Structural change in silicon from undercooled liquid state to crystalline state during crystallization

Nitrogen reaction with silicon: Investigation of Si undercooling and Si3N4 growth

“…The silicon droplets and the experimental vessel are assumed to be clean enough when the undercooling is higher than 250 K. This value is reasonable compared to the homogeneous nucleation undercooling supposed to occur at 25% of the melting point ($ 420 K, as measured by [14]). Most of the undercooling measured for pure silicon in our equipment is higher than 300 K, comparable to what is currently reported in the literature [14][15][16][17].…”

Does supercooled liquid Si have a density maximum?

“…In our analytical model of the supercooled l-Si structure, we used the model structure combined with the b-tin structure and diamond structure based on tetrahedral coordination to fit the experimental S(Q). 5 Our model structure for analysis is almost the same structure obtained by FPMD simulations based on tetrahedral coordination. However, it is well known that the density of l-Si is larger than that of crystalline Si, and the density increases with decreasing temperature, as shown in Fig.…”

“…Much work has been aimed towards exploring the transition between HDL (low tetrahedrality) and LDL (high tetrahedrality) in supercooled l-Si as well as in other liquids having tetrahedral coordination. 1,2 The density and structure of supercooled l-Si have been measured using various levitation techniques, including electromagnetic levitation (EML), [3][4][5] electrostatic levitation (ESL), [6][7][8][9] and conical-nozzle gas levitation (CNL) techniques. 10,11 However, experimental results have not shown evidence of the HDL-LDL transition in supercooled l-Si.…”

“…We also performed first-principles molecular dynamics (FPMD) simulations of supercooled l-Si to investigate the precise structure of supercooled l-Si by comparison with previous high-energy X-ray diffraction experiments combined with the EML technique. 4,5 From the measurement of density and structure, and the FPMD simulations, we discuss nucleation in supercooled l-Si and also discuss the existence of a density maximum of l-Si which is well known in water at 4 1C.…”