%59%6f%6e%67%73%6f%6f%20%59%61%6e%67
-
KAIST Directly Visualizes the Hidden Spatial Order of Electrons in a Quantum Material
<(Back row, from left) Yeongkwan Kim, SungBin Lee, Heejun Yang, Yongsoo Yang_(Front row, from left) Jemin Park, Seokjo Hong, Jaewhan Oh>
· Cryogenic 4D-STEM reveals how charge density waves form, fragment, and persist across a phase transition
· First direct measurement of electronic amplitude correlations uncovers strain-driven inhomogeneity and localized order above the transition temperature
Electronic order in quantum materials often emerges not uniformly, but through subtle and complex patterns that vary from place to place. One prominent example is the charge density wave (CDW), an ordered state in which electrons arrange themselves into periodic patterns at low temperatures. Although CDWs have been studied for decades, how their strength and spatial coherence evolve across a phase transition has remained largely inaccessible experimentally.
Now, a team led by Professor Yongsoo Yang of the Department of Physics at KAIST (Korea Advanced Institute of Science and Technology), together with Professors SungBin Lee, Heejun Yang, and Yeongkwan Kim, and in collaboration with Stanford University, has for the first time directly visualized the spatial evolution of charge density wave amplitude order inside a quantum material.
A New Way to See Electronic Order at the Nanoscale
Using a liquid-helium-cooled electron microscope setup combined with four-dimensional scanning transmission electron microscopy (4D-STEM), the researchers mapped how CDW order develops, weakens, and fragments as temperature changes. This approach allowed them to reconstruct nanoscale maps of the CDW amplitude, revealing not just whether the order exists, but how strong it is and how it is spatially connected.
This study is similar to filming the growth of ice crystals as water freezes using an ultra-high-magnification camera. In this case, however, the researchers observed electrons arranging themselves at cryogenic temperatures of around –253°C, and used an electron microscope capable of resolving features one hundred-thousandth the width of a human hair instead of a conventional camera. The results showed that the electronic patterns do not appear uniformly across the material. In some regions, clear patterns are visible, while in neighboring areas they are entirely absent, much like a lake that does not freeze all at once, with patches of ice interspersed with liquid water.
How Electronic Order Breaks Apart in Real Space
The team further demonstrated that this spatial inhomogeneity is closely linked to local strain inside the crystal. Even extremely small distortions that are far below optical resolution strongly suppress the CDW amplitude. This clear anticorrelation between strain and electronic order provides direct evidence that local lattice distortions play a decisive role in shaping CDW patterns.
Unexpectedly, the researchers also observed that localized regions of CDW order can persist even above the transition temperature, where long-range order is generally thought to disappear. These isolated pockets of electronic order suggest that the CDW transition is not a simple, uniform melting process, but instead involves gradual loss of spatial coherence.
A key advance of this work is the world’s first direct measurement of CDW amplitude correlations. By quantifying how the strength of electronic order at one location is related to that at another, the study reveals how CDW coherence collapses across the transition, while local amplitude remains finite. Such information could not be obtained with conventional diffraction or scanning probe techniques.
Toward a New Framework for Studying Electronic Order
Charge density waves are a central feature of many quantum materials and often coexist or compete with other electronic states. By directly accessing their spatial structure and correlations, this study provides a new experimental framework for understanding how collective electronic order forms and evolves in real materials.
Dr. Yongsoo Yang, who led the research, explained the significance of the results: “Until now, the spatial coherence of charge density waves was largely inferred indirectly. Our approach allows us to directly visualize how electronic order varies across space and temperature, and to identify the factors that locally stabilize or suppress it.”
[Figure 1] Schematic illustration of an experiment employing 4D-STEM to probe the spatial variations of charge density waves in the prototypical quantum material NbSe2 under a liquid-helium cryogenic environment (AI-generated image).
This research, with Seokjo Hong, Jaewhan Oh and Jemin Park of KAIST as co-first authors, was published online in Physical Review Letters on January 6th (Title: Spatial correlations of charge density wave order across the transition in 2H-NbSe2).
The study was mainly supported by the National Research Foundation of Korea (NRF) Grants (Individual Basic Research Program, Basic Research Laboratory Program, Nanomaterial Technology Development Program) funded by the Korean Government (MSIT).
2026.01.20 View 2205 -
Tracking Atoms during Fuel Cell Cycles: KAIST Team Reveals the Atomic-Scale Secret Behind Fuel Cell Catalyst Durability
<Professor Yongsoo Yang, Professor Eun-Ae Cho, Dr. Chaehwa Jeong, Dr. Joohyuk Lee, Dr. Hyesung Cho, Researcher Kwangho Lee from KAIST>
Hydrogen fuel cell vehicles have long been hailed as the future of clean mobility: cars that emit nothing but water while delivering high efficiency and power density. Yet a stubborn obstacle remains. The heart of the fuel cell, the platinum-based catalyst, is both expensive and prone to degradation. Over time, the catalyst deteriorates during operation, forcing frequent replacements and keeping hydrogen vehicles costly.
Understanding why and how these catalysts degrade at the atomic level is a longstanding challenge in the catalysis research. Without this knowledge, designing truly durable and affordable fuel cells for mass adoption remains out of reach.
Now, a team led by Professor Yongsoo Yang of the Department of Physics at KAIST (Korea Advanced Institute of Science and Technology), in collaboration with Professor Eun-Ae Cho of KAIST’s Department of Materials Science and Engineering, researchers at Stanford University and the Lawrence Berkeley National Laboratory, has successfully tracked the three-dimensional change of individual atoms inside fuel cell catalysts during thousands of operating cycles. The results provide unprecedented insight into the atomic-scale degradation mechanisms of platinum-nickel (PtNi) catalysts, and demonstrate how gallium (Ga) doping dramatically improves both their performance and durability.
A New Atomic “CT Scan” for Catalysts
To achieve this breakthrough, the team utilized a neural network-assisted atomic electron tomography (AET) technique. Much like a CT scan in a hospital reconstructs the inside of the human body from X-ray images, AET determines the positions of thousands of atoms inside nanomaterials from high-resolution electron microscopy images taken at many different angles. By combining these reconstructions with advanced AI-based data correction, the researchers were able to map the exact 3D coordinates and chemical identity of every atom in the nanoparticle catalysts.
This allowed them to directly observe—at single-atom resolution—how the catalysts changed in structure, chemical composition, and internal strain as they were cycled thousands of times under fuel cell operating conditions.
Key Findings: Why Gallium Makes a Difference
The researchers compared conventional PtNi catalysts with Ga-doped PtNi catalysts. The results revealed:
a) Shape stability: While undoped PtNi particles gradually lost their advantageous octahedral shape and became more spherical (i.e., the fraction of highly active {111} facets has been reduced), Ga-doped particles retained their octahedral shape even after 12,000 cycles.
b) Chemical stability: In PtNi catalysts, nickel atoms leached from both the surface and subsurface regions, driving structural instability. In Ga-doped catalysts, surface nickel was largely preserved, preventing collapse of the structure.
c) Strain preservation: The compressive strain in PtNi particles, crucial for optimizing oxygen reduction activity, relaxed substantially over time. In contrast, Ga-doped particles maintained near-optimal strain.
d) Catalytic performance: By integrating these factors, the researchers showed that while undoped PtNi catalysts lost ~17% of their oxygen reduction activity after 12,000 cycles, Ga-doped PtNi catalysts lost only ~4% and maintained superior activity throughout.
Dr. Yang, who led the research, explained the significance of the results: “These results represent the first time the true 3D atomic-scale degradation dynamics of fuel cell catalysts have been directly visualized. Our findings not only reveal why gallium doping works, but also establish a powerful framework for rationally designing durable, high-efficiency catalysts.”
Implications for a Hydrogen-Powered Future
The study demonstrates that neural network-assisted AET can reveal how nanomaterials evolve during real operating conditions, overcoming the limitations of traditional 2D imaging and ensemble-averaged methods. Beyond PtNi catalysts, the technique can be applied to a wide range of nanomaterials and catalytic systems, helping to design the next generation of energy materials with atomic precision.
For the hydrogen economy, this means that more durable catalysts could extend the lifetime of fuel cells, lower replacement costs, and accelerate the widespread adoption of hydrogen-powered vehicles and clean energy technologies.
[Figure 1] Three-dimensional atomic structures and catalytic activity of Ga-doped PtNi nanoparticles during potential cycling. The top row shows the 3D atomic structures at different stages (Pristine to 12,000 cycles; blue: platinum, pink: nickel). The bottom row visualizes oxygen reduction reaction (ORR) catalytic activity, where red indicates higher activity. Gallium doping stabilizes the octahedral geometry and preserves highly active {111} facets, enabling sustained catalytic performance even after extensive cycling.
This research, with Chaehwa Jeong, Juhyeok Lee, Hyesung Jo, KwangHo Lee from the KAIST as co-first authors, was published online in Nature Communications on August 28th (Title: Atomic-scale 3D structural dynamics and functional degradation of Pt alloy nanocatalysts during the oxygen reduction reaction).
The study was mainly supported by the National Research Foundation of Korea (NRF) Grants funded by the Korean Government (MSIT).
2025.09.15 View 3215 -
A 20-year-old puzzle solved: KAIST research team reveals the 'three-dimensional vortex' of zero-dimensional ferroelectrics
Materials that can maintain a magnetized state by themselves without an external magnetic field (i.e., permanent magnets) are called ferromagnets. Ferroelectrics can be thought of as the electric counterpart to ferromagnets, as they maintain a polarized state without an external electric field. It is well-known that ferromagnets lose their magnetic properties when reduced to nano sizes below a certain threshold. What happens when ferroelectrics are similarly made extremely small in all directions (i.e., into a zero-dimensional structure such as nanoparticles) has been a topic of controversy for a long time.
< (From left) Professor Yongsoo Yang, the corresponding author, and Chaehwa Jeong, the first author studying in the integrated master’s and doctoral program, of the KAIST Department of Physics >
The research team led by Dr. Yongsoo Yang from the Department of Physics at KAIST has, for the first time, experimentally clarified the three-dimensional, vortex-shaped polarization distribution inside ferroelectric nanoparticles through international collaborative research with POSTECH, SNU, KBSI, LBNL and University of Arkansas.
About 20 years ago, Prof. Laurent Bellaiche (currently at University of Arkansas) and his colleagues theoretically predicted that a unique form of polarization distribution, arranged in a toroidal vortex shape, could occur inside ferroelectric nanodots. They also suggested that if this vortex distribution could be properly controlled, it could be applied to ultra-high-density memory devices with capacities over 10,000 times greater than existing ones. However, experimental clarification had not been achieved due to the difficulty of measuring the three-dimensional polarization distribution within ferroelectric nanostructures.
The research team at KAIST successfully solved this 20-year-old challenge by implementing a technique called atomic electron tomography. This technique works by acquiring atomic-resolution transmission electron microscope images of the nanomaterials from multiple tilt angles, and then reconstructing them back into three-dimensional structures using advanced reconstruction algorithms. Electron tomography can be understood as essentially the same method with the CT scans used in hospitals to view internal organs in three dimensions; the KAIST team adapted it uniquely for nanomaterials, utilizing an electron microscope at the single-atom level.
< Figure 1. Three-dimensional polarization distribution of BaTiO3 nanoparticles revealed by atomic electron tomography. >(Left) Schematic of the electron tomography technique, which involves acquiring transmission electron microscope images at multiple tilt angles and reconstructing them into 3D atomic structures.(Center) Experimentally determined three-dimensional polarization distribution inside a BaTiO3 nanoparticle via atomic electron tomography. A vortex-like structure is clearly visible near the bottom (blue dot).(Right) A two-dimensional cross-section of the polarization distribution, thinly sliced at the center of the vortex, with the color and arrows together indicating the direction of the polarization. A distinct vortex structure can be observed.
Using atomic electron tomography, the team completely measured the positions of cation atoms inside barium titanate (BaTiO3) nanoparticles, a well-known ferroelectric material, in three dimensions. From the precisely determined 3D atomic arrangements, they were able to further calculate the internal three-dimensional polarization distribution at the single-atom level. The analysis of the polarization distribution revealed, for the first time experimentally, that topological polarization orderings including vortices, anti-vortices, skyrmions, and a Bloch point occur inside the 0-dimensional ferroelectrics, as theoretically predicted 20 years ago. Furthermore, it was also found that the number of internal vortices can be controlled depending on their sizes.
Prof. Sergey Prosandeev and Prof. Bellaiche (who proposed with other co-workers the polar vortex ordering theoretically 20 years ago), joined this collaboration and further proved that the vortex distribution results obtained from experiments are consistent with theoretical calculations.
By controlling the number and orientation of these polarization distributions, it is expected that this can be utilized into next-generation high-density memory device that can store more than 10,000 times the amount of information in the same-sized device compared to existing ones.
Dr. Yang, who led the research, explained the significance of the results: “This result suggests that controlling the size and shape of ferroelectrics alone, without needing to tune the substrate or surrounding environmental effects such as epitaxial strain, can manipulate ferroelectric vortices or other topological orderings at the nano-scale. Further research could then be applied to the development of next-generation ultra-high-density memory.”
This research, with Chaehwa Jeong from the Department of Physics at KAIST as the first author, was published online in Nature Communications on May 8th (Title: Revealing the Three-Dimensional Arrangement of Polar Topology in Nanoparticles).
The study was mainly supported by the National Research Foundation of Korea (NRF) Grants funded by the Korean Government (MSIT).
2024.05.31 View 13795 -
Observing Individual Atoms in 3D Nanomaterials and Their Surfaces
Atoms are the basic building blocks for all materials. To tailor functional properties, it is essential to accurately determine their atomic structures. KAIST researchers observed the 3D atomic structure of a nanoparticle at the atom level via neural network-assisted atomic electron tomography.
Using a platinum nanoparticle as a model system, a research team led by Professor Yongsoo Yang demonstrated that an atomicity-based deep learning approach can reliably identify the 3D surface atomic structure with a precision of 15 picometers (only about 1/3 of a hydrogen atom’s radius). The atomic displacement, strain, and facet analysis revealed that the surface atomic structure and strain are related to both the shape of the nanoparticle and the particle-substrate interface.
Combined with quantum mechanical calculations such as density functional theory, the ability to precisely identify surface atomic structure will serve as a powerful key for understanding catalytic performance and oxidation effect.
“We solved the problem of determining the 3D surface atomic structure of nanomaterials in a reliable manner. It has been difficult to accurately measure the surface atomic structures due to the ‘missing wedge problem’ in electron tomography, which arises from geometrical limitations, allowing only part of a full tomographic angular range to be measured. We resolved the problem using a deep learning-based approach,” explained Professor Yang.
The missing wedge problem results in elongation and ringing artifacts, negatively affecting the accuracy of the atomic structure determined from the tomogram, especially for identifying the surface structures. The missing wedge problem has been the main roadblock for the precise determination of the 3D surface atomic structures of nanomaterials.
The team used atomic electron tomography (AET), which is basically a very high-resolution CT scan for nanomaterials using transmission electron microscopes. AET allows individual atom level 3D atomic structural determination.
“The main idea behind this deep learning-based approach is atomicity—the fact that all matter is composed of atoms. This means that true atomic resolution electron tomogram should only contain sharp 3D atomic potentials convolved with the electron beam profile,” said Professor Yang.
“A deep neural network can be trained using simulated tomograms that suffer from missing wedges as inputs, and the ground truth 3D atomic volumes as targets. The trained deep learning network effectively augments the imperfect tomograms and removes the artifacts resulting from the missing wedge problem.”
The precision of 3D atomic structure can be enhanced by nearly 70% by applying the deep learning-based augmentation. The accuracy of surface atom identification was also significantly improved.
Structure-property relationships of functional nanomaterials, especially the ones that strongly depend on the surface structures, such as catalytic properties for fuel-cell applications, can now be revealed at one of the most fundamental scales: the atomic scale.
Professor Yang concluded, “We would like to fully map out the 3D atomic structure with higher precision and better elemental specificity. And not being limited to atomic structures, we aim to measure the physical, chemical, and functional properties of nanomaterials at the 3D atomic scale by further advancing electron tomography techniques.”
This research, reported at Nature Communications, was funded by the National Research Foundation of Korea and the KAIST Global Singularity Research M3I3 Project.
-Publication
Juhyeok Lee, Chaehwa Jeong & Yongsoo Yang
“Single-atom level determination of 3-dimensional surface atomic structure via neural network-assisted atomic electron tomography”
Nature Communications
-Profile
Professor Yongsoo Yang
Department of Physics
Multi-Dimensional Atomic Imaging Lab (MDAIL)
http://mdail.kaist.ac.kr
KAIST
2021.05.12 View 19976