SEARCH

SEARCH BY CITATION

Keywords:

  • capacitors;
  • electrochemistry;
  • nanostructures;
  • polymers;
  • renewable resources

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

Strong demand for high-performance energy-storage devices has currently motivated the development of emerging capacitive materials that can resolve their critical challenge (i.e., low energy density) and that are renewable and inexpensive energy-storage materials from both environmental and economic viewpoints. Herein, the pseudocapacitive behavior of lignin nanocrystals confined on reduced graphene oxides (RGOs) used for renewable energy-storage materials is demonstrated. The excellent capacitive characteristics of the renewable hybrid electrodes were achieved by synergizing the fast and reversible redox charge transfer of surface-confined quinone and the interplay with electron-conducting RGOs. Accordingly, pseudocapacitors with remarkable rate and cyclic performances (∼96 % retention after 3000 cycles) showed a maximum capacitance of 432 F g−1, which was close to the theoretical capacitance of 482 F g−1 and sixfold higher than that of RGO (93 F g−1). The chemical strategy delineated herein paves the way to develop advanced renewable electrodes for energy-storage applications and understand the redox chemistry of electroactive biomaterials.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

Electrochemical energy storage, such as that in supercapacitors (SCs) and lithium ion batteries (LIBs), is of paramount importance for a wide range of applications.1 Capacitive charge storage by SCs shows more desirable features than LIBs in terms of high power, fast charge/discharge rate, long cycle life, and safe operation,2 but the capacitance of SCs that deliver charges through nonfaradaic means (i.e., electrochemical double-layer capacitors, EDLCs) is too limited to be suitable for high-energy applications. Another type of SC, the pseudocapacitor, stores charge through a redox transfer at the electrode surface, and devices of this type have been shown to improve capacitance.3 Recently, metal oxides (e.g., RuO2, MnO2, NiO, and CuO2) and conducting polymers (e.g., polyaniline, polypyrrole, and polythiophene) have been intensively investigated as promising pseudocapacitive materials.1 Despite the higher capacitances of pseudocapacitors than those of EDLCs, their rate and cyclic performances can be degraded because of the relatively low electrical conductivity and (electro)chemical and mechanical stabilities.3 Thus, emerging pseudocapacitors that can resolve the aforementioned challenges should be exploited and, from both economic and environmental viewpoints, need to be both abundant and renewable for future industrial applications.

Bioinspiration, which is inspired by biological structures and functions in nature, is a platform technology that aims to achieve breakthroughs in surpassing the limitation of existing technology as a consequence of synergistically combining the tools of biology and nanotechnology.4 In biology, the redox functions of biomaterials are modulated in energy storage and harvest (e.g., respiration and photosynthesis) and signal transduction (e.g., neuronal action potential).5 The well-defined redox active centers found in plants, humus, bacteria, and animals are phenolics that reversibly transfer electrons and protons in a biologically programmed manner.6 As pseudocapacitive materials for bioinspired redox charge storage, we chose lignins (Ligs) that are the major components of vascular bundles with tubular structures and show a lattice structure consisting of aromatic carbons.7 In particular, Ligs and their derivatives are expected to be emerging renewable and inexpensive pseudocapacitive materials because they are biopolymers obtained from a by-product of paper processing.8

On the surface of carbon nanomaterials, such as carbon nanotubes and graphenes, adsorption and crystallization of small molecules, oligomers, polymers, and biomaterials are driven by van der Waals, π–π stacking, and electrostatic interactions,9 which offer significant insights to resolve key challenges (e.g., dispersion and separation of carbon nanomaterials) and modify physical, chemical, and biological properties.10 Considering the previous breakthroughs achieved by the nanoscale hybridization of pseudocapacitive and EDLC materials,1 a renewable hybrid electrode (RHE) consisting of Lig confined on reduced graphene oxides (RGOs) would overcome the limitations of SCs as a consequence of the unique interactions of electroactive biopolymer nanocrystals with the electron-conducting graphenes.

The goals of the chemical approach delineated herein are: 1) to synergize both features of electroactive Lig (used as pseudocapacitors) and highly electron-conducting RGOs (used as EDLCs) for renewable energy-storage materials with high energy and power and 2) to manipulate the nanoscale confinement of Ligs that interact strongly with RGOs for a fast redox charge transfer. To verify the pseudocapacitive charging in different modes of EDLCs and LIBs, the capacitive mechanisms of Lig nanocrystals confined on the RGO surface were investigated using both spectroscopic and electrochemical methods.

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

A series of Ligs hybridized with RGO nanosheets, which will henceforth be abbreviated as RL-X (where X denotes the weight percentage of Lig in the RHEs) were prepared by a straightforward self-assembly method (Figure 1 a). The Lig content confined in RHEs and their interplay were investigated by X-ray photoelectron spectroscopy (XPS), FTIR, and UV/Vis analyses (see the Supporting Information, Figures S1–S3). After the conversion of exfoliated graphene oxides (GOs) into RGOs, they can undergo irreversible aggregation because of the restacking of individual sheets.11 Notably, RLs with >40 wt % of Ligs revealed a stable black dispersion without any agglomeration, indicating that good processibility was attributed to the interaction with Ligs (see the Supporting Information, Figure S4). RLs with >60 wt % of Ligs were stable in dispersion for more than one month at room temperature. Even at the same composition of RL-60, the mixture of RGOs and Ligs with the mass ratio of 6:4, which was not fabricated by self-assembly but by physical mixing, showed less dispersibility than self-assembled RLs in deionized (DI) water. The chemical structure of Lig is composed of large aromatic rings and C[BOND]C bonds, together with various functional groups, such as thiol, sulfonate, carboxylate, and phenolic hydroxyl moieties.7 Accordingly, the strong π–π and hydrophobic interactions between the aromatic backbone of Lig and RGO sheets play an essential role in achieving a strong binding between them. Afterwards, the dispersion of RLs in DI water remained stable due to the functional groups with different pKa values provided by Ligs, which led to an electrostatic repulsion for suppressing the restacking between the RGO sheets.12 The morphologies of RLs were demonstrated by SEM analysis (see the Supporting Information, Figure S5). The agglomeration part of RGO in RLs gradually decreased as Lig content increased, which clearly attests to the good exfoliation of RGO when hybridizing with Lig.

thumbnail image

Figure 1. (a) Energy-storage mechanism for RLs originating from natural polymers. High-resolution XPS spectra of the O 1s spectral feature of RL-60 (b) before and (c) after charging.

Download figure to PowerPoint

As shown in field-emission (FE)-TEM images of RLs before reduction (see the Supporting Information, Figure S6 a), amphiphilic GO sheets were well-covered by Ligs, which were exfoliated into a single layer or a few layers, through π–π stacking interactions between the conjugation of GOs and aromatic carbons of Ligs and hydrogen bonding interactions between various functional groups of GOs and Ligs. After reduction into RGO, the insulating GO-Ligs were converted into electron-conducting RLs. Because π–π and hydrophobic interactions were prominent during reduction, the polymer chains of Ligs tethered by polar functional groups, which interact with the defects of GO, were located far from the hydrophobic surface of RGO. As a result, the crystalline domains of Ligs interacting with the conjugation of RGO were sparsely isolated through self-assembly and randomly distributed as discrete nanocrystals, whereas RGOs were restacked into a few layers (Figure 2 a and b). The interlayer d-spacing in the lattice of Lig is about 0.3 nm, which corresponds to the (002) interplanar distance of graphitic carbon in the direction perpendicular to the hexagonal graphene layers (see the Supporting Information, Figure S6 b).13 Furthermore, the nanocrystalline structure of Lig was proven by the mixed ring- and spot-like fast Fourier transform (FFT) pattern, which was obtained from the designated area of Lig (Figure 2 c). Although further study is needed to understand the formation mechanism of nanocrystalline Lig, the hybrid structure of RLs, in which the Ligs were isolated as a discrete nanocrystal and strongly confined on the surface of RGOs by specific interactions, is expected to be suitable for electrochemical reactions that require the homogeneous distribution of discrete electroactive nanomaterials with a large surface area and favorable interactions with the conductive substrate.

thumbnail image

Figure 2. (a) FE-TEM image of RL-60. Red arrows denote Lig nanocrystals confined on RGO. (b) Close-up of the individual nanocrystal and (c) FFT patterns of RL-60. (d) UV/Vis spectra of RGO and fluorescence spectra of Lig in the range of 400 to 650 nm. The concentrations of RGO and Lig were 0.5 mg mL−1. (e) Fluorescence spectra of Lig, RL-99, and RL-95. The concentrations of Lig, RL-99, and RL-95 were 0.5 mg mL−1. The excitation wavelength was 385 nm.

Download figure to PowerPoint

Raman, UV, and photoluminescence (PL) spectra of RLs were analyzed to address the underlying interactions between RGO and Lig (Figure 2 d and e; also see the Supporting Information, Figure S7). Because chemical doping or charge injection changes the Raman bands of graphene,14 the charge (i.e., electron or hole)-transfer interaction of RLs can be verified by Raman spectroscopy. In particular, graphene is known to be p-doped with electron-withdrawing groups (EWGs) of organic or inorganic molecules.9, 15 Accordingly, Lig is anchored onto the RGO sheet through combined π–π and hydrophobic interactions without disrupting the intrinsic structure of RGO, so then the charges can be transferred from the EWGs of Lig (i.e., sulfonate and carboxylate moieties) to RGOs. The hypsochromic shift of the G-band is induced by interactions with EWGs,16 which could influence a redox charge transfer of Lig. We observed the increase of the full width at half-maximum (FWHM) of the G-band (see the Supporting Information, Figure S7 b) and the decrease of the ratio of D-peak intensity to G-peak intensity (ID/IG) when the Lig content increased (see the Supporting Information, Figure S7 c).17 The charge transfer between RGO and Lig was further substantiated by UV/Vis and fluorescence spectra (Figure 2 d and e). The strong interactions of RLs for charge transfer were evidenced by a broad spectrum overlap between the absorption band of RGO and the emission band of Lig (Figure 2 d).18 As shown in the dramatic PL quenching and decay of RL-95 and RL-99 even with 5 and 1 wt % of RGO (Figure 2 e), RLs can show a direct charge transfer arising from strong interactions between RGOs and Ligs.

The fundamental electrochemical behaviors of RHEs were investigated using cyclic voltammetry (CV) analysis in a three-electrode configuration (Figure 3). Ag/AgCl and Pt wire were used as the reference and counter electrodes, respectively, in aqueous HClO4 (0.1 m). The CV spectrum of RGO in the potential range of 0.2–0.8 V was nearly featureless, having a rectangular shape, indicating an ideal electrochemical double-layer capacitive feature.3 Notably, a reversible redox wave of RLs was observed at 0.52 V of the formal potential. As verified by XPS analysis before and after charging (Figure 1 b and c), the change in the chemical circumstance of quinone moieties is attributed to the specific interaction with protons through the redox reaction of RHEs (Figure 1 a). The reversible faradaic reaction of RLs is illustrated in Equation (1):

  • equation image(1)
thumbnail image

Figure 3. (a) CV profiles of RL-60 at scan rates ranging from 1 to 100 mV s−1 in a three-electrode configuration. (a1) Plots of peak current versus the scan rate of RL-60 for charge (R2=0.9974) and discharge (R2=0.9973) curves. (b) CV spectrum at 1 mV s−1 with capacitive charge-storage elements (hashed area) of RL-60 in a three-electrode configuration. (c) Specific capacitances of RL-60 including the capacitance from capacitive elements (hashed area) and the insertion process derived from the Dunn method (at scan rates of 1, 10, and 100 mV s−1) and the Trasatti method. (d) Cathodic peak shift and peak separation of RL-60 with scan rate in three-electrode configuration. (e) CV curves of RLs measured at a scan rate of 10 mV s−1 in a three-electrode configuration.

Download figure to PowerPoint

where, in aqueous acidic solutions, quinone (Q)/hydroquinone (QH2) of Lig confined on RGO can store and release two electrons and protons during charging/discharging, respectively. Despite having the redox charge transfer in common, pseudocapacitance can be achieved in a different manner from charge storage by batteries; pseudocapacitors are kinetically facile for a surface-dominant redox reaction and show neither phase transitions (i.e., structural rearrangements) nor diffusion-controlled insertion on the time scale, resulting in distinct performances.19 Thus, the predominant pseudocapacitive nature of Lig moieties, corresponding to a reversible oxidation/reduction transition of Q/QH2 states in RHEs,20 can be based on the following reasons.

1) As proven by the linear dependence of capacitive peak currents on scan rates,21 the surface-confined electrochemistry of RHEs implies that pseudocapacitive charging/discharging was triggered by a reversible redox transition of electroactive quinone moieties confined on the surface of RHEs (Figure 3 a and a1). The surface confinement of Lig nanocrystals on RHEs was verified by TEM analysis. This surface redox chemistry of RHEs was promoted by the transfer of charges generated by the electroactive quinone moieties on the electrode surface and the cooperative interplay between RGO and Lig.

2) The charge storage of RHEs is dominated by surface-confined pseudocapacitance rather than by insertion and EDLC mechanisms. It is well known that the total charge is electrochemically stored by faradaic contributions from diffusion-controlled insertion and/or charge transfer processes with electroactive moieties at the surface or in the interlayer lattice plane (i.e., pseudocapacitance), as well as the nonfaradaic contribution from double-layer capacitance.22 The respective contributions of capacitive (k1v) and insertion (k2v1/2) mechanisms to the total stored charges can be quantitatively evaluated by Equation (2) as suggested by Dunn et al:23(2)

  • equation image(2)

where i(V) is the current at a fixed potential, v is the scan rate, and k1 and k2 are determined by the slope and y intercept values, respectively, when i/v1/2 is plotted against v1/2. In Figure 3 b, 89.8 % of the total charge was charged by both types of capacitive mechanisms (pseudocapacitance and double-layer charging), whereas only 10.2 % of the overall capacitance was achieved by the insertion mechanism. This result was in good agreement with those obtained at high scan rates (10 and 100 mV s−1) and from the Trasatti method (Figure 3 c; also see the Supporting Information, Figure S8).24

3) The surface-confined pseudocapacitive behavior of RLs was also corroborated by examining the changes in peak voltage and current shifts with respect to scan rates (Figure 3 d; also see the Supporting Information). As a consequence of the slow kinetics of the diffusion-controlled insertion mechanism, CV spectra can show large peak-voltage shifts and peak separation as well as slow current response for cathodic and anodic currents.23 Consequently, the slight changes in cathodic peak shift, small peak separation, and relatively fast current response of RL-60 indicate that total capacity is governed by a surface-dominant charge-transfer reaction showing capacitive behavior rather than insertion of ions.25

To further understand the capacitive behavior of RHEs, we investigated the influence of the chemical composition of RLs on capacitive behaviors using CV and galvanostatic charge/discharge (GCD) analyses (Figures 3 e and 4 a). In CV spectra of RHEs at a scan rate of 10 mV s −1, the specific capacitances of RLs (per total mass of RLs) monotonically increased with respect to Lig content (see the Supporting Information, Figure S9), reached a maximum of 432 F g−1, which is close to the theoretical capacitance of 482 F g−1 (see the Supporting Information) and sixfold higher relative to that of RGO (93 F g−1) at 60 wt % of Lig. Specific capacitance decreased as Lig content increased further. Such a trend of capacitance depending on Lig content was in good agreement with the values measured by GCD curves at a constant current of 1 A g−1 (Figure 4 a). The GCD profiles of RLs with long duration times and small internal resistances (IR drop) elucidate a fast and reversible pseudocapacitive behavior due to their excellent electrical conductivity and fast ion transport. By contrast, RGO revealed symmetric triangle-shaped profiles of EDLC behavior as seen in a rectangular CV spectrum. Two voltage slopes of RLs were observed at a transition of about 0.4 V in the discharge curves, which indicates that two modes of electrochemical charging/discharging were triggered by the RGO and quinone moieties (see the Supporting Information). This finding agreed with CV results (Figure 3 e) and previous reports.20 The contributions of RGO (CRGO based on the mass and slope of RGO) and Lig (CLig based on the mass and slope of Lig) to the overall capacitance (Coverall) of the RHEs can be decoupled from two distinct slopes of linear lines in the discharge curves (Figure 4 b and c; Table 1). As verified by the greater value of CLig relative to CRGO, Coverall of RHEs and even RL-20 with a small Lig content were dominated by pseudocapacitive CLig rather than the double-layer capacitive CRGO (Table 1). As the content of Lig increased from 20 to 60 wt % at a specific current of 0.5 A g−1, the ratio of CLig to CRGO was enhanced from 1.72 to 3.32. At the higher specific current of 20 A g−1, the ratio of CLig to CRGO in RHEs was lower than that at 0.5 A g−1 because the pseudocapacitance of Lig was not fully manifested at the high rate as a consequence of its slower kinetics rather than double-layer capacitance.26 As Lig content increased up to 80 wt %, the internal resistance of RLs calculated from the IR drop increased due to the low electrical conductivity of Lig (see the Supporting Information, Figure S10).27 Even bulk (or pristine) Ligs showed negligible capacitance (Figure 3 e). As discussed in the pseudocapacitance of RHEs, the highest specific capacitance of RL-60 was mainly attributed to the maximum utilization of pseudocapacitance arising from the redox chemistry of quinone. Considering the insulating property of Lig itself, by contrast, the performance decay of RHEs at the higher Lig content (i.e., RL-80) was due to the decline in the electrical conductivity of the electrodes and thus, pseudocapacitance was not fully revealed. Consequently, the capacitance of RHEs was greatly enhanced by synergizing the bioinspired redox electrochemistry of surface-confined quinone and the interplay with electron-conducting RGOs.

thumbnail image

Figure 4. (a) GCD curves of RLs measured at a constant specific current of 1 A g−1 in a three-electrode configuration. Respective specific capacitances of RGO (CRGO), Lig (CLig), and both (RGO+Lig, Coverall) decoupled from the slopes of the discharge GCD curve in a three-electrode system with (b) specific currents (for RL-60 only). (c) Respective specific capacitances of RGO (CRGO), Lig (CLig), and both (RGO+Lig, Coverall) decoupled from the slopes of the discharge GCD curve in a three-electrode system with Lig content (at 1 A g−1) in a three-electrode configuration.

Download figure to PowerPoint

Table 1. Overall specific capacitances and CLig/CRGO of RLs at 0.5 and 20 A g−1 and their rate capabilities in a three-electrode configuration.
Material0.5 A g−120 A g−1Rate capability
 Coverall [F g−1]CLig/CRGOCoverall [F g−1]CLig/CRGO[%]
RL-202011.701661.2682.5
RL-402582.142301.7189.3
RL-603793.322801.9874.0
RL-801552.35831.2653.8

To more accurately evaluate the SC performances of RHEs, such as the rate and cyclic capabilities for practical applications, GCD curves were obtained in a symmetric two-electrode configuration. The rate capability of RHEs was examined by varying the specific currents of GCD profiles from 0.5 to 30 A g−1 (Figure 5 a). The specific capacitance of RL-60 at 1 A g−1 was determined to be 203 F g−1, which is higher than that of RGO (80 F g−1). As specific current increased, the ratio of CLig to CRGO decreased, which means that the decrease in Coverall was attributed to the reduction of CLig, not to the improvement of CRGO (Figure 4 b and Table 1). In other words, CLig decreased, whereas CRGO was nearly constant at the high current. This finding is related to typical electrochemical behavior, in that EDLCs deliver charges faster than pseudocapacitors due to the discrepancy of the intrinsic charge-storage mechanism and the higher electronic conductivity of the former, especially for this system. Nevertheless, the rate capability of RHEs was reasonable in spite of pseudocapacitors; RHEs showed >75 % retention even at a high current (30 A g−1). In particular, RL-60 achieved the largest capacitance values over the entire current range; the specific capacitance of 161 F g−1 for RL-60 was two-fold greater than that of RGO (70 F g−1), even at the highest current of 30 A g−1. The corresponding capacitance retention of RHEs is a fairly good result relative to pseudocapacitors previously reported.27, 28 As shown in the Nyquist plot of Figure 5 b, the reduced charge-transfer resistance (RCT), which is attributed to the favorable interplay of RGO and Lig, promotes the redox reaction for improved pseudocapacitance.

thumbnail image

Figure 5. (a) Rate capability of RLs at the specific currents of 0.5 to 30 A g−1 in a two-electrode configuration. (b) Nyquist plots for RGO and RL-60 in a two-electrode configuration. Inset: magnification in the high-frequency region. (c) Cycle stability (open symbols) and coulombic efficiency (closed symbols) of RL-60 (inverted triangle) and RGO (square) at a constant current of 1 A g−1 in a two-electrode configuration over 3000 cycles.

Download figure to PowerPoint

After 3000 cycles of charging/discharging at 1 A g−1, RL-60 retained 96.1 % of initial capacitance with negligible capacitance attenuation, which was comparable to an initial capacitance retention of 94.0 % for RGO (Figure 5 c). Additionally, the coulombic efficiencies of both RL-60 and RGO in a two-electrode configuration were nearly preserved. Given that most pseudocapacitors suffer from poor cycle life, which is considered as a critical challenge, RHEs are as durable as an EDLC-type SC (RGO herein) while maintaining their good capacitances. Notably, the excellent capacitive characteristics of RHEs can be attributed to both the synergizing features of the redox chemistry of surface-confined quinone and the strong interplay between Lig and RGO.

The Ragone plots displayed in Figure 6 provide information about the high specific power (P) with acceptable specific energy (E) of RL SCs. In particular, RL-60 can deliver an E value of about 10 Wh kg−1 and a P value of about 40 kW kg−1, which are close to those of state-of-the-art pseudocapacitors3, 11, 27, 28b, 29 and EDLCs30 (Figure 6). The high E and P values demonstrate the synergistic performance of RHEs as a result of confinement of Lig nanocrystals and the favorable interplay between RGO and Lig, which leads to fast and reversible pseudocapacitance and long-term stability.

thumbnail image

Figure 6. Ragone plot of RL-60 with data previously reported in the literature (numbers in figures denote references).

Download figure to PowerPoint

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

We have demonstrated the highly reversible and fast pseudocapacitive behavior of confined Lig nanocrystals for renewable energy-storage materials with high power and energy. The unique pseudocapacitive behavior was comprehensively verified using both spectroscopic and electrochemical methods, which can be further applied for the fundamental understanding of energy-storage mechanisms in new systems. The synergistic features of pseudocapacitors were achieved by manipulating the bioinspired redox electrochemistry of surface-confined quinone and the favorable interplay of hybrids. Consequently, hybrid electrodes with good rate and cyclic performances showed a maximum capacitance of 432 F g−1, which is close to the theoretical capacitance of 482 F g−1 and sixfold higher relative to that of RGO (93 F g−1). Therefore, this research provides a deep understanding of the way hybrid nanostructures are integrated and synergized, which can greatly expand the utility and versatility of this innovative approach for the design and synthesis of emerging and advanced energy-storage materials.

Experimental Section

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

Synthesis of RLs

GO was prepared from natural graphite flakes (powder, <20 μm, Aldrich) using a modified Hummers method, the details of which were described elsewhere.31 RGO–Ligs were prepared by simple mixing of RGO and softwood sodium lignosulfonate (low sulfonate content with low sulfonate content, Mw≈10 000, Aldrich) at different weight ratios of 10:0 (RGO), 8:2, 6:4, 4:6, and 2:8. The following procedure was used for the preparation of RL-60. Other RLs were prepared using the same procedure except the weight ratios of RGO to Lig. GO (10 mg) was fully exfoliated in distilled water (20 mL, ca. 0.5 mg mL−1), and then Lig (15 mg) was added to the solution. A hydrazine solution (200 μL, 35 wt % in water, Aldrich) was readily added. The resultant stable suspension was placed in an oil bath thermostated at 60 °C for 12 h, yielding a homogeneous black dispersion. For other analytical characterization, the solution was filtered through an anodic alumina oxide (AAO) membrane filter (47 mm diameter, 0.2 μm pore size, Whatman) under vacuum.

Electrochemical measurements

Three- and two-electrode configurations were employed to evaluate electrochemical characteristics. The resulting RL solution was used to prepare their cast films on thin Au-film-coated poly(ethylene terephthalate) (PET) substrates as a current collector for electrochemical measurements. The suspension of RLs was dropped onto the Au-coated PET substrates. The RL coated electrode was dried in an oven at 80 °C for 12 h. An aqueous HClO4 solution (0.1 M) was used as an electrolyte. In the case of three-electrode systems, an Au-coated PET substrate was used as a working electrode, Ag/AgCl as a reference electrode, and Pt wire as a counter electrode. For two-electrode systems, an aqueous electrolyte-soaked separator (Whatman glass microfiber filter) was sandwiched between two electrodes. As-obtained devices were dried at room temperature prior to measurements. Electrochemical characteristics were evaluated by CV analyses using a CHI 760D electrochemical workstation (CH Instruments) and GCD analyses using an Iviumstat.XR electrochemical interface and impedance analyzer (Ivium Technologies). CV measurements were performed at different scan rates (1–100 mV s−1). GCD measurements were performed at different specific currents (0.5–30 A g−1). In GCD profiles, the specific capacitance in a two-electrode configuration was estimated using Equation (3):31(3)

  • equation image(3)

where I is the current applied, ΔVt is the slope of the discharge curve after the IR drop at the beginning of the discharge curve, and m is the average mass of two electrodes. In a two-electrode configuration, electrochemical impedance spectroscopy (EIS) measurements were performed over a frequency range from 106 to 10−2 Hz at a sinusoidal voltage amplitude of 10 mV using a Solartron 1260 impedance/gain-phase analyzer. Cyclic life tests were carried out by GCD cycles at a constant specific current of 1 A g−1 for 3000 cycles. P and E values were calculated using Equation (4):32(4), (5)

  • equation image(4)
  • equation image(5)

where ΔV, R, and C are the potential window of discharge, internal resistance from IR drop, and measured capacitance of both electrodes, respectively. The internal resistance was computed from the voltage drop at the beginning of each discharge:(5)

  • equation image(5)

where ΔVer. and i are the voltage drop between the first two points in the voltage drop at the top cut-off and the applied current, respectively.

Characterization

TEM images were collected on a JEM-2100F microscope (200 kV, JEOL). Raman spectra were recorded from 3500 to 100 cm−1 on a Bruker FT Raman spectrophotometer RFS 100/S using a 785 and 1064 nm dual-channel laser at a resolution of 1 cm−1. UV spectra were collected using a UV/Vis spectrometer (CARY 300 Bio, Varian). Fluorescence spectra were recorded using a QuantaMaster 40 spectrofluorometer (PTI) at an excitation of 385 nm. XPS data were obtained using a VG Multilab 2000 system (Thermo VG Scientific) with a monochromatic Mg KR X-ray source (=1253.6 eV) under a vacuum-analysis chamber (10−7 Torr). The high-resolution scans of C and low-resolution survey scans were analyzed for each sample at, at least, two separated locations. FTIR spectra of dried membranes and powder samples were recorded in the attenuated total reflectance (ATR) mode in the frequency range of 4000–650 cm−1 on a Nicolet 6700 instrument (Thermo Scientific, USA). Spectra were recorded as the average of 32 scans with a resolution of 8 cm−1. Each sample was put in equal physical contact with the sampling plate of the spectrometer accessory to avoid differences caused by pressure and penetration depth. The concentration of phenolic group in Lig was obtained using the periodate oxidation method as reported in the literature.33 To a solution of Lig (0.4 g) in cold water (6 mL) was added sodium periodate (0.8 g) and acetonitrile (internal standard) with stirring. The mixture was stored in the refrigerator (4 °C) and analyzed periodically for the quantitation of formed methanol by GC using an Agilent 7890N (FID detector) instrument, equipped with a HP-Innowax column (Agilent, 30 m×0.25 mm×25µm). Methanol formation leveled off after 2 d and methanol formation was 1.36 mmol per gram of Lig.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

This work was supported by the Energy Efficiency & Resources of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Knowledge Economy (No. 20122010100140), the National Research Foundation (NRF) funded by the Korean Government (MEST) (20090063004) and NRF-2010-C1AAA001-0029018, and Korea Institute of Science & Technology (KIST) institutional program. S.B.L. was supported (discussions on electrochemical analysis) as part of Nanostructures for Electrical Energy Storage (NEES), an Energy Frontier Research Center funded by the US Department of Energy, Office of Science, Office of Basic Energy Sciences under award number DESC000116.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
cssc_201301061_sm_miscellaneous_information.pdf837Kmiscellaneous_information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.