The relentless pursuit of higher performance and efficiency in modern aerospace engines places increasingly stringent demands on high-temperature structural materials. Components such as turbine disks and casings often operate at temperatures below the equi-cohesive temperature, where grain refinement is a critical pathway to significantly enhance their service performance, including fatigue life and mechanical strength. Among the various processing techniques, precision investment casting stands as a cornerstone for manufacturing complex, near-net-shape components from nickel-based superalloys. However, conventional gravity-fed precision investment casting often results in coarse, columnar, or large equiaxed grain structures due to the high pouring temperatures and excellent insulation properties of ceramic molds. This coarse microstructure leads to elemental segregation, the formation of detrimental secondary phases, and ultimately, inferior and scattered mechanical properties.
To address these limitations, dynamic grain refinement methods integrated into the precision investment casting process have emerged as a powerful and contamination-free solution. One such advanced technique involves imposing controlled agitation on the solidifying melt. This work delves into the effects of a specific dynamic refinement strategy—mold positive and negative rotation (P&NR)—on the solidification microstructure and consequent room-temperature mechanical properties of a widely used Ni-Fe based superalloy, K4169. This alloy is typically utilized for components like casings and turbine disks where fine, uniform equiaxed grains are desirable for optimal performance in the intermediate temperature range.
The principle behind dynamic methods like P&NR is the introduction of forced convection during the critical initial stages of solidification. In conventional casting, a stable boundary layer forms at the solid-liquid interface. By periodically reversing the direction of mold rotation, this stable boundary layer is disrupted, generating intense turbulence within the molten metal. This turbulent flow exerts shear forces on the growing dendrites, causing their fragmentation. These fragmented dendrites are then carried into the bulk liquid, acting as potent nucleation sites, thereby dramatically increasing the nucleation rate. Concurrently, the vigorous fluid flow enhances heat and mass transfer, leading to a more uniform temperature and solute field, which further promotes the formation of a fine, equiaxed grain structure.
This article systematically compares the outcomes of conventional gravity casting (CC) with those of the P&NR-enhanced precision investment casting process. We will explore the profound changes in macro/micro-structure, elemental segregation, secondary phase distribution, and the resulting room-temperature tensile properties. Theoretical frameworks involving nucleation kinetics and strengthening mechanisms will be employed to elucidate the observed phenomena.
Experimental Methodology and Material
The base material for this investigation was the cast superalloy K4169, with its nominal composition provided in Table 1. A characteristic annular component (inner diameter: 100 mm, outer diameter: 140 mm, height: 70 mm) was designed to simulate a typical rotational geometry encountered in engine hardware.
| Ni | Cr | Nb | Mo | Al | Ti | C | Fe |
|---|---|---|---|---|---|---|---|
| 52.0 | 19.1 | 5.03 | 3.11 | 0.61 | 0.94 | 0.05 | Bal. |
The molds were fabricated using the standard precision investment casting route with zirconia-based ceramic shells. All melting and casting were conducted under a protective atmosphere. The alloy was superheated to 1450°C and poured into a preheated mold (1000°C). Three distinct processing conditions were evaluated:
- Conventional Casting (CC): The mold remained stationary after pouring, allowing the metal to solidify under gravity.
- P&NR 4s: Immediately after pouring, the mold was subjected to alternating clockwise and counterclockwise rotation. Each rotation direction was maintained for a duration of 4 seconds before reversal. This sequence continued until the completion of solidification.
- P&NR 8s: Similar to P&NR 4s, but with each rotational direction sustained for 8 seconds before reversal.
Specimens for microstructural analysis were sectioned from the cast components. Macro-etching was performed to reveal grain boundaries, and the average grain size was determined using the linear intercept method. For dendrite morphology and phase analysis, standard metallographic preparation followed by appropriate etching was conducted. Microstructural examination was carried out using optical microscopy (OM) and scanning electron microscopy (SEM) equipped with energy-dispersive X-ray spectroscopy (EDS). Phase identification was confirmed via X-ray diffraction (XRD). The microhardness of the γ matrix and various secondary phases was measured. Room-temperature tensile tests were performed on specimens machined from the castings, and fracture surfaces were examined via SEM to identify failure mechanisms.
Results and Discussion: Transformation of Solidification Structure
1. Macrostructural Evolution: Dramatic Grain Refinement
The most striking effect of implementing the P&NR technique within the precision investment casting cycle is the exceptional refinement of the as-cast grain structure. As quantified in Table 2, the conventional casting process yielded very coarse equiaxed grains with an average size of (5.37 ± 0.21) mm. The application of mold agitation profoundly altered this outcome.
| Process | Average Grain Size, d (mm) | Ultimate Tensile Strength, UTS (MPa) | Elongation, EL (%) |
|---|---|---|---|
| Conventional Casting (CC) | 5.37 ± 0.21 | 638.4 ± 10.4 | 19.8 ± 1.2 |
| P&NR 4s | 0.27 ± 0.01 | 838.7 ± 3.5 | 29.8 ± 1.8 |
| P&NR 8s | 0.66 ± 0.05 | 808.2 ± 10.6 | 20.6 ± 0.2 |
The P&NR 4s condition resulted in the finest microstructure, with the grain size reduced by over an order of magnitude to (0.27 ± 0.01) mm. Interestingly, increasing the dwell time per rotation to 8 seconds (P&NR 8s) led to a partial coarsening of grains (0.66 ± 0.05 mm), indicating that the refinement efficacy is non-monotonic with respect to reversal time. This can be explained by the fluid dynamics at the solid-liquid interface. When the mold rotates in one direction for a shorter period (4s), a relatively thin stable boundary layer develops. Upon reversal, the generated turbulence effectively penetrates this layer, causing efficient dendrite fragmentation. With a longer rotation time (8s), a thicker, more stable boundary layer forms, which may shield the growing dendrites from the full effect of the subsequent turbulent flow upon reversal, thus reducing the fragmentation efficiency.
From a thermodynamic perspective, the application of centrifugal force during P&NR processing influences the nucleation barrier. The critical radius for nucleation (r*) and the associated critical nucleation work (ΔG*) under an applied pressure P can be expressed as:
$$
r^* = \frac{2 \gamma_{sl}}{\Delta G_v + k \epsilon P}
$$
$$
\Delta G^* = \frac{16 \pi \gamma_{sl}^3}{3(\Delta G_v + k \epsilon P)^2}
$$
where $\gamma_{sl}$ is the solid-liquid interfacial energy, $\Delta G_v$ is the volumetric free energy change, $k$ is a conversion factor, and $\epsilon$ is the volumetric shrinkage. Compared to gravity casting (P=0), the P&NR process effectively increases the pressure term, thereby decreasing both $r^*$ and $\Delta G^*$. This reduction in the energy barrier facilitates a higher nucleation rate (N), fundamentally enabling the observed grain refinement. The nucleation rate is governed by:
$$
N = K \cdot \exp\left(-\frac{\Delta G^*}{k_B T}\right) \cdot \exp\left(-\frac{Q}{k_B T}\right)
$$
where $K$ is a pre-exponential factor, $k_B$ is Boltzmann’s constant, $T$ is temperature, and $Q$ is the activation energy for diffusion. The decrease in $\Delta G^*$ directly leads to an exponential increase in N.
2. Microstructural Morphology: Dendrite Fragmentation and Degeneration
The microstructural refinement is not limited to the grain scale but extends to the morphology of the primary γ phase. In the CC condition, the microstructure exhibits well-developed dendritic morphology with clear primary, secondary, and even tertiary arms. This is characteristic of slow cooling with limited convection.
In contrast, the P&NR-processed samples, particularly the P&NR 4s condition, show a dramatically altered morphology. The coarse dendrites are replaced by a mixture of fragmented dendrites and rosette-like or granular crystals. This is direct evidence of the dendrite fragmentation mechanism induced by the fluid shear stresses during mold reversal. The fragmented dendrites, dispersed throughout the melt, become the primary sites for the formation of new grains, leading to the fine, equiaxed structure. The intense convection also reduces the temperature gradient in the liquid ahead of the solidification front and thins the solute diffusion boundary layer, creating conditions favorable for equiaxed growth over columnar development.
3. Mitigation of Elemental Segregation
Superalloys like K4169 are prone to microsegregation due to the partitioning of alloying elements during solidification. Elements like Nb, Mo, and Ti, which lower the melting point, are rejected into the liquid and enrich the interdendritic regions (positive segregation), while Cr and Fe tend to be enriched in the dendritic cores (negative segregation). The severity of this segregation significantly impacts the type, amount, and distribution of secondary phases. The segregation coefficient (k) for an element is defined as the ratio of its concentration in the solid ($C_s$) to that in the liquid ($C_l$) at the interface, $k = C_s / C_l$. In practice, the effective segregation is assessed by measuring the composition in dendrite core and interdendritic areas.
EDS analysis revealed that P&NR processing effectively alleviated microsegregation. As summarized in Table 3, the segregation ratios (Interdendritic/Dendritic concentration) for key elements like Nb and Mo were reduced compared to the CC condition.
| Feature | Conventional Casting (CC) | P&NR 4s Casting |
|---|---|---|
| Nb, Mo Segregation | Severe | Moderate |
| Laves Phase | Large, blocky, continuous networks | Smaller, more isolated, reduced fraction |
| MC Carbides | Large, script-like, lower fraction | Finer, more dispersed, slightly higher fraction |
| Dendrite Morphology | Coarse, well-developed | Fragmented, rosette-like |
This improvement is attributed to two main factors: First, the enhanced mass transport from forced convection promotes solute redistribution, homogenizing the liquid composition. Second, the drastic reduction in dendrite arm spacing (due to grain refinement) drastically shortens the diffusion path lengths for back-diffusion during the later stages of solidification, allowing for more compositional uniformity.
4. Evolution of Secondary Phases: Laves and Carbides
The solidification sequence of K4169 involves the primary precipitation of γ dendrites, followed by the formation of MC-type carbides (Nb, Ti)C in the interdendritic liquid, and finally a terminal L→γ + Laves eutectic reaction. The Laves phase (e.g., Fe2Nb) is a brittle, topologically close-packed intermetallic that consumes significant amounts of potent strengthening elements like Nb. Its presence, especially as a continuous network, is detrimental to mechanical properties. MC carbides, while hard and brittle, can be beneficial if they are finely dispersed, acting as barriers to dislocation motion.
XRD analysis confirmed that P&NR processing did not introduce new phases but altered their distribution and volume fraction. SEM examination coupled with image analysis revealed a marked transformation (see Table 3). In the CC sample, large, blocky Laves phases formed interconnected networks along coarse interdendritic channels. After P&NR 4s treatment, the Laves phase appeared as smaller, more isolated particles, and its overall area fraction decreased. Conversely, the population of MC carbides increased slightly and their morphology became more granular and dispersed.
This favorable shift is a direct consequence of reduced segregation and refined microstructure. With less Nb sequestered in the coarse, continuous Laves networks, more Nb is available in the matrix to form fine, strengthening γ” (Ni3Nb) precipitates during subsequent heat treatment and/or to participate in the formation of finer, discrete carbides. The confined interdendritic spaces in the fine-grained structure also physically restrict the growth of large Laves phases.
The hardness of these phases was measured, confirming that MC carbides are the hardest (~552 HV), followed by Laves phase (~466 HV), with the γ matrix being the softest (~249 HV). The transition from a microstructure dominated by continuous brittle Laves networks to one with finer, dispersed carbides and isolated Laves is a key contributor to enhanced ductility and strength.
Mechanical Property Enhancement and Fracture Analysis
1. Room-Temperature Tensile Properties
The profound microstructural modifications induced by P&NR precision investment casting translated into significant improvements in room-temperature mechanical performance, as detailed in Table 2. The ultimate tensile strength (UTS) increased from 638 MPa for CC to 839 MPa for P&NR 4s—an impressive increase of approximately 31.4%. The elongation to failure also saw a remarkable boost from 19.8% to 29.8%, indicating a simultaneous improvement in strength and ductility. The P&NR 8s condition also showed superior properties compared to CC, but they were slightly inferior to the optimal P&NR 4s condition, correlating with its relatively coarser grain size.
The strengthening mechanism can be rationalized primarily through the Hall-Petch relationship, which relates yield strength ($\sigma_y$) to grain size ($d$):
$$
\sigma_y = \sigma_0 + k_{HP} \cdot d^{-1/2}
$$
where $\sigma_0$ is the friction stress and $k_{HP}$ is the Hall-Petch slope. The dramatic reduction in grain size from ~5.37 mm to ~0.27 mm provides a substantial contribution to the increased strength due to grain boundary strengthening. Finer grains increase the density of grain boundaries, which act as effective barriers to dislocation motion, leading to higher stress required for plastic deformation.
The overall yield strength ($\sigma_s$) of a particle-containing alloy can be described by an extended model:
$$
\sigma_s = \sigma_i + k_{HP} \cdot d^{-1/2} + \sigma_{intra} + \sigma_{inter}
$$
Here, $\sigma_i$ is the intrinsic strength of the matrix, $\sigma_{intra}$ is the contribution from intragranular precipitates (like γ’ and γ”), and $\sigma_{inter}$ is the contribution from intergranular particles (like carbides and isolated Laves). While $\sigma_{intra}$ is crucial for precipitation-strengthened superalloys, the changes in $\sigma_{inter}$ are also significant. The refinement and dispersion of secondary phases improve the load transfer characteristics and reduce stress concentration sites. The reduction of continuous brittle Laves networks eliminates easy paths for crack initiation and propagation, directly benefiting both strength and ductility.
2. Fractography
Analysis of the tensile fracture surfaces provided clear evidence of the microstructural benefits. The fracture surface of the conventional cast (CC) specimen exhibited a mixed-mode morphology. While some areas showed dimples indicative of microvoid coalescence (a ductile fracture mechanism), other regions displayed quasi-cleavage features and relatively shallow dimples. This is consistent with fracture initiating at or propagating along the coarse, brittle Laves phase networks.
In stark contrast, the fracture surface of the P&NR 4s specimen was characterized by a uniform distribution of fine, deep, and equiaxed dimples. This morphology is classic for ductile fracture and indicates homogeneous plastic deformation throughout the gauge length. The absence of large cleavage facets confirms that the refined and homogenized microstructure effectively suppressed the early initiation of cracks at brittle secondary phases. The finer grain size promotes more grain boundary sliding and rotation, allowing for better accommodation of strain and leading to the observed superior elongation.
Conclusions and Perspectives
This investigation conclusively demonstrates the transformative potential of integrating dynamic grain refinement via mold positive and negative rotation (P&NR) into the precision investment casting process for high-performance superalloys like K4169. The key findings are:
- Exceptional Grain Refinement: The P&NR process, particularly with an optimal reversal period (4s in this study), can refine the as-cast grain size by over an order of magnitude compared to conventional gravity casting, changing it from millimeter-scale to sub-millimeter scale.
- Microstructural Homogenization: The technique promotes dendrite fragmentation, degrades dendritic morphology, and significantly reduces the microsegregation of key alloying elements such as Nb and Mo.
- Secondary Phase Control: It favorably alters the distribution of secondary phases, reducing the amount and continuity of the detrimental Laves phase while promoting a finer, more dispersed distribution of MC carbides.
- Enhanced Mechanical Properties: These microstructural improvements synergistically lead to a substantial increase in room-temperature tensile strength (up to 31.4%) and elongation (up to 50% increase), as explained by Hall-Petch strengthening and the elimination of brittle fracture pathways.
The success of this P&NR-enhanced precision investment casting technique hinges on the optimization of processing parameters, most notably the reversal timing, to maximize fluid turbulence at the solid-liquid interface for effective dendrite fragmentation without allowing the re-establishment of a stable boundary layer.
Future work should focus on exploring the effects of this technique on high-temperature properties, such as creep and fatigue resistance, which are critical for acro-engine components. Additionally, coupling this dynamic casting method with subsequent hot isostatic pressing (HIP) could further eliminate any residual microporosity and optimize the precipitation of strengthening phases, potentially unlocking even higher levels of performance. The P&NR approach represents a significant step forward in precision investment casting technology, offering a clean, efficient, and highly effective route to manufacturing superalloy components with refined, reliable, and superior microstructures and properties.