Evolution of Oxide Phases and Residual Stress in Haynes 282 Superalloy During Long-Term High-Temperature Oxidation

doi:10.21203/rs.3.rs-5257755/v1

The long-term oxidation behavior of Haynes 282 superalloy was investigated in air at temperatures ranging from 800 to 950°C for durations up to 720 hours. The oxide phases formed on the alloy surface were characterized using X-ray diffraction (XRD) and energy-dispersive X-ray spectroscopy (EDS). The predominant oxide phase was identified as rhombohedral-Cr₂O₃, with the presence of rutile-TiO₂, spinel-MnCr₂O₄, and perovskite-CoTiO₃ as secondary phases. The external oxide layer thickness increased with oxidation temperature and time, following parabolic kinetics. EDS mapping revealed the formation of an internal oxide layer, consisting of α-Al₂O₃ and TiO₂ precipitates beneath the external Cr₂O₃ layer. The activation energy for the long-term oxidation of Haynes 282 was calculated to be 210.05 ± 23.30 kJ mol^− 1. The residual stress in the Cr₂O₃ layer was measured using the average X-ray strain (AXS) method. Compressive residual stresses were observed in the Cr₂O₃ layer formed at 800°C, while tensile residual stresses were found in the layer formed at 950°C. The evolution of intrinsic stress with oxidation time and temperature was discussed in terms of the crystallite coalescence model and the Pilling-Bedworth ratio.

The significance of advanced ultra-supercritical (A-USC) coal-fired power generation in thermal power technology has been widely acknowledged, primarily due to its improved efficiency and reduced environmental impact compared to traditional coal-fired systems. A-USC systems operate at temperatures and pressures above the critical point of water, reaching up to 700°C and 35 MPa, resulting in efficiencies exceeding 45%. These operational efficiencies are responsible for decreasing coal consumption and significantly reducing CO₂ emissions per unit of electricity generated [1–3]. The Haynes 282 alloy, renowned for its exceptional high-temperature strength, corrosion resistance, and fabricability, is considered a crucial material in the operation of A-USC systems, particularly for critical power plant components exposed to the harshest conditions [4, 5]. The alloy's resistance to creep and oxidation at high temperatures is deemed essential for the long-term performance, durability, and reliability of A-USC power plants [6].

Extensive research has been conducted on the Haynes 282 alloy, a wrought, gamma-prime strengthened material, focusing on its microstructural evolution, mechanical properties, and response to various heat treatments [7, 8]. The microstructure of the alloy, especially the gamma prime (γ') precipitates crucial for its strengthening mechanism, undergoes significant transformations due to heat treatment temperatures. Moreover, the evolution of carbides from interconnected to discrete structures as a result of heat treatment substantially influences the alloy's mechanical behavior [7].

It is widely recognized that exposure to oxidizing environments significantly impacts the mechanical properties of materials. Oxidation behavior, characterized by the chemical reaction between the material and an oxidizing environment, often at elevated temperatures, typically leads to the formation of oxide layers or scales on the material's surface, considerably altering its properties and behavior. The material's ability to resist oxidation without substantial degradation is vital for its durability and stability, ensuring operational reliability and cost-effectiveness [9, 10]. While controlled oxidation processes have been observed to temporarily enhance certain mechanical properties in specific ceramics through the formation of a protective oxide layer, prolonged exposure to oxidizing conditions is generally detrimental [11]. The formation of oxide layers can induce internal stresses, potentially causing cracking or spallation of the oxide layer, exposing fresh material surfaces to further oxidation and accelerating material degradation [9, 10]. The inclusion of alloying elements such as Cr, Mn, Si, Al, and Ti, intended to improve oxidation resistance, may influence scaling rates and mechanical properties due to their interaction with oxidizing elements [12]. The oxidation resistance of Haynes 282 alloy, evaluated in flowing air for approximately 1000 hours at 927°C, is reported to be comparable or slightly superior to that of several other alloys [13]. The oxidation rates and behavior of Haynes 282 at 900°C and 1000°C are observed to be sensitive to changes in the system's oxygen partial pressure [14].

In addition to microstructure and environmental factors, surface treatments and coatings can provide further protection against oxidation and other forms of degradation in Haynes 282. For instance, the application of thermal barrier coatings (TBCs) can reduce the surface temperature of the alloy, slowing down the oxidation kinetics and extending the service life [15]. Other surface modification techniques, such as shot peening and laser surface melting, can introduce compressive residual stresses and refine the microstructure, resulting in improved oxidation resistance [16].

Incorporating oxidation data into service life prediction models is essential for accurately estimating the performance and durability of Haynes 282 in various applications. Miller et al. [17] proposed a Creep-Fatigue-Oxidation Microcrack Propagation Model for Thermomechanical Fatigue, which correlates the fatigue and oxidation components using the ΔJ parameter and incorporates an additional time dependence in the oxidation term. The study provides an understanding of the interplay between fatigue and oxidation in materials like Haynes 282, offering insights into their behavior and performance in high-temperature applications where oxidation is a significant degradation mechanism. Recently, it has been proposed that by incorporating oxidation damage, especially under long-life conditions, life prediction of the nickel-based Inconel 718 can be quantitatively improved [18].

To ensure the long-term reliability of advanced energy systems, a comprehensive study of Haynes 282's oxidation resistance at elevated temperatures is crucial, serving as a foundation for assessing its durability and performance in challenging environments. This study aims to evaluate the oxidation mechanism and investigate the microstructural evolution of the nickel-based superalloy, Haynes 282. Experiments are conducted in atmospheric conditions, with temperatures ranging from 800 to 950°C and exposure durations spanning from 1 to 720 hours, enabling a thorough analysis of its long-term alterations and adaptability. Furthermore, the impact of oxidation temperatures and durations on residual stress variations is examined. Meticulous observations of stress level changes are made, providing valuable insights into the complex mechanisms and the diverse effects of time and temperature on residual stress formation within the oxide layer.

2.1 Material and Sample Preparation

For this study, Haynes 282 alloy, provided by Haynes International, Inc., was utilized. The alloy's composition, as specified by the material supplier, is presented in Table 1. To ensure a uniform distribution of γ’ precipitates within the γ matrix, the as-received superalloy underwent a two-step aging process [19]. Specimens for oxidation experiments, measuring 10×10×1.2 mm³, were prepared using wire electrical discharge machining (EDM). Prior to initiating oxidation experiments, the specimens were subjected to wet grinding using SiC paper with grit sizes ranging from P80 to P4000. Subsequently, ultrasonic cleaning in acetone and ethanol was performed for 5 minutes each, followed by drying with nitrogen gas. Prior to high-temperature exposures, the initial mass (M_i) of the specimens was determined using a high-precision balance accurate to 0.01 mg. In addition, the surface area (A) of the samples was calculated through measurements of the width and length of the square coupons using an electronic Vernier caliper, which has an accuracy of 0.1 mm.

Table 1

The composition of Haynes 282 alloy in this study (at%)*.
Ni	Cr	Co	Mo	Al	Ti	Fe	C	Mn
Bal(56.04)	21.63	10.00	5.11	3.22	2.67	0.80	0.3	0.06
B	W	Ta	Zr	Cu	P	S	Si
0.03	0.01	0.004	0.002	< 0.009	< 0.004	< 0.004	< 0.103

*: The original composition proof of Haynes 282 is in the unit of weight percent. The atomic percent in this table is transformed from the original data.

2.2 High-temperature oxidation Tests

The high-temperature oxidation tests were performed in a conventional muffle furnace, with the ambient laboratory air acting as the oxidation medium. Specimens were arranged obliquely on an alumina crucible to guarantee uniform exposure to the high-temperature air on both surfaces throughout the test. This crucible was situated within the furnace’s zone of uniform temperature, where a variance of ± 1°C from the set temperature was maintained.

The heating regimen was executed as follows: the furnace was initially brought to 600°C at a heating rate of 15°C/min and held at this temperature for 10 minutes. Then, the temperature was escalated to the designated working temperatures of 800, 850, 900, and 950°C at a rate of 6°C/min. A 30-minute stabilization period at the working temperature preceded the onset of oxidation testing, with durations fixed at 6, 12, 24, 48, 96, 192, 360, and 720 hours. Specifically, short-term exposure represents heat treatment of less than 24 hours, mid-term exposure accounts for 48–192 hours of heat treatment, and long-term exposure comprises 360–720 hours of heat treatment. These categorizations ensure a thorough examination of the oxidation behavior over a range of durations, providing insights into the material's performance under different conditions. After reaching the predetermined exposure duration, the crucible along with the specimens was extracted from the furnace and allowed to cool in air to room temperature. The mass gain per unit area ($\:\frac{\varDelta\:M}{A}$) of the cooled specimens was determined and documented according to the formula:$\:\frac{{\Delta\:}M}{A}=\frac{{M}_{f}-{M}_{i}}{A}$, where M_f is the final mass of the specimen.

2.3 Microstructure and composition analysis

The phases and structures present on the external oxide layer of Haynes 282 were identified through X-ray diffraction (XRD) analysis, conducted using a Bruker D2 Phaser. The analysis encompassed both as-received and high-temperature oxidized specimens. The XRD employed a copper target's K_α X-ray as the illumination source, operating under a voltage of 30 kV and a current of 10 mA. The measurement range for θ/2θ extended from 20° to 95°, employing a scanning speed of 0.05°/s.

The surface morphology and composition of the Haynes 282 specimens, pre and post oxidation, were examined using high-resolution thermal field emission scanning electron microscopy (HRFEG-SEM) combined with energy-dispersive X-ray spectroscopy (EDS). These observations were conducted on a JEOL JSM-7100F FEG-SEM, which was equipped with an EDS system.

2.4 Cross-sectional analysis of oxidized-Haynes 282 specimens

For observing the layered oxide structure on Haynes 282 after the exposure at elevated temperatures, two different methods of cross-sectional analysis were utilized, starting with specimens having an oxide layer thickness greater than 1 µm. These specimens were first bisected using a precision low-speed diamond saw with a 4-inch diamond blade, operating at 40 rpm. Post-cutting, they were cleaned with tap water, treated with acetone to remove contaminants, and dried using a nitrogen gas air blow gun. The cleaned specimens were then bonded together using epoxy-phenolic strain gauge adhesive (M-Bond 610 Kit). The adhesive was applied thrice on one side of the specimens using a polymer brush. After a 10-minute air drying period for solvent evaporation, the specimens were clamped together and heated in an oven. The heating process involved ramping up to 175°C at 12°C/min, maintaining this temperature for one hour to cure the adhesive. Following the solidification process, the specimens were air-cooled to room temperature. This bonding procedure is critical for retaining the oxide layer during specimen preparation. The bonded specimens were then positioned upright in a 30mm-diameter, 6mm-height mold, and encapsulated with Buehler's EpoThin 2 epoxy resin. The surface was subsequently ground and polished, using a series of silicon carbide papers and Al₂O₃ on a ChemoMet synthetic polishing pad. Excess cold mounting media was removed with P80 SiC paper to avoid prolonged vacuuming in subsequent SEM observation. After water cleaning, the specimens were vacuumed in a molding vacuum desiccator. Prior to SEM observation, the specimen surface was gold-plated at 20 mA for 40 seconds using a JEOL JEC-3000FC sputter coater, completing the preparation for cross-sectional sample analysis. For specimens with an oxide layer thickness less than 1 µm, the aforementioned method would excessively damage the oxide layer. Additionally, HRFEG-SEM was used for microstructure observation of the specimen cross-sections. EDS was utilized to analyze the chemical composition of the oxide layer.

2.5 Residual stress measurement

Residual stress in the external oxide layer of Haynes 282, following oxidation at various temperatures and durations, was evaluated utilizing the XRD cos²αsin²ψ method [20]. This evaluation was conducted using a Bruker D8 XRD equipped with a 4-circle diffractometer and a ψ goniometer. The method entails measuring the shift in a particular diffraction peak across different ψ angles. The integration of this information with the elastic constant and Poisson's ratio (ν) of the main oxide phase facilitates the calculation of the overall residual stress of the external oxide layer. The (116) diffraction peak of rhombohedral-Cr₂O₃ (JCPDS #38-1479) was selected for the residual stress measurement of Haynes 282's external oxide layer due to rhombohedral-Cr₂O₃ being the primary phase of the oxide and its (116) peak offering adequate intensity for precise determination of the peak position. The Young's modulus (E) and ν values for the rhombohedral-Cr₂O₃, a principal component of the oxide layer on Haynes 282, were determined to be 138 GPa and 0.29, respectively [21].

3.1 Long term oxidation kinetics

The mass gain per unit area (∆M/A) of Haynes 282, which serves as an indicator of long-term oxidation kinetics, was investigated in relation to exposure time in still air at temperatures ranging from 800 to 950°C, as illustrated in Fig. 1. The presented data points represent the average mass gain per unit area obtained from two specimens subjected to the same oxidation conditions, with error bars denoting the variability between the tests. These results revealed a positive correlation between ∆M/A and both increasing temperature and prolonged exposure durations.

To further analyze the oxidation kinetics, a double logarithmic plot of mass gain versus time was constructed, as depicted in Fig. 2. This representation was employed to determine the exponent 'n' factor for the long-term oxidation of Haynes 282 across the investigated temperature range. The slopes of the linear relationships in Fig. 2, which correspond to the 'n' factor, are listed in Table 2. It was observed that as the oxidation temperature increased, the exponent factor 'n' deviated from the parabolic value of 0.5, suggesting a shift in oxidation kinetics at higher temperatures.

Table 2

The slope of the straight line with the R² of each fitting curve in double-log plot of mass gain per unit area of long-term oxidation vs. time.
Temperature	800°C	850°C	900°C	950°C
n	0.52	0.49	0.42	0.38
R²	0.985	0.973	0.994	0.991

3.2 Microstructure Analysis of Pre-/Post-Oxidized Haynes 282 for High-Temperature Oxidation

3.2.1 Crystal Structure of Pre-/Post-Oxidized Haynes 282

The crystallography of the Haynes 282 specimen was investigated using θ-2θ XRD analysis. The intensity-2θ patterns of Haynes 282 after oxidation in still air at temperatures ranging from 800 to 950°C are presented in Fig. 3 to Fig. 6. Each figure includes results from exposure times of 12 to 720 hours, as well as those of a non-oxidized, age-hardened Haynes 282 for comparison purposes. Four distinct diffraction peaks, corresponding to the (111), (200), (220), and (311) crystal orientations of face-centered cubic Ni (JCPDS #04-0850 [22]), were observed in the control specimen within the 2θ range of 20° to 96°. As the oxidation exposure time increased, a simultaneous decrease in the intensity of the four diffraction peaks associated with the FCC-Ni matrix phase was noted. For specimens oxidized at 950°C for 720 hours, these diffraction peaks of the Haynes 282 matrix completely disappeared, as shown in Fig. 6. The external oxide layer was found to be primarily composed of rhombohedral-Cr₂O₃ (JCPDS #38-1479 [23]), also known as chromia, as evidenced by the alignment of the majority of diffraction peaks across the temperature range of 800 to 950°C for oxidation durations of up to 720 hours with those of chromia. These chromia peaks were detectable in Haynes 282 specimens subjected to oxidation for only 12 hours at the specified temperatures of the study.

As the oxidation time progressed, the presence of a secondary distinct phase, rutile-TiO₂ (JCPDS #21-1276 [22]), became more pronounced. The appearance of rutile-TiO₂ diffraction peaks was observed to occur earlier and more prominently at higher temperatures. With increasing oxidation duration, some XRD peaks were found to align with those characteristic of spinel-type MnCr₂O₄ (JCPDS #75-1614 [24]) and/or perovskite-type CoTiO₃ (JCPDS #72-1069 [25]). It was observed that the volume fraction of these oxide phases increased with higher oxidation temperatures and longer durations, as evidenced by the increased number of identifiable phase peaks. Minor and infrequent peaks corresponding to cubic-NiO (JCPDS #47-1049 [26]) were also detected at oxidation temperatures ranging from 800 to 900°C.

3.2.2 Cross-Sectional Observation

Cross-sectional observations of Haynes 282, oxidized long-term with an oxide layer thickness at temperatures of 800, 850, 900, and 950°C, are presented in Fig. 7, presenting and backscattered-electron images (BEIs). From the top to the bottom of the oxidized alloys' cross-section, three distinct regions can be identified: the epoxy-phenolic strain gage adhesive, the external oxide layer, and the Ni-based matrix with discontinuously internal oxides near the interface as indicated. With increasing exposure time, an increase in both the external oxide layer's thickness and the depth of internal oxidation was observed across all tested temperatures. BEI offered a more pronounced contrast between oxide and matrix, with the external oxide layer invariably appearing slightly brighter than the inner oxide. This brightness differential suggests a variance in elemental composition between the external and inner oxide layers. BEIs were employed to measure both the thickness of the external oxide layer and the depth of the internal oxide in long-term oxidized specimens. The external oxide layer thickness, denoted as δ, was determined by averaging ten vertical line measurements within the external oxide layer using ImageJ, an image processing software. Figure 8 depicts the relationship between the δ and the exposure time of Haynes 282 in air at 800, 850, 900, and 950°C, with the error bar representing the standard deviation of the ten measurements for each condition. This pattern mirrors the evolution of mass gain per unit area, as shown in Fig. 1.

3.2.3 Compositional Depth-Profile

Figure 9 illustrate the EDS elemental mappings for cross-sections of Haynes 282, subjected to long-term oxidation at temperatures ranging from 800 to 950°C for 720 hours. The external oxide layer predominantly comprised chromium, manganese, and oxygen, with titanium present on the outermost surface of the external oxide layer and scattered within the matrix. Notably, the morphology of Ti coalescence within the matrix transitioned from spherical to larger clusters as oxidation time increased. Aluminum was consistently observed in strip or lump forms below the interface between the external oxide layer and the matrix across all oxidized specimens.

The cross-sectional mapping of a specimen oxidized at 800 and 850°C for 720 hours revealed cobalt and nickel co-localizing with titanium on the outermost surface, suggesting the formation of a Co-Ni-Ti-O layer. Aluminum was found to accumulate at the matrix/external oxide layer interface, forming a semi-continuous layer within the matrix just beneath this interface. In specimens oxidized at 900°C and 950°C for 720 hours, the accumulation of aluminum near the interface between matrix and external oxide layer was less pronounced. Cobalt was marginally detected and nickel was disappear on the outermost layer.

3.2.4 Residual Stress in the External Layer of Haynes 282

The measurement of residual stresses within the external oxide layer of Haynes 282, subjected to temperatures of 800 and 950°C, was conducted utilizing the Average X-ray Strain (AXS) method [27]. The results are presented in Fig. 10. It was found that the Cr₂O₃ layers, developed at both temperatures for durations extending up to 720 hours, exhibited compressive residual stresses (σ_AXS < 0). At an oxidation temperature of 800°C, an increase in the compressive residual stress within the chromia layer was observed with prolonged exposure times, indicating a cumulative enhancement of compressive stress as the material was subjected to extended periods under these specific conditions. Conversely, at the higher temperature of 950°C, an opposite trend was noted wherein the compressive residual stress diminished with an increment in exposure time. This distinction underscores a shift in the stress response of the chromia layer at elevated temperatures, potentially attributed to variations in oxide growth mechanisms, microstructural evolutions, or the impact of high-temperature phenomena on the material's characteristics

4.1 Evolution of External Oxide on Haynes 282

The oxide phase formed on Haynes 282 exposed to air at temperatures ranging from 800 to 950°C for 12 hours was characterized as rhombohedral-Cr₂O₃ using θ-2θ XRD analysis. The standard Gibbs free energy of formation per mole O2, ΔG_f⁰, for several binary oxides potentially forming on Haynes 282 are listed in Table 3 [28]. Although Al2O3 has the most negative ΔG_f⁰, suggesting a strong preference for formation, it was not detected in the XRD results. This absence is attributed to the low Al content (〜3 at%) in Haynes 282, insufficient for forming detectable ceramic crystal volumes using the in-house XRD method. In contrast, the high Cr content (〜21 at%) facilitates the formation of chromia, which has the second most negative ΔG_f⁰, enabling its identification in the short-term oxidized Haynes 282 at various temperatures. Rutile-TiO₂, with the third lowest ΔG_f⁰, was characterized in the oxidized Haynes 282 after 48 hours using XRD, despite the low Ti content (〜2 at%), possibly due to the high affinity between Ti and oxygen.

Cross-sectional EDS mapping indicated that Ni predominantly appears on the outermost surface of the external oxide layer at 800, 850, and 900°C, suggesting the formation of a thin NiO layer that dissipates at 950°C. This implies that NiO is thermodynamically unfavorable, consistent with its least negative ΔG_f⁰. Despite the high Ni composition and higher diffusivity of Ni in chromia compared to Cr [29], NiO is unstable relative to other thermodynamically favorable oxides like chromia and TiO₂, making it unobservable at higher oxidation temperatures.

The evolution of the oxide's crystalline structure, identified by XRD, with respect to oxidation temperature and exposure time, is consistent with the compositional distribution observed in EDS mapping. EDS analysis consistently found Mn distribution aligning with Cr distribution under all conditions, and XRD patterns of mid-term or long-term oxidized specimens showed diffraction peaks corresponding to the spinel-MnCr₂O₄ structure, despite the low Mn content (0.06 at%). MnCr₂O₄ can be formed through the reaction:

$$\:\text{M}\text{n}\text{O}+{{\text{C}\text{r}}_{2}\text{O}}_{3}\to\:\text{M}\text{n}{\text{C}\text{r}}_{2}{\text{O}}_{4}$$

MnCr₂O₄ is thermodynamically more stable than Cr₂O₃, as indicated by its more negative ΔG_f⁰ [30]. Moreover, Lobnig et al. found that Mn ions have higher diffusivity than Cr ions in Cr₂O₃ [29], enabling rapid Mn diffusion in the external oxide layer and accounting for the Mn content on the oxide surface. The formation of spinel in the external oxide layer increases steadily with exposure time or temperature. However, the low Mn content in as-received Haynes 282 and the time required for MnCr₂O₄ accumulation to allow detectable XRD diffraction peaks are responsible for the inconsistent results between XRD and EDS for short-term oxidized specimens. It is noted that cubic-CoO diffraction peaks were not observed, possibly due to the weaker thermodynamic preference for forming CoO compared to NiO, as the Gibbs formation free energy of CoO is slightly lower than that of NiO. However, CoTiO₃ formation was found in XRD results (Fig. 6). In the Co-Ti-O system, perovskite-type CoTiO₃ can thermodynamically exist below 1413K [31], and its ΔG_f⁰ is more negative than that of NiO and CoO [28]. CoTiO₃ can be produced through the reaction:

$$\:\text{C}\text{o}\text{O}+{\text{T}\text{i}\text{O}}_{2}\to\:\text{C}\text{o}\text{T}\text{i}{\text{O}}_{3}$$

The structure of CoTiO₃ corresponds to the observed Co distribution across the oxide thickness in the cross-section of Haynes 282 oxidized long-term at 950°C.

Table 3

The standard Gibbs free energy of formation (∆G_f) for various oxides observed in nickel-based superalloys, with values given in kilojoules per mole of oxygen (kJ/mol O₂) from 1000 K to 1300 K in increments of 100 K [22]. The oxides included in the table are NiO, Cr₂O₃, Rutile-TiO₂, Rhombohedral-Al₂O₃, and CoO.
Standard Gibbs free energy of formation $\:{\varDelta\:\mathbf{G}}_{\mathbf{f}}$ (kJ / mol O₂)
T (K)	Cr₂O₃	NiO	CoO	Rutile-TiO₂	Rhombohedral-Al₂O₃
1000	584.2	-298.4	-327.0	-762.5	-907.5
1100	567.4	-281.4	-313.2	-744.9	-885.4
1200	550.7	-264.4	-299.4	-727.2	-863.4
1300	-534	-247.4	-285.4	-709.4	-841.3

*Note: The original value of the standard Gibbs free energy of formation (∆G_f) for one mole compound from Barin’s book [22] is the ∆G_f of the formation of one mole compound. The ∆G_f values in this table are calculated per mole oxygen used in the formation of the compounds.

4.2 Evolution of Internal Oxide on Haynes 282

An internal oxide region, comprising TiO₂ and α-Al₂O₃, was observed beneath the external oxide layer in the cross-sectional EDS mapping results (Fig. 9). Unlike the continuous external oxide layer, alumina is located along grain boundaries close to the external oxide/matrix interface in all oxidized specimens. Titanium oxide is observed near alumina in most oxidized specimens, except those at 800°C. Prior research [32] indicated that these oxide phases are TiO₂ and Al₂O₃, constituting the internal oxide layer within superalloys. Rapp [33] reported four criteria for specific elements to form internal oxides in alloys:

The Gibbs free energy of formation of the internal oxide is more negative than that of the principal phase of the external oxide layer.
The Gibbs free energy of the internal oxidation reaction is negative.
The concentration of the alloying element is below the critical concentration for forming an external oxide layer.
The external oxide layer cannot prevent oxygen from diffusing into the alloy's matrix.

In Haynes 282, with chromia as the predominant external oxide phase, α- Al₂O₃ formation Gibbs free energy is more negative than that of Cr₂O₃, indicating that internal alumina is thermodynamically more stable than external chromia. The critical concentration of Al in the alloy,$\:\:{N}_{Al}^{\left(O\right)}$

$$\:{N}_{Al}^{\left(O\right)}\ge\:{\left[\frac{\pi\:{g}^{*}}{2V}{N}_{O}^{\left(s\right)}\frac{{D}_{O}{V}_{m}}{{D}_{Al}{V}_{OX}}\right]}^{1/2}$$

where represent the atomic fraction of oxygen in the alloy, the diffusivity of oxygen in the alloy, Al diffusivity in the alloy, molar volume of alumina, molar volume of the alloy, and the ratio of Al to O volume ($\:V=1.5$ for Al₂O₃), respectively. Wagner's theory suggests that the transition from internal to external scale formation is governed by the volume fraction of oxide reaching a critical value, $\:g=f\left({V}_{OX}/{V}_{m}\right)$, reaches a critical value, $\:{g}^{*}$. According to Zhao [34], the critical aluminum concentration, $\:{N}_{Al}^{\left(O\right)}$, necessary for the formation of a continuous external Al₂O₃ layer in Ni-based alloys falls within the range of 0.1 to 0.16 atomic fraction. In wrought Haynes 282, the Al content is reported to be 3.22 atomic percent [35], which is substantially lower than the critical concentration required for the development of a continuous external Al₂O₃ layer. The predominant presence of internal alumina along grain boundaries indicates that the nucleation of Al₂O₃is heterogeneous in nature.

As reported by Sabino et al. [36], the inward diffusion path of oxygen in polycrystalline Cr₂O₃ is along the grain boundaries. The presence of internal alumina suggests that oxygen atoms can penetrate through the chromia layer and into the metal matrix. The Ti content in the alloy, at 2.67 atomic percent, is slightly lower than that of Al. Furthermore, the formation Gibbs free energy of rutile-TiO₂ is more negative compared to that of Cr₂O₃, which is consistent with Rapp's criteria. Interestingly, TiO₂ is also detected at the outermost portion of the external oxide layer. It is hypothesized that the presence of TiO₂, rather than Al₂O₃, at the outermost portion may be attributed to the differing diffusivities of Ti and Al in chromia, although specific data on their diffusivities in chromia are not available. It is well-established that ionic diffusion in ceramic solids is related to the concentration of corresponding defects, which include both intrinsic and extrinsic defects, as described by the Kröger-Vink notation. The substitution of the cation lattice in chromia by doping with TiO₂ and Al₂O₃ can be represented as follows:

$$\:3\text{T}\text{i}{\text{O}}_{2}\underset{\iff\:}{{\text{C}\text{r}}_{2}{\text{O}}_{3}}3{{\text{T}\text{i}}_{\text{C}\text{r}}}^{\bullet\:}+6{{\text{O}}_{\text{O}}}^{\times\:}+{{\text{V}}_{\text{C}\text{r}}}^{{\prime\:}{\prime\:}{\prime\:}}$$

$$\:{\text{A}\text{l}}_{2}{\text{O}}_{3}\underset{\iff\:}{{\text{C}\text{r}}_{2}{\text{O}}_{3}}2{{\text{A}\text{l}}_{\text{C}\text{r}}}^{\times\:}+3{{\text{O}}_{\text{O}}}^{\times\:}$$

The provided formulas indicate that the incorporation of three TiO₂ molecules into the Cr₂O₃ lattice can result in the creation of a chromium cation vacancy, leading to an increase in the defect concentration within the ceramic. This increased defect concentration may contribute to a higher diffusivity of Ti⁴⁺ ions compared to Al³⁺ ions in chromia, assuming that these cations diffuse through the lattice. The presence of an Al accumulation region beneath the matrix/chromia interface may serve as evidence for the lower diffusivity of Al in comparison to Ti. This accumulation suggests that Al atoms are unable to diffuse as readily as Ti atoms through the chromia layer.

Kofstad [37, 38] reported that the formation of rutile-TiO₂ is governed by the inward diffusion of oxygen ions at temperatures below approximately 900°C. However, when the oxidation temperature surpasses 900°C, the dominant mechanism may transition to the outward diffusion of titanium cations. This shift in the dominant diffusion mechanism enables Ti atoms to diffuse through the chromia layer to the outermost position of the external oxide layer, where they can form the TiO₂ layer. Simultaneously, Ti atoms can also react with inwardly diffusing oxygen to form internal TiO₂ precipitates within the alloy matrix. The combination of outward titanium cation diffusion and inward oxygen ion diffusion contributes to the formation of both external and internal TiO₂ phases in the oxidized alloy.

4.3 Oxidation Kinetics of Haynes 282

Based on Wagner’s theory of oxidation kinetics, the rate of oxidation for protective oxides in Haynes 282 superalloys follows the parabolic law. This behavior can be described by the following equation:

$$\:\frac{\varDelta\:\text{M}}{\text{A}}=\sqrt{2{\text{K}}_{\text{P}}\text{t}}$$

1

Here, ∆M/A represents the mass gain per unit surface area, t is the oxidation time, and K_p is the parabolic rate constant. Figure 11 illustrates ∆M/A as a function of t^0.5, with a regression line fitted using Eq. (1). The parabolic rate constant can be expressed using the Arrhenius equation:

$$\:{\text{K}}_{\text{P}}={\text{K}}_{0}\text{e}\text{x}\text{p}\left(-\frac{\text{Q}}{\text{R}\text{T}}\right)$$

2

In this equation, K₀ is the pre-exponential constant, Q represents the activation energy of oxidation, R is the gas constant (8.314 J mol^− 1 K^− 1), and T denotes the oxidation temperature. The activation energy, Q, can be derived by taking the natural logarithm of both sides of Eq. (2), resulting in:

$$\:\text{l}\text{n}\left({\text{K}}_{\text{P}}\right)={\text{l}\text{n}(\text{K}}_{0})+(\frac{-\text{Q}}{\text{R}})\frac{1}{\text{T}}$$

3

By plotting ln(K_P) versus 1/T, the slope and intercept of the linear regression represent (-Q)/R and ln(K₀), respectively. Therefore, the value of Q is calculated by multiplying the slope by R, and K₀ is determined through the exponential of the intercept.

The slope, R-squared value, and K_P for each fitting curve are listed in Table 4. The Arrhenius plot for Haynes 282 after oxidation at temperatures ranging from 800 to 950°C is depicted in Fig. 12. The calculated activation energy Q for long-term oxidation is 210.05 ± 23.30 kJ mol^− 1, and K₀ is 4.78 × 10⁶ mg cm^− 2 h^− 0.5. It is noteworthy that all R-squared values in Table 5 surpass 0.96, suggesting a strong linear correlation with the square root of oxidation time. However, the exponent n value at 950°C appears to deviate from the typical range of 0.4–0.6 for the parabolic rate law.

According to Wagner's theory, the parabolic rate law is applicable when the oxide layer functions as a diffusion barrier and the oxidation reaction proceeds at a significantly faster rate than the diffusion of ions. Consequently, the oxidation behavior is considered to be governed by the diffusion of ions within the oxide scale. Under the assumption that all mass gain is a result of oxygen molecules reacting with chromium, and considering negligible mass loss due to oxide volatilization during the oxidation process, the relationship between ∆M/A of Haynes 282 and the external oxide thickness (δ) is expected to adhere to the following formula:

$$\:\frac{\varDelta\:\text{M}}{\text{A}}=0.1\:{\delta\:}\:{{\rho\:}}_{{\text{C}\text{r}}_{2}{\text{O}}_{3}}\:\frac{3\:{\text{M}}_{\text{O}}}{{\text{M}}_{{\text{C}\text{r}}_{2}{\text{O}}_{3}}}$$

4

Here, $\:{{\rho\:}}_{{\text{C}\text{r}}_{2}{\text{O}}_{3}}$is the density of Cr₂O₃ (5.22 g/cm³ [39]), M represents the molar mass in g/mol, and 0.1 is a unit conversion constant. Using values from the literature in Eq. (4), it simplifies to:

$$\:\frac{\varDelta\:\text{M}}{\text{A}}\:\left(\text{m}\text{g}\:{\text{c}\text{m}}^{-2}\right)=0.165\:{\delta\:}\:\left({\mu\:}\text{m}\right)$$

5

The linear relationship between ∆M/A and δ, as depicted in Fig. 8, is presented in Fig. 13. The slope and R-squared values for each fitting curve are provided in Table 5. It is observed that the average slope of ∆M/A versus δ in Fig. 13 is 0.164 ± 0.010 mg cm^− 2 µm^− 1, which closely aligns with the predicted value of 0.165 in Eq. (5). This finding lends support to the conclusion that the oxidation kinetics of Haynes 282 within the 720-hour timeframe is primarily governed by the oxidation of a single oxide phase, namely rhombohedral-Cr₂O₃.

Table 4

The slope of the straight line with the R² of each fitting curve and the value of $\:{K}_{P}$ of Haynes 282 for oxidation at the temperature range of 800 to 950°C
T (°C)	Slope (mg cm^− 2 h^− 0.5)	R²	$\:{\varvec{K}}_{\varvec{P}}$ (mg cm^− 2 h^− 0.5)
800	0.0222 ± 0.0009	0.990	0.00025
850	0.0426 ± 0.0028	0.974	0.00091
900	0.0723 ± 0.0033	0.988	0.00262
950	0.1036 ± 0.0042	0.987	0.00537

Table 5

The slope of the $\:\frac{\varDelta\:M}{A}$ versus δ straight line with the R² of each fitting curve of Haynes 282 for long term oxidation at the temperature range of 800 to 950°C
T (°C)	Slope (mg cm^− 2 µm^− 1)	R²
800	0.153 ± 0.012	0.9777
850	0.167 ± 0.011	0.9721
900	0.177 ± 0.015	0.9605
950	0.160 ± 0.013	0.9639

4.4 Generation and Relaxation of Intrinsic Stress in the Oxide Layer on Haynes 282

It is assumed that σ_AXS represents the total residual stress in the chromia. The total residual stress, σ_total, measured at room temperature (RT), was the sum of the thermal stress, σ_th, and the intrinsic stress, σ_in, and could be expressed by the equation:

$$\:{{\sigma\:}}_{\text{t}\text{o}\text{t}\text{a}\text{l}}=\:{{\sigma\:}}_{\text{t}\text{h}}+\:{{\sigma\:}}_{\text{i}\text{n}}$$

6

The origin of the thermal stress was attributed to the difference in the coefficient of thermal expansion (CTE) between the external oxide layer and the substrate. This discrepancy resulted in compressive strain in the Cr₂O₃ and tensile strain in the substrate during the cooling process from the oxidation temperature to RT. The thermal stress in Cr₂O₃ could be estimated using the equation:

$$\:{{\sigma\:}}_{\text{t}\text{h}}=\frac{{\text{E}}_{{\text{C}\text{r}}_{2}{\text{O}}_{3}}}{1-{{\nu\:}}_{{\text{C}\text{r}}_{2}{\text{O}}_{3}}}({{\alpha\:}}_{{\text{C}\text{r}}_{2}{\text{O}}_{3}}-{{\alpha\:}}_{\text{s}\text{u}\text{b}\text{s}\text{t}\text{r}\text{a}\text{t}\text{e}})\varDelta\:\text{T}$$

7

where $\:{{\alpha\:}}_{{\text{C}\text{r}}_{2}{\text{O}}_{3}}$ and $\:{{\alpha\:}}_{\text{s}\text{u}\text{b}\text{s}\text{t}\text{r}\text{a}\text{t}\text{e}}$were the linear expansion coefficients of Cr₂O₃ ($\:{{\alpha\:}}_{{\text{C}\text{r}}_{2}{\text{O}}_{3}}=9.55\times\:{10}^{-6}$ K⁻¹ [40]) and Haynes 282, respectively. $\:\varDelta\:\text{T}$ was the temperature difference between the oxidation temperature and the RT, i.e., $\:\varDelta\:\text{T}={\text{T}}_{\text{o}\text{x}\text{i}\text{d}\text{a}\text{t}\text{i}\text{o}\text{n}}-{\text{T}}_{\text{R}\text{T}}$. The mean value of linear CTE of Haynes 282 [41], $\:{{\alpha\:}}_{\text{H}\text{a}\text{y}\text{n}\text{e}\text{s}\:282}$, was utilized as the required $\:{{\alpha\:}}_{\text{s}\text{u}\text{b}\text{s}\text{t}\text{r}\text{a}\text{t}\text{e}}$ for calculating the thermal stress of the chromia. The thermal stress of the oxide layer $\:{\text{C}\text{r}}_{2}{\text{O}}_{3}$ at 800 and 950°C are − 807 ± 1 and − 1233 ± 1 MPa, respectively. Given that the thickness of the external oxide layer formed at 950°C constituted only 1.6% of the substrate's thickness, the thermal stress gradient across the external oxide was neglected, and a uniform thermal stress throughout the external oxide layer was presumed. The intrinsic stress, σ_in, was determined by subtracting the thermal stresses, σ_th, from the total residual stresses, σ_total, at the corresponding oxidation temperatures. Figure 14 illustrates the relationship between the intrinsic stresses of Cr₂O₃ and the exposure times at 800 and 950°C.

The intrinsic stress in chromia was initially tensile at 800°C. As the exposure time increased, the intrinsic tensile stress gradually decreased and eventually transformed into a compressive stress after 48 hours of oxidation. This compressive stress continued to intensify with prolonged oxidation time, but the rate of increase slowed down over time. After 720 hours of oxidation at 800°C, the intrinsic stress stabilized at approximately − 150 MPa. In contrast to the observations at 800°C, where the intrinsic stress transitioned from tensile to compressive, all intrinsic stresses measured at 950°C were tensile. However, the rate of increase in intrinsic tensile stress at 950°C also diminish ed over time, reaching approximately 761 MPa after 720 hours of oxidation.

It has been suggested that the tensile stress observed during the initial stages of polycrystalline film growth can be attributed to the process of crystallite coalescence [42, 43]. According to this model, which is based on thermodynamic principles, the free energy of the growing film is reduced through the coalescence of crystallites, subsequently leading to the development of intrinsic tensile stresses within the film. This decrease in Gibbs free energy favors the stabilization of the crystalline structure. The presence of tensile stress in the oxide is considered essential for the stabilization of polycrystalline film growth, as it facilitates the reduction of Gibbs free energy. Similar patterns of intrinsic tensile stress were observed by Tung and Stubbins during their analyses of residual stress in Cr₂O₃ formed on nickel-based superalloys, Inconel 617 and Haynes 230, during the early stages of oxidation within 6 h at elevated temperatures below 1000°C.[44, 45].

On the other hand, an alternative model [46] proposes that the initial stress in Cr₂O₃ is expected to be compressive, due to the fact that the Pilling-Bedworth ratio (PBR) of Cr₂O₃ is greater than 1. This high PBR indicates that the oxidation process is accompanied by volumetric expansion, which may induce compressive stress when subjected to the constraints of plastic deformation. Moreover, as demonstrated in a previous study [47], the diffusion of oxygen along grain boundaries in α-alumina scale may also contribute to the development of slight compressive stress within the oxide scale, especially when the oxidation mechanism is primarily governed by anionic diffusion at lower temperatures.

In the current study, it has been noted that the surface morphology of the oxides grown at 800°C and 950°C exhibits significant differences, as illustrated in Fig. 15. The oxides grown at 800°C displayed a relatively dense and smooth surface compared to those grown at 950°C. It is plausible that the relatively compact oxide layer, primarily composed of Cr₂O₃, may contribute to the development of compressive stress, where the Pilling-Bedworth mechanism plays a dominant role. Conversely, for the alloy exposed to 950°C, where severe oxidation occurs, the surface oxide exhibits a rugged morphology consisting of small crystals. This observation suggests that the process of crystallite coalescence was underway and subsequently dominated the development of tensile stress in the current study.

Another possible explanation for the greater residual tensile stress observed in the alloy exposed at 950°C is the significantly higher Ti content on the surface of the external oxide layer compared to that at 800°C. This suggests enhanced outward diffusion of Ti through chromia at 950°C. As mentioned in the Kröger-Vink notation section, the outward diffusion of Ti cations through the chromia lattice may induce tensile stress. Given the larger ionic radius of Cr³⁺ compared to Ti⁴⁺ [48], the substitution of Cr³⁺ by Ti⁴⁺ may cause lattice distortion in Cr₂O₃, leading to tensile stress. Furthermore, Kofstad's study [37, 38] indicates that the oxidation mechanism of TiO₂ at temperatures above 900°C is dominated by the outward diffusion of Ti⁴⁺, supporting increased Ti⁴⁺ diffusion at 950°C and introducing tensile stress into the chromia.

In this study, the long-term oxidation behavior of Haynes 282 was investigated at temperatures ranging from 800 to 950°C in still air for durations up to 720 hours. The oxidation kinetics were found to follow a parabolic rate law, with an activation energy of 210.05 ± 23.30 kJ mol^− 1 and a pre-exponential constant of 4.78 × 10⁶ mg cm^− 2 h^− 0.5. X-ray diffraction analysis revealed that the external oxide layer was primarily composed of rhombohedral-Cr₂O₃, with the presence of rutile-TiO₂, spinel-type MnCr₂O₄, and perovskite-type CoTiO₃ phases increasing with higher oxidation temperatures and longer durations. Cross-sectional observations and compositional depth-profiling indicated the formation of an Al-rich layer beneath the matrix/external oxide layer interface, while Ti was found to accumulate on the outermost surface of the external oxide layer and within the matrix. The intrinsic stress in the chromia layer was initially tensile at 800°C, gradually decreasing and transitioning to compressive stress after 48 hours of oxidation, while at 950°C, all intrinsic stresses were tensile. The differences in intrinsic stress behavior were attributed to the dominant oxidation mechanisms and the influence of Ti diffusion through the chromia layer at higher temperatures.

Acknowledgement

This research was funded by the Ministry of Science and Technology of Taiwan (ROC) under Contract No. MOST 109-2222-E-007-001-MY2. SEM were carried out at National Tsing Hua University. Special thanks for the support of SEM and XRD by Department of Material Research, National Atomic Research Institute. XRD for residual stress analysis were carried out at the National Chung Hsing University at Taichung, Taiwan and Department of Physics, National Atomic Research Institute, Taoyuan, Taiwan.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Kubushiro K, Nomura K, Matsuoka T, Nakagawa H, Muroki K. Development of boiler technology for 700 C A-USC plant. IHI Engineering review. 2016;49:34–43.
Bhaduri AK, Albert SK, Nagesha A, et al. Development and Qualification of Materials for Indian Advanced Ultra-Supercritical Coal-Fired Power Plant. Mater Perform Charact. 2022;11:88–108.
Lee S, Kim J, Tahmasebi A, Jeon CH, Liu Y, Yu J. Comprehensive technical review of the high-efficiency low-emission technology in advanced coal-fired power plants. Reviews in Chemical Engineering. 2023;39:363–386.
Tung DC, Lippold JC. Weld Solidification Behavior of Ni-Base Superalloys for use in Advanced Supercritical Coal-Fired Power Plants. Superalloys 2012. 2012;563–567. doi:10.1002/9781118516430.CH62.
A RP, RAMESH A. Numerical Analysis and Experimental Validation of Temperature Induced Creep and Fatigue life of Inconel 740H and Haynes 282. ARAI Journal of Mobility Technology. 2023;3:803–819.
Kim YJ, Park JH, Ahn YS. Comparison of Creep Properties of Cast and Wrought Haynes 282 Superalloy. Advances in Materials Science and Engineering. 2018;2018:2048959.
Joseph C. Microstructural characterization of Haynes 282 after heat treatment and forging. Chalmers Tekniska Hogskola; 2015.
Joseph C. Microstructure Evolution and Mechanical Properties of Haynes 282. Chalmers Tekniska Hogskola; 2018.
Kumar P, Lam TN, Tripathi PK, Singh SS, Liaw PK, Huang EW. Recent progress in oxidation behavior of high-entropy alloys: A review. APL Mater. 2022;10:120701.
Nowak WJ, Serafin D, Wierzba B. Effect of surface mechanical treatment on the oxidation behavior of FeAl-model alloy. J Mater Sci. 2019;54:9185–9196.
Cao C, Feng Y, Qiu T, et al. Effects of isothermal annealing on the oxidation behavior, mechanical and thermal properties of AlN ceramics. Ceram Int. 2017;43:9334–9342.
Pillai R, Romedenne M, Haynes JA, Pint BA. Oxidation Behavior of Candidate NiCr Alloys for Engine Exhaust Valves: Part I—Effect of Minor Alloying Elements. Oxidation of Metals. 2021;95:157–187.
Pérez-González FA, Garza-Montes-de Oca NF, Colás R. High Temperature Oxidation of the Haynes 282© Nickel-Based Superalloy. Oxidation of Metals. 2014;82:145–161.
Dudziak T, Deodeshmukh V, Backert L, et al. Phase Investigations Under Steam Oxidation Process at 800 °C for 1000 h of Advanced Steels and Ni-Based Alloys. Oxidation of Metals. 2017;87:139–158.
Padture NP, Gell M, Jordan EH. Thermal barrier coatings for gas-turbine engine applications. Science (1979). 2002;296:280–284.
Montross CS, Wei T, Ye L, Clark G, Ma Y-W. Laser shock processing and its effects on microstructure and properties of metal alloys: a review. Int J Fatigue. 2002;24:1021–1036.
Miller MP, McDowell DL, Oehmke RLT. A Creep-Fatigue-Oxidation Microcrack Propagation Model for Thermomechanical Fatigue. J Eng Mater Technol. 1992;114:282–288.
Wang J, Xue H, Wang Y. Oxidation behavior of Ni-based superalloy GH738 in static air between 800 and 1000 °C. Rare Metals. 2021;40:616–625.
Shin K-Y, Kim J-H, Terner M, Kong B-O, Hong H-U. Effects of heat treatment on the microstructure evolution and the high-temperature tensile properties of Haynes 282 superalloy. Materials Science and Engineering: A. 2019;751:311–322.
Ma C-H, Huang J-H, Chen H. Residual stress measurement in textured thin film by grazing-incidence X-ray diffraction. Thin Solid Films. 2002;418:73–78.
Wang A-N, Chuang C-P, Yu G-P, Huang J-H. Determination of average X-ray strain (AXS) on TiN hard coatings using cos2αsin2ψ X-ray diffraction method. Surf Coat Technol. 2015;262:40–47.
Swanson HE. Standard x-ray diffraction powder patterns. Circular of the Bureau of Standards. 1953;539:1–95.
McMurdie HF, Evans EH, Morris MC, et al. Standard X-Ray Diffraction Powder Patterns from the JCPDS Research Associateship. Powder Diffr. 1986;1:64–77.
Corliss LM, Hastings ; J M, Hastings JM, Bertaut EF. Magnetic Structure of Manganese Chromite. J Appl Phys. 1962;33:1138–1138.
Newnham RE, Fang JH, Santoro RP. Crystal structure and magnetic properties of CoTiO3. Acta Crystallogr. 1964;17:240–242.
Martin K, McCarthy G, Grant-in-Aid I. North Dakota State University. Fargo, ND, USA, ICDD, Grant-in-aid. 1991;
Wang A-N, Huang J-H, Hsiao H-W, Yu G-P, Chen H. Residual stress measurement on TiN thin films by combing nanoindentation and average X-ray strain (AXS) method. Surf Coat Technol. 2015;280:43–49.
Rumpf Dr-IngB. Thermochemical data of pure substances. Vet Immunol Immunopathol. 1997;55:359–360.
Lobnig RE, Schmidt HP, Hennesen K, Grabke HJ. Diffusion of cations in chromia layers grown on iron-base alloys. Oxidation of Metals. 1992;37:81–93.
Jian P, Jian L, Bing H, Xie G. Oxidation kinetics and phase evolution of a Fe–16Cr alloy in simulated SOFC cathode atmosphere. J Power Sources. 2006;158:354–360.
Yankin A, Vikhreva O, Balakirev V. P–T–x diagram of the Co–Ti–O system. Journal of Physics and Chemistry of Solids. 1999;60:139–143.
Lu X, Yang J, Yan M, Guo C, Wang J, Ma S. Isothermal oxidation behavior of Ni-Al-Co-Cr-Mo-Ti alloy at 850°C and 950°C. Procedia Eng. 2012;27:932–938.
RAPP RA. Kinetics, Microstructures and Mechanism of Internal Oxidation - Its Effect and Prevention in High Temperature Alloy Oxidation. Corrosion. 1965;21:382–401.
Zhao W, Gleeson BM. Steam effects on oxidation behavior of alumina-scale forming nickel-based alloys and a kinetics analysis of complex scale evolution during isothermal oxidation. ProQuest Dissertations and Theses. United States -- Pennsylvania; 2012.
Giggins CS, Pettit FS. Oxidation of Ni-Cr-Al Alloys Between 1000° and 1200°C. J Electrochem Soc. 1971;118:1782.
Sabioni ACS, Huntz AM, Millot F, Monty C. Self-diffusion in cr ₂ o ₃ III. Chromium and oxygen grain-boundary diffusion in polycrystals. Philosophical Magazine A. 1992;66:361–374.
Kofstad P. Note on the defect structure of rutile (TiO2). Journal of the Less Common Metals. 1967;13:635–638.
Kofstad P. High-temperature oxidation of titanium. Journal of the Less Common Metals. 1967;12:449–464.
Rumble J. CRC Handbook of Chemistry and Physics, 100th Edition. CRC Press. 2019;
Shackelford JF, Han Y-H, Kim S, Kwon S-H. CRC Materials Science and Engineering Handbook. CRC Materials Science and Engineering Handbook. 2016. doi:10.1201/b18971.
Kruger KL. HAYNES 282 alloy. In: Materials for Ultra-Supercritical and Advanced Ultra-Supercritical Power Plants. Elsevier; 2017:2017:511–545.
Dudziak T, Boron L, Gupta A, et al. Steam oxidation resistance and performance of newly developed coatings for Haynes® 282® Ni-based alloy. Corros Sci. 2018;138:326–339.
Nix WD, Clemens BM. Crystallite coalescence: A mechanism for intrinsic tensile stresses in thin films. J Mater Res. 1999;14:3467–3473.
Tung H-M, Stubbins JF. Incipient oxidation kinetics of alloy 617 and residual stress of the oxide scale formed in air at temperatures between 850 and 1000°C. Journal of Nuclear Materials. 2012;424:23–28.
Tung H-M, Stubbins JF. Incipient oxidation kinetics and residual stress of the oxide scale grown on Haynes 230 at high temperatures. Materials Science and Engineering: A. 2012;538:1–6.
Bedworth RE, Pilling NB. The oxidation of metals at high temperatures. J Inst Met. 1923;29:529–591.
Huntz AM, Boumaza A, Moulin G. Improvement of the oxidation resistance of superalloys by low-pressure, plasma-sprayed MCrAlYTa coatings. I. Oxidation of Metals. 1988;30:141–183.
Shannon RD. Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides. urn:issn:0567-7394. 1976;32:751–767.

No competing interests reported.

Evolution of Oxide Phases and Residual Stress in Haynes 282 Superalloy During Long-Term High-Temperature Oxidation

Status:

Version 1

Abstract

Figures

1. Introduction

2. Experimental details