2. Electron Parameters for Designation of the Al–9Si–Zr/Mo Alloys
The electron parameter, Mki, used in the compositional optimizations of Al alloys had been obtained from the electronic approach to alloys design.12–17) Mki is an s-orbital energy level existing above the Fermi energy level of an iAl18 cluster, containing an alloying element i and its surrounding Al atoms.15) So far, various parameters have been proposed to describe the alloying behavior, but the electronegativity and the atomic radius of elements were chosen here because they represent well the nature of chemical bonds between the atoms in the solids.15) There was a good correlation between the electronegativity or atomic radius and the Md parameter of d-orbital energy level on transition metals, as mentioned in section 1. For s, p simple metals like Al, the d-orbital energy level was no longer effective in the transition metal-based alloys. The Mki, which is an s-orbital energy level exciting above the Fermi energy level and was obtained by the discrete variational Xα30) (DV-Xα) cluster calculation. Consequently, alloying effects are inevitably involved in Mki.15) It is well known that the energy level obtained by the DV-Xα cluster calculation represents the electronegativity itself.14,15) The Mki value decreased with increasing electronegativity, whereas it increased with an increasing atomic radius of elements.14,15) Besides, the p-orbital energy level may be considered instead of the s-orbital energy level, but a spherical symmetrical s-orbital was probably better than a directional p-orbital for investigating the mechanical properties of Al alloys.15)
Most of the elements have a higher Mki value than Al except for the Si, Zn, Ga, and Ge. For alloys, two kinds of the average values of the s-orbital energy level, Mkt and ΔMk values were defined by taking the compositional average, using eqs. (1) and (2).
| \begin{equation}
Mk_{t} = \varSigma x_{i}Mk_{i}
\end{equation}
| (1) |
| \begin{equation}
\varDelta Mk = \varSigma x_{i}|Mk_{i} - Mk_{m}|
\end{equation}
| (2) |
Where
xi is the mole fraction of component
I in the alloy,
Mki, and
Mkm are the
Mk value for components
I and Al, respectively. In the case when all the elements in the alloy have a higher (or lower)
Mki value than Al, these two averages differ only by a constant bias of
Mkm value of 3.344 and hence there is no essential difference between them. However, when the elements with higher and lower
Mki values are mixed in the alloy, the two average values have different meanings.
15)
In this study, Si was selected as the second element will be also described focusing on the Mki value. As can be seen from the eq. (2), the alloying effect is reduced when an alloying element has a larger Mki value. The Si with a smaller Mki value than Al, which is suitable as an additive element. However, the addition of Si with a small Mki value than Al causes tensile strain in the crystal lattice of Al. It is necessary to add an element having a larger Mki value than Al to maintain the stability of the crystal lattice. Both Mo and Zr have a larger Mki value than Al.15) Besides, the atomic radius of Mo and Zr is close to Al. Iron is one of the most detrimental impurities of the Al alloys since it forms brittle and complex intermetallic compounds needle-like that significantly affect the mechanical properties.31) Manganese is the most common alloying element to suppress the development of needle particles by promoting the formation of a thermodynamically stable α-Al.31,32) Generally, alloying with a Fe:Mn ratio lower than 2 is recommended to encourage the α-Al precipitation.31,33) The addition amount of 9Si, 0.2Mo, and 0.2Zr is the upper limit of the component content in Castasil-37 alloy.7)
3. Experimental Methods
The pure Al, Al–Si, Al–Mn, Al–Zr, and Al–Mo master alloys were used for preparing experimental alloys. To avoid any masking effects or interactions with additional elements, neither Sr nor Na was added as modifier agents. In the production of ingots, it should be produced by the die-casting method according to the purpose of this research. However, to use ingots of simple shapes that do not require high precision, a low-cost gravity casting method was adopted. The melting temperatures were kept at 990–1007 K. Then, the melt was poured into a steel mold preheated at 426–488 K. The size of the ingots was 195 mm × 23 mm × 39 mm.
Specimens for microstructural observations were taken from each ingot at a position one-half of its length, width, and height. Specimens for microstructural observations were performed using optical microscopy (OM) and electron probe microanalyzer (EPMA), which were equipped with Wavelength Dispersive X-Ray Spectroscopy (WDS). The image-pro software was employed to measure the volume fractions of the dendrite (DA) areas and the inter-dendritic (ID) areas and the secondary dendrite arm spacings (SDAS) of the experimental alloys. At least 15 OM images were used to calculate the average volume fractions of DA and ID. The SDAS was measured by using OM images with an accurate scale bar. The line intercept method34) was used to measure the SDAS, and at least 100 dendrites were taken to determine the SDAS. Differential thermal analysis (DTA) with a heating-cooling rate of 5 K·min−1 was used to investigate solidification behaviors, and the liquidus temperature and the thermal effect of the exothermic cooling curve were also measured. Each test was repeated at least two times to ensure reproducibility of the detected temperature. The solidification path was measured by using point analysis in mass% from DA to ID with an interval of 50 µm. The analysis point analysis results were plotted on a triangular coordinate graph at 25 points, and the figure was shown as the solidification path.
The two-dimensional local number (LN2D) is defined as the number of centers of gravity of second phase particles in the measuring circle.35) When particles have a uniform random arrangement in two dimensions, the relative frequency distribution of LN2D exhibits the modified Poisson distribution for a random point field, and the probability for the LN2D is expressed by
| \begin{equation}
P(\textit{LN2D} = k + 1) = \frac{7^{k}}{k!}\mathit{exp}(-7)\quad k = 0, 1, 2\cdots
\end{equation}
| (3) |
The average and variance of this expected distribution are 7. On the base of this definition, an local number of 7 corresponds to the number density of all particles, when the spatial distribution is uniformly random. The average of LN2D (LN2Dav) and variance of LN2D (LN2Dvar) were used to evaluate the spatial distribution randomness of intermetallic compounds (IMCs).
Tensile test specimens with a diameter of 6 mm and a gauge length of 80 mm were machined from the cast ingots. The tensile tests were performed at an initial strain rate of 1.0 mm·min−1 at 293 K with a testing machine (DCS-R-5000, SHIMADZU, Japan). The tensile strain was accurately measured by using an extensometer until the necking. Four tensile tests were performed for each alloy, and a moderate level of tensile curve was selected. After the tensile tests, the fracture surfaces were observed using the scanning electron microscope to explain the alloy deformation characteristics. The Young’s modulus (E) in the tensile test was determined by measuring the slope of the stress-strain in the tensile curve when the strain was less than 0.2%. The Lankford value (r) was determined by measuring the width and length dimensions within the gauge length of the specimen before and after the tensile deformation and calculating as follows:
| \begin{equation}
r = \mathit{ln}(b/b_{0})/\mathit{ln}(b_{0}l_{0}/bl)
\end{equation}
| (4) |
Where
l0 and
b0 are the length and width dimensions of the specimen before the tensile deformation.
l and
b are the length and width dimensions of the specimen after the tensile deformation.
To reveal the mechanism of ductile deformation, 1% plastic strain by cold rolling was applied to the base and 0.2Mo+0.2Zr addition alloy specimens, and the specimens were observed by transmission electron microscopy (TEM, JEM-2010, JEOL, Japan). The dislocation density was also measured by the equal thickness fringe method using areas with a thickness of more than 100 nm in TEM specimens.
The nanoindentation experiments were conducted on the primary and eutectic α-Al phases of the base and 0.2Mo+0.2Zr addition alloys by an instrument (ENT-1100a, Elionix, Japan) using a Berkovich diamond indenter at room temperature. The measurements were performed at the holding load of 3 mN (load controlled), loading up in the 30 s, holding for 20 s at the peak load, and unloading in the 30 s on each specimen. The nanoindentation hardness (HIT) was calculated using the Oliver-Pharr method36) as shown in eqs. (5) and (6), respectively, according to load-depth curves.
| \begin{equation}
H_{\textit{IT}} = P_{\textit{max}}/A_{p}
\end{equation}
| (5) |
| \begin{equation}
A_{P} = 23.46 \times h_{c}^{2}
\end{equation}
| (6) |
Where
HIT is the nanoindentation hardness of the α-Al phase,
Pmax is the maximum indentation load,
AP is the projected area of the indentation, and
hc is the depth of indentation.
4. Results and Discussion
4.1 Solidified microstructures
The solidified microstructures of the base and 0.2Mo+0.2Zr addition alloys in the as-cast conditions are shown in Fig. 1(a) and (b). No micropores were observed throughout the samples due to good casting ability. The general features of those alloys were the primary α-Al phase, eutectic consisting of α-Al and Si phase, and some intermetallic compounds (IMCs) distributed in the eutectic regions. The primary α-Al phase was the dendrite feature, defined as DA, and the eutectic α-Al phase and eutectic Si phase were inter-dendrite features, defined as ID according to the solidification process in this study. The comparison of the SDAS and the volume fraction of ID as shown in Table 1 showed that a finer microstructure with SDAS values of 33.6 µm and a 72.2% volume fraction of the eutectic regions with a smaller size were obtained in the 0.2Mo+0.2Zr addition alloy.
Table 1 The volume fractions of the ID and area ratio between the DA and ID.
Figure 1(c) and (d) show the morphologies of IMCs in the base and 0.2Mo+0.2Zr addition alloys in the as-cast conditions via EPMA characterization. The representative chemical compositions of the investigated phases determined by WDS are given in Table 2. The Al15(Fe, Mn)3Si2 exhibited a “Chinese-script” morphology37) with a volume fraction of 2.4% obtained in the base alloy as shown in Fig. 1(c). The 0.2Mo+0.2Zr addition alloy showed the Al15(Fe, Mn, Mo)3Si2 with “Chinese-script” morphology,37) Al12Mo, and Al3Zr as shown in Fig. 1(d). The Al12Mo and Al3Zr with a mean size of 0.3 µm were observed in or near the Si phase, and Al12Mo and Al3Zr acted as nucleation sites for eutectic Si and reduced the size of eutectic Si with an average thickness of 2.1 µm. The volume fraction of ID with refined Si particles due to Al12Mo and Al3Zr was 72.2% in the 0.2Mo+0.2Zr addition alloy, which was higher than 66.7% of the base alloy. The Al15(Fe, Mn, Mo)3Si2 in the 0.2Mo+0.2Zr addition alloy were smaller in size and larger in volume fractions as shown in Table 2. The LN2D results for IMCs distribution of the base and 0.2Mo+0.2Zr addition alloys are shown in Fig. 1(e) and (f), respectively. The Poisson distribution was calculated to show the distribution condition. The Poisson distribution was the probability of occurrence when the number of particles in the measuring circle is n (n is a constant value), which is a discrete random distribution. The randomly distributed particles are consistent when having enough measuring circles (The number of measuring circles in this experiment was 106.). The variance is the parameter of the Poisson distribution, which describes the distance of a variable from its expected value. The variance for uniform random distribution was 7.0 in the definition of LN2D. If the variance of the experiment is less than 7.0, the particles are distributed in ordering (If the variance is 0, it is in perfect ordering). Conversely, if the variance is greater than 7.0, the particles are distributed in clustering. The base and 0.2Mo+0.2Zr addition alloy showed the same level of LN2Dav with 7.0, and the LN2Dvar with 5.7 and 7.2, respectively. The base alloy exhibited a small LN2Dvar probably due to the larger mean size of the Al15(Fe, Mn)3Si2, about 29.8 µm, which had some volume in overlapping. The 0.2Mo+0.2Zr addition alloy had a larger LN2Dvar than 7.0 for a random point filed predicted by eq. (3), which meant that all IMCs of the Al15(Fe, Mn, Mo)3Si2, Al3Zr, and Al12Mo with a smaller size in the 0.2Mo+0.2Zr addition alloy were clustered in or near the eutectic grains.
Table 2 Chemical compositions and the volume fraction of the base and 0.2Mo+0.2Zr addition alloys.
Typical DTA heating and cooling curves obtained from the base and 0.2Mo+0.2Zr addition alloys are shown in Fig. 2. The primary peak A showing liquidus temperature (TA) for the base alloy was the primary crystal reaction (7), which occurred at 864 K. Then, the eutectic reaction (8) occurred at TB of 844 K. The reaction of (9) at TC of 841 K appeared to be involved within the eutectic reaction peak in the DTA curves.
| \begin{equation}
\text{L}\to \text{primary $\alpha$-Al}
\end{equation}
| (7) |
| \begin{equation}
\text{L}\to \text{eutectic $\alpha$-Al+Si}
\end{equation}
| (8) |
| \begin{equation}
\text{L}\to \text{eutectic $\alpha$-Al+Si+Al$_{15}$(Fe,$\,$Mn)$_{3}$Si$_{2}$}
\end{equation}
| (9) |
In contrast, peak A was considered to be derived from the mixed exothermic reaction of Al3Zr, Al12Mo, and α-Al crystallization, and both Al3Zr and Al12Mo were formed before the crystallization of the primary α-Al phase in the 0.2Mo+0.2Zr addition alloy. The DTA curves revealed a clear trend where the TA shift to around 20 K due to the formation of the Al3Zr and Al12Mo in the 0.2Mo+0.2Zr addition alloy. There were different peaks, showing downward and upward convex shapes in the base and 0.2Mo+0.2Zr addition alloy, respectively, at a temperature between TA and TB. In addition, the thermal effect was calculated based on the area of the peaks. It was found that the ratio of the thermal effect between the primary peak A and eutectic peak B of each alloy was consistent with the area ratio between the DA and ID as seen in Fig. 1 and Table 1.
The solidification paths showed the changes in element composition from DA to ID after solidification. The solidification paths of the base and 0.2Mo+0.2Zr addition alloys are shown in Fig. 3. The composition deviated from the average composition due to non-equilibrium solidification, resulting in concentrated regions of the added elements during the solidification. Thereby, the Al12Mo and Al3Zr phases were formed, followed by the crystallization of the primary α-Al phase in the 0.2Mo+0.2Zr addition alloy, as shown by position ⓪ to ① to ② in Fig. 3(c). While the primary α-Al phase was crystallized at first in the base alloy without concentrated regions of the added elements as shown by position ⓪ to ① in Fig. 3(a). The eutectic reaction of the Si and α-Al phase with less solute element concentration than the primary α-Al phase was formed by repeating the non-isothermal form in the eutectic due to the redistribution of solute elements as shown by position ② and ③ in the base and 0.2Mo+0.2Zr addition alloys, respectively. The compositions of primary and eutectic α-Al phases in the base and 0.2Mo+0.2Zr addition alloys as magnified parts of (a) and (c) are shown in Fig. 3(b) and (d), respectively. The primary α-Al phase in the 0.2Mo+0.2Zr addition alloy had a low mean composition of 97.7Al–0.54Si–0.08Fe–0.10Mn–0.09Mo–0.13Zr with ΔMkpα of 5.4 eV than 98.3Al–0.62Si–0.05Fe–0.11Mn with ΔMkpα of 4.8 eV of the base alloy as shown in Fig. 3(b) and (d), which was caused by the Mo and Zr atoms in solid solution. However, the eutectic α-Al phase with a different composition of 99.1Al–0.33Si–0.04Fe–0.04Mn–0.2Mo–0.1Zr with ΔMkeα of 4.1 eV, compared with ΔMkpα of 5.4 eV, which was caused by diffusion modes with solute diffusion in the 0.2Mo+0.2Zr addition alloy in non-isothermal solidification. In addition, the formation of Al15(Fe, Mn)3Si2 and Al15(Fe, Mn, Mo)3Si2 was caused finally, as shown by position ③ and ④ in the base and 0.2Mo+0.2Zr addition alloys, respectively. The solidification paths were also in good agreement with the DTA results, as shown in Fig. 2.
The chemical composition analysis in the eutectic α-Al phase for the base and 0.2Mo+0.2Zr addition alloys is shown in Fig. 4. The refinement Si phase with an average thickness of 2.1 µm was obtained in the 0.2Mo+0.2Zr addition alloy, compared with one (4.6 µm) of the base alloy. The chemical composition analysis of the base and 0.2Mo+0.2Zr addition alloys was selected as the eutectic α-Al phase between two adjacent Si phases as shown in the BSE images. The direction of elemental analysis was from points A to B, and the results of the elemental analysis were calculated into the ΔMkeα, showing the indication of solid solution strengthening the level of eutectic α-Al phase by using the eq. (2). The results of the Al content and ΔMkeα in the base and 0.2Mo+0.2Zr addition alloys are shown in Fig. 4(c) and (d), respectively. The alloys exhibited a similar variation tendency in ΔMkeα values, being high close to the Si phase and low far away from the Si phase, which was reflected in the different solid solution strengthening levels in the eutectic α-Al phase. The center region with high Al content of 99.6% was narrow in the 0.2Mo+0.2Zr addition alloy, compared with that of 99.1% in the base alloy. In addition, the 0.2Mo+0.2Zr addition alloy exhibited a higher ΔMkeα than the base alloy caused by the solid solution of Mo and Zr atoms. The high content Al region had fewer solid solution atoms, showing a lower ΔMkeα value. The variation of ΔMkeα in the 0.2Mo+0.2Zr addition alloy implied that the Mo and Zr atoms with a non-isothermal system were redistributed or micro-segregated34) in the eutectic α-Al phase during the solidification process.
The nominal tensile stress-strain curves at 293 K for the experimental alloys in the as-cast conditions are shown in Fig. 5 and the characterization of tensile properties of the experimental alloys is shown in Table 3. The σ0.2, σUTS, and εf of the base alloy were 76 MPa, 145 MPa, and 6.1%, respectively. The 0.2Mo+0.2Zr addition alloy showed tensile properties with the σ0.2 of 89 MPa, σUTS of 160 MPa, and εf of 7.1%. The 0.2Mo+0.2Zr addition alloy showed an increment in the σUTS and εf by 8% and 14%, respectively, compared to the base alloy. The solid solution strengthening effect (ΔMk: 60.0 eV in the base and 62.4 eV in the 0.2Mo+0.2Zr addition alloys) caused by the addition of 0.2Mo+0.2Zr increased Young’s modulus (E) of the alloy from 73 GPa of the base alloy to 76 GPa. The other behaviors in S-S curves might be predominantly influenced by microstructural characteristics. The work hardening coefficient calculated from true S-S curves in the 0.2Mo+0.2Zr addition alloy was 0.3, which was greater than 0.2 of the base alloy, indicating that the 0.2Mo+0.2Zr addition alloy had good uniform deep drawing deformation ability caused by uniformly distributed eutectic regions influenced by Al3Zr and Al12Mo, as described in details in later section 4.5. The r-value of the 0.2Mo+0.2Zr addition alloy was 0.1 larger than that of the base alloy, which resulted in good drawing deformation ability. In addition, the 0.2Mo+0.2Zr addition alloy exhibited a higher work hardening rate and work hardening amount than the base alloy at the same strain rate at 293 K. This might be related to the formation of the small-sized eutectic α-Al phase during solidification and the degree of solid solution hardening, which is discussed in the later section. It is found that the 0.2Mo+0.2Zr addition alloy showed an improvement in flow stress and εf even at the as-cast conditions because of solid solution strengthening and microstructural control.
Table 3 The characterization of tensile properties of the base and 0.2Mo+0.2Zr addition alloys.
The fractured longitudinal sections of the base and 0.2Mo+0.2Zr addition alloys are shown in Fig. 6. The fracture surface of the base alloy was relatively flat with the Si interface, while the fracture surface of the 0.2Mo+0.2Zr addition alloy had larger undulation and irregularity of the Al interface. The high-magnification longitudinal cross-section of the fracture surface in the base and 0.2Mo+0.2Zr addition alloys are shown in Fig. 6(b) and (d), respectively. The fractured eutectic Al–Si interface could be observed in the base alloy, it was inconspicuous in the 0.2Mo+0.2Zr addition alloy, due to the smaller size of Si. However, the longitudinal section of the 0.2Mo+0.2Zr addition alloy exhibited malleable fractures in the eutectic α-Al phase. It is considered that the high Al or low solute elements regions in the eutectic α-Al phase as a continuous phase in the alloy due to heavy micro-segregation led to improvement in ductility and strength.
To clarify that the mechanical properties of the alloys were related to the α-Al phase in eutectic as the continuous phase, the nanoindentation measurements were performed on the primary and eutectic α-Al phases in base and 0.2Mo+0.2Zr addition alloys. Figure 7 shows the typical load (P) - depth (h) curves obtained from the nanoindentation experiments. The primary α-Al phase in the 0.2Mo+0.2Zr addition alloy showed a lower depth at the constant load than the base alloy. The degree of solid solution strengthening showed a higher level in ΔMkpα value for the primary α-Al phase in the 0.2Mo+0.2Zr addition alloy as shown in Fig. 3, which resulted in higher values in HIT and E. The HIT and E of the primary α-Al phase in the base and 0.2Mo+0.2Zr addition alloys were 825 MPa and 67 GPa, and 901 MPa and 71 GPa, respectively. The nanoindentation measurements were conducted on the eutectic α-Al phase between two adjacent eutectic Si phases, as seen in Fig. 8. The minimum value of the HIT (HIT, min) of the eutectic α-Al phase appeared in the central region between the Si phases, which exhibited high Al content as shown in Fig. 4(a) and (b) in both alloys. The typical P-h curves obtained from the nanoindentation measurements of the maximum HIT (HIT, max) and HIT, min are shown in Fig. 8(a) and (b). The HIT, max, Emax, and HIT, min, Emin, in the base alloy were 907 MPa, 71 GPa, and 801 MPa, 67 GPa, respectively. Its difference value (ΔHIT) between HIT, max, and HIT, min was 46.1 MPa. In contract, the HIT, max, Emax, and HIT, min, Emin, in the 0.2Mo+0.2Zr addition alloy were 912 MPa, 72 GPa, and 764 MPa, 67 GPa, respectively, and its ΔHIT was 96.1 MPa. The 0.2Mo+0.2Zr addition alloy showed two-fold higher ΔHIT, which resulted from the heterogeneity in the degree of solid solution strengthening through eutectic α-Al phase due to segregation in solidification. Further, the 0.2Mo+0.2Zr addition alloy with the smallest HIT, min showed the highest Al content layers or ΔMkea in the eutectic α-Al phase as continuous layers, as seen in Fig. 4(c) and (d), with a better deformation ability than the base alloy. Thereby, the εf of 0.2Mo+0.2Zr addition alloy was improved by 14% compared to the base alloy, as listed in Table 3.
It seemed that the improvement of the εf in 0.2Mo+0.2Zr addition alloy was mainly related to the compositional heterogeneity through the eutectic α-Al phase segregated by solute atoms such as Mo and Zr. The dark-field images of the eutectic α-Al phase in the base and 0.2Mo+0.2Zr addition alloys with 1% plastic strains by cold rolling are shown in Fig. 9. The base alloy showed a cell structure with a mean size of 820 nm. The 0.2Mo+0.2Zr addition alloy exhibited a certain number of small-sized cell structures with a mean size of 250 nm at 1% plastic strain. The size of the cell structures might correlate with one of the eutectic grains, as shown in Figs. 1 and 4. The plastic deformation resulted in the dislocations forming dense dislocation tangles in areas with few dislocations or in the walls surrounding the cell, which could call as cell structures.13) The segregated atoms might lead to the formation of the walls, and the higher solid solution strengthening level made the dislocations difficult to move, thereby a certain number of small-sized cell structures were observed in the 0.2Mo+0.2Zr addition alloy.
To reveal the distribution of Mo and Zr in the eutectic α-Al phase, the bright-field TEM image of the eutectic α-Al phase and element distribution of 0.2Mo+0.2Zr addition alloy prepared by focused ion beam equipment is shown in Fig. 10. The light and dark streaks corresponding to region “A” and “B” labels, respectively, were observed within the eutectic α-Al grains scaled by points 1 to 10. The points 1 to 10 in Fig. 10(a) correspond to that in Fig. 10(b). The dark streak exhibited the accumulation of dislocations due to the segregation of Mo and Zr atoms as shown in Fig. 10(b). The eutectic α-Al phase exhibited a non-uniform structure with segregation of solute atoms, which was affected by the solidification, which could improve the ductility of the Al alloys.38)
Both Al3Zr and Al12Mo were localized in the eutectic area of the 0.2Mo+0.2Zr addition alloy, as shown in Fig. 1, although they were heterogeneity nucleated for subsequent eutectic reaction, especially in the Si phase in the non-isothermal solidification process. Their IMCs were not acted as nucleation sites of the primary α-Al phase. They were randomly distributed in the eutectic region and acted as nucleation sites of the eutectic Si phase. The solute elements in liquid showed heterogeneity concentration profiles even eutectic period in the non-isothermal solidification, after primary α-Al phase crystallization. The IMCs of the Al3Zr and Al12Mo crystallized (Peak A) before the eutectic phase (Peak B) as shown in DTA curves, and the Al3Zr and Al12Mo with a mean size of 0.3 µm were observed in the eutectic Si phase as shown in Figs. 1(d) and 4(b). It was considered that solute elements such as additive Mo and Zr moved toward both Al3Zr and Al12Mo which acted as the nucleation site of eutectic Si. In contrast, Al content increased in the center area between Si phases in eutectic regions. The nucleation of the Si phase was caused on the surface of both IMCs (seen in Fig. 4(c)), and then the depleted zone of Mo and Zr was caused near Si. Thus, the α-Al phase with a low level of solute elements was nucleated on the Si phase. Figure 10(a) and (b) revealed the phenomena mentioned above. Point 1 near Si showed a high Al content level or low ΔMkeα, as shown in Fig. 10(a) and (b). Region A and B with high and low Al or low and high Mo and Zr contents corresponded to low and high levels in hardness, respectively, as shown in Fig. 8.
It is considered in the 0.2Mo+0.2Zr addition alloy with Al3Zr and Al12Mo of nucleation sites for eutectic Si, that the solutes enrichment strongly depending on their diffusion coefficients in liquid was caused by keeping the local equilibrium in liquid ahead of the solid and liquid interface, under same concentration between the solidified α-Al solid solution and infinity CL∞ in liquid representing by the effective equilibrium coefficient and a certain thickness of boundary layer depending on the solidification rate when CL∞ is equal to Cs in this model.39) The mixing of solute in liquid was conducted by just solute diffusion in this solidification process. These concentration profiles of Al, Mo, Zr, or ΔMkeα as shown in Fig. 10(b) corresponded to the initial or steady stages39) in solidification according to the solute enrichment model with constitutional supercooling mentioned above, which resulted in heavy micro-segregation. The solidification was non-isothermally performed even in the eutectic regions. The concentration difference of solute atoms showed by ΔCL between CL and CL∞ corresponded to a large Al or ΔMkeα difference amount in the shortened distance, as shown in Fig. 4(c), (d) and Table 4. The segregation degree (K) was approximated by eq. (10) using CAl, max, CAl, min and CAl, total, showing concentrations of maximum, minimum and total Al content in eutectic α-Al, respectively, as listed in Table 4.
| \begin{equation}
K = (C_{\textit{Al},\,\textit{max}}-C_{\textit{Al},\,\textit{min}})/C_{\textit{Al},\,\textit{total}}\times 100\%
\end{equation}
| (10) |
Table 4 The segregation coefficient K, CAl, max − CAl, min and ΔMkea, max − ΔMkea, min values of base and 0.2Mo+0.2Zr addition alloys at the as-cast conditions.
The 0.2Mo+0.2Zr addition alloy showed higher K with 2.9 and a difference ΔMkeα with 4.3 in shorter regions compared with the base alloy (1.9 and 2.6). In contrast, the base alloy showed a different concentration profile with higher CL∞ than Cs between the α-Al solid solution and infinity in liquid, which results from both effects of solute diffusion and convection, compared with the 0.2Mo+0.2Zr addition alloy. The base alloy showed a lower difference amount in the concentration of Al or ΔMkeα in larger regions, compared with the 0.2Mo+0.2Zr addition alloy, as shown in Fig. 4(c), (d) and Table 4, which might correspond to poor liquid flow and the refinement of the α-Al phase in the eutectic regions in the 0.2Mo+0.2Zr addition alloy. Figure 4 reveals the different levels (ΔCAl: 0.5%, 99.1%, 99.6%) of Al content between the base and 0.2Mo+0.2Zr addition alloys, respectively, mentioned above.
4.6 As-cast application possibility of the 0.2Mo+0.2Zr alloy
The Al–9Si–0.3Fe–0.15Mn alloy was chosen as the base alloy, and 0.2Mo+0.2Zr was added to the base alloy to maintain the ductility and improve the strength in tensile properties. The 0.2Mo+0.2Zr addition alloy showed improvement in both σUTS (160 MPa) and εf (7.1%) at the as-cast conditions, compared with those (145 MPa, 6.1%) of the base alloy. Figure 11 shows Vickers hardness in the DA, ID and typical wide area consisting of DA and ID of base and 0.2Mo+0.2Zr addition alloys. The mixture rule in hardness was roughly established in both alloys. The ID consisting of eutectic and IMCs showed maximum hardness in alloys, especially, their error bars were large, and their values varied widely in the 0.2Mo+0.2Zr alloy. It is considered that the characterizations of the eutectic structure are the main reason why both the strength and ductility, which are usually in a reciprocal relationship, are improved. Small IMCs such as Al3Zr, Al12Mo and Al15(Fe, Mn, Mo)3Si2 were clustered dispersedly in or near eutectic grains of the 0.2Mo+0.2Zr addition alloy, thus, the increase of the volume fraction of the Al3Zr and Al12Mo33) improved the tensile strength, because of the solid solution and precipitation hardening mechanisms. In contrast, for improvement in ductility, the inhomogeneity α-Al phase in eutectics with 72.2% as a continuous layer, exhibiting a high Al content layer or depleted area of solutes, which was formed by the clustered distribution of the Al3Zr and Al12Mo, which could improve the ductility of the 0.2Mo+0.2Zr addition alloy. Therefore, the inhomogeneity eutectic grains including of IMCs acted as the harmonic structure for the improvement in both strength and ductility.40) It was found on the basis of improvement in tensile properties that both usage of the gravity casting method and the addition of Mo and Zr, suggested the possibility for the as-cast application because the eutectic Si particles were refined due to the crystallization of both IMCs, and both IMCs acted as nucleation site for eutectic Si.41) And the eutectic α-Al phase was characterized by high Al content or low ΔMkeα region by micro-segregation of the Mo and Zr atoms.
