Solution - cast films of poly(3,4-ethylenedioxythiophene) as ion-to-electron transducers in all-solid-state ion-selective electrodes

Mercedes Vázquez , Petter Danielsson, Johan Bobacka , Andrzej Lewenstam and

Ari Ivaska

Åbo Akademi University, Process Chemistry Group,

c/o Laboratory of Analytical Chemistry, Biskopsgatan 8, FIN-20500, Åbo-Turku, Finland

Abstract

An aqueous dispersion of poly(3,4-ethylenedioxythiophene) (PEDOT) doped with poly(styrenesulfonate) (PSS) was cast on screen-printed gold substr 737e412h ates. PEDOT(PSS) was ionically (physically) crosslinked by multivalent cations, including Mg²⁺, Ca²⁺, Fe^2+/3+ and Ru(NH₃)₆^2+/3+, in order to decrease the water solubility of the PEDOT(PSS) films. The resulting Au/PEDOT(PSS) electrodes were characterized by cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS) and energy-dispersive X-ray analysis (EDXA). Ionic crosslinking of PEDOT(PSS) with Fe^2+/3+ and Ru(NH₃)₆^2+/3+ was faster than with Mg²⁺ and Ca²⁺ ions and resulted in PEDOT(PSS) films that were less soluble in water. Among the multivalent cations tested, Ru(NH₃)₆^2+/3+ resulted in PEDOT(PSS) films with the most stable potential. Incorporation of Ru into PEDOT(PSS) was shown by EDXA. Based on CV and EIS, ionic crosslinking of PEDOT(PSS) with the electroactive Ru(NH₃)₆^2+/3+ was found to increase the bulk redox capacitance of the PEDOT(PSS) film, compared to ionic crosslinking with electroinactive Mg²⁺ ions. Potentiometric measurements showed that PEDOT(PSS) ionically crosslinked with Ru(NH₃)₆^2+/3+ was less sensitive to CO₂ (pH) than the bare Au substrate. The Au/PEDOT(PSS) electrodes were found to work well as ion-to-electron transducer (solid contact) in all-solid-state K⁺-ISEs.

Keywords: conducting polymer; poly(3,4-ethylenedioxythiophene); Baytron P; ionic crosslinking; ion-to-electron transducer; all-solid-state ion-selective electrode.

1. Introduction

All-solid-state ion selective electrodes based on polymeric membranes containing neutral or charged ionophores (carriers) have attracted considerable interest since the discovery of the so-called coated-wire electrode (CWE) more than 30 years ago [1]. The main issue of concern in ion-selective electrodes (ISEs) with a solid internal contact (solid-contact ISEs, SCISEs) is the reversibility and stability of the ion-to-electron transduction process between the ionically conducting ion-selective membrane and the electronically conducting substrate [2]. Electroactive conjugated- and redox polymers such as poly(1-pyreneamine) [3], polypyrrole [4], poly(3-octylthiophene) [5], poly(vinyl ferrocene) [6], polyindole [7], polyaniline [8] and poly(3,4-ethylenedioxythiophene) [9] have been applied as an intermediate layer in solid-contact ISEs in order to improve the ion-to-electron transduction process. Polypyrrole has been applied as solid contact also in ion-selective microelectrodes [10] and miniaturized screen-printed ISEs [11]. Electroactive polymer layers are often deposited by electropolymerization [3-5, 7-11], but also solution casting has been employed [6].

Poly(3,4-ethylenedioxythiophene) (PEDOT) is known as one of the most stable conducting polymers available today [12]. By using PEDOT as solid contact material, it was shown that the potential stability of solid-contact ISEs is directly related to the redox capacitance of the solid contact material [9]. Additionally, PEDOT was found to be particularly suitable as solid-contact material in ISEs due to its low sensitivity to O₂ and CO₂ (pH) [13]. When used as ion-to-electron transducer, PEDOT doped with poly(styrenesulfonate) (PSS) was prepared by electrochemical polymerization of 3,4-ethylenedioxythiophene (monomer) on a glassy carbon disk electrode and subsequently coated by a plasticized PVC-based ion-selective membrane by solution casting [9, 13-16]. However, from the electrode manufacturing point of view, it may be advantageous to deposit the conducting polymer solid contact from solution by using a dispensing system [17], rather than by electropolymerization.

PEDOT doped with the PSS polyanion is commercially available in the form of an aqueous dispersion (Baytron P, Bayer AG, Germany). Baytron P is an aqueous dispersion of the oxidized (p-doped) PEDOT containing PSS as counterion, where the latter is in excess. The excess negative charge of PSS with respect to the positively charged chains of oxidized PEDOT makes the colloidal particles stable in the aqueous media [18,19]. Solution-cast films of PEDOT(PSS) (Baytron P) tend to swell and disintegrate in aqueous solutions of low-ionic-strength [21]. However, the excess of negative charge allows the material to be ionically (physically) crosslinked with multivalent cations such as Mg²⁺ resulting in a conducting hydrogel that is stable in aqueous media [18-22].

In the present work, polymer films were cast from the aqueous dispersion of PEDOT(PSS) (Baytron P) on screen-printed Au electrodes. In order to decrease the water-solubility of the resulting PEDOT(PSS) film, ionic crosslinking by multivalent cations was employed [18-22]. Electroactive multivalent cations were used for crosslinking in an attempt to stabilize the standard potential of the conducting polymer layer, which was subsequently coated with a plasticized PVC membrane containing valinomycin as K⁺ ionophore.

2. Experimental

2.1. Chemicals

Valinomycin (potassium ionophore I), potassium tetrakis(4-chlorophenyl)borate (KTpClPB), bis(2-ethylhexyl)sebacate (DOS), poly(vinyl chloride) of high molecular weight (PVC) and tetrahydrofuran (THF) were Selectophore reagents obtained from Fluka. The aqueous dispersion of the polymer poly(3,4-ethylenedioxythiophene) (PEDOT) doped with polystyrenesulfonate (PSS) known as Baytron P was obtained from Bayer AG. The polymer dispersion contains ca. 0.5 % (w/w) PEDOT and ca. 0.8 % (w/w) PSS, i.e. a slight molar excess of PSS compared to PEDOT on the basis of monomer units. The pH buffer solutions were obtained from Oy FF-Chemicals Ab. All other chemicals were analytical-reagent grade. ELGA ultrapure water (resistivity 18.2 MW cm) was used to prepare all solutions.

2.2. Screen-printed electrodes

The screen-printed Au electrodes used were obtained from Gwent Electronic Material Ltd., UK. They consisted of a series of six stripes of Au printed in parallel on an alumina substrate and subsequently coated with an insulating layer with 6 openings for the six Au electrodes (area 0.07 cm²) at one end, and the corresponding electrical contacts at the other end (Fig. 1a). On top of the screen-printed Au substrates, 10 ml of the aqueous dispersion of PEDOT(PSS) (Baytron P) was applied and allowed to dry (Fig. 1b). Assuming a density of 1 g/cm³ for PEDOT(PSS) films, the film thickness is on the order of 20 mm. The resulting electrodes are referred to as Au/PEDOT(PSS). In order to make the polymer stable in aqueous solution and prevent excessive swelling and disintegration, it was ionically crosslinked with a multivalent cation. Different cations and cationic redox-couples were tested: Mg²⁺, Ca²⁺, Fe²⁺/Fe³⁺ and Ru(NH₃)₆²⁺/ Ru(NH₃)₆³⁺. The cation solutions were prepared from MgSO₄, CaSO₄, FeCl₂/FeCl₃ and Ru(NH₃)₆Cl₂/ Ru(NH₃)₆Cl₃, respectively. The total concentration of the multivalent cations in the crosslinking solutions was 0.25 M, unless stated otherwise. The proportion of both cationic species of the redox-couples was 1:1, i.e. 0.125 M Fe²⁺ and 0.125 M Fe³⁺ in iron solutions. The Au/PEDOT(PSS) electrodes were immersed in 0.25 M Mg²⁺, Ca²⁺ and Fe²⁺/Fe³crosslinking solutions for at least 5 h. Alternatively, 10 ml of the crosslinking solution (0.25 M Ru(NH₃)₆²⁺/ Ru(NH₃)₆³⁺) was applied directly on top of the conducting polymer layer for at least 30 min. In some series of screen-printed electrodes crosslinking solutions containg only Ru(NH₃)₆³⁺ were also used and it will be indicated when necessary. Afterwards, the electrodes were thoroughly washed with ultrapure water and allowed to dry. Finally, the polymer layer was coated with 50 ml of the K⁺-selective membrane cocktail, i.e. membrane components dissolved in THF (Fig. 1c). The final composition of the ion-selective membrane was the following: 1.00 % (w/w) valinomycin, 0.22 % (w/w) KTpClPB, 65.85 % (w/w) DOS, and 32.93 % (w/w) PVC. The resulting all-solid-state K⁺-ISEs were conditioned in 0.1 M KCl at least 1d prior to and between measurements.

2.3. Potentiometric Measurements

The potentiometric measurements were performed with a homemade multi-channel mV-meter connected to a PC for data acquisition. The reference electrode was a Ag AgCl KCl (3 M) electrode. The potentiometric response of the different types of K⁺-ISEs fabricated was studied in KCl aqueous solutions with and without background electrolyte (0.1 M NaCl).

The sensitivity to CO₂ (pH) of all-solid-state K⁺-ISE was studied by bubbling alternatively CO₂ and Ar directly into the unbuffered solution. When being purged with CO₂, the pH of the aqueous solution decreases due to formation and dissociation of carbonic acid. Thus, a glass pH electrode (Orion Ross Sure-Flow Combination for pH) was used for simultaneous measurement of the solution pH.

All the measurements were performed at room temperature (23 2 C). The activity coefficients were calculated according to the extended Debye-Hückel equation [23]. No correction for the liquid-junction potential was applied (Henderson equation).

2.4. Electrochemical impedance spectroscopy (EIS) and cyclic voltammetry (CV)

Electrochemical impedance spectroscopy (EIS) and cyclic voltammetry (CV) measurements were performed by using an Autolab General Purpose Electrochemical System and Autolab Frequency Response Analyzer System (AUT20.FRA2-AUTOLAB, Eco Chemie, B.V., The Netherlands). EIS and CV measurements were performed by using a one-compartment three-electrode electrochemical cell where the screen-printed electrode under study was connected as the working electrode. A GC rod was used as auxiliary electrode and the reference electrode was a Ag AgCl KCl(3 M) electrode. The impedance spectra were recorded in the frequency range 100 KHz - 10 mHz by using an ac excitation amplitude of 10 mV. CV was performed in the potential range -0.5 to 0.5 V at a potential scan rate of 0.1 V s^-1. The measurements were performed in 0.1 M KCl solutions at room temperature (23 2 C).

2.5. Energy dispersive X-ray analysis (EDXA)

Energy dispersive X-ray analysis (EDXA) measurements were performed with a Leica Cambridge Ltd Stereoscan 360 scanning electron microscope. The sample was prepared by casting 500 ml of Baytron P on a PTFE substrate. After evaporation of water, the Baytron P film was conditioned in 0.1 M Ru(NH₃)₆Cl₂ / Ru(NH₃)₆Cl₃ solution for 24 h before EDAX measurement.

3. Results and Discussion

3.1. All-solid-state K⁺-ISEs

PEDOT(PSS) was deposited onto the gold electrodes by solution casting of the aqueous polymer dispersion (Baytron P). After drying at room temperature, the Au/PEDOT(PSS) electrodes were equilibrated with 0.25 M Mg²⁺, Ca²⁺, Fe²⁺/Fe³⁺ and Ru(NH₃)₆²⁺/ Ru(NH₃)₆³⁺ solutions before application of the ion-selective membrane. The calibration plots of the resulting all-solid-state K⁺-ISEs after 1 day and 14 days in 0.1 M KCl conditioning solution are shown in Fig. 2. The calculated slopes and standard potential values of the calibration plots are shown in Table 1. From the values of the slopes of the different calibration plots it can be seen that pretreatment with Fe²⁺/Fe³⁺ results in K⁺-ISEs showing near Nernstian behavior. The good reproducibility of the standard potential of the new electrodes (after 1 day in conditioning solutions) pretreated with Fe^2+/3+ solutions (E⁰= 630 ± 1, n = 4) is noteworthy. However, when considering the long-term stability, those electrodes experience a significant potential drift, and after 14 days in conditioning solution the standard potential drop is ~ 80 mV (n=4). In the case of electrodes treated with Mg²⁺ and Ca²⁺ solutions the potential drop after 14 days is even ~ 179 mV and ~ 147 mV (n=4), respectively. However, for electrodes treated with Ru(NH₃)₆^2+/3+ solutions, the potential drops only ~ 15 mV (n=4) in 14 days. Having a suitable redox potential, the ruthenium redox couple seems to contribute to the potential stability of the electrodes. The redox potential of the iron couple is relatively high and therefore the progressive reduction of Fe³⁺ to Fe²⁺ could cause the continuous negative drift of the standard electrode potential.

Ionic crosslinking of PEDOT(PSS) with Ru(NH₃)₆^2+/3+ and Fe^2+/3+ ions was very efficient as only 30 min in contact with the crosslinking solution was enough for the formation of a water-insoluble polymer. On the contrary, such a procedure was not successful for ionic crosslinking of the conducting polymer layer in Ca²⁺ and Mg²⁺ solutions. In the latter cases, the conducting polymer films tended to peal off the electrode surface when the electrodes were thoroughly washed with ultrapure water after the crosslinking step. Further trials were made with the Ru(NH₃)₆^2+/3+ couple by lowering its concentration in the crosslinking solution. As shown in Fig. 3, EDAX measurements revealed the presence of the Ru in the conducting polymer film after being conditioned in 0.1 M Ru(NH₃)₆^2+/3+ crosslinking solution for 24 h. Therefore, the concentration of Ru(NH₃)₆^2+/3+ cations can be lowered to 0.1 M with no risk of disintegration of the polymer film in aqueous solutions.

Considering the acceptable potentiometric response and good potential stability shown by electrodes treated with Ru(NH₃)₆^2+/3+, as well as the excellent procedure efficacy, the Ru(NH₃)₆^2+/3+ redox couple can be considered as a good candidate for ionic crosslinking compared to the rest of the ions tested.

A comparison of Figs. 2a and 2b reveal some differences in the calibration plots at the lowest concentration studied (10^-6 M), indicating some deterioration in electrode performance with time. Furthermore, similar all-solid-state K⁺-ISEs based on the same screen-printed substrates with electropolymerized PEDOT(PSS) as solid contact showed some redox sensitivity after prolonged conditioning in 0.1 M KCl. These results indicate that the adhesion between the ion-selective membrane and the insulating layer of the screen-printed substrate becomes a limiting factor in these all-solid-state ISEs after prolonged use. Optimization of the screen-printed Au electrodes to improve their long-term stability was not attempted, because the main objective of this work was to study the solution-cast PEDOT(PSS) as ion-to-electron transducer.

3.2. EIS and CV of Au/PEDOT(PSS)

Au/PEDOT(PSS) electrodes was studied by electrochemical impedance spectroscopy (EIS). Figs. 4a and 4b show the impedance plot of screen-printed Au/PEDOT(PSS) electrodes after ionic crosslinking of the conducting polymer with 0.25 M Ru(NH₃)₆^2+/3+ and Mg²⁺ solutions, respectively. The ionic crosslinking procedure applied was the same described in the previous section. The shape of the impedance spectra in Fig. 4 is similar to that observed earlier for electropolymerized PEDOT(PSS) [9,24], except for a more pronounced charge-transfer semicircle and more extended ~45º Warburg diffusion line. These differences may be partly ascribed to differences in film thickness, i.e. the solution-cast films PEDOT(PSS) studied here are thicker than the electropolymerized PEDOT(PSS) films studied earlier [9,24]. The low-frequency part of the impedance spectra show a characteristic ~90˚ line from the bulk redox capacitance of the polymer film. Approximate values for the redox capacitances (C_LF) of PEDOT(PSS) films ionically crosslinked with Ru(NH₃)₆^2+/3+ and Mg²⁺ ions were calculated from the impedance spectra recorded for different Au/PEDOT(PSS) electrodes as follows:

C_LF = 1 / (2 π f Z'') (1)

where f = 10 mHz and Z'' is the value of the imaginary impedance at that frequency. The redox capacitance values obtained were 5.10 ± 0.72 mF (n = 21 different film) for conducting polymer films treated with Ru(NH₃)₆^2+/3+ solutions, and 3.20 ± 0.29 mF (n = 28 different films) for those treated with Mg²⁺ solutions. Both types of films revealed large redox capacitances in the order of mF, though the ionic crosslinking with Ru(NH₃)₆^2+/3+ ions resulted in ca. 1.6 times increase in redox capacitance compared to ionic crosslinking with Mg²⁺ ions. These results agree with those obtained by CV for the same type of electrodes, as shown in Figs. 5a and 5b, where a larger current can be observed for films treated with Ru(NH₃)₆^2+/3+ ions. On the other hand, no big difference in CVs are observed for films treated with Ru(NH₃)₆^2+/3+ and Ru(NH₃)₆³⁺ solutions (Figs. 5b and 5c). This leads to the conclusion that Ru(III) ions and Ru(III)/Ru(II) redox couple are equally effective for ionic crosslinking of PEDOT(PSS) aqueous dispersion.

3.3. CO₂ sensitivity

In order to show a stable potential, all-solid-state ISEs should be insensitive to interferents such a dissolved gases present in the sample solution. Influence of CO₂ (pH) on the potential of PEDOT was previously found to be more important than the influence of O₂ [13]. The sensitivity to CO₂ of the screen-printed all-solid-state K⁺-ISEs was studied in 0.1 M KCl unbuffered solutions by bubbling alternatively Ar and CO₂ directly into the solution. The pH of the aqueous solution changed along the experiment covering the range from 7.5-5.5 when Ar was bubbled to ca. 3.8 when CO₂ was bubbled.

First, the potentiometric response to CO₂ of screen-printed electrodes was measured before and after covering the Au substrates with solution-cast PEDOT(PSS) ionically crosslinked by equilibration with 10 ml of 0.25 M Ru(NH₃)₆³⁺ solution for at least 30 min. (Fig. 6a). From these measurements it can be seen that after deposition of the conducing polymer on the Au substrates the sensitivity to CO₂ (pH) of the resulting Au/PEDOT(PSS) electrodes (Fig.6a, curve 2) has decreased drastically compared to that shown by the bare Au electrodes (Fig. 6a, curve 1). Furthermore, the Au/PEDOT(PSS) electrodes were found to show a more reproducible response to CO₂ than the bare Au electrodes that exhibited a broader range of responses. With the aging of the Au/PEDOT(PSS) electrodes, the sensitivity to CO₂ decreases slightly as can be seen in curve 3 of Fig. 6a.

Deposition of the ion-selective membrane on top of the Au/PEDOT(PSS) electrode reduces the sensitivity to CO₂ (pH) of the resulting all-solid-state K⁺-ISEs (Fig. 6b, curve 2) compared to the Au/PEDOT(PSS) electrodes (Fig. 6b, curve 1). Consequently, the sensitivity to CO₂ of the all-solid-state K⁺-ISEs is very low in freshly prepared electrodes, i. e. a potential jump of less than 5 mV is observed when CO₂ is bubbled into the sample solution. Without the ISM, the Au/PEDOT(PSS) electrodes exhibit an increment in the sensitivity to CO₂ in the order of 25-35 mV. These results agree well with those obtained for all-solid-state K⁺-ISEs with electrochemically polymerized PEDOT(PSS) as solid-contact material [13].

4. Conclusions

Solution-cast films of poly(3,4-ethylenedioxythiophene) doped with poly(styrenesulfonate), PEDOT(PSS), were found to work well as ion-to-electron transducer (solid contact) in all-solid-state K⁺-ISEs. Solution casting of PEDOT(PSS) from an aqueous dispersion (Baytron P) represents an advantage from the manufacturing point of view in comparison to electrochemical polymerization. Ionic (physical) crosslinking of PEDOT(PSS) with electroactive Ru(NH₃)₆^2+/3+ cations increased the total bulk redox capacitance of PEDOT(PSS) and improved the potential stability of the resulting all-solid-state K⁺-ISE, compared to PEDOT(PSS) crosslinked with electroinactive cations such as Mg²⁺. The overall long-term stability of the all-solid-state K⁺-ISE seems be limited by the adhesion of the plasticized PVC membrane to the screen-printed Au electrode substrates rather than by the PEDOT(PSS) itself.

Acknowledgments

Financial support from the National Technology Agency (TEKES) and Thermo Clinical Labsystems is gratefully acknowledged. The authors are grateful to Clifford Ekholm for EDXA measurements. This work is part of the activities at the Åbo Akademi Process Chemistry Group within the Finnish Centre of Excellence Programme (2000 - 2005) by the Academy of Finland.

References

R.W. Cattrall, H. Freiser, Anal. Chem. 43 (1971) 1905.

B.P. Nikolskii and E.A. Materova, Ion-Sel. Electrode Rev. 7 (1985) 3.

N. Oyama, T. Ohsaka, F. Yoshimura, M. Mizunuma, S. Yamaguchi, N. Ushizawa, T. Shimomura, J. Macromol. Sci., Chem. A25 (1988) 1463.

A. Cadogan, Z. Gao, A. Lewenstam, A. Ivaska, D. Diamond, Anal. Chem. 64 (1992) 2496.

J. Bobacka, M. McCarrick, A. Lewenstam, A. Ivaska, Analyst 119 (1994) 1985.

P.C. Hauser, D.W.L. Chiang, G.A. Wright, Anal. Chim. Acta 302 (1995) 241.

P.C. Pandey, R. Prakash, Sens. Actuators B 46 (1998) 61.

G. Cui, J.S. Lee, S.J. Kim, H. Nam, G.S. Cha, H.D. Kim, Analyst 123 (1998) 1855.

J. Bobacka, Anal. Chem. 71 (1999) 4932.

R.E. Gyurcsányi, A-S. Nybäck, K. Tóth, G. Nagy, A. Ivaska, Analyst 123 (1998) 1339.

R. Zielińska, E. Mulik, A. Michalska, S. Achmatowicz, M. Maj-Żurawska, Anal. Chim. Acta 451 (2002) 243.

G. Heywang, F. Jonas, Adv. Mater. (Weinheim, Ger.) 4 (1992) 116.

M. Vázquez, J. Bobacka, A. Ivaska, A. Lewenstam, Sens. Actuators B 82 (2002) 7.

J. Bobacka, T. Lahtinen, J. Nordman, S. Häggström, K. Rissanen, A. Lewenstam, A. Ivaska, Electroanalysis 13 (2001) 723.

J. Bobacka, A. Lewenstam, A. Ivaska, J. Electroanal. Chem. 509 (2001) 27.

J. Bobacka, T. Alaviuhkola, V. Hietapelto, H. Koskinen, A. Lewenstam, M. Lämsä, J. Pursiainen, A. Ivaska, Talanta 58 (2002) 341.

F. Song, J. Ha, B. Park, T.H. Kwak, I.T. Kim, H. Nam, G.S. Cha, Talanta 57 (2002) 263.

S. Ghosh, J. Rasmusson; O. Inganäs, Adv. Mater. 10 (1998) 1097.

S. Ghosh, O. Inganäs, Synth. Met. 101 (1999) 413.

S. Ghosh, O. Inganäs, Adv. Mater. 11 (1999) 1214.

S. Ghosh, O. Inganäs, J. Electrochem. Soc. 147 (2000) 1872.

S. Ghosh, O. Inganäs, Synth. Met. 121 (2001) 1321.

[23] J. Koryta, J. Dvorak, L. Kavan, Principles of Electrochemistry, 2^nd ed., Wiley, Chichester, 1993, p. 38.

J. Bobacka, A. Lewenstam, A. Ivaska, J. Electroanal. Chem. 489 (2000) 17.

Tables

Table 1. Slope and standard potential (mean values ± S.D., n = 4) calculated from the linear range of the calibration plots (10^-1-10^-4 M) for screen-printed all-solid-state K⁺-ISEs after conditioning in KCl solution for 1 and 14 days.

Figure captions

Fig. 1. Screen-printed gold substr 737e412h ates (a) bare; (b) after deposition of the PEDOT dispersion (Baytron P) layer; (c) after deposition of the ISM layer.

Fig. 2. Calibration plot for screen-printed all-solid-state K⁺-ISEs after 1d (a) and 14 days (b) in KCl conditioning solutions. Ionic crosslinking of Baytron P layer made in 0.25 M Mg²⁺ (■), Ca²⁺ ) and Fe²⁺/Fe³⁺ ( ) solutions for at least 5h, and with 10 ml of 0.25 M Ru(NH₃)₆²⁺/ Ru(NH₃)₆³⁺ ( ) solution for at least 30 min. Calibration plots were recorded in 10^-1-10^-6 M KCl solutions with a constant background electrolyte (0.1 M NaCl).

Fig. 3. EDAX measurement of the conducting polymer PEDOT(PSS) film (Baytron P) conditioned in 0.1 M Ru(NH₃)₆²⁺/Ru(NH₃)₆³⁺ solutions for 24 h.

Fig. 4. Impedance spectra of screen-printed Au/PEDOT(PSS) electrodes equilibrated (a) with 10 ml of 0.25 M Ru(NH₃)₆²⁺/ Ru(NH₃)₆³⁺ solutions for at least 30 min; (b) with 0.25 M Mg²⁺ solutions for at least 5h. Spectra recorded in deaerated 0.1 M KCl solutions at E_dc = 0.2 V and ΔE_ac = 0.1 V. Frequency range = 10 KHz - 0.01 Hz (inserted figures, frequency range = 10 KHz - 3.16 Hz).

Fig. 5. Cyclic voltamogramms (CVs) of screen-printed Au/PEDOT(PSS) electrodes equilibrated (a) with 0.25 M Mg²⁺ solutions for at least 5h, (b) with 10 ml of 0.1 M Ru(NH₃)₆²⁺/ Ru(NH₃)₆³⁺ solutions for at least 30 min, (c) with 10 ml of 0.1 M Ru(NH₃)₆³⁺ solutions for at least 30 min. CVs recorded in deaerated 0.1 M KCl solutions. Potential scan rate = 0.1 V s^-1.

Fig. 6. Potentiometric response to CO₂ and Ar in 0.1 M KCl unbuffered solution of: (a) (1) the screen-printed bare Au electrode; (2) the three-day old screen-printed Au/PEDOT(PSS) electrode; (3) the twenty-three-day old screen-printed Au/PEDOT(PPS) electrode; (b) (1) the screen-printed Au/PEDOT(PPS) electrode; (2) the screen-printed Au/PEDOT(PPS)/K⁺-ISM electrode. PEDOT(PPS) layer prepared by solution casting of the polymer dispersion Baytron P.

Fig. 1.

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

Fig. 6.