The Influence of Electrolyte Type on Kinetics of Redox Processes in the Polymer Films of Ni(II) Salen-Type Complexes

Electrodes modified with polymers derived from the complexes [Ni(salcn)], [Ni(salcn(Me))] and [Ni(salcn(Bu))] were obtained in order to study the kinetics of electrode processes occurring in polymer films, depending on the thickness of the films, the type of electrolyte and the solvent. FTIR and EQCM methods were used to determine the type of mass transported into polymer films during anode processes and the number of moles of ions and solvent. The rate of charge transport through films was determined by the cyclic voltammetry method, by the quantity cD1/2. It was shown that the charge transport was determined by the transport of anions. The kinetics were most efficient for poly[Ni(salcn(Bu))] modified electrodes, obtained from TBAPF6 and working in TBAClO4 and TBABF4. It was also shown that a solvent with a higher DN value and lower viscosity (MeCN) facilitated the transport of the charge through polymer films.

Our research focuses on electrodes modified with the (±)-trans-N,N -bis(salicylidene)-1,2-cyclohexanediaminenickel(II) complex ([Ni(salcn)]) and its tert-Bu ([Ni(salcn(Bu))]) and Me derivatives ([Ni(salcn(Me))]), substituted at the 3,3 -positions of the phenolate moieties (Scheme 1). To date, we have investigated the mechanism of complex oxidation in MeCN, CH 2 Cl 2 and DMSO [35], and the electropolymerization mechanism of the complexes and their electroactivity [36,37]. It was found that in DMSO, the central ion is oxidized in these complexes [35]. On the other hand, in non-coordinating solvents, in complexes containing tert-Bu substituents and in the complex without substituents, the ligand is oxidized to the phenoxyl radical, bis phenoxyl radical and phenoxonium cation [35,36,38]. However, in the complex with Me substituents, both the ligand and the central ion are oxidized [38]. As a result of these processes, and the subsequent polymerization reaction, phenyl-phenyl electroactive polymer films are obtained on the surface of the electrodes. They are conductive polymers in which charge transport takes place by π electron delocalization [36]. The present paper presents the effects of the type of supporting electrolyte and solvent on the kinetics of redox processes in poly[Ni(salcn)], poly[Ni(salcn(Me))] and poly[Ni(salcn(Bu))] films, derived from the above-mentioned complexes. FTIR ATR and EQCM methods were used for our investigations.

FTIR ATR of Polymer Films
FTIR ATR studies of electrodes modified with polymer films, neutral and oxidized, were carried out in order to determine the type of ions involved in charge transport during electrode processes. The analysis was performed on the basis of poly[Ni(salcn)] synthesized and investigated in TBAClO 4 . The most characteristic bands of the neutral film ( Figure 1a) were the bands at 2922 and 2881 cm −1 , corresponding to the valence vibrations of the -CH 2 and -CH 3 groups of the cyclohexane groups of the polymer and the cations of the supporting electrolyte present in the film structures. Another characteristic was the band at 1624(s) cm −1 corresponding to the valence vibrations of the imine groups, the bands at 1536(w), 1511(w) and 1475(m) cm −1 corresponding to the C-C phenyl ring in plane vibration, a band at 1322(m) cm −1 corresponding to υ C-O, a band at 1236(w) cm −1 derived from CH 2 out of plane bendings in CH 2 Cl 2 , and wide bands in the ranges 1220-1000(s) and 780-600(w) cm −1 with maxima at 1092(s) cm −1 and at 702(w) cm −1 , corresponding to the vibration of the Cl-O groups in ClO 4 − . The presence of bands corresponding to the C-C phenyl ring in plane vibration confirmed the presence of these groups in the structure of the investigated film. This was particularly important due to the lack of bands above 3000 cm −1 , which could be attributed to -CH stretching vibration in arenes. Most likely, these bands were superimposed on the CH 2 and -CH 3 bands of the cyclohexane groups of the polymer and the electrolyte cations. This explanation was confirmed by the location of the most characteristic and most intense band, corresponding to the valence vibrations of the imine groups. It occurred at slightly lower frequencies than those most frequently reported in the literature [39]. Due to the shift of the spectral bands, the band derived from CH 2 in CH 2 Cl 2 occurred at slightly lower frequencies than in the pure solvent [39]. In the oxidized film (Figure 1b), bands derived from υ CH 2 and CH 3 (2922(m) and 2881(m) cm −1 ), υ C=N (1624(s) cm −1 ) and δ CH 2 Cl 2 (1236(w) cm −1 ) did not change their intensity. The same intensity of the bands at 2922(m) and 2881(m) cm −1 indicated that the concentration of cations in the film after the anode process did not decrease, which means that they were not involved in charge transport to ensure electroneutrality. On the other hand, the unchanged intensity of the band originating from the valence vibrations of the imine groups was an effect resulting from the location of these groups outside the coupling area. Earlier studies have shown that delocalisation occurs through C-O and Ni (II) groups as a bridge connecting phenolate groups [36]. The remaining bands increased their intensity. The increase in the intensity of the bands from the C-C phenyl ring in plane vibration (1536(w), 1511(w) and 1475(m) cm −1 ) was related to the change in the surroundings of the aromatic ring, resulting from the charge delocalisation [40]. Such changes have been shown previously in the literature [24,41]. During the anode process, due to delocalisation, more quinoid connections were formed, as a result of which an increase in the intensity of the bands or a change in their position was observed. In the presented spectrum ( Figure 1b) the band was shifted from 1536(m) to 1566(s) cm −1 . On the other hand, the increase in the band intensity at 1322(m) cm −1 , corresponding to the C-O valence vibrations, resulted from the participation of these groups in the coupling. This delocalisation route has also been demonstrated in other salen-type complexes [24]. The bands corresponding to the vibrations of the Cl-O groups in ClO 4 − (1220-1000(s) and 780-600(w) cm −1 ) also increased significantly, indicating a higher concentration of these ions in the oxidized film than in the neutral film. This result showed that the electroneutrality condition during the anode process was achieved by transporting the anions into the polymer film. Analysis of the literature shows analogous charge transport mechanisms during the oxidation of polymer films of salen-type complexes [25,26,28]. However, a slightly reduced intensity of the band derived from CH 2 out of plane bendings in CH 2 Cl 2 (1236(w) cm −1 ) indicated a reduced solvent content in the oxidized polymer films compared to neutral films.
In the case of electrodes modified with films derived from the [Ni(salcn(Me))] and [Ni(salcn(Bu))] complexes, the spectra differed from the spectra of the poly[Ni(salcn)] modified electrodes in the intensity of the bands corresponding to the CH 2 vibration out of plane bendings in CH 2 Cl 2 . The results were exemplified by the FTIR ATR spectrum of a modified by poly[Ni(salcn(Me))] film electrode, synthesized and studied in TBABF 4 . The solvent band occurred at 1228(w) cm −1 ( Figure S1, ESI). Upon oxidation of the polymer film, this band increased, indicating diffusion of the solvent into the structures of the polymer film. Another effect of the solvent compared to poly[Ni(salcn)] films may be related to the presence of the substituents in monomers resulting in the polymerization towards only one active position of the phenolate moieties (5,5 ). After the oxidation of poly[Ni(salcn(Me))] and poly[Ni(salcn(Bu))] films, the band derived from anion vibrations, BF 4 − , was also increased, occurring within the range of 1150-950(m) cm −1 , with a maximum at 1044(m) cm −1 ( Figure S1, ESI). With the intensity of the bands corresponding to the vibrations of the -CH 2 and -CH 3 groups unchanged ( Figure S1, bands at 2935(m) and 2859(m) cm −1 , ESI) in the oxidized films, the increased band originating from BF 4 − indicated the sole contribution of the anion in realizing the electroneutrality condition during the anode process.

Voltammetric Curves
The voltammetric curves of the investigated polymers obtained, as a result of one electropolymerization cycle in the solutions of all basic electrolytes, were characterized by two oxidation peaks (Figures 2a, 3a, 4a, S2A-S7A, ESI). As previously shown, for poly[Ni(salcn)] and poly[Ni(salcn(Bu))], they derived from the oxidation of phenolate moieties to phenoxyl radicals and bis phenoxyl radicals [38]. In the case of poly[Ni(salcn(Me))], they also included the oxidation of Ni (II) to Ni (III) [38]. On the other hand, in poly[Ni(salcn(Bu))] films, at the lowest potentials there was an additional anode signal (Figures 4a and S6A, ESI), probably derived from a different geometric structure of the polymer chains, caused by the large steric effect of the substituents.   The different shape of the peaks of the studied polymers indicated that their electronic structure was sensitive to the surroundings in the phenolate moiety, which was also found in the case of other salen-type complexes [25,27]. The shape of the peaks and their potentials are also influenced by the type of supporting electrolyte. In TBABF 4 , a different nature of the curves was observed for each polymer than that recorded in the other two types of electrolytes ( Figures S3A, S5A and S7A, ESI). However, the FTIR ATR studies of the oxidized film, showing an increase in the intensity of the C-C phenyl ring in plane vibration bands in relation to the neutral film, confirmed that in the TBABF 4 solution, similarly to the others, oxidation of phenolate ligand groups took place.

Gravimetric Curves
Gravimetric curves were obtained on the basis of the Sauerbrey equation [42]. Literature studies show that salen-complex-type films are properly rigidly bound to the surface, which allowed the authors to apply a gravimetric interpretation based on the Sauerbrey equation [25,26]. Assuming that the investigated films are rigid or not sufficiently rigid, but viscoelasticity is not their dominant feature, the application of the Sauerbrey equation is justified [26,43].
The molar masses (M) of species exchanged with the polymer film during the oxidation processes were determined based on the slopes of the linear sections of the ∆m = f(Q) relationship, for the first and second step of the anode process, from the foot of a given anode peak to the value corresponding to its development. In the case of significant non-linearity in this part, the slopes were determined from a small area around the peak potential.
Dependencies ∆m = f(Q) for poly[Ni(salcn)], on the basis of the processes recorded at v = 0.05 V·s −1 (Figure 2b and Figures S2B and S3B, ESI) in the applied potential range, showed slight deviations from linearity in the end sections of curves and decreasing character in the initial sections of curves The decreasing nature of the curves ∆m = f(Q) in the initial sections of the curves may be related to at least three mechanisms of charge transport through the polymer film [44].
Analysis of these possibilities requires connecting them with the values of the obtained molar masses of species exchanged with the film during the oxidation process and with the results of FTIR ATR. The IR spectra showed an increase in the anion content of the oxidized film structures, which means that anions were involved in charge transport. The analysis of the literature also points to such a mechanism as dominant for salen-type-complex polymers [25,26,28]. However, the slopes of the linear parts of the ∆m = f(Q) relationships for the first step of the oxidation process for poly[Ni(salcn)], in solutions of all supporting electrolytes, did not correspond to the M values for pure anions resulting from Faraday's laws. The calculated M values (Table 1) were lower than expected for individual anions (145 g·mol −1 , 99.5 g·mol −1 and 87 g·mol −1 for PF 6 − , ClO 4 − and BF 4 − , respectively) and they were 0.5-0.7 of the M values. Such a result could indicate diffusion of the solvent from the polymer film during the first step of the oxidation process [44,45]. However, the curves ∆m = f(Q) obtained during the anode processes at scan rate v = 0.05 V·s −1 were not monotonic; in the initial sections of the curves the mass decreased (Figures 2b, S2B and S3B, ESI). The charge transport mechanism, based on the share of pure anion and solvent, should show a rectilinear relationship ∆m = f(Q) in the whole range [44,45]. Other possibilities to consider for the decreasing mass at the beginning of the curves shown in Figures 2b, S2B and S3B, ESI, are the mechanisms of "non-exchangeable" ions or ions association with the polymer matrix [44]. The first of these possibilities occurs because of the geometric aspects of the ions. Both this mechanism and the second can be excluded because the anions of the supporting electrolytes we used were not large and possessed poor coordinating properties, and the analysis of FTIR ATR spectra did not show a decreased share of cations in the oxidized films in relation to neutral species (Figure 1). In the case of an anion association, the relationship ∆m = f(Q) is linear in the whole range [44]. . The simultaneous decrease in these two values indicated that the film was not fully neutral and that, at the beginning of the anodic process carried out at the lowest potentials, a reduction takes place. Hence, the reduction in film mass was due to anions leaving the film. The confirmation of the hypothesis for the process carried out at v = 0.01 V·s −1 were the results obtained for the process carried out at v = 0.05 V·s −1 in the applied potential range preceded by a reduction lasting 40 s, at a potential of 0 V. The relationships ∆m = f(Q) (Figures 6, S10 and S11, ESI) in the initial sections of the curves did not show a decreasing character, and the molar masses of species participating in the oxidation process in step I, determined on the basis of the linear parts of these relationships, were higher, except for M in TBABF 4 . Taking into account the above considerations, the FTIR ATR spectra, and the fact that the obtained M values for the first step of the oxidation process occurring in poly[Ni(salcn)] were lower than those resulting from Faraday's laws for pure ions, it can be concluded that charge transport occurred through the penetration of anions into the polymer film and the diffusion of the solvent in the reverse direction. Such a mechanism is justified for poly[Ni(salcn)] films, most probably due to electropolymerization in the directions of both active positions of phenolate groups (ortho-and para-). The removal of the solvent accompanying the penetration of the anions was therefore intended to obtain the volume required for these anions.   The second step of the oxidation processes in poly[Ni(salcn)] was characterized by slopes of the linear parts of the curves ∆m = f(Q), indicating the dependence of the charge transport mechanism on the type of the primary electrolyte (Table 1, 0.05 V·s −1 after a reduction lasting 40 s). In the TBAClO 4 solution, the value of M exactly corresponded to that of the pure anion according to Faraday's laws. In the TBABF 4 solution, the M value was lower than expected for the BF 4 − ion. On the other hand, in the TBAPF 6 solution, the M value was higher than expected for the PF 6 − ion and indicated that the transport of the anion into the polymer film was accompanied by the transport of the solvent in the same direction [25,26].
Analyzing the molar masses for each supporting electrolyte, they were higher for the second step of polymer oxidation than for the first step ( Table 1) (Table 2) were higher than expected for pure anions (except for the M value obtained for the second step of the poly[Ni(salcn(Me))] process in TBABF 4 ). In light of the analysis carried out for poly[Ni(salcn)] films, this indicated a charge transport mechanism based on the penetration of anions into the polymer film, accompanied by solvent diffusion in the same direction. FTIR ATR studies showed an increased proportion of both anions and solvent in the oxidized films ( Figure S1, ESI). Molar mass values were comparable for both scan rates, which resulted in the independence of the charge transport mechanism from v. The molar masses for both poly[Ni(salcn(Me))] and poly[Ni(salcn(Bu))] were greater in the first step of the electrode process than in the second step ( Table 2), unlike that observed for poly[Ni(salcn)] (Table 1).    The dependence of M on the type of electrolyte was more visible in the case of poly[Ni(salcn(Me))], especially at the rate v = 0.01 V·s −1 ( Table 2). More moles of the solvent penetrated the poly[Ni(salcn(Bu))] films than poly[Ni(salcn(Me))] ones; in the case of poly[Ni(salcn(Bu))], it was the same in all electrolyte solutions, except for the second oxidation step in TBABF 4 . This may have been due to the different structural properties of the polymers resulting, inter alia, from the different steric effects of substituents of different sizes.
When comparing the M values in solutions of different electrolytes, the smallest number of moles of solvent was also transported in the TBABF 4 medium; in the second step even diffusion of the solvent from the polymer film was observed. The relationship of the number of moles of diffusing solvent on the type of electrolyte may be related to the size of anions. The BF 4 − ion was the smallest among the investigated anions (ion diameter BF 4 − = 0.334 nm [46]). The other ions, in the presence of which mass transport takes place to a greater extent, had larger sizes (ion diameters 0.370 and 0.435 nm for ClO 4 − and PF 6 − , respectively [46]).
The dependence of the charge transport on the type of solvent was demonstrated in the case of poly[Ni(salcn)].
In MeCN, the molar masses of the first step of the process were as expected for pure anions (Table 1 based on Figures 7b, S12 and S13B, ESI). However, in the second step of the process conducted in the MeCN, the solvent was transported along with the ions. For each type of supporting electrolyte, a much larger number of moles of solvent penetrated into the film structures than in the CH 2 Cl 2 process (Table 1). This result was most likely related to the lower MeCN viscosity (MeCN = 0.316, CH 2 Cl 2 = 0.43 mP·s), which facilitated the movement of ions and solvent.
The relations Γ = f(n) (n-number of electropolymerization cycles), for all polymers and in solutions of all supporting electrolytes, showed an increasing characteristic ( Figure 8). However, only for poly[Ni(salcn(Bu))] was this relationship linear over the whole range of electropolymerization cycles, which indicated that the conductivity of the film did not limit the electrolytic deposition under the investigated conditions. Such features of the polymer have also been noticed in the literature for another salen-type complex [23]. In the case of films of the other polymers, deviations from linearity were observed, the highest for poly[Ni(salcn)]. The reason was most likely the structure of this polymer, formed as a result of electropolymerization towards both active positions of the phenolate groups [36], which may have impeded conductivity in the thicker films. In the case of the poly[Ni(salcn(Me))] polymer, most probably due to the presence of substituents on the phenolate moieties, deviations from linearity occurred with thicker films. A decrease in electroactive surface coverage with an increase in film thickness was observed in the literature for other conductive polymers, such as polypyrrole, porphyrin complexes [48] and salen-type complexes [17,27]. Comparing the thin films of the investigated polymers, obtained as a result of one to three cycles, it can be seen that the poly[Ni(salcn)] films were characterized by the highest surface coverage. On the other hand, the lowest values of Γ occurred for poly[Ni(salcn(Bu))] films, despite the lack of limitations in electrodeposition of the film. The reason seemed to be less efficient electropolymerization [36], due to the large steric effect of the substituents.

Kinetic Analysis
The nature of the voltammograms of the investigated polymers depended on the film thickness, the polarization rate of the electrode, and the type of supporting electrolyte and solvent.
In case of poly[Ni(salcn)] in CH 2 Cl 2 , for its thin films, obtained by up to three electropolymerization cycles in TBAClO 4 solution, two-step electrode processes were observed at lower scan rates (Figure 9a). At higher scan rates, the first step of the electrode process was noticed only in the form of an inflection on the signal of the second step of the process Figure 9a). For thicker films, with higher scan rates, the process was only one-step ( Figure S14A, ESI). In contrast, in the TBABF 4 solution, in CH 2 Cl 2 , at higher scan rates, only one step of the electrode process was observed ( Figure S15A, ESI). This variability in the nature of the voltammograms indicated kinetic limitations that proceeded with the thickness of the films. Kinetic limitations were also observed for the poly[Ni(salcn(Me))] polymers. Thicker films, in the second step, at low rates, were oxidized to a very small extent ( Figure S16, ESI), and the first step of the electrode process became more and more electrochemically irreversible with increasing scan rate, both in TBAClO 4 and TBABF 4 (Figures 10a and S17A, ESI). On the other hand, kinetic limitations were least visible in the voltammograms obtained for poly[Ni(salcn(Bu))]. Regardless of the film thickness, they oxidized in two steps, and the anode and cathode peak potentials shifted less (Figures 11a and S18A, ESI) than was observed on voltammograms for poly[Ni(salcn(Me))] (Figures 10a and S17A, ESI). The exception was the TBABF 4 medium-at the highest scan rate the second anodic peak for poly[Ni(salcn(Bu))] disappeared ( Figure S19A, ESI).  The electrode processes in the films of all polymers were not electrochemically reversible. The greatest irreversibility was observed in the TBABF 4 solutions. The potential differences of the anode and cathode peaks for the poly[Ni(salcn(Me))] film were greater in TBABF 4 ( Figure S17A, ESI) than in TBAClO 4 (Figure 10a). In contrast, in the voltammograms of the films of poly[Ni(salcn)] ( Figure S15A, ESI) and poly[Ni(salcn(Bu))] ( Figure S19A, ESI) in TBABF 4 , the reduction peaks were absent or hardly visible. Despite significant electrochemical irreversibility, the polymer films were stable during the electrode processes. The successively recorded voltammetric curves did not show any significant differences ( Figures S20 and S21, ESI).
In MeCN, the voltammetric curves were less electrochemically irreversible ( Figures S22A and S23A, ESI) than the curves recorded in CH 2 Cl 2 (Figures 9a and S15A,  ESI), which was also shown on the basis of previous electrocatalytic studies [49].
The analysis of the kinetics of the electrode processes was carried out on the basis of cD 1/2 , values. For poly[Ni(salcn(Me))], they were determined only for the first step of the anode process and the corresponding step of the cathode process, due to the disappearance of the second step of the anode process with an increase in v. For all polymers, these parameters were determined for such film thicknesses, which to a relatively large extent showed the rectilinear relationship of the function i p = f(v 1/2 ), and thus ensured the diffusion process. Deviations from linearity in the initial section of the relationship i p = f(v 1/2 ) (e.g., Figure 9b) resulted from the surface process and indicated that a given film thickness at lower polarization rates of the electrode ensured oxidation of all active centres of the film ( [50], p. 595). On the other hand, deviations from linearity in the final segment of the relationship i p = f(v 1/2 (e.g., Figure S15B, ESI) resulted from kinetic constraints for the largest values of scan rates. They were especially visible for the polymer films of poly[Ni(salcn)] and poly[Ni(salcn(Me))].
Analyzing the values of cD 1/2 presented in Tables 3-5 on the basis of Figures 9b, 10b, 11b,c, S14B, S15B, S17-S23, S18C and S19C, ESI, it can be seen that, both in the anode and cathode processes, these values for each of the investigated polymers, obtained with the same number of cycles, were higher in the TBAClO 4 solution than in TBABF 4 , and in each of these solutions were lower than in TBAPF 6 ( Tables 3-5). Taking into account the different types of electrolytes used to electropolymerize a given complex, the c values in the films thus obtained (due to the same number of electropolymerization cycles) may differ, which makes it impossible to compare the cD 1/2 values. Therefore, in order to determine the dependence of kinetics on anion size, electropolymerization was carried out in TBAPF 6 solution, and redox switching of modified electrodes in solutions of all investigated electrolytes. For each polymer, the cD 1/2 values in TBAPF 6 /TBAClO 4 (electropolymerization in TBAPF 6 , redox switching in TBAClO 4 ) and in TBAPF 6 /TBABF 4 (electropolymerization in TBAPF 6 , redox switching in TBABF 4 ) were higher than the values obtained in TBAPF 6 (electropolymerization and redox switching in TBAPF 6 ) (Tables 3-5). The results showed that after electropolymerization carried out in the anion solution with the largest radius, the rate of electrode processes taking place on the modified electrodes was the highest in the solution of the smallest anions.   I  II  I  I  II  I  II  II  I  II  However, the least visible effect was in the poly[Ni(salcn)] films, both on the basis of cD a 1/2 and cD c 1/2 values, most likely due to electropolymerization towards both active positions of the phenolate moieties in the monomers. The cD a 1/2 values in TBAPF 6 /TBAClO 4 were only about 1.2 times higher than those obtained in TBAPF 6 ( Table 3). A slightly greater difference was obtained in TBAPF 6 /TBABF 4 versus TBAPF 6 , most likely due to the greater difference between the anion sizes used for electropolymerization and redox switching. In the case of the other polymers, these differences were greater (Tables 4 and 5). For poly[Ni(salcn)], both cD a 1/2 as well as cD c 1/2 values increased with increase in the number of electropolymerization cycles in each electrolyte solution, but to a lesser extent, indicating progressive kinetic limitations. For films obtained as a result of recording 15 and 20 cycles, values were even lower than for films obtained as the result of recording 10 cycles.      I  II  I  II  I  II  I  There were greater kinetic constraints during the cathode process than during the oxidation process. At low electrode polarization rates, 0.01-0.06 V·s −1 , a surface process was observed in the TBAClO 4 solution during the anodic process (Figure 9b), (the deviation from the linearity of the i pa = f(v 1/2 ) relationship, in the initial section), which did not not occur during the cathodic process. In addition, the cD c 1/2 values were lower than the corresponding cD a 1/2 values (Table 3), although, from the cD c 1/2 < cD a 1/2 relationship, it was difficult to unequivocally conclude that D c < D a , due to the electrochemical irreversibility of the electrode process, which means that the concentration of the active centres undergoing reduction may have been lower than those undergoing the previous oxidation reaction. However, EQCM studies showed that, in the case of this polymer, there was practically no solvent transport into the film structures during the oxidation processes (Table 1). Hence, it can be assumed that the concentrations of the active sites after oxidation will not have changed significantly and, from the cD c 1/2 < cD a 1/2 relationship, it was possible to conclude the D c < D a relationship.
The slower rate of reduction processes than oxidation processes in poly[Ni(salcn)] films was the opposite effect to that described in the literature for other salen-type complex polymers [23,25]. The explanation for our results seems to be the structure of the crosslinked poly[Ni(salcn)] polymer, which, during the transport of anions into the film by forced oxidation process, may have had limited scope to increase its volume. Hence, in the first step of the oxidation process, diffusion of the solvent from the films was observed in order to make room for the anions. The mass of such a film, increased during the oxidation process, undoubtedly reduced the free space of the film and may have hindered the transport of anions into the solution during the reduction process.
The  (Tables 4 and 5), which indicated that the optimal solution for increasing the speed of electrode processes was the greatest possible difference between the anions used for electropolymerization and redox switching of the modified electrodes.
The deviations from linearity in the initial sections of the relation i pa = f(v 1/2 ), both for anode and cathode processes (Figure 10b and Figure S17B, ESI), indicated easier kinetics in poly[Ni(salcn(Me))] than poly[Ni(salcn)] films, obtained by the same procedure. The results of the EQCM studies in TBAPF 6 and TBAClO 4 solutions showed that, both in the first and second step of the oxidation process, the transport of anions into the films was accompanied by diffusion of the solvent in the same direction (Table 2).
In the case of poly[Ni(salcn(Bu))], the type of electrolyte had the greatest impact on the kinetics of the redox processes, pointing to ion transport as a decisive step in the transport of charge through the polymer film. The lowest values of cD a 1/2 for modified electrodes in the TBAPF 6 solution became two and more times higher in TBAPF 6 /TBABF 4 (Table 5), while for poly[Ni(salcn(Me))] films in TBAPF 6 /TBABF 4 , these values were less than two times higher than TBAPF 6 ( Table 4). Efficient anode kinetics were also evidenced by the largest deviations from linearity in the initial sections of the i pa = f(v 1/2 ) relation, even for the thickest films (Figure 11b,c and Figure S18B,C, ESI), and no deviations from linearity in the final sections of these relationships were observed. Only in TBABF 4 did such deviations occur ( Figure S19B,C, ESI). The reason for the fastest kinetics of electrode processes in poly[Ni(salcn(Bu))] films, and the apparent influence of the type of ion on these kinetics, was probably the structure of the monomer. Furthermore, only in the case of this polymer in TBAPF 6 /TBAClO 4 was there a cD cI 1/2 > cD aII 1/2 relationship (Table 5), analogous to that described in the literature for electrodes modified with salen-type complexes [23,25], which showed the relationship D cI > D aII . The reason for the higher D cI values than D aII may have been the larger film volume obtained during oxidation, which facilitated the removal of ions and solvent from the film during the reduction process. The EQCM results showed that, in the case of this polymer, the greatest number of moles of the solvent penetrated into the structure of its films during oxidation processes four moles per one mole of anions in the first step of the process, about two moles in the second step (Table 2)).
Comparing the cD c 1/2 values obtained in the TBAClO 4 solution (Table 5), it can be seen that cD c1I 1/2 < cD cI 1/2 , which may indicate the effect of the reduced film volume, after the first step of the reduction process, as a result of removing ions and solvent from the film, on hindering charge transport in the next reduction step. On the other hand, when comparing the values of cD a 1/2 (Table 5), an inverse relationship, cD a1I 1/2 > cD aI 1/2 , in each electrolyte solution was observed. Taking into account the increase in the volume of films during the successive steps of the oxidation process, the dependence cD a1I 1/2 > cD aI 1/2 clearly showed the D a1I > D aI relationship. Here, the effect of reorganization of the film structure after the first step of the electrode process ( [50], p. 505) was most likely reflected, as a result of the interaction of functional groups ( [50], p. 505). Consequently, the film adopted a structure that was more conducive to charge delocalization ( [50], p. 505). This process was facilitated by the increased volume of the film after the first step of the oxidation process.
Voltammetric investigations in MeCN solutions allowed assessment of the influence of the solvent on the kinetics of the electrode processes. In MeCN, the cD a 1/2 values for poly[Ni(salcn)] polymer films were about 1.3 times higher than in CH 2 Cl 2 , relative to the electrolytes of the same type (Table 3). Moreover, for any of the tested film thicknesses, no decrease in the cD a 1/2 value was observed with increase in film thickness. The relationship i p = f(v 1/2 ) ( Figures S22B and S23B,C, ESI) did not substantially deviate from linearity in the end sections of these functions (but with slight deviations in TBABF 4 ). EQCM studies showed that the molar masses of the species transported in the films were also greater in MeCN than in CH 2 Cl 2 (Table 1). Moreover, for any of the steps in the oxidation process, no diffusion of the solvent from the polymer films was observed in order to make room for the electrolyte ions transported into the films. These features indicated easier charge transport in MeCN than in CH 2 Cl 2 . The reason for faster kinetics in MeCN than in CH 2 Cl 2 may have been the lower viscosity of MeCN, facilitating diffusion. Another reason may have been that the MeCN donor number (DN) was higher. This value for MeCN was 14.1, while for CH 2 Cl 2 it was 0. The consequence of a higher value of DN was an anion association which increased the anion radii and thus larger gaps in the polymer films obtained after the electropolymerization process, which may have had a positive effect on the conductive backbone, facilitating the relocation of the charge.

Instruments
Cyclic voltammetry was carried out using AUTOLAB PGSTAT 10 Eco Chemie in a three-electrode system. A platinum disk electrode (MINERAL), modified with polymer nickel complexes (Ptpoly[Ni(salcn(R))]), was used as a working electrode. The Ag/AgCl (1 mol·dm −3 KCl), connected to the bulk of the solution by a Luggin capillary, was used as a reference electrode. A platinum plate was used as a counter electrode. Measurements were recorded in CH 2 Cl 2 /TBAClO 4 and CH 2 Cl 2 /TBABF 4 (0.1 mol·dm −3 ), deoxidized with argon. All potentials are reported relative to the Ag/AgCl (1 mol·dm −3 KCl) electrode. The platinum disc electrode was cleaned in HNO 3 and aqueous suspension of 0.05µm alumina micropolish before each measurement.
EQCM measurements were performed using a module Autolab Electrochemical Quartz Crystal Microbalance for the AUTOLAB PGSTAT 302N, fitted with 6 MHz, AT-cut crystals coated with Pt. A platinum wire was used as a counter electrode and Ag/AgCl (1 mol·dm −3 KCl) was used as a reference electrode.
FTIR ATR spectra were performed on a Nicollet 8700 Thermo Scientific (Boston, MA, USA), on platinum disk electrodes, modified with the polymers obtained after electropolymerization process.

Electrode Modification Procedure
The platinum electrodes were modified by anodic electropolymerization. The process consisted in recording 1-30 cyclic voltammetric curves in solutions of complexes with a concentration of 10 −3 mol·dm −3 , in positive potential ranges. For the [Ni(salcn)] complex, a maximum of 20 cycles were recorded due to kinetic limitations. TBAClO 4 , TBABF 4 and TBAPF 6 were used as supporting electrolytes. The electropolymerization was carried out in the CH 2 Cl 2 medium, and, in the case of [Ni(salcn)], also in acetonitrile (MeCN). The other complexes did not dissolve in MeCN. The curves were recorded in the potential ranges covering the three-step complex oxidation process [36]. The films obtained under these conditions had higher concentrations of surface centres, especially poly[Ni(salcn(Bu))] films [38]. As a result of this process, yellow, electroactive polymer films, well-adhered to the electrode surface, were obtained. The electrodeposition processes were carried out at the polarization rate of the working electrode v = 0.05 V·s −1 . The thickness of the films was controlled by the number of electropolymerization cycles.

Procedures for Investigating the Modified Electrodes
Electrodes (platinum disc and quarzglas resonator) modified before electrochemical and electrochemical-gravimetric measurements were washed with a solvent and conditioned in a solvent for 10 min in order to remove as much of the complex as possible from the film structures. It was then conditioned in a supporting electrolyte solution for 5 min in order to stabilize the chemical equilibrium prior to the electrode process. For the electrodes prepared in this way, the first and the second voltammetric curve were basically the same.
The first cycles were recorded during electrochemical and electrochemical-gravimetric measurements in electrolyte solutions.
The number of electrons for the anode processes (z) was estimated according to the procedure described in [24], based on comparing the charge under the anodic part of the electropolymerization curve, obtained as a result of 1 cycle, and the charge under the anodic part of the voltammetric curve, recorded on the electrode modified in this way, in the electrolyte solution, on the basis of the relationship: Q pol a /Q a = (2 + z)/z, (Q pol a -charge under the anodic part of the electropolymerization curve; Q a -charge under the anodic part of the curve, recorded on the electrode modified in the electrolyte solution; 2-number of electrons related to complex oxidation). The curves were recorded at v = 0.01 V·s −1 in order to eliminate diffusion effects. The z values for poly[Ni(salcn)], poly[Ni(salcn(Me))] and poly[Ni(salcn(Bu))] were estimated to be 0.9, 0.8 and 0.6, respectively.
The study of the kinetics of the electrode processes taking place on modified electrodes was carried out using the cyclic voltammetry method, in TBAClO 4 and TBABF 4 solutions, at scan rates 0.01-1 V·s −1 , in the potential ranges including the two-step process of oxidation of polymer films, to bisphenoxyl radicals. The third step of polymer oxidation was hardly visible [37]. Due to the lack of electrochemical reversibility of the electrode processes, the potential range was increased along with the increase in the scan rate, which facilitated the precise determination of the currents of the last anode peaks and did not significantly The electroactive surface coverage Γ was determined on the basis of voltammograms recorded on modified electrodes, at low scan rates v = 0.01 V·s −1 , due to the achievement of conditions enabling complete oxidation of the polymer film.
The EQCM tests were carried out for modified electrodes obtained as a result of recording one electropolymerization cycle, at v = 0.05 V·s −1 , in TBAPF 6 , TBAClO 4 and TBABF 4 solutions. The voltammetric and gravimetric curves on the modified electrodes were recorded in the potential ranges covering the two-step process of oxidation of polymer films, with the electrode polarization rates v = 0.01 and 0.05 V·s −1 . The oxidation process was preceded by the reduction of films, carried out at potential E = 0 V for 10 s, and, in the case of poly[Ni(salcn)], for v = 0.05 V·s −1 also for 40 s. This ensured that inert films were obtained, and the mass and charge recorded during the anode process corresponded only to that exchanged during oxidation. The sensitivity coefficient of the quarzglas was determined based on silver electrodeposition tests.
FTIR ATR studies were carried out for modified electrodes obtained as a result of recording 10 electropolymerization cycles. Modified electrodes were washed with solvent (0.5 cm 3 ) and dried at room temperature for 20 min. Spectra were recorded for neutral and oxidized films, at a potential value equal to 0.8 V. Under these oxidation conditions, EQCM studies showed that it was easiest to compare the effect of the solvent on charge transport.
Charge transport through polymer films of modified electrodes was analyzed based on FTIR ATR spectroscopy, cyclic voltammetry and EQCM, in TBAPF 6 , TBAClO 4 and TBABF 4 solutions, in CH 2 Cl 2 and MeCN. It was shown that in CH 2 Cl 2 , anions and a solvent were transported into poly[Ni(salcn(Me))] and poly[Ni(salcn(Bu))] films during the oxidation process. Only anions were transported into the films of the cross-linked poly[Ni(salcn)] polymer, while the solvent in the first step of the oxidation process diffused in the opposite direction. In this way, this polymer provided a place for anions transported into its structure.
It was found that the optimal solution for effective charge transport was to use the electrolyte with the largest anion (TBAPF 6 ) for electropolymerization, and the smallest anion (TBABF 4 ) for redox switching of the modified electrodes. For each type of polymer, under these conditions, the cD 1/2 values were the highest.
Transport of the charge was the easiest through the films of poly[Ni(salcn(Bu))], most likely due to the high steric effect of the tert-Bu substituents. This was evidenced by the highest values of cD 1/2 , the largest amount of solvent diffusing into the structures of its films (n s ), and the largest deviations from linearity in the initial sections of the relation i pa = f(v 1/2 ), indicating significant participation of the surface process. The structures of poly[Ni(salcn(Bu))] films, regardless of their thickness and type of electrolyte, did not block the solvent movement, as indicated by the same n s values in the solutions of each of the electrolytes used, except for the second step of the electrode process in TBABF 4 .
The kinetics of the electrode processes in poly[Ni(salcn)] films were faster in MeCN than in CH 2 Cl 2 , most likely due to the lower viscosity of MeCN, facilitating diffusion, as evidenced by higher values of cD 1/2 and n s , and smaller deviations from linearity in the final sections of the relationship i pa = f(v 1/2 ) in MeCN. Another reason for the better quality of kinetics in MeCN was the higher DN value, which increased the association in this solvent, which caused the appearance of larger gaps in the poly[Ni(salcn)] films after the electropolymerization process, resulting from the transport of larger ions.