Fabrication of Mn–Co Alloys Electrodeposited on AISI 430 Ferritic Stainless Steel for SOFC Interconnect Applications

Abstract

Mn–Co alloys were electroplated on AISI 430 stainless steel using an electrodeposition technique with the aim to reduce oxidation and chromium volatilization. The electroplating parameters were designed to improve the coating quality. The increased current density with decreased MnSO₄ content resulted in a denser coating layer. A sample coated with 0.10 M CoSO₄ and 0.50 MnSO₄ at 350 mA cm⁻² showed the best oxidation resistance after being oxidized at 800 °C for 90 h. The X-ray diffraction (XRD) result revealed that the oxide growth on the surface of the coated samples mainly formed oxides of MnCo₂O₄, MnCr₂O₄, and Cr₂O₃. The chromium volatilization was evaluated by exposing the coated samples to humidified synthetic air at 800 °C for 96 h. The mass flux of Cr volatilization was on the order of 10⁻¹¹ g cm⁻² s⁻¹. Furthermore, different heat treatments in O₂ and CO₂ atmospheres were compared. Annealing in CO₂ at 800 °C for 4 h helped increase the Mn–Co coating density. The relationship between the porosity and its failure behavior was also discussed.

1. Introduction

A fuel cell is a clean power generator that directly and electrochemically converts chemical energy into electrical energy [1]. Solid oxide fuel cells (SOFCs) are considered a renewable energy source [2], and they have many advantages for a wide range of applications because of their high energy conversion efficiency, low greenhouse gas emissions, eco−friendliness, various inlet fuel choices, and long lifetime potential [3,4]. SOFCs comprise two porous electrodes (anode and cathode), separated by a dense layer of solid electrolyte, and an interconnect at each electrode to conduct electricity. To achieve higher power output, multiple SOFCs are heaped as repeated units and called an SOFC stack. An interconnect is a main component that is located between two adjacent repeated units, providing the electrical connection between the SOFCs, and simultaneously transports fuel gases (anode side) and air (cathode side). Ferritic stainless steels are promising materials to use for interconnects, instead of the traditional use of ceramics for interconnects, due to the cost–effectiveness and matching of the thermal expansion coefficient, which is compatible with the electrode materials of the SOFC in the operating temperature range of 650–800 °C [5,6,7,8]. Under this high operating temperature, the exposure of ferritic stainless steels in the cathode and anode atmospheres results in the formation of a chromium oxide (Cr₂O₃) layer at the surface. The Cr₂O₃ can then be further oxidized to form volatile Cr species (e.g., CrO₃ and CrO₂(OH)₂) over the oxide scale [9,10,11,12]. The volatile phases are induced by contamination at the cathode, known as Cr poisoning, which degrades SOFC performance [13,14,15,16]. Moreover, continuous exposure to oxidizing atmospheres eventually leads to the formation of a thick oxide scale. The formation and growth of Cr₂O₃ contribute to an increase in electrical resistance and spallation, which reduce the lifetime of SOFC operation [17].

Protective coatings have been considered to minimize the aforementioned problems, particularly for issues related to oxidation and Cr volatilization. Numerous studies reported that Mn–Co oxide spinel has high electrical conductivity, good oxidation resistance, and low chromium volatilization [18,19,20,21,22,23,24]. Various deposition methods have been proposed to apply Mn–Co oxide spinel on the surface of metallic interconnects, such as slurry coating [18,25], sol–gel [26,27,28], plasma spray [29,30,31], electrophoretic deposition [32,33,34], and electrodeposition [10,11,35,36,37,38,39,40]. Among the deposition methods above, the electrodeposition method is attractive for its economy and feasibility. It allows homogeneous coating over a large area and complex−shaped structures. Electrodeposition is typically used in cathodic deposition processes. Metal ions can be deposited on the cathode by the reduction of metal ions in the electrolyte solution. To form an alloy coating, a small difference of the standard reduction potential is required for multiple−element electrodeposition. If the reduction potentials of multiple elements are significantly different, a chelating agent is often introduced to adjust the reduction potentials of more noble metal to be more negative [36]. S.S. Abd El Rehim et al. [41] suggested that an acidic gluconate solution can be used as a chelating agent for co−electrodeposited Mn–Co alloys because of its environmentally friendly electrolyte. However, the results were different from the literature reported by J. Wu et al. [36]. They studied the effect of plating parameters, such as current density, CoSO₄ concentration, and bath composition, to optimize the deposition process. The target coating with Mn–Co near 1:1 could be obtained at a current density of 0.25 A cm⁻² with the composition of 0.5 M MnSO₄ and 0.1 M CoSO₄, 1.0 M H₃BO₃, 0.7 M gluconate, and 0.1 M (NH₄)₂SO₄. However, the concentration of MnSO₄ in the electrolytic bath should be studied to fulfill the investigation of Mn–Co alloy plating. The current density was adjusted in the range of previous work [40] when MnSO₄ was changed to give a dense deposited coating layer. Moreover, heat treatment processes were required to improve the densification of the coating layer. Y.-Z. Hu et al. [30] prepared Mn–Co spinel coatings via atmospheric plasma spray followed by heat treatments in reducing and subsequently oxidizing atmospheres. It was found that the adjacent metallic Co particles tended to be bonded together during the reducing process and transformed to Mn–Co spinel after the reoxidation process [30]. The dense coating reduced the inward diffusion of oxygen and the outward migration of chromium [30]. In our previous work, T. Thublaor et al. [40] prepared Mn–Co–electroplated coating layers annealed in argon followed by heat treatment in oxygen. The rather dense Mn–Co spinel coating helped reduce oxidation and chromium volatilization rates after being exposed to O₂–5%H₂O for up to 96 h at 800 °C, compared to one–step heat treatment in oxygen [11]. It was concluded that the heat treatment process effectively enhanced the densification of the coating, which further reduced the rate of oxidation and chromium volatilization.

In the present work, the electrodeposition method was used to deposit the Mn–Co alloy on AISI 430 stainless steel. The current density and the concentration of MnSO₄ in the electrolytic bath were varied, and the optimized result was used to prepare the coating for the oxidation and chromium volatilization test. The oxidation test was studied in air at 800 °C as a simulated SOFC operating condition on the cathode side. Cr species volatilization was also assessed in synthetic air–5%H₂O at the same temperature.

2. Materials and Methods

AISI 430 ferritic stainless steel was used in the present study. Its chemical composition was measured using optical emission spectroscopy (OES, Model ARL 3460 from Thermo ELECTRON CORPORATION, Ecublens, Switzerland), providing the element constituent in steels of Fe–16.41Cr–0.79Mn–0.20Si–0.10C–0.05Ni–0.03P in wt.%. The steel sheet was cut into rectangular pieces with sizes of 10.0 × 10.0 mm² and further cut to have curved ends to avoid spallation after the coating. The samples were ground with SiC paper up to 400 grit and then cleaned in an ultrasonic bath with deionized water (DI water) and ethanol. Finally, they were dried by an air dryer.

For the electrodeposition process, the electrolyte solution was first prepared by mixing 1.00 M H₃BO₃ with 0.70 M C₆H₁₁O₇Na in DI water. The 0.10 M CoSO₄ was added after stirring the solution until it was homogenous. The solution was further stirred continuously for 30 min and kept at room temperature for 24 h. It was then added with 0.1 M (NH₄)₂SO₄ and MnSO₄ under agitation and kept standing for 24 h. The pH was adjusted to 3.0 using 20.0 vol.% H₂SO₄. The electrodeposition was carried out using a two–electrode system comprising AISI 430 ferritic stainless steel as the cathode and AISI 304 austenitic stainless steel as the anode in the electrolyte solution. The two pieces of AISI 304 steel were cut into a rectangular shape with the dimensions of 25.0 × 30.0 × 1.0 mm³ and ground with SiC paper up to 1000 grits. Before electrodeposition, the surface of the sample was pickled in 25.0 vol.% HCl acid mixing 5.0 vol.% HNO₃ acid solution for 3 min and washed in ethanol with an ultrasonic machine and dried in air. Then, the sample was placed parallel to the anode. The current densities of the DC power supply were applied at 150, 250, and 350 mA cm⁻². The coating operated at room temperature for 20 min. After coating, the sample was dried for 20 min at 100 °C. To investigate appropriate parameters in Mn–Co electrodeposition, MnSO₄ concentrations of 0.50, 0.75, and 1.00 M were used.

For the oxidation test, the samples were placed inside an alumina crucible in a muffle furnace under an ambient air atmosphere at 800 °C for 30, 60, and 90 h. The samples were weighed using a precise balance before and after each test.

To investigate the Cr species volatilization of the coating in the simulated cathode atmospheres, humidified synthetic air (N₂–21%O₂–5%H₂O) was set for study at 800 °C for 96 h. The samples were cut into a rectangular shape (14.2 × 8.0 × 1.0 mm³), which had the curved ends of a semicircle, as described in our previous work [11]. The samples were prepared for electrodeposition and were coated in the abovementioned condition. Each sample was placed on a quartz tube in the horizontal furnace at room temperature. Then, it was heated to 800 °C in Ar. At this temperature, N₂–21%O₂ was the mixed gas carried into a water flask. A 5% H₂O content was controlled by maintaining the water temperature at 31 °C. The temperature was calculated according to the Clausius–Clapeyron equation using the enthalpy of water vaporization, 40,893.0 J mol⁻¹ [42]. The humidified synthetic air was fed into the furnace with a linear flow rate of 1.0 cm s⁻¹. It flowed through the sample surfaces along the furnace tube and condensed in the condenser column, which was additionally connected at an incline to the horizontal furnace tube at the exit. Then, it flowed to the lower end, immersed in water contained in a flask. After the test, the volatilized Cr species condensed in this water flask were collected. This condensate was called a concentration solution. The tube and condenser were cleaned with 0.1 M HCl, and this cleaning solution was added to the concentration solution, which was then taken to analyze the amount of Cr using inductively coupled plasma optical emission spectroscopy (ICP-OES, Agilent 5110, Agilent Technologies, Inc., Santa Clara, CA, USA).

To determine the effects of heat treatment atmospheres on Mn–Co spinel formation, the adhesion failure of the coating was studied. A bare steel sheet with a thickness of 1.0 mm was cut into pieces with a total length of 50 mm for the tensile test. The length and width of the gauge were 16 mm and 3 mm, respectively. This experiment has been detailed in our prior work [43]. Then, the sample surface was finished and coated with the same condition used previously. After being coated, the samples were loaded into a quartz tube in a horizontal furnace with O₂ and CO₂ atmospheres at 800 °C for 4 h for each condition. The linear velocity of the flowing gas was 1.0 cm s⁻¹. The sample was set in a tensile testing machine equipped with a high–speed CCD camera. The test was conducted at room temperature. A crosshead speed was set at 5 mm min⁻¹ with a tensile loading of 10 kN. The experimental force and deformation displacement were converted to stress–strain curves. During the tensile testing, the surface of the sample was monitored in situ using the CCD camera. The failure of the coating on the sample surface was observed and recorded as a video.

Standard scanning electron microscopy (SEM, QUANTA 450 from FEI^TM, Hillsboro, OR, USA) equipped with energy dispersive spectroscopy (EDS, EDAX-AMETEK, Inc., Berwyn, PA, USA) was used to characterize the morphology of the coating surface and cross–sectional images. EDS was used to determine the elemental composition of the coating sample. The formed phases in the coating were investigated using X-ray diffraction (XRD, Rigaku, Tokyo, Japan) with Cu Kα (α = 1.5406 Å) radiation. The diffraction pattern measured from the studied samples was matched with the standard XRD patterns organized by the International Center for Diffraction Data (ICDD, Newtown Square, PA, USA).

3. Results and Discussion

3.1. Surface Morphology of the As–Coated Sample

Figure 1 shows top–view photographs of the sample surfaces after coating using the electrodeposition technique with different concentrations of MnSO₄ (M) and the applied current densities (mA cm⁻²). The conditions of 1.00 M MnSO₄ at 150 mA cm⁻², 0.75 M MnSO₄ at 250 mA cm⁻², and 0.50 M MnSO₄ at 350 mA cm⁻² are well deposited without spallation. Among the coating conditions shown in Figure 1, the electrolytic bath with 0.5 M MnSO₄ provides dense coating with increasing current density. However, one sample has some spallation at the edge of the steel sample at low current density. To observe the microscopical morphology of the surface, the secondary electron image reveals a dense deposit with large grains of packed particles for the sample with the coating condition of 1.00 M MnSO₄ at 150 mA cm⁻² (Figure 2a), while Figure 2b,c exhibit small grains of particles for both conditions of 0.75 M MnSO₄ at 250 mA cm⁻² and 0.50 M MnSO₄ at 350 mA cm⁻². They have spongy or porous morphology. The porosity is evaluated and analyzed using ImageJ, as shown in Figure 2d,e. The values are 28.15% and 13.93%. In addition, the histograms of particle sizes on the coating are shown in Figure 2f–h. The average particle size of the as–coated sample produced using 1.00 M MnSO₄ at 150 mA cm⁻² was 2.43 μm. When applying the current densities of 250 mA cm⁻² in 0.75 M MnSO₄ (Figure 2g) and 350 mA cm⁻² in 0.50 M MnSO₄ (Figure 2h), the average particle sizes were 1.13 μm and 1.47 μm, respectively. It was noticed that increasing the current density from 250 to 350 mA cm⁻² resulted in the finer particle size and the less porous coating. These findings show that increasing the current density above 150 mA cm⁻² causes the morphology to shift from large grains to small grains of packed particles.

J. Wu et al. [36] studied Mn–Co electrodeposition in a similar Mn–Co sulfate bath. They reported that a spongy or porous morphology of the as–coated sample was obtained with the presence of a metal hydroxide—Co(OH)₂ and Mn(OH)₂—when a current density of at least 200 mA cm⁻² was applied. Given that Co²⁺ and Mn²⁺ from the electrolytic solution receive electrons at the cathode, Co and Mn solids are deposited on the steel surface according to Reactions (1) and (2) [36].

$${\mathrm{Co}}^{2+}+2{\mathrm{e}}^{\prime }\to {\mathrm{Co}}_{\left(\mathrm{s}\right)}$$

(1)

$${\mathrm{Mn}}^{2+}+2{\mathrm{e}}^{\prime }\to {\mathrm{Mn}}_{\left(\mathrm{s}\right)}$$

(2)

Meanwhile, the water can be split by adding hydrogen gas (as bubbles) and hydroxide ions $({\mathrm{OH}}^{\prime })$ as follows in Reaction (3). Co and Mn ions can react with hydroxide ions, yielding metal hydroxide as the reaction product [36].

$${2\mathrm{H}}_2{\mathrm{O}}_{\left(\mathrm{l}\right)}+2{\mathrm{e}}^{\prime }\to {\mathrm{H}}_2{}_{\left(\mathrm{g}\right)}+2{\mathrm{OH}}^{\prime }$$

(3)

The hydrogen bubbles formed on the surface electrode from Reaction (3) inhibit the reduction reaction and contribute to the coating porosity.

3.2. Oxidation Test and XRD Phase Identification

From Figure 1, the three dense coatings were selected to investigate oxidation durability. Figure 3 shows the top–view photographs of the surface coating after oxidizing in air at 800 °C for 30, 60, and 96 h. Spallation can be seen in every oxidation period for the coating under 1.00 M MnSO₄ at 150 mA cm⁻². Other conditions show fine coating layers covering the surfaces with spallation or peeling off. The weight of the coated samples without spallation before and after the oxidation test was investigated and converted to the weight per unit surface area; this was called the weight gain ($\mathsf{\Delta }\mathrm{m}/\mathrm{A}$). This quantity is plotted as a function of oxidation time (t) in Figure 4. The mass gain of the sample coated with 0.50 M MnSO₄ at 350 mA cm⁻² is lower than that coated with 0.75 M MnSO₄ at 250 mA cm⁻². It decreases by 2.0– to 2.6–fold after exposure in the oxidation test from 30 h to 90 h.

It may be related to oxygen adsorption caused by being exposed to ambient air, which is also the cause of porosity in the coating. Figure 2d,e provide the porosity of the as–coated samples with 0.75 M MnSO₄ at 250 mA cm⁻² and 0.50 M MnSO₄ at 350 mA cm⁻². The weight gain and porosity decrease similarly by 75%. In this case, it is implied that the oxidation rate is dominated by the porosity of the coating. In the present work, the sample coated with an electrolytic solution of 0.10 M CoSO₄ + 0.50 MnSO₄ at an applied current density of 350 mA cm⁻² showed the best oxidation resistance.

To identify the phase formation after oxidation at 800 °C in air for 30 h, the XRD result is reported in Figure 5a for the sample with the best oxidation resistance. We detected the presence of MnCo₂O₄ (ICDD 01-1130), MnCr₂O₄ (ICDD 75-1614), Cr₂O₃ (ICDD 38-1479), and Fe (ICDD 87-0721), a cubic I m–3 m ferritic phase as a steel substrate. Furthermore, to assess the thermodynamic stability of the oxides formed, the equilibrium oxygen partial pressure was calculated using the following formula: $p_{O_2,eq}=\exp \left(\mathsf{\Delta }{G}^{°}/\mathrm{R}T\right)$. $\mathsf{\Delta }{G}^{°}$ is the standard Gibbs free energy of oxide formation, while T and R are the absolute temperature (K) and the universal gas constant (8.314 J mol⁻¹ K⁻¹), respectively. Using Hess’s law, the standard Gibbs free energy of MnCr₂O₄ formation from their constituent elements and oxygen can be obtained from the standard Gibbs free energy of MnCr₂O₄ formation from Cr₂O₃ and MnO provided by Holcomb and Alman [44], and the standard Gibbs free energies of Cr₂O₃ and MnO formations which were calculated from the standard enthalpy and entropy of their constituent elements reported by Kubaschewski and Alcock [45]. For MnCo₂O₄ formation, Petric and Ling provide a standard Gibbs free energy of −30.3 kJ [22]. Using this assumption, it is derived from the reaction of Co₃O₄ and Mn₃O₄ [46]. $\mathsf{\Delta }{G}^{°}$ of MnCo₂O₄ formation was calculated from its constituent elements and oxygen. The calculating method is similar to that of MnCr₂O₄ formation. The data for investigating $\mathsf{\Delta }{G}^{°}$ of Co₃O₄ and Mn₃O₄ formation were obtained by Kubaschewski and Alcock [45]. At 800 °C, the standard Gibbs free energies per 1 mol oxygen of MnCr₂O₄, Cr₂O₃, and MnCo₂O₄ formation are −604.58, −560.72, and −343.70 kJ, respectively. These energies provide equilibrium oxygen partial pressures of 3.69 × 10⁻³⁰, 5.04 × 10⁻²⁸, and 1.85 × 10⁻¹⁷ bar for MnCr₂O₄, Cr₂O₃, and MnCo₂O₄, respectively. The actual oxygen partial pressure in the ambient air was 0.21 bar. This result is much higher than the equilibrium oxygen partial pressures of MnCr₂O₄, Cr₂O₃, and MnCo₂O₄, indicating that the formation of these oxides in air is thermodynamically possible.

3.3. Cr-Species Volatilization

At SOFC operating temperatures, the chromia can be volatilized, and the volatile species consequently contaminates the SOFC cathode, resulting in cell performance deterioration due to chromium poisoning [13,14,15,16]. The water vapor content in air (cathode side) is 3.6% at an ambient temperature of 25 °C, calculated using the standard thermodynamic data [45]. Moreover, it was reported that the CrO₂(OH)₂ is the dominant volatile species than CrO₃ under the humidified oxidizing atmosphere at a temperature lower than 1500 °C [9,10,11,12]. To hasten the volatilization of Cr–species, 5% water vapor content was added to air. The presence of oxides containing chromium, such as Cr₂O₃ and MnCr₂O_4, is detected on the surface of stainless steels exposed to air containing water vapor [47,48]. Figure 5b, the XRD result for the sample coated with 0.10 M CoSO₄ + 0.50 MnSO₄ at 350 mA cm⁻² after oxidation at 800 °C in humidified synthetic air (N₂–21%O₂–5%H₂O) for 96 h, shows that MnCr₂O₄ (ICDD 75-1614), Cr₂O₃ (ICDD 38-1479), MnCo₂O₄ (ICDD 01-1130), and Fe (ICDD 87-0721) are present. As Cr₂O₃ and MnCr₂O₄ are contained in the coatings, volatile Cr species can be liberated. The volatilized Cr–rich oxide can yield CrO₂(OH)₂ according to Reactions (4) and (5) [9,10,44,49].

$$\frac{1}{2}{\mathrm{Cr}}_2{\mathrm{O}}_{3(\mathrm{s}}{}_{)}+\frac{3}{4}{\mathrm{O}}_{2\left(\mathrm{g}\right)}{+\mathrm{H}}_2{\mathrm{O}}_{\left(\mathrm{g}\right)}\to {\mathrm{CrO}}_2{\left(\mathrm{OH}\right)}_{2\left(\mathrm{g}\right)}$$

(4)

$$\frac{1}{2}{\mathrm{MnCr}}_2{\mathrm{O}}_{4(\mathrm{s}}{}_{)}+\frac{3}{4}{\mathrm{O}}_{2\left(\mathrm{g}\right)}{+\mathrm{H}}_2{\mathrm{O}}_{\left(\mathrm{g}\right)}\to {\mathrm{CrO}}_2{\left(\mathrm{OH}\right)}_{2\left(\mathrm{g}\right)}+\frac{1}{2}{\mathrm{MnO}}_{(\mathrm{s}}{}_{)}$$

(5)

Young and Pint [50] presented the mass flux of Cr loss in terms of volatilization (J) as the following Equation (6) to predict the rate of Cr species volatilization on a flat surface [50].

$$J=\frac{M_{Cr}{k}_m}{RT}(p_s-p_b)$$

(6)

where $M_{Cr}$ is the atomic mass of Cr (51.9962 g mol⁻¹ [42]), $k_m$ is the mass transport coefficient (cm s⁻¹), R is the universal gas constant (8.314 J mol⁻¹ K⁻¹), T is the absolute temperature (K), and $p_s$ and $p_b$ are the partial pressure of gas volatile species at the solid surface and in the bulk gas, respectively. By considering that $p_b$ is insignificant compared with $p_s$, it is neglected in Equation (6). Additionally, $k_m$ is defined as Equation (7) [49,50]:

$$k_m=0.664{\left(\frac{{\rho }_A{V}^3{D}_{A-B}^4}{{\mu }_A{L}^3}\right)}^{\frac{1}{6}}$$

(7)

where ${\rho }_A$ is the density of the solvent gas (g cm⁻³), V is the linear velocity of the bulk gas (cm s⁻¹), L is the sample length (cm), ${\mu }_A$ is the viscosity of the solvent gas (g cm⁻¹ s⁻¹), and $D_{A-B}$ is the binary gas diffusion coefficient (cm² s⁻¹) between the solvent gas (air) and the Cr(VI) species gas (CrO₂(OH)₂). ${\mu }_A$ and $D_{A-B}$ are given as follows [50,51].

$${\mu }_A=2.6693\times {10}^{-5}\left(\frac{\sqrt{M_{A}T}}{{\sigma }_A^2{\mathsf{\Omega }}_{\mu ,A}}\right)$$

(8)

$$D_{A-B}=\frac{\left(1.858\times {10}^{-3}\right)T^{\frac{3}{2}}}{{\sigma }_{A-B}^2{\mathsf{\Omega }}_{\mu ,A-B}}\sqrt{\frac{1}{M_A}+\frac{1}{M_B}}$$

(9)

From the equation above, ${\sigma }_A$ is the collision diameter of the gas solvent (Å), and ${\mathsf{\Omega }}_{\mu ,A}$ is the dimensionless collision integral for viscosity at the absolute temperature. ${\mathsf{\Omega }}_{\mu ,A}$ can be calculated from $Tk/\epsilon$ as reported by Bird et al. [52]. $\epsilon$ is the characteristic energy of the interaction between molecules, and k is Boltzmann’s constant. The values of $\sigma$ and $\epsilon /k$ are given in

Table 1. Molecular mass (M) and the gas molecular interaction parameters [49,50].

Species	M (g mol–1)	σ (Å)	ε/k (K)
Air	28.97	3.617	97.0
CrO2(OH)2	118.01	4–5	230–450

. ${\sigma }_{A-B}$ and ${\mathsf{\Omega }}_{\mu ,A-B}$ are the collision diameter and the collision integral for the binary gas, which can be calculated as Equations (10) and (11), respectively [50,53].

$${\sigma }_{A-B}=\frac{1}{2}\left({\sigma }_A+{\sigma }_B\right)$$

(10)

$$\frac{{\epsilon }_{A-B}}{k}={\left(\frac{{\epsilon }_A}{k}·\frac{{\epsilon }_B}{k}\right)}^{\frac{1}{2}}$$

(11)

$p_s$ is partial pressure of CrO₂(OH)₂ (bar) over Cr₂O₃ and MnCr₂O₄, which according to Reactions (4) and (5) can be obtained by Equations (12) and (13) [49].

$$p_{{\mathrm{CrO}}_2{\left(\mathrm{OH}\right)}_2}=a_{C{r}_2{O}_3}^{1/2}{p}_{O_2}^{3/4}{p}_{H_{2}O}^{}\exp \left(-\frac{\mathsf{\Delta }{G}_4^{°}}{RT}\right)$$

(12)

$$p_{{\mathrm{CrO}}_2{\left(\mathrm{OH}\right)}_2}={\left(\frac{a_{MnC{r}_2{O}_4}}{a_{MnO}}\right)}^{\frac{1}{2}}{p}_{O_2}^{3/4}{p}_{H_{2}O}^{}\exp \left(-\frac{\mathsf{\Delta }{G}_5^{°}}{RT}\right)$$

(13)

In case of air, the partial pressure of oxygen is 0.21 $(p_{O_2}=0.21)$. The activities (a) of solid phases are equal to unity $(a=1)$. $\mathsf{\Delta }{G}_4^{°}$ is the standard Gibbs free energy of Reactions (4) as 49,150 + 45.13 T J, whereas $\mathsf{\Delta }{G}_5^{°}$ is 88,142.50 + 38.33 T J for the standard Gibbs free energy of Reactions (5), the thermodynamic data taken from the literature [9,11,44,45]. By inserting Equations (7), (12), and (13) into Equation (6), the mass flux (J) of Cr volatilization at different water vapor contents in air can be calculated and drawn on a logarithmic scale, as shown in Figure 6.

This figure plots mass fluxes of Cr volatilization from AISI 430 stainless steel [21,54,55], Crofer 22 APU [56,57], and AISI 441 stainless steel [58], after oxidation in air with different water vapor contents (air–H₂O). The mass fluxes of Cr volatilization from stainless steel are found in the gray zone, the region between a higher mass flux of Cr volatilization from MnCr₂O₄ and a lower mass flux from Cr₂O₃. Thus, the mass flux of Cr volatilization from stainless steel is obtained from the mass flux of Cr volatilization from MnCr₂O₄ combined with Cr₂O₃ [11]. All Mn–Co spinel–coated samples [21,59] have lower volatile Cr species than uncoated samples [21,54,55]. In this work, the average chromium volatilization rate of Mn–Co spinel–coated samples from two trials of the experiment is 4.39 × 10⁻¹¹ g cm⁻² s⁻¹ after exposure to synthetic air–5%H₂O for 96 h. The coating can reduce the volatilized rate compared to uncoated AISI 430 stainless steels. This comparison implies that the Mn–Co spinel coating potentially could reduce the chromium volatilization from the stainless steel.

3.4. O₂ and CO₂ Heat Treatment Atmospheres of the Coating Process

For the coating failure after oxidation in different heat treatment atmospheres, the plating condition for a dense coat and the best oxidation resistance, mixing 0.10 M CoSO₄ with 0.50 MnSO₄ at a current density of 350 mA cm⁻², was applied to AISI 430 stainless steel. The as–coated sample was heated at 800 °C for 4 h in O₂ and CO₂ atmospheres. During tensile testing, the failure was monitored using a high–speed CCD camera. The displacement of deformation was related to the strain of the sample and failure due to spallation. The first and continuing failure activities of the coating in the video were observed. From the video record, the average strain at first spallation is 0.59%, s.d. = 0.27%, for the sample heat treated in CO₂, indicating good resistance to failure, and 0.18%, s.d. = 0.05%, for the sample treated in an O₂ atmosphere. Figure 7a shows SEM images of the coated samples after annealing in O₂ and CO₂, after straining by 1%. Cracks can be seen on the coating layer of both samples. The EDS peaks of Mn, Co, and O were detected on both spots 1 and 2. Figure 7b shows SEM images of the areas near the broken necks after tensile testing. Plates or flakes of cracked coating layers are observed as a result of the applied load action. Some cracked layers remain on the surfaces, and the morphology is identical for both samples. The EDS spot analyzed results are shown in Figure 7b. For both samples, a strong detection of the Fe peak and a relatively low intensity of the O peak of spot 1 indicate the detection of a steel substrate. The relatively high intensity of O peaks of spots 2 and 3 of the cracked layers accompanied by Mn and Co detection corresponds to the coating layer.

In addition, the samples with post−heat treatment without tensile testing were cross−sectioned and analyzed using SEM and EDS mapping. The results are shown in Figure 8a,b for heat treatment in O₂ and CO₂, respectively. Both samples have continuous coating layers containing Mn, Co, and O. Furthermore, only a few porosities are apparent, with small cracks in the coating and at the coating/steel interface. The porosity of the sample annealed in O₂ and CO₂ seems to be difficult to observe from the cross−sectional SEM images. Figure 9 shows top−view surface SEM images of the samples annealed in CO₂ in Figure 9 (right), which have lower porosity than that annealed in O₂ in Figure 9 (left). From this evidence, it can be explained that the load from the tensile test is transferred to the Mn–Co coating layer. The load demolished the cohesive and adhesive binding forces of the coating, which appears as a mixed mode between cracking and spallation within the coating and coating/substrate interface [60]. Therefore, the heat treatment of the coating in a CO₂ atmosphere affects a dense coating by helping it avoid voids and gaps.

As a matter of discussion, the presence of intermetallic compounds at the interface between two different phases could reduce the interfacial adhesive strength [61,62,63]. In the case of stainless steel, the presence of Fe–Mo intermetallic compounds at the interface between the AISI 444 (Fe–18Cr–Mo–Ti–Nb) and its thermal oxide scale was found to help provoke the spallation [61]; however, such a phase was not observed here. The detachment of the coating from the studied steel, here, should be primarily due to the polishing and mounting processes. In addition, the coating layer can serve as a barrier to inhibit the oxygen inward diffusion and the outward diffusion of cationic species from the substrate [11,64,65]. As for the 800H Ni–based alloy oxidized at high temperature, J. Yang et al. [65] have recently found that the Cr coating could retard the Fe outward diffusion at a high temperature. For the AISI 430 stainless steel, we previously applied the Mn–Co oxide through slurry coating and conducted the oxidation test at 800 °C in O₂–5%H₂O [64]. It was found that the coating could help reduce the outward diffusion of Fe from the steel substrate, and Cr loss due to volatilization also tended to be reduced, compared with the result reported in literature [11]. We not only developed the Mn–Co coating through the slurry method but also through the electroplating method, and also found the beneficial effect that the Mn–Co coating had on reducing the Cr volatilization, as presented in Figure 6.

4. Conclusions

The electrodeposition of Mn–Co alloys was conducted on AISI 430 stainless steels. The effect of electroplating parameters on the coating quality was primarily studied. Furthermore, the samples were oxidized in air and humidified synthetic air (N₂–21%O₂–5%H₂O) at 800 °C to investigate oxidation resistance and Cr species volatilization, respectively. Heat treatments in O₂ and CO₂ atmospheres for 4 h at 800 °C were performed to investigate the failure of the coating in the last section. The following conclusions can be drawn.

1. Various electroplating current densities (150, 250, and 350 mA cm⁻²) and concentrations (0.50, 0.75, and 1.00 M MnSO₄) were investigated. An increased current density with a lower MnSO₄ concentration was required for a dense coating with good oxidation resistance without spallation. The optimized conditions are 1.00 M MnSO₄ at 150 mA cm⁻², 0.75 M MnSO₄ at 250 mA cm⁻², and 0.50 M MnSO₄ at 350 mA cm⁻².

2. The sample coated with an electrolyte solution of 0.10 M CoSO₄ mixed with 0.50 MnSO₄ at an applied current density of 350 mA cm⁻² demonstrated the best oxidation resistance in air. The low weight gain was a result of the lower porosity of the coating.

3. After 96 h of exposure, the volatilization of the Cr species of the coated samples was measured in humidified synthetic air at 800 °C. The results were on the order of 10⁻¹¹ g cm⁻² s⁻¹, in good agreement with the volatilization of Cr₂O₃ and MnCr₂O₄ reported in the literature.

4. Heat treatments in O₂ and CO₂ atmospheres were conducted to convert Mn–Co alloys to form the spinel of the coatings. The samples annealed in CO₂ had low porosity or void formation, resulting in greater resistance to failure than those annealed in O₂.