Analyst

PAPER

Publ

ishe

d on

14

Janu

ary

2014

. Dow

nloa

ded

by U

nive

rsity

of

Sout

hern

Cal

ifor

nia

on 0

2/04

/201

4 11

:24:

59.

View Article OnlineView Journal | View Issue

aInstitut de Recherche Interdisciplinaire (IRI

de la Haute Borne, 50 avenue de Halley, BPbTaras Shevchenko University, 60 VladimirscInstitute for Materials Research (IMO), Has

Diepenbeek, BelgiumdIMOMEC, IMEC, Wetenschapspark 1, 3590eInstitute of Physical Chemistry, Polish Acad

Warszawa, Poland

Cite this: Analyst, 2014, 139, 1726

Received 30th October 2013Accepted 11th January 2014

DOI: 10.1039/c3an02045b

www.rsc.org/analyst

1726 | Analyst, 2014, 139, 1726–1731

An impedimetric immunosensor based on diamondnanowires decorated with nickel nanoparticles

Palaniappan Subramanian,a Anastasiia Motorina,ab Weng Siang Yeap,c Ken Haenen,cd

Yannick Coffinier,a Vladimir Zaitsev,b Joanna Niedziolka-Jonsson,e

Rabah Boukherrouba and Sabine Szunerits*a

Nanostructured boron-doped diamond has been investigated as a sensitive impedimetric electrode for the

detection of immunoglobulin G (IgG). The immunosensor was constructed in a three-step process: (i)

reactive ion etching of flat boron-doped diamond (BDD) interfaces to synthesize BDD nanowires (BDD

NWs), (ii) electrochemical deposition of nickel nanoparticles (Ni NPs) on the BDD NWs, and (iii)

immobilization of biotin-tagged anti-IgG onto the Ni NPs. Electrochemical impedance spectroscopy

(EIS) was used to follow the binding of IgG at different concentrations without the use of any additional

label. A detection limit of 0.3 ng mL�1 (2 nM) with a dynamic range up to 300 ng mL�1 (2 mM) was

obtained with the interface. Moreover, the study demonstrated that this immunosensor exhibits good

stability over time and allows regeneration by incubation in ethylenediaminetetraacetic acid (EDTA)

aqueous solution.

1. Introduction

Immunoassays relying on antibody–antigen interactions providea promising means of analysis due to their specicity andsensitivity. High specicity is achieved mainly by the molecularrecognition of the target analyte by antibodies or antigensthrough the formation of a stable immunocomplex. The bindingaffinities are typically in the range of Kd¼ 10�3 to 10�7 M and aredictated by the type of interaction involved and the surface areaburied in the paratope–epitope interface.1 Immunosensing hastypically relied on standard sandwich bioaffinity assays inconjunction with labels such as enzymes, uorophores ornanoparticles.2–9 Electrochemical immunosensing based onelectrochemical impedance spectroscopy (EIS) has become awidely accepted alternative method for the detection of antigen–antibody interactions.10–16 By measuring the frequency depen-dence of the electrical impedance, EIS provides information thatcan be used to dissect the response of complex interfaces. Thislabel-free real time analytical method has shown to be capable ofproviding equivalent analytical characteristics in terms of selec-tivity and reliability with the additional advantages of being easyto perform, fast and cost effective. Furthermore, compared to

), CNRS USR 3078, Universite Lille1, Parc

70478, 59658 Villeneuve d'Ascq, France

kaya str., Kiev, Ukraine

selt University, Wetenschapspark 1, 3590

Diepenbeek, Belgium

emy of Sciences, Kasprzaka 44/52, 01-224

immunosensors using labels, electrochemical sensors offerhigher detection limits. Recently Kim and co-workers reported onultrasensitive antigen detection with a limit of detection of 100fg mL�1 using a reduced graphene oxide-based electrochemicalimmunosensor with horseradish peroxidase labeled secondaryantibodies as ampliers.4 An alkaline phosphatase (ALP)amperometric immunosensor with a detection limit of 0.3ngmL�1 has been developed by Einaga and it is based on the useof boron-doped diamond (BDD) electrodes modied with poly-(o-aminobenzoic acid).3 The outstanding properties of BDDinterfaces have been suggested to be responsible for the highsensitivity. Indeed, comparable experiments performed on glassycarbon (GC) electrodes resulted in a detection limit of 3.5 ngmL�1.3 Lately, we and others have demonstrated that nano-structured BDD electrodes have considerable advantages over atBDD interfaces in terms of sensitivity and selectivity.17–21 Theycombine the unique features of smooth BDD interfaces such ashigh chemical stability, low background current, wide potentialwindow, decreased biofouling and high biocompatibility,22,23

with an increased surface area. We have also shown that thelength of the wires inuences the sensitivity of the interface.21

Motivated by this nding, we investigated here the potentialapplication of long BDD nanowires (BDD NWs) as substrates forimpedimetric immunosensors using immunoglobulin G (IgG) asa model analyte. IgG, a 150 kDa monomer constituting approxi-mately 75% of the total circulating immunoglobulin, plays acrucial role in the immune system and is the main base of anyimmunosensor used for diagnostics. In this study, anti-bovineIgG was used as a model antibody to be immobilized onto theBDD NWs. Optimal immobilization of antibody molecules

This journal is © The Royal Society of Chemistry 2014

Fig. 1 (A) Formation of a BDDNWbased immunosensor, (B) SEM images of the initial BDD film (left), BDDNWs, (middle) and BDDNWs decoratedwith Ni nanoparticles (right).

Paper Analyst

Publ

ishe

d on

14

Janu

ary

2014

. Dow

nloa

ded

by U

nive

rsity

of

Sout

hern

Cal

ifor

nia

on 0

2/04

/201

4 11

:24:

59.

View Article Online

depends on the surface employed as well as on the respectivesurface tethering approach. Nanoparticle modied BDD NWshave shown to be highly appealing for electroanalytical applica-tions as they show improved catalytic activity due to the highsurface area, which in turn necessitates the use of much lessmaterial.20,24 An additional advantage is the enhanced diffusionobserved at nanoparticles due to convergent rather than lineardiffusion at slow scan rates at small particles.25 In this context,BDD NWs coated with nickel nanoparticles (BDD-NWs/Ni NPs),deposited electrochemically, are investigated as immobilizationplatforms for biotin-labeled IgG (Fig. 1A).

2. Experimental2.1. Materials

Potassium hydroxide, nickel(II) sulfate heptahydrate(NiSO4$7H2O, 99.999%), phosphate buffer saline tablets (PBS, pH7.4), bovine serum albumin (BSA), ethylenediaminetetraaceticacid (EDTA), Fe(CN)6

4�/3� and sodium hydroxide (NaOH) werepurchased from Sigma-Aldrich and used without purication.

Biotin conjugated rabbit anti-bovine IgG (H&L) and rabbitbovine IgG were obtained from Agrisera Antibodies, Sweden.

2.2. Preparation of boron-doped diamond nanowires (BDDNWs) decorated with nickel nanoparticles (BDD NWs/Ni NPs)

Boron-doped diamond (BDD) lms were grown by microwaveplasma-enhanced chemical vapor deposition from methane–hydrogen mixtures (1% CH4) in an ASTeX 6500 reactor. Thesubstrates were p-type doped (100) silicon wafers (thickness50–500 mm, resistivity from 1 to 20 U cm). Trimethyl borane gas

This journal is © The Royal Society of Chemistry 2014

was added during the growth with a ratio of 10 000 ppm B/C toCH4 to ensure good electrical conductivity.26 The nal lmthickness was 15 mm and the boron concentration was NA ¼ 8 �1019 B cm�2 according to SIMS measurements.

Boron-doped diamond nanowires (BDD NWs) were preparedusing reactive ion etching (RIE) of BDD using an oxygen plasma(Plasmalab 80plus) with a radiofrequency (RF) generator (13.56MHz) for 180 min. The operating oxygen pressure, ow speed,and plasma power were 150 mT, 20 sccm, and 350 W, respec-tively. The resulting BDD NWs were immersed for 15 min in anaqueous solution of HF (5% v/v) to dissolve the SiO2 depositedon the wires during the etching process.

Nickel nanoparticles (Ni NPs) were deposited onto BDD NWsby reduction of an aqueous solution of NiSO4 (5 mM) in PBSbuffer (0.01 M; pH 6.5) at �1.3 V for 200 s under moderatestirring, at 0 V for 1 min followed by another pulse at �1.3 V for200 s. This cycle was repeated 10 times. Aer deposition, themodied electrode was immersed in NaOH (0.1 M) and cycled300 times between 0 and 0.5 V to enrich the Ni(OH)2 deposit.

2.3. Binding of biotin-tagged anti-IgG onto BDD NWs/Ni NPs

The BDD NW/Ni NP interface was immersed for 100 min intobiotin-labeled anti-IgG (0–100 mg mL�1) in PBS buffer (0.01 M;pH 7.4) and then thoroughly washed three times under soni-cation with PBS. To limit the non-specic interaction of IgGwith the surface, the BDDNW/Ni NP-anti IgGmodied interfacewas furthermore immersed for 100 min into an aqueous bovineserum albumin (BSA, 100 mg mL�1 PBS solution) and thereaerthoroughly washed three times under sonication with PBS.

Analyst, 2014, 139, 1726–1731 | 1727

Analyst Paper

Publ

ishe

d on

14

Janu

ary

2014

. Dow

nloa

ded

by U

nive

rsity

of

Sout

hern

Cal

ifor

nia

on 0

2/04

/201

4 11

:24:

59.

View Article Online

Aer interaction with IgG, the antibody–antigen complexwas dissociated by incubation of the interface in acidic glycinebuffer (pH ¼ 3) for 30 min under sonication followed bywashing (three times) with PBS.

Bonded anti-IgG could be washed off the BDD NW/Ni NPinterface by incubation in an aqueous solution of ethyl-enediaminetetraacetic acid (EDTA 0.01 M) for 60 min undersonication followed by washing (three times) with PBS.

When not in use the immunosensor was stored in PBS(0.01 M, pH 7.4) at 4 �C.

2.4. Characterization

2.4.1. Scanning electron microscopy (SEM). SEM imageswere obtained using an electron microscope ULTRA 55 (Zeiss)equipped with a thermal eld emission emitter and a highefficiency In-lens, and ESB and SE detectors.

2.4.2. X-ray photoelectron spectroscopy (XPS). X-rayphotoelectron spectroscopy (XPS) experiments were performedusing a PHI 5000 VersaProbe-Scanning ESCA Microprobe(ULVAC-PHI, Japan/USA) instrument at a base pressure below 5�10�9 mbar. Monochromatic AlKa radiation was used and theX-ray beam, focused onto a diameter of 100 mm,was scanned on a250 � 250 mm surface, at an operating power of 25 W (15 kV).Photoelectron survey spectra were acquired using a hemi-spherical analyzer at a pass energy of 117.4 eV with a 0.4 eVenergy step. Core-level spectra were acquired at a pass energy of23.5 eV with a 0.1 eV energy step. All spectra were acquired with90� between the X-ray source and the analyzer and with the use oflow energy electrons and low energy argon ions for chargeneutralization. Aer subtraction of the Shirley-type background,the core-level spectra were decomposed into their componentswith mixed Gaussian–Lorentzian (30 : 70) shape lines usingCasaXPS soware. Quantication calculations were conductedusing sensitivity factors supplied by PHI.

2.4.3. Cyclic voltammetry. Cyclic voltammetry (CV) experi-ments were performed using an Autolab 20 potentiostat (EcoChimie, Utrecht, The Netherlands). The electrochemical cellconsisted of a working electrode (BDD NWs/Ni NPs), Ag/AgCl(Bioanalytical Systems, Inc.) as a reference electrode, and plat-inum wire as a counter electrode. Cyclic voltammetricmeasurements were performed at a scan rate v ¼ 0.05 V s�1.

2.4.4. Electrochemical impedance spectroscopy (EIS).Electrochemical impedance measurements were performedusing an Autolab 20 potentiostat (Eco Chimie, Utrecht, TheNetherlands). EIS experiments were carried out in an aqueoussolution of a mixture of 10 mM Fe(CN)6

4�/10 mM Fe(CN)63� in

PBS (0.01M) using the following parameters: amplitude of 10 mVat an open circuit potential with a frequency range of 100 kHz–0.1Hz. Impedance data were modeled using ZView2 soware.

3. Results and discussion3.1. Preparation and characterization of diamond nanowiresdecorated with nickel nanoparticles (BDD NW/Ni NPs)

Fig. 1A shows the different steps involved in the construction ofthe immunosensor. Micrometer long BDD NWs were produced

1728 | Analyst, 2014, 139, 1726–1731

using a mask-less etching process of polycrystalline boron-doped diamond thin lms of 15 mm thickness by reactive ionetching with oxygen plasma as previously reported by us.21

Fig. 1B shows SEM images of the BDD interface before and aerRIE for 180 min. While the polycrystalline facets stay clearlydistinguishable, long needle-like shaped BDD NWs of 3 � 0.2mm length and a tip diameter of 30 � 20 nm were formed usingthis etching protocol. The formed BDDNWs were thenmodiedwith Ni NPs by immersion into an aqueous solution of NiSO4

(5 mM) in 0.01 M PBS (pH 6.5) and by application of electro-chemical pulses. A representative scanning electron microscopy(SEM) image of the resulting BDD NW/Ni NP interface aercathodic deposition of Ni NPs at �1.3 V/Ag/AgCl using 10potential pulses of 200 s followed by electrochemical cycling in0.1 M NaOH to enrich the b-Ni(OH)2 phase is shown in Fig. 1B.The particles are 20 � 5 nm in diameter with a particle densityof 150 Ni NPs per mm2 for 200 s deposition time. This issomewhat larger than previously reported shorter BDD NWs.24

The chemical composition of the interface was investigatedusing X-ray photoelectron spectroscopy (XPS). The Ni 2p XPSspectrum (Fig. 2A) shows two major peaks centered at 856.3 and873.8 eV with a spin-energy separation of 17.5 eV, correspondingto Ni 2p3/2 and Ni 2p1/2, respectively. These bands are charac-teristic of the Ni(OH)2 phase and in agreement with other liter-ature reports.27–29 The two extra peaks centered at 861.8 and879.2 eV are due to Ni 2p3/2 and Ni 2p1/2 satellite peaks. Theatomic percentage of nickel present on the BDDNWs is 15.4 at%.

The oxidation state of the deposited nickel was conrmed bythe presence of a quasi-reversible redox peak at z0.47 V/Ag/AgCl in the cyclic voltammogram of the BDD NW/Ni NP elec-trodes in PBS (Fig. 2B) assigned to Ni(OH)2/NiOOH. Thehydrated nickel hydroxide could be present in two crystallineforms: the hydrated a-Ni(OH)2 and the anhydrous b-Ni(OH)2,the latter being the more stable and preferentially formed bycycling in NaOH (0.1 M).30

3.2. Preparation of the electrochemical immunosensor

The BDDNW/Ni NP electrode was used for the construction of anelectrochemical immunosensor. The strategy employed here forthe immobilization of anti-IgG is based on the preferentialcoordination of biotinylated anti-IgG on the Ni NPs as suggestedpreviously by Cosnier and co-workers using poly(pyrrole-nitrilo-triacetic acid)-Cu2+ lms.31 The advantage of this approach wouldbe the possibility of controlled immobilization of biotinylatedantibodies and antigens with the omission of avidin layers,inuencing the electrochemical behaviour of the electricalinterface. The affinity of the biotin-tag to the Ni NPs is alsobelieved to be weaker than the avidin–biotin interaction, allowingeasy regeneration of the interface through a simple EDTA wash.

Electrochemical impedance spectroscopy was employed toprobe the change in the interfacial properties upon binding ofbiotin-labeled anti-IgG to BDD NWs/Ni NPs. As the dynamics ofcharge transfer at electrode interfaces are strongly inuenced bythe nature of the electrode surface, immobilization of biomole-cules is anticipated to alter the interfacial electron-transferfeatures. Fig. 3A shows the Nyquist plot for the BBD NWs/Ni NPs

This journal is © The Royal Society of Chemistry 2014

Fig. 2 (A) Ni 2p high resolution XPS spectrum of Ni–BDD NWs; (B) thecyclic voltammogram of Ni–BDD NWs in 0.1 M NaOH.

Paper Analyst

Publ

ishe

d on

14

Janu

ary

2014

. Dow

nloa

ded

by U

nive

rsity

of

Sout

hern

Cal

ifor

nia

on 0

2/04

/201

4 11

:24:

59.

View Article Online

before and aer incubation with biotinylated anti-IgG (100 mMmL�1) for 100 min using Fe(CN)6

4�/3� as the redox probe. It wasfound that the capacitive changes were not as sensitive as

Fig. 3 (A) Electrochemical impedance spectrameasured between 100kHz and 0.1 Hz on BDD NWs/Ni NPs before (black) and after incuba-tion in biotin-tagged anti-IgG (100 mg mL�1): solution: 10 mMFe(CN)6

4�/10 mM Fe(CN)63� in PBS (0.01 M); (B) change of the charge

transfer resistance with an increased amount of biotinylated anti-IgG.

This journal is © The Royal Society of Chemistry 2014

electron transfer resistance (Ret), presented by the real compo-nent Re(z) of the impedance. The change in Ret is dependent onthe amount of biotin-labeled anti-IgG immobilized on the surface(Fig. 3B). Incubation of the electrode in biotin-labeled anti-IgG(100 mg mL�1) results in an increase of the charge transferresistance from 3953 � 95 U (BDD NWs/Ni NPs) to 5541 � 78 U

(BDDNWs/Ni NPs/anti-IgG) and thus amaximal change ofDRet¼1588 � 65 U (Fig. 3B). To avoid the non-specic interaction, theinterface was furthermore incubated in a BSA solution (100mg mL�1) for 100 min. Incubation in BSA resulted in a furtherincrease of charge transfer resistance by 150 U.

In a control experiment, BBD NW electrodes without Ni NPswere incubated in an identical way in biotin-tagged anti IgG(100 mg mL�1) solution and then washed several times with PBSunder ultrasonication. The recorded DRet was only 328 � 40 U,highlighting the importance of Ni NPs for a strong binding ofbiotin-labeled anti IgG.

3.3. Electrochemical detection of IgG binding by BDD NWs/Ni NPs/anti-IgG

The functional immunosensor was dipped into 0.01 M PBS(pH 7.4) containing various concentrations of IgG at roomtemperature for 40 min and then rinsed with PBS to remove anyunbound analyte. The immune-reaction between surface linkedanti-IgG and IgG in solution was monitored by EIS in the pres-ence of Fe(CN)6

4�/3� (0.01 M, PBS, pH ¼ 7.4) as the redox probe.The change in the charge transfer resistance of IgG recognition inthe range of 0–100 ng mL�1 IgG is shown in Fig. 4. A linearrelation between DRct and IgG concentration in the range of0.3–400 ng mL�1 was recorded with a correlation coefficient ofr ¼ 0.9996 according to DRet (kU) ¼ 0.02 + 0.0451 � [IgG]. Thedetection limit of IgG was determined to bez 0.3 ng mL�1 fromve blank noise signals (95% condent level). This detectionlimit is comparable to poly(o-aminobenzoic acid) modied

Fig. 4 Calibration plot for IgG concentrations on BDD NW/Ni NPinterfaces modified with 100 mg mL�1 biotin-labelled anti-IgG (0–600ng mL�1).

Analyst, 2014, 139, 1726–1731 | 1729

Table 1 Analytical performance of other electrochemical based immunosensors: poly-o-ABB: poly(o-aminobenzoic acid); poly(B.N.P.):copolymer carrying hydrophobic benzene ring, N-acryloxysuccinimide and poly(ethylene glycol) methacrylate, units; ALP: alkaline phosphatase,HRP: horseradish peroxidase; ANTA: N-a-bis(carboxymethyl)-L-lysine

Electrode Label Detection method LOD [ng mL�1] Ref.

Au/thioctic acid Biotin labeled protein–streptavidinnetwork

EIS 2000 15

Si/SiO2/Al2O3/Fe/carbonnanotubes

— EIS 200 11

ITO HRP–silica particles Amperometry 5 33Au/Au NPs-antibody HRP Amperometry 2 13BDD/poly-o-ABB ALP Amperometry 0.3 3ITO/rGO/poly(B.P.N.) HRP Amperometric 0.0001 4Polypyrrole/ANTA/Cu2+ — Differential pulse voltammetry 0.1 32BDD NWs/Ni NPs — EIS 0.3 This

work

Analyst Paper

Publ

ishe

d on

14

Janu

ary

2014

. Dow

nloa

ded

by U

nive

rsity

of

Sout

hern

Cal

ifor

nia

on 0

2/04

/201

4 11

:24:

59.

View Article Online

planar BDD interfaces using alkaline phosphate as the label withthe main advantage of a complete label-free detection in the caseof BDD NWs/Ni NPs (Table 1).3 The detection limit of BDD NWs/Ni NPs is lower than that reported for carbon nanotube array,11

and gold-nanoparticle based sandwiched electrochemicalimmunosensors.13 It is, however, less sensitive than the sensorrecently proposed by Kim and co-workers based on electro-chemically reduced graphene oxide modied with an N-acrylox-ysuccinimide-activated amphiphilic polymer (Table 1)4 andfunctional polypyrrole modied interfaces.32 The reproducibilityof the electrodes is expressed in terms of the relative standarddeviation which is found to be 5.3% at an IgG concentration of100 mg mL�1. The stability of the immunosensor was examinedaer 15 days storage in a refrigerator at 4 �C. The sensor retainedabout 97% of its initial sensitivity for 100 ng mL�1 IgG detection,indicating that the electrode has good stability.

The BDD NW/Ni NP interface could be regenerated in twodifferent ways. Polyclonal IgG populations will typically have Kd

values on the order of 10�10 to 10�7 M and can be disrupted upontreatment with 0.1MNaOH. The Ni NP–biotin interaction shouldbe weaker than biotin–streptavidin interactions (Kd ¼ 10�15 M).The anti-IgG could thus be detached from the BDD NWs/Ni NPsby incubation in an aqueous solution of ethylenediaminetetra-acetic acid (EDTA, 0.01 M) for 60 min and the BDD NWs/Ni NPscould be used for linking another antibody if desired.

4. Conclusions

The development of an immunosensor based on a boron-dopeddiamond nanowire electrode was demonstrated. The immobi-lization of biotin-labeled bovine anti-IgG was accomplishedthrough coordination of the biotin group to nickel nano-particles, electrochemically deposited onto the BDD NW elec-trode. This immobilization was found to be specic towards theNi NPs and proved to be as efficient as the conventional avidin–biotin system. The sensitivity toward the label-free detection ofIgG is notably comparable to poly(o-aminobenzoic acid) modi-ed planar BDD interfaces using alkaline phosphatase as thelabel. The immunosensor could be easily regenerated by

1730 | Analyst, 2014, 139, 1726–1731

incubation in a basic solution which results in the dissociationof the antigen–antibody complex. In addition, incubation in anaqueous solution of EDTA detached the biotin-labeled antibodyfrom the Ni NPs and recovered the initial BDD NW/Ni NPinterface. This will allow, in future, the immobilization ofdifferent anti-bodies using the same interface.

Acknowledgements

R.B, Y.C. and S.S. gratefully acknowledge the nancial supportfrom the Centre National de Recherche Scientique (CNRS), theUniversite Lille 1 and the Nord Pas de Calais region. Supportfrom the European Union through a FP7-PEOPLE-IRSES (PHO-TORELEASE) and FP7-PEOPLE-ITN (MATCON) are acknowl-edged. Many thanks to Mr Sobczak in obtaining the XPS data.

References

1 C. J. van Oss, Mol. Immunol., 1995, 21, 199.2 R. M. Lequin, Clin. Chem., 2005, 51, 2415.3 A. Preechaworapun, T. A. Ivandini, A. Suzuki, A. Fujishima,O. Chailapakul and Y. Einaga, Anal. Chem., 2008, 80, 2077.

4 A.-M. J. Haque, H. Park, D. Sung, S. Jon, S.-Y. Choi andK. Kim, Anal. Chem., 2012, 84, 1871.

5 F. Ricci, G. Volpe, L. Micheli and G. Palleschi, Anal. Chim.Acta, 2007, 605, 111.

6 C. A. Marquette and L. J. Blum, Biosens. Bioelectron., 2006, 21,1424.

7 F. Ricci, G. Adornetto and G. Palleschi, Electrochim. Acta,2012, 84, 74.

8 A. de la Escosura-Muniz, C. Parolo and A. Merkoci, Mater.Today, 2010, 13, 24.

9 J. F. Rusling, Chem. Rec., 2012, 12, 164.10 M. M. Billah, C. S. Hodges, H. C. W. Hays and P. A. Millner,

Bioelectrochemistry, 2012, 80, 49.11 Y. Yun, A. Bange, W. R. Heinemann, H. B. Halsall,

V. N. Shanov, Z. Dong, S. Pixley, M. Behbehani, A. Jazieh,Y. Tu, D. K. Y. Wong, A. Bhattacharya and M. J. Schulz,Sens. Actuators, B, 2007, 123, 177.

This journal is © The Royal Society of Chemistry 2014

Paper Analyst

Publ

ishe

d on

14

Janu

ary

2014

. Dow

nloa

ded

by U

nive

rsity

of

Sout

hern

Cal

ifor

nia

on 0

2/04

/201

4 11

:24:

59.

View Article Online

12 W. Yang, J. E. Butler, J. N. Russell and R. J. Hamers, Analyst,2007, 132, 296.

13 G. K. Ahirwal and C. K. Mitra, Biosens. Bioelectron., 2010, 25,2016.

14 Y. Xiao, C. M. Li and Y. Liu, Biosens. Bioelectron., 2007, 22,3161.

15 R. Pei, Z. Cheng, E. Wang and X. Yang, Biosens. Bioelectron.,2001, 16, 355.

16 I. Hafaid, S. Chebil, H. Korri-Youssou, F. Bessueille,A. Errachid, Z. Sasi, Z. Ali, A. Abdelghani and N. Jaffrezic-Renault, Sens. Actuators, B, 2010, 144, 323.

17 S. Szunerits, Y. Coffinier, E. Galopin, J. Brenner andR. Boukherroub, Electrochem. Commun., 2010, 12, 438.

18 D. Luo, L. Wu and J. Zhi, ACS Nano, 2009, 3, 2121.19 N. Yang, H. Uetsuka and C. E. Nebel, Adv. Funct. Mater.,

2009, 19, 887.20 N. Yang, W. Smirnov and C. E. Nebel, Electrochem. Commun.,

2013, 27, 89.21 Q. Wang, P. Subramanian, M. Li, W. S. Yeap, K. Haenen,

Y. Coffinier, R. Boukherroub and S. Szunerits, Electrochem.Commun., 2013, 34, 286.

22 S. Szunerits and R. Boukherroub, J. Solid State Electrochem.,2008, 12, 1205.

This journal is © The Royal Society of Chemistry 2014

23 McCreey, Chem. Rev., 2008, 108, 2646.24 P. Subramanian, J. Foord, D. Steinmuller, Y. Coffinier,

R. Boukherroub and S. Szunerits, Electrochim. Acta, 2013,110, 4.

25 S. E. W. Jones and R. G. Compton, Anal. Chem., 2008, 4, 177.26 S. D. Janssens, P. Pobedinskas, J. Vacik, V. Petrakova,

B. Ruttens, J. D'Haen, M. Nesladek, K. Haenen andP. Wagner, New J. Phys., 2011, 13/8, 083008.

27 M. W. Roberts and R. S. C. Smart, J. Chem. Soc., FaradayTrans., 1984, 80, 2957.

28 A. F. Carley, S. D. Jackson, J. N. O'Shea and M. W. Roberts,Surf. Sci., 1999, 440, L868.

29 A. P. Grosvenor, M. C. Biseinger, R. S. C. Smart andN. S. McIntyre, Surf. Sci., 2006, 600, 1771.

30 K. E. Toghill, L. Xiao, N. R. Stradiotto and R. G. Compton,Electroanalysis, 2010, 22, 491.

31 J. Baur, M. Holzinger, C. Gondran and S. Cosnier,Electrochem. Commun., 2010, 12, 1287.

32 S. Chebil, I. Hafaiedh, H. Sauriat-Dorizon, N. Jaffrezic-Renault, A. Errachid, A. Ali and H. Korri-Youssou,Biosens. Bioelectron., 2010, 26, 736.

33 Z. Y. Zhong, M. X. Li, D. B. Xiang, N. Dai, Y. qing, D. Wangand D. P. Tang, Biosens. Bioelectron., 2009, 24, 2246.

Analyst, 2014, 139, 1726–1731 | 1731