Study on Hot Corrosion of Low-Nickel Cladding Metals Containing Nitrogen in K₂SO₄-MgSO₄ Binary Molten Salt

Abstract

Molten salt is usually used as the energy storage medium for solar energy heat storage pipes, and 40wt% K₂SO₄ + 60wt% MgSO₄ is very suitable for use as a heat storage material for solar thermal power generation in tower and butterfly parabolic systems. The demand for high-temperature thermal energy storage systems has prompted research on low-cost alloys for use in high-temperature and corrosion-resistant environments. The 44% Ni-24% Cr-0.18N nitrogen-containing low-nickel flux-cored welding wire designed in this article has a corrosion resistance of up to 900 °C after welding repair, which is better than the repair ability of Inconel 625 flux-cored welding wire. Using the high-temperature static immersion corrosion method, the corrosion behavior of two deposited metals immersed in molten salt for 60 h at 900 °C was assessed. The corrosion product phase composition, corrosion morphology, and elemental distribution of the two deposited metals were systematically studied using X-ray diffraction (XRD) and a Gemini SEM 300 (Zeiss thermal field scanning electron microscope). The results showed that the corrosion weight loss of the deposited metals showed the same trend at 900 °C, with corrosion occurring slowly from 0 h to 10 h and increasing after 10 h to 60 h. It was found that 10 h was the boundary point for corrosion behavior, and the corrosion resistance of the low-nickel nitrogen-containing deposited metal is better than that of the Inconel 625 deposited metal. This was because the addition of N energy elements allowed the formation of a stable composite nitride layer to suppress corrosion.

1. Introduction

Nickel-based high-temperature alloy is a type of high-temperature alloy that exhibits excellent oxidation resistance, corrosion resistance, and other comprehensive properties within the range of 650 °C to 1000 °C [1]. Nickel-based alloys are one of the most important materials for chemical, energy, and marine engineering. Inconel 625 flux-cored wire is commonly used for repair work on solar storage units. Solar energy varies based on objective factors, such as time of day, cloud conditions, seasonal changes, and geographical location. Ephemeral solar energy also contributes to the mismatch between the demand and supply of power, which makes solar energy storage technologies particularly critical. Sulfate (K₂SO4-MgSO₄) has the characteristics of low cost, high intensity, high-temperature boiling point, low saturated vapor pressure, and excellent thermal instability, making it ideal for use as a heat storage material for solar thermal power generation in tower and butterfly epoxy systems. However, due to the strong corrosiveness of sulfate at high temperatures, the usage of Inconel 625 flux-cored welding wire has been increasing year by year, resulting in increased costs. Therefore, there is an urgent need to find an element that can replace Ni to reduce costs. Research has shown that adding N element to Ni Cr alloys can improve their corrosion resistance, as the addition of N element can reduce the dilution of Cr element, consume H⁺, and alter the local corrosive environment [2,3]. Hanninen et al. [4] investigated the impact of N alloying using various techniques and fabrication methods on the corrosion and wear properties of high-nitrogen stainless steels. The results showed that nitrogen alloys significantly improved the mechanical, wear resistance, and corrosion properties of high-nitrogen steel; however, when the N solubility limit was reached, nitride precipitation would affect these properties. Peter J. et al. [5] introduced the development concept and reported on the performance of new austenitic stainless steel. The main features of the new type of austenitic stainless steel are the high N levels and the lack of Ni. In addition to the absence of Ni, the steel also had excellent corrosion resistance and mechanical properties.

At present, the temperature employed for the use of sulfate molten salt as an energy storage medium is mainly 600 °C. The higher the temperature, the better the energy storage effect. However, as the temperature increases, the corrosion of sulfate salts gradually increases, leading to frequent pipeline failures. The nitrogen-containing low-nickel flux-cored welding wire designed in this article has a corrosion resistance of up to 900 °C after welding repair, which is better than the repair ability of Inconel 625 flux-cored welding wire. At the same time, it also provides new ideas for the material design of solar energy storage tubes and has strong commercial value.

2. Experiment

This experiment used the composition of Inconel 625 flux-cored welding wire as a reference. The composition of the flux-cored welding wire, labeled as 1#, 2#, and 3#, is shown in

Table 1. Chemical composition of low-nickel welding wire containing nitrogen (wt%).

Serial Number	Ni	Cr	Fe	Mn	Mo	Co	Cu	Ti	Nb	Al	W	N
1	44.0	20.0	14.40	3.8	5.1	0.7	0.2	1.4	5.6	1.7	2.7	0.21
2	44.0	22.0	12.85	3.4	6.4	0.7	0.2	1.5	5.4	1.2	2.0	0.15
3	44.0	24.0	12.62	2.2	6.3	0.7	0.2	1.1	5.3	1.1	2.3	0.18
4	64.7	22.0	0.9	0.23	8.6	0	0.02	0.4	3.15	0	0	0

, and the Inconel 625 deposited metal is labeled as 4#. In the experiment, MnN and CrN alloy powders were added to the core powder, and N element was partially substituted for Ni element. Ni, Mn, Mo, Cr, V, Al, Ti, Nb, and W powders with a purity of over 99.9% were selected to adjust their elemental ratios. The power mixer mixed the powder well and evenly, and an Inconel 718 steel strip was selected as the steel strip. The diameters of the freshly developed nitrogen-containing low-nickel flux-cored wires were 2.55 mm. This article used Molten active gas arc welding (MIG) welding to prepare Ni, Cr, and Nx (x = 0.15, 0.18, and 0.21) deposited metals with different N contents on a Q235 substrate. The size of the Q235 substrate is 100 mm × 10 mm × 10 mm, and its chemical composition is shown in

Table 2. Chemical composition of Q235 substrate (wt%).

Element	C	Mn	Si	S	P	Fe
Content	≤0.22	≤1.4	≤0.35	≤0.050	≤0.045	Bal

. The welding process parameters were as follows: welding current was 180 A, welding voltage was 28 V, welding speed was 8 m/h, gas flow was 10 L/min, and the shielding gas used in this test was 97% Ar + 3% N₂. In order to avoid the influence of the base material dilution rate, this experiment involves overlaying three layers and taking corrosion samples from the second layer, as shown in Figure 1a.

The thermophysical data for sulfate are shown in

Table 3. Thermophysical property data of K₂SO₄ and MgSO₄ corrosion salts.

Component	Melting Temperature (°C)	Melting Heat (J/g)	Density (g/m3)
K2SO4	(α→l) 1074	211.4	2.662
	(β→α) 583
MgSO4	(α→l) 1137	122	2.660
	(β→α) 1010

. The molten salt utilized in this test was a mixture of 40% K₂SO₄ + 60% MgSO₄ [6]. The above corrosive molten salt was configured and allowed to dry for 24 h at 110 °C. In the high-temperature immersion corrosion test, the test temperature was checked by placing the pyrometer in a drying tank, as shown in Figure 1b. The molten metal was placed in the molten salt for high-temperature heating while ensuring that the molten salt completely covered the samples. It should be kept in mind that the edges of the three samples were sharpened prior to performing the high-temperature molten salt immersion corrosion test, as shown in Figure 1c. The heating temperature was set to 900 °C, and the insulation times were 10 h, 20 h, 30 h, 40 h, 50 h, and 60 h. After the molten salt test, the sample was boiled in boiling water for half an hour, and the molten salt attached to the surface was removed by ultrasonic cleaning. The operating conditions of solar thermal storage tubes at 900 °C were characterized using the static corrosion method and compared with those of Inconel 625 deposited metal. After drying the sample, weigh the sample on a 0.1 mg electronic balance and compare it with the original weight to calculate the weight loss. To ensure the representativeness of the test data, each group included 3 samples, and the test results were reported as the average.

Following high-temperature molten salt immersion corrosion, the phase make-up of the corrosion-deposited metal was measured using an XRD-7000 X-ray diffractometer (Shimadzu, Kyoto, Japan). The specific parameters were as follows: pure Cu target material, 40 KV tube voltage, 30 mA current, 2°/min scanning velocity, and a 20–90° scanning field. The morphology and elemental distribution of the corrosion layer on the cross-section of the deposited metal were analyzed using a Gemini SEM 300 scanning electron microscope.

3. Experimental Results and Discussion

3.1. Corrosion Kinetics

In Table 2, it can be seen that the melting points of K₂SO₄ and MgSO₄ corrosion salts were higher than the experimental temperature. Therefore, at 900 °C, the sulfate did not melt during the experiment, and the corrosion salt was deposited on the surface of the sample, which will cause a weight gain for the sample. Figure 2 shows the corrosion weight increase curve of the nitrogen-containing nickel-based deposited metal and Inconel 625 deposited metal in 40wt% K₂SO₄ + 60wt% MgSO₄ corrosion salt at 900 °C. In Figure 2, it can be seen that the corrosion resistance of nitrogen-containing nickel-based deposited metal in sulfate corrosion salts was better than that of Inconel 625 deposited metal at 900 °C. At the experimental temperature, the mass of the deposited metal significantly increased in the first 10 h because the corrosion rate was faster in the early stage of corrosion. At 10–60 h, the increase in the quality of the deposited metal tended to be gradual, with only a slight increase. This was because as the corrosion time increased, substances with anti-corrosion properties were formed on the surface of the deposited metal, which can hinder the progression of the corrosion reaction. Thermal corrosion reactions in corrosive salts include oxidation and vulcanization, and the corrosive surface is also formed on the sample face [7,8,9,10]. In the corrosion weight increase curve, it can be seen that the corrosion resistance of 3# deposited metal was superior to other deposited metals, so 3# was the best possible configuration for nitrogen-containing low-nickel flux-cored welding wire. The subsequent high-temperature molten salt corrosion experiments mainly analyzed the corrosion behavior of 3# nitrogen-containing low-nickel deposited metal and Inconel 625 deposited metal.

The corrosion gravity loss curve gives a good picture of the continuous change in gravity over time due to corrosion, but it does not clearly reflect the time period during which the corrosion tolerance of the deposited metal undergoes a drastic change. Therefore, using the corrosion rates obtained with Formulas (1) and (2), the corrosion weight loss was converted, and the corrosion rate curve over time was plotted, as shown in Figure 3 [11]. As shown in Figure 3, at a test temperature of 900 °C, the corrosion rates of both nitrogen-containing nickel-based and Inconel 625 deposited metals showed a rapid increase in the first 10 h. At this time, the corrosion rates of the two deposited metals were 2.05 mm/year and 2.30 mm/year, respectively. This is due to the intense reaction between the surface of the deposited metal and the corrosive salt in the early stage of corrosion, when corrosion-resistant substances have not formed. As the corrosion time increased to 60 h after corrosion, the corrosion rates of the two deposited metals were 0.46 mm/year and 0.58 mm/year, respectively. As the temperature increased, the corrosion rate gradually decreased because corrosion-resistant substances formed on the surface of the deposited metal, hindering the progression of the corrosion reaction. In summary, the corrosion rate of Inconel 625 deposited metal was higher than that of nitrogen-containing nickel-based deposited metal, indicating the resistance of nitrogen-containing nickel-based deposited metal to “S” and “O” at different temperatures. The corrosiveness of “$\mathrm{S}{\mathrm{O}}_4^{2-}$” corrosion salts was better than that of Inconel 625 deposited metal.

$${\mathrm{v}}_{-}=\frac{{\mathrm{m}}_0-{\mathrm{m}}_1}{\mathrm{St}}$$

(1)

$${\mathrm{v}}_{\mathrm{L}}=\frac{{\mathrm{v}}_{-}\times 24\times 365}{1000\mathsf{\rho }}$$

(2)

In these formulas, v_- is the corrosion rate (g/m²h), m₀ is the original mass (g), m₁ is the mass (g), S is the surface area (m²), and t is the corrosion time (h). In addition, V_L represents the annual corrosion rate (mm/year), and ρ indicates the density (g/cm³).

3.2. Surface Product Phase of Hot Corrosion of Deposited Metal

According to the above corrosion curve, it can be found that the highest resistance of two deposited metals to “S” and “O” at different temperatures occurs at 10 h, which represents the transition point of “$\mathrm{S}{\mathrm{O}}_4^{2-}$” corrosion salt corrosion. Hence, in the subsequent analysis of corrosion products and corrosion mechanisms, samples subjected to 10 h of corrosion were selected for phase and element distribution analyses. Figure 4 shows the XRD spectra of the surface phase composition of two deposited metals after 10 h of corrosion at 900 °C. Jiang, H. et. al. [12] demonstrated that over a temperature range of 850–950 °C, the type of corrosion products does not change over 200 h. The presence of Cr₂O₃ and Ni₃S₂ indicated that hot corrosion involved two processes, namely oxidation and vulcanization. Research has shown that a layer of oxide film was first formed on the surface of the deposited metal, and then the electrons of S were converted to the interface layer between the corrosive salt and the sample and penetrated into the oxide film, indicating the occurrence of a sulfurization reaction [13,14]. The penetration of corrosive media can damage the protective Cr₂O₃ formed during the oxidation process. In Figure 4, it can be seen that the corrosion products of the two deposited metals at 900 °C were mainly composed of Cr₂O₃, NiCr₂O₄, Al₂O₃, and Ni₃S₂.

3.3. Corrosion Morphology and Element Distribution of Deposited Metal

Figure 5 displays the element distribution of a cross-section of a nitrogen-containing nickel-based deposited metal corroded at 900 °C for 10 h. It can be seen in Figure 5 that a thin corrosion layer was formed on the surface of the deposited metal, which was about 2.59 μm. Due to the presence of $\mathrm{S}{\mathrm{O}}_4^{2-}$ corrosive salts, a dense and complete oxide layer cannot be formed. The external corrosion layer is rich in Cr, O, and S elements. The corrosion layer of nitrogen-containing nickel-based deposited metal was mainly composed of white layered substances. Using scanning electron microscopy for element analysis, the corrosion products were identified as oxides of Cr and Ni. Combined with XRD analysis, it was found that the white layered substances were mainly Cr₂O₃, and the corrosion products also appear as spinel-structured NiCr₂O₄ and Ni₃S₂. The reaction generates a Cr₂O₃ layer that can provide a certain protective layer for the substrate from severe thermal corrosion. The generated Ni₃S₂ corrosion products indicated that the S element reacted with the Ni element on the skin of the deposited metal.

Figure 6 displays the element distribution of the cross-section of the Inconel 625 deposited metal after corrosion at 900 °C for 10 h. In Figure 5, it is seen that as the corrosion reaction progresses, the corrosion product layer of Inconel 625 deposited metal tends to be porous and loose, and S element can enter the matrix through the corrosion channel. After 10 h of corrosion at 900 °C, the primary corrosion products were Ni₃S₂, NiCr₂O₄, and Cr₂O₃. Cr element is not present in the internal area of Inconel 625 deposited metal. Due to the high mass fraction of Cr element in Inconel 625 deposited metal, Cr diffused from the inside of the deposited metal to the surface. Thus, an external shell of 12.94 μm Cr₂O₃ was formed. In the initial period of the corrosion reaction, the rapidly forming protective layer of Cr₂O₃ led to an increase in its mass, which was consistent with the corrosion weight gain curve.

4. Discussion

Usually, the hot corrosion process of alloys consists of an initial stage with a slow corrosion rate and a rapid catastrophic stage. In the initial stage, active elements with high affinity for oxygen in the alloy (such as Al, Cr, Si, etc.) oxidized with molten salt or O₂ in the environment. As the corrosion reaction progressed, the O element was continuously consumed at high temperatures, and the relative content of S element increased, beginning the initial stage of the sulfurization reaction. The formation of corrosion products on the alloy surface is closely related to the standard Gibbs free energy (∆G^θ) generated by the corresponding oxides and sulfides of each element. Figure 7 shows the standard Gibbs free energy of several main corrosion products at 200–950 °C. It can be seen in Figure 7 that the Al element in the nitrogen-containing low-nickel deposited metal first underwent selective oxidation to generate Al₂O₃ and Cr₂O₃. Whether Al₂O₃ and Cr₂O₃ can quickly cover the surface of the alloy and maintain its integrity is the key to determining the hot corrosion resistance of the alloy. To understand the mechanism of sulfide molten salt corrosion caused by nitrogen-containing low-nickel deposited metal and Inconel 625 deposited metal at high temperatures, the effects of S and N elements on corrosion were analyzed and explained.

The gradual penetration of elemental S through the metal/corrosive salt interface confirms that the vulcanization process follows an oxidation path. During the hot corrosion process, there is less entry of S element into the interior of the deposited metal due to the presence of N element. For nitrogen-containing nickel-based deposited metals, Cr₂N begins to precipitate along grain boundaries, and some Cr₂N even precipitates within the grains of nitrogen-containing nickel-based deposited metals [15]. Combined with organizational analysis, Cr₂N began to precipitate in the form of a nitride layered structure inside the grains, and the presence of N element reduced the diffusion and dissolution of Cr element. Compared with Inconel 625, the thickness of the corrosion outer layer reached 12.9 μm, as shown in Figure 5a. Therefore, nitrogen-containing nickel-based deposited metals are more resistant to “S” and “O” than Inconel 625 deposited metals due to corrosion of “$\mathrm{S}{\mathrm{O}}_4^{2-}$”. The N element enrichment interface at the oxide layer metal interface can form a stable composite nitride layer, inhibit corrosion, and promote the formation of a Cr₂O₃ protective layer on the surface of the deposited metal. The white layered corrosion products in the nitrogen-containing low-nickel deposited metal were the overlap of Ni and S in the corrosion layer, and it can be inferred that Ni₂S₃ appeared during the hot corrosion process. Both oxidation and sulfurization occur during the hot corrosion process. In the deposited metal, the hot corrosion process mainly consisted of three stages: the formation of the initial oxide layer, the formation of the external oxide layer and sulfide zone, and the development stage of the internal stripe sulfide structure [16]. The corrosion reaction stage varies with changes in the working temperature and the corrosion environment. During the hot corrosion process of nitrogen-containing nickel-based deposited metals, a Cr₂O₃ protective layer was first formed, similar to the high-temperature oxidation process [16]. As the hot corrosion time increases, “$\mathrm{S}{\mathrm{O}}_4^{2-}$” can form from the corrosion reaction reported in Formula (3):

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

(3)

In this equation, “$\mathrm{S}{\mathrm{O}}_4^{2-}$” forms from the “S” and “O” in corrosive salts. The S element released by “$\mathrm{S}{\mathrm{O}}_4^{2-}$” continuously attacks the nitrogen-containing nickel-based deposited metal and then reacts with Ni in the corrosion reaction described in Formula (4) to generate Ni₃S₂, resulting in internal sulfidation of the nitrogen-containing nickel-based deposited metal:

$$3\mathrm{N}\mathrm{i}+2\mathrm{S}={\mathrm{N}\mathrm{i}}_3{\mathrm{S}}_2$$

(4)

With further increases in the hot corrosion temperature and corrosion time, the sulfides formed in the reaction described in Formula (4) are easily oxidized by the corrosion reaction reported in Formula (5):

$${\mathrm{Ni}}_3{\mathrm{S}}_2+2{\mathrm{O}}_2=3\mathrm{N}\mathrm{i}+2\mathrm{S}{\mathrm{O}}_2$$

(5)

Based on the above experimental results and corresponding analysis, it can be confirmed that the hot corrosion process of nitrogen-containing nickel-based deposited metals is a combination of oxidation and sulfurization.

As the corrosion reaction progresses, the Cr element continuously diffuses and dissolves outward. After the formation of the Cr₂O₃ oxide film, the Cr element on the surface of the deposited metal undergoes volume expansion, and the expansion coefficient of the formed oxide film is inconsistent with that of the deposited metal. Therefore, as the stress in the alloy matrix increases, it will crack and fall off, causing local Cr-deficient areas [17]. On the unprotected surface of the deposited metal, S and O elements can enter the interior of the deposited metal along this area, which is consistent with the distribution of S elements found inside the deposited metal in Figure 4e. The corrosive ions entering the interior of the deposited metal will react with the interior of the deposited metal to form spinel-structured Ni and Cr oxides. Combined with XRD analysis, these results suggest that the corrosion product is NiCr₂O₄. These reasons have all led to the consumption of corrosion-resistant elements in the deposited metal, exacerbating the corrosion of Inconel 625 deposited metal.

During the corrosion reaction, in addition to the reaction described in Formula (5), the reaction reported in Formula (6) will also occur. Ni-Ni₃S₂ is a corrosion product of a low-melting-point eutectic, which has a lower melting point and accelerates the corrosion reaction. In addition, the S element that invades the deposited metal will also react with the Cr element of the deposited metal, as noted in Formula (7), consuming Cr element content and reducing its corrosion resistance.

$$\mathrm{N}\mathrm{i}+{\mathrm{Ni}}_3{\mathrm{S}}_2=\mathrm{Ni}-{\mathrm{Ni}}_3{\mathrm{S}}_2$$

(6)

$$\mathrm{Cr}+\mathrm{S}=\mathrm{CrS}$$

(7)

Figure 5 shows the cross-sectional element distribution of Inconel 625 deposited metal after 10 h of corrosion at 900 °C. Regarding the distribution of Cr element in Figure 5c, it can be seen that there is a lack of Cr element in the internal area of Inconel 625 deposited metal. Due to the high mass fraction of Cr element in Inconel 625 deposited metal, Cr diffuses from the inside of the deposited metal to the surface, forming an external shell layer of Cr₂O₃. Due to the limited content of Cr element, as the corrosion reaction progresses, Cr-poor areas begin to appear in the deposited metal, forming internal oxides and sulfides in the Cr element depletion zone. The distribution of S element in Figure 5e shows the presence of S element inside the deposited metal. The generated Cr₂O₃ protective shell will also dissolve the “S” and “O” in the “$\mathrm{S}{\mathrm{O}}_4^{2-}$” corrosive salt, and the corrosion reaction can be expressed as noted in Formula (8):

$$2\mathrm{S}{\mathrm{O}}_4^{2-}+{\mathrm{Cr}}_2{\mathrm{O}}_3+1/2{\mathrm{O}}_2=2\mathrm{Cr}{\mathrm{O}}_4^{2-}+2\mathrm{S}{\mathrm{O}}_2$$

(8)

Inconel 625 deposited metal is exposed to “S” and “O”. The formation of sulfides in the “$\mathrm{S}{\mathrm{O}}_4^{2-}$” corrosive salt is due to the presence of “S” and “O”. The S in the “$\mathrm{S}{\mathrm{O}}_4^{2-}$” corrosive salt reacts with the deposited metal component. The S in the “$\mathrm{S}{\mathrm{O}}_4^{2-}$” corrosion salt diffuses through cracks to the oxide skin/metal interface, where the O₂ potential is low and sulfides are easily formed. On the other hand, corrosive salts can also flow through cracks to the oxide skin/metal interface, where reduction reactions occur [18]. The corrosion reaction equation for “$\mathrm{S}{\mathrm{O}}_4^{2-}$” is expressed in Formula (9):

$$\mathrm{S}{\mathrm{O}}_4^{2-}+4{\mathrm{e}}^{-}={\mathrm{S}}^{2-}+2{\mathrm{O}}_2^{2-}$$

(9)

5. Conclusions

This article mainly describes the design of a nitrogen-containing low-nickel flux-cored welding wire to reduce costs while ensuring its high-temperature corrosion resistance. This provides an important reference and significance for the repair of solar energy storage pipes and the selection of materials. This article mainly studies the corrosion behavior of nitrogen-containing low-nickel deposited metal and Inconel 625 deposited metal in 40wt% K₂SO₄ + 60wt% MgSO₄ molten salt at 900 °C. By comparing the corrosion behavior of the two deposited metals and conducting a systematic analysis, the following conclusions can be drawn:

(1) The best recipe for nitrogen-containing low-nickel flux-cored wire was determined to be 44% Ni-24% Cr-0.18N based on a comparison of the high-temperature static immersion corrosion tests and by optimizing the design. Nitrogen-containing low-nickel deposited metal had superior corrosion performance to Inconel 625 deposited metal at 900 °C. The corrosion layer depth was 2.59 μm after 10 h. The corrosion layer of Inconel 625 deposited metal after 10 h of corrosion was 12.94 μm thick.

(2) After 10 h of corrosion of nitrogen-containing low-nickel deposited metal at 900 °C, the surface is mainly composed of Cr₂O₃, NiCr₂O₄, and Al₂O₃. The addition of N element can reduce the dilution of Cr, which is conducive to the formation of a Cr₂O₃ protective film and can improve its corrosion resistance. Due to the limited content of Cr element in the Inconel 625 deposited metal, as the corrosion reaction progresses, Cr-poor areas begin to appear in the deposited metal, forming internal oxides and sulfides in the Cr depletion zone.