The quest for advanced materials enabling weight reduction and performance enhancement in aerospace components, particularly within hot sections of aero-engines, is perpetual. Among the leading candidates, Titanium Aluminide (TiAl) alloys stand out due to their exceptional combination of high specific strength, excellent elevated-temperature mechanical properties, oxidation resistance, and burn resistance. To fully leverage these attributes in complex geometries such as diffusers, pre-swirl nozzles, and exhaust casings, precision investment casting has emerged as the principal near-net-shape manufacturing route. This process is uniquely capable of producing intricate, thin-walled, monolithic components, which are otherwise challenging or impossible to fabricate through conventional methods. However, the successful precision investment casting of TiAl alloys is fraught with significant challenges, primarily stemming from their inherent material characteristics: poor melt fluidity, a narrow solidification range, high linear shrinkage, and virtually zero room-temperature ductility in the as-cast state. These factors collectively elevate the risks of casting defects like misruns, cold shuts, hot tears, shrinkage porosity, and cracking, especially in delicate annular structures with thin connecting struts.
This investigation focuses on the precision investment casting of a representative double-ring thin-wall strut structure, a critical element in aero-engine architectures. The component features interconnected inner and outer rings linked by multiple thin struts, presenting a formidable test for casting integrity. The core objective is to systematically evaluate and optimize the casting process through integrated numerical simulation and experimental validation. Specifically, the influence of two pivotal factors is examined: the casting method (gravity pouring versus centrifugal casting) and the design of the gating system, particularly the positioning of the bottom gates relative to the struts. ProCAST, a sophisticated finite element analysis software dedicated to casting processes, serves as the primary tool for simulating filling patterns, solidification sequences, thermal stress development, and the prediction of shrinkage-related defects. The insights gleaned from these simulations are then rigorously compared against the outcomes of actual casting trials, thereby establishing a validated framework for process design in precision investment casting of complex TiAl alloy components.
1. Methodology and Simulation Setup
The target component is a double-ring casing with an approximate diameter of 460 mm and a height of 70 mm. Its defining feature is a thin-walled structure, with wall thicknesses ranging from 1 to 3 mm, connecting an inner and an outer ring. To mitigate the high risk of cracking induced by the substantial hindered contraction of TiAl alloy, the original design with four struts was modified to a configuration with eight struts, distributing thermal stresses more evenly. A bottom-gated filling system was adopted to ensure a tranquil and controlled melt entry into the mold cavity, which is crucial for minimizing turbulence and gas entrapment in precision investment casting.
Two distinct gating system layouts were designed for comparative analysis, as illustrated schematically below. In both designs, eight ingates are distributed around the lower flanges of both the inner and outer rings.
- Scheme 1 (Gate Below Strut): The bottom gates are positioned directly beneath the eight connecting struts.
- Scheme 2 (Gate Between Struts): The bottom gates are positioned at the midpoint between two adjacent struts.
The three-dimensional models of the casting and the two gating systems were imported into ProCAST for meshing and preprocessing. Accurate simulation requires precise thermophysical property data for both the TiAl alloy (assumed composition akin to Ti-47Al-2Cr-2Nb) and the ceramic mold (Y2O3-based shell). These parameters, detailed in Table 1, govern the heat transfer, fluid flow, and solidification behavior during the precision investment casting process.
| Material | Density (kg/m³) | Thermal Conductivity (W/m·K) | Specific Heat (kJ/kg·K) | Latent Heat (kJ/kg) | Solidification Range (°C) | Viscosity (Pa·s) | Interface Heat Transfer Coefficient (W/m²·K) |
|---|---|---|---|---|---|---|---|
| TiAl Alloy | 3872 | 15 – 28 | 0.6 – 0.8 | 400 | 1468 – 1523 | (4.2 – 5.4) × 10⁻³ | 1500 |
| Y2O3 Shell | 4200 | 2.1 – 2.4 | 0.7 – 1.0 | – | – | – | – |
Simulations were conducted for both gravity casting and centrifugal casting scenarios using the two gating schemes. For all cases, constant process parameters were set: a pouring temperature of 1600°C for the TiAl melt and a mold preheat temperature of 800°C. In the centrifugal casting simulations, a rotational speed of 300 rpm was applied. Given the high susceptibility of thin-walled TiAl castings to hot tearing, a dedicated hot-cracking indicator module within ProCAST was employed to assess the risk. The simulation workflow analyzed the sequential stages of mold filling, solidification, cooling, and the resultant stress state within both the casting and the shell.
2. Numerical Simulation Results and Analysis
2.1 Internal Stress and Hot Tearing Tendency
The evolution of thermal stress during solidification and cooling is a critical concern in precision investment casting of TiAl alloys. The simulation of effective stress (von Mises stress) revealed distinct patterns for the two gating schemes. For Scheme 2 (gate between struts), areas of high stress concentration (exceeding 330 MPa) were located at the center of the struts and at the fillet regions connecting the gates to the outer ring flange. For Scheme 1 (gate below strut), high stress zones were found at the root where the struts join the outer ring and at the gate/ring junctions. The hot tearing tendency analysis corroborated these findings, indicating a higher risk at the gate-fillet areas in Scheme 2 and at the gate-root areas in Scheme 1.
Furthermore, the stress state of the casting after shell removal was simulated. A significant finding was that the maximum residual stress in the casting for Scheme 1 was substantially lower (approximately 80 MPa) compared to Scheme 2 (over 330 MPa). This suggests that positioning the gate directly below the strut (Scheme 1) leads to a more favorable stress distribution, effectively reducing the risk of stress-induced cracking post-casting, a vital consideration for the brittle TiAl alloy.
2.2 Melt Filling Characteristics
The filling patterns, visualized through temperature field progression, were starkly different between gravity and centrifugal casting, and were also influenced by the gating scheme.
Gravity Casting Simulation: The melt filled the mold cavity in a bottom-up fashion. For both schemes, the inner and outer ring flanges were filled first from their respective bottom gates. The melt then progressed upward through the rings and into the connecting struts. A key difference emerged: In Scheme 1, the melt front in the outer ring was uneven, being higher near the struts and lower in the mid-span between gates, creating a risk of misrun. In Scheme 2, the melt front in the outer ring was more uniform. However, in Scheme 1, the faster meeting of melt streams from the inner and outer rings within the struts increased the risk of cold shuts due to confluence and possible gas entrapment.
Centrifugal Casting Simulation: The filling dynamics were governed by centrifugal force. The melt initially filled the sprue and runner away from the central axis, then filled the outer ring from its gates in an inward direction (toward the axis). Subsequently, the struts were filled by melt flowing backward from the outer ring, while the inner ring was filled from its gates. For Scheme 1, significant gas entrapment was predicted in the outer ring at the mid-span between two gates. For Scheme 2, the confluence point of melt in the outer ring coincided with the strut location, allowing trapped gas to escape through the strut, thereby markedly reducing the risk of cold shuts in the outer ring. A comparative summary is provided in Table 2.
| Casting Method | Gating Scheme | Primary Filling Direction | Major Filling Challenge / Risk | Key Advantage |
|---|---|---|---|---|
| Gravity | 1 (Below Strut) | Bottom-up | Misrun in outer ring mid-span; Cold shuts in struts. | Direct feeding to strut root. |
| 2 (Between Struts) | Bottom-up | Reduced misrun risk; Potential gate-root hot spot. | More uniform outer ring fill. | |
| Centrifugal | 1 (Below Strut) | Outer-in, Axial | Gas entrapment in outer ring mid-span. | Enhanced feeding pressure. |
| 2 (Between Struts) | Outer-in, Axial | Reduced gas entrapment in outer ring. | Excellent outer ring fill; Gas escape via struts. |
2.3 Prediction of Shrinkage Porosity Distribution
The Niyama criterion and shrinkage porosity modules in ProCAST were used to predict the location and severity of solidification shrinkage defects. The results followed a consistent logic related to thermal mass and feeding paths.
Gravity Casting Defect Prediction: For both gating schemes, defects were predicted in three main categories:
1. Small, dispersed microporosity within the strut bodies and on the surfaces of the inner and outer rings. This is typical of thin sections that solidify rapidly without adequate feeding.
2. Larger, isolated shrinkage pores at thermal hotspots: specifically, at the junctions where the gates meet the ring flanges and at the roots of the struts where they connect to the rings.
Comparing the two schemes under gravity, Scheme 2 showed fewer dispersed defects on the rings but larger defects at the strut roots. Scheme 1 provided better feeding to the strut roots due to the direct gate placement but had more ring defects.
Centrifugal Casting Defect Prediction: The defect distribution pattern was similar, but the size and severity of predicted defects were significantly reduced across the board due to the enhanced feeding pressure provided by centrifugal force. The pressure head $P_c$ in centrifugal casting can be expressed as:
$$ P_c = \rho \omega^2 r \Delta h $$
where $\rho$ is melt density, $\omega$ is angular velocity, $r$ is radius, and $\Delta h$ is the height difference in the feeding path. This pressure actively drives melt into incipient shrinkage cavities. Notably, the large shrinkage defects predicted at the gate-ring junctions in gravity casting were either eliminated or did not penetrate into the casting body in the centrifugal simulation. However, the defect size at the strut roots, especially for Scheme 2, remained relatively larger because the centrifugal feeding direction is radial/axial, not directly upward into the strut from the bottom gate.
3. Experimental Validation and Discussion
To validate the simulation findings, actual precision investment casting trials were conducted using Y2O3 shells and a TiAl alloy melt. Scheme 1 (gate below strut) was selected for production trials using both gravity and centrifugal casting methods.
3.1 Casting Integrity and Shape Filling
The experimental results confirmed the filling predictions with remarkable accuracy. The gravity-cast component exhibited a clear misrun defect on the outer ring, precisely in the mid-span region between two struts, matching the simulation. In contrast, the centrifugal-cast component was completely filled, with no misruns or cold shuts, demonstrating the superior filling capability of centrifugal casting for thin-walled TiAl structures. This validates the critical role of enhanced metal velocity and pressure in overcoming the poor fluidity of TiAl alloys during precision investment casting.
3.2 Internal Defect Analysis
X-ray radiography was performed on the cast components to inspect internal defects. The results strongly correlated with the simulation predictions regarding shrinkage and also highlighted an issue with gas porosity.
- Gravity Casting: The component showed a high population of gas pores in addition to shrinkage defects. Shrinkage porosity was concentrated at the predicted thermal hotspots: the strut-to-outer-ring junctions and the gate-to-inner-ring junctions. Dispersed micro-porosity was also visible on the rings and struts.
- Centrifugal Casting: The gas porosity content was drastically reduced. The shrinkage defects were significantly smaller and less severe. The large shrinkage cavities predicted for gravity casting at hotspots were either absent or minimal. This confirms the dual benefit of centrifugal casting: not only does the centrifugal pressure improve feeding and reduce shrinkage, but it also aids in the flotation and removal of entrained gas bubbles from the melt due to the increased buoyancy force in the rotating field.
The experimental defect distribution maps aligned well with the ProCAST predictions, affirming the software’s reliability as a tool for optimizing precision investment casting processes for complex TiAl components. The centrifugal process unequivocally produced a casting with far superior internal metallurgical quality.
4. Conclusion
This integrated study combining ProCAST numerical simulation and experimental validation provides definitive guidelines for the precision investment casting of complex, thin-walled TiAl alloy structures. The key conclusions are as follows:
- Casting Method is Paramount: Centrifugal casting is essential for achieving complete fill and high integrity in thin-walled TiAl components. Gravity casting poses a high risk of misruns and yields castings with substantially higher levels of both shrinkage and gas porosity.
- Gating System Design Influences Defect Distribution: The position of bottom gates relative to structural features like struts dictates the location of thermal hotspots and feeding efficiency.
- Gates located between struts promote more uniform filling of annular sections but create larger thermal masses at strut roots, leading to potentially larger shrinkage there.
- Gates located directly below struts provide more direct feeding to the strut root, potentially reducing shrinkage in that critical area, but may lead to less uniform ring filling and more dispersed micro-porosity.
Under centrifugal conditions, Scheme 2 (gate between struts) offers an additional advantage by providing a venting path for trapped gas through the struts.
- Stress and Cracking Risk: The gating scheme also affects residual stress. Placing gates below struts (Scheme 1) resulted in a simulated casting with significantly lower post-shell-removal residual stress compared to Scheme 2, reducing the risk of stress-related failure in the brittle TiAl casting.
- Simulation Accuracy: ProCAST simulations demonstrated high fidelity in predicting filling patterns, defect locations (shrinkage), and high-stress zones. This makes it an invaluable tool for de-risking and optimizing the precision investment casting process for challenging alloys like TiAl before committing to expensive production trials.
For the production of the double-ring thin-wall strut component, the optimal process identified is centrifugal precision investment casting with a carefully designed gating system. While Scheme 1 showed benefits for stress, a hybrid or optimized design that balances feeding to strut roots (like Scheme 1) with uniform filling and gas venting of the rings (benefits of Scheme 2) under centrifugal force would likely yield the best overall result. This research establishes a validated methodology for tackling the formidable challenges associated with the precision investment casting of advanced TiAl intermetallic components for aerospace applications.