Thermodynamics and Kinetics of The Crystallization of Phase Change Materials from In Situ Microscopy and Nanocalorimetry

The thermodynamics and kinetics of crystallization are key to understanding the stability of glass-forming materials, including poor glass formers such as phase change materials (PCMs). PCMs are semi-conducting alloys with distinct optical and electrical properties in the amorphous and crystalline phases that make them useful for memory applications. In memory devices, amorphous bits are crystallized in nanoseconds by either laser or Joule heating, but the amorphous phase must also be stable against crystallization for long-term data retention. Crystal growth rates relevant to memory devices span orders of magnitude and fundamental questions regarding PCM crystallization mechanisms remain open, partly due to the difficulty in measuring crystallization kinetics in certain temperature regimes.<br/><br/>The crystal growth rate, <i>u</i>, has been directly<b> </b>measured from the glass transition, T_g, to the melting temperature for good glass formers owing to their low <i>u</i>. In contrast, <i>u</i> in PCMs can exceed 10 m/s. Thus, it is challenging to measure <i>u</i> directly with microscopic methods, because the small grain sizes demand high spatial resolution and the grains impinge rapidly. This has led to the use of indirect<i> </i>methods to study PCM crystallization such as differential scanning calorimetry (DSC). In this work nanocalorimetry with high-frame-rate transmission electron microscopy (TEM) imaging was used to investigate the thermodynamics and kinetics of the crystallization of Ag₃In₄Sb₇₆Te₁₇ and GeTe above T_g. The PCMs were deposited on individually-calibrated nanocalorimeters designed to be operated in a TEM. Direct electron detectors capture the crystal growth in the milliseconds before impingement enabling direct measurement of <i>u</i>. The enthalpies of crystallization are calculated from nanocalorimetry measurements. Classical models for nucleation and growth are fit against these data and models that predict the growth rate solely from DSC data are compared to the direct growth rate measurements.