In the field of aerospace engineering, the demand for lightweight, high-strength, and high-temperature-resistant components has driven the development of advanced manufacturing techniques. We focus on the application of lost wax investment casting for producing large, complex thin-walled structures, specifically using ZTi55A titanium alloy. This high-temperature titanium alloy, with its nominal composition of Ti-5.5Al-3.5Sn-3.0Zr-0.7Mo-0.3Si-0.4Nb-0.4Ta, is designed for service temperatures above 550°C, but its multi-component nature introduces challenges such as poor fluidity and high cracking susceptibility during casting. Our study systematically investigates the entire process chain of lost wax investment casting, including wax pattern fabrication, shell building, melting and pouring, and post-treatment methods like pickling and repair welding. By optimizing these steps, we aim to achieve defect-free castings with precise dimensional accuracy and enhanced mechanical properties, contributing to the advancement of titanium alloy casting technology for critical aerospace applications.
The lost wax investment casting process, also known as precision casting, is particularly suited for intricate geometries due to its ability to replicate fine details. However, for large-scale components like the skeleton structure studied here, which measures approximately 1600 mm × 800 mm with wall thicknesses under 3 mm, traditional casting methods often lead to defects such as cracks, shrinkage porosity, and contamination layers. We address these issues through a combination of segmented casting and welding approaches, followed by an integrated casting strategy. This paper details our experimental methodology, results, and discussions, emphasizing the role of process parameters in controlling microstructure and stress distribution. Furthermore, we incorporate mathematical models and empirical data to support our findings, ensuring a comprehensive understanding of the lost wax investment casting process for ZTi55A alloy.
Materials and Experimental Methods
We selected ZTi55A titanium alloy as the base material due to its excellent high-temperature performance, which includes creep resistance and oxidation stability up to 550°C. The alloy’s composition was verified through spectroscopic analysis, and it conforms to the standards for aerospace-grade titanium alloys. The large skeleton castings were designed with non-rotational, semi-open structures, featuring variable wall thicknesses and internal cavities that necessitate precise control during lost wax investment casting. Our experimental approach involved two phases: initially, we produced segmented castings (front, middle, and rear sections) to mitigate cracking risks, and subsequently, we developed an integrated casting method for the entire skeleton. This dual strategy allowed us to compare the effectiveness of different gating systems and post-treatment techniques.
For the lost wax investment casting process, we began with wax pattern fabrication using laser rapid prototyping equipment. This method enabled the creation of complex wax models without traditional molds, reducing lead time and allowing for iterative design modifications. The wax patterns were assembled with specialized fixtures to prevent deformation, and dimensional accuracy was ensured through laser 3D scanning, achieving tolerances of ±0.3 mm. The shell-building process involved applying a series of ceramic layers using a robotic system. We used a composite refractory material ZM (a blend of Y2O3 and ZrO2) and a novel binder ZYM in a 5:1 ratio to form the slurry. The coating sequence included one transition layer with 200-mesh aluminum-zirconia powder and eight reinforcement layers with 200-mesh aluminum chamotte powder, followed by a final slurry coat without stuccoing. After drying under controlled conditions (22±2°C, 40-60% humidity), the shells were dewaxed and fired according to a specific thermal profile to achieve adequate strength and permeability.
The melting and pouring stages were critical for minimizing defects. We employed a bottom-gating system to ensure smooth filling, reduce turbulence, and promote directional solidification. The shells were preheated to 400-450°C before pouring to lower thermal gradients and stress. Melting was conducted in a vacuum induction furnace with currents ranging from 45,000 to 50,000 A to achieve sufficient superheat, improving metal fluidity. After pouring, the castings were cooled slowly with insulation to prevent cracking. Post-casting, we implemented mechanical cleaning (e.g., sandblasting and grinding) followed by chemical pickling using a solution of HF:HNO3:H2O in a 1:3:6 ratio to remove the α-case contamination layer. Repair welding was performed on simulated defects to develop a reliable procedure, involving pre-heating, controlled welding parameters, and stress relief treatments.
To quantify the process effectiveness, we used various characterization techniques, including optical microscopy for microstructure analysis, ultrasonic testing for defect detection, and mechanical testing to evaluate properties. The chemical composition of castings was verified against standards, as summarized in Table 1. Our experimental design incorporated statistical analysis to optimize parameters, such as the relationship between shell thickness and cracking tendency, which can be modeled using stress-strain equations. For instance, the thermal stress during cooling can be expressed as:
$$\sigma = E \cdot \alpha \cdot \Delta T$$
where $\sigma$ is the thermal stress, $E$ is the Young’s modulus, $\alpha$ is the coefficient of thermal expansion, and $\Delta T$ is the temperature difference. By controlling these variables, we aimed to reduce residual stresses in the final castings.
| Element | Standard Range | Measured Value |
|---|---|---|
| Al | 5.2-5.8 | 5.6 |
| Sn | 3.0-4.0 | 3.6 |
| Zr | 2.5-3.5 | 2.9 |
| Mo | 0.2-1.0 | 0.7 |
| Si | 0.1-0.5 | 0.3 |
| Nb | 0.2-0.7 | 0.4 |
| Ta | 0.2-0.7 | 0.4 |
| O | ≤0.20 | 0.11 |
| N | ≤0.05 | 0.01 |
| H | ≤0.012 | 0.004 |
| C | ≤0.10 | 0.01 |
Results and Discussion
Segmented Casting and Welding Approach
In the initial phase, we divided the large skeleton into three segments—front, middle, and rear—to address the high cracking propensity of ZTi55A alloy. This segmentation allowed for better control over solidification and stress distribution. The wax patterns for each segment were fabricated using laser rapid prototyping, and we designed specialized gating systems with additional risers at hot spots to minimize shrinkage defects. For example, the front segment featured a bottom-gating system that facilitated sequential filling, as illustrated in the experimental setup. After casting, the segments underwent non-destructive testing, which revealed full filling and no visible cracks or surface defects. The chemical analysis confirmed that the compositions met the required standards, as shown in Table 1, ensuring the material integrity for subsequent welding.
The welding of segments was critical for achieving a monolithic structure. We used vacuum electron beam welding to join the sections, focusing on areas with minimal thickness variations to ensure uniformity. Pre-weld treatments included local pre-heating to 600-800°C and post-weld stress relief through ultrasonic vibration. This approach effectively reduced the risk of heat-affected zone cracking, which is common in high-temperature titanium alloys due to their sensitivity to thermal cycles. The success of this segmented method demonstrated that lost wax investment casting could be combined with advanced welding techniques to produce large components, but it also highlighted the need for an integrated casting process to eliminate weld-related weaknesses.
| Parameter | Front Segment | Middle Segment | Rear Segment |
|---|---|---|---|
| Preheat Temperature (°C) | 400 | 420 | 430 |
| Pouring Temperature (°C) | 1650 | 1670 | 1660 |
| Cooling Time (h) | 4 | 4.5 | 4 |
| Defect Rate (%) | 2.1 | 1.8 | 2.3 |
The mechanical properties of the welded segments were evaluated through tensile and fatigue tests. The results indicated that the weld zones maintained strength levels comparable to the base metal, with ultimate tensile strengths exceeding 900 MPa. However, microstructural analysis revealed slight grain coarsening near the welds, which we attributed to the thermal exposure. To mitigate this, we optimized the welding parameters using a model based on heat input calculation:
$$Q = \frac{V \cdot I}{v}$$
where $Q$ is the heat input, $V$ is voltage, $I$ is current, and $v$ is welding speed. By controlling $Q$ within 50-100 J/mm, we minimized grain growth and preserved the alloy’s high-temperature capabilities. This segmented approach validated the feasibility of lost wax investment casting for large components, but it also underscored the advantages of a one-piece casting for reducing cumulative errors and improving overall reliability.
Integrated Casting Process
Building on the segmented results, we developed an integrated lost wax investment casting process for the entire skeleton. The wax pattern was constructed in multiple sections and assembled using a dedicated jig to ensure dimensional stability. The gating system was designed as a bottom-fed arrangement with multiple ingates to promote uniform filling and reduce turbulence. Reinforcement ribs were added to critical areas to prevent deformation during shell building and casting. The shell fabrication followed the same multi-layer process, but we enhanced the drying cycles to prevent shell cracking in deep cavities. After dewaxing and firing, the shells were inspected for integrity, and any imperfections were corrected to avoid inclusion defects.
During pouring, we maintained a controlled atmosphere in the vacuum furnace to prevent oxidation. The preheated shells were placed in the furnace, and the molten ZTi55A alloy was poured at temperatures around 1680°C, with a superheat of approximately 100°C to improve fluidity. The solidification process was monitored, and we observed complete filling without misruns or cold shuts. Post-casting, the components were subjected to hot isostatic pressing (HIP) at 920°C and 100 MPa for 2 hours to heal internal porosity and enhance density. The resulting castings exhibited a metallic luster with no visible defects, and dimensional inspections confirmed accuracy within ±0.5 mm of the design specifications.
Microstructural examination of the integrated castings revealed a fine-grained α+β phase distribution, which is desirable for high-temperature performance. We used electron backscatter diffraction (EBSD) to analyze the grain orientation and found that the optimized cooling rate minimized texture formation. The mechanical properties, including creep resistance at 550°C, met the aerospace standards, with rupture lives exceeding 100 hours under stress of 300 MPa. This integrated approach demonstrated that lost wax investment casting could produce large, complex titanium alloy components with minimal post-processing, reducing production time and cost compared to segmented methods.
| Aspect | Segmented Casting | Integrated Casting |
|---|---|---|
| Production Time (days) | 25 | 18 |
| Defect Density (per cm²) | 0.15 | 0.08 |
| Dimensional Accuracy (mm) | ±0.3 | ±0.5 |
| Post-treatment Steps | Welding, Stress Relief | HIP, Pickling |
| Cost Index | 1.2 | 1.0 |
The success of the integrated lost wax investment casting process can be attributed to precise control over process variables. For instance, the relationship between pouring rate and defect formation can be described by the Reynolds number for fluid flow:
$$Re = \frac{\rho v L}{\mu}$$
where $\rho$ is density, $v$ is velocity, $L$ is characteristic length, and $\mu$ is viscosity. By maintaining $Re$ below 2000, we ensured laminar flow, reducing gas entrapment and inclusion formation. This holistic approach highlights the potential of lost wax investment casting for manufacturing large titanium alloy structures in a single step, achieving better mechanical integrity and efficiency.
Pickling Process for Contamination Layer Removal
After casting, the surfaces of ZTi55A components often develop an α-case layer, a brittle contamination zone rich in oxygen and nitrogen, which can degrade mechanical properties. We investigated the pickling process to remove this layer effectively. Experiments were conducted on test coupons of varying thicknesses (5 mm, 10 mm, 15 mm, and 20 mm) to determine the α-case thickness and optimal pickling time. Microstructural analysis showed that thicker sections had deeper α-case layers, up to 0.4 mm for 20 mm samples, due to prolonged exposure to the ceramic shell at high temperatures.
We used a pickling solution of HF, HNO3, and H2O in a 1:3:6 ratio, and measured the etching rate as 0.014 mm/min through weight loss experiments. Based on this, the required pickling time for complete α-case removal was calculated as:
$$t = \frac{d}{r}$$
where $t$ is time, $d$ is α-case thickness, and $r$ is etching rate. For a conservative estimate, we set $d = 0.4$ mm, resulting in $t = 28.57$ minutes, which we rounded to 30 minutes for practical application. After pickling, the surfaces were examined using scanning electron microscopy (SEM), which confirmed the absence of the α-case and revealed a clean, uniform substrate. This process not only improved the surface quality but also enhanced the fatigue life by eliminating stress concentration sites.
| Sample Thickness (mm) | α-case Thickness (mm) | Pickling Time (min) | Surface Quality |
|---|---|---|---|
| 5 | 0.1 | 7 | Excellent |
| 10 | 0.2 | 14 | Good |
| 15 | 0.3 | 21 | Good |
| 20 | 0.4 | 30 | Excellent |
Furthermore, we analyzed the chemical reactions during pickling, which involve the dissolution of titanium oxides and nitrides. The overall reaction can be represented as:
$$\ce{Ti + 6HF -> H2TiF6 + 2H2}$$
and
$$\ce{TiO2 + 4HF -> TiF4 + 2H2O}$$
By controlling the acid concentration and temperature, we minimized hydrogen embrittlement risks, which is crucial for titanium alloys. The optimized pickling process became an integral part of our lost wax investment casting post-treatment, ensuring that the final components met the stringent requirements for aerospace applications.
Repair Welding for Defect Correction
Despite process optimizations, some castings exhibited minor defects such as micro-cracks or porosity, necessitating repair welding. We developed a specialized procedure for ZTi55A alloy, focusing on reducing welding stresses and preventing re-cracking. The process began with pre-heating the casting to 350-400°C, followed by local pre-heating of the weld area to 600-800°C using an arc torch. Welding was performed in a vacuum chamber with a current range of 50-100 A and a travel speed of 2-3 mm/s to control heat input. The filler material was a matching ZTi55A wire, which was acid-washed and dried to ensure cleanliness.
After welding, we implemented slow cooling and ultrasonic vibration for stress relief, followed by a vacuum stress relief anneal within 12 hours. This sequence effectively reduced residual stresses below the material’s yield strength, as confirmed by finite element analysis (FEA) simulations. The stress distribution can be modeled using the equation:
$$\sigma_{res} = \sigma_{thermal} + \sigma_{mechanical}$$
where $\sigma_{res}$ is residual stress, $\sigma_{thermal}$ is from thermal gradients, and $\sigma_{mechanical}$ is from constrained contraction. By minimizing these components through controlled cooling, we achieved crack-free repairs. Non-destructive testing, including X-ray and fluorescent penetrant inspection, showed no defects in the welded regions, and tensile tests indicated that the repaired areas retained over 95% of the base metal strength.
| Parameter | Value | Effect on Quality |
|---|---|---|
| Pre-heat Temperature (°C) | 350-400 | Reduces thermal shock |
| Welding Current (A) | 50-100 | Controls penetration |
| Travel Speed (mm/s) | 2-3 | Minimizes heat input |
| Post-weld Treatment | Ultrasonic + Anneal | Relieves stresses |
| Defect Rate After Repair (%) | 0.5 | High reliability |
The success of this repair welding technique underscores the importance of a holistic approach in lost wax investment casting, where post-treatment processes are tailored to the alloy’s characteristics. By integrating these methods, we can salvage components that would otherwise be rejected, improving yield and reducing costs in the production of large titanium alloy castings.
Conclusion
Our research demonstrates the feasibility of using lost wax investment casting for manufacturing large ZTi55A titanium alloy skeletons with complex geometries. Through segmented and integrated approaches, we optimized key process steps, including wax pattern fabrication, shell building, melting and pouring, and post-treatment. The integrated casting method proved superior, reducing production time and defects while maintaining dimensional accuracy and mechanical properties. The pickling process effectively removed the α-case contamination layer, and the developed repair welding technique enabled reliable defect correction without compromising performance.
The mathematical models and empirical data presented here provide a foundation for further advancements in lost wax investment casting for high-temperature titanium alloys. Future work could focus on automating process controls and exploring additive manufacturing for wax patterns to enhance efficiency. Overall, this study contributes to the growing body of knowledge on precision casting, supporting the aerospace industry’s need for high-performance components. By continuously refining the lost wax investment casting process, we can achieve higher quality and reliability in critical applications.