In Al-Mg alloys with Mg content 0 ≤ X_Mg ≤ 6 wt. %, the effects of X_Mg on dissolved hydrogen content ([H]) and melt quality were investigated. [H] was measured using the Closed-Loop Recirculation method, and the melt quality was quantified using the density index (DI), bifilm index (BI), and porosity measurement. [H] in the molten alloys increased with increasing X_Mg and melt temperature T_Melt; these trends agree with theoretical calculations for hydrogen solubility. The tendency of melt quality vs X_Mg was similar in DI, BI, and porosity measurements, and the poorest melt quality was observed in the Al-4Mg alloy that had X_Mg = 4 wt. %, whereas the highest [H] was obtained in the Al-6Mg alloy melt that had X_Mg = 6 wt. % Mg. During thermogravimetric/differential thermal analysis, rapid oxidation occurred in the Al-4Mg alloy melt during the holding time between 45 and 60 min at 800 °C, i.e., just before the molten metal was cast. The inferior melt quality of Al-4Mg alloy may have been caused by high-temperature oxidation.

The present study investigated the effect of the casting conditions for the reduced pressure test (RPT) on the melt quality of Al-Si alloy. The casting conditions considered in RPT were the atmospheric exposure during melting, sampling method, and mold pre-heating temperature. Density Index (DI) was measured to quantify the melt quality of the Al-Si alloy casted with the different casting conditions for the RPT. The sample with blocking atmospheric exposure during melting was 5.6% lower in DI than the one without blocking. DI showed a 1.9% gap between scooping-out and pouring sampling methods. Increasing mold pre-heating temperature from 100 °C to 250 °C increased the DI of the alloy from 8.5% to 18.7%. On the other hand, when the mold pre-heating temperature was 350 °C, the DI of the alloy dropped sharply to 0.9%. The melt quality of the alloys was analyzed by measuring the pores and microstructure and simulating the solidification of the samples. It was presumed that the oxides and inclusions in the molten alloys caused the difference in DI according to the atmospheric exposure and the sampling method. The difference in DI according to the mold pre-heating temperature could be understood by calculating the solidification starting time and hydrogen diffusion coefficient during the solidification of the alloys in RPT.

With the rise in the demand for eco-friendly and electric vehicles, welding and heat treatment are becoming very important to meet the necessary weight reduction, complexity, and high functionality of die castings. Pore-free (PF) die casting is an effective process that enables heat treatment and welding due to low gas porosities. Indeed, this process affords castings of low gas porosity, similar to those attained by high-vacuum die casting. In this study, we compared the gas porosities of different castings fabricated by PF die casting using varied injected oxygen amounts. The castings were all subjected to T6 heat treatment and analyzed by computed tomography (CT) to compare their microstructure and mechanical properties before and after T6 heat treatment. The results revealed that with the increasing injected oxygen amount, the gas porosity of the specimens decreased while their mechanical properties increased. In particular, the gas porosity was the lowest at 1.26 L. Moreover, the 1.26 L specimen displayed the best tensile strength, yield strength, and elongation results. Finally, Weibull distribution analysis revealed that the tensile strength and elongation repeatability and reproducibility increased with increasing injected oxygen amount.

In this study, the optimal conditions of gas bubbling filtration (GBF) treatment for securing highly-clean molten Al-Si-Mg-Cu alloy were identified. The effects of GBF treatment time and stabilization time on the degree of molten metal cleanliness were examined by measuring melt quality parameters such as density index, bifilm index, porosity, and the amount of dissolved hydrogen [H]. A high melt quality was achieved when GBF treatment was performed on 10 kg melt for more than 10 min (i.e., 1 L gas/kg melt). However, as the stabilization holding time after GBF treatment increased to 10, 20, and 30 min, the melt quality degraded. GBF treatment for 30 min had a similar effect to treatment for 10 min, and the degree of deterioration of melt quality during the stabilization time was also similar. Considering the economics, 10 min GBF treatment and short holding time are required. Observations of the shape and volume of the largest pore suggested the cause of defect formation and confirmed that the volume of the largest pore can be used as an index of the melt quality.

A laser power bed fusion (L-PBF) manufacturing process was optimized by analyzing the surface morphology and track width w of single scan tracks (SSTs) on Fe-3.4wt.%Si. An SST was evaluated under process conditions of laser power P, scan speed V, and energy density E = P/V. The SST surface shape was mainly affected by E; desirable thin and regular tracks were obtained at E = 0.3 and 0.4 J/mm. An L-PBF process window was proposed considering the optimal w of SST, and the appropriate range of E for the alloy was identified to be 0.24 J/mm to 0.49 J/mm. w showed a strong relationship with E and V, and an analytic model was suggested. To verify the process window derived from the appropriate w of SST, cubic samples were manufactured with the estimated optimal process conditions. Most samples produced had a high density with a porosity of <1%, and the process window derived from SST w data had high reliability. This study presents a comprehensive approach to enhancing additive manufacturing for Fe-3.4Si alloy, offering valuable insights for achieving high-quality samples without the need for time-intensive procedures.

In this study, regression-based machine learning models were developed to predict the melt pool width and depth formed during the Laser Powder Bed Fusion (LPBF) process for Fe-3.4Si and Fe-6Si alloys. Based on experimentally obtained melt pool width and depth data, a total of 11 regression models were trained and evaluated, and hyperparameters were optimized via Bayesian optimization. Key process parameters were identified through data preprocessing and feature engineering, and SHAP analysis confirmed that the input energy had the strongest influence on both melt pool width and depth. The comparison of prediction performance revealed that the support vector regressor with a linear kernel (SVR_lin) exhibited the best performance for predicting melt pool width, while the multilayer perceptron (MLP) model achieved the best results for predicting melt pool depth. Based on these trained models, a power–velocity (P-V) process map was constructed, incorporating boundary conditions such as the overlap ratio and the melt pool morphology. The optimal input energy range was derived as 0.45 to 0.60 J/mm, ensuring stable melt pool formation. Specimens manufactured under the derived conditions were analyzed using 3D X-ray CT, revealing porosity levels ranging from 0.29% to 2.89%. In particular, the lowest porosity was observed under conduction mode conditions when the melt pool depth was approximately 1.0 to 1.5 times the layer thickness. Conversely, porosity tended to increase in the transition mode and lack of fusion regions, consistent with the model predictions. Therefore, this study demonstrated that a machine learning-based regression model can reliably predict melt pool characteristics in the LPBF process of Fe-Si alloys, contributing to the development of process maps and optimization strategies.