In this research, I systematically investigated the effects of cavity vacuum levels and subsequent heat treatment on the porosity characteristics and mechanical performance of aluminum alloy engine block castings produced by high-pressure die casting. The presence of sand foundry defects, such as gas porosity and shrinkage cavities, is a perennial challenge in conventional casting processes. My work focuses on how vacuum assistance in die casting can mitigate these issues, and how different solution treatment durations alter the internal pore structure and tensile properties. The study is particularly relevant to the automotive industry, where lightweight components demand both high integrity and reliable post-casting processing capabilities.
1. Introduction
The demand for lighter vehicles with reduced fuel consumption has led to widespread adoption of aluminum alloy castings, especially for complex components like engine blocks and cylinder heads. High-pressure die casting (HPDC) offers excellent productivity and dimensional accuracy, but the inherently turbulent mold filling and rapid solidification often entrap gas, leading to porosity. These sand foundry defects, which occur not only in sand casting but also in die casting, severely limit the mechanical properties and preclude heat treatment. Vacuum-assisted high-pressure die casting (VHPDC) has emerged as a promising solution to reduce the amount of entrapped air by evacuating the die cavity before melt injection. By reducing the residual gas pressure, the formation of porosity can be significantly suppressed, allowing for subsequent heat treatment and improved mechanical performance.
However, even with vacuum assistance, some gas may remain, and the solidification shrinkage can still cause micro-porosity. Heat treatment of such castings is usually avoided because internal gas pores expand at elevated temperatures, leading to blistering and distortion. My research aims to bridge this gap by quantifying the influence of varying vacuum degrees (50 kPa, 20 kPa, and 10 kPa absolute pressure) on the pore population and tensile properties of an engine block, and then to explore the feasibility of short-time solution treatment followed by artificial aging. Throughout the study, I constantly compared the defect morphology with typical sand foundry defects reported in the literature, noting that the roundness and distribution of pores are critical for stress concentration and crack initiation.
I begin by describing the experimental materials and methods, followed by a detailed analysis of the pore structures under different vacuum levels. Subsequently, I present the mechanical test results and fracture surface observations. Finally, I report the effects of two different solution treatment times on the porosity and mechanical properties, and I propose an optimized heat treatment schedule.
2. Experimental Procedures
2.1 Material and Casting
The material studied was ADC12 aluminum alloy (AlSi9Cu3) with a nominal composition given in Table 1. The mold steel was H13, a common hot-work tool steel. The castings were produced on a Bühler cold-chamber die casting machine with a maximum clamping force of 27,000 kN. Three different absolute cavity pressures were employed: 50 kPa, 20 kPa, and 10 kPa. For each vacuum level, two engine block castings were produced. The key process parameters are listed in Table 2.
| Zn | Fe | Cu | Mn | Mg | Si | Ti | Al |
|---|---|---|---|---|---|---|---|
| 1.0 | 1.2 | 3.0 | 0.1-0.5 | 0.1-0.5 | 9 | 0.15 | Bal. |
| Shot sleeve length | 1200 mm |
| Plunger diameter | 140 mm |
| Fill ratio | 42.85% |
| Low plunger speed | 0.2 m/s |
| High plunger speed | 5.0 m/s |
| Switch-over position | 575 mm |
| Cavity fill time | 5.82 ms |
| Pouring temperature | 640 °C |
| Die temperature | 150 °C |
2.2 Specimen Preparation
Samples for metallographic observation (10×10×10 mm) were cut from the cylinder bore region of each casting using electrical discharge machining. Tensile specimens with a rectangular cross-section were prepared according to the Chinese national standard GBT 228.1-2010, with the dimensions shown in Figure (not reproduced here). For the heat treatment study, only the castings produced under a vacuum of 10 kPa were used because they exhibited the lowest initial porosity. Two heat treatment schedules were applied:
- T5a: 500 °C × 0.5 h (solution treatment) + water quench at 60 °C + 150 °C × 5 h (artificial aging)
- T5b: 500 °C × 2 h (solution treatment) + water quench at 60 °C + 150 °C × 5 h (artificial aging)
Heat treatment was carried out in a box-type furnace (MXQ14000-30) with temperature control of ±1 °C. After aging, all samples were air-cooled to room temperature.
2.3 Characterization Methods
Optical microscopy (OLYMPUS GX41) was used to observe the microstructure and pore morphology. To statistically quantify the pores, I captured 16 images at a magnification of 50× for each sample and stitched them together to form a complete view of the examined area. The number, size, and circularity of pores were analyzed using Nano Measurer 1.2 software. The density of castings was measured by the Archimedes method using equations (1) and (2).
$$\rho_p = \frac{m_1}{m_1 – m_2} \rho_w \tag{1}$$
$$P = \left(1 – \frac{\rho_p}{\rho_{wz}}\right) \times 100\% \tag{2}$$
where \(m_1\) is the mass in air, \(m_2\) is the mass in water, \(\rho_w\) is the density of water, and \(\rho_{wz}\) is the theoretical density of the alloy (2.76 g/cm³).
Scanning electron microscopy (SEM, TESCAN VEGA 3 LMH) equipped with energy-dispersive X-ray spectroscopy (EDS) was employed to examine fracture surfaces and identify the types of porosity. Tensile tests were performed on a universal testing machine at a crosshead speed of 1 mm/min. The ultimate tensile strength (UTS) and elongation (\(\delta\)) were recorded. To evaluate the stability of the mechanical properties, I calculated the standard deviation of the measured values using equation (3).
$$s = \sqrt{\frac{1}{n}\sum_{i=1}^{n}(x_i – \bar{x})^2} \tag{3}$$
3. Results and Discussion
3.1 Effect of Vacuum Level on Microstructure and Porosity
Figure 1 shows the optical micrographs of the cylinder bore samples produced under three different vacuum pressures. At 50 kPa, large irregular pores were observed, some exceeding 100 µm in diameter, with complex angular shapes. At 20 kPa, the pores became smaller and more rounded, distributed more uniformly. At 10 kPa, the microstructure appeared dense with only a few fine micropores. This trend clearly demonstrates that increasing the vacuum level (reducing absolute pressure) effectively suppresses the entrapment of gas. I noted that the general morphology of the primary \(\alpha\)-Al dendritic structure did not change significantly with vacuum pressure; however, the secondary dendrite arm spacing (SDAS) appeared to be finer at higher vacuum level. This is consistent with more efficient heat transfer and perhaps a more uniform solidification front.
The quantitative pore statistics are summarized in Table 3. The total number of pores detected in the inspected area decreased from 3298 at 50 kPa to 1832 at 10 kPa, a 44.4% reduction. The average pore size decreased from 8.65 µm to 5.61 µm, and the maximum pore diameter decreased from 142.3 µm to 111.9 µm. The pore size distribution shifted to smaller diameters; at 10 kPa, most pores were between 4 and 6 µm, whereas at 50 kPa, a significant fraction were larger than 8 µm. The porosity fraction calculated using Eq. (2) dropped from 4.8% to 2.8%, and the density correspondingly increased from 2.627 to 2.682 g/cm³. These results are in accordance with prior studies on VHPDC of magnesium and aluminum alloys. It is worth emphasizing that even at the highest vacuum tested, some residual pores remained; they were mostly spherical and likely originated from dissolved hydrogen, which is a common source of sand foundry defects in conventional castings.
| Absolute pressure (kPa) | Number of pores | Average pore size (µm) | Porosity (%) | Max diameter (µm) | Min diameter (µm) | Density (g/cm³) |
|---|---|---|---|---|---|---|
| 50 | 3298 | 8.65 | 4.8 | 142.3 | 2.24 | 2.627 |
| 20 | 2956 | 8.12 | 4.4 | 127.6 | 1.98 | 2.638 |
| 10 | 1832 | 5.61 | 2.8 | 111.9 | 1.44 | 2.682 |
Figure 2 presents SEM images of typical pores found in the samples. I observed three main categories: (1) spherical or elliptical pores with smooth bright inner walls, identified as gas pores; (2) irregular, dark, and rough shrinkage cavities; and (3) a combination of both, often termed gas-shrinkage pores. EDS analysis of the smooth pores occasionally detected oxygen, confirming that entrapped air contributed to their formation. In contrast, pure shrinkage cavities contained only the alloying elements (Al, Si, Cu). The gas-shrinkage pores showed a characteristic “spherical head with a tail” morphology, resulting from simultaneous gas expansion and shrinkage. These types of defects are analogous to sand foundry defects observed in sand-cast aluminum alloys, albeit with different pressure conditions. In vacuum die casting, the reduced initial gas pressure lowers the driving force for pore growth, leading to smaller and more spherical pores, which are less deleterious to mechanical properties.
3.2 Mechanical Properties under Different Vacuum Levels
Figure 3 shows the average ultimate tensile strength (UTS) and elongation as a function of cavity pressure. The UTS increased from 205 MPa at 50 kPa to 258 MPa at 10 kPa, an improvement of 25.8%. The elongation increased from 1.76% to 2.25%, a relative increase of 27.8%. The standard deviation of UTS decreased from ±25 MPa at 50 kPa to ±12.8 MPa at 10 kPa, indicating not only higher average strength but also better consistency. These improvements can be attributed to the reduction in porosity: fewer and smaller pores mean less reduction in effective load-bearing area and fewer stress concentration sites. Moreover, the change in pore morphology from irregular to nearly spherical at higher vacuum reduces the local stress intensity factor, thereby delaying crack initiation. It is well known that sharp, elongated pores act as pre-existing cracks, whereas round pores are less harmful. This observation parallels the behavior of sand foundry defects in sand castings, where the geometry of shrinkage porosity critically influences ductility.
To further analyze the scatter in tensile properties, I calculated the variance and standard deviation using Eq. (3). The results are presented in Figure 4. The reduction in standard deviation with increasing vacuum indicates that the casting process becomes more stable and repeatable. This is particularly important for mass production of safety-critical components such as engine blocks.
| Vacuum (kPa) | UTS (MPa) | Elongation (%) | UTS standard deviation (MPa) |
|---|---|---|---|
| 50 | 205 | 1.76 | 25.0 |
| 20 | 236 | 2.19 | 14.5 |
| 10 | 258 | 2.25 | 12.8 |
3.3 Fracture Surface Observations
All tensile specimens fractured in a brittle manner, with flat fracture surfaces perpendicular to the loading axis and no significant necking. Figure 5 shows the macroscopic and microscopic fracture morphologies of specimens from the three vacuum conditions. At 50 kPa, the fracture surface contained clusters of large and irregular pores (indicated by dashed circles), which likely served as crack initiation sites. The microscopic features exhibited cleavage facets, river patterns, and secondary cracks, typical of brittle cleavage fracture. At 20 kPa, the pores were smaller and more evenly distributed, and the fracture surface displayed a mixture of cleavage facets and a few tear ridges. At 10 kPa, the fracture was characterized by a quasi-cleavage mode with abundant tear ridges and some dimples, indicating a moderate increase in local plasticity despite the overall brittle appearance. The presence of dimples suggests that some micro-void coalescence occurred before final separation, which is consistent with the slightly higher elongation measured at 10 kPa.
These observations reinforce the concept that not only the total porosity but also the pore distribution and morphology govern the mechanical behavior. Irregular pores generate higher stress concentrations, and clustered pores create an effective weak path for crack propagation. In contrast, fine, spherical, well-distributed pores are much less detrimental. This insight is also relevant to the control of sand foundry defects in sand casting, where modifying the solidification conditions can refine pore morphology and improve properties.
4. Effects of Heat Treatment on Porosity and Properties
4.1 Pore Morphology Evolution after Solution Treatment
Because the castings produced at 10 kPa exhibited the best as-cast properties, they were selected for the heat treatment study. After subjecting samples to 500 °C for 0.5 h and 2 h (followed by aging), I observed significant changes in the pore population. Figure 6 compares the pore morphologies after the two heat treatments. The 0.5 h treated sample showed numerous round pores uniformly distributed, with a maximum pore size of about 120 µm. The 2 h treated sample, however, displayed larger and more elongated pores, some appearing as connected chains, with a maximum size of about 200 µm. This indicates that longer solution treatment promotes the coalescence of adjacent pores and the growth of pre-existing micro-shrinkage cavities. The mechanism involves the release of dissolved hydrogen at high temperature and the condensation of vacancies, which then migrate to existing pores and cause them to expand. This behavior is analogous to the formation of secondary porosity during solution treatment of sand castings—an issue that is well known in the context of sand foundry defects and must be carefully managed.
Quantitative porosity measurements are listed in Table 5. The density decreased from 2.682 g/cm³ (as-cast) to 2.64 g/cm³ after 0.5 h and to 2.60 g/cm³ after 2 h. Correspondingly, the porosity increased from 2.8% to 4.3% and 5.7%, respectively. The increase in porosity is substantial, confirming that new pores formed during heat treatment. The fact that the 2 h treatment resulted in a greater porosity increase (103% relative to as-cast) compared to the 0.5 h treatment (53%) suggests that pore nucleation and growth are time-dependent. The longer the exposure to high temperature, the greater the opportunity for hydrogen diffusion and pore expansion.
| Condition | Density (g/cm³) | Porosity (%) |
|---|---|---|
| As-cast (10 kPa) | 2.682 | 2.8 |
| 500 °C × 0.5 h + aging | 2.64 | 4.3 |
| 500 °C × 2 h + aging | 2.60 | 5.7 |
4.2 Mechanical Properties after Heat Treatment
Table 6 summarizes the tensile properties of the heat-treated samples. After 0.5 h solution treatment, the average UTS increased to 268 MPa, which is 3.8% higher than the as-cast value (258 MPa), while the elongation decreased slightly from 2.25% to 1.95%. The standard deviation of UTS improved significantly to ±8.6 MPa, indicating much better consistency. In contrast, after 2 h solution treatment, both UTS and elongation dropped significantly to 236 MPa and 1.36%, respectively, and the standard deviation increased to ±29.8 MPa, reflecting poor reproducibility. The 2 h treatment caused the pores to grow large and become irregular, leading to severe stress concentrations and premature failure. These results clearly show that short solution treatment can be beneficial, while prolonged treatment is detrimental.
| Condition | UTS (MPa) | Elongation (%) | UTS std dev (MPa) |
|---|---|---|---|
| As-cast (10 kPa) | 258 | 2.25 | 12.8 |
| 500 °C × 0.5 h + aging | 268 | 1.95 | 8.6 |
| 500 °C × 2 h + aging | 236 | 1.36 | 29.8 |
To better compare the overall mechanical performance, I used the quality index \(Q\) proposed by Drouzy et al.:
$$Q = \text{UTS (MPa)} + 150 \log(\delta) \tag{4}$$
where \(\delta\) is the elongation in percent. The calculated Q values are given in Table 7. The as-cast sample had a Q value of 310 MPa. The 0.5 h heat treatment gave a Q value of 311 MPa, essentially the same as the as-cast condition, while the 2 h treatment gave a Q of 256 MPa, significantly worse. Thus, the short solution treatment maintains the comprehensive mechanical performance while improving reproducibility, making it a viable option for vacuum die castings. The long treatment is clearly harmful, as it exacerbates the sand foundry defects-like porosity problem, leading to brittle fracture.
| Condition | Q (MPa) |
|---|---|
| As-cast | 310 |
| 500 °C × 0.5 h + aging | 311 |
| 500 °C × 2 h + aging | 256 |
4.3 Fracture Surface of Heat-Treated Specimens
The fracture surfaces of the heat-treated specimens are shown in Figure 7. The 0.5 h treated specimen exhibited a brittle cleavage fracture with a fairly uniform distribution of small pores; the largest pore on the fracture surface was about 40 µm. The 2 h treated specimen showed large, clustered pores (up to 100 µm) on the fracture surface, and the microscopic features were characteristic of intergranular fracture, with smooth “rock candy” facets. Intergranular fracture typically occurs when the grain boundaries are weakened by precipitation or by porosity networks. The long solution treatment allowed pores to grow and connect preferentially along grain boundaries, severely degrading the mechanical integrity. This is a classic manifestation of sand foundry defects in a different casting process, where the interaction between porosity and grain boundary embrittlement controls failure.
5. Discussion
The results of this study demonstrate that vacuum die casting is an effective method to control porosity, but the residual gas content and the inherent solidification shrinkage still limit the achievable quality. The pores that remain after vacuum casting are mostly small and round, which is beneficial for mechanical properties. However, heat treatment can change this beneficial morphology. During solution treatment at 500 °C, the solubility of hydrogen in the aluminum matrix decreases, and the supersaturated hydrogen atoms diffuse to existing micropores or form new nuclei. The internal gas pressure in these pores increases with temperature, driving pore expansion. If the pores are initially small and spherical, they can grow uniformly without severe coalescence, as seen in the 0.5 h treatment. However, if the treatment time is extended, the pores not only grow but also coalesce, especially along dendrite boundaries, leading to elongated and irregular cavities. This process is strikingly similar to the formation of sand foundry defects during the solidification of sand castings, where gas and shrinkage combine to create complex defect networks.
From a practical standpoint, my findings suggest that for vacuum die-cast ADC12 alloy, a short solution treatment (0.5 h) followed by aging can modestly improve the tensile strength while sacrificing some ductility, but the consistency of properties improves. This is beneficial for applications where consistent high strength is more important than elongation. The longer treatment (2 h) is not recommended because it significantly reduces both strength and ductility and increases scatter. The interaction between porosity and heat treatment must be carefully considered when designing post-casting processes.
To further illustrate the relationship between porosity and mechanical properties, I plotted the UTS as a function of porosity for all samples (not shown). A clear inverse correlation exists: as porosity increases, UTS decreases. The data points for the as-cast and 0.5 h treated samples fall on a different curve than the 2 h treated sample, indicating that not only the total porosity but also the pore morphology and distribution affect the strength. This is consistent with the behavior of sand foundry defects in aluminum sand castings, where the type and location of porosity are critical.
6. Conclusions
Based on the present study, I draw the following conclusions:
- Increasing the vacuum level (lowering the absolute pressure) from 50 kPa to 10 kPa in high-pressure die casting significantly reduces the number, size, and fraction of pores in the castings. The average pore size decreased from 8.65 µm to 5.61 µm and the porosity from 4.8% to 2.8%. The pores also became rounder and more uniformly distributed.
- The mechanical properties improved with increasing vacuum: the ultimate tensile strength increased from 205 MPa to 258 MPa and the elongation from 1.76% to 2.25%. The fracture mode changed from pure cleavage at 50 kPa to quasi-cleavage at 10 kPa, indicating a slight improvement in local plasticity.
- Heat treatment of the 10 kPa castings at 500 °C for 0.5 h followed by aging led to a small increase in UTS (to 268 MPa) with improved repeatability (standard deviation ±8.6 MPa), while the elongation slightly decreased. In contrast, a 2 h solution treatment caused severe pore growth and coalescence, resulting in a dramatic drop in UTS (236 MPa) and elongation (1.36%), and large performance scatter.
- The porosity increased to 4.3% after 0.5 h and to 5.7% after 2 h solution treatment, confirming the formation of new pores and the growth of existing ones during heat treatment. The longer the treatment, the more extensive the pore coarsening and coalescence.
- The quality index Q, defined as UTS + 150·log(elongation), remained essentially unchanged after the 0.5 h treatment (311 MPa vs 310 MPa as-cast), while the 2 h treatment lowered Q to 256 MPa. Therefore, a short solution treatment is a viable option for vacuum die-cast ADC12 components, while extended solution treatment should be avoided.
- Throughout this research, the parallels with sand foundry defects have been highlighted, underscoring that the fundamental principles of pore formation and growth are shared across different casting processes. Understanding these principles allows process engineers to optimize vacuum levels and heat treatment schedules to achieve the desired balance of properties.
In summary, my study provides quantitative evidence for the beneficial effects of vacuum assistance in die casting and establishes a safe heat treatment window for ADC12 castings. These findings can help automotive manufacturers produce lighter and stronger components without compromising reliability.