CoCrFeNi-high entropy alloys (HEAs) have gained increasing attention recently due to their novel properties, including high thermal stability 1±5 and great toughness at cryogenic temperatures. 6±8 This base alloy, with small proportion of Al, tends to crystallize into single face centered cubic (fcc) solid solution. However, when the mole fraction of Al exceeds 0.5, the alloy decomposes into fcc and B2-bcc phases. 9±32 Controlling the phase stability and transformation behaviors is the key factor that impacts the applications of HEAs at high temperature/corrosion environments. Recent studies showed that the corrosion resistance of a dual phase (fcc + B2) AlxCoCrFeNi alloy deteriorated due to the localized corrosion pits formed along B2/fcc interfaces. 15,33 Butler and Weaver 34 suggested, however, that ordered NiAl±B2 promotes the formation of stable Al 2 O 3 at the surface, working as an Al-reservoir. Therefore, understanding the microstructure and composition stabilities of HEAs during thermal aging is crucial; a comprehensive understanding about the critical role of Al addition and its concentration on microstructure stability can provide important implications for searching and preparing HEAs with high thermal stability and promising corrosion resistance.
CoCrFeNi serves as starting point for many HEAs, such as Al±CoCrFeNi, Mn±CoCrFeNi, and Cu±CoCrFeNi, and therefore, their phase stability has widely been studied using ex-situ bulk heating experiments at different heating temperatures with prolonged annealing 1,3±5,7,35 and thermodynamic calculations. 9,19 There is continuing discrepancy, however, regarding phase separation and whether this alloy should be considered as a metastable HEA.
Several studies have reported that equimolar CoCrFeNi remained a single-phase solid solution 3±5 after annealing at temperatures between 700 and 900 ○ C. Butler and Weaver, 34 however, examined the microstructure evolution of CoCrFeNi using high supercooling solidification (adopting a glass-fluxing method), and the results showed the formation of secondary, disordered bcc phases within the fcc matrix although the study lacked chemical data for secondary phases. He et al. 1 studied the phase stability of this alloy after prolonged annealing at 750 ○ C for 800 h. Using conventional and high-resolution Transmission Electron Microscopy (TEM) imaging, the study reported the formation of dark contrast regions with nanolattice distortion. While there is no evidence for chemical decomposition from the energy dispersive spectrometry (EDS) maps, the authors stated that CoCrFeNi is metastable at 750 ○ C. It is well known that the addition of further alloying elements, in minor concentration, such as Al or Mn to the base CoCrFeNi alloy, destabilizes the single fcc phase and accelerates the phase separation. Previous studies reported the formation of B2 with the increase of the Al addition. 16,19 Among many literature studies concerning this topic, we will concentrate here on understanding the critical role of Al and its concentration on the chemical and structural stability of the CoCrFeNi base alloy using an in situ TEM heating technique. In our previous work, 9 we designed a study concentrating on the acceleration of the precipitation kinetics in Al 0.3 CoCrFeNi via an in situ TEM heating technique, through the formation of secondary phases by means of a high density of surface nucleation sites in the thin foil. Our study revealed the formation of an additional intermediate phase, where hundreds of hours are predicted for this phase to be observed during ex-situ experiments. 9 Therefore, applying a similar in situ TEM technique here provides a platform to study the influence of Al on the stability of CoCrFeNi alloys, tracking the early, intermediate, and final stages of secondary phase formation.
In this work, we showed that the quaternary CoCrFeNi HEA is unstable upon in situ TEM annealing at intermediate temperatures at which Cr-rich bcc precipitates formed. Furthermore, we investigated the influence of Al on the thermal stability of the base chemistry alloy. At temperatures between 700 and 900 ○ C, more complex co-precipitates consisting of NiAl±B2 and Cr-rich bcc phases were formed when the Al content exceeded seven atomic percent (at. %). Density functional theory (DFT) calculations support these experimental results by demonstrating the HEA to be unstable with respect to the T ≙ 0 K convex hull, only becoming stable with the addition of configurational entropy at elevated temperature. Moreover, they show that increased Al addition results in increasingly favorable formation of Cr-rich bcc and NiAl±B2 phases.
These results, both experimental and first principles, compare favorably with CALPHAD (CALculation of PHAse Diagrams) modeling in the literature, 19 though, with some key differences discussed further with our results. Notably, while CALPHAD is a powerful tool for making broad predictions, we here demonstrate a more thorough mechanistic understanding of the phase stability in this HEA system.
HEAs with the nominal compositions of CoCrFeNi, Al 0.1 CoCrFeNi, Al 0.3 CoCrFeNi, and Al 0.5 CoCrFeNi (hereafter referred to as Al 0.0 , Al 0.1 , Al 0.3 , and Al 0.5, respectively) were created via vacuum-induction melting for this study. The ingots underwent hot isostatic pressing (HIP) at 1204 ○ C and 103 MPa for 4 h. The as-HIPed alloys were heat treated at 1250 ○ C for 2 h for chemical homogeneity. The as-homogenized ingots were then sliced into ∼300 μm-thick samples by electrical discharge machining (EDM), followed by traditional mechanical polishing to a final thickness of 80±100 μm, and 3 mm diameter disks were punched. The disks were electropolished to electron transparency in a double-jet polishing system using the electrolyte of 5% perchloric acid +95% methanol at -40 ○ C (weight or volume percent). Samples were in situ heat treated in a JEOL 2100 LaB 6 TEM, equipped with a high-resolution pole piece, using a Gatan heating holder at a ramp rate of 0.5 ○ /s up to 700 and 900 ○ C and held there for 10 and 60 and 120 min at temperatures between 700 and 900 ○ C. Microstructural analysis was carried out on as-homogenized and annealed samples using Scanning Transmission Electron Microscopy (STEM) on a JEOL 2100 field emission TEM equipped with energy dispersive x-ray spectroscopy (EDS). The precipitates were analyzed using the ImageJ software in conjunction with AdobePS™ across many areas and samples to assure a reasonable level of statistical significance.
To supplement the experimental results herein, we also performed DFT calculations to explore thermodynamic driving tendencies. Disordered phases were approximated via special quasirandom structures (SQSs). SQSs were generated using an Alloy Theoretic Automated Toolkit (ATAT) 36 and are provided in the supplementary material. Fcc SQSs were constructed for two, four, and five components, bcc SQSs were constructed for two components, and B2 SQSs were constructed for three components, with the latter only mixing two components on one sublattice. Structures contained 20±40 atoms as constrained by the compositions considered in reactions 1 and 2. Among all structures, the maximum mean absolute error in cluster correlations from the random alloy up to tenth nearest neighbor pair and fourth nearest neighbor triplet was 0.037. Nonmagnetic DFT calculations were performed with the generalized-gradient approximation within the Perdew±Burke±Ernzerhof (GGA-PBE) 37±39 implementation using the Vienna ab initio simulation package (VASP). 40,41 A k-point mesh of 8000 k-points per reciprocal atom and a plane-wave energy cutoff of 520 eV were used. Methfessel±Paxton smearing of order one, with a smearing width of 0.1 eV, was used during the relaxation, and atomic positions were considered relaxed when forces on each atom were less than 0.01 eV/Å. The relaxations were followed by static calculations using the tetrahedron method with Blöchl corrections.
An in situ TEM heating technique was utilized to evaluate the stability of the single-phase CoCrFeNi HEAs at intermediate and high annealing temperatures, including the influence of Al on phase separation. We find that the increase in the Al content stimulates the formation of NiAl±B2 and Cr-rich phases. The knowledge about phase transition is a key factor in advancement and application of HEAs, where properties, in particular corrosion and mechanical behavior, can be highly influenced.
TEM characterizations revealed that all the alloys examined in this study (Al 0.0 , Al 0.1 , Al 0.3 , and Al 0.5 ) are single fcc phase alloys in the as-homogenized state. The corresponding Selected Area Diffraction (SAED) patterns (Fig. 1) from [112∥ Al0.0 , [113∥ Al0.1 , [013] Al0.3 , and [001] Al0.5 show the reflections of only an fcc phase. Our results are consistent with those reported in the literature, which show that Al 0.0 , Al 0.1 , 1 and Al 0.3 16,22 are single fcc phases in the as-received and homogenized state. Although, for the Al 0.5 alloy, no secondary phases are present in the TEM foil, the alloy was previously examined using scanning electron microscopy (coupled with EDS), and the precipitation of the NiAl±B2 phase was confirmed in the as-homogenized condition. 42 The EDS-measured chemical composition of as-homogenized AlxCoCrFeNi is listed in Table I.
Figure 2(a) shows the representative high-angle annular darkfield (HAADF)-STEM and EDS elemental maps of CoCrFeNi HEAs at 750 ○ C for 10 min. As shown in Fig. 2, the alloy is no longer single phase and phase separation occurred by the formation of secondary precipitates in the grain interior. Based on EDS analysis, the precipitates are 31 at. % Cr, 28 at. % Fe, 22.2 at. % Ni, and 18.37 at. % Co. These precipitates are either blocky or spherical morphology with an average size of ∼204 nm. To further analyze these precipitates, SADPs (supplementary material, Fig. 1) were utilized and show the reflection of [113∥ bcc refection from the Cr-rich phase along with [001] fcc reflection from the matrix, indicating the formation of the secondary bcc phase. Our results are consistent with the previous study by He et al., 1 which shows that CoCrFeNi is metastable after heating at 750 ○ C for 800 h. The study reported the formation of dark contrast regions (with minor lattice distortion, 0.1 nm) after aging. To confirm the phase separation and define these features, the study attempted to accelerate the phase separation by adding 2 at. % Al to the base alloy, and after heating at 750 ○ C for 800 h, Cr-rich precipitates formed along grain boundaries. In our study, by contrast, we utilized in situ TEM heating technique to track the formation of secondary phases in early, intermediate, and final stages. The technique has proven to accelerate the kinetics of phase separation by means of high density of nucleation sites within the thin film, 9 facilitating the formation of phases where the time required to form during conventional methods is still unknown. Using in situ TEM heating, Cr-rich bcc precipitates formed after 10 min heating at 750 ○ C, proving that the base CoCrFeNi alloy is unstable at intermediate temperature. The formation of bcc secondary phases in the base CoCrFeNi was previously reported in the as-solidified highly undercooled CoCrFeNi alloy. 7 The bcc precipitates enhanced the mechanical properties, where the compressive yield strength increased from 137 MPa for the single fcc phase CoCrFeNi HEAs to 455 MPa after the bcc phase formation. 8 As in the Al 0.0 case, secondary phases precipitated out of the disordered fcc in the Al 0.1 alloy after in situ TEM heating at 900 ○ C for 10 min [Fig. 2(b)]. The average size of these precipitates was around 299 nm. Additionally, the Cr concentration within the precipitates increased slightly to about 38 at. %, in comparison with Al 0.0 . The SADPs (supplementary material, Fig. 1) from the matrix show the reflections of [113∥ bcc zone axes from the Cr-rich precipitate along with [013] fcc from the matrix. Increasing the Al addition, a more complex precipitate, a coprecipitate, started to form, which was classified as a Cr-rich bcc structure encased by a NiAl±B2 phase. Figures 2(c) and 2(d) show the representative HAADF-STEM images, EDS, and SAED of Al 0.3 after in situ TEM heating at 700 ○ C for 1 and 2 h and at 900 ○ C for 1 h, respectively. The precipitation of secondary phases was also observed, similar to previous alloy compositions. In the 700 ○ C 1 h condition, secondary phases with an average size of 160 nm were formed in grain interiors. The chemical composition of these precipitates was examined by EDS and found to be 48.35 at. % Cr, 18.9 at. % Co, 15.8 at. % Fe, and 16.5 at. % Ni. To further understand the phase stability of this alloy,
scitation.org/journal/apm the sample was held at 700 ○ C for an additional hour, where coprecipitates with an average size of ∼268 nm were seen in grain interiors.
Based on EDS analysis, these precipitates are classified into Cr-rich and NiAl precipitates. At higher annealing temperature, 900 ○ C-1 h, coprecipitates (similar to the 700 ○ C-2 h condition) with CrFe and NiAl are formed in the grain interior(s). The average size of these precipitates increased significantly in comparison with the 700 ○ C-2 h condition to around ∼1167 nm. The SADPs (supplementary material, Fig. 1) from Cr-rich precipitates show the reflection of ∥111∥ bcc zone axes (and [011] bcc together with ∥122∥ fcc zone axes) from the matrix. In addition, the SADPs from the NiAl precipitates exhibit the reflections of [023] bcc zone axes together with [113∥ (supplementary material, Fig. 1) from the matrix. Interestingly, the Cr concentration of Cr-rich bcc precipitates increased significantly up to 70 at. % at. 700 ○ C and 85 at. % at 900 ○ C in comparison to low Al HEAs. Figures 2(e) and 2(f) show the representative HAADF-STEM images of the Al 0.5 alloy after in situ TEM heating at 700 and 900 ○ C for 1 h. Similar to previous alloys, complex precipitates were formed. For the 700 ○ C-1 h condition, coprecipitates with an average size of 288 nm were seen in the fcc phase. Based on EDS elemental maps, the coprecipitates can be classified into a phase rich in Ni and Al and Cr-rich phase with an average size of ∼250 nm. Increasing the annealing temperature to 900 ○ C [Fig. 2(f)], the average size of coprecipitates increased to around 294 nm. The chemical composition and crystal structure of these precipitates were similar to those observed in the 700 ○ C-1 h condition.
The coprecipitates have been reported previously in a Febased 43 alloy and ultra-high strength steel. 44 Similar to our precipitates, the B2 phase encases the bcc phase. 44 The authors observed enhancement in mechanical properties of the aged alloy since these coprecipitates provide a complex obstacle to dislocation movement. 43,44 Similar observations were seen in the case of the Al 0.5 alloy after in situ TEM heating at 700 and 900 ○ C for 1 and 2 h, respectively, where coprecipitates (Cr-rich bcc and NiAl±B2 phases) formed. However, it is interesting to note that the precipitation of the B2 phase in Al 0.5 required less exposure time at 700 ○ C (only 1 h) in comparison with the Al 0.3 alloy (2 h), where Al is well known to stabilize the B2 phase. 19 These observations broadly agree with previous thermodynamic modeling using CALPHAD. Specifically, CALPHAD predicts that an fcc HEA matrix with NiAl±B2 precipitates first forms above 0.12 mol fraction Al at 700 ○ C and 0.2 mol fraction Al at 900 ○ C; however, it, instead, predicts an additional σ phase at 700 ○ C with greater than 0.18 mol fraction Al and does not predict a Cr-rich bcc phase until the Al content above 0.7 mol fraction at 900 ○ C. 19 The Cr-rich bcc phase, however, was predicted to form in Al 0.3 using Thermocalc software, in conjunction with the Thermocalc TCHEA1thermodynamic databases at the temperature below 700 ○ C, while it was observed experimentally at 900 ○ C using in situ and ex-situ heating techniques. 9 The study highlighted the role of diffusion kinetics in the formation and the sequence of secondary phases, where the phase containing elements with the highest diffusivity (here is Cr based on the previous study by Dabrowa et al. 45 ) will form first. This trend is consistent with the Representative HAADF-STEM/EDS images of (a) Al 0.0 , (b) Al 0.1 , (c) and (d) Al 0.3 , and (e) and (f) Al 0.5 after in situ TEM heating at 750, 900, 700, 900, 700, and 900 ○ C, respectively. (g) T = 0 K reaction energy calculations derived from DFT calculations of phases similar to those in (a)±(f). These calculations show that NiAl±B2 formation is energetically favorable at the Al content above Al 0.3 . The heating parameters of in situ annealing were chosen based on the experimental evolution of maximum heating conditions that samples can withstand before surface diffusion occurs.
current in situ TEM results where the Cr-rich bcc phase formed in all alloys investigated in this study. 9 To build a more robust understanding of the phase stability window of AlxCoCrFeNi HEAs and to examine the role of alloy composition on phase separation, we also performed DFT calculations to explore thermodynamic driving tendencies. In particular, we examined the decomposition of the HEA solid solution to simple products via the following reactions: AlxCoCrFeNi (fcc) → CrFe (bcc) + CoNi 1-x(fcc) + xAlNi (B2) , ( 1)
where the fcc and bcc subscripts denote disordered solid solutions of each respective phase and B2 denotes the ordered CsCl±B2 structure. For computational simplicity, we calculated compositions with mole fractions x ≙ 0, 0.25, 0.5, 0.75, and 1. These particular reactions are motivated by the compositions in Table I, which show experimentally identified Cr-rich bcc and NiAl±B2 phases at high Al concentrations. Additionally, in the case of reaction 2 where a more complex Al(Fe,Ni) B2 precipitate was used, to ensure that this was a reasonable composition choice, the site mixing of Co, Cr, and Fe was considered on both the Al and Ni sublattices, as given in Figs. 2 and 3 of the supplementary material. Figure 2(g) shows the DFT calculated T ≙ 0 K reaction energies as a function of Al composition. The HEA in the left-hand side of reactions 1 and 2 is the reference state for all these energies, so it is plotted as the zero of energy in Fig. 2(g). The DFT calculations show that the HEA solid solution is energetically favored compared to the decomposition products at low Al content but becomes unstable with respect to decomposition as the Al content is increased. This transition occurs at slightly different Alx depending on the reaction, around x ≙ 0.25 for reaction 1 and x ≙ 0.5 for reaction 2. These calculated energies support the experimental observation of Cr-rich bcc and NiAl±B2 precipitates in Al 0.3 and Al 0.5 samples. Figures 3(a)±3(c), in addition to Figs. 2 and3 of the supplementary material, show mixing energy plots for each of the products and demonstrate that it is the formation of Cr-rich BCC and NiAl±B2 phases that drives the HEA decomposition. While both Cr-rich bcc and NiAl±B2 phases have negative mixing energies, the substantially larger negative mixing energy of NiAl±B2 dominates, and as more Al is added, Ni is increasingly pulled out of the fcc HEA matrix, further destabilizing the matrix microstructure. In contrast, the positive mixing energy of CoNi fcc indicates that they stabilize the fcc HEA matrix. Each element's effect on phase (de-)stabilization of the AlxCoCrFeNi HEA agrees with phenomenological observations from CALPHAD. 19 We also compare the energetics of the decomposition products in reactions 1 and 2 with the T ≙ 0 K ground state convex hull (i.e., the lowest-energy linear combination of compounds at each composition). A comparison with the convex hull computed from the Open Quantum Materials Database (OQMD) 46,47 shows the products in reactions 1 and 2 to be higher in energy than the T ≙ 0 K convex hull. This conclusion seems natural since the products in reactions 1 and 2 involve solid solutions only stabilized at finite temperature, and our comparison is for T ≙ 0 K energetics. Hence, we incorporated a simplistic estimate of the ideal configurational entropy in the solid solution phases in order to understand the temperature dependence of these reaction free energies. The solution phases, and particularly the HEA, are stabilized by entropic contributions. Considering only T ≙ 0 K energetics, the HEA is significantly unstable with respect to the OQMD convex hull at all considered compositions. However, at elevated temperature with ideal entropy added, the HEA becomes stable (around 500 K for CoCrFeNi and 1200 K for AlCoCrFeNi; see Fig. 4 of the supplementary material). This is in qualitative agreement with what has been found by the authors previously via experiments. 9 However, we note that the inclusion of ideal configurational entropy does not cause either considered reaction to become lower in energy than the hull at any composition while also remaining favorable with respect to the solid solution HEA (see Fig. 4 of the supplementary material). We do not fully understand the source of this apparent discrepancy at present, but there are several possible causes. The ideal entropy is a crude approximation and is an upper bound of the true configurational entropy, and the true free energy will, in fact, be lower than this ideal model due to the presence of chemical short-range order. Since the HEA benefits comparatively more from the inclusion of configurational entropy, this would serve to stabilize the decomposition products. In addition, the size of other, neglected, contributions to the entropy (e.g., vibrational) is currently unknown. That neither of the considered reactions penetrate the convex hull while favoring the decomposition of the HEA indicates that these contributions could, in fact, be large enough to account for the formation of NiAl±B2 and Cr-rich bcc phases at intermediate and high temperatures. Finally, we should note that the decomposition products in reactions 1 and 2 are simple approximations to the observed decomposition products, as shown in Table I.
The addition of alloying elements in minor concentration (i.e., Al) to the base chemistry of CoCrFeNi HEAs is known to destabilize the microstructure and accelerate the phase separation. 19 In the current study, we find an important impact of Al addition on the chemical composition and density of the secondary phases.
The Cr content (Fig. 4) of Cr-BCC phases increased dramatically by increasing the Al fraction, starting from 31 at. % for the Al 0.0 alloy, reaching around 90 at. % in the case of the Al 0.5 alloy. When the NiAl±B2 phase starts to nucleate in the high Al content HEAs (Al 0.3 and Al 0.5 ), the Cr concentration increases gradually in the matrix and, as a result, boosts the Cr content of the Cr-rich precipitates [Fig. 4(b)]. Additionally, the Cr content significantly increases by increasing the heating temperature and exposure time due to the formation of the NiAl±B2 phase and the high diffusivity of Cr in the AlxCoCrFeNi HEAs. 45,48 Our chemical analysis of Crrich phases is found to be indistinguishable to the ones that observed recently in CoCrFeNiMn after 500 days of exposure at 500 ○ C, where the Cr content was about 86 at. % and showed a bcc structure. 49 Overall, the chemical composition of the B2 phase, however, was not clearly affected by alloy composition and heating conditions, where the concentration of Al and Ni found to vary within the range of 30±40 at. %. The Cr-rich bcc phase was recently reported in AlCuCoCrFeNi, 50 Al 1.5 CoCrFeNi, 51 and CoCrFeNiMn 49 and were labeled as an A2 structure. 15 In AlCuCoCrFeNi, 50 the Cr-bcc phase is mainly enriched with Cr±Co, which is similar to the precipitates that formed in Al 0.1 and Al 0.3 (700-1 h) alloys during in situ TEM annealing experiments. In addition, the Cr-rich bcc phase that was reported in Al 1.5 CoCrFeNi 51 contains Cr and Fe as principal elements, which is consistent with our Cr±Fe precipitates that formed in Al 0.0 , Al 0.3 (at 700 ○ C-2 h and 900 ○ C-1 h), and Al 0.5 (at 700 ○ C-1 h). Additionally, our chemical analysis of Cr-rich bcc precipitates in Al 0.3 and Al 0.5 alloys (the heating condition: 900 ○ C-1 h) nearly matches the precipitates observed in CoCrFeNiMn after 500 days of exposure at 500 ○ C, where the Cr content was about 86 at. %. 49 For the B2 phase, our results are consistent with previous studies where the B2 phase formed at relatively higher annealing temperatures (700±900 ○ C) during in situ and ex-situ TEM heating. Shun and Du 32 reported the formation of the B2 phase in Al 0.3 CoCrFeNi after ex-situ TEM heating at 900 ○ C for 72 h with a chemical composition similar to our B2 phase. The density of secondary phases is also influenced by the Al content, where it increases by increasing the content of Al. The trend is clearly visible when comparing Al 0.3 with Al 0.5 alloys [Fig. 4(c)]. Specifically, the precipitate density of the B2 phase (fractional area) increased from 3.3% for the Al 0.3 alloy to ∼5.7% for the Al 0.5 alloy after in situ TEM heating at 700 ○ C and from 5.1% to ∼7% by increasing the heating temperature to 900 ○ C for Al 0.3 and Al 0.5 , respectively, where the formation of the B2 phase is dominant with increasing the Al addition and heating temperature (up to 900 ○ C). The summary details of the phase evolution regarding the precipitate's density and size over different annealing conditions are listed in Table II.
Thermal stability of AlxCoCrFeNi (x ≙ 0.0±0.5) was studied using an in situ TEM heating technique at the temperature between 700 and 900 ○ C. The alloys (Al 0.0 , Al 0.1 , Al 0.3 , and Al 0.5 ) are identified to be single fcc phases after the homogenization treatment, and multiple secondary phases precipitated after in situ TEM heating at temperatures between 700 and 900 ○ C. Our study showed that the single fcc phase CoCrFeNi alloy is unstable after 10 min exposure at 750 ○ C where a Cr-rich bcc phase formed. We found that the addition of Al has two major impacts on the microstructure. First, the addition of Al destabilized the microstructure and induced the precipitation of an additional secondary phase that was enriched mainly with Ni and Al (B2). Interestingly, the formation of the B2 phase required less exposure time at 700 ○ C when the Al content increased from 6 at. % (Al 0.3 alloy) to ∼11 at. % (Al 0.5 alloy), consistent with Al-additions stabilizing the formation of NiAl±B2. Second, the content of Cr of Cr-rich bcc precipitates increased dramatically by increasing the Al content, up to 90 at. % for the Al 0.5 alloy, where the Cr concentration increased gradually in the matrix when B2 formed and, as a result, boosted the Cr content of Cr-rich precipitates. Additionally, the overall density of the secondary phases increases by increasing the addition of Al, where Al induced the precipitation of the NiAl±B2 phase along with the Cr-rich bcc phase. DFT calculations confirm the base alloy to be energetically unstable at low temperature. Moreover, they provide additional evidence that the Cr-rich bcc and NiAl±B2 secondary phases become energetically favorable as the Al content is increased. Ideal configurational entropy contributions tend to favor the multicomponent HEA over the decomposition products, hence pushing the Al content where decomposition becomes favored to higher values. Overall, while the formation of these secondary phases (in the form of single precipitates and/or coprecipitates) helps maintain mechanical properties, the effect on corrosion resistance is unknown. The knowledge about these phases should motivate conducting more intensive studies to evaluate the structural and corrosion properties of these alloys, predicting an innovative design for maintaining thermal, mechanical, and oxidation stabilities.
See the supplementary material for the SQS structures and cluster correlation information of calculations used in this work that ARTICLE scitation.org/journal/apm can be found at https://github.com/ncs7/Role-of-Al-Additions-in-Secondary-Phase-Formation-in-CoCrFeNi-High-Entropy-Alloys.
© Author(s) 2022