Potentiometric Sensors for Iodide and Bromide Based on Pt(II)-Porphyrin

Pt(II) 5,10,15,20-tetra(4-methoxy-phenyl)-porphyrin (PtTMeOPP) was used in the construction of new ion-selective sensors. The potentiometric response characteristics (slope and selectivity) of iodide and bromide-selective electrodes based on (PtTMeOPP) metalloporphyrin in o-nitrophenyloctylether (NPOE), dioctylphtalate (DOP) and dioctylsebacate (DOS) plasticized poly(vinyl chloride) membranes are compared. The best results were obtained for the membranes plasticized with DOP and NPOE. The sensors have linear responses with near-Nernstian slopes toward bromide and iodide ions and good selectivity. The membrane plasticized with NPOE was electrochemically characterized using the EIS method to determine its water absorption and the diffusion coefficient into the membrane.


Introduction
Porphyrins and their derivatives have versatile physicochemical properties and structural features affording functionalization that highly recommend them as sensitive compounds for sensors design [1]. Porphyrins can recognize both positive and negative ions and also neutral molecules [2] being used for testing of biologically active substances, environmental pollutants, heavy metal ions [3] and explosives [4,5]. As a result of interaction with an analyte, the porphyrin modifies its structure and performs a change in colour, optical, fluorescent or electrical properties [6].
Potentiometric ion selective membrane electrodes [7] based on porphyrins [8] offer many advantages amongst which are: good selectivity and sensitivity, user-friendliness, lack of toxicity. Free base porphyrins perform as neutral carriers when incorporated into polymeric membranes and can detect the metallic cations by generating sitting-a-top complexes [9].
Metalloporphyrins act by selective axial coordination and can recognize and quantify several types of anions [10] and the anion selectivity can be enhanced by changing either the central metal ion or the peripheral substitution of the macrocycle [11].
Reports focusing on metalloporphyrins with bivalent central metal ions as ionophores for the construction of anion-selective membrane electrodes have rarely appeared in the literature and are attracting more and more attention [12].
Amongst the metalloporphyrins, platinum derivatives are receiving interest only in the last years, being successfully used for oxygen detection [13,14], perchlorate ions determination [15] and triiodide/iodide is a key redox couple used in solar-energy-harvesting systems [31] so its quantity has to be controlled. Different methods were used to determine levels of bromide, respective iodine ions in milk and other food products, drinking water and urine [32,33] Thus, kinetic colorimetric ceric-arsenic assay [34] gas chromatography (GC) and ion chromatography (IC) [35], high performance liquid chromatography (HPLC) [36], mass spectrometry (MS) [37], inductively coupled plasma mass spectrometry (ICP-MS) [38] were used. The GC method showed higher accuracy compared to the IC procedure and provided a detection limit of 0.4 mg/L when evaluating the levels of bromide ion in urine samples of workers exposed to methyl bromide [35].
In the last decades, due to the vital role of selective iodide and bromide determination in various areas, non-expensive investigations using selective membrane electrodes based on different ion carriers have been done. The most extensively studied systems for sensing carriers for iodide detection are in the range of concentrations from 1 × 10 −1 to 1 × 10 −6 M are metallophthalocyanines [39] that can be used over a wide pH range of 3.0 ± 8.0 [40]. Approximately the same performances with metallophthalocyanines are provided by Mn(III)-metaloporphyrins [41], Co(II) and Ni(II) complexes of cyclam derivative [42], Cu (II) salicylidene complex [43] and Schiff base complexes of Cd(II) metal ions [44]. Although not improving the detection limits, another iodine sensor system focused on free base tetraphenylporphyrin incorporated into a glass-like silicone ladder-type polymer is remarkable because it provides constant selectivity coefficients for half a year [45].
A detection limit of 7.4 × 10 −6 M for iodide ions and a fast response time of 15 s was obtained when Zn(II)-5,10,15,20-tetrakis(-4-pyridyl)porphyrin was used as ionophore in a liquid/polymeric membrane electrode having 20 mol % of TDMACl as additive in the membrane [46].
Porphyrins exhibit rich coordination properties having the capacity to distort their planar conformation through structural modification so that the internal NH groups become properly disposed for selective anion binding [47]. The group of Mamardashvili reported that the diprotonated form of 2,8,12,18-tetramethyl-3,7,13,17-tetrabutyl-porphyrin functioned as a bromide selective receptor, forming stable complexes with bromide in acetonitrile at 25 • C. Using the UV-vis spectrometric method, the limit of detection for bromide ions was of 3 × 10 −8 M [48].
The high affinity of halide ions in complexes of 1:1 and 2:1 ratios formed with diprotonated porphyrins was also reported [49] and the binding is taking place between I − ions and N + H groups from diprotonated porphyrins. During the halide ions binding, the porphyrin is maintaining its monomeric structure.
Because the selective iodide and/or bromide determinations are challenges in medical, biological, environmental and food monitoring, the purpose of this research is to develop new non-toxic Pt(II) porphyrins as stable ionophores providing a wide linear response to these anions.
As can be seen from Figure 1, the Soret band of the Pt-metalloporphyrin is significantly hypsochromically shifted and has hyperchromic effect in comparison with porphyrin-base. The same phenomena are to be seen regarding the Q bands. The method of synthesis is described, as follows: to 0.1785 g (2.429 × 10 −4 mole) 5,10,15,20-tetra-(p-methoxy-phenyl)porphyrin (TMeOPP) dissolved in chlorobenzene (30 mL) was added a solution comprising 0.165 g (1.245 × 10 −3 mole) CH3COONa × 3H2O [50] and 0.172 g (3.644 × 10 −4 mole) PtCl2(PhCN)2 dissolved in 20 mL chlorobenzene. The reaction mixture was heated to reflux for one hour under atmospheric conditions. The metallation was monitored by UV-vis spectroscopy, until the four Q bands of the porphyrin base were reduced to only two. The reaction mixture was then cooled to room temperature and filtered. The precipitate was repeatedly washed with hot water, then collected and dried in vacuum drying oven at 90 °C for 30 h. The recrystallization was performed using CH2Cl2. Brownish-orange crystals of Pt-TMeOPP were obtained.
As can be seen from Figure 1, the Soret band of the Pt-metalloporphyrin is significantly hypsochromically shifted and has hyperchromic effect in comparison with porphyrin-base. The same phenomena are to be seen regarding the Q bands.   From the shape and position of the signals in 1 H-NMR it can be stated that the Pt-porphyrin is distorted, most probably in saddle type conformation and has no longer a planar structure.

Sample Preparations
The performance of each sensor was investigated by measuring its potential in the concentration range 10 −6 -10 −1 M of different anionic solutions. Stock solutions of 0.1 M were prepared by dissolving sodium or potassium salts in 4-morpholinoethanesulfonic acid (MES) of pH = 5.5. All working solutions were prepared by gradual dilution of the stock solutions.

Electrode Membrane Preparation and Measurements
The membranes have the composition 1% ionophore, 33% PVC and 66% plasticizer. Tridodeocylmethylammonium chloride (TDMACl) was used as additive (20 mol % relative to ionophore). The electroactive material and the solvent mediator were mixed together and then the PVC and the appropriate amount of THF were added. The ingredients were intensively stirred for about 20 min until all of them were dissolved, obtaining a transparent solution. This solution was transferred onto a glass plate of 20 cm 2 and the THF was allowed to evaporate at room temperature leaving a tough, flexible membrane embedded in a PVC matrix. An 8 mm diameter piece of membrane was cut out and assembled on the Fluka electrode body.

Apparatus and Electrodes
The measurements were carried out at room temperature using a Hanna Instruments HI223 pH/mV-meter by setting up the following cell: Electrochemical Impedance Spectroscopy (EIS) investigations were performed with the help of Autolab 302N EcoChemie equipped with the FRA2 impedance module. The sinusoidal potential amplitude was 10 mV and the tested frequency range was from 0.1 Hz to 100 kHz. All electrochemical measurements were performed at room temperature in a conventional one-compartment three-electrode cell, equipped with two stainless steel counter electrodes and Ag/AgCl as reference electrode. The working electrode is represented by the membrane fixed to a Fe electrode (active surface equal to 0.785 cm 2 ). In order to determine the absorption of water, the membrane was immersed in a buffer of 2-(N-morpholino)ethanesulfonic acid (MES) solution. During immersion EIS spectra are recorded at OCP potential.

Response Characteristics of the Electrodes
In the case of platinum porphyrins, the neutral response is expected due to the 2+ oxidation state of the central ion. This is the reason why membranes with cationic additive were prepared and their responses to a number of eight anions were evaluated. The influence of three different plasticizers on the potentiometric response was also tested. The composition of the prepared membranes is presented in Table 1. Each membrane contains 20 mol % TDMACl relative to the ionophore.         Analysing Figures 2-4, it can be seen that the plasticizer plays an important role in the formulation of membranes. In the case of sensor A, plasticized with DOS, there is no significant potentiometric answer to any of the tested anions. It was observed that in the case of membranes containing plasticizers having a low dielectric constant such as DOS (ε = 4), the aggregation of hydrophobic porphyrins occurs and the free ability of ionophores and their complexes to move is thus constrained. In these DOS plasticized membranes the ionophores capacity to form anion-ligand complexes is diminished, conducting to randomly potentiometric response for all the tested anions ( Figure 2). As a consequence, DOS plasticizer is not recommended in formulations of membranes based on this platinum (II) porphyrin destined to anions detection.
The sensor B, plasticized with DOP, shows a near-Nernstian potentiometric answer to bromide in the range of 10 −1 to 10 −5 M with a slope of (64.4 ± 0.4) mV/decade, that is presented in Figure 5. The sensor was intensively used for 4 weeks and after this period the slope was measured again and it is presented in Figure 6. There was a little decrease of the slope in time, (61.6 ± 0.3) mV/decade but it still remains in the analytical useful range. Analysing Figures 2-4, it can be seen that the plasticizer plays an important role in the formulation of membranes. In the case of sensor A, plasticized with DOS, there is no significant potentiometric answer to any of the tested anions. It was observed that in the case of membranes containing plasticizers having a low dielectric constant such as DOS (ε = 4), the aggregation of hydrophobic porphyrins occurs and the free ability of ionophores and their complexes to move is thus constrained. In these DOS plasticized membranes the ionophores capacity to form anion-ligand complexes is diminished, conducting to randomly potentiometric response for all the tested anions ( Figure 2). As a consequence, DOS plasticizer is not recommended in formulations of membranes based on this platinum (II) porphyrin destined to anions detection.
The sensor B, plasticized with DOP, shows a near-Nernstian potentiometric answer to bromide in the range of 10 −1 to 10 −5 M with a slope of (64.4 ± 0.4) mV/decade, that is presented in Figure 5.  Analysing Figures 2-4, it can be seen that the plasticizer plays an important role in the formulation of membranes. In the case of sensor A, plasticized with DOS, there is no significant potentiometric answer to any of the tested anions. It was observed that in the case of membranes containing plasticizers having a low dielectric constant such as DOS (ε = 4), the aggregation of hydrophobic porphyrins occurs and the free ability of ionophores and their complexes to move is thus constrained. In these DOS plasticized membranes the ionophores capacity to form anion-ligand complexes is diminished, conducting to randomly potentiometric response for all the tested anions ( Figure 2). As a consequence, DOS plasticizer is not recommended in formulations of membranes based on this platinum (II) porphyrin destined to anions detection.
The sensor B, plasticized with DOP, shows a near-Nernstian potentiometric answer to bromide in the range of 10 −1 to 10 −5 M with a slope of (64.4 ± 0.4) mV/decade, that is presented in Figure 5. The sensor was intensively used for 4 weeks and after this period the slope was measured again and it is presented in Figure 6. There was a little decrease of the slope in time, (61.6 ± 0.3) mV/decade but it still remains in the analytical useful range. The sensor was intensively used for 4 weeks and after this period the slope was measured again and it is presented in Figure 6. There was a little decrease of the slope in time, (61.6 ± 0.3) mV/decade but it still remains in the analytical useful range. In the case of sensor C, plasticized with NPOE, a potentiometric answer to iodide was obtained in the range 10 −1 to 10 −5 M with a near-Nernstian slope of (52.3 ± 0.2) mV/iodide decade and it is presented in Figure 7. The sensor was also used for a period of four weeks with no significant modification of the slope value.
The nature of the plasticizer influences the dielectric constant of the membrane (o-nitrophenyloctylether (NPOE, ε = 24), dioctyl phtalate (DOP, ε = 7) dioctyl sebacate (DOS, ε = 4)) and the mobility of the ionophore and its complex. The porphyrins aggregation in membrane is most likely induced by the nature of plasticizer (low polar DOP and high polar NPOE plasticizer). DOP-a plasticizer with lower polarity-seems to favour the aggregation of porphyrins macrocycles and this behaviour explains the value of the Nernstian slope (the deviations from the Nernstian slope) and the preferences for bromide, an ion with a lower radius than iodide. In membranes with a polar plasticizer such as NPOE, the aggregation of hydrophobic porphyrins is hindered and the mobility of ionophores and of their complexes is increased, so that they have the capacity to form larger ion-ligand complexes, binding in this way anions with larger radius. That is the reason why this membrane gives a Nernstian response for iodide [26,52]. In the case of sensor C, plasticized with NPOE, a potentiometric answer to iodide was obtained in the range 10 −1 to 10 −5 M with a near-Nernstian slope of (52.3 ± 0.2) mV/iodide decade and it is presented in Figure 7. In the case of sensor C, plasticized with NPOE, a potentiometric answer to iodide was obtained in the range 10 −1 to 10 −5 M with a near-Nernstian slope of (52.3 ± 0.2) mV/iodide decade and it is presented in Figure 7. The sensor was also used for a period of four weeks with no significant modification of the slope value.
The nature of the plasticizer influences the dielectric constant of the membrane (o-nitrophenyloctylether (NPOE, ε = 24), dioctyl phtalate (DOP, ε = 7) dioctyl sebacate (DOS, ε = 4)) and the mobility of the ionophore and its complex. The porphyrins aggregation in membrane is most likely induced by the nature of plasticizer (low polar DOP and high polar NPOE plasticizer). DOP-a plasticizer with lower polarity-seems to favour the aggregation of porphyrins macrocycles and this behaviour explains the value of the Nernstian slope (the deviations from the Nernstian slope) and the preferences for bromide, an ion with a lower radius than iodide. In membranes with a polar plasticizer such as NPOE, the aggregation of hydrophobic porphyrins is hindered and the mobility of ionophores and of their complexes is increased, so that they have the capacity to form larger ion-ligand complexes, binding in this way anions with larger radius. That is the reason why this membrane gives a Nernstian response for iodide [26,52]. The sensor was also used for a period of four weeks with no significant modification of the slope value.
The nature of the plasticizer influences the dielectric constant of the membrane (o-nitrophenyloctylether (NPOE, ε = 24), dioctyl phtalate (DOP, ε = 7) dioctyl sebacate (DOS, ε = 4)) and the mobility of the ionophore and its complex. The porphyrins aggregation in membrane is most likely induced by the nature of plasticizer (low polar DOP and high polar NPOE plasticizer). DOP-a plasticizer with lower polarity-seems to favour the aggregation of porphyrins macrocycles and this behaviour explains the value of the Nernstian slope (the deviations from the Nernstian slope) and the preferences for bromide, an ion with a lower radius than iodide. In membranes with a polar plasticizer such as NPOE, the aggregation of hydrophobic porphyrins is hindered and the mobility of ionophores and of their complexes is increased, so that they have the capacity to form larger ion-ligand complexes, binding in this way anions with larger radius. That is the reason why this membrane gives a Nernstian response for iodide [26,52].

Potentiometric Selectivity
One of the most important characteristics of a sensor is the selectivity coefficient, showing the potentiometric answer of the sensor to the interfering anions toward the primary one. The selectivity coefficients of the sensors B and C, calculated using the separate solution method is presented in Table 2. Analysing the obtained results, it can be concluded that both sensors present good selectivity compared to the interfering tested anions.

Effect of the pH and the Response Time of the Sensors
The pH functions of the two sensors were obtained by using NaOH and HCl 2 M solutions. These were added in drops to the 10 −2 M solution of each primary anion and the behaviour is presented in Figures 8 and 9.

Potentiometric Selectivity
One of the most important characteristics of a sensor is the selectivity coefficient, showing the potentiometric answer of the sensor to the interfering anions toward the primary one. The selectivity coefficients of the sensors B and C, calculated using the separate solution method is presented in Table 2. Analysing the obtained results, it can be concluded that both sensors present good selectivity compared to the interfering tested anions.

Effect of the pH and the Response Time of the Sensors
The pH functions of the two sensors were obtained by using NaOH and HCl 2 M solutions. These were added in drops to the 10 −2 M solution of each primary anion and the behaviour is presented in Figures 8 and 9.  As displayed in Figure 8, the pH range of the bromide-selective sensor is from 6-12. Figure 9 shows that the iodide-selective sensor works properly in a pH range from 3 to 12.

Potentiometric Selectivity
One of the most important characteristics of a sensor is the selectivity coefficient, showing the potentiometric answer of the sensor to the interfering anions toward the primary one. The selectivity coefficients of the sensors B and C, calculated using the separate solution method is presented in Table 2. Analysing the obtained results, it can be concluded that both sensors present good selectivity compared to the interfering tested anions.

Effect of the pH and the Response Time of the Sensors
The pH functions of the two sensors were obtained by using NaOH and HCl 2 M solutions. These were added in drops to the 10 −2 M solution of each primary anion and the behaviour is presented in Figures 8 and 9.  As displayed in Figure 8, the pH range of the bromide-selective sensor is from 6-12. Figure 9 shows that the iodide-selective sensor works properly in a pH range from 3 to 12. As displayed in Figure 8, the pH range of the bromide-selective sensor is from 6-12. Figure 9 shows that the iodide-selective sensor works properly in a pH range from 3 to 12.
The average time of both sensors to reach the final potential value after successive immersions in iodide and bromide solutions, each having a 10-fold difference in concentration from low to high and reverse was measured. The results are presented in Figures 10 and 11. The average times for both sensors to reach 95% of the final potential value [53] were obtained for solutions from 10 −3 to 10 −2 M and were of 80 s for the bromide-selective sensor, respectively 60 s for the iodide one. The average time of both sensors to reach the final potential value after successive immersions in iodide and bromide solutions, each having a 10-fold difference in concentration from low to high and reverse was measured. The results are presented in Figures 10 and 11. The average times for both sensors to reach 95% of the final potential value [53] were obtained for solutions from 10 −3 to 10 −2 M and were of 80 s for the bromide-selective sensor, respectively 60 s for the iodide one.  The detection limit of the electrodes was established at the point of intersection of the extrapolated linear mid-range and final low concentration level segments of the calibration plot and it is 9 × 10 −6 M for the iodide-selective sensor and 8 × 10 −6 M for the bromide-selective sensor. This range has relevance for the medical detection [28,29].

EIS Characterization of NPOE Membrane
In order to determine the absorption of water, the NPOE membrane was maintained in a MES buffer solution and EIS spectra were recorded during immersion at OCP, at 24 °C. The decrease of coating impedance during the immersion is expected because water and ions can penetrate into the pores of the membrane. When water penetrates the membrane, its capacitance increases as the dielectric constant increases. As a result, the capacitance can be used to measure the water absorbed by the membrane.
The Nyquist and Bode diagrams are presented in Figure 12a,b. The immersion time for each curve is presented in Table 3.  The average time of both sensors to reach the final potential value after successive immersions in iodide and bromide solutions, each having a 10-fold difference in concentration from low to high and reverse was measured. The results are presented in Figures 10 and 11. The average times for both sensors to reach 95% of the final potential value [53] were obtained for solutions from 10 −3 to 10 −2 M and were of 80 s for the bromide-selective sensor, respectively 60 s for the iodide one.  The detection limit of the electrodes was established at the point of intersection of the extrapolated linear mid-range and final low concentration level segments of the calibration plot and it is 9 × 10 −6 M for the iodide-selective sensor and 8 × 10 −6 M for the bromide-selective sensor. This range has relevance for the medical detection [28,29].

EIS Characterization of NPOE Membrane
In order to determine the absorption of water, the NPOE membrane was maintained in a MES buffer solution and EIS spectra were recorded during immersion at OCP, at 24 °C. The decrease of coating impedance during the immersion is expected because water and ions can penetrate into the pores of the membrane. When water penetrates the membrane, its capacitance increases as the dielectric constant increases. As a result, the capacitance can be used to measure the water absorbed by the membrane.
The Nyquist and Bode diagrams are presented in Figure 12a,b. The immersion time for each curve is presented in Table 3. The detection limit of the electrodes was established at the point of intersection of the extrapolated linear mid-range and final low concentration level segments of the calibration plot and it is 9 × 10 −6 M for the iodide-selective sensor and 8 × 10 −6 M for the bromide-selective sensor. This range has relevance for the medical detection [28,29].

EIS Characterization of NPOE Membrane
In order to determine the absorption of water, the NPOE membrane was maintained in a MES buffer solution and EIS spectra were recorded during immersion at OCP, at 24 • C. The decrease of coating impedance during the immersion is expected because water and ions can penetrate into the pores of the membrane. When water penetrates the membrane, its capacitance increases as the dielectric constant increases. As a result, the capacitance can be used to measure the water absorbed by the membrane.
The Nyquist and Bode diagrams are presented in Figure 12a,b. The immersion time for each curve is presented in Table 3.  Figure 12b shows the modulus and the phase angle for the membrane during immersion. The membrane exhibited a well-defined time constant at lower frequencies (the peak in the phase angle plot). Increasing the immersion time the peak seems to move to lower frequencies. Membrane also exhibits a second peak at higher frequencies but this peak extends over the range of measured frequencies (10 5 Hz). The experimental data were fitted to the equivalent electrical circuit by a complex non-linear least squares method, using the ZView-Scribner Associates Inc. (Southern Pines, NC, USA and Solartron Analytical, Oak Ridge, TN, USA) software. The equivalent electric circuit (EEC) used for the NPOE electrode is presented in Figure 13. The obtained results are presented in Table 3 and were used for water absorption investigation.  Figure 12b shows the modulus and the phase angle for the membrane during immersion. The membrane exhibited a well-defined time constant at lower frequencies (the peak in the phase angle plot). Increasing the immersion time the peak seems to move to lower frequencies. Membrane also exhibits a second peak at higher frequencies but this peak extends over the range of measured frequencies (10 5 Hz). The experimental data were fitted to the equivalent electrical circuit by a complex non-linear least squares method, using the ZView-Scribner Associates Inc. (Southern Pines, NC, USA and Solartron Analytical, Oak Ridge, TN, USA) software. The equivalent electric circuit (EEC) used for the NPOE electrode is presented in Figure 13. The obtained results are presented in Table 3 and were used for water absorption investigation.   Figure 13. Equivalent electric circuit.
The EEC includes the resistance of electrolyte Rs, in series with a parallel connexion of capacitance C1 (membrane capacitance) and the resistance R1 (membrane resistance, similar with ionic transfer in the membrane pores) and also in series with a parallel connexion of CPE2 (representing the charge/discharge process that occurs at the substrate/electrolyte interface) and the polarization resistance R2. Constant phase element (CPE) was introduced to represent the double layer capacitance, for the reason of non-ideal behaviour. The impedance of a CPE can be expressed by ZCPE = [T(jω) n ] −1 , where ω is frequency, T is the CPE magnitude and the exponent n is between 0 and 1. The obtained values of the electric circuit elements for the NPOE electrode are presented in Table 3.
The measurement of the water absorption or water permeability using EIS techniques is based on the determination of the changes of the coating capacitance. The absorbed water, W (volume fraction), was obtained from Equation (2) [51,54] using the membrane capacitance resulted from EIS modelling (Ct at different time, t, C0 for t = 0) and the dielectric constant of water εH2O = 80, typically at room temperature.
The calculated data indicate an increase of the water absorption during time for the tested membrane. The water absorption (W) increase is linear until 320 min, as can be seen in Figure 14. Saturation is considered to be reached after 1495 min (~25 h). The maximum volume fraction of the absorbed water determined was 0.290.  The EEC includes the resistance of electrolyte R s , in series with a parallel connexion of capacitance C 1 (membrane capacitance) and the resistance R 1 (membrane resistance, similar with ionic transfer in the membrane pores) and also in series with a parallel connexion of CPE 2 (representing the charge/discharge process that occurs at the substrate/electrolyte interface) and the polarization resistance R 2 . Constant phase element (CPE) was introduced to represent the double layer capacitance, for the reason of non-ideal behaviour. The impedance of a CPE can be expressed by Z CPE = [T(jω) n ] −1 , where ω is frequency, T is the CPE magnitude and the exponent n is between 0 and 1. The obtained values of the electric circuit elements for the NPOE electrode are presented in Table 3.
The measurement of the water absorption or water permeability using EIS techniques is based on the determination of the changes of the coating capacitance. The absorbed water, W (volume fraction), was obtained from Equation (2) [51,54] using the membrane capacitance resulted from EIS modelling (C t at different time, t, C 0 for t = 0) and the dielectric constant of water ε H2O = 80, typically at room temperature.
The calculated data indicate an increase of the water absorption during time for the tested membrane. The water absorption (W) increase is linear until 320 min, as can be seen in Figure 14. Saturation is considered to be reached after 1495 min (~25 h). The maximum volume fraction of the absorbed water determined was 0.290. The EEC includes the resistance of electrolyte Rs, in series with a parallel connexion of capacitance C1 (membrane capacitance) and the resistance R1 (membrane resistance, similar with ionic transfer in the membrane pores) and also in series with a parallel connexion of CPE2 (representing the charge/discharge process that occurs at the substrate/electrolyte interface) and the polarization resistance R2. Constant phase element (CPE) was introduced to represent the double layer capacitance, for the reason of non-ideal behaviour. The impedance of a CPE can be expressed by ZCPE = [T(jω) n ] −1 , where ω is frequency, T is the CPE magnitude and the exponent n is between 0 and 1. The obtained values of the electric circuit elements for the NPOE electrode are presented in Table 3.
The measurement of the water absorption or water permeability using EIS techniques is based on the determination of the changes of the coating capacitance. The absorbed water, W (volume fraction), was obtained from Equation (2) [51,54] using the membrane capacitance resulted from EIS modelling (Ct at different time, t, C0 for t = 0) and the dielectric constant of water εH2O = 80, typically at room temperature.
The calculated data indicate an increase of the water absorption during time for the tested membrane. The water absorption (W) increase is linear until 320 min, as can be seen in Figure 14. Saturation is considered to be reached after 1495 min (~25 h). The maximum volume fraction of the absorbed water determined was 0.290.  This dependence, presented in Figure 15, is caused by the penetration of water into the membrane. The presence of water into the membrane can determine the diffusion of the electrolytes from the buffer solution into the membrane, causing also the decrease of the membrane electric resistance.
The electrical resistance of the membrane is also influenced by the time of immersion in MES buffer solution, as the electrolyte solution penetrates through the membrane pores. It linearly decreases with the time in the first 200 min of immersion in the electrolyte solution. With the increase of immersion time the membrane resistance reaches a plateau, attributed to the saturation with water.
This dependence, presented in Figure 15, is caused by the penetration of water into the membrane. The presence of water into the membrane can determine the diffusion of the electrolytes from the buffer solution into the membrane, causing also the decrease of the membrane electric resistance. The diffusion coefficient, D, into the membrane was calculated according to Equation (3) [55] from the time dependence of the membrane capacitance (presented in Figure 16). Ct is the capacitance at the moment t, C0 at t = 0 and Cf at the final moment and L the membrane thickness. For the NPOE membrane having 180 μm, the diffusion coefficient is D = 5.13 × 10 −7 cm 2 s −1 .    The diffusion coefficient, D, into the membrane was calculated according to Equation (3) [55] from the time dependence of the membrane capacitance (presented in Figure 16). C t is the capacitance at the moment t, C 0 at t = 0 and C f at the final moment and L the membrane thickness. For the NPOE membrane having 180 µm, the diffusion coefficient is D = 5.13 × 10 −7 cm 2 s −1 .
The electrical resistance of the membrane is also influenced by the time of immersion in MES buffer solution, as the electrolyte solution penetrates through the membrane pores. It linearly decreases with the time in the first 200 min of immersion in the electrolyte solution. With the increase of immersion time the membrane resistance reaches a plateau, attributed to the saturation with water.
This dependence, presented in Figure 15, is caused by the penetration of water into the membrane. The presence of water into the membrane can determine the diffusion of the electrolytes from the buffer solution into the membrane, causing also the decrease of the membrane electric resistance. The diffusion coefficient, D, into the membrane was calculated according to Equation (3) [55] from the time dependence of the membrane capacitance (presented in Figure 16). Ct is the capacitance at the moment t, C0 at t = 0 and Cf at the final moment and L the membrane thickness. For the NPOE membrane having 180 μm, the diffusion coefficient is D = 5.13 × 10 −7 cm 2 s −1 .    The prepared membranes with DOP and DOS additives absorb water in a volumetric content almost identical with the NPOE membrane (the differences are no more than 0.035% comparative to NPOE).
The performed EIS measurements emphasized the necessity to introduce a membrane conditioning step, before the detection step and to establish a minimum immersion time of 25 h in MES solution.

Analytical Applications
Both of the sensors were used with good results for the detection of bromide and iodide from synthetic and pharmaceutical samples and the obtained values are presented in Table 4. Table 4. Analytical applications of sensors from synthetic samples. Additionally, the iodide-selective sensor was used for the determination of iodide in pharmaceutical potassium iodide tablets. The samples were prepared as previously described and the iodide assay was carried out by direct potentiometry. The results are presented in Table 5 (average of three measurements). The results are in good agreement with the declared label amount. Table 5. Determination of iodide in pharmaceutical formulations.

Conclusions
There is an increased interest for the determination of iodide and bromide in therapeutic monitoring. Therefore, the objective of this study was to develop new iodide and bromide ion-selective membrane sensors.
By EIS method the water absorption of the membranes was determined in MES buffer solution. The maximum volume fraction of the adsorbed water determined was 0.290, reached after 1495 min. The electrical resistance of the membrane decreased with the increasing of the water absorption. The performed measurements emphasized the necessity to introduce a membrane conditioning step, before the detection step and to establish a minimum immersion time of 25 h in MES solution.
A fast, sensitive and reliable potentiometric method for the determination of iodide and bromide, with relevance in the medical monitored concentration range, was developed and validated. New approaches are created by introducing the versatile PtTMeOPP as ionophore in ion-selective sensors formulations.