poly(3,4-ethylenedioxypyrrole) enwrapped by reduced graphene oxide: how conduction behavior at...

Published: August 11, 2011

r 2011 American Chemical Society 18354 dx.doi.org/10.1021/jp205551k | J. Phys. Chem. C 2011, 115, 18354–18365

ARTICLE

pubs.acs.org/JPCC

Poly(3,4-Ethylenedioxypyrrole) Enwrapped by Reduced GrapheneOxide: How Conduction Behavior at Nanolevel Leads to IncreasedElectrochemical ActivityB. Narsimha Reddy,† Melepurath Deepa,*,† Amish G. Joshi,‡ and Avanish Kumar Srivastava‡

†Department of Chemistry, Indian Institute of Technology, Hyderabad, Yeddumailaram-502205, Andhra Pradesh, India‡National Physical Laboratory, Dr. K. S. Krishnan Road, New Delhi-11001, India

bS Supporting Information

1. INTRODUCTION

Conducting polymers are potentially useful for a variety ofnew electronic devices owing to ease of processability, highelectronic conductivity, and cost effectiveness. The family ofpoly(alkylenedioxypyrrole)s in particular has aroused muchattention due to a dramatically lowered oxidation potentialamong p-doped polymers, inherently large electrical conductivity,good environmental stability, and low band gap.1�3 Amongthese, poly(3,4-ethylenedioxypyrrole) or PEDOP owing to afairly low oxidation potential (E1/2 of�0.5 V versus Ag/Ag+) andthe ability to undergo a reversible and perceptible change inoptical properties (as it switches between deep red in reducedstate and a pale transparent blue-gray in the oxidized state)render it most suitable for diverse electrochromic devices such asinformation displays, smart windows, antiglare rearview mirrors,helmet visors, and so forth.3�6 Most of the research effortsin electro-optical conducting polymers are now focused on(a) extending the ability of conducting polymers to modulate solarradiation from visible (or photopic) to near-infrared (or NIR)spectral regions, (b) improving the redox cycling capability,(c) widening the operational electrochemical potential stability

window, and (d) further amplifying the electrochromic contrastratio and reducing switching times. To this end, in the recentpast, nanohybrids or composites of conducting polymers withcarbon nanostructures (carbon fibers/tubes/aerogels and acti-vated carbon) have been synthesized and used for variouselectrochemical applications as they offer an attractive combina-tion of properties such as high surface area, low toxicity, longcycle life, chemical and thermal inertness and ease of electrontransport.7�12 Specifically, colloidal dispersions of functionalizedcarbon nanotubes (CNTs), when used with monomers likeaniline, substituted thiophenes, and pyrrole, serve as counterionsduring in situ electropolymerization, thereby yielding a CNT-conducting polymer composite. Such a composite can be easilyfabricated, in a thin film form and that too, in a single step atambient temperature.9,13 However, a major drawback of usingCNTs, is that the syntheses of CNTs typically entail the use of(i) elevated temperatures and (ii) capital intensive equipment.

Received: June 14, 2011Revised: August 6, 2011

ABSTRACT:Composite films of poly(3,4-ethylenedioxypyrrole)(PEDOP) enwrapped by reduced graphene oxide (RGO) andflanked by an ionic liquid (IL: dialkyl substituted imidazoliumimide) have been synthesized. To study the effect of functionalizedRGO on the polymer, the structure, conduction properties, andredox chemistry of the PEDOP-RGO/IL composite films havebeen compared with that of the control PEDOP-IL film, dopedonly by the ionic liquid imide anion. Evidence for the successfulinclusion of RGO/IL in PEDOP was obtained by X-ray photo-electron spectroscopy and high resolution transmission electronmicroscopy, in the form of modified C1s signals, new signals due to nitrogen and sulfur, and also the revelation of a quasi-hexagonalassembly of atoms in theRGO/ILnanosheets. Conductive-atomic forcemicroscopy revealed the role ofRGO/IL in completelymodifyingthe charge carrying ability of PEDOP, as unusually high current values and a largely uniform current distribution were achieved in thePEDOP-RGO/IL composite in contrast to control PEDOP-IL. Point contact I�V profiles showed that the population of insulatingdomains flanking the conducting polymer grains is drastically reduced on replacing the IL dopant with RGO/IL in PEDOP. A reducedFermi edge to valence band gap, an exemplary electrochromic coloring efficiency of 477 cm2 C�1 (λmax = 509 nm), a 2-fold increment indc electronic conductivity, improved switching kinetics, an amplified charge insertion�extraction capacity, as opposed to the case forcontrolfilm,mirrored the role of RGO/IL in controlling the charge transport dynamics of PEDOP.The results show that the entrapping ofa conducting polymer byRGObased nanostructures affords tunable redox, electronic and optical properties, thus providing a paradigm formapping the current distribution in similar nanocomposites.

18355 dx.doi.org/10.1021/jp205551k |J. Phys. Chem. C 2011, 115, 18354–18365

The Journal of Physical Chemistry C ARTICLE

The use of graphene oxide (GO) as a viable alternate to CNTsfor fabricating composites with conducting polymers putativelycircumvents the aforesaid limitations, owing to low processingtemperatures and ease of preparation.14�16 GO is easily proces-sable by themodifiedHummers’method, which usually proceedsvia exfoliation of graphite in acids, followed by rigorous reductionunder appropriate conditions to yield a material with minimumresidual oxygen functionalities and maximum restoration of thecarbon�carbon network.17�19 In a previous report, grapheneoxide has been functionalized with poly(diallyldimethylam-monium chloride) and used for the immobilization ofhemoglobin.20 Likewise, surfactant based intercalants have alsobeen used in the past for creating sandwich structures comprisingnegatively charged GO sheets and positively charged micelles.21

Previously, electro-methods have been used to generate exfo-liated graphene with IL moieties from graphite rods.22,23 Theensuing GO/IL composite is easily dispersed in organic solventsand after monomer (pyrrole or thiophene or aniline) addition, itcan then serve as a counterion for electropolymerization of thelatter. To recover the desirable properties of graphene such aslow band gap (because the graphene monolayer is a zero gapmaterial), high electronic conductivity, extraordinary mechan-ical, and chemical strength, and large effective surface area, strongchemical reduction of GO is imperative.15 Such a protocol,involving the use of an IL, overrides the effect of oxygen moietiesconsiderably and also introduces the ionic liquid cation and anionto the reduced graphene oxide in a controlled way.

There are hardly any reports on graphene/RGO/GO-PEDOPbased composites, but reports on its thiophene analogue, namely,poly(3,4-ethylenedioxythiophene) or PEDOTwith GO do exist.Spin coated films of PEDOT/graphene showed a conductivityimprovement by 2 times of that of neat PEDOT, and a 6-foldenhancement in mechanical strength.24 In another report, gra-phene-PEDOT:poly(styrenesulfonate) or PSS hybrid filmsshowed a conductivity of about 0.2 S cm�1 and a high opticaltransparency and flexibility, thus enabling its use as a functionaltransparent conducting electrode.25 Another report on a flexibledevice of chemically derived graphene and PEDOT:PSS asseparate electrodes showed a novel yellow light emission andgood efficiency.26 Similarly, enhanced performance characteris-tics have been achieved, for supercapacitor electrodes of conductingpolymers when coupled with graphene based materials.27,28 In anearlier study, we also achieved a superior electrochromic re-sponse and redox activity for PEDOT films doped with graphenefunctionalized by a fluoroalkylphosphate (FAP) ionic liquid,29

but PEDOP has a lower oxidation potential and switchesbetween blue and red hues instead of the monochrome bluehues that PEDOT offers. Furthermore, a dialkylimidazloiumimide based IL is less viscous and more conductive and thereforeis easier to handle than an FAP-IL. More importantly, we studyhere, for the first time, the ability of functionalized RGO tocompletely alter the charge transport dynamics of PEDOP,by use of conductive-atomic force microscopy as a probe. Wereport the synthesis of PEDOP-RGO/IL films, and how the

Figure 1. Current (O)/charge (—) versus time transients for oxidative electropolymerization of (a) 0.1 M EDOP in ionic liquid solution (photographof solution in inset) and (b) EDOP in a RGO/IL dispersion under a constant dc potential of +1.5 V. The inset of (b) from left to right shows photographsof GO powder, GO in acetonitrile, RGO/IL bucky gel, RGO/IL in acetonitrile, and EDOP-RGO/IL in acetonitrile. Frequency change (O)/polymermass per unit surface area (0) versus time plots recorded during electropolymerization of (c) 0.1MEDOP in neat ionic liquid and (d) 0.1M EDOP in aRGO/IL dispersion, in chronoamperometric mode.



extraordinary current carrying ability of PEDOP-RGO/IL filminfluences its electrochemical and optical properties in contrastto that of the control PEDOP-IL films grown in the neat ionicliquid.

2. EXPERIMENTAL SECTION

2.1. Chemicals. 3,4-Ethylenedioxypyrrole or EDOP (2% w/vsolution in tetrahydrofuran), graphite platelets (width 50�250 nmand length 0.5�5 μm), and poly(methylmethacrylate) (PMMA,MW 996 000) from Aldrich and ionic liquid (IL) 1-ethyl-3-methylimidazolium bis(trifluoromethylsylfonyl)imide, poly-(ethyleneglycol)400 (PEG400), potassium hexacyanoferrate(K3[Fe(CN)6]) and ferric chloride (FeCl3) from Merck wereused as received. Inorganic transparent electrodes of SnO2:F-coated glass (Pilkington, sheet resistance 14Ω/sq) were cleanedin a soap solution, double distilled water, and acetone prior touse. Ultrapure water (resistivity ∼ 18.2 MΩ cm) obtainedthrough a Millipore Direct-Q 3 UV system and acetonitrile(Merck) were used as solvents. Sulfuric acid (H2SO4), potassiumpermanganate (KMnO4), nitric acid (HNO3), hydrochloric acid(HCl), hydrazine (N2H4), and ammonia (NH3), all procuredfrom Merck, were used as received.2.2. Reduced Graphene Oxide. Graphite platelets (0.5 g)

were dispersed in H2SO4 (11.5 mL), and KMnO4 (0.5 g) wasadded gradually under continuous stirring; the temperature ofthe slurry was maintained at approximately 20 �C. The slurry wasmagnetically stirred at 35 �C for 2 h, followed by addition ofdeionized water (2 mL) and further stirring for 1 h. The reactionwas terminated by addition of deionized water (70 mL). Thesuspension was washed repeatedly using a 1:10 (HCl: Water,v/v) solution to remove unwanted metal ions until the super-natant liquid showed a pH close to neutral. Graphene oxide wasextracted by filtration and dried in an oven at ∼60 �C until acompletely dry product was obtained. Graphene oxide (0.5 g)was dispersed in deionized water (10 mL), hydrazine (0.6 mL),and ammonia solution (7 mL) and refluxed for 1 h in an oil bathat 95 �C. The suspension was then washed thoroughly withdeionized water and dried in oven at ∼60 �C for a few hours.Hydrogen gas was purged through the solid for 10 min to yieldreduced graphene oxide (RGO), and it was stored in a vacuumdesiccator. RGO (0.08 g) was ultrasonicated with 1-ethyl-3-methylimidazolium bis(trifluoromethylsylfonyl)imide (2 g) for2 h, and the resulting bucky gel was subjected to iterative washingwith acetonitrile and filtration to remove the superfluous ionicliquid until RGO/IL composite was obtained. The photographsof GO, RGO/IL bucky gel, and suspensions formed from GOand RGO/IL in acetonitrile are shown in the inset of Figure 1b.2.3. PEDOP-RGO/IL and PEDOP-IL films. A colloidal suspen-

sion containing EDOP (0.1 M), the bucky gel of RGO/ILcomposite in acetonitrile (12 mL) and PEG400 (2 mL) wasused for the preparation of the PEDOP-RGO/IL films (thephotograph of the precursor is shown in the inset of Figure 1b).The solution was ultrasonicated prior to film deposition tominimize aggregation and settling down of particles. A clear solu-tion of 1-ethyl-3-methylimidazolium bis(trifluoromethylsylfonyl)-imide (2 g) and EDOP (0.1 M) in acetonitrile (12 mL) andPEG400 (2 mL) was used for synthesizing the control PEDOP-IL films (photograph of this formulation is shown in the inset ofFigure 1a). A three-electrode single compartment cell with aplatinum sheet as counter electrode, Ag/AgCl/KCl as referenceelectrode, and a transparent conducting SnO2:F coated glass

substrate as the working electrode was utilized for film deposi-tion. A constant potential of +1.5 V was applied to the workingelectrode, which was immersed in EDOP-RGO/IL dispersion orEDOP-IL solution for 300 s at room temperature. The resultingbluish-black colored composite films of PEDOP-RGO/IL orcontrol PEDOP/IL were immediately rinsed with acetonitrileand dried in air for 2 h and stored in air. Prussian blue films werefabricated from a solution of K3[Fe(CN)6] (10 mM) and FeCl3(10 mM) in HCl (0.01 N) in galvanostatic mode by applicationof 10 μA cm�2 to a SnO2:F coated glass substrate for 8 min. Apolymeric gel electrolyte was prepared by dissolving 15 wt % ofPMMA in IL. Protocols for assembling devices composed ofPEDOP-RGO/IL-PB and PEDOP-IL-PB are summarized else-where in detail.6,12

2.4. Characterization Techniques. Simultaneous measure-ment of Coulombic charge and the polymer mass depositedduring electropolymerization was done with a electrochemicalquartz crystal microbalance, on a 6 MHz, AT-cut quartz crystalcoated with 100 nm polished gold layer of 8 mm diameter. A goldwire functioned as a counter microelectrode and an Ag/AgCl/KCl electrode served as the reference. The frequency resolutionwas 0.07 Hz, and the mass detection limit was well within thesubmicrogram range. Surface morphology of films/solids wasexamined using a scanning electron microscope (SEM, ZeissEVO MA10). SEM images of GO, RGO/IL, control, and com-posite samples are displayed in Figure S1 and S2 (SupportingInformation). For high resolution transmission electron micro-scopy (HRTEM), a thin layer of the sample was carefullyextracted using forceps in deionized water and then transferredonto a carbon coated copper grid, 3.05 mm in diameter, and thesolvent was evaporated at room temperature before use. AnHRTEM FEI Tecnai G2 F 30 STWIN with a FEG sourceoperating at 300 kV was used. Conductive AFM (Veeco Multi-mode 8) with a current sensingmodule was used in contact modeto measure current images, and topographical images wererecorded in the tapping mode. For conductive AFM, cantileversmade of antimony doped silicon (n-doped; resistivity ∼0.01�0.025Ω cm) and coated with platinum�iridium (20 nm) on thefront and backside were purchased from Veeco. A bias voltage of∼100 mV was applied to the conducting tip (which was grounded)during all imaging measurements. The current sensitivity was 1nA/V and a load force of 57 nN was maintained between the tipand the sample. The sample (control PEDOP or PEDOP-RGO/IL deposited on SnO2:F coated glass, size 4 mm � 3 mm) wasfirst tethered to a stainless steel support disk with a thinconducting carbon tape. The films were the same as the onesused for evaluating electrochromic/electrochemical performance.Silver paste was applied uniformly and carefully from the topsurface of the sample, along the edge of the conducting glasssubstrate to the stainless steel mount, to enable the measure-ment, which was performed under ambient conditions. Due carewas taken to ensure that no cracks were formed on the silverpaste coating, upon drying, to maintain a continuous contact,during the measurement. I�V characteristics for both sampleswere recorded in the point contact mode between +3 and�3 V.X-ray photoelectron spectroscopy (XPS) was carried out on a

Perkin-Elmer 1257 model operating at a base pressure in therange 2.0 � 10�8 to 4.2 � 10�8 Torr (100 W, 15 kV) with anonmonochromatized Al Kα line at 1486.6 eV, an analyzer passenergy of 71.55 eV, and a hemispherical sector analyzer capableof 25 meV resolution. The overall instrumental resolution wasabout 0.3 eV. For all fitting multiplets, the FWHMs were fixed



accordingly. Corrections due to charging effects were taken careof by using C(1s) as an internal reference and the Fermi edge of agold sample. The C1s signal was fixed at 284.6 eV. The core levelspectra were smoothened and deconvoluted using a nonlineariterative least-squares Gaussian fitting procedure. The JandelPeak FitTM (version 4.01) program was used for the analyses.For recording the valence band spectra, the step size was0.026 eV and the time per step was 100 ms and ten sweeps wereperformed for each sample. Cyclic voltammetry (CV) for thefilms was performed in a conventional three electrode electro-chemical cell within (1/( 3 V, wherein a composite/controlfilm was used as the working electrode, Ag/AgCl/KCl wasemployed as the reference electrode and a Pt wire was used asthe auxiliary electrode in the neat ionic liquid: 1-ethyl-3-methylimi-dazolium bis(trifluoromethylsylfonyl)imide working as an electro-lyte. The electronic conductivities of PEDOP-RGO/IL andPEDOP-IL films were obtained by transferring the film from

the substrate to a stainless steel (SS) surface and by placinganother SS flat electrode as an overlying layer on the sample. I�Vcharacteristics were obtained by linear sweep voltammetry(LSV), wherein the voltage was swept from �3 V to +3 V.The optical density for coloration efficiency calculations wasmeasured in situ, in a device configuration on a ShimadzuUV�visible-NIR 3600 spectrophotometer under dc potentialsof different magnitudes (applied for a 90 s duration). Ramanspectra were recorded on a Bruker Senterra Dispersive RamanMicroscope spectrometer, the laser excitation wavelength wasfixed at 785 nm. All electrochemical measurements and electro-polymerization was performed on an Autolab PGSTAT 302Ncoupled with a NOVA 1.6 software.

3. RESULTS AND DISCUSSION

3.1. I�t plots and Electrogravimetry. The current�timetransients recorded during electropolymerization from suspen-sions of (a) EDOP and IL and (b) EDOP and RGO/IL under afixed dc potential of +1.5 V are shown in Figure 1a,brespectively. The initial spike followed by a plateau likeresponse, observed for the formation of both films is char-acteristic of (a) oxidation of monomer and (b) coupling ofradical cations and precipitation of the oligomers onto thesubstrate to yield PEDOP nuclei. The plateau region com-mensurates with the PEDOP chain propagation. The corre-sponding charge involved during electropolymerization isalso shown. For the control PEDOP-IL film, the Coulombiccharge (Q) deposited increases almost linearly with time, andat the point of culmination of deposition time, i.e., at t = 300 s,Q was ∼0.8 mC. A similar variation in current/charge withtime was observed for potentiostatic synthesis of PEDOTdoped by perchlorate ions.30 In contrast for the PEDOP-RGO/IL, the Q versus t plot (Figure 1b) shows a lineardependence from t = 0 s to t = 80 s, and at this juncture, Q is0.13 mC; thereafter Q saturates and does not vary much untilthe end of the deposition duration (t = 300 s). The magnitudeof Coulombic charge deposited on the substrate from theEDOP-RGO/IL dispersion is obviously smaller compared tothe charge deposited from the EDOP-IL solution. Thisdifference may be attributed to the bulkier nature of theRGO/IL flakes, which are only dispersed and not dissolved asthe ionic liquid anion or cation are, in the EDOP-IL solution.As a consequence, the movement of RGO/IL nanosheets,which serve as counterions for charge compensation onto theradical cations formed on PEDOP, is sluggish and thereforelesser charge is deposited. The corresponding Δf versus timeplots (Figure 1c,d) for the two samples reveal a largerfrequency decrease in the growing PEDOP-IL film as com-pared to that for the PEDOP-RGO/IL system, which alsoindicates a higher amount of polymer deposited in the formercase. The corresponding mass deposited onto the substrate isalso considerably higher for the control PEDOP-IL film(Figure 1c) as compared to that for the PEDOP-RGO/ILcomposite. The Sauerbrey equation was employed for deter-mination of mass, where Δf is the resonant frequency of thequartz crystal, fo is the fundamental frequency, F is the densityof the quartz crystal (2.648 g cm�2), μ is the shear modulusof quartz (2.947 � 1011 g cm�1 s�2), n is the harmonicnumber of oscillation (∼1), and Cf is the sensitivity factor(Hz cm2 μg�1).

Figure 2. (a) TEM image of wrinkled pristine RGO/IL nanosheets(scale bar = 50 nm). HRTEM images of (b) RGO/IL nanosheetsjuxtaposed on one another (the block encapsulated by dotted linesreveals the defects in the structure) (scale bar = 5 nm), (c) wavylattice fringes in overlapping nanosheets of RGO/IL (scale bar =5 nm) (inset: corresponding spotty EDP generated by FFT),(d) blown up view of a small defect free region from (c) showingthe fringe separation, (e) quasi-hexagonal lattice in RGO/IL indi-cated from the regions enclosed by ellipses (scale bar = 2 nm), and(f) atomic resolution image (scale bar = 1 nm) derived from (e).



Δf ¼ � 2fo2mn=ðFμÞ1=2 ¼ � Cfm ð1Þ

A direct comparison between the masses of the two samples isnot justified; as in the case of the PEDOP-IL film, the massdeposited includes the mass of the solvent as well, for the ionicliquid cation/anion is highly solvated in solution and thereforemigrates as a whole onto the growing film. In contrast, for theformation of PEDOP-RGO/IL, the counterion is least likely tobe solvated and therefore the mass involved is significantlylowered. In an earlier report, a similar, almost linear variationfor polymer mass per unit surface versus time was obtained for aPEDOT-ClO4 system.31 However, despite a higher mass/charge,themaximum charge that can be deposited onto the substrate at agiven time, is attained in a much shorter time span in the case ofthe PEDOT-RGO/IL film. This film is surprisingly found to be

more efficient at electron conduction than the PEDOP-IL film,indicating that a lower charge value does not necessarily implypoor conduction ability.3.2. HRTEM Studies. The HRTEM image of neat RGO/IL

shown in Figure 2a reveals crumpled silk veil-like waves of RGO/IL nanosheets with a lateral size of about a few hundrednanometers. The undulations on the nanosheets are due to de-fects (caused by excessive sonication), and the region enclosedby a rectangle in Figure 2b shows the point where two nanosheetsoverlap. The folds and a rather high density of edge dislocationsproduce the wavy structure in RGO/IL (Figure 2c). Fast Fouriertransformation (FFT) performed on the same, yielded theelectron diffraction pattern (EDP) which is shown as an insetof Figure 2c. The interfringe spacing determined from the EDP,is∼0.15 nm. A blown-up view of a small region from this image isextracted onto Figure 2d. The separation between the two fringesis ∼0.14 nm and it is close to the value determined from EDP.Our value agrees well with that previously reported for a singlelayer of graphene prepared by micromechanical cleavage.32

Regions in RGO/IL, where the quasi-hexagonal lattice couldbe perceived are encircled in ellipses in Figure 2e. A magnifiedview of one such region from Figure 2e is shown in Figure 2f; itdistinctively illustrates the atomic level resolution attained in aRGO/IL layer and also concurs well with a reported TEM viewfor a graphene membrane.32 HRTEM of the control PEDOP-ILfilm, shown in Figure 3a, shows an amorphous structure, an usualfeature of conducting polymer.33 Low magnification image(Figure 3b) of the PEDOP-RGO/IL composite again revealscompactly packed crumpled sheets, folded and curled up inmanyregions. This image shows that the polymer particles are ad-sorbed onto the surface of RGO/IL nanosheets and the pre-dominant dark hues in the image indicate that almost a full use ofthe large specific area of RGO/IL nanosheets has been made.RGO/IL nanosheets act as scaffolds first for the adsorption of themonomer and later for the growth of the polymer chains, uponanodic oxidation, via formation of radical cations compensatedby the negatively charged imide and carboxylate functionalitiesflanked to graphene oxide. The lattice fringes for two differentlayers of RGO/IL labeled as layers I and II and in direct contactwith the polymer can be perceived in Figure 3c. Furthermagnification of the same yielded Figure 3d, which shows adistinct interface (shown with a dashed line) formed between thedisordered polymer (seen in the right portion of the panel) andthe structural ordering of the crystalline RGO/IL nanosheetstructure (visible on the left side). The structure of RGO/ILcontinues to be preserved even in the PEDOP-RGO/IL compo-site, as dark and light fringe patterns could be identified in someportions of the film as well (Figure 3e). The separation betweentwo similar lines was ∼0.32 nm, which is close to a latticeconstant of 0.34 nm attained for graphene films grown bychemical vapor deposition.34,35 FFT on the Figure 3e is shownas an inset, and this too conforms to a d-spacing of 0.34 nm. Theincrease in the layer spacing on going from neat RGO/IL toPEDOP-RGO/IL could be due to the incorporation of thepolymer. The corrugated structure of RGO/IL interspersed withdefects (the latter evident from the curved and discontinuous linepattern in Figure 3f) in contact with the featureless polymer isalso seen. Such a structure made of PEDOP chains sandwichedbetween RGO/IL layers has a large active surface area, which isfavorable for charge transport.3.3. XPS Studies. The deconvoluted C1s and O1s core level

spectra of GO and RGO are shown in Figure 4. Consistent with

Figure 3. HRTEM images of (a) amorphous, featureless structure ofcontrol PEDOP-IL film (scale bar = 10 nm) (inset: corresponding lowmagnification view (scale bar = 50 nm)); (b) multiple layers of RGO/ILnanosheets enwrapped PEDOP in the composite PEDOP-RGO/IL(scale bar = 50 nm), (c) one nanosheet (I) of RGO/IL overlying onanother nanosheet (II) in PEDOP-RGO/IL (scale bar = 10 nm), (d) aclear interface formed between the amorphous polymer and the orderedstructure of RGO/IL, represented by a dotted line (scale bar = 10 nm),(e) an interfringe spacing of 0.32 nm achieved in a high quality crystalliteof RGO/IL retained in PEDOP-RGO/IL (inset: correspondingspeckled EDP produced by FFT), and (f) corrugated sheets of RGO/IL in direct contact with the amorphous polymer (scale bar = 2 nm).



Gao et al.’s C1s spectrum of hydrophilic GO produced by amodified Hummers’ method,36 here we also obtained twoseparated peaks in the C1s spectrum (Figure 4a), a consequenceof a high concentration of oxygen functionalities. The two peakswere resolved into six components (fwhm = 1.75 eV) at 284.7,286.5, 288.2, 293.6, 295.3, and 296.8 eV attributable to C—C,C—OH (hydroxyl), C(O)C (epoxy), C—O—C (ether), CdO(carbonyl), and COO� (carboxylate) groups. The deconvolu-tion of the slightly asymmetric O1s peak yielded two compo-nents at 532.4 and 533.8 eV (fwhm = 2.0 eV) ascribable to C—Oand C—OH links (Figure 4d). The broadness of the C1s peak isconsiderably reduced, and the two peaks now appeared as one, inthe RGO/IL sample (Figure 4b). A similar transition from ahighly asymmetric to an almost nonasymmetric C1s signal wasobserved previously on going from GO to RGO by Stankovichet al.37 Here, the single dominant C1s peak peak observedfor RGO indicates a good degree of restoration of the CdCnetwork, after chemical reduction. The spectrum shows fourcomponents at 284.7, 286.6, 288.9, and 291.1 eV, after fixing thefwhm at 1.92 eV, corresponding to contributions from C—C,C—N, C—O, and CdO linkages (Figure 4b). The additionalC—N component shows that oxygen removal is accompanied bynitrogen incorporation in graphene during reduction. A similarsignal for carbon bound to nitrogen has been observed by Yanget al.38 and Dreyer et al.39 for graphene functionalized with anamine terminated IL and for graphene reduced by hydrazine.Also notable is the reduction of the intensity of the O1s signal toabout 34% of its original, on going fromGO to RGO/IL, which issuggestive of the removal of oxygen containing groups(Figure 4e). This signal is again resolved into two componentsat 531.5 and 533.5 eV (fwhm = 2.55 eV) due to C—O and C—OH. The N1s peak of RGO/IL (Figure 4c) was deconvolutedinto main components at 399.2 and 401.65 eV; the former isfrom the S—N of the imide ion and the latter arises from C—N

(of the imidazolium cation and between the carbon of GO and theN generated during hydrazine treatment). A similar split of theN1s signal for the pure ionic liquid 1-butyl-3-methylimidazoliumcation and imide anion has been observed earlier.40 Here, thefwhmwas 1.88 eV, I(S�N)was 0.41, and I(C�N) was 0.59. TheS2p spin�orbit splitting in RGO/IL (Figure 4f) with the S2p3/2and S2p1/2 at 163.3 and 167.58 eV (fwhm = 2.93 eV) and a peakseparation of about 1.2 eV shows an intensity ratio of ∼3:1,which agrees reasonably well with the reported values for the S2pcomponents.12 This sulfur signal is attributable to the sulfur onthe imide anion. Similarly, the F1s signal (inset of Figure 4f) withfwhm of 2.3 eV was symmetric and peaked at 690.4 eV; this peakarises from the C�F linkages of the �CF3 groups on the imideanions. The distinct signals due to N, F, and S in the RGO/ILsample affirm the inclusion of both the imidazolium ion and theimide anion within the RGO layers. The incorporation of RGO/IL in PEDOP in the composite and IL in control PEDOP wasconfirmed from the core level spectra of the control and com-posite. These spectra are shown in Figures S3 and S4 (SupportingInformation).For neat GO (prior to IL incorporation and reduction), the

I(C—C)(C1s)/I(Ctotal) ratio was 0.31, which is the proportion ofGO that remains deoxygenated. For the same sample, theI(C—O)(C1s)/I(Ctotal) is 0.69, which is a measure of the degreeof oxygenation prevalent in GO. A high value for this ratio issuggestive of a high oxygen content in GO. Subsequent totreatment with the IL and hydrazine reduction, for the RGO/IL sample, the I(C—C)(C1s)/I(Ctotal) ratio increased from 0.31 to0.64, indicating a fairly high restoration of the sp2 bonded pristinecarbon framework. The successful elimination of oxygen func-tionalities was also reflected in the low but nearly equivalentratios of I(C—O)(C1s)/I(Ctotal) and I(C—N)(C1s)/I(Ctotal). The valuewas∼0.36 for neat RGO/IL, as opposed to a rather large value of∼0.69 (I(C—O)(C1s)/I(Ctotal)), observed for neat GO. For control

Figure 4. Deconvoluted core level spectra of (a) C1s and (d) O1s for GO before reduction, (b) C1s, (c) N1s, (e) O1s, and (f) S2p of RGO/ILcomposite. The inset of (f) shows the F1s peak from RGO/IL.



PEDOP-IL, the dopant level, which is a measure of the imide ionloading in the polymer, was estimated from the I(SdO)(O1s)/I(Ctotal) ratio, and it was found to be ∼0.37. From the N1sspectrum of the same sample, the I(S—N)(N1s)/I(Ctotal) ratio wasdeduced to be∼0.32. Because the two values are not far apart, itis a reasonably good estimate of the doping percentage in thecontrol PEDOP-IL. From the N1s of control PEDOP-IL, theI(C—N)(N1s) contribution was found to be quite high, ∼0.43,which is as anticipated, as it is predominantly arising from C—NHlinks on the PEDOP chains. Most striking was the ratio of theI(S—N)(N1s)/I(C—N+)(N1s) ratio; this is a ratio of the imide ion toimidazolium ion content in the film. It was∼1.28, indicating thatthe proportion of imidazolium cation incorporated in the poly-mer is only slightly lower than the proportion of imide ionincluded in the polymer. This ratio clearly indicates the incor-poration of the imidazolium cation along with the anion in thecontrol PEDOP-IL film, a feature that is absent in conductingpolymer films doped by conventional anions.12 In PEDOP-RGO/IL composite, from the O1s spectrum, the {I(SdO)(O1s) +I(C—OH)(O1s)}/I(Ctotal) ratio was found to be ∼0.41, which is ameasure of the doing level of RGO/IL in PEDOP. For the samesample, from the N1s spectrum, the I(S—N)(N1s) and I(C�N+)(N1s)

values were ∼0.11 and 0.15, respectively. Because the imidazo-lium ion was found to be in a higher proportion than the imideion here, it is apparent that the imidazolium ion has a greateraffinity for GO than the imide. Again, the S—N or C—N+ pro-portions are lower in contrast to their values in control PEDOP-IL, for here the IL is flanked by GO and is not the sole counterionduring electropolymerization of PEDOP. A schematic showingthe interactions that prevail in the PEDOP-RGO/IL compositeis shown in Figure 5.

3.4. Valence Band and Raman Spectra. The experimentalphotoemission spectrum of the valence band region for neat GO,neat RGO/IL, control PEDOP-IL, and PEDOP-RGO/IL areshown in Figure 6. The density of states (DOS) at the Fermi level(EF, corresponding to a zero binding energy) is finite for allsamples. The DOS in the vicinity of EF has a larger magnitude(inferred from a higher intensity of the signal in the inset ofFigure 6a) in RGO/IL as compared to that for GO. In an earlierreport,41 a similar increase in DOS was observed as the level ofhydrogenation (or reduction) was increased in graphene. Theauthors41 postulated that the increase in DOS is caused by achange in the geometry: as the hydrogenated carbon atomsmoveout of the graphene plane, the lattice is distorted and thesymmetry is broken, and as a result, more k points in the Brillouinzone contribute to DOS. The same line of reasoning can be usedto explain the DOS increment on going from GO to RGO/IL.From the valence band diagram, the EF to valence band gap forGO was 2.05 eV and it reduced to 1.62 eV in RGO/IL, indicatingthe formation of a more conductive material upon reduction. Forcontrol PEDOP-IL, this gap was 1.76 eV and for the PEDOP-RGO/IL composite it was 1.26 eV (Figure 6b); this narrowing ofthe energy gap could be due to the introduction of additionallevels in the band gap by RGO/IL, thus enabling the formation oflower gap conducting polymer.Raman spectra of GO and RGO are shown in Figure 7. Both

spectra show the D band and the G band. For neat GO, the G

Figure 5. Schematic showing the interactions in the PEDOP-RGO/ILcomposite film.

Figure 6. Valence band spectra of (a) (0) neat GO before reductionand (O) RGO/IL composite and (b) (0) control PEDOP-IL film and(O) PEDOP-RGO/IL composite film. The inset of (a) shows thedifference inDOS betweenGO and RGO/IL at 0 eV. The dotted lines in(a) and (b) denote the valence band to Fermi level gaps.



peak is seen at 1591 cm�1 and after reduction it shows a blue shiftto 1584 cm�1. The D band observed at 1353 cm�1 in pristineGO, which is ascribed to surface defects and smaller sizedgraphene layers and indicates the presence of other functional-ities, also shows a blue shift to 1348 cm�1 upon reduction. The Gpeak is due to the in-plane vibration of sp2 hybridized carbonatoms, which is the doubly degenerate zone center E2g mode.

23,42

The intensity of both D and G bands decreases on going fromGO to the reduced sample, indicating a lower defect densityupon reduction. Similar hypsochromic shifts were also registeredby authors for reducedGOproducts in previous work.23,36,43 The2D overtone band, which is visibly of a lower intensity than thecorresponding D band, is at 2697 cm�1 in neat GO and down-shifts to 2674 cm�1 in the reduced sample. Such a shift has beenobserved previously on going from exfoliated GO to reducedgraphene.36

3.5. Conductive-AFM Studies. Typical current and thecorresponding topographical images as well as the cross-sectionalcurrent profiles of control PEDOP and PEDOP-RGO/IL filmsare shown in Figure 8. Figure S5 (Supporting Information)shows the images and I�V trace of the neat SnO2:F coated glass,which ratify that the plots obtained in Figure 8 arise from theelectroactive coating and not the underlying substrate. Thecurrent maps (Figure 8a,b) represent the current flowing verti-cally through the film, and these were imaged by applying a fixedvoltage of 100 mV between the conducting tip and the substrate(on which the sample is mounted), while the tip simultaneouslyscans the surface of the film in a horizontal manner. The verticalcurrent flowing through the film, as estimated by this method, ismore relevant than the bulk dc conductivity determined by two/four-probe methods,44�46 for in most of the practical electronicdevices, charge flows vertically through the electroactive material.In the current maps (Figure 8a,b), the bright regions are attributedto regions of high current, and the dark domains correspond tolow currents. The maps are slightly noisy, and this originatesfrom a rather high rms roughness of the films, as for controlPEDOP-IL, it is 28 nm, and it is 16.2 nm for the PEDOP-RGO/IL film. The current map of the control PEDOP film (Figure 8a)shows the bright spots to be randomly scattered across the film,and these are surrounded by relatively large domains of darkregions. On the other hand, the current map of the PEDOP-RGO/IL nanocomposite film shows very few dark domains,indicating that current tends to flow more uniformly across thesurface of the PEDOP-RGO/IL film, as the conducting regions

(denoted by bright regions) are enclosed by insulating darkdomains to the extent they are in the control PEDOP film. It isapparent that the RGO/IL dopant tends to connect the con-ducting polymer (PEDOP) grains by providing conducting path-ways, thus maintaining a uniform current distribution. In con-trast, in the control PEDOP film, the ionic liquid dopant iselectronically insulating and, therefore, the IL abundant regionsin the film yield the dark regions and they obstruct the currentflow. The maximum current flowing at the bright spots is about2 nA in the control PEDOP film, which is significantly lower thana maximum current of 550 nA detected for the PEDOP-RGO/ILfilm, under the same bias voltage of 100mV. The role of RGO/ILis furnishing vertical pathways for charge percolation and pro-pagation, thus improving the overall charge transport behavior ofPEDOP.Further proof to support this claim, was obtained in the form

of the point contact I�V characteristics of control PEDOP andPEDOP-RGO/IL films (Figure 8e,f), recorded from the brightregions (spots: A1 and B1) and the dark regions (spots: A2 andB2). Because current profiles recorded from different bright/dark regions from the current map of a given film were roughlysimilar, only representative plots are shown here, chosen so thatthey represent the average response of the film. For the controlPEDOP sample, the current profile recorded from a bright spotA1, resembled that of a conventional semiconductor, in thepotential range�0.67 to +0.87 V. As can be seen from the figure,current saturates at∼12 nA, in the bright conducting spots, in thecontrol PEDOP film. Although the response is nonlinear, butassuming a linear behavior, the slope is ∼15 nA/V, indicatingpoor conduction in the sample. The current response in the darkregion shows an almost zero slope in the �0.7 to +0.7 V range,implying that very low or almost zero currents flow in the darkdomains. Beyond this bias range, the response shows manycurrent spikes, more dominant above +1.5 and�1.5 V. A similarI�V curve, characterized by many spikes, has been observedearlier from the dark region of an electropolymerized polypyrrolefilm.44 Authors44,46 assigned such spikes to noncontinouusdensity of states available for charge flow, which here, could bedue to inhomogeneous distribution of the conducting PEDOPand the insulating IL dopant grains. In the case of the PEDOP-RGO/IL film, the slope in the �0.25 V to +0.25 V region wassteep and, astonishingly, the traces were almost similar for bothbright and dark spots, thus implying that insulating domains wererather scarce in this nanocomposite. As mentioned earlier, I�Vtraces were obtained from different bright and dark regions, andthe current profile and the overall magnitudes of currents were ofthe same order. The current was found to saturate at∼500 nA, indifferent regions of this film, thus suggestive of facile electrontransport across the film. The slope of the curve was approxi-mately 2000 nA/V, which is greater by 2 orders of magnitudecompared to that of the control PEDOP film. For highly dopedPEDOT films, the slope estimated from conductive AFM wasfound to be in the range 3000�9000 nA/V. Here, because theresponse measured is for the as-fabricated films and not forelectrolytically oxidized films, the difference is accounted for.The enhanced conduction ability of PEDOP-RGO/IL was

reaffirmed from the I�V characteristics of measured by linearsweep voltammetry method (Figure S6, Supporting Informa-tion). The electronic conductivity of GO was deduced to be0.08 S cm�1 and for RGO/IL, it was 0.11 S cm�1. For thePEDOP-RGO/IL composite, the conductivity was 0.025 S cm�1

and for the control PEDOP-IL sample, it was 0.0007 S cm�1.

Figure 7. Raman spectra (solid scan, excitation wavelength = 785 nm)of (a) reduced GO (black line) and neat GO (red line). (b) and (c) arethe Gaussian fits of D and G bands from (a).



These values also complement the current sensing AFM results,as RGO/IL increases the conductivity of PEDOP by 2-fold times,an attribute, which is of paramount significance for improving theredox response.3.6. Spectroelectrochemistry of PEDOP-IL and PEDOP-

RGO/IL. The absorption plots of control PEDOP-IL and PED-OP-RGO/IL composite devices for oxidation potentials variedfrom +0.5 to +2.0 V (at intervals of +0.5 V) and reductionpotentials from �0.2 to �3.0 V (at intervals of �0.1 V) areshown in Figure 9a,b. The reference potential for determiningthe optical density change at any given wavelength (ΔOD(λ))was fixed at +1.0 V for both devices. The redox switchingresponse was studied by cyclic voltammetry, and these plotsare shown in Figures S7 and S8 (Supporting Information). Thebipolaronic peak observed in the oxidized state of the controlPEDOP-IL in the 600�800 nm wavelength range paves way forthe π�π* absorption peak, observed under reduction potentials.The peak corresponding to π�π* transitions is observed in thewavelength range 490�495 nm, and it becomes predominant

only at E > �2.0 V. But notable is the retention of a high butrather flat absorption band in the 900�1800 nm wavelengthregion. This absorption in the near-infrared (NIR) region variesin a manner exactly similar to the bipolaronic peak; i.e., theintensity of this flat band decreases with increasing reductionpotential, thus indicating the ability of the control film tomodulate not only visible light but also NIR radiation. For thePEDOP-RGO/IL composite (Figure 9b), the trends for opticaldensity variation with wavelength/potential are akin to that ofthe control PEDOP-IL device, albeit a few differences. Theπ�π* absorption peak also tends to preponderate at E > �2.0 V,but it shows a bathochromic shift relative to the control PEDOP-IL,and it is seen in the wavelength range 500�510 nm. TheΔODmax (ΔODmax = ΔOD(�3.0V) � ΔOD(+1.0V)) in the visibleregion at λmax ∼ 500 nm for the PEDOP-RGO/IL compositewas 0.94 and for the control PEDOP-IL, at a λmax of 495 nm,ΔODmax was 0.69. Our value for the composite is higher than aΔOD of ∼0.65 observed for a PEDOP film doped by ClO4

�

Figure 8. Representative current and topography images of control PEDOP-IL (a (height = 2 nA) and c (height = 190 nm)) and PEDOP-RGO/IL(b (height = 550 nA) and d (height = 142 nm)). Adjoining (c) and (d) are the current cross-sectional profiles, measured along the lines shown in (a) and(b). I�V curves obtained from the points A1 and A2 in control PEDOP-IL (e) and from points B1 and B2 in PEDOP-RGO/IL nanocomposite (f).



ions.3 In the PEDOP-RGO/IL device, a peak with a λmax of∼733 nm was seen distinctly in the high reduction potentialrange from�1.9 to�3.0 V, with maximum intensity at E =�1.9 V.Furthermore, the intensity of the flat absorption band inthe NIR region (at λ g 900 nm), for the composite, showedthe expected trend; i.e., it diminished with increasing reductionpotential only in this narrow potential range from�1.9 to�3.0 V.For the remaining potentials of +0.5 to +2.0 V and �0.2 to�1.8 V the trend was just the opposite; the absorption intensityof the flat band increased with increasing reduction potential.Although it is difficult to explain at this stage, it is probably theRGO/IL nanosheets that cause the digression from the typicaltrend, as the control PEDOP-IL device does not show thisunusual behavior.Coloration efficiency (CE) plots of control PEDOP-IL and

PEDOP-RGO/IL composite devices are shown in Figure 9c,d.For the control PEDOP-IL, a CEmax(visible) of 263 cm

2C�1, in thevisible region was attained under a potential of �2.7 V at aλmax(visible) of 467 nm. In addition to a CEmax(visible) in the visibleregion, another CEmax(NIR) was observed in the NIR region, andit was higher than its visible counterpart for applied reductionpotentials of �1.8 to �3.0 V. For reduction potentials in therange �0.8 to �1.7 V, CEmax(NIR) was lower than CEmax(visible).For the same film, under �2.7 V at a λmax(NIR) of 1275 nm, aCEmax(NIR) of 568 cm2 C�1 was registered. For the PEDOP-RGO/IL composite, a remarkably high CEmax(visible) of 477 cm

2

C�1 was achieved at a λmax(visible) of 509 nm, under a reductionpotential of only �1.0 V. For the composite, in general, theCEmax(NIR) was lower than CEmax(visible), for nearly all valuesof applied potential under consideration. The CEmax in the

photopic region is higher for the composite than for the controlfilm and it is also attained under a significantly lower value ofapplied potential, which implies that a fairly low injected chargedensity suffices to bring about a larger electrochromic contrast inthe composite. This is most useful for increasing the cycling life ofthe device. The control PEDOP-IL is bereft of any suchadvantage, as the potential at which CEmax is obtained is muchhigher (E = �2.7 V). This is also augmented by the low CEshown by the composite device at high negative potentials.Although the control film shows a higher CEmax(NIR) than thecomposite, but because, by and large, the primary function ofelectrochromic smart windows is to tune visible radiation, thisfunction is far superior for the composite while simultaneously italso retains a fairly large coloring efficiency in the NIR region. Inthis manner, the composite achieves an optimal balance betweenvisible and NIR coloring efficiencies at charge densities lowerthan that required for acquiring CEmax in the control PEDOP-IL.The PEDOP-RGO/IL composite, under a slightly lower reduc-tion potential of �0.8 V, shows a CE(visible) of 459 cm

2 C�1 at aλmax(visible) of 532 nm, indicating the capability of the film toretain a fairly high CE, even at potentials less than �1.0 V. TheRGO/IL nanosheets facilitate electron transport, as observedearlier by conductive-AFM, for there exist very few insulatingdomains in the composite, and therefore electron movementthrough the film is efficient and almost unhindered. On the otherhand, in the control PEDOP-IL film, the prevelance of insulatingregions in the absolute vicinity of the conducting regions hinderselectron transport. The greater the homogeneity of conductingareas, as in PEDOP-RGO/IL, the easier it would be for electrons,to move out during oxidation and to move in during reduction

Figure 9. Absorption spectra of (a) control PEDOP-IL and (b) PEDOP-RGO/IL composite based devices under oxidation potentials of +0.5, +1.0,+1.5, and +2.0 V and under reduction potentials of �0.2 to �3.0 V, in steps of �0.1 V in the 300�1800 nm wavelength range. Coloration efficiencyversus wavelength plots for (c) control PEDOP-IL and (d) PEDOP-RGO/IL composite based devices, under reduction potentials of�0.8 to�3.0 V, insteps of �0.1 V. The absorption spectrum under +1.0 V has been chosen as reference in both cases.



through the cross-section of the film. For every electron ejectedfrom the film, a charge compensating anion is inserted into thefilm from the electrolyte; so the greater the ease of electronmovement in the film, the greater would be the number of ionsadsorbed by the film from the electrolyte and thus larger wouldbe the contrast and therefore the coloration efficiency. The CEresults markedly reflect the role of RGO/IL nanosheets inenabling greater and easier access of the electrolyte ions to theelectrochemically active sites in the composite. Although forPEDOP the exact value of CE is not reported, in our previousstudy on a PEDOP-Au nanocomposite,6 we obtained a CE of270 cm2 C�1 at λ = 458 nm. For PEDOP-RGO/IL, the CE ismuch higher in the visible region. On comparing our values withthe thiophene analogue, i.e., PEDOT, we find that Gaupp et al.47

achieved a CE of 206 cm2 C�1, at a λmax of 585 nm for 80% of thefull switch. For another polychromic polymer, containing EDOTunits, a CE of 680 cm2 C�1 was reported at 535 nm.48 For apoly(3,4-propylenedioxythiophene)-Et2 film, a CE of 505 cm2

C�1 has been observed, which is slightly greater than our valuefor the composite.49

Electrochromic switching response of the two devices, per-formed under a square wave potential of (3 V, at a fixedwavelength of 500 nm, at 0.05 Hz, are shown in Figure 10a,c.The absorptive modulation of the control PEDOP-IL is compar-able to that of the composite PEDOP-RGO/IL, but the compo-site acquires a saturated hue in the bleached state in a lesser timespan, as current propagation is more homogeneous in this film.Although our color-bleach time values are longer than the colorand bleach times of 4.5 s shown for a copolymer of terthiopheneand EDOP corresponding to a full switch,4 but because a goodoptical contrast (shown in Figure 10b,d) is acquired in about

10 s, the devices are conducive for transmissive electrochromicapplications.

4. CONCLUSIONS

Composites of reduced graphene oxide layers with PEDOPwere prepared by an electrochemical method. We demonstratedthe incorporation of RGO/IL in PEDOP by XPS, on the basis ofnew signals due to S, N, and a completely altered C signal and byHRTEM, which revealed the composite to be composed ofrippled sheets of RGO-IL encapsulating the polymer betweenthe RGO/IL layers and extending over several hundred squaresof nanometers. An altered electronic structure for the compositewas also evidenced from the decreased valence band to the Fermiedge gap, on going from control PEDOP-IL to PEDOP-RGO/IL.Conductive-AFM showed that the high current flowing regionsdominate the surface of the PEDOP-RGO/IL composite, in starkcontrast to control PEDOP-IL, wherein the conducting regionsare predominantly flanked by insulating domains, which obstructelectron propagation. RGO/IL, in addition to serving as a dopantfor forming the polymer film, acts as an electron conduitconnecting the PEDOP grains, thus leading to a homogeneousdistribution of conducting regions in the PEDOP-RGO/ILcomposite. This was also supported by a 2-fold increment indc electronic conductivity, measured independently, andachieved in the PEDOP-RGO/IL composite. The benefit ofless-hindered and more uniform current pathways available tothe PEDOP-RGO/IL composite was clearly reflected in theimproved optical contrast, as the ease of electron ingress andegress during reduction and oxidation of the electrochromiccomposite allows higher ion uptake, thus imparting larger color-ing efficiency in the visible region, faster bleaching rate, andhigher electrochemical ion-insertion�extraction capacity incomparison to the case for control PEDOP-IL film. Furtherstudies are required to discern the role of the conducting polymerand RGO and IL interfaces in such composites, in modifying theelectronic transport behavior of the polymer.

’ASSOCIATED CONTENT

bS Supporting Information. Discussion and figures of SEMimages, XPS core level spectra, topography and conducting-AFMimages, I�V plots, and cyclic voltammograms of films. Thismaterial is available free of charge via the Internet at http://pubs.acs.org.

’AUTHOR INFORMATION

Corresponding Author*Tel. +91-4023016024. Fax. +91-40-23016003. E-mail: [email protected].

’ACKNOWLEDGMENT

Financial support from Department of Science & Technology(DST/TSG/PT/2007/69) is gratefully acknowledged. We thankMr. K. N. Sood for SEM measurements.

’REFERENCES

(1) Schottland, P.; Zong, K.; Gaupp, C. L.; Thompson, B. C.; Thomas,C. A.; Giurgiu, I.; Hickman, R.; Abboud, K. A.; Reynolds, J. R. Macro-molecules 2000, 33, 7051–7061.

Figure 10. Color-bleach characteristics of (a) control PEDOP-IL and(c) PEDOP-RGO/IL based devices recorded at a step time of 10 s atλ = 500 nm under( 3.0 V. (b) and (d) are the corresponding devices inoxidized (blue) and reduced (red) states.



(2) Sonmez, G.; Schottland, P.; Zong, K.; Reynolds, J. R. J. Mater.Chem. 2001, 11, 289–294.(3) Gaupp, C. L.; Zong, K.; Schottland, P.; Thompson, B. C.;

Thomas, C. A.; Reynolds, J. R. Macromolecules 2000, 33, 1132–1133.(4) Seol, H.; Jeong, H.; Jeon, S. J. Electroanal. Chem. 2009, 636,

107–112.(5) Walczak, R.M.; Reynolds, J. R.Adv. Mater. 2006, 18, 1121–1131.(6) Kharkwal, A.; Deepa, M.; Joshi, A. G.; Srivastava, A. K. Chem-

PhysChem 2011, 12, 1176–1188.(7) Lota, G.; Fic, K.; Frackowiak, E. Energy Environ. Sci. 2011,

4, 1592–1605.(8) Hou, Y.; Cheng, Y.; Hobson, T.; Liu, J. Nano Lett. 2010, 10,

2727–2733.(9) Peng, C.; Jin, J.; Chen, G. Z. Electrochim. Acta 2007, 53,

525–537.(10) Nystrm, G.; Mihranyan, A.; Razaq, A.; Lindstrm, T.; Nyholm,

L.; Strømme, M. J. Phys. Chem. B 2010, 114, 4178–4182.(11) Choi, Y. S.; Joo, S. H.; Lee, S.-A.; You, D. J.; Kim, H.; Pak, C.;

Chang, H.; Seung, D. Macromolecules 2006, 39, 3275–3282.(12) Bhandari, S.; Deepa, M.; Srivastava, A. K.; Joshi, A. G.; Kant, R.

J. Phys. Chem. B 2009, 113, 9416–9428.(13) Peng, C.; Snook, G. A.; Fray, D. J.; Shaffer, M. S. P.; Chen, G. Z.

Chem. Commun. 2006, 4629–4631.(14) Chang, H.; Wang, G.; Yang, A.; Tao, X.; Liu, X.; Shen, Y.;

Zheng, Z. Adv. Funct. Mater. 2010, 20, 2893–2902.(15) Green, A. A.; Hersam, M. C. J. Phys. Chem. Lett. 2010, 1,

544–549.(16) Park, S.; Ruoff, R. S. Nature Nanotechnol. 2009, 4, 217–224.(17) Si, Y.; Samulski, E. T. Nano Lett. 2008, 8, 1679–1683.(18) Jiao, L.; Wang, X.; Diankov, G.; Wang, H.; Dai, H. Nature

Nanotechnol. 2010, 5, 321–325.(19) Wang, Y.; Yao, H.-B.; Wang, X.-H.; Yu, S.-H. J. Mater. Chem.

2011, 21, 562–566.(20) Liu, K.; Zhang, J.; Yang, G.; Wang, C.; Zhu, J.-J. Electrochem.

Commun. 2010, 12, 402–405.(21) Zhang, L. L.; Zhao, S.; Tian, X. N.; Zhao, X. S. Langmuir 2010,

26, 17624–17628.(22) Liu, N.; Luo, F.; Wu, H.; Liu, Y.; Zhang, C.; Chen, J. Adv. Funct.

Mater. 2008, 18, 1518–1525.(23) Lu, J.; Yang, J.-X.; Wang, J.; Lim, A.; Wang, S.; Loh, K. P. ACS

Nano 2009, 3, 2367–2375.(24) Choi, K. S.; Liu, F.; Choi, J. S.; Seo, T. S. Langmuir 2010,

26, 12902–12908.(25) Xu, Y.; Wang, Y.; Liang, J.; Huang, Y.; Ma, Y.; Wan, X.; Chen, Y.

Nano Res. 2009, 2, 343–348.(26) Matyba, P.; Yamaguchi, Y.; Chowalla, M.; Edman, L.; Robinson,

W. D. ACS Nano 2010, 4, 637–642.(27) Zhang, L. L.; Zhou, R.; Zhao, X. S. J. Mater. Chem. 2010, 20,

5983–5992.(28) Wang, D.-W.; Li, F.; Zhao, J.; Ren, W.; Chen, Z.-G.; Tan, J.; Wu,

Z.-S.; Gentle, I.; Lu, G. Q.; Cheng, H.-M. ACS Nano 2009, 3, 1745–1752.(29) Saxena, A. P.; Deepa, M.; Joshi, A. G.; Bhandari, S.; Srivastava,

A. K. ACS Appl. Mater. Interfaces 2011, 3, 1115–1126.(30) Pigani, L.; Heras, A.; Colina, A.; Seeber, R.; L-Palacois, J.

Electrochem. Commun. 2004, 6, 1192–1198.(31) Efimov, I.; Winkels, S.; Schultze., J. W. J. Electroanal. Chem.

2001, 499, 169–175.(32) Meyer, J. C.; Geim, A. K.; Katsnelson, M. I.; Novoselov, K. S.;

Booth, T. J.; Roth, S. Nature 2007, 446, 60–63.(33) Bhandari, S.; Deepa, M.; Pahal, S.; Joshi, A. G.; Srivastava, A. K.;

Kant, R. ChemSusChem 2010, 3, 97–105.(34) Reina, A.; Jia, A.; Ho, J.; Nezich, D.; Son, H.; Bulovic, V.;

Dresselhaus, M. S.; Kong, J. Nano Lett. 2009, 9, 30–35.(35) Kim, K. S.; Zhao, Y.; Lee, S. Y.; Kim, J. M.; Ahn, J.-H.; Kim, P.;

Choi, J.-Y.; Hong, B. H. Nature 2009, 457, 706–710.(36) Gao, W.; Alemany, L. B.; Ci, L.; Ajayan, P. M. Nature Chem.

2009, 1, 403–408.

(37) Stankovich, S.; Dikin, D. A.; Piner, R. D.; Kohlhaas, K. A.;Kleinhammes, A.; Jia, Y.; Wu, Y.; Nguyen, S-B. T.; Ruoff, R. S. Carbon2007, 45, 1558–1565.

(38) Yang, H.; Shan, C.; Li, F.; Han, D.; Zhang, Q.; Niu, L. Chem.Commun. 2009, 3880–3882.

(39) Dreyer, D. R.; Park, S.; Bielawski, C. W.; Ruoff, R. S. Chem. Soc.Rev. 2010, 39, 228–240.

(40) Caporali, S.; Bardi, U.; Lavacchi, A. J. Electron Spectrosc. Relat.Phenom. 2006, 151, 4–8.

(41) Chandrachud, P.; Pujari, B. S.; Haldar, S.; Sanyal, B.; Kanhere,D. G. J. Phys.: Condens. Matter 2010, 22, 465502 (1–11).

(42) Ferrari, A. C.; Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri,M.; Mauri, F.; Piscanec, S.; Jiang, D.; Novoselov, K. S.; Roth, S.; Geim,A. K. Phys. Rev. Lett. 2006, 97, 187401–187404.

(43) Chen, W.; Yan, L. Nanoscale 2010, 2, 559–563.(44) Lee, H. J.; Park, S.-M. J. Phys. Chem. B 2004, 108, 1590–1595.(45) Han, D.-H.; Kim, J.-W.; Park, S.-M. J. Phys. Chem. B 2006,

110, 14874–14880.(46) Lee, H. J; Lee, J.; Park, S.-M. J. Phys. Chem. B 2010,

114, 2660–2666.(47) Gaupp, C. L.; Welsh, D. M.; Rauh, R. D.; Reynolds, J. R. Chem.

Mater. 2002, 14, 3964–3970.(48) Sonmez, G.; Meng, H.; Wudl, F. Chem. Mater. 2004,

16, 574–580.(49) Gaupp, C. L.; Welsh, D. M.; Reynolds, J. R. Macromol. Rapid

Commun. 2002, 23, 885–889.

poly(3,4-ethylenedioxypyrrole) enwrapped by reduced graphene oxide: how conduction behavior at...

Documents