1. Introduction
Additive manufacturing (AM) technologies are rapidly spreading in several industrial fields. Therefore, the request for metallic powder with specific characteristics for techniques, such as laser powder bed fusion (LPBF), is constantly increasing.
Atomization is undoubtedly the leading technology used to obtain metallic powder suitable for AM processes. The most established commercial atomization methods are water atomization, plasma atomization and gas atomization. With water atomization, it is possible to obtain irregular particles with a high production rate at a low cost [
1]. With plasma atomization, perfectly spherical particles can be produced with a high yield of fine particles, suitable for LPBF technology [
2]. Notwithstanding this, plasma atomization presents high costs and probably for this reason the sector of AM metallic powder production is still led by gas atomization [
3]. This technology allows the production of spherical particles with a good yield of fine particles, with a lower cost than plasma atomization and the advantage of requiring raw materials in an ingot form [
1]. A gas atomizer consists of a furnace for melting the raw material under a protective atmosphere in an atomization chamber. Here, a thin flow of melted liquid is introduced and then dispersed into small droplets under the high pressure of inert gas. During the fall, the droplets can solidify, becoming spherical particles [
3]. On the basis of the atomization parameters (i.e., viscosity of the melted metal and its flow rate [
4], superheating temperature [
5], gas pressure [
6] and nozzle diameter [
7]), it is possible to control the powder characteristics, such as the particle size distribution (PSD), the impurities content, the satellites and irregularly shaped particles presence.
To define whether a metallic powder is suitable for AM processes, the standard ASTM F3049-14 can be considered [
8]. This standard summarizes the main features to consider before processing a new powder: PSD, particle morphology, rheological behavior (flowability, tap density and apparent density) and oxygen content. These analyses can give an idea of the general behavior of the powders and generally, in the literature, it has been demonstrated that they are useful to correlate the powder properties to the AM processability of the powder [
8].
Mussatto et al., for example, performed powder characterization analyses on 316L powders produced by different suppliers. Their analysis included PSD, morphology and flowability, with the purpose of understanding which supplier was the best choice, in terms of powder quality [
9].
Generally, in powder characterization, the first feature to evaluate is the PSD, which significantly influences the final components’ quality [
10]. For example, Averardi et al. asserted that the PSD has a substantial impact on the powder layer packing density, in correlation with the layer thickness and the geometric resolution of the components. In particular, they stated that, particles smaller than 10 µm tend to agglomerate, negatively affecting the flowability [
11]. Hannah et al. studied the influence of the finer particles on the properties of the powders. In their work, they mixed a batch of coarse particles of 316L with finer ones. They found that the mixed powder presented poor flow characteristics (Hausner ratio), with respect to the initial batch of large particles [
12]. Another important contribution of the PSD, is in the laser absorption coefficient. Boley et al. asserted that the size distribution of the powder and their geometrical arrangement are essential factors to control for LPBF process optimization. In particular, in their study, they found that a bimodal PSD can drastically increase the absorptivity of the material, with respect to a monomodal one. This result is more evident in materials with a high reflectivity to the laser [
13].
Both the PSD and the particle morphology, together with many other factors, have a substantial effect on the powder flowability, which significantly influences the LPBF processability. Engeli et al., for example, characterized several batches of IN738LC powder, to understand the influence of powder properties on their processability. They found that the powder with an insufficient flow behavior and a low apparent density presented some issues during the recoating phase, which negatively affected the densification of the part [
14].
Finally, a significant feature of the powder that directly influences the AM parts consolidation and properties, is its chemical composition and, in particular, the impurities content (e.g., oxygen). Leung et al. demonstrated that the powder oxide content could be related to the handling and inadequate storage conditions. The presence of an oxide layer surrounding the particles, can alter the thermodynamics of the melt pool formation, since it can behave as a nucleation site for the pores, negatively affecting the densification of the components [
15].
Following these considerations, it is clear that a complete characterization is mandatory to fully determine the powders’ properties. The AM literature covers the powder characterization and the influence on the final components’ properties, making comparisons between the different powder batches [
5,
10], but also between the different powder production techniques [
16,
17]. However, most of the powders characterized in the literature, are produced by different industrial suppliers that work with large-scale atomization plants, used for massive production. Moreover, several studies were conducted on AlSi10Mg powders produced with laboratory-scale gas atomizers, but these studies focused on the atomization process [
15,
16,
17,
18]. Furthermore, a recent study described the advantages, in terms of the control and stability of the atomization conditions achievable with a laboratory-scale gas atomizer [
19]. In an another study, it was highlighted that the powder produced with a laboratory-scale atomizer fulfilled all of the necessary requirements for the AM processes. In addition, it was asserted that, in the laboratory-scale setup, the oxygen control of the powder can be further improved thanks to the freedom of choice of the higher purity raw materials and processed gas [
20]. Despite this, to the best of the authors’ knowledge, no work, on the comparison between a laboratory-scale produced powder and a commercial one, has yet been published.
Otherwise, this work focuses on the comparison between the commercial-grade and a laboratory-scale AlSi10Mg gas atomized powders. In particular, the two powders that were obtained through the same process but at different scales, were characterized to underline the differences in their properties and in the characteristics of the bulk LPBF samples.
Using a laboratory-scale gas atomizer allows a better control of the process and the successive post-processing steps, compared to an industrial-scale production. This can significantly affect the powders characteristics and, as a consequence, the final properties of the bulk samples.
4. Conclusions
This work aimed to analyze two powders of the same composition but with different origins: a commercial-grade gas atomized AlSi10Mg (CL31) in its “as-received” state and a laboratory-scale gas atomized counterpart (HM), after sieving.
For both powders, the main characteristics of a powder useful for AM, were measured (following the standard ASTM F3096-14) and the bulk samples were produced via LPBF.
The main differences between the CL31 and HM powders resulted in the PSD number and oxidation level. In fact, the HM powder presented a significantly lower amount of finer particles with a lower oxidation level. These results led to better powder packing, due to less friction between the particles, which is improved by a low amount of fine particles and a low level of surface oxidation. These improvements derived from the used gas atomization process parameters, the controlled sieving step and handling operations. As a consequence, better powder packing led to a higher level of densification of the bulk samples during the LPBF process with an improvement in their yield strength.
Considering that a powder with a smaller amount of finer particles is less sensitive to a layer thickness variation [
17], using the HM powder would allow the increase in the LPBF production rate by increasing the layer thickness and consequently decreasing the building time of the components.