ANALYTICAL SCIENCES JULY 2015, VOL. 31
603
2015 © The Japan Society for Analytical Chemistry
Fabrication and Characterization of Ultrathin-ring Electrodes for Pseudo-steady-state Amperometric Detection Yuki KITAZUMI,† Katsumi HAMAMOTO, Tatsuo NODA, Osamu SHIRAI, and Kenji KANO Division of Applied Life Science, Graduated School of Agriculture, Kyoto University, Kitashirakawaoiwake-cho, Sakyo, Kyoto 606–8502, Japan
The fabrication of ultrathin-ring electrodes with a diameter of 2 mm and a thickness of 100 nm is established. The ultrathin-ring electrodes provide a large density of pseudo-steady-state currents, and realize pseudo-steady-state amperometry under quiescent conditions without a Faraday cage. Under the limiting current conditions, the current response at the ultrathin-ring electrode can be well explained by the theory of the microband electrode response. Cyclic voltammograms at the ultrathin-ring electrode show sigmoidal characteristics with some hysteresis. Numerical simulation reveals that the hysteresis can be ascribed to the time-dependence of pseudo-steady-state current. The performance of amperometry with the ultrathin-ring electrode has been verified in its application to redox enzyme kinetic measurements. Keywords Amperometry, pseudo-steady-state current, microband electrode, kinetic measurements (Received February 25, 2015; Accepted April 22, 2015; Published July 10, 2015)
Introduction The amperometric sensor is a device that detects the concentration of analytes on the basis of the Faradaic current of the electrode reaction. The current reflects the reaction rate on the electrode surface and is proportional to the concentration of the analyte. At conventional electrodes, the electrode reaction increases the depletion of the analyte around the electrode with time, that is, the concentration polarization. The time-dependent characteristic of the diffusion current at conventional electrodes is inconvenient for amperometric analysis and limits the practical application. Therefore, user-friendly and convenient amperometric sensors require the time-independent characteristics of the amperometric response, that is, the steady-state current. Several concepts are adopted to achieve steady-state currents.1 The first one is to control the current by the catalytic reaction rate. This kind of the steady-state current reflects the rate of catalytic reactions, especially of redox enzyme reactions in the presence of excess amounts of enzyme substrates.2,3 The second one is to control the mass transfer of the analyte by transport (or permeation) across a membrane on the electrode surface. This approach has been well and successfully realized in the Clark-type oxygen electrode.4 The third one is convective mass-transfer by using a rotating disc electrode or a thin-layer flow cell. This well-known technique provides useful information about the electrode reaction as well as amperometry.5 The forth one is utilization of the microelectrode based on sufficient mass-transfer by non-linear diffusion.6–9 A microdisc electrode provides a steady-state limiting current (Is,lim) after a suitable electrolysis period under quiescent To whom correspondence should be addressed. E-mail: [email protected]
†
conditions, as expressed by the following equation:6 Is,lim = 4nFDrc0,
(1)
where n, F, D, c0, and r are the number of electrons, the Faraday constant, the diffusion coefficient, the concentration in the bulk phase, and the radius of the electrode, respectively. Unfortunately, the signals of microdisc electrodes are quite small (typically less than a few nanoampere) as compared to the noise from the external disturbance. Therefore, measurements with microdisc electrodes require a Faraday cage. One of the ideas to enhance the signal intensity with the microdisc electrode is to construct of an electrode array. However, the usual microelectrode arrays do not give steady-state currents but a diffusion current with time-dependent characteristics, since the diffusion layer at the microdisc electrodes overlap with each other.10,11 On the other hand, coarsely arranged microdisc electrode arrays are proposed to overcome this problem.12–14 However, the fabrication of the coarsely arranged microdisc electrode array is technically difficult and is not suitable for mass production for practical applications. Another idea to obtain a large steady-state signal is to use a microband electrode.6 The current response of the microband electrode under the limiting current conditions (I(t)lim) is formulated by Szabo et al. as follows:15 I (t ) lim = nFDc 0 l
1 + 1 (τ < 2 / 5) πτ
=
π exp(−2 / 5 πτ ) 4 πτ
+
π (τ > 2 / 5), ln[ 64τ exp(−0.5772156) + exp(5 / 3)]
(2)
where τ (≡ Dt/w2), l, and w are the dimensionless time, the length of the microband, and the thickness of the microband,
604 respectively. Although it takes an extremely long time to reach a true steady-state current at a microband electrode, the pseudo-steady-state current increases with an increase in l.6,15 The density of the pseudo-steady-state current (I(t)lim/lw [t = 10 s, l = 6.28 mm, w = 0.1 μm, D = 1 × 10–9 m2 s–1, c0 = 1 mmol dm–3] = 35 mA cm–2) is much larger than a steady-state current density at a typical microdisc electrode (Is,lim/πr2 [r = 10 μm] = 1.2 mA cm–2). This is a great advantage in practical amperometric applications over the microdisc electrode method. In order to increase l in a rod-type style electrode, thin-ring electrodes can be proposed.16,17 In addition, an ultrathin-ring electrode should also be effective for applications, because a decrease in w should decrease the response time. Although ultrathin-ring electrodes will bring out their advantages in the quiescent solution, ultrathin-ring electrodes for an amperometric detector have only been reported in flow systems.18,19 Moreover, the reported responses of ultrathin-ring electrodes were less than half the theoretical value and largely depend on polishing of the electrode.19 The origin of non-ideal characteristics were predicted discontinuities in the metal surface.19 Therefore, more suitable fabrication of ultrathin-ring electrodes and the characterization remain to be elucidated. To order to solve and clarify these problems about ultrathinring electrodes, the fabrication of an ultrathin-ring electrode is proposed. Furthermore, the current response is analyzed from a theoretical viewpoint. The proposed amperometric method with the ultrathin-ring electrode is applied to the detection of the concentration change due to a redox enzyme reaction under quiescent conditions.
Experimental Preparation of the ultrathin-ring electrode A schematic view of the ultrathin-ring electrode is shown in Fig. 1A. A rod formed by a UV-curable resin (Sunny, UV-3421) with a diameter of 2.0 mm (a) was bound with a copper wire (b). A gold film (c) was prepared at a thickness of 100 nm by sputtering on the side surface of the rod on a desk-top quick coater (SC-704, Sanyu Electron). The sputtering was carried out in twice on each semicylindrical surface of the rod under control of the current and time. The electrode side of the rod was hardened with the UV-curable resin (d). The UV-curable resin was selected on the basis of the adhesion property to gold. The compatibility of materials pinching the gold film was necessary to avoid any cracking at the electrode surface. An epoxy resin (Huntsman, Araldite® 2020) was cast as the body of the electrode (e). The tip of the electrode was cut with a lathe and ground with running water using emery papers (#1500 and then #4000). A photograph in Fig. 1B shows the tip of a fabricated ultrathin-ring electrode. The performance of the ultrathin-ring electrodes was checked by amperometry and cyclic voltammetry. According to this improved procedure, the yield fabricating of the ultrathin-ring electrodes was over 70% within 1% of the variability in the current response. Note here that polishing of the ultrathin-ring electrode with abrasive particles depresses the performance. The established processes are easy to be carried out, because every step includes no delicate operations and is carried out at room temperature. Chemicals 1,1′-Ferrocenedimethanol (Fc(CH2OH)2), potassium chloride, glucose, piperazine-1,4-bis(2-ethanesulfonic acid) (PIPES), and pyrroloquinoline quinone dependent glucose dehydrogenase
ANALYTICAL SCIENCES JULY 2015, VOL. 31
Fig. 1 (A) Schematic view of an ultrathin-ring electrode, where (a) rod formed by UV cured resin, (b) lead wire connected to the gold thin layer, (c) spattered gold layer, (d) tip formed by UV cured resin, and (e) body formed by epoxy resin. (B) A photograph showing the tip of a prepared ultrathin-ring electrode.
(PQQ-GDH, Toyobo) were purchased commercially and used without further purification. A stock solution of 1,1′-bis (hydroxymethyl)ferrocenium (Fc(CH2OH)2+) was prepared as follows. Fc(CH2OH)2 was oxidized with two equivalents of AgNO3 in water. After filtration to remove the product Ag, any excess of Ag+ in the solution was removed as AgCl by the addition of KCl. The resultant solution was used as the stock solution of Fc(CH2OH)2+. Apparatus Electrochemical measurements were carried out by using an electrochemical analyzer (BAS50W, BAS). An Ag | AgCl | sat. KCl electrode and a platinum wire were used as a reference electrode and a counter electrode, respectively. Time-causes of the absorbance during an enzyme reaction were recorded on a UV-Vis spectrophotometer (UV2550, Shimadzu). All measurements were carried out at room temperature. Numerical simulation The reversible electrode reaction at ultrathin-ring electrodes and diffusion equations of redox species were numerically simulated by using a commercial package of the finite-element method, COMSOL 5.0 (COMSOL).
Results and Discussion To investigate the pseudo-steady-state current recorded at the ultrathin-ring electrode, potential-step chronoamperometry was carried out. The solid line in Fig. 2 shows an amperogram recorded under the limiting conditions for the oxidation of 1 mmol dm–3 Fc(CH2OH)2 in 100 mmol dm–3 of KCl at an ultrathin-ring electrode. The current reaches a practical steady-state value at about 10 s. The closed circles in Fig. 2 show the current value calculated on Eq. (2) at w = 100 nm, l = 6.28 mm, and D = 6.8 × 10–10 m2 s–1. Here, the value of D used here for Fc(CH2OH)2 was determined separately by voltammetry on a conventional disc electrode (data not shown). The good agreement between the experimental and the theoretical values indicate that the ultrathin-ring electrode fabricated in the proposed method behaves as a typical microband electrode with
ANALYTICAL SCIENCES JULY 2015, VOL. 31
Fig. 2 (Solid line) A potential-step chronoamperogram for the oxidation of 1 mmol dm–3 Fc(CH2OH)2 in 100 mmol dm–3 KCl. (Closed circles) The current calculated on Eq. (2) with n = 1, D = 6.8 × 10–10 m2 s–1, c0 = 1 mmol dm–3, w = 100 nm, and l = 6.28 mm. (Broken line) Time dependence of the relative change in the current (Δ).
a sufficiently large l value. Although the ultrathin-electrode does not give a true steady-state current within 20 s, as shown in Fig. 2, we may use a pseudo steady-state current for practical applications in amperometry. Here, the relative change in the current, Δ = (dI(t)lim/dt)/I(t), is introduced to estimate the pseudo-steady state of the current. The broken line in Fig. 2 shows the calculated value of Δ. When Δ is set to 1% as a threshold considering the steady-state current, we can conclude the current reaches a pseudo-steady-state value at 5 s after potential step. This quick response of the ultrathin-ring electrode is due to ultrathin characteristics. The pseudo-steady-state current at t = 5 s of the amperograms in Fig. 2 is about 150 nA. The current is about 50-times larger than that at the conventional microdisc electrode with r = 10 μm (see Eq. (1)). The current-potential characteristics of the pseudo-steady-state current at the ultrathin-ring electrode were analyzed on the log-plot analysis (Fig. 3), where the pseudo-steady-state current at 10 s (I(10)) at various potentials was used. In this analysis, the current recorded at 500 mV was defined as the limiting current (I(10)lim). The dotted line in Fig. 3 indicates the guide of the slope for the reversible current response of one-electron oxidation. The experimental data agree well with the guide line. This agreement demonstrates that the reversible and one-electron redox reaction of Fc(CH2OH)2 occurs at the ultrathin-ring electrode. The electrical resistance in the electrode is vanishingly small. The estimated half-wave potential from this analysis is 270 mV, which is in good agreement with the literature value.20 The cyclic voltammetric characteristics at the ultrathin-ring electrode were also examined. Figure 4 shows voltammograms for the oxidation of Fc(CH2OH)2 at various concentrations. The voltammograms have a sigmoidal shape. The half-wave potential of the voltammograms is independent of the concentration of Fc(CH2OH)2 and agrees with that obtained by the log-plot analysis. The inset shows the concentration dependence of the currents at 500 mV on the voltammograms. The currents are proportional to the concentration of Fc(CH2OH)2 with a sensitivity of 150 nA mmol–1 dm3 (Fig. 4, inset) with a detection limit of 10 μmol dm–3. The sensitivity of the ultrathin-ring electrode is large enough to measure the current without a Faraday cage. However, the voltammograms in Fig 4 are somewhat different from the ideal sigmoid curve, as judged from the hysteresis
605
Fig. 3 Logarithm plot of the pseudo-steady state currents recorded at 10 s from the potential step. The limiting current was recorded at 500 mV. The solid line shows the guide for a reversible response.
Fig. 4 Cyclic voltammograms for the oxidation of Fc(CH2OH)2 with the ultrathin-ring electrode recorded at a scan rate of 10 mV s–1. The concentration of Fc(CH2OH)2 is (a) 0, (b) 0.11, (c) 0.22, (d) 0.33, (e) 0.55, (f) 0.77, and (g) 0.99 mmol dm–3. The inset shows the concentration dependence of the current at 500 mV.
characteristics in the forward and reverse scans. Since the hysteresis depends on the concentration of Fc(CH2OH)2, the hysteresis is not caused by the non-faradaic process, but due to the mass-transfer characteristics. Figure 5A shows a numerically simulated cyclic voltammogram at 10 mV s–1 for the reversible reaction at an ultrathin-ring electrode with a diameter of 2 mm and a thickness of 100 nm. The diffusion coefficient, standard redox potential, and concentration for the redox species used here are set as: 6.8 × 10–10 m2 s–1, 0 V, and 1 mmol dm–3, respectively. The hysteresis between the forward and revers scans is well reproduced in the simulated voltammogram. This reveals that the hysteresis is caused by the mass-transfer around the ultrathin-ring electrode. The solid line in Fig. 5B shows a simulated amperogram stepped to the limiting current conditions. The amperogram agrees with the current calculated on Eq. (2). The dotted line in Fig. 5B demonstrates the unfolded view of the simulated voltammogram. Here, the voltammogram is compared with the amperogram with an assumption that the time-dependence of the diffusion current in the linear potential sweep is close to that of the potential step at a time corresponding to that at which the
606
ANALYTICAL SCIENCES JULY 2015, VOL. 31
Fig. 6 Time courses for the relative concentration of Fc(CH2OH)2+ by PQQ-GDH-catalyzed glucose oxidation simultaneously monitored by (a) photometry and (b) amperometry. The thin lines show the guide for the slope at 90 s. The arrow indicates the addition of PQQ-GDH.
Fig. 5 (A) Numerically simulated cyclic voltammograms for the reversible reaction (solid line) at an ultrathin-ring electrode and (triangles) at a microband electrode when v = 10 mV s–1, n = 1, D = 6.8 × 10–10 m2 s–1, c0 = 1 mmol dm–3, w = 100 nm, and l = 6.28 mm. (B) Numerically simulated potential-step chronoamperogram when the potential was stepped from –200 to 200 mV at t = 20 s. The dotted line shows the unfolded voltammogram at the ultrathin-ring electrode given in Panel A.
potential scan passes the standard potential. Under limited conditions, the voltammogram at 10 mV s–1 and the potential step amperogram stepped after 20 s overlap with each other. The overlap indicates that the shape of the voltammogram is caused by the pseudo-steady state characteristics of the ultrathin-ring electrode and that the potential dependence of the pseudo-steady-state current is Nernstian for Fc(CH2OH)2. Since the current density at the ultrathin-ring electrode is very large compared with that of the conventional microelectrode, the signal/noise ratio should also be very large. However, the charging current of the electrode was quite higher than the expected value. For example, the charging current of curve a in Fig. 4 is about 0.4 nA. According to this charging current, the estimated surface capacitance is about 6000 μF cm–2. The estimated value is about 60-times larger than the typical double layer capacitance at the gold surface with a corresponding surface area.21 The quite large surface capacitance will be caused by the roughness of the electrode surface and the partial penetration of the solution into slight gaps between the conductor and the insulator. Although the ultra-thin ring electrode gives non-ideal steady-state characteristics in the current response, the pseudo-steady-state current can be utilized in amperometry for practical purposes, as judged from the fact that Δ quickly decreases down to 1% within 5 s after a potential step. Here, we applied this method to the detection of the concentration change due to a redox enzyme. PQQ-GDH catalyzes the glucose oxidation by several artificial electron acceptors. We
used Fc(CH2OH)2+ as an electron acceptor in this work. The reduction of Fc(CH2OH)2+ by PQQ-GDH-catalyzed glucose oxidation was simultaneously monitored by amperometry and photometry under quiescent conditions. Curves a and b in Fig. 6 show the time course of the normalized value of the Fc(CH2OH)2+ concentration determined from the absorbance at 638 nm and the reduction current at 0 mV, which locates in the limiting region for the reduction of Fc(CH2OH)2+. A stock solution of the enzyme was added into an optical cell filled with 3 cm3 of 25 mmol dm–3 PIPES pH 7 containing 0.29 mmol dm–3 Fc(CH2OH)2+, 100 mmol dm–3 glucose, and 100 mmol dm–3 KCl under stirring at t = 60 s (shown as arrow). The stirring was continued until the solution became uniform. The addition of 0.2 μg cm–3 PQQ-GDH induces the reduction of Fc(CH2OH)2+. The slope of the curve indicates to the reaction rate of the enzymatic reduction in the solution. The initial slope corresponds to the initial rate in the enzymatic kinetics. The estimated value of the slopes at t = 90 s were 2.3 and 2.2 μmol dm–3 s–1 for curves a and b, respectively. The good agreement of the slopes supports that the amperometric method with the ultrathin-ring electrode can be applied to the determination of the redox enzyme activity. The electrochemical method can be applied to colored or turbid samples under quiescent conditions without a Faraday cage.
Conclusions The convenient fabrication method of an ultrathin-ring electrode has been established. The fabricated ultrathin-ring electrode provides a large pseudo-steady-state current with a large current density. The characteristics realize simple amperometric measurements without stirring and without a Faraday cage. Becase of the ultrathin layer of the electrode, the current reachs a pseudo-steady-state within 5 s. This property make it possible The to measure the concentration change monitoring. comparison between the amperometric and photometric monitorings of the concentration change by an enzyme reaction supports the good performance of the amperometric method with the ultrathin-ring electrode proposed here. Cyclic voltammetric response with the non-ideal steady-state property has been clearly characterized. The ultrathin-ring electrode will have wide applications including stripping voltammetry without
ANALYTICAL SCIENCES JULY 2015, VOL. 31 stirring, continuous amperometric monitoring of environmental markers, and quick monitoring of food samples.
References 1. M. V. Mirkin and A. J. Bard, Anal. Chem., 1992, 64, 2293. 2. A. E. G. Cass, G. Davis, G. D. Francis, H. A. O. Hill, W. J. Aston, I. J. Higgins, E. V. Plotkin, L. D. L. Scott, and A. P. F. Turner, Anal. Chem., 1984, 56, 668. 3. R. Matsumoto, K. Kano, and T. Ikeda, J. Electroanal. Chem., 2002, 535, 37. 4. L. C. Clark Jr, R. Wolf, D. Granger, and Z. Taylor, J. Appl. Physiol., 1953, 6, 189. 5. A. J. Bard and L. R. Faulkner, “Electrochemical Methods Fundamentals and Applications”, 2nd ed., 2001, John Wily & Sons, New York. 6. Y. Saito, Rev. Polarogr., 1968, 15, 177. 7. K. Aoki and J. Osteryoung, J. Electroanal. Chem., 1981, 122, 19. 8. A. M. Bond, K. B. Oldham, and C. G. Zoski, Anal. Chim. Acta, 1989, 216, 177. 9. R. J. Forster, Chem. Soc. Rev., 1994, 289. 10. B. R. Scharifker, J. Electroanal. Chem., 1988, 240, 61. 11. T. J. Davies and R. G. Compton, J. Electroanal. Chem.,
607 2005, 585, 63. 12. M. Kudera, Y. Nakagawa, S. Fletcher, and H. A. O. Hill, Lab Chip, 2001, 1, 127. 13. T. Noda, K. Hamamoto, M. Tsutsumi, S. Tsujimura, O. Shirai, and K. Kano, Electrochem. Commun., 2010, 12, 839. 14. T. N. Huan, L. Q. Hung, V. T. T. Ha, N. H. Anh, T. V. Khai, K. B. Shim, and H. Chung, Talanta, 2012, 94, 284. 15. A. Szabo, D. K. Cope, D. E. Tallman, P. M. Kovach, and R. M. Wightman, J. Electroanal. Chem., 1987, 217, 417. 16. M. Fleischmann, S. Bandyopadhyay, and S. Pons, J. Phys. Chem., 1985, 89, 5537. 17. D. R. MacFarlane and D. K. Y. Wong, J. Electroanal. Chem., 1985, 185, 197. 18. S. B. Khoo, H. Gunasingham, K. P. Ang, and B. T. Tay, J. Electroanal. Chem., 1987, 216, 115. 19. J. W. Bixler, M. Fifield, J. C. Poler, A. M. Bond, and W. Thormann, Electroanalysis, 1989, 1, 23. 20. R. Szentrimay, P. Yeh, and T. Kuwana, in “Electrochemical studies of biological systems. ACS symposium series; 38”, ed. D. T. Sawyer, 1977, American Chemical Society, Washington, D.C., 143. 21. P. S. Germain, W. G. Pell, and B. E. Conway, Electrochim. Acta, 2004, 49, 1775.
603
2015 © The Japan Society for Analytical Chemistry
Fabrication and Characterization of Ultrathin-ring Electrodes for Pseudo-steady-state Amperometric Detection Yuki KITAZUMI,† Katsumi HAMAMOTO, Tatsuo NODA, Osamu SHIRAI, and Kenji KANO Division of Applied Life Science, Graduated School of Agriculture, Kyoto University, Kitashirakawaoiwake-cho, Sakyo, Kyoto 606–8502, Japan
The fabrication of ultrathin-ring electrodes with a diameter of 2 mm and a thickness of 100 nm is established. The ultrathin-ring electrodes provide a large density of pseudo-steady-state currents, and realize pseudo-steady-state amperometry under quiescent conditions without a Faraday cage. Under the limiting current conditions, the current response at the ultrathin-ring electrode can be well explained by the theory of the microband electrode response. Cyclic voltammograms at the ultrathin-ring electrode show sigmoidal characteristics with some hysteresis. Numerical simulation reveals that the hysteresis can be ascribed to the time-dependence of pseudo-steady-state current. The performance of amperometry with the ultrathin-ring electrode has been verified in its application to redox enzyme kinetic measurements. Keywords Amperometry, pseudo-steady-state current, microband electrode, kinetic measurements (Received February 25, 2015; Accepted April 22, 2015; Published July 10, 2015)
Introduction The amperometric sensor is a device that detects the concentration of analytes on the basis of the Faradaic current of the electrode reaction. The current reflects the reaction rate on the electrode surface and is proportional to the concentration of the analyte. At conventional electrodes, the electrode reaction increases the depletion of the analyte around the electrode with time, that is, the concentration polarization. The time-dependent characteristic of the diffusion current at conventional electrodes is inconvenient for amperometric analysis and limits the practical application. Therefore, user-friendly and convenient amperometric sensors require the time-independent characteristics of the amperometric response, that is, the steady-state current. Several concepts are adopted to achieve steady-state currents.1 The first one is to control the current by the catalytic reaction rate. This kind of the steady-state current reflects the rate of catalytic reactions, especially of redox enzyme reactions in the presence of excess amounts of enzyme substrates.2,3 The second one is to control the mass transfer of the analyte by transport (or permeation) across a membrane on the electrode surface. This approach has been well and successfully realized in the Clark-type oxygen electrode.4 The third one is convective mass-transfer by using a rotating disc electrode or a thin-layer flow cell. This well-known technique provides useful information about the electrode reaction as well as amperometry.5 The forth one is utilization of the microelectrode based on sufficient mass-transfer by non-linear diffusion.6–9 A microdisc electrode provides a steady-state limiting current (Is,lim) after a suitable electrolysis period under quiescent To whom correspondence should be addressed. E-mail: [email protected]
†
conditions, as expressed by the following equation:6 Is,lim = 4nFDrc0,
(1)
where n, F, D, c0, and r are the number of electrons, the Faraday constant, the diffusion coefficient, the concentration in the bulk phase, and the radius of the electrode, respectively. Unfortunately, the signals of microdisc electrodes are quite small (typically less than a few nanoampere) as compared to the noise from the external disturbance. Therefore, measurements with microdisc electrodes require a Faraday cage. One of the ideas to enhance the signal intensity with the microdisc electrode is to construct of an electrode array. However, the usual microelectrode arrays do not give steady-state currents but a diffusion current with time-dependent characteristics, since the diffusion layer at the microdisc electrodes overlap with each other.10,11 On the other hand, coarsely arranged microdisc electrode arrays are proposed to overcome this problem.12–14 However, the fabrication of the coarsely arranged microdisc electrode array is technically difficult and is not suitable for mass production for practical applications. Another idea to obtain a large steady-state signal is to use a microband electrode.6 The current response of the microband electrode under the limiting current conditions (I(t)lim) is formulated by Szabo et al. as follows:15 I (t ) lim = nFDc 0 l
1 + 1 (τ < 2 / 5) πτ
=
π exp(−2 / 5 πτ ) 4 πτ
+
π (τ > 2 / 5), ln[ 64τ exp(−0.5772156) + exp(5 / 3)]
(2)
where τ (≡ Dt/w2), l, and w are the dimensionless time, the length of the microband, and the thickness of the microband,
604 respectively. Although it takes an extremely long time to reach a true steady-state current at a microband electrode, the pseudo-steady-state current increases with an increase in l.6,15 The density of the pseudo-steady-state current (I(t)lim/lw [t = 10 s, l = 6.28 mm, w = 0.1 μm, D = 1 × 10–9 m2 s–1, c0 = 1 mmol dm–3] = 35 mA cm–2) is much larger than a steady-state current density at a typical microdisc electrode (Is,lim/πr2 [r = 10 μm] = 1.2 mA cm–2). This is a great advantage in practical amperometric applications over the microdisc electrode method. In order to increase l in a rod-type style electrode, thin-ring electrodes can be proposed.16,17 In addition, an ultrathin-ring electrode should also be effective for applications, because a decrease in w should decrease the response time. Although ultrathin-ring electrodes will bring out their advantages in the quiescent solution, ultrathin-ring electrodes for an amperometric detector have only been reported in flow systems.18,19 Moreover, the reported responses of ultrathin-ring electrodes were less than half the theoretical value and largely depend on polishing of the electrode.19 The origin of non-ideal characteristics were predicted discontinuities in the metal surface.19 Therefore, more suitable fabrication of ultrathin-ring electrodes and the characterization remain to be elucidated. To order to solve and clarify these problems about ultrathinring electrodes, the fabrication of an ultrathin-ring electrode is proposed. Furthermore, the current response is analyzed from a theoretical viewpoint. The proposed amperometric method with the ultrathin-ring electrode is applied to the detection of the concentration change due to a redox enzyme reaction under quiescent conditions.
Experimental Preparation of the ultrathin-ring electrode A schematic view of the ultrathin-ring electrode is shown in Fig. 1A. A rod formed by a UV-curable resin (Sunny, UV-3421) with a diameter of 2.0 mm (a) was bound with a copper wire (b). A gold film (c) was prepared at a thickness of 100 nm by sputtering on the side surface of the rod on a desk-top quick coater (SC-704, Sanyu Electron). The sputtering was carried out in twice on each semicylindrical surface of the rod under control of the current and time. The electrode side of the rod was hardened with the UV-curable resin (d). The UV-curable resin was selected on the basis of the adhesion property to gold. The compatibility of materials pinching the gold film was necessary to avoid any cracking at the electrode surface. An epoxy resin (Huntsman, Araldite® 2020) was cast as the body of the electrode (e). The tip of the electrode was cut with a lathe and ground with running water using emery papers (#1500 and then #4000). A photograph in Fig. 1B shows the tip of a fabricated ultrathin-ring electrode. The performance of the ultrathin-ring electrodes was checked by amperometry and cyclic voltammetry. According to this improved procedure, the yield fabricating of the ultrathin-ring electrodes was over 70% within 1% of the variability in the current response. Note here that polishing of the ultrathin-ring electrode with abrasive particles depresses the performance. The established processes are easy to be carried out, because every step includes no delicate operations and is carried out at room temperature. Chemicals 1,1′-Ferrocenedimethanol (Fc(CH2OH)2), potassium chloride, glucose, piperazine-1,4-bis(2-ethanesulfonic acid) (PIPES), and pyrroloquinoline quinone dependent glucose dehydrogenase
ANALYTICAL SCIENCES JULY 2015, VOL. 31
Fig. 1 (A) Schematic view of an ultrathin-ring electrode, where (a) rod formed by UV cured resin, (b) lead wire connected to the gold thin layer, (c) spattered gold layer, (d) tip formed by UV cured resin, and (e) body formed by epoxy resin. (B) A photograph showing the tip of a prepared ultrathin-ring electrode.
(PQQ-GDH, Toyobo) were purchased commercially and used without further purification. A stock solution of 1,1′-bis (hydroxymethyl)ferrocenium (Fc(CH2OH)2+) was prepared as follows. Fc(CH2OH)2 was oxidized with two equivalents of AgNO3 in water. After filtration to remove the product Ag, any excess of Ag+ in the solution was removed as AgCl by the addition of KCl. The resultant solution was used as the stock solution of Fc(CH2OH)2+. Apparatus Electrochemical measurements were carried out by using an electrochemical analyzer (BAS50W, BAS). An Ag | AgCl | sat. KCl electrode and a platinum wire were used as a reference electrode and a counter electrode, respectively. Time-causes of the absorbance during an enzyme reaction were recorded on a UV-Vis spectrophotometer (UV2550, Shimadzu). All measurements were carried out at room temperature. Numerical simulation The reversible electrode reaction at ultrathin-ring electrodes and diffusion equations of redox species were numerically simulated by using a commercial package of the finite-element method, COMSOL 5.0 (COMSOL).
Results and Discussion To investigate the pseudo-steady-state current recorded at the ultrathin-ring electrode, potential-step chronoamperometry was carried out. The solid line in Fig. 2 shows an amperogram recorded under the limiting conditions for the oxidation of 1 mmol dm–3 Fc(CH2OH)2 in 100 mmol dm–3 of KCl at an ultrathin-ring electrode. The current reaches a practical steady-state value at about 10 s. The closed circles in Fig. 2 show the current value calculated on Eq. (2) at w = 100 nm, l = 6.28 mm, and D = 6.8 × 10–10 m2 s–1. Here, the value of D used here for Fc(CH2OH)2 was determined separately by voltammetry on a conventional disc electrode (data not shown). The good agreement between the experimental and the theoretical values indicate that the ultrathin-ring electrode fabricated in the proposed method behaves as a typical microband electrode with
ANALYTICAL SCIENCES JULY 2015, VOL. 31
Fig. 2 (Solid line) A potential-step chronoamperogram for the oxidation of 1 mmol dm–3 Fc(CH2OH)2 in 100 mmol dm–3 KCl. (Closed circles) The current calculated on Eq. (2) with n = 1, D = 6.8 × 10–10 m2 s–1, c0 = 1 mmol dm–3, w = 100 nm, and l = 6.28 mm. (Broken line) Time dependence of the relative change in the current (Δ).
a sufficiently large l value. Although the ultrathin-electrode does not give a true steady-state current within 20 s, as shown in Fig. 2, we may use a pseudo steady-state current for practical applications in amperometry. Here, the relative change in the current, Δ = (dI(t)lim/dt)/I(t), is introduced to estimate the pseudo-steady state of the current. The broken line in Fig. 2 shows the calculated value of Δ. When Δ is set to 1% as a threshold considering the steady-state current, we can conclude the current reaches a pseudo-steady-state value at 5 s after potential step. This quick response of the ultrathin-ring electrode is due to ultrathin characteristics. The pseudo-steady-state current at t = 5 s of the amperograms in Fig. 2 is about 150 nA. The current is about 50-times larger than that at the conventional microdisc electrode with r = 10 μm (see Eq. (1)). The current-potential characteristics of the pseudo-steady-state current at the ultrathin-ring electrode were analyzed on the log-plot analysis (Fig. 3), where the pseudo-steady-state current at 10 s (I(10)) at various potentials was used. In this analysis, the current recorded at 500 mV was defined as the limiting current (I(10)lim). The dotted line in Fig. 3 indicates the guide of the slope for the reversible current response of one-electron oxidation. The experimental data agree well with the guide line. This agreement demonstrates that the reversible and one-electron redox reaction of Fc(CH2OH)2 occurs at the ultrathin-ring electrode. The electrical resistance in the electrode is vanishingly small. The estimated half-wave potential from this analysis is 270 mV, which is in good agreement with the literature value.20 The cyclic voltammetric characteristics at the ultrathin-ring electrode were also examined. Figure 4 shows voltammograms for the oxidation of Fc(CH2OH)2 at various concentrations. The voltammograms have a sigmoidal shape. The half-wave potential of the voltammograms is independent of the concentration of Fc(CH2OH)2 and agrees with that obtained by the log-plot analysis. The inset shows the concentration dependence of the currents at 500 mV on the voltammograms. The currents are proportional to the concentration of Fc(CH2OH)2 with a sensitivity of 150 nA mmol–1 dm3 (Fig. 4, inset) with a detection limit of 10 μmol dm–3. The sensitivity of the ultrathin-ring electrode is large enough to measure the current without a Faraday cage. However, the voltammograms in Fig 4 are somewhat different from the ideal sigmoid curve, as judged from the hysteresis
605
Fig. 3 Logarithm plot of the pseudo-steady state currents recorded at 10 s from the potential step. The limiting current was recorded at 500 mV. The solid line shows the guide for a reversible response.
Fig. 4 Cyclic voltammograms for the oxidation of Fc(CH2OH)2 with the ultrathin-ring electrode recorded at a scan rate of 10 mV s–1. The concentration of Fc(CH2OH)2 is (a) 0, (b) 0.11, (c) 0.22, (d) 0.33, (e) 0.55, (f) 0.77, and (g) 0.99 mmol dm–3. The inset shows the concentration dependence of the current at 500 mV.
characteristics in the forward and reverse scans. Since the hysteresis depends on the concentration of Fc(CH2OH)2, the hysteresis is not caused by the non-faradaic process, but due to the mass-transfer characteristics. Figure 5A shows a numerically simulated cyclic voltammogram at 10 mV s–1 for the reversible reaction at an ultrathin-ring electrode with a diameter of 2 mm and a thickness of 100 nm. The diffusion coefficient, standard redox potential, and concentration for the redox species used here are set as: 6.8 × 10–10 m2 s–1, 0 V, and 1 mmol dm–3, respectively. The hysteresis between the forward and revers scans is well reproduced in the simulated voltammogram. This reveals that the hysteresis is caused by the mass-transfer around the ultrathin-ring electrode. The solid line in Fig. 5B shows a simulated amperogram stepped to the limiting current conditions. The amperogram agrees with the current calculated on Eq. (2). The dotted line in Fig. 5B demonstrates the unfolded view of the simulated voltammogram. Here, the voltammogram is compared with the amperogram with an assumption that the time-dependence of the diffusion current in the linear potential sweep is close to that of the potential step at a time corresponding to that at which the
606
ANALYTICAL SCIENCES JULY 2015, VOL. 31
Fig. 6 Time courses for the relative concentration of Fc(CH2OH)2+ by PQQ-GDH-catalyzed glucose oxidation simultaneously monitored by (a) photometry and (b) amperometry. The thin lines show the guide for the slope at 90 s. The arrow indicates the addition of PQQ-GDH.
Fig. 5 (A) Numerically simulated cyclic voltammograms for the reversible reaction (solid line) at an ultrathin-ring electrode and (triangles) at a microband electrode when v = 10 mV s–1, n = 1, D = 6.8 × 10–10 m2 s–1, c0 = 1 mmol dm–3, w = 100 nm, and l = 6.28 mm. (B) Numerically simulated potential-step chronoamperogram when the potential was stepped from –200 to 200 mV at t = 20 s. The dotted line shows the unfolded voltammogram at the ultrathin-ring electrode given in Panel A.
potential scan passes the standard potential. Under limited conditions, the voltammogram at 10 mV s–1 and the potential step amperogram stepped after 20 s overlap with each other. The overlap indicates that the shape of the voltammogram is caused by the pseudo-steady state characteristics of the ultrathin-ring electrode and that the potential dependence of the pseudo-steady-state current is Nernstian for Fc(CH2OH)2. Since the current density at the ultrathin-ring electrode is very large compared with that of the conventional microelectrode, the signal/noise ratio should also be very large. However, the charging current of the electrode was quite higher than the expected value. For example, the charging current of curve a in Fig. 4 is about 0.4 nA. According to this charging current, the estimated surface capacitance is about 6000 μF cm–2. The estimated value is about 60-times larger than the typical double layer capacitance at the gold surface with a corresponding surface area.21 The quite large surface capacitance will be caused by the roughness of the electrode surface and the partial penetration of the solution into slight gaps between the conductor and the insulator. Although the ultra-thin ring electrode gives non-ideal steady-state characteristics in the current response, the pseudo-steady-state current can be utilized in amperometry for practical purposes, as judged from the fact that Δ quickly decreases down to 1% within 5 s after a potential step. Here, we applied this method to the detection of the concentration change due to a redox enzyme. PQQ-GDH catalyzes the glucose oxidation by several artificial electron acceptors. We
used Fc(CH2OH)2+ as an electron acceptor in this work. The reduction of Fc(CH2OH)2+ by PQQ-GDH-catalyzed glucose oxidation was simultaneously monitored by amperometry and photometry under quiescent conditions. Curves a and b in Fig. 6 show the time course of the normalized value of the Fc(CH2OH)2+ concentration determined from the absorbance at 638 nm and the reduction current at 0 mV, which locates in the limiting region for the reduction of Fc(CH2OH)2+. A stock solution of the enzyme was added into an optical cell filled with 3 cm3 of 25 mmol dm–3 PIPES pH 7 containing 0.29 mmol dm–3 Fc(CH2OH)2+, 100 mmol dm–3 glucose, and 100 mmol dm–3 KCl under stirring at t = 60 s (shown as arrow). The stirring was continued until the solution became uniform. The addition of 0.2 μg cm–3 PQQ-GDH induces the reduction of Fc(CH2OH)2+. The slope of the curve indicates to the reaction rate of the enzymatic reduction in the solution. The initial slope corresponds to the initial rate in the enzymatic kinetics. The estimated value of the slopes at t = 90 s were 2.3 and 2.2 μmol dm–3 s–1 for curves a and b, respectively. The good agreement of the slopes supports that the amperometric method with the ultrathin-ring electrode can be applied to the determination of the redox enzyme activity. The electrochemical method can be applied to colored or turbid samples under quiescent conditions without a Faraday cage.
Conclusions The convenient fabrication method of an ultrathin-ring electrode has been established. The fabricated ultrathin-ring electrode provides a large pseudo-steady-state current with a large current density. The characteristics realize simple amperometric measurements without stirring and without a Faraday cage. Becase of the ultrathin layer of the electrode, the current reachs a pseudo-steady-state within 5 s. This property make it possible The to measure the concentration change monitoring. comparison between the amperometric and photometric monitorings of the concentration change by an enzyme reaction supports the good performance of the amperometric method with the ultrathin-ring electrode proposed here. Cyclic voltammetric response with the non-ideal steady-state property has been clearly characterized. The ultrathin-ring electrode will have wide applications including stripping voltammetry without
ANALYTICAL SCIENCES JULY 2015, VOL. 31 stirring, continuous amperometric monitoring of environmental markers, and quick monitoring of food samples.
References 1. M. V. Mirkin and A. J. Bard, Anal. Chem., 1992, 64, 2293. 2. A. E. G. Cass, G. Davis, G. D. Francis, H. A. O. Hill, W. J. Aston, I. J. Higgins, E. V. Plotkin, L. D. L. Scott, and A. P. F. Turner, Anal. Chem., 1984, 56, 668. 3. R. Matsumoto, K. Kano, and T. Ikeda, J. Electroanal. Chem., 2002, 535, 37. 4. L. C. Clark Jr, R. Wolf, D. Granger, and Z. Taylor, J. Appl. Physiol., 1953, 6, 189. 5. A. J. Bard and L. R. Faulkner, “Electrochemical Methods Fundamentals and Applications”, 2nd ed., 2001, John Wily & Sons, New York. 6. Y. Saito, Rev. Polarogr., 1968, 15, 177. 7. K. Aoki and J. Osteryoung, J. Electroanal. Chem., 1981, 122, 19. 8. A. M. Bond, K. B. Oldham, and C. G. Zoski, Anal. Chim. Acta, 1989, 216, 177. 9. R. J. Forster, Chem. Soc. Rev., 1994, 289. 10. B. R. Scharifker, J. Electroanal. Chem., 1988, 240, 61. 11. T. J. Davies and R. G. Compton, J. Electroanal. Chem.,
607 2005, 585, 63. 12. M. Kudera, Y. Nakagawa, S. Fletcher, and H. A. O. Hill, Lab Chip, 2001, 1, 127. 13. T. Noda, K. Hamamoto, M. Tsutsumi, S. Tsujimura, O. Shirai, and K. Kano, Electrochem. Commun., 2010, 12, 839. 14. T. N. Huan, L. Q. Hung, V. T. T. Ha, N. H. Anh, T. V. Khai, K. B. Shim, and H. Chung, Talanta, 2012, 94, 284. 15. A. Szabo, D. K. Cope, D. E. Tallman, P. M. Kovach, and R. M. Wightman, J. Electroanal. Chem., 1987, 217, 417. 16. M. Fleischmann, S. Bandyopadhyay, and S. Pons, J. Phys. Chem., 1985, 89, 5537. 17. D. R. MacFarlane and D. K. Y. Wong, J. Electroanal. Chem., 1985, 185, 197. 18. S. B. Khoo, H. Gunasingham, K. P. Ang, and B. T. Tay, J. Electroanal. Chem., 1987, 216, 115. 19. J. W. Bixler, M. Fifield, J. C. Poler, A. M. Bond, and W. Thormann, Electroanalysis, 1989, 1, 23. 20. R. Szentrimay, P. Yeh, and T. Kuwana, in “Electrochemical studies of biological systems. ACS symposium series; 38”, ed. D. T. Sawyer, 1977, American Chemical Society, Washington, D.C., 143. 21. P. S. Germain, W. G. Pell, and B. E. Conway, Electrochim. Acta, 2004, 49, 1775.
