Influence of the Substrate on the Performance of a Co-Prussian Blue Catalyst towards the Water Oxidation under Mild Conditions

Abstract

Efforts in developing water oxidation reaction (WOR) catalysts mainly focus on Earth-abundant elements with high stability and efficiency. However, the behavior of these materials may vary depending on the electrochemical substrate with which they interact. Understanding the interaction between the catalyst and the substrate can provide valuable insights to advance the design of highly efficient water-splitting devices. In this context, we investigated the influence of different electrochemical substrates on the performance of cobalt-Prussian blue analog (Co-PBA) in WOR under neutral conditions (pH = 7). The results highlight the role of substrate type, including conductive glass (fluorine-doped tin oxide (FTO)), 3D-printed electrodes (3DPE), and Ni-foam, on the electrocatalytic efficiency of Co-PBA. Two methods of Co-PBA synthesis (electrodeposition and drop casting) were employed to understand their effects on morphology, electrochemical properties, and performance in WOR. The results revealed that the substrate and preparation method of Co-PBA can affect the electrochemical behavior, demonstrating differences in electrocatalytic activity, stability, and charge transfer.

Keywords:
Prussian blue analogues; oxygen evolution reaction; water oxidation; green hydrogen; 3D printed electrodes; Ni-foam

Introduction

Global warming and its consequences are leaving the pages of the climate warning reports to be part of our daily lives.¹ Anthropogenic activity can be considered the most important factor in increasing global warming, especially when considering greenhouse gas (GHG) emissions, such as CO₂.² Given this scenario, we must seek sustainable alternatives to processes and economic activities that generate high volumes of CO₂. In this case, using green hydrogen (g-H₂) an as energy carrier or raw material for the chemical industry could be a viable alternative to mitigate greenhouse gas emissions.³

Hydrogen plays a fundamental role in our society, as it can be used as a chemical input in the petrochemical industry, in the production of ammonia (fertilizers), as an industrial reducing agent, and as a clean fuel. In addition to its unique chemical properties, the hydrogen’s combustion byproduct is water.³,⁴,⁵ However, the environmental benefits of the hydrogen depend on its method of production, as shown in our previous paper.⁵ Almost 95% of the hydrogen produced in the world comes from fossil fuels through the methane steam reforming process followed by the water-shift reaction, which produces grey hydrogen. Oil and coal can also be used to produce H₂. All these raw materials contribute to a large production of CO₂, that is, the more hydrogen is produced through these processes, the more GHG is released into the atmosphere. Therefore, the viable alternative to minimize CO₂ emissions is hydrogen production through the electrolysis process, also called water splitting. This process is considered environmentally friendly when H₂ is produced from water and uses renewable energy as an energetic input (i.e., solar or wind). In this case, the green hydrogen is obtained.⁵,⁶,⁷

A challenge for engineers and scientists is to develop a strategy to produce g-H2 at competitive prices since its production costs are over three times higher than those of H₂ obtained from fossil fuels.⁵,⁸,⁹ From a chemical perspective, the task presents a significant challenge, as water electrolysis, also known as water splitting, is a process that is thermodynamically unfavorable (ΔG = +237 kJ mol⁻¹) and kinetically sluggish. It is important to note that the water oxidation step represents the primary bottleneck in this reaction.¹⁰,¹¹,¹²

The motivation for studying water oxidation catalysts is to produce materials with good performance under low overpotential conditions, high stability, and robustness and use Earth-abundant metals. Among the most efficient catalysts in the literature for water oxidation at pH = 7, we can highlight the Prussian blue analogues containing cobalt, more specifically the Co₃[Co(CN)₆]₂.¹²,¹³,¹⁴ Cobalt is the coordination site for water and the catalytic active site for the water oxidation process to molecular oxygen production. Furthermore, these catalysts can have different behavior depending on the kind of electrochemical substrate. Thus, the properties of the set catalyst and substrate can bring additional information to support the development of highly efficient devices for hydrogen production. Given this motivation, in this work, we are investigating the influence of the electrochemical substrate (working electrode) in the performance of water oxidation by Co-Prussian blue (Co₃[Co(CN)₆]₂) obtained for two different ways (electrodeposition and drop casting) under mild conditions (pH = 7). The studied substrates were conductive glass, 3D-printed electrodes (3DPE), and Ni-foam. The first one is a conductive glass widely used in (photo)electrochemical experiments called FTO (fluorine-doped tin oxide (SnO2:F)). 3DPE were produced from Black Magic® filament composed of polylactic acid and graphene through the 3D printing process. Ni-foam is a kind of 3D electrode formed by open pore foam type.

Experimental

Chemicals

All chemicals: KCl, KNO₃, [Ru(NH₃)₆]Cl₃, urea, CoSO₄7H₂O, Co(NO₃)₂, K₃[Co(CN)₆], and the FTO are purchased from Sigma-Aldrich Brasil Ltda (Cotia-SP, Brazil) and utilized without previously purification. Sulfuric and hydrochloric acid were acquired from Synth, Diadema-SP, Brazil. Distilled water in reverse osmosis filter Milli-Q with resistivity at 25 °C 18.5 MΩ cm was utilized in all aqueous solutions preparation.

Equipment

All electrochemical experiments were performed in an Autolab PGSTAT12 potentiostat/galvanostat (Metrohm, Netherlands). The reference electrode was a Metrohm Ag/AgCl with 3.5 mol L⁻¹ KCl electrolyte solution and the counter-electrode was a large surface area platinum wire. The microscopy images and energy dispersive X-ray spectroscopy (EDS) were performed in a JEOL JMS-6340F with (JEOL Ltd., Tokyo, Japan) operating with an accelerating voltage of 15 kV. Infrared spectra were obtained in an Agilent CARY 630 FTIR spectrophotometer by attenuated total reflectance (ATR) mode with a Ge crystal in the range of4000-400 cm⁻¹ with a resolution of 4 cm⁻¹ (Agilent Technologies, Inc, USA). For this, the surface layer of the CoM-PBA and CoH-PBA film formed on the FTO electrode was scraped. The X-ray powder diffraction patterns were recorded using a Shimadzu X-ray diffractometer (Cu Ka = 1.5418 Å radiation) in a 26 range of 5-70° at scan rate of 2.00° min⁻¹ and step size of 0.02° (Tokyo, Japan). The Raman analysis was performed on the electrode surface on a confocal Horiba Jobin Yvon T64000 Raman (Horiba Jobin Yvon, France) confocal spectrometer using 532 nm laser and ×50 and ×100 lens. The Raman mappings were made following the cyanide band in a range of 2100-2200 cm⁻¹ in an area of 15 μn² of all electrode surface. The mapping acquisition was done with incidence on each of the 225 points for 15 s with a 532 nm laser at 7.5 mW power.

Preparation of electrodes

FTO

FTO is a conductive glass formed by fluorine-doped tin oxide (SnO₂:F). The FTO electrodes were adjusted to 1.0 cm × 2.5 cm and were washed in a process analogous to that described by Pires et al.¹⁵ and heated at 400 °C for 30 min. An area of 1.0 cm² was isolated on the conductive face of the FTO before the modification.

3D printed electrode (3D)

The printed electrodes were treated with an aqueous hydrazine solution (10 μL/30 mL of water). The system was kept under heating at 65 °C for 2 h. After this period, the electrodes were washed with distilled water and allowed to dry in the environment. The dry electrodes were electrochemically activated in 0.1 mol L⁻¹ of HCl/KCl solution applying 1.8 V for 900 s followed by applying −1.8 V for 90 s.

Nickel foam (Ni-foam)

1.0 cm × 2.0 cm electrodes were etched by immersion in a 3.5 mol L⁻¹ HCl solution for 30 min to remove oxide layers. Afterward, the electrodes were subjected to a sonic bath in ethanol for 15 min and dried at room temperature before use.

Preparation of cobalt Prussian blue analogue (Co-PBA)

The FTO, 3DPE and Ni-foam were used as working electrodes. The formation of Co-PBA on the surface of these electrodes was carried out in two different ways.

Method 1-electrodeposition (CoM-PBA)

The working electrodes were immersed in a 0.1 mol L⁻¹ solution of cobalt sulfate acidified with sulfuric acid to obtain pH 4 in an electrochemical cell with the electrodes arranged in a three-electrode system. Then, −1.6 V vs. Ag/AgCl (3.5 mol L⁻¹ KCl) was applied for variable periods for each electrode to form a metallic film. After that, K₃[Co(CN)₆] was added to the solution and the film was oxidized by applying +0.5 V vs. Ag/AgCl (3.5 mol L⁻¹ KCl) to form the Co-PBA as presented in Table 1. The electrodes modified with Co-PBA were washed with distilled water and dried in an oven at 80 °C for 2 h to stabilize the film.

Table 1.
Optimized parameters for potentiostatic Co-PBA film formation

Electrode	Reduction (−1.8 V) time / s	Oxidation (+0.5 V) time / s
FTO	40	200
3DPE	1200	2700
Ni-foam	60	120

• FTO: fluorine-doped tin oxide; 3DPE: 3D-printed electrodes.

Method 2-drop casting (CoH-PBA)

78 mg of Co(NO₃)₂ and 81 mg of urea were solubilized in 15 mL of deionized water and subjected to a hydrothermal reaction system for 12 h at 120 °C to obtain Co(OH)₂. Aliquots of 50 pL of a suspension of Co(OH)₂ (5 mg Co(OH)₂ + 120 μL distilled water + 360 μL of dimethylformamide + 20 μL Nation®) were dripped onto the surface of the electrodes leaving them dry for 12 h. The electrodes were then immersed in a sulfuric acid solution (pH = 4) with K₃[Co(CN)₆] for 2 h, washed with distilled water, and left to dry at room temperature.

Electrochemical features

Electroactive surface area (ECSA) and heterogeneous rate constant (k_obs) determination by cyclic voltammetry

The determination of the electroactive area and the electronic transfer constant were performed using a solution of 1 mmol L⁻¹ [Ru(NH₃)₆]Cl₃ containing 0.1 mol L⁻¹ KCl as support electrolyte. Cyclic voltammograms were recorded at potentials from −0.6 to 0.2 V vs. 3.5 mol L⁻¹ Ag/AgCl reference electrode at scanning speeds ranging from 500 to 5 mV s⁻¹. The solutions were deoxygenated by purging them with nitrogen gas for 20 min previously the experiments.

To determine the ECSA, the linear relation between the peak current (anodic or cathodic) and the scan rate expressed by the Randles-Sevcik equation (equation 1) to reversible system is:

(1) ${\text{i}}_{\text{p}}=0.4463{\text{A}}_{\text{e}}\text{C}{\left(\text{nF}\right)}^{\frac{3}{2}}{\left(\frac{\text{vD}}{\text{RT}}\right)}^{\frac{1}{2}}$

where i_p is the peak current, n is the number of electrons involved in the reaction, C is the concentration and D is the diffusion coefficient (for [Ru(NH₃)₆]³⁺, C = 9.10 x 10⁻⁶ cm² s⁻¹), F is the Faraday constant (98485.3 C mol⁻¹), R is the gas constant (8.314 J K⁻¹mol⁻¹), T is the temperature (298.15 K), v is the scan rate (mV s⁻¹) and A_e is the electroactive surface area.

The heterogeneous rate constant values were obtained from a modification of Nicolson method¹⁶ by the following equation (equation 2):

(2) ${\text{k}}_{\text{obs}}=\left[2.18{\left(\frac{\text{DαnFη}}{\text{RT}}\right)}^{\frac{1}{2}}\right]\exp \left[-\left(\frac{{\text{α}}^2\text{nF}}{\text{RT}}\right)\Delta {\text{E}}_{\text{P}}\right]$

where k_obs is the heterogeneous rate constant, D is the diffusion coefficient, a is the asymmetry barrier factor (assumed to correspond to 0.5), n is the number of transferred electrons, F is the Faraday constant, R is the gas constant, T is the temperature and the ΔE_p is the peak-to-peak separation.

Water oxidation experiments

Experiments of electrocatalytic water oxidation were performed using in a 0.5 mol L⁻¹ of KNO₃ aqueous solution and adjusted to pH = 7 as supporting electrolyte at room temperature and bubbled with nitrogen for 20 min to deoxygenate the solution before the measurements. The linear sweep voltammetry (LSV) and cyclic voltammetry experiments are performed in a range of 0 to 2.04 V vs. Ag/AgCl with a scan rate set at 5 mV s⁻¹. The iR drop correction was applied to all data before analysis using a feedback positive method to compensate the solution resistance. The chronoamperometric experiments were conducted in the respective potential applied for water oxidation previously extracted in the LSVs measurements.

Electrochemical impedance spectroscopy (EIS)

Electrochemical impedance spectroscopy experiments were performed in a [Ru(NH₃)₆]³⁺ 10 mmol L⁻¹ containing 0.1 mol L⁻¹ KNO₃ as support electrolyte. A constant potential of +0.3 V vs. Ag/AgCl are applied in range of frequencies from 1.0 × 10⁵ to 0.1 Hz. All solutions were deoxygenated by purging them with nitrogen gas for 20 min before the experiments. The capacitance of the double layer (C_DL) of the working electrodes and the solution resistance (R_S) measurements were carried out by EIS.

Results and Discussion

Cobalt Prussian blue analog (Co-PBA) films were deposited on the surface of the FTO electrode by two different pathways as described before where CoM-PBA was prepared by electrodeposition and CoH-PBA by leaching of a cobalt hydroxide film in an acidified solution of hexacyanocobaltate(III). The obtained materials were characterized using IR spectroscopy and X-ray diffraction, to confirm the composition and structure. In order to obtain information about the functional groups present in the Co-PBA films, infrared spectra by ATR were obtained from both modified FTO electrodes. Such spectra are shown in Figure 1.

Figure 1.
Infrared spectra (ATR) for cobalt Prussian blue analog obtained starting from cobalt metallic film (CoM-PBA) and cobalt hydroxide film (CoH-PBA). Spectra were obtained over the FTO electrodes by ATR method with Ge crystal.

From the infrared spectra, it is possible to visualize some similarities between the format and position of the bands of both spectra, which indicates that both films have similar functional groups. The highest intensity band in the spectra is found at 2172 and 2168 cm⁻¹ and can be assigned to the stretching (v) of the (–C≡N) group. A small band can be observed at 3637 cm⁻¹ and it is associated with the stretching of occluded water molecules within the porous structure of the Prussian blue like a zeolite structure.¹³,¹⁷ The bands at 3400 and 3396 cm⁻¹ are assigned with v(O-H) is related to hydrogen bonding from water molecules adsorbed on the catalysts. In addition, the band at 1614 cm⁻¹ can be attributed as δ(H–O–H). The band present at 490 cm⁻¹ is reported as the v(Co–CN–Co).¹³,¹⁸ In the CoM-PBA spectra, a different band appears at 1350 cm⁻¹. This band can be assigned with a v_as(S=O) of sulfate ion reminiscent from the CoSO₄ used in the electrodeposition of the Co metallic film.¹⁹ Since both spectra show similarly shaped bands in characteristic regions of Prussian blue analogs compounds, spectroscopy shows evidence of the effectiveness of both surface modification methods of working electrodes with Co-PBA films.

To obtain relevant information about the structure of both CoM-PBA and CoH-PBA, X-ray powder diffraction was performed. The diffractogram (Figure 2) showed that the Co₃[Co(CN)₆]₂ is iso-structural to the Prussian blue crystal structure with space group Fm3m adopting a face-centred cubic (FCC) form. The main peaks that can be observed in the diffractogram for both CoM-PBA and CoH-PBA are associated with the crystalline planes (200), (220), (400), (420) and (422), revealing a single-phase of FCC structure being another strong indication that both methods of surface modification of the electrode result in the expected Co-PBA film.¹³

Figure 2.
X-ray powder diffractogram for cobalt Prussian blue analogue obtained starting from cobalt metallic film (CoM-PBA) and cobalt hydroxide film (CoH-PBA).

The morphology of the catalysts deposited over the surface of electrodes (Figure 3) was analyzed by scanning electron microscope (SEM) and energy-dispersive X-ray spectroscopy (EDS) to confirm the composition and dispersion of the materials on the conductive substrate. Based on these images, it is possible to observe the formation of a material with irregular morphology, showing cubes with different sizes clustered on the surface of the electrodes. The irregularity in the formation of Co-PBA particles can be attributed to the high concentrations of the complex anion ([Co(CN)₆]³⁻) used to obtain the films. When releasing Co²⁺ ions from the electrode surface, either by oxidation of the metallic film or by leaching of the hydroxide film, this instantly coordinates with the complex anion forming regions of agglomeration, which hinders the slow and orderly growth of the crystals.²⁰

Figure 3.
SEM images and EDS analysis for CoM-PB and CoH-PB in different modified electrodes.

The correlation between particle size and film formation time is easily explained if we compare the methods of obtaining. In the case of CoM-PBA films that require the application of a potential to promote the oxidation of the cobalt film, the particles formed on the surface of the 3D electrode are considerably larger compared to FTO and nickel foam electrodes since it is less conductive, it takes longer to oxidize the mentioned film. On the other hand, for CoH-PBA films that depend almost exclusively on the diffusion of anions in solution, the particle size does not vary as much according to the electrode. According to the EDS, it appears that the material has cobalt, carbon and nitrogen in its composition, which indicates that the structure is the Co-PBA for both electrodes whose substrate was the metallic film, and the substrate was the cobalt hydroxide.

In addition to the microscopy analysis, a surface mapping by Raman spectroscopy was performed on the modified electrodes to understand how the material is distributed, as shown in Figure 4.

Figure 4.
Mapping by Raman spectroscopy of the different modified electrodes (FTO, 3D is the 3DPE and Ni-foam) surface with CoHPBA and CoM-PBA. The scale colors represent the v(CN) band intensity.

According to these results, it is possible to observe some patterns that indicate the CoM-PB and CoH-PB are distributed over the electrodes. The growth of the Co-PBA film, in both modification pathways, now occurs from “islets” that spread three-dimensionally from these points, which generates sections of the electrode with a higher concentration of the Prussian blue analog represented by the red regions of the image (regions of greater intensity of the cyanide band).

For the FTO electrode, there is a uniform distribution over its entire surface, with a few regions with no filling by the catalyst. Since the surface of the FTO is flat and, when the surface is modified, the nucleation points are better distributed over the entire surface since the diffusion of anions [Co(CN)₆]³⁻ coming from the bulk and anchor on the electrode occurs in parallel and thus generates a surface completely covered with Co-PB regardless of the method of obtaining used. On the other hand, the 3D electrodes do not allow such a homogeneous distribution of the catalyst on their surface. For the case of CoM-PBA as the first stage involves the potentiostatic deposition of a metallic film, the electrode surface may have become uniform by the formation of the Co⁰ film, and during the modification of this to Co-PBA only the outermost layer of Co⁰ can be oxidized, and generate Co₃[Co(CN)₆]₂ while the innermost layers can remain filling the electrode grooves. As for the case of CoH-PBA whose first stage involves the dripping of cobalt hydroxide on the electrode surface, there is a tendency for the solution to drain to fill inhomogeneities such as grooves, immediately during the conversion of Co(OH)₂ to Co₃[Co(CN)₆]₂ these regions will present a greater accumulation of material and therefore, greater intensity of the v(CN) band, while sections of the electrode whose relief is higher do not fix a large amount of material presenting a low signal intensity which causes a more heterogeneous distribution in this system.

In the case of the Ni-foam electrode, for both methods, the material is distributed locally on the electrode surface. Such observation occurs because the material is deposited on the metallic structure of the foam leaving the pores free from the presence of the material. It is also possible to observe that there is no significant difference between CoM-PB and CoH-PB about the electrode coating mode, with both points showing the high signal intensity of the cyanide band with subsequent attenuation of the signal in the region radial to these points which can indicate that the formation occurs from a nucleus present on the surface of the electrodes as in the other electrodes.

Electrochemical characterization

The influence of the electrochemical features, according to the type of electrode modified with Co-PBA, was evaluated through the measurement of the electron transfer constant values. In these experiments, the cyclic voltammetry is performed at different scan rates using [Ru(NH₃)₆]³⁺ as an electrochemical probe.

Cyclic voltammograms are presented in Figure 5 and give us an idea of the reversibility of the one-electron redox process for both Co-PBA precursors in each electrode substrate. In this system, we consider that the mass transport and ohmic drop polarizations are small enough that the only effective contribution is the activation polarization.

Figure 5.
Cyclic voltammetry in [Ru(NH₃)₆]³⁺/KCl (1 mmol L⁻¹ 70.1 mol L⁻¹) for modified electrodes with cobalt Prussian blue based in metallic film (CoM-PBA) and hydroxide film (CoH-PBA).

Based on the results presented in Figure 5, it is possible to observe some characteristics in the shape of cyclic voltammograms. For the 3D and Ni-foam electrodes modified with Co-PBA by both methods, there is a separation between the cathode and anode peaks greater than 59 mV, increasing with the increase of the scanning speed, which characterizes a quasi-reversible process. As for the FTO electrode, regardless of the obtained method, the peak-to-peak separation is less than 59 mV and does not depend on the scan rate applied to the system, indicating to be a reversible process.

Another interesting observation goes through the definition of the peaks concerning the film deposited on the surface of the electrodes. The electrodes containing CoM-PBA film show broader peaks of the redox processes compared to cyclic voltammograms for the electrodes modified with the CoH-PBA film. An explanation for this characteristic can be related to the morphological profile of the constituent particles of the Co-PBA films. Smaller and less irregular particles, as in the case of CoH-PBA film, generate microporous cavities that hamper electrolyte access, thus reducing the interfacial area with the electrolyte, while for CoM-PBA with larger and more irregular particles, they generate regions that increase the area in contact with the solution and in this way increases the capacitive profile of the Prussian blue analog film.²¹ In addition, the voltammograms expressed in Figure 5 were used together with equations 1 and 2 to obtain the electrochemically active surface area (ECSA) and electron transfer constant (k_obs) parameters presented in Table 2.

Table 2.
Electrochemical parameters obtained from different modified electrodes with CoH-PBA and CoM-PBA films

Catalyst film	Electrode	ηonset / mV	η / (10 mA cm−2)	Tafel / (mV dec−1)	ECSA / cm2	kobs / (10−3 cm s−1)
CoH-PBA	FTO	320	559	93	0.059	4.87
	3DPE	371	698	281	0.039	1.40
	Ni-foam	278	357	177	0.412	1.95
CoM-PBA	FTO	398	798	124	0.348	3.13
	3DPE	412	425	175	0.031	2.14
	Ni-foam	407	711	192	0.342	4.81

• η: overpotential; ECSA: electrochemically active surface area; k_obs: electron transfer constant; FTO: fluorine-doped tin oxide; 3DPE: 3D-printed electrodes.

In addition, cyclic voltammetry experiments in a potential range of water oxidation were made to evaluate the electrocatalytic performance of the electrodes and materials. Figure 6 shows the voltammograms obtained for water oxidation in 0.5 mol L⁻¹ KNO3 solution (pH = 7). For all modified electrodes, there is a significant increase in the faradaic current density of the voltammetry curve when compared with the bare electrode. This fact demonstrates that the electrode modification with Co₃[Co(CN)₆]₂ catalyst enhances the water oxidation reaction under mild conditions

Figure 6.
Cyclic voltammetry of modified electrodes (Ni-foam, 3DPE and FTO) with Co3[Co(CN)₆]₂ performed in 0.5 mol L⁻¹ KNO3 at 5 mV s⁻¹. Co₃[Co(CN)₆]₂ was produced by the follow methods: (a) cobalt metallic film (CoM-PBA) and (b) cobalt hydroxide (CoH-PB). The curves were normalized using ECSA.

For FTO and 3DPE electrodes modified with the catalyst, the increase of faradaic current is often higher compared to the bare electrode, since these electrodes alone have low conductivity and surface area, and the modification with Co-PBA over their surface promotes the electroactivity toward the water oxidation reaction. On the other hand, for the Ni-foam electrode, the increase is less significant, generating an increase of about 3 times when compared with the bare electrode again. This observation can be justified due to the high surface area and conductivity of Ni-foam, presenting catalytic properties alone. Nevertheless, the modification of the porous electrode surface still significantly enhances the catalytic current in the water oxidation process. A direct comparison between the effectiveness of the two methods of obtaining the PBA can be seen in Figure 6. A comparison of the effectiveness of the two methods of obtaining the Co-PBA and their performances in different substrates can be found in the LSV studies presented in Figure S1 (Supplementary Information (SI) section).

The LSV curves show that the 3D and Ni-foam electrodes modified present a capacitive current contribution in intermediate potential regions represented by the current value ranging from 1.2 to 1.7 V vs. reversible hydrogen electrode (RHE). However, this fact was not observed for electrodes containing CoH-PBA film. CoM-PBA films in Ni-foam and 3DPE could have a high initial oxidation potential due to using Nafion in the ink applied to modify the electrode. Furthermore, in these electrodes, the overpotential for starting the catalytic process is lower than in comparison with the same electrodes modified with CoH-PBA film, which indicates that the speed with which the transfer of electrons from the water molecules adsorbed to the catalytic sites occurs is faster. On the other hand, in the FTO electrodes, there are no significant differences in the curves regarding capacitive components of current and overpotential in the starter of the water oxidation reaction. In this case, the similar shape of the curves most likely occurs because the interfacial areas of the films in contact with the electrolyte are smaller than the others, so the accumulation of charges in the double layer is reduced, making the current less significant.

The heterogeneous rate constant presented in the Table 2 showed us the electronic transfer rate between electrode/solution changes according to the conductivity of the electrodes, as well as to the methodology used to obtain the applied films. More conductive electrodes like Ni-foam and FTO show the highest values while the 3D electrode shows the lowest values, which reinforces the choice of an adequate substrate.

Regarding the preparation method, the more uniform distribution of CoM-PBA films over the electrode’s surface results in a more covered area of the substrate’s surface, improving the electronic transfer. Tafel slope values (presented in Table 2) were calculated according to the procedure described in the literature.¹² The Tafel slope gives us an indication of how fast is the kinetics of the evaluated electrocatalyst. In general, there are no significant differences in the kinetics of the materials formed from cobalt metal film or cobalt hydroxide film, except for the 3D/CoH-PBA electrode that has the highest Tafel slope value, which implies the lowest kinetic in comparison to the other electrodes.

Electrochemical impedance spectroscopy brings relevant information about the electrochemical processes involved in the surface of the modified electrodes. A complete discussion can be found in the SI section. An important information brought for this technique is the charge transfer resistance which is a measure of the difficulty encountered in transferring electrons from the electrode to the substrate. Thus, it is possible to observe in Table 3 that the deposition of Co-PBA on 3DPE increases the R_CT value which could be associated with the blocking of charge transfer. Printed electrodes are sensitive to the modification of the surface, because of this they request an activation of the surface before the measurements. Ni-foam electrodes have not the charge transfer resistance (R_CT) changed by the modification with Co-PBA. On the other hand, the FTO had the RCT decreased by the Co-PB modification. This analysis is not trivial because different equivalent circuits were used for the electrocatalysts, but it can show the tendency of the values. All data from EIS are shown in Table 3.

Table 3.
EIS parameters for all available electrodes in the Faradaic region

Electrode	Surface	Rs / Ω	RCT / Ω	CDL / F
3DPE	bare	100	31.5	416.9 × 10−6
	CoH-PBA	150	862.0	2.7 × 10−6
	CoM-PBA	313	710.0	0.5 × 10−6
FTO	bare	30.5	93.7	0.6 × 10−6
	CoH-PBA	16.4	8.2	22.8 × 10−6
	CoM-PBA	19.6	9.2	11.8 × 10−6
Ni-foam	bare	5.2	29.7	7.6 × 10−3
	CoH-PBA	5.8	31.0	0.3 × 10−3
	CoM-PBA	2.9	17.5	2.0 × 10−3

• R_s: solution resistance; R_CT: charge transfer resistance; C_DL: double layer capacitance; FTO: fluorine-doped tin oxide; 3DPE: 3D-printed electrodes.

The stability study of the catalyst in water oxidation reaction has great importance in elucidating the viability and the time of operation and is particularly important to guide future developments to improve the performance of stable catalysts. To evaluate the stability of modified electrodes, a chronoamperometric experiment was made applying the potential necessary to drive a current of 10 mA as represented in Figure 7.

Figure 7.
Chronoamperometric stability examination of the catalysts during the oxygen evolution reaction (OER) process immobilized in (a) 3D printed electrode, (b) FTO electrode and (c) Ni-foam electrode.

As seen in Figure 7, both the 3D electrode and FTO modified with CoM-PBA exhibit a similar behavior, where the increase in the time scale leads to a decrease in the current, indicating degradation of the material during the electrolysis process. For both modified Ni-foam electrode and FTO/CoH-PBA modified electrode, variations in current at early time indicate instability of the material due to conversion in another species in conditions of water oxidation. A general briefing of this experiment is shown in Table 4.

Table 4.
Stability current and percentage of current loss for both 3D and FTO modified electrode

Electrode	Current (t = 600 s) / A	Current (t = 7200 s) / A	Lost current /%
3D/CoH-PBA	0.00120	0.0009	−18.30
3D/CoM-PBA	0.0024	0.0016	−33.30
FTO/CoH-PBA	0.0022	0.0007	−68.50
FTO/CoM-PBA	0.0015	0.0011	−26.70
Ni-foam/CoH-PBA	0.030	0.041	+36.7
Ni-foam/CoM-PBA	0.030	0.02	−33.3

Table 4 shows that materials immobilized in 3D electrodes are more stable in comparison with the materials immobilized in FTO electrodes, and show a less percentual of loss current in 2 h. We suppose that the presence of oxygenated groups, on the 3D electrode, coordinates with cobalt sites, resulting in particles of Co-PBA more strongly adhered to the substrate. The materials immobilized in Ni-foam electrodes do not show long-term stability. The instability can be related to the continuous oxidation of the metallic nickel in Ni²⁺ with the formation of nickel hydroxide in the bulk solution when the potential of 10 mA is applied in the system, as demonstrated by the presence of the Gerischer element in the impedance experiments.

Conclusions

As a conclusion of this work, it can be inferred that the methodology applied to obtain the cobalt Prussian blue analog over different substrate surfaces has an important role in the material properties as a catalyst for water oxidation under mild conditions. In this context, the formation of agglomerates of Co with different sizes, as obtained by the potentiostatic method, leads to a more capacitive electrochemical behavior and is susceptible to diffusion effects, which is not as observable for the films obtained by the drop-casting method. The growth of Co-PBA crystals in different substrates was predominantly governed by the geometry and conductivity of the substrate’s surfaces, as evidenced in the Raman mapping. According to the substrate’s conductivity, it may result in different contact resistances in the interface substrate/electrocatalyst, which directly affects the evaluated electrochemical properties and activity. In addition, the nature of the substrate was a major factor in the stability of the catalyst against the action of the water oxidation reaction under mild conditions (pH = 7). In this way, the results obtained in this work reinforce to take into consideration of the choice of adequate substrates, to truly access the electrochemical parameters and activity of the material under investigation.

Supplementary Information

Supplementary information (Figures S1-S5) is available free of charge at http://jbcs.sbq.org.br as PDFfile.

Acknowledgments

The authors are grateful for the financial support of Brazilian Funding Agencies. This work was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001, Conselho Nacional de Desenvolvimento Científico e Tecnológico CNPq (grant No. (308203/2021-6) and Fundação de Amparo à Pesquisa do Estado de São Paulo, FAPESP (grant No. 2021/05976-2 and grant No. 2017/23960-0).

References

• ¹
  Cho, H. H.; Strezov, V; Evans, T. J.; Sustainable Mater. Technol. 2023, 35, e00567. [Crossref]
  » Crossref
• ²
  Rehman, A.; Alam, M. M.; Ozturk, I.; Alvarado, R.; Murshed, M.; Içik, C.; Ma, H.; Environ. Sci. Pollut. Res. 2023, 30, 9699. [Crossref]
  » Crossref
• ³
  von Zuben, T. W.; Moreira, D. E. B.; Germscheidt, R. L.; Yoshimura, R. G.; Dorretto, D. S.; de Araujo, A. B. S.; Salles Jr., A. G.; Bonacin, J. A.; J. Braz.. Chem. Soc. 2022, 33, 824. [Crossref]
  » Crossref
• ⁴
  Guarieiro, L. L. N.; dos Anjos, J. P.; da Silva, L. A.; Santos, A. A. B.; Calixto, E. E. S.; Pessoa, F. L. P.; de Almeida, J. L. G.; Andrade Filho, M.; Marinho, F. S.; da Rocha, G. O.; de Andrade, J. B.; J. Braz.. Chem. Soc. 2022, 33, 844. [Crossref]
  » Crossref
• ⁵
  Germscheidt, R. L.; Moreira, D. E. B.; Yoshimura, R. G.; Gasbarro, N. P.; Datti, E.; dos Santos, P. L.; Bonacin, J. A.; Adv. Energy Sustainability Res. 2021, 2, 2100093. [Crossref]
  » Crossref
• ⁶
  Hassan, N. S.; Jalil, A. A.; Rajendran, S.; Khusnun, N. F.; Bahari, M. B.; Johari, A.; Kamaruddin, M. J.; Ismail, M.; Int. J. Hydrogen Energy 2024, 52, 420. [Crossref]
  » Crossref
• ⁷
  El-Shafie, M.; Results Eng. 2023, 20, 101426. [Crossref]
  » Crossref
• ⁸
  Moritz, M.; Schõnfisch, M.; Schulte, S.; Int. J. Hydrogen Energy 2023, 48, 9139. [Crossref]
  » Crossref
• ⁹
  Franzmann, D.; Heinrichs, H.; Lippkau, F.; Addanki, T.; Winkler, C.; Buchenberg, P.; Hamacher, T.; Blesl, M.; LinBen, J.; Stolten, D.; Int. J. Hydrogen Energy 2023, 48, 33062. [Crossref]
  » Crossref
• ¹⁰
  Linnemann, J.; Kanokkanchana, K.; Tschulik, K.; ACS Catal. 2021, 11, 5318. [Crossref]
  » Crossref
• ¹¹
  Millet, P. In Hydrogen Production; Godula-Jopek, A., ed.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2015, p. 33-62. [Crossref]
  » Crossref
• ¹²
  Zambiazi, P. J.; de Moraes, A. T. N.; Kogachi, R. R.; Aparecido, G. O.; Formiga, A. L. B.; Bonacin, J. A.; J. Braz. Chem. Soc. 2020, 31, 2307. [Crossref]
  » Crossref
• ¹³
  Alsaç, E. P.; Ülker, E.; Nune, S. V. K.; Dede, Y.; Karadas, F.; Chem. - Eur. J. 2018, 24, 4856. [Crossref]
  » Crossref
• ¹⁴
  Azhar, A.; Li, Y.; Cai, Z.; Zakaria, M. B.; Masud, M. K.; Hossain, M. S. A.; Kim, J.; Zhang, W.; Na, J.; Yamauchi, Y.; Hu, M.; Bull. Chem. Soc. Jpn. 2019, 92, 875. [Crossref]
  » Crossref
• ¹⁵
  Pires, B. M.; dos Santos, P. L.; Katic, V.; Strohauer, S.; Landers, R.; B. Formiga, A. L.; Bonacin, J. A.; Dalton Trans. 2019, 48, 4811. [Crossref]
  » Crossref
• ¹⁶
  Bonacin, J. A.; dos Santos, P. L.; Katic, V.; Foster, C. W.; Banks, C. E.; Electroanalysis 2018, 30, 170. [Crossref]
  » Crossref
• ¹⁷
  Netskina, O. V.; Pochtar, A. A.; Komova, O. V.; Simagina, V. I.; Catalysts 2020, 10, 201. [Crossref]
  » Crossref
• ¹⁸
  Kettle, S. F. A.; Diana, E.; Marchese, E. M. C.; Boccaleri, E.; Croce, G.; Sheng, T.; Stanghellini, P. L.; Eur. J. Inorg. Chem. 2010, 3920. [Crossref]
  » Crossref
• ¹⁹
  Pavia, D. L. In Introduction to Spectroscopy, 4^th ed.; International Student Ed.; Brooks-Cole: Belmont, USA, 2009.
• ²⁰
  Yoshimura, R. G.; Ferraz, T. V. B.; J. Zambiazi, P.; Bonacin, J. A.; Energy Adv. 2024, 3, 495. [Crossref]
  » Crossref
• ²¹
  Frackowiak, E.; J. Braz. Chem. Soc. 2006, 17, 1074. [Crossref]
  » Crossref

Edited by

Editor handled this article: André Galembeck (Guest)

Publication Dates

• Publication in this collection
  22 July 2024
• Date of issue
  2024

History

• Received
  09 Feb 2024
• Accepted
  28 May 2024