SnO2 Quantum Dots-Reduced Graphene Oxide Composite forEnzyme-Free Ultrasensitive Electrochemical Detection of UreaDipa Dutta,† Sudeshna Chandra,† Akshaya K. Swain,‡ and Dhirendra Bahadur*,†

†Department of Metallurgical Engineering & Materials Science, Indian Institute of Technology Bombay, Mumbai, Maharashtra400076, India‡IITB Monash Research Academy, Department of Metallurgical Engineering & Materials Science, Indian Institute of TechnologyBombay, Mumbai, Maharashtra 400076, India

*S Supporting Information

ABSTRACT: Most of the urea sensors are biosensors andutilize urease, which limit their use in harsh environments.Recently, because of their exceptional ability to endorse fasterelectron transfer, carbonaceous material composites andquantum dots are being used for fabrication of a sensitivetransducer surface for urea biosensors. We demonstrate anenzyme free ultrasensitive urea sensor fabricated using a SnO2quantum dots (QDs)/reduced graphene oxide (RGO)composite. Due to the synergistic effect of the constituents,the SnO2 QDs/RGO (SRGO) composite proved to be anexcellent probe for electrochemical sensing. The morphologyand structure of the composite was characterized by varioustechniques, and it was observed that SnO2 QDs are decoratedon RGO layers. Electrochemical studies were performed to evaluate the characteristics of the sensor toward detection of urea.Amperometry studies show that the SRGO/GCE electrode is sensitive to urea in the concentration range of 1.6 × 10−14−3.9 ×10−12 M, with a detection limit of as low as 11.7 fM. However, this is an indirect measurement for urea wherein the analyticalsignal is recorded as a decrease in the amperommetric and/or voltammetric current from the solution redox species ferrocyanide.The porous structure of the SRGO matrix offers a very low transport barrier and thus promotes rapid diffusion of the ionicspecies from the solution to the electrode, leading to a rapid response time (∼5 s) and ultrahigh sensitivity (1.38 μA/fM). Goodanalytical performance in the presence of interfering agents, low cost, and easy synthesis methodology suggest that SRGO can bequite promising as an electroactive material for effective urea sensing.

Composite materials containing two or more constituentsare of great interest as they can jointly exhibit unique

physical and chemical properties for varied applications.Quantum dots are zero dimensional materials having size-dependent properties which make them quite interesting.Among the wide band gap semiconductors, tin dioxidequantum dots (SnO2 QDs) have attracted enormous researchinterest for their widespread applications in electronics andoptics due to their excellent electrical and electrochemicalproperties.1,2 On the other hand, graphene has also attractedgreat attention due to its outstanding electrical, thermal, andoptical properties and a theoretical high surface area.3,4 Thenumber of graphene layers in reduced graphene oxide (RGO)significantly influences the properties of RGO. However, thesegraphene layers tend to aggregate in solution due to lack ofoxygen-containing groups which can be overcome byincorporating different metal oxide nanoparticles between thelayers.4

Metal oxide/graphene composites are expected to exhibitexcellent material properties of parent components, due to thesynergistic effect of both. Uniform loading of metal oxide

nanocrystals on the graphene surface is quite challenging due tothe uneven distribution of functional groups on its surface.4,5

Presently, tin oxides/graphene composites have aroused theinterest of researchers, since they are efficient composites forlithium ion batteries,4,6 supercapacitors,7 dye degradation,8 andbiosensing.9 The SnO2/graphene composites can be used toconstruct electrochemical sensors due to high active surfacearea, high electrocatalytic activity, chemical stability, andinterface dominated properties. In line with this, one maythink of using the composites for sensing small molecules likeurea which is an end product of protein degradation andnitrogen metabolism. Urea is also a critical indicator of liver andkidney malfunction in the human body.10 The normalconcentration of urea in blood serum is 3−7 mM (15−40mg/dL).11 The presence of urea in the environment is also amatter of great concern for ecosystem conservation and humanhealth protection.12 Due to its relatively higher water solubility,

Received: February 25, 2014Accepted: May 15, 2014

Article

pubs.acs.org/ac

© XXXX American Chemical Society A dx.doi.org/10.1021/ac5007365 | Anal. Chem. XXXX, XXX, XXX−XXX

a considerable amount of the applied urea herbicides is washedout in the aquatic environment. It can pollute the surface andthe groundwater into which it drains.13 It is therefore, essentialto analyze urea in the environment, drinking water, and food.12

Electrochemical analysis has inherent advantages of simplicity,high sensitivity, and relatively low cost over other sophisticatedmethods like high performance liquid chromatography,chemiluminescence, fluorimetry, and the like.Most of the urea sensors are biosensors, which utilize urease

(Urs) as a sensing element. However, their complicatedimmobilization procedures, activity, stability, high cost ofenzymes, and critical operating condition have somelimitations, and they are not suitable in harsh environments.14

Recently, metal oxide matrices such as NiO nanoparticles,10

zinc oxide nanostructures,15 metal oxide−chitosan composite,16

carbonaceous materials composites,17 and quantum dots18 havebeen used for fabrication of a sensitive transducer surface forurea biosensors because of their exceptional ability to endorsefaster electron transfer between the electrolyte and theelectrode. So far, no attempt has been made to fabricateenzymeless chemical sensors for lower level urea detection.SnO2 QDs (SQDs) and RGO composite based chemicalsensors can be used as potential probes for detection of urea inenvironmental samples.Herein, we report a new strategy for fabrication of SQDs/

RGO (SRGO) from a hybrid complex [(N2H4)m(SnCl4)n].Different morphological characterization techniques were usedto confirm the formation of the SRGO composite. Electro-chemical studies were performed to investigate its performanceas a proposed enzymeless chemical sensor for urea. It showed adetection limit which is much lower than the previouslyreported sensors that are based on different matrixes such asmetal oxides, carbon materials, and quantum dots.16−18 SRGOhas synergistic effects of the QDs and the two matrixes, namely,SnO2 and RGO, which results in highly sensitive sensormaterial.

■ EXPERIMENTAL SECTIONMaterials Preparation. A hybrid SRGO composite was

fabricated in two simple steps. In the first step, graphite oxide(GO) was prepared from natural graphite power, and in thesecond step, it was reduced to form reduced graphene oxide.Natural graphite powder (trace metals <100.0 ppm, purity>99.99%, average particle size <45 μm), SnCl4·5H2O (98%),and hydrazine hydrate (98%) from Sigma-Aldrich and H2SO4(98%), HCl, KMnO4, H3PO4, and H2O2 from Thomas Bakerwere used for this synthesis. All reagents were used withoutfurther purification. Natural graphite powder was oxidized by amodified Hummers method.19,20

40 mg of GO was added to 100 mL of Milli-Q water andsonicated for 30 min to form a brown homogeneoussuspension. 1 mL of N2H4·H2O (98%) and 10 mL ofNH4OH (25%) were added to this suspension and maintainedat 100 °C. In an alkaline medium, N2H4·H2O serves as a strongreductant which reduces GO to RGO. For the synthesis ofSRGO composites, a hybrid complex (N2H4)m(SnCl4)n wasadded to the mixture after 90 min and maintained at 100 °C foranother 18 h. The preparation of this hybrid complex isreported in our previous work.21 The product, thus obtained,was centrifuged at 10 000 rpm (9838g) for 10 min and washedwith ethanol followed by deionized water; this washing isrepeated six times and then dried overnight at 60 °C in air. Theproposed mechanism of the formation of the SRGO composite

can be explained with the fact that the surface of the GO iscovered with oxygen-containing groups, such as hydroxyl andepoxy at basal plane and carbonyl and carboxyl acid at theedges.22 Along with the reduction of GO, the hybrid complex((N2H4)m(SnCl4)n) simultaneously decomposes to Sn4+,23

which gets attached to the surface of RGO through theoxygen-containing groups. The attached Sn4+ then getconverted to SnO2 in an aqueous medium.24

The bare RGO was synthesized by the same methoddescribed above without addition of the hybrid complex[(N2H4)m(SnCl4)n]. The methodology for the preparation ofbare SQDs has been described elsewhere.21

Materials Characterization. The crystallinity and phasepurity of the products were verified by X-ray diffraction (XRD)patterns using a Philips powder diffractometer PW3040/60with Cu Kα (1.5406 Å) radiation. The surface compositionsand chemical states of the samples were determined by X-rayphotoelectron spectroscopy (XPS) (MULTILAB from ThermoVG Scientific) using monochromatic Al Kα X-rays (1486.6 eV).The size and morphology of the samples were investigatedusing a high resolution transmission electron microscopy(HRTEM), JEOL JEM 2100F (200 kV), and field emissionscanning electron microscope (FESEM, JEOL, JSM-7600F).Raman scattering measurement was performed at roomtemperature on Horiba Jobin Yvon LabRam series HR(800)spectrometer using an excitation of 514.5 nm from an Ar laser.The porosity and surface area measurement were performedafter thoroughly degassing the samples at 150 °C for 4 h, usingan N2 absorption desorption isotherm in a surface area andporosity analyzer (Micromeritics ASAP 2020).

Electrochemical Studies. Electrochemical behavior ofSRGO and its constituents (RGO and SQDs) were studiedin detail. Cyclic voltammetry (CV), amperometry, andelectrochemical impedance spectroscopy (EIS) analysis wereconducted on a CH Instruments Model 660D electrochemicalanalyzer (CH Instruments Inc., US) using a three-electrodesystem with modified glassy carbon electrode (GCE) as theworking electrode (3.14 mm2), a platinum wire as the counterelectrode, and Ag/AgCl as the reference electrode in 0.1 Mpotassium ferricyanide/potassium ferrocyanide ([Fe-(CN)6]

3−/4−) couple solution. The EIS was performed in 0.1M [Fe(CN)6]

3−/4− solution at a potential of 0.32 V in afrequency range of 10−106 Hz at a signal amplitude of 10 mV.GCEs were polished with slurries of 1, 0.3, and 0.05 μmalumina powders and rinsed thoroughly with doubly distilledwater. Then, the electrodes were sonicated alternatively inmethanol and distilled water for 15 min, and the whole processwas repeated six to seven times. A thin film of the materials (40μg) was deposited on the surface of the working electrodes bythe drop-casting method. The electrodes were left overnight todry at room temperature. The electrodes were polished andcleaned before every deposition step.

■ RESULTS AND DISCUSSIONCharacterization of the Samples. XRD patterns of the

SRGO composite as well as GO and RGO are shown in FigureS-1(Supporting Information). The XRD of GO shows anintense peak at 2θ = 10.9°, which corresponds to (002)reflection of stacked GO. After reduction, the sharp peak of GOdisappeared and a broad peak appeared at 2θ = 23° indicatingreduction of GO to RGO.8 The XRD analysis of bare SQDs(Figure S-2, Supporting Information) and SRGO reveals thatall diffraction peaks can be indexed to the tetragonal SnO2

Analytical Chemistry Article

dx.doi.org/10.1021/ac5007365 | Anal. Chem. XXXX, XXX, XXX−XXXB

structure (ICDD file no. 41-1445). The values of crystallite size(D) of pristine SQDs and QDs in SRGO were estimated fromWilliamson and Hall theorem.21 Pristine SQDs and QDs inSRGO exhibit a single phase rutile structure with an averagecrystallite size of ∼2 and ∼2.2 nm (averaged from (110), (101),and (211) peaks), respectively.The surface chemistry of the SRGO was investigated by XPS

analysis. Carbon 1s core-level XPS signals of GO and RGO aredeconvoluted into four components, as shown in Figure 1a,b,respectively. The main peak of GO and RGO due to graphiticsp2 carbon atoms is centered at about 284.6 eV. The otherpeaks of GO at 285.8 (hydroxyl carbon, C−OH), 287.0(carbonyl carbon, CO), and 288.7 eV (carboxylate carbon,OC−OH) are due to carbon atoms connecting with oxygencontaining groups.3,8 The C 1s XPS spectrum of the RGO alsohas four components: nonoxygenated ring C (284.6 eV), C−OH species (285.5 eV), CO species (287.0 eV), andcarboxylate carbon (289.3 eV). The intensity of the peaks ofRGO due to oxygen containing groups decreases as comparedto GO peaks, indicating the reduction of GO to RGO.22 TheXPS results show that most oxygen containing functionalitieswere removed during the reactions. Only small amounts ofresidual oxygen containing functionalities are left on RGO,which is also confirmed by the O 1s XPS peaks at 531.7 and532.2 eV (Figure S-3, Supporting Information). Remnantoxygen containing groups on RGO are believed to be helpful

for maintaining an even dispersion of SQDs on the RGO sheetsthrough electrostatic attraction and hydrogen bonding.22,25

The core-level XPS signals of Sn 3d as shown in Figure S-4(Supporting Information) exhibit Sn 3d5/2 and 3d3/2 doublet at487 and 495.4 eV, respectively, which corresponds to Sn4+ inthe tetragonal rutile structure. This suggests the formation ofSnO2 on the surface of RGO sheets. The separation betweenthe Sn 3d5/2 and Sn 3d3/2 levels (8.4 eV) and the area ratio(1.5) of these two peaks are exactly the same as reportedearlier.21,25 The formation of SnO2 on the surface of RGOsheets is further confirmed by the O 1s peak at binding energyof 530.8 eV in the oxygen 1s spectra of SRGO (Figure S-3,Supporting Information) which implies the presence of the O1s core level of SnO2.

25 The oxygen 1s spectra of SRGOexhibited an asymmetric peak which is deconvoluted into threepeaks.Figure 1c,d presents HRTEM micrographs of the composite

indicating SQDs decorated on the RGO sheet and their averageparticle size which is in the range of 2−3.3 nm; this is inagreement with the average crystallite size estimated from XRDline broadening. At higher magnification, SQDs (red circle) canbe seen on the wrinkled RGO surface. The SQDs show latticespacing of 0.35 nm which corresponds to the (110) plane ofrutile SnO2. The number of wrinkled layers at highermagnification, marked as red lines in Figure 1d, indicates thenumber of graphene layers in RGO. Restacking of carbona-ceous materials is a hindrance in the fabrication of such

Figure 1. Core level XPS spectra of C 1s for (a) GO and (b) SRGO. (c) Low magnification HRTEM image of the SRGO. SAED pattern of SRGO(inset of (c)), showing both the diffraction rings of SQDs and elongated diffraction spots of RGO. (d) HRTEM image of SRGO exhibitingcrystalline SQDs with sizes of 2−3 nm, adhered on the RGO. (e) Elemental mapping of SRGO depicting the even distribution of C, O, and Sn.

Analytical Chemistry Article

dx.doi.org/10.1021/ac5007365 | Anal. Chem. XXXX, XXX, XXX−XXXC

materials like RGO. In our case, SQDs being decorated onRGO play a pivotal role to prevent the restacking of RGO inthe multilayer. The selected area electron diffraction (SAED)(inset of Figure 1c) of SRGO also confirms the crystallinenature of SQDs, which shows three diffraction rings of rutileSnO2 accompanied by elongated diffraction spots arising fromthe RGO sheets.26,27 These findings serve as explicit evidenceconfirming the formation of the SRGO composite. FEGSEMimages (Figure S-5, Supporting Information) show uniformdistribution of SQDs on or between the RGO sheets, as is alsoevident from the HRTEM study. The energy dispersivespectrum of this region confirms the presence of C, Sn, andO (Figure S-6, Supporting Information). The elementalmapping (Figure 1e) of SRGO confirms uniform decorationof SQDs on RGO. Red, white, and green represents theelements C, O, and Sn, respectively.GO and SRGO composite are investigated by Raman

spectroscopy (Figure S-7, Supporting Information). TheRaman spectrum of GO contains both G (1599 cm−1) and D(1345 cm−1) bands. The Raman spectra of the SRGOcomposites exhibit both G and D bands at 1591 and 1347cm−1, respectively, along with a weaker band at 570 cm−1,which is due to the surface defects of SQDs.21 The Ramanresults reveal the amalgamation of the SQDs onto the RGOsurface, which is consistent with the HRTEM results.In order to compare the textural properties such as surface

area and pore volume of SQDs, RGO, and the SRGOcomposite, Brunauer−Emmett−Teller (BET) surface area andpore volume measurements based on the nitrogen adsorption−desorption isotherms were carried out. Figure 2a−c shows theN2 physisorption isotherms of the samples and theircorresponding BJH pore size distribution curves. The SRGOsample exhibits a hysteresis loop at the P/P0 range of 0.42−1.0.The hysteresis loop of the SRGO resembles the H1 type as perthe IUPAC classification. However, a hysteresis loop in bareSQDs and RGO is H2 type.28 SRGO has the highest BETsurface area of 272 m2/g followed by SQDs (214 m2/g) andRGO (27 m2/g). The surface area of SRGO is larger than therecently reported SQDs and graphene nanosheet composite.29

The higher surface area of SRGO is due to decrease in therestacking of the RGO sheets, resulting from the decoration ofSQDs that have a higher surface to volume ratio. The surfacearea of RGO is much less than the composite due to restackingof graphene layers. The pore size distribution calculated fromthe desorption branch using the BJH model for SQDs, RGO,and SRGO clearly indicates the presence of mesoporesaccording to the IUPAC nomenclature. The presence ofpores in SRGO is attributed to the interstitial space betweenthe QDs, the interlayer space of RGO, and voids between QDsand graphene terrain. The total pore volume (single pointadsorption) of SRGO is 0.67 cm3/g, which is much larger thanthat of RGO and SQDs. The details of texture analysis ofSQDs, RGO, and SRGO are listed in Table 1. The largersurface area and pore volume of SRGO can play a significantrole in electrochemical analysis as compared to its counterparts.Electrochemical Detection of Urea. In order to under-

stand the electrochemical properties of the SRGO compositeand its use as electrode material in electrochemical sensing ofanalytes, cyclic voltammetry (CV) was performed. Theelectroactivity of the composite was tested for selectivedetermination of urea using SRGO modified GCE by varioustechniques, viz., CV, amperometry, and EIS. To elucidate thecomparative electrochemical performance of SRGO and its

counterparts (RGO and SQDs), electrochemical measurementsof all the three materials were performed. Figure 3 shows thecyclic voltammograms of bare GCE, SRGO, RGO, and SQDsat a scan rate of 50 mV/s, within the potential window of −0.8to 1 V, in 0.1 M potassium ferricyanide/ferrocyanide[Fe(CN)6]

3−/4− electrolyte. Well-resolved anodic and cathodicpeaks are observed for all three samples. The potentialseparation (ΔE) of the two peaks for SRGO/GCE, RGO/GCE, and SQDs/GCE are 235, 203, and 563 mV, respectively.A lower value of ΔE indicates a fast electron transfer process.30

The SRGO/GCE, RGO/GCE, and SQDs/GCE showedanodic shift in the peak potential by 242, 261, and 83 mV,respectively, with respect to bare GCE. These two observationsaptly suggest that SRGO/GCE and RGO/GCE have a higherelectrocatalytic activity than SQDs/GCE.

Figure 2. (a−c) Nitrogen adsorption−desorption isotherm curves ofSQDs, SRGO, and RGO, respectively, with their corresponding poresize distribution curves.

Table 1. SBET, Pore Volume, and Pore Size of SQDs, RGO,and SRGO

samples

BETsurfacearea

(m2/g)

single point adsorptiontotal pore volume of pores

(cm3/g)

BJH desorptionaverage pore diameter

(4 V/A) Å

SQDs 214 0.16 32.4RGO 27 0.10 25.5SRGO 272 0.67 100.1

Analytical Chemistry Article

dx.doi.org/10.1021/ac5007365 | Anal. Chem. XXXX, XXX, XXX−XXXD

The surface concentration of ionic species per unit surfacearea on SRGO/GCE, RGO/GCE, and SQDs/GCE wasestimated from the plot of Ip versus scan rate (ν) (Figure S-8, Supporting Information) using the Brown−Anson model.10

The calculated values of the surface concentration of the ionicspecies (Γ) on the SRGO/GCE, RGO/GCE, and bare SQDs/GCE electrodes are approximately 9.56 × 10−9, 9.75 × 10−9,and 7.04 × 10−9 mol cm−2, respectively. The larger value ofsurface concentration indicates that large numbers of redoxspecies are available in the electrode surface for oxidation,resulting in higher Faradaic current. The anodic and thecathodic peak currents for the three materials increased withthe increase in the scan rate (Figure S-9, SupportingInformation). The peak currents vary linearly with the squareroot of the scan rate (ν) over the entire range of 5−1000 mV/s(Figure S-10, Supporting Information), suggesting diffusioncontrolled mass transfer reactions.16,31 Diffusion coefficients arecalculated using the Randles-Sevcik equation.16 The calculatedvalues are 1.176 × 10−4, 1.242 × 10−4, and 0.667 × 10−4 cm2/sfor SRGO/GCE, RGO/GCE, and SQDs/GCE, respectively.The sensing response of SRGO/GCE, RGO/GCE, and

SQDs/GCE electrodes toward urea was estimated from thechanges in the anodic peak current in the presence of urea.Addition of 4 mM urea in the electrolyte solution results in anabrupt change in the anodic peak current of SRGO/GCEelectrode (inset of Figure 4), while for RGO/GCE and SQDs/GCE electrodes, the anodic peak currents do not register any

significant change. Figure 4 shows the percentage changes ofcurrent with the addition of urea for the three electrodes as abar graph. The anodic peak current changes by 0.89%, 3.5%,and 48.11% for SQDs/GCE, RGO/GCE, and SRGO/GCE,respectively.This observation aptly suggests that SRGO/GCE has the

best electroanalytical applicability toward urea, even though thesurface concentration and diffusion coefficient for RGO/GCEand SRGO/GCE electrodes are nearly the same. As indicatedby negative zeta potential of SQDs, RGO, and SRGO, urea iselectostatically absorbed on the working electrode. Thus,electrostatically absorbed urea on the electrodes surface formsinsulating layers. These layers act as a barrier for the interfacialelectron transfer, and the current decreases with the addition ofthe urea. Though zeta potential may play a vital role for bettersensing, in this case, the values of zeta potential values are verysimilar for all the three materials: −36.9, −35.5, and −38.7 mVfor SQDs, RGO, and SRGO, respectively. Further, the surfacearea of the sensing electrode material plays a crucial role as canbe evidenced from the BET surface area of SRGO which is 10times higher than that of RGO with a high pore volume, whichmay allow the larger amount of urea absorption. Moreover, thepore diameter may also be responsible for the electrochemicaldetection of urea as SRGO has the largest pore diameter amongall other materials.To evaluate the detection limit, sensitivity, and response time

of the fabricated SRGO/GCE system toward sensing of urea,amperometric measurements were carried out. Figure 5adescribes the amperometric current response as a function oftime, and Figure 5b depicts the steady-state current depend-ence calibration curve with the urea concentration ranging from1.6 × 10−14 to 2.5 × 10−13 M and from 3.2 × 10−13 to 3.9 ×10−12 M. The inset of Figure 5b shows an amplified calibrationcurve ranging from 1.6 × 10−14 to 2.5 × 10−13 M. A constantpotential −0.344 V (vs Ag/AgCl) was applied to the SRGO/GCE electrode.The current was recorded as a function of timeonce the steady baseline was reached. The current−time profileof the amperometric response was studied with successiveaddition of 100 μL (1 pM) of urea in 6 mL of 0.1 M[Fe(CN)6]

3−/4− electrolyte at an interval of 100 s with constantstirring (300 rpm) using a magnetic bead.Instead of a single calibration curve for the whole

experimental concentration range of urea, two curves wereobtained with good linearity for two different concentrationranges. Similar observations were also reported by Lian et al.14

for the urea sensor based on molecularly imprinted chitosanfilm doped with CdS quantum dots.The regression equations are

× = × −

= × ×

− −

− −

I C

R

( 10 A) 13.8 10 (fM) 8.93

( 0.992, 1.6 10 to 2.5 10 M)

p4 3

urea

2 14 13 (1)

× = × −

= × ×

− −

− −

I C

R

( 10 A) 6.87 10 (fM) 5.68

( 0.999, 3.2 10 to 3.9 10 M)

p4 4

urea

2 13 12(2)

The developed sensor exhibited a wide linear currentresponse to the urea concentration ranging from 1.6 × 10−14

to 2.5 × 10−13 M (R2 = 0.992) and 3.2 × 10−13 to 3.9 × 10−12

M (R2 = 0.999). The detection limit is 11.7 fM as calculatedfrom the signal-to-noise ratio of 3,32 which is much lower thanthe previously reported carbonaceous materials and itscomposite based urea biosensor.17,33

Figure 3. Cyclic voltammograms of modified electrodes in 0.1 M[Fe(CN)6]

3−/4− electrolyte at a scan rate of 50 mV/s.

Figure 4. Percentage changes of current with the addition of 4 mMurea solution for SQDs/GCE, RGO/GCE, and SRGO/GCE electro-des. Inset shows the cyclic voltammogram of SRGO/GCE with urea(black) and without urea (red).

Analytical Chemistry Article

dx.doi.org/10.1021/ac5007365 | Anal. Chem. XXXX, XXX, XXX−XXXE

The detection limit is around 100 times lower than therecently reported urea electrochemical sensor based on otherQDs composite.14 A rapid response time (∼5 s) is observed fordetection. The porous structure of the SRGO matrix offers avery low transport barrier and, thus, provides a rapid diffusionof ionic species from the solution to the electrode, which leadsto a rapid response time. The sensitivity of the SRGO/GCEelectrode calculated from the slope of the calibration curve hasbeen found to be 1.38 μA/fM, which is much larger than theurea sensors’ sensitivity reported previously.10,16 Successiveaddition of urea probably blocks the pores of SRGO, whichprevents further absorption of urea and hinders electrontransfer between the electrode and the electrolyte.This alsodecreases the sensitivity (68.7 μA/pM), as can be seen from thecalculations of the slope value from the second calibrationcurve. This result confirms that the textural properties of SRGOplay a vital role in the sensing of urea. We also obtained a newset of calibration data with decreasing urea concentrationsranging from 40 to 10 fM (Figure S-11, SupportingInformation). The nature of the calibration curve remains thesame as that obtained earlier with increasing concentrations ofurea; it rules out the possibility of irreversible electrode foulingdue to urea adsorption.To have further insight into the detection of urea by the

fabricated electrode system, electrochemical impedance spec-troscopy was performed. With regard to sensors, EIS isparticularly well-suited for the analysis of the binding events onthe transducer surface and detection of molecules. Besides this,EIS can also be used to characterize surface modifications whichoccur during the recognition process on the transducers.34 Inthis work, EIS was used to monitor the impedance changes ofthe SRGO-modified electrode surface and its interfacialproperties in the presence of different concentrations of urea.EIS was performed between 106 and 10 Hz at AC amplitude of10 mV and at DC potential (Eocp = 0.32 V) in 0.1 M[Fe(CN)6]

3−/4− electrolyte. The corresponding semicirclesrepresent the best-fit curves on the equivalent circuit modelingof R(CR(QR))(CR) as shown in Figure 6. The good fit impliesthat the equivalent circuit model reasonably reveals theelectrochemical processes occurring on the modified electrodes.The electrochemical parameters of the electrode/electrolytesystem were evaluated with the ZSimpWin program. Theprocess at the electrode/electrolyte interface is represented byan equivalent circuit composed of resistance R, capacitance C,

and constant phase elements (Q), combined in parallel or inseries. The solution resistance Rs in the range of 4.535−6.749Ω, is very low, and is unaffected by the applied potentials. Thediameter of the semicircular region in the Nyquist plot (Figure6) increases with increasing urea concentration. In presence ofurea, the modified electrode shows higher values of resistanceR3. The components C1 and C2 are the double layer capacitanceof the system. However, the impedance response of the doublelayer at a modified electrode does not show the ideal behavioras pure capacitance. It rather shows a constant phase element(CPE) with a phase angle less than 90°. This infers that theimpedance of such a nonideal double layer is represented by aconstant phase element Q which is usually employed in a modelin place of a capacitor to compensate for the nonhomogeneityof the system. For an ideal electrode system, Q is equal todouble layer capacitance and n = 1. However, in this case, thevalue of n was found to be 0.88. As this value lies between 0.5and 1, the effect may be attributed to surface roughness or dueto surface heterogeneity caused by the deposition of urea.35

Such surface occurs as a result of surface coupling betweennonhomogeneous double layer capacitance and the electrolyteresistance. R1 and R2 are in the parallel connection in thecircuit, and R2 is the charge transfer resistance of the system. R3

Figure 5. (a) Current−time profile of the amperometric response of the developed sensor. (b) Calibration plots with the concentration of urearanging from 1.6 × 10−14 to 2.5 × 10−13 M and from 3.2 × 10−13 to 3.9 × 10−12 M. The inset shows an amplified calibration curve with theconcentration of urea ranging from 1.6 × 10−14 to 2.5 × 10−13 M.

Figure 6. Nyquist plots of SRGO modified electrodes with varyingconcentration of urea; the bottom shows the equivalent circuit diagramused to model impedance data.

Analytical Chemistry Article

dx.doi.org/10.1021/ac5007365 | Anal. Chem. XXXX, XXX, XXX−XXXF

may be considered as the ionic/electronic charge transferresistance of the system or surface charge transfer resistancecaused due to Faradaic reaction at the electrode/electrolyteinterface.This type of resistance (R3) appears due to absorbed

chemical species from the electrolyte,36 and it has been foundthat R3 increases linearly with urea concentration in the rangeof 7.9 × 10−15 to 0.3 × 10−12 M. This indicates migration andabsorption of urea from the electrolyte to the SRGO electrode/electrolyte interface. The presence of urea leads to an additionalinterfacial impedance which can be treated as a platform todetermine the sensitivity of the system toward detection of theanalyte. The limit of detection (LOD) was calculated to be 23.7fM (defined as S/N = 3). The regression equation is R3 (ohm)= 0.63 × Curea (fM) + 227.97, where C is the concentration ofurea in fM. The correlation coefficient is 0.974.Stability and Selectivity of the SRGO/GCE Electrode.

Shelf life of the electrode was studied for more than 2 monthsby measuring the current with respect to time at an interval of 1week. No considerable change in response characteristics (datanot shown) was observed even after 2 months which confirmsstorage capability for a long period, stability, and reproduci-bility.The selectivity of the SRGO/GCE electrode with respect to

other interfering agents, viz., ascorbic acid, uric acid, andglucose, has been studied by comparing the shape andmagnitude of the anodic current response (Figure S-12,Supporting Information). The voltammograms of urea (4mM) were obtained in the presence of various interferingagents of 40 mM each (10 times higher concentration thanurea). The interference concentration used for this study ismuch more as compared to their normal physiological values inhuman serum. No substantial changes in peak position andshape of the voltammograms were observed in the presence ofascorbic acid and uric acid; however, a negligible decrease incurrent intensity is observed. In the case of glucose, the currentintensity of the voltammogram of urea remains the same but aslight change in the peak position is observed. This may be dueto the competitive effect of the electrostatic interaction of theSRGO with the −OH groups of glucose. When one takes intoaccount the low concentration of the interfering agents inhuman serum, the proposed sensor can be used for detection ofurea in real samples.

Table 2 shows a comparative analysis of previously reportedurea sensors with the proposed enzymeless sensor taking intoaccount various electrochemical methods and their perform-ance.

■ CONCLUSIONIn conclusion, a two-step simple method has been developed tofabricate the SRGO composite, in which RGO sheets areloaded with SnO2 QDs. These QDs prevent the restacking ofgraphene layers which always remains as a big challenge to afabricator. Decrease in the restacking of graphene sheets resultsin a higher surface area for SRGO, which is a favorable criterionfor any good sensor. The SRGO based enzyme-free sensor hasbeen successfully exploited for the understanding of a proficientsensor for urea. It showed improved characteristics, such as alinear current response (R2 = 0.992), to the urea concentrationranging from 1.6 × 10−14 to 3.9 × 10−12 M and a detection limitas low as 11.7 fM (S/N = 3) by amperometry. The sensorexhibited an ultrahigh sensitivity of 1.38 μA/fM with a rapidrespose time of 5 s. Besides this, we provide a thoroughunderstanding of the mechanism of detection of urea using theSRGO electrode by impedance spectroscopy. The lowdetection limit promises its practical use in the detection ofurea in environmental samples with great ease.

■ ASSOCIATED CONTENT*S Supporting InformationXRD of GO, RGO, SRGO and SnO2 QDs, XPS data of theoxygen 1s spectra and Sn 3d of SRGO, SEM image of theSRGO, EDS of SRGO, Raman spectra of GO and SRGO, andelectrochemical data of the composite. This material is availablefree of charge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: dhirenb@iitb.ac.in. Fax: (022)25723480. Tel: (022)25767632.Author ContributionsThe manuscript was written through contributions of allauthors. All authors have given approval to the final version ofthe manuscript.NotesThe authors declare no competing financial interest.

Table 2. Sensing Characteristics of Urea Sensors Reported in Recent Literature Compared to Present Results

electrode material methods sensitivity linear range detection limitresponsetime, s ref

Urs-GLDH/CH-ZnO/ITO

cyclic voltammetry 0.13 μA/mM cm2 5−100 mg/dL 0.5 mM 10 16

Ur/NiO-NP/ITO/glass cyclic voltammetry 21.3 μA/mM cm2 0.83−16.65 mM 0.88 mM 5 10Urs/MWCNTs/SiO2/ITO

potentiometry 23 mV/decade/cm2 2.18 × 10−5−1.07 × 10−3 M 10−25 33

Urs-GLDH/Nano-ZnO/ITO

cyclic voltammetry 1.44 μA/mg/dL cm2 10−80 mg/dL 2.25 mM 15

Urs-GLDH/MLG/ITO cyclic voltammetry 5.43 μA/mgdL/cm−2 10−100 mg/dL 0.65 mM 10 17Urs/H40−Au/ITO amperometry 7.48 nA/mM 0.01−35 mM 1 × 10−5 M 3 37Ur-GLDH/CH-Fe3O4/ITO

differential pulsevoltammetry

12.5 μA/(mMcm−2) 5−100 mg/dL 8.3 × 10−5 M 10 38

CdS QDs−MIP differential pulsevoltammetry

5.0 × 10−12−7.0 × 10−8 M 1.0 pM 300 14

SRGO/GCE amperometry 1.38 μA/fM 1.6 × 10−14−3.9 × 10−12 M 11.7 fM 5 presentwork

Analytical Chemistry Article

dx.doi.org/10.1021/ac5007365 | Anal. Chem. XXXX, XXX, XXX−XXXG

■ ACKNOWLEDGMENTS

We gratefully acknowledge the financial support by nano-mission of DST, Govt. of India. D.D. acknowledges CSIR,India, for the award of Senior Research Fellowship (SRF).A.K.S. acknowledges IITB-Monash Research Academy forfinancial support. We acknowledge the central characterizationfacilities at IIT Bombay.

■ REFERENCES(1) Kilic, C.; Zunger, A. Phys. Rev. Lett. 2002, 88, 095501-1.(2) Cui, H.; Liu, Y.; Ren, W.; Wang, M.; Zhao, Y. Nanotechnology2013, 24, 345602.(3) Zhu, C.; Fang, Y.; Wen, D.; Dong, S. J. Mater. Chem. 2011, 21,16911−16917.(4) Zhang, L.-S.; Jiang, L.-Y.; Yan, H.-J.; Wang, W. D.; Wang, W.;Song, W.-G.; Guo, Y.-G.; Wan, L.-J. J. Mater. Chem. 2010, 20, 5462−5467.(5) Wang, X. R.; Tabakman, S. M.; Dai, H. J. J. Am. Chem. Soc. 2008,130, 8152−8153.(6) Su, Y.; Li, S.; Wu, D.; Zhang, F.; Liang, H.; Gao, P.; Cheng, C.;Feng, X. ACS Nano 2012, 6, 8349−8356.(7) Li, F.; Song, J.; Yang, H.; Gan, S.; Zhang, Q.; Han, D.; Ivaska, A.;Niu, L. Nanotechnology 2009, 20, 455602−455607.(8) Seema, H.; Kemp, K. C.; Chandra, V.; Kim, K. S. Nanotechnology2012, 23, 355705−355712.(9) Zhu, C.; Fang, Y.; Wen, D.; Dong, S. J. Mater. Chem. 2011, 21,16911−16917.(10) Tyagi, M.; Tomar, M.; Gupta, V. Biosens. Bioelectron. 2013, 41,110−115.(11) Ansari, S. G.; Wahab, R.; Ansari, Z. A.; Kim, Y.-S.; Khang, G.;Al-Hajry, A.; Shin, H.-S. Sens. Actuators, B 2009, 137, 566−573.(12) Velichkova, Y.; Ivanov, Y.; Marinov, I.; Ramesh, R.; RamuduKamini, N.; Dimcheva, N.; Horozova, E.; Godjevargova, T. J. Mol.Catal. B: Enzym. 2011, 69, 168−175.(13) Luong, J. H. T.; Bouvrette, P.; Male, K. B. Trends Biotechnol.1997, 15, 369−377.(14) Lian, H.-T.; Liu, B.; Chen, Y.-P.; Sun, X.-Y. Anal. Biochem. 2012,426, 40−46.(15) Ali, A.; Ansari, A. A.; Kaushik, A.; Solanki, P. R.; Barik, A.;Pandey, M. K.; Malhotra, B. D. Mater. Lett. 2009, 63, 2473−2475.(16) Solanki, P. R.; Kaushik, A.; Ansari, A. A.; Sumana, G.; Malhotra,B. D. Appl. Phys. Lett. 2008, 93, 163903−163903.(17) Srivastava, R. K.; Srivastava, S.; Narayanan, T. N.; Mahlotra, B.D.; Vajtai, R.; Ajayan, P. M.; Srivastava, A. ACS Nano 2012, 6, 168−175.(18) Ruedas-Rama, M. J.; Hall, E. A. H. Anal. Chem. 2010, 82, 9043−9049.(19) Marcano, D. C.; Kosynkin, D. V.; Berlin, J. M.; Sinitskii, A.; Sun,Z.; Slesarev, A.; Alemany, L. B.; Lu, W.; Tour, J. M. ACS Nano 2010, 4,4806−4814.(20) Swain, A. K.; Li, D.; Bahadur, D. Carbon 2013, 57, 346−356.(21) Dutta, D.; Bahadur, D. J. Mater. Chem. 2012, 22, 24545−24551.(22) Zhang, J.; Xiong, Z.; Zhao, X. S. J. Mater. Chem. 2011, 21,3634−3640.(23) Zhu, H.; Yang, D.; Yu, G.; Zhang, H.; Yao, K. Nanotechnology2006, 17, 2386−2389.(24) Liang, J.; Wei, W.; Zhong, D.; Yang, Q.; Li, L.; Guo, L. ACSAppl. Mater. Interfaces 2012, 4, 454−459.(25) Song, H.; Zhang, L.; He, C.; Qu, Y.; Tian, Y.; Lv, Y. J. Mater.Chem. 2011, 21, 5972−5977.(26) Zhong, C.; Wang, J.; Chen, Z.; Liu, H. J. Phys. Chem. C 2011,115, 25115−25120.(27) Wang, X.; Zhou, X.; Yao, K.; Zhang, J.; Liu, Z. Carbon 2011, 49,133−139.(28) Salmas, C. E.; Androutsopoulos, G. P. Langmuir 2005, 21,11146−11160.

(29) Chen, C.; Wang, L.; Liu, Y.; Chen, Z.; Pan, D.; Li, Z.; Jiao, Z.;Hu, P.; Shek, C.-H.; Wu, C. M. L.; Lai, J. K. L.; Wu, M. Langmuir2013, 29, 4111−4118.(30) Yan, J.; Sun, W.; Wei, T.; Zhang, Q.; Fan, Z.; Wei, F. J. Mater.Chem. 2012, 22, 11494−11502.(31) Prakash, A.; Chandra, S.; Bahadur, D. Carbon 2012, 50, 4209−4219.(32) Chandra, S.; Arora, K.; Bahadur, D. Mater. Sci. Eng., B 2012,177, 1531−1537.(33) Ahuja, T.; Kumar, D.; Singh, N.; Biradar, A. M.; Rajesh. Mater.Sci. Eng., C 2011, 31, 90−94.(34) Lisdat, F.; Schaf̈er, D. Anal. Bioanal. Chem. 2008, 391, 1555−1567.(35) Ates, M.; Uludag, N. J. Appl. Polym. Sci. 2012, 124, 4655−4662.(36) Sujaya, C.; Sashikala, H. D.; Umesh, G.; Hegde, A. C. J.Electrochem. Sci. Eng. 2012, 2, 19−31.(37) Tiwari, A.; Aryal, S.; Pilla, S.; Gong, S. Talanta 2009, 78, 1401−1407.(38) Kaushik, A.; Solanki, P. R.; Ansari, A. A.; Suman, G.; Ahmad, S.;Malhotra, B. D. Sens. Actuators, B 2009, 138, 572−580.

Analytical Chemistry Article

dx.doi.org/10.1021/ac5007365 | Anal. Chem. XXXX, XXX, XXX−XXXH