Effect of Heat Treatment on Thermal Corrosion Behavior of Laser Selective Melting GH4099 Nickel-Based High-Temperature Alloy at 900°C (75% Na₂SO₄ + 25% NaCl)
Release time:
2025-11-07
Currently, with the rapid development of China's aerospace and defense industries, the demand for high-performance aircraft is growing increasingly urgent. As the core component of an aircraft, the performance of an aero engine directly determines the upper limit of the entire aircraft's capabilities. However, current aeroengine performance is primarily constrained by material limitations and design/R&D capabilities. GH4099, a nickel-matrix alloy with Co, Mo, Cr, and W as solid solution strengthening elements and Al and Ti as γ9 phase precipitation strengthening elements, exhibits excellent mechanical properties and corrosion resistance at elevated temperatures. It is an ideal material for hot-end components in aeroengines [1,2].
Selective Laser Melting (SLM) technology is a form of Additive Manufacturing (AM) [3]. By sequentially melting metal powder layer by layer, SLM not only enables precise control over component geometry but also allows optimization of alloy microstructure through process parameter adjustments [4,5]. Compared to traditional casting and machining techniques, SLM offers significant advantages such as the ability to produce complex geometries, improved material utilization, and reduced production cycles. It eliminates many geometric constraints, thereby providing greater design freedom [6]. This technology demonstrates immense application potential in fields like aerospace where material performance demands are extremely high, meeting the design requirements for complex structures in aeroengines.
To ensure SLM-manufactured GH4099 alloys meet service requirements, research has focused on process parameters, microstructure, and mechanical properties. Zhang et al. [7] investigated the effects of scan speed on microstructure and mechanical properties, demonstrating that increasing scan speed refines grains, reduces cellular grain size, and decreases texture strength, thereby enhancing overall strength. Li Fan et al. [8] identified a process window for SLM GH4099 that produces crack-free parts with low porosity. After heat treatment, its tensile properties at room temperature and 700°C outperformed cold-rolled sheet processes, though high-temperature tensile plasticity at 900°C was inferior to cold-rolled sheet. Zhang et al. [9] demonstrated that combining hot isostatic pressing with heat treatment enhances the mechanical properties of SLMGH4099 alloy. Lu et al. [10] measured warpage in SLM-produced GH4099 components and those annealed at different temperatures using a 3D scanner, indicating that internal residual stresses are nearly eliminated when annealing temperatures exceed 1000°C. In summary, current research primarily focuses on the effects of process parameters and heat treatment on the microstructure and mechanical properties of GH4099. However, for GH4099 alloy components operating in complex service environments, in addition to possessing adequate high-temperature mechanical properties, excellent corrosion resistance is also essential [11]. Since GH4099 alloy components operate primarily under high-temperature and complex stress conditions, prolonged exposure to high-temperature sulfur-containing fuels and saline environments accelerates oxidation due to sulfate deposits formed on the alloy surface during fuel combustion, making them susceptible to thermal corrosion [12,13]. The microstructure of SLM GH4099 alloy produced by selective laser melting (SLM) differs significantly from that of heat-treated SLM GH4099, resulting in notable variations in mechanical properties. However, no studies have yet investigated the corrosion resistance of SLMGH4099. Therefore, it is necessary to examine the thermal corrosion behavior of SLM-processed GH4099 alloy from a service environment perspective, elucidate its thermal corrosion mechanism, and provide theoretical foundations and technical support for the practical application of SLMGH4099 components.
In this work, commercially available GH4099 powder printed with identical process parameters was used as the sample. The thermal corrosion behavior of SLMGH4099 alloy before and after heat treatment was investigated in a 75% Na₂SO₄ + 25% NaCl mixed salt solution at 900°C. Thermal corrosion products and corrosion morphologies were characterized using optical microscopy (OM), X-ray diffraction (XRD), and scanning electron microscopy (SEM)/energy-dispersive X-ray spectroscopy (EDS), with the thermal corrosion mechanism investigated.
1 Experimental Methods
Spherical GH4099 alloy powder prepared via vacuum induction gas atomization (VIGA) exhibits particle sizes ranging from 15 to 53 μm. The alloy composition of SLMGH4099 (atomic fraction, %) is: Cr 18.43, Co 6.49, W 6.06, Mo 4.35, Al 1.98, Ti 1.19, Fe 0.05, C 0.046, Ni balance. SLMGH4099 was fabricated using Guangdong Hanbang Laser Technology Co., Ltd.'s SLM equipment HBD150. The forming process parameters were: laser power 260W, scan rate 1050 mm/s, scan spacing 70 μm, powder layer thickness 40 μm. A strip scanning strategy was employed, with the scan direction rotated 67° after each completed layer. Concurrently, the substrate was preheated to 100°C. Throughout the manufacturing process, the O₂ content within the build chamber was strictly controlled to ensure it remained below 200 mg/L. The HTGH4099 specimen was obtained by subjecting the SLMGH4099 specimen to a 1-hour solution treatment at 1110°C, followed by air cooling to room temperature, then tempering at 800°C and aging for 8 hours, after which it was air-cooled. After cutting, grinding, and polishing both pre- and post-heat-treated specimens, the microstructures were examined using a metallographic etchant comprising 5 g CuSO₄ + 100 mL C₂H₅OH + 100 mL HCl.
SLM and HT specimens were machined into 10 mm × 10 mm × 2 mm thin sections via electrical discharge machining (EDM). These sections were then ground with abrasive paper of increasing grit sizes up to 2000 grit. Specimen surfaces were cleaned using anhydrous ethanol ultrasonic cleaning and dried for subsequent use. For thermal corrosion testing, the salt coating method was employed. A saturated salt solution was prepared by weighing and mixing 75% Na₂SO₄ and 25% NaCl according to the specified ratio. After weighing, the specimens were placed on a metal heating plate (approximately 120°C). The saturated salt solution was evenly sprayed onto the specimen surfaces using a spray bottle. After complete evaporation of the moisture, a thin salt film adhered to the specimen surfaces. The salt coating content was controlled at (1.0±0.1) mg/cm². The coated specimens were placed in a box-type resistance furnace and heated to 900°C. To prevent salt volatilization during thermal corrosion, specimens were removed every 20 hours, washed with boiling water, dried, weighed, and recoated to continue the thermal corrosion cycle. The holding times were 5, 10, 20, 40, 60, 80, 100, and 120 hours, with two parallel specimens used for each test condition.
Mass measurements before and after corrosion were performed using an analytical balance with an accuracy of 0.1 mg. Corrosion kinetics curves were plotted with mass change per unit area on the vertical axis and corrosion time on the horizontal axis. A JSM-IT800 SEM equipped with EDS was employed to observe and analyze the microstructure of specimens before and after corrosion in backscattered electron (BSE) mode, along with the distribution of product elements. An F200X TEM was used for analysis and characterization of nanoscale precipitation phases. The phase composition of corrosion products was analyzed and identified using an Ultima VI XRD.
2 Results and Discussion
2.1 Microstructure of Specimens Before and After Heat Treatment
Figure 1 shows the microstructural morphology observed under scanning electron microscopy before and after heat treatment. The microstructure of SLMGH4099 exhibits typical laser selective melting characteristics, featuring numerous fine cellular subgrains within the grains. These subgrains form during SLM due to the extremely rapid solidification rate of the micro-melting pool, as shown in Figures 1a and 1b. Fine MC carbides, primarily enriched with Ti and C [14], are diffusely distributed near the cellular subgrains. As shown in Figures 1c and d, after heat treatment, the grains recrystallized with extensive twinning. Irregular M23C6 particles, rich in Cr, Mo, and C [15], precipitated along grain boundaries. Additionally, irregular M23C6 particles were dispersed within grains, and nanoscale γ9 phase was precipitated. Comparing SLM and HT specimens, heat treatment resulted in increased grain size, disappearance of MC carbides, precipitation of M23C6 along grain boundaries and within grains, and precipitation of γ9 phase. These microstructural changes inevitably affect the alloy's properties. As shown in Figure 2a, TEM analysis of the nano-precipitated phases reveals that the matrix phase of the SLM specimen is γ. No secondary phase diffraction patterns are observed in the selected area electron diffraction (SEAD) pattern. The EDS elemental distribution map shows minor Ti enrichment at subgrain boundaries and trace amounts of nano-MC carbides, consistent with SEM findings. After heat treatment, as shown in Figure 2b, γ9 phase was uniformly dispersed in the matrix, and granular M23C6 carbides precipitated at grain boundaries.
2.2 Corrosion Kinetics
Figure 3 shows the corrosion kinetics curves of SLMGH4099 alloy specimens before and after heat treatment, exposed to a 75% Na₂SO₄ + 25% NaCl mixed salt solution at 900°C for 120 hours: During the initial corrosion phase (0–20 hours), SLMGH4099 specimens before and after heat treatment exhibited similar behavior, showing no significant mass change after 20 hours of exposure in the molten salt hot corrosion environment. This is attributed to the minimal corrosion product formation during the initial stage, where the amount of corrosion products shed during air cooling and cleaning was insufficient to cause a noticeable weight change. During the mid-stage (20–40 h), after a second salt coating and 40 h of thermal exposure at 900°C, the untreated SLM specimen exhibited severe weight loss with the most pronounced loss trend. The heat-treated specimen also lost weight but to a significantly lesser extent, indicating that heat treatment enhanced the alloy's resistance to mid-stage corrosion weight loss. During the late corrosion stage (40–120 h), after 100 h of cyclic thermal corrosion, the untreated specimen exhibited a mass change per unit area of approximately -1.76 mg·cm⁻², while the heat-treated specimen HT showed approximately -1.35 mg·cm⁻², indicating markedly lower weight loss for the heat-treated specimen. Throughout the process, both types of specimens initially experienced mass loss due to reactions between the molten salt and the matrix, as well as the shedding of corrosion products during air cooling and cleaning. As cyclic salt coating and high-temperature exposure continued, corrosion intensified and loose products increased, leading to corrosion-induced weight loss. However, the heat-treated specimens exhibited superior corrosion kinetics at all stages, demonstrating the positive effect of heat treatment on enhancing the thermal corrosion resistance of the SLMGH4099 alloy.
2.3 Corrosion Product Analysis
2.3.1 XRD Analysis
Figure 4a shows the XRD patterns of SLMGH4099 specimens before and after heat treatment. The matrix phases primarily consist of γ and γ9 phases, with the SLM specimen exhibiting strong <001> orientation texture. This is attributed to the high cooling rate during SLM and the inherent characteristics of nickel-based alloys [16]. Figure 4b shows the XRD patterns of SLM and HT specimens after 60 h of thermal corrosion. After three cycles of salt coating corrosion in a mixed Na₂SO₄ and NaCl salt solution at 900°C for 60 h, the peak intensities of corrosion products indicate that only trace amounts of TiO₂, CoO, NiO, NiCr₂O₄, Cr₂O₃ on the SLM specimen surface. The prominent γ/γ9 peaks in the matrix indicate that the SLM specimen failed to form an effective protective film. In contrast, the HT specimen exhibited higher XRD diffraction peaks for corrosion products such as TiO₂, NiO, Cr₂O₃, NiCr₂O₄, and CoCr₂O₄ after heat treatment. Figure 4c shows the XRD surface patterns of SLM and HT specimens after 120 hours of thermal corrosion. Both specimens exhibit identical corrosion products: TiO₂, NiO, NiCr₂O₄, CoCr₂O₄, and Cr₂O₃. Comparing Figures 4b and 4c reveals that after 60 and 120 hours of thermal corrosion, the SLM specimen exhibits an extremely strong γ matrix phase peak, indicating that the matrix was exposed at the surface without forming a dense oxide film to protect it. In contrast, the HT specimen showed multiple oxide products on its surface after 60 hours of thermal corrosion testing. Following 120 hours of thermal corrosion, repeated thermal corrosion cycles caused the protective oxide film to fail.
2.3.2 Surface Morphology Analysis
Figure 5 shows the macro- and micro-morphologies of the specimens before and after heat treatment following 60 hours of cyclic thermal corrosion in a mixed Na₂SO₄ and NaCl salt solution. The corresponding EDS results for the respective regions are presented in Table 1. As shown in Figures 5a and 5b, the oxide films on both specimen surfaces have extensively peeled off, with only a small amount of blocky corrosion products remaining. The substrate is exposed on the surface, revealing fine pores on its surface. Figures 5c and d present high-magnification SEM images of the exposed substrate beneath the detached oxide layers on both specimens. Micro-pores ranging from 1 to 5 μm in size are discernible within the substrate. EDS analysis of the SLM specimen in region A indicates that the exposed substrate beneath the detached oxide layer consists primarily of Ni-rich oxidation products. Similarly, the exposed substrate beneath the detached oxide layer in region D of the HT specimen is dominated by Co-rich oxides. Figures 5e and f present high-magnification SEM images of the oxide layers on the surfaces of the two specimens. EDS analysis of the SLM specimen's oxide layer indicates it primarily contains O, Al, Ti, Cr, Co, and Ni, likely comprising Al₂O₃, TiO₂, Cr₂O₃ oxides, and NiMoO₄, NiCr₂O₄, CoCr₂O₄ spinels. Figure 6 shows the macro- and micro-morphologies after 120 h of thermal corrosion in a mixed Na₂SO₄-NaCl salt solution. Compared to other stages, the surface retains more oxidation products, consistent with corrosion kinetics and XRD results. As shown in Figures 6a and 6b, both specimens exhibit flaking of oxide products. Compared to the 60-hour specimens, the NiCr₂O₄ and CoCr₂O₄ spinels are finer-grained, with an oxide layer beneath them.
2.3.3 Cross-Section Analysis of Corrosion
Figure 7 presents the SEM image and EDS spectrum of the cross-section of the SLM specimen after 120 hours of corrosion in a mixed Na₂SO₄ and NaCl salt solution at 900°C. Combined with the distribution of different corrosion products along the longitudinal section shown in Figure 8, the results indicate that after thermal corrosion, the matrix and oxides of both SLM and HT specimens form four distinct layers: Ni and Co are distributed in the outermost layer, identified by XRD analysis as NiMoO₄, NiCr₂O₄, and CoCr₂O₄ spinels; Cr and Al are distributed in the intermediate layer, with the Cr₂O₃ layer overlying the Al₂O₃ layer. The bottom layer is a Ni-rich layer, where Ni₃S₂ diffuses near the grain boundaries. The exposed substrate exhibits a Ni-rich layer approximately 100 μm deep, composed of numerous pores. This phenomenon is not observed in areas protected by the oxide film. EDS results in Figures 7 and 9 reveal grain boundary corrosion in the matrix of both specimens, with the corroded grain boundary regions primarily composed of S, O, Al, and Ni elements.
2.4 Analysis and Discussion
The microstructure of GH4099 alloy produced by selective laser melting exhibits unique characteristics, featuring only a small amount of MC carbides and no significant element segregation. This is attributed to the extremely high cooling rate of the molten pool during SLM (10⁶–10⁸ K/s [17]), which effectively reduces solute atom segregation and suppresses the precipitation of γ, and carbides. Consequently, the alloy remains in a supersaturated state with high solute element content in the matrix, resulting in significant differences in carbide number density and size compared to heat-treated alloys [18].
Thermal corrosion kinetics curves indicate that SLM specimens exhibit faster weight loss than HT specimens within 0–60 h. After initial thorough reaction between molten salt and specimen surfaces, corrosion products detach, forming pits and micro-pores. Subsequent cyclic thermal corrosion allows molten salt to penetrate these defects, accelerating corrosion. Both specimens show significant corrosion after 120 h of thermal corrosion at 900°C.
Due to the high Mo content in GH4099 alloy, Mo readily forms MoO₃ at 900°C during oxygen-containing thermal corrosion. This MoO₃ reacts with metal oxides, compromising the oxide layer's integrity and causing the corrosion layer to crack and peel off. Additionally, oxide dissolution occurs at the interface between the oxide layer and molten salt, as described by Equations (1) and (2).
Furthermore, at the molten salt-atmosphere interface during thermal corrosion, oxide precipitation occurs as shown in Equation (3):
This causes the dense oxide layer to dissolve on the inner surface of the molten salt while forming loose corrosion products on the outer surface.
As shown in Figures 7–9, the corrosion products form four distinct layers. The outermost layer consists of spinel phases NiMoO₄, NiCr₂O₄, and CoCr₂O₄, which are formed through solid-state reactions between metal oxides [19]. This layer is ineffective at preventing molten salt from penetrating the substrate. The molten salt may then penetrate this layer to reach the Cr₂O₃ layer, where it reacts with molten Na₂SO₄ to decompose into Na₂O, S, and O₂. This molten alkaline reaction with Cr₂O₃ causes the originally dense Cr₂O₃ layer to become loose and porous. Furthermore, NaCl penetrating the substrate reacts with it to generate volatile chlorine gas. The accumulated high-pressure chlorine gas breaks through the oxide film and diffuses outward, further promoting cracking of the oxide film [20], as shown in Equations (4) and (5):
During cyclic thermal corrosion, whenever the specimen is removed from the furnace and cooled in air, the Cr₂O₃ layer—already eroded by the molten salt—fractures and peels away from its internal pores. This exposes large areas of the substrate, a phenomenon clearly visible in Figures 5 and 7. During the subsequent salt-coated thermal corrosion cycle, the substrate re-engages with the molten salt, perpetuating corrosion deterioration and progressively impairing material performance.
EDS analysis results in Figures 8 and 9 reveal that the primary elements in the corroded grain boundary regions include S, O, Al, and Ni. Both SLM and HT specimens exhibited severe grain boundary corrosion, clearly indicating sulfidation and oxidation reactions at these interfaces. This reasonably suggests that during thermal corrosion, sulfur and oxygen rapidly diffuse within the alloy via existing defects, significantly accelerating corrosion rates. Moreover, the chlorine gas generated during these reactions causes cracking and spalling of the corrosion layer. This further promotes the diffusion of S and O through defects into the matrix, continuously deepening the corrosion. Ultimately, this leads to the formation of internal oxidation and sulfidation zones within the alloy, severely compromising its performance and structural integrity.
During cyclic thermal corrosion, pores and cracks form on the specimen surface after the first cycle. Subsequent thermal corrosion cycles with salt recoating exhibit significantly accelerated corrosion rates. Compared to SLM specimens, heat-treated HT specimens demonstrate superior thermal corrosion resistance. This may be attributed to the precipitation of M23C6 and γ phases during heat treatment. These phases create additional interfaces between the phases and the matrix, establishing rapid transport pathways for Cr and Al diffusion. These precipitates, rich in Cr, Al, and Ti, can rapidly migrate to the surface and aggregate, forming a dense oxide film that effectively inhibits further corrosion.
As key pathways for elemental diffusion, the equiaxed grain structure formed after heat treatment provides more efficient transport routes for elemental diffusion. Simultaneously, Cr and Al from the precipitates diffuse into the matrix and react with O₂. The elemental gradient distribution between the phase and matrix accelerates this reaction, ultimately forming a mixed Cr₂O₃ and Al₂O₃ oxide layer in the HT samples, enhancing the material's corrosion resistance.
3 Conclusions
(1) During thermal corrosion in 75% Na₂SO₄ + 25% NaCl solution, the corrosion products formed in laser-selective melting GH4099 alloy primarily consist of oxides and sulfides of Ni, Ti, Al, and Cr, along with reaction products between oxides such as NiCr₂O₄ and CoCr₂O₄ spinels. The SLM GH4099 alloy exhibits a unique microstructure featuring cellular grains and fine MC carbides, which undergo significant changes after heat treatment, such as grain growth and phase precipitation. Regarding thermal corrosion performance, heat-treated specimens demonstrate relatively better resistance due to the presence of M23C6 and γ′ phases, while SLM specimens experience faster initial weight loss owing to their supersaturated state and microstructural characteristics.
(2) The thermal corrosion mechanism of SLM GH4099 alloy before and after heat treatment follows a typical acidic dissolution corrosion pathway. During corrosion, substances like MoO₃ disrupt the oxide layer, while molten salt reacts with the substrate, causing the oxide layer to crack and peel off. Sulfidation and oxidation reactions occur at grain boundaries, with S and O diffusion accelerating corrosion. Generated chlorine gas further promotes changes in the corrosion layer, and cyclic thermal corrosion continuously worsens the specimen's condition.
Reference: Journal of Chinese Corrosion and Protection, Vol. 45, No. 6 Effect of Heat Treatment on the Thermal Corrosion Behavior of Laser Selective Melting GH4099 Nickel-Based High-Temperature Alloy at 900°C (75% Na₂SO₄ + 25% NaCl)
As a national high-tech enterprise, Stardust Technology utilizes radiofrequency plasma spheroidization technology to produce spherical powders of rare refractory metals for its selective laser melting (SLM) process. The product portfolio encompasses elemental and alloy powders of tungsten, molybdenum, tantalum, niobium, and others. These powders feature high purity with low oxygen content, high sphericity, controllable particle size, and excellent flowability, with minimal internal defects and no satellite spheres. These powders are optimized for SLM processes and related forming requirements, finding extensive applications in aerospace, defense, medical devices, and other sectors. We provide integrated solutions from powder preparation to application, meeting the stringent material performance demands of high-end manufacturing. For detailed product specifications, please contact our specialist, Manager Cathie Zheng, at +86 13318326187.
News
Stardust Technology (Guangdong) Co., Ltd.
101, Building 1, Liandong Youzhi Zone, Senshuji Road, Nansha Community, Danzao Town, Nanhai District, Foshan City,Guangdong Pro.,China
Tel：13318326187; 13378626726；
QR code