A New Principle Distinguishes Thermal Operations in Quantum Systems

Abstract Thermal operations are quantum channels that play a central role in deriving thermodynamic limitations in quantum systems. However, they were originally defined by implementation procedures rather than by fundamental principles. Alternative models of thermal processes have been proposed, but they obscure the resources required for implementation. Here, we identify the universal principle that separates these models. We show that thermal operations are uniquely characterized by a purely quantum information-theoretic property, which we call informational equilibrium. This formulation does not assume a particular Hamiltonian or even a notion of temperature: a channel is a thermal operation precisely when it leaves its environment invariant whenever the system is prepared in equilibrium. In this way, the usual energetic description of equilibrium (i.e., Gibbs states) emerges as a representation of an underlying informational constraint. Extending this framework, we note that enforcing environment invariance for all system inputs leads to catalytic channels, which idealize perfectly reusable heat baths. This informational equilibrium perspective culminates in a hierarchy for doubly-stochastic quantum channels, with strict separations between the classes. Our refined hierarchy sharpens the failure of the Birkhoff–von Neumann theorem in the quantum regime, and highlights the richer structural landscape of quantum channels compared to classical stochastic processes that cannot be captured by state-convertibility relations. A quantum process can preserve thermal equilibrium without necessarily being achievable through interactions with a heat bath alone. Professor Seok Hyung Lie of the Department of Physics at UNIST, working with colleagues in Singapore and Germany, has identified a fundamental principle that explains this distiction. The study introduces informational equilibrium, a condition that characterizes thermal operations without requiring temperature or energy levels to be specified in advance. Thermal operations describe processes in which a quantum system exchanges energy with an environment in thermal equilibrium while conserving their combined energy. They belong to a broader class known as Gibbs-preserving maps, which leave a system unchanged when it is already in thermal equilibrium. Yet not every Gibbs-preserving map can be implemented as a thermal operation. The researchers found that the distinction depends on whether the environment can remain unchanged. They proved that a quantum process is a thermal operation precisely when it can be implemented without altering the environment whenever the system starts in equilibrium. This condition, termed informational equilibrium, provides a mathematical characterization of thermal operations independent of their conventional energy-based definition. For Gibbs-preserving maps that fall outside this class, preserving the system's equilibrium state necessarily comes at a cost: the environment must change. Such processes therefore require resources beyond those available through thermal operations alone. The framework also clarifies the limits of **quantum catalysis**, in which an auxiliary quantum system enables a transformation while returning to its original state. The researchers showed that a catalyst cannot extend the range of thermal operations if it must be restored exactly even when the system's initial state contains small errors. Any catalytic advantage beyond those limits must therefore depend on sensitivity to those errors. “Our study shows that thermal operations can be understood through an informational condition, without specifying temperature or the energy levels of individual quantum states,” said Professor Lie. “This distinction could help establish the resources needed to operate quantum devices and assess the performance limits of quantum heat engines.” Published in Physical Review Letters on July 13, the study also involved Professor Nelly H. Y. Ng of Nanyang Technological University in Singapore, researcher Jeongrak Son of Freie Universität Berlin, Professor Henrik Wilming of Leibniz University Hannover, and independent researcher Paul Boes. The research was supported by the Ministry of Science and ICT (MSIT), the Institute of Information & Communications Technology Planning & Evaluation (IITP), and the National Research Foundation of Korea (NRF). Journal Reference Seok Hyung Lie, Jeongrak Son, Paul Boes, et al., "Thermal Operations from Informational Equilibrium," Phys. Rev. Lett., (2026).

2026.10.15

Molecular Design Offers Greater Control Over Semiconductor Patterning

Abstract Hybrid tin oxo clusters (SnOCs) have emerged as promising molecular resists for electron-beam (EB) and extreme-ultraviolet (EUV) lithography; However, their patterning mechanisms remain poorly understood due to the coupled effects of ligand electronics, cluster nuclearity, and electron-induced fragmentation. In this study, we demonstrate a ligand-mediated nuclearity tuning strategy that enables the selective synthesis of Sn6- and Sn3-based clusters using electron-donating carboxylate and electron-withdrawing diphenylphosphinate ligands, respectively. Density functional theory calculations, electrostatic potential analysis, and bond dissociation energy evaluations show that ligand electronic properties dictate the preferred electron-driven cleavage pathways. Sn6 clusters predominantly undergo Sn–alkyl bond scission, whereas aromatic phosphinate ligands in Sn3 clusters stabilize the framework and suppress ligand fragmentation. These trends are supported by As a result, Sn6 clusters exhibit high EB sensitivity (D0 = 285 µC cm−2) and sub-30 nm resolution, while Sn3 clusters deliver ultra-low line-edge roughness (1.02 nm) and excellent pattern fidelity. This work identifies nuclearity as a chemically tunable parameter for rational molecular-resist design. The organic molecules surrounding tin atoms can determine how a photoresist responds to electron exposure—and how precisely it reproduces a semiconductor pattern. A research team, led by Professor Ji-Hyun Jang of the School of Energy and Chemical Engineering at UNIST found that changing these molecules alters the structure and patterning behavior of tin-based resists being explored for next-generation extreme-ultraviolet (EUV) lithography. Tin-based photoresists have attracted interest for EUV lithography because they absorb EUV radiation efficiently and can withstand the etching processes used to transfer circuit patterns onto semiconductor wafers. Yet the chemical reactions that govern their sensitivity and pattern quality have remained difficult to understand. To investigate these reactions, the researchers compared two ligands with contrasting electronic properties. An electron-donating ligand derived from adamantane-1-carboxylic acid produced clusters containing six tin atoms (Sn6), while an electron-withdrawing diphenylphosphinate ligand produced clusters containing three (Sn3). The difference in molecular structure translated into distinct patterning behavior. The Sn6-based resist responded at a lower electron-beam dose, producing patterns with a designed line width of 20 nanometers. The Sn3-based resist required a higher dose but produced smoother edges, with a line-edge roughness of just 1.02 nanometers in 50-nanometer-wide lines. The team then examined how the clusters reacted to electron exposure. In Sn6, tin–alkyl bonds broke readily, promoting crosslinking between organic and inorganic components. This reaction helped explain the material's higher sensitivity. Sn3 behaved differently. Its aromatic phosphinate ligands resisted fragmentation and helped preserve the tin–oxygen–phosphorus framework, limiting the chemical changes that can disrupt pattern formation. This stability contributed to greater pattern fidelity and smoother edges. Computational analysis and chemical measurements supported these findings, linking the ligands' electronic properties to the different reactions observed in the two materials. “Our findings show that photoresist sensitivity and pattern precision depend on ligand chemistry as well as tin content,” said Professor Jang. “This could help guide the design of EUV photoresists for different manufacturing requirements.” Published online in Small on August 25, the study was also selected for the journal's back cover. It was supported by the National Research Foundation of Korea (NRF), the Institute of Information & Communications Technology Planning & Evaluation (IITP), the 2026 Research Fund of UNIST, and the InnoCORE program of the Ministry of Science and ICT (MSIT). Journal Reference Seong-Ji Ha, Jin Young Lee, Bo Kyu Kwon, et al., "Ligand-Mediated Nuclearity Tuning of Tin Oxo Clusters for Advanced Lithography," (2026).

2026.10.14

Vacuum-Deposited Passivation Layer Boosts Perovskite Solar Cell Performance

Abstract Controlling the top perovskite/electron-transport-layer interface is essential, as it dominates the loss pathways limiting efficiency and reproducibility in inverted perovskite solar cells. Through a systematic comparative study of organic ammonium passivators, we identify vacuum-deposited piperazinium diiodide (PDI-Vac) as an effective route for regulating the perovskite/C60 interfacial organization, enabled by its favorable kinetic and thermodynamic volatility. This approach forms a conformal and homogeneous passivation layer, which significantly reduces trap-assisted non-radiative recombination losses, as evidenced by enhanced photoluminescence, increased photoluminescence quantum yield, prolonged carrier lifetimes, and reduced quasi-Fermi level splitting and associated energy losses. Consequently, PDI-Vac devices achieve a power conversion efficiency (PCE) of 26.05% with markedly improved device-to-device consistency. The benefits persist in wide-bandgap (1.68 eV) perovskites and extend to monolithic perovskite/silicon tandem cells, delivering a PCE of 33.37% (certified 33.09%). The PDI-Vac strategy can serve as a key contributor to a scalable pathway toward commercial viability, delivering reproducible, state-of-the-art performance in perovskite-based photovoltaics. How passivation molecules are arranged at the surface of a perovskite solar cell (PSC) can affect both energy loss and device-to-device consistency. Researchers at UNIST have found that molecular volatility can be used to control this process through vacuum deposition, producing a more uniform interface and more consistent device performance. A research team, led by Professors Changduk Yang of the School of Energy and Chemical Engineering and Kyoung Jin Choi of the Department of Materials Science and Engineering identified piperazinium diiodide (PDI) as particularly well suited to vacuum deposition. The material forms a thin, uniform passivation layer at the perovskite/C60 interface, suppressing defects that would otherwise contribute to energy loss. Passivation materials are typically applied from solution, but uneven molecular adsorption and assembly can create variations across the interface and increase non-radiative recombination. Rather than simply testing different passivators for their ability to suppress defects, the researchers examined molecular volatility as a design parameter for vacuum deposition. After comparing six organic ammonium passivators, the team found that PDI offered a favorable balance of thermal stability and volatility. Deposited at 0.01 nanometers per second, PDI formed a conformal, homogeneous layer at the perovskite/C60 interface, reducing trap-assisted non-radiative recombination. Inverted PSCs using vacuum-deposited PDI reached a power conversion efficiency of 26.05%, compared with 25.43% for cells treated with PDI from solution. Tests across 12 devices also showed less variation in efficiency, indicating improved device consistency. The strategy also proved effective in wide-bandgap perovskites used for perovskite/silicon tandem solar cells. The tandem devices reached 33.37% efficiency, with a certified efficiency of 33.09%. Stability improved as well. Under continuous illumination at 65 °C, cells treated with vacuum-deposited PDI retained 90% of their initial efficiency for 1,005 hours—about 55% longer than cells treated with PDI from solution. “By treating molecular volatility as a design parameter, we were able to precisely control the deposition process and form a uniform passivation layer on the perovskite surface,” said Jeewon Park, a co-first author of the study. “This approach improves efficiency and stability while also reducing performance variation between devices,” said Professor Yang. “Achieving a certified efficiency above 33% in a perovskite/silicon tandem cell demonstrates its potential for producing high-performance solar cells with greater consistency.” Jeewon Park and Jieun Kim of the Department of Energy and Chemical Engineering and Young Im Noh of the Department of Materials Science and Engineering served as co-first authors of the study. Their findings were published in Energy & Environmental Science on September 1, 2026. The research was supported by the National Research Foundation of Korea (NRF), the Korea Institute of Energy Technology Evaluation and Planning (KETEP), and the Ministry of Climate, Energy and Environment (MCEE). Journal Reference Jeewon Park, Jieun Kim, Young Im Noh, et al ., “Volatility-governed interfacial organization for low-loss perovskite tandems,” Energy Environ. Sci. , (2026).

2026.10.13

Organ Environment Shapes How Metastatic Cancer Cells Adapt

Abstract The extracellular matrix (ECM) plays a central role in regulating tumor progression and metastatic colonization by providing biochemical and mechanical signals that shape cancer cell fate. However, most organoid culture systems rely on basement membrane extracts that fail to reproduce the tissue-specific extracellular environments encountered during metastasis. Here, we develop tissue-derived decellularized matrix hydrogels to reconstruct organ-specific microenvironments and investigate epigenetic adaptation to ECM cues during metastatic colonization. Patient-derived colorectal cancer organoids cultured in colon-derived matrices exhibited enhanced maintenance of stem-like phenotypes and colon-specific chromatin accessibility landscapes compared with cultures grown in basement membrane extracts, demonstrating improved physiological relevance for primary tumor modeling. When exposed to matrices derived from secondary organs, the organoids showed distinct growth phenotypes accompanied by rapid, tissue-dependent chromatin accessibility remodeling, indicating that ECM composition alone can reshape regulatory programs governing metastatic adaptation. Notably, liver-derived matrices selectively activated hepatocyte nuclear factor 4 alpha (HNF4A)–associated transcriptional networks and created a context-specific dependence on c-MET signaling for survival. Functional perturbation of HNF4A or c-MET signaling confirmed that both are required for organoid formation specifically within the liver matrix environment. Together, these findings establish tissue-derived matrix hydrogels as instructive bioactive materials that actively regulate cancer cell epigenetic states and reveal microenvironment-specific therapeutic vulnerabilities during early metastatic colonization. Colorectal cancer cells can behave differently depending on the organ environment they encounter. New research shows that the composition of the extracellular matrix (ECMs) can reshape gene regulation as cancer cells adapt to a new site, revealing a distinct vulnerability in the liver environment. A research team, led by Pofessors Seung Woo Cho and Tae-Eun Park of the Department of Biomedical Engineering at UNIST has developed tissue-derived hydrogels that recreate key features of the ECM surrounding cells in different organs. Using patient-derived colorectal cancer organoids, the researchers examined how these environments influenced cancer cells during the early stages of metastatic colonization. The team created hydrogels from decellularized colon, liver, lung, and brain tissue, preserving the distinct ECM composition of each organ. Organoids grown in colon-derived ECM better maintained stem-like characteristics and colorectal cancer–specific chromatin states than those cultured in conventional basement membrane extracts. When exposed to ECM from other organs, however, the cancer cells undergo rapid, tissue-dependent changes in chromatin accessibility, indicating that the surrounding matrix can reshape the regulatory programs involved in metastatic adaptation. Liver-derived ECM produced a particularly distinct response. It activated regulatory programs associated with HNF4A, a transcription factor involved in gene regulation. Suppressing HNF4A significantly impaired organoid formation in the liver-derived matrix, showing that this regulatory program had become important to the cells in that environment. The liver environment also revealed a vulnerability involving c-MET, a receptor involved in cell growth and survival. Blocking c-MET sharply reduced cancer cell survival in liver-derived ECM, while having little effect under conventional culture conditions. The finding suggests that the environment cancer cells encounter after spreading to another organ can influence not only how they adapt, but also which signaling pathways they come to depend on. “Early adaptation to a new organ is a critical stage of metastasis, but it is difficult to study in patients because metastatic tumors are usually detected only after they have become established,” the researchers said. “Our platform allows us to examine this early transition in an organ-specific environment and identify vulnerabilities that emerge as cancer cells adapt.” The team plans to expand the platform to other cancer types and incorporate immune cells to investigate how tumor–immune interactions differ across organ environments and how those differences affect responses to immunotherapy. The study, with Heejeong Yoon and Dayoung Kim of UNIST as co-first authors, was published online in Bioactive Materials on August 26, 2026. The work was supported by the National Research Foundation of Korea (NRF), and the Korean ARPA-H Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health and Welfare (MOHW). Additional support was provided by the Ministry of Education, Ulsan Metropolitan City, the Institute for Basic Science (IBS), and UNIST. Journal Reference Heejeong Yoon, Dayoung Kim, Hye-Jin Jeong, et al., “Tissue-derived extracellular matrix hydrogels instruct epigenetic adaptation in metastatic colonization,” Bioact. Mater. , (2026).

2026.10.12

UniSim-SLAM Keeps Robots on Track in Unfamiliar Spaces

Abstract Recent geometric foundation models enable feed-forward inference for SLAM, but their predictions are strongly dependent on the input view set, which leads to geometric inconsistencies and trajectory drift when results are chained over long sequences. Online deployment further exposes a trade-off between the low latency of two-view tracking and the constraint richness of multi-view inference. We introduce UniSim-SLAM, an integrated system that runs lightweight two-view keyframe tracking in the frontend and performs periodic multi-view submap refinement in the backend. To combine predictions defined in heterogeneous local coordinates with inconsistent scales, we formulate a unified multi-level factor graph on Sim(3) that jointly optimizes global keyframe poses and submap poses. The graph integrates temporal view-to-view odometry edges, view-to-submap bridge edges with depth-statistics scale anchoring, and submap-to-submap tie and scale constraints to enforce consistent similarity relations across submaps. Experiments on TUM RGB-D and 7-Scenes show that UniSim-SLAM achieves state-of-the-art accuracy in the uncalibrated setting, reducing trajectory error by 38.5% on TUM RGB-D and 45.9% on 7-Scenes compared to prior best results. Robots navigating unfamiliar environments must track their position while continuously mapping the space around them. AI has made both tasks faster, but a persistent trade-off remains: using more camera views can improve accuracy, but requires more processing time. Professor Kyungdon Joo and his research team from the Graduate School of Artificial Intelligence at UNIST have developed UniSim-SLAM, a system that combines rapid camera tracking with periodic multi-view refinement. The approach allows robots to update their position quickly while correcting errors that would otherwise accumulate as they move. Known as simultaneous localization and mapping (SLAM), the underlying technology enables robots and other autonomous systems to determine their location while reconstructing their surroundings. Recent AI-based approaches can infer 3D geometry directly from camera images, but small differences between successive estimates can compound over long sequences, gradually pushing the reconstructed path away from the robot's actual trajectory. UniSim-SLAM tackles the problem by giving fast tracking and detailed reconstruction different roles. It compares two camera views at a time to track movement, then periodically brings multiple views together to refine the local 3D map. A unified optimization framework aligns estimates produced at different scales and orientations, keeping the reconstructed environment and camera trajectory consistent over time. On two standard indoor benchmarks achieved, UniSim-SLAM state-of-the-art results in the uncalibrated setting. With the previous best Compared methods, it reduced trajectory error by 38.5% on TUM RGB-D and 45.9% on 7-Scenes. The system was also considerably faster than a multi-view alternative evaluated in the study. VGGT-SLAM required 3.41 seconds to produce a result, compared with 0.197 seconds for UniSim-SLAM. The new approach also improved the accuracy of 3D reconstruction. “UniSim-SLAM brings fast tracking and accurate multi-view analysis into the same system,” said Professor Joo. “This could benefit applications that need reliable localization and real-time 3D reconstruction, from robotics and autonomous driving to augmented reality.” The study, with Inha Lee of UNIST as first author, was presented at the European Conference on Computer Vision (ECCV 2026), held September 8–12 in Malmö, Sweden. The work was supported by the Ministry of Science and ICT (MSIT) through the Institute of Information & Communications Technology Planning & Evaluation (IITP), including the Artificial Intelligence Graduate School Program at UNIST and programs in collaborative AI and spatial immersive technologies, as well as the InnoCORE program. Journal Reference Inha Lee, Dongjae Jeong, Junhee Lee, and Kyungdon Joo, "UniSim-SLAM: Feed-Forward SLAM with Unified Sim(3) Optimization," ECCV'26 (2026).

2026.10.08

A More Efficient Energy Strategy for Power-Hungry AI Data Centers

AI data centers consume enormous amounts of electricity—not only to run thousands of GPUs, but also to keep them cool. A new study suggests that recovering waste heat while integrating on-site power generation and carbon capture could substantially reduce both their cost and carbon footprint. Led by Professor Hankwon Lim of the Graduate School of Carbon Neutrality at UNIST, a research team evaluated a data-center energy system combining solid oxide fuel cells (SOFCs), carbon capture, and organic Rankine cycle (ORC) cooling. Compared with a conventional grid-powered, air-cooled system, the proposed approach could reduce total system costs by approximately 48% and lifecycle greenhouse gas emissions by up to 72%. The system puts waste heat to work in two ways. High-temperature heat from the fuel cells supplies energy for the carbon-capture process, while heat removed from servers is recovered through an ORC to generate additional electricity. This reduces both the energy required for cooling and the amount of electricity the data center must draw from the grid. The researchers modeled a data center with 20,000 GPUs and compared eight combinations of power supply, cooling, and carbon capture using energy prices and electricity mixes in Korea, the United States, and the European Union. Across a broad range of cost and emissions priorities, the combination of SOFC power, carbon capture, and ORC cooling consistently ranked highest. ORC cooling performed particularly well across all three regions, reducing annual cooling-related electricity costs by as much as $19 million. Lower electricity demand also cuts annual greenhouse gas emissions by an estimated 80,000 to 150,000 metric tons, potentially avoiding $5 million to $10 million in annual carbon costs under carbon-pricing schemes such as the EU Emissions Trading System. The preferred source of electricity varied by region. Relatively inexpensive natural gas made SOFC generation economically attractive in the United States, while in Korea, where natural gas is more expensive, grid electricity remained more economical when cost was prioritized. “As AI data centers expand, electricity supply, cooling, and carbon emissions need to be considered together,” said Donghyoun Lee from the Graduate School of Carbon Neutrality at UNIST. “Our analysis shows how power generation, waste-heat recovery, and carbon capture can be evaluated as a single energy system.” “The best strategy will depend on local energy prices and the carbon intensity of electricity,” said Professor Lim. “This framework can help identify energy systems that balance cost and emissions for data centers of different sizes and in different regions.” The findings of this study have been published online in Chemical Engineering Journal on August 14, 2026. The research was supported by the National Research Foundation of Korea (NRF) and the Ministry of Science and ICT (MSIT). Journal Reference Donghyoun Lee, JeHyeon Seong, and Hankwon Lim, "An integrated solution for data center power supply, thermal management, and carbon capture evaluated through a multi-criteria decision analysis framework," Chem. Eng. J., (2026).

2026.10.07