Macrosegregation Evolution in Eutectic Al-Si Alloy under the Influence of a Rotational Magnetic Field

Abstract

Using magnetic stirring during solidification provides a good opportunity to control the microstructure of alloys, thus controlling their physical properties. However, magnetic stirring is often accompanied by a change in local concentrations, and new structures form which could harm the physical properties. This research paper investigated the effect of forced melt flow by a rotating magnetic field (RMF) on the macrostructure of an Al-Si eutectic alloy. To serve this purpose, Al-12.6 wt% Si alloy samples were solidified in a vertical Bridgman-type furnace equipped with a rotating magnetic inductor to induce the flow in the melt. The diameter and length of the sample are 8 mm and 120 mm, respectively. The solidification parameters are a temperature gradient (G) of 6 K/m, and the solid/liquid front velocity (v) of 0.1 mm/s. These samples were divided into parts during the solidification process, where some of these parts are solidified under the effect of RMF stirring while others are solidified without stirring. The structure obtained after solidification showed a distinct impact of stirring by RMF; new phases have been solidified which were not originally present in the structure before stirring. Besides the eutectic structure, the new phases are the primary aluminum and the primary silicon. The Si concentration and the volume fraction of each phase were measured using Energy-Dispersive Spectroscope (EDS)and new image processing techniques. The experimental results reveal that applying the RMF during the solidification has a distinct effect on the macrostructure of Al-Si eutectic alloys. Indeed, the RMF provokes macro-segregation, reduces the amount of eutectic structure, and changes the sample’s Si concentration distribution.

1. Introduction

Whether it is natural or forced convection, fluid flow can significantly affect the microstructure that develops during solidification. The concentration and temperature variation between the liquid phase in the mushy zone and the pure liquid causes a density difference which might produce natural convection during unidirectional solidification on Earth because of the gravitation. On the other hand, forced convection occurs due to externally applied conditions [1]. During the past two decades, there has been significant interest in forced flow during the solidification process due to its traceable effect on the alloys’ microstructure and, thus, their physical properties. A time-varying RMF can be used to provoke the melt flow. The flexibility of electromagnetic fields makes them an attractive research field. They are utterly contactless with the melt. The stirring intensity can be easily controlled by adjusting the electric control parameters, and the flow pattern can be modulated through coil configurations [2]. Previous studies revealed that the use of RMF has been associated with the following effects: columnar to equiaxed dendritic growth transition (CET) [3], grain refinement [4], and macro-segregation [5]. However, the exact mechanism by which magnetic stirring affects the macro-and microstructure remains unclear.

It is essential to understand the fluid flow underlying rotating magnetic stirring to fully realize the effect of magnetic stirring during solidification. Many experiments and simulations have been conducted to predict fluid flow during directional solidification [6,7]. These experiments are often carried out on an axis-symmetric cylinder mold placed inside a furnace equipped with a rotating magnetic inductor. To ensure unidirectional solidification, the sample was pulled out in one direction, often downward. During solidification, the rotating magnetic inductor was operated. The movement of the solidification front was in the opposite direction of the heat extraction.

Zimmermann and Weiss [1] state two flows when applying RMF during solidification: (i) azimuthal primary flow around the mold axis. Changing currents in the stator windings generate a rotating magnetic field radially around the rotor. A rotating magnetic field causes eddy currents. The induction process between eddy currents and the free electrical charges in the melt produces the Lorentz force, which is the primary cause of moving the liquid mass around the sample’s center [2]. (ii) A secondary meridional flow is generated by the imbalance of the centrifugal force and the radial pressure gradient near the horizontal walls. The secondary flow has a form of two toroidal vortices in the r-z plane (the plane in the radial r and length z directions). These vortices transport the alloying element from the mushy zone ahead of the growing front to the pure liquid (Figure 1a).

As a result of combining these two flows, a complex heat and mass transport in the melt exists in a spiral motion, which leads to a significant change in the growth morphology (Figure 1b).

The RMF is considered an efficient tool to change the fluid flow and, consequently, the solute distribution. This does not, however, imply that the solute concentration has become homogeneous. According to Wang [8], applying a steady electromagnetic stirring does not eliminate macro-segregations but rather modifies their locations and may potentially cause and promote segregation.

The role of forced convection during solidification is often studied using low melting point alloys with pure or partly dendritic microstructure, such as with the Al-7 wt% Si hypoeutectic alloy. When the Al-Si hypoeutectic alloy is directionally solidified in a cylindrical mold under the effect of RMF, three main results are confirmed: (i) An earlier occurrence of CET [9,10,11]. (ii) The complicated spiral flow carries the alloying element from the edges to the center of the sample due to that eutectic structure solidifying at the center axis of the sample. (iii) The secondary flow carries the solute from the mushy zone to the pure liquid, where the solute accumulates and forms a silicon-rich liquid [3,5,12,13,14].

The previous observations are valid only when there is a laminar flow. But when the magnetic intensity is high enough, a turbulent flow occurs. Lim states that turbulent flow occurs when the Taylor number exceeds the critical value 10⁵ [15]. When turbulent flow is present, the macro-segregation changes its pattern to a Christmas tree-like (CTL) shape (Figure 1c). Turbulent flow activates Taylor and Gortler (T-G) vortices, which appear randomly along the sidewall of the cylinder. These vortices transfer the silicon in two parallel reverse directions: from the mushy zone to the liquid and vice versa, in a recurring inversion manner. The plash of the (T-G) vortices in the mushy zone leads to a radial flow, carrying silicon with it. As a result, local remelting occurs at the solidification front, causing it to change to a wavy shape, and the segregation takes place in the form of a Christmas tree.

The impact of RMF on dendritic alloys was given great attention during previous researches. However, only a few have investigated the growth of eutectic alloys under the effect of forced melt flow.

It is commonly known that eutectic solidification entails the isothermal transition of a homogenous liquid solution into two distinct solid phases, that is, L → α + β. The two solid phases grow in an alternating pattern and proceed through a diffusion couple zone, providing the most effective solute transport. Depending on the growth parameters of the solid phases, eutectic phases may emerge in various morphologies. We may classify the eutectic structure into two types: regular and irregular eutectic [16].

The Al-Si eutectic structure is an excellent example of an irregular eutectic structure, with aluminum and silicon as the non-faceted and faceted phases, respectively. The eutectic structure contains soft Al as a matrix and Si as plate-like. The solidification of the irregular eutectic occurs in an asymmetric couple zone. The nature of the eutectic solidification mechanism dictates that one phase should solidify ahead of the solid-liquid phase. This phase is also known as the leading phase.

Shingu and his fellows [17] conducted a solidification experiment in which an Al-Si eutectic alloy was solidified unidirectionally under the forced convective flow of the melt by RMF. They indicate the occurrence of the “eutectic separation phenomenon’’. This phenomenon is described by the simultaneous growth of aluminum and silicon phases separated by several centimeters, aided by the macro transport of the solid’s nuclei via fluid flow. The authors proposed the S parameter to analyze the influence of the melt flow on the solidified structure. The S parameter can be given by the ratio between the flow and the growth rates (S = flow rate/growth rate). When S is greater than 3 × 10⁴, a notable propensity for macro-segregation is found. The separated eutectic phenomenon was attributed to the fact that the leading phase is altered due to fluid flow, where it is silicon when there is no forced convection and aluminum in the case of RMF. The Si phase nucleates and solidifies in the liquid ahead of the solid-liquid front, making it the leading phase. However, when applying RMF, the convective flow separates the nucleated Si particles from the solidification front; consequently, the solidification front will be constituted by the other eutectic phase, aluminum.

Zhongming and Jinze [18] studied the formation of separated eutectic phenomenon in Al-Si eutectic ingots under electromagnetic stirring (EMS). It was found that the walls of ingots play a significant role in the separated eutectic, where the eutectic silicon detached from the aluminum and concentrated around the ingot’s perimeter, producing a silicon-rich layer. During a quenching experiment, the authors observed that the silicon-rich layer extended into the liquid across the solidification front. This indicates that the Si-layer originated in the liquid prior to the Solid/Liquid (s/l) front, i.e., before solidification occurred there. This finding approves the results concluded by Shingu [17], who claimed that the secondary convective flow swept away the Si nuclei into the liquid ahead of the s/l front.

Kim and Shingu [19] studied the effect of fluid flow intensity on the macro-segregation evolution of an Al-Si eutectic alloy. Macro-segregation is shown to be promoted by greater fluid flow intensity; nevertheless, there is a tipping point beyond which no more macro-segregation is attained. They attribute it to the limited ability of the fluid flow to decrease and penetrate the diffusion boundary layer ahead of the s/l front.

Zhejiang and Junze [20] conducted over 26 solidification experiments to study the influence of forced melt flow on simple and complex binary eutectic alloys. They revealed that separate eutectic occurrences could be interpreted using the eutectic system’s thermodynamic data. For example, the separated eutectic phenomenon occurs exclusively in the irregular eutectic structure, where the leading phase is faceted. Also, they revealed that a significant eutectic separation would happen when one of the irregular eutectic phases has a solution entropy of more than 23 J/mol.K.

This study aims to investigate the effect of RMF on the macro-segregation evolution during the unidirectional solidification of an Al-Si eutectic alloy using different values of magnetic induction (B).

2. Materials and Methods

2.1. Alloy

Al-12.6 wt% Si eutectic alloy samples were used for the solidification experiments. Al (purity is 99.95 wt%), Si (purity is 99.95 wt%).

2.2. Solidification Facility

The solidification experiments were carried out in a vertical Bridgman-type tube furnace (Figure 2) which consists of the following parts [21]:

• The capsule with the thermocouples and a solidified sample; the sample (Part 1) is solidified inside the capsule. The capsule contains three parts; the upper and middle parts are made from alumina ceramic (Part 2), and the lower part is copper (Part 4) due to its high thermal conductivity. The thickness of the ceramic is 0.2 mm, which makes no significant temperature difference inside the mold where the sample is located and outside the ceramic mold wall during the solidification process. Long, narrow apertures were placed in longitudinal directions on the surface of the capsule to insert 13 pieces of K-type thermocouples around the capsule at different heights. To avoid direct contact with the coolant, a quartz tube (Part 3) was used to cover these 13 thermocouples.
• The furnace with a programmable temperature controller: the melting process for the sample takes place in a vertical tube furnace. This furnace has four separate heating zones (Part 5). Each zone can be adjusted independently by using a programmable temperature controller.
• Control unit: This unit controls the samples’ movement toward the cooling unit under the furnace. A step motor can adjust the velocity of movement (Part 6).
• The MagnetoHydroDynamic stirring (MHD) inductor: its task is to apply an external RMF to the molten liquid during the solidification experiments (Part 7).
• Cooling unit (Part 8): Water was used as the coolant. Cooling the sample is accomplished by submerging the copper block into water.
• Basement: It holds the whole solidification facility (Part 9).

Figure 2. Sketch of solidification facility; 1: Sample, 2: alumina capsule, 3: quartz tube, 4: copper cooling core, 5: furnace with four heating zones, 6: step motor, 7: RMF inductor, 8: water cooling, 9: basement [5].

Figure 2. Sketch of solidification facility; 1: Sample, 2: alumina capsule, 3: quartz tube, 4: copper cooling core, 5: furnace with four heating zones, 6: step motor, 7: RMF inductor, 8: water cooling, 9: basement [5].

2.3. Solidification Experiments

According to the vertical upward Bridgman method, the Al-Si eutectic alloy samples were unidirectionally solidified in the above-mentioned solidification facility. The dimensions of the cylindrical samples to be solidified were ∅ 7.1 mm × 110 mm. The unidirectional solidification was realized by the relative vertical transition of the sample from the furnace chamber to the cooling bath.

During the experiments, the Al-Si eutectic alloy samples were solidified in four steps considering different magnetic induction (B) values in each sample (

Table 1. Magnetic induction (B) profile during solidification.

Sample	B (mT)
	Step 1	Step 2	Step 3	Step 4
A	0	20	0	40
B	0	60	0	80
C	0	100	0	120
D *	0	150	N/A	N/A

*: During the solidification of the D sample, only steps 1 and 2 were carried out, but steps 3 and 4 were not performed.

and Figure 3). As stated earlier, the aim is to study the effects of stirring by RMF with different magnetic inductions on macro-segregation. For that, the other solidification parameters were maintained to be constant: frequency of magnetic stirring (f) of 50 Hz, average s/l front velocity (v) of 0.1 mm/s and average temperature gradient (G) of 6 K/mm. v and G were calculated from the measured cooling curves at the solidification front of the eutectic structure.

2.4. Measuring Methods

According to standard metallographic procedures, the solidified samples were sectioned longitudinally and prepared by grinding, polishing, and etching in an aqueous solution of Hydrogen Fluoride (HF) acid. The microstructures were examined in the etched condition using optical microscopy. Three parameters were characterized.

• The macro-segregation: different macro-segregation patterns result from changes in magnetic induction values (B). The macro-segregation evolution in each sample is observed by analyzing the macroscopic photos.
• The average of the area fraction of the eutectic was measured by the Image J software by all stirred and non-stirred steps. The measurement process is performed on the macroscopic photos.
• The distribution of Si concentration is analyzed using a Scanning Electron Microscope (SEM) equipped with an Energy-Dispersive Spectroscope (EDS). The analysis is performed along the axis of the samples at different locations. The samples were divided into 16 parts with 0.45 mm × 2 mm dimensions along the diameter. The average Si concentration was measured in every part.

3. Results

3.1. Qualitative Analysis of Microstructure

In step 1, all samples were solidified without stirring (0 mT), which means there was no solute redistribution or forced fluid flow during solidification (Figure 4). As a result, the well-known irregular eutectic structure solidified without macro-segregation. The micrograph has several white areas that reflect the degenerated eutectic structure.

In step 2; the macrostructure of the solidified samples is shown in Figure 5. The RMF was operated during the solidification of the samples with different B values on each sample. Applying a magnetic field destroys the coupled growth of the eutectic phases, replacing it with distinct macro-segregation patterns. The segregation patterns were highly related to the B values. The following segregation patterns are observed:

• A distinct segregation pattern is observed in all samples, in which the primary aluminum (white areas) and eutectic structure (gray areas) solidified at the sample’s edges. In contrast, there is an almost pure eutectic structure in the center channel areas of the samples (Figure 5).
• One may note that the pure eutectic structure in the center channel areas has the shape of the CTL segregation pattern. However, the side arm freckles of the CTL have an opposite direction to that reported in the literature by Roósz [22].
• It can be noticed that sample A (Figure 5A), which is solidified under 20 mT of magnetic induction, had more eutectic structure and less primary aluminum at the edges than samples B and C, which are solidified at 60 and 100 mT, respectively (Figure 5B,C).
• The primary silicon solidification is different in each sample. i.e., the primary silicon is not solidified in the second part of the sample A when 20 mT of magnetic induction is applied (Figure 5A). However, in samples B and C, the primary silicon is solidified into unique periodic “arc-like colonies” (black areas in Figure 5B,C). The arc-like colonies are more apparent in the case of sample C (B = 100 mT) than in sample B (B = 60 mT) due to the reason that the applied magnetic induction is higher. On the other hand, the primary silicon in sample D is solidified between the eutectic side arm freckles of the CTL segregation pattern (Figure 5D).
• Figure 5C shows that the amount of the primary silicon in the periodic arc-like colonies decreases over time. However, the amount of the eutectic structure increases in the CTL segregation before each colony.

In step 3, the RMF is turned off during solidification experiments. There is no macro-segregation in any of the samples, and the macrostructure is similar to that obtained in step 1 (Figure 4). However, it is believed that the Si concentration in step 3 is higher than 12.6%. This assumption is based on the fact that in step 2, the RMF-fluid flows continuously transfer the rejected Si atoms into the pure liquid when the primary aluminum grows at the edges. As a result, the average Si concentration in step 2 is less than 12.6% (eutectic concentration), whereas it is more than 12.6% in the pure liquid. As an outcome, the coupled eutectic structure solidified in step 3 is an over-eutectic Si composition.

In step 4, the RMF is turned on again, and the eutectic structure is stirred for the second time in a row. The macrostructure of the solidified samples is shown in Figure 6. It can be seen that the three samples A, B, and C are solidified in similar segregation patterns, where the primary aluminum phase and the eutectic structure solidified at the edges of the samples. The center channels of the samples mainly consisted of a pure eutectic structure. The eutectic structure is solidified into a CTL segregation pattern. The primary silicon is solidified between the eutectic side arms freckles in the CTL segregation.

When comparing the macrostructures of the different samples in step 4, two main differences can be observed: (i) the amount of the primary silicon in sample A (Figure 6A), where 40 mT of magnetic induction is applied, is much less than in the other samples (B and C) where 80 and 120 mT are used, respectively (Figure 6B,C). (ii) In samples B and C, the solidification of the primary silicon coincided with the operation of the RMF. However, the primary aluminum phase is the one that solidified when RMF is turned on in sample A.

3.2. Quantitative Analysis of Microstructure

The volume fractions of the eutectic structure and the primary phases (primary aluminum and silicon together) are measured by using image J processing software (Image J 1.52a, National Institutes of Health, LOCI, USA). The measurements revealed a distinct effect of magnetic field induction (B) on the volume fraction amount. In step 1, the samples are solidified without stirring by RMF, and due to that, the solidified structures are entirely eutectic.

During step 2, the RMF is operated with different magnetic induction (B) values on each sample. Figure 7 demonstrates that higher values of B produce more primary phases and reduce the amount of the eutectic structure. However, an opposite result is obtained when the samples are stirred for the second time in step 4 of the experiments.

In step 4, the volume fraction of primary phases decreased with increasing the B value. Consequently, the amount of eutectic structure increases. Such results present a promising way to reduce macro-segregation when using RMF. The reason could be that the molten liquid is stirred twice, which means that the Si concentration is spread out more evenly.

3.3. The Si Concentration Distribution

Figure 8 shows the Si concentration distribution along the diameter of the C sample in steps 1 and 2. In step 1, the sample is solidified without RMF. Thus, the average Si measured concentration is 12.6% along the diameter of the sample (at 5 mm). Moving forward to step 2, where the stirring starts, a solute redistribution occurs by the act of the primary and secondary flow. At line 27 mm, the primary aluminum and eutectic structure solidified at the edges; because of that, the Si concentration at the edges is less than the center, which contains only a pure eutectic structure with eutectic composition (12.6%). On the other hand, the aforementioned arc-like segregation patterns are confirmed by lines 28.5 and 31.5 mm. The line at 28.5 mm is chosen to describe the edges of the arc, where it can be observed that the edges have a higher Si concentration than the center of the sample. The center area of the arc is described by the line at 31.5 mm, which shows more silicon in the middle than on the edges.

Figure 9 shows the Si concentration distribution in steps 3 and 4 of sample C. In step 3, the sample is solidified without stirring; the average measured Si concentration at line 5 mm is 13.6 wt%. This average concentration is higher than in step 1 because the sample was stirred previously during step 2. Line 0.25 mm shows that when stirring starts at step 4, the primary silicon solidifies instantly at the edges. The measurements show higher Si concentrations at the edges and eutectic concentrations at the center of the sample. Moving forward to the 15.8 mm line, the three phases coexist along the radius of the sample. The primary silicon is randomly distributed along with the sample.

4. Discussion

In step 1, the samples are solidified into a pure eutectic structure. The eutectic solidification is characterized as cooperative growth. The two solid phases grow simultaneously in an alternating manner. The Al-Si eutectic structure consists of the α-Al and plate-like Si phases. The Al atoms are rejected into the liquid as the Si phase grows, and the Si atoms are rejected into the liquid as the α-Al phase grows (Figure 10a). The growth of the eutectic structure imposes the solidification of one phase before the other; that phase is known as the leading phase. The leading phase is the eutectic Si when no forced convection is present [19,20].

In step 2, the Al-Si eutectic samples are solidified under the influence of RMF. During unidirectional solidification under RMF, two flows affect the solidification structure. One is the primary flow around the sample axis, and the other is the secondary flow caused by the imbalance of the centrifugal force and the radial pressure gradient near the horizontal walls. A complex spiral flow results from combining these two flows.

Veres [23] mentioned that the spiral flow velocity of the melt is close to zero along the sample axis and the walls if the sample axis coincides with the axis of the rotating magnetic field. Along the radius of the sample, the velocity varies between the wall and the center in a curve profile. Because of that, it can be observed that the RMF significantly effects the edges of the samples where the primary aluminum and eutectic structure solidified. In contrast, the center channel area is formed of a pure eutectic structure because the influence of the RMF is at minimum (Figure 10b).

According to Shingu [19] and Zhejiang [20], when an Al-Si eutectic alloy solidifies under the effect of RMF, the leading phase changes from the Si phase to the α-Al phase. Based on that, the following mechanism is suggested to explain the solidification of the different phases in stirred parts in step 2:

(i)

At the edges of the samples, the melt flow velocity is high. For that, the eutectic coupled growth will be destroyed and replaced by the separated growth of eutectic phases. The leading phase is the α-Al, which rejects the Si atoms as it grows. The fluid flows transport the rejected Si atoms from the s/l front to the pure liquid, for which, there is an increase in the Si concentration in the liquid and a decrease in its concentration near the s/l front at the edges. Therefore, the primary aluminum solidified along the edges of the samples (Figure 11II.a). The primary aluminum phase grows in a dendritic growth manner. The dendrites develop secondary arms which reject the Si atoms as they grow. The melt flow has a limited ability to penetrate the dendritic structure and transport the Si atoms between the dendrites, which in turn will cause the Si concentration between the aluminum dendrites to increase. As a result, an inter-dendritic eutectic structure solidifies at the edges.

(ii)

The melt flow velocity is zero at the sample’s center channel area, which maintains the coupled eutectic growth mechanism, and solidifies a pure eutectic structure.

(iii)

The eutectic structure is solidified as a CTL segregation pattern due to the turbulent flow (Figure 11II.b). It was mentioned before that Taylor-Gortler vortices appear along the sidewall of the cylindrical mold when the laminar flow loses its stability and becomes turbulent. The T-G vortices transport the rejected solute in both directions between the mushy zone and the pure liquid. Willer reports a significant amplification of the radial flow when the T-G vortex impinges on the mushy zone, which means that the transported solute by the radial flow is enhanced momentary along the mushy zone [24].

(iv)

Over time, the concentration of the Si will significantly increase in the pure liquid due to the cumulative transportation of the rejected Si atoms from the s/l front. When the concentration is increased above a critical value, the solidification of the primary silicon occurs. The results reveal that the segregation pattern of the primary silicon is highly related to the value of the magnetic induction (B). Higher magnetic induction means more transported Si atoms and a higher Si concentration in the liquid, therefore solidifying the primary silicon in distinct segregation patterns. For example, during step 2 of the C sample solidification (B = 100 mT), the primary silicon solidified into periodic arc-like colonies (Figure 11II.c). In contrast, when the magnetic induction is low, the Si concentration in the liquid doesn’t reach the critical value that allows the primary silicon solidification as in the case of sample A when 20 mT was applied in step 2.

The stirring process by RMF includes transferring the solute by recurring inversions of flow between the mushy zone and the pure liquid, which results in solidifying the primary silicon in the form of specific segregation patterns. Nevertheless, these segregation colonies don’t represent all of the transferred solute atoms. Roplekar [25] and Roosz [22] report a continuous increment of the Si concentration along with the sample. In step 3, the RMF is turned off, which means that the solidification process is pure eutectic solidification, so no primary phases or macro-segregation patterns are found. However, the average Si concentration is higher than the similar eutectic in step 1. The change in Si concentration is accompanied by a change in the leading phase in eutectic solidification. Therefore, the solidification of the eutectic Si phase takes the lead instead of the α-Al. This effect is invisible in the microstructure, but it can be attested by the primary silicon solidification at the beginning of stirring in step 4 of the solidification experiments. Figure 6 states clearly that the solidification of the primary silicon coincided with the operation of the RMF in step 4. Conversely, step 2 shows instant solidification of the primary aluminum dendrites. Both results are attributed to the fact that the leading phase in the prior solidification step determines which primary phase will solidify when RMF is turned on. In the case of step 2, the α-Al phase is the leading phase, but in the case of step 4, the eutectic Si phase is the leading phase due to the increment of Si concentration.

Another conclusion can be obtained from Figure 6 that after the solidification of the primary silicon phase at the edges, the overall Si-concentration turns back to the eutectic concentration. i.e., 12.6%. The leading phase is greatly related to the solute concentration, and due to that, the α-Al phase takes the lead again. However, the stirring process is still going on and the underlying flow that carries the solute-rich melt persists. This led to the same result as before, which is, solidifying the primary aluminum at the edges and the eutectic structure in the center axis of the sample.

5. Conclusions

The effect of RMF on the Al-Si eutectic structure has been investigated by solidifying four axisymmetric cylindrical samples with various magnetic induction (B) profiles at the front velocity of v = 0.1 mm/s and at a temperature gradient of G = 6 K/mm. The macro-segregation, Si concentration distribution, and volume fraction of the samples were investigated by using new measuring methods.

The RMF destroys the coupled growth of the α-Al and Si eutectic phases at the edges and alters the leading phase from the Si to the α-Al phase. In general, the fluid flows reduce the Si concentration at the s/l front at the edges, causing the primary aluminum phase to solidify with an inter-dendritic eutectic. On the other hand, the effect of RMF in the center channel areas of the samples is minimal. Due to that, a pure eutectic structure is solidified there. Due to the turbulent flow, a CTL segregation pattern was found.

The primary silicon segregation patterns were different based on the induced magnetic intensity. At critical values of magnetic induction, the primary silicon tends to accumulate and solidify in arc-like colonies. However, below these values, the primary silicon doesn’t solidify. If the magnetic intensity is above the critical values, the primary silicon is randomly distributed along the axis of the sample. Determining these critical values requires a wide scale of magnetic induction experiments.

During the early stages of stirring, the α-Al matrix was considered as the leading phase. However, after the stirring, the overall Si-concentration increased, and consequently, the leading phase became the Si-eutectic phase. It has been found that the leading phase has a significant effect on the resultant macro-segregation patterns. For example, if the leading phase is the α-Al phase, then the primary aluminum dendrite solidifies before the primary silicon, and vice versa.

Increasing the magnetic induction leads to increasing the amount of the primary phase at the expense of the eutectic structure. However, stirring the sample twice reflects the result., i.e., the eutectic structure amount increases, and the primary phases decrease.