Hot Corrosion Behavior of Inconel 625 in Na₂SO₄ and V₂O₅ Molten Salt System

Abstract

This study aimed to examine the corrosion behavior of Inconel 625 in a molten salt system of sodium sulfate and vanadium pentoxide at varying temperatures and durations. The corrosion products, microstructure, and element distribution of hot extruded Inconel in Na₂SO₄ and V₂O₅ molten salt systems were analyzed using X-ray diffraction (XRD), scanning electron microscopy (SEM), and energy-dispersive spectroscopy (EDS) analyses. This study demonstrates that corrosion of the alloy increases with time at a constant temperature. During the initial stage of corrosion, the surface of the alloy is primarily composed of a dense oxide layer consisting of Cr₂O₃ and NiO. However, after exposure to the salt bath for 24 h, a chemical reaction occurs between the alloy and vanadium (V), resulting in the formation of CrVO₄ and Ni₃V₂O₈. Furthermore, the intrusion of sulfur (S) element into the matrix leads to the formation of internal sulfides, including Ni-, Cr-, and Mo-based sulfides, which accelerate intergranular and intracrystalline corrosion. As the corrosion temperature rises, the surface microstructure of the corrosion layer transforms from powder to salt particles and then to massive particles. The corrosion products exhibit a clear stratification, while the alloy undergoes simultaneous oxidation and vulcanization processes.

1. Introduction

Since the end of the 20th century, research studies on 700 °C, supercritical, oil-fired power generation technology have been continuously conducted. The goal is to build 37.5 MPa/700 °C advanced ultra-supercritical (A-USC) demonstration power stations to improve thermal efficiency. A 700 °C class A-USC thermal power unit requires high-temperature materials with good high-temperature creep strength, excellent resistance to steam oxidation on the inner wall of the tube, and excellent processing performance [1]. Inconel 625 has good corrosion resistance and oxidation resistance in high-temperature and high-pressure environments; therefore, it has been considered to be one of the excellent candidate materials for the key components of advanced supercritical boiler superheaters and reheaters [2,3,4,5,6]. However, advanced supercritical oil-fired boilers usually use high-sulfur and high-vanadium crude oil as fuel, and the sulfur and vanadium in the fuel will cause a high-temperature corrosion of the boiler [7,8,9,10]. In order to simulate the service of Inconel 625 in high-sulfur and high-vanadium corrosion environments at high temperatures, a large number of studies [11,12,13,14] have been carried out, which have mainly focused on the influence of corrosive media and temperature on the corrosion behavior; these studies have proposed corrosion models such as the “acid-base melting mechanism” and “vulcanization mechanism”.

Fresnillo et al. [15,16] found that a thin chromium-based oxide layer was formed on the surface of the alloy after long-term oxidation in Ar-50% H₂O at 700–800 °C. The outside of the oxide layer is a Cr/Mn spinel, and the inside is SiO₂. The phases found in the bulk alloy after long-term exposure were mainly the needle-shaped δ-Ni₃(Nb, Mo) phase, μ-phase, and Si-rich η-M₆C carbide. Guo et al. [17,18] investigated the hot corrosion behavior of iron- and nickel-based alloys in mixed sulfuric acid molten salt and found that internal vulcanization–internal oxidation is the main damage form of hot corrosion for these alloys. When hot corrosion occurs in a nickel base alloy, a layer of Cr₂O₃ oxide film will first generate on the surface of the alloy; however, sulfur element will invade into the alloy through this layer of oxide film and cause internal sulfuration. Lu et al. [19] studied the hot corrosion behavior of a high-chromium nickel-based alloy in molten sulfate and found that the oxidation and sulfurization of the alloy were carried out simultaneously. With the increase in temperature, the depth and size of oxidation and inner sulfurization zone increased. Ye et al. [20] analyzed the corrosion mechanism of Inconel 625 in mixed salt sulfuric acid for 60 h using the salt layer method. The obvious internal oxidation and internal sulfide phenomenon was observed because the O and S elements had obvious erosion, which provided a theoretical basis for the sulfate corrosion resistance of the nickel base alloy. However, the corrosion behavior in the later stage was not deeply explored due to the short experimental time.

There are currently two views on the mechanism of vanadium corrosion: a. Corrosion caused by the destruction of the oxide film on the metal surface by the molten vanadium with a low melting point: Tiwari [21] studied the chemical reaction between NaSO₄ and V₂O₅, which mainly produced low-melting vanadates and released SO₃ at the same time.

$${\mathrm{N}\mathrm{a}}_2{\mathrm{S}\mathrm{O}}_4+{\mathrm{V}}_2{\mathrm{O}}_5\to 2{\mathrm{N}\mathrm{a}\mathrm{V}\mathrm{O}}_3+{\mathrm{S}\mathrm{O}}_3$$

(1)

When the ratio of Na/V is different, the products are also different:

$$2{\mathrm{N}\mathrm{a}}_2{\mathrm{SO}}_4+4{\mathrm{V}}_2{\mathrm{O}}_5\to 2{\mathrm{N}\mathrm{a}\mathrm{V}\mathrm{O}}_3+{\mathrm{N}\mathrm{a}}_2\mathrm{O}·3{\mathrm{V}}_2{\mathrm{O}}_5+2{\mathrm{SO}}_3$$

(2)

$$2{\mathrm{N}\mathrm{a}}_2{\mathrm{S}\mathrm{O}}_4+9{\mathrm{V}}_2{\mathrm{O}}_5\to {\mathrm{Na}}_2\mathrm{O}·6{\mathrm{V}}_2{\mathrm{O}}_5+{\mathrm{N}\mathrm{a}}_2\mathrm{O}·3{\mathrm{V}}_2{\mathrm{O}}_5+2{\mathrm{S}\mathrm{O}}_3$$

(3)

Foster et al. [22] Studied the Na₂SO₄–V₂O₅ system theoretically from the perspective of phase equilibrium. The results show that the potentially corrosive substance in the putty are NaVO₃ and Na₂O·3V₂O₅; when the molar ratio of Na/V is 1:6, a highly corrosive vanadate Na₂O·V₂O₄·5V₂O₅ will be formed. This compound will absorb oxygen during melting and release oxygen during solidification:

$$\frac{1}{2}{\mathrm{O}}_2+{\mathrm{N}\mathrm{a}}_2\mathrm{O}·{\mathrm{V}}_2{\mathrm{O}}_4·5{\mathrm{V}}_2{\mathrm{O}}_5\leftrightarrow {\mathrm{Na}}_2\mathrm{O}·6{\mathrm{V}}_2{\mathrm{O}}_5$$

(4)

These low-melting vanadium compounds can dissolve the oxides on the metal surface. For example, NaVO₃ can destroy the protective oxide layer on the surface of Ni–Cr alloy through the following reaction, thereby causing accelerated corrosion:

$$2{\mathrm{N}\mathrm{a}\mathrm{V}\mathrm{O}}_3+\mathrm{N}\mathrm{i}\mathrm{O}\to {\mathrm{N}\mathrm{a}}_2{\mathrm{N}\mathrm{i}\mathrm{O}}_2+{\mathrm{V}}_2{\mathrm{O}}_5$$

(5)

$$2{\mathrm{N}\mathrm{a}\mathrm{V}\mathrm{O}}_3+{\mathrm{C}\mathrm{r}}_2{\mathrm{O}}_3\to 2{\mathrm{C}\mathrm{r}\mathrm{V}\mathrm{O}}_4+{\mathrm{N}\mathrm{a}}_2\mathrm{O}$$

(6)

Oxygen absorbed by molten vanadate further accelerates metal oxidation:

Cunningham and Brasunas expanded the research results of Foster [23]. It was found that the most corrosive putty contains 15–20% Na₂SO₄. At the same time, the vanadate absorbs oxygen in an aerobic environment and releases oxygen during solidification, and the degree of corrosion is related to the oxygen absorption of the mixture. The reaction formula is as follows:

$${\mathrm{m}\mathrm{N}\mathrm{a}}_2\mathrm{O}·{\mathrm{n}\mathrm{V}}_2{\mathrm{O}}_5\leftrightarrow {\mathrm{m}\mathrm{N}\mathrm{a}}_2\mathrm{O}·\left(\mathrm{n}-\mathrm{p}\right){\mathrm{V}}_2{\mathrm{O}}_5·{\mathrm{p}\mathrm{V}}_2{\mathrm{O}}_4+\frac{1}{2}{\mathrm{p}\mathrm{O}}_2$$

(7)

Although nickel-based alloys have been deeply studied in a sulfur corrosion environment, there are few research studies on the high-temperature vulcanization and vanadium corrosion of nickel-based alloys. In this work, in order to study the corrosion mechanism of Inconel 625 in a supercritical oil boiler, the corrosion behavior of Inconel 625 in a sodium sulfate and vanadium pentoxide molten salt system at different temperatures and times was characterized using X-ray diffraction (XRD) (BRKR Inc., Karlsruhe, Germany), scanning electron microscope (SEM) (JEOL Inc., Tokyo, Japan), and an energy-dispersive spectrometer (EDS) (JEOL Inc., JPN).

2. Experiment

2.1. Materials

The chemical composition of Inconel 625 is listed in

Table 1. Chemical composition of experimental alloy (wt.%).

C	Cr	Ni	Co	Mo	Al	Ti	Fe	Nb	Si	Mn	S	P	Cu
0.042	21.77	60.63	0.19	8.79	0.21	0.40	3.68	3.75	0.12	0.2	0.0006	0.006	0.06

. The specimens were first deformed by the following extrusion parameters: extrusion ratio 3.61, extrusion temperature 1163 °C, extrusion speed 45 mm/s, and lubrication with glass lubricant. Then, they were cut using wire electrical discharge machining to gain the size of 10 × 10 × 5 mm, and then the burrs, debris, and scratches on the surface of specimens were polished using 2000, 3000, and 5000 grit silicon carbide papers. Finally, the sample was washed with acetone and alcohol.

2.2. Corrosion Method and Weight Loss Measurement

The hot corrosion tests of polished samples were carried out using the crucible landfill method. First, Na₂SO₄ and V₂O₅ powders were mixed into corundum crucible at the proportion of 20% Na₂SO₄ and 80% V₂O₅. Then, the samples were immersed in the mixed molten salt powders. Subsequently, the crucible was heated in a box-type resistance furnace at 700 °C, 750 °C, and 800 °C for 24, 48, 72, 96, and 120 h, respectively. The specimens were taken out of the furnace and cooled to room temperature with the molten salt after the immersion tests. Vanadium compounds on the surface of the corroded specimens were rinsed with 10% sodium hydroxide (NaOH) aqueous solution [24]. The corrosion rate of the specimens was calculated using the weight loss method:

$$v_{loss}=\frac{m_0-m_1}{st}$$

(8)

where $v_{loss}$ is the corrosion rate when using the weight-loss method, measured in g/(cm² × h); m₀ is the initial mass of the specimen in grams; m₁ is the mass of the specimen after corrosion in grams; s is the initial area of the specimen in cm²; and t is the corrosion time in hours. The metal corrosivity rating is usually determined according to the corrosion depth. Equation (9) was applied for converting the corrosion rate from the weight loss method to depth method:

$$v_{depth}=\frac{v_{loss}\times 24\times 356}{1000\times \rho }=\frac{8.76v_{loss}}{\rho }$$

(9)

where $v_{depth}$ is the corrosion rate when using the depth method and ρ stands for the metal density; in the equation, the unit of $v_{depth}$ is mm/a. “mm/a” is the unit of the corrosion rate, where “mm” is the millimeter representing the depth of corrosion and “a” is the year representing the corrosion time.

2.3. Surface Characterization Method

The hot corrosion experiment was conducted using the crucible method. A processed sample was placed into a crucible along with 3 g of Na₂SO₄ and 17 g of V₂O₅, both having a purity of at least 99.9%. The crucible was then heated in an electric furnace at a rate of 5 °C/min until it reached 700 °C (750 °C and 800 °C temperatures were also used). Atmospheric pressure and constant temperature hot corrosion tests were then performed. Samples were collected every 24 h, cooled in air to room temperature, cleaned with alcohol, dried, and weighed. The corrosion products were analyzed using a D8ADVANCE X-ray diffractometer. The corrosion morphology and composition of the samples were analyzed using a JSM-6700 field emission scanning electron microscope (SEM) and EDAX spectrometer, and a cross-section of the sample after corrosion was observed using an SEM.

3. Results and Discussion

3.1. Evaluation of Hot Corrosion Rate

As shown in Figure 1a, the three curves represent the corrosion rates of the samples at 700 °C, 750 °C, and 800 °C, respectively, where the X-axis of the coordinate system is time, in hours; the Y-axis of the coordinate system represents the mass change of the sample, expressed in mg/cm². The corrosion curves of the alloy at 700 °C and 750 °C conform to the parabola law, and the corrosion rate at 700 °C and 750 °C is obviously lower than that at 800 °C. As shown in Figure 1b–d, the corrosion kinetics curve within 48 h of 800 °C corrosion is close to the linear law.

In high-temperature corrosive environments, alloys generate an oxide protective film to safeguard the alloy substrate. However, corrosion gradually damages and consumes the oxide film. This process involves high-temperature oxidation, the dissolution of the oxide layer, re-oxidation, and re-dissolution. At 800 °C, if the corrosion time prolongs, the Cr₂O₃ on the material surface gradually decreases, indicating that the protective oxide film, such as Cr₂O₃, which is produced by the extruded Inconel 625 alloy during 800 °C corrosion, is not sufficient to resist the corrosion of V₂O₅ and Na₂SO₄. As a result, Cr₂O₃ gradually dissolves and is consumed during the corrosion process, allowing the S element to invade the matrix and generate sulfide. The study found that the protective effect of the oxide film is good at 700 °C and 750 °C as it prevents the corrosive medium from coming into contact with the substrate. As shown in Figure 1, the corrosion rate is much higher at 800 °C. At this temperature, there is no formation of a Cr₂O₃ oxide layer in the initial stage of corrosion, resulting in a faster corrosion rate. However, a protective Cr₂O₃ oxide layer is formed in the later stage of corrosion, resulting in a change from a straight line to a parabolic line. The corrosion rate of the alloy according to Equation (9) was calculated to be 0.135 mm/a at 700 °C for 120 h, 0.152 mm/a at 750 °C for 120 h, and 0.431 mm/a at 800℃ for 120 h, respectively. In light of the ten grade evaluation system, the corrosion rate is grade six.

3.2. Hot Corrosion Morphology Analysis

Figure 2 shows the corrosion morphology of the specimens at 700 °C at different times under 20% Na₂SO₄ and 80% V₂O₅ molten salt. As can be seen from the corrosion morphologies of samples at 700 °C for 24 h, a flat and dense corrosion layer is formed on the surface with a trapezoidal distribution. There is no obvious granular oxide distribution of the surface. After 48 h of corrosion, a gap occurs between the corrosion layer and substrate, and an obvious corrosion belt appears, leading to poor bonding with the substrate. At the same time, irregularly shaped particles are distributed over the corrosion belt. According to the element distribution after the corrosion was characterized using EDS analysis, the corrosion layer mainly contains Ni, Cr, O, and V elements. Compared with the surface, the content of Ni elements in the inside of the corrosion belts increases while the content of Cr elements decreases, indicating that the granular material is a vanadate containing Ni (CrVO₄). After 120 h of corrosion, the corrosion layers become loose, and the corrosion layers in some corrosion areas peel off due to the influence of thermal stress.

Figure 3 illustrates the morphology of the sample after corrosion at 750 °C under a molten salt mixture of 20% Na₂SO₄ and 80% V₂O₅ for different time intervals. The corrosion morphology of the sample after 24 h at 750 °C reveals the formation of a flat and dense corrosion layer with a trapezoidal distribution on the surface. Additionally, there is no apparent granular oxide distribution on the surface. An energy spectrum analysis was conducted on various areas of the corrosion layer 48 h after the sample’s corrosion (Figure 3b,d). The analysis revealed that the precipitates in the area are rich in O and V elements. It was hypothesized that the formation of vanadate Ni₃V₂O₈ containing Ni occurred in this region. After being subjected to corrosion at a temperature of 750 °C for a period of 120 h, the surface of the sample displays a corrosion layer that mostly consists of loose powder particles. The presence of pits in certain areas suggests that the sample’s interior is relatively dense, albeit with some cracks, as depicted in Figure 3c.

Figure 4 shows the corrosion morphology of the specimens at 800 °C at different times under 20% Na₂SO₄ and 80%V₂O₅ molten salt. Figure 4a shows the corrosion morphology of extruded Inconel 625 alloy in a V₂O₅–Na₂SO₄ system at 800 °C for 24 h. The corrosion layer is relatively flat, but there are pits on the surface, along with the distribution of spherical iron segregation and massive Ni₃(VO₄)₂ particles. After corrosion at 800 °C for 24 h, pits appear on the surface of the corrosion layer, the surface is in a loose and porous state, and a large number of corrosion particles are distributed on the surface. After 48 h of corrosion, most of the particles on the surface of the corrosion layer gather into clusters, and a small number of particles are dispersed on the surface of the corrosion layer. As written in the paper by Wang et al., with the increase in corrosion time, Na₂SO₄ further decomposes to form O²⁻ at the molten salt/substrate interface according to Equation (10) throughout the hot corrosion process [25]. Subsequently, a large number of Cr₂O₃ oxide films decompose and react with molten salt to form vanadates. NiO and Cr₂O₃ combine to form NiCr₂O₄ oxide with a spinel structure on this basis; this phenomenon also appeared in the paper by Malafaia et al. [26].

$${\mathrm{S}\mathrm{O}}_4^{2-}={\mathrm{S}\mathrm{O}}_2+\frac{1}{2}{\mathrm{O}}_2+{\mathrm{O}}^{2-}$$

(10)

Moreover, spheroidal iron segregation (one point in Figure 4b) can be observed inside the substrate, which is exposed, cracked, and porous due to the exfoliation of the corrosion layers and oxidation layers. After corrosion for 120 h at 800 °C, the surface of the sample is rugged and undulating, showing mountain-shaped characteristics, and a large number of massive and granular corrosion products are formed.

The cross-sectional micrographs of the sample in the molten salt system of 20% Na₂SO₄ and 80% V₂O₅ at different temperatures with different corrosion times are shown in Figure 5. The corrosion layer is a dense oxide layer and combines well with the substrate at 700 °C for 24 h. As for 48 h, corrosion occurs at the grain boundary, and the corrosion layer is a Cr-rich layer with crack initiation. After 120 h of corrosion, the grain boundary corrosion is more serious. In addition, there are obvious stratification phenomena in the corrosion layer and corrosion particles in the outer layer, and the corrosion cracks in the inner layer gradually expand into large spallation.

Compared with 700 °C, after corrosion at 800 °C for 24 h, it can be seen that the corrosion is a porous oxide layer and that there are a large number of corrosion voids in the inner corrosion layer. The cross-section morphology of the samples after 48 h shows different corrosion zones (Figure 5e), where the corrosion products are relatively dense in zone A, abundant voids in zone B, and a denser film between zone A and zone B; the alloy substrate is in zone C. The intergranular corrosion can be clearly observed in the corrosion section diagram of the samples after 120 h.

The phenomenon described above is due to the dissolution of the oxide layer by the corrosive medium and thermal stress during the corrosion process, which causes an oxidation layer void, indentation, and other weak links. So that the S element can corrode the substrate through the oxide layer, it regenerated the oxide film and substrates, repeating until the substrate surface area of the Cr element was consumed.

Simultaneously, the V₂O₅ and O²⁻ quickly convert to ${\mathrm{V}\mathrm{O}}_4^{3-}$ at the interface of the molten salt according to Equation (11). However, the undissolved V₂O₅ and vanadate have strong acidity that can dissolve the protective oxide film that is formed on the surface of the alloy [27,28,29].

$${\mathrm{V}}_2{\mathrm{O}}_5+3{\mathrm{O}}^{2-}=2{\mathrm{V}\mathrm{O}}_4^{3-}$$

(11)

The experimental results schematically show that the dissolution rate of molten salt is faster in the first 24 h of hot corrosion. Not only was the Cr₂O₃ oxide film quickly dissolved, but the nickel was also directly involved in the hot corrosion, forming a rich area of chemical dissolution corrosion. During the period of 24 h to 120 h, a loose oxide layer formed on the surface of the sample because of the continuous accumulation of NiO, Cr₂O₃, and other corrosion products, which could resist the direct invasion of molten salt. To some extent, the loose oxide layer is beneficial to the formation of a continuous protective oxide film, whose hot corrosion process mainly includes Equation (12).

$${\mathrm{S}\mathrm{O}}_4^{2-}+\frac{9}{2}\mathrm{N}\mathrm{i}=3\mathrm{N}\mathrm{i}\mathrm{O}+\frac{1}{2}{\mathrm{N}\mathrm{i}}_3{\mathrm{S}}_2+{\mathrm{O}}^{2-}$$

(12)

3.3. Hot Corrosion Products Analysis

The X-ray diffraction pattern of corrosion products on the sample surface is shown in Figure 6. The main diffraction peaks are the metal substrate, and the corrosion products are mainly vanadate, oxide, and sulfide. The diffraction peaks of Cr₂O₃, Ni₃V₂O₈, and CrVO₄ are stronger than those of 700 °C and 800 °C after corrosion at 750 °C for 24 h, and the diffraction peaks of NiCr₂O₄ are more obvious at high temperature. After 120 h of corrosion, the type of corrosion products did not obviously change; however, with the extension of corrosion time, the intensity of the diffraction peak of some phases was different. The intensity of diffraction peak Cr₂O₃ after corrosion at 700 °C for 120 h is stronger than that after corrosion at 750 °C and 800 °C for 120 h, and the diffraction peak intensity of sulfide after corrosion at 800 °C for 120 h is also more obvious than that after corrosion at 700 °C and 750 °C. When the corrosion temperature is 700 °C, the content of Cr₂O₃ obviously increases with the increase in corrosion time, and the vanadate of Ni and Cr increases. At 800 °C, the content of Cr₂O₃ decreases, and the sulfide content of Ni, Cr, and Mo increases with the increase in corrosion time.

Figure 7 shows the distribution of cross-sectional corroded surface elements after corrosion at different temperatures and times in 20% Na₂SO₄ and V₂O₅ molten salt. It is obvious that after the corrosion at 700 °C for 48 h, the corrosion layer was mainly composed of O elements, V elements, and a small amount of Cr elements. After 120 h of corrosion, with the increase in O elements, V elements, and Cr elements in the corrosion layer, corrosion occurred at the grain boundary.

The cross-section diagram of the corroded sample at 800 °C for 24 h reveals the presence of numerous holes in the corroded layer. Additionally, the content of Cr in the corroded layer and matrix below it is substantially reduced, indicating a significant consumption of Cr. The corrosion layer is primarily composed of sulfide corrosion, as evidenced by the enrichment of Ni, S, and Mo elements. In the event of corrosion, an oxide layer forms on the surface to safeguard the matrix from further corrosion. However, due to the corrosive medium and thermal stress, weak links such as holes, pits, and cracks appear in the oxide layer. This allows element S to pass through the oxide layer and corrode the matrix. The matrix will then regenerate the oxide film, and the process repeats until the Cr element in the surface area of the substrate is exhausted. S continues to vulcanize with the Ni and Mo elements in the substrate, resulting in a sulfide corrosion layer between the oxide film and the substrate. The failure to detect the enrichment of V and O elements outside the enrichment area of S elements may be attributed to the spallation of the oxide layer in certain areas under the influence of external force or thermal stress. This exposes the sulfide accumulated corrosion layer to the surface, leading to the undetected enrichment area of the V and O elements.

Nevertheless, after the corrosion at 800 °C for 48 h, the corrosion surface is evidently stratified. From the section diagram of the sample in Figure 5e and Figure 7c, it can be concluded that there are mainly V elements and O elements in region A, as well as a few Cr elements and Mo elements; region B mainly contains S element. When taken together with the XRD analysis, region A is the vanadium corrosion area, region B is the sulfur corrosion area, and region C is the alloy substrate (Figure 5e), and the dense film between zone A and B is mainly composed of Cr₂O₃. The results indicate that vanadium corrosion consumes the Cr₂O₃ oxide film on the metal surface, causing the Cr elements to mainly gather on the material surface in the form of vanadate and chromate; meanwhile, the S element is vulcanized with Ni, Mo, and other elements in the corrosion layer [30,31]. Combined with the XRD analysis, the outer layer of the corrosion layer is mainly a vanadate of the CrVO₄ structure type and a small amount of Ni₃V₂O₈ particles; the pores of the inner layer of the corrosion layer are mainly a Ni-containing sulfide.

The corrosion cross-section diagram of the sample after 120 h of corrosion at 800 °C clearly displays the stratification of the corrosion layer. The outermost layer is primarily composed of O, V, and Cr elements with a small amount of Ni. In contrast, the inner layer is enriched with S and Mo elements, with no presence of Cr in the area of S element enrichment. The corrosion layer can be categorized into two distinct layers, namely the vanadium corrosion layer and the loose sulfide corrosion layer, which are arranged from the outside to the inside. Vanadium corrosion leads to the consumption of the original Cr₂O₃ oxide film on the metal surface. As a result, vanadate and chromate become the primary forms of Cr element on the material surface. Sulfide corrosion occurs with Ni, Mo, and other elements as the S element enters the corrosion layer.

As shown in Figure 7d, after 120 h of hot corrosion, the chromium content in the corrosion layer is very small, and the main reaction changes as the hot corrosion continues to expand. The Cr elements content is too small to form a protective Cr₂O₃ oxide film at that time. As the influence of thermal stress increases, the surface corrosion layer begins to peel off as per Equations (13) and (14). When the protective layer on the surface completely disappears, it begins to undergo rapid internal oxidation and internal vulcanization.

$${\mathrm{C}\mathrm{r}}_2{\mathrm{O}}_3+2{\mathrm{O}}^{2-}+\frac{3}{2}{\mathrm{O}}_2=2{\mathrm{C}\mathrm{r}\mathrm{O}}_4^{2-}$$

(13)

$$\mathrm{N}\mathrm{i}\mathrm{O}+{\mathrm{O}}^{2-}={\mathrm{N}\mathrm{i}\mathrm{O}}_2^{2-}$$

(14)

Through the above analysis, it is concluded that the process of the corrosion of the extruded Inconel 625 in the molten salt system of Na₂SO₄ and V₂O₅ is mainly a high-temperature oxidation, dissolution, re-oxidation, and re-dissolution of the oxide layer.

3.4. Hot Corrosion Model

Based on the above results and analyses, the hot corrosion process of extruded Inconel 625 in Na₂SO₄ and V₂O₅ molten salt systems can be described as the following four stages, as shown in Figure 8:

4. Conclusions

(1) When Inconel 625 corrodes in a molten salt system of Na₂SO₄ and V₂O₅ at high temperature, the surface micromorphology of the corrosion layer will transition from a powder to a salt granule and bulk granule with the increase in corrosion temperature;

(2) At the same corrosion temperature, a smooth and dense corrosion layer forms on the surface of the alloy at the initial stage of corrosion. Part of the corrosion layer is peeled off due to the influence of thermal stress, and there is a layered phenomenon inside, which is distributed in a stepped manner. In the later stage of corrosion, the grain boundaries are severely damaged, forming a large number of massive corrosion products with poor bonding to the substrate that are accompanied by cracks and cavities;

(3) In the early stage of corrosion, an oxide film that is mainly composed of Cr₂O₃ and NiO will form on the surface of the alloy. As the corrosion progresses further, the element V will consume the oxide film to generate Ni₃V₂O₈ and CrVO₄, making the oxide layer loose. At the same time, the element S will pass through the oxide film and enter the inside of the substrate, causing internal sulfidation and generating sulfides that are mainly composed of Ni, Cr, and Mo, leading to accelerated intergranular corrosion and intragranular corrosion.