Effect of Heat Dissipation Rate on Microstructure and Mechanical Properties of Al_0.5FeCoCrNi High-Entropy Alloy Wall Fabricated by Laser Melting Deposition

Abstract

High-entropy alloys (HEAs) are a new type of multi-component alloy. The design of the compositions breaks the design ideas of traditional alloys and shows many excellent properties. Therefore, an Al_0.5FeCoCrNi HEA with face-centered cubic (FCC) and body-centered cubic (BCC) dual-phase structure was used in this paper. During the additive manufacturing process, the heat dissipation rate gradually changes with the increase in wall height. As a result, the composition of the phases changes, resulting in differences in mechanical properties. Here, we designed laser melting deposition (LMD) on T-beams of different heights to change the heat dissipation rate of the wall, and the effects of the heat dissipation rate on the microstructure and mechanical properties of Al_0.5FeCoCrNi HEAs were studied. The experimental results showed that increasing the height of the T-beam would gradually slow down the heat dissipation rate of the wall. The above phenomena not only led to a gradual reduction of the BCC phase under the influence of heat accumulation but also increased the length of columnar crystals in the wall with the slowing of heat dissipation. Heat accumulation hindered the nucleation during solidification and eventually led to the growth of grains across the deposition layer. Furthermore, the slow heat dissipation rate changed the grain number and BCC phase content, which gradually decreased the strength and hardness, while the ductility of the samples improved.

1. Introduction

Although high-entropy alloys (HEAs) are composed of various components, a simple single-phase solid solution structure without intermetallic compounds is shown in the alloys, such as a face-centered cubic structure (FCC), body-centered cubic structure (BCC), or hexagonal close-packed structure (HCP). It exhibits excellent mechanical properties [1,2,3], thermal stability [4], friction and wear properties [5], corrosion resistance [6], and some other novel properties. Therefore, HEAs have high academic research value and industrial application potential.

Additive manufacturing (AM) is an emerging technology that has a non-negligible contribution to modern production [7,8]. Among many additive fabricating technologies, laser melting deposition (LMD) technology has the advantages of high forming efficiency and complex forming structure and has become one of the most promising metal additive manufacturing technologies [8]. It is worth noting that in 2015, Joseph et al. [9] proved that the laser melting deposition technique is a process capable of successfully fabricating high-entropy alloys, laying a foundation for the research of additive manufacturing of multiple systems of HEAs. Al_0.5FeCoCrNi HEAs occupy an important position in the Al_xFeFeCoCrNi HEA system that gradually appears in the vision of researchers. Because Al_0.5FeCoCrNi HEAs have an FCC + BCC dual-phase structure, they can make up for the problem that in the traditional HEAs system, a single FCC solid solution phase or a single BCC solid solution phase can only achieve the mechanical response of high plasticity and low strength or high strength and low plasticity. HEAs with multiphase structures can obtain an excellent combination of strength and plastic toughness through reasonable composition design, which has important reference value for the research [10,11,12,13,14]. Sokkalingam et al. [13] fabricated Al_0.5CoCrFeNi HEAs by gas tungsten arc welding and found that the rapid heating and cooling during fabrication prevented the formation of the BCC phase. Then, in the mechanical property test, it was found that the hardness and the degree of work in hardening the samples decreased slightly, and the ductility increased due to the decrease of the BCC phase content. Niu et al. [12] fabricated an Al_0.5CoCrFeNi HEA by arc melting. After 1 h of heat treatment, the increase in the BCC phase led to an increase in strength. Lin et al. [11] studied the microstructure and hardness properties of an Al_0.5CoCrFeNi HEA with FCC solid solution structure. However, after an aging treatment at 350~950 °C, the Al_0.5CoCrFeNi HEA transformed into an FCC + BCC dual-phase solid solution structure. The aged Al_0.5CoCrFeNi alloy also realized an improvement in hardness. It was found that the thermal effects of different processes affected the phase composition of the Al_0.5CoCrFeNi HEA, which changed the mechanical property. During additive manufacturing, the height of the wall increases, and the rate of heat dissipation changes. At present, there are few studies on the LMD of Al_0.5FeCoCrNi HEAs, and the effect and mechanism of heat dissipation during additive manufacturing on microstructure and mechanical properties are still unclear.

Therefore, different heights of T-beams were used to change the heat dissipation rate of the wall, and the effect of heat dissipation rate on the microstructure and mechanical properties of Al_0.5FeCoCrNi HEAs was studied. Moreover, the strengthening mechanism of Al_0.5FeCoCrNi HEA walls was analyzed. The research results in this paper provide an experimental basis for the fabrication and application of additive manufacturing Al_0.5FeCoCrNi HEAs.

2. Experimental Procedures

2.1. Materials

The substrate used in the laser melting deposition process was 316 stainless steel (120 × 40 × 10 mm³) and its chemical composition is shown in

Table 1. Chemical composition of 316 stainless steel (wt.%).

Si	Mn	Cr	Ni	Mo	C	Fe
0.63	1.19	17.99	12.84	2.56	0.07	Bal.

. Pre-alloyed powder of Al_0.5FeCoCrNi high-entropy alloy was prepared by argon atomization (202107-PRD-64, Jiangsu Vilory Co., Ltd., Xuzhou, China) and used in the laser melting deposition process. Its chemical composition is shown in

Table 2. Chemical composition of Al_0.5FeCoCrNi HEA (wt.%).

Al	Fe	Co	Cr	Ni
5.81	23.17	Bal.	22.26	24.41

. The molar ratio of each constituent element was 0.5:1:1:1:1 (Al:Fe:Co:Cr:Ni). Figure 1a shows the scanning electron microscope (SEM, HITACHI, Tokyo, Japan) image of the Al_0.5FeCoCrNi HEA raw powders. The powder particles were spherical, which was beneficial to fluidity during the powder feeding process. The particle size distribution of the Al_0.5FeCoCrNi HEA powder ranged from 45~150 μm, with an average particle size of 122 μm, and the specific size distribution and density distribution are shown in Figure 1b.

2.2. Measurement of Thermal Cycling of LMD-Al_0.5FeCoCrNi Walls

The laser melting deposition was conducted using a HANSGS-RJ0016-F3K processing system, and the schematic diagram of the appearance is shown in Figure 2a. In this experiment, a single scanning strategy was adopted; that is, unidirectional scanning was performed according to the specified starting point to the endpoint to avoid the influence of heat accumulation. Figure 2b is a schematic diagram of the single scanning strategy. There was a total of 10 scanned layers. After each layer was scanned by a laser, the height of the laser head was raised by a distance of 1 mm. According to the previous preparation experience [15], the Al_0.5FeCoCrNi HEA wall was fabricated with a laser power of 1000 W in this experiment. The whole scanning process moved at a scanning speed of 300 mm/min. The powder passed through the closed-loop powder feeding device and was finally fed into the molten pool from the coaxial nozzle. Feeding speed was controlled at 30 g/min. During the whole scanning process, the shielding gas (argon) flow rate used was kept at 5 NL/min to ensure that the oxygen content was limited to below 10 ppm.

In order to explore the effects of different heat dissipation rates on the microstructure and mechanical properties of the Al_0.5FeCoCrNi HEA fabricated by LMD, we designed an experiment that would deposit Al_0.5FeCoCrNi HEA walls on T-beams of different heights, as shown in Figure 3a. The size of the bottom plates of the T-beams was 120 × 75 × 5 mm³, and beams with a length of 120 mm, a thickness of 5 mm, and heights of 20 mm, 40 mm, and 60 mm were vertically welded to the bottom plate. At the same time, to clarify the temperature change of the samples and the thermal history of the entire deposition process of the T-beams at different heights during the LMD, a K-type thermocouple was used for measurement in this experiment. The positions of the thermocouples placed during the experiment are shown in Figure 3b. To prevent the influence of laser irradiation or unstable heat conduction on the thermocouple probe during the deposition, for the substrate with a beam height of 0 mm, the thermocouple was placed in a central position on both sides of the substrate and 12 mm away from the wall. For the baseplates with beam heights of 20 mm, 40 mm, and 60 mm, the thermocouples were placed in a central position on both sides of the T-beams and 12 mm away from the top of the T-beams. In addition, the samples were named T-0 sample, T-20 sample, T-40 sample, and T-60 sample according to the heights of the T-beams when the samples were fabricated.

2.3. Microstructure Characterization and Mechanical Properties

In order to analyze and compare the microstructures and grain growth of the Al_0.5FeCoCrNi high-entropy alloy wall samples, a DK77 Electric Discharge Cutting Machine was used to cut the walls along the deposition direction. To reveal the microstructure, the polished metallographic samples were corroded with aqua regia (the volume ratio of concentrated hydrochloric acid and concentrated nitric acid was 3:1) for 5–15 s. The OLYMPUS GX51 optical microscope (OM, Olympus Co., Ltd., Tokyo, Japan) was used to observe the sampling position as shown in Figure 4a, and then the OLYMPUS M3 software was used for further analysis to obtain the microstructures and grain growth of the Al_0.5FeCoCrNi HEA wall samples. The phase analysis of the Al_0.5FeCoCrNi HEA wall samples was carried out by X-ray Diffraction (XRD, Rigaku D/max/2500PC, Hitachi Co., Ltd., Tokyo, Japan). The sampling location is shown in Figure 4d. The phase identification was carried out using a SmartLab advanced X-ray diffractometer at a scanning speed of 4°/min and a range of 20° to 100°. The analysis for the XRD pattern was performed using the Jade 6.5 software. Electron Backscattered Diffraction (EBSD) analysis was performed on a TESCAN MAIA3 (Tisken (China) Co., Ltd., Brno, Czekh) super-resolution field scanning emission electron microscope, and the sampling location is shown in Figure 4a. In the process of EBSD information acquisition, the accelerating voltage of the electron microscope was 15 kV, and the step sizes were 0.3 μm and 4 μm according to the needs.

The microhardness of the wall samples was tested by the HMV-2T Vickers microhardness tester (Shimadzu Co., Ltd., Tokyo, Japan). From the substrate plane to the top of the samples, the distance between the two adjacent points was 1 mm as shown in Figure 4b. At the same height, the average hardness of the Al_0.5FeCoCrNi HEA wall sample was the average value obtained after 5 measurements. The applied load was 2.942 N and the loading time was 15 s. The tensile sampling location is shown in Figure 4a, and its dimension is shown in Figure 4c. The tensile properties of the samples were tested by a DDL 10 universal tensile testing machine, and the stress-strain curves of the samples were obtained. The stretching speed was 10⁻³ s⁻¹. The tensile fracture morphology of the samples was characterized and analyzed by an S-3400N scanning electron microscope (SEM, Hitachi Co., Ltd., Tokyo, Japan).

3. Results and Discussion

3.1. Heat Dissipation Rates Analysis of LMD-Al_0.5FeCoCrNi Walls

In the process of laser melting deposition, the molten pool temperature of the laser scanning would rise sharply. After each layer was deposited, due to the rapid cooling process, the deposition layer temperature would drop rapidly. Therefore, in the process of the LMD, with the increasing number of sedimentary layers, the temperature curves of the deposited samples are shown in Figure 5. It can be seen that the peak temperature of each pass in the temperature curves of the four groups shows a gradual increase. This is because the first layer was deposited on the T-beam at room temperature and had the best heat conduction condition. However, with the layer-by-layer deposition of the HEAs wall, the laser transmitted heat to the new layer of the wall, where the heat input in the molten pool was more than the heat loss, causing the temperature to rise layer by layer [16]. This shows that there was obvious heat accumulation in the wall during the process of the LMD. Furthermore, specimens deposited at different beam heights had different heat accumulations. The heat accumulation of the T-60 was the most serious, the heat accumulation of the T-40 was the second, the heat accumulation of the T-20 was weak, and the heat accumulation of the T-0 was the weakest. By comparing the temperature curves of these four samples, it can be seen that the peak of the temperature curves began to stabilize almost around the 8th layer. This shows that in the process of laser melting deposition, the heat input from the laser to the samples and the heat dissipation caused by the increase in the height of the samples reached a dynamic balance process [17].

To better compare the difference in heat dissipation rates, the derivation of the first-order function was performed for the temperature curves in Figure 5. The corresponding heating and cooling rate diagrams of Al_0.5FeCoCrNi HEA wall samples under different T-beam heights were obtained, as shown in Figure 6. It can be seen that the heating rates of the T-20 sample, T-40 sample, and T-60 sample were consistent. For example, in the process of LMD of the first layers, the heating rates were 15 °C·s⁻¹, and the heating rates of the second layers were also about 12 °C·s⁻¹. The above derivation shows that the laser heating efficiencies were the same under three different beam heights and had a similar thermal cycling process. Although there was a certain similarity in heating, the cooling rates of the T-20 sample, T-40 sample, and T-60 sample were different. The cooling rate of the T-20 sample was the fastest, followed by the T-40 sample, and the T-60 sample was the slowest. The differences in cooling rates show that the sample was affected by the heat dissipation method [18]. In the heat dissipation process of metal additive manufacturing, the heat dissipation methods are mainly heat convection, heat radiation, and heat conduction. Compared with heat convection and heat radiation to the air, heat conduction has a higher heat dissipation efficiency and is the most important process of heat dissipation [17,18,19]. When deposited on a T-beam with a beam height of 60 mm, the heat that could be transferred by thermal conduction was the least, so the T-60 sample had the slowest cooling rate. Bai et al. [18] established a 3D transient model and obtained the same results by comparing and discussing the computational and experimental results for layer 1, layer 2, and layer 21 depositions. The T-0 sample dissipated directly through the base plate downwards. The heat transfer rate of this heat conduction was the most efficient, which resulted in rapid heat dissipation even at higher heating rates. By comparing the cooling curves of the T-0 sample with the T-60 sample, T-40 sample, and T-20 sample in the later stage, it was found that the T-0 sample still had a very high cooling rate, about 5 °C·s⁻¹. However, the cooling rates of the T-60 sample, T-40 sample, and T-20 sample changed from 5 °C·s⁻¹ at the beginning to 1 °C·s⁻¹. It can be seen that the heat dissipation mode through the substrate was very efficient and continued throughout the heat dissipation process of the T-0 sample.

3.2. Effect of Heat Dissipation Rate on the Microstructure of LMD-Al_0.5FeCoCrNi Wall

3.2.1. Phase Analysis of LMD-Al_0.5FeCoCrNi Walls

To explore the effect of different heat dissipation rates on the phase of Al_0.5FeCoCrNi high-entropy alloy walls, EBSD and XRD techniques were used to characterize the samples. Figure 7 is the EBSD phase distribution diagrams of the Al_0.5FeCoCrNi HEA walls. It can be seen that there were FCC + BCC dual-phases in the samples under four different heat dissipation rates from T-0 to T-60 (Figure 7a–d), in which the FCC phase represented by the blue area was dominant, and the BCC phase (red) gradually decreased. In T-0, T-20, T-40, and T-60 samples, the contents of the BCC phase were 12.5%, 7.8%, 6.5%, and 2.7%, respectively. In order to see the distribution of the two phases more clearly, we photographed the local area of the T-0 sample at a high magnification, and it was found that the FCC phase was generally distributed in the intragranular area, while the BCC phase was mostly distributed between dendrites, as shown in Figure 7e,f.

The XRD results of the Al_0.5FeCoCrNi samples are shown in Figure 8. It can be seen that the FCC + BCC dual-phase existed in the Al_0.5FeCoCrNi walls with different heat dissipation rates. The XRD of each sample was further analyzed and calculated, and the phase ratios of the Al_0.5FeCoCrNi samples under different heat dissipation rates were obtained, as shown in Figure 8b. Through the analysis and comparison, it is found that the percentage of the BCC phase decreased with the gradually slowing heat dissipation rate, which is consistent with the previous results obtained in the EBSD data (Figure 7). The reason for this phenomenon was the difference in heat dissipation rates. According to Figure 5, there was a significant heat accumulation on the Al_0.5FeCoCrNi HEA walls during the LMD process. The effect of the heat dissipation method means that the higher the height of the beam, the more serious the heat accumulation would be. Heat accumulation caused the wall to maintain a high temperature for a period of time. At high temperatures, the BCC phase tends to transform into the FCC phase [14,20]. Therefore, with the continuous increase in beam height, the percentage of BCC phase in the sample decreased continuously.

Furthermore, it can be seen from Figure 8a that with the gradual slowing of the heat dissipation rate, the diffraction peak in the (200) direction gradually increased, indicating that the preferential growth of the samples became stronger. Since the detection results of XRD alone may bring errors, EBSD technology was used to further determine the preferred orientation of the samples. The pole maps of the T-0, T-20, T-40, and T-60 specimens are shown in Figure 9, respectively. It can be seen from the pole maps that the four walls all formed the <001>//ND silk texture orientation, and the texture was very obvious. This is because, in the process of the LMD, the laser was emitted perpendicular to the molten pool so that there was a maximum temperature gradient perpendicular to the bottom of the molten pool. In this case, the grains would preferentially grow along the direction of the maximum temperature gradient, so the orientation of the grains tended to coincide with the direction of the maximum temperature gradient in the molten pool [21]. After quantitative measurement, the texture intensities of T-0, T-20, T-40, and T-60 were 22.6 (×random), 27.3 (×random), 30.2 (×random), and 31.6 (×random), respectively. The results show that with the increase in beam height, the heat dissipation rate gradually slowed down, and the strength of the formed texture gradually increased.

3.2.2. Microstructures of LMD-Al_0.5FeCoCrNi Walls

The microstructures of the Al_0.5FeCoCrNi high-entropy alloy walls at different heat dissipation rates are shown in Figure 10. It can be seen that the microstructures of the Al_0.5FeCoCrNi HEA fabricated by different processes are dominated by columnar crystals with different morphologies, and the microstructures in each sample were almost uniform in the vertical direction. For the T-0 sample, due to its faster heat dissipation rate, the grains had been solidified before they grew up, and the microstructure showed relatively short columnar crystals. With the increase in the height of the T-beam, the heat dissipation rate of the wall gradually slowed down and the length of the columnar crystals in the sample gradually became longer. Therefore, the differences in grain morphologies were mainly related to the heat dissipation rates during the deposition process.

According to the Johnson–Mehl equation [22], the relationship between the number of grain nuclei $P\left(t\right)$ formed in time $t$, nucleation rate $N$, and growth rate $V_g$ can be expressed as follows:

$$P\left(t\right)=k{\left(\frac{N}{v_g}\right)}^{3/4}$$

(1)

In the above equation, $k$ is a constant, which is related to the shape of the crystal nucleus; $P\left(t\right)$ is inversely proportional to the grain size $d$. It can be seen from the above equation that the larger the nucleation rate $N$, the finer the grain; the larger the crystal growth rate $V_g$, the coarser the grain. In the same material, both $N$ and $V_g$ depend on the degree of undercooling because $N\propto \exp \left(-\frac{1}{\Delta T^2}\right)$, and when it grows continuously, $v_g\propto \Delta T$. It can be seen that the amount of nucleation per unit time is related to the degree of supercooling. From the point of view of heat dissipation, in the process of LMD of Al_0.5FeCoCrNi walls at different beam heights, with the increase in beam height, the heat dissipation rate gradually slowed down and heat accumulated. Thus, in the process of depositing a new layer, the temperature difference between the new layer and the previous layer decreased with the increase in the height of the T-beam, thereby reducing the degree of subcooling during solidification. According to Equation (1), when the degree of supercooling reduces, $N$ decreases rapidly and faster than $v_g$. Therefore, heat accumulation reduces the degree of subcooling and prevents the nucleation of the new deposition layer, resulting in the continuous growth of grains without encountering resistance from other nuclei during the growth process [23]. Therefore, there were more elongated columnar grains in the T-60 sample.

Furthermore, according to the study by Hou et al. [24], the FCC phase is mainly distributed within the grains, while the BCC phase is mainly distributed in the intergranular region. At the same time, in the solidification process of the BCC phase, the solidification of the FCC phase will play a role in contact inhibition. According to Figure 7b, the content of the BCC phase in the T-0 sample was the highest, so the contact inhibition during the growth of the FCC phase was the strongest; that is, the length of the columnar crystals in the T-0 sample was the shortest.

Although the microstructures were mainly composed of columnar crystals at different heat dissipation rates, the growth was different in the deposition process. Figure 10 shows that the grains gradually grew across the deposited layer boundary (LB) during the deposition of two adjacent layers at different heat dissipation rates.

In the solidification process, the solidification of grains undergoes two processes: nucleation and growth. Extremely high solidification rates and molten pool temperatures are not conducive to spontaneous nucleation. Moreover, the process of the LMD is exposed to a complex environment. In metal solidification, the deposition process tends to introduce small amounts of impurities. Therefore, the grains usually occur through non-uniform nucleation during the solidification process. The non-uniform nucleation work $\Delta G$ can be expressed as [25]:

$$\Delta G=\frac{16\pi {\sigma }^3{T}_m^2}{3{\left(L_m\Delta T\right)}^2}\left(\frac{2-3\cos \theta +{\cos }^3\theta }{4}\right)$$

(2)

In the above equation, σ represents the specific surface energy of the molten metal, L_m represents the heat of fusion, ΔT represents the degree of subcooling in the process, θ represents the wetting angle during nucleation, and T_m represents the melting point of the metal during crystallization. It can be seen from the above expression that nucleation work is mainly related to the degree of undercooling during the solidification process. After nucleation is completed, the grains begin to enter the growth process. For most metallic materials, the dynamic undercooling is very small, so the average grain growth rate is mainly proportional to the undercooling during solidification. The average grain growth rate R can be expressed as [25]:

$$R={\mu }_1\Delta T$$

(3)

In the above equation, ${\mu }_1$ is a proportional constant, the value of which depends on the type of material, and ΔT represents the degree of undercooling during solidification. According to the grain nucleation work and average growth rate equation, the solidification process of metal is mainly related to its undercooling degree. The Al_0.5FeCoCrNi HEA walls were obtained by layer-by-layer deposition. Due to the existence of the dilution phenomenon, the surface of the previously solidified molten pool would be remelted due to the deposition of the new layer. With the accumulation of heat on the deposition layers, the subcooling of the molten pools and the substrate surfaces decreased, which hindered the nucleation of grains in the new deposition layers. Thus, the original columnar grains could grow continuously along the deposition direction without interruptions between layers. With the continuous increase of the deposition height of the samples, the columnar crystals had been continuously grown [26]. Therefore, the grains’ growth morphologies shown in Figure 10 were exhibited.

3.3. Effect of Heat Dissipation Rates on Mechanical Properties of LMD-Al_0.5FeCoCrNi Walls

Based on the analysis of the above microstructures and heat dissipation methods, the mechanical properties of the T-0 sample, T-20 sample, T-40 sample, and T-60 sample were further tested. Since the Al_0.5FeCoCrNi walls experienced different heat dissipation rates in the laser melting deposition process, the mechanical properties would also be different due to their structures.

3.3.1. Microhardness Analysis of LMD-Al_0.5FeCoCrNi Walls

The microhardness test results are shown in Figure 11. The hardness values were lower near the junction of the substrates and the deposited layers. This is because, in the deposition of the first layer, dilution of the composition between the deposited layer and the substrate occurred. The hardness test results obtained near the junction became lower under the influence of the substrate with low hardness. After that, the hardness increased significantly when the deposition height was 1 mm. Due to the increase in deposition height, the dilution effect of components was weakened, and the hardness value suddenly increased. However, with the continuous progress of the LMD process, the hardness values decreased gradually. Overall, there were also obvious differences in the hardness values of different beam heights. The higher the beam height, the lower the hardness of the specimen.

3.3.2. Tensile Properties Analysis of LMD-Al_0.5FeCoCrNi Walls

The engineering stress-strain curves obtained from the uniaxial tensile test are shown in Figure 12. It can be found that LMD-Al_0.5FeCoCrNi HEAs exhibited different tensile properties. As the beam height increased, the tensile strength decreased from 683 MPa to 645 MPa, and the ductility increased from 23.0% to 26.8%.

In general, there should be a large number of dimples in the fractures of tensile specimens with different beam heights, and the dimples are small and dense. We obtained the tensile fracture morphology images of the Al_0.5FeCoCrNi HEA walls by SEM, as shown in Figure 13. It is found that many fractured columnar crystals appeared inside the samples (Figure 13b–d). In addition, there were no obvious spherical impurity particles in the morphology of the dimples, indicating that the powder was sufficiently melted during the LMD process.

The reasons and mechanisms of the alloy fracture could be inferred by analyzing the fracture morphology characteristics. This was closely related to the phase distributions of Al_0.5FeCoCrNi HEAs. Generally, FCC-type HEAs are soft and tough, while BCC-type HEAs are hard and brittle [27,28]. According to Su et al. [29], the BCC phase exhibits poor dislocation carrying capacity due to its high inherent brittleness. It becomes the source of crack initiation of Al_0.5FeCoCrNi HEAs during tensile deformation. Therefore, the cracks start from inside the BCC phase and rapidly propagate to the entire BCC single-phase region. It can be seen from Figure 7 that the FCC phase was distributed inside the grains, while the BCC phase was distributed at the grain boundaries. Therefore, a large number of gullies with columnar crystal morphology appeared on the fracture surfaces. Further analysis of the fracture types shown in Figure 13 shows that there were obvious dimples in the inner region of the grains, while cleavage planes existed at the grain boundaries. We have found in Section 3.2 that with the decrease in heat dissipation rate, the BCC phase decreased and the FCC phase increased, which improved the ductility of the sample. Therefore, there were a large number of dimples at the fracture of the sample, showing an obvious ductile fracture.

3.4. Strengthening Mechanism of LMD-Al_0.5FeCoCrNi

3.4.1. Strength Evolution Mechanism

The KAM diagram can qualitatively reflect the degree of plastic deformation of the grains of the material. Kamaya et al. found that the local misorientation is consistent with the geometrically necessary dislocation density [30]. On this basis, we obtained the KAM distributions (Figure 14) of Al_0.5FeCoCrNi high-entropy alloy walls under different heat dissipation rates. The green areas indicate larger misorientation. By comparison, it was found that large local misorientation often occurred at and near the phase boundaries between the FCC phase and the BCC phase (Figure 14e). This was due to the difference in the structural compositions of the FCC phase and the BCC phase. Serious atomic dislocations often occurred at the junction of the two phases, resulting in a severe distortion of the lattice and thus the formation of an elastic stress field, which was one of the reasons for the formation of dislocations. In addition, due to the structural differences between the two phases (FCC + BCC), the deformation and relative movement between the two phases were restricted, and more energy was often needed for the dislocations and other defects to pass through the joints of the twisted phase. Therefore, the FCC + BCC phase further improved the mechanical properties of the Al_0.5FeCoCrNi HEA due to the differences in the two-phase structure [31].

In addition, the KAM data can be used to quantitatively calculate the geometrically necessary dislocation density to obtain the dislocation content. The detection conditions and step size in the detection process of EBSD have a great influence on the KAM map. Therefore, all measurements in this experiment were carried out under the same conditions with a step length of 4 μm. The geometrically necessary dislocation density (${\rho }^{GND}$) can be calculated using the following equation [32]:

$${\rho }^{GND}=\frac{2KA{M}_{ave}}{\mu b}$$

(4)

In the above equation, μ represents the step size, which is 4 μm; b represents the Burger vector, which is 2.55 × 10⁻¹⁰; and $KA{M}_{ave}$ represents the weighted average KAM value of the EBSD shooting area. The KAM value distributions of the selected areas are shown in Figure 15. At different heat dissipation rates, the average KAM values of the Al_0.5FeCoCrNi HEAs were 0.0453°, 0.0591°, 0.0562°, and 0.0601°, respectively. According to the geometrically necessary dislocation equation, the geometrically necessary dislocation densities of the Al_0.5FeCoCrNi HEAs at different heat dissipation rates were 1.55 × 10¹² m⁻², 2.02 × 10¹² m⁻², 1.93 × 10¹² m⁻², and 2.06 × 10¹² m⁻², respectively. Unlike traditional metal manufacturing methods, laser melting deposition technology has an extremely fast cooling rate (10³~10⁸ K·s⁻¹) [33]. This extremely fast cooling rate was very conducive to the generation of massive dislocations [34,35,36], which further led to complex dislocation interactions during deformation. With the continuous deformation of the metal, the number of dislocations continued to increase, and the interaction between dislocations would further hinder the movement of dislocations, making the deformation hard to continue, thus improving the strength of the materials [35]. In other words, a large number of dislocations increased the strength of the matrix.

Similarly, different heat dissipation rates had an impact on the phase content inside the sample (Figure 8b), which was also the main reason for the differences in mechanical properties. Both FCC and BCC structures include 12 slip systems. Each sliding surface of the FCC structure has three sliding directions, while each sliding surface of the BCC structure has only two sliding directions. Therefore, for the same material, it is easier for the FCC structure to reach the critical resolved shear stress in plastic deformation than the BCC structure. Furthermore, the atomic packing density of the FCC slip surface is larger and its slip resistance is smaller. Thus, the FCC phase has better ductility. Among the samples in this experiment, the T-0 sample showed better strength because it contained a higher BCC phase inside. As the heat dissipation rate slowed down, the BCC content inside the sample gradually decreased, so the strength of the sample showed a gradual decrease.

Furthermore, different heat dissipation rates affected the number of grains in the Al_0.5FeCoCrNi HEA wall samples. Within the range of 400 × 400 μm² in Figure 10, the average grain area and the aspect ratio of the grains inside the samples were calculated. The average grain areas of the T-0 specimen, T-20 specimen, T-40 specimen, and T-60 specimen were 919.72, 1311.16, 3890.31, and 18,077.21 (μm²), and the length-diameter ratios of the four specimens were 3.84, 5.34, 18.42, and 84.63, respectively, as shown in Figure 16.

It can be seen that as the heat dissipation rate gradually slowed down, the average area of the grains increased. In addition, when the width of the grains in the four samples was similar, the aspect ratio increased significantly. Generally speaking, the finer the grain size in the alloy system, the more grain boundaries in the unit volume. The existence of grain boundaries facilitates the release of dislocations during the deformation process, thus reducing the possibility of stress concentration in the local areas to a certain extent, so the possibility of cracking is greatly reduced. In this case, the metal withstands more external forces and thus exhibits better strength. Therefore, with the decrease in heat dissipation rate, the strength of the sample also showed a gradual weakening.

3.4.2. Ductility Evolution Mechanism

In the process of metal deformation, the slip plane and slip direction are often the most densely arranged crystal planes and crystal orientations in the metal crystal. The reason is that the crystal plane with the largest atomic density has the widest interplanar spacing and the least lattice resistance, so slipping along such a plane is more likely to occur. In addition, since the atomic spacing in the closest-packed direction is the shortest, the Burger vector of dislocation is the smallest, and the slip direction is the direction with the largest atomic density. The shear stress $\tau$ in the slip direction of the external force $F$ on the slip surface is [37]:

$$\tau =\frac{F}{A}\cos \phi \cos \lambda$$

(5)

In the above equation, $\cos \phi \cos \lambda$ is the Schmid factor (SF). Generally speaking, the SF of hard orientation is less than 0.35, and the SF of soft orientation is more than 0.40. Figure 17 shows the SF distribution histogram in the (111) <110> direction. The SF ranges from 0 to 0.5, indicating that the difficulty of opening the slip system gradually decreases. The larger the orientation coefficient, the more likely the grain deformation will occur [37]. In Figure 17, the SF was concentrated in the region of 0.4~0.5, indicating that the grains inside the samples were mainly soft-oriented grains, which were more likely to open the slip system and cause plastic deformation under the action of external force. Therefore, the samples under different heat dissipation rates exhibited good ductility. However, due to the mutual restriction relationship between strength and ductility, the tensile test showed that the ductility gradually decreased with the strength increase.

In conclusion, compared with dislocation strengthening, grain boundary strengthening and BCC phase content played a decisive role in the strength of the samples.

4. Conclusions

In the study, T-beams with different heights were used to explore the effect of different heat dissipation rates on the microstructure and mechanical properties of the Al_0.5FeCoCrNi high-entropy alloy walls fabricated by laser melting deposition. Through analysis and comparison, the following main conclusions are drawn:

(1) Different T-beam heights changed the heat dissipation rate of the Al_0.5FeCoCrNi high-entropy alloy wall. With the increase in T-beam height, the heat dissipation rate of the wall gradually slowed down, resulting in differences in the walls.

(2) Under the different heat dissipation rates, the Al_0.5FeCoCrNi high-entropy alloy walls had an FCC + BCC dual-phase structure, in which the FCC phase was dominant. As the heat dissipation rate gradually slowed down, the content of the BCC phase gradually decreased.

(3) The microstructure of the Al_0.5FeCoCrNi high-entropy alloy wall specimen was mainly composed of columnar crystals. At the same time, as the heat dissipation rate slowed down, the growth mode of the columnar crystals across the deposition layers became stronger.

(4) The heat dissipation gradually slowed down, and the strength and hardness of the Al_0.5FeCoCrNi high-entropy alloy wall samples gradually decreased. Compared with dislocation strengthening, grain boundary strengthening and BCC phase content played a decisive role in the strength of the samples, and the ductility gradually increased with the decrease of the strength.