In my research, I focused on the improvement of mechanical properties of ductile iron produced by lost foam casting through the application of mechanical vibration. Lost foam casting is a well-developed special casting technology, but due to the use of dry sand to fill the flask, the heat dissipation rate is relatively slow, resulting in coarse dendrites and lower mechanical properties. To solve this problem, I introduced mechanical vibration during the solidification process of the melt. The vibration can break dendrites, promote the dissociation of crystal nuclei, and refine the microstructure. In this paper, I systematically studied the effects of vibration frequency and amplitude on the microstructure, tensile properties, low-temperature impact toughness, and wear resistance of QT400-18 ductile iron prepared by lost foam casting. The experimental results clearly show that appropriate mechanical vibration can significantly refine graphite nodules, ferrite grains, and pearlite lamellae, thereby improving the overall mechanical performance of the alloy.
Experimental Materials and Procedures
I prepared the QT400-18 ductile iron using pig iron, scrap steel, and ferrosilicon as raw materials. The chemical compositions of the raw materials are summarized in Table 1.
| Element | C | Si | Mn | S | P | Fe |
|---|---|---|---|---|---|---|
| Pig iron | 4.07 | 0.72 | 0.06 | 0.012 | 0.026 | Balance |
| Scrap steel | 0.05 | 0.01 | 0.26 | 0.005 | 0.014 | Balance |
| Ferrosilicon | 0.035 | 72.15 | — | 0.003 | 0.008 | Balance |
For spheroidization and inoculation, I used 1.6 wt.% FeSiCaMgRE as the nodulizer and 0.5 wt.% CaBa-FeSi as the inoculant. Their compositions are given in Table 2.
| Material | Si | Al | Mg | Ca | RE | Ba | Fe |
|---|---|---|---|---|---|---|---|
| Nodulizer | 45.96 | 0.72 | 5.69 | 1.09 | 0.89 | — | Balance |
| Inoculant | 72.99 | 1.15 | — | 1.73 | — | 2.25 | Balance |
The lost foam casting process was carried out on a mechanical vibration platform. I selected vertical vibration because it exerts the most significant influence on the compaction of quartz sand and the solidification behavior of the alloy. The vibration frequency was controlled by a frequency converter, and the amplitude was adjusted by changing the angle of the eccentric blocks. The experimental setup is illustrated in Figure 1.
In the first series of experiments, the amplitude was fixed at 1.5 mm, and the vibration frequency was set to 0, 30, 40, and 50 Hz. In the second series, the vibration frequency was fixed at 40 Hz, and the amplitude was set to 0, 1, 1.5, and 2 mm. After pouring, the negative pressure was maintained at 0.06 MPa for 10 minutes. Standard tensile specimens were machined from the castings, and the tensile tests were performed at a displacement rate of 1 mm/min. The elongation was calculated using the following formula:
$$ A = \frac{L_u – L_0}{L_0} \times 100\% $$
where \(A\) is the percentage elongation after fracture, \(L_u\) is the final gauge length, and \(L_0\) is the original gauge length. The microstructure was characterized by optical microscopy after etching with 4% nitric acid alcohol solution. For the low-temperature impact tests, standard Charpy V-notch specimens with dimensions of 55 mm × 10 mm × 10 mm were used, and the tests were conducted at 20, 0, -20, -40, -50, and -60 °C. The wear tests were carried out on a reciprocating friction and wear tester with a tungsten carbide ball of 6 mm diameter, under a load of 50 N, frequency of 2 Hz, sliding distance of 10 mm, and wear time of 30 minutes.
Effect of Vibration Frequency on Microstructure and Tensile Properties
Graphite morphology
With the amplitude fixed at 1.5 mm, I prepared ductile iron samples at 0, 30, 40, and 50 Hz. The graphite morphology changed significantly with increasing vibration frequency. In the sample without vibration, the graphite nodules were relatively large and irregularly shaped, and the graphite content was low. When the vibration frequency increased to 30 Hz, more graphite nodules appeared and their average size decreased. At 40 Hz, the graphite nodules became finer, more uniform, and the spheroidization rate reached a high level. However, at 50 Hz, excessive vibration destroyed the growth environment of spherical graphite, and the spheroidization rate decreased even though the graphite size continued to decrease slightly.
I quantitatively analyzed the graphite characteristics using image analysis software. The results are summarized in Table 3.
| Sample | Frequency (Hz) | Spheroidization rate (%) | Spheroidization grade | Average graphite diameter (μm) | Graphite content (%) |
|---|---|---|---|---|---|
| a | 0 | 78.8 | 4 | 33.86 | 13.58 |
| b | 30 | 83.1 | 3 | 29.61 | 15.21 |
| c | 40 | 85.9 | 3 | 27.09 | 16.26 |
| d | 50 | 82.3 | 3 | 26.44 | 16.41 |
From Table 3, it is evident that the spheroidization rate increased from 78.8% to 85.9% when the vibration frequency rose from 0 to 40 Hz, while the average graphite diameter decreased from 33.86 μm to 27.09 μm. The graphite content also increased from 13.58% to 16.26%. At 50 Hz, the spheroidization rate dropped to 82.3%, but the graphite content still slightly increased to 16.41%. This indicates that moderate vibration promotes graphite nucleation and growth in a spherical manner, but excessive vibration can destabilize the growth front and reduce nodularity.
The mechanism behind this behavior can be explained by the interaction between vibration and nucleation. During solidification, mechanical vibration introduces periodic tension and compression into the melt, which reduces the wetting angle at the nucleation interface and promotes heterogeneous nucleation. Vibration also detaches the initially formed dendritic fragments from the mold wall and the solidification front, turning them into new substrates for graphite precipitation. Moreover, vibration accelerates the formation of the austenite shell around the graphite nuclei, which protects the graphite from irregular growth and favors the formation of well-rounded nodules.
Ferrite grain refinement
Figure 2 shows the matrix microstructures of ductile iron prepared at different vibration frequencies. As the frequency increased, the ferrite content gradually increased, and the ferrite grain size became finer. I used the three-circle intercept method to measure the ferrite grain size according to the standard GB/T 6394-2002. The volume fraction of ferrite in the matrix was calculated from the area fraction of ferrite in the images. The ferrite volume fraction can be expressed as:
$$ V_{V\alpha} = (1 – f_g) \times f_{\alpha} $$
where \(f_g\) is the graphite content and \(f_{\alpha}\) is the area fraction of ferrite in the etched matrix. The calculated values for the four samples are presented in Table 4.
| Sample | Frequency (Hz) | Ferrite area in matrix (%) | Graphite content (%) | Ferrite volume fraction (%) |
|---|---|---|---|---|
| a | 0 | 79.01 | 13.58 | 68.28 |
| b | 30 | 83.23 | 15.21 | 70.57 |
| c | 40 | 85.29 | 16.26 | 71.42 |
| d | 50 | 86.15 | 16.41 | 72.01 |
The mean intercept length of the ferrite grains was calculated using the following equation:
$$ \bar{l}_{\alpha} = \frac{V_{V\alpha} L}{M N_{\alpha}} $$
where \(L\) is the total test line length (500 mm), \(M\) is the magnification, and \(N_{\alpha}\) is the number of intersections with ferrite grain boundaries. The calculated intercept lengths and the corresponding ASTM grain size numbers are given in Table 5.
| Sample | Frequency (Hz) | Average intersections, \(N_{\alpha}\) | Mean intercept, \(\bar{l}_{\alpha}\) (mm) | Grain size number, \(G\) |
|---|---|---|---|---|
| a | 0 | 58 | 0.029598 | 6.87 |
| b | 30 | 65 | 0.026951 | 7.14 |
| c | 40 | 72 | 0.024843 | 7.37 |
| d | 50 | 74 | 0.024542 | 7.41 |
The results clearly show that mechanical vibration refines the ferrite grains. The grain size number increased from 6.87 for the non-vibrated sample to 7.41 for the sample vibrated at 50 Hz. This refinement is attributed to the fragmentation of primary dendrites and the increased number of nucleation sites caused by vibration-induced convection and cavitation. The finer ferrite grains contribute to higher strength and toughness.
Pearlite morphology
I also observed the pearlite morphology in the as-cast samples. Without vibration, pearlite existed mainly as coarse lamellae. With increasing vibration frequency, the lamellar spacing decreased, and short-rod-like or granular pearlite gradually appeared. At 40 Hz, the pearlite became much finer, and at 50 Hz, more granular pearlite formed. I measured the pearlite lamellar spacing, and the results are shown in Figure 3.
$$ \text{Pearlite spacing} = \frac{1}{n} \sum_{i=1}^{n} d_i $$
The measured average lamellar spacings are listed in Table 6.
| Frequency (Hz) | 0 | 30 | 40 | 50 |
|---|---|---|---|---|
| Lamellar spacing (μm) | 0.34 | 0.27 | 0.22 | 0.29 |
The lamellar spacing first decreased and then increased with rising frequency. The minimum spacing of 0.22 μm was obtained at 40 Hz. Vibration promotes carbon diffusion and accelerates the breakdown of cementite, resulting in the spheroidization of pearlite. However, at excessive vibration intensity, the perturbation may disturb the cooperative growth of ferrite and cementite, leading to a slight coarsening of the lamellae.
Tensile properties
Tensile tests were performed on the samples prepared at different vibration frequencies. The stress-strain curves are displayed in Figure 4. The average values of tensile strength and elongation are summarized in Table 7.
| Frequency (Hz) | Tensile strength (MPa) | Elongation (%) |
|---|---|---|
| 0 | 419.5 | 18.3 |
| 30 | 446.5 | 21.1 |
| 40 | 457.1 | 22.2 |
| 50 | 452.3 | 20.6 |
Compared with the non-vibrated sample, the tensile strength increased by 6.4%, 9.0%, and 7.8% for 30, 40, and 50 Hz, respectively. The elongation also increased from 18.3% to 22.2% at 40 Hz. The improvement is attributed to the combined effects of finer graphite, higher nodularity, refined ferrite grains, and finer pearlite lamellae. At 50 Hz, the deterioration of nodularity and the increase in irregular graphite caused a slight reduction in both strength and elongation.
The tensile fracture surfaces were examined by scanning electron microscopy. The non-vibrated sample showed coarse graphite nodules and some cleavage facets with river patterns. With increasing frequency, the number of dimples around graphite nodules increased, and the cleavage features gradually disappeared. At 40 Hz, the fracture surface exhibited a fully ductile morphology with fine dimples. At 50 Hz, some irregular graphite and a few cleavage facets reappeared. This observation is consistent with the tensile results.
I also performed three-dimensional reconstruction of the tensile fracture surfaces using a confocal laser microscope. The height difference across the fracture surface can reflect the degree of plastic deformation. The height differences are listed in Table 8.
| Frequency (Hz) | 0 | 30 | 40 | 50 |
|---|---|---|---|---|
| Height difference (μm) | 711.9 | 838.7 | 934.2 | 825.3 |
The surface roughness parameter \(S_a\) was calculated from the reconstructed surfaces using:
$$ S_a = \frac{1}{A} \iint_A |z(x,y)| \, dx \, dy $$
where \(A\) is the measured area and \(z(x,y)\) is the height function. The trend of \(S_a\) with vibration frequency is shown in Figure 5. The roughness increased first and then decreased, which correlates well with the elongation results.
Effect of Amplitude on Microstructure and Tensile Properties
Graphite morphology under different amplitudes
In the second series of experiments, the vibration frequency was fixed at 40 Hz, and the amplitude was varied from 0 to 2 mm. The graphite morphology improved progressively with increasing amplitude. The quantitative results are given in Table 9.
| Sample | Amplitude (mm) | Spheroidization rate (%) | Spheroidization grade | Average graphite diameter (μm) | Graphite content (%) |
|---|---|---|---|---|---|
| a | 0 | 78.8 | 4 | 33.86 | 13.58 |
| b | 1 | 81.3 | 3 | 30.79 | 14.66 |
| c | 1.5 | 85.9 | 3 | 27.09 | 16.26 |
| d | 2 | 86.2 | 3 | 26.87 | 16.32 |
The graphite content increased from 13.58% to 16.32% when the amplitude increased from 0 to 2 mm. The average graphite diameter decreased from 33.86 μm to 26.87 μm, and the spheroidization rate increased from 78.8% to 86.2%. Larger amplitudes generate stronger agitation in the melt, which improves the homogeneity of temperature and solute distribution, and enhances the fragmentation of dendrites. This results in a larger number of graphite nuclei and a finer graphite size.
Ferrite refinement under different amplitudes
The ferrite volume fraction and grain size parameters for different amplitudes are summarized in Tables 10 and 11.
| Amplitude (mm) | Ferrite area in matrix (%) | Graphite content (%) | Ferrite volume fraction (%) |
|---|---|---|---|
| 0 | 79.01 | 13.58 | 68.28 |
| 1 | 82.94 | 14.66 | 70.78 |
| 1.5 | 85.29 | 16.26 | 71.42 |
| 2 | 85.98 | 16.32 | 71.95 |
| Amplitude (mm) | Average intersections, \(N_{\alpha}\) | Mean intercept, \(\bar{l}_{\alpha}\) (mm) | Grain size number, \(G\) |
|---|---|---|---|
| 0 | 58 | 0.029598 | 6.87 |
| 1 | 65 | 0.027416 | 7.09 |
| 1.5 | 72 | 0.024843 | 7.37 |
| 2 | 75 | 0.024023 | 7.47 |
The ferrite grain size number increased from 6.87 to 7.47 when the amplitude increased from 0 to 2 mm. The refinement of ferrite grains enhances the strength and toughness of the ductile iron by increasing the grain boundary area and impeding dislocation movement.
Pearlite refinement under different amplitudes
The pearlite lamellar spacing decreased with increasing amplitude, as shown in Table 12.
| Amplitude (mm) | 0 | 1 | 1.5 | 2 |
|---|---|---|---|---|
| Lamellar spacing (μm) | 0.34 | 0.31 | 0.22 | 0.19 |
At 2 mm amplitude, the pearlite became very fine, and the content of granular pearlite increased. The increase in vibration amplitude accelerates the heat transfer and reduces the diffusion distance of carbon atoms, leading to a finer pearlite structure.
Tensile properties under different amplitudes
The tensile results are summarized in Table 13.
| Amplitude (mm) | Tensile strength (MPa) | Elongation (%) |
|---|---|---|
| 0 | 419.5 | 18.3 |
| 1 | 440.3 | 20.8 |
| 1.5 | 457.1 | 22.2 |
| 2 | 462.8 | 22.8 |
At 2 mm amplitude, the tensile strength reached 462.8 MPa and the elongation reached 22.8%, which are the highest values among all the tested conditions. The improvement from 1.5 mm to 2 mm was relatively small, suggesting a saturation effect in the vibration refinement. Further increasing the amplitude may not bring additional benefits and could introduce defects.
The fracture surfaces of the tensile samples at different amplitudes showed a transition from mixed fracture to fully ductile fracture. The 2 mm amplitude sample exhibited the most uniform and deepest dimples. The three-dimensional surface roughness also increased with amplitude, confirming the enhanced plastic deformation capability.
Thermodynamic Analysis of Graphite Precipitation
I also analyzed the thermodynamic characteristics of the solidification process. The eutectic transformation temperature of the vibrated samples was lower than that of the non-vibrated samples. This means that with mechanical vibration, the eutectic reaction starts at a lower undercooling, which reduces the atomic diffusion rate and shortens the diffusion time. Consequently, the critical nucleus radius of graphite becomes smaller. The relationship between the critical nucleus radius \(r_c\) and the undercooling \(\Delta T\) is given by:
$$ r_c = \frac{2 \sigma_{SL} T_m}{\Delta H_v \Delta T} $$
where \(\sigma_{SL}\) is the solid-liquid interfacial energy, \(T_m\) is the melting temperature, and \(\Delta H_v\) is the volumetric latent heat. Because vibration increases the effective undercooling, the critical nucleus radius decreases, promoting the formation of more graphite nuclei.
In addition, the growth of graphite is controlled by the diffusion of carbon atoms. During the growth stage, carbon atoms in the melt diffuse toward the graphite nucleus and are consumed. The growth rate of graphite can be expressed as:
$$ \frac{dr}{dt} = D \frac{C_m – C_i}{r (C_g – C_i)} $$
where \(D\) is the diffusion coefficient of carbon in the melt, \(C_m\) is the carbon concentration in the bulk liquid, \(C_i\) is the carbon concentration at the interface, and \(C_g\) is the carbon concentration in graphite. Vibration enhances convection and homogenizes the liquid, thereby increasing the concentration gradient and promoting the uniform growth of spherical graphite. The austenite shell then quickly encloses the graphite, limiting further carbon supply and ensuring a fine and regular graphite structure.
Low-Temperature Impact Properties
To evaluate the toughness of the ductile iron, I selected the samples prepared at 40 Hz and 2 mm amplitude and compared them with the non-vibrated samples. Charpy impact tests were conducted at 20, 0, -20, -40, -50, and -60 °C. The impact energy values are shown in Table 14.
| Temperature (°C) | Without vibration (J) | With vibration (J) | ||||
|---|---|---|---|---|---|---|
| Specimen 1 | Specimen 2 | Average | Specimen 1 | Specimen 2 | Average | |
| 20 | 18.31 | 18.09 | 18.23 | 20.07 | 19.79 | 19.86 |
| 0 | 16.34 | 16.12 | 16.17 | 17.36 | 17.20 | 17.18 |
| -20 | 14.01 | 14.15 | 14.16 | 15.61 | 15.39 | 15.46 |
| -40 | 12.31 | 12.21 | 12.19 | 13.35 | 13.08 | 13.17 |
| -50 | 8.33 | 8.10 | 8.14 | 11.66 | 11.99 | 11.74 |
| -60 | 4.37 | 4.49 | 4.48 | 7.61 | 7.85 | 7.67 |
The impact energy decreased with decreasing temperature for both conditions. The vibrated samples always showed higher impact energy than the non-vibrated ones. At 20 °C, the impact energy increased from 18.23 J to 19.86 J, an improvement of 1.13 J. The ductile-to-brittle transition temperature was about -40 °C for the non-vibrated sample and -50 °C for the vibrated sample. This indicates that mechanical vibration not only improves the strength but also delays the ductile-to-brittle transition, allowing the material to be used at lower service temperatures.
The fracture surfaces of the impact specimens were examined. At 20 °C, the fracture surfaces were covered with fine dimples around graphite nodules. As the temperature decreased, cleavage facets gradually appeared. At -40 °C, the non-vibrated sample showed a large number of cleavage facets, whereas the vibrated sample still exhibited a mixed morphology. At -60 °C, both samples failed in a brittle manner, but the vibrated sample still had a slightly higher impact energy. The three-dimensional reconstruction of the fracture surfaces also showed a higher height difference for the vibrated samples, indicating greater plastic deformation.
The surface roughness \(S_a\) of the impact fracture surfaces was measured at each temperature, and the results are presented in Table 15.
| Temperature (°C) | 20 | 0 | -20 | -40 | -50 | -60 |
|---|---|---|---|---|---|---|
| Without vibration \(S_a\) (μm) | 121.4 | 113.8 | 102.6 | 88.3 | 67.5 | 51.2 |
| With vibration \(S_a\) (μm) | 134.7 | 126.9 | 119.8 | 107.4 | 89.6 | 68.4 |
The higher roughness values for the vibrated samples indicate more energy absorption during fracture, which aligns with the higher impact energy.
The better low-temperature impact toughness of the vibrated ductile iron is mainly due to the finer ferrite grains and the higher nodularity of graphite. Finer grains provide more grain boundaries that can impede crack propagation and reduce the cleavage fracture stress. Well-nodular graphite produces less stress concentration than irregular graphite, thereby reducing the probability of crack initiation. As a result, the vibrated samples can maintain higher impact energy at lower temperatures.
Friction and Wear Properties
I also investigated the wear resistance of the ductile iron prepared with optimal vibration parameters (40 Hz, 2 mm) and compared it with the non-vibrated sample. The wear tests were performed under a load of 50 N, frequency of 2 Hz, sliding distance of 10 mm, and wear time of 30 minutes.
Hardness before and after wear
The Brinell hardness values are listed in Table 16.
| Condition | Before wear hardness (HB) | Average before wear (HB) | After wear hardness (HB) | Average after wear (HB) |
|---|---|---|---|---|
| Without vibration | 148.3, 146.7, 152.2 | 149.1 | 153.4, 154.3, 158.5 | 155.4 |
| With vibration | 141.6, 147.2, 144.3 | 144.4 | 154.4, 159.7, 161.3 | 158.5 |
The non-vibrated sample had a slightly higher initial hardness (149.1 HB) than the vibrated sample (144.4 HB), because the non-vibrated sample contained more pearlite. After the wear test, both samples showed an increase in hardness due to work hardening. The vibrated sample exhibited a larger hardness increase (14.1 HB) compared with the non-vibrated sample (6.3 HB). The more stable work-hardened layer in the vibrated sample can be attributed to the higher graphite nodularity and more uniform stress distribution, which prevents early cracking of the hardened layer.
Wear loss and friction coefficient
The weight loss results are listed in Table 17.
| Condition | Weight loss (g) | Average weight loss (g) |
|---|---|---|
| Without vibration | 0.04331, 0.04357, 0.04418 | 0.04369 |
| With vibration | 0.04145, 0.04261, 0.04167 | 0.04191 |
The vibrated sample exhibited a lower average weight loss than the non-vibrated sample. The friction coefficient curves are shown in Figure 6. The average friction coefficient of the vibrated sample was 0.803, while that of the non-vibrated sample was 0.912. The lower friction coefficient of the vibrated sample is mainly due to the higher graphite content and better graphite distribution, which provides a continuous lubricating film on the worn surface. The regular spherical graphite is less likely to create stress concentrations, so the hardened layer remains intact longer and reduces the wear rate.
The worn surface morphologies were observed by scanning electron microscopy. In the running-in stage, both samples showed plastic deformation and oxide layers. The non-vibrated sample had more severe surface damage and more iron oxide particles. In the steady-state wear stage, the non-vibrated sample exhibited deep grooves and spalling pits, while the vibrated sample had a relatively smooth worn surface with shallow grooves and a stable hardened layer.
Three-dimensional reconstruction of the worn surfaces was performed using confocal microscopy. The maximum wear depth of the non-vibrated sample was larger than that of the vibrated sample. The height differences along the wear track are listed in Table 18.
| Condition | Maximum wear depth (μm) | Average roughness of wear track (μm) |
|---|---|---|
| Without vibration | 72.3 | 18.7 |
| With vibration | 55.8 | 13.9 |
These results confirm that mechanical vibration improves the wear resistance of lost foam casting QT400-18 ductile iron.
Conclusion
From my systematic study on the effect of mechanical vibration on the microstructure and properties of QT400-18 ductile iron produced by lost foam casting, I have drawn the following conclusions:
1. Mechanical vibration during lost foam casting significantly refines the graphite structure. With increasing vibration frequency from 0 to 40 Hz, the graphite nodularity and content increase while the average graphite size decreases. At 50 Hz, the nodularity decreases due to excessive vibration disturbing the spherical growth environment.
2. The ferrite grains are remarkably refined by mechanical vibration. The ferrite grain size number increases from 6.87 for the non-vibrated sample to 7.47 for the sample prepared at 40 Hz and 2 mm amplitude. The pearlite lamellar spacing decreases from 0.34 μm to 0.19 μm, and more granular pearlite forms.
3. The optimal tensile properties are obtained at a vibration frequency of 40 Hz and an amplitude of 2 mm. The tensile strength reaches 462.8 MPa and the elongation reaches 22.8%, which are respectively 43.3 MPa and 4.5% higher than those of the non-vibrated sample.
4. The low-temperature impact toughness is improved by mechanical vibration. At 20 °C, the impact energy increases from 18.23 J to 19.86 J. The ductile-to-brittle transition temperature is lowered from -40 °C to -50 °C, enabling the material to be used at lower temperatures.
5. The wear resistance of the vibrated sample is better than that of the non-vibrated sample. The average weight loss decreases from 0.04369 g to 0.04191 g, and the friction coefficient decreases from 0.912 to 0.803 under the tested conditions. The finer and more spherical graphite in the vibrated sample provides a more stable lubricating film and reduces the tendency for crack initiation, leading to improved wear resistance.
In conclusion, the application of mechanical vibration to lost foam casting is a simple, economical, and environmentally friendly method to enhance the microstructure and mechanical properties of QT400-18 ductile iron. These findings provide valuable guidance for optimizing the lost foam casting process for ductile iron components used in demanding engineering applications.