Introduction

The development of advanced materials capable of withstanding the extreme conditions in nuclear fusion reactors represents a critical challenge for achieving sustainable fusion energy1,2. Plasma-facing materials (PFMs), which are directly exposed to intense particle fluxes and extreme heat from high-temperature plasma, must simultaneously resist high-energy neutron irradiation and withstand helium (He) and hydrogen (H) fluxes from the plasma environment3.

Tungsten (W) has been recognized as a leading plasma-facing material due to its remarkable properties. Its refractory nature, high melting point, excellent thermal conductivity, low erosion rates, and outstanding thermomechanical stability make it highly suitable for use as the first wall material in ITER and future fusion reactors3,4,5. However, the formation of helium (He) bubbles during helium irradiation raises significant concerns about the long-term structural integrity of plasma-facing components manufactured from pure tungsten4,6. In recent years, high-entropy alloys (HEAs) have been identified as promising alternatives to traditional materials for use in extreme environments7,8,9,10,11,12,13,14,15,16. Among these, W-based HEAs have demonstrated exceptional properties, including high melting points and superior mechanical performance at elevated temperatures, surpassing those of Ni-based superalloys and nanocrystalline tungsten12,13. Additionally, W-based HEAs have exhibited significantly enhanced radiation resistance compared to pure tungsten7,8,9,10,11,14,15,17. For instance, alloys such as MoNbTaVW and MoNbTaTiW have shown greater irradiation resistance than pure tungsten7,11. However, the inclusion of high-activation elements such as Mo and Nb in these alloys raises concerns regarding their long-term applicability as plasma-facing materials (PFMs). To address this issue, W-Ta-Cr-V HEAs have been developed as a potential solution8,9,10,14,15. An experimental study on W38Ta36Cr15V11 revealed remarkable irradiation resistance when exposed to 1 MeV Kr+2 ions, even at high irradiation doses of up to 8 displacements per atom (dpa) at 800 °C9. Unlike pure tungsten, this alloy exhibited no evidence of dislocation loop formation under these conditions9. Furthermore, this HEA exhibited exceptional resistance to helium (He) bubble damage at 1223 K, with small (~ 2–3 nm) bubbles growing uniformly at a slow rate and no preferential formation on grain boundaries14. Molecular dynamics (MD) simulations further revealed that, although the number of Frenkel pairs (FPs) created in the primary damage state of W38Ta36Cr15V11 is higher than in pure tungsten, the interstitial cluster size and dislocation loop density are significantly lower17. These findings highlight the superior radiation resistance of this HEA17. Furthermore, ab initio simulations have attributed the radiation resistance of WTaCrV alloy to the slowed interstitial diffusion and the high recombination probability of interstitials and vacancies in the alloy18. An experimental study on equimolar WTaCrV HEA has demonstrated the formation of helium (He) bubbles with diameters below 1 nm during helium irradiation, indicating exceptional resistance to He irradiation-induced defect accumulation10. These results highlight the alloy’s microstructural stability and resistance to irradiation hardening. Additionally, a recent atomistic simulation study demonstrated the exceptional resistance of equimolar WTaCrV HEA to surface modifications caused by energetic helium ions, further highlighting their potential for plasma-facing materials8.

The majority of studies on HEAs have focused on their radiation response to heavy ion, neutron, or He irradiation. However, to evaluate the radiation response of materials under fusion reactor-relevant conditions—characterized by fast neutrons and gas production from transmutation reactions—a more complex irradiation environment is required. Atwani et al. addressed this challenge by mimicking such conditions through dual-beam irradiation with 1 MeV Kr⁺ and 16 keV He⁺ ions to test the radiation resistance of W29.4Ta42Cr5.0V16.1Hf7.5 and W31Ta34Cr5.0V27Hf3.0 at 1173 K15. Their findings revealed no dislocation loops, even after 8.5 displacements per atom (dpa) and 9.13% He implantation. However, cavities were observed under these conditions15.

Despite these promising results, no fundamental studies have yet been conducted to investigate the formation and evolution of defects in W-Ta-Cr-V HEAs under complex irradiation environments. Specifically, the combined effects of preexisting helium (introduced through He injection) followed by neutron or ion irradiation remain unexplored. In this work, we present a direct comparison of the radiation resistance of tungsten and equimolar WTaCrV HEA with 0%, 1%, and 2% He, added during the simulation setup. Overlapping displacement cascade simulations were employed to mimic neutron irradiation. These simulations, widely used to model radiation damage in metals and alloys, provide a robust platform for studying the production and evolution of defects11,19,20. Utilizing a recently developed interatomic potential—constructed based on density functional theory (DFT) data for WTaCrV-He properties and validated against experimental observations of He bubble growth8,16—MD simulations provide a reliable platform for evaluating the comparative performance of these materials. The insights obtained from this study will enhance our understanding of the behavior of plasma-facing components (PFCs) under complex irradiation conditions. Furthermore, this work evaluates whether W-Ta-Cr-V HEAs remain strong candidates for advanced plasma-facing materials in the demanding environments of fusion reactors.

Results

Figure 1 compares the evolution of defects in pure tungsten (W) and WTaCrV HEA in the absence of helium as a function of dose, up to 550 successive cascades, corresponding to 0.2 displacements per atom (dpa) based on the NRT formula at 300 K. It should be noted that in Figs. 1, 2 and 3, the ‘maximum size’ of defect clusters refers to the number of defects (vacancies, interstitials, or He atoms) contained within the largest cluster observed. The results indicate that, while the concentration of Frenkel pairs (FPs) is higher in the HEA compared to pure W (Fig. 1a), the tendency of interstitials to form clusters is significantly lower in the WTaCrV HEA (Fig. 1b, d, and e). In pure W, large interstitial clusters and dislocation loops are observed, whereas in the WTaCrV HEA, no dislocation loops or interstitial clusters larger than 10 atoms are detected. Notably, within the dose range of 0.15–0.2 dpa, the mean maximum interstitial cluster size in pure W is approximately 26 times larger than that in the WTaCrV alloy (Fig. 1b). For vacancies, the majority in both W and the HEA appear as single vacancies (Fig. 1f,g). Additionally, approximately 20% of the vacancies in both materials form vacancy clusters, typically ranging in size from 2 to 4 vacant sites. Larger vacancy clusters are observed in the HEA but represent only a small fraction of the total vacancy population.

Fig. 1
figure 1

Analysis and visualization of defect evolution in W and WTaCrV HEA under overlapping cascade simulations: (a) Concentration of Frenkel pairs as a function of dose. (b) Maximum size of interstitial clusters as a function of dose. (c) Maximum size of vacancy clusters as a function of dose. Fraction of interstitials in different cluster size ranges as a function of dose in (d) WTaCrV HEA and (e) W. Fraction of vacancies in different cluster size ranges as a function of dose in (f) WTaCrV HEA and (g) W. Visualization of defects at 0.2 dpa, including vacancies (blue), interstitials (red), and dislocations (green for ½ < 111 > loops and pink for < 100 > loops) for (h) WTaCrV HEA and (i) W.

Full size image
Fig. 2
figure 2

Analysis and visualization of defect evolution in W and WTaCrV HEA with 1% He atoms inserted, under overlapping cascade simulations: (a) Concentration of Frenkel pairs as a function of dose. (b) Maximum size of interstitial clusters as a function of dose. (c) Maximum size of vacancy clusters and He clusters as a function of dose. Fraction of interstitials in different cluster size ranges as a function of dose in (d) WTaCrV HEA and (e) W. Fraction of vacancies in different cluster size ranges as a function of dose in (f) WTaCrV HEA and (g) W. Visualization of defects at 0.2 dpa for (h) WTaCrV HEA and (i) W: one frame displays vacancies (blue), interstitials (red), and dislocations (green for ½ < 111 > loops and pink for < 100 > loops), while another frame shows the positions of He atoms (light blue).

Full size image
Fig. 3
figure 3

Analysis and visualization of defect evolution in W and WTaCrV HEA with 2% He atoms inserted, under overlapping cascade simulations: (a) Concentration of Frenkel pairs as a function of dose. (b) Maximum size of interstitial clusters as a function of dose. (c) Maximum size of vacancy clusters and He clusters as a function of dose. Fraction of interstitials in different cluster size ranges as a function of dose in (d) WTaCrV HEA and (e) W. Fraction of vacancies in different cluster size ranges as a function of dose in (f) WTaCrV HEA and (g) W. Visualization of defects at 0.2 dpa for (h) WTaCrV HEA and (i) W: one frame displays vacancies (blue), interstitials (red), and dislocations (green for ½ < 111 > loops and pink for < 100 > loops), while another frame shows the positions of He atoms (light blue).

Full size image

We introduced 1% helium atoms into the simulation box for each system and subsequently performed hundreds of overlapping cascade simulations to reach 0.2 dpa. Figure 2 compares the evolution of defects in W and WTaCrV HEA with 1% helium atoms as a function of dose. The results indicate that the concentration of FPs is significantly higher in W compared to the HEA (Fig. 2a). Comparing this with Fig. 1a reveals that the introduction of 1% He dramatically increases the FP concentration in W at 0.2 dpa (approximately five times higher), while the FP concentration in WTaCrV HEA increases by only 1.5 times compared to the case without He. Figure 2b, d, e and h show that in the HEA, no dislocation loops or large interstitial clusters are observed, whereas in W, a significant fraction of interstitials forms large dislocation loops. Comparing these results with Fig. 1 reveals that the introduction of 1% He notably increases the fraction of interstitials in large loops in W at 0.2 dpa. Figure 2c, f, and g provide insights into vacancy and He clusters. Figure 2c shows that the growth of the largest vacancy clusters is proportional to the growth of the largest He clusters, as vacancies are a primary source of He trapping. This leads to the formation of VnHem bubbles, where n represents the number of vacancies and m represents the number of helium atoms. The presence of 1% He increases the fraction of vacancies in larger clusters for both W and the HEA, as shown by comparing Fig. 1f,g with Fig. 2f,g. In WTaCrV, vacancies predominantly form clusters of 11–20 and 20–51 vacant sites, whereas in W, the fraction of vacancies in 5–10 site clusters increases with dose. Figure 2h illustrates that in the HEA, large vacancy clusters are closely associated with He clusters, forming VnHem bubbles. These bubbles are spatially well-separated, with surrounding regions containing single He atoms or very small clusters (2–4 sites), and exhibit spherical shapes. In contrast, Fig. 2i shows that in W, He bubbles are more heterogeneous in size, morphology, and spatial distribution.

Figure 3 illustrates the radiation response of both systems under cascade simulations with 2% helium atoms introduced into the simulation box at the start of the simulations. Increasing helium concentration to 2% significantly impacts FP formation. At 0.2 dpa, FP concentrations in W are 2.4 times higher than in the HEA (Fig. 3a). Comparing Fig. 2a with Fig. 3a shows that increasing the helium content from 1% to 2% raises the FP concentration in W by 2.4 times and in WTaCrV HEA by 2.6 times at 0.2 dpa. Notably, dislocation loops are absent in the HEA even at 0.2 dpa with 2% helium (Fig. 3h). However, the formation of larger interstitial clusters is observed in this system, with an increased fraction of clusters in the 5–20 atom size range (Fig. 3d). Additionally, a small fraction of interstitials forms a cluster containing 289 interstitials (Fig. 3b). By comparing Figure 2c, f, and h with Figure 3c, f, and h, it becomes clear that increasing helium content leads to the growth of VnHem clusters in the HEA under irradiation. Figure 3f reveals that, in the HEA with 2% He and irradiated up to 0.2 dpa, the majority of vacancies were concentrated in clusters ranging from 21 to 50 vacancies. The growth of VnHem clusters with increasing helium content is also observed in W. Figure 3c shows that the largest vacancy clusters in W are comparable in size, or even slightly larger, than those in the HEA after 0.1 dpa. Additionally, Figure 3g indicates that the fractions of vacancies as single vacancies, 2–4 vacancy clusters, 5–10 vacancy clusters, and 11–20 vacancy clusters are nearly identical. However, a slightly smaller fraction is observed to accumulate in the 21–50 vacancy cluster size range.

In W, the presence of 2% helium significantly promotes the formation of massive interstitial clusters during overlapping cascade simulations. Figure 3b and e show the development of a large interstitial network with increasing dose, reaching a size of up to 12,869 atoms at 0.2 dpa and forming a distinct cross-like structure (Fig. 3i). The detailed progression of these interstitial clusters can be observed in Movie S1, while Fig. 4d–g provide two-dimensional snapshots showing the formation and evolution of this extensive polygonal interstitial network. Dislocation behavior in W is further analyzed in Fig. 4. As shown in Fig. 4a–c, increasing helium content leads to higher total dislocation densities. However, in W with 2% He, a noticeable drop in dislocation density begins at 0.05 dpa, ultimately leading to the formation of a large polygonal interstitial network by 0.19 dpa (Fig. 4d–g). Figure 4b,c indicate that both ½ < 111 > and < 100 > dislocation loops form in W, with ½ < 111 > dislocations being the predominant type in all cases.

Fig. 4
figure 4

Comparison of (a) total dislocation lengths, (b) ½ < 111 > dislocation lengths, and (c) < 100 > dislocation lengths as a function of dose for tungsten with 0%, 1%, and 2% preexisting He. (d–g) Two-dimensional frames illustrating the evolution of defects, including vacancies (blue), interstitials (red), and dislocations (green for ½ < 111 > loops and pink for < 100 > loops), in W containing 2% He at four points indicated in (a). For better visualization, the simulation box is replicated twice in the x and z directions using periodic boundary conditions.

Full size image

Discussion

Figure 1 demonstrates that the concentration of Frenkel pairs (FPs) in WTaCrV HEA is higher than in W up to 0.2 dpa. However, while interstitial clusters in W grow to significantly larger sizes compared to the HEA, vacancy clusters are slightly larger in the HEA. These results align with recent molecular dynamics (MD) simulations, which utilized a newly developed classical interatomic potential for WTaCrV HEA17. The primary radiation damage in W and W38Ta36Cr15V11 was compared under various primary knock-on atom (PKA) energies ranging from 1 to 100 keV. The simulations revealed a similar trend to what was observed in the overlapping cascade simulations at the primary stage of damage17. Additionally, a recent MD study employing overlapping cascade simulations with a machine learning-based interatomic potential investigated defect evolution up to 0.4 dpa in W and several medium- and high-entropy W-based alloys, among which WTaV is the closest in composition to the WTaCrV HEA studied here11. The results from that study are broadly consistent with ours, with one notable difference: in W, the concentration of FPs is lower than in WTaV up to ~ 0.1 dpa, whereas in our case, this trend persists up to 0.2 dpa. This discrepancy arises from differences in simulation size and the displacement energy (Ed) values used, which affect dose calculations in NRT simulations. Despite this, the saturation FP concentration in WTaV (~ 0.3–0.35%) observed in that study aligns closely with the saturation observed for WTaCrV in our study11. Moreover, the maximum sizes of vacancy and interstitial clusters in WTaCrV from our simulations are comparable to those reported for WTaV. In contrast, in W, the maximum vacancy cluster size is smaller than in WTaV (at least up to 0.1 dpa), while the maximum interstitial cluster size in W is significantly larger than in WTaV, consistent with our observations11.

When overlapping cascade simulations are performed in W with preexisting helium, the formation of interstitial clusters and the density of dislocation loops increase significantly (Figs. 1, 2, 3 and 4). Experimentally, the formation of dislocation loops and lines in W under dual-beam irradiation (involving He ion and heavy ion beams) has been observed, while the formation of bubbles strongly depends on helium content and temperature21,22. For instance, in an experimental study with 0.3% helium in the peak damage region under dual-beam irradiation (He⁺ and Fe⁺), bubbles were only observed at temperatures of 500 °C and above21. In our simulations, increasing helium content enhances the tendency of vacancies to form larger clusters. The formation and growth of VnHem clusters are expected to reduce interstitial-vacancy recombination and, consequently, lead to a higher interstitial concentration in W.

The formation of polygonal interstitial networks (Fig. 4d–g) in W containing 2% He during cascade simulations is a notable observation in this study. We hypothesize that these planar interstitial networks may result from dislocation loop rafting, a phenomenon reported experimentally in irradiated W under both single23 and dual-beam irradiation conditions (involving He and heavy ion beams)24. At room temperature, the primary mechanism driving dislocation loop rafting involves the prismatic glide of glissile interstitial clusters and the rotation of their Burgers vectors under irradiation conditions23. In our simulations, as the dose increases, the system accommodates the high defect density by aligning interstitial clusters and dislocations into planar, joint structures. Helium plays a critical role in this process by stabilizing defects and modifying local stress fields. Additionally, helium may act as pinning sites that guide the planar alignment of interstitial clusters and could increase internal stresses that facilitate the development of these polygonal networks.

When comparing the evolution of defects in WTaCrV HEA with 0%, 1%, and 2% helium, as shown in Figs. 1, 2, and 3, several observations become apparent. First, no dislocation loops are formed in any of these cases. This result is consistent with findings reported in the experimental literature. Specifically, it has been demonstrated that for W38Ta36Cr15V11 irradiated with heavy ions up to 8 dpa at high temperatures, no dislocation loops were observed when helium content was 0%9. Similarly, in another experimental study using dual-beam ion irradiation with Kr⁺ and He⁺ ions, no dislocation loops were observed in W29.4Ta42Cr5.0V16.1Hf7.5 and W31Ta34Cr5.0V27Hf3.0 HEAs, even after 8.5 dpa at elevated temperatures, despite the helium concentration reaching 9.13%15. The underlying mechanism for this behavior can be attributed to DFT calculations conducted by Zhao18, which highlight the slowed interstitial diffusion in the WTaCrV alloy. This reduced mobility inhibits the clustering of interstitials necessary for dislocation loop formation.

The second observation is that the fraction of vacancies in large vacancy clusters, as well as helium clusters, increases with higher helium content. The growth of VnHem clusters follows a distinct pattern: the clusters are well-spaced, with surrounding regions containing isolated He atoms or very small helium clusters (2–4 sites), and they exhibit a spherical morphology. In contrast, the bubble distribution morphology in W differs, featuring helium clusters of varying sizes that are distributed with similar populations. While WTaCrV HEA is resistant to the formation of large cavities under helium ion irradiation10,14, cavity formation in this family of alloys has been reported under dual-beam irradiation at 1173 K15. Several mechanisms explain the growth of bubbles in WTaCrV HEA under overlapping cascade simulations. First, helium atoms exhibit lower diffusivity in WTaCrV alloy compared to W. This reduced mobility is evident in Fig. 5, which shows the trajectory paths of a He atom in both systems. This low diffusivity is one of the reasons why experimentally, only small bubbles are observed in WTaCrV when irradiated solely with helium atoms.

Fig. 5
figure 5

Trajectory paths of a He atom in W (upper panel) and WTaCrV HEA (lower panel) at different temperatures. Simulations were conducted for up to 4 ns.

Full size image

Another key factor is the dependence of He cluster formation on vacancy availability. In WTaCrV, the binding energy of vacancies to He clusters is relatively high, facilitating bubble growth. Studies have shown that the binding energy between helium atoms (in the absence of vacancies) is negligible in WTaCrV HEA, whereas in W, it becomes significant, particularly as the number of helium atoms in the cluster increases8. However, for VnHem clusters with n > 2, the mean binding energy ranges from 1–2 eV, enhancing stability and promoting growth in the HEA8,16. Figure 6 demonstrates that in WTaCrV, even at 1800 K where He atoms can diffuse effectively, He clusters do not form without vacancies, whereas in W, He clusters can form even in the absence of vacancies. Additionally, Fig. 6 illustrates the formation of VnHem clusters in the presence of 9 vacancies at different temperatures for both systems. In WTaCrV HEA, formation of bubbles and growth under overlapping cascade simulations can be described in three distinct phases. Initially, vacancies are created via displacement cascade mechanisms, and helium atoms diffuse into these vacancies to form VnHem bubbles. Helium migration in WTaCrV alloy is limited due to the high migration barrier of He atoms, and mobility is primarily induced by the high energy of recoil atoms and localized thermal spikes generated during displacement cascades. As the simulation progresses, bubble growth occurs through two primary mechanisms. First, single helium atoms are incorporated into existing VnHem clusters, facilitated by vacancy trapping sites. Helium atoms can displace lattice atoms during this process, increasing the number of vacancies in the cluster. Second, nearby VnHem bubbles can coalesce under the influence of energy from displacement cascades, resulting in larger bubbles with higher helium content (Fig. 7a). Finally, the stability of large VnHem bubbles becomes apparent. Due to their size and helium content, these bubbles are energetically favorable and highly resistant to deformation or shrinkage under further displacement cascades. This stability, as shown in Fig. 7b, is critical for understanding bubble evolution under irradiation conditions.

Fig. 6
figure 6

Evolution of 20 He atoms (1%) at different temperatures in WTaCrV HEA and W: (a) WTaCrV HEA without preexisting vacancies, (b) WTaCrV HEA with 9 preexisting vacancies, (c) W without preexisting vacancies, and (d) W with 9 preexisting vacancies.

Full size image
Fig. 7
figure 7

(a) Snapshots illustrating how displacement cascade induces the coalescence of two small VnHem bubbles. (b) Snapshots demonstrating the stability of a large bubble under displacement cascades. To visualize the displacement cascade area, only metallic atoms with kinetic energies exceeding 0.87 eV are shown.

Full size image

In summary, we investigated defect evolution in W and WTaCrV HEA with and without preexisting helium atoms under overlapping cascade simulations up to 0.2 dpa at 300 K using molecular dynamics. In W, we observed the formation of dislocation loops and large interstitial clusters, regardless of preexisting helium. Increasing helium content and dose resulted in higher dislocation densities and the development of a polygonal interstitial network. The size and quantity of helium clusters grew with increasing dose and helium content, with notable differences in their size and spatial distribution between W and WTaCrV alloy. While WTaCrV HEA exhibited strong resistance to dislocation loop formation and large interstitial clusters, it was less resistant to bubble formation, particularly at higher helium concentrations. These findings suggest the necessity of revisiting the design principles for HEAs to achieve better radiation resistance under such complex irradiation environments.

Methods

All molecular dynamics simulations in this study were conducted using the Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS)25. We employed the embedded atom method (EAM) potential developed by Zhou et al. to describe interatomic forces between metallic atoms8,26. The interactions among He atoms were modeled using the pair potential developed by Becker27, while the He-metal interactions were based on the pair potential formalism introduced by Juslin and Nordlund (JN)8,28,29. To accurately capture short-range interactions, all potentials were smoothly integrated with the Ziegler–Biersack–Littmark (ZBL) repulsive potential8,30. We created multiple simulation boxes for overlapping cascade simulations in W and WTaCrV systems, with and without preexisting He. Each box consisted of 55 × 55 × 55 bcc unit cells aligned along the < 100 > , < 010 > , and < 001 > directions, containing 332,750 W atoms. HEA configurations were generated by randomly replacing W atoms with Ta, Cr, and V atoms to achieve an equimolar composition (W:Ta:Cr:V = 1:1:1:1). Helium atoms were then added randomly at 1% and 2% concentrations relative to the total number of atoms for He-containing systems. Periodic boundary conditions were applied in all directions, and energy minimization was performed using the conjugate gradient (CG) method to stabilize the system. The system was then equilibrated at 300 K and zero pressure using the NPT ensemble for 100 ps. To simulate irradiation, a central metallic atom was assigned a recoil energy of 10 keV in a random direction, initiating a collision cascade. A recoil energy of 10 keV was selected, as it lies within the range frequently observed for recoils in tungsten under fusion-relevant conditions31, and is commonly used in overlapping cascade simulations of BCC metals and alloys to balance damage representativeness and computational tractability11,32,33,34. The simulations were performed under constant volume conditions using the microcanonical (NVE) ensemble. To dissipate the heat generated during the cascade, only the outer 5 Å boundary region of the simulation box was coupled to a Berendsen thermostat35 set at 300 K for 20 ps. Subsequently, the entire box was stabilized at constant pressure using a Berendsen barostat for an additional 10 ps. This irradiation process was repeated 550 times per simulation cell, achieving ~ 0.2 dpa based on the NRT approximation36 i.e. \(\left(\frac{n}{N}\right)\times \frac{{E}_{recoil}}{2{E}_{d}}\), where n indicates the number of cascades, N represents the total number of atoms in the box, Erecoil refers to the energy of recoiled atom, and \({E}_{d}\) signifies the displacement threshold energy. Ed = 40 eV was chosen for all the metallic elements in accordance with Ref.9. To achieve uniform irradiation, the simulation box was randomly shifted across periodic boundaries following each cascade. Vacancy and interstitial concentrations were assessed using the Wigner–Seitz method37,38. Frenkel pairs were defined as the number of vacancies or interstitials, which are equal following displacement cascades. Cluster identification followed established conventions for BCC metals39: vacancy clusters were identified using a cutoff equal to the average of the second and third nearest-neighbor distances, and interstitial clusters using the average of the third and fourth nearest-neighbor distances. For helium clusters, a cutoff equal to the lattice constant of the crystal was used. Dislocation structures were analyzed using the Dislocation Extraction Algorithm (DXA)40, with all analyses conducted in OVITO software41. For the simulations corresponding to Figs. 5 and 6, smaller simulation boxes containing approximately 2000 atoms were created, with dimensions of 30.764 Å × 30.764 Å × 30.764 Å along the < 100 > , < 010 > , and < 001 > directions. The same procedures used for generating the WTaCrV HEA composition, as well as the energy minimization and equilibration steps applied to the larger samples, were employed for these smaller simulation boxes.