Litcius/Paper detail

Optimized IGZO FETs for Capacitorless DRAM with Retention of 10 ks at RT and 7 ks at 85 °C at Zero V<sub>hold</sub> with Sub-10 ns Speed and 3-bit Operation

Qianlan Hu, Qijun Li, Shenwu Zhu, Chengru Gu, Shiyuan Liu, Ru Huang, Yanqing Wu

20222022 International Electron Devices Meeting (IEDM)43 citationsDOI

Abstract

The emerging capacitorless DRAM based on amorphous oxide semiconductor shows encouraging retention performance, while great challenges still exist in the need of a negative hold voltage and the low on-current with slow write speed. In this work, a BEOL-compatible capacitorless (2T0C) DRAM cell based on optimized amorphous IGZO FETs as write transistors is demonstrated, with a record high retention time and fastest write operation. optimized IGZO transistors at high positive threshold voltage (V <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</inf> ) over 1.2 V exhibit improved on-current (I <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">on</inf> ) of 24 $\mu$A/$\mu$m owing to the insertion of a thin In-rich ITO interlayer at source/drain area with greatly reduced contact resistance, enabling sub-ten-nanosecond speed. The high V <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</inf> enables remarkable retention time for the 2T0C DRAM cell at zero hold voltage, capable of keeping data over a record-long 10 ks and 7 ks retention time at room temperature and 85°C, respectively. Furthermore, 3-bit memory operations with high linearity are achieved by changing the WBL voltage or WWL voltage at temperatures up to 85°C.

Topics & Concepts

DramData retentionElectrical engineeringMaterials scienceTransistorOptoelectronicsVoltageAmorphous solidAnalytical Chemistry (journal)Topology (electrical circuits)PhysicsChemistryEngineeringCrystallographyChromatographySemiconductor materials and devicesThin-Film Transistor TechnologiesAdvanced Memory and Neural Computing