2022 Volume 63 Issue 10 Pages 1469-1476
The effects of ultrasonic treatment (UST) using a titanium (Ti) sonotrode in a fully liquid stage on the grain refinement and mechanical properties of as-cast Al–Mg alloy billets were investigated. To clarify the grain refinement mechanism, the grain size (GS) and dendrite arm spacing (DAS) were examined as functions of the growth restriction factor (Q) dependent on the Ti dissolution from the sonotrode, and compared to those in the as-cast and re-melted states of the samples inoculated using an Al–10 mass%Ti master alloy. In addition, the formation and dissolution behavior of Al3Ti intermetallic particles acting as a heterogeneous nuclei was indirectly observed by measuring the electrical resistivity during isochronal annealing. In comparison with the chemical refiner inoculation, the UST effectively refined not only the GS but also the DAS, both of which showed similar slightly concave upward curves with an increasing slope against 1/Q. Electrical resistivity measurement results provided indirect evidence that the dissolved Ti was present as a solute during the solidification stage. The GS, DAS, and electrical resistivity results all suggest that the dissolved Ti refined the GS primarily by solute-induced growth restriction effect rather than by providing heterogeneous nucleation sites. The UST effect on the microstructure refinement was efficient when the Ti-dissolution content was as low as less than 0.05 mass%. The refinement of grains, Al3Fe particles, dendrites, and pores by the UST significantly improved mechanical properties, especially the elongation at break.
Fig. 5 Variations in (a) GS and (b) DAS with UST application, chemical inoculation, and re-melting as functions of the inverse growth restriction factor, 1/Q.
The grain refinement is very significant materials processing in aluminum casting, by which the down-stream processing behavior for as-cast intermediate products, mechanical properties for finished products, and structural uniformity for consistent performance are improved.1,2) Especially, in non-heat-treatable aluminum alloys, obtaining a fine, uniform and equiaxed grain structure during the casting process is regarded as a subject of utmost importance, because their final properties are largely determined by the microstructure and the cold working degree as well as the solute content.3,4) The UST has long attracted great interest as a simple, efficient, and eco-friendly method to refine grains of aluminum alloys compared to other approaches. But a lack of correct understanding has limited its industrial applications.5) To explain the grain refinement effect through ultrasonification, two major mechanisms have been proposed: cavitation-induced dendrite fragmentation and cavitation-induced heterogeneous nucleation.6,7) However, despite a lot of research conducted over the past decades, there is still no convincing evidence regarding which mechanism plays the most important role in different solidification stages.8,9) In particular, the application of UST in a fully liquid stage, which is advantageous for industrial applications such as degassing, as well as large-sized casting and continuous casting, has been relatively less studied. Studies on the effect of UST in the fully liquid stage on the grain refinement is mainly classified into the following two categories: first, the application of UST on the alloys, which contain transition metals above a peritectic composition, and thus intermetallic compounds such as Al3Ti acting as heterogeneous nuclei can be formed,5,10,11) and second, the simultaneous application of UST and a grain refiner inoculation.12–16) By contrast, as for applying ultrasound on alloys containing transition metals below a peritectic composition without a separate process for refiner feeding in a fully liquid stage, only several studies have been reported, and there have been very diverse views on the effect of ultrasound on grain refinement.
Some researchers have recently begun to take a different view when discussing the role of ultrasound in grain refinement, because ultrasound is mainly injected into molten metals through sonotrodes made of titanium (Ti) alloys, which have been preferred over Nb alloy sonotrodes for the ultrasonic treatment of light alloy melts owing to their low price. When Ti-alloy sonotrodes are exposed to molten metals under cavitation, they erode and introduce Ti into the melts, and it has been reported that grains are able to be refined without a separate process for refiner feeding by making good use of this undesirable phenomenon.6,17,18) However, data on the performance of Ti-alloy sonotrodes in terms of grain refinement by ultrasonicating under fully liquid conditions are still lacking in the literature, and views on the grain refining mechanism are debatable. In Mg–(3–9) mass%Al alloys, Youn et al. observed that the grain was effectively refined by simply applying ultrasound in a complete melt state using Ti sonotrodes without inoculating the grain refiners. It was explained that the atoms and/or clusters parted from the sonotrode by the cavitation erosion react with the melt to form proper intermetallic compounds such as Al3Ti that can act as heterogeneous nucleation sites.18) In Al alloys, Puga et al.,19) Mills et al.,20) and Zou et al.21) reported that grain refinement was achieved by applying ultrasound using Ti-alloy sonotrodes above the liquidus temperatures in Al–Si, Al–Mg, and Al–Zn–Mg alloys, respectively. However, despite attempts to explain the experimental results with a cavitation-enhanced nucleation mechanism, little attention has been paid to the effect of Ti as a sonotrode material. In addition, the experimental conditions, in which the ultrasonification temperatures were too close to the liquidus temperatures and/or unheated sonotrodes were used, may have unintentionally contributed to the grain refinement through cavitation-induced dendritic fragmentation. Eskin5) and Huang et al.22) reported that the dissolution of Ti from the sonotrodes into the melts above the liquidus temperatures during the ultrasonification refines the as-cast grain structures of dilute Al alloys. As an interesting aspect, although the Ti contents were similar to each other (0.02 mass% and 0.026 mass%), they attributed the grain refinement to different reasons, i.e., a simple alloying effect of a solute element with the highest growth restriction factor (Q) and the formation of Al3Ti intermetallic particles on the sonotrode surface. Jiang et al.6) and Jung et al.23) investigated the mechanism of grain refinement in Al–Zn–Mg and Al–Si alloys with intermediate to high solute contents under ultrasound vibration, focusing on dissolved Ti and considering the effects of cavitation-enhanced nucleation and a chilled radiating surface together. However, although the ultrasonification in a complete melt state using Ti-alloy sonotrodes increased the Ti content by up to 0.046 mass%6) and 0.2 mass%,23) respectively, grain refinement was not observed6) or the level of grain refinement did not exceed a large standard deviation,23) which is presumed to be caused by high amounts of Si. As such, it appears that despite the observed influence of UST applied in the complete liquid stage using Ti-alloy sonotrodes on the grain refinement, a detailed understanding of the grain refinement mechanism is still lacking.
In this study, the effect of UST using a Ti sonotrode under a fully liquid condition, where solidification does not occur, on the microstructure refinement and mechanical properties of A5052 aluminum alloy, a non-heat treatable Al–Mg base aluminum alloy with intermediate solute content, was investigated. To verify the grain refinement mechanism, the grain size (GS) and dendrite arm spacing (DAS) were evaluated as a function of the Q dependent on the Ti dissolution from the sonotrode and were compared to those in the as-cast and re-melted states of the samples grain refined with an Al–10 mass%Ti master alloy. In addition, to explore whether Al3Ti intermetallic particles were formed during the UST process, electrical resistivity measurements were conducted.
A commercial-grade A5052 aluminum alloy was used as an experimental material and the chemical compositions are shown in Table 1. Approximately 10 kg of the alloy ingots were first melted inside a clay-graphite crucible with a 180 mm diameter and 230 mm height using an induction furnace, and heated to ultrasonicating temperatures of 700°C, 740°C, and 780°C. The melt crucibles were then removed from the induction furnace and placed inside an electric resistance furnace pre-heated to these ultrasonicating temperatures.
The UST of the melts was conducted in the preheated electric resistance furnace equipped with an ultrasonic system, which consisted of a 5 kW ultrasound generator (USGC-5-22 MS), a water cooled 20 kHz magnetostrictive transducer (MST-5-18), and a sonotrode made of a commercial purity titanium with a 40 mm diameter. The Ti sonotrode was preheated to the above described ultrasonication temperatures prior to UST processing, and immersed into melts at a depth of 30 mm from the melt surface. The UST experiments were conducted at a fixed power input of 4.4 kW and a fixed frequency of 17.8 kHz by varying the ultrasonication temperature (700°C, 740°C, and 780°C) and time (0, 1, 3, 5 min). To investigate the grain refinement effect under conditions excluding cavitation-induced fragmentation of the primary crystals, all ultrasonication temperatures were set to at least 50°C higher than the liquidus temperature (30°C higher than the Al7Cr crystallization temperature). The ultrasonicated melts were then poured into a steel mold with a cavity of 100 mm in diameter and 400 mm in height that had been preheated to 150°C.
To clarify the mechanism of grain refinement by UST applied in a fully liquid condition using the Ti sonotrode, reference A5052 aluminum alloy billets, which were grain refined using a chemical grain refiner and contained equal amounts of Ti as the ultrasonicated samples, were prepared. Because Al–Ti–B based grain refiners with the best refinement effect possess different nucleant particles and a solute element such as TiB2 or B, which cannot be formed or introduced into the aluminum melt through UST using the Ti sonotrode, an Al–10 mass%Ti master alloy was used as a grain refiner instead. After inoculating pre-weighed quantities of an Al–10 mass%Ti master alloy using a steel plunger, the aluminum alloy melts were mechanically stirred and degassed through Ar gas bubbling filtration (GBF) for 10 min, and then poured into the preheated steel mold to solidify under identical cooling conditions as applied with the ultrasonicated samples.
2.2 Evaluation and analysisThe average Ti contents of the billets were detected using spark optical emission spectroscopy (OES, OBLF QSN-750). To evaluate GS and DAS as functions of Q, the Q for each alloy was calculated using the analyzed chemical compositions of the billets and the partition coefficient and liquidus slope data for binary aluminum alloys,24) assuming each solute in a multi-component system behaves as if in a binary alloy and that the effects of the solutes are additive.25) The specimens for the grain structure and microstructure examinations were cut along the cross sections at 10 mm above the bottom, at the middle height, and at 10 mm below the riser of each billet. These were mechanically polished and etched with a Poulton’s reagent solution (60 ml HCl, 30 ml HNO3, 5 ml HF, 5 ml H2O). The macro- and microstructures were observed using a combination of optical microscopy (OM), polarized light microscopy (PLM), and scanning electron microscopy (SEM). The GS and DAS were measured using a linear intercept technique (ASTM E112).
The tensile mechanical properties of the as-cast billets were determined according to the ASTM E-8 standard using an Instron 5985 tensile test machine. The strain rate was 2.08 × 10−4/s.
Whether Al3Ti intermetallic particles were formed during the UST process was indirectly investigated by measuring the electrical resistivity of as-cast billet specimens during an isochronal annealing up to 480°C with a heating rate of 5°C/min. The relative change in electrical resistivity (ρ), which was normalized to the electrical resistivity in an as-cast state at room temperature, was used for analysis.
Phase characterization was carried out through phase equilibria modelling (JMatPro 5.0, Sente Software Ltd.) and combined analysis of X-ray diffraction (XRD; Rigaku, Japan), and a field emission scanning electron microscope (FE-SEM, FEI Quanta 200F) equipped with an energy-dispersive X-ray spectroscopy (EDS) probe.
The casting of A5052 aluminum alloy billets was carried out after the UST, which was conducted at three different temperatures and three processing times in a fully liquid state under a fixed power input and frequency condition using a Ti sonotrode. In each case with the same superheat, a cast billet without the application of UST was also produced for comparison. The change in cross-sectional grain structures of as-cast billets before and after the UST are shown in Fig. 1 as a function of the ultrasonicating temperature, and quantitative information on the GSs is summarized in Fig. 2. In the untreated samples, which hereafter are samples that were not ultrasonicated and not inoculated, the increase in melt pouring temperature decreased the cooling rate, resulting in an extension of the columnar zone and the increase in the GS. When the UST was applied, fine equiaxed zones were formed over the entire cross-sectional area of all samples, and the increase in ultrasonicating temperature slightly reduced the GS despite the slower cooling rate owing to the increase in pouring temperature. As a result, as the ultrasonicating temperature increased from 700°C to 780°C under the same ultrasonicating time of 5 min, the GS reduction rate relative to the GS values before applying the UST increased from 50% to 60%, as shown in Fig. 2(b). By contrast, the effect of the ultrasonicating time on the macrostructure and grain refinement was observed to be not as pronounced as that of the temperature; as the ultrasonicating time increased up to 5 min at a same ultrasonicating temperature of 780°C, the GS decreased with a diminishing GS reduction rate per unit time.
Photographs showing as-cast grain structures of untreated (left) and ultrasonicated (right) A5052 aluminum alloy billets as functions of ultrasonicating and pouring temperature: (a) 700°C, (b) 740°C, and (c) 780°C.
Variations in (a) GS and (b) GS reduction rate as a function of ultrasonicating and pouring temperature for different ultrasonication times.
To study the effect of UST on mechanical properties of A5052 aluminum alloy in as-cast state, the tensile tests were carried out at room temperature and the results were presented in Fig. 3. Similar to the effect on grain refinement, the change between with and without UST was very significant, whereas the change according to UST application temperature and time was not so prominent within the experimental conditions. However, it was clear that the UST application improved both the ultimate tensile strength (UTS) and elongation. In particular, the increase of elongation was more remarkable, which would be very beneficial for parts forming processes using A5052 aluminum alloy, such as rolling and extrusion. Compared with the untreated samples, the UTS and elongation of the ultrasonicated samples were increased on average around by 11% and 26%, respectively. The improvement of mechanical properties was obviously attributed to the grain refinement induced by the UST. In addition to that, the UST application appears to have had a significant influence on the improvement of mechanical properties by causing the refinement of brittle Al3Fe intermetallic particles with sharp ends that crystallized along grain boundaries (see Fig. 9) as well as the reduction of DAS and porosity (see Figs. 5 and 6).
Variations in (a) ultimate tensile strength and (b) elongation as a function of ultrasonicating and pouring temperature for different ultrasonication times.
The amount of Ti dissolution from the Ti sonotrode into the aluminum alloy melt was measured to quantitatively investigate the effect of the UST application on the grain refinement and is summarized in Fig. 4. The Ti contents of all ultrasonicated samples exceeded 0.005–0.01 mass%, which is the Ti inoculation level where the grains of wrought Al alloys are adequately refined by inoculating the chemical grain refiners.26) As the ultrasonicating temperature increased, the amount of Ti dissolution rapidly increased. Considering that, as shown in Fig. 2, the GS reduction rate significantly increased despite the decrease in the cooling rate as the ultrasonicating temperature increased, it can be seen that the large amount of Ti dissolution played an important role in the grain refinement. Meanwhile, in contrast to the diminishing behavior of the GS reduction rate with the ultrasonication time shown in Fig. 2, in the samples ultrasonicated at 780°C for different ultrasonication times, the amount of Ti dissolution increased almost in proportion to the ultrasonication time.
Variations in Ti dissolution amount from the Ti sonotrode into the aluminum melt as a function of ultrasonicating temperature for different ultrasonicating times.
Deep insight regarding the grain refinement of as-cast aluminum alloy can be gained by plotting the GS against Q by which the segregating power of the solute and the constitutional undercooling that initiates the nucleation are quantified.27) The value of Q is inversely proportional to the growth rate of the dendrites and grains.28) By calculating Q for the alloy composition (the chemical composition of A5052 Al alloy used in the experiment plus the contents of the dissolved Ti), the relationship between the GS and 1/Q is presented in Fig. 5(a). In the ultrasonicated samples, a slightly concave upward curve with an increasing slope against 1/Q was observed. A similar curved GS versus the 1/Q relationship has also been reported in Du et al.’s numerical simulation on an Al–Ti binary alloy system using a CALPHAD-coupled Maxwell–Hellawell GS prediction model.29) For the growth restriction effect of solutes in aluminum alloys, in analytical theories, the GS was reported to be linearly related to 1/Q,30) whereas Du et al. proposed based on their numerical simulation results that the relationship between the GS and 1/Q is non-linear, depending on solute species. In Fig. 5(a), the increasing curvature can be interpreted as a result of the growth restriction parameter having a greater effect on the grain refinement at low levels of Ti content. By contrast, in the inoculated samples containing amounts of Ti equivalent to those of the ultrasonicated samples, the GS continued to decrease without the diminishing behavior of the GS reduction rate according to the decrease in 1/Q (in other words, the increasing Ti content), resulting in half that of the ultrasonicated sample when 1/Q was below 0.045 (compare Figs. 6(b) and 6(d)). The remarkable decrease in GS at the same 1/Q is apparently due to an increase in the number of Al3Ti intermetallic particles acting as heterogeneous nucleation sites. For this reason, after the re-melting process including melt holding for 1 h, which is sufficiently long for the fading of Al3Ti particles to occur,31) the GS of the inoculated sample increased to approximately the same level as that of the ultrasonicated sample. These above variations in GS depending on the Ti content according to the UST application and chemical inoculation, and whether a re-melting occurred or not, indicate that in refining the grains of the ultrasonicated samples, the growth restriction effect of the solutes, and not the heterogeneous nucleation effect, prevailed. In addition, it is noteworthy that the Ti solute induced growth restriction effect was very efficient at a higher 1/Q region, i.e., when Ti content was less than 0.04 mass%, the refining effect was almost similar to that of the inoculation of the Al–Ti master alloy grain refiner.
Variations in (a) GS and (b) DAS with UST application, chemical inoculation, and re-melting as functions of the inverse growth restriction factor, 1/Q.
PLM and OM images of macro- (left) and microstructures (right) of (a), (b) ultrasonicated and (c), (d) grain refiner inoculated samples with different Ti dissolution amounts: (a), (c) 0.02 mass% and (b), (d) 0.07 mass%.
Variations of the DAS, which is not dependent on the GS but on the cooling rate and the alloy composition,32,33) also supported that the ultrasonicated samples were refined primarily by the growth restriction effect of the solutes. Figure 5(b) shows the variation of the DAS with the UST application and the chemical inoculation. The DASs of the ultrasonicated samples showed a slightly concave upward curve with an increasing slope against 1/Q, which is extremely similar to the GS behavior according to the UST application. Interestingly, however, they were smaller than the DASs in the inoculated samples. It can be interpreted that Ti introduced from the sonotrode was mainly present in the aluminum melt as a solute, effectively contributing to retarding the dendrite growth, whereas Ti introduced from the grain refiner was mostly present as an Al3Ti intermetallic compound and the solute level in the aluminum melt was relatively lower. The representative microstructures of the as-cast billets that were influenced by the UST application and the chemical refiner inoculation, respectively, are compared in Fig. 6 as a function of the equivalent Ti content. In the inoculated samples, the DAS reduction rate with an increase in the Ti content was small, whereas the GS reduction rate was large, and thus as the Ti content increased to 0.07 mass%, the grain morphology changed from dendritic to rosette-like or cellular, making it impossible to get a meaningful measurements of the DAS. By contrast, the ultrasonicated samples showed equiaxed dendritic grain structures in all cases. In addition, it was noticeable that the amounts of pores were markedly lower than those of the inoculated and subsequently Ar GBF treated samples, indicating that the UST application in a fully liquid state was effective in degassing. All of the above solidification microstructural features such as the GS, DAS, and grain morphology suggested that Ti introduced into the melt during the UST applied under a fully liquid condition using the Ti sonotrode was mostly present in the form of a solute, and the grains were refined primarily by the growth restriction effect of the solutes rather than by the heterogeneous nucleation effect.
The reason why the effect of Ti on the grain refinement in UST remains an open question is that it is extremely difficult to directly observe the nuclei, which are quite small and moreover obscured by subsequent crystal growth.6,34,35) Therefore, in this study, electrical resistivity measurements were conducted to indirectly observe whether Al3Ti intermetallic particles were formed during the UST process. Although the electrical resistivity measurement is an indirect method, it is possible to measure in situ the microstructural evolution of even nano-sized phases compared to direct methods (e.g., XRD, TEM, etc.).36) A remarkable contrast in the changes in electrical resistivity was observed between the ultrasonicated sample and the untreated sample during an isochronal annealing up to 480°C, as shown in Fig. 7. In the untreated sample, which was not ultrasonicated using the Ti sonotrode and not inoculated using Al–10 mass%Ti as well, a monotonically increasing section followed by a plateau was observed with increasing temperature. By contrast, the sample ultrasonicated at 780°C for 5 min was typified by a region with a lower resistivity compared to the untreated sample (stage I), and an inflection point (stage III), which were able to be explained by the precipitation and decomposition of the Al3Ti phase, respectively. According to the combined XRD (Fig. 8) and SEM-EDS (Fig. 9) analyses and numerical simulation of the solidification process using the non-equilibrium Scheil model (Fig. 10(a)), there were no differences in the phase type and phase fraction in the as-cast state owing to an increase in the Ti content of 0.07 mass% caused by the UST processing. The main secondary phases were Al3Mg2 and Al3Fe in both samples. However, according to the temperature versus equilibrium phase fraction curve of Fig. 10(b), calculated using the equilibrium model to simulate the heat treatment process, the Al3Ti phase that could not be crystallized during the solidification process was precipitated within a low-temperature range of below approximately 320°C. Therefore, it is believed that because Ti has more than 9 times a greater effect on the electrical resistivity decrease per weight in aluminum owing to precipitation in comparison to other alloying elements such as Mg,37) even with an amount of Ti of as small as 0.07 mass%, the increase in the electrical resistivity was slowed by the precipitation of the Al3Ti phase. The inflection point was observed because very fine Al3Ti particles, which were precipitated through solid state transformation at low temperature from an Al matrix containing Ti much less than the peritectic composition, must have been rapidly decomposed near 350°C where the Al3Ti phase was thermodynamically unstable and the temperature was relatively high, resulting in a steep increase in the electrical resistivity. Overall, as the temperature increased in the ultrasonicated sample, the effect of decomposition of fine secondary phase particles such as Al7Cr and Al3Mg2 was added to the temperature rise effect, and thus the increment rate of electrical resistivity tended to increase. In the untreated sample, because the decomposition effect was relatively small compared to the ultrasonicated sample owing to the absence of Al3Ti phase and the presence of coarse secondary phase particles, which were developed along the coarse grain structure, the electrical resistivity increased gradually without an increasing slope as the temperature increased up to approximately 360°C (stages I–III). In addition, as the temperature further increased, the solute homogenization effect owing to an increase in diffusivity was added, and thus a plateau section (stage IV) was observed, where the increase in electrical resistivity owing to the temperature rise was balanced with the decrease in electrical resistivity owing to solute homogenization.36) That is, as described above, the notable contrast in electrical resistivity between the ultrasonicated and the untreated samples provided indirect evidence that Ti, which had been dissolved from the sonotrode during the UST applied under complete liquid conditions, was present as a solute at least during the solidification stage.
Variations in electrical resistivity, which was normalized to that in an as-cast state, with and without the application of UST during continuous heating at a rate of 5°C/min.
X-ray diffraction patterns of ultrasonicated and untreated A5052 aluminum alloy samples in as-cast state.
Scanning electron microscopy [SEM, backscattered electrons (BSE)] microstructural images and EDS analysis results of (a), (b) untreated and (c), (d) ultrasonicated samples: (a), (c) low magnification images and (b), (d) high magnification images.
Phase equilibria calculated using JmatPro for A5052 aluminum alloy with 0.07 mass% Ti using (a) the non-equilibrium Scheil model and (b) the equilibrium model.
In a recent report by Dong et al. on an SEM examination on the surface of a Ti alloy sonotrode used for the UST in the continuous casting of an Al–Zn–Mg aluminum alloy (7050), it has been reported that the Al3Ti intermetallic compound was produced during cavitation erosion. Because the Al3Ti particles are harder and more brittle than the Ti matrix, Al3Ti particles of a few micrometers in size were easily extracted from the sonotrode surface under the cavitation impact and dispersed into the aluminum melt.17) Al3Ti particles most typically act as heterogeneous nucleation sites that refine the grains in Al alloys. In our investigation, however, the experimental results of the GS, DAS, and electrical resistivity all suggest that a growth restriction effect owing to the Ti solutes played a more dominant role in refining the grain than the heterogeneous nucleation effect. The reason why the heterogeneous nucleation effect was not observed in our investigation can be explained by the dissolution reaction of Al3Ti particles. According to previous studies related to the fading mechanism of grain refinement of aluminum alloy using Al–Ti based grain refiners, the dissolution time of the Al3Ti particles is proportional to the square of the size,38) and thus it would take 10 s for the dissolution of particles with a 1 µm diameter and 1000 s for the dissolution of a particle with a 10 µm diameter.39) Therefore, it is estimated that Al3Ti particles of a few micrometers in size, which were produced during cavitation erosion of the Ti sonotrode and dispersed into an aluminum melt, must have dissolved within a couple of minutes. A cavitation induced vigorous melt stream probably shortened the dissolution time further because the dissolution of the intermetallic compounds in the melt is a diffusion-controlled process.40) Because the Ti content was less than a peritectic composition (0.15 mass%), this dissolution of Al3Ti particles must have been irrecoverable until the subsequent solidification process was completed. However, in the inoculated samples containing equal amounts of Ti as in the ultrasonicated samples, although the Ti content was less than the peritectic composition, the size of the Al3Ti particles introduced into the molten metal was hundreds to tens of micrometers,29) and thus the Al3Ti particles were able to survive during the inoculating, stabilizing, and casting processes up to 30 min and become heterogeneous nuclei. Based on the experimental results and theoretical considerations, we concluded that when UST was applied to the Al–Mg base aluminum alloy (A5052) with an intermediate solute content using the Ti sonotrode under a fully liquid condition, the dissolved Ti refined the GS primarily by a solute-induced growth restriction effect rather than by providing heterogeneous nucleation sites.
The effect of UST applied using a Ti sonotrode in a fully liquid stage on the microstructure refinement and mechanical properties of A5052 aluminum alloy, which is a non-heat-treatable Al–Mg based aluminum alloy with an intermediate solute content, was investigated. The conclusions of this study are as follows:
This work was conducted with the support of the Korea Institute of Industrial Technology as “Development of root technology for multi-product flexible production (KITECH EO-22-0006)”.
