High Temperature Oxidation Behavior of Nickel and Iron Based Superalloys in Helium Containing Trace Impurities

Abstract

A high-temperature gas-cooled reactor (HTGR) is recognized as the best candidate reactor for next generation nuclear reactors. Helium is used to be the coolant in the core of the HTGR with temperature expected to exceed $900^{\circ}C$ at the core outlet. Several iron- and nickel-based superalloys, including Alloy 800H, Hastelloy X, and Alloy 617, are potential structural materials for intermediate heat exchanger (IHX) in an HTGR. Oxidation behaviors of three selected alloys (Hastelloy X, Alloy 800H, and Alloy 617) were investigated at four different temperatures from $650^{\circ}C$ to $950^{\circ}C$ under helium environments with various concentrations of $O_2$ and $H_2O$. Preliminary results showed that chromium oxide as the primary protective layer was observed on surfaces of the three tested alloys. Based on results of mass gain and SEM analyses, Hastelloy X alloy exhibited the best corrosion resistance in all corrosion tests. Further details on the oxidation mechanism of these alloys are presented in this study.

1. Introduction

A high-temperature gas-cooled reactor (HTGR) is a Generation IV nuclear reactor based on the design devel-oped in the 1950s. This type of system is designed to attain the aim of high efficiency and produce massive amounts of hydrogen. An inert gas, helium, is used in the coolant system and the outlet temperature exceeds 900 °C[1]. Accordingly, the alloy materials selected for an inter-mediate heat exchanger (IHX) in an HTGR should have high degradation resistance and superior mechanical strength over a long lifetime, about 60 years. The helium used in HTGR coolant system predictably contains low levels of impurities during reactor operation. Impurities in the helium coolant, such as oxygen and water vapor, may increase the corrosion rate of alloys at high temper-atures [2-7]. Several nickel- and iron-based alloys are po-tential candidates for structural materials in an HTGR. Furthermore, more severe oxidation of the alloys used for IHX could result in integrity breach of the IHX structure during an air ingress event. Although the formation of stable oxides is preferred for preventing more oxygen from reacting with substrate to degrade alloy materials, unexpected spallation of thick oxides may harm the long-term integrity and hence the mechanical strength of the materials during prolonged operation of an HTGR. In this study, two nickel-based alloys (Hastelloy X and Alloy 617) and one iron-based alloy (Alloy 800H) were selected as target materials for undergoing corrosion test in helium gas environments with various amounts of im-purities (i.e. oxygen and water vapor) at different temper-atures from 650 ^oC to 950 ^oC for 48hTests results on the mass changes, surface morphologies, and composition of oxides of the samples are reported.

2. Materials

Table 1 shows the chemical composition of the tested materials of Alloy 800H, Hastelloy X, and Alloy 617. These alloys were cut into samples in square disk shape with a dimension of 20 mm × 20 mm × 1 mm. No thermal treatment was carried out before the corrosion test. Both sides of the samples were mechanically polished by abra-sive paper starting from 80 grit up to 1200 grit. Before undergoing corrosion test, the specimens were ultrasoni-cally washed in acetone (CH₃COCH₃), dried in air, and weighted using an electronic balance.

Table 1 Chemical composition of the alloys used in this study (wt%)

3. Experiment

3.1 Test loops

A conceptual schematic of the experimental system is shown in Fig. 1. In each corrosion test, six samples, two from each alloy, were exposed to designated flowing gas mixtures in a high-temperature quartz tube furnace. The sample holder was made of ceramic, which is an inert material. Gas input was controlled using a mass flow con-troller set with a flow rate of 250 ml/min. The samples were heated up at a rate of 15 ^oC/min and cooled down at a rate of 7.5 ^oC/min in pure argon at a constant flow rate of 250 ml/min.

Fig. 1 Conceptual Schematic of the experimental system.

3.2 Test conditions

The experiments in this study used six gaseous con-ditions to simulate normal and various air-ingress environ-ments in an HTGR, including pure dry helium, 99% heli-um with 1% oxygen, 90% helium with 10% oxygen, 90% helium with 10% air, pure helium with 10% humidity, and pure helium with 50% humidity. Table 2 lists the impurity levels in pure helium used in this study. The testing temperature was set at 650 °C, 750 °C, 850 °C, and 950 °C in each experimental condition. Corrosion test duration was 48 hours.

Table 2 Level of impurities (ppm) of pure helium gas used in this study (He > 99.9995%)

3.3 Sample Observation and Analyses

After the corrosion tests, the mass gain of each sample was measured using a high-precision balance. A scanning electron microscope (SEM) was used to analyze the oxides on the surface and cross-sectional microstructure of sam-ples in different conditions. The elemental composition of the oxide was analyzed using energy-dispersive X-ray spectroscopy (EDX). The crystal structure and chemical composition of oxide were characterized by X-ray dif-fraction (XRD), confirming the results of EDX.

4. Result and Discussion

4.1 Mass gain in high-temperature gas environments

Fig. 2shows the mass gains of Alloy 800H as a function of the test temperature after isothermal oxidation up to 48 hours in various controlled helium environments. Similar oxidation behavior appeared at 650 °C, 750 °C and 850 °C in all gas environments, and no significant differences in mass gain of this alloy were observed. The oxidation rate at 950 °C exhibited higher than those at other temperatures. At 850 ^oC, the mass gain of Alloy 800H in 90% helium with 10% oxygen was the highest due to the higher oxygen partial pressure. As up to 950 ^oC, the mass gains in both wet 10% and 50% humidity environments were similar with that in 90% helium with 10% oxygen. No oxide spallation was observed on the surface of Alloy 800H in this study.

Fig. 2 Mass gains of Alloy 800H at various testing conditions.

Fig. 3 shows the mass gain of Alloy 617 after being oxidized at four test temperatures up to 48 hours in various controlled helium environments. It is known that the alloy oxidation is faster in water vapors or humid environments than in dry environments, therefore, a higher mass gain is expected in wet helium than in the helium environment containing air or oxygen at the same temperature.

Fig. 3 Mass gains of Alloy 617 at various testing conditions.

Fig. 4 indicated that Hastelloy X had the same tendency of oxidation behavior at various temperatures from 650 °C to 950 °C. At four test temperatures, a lower mass gain in both humid environments. The lack of obvious spallation on the surface suggested that the major source of lower mass gain might be caused by the evaporation of Cr-rich oxide layers, which become unstable and react with water vapors [8-14]. Hastelloy X apparently has the ability to resist oxidation at high temperatures, but chro-mic volatilization may occur in humid environments. Furthermore, the mass gain of Alloy 800H was the high-est, and that of Hastelloy X was the lowest among the three alloys under all test conditions.

Fig. 4 Mass gains of Hastelloy X at various testing conditions.

4.2 Morphology of Alloy 800H

Fig. 5 shows the cross-sectional oxide morphologies of Alloy 800H tested at 950 °C for 48 hours in pure dry helium, 99% helium with 1% oxygen, 90% helium with 10% oxygen, 90% helium with 10% air, pure helium with 10% humidity, and pure helium with 50% humidity. Fig. 6 shows the EDX mapping analysis of Alloy 800H for 48 hours at 950 °C in a 90% helium with 10% oxygen environment. The elements of iron, chromium, and man-ganese appear in the oxide layer of Alloy 800H, and Cr-rich oxide forms an inner layer above the substrate. The scattered internal oxides of Al2O₃ were observed be-neath the continuous Cr2O3layer. Because manganese is soluble in Cr2O3at high temperature, it exists in the inner oxide layer to forms manganese spinel, which is more stable than MnO and Cr2O3, at the interface of chromium oxide and the spinel phase [15,16]. Furthermore, man-ganese diffuses faster than iron or nickel in the Cr2O₃ layer via the lattice diffusion, which explains how man-ganese concentrates on top of the spinel scale [17,18]. Thus, three oxide layers were formed by a spinel phase (NiFeO₄, MnCr₂O₄, Mn₃O₄and Fe₃O₄) on the surface, a Cr2O₃ layer below the external oxide layer, and an internal oxide of an Al2O3layer below the chromium-rich layer. These oxides were confirmed by XRD and EDX analysis. EDX analysis also confirmed the depletion of chromium in the matrix as Cr2O₃ grew on the top of the metal.

Fig. 5 SEM cross-sectional micrographs for Alloy 800H at 950 °C for 48 hours: (a) in 100% pure dry helium, (b) in 99% helium with 1% oxygen, (c) in 199% helium with 10% air, (d) in 99% helium with 10% oxygen, (e) in pure helium with 10% humidity, (f) in pure helium with 50% humidity.

Fig. 6 EDX mapping for the cross-sectional of Alloy 800H at 950 °C for 48 hours in a 90% helium with 10% oxygen environment.

4.3 Morphology of Alloy 617

Fig. 7 shows SEM cross-sectional micrographs of Alloy 617 at 950 °C for 48 hours in various controlled helium environments. The EDX mapping analysis of Alloy 617 for 48 hours at 950 °C in a 90% helium with 10% oxygen environment is shown in Fig. 8. Titanium and manganese existed on the top layer of the oxide in the form of TiO₂ and MnCr₂O₄, as confirmed by XRD analysis. The surface oxides of Alloy 800H and Hastelloy X exhibited the chro-mium oxide formed as Cr₂O₃. It was noted that aluminum particles concentrated below the metal-scale interface as Al-oxide or Al₂O₃. Therefore, during the SEM, XRD, and EDX analysis, three oxide layers were formed by a layer of spinel phase and TiO₂ on the top surface, an inner oxide layer consisting of Cr₂O₃ below the spinel, and aluminum concentrating in the metal layer, forming Al₂O₃ internal oxidation.

Fig. 7 SEM cross-sectional micrographs for Alloy 617 at 950 °C for 48 hours: (a) in 100% pure dry helium, (b) in 99% helium with 1% oxygen, (c) in 199% helium with 10% air, (d) in 99% helium with 10% oxygen, € in pure helium with 10% humidity, (f) in pure helium with 50% humidity.

Fig. 8 EDX mapping the cross-sectional of Alloy 617 at 950 °C for 48 hours in a 90% helium with 10% oxygen environment.

4.4 Morphology of Hastelloy X

Fig. 9 shows SEM cross-sectional micrographs of Hastelloy X tests at 950 °C for 48 hours in various con-trolled helium environments. The SEM micrographs of this alloy in various gas conditions appear similar, and it is consistent with the results of mass gain. According to the EDX mapping analysis of the sample shown in Fig. 10, which shows the results tested at 950 °C for 48 hours in a 90% helium with 10% oxygen environment, the oxide on the surface consists of an inner layer of Cr₂O₃, and silicon collects below the Cr₂O₃inner layer. The manganese exists in both the inner and outer oxide layers since Mn diffuses faster than Cr, Fe, and Ni and is solute in Cr₂O layer [17,18]. Furthermore, manganese oxide forms a stable spinel phase, and co-exist with Cr₂O₃[15,16]. Conversely, Fe and Ni are mostly blocked in the matrix by Cr₂O₃even though in severe condition. Due to the results of XRD and EDX analysis, the oxide layer was mainly composed of a thin spinel layer composed of MnCr₂O₄on the surface, a Cr₂O₃ layer that accounts for most of the oxides above the substrate, and silicon forming an internal oxide as SiO₂below the chro-mium-rich layer.

Fig. 9 SEM cross-sectional micrographs for Hastelloy X at 950 °C for 48 hours: (a) in 100% pure dry helium, (b) in 99% helium with 1% oxygen, (c) in 199% helium with 10% air, (d) in 99% helium with 10% oxygen, € in pure helium with 10% humidity, (f) in pure helium with 50% humidity.

Fig. 10 EDX mapping for the cross-sectional of Hastelloy X at 950 °C for 48 hours in a 90% helium with 10% oxygen environment.

4.5 Corrosion mechanism

The composition of elements is very crucial to deciding the properties and characteristics of an alloy. Alloys may maintain long-term integrity exposed to high temperature gas environments due to the presence of a protective oxide layer on the alloy surface. This avoids or reduces the dif-fusion rate of anions inward to the metal or cation outward to the surface [19]. In this study, Alloy 800H is iron-based alloy, and Hastelloy X and Alloy 617 are nickel-based alloys. The Cr contents in the selected alloys ranged from 18% to 22%, which surpass the minimum value of critical chromium concentration to form continuous oxide scales [8,20-22]. Because chromium oxide grows at a slow rate and is considerably more stable than oxides of iron, nickel, and cobalt, Cr₂O₃ is formed preferentially based on ther-modynamics [19]. Therefore, a Cr₂O₃ layer became the major structure on the surface of the tested alloy at the early stage of the corrosion process. The concentration of aluminum in the tested alloys in this study was less than 2% and prevent it from forming scales even if Al₂O₃ has a lower Gibbs free energy (ΔGf =-858.30 kJ/mol) than that of Cr₂O₃ (ΔGf =-546.86 kJ/mol). In this case, the aluminum caused internal oxidation underneath the sur-face oxide layer, leading to reduction in creep resistance of the alloy due to the tensile stress in the perpendicular direction. Cr₂O₃would be the main protective oxide, and the internal oxidation of Al would cause alloy failure [19].

Alloy oxidation becomes faster in the wet helium envi-ronments compared to pure air helium and the helium en-vironment containing oxygen or air. This study shows that the mass gain of Alloy 617 in wet helium environments was greater that in a mixture of helium and oxygen at 850^oC and 950^oC, and the variation of mass gain in all tested environments is not significant at 650^oC and 750^oC. Alloy 800H had a slightly higher mass gain in wet helium environments than the helium environment containing oxidants except 10% oxygen. For Hastelloy X, the mass gain in wet helium environments was smaller than that in pure dry helium and helium environment con-taining oxidants. Alloy in a humid helium environment may react with steam in several oxidizing ways. Steam may react with metal to produce oxide and hydrogen on the metal surface. Alternatively, this reaction produces volatile hydrated species [23,24]. For example, chromium may form Cr₂O₃ and H₂ or CrO2(OH)2, as determined by the following reactions:

(1)

(2)

CrO₂(OH)₂ is a volatile chromium oxide, and then re-duce the mass gain of the alloy. Although FeO, Fe₃O₄, Fe₂O₃, NiO, and Al₂O₃ also form volatile compounds by reacting with steam, the hydroxide formation reaction of Cr₂O₃ has lower free energy than the above reactions, and CrO₂(OH)₂ is more stable than other hydroxides [25,26]. At Fig.9, the oxide thickness in the wet helium environ-ments was thinner than those in other helium conditions. It was also found that a higher humid helium environment would reduce the mass gain of Hastelloy X. This can be explained by the formation of volatile chromium hydrox-ide on these alloys. The behavior of volatilization reduces the lifetime of alloys due to the thinner protective layer [19].

5. Conclusion

In this study, isothermal oxidation tests on Alloy 800H, Hastelloy X, and Alloy 617 were conducted at different temperatures from 650 °C to 950 °C for 48 hours in six controlled helium environments. The alloys in this study are potential superalloys for an HTGR involving inter-mediate heat exchanger and recuperator applications. Based on the results of oxidation tests, SEM and XRD analyses, this study presents the following findings:

1. At 950 °C, a continuous Cr₂O₃ layer was observed on the surfaces of all tested alloys in the six selected environments.

2. Hasetlloy X exhibited the lowest mass gain in all test conditions among the tested alloys. This alloy ex-hibited the simultaneous formation of Cr₂O₃ and MnCr2O4on the surface of the alloy. In the meantime, Hastelloy X demonstrated lower mass gains in humid helium envi-ronments than those in dry helium environments contain-ing of oxygen or dry air.

3. Both Alloy 800H and Alloy 617 suffered from in-ternal oxidation after the oxidized test at 950 °C for 48 hours. Alloy 617 exhibited more aluminum oxides below the Cr₂O₃ inner layer and a poorer resistance to internal oxidation.