Insertion of Ag Layer in TiN/SiN_x/TiN RRAM and Its Effect on Filament Formation Modeled by Monte Carlo Simulation

Effect of Ag source layer thickness on the switching mechanism of TiN/Ag/SiN x /TiN conductive bridging random access memory observed at sub-µA current

“…Furthermore, figure 5 provides a clue to explain the difference in pristine conductance between CBRAMs with Ag source layers deposited at 5 nm, 10 nm, and 40 nm thicknesses. According to a previous study, Ag atoms dissolve and seep into the dielectric as deposited onto it [34]. The EDS profile in figure 1(c) also shows that some of Ag atoms are detected in the SiN x layer.…”

Electric-Field-Induced Metal Filament Formation in Cobalt-Based CBRAM Observed by TEM

“…Figure f compares the pristine conductances measured for the Co/SiO x (5 nm)/TiN and Ag/SiO x (5 nm)/TiN material stacks. A previous study reported that Ag atoms in Ag CBRAM dissolve into a neighboring dielectric layer after approximately 3–5 nm of deposition, even without electrical bias . Thus, if the Co pillar in Figure a is formed spontaneously during metal deposition, the Co/SiO x (5 nm)/TiN CBRAM cell will exhibit a leaky pristine conductance similar to that of the Ag/SiO x (5 nm)/TiN control CBRAM cell.…”

“…A previous study reported that Ag atoms in Ag CBRAM dissolve into a neighboring dielectric layer after approximately 3−5 nm of deposition, even without electrical bias. 27 Thus, if the Co pillar in Figure 3a is formed spontaneously during metal deposition, the Co/SiO x (5 nm)/TiN CBRAM cell will exhibit a leaky pristine conductance similar to that of the Ag/SiO x (5 nm)/TiN control CBRAM cell. However, Figure 3f shows the perfect insulation of the corresponding Co sample, indicating that the 5 nm long Co pillar in Figure 3a is formed by electrical bias rather than a natural process during or after device fabrication.…”

“…With the increasing demand for data parallelism to save the time and energy cost of accessing data, RRAM plays a significant role in in-memory computing research, including neuromorphic systems. − RRAM is mainly classified into valence change memory (VCM) or electrochemical metallization memory (ECM) according to the type of ions used during resistive switching (RS). , ECM, also known as conductive-bridging random access memory (CBRAM), alters its resistive state by forming a cation-type conductive path composed of metal atoms from the electrochemically active metal electrode. CBRAM is superior to other types of RRAM in terms of on/off resistance ratio, switching speed, and programming voltage. − Conventional CBRAM cells contain Ag, Cu, and Ni active metal electrodes and have been widely investigated since the introduction of ECM. − …”

Resistive random access memory: introduction to device mechanism, materials and application to neuromorphic computing

“…But, the intermediate functional resistive switching layer (RS layer) provides the switching region. Therefore, the performance of the RRAM device decisively relies on the thin film materials' mechanism [130][131][132][133][134][135]. However, the proper understanding of these materials and mechanisms governing their operations are currently vital issues.…”