Achieving 100% Utilization of Reduced Graphene Oxide by Layer-by

Feb 26, 2014 - Achieving 100% Utilization of Reduced Graphene Oxide by Layer-by-Layer Assembly: Insight into the Capacitance of Chemically Derived Gra...
0 downloads 13 Views 2MB Size
Article pubs.acs.org/JPCC

Achieving 100% Utilization of Reduced Graphene Oxide by Layer-byLayer Assembly: Insight into the Capacitance of Chemically Derived Graphene in a Monolayer State Zhongwei Lei,† Takahiro Mitsui,† Hiroki Nakafuji,‡ Masayuki Itagaki,‡ and Wataru Sugimoto*,† †

Faculty of Textile Science and Technology, Shinshu University, 3-15-1 Tokida, Ueda, Nagano 386-8567, Japan Faculty of Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan



S Supporting Information *

ABSTRACT: Face-to-face restacking is one of the main reasons for low surface utilization of multilayered graphene. In this study, multilayered reduced graphene oxide/polymer architecture was fabricated by gas phase or chemical reduction of thin films composed of graphene oxide monolayers using poly(diallyldimethylammonium) as the cationic binder deposited by layer-by-layer self-assembly. The electrochemical behavior of the thin films in acidic and neutral electrolytes was investigated by using cyclic voltammetry and electrochemical impedance spectroscopy. A transmission line model was adopted to simulate the electrochemical impedance data. The electrochemical data were analyzed and deconvoluted into charge storage due to non-Faradaic electrical double layer capacitance and pseudocapacitance arising from Faradaic surface redox reactions. Pseudocapacitance observed in acidic electrolyte is proportional to the amount of surface functional groups. An overall volumetric capacitance as high as 364 F cm−3 was achieved for the nanoarchitecture, and it is shown that the electrical double layer capacitance of a monolayer of graphene oxide is 20 μF cm−2, regardless of the number of layers deposited. This can be interpreted as full capacitive utilization of reduced graphene oxide sheets in the multilayered reduced graphene oxide films.

1. INTRODUCTION Graphene has recently attracted much interest as a promising electrode material for electrochemical capacitors owing to its theoretically high specific surface area and good electronic conductivity.1−7 A number of studies have been reported for the fabrication of electrochemical capacitors with use of graphenebased materials as the electrode.8−13 However, the achieved capacitance of all these studies was far from the maximum theoretical value of ∼550 F g−1 calculated based on the theoretical surface area of 2630 m2 g−1 of graphene and the commonly accepted electrical double layer capacitance value of ∼20 μF cm−2 for electrodes in general.9,14,15 Naturally, one of the reasons that the theoretical capacitance cannot be achieved is that the surface area of graphene-based electrodes is smaller than its theoretical value due to restacking. In addition, the capacitive performance of graphene-based materials is known to be affected by a number of factors, such as the degree of exfoliation (number of graphene layers),16,17 lateral size of individual sheets,18,19 defects,2,20 degree of oxidation,21 and so on.22 Such factors have inhibited characterization of the intrinsic properties of these materials. The capacitance of high-quality graphene obtained via mechanical cleavage or chemical vapor deposition has pointed out the importance of quantum capacitance in addition to the classical electric double layer capacitance.23−25 The quantum capacitance is influenced by a number of parameters including © 2014 American Chemical Society

the number of graphene layers, impurities, and defects. These well-defined model electrode studies have shown that the quantum capacitance of monolayer graphene may dominate that overall charge storage of such devices. In contrast to these single graphene measurements of physically derived graphene, much of the work on graphene-based electrochemical capacitors is reported using chemically derived graphene, i.e., graphene derived by reduction of exfoliated graphite oxide nanosheets. The chemically derived graphenes have disadvantages compared to physically derived graphene, including higher impurity and defect content, as well as restacking during the reduction process. Nonetheless, advantages including higher surface area, higher volumetric density, and scalability allow chemically derived graphene to be more practically feasible for energy storage applications. The complexity of electrodes constructed from chemically derived graphene has so far hindered the understanding of the fundamental capacitive behavior of chemically derived graphene, particularly in a monolayer state. Thus, a simplified model electrode study with well-defined, countable graphene layers should contribute to the understanding of the capacitive behavior of chemically derived graphene. Received: December 23, 2013 Revised: February 26, 2014 Published: February 26, 2014 6624

dx.doi.org/10.1021/jp412570s | J. Phys. Chem. C 2014, 118, 6624−6630

The Journal of Physical Chemistry C

Article

We chose to adopt the electrostatic layer-by-layer (LbL) selfassembly technique, a widely used approach for fabrication of well-defined and controlled thin films.26 By alternately depositing positively and negatively charged nanomaterials, one can readily fabricate multicomponent materials in a controllable manner. As LbL affords well-defined thin films of manipulable and countable alternating layers of graphene oxide, we anticipated that the LbL self-assembly approach to fabricate mono- to multilayered graphene oxide thin films would serve as an ideal model system to achieve a spaced nanoarchitecture of graphene electrodes for charge storage applications. LbL selfassembly has been utilized by several groups for fabrication of thin films consisting of graphene oxide or similar materials.27−33 These studies have proven that combining graphene with other electrode materials such as carbon nanotubes, conducting polymers, and metal oxides results in enhanced capacitive performance. Unfortunately, graphene is used merely as a conducting agent or a conductive support in these composites, not differing much from other traditional carbonaceous materials. Here, we report the fabrication of thin films composed of monolayered reduced graphene oxide suitable for characterizing the capacitance of a single layer of graphene. The LbL technique was employed to fabricate thin films composed of monolayered graphene oxide (GO) and a polycation, followed by reduction of GO to obtain reduced graphene oxide (rGO). The polycation is anticipated to act as a polymeric binder preventing individual rGO from restacking and allow penetration of electrolyte, thus making full use of its surface for application as electrodes for electrochemical capacitors. A series of thin films with 1 to 10 monolayers of rGO were investigated in detail using cyclic voltammetry and electrochemical impedance spectroscopy in acidic and neutral electrolyte in order to elucidate the fundamental capacitive behavior of the multilayered rGO electrodes.

The deposited films were reduced either by gas-phase or chemical reduction. Gas-phase reduction was conducted by thermal treatment at 200 °C under H2/N2 = 1/9 flow for 2 h. Chemical reduction was conducted by dipping (PDDA/GO)n films in an aqueous solution of 0.1 M N2H4 (Wako Pure Chemical) for 24 h at 60 °C, followed by washing with ultrapure water. The H2- and N2H4-reduced samples will be denoted (PDDA/rGO)n-H2 and (PDDA/rGO)n-N2H4, respectively. The surface morphology of (PDDA/GO)1 deposited on Si wafer was observed with atomic force microscopy (AFM, SPM400, Seiko Instruments Inc.). UV−vis spectroscopy (U-4100, Hitachi High-Technologies Corp.) was employed for characterization of the amount of GO deposited on quartz glass by each deposition cycle. X-ray photoelectron spectrometry (XPS, AXISULTRA DLD, Kratos Analytic Ltd.) was conducted for chemical analysis of powder samples of rGO synthesized by drying colloidal GO at 60 °C, and reduced accordingly. Cyclic voltammetry (HZ-5000, Hokuto Denko Corp.) was carried out with a three-electrode beaker cell, in which the working electrode was (PDDA/rGO)n (n = 1−10) deposited on one side of a flat gold substrate of 1 × 1 cm2 with the back side masked. An Ag/AgCl (sat. KCl) electrode was used as the reference electrode, and a Pt mesh was employed as the counter electrode. A sample of 0.5 M H2SO4 or Na2SO4 was used as the electrolyte. Potential scans were conducted between 0.2 and 1.2 V vs RHE at 500, 200, 50, 20, 5, and 2 mV s−1 after 500 break-in cycles at 50 mV s−1. Electrochemical impedance spectroscopy (1287 Electrochemical Interface, 1255B Frequency Response Analyzer, Solartron) was carried out after cyclic voltammetry. A Pt wire was used as the reference electrode in order to minimize IR loss at high frequency. Impedance measurements were conducted at 0.2, 0.4, 0.6, 0.8, 1.0, and 1.2 V vs RHE by sweeping the frequency from 100 kHz to 10 mHz with an amplitude of 5 mV.

2. EXPERIMENTAL SECTION Graphite oxide was prepared by oxidization of graphite powder (Z-5F, ITO GRAPHITE Co., Ltd.), following the Hummers method.34 Briefly, graphite powder was mixed with NaNO3 and added into concentrated H2SO4. KMnO4 was added slowly into the slurry with stirring and aged for 4 days at room temperature. After completion of the oxidization of graphite, water was added to the slurry, which was then filtered and washed thoroughly with diluted HCl and methanol in sequence. The obtained brown matter (graphite oxide) was dried overnight and ground to a powder state. Graphite oxide (20 mg) was dispersed in 20 mL of ultrapure water (>18 MΩ cm) and exfoliated with the assistance of mild ultrasonic treatment. Colloidal monolayer graphene oxide was obtained by removal of nonexfoliated graphite oxide via centrifugation (2000 rpm for 30 min). Poly(diallyldimethylammonium chloride) (PDDA, 20 wt % aqueous solution, Aldrich) was used as the counter polycation to electrostatically deposit GO layer-by-layer. The LbL deposition was conducted with a DC 4200 dip-coating system (AIDEN CO., Ltd.). Briefly, the substrate (Au plate, quartz glass or silicon wafer) was dipped in a diluted PDDA solution (2 wt %) for 10 min, washed carefully with ultrapure water and dried in air, then dipped in colloidal GO (diluted to 0.2 g L−1) for 20 min, washed again with ultrapure water and dried in air. One cycle of this procedure gives a film composed of a monolayer graphene oxide, (PDDA/GO)1, and repetitive cycles afford multilayered graphene oxide films, (PDDA/GO)n with n = 2−10.

3. RESULTS AND DISCUSSION AFM images of (PDDA/GO)1 (Figure 1) revealed that the majority of GO nanosheets have dimensions in lateral size of several micrometers, in agreement with the particle size of the parent graphite particles (∼4 μm). Previous studies have shown that the edge effect on electrochemical capacitance of reduced graphene oxide is observed when the lateral size of nanosheets is below a few hundred nanometers.18,19 Hence, the edge effect in the present system can be neglected. The thickness of individual GO was ∼1 nm, indicating successful exfoliation, with a negligible amount of nonexfoliated graphite oxide or few-layer sheets. Under the film processing conditions applied, over 90% of the substrate was covered by GO, with