Study on the Refining Behavior of Al-Si Alloys Doped with Al-Ti-C Intermediate Alloys at Different B Doping Temperatures

Abstract

In this paper, a novel Al-5Ti-0.3C-0.2B intermediate alloy with an efficient refining effect on Al-Si alloys by a low-cost and high-performance fabrication process is developed. The results show that the addition of only 0.15 wt.% of Al-5Ti-0.3C-0.2B intermediate alloy is able to refine the average grain size of Al-Si alloys to 120 ± 5 μm, which is 28.7% higher than the reference value of 0.2 wt.% addition. Additionally, TiAl₃ in bulk form has a faster response and more stable refinement than TiAl₃ in needle form. The blocky TiAl₃ generated at low temperature will turn acicular or elongated as the temperature increases, slowing the response speed and increasing the time required to achieve the ideal refinement effect. We are convinced that this work will provide a reference for developing new aluminum alloy refinement intermediate alloys.

1. Introduction

Aluminum–silicon casting alloys play an important role in the field of industrial production [1]. The A356 alloy was developed as a series of aluminum alloys in the United States in the mid-twentieth century and is widely used in many fields due to its excellent castability, high strength, and low density [2,3,4,5]. However, if the α-Al phase in the A356 alloy has a coarse dendritic morphology accompanied by coarse acicular or lamellar eutectic Si phases, the mechanical properties of the alloy can be seriously weakened [6,7]. Therefore, to improve the mechanical properties of sub-eutectic Al-Si alloys, it is important to investigate how to refine the grains of the α-Al phase and metamorphose the eutectic Si phase.

In the past few decades, the grain refinement effects of Al-Ti-B ternary intermediate alloys, especially Al-5Ti-1B intermediate alloys, have been excellent in continuous casting and semicontinuous casting deformed alloys but poor in casting alloys, especially in Al-Si alloys with Si contents greater than 3% [8,9], which is due to the fact that Si in the melt reacts with Ti to form a Ti-Si phase at the expense of TiAl₃ particles, thus affecting the particle refinement efficiency of the intermediate alloys [8,10]. After the Al-Ti-C intermediate alloy is added to the aluminum melt, its refining effect shows a significant decline or even disappears completely with the extension of the holding time [11]. The synthesis of stoichiometric TiC (C/Ti = 1) is known to be very difficult because of the vacancies that are always present during high temperature synthesis, and therefore C/Ti ratios are normally in the range from 0.48 to 1 [12]. The existence of vacancies also provides conditions for the doping of dissimilar atoms, especially elements close to C, with a smaller radius and electronic structure similar to C and B, thus improving the stability of TiC_x, which can effectively solve the refining agent “Si poisoning” phenomenon [8]. In the previous preparation process of Al-Ti-C-B, the raw material used was pure Ti (99.8%) [13], with a melting point of 1667 °C. At low temperatures, Ti is difficult to dissolve and absorb into the aluminum melt; if the temperature is too high to reach the melting point of Ti, it has strict requirements for the equipment and poses a significant safety hazard.

We have proposed an effective method for refining the alloy based on the problems faced by its production. The main purpose of this article is to use the melt reaction method to replace Ti with K₂TiF₆ to provide a Ti source. In the process of preparing Al-5Ti-0.3C intermediate alloy, B was introduced in the form of Al-3B intermediate alloy to prepare Al-5Ti-0.3C-0.2B. The grain refinement performances of intermediate alloys on A356 alloy were investigated.

2. Materials and Methods

An Al-5Ti-0.3C-0.2B intermediate alloy was prepared with industrially pure aluminum (99.7%, 5 μm), graphite powder (99.9%, 1 μm), potassium fluorotitanate (99.85%), and commercially available Al-3B intermediate alloy using the intermediate frequency furnace melt reaction method. Two seed alloys were prepared with the same process at low (named M1 alloy) and high temperatures (named M2 alloy), with the percentage difference in temperature being 24.1%. They were poured into a graphite mold with dimensions of φ 30 mm×15 mm. The main differences between these two alloys were the morphology and distribution of TiAl₃ particles in the melt. M1 and M2 seed alloys at 0.15 wt.% were added to A356 molten bath to compare their refining effects. The chemical composition of A356 alloy is shown in

Table 1. The chemical composition of the A356 alloy.

	Si	Mg	Fe	Ti	Sr	Al
A356	6.34	0.33	0.07	0.12	0.024	Balance

. First, the base alloy was melted in a resistance furnace at 730 °C and held for 30 min, and then the seed alloys were added and stirred well; then, it was poured into preheated KBI cylindrical steel molds with sizes of φ 35 mm × 10 mm for 15 min, 90 min, and 180 min and poured on the refractory bricks coated with graphite powder on the surface. After curing, the bottoms of the refined specimens in contact with the bricks were etched with aqua regia with HCl:HNO₃ at 3:1 (volume fraction composition). Macrostructural photographs of each sample were taken with a high magnification microscope, and the average particle size was determined using the linear intercept method. Metallographic specimens were taken from the center of the cross-section of each specimen, and the composition of the phases of the extracted particles was identified by X-ray diffraction (XRD, Rigaku D/maxrb). The phases and the microstructures were characterized using a scanning electron microscope (SEM).

3. Results and Discussion

Figure 1a–f shows the microstructures and elemental distributions of Al-5Ti-0.3C-0.2B. Figure 1a shows the image of the bulk intermediate alloy. Figure 1b,c depicts backscattering scanning electron (BSE) images of M1 at low and high magnification, respectively. As shown in Figure 1d–f, TiAl₃ and TiC are detected on the α-Al matrix, in addition to a trace amount of B-doped TiC. The statistical results show that the average diameter of B-doped TiC is 3 μm ± 1 μm, and the average diameter of TiAl₃ is 8 μm ± 2 μm. In addition, as shown by the enlarged SEM images and corresponding EDS results(Figure 1c,f), the particles are constituted by two types of compounds, which simultaneously contain Ti, C, and B elements. Combined with the XRD spectra (Figure 1d) and compared with the similar literature [14], it can be determined that the particles are an endo-doped B-TiC complex.

In Figure 2a,b, it can be noticed that the morphology of TiAl₃ changed from massive at low temperature to needle-like at high temperature. It can be found that different temperatures and holding times had a great influence on the morphology and distribution of the particles in the second phase (i.e., TiAl₃ and TiC). M1 seed alloy was prepared at low temperatures, TiAl₃ is unstable during the early stage of its formation, which is easily dissolved in the aluminum liquid. As the reaction progresses, undissolved blocky TiAl₃ in the aluminum molten and Ti-rich area with composition fluctuation were formed around TiAl₃. Therefore, the aluminum molten is not uniform. During the solidification process, the precipitated TiAl₃ either attached to the existing TiAl₃ particles or nucleated independently. Due to the presence of a supersaturated region, its preferred growth direction was not obvious, so it grew uniformly, resulting a polyhedral block or ball. M2 seed alloy was prepared at high temperature, and the TiAl₃ which formed in the early stage all dissolved. The concentration of Ti in the aluminum melt was evenly distributed, and the melt was in a uniform state, so there was no obvious aggregation of the TiAl₃ phase in the alloy. However, the high temperature promoted the formation of the TiAl₃ phase, which had an obvious preferential orientation and gradually changed from a blocky shape to long strips and needle sheets.

Figure 3a–j show the grain macrostructures of A356 refined by 0.15 wt.% Al–Ti–C–B after holding for different times. The unrefined A356 grains are very coarse, and the average grain size is about 1570 μm, as shown in Figure 3a. Figure 3b shows the average grain sizes for different holding times with the addition of two intermediate alloys (i.e., M1 and M2). Comparing Figure 3c–f with Figure 3g–j, it is obvious that the M1 alloy has a better refining effect that can refine the grains of A356 alloy to 120 μm ± 5 μm. It is 28.7% higher than that of adding 0.2 wt.%, which can refine the grains of the A356 alloy to 167 μm ± 8 μm [13]. Figure 3g shows that there is still a certain gap in the refining effect between M1 and M2 at the early stage of the addition of M2 and M1, because there are acicular TiAl₃ phases in the M2. The TiAl₃ phase with the blocky morphology can provide more nucleation planes, which are easy to dissolve and react faster. The acicular TiAl₃ phase only has the (110) plane facing the melt to provide nucleation sites, and the dissolution and reaction speed during the refining process is much slower than that of the blocky TiAl₃ phase. When the needle-like TiAl₃ is completely dissolved in the melt as the holding time increases, the M2 seed alloy will also have the similar refinement effect to that of the M1, as shown in Figure 3f,j. During the refining process, the M1 alloy always has a good refining effect, so it has more practical advantages than the M2 alloy.

The microscopic particles present in the intermediate alloy Al-5Ti-0.3C-0.2B are predominantly TiB₂, TiC, and B-doped TiC particles. It is shown that TiB₂ particles in aluminum alloys have a strong tendency to aggregate relative to TiC particles, but in the intermediate alloy Al-5Ti-0.3C-0.2B, the agglomeration of TiB₂ particles is significantly reduced and uniformly distributed. Excellent stable TiCB particles were obtained by doping trace amounts of B into the TiC particles in the intermediate alloy. The addition of refiners to the A356 melt provided more substrates for α-Al nucleation and improved the refining effect and recession resistance of the A356 alloy. When the intermediate alloy Al-5Ti-0.3C-0.2B is added to the aluminium melt, due to the relatively small mismatch of the lattice of the microstructure between the two TiC and α-Al [15,16], and with the assistance of the bulk TiAl₃ the second phase particles TiC and TiB₂ have a good activity and stability. The increase in the amount of TiC particles as α-Al heterogeneous nuclei is followed by an increase in the grain refinement efficiency.

In general production processes, this often involves the use of complex processing equipment or high-temperature, high-pressure, and inert gas environments, which not only makes the production process more complex but also significantly increases the production costs and time [17,18,19,20]. The refining process in this paper avoids high-temperature, high-pressure and special gas protection environments and still achieves good refining results, meaning that it has positive potential for commercial applications.

4. Conclusions

A novel Al-5Ti-0.3C-0.2B intermediate alloy with an efficient refinement behavior for Al-Si alloy through a kind of low-cost and excellent fabrication process was developed and its refinement effect and mechanism were discussed, drawing the following conclusions:

(1)

The average diameters of TiAl₃ and TiC were 8 μm ± 2 μm and 3 μm ± 1 μm, respectively, which were uniformly distributed on the α–Al matrix;

(2)

The blocky TiAl₃ generated at low temperature became acicular or elongated with the increase in temperature, slowing the response speed and increasing the time required to achieve the ideal refinement effect;

(3)

By adding 0.15 wt.% Al-5Ti-0.3C-0.2B intermediate alloy, the grain size of A356 alloy was refined to 120 μm ± 5 μm and maintained for 180 min without recession.