Pe

Sa

b

h

•

•

•

•

a

ARRAA

KPSIBC

0h

Analytica Chimica Acta 795 (2013) 8– 14

Contents lists available at ScienceDirect

Analytica Chimica Acta

jou rn al h om epage: www.elsev ier .com/ locate /aca

olyaniline-iron oxide nanohybrid film as multi-functional label-freelectrochemical and biomagnetic sensor for catechol

udeshna Chandraa, Heinrich Langb, Dhirendra Bahadura,∗

Department of Metallurgical Engineering & Materials Science, Indian Institute of Technology Bombay, Powai, Mumbai 400076, Maharashtra, IndiaLehrstuhl für Anorganische Chemie, Institut für Chemie, Technische Universität Chemnitz, Straße der Nationen 62, D-09111 Chemnitz, Germany

i g h l i g h t s

Synthesis of mesoporouspolyaniline-iron oxide magnetic(Fe3O4@PANI) nanohybrid with highsurface area of 94 m2 g−1.Ligand-to-metal charge transfer isseen between p� orbitals of catecholand d�* metal orbital of nanohybrid.Highly sensitive response(312 �A �L−1) toward catecholwith low detection limit (0.2 nM)was obtained by amperometry.Decrease in peak frequency ofAC susceptibility shows that theFe3O4@PANI nanohybrid be used asbiomagnetic sensor for catechol.

g r a p h i c a l a b s t r a c t

r t i c l e i n f o

rticle history:eceived 8 April 2013eceived in revised form 8 July 2013ccepted 16 July 2013vailable online 31 July 2013

eywords:ANI-iron oxide nanohybridpectroelectrochemistry

a b s t r a c t

Polyaniline-iron oxide magnetic nanohybrid was synthesized and characterized using various spectro-scopic, microstructural and electrochemical techniques. The smart integration of Fe3O4 nanoparticleswithin the polyaniline (PANI) matrix yielded a mesoporous nanohybrid (Fe3O4@PANI) with high surfacearea (94 m2 g−1) and average pore width of 12.8 nm. Catechol is quasi-reversibly oxidized to o-quinoneand reduced at the Fe3O4@PANI modified electrodes. The amperometric current response toward catecholwas evaluated using the nanohybrid and the sensitivity and detection limit were found to be 312 �A �L−1

and 0.2 nM, respectively. The results from electrochemical impedance spectroscopy (EIS) indicated thatthe increased solution resistance (Rs) was due to elevated adsorption of catechol on the modified elec-

mpedanceiomagneticatechol

trodes. Photoluminescence spectra showed ligand-to-metal charge transfer (LMCT) between p-� orbitalsof the phenolate oxygen in catechol and the d-�* metal orbital of Fe3O4@PANI nanohybrid. Potentialdependent spectroelectrochemical behavior of Fe3O4@PANI nanohybrid toward catechol was studiedusing UV/vis/NIR spectroscopy. The binding activity of the biomagnetic particles to catechol throughBrownian relaxation was evident from AC susceptibility measurements. The proposed sensor was usedfor successful recovery of catechol in tap water samples.

∗ Corresponding author. Tel.: +91 22 25767632.E-mail address: dhirenb@iitb.ac.in (D. Bahadur).

003-2670/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.aca.2013.07.043

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Catecholamines are biogenic sympathomimetic hormones pro-

duced by adrenal glands in response to stress which play animportant role in the central nervous system. Catechols are typicalneurotransmitters that affect the regulation of blood pressure andmetabolic processes, and are used as an index for several diseases,

a Chim

ssAcddricst

temaosta

ossobuTcbchgs

wncfbaltnpcattrot

(sawboeplcPnt

S. Chandra et al. / Analytic

uch as hypertension, phaeochromocytoma, neuroblastoma septichock, Parkinson’s disease and pharmacological vasodilatation [1].ny abnormality in the concentration of catecholamines in biologi-al fluids, such as urine, plasma, or serum sends a warning signal ofiseases. The pronounced significance of catecholamines in clinicaliagnosis and medical treatment therefore, makes its quantitative,apid and sensitive determination important. Though there are var-ous chromatographic and spectroscopic methods for determiningatecholamines, unfortunately their high cost, low sensitivity andelectivity, requirement of extraction and derivation steps makehe system less operative [2].

We have sought to address this problem through spectroelec-rochemical, impedimetric and biomagnetic sensor based on anlectroactive porous nanohybrid with the combined benefits ofagnetic iron oxide [3], capable of oxidizing the catechol, and

n electrically conducting polyaniline. Electrochemical oxidationf catechol in the mixed organic solvents is well reported whichhowed one anodic and a cathodic peak that correspond to theransformation of catechol to benzoquinone and vice versa within

quasi-reversible two-electron process [4].Porous materials with high surface area have a wide range

f applications in adsorption and separation processes, catalysis,ensors and electrodes, biotechnology, etc. Commonly repre-ented porous materials are carbon, glass, silicates (zeolites),xides, metals, and polymers. Usually, nanoporous materials cane obtained by direct copolymerization of crosslinking monomerssing porogenic solvents or by the hypercrosslinking method.he latter involves dissolution of non-crosslinked or marginallyrosslinked polymers followed by immobilization of the poresy a secondary material. Germain et al. [5] followed the hyper-rosslinking approach for preparing nanoporous PANI with highydrogen storage capacity. So far the synthetic approach involvedrafting of PANI on porous materials so as to produce porous micro-pheres using interfacial polymerization.

Of late, efforts have been put to synthesize porous nanohybridsith an aim to synergistically combine the merits of each compo-ent in the hybrid material [6,7]. Smart intercalation of respectiveomponents can form nanohybrids with much more enhancedunctions and capability. The expected advancement in the nanohy-rids depends on the polymer-nanoparticle interfacial chemicalnd electronic interactions arising due to polymer wrapping, cova-ent/noncovalent interaction of molecules, etc. This may give riseo different surface chemistries and electronic structures to theanohybrids [8]. It is indeed a great challenge to develop optimalorous nanohybrids with a high specific surface area favorable foratalytic sensing and adsorption. Recently, Chen et al. [6] designed

Pt/C@PANI core–shell-structured catalyst for fuel cell in whichhe carbon support is decorated by a layer of PANI and Pt nanopar-icles are anchored to the carbon support. Qiao and co-workers [7]eported a mesoporous nanostructured PANI/TiO2 composite withptimized biocatalytic and electrocatalytic properties for applica-ion in microbial fuel cell.

We herein report tailoring of a porous nanohybridFe3O4@PANI) fabricated from iron oxide nanoparticles andoluble PANI, wherein the Fe3O4 nanoparticles are looked upons biocompatible, stable, and environmental friendly nanoparticleith attractive electronic and magnetic properties which has

een recently explored for biosensing application [9]. On thether hand, PANI with a �-conjugated structure provides highlectrical and proton conductivity in acidic environments andossesses unique redox properties. However, its application is

imited by its intractable nature of being insoluble if synthesized

hemically; and brittle and thin if obtained electrochemically [10].roposed biosensing technology in this work exploits Fe3O4@PANIanohybrids and their optical, magnetic, electrochemical and spec-roelectrochemical properties for enhancing sensing efficiency.

ica Acta 795 (2013) 8– 14 9

Morphological characterization of the nanohybrid plays animportant role since their properties affects the performance ofthe sensor. Besides size and structure, other parameters whichinfluence the properties of hybrid material are surface charge,porosity, type and density of reactive surface groups. The stabilityof the nanomaterials under varied temperature and pH conditionsis correctly oriented by choosing blocking agents and carryingout post-immobilization modification. In the present case, theelectroactive PANI has been used as matrix materials in in situsynthesis of Fe3O4 nanoparticles which not only provided stabilityto the nanohybrid but also provided efficient electron transfer forcatechol oxidation. Unlike conventional co-precipitation method,the proposed modified method ensures control over porousnature and matrix-natured nanostructures with enhanced elec-tronic properties. The Fe3O4@PANI nanohybrids were found to beextremely suitable mesoporous matrices for catechol sensing dueto increased internal porosity, high charge densities, super elec-trochemical behavior and rapid charge transfer kinetics. Anotherinnovative aspect of the proposed technology is the biomagneticsensing using the nanostructures wherein Fe3O4@PANI can beused for detecting the presence of small amounts of catechol. Theresulting nanohybrid has been studied in detail with respect to itselectronic, magnetic and optical properties, surface morphology,porosity, and applicability in electrochemical, impedimetric andbiomagnetic sensing of catechols (CA).

2. Materials and methods

2.1. Materials and equipment

Ferric chloride hexahydrate (FeCl3·6H2O), and ferrous chloridetetrahydrate (FeCl2·4H2O) were purchased from Sigma AldrichChemical Co. All other chemicals were of analytical grade and usedas received. The phase purity and identification of the magneticnanoparticles were done by X-ray diffraction (XRD), and the pat-tern of each sample was recorded with a PanAnalytical X-Pertdiffractometer using a monochromatized X-ray beam with nickel-filtered Cu-K� radiation with 4◦ min−1 scan rate. A continuous scanmode was used to collect 2� data from 5◦ to 70◦. Fourier transforminfrared (FT-IR) spectra were obtained using a Jasco, FT-IR 300Espectrometer with a resolution of 4 cm−1. Scanning electron micro-scope (SEM) with Energy Dispersive Spectroscope (EDS) S-3400NHitachi was used to image PANI and Fe3O4@PANI nanohybrids atan acceleration voltage of 3 kV. Transmission electron microscopyalong with Electron energy loss spectroscopy (EELS) was used toestablish the distribution of iron oxide nanoparticles on PANI. EELSwere performed with a TEM 200 kV that incorporates an elec-tron beam JEOL JEM 2010 F and an energy loss spectrometer. Thespecific surface area, pore volume and pore size distribution ofthe hybrids were measured by ASAP 2020 Micromeritics instru-ment. Specific surface areas were determined by the multipointBrunauer–Emmet–Teller (BET) method. The corresponding poresize distribution and total pore volume were determined by theBrunauer–Joyner–Hallenda (BJH) method applied to the desorp-tion branch. Prior to measurements, the samples were outgassedat 40 ◦C with a heating rate of 10 ◦C min−1 for 1 h and then thetemperature was raised up to 50 ◦C and maintained overnight. Pho-toluminescence of all the samples was studied by Cary EclipseFluorescence Spectrometer in the range of 300–700 nm.

AC magnetic susceptibility under different frequencies, and thedependence of magnetization with the applied magnetic field were

measured by using a Physical Properties Measurement System(PPMS from Quantum Design).

The electrochemical measurements (Cyclic Voltammetry andAmperometry) were conducted in a 3-electrode single-cell

1 Chim

sGAetcshoib

2

itFstnisebTmwahrscsn

2

oedtimaagacv(eA

2

bwetaum

0 S. Chandra et al. / Analytica

ystem in phosphate buffer along with 0.1 M KCl as electrolyte.lassy carbon electrode (GCE, diameter ϕ = 2 mm), Pt-wire andg/AgCl electrodes were used as working, counter and referencelectrodes, respectively with CHI1140A electrochemical worksta-ion (CHI110, Austin, TX). All electrochemical measurements werearried out at room temperature. To eliminate the effect of dis-olved oxygen, the electrolyte was purged with nitrogen gas foralf an hour. The Electrochemical impedance measurements (EIS)f the fabricated electrodes were measured by the (Gamry EIS 300)nstrument in a single compartment 3-electrode cell and analyzedy Echem Analyst, Gamry software.

.2. Synthesis of Fe3O4@PANI nanohybrid

The most common method to prepare Fe3O4@PANI nanohybrids chemical/electrochemical oxidation polymerization, wherein,he aniline monomer is polymerized in situ in the presence ofe3O4 nanoparticles and surfactant [11] or by the in situ emul-ion polymerization/reverse micelle method [12]. However, forhese methods challenges like preventing aggregation of the Fe3O4anoparticles, solubility of the nanohybrids, and processability

ssue which arise due to polymerization had to be managed. Weought to address the problem by developing a simple yet highlyfficient and easy method for preparing the Fe3O4@PANI nanohy-rid which retained the unique behavior of both the components.he nanohybrid were synthesized by a modified co-precipitationethod in which FeCl2·4H2O (0.86 g) and FeCl3·6H2O (2.35 g)ere dissolved in toluene–isopropanol (2:1 mixture) containing

n already synthesized PANI [10,13]. As the reaction solution waseated to 80 ◦C, 2 mL hydrazine hydrate was added slowly and theeaction was allowed to proceed for 1 h with constant and vigoroustirring to produce a black suspension. The reaction mixture wasooled to ambient temperature, and purified by repeated washingequentially with ethanol and acetone using a permanent mag-et.

.3. Fabrication of Fe3O4@PANI electrode

Glassy carbon working electrodes were cleaned by polishingn aqueous slurries of 0.3 �m and 0.05 �m alumina powder. Thelectrode was then rinsed with deionized water and was kept inilute sulphuric acid for 5 min followed by repeated ultrasonica-ion in methanol and water for 20 min each. After final sonicationn ethanol, the electrode was left to dry in ambient air for a few

inutes. Fe3O4@PANI nanohybrids were dispersed in Milli-Q waternd drop-casted onto the surface of the working electrode andir-dried to obtain a thin layer of coating. The electrode was thenently washed with deionized water to remove unbound particlesnd used fresh in electrochemical experiments. The electrochemi-al behavior of the modified electrode was investigated by cyclicoltammetry (CV) and electrochemical impedance spectroscopyEIS). Amperometric i − t measurements were used to detect cat-chol quantitatively at a fixed operating potential of 800 mV vs.g/AgCl in 0.1 M PBS/0.1 M KCl electrolyte.

.4. Real sample analysis

Modified Fe3O4@PANI electrodes were tested for their applica-ility for analysis of catechol in tap water samples. The samplesere spiked with known concentration of catechol and recovery

xperiments were performed by measuring the CV responses to

he samples in which the known concentration of catechol weredded. The results were also compared with high performance liq-id chromatography (HPLC) using water:methanol (ratio 90:10)obile phase with C18 column and UV detector.

ica Acta 795 (2013) 8– 14

3. Results and discussion

3.1. Structural analysis of the nanohybrid

The X-ray diffraction pattern of Fe3O4@PANI nanohybrid exhib-ited well resolved peaks of crystalline iron oxide (Figure S1 in SI).The crystallite size as calculated from the most intense peak wasfound to be 16.7 nm. The FTIR spectrum of Fe3O4@PANI nanohy-brid exhibited similar IR bands as that of PANI (Figure S2 in SI). Thedecrease in the intensity of the NH stretching bands (1170 cm−1)in Fe3O4@PANI nanohybrid is attributed to NH· · ·Fe3O4 inter-actions [14]. The strong band at 574 cm−1 is assigned to Fe Ostretching vibrations [15].

Surface morphology of PANI and Fe3O4@PANI nanohybrids wasdetermined by Scanning electron microscopy (SEM) which showsporous and irregular PANI (Figure S3a). Images of Fe3O4@PANInanohybrids (Figure S3b) show the presence of Fe3O4 nanoparticlesas white balls and patches on the PANI matrix. The energy disper-sive X-ray (EDX) shows that the white spots (marked as # in FigureS3b) are Fe3O4 with weight % as C (19.02), N (7.64), O (27.30) andFe (44.68). In contrast to the above, these white spots are absentin PANI (Figure S3a) and the EDX gives an weight % as C (56.64), N(31.67), and O (11.58).

Transmission electron microscopy (TEM) along with electronenergy loss spectroscopy (EELS) was used to establish the dis-tribution of iron oxide nanoparticles on PANI which showedmonodispersed Fe3O4 nanoparticles within the PANI matrix (FigureS4). EELS were performed with a TEM 200 kV that incorporates anelectron beam JEOL JEM 2010 F and an energy loss spectrometer. Inthe EELS spectrum, K-edges of C, N and O can be clearly identified.The sharply defined �* and �* of the C K and N K-edges are char-acteristic of sp2 bonding of PANI network. The O-K edges of Fe3O4showed two peaks: one at 533 eV derived from O 1s to 2p core levelhybridized with Fe 3d orbital and the other at 548 eV which origi-nates from O 2p states hybridized with the Fe 4s and 4p states. Thepeaks at 562 and 568 eV are due to the scattering of the third andthe first oxygen coordination shells by outgoing or backscatteringelectrons [16,17]. The EELS analysis also confirms the presence ofthe Fe L2 and L3 edge signals at 715 and 727 eV which correspond toexcitations from the spin-orbit level transitions 2p3/2 → 3d3/2 3d5/2

and 2p1/2 → 3d3/2, respectively. Thus, from the above observationsof TEM-EELS, it may be inferred that the Fe3O4 particles interactedwith the PANI matrices through non-covalent bonds, such as �–�stacking and hydrophobic wrappings [18,19].

The BET surface area of the Fe3O4@PANI nanohybrids was cal-culated based on adsorption data in the partial pressure (P/Po)range 0.01–0.99. The N2 adsorption–desorption isotherm has ahysteretic loop in 0.7–0.9 range of relative pressure, which is a char-acteristic of the adsorption–desorption of a mesoporous material(Fig. 1). The hysteresis loop remains unsaturated even at high rel-ative pressure (H3 type) which suggests slit shaped pores (as alsoevidenced from the hump in the pore distribution) that may bedue to the deposition of Fe3O4 nanoparticles on the PANI-matrix.The external and the BET surface area of the nanohybrid wereexperimentally found out to be 133 and 94 m2 g−1, respectivelywith the adsorption average pore width of 12.8 nm. The high sur-face area and the narrow pore size distribution could render thenanohybrid very useful for adsorption of any analyte. It can alsobe anticipated that the hydrogen acceptors (NH ) of the PANImay act as stabilizers for the Fe3O4 nanoparticles and can pro-vide a bed matrix for its deposition. As expected, the wealth ofa simple synthetic protocol of mesoporous nanohybrids with high

surface area might lead to many prospective studies on biosens-ing, which prompted us to examine its sensing capability towardcatechol, a neurotransmitter and a prognostic marker for severaldiseases.

S. Chandra et al. / Analytica Chimica Acta 795 (2013) 8– 14 11

Fo

ownt4bwPatbTcepmTait[

Fr

Fig. 3. Cyclic voltammograms of modified electrodes in the presence of 10−6 M cate-

ig. 1. Nitrogen adsorption/desorption isotherm and pore size distribution (inset)f the Fe3O4@PANI nanohybrids.

Photoluminescence (PL) spectroscopy was used to study theptical and electronic interaction of Fe3O4@PANI nanohybridsith catechol and for elucidation of the probable mecha-ism of the sensing system (Fig. 2). The spectrum of PANI inoluene–isopropanol (2:1 mixture) shows emission peaks at 322,09 and 426 nm and a band around 450–500 nm which are causedy the reduced benzenoid group and localized exciton associatedith the quinoid ring, respectively [20]. The room temperature

L spectrum of Fe3O4@PANI nanohybrids shows emission bandst ∼322 and 426 nm corresponding with that of PANI, however,he emission intensity is higher than that of PANI which is proba-ly due to the introduction of Fe3O4 nanoparticles in the hybrid.he Fe3O4 nanoparticles increase the density of the carrier andonfine the non-radiative decay of PANI thereby improving themission efficiency of material [21]. The band at 409 nm disap-ears and new bands appear at ∼360, 502, and 531 nm, whichay be due to the interfacial states of the Fe3O4 and PANI [22].

here is also a significant increase in the intensity of the weak bandt 502 nm due to increased �-electron mobility in the compos-

te favoring formation of singlet excitons which decay radiativelyo the ground state resulting in an enhanced photoluminescence23]. Interestingly, the PL intensity is enhanced on addition of

300 35 0 40 0 45 0 50 0 55 0 60 0 65 0 70 0

0

5

10

15

20

25

Inte

ns

ity

/-

Wave l eng th (nm)

Fe3O

4@PANI

PANI

Fe3O

4@PANI-Catechol

32

2

36

0

42

6

50

2

53

1

40

9

ig. 2. Photoluminescence spectra of PANI, Fe3O4@PANI nanohybrids and itsesponse toward catechol.

chol; scan rate 50 mV s−1, inset shows calibration curve obtained from amperometrywith varying [catechol].

catechol to the solution of Fe3O4@PANI nanohybrids, which maybe due to the combined effect of ligand-to-metal charge transfer(LMCT) between the p� orbitals of the phenolate oxygen of thecatechol and the d�* metal orbital of the Fe3O4@PANI nanohybridsand reduction of non-radiative pathways.

3.2. Electrochemical performance of the nanohybrids

Cyclic voltammograms (CVs) of PANI and Fe3O4@PANI nanohy-brids in 0.1 M PBS/0.1 M KCl are given in Figure S5. CVs were thenrecorded to examine the electrochemical response of neat PANIand Fe3O4@PANI nanohybrids toward catechol in 0.1 M PBS/0.1 MKCl at a scan rate of 50 mV s−1. Buffer mixtures are of particularsignificance with respect to chemical processes in physiology, andsince we propose to use the developed biosensor for detection ofcatechol in living systems, we used PBS as electrolyte. In our pro-posed application, buffer system is adjusted to a physiological pHof 7.4, which means while, on one hand, PBS can act as an elec-trolyte, and on the other hand a definite pH is maintained by itwhich is of particular importance in living system. Unlike bare GCE,on all the modified electrodes, the oxidation and the reductionpeak of catechol appeared which indicates that the catechol canbe oxidized to o-quinone and reduced at the electrodes (Fig. 3).However, there is a remarkable increase in the peak oxidation cur-rent of catechol using Fe3O4@PANI nanohybrid electrodes whichshows that Fe3O4@PANI nanohybrid have a good electrocatalyticactivity toward analyte detection. Catechol exhibited one-electronredox couple at 0.23 and 0.5 V and an irreversible oxidation peakat 0.78 V, with a peak-to-peak separation potential (�Ep) of 60 mV.The first redox peak corresponds to the quasi-reversible oxidationof catechol to semiquinone radical, while the second reductionpeak is attributed to the reduction of the semiquinone radicalfollowed by proton abstraction to give catechol. Probably, Fe3O4helps in adsorbing and facilitating catechol to reach the electrodesurface and PANI enhances the electron transfer and acceleratesthe reaction rate. Thus, the positive shift in oxidation potential isattributed to the binding of catechol with the nanohybrids whichfavors the reaction at lower potential. It is quite possible that dur-ing oxidation of catechol, electron transfer takes place through the

bonds between PANI–NH· · ·Fe3O4 and the porous structure of thenanohybrids which further augments the interface between them,thus resulting in an improved electrocatalytic performance [24]. Itmay be assumed that the external surface area of the nanohybrids

12 S. Chandra et al. / Analytica Chimica Acta 795 (2013) 8– 14

-0.2 0.0 0.2 0.4 0.6 0.8 1.0

-3.0 x10-5

0.0

3.0 x10-5

6.0 x10-5

9.0 x10-5

1.2 x10-4

1.5 x10-4

Cu

rre

nt

(A)

Potential (V)

Catecho l

asc orbic acid

dopa mine

Fig. 4. Cyclic voltammograms of 1.0 × 10−6 M catechol in the presence of1ro

as

Ft5ttmcp5Iarieqoie

3

arortmpicd(ctectosd

.0 × 10−4 M dopamine and 1.0 × 10−4 M ascorbic acid. The voltammograms rep-esents selective determination of catechol in the presence of higher amounts ofther interferents.

fforded by the pores increases the active site on the electrodeurface thereby playing a vital role in the catalytic reaction.

Cyclic voltammetry for oxidation–reduction of catechol ate3O4@PANI nanohybrid modified GCE was also carried out inhe potential range −0.1 to 1.0 V vs Ag/AgCl at scan rates of–200 mV s−1 (Figure S6 in SI). With increase in the scan rate,he anodic (Ipa) and cathodic (Ipc) peak currents increased. Also,he Ipa moved to more positive potential while the Ipc shifted to

ore negative values. The separation of the peak potential of cate-hol increased with increasing scan rates. The peak currents wereroportional to the square root of scan rate (�) in the range of–200 mV s−1 with R2 value of 0.99591 and 0.99709 for Ipa and

pc, respectively, suggesting an adsorption-controlled process. It islso seen that proportional to the augmentation of potential sweepate, there is an increase in the height of cathodic peak indicat-ng that the overall mechanism involved a chemical step after thelectron transfer [25]. An EC mechanism is proposed in which theuasi-reversible chemical and electrochemical reactions take placen the modified electrodes in the presence of catechol. The decreasen current ratio (Ipa/Ipc) with increase in scan rate indicates goodlectroactivity of the material toward catechol.

.3. Interference studies of dopamine (DA) and ascorbic acid (AA)

Both AA and DA coexist with catechol in physiological fluidsnd possess oxidation potentials close to each other at electrode,esulting in the electrochemical response of catechol being almostverlapped by AA or DA. In this study, it was found that wellesolved redox peak of catechol in the presence of higher concen-ration of DA and AA could be obtained by using the Fe3O4@PANI

odified GCE. In order to confirm the sensing of catechol in theresence of AA and DA, we scanned a mixture solution contain-

ng 1.0 × 10−6 M catechol, 1.0 × 10−4 M AA, and 5.0 × 10−4 M DA byyclic voltammetry. As is shown in Fig. 4, the cyclic voltammogramsemonstrated that the coexistence of AA and DA with catecholpH 7.0; PBS) produced no obvious change in peak potentials ofatechol, though slight lowering of peak current was observed. Fur-her, in the presence of only dopamine, the Fe3O4@PANI-modifiedlectrode showed a cathodic peak separation of about 90 mV foratechol-DA which indicates that DA could produce no substan-

ial effect on the assay of catechol. Similarly, there was no overlapf peaks of catechol with that of ascorbic acid suggesting the fea-ibility of reliable determination of catechol. Based on the aboveiscussion, it may be concluded that neither DA nor AA will

Fig. 5. Nyquist plots of Fe3O4@PANI modified electrodes with varying concentrationof catechol; the inset shows equivalent circuit diagram used to model impedancedata.

interfere in the sensing of catechol and the proposed system canbe well used for selective determination of catechol in the mixturesolution containing other possible interferents.

A typical steady state current response of the Fe3O4@PANInanohybrids modified GCE toward successive addition of catecholin a stirred solution of in 0.1 M PBS/0.1 M KCl electrolyte is shownin Figure S7 in SI. For this, 20 �L of catechol was added to the elec-trolyte solution and amperometric detection was carried out at astirring of 250 rpm for a period of 600 s. A fast and stable currentresponse was observed and the sensitivity of catechol was foundto be about 312 �A �L−1 with a detection limit of 0.2 nM (inset ofFig. 3).

3.4. Impedimetric and spectroelectrochemical characteristics ofthe nanohybrids

Electrochemical impedance spectroscopy (EIS) was used tomonitor the impedance changes of Fe3O4@PANI modified electrodesurface and its interfacial properties in the presence of differentconcentrations of catechol. The frequency was varied from 0.1to 105 Hz with 20 point impedance readings per decade. Nyquistdiagram (Fig. 5) displays that the modified electrode showed anincrease in Rs values (from 31.43 to 182.7 �) with increase in con-centration of catechol in solution where the analyte tends to adsorbfaster on the electrodes giving rise to more Rs indicating that theelectrochemical impedance signal is directly related to the amountof catechol adsorbed onto the electrode. A good linear relationshipwas obtained with a correlation coefficient of 0.9894, the limit ofdetection (LOD) was calculated to be 16.5 nM (S N−1 = 3).

To compliment the findings and to gain better insight into themechanism of the interaction between Fe3O4@PANI nanohybridsand catechol, spectroelectrochemical studies were performed.Potential dependent optical absorption spectra of Fe3O4@PANInanohybrids and Fe3O4@PANI nanohybrids-catechol in 0.1 Mdichloromethane solutions of [NBu4][B(C6F5)4] as supporting elec-trolyte was studied using UV/vis/NIR spectroscopy. For this, astepwise increase (100 mV) of the potential from −100 mV to700 mV vs. Ag/AgCl in an optically transparent thin-layer electro-chemistry (OTTLE) cell has been applied (Fig. 6).

Fe3O4@PANI nanohybrids showed the soret band at 380 nm

which increased in intensity on increasing the potential from −100to 600 mV. Fe3O4@PANI nanohybrids-catechol showed strongabsorption bands at 275 nm in the absence of any potential. Onapplying 100 mV, a new broad band appears at ∼390 nm. The

S. Chandra et al. / Analytica Chimica Acta 795 (2013) 8– 14 13

500 1000 1500 2000 2500

0.00

0.25

0.50

0.75

1.00

1.25

1.50

200 30 0 40 0 50 0 60 0 70 0

0.0

0.3

0.6

0.9

1.2

Wave l eng th (nm)

withou t p otential

-100 mV

0 mV

100 mV

200 mV

300 mV

400 mV

500 mV

600 mV

700 mVAb

so

rpti

on

/-

Ab

so

rpti

on

/-

Wave length (nm)

Fig. 6. UV/vis/near-IR spectroelectrochemical measurements of modified elec-te

ipTtTrPcsotwce

utftdsr

0 200 0 400 0 600 0 800 0 1000 0

8.1x1 0-5

8.4x1 0-5

8.7x10-5

9.0x1 0-5

9.3x1 0-5

9.6x1 0-5

9.9x1 0-5

Fe3O

4@PANI-

catecho l

" (

B/O

e)

Frequ ency (H z)

Fe3O

4@PANI

χμ

TR

TC

rodes in the presence of catechol, recorded in an OTTLE cell. Inset shows thexpanded region from 200–700 nm.

ntensity of the band at 275 nm started decreasing on applying aotential of 200 mV with a rise in intensity of the band at 390 nm.he change in the intensity of the absorption continued till a poten-ial of 400 mV, after which the band at 275 nm started diminishing.his transition indicates a reduction in the concentration of theeduced species and their conversion into the oxidized species.olarization of the electrode takes place from 0.2 to 0.7 V whichorresponds to the oxidation state of the catechol resulting in ateady growth of the band at 390 nm which grows at the expensef the band at 275 nm. The intervalence charge transfer band due tohe semiquinone-catecholate transition was observed at 1900 nmhich is shifted to lower energy with increase in potential which

ould be attributed to various substituents in the catecholate moi-ty [26].

We further explored this sensing through Brownian relaxationpon binding of the nanohybrids with the catechol. Fig. 7 showshe imaginary part of AC magnetic susceptibility as a function ofrequency in the presence of an AC magnetic field. On interac-ion with the catechols, the peak frequency of the nanoparticles

ecreased considerably along with a frequency shift toward higheride [27]. The interaction led to an increase in the hydrodynamicadii of the nanoparticles and their rotational motion is blocked

able 1ecoveries of catechol concentration in tap water samples by the proposed electrode.

Sample By HPLC (g g−1)

Added (�M) Found (�M)

Tap water 1 100 105

Tap water 2 120 123

able 2omparison of various catechol biosensors based on different electrode materials in term

Sl. No Electrode material Technique Limit of det

1 Graphene DPV 1.8 × 10−8 M2 Ferrocene Amperometry 10.8 �M

3 Poly(malachite green)coated MCNT film DPV 5.8 �M

4 DeniLite Laccase Amperometry 70 nM

5 Graphite SWV 2.9 × 10−7 M6 Au/TiO2 Amperometry 1 × 10−6 M

7 PANI Amperometry 0.05 �M

8 Mesoporous Pt DPV NM

9 Fe3O4@PANI/GCE AmperometryEIS

0.2 nM16.5 nM

Fig. 7. Imaginary part of the AC magnetic susceptibility as a function of frequencyfor Fe3O4@PANI (solid symbols) and Fe3O4@PANI-catechol (open symbols).

which diminished the peak frequency. Since the peak frequency ofthe AC susceptibility is inversely proportional to the particle vol-ume, this scheme can be used to monitor the change in the particlevolume upon binding of catechol to Fe3O4@PANI. The use of Brown-ian relaxation time measurements in frequency domain provides aplatform for developing a biomagnetic sensor.

3.5. Real sample analysis

The proposed sensing method was used to determine catechol intap water collected from two different sources and a known amountof catechol was added. Recovery experiments were performed andthe quantitative recoveries were evaluated from the calibrationcurve obtained by cyclic voltammetry. The results were also com-pared using HPLC and are tabulated in Table 1. The results show thatthe two methods are in good agreement and the Fe3O4@PANI mod-ified electrodes can be used for effective determination of catecholin the water samples.

Table 2 gives an overview of various materials and techniqueslike amperometry, square wave voltammetry (SWV), differential

pulse voltammetry (DPV) and cyclic voltammetry (CV) used forsensing and detection of catechol. It also provides a comparativeperformance of various catechol sensors in terms of detection limit,

By cyclic voltammetry using Fe3O4@PANI modified GCE Recovery (%)

Added (�M) Found (�M)

50 56 112100 95 95

s of performance.

ection Sensitivity Interference Reference

NM None with hydroquinone [28]1.14 �A mM−1 None with UA [29]1.3 mA cm−2 mM−1 None with quinol [30]70 nA �M−1 None with AA [31]

NM None with AA [32]51.58 �A �M−1 NM [33]NM None with AA, DA [2]1.20 mA cm−2 mM−1 NM [34]312 �A �L−1 None with AA, DA Present work

1 Chim

ss

4

tafFaTmnBanb

A

eG

A

t

R

[[[[

[

[[

[[[

[

[[

[

[

[

[

[[[[30] Y. Umashankar, A.P. Periasamy, S.-M. Chen, Anal. Biochem. 411 (2011) 71–79.

4 S. Chandra et al. / Analytica

ensitivity and interference which clearly shows that the presentystem is much better than the previously reported ones.

. Conclusion

In summary, we present a synthetic strategy for easy evolu-ion of a mesoporous Fe3O4@PANI nanohybrid and demonstraten ultra-sensitive electro- and spectroelectrochemical platformor label-free sensing of catechol. With the unique distribution ofe3O4 on the conducting PANI matrix, the catechol detection waschieved at nanomolar levels (the limit of detection was 0.2 nM).he decrease in the peak frequency of the imaginary part of the ACagnetic susceptibility demonstrates the feasibility of using the

anohybrid as biomagnetic sensor based on Brownian relaxation.esides this, we provide a thorough understanding of the inter-ction of catechols with the nanohybrid system, thereby openingew horizons in the area of spectroelectrochemical analysis andiosensing.

cknowledgements

Authors are grateful to the Nanomission of Department of Sci-nce and Technology, Govt. of India, and Humboldt Foundation,ermany for providing fellowships.

ppendix A. Supplementary data

Supplementary data associated with this article can be found, inhe online version, at http://dx.doi.org/10.1016/j.aca.2013.07.043.

eferences

[1] L.T. Zheng, G.-M. Ryu, B.-M. Kwon, W.-H. Lee, K. Suk, Eur. J. Pharmacol. 588(2008) 106–113.

[2] Y. Tan, J. Kan, S. Li, Sens. Actuators B 152 (2011) 285–291.[3] S. Chandra, K.C. Barick, D. Bahadur, Adv. Drug Deliv. Rev. 63 (2011) 1267–1281.

[[[[

ica Acta 795 (2013) 8– 14

[4] D. Nematollahi, A.R. Atlasi-Pak, R. Esmaili, HeIv. Chim. Acta 95 (2012)1605–1612.

[5] J. Germain, J.M.J. Frechet, F. Svec, J. Mater. Chem. 17 (2007) 4989–4997.[6] S. Chen, Z. Wei, X. Qi, L. Dong, Y. Guo, L. Wan, Z. Shao, L. Li, J. Am. Chem. Soc.

134 (2012) 13252–13255.[7] Y. Qiao, S.-J. Bao, C.M. Li, X.-Q. Cui, Z.-S. Lu, J. Guo, ACS Nano 2 (2008) 113–119.[8] Y. Ma, S.R. Ali, L. Wang, P.L. Chiu, R. Mendelsohn, H. He, J. Am. Chem. Soc. 128

(2006) 12064–12065.[9] S. Chandra, N. Barola, D. Bahadur, Chem. Commun. 47 (2011) 11258–11260.10] S. Shreepathi, R. Holze, Chem. Mater. 17 (2005) 4078–4085.11] W. Shen, M. Shi, M. Wang, H. Chen, Mater. Chem. Phys. 122 (2010) 588–594.12] Q. Xiao, X. Tan, L. Ji, J. Xue, Synth. Met. 157 (2007) 784–791.13] S. Dietrich, S. Chandra, C. Georgi, S. Thomas, D. Makarov, S. Schulze, M.

Hietschold, M. Albrecht, D. Bahadur, H. Lang, Mater. Chem. Phys. 132 (2012)292–299.

14] S. Radhakrishnan, C.R.K. Rao, M. Vijayan, J. Appl. Polym. Sci. 122 (2011)1510–1518.

15] S. Nigam, K.C. Barick, D. Bahadur, J. Magn. Magn. Mater. 323 (2011) 237–243.16] Y.-L. Chueh, M.-W. Lai, J.-Q. Liang, L.-J. Chou, Z.L. Wang, Adv. Funct. Mater. 16

(2006) 2243–2251.17] Y. Liu, L. Liao, C. Pan, J. Li, Y. Dai, W. Chen, J. Phys. Chem. C 112 (2008) 902–910.18] Y. Liang, Y. Li, H. Wang, H. Dai, J. Am. Chem. Soc. 135 (2013) 2013–2036.19] J. Liang, Y. Xu, D. Sui, L. Zhang, Y. Huang, Y. Ma, F. Li, Y. Chen, J. Phys. Chem. C

114 (2010) 17465–17471.20] K.R. Reddy, K.P. Lee, A.I. Gopalan, Colloids Surf. A: Physicochem. Eng. Aspects

320 (2008) 49–56.21] S. Xiong, Q. Wang, Y. Chen, Mater. Chem. Phys. 103 (2007) 450–455.22] S. Pethkara, R.C. Patila, J.A. Kherb, K. Vijayamohanan, Thin Solid Films 349

(1999) 105–109.23] M. Amrithesh, S. Aravind, S. Jayalekshmi, R.S. Jayasree, J. Alloys Compd. 449

(2008) 176–179.24] Q. Ximin, D. Yanfang, Z. Guirong, Z. Peng, L. Jiaxing, Front. Chem. China 3 (2006)

345–349.25] L.J. Nùnez-Vergara, S. Bollo, J. Fuentealba, J.C. Sturm, J.A. Squella, Pharm. Res.

19 (2002) 522–529.26] N. Shaikh, S. Goswami, A. Panja, X.-Y. Wang, S. Gao, R.J. Butcher, P. Banerjee,

Inorg. Chem. 43 (2004) 5908–5918.27] S.H. Chung, A. Hoffmann, S.D. Bader, Appl. Phys. Lett. 85 (2004) 2971–2973.28] H. Du, J. Ye, J. Zhang, X. Huang, C. Yu, J. Electroanal. Chem. 650 (2011) 209–213.29] A.S. Kumar, P. Swetha, K.C. Pillai, Anal. Methods 2 (2010) 1962–1968.

31] D. Quan, W. Shin, Electroanalysis 16 (2004) 1576–1582.32] G.A.M. Mersal, Int. J. Electrochem. Sci. 4 (2009) 1167–1177.33] M. Wei, Y. Liu, Z.-Z. Gu, Z.-D. Liu, J. Chin. Chem. Soc. 58 (2011) 516–521.34] M.A. Ghanem, Electrochem. Commun. 9 (2007) 2501–2506.