Silver Chloride / Ferricyanide-Based Quasi-Reference Electrode for Potentiometric Sensing Applications

: Processes’ occurring at the Ag / AgCl / Cl – , ([Fe(CN) 6 ] 3– / 4– ) ions interface study results are presented. Conditions are selected for the mixed salts’ precipitate formation on the silver surface. It has been shown that the potential of a silver screen-printed electrode (AgSPE) coated with a mixed precipitate containing silver chloride / ferricyanide is stable in the presence of [Fe(CN) 6 ] 3– / 4– . The electrode can serve as a quasi-reference electrode (QRE) in electrochemical measurements in media containing ions [Fe(CN) 6 ] 3– / 4– . The electrode is formed during polarization of AgSPE (0.325 V vs. Ag / AgCl / KCl, 3.5 M) in a solution containing chloride- and ferri / ferrocyanides ions. The results of the obtained QRE study by potentiometry, scanning electron microscopy and cyclic voltammetry are presented. The proposed QRE was used in a sensor system to evaluate the antioxidant activity (AOA) of solutions by hybrid potentiometric method (HPM). The results of AOA assessment of fruit juices and biofluids obtained using new QRE and commercial Ag / AgCl RE with separated spaces do not differ.


Electrodes
A platinum electrode (disk diameter 3 mm) in a polyether ether ketone case type 6.1204.310 (Metrohm AG, Switzerland) was used as a control indicator electrode (Pt). Before starting work, Pt was polished using aluminum oxide powder (first used, a grain size of 0.3 µm, then 0.05 µm) deposited on a polishing cloth, and additionally subjected to a cyclic polarization in the potential range from -0.2 to 1.5 V and scanning rate 0.1 V/s in 0.1 M H 2 SO 4 solution until a stable cyclic voltammogram was obtained [56]. After each regeneration step, Pt was washed with DI water. Control reference electrode was a silver/silver chloride electrode Ag/AgCl/KCl (3.5 M) type 6.0728.040 completed with an electrolyte vessel type 6.1245.010 (Metrohm AG, Switzerland). A silver/silver chloride electrode Ag/AgCl/KCl (3.5 M) type EVL-1M3.1 (JSC Gomel Plant of Measuring Devices, Belorussia) was used in the measurements, unless otherwise indicated. Potential of the EVL-1M3.1 at 25 • C is 0.204 ± 0.003 V relative to SHE [57]. Potential of the EVL-1M3.1 was checked daily in relation to the reference electrode (a deviation of ± 3 mV was considered acceptable). Glassy carbon rod GC-2000 3 mm/100 mm (JSC RI Grafit, Russia) was used as an auxiliary electrode in voltammetric measurements.
AgSPE was prepared by applying two layers of silver paste on a ceramic plate washed with acetone, ethanol and deionized water. Each layer of paste was hardened as follows: heated to 850 • C for 30 min, cooled to 150 • C for 30 min, dried at 150 • C for 15 min and cooled to room temperature. Double paste application provided an average silver layer thickness of 30 µm. The plate was cut and electrodes 5 × 40 mm in size were obtained. The middle part of AgSPE separating the working and contact zones was isolated with a Cementit-acetone mixture in a ratio of 1:5 by v/v, so that the electrode working area was 20 mm 2 (5×4 mm). AgSPEs were modified by forming a precipitate on its surface by electrolysis of a solution containing [Fe(CN) 6 ] 3-/4under various conditions (see Section 3.2.).

Potentiometric Measurements
For potentiometric studies, the measurement circuit proposed in this work and shown in Figure 1, was applied. The advantage of such a circuit over the traditional two-electrode configuration is the implementation of identical conditions for comparing the electrodes and the reduction of time, since it provides the possibility to obtain the results of two measurements in one experiment.
Electrode stability in the auxiliary solution was evaluated by potentiometric method using measuring circuit shown in Figure 1a. Following parameters were determined:
E 1 -steady-state value of potential, mV;  6 ] in the cell is 9.1 mM, which models the upper value of antioxidant content in the sample). 5. E 3 -potential value, mV, was established when electrode is returned to the initial solution.
These parameters were evaluated based on the study of four modified AgSPEs (Table 2) and four Ag/AgCl QREs ( Table 4). The most stable modified AgSPE was considered as a QRE mix .
Solution's (fruit juices and biofluids) AOA were evaluated at room temperature by HPM [1][2][3][4][5]. The measuring circuit, shown in Figure 1b, was used. In the circuit, PtSPE vs. EVL-1M3.1 served as an electrode pair, and PtSPE vs. QRE mix served as a sensor system. Electrode stability in the auxiliary solution was evaluated by potentiometric method using measuring circuit shown in Figure 1a. Following parameters were determined: 1. τ -potential stabilization (establishment) time, s; 2. E1 -steady-state value of potential, mV; 3. ΔE/Δt -drift of the potential, mV/h; 4. E2 -potential value, mV, was obtained when 0.4 mL 0.1 M K4[Fe(CN)6] was added to 4 mL of auxiliary solution (the final concentration of K4[Fe(CN)6] in the cell is 9.1 mM, which models the upper value of antioxidant content in the sample). 5. E3 -potential value, mV, was established when electrode is returned to the initial solution.
These parameters were evaluated based on the study of four modified AgSPEs (Table 2) and four Ag/AgCl QREs ( Table 4). The most stable modified AgSPE was considered as a QREmix.

Voltammetric and Scanning Electron Microscopy Measurements
Normal three-electron cell was used, consisting of AgSPE or Pt, EVL-1M3.1 or QREmix and GC-2000 as working electrode, reference electrode, and auxiliary electrode. Potential scanning rate was 0.05 V/s. Scanning electron microscopy (SEM) measurements were performed at 20 kV in vacuum. Data were used for electrodes surface morphology characterization (see Section 3.4).

Potentiometric Sensor System Assembly
Platinum screen-printed electrode (PtSPE) was used as an indicator in the evaluation of AOA of fruit juices and biofluids. Applicability of PtSPE in food analysis [4] and biofluids [1,2,5] has been confirmed earlier. After operation in food and biological matrices, PtSPE was regenerated by annealing at a temperature of 750 °С for 1 hour [58]. The most stable modified AgSPE (i.e., QREmix) served as a reference electrode. The results obtained are presented in Section 3.6.

Sampling and Sample Preparation
Samples of apple juices of the Dobryj, Rich, J7 brands and fresh apples of Smit and Fuji varieties were purchased at a local supermarket. The juices taken from the packages opened before analysis and freshly squeezed juices were examined. Saliva was taken into a plastic container 10 minutes before analysis, while the respondent refrained from eating, drinking, smoking and brushing his teeth for at least an hour. Blood was collected by venipuncture at the bend of the elbow joint into a polyethylene terephthalate tube containing a blood coagulation activator (SiO2). To obtain serum,

Voltammetric and Scanning Electron Microscopy Measurements
Normal three-electron cell was used, consisting of AgSPE or Pt, EVL-1M3.1 or QRE mix and GC-2000 as working electrode, reference electrode, and auxiliary electrode. Potential scanning rate was 0.05 V/s. Scanning electron microscopy (SEM) measurements were performed at 20 kV in vacuum. Data were used for electrodes surface morphology characterization (see Section 3.4).

Potentiometric Sensor System Assembly
Platinum screen-printed electrode (PtSPE) was used as an indicator in the evaluation of AOA of fruit juices and biofluids. Applicability of PtSPE in food analysis [4] and biofluids [1,2,5] has been confirmed earlier. After operation in food and biological matrices, PtSPE was regenerated by annealing at a temperature of 750 • C for 1 h [58]. The most stable modified AgSPE (i.e., QRE mix ) served as a reference electrode. The results obtained are presented in Section 3.6.

Sampling and Sample Preparation
Samples of apple juices of the Dobryj, Rich, J7 brands and fresh apples of Smit and Fuji varieties were purchased at a local supermarket. The juices taken from the packages opened before analysis and freshly squeezed juices were examined. Saliva was taken into a plastic container 10 min before analysis, while the respondent refrained from eating, drinking, smoking and brushing his teeth for at least an hour. Blood was collected by venipuncture at the bend of the elbow joint into a polyethylene terephthalate tube containing a blood coagulation activator (SiO 2 ). To obtain serum, whole blood samples were centrifuged at 3500 rpm for 15 min. The resulting blood serum was frozen and stored at −18 • C. The ejaculate was taken into a plastic container by natural masturbation after 2-3 days of abstinence. The selected ejaculate samples were kept for 40 min at room temperature, and then they were frozen and stored at −18 • C. Before analysis, serum and ejaculate samples were thawed for 40 min at room temperature.

Statistical Analysis
All measurements were repeated 4 times. Statistical analysis was performed in Microsoft Excel 2010 with an accepted significance level of α = 0.05. The data are presented as X ± ∆X, where X is the average value, ∆X is the standard deviation. Validation of the results of the evaluation of AOA solutions obtained on the developed reference electrode (QRE mix ) was performed in relation to the results obtained on a commercial reference electrode (EVL-1M3.1), based on F-and t-tests. before analysis, while the respondent refrained from eating, drinking, smoking and brushing his teeth for at least an hour. Blood was collected by venipuncture at the bend of the elbow joint into a polyethylene terephthalate tube containing a blood coagulation activator (SiO2). To obtain serum, whole blood samples were centrifuged at 3500 rpm for 15 minutes. The resulting blood serum was frozen and stored at −18°C. The ejaculate was taken into a plastic container by natural masturbation after 2-3 days of abstinence. The selected ejaculate samples were kept for 40 minutes at room temperature, and then they were frozen and stored at -18°С. Before analysis, serum and ejaculate samples were thawed for 40 minutes at room temperature.

Statistical Analysis
All measurements were repeated 4 times. Statistical analysis was performed in Microsoft Excel 2010 with an accepted significance level of α = 0.05. The data are presented as X ± ΔX, where X is the average value, ΔX is the standard deviation. Validation of the results of the evaluation of AOA solutions obtained on the developed reference electrode (QREmix) was performed in relation to the results obtained on a commercial reference electrode (EVL-1M3.1), based on F-and t-tests.  Taking into account the low solubility of silver salts (Table 1), it should be concluded that the currents in cyclic voltammograms (Figure 2) are due to the following processes:  In a mixture of components, redox processes considered can take place in parallel or as competing ones. Composition of the precipitate formed on AgSPE depends on concentration of anions in the solution, solubility product of silver compounds, and electrode potential. In the potential range 0.05 V < E < 0. 15 6 ] and Ag 3 PO 4 is possible. Since, predominantly, those poorly soluble compounds are formed, in equilibrium with which concentration of silver ions is minimal ( Table 1), formation of Ag 3 PO 4 under these conditions can be neglected. Table 2 presents modification conditions of AgSPE. AgSPE polarization modes are selected according to the cyclic curves shown in Figure 2.   (Figure 3b). Minimum τ values are observed after modifying electrodes for 120 s. Results in which standard deviation of one or more parameters E 1 , E 2 , and E 3 exceeds 3 mV are excluded from further consideration, since they do not satisfy necessary conditions. Results in which standard deviation of the parameters E 1 , E 2 and E 3 is less than 3 mV are shown in Table 3.   It follows from Table 3 that optimal conditions are: potentiostatic polarization mode at 0.325 V for 120 s. Further, the electrode formed under these conditions was considered as a new QRE and defined as QREmix, which indicated presence of a mixed precipitate on its surface. In this case, precipitate consisted mainly of AgCl and Ag3[Fe(CN)6] (see Section 3.1). Table 4. The results of a comparative study of Ag/AgCl QRE and QREmix in solution containing [Fe(CN)6] 3-/4-stability are shown in Figure 4. It can be seen from Figure 4 that QREmix is the most stable. In this case, parameter's (τ, ΔE/Δt, E1, E2, E3) values that determine operability of the electrode are minimal. QREmix was stored in air at room temperature without direct sunlight. Parameters preserve normal values for application as a reference electrode for 7 days (Figure 4).   It follows from Table 3 that optimal conditions are: potentiostatic polarization mode at 0.325 V for 120 s. Further, the electrode formed under these conditions was considered as a new QRE and defined as QRE mix , which indicated presence of a mixed precipitate on its surface. In this case, precipitate consisted mainly of AgCl and Ag 3 [Fe(CN) 6 ] (see Section 3.1). Table 4. The results of a comparative study of Ag/AgCl QRE and QRE mix in solution containing [Fe(CN) 6 ] 3-/4stability are shown in Figure 4. It can be seen from Figure 4 that QRE mix is the most stable. In this case, parameter's (τ, ∆E/∆t, E 1 , E 2 , E 3 ) values that determine operability of the electrode are minimal. QRE mix was stored in air at room temperature without direct sunlight. Parameters preserve normal values for application as a reference electrode for 7 days (Figure 4).    (Figure 5d,f). Probably a fine-crystalline precipitate leave out, mainly of Ag3[Fe(CN)6], leave out as shown earlier (see Section 3.1), fills pores between large AgCl crystals, which leads to an improvement of QREmix stability compared to Ag/AgCl QREs.

QREmix Characterization by Cyclic Voltammetry
Cyclic voltammograms recorded on Pt in a solution containing 1 mM [Fe(CN)6]3-/4-(1:1) and 1 M KCl using EVL-1M3.1 оr QREmix as a reference electrode are shown in Figure 6. It is seen that the obtained cyclic voltammograms are almost identical and differ only in a parallel shift along the potential axis by 45 ± 3 mV (n = 4), which corresponds to the potential difference between EVL-1M3.1 and QREmix. This is one more evidence of QREmix stable operation in a solution containing [Fe(CN)6]3-/4-. It may contribute to the use of this electrode for other electrochemical applications.

Potentiometric Sensor System in the Determination of AOA of Solutions
The QREmix proposed was used as part of the sensor system to evaluate AOA of fruit juices and biofluids by HPM in comparison with commercial Ag/AgCl RE. Measurement circuit shown in Figure 1b (IE = PtSPE, RE 1 = EVL-1M3.1 and RE 2 = QREmix) was used. The results are presented in Table 5. It can be seen from Table 5 that values of F-and t-tests are less than theoretical ones, which proves the same reproducibility and statistically insignificant differences between the results. The data presented indicate correctness of using of QREmix in the real samples' analysis.

QREmix Characterization by Cyclic Voltammetry
Cyclic voltammograms recorded on Pt in a solution containing 1 mM [Fe(CN)6] 3-/4-(1:1) and 1 M KCl using EVL-1M3.1 or QREmix as a reference electrode are shown in Figure 6. It is seen that the obtained cyclic voltammograms are almost identical and differ only in a parallel shift along the potential axis by 45 ± 3 mV (n = 4), which corresponds to the potential difference between EVL-1M3.1 and QREmix. This is one more evidence of QREmix stable operation in a solution containing [Fe(CN)6] 3-/4-. It may contribute to the use of this electrode for other electrochemical applications.

Potentiometric Sensor System in the Determination of AOA of Solutions
The QREmix proposed was used as part of the sensor system to evaluate AOA of fruit juices and biofluids by HPM in comparison with commercial Ag/AgCl RE. Measurement circuit shown in Figure 1b (IE = PtSPE, RE 1 = EVL-1M3.1 and RE 2 = QREmix) was used. The results are presented in Table 5. It can be seen from Table 5 that values of F-and t-tests are less than theoretical ones, which

Potentiometric Sensor System in the Determination of AOA of Solutions
The QRE mix proposed was used as part of the sensor system to evaluate AOA of fruit juices and biofluids by HPM in comparison with commercial Ag/AgCl RE. Measurement circuit shown in  Table 5. It can be seen from Table 5 that values of F-and t-tests are less than theoretical ones, which proves the same reproducibility and statistically insignificant differences between the results. The data presented indicate correctness of using of QRE mix in the real samples' analysis.

Conclusions
The results of a study of the processes occurring at the electrode/multianionic interface, some of which form sparingly soluble or complex compounds with the electrode material, are of general interest. The data obtained in this work make a definite contribution to this area: they justify research paths and provide the information needed for creating QREs for sensor systems operating in the presence of interfering ions, [Fe(CN) 6 ] 3-/4in particular. The latter allowed us to solve a very important problem-to develop sensory systems for monitoring AOA/OA of various objects, including vital biological ones. The QRE proposed consists of an Ag screen-printed substrate electrochemically coated with a mixed precipitate containing silver chloride/ferricyanide. The sediment was studied by potentiometry, scanning electron microscopy and cyclic voltammetry. A new electrochemical scheme is proposed and used, which allows comparing electrodes in one measurement. This approach increases the accuracy of the research results and reduces time required to obtain them. Performance of the sensor system with the new QRE is illustrated by its application for determination of AOA of fruit juices and body liquids. The results obtained have good analytical characteristics, which allows us to predict the widespread use of QRE described in this paper. Additional strategies aimed at improving the stability of the developed QRE may include, but are not limited to, the use of protective coatings. The approaches and methods described may be useful for creating QREs working in the presence of other interfering substances.