Effect of Al Addition on the Microstructural Evolution of Equiatomic CoCrFeMnNi Alloy

The present investigation reports the effect of aluminum addition on the microstructural evolution in the equiatomic CoCrFeMnNi high-entropy alloy. Aluminum was added to the alloy in varying quantity (0 ≤ Al ≤ 10 at.%) using the vacuum arc melting technique, and phase formation was probed using X-ray diffraction, scanning electron microscopy and transmission electron microscopy. The results indicate that the FCC phase in the alloy remains unaltered up to Al of 5 at.%. The higher amount of Al addition leads to the precipitation of B2 Al(Ni, Cr, Fe, Co, Mn) in the FCC matrix. For Al ≥ 7%, typical phase-separated microstructure consisting of FCC, and B2 phases have been observed. The microstructural changes lead to hardness variation from 1.3 to 2.2 GPa, mainly due to precipitation and solute solution hardening of FCC phase. For FCC phase, Al atoms being larger in size can lead to lattice distortion and improve yield strength. The results have been explained by detailed thermodynamic analysis using HEA3 database.


Introduction
In last 15 years, research and development of the novel multicomponent multi-principle alloys have matured from the perspective alloy design with emphasis on engineering novel microstructure to obtain unique properties 1 . These alloys, popularly known HEAs (high entropy alloys) exhibit an excellent combination of properties, providing impetus to novel alloy development with single or multiphase microstructure. Although these alloys contain 5 or more components, they show FCC; BCC or combination of FCC/BCC solid solution phase in the mixture, devoid of brittle intermetallic compounds (IMCs) [1][2][3][4]. One such important alloy system in equiatomic CoCrFeMnNi, popularly known as Cantor alloy, named after the discoverer of the alloy [5,6]. It shows FCC solid solution (a= 0.33605 nm), exhibiting reasonable strength (yield strength = 160 MPa), high ductility (50 -60%) and toughness (>217 MPa.m 1/2 ) [7,8]. However, the application of this alloy is limited by lack of sufficient high yield and ultimate tensile strength, even though it has high plasticity and toughness. This has been ascribed primarily to single phase microstructure and insufficient lattice distortion [9].
One of the ways to improve strength involves the addition of certain elements in the Cantor alloy to create microstructure providing strengthening of the existing FCC phase. Extensive literature survey reveals that Al addition to equiatomic CoCrFeMnNi alloy can definitely affect the microstructural evaluation, providing avenues to design novel microstructures.
Previous studies Al addition on various alloys TiCoCrFeMn [10], CoCrFeNbNi [11] as well as CoCrFeMnNi [5] categorically show that Al addition affects the phase formation and microstructural evaluation in these alloys. After being introduced for the first time by Cantor et al. [5], the equiatomic CoCrFeMnNi HEA which is a single phase FCC solid solution [1] has been studied extensively owing to its microstructure [12], thermodynamic stability [13], sluggish diffusion [14], ductility, malleability, low-temperature mechanics [15][16][17], etc. The logic behind the use of Al as an alloying addition is the dualism of Al as metal as well as a non-metal which has also led the researchers to use it for the same [18,19]. The cause of this duality is attributed to the special electronic structure of Al. Even though Al has a FCC structure, its effects on the microstructure of HEAs is contrasting. It is reported to promote the formation of hard BCC crystal structure [20,21]. In addition, Al addition is required from the perspective of understanding intrinsic lattice distortion in the HEAs. Lattice distortion is considered to be a unique feature in HEAs, controlling deformation micro-mechanism and solid solution strengthening. The lattice distortion, controlling the solid solution strengthening is dictated by relative size difference of atomic species. The atomic size difference is equiatomic CoCrFeMnNi alloy is small, and the effects will not be significant to be proved by experiments. Hence, Al, which has substantially different atomic size, can eventually be incorporated in the Cantor alloy. In order to create sufficiently large lattice distortion, the FCC solid solution must without any solute partitioning. Therefore, careful and optimum level of Al -addition must be maintained to achieve it. Therefore, it is vital to study the effect of Al addition to FCC CoCrFeMnNi alloy to understand the above-mentioned aspects. In the present investigation, Al x CrCoFeMnNi alloy with an equiatomic composition of Cr, Co, Fe, Mn, Ni and different concentrations of Al are fabricated via vacuum arc melting under high purity Ar atmosphere on a water cooled copper hearth with-consumable tungsten electrode. The different amount of Al (0.5 to 10 atom %) has been added to equiatomic CoCrFeMnNi alloy via the vacuum arc melting route. The microstructural evaoluation has been probed using XRD, SEM, and TEM. The experimental results have been verified with predictions from CALPHAD to determine the ability and efficiency of thermodynamic modeling for predicting phase evolution in HEA

Samples preparation and characterization
Alloy buttons with a composition of (CrCoFeMnNi) x Al 100-x (x = 0, 0.5, 2, 3, 4, 5, 6, 7, 8, 10 atom %) were synthesized in a vacuum arc melting furnace under high purity Ar atmosphere on a water cooled copper hearth with non-consumable tungsten electrode. The alloy buttons were flipped and re-melted at least five times to ensure chemical homogeneity. Subsequently, the alloys buttons were homogenized sealed in a quartz tube under an argon gas atmosphere and homogenized at 1273 K for 24 hours, followed by quenching in water. The phase formation and microstructural evolution has been studied using X-ray diffraction ((Cu K radiation,  = 0.154056 nm) using Empyrean Panalytical diffractometer), Scanning electron microscope (SEM, Carl Zeiss, EVO50) equipped with energy dispersive spectroscopy (EDS, Oxford Instruments) and transmission electron microscope (TEM, FEI Tecnai UT20 with accelerating voltage of 200 kV) . The elemental concentration and distribution were determined using EDS analysis. TEM sample was prepared by cutting the 3 mm diameter disk from bulk sample followed by dimpling to reduce the thickness in the center of the disk (40 m) and electro-jet polishing using a solution containing 90 vol.% methanol and 10 vol.% perchloric acid till perforation in the sample had been made. Hardness measurements were performed on homogenized samples using Tinius-Olsen FH-10 Universal Hardness Testing Machine with a Vickers's Indenter with a load of 5kgf.

CALPHAD
Computational software and databases with recorded thermodynamic values are necessary to use the CALPHAD method to obtain a phase/property diagram which enables us to predict the thermodynamic properties of regions which haven't been experimentally performed.
Extrapolation of the binary and ternary systems allows data calculation for higher-order systems. Using the HEA database (TC HEA1 v3.0) the property diagrams of each of the alloy's composition were obtained. Subsequently, the pseudo binary phase diagram was also constructed using the same database. The property diagrams for alloys with Al concentrations (0, 0.5, 4, 6 and 8 at. %) were constructed to obtain the number of phases present for different samples at a specific temperature.

X-Ray Diffraction
The X-ray diffraction pattern of homogenized samples at 1000 o C/24 h with different concentration of Aluminium addition (Al atomic percent = 0, 0.5, 2, 3, 4, 5, 6, 7, 8, 10) have been recorded as shown in Figure 1(a). The lattice parameters of the phases have been calculated using precise lattice parameter calculation (Nelson-Riley function) [22] as shown in Figure 1(b). The lattice parameter was significantly increased on the addition of progressively Al in Cantor alloy and after 7 at. % it starts to decreases as well as the BCC phase formation observed (shown in Figure 1 Al clearly shows the presence of an additional B2 phase. A weak superlattice 100 peak is also observed in the sample containing 8 at.% Al indicating the presence of B2 phase. It is to be mentioned here that ordering in HEA phases has been reported earlier in the literature [10,[23][24][25][26][27]. It is thus clear that with progressive addition of aluminium up to six atomic percent (Figure 1 a), eqiatomoc CoCrFeMnNi HEA remains single phase albeit undergoing lattice expansion and therefore, increase in lattice distortion. The alloy with 8 at.% Al shows a decrease in the lattice constant for the FCC phase and the presence of a B2 phase. There is an increase in the lattice parameter of both the phases for the sample containing 10 at.% Al. This can be obtained on the basis of the higher atomic size of aluminum as well as its ambiguous metalloid character that contributes to higher distortion, thereby increasing the entropy of the more open BCC phase compared to close-packed FCC high entropy phase.

Microstructure and morphology using SEM
The segregation of Al element in the cantor alloys also have been observed using SEM-EDS mapping and shown in Figure 2(b). Al has been observed to be homogeneously distributed in the alloy for alloys containing Al up to 4 at.% (Fig. 2b(ii)). For HEAs with Al concentration of 5 at.% or higher, distinct segregation of Al has been observed in the homogenized samples ( Fig.2b(iii)). For 7 at.% Al, these Al-segregated zones are distinct and large and hence, can be considered as a separate phase. For alloys with 8 or 10 at.% Al, a feature of phase separated zones containing (Al, Ni,Fe, Cr) and (Mn, Ni) are evident. Hence, Al maps, obtained by EDS measurements clearly suggests the stability and new phase formation in the Al -containing HEAs.

Transmission Electron Microscopic (TEM) analysis
The finer scale microstructural characterization of anll the Al-containing samples have been carried out to understand the formation of different phases, size, and their morphologies.
However, we shall discuss here the salient results. The Figures 3(a-d) shows the bright field image of the equiatomic CoCrFeMnNi alloy as well as HEAs with different Al concentration.  Fig. 3b) and 5 at.% Al ( Fig. 3c). In fact, single phase FCC grains have been observed for HEAs containing 6 at.% Al. Figure 3d shows a bright field image of 8 at.% Al, revealing the second phase as circular precipitation and corresponding ring diffraction pattern confirmed the BCC second phase formation. Therefore, it is evident that BCC/B2 precipitates form for alloys containing Al up to 7 at.% or higher.
In order to corroborate the experimental findings, thermodynamic modeling was carried out using CALculation of PHAse Diagram approach using CALPHAD software with TCHEA 3 database. The results of these calculations have been provided in terms of property diagram as well as a pseudobinary phase diagram. Figures 4(a-e)

Hardness Measurements
In order to probe the effect of Al addition on the mechanical property, the hardness of the homogenized samples has been measured and reported in Figure 5. It shows that hardness monotonically increases up to 6 at.% Al and this is followed by a rapid increase of increase up to 10 at.% Al. The first increase may be assigned to solutelid solution hardening where as the later part (as marked on the Fig.5) is due to the formation of B2 precipitates in the FCC matrix. Hence, Al-addition causes hardness change due to two distinct mechanisms. In the former, the solid solution hardening and associated lattice distortion in the multielemental matrix because of distinctly smaller atomic diameter as compared to other elements ( Co, Cr, Fe, Mn, and Ni). It has been shown earlier that HEAs are characterized by higher lattice distortion compared to conventional metals and alloys. This leads to pinning of dislocations by a local solute environment that contributes to higher lattice friction stress. It is expected that a significantly larger aluminum atom in the CoCrFeMnNi solid solution will cause significant lattice distortion and contribute to higher friction stress thereby contributing to an increase in hardness. It is to be mentioned here that other contributions to hardness or

Summary and conclusions
The present study has shown that Al addition to equiatomic CoCrFEMnNi HEA affects the phase formation and microstructural evolution significantly and according to the mechanical properties can be tuned by adding a different amount of Al to the alloys. The following conclusions can be drawn from the study: 1. FCC HEA phase in the Cantor is stable up to 6 at.% of Al addition. Al can dissolve in the FCC phase and expand the lattice causing lattice distortion.
2. For the higher amount of Al (6 -10 at.%), (Al,Ni,Fe)-rich BCC/B2 phase precipitates in the FCC matrix. The precipitates show near spherical shape with nanometric size.
3. Thermodynamical calculations using HEA3 database reveal that FCC HEA phase is stable at a higher temperature (800-1250 0 C) in the Cantor alloy as well as with low