Multi-level programming in Phase-Change Memories (PCMs) requires adequate understanding of the phenomena affecting the stability of the programmed levels. Although crystallization in GST (Ge2Sb2Te5) has been extensive...
详细信息
ISBN:
(纸本)9781479942428
Multi-level programming in Phase-Change Memories (PCMs) requires adequate understanding of the phenomena affecting the stability of the programmed levels. Although crystallization in GST (Ge2Sb2Te5) has been extensively studied, further analysis is needed to understand the phase distribution in intermediate states between the amorphous and the crystalline GST. In this paper, we carry out a drift-driven analysis on a PCM array to investigate phase distribution of intermediate resistance states obtained with partial-set programming. The analysis shows that the phase distribution after a partial-set pulse can be more conveniently described by means of a parallel-like model having a crystalline path inside an amorphous cap.
A key issue in nonvolatile storage is long-term data retention. This aspect is even more important in innovative storage technologies such as phase-change memory (PCM), which promises better performance and easier sca...
详细信息
A key issue in nonvolatile storage is long-term data retention. This aspect is even more important in innovative storage technologies such as phase-change memory (PCM), which promises better performance and easier scalability with respect to traditional Flash memory and potential for multilevel storage. In this respect, we experimentally investigated the stability of intermediate states obtained by means of partial-set programming. To this end, we analyzed the effects of the width and the amplitude of the programming pulses on the degradation of intermediate programmed resistance levels over time in PCM cells. Our study was carried out by considering the average behavior of an array of PCM cells, showing that data-retention properties degrade as the programming thermal stress increases.
暂无评论