High-Temperature Corrosion Behavior of Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ Coating Prepared by rf Magnetron Sputtering

Abstract

Using the rf magnetron sputtering technique, Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ coatings were formed and obtained as a thin film on Hastelloy substrates. When subjected to high-temperature conditions, the effect of lanthanum on the anti-corrosive properties of the coatings was investigated. The anti-corrosive response was evaluated by electrochemical impedance spectroscopy and potentiodynamic curves, which are rarely reported. Hot corrosion occurs through the electrochemical mechanism, and more information can be obtained through electrochemical corrosion tests, which are very effective and fast. The electrochemical behavior at high temperatures was studied via molten salt corrosion tests, potentiodynamic polarization curves, and electrochemical impedance spectroscopy. Additionally, the coatings were evaluated via scanning electron microscopy and transmission microscopy to determine their morphology. With X-ray diffraction, the crystallinity of the films was determined. It was determined that the corrosion rate directly correlates with the temperature, attributed to the mechanisms induced by the Na₂SO₄ and V₂O₅ salts that generated condensation. As the temperature increases, the density of the corrosion current increases in the thin films of Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂. When comparing the two compounds, it is determined that the increase in lanthanum alters the positive acid character, thus reducing the dissolution of the oxides and increasing protection.

1. Introduction

The composite of bismuth, titanium, and oxygen in the Bi₄Ti₃O₁₂ phase is a material that contributes to the phase-structured ferroelectrics family [1]; it consists of three units of a perovskite-like structure. It has a high dielectric constant and high breakdown strength; the high Curie temperature is reported to be 675 °C [2], making it useful for applications as a converter for pyro-electric devices over a wide temperature range (20 to 600 °C). Bi₄Ti₃O₁₂’s perovskite structure comprises three layers of TiO₆ octahedra with Bi₃⁺ ions. The layers of octahedra and (Bi₂O₂)₂⁺ are observable when Trivalent elements such as lanthanum are used to replace bismuth [3,4]. Bi₂O₃ applications have been primarily used in sensors and solid electrolytes [5,6]. However, there is a broad spectrum of applications for bismuth-based ternary and quaternary compounds containing metal oxides, and uses have been found in areas such as pigments, superconductors, sensors, scintillators, oxidation catalysis, photocatalysis, high-temperature electrolytes, and thermal barrier materials [7].

Technological advances allow the development of coatings for applications in high-performance devices, where they are deposited on metal parts exposed to high temperatures and corrosive environments [8]. In the process of high-temperature corrosion, the degradation of materials does not require the presence of a liquid electrolyte [9]. In most industrial environments, high-temperature oxidation, regardless of the predominant mode of corrosion, has demanded the development of materials and techniques to match its needs for device fabrication [10], leading to the research and production of materials in particular designs and arrangements. The corrosion rate depends on the physicochemical properties of the material, the corrosive atmosphere in which it is exposed, and the corrosion products [11]. In a variety of engineering application systems where high temperatures are used, contact between the material and the molten salt film occurs, generating a considerable increase in the corrosion rate of the materials and thus causing their deterioration [12]. High-temperature corrosion phenomena, due to salt deposits caused by combustion ashes, have usually resulted in the modification of the reaction rate of the metal with the environment. This produces molten salt baths or salt films with high adhesion to be in contact with the metal at elevated temperatures [13].

Consequently, there is an acceleration of corrosion that often occurs on the contact of the molten salt electrolyte with the metal or its protective oxide film. It is evident, then, that the combustion of ash is a significant factor in the corrosive attack. Sodium sulfate (Na₂SO₄) and vanadium pentoxide (V₂O₅) are the most common salts in deposits involved in molten salt corrosion and come from fuels containing contaminants such as sulfur (S) and vanadium (V), or in environments that may contain sodium chloride (NaCl) [14]. When fuels are burned with excess oxygen, Na₂SO₄, V₂O₅, and NaVO₃, among others, are formed. Vanadium-containing substances are very harmful and contribute to increased corrosion in combustion process equipment. Since many studies have been carried out to examine the mechanisms of Na₂SO₄-induced hot corrosion, researchers have found that the condensation of Na₂SO₄ is necessary to accelerate corrosion [15].

In this study, rf magnetron sputtering was used to fabricate coatings. Hot corrosion occurs through the electrochemical mechanism, and more information can be obtained through electrochemical corrosion testing, which is very effective and quick [16]. Electrochemical impedance spectroscopy (EIS) is a technique that has proven effective in investigating the mechanisms and reaction kinetics in hot corrosion induced by molten carbonates. Electrochemical testing of the Tafel determination of the corrosion rate was performed using a method known as Tafel extrapolation [17]. As with all electrochemical studies, the environment must be electrically conductive. Although potentiodynamic curves and electrochemical impedance spectroscopy measurements are widely used to evaluate the corrosion resistance or reveal the corrosion mechanism at room temperature [18], electrochemical measurements are rarely used at high temperatures. In this study, high-temperature corrosion behavior was measured by electrochemical impedance spectroscopy measurements and potentiodynamic curves [19,20].

2. Materials and Methods

Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ thin films on commercial Hastelloy X Nickel-based superalloy substrates were fabricated by rf magnetron sputtering. The details of the deposition process and conditions were as follows: Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ targets [15], rf power of 40 W, argon deposition atmosphere, deposition time 65 min, target to substrate distance 65 mm. Samples were ultrasonically cleaned in acetone for 15 min before the X-ray diffraction (XRD) and scanning electron microscopy (SEM) characterization. Coating surfaces and their cross-sections were examined with the JSM 6490LV JEOL SEM to determine the morphological and microstructural modifications before and after the hot corrosion treatment. X-ray diffraction (XRD) was used for local element analyses, and for detecting corrosion products both on the surface and throughout the thickness of the sample. These coatings were structurally characterized by XRD before and after the hot corrosion test and analyzed by Rietveld’s method. X-ray diffraction patterns were collected at room temperature using 0.0197° (2θ) scanning steps over a range of 9.99720° < 2θ < 90.00001° on a Panalytical X’Pert Pro MRD with Cu Kα (1.541874 Å) radiation.

For electrochemical impedance spectroscopy measurements and potentiodynamic curves, the sample was placed in an induction furnace and heated from room temperature to a maximum of 1000 °C. Measurements were made at three temperatures: 700 ± 3 °C, 850 ± 3 °C, and 1000 ± 3 °C. These temperatures were maintained for 45 min in an air atmosphere while electrochemical measurements were performed. Afterward, the specimen was cooled to room temperature to reduce thermal shocks. The experimental setup for the EIS analysis is presented in Figure 1.

Electrochemical measurements were performed using a Gamry Instrument Interface 1010E potentiostat frequency response analyzer. Electrochemical impedance spectroscopy (EIS) curves were obtained in situ at 700 ± 3 °C, 850 ± 3 °C, and 1000 ± 3 °C, using a cell containing a working electrode with an exposed area of 1 cm², a platinum wire as a counter electrode (CE), and a platinum wire as a reference electrode (RE). In the laboratory conditions, a mixture of vanadium pentoxide powder (V₂O₅) and sodium sulfate (Na₂SO₄) in a ratio of 50:50 (wt%) at a rate of 90 g/cm² was used as the corrosive salt for the simulation of the accelerated hot corrosion process. Bode plots were recorded using a sinusoidal signal amplitude of 5 mV applied to the working (specimen) and reference electrodes in a frequency range from 0.1 Hz to 300 kHz. Polarization curves were obtained at a sweep speed of 1 mV/s in a voltage range between −0.25 and 0.25 V against the electrode reference potential, defined for the open-circuit potential.

3. Results and Discussion

3.1. High-Temperature Corrosion

Figure 2 shows the potentiodynamic polarization curves for the substrate as a function of temperatures of 700 °C, 850 °C, and 1000 °C, respectively. Again, the variation of the corrosion potentials is observed. The corrosion potential generally tends to take more positive values at 850 °C. The values of corrosion potential, corrosion current, and corrosion rate for the substrate, according to the heat treatment temperatures, are consolidated in

Table 1. The corrosion potential, current density, and corrosion rate were obtained for the steel substrate at 700 °C, 850 °C, and 1000 °C.

Temperature (°C)	Error (V vs. Pt)	Icorr (μA/cm2)	Corrosion Rate (μm/Year)
700	−1.47	8.18 × 10−6	94.28
850	−1.44	16.95 × 10−6	195.36
1000	−1.49	77.44 × 10−6	892.55

.

The responses of the coatings at high temperatures are evaluated by employing two tests; the first one, to determine the rate of corrosion and the corrosion kinetics, is the potentiodynamic polarization technique (Figure 3 and Figure 4). The evaluation of the oxidation of the coatings in molten salts simulates severe conditions in atmospheres, with the presence of contaminants and materials used in pipes that must withstand high temperatures in combustion processes (580 °C) [21,22]. Hence, a coated film aims to protect the base material, which is the same steel previously used for this purpose. Therefore, the degree of protection of the metal–coating interface must be determined. In Figure 2 and Figure 3, the corrosion rate was determined to have a direct relationship with the temperature where the mechanisms induced by the Na₂ SO₄ and V₂O₅ salts have generated condensation. By increasing the temperature, the corrosion current density increases in the thin films of Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂. By comparing the two compounds, it is determined that the increase in La alters the positive acid character, thus decreasing the dissolution of the oxides and increasing the protection [23]. The concentrations of Titanium are kept constant, and the concentrations of Lanthanum are varied, the results show that increasing the content of Lanthanum in the coatings improves the anticorrosive protection of the substrate, as observed when comparing the potentiodynamic curves. This curve indicates that the electrochemical action involving combustion gases and corrosive salts depends on the compounds formed at the test temperatures. The corrosion rates in the so-called high-temperature elements are engineered in systems with temperatures above 600 °C; the base steel performs adequately under this temperature. However, at temperatures above 700 °C, layers of molten solids are formed, causing an increase in both the corrosion rate and the deterioration of the substrate. The coating protects the substrate from accelerated corrosion processes [24].

Figure 3 and Figure 4 correspond to the potentiodynamic curves of the Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ coatings. The results show that the damage is minimal at temperatures of 700 °C. When increasing to 850 °C, significant damage to the substrate-coating system is evident. At 1000 °C, damages are catastrophic due to increased corrosion current density. In all the evaluated cases, it was determined that a general dissolution exists in the anodic region. There was no evidence of pitting corrosion, typical of steel, which indicates an adequate barrier. When a general dissolution is generated, this is associated with the thin film having contact with the molten salt composed of vanadium, sodium, and sulfur. The more ashes of these compounds are generated by the evaluation temperature, the more degradation is present, and therefore, these compounds are related to a condensed corrosion-causing phase at high temperatures [25].

The results based on the treatment temperatures, the corrosion potential, current, and corrosion rate values for Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ are consolidated in

Table 2. The corrosion potential, current density, and corrosion rate obtained for Bi_3.75La_0.25Ti₃O₁₂ at 700 °C, 850 °C, and 1000 °C.

Temperature (°C)	Error (V vs. Pt)	Icorr (μA/cm2)	Corrosion Rate (μm/Year)
700	−0.09	1.01 × 10−6	1.12
850	−0.10	0.68 × 10−6	7.82
1000	−0.11	16.10 × 10−6	185.55

and

Table 3. The corrosion potential, current density, and corrosion rate obtained for Bi₃La₁Ti₃O₁₂ at 700 °C, 850 °C, and 1000 °C.

Temperature (°C)	Error (V vs. Pt)	Icorr (μA/cm2)	Corrosion Rate (μm/Year)
700	0.21	0.10 × 10−6	1.18
850	0.18	0.59 × 10−6	4.47
1000	0.16	8.14 × 10−6	93.83

, respectively.

3.2. Electrochemical Impedance Spectroscopy

Figure 5 and Figure 6 present the Bode diagrams, which are the response of the electrochemical impedance spectroscopy technique using 5 mV and a potential signal applied to Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ coatings, evaluated in the temperature range of 700 °C to 1000 °C, in 150 °C increments, where these films are undergoing corrosion processes. The response is measured in current, thus determining the material’s response to the degradation processes [26]. Bode diagram analysis allows observation by the logarithm of the absolute value of the impedance versus the frequency. In the two systems, it was determined that at high temperatures, the impedance response is low, and as a result, a series of impedance values is obtained for each frequency studied. This relationship between frequency and impedance is known as the impedance spectrum, where the system with the lowest amount of lanthanum response is about 10 ohm (Figure 5).

The system with the highest lanthanum content evaluated at 1000 °C shows a six-fold increase to 67 ohm (Figure 6) [27,28]. These plots indicate similar results for the systems evaluated at 700 °C and 850 °C due to the material’s temperature increase. The resulting signals were obtained due to the frequency variation between 10⁻² Hz and 10¹ Hz, which allowed the diagrams to be adjusted by applying the same input variable to the physical interpretation and the chemical reactions. The diagrams indicate that the generated reactions allow charge transport, i.e., the corrosive salts are transported through the material. As the temperature increases, the rate of charge transport increases; regarding the barrier generated by the formation of a surface coating on the substrates, and the flow of charge through this protective layer, the Bode diagrams indicate that at both 700 °C and 850 °C the system is suitable. The spectrum does not allow corrosive salts to permeate because of pores that have not interconnected. However, at 1000 °C, these pores interlock and form cracks and can be highly brittle [29]. The phase that generates the protection is due to the characteristics of the bismuth lanthanum titanate thin film. In this case, the composite with a higher amount of lanthanum generates 10 times more protection; however, after surpassing 850 °C, the transfer of charge through the discontinuities and defects of the thin film causes the breakage of the protective layer, resulting in continuous failure [30,31].

The equivalent circuit proposed to model the behavior of the Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ coatings has been schematized in Figure 7. The model considers that the corrosion of the films deposited on the substrate is located in the permeable pores where corrosive salts can penetrate them and reach the surface of the substrate. Therefore, the coating can be considered a leaky capacitor (constant phase element). The constituent elements of the electrical circuit in Figure 7 are as follows: R.E. the reference electrode, W.E. the working electrode, Rs solution resistance, Rp1 porosity charge transfer resistance, CP1 coating capacitance, CP2 the capacitance of the exposed metal, and Rp2 the charge transfer resistance of the substrate–coating interface.

Table 4. Values of the electrochemical parameters of the equivalent electrical circuit for Bi_3.75La_0.25Ti₃O₁₂ coatings.

Temperature (°C)	Rs (Ω cm2)	CPE1 (μF cm−2 s−(1−α1)	α1	Rp1 (Ω cm2)	CPE2 μF cm−2 s−(1−α2)	α2	Rp2 103(Ω cm2)
700	4.10 (0.3%)	92.56 (4%)	0.75	19.20 (3%)	342.83 (3%)	0.79	2.43 (0.2%)
850	5.24 (0.2%)	87.12 (4%)	0.69	5.31 (2%)	546.23 (3%)	0.75	1.02 (0.3%)
1000	5.91 (0.3%)	92.53 (5%)	0.74	2.84 (2%)	653.76 (4%)	0.67	0.05 (0.2%)

and

Table 5. Values of the electrochemical parameters of the equivalent electrical circuit for Bi₃La₁Ti₃O₁₂ coatings.

Temperature (°C)	Rs (Ω cm2)	CPE1 (μF cm−2 s−(1−α1)	α1	Rp1 (Ω cm2)	CPE2 μF cm−2 s−(1−α2)	α2	Rp2 103(Ω cm2)
700	1.31 (0.4%)	121.05 (2%)	0.54	5.3 (0.1%)	564.21 (3%)	0.56	0.08 (0.01%)
850	3.22 (0.6%)	243.13 (3%)	0.59	2.3 (0.2%)	853.28 (2%)	0.67	0.04 (0.02%)
1000	2.54 (0.6%)	322.25 (3%)	0.64	1.3 (0.3%)	1035.25 (2%)	0.64	0.01 (0.02%)

show the values of the electrochemical parameters of the equivalent electrical circuit in Figure 7 for the Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ coatings, respectively. Low values of solution resistance (Rs) are observed in the results due to the contribution of the electrolyte. The values for Rp2 (resistance to polarization to charge transfer of the substrate–coating interface) of the substrate are lower. This shows that the films act as an anti-corrosive barrier against attack by the electrolytic solution.

3.3. X-ray Diffraction

Figure 8 and Figure 9 represent the diffraction pattern obtained for lanthanum bismuth titanate thin films deposited on Hastelloy X substrates. The results were compared with the Joint Committee on Powder Diffraction Standards (JCPDS 73-2128) [32,33]. The effect of temperature on the crystallinity of the coatings was studied. For this purpose, the samples were evaluated at room temperature (25 °C), 700 °C, 850 °C, and 1000 °C. The thin films present diffraction planes mainly at (118) (262), Figure 8, and with less intensity at (317), Figure 9, belonging to the orthorhombic crystalline system.

Figure 8 and Figure 9 show that for temperatures of 1000 °C, coatings Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ are amorphous. At 850 °C, coating Bi₃La₁Ti₃O₁₂ shows a crystalline structure in the crystallographic plane (317). However, at this same temperature, the thin films Bi_3.75La_0.25Ti₃O₁₂ are entirely amorphous. Due to the heating temperature, the crystalline phases were observed, possibly due to the corrosive salts’ evaporation. Consequently, the high-temperature corrosion influenced the crystallization orientation due to changes in the crystalline planes in the coating. Furthermore, the effect of the element lanthanum as a dopant was evident, since, at 1000 °C, the crystallization of the film usually occurs if thermal annealing is adopted, whereas the composite with a lower amount of lanthanum generated a high quantity of amorphous phase [34].

3.4. Transmission Electron Microscopy

Figure 10a,b are images obtained by transmission electron microscopy of the cross-section of the Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ films, respectively. The results confirm the growth of a columnar structure due to the intense atomic bombardment. In addition, good adhesion of the coatings to the substrate is also observed.

3.5. Scanning Electron Microscopy

Figure 11 shows the surface characterization via scanning electron microscopy of the Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ coatings as a function of the sintering temperature. In the Bi_3.75La_0.25Ti₃O₁₂ coatings, rough surfaces are evident due to the sputtering process when forming the coatings. The presence of pores of varied sizes is also evident. In the Bi₃La₁Ti₃O₁₂ coatings, the surface is more homogeneous, and the presence of pores is not evident.

Figure 11 shows the micrographs of the Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ films deposited on the steel. The micrograph represents the specimen with the highest lanthanum content, where deterioration is observed in the central and lower region. However, part of the coating was not affected, as seen in the upper part of the micrograph. For the Bi_3.75La_0.25Ti₃O₁₂ film, it can be observed that the percentage of lanthanum is 0.25. It can be seen that the films are heterogeneous; they present some discontinuities, such as pores (dark regions) [35]. Homogeneity is evidenced in the micrographs of the films with higher lanthanum content and higher intensity values in the X-ray diffraction. Unlike the films with lower lanthanum content, the micrographs of the films deposited under the same conditions but varying the amount of lanthanum on steel substrates also showed the same Miller indices. However, they decreased intensity, allowing greater amorphism for the system with lower lanthanum. Figure 11 shows the presence of pores and partial detachment of the coating at 1000 °C, so the coatings are distinguished as compromised. It is observed that the behavior against the attack generated by the electrolyte at the time of the evaluation of the corrosion resistance will be lower for the film with lower lanthanum content. As obtained in the electrochemical tests, these systems are 10 times more susceptible to corrosion damage at high temperatures [36].

The elemental chemical analysis was developed for the two types of coatings. The analyzed areas correspond to the micrographs observed in Figure 11. The results found are recorded in

Table 6. Chemical analysis for Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ coatings and the substrate at a temperature of 700 °C.

System	C	O	Na	Si	S	La	Ti	Cr	Mn	Fe	Ni	V	Bi
	(%)
Substrate	1	16	4	3	6	0	0	14	3	39	4	10	0
Bi3.75La0.25Ti3O12	1	15	1	2	1	6	18	7	1	10	5	8	25
Bi3La1Ti3O12	2	13	1	2	1	9	19	8	1	9	5	7	23

,

Table 7. Chemical analysis for Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ coatings and the substrate at a temperature of 850 °C.

System	C	O	Na	Si	S	La	Ti	Cr	Mn	Fe	Ni	V	Bi
	(%)
Substrate	1	22	4	1	2	0	0	4	1	49	5	16	0
Bi3.75La0.25Ti3O12	3	19	14	2	4	3	14	2	0	4	5	21	12
Bi3La1Ti3O12	4	18	12	1	5	4	16	1	0	7	1	19	12

and

Table 8. Chemical analysis for Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂ coatings and the substrate at a temperature of 1000 °C.

System	C	O	Na	Si	S	La	Ti	Cr	Mn	Fe	Ni	V	Bi
	(%)
Substrate	1	18	2	1	2	0	0	2	1	46	2	25	0
Bi3.75La0.25Ti3O12	1	22	16	1	2	2	12	0	0	2	1	29	12
Bi3La1Ti3O12	1	25	19	2	1	1	11	0	0	2	1	22	15

.

In the chemical analysis, elements such as Ti, Bi, and O are evident, which correspond to the films. In addition, elements such as the substrate’s Cr, Fe, and Ni components are also reported.

3.6. Energy Activation

In general, the estimate of the activation energy is determined by the Arrhenius equation. This study used three temperatures of 700 °C, 850 °C, and 1000 °C. Activation energies were 193 kJ/mol for the substrate, 69.8 kJ/mol for films of the Bi_3.75La_0.25Ti₃O₁₂ system, and 87.3 kJ/mol for Bi₃La₁Ti₃O₁₂ coatings. The variation in the activation energy values results is related to the effect of the concentrations of lanthanum that make up the coatings.

4. Conclusions

The corrosion rate was determined to have a direct relationship with the temperature because the mechanisms induced by the Na₂SO₄ and V₂O₅ salts have generated condensation. Therefore, increasing the temperature increases the corrosion current density in the thin films of Bi_3.75La_0.25Ti₃O₁₂ and Bi₃La₁Ti₃O₁₂. Furthermore, by comparing the two compounds, it is determined that the increase in La alters the positive acid character, thus decreasing the dissolution of the oxides and increasing the protection.

The potentiodynamic polarization curves for the coatings determined minimal damage for a temperature of 700 °C—the damage increases as a function of the sintering temperature of the thin films. However, at 1000 °C, the damage is catastrophic due to the increased corrosion current density. In addition, the existence of general dissolution in the anodic region was determined in the two systems studied.

The films have crystalline characteristics that are preserved up to 850 °C. However, for the coating with higher lanthanum content, a structure is obtained up to 1000 °C, owing to the generated phases. Once cooled, the coating is affected. Amorphous processes are generated due to the increase in corrosion current density.

Homogeneity is evidenced in the micrographs of the films with higher lanthanum content and higher intensity values in the X-ray diffraction.

The results of the activation energy values are related to the effect of the concentrations of lanthanum that make up the coatings.

The results allowed us to establish that the Bi₃La₁Ti₃O₁₂ coatings evaluated at temperatures of 700 °C, 850 °C, and 1000 °C exhibit better stability at relatively high temperatures.