Ultrahigh Storage Densities via the Scaling of Patterned Probe Phase-Change Memories
IEEE Transactions on Nanotechnology
Institute of Electrical and Electronics Engineers (IEEE)
Copyright (c) 2017 IEEE. Personal use of this material is permitted. However, permission to use this material for any other purposes must be obtained from the IEEE by sending a request to firstname.lastname@example.org.
The scaling potential of patterned probe phase-change memory (PP-PCM) cells is investigated, down to single-nanometer dimensions, using physically realistic simulations that combine electro-thermal modelling with a Gillespie Cellular Automata (GCA) phase-change model. For this study, a trilayer TiN/Ge 2 Sb 2 Te 5 /TiN cell structure (isolated by a SiO 2 insulator) was preferred, due to its good performance and practicability, over previously investigated probe-based structures such as those that used diamond-like carbon capping layers or immersion in an inert liquid to protect the phase-change layer (while still allowing for electrical contact). We found that PP-PCM cells with dimensions as small as 5 nm could be successfully amorphized and re-crystallized (RESET and SET) using moderate voltage pulses. The resistance window between the RESET/SET states decreased with a reduction in cell dimensions, but it was still more than order of magnitude even for the smallest cells, predicting that PP-PCM cells are indeed scalable and operable in the sub-10 nm region. Most importantly, it was found that the storage density could be increased by cell size scaling with storage densities as high as 10 Tb/in 2 being achieved, which is significantly higher than the storage densities previously reported in phase-change probe storage, and other probe-based technologies such as thermomechanical, magnetic and ferroelectric probe storage.
Hasan Hayat would like to thank the College of Engineering, Mathematics and Physical Sciences at the University of Exeter for PhD scholarship funding while carrying out this work. C. David Wright would like to thank the EPSRC for funding via grant EP/M015130/1.
This is the author accepted manuscript. The final version is available from IEEE via the DOI in this record.
Published online 31 March 2017