DocumentCode
172632
Title
Impact of vertical RRAM device characteristics on 3D cross-point array design
Author
Pai-Yu Chen ; Shimeng Yu
Author_Institution
Arizona State Univ., Tempe, AZ, USA
fYear
2014
fDate
18-21 May 2014
Firstpage
1
Lastpage
4
Abstract
Resistive switching random access memory (RRAM) is one of the most promising candidates for the next-generation non-volatile memory. To enable large capacity RRAM array, the 3D architecture is needed. In this work, the analyses of different device characteristics, including the RRAM ON/OFF-state resistance (Ron/Roff), I-V nonlinearity ratio and the interconnect material, are performed to find the optimal design point of the 3D RRAM array. The results show that insufficient current drivability of the vertical transistor causes a significant amount of voltage drop on the transistor. The write and read margin conflicts with each other. A high Ron is preferred over a high nonlinearity to improve the write/read margin. A RRAM cell with Ron=500kΩ and nonlinearity=5x can satisfy the specified programming and read sense criterion for a 128×128×16 array, Moreover, the energy consumption (<;nJ/bit) of the write operation outperforms that in NAND FLASH. Finally, a feature size (F) of 30nm and 60nm with equal array densities are compared, which suggests that the downscaling of feature size might not be as crucial as believed and the design constraint can be relaxed as a result.
Keywords
electric resistance; random-access storage; three-dimensional integrated circuits; 3D RRAM array; 3D architecture; 3D cross-point array design; I-V nonlinearity ratio; ON-OFF-state resistance; energy consumption; interconnect material; large capacity RRAM array; next-generation nonvolatile memory; read margin; resistance 500 kohm; resistive switching random access memory; size 30 nm; size 60 nm; vertical transistor; write margin; write operation; Arrays; Electrodes; Flash memories; Nonvolatile memory; Three-dimensional displays; Transistors; 3D vertical RRAM; Rom; cross-point array; energy consumption; feature size; interconnect; nonlinearity;
fLanguage
English
Publisher
ieee
Conference_Titel
Memory Workshop (IMW), 2014 IEEE 6th International
Conference_Location
Taipei
Print_ISBN
978-1-4799-3594-9
Type
conf
DOI
10.1109/IMW.2014.6849382
Filename
6849382
Link To Document