Nanoscale
PAPER
Publ
ishe
d on
22
Aug
ust 2
013.
Dow
nloa
ded
by U
nive
rsity
of
Vir
gini
a on
17/
11/2
013
18:1
7:16
.
View Article OnlineView Journal | View Issue
aCSIR-National Physical Laboratory, Dr K.
India. E-mail: [email protected] of Applied Chemistry, Delhi
Delhi, 110042, India
Cite this: Nanoscale, 2013, 5, 10494
Received 18th May 2013Accepted 16th August 2013
DOI: 10.1039/c3nr02575f
www.rsc.org/nanoscale
10494 | Nanoscale, 2013, 5, 10494–1
Microstructural and electrochemical impedancecharacterization of bio-functionalized ultrafine ZnSnanocrystals–reduced graphene oxide hybrid forimmunosensor applications
Sujeet K. Mishra,ab Avanish K. Srivastava,a Devendra Kumar,b Ashok M. Biradara
and Rajesh*a
We report a mercaptopropionic acid capped ZnS nanocrystals decorated reduced graphene oxide (RGO)
hybrid film on a silane modified indium-tin-oxide glass plate, as a bioelectrode for the quantitative
detection of human cardiac myoglobin (Ag-cMb). The ZnS nanocrystals were anchored over
electrochemically reduced GO sheets through a cross linker, 1-pyrenemethylamine hydrochloride, by
carbodiimide reaction and have been characterized by scanning electron microscopy, transmission
electron microscopy and energy dispersive X-ray spectroscopy. The transmission electron microscopic
characterization of the ZnS–RGO hybrid shows the uniform distribution of ultra-fine nanoparticles of
ZnS in nano-sheets of GO throughout the material. The protein antibody, Ab-cMb, was covalently linked
to ZnS–RGO nanocomposite hybrid for the fabrication of the bioelectrode. A detailed electrochemical
immunosensing study has been carried out on the bioelectrode towards the detection of target Ag-
cMb. The optimal fitted equivalent circuit model that matches the impedance response has been
studied to delineate the biocompatibility, sensitivity and selectivity of the bioelectrode. The bioelectrode
exhibited a linear electrochemical impedance response to Ag-cMb in a range of 10 ng to 1 mg mL�1 in
PBS (pH 7.4) with a sensitivity of 177.56 U cm2 per decade. The combined synergistic effects of the high
surface-to-volume ratio of ZnS(MPA) nanocrystals and conducting RGO has provided a dominant charge
transfer characteristic (Ret) at the lower frequency region of <10 Hz showing a good biocompatibility
and enhanced impedance sensitivity towards target Ag-cMb. The impedance response sensitivity of the
ZnS–RGO hybrid bioelectrode towards Ag-cMb has been found to be about 2.5 fold higher than that of
a bare RGO modified bioelectrode.
Introduction
Nanomaterials, including carbon-allotrope materials havereceived great attention due to their attractive properties andapplications in many areas, including catalysis, sensing, elec-tronics and photonics. Graphene, a one-atom-thick planar sheetof sp2-bonded carbon atoms with a dense honeycomb crystalstructure1 and unique physicochemical properties, includinghigh surface area, excellent electrical conductivity, strongmechanical strength, biocompatibility, ease of functionaliza-tion and mass production, which make it a popular researchsubject for electro-chemical applications including sensors,2–4
electric devices5 transistors,6 and fuel cells7 to name only a few.Graphene is considered to be more advantageous when
S. Krishnan Road, New Delhi, 110012,
Technological University, Bawana Road,
0503
compared to other multidimensional carbon allotropes, such ascarbon nanotubes, due to the fact that graphene does notcontain metal impurities that dominate and foul the electro-chemical behavior of the transducer.8 Graphene oxide (GO), ahighly oxidized form of graphene containing groups such ascarboxylic groups at the edges and phenol, hydroxyl and epoxygroups on the basal planes has emerged as a well knownprecursor for graphene.9 GO is an electrically insulating mate-rial due to its randomly allocated nonconductive sp3 carbonsites and disrupted sp2 bonding networks. Recently, Hayamiet al.10 explained the mechanism of proton conduction in GOciting the proton propagation of the adsorbed water lm viahydrogen-bonding networks of the hydrophilic sites present inGO as –O–, –OH, and –COOH functional groups. They observedthat the conductivity of GO is 2–3 orders of magnitude greaterthan that of bulk graphite oxide. The ndings specify thepossible applications of GO-based perfect two-dimensionalproton-conductive materials in fuel cells, sensors, and so on.Among the different methods reported for reduction of GO
This journal is ª The Royal Society of Chemistry 2013
Paper Nanoscale
Publ
ishe
d on
22
Aug
ust 2
013.
Dow
nloa
ded
by U
nive
rsity
of
Vir
gini
a on
17/
11/2
013
18:1
7:16
. View Article Online
sheets, the electrochemical reduction method is found to be apromising route for preparing reduced graphene oxide (RGO)modied electrode surface because it is simple, fast, inexpen-sive and more efficient than other methods such as chemicaland thermal reduction, with no additional element such as Nincorporated into the obtained RGO lm.11 Recent applicationsof RGO includes its use as a kind of novel electrochemicalsensing material in biological systems, such as detection ofDNA,4 proteins and pathogens,12 design of cell/bacterial nano-devices13 and drug delivery carriers.14
To obtain enhanced mechanical, thermal and electro-chemical properties, the RGO surface has been functionalizedwith different materials such as polymers, inorganic metal/semiconducting nano materials and biomaterials.15–17 Theability to retain the native structure of RGO while enabling thebioactivity of the functionalizing moiety through a surfaceconned process, as well as effective direct electron transferreaction properties, means that RGO is a suitable material forconstruction of electrochemical substrates. Wang et al. recentlyprepared a glucose oxidase biosensor based on graphene andCdS nanocrystals.18 For “green chemistry”, ZnS is very suitableas a good substitute of CdS due to its nontoxicity to the humanbody and low cost. Jian Du et al. have done a comparison studyof the electrochemical and photo electrochemical behaviors ofthree biosensors, based on the use of Au, CdS, and ZnS nano-particles–glucose oxidase (GOD) systems and they found thatbiosensors based on ZnS NPs are more sensitive and much lesstoxic to humans and the environment than CdS NPs.19 Recentapplication of ZnS in the eld of biosensor includes theformation of single-walled carbon nanotube based chemir-esistive label-free DNA sensors.20 The purpose of using water-soluble ZnS nanoparticles coated with carboxyl capping agentsis that it can allow greater affinity of binding or interaction withthe bio-target molecules, and the biocompatible nature of theZnS–RGO nanocomposite lm can provide a favorable micro-environment to retain the activity of the immobilized proteins.
According to the WHO, acute myocardial infarction (AMI) isthe result of a sudden occlusion that decreases blood ow to aportion of the myocardium, causing cell death. Symptoms ofAMI include chest pain, pressure, shortness of breath, and/ornausea but these symptoms may also be seen with non-heart-related conditions. So it is very crucial that physicians areprovided with additional information in a short space of time,enabling them to carry out quick and accurate diagnoses.Cardiac biomarker tests are intended to help detect AMI and toevaluate its severity as soon as possible so that appropriatetherapy can be initiated. Cardiac biomarkers are substancesthat are released into the blood when there is damage to theheart muscle. Typical cardiac markers used for diagnosis of AMIare cardiac myoglobin, creatine kinase-MB and cardiac tropo-nins I and T. Although not cardiac-specic, myoglobin (cMb) isone of the very earliest known markers that increase aer acutemyocardial infarction, and its rapid screening under acutephysiological conditions is fundamental. Due to its small size(17.8 kDa), facilitating its quick release into circulation (as earlyas 1–3 h upon symptom onset), as well as its high sensitivity andhigh predictive value, cMb is considered as a valuable early
This journal is ª The Royal Society of Chemistry 2013
screening test for AMI.21 The “cut-off” concentrations of cardiacmyoglobin may vary from 50 ng mL�1 (Behring Diagnosticsmethod, Nanogen cardiac STATus panel) and 56 ng mL�1
(Stratus CS STAT, for female) to 170–200 ngmL�1 (triage cardiacpanel22) with majority of researchers holding the “cut-off” toabout 100 ng mL�1.23 Several conventional methods have beenemployed to detect and quantify Mb. These include enzymelinked immunosorbent assay (ELISA),24 and chromatographic25
or spectrophotometric methods.26 These methods, however,lack the required specicity and/or involve several steps, aretime consuming and require very expensive reagents.
As one of the electrochemical technologies, electrochemicalimpedance spectroscopy (EIS) represents a powerful method forthe investigation of bulk and interfacial electrical properties ofany kind of solid or liquid material which is connected to, orpart of, an appropriate electrochemical transducer providing asensitive, non-destructive, and rapid electrochemical sensingmethod for the characterization of the electrical properties ofbiological interfaces. EIS measures the response (current and itsphase) of an electrochemical system to an applied oscillatingpotential as a function of the frequency. It is an effectivemethod to detect antigen–antibody complex formation, biotin–avidin complexation and oligonucleotide–DNA interaction,27 ascompared with other methods such as radiochemical, colori-metric and chemiluminescent methods.
This work demonstrates a facile strategy to synthesize a ZnS–RGO nanocomposite consisting of 3-mercaptopropionic acid(MPA) capped ZnS nanocrystals, ZnS(MPA), anchored on RGOsheets through a linker and deposited onto silane modiedindium-tin-oxide (ITO) glass plate for the fabrication of a bio-electrode. In this work, we have utilized large surface ZnS (MPA)nanocrystals, where the surrounding carboxyl functional groupsprovided a high loading of protein antibody, Ab-cMb, moleculesthrough strong carbodiimide linkage. The composition,morphology and the microstructure of the as-obtainedZnS(MPA)–RGO nanocomposite was characterized usingvarious instrumental techniques such as SEM, TEM and elec-trochemical techniques. The impedimetric sensing perfor-mance of the bioelectrode with ZnS–RGO nanocompositetowards the quantitative detection of target protein antigen, Ag-cMb, in phosphate buffer solution (PBS; pH 7.4) was studiedand compared with that of native RGO sheets without ZnSnanoparticles, to highlight the contribution of the ZnS nano-particles in the overall enhanced immunosensing performance.
ExperimentalChemicals and reagents
Ab-cMb (Cat4M2 MAb 4E2) & Ag-cMb (Cat 8M50) were obtainedfrom Hytest (Turku, Finland). Mouse immunoglobulin-G (Ag-IgG) (Cat IGP3) was obtained from GENEI, Bangalore, India.3-Aminopropyltriethoxysilane (APTES) was purchased fromMerck chemicals (Germany). N-(3-Dimethylaminopropyl)-N0-ethyl carbodiimide hydrochloride (EDC) and N-hydroxy succi-nimide 98% (NHS), zinc nitrate hexahydrate (Zn(NO3)2$6H2O),sodium sulde nonahydrate (Na2S$9H2O), 1-pyrenemethyl-amine hydrochloride (PyMe-NH2), 1-pyrene butanoic acid
Nanoscale, 2013, 5, 10494–10503 | 10495
Nanoscale Paper
Publ
ishe
d on
22
Aug
ust 2
013.
Dow
nloa
ded
by U
nive
rsity
of
Vir
gini
a on
17/
11/2
013
18:1
7:16
. View Article Online
succinimidyl ester and 3-mercapto propionic acid (MPA) wereobtained from Sigma-Aldrich Corp. All other chemicals were ofanalytical grade and used without further purication.
Fig. 1 Electrochemical reduction of GO/APTES/ITO-glass surface in a deaeratedsolution of 0.5 M KCl, at a scan rate of 50 mV s�1.
Apparatus
Contact angles were recorded on Drop Shape Analysis System;model DSA10MK2 from Kruss GmbH, Germany. High resolu-tion transmission electron microscopy (HR-TEM) was doneusing an FEI model: Tecnai G2 F30 and JEOLmodel: JEM 2100F.Scanning electronmicroscopy (SEM) images were obtained witha LEO 440 PC; UK based digital scanning electron micrograph.Cyclic voltammetry and EIS measurements were done on aPGSTAT302N, AUTOLAB instrument from Eco Chemie, TheNetherlands. All measurements were carried out in a conven-tional three-electrode cell conguration consisting of theproposed modied electrode as working electrode, Ag/AgCl as areference electrode and a platinum wire as a counter electrode.Electrochemical impedance spectroscopy was conducted inPBS (pH 7.4, 0.1 M KCl) solution containing 2 mM [Fe(CN)6]
3�/[Fe(CN)6]
4� in the frequency range from 1 Hz to 100 kHz at anAC voltage of 0.05 V.
Preparation of biofunctionalized of Ab-cMb(BSA)/ZnS(MPA)–RGO/APTES/ITO-glass bioelectrode
The ITO coated glass plates (10 U ,�1) were cleaned bysequential ultrasonic cleaning in dextran, acetone, ethanol,isopropyl alcohol and DI for 10 min each and dried invacuum. Then, the cleaned ITO glass plates were exposed tooxygen plasma for 5 minutes in a plasma chamber. The ITOglass plates were immersed in 2% APTES solution preparedin ethanol for 1.5 h, under the ambient conditions, to form aself assembled monolayer (SAM) of APTES. These glass plateswere then rinsed with ethanol in order to remove non-bondedAPTES from the surface of the substrate and dried under N2
gas ow. The APTES modied ITO glass plates were thenimmersed in the GO solution (0.3 mg mL�1) for a period of 1h followed by washing with distilled water and dried underN2 to form the GO/APTES/ITO-glass electrodes. Negativelycharged GO akes were deposited on the positively chargedamino modied APTES/ITO-glass plates due to electrostaticinteractions and were not removed even aer repeatedwashing. The GO/APTES/ITO-glass electrodes were then elec-trochemically reduced by cyclic voltammetry (CV) in 0.5 MKCl solution saturated with N2 gas from 0.7 to �1.1 V for 3CV cycles, at a scan rate of 50 mV s�1 (Fig. 1) to reducedgraphene oxide (RGO). The large reduction current at �1.1 Vcorresponds to the reduction of the surface oxygen groupsonly and not water since the reduction of water to hydrogenoccurs at more negative potential (e.g., �1.5 V) as shown inthe scheme.
GO + aH+ + be� / RGO + cH2O
However, this reduction peak at �1.1 V disappeared in thenext two consequent scans and is therefore irreversible,
10496 | Nanoscale, 2013, 5, 10494–10503
indicating the reduction of surface oxygenated species in therst cycle only with the formation of RGO. The remaining twoCV cycles correspond to the electrochemical behavior of theresulting RGO/APTES/ITO.
The RGO/APTES/ITO-glass electrodes were then immersed in6 mM solution of 1-pyrenemethylamine hydrochloride (PyMe-NH2) in DMF, for 2 h, at room temperature, and thereaerwashed extensively with DMF and dried under N2 gas ow. TheZnS(MPA) nanocrystals were synthesized in aqueous solution,at room temperature, by a method reported earlier.20 1-Pyr-enemethylamine functionalized RGO/APTES/ITO-glass elec-trodes were treated with a 1 mg mL�1 aqueous solution ofZnS(MPA) nanocrystals containing 0.1 M EDC and 0.05 M NHSfor 2 h and were rinsed thoroughly with double distilled water toobtain the ZnS(MPA) functionalized RGO/APTES/ITO-glasselectrodes. Ab-cMb was then covalently immobilized onZnS(MPA)-RGO/APTES/ITO-glass electrodes by immersing themin PBS buffer (pH 7.4) containing 100 mg mL�1 Ab-cMb over-night at 4 �C, followed by washing with PBS and drying with N2
gas ow. The protein antibody immobilized electrodes werefurther immersed in 1% BSA (W/V) solution to block thenonspecic binding sites and the remaining unbound freecarboxyl groups as well, followed by washing with PBS to removeany physically adsorbed antibodies and nally dried under N2
ow and stored at 4 �C. Ab-cMb was covalently immobilizeddirectly over the RGO sheets through 1-pyrene butanoic acidsuccinimidyl ester (PyBtNHS) without undergoing any func-tionalization with ZnS nanocrystals to fabricate Ab-cMb/RGO/APTES/ITO-glass for comparative study. The stepwiseconstruction of the prototype assembly is represented inScheme 1.
Results and discussionsContact angle measurement
Contact angle measurements based on the sessile drop methodwere used to determine the hydrophobic/hydrophilic characterof the surface. The image of the drop deposited on the modiedITO electrode surface was recorded by a video camera and animage-analysis system calculates the contact-angle (q) from theshape of the drop. Measurements were repeated with four drops
This journal is ª The Royal Society of Chemistry 2013
Scheme 1 Schematic representation of the stepwise fabrication of thebioelectrode.
Fig. 2 Contact angle measurement images of (a) ITO coated glass plate; (b)APTES/ITO-glass; (c) GO/APTES/ITO-glass; (d) RGO/APTES/ITO-glass; (e)ZnS(MPA)-RGO/APTES/ITO-glass; and (f) Ab-Mb/ZnS(MPA)-RGO/APTES/ITO-glass.
Fig. 3 SEM images of (a) RGO/APTES/ITO-glass at the magnification of 5 k� and(b) ZnS(MPA)-RGO/APTES/ITO-glass at the magnification of 20 k�.
Paper Nanoscale
Publ
ishe
d on
22
Aug
ust 2
013.
Dow
nloa
ded
by U
nive
rsity
of
Vir
gini
a on
17/
11/2
013
18:1
7:16
. View Article Online
of water as a test liquid probe at different regions of themodied surface and are shown in Fig. 2.
The obtained contact angle value for bare ITO-glass wasfound to be 40.15� � 2� (Fig. 2a) analogous to a hydrophilicsurface with hydroxyl groups present on it. A signicantincrease in contact angle, 73.42� � 2� (Fig. 2b) for the APTESmodied ITO-glass was observed due to the presence ofhydrophobic alkyl chains of APTES molecules. The drasticreduction of the hydroxyl groups on ITO-glass upon silanizationby APTES molecules results in a decrease of the surface freeenergy of the APTES/ITO-glass plate with respect to bare ITO-glass (the polar component more than the dispersivecomponent) leading to an increase in the contact angle. Thehydrophobic behavior altered to hydrophilic aer the coating ofthe APTES/ITO-glass surface with GO containing hydrophilicgroups like –OH and –COOH, reduces the contact angle to59.11� � 1� (Fig. 2c). Aer electrochemical reduction of thehydrophilic groups of GO/APTES/ITO-glass the contact angleagain increased to 78.70� � 2� (Fig. 2d). Further, aer modi-cation with ZnS (MPA), the contact angle of ZnS(MPA)–RGO/APTES/ITO-glass slightly decreased to 65.25� � 1� (Fig. 2e) dueto the introduction of hydrophilic free carboxyl groups available
This journal is ª The Royal Society of Chemistry 2013
on ZnS (MPA) nanocrystals. However, upon immobilizationof hydrophobic protein antibody, Ab-cMb, molecules onZnS(MPA)–RGO/APTES/ITO-glass the contact angle signicantlyincreased to 97.24� � 1�(Fig. 2f) indicating the formation of thebioelectrode.
Microstructural characteristics
Fig. 3 shows the SEM images RGO/APTES/ITO-glass andZnS(MPA)–RGO/APTES/ITO-glass. The isolated akes of GO asobserved in Fig. 3a of SEM image demonstrate that many RGOsheets have been uniformly dispersed throughout the structurewithout aggregation. In contrast, the SEM of ZnS–RGO (Fig. 3b)shows a dense particulate feature covering the entire RGOsheets indicating a strong interaction between ZnS (MPA)nanoparticles and RGO akes spreading well all over theelectrode.
High resolution transmission electron microscopy (HR-TEM), using FEI model: Tecnai G2 F30 STWIN and JEOL model:JEM 2100F, was performed on two samples, viz., GO andnanoparticles of ZnS attached on GO. In general the micro-structure of GO was noted as ultra-thin rolled layers of GO in anaggregate. The nano-sheets of GO overlap each other withdistinct boundaries. Two such distinct nano-sheets of GO aremarked as I and II and their contours are separated with whitedotted lines (Fig. 4a). An atomic scale image of such nano-sheets reveals a hexagonal honeycomb-like structure of a typicalgraphene with fringe spacing of about 0.34 nm (inset A inFig. 4a). Another inset B in Fig. 4a shows a correspondingselected area electron diffraction pattern (SAEDP) elucidatingthe spots in hexagonal array with an obvious c-axis orientationof the GO nano-sheet. A plane with Miller indices of 110 (hkl) ofa hexagonal crystal structure in reciprocal space is marked onthe SAEDP (inset B in Fig. 4a). In a composite microstructure ofZnS nanoparticles with GO nano-sheets the ZnS nanoparticlesare distributed uniformly in the matrix of GO (Fig. 4b). Multiplesheets of GO co-existing with ZnS particles are clearly visible inthe micrograph. A corresponding SAEDP of a nano-composite ofGO and ZnS exhibits the presence of prominent diffractionspots of GO (inset in Fig. 4b), similar to Fig. 4a, along with faintspots (hkl: 111 with inter-planar, d-spacing of 0.31 nm) of acubic zinc-blende type structure (reference: JCPDS card no. 80-0020, lattice constant: a ¼ 0.5345 nm) as marked in inset ofFig. 4b. An atomic scale high resolution image recorded fromthe coexisting GO and ZnS shows the presence of nanoparticlesof ZnS and fringes of GO. Fig. 4c reveals such a nanoparticle of
Nanoscale, 2013, 5, 10494–10503 | 10497
Fig. 4 (a) HR-TEM image showing the nano-sheets of graphene oxide (GO).Insets: (A) an atomic scale image of GO and (B) the corresponding selected areaelectron diffraction pattern; (b) composite of nano-sheets of graphene oxide (GO)with nanoparticles of ZnS. Inset: corresponding selected area electron diffractionpattern from GO and ZnS and (c) atomic scale image revealing the inter-planarspacing of both GO and ZnS.
Nanoscale Paper
Publ
ishe
d on
22
Aug
ust 2
013.
Dow
nloa
ded
by U
nive
rsity
of
Vir
gini
a on
17/
11/2
013
18:1
7:16
. View Article Online
ZnS of about 6 nm in size with well-oriented atomic planes at aninter-planar spacing of 0.31 nm and a partial overlap with GO ata fringe spacing of 0.34 nm. In this micrograph, a nanoparticleof ZnS with an inter-planar spacing of 0.27 nm (hkl: 200), is alsoindicated. It is clear from the atomic scale images that theindividual nanoparticles of ZnS are normally single crystalsstacked in well oriented atomic planes.
To study the chemical homogeneity of ZnS in the matrix ofGO nano-sheets, spectroscopy (EDXS attached with TEM) wasperformed (Fig. 5). Fig. 5a is a bright eldmicrograph of a nano-composite of GO with ZnS. The EDXS spectrum exhibits thepresence of the peaks corresponding to energy levels of C (0.277keV; Ka), Zn (1.011 keV; La and 8.631 keV; Ka), S (2.30 keV; Ka)
Fig. 5 HR-TEM image and corresponding EDXS chemical measurements;showing (a) compositemicrostructure of nano-sheets of graphene oxide (GO)withnanoparticles of ZnS, (b) corresponding EDXS spectrumwithX-axis: 0 to 10 keVandY-axis: intensity in arbitrary units, (c) elemental map of C, (d) elemental map of Zn,(e) elemental map of S and (f) overlapped elemental maps of C, Zn and S.
10498 | Nanoscale, 2013, 5, 10494–10503
and Cu (0.929 keV; La and 8.0431 keV; Ka), as displayed inFig. 5b. The presence of Cu in the spectrum is due to the 200-mesh size Cu-TEM grids used in microscopy experiments. TheEDXS spectrum was further utilized to interpret the image-spectrum and therefore the different elemental maps (C, Zn, S)corresponding to the bright eld micrograph were determinedand are displayed in Fig. 5c–e. Moreover an overlappedelemental map encompassing all three elements (C, Zn, S) hasalso been plotted (Fig. 5f). It is evident from these chemicalmaps of different elements that the distribution of ultra-nenanoparticles of ZnS in nano-sheets of GO is uniformthroughout the material. In general the electron microscopyobservations have delineated that the nano-sheets of GO andultrane nanoparticles of ZnS are integrated uniformlythroughout the microstructure of these nano-composites.
Electrochemical characterization of Ab-cMb(BSA)/ZnS (MPA)–RGO/APTES/ITO-glass bioelectrode
Alternating current (AC) impedance or EIS is a powerful tool toprobe the features of a surfacemodied electrode. EIS combinesthe analysis of both the resistive and capacitive properties ofmaterials, based on the perturbation of a system at equilibriumby a small amplitude sinusoidal excitation signal. The generalRandles electronic equivalent circuit (inset Fig. 6a), which is
Fig. 6 (a) Nyquist plots obtained for bare ITO glass; APTES/ITO; GO/APTES/ITO;RGO/APTES/ITO; ZnS(MPA)–RGO/APTES/ITO; and Ab-cMb(BSA)/ZnS(MPA)-RGO/APTES/ITO in PBS (pH 7.4, 0.1 M KCl) solution containing 2 mM [Fe(CN)6]
3�/[Fe(CN)6]
4� and (b) corresponding Bode plots.
This journal is ª The Royal Society of Chemistry 2013
Paper Nanoscale
Publ
ishe
d on
22
Aug
ust 2
013.
Dow
nloa
ded
by U
nive
rsity
of
Vir
gini
a on
17/
11/2
013
18:1
7:16
. View Article Online
oen used to model electrochemical transformations occurringat the electrode/electrolyte interface, includes the following fourelements: (i) the ohmic resistance of the electrolyte solution, Rs;(ii) the Warburg impedance, Zw, resulting from the diffusion ofions from the bulk electrolyte to the electrode interface; (iii) theelectron transfer resistance, Ret, and (iv) the interfacial doublelayer capacitance (Cdl) between an electrode and a solution,relating to the surface condition of the electrode. Since thesurface of the electrodewas very rough, it had a larger real surfacearea; therefore, we used a constant phase element (CPE) insteadof the classical capacitance to t the impedance data which isgenerally ascribed to surface inhomogeneity, roughness orfractal geometry, electrode porosity, and to current and potentialdistributions associated with electrode geometry. The imped-ance of CPE can be expressed by:
ZCPE(u) ¼ Z0(ju)�n (1)
where Z0 is a constant, j is the imaginary number, u the angularfrequency (u ¼ 2pf, in Hz) and n is the CPE exponent which canbe used as a gauge of the heterogeneity and gives details aboutthe degree of surface inhomogeneity (roughness). Dependingon the value of n, CPE can represent resistance (n ¼ 0, Z0 ¼ R),capacitance (n ¼ 1, Z0 ¼ 1/C), inductance (n ¼ �1, Z0 ¼ L) or aWarburg element (n ¼ 0.5). As shown in Table 1, the value of ‘n’for different modied electrodes is close to 1, indicating thepresence of minimal defects in the modied layer on the elec-trode surface and that CPE in this case resembled closely acapacitor. In the modied tting equivalent circuit, parallelelements (CPE and Zw + Ret) were introduced since the totalcurrent through the working interface was the sum of respectivecontributions from the Faradaic process and double layercharging. The two components of the equivalent circuit, Rs andZw are not affected signicantly by modications to the elec-trode surface, while the remaining two Cdl and Ret are controlledby the surface modication of the electrode. In our case, anegligible change in Rs was observed during the modicationprocess so it has been ignored. At the same time, as can also beseen in Table 1, the changes in Ret values were much larger thanthose in other impedance components. Thus, changes in the Retvalue were taken as a suitable signal for sensing the interfacialproperties of the prepared bioelectrode during all the surfacemodication steps and for immunoreaction as well.
Table 1 CV and EIS characteristics parameters at various stages of electrode fabric
Electrodes DEp/mV Ret/U cm2
CPE
Z0/mF
Bare ITO-glass 180 184.8 3.65APTES/ITO 157 99.8 4.79GO/APTES/ITO 318 416.3 3.75RGO/APTES/ITO 182 139.0 5.24ZnS(MPA)–RGO/APTES/ITO 327 578.5 4.48Ab-cMb-ZnS(MPA)–RGO/APTES/ITO 347 699.0 4.20
a DEp ¼ redox potential; Ret ¼ charge transfer resistance; CPE¼ constant pZw ¼ Warburg resistance.
This journal is ª The Royal Society of Chemistry 2013
The total impedance (Z), which is the current–voltage ratio,can be represented by eqn (2) as a sum of in-phase (Z0) and out-of-phase (Z0 0) impedances:
Z(u) ¼ Z0(u) + iZ00(u) (2)
Fig. 6a represents the Nyquist diagram of the electrodesprepared at various stages of surface modication. For a revers-ible reaction at a solid interface in solution, these plots typicallyexhibit two important regions: (i) a semicircle at high frequen-cies which is dictated by solution resistance, charge transferresistance and the capacitance of the electrochemical cell wherethe semicircle diameter equals Ret; (ii) a straight region at lowerfrequencies, depicting the diffusion-limited transport of theredox species from the electrolyte to the electrode interface. Thesimulated values of the equivalent circuit elements are summa-rized in Table 1. The Chi-squared function (c2) provides a goodindication of the quality of the t. Generally, the c2 statisticshould be less than 10�6 if the data are said tot within the noiselevel of the measurement. Therefore, a value of 10�5 to 10�4
indicates a reasonably good t. Small values of c2 (Chi-squared)on the order of 10�4, as obtained in our case (Table 1), suggestthat the chosen selected tting circuit model is the mostappropriate. We should noted that the Ret value of 99.8 U cm2
(Table 1) for the APTES/ITO-glass electrode is much lower thanthat of a bare ITO-glass electrode (Ret ¼ 184.8 U cm2) indicatingeasy electronic transport at the electrode solution interface aerthe formation of a SAM of APTES. Aer the modication of theAPTES/ITO-glass electrode surface by GO the Ret value increasedsharply to 416.3 U cm2. This may be attributed to mainly tworeasons rst, GO is less electrically conducting due to theoxidized surface and secondly, the negatively charged carboxyl(COO�) and hydroxyl (OH�) functional groups present on thesurface generates the repulsive force to the negatively charged[Fe(CN)6]
3�/4� probe. The reduction of the most of these func-tional groups from the GO/APTES/ITO-glass electrode surfaceaer electrochemical reductionprovided adrop off inRet value to139 U cm2. The incorporation of ZnS(MPA) nanoparticles on theRGO/APTES/ITO-glass electrode exhibited an increased Ret valueof 578.5U cm2 due to the combined effect of the semiconductingbehavior of ZnS(MPA) and the repulsive interaction between theCOO� groups of ZnS (MPA) and the anionic probe [Fe (CN)6]
3�/4�
at the electrode/solution interface. Aer the immobilization of
ationa
k0 (�10�4)/m s�1
Zw (�10�4)/U cm2
c2
(�10�4) d/mmcm�2 n
0.95 2.87 8.71 1.20 67.350.90 5.32 9.05 1.32 62.890.93 1.27 12.5 1.10 89.510.90 3.82 8.61 1.41 67.720.92 0.91 8.77 1.70 90.770.93 0.83 7.06 1.95 93.51
hase element; k0 ¼ apparent rate constant; d¼ diffusion layer thickness;
Nanoscale, 2013, 5, 10494–10503 | 10499
Fig. 7 CV of Ab-cMb/ZnS(MPA)–RGO/APTES/ITO electrode as a function of scanrate in PBS (pH 7.4,0.1 M KCl) containing 2 mM [Fe(CN)6]
3�/4�. Inset: plot of redoxpeak current vs. n1/2 (mV s�1).
Nanoscale Paper
Publ
ishe
d on
22
Aug
ust 2
013.
Dow
nloa
ded
by U
nive
rsity
of
Vir
gini
a on
17/
11/2
013
18:1
7:16
. View Article Online
protein antibody, Ab-cMb, on the surface of the ZnS(MPA)–RGO/APTES/ITO-glass electrode and on subsequent treatment with ablocking protein, bovine serum albumin (BSA), the Ret furtherincreased to 699 U cm2. This may be attributed to the formationof a protein antibody layer on the electrode, which acts as theelectron communication and mass-transfer blocking layer,thereby hindering the access of redox probes towards the elec-trode surface.
The behavior of different circuit elements of the tted Ran-dles' circuit with the change in frequency is more pronouncedin the Bode plot (Fig. 6b).
The Bode plot has some distinct advantages over the Nyquistplot. Since frequency appears as one of the axes, it's easy tounderstand from the plot how the impedance depends on thefrequency and hence it can provide information about certainkinetic phenomena occurring at the electrode/solution interfaceat a different range of applied frequencies. At very highfrequencies ranging from 10 kHz to 100 kHz, the only imped-ance is the ohmic resistance of the electrolyte solution (Rs) forwhich negligible changes were observed at different stages ofsurface modication, indicating that the Rs is independent offrequency. As the frequency drops, the contribution of thecapacitive element CPE becomes signicant and so in theintermediate range of frequency i.e. from 50 Hz to 10 kHzalmost a straight line curve is obtained which is indicative ofcapacitance behavior. Finally, in the low frequency region (<50Hz) the impedance corresponding to CPE becomes very high sothe only current that ows is through Ret and Zw. It is in this lowfrequency region that we observed prominent changes inimpedance with each surface modication step of the ITO-glasselectrode leading to the formation of the bioelectrode. As can beseen from the Fig. 6b the Ret behavior of the bioelectrode hasshied towards the lower frequency region which is indicativeof the good biocompatible nature of the bioelectrode
The heterogeneous electron transfer rate constant (k0) valuesof the [Fe(CN)6]
3�/4� redox couple for the unmodied bare ITO-glass electrode and aer surface modication were determinedby using Ret values obtained from their corresponding imped-ance plots. The corresponding k0 values of the modied elec-trodes were calculated by using charge transfer kinetics:
k0 ¼ RT/n2F2ARetC (3)
whereR is the gas constant,T is the temperature, n is the electrontransfer constant of the redox couple, F is the Faraday constant, Ais the area of the electrode (0.25 cm2), and C is the concentrationof the redox couple in the bulk solution. Table 1 shows the cor-responding k0 values obtained at different stages ofmodicationof the bare ITO electrode using eqn (3). Notable amongst these isthe almost three fold increase in the k0 value aer the electro-chemical reduction of the GO/APTES/ITO-glass electrode, indi-cating a faster electron transport at the RGO/APTES/ITO-glasselectrode/solution interface. The signicant decrease in k0 valuesaer modication of the RGO/APTES/ITO-glass electrode withZnS nanoparticles and on further subsequent protein immobi-lization is due to the sluggish electron transport resulting fromthe repulsive forces of the negatively charged functional groups
10500 | Nanoscale, 2013, 5, 10494–10503
and insulating behavior of the protein molecules, respectively,at the electrode/solution interface. According to Davies andCompton, the diffusion layer thickness, d, which helps to cate-gorize the type of diffusion occurring at the electrodes, can beobtained from themodied Einstein equation for the root meansquare displacement of diffusing particles:28
d ¼ (2DDE/y)1/2 (4)
where D is diffusion coefficient of aqueous ferrocyanide (6.3 �10�6 cm2 s�1), DE is the potential width of the voltammogramand y is the scan rate (0.05 V s�1). The corresponding d valuesfor each step of surface modication are given in Table 1. Fromthe d values (Table 1), it can be observed these correspond to thecase 3 behavior of the voltammetric responses at spatiallyheterogeneous electrodes, which is associated with an overlapof adjacent diffusion layers resulting from the small size of theinert part of the electrode.28
The nature of the redox process occurring at the electrode/electrolyte interface and the surface protein concentration ofthe Ab-cMb/ZnS(MPA)–RGO/APTES/ITO-glass bioelectrode wasdetermined by taking various CVs of the bioelectrode atdifferent scan rates in PBS (pH 7.4, 0.1 M KCl) solution con-taining 2 mM [Fe(CN)6]
3�/4� (Fig. 7). The inset of Fig. 7 showsthe plot of the anodic (Ipa) and cathodic peak currents (Ipc)versus square root of the scan rates (n1/2). Both the Ipa and Ipccurrents are proportional to the n1/2, suggesting a diffusioncontrolled process at the electrode surface.
The linear relationship between the redox peak current Ipaand n1/2 which can be expressed by the eqn (5),
Ipa(n1/2) ¼ bn1/2 + 14.16 (5)
where the slope, b ¼ 8.04 � 0.02, has a correlation coefficient of0.996.
The protein concentration at the bioelectrode surface wasdetermined by using eqn (6) of the Brown–Anson model29
I ¼ n2F2GAn/4RT (6)
This journal is ª The Royal Society of Chemistry 2013
Paper Nanoscale
Publ
ishe
d on
22
Aug
ust 2
013.
Dow
nloa
ded
by U
nive
rsity
of
Vir
gini
a on
17/
11/2
013
18:1
7:16
. View Article Online
where n is the number of electrons transferred, F is the Faradayconstant (96 485.34 C mol�1), A is the surface area (0.25 cm2), Ris the gas constant (8.314 J mol�1 K�1), G is the surface proteinconcentration of the bioelectrode (mol cm�2), T is 298 K and I/nis the slope of the calibration plot. The surface proteinconcentration was found to be 13.6 � 10�10 mol cm�2.
Electrochemical impedance response to protein antigen Ag-cMb
The electrochemical impedance response of the fabricated Ab-cMb(BSA)/ZnS(MPA)–RGO/APTES/ITO-glass bioelectrode for thedetection of the target protein antigen, Ag-cMb, was carried outin PBS (pH 7.4), containing 0.1 M KCl and 2 mM [Fe(CN)6]
3�/4�
as redox probe, at scanning frequencies from 1 to 100 kHz. Theintersheet hopping mechanism of the electrical charge occur-ring within the stacked layers of the RGO imparts a greatersensitivity to the Ret value in the EIS sensing system30 where thecharge transfer mechanism is governed by the antibody–antigen interaction.
The EIS response of the bioelectrode corresponding to theaddition of different concentrations of Ag-cMb is shown inFig. 8, wherein a response for a sample solution containing notarget Ag-cMb was considered as a control sample response. Itwas found that aer each successive addition of the aliquots of
Fig. 8 (a) Faradaic impedance spectra of the Ab-cMb/ZnS(MPA)–RGO/APTES/ITO-glass bioelectrode before and after incubation with different concentrationsof Ag-Mb in PBS (pH 7.4) with a 0.1 M KCl solution containing 2 mM[Fe(CN)6]
3�/4� and (b) the corresponding Bode plots.
This journal is ª The Royal Society of Chemistry 2013
different concentrations of target Ag-cMb the Ret value changes(Table 2) considerably with a change in the diameter of thesemicircle portion of the Nyquist plot (Fig. 8a). The increase inRet value can be explained on the basis of the resulting immu-noreaction at the bioelectrode which takes place between anti-body–antigen interactions aer every successive addition ofAg-cMb. When antigens bind to the surface-immobilized anti-bodies, the access of the redox couple is hindered to a higherdegree than in the absence of antigens. As the Faradaic reactionof a redox couple becomes increasingly hindered, the electrontransfer resistance increases and the capacitance decreasesaccordingly. It is known that the antigen–antibody complex actsas a layer, disturbing ion diffusion and changing electricalcapacitance. Both factors affect signicantly the electro-chemical impedance of electrodes during formation of theantigen–antibody complex.
A more clear description of the electrochemical system'sfrequency-dependent behavior is provided by using frequency(f) dependent Bode impedance (Z) and phase (F) curves,shown in Fig. 8b. The frequency region from 10 kHz and abovewhere phase angle (F) is nearly zero is representative ofsolution resistance (Rs). In the intermediate range of frequencyfrom 10 Hz to 10 kHz a straight line with a phase angle greaterthan 70� but less than 90� was obtained, indicating thepseudo-capacitive nature of the circuit element. At the lowerend of the frequency region (<10 Hz), the phase angle wasfound to be approaching zero but not completely zero, due tothe presence of both Ret and Zw. It is important to note herethat insignicant changes observed in the capacitive region ofthe Bode plot on immunoreaction indicated the non-capacitative behavior of the bioelectrode. As can be seen fromthe plot, the maximum changes in the impedance wereobserved on immunoreaction at a frequency less than 10 Hz,which is the region mainly dominated by Ret, therefore we tookRet as the main sensing element in the impedance measure-ment of the immunoreaction.
Fig. 9 shows a linear relationship between the changein specic electron charge transfer resistance (DRet ¼(Ret)aer immunoreaction � (Ret)control) and logarithmic value ofAg-Mb concentration in the range of 10 ng to 1 mg mL�1 and isrepresented by the eqn (7):
DRet(log[Ag–Mb]) ¼ blog[Ag–Mb] + 435.5 (7)
The bioelectrode shows an Ret sensitivity (slope b of thecalibration curve) of 177.56 U cm2 per decade of Ag-cMb havinga correlation coefficient 0.989 (n ¼ 5). Though no saturation inthe impedance response was observed even beyond 1 mg mL�1
of Ag-cMb, the experimental data was concluded only up to thisconcentration keeping in view the physiological range of Ag-cMb in human body. This also highlighted the good biocom-patibility of the constructed bioelectrode.
The selectivity of the bioelectrode towards Ag-cMb was alsotested by carrying out an immunoreaction with the non specicprotein antigen, Ag–IgG, in a range of 10 ng to 1 mg mL�1 underidentical conditions. However, no considerable changes in the
Nanoscale, 2013, 5, 10494–10503 | 10501
Table 2 EIS characteristics parameters of the bioelectrode on immunoreaction with different concentration of target Ag-cMb
Concentrationof Ag-cMb
Ret/U cm2
CPEk0 (�10�4)/m s�1
Zw (�10�4)/U cm2
c2
(�10�4)Z0/mF cm�2 n
Control 699 4.20 0.932 0.75 7.06 1.950.01 mg mL�1 800 4.10 0.933 0.66 6.58 1.950.05 mg mL�1 900 4.08 0.931 0.59 5.27 1.940.10 mg mL�1 1010 4.04 0.934 0.52 4.65 2.210.50 mg mL�1 1093 4.02 0.933 0.48 4.59 2.321.00 mg mL�1 1170 3.90 0.936 0.45 4.29 2.49
Fig. 9 Concentration dependent calibration curve for Ab-cMb/ZnS(MPA)–RGO/APTES/ITO and Ab-cMb(BSA)/RGO/APTES/ITO bioelectrode; the error barsrepresent the standard deviation from three separate experiments.
Nanoscale Paper
Publ
ishe
d on
22
Aug
ust 2
013.
Dow
nloa
ded
by U
nive
rsity
of
Vir
gini
a on
17/
11/2
013
18:1
7:16
. View Article Online
Ret were observed with added aliquots of the increasingconcentration of IgG with respect to the control sample withoutIgG, showing almost no antibody–antigen interaction at theelectrode surface, thereby indicating the selectivity of the bio-electrode only towards the specic target Ag-cMb. In order tocompare the performance of the ZnS/RGO composite materialsover RGO, a control immunoreaction experiment was carriedout with a bare RGO bioelectrode.
The results (Fig. 9) indicate that the sensitivity of theZnS(MPA)/RGO modied bioelectrode towards Ag-cMb is about2.5 fold higher than that of the bare RGO bioelectrode (73.61 U
cm2 per decade). This sensitivity was found to be higher thanthose of the recently reported bioelectrodes for Ag-cMb detec-tion.31–33 This shows that the reduced graphene oxide acts asbetter matrix for semiconductor/metal nanohybrid basedimmunosensors than the earlier reported matrices includingconducting polymers34,35 because of the uniform distribution ofcolloidal nanocrystals and coherent ionic diffusion. Thestability of the bioelectrode was also investigated by repeatedlycarrying out impedimetric response measurements on thebioelectrode for the same sample of Ag-cMb under identicalconditions. With the appearance of no signicant changes inthe Ret values, even aer the 10 repeated impedance measure-ments, it could be concluded that the ZnS–RGO nanocompositematrix has good biocompatibility both in the solution and inthe open as well.
10502 | Nanoscale, 2013, 5, 10494–10503
Conclusion
We have demonstrated the fabrication of electrochemicallyreduced graphene oxide–ZnS nanocrystal (ZnS–RGO) hybridsdeposited onto the surface of an indium-tin-oxide glass plate toconstruct a bioelectrode. The effectiveness of the ZnO–RGO forthe immobilization of a protein antibody, Ab-cMb, in nativeconguration has been delineated with the impedimetricresponse of the bioelectrode towards the detection of the targetprotein antigen, Ag-cMb in PBS (pH 7.4). The bioelectrodeprovided a linear range of impedimetric detection of Ag-cMbfrom 10 ng to 1 mg mL�1 with a sensitivity of 177.56 U cm2 perdecade. The ZnO–RGO hybrid modied electrode exhibited anenhanced sensitivity of about 2.5 fold higher than the bare RGOmodied bioelectrode, indicating a strong antibody–antigeninteraction at the hybrid surface. The combined synergisticeffects of the high surface-to-volume ratio of ZnS(MPA) nano-crystals with high protein loading and conducting RGO lead tothe construction of a selective, highly sensitive and biocom-patible electrode for Ag-cMb detection. This ZnO–RGO hybridplatform could be adapted to immobilize other biomoleculesfor various biosensing applications.
Acknowledgements
We are grateful to Prof. R. C. Budhani, Director, NationalPhysical Laboratory, New Delhi, India for providing facilities. S.K. Mishra is thankful to Council of Scientic and IndustrialResearch, India for providing a senior research fellowship(SRF). We also acknowledge Mr V. Sharma and V. K. Tanwar fortechnical assistance.
References
1 M. J. Allen, V. C. Tung and R. B. Kaner, Chem. Rev., 2010, 110,132.
2 X. C. Dong, W. Huang and P. Chen,Nanoscale Res. Lett., 2011,6, 60.
3 Z. Y. Yin, X. Huang, J. Zhang, S. X. Wu, P. Chen, G. Lu,P. Chen, Q. C. Zhang, Q. Y. Yan and H. Zhang, Nanoscale,2012, 4, 293.
4 Z. Wang, J. Zhang, P. Chen, X. Zhou, Y. Yang, S. Wu, L. Niu,Y. Han, L. Wang, P. Chen, F. Boey, Q. Zhang, B. Liedberg andH. Zhang, Biosens. Bioelectron., 2011, 26, 3881.
This journal is ª The Royal Society of Chemistry 2013
Paper Nanoscale
Publ
ishe
d on
22
Aug
ust 2
013.
Dow
nloa
ded
by U
nive
rsity
of
Vir
gini
a on
17/
11/2
013
18:1
7:16
. View Article Online
5 B. Li, X. Cao, H. G. Ong, J. W. Cheah, X. Zhou, Z. Yin, H. Li,J. Wang, F. Boey, W. Huang and H. Zhang, Adv. Mater., 2010,22, 3058.
6 Q. He, S. Wu, S. Gao, X. Cao, Z. Yin, H. Li, P. Chen andH. Zhang, ACS Nano, 2011, 5, 5038.
7 Y. Shao, S. Zhang, C. Wang, Z. Nie, J. Liu, Y. Wanga andY. Lin, J. Power Sources, 2010, 195, 4600.
8 M. Pumera, A. Ambrosi, A. Bonanni, E. L. K. Chng andH. L. Poh, TrAC, Trends Anal. Chem., 2010, 29, 954.
9 S. Park and R. S. Ruoff, Nat. Nanotechnol., 2009, 4, 217.10 M. R. Karim, K. Hatakeyama, T. Matsui, H. Takehira,
T. Taniguchi, M. Koinuma, Y. Matsumoto, T. Akutagawa,T. Nakamura, S. Noro, T. Yamada, H. Kitagawa andS. Hayami, J. Am. Chem. Soc., 2013, 135, 8097.
11 Z. Wang, S. Wu, J. Zhang, P. Chen, G. Yang, X. Zhou,Q. Zhang, Q. Yan and H. Zhang, Nanoscale Res. Lett., 2012,7, 161.
12 H. Chang, L. Tang, Y. Wang, J. Jiang and J. Li, Anal. Chem.,2010, 82, 2341.
13 L. Feng, Y. Chen, J. Ren and X. Qu, Biomaterials, 2011, 32,2930–2937.
14 X. Sun, Z. Liu, K. Welsher, J. T. Robinson, A. Goodwin,S. Zaric and H. Dai, Nano Res., 2008, 1, 203.
15 J. M. Lee, Y. B. Pyun, J. Yi, J. W. Choung and W. I. Park,J. Phys. Chem. C, 2009, 113, 19134.
16 W. Zou, J. Zhu, Y. Sun and X. Wang, Mater. Chem. Phys.,2011, 125, 617.
17 J. Shia, H. Zhang, A. Snyder, M. Wang, J. Xie,D. M. Porterelda and L. A. Stanciu, Biosens. Bioelectron.,2012, 38, 314.
18 K. Wang, Q. Liu, Q. Guan, J. Wu, H. Li and J. Yan, Biosens.Bioelectron., 2011, 26, 2252.
19 J. Du, X. Yu and J. Di, Biosens. Bioelectron., 2012, 37, 88.
This journal is ª The Royal Society of Chemistry 2013
20 Rajesh, B. K. Das, S. Srinives and A. Mulchandani, Appl. Phys.Lett., 2011, 98, 013701.
21 A. S. Jaffe, L. Babuin and F. S. Apple, J. Am. Coll. Cardiol.,2006, 48, 1.
22 S. M. Ng, P. Krishnaswamy, R. Morissey, P. Clopton,R. Fitzgerald and A. S. Maisel, Am. J. Cardiol., 2001, 88, 611.
23 F. D. Serio, G. Antonelli, P. Trerotoli, M. Tampoia,A. Matarrese and N. Pansini, Clin. Chim. Acta, 2003, 333, 185.
24 F. Torabi, H. R. M. Far, B. Danielsson and M. Khayyami,Biosens. Bioelectron., 2007, 22, 1218.
25 B. M. Mayr, O. Kohlbacher, K. Reinert, M. Sturm, C. Gropl,E. Lange, C. Klein and C. G. Huber, J. Proteome Res., 2006,5, 414.
26 Q. Zheng, Z. Liu and R. Cai, Spectrochim. Acta, Part A, 2005,61, 1035.
27 B. Derkus, E. Emregul, C. Yucesan and K. C. Emregul,Biosens. Bioelectron., 2013, 46, 53.
28 S. A. Mamuru and K. I. Ozoemena, Electroanalysis, 2010, 22,985.
29 M. J. Klink, E. I. Iwuoha and E. E. Ebenso, Int. J. Electrochem.Sci., 2011, 6, 2429.
30 K. S. Kim, Y. M. Um, J. Jang, W. Choe and P. J. Yoo, ACS Appl.Mater. Interfaces, 2013, 5, 3591.
31 N. Puri, S. K. Mishra, A. Niazi, A. M. Biradar and Rajesh,Synth. Met., 2013, 169, 18.
32 F. Darain, P. Yager, K. L. Gan and S. C. Tjin, Biosens.Bioelectron., 2009, 24, 1744.
33 T. M. O'Regan, L. J. O'Riordan, M. Pravda, C. K. O'Sullivanand G. G. Guilbault, Anal. Chim. Acta, 2002, 460, 141.
34 E. G. Matveeva, Z. Gryczynski and J. R. Lakowicz, J. Immunol.Methods, 2005, 302, 26.
35 S. K. Mishra, R. Pasricha, A. M. Biradar and Rajesh, Appl.Phys. Lett., 2012, 100, 053701.
Nanoscale, 2013, 5, 10494–10503 | 10503
