Heat Treatments and Critical Quenching Rates in Additively Manufactured Al–Si–Mg Alloys

Leonhard Hitzler,Andreas Öchsner,Helmut Clemens,Ewald Werner,Stephan Hafenstein,Enes Sert,Markus Merkel,Francisca Mendez Martin

doi:10.3390/ma13030720

Leonhard Hitzler, Andreas Öchsner + Show 6 more

Open Access

https://doi.org/10.3390/ma13030720

Copy DOI

Abstract

Laser powder-bed fusion (LPBF) has significantly gained in importance and has become one of the major fabrication techniques within metal additive manufacturing. The fast cooling rates achieved in LPBF due to a relatively small melt pool on a much larger component or substrate, acting as heat sink, result in fine-grained microstructures and high oversaturation of alloying elements in the α-aluminum. Al–Si–Mg alloys thus can be effectively precipitation hardened. Moreover, the solidified material undergoes an intrinsic heat treatment, whilst the layers above are irradiated and the elevated temperature in the built chamber starts the clustering process of alloying elements directly after a scan track is fabricated. These silicon–magnesium clusters were observed with atom probe tomography in as-built samples. Similar beneficial clustering behavior at higher temperatures is known from the direct-aging approach in cast samples, whereby the artificial aging is performed immediately after solution annealing and quenching. Transferring this approach to LPBF samples as a possible post-heat treatment revealed that even after direct aging, the outstanding hardness of the as-built condition could, at best, be met, but for most instances it was significantly lower. Our investigations showed that LPBF Al–Si–Mg exhibited a high dependency on the quenching rate, which is significantly more pronounced than in cast reference samples, requiring two to three times higher quenching rate after solution annealing to yield similar hardness results. This suggests that due to the finer microstructure and the shorter diffusion path in Al–Si–Mg fabricated by LPBF, it is more challenging to achieve a metastable oversaturation necessary for precipitation hardening. This may be especially problematic in larger components.

Highlights

Various additive manufacturing methods for direct metal fabrication have emerged in the last decade, which can be used to fabricate directly deployable components without the necessity of a post-densification process [1,2]
In comparison to the cast samples, an increased silicon content was noted for the Laser powder-bed fusion (LPBF) samples; see Table 4
In comparison to the immediately aged samples, a dwell time of 15 min at room temperature resulted in a substantial decrease in hardness (8–15 HBW) for LPBF samples, whilst cast samples decreased by 4–8 HBW (Figure 4b). These deviations vanished after longer dwell times; samples aged after a 7 d dwell time at room temperature showed consistent hardness and a similar hardness reduction compared to the immediately aged condition. These results show that immediate aging is applicable and beneficial on LPBF samples

Summary

Introduction

Various additive manufacturing methods for direct metal fabrication have emerged in the last decade, which can be used to fabricate directly deployable components without the necessity of a post-densification process [1,2]. Some of the most prominent representatives are the powder-bed fusion techniques, whereby a laser or electron beam is utilized to repetitively melt sections in a powder layer, resembling the sliced approximation of the component to be manufactured, and gradually, in a layer by layer approach, fabricate the entire component. Al–Si–Mg alloys exhibit a unique microstructure in their as-fabricated state, whereby the macroscopic anisotropy characteristics are predominantly governed by the localized formation of silicon segregations; see Figure 1 [3,4]. This is in strong contrast to other alloys, like austenitic stainless steels or Inconel, with their dominating characteristic being their grain morphology [5,6]. The cause for the major impact of the Si-segregations is thought to be their location, as they occur most prevalently in remolten areas, i.e., in the bonding areas between single scan tracks and subsequent layers [7]

Methods

Discussion

Conclusion