Failure analysis of large area pt/hfo 2 /pt capacitors using multilayer perceptrons
Loading...
Files
Accepted version
Date
2021-10
Authors
Muñoz-Gorriz, J.
Monaghan, Scott
Cherkaoui, Karim
Suñé, Jordi
Hurley, Paul K.
Miranda, Enrique
Journal Title
Journal ISSN
Volume Title
Publisher
IEEE
Published Version
Abstract
In this work, we investigated the spatial distribution of failure sites in large area Pt/HfO 2 /Pt capacitors using simple neural networks as classifiers. When an oxide breakdown (BD) occurs due to severe electrical stress, a mark shows up in the top metal electrode at the location where the failure event took place. The mark is the result of a microexplosion occurring inside the dielectric film. Large area devices need to be studied because the number of generated spots must be the required for statistical analysis. The obtained results using multilayer perceptrons with different number of neurons and hidden layers indicate that the largest breakdown spots tend to concentrate towards the center of the device. This observation is consistent with previous exploratory analysis carried out using spatial statistics techniques. This exercise shows the suitability of multilayer perceptrons for investigating the distribution of failure sites or defects on a given surface.
Description
Keywords
Neural networks , Perceptron , Spatial statistics , MIM , Dielectric breakdown , Reliability
Citation
Munoz-Gorriz, J., Monaghan, S., Cherkaoui, K., Sune, J., Hurley, P.K. and Miranda, E. (2021) ‘Failure analysis of large area pt/hfo 2 /pt capacitors using multilayer perceptrons’, in 2021 IEEE International Symposium on the Physical and Failure Analysis of Integrated Circuits (IPFA), Singapore, pp. 1–5. doi: 10.1109/IPFA53173.2021.9617281
Link to publisher’s version
Copyright
© 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.