Field-Effect-Transistor Sensor Based on Enzyme-Functionalized

Jul 23, 2008 - We describe the detection of glucose based on a liquid-ion gated field-effect transistor configuration in which enzyme-functionalized p...
32 downloads 14 Views 1MB Size
Download PDF
9992

J. Phys. Chem. B 2008, 112, 9992–9997

Field-Effect-Transistor Sensor Based on Enzyme-Functionalized Polypyrrole Nanotubes for Glucose Detection Hyeonseok Yoon, Sungrok Ko, and Jyongsik Jang* School of Chemical and Biological Engineering, Seoul National UniVersity, 599 Gwanangno, Gwanakgu, Seoul 151-742, Korea ReceiVed: January 21, 2008; ReVised Manuscript ReceiVed: April 23, 2008

We describe the detection of glucose based on a liquid-ion gated field-effect transistor configuration in which enzyme-functionalized polypyrrole nanotubes are employed as the conductive channel. First of all, carboxylated polypyrrole nanotubes (CPNTs) were successfully fabricated by the chemical polymerization of an intrinsically functionalized monomer (pyrrole-3-carboxylic acid, P3CA) without degradation in major physical properties. The CPNTs possessed not only well-defined functional groups but also electrical properties comparable to nonsubstituted polypyrrole. Importantly, the carboxylic acid functional group can be utilized for various chemical and biological functionalizations. A liquid-ion gated FET sensor was readily constructed on the basis of the chemical functionality of CPNTs. In the first stage, the CPNTs were immobilized onto a microelectrode substrate via covalent linkages. It was noteworthy that the covalent immobilization allowed high-quality contact between the nanotubes and the microelectrodes in the liquid phase. The second stage involved the covalent binding of glucose oxidase (GOx) enzyme to the nanotubes. The covalent functionalization generally provides excellent enzymatic activity and thermal stability. The fabricated FET sensor provided real-time response (an increase in source-drain current) and high sensitivity toward the various concentrations (0.5-20 mM) of glucose. The enzymatic reaction product, hydrogen peroxide, played pivotal roles in modulating the charge transport property of CPNTs. Introduction Incorporation of molecular recognition elements into signal transducers makes it possible to pinpoint the presence of target molecules. Versatile signal transducers coupled with bioreceptors such as antibodies, enzymes, nucleic acids, cells, and tissues have been employed for biocatalysis-, bioaffinity-, and microbebased sensor systems.1–4 A notable example involves enzymebased electrochemical glucose detection for diagnosing diabetes.5–7 The majority of glucose biosensors are based on amperometric detection with glucose oxidase (GOx) enzyme electrodes.8–11 GOx is a homodimer containing two flavine adenine dinucleotide (FAD) moieties as the cofactor. In an amperometric glucose sensor, the current is a direct measure of the rate of the enzymatic regeneration reaction carried out by oxygen or artificial redox mediators, providing quantitative information on the glucose level. Conducting polymers such as polyaniline, polypyrrole, and poly(3,4-ethylenedioxythiophene) have been utilized as the immobilization matrix for enzymes in glucose sensors.8–13 The enzymes are incorporated into a conducting polymer layer on the electrode via covalent attachment, adsorption, entrapment, and so forth. In particular, the covalent attachment is an attractive approach for achieving excellent enzymatic activity and thermal stability. Field-effect transistors (FETs) have attracted increasing interest as primary candidates for fabricating state-of-the art sensor platforms because they are capable of achieving high current amplification and sustaining an enhanced signal-to-noise ratio.7,14–16 Typically, the integration of a single carbon nanotube (CNT) with an FET has opened up new routes toward the development of highly efficient sensors.17 In most cases, the * Corresponding author. Telephone: 82-2-880-7069. Fax: 82-2-888-7295. [email protected].

FETs based on a single CNT displayed improved sensor performance compared with conventional counterparts. Unfortunately, however, the use of a single CNT has significant drawbacks in terms of surface functionality and position controllability, often resulting in low device yield and irreproducible sensor response. Conducting polymers have been widely utilized in the areas of bioanalytical science due to their inherent charge transport properties and biocompatibility.18 Recently, remarkable progress in synthesis and characterization of conducting polymer nanomaterials has offered a great possibility for novel applications in various fields.19–25 In particular, onedimensional (1D) conducting polymer nanomaterials have attracted much attention as promising building blocks for FET sensor applications.26–28 Compared with conventional films, 1D conducting polymer nanomaterials have remarkable characteristics derived from anisotropic electronic properties, high surface area, and small dimensions.9,10,29–31 When used as the conductive channel of FETs, 1D conducting polymer nanomaterials can offer highly sensitive responses through depletion or accumulation of charge carriers in the bulk of the nanomaterials (versus only the surface of conventional films). However, since the detection of biological species is commonly carried out in solution, conducting polymer nanomaterials need a delicate and time-consuming process for adhering to a patterned electrode. The interaction between polymer transducers and bioreceptors has also a major impact on the quality and performance of biosensors. Therefore, it is important to develop an efficient fabrication route to biosensor platforms based on polymeric nanomaterial transducers. Herein, we present the successful demonstration of a liquidion gated FET sensor using enzyme-conjugated polypyrrole nanotubes to detect the presence of glucose. We describe a simple and versatile approach, based on the use of a function-

10.1021/jp800567h CCC: $40.75  2008 American Chemical Society Published on Web 07/23/2008

FET Sensor Based on GOx-CPNTs alized repeating unit, for immobilizing the nanomaterials onto an electrode substrate and binding biomolecules to conducting polymer nanomaterials. This strategy takes advantage of the inherent versatility of functionalized repeating units, which allows chemical functionalization without any sophisticated surface treatment and degradation in physical properties. Having demonstrated the feasibility of the above strategy, we applied it to construct an enzymatic FET sensor platform for detecting glucose. As mentioned above, most glucose sensors have depended on amperometric detection using GOx electrodes and conducting polymer films have been commonly utilized as the immobilization matrix for GOx. Therefore, to the best of our knowledge, this is the first example of an enzyme FET sensor based on functionalized conducting polymer nanomaterials. Experimental Methods Pyrrole-3-carboxylic acid (P3CA) was purchased from Acros Organics. β-D-glucose (97%) and GOx (from Aspergillus niger, 15 500 units g-1) were obtained from Sigma. Horseradish peroxidase (113 purpurogallin units mg-1) and o-dianisidine were purchased from Sigma. Carboxylated polypyrrole nanotubes (CPNTs) were prepared by vapor deposition polymerization inside the pores of an alumina membrane (100 nm pore diameter).32,33 Iron cations (oxidizing agent) were adsorbed within the pores of the alumina membrane by immersing the membrane in 0.1 M aqueous ferric chloride solution. The iron-adsorbed membrane was placed in a reactor containing 0.18 mmol of pyrrole-3-carboxylic acid (P3CA). The internal pressure was reduced to 10-2 Torr (1 Torr ) 133.322 Pa) for the vaporization of P3CA, and the vapor deposition polymerization proceeded at 150 °C for 5 h. Subsequently, the alumina membrane was removed with 5 M HCl solution and the resulting product was obtained after washing with distilled water. Transmission electron microscopy (TEM) images were obtained with a JEOL EM-2000 EXII microscope. The samples were dispersed in an ethanol solution and were deposited on a carbon mesh foil supported on a copper grid. Scanning electron microscopy (SEM) observation was performed with a JEOL JSM-6700. A specimen was coated with a thin layer of gold to eliminate charging effects. Fourier transform infrared (FT-IR) spectra were taken with a Bomem MB 100 spectrometer. An efficient condensing agent, 4-(4,6-dimethoxy-1,3,5-triazin2-yl)-4-methylmorpholinium chloride (DMT-MM), was synthesized.34 Briefly, 2-chloro-4,6-dimethoxy-1,3,5-triazine was reacted with N-methylmorpholine in tetrahydrofuran (THF) for 30 min (by 1:1 molar ratio). The resulting white precipitates were washed with THF, and then completely dried. The final product was stored in a freezer under -20 °C. Importantly, DMT-MM is capable of providing excellent yields in water as well as alcohol phases for the condensation reaction between carboxylic acid and amine groups. A microarray consisting of a pair of gold interdigitated microelectrodes with 40 fingers (finger dimensions: width 10 µm, length 4000 µm, interfinger spacing 10 µm) was patterned on a glass substrate via a photolithographic process. The microelectrode substrate was treated with 5 wt % aqueous (3aminopropyl)trimethoxysilane solution for 6 h and then exposed to a mixture of 0.1 wt % CPNT solution (10 µL) and 10 wt % DMT-MM solution (10 µL) for 12 h. CPNTs were immobilized on the microelectrode substrate through condensation reactions between carboxyl groups of CPNTs and amino groups of the substrate surface. The resulting CPNTs-immobilized microelectrode substrate was rinsed with distilled water. The coupling

J. Phys. Chem. B, Vol. 112, No. 32, 2008 9993 reaction for covalent binding of GOx enzyme to CPNTs proceeded using a mixture of 10 wt % GOx solution (4 µL) and 10 wt % DMT-MM solution (10 µL) for 12 h. Subsequently, the microelectrode substrate was rinsed with distilled water and dried with a stream of nitrogen gas. A colorimetric assay was performed to calculate the loading amount of GOx on the nanotubes.35 First, the microelectrode substrate was immersed in 1.0 mL of glucose solution (0.1 M) for 10 min, and then a mixture of 0.1 mL of 0.2 wt % peroxidase solution and 0.2 mL of 0.2 mM o-dianisidine solution were added into the solution. The mixture solution was incubated for 5 min, and afterward, the absorbance of the solution was measured at 416 nm (the maximum absorption band of H2O2) using an ultraviolet-visible spectrometer (Perkin-Elmer Lambda 20). A standard curve of absorbance versus H2O2 concentration was plotted and used to determine the loading amount of GOx. The whole reactions were carried out in a pH 7 phosphate buffer solution at 25 °C. A solution chamber (volume 10 mL) was designed and employed for all solution-based measurements. The FET sensor substrate was immersed with a Pt counter electrode and Ag/ AgCl reference electrode in 10 mM phosphate buffered saline solution (pH 7.0). All electrical measurements were conducted with a Keithley 2400 SourceMeter and a Wonatech WBCS 3000 potentiostat. The response time was defined as the time required for sensor signal to reach 90% of the maximum value (sensitivity). Results and Discussion Above all, CPNTs were fabricated via vapor deposition polymerization combined with template synthesis. The intrinsically functionalized monomer, i.e., P3CA, was chemically polymerized inside the cylindrical pores of an alumina membrane. The ability of P3CA repeating unit to delocalize π-electrons along polymer chains is comparable to that of nonsubstituted pyrrole, presenting a great advantage over Nand 2-substituted pyrroles from the viewpoint of charge transport. In addition, the CPNTs consisting of P3CA repeating units offer a defined number of homogeneous carboxyl groups on their surface. Importantly, these surface functional groups can act as chemically reactive sites for covalent functionalization. Figure 1 shows schematically the overall reaction steps for the immobilization of CPNTs onto a microelectrode substrate, followed by the functionalization of the CPNTs with GOx. First, the microelectrode substrate was readily modified with amino groups using an aminosilane coupling agent (3-aminopropyltrimethoxysilane, APS). Subsequently, the CPNTs were anchored onto the substrate through a condensation reaction between carboxyl moieties of the nanotubes and amine moieties of the substrate surface. GOx contains terminal amino groups on its lysine residues, and thus a similar condensation reaction can be made for linking GOx enzyme to CPNTs. Finally, CPNTs could be covalently functionalized with GOx via a condensation reaction using DMT-MM. TEM observation was carried out to directly demonstrate the GOx functionalization of CPNTs through the condensation reaction. CPNTs were functionalized with GOx in an aqueous solution using the same recipe. Figure 2 displays TEM images of CPNTs before and after the immobilization of GOx on the nanotubes. Before GOx immobilization (Figure 2a), CPNTs had a smooth outer surface, which corresponded to a common feature of nanomaterials prepared from template synthesis, and their average wall thickness was 20 nm. On the other hand, the surface of CPNTs became considerably rugged after GOx immobilization, and the wall thickness also increased by 5-10 nm (Figure 2b). Since

9994 J. Phys. Chem. B, Vol. 112, No. 32, 2008

Yoon et al.

Figure 1. Schematic illustration of reaction steps for the fabrication of a sensor platform based on CPPy nanotubes: (a) microelectrode substrate, (b) aminosilane-treated substrate, (c) immobilization of the nanotubes onto a substrate, and (d) binding of GOx to the nanotubes.

Figure 3. FT-IR spectra of (a) DMT-MM, (b) GOx, (c) CPNTs, and (d) GOx-CPNTs.

Figure 2. Typical TEM images of CPNTs (a) before and (b) after GOx immobilization. For GOx immobilization, 10 µL of an aqueous solution containing 0.1 wt % CPNT was reacted with 4 µL of 10 wt % GOx solution and 10 µL of 10 wt % DMT-MM solution for 12 h.

the size of a dry GOx molecule is ca. 2-3 nm,36 it is evident that GOx molecules are incorporated into the polymer chains of CPNTs. Chemical characterization of the samples was carried out using FT-IR spectroscopy (Figure 3). First, FT-IR spectrum of the condensing agent, DMT-MM, displayed characteristic peaks at 3450, 3012, 2950-2875, and 1600-1500 cm-1, which were

assigned to stretching vibrations of amine groups, aromatic ring, C-H, and C-N in DMT-MM, respectively (Figure 3a). For GOx enzyme (Figure 3b), the absorption peaks at 2930 and 1650 cm-1 were attributed to stretching vibrations of alkyl and carbonyl groups, and the absorption peaks in the range of 1595-1000 cm-1 were ascribed to bending vibrations of amide groups, methyl and methylene groups, and aromatic rings in GOx. The FT-IR spectrum of CPNTs indicated a carboxylic acid stretching peak at 1701 cm-1, a pyrrole stretching peak at 1548 cm-1, and a conjugated C-N stretching peak at 1473 cm-1 (Figure 3c). When GOx was immobilized onto CPNTs, the absorption peaks originating from amide I (at 1630 cm-1) and amide II (at 1525 cm-1) were indistinguishable in the spectrum of GOx-CPNTs due to the superposition of stretching vibrations from several chemical groups (Figure 3d). However, the lowintensity peaks at 2930, 1448, and 1105 cm-1 were observed in the spectrum of GOx-CPNTs, which corresponded to the characteristic peaks of GOx. This fact indicated that GOx was effectively immobilized on the CPNTs. The nanotubes form networks on the microelectrode substrate, as shown in Figure 4a. Compared with a single nanotube, the nanotube networks offer remarkable advantages such as great electrical reliability and simple fabrication.37 To estimate the

FET Sensor Based on GOx-CPNTs

Figure 4. (a) Typical FE-SEM image of CPNTs deposited on the interdigitated microelectrodes and (b) I-V characteristics of CPNTs before and after GOx immobilization (scan rate 10 mV s-1).

number of CPNTs deposited on the interdigitated microelectrode array (40 fingers, interfinger gap 10 µm), 10 SEM images were taken from different areas of the electrodes. The nanotubes in these images were counted, and the resulting number of nanotubes was normalized to the total electrode area. From this procedure, it was determined that the nanotube coverage was approximately 1 nanotube per 85 µm2. The area of the interdigitated microelectrode array used was 4000 × 1600 µm2. The CPNTs immobilized were retained more than 90% without deformation even after sonication for 30 s, indicating strong anchoring of the nanotubes to the microelectrode substrate. The loading amount of GOx on the nanotubes calculated using a colorimetric assay was 5.4 × 10-3 unit, where 1 unit of enzyme oxidizes 1.0 µmol of β-D-glucose to D-gluconolactone and H2O2. This loading amount is considerably higher than that of other conventional enzyme electrodes,38,39 because the CPNTs offer a high surface area as well as a dense array of surface functionality. The high enzyme loading on the microelectrode substrate enhances the rate of turnover of the enzyme, providing a possibility toward high-performance miniature sensors. Figure 4b displays current-voltage (I-V) characteristics of CPNTs and GOx-functionalized CPNTs (GOx-CPNTs) deposited on the interdigitated microelectrode array. CPNTs with no GOx immobilized on their surface exhibited an ohmic contact and their dI/dV value was as high as 4.2 × 10-5 S. The dI/dV value slightly decreased to 2.0 × 10-5 S after GOx immobilization. For GOx immobilization, CPNTs inevitably went through a condensation reaction, followed by a washing step. During the process, the oxidation level of CPNTs may be affected by the chemicals used, and the counterions doped can be physically detached from the polymer backbone during the washing process. Consequently, it is believed that these possible events played a role in causing a change in the dI/dV. Nevertheless, after GOx immobilization, the linearity in I-V characteristic was still preserved and the dI/dV value was also maintained within the same order of magnitude. Therefore, it can be considered that GOx molecules are effectively incorporated into CPNTs without any deterioration in electrical contact and conductivity. An FET sensor configuration was constructed with the interdigitated microelectrode array using a phosphate-buffered

J. Phys. Chem. B, Vol. 112, No. 32, 2008 9995

Figure 5. (a) Schematic representation of an enzyme FET sensor platform based on GOx-CPNTs. The source (S), drain (D), liquid-ion gate (G), Ag/AgCl reference electrode (R), and Pt counter electrode (C) are labeled. (b) ISD-EG curve of the FET measured at (10 mV s-1 (VSD ) -10 mV) (inset: expanded ISD-EG curve between 0 and 100 mV). The redox state of CPNTs depending on the EG is shown as well as the working potential range suitable for a sensing test.

electrolyte (pH 7.0) as a liquid-ion gate. Figure 5a schematically depicts an enzyme FET sensor platform based on GOx-CPNTs. As described above, GOx-CPNTs were immobilized via covalent linkages on the microelectrode substrate. Two gold interdigitated microelectrode bands were employed as source (S) and drain (D) electrodes, respectively. In addition, a reference electrode (R) and a counter electrode (C) were immersed in the electrolyte, and the gate potential (EG) was applied between the reference electrode and the source electrode through the buffer solution. The applied EG is dependent on the distance between the reference electrode and the source electrode due to the resistivity of the buffer solution. Therefore, the distance between the electrodes was fixed at 2 mm. In order to characterize the charge transport property of GOx-CPNTs in the FET configuration, the dependence of the source-drain current (ISD) versus EG was investigated at a constant source-drain voltage (VSD). As shown in Figure 5b, the ISD-EG curve was swept from negative EG to positive EG and back to negative EG. Overall, the ISD was modulated by varying EG in a p-type FET response mode. GOx-CPNT displayed an irreversible single peak at approximately +0.9 V, due to the oxidation of the polymer backbone, which was comparable to that of nonsubstituted polypyrrole.40 This result suggests that the 3-substitution of pyrrole repeating units has little effect on their electrochemical properties. The change in ISD was reasonably linear in the EG range between -50 and 400 mV, indicating that there were few side reactions preceding and/or following the charge transfer. The positive EG is favorable to the oxidation reaction of glucose and H2O2 although it gives rise to the dedoping of conducting polymers. Therefore, the positive yet low EG values (0-100 mV) may be suitable for a glucose sensing test. In this bias range, the transconductance (gm ) dISD/dEG) ranged from 0.4 × 10-5 to 1.2 × 10-5 S. Figure 6 displays the response of a GOx-CPNT FET sensor upon exposure to glucose. The ISD was monitored in real time

9996 J. Phys. Chem. B, Vol. 112, No. 32, 2008

Figure 6. Response of a GOx-CPNT FET sensor to glucose at VSD ) -10 mV: (a) real-time ISD change upon consecutive addition of 2-20 mM glucose; (b) calibration curve of sensitivity versus glucose concentration. (The sensitivity was determined as the maximum ∆I/I0 recorded after exposure to GOx).

at VSD ) -10 mV (EG ) 10 mV), a low operating voltage. As shown in Figure 6a, various concentrations of glucose (2-20 mM) were added into the buffer solution. Upon each addition of glucose, the ISD increased rapidly and then decreased gradually. The decrease in ISD change toward the initial level is attributed to the reduction (deprotonation) of CPNTs by the positive EG. As mentioned above, applying a positive EG causes the electrochemical dedoping of conducting polymers by driving cations from the electrolyte to the conducting polymers, resulting in the decrease in ISD. The response time was approximately 5-10 s, which was tremendously faster than that (ca. 10 min) of a 10 µm gap electrode coated with a conventional polypyrrole film.41 The modulation in ISD was dependent on the concentration of glucose. Figure 6b exhibits the sensitivity change of the FET sensor as a function of glucose concentration (0.5-20 mM): these concentrations encompass normal (3.6-6.1 mM) as well as abnormal (20 mM) glucose levels in human blood.7 The sensitivity was determined as the normalized current change ([∆I/I0]SD) measured when the maximum value is reached after exposure to GOx. The sensitivity became higher with increasing concentration of glucose solutions and tended to slightly saturate at concentrations of more than 20 mM. Compared with existing glucose sensors based on CNTs,36,42 the GOx-CPNT FET sensor presented 1-2 orders of magnitude higher sensitivity. For example, the GOx-CPNT FET sensor had a sensitivity of 1.7 for 2 mM glucose while a previously reported glucose sensor based on CNT nanoelectrodes gave a sensitivity of 0.2 for the same concentration.42 A control experiment was performed using CPNTs with no GOx attached. The FET sensor did not give any remarkable changes in ISD when exposed to glucose. This result demonstrates that the sensor response originates from the specific interaction between GOx and glucose, not direct electrochemical oxidation of glucose. The enzyme GOx catalyzes the oxidation of glucose. The enzymatic reactions can temporarily change the protonation state of conducting polymers and thus affect the ISD in a manner

Yoon et al.

Figure 7. Response of a GOx-CPNT FET sensor to H2O2 at VSD ) -10 mV: (a) real-time ISD change upon consecutive addition of 2-20 mM H2O2; (b) calibration curve of sensitivity versus H2O2 concentration. (The sensitivity was determined as the maximum ∆I/I0 recorded after exposure to H2O2).

similar to the effect of applying EG. The general reaction steps are as follows:

GOx(FAD) + β-D-glucose f GOx(FADH2) + D-glucono-1,5-lactone (1) GOx(FADH2) + O2 f GOx(FAD) + H2O2 (2) D-glucono-1,5-lactone + H2O f D-gluconic acid (3) The GOx converts glucose into gluconolactone through reduction of the FAD prosthetic group. The reduced cofactor (FADH2) is reoxidized by the final electron acceptor (oxygen), yielding H2O2 as a reaction product. The generated H2O2 can alter the charge transport property of conducting polymers. Most glucose sensors are based on the detection of H2O2.5 Accordingly, first of all, the response of the FET sensor to H2O2 was investigated to clarify the sensing mechanism. Figure 7 exhibits the response of a GOx-CPNT FET sensor to H2O2. The ISD was found to be extremely sensitive to the presence of H2O2. As illustrated in Figure 7a, the FET sensor revealed an abrupt increase in ISD when exposed to H2O2. In Figure 7b, the calibration curve shows that more prominent signals were detected with increasing H2O2 concentration. The applied EG (10 mV) is below the oxidation potential of H2O2, and thus the modulation in ISD is not due to the electrochemical oxidation of H2O2. Owing to its oxidizing properties, H2O2 can play a role in enhancing the oxidation level ([N+]/[N] ratio) of polypyrrole backbone, facilitating charge transport in the material. During the enzymatic reaction, GOx produces H2O2 in the presence of oxygen. When a control experiment was performed under anaerobic condition, the FET sensor was relatively insensitive to glucose. Therefore, it is evident that the H2O2 generated has an effect on the change in ISD. In the enzymatic reaction cycle, the gluconolactone intermediate undergoes hydrolysis to give gluconic acid. The gluconic acid (C6H12O7), which forms gluconate anion (C6H11O7-) in

FET Sensor Based on GOx-CPNTs

Figure 8. Response of a GOx-CPNT FET sensor to gluconic acid at VSD ) -10 mV: real-time ISD change upon consecutive addition of 5-50 mM gluconic acid.

water, can play the role of a counterion for polypyrrole and acidify the CPNT/gate interface through proton dissociation.43 As shown in Figure 8, however, the FET sensor had no appreciable signals up to 10 mM and then detectable signals at higher concentrations when exposed to gluconic acid. Compared with the response of the FET sensor to H2O2 at the same concentration, its response to gluconic acid was negligible, especially at low concentrations. Judging from this result, it is considered that the gluconic acid made little contribution to the change in ISD and the H2O2 played a crucial role in modulating ISD. Conclusions An enzyme FET sensor for glucose detection was fabricated on the basis of the inherent functionality of CPNTs. CPNTs were covalently anchored on the FET substrate for high-quality electrical contact, and then GOx enzyme was readily tethered onto the nanotubes via covalent linkages without sophisticated surface treatment. This simple yet versatile approach may open up a new route to the fabrication of various types of biosensor platforms based on polymer nanomaterials. The enzyme FET sensor gave an increase in ISD upon exposure to glucose. The sensing mechanism involved the variations in the protonation state of CPNTs by the H2O2 generated during the enzymatic reaction. Importantly, the high surface area and abundant functional groups of CPNTs allow the loading of large amounts of enzyme on the microelectrode substrate, and the FET configuration can amplify the change in ISD. Therefore, it is expected that the enzyme FET sensors based on functionalized polymer nanomaterials may be a promising candidate for highperformance miniature sensors after further optimization. Acknowledgment. This work was supported by grants from the Center for Advanced Materials Processing under the 21C Frontier Programs and from the Fundamental R&D Program for Core Technology of Materials funded by the Ministry of Knowledge Economy, Republic of Korea. References and Notes (1) Katz, E.; Willner, I. ChemPhysChem 2004, 5, 1084.

J. Phys. Chem. B, Vol. 112, No. 32, 2008 9997 (2) Wallace, G. G.; Kane-Maguire, L. A. P. AdV. Mater. 2002, 14, 953. (3) Wang, J. Small 2005, 1, 1036. (4) Song, C.; Pehrsson, P. E.; Zhao, W. J. Mater. Res. 2006, 21, 2817. (5) Wang, J. Chem. ReV. 2007, 108, 814. (6) Zayats, M.; Kharitonov, A. B.; Katz, E.; Bu¨ckmann, A. F.; Willner, I. Biosens. Bioelectron. 2000, 15, 671. (7) Forzani, E. S.; Zhang, H.; Nagahara, L. A.; Amlani, I.; Tsui, R.; Tao, N. Nano Lett. 2004, 4, 1785. (8) Zayats, M.; Katz, E.; Baron, R.; Willner, I. J. Am. Chem. Soc. 2005, 127, 12400. (9) Ekanayake, E. M. I. M.; Preethichandra, D. M. G.; Kaneto, K. Biosens. Bioelectron. 2007, 23, 107. (10) Li, J.; Lin, X. Biosens. Bioelectron. 2007, 22, 2898. (11) Zhu, L.; Yang, R.; Zhai, J.; Tian, C. Biosens. Bioelectron. 2007, 23, 528. (12) Raitman, O. A.; Katz, E.; Bu¨ckmann, A. F.; Willner, I. J. Am. Chem. Soc. 2002, 124, 6487. (13) Kros, A.; Nolte, R. J. M.; Sommerdijk, N. A. J. M. AdV. Mater. 2002, 14, 1779. (14) Janata, J.; Josowicz, M. Nat. Mater. 2003, 2, 19. (15) Freeman, R.; Gill, R.; Willner, I. Chem. Commun. 2007, 3450. (16) Star, A.; Joshi, V.; Han, T.-R.; Altoe´, M. V. P.; Gru¨ner, G. Org. Lett. 2004, 6, 2089. (17) Allen, B. L.; Kichambare, P. D.; Star, A. AdV. Mater. 2007, 19, 1439. (18) Geetha, S.; Rao, C. R. K.; Vijayan, M.; Trivedi, D. C. Anal. Chim. Acta 2006, 568, 119. (19) Jang, J. AdV. Polym. Sci. 2006, 199, 189. (20) Zhang, X.; Zhang, J.; Song, W.; Liu, Z. J. Phys. Chem. B 2006, 110, 1158. (21) West, K.; Bay, L.; Nielsen, M. M.; Velmurugu, Y.; Skaarup, S. J. Phys. Chem. B 2004, 108, 15001. (22) Song, H.-K.; Palmore, G. T. R. J. Phys. Chem. B 2005, 109, 19278. (23) Virji, S.; Kaner, R. B.; Weiller, B. H. J. Phys. Chem. B 2006, 110, 22266. (24) Ramamathan, K.; Bangar, M. A.; Yun, M.; Chen, W.; Myung, N. V.; Mulchandani, A. J. Am. Chem. Soc. 2005, 127, 496. (25) Tabard-Cossa, V.; Godin, M.; Gru¨tter, P.; Burgess, I.; Lennox, R. B. J. Phys. Chem. B 2005, 109, 17531. (26) Yoon, H.; Kim, J.-H.; Lee, N.; Kim, B.-G.; Jang, J. ChemBioChem 2008, 9, 634. (27) Wanekaya, A. K.; Chen, W.; Myung, N. V.; Mulchandani, A. Electroanalysis 2006, 18, 533. (28) Wanekaya, A. K.; Bangar, M. A.; Yun, M.; Chen, W.; Myung, N. V.; Mulchandani, A. J. Phys. Chem. C 2007, 111, 5218. (29) Jang, J.; Chang, M.; Yoon, H. AdV. Mater. 2005, 17, 1616. (30) Yoon, H.; Chang, M.; Jang, J. J. Phys. Chem. B 2006, 110, 14074. (31) Yoon, H.; Chang, M.; Jang, J. AdV. Funct. Mater. 2007, 17, 431. (32) Jang, J.; Lim, B. Angew. Chem., Int. Ed. 2003, 42, 5600. (33) Ko, S.; Jang, J. Angew. Chem., Int. Ed. 2006, 45, 7564. (34) Kunishima, M.; Kawachi, C.; Hioki, K.; Terao, K.; Tani, S. Tetrahedron 2001, 57, 1551. (35) Guidotti, G.; Colombo, J.-P.; Foa, P. P. Anal. Chem. 1961, 33, 151. (36) Besteman, K.; Lee, J.-O.; Wiertz, F. G. M.; Heering, H. A.; Dekker, C. Nano Lett. 2003, 3, 727. (37) Yoon, H.; Hong, J.-Y.; Jang, J. Small 2007, 3, 1774. (38) Sun, S.; Ho-Si, P. H.; Harrison, D. J. Langmuir 1991, 7, 727. (39) Bartlett, P. N.; Astier, Y. Chem. Commun. 2000, 105. (40) Yon-Hin, B. F. Y.; Lowe, C. R. J. Electroanal. Chem. 1994, 374, 167. (41) Nishizawa, M.; Matsue, T.; Uchida, I. Anal. Chem. 1992, 64, 2642. (42) Lin, Y.; Lu, F.; Ren, Z. Nano Lett. 2004, 4, 191. (43) Kharitonov, A. B.; Zayats, M.; Lichtenstein, A.; Katz, E.; Willner, I. Sens. Actuators, B 2000, 70, 222.

JP800567H