In Situ Ellipsometry and EIS Study of Potentiostatic Synthesis of Pseudocapacitive MnO_x

The electrochemical apparatus consisted of a three-electrodes configuration in a one-compartment cell connected to a potentiostat Gamry 1010E. The counter electrode was a platinum sheet and the reference was a saturated calomel electrode (Hanna Instruments). The working electrode consisted of a 304L stainless steel sheet (Goodfellow, UK) connected to a copper wire through colloidal silver suspensions, and encased in epoxy resin (EpoKwick^TM FC, Buehler, USA) forming a disk with an exposed geometric area of 1 cm². The polishing of the working electrode was initiated using a metallographic polisher with sandpapers with a successively smaller grain size down to SiC grit 4000. Then, the electrode was polished manually on a polishing cloth using alumina suspensions with grain sizes 1, 0.3 and 0.05 μm until obtaining a mirror-like surface. Finally, the electrode was cleaned with abundant acetone and Millipore water to remove polishing impurities.

Electrochemical experiments were performed inside a Faraday cage at room temperature and without deaeration of solutions.

The electrodeposition of MnO_x was carried out in an aqueous solution with a pH of 6.5 of 0.1 M LiClO₄, as supporting electrolyte, and 0.3 M MnSO₄.H₂O. For the study of the applied potential effect, it was used a total charge density of 40 mC cm⁻². In-situ ellipsometric experiments were performed during SS 304 oxidation and MnO_x electrodeposition through a monochromatic ellipsometer with a rotating analyser (Sentech SE 400), which is equipped with a He-Ne laser (λ = 6328 Å), at an incident angle of 70°. The refractive index of the electrolytic solution (n_e = 1.3405) was determined prior to the ellipsometric tests with an Abbe "60" refractometer.

The electrochemical characterization was performed in an aqueous solution of 1 M Na₂SO₄. EIS measurements were executed with a 10 mV rms potential perturbation around OCP, in the frequency range from 100 kHz to 10 mHz, and with 10 points per decade with a logarithmic distribution. The electrochemical response of the modified electrodes was determined by cyclic voltammetry and galvanostatic charge-discharge (GCD) tests and the materials stability was accessed by GCD for 5000 cycles.

In the case of using an active substrate, such as stainless steel, the possible formation of a resistive oxide layer on the electrode must be considered. Therefore, before starting the synthesis of manganese oxide films, the behavior of the stainless steel 304 electrode under a potentiostatic regime, without the presence of the manganese salt, was investigated through in situ ellipsometry. Ellipsometry is based on the variation of the polarization of a specularly reflected light beam after interacting with a given surface. In an ellipsometric test, the polarization change is determined by the phase difference — Δ — and the magnitude ratio — Ψ. Phase difference is a very sensitive parameter to the thickness variation. Figure 1a presents the evolution of phase shift with time, at different applied potentials — 0.8, 0.9 and 1.0 V vs SCE — in a 0.1 M LiClO₄ solution. The small variation of Δ (3°–6°) after 10 min of anodic pulse suggests the formation of a very thin layer, which increases in thickness with increasing the applied potential. The current transient, with the simultaneously recorded Δ evolution at 0.8 V (vs SCE) for the first 10 s, is represented in Fig. 1b. The increase in the thickness of the native oxide occurs substantially in the first seconds of oxidation and then evolves very slowly due to the resistive nature of the passive layer.

Zoom In Zoom Out Reset image size

A three-phase model (solution, film and substrate) was used to simulate the growth of a single homogeneous layer onto the substrate. This procedure consists in determining the complex refractive index of the film (${\hat{n}}_{f}={n}_{f}-i{k}_{f}$) that allows obtaining the best adjustment between the theoretical and experimental ellipsometric curves — Δ vs Ψ — without losing the physical meaning of the system. The model used to calculate the optical parameters and thickness (d) of the film is based on the fundamental equation of ellipsometry, which relates the ellipsometric parameters — Δ and Ψ — with the Fresnel reflection coefficients — ${\hat{r}}_{p}$ and ${\hat{r}}_{s}:$

$$\begin{eqnarray}&&\tan {{\rm{\Psi }}e}^{i{\rm{\Delta }}}=\frac{{\hat{r}}_{p}}{{\hat{r}}_{s}}\end{eqnarray} \tag{ 1 }$$

In practice, the ellipsometric parameters are a function of the optical parameters of the 3 phases and the thickness of the film. The refractive index of the electrolyte solution is determined using a refractometer, while that of the substrate is determined before electrodeposition. As such, we get an equation with two variables (Ψ and Δ) and three unknowns (n_f , k_f and d_f ), which is not possible to solve analytically. Therefore, it is necessary to estimate one of the parameters, which is usually the film thickness.

The determination of the thickness of the oxide layer on stainless steel was performed assuming a constant refractive index — n — of 2.3, as reported by several authors. ^15,16

The extinction coefficient — k — of the films decreases with increasing applied potential, evidencing the increase in resistivity. After about 5 s of pulse, the substrates subjected to 0.8, 0.9 and 1.0 V present an oxide layer with a thickness of 8, 12 and 15 Å, respectively. After this initial stage, the rate of growth becomes independent of the applied potential, presenting a value of approximately 0.18 nm s⁻¹. Table S1 (see supplementary material) summarizes the optical parameters and thickness obtained for the different experimental conditions.

The current-time transients of the MnO_x electrodeposition at different applied potentials, for a total charge density of 40 mC cm⁻², are shown in Fig. 2.

Zoom In Zoom Out Reset image size

The sharp decrease in current at the early stages is due to the formation of the double layer and oxidation of the electroactive species in the solution. After this initial stage, the current evolution evidences a slow decrease, reaching a non-zero limiting value, which is characteristic of a resistive film.

The potentiostatic growth of MnO_x was followed by in situ ellipsometry. The measurement started from the ellipsometric parameters — Δ and Ψ — characteristic of bare substrate, which depend on its nature and on the surface condition (polishing degree and presence of contaminants). After the pulse application, their evolution with time began immediately, as shown in Fig. 3, showing that the film formation is also immediate. The faster evolution with increasing the applied potential suggests an increase in the kinetics of the nucleation process. In the case of growths carried out at the lowest potentials — 0.70, 0.75 and 0.80 V_SCE — it is possible to identify the formation of a passive layer on the substrate (inset of Fig. 3a), as predicted by the initial characterization of the substrate by ellipsometry, while at higher potentials the passive layer formation it is not observable, possibly due to the lower number of experimental points in the initial stage of synthesis.

Zoom In Zoom Out Reset image size

Again, the simulation of the growth of the MnOx films onto stainless steel was performed through the three-phase model (solution, film and substrate), previously described. However, for these films, it was necessary to consider the formation of different homogeneous layers to achieve a reasonable fitting of experimental data. The ellipsometric experimental values and the computationally simulated curves for the potentiostatic growths of MnO_x obtained at 0.70, 0.80, 0.90 and 1.00 V vs SCE are shown in Fig. 4. Clearly, modified electrodes are multilayered materials consisting of an inner passive layer (connoted as L_ox) and two subsequent layers of manganese oxide (L_in and L_out). Films grown at 0.70 and 0.75 V do not form the outer layer of MnO_x.

Zoom In Zoom Out Reset image size

Figure 5 shows the experimental parameters — refractive index, extinction coefficient and rate of growth — of both MnO_x layers, determined from the simulated Ψ vs Δ curves, as a function of the synthesis potential of MnO_x films (See Table S2). The increase in the applied potential leads to an increase of the refractive index and rate of growth of both layers, which present a similar evolution of such parameters (Fig. 5a). This suggests a dependency of the structural evolution of the outer layer with the inner one. The increasing film density with increasing synthesis potential can be explained by the formation of a higher number of nuclei with smaller sizes. The inner layer is more compact (higher n) and clearly more resistive (lower k) than the outer one, and presents a decreasing thickness with increasing applied potential (Fig. 5d). The outer layer increases with increasing potential of synthesis. The differences between the two layers, namely the extinction coefficient, are in agreement with the formation of two different layers.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The morphology of the MnO_x films grown potentiostatically at 0.8, 0.9, 1.0 and 1.1 V vs SCE was analyzed by scanning electron microscopy (Hitachi SU-70), as shown in Fig. 6. MnO_x films present a petal-shaped morphology. This structure, which depends on the synthesis conditions, ^11,17 presents reduced electrical resistance that can improve the double-layer capacitance and charge transfer processes, and has a porous architecture that can facilitate electrolyte diffusion into the film matrix. ¹⁷ The increase in the applied potential leads to a decrease in the size of nuclei and an increase in their number, which promotes the formation of films with an increasing available area of active material. An increase in the overpotential usually leads to an increase in nuclei density. ¹⁸

The electrochemical behavior of MnO_x films was studied by galvanostatic charge-discharge curves and cyclic voltammetry in a 1 M Na₂SO₄ electrolyte in a working potential window between 0.0 and 1.0 V vs SCE. The areal capacitance, Ca, was calculated from the galvanostatic discharge curves according to the equation:

$$\begin{eqnarray}&&{C}_{a}=\frac{i.{t}_{d}}{A.\unicode{x02206}E}\end{eqnarray} \tag{ 2 }$$

where i is the applied current, t_d is the measured discharge time, A is the geometric area of the film, and ΔE is the potential range during the discharge experiments. For comparative purposes, the specific capacitance was estimated using the Faraday equation, assuming that the electrodeposited films are exclusively constituted by MnO₂ and that all electrons were consumed for the formation of the film. It should be noted that these assumptions will lead to an overestimated mass. Films grown for 5, 7, 10, 20, 40, 70 and 100 mC cm⁻² will have 2.25, 3.16, 4,51, 9.02, 18.00, 22.5, 31.6 and 45.1 μg respectively.

The GCD curves obtained at 36 μΑ cm⁻² (2 A g⁻¹) of MnO_x films grown at different potential pulses, which are represented in Fig. 7a, show a nearly triangular shape with a low ohmic drop and present a Coulombic efficiency close to 100%. The evolution of the capacitance and ohmic drop with the synthesis potentials are represented in Figs. 7c and 7d, respectively. Increasing the synthesis potential from 0.7 to 1.0 V_SCE leads to an increase in the specific capacitance. For potentials greater than 1.0 V the capacitance remains constant. This evolution of the capacitance with the growth potential can be explained by the decrease of the inner MnO_x layer thickness and by the increase in the available area of the films grown at higher potential until the increase in the density starts to hinder the ionic diffusion through the film matrix. Electrodes modified at higher potentials (from 1.0 to 1.2 V) present good electrochemical performance, namely an areal capacitance of about 6 mF cm⁻² at a current density of 36 μΑ cm⁻² (about 330 F g⁻¹ at a specific current of 2 A g⁻¹). However, the increase in the potential pulse promotes the formation of more resistive films (Fig. 7d) with lower capacitance retention (Fig. 7b).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The cyclic voltammograms at different scan rates (from 10 to 500 mV s⁻¹) of the MnO_x film grown at 1.0 V vs SCE for 40 mC cm⁻², shown in Fig. 7e, present an almost rectangular shape, even at high scan rates, indicating a very good pseudocapacitive behavior and high reversibility during charge and discharge.

In terms of rate capability, all the electrodes showed excellent performance, presenting a small increase in the areal capacitance after 5000 GCD cycles at an applied current of 360 μΑ cm⁻² (20 A g⁻¹), as shown in Fig. 7f. The capacitance retention of the MnO_x films was not affected by the potential of growth.

Afterwards, the electrochemical response of MnOx films grown at 1.0 V for different charge of synthesis (5, 7, 10, 20, 50 and 100 mC cm⁻²) was studied. Figure 7h presents the areal capacitance as a function of the discharge current for the different modified electrodes. According to the results obtained by ellipsometry, the transition between the formation of the inner and the outer MnOx layer of the film grown at 1.0 V_SCE occurs around 10 and 15 mC cm⁻² of synthesis charge. The evolution of capacitance (Fig. 7i) and ohmic drop (Fig. 7j), at a current density of 40 μA cm⁻², with the charge of synthesis presents two different regions with the transition between those at around 10 and 15 mC cm⁻² of synthesis charge. Initially, the increase in capacitance is associated with the formation of MnOx crystallites, which being more conductive than the native oxide layer, lead to a decrease in the ohmic drop. From 20 mC cm⁻², the ohmic drop starts to increase and the specific capacitance reaches a maximum (inset of Fig. 7i), which can be explained by the increase in the film density during the transition between the formation of the first and second layer, due to preferential lateral growth of crystallites. Then, the film evolves with a petal-shaped morphology, as shown in SEM images, leading to the formation of the outer layer.

After electrodeposition, the modified electrodes were analyzed by electrochemical impedance spectroscopy in a 1 M Na₂SO₄ electrolyte.

The electrical equivalent model used to interpret the EIS data (shown in Fig. 8c) was based on the ellipsometric results assuming that MnO_x films are composed of two distinct layers: a more compact and resistive inner layer, and a porous and more conductive outer layer. The stainless-steel passive layer is considered to be part of the inner MnO_x layer as it is a very thin and compact film and is equally resistive. In this model, R_p,out and CPE_dl,out are the electrolyte resistance inside the pores and the capacitive behavior of the MnO_x outer layer, respectively, whereas R_ct,in represents the charge transfer resistance of redox processes within the pores and CPE_in is associated with the capacitive response of the inner layer. The Nyquist diagrams of the potentiostatically modified electrodes at different applied potentials (0.8, 0.9, 1.0, 1.1 and 1.2 V_SCE), for 40 mC cm⁻², and for different charges of synthesis (5, 7, 10, 20, 50 and 100 mC cm⁻²), at 1.0 V vs SCE, are represented in Figs. 8a and 8b, respectively.

Zoom In Zoom Out Reset image size

The impedance spectra present a depressed semi-circle at high frequencies that is followed by an almost vertical capacitive line at low frequencies. Tables I and II present the electrical parameters obtained by fitting of impedance spectra of different samples modified at different potentials and for different stages of growth, at 1.0 V vs SCE, respectively.

Figure 9 presents the fitting parameters of impedance data from manganese oxide films as a function of the growth potential, and as a function of the charge of synthesis. In the case of growth potential study, the increase in R_p,out (Fig. 9b) with the increase in the potential presents a similar trend with the variation of the ohmic drop (Fig. 7d) suggesting the formation of thicker and less porous films, which difficult the diffusion of electrolyte into the film bulk. The capacitance is directly proportional to the admittance determined for the constant phase element, according to the Brug equation. ¹⁹

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The capacitance of the double layer practically does not vary with the increase in the synthesis potential (Fig. 9a) because although the surface area increases, the pore resistance also increases.

As can be seen by the variation of Y_in, shown in Fig. 9c, which has a strong correlation with the evolution of capacitance obtained by the GCD tests (Fig. 7c), and taking into account that the capacity is inversely proportional to the layer thickness (C = εε₀ /l), the capacitive behavior of these materials is determined by the inner layer, which decreased in thickness with the increase in the growth potential, as observed in the in situ ellipsometry results, as well as a decrease in its resistivity, which is shown in Fig. 9d.

In the case of the growth study at 1.0 V vs SCE for different charges of synthesis, the double layer capacitance decreases at the beginning of the outer layer formation and after 20 mC cm⁻² becomes constant. The pore resistance decreases during the formation of the first nuclei, up to about 10 mC cm⁻², and then starts to increase as the outer layer thickness increases (Fig. 9f). Once again, the pore resistance variation showed a great similarity with the ohmic drop evolution with the charge of synthesis, obtained in GCD tests (Fig. 7j). The capacitance of the inner layer increases as the film evolves (Fig. 9g) and presents two regions. The resistance of the inner layer decreases sharply at the beginning until it converges to a minimum value, as seen in Fig. 9h, which may be due to the initial formation of a more conductive film than the oxide layer on the substrate and to the possible continuous conversion of the more resistive MnOOH into MnO₂. The capacitance evolution with the charge of synthesis also presents a trend similar to that of the capacitance determined from the discharge curves (Fig. 7i).

These results are in agreement with those obtained through the characterization by in situ ellipsometry and GCD analysis. The capacitance variation as a function of the synthesis potential follows the same trend as the capacitance calculated from the impedance data. The increase in capacitance may be due to an increase in the number of nuclei with smaller size and a decrease in the thickness of the inner layer of manganese oxide. The ohmic drop is related to the resistance in the pores of the outer layer of MnO_x. Its increase is explained by the increase in the density and thickness of the outer layer, which hinders the diffusion of ions through the film matrix.