Kinetics of Crystal Nucleation and Growth in Thin Films of Amorphous Te Alloys measured by Atomic Force Microscopy
- PDF / 646,908 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 46 Downloads / 181 Views
HH3.8.1
Kinetics of Crystal Nucleation and Growth in Thin Films of Amorphous Te Alloys measured by Atomic Force Microscopy J. Kalb1,2, F. Spaepen1 and M. Wuttig2, a) 1 Division of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138, U.S.A. 2 I. Physikalisches Institut der RWTH Aachen, 52056 Aachen, Germany ABSTRACT Both the crystal nucleation rate and the crystal growth velocity of sputtered amorphous Ag0.055In0.065Sb0.59Te0.29 and Ge4Sb1Te5 thin films used for optical data storage were determined as a function of temperature. Crystals were directly observed using ex-situ atomic force microscopy, and their change in size after each anneal was measured. Between 140°C and 185°C, these materials exhibited similar crystal growth characteristics, but differed in their crystal nucleation characteristics. These observations provide an explanation for the different recrystallization mechanisms observed upon laser-induced crystallization of amorphous marks. INTRODUCTION In order to increase the data transfer rate of optical data storage media, the time-limiting factor, which is the re-crystallization of an amorphous bit, has to be accelerated. Therefore, it is essential to understand the mechanisms of re-crystallization of phase change materials. Two mechanisms of re-crystallization depending on the composition of the alloy have been observed. For instance, AgIn-doped Sb2Te re-crystallizes by the growth of the crystalline phase from the rim of the amorphous bit towards its center [1]. In contrast, Ge4Sb1Te5 (an alloy close to the GeTe-Sb2Te3 pseudobinary line [2]) re-crystallizes by nucleation and subsequent growth of crystals inside the amorphous bit [3]. The atomistic basis for this difference is still not clearly understood. Even though several research groups have assumed that these alloys differ in their crystal nucleation rate and crystal growth velocity, systematic measurements of these two quantities as a function of temperature have to the best of our knowledge so far not been performed. Therefore, in many modeling studies of re-crystallization the fitting parameters have no direct experimental justification [4-6]. Additional difficulties occur because the glass transition temperature Tg is usually accompanied by an abrupt change in the temperature derivative of several physical parameters [7-9]. Therefore, crystallization parameters (in particular activation energies) determined experimentally from the easily accessible amorphous phase (below Tg) cannot be extrapolated into the undercooled liquid (above Tg). Only experimental data collected above Tg are useful for the simulation and understanding of the recrystallization process of phase change media, which also occurs above Tg. Therefore, there is a strong demand of systematic measurements of the crystallization parameters in the regime of the undercooled liquid (rather than the amorphous phase). a)
electronic mails: [email protected], [email protected], [email protected]
HH3.8.2
EXPERIMENTAL PROCEDURE Phase change
Data Loading...
