Corrosion Resistance of Fe-Cr-Al Intermetallic Coatings Obtained by Aluminizing

: The growing interest in intermetallic and metal–intermetallic materials and coatings is based on the number of favorable properties they possess, primarily mechanical. However, the lack of data on their corrosion resistance has largely limited their scope of application. In this study, the corrosion destruction mechanisms of coatings formed on substrates made of AISI 321 steel and Aluchrom W (fechralloy) were investigated. The coatings were created by alloying in an aluminum melt followed by diffusion annealing to form the ultimate intermetallic structure. Corrosion resistance was studied under cyclic exposure to a humid marine atmosphere simulator and potentiostatic tests in an aqueous NaCl solution. Corrosion destruction parameters were determined, and mechanisms for each type of coating were revealed. The conducted studies allowed us to determine the electrochemical parameters of the corrosion destruction process and its mechanisms. It was shown that the corrosion rates during potentiostating for coatings on substrates Cr15Al5 and 12Cr18Ni10Ti differed by almost twofold. Two different mechanisms of corrosion are proposed. The ﬁrst is associated with the formation of Al 2 O 3 and MgO oxide ﬁlms, which at the initial stage protect only local areas of the coating surface on Cr15Al5. The second is determined by the diffusion of titanium atoms during annealing to the coating surface on a 12Cr18Ni10Ti steel substrate with the formation of TiC carbide at the grain boundaries.


Introduction
Due to their favorable performance properties at high temperatures, Fe-Cr-Al alloys (sometimes referred to as 'fechrals') are actively used as heat-resistant materials that ensure the stable operation of parts under high temperatures (up to 900-1100 • C) and corrosive environments (water vapor, exhaust gases, etc.) typical in the operation of internal combustion engines and power plants, especially when using biofuels [1][2][3][4][5].According to the study conducted by Zhang et al. [1], such a coating successfully withstood heating to a temperature of 900 • C in air for 4 h.As a result, the microstructure of the composite coating became three-layered.The first layer consisted of a solid solution of Cr and Fe in aluminum and an intermetallic compound FeAl.The second layer represented only one phase of the aluminide.The third layer in the composition corresponded to the original Fe-Cr-Al coating.The main factor determining the heat resistance of fechrals is the content of aluminum.Its presence ensures the formation of a dense and inert corundum (Al 2 O 3 ) film on the surface of the material.Engkvist et al. [2] used modern analysis methods (SEM-scanning electron microscopy, EDX-energy-dispersive X-ray spectroscopy, TEM-transmission electron microscopy, XRD-X-ray diffraction, AES-atomic emission spectroscopy, and SIMS-secondary-ion mass spectrometry) to investigate the structure of the coating formed on the surface of an FeCrAl alloy under exposure to dry and wet O 2 for 72 h at a temperature of 900-1100 • C. The oxide films in all oxidation variants were two-layered α-Al 2 O 3 Metals 2023, 13, 1883 2 of 13 flakes with an upper layer of equiaxed grains and a lower layer containing elongated grains.According to the authors, these two layers were separated by a zone rich in chromium formed from the oxide present on the alloy surface even before exposure to oxygen.
An increase in the aluminum content above 5 wt.% leads to a significant level of brittleness in the alloy.This occurs due to growth in the number of brittle aluminum intermetallic compound inclusions with iron and chromium in the structure.Thus, it creates problems during pressure treatment and limits the possible areas of application [6,7].
It was found in [8] that it is possible to increase the service life of products operating under corrosive environments at elevated temperatures by forming an intermetallic Fe-Cr-Al coating on their surface using the thermal diffusion aluminizing method.
Currently, different methods of applying aluminum to the surface of various alloys are actively used to increase resistance to oxidation and corrosion, including powder spraying, electrolytic aluminizing, aluminizing by hot dipping, and aluminizing in a gaseous atmosphere [9][10][11][12][13].The method of hot dipping is especially relevant because of its simplicity, its high efficiency, the absence of a need for expensive equipment, and its ability to process large products.Its application allows one to form a uniform defectless coating over the entire surface of the product on the basis of intermetallic Fe 2 Al 5 , Al 3 Fe, and Al 7 Cr phases dispersed in an aluminum matrix [14].
The high heat resistance of coatings based on iron aluminides [15,16] is determined by the possibility of forming a uniform oxide film of aluminum and certain alloying elements that prevents the diffusion of oxygen to the substrate at high temperatures [17][18][19].At the same time, the presence of aluminum in the coating contributes to the formation of Al 2 O 3 upon contact with the atmosphere, while the alloying components normalize its α-modification, which is characterized by the highest density and resistance at high temperatures [20][21][22].Bülbül et al. [20] investigated the corrosion resistance of high-alloy steels containing 0.4% C, 29% Cr, and 4-20% Ni in 10% HCl solutions.Their studies showed the relatively active destruction of the surface and subsurface layers of these steels both in the as-cast condition and after heat treatment.The weight loss values of the alloys exposed to age heat treatment decreased with an increasing nickel content.The nickel content increase from 4 to 20% led to a decrease in the corrosion rate of about 30-fold in the heat-treated steels but practically did not affect the corrosion resistance of the steel castings.
It has been established that austenitic corrosion-resistant chromium-nickel steels (for instance, AISI 321) are widely used for the manufacture of equipment to be employed in aggressive environments.
However, the tendency for local and intercrystalline corrosion under certain conditions makes the issue of increasing the corrosion resistance of such steels extremely relevant, especially when exposed to environments containing products from the processing or combustion of hydrocarbons and sulfur compounds or a humid seaside atmosphere [23,24].Pugacheva et al. [8] also showed the possibility of applying aluminide coatings to the surface of austenitic corrosion-resistant steel (the Russian analogue of AISI 304 steel).Aluminum diffusion was carried out in a vacuum chamber at temperatures of 1000-1100 • C, resulting in a coating with a thickness of about 100-160 µm.The outer coating zone comprised ordered B2-phase (Fe,Cr,Ni)Al with a thickness of about 20 µm, in which the aluminum content reached 25-28 wt.%.This was followed by a nickel-enriched thin layer in the form of a mechanical mixture of (Fe,Ni)Al and (Fe,Ni) 3 Al phases.The third zone from the surface consisted of an α-solid solution of Al and Ni in a body-centered cubic lattice of Fe(Cr) and dispersed particles of (Fe,Cr,Ni)Al.The thickness of this zone was 80-120 µm.
One of the most frequently applied ways to increase corrosion resistance is coating.Nevertheless, currently published data on the testing of Fe-Al intermetallic coatings under exposure to a humid marine atmosphere indicate their relatively high rate of corrosion damage [25], which makes the search for means to increase the corrosion resistance of these coatings extremely relevant.One of the most promising ways to increase the corrosion resistance of such coatings may be alloying, due to the diffusion of the necessary elements from the substrate material.However, the issue of coating resistance based on alloyed iron aluminides under exposure to electrochemical corrosion in humid atmospheres or salt solutions has barely been considered.There are also no data on the electrochemical parameters of the corrosion processes, soluble phases, and their electrode potentials.
The purpose of this work was to study corrosion resistance of Fe-Cr-Al intermetallic coatings obtained by aluminizing of Aluchrom W (fechralloy) and austenitic 12CrNi10Ti chromium-nickel steel (the Russian analogue of AISI 321 steel) by hot dipping in an aluminum melt followed by diffusion annealing and to determine the electrochemical potentials of the phases forming the coating.

Materials under Study
The formation of intermetallic coatings on the surface of the substrates was carried out in two stages.The first stage provided the formation of a thin layer of aluminum on the surface of the substrate by hot dipping aluminization.At the second stage, diffusion heat treatment was performed to form the ultimate intermetallic structure.As substrates, plates made of austenitic 12CrNi10Ti steel and Cr15Al5 fechralloy (the Russian analogue of Aluchrom W fechralloy) were used.The chemical composition of the substrate before aluminizing was monitored by means of an optical emission PMI-Master Smart UVR spectrometer (firm WASAG group Oxford Instruments, Wiesbaden, Germany) in a Spark regime with use of argon.The chemical composition of the alloys determined by optical emission analysis is given in Table 1.The plates were aluminized by hot dipping in technically pure aluminum AD1 (the analogue of ENAW-1235).The chemical composition of pure aluminum AD1 is given in Table 1.Before aluminizing, the surface of the samples was subjected to abrasive treatment with grinding skins with a grain size of abrasive (silicon carbide) 2000 mesh, followed by ultrasonic cleaning in a bath with a mixture of acetone and ethanol.The aluminum melt for aluminizing was heated in ceramic crucibles in a SNOL 8.2/1100 chamber furnace of resistance (Snol-Term, Tver, Russia) to a temperature of 700-780 • C. The duration of dipping the substrates into an aluminum melt was 5 min.In order to obtain an intermetallic coating, diffusion annealing was carried out in the same furnace at a temperature of 1050 • C for 24 h.

Applied Research Methods
The accelerated corrosion tests in an environment simulating the effects of a humid marine atmosphere and calculation of corrosion parameters were conducted in accordance with ISO 11130-2017 "Corrosion of metals and alloys-Alternate immersion test in salt solution" with a cyclic immersion in an electrolyte (a 3% sodium chloride solution with hydrogen peroxide) at a temperature of 18-25 • C on a specially designed computerized test bench with an hourly repeating cycle with immersion in the solution for 10 min with a subsequent extraction in air for 50 min.Each sample was immersed in an individual vessel with a volume of 800 mL, in which the concentration of the solution was kept constant.The tests were carried out for 90 days.Every 15 days, the samples were extracted, cleaned mechanically from corrosion products with a brush, rinsed with deionized water, and dried.After these actions, the change in their mass was measured.Vibra HT 224 RCE electronic scales (Vibra, Tokyo, Japan) with an accuracy of 0.001 g were used to weigh the samples.To eliminate the measurement error, the tests were carried out on a series of samples of 20 pieces.The results were averaged.
Material corrosion resistance was determined from the weight loss and corrosion rate of the samples.
The corrosion rate was calculated using the following equation: where ∆m is the weight loss; τ is the exposure time of the corrosion medium.
The weight loss was calculated using the following equation: where m 0 is the weight of the sample before testing, g; m 1 is the weight of the sample after the cycle of the tests and the removal of corrosion products, g; S is the surface area of the sample, m 2 .
To study the microstructure of the metallographic rectangular sections with a size of 10 × 20 mm and a thickness of 5-7 mm, optical microscopy at ×50-500 magnifications on a modular motorized optical Olympus BX-61 microscope (Olympus, Tokyo, Japan) was used.The depth of corrosion damage was determined by the optical method by changing the depth of field.In order to obtain a three-dimensional surface of the corroding coating, the samples were photographed with an optical microscope using a DP-12 digital camera (Olympus, Tokyo, Japan), and then the package of the digital photographs was processed using the free computer program CombineZP v.4.0.0.2 (by Alan Hadley, Worthing, UK).The samples were etched due to exposure to a corrosive environment.
Potentiostating was carried out using the P-40X potentiostat-galvanostat with an FRA-24M electrochemical impedance measurement module (Electrochemical instruments, Chernogolovka, Russia).The measuring cell was assembled according to a three-electrode scheme (Figure 1) using an ASr-10105 silver-chloride reference electrode and an A-5S graphite working electrode.The potential range was from −2000 to +2000 mV.The samples were immersed in a 3% aqueous NaCl solution on the nichrome current leads with a diameter of 3 mm.A part of the current lead and the spot where it was attached to the sample were covered with a non-conductive protective varnish.The environmental temperature was 20-25 • C. Based on the experimental results, the polarization curves were plotted both in the ordinary coordinates and the Tafel coordinates, and the potentials of the phases involved in electrochemical processes were determined.
After annealing and corrosion tests, the structure and local chemical composition of the coatings were observed by scanning two-beam electron-ion microscopy (Versa3D, DualBeam system, FEI Company, Eindhoven, The Netherlands) in conjunction with an energy-dispersive spectrometer (INCA X-Max, Oxford Instruments).The structure was assessed at magnifications from ×132 to ×1000 with potential of the electron gun of 20 kV.The average error of energy dispersion analysis in determining the chemical composition was about 5%.
A Bruker D8 Advance Eco X-ray diffractometer (Bruker AXS GmbH, Karlsruhe, Germany) was used to compare the results of the study when using optical and electron microscopy with EDS analysis and the results of the study of the phase composition of the coatings.
The interpretation of the analysis of the phases was carried out using the licensed versions of the programs Diffrac.Eva v.6 and Diffrac.Topas v.4.2 and the licensed database PDF-2.
and an A-5S graphite working electrode.The potential range was from −2000 to +2000 mV.The samples were immersed in a 3% aqueous NaCl solution on the nichrome current leads with a diameter of 3 mm.A part of the current lead and the spot where it was attached to the sample were covered with a non-conductive protective varnish.The environmental temperature was 20-25 °C.Based on the experimental results, the polarization curves were plotted both in the ordinary coordinates and the Tafel coordinates, and the potentials of the phases involved in electrochemical processes were determined.

Results and Discussion
The intermetallic coatings with a thickness of 500-700 µm were formed on the surface of the plates after heat treatment of the aluminized samples at a temperature of 1050 • C for 24 h.It was previously shown by the authors in [25] that the coatings on a fechral surface had a composition differing from the surface into the depth: FeAl(Cr) → Fe 3 Al(Cr) → Fe(Al,Cr).The coatings on 12Cr18Ni10Ti steel have a layered structure after annealing.The upper layer is a solid solution based on FeAl(Cr,Ni) intermetallic, and the lower layer is an α-Fe(Al,Cr,Ni) solid solution with a gradual decrease in the content of aluminum towards 12Cr18Ni10Ti steel.The coating surface based on the fechral had two distinctive areas (Figure 2a): a light base, which was examined by backscattered electrons, and dark spots in disorder.The energy dispersion analysis in combination with XRD established (Table 2) that the light base is the FeAl(Cr) phase, which corresponds to the data in [25], whereas the dark spots are the same phase covered with MgO and Al 2 O 3 oxide films.Thus, the oxide films protect only a part of the surface of the intermetallic coating from corrosion.After annealing and corrosion tests, the structure and local chemical composition o the coatings were observed by scanning two-beam electron-ion microscopy (Versa3D DualBeam system, FEI Company, Eindhoven, Netherlands) in conjunction with an en ergy-dispersive spectrometer (INCA X-Max, Oxford Instruments).The structure was as sessed at magnifications from ×132 to ×1000 with potential of the electron gun of 20 kV The average error of energy dispersion analysis in determining the chemical composition was about 5%.
A Bruker D8 Advance Eco X-ray diffractometer (Bruker AXS GmbH, Karlsruhe Germany) was used to compare the results of the study when using optical and electron microscopy with EDS analysis and the results of the study of the phase composition o the coatings.
The interpretation of the analysis of the phases was carried out using the licensed versions of the programs Diffrac.Eva v.6 and Diffrac.Topas v.4.2 and the licensed data base PDF-2.

Results and Discussion
The intermetallic coatings with a thickness of 500-700 µm were formed on the sur face of the plates after heat treatment of the aluminized samples at a temperature of 1050 °C for 24 h.It was previously shown by the authors in [25] that the coatings on a fechra surface had a composition differing from the surface into the depth FeAl(Cr)→Fe3Al(Cr)→Fe(Al,Cr).The coatings on 12Cr18Ni10Ti steel have a layered structure after annealing.The upper layer is a solid solution based on FeAl(Cr,Ni) in termetallic, and the lower layer is an α-Fe(Al,Cr,Ni) solid solution with a gradual de crease in the content of aluminum towards 12Cr18Ni10Ti steel.The coating surface based on the fechral had two distinctive areas (Figure 2a): a light base, which was examined by backscattered electrons, and dark spots in disorder.The energy dispersion analysis in combination with XRD established (Table 2) that the light base is the FeAl(Cr) phase which corresponds to the data in [25], whereas the dark spots are the same phase covered with MgO and Al2O3 oxide films.Thus, the oxide films protect only a part of the surface of the intermetallic coating from corrosion.The basis of the coating surface on the plates made of 12Cr18Ni10Ti steel is an FeAl(Cr,Ni) intermetallic compound (Figure 2b, Table 2, point 4).The separate light in clusions containing about 35% titanium (Figure 2b, Table 2, point 3) are most likely to be TiC carbides on the FeAl(Cr) surface.Indirectly, this is confirmed by a much lighter shade  The basis of the coating surface on the plates made of 12Cr18Ni10Ti steel is an FeAl(Cr,Ni) intermetallic compound (Figure 2b, Table 2, point 4).The separate light inclusions containing about 35% titanium (Figure 2b, Table 2, point 3) are most likely to be TiC carbides on the FeAl(Cr) surface.Indirectly, this is confirmed by a much lighter shade of inclusions due to the presence of light carbon atoms in this phase.The Fe 2 Al 5 intermetallic compound, which is usually formed in aluminized layers, is absent in the structure of the coatings under study due to the dissolution of aluminum in the substrate during prolonged diffusion annealing.As a result, there are no zones with a ratio of iron and aluminum sufficient for the formation of this intermetallic compound.The cross-sections of the coatings in the initial state and after corrosion were given by the authors in their previous work [25].In this state, the samples were subjected to the tests described below.
The tests simulating the effects of a humid seaside atmosphere conducted for 90 days showed that in this environment the corrosion resistance of the coating on the plates made of 12Cr18Ni10Ti steel is 2.5 times higher (the corrosion rate was 0.0001 (g/m 2 )/day) than on the plates made of Cr15Al5 fechralloy (the corrosion rate was 0.0025 (g/m 2 )/day) [25].The corrosion resistance of austenitic 12Cr18Ni10Ti steel both in the initial state and after coating remained quite high throughout the entire test period.The nature of the corrosion rate increase was close to linear, which makes it possible to predict with high accuracy the service life of parts and structures made of this material in an aggressive environment.The samples from fechral with an aluminide coating had a sharp increase in the corrosion rate after 45 days of exposure to the electrolyte.The microstructure study of the aluminized fechral near-surface layers after corrosion tests showed that the corrosion process developed from the surface foci deep into the material along the grain boundaries.Such a significant difference in corrosion resistance is explained by distinctions in the mechanisms of corrosion destruction caused by the structure of the coatings after aluminizing and subsequent heat treatment.Fechral is characterized by local corrosion areas with penetration to a depth of up to 1000 µm.The lesion of the coating on the 12Cr18Ni10Ti steel plate is localized in a layer up to 200 µm deep.The obtained data on the corrosion rate make it possible to accurately predict the lifetime of the elements of equipment and structures coated on aluminides operating under exposure to a humid coastal atmosphere (on coastal structures, offshore platforms, and ships).
The potentiostatic tests in the voltage range from −2000 to +2000 mV in a 3% aqueous NaCl solution were carried out to reveal more precisely the corrosion mechanisms of the coatings under consideration and determine the electrochemical parameters of the process.The obtained polarization curves (Figure 3) show that the corrosion processes of both types of the coatings proceeded almost identically.The passivation and repassivation areas almost coincide, and the coating on 12Cr18Ni10Ti steel shows a much higher resistance in the area of active dissolution.The determination of the destruction rate by weight loss also showed that during the test the coating on the fechral was destroyed faster (0.53 g/s) than the coating on the steel (0.27 g/s), which is noticeably higher than when tested by a cyclic immersion in the same environment [25].Since the dependence of the current on overvoltage is exponential in a wide range of potentials, to determine the key electrochemical parameters of corrosion [26], the polarization curves for the full cycle were rearranged in the semi-logarithmic (Tafel) coordinates E(log(j)) (Figure 4).This made it possible to identify the exchange current log(j₀) and the equilibrium potential E₀ of the coating dissolution by plotting tangent lines at characteristic points.The equilibrium potential values for both steel plates have very close values of ~400 mV, which indicates the predominant dissolution of the same phase.Since the dependence of the current on overvoltage is exponential in a wide range of potentials, to determine the key electrochemical parameters of corrosion [26], the polarization curves for the full cycle were rearranged in the semi-logarithmic (Tafel) coordinates E(log(j)) (Figure 4).This made it possible to identify the exchange current log(j 0 ) and the equilibrium potential E 0 of the coating dissolution by plotting tangent lines at characteristic points.The equilibrium potential values for both steel plates have very close values of ~400 mV, which indicates the predominant dissolution of the same phase.
The structural and chemical composition analysis of the samples' coating surface made from Cr15Al5 fechralloy after potentiostating showed that there was a complete dissolution of the phases formed on the outer layers of the coating (Figure 5, Table 3, points 1, 2).At the same time, the corrosion damage reached a deeper Fe 3 Al(Cr) phase, and its dissolution signs were not detected.A large number of traces of the point corrosion damage indicates that between the FeAl(Cr) and Fe 3 Al(Cr) layers there is a transitional two-phase zone consisting of the mixture of these phases.The dissolution of the FeAl(Cr) grains going deeper formed pittings, which in some cases (Figure 5, Table 3, points 3, 4) reached the depth of the third layer (Fe(Al,Cr)) formed during diffusion while being annealed.Thus, the comprehensive study using potentiostating and the chemical analysis of the samples before and after the tests allowed us to determine the type of phase dissolving during electrochemical corrosion and its dissolution potential.This made it possible to create a cathodic protection system to prevent the destruction of intermetallic Fe-Cr-Al coatings obtained using the described technology.As shown, a potential at the level of ~0.4 V is needed.The structural and chemical composition analysis of the samples' coating surface made from Cr15Al5 fechralloy after potentiostating showed that there was a complete dissolution of the phases formed on the outer layers of the coating (Figure 5, Table 3, points 1, 2).At the same time, the corrosion damage reached a deeper Fe₃Al(Cr) phase, and its dissolution signs were not detected.A large number of traces of the point corrosion damage indicates that between the FeAl(Cr) and Fe₃Al(Cr) layers there is a transitional two-phase zone consisting of the mixture of these phases.The dissolution of the FeAl(Cr) grains going deeper formed pittings, which in some cases (Figure 5, Table 3, points 3, 4) reached the depth of the third layer (Fe(Al,Cr)) formed during diffusion while being annealed.Thus, the comprehensive study using potentiostating and the chemical analysis of the samples before and after the tests allowed us to determine the type of phase dissolving during electrochemical corrosion and its dissolution potential.This made it possible to create a cathodic protection system to prevent the destruction of intermetallic Fe-Cr-Al coatings obtained using the described technology.As shown, a potential at the level of ~0.4 V is needed.The structural and chemical composition analysis of the samples' coating surface made from Cr15Al5 fechralloy after potentiostating showed that there was a complete dissolution of the phases formed on the outer layers of the coating (Figure 5, Table 3, points 1, 2).At the same time, the corrosion damage reached a deeper Fe₃Al(Cr) phase, and its dissolution signs were not detected.A large number of traces of the point corrosion damage indicates that between the FeAl(Cr) and Fe₃Al(Cr) layers there is a transitional two-phase zone consisting of the mixture of these phases.The dissolution of the FeAl(Cr) grains going deeper formed pittings, which in some cases (Figure 5, Table 3, points 3, 4) reached the depth of the third layer (Fe(Al,Cr)) formed during diffusion while being annealed.Thus, the comprehensive study using potentiostating and the chemical analysis of the samples before and after the tests allowed us to determine the type of phase dissolving during electrochemical corrosion and its dissolution potential.This made it possible to create a cathodic protection system to prevent the destruction of intermetallic Fe-Cr-Al coatings obtained using the described technology.As shown, a potential at the level of ~0.4 V is needed.
Furthermore, the combination of data on the microstructure and chemical composition of the coating in the initial state and after corrosion tests, as well as the data from the previous studies [25], allowed us to describe the probable mechanism of corrosion destruction of these coatings.Furthermore, the combination of data on the microstructure and chemical composition of the coating in the initial state and after corrosion tests, as well as the data from the previous studies [25], allowed us to describe the probable mechanism of corrosion destruction of these coatings.
The hypothetical mechanism of corrosion destruction of the coating on the fechral is shown in Figure 6.At the initial stage, the dissolution of the Fe-Al phase began on the areas not covered with an Al 2 O 3 oxide film, and then the process continued both over the entire surface and deep into it.Penetrating into the inner layers, the corrosion process stopped, when the deeper phase of Fe 3 Al came into contact with the environment.The data analysis on the structure and chemical composition of the coating surface on 12Cr18Ni10Ti after potentiostatic tests (Figure 7, Table 4) showed that the intense dissolution of the FeAl(Cr,Ni) phase located on the coating surface occurred during potentiostating, as in the case of the coating on a fechral substrate.The penetration of corrosion lesions along the FeAl grains right up to a diffusion layer consisting of Fe(Al,Cr) was recorded in some cases.At the same time, the corrosion process did not develop further.
The data analysis on the structure and chemical composition of the coating surface on 12Cr18Ni10Ti after potentiostatic tests (Figure 7, Table 4) showed that the intense dissolution of the FeAl(Cr,Ni) phase located on the coating surface occurred during potentiostating, as in the case of the coating on a fechral substrate.The data analysis on the structure and chemical composition of the coating surface on 12Cr18Ni10Ti after potentiostatic tests (Figure 7, Table 4) showed that the intense dissolution of the FeAl(Cr,Ni) phase located on the coating surface occurred during potentiostating, as in the case of the coating on a fechral substrate.However, the mechanisms of corrosion destruction differ significantly.This difference is due to the fact that the dissolution of the FeAl(Cr,Ni) phase was accompanied by intercrystalline corrosion at the sites of the TiC inclusion formation with precipitation of the latter.Visually, there are a large number of pittings along the boundaries of the main phase (Fe₃Al(Cr,Ni)) grains, which have come to the surface.The proposed scheme of the   However, the mechanisms of corrosion destruction differ significantly.This difference is due to the fact that the dissolution of the FeAl(Cr,Ni) phase was accompanied by intercrystalline corrosion at the sites of the TiC inclusion formation with precipitation of the latter.Visually, there are a large number of pittings along the boundaries of the main phase (Fe 3 Al(Cr,Ni)) grains, which have come to the surface.The proposed scheme of the corrosion destruction mechanism is shown in Figure 8.

Conclusions
The conducted studies have shown that intermetallic Fe-Cr-Al coatings exhibit high corrosion resistance and withstand long-term tests successfully under exposure to a humid marine atmosphere.The corrosion rate does not exceed 0.0003 (g/m 2 )/day.

Conclusions
The conducted studies have shown that intermetallic Fe-Cr-Al coatings exhibit high corrosion resistance and withstand long-term tests successfully under exposure to a humid marine atmosphere.The corrosion rate does not exceed 0.0003 (g/m 2 )/day.
It was found that the corrosion rate of the aluminized coating formed on 12Cr18Ni10Ti steel is twofold lower than that of the coating on the plate made of Cr15Al5 fechralloy.This is explained by the various mechanisms of corrosion damage.The corrosion destruction of the coating on the fechral plate to a depth of more than 1000 µm occurs due to intercrystalline corrosion.The corrosion destruction of the aluminized coating on steel 12Cr18Ni10Ti is concentrated in the surface layers of the coating and has a focal nature.The corrosion process stops when the layer of the Fe(Al,Cr,Ni) solid solution containing about 18% chromium is reached.
The potentiostatic tests conducted in a corrosive environment of a 3% aqueous sodium chloride solution combined with the electron microscopy and energy dispersion analysis of the coatings allowed us to determine the key electrochemical parameters of the corrosion destruction process and to reveal its mechanisms.FeAl intermetallic compounds actively dissolve when exposed to an aggressive environment on both types of coatings.The dissolution potential of this phase is ~−400 mV.The current density for the beginning of the dissolution process for the coating on the plate made of 12Cr18Ni10Ti steel is lower than on the plate made of Cr15Al5 fechralloy.This indicates a more intensive course of the corrosion process in the latter case.The corrosion rate during potentiostating for the coating on the 12Cr18Ni10Ti substrate amounted to 0.27 g/s, and on the Cr15Al5 substrate it was 0.53 g/s.
The difference in the corrosion rate is explained by the difference in the mechanism of its development.
It has been established that the surface of the coating on the Cr15Al5 substrate consists of an FeAl(Cr) intermetallic compound.The formation of the zones with a high magnesium content led to the formation of Al 2 O 3 and MgO oxide films on the surface, which protected only local areas of the surface at the initial stage.The corrosion process on the unprotected areas of FeAl alloy developed intensively depthward reaching the layers of Fe 3 Al(Cr) and Fe(Al,Cr) located below.At the same time, corrosion penetrated under the oxide films and thus gradually captured the entire surface of the coating.
The mechanism of the coating destruction on the substrate made of 12Cr18Ni10Ti steel had a different nature.During annealing, a large number of titanium atoms, which are most likely to form TiC at the grain boundaries, diffused to the surface of the coating.The difference in the electrochemical potentials between the matrix of FeAl(Cr,Ni) and TiC carbides initiated the process of intercrystalline corrosion with dissolution of the matrix around the carbide inclusions with further precipitation of the latter.This type of destruction is evidenced by the structure of the coating after potentiostating, which represents a base of Fe 3 Al(Cr,Ni) intermetallic with pores remaining after precipitation of carbide inclusions.
The practical significance of the results obtained consists in the possibility of predicting the lifetime of coatings based on Fe-Cr-Al aluminides when operating under exposure to a humid marine atmosphere, which is known to be one of the most common and active corrosive environments.

Figure 2 .
Figure 2. The initial structure of the surface during the study by backscattered electrons and the areas for determining the chemical composition of coatings on the plates made of Cr15Al5 fe chralloy (a) and 12Cr18Ni10Ti steel (b).

Figure 2 .
Figure 2. The initial structure of the surface during the study by backscattered electrons and the areas for determining the chemical composition of coatings on the plates made of Cr15Al5 fechralloy (a) and 12Cr18Ni10Ti steel (b).

Figure 3 .
Figure 3. Polarization curves j(E) for the coating samples on the plates made of Cr15Al5 fechralloy (a) and 12Cr18Ni10Ti steel (b).

Figure 3 .
Figure 3. Polarization curves j(E) for the coating samples on the plates made of Cr15Al5 fechralloy (a) and 12Cr18Ni10Ti steel (b).

Figure 4 .
Figure 4. Tafel diagrams for the coatings on the plates made of Cr15Al5 fechralloy (a) and 12Cr18Ni10Ti steel (b), log(j 0 )-exchange current, E 0 -equilibrium potential.

Figure 5 .
Figure 5.The microstructure of the corrosion damage area and the determination points (1−4) of chemical composition of the corrosion damage areas of the coating on Cr15Al5 fechralloy.Figure 5.The microstructure of the corrosion damage area and the determination points (1-4) of chemical composition of the corrosion damage areas of the coating on Cr15Al5 fechralloy.

Figure 5 .
Figure 5.The microstructure of the corrosion damage area and the determination points (1−4) of chemical composition of the corrosion damage areas of the coating on Cr15Al5 fechralloy.Figure 5.The microstructure of the corrosion damage area and the determination points (1-4) of chemical composition of the corrosion damage areas of the coating on Cr15Al5 fechralloy.

Figure 6 .
Figure 6.The mechanism of corrosion destruction of Cr15Al5 fechralloy: the initial state after aluminizing and annealing (a), the beginning of corrosion destruction and the precipitation of FeAl (b), the precipitation of Fe3Al (c), the three-dimensional image of the surface of the coating after corrosion obtained using the CombineZP v.4.0.0.2 software (d).

Figure 6 .
Figure 6.The mechanism of corrosion destruction of Cr15Al5 fechralloy: the initial state after aluminizing and annealing (a), the beginning of corrosion destruction and the precipitation of FeAl (b), the precipitation of Fe 3 Al (c), the three-dimensional image of the surface of the coating after corrosion obtained using the CombineZP v.4.0.0.2 software (d).

Figure 6 .
Figure 6.The mechanism of corrosion destruction of Cr15Al5 fechralloy: the initial state after aluminizing and annealing (a), the beginning of corrosion destruction and the precipitation of FeAl (b), the precipitation of Fe3Al (c), the three-dimensional image of the surface of the coating after corrosion obtained using the CombineZP v.4.0.0.2 software (d).

Figure 7 .
Figure 7.The microstructure of the coating surface and the determination areas (а) and point (b) of chemical composition on 12Cr18Ni10Ti steel after corrosion tests.

Figure 7 .
Figure 7.The microstructure of the coating surface and the determination areas (a) and point (b) of chemical composition on 12Cr18Ni10Ti steel after corrosion tests.

Table 4 .
Chemical composition of the surface of the coating on 12Cr18Ni10Ti steel in the initial state and after corrosion tests.

Figure 8 .
Figure 8.The possible mechanism of corrosion destruction of 12Cr18Ni10Ti: the initial state after annealing (a), the beginning of corrosion destruction accompanied by the precipitation of FeAl and intercrystalline corrosion at the sites of the TiC inclusion formation (b), the precipitation of Fe₃Al(Cr,Ti) (c), the three-dimensional image of the surface of the coating after corrosion obtained using the CombineZP v.4.0.0.2 software (d).

Figure 8 .
Figure 8.The possible mechanism of corrosion destruction of 12Cr18Ni10Ti: the initial state after annealing (a), the beginning of corrosion destruction accompanied by the precipitation of FeAl and intercrystalline corrosion at the sites of the TiC inclusion formation (b), the precipitation of Fe 3 Al(Cr,Ti) (c), the three-dimensional image of the surface of the coating after corrosion obtained using the CombineZP v.4.0.0.2 software (d).

Table 1 .
Chemical composition of the used alloys (% at.).

Table 2 .
Chemical composition of the surface of the coating in the initial state in Figure2.

Table 3 .
Chemical composition of the corrosion damage areas of the coating on Cr15Al5 fechralloy after corrosion tests.

Table 4 .
Chemical composition of the surface of the coating on 12Cr18Ni10Ti steel in the initial state and after corrosion tests.