An official website of the United States government
Abstract Many energy conversion, sensing, and microelectronic applications based on ferroic materials are determined by the domain structure evolution under applied stimuli. New hyperspectral, multidimensional spectroscopic techniques now probe dynamic responses at relevant length and time scales to provide an understanding of how these nanoscale domain structures impact macroscopic properties. Such approaches, however, remain limited in use because of the difficulties that exist in extracting and visualizing scientific insights from these complex datasets. Using multidimensional band‐excitation scanning probe spectroscopy and adapting tools from both computer vision and machine learning, an automated workflow is developed to featurize, detect, and classify signatures of ferroelectric/ferroelastic switching processes in complex ferroelectric domain structures. This approach enables the identification and nanoscale visualization of varied modes of response and a pathway to statistically meaningful quantification of the differences between those modes. Among other things, the importance of domain geometry is spatially visualized for enhancing nanoscale electromechanical energy conversion.
more »
« less
Why it is Unfortunate that Linear Machine Learning “Works” so well in Electromechanical Switching of Ferroelectric Thin Films
Abstract Machine learning (ML) is relied on for materials spectroscopy. It is challenging to make ML models fail because statistical correlations can mimic the physics without causality. Here, using a benchmark band‐excitation piezoresponse force microscopy polarization spectroscopy (BEPS) dataset the pitfalls of the so‐called “better”, “faster”, and “less‐biased” ML of electromechanical switching are demonstrated and overcome. Using a toy and real experimental dataset, it is demonstrated how linear nontemporal ML methods result in physically reasonable embedding (eigenvalues) while producing nonsensical eigenvectors and generated spectra, promoting misleading interpretations. A new method of unsupervised multimodal hyperspectral analysis of BEPS is demonstrated using long‐short‐term memory (LSTM) β‐variational autoencoders (β‐VAEs) . By including LSTM neurons, the ordinal nature of ferroelectric switching is considered. To improve the interpretability of the latent space, a variational Kullback–Leibler‐divergency regularization is imposed . Finally, regularization scheduling of β as a disentanglement metric is leveraged to reduce user bias. Combining these experiment‐inspired modifications enables the automated detection of ferroelectric switching mechanisms, including a complex two‐step, three‐state one. Ultimately, this work provides a robust ML method for the rapid discovery of electromechanical switching mechanisms in ferroelectrics and is applicable to other multimodal hyperspectral materials spectroscopies.
more »
« less
- Award ID(s):
- 1839234
- PAR ID:
- 10375802
- Publisher / Repository:
- Wiley Blackwell (John Wiley & Sons)
- Date Published:
- Journal Name:
- Advanced Materials
- Volume:
- 34
- Issue:
- 47
- ISSN:
- 0935-9648
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Piezoresponse force microscopy (PFM) is routinely used to probe the nanoscale electromechanical response of ferroelectric and piezoelectric materials. However, many challenges remain in the interpretation of the recovered signal. Specifically, many non‐ferroelectric contributions affect the measured response, ranging from electrostatics, to charge injection and trapping, and topographic cross‐talk. Recently, machine learning (ML) has been utilized to identify multiple contributors within complex data systems, such as PFM response. A substantial advancement in ML approaches for PFM techniques is offered by dimensional stacking, enabling encoding of physical and/or chemical correlations within the materials' response across different data dimensions spanning varying ranges. However, dimensional stacking requires appropriate scaling for each dimension (before ML analysis) to minimize undesired information loss. Here, the impact of clustering globally and locally scaled parameters in polarization switching experiments via resonant PFM (RPFM) are discussed. Specifically, dimensional stacking of scaled parameters can mask or enhance ferroelectric and non‐ferroelectric behaviors, and aid identification of various physical phenomena contributing to the measured RPFM response. This study highlights the importance of data curation for ML, and its role in identifying signal contributors to scanning probe microscopy (SPM)‐based techniques with multidimensional data, such as resonant and/or spectroscopic SPM.more » « less
-
Abstract Scanning Probe Microscopy (SPM) based techniques probe material properties over microscale regions with nanoscale resolution, ultimately resulting in investigation of mesoscale functionalities. Among SPM techniques, piezoresponse force microscopy (PFM) is a highly effective tool in exploring polarization switching in ferroelectric materials. However, its signal is also sensitive to sample‐dependent electrostatic and chemo‐electromechanical changes. Literature reports have often concentrated on the evaluation of theOff‐fieldpiezoresponse, compared toOn‐fieldpiezoresponse, based on the latter's increased sensitivity to non‐ferroelectric contributions. Using machine learning approaches incorporatingboth Off‐andOn‐fieldpiezoresponse response as well asOff‐fieldresonance frequency to maximize information, switching piezoresponse in a defect‐rich Pb(Zr,Ti)O3thin film is investigated. As expected, one major contributor to the piezoresponse is mostly ferroelectric, coupled with electrostatic phenomena duringOn‐fieldmeasurements. A second component is electrostatic in nature, while a third component is likely due to a superposition of multiple non‐ferroelectric processes. The proposed approach will enable deeper understanding of switching phenomena in weakly ferroelectric samples and materials with large chemo‐electromechanical response.more » « less
-
Abstract Memristors with excellent scalability have the potential to revolutionize not only the field of information storage but also neuromorphic computing. Conventional metal oxides are widely used as resistive switching materials in memristors. Interface‐type memristors based on ferroelectric materials are emerging as alternatives in the development of high‐performance memory devices. A clear understanding of the switching mechanisms in this type of memristors, however, is still in its early stages. By comparing the bipolar switching in different systems, it is found that the switchable diode effect in ferroelectric memristors is controlled by polarization modulated Schottky barrier height and polarization coupled interfacial deep states trapping/detrapping. Using semiconductor theories with consideration of polarization effects, a phenomenological theory is developed to explain the current–voltage behavior at the metal/ferroelectric interface. These findings reveal the critical role of the interaction among polarization charges, interfacial defects, and Schottky interface in controlling ferroelectric resistive switching and offer the guidance to design ferroelectric memristors with enhanced performance.more » « less
-
Abstract Integration of ferroelectric materials into novel technological applications requires low coercive field materials, and consequently, design strategies to reduce the ferroelectric switching barriers. In this first principles study, we show that biaxial strain, which has a strong effect on the ferroelectric ground states, can also be used to tune the switching barrier of hybrid improper ferroelectric Ruddlesden–Popper oxides. We identify the region of the strain-tolerance factor phase diagram where this intrinsic barrier is suppressed, and show that it can be explained in relation to strain-induced phase transitions to nonpolar phases.more » « less
