Electrochimica Acta 176 (2015) 880–886
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Insight on the effect of surface modification by carbon materials on the Ionic Liquid Electric Double Layer Charge Storage properties Renata Costa, Carlos M. Pereira* , A. Fernando Silva Faculdade de Ciências da Universidade do Porto, Departamento de Química e Bioquímica, CIQUP-Physical Analytical Chemistry and Electrochemistry Group Rua do Campo Alegre, s/n 4169–007 Porto, Portugal
A R T I C L E I N F O
A B S T R A C T
Article history: Received 22 April 2015 Received in revised form 30 June 2015 Accepted 30 June 2015 Available online 6 July 2015
Development of new and better performing energy storage devices rely on the use of innovative materials. The combination of nanostructured materials with high specific area, such as graphene, and conventional electrode materials such as Glassy Carbon (GC) and the use of the composite electrode in ionic liquids (IL) are considered to be a promising strategy for improved energy storage devices. Following our previous studies of electrochemical interfaces involving 1-butyl-3-methylimidazolium (tris(pentafluoroethyl) trifluorophosphate) ([C4MIM][FAP]) ionic liquid and Hg, Au, Pt and GC we extended the search for better electrochemical performance by preparing GC electrode surfaces modified with different carbon materials (reduced graphene oxide, reduced graphite oxide and graphite). Cyclic voltammetry of these electrode surfaces in [C4MIM][FAP] ionic liquid shows a 100 fold increase in the capacitive current of the composite electrode when compared with plain GC electrode. ã 2015 Elsevier Ltd. All rights reserved.
1. Introduction The interaction of ionic liquids with charged surfaces is still relevant for the understanding of the specific interactions on the molecular arrangement near surface region [1,2,3,4]. Ionic liquids enclose chemical and structural diversity with multiple molecular interactions. The complexity of ionic liquids at electrified interface stems from on the complex structure and nature of the ions [5], its purity [6] and is particularly dependent on the specific interactions between the electrode material and the IL [7,8]. The relationship between differential capacitance curves behavior with the EDL structure has become the centerpiece to many IL EDL models [9,10,11,12,13,14]. Several theoretical works have been performed in order to evaluate the influence of electrode roughness on the double layer formation in ionic liquids [15,16,17,18]. From the experimental point of view, it has been demonstrated that the nature of the electrode surface significantly affect the overall shape of the C(E) curves, and consequently, the arrangement of the ions at the electrified surface [19]. Very recently, a study aimed to assess the effect of the electrode surface structure roughness on EDL was published [20]. The authors compared a oriented surface of thin Au (111) film on mica with a polycrystalline
* Corresponding author. Tel.: +351 220402659; Fax.: +351 220402613 E-mail address:
[email protected] (A. F. Silva). http://dx.doi.org/10.1016/j.electacta.2015.06.142 0013-4686/ ã 2015 Elsevier Ltd. All rights reserved.
Au disc surface in contact with the non-aromatic pyrrolidinium (1butyl-1-methylpyrrolidinium bis(trifluoromethane) sulfonimide) ionic liquid and found that the electrode surface structure and roughness have weak effect on the fast process occurring on the double layer formation while having a strong influence on the slower processes. Despite the various studies reported in the literature, the influence of the electrode surface properties on the Electric Double Layer (EDL) structure is still only partially understood [2,21]. Models that do not accommodate the molecular-scale effects of ion geometry (e.g. finite volume of ions) and also the impact of the electrode surface nature and structure on the C(E) curves, will fail to give a realistic theoretical picture of the RTIL/electrode interface. Controlling the electrode nature and structure (topography and geometry) can be a route to obtain relevant information on the EDL as indicated by its fundamental property such as differential capacitance. Over the past decade, ionic liquids and graphene type materials have received a great deal of attention. Many experimental and theoretical works emerged focusing on the dispersion of carbon nanomaterials in ILs and on the importance of the intermolecular interactions between graphene and the IL [22]. Graphene is composed of a honeycomb arrangement of p-electron-rich carbon material, sp2– hybridized with two-dimensional nanostructure [23,24]. Graphene properties and characteristics have been widely described [25,26,27,28,29,30,31,32]. Recently, Bianco [33] et al. proposed a nomenclature to a more precise description of the graphene based
R. Costa et al. / Electrochimica Acta 176 (2015) 880–886
subject materials so that confusion and inconsistency would be minimized in order to get a better systematization of terms associated with materials based on graphene. The effect of surface modification with high specific area materials such graphene is commonly discussed in the literature and is of great importance for systems which require an electrode with high and more accessible surface area such as supercapacitors [34,35]. The microscopic interaction between graphene nanomaterial and ILs can be explained by short-range interactions involving ILs cation- p interaction of carbon nanomaterials and by the long range dispersion interactions. Very recently, Fedorov and Lynden-Bell have been probing the mechanisms of the interfacial layer formation of the neutral graphene/1,3-dimethyimidazolium chloride (ionic liquid) interface by molecular dynamics simulation [36]. Ivaništšev et al. [37] published a very interesting approach using a molecular dynamics-based simulation to obtain the free energy profiles in order to evaluate the influence of two ionic liquid solvents (1-butyl-3-methylimidazolium tetrafluoroborate and 1, 3-dimethylimidazolium chloride) on the interaction of dissolved ions with a graphene electrode surface. The capacitance-potential curves of the EDL obtained from a computational study performed by Paek et al. [38] on the interfacial structure of graphene and [C4MIM][PF6] IL were found to exhibit convex- or bell-shape. The MD simulations clearly demonstrate the distinct alternative layering of [C4MIM]+ and [PF6]- in the vicinity of an electrified graphene surface with the authors assigning particular importance of the quantum capacitance contribution to the total interfacial capacitance between graphene and [C4MIM][PF6]. In line with a previous work centered on the role of the electrode material on the IL EDL structure, this work was outlined in order to provide a relatively inexpensive way of improving the energy storage properties of electrochemical devices through the modification of a glassy carbon electrode with carbon materials in contact with an ionic liquid ([C4MIM][FAP]). Previous experience on the use of [C4MIM][FAP] for studies on electrified interfaces [19] indicate the suitability of this liquid since it demonstrates to have excellent hydrolytic stability, low viscosity and high electrochemical and thermal stability. The reported results underline the importance of combining high specific area materials such as reduced graphene oxide modified electrodes with distinctive ILs to achieve higher density currents and higher capacitances. Detailed experimental molecular information on the surface chemistry of graphene is limited, however this work contributes to highlight the importance of ionic liquid- reduced graphene oxide interactions at a charged and planar electrode. This may be very important to develop applications stemmed from surface interactions of high specific area carbon materials using ionic liquids as electrolytes for advanced electrochemical storage devices. 2. Experimental The hydrophobic ionic liquid [C4MIM][FAP] was purchased to Merck with the highest purity grade available (higher than 99 %). Graphite (Aldrich, 1-2 micron), hydrazine hydrate (Aldrich, 98 %), methanol (Merck, 99.9 %), sulfuric acid (Aldrich, 95-97 %), potassium chloride (Merck, 99.5 %), potassium permanganate (Merck), dimethyl formamide (DMF, Merck, 99.8 %) were all used without further purification. Graphite oxide was prepared from graphite powders by following a method described by Hummers and Offeman [39]. Briefly, 2.0 g of graphite powder was first added into 100 mL concentrated H2SO4 at room temperature. Under stirring, the mixture was cooled to 5 C using an ice bath, and the temperature of the mixture was kept below 5 C for 30 min. KMnO4 (8.0 g) was then added gradually under stirring and cooling. 100 mL distilled water was added into the mixture, stirred for 1 h, and further
881
diluted to approximately 300 mL with distilled water. After that, 20 mL of 30% H2O2 was added to the mixture to reduce the residual KMnO4. The solid was filtered, washed with 5% HCl aqueous solution (800 mL) to remove metal ions and with ultrapure water until the pH was 6. The resulting graphite oxide was dried at 30 C for 24 h. The chemical reduction of graphite oxide was adapted from reference [40]. The chemically reduced graphene oxide was prepared by were prepared by dispersing the graphite oxide with ultrapure water (3 mg/mL) to yield a yellow-brown suspension followed by sonication of this mixture in 1 L ultrasonic bath (Bandelin Sonorex digitec) for 3 h. The sonicated yellow-brown suspension was then treated with hydrazine hydrate, and the mixture was heated in an oil bath at about 95 C in a water-cooled condenser for about 12 h. The reduced graphene oxide was obtained as a black solid and was filtered and washed with deionized water (5 100 mL) and methanol (5 20 mL). Following this, the precipitate was dried at room temperature for about 24 h. The immobilization of graphene was preceded by the preparation of a dispersion of 10 mg of graphene particles in 1 mL of DMF followed by ultrasonication for about 6 h to facilitate the complete dispersion of rGO. Different volumes (0.5, 1.0 and 1.5 mL) of this dispersion were then spread on the glassy carbon electrode surface using a micropipette. The solvent evaporation was performed at room temperature followed by a more thorough drying step placing the electrode in a high vacuum line for a period of 12 h. The glassy carbon electrode was then ready to be immersed in the ionic liquid. IL purification procedure included the washing of the ionic liquid several consecutive times using ultra-pure water under stirring in order to dissolve and remove water soluble impurities. To reduce the water content to an acceptable minimum level, the IL was heated for several hours at 70-80 C under vacuum ( 10 Pa) and under continuous stirring. The final water content, according to Karl-Fisher titration (831 KF coulometer Metrohm) was below 30 ppm. Before each experiment all glass material was washed with concentrated sulfuric acid followed by abundant washing with ultra-pure water and finally with boiling ultra-pure water. The ultra-pure water was obtained by filtration through the purification system Milli-Q deionized water with a volume resistivity of not less than 18.2 MV cm The X-ray diffraction patterns were obtained in a Siemens D5000 X-ray difractometer and were carried out with Cu Ka radiation (l=1.54056) using an operating voltage of 40 kV, a step size of 0.01. This equipment consists of a theta/2theta diffraction instrument operating in the reflection geometry, focused by a primary Ge crystal monochromator. The detector is a standard scintillation counter. Surface area was measured using the BET Brunauer-EmmettTeller (BET) Analyser Micromeritics Tristar II analyzer through nitrogen gas adsorption–desorption isotherms at 77.3 K. Prior to measurements, the sample was out gassed at 80 C for 1 h and at 100 C for 5 h. Electrochemical experiments were carried out in a water jacketed three-electrode cell and the temperature was kept constant by the use of a thermostated bath. The experiments were performed inside a N2-filled glove box to prevent oxygen and moisture interference. The working electrode used was a GC electrode (Metrohm), the counter electrode was made of glassy carbon, and a silver wire was used as a quasi-reference electrode connected to an Autolab 302N. During the experiments, N2 was kept over the electrolyte without disturbing cell configuration. Capacitance and current intensities were normalized using the GC working electrode geometrical area (0.0314 cm2). Before each study, the electrode was polished with 1 micron diamond paste
882
R. Costa et al. / Electrochimica Acta 176 (2015) 880–886
Buehler during 2 minutes and washed with abundant Millipore water to remove any diamond paste trace remaining in the electrode surface. 3. Results 3.1. Carbon materials characterization Fig. 1 a) shows a SEM image of a flake of exfoliated hydrazinereduced graphene oxide with a transparent aspect and a slight thin wrinkled structure typically observed in graphene sheets [41]. The wrinkled and rippled structure may be a result of deformation upon the exfoliation process. SEM images in Fig. 1 b,c and d) were obtained after surface immobilization of the reduced graphene oxide flakes and point to the possibility of rGO flakes being overlapped or aggregated. Under this hypothesis the immobilization of rGO on the glassy carbon electrode will maintain a high fraction of electrode surface non-covered by the carbon material thus in direct contact with the electrolyte. The structural properties of graphite and the reduced material rGO (with and without exfoliation) were characterized by X-ray diffraction (XRD) and the XRD patterns are presented in Fig. 2. For the graphite powder, the XRD profile presents a sharp and intense peak at 2u = 26.6 which has been attributed to the (002) plane and is in agreement with the value found in the literature [42]. By analyzing the XRD pattern of rGO, after chemical reduction by hydrazine, the sharp (002) peak observed for graphite powder disappeared to give place to a broad peak with markedly lower intensity. For the rGO sample the intensity of the broad peak is lower than the reduced graphite oxide sample. This broad peak has been ascribed to the partial restacking of the graphene layers and may indicates disorder created during synthesis by modified Hummer’s method [43]. The BET surface area was obtained from the nitrogen adsorption–desorption isotherms shown in Fig. 3. The sample has a BET surface area of 52 5 m2/g and a pore volume of 0.00214 cm3/g measured by nitrogen adsorption at 77 K. The
Fig. 2. XRD patterns of graphite, reduced graphite oxide and reduced graphene oxide samples.
adsorption isotherm shows a slight adsorption in the low pressure region (
