Resistive Electrical Switching of Cu and Ag based Metal-Organic Charge Transfer Complexes
- PDF / 1,569,832 Bytes
- 12 Pages / 612 x 792 pts (letter) Page_size
- 17 Downloads / 172 Views
1071-F06-04
Resistive Electrical Switching of Cu+ and Ag+ based Metal-Organic Charge Transfer Complexes Robert Mueller1, Joris Billen1,2, Aaron Katzenmeyer1, Ludovic Goux3, Dirk J. Wouters3, Jan Genoe1, and Paul Heremans2,4 1 PT\SOLO\PME, IMEC v.z.w., Kapeldreef 75, Leuven, 3001, Belgium 2 ESAT, Katholieke Universiteit Leuven, Kasteelpark Arenberg 10, Leuven, 3001, Belgium 3 PT\CMOSRD\MEMORY, IMEC v.z.w., Kapeldreef 75, Leuven, 3001, Belgium 4 PT\SOLO, IMEC v.z.w., Kapeldreef 75, Leuven, 3001, Belgium
ABSTRACT Memory cells based on Cu+ and Ag+ metal-organic charge-transfer complexes, as for example CuTCNQ (where TCNQ denotes 7,7',8,8'-tetracyanoquinodimethane), are well known for their bistable resistive electrical switching since 1979. The switching mechanism however remained unclear for very long time. In this contribution we describe the different views (bulk vs. interfacial switching), give evidence for interfacial switching in the case of CuTCNQ, and present a model allowing explaining the bipolar resistive electrical switching by an interfacial effect, even for experiments considered until now as proof for bulk switching. The proposed switching mechanism is based on bridging of an ion-permeable layer (or gap) by conductive Cu channels, which are formed and dissolved by an electrochemical reaction implying monovalent Cu+ cations, originating from a solid ionic conductor (as for example CuTCNQ). The model was furthermore generalized to other memory systems consisting of a permeable layer and a solid ionic conductor, including also inorganic solid ionic conductors as for example Ag2S. INTRODUCTION Downscaling of traditional charge-storage based non-volatile memory technology (Flash) becomes more and more challenging due to physical limitations and increasing processing complexity. According to the International Technology Roadmap for Semiconductors [1] NOR and NAND Flash Memories will encounter scaling issues around the years 2009 and 2013, respectively (Fig. 1). Resistive switching memories are not expected to suffer from these scaling issues and are therefore currently investigated as potential candidates for future memory applications. Various kinds of resistive switching memories have been proposed so far, as for example the oxide resistive random access memory [2], the phase change memory [3], the programmable metallization cell [4,5] and the chalcogenide based "Nanobridge®" [6]. The working principle of these different memory types have been compared in a recent paper [7].
100 90 9 80 8 70 7
NOR
Feature size F (nm)
60 6 50 5
Scaling issue: tunnel oxide thickness
NAND
40 4
30 3
Scaling issue: interpoly dielectric thickness
2 20
10
2006
2008
2010
2012
2014
2016
2018
2020
Year
Figure 1. International Technology Roadmap for Semiconductors (redrawn from the 2006 update [1]). Several organic materials have been investigated for resistive switching memory cells as well (see [8-10] for review). CuTCNQ (where TCNQ denotes 7,7',8,8'-tetracyanoquinodimethane) is one of the most known organic ma
Data Loading...
