In the field of biomedical engineering, metallic materials play a crucial role in implant applications due to their mechanical strength and durability. Among these, cobalt-based alloys, particularly CoCrMo alloys compliant with ASTM F75 standards, have garnered significant attention for their exceptional corrosion resistance, wear resistance, and biocompatibility. These properties make them ideal for long-term orthopedic implants, such as joint replacements and dental prostheses. However, when produced via investment casting—a common manufacturing method for complex-shaped casting parts—CoCrMo alloys often exhibit coarse grain structures, which can compromise mechanical performance, leading to reduced fatigue life and ductility. To address this, grain refinement techniques are essential, as they enhance microstructure homogeneity and mitigate defects like porosity and segregation. In this study, I investigate the application of electromagnetic stirring (EMS) during the solidification of CoCrMo alloy investment casting parts, aiming to achieve significant grain refinement and promote the columnar-to-equiaxed transition (CET). By combining experimental analysis with numerical simulations, I explore the underlying mechanisms and optimize process parameters for improved microstructure control.
Investment casting, also known as lost-wax casting, is widely used to fabricate intricate casting parts with high dimensional accuracy. For CoCrMo alloys, this process involves creating a ceramic mold around a wax pattern, melting the alloy, and pouring it into the mold. While cost-effective for producing complex geometries, the solidification process in investment casting often results in large columnar grains, especially in thick sections, due to directional heat transfer. This coarse microstructure can lead to anisotropic properties and susceptibility to cracking. Therefore, refining the grain size is critical to enhancing the overall quality of casting parts. Various methods exist for grain refinement, including chemical inoculation and physical techniques like mechanical stirring, ultrasonic treatment, and electromagnetic fields. Among these, EMS stands out due to its non-contact nature, efficiency, and ability to induce forced convection in the melt, thereby altering solidification dynamics. In EMS, a rotating magnetic field is applied to the molten metal, generating Lorentz forces that drive fluid flow. This flow disrupts the thermal and solute fields, leading to dendrite fragmentation and increased nucleation, which collectively refine the grain structure. Previous studies have demonstrated EMS’s effectiveness in steels and aluminum alloys, but its application to CoCrMo investment casting parts remains underexplored. This research aims to fill that gap by systematically evaluating EMS’s impact on CoCrMo alloy microstructure, with a focus on grain size reduction and CET promotion.
The experimental setup involved a custom-built EMS system consisting of a rotating magnetic field power supply and a field generation device. The power supply delivered three-phase alternating current with adjustable effective current (5–350 A) and frequency (5–50 Hz). The field generator comprised six copper coils (50 turns each) wound around an iron core made of stacked silicon steel sheets, along with a cooling water circuit. The coils were arranged in a star configuration to produce a rotating magnetic field when energized. For the casting parts, a standard investment casting mold was prepared using a wax pattern designed to produce test bars and runners, simulating typical implant geometries. The mold was preheated to a specified temperature, and CoCrMo alloy (composition compliant with ASTM F75) was melted in a vacuum induction furnace. Once molten, the alloy was poured into the mold, which was then transferred to the EMS coils within 15 seconds. The EMS parameters were set prior to placement, and stirring was applied for designated durations. Four experimental conditions were tested: one without EMS (control) and three with EMS at varying stirring times and conditions, as summarized in Table 1. After EMS treatment, the casting parts were allowed to cool naturally, and samples were sectioned for metallographic analysis.
| Sample | Current (A) | Frequency (Hz) | Current Holding Time (s) | Commutation Time (s) | Stirring Time (min) | Stirring Conditions |
|---|---|---|---|---|---|---|
| Control | N/A | N/A | N/A | N/A | N/A | No EMS |
| EMS-1 | 100 | 10 | 10 | 30 | 4 | EMS started after mold placement |
| EMS-2 | 100 | 10 | 10 | 30 | 4 | EMS always on |
| EMS-3 | 100 | 10 | 10 | 30 | 8 | EMS always on |
Macrostructural analysis revealed significant differences between the control and EMS-treated casting parts. The control sample exhibited coarse grains, with an average grain size exceeding 2 mm in some regions, while EMS-treated samples showed refined grains, particularly in the runner sections. For instance, in EMS-3 (8-minute stirring), the grain size in the test bar was reduced to below 1 mm, and the equiaxed grain fraction increased to 31%. The refinement effect was more pronounced in thicker sections, such as runners, due to enhanced fluid flow. Microstructural examination using optical microscopy confirmed dendritic structures in all samples, with CET observed from the edges to the centers. The EMS-treated casting parts displayed finer columnar dendrites and more equiaxed grains, with secondary dendrite arm spacing (SDAS) measurements indicating faster cooling in EMS-3 (15.2 ± 1.2 µm) compared to the control (19.8 ± 2.0 µm). These findings suggest that EMS not only refines grains but also improves thermal uniformity during solidification.
To elucidate the mechanisms behind grain refinement, numerical simulations were conducted using COMSOL Multiphysics software. The model simulated electromagnetic fields, induced currents, Lorentz forces, and fluid flow within the casting parts during EMS. The governing equations include Maxwell’s equations for electromagnetism and the Navier-Stokes equations for fluid dynamics. The magnetic flux density B and electric field E are described by:
$$ \nabla \times \mathbf{E} = -\frac{\partial \mathbf{B}}{\partial t} $$
$$ \nabla \times \mathbf{H} = \mathbf{J} + \frac{\partial \mathbf{D}}{\partial t} $$
where H is the magnetic field intensity, J is the current density, and D is the electric displacement field. For the molten metal, the Lorentz force F is calculated as:
$$ \mathbf{F} = \mathbf{J} \times \mathbf{B} $$
This force drives fluid flow, with the velocity field u obeying:
$$ \rho \left( \frac{\partial \mathbf{u}}{\partial t} + (\mathbf{u} \cdot \nabla) \mathbf{u} \right) = -\nabla p + \mu \nabla^2 \mathbf{u} + \mathbf{F} $$
where ρ is density, p is pressure, and μ is dynamic viscosity. The simulation results showed that the magnetic field rotated periodically with the alternating current, inducing eddy currents in the casting parts. The Lorentz force magnitude varied cyclically, peaking in central regions due to higher current density. Fluid flow velocities increased over time, stabilizing after several seconds, with maximum flow in thick sections like runners. This aligns with the experimental observation that refinement was more effective in areas with stronger convection. The simulated flow patterns indicated turbulent mixing, which promotes dendrite fragmentation and temperature homogenization.
The grain refinement mechanism in EMS can be attributed to two primary factors: dendrite fragmentation due to melt flow and enhanced heterogeneous nucleation from electromagnetic effects. The Lorentz force-induced flow creates shear stresses that break off dendrite arms, especially in the mushy zone. These fragments act as nucleation sites, increasing grain density. Additionally, the electromagnetic field may lower the critical nucleation radius by increasing undercooling. The relationship between undercooling ΔT and nucleation radius r* is given by:
$$ r^* = \frac{2 \sigma_{sl}}{L_m \Delta T} $$
where σsl is the solid-liquid interfacial energy and Lm is the latent heat of fusion. As ΔT increases, r* decreases, facilitating more nucleation events. In this study, the dominant mechanism appears to be dendrite fragmentation, as evidenced by the correlation between flow intensity and refinement extent. For instance, in EMS-3, prolonged stirring led to finer grains and higher equiaxed fraction, consistent with continuous dendrite disruption. The contribution of electromagnetic nucleation is secondary but non-negligible, particularly in regions with high current density. To quantify these effects, I derived an expression for the Lorentz force magnitude F in a rotating field:
$$ F \approx \frac{\pi^2 f \mu \sigma k U^2 r^{-1}}{2} e^{-\pi^2 f \mu \sigma r^2} $$
where f is frequency, μ is magnetic permeability, σ is electrical conductivity, k is a coil constant, U is voltage, and r is radial distance. This force drives flow velocity v, which scales with stirring time t:
$$ v(t) = v_0 (1 – e^{-t/\tau}) $$
where v0 is steady-state velocity and τ is a time constant. The grain size d can be related to flow parameters through empirical models, such as:
$$ d = A v^{-\alpha} + B $$
where A, B, and α are material constants. For CoCrMo alloy, my data suggests α ≈ 0.5, indicating significant refinement with increased flow. These equations highlight the interplay between EMS parameters and microstructure outcomes.
Further analysis of the casting parts involved measuring grain size and equiaxed fraction across different regions, as shown in Table 2. The control sample had large grains (up to 2.5 mm) and low equiaxed fraction (13.1%), while EMS-treated samples exhibited progressive improvement. EMS-3 achieved the best results, with grain sizes below 1 mm and equiaxed fraction reaching 30.9%. This underscores the importance of stirring duration in optimizing refinement. Additionally, SDAS decreased with longer EMS, confirming enhanced cooling rates. The numerical simulation provided insights into why thick sections refined more: higher Lorentz forces and flow velocities in those zones, as calculated from the model. For example, in the runner, simulated velocities exceeded 0.1 m/s, compared to 0.05 m/s in thin sections, leading to more effective dendrite fragmentation. This explains the superior refinement in runners observed experimentally.
| Sample | End 1 Grain Size (µm) | Middle Grain Size (µm) | End 2 Grain Size (µm) | Equiaxed Grain Rate (%) |
|---|---|---|---|---|
| Control | 2201 ± 1216 | 1080 ± 724 | 2470 ± 756 | 13.1 |
| EMS-1 | 665 ± 96 | 778 ± 97 | 1745 ± 876 | 27.8 |
| EMS-2 | 847 ± 302 | 673 ± 305 | 1063 ± 202 | 25.0 |
| EMS-3 | 594 ± 74 | 335 ± 81 | 928 ± 256 | 30.9 |
The implications of this research extend beyond CoCrMo alloys to other investment-cast metals. By refining grains, EMS can improve mechanical properties such as tensile strength, fatigue resistance, and ductility in casting parts. For biomedical implants, this translates to longer service life and reduced failure risks. Moreover, EMS is environmentally friendly compared to chemical refiners, as it avoids introducing foreign elements. However, challenges remain in scaling up the process for industrial production. Factors like mold design, alloy composition, and EMS parameter optimization need further study. My simulations suggest that increasing current or frequency can enhance refinement, but excessive stirring may cause defects like gas entrapment. Therefore, a balanced approach is essential. Future work could explore combined methods, such as EMS with inoculants, to achieve even finer microstructures.
In conclusion, electromagnetic stirring is a highly effective technique for grain refinement in CoCrMo alloy investment casting parts. Through experimental and numerical investigations, I demonstrated that EMS promotes the columnar-to-equiaxed transition, reduces grain size to below 1 mm, and increases equiaxed fraction up to 31%. The refinement is driven primarily by dendrite fragmentation from forced convection, with secondary contributions from electromagnetic nucleation. Longer stirring times and optimized parameters yield better results, especially in thick sections. This study provides a foundation for improving the quality of biomedical casting parts, offering a pathway to enhanced performance and reliability. As demand for advanced implants grows, such physical refinement methods will play a key role in manufacturing high-integrity components.
To support these findings, I present additional data on the relationship between EMS parameters and microstructure metrics. The following equations summarize key relationships derived from this work. The undercooling ΔT induced by EMS can be estimated as:
$$ \Delta T = \frac{\mu_0 \chi}{2 \Delta H_m} \left( U_1^2 e^{-\pi^2 f_1 \mu \sigma r^2} – U_2^2 e^{-\pi^2 f_2 \mu \sigma r^2} \right) $$
where μ0 is vacuum permeability, χ is magnetic susceptibility, and ΔHm is enthalpy of fusion. This undercooling reduces the critical radius, enhancing nucleation. Additionally, the grain refinement efficiency η can be defined as:
$$ \eta = \frac{d_0 – d}{d_0} \times 100\% $$
where d0 is grain size without EMS and d is with EMS. For EMS-3, η reached ~70% in some regions. These quantitative insights aid in process control for investment casting parts.
Overall, this research underscores the value of electromagnetic stirring in advancing casting technology. By integrating experimental validation with computational modeling, I have elucidated the mechanisms behind grain refinement, paving the way for optimized production of high-performance CoCrMo alloy components. As I continue to explore EMS applications, the goal is to develop standardized protocols for various alloy systems, ensuring consistent quality in critical casting parts for medical and industrial use.