A Mechanism for Origin of Reversible Redox Transitions of Molybdenum/Surface Molybdenum Oxides

Figure 1 shows CVs of Mo/glass electrodes recorded at various pH levels and with the three electrolytes bearing different anions. Peaks corresponding to two redox processes emerged. In SO₄²⁻, the major redox peaks (a₁ and c₁) were evident at pH 1.5, 2.5, 3.0 and 3.5. A similar variation was observed in Cl⁻, even though a clear redox peak was not detected at pH 3.5 at scan rate (ω) ≥ 0.10 V.s⁻¹; whereas in ClO₄⁻, the redox peak was not apparent when the pH exceeded 2.5. The minor redox peaks (a₂ and c₂) in general underwent analogous variation to the major redox peaks. The redox transitions were not distinct in neutral media.

Zoom In Zoom Out Reset image size

The linear dependence of i_p of peak c₁ with ω confirmed that the peak was associated with the surface oxide species and the electrode process was adsorption controlled (Figure 2).⁸ The i_p and v_p of peak a₁ showed a similar variation (Supporting information, Figure S3). At pH 1.5, the difference between the v_p of a₁ and c₁ peaks in Cl⁻, SO₄²⁻ and ClO₄⁻ were always less than 0.06 V and the peak current ratios (i_pc₁/i_pa₁) were near unity: all of which are characteristics of reversible redox processes. The corresponding peak separations at pH 2.5 were 0.059, 0.053 and 0.080 V (at 0.10 V.s⁻¹), respectively. The c₁ and a₁ peak separation showed a proportional increment with pH increased, which is indicative of the deviation more from reversibility.⁸

Zoom In Zoom Out Reset image size

The role of cations and anions on the major redox transition can be understood from Figures 1 and 2. The decreased i_p of the c₁ peak and the shifting of v_p to more negative values with increased pH can be attributed to the decreased H⁺ concentration in the electrolyte (main reducing agent). The maximum i_p and most positive v_p of the c₁ peak were recorded in SO₄²⁻ except at pH 3 where the i_p of the c₁ peak in Cl⁻ was higher than the corresponding value in SO₄²⁻. The i_p of c₁ peak were lowest in ClO₄⁻ with a more negative v_p irrespective of the pH level. These variations can be explained as due to a more effective interaction/adsorption of SO₄²⁻ and Cl⁻ at the surface oxides.³ The absence of redox peaks and the decreased overall reduction current above pH 2.5 in ClO₄⁻ can be correlated to its adsorption capabilities.^3,9,10 Figures S4 and S5 (Supporting information) compares CV, i_p and v_p with the nature of anions. The odd behavior in Cl⁻ at pH 3 (where i_p of c₁ peak was higher than those at pH 1.5 or 2.5) can be explained by the complex Cl⁻ chemistry and its ability to form strong interaction with the hydrated surface oxides.^3,11 The passive surface oxides at pH 1.5 and 2.5 is likely to be different from those at pH 3 or 3.5 (Supporting information, Figure S6). It was expected that the MoO₂/MoO(OH)₂ ratio of the passive film^12,13 would decrease at pH 3 and hence excess hydroxide facilitated more effective Cl⁻ interaction/adsorption. According to the passive film structure proposed by Lu and Clayton,¹² the film at the potential range of redox transitions was composed mainly of Mo(IV) oxide/hydroxides (MoO₂ and MoO(OH)₂). The outer portion of the film is expected to be rich in the hydroxide fraction and the inner layer rich in MoO₂.

Interestingly, there was an absence of redox transitions in neutral media. We primarily attributed this to the thin passive surface oxide and its instability in neutral/alkaline media leading to the formation of soluble surface species.² The Na⁺ or anions in the electrolyte would have contributed to the redox peak provided a stable passive surface oxide or a much thicker oxide layer existed on Mo. Despite the apparent instability, presence of a thicker oxide layer may provide ample catalytically active centers for the redox process. Because of the excess hydroxides in the passive film¹² and the reduced H⁺ activity in the electrolyte, anion interactions can be facilitated at higher pH levels provided the thin passive surface oxides remains stable. The absence of distinct redox behavior in Cl⁻ at pH 3.5 at ω ≥ 0.1 V.s⁻¹ (Figure 1) may also be related to its corrosive interactions. Time-dependent anion adsorption (at OCP) or anion adsorption during repeated CV scans was not found to have any significant impact on the major redox transition (Supporting information, Figure S7).

OCP transient studies coupled with Ar gas bubbling experiments and EIS phase angle plots (Supporting information, Figure S8 and S9 respectively) were employed in order to gain insight as to the extent and nature of the interactions between the cations/anions and the surface oxides. Representative OCP transient plots were recorded in SO₄²⁻ (Supporting information, Figure S8). The OCP was initially recorded after immersion of the Mo electrode in the de-aerated electrolyte for 10 min (line a). Next the electrode was subjected to an LSV from –0.05 to –0.60 V (the potential range corresponding to the redox transition) and the OCP was recorded thereafter (line b). Subsequently Ar bubbling was performed for 2 min prior to additional OCP recording (line b^l). The lines c and c^l corresponds to similar experiment where LSV was extended up to –0.90 V (from –0.05 to –0.90 V) on a new electrode. The OCP shift was significant after the LSV corresponding to the redox transition (lines a and b). However, soon after the Ar bubbling, the OCP retained potentials close to the original value (line b^l is close to line a). Such a behavior suggests that the ionic (anion and/or cation) interaction/adsorption at the surface oxides that occurred during the LSV corresponding to the redox transition was weak. Perhaps this indicates that the anion interaction during the potential range of redox transition is predominantly electrostatic in nature. The figure also indicates that the cation interaction at the potential range of redox transition mainly transpires at the surface/near-surface of the passive oxides and significant irreversible ingression of H⁺/cation into the oxides occurs only after the potentials corresponding to the redox transitions. This was evident when the LSV was performed to more negative values (lines c and c^l), where Ar bubbling does not considerably affect the OCP variation. Under these conditions, the electrode was unable to retain the original OCP (line a). Similar variations were observed for the corresponding Cl⁻ and ClO₄⁻ electrolytes (data not shown).

The difference between the lines a and b^l can be associated with the electron transfer processes and/or the ion's adsorption/interaction. Considering the adsorption factor, one can suggest that only a few anions would have specifically adsorbed and/or a few cations would have irreversibly adsorbed during the reduction process corresponding to the redox transition. We propose that most of the anion interaction would have happened via electrostatic attraction possibly via the hydrated oxide (MoO(OH)₂). The extent of chemisorbed Cl⁻ and SO₄²⁻ are expected to be higher than that of ClO₄⁻. To the best of our knowledge, there are only a few studies in which experimental evidences were provided for the existence of anion (Cl⁻) on Mo passive film through in-situ XPS studies.^7,12 When compared to the anion interaction, cation migration to the surface oxides is a much more facile process and the cations can be intercalated into the film of oxide layer. Both the anions and cations in the electrolyte can contribute to the redox transition through interaction/adsorption at the surface oxides.³

EIS phase angles plots corresponding to the redox transition, recorded before and after the LSV, provided noticeable differences among the three electrolytes (Supporting information, Figure S9). At low frequencies in ClO₄⁻, the decrease in phase angles provides an evidence for the higher concentration of H⁺/cations charging into the surface oxides. The higher H⁺/cations charging in ClO₄⁻ can be associated with a lower anionic barrier effect (due to the lower adsorption tendency of ClO₄⁻),³ and/or due to the possible formation of a marginally thicker surface oxide (due to the higher oxidizing nature of HClO₄). The second factor was evident through variation in the high frequency region where higher phase angles were always recorded in ClO₄⁻. Conversely, the decline in phase angles at low frequencies indicates the deterioration of the protective properties of the passive film. This supports the lower adsorption capabilities of ClO₄⁻ (faster H⁺ ingression) and the associated lower i_p of c₁ peak in Figure 1. Both the SO₄²⁻ and Cl⁻ exhibited a similar variation. Further discussion on EIS is provided in Section - EIS studies.

Both the major and minor redox transitions were expected to change with the surface species (cations/anions interacting at the surface oxides). Several alternative electrochemical experiments were conducted in order to reveal any significant variation in the minor redox peak. Among them, the time-dependent studies in Cl⁻ revealed some interesting results (Figure 3). The peak a₂ showed a noteworthy enhancement in the CVs recorded after 6 h. In order to study the variation of the redox peaks, the electrode was kept immersed in the same electrolyte for 8 h (t = 8 h) following the initial CV scan (t = 0 h). The i_p vs. ω plot of the CVs recorded at t = 8 h for peaks c₁, a₁ and a₂ are provided (Supporting information, Figure S10). The c₂ peak was not present to a significant level. Both the c₁ and a₁ peaks showed adsorption controlled behavior (similar to Figure 2). However, the i_p vs. ω plot of peak a₂ showed an apparent deviation from the linear behavior.

Zoom In Zoom Out Reset image size

Figure 4 shows the CVs (after t = 8 h) scanned at different potential ranges. The minor redox transition was not seen when the maximum negative potential was –0.20 or –0.35 V (Figure 4a and 4b). However, peak a₂ was observed once the cathodic limiting potential was extended to –0.40 V (Figure 4c). Both the peaks a₁ and a₂ become evident upon extension of the scan to include more negative potentials, but only after the appearance of the peak c₁ (Figure 4d and 4e). The corresponding plots at a lower ω of 0.01 V/s are also provided (Figure 4f – 4i). Here, the i_p of peak a₂ was significantly higher than that of peak a₁. Initially it appeared as if the i_p of peak a₁ varied in proportion to the i_p of peak a₂ over time (Figure 3c). However, this was not the case as the variation of peak a₁ was in a similar line to when i_p of peak a₁ remained constant after an initial decrease at t = 2 h (Supporting information, Figure S7). A gradual decrease in the i_p of the a₁ or c₁ peak with increased CV scans is characteristic of the system of Mo with passive surface oxides and can be correlated with the decreased availability of the surface oxides for the redox process due to the initial CV scans.²

Zoom In Zoom Out Reset image size

The oxidation/desorption process corresponding to the peak a₂ should have been clear with the increase in the number of CV scans. Yet, the similar variations were not observed when the experiments were repeated in SO₄²⁻ or ClO₄⁻ irrespective of the cationic species used (Na⁺ or K⁺). Representative plots recorded in SO₄²⁻ are provided in Figure 3d. The data suggest that the variation observed for the peak a₂ (Figure 3b and 3c) is associated with either Cl⁻ or cations or both. By increasing the time from t = 0 to t = 8 h, a small number of adsorbed Cl⁻ or cations should have been able to participate in the oxidation process. Due to the thin passive oxide layer, such a process may not occur at t = 0 h. This is further supported by our multiple cycle CV experiments recorded at t = 0 h (Figures 5 and 6). The variation of the major redox transition with electrolyte anions is further evident from Figure 5a–5c. The minor redox transition was well resolved in Cl⁻ and SO₄²⁻ at higher pH (Figure 5e–5g). The continuous increase expected in overall reduction current, resulting from an increase in the number of CV cycle was not observed due to the thin passive surface oxide. After the first CV cycle, the overall reduction current showed an abrupt decrease followed by a marginal gradual increase. Figure 6 shows corresponding plots recorded from a more positive potential (scan started from 0.35 V). It was expected that the electrode would have a thicker transpassive Mo(VI) surface oxide due to the positive potential range used. The CVs vary significantly with anion content in the electrolyte (Figure 6a–6c). A continuous increment in the overall reduction current was not observed in SO₄²⁻ (Figure 6b), while it was for Cl⁻ (Figure 6a). This is perhaps due to the formation of a more effective SO₄²⁻ anionic barrier layer at the surface oxide.³ The absence of such an effective anionic barrier layer in Cl⁻ and ClO₄⁻ would facilitate the continuous ingression of cations to the thicker surface oxide. Change in the overall reduction current may also depend on the variation in the anion reduction at the surface oxides. The data indicate that with an increase in the number of CV cycle, the reduction tendency of anion is in the order of Cl⁻ > ClO₄⁻ > SO₄²⁻. Considering the faster cation reduction/intercalation kinetics and the presence of thicker oxide layer, we theorize that the contrasting behavior in the overall reduction current variation in the three electrolytes contains significant contribution from the restricted cation reduction kinetics by the SO₄²⁻ anionic barrier layer formation (Figure 6a–6c).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The appearance of the two oxidation peaks (Figure 6a) is interesting and is analogous to that of Figures 3 and 4. At higher ω, the two oxidation peaks seem to correspond to a single reduction peak. However, the equivalent reduction peaks became evident at a low ω (Figure 6f). The redox peaks were very close to each other. The anodic peaks were well resolved in Cl⁻ whereas it was least resolved in SO₄²⁻. The SO₄²⁻ barrier layer seems to prevent the oxidation/desorption corresponding to the redox transition. This suggests that the cations play a pronounced role in the appearance of the oxidation peaks. Well-defined anodic peaks in Cl⁻ (Figure 6a) when compared to ClO₄⁻ (Figure 6c) suggest that Cl⁻ interaction/desorption also contributed to the oxidation peak.

When the redox transitions are attributed to the ion's adsorption/interaction at the surface oxides, it is reasonable to think that the appearance of the two redox peaks can be associated with the variations in the adsorption strength and/or adsorption sites. Considering the passive oxide structure,^12,14–16 and the results of the present and our earlier³ studies, we propose that the major and minor redox transitions are associated with two thermodynamically non-equivalent adsorption sites in the surface oxides. The following description is based on the consideration that the major passive surface oxide of Mo is of the structured MoO₂ or its hydrated forms.^14–16

Mo(IV) oxide crystallizes normally in monoclinic systems. The structure of its constituent building blocks is distorted [MoO₆], where the Mo-O bonds vary in length (between 1.97 and 2.07Å) and the O-Mo-O angles are in the range of 78–97°. In MoO₂ structures, the two types of oxygen centers are each coordinated to three Mo.^14–16 The MoO₆ octahedra can be cut in different ways. This gives rise to the formation of different surface terminations with various surface active centers. The atomic layers parallel to the surface are ordered as O^'-MoO-O-O^' on the surface of the most commonly studied (011) monoclinic system. Hence such a surface can have different terminus with exposed O and O^' centers as wells as Mo and O centers.¹⁶ The Mo centers in the surface oxides bear a positive metallic character since only a limited number of Mo atoms/unit cell can have d electrons.¹⁷ Density of states plots of MoO₂ showed that the Fermi level was located within the band dominated by Mo orbitals, and the system exhibited a clear metallic behavior.¹⁶ During the reduction process, the Mo centers accept electrons and the electrolyte ions are attracted mainly towards the O centers.

The redox transitions can be related to the adsorption/interaction of anions/cations to two different O centers of the surface oxides (Scheme 1). Analogous to the weak and strong H⁺ adsorption peak on Pt,¹⁸ the minor and major redox peaks can be correlated to a stronger and weaker ion interaction/adsorption, respectively. Thus the major redox peak can be attributed to the electrostatic interaction of anions with the O or -OH centers. The hydrated outer oxide layer is rich in MoO(OH)₂, which can favor such a process.¹² It is expected that many of the coordinated oxygen species exist as terminal -OH or OH₂. This results in a far less cross-linked structure that is much more dispersed during the reduction half cycle. It is worthy of note that H⁺/cations can be easily adsorbed and even migrate to the interior of the oxides without much difficulty. The adsorption of the cations is expected to enhance the major redox peaks in much the same manner as that of the anions. As the minor redox transition is appearing at the initial potential ranges, it can be correlated to faster ion adsorption occurring preferably at the most easily accessible O or O^' centers. The reduced i_p associated with the minor redox peaks can be explained as due to the lower extent of ion adsorption/interaction with the O^' centers. This is perhaps the reason for the pronounced i_p of peak a₂ during the Ar bubbling experiment in which the maximum desorption of specifically bound anions (Cl⁻) would have occurred.

Scheme 1. A suggested mechanism for the interaction/adsorption of electrolyte anion/cation at the surface oxides contributing to the redox transitions. The passive oxide film is represented by the monoclinic structure of MoO₂.^12,14–16 The redox transitions were correlated with the ion's interaction/adsorption at thermodynamically non-equivalent sites. The MoO₂ structure was taken from ref. 16 (Copyright Elseveir, 2011), where (011) surface was modeled. The process represented at the surface is expected to occur in the same sequence in the film of the oxide.

The oxidation peaks a₁ and a₂ can be attributed to the desorption processes about the O and O^' centers. The variation from linearity of i_p vs. ω plot of peak a₂ (Supporting information, Figure S10) suggests that the extent of desorption/oxidation process at peak a₂ may involve contributions from processes other than desorption of surface anions or cations, such as desorption of cations trapped/intercalated at the surface oxides. The higher i_p of peak a₂ when compared to that recorded for peak a₁ at a low ω (Figure 4i) can be associated with this. The time provided by the slow ω would have continuously allowed a greater number of cations to desorb, which in turn would have resulted in larger i_p of peak a₂. These observations suggest that a small number of anions/cations that were firmly linked with O centers at peak a₁ may be desorbed at peak a₂ in addition to those from O^' centers. The redox transitions and the associated pseudocapacitance of the surface oxides, corresponds primarily to the surface or near-surface reversible redox reactions.^1,2 The cations intercalation in the gap between the interlayer spacing of the layer-layer structure through single phase faradaic reaction is not expected to affect the i_p of cathodic peaks significantly (not the overall variation of the cathode current density); however, it may affect the variation in the i_p of anodic peaks as discussed below.

During oxidation, high resistance is expected between the compact oxide/electrolyte interface. Because of the dispersed nature of the reduced film, however, electroactive species can freely enter/adsorb into the film throughout the reduction half cycle, while the oxidation of the adsorbed species can occur during the reverse scan. The higher i_p of peak a₂ (Figures 3 and 4) suggests that during the reduction half cycle an increasing number of cations and Cl⁻ were able to get trapped/intercalated in the oxide. These species were later oxidized/desorbed during the reverse cycle. At first glance the negligible presence of peak c₂ (Figures 3b–3c and 4) is contrary to this. However, careful examination of Figures 1, 3 and 4 reveals that the reduction process contributing to peak c₂ is happening throughout the reduction half cycle whereas the oxidation process is particularly sensitive to the desorption potential. Due to the compact oxide film present during the oxidation half cycle, the desorption process is specific and stringently occurred at the oxidation potential corresponding to the peak a₁ and a₂. This fact is further supported by the integrated area calculation. The sum of the integrated area under anodic peaks a₁ and a₂ was ∼54% higher than that of the cathodic peaks (Figure 4d). At low ω (Figure 4i), this value increased to ∼63%.

The two oxidation peaks obtained in multiple cycles CV experiments (Figure 6) and in time-dependent studies (Figures 3b–3c and 4) vary significantly in potentials. In multiple cycles CVs, the thicker Mo(VI) oxide becomes the controlling factor as the scan started from transpassive potentials. Due to the thicker oxide layer, the role of cations dominates the redox transitions whereas the anions become less significant. In this case the two anodic peaks can be mainly attributed to the oxidation/bond breaking/desorption of cations from the two adsorption sites in the surface oxides. The reason for the pronounced appearance of the oxidation peak in Cl⁻ (Figure 6a) may be related to its ability to form strong bonds with the surface oxides. It is well known that halides have the ability to complex with Mo species of different valence.^19,20

Studies on bulk Mo oxides (not metallic Mo) resulted in different views on the origin of the redox peaks.^5,6 Anbananthan et al. assigned the minor redox peak to Mo(VI)/Mo(V) reduction and the major redox peak to the formation of Mo bronzes due to the removal of O ions and the associated structural relaxation.⁵ Studies by Wang and Dong⁶ on Mo(VI) oxides suggest that the two redox transitions corresponds to the formation of two Mo oxide bronzes. Potential contribution via sub-stoichiometric lower Mo oxides formation was also suggested.⁶ The formation of bronzes such as blue (A_0.3MoO₃), red (A_0.33MoO₃) and purple (A_0.9Mo₆O₁₇),^1,21–23 can take place during the reduction half cycle and that primarily depends on the availability (thickness) of the oxide layer and the concentration of the electrolyte cations. While we have not observed a color change corresponding to the bronze formation on the electrode's surface, this is likely due to the presence of thinner surface oxides. However, when the CV scan was started from more positive potentials where thicker transpassive surface oxides formed, distinct blue color was formed in the potential range of the study.³

Some aspects of the present work are in agreement with the above views.^5,6 On the other hand, we explained the phenomenon in terms of interaction/adsorption of electrolyte ions at the surface oxides. The redox transition corresponds to surface species and is adsorption controlled. O ion removal, ion reduction, bronzes formation and electron accepting process are expected to be taking place throughout the reduction half cycle. These processes involve a significant contribution from electrolyte cations/anions and the way they interact with the surface oxide. The extent of enhancement of the i_p of the redox transition can be employed for detecting a particular electrolyte species in trace amounts. The species adsorbing at the electrode surface can change the i_p through non-faradaic process with a relative shifting of the v_p. Standardizing the variation of the redox transition with and without the species to be detected can yield advantageous results. In general, voltammetric sensors examine the concentration effect of the detecting species on the current-potential characteristics of the reduction or oxidation reaction involved and the height of the i_p can be used for the quantification of the concentration of the oxidation or reduction species.²⁴ The following EIS studies further substantiate the variation of the interfacial behavior with different electrolyte ions.

EIS studies were conducted at two different potentials analogous to the major and minor redox transitions in order to identify the interfacial behavior in the three selected electrolytes. In addition to the de-aerated media (DA), the experiments were also conducted in oxygenated media (OX) in order to study the variation of the EIS plots with different electrolyte anions. EIS plots recorded in DA were curve fitted and the C and the R_ox were compared.

The protective anionic barrier layer formation of SO₄²⁻ is prominent (Figure 7). The behavior of SO₄²⁻ and Cl⁻ was similar in de-aerated media although a marginally higher oxide film resistance (R_ox) was observed in the case of SO₄²⁻ (Figure 7g). A higher R_ox in SO₄²⁻ may indicate a more effective anionic barrier layer formation, while the lower R_ox in ClO₄⁻ can be associated with faster H⁺ ingression. The complex interaction of Cl⁻ is especially evident in the presence of dissolved oxygen. This is indicated by the appearance of depressed semicircle in the Nyquist plots. These variations were pronounced at –0.50 V, where the oxide film was expected to be more conductive and porous.

Zoom In Zoom Out Reset image size

A lower capacitive behavior with the reduced resistance in the presence of dissolved oxygen may correspond to the formation of a thicker surface oxide layer (Figure 7). SO₄²⁻ exhibited the typical behavior in the oxygenated media in accord with the protective passive layer formation, whereas Cl⁻ showed the depressed semicircles. Higher R_s observed in Cl⁻ at –0.50 V (evident from the Nyquist plot in Figure 7d) can be attributed to the corrosive involvement of Cl⁻. Representative Bode plots were provided (Supporting information, Figure S11).

The lowest R_ox was recorded in ClO₄⁻ and can be explained in terms of the oxidizing power of the acids, analogous to the behavior of Nyquist plots in OX media. Higher capacitive behavior and reduced R_ox in ClO₄⁻ may be associated with the comparatively higher oxidizing power of HClO₄ (Figures S9 and 7). If so, the Mo electrode could have a marginally thicker passive oxide layer in ClO₄⁻ when compared to Cl⁻ or SO₄²⁻ and the i_p of the c₁ peak would have a maximum value in ClO₄⁻ due to the higher availability of the surface oxides. However this was not case. This further supports our postulated mechanism on the redox transition. This also suggests that the peak c₁ does not completely correspond to the extent of cation ingression or bronze formation at the surface oxides.

The variation of the Nyquist plots as a function of electrolyte pH (Figure 8) reflects the decreased passive film stability in the neutral medium. Here, the reduced impedance in the neutral medium (Figure 8a–8c) suggests the formation of soluble surface species. This, in effect, would hinder the appearance of the redox transitions (due to the nanometer scale thickness of passive surface oxides). The R_ox in ClO₄⁻ was always lower than that of Cl⁻ and SO₄²⁻ at lower pH (Figure 7g and 8e). However, in the neutral medium the opposite was observed (Figure 8e). A similar pattern was observed when the experiment was repeated at more negative potentials (Figure 8d). In the neutral medium, R_ox showed a significant increase at –0.50 V as compared to the –0.15 V experiment (Figure 8d). This is in contrast to Figure 7 where R_ox at –0.50 V was always lower than that of –0.15 V. This behavior can be attributed to the formation of a thicker oxide/hydroxide layer on the electrode's surface due to the unstable surface oxides in the neutral pH.^2,13 Such a process can be facilitated by the conductive porous oxides formed at more negative potentials.

Zoom In Zoom Out Reset image size

A comparison of the interfacial C was provided (Figure 7h and 8f). The total interfacial C of Mo-oxides can have contributions from both non-faradaic double layer (expected to be low; 20–50 μF cm⁻²) and faradaic pseudocapacitance (more than ten times the pure double layer capacitance). The faradaic charge storage is expected to be taken place primarily by cation (H⁺/Na⁺) adsorption onto the oxide surface by redox processes and/or through cation intercalation in the interlayer spacing of the layered oxide structure through single phase faradaic reaction.²⁵ As the cathodic i_p of the redox transition was directly proportional to ω, the contribution from intercalation pseudocapacitance can be suggested to be not significant as in such a case charge storage is limited by solid state diffusion and therefore the i_p would scale with square root of ω, or it can be assumed that the intercalation process is taking place mainly in the near-surface region. A higher C obtained at –0.50 V when compared to that at –0.15 V can be attributed to a higher cation adsorption/intercalation at the surface oxides.³ The decreased C in the neutral pH is associated with the decrease of H⁺ concentration in the electrolyte where the faradaic process can be largely attributed to Na⁺. The marginal variation of C with different electrolyte anion can be correlated to the variation in the anionic barrier layer effect.³

XRD of Mo electrode, model showing peak fitting in CV, i_p and v_p of peak a₁, comparison of CVs with nature of anions, time-dependent variation of i_p of c₁ and a₁ peaks, OCP decay experiments during Ar gas bubbling experiments, EIS phase angle plots before and after LSV, i_p vs.ω plot corresponding to Figure 3b and representative EIS Bode plots were provided.