article·2012-11-09

Thermal-stability optimization of Al2O3/Cu–Te based conductive-bridging random access memory systems

L. Goux, Karl Opsomer, Alexis Franquet, Gouri Sankar Kar, N. Jossart, Olivier Richard, D. J. Wouters, Robert Müller, Christophe Detavernier, M. Jurczak, J. A. Kittl

Publisher page

Abstract

In this article we study the thermal stability of Al2O3Cu–Te bi-layers up to temperatures used in the back-end-of-line integration process flow of conductive-bridging random-access-memory (CBRAM) technology. We investigate the temperature dependence of the microstructure and morphology of the CuxTe1 − x layers for 0.2 < x < 0.8. For x > 0.7, phase separation is observed for as-deposited layers, resulting in rough morphology. Te-rich CuxTe1 − x layers (x < 0.5) show large segregation processes during post-deposition annealing. Cu–Te phase restructuration phenomena are also observed during annealing in the range 0.5 < x < 0.7, however affecting little the roughness of the layer. On the other hand, both the Cu and Te elements diffuse into the Al2O3 layer already at moderate temperatures. These in-diffusion processes are efficiently reduced by inserting thin Ti layer at the Al2O3Cu–Te interface. By means of secondary ion-mass spectroscopy analysis, we show that the thickness of the Ti layer allows obtaining either Cu-barrier (6 nm-thick Ti) or Cu-buffer (3 nm-thick Ti) properties. The optimized thermal stability achieved both by tuning the Cu–Te composition and by inserting 3 nm-thick Ti layer results in excellent and thermally stable CBRAM functionality after stack integration. Highlights ► We investigate phase restructuration processes during annealing of Cu–Te layers. ► We identify the Cu–Te composition for optimum thermal stability of the layer. ► We engineer a 3 nm-thick Ti buffer layer at the Al2O3Cu–Te interface. ► The integrated 90 nm-large resistive-switching cell shows excellent functionality.
Source: expertdata · Quality: candidate unverified · Source record ↗

Research context

OpenAlex topics: Advanced Memory and Neural Computing · Ferroelectric and Negative Capacitance Devices · Phase-change materials and chalcogenides

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References

Reference metadata was not supplied for this record.