Automated scanning probe microscopy of combinatorial ferroelectric libraries: Gaussian-process-guided exploration and noise-aware experiment planning

Liu, Yu [University of Tennessee, Knoxville, TN (United States)] (ORCID:0000000198719301); Pant, Rohit [University of Maryland, College Park, MD (United States)] (ORCID:0000000245915752); Takeuchi, Ichiro [University of Maryland, College Park, MD (United States)] (ORCID:0000000326250553); Spurling, R. Jackson [The Pennsylvania State University, University Park, PA (United States)]; Maria, Jon-Paul [The Pennsylvania State University, University Park, PA (United States)]; Ziatdinov, Maxim [Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)] (ORCID:0000000325704592); Kalinin, Sergei V. [University of Tennessee, Knoxville, TN (United States)] (ORCID:0000000153546152)

doi:10.1063/5.0341018

DOE OSTI · 3536194

Automated scanning probe microscopy of combinatorial ferroelectric libraries: Gaussian-process-guided exploration and noise-aware experiment planning

Liu, Yu [University of Tennessee, Knoxville, TN (United States)] (ORCID:0000000198719301)·Pant, Rohit [University of Maryland, College Park, MD (United States)] (ORCID:0000000245915752)·Takeuchi, Ichiro [University of Maryland, College Park, MD (United States)] (ORCID:0000000326250553)·Spurling, R. Jackson [The Pennsylvania State University, University Park, PA (United States)]·Maria, Jon-Paul [The Pennsylvania State University, University Park, PA (United States)]·Ziatdinov, Maxim [Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)] (ORCID:0000000325704592)·Kalinin, Sergei V. [University of Tennessee, Knoxville, TN (United States)] (ORCID:0000000153546152)

View original source ↗Read source document ↗Export metadata ↓

Abstract

Combinatorial materials libraries provide an efficient route for mapping composition–property relationships, but their broader impact depends on rapid, quantitative, and functionally relevant characterization. Scanning Probe Microscopy (SPM), including piezoresponse force microscopy (PFM), offers significant potential for quantitative, functionally relevant combi-library readouts. Here, we implement a fully automated SPM workflow for ferroelectric combinatorial libraries and benchmark Gaussian-process-based Bayesian optimization strategies for autonomous experiment planning. The workflow integrates automated probe motion, contact optimization, imaging, and dual amplitude resonance tracking-PFM spectroscopy, and uses scalarized spectroscopic observables to guide subsequent measurements. Stage motion, probe engagement, in-contact tuning, imaging, spectroscopy, and the choice of the next measurement location all proceed without human input. We demonstrate the approach on Sm-doped BiFeO 3 and Zn x Mg 1−x O libraries. By comparing vanilla Bayesian optimization with a measured-noise variant, we show that explicit treatment of local reproducibility can improve modeling of composition-dependent response when the measured variance is physically meaningful, but can also reduce robustness when variability is dominated by outliers or topographic artifacts. Furthermore, these results establish automated SPM as a bridge between combinatorial synthesis and quantitative functional characterization.

Keep this discovery

BibTeX RIS

Liu, Yu [University of Tennessee, Knoxville, TN (United States)] (ORCID:0000000198719301), Pant, Rohit [University of Maryland, College Park, MD (United States)] (ORCID:0000000245915752), Takeuchi, Ichiro [University of Maryland, College Park, MD (United States)] (ORCID:0000000326250553), Spurling, R. Jackson [The Pennsylvania State University, University Park, PA (United States)], Maria, Jon-Paul [The Pennsylvania State University, University Park, PA (United States)], Ziatdinov, Maxim [Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)] (ORCID:0000000325704592), Kalinin, Sergei V. [University of Tennessee, Knoxville, TN (United States)] (ORCID:0000000153546152). 2026-08-27. Automated scanning probe microscopy of combinatorial ferroelectric libraries: Gaussian-process-guided exploration and noise-aware experiment planning. https://doi.org/10.1063/5.0341018

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related reports

DOE OSTI2026

Scan‐Path‐ and Initial‐State‐Dependent Superdomain Switching in (111)‐Oriented PZT

Polarization switching in ferroelectric materials arises from the collective evolution of complex domain hierarchies, yet deterministic control over these processes remains challenging. Here, we investigate scan-path- and initial-state-dependent switching in epitaxial (111)-oriented PbZr 0.2 Ti 0.8 O 3 thin films using automated AFM-based writing combined with quantitative 3D piezoresponse force microscopy. We show that the scan trajectory acts as an experimentally accessible control parameter for superdomain formation. Box-in-box raster scans reproducibly stabilize ordered stripe superdomains with a reduced subset of symmetry-allowed variants, whereas spiral trajectories generate frustrated mixed-variant states with a broader distribution of final microstructures. Automated pulsing experiments further show that the local superdomain configuration at the nucleation site strongly influences the final written morphology. Phase-field modeling qualitatively reproduces the contrast between representative initial-state geometries and supports the role of compatibility constraints among competing ferroelastic pathways. These findings establish scan-path and initial-state engineering as practical handles to program ferroic order in hierarchical ferroelectric domain structures.

Vasudevan, Rama K. [Oak Ridge National Laboratory ↗

DOE OSTI2024

Adaptive Scalpel Scanning Probe Microscopy for Enhanced Volumetric Sensing in Tomographic Analysis

Controlling nanoscale tip‐induced material removal is crucial for achieving atomic‐level precision in tomographic sensing with atomic force microscopy (AFM). While advances have enabled volumetric probing of conductive features with nanometer accuracy in solid‐state devices, materials, and photovoltaics, limitations in spatial resolution and volumetric sensitivity persist. This work identifies and addresses in‐plane and vertical tip‐sample junction leakage as sources of parasitic contrast in tomographic AFM, hindering real‐space 3D reconstructions. Novel strategies are proposed to overcome these limitations. First, the contrast mechanisms analyzing nanosized conductive features are explored when confining current collection purely to in‐plane transport, thus allowing reconstruction with a reduction in the overestimation of the lateral dimensions. Furthermore, an adaptive tip‐sample biasing scheme is demonstrated for the mitigation of a class of artefacts induced by the high electric field inside the thin oxide when volumetrically reduced. This significantly enhances vertical sensitivity by approaching the intrinsic limits set by quantum tunneling processes, allowing detailed depth analysis in thin dielectrics. The effectiveness of these methods is showcased in tomographic reconstructions of conductive filaments in valence change memory, highlighting the potential for application in nanoelectronics devices and bulk materials and unlocking new limits for tomographic AFM.

36 MATERIALS SCIENCE↗

NASA NTRS2026

Benchmarking Bayesian Optimization Frameworks and Acquisition Strategies for Materials Discovery and Autonomous Laboratories

Bayesian optimization (BO) can accelerate materials discovery by guiding expensive experiments toward the most promising processing conditions. We systematically compare five BO surrogate and framework combinations (Gaussian processes in Ax, Gaussian processes and Monte-Carlo neural networks in BayBE, random forests in Lolopy, and tree-structured Parzen (TPE) estimators in Hyperopt) on three benchmarks that mimic common materials design tasks (a discrete solid-electrolyte composition space, a hybrid discrete/continuous laminate-composite design problem solved with micromechanics modeling, and the continuous Ishigami analytic function which is a standard optimization benchmark). Each BO surrogate is paired with posterior mean, probability of improvement, and expected improvement acquisition functions and run for 100 trials from randomized initial samples with uniform random search providing a control. Across five random seeds per setting, BayBE’s Gaussian-process surrogate with expected improvement consistently reached ≥95 % of the known optimum in the fewest evaluations, while Lolopy’s random forest matched or exceeded GP performance on purely categorical or mixed spaces at a higher computational cost. Posterior mean alone often stagnated at local optima, underscoring the need for exploration, whereas probability and expected improvement balanced exploration and exploitation leading to better optimization in fewer trials. Execution times ranged from milliseconds for TPE to minutes for neural-network and random-forest surrogates. These results establish baseline expectations for BO in automated materials laboratories and highlight expected improvement with Gaussian processes as a reliable first choice, with random forests offering a strong alternative when categorical variables dominate. The benchmark suite and code are released to facilitate future surrogate, acquisition, and constraint-handling research in data-driven materials optimization.

Bayesian optimization↗