Pfs

Ka

b

a

ARRAA

KNRCPH

1

picnHo

o[tgstlinurt

0h

Sensors and Actuators B 195 (2014) 197–205

Contents lists available at ScienceDirect

Sensors and Actuators B: Chemical

jo ur nal home page: www.elsev ier .com/ locate /snb

t-CuO nanoparticles decorated reduced graphene oxide for theabrication of highly sensitive non-enzymatic disposable glucoseensor

eerthy Dharaa, John Stanleyb, Ramachandran Ta, Bipin G. Nairb, Satheesh Babu T.Ga,∗

Department of Sciences, Amrita School of Engineering, Amrita Vishwa Vidyapeetham, Amritanagar, Coimbatore 641112, IndiaAmrita School of Biotechnology, Amrita Vishwa Vidyapeetham, Amritapuri, Clappana P.O. , Kollam 690525, India

r t i c l e i n f o

rticle history:eceived 6 November 2013eceived in revised form 6 January 2014ccepted 12 January 2014vailable online 21 January 2014

a b s t r a c t

Platinum nanocubes and copper oxide nanoflowers decorated reduced graphene oxide (rGO) obtained byone step chemical process. X-ray crystallographic analysis confirms that CuO in monoclinic form and Ptin cubic crystal form. Pt-CuO/rGO nanocomposite dispersed in N,N-dimethylformamide (DMF) was dropcasted onto the working electrode of an indigenously fabricated screen printed three electrode system.Oxidation of glucose on the Pt-CuO/rGO nanocomposite modified screen printed electrode (SPE) was

eywords:on-enzymatic glucose sensoreduced graphene oxideopper oxide nanoflowerlatinum nanocubes

occurred at +0.35 V. The sensor showed a limit of detection 0.01 �M (S/N = 3) and very high sensitivityof 3577 �A mM−1 cm−2 with linear response upto 12 mM. The sensor was highly selective to glucosein the presence of commonly interfering species like ascorbic acid (AA), dopamine (DA), uric acid (UA)and acetaminophen. The sensor was employed for the testing of glucose in blood serum and the resultsobtained were comparable with other standard test methods.

ummer’s method

. Introduction

Diabetic mellitus, considered as the second major public healthroblem worldwide, is a metabolic disorder due to the deficiency of

nsulin resulting in elevated blood glucose levels. Long term compli-ations of diabetes include retinopathy with potential loss of vision;ephropathy leading to renal failure and cardiovascular diseases.ence there is a great demand for a fast, reliable and low cost pointf care monitoring of blood glucose level [1–3].

Development of glucose sensors became one of the hot areasf research after the pioneering work by Clarks and Lyons in 19624,5]. Due to their excellent selectivity and high sensitivity towardshe oxidation of glucose, glucose oxidase (GOx) based sensorsained immense popularity [6–8]. However enzyme based sensorsuffer from draw backs including instability to changes in the pH,emperature, etc., and the need for complicated enzyme immobi-ization procedures [9,10] and higher cost associated with enzymesolation and purification. To address these limitations, an alter-ative approach based on the direct electro-oxidation without the

se of enzyme, ‘non-enzymatic’ has been tried [11]. Choosing theight catalyst for direct electrochemical activity is the key step inhe fabrication of non-enzymatic glucose sensors.

∗ Corresponding author. Tel.: +91 944236863; fax: +91 422 2656274.E-mail addresses: tgsatheesh@gmail.com, tg satheesh@cb.amrita.edu (S.B. T.G).

925-4005/$ – see front matter © 2014 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.snb.2014.01.044

© 2014 Elsevier B.V. All rights reserved.

The direct electrooxidation of glucose on different noble metalsubstrates such as platinum [12], gold [13], palladium [14] andalloys of Pt, Pd, Pb and Rh [15–17] has been explored for thedevelopment of non-enzymatic glucose sensor. However, theseelectrodes suffer from low sensitivity, poor selectivity and areprone to loss of activity due to absorbed intermediates and poi-soning by chloride ions [18]. Metals such as copper, nickel, zincand manganese have attracted the attention of researchers dueto their low cost, good electrocatalytic activity and the possibil-ity of promoting electron transfer reactions at lower overpotential[19,20]. Due to narrow band gap and easy to tune nano architec-ture, a wide variety of copper oxide based materials have beeninvestigated for their glucose sensing capabilities. Copper oxidein different morphologies, such as nanoporous [21], nanoparticles[22], nanospheres [23], nanowires, nanoflowers [24], nanorods [25]and ball-in ball microspheres [26] have been explored. In orderto improve the performance of these sensor, the role of carbonnanotubes (CNT) as support matrices for metal and metal oxidenanoparticle composite [27–31] has been investigated.

Graphene one atom thick two dimensional array of carbonatoms packed in a dense honeycomb crystal structure attractedtremendous attention due to its excellent electrical and mechan-ical properties [32], large surface area, high stability and low cost

[33,34]. They have also been employed for the development of elec-trochemical super capacitor [35], photovoltaic cells [36], biosensors[37,38] and field effect transistors [39]. Graphene decorated withmetals oxides of copper [40], nickel [41] and metal nanoparticles

1 Actua

spopo

nastlnhdccco

2

2

(hpFgcifEm

2

wFtuXumsosVs

tebp

rmu

fdw0

98 K. Dhara et al. / Sensors and

uch as platinum and gold [42], and alloy nanoparticles like nickel-latinum [43] have been employed for the direct electrooxidationf glucose. So far there is no report on graphene decorated with cop-er oxide and platinum nanoparticles for the amperometric sensingf glucose.

This paper reports the one step synthesis of copper oxideanoflowers and platinum nanocubes decorated graphene and itspplication in the amperometric sensing of glucose. The sensorhowed very high sensitivity (3577 �A mM−1 cm−2) and the oxida-ion of glucose occurs at a low potential (+0.35 V). Graphene witharge surface area is an excellent platform for accommodating largeumber of catalytic sites. One step fabrication process coupled withigh sensitivity and selectivity together with affordable cost foreveloping markets makes this sensor highly suitable for point ofare glucose monitoring. Through the well known mechanism [44],opper oxide catalyzes glucose oxidation and platinum nanoparti-les act as a co-catalyst to enhance the electron transfer during thexidation of glucose.

. Materials and methods

.1. Chemicals and reagents

Graphite powder (<20 �m), �-d-(+)-glucose, l-ascorbic acidAA), dopamine (DA), acetaminophen and hexachloroplatinateexahydrate (H2PtCl6·6H2O) were purchased from Aldrich. Cop-er(II) nitrate trihydrate (Cu(NO3)2·3H2O) was purchased frominar chemicals (India). All other chemicals were of analyticalrade and used as received without further purification. Medi-al grade conductor paste (carbon ink-BQ242, Ag/AgCl-5874, silvernk-5064) and the thinner for the respective inks were purchasedrom Du Pont Company Pte Ltd., Singapore. De-ionised water fromlix-10 system (Millipore, Germany) was used for all the experi-ents.

.2. Characterization and instrumentation

Absorption spectra in the range of 200–800 nm was recordedith UV–vis spectrophotometer (Pharmaspec 1700, Shimadzu).

ourier transform infrared spectra (FTIR) were recorded in theransmission mode on a Thermo Nicolet, iS10 FTIR spectrometersing the KBr pellet and ATR mode for the range 400-4000 cm−1.-ray photoelectron spectroscopic (XPS) analysis was carried outsing KRATOS Axis Ultra (Kratos Analytical, United Kingdom). Aonochromated Al K� radiation (1486.6 eV) was focused on a

ample surface area of 0.7 mm × 0.3 mm. The survey scans werebtained by employing three passing sweeps of 15 eV and a stepize of 0.1 eV. Raman spectra were recorded on a LabRAM HR UV-IS-NIR Raman microscope from HORIBA Jobin-Yvon (633 nm laserource).

All electrochemical measurements were carried out using elec-rochemical analyzer CHI608D (CH Instruments, TX, USA). A threelectrode system consisting of an Ag/AgCl reference electrode, car-on working and counter electrodes was screen printed onto aolyethylene terephthalate (PET) substrate.

High resolution scanning electron micrograph (HRSEM) wasecorded using FEI quanta FEG 200-HRSEM. High resolution trans-ission electron microscopic (HRTEM) analysis was carried out

sing the JEOL JEM 2100 TEM.Small angle X-ray diffraction (SA-XRD) measurements were per-

ormed at room temperature (25 ◦C) on a Rigaku Miniflex II X-rayiffractometer, using K� foil filtered Cu K� radiation (� = 1.541 A)ith a scintillator detector (40 kV, 15 mA). The step time was 1 s at

.04◦/step in a 2� range of 5–90◦.

tors B 195 (2014) 197–205

2.3. Synthesis of Pt-CuO/rGO nanocomposite

Graphite oxide was prepared by following the Hummer’smethod [45] which involves the oxidation of graphite flakes usingKMnO4 in acidic medium. Graphene oxide (GO) was obtainedby exfoliating 100 mg of graphite oxide in 200 mL of distilledwater (0.5 mg/mL) by sonication. To this dispersion, 5.81 mL of 1 Mof Cu(NO3)2·3H2O and 0.275 mL of 0.0772 M H2PtCl6·6H2O wereadded which found to be optimum in terms of the sensing proper-ties of the composite. After 20 min stirring, 50 mL of 0.79 M sodiumborohydride was added drop wise to the reaction mixture andstirred at room temperature for several hours. The precipitate wasfiltered and washed with distilled water several times to removethe unreacted contents. The product obtained was dried at 70 ◦C.Finally the nanocomposite was annealed at 300 ◦C for 4 h to convertCu in the nanocomposite to CuO. The nanocomposite obtained wastermed as Pt-CuO/rGO.

2.4. Fabrication of sensor electrode (Pt-CuO/rGO/SPE)

The screen printed electrodes (SPE) consist of three electrodes(working, pseudo reference and auxiliary) were fabricated by tradi-tional screen printing technique. The SPE was activated by 20 cyclesin the potential range of −0.2 to +1.8 V at a scan rate of 0.1 V/s in 1 Msulphuric acid. 10 mg of Pt-CuO/rGO nanocomposite was dispersedin 2 mL of DMF using ultrasonication. 10 �L of the suspension wasdrop cast on an activated screen printed electrode (2 mm dia) anddried.

2.5. Electrochemical studies

The electrocatalytic behavior of the sensor electrode was stud-ied using linear sweep voltammetry (LSV) at a potential window of−0.2 to +1.0 V at various scan rates in 0.1 M NaOH with and with-out glucose. The steady state current response (amperometry) wasobtained at a constant potential of +0.6 V with reference to pseudoreference electrode Ag/AgCl. Electrochemical impedance spectro-scopic (EIS) analysis was carried out in 0.1 M NaOH solution at itsopen circuit potentials, in the frequency range of 0.01 Hz to 1 MHzwith potential amplitude of 5 mV.

3. Results and discussions

3.1. Characterization of Pt-CuO/rGO nanocomposite

Fig. 1A represents the UV-Visible spectrum of GO (curve a) andPt-CuO/rGO nanocomposite (curve b). The peak at 230 nm in GO isattributed to the pi-pi excitation and the peak seen at 295 nm isthe n-pi excitation of carboxylic moieties (C O) (curve a). The peakat 295 nm in GO disappeared in Pt-CuO/rGO due to the reductionof C O and the peak at 230 nm red shifted to 270 nm due to theextension of conjugation (curve b).

In the FTIR spectra (Fig. 1B), the peak at 1725 cm−1(curve a) dueto the stretching mode of carboxylic moieties (C O group) presentsin GO disappears in Pt-CuO/rGO (curve b) indicates the reduction ofGO to rGO. Further, the C C stretch of GO at 1628 cm−1 was shiftedto 1637 cm−1 in Pt-CuO/rGO. The intensity of the peak at 3407 cm−1

(O-H stretch) was decreased due to the reduction of GO to rGO. Thepeak observed at 3436 cm−1 in Pt-CuO/rGO nanocomposite is dueto the presence of CuO. Further, the characteristic peak of CuO at532 cm−1 is shifted to 542 cm−1 in the nanocomposite [46].

Fig. 1C depicts the Raman spectrum of GO (curve a) and Pt-

CuO/rGO (curve b). The predominant peaks obtained at 1344 and1591 cm−1 are the D and G bands respectively. The frequency of Dand G bands in GO and Pt-CuO/rGO are found to be same and theratio of intensity D/G is also same [GO (1.04), Pt-CuO/rGO (1.03 s)]

K. Dhara et al. / Sensors and Actuators B 195 (2014) 197–205 199

rum o

wosn2v

FtsoacocvppTadprtaip9fpP

o

Fig. 1. (A) UV-Visible spectrum of GO (a) and Pt-CuO/rGO (b), (B) FTIR spect

hich confirms that the method of synthesis does not affect the sizef the in-plane sp2 domains greatly [47]. From Raman spectroscopictudies it is evident that the electrical conductivity of graphene isot altered during the composite formation. The peaks obtained at68, 330 and 610 corresponding to the Ag and two Bg modes ofibrations of CuO respectively [48].

The XPS spectrum of the GO and Pt-CuO/rGO is show in Fig. 2.ig. 2A shows the wide scan (0–1000 eV) XPS spectrum of GO (blue),wo predominant peaks were obtained, one at 282–290 eV corre-ponding to carbon and the other at 525–540 eV corresponding toxygen. In the composite spectrum (purple), the peak appearingt 76 eV corresponds to Pt and the peak at 942.4 eV corresponds toopper 2P3/2, which is greater than that of standard value (932.7 eV)f metallic copper [40]. This shift towards high binding energy valueonfirms the formation of CuO after thermal treatment. The decon-oluted spectrum of C1 and O1 in GO contains sp2 hybridized C Ceak at 284.6 eV (red), C–OH peak at 286.2 eV (cyan), epoxy C–O–Ceak at 287.1 eV (green) and C O peak at 288.2 eV (blue) (Fig. 2B).he C–O–C peak at 532.1 eV (green), C O peak at 531.5 eV (cyan)nd C–OH peak at 533.5 eV (blue) were observed (Fig. 2C) [49]. Theeconvoluted spectrum of C1 in Pt-CuO/rGO has a predominanteak at 285 eV corresponding to C C, and all other peak intensityeduced compared to C1 spectrum of GO; confirm the reduction ofhe GO (Fig. 2D). The deconvoluted peak of O1 contains three peakst 529.9 eV (green), 531.7 eV (cyan) and 533.6 eV (blue) correspond-ng to Cu–O, C O and C–O respectively (Fig. 2E). The deconvolutedeak of copper 2P3/2 contains two peaks at 934.06 eV (red) and36.62 eV (green) that correspond to CuO and presence of oxygenrom air (Fig. 2F and G). The deconvoluted peak of Pt contains two

eaks at 73.8 eV (red) and 76.9 eV (green) correspond to Pt7/2 andt5/2 respectively (Fig. 2H and I).

Fig. S1 shows the powder X-ray diffractogram featuresf graphite, GO and Pt-CuO/rGO. The characteristic peak of

f GO (a), Pt-CuO/rGO (b) and (C) Raman spectrum of GO (a), Pt-CuO/rGO (b).

hexagonal graphite corresponding to a d-spacing of 0.33 nm wasfound at 26.55◦ in graphite (Fig. S1A). Upon conversion of thegraphite into GO, the peak shifted to 9.83◦ and the interlayerspacing increased to 0.89 nm (Fig. S1B). This increase in d-spacingis due to the intercalation of carboxyl, epoxy and hydroxyl func-tional groups in between the graphene layers. In Pt-CuO/rGO, themajor peaks at 32.28 (−110), 35.48 (−111), 38.96 (200), 48.73(−202), 61.46 (−113), 65.8 (022), 75.3(−222) and 89.7 (−131)confirms the presence of CuO in monoclinic system (JCPDS 45-0937). Further, peaks at 44.6 (200), 78.5 (311), 82.9 (222) inferthat Pt is in cubic form (JCPDS 87-0644) (Fig. S1C). XRD fromPt/rGO, CuO/rGO nanocomposites helped better understating ofPt-CuO/rGO nanocomposite result (Fig. S1D).

3.2. Morphological characterization

Fig. 3 shows the HRSEM images of GO and Pt-CuO/rGO. Fig. 3Areveals the morphology of the GO as neatly laid out layer with wrin-kles. Fig. 3B represents that the graphene layer is decorated withCuO nanoflower and platinum nanocubes in a well dispersed man-ner. It has already been reported that the reduction of copper(II)salts with sodium borohydride results in copper oxide nanoflowers[50].

Fig. 4 Further the morphology of GO and Pt-CuO/rGO was con-firmed by HRTEM analysis. Fig. 4A shows the morphology of GO,which is in good agreement with HRSEM results. Fig. 4B–D repre-sents the HRTEM images Pt-CuO/rGO at different magnifications.Selected area electron diffraction (SAED) patterns of GO and Pt-CuO/rGO are shown in Fig. 4E and F. Presence of crystalline planes

in the Pt-CuO/rGO was confirmed with the diffraction patterns.From the HRTEM images and SAED patterns d-spacing values wereobtained. The d-spacing value of 0.27 nm is consistent with the(−110) lattice spacing of end-centered monoclinic flower form

200 K. Dhara et al. / Sensors and Actuators B 195 (2014) 197–205

Fig. 2. XPS analysis: (A) GO (blue) and Pt-CuO/rGO (purple); (B) the GO deconvoluted spectrum of C1 [C C (red), C-OH (cyan), C-O-C (green), C O (blue)]; (C) the deconvolutedspectrum of O1 [C-O-C (green), C O (cyan) and C-OH (blue)]; (D) Pt-CuO/rGO deconvoluted spectrum C1 [C C (red)]; (E) deconvoluted spectrum of O1 [Cu-O (green), C O(cyan) and C-O (blue)]; (F) the deconvoluted Cu 2P3/2 peak; (G) deconvoluted spectra of Cu 2P3/2 [CuO (red), oxygen from air (green)]; (H) convoluted Pt peak; I. deconvolutedpeak of Pt [Pt7/2 (red), Pt5/2 (green)]. (For interpretation of the color information in this figure legend, the reader is referred to the web version of the article.)

Fig. 3. HRSEM images of (A) GO, (B) Pt-CuO/rGO.

K. Dhara et al. / Sensors and Actuators B 195 (2014) 197–205 201

gnific

occg

3

3

a0

Fa0

Fig. 4. HRTEM images of GO (A), Pt-CuO/rGO at different ma

f copper oxide and 0.12 nm, is the (200) lattice spacing of face-entered cubic form of Pt. These d-spacing value and calculatedrystal planes are compared with XRD data and are found to be inood agreement.

.3. Electrochemical studies

.3.1. Voltammetric analysisThe linear sweep voltammograms obtained on the bare SPE,

ctivated SPE, Pt/rGO/SPE, CuO/rGO/SPE and Pt-CuO/rGO/SPE in.1 M NaOH containing 3 mM glucose are shown in Fig. 5. There is

ig. 5. LSV of the bare SPE (a), activated SPE (b), Pt/rGO/SPE (c), CuO/rGO/SPE (d)nd Pt-CuO/rGO/SPE (e) in 0.1 M NaOH containing 3 mM glucose at a scan rate of.1 V/s.

ations (B, C, D), SAED pattern of GO (E) and Pt-CuO/rGO (F).

no well defined anodic peak corresponding to the oxidation of glu-cose was observed on the bare SPE (curve a), activated SPE (curve b)and Pt/rGO/SPE (curve c). A well defined anodic peak was observedat +0.35 V on CuO/rGO/SPE (curve d) and Pt-CuO/rGO/SPE (curvee). But the peak current observed on Pt-CuO/rGO/SPE is 267 �Ahigher than that observed on CuO/rGO/SPE. From this observationit is obvious that though CuO is responsible for electrocatalytic oxi-dation of glucose, Pt is necessary to increase the sensitivity. Further,the oxidation potential of glucose on the Pt-CuO/rGO/SPE is lowerwhen compared to published reports [40,44].

Fig. 6A shows the linear sweep voltammograms on the Pt-CuO/rGO/SPE in 0.1 M NaOH containing 3 mM glucose solution atdifferent scan rates. The anodic peak current increases linearly withincrease in scan rate from 0.01 to 0.15 V/s with the regression equa-tion IP (�A) = 142.0 + (5.92)� (mV/s) (r = 0.9995) with a standarddeviation of 7.96 (n = 15). This confirms that the oxidation of glucoseon the Pt-CuO/rGO/SPE is a surface confined process. Similar resultshave already been reported based on graphene modified electrodes[40]. Thus oxidation of glucose on the modified electrode obeysadsorptive electroactive Nernstian condition for adsorption con-trolled reactions, iP = (n2F2�A� *

O/4RT) where iP is the peak current,n is the number of electrons, F is Faraday constant (in coulomb), Ris gas constant (J mol−1 K−1), A is the area (in cm2), � is the scanrate (in V/s) [51].

Linear sweep voltammograms of Pt-CuO/rGO/SPE in 0.1 M NaOHsolution with different concentrations of glucose at a scan rate0.1 V/s is shown in Fig. 6B. Curves ‘a’ without glucose and curveb to m correspond to 1 mM each of glucose for 12 additions. It

is seen that as the concentration of glucose increases the anodicpeak current also increases linearly with the regression equationIP (�A) = 44.96 + (81.89)C (mM) with a relative standard deviation(�) of 1.73, n = 13 and correlation coefficient r = 0.999 throughout

202 K. Dhara et al. / Sensors and Actuators B 195 (2014) 197–205

F iffere( scan r

to

3

Ps(aCRWascterPsts

3

aIi

Fpa

ig. 6. LSVs on the Pt-CuO/rGO/SPE in 0.1 M NaOH, containing 3 mM glucose with dA) and with increase in concentrations of glucose from 1 mM to 12 mM (b-m) at a

he concentration range. A slight shift in anodic peak potential wasbserved at higher concentrations.

.3.2. Electrochemical impedance studiesThe electron transfer behavior of the bare SPE, activated SPE,

t/rGO/SPE, CuO/rGO/SPE and Pt-CuO/rGO/SPE electrodes weretudied by electrochemical impedance spectroscopy in 0.1 M NaOHFig. S2). The Nyquist plots obtained are fitted with an equiv-lent circuit (Fig. S2 inset ‘a’), enlarged images of Pt/rGO/SPE,uO/rGO/SPE and Pt-CuO/rGO/SPE (Fig. S2 inset ‘b’). Where, R1 and2 are respectively the bulk and charge transfer resistance, W thearburg impedance. In order to understand the role of graphene

nd Pt nanoparticles EIS was carried out after every step of sen-or fabrication. The R2 value of the activated SPE decreased afteroating Pt/rGO on SPE. This can be attributed to the enhanced elec-ron transfer behavior of the Pt/rGO/SPE compared to the activatedlectrode. CuO/rGO/SPE electrode showed very high charge transferesistance (114 �) compared to Pt/rGO/SPE (62 �). The presence oft in the composite (Pt-CuO/rGO/SPE) reduced the R2 value (71 �)ignificantly than CuO/rGO/SPE. This study clearly demonstrateshe significance of Pt nanoparticles in the fabrication of proposedensor.

.3.3. Amperometry

Different potentials (+0.45 V, +0.50 V, +0.55 V, +0.60 V) were

pplied and tested independently for glucose oxidation (Fig. 7A).t is obvious from the figure that the response obtained at +0.6 Vs higher compared to other potentials in terms of linearity and

ig. 7. (A) Steady state current response of Pt-CuO/rGO/SPE to 0.2 mM glucose additions

otentials). (B) Plot of response current vs concentration obtained on Pt-CuO/rGO/SPE

dditions).

nt scan rate from 0.01 V/s to 0.15 V/s (a-o) (inset: plot of peak current vs. scan rate)ate of 0.1 V/s (inset: plot of peak current vs. concentration) (B).

sensitivity. Hence all further amperometric experiments were car-ried out at +0.6 V. Fast (less than 3 s) and stable response wasobtained on adding glucose. Response current was linear with con-centration (Inset in Fig. 7A). The linear regression equation forglucose oxidation at +0.6 V is given by Ip(�A) = 7.465 + (112.3)C(mM) with � = 4.825, n = 16 and correlation coefficient of r = 0.9991.The sensitivity was calculated from the slope of the calibrationplot and was found to be 3577 �A mM−1 cm−2. The sensor showeda lower detection limit (S/N = 3) of 0.01 �M. Chronoamperometrywas conducted at +0.6 V potential by injecting 2 mM glucose solu-tion each and found that the sensor is linear upto 12 mM of glucosewith a linear regression equation is given by Ip = 6.65 + (18.5)C (mM)with � = 4.93, n = 7 and correlation coefficient of r = 0.9984 (Fig. 7B).

3.4. Effect of interfering species

Biomolecules like AA, DA and UA commonly exist with glu-cose and also get easily oxidized at the applied potential on thesensor electrode. Hence these biomolecules (at their physiologi-cal concentrations) were also tested amperometrically at +0.6 Valong with glucose on the Pt-CuO/rGO/SPE (Fig. 8). It is obviousfrom the Fig. 8 that the response obtained for 1.25 �M UA, 125 �Macetaminophen, 125 �M AA and 125 �M DA were less than 1% ofthat obtained for glucose.

Table 1 shows the comparison of the present work withthat already reported on the glucose sensors based on graphenenanocomposites. The sensor electrode chosen for comparisonwas constructed using CuO/graphene [40], PtNi/graphene [43],

in 0.1 M NaOH under various potentials (inset: calibration plot at different appliedat +0.6 V in 0.1 M NaOH solution (inset: amperometric response to 2 mM glucose

K. Dhara et al. / Sensors and Actuators B 195 (2014) 197–205 203

Table 1Performance of Pt-CuO/rGO/SPE glucose sensor compared with other reported non-enzymatic glucose sensors.

Electrode Potential (V) Medium Sensitivity (�A mM−1 cm−2) Linear range of detection Lower limit of detection Reference

CuO/graphene +0.6 0.1 M NaOH 1065 1 �M–8 mM 1 �M [40]PtNi/graphene −0.35 PBS, pH 7.4 20.42 35 mM 0.01 mM [43]GNS/NiO/DNA +0.6 0.1 M NaOH 9.0

14.31 �M–8 mM 1 �M–200 �M 2.5 �M [41]

Graphene-CuO +0.4 0.1 M NaOH 1490 2 �M–0.06 mM 0.29 �M [48]Graphene/CuO nanocubes +0.55 0.1 M NaOH 1360 2 �M–4 mM 0.7 �M [44]SWCNT/Cu/Nafion +0.65 0.05 M NaOH – 0.5 mM 0.25 �M [52]GO/Pt nanoflowers +0.47 PBS, pH 7.4 1.26

0.64Pt-CuO/rGO/SPE +0.6 0.1 M NaOH 3577

F2A

G[dtg

3

bcFog

F1

ig. 8. Amperometric response of Pt-CuO/rGO/SPE in 0.1 M NaOH at +0.6 V with mM glucose additions in presence of 1.25 �M UA, 125 �M acetaminophen, 125 �MA and 125 �M DA.

NS/NiO/DNA [41], Graphene-CuO [48], Graphene/CuO nanocubes44], SWCNT/Cu/Nafion [52], GO/Pt nanoflowers [53]. The sensoreveloped in this work has wide linearity, better limit of detec-ion and higher sensitivity compared to that reported already onraphene based non-enzymatic glucose sensors.

.5. Real sample analysis

The practical application of Pt-CuO/rGO/SPE was demonstratedy testing the glucose in blood serum samples. Blood samples were

entrifuged for 20 min at 12,000 rpm to separate out the serum.ig. 9 (A and B for patient 1 and 2 respectively) shows the amper-metric response curves obtained by injecting 10 �L of 50 mMlucose (a, b, e and f) and 100 �L blood serum (c and d) into 4 mL of

ig. 9. Steady state current response of Pt-CuO/rGO/SPE to glucose and serum samples (A25 �M glucose solutions and c, d are blood serum.

2 �M–10.3 mM10.3–20.3 mM 2 �M [53]

0.5 �M–12 mM 0.01 �M Current work

constantly stirred 0.1 M NaOH at a constant potential of +0.6 V. Theconcentration of glucose in blood serum was calculated by compar-ing the current response obtained for glucose solution and bloodserum. The concentration of glucose in blood is 5.97 and 8.72 mMrespectively for patient 1 and 2. These obtained results were com-pared with commercially available glucometer and test strips (OneTouch Ultra, Johnson and Johnson) and found that they are well inagreement (less than 2% variation).

3.6. Reproducibility and stability

Hundreds of sensor strips were fabricated (as in Section 2.4)and tested in 0.1 M NaOH (as described in Section 2.5). Responseobtained by the amperometric and voltammetric analysis showedthat the variation in current response was less than 5%. The longterm on shelf stability of the sensor was tested by preserving thesensors in air tight container for 3 months. It was found that thedecrease in current response was less than 2%. In order to studythe effect of humidity and other environmental parameters duringtesting, the electrode was exposed to air for three days and testedunder similar conditions. The current response was decreased byabout 4%. These studies corroborate the excellent reproducibilityof the sensor fabrication procedure and on shelf storage stability.

4. Conclusion

Low cost screen printed electrodes were fabricated and used fornon-enzymatic glucose sensing applications. The electrocatalyst,Pt-CuO/rGO nanocomposite was synthesized by a simple one step

chemical reduction. Pt exists as nanocubes and CuO as nanoflowersin the composite. Glucose oxidation on the modified electrodeoccurs at +0.35 V and it was a surface confined process. Thesensor exhibited excellent sensitivity of 3577 �A mM−1 cm−2 and

and B correspond to patient 1 and 2) tested in 0.1 M NaOH at 0.6 V. a, b, e, f are for

2 Actua

s1ctt

A

o(2Xit

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

04 K. Dhara et al. / Sensors and

electivity towards glucose. Sensor showed linear response upto2 mM glucose with lower detection limit of 0.01 �M. Glucoseoncentration in blood serum was successfully determined withhe sensor. The easy fabrication and no enzyme limitation pavehe way for commercial application.

cknowledgement

Keerthy Dhara thanks to Department of Science and Technol-gy (DST) for the financial support under DST-INSPIRE Fellowshipsanction order No. DST/INSPIRE Fellowship/2011/[4] & September6th, 2011). Service rendered by Amrita centre for Nanosciences forPS analysis, PSG institute of advanced studies for HRTEM analysis

s greatly acknowledged. The author expresses her sincere thankso S. Ramakrishnan (PhD scholar) for his timely help and support.

eferences

[1] J.D. Newman, A.P.F. Turner, Home blood glucose biosensors: a commercial per-spective, Biosens. Bioelectron. 20 (2005) 2435–2453.

[2] G.S. Wilson, R. Gifford, Biosensors for real-time in vivo measurements, Biosens.Bioelectron. 20 (2005) 2388–2403.

[3] J. Wang, Electroanalysis 17 (2005) 7–14.[4] L.J. Clark Jr., Electrode systems for continuous monitoring in cardiovascular

surgery, Ann. N.Y. Acad. Sci. 102 (1962) 29.[5] J. Wang, Electrochemical glucose biosensors, Chem. Rev. 108 (2008) 814–825.[6] H. Tang, J. Chen, S. Yao, L. Nie, G. Deng, Y. Kuang, Amperometric glucose biosen-

sor based on adsorption of glucose oxidase at platinum nanoparticle-modifiedcarbon nanotube electrode, Anal. Biochem. 331 (2004) 89–97.

[7] S.H. Lim, J. Wei, J. Lin, Q. Li, J. KuaYou, A glucose biosensor based on electrodepo-sition of palladium nanoparticles and glucose oxidase onto Nafion-solubilizedcarbon nanotube electrode, Biosens. Bioelectron. 20 (2005) 2341–2346.

[8] G.L. Luque, M.C. Rodríguez, G.A. Rivas, Glucose biosensors based on the immo-bilization of copper oxide and glucose oxidase within a carbon paste matrix,Talanta 66 (2005) 467–471.

[9] Y. Zhu, H. Zhu, X. Yang, L. Xu, C. Li, Sensitive biosensors based on (dendrimerencapsulated pt nanoparticles)/enzyme multilayers, Electroanalysis 19 (2007)698–703.

10] R. Wilson, A.P.F. Turner, Glucose oxidase: an ideal enzyme, Biosens. Bioelectron.7 (1992) 165–185.

11] K.E. Toghill, R.G. Compton, Electrochemical non-enzymatic glucose sensors: aperspective and an evaluation, Int. J. Electrochem. Sci. 5 (2010) 1246–1301.

12] S. Park, T.D. Chung, H.C. Kim, Nonenzymatic glucose detection using meso-porous platinum, Anal. Chem. 75 (2003) 3046–3049.

13] S. Cherevko, C.-H. Chung, Gold nanowire array electrode for non-enzymaticvoltammetric and amperometric glucose detection, Sens. Actuators B: Chem.142 (2009) 216–223.

14] H. Bai, M. Han, Y. Du, J. Bao, Z. Dai, Facile synthesis of porous tubular palladiumnanostructures and their application in a nonenzymatic glucose sensor, Chem.Commun. 46 (2010) 1739–1741.

15] Y. Sun, H. Buck, T.E. Mallouk, Combinatorial discovery of alloy electrocatalystsfor amperometric glucose sensors, Anal. Chem. 73 (2001) 1599–1604.

16] S. Park, H. Boo, T.D. Chung, Electrochemical non-enzymatic glucose sensors,Anal. Chim. Acta 556 (2006) 46–57.

17] J. Wang, D.F. Thomas, A. Chen, Nonenzymatic electrochemical glucose sensorbased on nanoporous PtPb networks, Anal. Chem. 80 (2008) 997–1004.

18] Y.B. Vassilyev, O.A. Khazova, N.N. Nikolaeva, Kinetics and mechanism of glu-cose electrooxidation on different electrode-catalysts: Part I. Adsorption andoxidation on platinum, J. Electroanal. Chem. Interfacial Electrochem. 196 (1985)105–125.

19] Y. Zhao, J. Zhao, D. Ma, Y. Li, X. Hao, L. Li, et al. Synthesis, growth mechanismof different Cu nanostructures and their application for non-enzymatic glucosesensing, Colloids Surf. A: Physicochem. Eng. Aspects, 409 (2012) 105–111.

20] N.K. Singh, B. Jain, S. Annapoorni, ZnO modified gold disc: A new route toefficient glucose sensing, Sens. Actuators B: Chem. 156 (2011) 383–387.

21] T.G.S. Babu, T. Ramachandran, Development of highly sensitive non-enzymaticsensor for the selective determination of glucose and fabrication of a workingmodel, Electrochim. Acta 55 (2010) 1612–1618.

22] J. Zhu, D. Li, H. Chen, X. Yang, L. Lu, X. Wang, Highly dispersed CuO nanopar-ticles prepared by a novel quick-precipitation method, Mater. Lett. 58 (2004)3324–3327.

23] E. Reitz, W. Jia, M. Gentile, Y. Wang, Y. Lei, CuO nanospheres based nonenzy-matic glucose sensor, Electroanalysis 20 (2008) 2482–2486.

24] G. Wang, Y. Wei, W. Zhang, X. Zhang, B. Fang, L. Wang, Enzyme-free ampero-metric sensing of glucose using Cu-CuO nanowire composites, Microchim. Acta

168 (2010) 87–92.

25] C. Batchelor-McAuley, Y. Du, G.G. Wildgoose, R.G. Compton, The use of cop-per (II) oxide nanorod bundles for the non-enzymatic voltammetric sensingof carbohydrates and hydrogen peroxide, Sens. Actuators B: Chem. 135 (2008)230–235.

tors B 195 (2014) 197–205

26] F. Wang, A. Kalam, L. Chang, D. Xie, A.S. Al-Shihri, G. Du, Rapid microwave-assisted synthesis of ball-in-ball CuO microspheres and its application as a H2O2

sensor, Mater. Lett. 92 (2012) 96–99.27] Y.-Q. Dai, K.-K. Shiu, Glucose biosensor based on multi-walled carbon nanotube

modified glassy carbon electrode, Electroanalysis 16 (2004) 1697–1703.28] L.-Q. Rong, C. Yang, Q.-Y. Qian, X.-H. Xia, Study of the nonenzymatic glucose sen-

sor based on highly dispersed Pt nanoparticles supported on carbon nanotubes,Talanta 72 (2007) 819–824.

29] X. Kang, Z. Mai, X. Zou, P. Cai, J. Mo, Glucose biosensors based on platinumnanoparticles-deposited carbon nanotubes in sol–gel chitosan/silica hybrid,Talanta 74 (2008) 879–886.

30] L. Meng, J. Jin, G. Yang, T. Lu, H. Zhang, C. Cai, Nonenzymatic electrochemi-cal detection of glucose based on palladium- single-walled carbon nanotubehybrid nanostructures, Anal. Chem. 81 (2009) 7271–7280.

31] C.B. Jacobs, M.J. Peairs, B.J. Venton, Review carbon nanotube based electrochem-ical sensors for biomolecules, Anal. Chim. Acta 662 (2010) 105–127.

32] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, et al.,Electric field effect in atomically thin carbon films, Science 306 (2004) 666–669.

33] A.K. Geim, K.S. Novoselov, The rise of graphene, Nature Mater. 6 (2007)183–191.

34] R.F. Service, Carbon, Sheets an atom thick give rise to graphene dreams, Science324 (2009) 875–877.

35] Z. Xu, Z. Li, C.M.B. Holt, X. Tan, H. Wang, B.S. Amirkhiz, et al., Electrochemicalsupercapacitor electrodes from sponge-like graphene nanoarchitectures withultrahigh power density, J Phys. Chem. Lett. 3 (2012) 2928–2933.

36] Z. Liu, Q. Liu, Y. Huang, Y. Ma, S. Yin, X. Zhang, et al., Organic photovoltaicdevices based on a novel acceptor material: graphene, Adv. Mater. 20 (2008)3924–3930.

37] M. Zhou, Y. Zhai, S. Dong, Electrochemical sensing and biosensing plat-form based on chemically reduced graphene oxide, Anal. Chem. 81 (2009)5603–5613.

38] C.-H. Lu, H.-H. Yang, C.-L. Zhu, X. Chen, G.-N. Chen, A graphene platform forsensing biomolecules, Angew. Chem. Int. Ed. 48 (2009) 4785–4787.

39] S. Gilje, S. Han, M. Wang, K.L. Wang, R.B. Kaner, A chemical route to graphenefor device applications, Nano Lett. 7 (2007) 3394–3398.

40] Y.-W. Hsu, T.-K. Hsu, C.-L. Sun, Y.-T. Nien, N.-W. Pu, M.-D. Ger, Synthesisof CuO/graphene nanocomposites for nonenzymatic electrochemical glucosebiosensor applications, Electrochim. Acta 82 (2012) 152–157.

41] W. Lv, F.-M. Jin, Q. Guo, Q.-H. Yang, F. Kang, DNA-dispersed graphene/NiOhybrid materials for highly sensitive non-enzymatic glucose sensor, Elec-trochim. Acta 73 (2012) 129–135.

42] C. Shan, H. Yang, D. Han, Q. Zhang, A. Ivaska, L. Niu, Graphene/AuNPs/chitosannanocomposites film for glucose biosensing, Biosens. Bioelectron. 25 (2010)1070–1074.

43] H. Gao, F. Xiao, C.B. Ching, H. Duan, One-step electrochemical synthesis ofPtNi nanoparticle-graphene nanocomposites for nonenzymatic amperometricglucose detection, ACS Appl. Mater. Interfaces 3 (2012) 3049–3057.

44] L. Luo, L. Zhu, Z. Wang, Nonenzymatic amperometric determination of glucoseby CuO nanocubes graphene nanocomposite modified electrode, Bioelectro-chemistry 88 (2012) 156–163.

45] W.S. Hummers, R.E. Offeman, Preparation of graphitic oxide, J. Am. Chem. Soc.80 (1958), 1339-1339.

46] M.R. Johan, M.S.M. Suan, N.L. Hawari, H.A. Ching, Annealing effects on theproperties of copper oxide thin films prepared by chemical deposition, Int. J.Electrochem. Sci. 6 (2011) 6094–6104.

47] J.-J. Shi, J.-J. Zhu, Sonoelectrochemical fabrication of Pd-graphene nanocom-posite and its application in the determination of chlorophenols, Electrochim.Acta 56 (2011) 6008–6013.

48] Y. Li, F. Huang, J. Chen, T. Mo, S. Li, F. Wang, et al., A high performanceenzyme-free glucose sensor based on the graphene-CuO nanocomposites, Int.J. Electrochem. Sci. 8 (2013) 6332–6342.

49] H.-K. Jeong, H.-J. Noh, J.-Y. Kim, M. Jin, C. Park, Y. Lee, X-ray absorption spec-troscopy of graphite oxide, EPL (Europhys. Lett.) 82 (2008) 67004.

50] D. Volanti, D. Keyson, L. Cavalcante, A. Simoes, M. Joya, E. Longo, et al., Synthesisand characterization of CuO flower-nanostructure processing by a domestichydrothermal microwave, J. Alloys Comp. 459 (2008) 537–542.

51] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and Applica-tions. Wiley, New York, 1980.

52] K.B. Male, S. Hrapovic, Y. Liu, D. Wang, J.H. Luong, Electrochemical detection ofcarbohydrates using copper nanoparticles and carbon nanotubes, Anal. Chim.Acta 516 (2004) 35–41.

53] G.-h. Wu, X.-h. Song, Y.-f. Wu, X.-m. Chen, F. Luo, X. Chen, Non-enzymaticelectrochemical glucose sensor based on platinum nanoflowers ssupported ongraphene oxide, Talanta 105 (2013) 379–385.

Biographies

Ms. Keerthy Dhara a PhD student at Amrita University completed her B.Sc. from

Kakatiya University in 2006, M.Sc. in organic chemistry from Periyar Universityin 2008. Her research interests include synthesis of metal nanoparticles decoratedgraphene nanocomposites for Biosensing application.

Mr. John Stanley completed his bachelor’s and master’s in biotechnology fromBharathiar University in 2005 and 2007 respectively. He obtained an M.Tech. in

Actua

Bhed

Dv1UB

Ds

2001. He received his master’s from Gandhigram Rural Institute in 2003. He obtainedhis PhD from Amrita University in 2010. Currently he is serving as an Assistant

K. Dhara et al. / Sensors and

iomedical Engineering from Amrita University in 2009. He is currently pursuingis Ph.D. at Amrita University. His research focus includes development of non-nzymatic biosensors and engineering Lab on a chip device for multiple analyteetection.

r. Ramachandran T is currently Professor, Department of Sciences Amrita Uni-ersity. He completed his B.Sc., M.Sc. and Ph.D. degrees from Madras University in962, 1964 and 1988. He has post doctoral research experience from Georgetown

niversity (Washington, USA). Industrial electrochemical processes, Fuel cells andiosensors are his research interests.

r. Bipin G Nair, Professor and Dean, School of Biotechnology, Amrita Univer-ity graduated from Gujarat University in 1979. He obtained his master’s and

tors B 195 (2014) 197–205 205

doctoral degrees from Maharaja Sayajirao University in 1981 and 1986 respectively.He received his Post doctoral research training from University of Tennessee, Mem-phis. His research interests include mechanism underlying impaired wound healingand development of low cost biomedical devices.

Dr. Satheesh Babu T.G completed his bachelor’s from the University of Calicut in

Professor in the department of Sciences at Amrita University. His research focus isdirected towards synthesis of nanomaterials and their application in the develop-ment of non-enzymatic biosensors and Lab on a chip device.