Development of a sensor prepared by entrapment of MIP particles in electrosynthesised polymer films for electrochemical detection of ephedrine

Available online at www.sciencedirect.com

Biosensors and Bioelectronics 23 (2008) 1152–1156

Short communication

Development of a sensor prepared by entrapment of MIP particles in electrosynthesised polymer films for electrochemical detection of ephedrine夽 E. Mazzotta a , R.A. Picca a , C. Malitesta a,∗ , S.A. Piletsky b , E.V. Piletska b a

Laboratorio di Chimica Analitica, Dipartimento di Scienza dei Materiali, Universit`a del Salento, Via Arnesano, 73100 Lecce, Italy b Cranfield Health, Cranfield University, Silsoe, Bedfordshire, MK45 4DT, UK

Received 25 July 2007; received in revised form 25 September 2007; accepted 27 September 2007 Available online 2 October 2007

Abstract A voltammetric sensor for (−)-ephedrine has been prepared by a novel approach based on immobilisation of an imprinted polymer for ephedrine (MIPE) in an electrosynthesised polypyrrole (PPY) film. Composite films were grown potentiostatically at 1.0 V vs. Pt (QRE) on a glassy carbon electrode using an unconventional “upside-down” (UD) geometry for the three-electrode cell. As a consequence, a high MIP loading was obtained, as revealed by SEM. The sensor response was evaluated, after overoxidation of PPY matrix, by cyclic voltammetry after pre-concentration in a buffered solution of analyte in 0.5–3 mM concentration range. An ephedrine peak at ≈0.9 V increasing with concentration and saturating at high concentrations was evident. PPY-modified electrode showed a response, which was distinctly lower than the MIP response for the same concentration of the template. The effect of potential interferences including compounds usually found in human fluids (ascorbic acid, uric acid, urea, glucose, sorbitol, glycine, dopamine) was examined. © 2007 Elsevier B.V. All rights reserved. Keywords: Molecular imprinting; Electrochemical sensor; Polypyrrole; Ephedrine

1. Introduction Molecular imprinting is a useful technique for the preparation of synthetic polymers with molecular recognition properties where binding sites specific for the template molecule are formed during the polymerisation process (Haupt and Mosbach, 2000). Molecularly imprinted polymers (MIPs) are used in several applications such as chromatography (Tamayo et al., 2005), solid phase extraction (Zhu et al., 2005), catalysis (Wulff, 2002), immunoassays (Lavignac et al., 2004) and sensors (see e.g. Kriz and Mosbach, 1995; Kriz et al., 1995; Dickert et al., 1998; Jakush et al., 1999; Malitesta et al., 1999; Peng et al., 2000; Blanco-Lopez et al., 2004; Shoji et al., 2003; Feng et al., 2004; Liao et al., 2004; Ho et al., 2005; Yeh and Ho, 2005; Ulyanova et al., 2006). MIP-based sensors attract increasing attention on

夽 ∗

Part of the Biosensors 2006 Special Issue. Corresponding author. E-mail address: [email protected] (C. Malitesta).

0956-5663/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.bios.2007.09.020

account of their many advantages such as low cost, very high specificity, stability and robustness. The use of MIPs in extreme environments, such as organic solvents or at high temperatures, makes them ideal recognition elements for sensors and a possible alternative to unstable natural receptors. A particularly important aspect in the design of a MIP sensor is MIP integration with the transducer. Many attempts have been made in the field in order to address this key issue. Among these, surface coating (spin and spray coating) is a method of synthesising and anchoring MIPs onto a transducer surface. This approach was firstly used with optical (Jakush et al., 1999) and acoustic (Dickert et al., 1998) transducers, and more recently with electrochemical sensors (Blanco-Lopez et al., 2004; Shoji et al., 2003). In this case, MIP synthesis and its deposition on the transducer surface take place in a single step, allowing an easy and less time-consuming sensor preparation. Another interesting approach is represented by the electropolymerisation of MIP films at the transducer surface, both in electrochemical (Blanco-Lopez et al., 2004; Liao et al., 2004;

E. Mazzotta et al. / Biosensors and Bioelectronics 23 (2008) 1152–1156

Yeh and Ho, 2005; Ulyanova et al., 2006) and piezoelectric (Malitesta et al., 1999; Peng et al., 2000; Feng et al., 2004; Liao et al., 2004) sensors. This is a simple and rapid way to deposit a uniform molecularly imprinted layer with good adherence to the transducer surface (Malitesta et al., 1999); moreover, it allows an easy control of the film thickness by varying the amount of circulated charge. In both approaches, best conditions for imprinting and entrapment stages cannot be separately selected, so, the simultaneous optimisation of parameters for both processes may be required. Entrapment of MIP particles into gels (Kriz and Mosbach, 1995) or membranes (Kriz et al., 1995) has been proposed for electrochemical transducers. Alternatively, a suspension of MIP particles and conductive materials (graphite or carbon black) in a solution of an inert soluble polymer (PVC) has been spin-coated on the electrode surface (Blanco-Lopez et al., 2004). However, these approaches suffer from some limitations such as slow diffusion kinetics, high response time, non-specific binding, suppressed binding capacity, difficult control of layer thickness and electrochemical area (Haupt and Mosbach, 2000; Blanco-Lopez et al., 2004). In this work, a further extension of the last approach is proposed, i.e. the realisation of an electrochemical sensor based on the immobilisation of chemically-synthesised MIP particles in an electrosynthesised polymeric matrix. This new strategy couples the advantages of electrochemical preparation to the one related to the decoupling of MIP synthesis and immobilisation steps. The first paper briefly describing this approach has been reported recently (Ho et al., 2005). In the present work, the approach was explored in relation to the electrochemical detection of ephedrine. (−)-Ephedrine is an alkaloid commonly employed as a stimulant, appetite suppressant, concentration aid, decongestant and to treat hypotension; it is widely used as an energising and stimulating substance for athletes and in various diet foods designed to excite the central nervous system. The extensive use of this chemical has generated a significant interest of international athletes associations and has encouraged the development of new reliable methods for its determination in health foods, pharmaceutical products and human fluids of athletes for anti-doping control (Nikolelis et al., 2005). Up to now, many powerful sophisticated analytical methods such as HPLC (Weiping et al., 2006) and GC (Van Eenoo et al., 2001) have been reported. Electrochemical techniques (Chicarro et al., 1993; Hernandez et al., 1997; Platts et al., 2006) have been also used partly employing different modified electrodes (Chicarro et al., 1994, 1995; Cookeas and Efstathiou, 2000; Nikolelis et al., 2005). They allow low detection limits but suffer from poor selectivity and/or analytical device preparation is often complicated and time-consuming. The present work describes the development of an ephedrine electrochemical sensor based on MIP-modified electrode. MIP for (−)-ephedrine (MIPE) has been incorporated into a polypyrrole (PPY) film potentiostatically grown on a glassy carbon electrode using an unconventional cell configuration. Microscopic properties of the composite films and analytical performances of the sensor are reported.

1153

2. Experimental 2.1. Materials MIPE was prepared according to the procedure reported elsewhere (Piletsky et al., 2001), (−)-ephedrine was supplied by Chemical Development, GlaxoSmithKline R&D; pyrrole, acetonitrile (ACN), tetrabutylammonium perchlorate (TBAP), alumina nanopowder were purchased from Aldrich and used as received. Ultra pure water (18.2 , Millipore) was used for buffer preparation. Britton–Robinson buffer at pH 9.2 was prepared using H3 BO3 0.04 M, H3 PO4 0.04 M, CH3 COOH 0.04 M, NaOH 0.2 M. Stock solutions (5 mM) of ephedrine were prepared in the buffer and stored under refrigeration. 2.2. Apparatus Electrochemical measurements were performed with a computer-controlled potentiostat CHI 620A (CH Instruments). 2.3. Preparation of MIP-modified electrode The preparation of MIPE/PPY-modified electrode was carried out with a three-electrode cell in “upside-down” (UD) geometry (Losito, 1997). In this configuration, the electrolytic solution was deposited on the surface of a 3 mm diameter glassy carbon (GC) working electrode. Two Pt wires worked as counter and quasi-reference electrode, respectively being immersed in the drop on the GC surface. In particular, a drop of MIPE suspension (10 mg/mL in 0.1 M TBAP in ACN) was first deposited on the GC electrode surface and then dried with a gentle nitrogen flux. A solution of pyrrole 0.15 M in 0.1 M TBAP in ACN was applied afterwards. PPY films were potentiostatically grown at 1.0 V until a 70 mC charge was passed, corresponding to a 5 ␮m film thickness (Rosenthal et al., 1985). After synthesis, MIPE/PPY-modified electrodes were thoroughly washed with ACN. Film overoxidation was then performed by cyclic voltammetry (CV) from −0.5 to 2.0 V at a 0.1 V/s scan rate in Britton–Robinson buffer (pH 9.2) with a conventional threeelectrode cell using the MIPE/PPY-modified electrode, a Pt wire and a SCE as working, counter and reference electrodes, respectively. The same protocol was followed for PPY-modified electrode except for MIPE particles deposition. Before film deposition and for its removal, the GC electrode was polished with alumina nanopowder and ultra pure water. 2.4. SEM characterisation SEM characterisation was performed with a JEOL 541OLV scanning electron microscope at 20 kV accelerating voltage and 80 ␮A current beam at pressure of 7 Pa. 2.5. Evaluation of sensor response For the evaluation of sensor response, the MIPE/PPYmodified electrode was placed in a vial filled with quiescent ephedrine solutions in Britton–Robinson buffer for 3 h. Five

1154

E. Mazzotta et al. / Biosensors and Bioelectronics 23 (2008) 1152–1156

concentrations in range 0.5–3 mM were tested. After contact, the modified electrode was simply washed with ultra pure water to remove the ephedrine loosely adsorbed on electrode surface. The electrochemical performance of sensor was evaluated by CV in −0.5 –1.2 V range at a scan rate of 0.1 V/s in Britton–Robinson buffer using a conventional three-electrode cell (WE: MIPE/PPY-modified electrode; RE: SCE; CE: Pt wire) (Chicarro et al., 1993). Also, unspecific PPY-modified electrode response was evaluated in the same way. To regenerate the sensor for further measurements, it was subjected to CV in the same conditions for 60 cycles, sufficient to eliminate analyte signal. Interfering compounds (ascorbic acid, uric acid, urea, glucose, sorbitol, glycine, dopamine) that could compromise ephedrine determination (Nikolelis et al., 2005) were examined to evaluate selectivity of sensor response. For this purpose, the same scheme described above for ephedrine was followed and the same experimental conditions used for analyte determination were adopted if not otherwise specified. 3. Results and discussion 3.1. Sensor preparation and SEM characterisation The UD configuration employed for the preparation of the modified electrode allows in principle a high MIP loading. Fig. 1 shows a large number of MIPE particles (larger than 10 ␮M) emerging from the 5 ␮M thick PPY film and roughly homogeneously distributed. This aspect is not evaluated in Ho et al.’s work (2005), where the conventional cell geometry employed should produce a lower loading. 3.2. Voltammetric detection of ephedrine A typical calibration curve for MIPE/PPY-modified electrode is shown in Fig. 2. It exhibits the expected plateau at high concentrations indicating saturation of MIPE binding sites.

Fig. 1. SEM micrographs of the MIPE/PPY film on the GC surface.

An example of voltammetric response to ephedrine 2 mM is shown in the inset. Cyclic voltammograms were recorded in Britton–Robinson buffer after 3 h contact with the analyte (the minimum time required to detect a sensor response). A longer incubation time (21 h) did not produce significant differences in ephedrine signal indicating that a 3 h time is enough for ephedrine to diffuse into the polymer and bind. Shorter incubation time could be obtained by using smaller MIPE particles. An anodic peak increasing with ephedrine concentration is observed at ≈0.9 V, which is not present before contact (dashed line). The lowest detectable concentration was 0.5 mM. This feature should allow the use of the sensor for the determination of ephedrine in anti-doping control (prohibited urine ephedrine concentration ≥60 ␮M) following a detection scheme employing a cleanup/pre-concentration step. The anodic peak is shifted toward higher potential for the MIP-modified electrode in comparison to the bare GC electrode (≈0.8 V); this result can be tentatively explained by the presence of the polymeric layer that limits diffusion to the electrode surface. Also, PPY-modified electrodes exhibit a response to analyte, probably captured by C O and COOH groups, formed during polymer overoxidation (Palmisano et al., 1995). These moieties can interact with nitrogen of ephedrine causing unspecific binding. However, PPY response to ephedrine is lower than the MIPE/PPY one (Fig. 3) for each studied concentration. In fact,

Fig. 2. Calibration curve for the MIPE/PPY-modified electrode. Each point refers to the mean of triplicate measurements. The same electrode was employed for the full set of data. Inset: CV (first cycle) sensor response to (−)-ephedrine 2 mM after 3 h contact. Scan rate 0.1 V/s.

E. Mazzotta et al. / Biosensors and Bioelectronics 23 (2008) 1152–1156

1155

0.44, respectively. Uric acid was clearly the strongest interferent. In order to enhance MIPE/PPY selectivity, we changed the pH (7.0 instead of 9.2) (Centonze et al., 1992) of Britton–Robinson buffer used for PPY overoxidation. This allowed a remarkable decrease in peak height for interfering compound and, consequently, a drastic reduction of interference ratio from 2.8 to 0.43. Nevertheless, further efforts are necessary for total suppression of PPY response to uric acid. Interference analysis accomplished in the present work represents another important improvement in respect of Ho et al.’s study (2005), where selectivity was evaluated using a compound that does not actually interfere with the analyte detection, since it is not electroactive in the adopted experimental conditions. 4. Conclusions Fig. 3. Comparison between MIPE/PPY (solid line) and PPY (dotted line) response to ephedrine 1 mM. Scan rate 0.1 V/s.

the ratio between MIPE/PPY and PPY responses is 1.7 in average, indicating the fundamental role played by the imprinted polymer in determining the recognition capabilities of sensor. The increase in sensor response observed for imprinted electrode most likely resulted from pre-concentration of the template by entrapped MIP. This result represents a distinct improvement in comparison to Ho et al.’s paper (2005) where a maximum ratio between responses obtained for MIP and blank polymers was 1.3. The MIPE/PPY sensor appears quite stable, since after 3 days its response decreased only by 6%. An average repeatability (same electrode) of 4% was observed. Average reproducibility (different electrodes) was distinctly higher (43%). While the variation in actual amount of MIPE entrapped in the polymeric matrix could be responsible for the observed STD, the possible different extent of PPY matrix overoxidation, leading to a different degree of non-specific binding could be a contributing factor. This can be seen from the average reproducibility of PPY response (measured for different electrodes) where STD is 36%.

A new MIP-based electrochemical sensor for ephedrine has been prepared by the immobilisation of chemically-synthesised MIP particles in an electrosynthesised and overoxidised PPY matrix. This novel approach decouples the synthesis of MIP and its integration with the transducer surface; as a result, the two steps can be separately optimised. Moreover, in order to achieve a high MIP loading on the electrode surface, an unconventional UD configuration cell has been used. SEM micrographs of the as-prepared composite films have revealed the successful MIP entrapment. The assembled sensor has shown promising results in the ephedrine detection at mM concentrations, but it suffers from some unspecific binding of ephedrine and interfering compounds by PPY matrix. The proposed sensor exhibits a significant selectivity being able to reject ascorbic acid, urea, glucose, sorbitol and glycine. Work is in progress to eliminate the residual interference of uric acid and dopamine. Moreover, further work is necessary to reduce or fully eliminate unspecific contribution to sensor response by employing other electropolymerisable monomers or chemical derivatisation of PPY, in order to inactivate chemical groups responsible for the unspecific binding. All this optimisation work will certainly improve detection limit also.

3.3. Interference study

References

Interferences were evaluated by analysing the CV sensor response to compounds commonly found in real samples together with ephedrine. In all cases, apart of uric acid and dopamine, no significant interference was observed. It should be mentioned that among interferents tested, only ascorbic acid, uric acid and dopamine are electroactive under employed experimental conditions. In fact, they show an oxidation peak at 0.3, 0.4 and 0.2 V vs. SCE on bare GC, respectively. Thus, these compounds could act as interferents in amperometric detection of ephedrine at 0.9 V, but not in voltammetric analysis, since separation between peaks is large enough to allow the ephedrine signal to be discriminated from interferents. Interference ratio (ip interference /ip ephedrine ), evaluated at the same concentration (0.5 mM) for interference and for ephedrine, resulted equal to zero for each tested interference except for uric acid and for dopamine, for which it was equal to 2.8 and

Blanco-Lopez, M.C., Gutierrez-Fernandez, S., Lobo-Castanon, M.J., MirandaOrdieres, A.J., Tunon-Blanco, P., 2004. Anal. Bioanal. Chem. 378, 1992. Centonze, D., Guerrieri, A., Malitesta, C., Palmisano, F., Zambonin, P.G., 1992. Fresenius’ J. Anal. Chem. 342, 729. Chicarro, M., Zapardiel, A., Bermejo, E., Perez, J.A., Hernandez, L., 1993. Anal. Chim. Acta 273, 361. Chicarro, M., Zapardiel, A., Bermejo, E., Perez, J.A., Hernandez, L., 1994. Anal. Lett. 27, 1809. Chicarro, M., Zapardiel, A., Bermejo, E., Perez, J.A., Hernandez, L., 1995. Analusis 23, 131. Cookeas, E.G., Efstathiou, C.E., 2000. Analyst 125, 1147. Dickert, F.L., Forth, P., Lieberzeit, P., Tortschanoff, M., 1998. Fresenius’ J. Anal. Chem. 360, 759. Feng, L., Liu, Y., Tan, Y., Hu, J., 2004. Biosens. Bioelectron. 19, 1513. Haupt, K., Mosbach, K., 2000. Chem. Rev. 100, 2495. Hernandez, L., Zapardiel, A., Bermejo, E., Perez, J.A., Chicarro, M., Garijo, M.J., 1997. Electroanalysis 9, 1214. Ho, K.-C., Yeh, W.-M., Tung, T.-S., Liao, J.-Y., 2005. Anal. Chim. Acta 542, 90.

1156

E. Mazzotta et al. / Biosensors and Bioelectronics 23 (2008) 1152–1156

Jakush, M., Janotta, M., Mizaikoff, B., Mosbach, K., Haupt, K., 1999. Anal. Chem. 71, 4786. Kriz, D., Mosbach, K., 1995. Anal. Chim. Acta 300, 71. Kriz, D., Ramstrom, O., Svensson, A., Mosbach, K., 1995. Anal. Chem. 67, 2142. Lavignac, N., Allender, C.J., Brain, K.R., 2004. Anal. Chim. Acta 510, 139. Liao, H., Zhang, Z., Li, H., Nie, L., Yao, S., 2004. Electrochim. Acta 49, 4101. Losito, I., 1997. PhD Thesis, 40. Bari. Malitesta, C., Losito, I., Zambonin, P.G., 1999. Anal. Chem. 71, 1366. Nikolelis, D.P., Raftopoulou, G., Siontorou, C.G., 2005. Electroanalysis 17, 1870. Palmisano, F., Malitesta, C., Centonze, D., Zambonin, P.G., 1995. Anal. Chem. 67, 2207. Peng, H., Liang, C.D., Zhou, A.H., Zhang, Y.Y., Xie, Q.J., Yao, S.Z., 2000. Anal. Chim. Acta 423, 1513. Piletsky, S.A., Karim, K., Piletska, E.V., Day, C.J., Freebairn, K.W., Legge, C., Turner, A.P.F., 2001. Analyst 126, 1826.

Platts, M., Smith, R.B., Mould, N., Davis, J., 2006. Electrochem. Commun. 8, 633. Rosenthal, M.V., Skotheim, T.A., Melo, A., Florit, M.I., 1985. J. Electroanal. Chem. 185, 297. Shoji, R., Takeuchi, T., Kubo, I., 2003. Anal. Chem. 75, 4882. Tamayo, F.G., Titirici, M.M., Martin-Esteban, A., Sellergren, B., 2005. Anal. Chim. Acta 542, 38. Ulyanova, Y.V., Blackwell, A.E., Minteer, S.D., 2006. Analyst 131, 257. Van Eenoo, P., Delbeke, F.T., Roels, K., De Backer, P., 2001. J. Chromatogr. B 760, 255. Weiping, W., Cunhong, L., Ying, L., Zhide, H., Xingguo, C., 2006. J. Chromatogr. A 1102, 273. Wulff, G., 2002. Chem. Rev. 102, 1. Yeh, W.-M., Ho, K.-C., 2005. Anal. Chim. Acta 542, 76. Zhu, X., Yang, J., Su, Q., Cai, J., Gao, Y., 2005. J. Chromatogr. A 1092, 161.