ABSTRACT
The easily vitrified materials such as natural silicates, some chalcogenides and polymers are characterized by high melt viscosities T], of approximately 102 -103 Pas at temperatures higher than their equilibrium melting temperature, T m· Glass formation is promoted by the additional rapid increase of viscosity on undercoating. By the conventional glasses low or moderate cooling rates of 0.01 to 5 K s·1 are sufficient to avoid crystallization because of the very slow rate of nucleation and growth of crystals in their undercooled melts. Due to the fact that the viscosity of molten metallic alloys is very low, e.g., IQ-3 Pas, extremely high cooling rates of 106 K.s-1 or more are needed
to avoid their crystallization (1-4). After the splat quenching was developed in 1960 by Klement et al. (5,6), cooling rates up to I 07 Ks-1, i.e., high above the cooling rates, normally applied in metallurgical practice, have been achieved. This has resulted in expanding the number and types of glass forming materials to include a variety of compositions of metal-metaloid and metal-metal amorphous alloys (7). Amorphous or glassy metallic alloys represent a new class of materials, which show unusual combination of disordered amorphous structure, typical for conventional silicate glasses, with physical properties, typical for metals and alloys with crystalline structure. The subsequent development of new methods for quenching from the melt such as Chill-Block Melt Spinning (CBMS) (8,9,10) and Planar Flow Casting (PFC) (11, 12,13 ), capable of large-scale production of glassy metallic ribbons of practically unlimited length by widths more than 300 mm and thickness of 0.02 to 0.05 mm, has shifted these materials in the foreground of both material scientists and condensed matter physicists. This interest has primarily arisen due to the fact that many properties of the metallic glasses differ considerably from those of their crystalline counterparts and make them very promising for industrial applications (14-24). The possibility to produce new types of metallic glasses including also various contents of the components, has given rise to the fundamental research of atomic arrangements and rearrangements in condensed matter with respect to the glass-forming ability and properties of the metallic alloys. In studying the nature of glass formation, the kinetic approach (25) is often used, which assumes that all materials, including metals and alloys, can be vitrified, provided their melts are cooled down with sufficiently high cooling rate, which has to be at least equal or somewhat greater than the critical cooling rate Tcr' specific for the material under consideration. Uhlman (26, 27) has proposed a method of estimating the critical cooling rate of glass forming systems using accepted theories of nucleation, crystal growth and volume transformation kinetics (28). He has shown that the time, t, for a small fraction, s (usually accepted as I0-6), crystallized at temperature T, equals
t = {9.3 11 (T)/kT} [(a0 <;/f3N) { exp (1.024 IT/ D.T/) }I
(1)
where D. T, = (T m -T ), T, = TIT m' a0 is the average atomic diameter Nv is the number of atoms per unit volume, f is the fraction of sites at the melt/crystalline interface where atoms are preferentially added or removed and D.Hm is the molar enthalpy of fusion. The critical cooling rate is taken as
(2)
where Tn and t0 are the temperature and the time at the nose of the time-temperature-transformation diagram showing the temperature dependence of the time, needed for reaching I0-6 volume fraction of crystallized melt. Such a TIT-diagram is schematically shown in Figure 1. Computation of the TTT-curve requires the temperature dependence of the melt viscosity 11 (T) within the temperature range from melting temperature Tm to the glass transition temperature Tg to be known.
