Abstract:We report the development and implementation of a hypothesis learning based automated experiment, in which the microscope operating in the autonomous mode identifies the physical laws behind the material's response. Specifically, we explore the bias induced transformations that underpin the functionality of broad classes of devices and functional materials from batteries and memristors to ferroelectrics and antiferroelectrics. Optimization and design of these materials require probing the mechanisms of these transformations on the nanometer scale as a function of the broad range of control parameters such as applied potential and time, often leading to experimentally intractable scenarios. At the same time, often the behaviors of these systems are understood within potentially competing theoretical models, or hypotheses. Here, we develop a hypothesis list that covers the possible limiting scenarios for the domain growth, including thermodynamic, domain wall pinning, and screening limited. We further develop and experimentally implement the hypothesis driven automated experiment in Piezoresponse Force Microscopy, autonomously identifying the mechanisms of the bias induced domain switching. This approach can be applied for a broad range of physical and chemical experiments with relatively low dimensional control parameter space and for which the possible competing models of the system behavior that ideally cover the full range of physical eventualities are known or can be created. These include other scanning probe microscopy modalities such as force distance curve measurements and nanoindentation, as well as materials synthesis and optimization.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to identify the physical mechanisms behind domain switching in ferroelectric materials through autonomous experiments. Specifically, the author has developed and implemented an automated experimental method based on hypothesis learning, aiming to explore the bias - induced domain - switching phenomenon and optimize the mechanisms of these conversions. ### Problem Background Ferroelectric materials are widely used in a variety of functional devices, such as non - volatile random - access memories, ferroelectric tunnel barriers, multiferroic materials, and field - effect devices. These applications rely on the bias - induced polarization reversal or domain - conversion processes. However, these conversion processes are affected by multiple factors, including domain nucleation, domain - wall motion and pinning effects, as well as the dynamic behavior of surface - screening charges. In order to optimize and design these materials, it is necessary to study these conversion processes at the nanoscale and understand their variation with control parameters (such as applied voltage and time). ### Limitations of Traditional Methods Traditional experimental methods usually require intensive sampling of high - dimensional parameter spaces, which is very difficult and time - consuming experimentally. In addition, for local characterization, since it is necessary to probe a very high - dimensional parameter space, it is difficult to implement the experiment. Therefore, it has become particularly important to develop an automated experimental process that can efficiently explore relevant behaviors. ### Solution To solve the above problems, the author proposes an automated experimental method based on hypothesis learning. This method gradually narrows the range of possible physical mechanisms by pre - establishing a series of possible physical models (hypotheses) and continuously updating the probability distributions of these models during the experimental process, and finally determines the model that best fits the experimental data. ### Specific Implementation 1. **List of Hypotheses**: The author lists four possible domain - growth - limiting mechanisms: - Thermodynamic control (Model I and Model II) - Domain - wall - pinning dynamics (Model III) - Surface - screening - charge transport (Model IV) 2. **Automated Experimental Process**: By integrating multiple software and hardware systems, an automated experimental process is realized. During the experiment, the algorithm predicts the next experimental parameters based on the existing experimental results and continuously optimizes the model selection through a feedback mechanism. 3. **Experimental Verification**: An automated experiment based on hypothesis learning was carried out using BaTiO₃ (BTO) thin films as a model system. The experimental results show that domain growth is mainly controlled by the dynamics of domain - wall motion, rather than by the thermodynamic screening process. ### Conclusion This study demonstrates an automated experimental method based on hypothesis learning, which can efficiently explore the domain - conversion mechanisms in ferroelectric materials. This method is not only applicable to the research of ferroelectric materials, but can also be extended to other types of scanning - probe - microscope experiments and other fields of material synthesis and optimization. In this way, researchers can gain a deeper understanding of physical mechanisms in a shorter time, thereby accelerating the research and development process of new materials and new devices.

Autonomous scanning probe microscopy with hypothesis learning: Exploring the physics of domain switching in ferroelectric materials

Ferroelectric Domain and Switching Dynamics in Curved In2Se3: First-Principles and Deep Learning Molecular Dynamics Simulations.

Autonomous Experiments in Scanning Probe Microscopy and Spectroscopy: Choosing Where to Explore Polarization Dynamics in Ferroelectrics

Hypothesis Learning in Automated Experiment: Application to Combinatorial Materials Libraries

Laboratory Experiments Based on Tip Probe - Scanning Probe Detection and Regulation of Ferroelectric Domains and Their Microscopic Physical Properties

Experimental discovery of structure–property relationships in ferroelectric materials via active learning

"Laboratory experiments based on the tip probe"-detection and manipulation of ferroelectric domains and microscopic physical properties by scanning probe microscopy

Automated Experiments of Local Non-Linear Behavior in Ferroelectric Materials.

Physics discovery in nanoplasmonic systems via autonomous experiments in Scanning Transmission Electron Microscopy

Sub-10 nm Probing of Ferroelectricity in Heterogeneous Materials by Machine Learning Enabled Contact Kelvin Probe Force Microscopy

Probing Local Bias-Induced Transitions Using Photothermal Excitation Contact Resonance Atomic Force Microscopy and Voltage Spectroscopy

Bayesian inference in band excitation scanning probe microscopy for optimal dynamic model selection in imaging

Exploring Physics of Ferroelectric Domain Walls in Real Time: Deep Learning Enabled Scanning Probe Microscopy

Watching domains grow: In-situ studies of polarization switching by combined scanning probe and scanning transmission electron microscopy

Scanning Probe Manipulation of Magnetism at the LaAlO3/SrTiO3 Heterointerface.

Latent mechanisms of polarization switching from in situ electron microscopy observations

Causal analysis of competing atomistic mechanisms in ferroelectric materials from high-resolution scanning transmission electron microscopy data

Differentiating Ferroelectric and Nonferroelectric Electromechanical Effects with Scanning Probe Microscopy.

Bayesian Active Learning for Scanning Probe Microscopy: from Gaussian Processes to Hypothesis Learning

Local and correlated studies of humidity-mediated ferroelectric thin film surface charge dynamics

Probing Local Electromechanical Effects in Highly Conductive Electrolytes.