This paper presents our recently developed AI-powered scanning probe microscope (AI-SPM) for autonomous atomic-scale measurements. The system can recognize single-atom positions with high precision and autonomously perform tasks such as spectroscopic measurements and atomic manipulations. It not only identifies individual atoms but also detects surface defects, avoiding them when necessary to perform measurements on target atoms. Additionally, it autonomously corrects for thermal drift and repairs probe tips, common challenges in SPM experiments. We demonstrated the effectiveness of the system using a room-temperature scanning tunneling microscope (STM) on a Si(111)-(7 × 7) surface. The AI-SPM performed numerous I-V measurements at four different Si adatom sites, revealing differences in electronic density of states. A large dataset, essential for reliable material property assessments, was generated, showcasing AI-SPM’s ability to significantly enhance data acquisition quality. This achievement represents a step towards more effective, precise, and reliable atomic-level surface analysis, potentially bringing substantial advancements to material characterization techniques.

To observe samples by transmission electron microscopy (TEM), it is necessary to irradiate them with a high-energy electron beam. This irradiation, however, causes significant damage to organic materials. To address this issue, low electron-dose imaging is essential for preserving the original structure of the samples. Under such low-dose conditions, the limited electron quantity results in noisy images, complicating structural analysis. Recently, advanced image analysis techniques using machine learning have emerged as promising tools for extracting signals from noisy images. This paper introduces a 3D tensor decomposition method to effectively remove noise from electron wave interference patterns (holograms) obtained under low-dose conditions. This approach facilitates the clear observation of the electric potential distribution in organic electroluminescence (EL) devices using 1/60 of the conventional electron dose. The technical approach and corresponding experimental results are discussed in detail.

This paper explores the application of informatics technologies to “measurement,” a core aspect of research in the natural sciences. Specifically, it focuses on the analysis of spectrum image (SI) data acquired through the combination of scanning transmission electron microscopy (STEM) with electron energy loss spectroscopy (EELS) and energy-dispersive X-ray spectroscopy (EDS). The study emphasizes the reliability and limitations of chemical imaging achieved using non-negative matrix factorization (NMF). Through the example of label-free chemical imaging of polymer blends via the STEM-EELS-SI method, the research demonstrates the construction of a sparse, multidimensional descriptor space that captures the intrinsic information embedded in experimental data. This methodology addresses the challenges of NMF and provides a physically interpretable approach to chemical imaging. Lastly, the paper proposes practical recommendations for integrating informatics technologies into measurement processes, enhancing the accuracy and interpretability of data analysis in materials science.

In recent years, the evolution of measurement and analysis techniques and the dramatic improvement of computational power have made it possible to obtain large amounts of complex data, and informatics methods are being actively used to extract new meaning and knowledge from these data. Our research group focuses on the integration of metrology and informatics, a field we term Metrology Informatics, with primary applications in the analysis of semiconductor materials and devices. In this paper, we present two examples from our work. The first involves applying one-sided orthogonal non-negative matrix factorization to three-dimensional scanning electron microscopy-cathodoluminescence (SEM-CL) spectral images, enabling the analysis of emission modes using statistical insights. The second example showcases a multimodal analysis approach, employing non-rigid registration methods to integrate 3D microscopy data from 3D atom probe tomography (3DAP) and electron tomography (ET), which have differing data structures.

The demand for advanced spectroscopic techniques capable of direct observation and analysis at the nanoscale is growing rapidly, driven by progress in nanomaterials research, which is expected to play a key role in developing innovative devices based on the quantum properties of matter, as well as in biotechnology and biomedical engineering. Nano-spectroscopic imaging has become an indispensable experimental method for uncovering novel properties and functions of nanomaterials and understanding their underlying mechanisms. It enables precise evaluation of material structures, chemical compositions, and electrical and optical properties with nanoscale spatial resolution. This article introduces the latest advancements in infrared scattering-type near-field optical microscopy (IR-SNOM), a powerful nano-spectroscopic imaging technique, along with recent research from the author’s group. IR-SNOM extends the capabilities of infrared spectroscopy—a highly effective method for chemical analysis and property evaluation—by achieving spatial resolution beyond the diffraction limit. We describe highly sensitive vibrational spectroscopy of single proteins and the direct observation of the insulator-to-metal phase transition in vanadium dioxide nanoparticles.

Digital bioanalysis, which can quantify target molecules at the single-molecule level, is gaining attention as the next-generation bioanalysis method due to its high detection sensitivity and accuracy. Currently, miniaturization and cost reduction of analytical devices are challenges for the widespread adoption and practical use of digital bioanalysis. Here, we introduce the low-cost and portable wide-field fluorescence imaging device (COWFISH) recently developed for digital bioanalysis and present the future prospects of these advances.

This paper describes the positioning of the ultra-low accelerating voltage scanning electron microscope (ULV-SEM) equipped with multiple imaging detectors in the history of SEM development. ULV-SEM provides rich information once the user finds the “sweet spot” for both secondary electron and backscattered electron images based on an understanding of the signal acceptance of the instrument. Use of multiple imaging detectors allows acquisition of various images with a single scan. X-ray microanalysis under the same experimental conditions as the observation “sweet spot” has become possible with a windowless X-ray spectrometer optimized for use at a short working distance.

Fluorophores widely used in fluorescence microscopy have substance-specific excitation and fluorescence spectra. Fluorescence lifetimes are also fluorescent substance-specific values. These fluorescence lifetimes allow biosensors to measure microenvironmental parameters such as temperature, molecular interaction and ion-concentration, as well as multi-color imaging by separating fluorescence. Recently, it has also been used to separate fluorescence in super-resolution microscopy and improve resolution. This paper introduces the usefulness of fluorescence lifetime.