Electrochimica Acta 161 (2015) 108–114

Rudimentary simple method for the decoration of graphene oxide withsilver nanoparticles: Their application for the amperometric detectionof glucose in the human blood samples

Aditee C. Joshi, Ganesh B. Markad, Santosh K. Haram*Department of Chemistry, University of Pune, Pune 411 007, India

A R T I C L E I N F O

Article history:Received 18 October 2014Received in revised form 1 January 2015Accepted 8 February 2015Available online 11 February 2015

Keywords:Glucose sensorGraphene OxideSilver nanoparticlesCyclic Voltammetry

A B S T R A C T

Graphene oxide decorated with silver nanoparticles (GO-Ag) was prepared by anodic dissolution of silverin the aqueous dispersion of GO. The composites were characterized by XRD, XPS, TEM, AFM, and Ramanspectroscopy. The electrooxidation of glucose on GO-Ag modified electrodes have been tested by cyclicvoltammetry and chronoamperometry. The detail mechanism of redox processes on the GO-Agelectrodes has been studied. A few mg loading of silver has demonstrated to give current in mA for the mMconcentration of glucose. A linear relationship between peak height in the voltammograms and glucoseconcentration in the range 1–14 mM has been proposed for amperometric detection of glucose. From theresults, the detection limit for glucose sensing is estimated to be as small as 4 mM. The selectivity forglucose in presence of interfering molecules viz. ascorbic and uric acids is tested. Proof-of concept ispresented by carrying out the measurements in real human blood samples.

ã 2015 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Electrochimica Acta

journal homepa ge: www.elsev ier .com/locate /e lectacta

1. Introduction

Glucose estimation in blood samples has been exceedinglyimportantwith an advent of alarmic rate of Diabetes mellitus affectedpatients (�150 million people), world-wide [1]. Self-monitoring ofthe blood glucose level is one of the key-factors in controlling thisdecease. In this context, the devices based on glucose oxidase (GOx)modified electrodes have been extremely popular for last severaldecades [2]. Unfortunately, GOx, being protein is vulnerable todenature at high temperature, low humidity and change in pH. Inthis scenario, non-enzymatic glucose sensors are highly desirableand thus deserves significant attention. Glucose is known to getelectrochemically oxidize over various metal oxide electrodes, viz.CuO, NiO, Ag/CuO and Co3O4 [3–6], which leads to a detectableamperometric signal. Being poor conductor most of these oxidespossess limitation to develop good electrochemical sensors. One ofthe ways to overcome this limitation is to disperse them on theconducting support which would not only prevent them toagglomerate but also would provide conducting pathway. Amongvarious materials, crystalline carbons viz. carbon nanotubes andgraphene would be the most appropriate choice for this purpose.

* Corresponding author. Tel.: +91 2025691373; fax: +91 20 2569 3981.E-mail address: haram@chem.unipune.ac.in (S.K. Haram).

http://dx.doi.org/10.1016/j.electacta.2015.02.0770013-4686/ã 2015 Elsevier Ltd. All rights reserved.

Between these two, lots of studies are available in the literature,about the use of CNT-Cu [7], CNT-Pt [8] and MWCNT-Ni [9]composites for glucose sensing applications. However, the litera-ture regarding silver graphene based materials are relatively rareand lot of room for further development is available. With thisbackground, we have undertaken the detail investigation regardingdevelopment of GO-silver nanoparticle composite based materialfor the amperometric detection of glucose. Following properties ofindividual Ag and GO and their likely synergistic interaction hasmotivated us to choose this combination.

In case of Ag, its oxides are formed in situ at sufficiently positivepotentials [10,11]. AgO so-formed, chemically reacts quantitativelywith glucose to yield Ag2O, which shows distinct anodic peak in areverse cycle; proportional to the glucose concentration. Moreover,kinetics of formation of silver oxides is very facile [12] which isadvantageous in minimizing the response time. Furthermore, Ag2Ohas inherent good electrical properties (ca. 20 S cm�1) due to theoxygen vacancies at room temperature, which would be advanta-geous to minimize the current losses and help in improving thesensitivity.

Graphene and its analogues i.e.graphene oxide and reducedgraphene oxide have several desirable properties such as highsurface area, electronic conductivity and, chemical stability in theoxidative environment. GO provides chemical and mechanicalstable marix for the metal nanoparticles. Thus, most of the surface

A.C. Joshi et al. / Electrochimica Acta 161 (2015) 108–114 109

area of the nanoparticles available for the reaction whichotherwise would have occupied by capping agents. Moreover,GO and r-GO are demonstrated to undergo charge transfer withvarious metals and semiconductor nanoparticles. This synergisticproperty of metal nanoparticles with GO make them promisingcombination in the electrocatalysis [13].

In the current work, we demonstrate preparation of GO-Agcomposite with rudimentary simple electrochemical approachwhich we have already successfully tested before for MWCNTs-Agcomposites [14]. GO-Ag electrode showed excellent selectivity andrapid detection for glucose with good sensitivity, indicatingpotential use of GO-Ag in fabricating enzyme-free glucose sensor.

2. Experimental

2.1. Preparation of Graphene Oxide (GO)

GO samples were prepared by Hummers method [15] with aslight modification [16]. In brief, graphite powder (0.5 g), sodiumnitrate (0.5 g) and sulfuric acid (23 ml) were mixed in an ice- bathunder a continuous stirring. Potassium permanganate (3.0 g) wasslowly added into the reaction mixture at 20 �C. Flask was thentransferred to water bath (35 � 5 �C) and solution was stirred for anhour to get thick pasty product. 100 ml water was added andtemperature of the bath was raised to 90 � 5 �C under constantstirring for another 15 min. The solution was diluted by adding500 ml water and 3 ml H2O2 (30%) which led to color change fromdark brown to yellow. The mixture was filtered and washed severaltimes with hot water to eliminate the acid residue. The resultantsolid was dried under vacuum and stored in a desiccator forsubsequent use.

2.2. Preparation of Graphene Oxide decorated with Silvernanoparticles.

GO-Ag composite was prepared by electrochemical depositionof silver nanoparticles on graphene oxide sheets. Dispersion of GOwas prepared by sonicating GO sheets (0.1 mg ml�1) in MilliQ1

water for 30 minutes. Silver electrodes (diameter ca. 2 mm) werepolished using emery paper and rinsed with DI water. Thecomposites were prepared by electrolysis of GO dispersion withsilver electrodes at of 20 V for 30 minutes. The color of solution waschanged from brown to colorless, followed by grey deposit on thecathode (negative electrode) and dark brown adherent film onanode (positive electrode). The cathode deposit was separatedfrom solution by centrifuging at 5000 rpm for 10 minutes. In orderto compare the influence of graphene oxide, the control experi-ment was performed on silver nanoparticles (AgNps) without GO.For that, AgNps were prepared in water keeping all otherparameters same without adding graphene oxide. In this experi-ment as the reaction progresses the solution became pale yellow incolor indicating formation of silver nanoparticles.

2.3. Material Characterization

UV–visible spectra were recorded using an Agilent 8453 diodearray single beam spectrophotometer. Powder X-ray diffracto-grams (XRD) were recorded on the dried product using a Bruker,D8-Advance, X-ray diffractometer (Cuka, 40 kV and 40 mA).Transmission electron microscopic (TEM) images were recordedon the samples using a Technai G2 20 V TWIN transmissionelectron microscope (20–200 kV). X-ray photoelectron spectros-copy (XPS) was performed using ESCA-3000 VG Scientific Ltd.(England) spectrometer with monochromated Al Ka source(1486.7 eV). AFM images were recorded using Nano Wizard13-AFM (JPK instruments) on a mica substrate. Raman spectra were

recorded at room temperature using the 633 nm line, from a He-Nelaser. The laser beam was focused onto a spot �1 mm in diameter,and the laser power at the sampling position was 10 mW. TheRaman band of a silicon wafer at 520 cm�1was used to calibrate thespectrometer, and the accuracy of the spectral measurement wasestimated to be better than 1 cm�1. For the Raman measurements,the sample solutions were taken in a standard 1 �1 cm2 cuvette oron glass slides and the Raman signal was collected at 180�

scattering geometry or with a 50X LWD (long working distance)objective and detected using a CCD (Synapse, Horiba Jobin Yvon)based monochromator (LabRAM HR800, Horiba Jobin Yvon,France) together with an edge filter, covering a spectral range of200–1800 cm�1.

2.4. Electrochemical Measurements

The electrochemical measurements were performed with BioLogic Potentiostat/galvanostat (SP-300) workstation in a typicalthree electrodes system; Glassy carbon as a working electrode, Hg/HgO/sat. Ca (OH)2 as a reference electrode and Pt-wire as a counterelectrode.

2.5. Preparation of GO-Ag modified electrode.

Electrochemical activity for composite was studied by dropcasting the dispersion on GC electrode. For this purpose, GCelectrode (3 mm diameter) was polished with alumina powder(0.5 mm) and rinsed with MilliQ1 water. GO-Ag deposit formed atcathode was transferred in MilliQ1 water (1 mg/ml) and sonicatedtill a uniform dispersion was formed. From this an aliquot of 40 mlwas drop casted onto precleaned GC and vacuum dried at roomtemperature. Similarly, an aliquot of 40 ml AgNps was drop castedon the GC electrode and electrochemical measurements wereperformed.

2.6. Determination of blood glucose using screen printed electrodes.

To verify feasibility of sensor, GO-Ag composite was employedto estimate glucose in human blood samples. For this purposescreen printed three electrode system (Zensor TE100 SPE’s)comprising of carbon as working electrode, Ag/AgCl as referenceelectrode and Platinum as counter electrode were used. Thecomposite was drop-casted (5 ml volume) on working electrodeand dried at room temperature. Onto it 20 ml of 0.1 M NaOH wasadded and background was recorded. Further 20 ml blood drop wasadded to 20 ml of 0.1 M NaOH and amperometric response forblood glucose was recorded at 0.6 V. All the safety protocols arefollowed in handling, testing and disposing of human bloodsamples.

3. Result and Discussion

3.1. Material Characterization of GO-Ag.

Formation of GO-Ag was confirmed by performing variousstructural characterizations. The morphology of the productexamined by TEM, as indicated in Fig. 1a revealed a decorationof about 50 � 5 nm diameter spherical silver nanoparticles overgraphene oxide sheets. Close examination revealed that someparticles are having hexagonal shape and their facets are clearlylegible in the micrograph (Inset Fig. 1). This was further supportedby AFM results of the composite showing graphene sheetsdecorated uniformly with silver nanoparticles (Fig. 1b).

Crystallographic studies of composite were carried out by XRDanalysis. Fig. 1c shows a typical powder diffractogram recorded onGO-Ag sample. The observed peaks at 38.33�, 44.48�, 64.68�, and

Fig. 2. UV–Visible spectrum of aqueous dispersion of GO-Ag composite.

Fig. 1. Material characterization performed on GO-Ag composite. (a) TEM image with an inset shows higher magnification portion. (b) AFM image, (c) XRD and (d) Ramanspectra.

110 A.C. Joshi et al. / Electrochimica Acta 161 (2015) 108–114

77.6�are indexed to (111), (2 0 0), (2 2 0), and (311) planes of facecentered cubic silver (JCPDS-4-783), respectively. The expectedpeaks due to GO at 11� were not observed in XRD which perhapsgot masked due to strong reflection from silver. The chemicalstructure of GO and GO-Ag is studied by Raman Spectroscopy. Asshown in Fig. 1d GO has two peak positions at 1358 cm�1 (D band)and 1601 cm�1 (G band). After GO-Ag composite formationincrease in intensity ratio for D/G (1.1) is observed. This confirmssuccessful deposition of silver on graphene oxide sheets.

Fig. 2 shows a typical UV–vis spectrum recorded on the aqueousdispersion of GO-Ag. Prominent peaks at 256 nm and 405 nm areattributed to GO and surface plasmon resonance (SPR) for thesilver, respectively [16,17]. Fig. 3 shows XPS spectra recorded oncomposite in 3d spectral region of silver. The peaks at 368.5 eV and374.07 eV are assigned to 3d5/2 and 3d3/2 of Ag respectively. Theobserved peaks are de-convoluted into two different pairs; strongpeaks at 367.9 eV and 374.2 eV are attributed to the Ag (I) state andweak peaks at 368.2 eV and 373.6 eV are attributed to the Ag (0)phase [18]. Occurrence of Ag (I) phase suggests the presence ofAg-O species on GO-Ag surface. All these results confirmsformation of GO-Ag nanocomposite in which ca. 50 nm diametersilver nanoparticles are decorated on GO surface and having Ag-Ospecies on surface.

3.2. Electrochemical Characterization of GO-Ag

The electrocatalytic activity of GO-Ag and AgNps was investi-gated by cyclic voltammetry (CV) and chronoamperometry (CA).For this purpose aliquots of the dispersed sample were drop-casted

onto pre-cleaned GC electrode and measurements wereperformed. Fig. 4 represents comparative CV recorded on boththe samples in potential range �0.7 V to 0.9 V at 100 mV s�1 scanrate. For AgNps (top) four distinct anodic peaks (marked as A1, A2,A0

2, A3) were observed in the forward scan. Similar observation wasmade in case of GO-Ag samples (bottom) except peak A1. In thereverse scan the anodic peak around 0.6 V is observed in both the

Fig. 3. XPS spectra in Ag-3d region for GO-Ag composite. Hollow black circlesrepresent the experiemental data. The yellow and purple curves are fittings in theAg(I) and green, and blue curves are the fitting for Ag(0) phases. The black solid lineindicates a resultant of the best fit of all the four curves. (For interpretation of thereferences to color in this figure legend, the reader is referred to the web version ofthis article.)

A.C. Joshi et al. / Electrochimica Acta 161 (2015) 108–114 111

samples (peak A4). In further scanning towards negative potentialsboth the samples shows two peaks namely C1 and C2 and ashoulder, C0

2. Control sample (GO without silver) did not showsimilar peaks (Fig. S1)

The peak A1 observed prominently in AgNps is ascribed toformation of Ag (OH)2 layer by electrochemical reaction of Ag withOH� ions from the solution. This peak is not prominent in case ofGO-Ag. Most probably OH� group from graphene are alreadyinvolved with GO-Ag because of which further electrochemicalreaction of that Ag with OH� ions is less likely to be occurred.Therefore, the Ag associated with GO is more or less occupied by

Fig. 4. CVs recorded on (a) Ag nanoparticles and (b) GO-Ag composite drop-castedon GC electrode. The electrolyte was 0.1 M NaOH. The scan rate was 100 mV s�1.

Ag-O-C species and very less amount of native Ag is available forelectrochemical formation of Ag (OH)2.

The peak A2 and A02 observed in both the samples are indicative

of formation of Ag2O monolayer and further formation of Ag2Omultilayer respectively. In case of GO-Ag shift of peak by +0.08 V asthat of AgNps is attributed to formation of multilayer Ag2O whichis more difficult to form in case of GO-Ag and again attributed tointeraction of Ag with GO. Due to similar reason, the oxidation ofAg2O to AgO at A3 is shifted by +0.08 V. In reverse potential sweep,the observed peak at 0.6 V (marked as A4) is attributed toreoxidation of Ag2O to AgO. The peak C1 is complementary to peakA3 that is assigned to formation of Ag2O (plausibly multilayer) fromAgO. Up on further scanning the potential toward more negativepotential, a prominent cathodic peak, convoluted with a shoulder(marked as C2 and C’2) is seen. Potential clipping CV measurements(Refer supporting information Fig. S3) suggested that C2 and C2’ arecomplimentary to the anodic peaks A2 and A’2, respectively.Therefore, the reduction of Ag2O (monolayer) and Ag2O (multi-layers) to Ag(0) phase at C2 and C0

2, respectively is concluded.Fig. 5a shows a CV response recorded on AgNps before and

after addition of glucose. Prominent increase in peak heights of A1,A2 and A0

2 were observed upon glucose addition. We attributethis increase to the chemical reaction of glucose with nativesilver oxide which leads to the formation of more Ag on the surface.This silver subsequently gets oxidised into silver oxides accompa-nying with increase in the anodic current at A1, A2 and A0

2.Furthermore, upon addition of glucose the disappearance of peakC2 was noted. The potential clipping experiment in this region(Refer supporting information figure S3) suggested that peak C2 iscomplimentary to A2 at which monolayer of Ag2O was formedduring anodic cycle. Thus disappearance of C2 suggests thatmonolayer of Ag2O get vanished upon glucose addition. Perhapsglucose reduces monolayer preferentially immediately after itsformation.

Fig. 5. CVs recorded on (a) silver nanoparticles (b) GO-Ag composite, drop-castedon glassy carbon electrode. Black curve is for without glucose and the red curve isfor 4.0 mM glucose. The electrolyte was 0.1 M NaOH. (For interpretation of thereferences to color in this figure legend, the reader is referred to the web version ofthis article.)

112 A.C. Joshi et al. / Electrochimica Acta 161 (2015) 108–114

Fig. 5b shows CVs obtained on GO-Ag in presence and absenceof glucose. Interestingly peak A2 and A0

2 were not disappearedcompletely and still seen after glucose addition. This observation isopposite to that of noted in case of AgNps (Refer Fig. 5a anddiscussion therein). In case of GO-Ag the presence of Ag-O complexwas noted in XPS (Refer Fig. 3 and discussion therein). Probably,glucose is not able to reduce this complex thus formation of silverupon addition of glucose is discouraged.

In both the cases decrease in peak height of A3 upon of glucoseaddition is observed. Ag2O so formed gets reduced chemically byglucose to Ag. So relatively less Ag2O is available for the furtheroxidation.

In presence of glucose, peak A4 became more prominent in boththe cases. This increase in peak height indicates formation of moreAg2O due to the chemical reduction of AgO by glucose. The peakheight A4 is further found to be proportional to the glucoseconcentration.

Fig. 6 shows CV recorded with varied glucose concentrations(1–14 mM) for AgNps (40 mg in 40 ml) (Fig. 6a) and GO-Ag (Fig. 6b).Since chemical reaction of glucose with silver oxides is continu-ously occurring in solution AgO formed at A4 get reducedchemically which impact on peak height of C1 and subsequentlyC2 and C‘2. These changes are more prominent in case of GO-Agthan AgNps which suggests that oxidation of glucose on GO-Ag ismore facile than Ag Nps. All these measurements have been carriedout with 12 mg loading of GO-Ag which suggests outstandingsensitivity of composite for the glucose detection.

3.3. Amperometric Detection of Glucose

For the amperometric detection experiment, a potential step of+0.60 V was applied and the current response was monitored bystandard addition of glucose in 0.1 M NaOH solution at a fixed timeinterval. Fig. 7a shows amperometric response of GO-Ag for

Fig. 6. CV’s recorded for various glucose concentrations in the range of 1–14 mM for(a) Ag nanoparticles (b) GO-Ag composite drop-casted on glassy-carbon electrode.The inset shows magnified view in region of peak A4 for curves.

Fig. 7. Amperometric response for 1–14 mM glucose concentrations on (a) GO-Agcomposite (b) Ag nanoparticles. Insets show calibration curves for respectivesamples showing current variation with concentration and per mg loading of silver.

glucose concentration ranging from 1–14 mM. The limiting currentvalues were noted after each addition and normalized to currentper mg of silver loaded is plotted (refer to Fig. 7 Inset) againstglucose concentration. The current was found to increase linearlyand data is fitted in a straight line (R2 = 0.9741). The sensitivity andlimit of detection were noted to be 11 mA mM�1 cm�2mg�1 and4 mM, respectively. In a contrary, in case of pure AgNps (Fig. 7b)linear variation of current was obtained for 1–14 mM (R2 = 0.9986)with sensitivity and detection limit of 0.5 mA mM�1 cm�2mg�1and6 mM, respectively.

The comparison of our results with the recent reports on GObased non-enzymatic glucose sensor are presented in Table 1. Themoderate sensitivity in this case is attributed to the in situformation of silver oxides which make system dynamic but slow. Inthe reported cases, the metal-oxides were used and glucose gets

Table 1Comparison of different non- enzymatic glucose sensors, based on Graphene Oxidecomposites.

Materials Sensitivity Detection Limit Ref.

Ag/CuO NF 1347 mA mM�1 cm�2 51 nM [5]3-D Graphene-Co3O4 3.39 mA mM�1 cm�2 25 nM [6]rGO-/Pt-CuO 3577 mA mM�1 cm�2 0.01 mM [19]Pt/N-GSS 6.36 mA mM�1 cm�2 1 mM [20]Ni(OH)2/rGO 2042 mA mM�1 cm�2 2.7 mM [21]GO-Ag 135 mA mM�1/ cm�2

(11 mA/mM cm�2mg)4 mM Our work

Fig. 8. Amperometric response for human blood sample on Ag-GO modified screenprinted electrodes. (a) Red curve for Ag-GO composite. The black curve is theresponse recorded on un-modified screen printed electrode (the control) (b) Thecomparision of current response from blood sample with glucose solution on Ag-GOmodified screen printed electrode. (For interpretation of the references to color inthis figure legend, the reader is referred to the web version of this article.)

A.C. Joshi et al. / Electrochimica Acta 161 (2015) 108–114 113

oxidize directly on them. The advantages of GO-Ag system,however has been few mgs of silver-loading is sufficient to seethe current response of �135 mA mM�1. Furthermore, repeatabilityobserved in the potential-cycles assures that the catalyst is stableand self-regenerative during repeated oxidation.

In addition to this an attempt was made to study response ofGO-Ag composite in detecting blood glucose. It is very crucial todetect glucose in a smaller volume of blood. Fig. 8a representscurrent response on GO-Ag modified electrode after addition of20 ml blood drop. After addition of blood sample substantialincrease in current response was observed in case of GO-Ag, similarresponse was not observed in case of electrode without GO-Ag

Fig. 9. Amperometric response of the GO-Ag composite to the stepwise addition of0.1 mM uric acid and ascorbic acid followed by addition of 1 mM glucose.

(Refer Fig. 8a bottom black curve). Fig. 8b shows chronoampero-grams obtained for blood glucose response overlapped with100 mM stock glucose concentration. Current response obtainedfor blood glucose was similar to 6 mM standard glucoseconcentration. This indicates the total amount of glucose presentin blood is nearly equal to 6 mM. For smaller volume of blood (FigS2) the approximate estimation of glucose was around 4 mM.

3.4. Interference Studies on GO-Ag Modified Electrode

In human blood serum concentrations of ascorbic acid (AA) anduric acid (UA) are significant (ca. 0.1 mM and 0.02 mM, respective-ly), they interfere in the measurements due to high rate constantsfor their oxidations. Fig. 9 shows a current-time profile for GO-Agmodified electrode for 0.1 mM of AA and UA and subsequentaddition of 1 mM glucose. The response for UA and AA wasmarginal whereas for glucose total current changed by a factor of 4.The results indicate high selectivity of GO-Ag electrode towardsglucose against the interfering components.

4. Conclusion

We demonstrate a simple electrochemical route for synthesis ofsilver nanoparticles decorated on graphene oxide. The compositehas proved to be a superior catalyst material for electrochemicalglucose oxidation. The total response of the sensor has improvedsignificantly as compared to pure silver nanoparticles withsensitivity 11 mA cm�2mM�1mg�1 and limit of detection 4 mM.This is mainly due to synergistic interaction between the silvernanoparticles and GO. The sensor was found to be specific towardsglucose and displayed a linear range between 1–14 mM. Further-more, sensor showed feasibility towards detection of glucose inreal blood samples. The results prove potential use of GO-Ag indeveloping enzyme-less amperometric glucose sensors.

Acknowledgment

Authors are thankful to Dr. Kashinath Patil from NCL for XPSanalysis. Aditee Joshi would like to acknowledge UGC-.Dr. D.SKothari PDF program for the financial support. Ganesh Markadthanks BARC-Pune University Collaborative Program for thefellowship. Author Thanks BRNS for the financial support.

Appendix A. Supplementary data

Supplementary data associated with this article can befound, in the online version, athttp://dx.doi.org/10.1016/j.electacta.2015.02.077.

References

[1] Z. Zhu, L. Garcia-Gancedo, A. Flewit, H. Xie, F. Moussy, W. Milne, Sensors 12(2012) 5996.

[2] R. Prehn, M.C. Puig, F.X. Muñoz, J. Electrochem. Soc. 159 (2012) F134.[3] C.L. Sun, W.L. Cheng, T.K. Hsu, C.W. Chang, J.L. Chang, J.M. Zen, Electrochem.

Commun 30 (2013) 91.[4] G. Wang, X. Lu, T. Zhai, Y. Ling, H. Wang, Y. Tong, Yat Li, Nanoscale 4 (2012) 3123.[5] B. Zheng, G. Liu, A. Yao, Y. Xiao, J. Du, Y. Guo, D. Xiao, Qin Hu Choi, M.F. Martin,

Sens. Actuators B 195 (2014) 431.[6] X.C. Dong, H. Xu, X.W. Wang, Y.X. Huang, B. Chan-Park, Mary, H. Zhang, L.H.

Wang, W. Huang, P. Chen, ACS Nano 6 (2012) 3206.[7] J. Zhao, L. Wei, C. Peng, Y. Su, Z. Yang, L. Zhang, H. Wei, Y. Zhang, Biosens.

Bioelectron. 47 (2013) 86.[8] H.W.Chang,Y.C.Tsai,C.W.Cheng,C.Y.Lin,P.H.Wu,Sens.ActuatorsB183(2013)34.[9] K.C. Lin, Y.C. Lin, S.M. Chen, Electrochim. Acta 96 (2013) 164.

[10] T.P. Tougas, M.J. Debenedetto, J.M. DeMott Jr., Electroanalysis 5 (1993) 669.[11] Demott Jr., E.D.G.E. Jahngen, Electroanalysis 17 (2005) 599.[12] H. Ng Kwok, H. Liu, R.M. Penner, Langmuir 16 (2000) 4016.[13] I.V. Lightcap, T.H. Kosel, P.V. Kamat, Nano Lett. 10 (2010) 577.[14] K.M. Samant, V.R. Chaudhari, S. Kapoor, S.K. Haram, Carbon 45 (2007) 2126.

114 A.C. Joshi et al. / Electrochimica Acta 161 (2015) 108–114

[15] W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc. 80 (1958) 1339.[16] R. Chadha, N. Biswas, G.B. Markad, S.K. Haram, T. Mukherjee, S. Kapoor, Chem.

Phy. Lett. 529 (2012) 54.[17] J. Li, C.Y. Liu, Y. Liu, J. Mater. Chem. 22 (2012) 8426.[18] S. Maenoson, J. Lee, A.T.N. Dao, D. Mott, Surf. Interface Anal. 44 (2012) 1611.

[19] K. Dhara, J. Stanley, T. Ramachandran, B.G. Nair, S. Babu, T.G. Sens, Actuators B195 (2014) 197.

[20] L. Meng, Y. Xia, W. Liu, P. Li Zhang a, Zou, Y. Zhang, Electrochimica Acta 152(2015) 330.

[21] W. Gao, W. Weei Tjiu, J. Wei, T. Liu, Talanta 120 (2014) 484.