In this letter, we explore the influence of the CuxTe1-x layer composition (0.2 < x < 0.8) on the resistive switching of CuxTe1−x/Al2O3/Si cells. While x > 0.7 leads to large reset power, similar to pure-Cu electrodes, x < 0.3 results in volatile forming properties. The intermediate range 0.5 < x < 0.7 shows optimum memory properties, featuring improved control of filament programming using <5 μA as well as state stability at 85 °C. The composition-dependent programming control and filament stability are closely associated with the phases in the CuxTe1−x layer and are explained as related to the chemical affinity between Cu and Te.
We demonstrate for the first time record low Leakage-EOT (3.5x10 -7 A/cm 2 at 1V, EOT=0.49 nm) MIM capacitors fabricated using a low temperature (250 o C) ALD SrTiO 3 (STO) deposition process on ALD TiN bottom electrode. While most previous work on STO used deposition techniques not compatible with high aspect ratio DRAM applications, recent work on ALD STO showed promise on noble-like metal electrodes (Ru, Pt) [1,2]. In this work, a low temperature ALD process with alternative precursor set and carefully optimized deposition and processing conditions enables the use of low-cost, manufacturablefriendly TiN electrode MIMcaps for future DRAM nodes. Composition (Sr-rich) and process optimization allowed minimization of interfacial EOT penalties and leakage reduction by decreasing the density of leakier STO grains.
IntroductionMIMcaps with EOTs 0.5 nm and low leakage are required for future DRAM nodes. Alternatives beyond ZrO 2 /Al 2 O 3 /ZrO 2 are needed. STO is a promising candidate, but much of previous work focused on nonconformal deposition techniques. As exception, ALD STO using Sr(thd) 2 precursor for Sr has been reported [1,2] with promising results on noble like metal electrodes such as Ru and Pt. However, these processes required either high deposition temperature and/or post-deposition anneals in oxidizing ambients [1,2], making STO incompatible with TiN. By using an alternative ALD precursor system and optimizing carefully deposition variables, composition and post-deposition processing, we demonstrate for the first time excellent results for STO/TiN.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.