Crystallization supercooled silicon

Figure 23. The ph ase diagram of supercooled silicon in pressure temperature (P, T) plane obtained from simulations using the SW potential. The phase diagram shows the location of (i) the liquid-crystal phase boundary [115]—thick solid line, (ii) the liquid-gas phase boundary and critical point—line and a star, (iii) the liquid-liquid phase boundaiy and critical point—filled diamond and a thick circle, (iv) the liquid splnodal—filled circle (v) the tensile limit—open circle (vi) the density maximum (TMD) and minimum (TMinD) lines— filled and open squares, and (vii) the compressibility maximum (TMC) and minimum (TMinC) line—filled and open circle. Lines joining TMD and TMinD (dot-dashed), TMC and TMinC (solid), Spinodal (black dotted line) are guides to the eye.

Dynamics One of the biggest challenges in the study of supercooled silicon is that, with deep undercooling the system, not only becomes prone to rapid crystallization but also the relaxation times increase very rapidly. Vasisht et al. studied the relaxation times at different parts of the phase diagram and found that ap proaching the liquid liquid transition line (below the critical temperature) and the compressibility maxima line (above the critical temperature), the relaxation time increases in a non Arrhenius manner. In Fig. 25a, the relaxation time as a function of temperature at two different pressme values (above — P = 0 GPa and below — P = —1.88 GPa critical point) is shown. Figure 25b shows relaxation times as a function of pressure for three different temperatures values. 

Pollitt and Brown (P2) were unable to prepare an analogue of the orthorhombic phase with as the sole substituent, but obtained evidence that could stabilize it in clinker, probably because other substituents were also present. Maki (M12) also failed to prepare K forms of the orthorhombic or monoclinic phases under equilibrium conditions, but by moderately rapid cooling of melts he obtained orthorhombic crystals having cell parameters close to those of the corresponding sodium-containing phase. He considered that the presence of silicon in the clinker liquid would favour supercooling and thereby also non-equilibrium formation of the orthorhombic or monoclinic phase. 

---