Microstructure and Mechanical Properties of Y-Doped AlCoCrFeNi_2.1 Eutectic High-Entropy Alloy Fabricated by PBF-LB/M

Abstract

A Y-doped AlCoCrFeNi_2.1 eutectic high-entropy alloy was fabricated via powder bed fusion-laser melting/metal (PBF-LB/M), and the effects of the rare-earth element Y on its microstructure and mechanical properties were investigated. The results indicate that Y addition preserves the fine eutectic microstructure inherent to the PBF-LB/M process, while inducing lattice distortion within the face-centered cubic (FCC) matrix and promoting grain refinement. During solidification, Y facilitates heterogeneous nucleation and, due to its strong affinity with Al, increases both the volume fraction of the body-centered cubic (BCC) phase and the proportion of high-angle grain boundaries. X-ray diffraction (XRD) analysis further confirms that Y suppresses the formation of the ordered B2 phase. Tensile testing reveals that Y doping improves the tensile strength from 1383 MPa to 1475 MPa and enhances the elongation from 13.0% to 16.3%. Fractography shows a transition from quasi-cleavage to ductile fracture mode, indicating that Y significantly enhances the strength–ductility synergy of the alloy.

1. Introduction

AlCoCrFeNi_2.1, as the first reported eutectic high-entropy alloy (EHEA), has demonstrated an excellent combination of mechanical properties, featuring high tensile strength (>1 GPa) and considerable ductility (>10%) [1]. Its superior fluidity during solidification enables the formation of fine and well-aligned lamellar microstructures in as-cast samples [2]. Unlike conventional processing methods, PBF-LB/M introduces extremely high cooling rates (typically 10³–10⁶ K/s), which significantly suppress elemental diffusion, refine eutectic lamellar spacing, and allow the fabrication of complex geometries with near-net-shape precision [3]. Guo et al. [4] fabricated AlCoCrFeNi_2.1 samples with a relative density of up to 99.73% using the PBF-LB/M method. Compared with the as-cast counterpart, the yield strength was increased by approximately 20%, and the elongation reached 22.5%. Furthermore, they investigated the influence of the laser processing parameter window on the microstructural morphology and found that as the volumetric energy density (VED) decreased, the fraction of cellular eutectic structures increased. Similar results have also been reported in other PBF-LB/M studies on this alloy [5,6]. However, PBF-LB/M-fabricated EHEAs still face challenges such as defect formation, residual stress accumulation, and insufficient high-temperature stability, which critically restrict their application in advanced engineering scenarios [7].

Rare-earth (RE) elements have long been recognized as efficient microalloying additives in conventional alloys, where they enhance overall properties through mechanisms including interfacial segregation, grain refinement, and secondary phase regulation [8]. Compared with traditional modification approaches involving thermal and/or pressure treatments, RE doping provides higher efficiency and greater potential for performance enhancement [9]. This is particularly relevant for PBF-LB/M-fabricated alloys, where the as-built ultrafine microstructure is initially beneficial but is also highly prone to grain coarsening during subsequent heat or pressure treatments, which weakens the fine-grain strengthening effect [10]. In recent years, RE doping has been introduced into high-entropy alloy (HEA) systems. Owing to their unique 4f electron configuration and high electronegativity, RE elements tend to react with impurity elements such as oxygen and sulfur, purifying grain boundaries and improving alloy densification [11,12].

Yttrium and Hafnium are among the most widely used RE dopants in HEAs, owing to their relatively high melting points, strong affinity with oxygen/sulfur for grain-boundary purification, and large atomic radii that promote lattice distortion strengthening. Lu et al. [13] reported that Yttrium doping in AlCoCrFeNi_2.1 significantly enhanced its high-temperature oxidation resistance. Other studies have also confirmed that RE addition can improve the mechanical performance of HEAs [14,15]. For instance, Zhou et al. [16] demonstrated that adding Y to CrFeNi₂ medium-entropy alloys induced the precipitation of a high-hardness HCP phase, resulting in pronounced grain refinement and substantial strength enhancement predominantly through second-phase strengthening. Likewise, El Garah et al. [17] reported that appropriate Y addition to TiTaZrHfW high-entropy films modified the phase constitution, refined the microstructure, and improved mechanical properties via the formation of nanocrystalline L1₂ and hexagonal Y-rich phases. These findings collectively demonstrate that RE element addition is an effective strategy for tailoring the microstructure and enhancing the properties of high-entropy alloys.

However, only limited studies have explored RE doping in EHEAs under additive manufacturing conditions, and most existing reports concern cast alloys or coatings rather than PBF-LB/M-fabricated EHEAs [13,18,19]. To address this gap, the present study investigates for the first time the effects of Y addition on the microstructure and mechanical performance of PBF-LB/M-processed AlCoCrFeNi2.1, using a direct comparison with the undoped counterpart to elucidate the role of Y in tailoring microstructure and properties.

2. Materials and Methods

The general powder fabrication process and scanning strategy were consistent with our earlier work on laser spot diameter effects [20]. Y-doped (0.05 at%) and undoped AlCoCrFeNi_2.1 EHEA powders were synthesized via a plasma rotating electrode atomization system (AVI-PREP-4W). After sieving to a particle size range of 15–53 μm, the powders—exhibiting Hall flow rates below 35 s/50 g—were used for PBF-LB/M, also referred to as laser powder bed fusion, LPBF; hereafter “printing”). All raw materials had a certified purity exceeding 99.9 wt% according to the supplier’s analysis, and this was further confirmed by inductively coupled plasma optical emission spectrometry (ICP-OES). The powder morphology and characteristics are shown in Figure 1. Prior to printing, the powders were dried at 80 °C.

Cubic specimens (10 × 10 × 10 mm³) were manufactured on GH4169 substrates using a LiM-X150A PBF-LB/M system equipped with an IPG 500 W fiber laser, under vacuum with the oxygen partial pressure reduced to below 120 ppm. The selection of process parameters was based on prior experimental experience to ensure high densification and stable melt pool formation, and the details are summarized in

Table 1. PBF-LB/M processing parameters.

Parameters	Laser power P, [W]	Scanning speed v, [mm/s]	Scanning spacing d, [mm]	VED, [J/mm−3]
Value	130~180	500~800	0.09	60.2~133.3

. All builds employed the same scanning strategy with a 67° rotation between successive layers and a layer thickness of 0.03 mm. A schematic of the process and the sampling positions for microstructural and mechanical characterization are shown in Figure 2.

The volumetric energy density (VED) was calculated as:

$$VED={\displaystyle \frac{P}{v·h·t}}$$

where P is the laser power (W), v the scanning speed (mm/s), h the hatch spacing (mm), and t the layer thickness (mm). The density of the printed specimens was measured using the Archimedes drainage method according to:

$${\rho }_{measured}={\displaystyle \frac{m_{air}}{m_{air}-m_{water}}}·{\rho }_{water}$$

where m_air and m_water are the specimen weights in air and water, respectively, and ρ_water is the density of water at the test temperature. The relative density was then calculated as:

$$Relative Density\left(\%\right)={\displaystyle \frac{{\rho }_{measured}}{{\rho }_{theoretical}}}\times 100\%$$

For each processing condition, three specimens were measured and the average density was calculated. The set of parameters yielding the highest average relative density was identified as the optimal condition, and the corresponding specimens were selected for subsequent microstructural and mechanical characterization.

Metallographic samples were sectioned from the interior of the printed blocks by electrical discharge machining (EDM), with the observation plane (X-Y) perpendicular to the build direction. The specimens were sequentially ground with SiC papers from 80# to 4000# and subsequently polished with diamond suspensions. Phase identification was conducted via X-ray diffraction (XRD, D8 ADVANCE) (Bruker AXS GmbH, Karlsruhe, Germany) at a scan rate of 2°/min over a 2θ range of 10–100°. Macro-defects and microstructural features were examined using an optical microscope (VHX-1000) (Keyence Corp, Osaka, Japan) and a scanning electron microscope (SU-5000) (Hitachi High-Tech Corp, Tokyo, Japan), while electron backscatter diffraction (EBSD, Oxford C-Nano) (Oxford Instruments, Abingdon, UK) was performed with a step size of 0.08 μm to determine phase fractions, grain boundary character, and crystallographic orientations. Tensile specimens were fabricated under the optimized PBF-LB/M parameters to evaluate high-temperature performance and fracture behavior. All tensile tests were performed on an INSTRON 5982 universal testing machine (Instron Corp., Norwood, MA, USA) at a constant strain rate of 0.05 s⁻¹, with each test repeated three times and the average values reported.

3. Results and Discussion

3.1. Relative Density and Processing Parameters

Figure 3 presents the relationship between VED and the relative density of AlCoCrFeNi_2.1 EHEA samples with and without 0.05 at% Y doping. The relative density of the specimens increases initially and then decreases with rising VED. The Y-doped sample achieves the highest relative density of 99.23% at a lower VED of 79.4 J·mm⁻³, whereas the undoped counterpart requires a higher VED of 91.2 J·mm⁻³ to reach the same density level. In addition, the Y-doped samples exhibit a more gradual decline in relative density in the high-energy regime (VED > 111.1 J·mm⁻³), indicating a broader processing window and enhanced tolerance to energy input fluctuations.

Further insights into the observed densification behavior can be obtained by examining the evolution of defect types through metallographic analysis. At lower VED levels (62.5 J·mm⁻³), specimens exhibit prominent lack-of-fusion pores and partially incomplete melted powder particles, indicating insufficient energy input during laser processing. This leads to incomplete powder melting and poor melt pool wettability, resulting in interlayer bonding defects. Such defects significantly impair metallurgical continuity between layers and constitute a major obstacle to achieving full densification. In contrast, when the VED is excessively high (129.6 J·mm⁻³), severe localized evaporation of the alloy occurs, making it difficult for the generated metal vapor to escape in time, thereby forming gas pores. Additionally, the associated vapor recoil and spatter—as well as melt pool instabilities—can deteriorate interlayer integrity and reduce the overall relative density [21].

The addition of Y exhibits a pronounced optimizing effect on the densification behavior of the alloy. At relatively low energy input conditions, Y is believed to improve the laser energy absorption efficiency by altering the surface characteristics of the powder particles or through the formation of Y-containing surface oxides, thereby enhancing the coupling between the laser and the material [22]. This leads to more effective energy utilization and facilitates the formation of a stable and continuous melt pool, even under low VED conditions. Furthermore, Y also plays a role in modifying the surface tension and wettability of the molten metal, improving the wetting between the melt pool and adjacent unmelted powders, and strengthening interfacial bonding [23]. Under high VED conditions, Y may influence the surface tension gradient and suppress recoil pressure by stabilizing melt pool dynamics. Its strong affinity for oxygen contributes to the formation of thermodynamically stable Y₂O₃ particles, which further reduces the oxygen content in the melt and suppresses pore formation by minimizing vapor-induced spatter [24].

3.2. Microstructural Morphology and Phase Analysis

Figure 4 shows the representative eutectic microstructure of the 0.05 at% Y-doped AlCoCrFeNi_2.1 EHEA fabricated by PBF-LB/M. As shown in Figure 4a, the center of the melt pool is predominantly occupied by cellular eutectic structures, while lamellar eutectics are mainly distributed near the melt pool boundaries, growing perpendicular to the local thermal gradient and forming well-aligned banded morphologies. These lamellar and cellular morphologies are further highlighted in Figure 4d and Figure 4e, respectively. The distinct banded alignment indicates strong thermal gradient control during rapid solidification, and the transformation of some lamellar regions into fine equiaxed cellular structures suggests the influence of thermal cycling-induced localized recrystallization, particularly at overlapping scan zones [18,25]. Compared to the coarse and irregular lamellar eutectics observed in as-cast AlCoCrFeNi_2.1 alloys [26], the rapid cooling rate intrinsic to the PBF-LB/M process results in significant refinement of the eutectic microstructure and promotes a more uniform spatial distribution [18,19,25].

In addition to this overall refinement effect, more detailed morphological differences are observed between the Y-doped and undoped samples, as illustrated in Figure 4b and Figure 4c, respectively. The eutectic structure in the Y-doped alloy appears significantly finer, with shortened and narrowed lamellar features and more compact eutectic colonies. This structural refinement may be attributed to the influence of Y on the solidification process, where Y atoms or Y₂O₃ nanoparticles potentially act as heterogeneous nucleation sites or impede the growth of eutectic fronts, thereby suppressing coarse structural development [18]. Moreover, the lamellar eutectic fraction increased from 61.3% in the undoped sample to 74.8% in the Y-doped sample (values obtained by ImageJ 1.54p area-fraction analysis), suggesting that Y addition stabilizes the lamellar growth mode under rapid solidification conditions. This could result from localized compositional adjustments or interface energy modifications induced by Y [18,27,28]. Additionally, eutectic boundaries in the Y-doped sample (Figure 4c) are noticeably more regular and continuous, indicating enhanced interfacial stability and solidification control [19]. In contrast, the undoped sample exhibits more irregular colony boundaries and coarser cellular regions. These observations collectively demonstrate that Y addition not only promotes refinement of the eutectic microstructure—as confirmed by ImageJ-based measurements showing a reduction in average lamellar spacing from ~0.65 µm in the undoped alloy to ~0.48 µm in the Y-doped alloy—but also enhances its uniformity, manifested as more continuous and regularly aligned eutectic boundaries (Figure 4c). In addition, Y doping contributes to improved morphological stability, here referring to the ability of the lamellar structure to resist coarsening and maintain its morphology under subsequent thermal exposure or mechanical loading, which is critical for ensuring structural reliability during service [15,17]. This stabilization is closely related to the influence of PBF-LB/M processing on nucleation dynamics and interfacial characteristics: the high cooling rates and steep thermal gradients inherent to PBF-LB/M promote dense nucleation and sharp interfaces, while Y segregation further reduces interfacial energy and stabilizes eutectic growth [27,29].

Figure 5 displays the XRD patterns of samples with and without 0.05 at% Y doping. Both samples exhibit characteristic diffraction peaks corresponding to a dual-phase structure consisting of face-centered cubic (FCC) and body-centered cubic (BCC) phases. In the Y-doped sample, a noticeable leftward shift in the FCC main peak is observed, which is attributed to lattice expansion caused by the incorporation of large-radius Y atoms (1.80 Å) into the FCC matrix. Moreover, a distinct superlattice reflection near 2θ ≈ 44.5°, associated with the ordered B2 phase, is clearly visible in the undoped sample but becomes significantly attenuated after Y addition. This observation suggests that Y may introduce lattice distortion and slow diffusion—two factors known to impede the kinetics of long-range ordering [30], and similar Y-induced reductions in B2 ordering have been documented in NiAl-based B2 alloys [31]. Thus, Y doping could plausibly suppress B2 ordering here, which may improve structural stability under thermal exposure and delay creep-related degradation while reducing brittleness and enhancing ductility [11,17].

To elucidate the influence of Y addition on elemental distribution within the eutectic microstructure, point EDS analyses were conducted on the FCC and BCC phases in both 0.05 at% Y-doped and undoped AlCoCrFeNi_2.1 alloys, as shown in Figure 6 and

Table 2. Element Point Scanning by EDS.

Point	Element, at%
	Al	Co	Cr	Fe	Ni
Site 1	0.94	23.34	22.57	23.38	29.77
Site 2	12.13	15.27	7.12	13.25	52.23
Site 3	1.16	23.26	24.82	23.54	27.22
Site 4	9.81	16.64	7.88	13.13	52.54

. In both conditions, the FCC phase (sites 1 and 3) is relatively enriched in Co, Cr, and Fe, while the BCC phase (sites 2 and 4) exhibits higher concentrations of Al and Ni—consistent with the typical partitioning behavior observed in this alloy system [25,32]. Specifically, the addition of Y promotes the enrichment of Al in the BCC phase. Compared to the undoped alloy, the Al content exhibits a more pronounced variation in both phases upon Y doping (FCC: 1.16 → 0.94 at%; BCC: 9.81 → 12.13 at%), along with observable changes in the distribution of Co and Cr. These results suggest that Y incorporation affects solute partitioning during solidification, particularly enhancing the segregation tendency of Al toward the BCC phase. This redistribution may favor a more robust Al–Ni coupling environment, thereby contributing to improved structural ordering and chemical stability of the alloy.

These compositional redistributions, though modest, carry important implications for phase stability and atomic-scale interactions within the alloy. In particular, the elevated Al content in the BCC phase of the Y-doped sample, combined with the retention of high Ni concentration (~52 at%), suggests a more favorable Al–Ni chemical environment that may stabilize the ordered BCC lattice [33]. Thermodynamically, Al and Ni have a strong tendency to form B2-type structures, which constitute the foundational framework of the BCC phase in eutectic high-entropy alloys [26,33]. Furthermore, Yttrium’s strong affinity for both Al and oxygen likely reduces the local oxygen activity, suppressing premature Al oxidation and improving the availability of Al for lattice incorporation [34]. Y segregation at the solid–liquid interface can reduce interfacial energy and alter melt viscosity, which facilitates smoother solute redistribution and more efficient atomic packing during rapid solidification [18,19]. Collectively, these effects potentially contribute to improved structural ordering and chemical stability of the BCC phase, which are beneficial for strengthening and thermally stabilizing the alloy under service conditions [19,21].

Figure 7a and Figure 7c present the EBSD inverse pole figure (IPF) maps of the undoped and Y-doped samples, respectively. No significant difference is observed in the overall grain orientation distribution between the two samples, indicating that Y addition does not fundamentally alter the macroscopic crystallographic texture of the alloy (grains are defined by EBSD with misorientation >15°, and size was measured by the equivalent circle diameter method). The phase distribution maps (Figure 7b,d) reveal that Y addition significantly increases the volume fraction of the BCC phase. This can be attributed to the strong chemical affinity between Y and Al, which promotes the formation of stable coordination structures, in contrast to the relatively weaker interactions between Y and Fe or Cr [18,29]. Despite the increased BCC content, the intensity of the B2 superlattice reflection in the XRD patterns is markedly reduced. This apparent contradiction suggests that while Y enhances the formation of BCC phases, it simultaneously hinders their long-range ordering, leading to a greater proportion of disordered BCC structures [26,31]. Considering the rapid solidification and elemental segregation characteristic of the PBF-LB/M process, localized enrichment of Y may interfere with the coordinated arrangement of Ni and Al atoms, thereby disrupting the formation of the ordered B2 phase and diminishing its corresponding diffraction intensity.

However, grain size statistics (Figure 8a) reveal that Y doping leads to notable grain refinement, with the average grain size decreasing from 0.72 μm to 0.66 μm (measured from EBSD maps using AztecCrystal). This refinement may result from solute undercooling, grain boundary pinning, or the inhibition of grain boundary migration during solidification induced by Y atoms [35]. In addition, EBSD results indicate a reduced proportion of low-angle grain boundaries (LAGBs: 2–10°) after Y incorporation, decreasing from 23.1% in the undoped alloy to 16.3% in the Y-doped alloy. Although rare earth doping is often considered to stabilize dislocations and suppress recovery, the interfacial segregation behavior of Y under rapid solidification conditions in this study is more likely to facilitate heterogeneous nucleation [36]. This mechanism promotes the formation of grains with greater misorientation, thereby increasing the fraction of high-angle grain boundaries (HAGBs: >10°).

3.3. Tensile Properties and Analysis of Fracture Morphology

The tensile behavior and corresponding fracture morphologies of the 0.05 at% Y-doped and undoped AlCoCrFeNi_2.1 eutectic alloys are presented in Figure 9 and Figure 10, respectively. As shown in the engineering stress–strain curves (Figure 9a), both sets of samples exhibit high strength levels, with yield strengths exceeding the typical values reported for cast counterparts (>800 MPa). Notably, the Y-doped samples show a slight increase in yield strength (from 1019 MPa to 1103 MPa) and ultimate tensile strength (from 1383 MPa to 1475 MPa). More notably, elongation is improved from 13.0% to 16.3% after Y doping (Figure 9b), indicating a significant enhancement in plasticity. Fractographic observations further support the mechanical data. As shown in Figure 10a,b, the undoped alloy exhibits a relatively flat fracture surface dominated by quasi-cleavage features, including cleavage steps and river patterns, which are characteristic of brittle or mixed-mode failure. A few small dimples are occasionally observed, suggesting localized plastic deformation. In contrast, the Y-doped sample reveals a markedly different morphology (Figure 10c,d), with abundant and uniformly distributed shallow dimples, tear ridges, and ductile shear features, indicative of microvoid coalescence and a transition toward ductile fracture behavior.

The improved mechanical performance in the Y-doped alloy can be rationalized by considering the microstructural and phase evolution induced by Y addition. The reduction in average grain size and the significant decrease in the fraction of low-angle grain boundaries (from 23.1% to 16.3%) enhance intergranular plastic compatibility, enabling more uniform slip transfer between neighboring grains. This effect can be described in terms of Taylor’s hardening relationship [37]:

$$\sigma =M\alpha Gb\sqrt{\rho }$$

where σ is the flow stress, M is the Taylor factor, G the shear modulus, b is the Burgers vector, and ρ is the dislocation density. Using the Hall–Petch relation with k ≈ 0.4 MPa·m^1/2, the refinement from 0.72 µm to 0.66 µm corresponds to an additional ~15–20 MPa in yield strength. This compensates for the softening from B2 disordering and provides a significant share of the observed strengthening. A refined and less misoriented grain structure suppresses localized dislocation pile-up, reducing strain concentration and enhancing macroscopic ductility.

XRD results further indicate the destabilization of the ordered B2 structure upon Y doping, as reflected by the attenuation of superlattice peaks. The increase in the disordered BCC fraction facilitates dislocation glide and slip system activation. However, local strengthening is still retained due to residual nanoscale order or interphase barriers, which may introduce Orowan-type dislocation bypassing [38]:

$${\sigma }_{Orowan}={\displaystyle \frac{0.4Gb}{\mathsf{\lambda }}}ln\left({\displaystyle \frac{r}{b}}\right)$$

where λ is the interparticle spacing and r is the obstacle radius. Based on lamellar spacing refinement from ~0.65 µm to ~0.48 µm (measured by ImageJ from Figure 4), the Orowan contribution is estimated at ~15–20 MPa. This compensates for the softening from B2 disordering and provides the largest share of the observed strengthening. The partial disordering of B2 leads to refined structural units that continue to act as effective barriers for dislocation motion.

Finally, the addition of Y introduces significant atomic-scale lattice distortion due to its large atomic radius (1.80 Å), which alters local bonding environments and suppresses oxidation-driven defect formation at grain boundaries [27]. The magnitude of size mismatch can be estimated by the atomic misfit parameter δ:

$$\delta =\sqrt{\sum c_i{\left(1-{\displaystyle \frac{r_i}{\overline{r}}}\right)}^2}$$

where c_i and r_i are the atomic fraction and radius of each constituent, and $\overline{r}$ is the average radius [39]. For the present alloy, δ increases from ~6.1% (undoped) to ~6.5% (Y-doped), yielding an additional solid-solution strengthening of ~20–25 MPa. The larger δ promotes local stress fields and impedes dislocation glide.

4. Conclusions

In this study, Y-doped AlCoCrFeNi_2.1 eutectic high-entropy alloys were successfully fabricated via PBF-LB/M, and their microstructure and mechanical properties were systematically investigated. The main conclusions are as follows:

• Y addition significantly refines the eutectic microstructure and increases the volume fraction of the BCC phase, while simultaneously suppressing the long-range ordering of the B2 structure due to lattice distortion and solute redistribution effects.
• EBSD analysis reveals a reduced grain size and a notable decrease in the proportion of low-angle grain boundaries, indicating enhanced intergranular misorientation and refined grain boundary structures, which contribute to both strength and ductility improvement.
• Tensile tests demonstrate a simultaneous enhancement in mechanical performance, with yield strength increasing from 1019 MPa to 1103 MPa, ultimate tensile strength from 1383 MPa to 1475 MPa, and elongation from 13.0% to 16.3%. These improvements are attributed to multiple strengthening mechanisms including dislocation hardening, grain boundary strengthening, and second-phase strengthening.
• Fractography confirms a transition from quasi-cleavage to ductile fracture mode upon Y addition, with the fracture surface showing abundant dimples and tear ridges, indicative of enhanced microvoid coalescence and improved plastic deformability.

Overall, the results demonstrate that trace Y doping is an effective strategy to optimize the microstructure and mechanical performance of EHEAs processed via PBF-LB/M, offering valuable insights for the development of high-performance structural materials in critical domains such as nuclear power, aerospace, and submarine engineering, where enhanced strength–ductility synergy and stability under extreme environments are required. Nevertheless, comprehensive evaluation of high-temperature mechanical behavior, oxidation resistance, and long-term thermal stability is still required, which will be the focus of our future investigations.