Alloy Phase Research Articles

TEM Studies of a New Modulated Structure in Mn2RuSn Alloy and Intermetallic Phases in Fe3+xCo3-xTi2 (x = 0, 1, 2, 3) Alloys.

TEM Studies of a New Modulated Structure in Mn2RuSn Alloy and Intermetallic Phases in Fe3+xCo3-xTi2 (x = 0, 1, 2, 3) Alloys.

Improving tensile properties and corrosion resistance of Mg–Zn–Ca–Mn alloy via Gd addition

Effects of various Gd additions on the microstructure evolution, tensile properties and corrosion resistance of Mg–6Zn-0.5Ca–1Mn (ZMX6105) alloys were investigated. Results showed that the second phases of as-cast ZMX6105-xGd alloys (x = 0, 0.2, 1.0 wt%) were mainly composed of Mg7Zn3 phase, Ca2Mg6Zn3 phase, Mg3Zn6Gd phase. Gd addition promoted the formation of Ca2Mg6Zn3 phase during solidification, but decreased the fraction of Mg7Zn3 phase. Microstructure observation of as-extruded alloys revealed that Gd addition inhibited dynamic recrystallization, refined dynamic recrystallized (DRXed) grains and strengthened basal texture. As-extruded ZMX6105-0.2Gd alloy exhibited excellent tensile properties with yield strength of 276 MPa, ultimate tensile strength of 354 MPa and elongation of 11%, and simultaneously had optimal corrosion resistance with corrosion rate of 22.06 mm/year based on the weight loss tests in 3.5 wt% NaCl solution.

Hydrogen storage property improvement of ball-milled Mg2.3Y0.1Ni alloy with graphene

In this paper, Mg2.3Y0.1Ni + x wt.% graphene (x = 1, 3, 5, 7) composite hydrogen storage alloys were fabricated by mechanical ball milling of Mg2.3Y0.1Ni alloy powder and graphene. The experimental results show that the phases of the composite alloy consist mainly of Mg2Ni, Mg, MgNi2 and Ni3Y phases, and graphene tends to be distributed at the edges of the alloy, increasing the phase boundary resulting from the contact between the C-boundary and the Mg2.3Y0.1Ni alloy, providing more channels for the diffusion of H2. In which 7 wt.% of graphene is added, the gestation period of the composite alloy completely disappears. And the ball-milled alloy with 3 wt.% graphene at different temperatures has the highest hydrogen uptake in 2 min and the highest hydrogen release in 10 min. At 573 K, the composite alloy has a hydrogen absorption capacity of 3.251 wt.% H2 in 2 min and reaches 90.21% of the maximum hydrogen absorption, while it is able to release 3.389 wt.% H2 in 10 min and reaches 99.04% of the maximum hydrogen release. And the composite alloy with 3 wt.% graphene has the smallest dehydrogenation activation energy. The absolute value of the enthalpy change is reduced from 60.1 kJ/mol to 54.8 kJ/mol, and the desorption enthalpy change is reduced from 73.4 kJ/mol to 67.5 kJ/mol.

Effects of grain boundaries and temperature on spinodal decomposition in a binary Fe-Cr alloy: A phase-field simulation

Due to its excellent oxidation resistance, corrosion resistance and mechanical properties, Fe-Cr based stainless steel is a promising candidate for advanced nuclear power systems. At the temperature of 300–600 °C, spinodal decomposition usually occurs in Fe-Cr alloys. However, irradiation can greatly accelerate this process. Fe-Cr based stainless steel decompose into a Cr-enriched α′ phase and a Fe-enriched α phase during spinodal decomposition, and the Cr-enriched α′ phase plays an important role in the hardening/embrittlement of Fe-Cr based stainless steels. In the present work, the evolution of vacancy and the Cr-enriched α′ phase in Fe-Cr alloy is investigated using phase-field method. The simulation results show that the vacancy concentration has a great influence during the spinodal decomposition. The Cr-enriched α′ phase occurs initially at the positions with high vacancy concentration. During spinodal decomposition, the vacancy forms a defect concentration loop around the Cr-enriched α′ phase. In addition, with the increase of temperature, the concentration of Cr atoms in the center of the Cr-enriched α′ phase decreases, and the coarsening rate increases, while the number of the Cr-enriched α′ phase formed in the alloy decreases. At low temperature, α′ phase is mainly coarsened by interlinking; while at high temperature, α′ phase is mainly coarsened by Oswald maturation. Moreover, the Cr-enriched α′ phase occurs at the grain boundary (GB) and the Fe-enriched α phase occurs at both sides of the GB during the spinodal decomposition. At the same time, in polycrystalline system, GB movement during grain growth is inhibited during the spinodal decomposition.

Prediction of phases in high entropy alloys using machine learning

Predicting the phases of high entropy alloys (HEAs) is a complex task as it depends on a large number of thermal, physical and geometrical parameters. The phases consist of intermetallic (IM) and amorphous (AM) forms. As, the trial-and-error approach is exhaustive and time-consuming process for both experiments and numerical simulations. Standard statistical models such as XGboost, random forest, adaboost, decision tree, logistic regression, support vector machine (SVM) and k-nearest neighbors (KNN) were adopted. The XGboost algorithm yields the better accuracy (90 %). Machine learning methods well predicted the phases and also identified the main significant variables such as mean atomic radius, atomic size difference, mixing enthalpy, ideal mixing entropy and valence electron concentration, which are in consistence with experimental results. As a whole, it is recommended that the widespread use of machine learning approaches could accelerate HEA process with tailored configurations. The use of multiple models in ensemble learning reduces the risk of overfitting, increases the accuracy of predictions, and enhances the generalizability of models.

Processing and microstructure–property relations of Al-Mg-Si-Fe crossover alloys

This study introduces new alloys, which combine the age-hardening capability of Al-Mg-Si alloys with the microstructure-controlling effect on processing of primary Fe-rich intermetallic phases used in foil stock. In detail, the processing and microstructure–property relations in new crossover aluminum alloys derived from 6xxx and 8xxx foil stock alloys, is shown. A highly Fe-rich intermetallic phase content was deployed to conceptually mimic high scrap content. Fast and slow solidification rates were applied to represent thin strip and direct chill casting, respectively. The effects of adding Fe and Mn to alloy 6016 were examined, while the Si consumed in primary phases was partly adjusted to maintain age-hardening potential. It was shown that upon thermomechanical processing, primary intermetallic phases in the new alloys are finely fragmented and well dispersed, resulting in strong grain refinement and a uniform texture. Attractive combinations of strength and ductility were revealed, also in material processed under direct chill casting conditions. The new alloys’ high elongation values of up to 30%, and their age-hardening response, were similar to those seen in commercial alloy 6016, while their strain hardening capacity was significantly greater. This can be attributed mainly to the formation of geometrically necessary dislocations near primary Fe-rich intermetallic phases. The study discusses microstructure refinement on the basis of particle stimulated nucleation. It uses a simple model to describe the individual contributions to yield strength, including the effect of primary phases. It also models the effect of these particles on increased strain hardening and ductility.

Magnetocaloric effect and phase transformation of LaFe11.0M0.5Si1.5 (M = Cu, Ni, Mg) alloys

La (Fe,Si)13 is a promising magnetic refrigerant material. In this paper, the effects of Cu, Ni and Mg substitution of Fe on the magnetocaloric effect and phase transition properties of LaFe11.5Si1.5 alloys were studied by first-principles calculations combined with experiments. Through scanning electron microscopy–energy-dispersive spectroscopy analysis, it was concluded that after Cu partially replaced Fe, the content of the 1:13 main phase significantly decreased to 61.49%, and the contents of the α-Fe and LaFeSi phases increased. Cu was not conducive to the formation of the main phase. After Mg partially replaced Fe, the content of the 1:13 main phase increased to 99.12%, and Mg promoted the formation of the main phase. The electronic structure of the alloy was reconstructed by density functional theory (DFT), and the effects of the doping atoms on the magnetic properties of the alloys were studied. The calculation results showed that Ni atoms had the same weak magnetism as that in the magnetic moment vector direction of Fe atoms, Cu and Mg atoms had magnetism opposite to that in the magnetic moment vector direction of Fe atoms, and all alloys showed a lower total magnetic moment. According to the Arrott curve and the field dependence method of the magnetocaloric effect (MCE), when Cu = 0.5 and Ni = 0.5, the alloy phase transition changed from a first-order phase transition to a second-order phase transition, and when Mg = 0.5, the alloy still had a first-order phase transition.

Effect of ultrasonic surface shot peening on the microstructure and properties of Cu-8Cr alloy

Surface nanocrystalline layers were prepared on Cu-8Cr alloy using ultrasonic shot peening (USSP) technique by controlling the pellet size and shot peening time. The tissue morphology, microstructure, surface hardness, surface strength and corrosion resistance of the generated surface nanolayers were systematically characterized using x-ray diffractometer (XRD), scanning electron microscope (SEM), and the corresponding test methods. The results showed that the USSP treatment did not change the surface phase of Cu-8Cr alloy, but only refined the surface grains into nanocrystals. The pellet size and shot peening time had a large effect on the surface properties of Cu-8Cr alloy. The best overall surface properties were obtained with a shot size of 2 mm and a shot time of 30 min The average grain size of the nano-layer is about 66 nm, the surface hardness is increased to 126.69 HV0.05, the tensile strength reached 417 MPa, and the corrosion resistance of the alloy is significantly enhanced. The USSP treatment provides a new method for the preparation of CuCr alloys with excellent surface properties.

Kinetic Analysis of Boron and Phosphorus Removal from Si-Fe Alloy by CaO-Al2O3-SiO2-Na2O Slag

A hybrid process of slag and solvent refining was used to remove boron and phosphorus from silicon. Quaternary slag of CaO-Al2O3-SiO2-Na2O was employed to remove boron (B) and phosphorus (P) from Si-20 wt% Fe alloy at 1300 °C. A slag-to-metal ratio of one was used at different reaction times. The mass transfer coefficient of B and P in the slag and alloy phases was calculated to determine the rate-limiting step. The mass transfer coefficients of B in the alloy and slag phases were 6.6 × 10−7 ms−1 and 2.8 × 10−7 ms−1, respectively. The mass transfer coefficients of P in the alloy and slag phases were determined to be 7.5 × 10−8 ms−1 and 3.5 × 10−7 ms−1, respectively. The rate-limiting stage of the slag–alloy reaction kinetics was mass transport in the liquid slag for B and mass transport in the alloy phase for P.

Micro-encapsulation of a low-melting-point alloy phase change material and its application in electronic thermal management

The application of liquid metal (LM) with high thermal conductivity in electronic thermal management is an emerging hotspot, but leakage of LM can cause environmental pollution and lead to corrosion of electronic equipment, thereby affecting the safety of its operation. Moreover, large temperature fluctuation will be caused for the limited thermal storage capacity of LM. To solve the above problems, it is urgent to develop LM-based microcapsules with high thermal conductivity and high thermal storage capacity. Herein, a novel micro-encapsulated phase change material (MEPCM) with an internal void, composed of liquid metal Sn–Bi–In core and PMMA shell, was firstly fabricated via liquid phase micro-encapsulation method. The results showed that microcapsules maintained good morphology and efficient thermal storage capacity (246.46 J cm−3) after 300 thermal cycles, 97.12% of that before the cycle (253.77 J cm−3). Subsequently, the MEPCM-based composite was prepared by mixing MEPCM into thermal gel. The enthalpy of 50 wt% MEPCM composite was 126.72 J cm−3, and the thermal conductivity of it was 3.23 W m−1·K−1, while that of thermal gel was 0.98 W m−1·K−1. Then the 50 wt% MEPCM-based composite was applied for LED chip, and the temperature of chip was 71.1 °C, 11.2 °C lower than that of thermal gel. Therefore, MEPCM-based composites with high thermal conductivity and thermal storage capacity are considerably potential for electronic thermal management.

Pgm: A Python package for free energy calculations within the phonon gas model

The quasi-harmonic approximation (QHA) is a powerful method that uses the volume dependence of non-interacting phonons to compute the free energy of materials at high pressures (P) and temperatures (T). However, anharmonicity, electronic excitations in metals, or both, introduce an intrinsic T-dependence on phonon frequencies, rendering the QHA inadequate. Here we present a Python code, pgm, to compute the free energy and thermodynamic property within the phonon gas model (PGM) that uses T-dependent phonon quasiparticle frequencies. In this case, the vibrational contribution to the Helmholtz free energy is obtained by integrating the vibrational entropy, which can be readily calculated for a system of phonon quasiparticles. Other thermodynamic properties are then obtained from standard thermodynamic relations. We demonstrate the successful applications of pgm to two cases of geophysical significance: cubic CaSiO3-perovskite (cCaPv), a strongly anharmonic insulator and the third most abundant phase of the Earth's lower mantle, and NiAs-type (B8) FeO, a partially covalent-metallic system. This is the oxide endmember of a recently discovered iron-rich FenO alloy phase likely to exist in the Earth's inner core. Program summaryProgram Title:pgmCPC Library link to program files:https://doi.org/10.17632/8rfw6syvzp.1Developer's repository link:https://github.com/MineralsCloud/pgmLicensing provisions: GNU General Public License 3Programming language: Python3Nature of problem: The classic quasi-harmonic approximation (QHA) method to compute the vibrational free energy does not apply to physical systems when phonon frequencies have an intrinsic and non-negligible temperature (T) dependence. Examples are anharmonic systems or metals with abundant electronic thermal excitations. Both cases introduce an intrinsic T-dependence on phonon frequencies.Solution method: The method implemented in pgm is based on the phonon gas model where the entropy is well defined for T-dependent phonon quasiparticle. The free energy is calculated by integrating the entropy, making it suitable for anharmonic systems or systems with extensive thermal electronic excitations affecting phonon frequencies. The static free energy, the vibrational density of states (VDoS), and the entropy are first computed on sparse T- and P-grids. The entropy is then suitably interpolated on denser user-specified grids for integration.Additional comments, including restrictions and unusual features: The package allows it to be run directly in the command line. It can also be incorporated into other programs. We implemented Just-in-time (JIT) compiling and parallel computing techniques [1] in pgm Python code to speed up the numerical calculation.

Investigation of the Pd(1−x)Zn x alloy phase diagram using ab initio modelling approaches

The identification of the stable phases in alloy materials is challenging because composition affects the structural stability of different intermediate phases. Computational simulation, via multiscale modelling approaches, can significantly accelerate the exploration of phase space and help to identify stable phases. Here, we apply such new approaches to understand the complex phase diagram of binary alloys of PdZn, with the relative stability of structural polymorphs considered through application of density functional theory coupled with cluster expansion (CE). The experimental phase diagram has several competing crystal structures, and we focus on three different closed-packed phases that are commonly observed for PdZn, namely the face-centred cubic (FCC), body-centred tetragonal (BCT) and hexagonal close packed (HCP), to identify their respective stability ranges. Our multiscale approach confirms a narrow range of stability for the BCT mixed alloy, within the Zn concentration range from 43.75% to 50%, which aligns with experimental observations. We subsequently use CE to show that the phases are competitive across all concentrations, but with the FCC alloy phase favoured for Zn concentrations below 43.75%, and that the HCP structure favoured for Zn-rich concentrations. Our methodology and results provide a platform for future investigations of PdZn and other close-packed alloy systems with multiscale modelling techniques.

Flexible core–shell structured Al-Cu alloy phase change materials for heat management

The development of core–shell structured phase change materials (PCMs) with tunable melting points has been attracted much attention for thermal storage of solar energy, and a flexible shell is crucial for their practical applications. Herein, we prepared microencapsulated Al-Cu alloy PCMs (MEPCMs) with a flexible shell by a deposition-oxidizing calcination method. The Cu nanoparticles are chemically deposited on the surface of Al powders by a rapid replacement reaction, which further promotes the formation of Al-Cu alloy microspheres and the oxidation of Al to form Al2O3 shell during oxidation calcination. The prepared MEPCMs show excellent thermal cyclic performance and tunable melting point (549–592 °C). The presence of a thin film between the Al2O3 shell and Al-Cu alloy core may act as a buffer to accommodate volume change during melting-freezing process, obtaining a flexible core–shell structure. The heat storage density of Al-Cu alloy MEPCMs decreases by only 0.09% after 100 melting-freezing cycles, indicating high thermal cycling stability. In addition, the microencapsulated Al-Cu alloys present high latent heat (208.6–222.3 J·g−1) and relatively high thermal conductivity (∼1.8 W·(m·K)−1), which makes it attractive for application in high-temperature thermal storage systems.

Co, In, and Co–In alloyed Cu6Sn5 interconnects: Microstructural and mechanical characteristics

The mechanical reliability of the future miniaturized interconnects is mainly governed by the intermetallic compounds such as Cu6Sn5. Alloyed Cu6Sn5 with various elements, including Co and In, have been introduced and attracted attention for different reasons, such as the enhancing mechanical reliability and lowering the bonding temperature. Hence, this work aimed to evaluate the microstructural and mechanical properties of Cu6Sn5-, Cu6(Sn,In)5-, (Cu,Co)6Sn5-, and (Cu,Co)6(Sn,In)5-interconnects. The grain size, grain orientation, and crystal structure of the pure and alloyed Cu6Sn5 phases were analyzed using electron backscatter diffraction. The results revealed that all the joints contained monoclinic and hexagonal crystal structures arbitrarily formed across the bond-line. Furthermore, the Cu6Sn5 grains exhibited random grain orientation and there was no discernible difference between the pure and alloyed Cu6Sn5 interconnects other than Cu6(Sn,In)5 grains elongated along the perpendicular direction to the bonding interface. However, it was found that alloying elements altered the grain sizes. In alloying refined and elongated the Cu6Sn5 grains while the Co alloying enlarged the Cu6Sn5 grains. The mechanical properties of the interconnects were examined using nanoindentation test. The results indicated that the hardness (H) and Young's modulus (Ei) values of Cu6Sn5 is increased with the alloying elements. (Cu,Co)6(Sn,In)5 showed the highest Ei/H value which indicates its highest plasticity.

Synergistic effect for improving the hydrogen storage capability and electrochemical performance of AB2 Laves phase alloys

The structure of a series AB2 Laves-type metal hydride alloys, in which Ce was doped into Ti0.24Zr0.76Ni1.15Mn0.7V0.15 and annealed the alloys, was studied and correlated to hydrogen storage properties and electrochemical performances. X-ray diffraction, scanning electron microscope, and energy-dispersive spectroscopy results revealed that Ce formed a CeNi3 secondary phase instead of entering the Laves phases. The synergistic effect of the CeNi3 and (Ti,Zr)Ni phase improved the hydrogen storage and electrochemical discharge properties of the Ce-doped (Ce0.02) alloy. Especially, the secondary phases could decrease the hydrogen absorption plateau, reduce the hysteresis phenomenon, and increase the hydrogen storage capacity (from 0.9 wt% to 1.4 wt% at 25 ℃). The formation of CeNi3Hx hydride on the surfaces of alloy electrodes led to a higher discharge capacity (∼500 mAh/g) and a faster activation rate. The Ce0.02 alloy also showed a good cyclic stability ∼82.08%, after 500 charge-discharge cycling, which could be ascribed to the good corrosion resistance of the CeNi3 phase on the surface of alloy. In addition, annealing treatment promoted the depletion of (Ti,Zr)Ni secondary phase and the transformation from the C14 to C15 phase. The Ce-doped AB2 alloy can be used as a hydrogen storage and hydride battery electrode material.

Preparation of Al-Ni Alloys of Different Phases Through Direct Electro-Deoxidation of Al2O3-NiO in CaCl2-NaCl Molten Salt

Al-Ni alloys have several common thermodynamically stable alloy phases. These alloys showed distinct physical properties and were applied in different fields, including aerospace and electrochemical catalysis. However, it is difficult to prepare Al-Ni alloys with the certain phase composition through traditional preparation methods, such as casting and mechanical alloying. Herein, Al-Ni alloys were prepared through electro-deoxidation in CaCl2-NaCl molten salt. Pure Al-Ni alloys phases (AlNi3, AlNi and Al3Ni2) were obtained after optimizing the electro-deoxidation conditions, such as the molar ratio of Al and Ni in the original tablet, the sintering temperature of the precursor, electro-deoxidation time and the molten salt temperature. The deoxidation processes of the Al2O3-NiO composite were investigated through various electrochemical methods, such as cyclic voltammetry, chronoamperometric curve and constant potential electro-deoxidation experiment. The loss of Al2O3 in the molten salt was confirmed to be responsible for the loss of Al in the electro-deoxidation processes. Meanwhile, it was proposed that the loss of Al could be efficiently delayed by increasing the precursor’s sintering temperature and the molten salt temperature. The obtained Al-Ni alloys (AlNi3, AlNi and Al3Ni2) could be used as efficient additives in the cathode of lithium–sulfur batteries to improve the charging/discharging capacity.

Evolution behavior of δ phase in Cu12Sn2Ni alloy and the corresponding effects on alloy property

AbstractWith the increase of tin content in tin bronze, the rise of δ phase made the strength, hardness of tin bronze increase and the ductility decrease sharply, that difficult to process. In this paper, the Cu12Sn2Ni alloy was prepared by centrifugal casting, the microstructure and phase formation before and after heat treatment were observed by x‐ray diffraction, scanning electron microscope, and transmission electron microscope. The results showed that the as‐cast sample microstructure was composed of equiaxed grains rather than coarse dendrites. centrifugal casting inhibits tin diffusion to form metastable phase β′‐Cu13.7Sn. The as‐cast sample had good deformability and its tensile strength and elongation were 381.9 MPa and 12.4 %, respectively, which are higher than the mechanical properties of gravity casting. The tensile strength and elongation of the sample after furnace cooling at 620 °C/8 min are 439.5 MPa and 24.4 %, respectively, the increase was 16.6 % and 85.07 %, compared with the as‐cast samples, due to the solid solution strengthening, the second phase strengthening and the homogenization of the microstructure.

Study on the influences of different process temperatures on microstructural evolution of Mg-11%Al-2%Zn-1%Si alloy through isothermal semi-solid treatment

Due to the eutectic reaction, the Mg2Si phase is inclined to creating an unfavorable fishbone shape at low solidification rates. This can significantly degrade the mechanical characteristics and become the main barrier to the practical application of Mg-Si based alloy. It is regarded that isothermal semi-solid treatment is an efficient technique to modulate the fishbone-shape eutectic Mg2Si phases in Mg-Si based alloys. In this work, a new type of Mg-11%Al-2%Zn-1%Si alloy is successfully fabricated through isothermal semi-solid treatment, and the influences of different process temperatures on microstructural evolution of Mg-11%Al-2%Zn-1%Si alloy are studied. The findings illustrate that the mean size of the α-Mg grains increases and their trendency of spheroidization becomes more evidently after held for 30 min at different process temperatures from 550 °C to 580 °C. Additionally, through isothermal semi-solid treatment, the eutectic Mg2Si phases in experimental alloy are totally changed from their original fishbone shape into polygon or granule shapes. With increasing the process temperatures, the mean size and morphologies of eutectic Mg2Si phases have no significantly changes. The morphological change theory of eutectic Mg2Si phases through isothermal semi-solid treatment is also studied.

Achieving ultrahigh strength in pre-ageing-extruded Mg-Gd-Y-Zn-Mn alloys via ageing treatment

Ultrahigh-strength Mg-9.2Gd-4.4Y-1.0Zn-0.8Mn (wt%) alloys were produced through isothermal ageing at 200 °C applied to pre-ageing-extruded bars. The remarkable thermal stability exhibited by the β and LPSO phases ensures that the recrystallization state of α-Mg grains and the morphology and distribution of secondary phases in peak-aged alloys resemble those observed in the as-extruded alloys at the micron scale. The microstructure of peak-aged alloys is characterized by bimodal α-Mg grains with a pronounced fibrous texture, with the major strengthening phases consisting of prismatic β′ and basal γ′ phases precipitated during ageing. Through solid-solution and pre-ageing treatments, the extruded alloy at peak age demonstrated the highest tensile strength, with an ultimate tensile strength (UTS) of 555 MPa, a tensile yield strength (TYS) of 488 MPa, and an elongation of 5.8%. The considerable enhancement in tensile strength with satisfactory ductility in peak-aged alloys can be attributed to dense number distribution co-precipitation, which increases both basal and non-basal critical resolved shear stresses (CRSS) and inhibits microcrack nucleation and propagation. These findings have significant implications for the advancement of high-strength Mg-Gd-Y-Zn alloys with improved mechanical properties and performance in practical applications.

An end-to-end machine learning framework exploring phase formation for high entropy alloys

Exploring the rules of high entropy alloys (HEAs) phase formation has clear guiding significance for the design of new alloys. An end-to-end framework was proposed to select the feature subset and machine learning (ML) model from the feature pool and model pool, respectively. In this framework, each model in the pool is to determine its materials feature subset based on the feature importance. The final model was confirmed through the evaluation of the fitting result of every model and its feature subset. This method extracts important factors affecting the phase formation of HEAs. The results show that the chosen model could classify 430 HEAs into five phases, with test accuracy of 87.8%. And the model analysis suggests that the formation of single-phase solid solution is often inhibited when the atomic size difference is greater than 8.295%.