Effect of Solid Solution Treatment Routes on the Microstructure Configuration of a Third-Generation Ni-Based Superalloy

Abstract

The step and ramp heat treatments can improve the homogenization degree of the structure in superalloys. However, the comparison between them using the same temperate range and total holding time on the microstructure and the microstructural stability is still vague. The effects of step and ramp heat treatment in the same solution temperature range and the same hold time on the microstructural stability in a third-generation superalloy were studied. The sizes of γ′ phases at dendrite cores and dendrite edges are different, which is caused by the dissolution degree of the γ′ phase during solution treatment. A low degree of γ′ phase dissolution aggravates the dendritic segregation. Dendrite segregation makes refractory elements such as Co, Mo, W, and Re aggregate in the dendrite core, and the TCP phase is easy to nucleate and precipitate during long-term thermal exposure, resulting in decreased structural stability. The investigations in this study compare the differences and obtain optimized processes and parameters for the improvement of superalloys.

1. Introduction

Nickel-based single-crystal superalloys are widely used in the aerospace field due to their excellent mechanical properties, microstructural stability, and oxidation resistance [1,2,3]. Adding refractory elements is the main strategy for developing nickel-based single-crystal superalloys [4,5,6,7]. They can improve the service life of the alloy as the increase in solid solution has an effect [8,9]. However, with the excessive addition of refractory elements, the topologically close-packed (TCP) phase will precipitate for local supersaturation in the matrix [10,11,12,13]. The precipitation of the brittle TCP phase can consume the strengthening elements within the matrix and cause stress concentrations. These will decrease the service life of the alloy [14,15].

Nickel-based single-crystal superalloys are commonly fabricated using the directional solidification method [16,17,18,19]. In such a complex alloy system, the microstructure formed after solidification is often non-uniform. As a result, the as-cast alloy formed by this method exhibits a dendritic structure, along with the formation of a eutectic phase [20,21,22]. The solid-solution heat treatment is a major process used to eliminate the dendritic and eutectic structures [23,24,25]. Typically, two types of solid-solution heat treatment are used. One is the step-by-step heat preservation treatment (denoted as step heat treatment) at different temperatures, and the other is the continuous heating treatment (denoted as ramp heat treatment) without staged holding [26,27,28].

The studies on the step heat treatment mainly focus on the influence of different parameters on the degree of homogeneity. The results indicate that a longer holding time at higher temperatures can produce a better homogenization effect. For example, as Pang et al. [29] investigated the holding time of the CMSX-10 alloy, they found that the area of the inter-dendritic region decreased from 21.7% to 0.02% by adjusting the time from 15 min to 20 h in the temperature range of 1354 °C to 1365 °C. However, under the same temperature range and holding time, previous studies have reported inconsistent conclusions regarding the effects of step treatment and ramp heat treatment on the homogenization of superalloy microstructures. Zhang et al. compared the step heat treatment and the ramp-couple step heat treatment in the same temperature range and total holding time. They found the segregation coefficients of Re and W in the dendrite structure to be much higher after the step heat treatment [27]. Similarly, Liu et al. also compared the step heat treatment and the ramp-couple step heat treatment in the same temperature range and nearly the same total holding time. They found the segregation coefficients of Al, Ta, Re, and W elements in the dendrite structure to be almost the same [28]. Hedge et al. compared the microstructures resulting from step heat treatment and ramp heat treatment within nearly the same temperature range but with different total holding times. The results indicate that the sample subjected to step heat treatment exhibited a lower volume fraction of the eutectic structure despite the shorter total holding time [26].

Both the step and ramp heat treatments can affect the degree of homogenization and the microstructure of the superalloy. However, within the same heat treatment temperature range and holding time, the impact of different solution treatment processes on microstructure stability and its underlying mechanism remained unclear. In this study, we examined the effects of various solution treatment processes on microstructure stability and optimized the processes and parameters needed to enhance the properties of superalloys.

2. Materials and Methods

In this study, a third-generation nickel-based single-crystal superalloy was used, with its nominal chemical composition listed in

Table 1. Nominal chemical composition of the alloy (wt. %).

Elements	Co	Cr	Mo	W	Re	Al	Ti	Ta	Ni
wt. %	10.0	4	1	6.0	5	6	1	8.5	Bal.

. The alloy was prepared by the directional solidification method after pouring it into a vacuum-induction melting furnace. The [001] oriented single-crystal bars were solidified by the crystal selection method. The size of the bars was Φ16 mm × 160 mm, and the alloy was cut into cylindrical samples of Φ16 mm × 5 mm using the electric spark cutting technology.

The solution heat treatment procedures are systematically outlined in

Table 2. Heat treatment procedure of different step and ramp heat treatments.

Type	Symbol	Heat Treatment Procedure
Step solution	Step-1	1335 °C/2 h + 1345 °C/4 h + 1355 °C/6 h
	Step-2	1335 °C/4 h + 1345 °C/4 h + 1355 °C/4 h
	Step-3	1335 °C/6 h + 1345 °C/2 h + 1355 °C/4 h
Ramp solution	Ramp-1	1335 °C→1.67 °C/h→1345 °C→1.67 °C/h→1355 °C
	Ramp-2	1335 °C→2.5 °C/h→1345 °C→1.25 °C/h→1355 °C
	Ramp-3	1335 °C→1.25 °C/h→1345 °C→2.5 °C/h→1355 °C

, where different heat treatment processes are denoted by specific symbols for clarity. All procedures were conducted within a controlled temperature range of 1335 °C to 1355 °C to ensure consistent thermal exposure, with a total holding time of 12 h. The measurement error for the temperature was within ±3 °C. After heat treatment, each specimen underwent prolonged high-temperature exposure at 1100°C for up to 500 hours to assess microstructural stability. Scanning Electron Microscope (SEM) studies were conducted using a Quanta 650F (Thermo Fisher Scientific, Waltham, MA, USA) instrument operated at 30 kV of acceleration voltage performed for the characterization of the microstructure under different heat treatment conditions and after long heat exposure. The SEM is equipped with an Energy-Dispersive Spectroscopy (EDS) and model EDAX Octane Elect EDS System (EDAX, Beijing, China). Transmission Electron Microscopy (TEM) studies were conducted using a Talos F200X-G2 (Thermo Fisher Scientific, MA, USA) instrument, operating at 200 kV with a resolution limit of 0.14 nm. Partial TEM characterization was performed using High-Angle Annular Dark Field (HAADF) imaging in Scanning Transmission Electron Microscopy (STEM) mode, along with EDS analysis. Thermo Scientific Velox software is used to enable fast and easy data collection and analysis. The EDS module is equipped with four in-column SDD Super-X detectors. The EDS instrument was calibrated using the traditional “Cliff-Lorimer” method with the k-factor. It provides a maximum energy resolution of 136 eV and an effective detection area of 120 mm². This setup enables rapid and precise nanoscale characterization, with composition measurements regularly calibrated to achieve a precision of 0.1 wt. % for common metallic elements. The SEM specimens were chemically etched in a solution composed of 5 g CuSO₄, 20 mL HCl, and 100 mL H₂O to reveal microstructural details. The TEM specimens were prepared using twin-jet electrochemical polishing in an electrolyte consisting of 5% HClO₄ and 95% C₂H₅OH at −30 °C, ensuring high-quality thinning for TEM analysis. To quantify the γ′ phase size, Image-Pro Plus (Version 6.0, Media Cybernetics, Sliver Spring, Rockville, MD, USA) software was employed for precise image analysis and statistical measurements. Each data point was obtained by measuring more than 100 γ′ phases.

3. Results and Discussions

Figure 1 presents the SEM image showing the morphology of the dendritic structure, along with the corresponding elemental distribution maps obtained by EDS. The dendritic core (marked with a white dotted line) exhibits a cross-shaped morphology, with the inter-dendritic regions located at the junctions between multiple dendritic cores. Due to significant compositional variations between the dendritic core and the inter-dendritic regions, the dendritic cores appear light gray in the image, whereas the inter-dendritic regions appear dark gray. Cr, Co, W, and Re are preferentially concentrated in the dendritic core, while Al, Ta, and Ti are preferentially enriched in the inter-dendritic regions. The enrichment of Cr, Mo, W, and Re in the dendritic core occurs due to their strong segregation tendencies during solidification. With their high melting points, these elements solidify earlier and become trapped in the core. Additionally, their low diffusion rates restrict redistribution, resulting in their sustained concentration in the core region.

After the solution heat treatment, the uniformity of the alloy improved, and the contours of the dendritic structure became blurred, although they remained distinguishable under the SEM observation. Figure 2 presents the microstructures resulting from different solution heat treatment conditions. They all consist of the γ′ and γ phases, and the sizes of γ′ phase are larger in the dendritic edge than in the dendritic core. The differences are that they exhibit different morphologies and sizes of the γ′ phase. Among them, their γ′ phase is in cuboidal shape after the step-1 and ramp-2 solution heat treatments, as shown in Figure 2a,b,i,j. The γ′ phase is in a cuboidal shape with round corners after the step-2, step-3, and ramp-1 solution heat treatment, as shown in Figure 2c–h. Mostly, the γ′ phase is in the butterfly shape after the ramp-3 solution heat treatment, as shown in Figure 2k,l.

Figure 3 shows the sizes of the γ′ phase in the dendritic core and edge regions of Figure 2. The sizes of the γ′ phase follow the subsequent order: cuboidal shape < the cuboidal shape with round corner < the butterfly shape. This is also consistent with the evolution of morphologies during the coarsening of the γ′ phase. After moderate thermal treatment, the morphological and size differences in the γ′ phase between the dendritic core and edge regions become distinct and readily discernible. The size and distribution of the γ′ phase in the dendritic core and edge regions tend to be the same. The size of the γ′ phase varies under different treatment processes, and the size variation in the γ′ phase under different treatment is as follows: Step-1 < Ramp-2 < Step-2 < Step-3 < Ramp-1 < Ramp-3.

The difference in the sizes of γ′ phase reflects the segregation in the dendritic structure. Figure 4a shows the EDS mapping of the microstructure in the dendritic core of the alloy after the step-1 heat treatment. Ni, Al, and Ta are rich in the γ′ phase, whereas Co, Re, Cr, Mo, and W are rich in the γ phase. Figure 4b shows the partition ratios of the elements between the dendrite core and dendrite edge in the alloys after step-1 and ramp-3 solution heat treatments. The partition ratio is $\mathrm{K}={\mathrm{C}}_{\mathrm{dendrite}\text{-}\mathrm{core}}/{\mathrm{C}}_{\mathrm{dendrite}\text{-}\mathrm{edge}}$. According to the dendritic edge containing a higher density of the γ′ phase, the γ-forming elements (Co, Re, Cr, Mo, and W) all have K > 1 (enriched in the dendrite core), and the γ′-forming elements (Ta and Al) all have K < 1 (depleted in the dendrite core). Consistent with the difference in the size of the γ′ phase, the partition ratio of all the elements is much higher in the alloy after the ramp-3 solution heat treatment than the step-1 solution heat treatment.

To induce the formation of the TCP phase, the solution heat-treated alloys were subjected to thermal exposure at 1100 °C. Figure 5 shows the evolution of the TCP phase during thermal exposure. The sizes of the γ′ phase increased with prolonged thermal exposure, and the morphologies evolved from a cuboidal to an irregular strip shape. Based on the precipitation time of the TCP phase, the alloys can be divided into three types. The first type is the step-1 solution heat treatment alloy, with no TCP phase precipitate even after 500 h of thermal exposure. The second type is step-2, step-3, and ramp-2 solution heat treatment alloys, which precipitate the TCP phase after 400 h. The third type is the ramp-1 and ramp-3 solution heat treatment alloys, which precipitate the TCP phase after 300 h. These differences indicate that the solid solution heat treatment has an important effect on the stability and precipitation behavior of the TCP phase. This effect may be attributed to dendritic segregation in the superalloy following solution treatment.

To explore the precipitation location of the TCP phase in the alloy, Figure 6 presents the microstructure of the heat-treated alloy after 500 h of treatment at 1100 °C. Figure 6a,b show the distribution of the TCP phase under low and high magnification, indicating that the TCP phase precipitates in the dendrite core region. The TCP phase was further characterized using TEM, as presented in Figure 6c. Diffraction calibration identified the TCP phase as the μ phase. Furthermore, Figure 6d shows the EDS image of the TCP phase and its surrounding region, where the results indicate that the TCP phase is enriched in Mo, Cr, Co, W, and Re elements. This result is highly consistent with the element partition ratio shown in Figure 4. The enrichment of Mo, Cr, Co, W, and Re in the dendrite core provides favorable conditions for TCP phase formation.

The different solid solution treatments create different levels of residual segregation in the dendritic structure. This promotes the precipitation of the TCP phase in the dendritic core of samples that contain higher residual segregation. This can help establish the relationship between the heat treatment procedure and residual segregation. Figure 7 shows the schematic image of different heat treatment procedures. The figure illustrates the integral area (S) of temperature changes over time under different heat treatment regimes. By comparing the temperature–time integral areas (S) after various heat treatment processes, it can be concluded, as shown in Figure 7a as S_step-1 > S_step-2 > S_step-3, and as shown in Figure 7b as S_ramp-2 > S_ramp-1 > S_ramp-3. Notably, a larger integral area corresponds to a small γ′ phase size, reduced dendritic segregation, and a low tendency for TCP phase precipitation. For instance, as illustrated in Figure 7, the temperature–time integral areas S_step-1 > S_ramp-2 and S_step-3 > S_ramp-3 suggest that the size variations in the γ′ phase follow the corresponding order: Step-1 < Ramp-2 and Step-3 < Ramp-3. This finding is consistent with the observations in Figure 2 and the statistical analysis in Figure 3. The correlation between the temperature–time integral area, the γ′ phase size, and TCP phase precipitation behavior in the solution heat treatment procedure diagram confirms its relationship with dendritic segregation. Specifically, a larger integral area corresponds to lower dendritic segregation and a more uniform alloy composition. Therefore, regulating the temperature–time integral area plays a crucial role in controlling residual segregation within the dendritic structure.

4. Conclusions

Our study systematically investigated the difference in the stability of alloys with different step and ramp heat treatments. The microstructure after solution heat treatment and the change in microstructure during long-term thermal exposure were characterized, and the influence of two different solution heat treatments on the stability of the alloy microstructure was revealed. This provides a new perspective for the formulation of the alloy solution process and the inhibition of the TCP phase precipitation. The main conclusions are the following:

(1)

After different heat treatment procedures, the shape and size of the γ′ phase between the dendritic core and edge regions tend to be the same, with the size difference between these two regions represented by the following sequence: Step-1 < Ramp-2 < Step-2 < Step-3 < Ramp-1 < Ramp-3.

(2)

The dissolution degree of the γ′ phase is determined by the thermal solid solution, and the high dissolution degree of γ′ phase decreases the size difference in the γ′ phase between the dendrite core and edge after solid solution treatment.

(3)

The degree of dendritic segregation is related to the size difference in the γ′ phase between the dendrite core and edge. The size variation in the γ′ phase is large in the alloy with severe dendritic segregation.

(4)

Dendrite segregation leads to the segregation of refractory elements such as Co, Mo, W, and Re at the dendrite core, which makes the TCP phase easy to nucleate and precipitate at this position during long-term thermal exposure.

(5)

The alloy with a long thermal solution time has improved microstructure stability.