Study of mechanical properties of 2024-RAM2C and 6061-RAM2C aluminium alloys made by laser-based powder bed fusion of metals

Traditional aluminium alloys from 2000 and 6000 series are commonly used for wrought processing. Due to their good properties such as low density, high corrosion, strength and fatigue resistance, wrought aluminium alloys are widely used in the aerospace industry (Varmus et al., 2023; Pantelakis and Alexopoulos, 2008). A well-known example of former series is Aluminium 2024, which primarily contains copper as its alloying element. After being subjected to different heat treatments, it is highly suitable for aircraft applications, such as components of wing and fuselage under tension (El Garchani and Kabiri, 2023). On the other side, an example from the 6000 series is Aluminium 6061, which has magnesium and silicon as major alloying elements (Elementum 3D, 2025). This alloy is more frequently used in heavy-duty structural applications such as truck frames, shipbuilding parts, hi-tech bicycles and motorcycle components.

Additive manufacturing (AM) technology laser-based Powder Bed Fusion of metals (PBF-LB/m) is used to produce complex three-dimensional components with satisfactory mechanical properties in layerwise manner. However, the problem during AM of aluminium alloys 2024 and 6061 is the tendency of these alloys to hot cracking. Hot cracking takes place during the solidification process due to high cooling rate in selective laser melting process (between 10³ and 10⁸ Ks⁻¹) (Tan et al., 2018; Zhang et al., 2019; Aboulkhair et al., 2019). The hot cracks lead to poor mechanical properties and performance of the manufactured components, not even being solved using the post-processing techniques such as hot isostatic pressing (Kotadia et al., 2021).

To diminish the hot cracking phenomenon of aluminium alloys in AM, most of research work has focused on several principal approaches. Process optimisation (Kaufmann et al., 2016; Pekok et al., 2021) is a widespread approach, but it has limited effectiveness in solving hot cracking. The same applies to hybrid processing (Wu et al., 2020; Liu et al., 2022b), which involves remelting and is more commonly used in Directed Energy Deposition with laser beam (DED-LB) than in PBF-LB. The effort to reduce the high cooling rate by increasing the temperature of the powder bed and/or the start plate (preheating) was more successful (Uddin et al., 2018a; Uddin et al., 2018b). The benefit of this approach is a reduction in the thermal gradient between the working temperature of the melt pool and the temperature of the consolidated component during the build. The principal disadvantage of this approach is the inconsistency across the built component, contingent on the thermal conductivity of the material, which gives rise to discrepancies at varying build levels.

An alternative methodology to solve hot cracking is the utilisation of laser beam shaping techniques (Paillier and Prätzsch, 2024; Bakhtari et al., 2024). Beam shaping allows for the precise tailoring of energy delivery to the material, which can consequently result in a notable reduction in the cooling rate. Current limitation of this approach is the cost of the shaping equipment and the low technology readiness level for industrial application.

The final and most spread approach is to modify the cooling rate by modifying the chemical composition, using different additives that promote melt pool stability and lower cooling rates. Various authors have employed a range of additives in their respective studies, including Zr (Mehta et al., 2021a), Zr/Sc (Michi et al., 2022), ZrH₂ (Xu et al., 2023), yttrium stabilised zirconia (Opprecht et al., 2020) and others. A computational design of crack-free aluminium alloys has also been in research focus (Dreano et al., 2022), however it brings to new chemical compositions, which can not have direct industrial application, since those new compositions should first be certified in a long and tedious process.

Some of these material mixtures have already reached the market. The company Elementum 3D (USA) has developed a proprietary additive composition, which works with a significant number of aluminium alloys. Elementum 3D offers advanced gas atomised dispersion-strengthened aluminium AM powders enabled by their proprietary reactive additive manufacturing (RAM) process. The RAM™ process allegedly significantly improves the printability of aluminium alloys 2024 and 6061, making them compatible with the PBF-LB/m technology (Elementum 3D, 2025). This innovative powder-enrichment process inoculates alloys against hot tearing and produces an equiaxed fine-grained microstructure with mechanical properties comparable to those produced with conventional manufacturing technologies. The utilisation of exothermic chemical reactions during the AM process, as facilitated by RAM technology, serves to enhance both the printability and the material properties of the resultant product. The formation of dispersion-strengthened aluminium metal matrix composites within the melt pool, enabled by these reactions, results in the generation of unique and advantageous combinations of product properties, including high ductility, strength, toughness, stiffness, fatigue resistance and high-temperature performance (Elementum 3D, 2025).

Given the numerous approaches available, the use of commercial alloys appears to be the most readily implementable at the industrial level. In accordance, this study presents the findings of a systematic and methodical process development of the provided alloys, including the chemical, static and dynamic testing of the manufactured samples.

Some research on these two commercial alloys is already available in the literature, both in the case of Aluminium 2024-RAM2C ^[1] (Varmus et al., 2023; Panda et al., 2025; Bona and Grande, 2024) and Aluminium 6061-RAM2C (Waller et al., 2025; Torbati-Sarraf et al., 2020). However, in the present work, we extended the study focusing on two main aspects not previously dealt with:

• the process parameterisation has been carried out beyond the full density criterion, also including a preliminary dynamic performance evaluation as a criterion for the final selection of process parameters; and

• a complete mechanical test is offered before and after the heat treatment, showing the influence of the post-treatment on the properties of both 2024-RAM2C and 6061-RAM2C.

As will be shown in this work, the former is relevant since the thermally induced stresses make a significant difference in the quality of the full density materials manufactured with different parameter sets. Although the set of data (two materials, two orientations and three process parameter sets) is not sufficiently big for a conclusive remark, these results allow the authors to pose the hypothesis that the adoption of final process parameters in PBF-LB/m should never end with the density and microstructure evaluation, but it should also include quantitative dynamic testing to prevent the premature failure of the components in exploitation.

In this study, gas atomised powders made of aluminium alloys A2024-RAM2C and A6061-RAM2C, containing specific undisclosed proprietary additives (Elementum 3D, USA), were used as supplied by the manufacturer. The chemical composition of the powder is a proprietary knowledge and was not disclosed by the powder manufacturer. However, the authors have focused on measuring the chemical composition of the processed solid material, which is of greater importance for the intended application and will be discussed subsequently.

The samples were manufactured using the Samylabs Alba 300 metal 3D printer, which is equipped with a 300 W ytterbium fibre laser. Argon was used as a protective atmosphere during the process and the oxygen content was continuously kept below 0.10%, to exclude any influence of the process on fragilization of the material. The compact dimensions of the building chamber (160 mm in diameter by 200 mm in height) and the associated necessity for a modest quantity of powder that can be continuously collected, sieved, and returned to production via bottle system render this equipment ideal for developmental purposes. It is also noteworthy that this equipment is an open system that enables online or on-the-fly interventions, such as modifications to production parameters (e.g., laser power, speed, etc.) or even the feed rate of the powder. This was, however, rarely utilised in this research, as the design of experiment (DoE) of the parameters was precisely conducted prior to the commencement of the build job execution. Figure 1 (a and b), shows two layouts of test samples for production in Samystudio 5.7 software.

Regarding post-treatment, all manufactured samples were subjected to heat treatment for stress relief in an argon-protected oven for a period of 2 h at a temperature of 350°C. This is a standard internal procedure used for aluminium alloys, which has shown to be successful in the AlMgSi₁₀ alloy. Subsequently, the samples were sandblasted using glass microbeads (SiO₂, 150–250 µm) in a blasting cabinet. This type of post-treatment will be referred to as Type 1 in the following chapters. Some of the samples for tensile and Charpy testing were additionally subjected to T6 heat treatment – therefore, Type 2 treatment refers to a sequence of stress relief treatment and T6 heat treatment, performed one after another.

In the case of T6 heat treatment, the following procedures have been used:

• Aluminium 2024-RAM2C: It was first heated to 495°C, held for 2 h, then cooled and artificially aged at 190°C for 10 h (Gairola et al., 2024).

• Aluminium 6061-RAM2C: It was first heated to 535°C, held for 2 h, then cooled and artificially aged at 180°C for 10 h (Liu et al., 2022a; Mehta et al., 2021b).

The selected methodology consisted in the following steps:

• In three occasions during the manufacturing process, the powder size distribution (PSD) of the powder samples was analysed using the 3P Instruments Bettersizer 2600 and corresponding software.

• Measurement of density of the cube samples, resulting from the process parametrization DoE, has been carried out using two dozen 10- × 10- × 5-mm samples, the density of which was determined using the Archimedes principle, in accordance with the ASTM B311 standard.

• Investigation of internal defects, microstructure and (EDX) chemical compositions was carried out using a scanning electron microscope (SEM; Tescan VEGA3) coupled with an energy-dispersive X-ray (EDX) analysis system (Oxford Instruments Ultim max40). The surfaces tested were polished using a sequence of SiC polishing paper 2000, 1000, 220 and then with an MDDAK/CHEM polishing pad.

• The chemical composition of the processed solid material was measured by spark emission spectrometry using the internal standard PE-AQ 112. This was performed at an external partner.

• Hardness of the polished cross-section was measured using a Frank HF1 tester. The maximal load applied was 10 N, and the results are reported in Vickers hardness (HV1) values. The distance between adjacent indentations was meticulously maintained, to prevent any mutual influence (ISO 6507).

• Tensile test has been carried out on horizontal samples, manufactured in a single build and post-treated in both Type 1 and Type 2 conditions, respectively. Testing was performed on the universal testing machine according to the EN ISO 6892-1:2019.

• Charpy impact test was performed, also on horizontal samples manufactured in the same build as the tensile samples, according to EN 148-1:2009 standard. Like the tensile tests, it was performed by an external partner.

• The dynamic (fatigue) test was performed on a mini fatigue testing equipment, MINI FP2 (TP Engineering s.r.l., Italy), using batches of 10 non-standard samples sized as 22 × 7 × 5 mm with a notch in the middle. The method was invented at the University of Parma especially for testing AM materials in a fast and efficient way (Nicoletto, 2021). The testing machine operated at a frequency of 50 Hz with a stress ratio of R = 0. The samples were manufactured using three different sets of process parameters, each one having produced material of >99.90% density. These sets were labelled as HD (the one with the highest relative density), FAST (theoretically, the most productive one among them) – which had been already used to manufacture the static test samples – and ECO (the one that consumes nominally less energy per unit of time). In total, 120 samples have been tested, representing 10 samples manufactured in both the horizontal and vertical orientation (×2), material (×2) and parameters set (×3). The selection of horizontal and vertical orientation is a common practice in this testing procedure, since these two orientations are considered to represent the two extremes of performance, with all other orientations falling somewhere in between. This was sufficient to generate 12 Wohler curves. The samples were subjected only to post-treatment Type 1, which was sufficient for a qualitative comparison between different manufacturing conditions.

During the initial powder spreading test in the printing machine a relatively poor powder spreadability of Aluminium 2024-RAM2C was observed. Figure 2 (b) shows the results of virgin powder particle size distribution, while Figure 2 (a) shows SEM analysis of powder shape and size.

SEM analysis revealed a considerable quantity of powder particles with an irregular morphology, accompanied by a notable number of satellites. The PSD analysis confirmed the presence of up to 21% of particles with a diameter below 0.5 μm, more precisely 10% of the particles are less than 0.070 μm. This was identified as the main cause of poor flowability due to the formation of satellite clusters around the larger particles.

However, after the first sieving, the powder exhibited highly improved flow properties, a finding that was corroborated by additional PSD analysis [Figure 3 (a)]. It is evident that sieving of the powder and its subsequent use resulted in rapid elimination of the fine satellites, thereby significantly enhancing the powder flow [Figure 3 (b)]. The most likely explanation for how the sieving eliminated the satellites is that the light weight of the particles caused them to lift and stick to the walls of the moisture-impregnated lid and net. Since the net and the lid were thoroughly vacuum-cleaned after each sieving batch, and the sieving was made in small batches, due to the low throughput of the experimental sieving equipment (Woven Wire Mesh Sieves, Retsch), the satellites were thoroughly eliminated.

As a general remark, the PSD analysis revealed the presence of a powder with a true PSD of 10–75 μm. The irregular and diverse particle shapes on SEM images do not provide sufficient evidence to ascertain whether the powder production process is a simple gas atomisation or a combination of powder milling and spheroidization.

The situation with Aluminium 6061-RAM2C was different. Based on prior experience with Aluminium 2024-RAM2C, the authors conducted a PSD analysis of the fresh powder before initiating process development [Figure 4 (a and b)]. The results demonstrated the complete absence of minute satellites, a finding that was corroborated by enhanced flowability from the outset. As the production process is very similar to that of Aluminium 2024-RAM2C, the presence of satellites in the former alloy was obviously result of a production problem in a particular batch, rather than a general characteristic of the powder production process. Additionally, the powder particle size in Aluminium 6061-RAM2C was also measured in the range of 10–75 μm, like the Aluminium 2024-RAM2C powder.

The authors acknowledge that both powders have a 10–75 μm PSD with a good flowability, although the powder manufacturing process could not be clearly determined.

To facilitate the comparison of the process parameters in both cases, several common parameters have been selected. These include a layer thickness of 25 μm, which was successfully used with AlMgSi₁₀ to produce high-quality parts with the similar PSD (25-63µm), so it was kept despite a slightly higher upper range of the PSD.

The authors typically combine a line test and a basic DoE for their procedure. The line test melts single weld seams on a single powder layer across various energy densities (ED) to find the optimal result through metallurgical analysis. Once the ideal ED is identified, the basic DoE determines the best combination of laser power, scanning speed and hatching distance to achieve that ED. If this isn’t successful and the resulting densities are low, especially with experimental alloys, statistical modelling is used, a factorial experiment table developed in R.

However, in this specific case, the powder manufacturer provided the ED values (42,6 J/mm³ and 51,3 J/mm³, respectively), and the basic DoE yielded numerous parameter combinations resulting in over 99.7% relative density, making statistical modelling unnecessary. In the case of Aluminium 2024-RAM2C, 18 out of 20 samples exhibited a relative density superior to 99.70%, which was established as a lower desired limit. In the case of Aluminium 6061 RAM2C, only one parameter set yielded a relative density value below 99.70%. This shows that the material exhibits high flexibility to different process parameter sets.

The following process parameters were selected to continue manufacturing the test samples, since they had yielded the highest relative density:

• for Aluminium 2024-RAM2C laser power of 240 W, scanning speed of 1,080 mm s⁻¹ and hatching distance of 0.15 mm;

• for Aluminium 6061-RAM2C laser power of 240 W, scanning speed of 1,315 mm s⁻¹ and hatching distance of 0.15 mm.

They were saved as HD parameters in both cases.

The first step in the material characterisation was to measure the chemical composition of the two alloys in the post-treated Type 1 condition and compare it with the standard for both alloys. The results are shown in following Table 1.

The comparison with the requirements of the material norm shows a coincident difference in a significant increase of titanium (Ti) and boron (B) content, which most likely form the base of the additive(s) used in the RAM™ process. However, the chemical analysis also showed quite lower values of Mg in both alloys, as well as of Mn and Cu in Aluminium 2024-RAM2C, clearly laying out of the limits defined by the standard. This should be noted when considering these AM-made alloys for a specific application, since the acceptance of the components could be called in question.

The SEM images in Figures 5–8 illustrate results of investigation of internal defects, microstructure and chemical compositions (EDX) of Aluminium 2024-RAM2C and 6061-RAM2C, in both Type 1 and Type 2 conditions.

Figure 5 (a) shows the presence of a few pores, but without any hot cracks, which most often occur when processing aluminium alloys of series 2000 using PBF-LB/m technology. There are also isolated areas of pure Ti and B, detected by integrated EDX analysis, but also clearly visible as bigger white/black spots. This confirms the results of the chemical analysis. In comparison to Type 1 post-treatment, the SEM images of samples after Type 2 post-treatment [Figure 5 (b)] have areas with significantly more white spots. In Figure 6 (a), at higher magnification, EDX analysis confirmed that apart from Ti and B, some of these white spots are made up of aluminium and copper compound (Al₂Cu), which are exhibited as smaller plate-like forms, causing precipitation hardening. Figure 6 (b-d) shows spectrum analysis which confirms this.

In the case of Aluminium 6061-RAM2C [Figure 7 (a) with Type 1 and Figure 7 (b) with Type 2 treatment], the presence of a few pores and the absence of hot cracks can be observed at lower magnification. Like Aluminium 2024 RAM2C, clearly visible bigger white/black spots represent Ti and B [also seen on Figure 8 (a)]. In Figure 8 (b), EDX has also confirmed that some of the smaller plate-likes spots are made up of aluminium, iron, silicon and magnesium, which can form AlFeSi and AlMg precipitates and cause hardening after Type 2 heat treatment.

Figures 9 and 10 show the comparative view of different materials properties obtained by hardness, Charpy and tensile tests after ANOVA, made using GraphPad Prism software. For clearer view, given different orders of magnitude, hardness as well as yield and tensile strength are shown on Figure 9, while impact energy, elongation and area reduction are shown in Figure 10. Table 2 shows the measurement values for all parameters, along with the conventional values found in the literature.

To demonstrate the effect of precipitation hardening during the T6 heat treatment (part of Type 2), also referred to in the literature as artificial ageing, and its influence on mechanical properties, samples were first subjected to hardness measurements in both Type 1 and Type 2 post-treatment conditions. The results obtained for Aluminium 2024-RAM2C in the Type 1 condition (stress relief only) show a slightly lower hardness than that of conventionally produced Aluminium 2024 alloy (ASM Aerospace Specification Metals Inc, 2025a), although quite like the previous study of Aluminium 2024-RAM2C with AM technology (Panda et al., 2024). On the other hand, the samples after Type 2 treatment had an increased hardness of 23%. The ANOVA showed an adjusted p-value of 0.0014, making this increase significant.

Similarly, the hardness of the Type 1 Aluminium 6061-RAM2C, with a mean value of 105 HV1, is very similar to the conventional material (107 HV1) (ASM Aerospace Specification Metals Inc., 2025b). In addition, the Type 2 treatment has increased the hardness by 14.3%. However, the ANOVA show an adjusted p-value of 0.1543, which makes this increase insignificant.

The mechanical properties of the developed materials were evaluated by tensile and Charpy impact tests, carried out as part of the static testing procedure. The tests were carried out on horizontal specimens produced in a single build and post-treated in both Type 1 and Type 2 conditions.

To this respect, the graphical view shows that the Type 2 heat treatment had an unanimously good effect on the tensile and yield strength of both 2024-RAM2C and 6061-RAM2C Aluminium, as all parameters increased their average values. This perception of the heat treatment effect was confirmed by ANOVA, which resulted in an adjusted p-value of <0.0001 in all cases. However, it should be highlighted that the tensile and yield strength of both alloys produced by PBF-LB/m are lower than those produced conventionally (Mathweb.com, 2025).

On the other hand, the ANOVA did not show a significant effect of the Type 2 treatment on elongation and area reduction, except in the case of area reduction for 6061-RAM2C, where there is a significant reduction in the T2 condition (adjusted p-value of 0.0375).

Regarding the Charpy test, the impact energy shows a non-significant increase in ductility for 2024- RAM2C Aluminium, but the opposite effect for 6061-RAM2C Aluminium. The adjusted p-value in the latter case was <0.0001.

The fracture area of tensile and Charpy samples have been evaluated by SEM and the following figures show the fracture images. The characteristic tearing dimples, showing a relatively brittle fracture area, can be observed in Figure 11 and 12 [(a and b) in both figures showing Charpy and tensile specimens, respectively].

The use of fatigue testing in process parameters’ development is based on the hypothesis that static behaviour of the samples is predominantly influenced by internal defects in the material, such as porosity, cracks and delamination. In contrast, dynamic behaviour is additionally affected by residual thermal stresses accumulated in the material (Nicoletto, 2018). Consequently, different process parameter sets may yield a metallurgy of comparable relative density and yet exhibit different internal stress distributions, significantly affecting the dynamic behaviour of the material. Since most of the industrial applications imply dynamic loads, finishing the process development without dynamic testing does not seem to be adequate.

However, in typical circumstances, assessing fatigue characteristics of diverse process parameter configurations does not seem feasible, given the extended duration and substantial expenses associated with conducting fatigue tests and obtaining the six S-N curves presented in this study. Nevertheless, the advantage of the Mini FP equipment, originally conceived at the University of Parma and subsequently commercialised by TP Engineering, has facilitated the qualitative analysis of fatigue properties of samples manufactured under different conditions (Nicoletto, 2018). These may include process parameters or post-processing techniques (Varmus et al., 2023). The principal benefit is that the S-N curve can be generated in approximately three to four weeks. This is not a standardised fatigue testing procedure and cannot be used for quantitative evaluation of materials and their dynamic behaviour. However, if the quantitative and normative S-N values are not pursued, it provides a valuable tool for qualitative comparison of different processing conditions.

Figures 13–16 present a comparative view of the S-N curves of two aluminium alloys under study, processed using three different sets of process parameters. The process parameters were selected according to the criteria that the material has a relative density greater than 99.90%. The parameter sets are labelled as HD (the one with the highest relative density), FAST (theoretically, the most productive one among them) and ECO (the one that consumes nominally less energy per unit of time).

The resulting S-N points in each curve have created an extrapolated potential curve. The potential law was selected according to the previous experience and indication of the experts behind the testing technology (Prof. Nicoletto et al.). After the extrapolation, for each set of data minimal absolute error (MAE) has been calculated according to the formula:

Afterwards, the corresponding load for 5.000.000 cycles has been estimated using the potential curves’ formulae also shown at the graph. The difference between the three values is expressed using the highest estimated value as reference.

In addition, an average relative value of MAE is calculated as:

where 12 stands for 12 different curves, showing a relatively small relative MAE, attributed to the test procedure and aligned with the previous experience in literature.

It can be observed that the ECO parameters in 2024-RAM2C produce vertical samples with significantly lower dynamic performance than the other two parameter sets (22.2% lower). This result cannot be attributed in any case to the statistical error of the measurement and are conclusively product of higher level of residual thermal stresses. Additionally, there is a 7.8% decrease in performance of the ECO and HD parameters respect to the FAST parameters in horizontal 2024-RAM2C, as well as a 9.4% decrease in performance of the FAST and ECO parameters compared to the ECO parameters in vertical 6061-RAM2C samples. These deviations cannot be unequivocally attributed to a higher level of residual stress, but the authors consider it likely to be the case. The decline in performance in both cases is of a similar order of magnitude to the corresponding MAE, thus precluding any definitive assertion regarding the inferiority of the resulting material and dynamic properties.

This research thoroughly investigated Aluminium 2024-RAM2C and 6061-RAM2C alloys, manufactured via PBF-LB/m. The study focused on process development and extensive material characterisation, using process parameter insights to analyse their effect on thermal stress accumulation and dynamic properties.

Powder analysis revealed good flowability for both materials, evidenced by uniform particle size distribution and no agglomerates. Although initial 2024-RAM2C samples showed ultra-small satellites, initial sieving resolved this. Density measurements after the parameter DoE confirmed high relative density (above 99.70%) across most parameter combinations, which makes both these alloys very process-flexible. The authors have shown the beneficial impact of the reinforcement by enabling manufacturing without hot cracks and exhibiting excellent, homogeneous reinforcement dispersion.

Chemical composition indicated significant Ti and B, likely forming the RAM™ stabilisation additive. Post-Type 2 heat treatment EDX analysis confirmed Al2Cu, AlFeSi and AlMg precipitates, common in precipitation hardening of Aluminium 2024 and 6061, and beneficial for mechanical properties. Nevertheless, discrepancies in Mg, Mn and Cu content with the standard alloys raised questions about classifying these as Aluminium 2024 and 6061 alloys.

Mechanical tests, endorsed by ANOVA, showed that Type 2 heat treatment significantly increased hardness, yield strength and tensile strength in both alloys. Yet, effects on elongation, area reduction and impact energy varied. While these properties increased in 2024 RAM2C, ANOVA did not confirm its significance. In 6061 RAM2C, they even decreased, indicating increased brittleness, possibly due to a more heterogeneous microstructure. Further crystal structure analysis of 6061 is recommended.

The second part of the study analysed dynamic properties. The fatigue testing method used was non-standardised and suitable only for qualitative comparisons between production conditions, but fast and efficient. The study compared fatigue properties of samples produced with varied process parameters within a single system, with a purpose of correlating them with heat inputs and thermal stresses. Results largely showed consistency across parameter sets, but differences among different sets were noted: in vertical 2024-RAM2C samples, the ECO set showed −22% less resistance load at 5 M cycles, likely due to higher thermal stress accumulation at slower speeds. A performance decrease, greater than MAE, was also observed in horizontal 2024-RAM2C (ECO and HD vs FAST) and vertical 6061-RAM2C (FAST and HD vs ECO).

These findings confirm the authors’ hypothesis that different parameter sets lead to varying residual thermal stresses and significantly impact dynamic performance, which is why it is important to include dynamic assessment before adopting final parameter set as a general rule in PBF-LB. However, more comprehensive investigation with additional alloys is needed to substantiate this hypothesis. Hence, the authors plan to collect more data to support this.

This research was conducted in the framework of the AMNextGenMet project, which was funded by the Austrian Federal Ministry of Climate Action, Environment, Energy, Mobility, Innovation and Technology (BMK). The authors would also like to thank the Elementum 3D Inc. for their generous support.