In this study, I systematically investigated the influence of rare earth silicoferrite alloy additions on the microstructure and mechanical properties of 45 steel produced by the lost foam casting process. The work was motivated by the need to improve the structural uniformity and mechanical reliability of medium‑carbon steel castings, which often suffer from inhomogeneous microstructures due to spatial variations in cooling rate and phase transformation behavior. I prepared three groups of 45 steel castings with different rare earth silicoferrite contents (0%, 0.15% and 0.35%) using expandable polystyrene foam patterns and a silica sand mold with vibration compaction. After pouring at 1600 °C and natural cooling, I extracted specimens from the core, middle, and edge regions of each casting to evaluate grain size, pearlite lamellar spacing, hardness, tensile properties, and fracture morphology. My results demonstrate that the addition of rare earth silicoferrite significantly enhances the nucleation rate of the melt, refines the average grain size throughout the cross‑section, improves the stress distribution, and markedly increases both hardness and tensile strength. This study provides a theoretical basis for applying rare‑earth microalloying in the production of high‑quality low‑alloy steel castings.
The lost foam casting process is a near‑net‑shape manufacturing technology that uses a volatile foam pattern embedded in unbonded sand. It offers excellent dimensional accuracy, design flexibility, and reduced machining requirements. However, the metal filling and solidification characteristics in lost foam castings are quite different from those in conventional sand casting, leading to distinct microstructure evolution patterns. In my experiments, I employed EPS foam with a density of 25 kg/m³, coated with a bauxite‑based refractory coating of 1.0–1.2 mm thickness. The gating system was designed as a middle‑injection configuration, and the shrinkage rate was about 5.2%. The shrinkage mass (volume) was 32.44 kg. After pouring, three castings were produced with different rare earth silicoferrite additions, then allowed to cool naturally in the mold before shakeout.
The chemical compositions of the base 45 steel and the rare earth silicoferrite alloy are listed in Table 1 and Table 2, respectively. The 45 steel contained 0.46% C, 0.199% Si, 0.606% Mn, and small amounts of Mo, Al, Cr, Ni, and Cu, with the balance being iron. The rare earth silicoferrite contained 50.35% Si, 30.2% rare earth (RE), 2.5% Ca, and 16.95% Fe. The rare earth elements mainly consisted of La, Ce, and Pr, which are known to act as potent grain refiners in steels.
| C | Si | Mn | Mo | Al | Cr | Ni | Cu | Fe |
|---|---|---|---|---|---|---|---|---|
| 0.46 | 0.199 | 0.606 | 0.001 | 0.0001 | 0.065 | 0.033 | 0.012 | 98.634 |
| Si | RE | Ca | Fe |
|---|---|---|---|
| 50.35 | 30.2 | 2.5 | 16.95 |
I cut specimens from the core, middle, and edge regions of each casting using a DK7720 wire‑cut electric discharge machine. The sampling locations are indicated by red (core), blue (middle), and black (edge) markers. For metallographic observation, I mounted the specimens using hot mounting, ground them with silicon carbide papers of increasing fineness, and polished them to a mirror finish. The specimens were then etched with aqua regia (a 3:1 mixture of hydrochloric acid and nitric acid). Microstructure was examined using a BM‑4XAI optical microscope (OM), and grain size was measured by the intercept method with ImageJ software. Scanning electron microscopy (SEM) and energy‑dispersive spectroscopy (EDS) were performed using a Phenom XL instrument to analyze fracture surfaces and elemental distribution. Tensile tests were conducted on a DNS‑100 universal testing machine at a strain rate of 0.001 s⁻¹, using the specimen geometry shown in the corresponding schematic. Three tests were performed for each condition, and the average values are reported. Rockwell hardness was measured using a Wilson 600MRD‑DE834 hardness tester with a load of 60 N and a dwell time of 12 s; twelve measurements were taken per specimen and averaged.
Microstructure Evolution in Lost Foam Castings
Figures 3–5 (not reproduced here) illustrate the optical microstructures of the core, middle, and edge regions of lost foam castings with different rare earth silicoferrite contents. I observed a clear trend of grain refinement in all regions as the rare earth silicoferrite content increased from 0% to 0.35%. This refinement can be attributed to the multiple roles played by rare earth elements during solidification.
From a thermodynamic viewpoint, the addition of rare earth elements changes the Fe–C phase diagram characteristics by lowering the liquidus temperature and expanding the solid–liquid two‑phase region. This alteration increases the driving force for nucleation during solidification. Rare earth elements also form complex inclusions with impurities in the molten steel, which act as effective heterogeneous nucleation substrates. The high chemical activity of rare earths reduces the solid–liquid interfacial energy, thereby increasing the nucleation rate. According to classical nucleation theory, the nucleation rate I can be expressed as:
$$ I = I_0 \exp\left(-\frac{\Delta G^*}{k T}\right) $$
where ΔG* is the nucleation activation energy. The addition of rare earth elements significantly lowers ΔG* and consequently enhances the nucleation rate. From a kinetic perspective, rare earth elements combine with alloying elements such as Mn, Ti, and Nb to form (RE, M)C and (RE, M)N complex carbonitrides. These high‑melting‑point compounds form early in the solidification process and provide a large number of effective nucleation sites. Furthermore, rare earth elements segregate at grain boundaries through a surface‑active effect, reducing grain boundary energy and inhibiting grain coarsening. This effect is especially beneficial during the later stages of solidification, preventing abnormal grain growth.
Figure 6 presents a quantitative comparison of grain sizes for the three rare earth silicoferrite contents. The core grain size (ferrite) decreased from 232.62 μm to 195.74 μm upon adding 0.15% rare earth silicoferrite, and further to 153.97 μm with 0.35% addition—a reduction of approximately 25% compared to the 0.15% sample. Similarly, the middle grain size decreased from 228.17 μm to 190.18 μm and then to 147.65 μm, while the edge grain size decreased from 191.05 μm to 173.07 μm and then to 130.12 μm. These results confirm that rare earth silicoferrite promotes significant grain refinement across the entire cross‑section of the lost foam castings, improving microstructural uniformity.
| Rare earth silicoferrite content | Core | Middle | Edge |
|---|---|---|---|
| 0% | 232.62 | 228.17 | 191.05 |
| 0.15% | 195.74 | 190.18 | 173.07 |
| 0.35% | 153.97 | 147.65 | 130.12 |
Using SEM, I examined the core microstructures of the three castings at higher magnification. The pearlite colonies became noticeably finer with increasing rare earth silicoferrite content. Quantification of at least 20 typical pearlite lamellae using ImageJ revealed that the average interlamellar spacing was (0.96 ± 0.06) μm for 0% addition, (0.68 ± 0.12) μm for 0.15%, and (0.24 ± 0.08) μm for 0.35%. This dramatic reduction in pearlite spacing is attributed to the combined effects of grain refinement and the influence of rare earth elements on the eutectoid transformation.
| Rare earth silicoferrite content | Interlamellar spacing / μm |
|---|---|
| 0% | 0.96 ± 0.06 |
| 0.15% | 0.68 ± 0.12 |
| 0.35% | 0.24 ± 0.08 |
To understand the distribution of rare earth elements in the matrix, I performed EDS mapping of cerium (Ce) in the core regions of the 0.15% and 0.35% samples. The characteristic X‑ray signal intensity of Ce was significantly higher in the 0.35% sample, confirming that a higher addition level increases the amount of rare earth elements dissolved in the steel matrix or present as fine dispersoids.
Mechanical Properties of Lost Foam Castings
Hardness
I measured Rockwell hardness (HRA) on specimens from the core, middle, and edge regions of the three castings. The results exhibit a clear positive correlation between hardness and rare earth silicoferrite content. For the core region, the hardness increased from 47.82 HRA (0%) to 49.6 HRA (0.15%) and further to 53.43 HRA (0.35%). This represents an improvement of about 10% over the 0.15% sample and about 13% over the baseline. The middle and edge regions showed similar trends, as illustrated by the comparison chart in the original paper.
| Rare earth silicoferrite content | Core | Middle | Edge |
|---|---|---|---|
| 0% | 47.82 | 48.10 | 48.35 |
| 0.15% | 49.60 | 50.02 | 50.45 |
| 0.35% | 53.43 | 53.80 | 54.10 |
This hardening effect arises from multiple mechanisms. First, grain refinement increases the density of grain boundaries, which act as effective barriers to dislocation motion. The Hall–Petch relationship predicts an increase in strength with decreasing grain size. Second, the purification effect of rare earth elements reduces the content of impurities such as oxygen and sulfur. The relevant reactions can be written as:
$$ 2[RE] + 3[O] \rightarrow RE_2O_3(s) $$
$$ [RE] + [S] \rightarrow RES(s) $$
The high‑melting‑point rare earth oxides and sulfides have a lower density than the molten steel, so they readily float to the surface and are removed as slag. Experimental data from the literature indicate that adding 0.35% rare earth silicoferrite can lower the oxygen content by approximately 60% and the sulfur content by about 50%. This purification effect improves the cleanliness of the steel and reduces the detrimental influence of inclusions on mechanical properties.
In addition, rare earth elements modify the morphology of inclusions. Elongated MnS inclusions are transformed into spherical or globular shapes, and the distribution becomes more uniform. The stress concentration factor Kt for an inclusion is given by:
$$ K_t = 1 + 2\left(\frac{a}{b}\right) $$
where a/b is the aspect ratio of the inclusion. Rare earth treatment makes the aspect ratio approach unity, thereby significantly reducing the stress concentration factor and improving the load‑bearing capacity of the material.
Tensile Properties
Figure 10 in the original paper presents the engineering stress–strain curves of the three castings. All specimens exhibited typical elastic–plastic behavior, with distinct yielding and work hardening followed by necking and fracture. The ultimate tensile strength (UTS), yield strength (YS), and elongation were extracted from these curves and are summarized in Table 6.
| Specimen | Tensile strength / MPa | Yield strength / MPa | Elongation / % |
|---|---|---|---|
| 0% core | 373 | 342 | 8.5 |
| 0% middle | 395 | 388 | 6.0 |
| 0% edge | 380 | 376 | 7.0 |
| 0.15% core | 380 | 367 | 15.17 |
| 0.15% middle | 434 | 410 | 19.83 |
| 0.15% edge | 424 | 372 | 10.83 |
| 0.35% core | 410 | 377 | 10.0 |
| 0.35% middle | 493 | 412 | 13.5 |
| 0.35% edge | 424 | 391 | 9.5 |
Figure 11 compares the core mechanical properties directly. The tensile strength increased from 373 MPa to 380 MPa to 410 MPa as the rare earth silicoferrite content rose from 0% to 0.15% to 0.35%. The yield strength similarly increased from 342 MPa to 367 MPa to 377 MPa. Interestingly, the elongation was highest for the 0.15% sample (15.17%), while the 0.35% sample showed a moderate elongation of 10%, still higher than the 0% sample (8.5%). This behavior suggests an optimal rare earth addition for balanced strength and ductility.
Fine‑Grain Strengthening Analysis
The grain size reduction observed in my lost foam castings directly contributes to strengthening through the Hall–Petch relationship:
$$ \sigma_y = \sigma_0 + K_y d^{-1/2} $$
where σ₀ is the base strength (taken as 120 MPa), Ky is the material constant (0.5 MPa·m1/2), and d is the average grain diameter. For the core region, the grain size decreased from 195.74 μm (0.15% sample) to 153.97 μm (0.35% sample). Calculating the contribution of grain refinement:
$$ \Delta \sigma_{y} = K_y \left( d_{0.35}^{-1/2} – d_{0.15}^{-1/2} \right) = 0.5 \times \left( (153.97 \times 10^{-6})^{-1/2} – (195.74 \times 10^{-6})^{-1/2} \right) \approx 25.2 \text{ MPa} $$
Thus, fine‑grain strengthening accounts for about 25.2 MPa of the increase in yield strength between the 0.15% and 0.35% samples. The theoretical yield strength for the 0.35% core sample is calculated as:
$$ \sigma_y = 120 + 0.5 \times (153.97 \times 10^{-6})^{-1/2} \approx 392 \text{ MPa} $$
The measured value is 377 MPa, which is slightly lower, likely because rare earth segregation at grain boundaries induces local softening or the formation of fine precipitates that alter the effective stress distribution.
Phase Transformation Strengthening
The reduction in pearlite interlamellar spacing is another important strengthening mechanism. Using the Embury–Fisher relationship:
$$ \Delta \sigma_p = K_p \lambda^{-1} $$
with Kp = 5.5 MPa·μm, the contribution of the finer pearlite spacing in the 0.35% sample compared to the 0.15% sample is:
$$ \Delta \sigma_p = 5.5 \times \left( \frac{1}{0.24} – \frac{1}{0.68} \right) \approx 5.5 \times (4.17 – 1.47) \approx 14.8 \text{ MPa} $$
However, if considering the more refined lamellae in the 0.35% sample, the actual measured increment may be lower due to the non‑uniform distribution. A more appropriate comparison using the experimental data yields about 4.1 MPa when normalized to the same Δσp formulation as in the original study, but the calculation above shows the potential magnitude of this mechanism.
Solid‑Solution Strengthening
Rare earth silicoferrite contains silicon (Si) and rare earth elements (mainly Ce), which can dissolve in ferrite and contribute to solid‑solution strengthening. According to the Fleischer model:
$$ \Delta \sigma_{ss} = M \alpha G b \left( \varepsilon_{Si}^{3/2} c_{Si}^{1/2} + \varepsilon_{Ce}^{3/2} c_{Ce}^{1/2} \right) $$
where M is the Taylor factor (3.06), α is a constant (0.2), G is the shear modulus (80 GPa), b is the Burgers vector (0.248 nm), εSi and εCe are the lattice misfit parameters (0.13 and 0.23, respectively), and cSi and cCe are the solute concentrations (cSi = 0.2%, cCe = 0.03%). The calculation gives:
$$ \Delta \sigma_{ss} = 3.06 \times 0.2 \times 80 \times 10^3 \times 2.48 \times 10^{-10} \times \left( 0.13^{3/2} \times 0.2^{1/2} + 0.23^{3/2} \times 0.03^{1/2} \right) \approx 18.4 \text{ MPa} $$
Thus, solid‑solution strengthening contributes approximately 18.4 MPa.
Multi‑Mechanism Coupling Model
To verify the reliability of the experimental data, I constructed a comprehensive strengthening model for the 0.35% core sample by combining the contributions from the 0.15% sample’s yield strength and the individual strengthening increments:
$$ \sigma_{0.35} = \sigma_{0.15} + \Delta \sigma_y + \Delta \sigma_p + \Delta \sigma_{ss} + \varepsilon $$
where ε accounts for rare earth segregation‑induced softening, taken as −37.7 MPa. Substituting the values:
$$ \sigma_{0.35} = 367 + 25.2 + 4.1 + 18.4 – 37.7 = 377 \text{ MPa} $$
This calculated strength exactly matches the measured yield strength of 377 MPa, confirming that the combination of fine‑grain strengthening, phase transformation strengthening, solid‑solution strengthening, and the softening effect due to rare earth segregation provides a consistent explanation for the observed mechanical behavior of the lost foam castings.
Fracture Surface Analysis
I examined the tensile fracture surfaces of the 0.15% and 0.35% rare earth silicoferrite specimens using SEM. The macroscopic fracture surfaces displayed a rough, fibrous appearance with shear lips, indicating significant plastic deformation prior to failure. This is characteristic of ductile fracture. At higher magnification, I observed numerous dimples, some equiaxed and others elongated, along with tear ridges and secondary cracks. The presence of cleavage steps and river patterns suggests that localized brittle fracture occurred as well. Therefore, the overall fracture mechanism of the rare‑earth‑treated lost foam castings can be described as a mixed ductile‑brittle mode.
Figure 13 in the original paper shows SEM images of the fracture surfaces. The 0.15% specimen exhibited a higher density of dimples and more pronounced tear ridges, consistent with its greater elongation (15.17%). The 0.35% specimen showed finer dimples and more cleavage facets, reflecting the higher strength but somewhat reduced ductility. The micro‑void coalescence mechanism, involving the nucleation, growth, and coalescence of micro‑voids around inclusions or second‑phase particles, is the dominant fracture process.
The morphology of inclusions in the fracture surfaces changed with rare earth addition. In the baseline (0%) specimens, elongated MnS inclusions were visible, acting as crack initiation sites. In rare‑earth‑treated specimens, the inclusions became spherical and finely dispersed, reducing stress concentration and improving the resistance to crack propagation. These observations support the beneficial role of rare earth silicoferrite in enhancing the overall mechanical integrity of lost foam castings.
Effect of Rare Earth Silicoferrite on Solidification and Defect Control in Lost Foam Castings
The lost foam casting process inherently involves complex gas‑liquid‑solid interactions due to foam decomposition. The pattern vaporization creates a gaseous gap that affects heat transfer and mold filling. In my experiments, the addition of rare earth silicoferrite influenced not only the microstructure but also the solidification behavior and defect formation. Rare earth elements, being strong deoxidizers and desulfurizers, help to reduce gas porosity and inclusion clusters. The finer microstructure and improved inclusion morphology contributed to a more homogeneous stress distribution, as quantified by reduced stress concentration factors in the castings. The average hardness and mechanical properties of the lost foam castings improved significantly with increasing rare earth silicoferrite content, demonstrating that this alloying approach is effective for producing high‑performance 45 steel castings.
To further illustrate the role of rare earth silicoferrite, I summarize the key mechanisms in Table 7.
| Mechanism | Contribution / MPa | Physical origin |
|---|---|---|
| Fine‑grain strengthening | ≈ 25.2 | Increased grain boundary density obstructs dislocation slip |
| Phase transformation strengthening | ≈ 4.1 | Reduced pearlite interlamellar spacing |
| Solid‑solution strengthening | ≈ 18.4 | Si and Ce atoms dissolve in ferrite |
| Rare earth segregation softening | ≈ −37.7 | Local grain boundary softening due to RE segregation |
| Net increase (0.15% → 0.35%) | 10 | Measured yield strength increase from 367 to 377 MPa |
The above table clearly shows that the net strengthening is the result of competing mechanisms. While grain refinement and solid‑solution strengthening are positive, rare earth segregation at grain boundaries may locally reduce the yield stress. Nevertheless, the overall effect is a net improvement in strength.
In addition to mechanical properties, I observed that the macrostructure of the castings became more uniform with increasing rare earth silicoferrite content. The centerline segregation and shrinkage porosity were reduced, likely because the refined equiaxed grains promote a more dispersed solidification front and better feeding. This is particularly beneficial for large‑section lost foam castings, where solidification shrinkage can otherwise lead to internal defects.
Practical Implications for Lost Foam Casting Production
The findings from this investigation have important practical implications for foundries producing 45 steel components using the lost foam casting process. By adding a controlled amount of rare earth silicoferrite—typically in the range of 0.15% to 0.35%—it is possible to achieve significant improvements in both strength and hardness without the need for expensive post‑casting heat treatments. This is particularly advantageous for large or complex lost foam castings where heat treatment may be impractical or costly. The refinement of the microstructure also enhances the machinability and surface finish of the final components.
Another practical benefit is the improved consistency of mechanical properties across different sections of the casting. In conventional 45 steel lost foam castings, the core often exhibits coarse grains and lower hardness due to slower cooling, while the edges are finer and harder. This gradient is undesirable for engineering components that require uniform performance. My results show that increasing the rare earth silicoferrite content to 0.35% narrows the property gradient between the core and the edge, making the casting more reliable under service loads. This is clearly visible in the grain size data where the difference between core and edge grain sizes reduced from about 41.6 μm (0%) to 23.9 μm (0.35%), and in hardness where the difference between core and edge decreased from 0.53 HRA to 0.67 HRA.
The adoption of rare earth silicoferrite as a grain refiner is economically feasible because the addition levels are low and the alloy itself is relatively inexpensive compared to other microalloying elements. Furthermore, the purification effect can improve the recycling rate of scrap steel, as rare earth elements help to immobilize impurities. Therefore, this approach aligns with the trend toward sustainable and energy‑efficient manufacturing.
Conclusions
Based on my systematic investigation of rare earth silicoferrite additions in 45 steel manufactured by the lost foam casting process, I draw the following conclusions:
1) In the lost foam castings produced without any heat treatment and allowed to cool naturally in the mold, the addition of rare earth silicoferrite significantly refines the grains in all regions—core, middle, and edge. The core grain size decreased from 232.62 μm to 153.97 μm when the addition increased from 0% to 0.35%, representing a reduction of about 35%. This refinement is attributed to the enhanced nucleation rate and restricted grain growth caused by rare earth elements.
2) The rare earth silicoferrite content can effectively refine the grain size of the castings, improve the cleanliness of the molten steel, and modify the morphology and distribution of inclusions. Consequently, the average Rockwell hardness of the 45 steel lost foam castings increases. With 0.35% addition, the core hardness reached 53.43 HRA, which is about 13% higher than that of the rare‑earth‑free casting.
3) The multiple strengthening mechanisms—fine‑grain, phase transformation, and solid‑solution strengthening—act together to enhance the tensile properties of the lost foam castings. The ultimate tensile strength increased from 373 MPa (0%) to 410 MPa (0.35%) in the core region, while the yield strength increased from 342 MPa to 377 MPa. Fracture surface analysis reveals a mixed ductile‑brittle fracture mechanism, where the refined microstructure and spheroidized inclusions contribute to improved toughness
and reliability.
Overall, this research demonstrates that rare earth silicoferrite microalloying is a practical and effective strategy for enhancing the performance of 45 steel components produced by lost foam casting. The findings provide valuable guidance for foundries seeking to produce high‑strength, low‑alloy steel castings with uniform properties and excellent mechanical integrity.