Memory-State Dependence of Random Telegraph Noise of $ \hbox{Ta}_{2}\hbox{O}_{5}/\hbox{TiO}_{2}$ Stack ReRAM

Abstract

Memory-state [low-resistance state (R L ) and highresistance state (R H )] dependence of random telegraph noise (RTN) of Ta 2 O 5 /TiO 2 resistive random access memory is investigated. The conduction mechanism of both memory states and a limit of resistance controllability are also investigated to clarify the difference in the RTN mechanism of both states. The boundary between the R L and R H states was found at 5-20 kΩ, and the conduction mechanism much depended on the memory state. The noise also depended on the memory state. The noise amplitude in the R H state was larger than that in the R L state. In the R H state, a tunnel barrier was generated to cut off a conduction path (filament), and traps inside the tunnel barrier were supposed to increase the noise amplitude. Moreover, the composition of the following degraded the noise distribution in the R H state: 1) capture and emission of charges with traps and 2) instability of these traps against the bias temperature stress.Index Terms-Filament, noise, random telegraph noise (RTN), resistive random access memory (ReRAM), resistive switch, RRAM, Ta 2 O 5 , TiO 2 .

(Invited) Random Telegraph Noise: From a Device Physicist's Dream to a Designer's Nightmare

“…In this case, the trapping is associated with p-n junction leakage current and with traps in the silicon depletion region (37,38). Recent results have demonstrated that the RTN issue will become even more pronounced for future Resistive RAM (ReRAM) (39)(40)(41)(42) or Phase-Change Memories (43)(44)(45). , 39 (1) 3-15 (2011) It is a we follows a that this observed the 1/f no to a Lore distribute demonstr that when more Lo indicates determin dominate Figure 12 (pulse lik signature noise spe explained (13) , 39 (1) 3-15 (2011) …”

Investigation of random telegraph noise amplitudes in hafnium oxide resistive memory devices

“…Random telegraph noise (RTN) in read current arising from charge trapping/detrapping in an RRAM dielectric, however, has posed to be a serious reliability issue. Previous studies have shown that RTN amplitudes in RRAM scatter over a wide range, especially in high-resistance state (HRS) [4,5]. However, rare efforts have been made to study the cause of large RTN amplitudes.…”

Ultra-large resistance ratio of silver programmable metallization cell with stacked silicon oxide films

“…In some RRAMs, especially for the devices with CF of oxygen vacancies, the resistance ratio can be as high as five to six orders of magnitude with stable retention and sufficient endurance properties [20]. Such a huge resistance ratio can be ascribed to the large resistance at high resistance state (>10 8 U) by completely cutting off the filament with a tunnel barrier of a quite small MIM devices (<10 mm 2 ) [20,21], providing the possibility of realizing a high-density nonvolatile memory by means of the multi-bit storage [11]. Nevertheless, for the conventional PMCs, the resistance ratio is presented to be lower than 10 4 [14e16], which limits the use for multi-level cell operation.…”