Hardware-assisted 3D TCAD for predictive capacitance extraction in 32nm SOI SRAMs

A comprehensive technology-node/process/layout-independent TCAD flow for guiding FEOL/BEOL analysis/design of 32nm SOI SRAMs is presented using, for the first time, iterative 3D TCAD capacitance extraction assisted by hardware data. The methodology is unique in the industry, giving insight into FEOL...

Full description

Saved in:
Bibliographic Details
Main Authors: Bhoj, A. N., Joshi, R. V., Polonsky, S., Kanj, R., Saroop, S., Tan, Y., Jha, N. K.
Format: Conference Proceeding
Language:English
Subjects:
Capacitance
Capacitance measurement
Hardware
Layout
Optimization
Random access memory
Three dimensional displays
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A comprehensive technology-node/process/layout-independent TCAD flow for guiding FEOL/BEOL analysis/design of 32nm SOI SRAMs is presented using, for the first time, iterative 3D TCAD capacitance extraction assisted by hardware data. The methodology is unique in the industry, giving insight into FEOL/BEOL components independently, when total capacitance is the only experimentally measurable quantity. 32nm SOI simulations from the flow are in excellent agreement with hardware data for two different 6T SRAM macros in the same process. In particular, they isolate the FEOL component, consisting mainly of junction capacitance, as the dominant factor affecting total bitline capacitance variation across wafers, owing to the sensitivity to body doping. Leveraging hardware data, the method is able to effectively predict other key capacitances (e.g., wordline) of generic layouts in the same process, thereby reducing the silicon footprint (cost) for test structures during early phases of technology development.
ISSN:0163-1918
2156-017X
DOI:10.1109/IEDM.2011.6131673