilic ends towards the water incorporated in the hydrogel), resulting in an enhancement of both the vehicular and Grotthuss type of charge transport within the solid ionic conductor. This was also confirmed by ab initio computational studies.24

34 ACS Paragon Plus Environment

Page 35 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 7. EIS spectra of graphene-grafted-PPy prepared by e-ATRP [40] (A), carbon-graftedPPy synthesized using different methods and incorporated into the hydrogel containing KCl and tested in a 2-electrode system (Figure S2, Supporting information) (B), carbon-grafted-PPy synthesized using different methods tested in a 3-electrode cell with KCl solution (C).

Ionic mobility in carbon-grafted-PPy electrodes analysed by electrochemical impedance spectroscopy: effect of electrode polarization Figure 8 shows the ionic mobility as a function of applied potential bias calculated from Equation 8, and taking into account the electrical parameters estimated from EIS spectra. Clearly, the highest ionic mobility is observed in KCl, followed by NaCl in the potential range from 0.522 to 0.922 V (where a synergistic effect of double-layer and pseudo-capacitance is expected). At polarizations higher than 0.922 V, all chlorides show almost the same ionic mobility. Its decrease is related to the saturation of polymer fraction with doping anions (also called the electrolyte starvation42). A further important observation is mobility of bromides, which for all cases was the lowest at potentials below 0.922 V, and significantly higher above 0.922 V. Regarding the mobilities at potentials below 0.922 V, the rate of polymer doping with bromide counter-anions is slower as compared to chloride. This is simply related to lower ionic mobility of the bromide in both aqueous and organic solvents, which is well-known and reported elsewhere.43 The major increase in ionic mobility for all bromides at higher polarization may be related to the formation of tribromide (Br2 + Br- → Br3-). We further confirmed this by the UV-Vis analysis of KBr solution that was subjected to polarization within the range of 0.4 to 1 V. Clearly, the formation of yellow solution at the electrode surface (Br2) was observed during the measurements (Figure S3, 35 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 54

supporting information). The increase of absorption intensity at 253 nm (slightly shifted towards higher wavelengths at higher concentration) upon increasing polarization indicates on the formation of Br2 and Br3-. In addition to this effect, the computational analysis performed by another research group44 showed the distinctive difference in residence times around the cation between chloride and bromide, and led to a model that classifies halides as "free" and "bound", where bromide is considered as “bound”. In addition, the interaction of sodium chloride and sodium bromide, being simultaneously present in the aqueous solution with the same cation, demonstrated the selectivity effect, where bromide ions tend to replace chloride in the immediate vicinity of the positively charged molecules. This also suggests higher affinity of bromides to the charged surface, resulting in the substantial variation in bromide mobility at higher polarizations observed in this work. All Li salts showed slightly lower mobilities than KCl, but higher ones than sodium-containing electrolytes. This can be attributed to the extent of co-adsorption of cations during doping of PPy. As demonstrated by theoretical predictions, the cation co-adsorption associated with the redox activity of PPy was not observed for Li salts (Figure 2, structure of HOMO orbitals in LiBr). This was further confirmed by the electro-gravimetric studies carried out using EQCM (Figure 3), and the quantification of cation co-adsorption shown in Table 1 for all electrolytes. This leads to the conclusion that the contribution of co-adsorbed cation to increase in total capacitance (this co-adsorption contributes to the increase in double-layer capacitance, but not in pseudocapacitance), in the potential range corresponding to the redox activity of PPy (maximum pseudo-capacitance) is small, within the same order of magnitude. Possibly, this co-adsorption effect will be stronger for composites that have significantly higher specific surface area as compared to materials synthesized in this study.

36 ACS Paragon Plus Environment

Page 37 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 8. Ionic mobility calculated from Equation 8 for the MWCNT-grafted-PPy in various liquid electrolytes (0.1 M) as a function of potential bias. Conclusions We analyzed the influence of electrolyte on the electrochemical performance of capacitor electrodes made from MWCNT, graphene and SWCNH grafted with PPy, using oxidative radical polymerization and two different living polymerizations. Electrochemical investigations supported by theoretical modeling led to the following conclusions: i) Computational studies on a 5-membered PPy chain with a single charged pyrrole carried out using the MMFFaq and DFT-B3LYP 6-31G models revealed a simultaneous cation co-adsorption, 37 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 54

occurring during polymer oxidation (and associate anion doping) for potassium and sodium halides, but not for lithium salts. The lowest minimum energy of cation co-adsorption is observed for KCl. This co-adsorption is also possible in NaCl, NaBr and KBr solutions, but the effect is weaker as compared to the KCl solution. ii) Electro-gravimetric analysis of the mass change on the polarized material showed an increase in electrode mass related to cation co-adsorption on the oxidized polymer in KCl; the coadsorption generates additional double-layer capacitance and contributes to the total increase in gravimetric capacitance for all type of electrodes measured in KCl; this phenomenon was not observed in EQCM scans for various electrodes carried out in LiCl solutions. The calculated number of monolayers of adsorbed ions (sum of cations and anion‚) were higher than unity for KCl and NaCl, indicating co-adsorption processes at the polarized polymer. iii) The highest ion mobility investigated using electrochemical impedance spectroscopy was found for MWCNT-grafted-PPy prepared by oxidative radical polymerization in liquid electrolytes. The same system prepared by living radical polymerizations (RAFT and e-ATRP) featured lower mobilities, followed by the capacitive material synthesized using the same method with graphene and SWCNHs. This result is related to the size and morphology of carbon particles; for single-walled carbon nanohorns, carbon structures are too small (range of 20 nm) to host larger PPy particles, resulting in polymer agglomeration; the synthesis procedure has to be further optimized for graphene in order to avoid the stacking of bare graphene sheets (which also caused PPy agglomeration). The synthesis procedures of composites with graphene and SWCNHs will be further optimized in order to better utilize the carbon fraction. iv) The ionic mobilities for various carbon-grafted-PPy embedded in hydrogel electrolyte were comparable, except for the graphene-based electrodes (one order of magnitude lower than 38 ACS Paragon Plus Environment

Page 39 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

MWCNT- and SWCNH-grafted-PPy); the solution resistance of the hydrogel-KCl was slightly lower than that of the liquid phase, due to the addition of Nafion. The ionic conductivity of hydrogels was improved due to the formation of water percolation channels from protonated sulfonic groups in Nafion. v) The ionic mobility is a potential-dependent entity; the highest value was observed for KCl in the range of redox activity of PPy (from 0.522 to 0.922 V), and decreases for chlorides at higher potentials due to the saturation of the polymer (maximum doping). The Li salts showed smaller ion mobilities, but in the same order of magnitude as K and Na chlorides, which is related to the size of ionic radii (for the doping of PPy smaller anions, e. g., Cl-, are favoured and larger cations, e.g. K+, participate in the co-adsorption). All bromides showed unusually high mobilities at potentials above 0.922 V, owing to additional electrochemical activity that leads to the formation of Br3-. vi) Cyclic voltammetry revealed the highest gravimetric capacitance for the MWCNT-graftedPPy incorporated in hydrogels and synthesized via oxidative radicals (516.86 F g-1) and e-ATRP (456.31 C g-1), which was significantly higher than for the same electrode measured in the liquid electrolyte (350.49 F g-1 for MWCNT-grafted-PPy fabricated by oxidative radical polymerization, and 338.43 F g-1 by e-ATRP). Poly-acrylamide based hydrogels are promising electrolytes for leakage-free, flexible and light-weight energy harvesting systems or printed electronics. Supporting information: the chemical structures of carbon-grafted-PPy, images of

the

electrochemical set-up used for hydrogel-embedded electrodes, UV-Vis spectra of the bromine intermediate product and a photo of the electrolyte solution, the visualization of folding of the PPy chain, an example of impedance spectra, the complementary method (based on impedance 39 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 54

data and the linear correlation of susceptance as the function of log(frequency)) for the quantification of specific capacitances with relevant mathematical models are given in supporting information. Acknowledgments This work was carried out with the financial support of New Brunswick Foundation for Innovation, Frank J and Norah Toole Graduate Scholarship, UNB President and NSERC Discovery grant No: RGPIN-2016-03620. References

(1)

Verzijlbergh, R. A.; De Vries, L. J.; Dijkema, G. P. J.; Herder, P. M. Institutional Challenges Caused by the Integration of Renewable Energy Sources in the European Electricity Sector. Renew. Sustain. Energy Rev. 2016, 75 (November 2015), 0-1.

(2)

Curve, D.; Ahead, A. W. Electricity Currents: A Survey of Current Industry News and Developments California’s “Duck Curve” 2016, 29, 71-72.

(3)

Denholm, P.; O’Connell, M.; Brinkman, G.; Jorgenson, J. Overgeneration from Solar Energy in California: A Field Guide to the Duck Chart (NREL/TP-6A20-65023). 2015, No. November, 46.

(4)

Miller, J. R.; Outlaw, R. A.; Holloway, B. C. Graphene Double-Layer Capacitor with Ac Line-Filtering Performance. Science 2010, 329 (5999), 1637-1639.

(5)

Conte, M. Supercapacitors Technical Requirements for New Applications. Fuel Cells 2010, 10 (5), 806-818. 40 ACS Paragon Plus Environment

Page 41 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

(6)

Zhong, C.; Deng, Y.; Hu, W.; Qiao, J.; Zhang, L.; Zhang, J. A Review of Electrolyte Materials and Compositions for Electrochemical Supercapacitors. Chem. Soc. Rev. 2015, 44 (21), 7484-7539.

(7)

Roberts, A. J.; Slade, R. C. T. Effect of Specific Surface Area on Capacitance in Asymmetric Carbon-MnO2 Supercapacitors. Electrochim. Acta 2010, 55 (25), 7460-7469.

(8)

Wang, Q.; Wen, Z.; Li, J. A Hybrid Supercapacitor Fabricated with a Carbon Nanotube Cathode and a TiO2-B Nanowire Anode. Adv. Funct. Mater. 2006, 16 (16), 2141-2146.

(9)

Zhu, J.; Xu, Y.; Wang, J.; Lin, J.; Sun, X.; Mao, S. The Effect of Various Electrolyte Cations on Electrochemical Performance of Polypyrrole/RGO Based Supercapacitors. Phys. Chem. Chem. Phys. 2015, 17 (43), 28666-28673.

(10)

Choudhury, N.; Sampath, S.; Shukla, A. K. Hydrogel-Polymer Electrolytes for Electrochemical Capacitors: An Overview. Energy Environ. Sci. 2009, 2 (1), 55-67.

(11)

Yang, J.; Liu, Y.; Liu, S.; Li, L.; Zhang, C.; Liu, T. Conducting Polymer Composites: Material Synthesis and Applications in Electrochemical Capacitive Energy Storage. Mater. Chem. Front. 2017, 251-268.

(12)

Shi, Y.; Peng, L.; Ding, Y.; Zhao, Y.; Yu, G. Nanostructured Conductive Polymers for Advanced Energy Storage. Chem. Soc. Rev. 2015, 44 (19), 6684-6696.

(13) Jung, H. J.; Kim, Y.; Han, J. H.; Yudasaka, M.; Iijima, S.; Kanoh, H.; Kim, Y. A.; Kaneko, K.; Yang, C. Thermal-Treatment-Induced Enhancement in Effective Surface Area of

41 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 54

Single-Walled Carbon Nanohorns for Supercapacitor Application. J. Phys. Chem. C 2013, 117, 25877-25883.

(14)

Han, Z. J.; Seo, D. H.; Yick, S.; Chen, J. H.; Ostrikov, K. (Ken). MnOx/carbon Nanotube/reduced Graphene Oxide Nanohybrids as High-Performance Supercapacitor Electrodes. NPG Asia Mater. 2014, 6 (10), e140.

(15)

Shiraishi, S.; Kibe, M.; Yokoyama, T.; Kurihara, H.; Patel, N.; Oya, A.; Kaburagi, Y.; Hishiyama, Y. Electric Double Layer Capacitance of Multi-Walled Carbon Nanotubes and B-Doping Effect. Appl. Phys. A Mater. Sci. Process. 2006, 82 (4 SPEC. ISS.), 585-591.

(16)

Volfkovich, Y. M.; Mikhailin, A. A.; Bograchev, D. A.; Sosenkin, V. E.; Bagotsky, V. S. Studies of Supercapacitor Carbon Electrodes with High Pseudocapacitance. Recent Trend Electrochem. Sci. Technol. 2012, 159-182.

(17)

Ren, Z.; Li, J.; Ren, Y.; Wang, S.; Qiu, Y.; Yu, J. Large-Scale Synthesis of Hybrid Metal Oxides through Metal Redox Mechanism for High-Performance Pseudocapacitors. Sci. Rep. 2016, 6 (September 2015), 20021.

(18)

Zou, R.; Yuen, M. F.; Yu, L.; Hu, J.; Lee, C.-S.; Zhang, W. Electrochemical Energy Storage Application and Degradation Analysis of Carbon-Coated Hierarchical NiCo2S4 Core-Shell Nanowire Arrays Grown Directly on Graphene/Nickel Foam. Sci. Rep. 2016, 6 (August 2015), 1-9.

42 ACS Paragon Plus Environment

Page 43 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

(19)

Medina-Llamas, J. C.; Chávez-Guajardo, A. E.; Andrade, C. A. S.; Alves, K. G. B.; de Melo, C. P. Use of Magnetic Polyaniline/Maghemite Nanocomposite for DNA Retrieval from Aqueous Solutions. J. Colloid Interface Sci. 2014, 434, 167-174.

(20)

Iwakura, C.; Murakami, H.; Nohara, S.; Furukawa, N.; Inoue, H. Charge-Discharge Characteristics of Nickel/zinc Battery with Polymer Hydrogel Electrolyte. J. Power Sources 2005, 152 (1–2), 291-294.

(21)

Radtke, M.; McMillan, D. G. G.; Schroeter, B.; Hoeppener, S.; Dietzek, B.; Schubert, U. S.; Ignaszak, A. The Effect of 3-Amino Benzoic Acid Linker and the Reversal of DonorAcceptor Pairs on the Electrochemical Performance and Stability of Covalently Bonded Polypyrrole Nanotubes. Polymer 2015, 77, 289-296.

(22)

Radtke, M.; Ignaszak, A. Grafting of the Carbon Allotropes and Polypyrrole via a KevlarType Organic Linker: The Correlation of Carbon Structure/morphology with Electrochemistry of the Composite Electrode. Mater. Renew. Sustain. Energy 2017, 6 (1), 1-9.

(23)

Fic, K.; Lota, G.; Meller, M.; Frackowiak, E. Novel Insight into Neutral Medium as Electrolyte for High-Voltage Supercapacitors. Energy Environ. Sci. 2012, 5 (2), 58425850.

(24)

Luduea, G. A.; Koehne, T. D.; Sebastiani, D. Mixed Grotthuss and Vehicle Transport Mechanism in Proton Conducting Polymers from Ab Initio Molecular Dynamics Simulations. Chem. Mater. 2011, 23 (6), 1424-1429.

43 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(25)

Page 44 of 54

Garcia-Belmonte, G.; Munar, A.; Barea, E. M.; Bisquert, J.; Ugarte, I.; Pacios, R. Charge Carrier Mobility and Lifetime of Organic Bulk Heterojunctions Analyzed by Impedance Spectroscopy. Org. Electron. Physics, Mater. Appl. 2008, 9 (5), 847-851.

(26)

Bisquert, J. Chemical Capacitance of Nanostructured Semiconductors: Its Origin and Significance for Nanocomposite Solar Cells. Phys. Chem. Chem. Phys. 2003, 5 (24), 5360-5364.

(27)

Radtke, M.; Ignaszak, A. Carbon allotropes grafted with poly (pyrrole) derivatives via living radical polymerizations: electrochemical analysis of nano-composites for energy storage. RSC Advances 2017, 7, 35060-35074.

(28)

Dante, A. B. I; Tong, M.; Wu, G.; Furlani, E. P. Numerical Analysis of electric Double Layer Capacitors with Mesoporous Electrodes: Effect of Electrode and electrolyte Properties. J. Phys. Chem. C 2015, 119, 25235-25242.

(29)

Weidlich, C.; Mangold, K. M.; Jüttner, K. EQCM Study of the Ion Exchange Behaviour of Polypyrrole with Different Counterions in Different Electrolytes. Electrochim. Acta 2005, 50 (7–8), 1547-1552.

(30)

Vlad, A.; Singh, N.; Rolland, J.; Melinte, S.; Ajayan, P. M.; Gohy, J.-F. Hybrid Supercapacitor-Battery Materials for Fast Electrochemical Charge Storage. Sci. Rep. 2014, 4, 4315-4322.

44 ACS Paragon Plus Environment

Page 45 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

(31)

Syritski, V.; Öpik, A.; Forsén, O. Ion Transport Investigations of Polypyrroles Doped with Different Anions by EQCM and CER Techniques. Electrochim. Acta 2003, 48 (10), 14091417.

(32)

Brânzoi, V.; Brânzoi, F.; Pilan, L. Electrochemical Characterization of Polypyrrole Modified Electrodes Doped with Large Organic Anions. Rev. Roum. Chim. 2007, 52 (1–2), 169-179.

(33) Pensado, A. S.; Malberg, F.; Gomes, M. F. C.; Padua, A. A. H.; Fernandez, J.; Kirchner, B. Interactions and Structure of Ionic Liquids on Graphene and Carbon Nanotubes Surfaces. RSC Adv. 2014, 4 (35), 18017-18024.

(34)

Le, T.-H.; Kim, Y.; Yoon, H. Electrical and Electrochemical Properties of Conducting Polymers. Polymers (Basel). 2017, 9 (4), 1-32.

(35)

Brédas, J. L.; Beljonne, D.; Coropceanu, V.; Cornil, J. Charge-Transfer and EnergyTransfer Processes in π-Conjugated Oligomers and Polymers: A Molecular Picture. Chem. Rev. 2004, 104 (11), 4971-5003.

(36)

Bay, L.; Jacobsen, T.; Skaarup, S.; West, K. Mechanism of Actuation in Conducting Polymers: Osmotic Expansion. J. Phys. Chem. B 2001, 105 (36), 8492-8497.

(37)

Mosch, H.; Höppener, S.; Paulus, R. M.; Schröter, B.; Schubert, U. S.; Ignaszak, A. The correlation of the binding mechanism of the polypyrrole–carbon capacitive interphase with electrochemical stability of the composite electrode. Phys. Chem. Chem. Phys. 2015, 17, 13323-13332.

45 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(38)

Page 46 of 54

Lam, V. W. S.; Kannangara, D. C. W.; Alfantazi, A.; Gyenge, E. L. Electrochemical Quartal Crystal Microbalance Study of Borohydride Electro-Oxidation on Pt: The Effect of Borohydride Concentration and Thiourea Adsorption. J. Phys. Chem. C 2011, 115 (6), 2727-2737.

(39)

Dubal, D. P.; Ayyad, O.; Ruiz, V.; Gómez-Romero, P. Hybrid energy storage: the merging of battery and supercapacitor chemistries. Chem. Soc. Rev. 2015, 44, 1777-1790.

(40)

Radtke, M.; Ignaszak, A. Stretchable and electrically/ionic conductive hydrogels with well dispersed pristine nanostructured carbons for applications as supercapacitors and energy harvesting media. Manuscript in preparation

(41)

Kleemann, H. Theory of the Impedance of Electron Diffusion and Recombination in a Thin Layer. 2010, 325-333.

(42)

Gu, W.; Yushin, G. Review of Nanostructured Carbon Materials for Electrochemical Capacitor Applications: Advantages and Limitations of Activated Carbon, CarbideDerived Carbon, Zeolite-Templated Carbon, Carbon Aerogels, Carbon Nanotubes, Onionlike Carbon, and Graphene. Wiley Interdiscip. Rev. Energy Environ. 2014, 3 (5), 424-473.

(43)

Harris, D. C. Quantitative Chemical Analysis, 9th Edition, W. H. Freeman & Company, 2016, page 344.

(44) Watson, K.; Farré, M. J.; Knight, N. Strategies for the removal of halides from drinking water sources, and their applicability in disinfection by-product minimisation: a critical review. J. Environ. Manag. 2012, 110, 276-298.

46 ACS Paragon Plus Environment

Page 47 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

(45) Tsai, W.Y.; Taberna, P.L.; Simon, P. Electrochemical Quartz Crystal Microbalance (EQCM) Study of Ion Dynamics in Nanoporous Carbons. J. Am. Chem. Soc. 2014, 136, 8722-8728. (46) Jeon, I.Y.; Choi, H.J.; Tan L-S.; Baek, H.B. Nanocomposite prepared from in situ grafting of polypyrrole to aminobenzoyl-functionalized multiwalled carbon nanotube and its electrochemical properties. J. Polym. Sci. A Polym. Chem. 2011, 49, 2529-2537.

47 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Fig1 275x192mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 48 of 54

Page 49 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Fig2 1000x1016mm (120 x 120 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Fig3 207x270mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 50 of 54

Page 51 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Fig4 241x196mm (300 x 300 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Fig5 215x212mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 52 of 54

Page 53 of 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Fig7 157x124mm (300 x 300 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

graphical abstract 154x103mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 54 of 54