Show Summary Details
More options …

# High Temperature Materials and Processes

Editor-in-Chief: Fukuyama, Hiroyuki

Editorial Board: Waseda, Yoshio / Fecht, Hans-Jörg / Reddy, Ramana G. / Manna, Indranil / Nakajima, Hideo / Nakamura, Takashi / Okabe, Toru / Ostrovski, Oleg / Pericleous, Koulis / Seetharaman, Seshadri / Straumal, Boris / Suzuki, Shigeru / Tanaka, Toshihiro / Terzieff, Peter / Uda, Satoshi / Urban, Knut / Baron, Michel / Besterci, Michael / Byakova, Alexandra V. / Gao, Wei / Glaeser, Andreas / Gzesik, Z. / Hosson, Jeff / Masanori, Iwase / Jacob, Kallarackel Thomas / Kipouros, Georges / Kuznezov, Fedor

IMPACT FACTOR 2018: 0.427
5-year IMPACT FACTOR: 0.471

CiteScore 2018: 0.58

SCImago Journal Rank (SJR) 2018: 0.231
Source Normalized Impact per Paper (SNIP) 2018: 0.377

Open Access
Online
ISSN
2191-0324
See all formats and pricing
More options …
Volume 37, Issue 1

# Study of the Boron Distribution and Microstructure of Solidified Al-Si Alloy During the Process of Silicon Purification

Yanlei Li
• Corresponding author
• Energy-Saving Building Materials Collaborative Innovation Center of Henan Province, Xinyang Normal University, Xinyang 464000, P. R.China
• Email
• Other articles by this author:
• De Gruyter OnlineGoogle Scholar
/ Jian Chen
• Institute of Applied Technology, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031,P. R. China
• Other articles by this author:
• De Gruyter OnlineGoogle Scholar
/ Songyuan Dai
• State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources, North China Electric Power University, Beijing 102206,P. R. China
• Other articles by this author:
• De Gruyter OnlineGoogle Scholar
Published Online: 2017-02-28 | DOI: https://doi.org/10.1515/htmp-2016-0090

## Abstract

The Al-Si melts that contain different silicon contents were solidified with a series of cooling rates, and the boron contents in primary silicon phases and eutectic silicon phases were measured and discussed. The results indicate that the boron content in the eutectic silicon phases is higher than that in the primary silicon phases when the cooling rate is constant. When the cooling rate decreases, the boron content in the primary silicon phases decreases, but the boron content in the eutectic silicon phases increases. The microstructure observations of solidified ingots show that there is an interface transition layer beside the primary silicon phase, and the average width of the interface transition layer increases with decreasing cooling rate.

PACS: 81.10.Fq

## Introduction

Solar cell is an important device to utilize renewable energy. Solar grade silicon is the raw materials for it, and its purity is 99.9999 %. Traditional processes of purifying silicon use chemical methods [1]. They change silicon into compounds such as SiHCl3 and SiH4. These compounds are purified, and then they are reduced into silicon of high purity. These processes are complicated and with high energy consumption. New methods to purify silicon with simple procedures and low energy cost, for example via metallurgical processes, are needed.

When silicon is purified by the metallurgical processes, boron is difficult to remove because of its high segregation coefficient (value of 0.8) in silicon [2]. However, it can be removed effectively by solidification of hypereutectic Al-Si melt because thermodynamics calculations show that its segregation coefficient decreases in the Al-Si melt [3, 4]. After the solidification of the Al-Si melt, silicon exists in two kinds of phases, primary silicon phase and eutectic silicon phase. They form in different stages of the solidification process. The primary silicon phase is the product of uniform grain growth in the early stage, and the eutectic silicon phase forms in the eutectic reaction at the final solidification stage. The boron content in the primary silicon phase has been studied by many researchers [5, 6, 7, 8, 9, 10, 11, 12, 13, 14], but the boron content in the eutectic silicon phase has never been studied. So it is investigated in this work.

In this work, the Al-Si melts with different silicon contents were solidified with a series of cooling rates. The boron contents in the primary silicon phases and eutectic silicon phases were measured and discussed. The macrostructures and microstructures of solidified samples were investigated, and the interface transition layer beside the primary silicon phase was analyzed and discussed.

## Experimental

The raw materials used in this work are industrial grade aluminum and metallurgical grade silicon, and their impurities contents are shown in Table 1.

Table 1:

Impurity content in the raw materials (ppmw).

Approximate 250 g Al-Si alloy (the silicon content is shown in Table 2) was put in an alumina crucible (I.D.: 55 mm, and depth: 100 mm), and the crucible was placed in a resistance heating furnace. The temperature of the furnace rises up to 1,323 K, and the alloy melts. After that, the melt was stirred by a quartz rod to mix it completely, and then the melt was cooled to 830 K (20 K below the eutectic temperature) with the cooling rate shown in Table 2. After that, the sample was furnace cooled to room temperature.

Table 2:

The silicon contents and cooling rates of the samples.

The solidified ingot was cut into two parts along its vertical axis. The cutting surface of one part was scanned by an optical scanner for morphology analysis. Then the surface was ground and polished, and it was etched with 0.5 % HF solution for about 50 s. Microstructure was analyzed by optical microscopy (OM). The other part was put in hydrochloric acid to dissolve eutectic aluminum. After the reaction, primary silicon phases (the shape of flake) and eutectic silicon phases (the shape of powder) appear. Then these primary silicon phases and eutectic silicon phases were washed with deionized water and dried. Then they were separated by the sieve of 0.5 mm’s mesh size. The boron contents in them were determined by inductively coupled plasma optical emission spectrometry (ICP-OES).

## Boron distribution in the primary silicon phases and eutectic silicon phases

Figure 1 shows the boron contents in the primary silicon phases and eutectic silicon phases in the Al-30 % Si alloys. It can be found that when the cooling rate decreases from 31.89 to 0.55 mK/s, the boron content in the primary silicon phases decreases from 4.9 to 2.0 ppmw, and the boron content in the eutectic silicon phases increases from 10.3 to 12.1 ppmw. In other words, when the cooling rate decreases, the boron content in the primary silicon phases decreases, but the boron content in the eutectic silicon phases increases. In addition, when the cooling rate is constant, the boron content in the eutectic silicon phases is higher than that in the primary silicon phases. Figure 2 shows the boron contents in the primary silicon phases and eutectic silicon phases in the Al-40 % Si alloys, and it shows the same trend.

Figure 1:

Boron contents in the primary silicon phases and eutectic silicon phases in the Al-30 % Si alloys.

Figure 2:

Boron contents in the primary silicon phases and eutectic silicon phases in the Al-40 % Si alloys.

In hypereutectic Al-Si melt, the primary silicon phase begins to grow when temperature falls below its liquidus. At the same time, boron atoms are excluded because its segregation coefficient is smaller than 1. So boron atoms are enriched beside the primary silicon phase, and then they diffuse into the remaining melt. The melt zone beside the primary silicon phase can be considered as a boundary layer. The boron content in the primary silicon phase is determined by the boron segregation coefficient in the Al-Si melt and the boron content in the boundary layer, and it is shown in eq. (1). ${C}_{s}=k{C}_{i}$(1)

where Cs, k, Ci are the boron content in the primary silicon phase, the boron segregation coefficient in the Al-Si melt and the boron content in the boundary layer respectively.

When the cooling rate decreases, the degree of supercooling of the melt decreases, and the growth rate of the primary silicon phase decreases. The boron atoms in the boundary layer have enough time to diffuse into the remaining melt, so the boron concentration in the boundary layer decreases. According to eq. (1), the boron content in the primary silicon phase decreases.

When the temperature of the melt decreases, the primary silicon phase grows, and the composition of the remaining melt changes according to the liquidus line. When the temperature is 850 K and the silicon weight ratio is 12.6 %, the eutectic reaction happens [15], and the eutectic silicon phase and the eutectic aluminum phase form. If the cooling rate is small, the boron content in the primary silicon phases will be low, and the boron concentration in the melt will be high. Besides, the boron redistribution during the eutectic reaction is not large because the eutectic reaction happens at the eutectic temperature with relatively fast rate, and there is no much time for boron atoms to diffuse. Then the boron content in the eutectic silicon phase is high. On the contrary, if the cooling rate is large, the boron content in the eutectic silicon phase is low.

The primary silicon phases and eutectic silicon phases form in different solidification stages. The primary silicon phases grow first and reject boron atoms. Then the eutectic reaction happens, and the eutectic silicon phases form. At the time, the boron concentration in the melt is higher than that in the primary silicon phases. Besides, the boron redistribution during the eutectic reaction is not large as discussed before. Then the boron content in the eutectic silicon phases is higher than that in the primary silicon phases.

## The macrostructure of the solidified samples

Figure 3 shows the macrostructures of the Al-40 % Si with different cooling rates. Primary silicon phases (A in Figure 3(b)) and eutectic Al-Si matrix (B in Figure 3(b)) can be found in the images. The macrostructure observations indicate that when the cooling rate decreases, the quantity of the primary silicon phases decreases, and their widths increase.

Figure 3:

Scanning images of cross sections of the Al-40 % Si with different cooling rates (a, b, c, d, and e cooled at 46.78, 19.49, 6.50, 2.44 and 0.81 mK/s respectively).

Crystal growth has two steps, nucleation and growth. In hypereutectic Al-Si melt, the primary silicon phases begin to grow when temperature drops below its liquidus. When the cooling rate decreases, the degree of supercooling of the melt decreases, and the amount of nucleation decreases. Besides, the primary silicon phases grow slowly at the same time. Then the silicon atoms around the primary silicon phases have more time to diffuse to the primary silicon phases, and they are absorbed. According to the level rule and the Al-Si phase diagram [15], the mass fraction of the primary silicon phase is constant for Al-40 % Si. So when the cooling rate decreases, the quantity of the primary silicon phases decreases and their widths increase.

## Interface transition layer

Figure 4 shows OM image of the Al-40 % Si alloy whose cooling rate is 0.81 mK/s. The large primary silicon phase (A), irregular small eutectic silicon phases (B) and eutectic aluminum matrix (C) can be seen in this image. The layer (D) beside the primary silicon phase is an area with no eutectic silicon phase, and it is marked with a rectangle. It can be considered as an interface transition layer. The average width of this layer was calculated by averaging widths of more than 40 layers in the same sample.

Figure 4:

OM image of the Al-40 % Si alloy cooled at 0.81 mK/s.

The reason of the formation of the interface transition layer can be explained as following. When the temperature drops below the liquidus temperature, a silicon crystal nucleus forms in the hypereutectic Al-Si melt. It will absorb silicon atoms from the surrounding melt and grow into a primary silicon phase. In other words, the silicon atoms around this solidified phase will diffuse to it, while aluminum atoms diffuse away from it. At last, the layer beside this primary silicon phase becomes rich in aluminum and poor in silicon. When the eutectic reaction happens, the eutectic silicon phase can not form there because of lack of silicon and non-equilibrium solidification. When the melt solidifies, it appears as an interface transition layer.

Figure 5 shows the average width of the interface transition layer. In Al-40 % Si alloy, when the cooling rate decreases from 46.78 to 0.81 mK/s, the average width of the interface transition layer increases from 75 to 113 μm, and it indicates that the average width of the interface transition layer increases with the decrease of cooling rate. The same trend can also be found in Al-30 % Si alloy. In addition, silicon content affects the average width of the interface transition layer.

Figure 5:

Average width of the interface transition layer.

If the cooling rate is slow, the primary silicon phase grows slowly. Silicon atoms have enough time to diffuse to the primary silicon phase, and aluminum atoms have enough time to diffuse away from it. As a result, the average width of the interface transition layer increases.

As discussed before, the primary silicon phase grows slowly at low cooling rate, and the silicon and aluminum atoms have enough time to diffuse. When silicon content increases form 30 % to 40 %, there are more silicon atoms around the primary silicon phase. Only those which are close to the primary silicon phase can diffuse to it, and then they are absorbed. So the average width of the interaction transition layer decreases. But when cooling rate is large, the primary silicon phase grows fast, and the silicon and aluminum atoms around the primary silicon phase do not have enough time to diffuse. Moreover, according to the Al-Si phase diagram [15], the temperature ranges of primary silicon solidification are from 850 K to 1,117 K for Al-30 % Si, and from 850 K to 1,251 K for Al-40 % Si alloy. When the solidification temperature increases, the diffusion coefficients of silicon and aluminum in the Al-Si melt increase. When silicon content increases form 30 % to 40 %, the solidification temperature increases, and the silicon and aluminum atoms diffuse faster. So the average width of the interaction transition layer increases.

## Conclusion

Boron distribution in the primary silicon phases and eutectic silicon phases after the solidification of the Al-Si melts is studied in this work. The results show that when the cooling rate is constant, the boron content in the eutectic silicon phases is higher than that in the primary silicon phases. When the cooling rate decreases, the boron content in the primary silicon phases decreases, but the boron content in the eutectic silicon phases increases. The macrostructures of solidified samples demonstrate that when the cooling rate decreases, the quantity of the primary silicon phases decreases, but their widths increase. In addition, the average width of the interface transition layer increases with the decrease of cooling rate.

## References

• [1]

P. Woditsch and W. Koch, Sol. Energ. Mat. Sol. C., 72 (2002) 11–26.

• [2]

F.A. Trumbore, Bell System Tech. J., 39 (1960) 205–233.

• [3]

K. Tang, E.J. Øvrelid, G. Tranell and M. Tangstad, Mater. Trans., 50 (2009) 1978–1984.

• [4]

L.X. Zhao, Z. Wang, Z.C. Guo and C.Y. Li, Trans. Nonferrous Met. Soc. China, 21 (2011) 1185–1192.

• [5]

J.W. Li, X.L. Bai, Y.L. Li, B.Y. Ban and J. Chen, High Temp. Mater Processes, 34 (2015) 833–838. Google Scholar

• [6]

Y.L. Li, J. Chen, B.Y. Ban, T.T. Zhang and S.Y. Dai, High Temp. Mater Processes, 34 (2015) 43–49. Google Scholar

• [7]

X. Gu, X.G. Yu and D.R. Yang, Sep. Purif. Technol., 77 (2011) 33–39.

• [8]

M.W. Ullah and T. Carlberg, J. Cryst. Growth, 318 (2011) 212–218.

• [9]

P.P. Wang, H.M. Lu and Y.S. Lai, J. Cryst. Growth, 390 (2014) 96–100.

• [10]

W.Z. Yu, W.H. Ma, G.Q. Lv, H.Y. Xue, S.Y. Li and Y.N. Dai, J. Cryst. Growth, 405 (2014) 23–28.

• [11]

Y.Q. Li, Y. Tan, J.Y. Li, Q. Xu and Y. Liu, J. Alloy. Compd., 583 (2014) 85–90.

• [12]

T. Yoshikawa, K. Arimura and K. Morita, Metall. Mater. Trans. B, 36 (2005) 837–842. Google Scholar

• [13]

T. Yoshikawa and K. Morita, ISIJ Int., 47 (2007) 582–584.

• [14]

J.W. Li, Z.C. Guo, H.Q. Tang, Z. Wang and S.T. Sun, Trans. Nonferrous Met. Soc. China, 22 (2012) 958–963.

• [15]

J.L. Murray and A.J. Mcalister, Bull. Alloy Phase Diag., 5 (1984) 74–84.

## About the article

Accepted: 2017-01-12

Published Online: 2017-02-28

Published in Print: 2018-01-26

This work was financially supported by 2016 Key Research Program of University in Henan Province (No. 16A430006), 2015 Program of Educational Technique & Equipment and Practical Education of Henan Province (No. GZS142), Doctoral Scientific Research Foundation of Xinyang Normal University, and Nanhu Scholars Program for Yong Scholars of XYNU.

Citation Information: High Temperature Materials and Processes, Volume 37, Issue 1, Pages 69–73, ISSN (Online) 2191-0324, ISSN (Print) 0334-6455,

Export Citation

© 2018 Walter de Gruyter GmbH, Berlin/Boston.