Experimental And Simulation Study On Grain Refinement Of Mg-9Al Alloy By Nano-SiC Particles

The addition of nanoparticles to magnesium alloys can significantly refine the grain size of alloy while improving the strength and plasticity properties.It is important to the fast development of aviation,aerospace,automotive,machinery and electronics industries.In this paper,nano-silicon carbide particles(n-Si C_p)reinforced Mg-9Al alloys were prepared by semi-solid stirring and high-energy ultrasonic dispersion methods.Through the analysis of grain structure,the grain size is refined gradually with the increase of the amount of nanoparticles.Compared to the added amount of 0vol.%,0.5 vol.%and 1.2 vol.%,with the amount 1.0 vol.%,the grain size(about 52?m)is the most refined,which is about 64%smaller than that of the matrix alloy(144?m).The microstructure of 1.0 vol.%n-Si C_p/Mg-9Al composite material is analyzed.Most of the nanoparticles distribute near the grain boundary and locate inside the?-Mg₁₇Al₁₂ phase.According to the interaction model between nanoparticles and the solid-liquid interface that assumes particles are in the equilibrium position with the solid-liquid interface,the growth velocity of the dendrite tip is much smaller than the calculated critical velocity.Thus,nanoparticles will be pushed by the solid-liquid interface to the vicinity of grain boundary,which explains why nanoparticles disperse around grain boundary.The nanoparticles near the grain boundary form a multilayer coating,inhibiting solute diffusion and then slowing down grain growth.A two-dimensional sharp-interface model was developed to simulate the dendritic growth of?-Mg and microstructure formation.A cellular automation(CA)technique was used to describe dendritic development.A mesh anisotropy reduction curvature algorithm(MAR)was proposed that is capable of overcoming the influence of mesh anisotropy on the simulated results.Compared with the variation of unit vector normal(VUVN),the proportion of unreasonable cells in the curvature calculation with the MAR algorithm is less than the results with the VUVN algorithm.The equivalent solute diffusion coefficient(D_l')was determined by the experiments.The equivalent solute diffusion coefficient is substituted into the sharp-interface model for microstructure simulation.The simulations show that the grain size decreases gradually with the decrease of D_l',which is in line with the experimental grain-refinement tendency.Compared with Mg-9Al alloy without nanoparticles,1.0vol.%n-Si C_p/Mg-9Al composite has a lower solute diffusion coefficient.With the addition of nanoparticles,the degree of solute diffusion reduces,effective undercooling increases,dendrite growth velocity decreases,and the latent heat release slows.Thus,many nuclei are activated and the grain refinement is achieved.In my thesis,the experimental characterizations and simulations were used to reveal the mechanism of grain refinement of n-Si C_p/Mg-9Al composites.The prevention of solute diffusion around growing dendrites with the formation of multilayer-nanoparticle-coating plays a key role.The obtained results will provide a theoretical reference for grain refinement of magnesium alloy.