Deep learning for exploring ultra-thin ferroelectrics with highly improved sensitivity of piezoresponse force microscopy

Panithan Sriboriboon; Huimin Qiao; Owoong Kwon; Rama K. Vasudevan; Stephen Jesse; Yunseok Kim

doi:10.1038/s41524-023-00982-0

Deep learning for exploring ultra-thin ferroelectrics with highly improved sensitivity of piezoresponse force microscopy

Panithan Sriboriboon, Huimin Qiao, Owoong Kwon, Rama K. Vasudevan, Stephen Jesse, Yunseok Kim

Research output: Contribution to journal › Article › peer-review

6 Scopus citations

Abstract

Hafnium oxide-based ferroelectrics have been extensively studied because of their existing ferroelectricity, even in ultra-thin film form. However, studying the weak response from ultra-thin film requires improved measurement sensitivity. In general, resonance-enhanced piezoresponse force microscopy (PFM) has been used to characterize ferroelectricity by fitting a simple harmonic oscillation model with the resonance spectrum. However, an iterative approach, such as traditional least squares (LS) fitting, is sensitive to noise and can result in the misunderstanding of weak responses. In this study, we developed the deep neural network (DNN) hybrid with deep denoising autoencoder (DDA) and principal component analysis (PCA) to extract resonance information. The DDA/PCA-DNN improves the PFM sensitivity down to 0.3 pm, allowing measurement of weak piezoresponse with low excitation voltage in 10-nm-thick Hf_0.5Zr_0.5O₂ thin films. Our hybrid approaches could provide more chances to explore the low piezoresponse of the ultra-thin ferroelectrics and could be applied to other microscopic techniques.

Original language	English
Article number	28
Journal	npj Computational Materials
Volume	9
Issue number	1
DOIs	https://doi.org/10.1038/s41524-023-00982-0
State	Published - Dec 2023

Funding

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. NRF-2021R1A2C2009642) and Basic Science Research Program through the NRF funded by the Ministry of Education (No. 2019R1A6A1A03033215). Some coding and analysis research was supported by the Center for Nanophase Materials Sciences (NMS), which is a US Department of Energy, Office of Science User Facility at Oak Ridge National Laboratory.

Funders	Funder number
Center for Nanophase Materials Sciences
NMS
U.S. Department of Energy
Office of Science
Oak Ridge National Laboratory
Ministry of Education	2019R1A6A1A03033215
Ministry of Science, ICT and Future Planning	NRF-2021R1A2C2009642
National Research Foundation of Korea

Access to Document

10.1038/s41524-023-00982-0

Cite this

@article{f5a364db320143f3b4e7f9497fb6e4f5,

title = "Deep learning for exploring ultra-thin ferroelectrics with highly improved sensitivity of piezoresponse force microscopy",

abstract = "Hafnium oxide-based ferroelectrics have been extensively studied because of their existing ferroelectricity, even in ultra-thin film form. However, studying the weak response from ultra-thin film requires improved measurement sensitivity. In general, resonance-enhanced piezoresponse force microscopy (PFM) has been used to characterize ferroelectricity by fitting a simple harmonic oscillation model with the resonance spectrum. However, an iterative approach, such as traditional least squares (LS) fitting, is sensitive to noise and can result in the misunderstanding of weak responses. In this study, we developed the deep neural network (DNN) hybrid with deep denoising autoencoder (DDA) and principal component analysis (PCA) to extract resonance information. The DDA/PCA-DNN improves the PFM sensitivity down to 0.3 pm, allowing measurement of weak piezoresponse with low excitation voltage in 10-nm-thick Hf0.5Zr0.5O2 thin films. Our hybrid approaches could provide more chances to explore the low piezoresponse of the ultra-thin ferroelectrics and could be applied to other microscopic techniques.",

author = "Panithan Sriboriboon and Huimin Qiao and Owoong Kwon and Vasudevan, {Rama K.} and Stephen Jesse and Yunseok Kim",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s).",

year = "2023",

month = dec,

doi = "10.1038/s41524-023-00982-0",

language = "English",

volume = "9",

journal = "npj Computational Materials",

issn = "2057-3960",

publisher = "Nature Research",

number = "1",

}

TY - JOUR

T1 - Deep learning for exploring ultra-thin ferroelectrics with highly improved sensitivity of piezoresponse force microscopy

AU - Sriboriboon, Panithan

AU - Qiao, Huimin

AU - Kwon, Owoong

AU - Vasudevan, Rama K.

AU - Jesse, Stephen

AU - Kim, Yunseok

PY - 2023/12

Y1 - 2023/12

N2 - Hafnium oxide-based ferroelectrics have been extensively studied because of their existing ferroelectricity, even in ultra-thin film form. However, studying the weak response from ultra-thin film requires improved measurement sensitivity. In general, resonance-enhanced piezoresponse force microscopy (PFM) has been used to characterize ferroelectricity by fitting a simple harmonic oscillation model with the resonance spectrum. However, an iterative approach, such as traditional least squares (LS) fitting, is sensitive to noise and can result in the misunderstanding of weak responses. In this study, we developed the deep neural network (DNN) hybrid with deep denoising autoencoder (DDA) and principal component analysis (PCA) to extract resonance information. The DDA/PCA-DNN improves the PFM sensitivity down to 0.3 pm, allowing measurement of weak piezoresponse with low excitation voltage in 10-nm-thick Hf0.5Zr0.5O2 thin films. Our hybrid approaches could provide more chances to explore the low piezoresponse of the ultra-thin ferroelectrics and could be applied to other microscopic techniques.

AB - Hafnium oxide-based ferroelectrics have been extensively studied because of their existing ferroelectricity, even in ultra-thin film form. However, studying the weak response from ultra-thin film requires improved measurement sensitivity. In general, resonance-enhanced piezoresponse force microscopy (PFM) has been used to characterize ferroelectricity by fitting a simple harmonic oscillation model with the resonance spectrum. However, an iterative approach, such as traditional least squares (LS) fitting, is sensitive to noise and can result in the misunderstanding of weak responses. In this study, we developed the deep neural network (DNN) hybrid with deep denoising autoencoder (DDA) and principal component analysis (PCA) to extract resonance information. The DDA/PCA-DNN improves the PFM sensitivity down to 0.3 pm, allowing measurement of weak piezoresponse with low excitation voltage in 10-nm-thick Hf0.5Zr0.5O2 thin films. Our hybrid approaches could provide more chances to explore the low piezoresponse of the ultra-thin ferroelectrics and could be applied to other microscopic techniques.

UR - http://www.scopus.com/inward/record.url?scp=85149173824&partnerID=8YFLogxK

U2 - 10.1038/s41524-023-00982-0

DO - 10.1038/s41524-023-00982-0

M3 - Article

AN - SCOPUS:85149173824

SN - 2057-3960

VL - 9

JO - npj Computational Materials

JF - npj Computational Materials

IS - 1

M1 - 28

ER -

Deep learning for exploring ultra-thin ferroelectrics with highly improved sensitivity of piezoresponse force microscopy

Abstract

Funding

Access to Document

Other files and links

Fingerprint

Cite this