Microtubule Sensors and Sensor Array Based on Polyaniline

Microtubule sensors for glucose, urea, and triglyceride were fabricated based ... spaced devices and its response was found to be free from cross-inte...
55 downloads 0 Views 105KB Size
Download PDF
Anal. Chem. 2003, 75, 5673-5679

Microtubule Sensors and Sensor Array Based on Polyaniline Synthesized in the Presence of Poly(styrene sulfonate) Mandakini Kanungo, Anil Kumar, and A. Q. Contractor*

Department of Chemistry, Indian Institute of Technology, Bombay, Mumbai-76, India

Microtubule sensors for glucose, urea, and triglyceride were fabricated based on poly(styrene sulfonate)-polyaniline (PSS-PANI) composites synthesized within the pores of track-etched polycarbonate membranes. The synthesis of a sufficiently thick and conducting PSS-PANI film at pH 5 provided the advantage of immobilizing enzymes during polymerization. This resulted in the improvement of sensor response for urea and triglyceride by a factor of ∼102 with a significant increase in the linear region of response compared to polyaniline-based sensors, where the enzymes were immobilized by physical adsorption after the polymerization. The sensors based on urea and triglyceride were found to have a higher linear range of response, better sensitivity, improved multiple use capability, and faster response time compared to the potentiometric and amperometric sensors based on polyaniline. A microtubule sensor array for glucose, urea, and triglyceride based on PSS-PANI was fabricated by immobilization of three different sets of enzymes on three closely spaced devices and its response was found to be free from cross-interference when a sample containing a mixture of the above analytes was analyzed in a single measurement. Organic conducting polymers have emerged as promising materials in the development of compact and portable probes for the detection of biologically significant molecules.1,2 Early reports of using conducting polymer matrix for the immobilization of biological substrates were concerned with the use of polypyrrole for the amperometric detection of glucose.3 Since then, various research pursuits have resulted in a variety of sensors based on conducting polymers. Among various conducting polymers, polyaniline has a unique position due to its easy synthesis, environmental stability, and reversible acid-base chemistry in aqueous solution. The application of polyaniline in biosensors is very promising. Polyaniline has been used both as an immobilization matrix and as a physicochemical transducer to convert a chemical signal into an electrical signal. A change in the pH of the microenvironment or a change in the conformation of the polymer * Corresponding author. E-mail: [email protected]. (1) McQuade, D.; Tyler, Pullen. A. E.; Swager, T. M. Chem. Rev. 2000, 100, 2537. (2) Bartlett, P. N.; Astier, Y. Chem. Commun. 2000, 105. (3) Umana, M.; Waller, J. Anal. Chem. 1984, 106, 7389. 10.1021/ac034537h CCC: $25.00 Published on Web 09/12/2003

© 2003 American Chemical Society

caused by a binding event in the polymer matrix results in a change in the electronic conductivity of the polymer. This property of polyaniline has been explored in fabrication of conductometric sensors for the determination of various biomolecules/ions in our laboratory.4-7 Polyaniline loses its electrochemical activity in solutions of pH greater than 4.8,9 Therefore, adaptation of polyaniline to neutral pH is an important problem. Attempts have been made to modify the properties of polyaniline in order to extend its conductivity to neutral pH. The first attempt in this direction was by Epstein and co-workers wherein sulfonic acid groups were introduced on the polyaniline backbone by sulfonation of the emeraldine and leucoemeraldine states of polyaniline to get self-doped polyaniline.10-12 Self-doped polyanilines were also synthesized electrochemically by copolymerization of aniline and metanilic acid where the redox activities of the polymers was maintained up to pH 9.13 Recently Kumar and co-workers have reported the successful homopolymerization of metanilic acid to get 100% sulfonated polyaniline.14 Another strategy in which the conductivity of polyaniline can be extended to neutral pH is to synthesize the polymer in the presence of anionic polyelectrolyte with a sulfonate group.15 Several potentiometric and amperometric sensor devices have been developed using the modified polyanilines, which are responsive to glucose, urea, and NADH and can be operated in pH 7 buffer solution.16-20 (4) Hoa, D. T.; Suresh Kumar, T. N.; Srinivasa, R.S.; Lal, R.; Punekar, N. S. Contractor, A. Q. Anal. Chem. 1992, 64, 2645. (5) Sangodhkar, H.; Sukeerthi, S.; Lal, R.; Srinivasa, R. S.; Contractor, A. Q. Anal. Chem. 1996, 68, 779. (6) Sukeerthi, S.; Contractor, A. Q. Anal. Chem. 1999, 71, 2231. (7) Dabke, R. B.; Singh, G. D.; Dhanabalan, A.; Lal, R.; Contractor, A. Q. Anal. Chem. 1997, 69, 724. (8) Gospodinova, N.; Terlemyzyan, L.; Mokreva, P.; Kossev, K. Polymer 1993, 34, 2434. (9) Gospodinova, N.; Mokreva, P.; Terlemezyan, L. Polymer 1994, 35, 3102. (10) Yue, J.; Epstein, A. J.J. Am. Chem. Soc. 1990, 112, 2800. (11) Yue, J. W.; Jhao, H.; Cromack, K. R.; Epstein, A. J.; MacDiarmid, A. G. J. Am. Chem. Soc. 1991, 113, 2265. 12. (12) Wei, X.; Wang, Y. Z.; Lang, S. M.; Bobeczcko, C.; Epstein, A. J. J. Am. Chem. Soc. 1996, 118, 2545. (13) Karyakin, A. A.; Strakhova, A. K.; Yatimirsky, A. K. J. Electroanal. Chem. 1994, 371, 259. (14) Krishnamoorthy, K.; Contractor, A. Q.; Kumar, A. Chem. Commun. 2002, 240. (15) Austria, G.; Jang, G. W.; MacDiarmid, A. G.; Doblhofer, K.; Zhang, C. Ber. Bunsen-Ges. Phys. Chem. 1991, 95, 1381. (16) Castillo-Ortega, M. M.; Rodriguez, D. E.; Encinas, J. C.; Plascencia, M.; Mendez-Velarde, F. A.; Olayo, R. Sens. Actuators, B 2002, B85(1-2), 19.

Analytical Chemistry, Vol. 75, No. 21, November 1, 2003 5673

Since conducting polyaniline could only be synthesized from acidic solutions, most of the sensor devices are fabricated by the physical adsorption of the enzyme/host species after the polymerization. This restricts the amount of loading, which then results in poor sensitivities. Furthermore, the multiple-use capability of these devices is also very poor due to the easy leaching out of the enzyme/host species because of the poor interaction due to the physical adsorption. These problems can be circumvented if the enzyme/host species could be loaded during the polymerization, which will then result in higher loading levels and better entrapment. This cannot be achieved because most of the enzymes are unstable in acidic conditions required for the polymerization of aniline. Recently, Tripathy and co-workers have come up with a novel strategy for enzymatic synthesis of conducting polyaniline in the presence of a strong acidic polyelectrolyte such as poly(styrene sulfonate) at pH 4.3-5.5 phosphate buffer in the presence of the enzyme horseradish peroxidase.21-23 Since acidic polyelectrolytes attract hydrogen ions electrostatically, the pH at the acidic polyelectrolyte surfaces is much lower than that of bulk aqueous medium.24 This provides a suitable microenvironment for the formation of conducting polyaniline. On the basis of these reports, we envisaged that if the polyaniline could be synthesized at higher pH then this would enable us to immobilize the enzymes during polymerization. This should then result in significant improvement in sensitivity and recycling ability. In this paper, we report on the synthesis and characterization of a sufficiently thick conducting film at a biocompatible pH 5. Since most of the enzymes are fairly stable at this pH, we could immobilize the enzymes during polymerization. This goal was achieved by synthesizing the polyaniline in the presence of an anionic polyelectrolyte, poly(styrene sulfonate) (PSS). A strong acidic polyelectrolyte such as PSS is the most favored because it provides a sufficiently low local pH microenvironment for the formation of a conducting polymer film at high bulk pH. The resulting devices showed a significant increase in the sensitivity and linear range of response especially in the case of urea and triglyceride compared to that of polyaniline. A sensor array was fabricated, which was used to determine the concentration of glucose, urea, and triglyceride in a single measurement. The sensors based on urea and triglyceride were found to have a higher linear range of response, better sensitivity, improved recycling ability, and faster response time compared to the potentiometric and amperometric sensors based on polyaniline. (17) Bartlett, P. N.; Birkin. P. R.; Wallace, E.N. K. J. Chem. Soc., Faraday Trans. 1997, 93, 1951. (18) Tatsuma, T.; Ogawa, T.; Sato, R.; Oyama, N. J. Electroanal. Chem. 2001, 501 (1-2), 180. (19) Karyakin, A. A.; Lukachova, L. V.; Karyakina, E. E.; Orlov, A. V.; Karpachova, G.; Wang, J. Anal. Chem. 1999, 71, 2534. (20) Raitman, O. A.; Katz, E.; Bueckmann, A. F.; Willner, I. J. Am. Chem. Soc. 2002, 124 (22), 6487. (21) Liu, W.; Kumar, J.; Tripathy, S. K.; Senecal, K. J.; Samuelson, L. J. Am. Chem. Soc. 1999, 121, 71. (22) Liu, W.; Cholli, A. L.; Nagarajan, R.; Kumar, J.; Tripathy, S.; Bruno, F. F.; Samuelson, L. J. Am. Chem. Soc. 1999, 121, 11345. (23) Nagarajan, R.; Tripathy, S.; Kumar, J.; Bruno, F. F.; Samuelson, L. Macromolecules 2000, 33, 9542. (24) (a) Manning, G. S. Acc. Chem. Res. 1979, 12, 443. (b) Manning, G. S. J. Chem. Phys. 1988, 89, 3722.

5674

Analytical Chemistry, Vol. 75, No. 21, November 1, 2003

EXPERIMENTAL SECTION Chemicals and Materials. Freshly distilled aniline (Merck) was used for preparing monomer solution. The sulfuric acid used was MOS grade with 99.9% purity. The sodium salt of poly(styrene sulfonate) (Mw ) 70,000) was obtained from Aldrich. The enzymes glucose oxidase from Aspergillus niger (EC 1.1.3.4), urease from Jack bean meal (EC 3.5.1.5), and peroxidase (EC 1.11.1.7) were obtained from Sigma. Lipase from triglycerol acylhydrolase and glucose, urea, and triolein was obtained from SRL (Sisco Research Laboratories). Salts used for the preparation of buffers were of analytical grade and were used without further purification. Tracketched polycarbonate membranes having a pore diameter of 1.2 µm and thickness of 10 µm were obtained from the Millipore Inc. The pore densities of the membrane were calculated with the help of the scanning electron micrographs of the polycarbonate membranes. Around six pictures of various regions were taken, and the numbers of pores per cubic centimeter were calculated. The average pore densities were found to be 1.2 × 107 pores/ cm2. Fabrication of Sensor Devices. Track-etched polycarbonate membranes were used for the fabrication of sensor devices. Gold films were deposited on the two sides of the membrane by vacuum evaporation using a mask in a homemade vacuum evaporation system.25 The mask was made by cutting equidistant lines of 1-mm width separated by 1 mm on an aluminum sheet. The two gold lines at the opposite faces of the membrane exactly overlapped with each other and were used as two electrodes for the growth of the polyaniline. Each gold line was used as an electrode. The electrodes were held by a plastic clip holder with platinum contacts from which connections to the instruments were made. PSSpolyaniline (PANI) was synthesized from pH 0.6 solution (0.1 M aniline in 0.5 M H2SO4 + 50 mM PSS) and from pH 5.0 solution (0.1 M aniline in pH 5 buffer + 50 mM PSS) within the pores (1.2 µm) of gold-coated polycarbonate membranes. PSS-PANI was deposited electrochemically from pH 0.6 by scanning the potential between -0.2 and 0.8 V versus SCE at a scan rate of 50 mV/s. For the polymerization from pH 5 solution, the potentiodynamic growth of the polymer was carried out by scanning the potential between -0.2 and 1.0 V versus SCE. However, the potentiodynamic growth of the polymer at pH 5 phthalate buffer was very slow and therefore a potentiostatic growth of the polymer at 1.0 V was preferred. It took ∼25 min to bridge the two sides of the membrane. Polyaniline films in the absence of PSS were synthesized by scanning the potential between -0.2 and 0.8 V versus SCE at a scan rate of 50 mV/s for 10 min. The synthesis of polyaniline in the absence of PSS at pH g4 resulted in a very slow kinetics of polymerization with the formation of a very thin and poorly conducting film. The sensor devices were fabricated in two ways. For PSSPANI-A and PANI devices, the electropolymerization was done at pH 0.6. The immobilization of glucose oxidase (GOx) was carried out by formation of a two-layer film, polymer, and polymer + GOx.4,5 The immobilization of urease and lipase was carried out by physical adsorption after polymerization in the same manner as described earlier.4,5 For PSS-PANI-B devices, the electropolymerization was done at pH 5 in the presence of the enzyme by potentiostatic polymerization at 1.2 V versus SCE for (25) Sukeerthi, S.; Contractor, A.Q. Chem. Mater. 1998, 10, 1412.

∼35 min. The potential was kept at 1.2 V because the presence of enzymes in the monomer solution slows the rate of polymerization, probably due to inhibition by adsorption of enzyme. It should be noted that phthalate buffer was used for the chracterization of the polymer but the buffer was changed to phosphate for biosensor studies because our earlier work on the biosensor used this buffer. Changing the buffer from phthalate to phosphate did not have any effect in the device studies. Fabrication of Sensor Array for Glucose, Urea, and Triglyceride. Microtubular sensor arrays were fabricated on one single device consisting of glucose, urea, and triglyceride immobilized on polyaniline synthesized in the presence of PSS at pH 5. Four gold lines were taken and were held by a plastic clip holder with platinum contacts. Among the four gold lines, three lines were used as sensor devices and the fourth line was used as a reference sensor. The reference sensor was coated with PSSPANI in the absence of enzyme. Glucose oxidase, urease, and lipase were immobilized on individual lines one after another by synthesizing PSS-PANI from pH 5 buffer in the presence of the respective enzymes. All the other neighboring lines coated with PSS-PANI (in case of the reference device) or PSS-PANIenzyme films were maintained at -0.2 V. At -0.2 V, the polymer remains in a compact form and hence it minimizes the possibility of cross-immobilization of enzymes on the polymer matrix. Characterization of the Polymer. PSS-PANI films were characterized by cyclic voltammetry, in situ conductance measurements, spectroelectrochemistry, and scanning electron microscopy. Cyclic voltammograms of the polymer were recorded with the help of an EG & G PARC 362 potentiostat/galvanostat coupled to a Linseis XY-t recorder. In situ resistance measurements were carried out on polymer formed on gold-coated 1.2-µm-pore diameter polycarbonate membranes in the transistor mode. The two sides of the membrane electrode act as “source” and “drain” of the electrochemical transistor. The conductance of the polymer was taken as the reciprocal of resistance. An AFRDE4 Pine bipotentiostat coupled with a Philips multimeter was used for carrying out the resistance/conductance measurements. The in situ UV-visible spectroscopy of the PSS-PANI film was carried out with the help of a Shimazdu 2100 UV-visible spectrophotometer. The potential was controlled with the help of an EG &G PARC 362 potentiostat. The scanning electron microscopy of the polymer tubules was recorded with the help of a JEOL JSM6400 microscope after dissolving the template membrane in dichloromethane. Activity of the Immobilized Enzymes. The immobilized enzymes on PANI and PSS-PANI films were checked for activity. To check the activity of the polymer-GOx film (0.5 cm2), the films were immersed in 2.6 mL of 0.1 M sodium phosphate buffer (pH 7) containing 2 units of peroxidase (POD), 50 µmol of D-glucose and o-anisidine.26 The film was maintained for 30 min at 298 K with vigorous stirring and was further incubated for 30 min at 310 K. The change in absorbance was recorded at 520 nm. The polymer-urease was immersed in pH 5.2 acetate buffer containing 50 µmol of urea. The film was maintained for 30 min at 298 K with stirring followed by 30-min incubation at 310 K. (26) Kunst, A.; Drager, B.; Ziegenhorn, J. In Methods of Enzymatic Analysis; Bergmeyer, H. U., Bergmeyer, J., Grassi, M., Eds.; Verlag Chemie: Weinheim, Germany, 1984; Vol. 6, p 178.

Table 1. Activities of the Immobilized Enzymes enzyme activity (µmol) polymer

urea

lipase

glucose

PANI PSS-PANI-B

0.8 2.6

0.6 1.7

2 3.2

Then the assay for urea was done by the conventional procedure.27 A titrimetric assay was carried out to check the activity of immobilized lipase on the polymer film.28 The polymer-lipase film was immersed in 5 mL of 20-µmol triolein solution in 0.01% Triton X for 30 min, and 5 mL of methanol was added to stop the reaction. The liberated fatty acid was titrated with 0.1 M KOH. Control experiments were done in all the above cases for polymer film in the absence of enzyme, and the values obtained were subtracted from the experiments conducted in the presence of enzyme to calculate the activity of immobilized enzyme. Enzyme activity was calculated in terms of the moles of substrate oxidized in the above assay. The immobilized enzyme activity for the various enzymes into different polymer devices was determined, and the values are reported in Table 1. In the case of PANI and PSS-PANI-A, enzyme immobilization was carried out for 2 h by a physical absorption method. In the case of PSS-PANI-B, the enzyme was loaded during the polymerization (pH 5) that was carried out for 40 min, i.e., the time taken to form a sufficiently thick film on the electrode surface. As can be seen in the Table 1, the enzyme activities are significantly higher in PSS-PANI-B devices though the loading time was smaller. This further confirms the efficient loading of the enzymes during polymerization. Enzyme activities in PSS-PANI-A devices, where the enzymes were loaded after the polymerization by physical adsorption, were comparable to that of PANI. Sensor Measurements. Sensor response for glucose is represented by ∆g/go, where go is the conductance of the sensor in the absence of the substrate and ∆g ) g - go, where g is the conductance in the presence of the substrate. Representing the response in this manner normalizes it for variations in conductance from sensor to sensor. The response of urea and triglyceride sensors is represented by ∆r/ro, where ro is the resistance of the sensor in the absence of the substrate and ∆r ) r - ro, where r is the resistance in the presence of the substrate. The variable g was chosen in the case of glucose and r in the case of urea and triglyceride so that the sensor response had a positive slope when plotted versus concentration of the substrate in all the three cases. RESULTS AND DISCUSSION Synthesis and Characterization of PSS-PANI Film. PSSPANI film was synthesized both at pH 0.6 (PSS-PANI-A) and at pH 5 (PSS-PANI-B). Here, PSS acts as a charge compensator and the rate of polymerization of aniline is enhanced in the presence of PSS.29 PSS-PANI film was synthesized from pH 5 (27) Kerscher, L.; Ziegenhorn, J. In Methods of Enzymatic Analysis, 3rd ed.; Bergmeyer, H. U., Editor-in-Chief, Bergmeyer, J., Grabi, M., Eds.; VCH Publishers: Weinheim, Germany, 1985; Vol. 8, p 449. (28) Borgstrom, B.; Brockman, L. H. Lipases; Elsevier Science Publishers: Amsterdam, The Netherlands, 1984. (29) Michaelson, J. C.; McEnvoy, A. J.; Kuramoto, N. Reactive Polym. 1992, 17, 197.

Analytical Chemistry, Vol. 75, No. 21, November 1, 2003

5675

Figure 2. Change in conductance of 1.2-µm PSS-PANI and PANI tubules as a function of pH at various gate potentials. [, 9, and 2 are for PSS-PANI at a gate potential of -0.2, 0.0, and 0.2 V, respectively, and b for PANI at a gate potential of 0.2 V. PSS-PANI was polymerized from pH 5 buffer solutions.

Figure 1. Scanning electron micrographs of 1.2-µm PSS-PANI-B tubules formed on the pores of polycarbonate membranes. The tubules are 8-9 µm in length. The thickness of the polycarbonate membrane used was 10 µm.

buffer solution by potentiostatic polymerization at 1.0 V versus SCE. The formation of conducting polyaniline at pH 5 was attributed to the presence of aniline in the regions, which are locally more acidic than the bulk solution.30 In other words, PSS provides the necessary alignment of the aniline monomer and a sufficiently low local pH for the formation of a conducting polyaniline.21-23 Since the bulk pH is high enough to prevent the denaturation of enzymes, they can be added to the monomer solution and therefore can be immobilized into the polymer films during polymerization. However, the synthesis of a conducting polyaniline in the presence of PSS was also found to depend on the pH of the monomer solution. In our case, conducting polyaniline film could not be formed at pH greater than 5.3. This was further supported by the observation of Liu et al., who have reported the formation of a highly branched and insulating polyaniline at pH g6.21,22 Figure 1 shows the micrographs of PSSPANI-B tubules formed inside the pores of the membrane. The tubules are 8-9 µm in length. The formation of the polymeric tubules inside the pores of the membrane indicates that PSSPANI synthesized at pH 5 phthalate buffer solution is sufficiently conducting in nature to allow the growth of a bridge across the membrane. The in situ conductance measurements of the PSSPANI-B synthesized at pH 5 were carried out at different potentials and pHs. Figure 2 shows the change in conductance as a function of pH for 1.2-µm PSS-PANI-B tubules at different gate potentials. Similar trends were observed with PSS-PANI-A film synthesized at pH 0.6. The conductance of the PSS-PANI decreases gradually with increase in pH. This is in contrast to PANI, where the conductance of the polymer film decreases sharply with increase in pH from pH 1 to 6 after which there is no significant change in the conductance of the polymer with pH. However, unlike the sulfonated PANI films, the conductance of the PSS-PANI is not entirely independent of pH.12 This may be because in sulfonated polyaniline sulfonate groups are bound covalently to the polymer. Therefore, the polymer will not be in a completely doped state at pH 5. This results in an increase in conductance of around 1-1.5 (30) Kuramoto, N.; Michaelson, J. C.; McEnvoy, A. J. Gratzel, M. J. Chem. Commun. 1990, 1478.

5676 Analytical Chemistry, Vol. 75, No. 21, November 1, 2003

Figure 3. Response of glucose microtubules based on PANI (×), PSS-PANI-A (2), and PSS-PANI-B (b) (PSS-PANI-A and PANI shown in the inset) along with the error bars (glucose in phosphate buffer of pH 7). Data are given as mean ( (variation from the mean) for three experiments.

order of magnitude while going from pH 5 to pH 1. The in situ UV-visible spectrum of PSS-PANI-B also shows the formation of strong absorption bands at 430-440 and 800 nm, which are indicative of the formation of a conducting film. The 800-nm band increases with increase in potential until 0.3 V, after which it decreases because of the presence of a fully oxidized PANI at higher potential. Sensor Response. The responses of the membrane devices were measured by dc conductance measurements as described earlier.25 The sensor response was measured at three different gate potentials -0.2, 0, and 0.2 V versus SCE with a 20-mV drain potential. The sensitivity (defined as change in response of the sensor per millimolar change in the concentration of the substrate in the linear region of response) was found to be highest at 0 V for PSS-PANI-based sensors and at 0.2 V for PANI-based sensors. All the experiments were repeated three to four times with films prepared under nominally identical conditions. The extents of variation between the experiments are mentioned in the graphs as the +ve and -ve variation from the mean values of measurement obtained from three different experiments. In the case of the glucose sensor, the enzyme-catalyzed reaction results in the formation of gluconic acid and H2O2. The production of gluconic acid results in a decrease in the pH of the microenvironment and in turn an increase in the conductance of the polymer. Figure 3 shows the response based on PSS-PANI-B, PSS-PANI-A, and PANI devices.

Table 2. Comparison of Sensitivity and Linear Range of Glucose, Urea, and Triglyceride Sensors Based on PANI and PSS-PANI Microtubules substrate urea

triglyceride

glucose

polymer

sensitivity (mM-1)

linear range (mM)

sensitivity (mM-1)

linear range (mM)

sensitivity (mM-1)

linear range (mM)

PANI PANI-A PANI-B

0.02 ( 0.001 0.09 ( 0.003 1.72 ( 0.056

0-30 0-40 0-60

0.03 ( 0.004 0.15 ( 0.005 0.98 ( 0.037

0-30 0-40 0-60

2.4 ( 0.081 0.19 ( 0.011 0.72 ( 0.033

0-50 0-50 0-60

Figure 4. Response of the urea microtubule sensor based on PSSPANI-B (b), PSS-PANI-A (2), and PANI (×) (PSS-PANI-A and PANI shown in the inset), urea in pH 5.2 buffer. Data are given as mean ( (variation from the mean) for three experiments.

Figure 5. Response of triolein microtubule sensor based on PSSPANI-B (b), PSS-PANI-A (2), and PANI (×) (PSS-PANI-A and PANI shown in the inset), triolein in pH 5.2 buffer solubilized by 0.01% Triton X 100. Data are given as mean ( (variation from the mean) for three experiments.

The enzyme-catalyzed reaction for the urease-loaded film in the presence of urea results in the formation of NH3. The production of NH3 (pKa ) 9.25) would raise the pH of the microenvironment of the polymer matrix and would consequently lower the conductance of the film. The sensor response of urea based on PSS-PANI-B, PSS-PANI-A, and PANI devices is shown in Figure 4. The enzyme-catalyzed reaction for lipase-loaded film results in the formation of oleic acid and glycerol. Oleic acid, being insoluble in water, either forms micelles or remains solubilized by 0.01% Triton X 100 while the glycerol goes into the solution. An independent control experiment shows that addition of glycerol and Triton X into the buffer in the concentration similar to that used here results in an increase in the pH of the solution and hence a lowering of the conductance of the film. This is in agreement with our present observation. Figure 5 shows the sensor response of triolein for PSS-PANI-B, PSS-PANI-A, and PANI films. Control experiments for PANI and PSS-PANI were carried out in all the above cases. The polymer film was exposed to different concentrations of the analyte in the absence of enzyme. The small and constant response of the reference film shows that the sensor response is due to the presence of enzyme-catalyzed reaction on the polymer matrix. The sensitivity and linear range of response of glucose, urea, and triglyceride sensors based on PSS-PANI-B, PSS-PANI-A, and PANI films are given in Table 2. It is clear from the observations that the sensitivity for urea and triglyceride sensors is highest for PSS-PANI-B followed by PSS-PANI-A and PANI film. On the other hand, for the glucose sensor, the sensitivity

was found higher in PANI followed by PSS-PANI-B and PSSPANI-A. The linear range of response for urea and triglyceride sensors were found more for PSS-PANI-based sensors compared to that of PANI. The sensor response depends on multiple factors: amount of active enzyme immobilized on the polymer matrix, pH-dependent conductance of the polymer, and diffusion of substrates into the polymer film. In the present case, urease and lipase were immobilized into PANI and PSS-PANI-A films by physical adsorption after polymerization and into PSS-PANI-B film during polymerization. For PSS-PANI, the conductance changes with pH are sharper on going to higher pH from pH 5 in comparison to that of PANI (Figure 2). Thus, there is an increase in the sensitivity of a sensor based on PSS-PANI-A film compared to PANI film both for urea and triglyceride sensors. It should be noted here that, in both the cases, the enzyme immobilization was carried out by physical adsorption. The sensitivity of the sensor further increases in the case of PSS-PANI-B, and this was attributed to the higher enzyme loading on the polymer matrix by immobilizing the enzymes during polymerization. On the other hand, for the glucose sensor, the sensitivity was found highest for the PANI-based sensor followed by PSS-PANI-B and PSSPANI-A. This is because in the glucose sensor, the pH of the microenvironment decreases due to the production of gluconic acid and the conductance of PANI shows a sharp increase with pH in the region of pH 1-6 (Figure 2). Though the enzyme loading was found more in PSS-PANI-B compared to PANI film, the sharp change of conductance of PANI in the range of pH 1-6 explains the increase in sensitivity of the sensor in the latter. For PSS-PANI-A film, the sensitivity was found still lower. Therefore, Analytical Chemistry, Vol. 75, No. 21, November 1, 2003

5677

Figure 6. Repeatability of the urea and lipid microtubule sensors to 20 independent measurements (a) Response of the urea sensors to 30 mM urea solution ([). ((b) Response of the lipid sensor to 20 mM lipid solution (9).

Figure 7. Sensor response for urea for PSS-PANI-B (9) and PANI ([) as a function of number of days. The sensors were stored in buffer at 4 °C.

the sensitivity and linear range of urea and triglyceride sensors can be improved by modifying the pH-dependent conductance behavior of the polymer and further by increasing the effective enzyme loading in the polymer matrix by immobilizing the enzymes during polymerization. The current increases (in the case of a glucose sensor) or decreases (in the case of urea and triglyceride sensors), when the enzyme-modified electrode was exposed to the appropriate analyte solution, then it reaches a stable

Figure 8. Response of the microtubule sensor array for PSS-PANI-B. 5678

Analytical Chemistry, Vol. 75, No. 21, November 1, 2003

value (less than 5% variation over a period of 3 min) in 6-8 s for glucose and urea and 10 s for triglyceride. The time taken for the sensor to reach this stable value is called here the response time of the sensor. The slower response time found in the triglyceride sensor may be attributed to the slow diffusion of the triglyceride molecule to the polymer electrode because of the solubilization of triglyceride by Triton X 100. To study the reproducibility and the stability of the sensor devices, the PSS-PANI-B sensors were used for repeated measurements for a 30 mM urea and 20 mM triglyceride solution. The results are shown in Figure 6. Each measurement was preceded by rinsing the sensor in buffer solution. The sensor response was fairly reproducible for both the urea and triglyceride sensors even after 20 runs with a low standard deviation of 4%, which indicates that the enzyme is trapped in the polymer film and does not leach out. The urea sensors based on PSS-PANI-B and PANI were tested for shelf life by keeping the polymer-enzyme film at 4 °C in buffer solutions. PSS-PANI-based urea sensors were found to maintain 75-80% activity even after 7 days. On the other hand, the PANIbased urea sensors retained only 15-20% activity after 7 days. The PSS-PANI-B sensor devices were found to retain 50% of their activity after storing the film in the buffer solution for 2 months. Figure 7 shows the sensor response of urea for PSS-PANI-B and PANI as a function of number of days. Thus, the stability and the shelf life of the sensors were found to be improved significantly by immobilizing the enzyme during polymerization. Response of Sensor Array for Glucose, Urea, and Triglyceride. The detailed explanation about the fabrication of the sensor array is given in the Experimental Section. The sensor array was exposed to a mixture containing all three substrates. Data for three such mixtures are presented here, and the compositions of the three mixtures are given in Table 3. The sensors were addressed sequentially while maintaining the sensors that were not being addressed at -0.2 V to minimize leaching out of the respective enzyme. The concentration of each component was estimated by comparison with the respective calibration plot. Concentrations thus obtained are plotted versus the known concentrations of that component and shown in Figure 8. The

Table 3. Response of Sensor Array for Glucose, Urea, and Triglyceride concentrations as prepared (mM)

concentrations as measured (mM)

solution glucose urea triglyceride glucose urea triglyceride A B C

10 30 40

20 40 30

30 10 20

8 27 37

22 38 27

28 11 18

data points are in good agreement with a line of unity slope that would be expected ideally. CONCLUSIONS In this paper, we have shown that conducting polyaniline can be synthesized at a more biocompatible pH 5 in the presence of poly(styrene sulfonate). The PSS-PANI films showed a different pH-dependent conductivity compared to polyaniline. Here the conductivity change with pH was more gradual compared to polyaniline. The immobilization of enzymes during polymerization increased the effective enzyme loading, sensitivity, and linear response of the sensor devices. The linear range (0-60 mM) and response time (6-8 s) of the present urea sensors based on PSSPANI were found to be significantly higher compared to the potentiometric urea sensor based on processible polyaniline (linear range of 1-10 mM with response time of 50 s).19 On the other (31) Bartlett, P. N.; Wang, J. H. J. Chem. Soc., Faraday. Trans. 1996, 92, 4137.b

hand, the linear range (0-60 mM) and response time (6-8 s) of the glucose sensors were also found to be better than the reported linear range of 0.1-30 mM with a response time of 1-2 min.19 Furthermore, the linear range of our glucose sensor is better than the linear range of some commercial instruments that have an assay range of 20-600 mg/dL (1.1-33.3 mM). Bartlett and Wang reported on glucose sensors based on PSS-PANI composite films operated at neutral pH 7.31 The response time of the sensors was 100 s, and the sensors showed good stability for 58 consecutive measurements. On the basis of these facts, we can conclude that the loading of the enzymes during polymerization improves the loading of the enzymes, which in turn enhances the performance of the devices. It is interesting to note that the sensor response is the result of the change in conductivity resulting from a change in pH though the solutions are buffered. This is because the change in conductance is due to the polymer deposited on the pore wall, and since the pores are extremely small (diameter 1.2 µm and length 10 µm), movement and equilibration of H+ ions with the bulk solution is very slow. Therefore, though the pH in the bulk is buffered, that in the pores is not. ACKNOWLEDGMENT We thank MHRD, India for the financial help and Prof. Rakesh Lal for numerous discussions. Received for review May 21, 2003. Accepted August 14, 2003. AC034537H

Analytical Chemistry, Vol. 75, No. 21, November 1, 2003

5679