【Article source】
Special Casting & Non-ferrous Alloys, Vol. 44, No. 4, 2024
【Citation Format】
程颖，洪涛，易浩，等. TiB2粒子尺寸分布对Al-5Ti-1B晶粒细化效果的影响[J]. 特种铸造及有色合金，2024，44（4）：527-531.
CHENG Y，HONG T，YI H，et al. Influence of TiB2 particle size distribution on grain refinement of Al-5Ti-1B［J］. Special Casting & Nonferrous Alloys，2024，44（4）：527-531.
Guide
As the main phase in Al-5Ti-1B, the size distribution of TiB2 has an important influence on the grain refinement effect. The method of adding TiB2 was used to solve the problem that the size distribution of TiB2 phase is wide and difficult to accurately control and adjust when TiB2 is generated in situ by the traditional fluoride salt method, and the preparation of Al-5Ti-1B with controllable TiB2 size was realized, and the effect of TiB2 size distribution on the refinement effect of Al-5Ti-1B on pure aluminum was studied by combining refinement test and model calculation. The results showed that the refinement effect of Al-5Ti-1B prepared with median particle size of 2.5 μm and 1.4 μm on pure aluminum was 149.2 μm and 137.0 μm respectively under the addition of 0.2% TiB2 and the cooling rate of 1.5 °C/s, which was 46% and 50% lower than that of Al-5Ti-1B prepared by fluoride salt method at 275.6 μm, respectively. Adjusting the size distribution of TiB2 to a finer concentration will further improve the refinement effect of Al-5Ti-1B.
【Background】
Aluminum alloy is a lightweight material with broad application prospects, and refining grains is an important way to improve its casting performance, strength and plasticity. Among the many grain refinement methods, adding a refiner to the melt for inoculation treatment is the most commonly used grain refinement method, among which Al-5Ti-1B is the most widely used commercial aluminum grain refiner. The phases of Al-5Ti-1B are TiAl3 and TiB2, and the nucleation particles with high nucleation potential (TiB2 covered with TiAl3 modified layer) and solute elements (Ti) with high growth limiting factors are the key reasons why Al-5Ti-1B can effectively refine the grain size of aluminum alloys.
Although Al-5Ti-1B has achieved good refinement results on aluminum alloys, only a small amount of TiB2 particles can play the role of heterogeneous nucleation substrate in the refinement process. In order to solve the problem of low utilization efficiency of TiB2 particles, researchers have carried out a lot of work to analyze the causes of TiB2 particles, among which the free growth theory has the highest recognition. According to the free growth theory, the supercooling degree required for the formation of α-Al on large-size TiB2 particles is lower than that of small-size TiB2 particles, and when the latent heat of crystallization released by the grains formed first on large-size TiB2 particles during growth is higher than that of the heat dissipated by cooling and causing reglow, the supercooling degree of the melt cannot be further reduced, and the small-size TiB2 particles cannot become an effective heterogeneous nucleation substrate because the maximum supercooling degree that can be achieved by the melt still cannot meet the supercooling conditions required for its free growth.
【Research Highlights】
At present, the commercial Al-5Ti-1B refiners are prepared by fluoride salt method, in which the size distribution range of TiB2 particles is wide and has the characteristics of long tail, that is, the number of small-sized particles is large and the size distribution is concentrated, and the number of large-size particles is small and the size distribution is relatively scattered. Therefore, grains can only be formed on a small number of large-size TiB2 particles during the refinement process. In order to further improve the refinement effect of Al-5Ti-1B, the size distribution of TiB2 particles needs to be improved, so that more TiB2 particles can be activated into effective heterogeneous nucleation substrates during the refinement process. In order to solve the problem that the size range of TiB2 particles generated in situ in situ in the fluoride salt process is wide and difficult to accurately control and adjust, the joint research team of the Institute of Frontier Science and Technology Innovation of Beijing University of Aeronautics and Astronautics, Ningbo Innovation Research Institute of Beijing University of Aeronautics and Astronautics, and Zhucheng Hangda New Material Technology Co., Ltd. published an article entitled "The Effect of TiB2 Particle Size Distribution on the Grain Refinement Effect of Al-5Ti-1B" in the journal Special Casting and Nonferrous Alloys, Volume 44, Issue 4, 2024. In order to provide a reference for optimizing the size distribution of TiB2 particles and improving the refinement performance of Al-5Ti-1B, the authors prepared Al-5Ti-1B by adding TiB2 particles.
Click on the image to read the original article
【Methodology】
The raw materials used in this test mainly include commercial pure aluminum (CP-Al), TiB2 powder, Ti powder (325 mesh) and Al-5Ti-1B refiner. Among them, TiB2 powder was pre-sieved into two sizes: 2.5 μm and 1.4 μm with median particle size (D50), respectively, and the size distribution is shown in Table 1. The raw materials and dosages for the preparation of Al-5Ti-1B are shown in Table 2.
According to the refiner composition, the required quality of aluminum ingots, pre-sieved TiB2 powder and Ti powder were weighed, dried at 75 °C for 1 h before use, and then the two powders were mixed evenly and wrapped in aluminum foil for later use. First, the aluminum ingot is heated and melted in a resistance furnace, and after the aluminum ingot is completely melted and heated to 900 °C, the mixed powder of TiB2 and Ti wrapped in aluminum foil is added to the melt, kept warm for 30 min, and the melt is stirred with a graphite stir rod every 10 min to ensure that the powder is evenly mixed into the melt. After the heat preservation is completed, the melt is cooled to 800 °C, the scum on the melt surface is skimmed, and the melt is poured into a metal mold with a preheating temperature of 175 °C, and Al-5Ti-1B refiner is obtained after cooling and solidification.
Pure aluminum was used as the refinement object to evaluate the refinement effect of the refiner. As a comparison, a commercially available Al-5Ti-1B refiner was used for refinement tests. Firstly, the aluminum ingot to be refined is heated and melted in a resistance furnace, and after the aluminum ingot is completely melted and heated to 700 °C, 0.2% refiner is added to the melt and continued to be kept warm for 10 min, and then the melt is poured into a metal mold with a preheating temperature of 175 °C, and the refined pure aluminum is obtained after cooling and solidification at a cooling rate of 1.5 °C/s.
A scanning electron microscope (Zeiss EVO 10) with EDS spectrometer was used to observe and analyze the microstructure of the refiner, and the size of TiB2 particles in the refiner was counted. Image-Pro Plus 6.0 software was used for grain size statistics and calculations. In order to facilitate the differentiation of different grains, the samples were anodically coated before observation, and the polarized light mode of the SG-51 metallurgical microscope was observed and the tissue photos were taken, which were imported into the Image-Pro Plus 6.0 and the average grain size was calculated by the truncation method.
[Graphic Analysis]
The composition of the second phase and the morphology and size characteristics of each phase in Al-5Ti-1B prepared by TiB2 are the same as those of the commercially available Al-5Ti-1B refiner prepared by fluoride salt method. In addition to α-Al as the matrix, there are also two phases of TiAl3 and TiB2 in Al-5Ti-1B, in which TiAl3 is a relatively large block and TiB2 is a relatively small granular. Among them, the average particle size of TiB2 particles in No. 1 refiner was the smallest, which was 1.7 μm, and the standard deviation σ was 2.56, and the average particle size of TiB2 particles was 2.8 μm, and the standard deviation σ was 3.46, and the average particle size of TiB2 particles was the largest, which was 3.2 μm, and the distribution was the widest, with a standard deviation σ of 5.31. Different from the chemical reaction between KBF4, K2TiF6 and Al in the fluoride salt process, the TiB2 particles in Al-5Ti-1B prepared in this study were introduced by additive. Due to the complexity of the in-situ generation process of TiB2 in the fluoride salt method, it is still difficult to accurately control and adjust the size of TiB2 particles by process means. In contrast, the introduction of TiB2 by external method can accurately control its particle size distribution by prior screening, and the TiB2 particles with smaller average size and more concentrated particle size distribution can be successfully obtained in Al-5Ti-1B.
Fig.1 Microstructure and EDS spectroscopy of different Al-5Ti-1B refiners
Fig.2 Particle size distribution of TiB2 particles in different Al-5Ti-1Bs
Fig. 3 is the grain structure of pure aluminum after treatment with different Al-5Ti-1B refiners, and Fig. 4 is the statistical results of the corresponding average grain size. It can be seen that the grain size of pure aluminum without refiner is very coarse after cooling and solidification, with an average grain size of 1 000 μm. After adding 0.2% commercially available Al-5Ti-1B refiner, the as-cast microstructure of pure aluminum was significantly refined, and the average grain size was reduced to 275.6 μm, but the refinement effect was slightly worse than that reported at 220 μm. This is because in addition to the properties of the refiner itself, the refinement effect is also closely related to the cooling rate during the solidification process, and the grain size increases with the decrease of the cooling rate. In this study, the refinement test was carried out at a relatively low cooling rate (1.5 °C/s), so the grain size after refinement was larger than that reported by Morty B S et al.
Fig.3 Pure aluminum grain structure after treatment with different Al-5Ti-1B refiners
Fig.4 Average grain size of pure aluminum after different Al-5Ti-1B refinement treatments
Compared with No. 3 refiner, Al-5Ti-1B prepared with external TiB2 had a better refinement effect under the same addition amount (0.2%) and cooling rate (1.5 °C/s). The average grain sizes of No. 1 and No. 2 refiners were 149.2 μm and 137.0 μm, respectively, which were 46% and 50% lower than those of commercially available No. 3 refiners, respectively. The results show that the refinement effect of Al-5Ti-1B can be significantly improved by reasonably adjusting the size distribution of TiB2. Combined with the size distribution of TiB2 in Figure 2 and the refinement results in Figure 4, it can be seen that the No. 2 refiner with a smaller median particle size of D50 and a more concentrated size distribution of TiB2 particles has a better effect on the size refinement of pure aluminum grains.
In order to further study the influence of TiB2 size distribution on the refinement effect of Al-5Ti-1B, a grain size prediction model was established based on the free growth theory, and the quantitative relationship between TiB2 size distribution and grain refinement effect was analyzed, and the optimization basis of TiB2 size distribution was theoretically given.
The basic idea of the model is as follows: the melt solidifies at a fixed cooling rate (dT/dt) and the temperature of all parts of the melt is equal during the cooling process, and the grains are based on TiB2 as the heterogeneous nucleation substrate. In each time step (t), the number of newly formed grains and the crystallization latent heat released by the growth of formed grains are calculated sequentially according to the current melt supercooling, and the melt supercooling degree of the next time step is determined according to the values of crystallization latent heat and cooling heat dissipation. This process is repeated until the melt is reglowing, after which no new grains are formed, only existing grains grow. The average grain size can be calculated based on the number of grains formed when the melt is reglowed. The main parameters used in the calculation of the model are shown in Table 4.
Fig.5 Effect of TiB2 particle size distribution on grain refinement
It can be seen that when the geometric average size d0 of TiB2 is within 1.0~3.0 μm, the average grain size gavg decreases with the decrease of d0. Under the same d0 condition, the gavg decreases with the decrease of the geometric standard deviation of the TiB2 size distribution. The results of the model show that when the size distribution of TiB2 as a heterogeneous nucleated substrate in Al-5Ti-1B is more fine and concentrated, the grain refinement effect of α Al-5Ti-1B is better, which is consistent with the grain refinement test results. As can be seen from Figure 5, the grain refinement effect of No. 2 refiner (137.0 μm) is better than that of No. 1 refiner (149.2 μm), and the difference between the two refiners is mainly reflected in the use of TiB2 powder with different size distributions in preparation. It can be seen from Table 1 that the size of TiB2 added to the preparation of No. 1 refiner is relatively large, and the size distribution is relatively more dispersed (≥97% of the TiB2 particles are distributed in the size range of 0.2~10 μm). When preparing No. 2 refiner, the size of the external TiB2 was relatively small, and the size distribution was relatively more concentrated (≥97% of the TiB2 particles were distributed in the size range of 0.1~5 μm). Therefore, both the model and the experimental results show that adjusting the size distribution of TiB2 particles to a finer concentration will be conducive to the further improvement of the refinement effect of Al-5Ti-1B.
【Main conclusions】
(1) The size distribution of TiB2 particles is the main factor affecting the refinement effect of Al-5Ti-1B, and the preparation of Al-5Ti-1B alloy by adding TiB2 can achieve accurate control and adjustment of TiB2 size, which is a feasible way to optimize the size distribution of TiB2 and improve the refinement effect of Al-5Ti-1B. Under the conditions of 0.2% TiB2 addition and 1.5 °C/s cooling rate, the pure aluminum prepared by TiB2 with median particle sizes of 2.5 μm and 1.4 μm was refined to 149.2 μm and 137.0 μm, respectively, which was 46% and 50% lower than that of Al-5Ti-1B prepared by conventional fluoride salt method, which was reduced by 46% and 50%, respectively.
(2) Under the same conditions, the smaller the average size and size dispersion of TiB2 in Al-5Ti-1B, the greater the number of TiB2 that can become an effective heterogeneous nucleation substrate during the refinement treatment, and the smaller the average grain size in the as-cast structure formed after the refinement treatment. In order to obtain a better grain refinement effect, it is necessary to adjust and optimize the size distribution of TiB2 in Al-5Ti-1B to a finer concentration.