Abstract
In-situ magnesium silicide/aluminum (Mg2Si/Al) composites were fabricated by laser deposition. A dynamic model of in-situ Mg2Si/Al composites was established. The laser power (system temperature), Mg-rich layer thickness, Si particle size, and Al content were identified as the main factors affecting the reaction rate and degree. Results show that increasing the laser power (system temperature) and reducing the Mg-rich layer thickness, Si particle size, and Al content accelerate the reaction rate and degree.
Science Press


Aluminum (Al) matrix composites have wide application prospects in the field of light rail transportation, due to their high specific strength, stiffness, modulus, wear resistance and thermal resistanc
Compared with conventional fabrication methods for in-situ composites, such as mechanical alloyin
In this study, a dynamic model of in-situ Mg2Si/Al aluminum matrix composites was established, and the effects of laser power (P) on the microstructure of in-situ Mg2Si/Al composite were investigated.
Pure Al, Mg, and Si powders were used with the average size of ~74 μm. The Mg-to-Si powder ratio corresponded to the stoichiometric ratio of Mg2Si. The powders were mixed in a high-energy ball mill filled with Ar gas for 2 h.

Fig.1 Morphology of powder mixtures after milling
The laser deposited samples with dimension of 70 mm×10 mm×10 mm were obtained. Metallographic samples were obtained by cutting transversely from the middle of the deposited samples and then polished and etched with 0.5vol% HF solution for scanning electron microscopy (SEM) observation with the Tesan VEGAII Lmh system. The phase constitution was determined by SEM with energy dispersive spectroscopy (EDS). The phases were analyzed by X-ray diffraction (XRD) at 40 kV and 40 mA using Cu-Kα radiation (Max-2000X) at 1873 K for 1 h in argon atmosphere.

Fig.2 Microstructure of 10wt%Mg2Si/Al composite with laser power of 800 W (a), 1000 W (b), and 1200 W (c)

Fig.3 Microstructure of 15wt%Mg2Si/Al with laser power of 900 W
In the present research, the setting that Mg-to-Si powder ratio corresponds to the stoichiometric ratio of Mg2Si should result in no Mg and Si in the microstructure. However, the results in Fig.4 show the presence of Si without Mg in the microstructure. A dynamic model of in-situ synthesis of Mg2Si is established for theoretical calculations, as shown in

Fig.5 Dynamic model of in-situ synthesis of Mg2Si

Table 1 EDS results of points A and B in Fig.3 (at%)
The following hypotheses are put forward to simplify the calculation: (1) the Si powders are globular and distributed uniformly in the melt; (2) there is a Mg-Al middle layer with the thickness of δ around the Si powder; (3) the diffusion coefficient D=D0exp(-Q/RT) of Mg atoms-to-Si atoms is only related to the temperature; (4) the interfacial reaction rate of Mg on the surface of Si particles is much greater than the diffusion rate of Mg in the middle layer because the diffusion coefficient of Mg in liquid Al is only about ~1
Therefore, the reaction equation in the melt is 2Mg+Si+xAl=Mg2Si+xAl where x is the mole number of Al.
Accord to the hypothesis (3), the diffusion flux J based on Fick’s second law can be expressed as follows:
(1) |
where A is the spherical surface area of any middle layer (
The diffusion flux J based on Fick’s first law can be defined as follows:
(2) |
where D is the diffusion coefficient of Mg in the Al-Mg melt (
According to
=0 | (3) |
Based on the hypothesis (4), the concentration of Mg on the surface of Si particles is 0. Thus, the boundary condition of
(4) |
(5) |
where rSi is the radius of the Si particle at a certain time (m), δ is the distance showed in
Therefore, the Mg concentration at the position with radius r can be obtained using Eq.(3)~
(6) |
where dSi is the diameter of the Si particle at a certain time (m).
Thus, the diffusion rate of Mg (VMg) across the middle layer to the surface of Si particle is as follows:
(7) |
where SSi=4π
The depletion rate of Si particles during in-situ synthesis can be defined as a volume reduction of Si particles:
(8) |
where t is the time.
According to
(9) |
The diameter of Si particles is d0 when t=0, and the melt consists of 1 mol Si+2 mol Mg+x mol Al. When t>0, the mole number of Al, Mg, Si and Mg2Si is x, , , and , respectively; the molar volume of Al, Mg, Si, and Mg2Si is , , , and , respectively. ρSi, ρAl, ρMg, and are the density of Si, Al, Mg and Mg2Si (kg/
Therefore, the molar concentration of Mg in the melt at a certain time t is as follows:
(10) |
The forming rate of Mg2Si (or the depletion rate of Si particles) can be obtained by incorporating
(11) |
It can be seen from
The reaction degree η of the Si particle is defined as follows:
(12) |
with
The diffusion coefficient of Mg depends on the temperature. The temperature, which is related to the laser powder and scanning velocity, is as follows:
(13) |
where Pe is the laser power (W) and v is the scanning velocity (mm/s). When the scanning velocity is 200 mm/min, the corresponding temperature is 980, 1000, 1019, 1038, and 1057 ℃ for the laser power of 800, 900, 1000, 1100, and 1200 W, respectively.


The effects of laser power on the forming rate of Mg2Si and the reaction degree of Si particles are shown in Fig.6. The forming rate of Mg2Si and the reaction degree of Si particles increase with increasing the laser power.

The effects of middle layer thickness δ on the forming rate of Mg2Si and the reaction degree of Si particles are shown in Fig.7. The forming rate and the reaction degree decrease as the thickness of the middle layer increases. This is attributable to the increase in the diffusion distance from Mg atoms to the Si particles.

The effect of the diameter of Si particles d0 on the forming rate of Mg2Si and the reaction degree of Si particles is shown in Fig.8. The forming rate and reaction degree decrease as the diameter of the Si particles increases. The smaller the diameter of the Si particles, the quicker the formation of Mg2Si. However, the flow of the powder is disrupted, resulting in a discontinuous powder feed. In this study, the optimum diameter of Si particles is 74 μm.
Fig.9 shows the effect of Al content on the forming rate of Mg2Si and the reaction degree of Si particles. The forming rate and the reaction degree decrease with increasing the Al content. This is attributed to the low Mg content. Thus, the interaction between Mg and Si is less likely.
Fig.6b, 7b, 8b, and 9b all indicate that Si particles have long reaction time. However, the solidification rate is very fast (1
1) The microstructure of in-situ Mg2Si/Al composites fabricated by laser deposition consists of α-Al phase, Si phase, and Mg2Si particles.
2) The forming rate of Mg2Si and the reaction degree of Si particles are related to the laser power (system temperature), Mg-rich layer thickness, Si particle size, and Al content.
3) Increasing the laser power (system temperature) accelerates the reaction rate and reaction degree. With increasing the Mg-rich layer thickness, Si particle size, and Al content, the reaction rate and the reaction degree reduce.
References
Ghasali E, Pakseresht A H, Alizadeh M et al. Journal of Alloys and Compounds[J], 2016, 688: 527 [Baidu Scholar]
Rahimian M, Parvin N, Ehsani N. Materials Science and Engineering A[J], 2010, 527(4-5): 1031 [Baidu Scholar]
Omyma E K, Fathy A. Materials and Design[J], 2014, 54: 348 [Baidu Scholar]
Wang Z D, Wang X W, Wang Q S et al. Nanotechnology[J], 2009, 20(7): 75 605 [Baidu Scholar]
Selvam J D R, Dinaharan I. Engineering Science and Technology[J], 2017, 20(1): 187 [Baidu Scholar]
Gao Q, Wu S S, Lü S L et al. Materials & Design[J], 2016, 94: 79 [Baidu Scholar]
Tochaee E B, Hosseini H R M, Reihan S M S. Materials Science and Engineering A[J], 2016, 658: 246 [Baidu Scholar]
Khanra A K, Pathak L C, Mishra S K et al. Materials Letters[J], 2004, 58(5): 733 [Baidu Scholar]
Li S F, Kondoh K, Imai H et al. Materials & Design[J], 2016, 95: 127 [Baidu Scholar]
Zhu H G, Wang H Z, Ge L Q. Wear[J], 2008, 264(11-12): 967 [Baidu Scholar]
Zhao L Z, Zhao M J, Li D Y et al. Applied Surface Science[J], 2012, 258(8): 3368 [Baidu Scholar]
Attar H, Bönisch M, Calin M et al. Acta Materialia[J], 2014, 77(1): 13 [Baidu Scholar]
Shamsaei N, Yadollahi A, Bian L K et al. Additive Manufacturing[J], 2015, 8: 12 [Baidu Scholar]
Yu X T, Wang H M. Acta Materiae Compositae Sinica[J], 2008, 25(4): 113 (in Chinese) [Baidu Scholar]
Zhang J, Fan Z, Wang Y Q et al. Materials Science and Technology[J], 2001, 17(5): 494 [Baidu Scholar]
Du Y, Chang Y A, Huang B Y et al. Materials Science and Engineering A[J], 2003, 363: 140 [Baidu Scholar]
Zhang H, Pan Y, He Y Z et al. JOM[J], 2014, 66(10): 2057 [Baidu Scholar]