High-sensitivity paracetamol sensor based on Pd/graphene oxidenanocomposite as an enhanced electrochemical sensing platform

Junhua Li a,b,c, Jinlong Liu a, Gongrong Tan a, Jianbo Jiang a, Sanjun Peng a, Miao Deng a,Dong Qian a,c,n, Yonglan Feng b, Youcai Liu a,n

a College of Chemistry and Chemical Engineering, Central South University, Changsha 410083, PR Chinab Department of Chemistry and Material Science, Hengyang Normal University, Hengyang 421008, PR Chinac State Key Laboratory of Powder Metallurgy, Central South University, Changsha 410083, PR China

a r t i c l e i n f o

Article history:Received 16 August 2013Received in revised form21 October 2013Accepted 1 November 2013Available online 8 November 2013

Keywords:Pd nanoparticlesGraphene oxideElectrochemical sensorParacetamol

a b s t r a c t

Well-dispersed Pd nanoparticles were facilely anchored on graphene oxide (Pd/GO) via a one-potchemical reduction of the Pd2þ precursor without any surfactants and templates. The morphology andcomposition of the Pd/GO nanocomposite were characterized by scanning electron microscopy (SEM),transmission electron microscopy (TEM), Fourier transform infrared spectroscopy (FTIR) and energydispersive analysis of X-ray (EDX). The stepwise fabrication process of the Pd/GO modified electrode andits electrochemical sensing performance towards paracetamol was evaluated using electrochemicalimpedance spectroscopy (EIS), cyclic voltammetry (CV) and differential pulse voltammetry (DPV).The experimental results indicate that the as-synthesized Pd nanoparticles are relatively uniform insize (5–10 nm) without large aggregation and uniformly distributed in the carbon matrix with the overallPd content of 28.77 wt% in Pd/GO. Compared with the GO modified electrode, the Pd/GO modifiedelectrode shows a better electrocatalytic activity to the oxidation of paracetamol with lower oxidationpotential and larger peak current, so the Pd/GO nanocomposite can be used as an enhanced sensingplatform for the electrochemical determination of paracetamol. The kinetic parameters of the para-cetamol electro-oxidation at Pd/GO electrode were studied in detail, and the determination conditionswere optimized. Under the optimal conditions, the oxidation peak current is linear to the paracetamolconcentration in the ranges of 0.005–0.5 μM and 0.5–80.0 μM with a detection limit of 2.2 nM. Based onthe high sensitivity and good selectivity of the Pd/GO modified electrode, the proposed method wassuccessfully applied to the determination of paracetamol in commercial tablets and human urines, andthe satisfactory results confirm the applicability of this sensor in practical analysis.

& 2013 Elsevier B.V. All rights reserved.

1. Introduction

Paracetamol (N-acetyl-p-aminophenol) is an effective and impor-tant analgesic agent used worldwide for the relief of pain associatedwith arthralgia, neuralgia, cephalagra, cancer pain, headache, back-ache, and postoperative pain (Yin et al., 2011). In general, paraceta-mol seems to be safe and appears to have no toxic effects on human'shealth when taken in normal therapeutic doses (Tsierkezos et al.,2013). However, large doses and chronic use of paracetamol, orconcomitant use with alcohol or other drugs can cause skin rashes,liver disorders, nephrotoxicity and inflammation of the pancreas(Prabakar and Narayanan, 2007). It is also worth noting that someepidemiological studies suggest a possible association between the

use of paracetamol during pregnancy and an increasing appearanceof asthma in children (Andersen et al., 2012). Furthermore, para-cetamol might be reentering the food chain by the contaminatedfeather meal which can be used as an additive in animal feed (Loveet al., 2012). Thus, it is imperative to develop a simple, fast andaccurate quantification method for paracetamol as it would directlyimpact clinical diagnosis, food safety and quality of paracetamol-containing medicines.

Until now, various methods have been developed for thedetermination of paracetamol, such as capillary electrophoresis(CE) (Maher et al., 2013), high performance liquid chromatography(HPLC) (Abdelaleem and Abdelwahab, 2013), UV-spectrophoto-metry (Abirami and Vetrichelvan, 2013), HPLC-tandem mass spec-trometry (Modick et al., 2013) and thermogravimetric analysis(TGA) (Khanmohammadi et al., 2012). The determinations of para-cetamol using these techniques show high sensitivities, but thesemethods are generally performed at centralized laboratories, requiringexpensive instruments, skilled operators, multi-step sample

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/bios

Biosensors and Bioelectronics

0956-5663/$ - see front matter & 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.bios.2013.11.001

n Corresponding authors at: College of Chemistry and Chemical Engineering,Central South University, Changsha 410083, PR China. Tel./fax: þ86 731 88879616.

E-mail addresses: qiandong6@vip.sina.com (D. Qian),liuyoucai@126.com (Y. Liu).

Biosensors and Bioelectronics 54 (2014) 468–475

preparations and complicated analysis procedures, which limittheir applications for the on-site determination. In recent years,electroanalytical techniques have been receiving considerableattention in the detection of drugs and environmental pollutantsbecause of their high sensitivities, excellent selectivities, lowcosts, fast responses, and timesaving and simple operations. Dueto its electroactive groups of hydroxy and acetamido in benzenering, paracetamol can be detected by the electrochemical methods.Nevertheless, this analyte is usually hard to be detected directly atconventional working electrodes because of the poor electrochemi-cal responses. Toward a general solution to this problem, variousmaterials with excellent electrocatalytic activities such as nanopar-ticles of metals and metallic oxides (Ye et al., 2012; Atta et al., 2009,2011; Zhang et al., 2010; Devaraj et al., 2013; Bui et al., 2012),conductive polymers (Ghadimi et al., 2013), mesoporous materials(Mazloum-Ardakani et al., 2012) and various carbon-based nano-materials (Olivé-Monllau et al., 2013; Kutluay and Aslanoglu, 2013;Arvand and Gholizadeh, 2013; Fan et al., 2011; Kang et al., 2010;Habibi et al., 2011), have been employed to modify the workingelectrodes for developing highly sensitive and selective methods forthe paracetamol quantification. Among these materials, graphene,due to its favorable characteristics such as high electrical conduc-tivity, large surface-to-volume ratio and excellent chemical toler-ance, is regarded as an attractive matrix for the preparations ofpreferable and multifarious sensors (Fan et al., 2011; Kang et al.,2010; J. Li et al., 2012, 2013). From a series of literatures, it is wellknown that the uses of graphene and its hybrids for modifyingelectrodes can effectively improve the detection limits and detec-tion ranges for lots of analytical targets (Li et al., 2012a, 2013).However, the detection limits of the graphene-based electrochemi-cal sensors for paracetamol are still relatively high in recent reports.For instance, the graphene-based and Nafion/TiO2/graphene-based electrochemical sensors for paracetamol with respectivedetection limits of 32 nM and 210 nM were reported (Kang et al.,2010; Fan et al., 2011). On the other hand, the construction of astable and enhanced sensing interface based on the functiona-lized graphene is still a challenge in the developments ofeffective and practical sensors because the graphene with anexclusive two-dimensional structure often tends to form con-glomeration or even re-graphitizes to graphite through the vander Waals interaction and strong π–π piling, leading to releasingthe high surface area of graphene.

Noble metal nanoparticles (such as Au, Ag, Pt and Pd) haveattracted great interests owing to their high catalytic activities formany chemical reactions. With an ease of miniaturization tonanoscale dimensions and the multifunctional abilities, thesenanoparticles are usually used to construct various chemical/biochemical sensing platforms. Among these nanoparticles, Pdnanoparticles are one of the hot research topics due to their higherabundance and lower cost in comparison with Pt and Au nano-particles. To promote the performance of the Pd-based catalyst,it is very important to select an appropriate supporting matrix,which needs to be stable and cheap, and exhibits a possiblesynergistic effect contributing to the enhancement of the electro-catalytic activity of Pd. In this regard, a variety of Pd/graphenehybrids have been prepared and used as the electrocatalysts forthe oxidation of glucose (Lu et al., 2011; Wang et al., 2012;Claussen et al., 2012), hydrogen peroxide (Chen et al., 2013; Youet al., 2013; Nandini et al., 2013), ascorbic acid (Wu et al., 2012),dopamine (Palanisamy et al., 2013) and formaldehyde (Qiao et al.,2013). These researches provide a hint that graphene could be anideal substrate for growing and anchoring Pd nanoparticles. But inprevious works, the Pd nanoparticles used as sensing materialswere ordinarily obtained by electrodeposition (Z. Li et al., 2012;Thiagarajan et al., 2009), electrospinning (Huang et al., 2008) orspontaneous redox reaction (Wu et al., 2012; Chen et al., 2013),

in which the metal content and particle size in composites arehard to be manipulated. In addition, some reported preparationprocesses need multi-step operations or complicated conditionswithout ultrafine, well-dispersed and high-yield productions(Thiagarajan et al., 2009; Huang et al., 2008).

In this paper, we demonstrate a facile one-step chemicalreduction method to obtain a high quality Pd/graphene oxide(Pd/GO) nanocomposite, by which well-dispersed Pd nanoparti-cles in situ grew on the surfaces of GO through the chemicalreduction of the Pd2þ precursor without any surfactants andtemplates. And then the as-synthesized nanocomposite withoutlarge aggregation was employed to construct an enhanced sensingplatform for the electrochemical detection of paracetamol incommercial tablets and human urines. The constructed sensordisplays fast electron transfer, prominent catalytic ability andexcellent analytical performance to paracetamol with a wide linearrange, a low detection limit and satisfactory recoveries. Therefore,this work opens a new avenue to detect paracetamol via theelectrochemical method, benefits the studies of the redox meta-bolism of paracetamol in aqueous solution and broadens theapplications of carbon-metal composite materials in electroanaly-tical chemistry.

2. Experimental

2.1. Reagents and apparatus

Full reagent and apparatus details can be found in the SupportingInformation (SI).

2.2. Synthesis of the Pd/GO nanocomposite

To anchor the Pd nanoparticles uniformly on GO, the Pd/GOnanocomposite was prepared by an in situ chemical method usingPd2þ and GO as the precursors. GO was first synthesized fromnatural graphite powders by a modified Hummers' method asdescribed in our previous publication (Liang et al., 2011). In a typicalpreparation process of Pd/GO nanocomposite, 10 mg of GOwas firstlydispersed into 10 mL water with ultrasonic oscillation for 1.0 h toobtain GO precursor, and then 1.8 mg of PdCl2 was slowly added intothe GO suspension under stirring. Subsequently, 10 mg of NaBH4 wasadded into the suspension to reduce the Pd2þ in situ without anysurfactants and templates, and the reduction reaction was lasted for7.0 h under vigorously stirring at ambient temperature and atmo-spheric pressure. After that, the mixture was rested for 10 h atambient temperature. Finally, the precipitates were collected bycentrifugation, washed with deionized water several times to removethe unreacted metal ions and dried at 60 1C for 12 h in a vacuumoven to obtain the Pd/GO nanocomposite.

2.3. Preparation of the modified electrodes

The Pd/GO modified GCE (Pd/GO/GCE) and GO modified GCE(GO/GCE) were prepared by a general drop-coating method andthe detailed description is listed in SI.

3. Results and discussion

3.1. Morphological characterization of the Pd/GO nanocomposite

The detailed morphology and composition of the as-generatedGO and Pd/GO were examined by SEM, TEM, EDX and FTIR. As canbe seen from SEM images, GO is formed with a typical crumpledsurface structure due to the chemical exfoliation (Fig. 1A).

J. Li et al. / Biosensors and Bioelectronics 54 (2014) 468–475 469

Nevertheless, the crumpled surfaces of GO change to multilayeredand defective surfaces with the Pd nanoparticles decorated(Fig. 1B). The rougher surfaces and layered structure of the Pd/GO nanocomposite could significantly increase the effective elec-trode surface and facilitate the diffusion of analytes into the Pd/GOsheets. The TEM image of the pristine GO (Fig. 1C) shows that thepristine GO has a clean and hyaline surface with wrinkled textures,while the TEM image of the Pd/GO nanocomposite (Fig. 1D)exhibits that the Pd nanoparticles are well formed and highlydispersed on GO despite of some slight aggregation. The HRTEMimage of the Pd/GO nanocomposite (Fig. 1E) further reveals thatthe Pd nanoparticles are firmly anchored on the surface of GO witha highly ordered crystalline structure and in the size of 5–10 nm,which is much smaller than those of the Pd nanoparticles directlyelectrodeposited or electrospinned on the surfaces of C60 (10.5 nm)(Z. Li et al., 2012), ITO (39–78 nm) (Thiagarajan et al., 2009), carbon

nanofibers (73 nm) (Huang et al., 2008) and reduced GO (100–200 nm) (Palanisamy et al., 2013). Undoubtedly, the smaller nano-particles are beneficial to increasing the surface area and catalyticability of the Pd/GO nanocomposite. The Pd loading on the GOsheets is estimated to be 28.77 wt% (or 4.53% in atomic weight)based on the EDX data of the Pd/GO nanocomposite as shown inFig. 1F. Changes in the functionalized surfaces of samples were alsoobserved by FTIR spectroscopy (Fig.S1). The bands centered at 1052,1096, 1400, 1609, 1723 and 3402 cm�1 can be ascribed to theoxygen-containing functional groups on GO (Fig. S1a), indicat-ing the successful oxidation of graphite. However, as presented inFig. S1b, the peak intensity of the oxygen-containing functionalgroups in the Pd/GO nanocomposite show a greater decrease due tothe decoration of Pd onto the GO surfaces. Moreover, the functionaloxygen-containing groups on GO, such as –OH, serve not only as thenucleation sites but also as the reductants, resulting in more Pd

Fig. 1. SEM images of the GO (A) and Pd/GO (B) sheets, TEM images of the GO (C) and Pd/GO (D) sheets, and HRTEM image (E) and EDX (F) of the Pd/GO sheets.

J. Li et al. / Biosensors and Bioelectronics 54 (2014) 468–475470

nanoparticles closely anchored on the GO sheets without anydislocation (Chen et al., 2011; Wang et al., 2013). The enhancedelectrocatalytic property can also be attributed to the reduction ofthe surface oxygen groups (You et al., 2013). As a result, the Pd/GOnanocomposite without large aggregation and lots of oxygen-containing groups has a preferable film-forming property and couldgreatly improve the electrocatalytic and electroanalytical capabil-ities of the prepared sensors.

3.2. Electrochemical characterization of the Pd/GO modifiedelectrode

The cyclic voltammetry (CV) and electrochemical impedancespectroscopy (EIS) were employed to study the stepwise fabrica-tion process and electrochemical characterization of the modifiedelectrodes, and the results (see Fig. S2) confirmed that Pd/GO hasbeen successfully immobilized on the surface of GCE (see SI for thediscussion). According to the Randles–Sevcik equation (Bard andFaulkner, 2001), the electroactive surface areas of the bare GCE,GO/GCE and Pd/GO/GCE are calculated to be 0.0755, 0.182 and0.251 cm2, respectively (see SI for the detailed calculation process).

3.3. Electrocatalytic behaviors of paracetamol at Pd/GO/GCE

Fig. 2A illustrates CV responses of GCE (a), GO/GCE (b) and Pd/GO/GCE (c) in 0.1 M PBS (pH¼6.8). No redox peaks are observed atGCE and GO/GCE in PBS, implying that GCE and GO/GCE are non-electroactive in the investigated potential region. In contrast, theredox peaks appeared at �0.309 V and 0.009 V for Pd/GO/GCE,which can be ascribed to the electrochemical redox process of thePd nanoparticles and further confirms that the Pd nanoparticleshave been firmly decorated on the GO surfaces. Fig. 2B displays CVresponses of GCE (a), GO/GCE (b) and Pd/GO/GCE (c) in thepresence of 10 μM paracetamol. It can be seen that paracetamolshows a quasi-reversible redox behavior with relatively weakredox peaks at the bare GCE, which indicates that the directelectron transfer of paracetamol at the bare GCE is very slow.At GO/GCE, a pair of well-defined and quasi-reversible redox peakscorresponding to the electrochemical reactions of paracetamol areobserved with the oxidation peak at 0.475 V. The significantlyincreased redox peak currents of GO can be reasonably attributedto its increased surface area. Furthermore, a distinct well-definedredox peaks appear at Pd/GO/GCE with the oxidation peakpotential negatively shifted to 0.463 V, of which the oxidationpeak current is ca. 12.17 and 2.08 times than those of the bare GCEand GO/GCE, respectively. On the whole, redox peaks of para-cetamol were enhanced due to the wonderful conductivity of Pdand the larger surface area of Pd/GO modified electrode. Further,the decrease of the oxidation overpotential is a clear evidence ofthe greater electrocatalytic activity of the Pd/GO nanocompositetowards the oxidation of paracetamol, which could be furtherconfirmed by chronoamperometry (CA) (see Fig. S3). From theresults of CA, it can be seen that the response time of Pd/GO/GCE tothe paracetamol oxidation is less than 10 s, and the catalytic rateconstant is calculated to be 5.18�103 M�1 s�1 by Galus'sequation (Galus, 1994) (see SI for the detailed calculation process).Without any doubt, the synergetic effect of GO and Pd nanopar-ticles make contributions to the higher current response anddetection sensitivity of paracetamol.

3.4. Effect of the pH values

The effect of the PBS solution pH in the range of 4.0–9.0 on theredox response of 10 μM paracetamol was investigated in 0.1 M PBSby CV with the results exhibited in Fig. 3A, and the relationships of theoxidation peak current/potential with the solution pH are depicted in

Fig. 3B and C, respectively. From Fig. 3B, it can be seen that the redoxpeak current signal increases with the solution pH ranging from 4.0 to6.8, and beyond this pH range, the decrease of the redox peak currentsignal is observed. Considering the determination sensitivity, the pHvalue of 6.8 in the PBS solution was chosen in the followinginvestigation. Interestingly, the anodic and cathodic peak potentialsshift to more negative values with increasing the pH values as shownin Fig. 3A. This observation can be explained by the protons taking partin the electrochemical reactions (Kutluay and Aslanoglu, 2013; Arvandand Gholizadeh, 2013). Moreover, a linear relationship can be obtainedbetween the anodic peak potential (Epa) and pH value, and theregression equation is plotted as Epa(V)¼�0.0552 pHþ0.8593(R2¼0.9835). In this case, the slope value of �55.2 mV pH�1 is veryclose to the theoretical value of �59.0 mV pH�1 from the Nernstequation, indicating that the equal amounts of proton and electrontransfer occur in the electrode reactions. This result agrees well withthose obtained at multi-walled carbon nanotubes (MWCNT) (Kutluayand Aslanoglu, 2013), MWCNTs/graphene (Kutluay and Aslanoglu,2013), graphite/polystyrene (Khaskheli et al., 2013), Poly(4-vinylpyr-idine)/MWCNT (Ghadimi et al., 2013), Nafion/TiO2-graphene (Fanet al., 2011) and single-walled carbon nanotubes (SWCNT) (Goyalet al., 2010) modified electrodes.

3.5. Effect of the potential scan rate

The effect of the potential scan rate on the electro-oxidation ofparacetamol at Pd/GO/GCE was also investigated by CV. With theincrease of the potential scan rate from 20 to 300 mV s�1, the

Fig. 2. CVs of GCE (a), GO/GCE (b) and Pd/GO/GCE (c) in 0.1 M PBS (pH¼6.8) inabsence (A) and presence (B) of 10 μM paracetamol.

J. Li et al. / Biosensors and Bioelectronics 54 (2014) 468–475 471

oxidation peak potential shifts to more positive direction and theoxidation peak current increases gradually (Fig. 4A). A straight line(Fig. 4B) can be obtained for the plot of the oxidation peak current(ipa) vs. the potential scan rate (v), following the regressionequation: ipa (μA)¼0.05774v (mV s�1)þ2.0550 (R2¼0.9959). Thissuggests that the electrode process is controlled by the adsorptionrather than the diffusion of paracetamol on the electrode surface,and the result is in agreement with the conclusions reported inrecent literatures (Arvand and Gholizadeh, 2013; Ghadimi et al.,2013; Fan et al., 2011; Goyal et al., 2010). According to this result,the surface coverage of paracetamol on the electrode surface iscalculated to be 6.12�10�8 mol cm�2 by using Laviron'sequation (Laviron, 1979) (see SI for the detailed calculationprocess). Hence, the calculated surface coverage is much higherthan those of the electrodeposited nano-Pd film (5.70�10�10

mol cm�2) (Thiagarajan et al., 2009), diglycolic acid polymer(1.43�10�9 mol cm�2) (Xu et al., 2012) and Nafion/TiO2-gra-phene film (3.24�10�9 mol cm�2) (Fan et al., 2011). The obtainedsurface coverage is also higher than that of a monomolecular layer(J. Li et al., 2012), indicating that the multilayer adsorption ofparacetamol appeared at Pd/GO film due to its specific surfacestructure. In brief, Pd/GO has an excellent adsorption capacity toparacetamol.

As shown in Fig. 4C and Fig. S4, with increasing the potential scanrate, the redox peak potentials (Epa and Epc) have linear relationshipswith the Napierian logarithm of the potential scan rate (ln v). In theselected potential scan rate range, the linear regression equations canbe expressed as Epa¼0.02225 ln vþ0.3793 (R2¼0.9907) and Epc¼�0.02305 ln vþ0.5059 (R2¼0.9762). According to Laviron's model(Laviron, 1974; Laviron, 1979), the number of the transferred electrons

(n) and charge-transfer coefficient (α) can be deduced to be 2.26 and0.49, respectively (see SI for the detailed calculation process). There-fore, it can be concluded that there is a two-proton and two-electronprocess for the paracetamol electro-oxidation at Pd/GO/GCE, and thepossible reaction mechanism is proposed according to other reports(Kutluay and Aslanoglu, 2013; Arvand and Gholizadeh, 2013; Ghadimiet al., 2013; Fan et al., 2011) and listed in Fig. S5. Moreover, the value ofapparent rate constant (ks) was evaluated to be 2.706 s�1 according toLaviron's model (see SI for the detailed calculation process), and it issignificantly higher than those previously reported for some othermodifiers such as nitrogen-doped MWCNT (0.0036 s�1) (Tsierkezoset al., 2013) and MWCNTs/graphene (1.8 s�1) (Arvand and Gholizadeh,2013), displaying a quick electron-transfer reaction at the developedsensor.

3.6. Optimization of the accumulation conditions and Pd/GOconcentration

The optimized accumulation step in the experiments was set at0.2 V with 40 s under motionless condition. Moreover, the con-centration of 0.5 mg mL�1 Pd/GO was chosen to modify GCE forthe electrochemical detection of paracetamol, and the detailedinformation is given in SI with Fig. S6 and S7.

3.7. Analytical performance characteristics

As known to all, differential pulse voltammetry (DPV) hasadvantages of an increase in sensitivity and better resolution forthe quantitative analysis; therefore, DPV was employed to study

Fig. 3. CVs of 10 μM paracetamol at Pd/GO/GCE under different pH. Curves (a–f):4.0, 5.0, 6.8, 7.5, 8.0 and 9.0 (A); effect of pH on the oxidation peak currents (B);and the relationship between pH and Epa (C).

Fig. 4. CVs of 10 μM paracetamol at Pd/GO/GCE at different potential scan rates.Curves (a, b, c, d, e, f, g, h, i, and j) are obtained at 20, 40, 60, 80, 100, 120, 150, 200,250 and 300 mV s�1, respectively (A); dependence of the oxidation peak current(ipa) on the potential scan rate (v) (B); and the relationship between Epa andln v (C).

J. Li et al. / Biosensors and Bioelectronics 54 (2014) 468–475472

the linear concentration range and detection limit of paracetamolat Pd/GO/GCE. Moreover, optimization of DPV parameters usingthe prepared sensor shows the best results at a pulse amplitude of25 mV, a pulse width of 50 ms and a pulse period of 200 ms. Underthe all optimized experiment conditions, the anodic peak currentis proportional to the concentration of paracetamol in a widerrange of 5.0 nM to 80.0 μM as shown in Fig. 5. It can be seen thatthe calibration plots of peak current vs. concentration of para-cetamol show two linear dynamic ranges from 0.005–0.5 μM(Fig. 5A) and 0.5–80 μM (Fig. 5B). The linear regression equationsfor these two regions are ipa (μA)¼0.05561C (nM)þ1.8284(R2¼0.9946) and ipa (μA)¼1.4496C (μM)þ33.426 (R2¼0.9923).Also the detection limit is calculated to be 2.2 nM by the equationof 3sb/m (where sb is the standard deviation of the blank andm is the slope of the calibration plot). In order to investigatethe analytical characteristics of the proposed method, Table S1summarizes a careful comparison of Pd/GO/GCE with other elec-trochemical sensors reported for the paracetamol determination

(Olivé-Monllau et al., 2013; Tsierkezos et al., 2013; Baś et al., 2013;Arvand and Gholizadeh, 2013; Khaskheli et al., 2013; Ghadimiet al., 2013; Mazloum-Ardakani et al., 2012; Švorc et al., 2012; Yeet al., 2012; Yang et al., 2012; Yin et al., 2011; Fan et al., 2011; Öcanand Şahin, 2011; Zhang et al., 2010; Kang et al., 2010; Nair et al.,2009; Atta et al., 2009; Lourenção et al., 2009; Kachoosangi et al.,2008; Keeley et al., 2012; Devaraj et al., 2013; Bui et al., 2012). Inparticular, sensitivities listed in Table S1 are expressed as theslopes of the calibration graphs. As can be seen from Table S1, thesensitivity of this method is higher than those for most othersensors just except the electrochemically treated pencil graphiteelectrode (Öcan and Şahin, 2011). It is worth noting that the lowestdetection limit and lower dynamic range are achieved for theparacetamol determination by this novel approach. The comparedresults further confirm that the synergetic effect of Pd and GO cangreatly enhance the electroanalytical performances of the preparedsensor which could potentially be used to sensitively detect para-cetamol in practical matrixes.

3.8. Reproducibility, reusability and stability

The assay reproducibility of Pd/GO/GCE was studied throughrepeating the determination of 10 μM paracetamol by DPV. Aftereach determination, the modified electrode used underwent 10successive CV sweeps between 0.2 and 0.7 V at the rate of100 mV s�1 in the blank PBS (pH 6.8) to remove any adsorptivesubstances and was regenerated. The 10 measurements achievegood reproducibility with the relative standard deviation (RSD) of2.1%. The fabrication reproducibility of Pd/GO/GCE was also inves-tigated by DPV for detecting 10 μM paracetamol. Ten Pd/GO/GCEswere prepared independently by the same procedure, and the RSDresult is 3.9%, confirming an acceptable reproducibility for theelectrode fabrication.

The stability of Pd/GO/GCE was first checked by recordingsuccessive CV sweeps. After 200 cycles, no change was observedin the voltammetric profiles of the modified electrode. After that,this modified electrode was used to detect 10 μM paracetamol byDPV, and no significant change was observed in 30 successivedetections. While the obvious decrease of DPV signal appeared,Pd/GO/GCE could be reactivated in blank PBS (pH 6.8) with CVscanning several times between 0.0 and 1.0 V to eliminate theimpurities at electrode surface. Apparently, the prepared electrodeexhibited good reusability with convenience of reuse. Further-more, when the modified electrode was stored in a refrigerator at4 1C, it exhibited no obvious decrease in the current response inthe first week and maintained about 90% of its initial value afterthree weeks. This investigation discloses that the sensor possessesgood stability and can be employed for the practical application.

3.9. Effect of the interferences

An important parameter for a sensor is its ability to discrimi-nate between the interfering species commonly present in similarphysiological environment and the target analyte. Voltammetricresponses of Pd/GO/GCE to 10 μM paracetamol were examined inthe presence of some possible interfering biomolecules andorganic compounds, and the results are listed in Table S2. It canbe seen that most of the selected coexisting analytes are electro-active, but their oxidation potentials are quite different from thatof paracetamol (0.463 V), which illustrates that Pd/GO/GCE hasfavorable properties of discrimination and anti-interference in thecomplex matrix. Furthermore, it was found that the anodic peakcurrent of paracetamol is not distinctly affected in the presence of10-folds of ascorbic acid, dopamine, uric acid and 4-aminophenol,20-folds of tyrosine, tryptophan and guanine, 30-folds of adenineand vitamin B1, and 50-folds of glucose. Finally, the influence of

Fig. 5. (A) DPVs obtained at Pd/GO/GCE in 0.1 M PBS (pH 6.8) containing lowconcentrations of paracetamol (a–j: 0.0, 5.0, 10.0, 20.0, 30.0, 80.0, 120.0, 200.0, 300.0and 500.0 nM), and the inset is the calibration plot (the RSD for the determination is lessthan 4.7% for n¼5); (B) DPVs obtained at Pd/GO/GCE in 0.1 M PBS (pH 6.8) containinghigh concentrations of paracetamol (a–f: 0.5, 3.0, 18.0, 32.0, 55.0 and 80.0 mM), and theinset is the calibration plot (the RSD for the determination is less than 3.5% for n¼5).DPV conditions: pulse amplitude¼25mV, sample width¼17ms, pulse width¼50ms,pulse period¼200 ms and quiet time¼2 s.

J. Li et al. / Biosensors and Bioelectronics 54 (2014) 468–475 473

common inorganic ions was also studied. It was determined that100-folds of Kþ , Ca2þ , Cu2þ , Cd2þ , Pb2þ , Al3þ , Fe3þ , SO4

2� andCl� have almost no influence on the response signal of paraceta-mol with the deviation below 5%. Therefore, the present methodhas an excellent selectivity towards paracetamol.

3.10. Detection of paracetamol in real samples

Commercial pharmaceutical samples containing paracetamol wereemployed to evaluate the practical performance of the developedsensor. First, three paracetamol commercial tablets were ground topowders, dissolved in 0.1 M PBS (pH 6.8) and then diluted to aworking concentration range for the electrochemical determination.It was found that the average concentration of paracetamol wasdetected to be 489mg/tablet, which is in good agreement with thecontent of paracetamol (500 mg/tablet) provided by the manufacturer.In order to further confirm the applicability of Pd/GO/GCE, thecontents of paracetamol in human urine samples were also detectedusing the standard addition method. The urine samples were collectedfrom patients after 3 h of the oral administration of commercial tabletscontaining 500 mg of paracetamol. Prior to analysis, the urine sampleswere diluted 50 times with 0.1 M PBS (pH 6.8) to reduce the matrixcomplexity. The results obtained from three kinds of urine samples,before and after spiking a known concentration of paracetamol, arelisted in Table S3. The acceptable recoveries were obtained in therange of 98.1–103.5%, indicating that there are no important matrixinterferences for the samples analyzed by the proposed DPV method.These findings reveal that the Pd/GO/GCE fabricated in this work canbe used as an effective and reliable sensing platform for detectingparacetamol in drugs and biological fluids.

4. Conclusions

In this work, the ultrafine and high-yield Pd nanoparticles werewell dispersed on the GO sheets via the one-step chemical reductionunder mild conditions without any surfactants and templates. Then anovel electrochemical sensing approach based on the prepared Pd/GOnanocomposite has been proposed for the sensitive detection ofparacetamol. The Pd/GO-based sensor shows enhanced electrontransfer property and favorable electrocatalytic performance towardsparacetamol. Wide linear concentration ranges (0.005–0.5 and 0.5–80 μM), low detection limit (2.2 nM), and excellent reproducibility andstability are achieved on the resulting sensor, indicating the hybridsheets could be used as a promising sensing platform for the analyticalapplication. This work provides a new way for the detection ofparacetamol by the electrochemical method, and broadens the appli-cation of graphene-based materials in the electrochemical sensor andpharmaceutical analysis.

Acknowledgments

This work was financially supported by National NaturalScience Foundation of China (No. 21171174), Provincial NaturalScience Foundation of Hunan (Nos. 09JJ3024 and 13JJ3112), Scien-tific Research Project of Education Department of Hunan Province(Nos. 12C0536 and 13K105), Provincial Environmental Science andTechnology Foundation of Hunan, and the Open-end Fund for theValuable and Precision Instruments of Central South University.

Appendix A. Supplementary material

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.bios.2013.11.001.

References

Abdelaleem, E.A., Abdelwahab, N.S., 2013. Anal. Methods 5, 541–545.Abirami, G., Vetrichelvan, T., 2013. Int. J. Pharm. Pharm. Sci. 5, 488–492.Andersen, A.B., Farkas, D.K., Mehnert, F., Ehrenstein, V., Erichsen, R., 2012. Clin.

Epidemiol. 4, 33–40.Arvand, M., Gholizadeh, T.M., 2013. Colloids and Surf. B 103, 84–93.Atta, N.F., El-Kady, M.F., Galal, A., 2009. Sens. Actuators B: Chem. 141, 566–574.Atta, N.F., Galal, A., Abu-Attia, F.M., Azab, S.M., 2011. J. Mater. Chem. 21,

13015–13024.Bard, A.J., Faulkner, L.R., 2001. Electrochemical Methodssecond ed. Wiley–VCH,

New York.Baś, B., Bugajna, A., Jakubowska, M., Reczyński, W., Smalec, A., 2013. Electrochim.

Acta 99, 190–197.Bui, M.P.N., Li, C.A., Han, K.N., Pham, X.H., Seong, G.H., 2012. Sens. Actuators B:

Chem. 174, 318–324.Chen, X., Cai, Z., Huang, Z., Oyama, M., Jiang, Y., Chen, X., 2013. Electrochim. Acta 97,

398–403.Chen, X., Wu, G., Chen, J., Xie, Z., Wang, X., 2011. J. Am. Chem. Soc. 133,

3693–3695.Claussen, J.C., Kumar, A., Jaroch, D.B., Khawaja, M.H., Hibbard, A.B., Porterfield, D.M.,

Fisher, T.S., 2012. Adv. Funct. Mater. 22, 3399–3405.Devaraj, M., Deivasigamani, R.K., Jayadevan, S., 2013. Anal. Methods 5,

3503–3515.Fan, Y., Liu, J.H., Lu, H.T., Zhang, Q., 2011. Colloids Surf. B 85, 289–292.Galus, Z., 1994. Fundamentals of Electrochemical AnalysisEllis Horwood,

New York.Ghadimi, H., Tehrani, R.M.A., Ali, A.S.M., Mohamed, N., Ghani, S.A., 2013. Anal. Chim.

Acta 765, 70–76.Goyal, R.N., Gupta, V.K., Chatterjee, S., 2010. Sens. Actuators B: Chem. 149,

252–258.Habibi, B., Jahanbakhshi, M., Pournaghiazar, M.H., 2011. Microchim. Acta 172,

147–154.Huang, J., Liu, Y., Hou, H., You, T., 2008. Biosens. Bioelectron. 24, 632–637.Kachoosangi, R.T., Wildgoose, G.G., Compton, R.G., 2008. Anal. Chim. Acta. 618,

54–60.Kang, X., Wang, J., Wu, H., Liu, J., Aksay, I.A., Lin, Y., 2010. Talanta 81, 754–759.Keeley, G.P., McEvoy, N., Nolan, H., Kumar, S., Rezvani, E., Holzinger, M., Cosnier, S.,

Duesberg, G.S., 2012. Anal. Methods 4, 2048–2053.Khanmohammadi, M., Soleimani, M., Morovvata, F., Bagheri Garmarudi, A., Kha-

lafbeigi, M., Ghasemi, K., 2012. Thermochim. Acta 530, 128–132.Khaskheli, A.R., Fischer, J., Barek, J., Vyskočil, V., Sirajuddin, Bhanger, M.I, 2013.

Electrochim. Acta 101, 238–242.Kutluay, A., Aslanoglu, M., 2013. Sens. Actuators B: Chem. 185, 398–404.Laviron, E., 1974. J. Electroanal. Chem. 52, 355–393.Laviron, E., 1979. J. Electroanal. Chem. 101, 19–28.Li, J., Kuang, D., Feng, Y., Zhang, F., Xu, Z., Liu, M., 2012. J. Hazard. Mater. 201–202,

250–259.Li, J., Kuang, D., Feng, Y., Zhang, F., Xu, Z., Liu, M., Wang, D., 2013. Biosens.

Bioelectron. 42, 198–206.Li, Z., Zhang, J., Zhou, Y., Shuang, S., Dong, C., Choi, M.M.F., 2012. Electrochim. Acta

76, 288–291.Liang, R.L., Cao, H.Q., Qian, D., Zhang, J.X., Qu, M.Z., 2011. J. Mater. Chem. 21,

17654–17657.Lourenção, B.C., Medeiros, R.A., Rocha-Filho, R.C., Mazo, L.H., Fatibello-Filho, O.,

2009. Talanta 78, 748–752.Love, D.C., Halden, R.U., Davis, M.F., Nachman, K.E., 2012. Environ. Sci. Technol. 46,

3795–3802.Lu, L.M., Li, H.B., Qu, F., Zhang, X.B., Shen, G.L., Yu, R.Q., 2011. Biosens. Bioelectron.

26, 3500–3504.Mazloum-Ardakani, M., Sheikh-Mohseni, M.A., Abdollahi-Alibeik, M., Benvidi, A.,

2012. Sens. Actuators B: Chem. 171–172, 380–386.Modick, H., Schütze, A., Pälmke, C., Weiss, T., Brüning, T., Koch, H.M., 2013. J.

Chromatogr. B 925, 33–39.Nair, S.S., John, S.A., Sagara, T., 2009. Electrochim. Acta 54, 6837–6843.Nandini, S., Nalini, S., Manjunatha, R., Shanmugam, S., Melo, J.S., Suresh, G.S., 2013. J.

Electroanal. Chem. 689, 233–242.Öcan, A., Şahin, Y., 2011. Anal. Chim. Acta 685, 9–14.Olivé-Monllau, R., Muñoz-Pascual, F.X., Baldrich, E., 2013. Sens. Actuators B: Chem.

185, 685–693.Palanisamy, S., Ku, S., Chen, S.M., 2013. Microchim. Acta 180, 1037–1042.Prabakar, S.J.R., Narayanan, S.S., 2007. Talanta 72, 1818–1827.Qiao, J., Guo, Y., Song, J., Zhang, Y., Sun, T., Shuang, S., Dong, C., 2013. Anal. Lett. 46,

1454–1465.Sultan, Maher, M.A., Alzoman, H.M., Alshehri, N.Z., Rizk, M.M., Elshahed, M.S., Olah,

I.V., M.S., 2013. J. Chromatogr. Sci. 51, 502–510.Švorc, Ľ., Sochr, J., Tomčík, P., Rievaj, M., Bustin, D, 2012. Electrochim. Acta 68,

227–234.Thiagarajan, S., Yang, R.F., Chen, S.M., 2009. Bioelectrochem. 75, 163–169.Tsierkezos, N.G., Othman, S.H., Ritter, U., 2013. Ionics 19, 1897–1905.Wang, Q., Cui, X., Chen, J., Zheng, X., Liu, C., Xue, T., Wang, H., Jin, Z., Qiao, L., Zheng,

W., 2012. RSC Adv. 2, 6245–6249.Wang, Q., Cui, X., Guan, W., Zheng, W., Chen, J., Zheng, X., Zhang, X., Liu, C., Xue, T.,

Wang, H., Jin, Z., Teng, H., 2013. J. Phys. Chem. Solids 74, 1470–1474.

J. Li et al. / Biosensors and Bioelectronics 54 (2014) 468–475474

Wu, G., Wu, Y., Liu, X., Rong, M., Chen, X., Chen, X., 2012. Anal. Chim. Acta 745,33–37.

Xu, F., Ru, H.Y., Sun, L.X., Zou, Y.J., Jiao, C.L., Wang, T.Y., Zhang, J.M., Zheng, Q., Zhou,H.Y., 2012. Biosens. Bioelectron. 38, 27–30.

Yang, G., Yu, L., Jia, J., Zhao, Z., 2012. J. Solid State Electrochem. 16, 1363–1368.Ye, D., Xu, Y., Luo, L., Ding, Y., Wang, Y., Liu, X., 2012. J. Solid State Electrochem. 16,

1635–1642.

Yin, H., Shang, K., Meng, X., Ai, S., 2011. Microchim. Acta 175, 39–46.You, J.M., Kim, D., Kim, S.K., Kim, M.S., Han, H.S., Jeon, S., 2013. Sens. Actuators B:

Chem. 178, 450–457.Zhang, Y., Luo, L., Ding, Y., Liu, X., Qian, Z., 2010. Microchim. Acta 171,

133–138.

J. Li et al. / Biosensors and Bioelectronics 54 (2014) 468–475 475