Investigation of Reverse Filament Formation in ITO/HfO2-based RRAM

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceeding


To overcome the large discrepancy in speed between computational devices and that of contemporary large capacity non-volatile memory (NVM) technologies, resistive random access memory (RRAM) technologies are seen as promising candidates, offering speed/energy improvements in several orders of magnitude while being 3D integration compatible [1]. Indium-Tin-Oxide (ITO) has several unique properties for RRAM operation, perhaps most prominently the self-compliance and an ultra-low switching voltage (±200 mV) [2]. We report on considerations for ITO electrical bottom electrode (BE) RRAM where we vary the ALD oxide deposition parameters in order to improve the reverse filament formation (RFF) occurring at large reset voltages. RFF is when the conducting filament is reformed. One of the key parameters of RRAM is the endurance, how many times it can switch before failure. The RFF is one of the limitations in the number of switches until a device reaches failure and it is thus of high importance to ensure a sufficient margin between the highest applied reset voltage and the RFF voltage not to compromise the endurance. We optimized the oxide to improve the RFF properties.
Original languageEnglish
Title of host publication2019 Device Research Conference (DRC)
PublisherIEEE - Institute of Electrical and Electronics Engineers Inc.
Number of pages2
ISBN (Electronic)978-1-7281-2112-3
ISBN (Print)978-1-7281-2111-6
Publication statusPublished - 2019 Jun
Event77th Device Research Conference (DRC) - University of Michigan, Ann Arbor, United States
Duration: 2019 Jun 232019 Jun 26


Conference77th Device Research Conference (DRC)
Country/TerritoryUnited States
CityAnn Arbor

Subject classification (UKÄ)

  • Nano Technology
  • Computer Systems


Dive into the research topics of 'Investigation of Reverse Filament Formation in ITO/HfO2-based RRAM'. Together they form a unique fingerprint.

Cite this