AlSi12

The AlSi12 alloy has been, in general, less studied than AlSi10Mg. Only a few works have compared the corrosion resistance of the additively manufactured specimens with that of the conventional cast alloy [62,63]. Yang et al. [63] reported a higher corrosion resistance for the AM material compared to the cast alloy. Their experimental data in NaCl solution showed not only a better electrochemical performance during open circuit potential and potentiodynamic polarization tests for the AM AlSi12 sample than for the cast alloy, but also a much lower weight loss during immersion in the same solution for the AM sample compared to the cast material [63]. On the other hand, other researchers reported that the cast and the AM AlSi12 alloy display similar corrosion behaviour in the HNO3 solution [62]. Weight loss measurements after immersion in 0.1 M HNO3 showed an almost perfect overlap between the curves of the cast and additively manufactured AlSi12 alloy [62].

#### *3.2. E*ff*ect of Heat Treatment on Microstructure and Corrosion*

As mentioned above, the unique processing conditions during MAM cause rapid solidification of melt pools due to the extremely high cooling rates reached (~10<sup>5</sup> K/s) [39,58]. This leads to a very fine cellular microstructure far from equilibrium, consisting of a primary Al-rich phase [39,64] supersaturated in Si (up to 11 wt.%) [52] with residual Si segregated at the cellular boundaries [39]. The extremely fine Si network gives a large total interfacial energy, resulting in a high driving force for Si coarsening [38]. As thermal gradients and grain growth rates decrease towards the melt pool borders, coarser cellular structures can be found there [36]. In addition, the fast cooling rates can introduce residual thermal stresses that can lead to dimensional inaccuracy and distortion [55]. A variety of heat treatments can be performed to act on these factors. Due to the presence of Mg in most Al-Si alloys, artificial aging (150–180 ◦C) [65] can be performed to induce age hardening through Mg2Si precipitation [34,52,55]. The increase in the solid solubility of Si in Al due to the rapid solidification can enhance the effect of solution hardening and strengthening [66]. Partial annealing treatments can be performed to release residual stress [50,52–55,58,67,68]. Alternatively, the building platform can also be heated during manufacturing [34,52].

Fiocchi et al. [69] performed Differential Scanning Calorimetry (DSC) experiments to determine the phase transformations occurring in an AlSi10Mg alloy prepared by selective laser melting. Several heating rates (2, 5, 10, 20, and 30 ◦C/min) were applied for a temperature range of 0 to 500 ◦C. Two exothermic transformations were identified for the SLM material. The first peak was found to be common to SLM material, unmolten powder, and cast alloy of similar composition. It was determined to result from the alloying elements and was identified as Mg2Si precipitation. The second peak was found to be unique to the SLM material. It was ascribed to the rupture and spheroidization of the Si network, which is characteristic of SLM parts. The isothermal endpoint temperatures of these transformations were calculated using second-degree polynomial regressions, resulting in 263 ◦C for Mg2Si precipitation and 294 ◦C for Si rupture and spheroidization [69]. Similar DSC experiments on SLM AlSi10Mg alloy were performed by Rafieazad et al. [55] for a temperature range of 0 to 550 ◦C. They determined the peak values for the two exothermal transformations at several heating rates (2, 5, 10, and 20 ◦C/min) and calculated the isothermal temperatures using second-degree polynomial regressions. These were determined to be 232.9 and 273.2 ◦C. Similar to the work of Fiocchi et al. [69], these were interpreted as being respectively the Mg2Si β phase precipitation and the Si phase precipitation via solid-state diffusion [55].

#### 3.2.1. Effect on Microstructure

#### Stress Release Heat Treatments

Several studies have focused on the effect of stress-relieving heat treatments on the microstructure. Cabrini et al. [52] studied the effect of stress-release at 200, 300, 400, and 500 ◦C for 2 h. Similarly, Rubben et al. [54] investigated the effect of stress-release at 250 and 300 ◦C for 2 h, while Rafieazad et al. [55] investigated heat treatments at 200, 300, and 350 ◦C for 3 h. Other papers discuss annealing with similar conditions [68] but without the expressed aim of stress release.

Cabrini et al. [52] observed that stress release heat treatments up to 300 ◦C for 2 h do not cause a significant change in the characteristic melt pool macrostructure. It should be noted however that although melt pool tracks are retained, significant microstructure alteration can still occur. For the heat treatment of 200 ◦C for 3 h, no microstructure change was noticed [55]. This is attributed to the temperature being significantly lower than the minimum required for Si interdiffusion in Al [55]. After a stress release at 250 ◦C for 2 h, a limited breakup of the Si network was observed [54] (see Figure 8b). The limited nature of this breakup can be explained by considering that the temperature of stress release was below the peak temperature value of Si network coarsening [55], resulting in slow kinetics. More significant changes occurred after stress release at 300 ◦C for 2 h. The morphology of the Si phase changed completely, going from a continuous silicon network towards separate Si precipitates

(see Figure 8c) [52,54,68]. Despite this, larger microstructures were still observed at MPBs. As a result, the melt pool macrostructure was retained. After stress release at 400 ◦C for 2 h, the MPBs became harder to distinguish [52]. Rafieazad et al. [55] reported a similar diminishing of MPBs after 3 h at 350 ◦C. In both cases, an α-Al matrix with rounded silicon particles was revealed at higher magnification [52,55,68]. When the temperature was increased to 500 ◦C, the characteristic melt pool macrostructure completely disappeared. Silicon particle coalescence occurred, leading to much larger silicon particles [52].

**Figure 8.** Secondary electron image showing the microstructure of (**a**) an as-built AM Al-Si sample (AlSi10Mg), as well as two stress-released specimens: one at 250 ◦C for 2 h (**b**), and another at 300 ◦C for 2 h (**c**). MPB refers to the melt pool border, while HAZ stands for the heat-affected zone. (Adapted from reference [54]).

#### Heating of Building Platform

Heating the building platform during manufacturing can be used as an alternative to the stress release heat treatment. Residual stresses can be reduced by increasing the building platform temperature without affecting the mechanical strength [52]. Cabrini et al. [52] studied the effect of two different building platform temperatures, 35 and 100 ◦C. The temperature change did not cause a noticeable change in the macrostructure as both show similar melt pools and heat-affected zones. At higher magnification, some small changes were noticed in the melt pools. Limited formation of the eutectic-like phase by silicon particles was more visible for specimens built at 100 ◦C. The small changes in microstructure explain the lack of change in mechanical strength. A higher building temperature of 300 ◦C was investigated by Brandl et al. [34]. No significant change in macrostructure was obtained. It should be noted that the used magnification was not high enough to be able to observe variations in the cellular network.
