Research

A Hidden Phase of In2Se3 Emerges as a Wafer-Scale Logic Semiconductor

[POSTECH-KAIST Team Demonstrates 4-Inch Wafer-Scale κ-Phase In2Se3 Transistors Fabricated at 250 °C] Just as carbon atoms arranged in different ways can form either graphite or diamond—the hardest natural material—semiconductor materials can also exhibit completely different properties depending on their atomic arrangements. Now, a previously overlooked crystal structure of indium selenide (In2Se3) has been brought to life as a promising material for next-generation logic electronics. A research team led by Professor Yong-Young Noh from the Department of Chemical Engineering at POSTECH, including graduate student Jaeyun Lee, in collaboration with Professor Jimin Kwon and Dr. Yongwoo Lee from the School of Electrical Engineering at KAIST, has successfully realized wafer-scale logic transistors based on the rarely explored κ-phase In2Se3. The study has been published in the international journal Nature Communications. In2Se3 has long been considered a promising next-generation semiconductor material because of its atomically thin structure and excellent charge transport properties. However, previously studied crystal phases have shown limitations. Their ability to retain a previous electrical state after voltage removal—known as a “memory effect”—is beneficial for memory applications, but can cause instability in logic circuits that require identical outputs from identical inputs. In addition, although high performance has been demonstrated in small exfoliated flakes, achieving uniform and high-quality In2Se3 films over large wafer areas has remained a major challenge for practical semiconductor manufacturing. The research team turned their attention to the previously underexplored κ-phase In2Se3. Unlike conventional In2Se3 phases, the unique atomic arrangement of the κ-phase suppresses ferroelectric memory effects, making it more suitable for stable logic operation. Using an industrially compatible thermal evaporation process followed by a low-temperature annealing treatment at 250 °C, the researchers successfully fabricated approximately 10 nm-thick κ-In2Se3 films uniformly across an entire 4-inch wafer. The low processing temperature, below 450 °C, also satisfies the thermal requirements for three-dimensional semiconductor integration, allowing additional electronic layers to be built on top of existing silicon circuits without damaging them. The fabricated devices demonstrated outstanding electrical performance and uniformity. A total of 576 transistors (24 × 24 array) distributed across the wafer exhibited highly consistent characteristics regardless of their position. The devices achieved an average electron mobility of 39.3 cm2V−1s−1 and an on/off current ratio exceeding 108, representing a high level of performance among large-area thin-film two-dimensional semiconductor devices. Furthermore, the transistors maintained stable operation under 10 million repeated switching cycles using ultrashort 100 ns electrical pulses, demonstrating both fast switching capability and long-term operational reliability required for logic applications. To further demonstrate its potential as a logic semiconductor, the team integrated κ-In2Se3 with a p-type selenium-alloyed tellurium oxide (Se-TeOx) semiconductor featuring complementary charge transport characteristics. By combining n-type κ-In2Se3 and p-type Se-TeOx devices, the researchers successfully implemented a complementary inverter circuit, a fundamental building block of modern logic electronics. This result provides direct evidence that κ-In2Se3 can serve not only as a memory-related material but also as a stable platform for logic semiconductor applications. This study demonstrates that In2Se3, previously regarded mainly as a memory-oriented material, can be transformed into a wafer-scale logic semiconductor by utilizing an alternative crystal phase with suppressed memory effects. The ability to fabricate uniform large-area films at low temperatures is particularly significant for future three-dimensional integrated electronics, where new semiconductor functions are stacked vertically on existing silicon chips. This work opens a pathway toward smaller, more powerful, and highly integrated semiconductor technologies. Jaeyun Lee, an integrated M.S./Ph.D. student at POSTECH who led the study, stated, “In2Se3 has mainly been investigated as a ‘memory semiconductor’ due to its intrinsic ferroelectric properties. Our work expands its potential by introducing the less-explored κ-phase as a stable material platform for logic semiconductor applications.” This research was supported by the Ministry of Science and ICT, the National Research Foundation of Korea, and the Glocal University 30 Project. ▶️ DOI: https://doi.org/10.1038/s41467-026-72553-y

Put the Spectrometer Away: A Tiny Sensor That Detects Disease with a Single Camera

[POSTECH Researchers Develop Portable, Imaging-Based Metasurface Biosensor for Spectrometer-Free Molecular Detection] What if signs of disease could be detected using nothing more than a tiny chip that fits in the palm of your hand and a camera, without ever visiting a hospital? A scene that once seemed possible only in science fiction has now moved one step closer to reality. A research team led by Professor Junsuk Rho of the Departments of Mechanical Engineering, Chemical Engineering, Electrical Engineering, and the Graduate School of Convergence Science and Technology at POSTECH, together with Dr. Hongyoon Kim of the Department of Mechanical Engineering and Heechang Yun and Sebin Jeong of the Department of Chemical Engineering, has developed a metasurface biosensor capable of precisely detecting biomolecules through imaging without the need for an expensive spectrometer. The research was conducted in collaboration with Professor Hyomin Lee’s group in the Department of Chemical Engineering at POSTECH and Professor Yong-Sang Ryu’s group in the Department of Biomedical Engineering at Korea University. The study was published in Nature Communications. Biosensors are essential tools in healthcare, with applications ranging from disease diagnosis and infectious disease testing to cancer biomarker detection and genetic analysis. Among them, optical biosensors offer a particularly important advantage. They can directly detect biomolecules without fluorescent labels by measuring subtle optical changes that occur when target molecules bind to the sensor surface. The challenge, however, has been the size and complexity of conventional systems. Detecting such minute optical changes typically requires a high-resolution spectrometer together with a broadband light source. Because these instruments are bulky and expensive, it has been difficult to bring high-performance optical biosensing beyond laboratories and hospitals. The POSTECH-led team approached the problem from a different perspective. Instead of measuring a spectrum with a spectrometer, they designed the sensor to convert spectral information into spatial information that can be directly read through imaging. At the heart of the technology is a specially engineered continuous geometric-gradient metasurface. The nanoscale structures that make up the metasurface gradually change in geometry along one direction of the chip, causing different positions to respond to different wavelengths of light. Just as the keys of a piano produce different notes depending on their position, different locations on the metasurface resonate with different wavelengths. When the metasurface is illuminated with a compact single-wavelength laser, a dark line appears in the captured image at the position where the local optical resonance matches the laser wavelength. At this location, light transmission is strongly suppressed. When biomolecules bind to the sensor surface, they slightly alter the surrounding refractive index. This shifts the optical resonance and causes the dark line to move. By tracking this spatial displacement through imaging, the researchers were able to accurately measure molecular binding signals. In other words, the system converts an extremely small spectral shift into a spatial displacement that can be directly measured through imaging, eliminating the need for conventional spectral measurements. Despite its simple imaging-based readout scheme, the sensor demonstrated high performance. Within an area only about 300 micrometers (μm) across, roughly comparable to the width of several human hairs, the device simultaneously achieved a broad sensing range and fine spectral readout resolution of approximately 0.1 nanometers (nm). Experiments involving proteins and DNA further demonstrated that the platform could detect biomolecules at concentrations down to the hundreds-of-picomolar (pM) range. The researchers also showed that the sensor could selectively recognize specific DNA sequences, demonstrating its potential for molecular diagnostics and genetic testing. One of the greatest advantages of this technology is its simplicity. Instead of relying on a bulky, costly spectrometer and a broadband light source, the system requires only a compact single-wavelength laser and an imaging sensor. This approach could pave the way for miniaturizing precision optical biosensing systems traditionally confined to laboratories, into compact and portable diagnostic devices. The researchers expect the platform to have broad potential for point-of-care diagnostics, including infectious disease testing, cancer biomarker detection, genetic analysis, and liquid biopsy. Professor Junsuk Rho of POSTECH said, “This work demonstrates how the high optical sensitivity of metasurfaces can be combined with the simplicity of imaging-based readout to overcome key limitations of conventional spectrometer-based biosensors. The platform has strong potential to evolve into a compact, low-cost biosensing technology for point-of-care diagnostics.” The research was supported by a grant from the Korean ARPA-H Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea, and by a National Research Foundation grant funded by the Ministry of Science and ICT of the Korean government. ▶️ DOI: https://doi.org/10.1038/s41467-026-74749-8

A New Culture Membrane Mimics the Native Intestinal Environment

[A joint research team from POSTECH and CHA University develops a culture membrane that closely mimics native intestinal tissue using components extracted from porcine colon] The human intestine is a soft and flexible tissue. Yet the environments traditionally used to culture intestinal cells in the laboratory are strikingly different from those found in the body. Conventional culture platforms are made of synthetic materials that are far stiffer than native intestinal tissue. Could we instead provide cells with an environment that more closely resembles the intestine? A Korean research team has now developed a new cell culture membrane that recreates not only the tissue-specific components of the intestine but also its soft physical environment. The research team led by Professor Dong Sung Kim of the Department of Mechanical Engineering and Graduate School of Convergence Science and Technology at POSTECH, together with Dr. Hyeonji Kim of the Department of Mechanical Engineering, collaborated with Professor Jaeseung Youn of CHA University School of Medicine, a POSTECH alumnus, and Professor Hyun Jung Kim of Cleveland Clinic to develop a new culture membrane that simultaneously mimics the biochemical composition and physical properties of native intestinal tissue. The study was recently published online in Advanced Healthcare Materials. Intestinal organoids are miniature, stem cell-derived models of the intestine. Because they reproduce diverse cell types and functions of the native intestine, they are widely used to study intestinal diseases and evaluate drug responses. For greater experimental convenience, cells derived from intestinal organoids can be grown as a flat layer on a thin culture membrane. The problem, however, lies in the membrane itself. Conventional synthetic polymer membranes are much stiffer than native intestinal tissue and fail to fully reproduce its tissue-specific extracellular matrix (ECM). In other words, while the cells may represent the intestine, the environment in which they grow does not. This difference goes beyond how the culture surface simply “feels.” Cells respond not only to biochemical signals such as growth factors and proteins but also to physical cues, including the stiffness and structural properties of their surroundings. Because these physical properties can influence cell proliferation, differentiation, and tissue formation, recreating a physiologically relevant intestinal model requires consideration of both the biochemical composition and physical properties of the native tissue. The research team therefore sought to integrate both features into a single culture membrane. Using electrospinning, they first fabricated an ultrathin nanofibrous scaffold composed of fibers far thinner than a human hair. They then incorporated decellularized extracellular matrix (dECM) derived from porcine colon tissue, from which cellular components had been removed while preserving extracellular matrix components. The resulting C-NaDE membrane was designed to combine the structural stability of a nanofibrous scaffold with the tissue-specific biochemical components of colon-derived dECM. Unlike conventional synthetic culture membranes, it also provided a soft physical environment closer to that of native intestinal tissue, allowing cells to experience a microenvironment that more closely resembles their physiological surroundings. The effects were evident at the cellular level. When human colon organoid-derived epithelial cells were cultured on the newly developed membrane, the researchers observed enhanced characteristics associated with intestinal stem cell maintenance and cell proliferation. Genetic markers associated with certain secretory cell lineages, which are relatively difficult to observe under conventional culture conditions, were also increased. In addition, the resulting epithelial layer exhibited morphological characteristics more closely resembling those of native intestinal tissue. The key advance of the study lies in moving beyond conventional approaches that primarily focus on supplementing individual proteins or growth factors. Instead, the researchers recreated both the biochemical composition and physical environment in which intestinal cells naturally reside. By integrating tissue-derived biochemical cues with biophysical cues such as physiologically relevant softness, the new membrane helps bridge the gap between conventional synthetic culture substrates and native intestinal tissue. The technology can also be incorporated into conventional cell culture systems as well as microfluidic organ-on-a-chip platforms. The researchers expect that the membrane could be further developed as a platform for intestinal disease modeling, drug response testing, intestinal barrier assessment, and studies of interactions between the intestinal epithelium and gut microbiota. Professor Dong Sung Kim of POSTECH said, “Intestinal organoid models offer the advantage of reproducing diverse intestinal cell types, but when they are cultured as a monolayer, they typically rely on synthetic substrates whose biochemical and physical properties differ substantially from those of native tissue.” He added, “The significance of this study lies in integrating the tissue-specific components of the intestine and a physiologically relevant physical environment within a single culture membrane.” Professor Jaeseung Youn of CHA University, the first author of the study, said, “Cells respond not only to biochemical factors such as growth factors and proteins but also to physical properties of their surroundings, including stiffness.” He explained, “By simultaneously recreating the biochemical and biophysical microenvironment of intestinal tissue, our approach provides a foundation for developing intestinal models that more closely resemble the in vivo environment.” He added, “Because the membrane can be incorporated into existing culture systems and microfluidic chips, we expect it to serve as a next-generation platform for intestinal disease research, drug evaluation, and studies of interactions between the gut microbiota and intestinal tissue.” This research was supported by the Mid-Career Researcher Program, the National Research Laboratory (NRL) Program, and the Sejong Science Fellowship Program funded by the Ministry of Science and ICT and the National Research Foundation of Korea, as well as the Korea Health Technology R&D Project funded by the Ministry of Health and Welfare and the Korea Health Industry Development Institute. ▶️ DOI: https://doi.org/10.1002/adhm.202501510

“Hidden Order in Disorder” Makes Nanodevices Easier to Design

[POSTECH researchers develop a new analytical framework that reveals hidden periodicity in complex metasurfaces] Augmented reality (AR) glasses, lenses thinner than a human hair, and holograms floating above your fingertips may sound like technologies from science fiction. At the heart of these emerging technologies, however, lies a nanoscale optical device known as a “metasurface.” Made of nanostructures smaller than the wavelength of light, metasurfaces can precisely control the direction, color, and other properties of light. Despite their remarkable capabilities, their complex structures have long posed a major challenge to researchers. A research team led by Professor Junsuk Rho and Dr. Seokwoo Kim of POSTECH has discovered hidden repeating patterns within seemingly disordered nanostructures, opening a new route toward faster and more accurate analysis and design of metasurfaces. The study was published in the international journal Nature Communications. To achieve a desired optical function, metasurfaces are designed by gradually varying the size, shape, or orientation of individual nanostructures. As these structures become increasingly different from one another, however, the periodicity of the overall surface is lost, making it difficult to calculate how the device interacts with light. If a structure is periodic, researchers can analyze only a small portion of it and predict the behavior of the entire device. In contrast, an aperiodic structure may require an enormous number of individual nanostructures to be calculated separately. A metasurface measuring several centimeters across can contain more than one billion nanostructures, making rigorous computer simulations extremely time-consuming. Conventional approaches have therefore often relied either on simplified assumptions or on computationally demanding large-scale simulations. The research team focused on the idea that even a structure that appears highly complex may contain an underlying repeating order. They found a clue in the Moiré pattern1, a large-scale interference pattern that appears when two regular patterns with slightly different alignments or spacings are overlaid. A familiar example can be seen when two layers of mosquito netting or thin curtains are placed on top of each other. The researchers interpreted the basic lattice of a metasurface and the additional spatial variation introduced to redirect light as two overlapping patterns. They then mathematically demonstrated that when the orientation and spacing of these patterns satisfy specific conditions, a large-scale repeating pattern emerges even within a structure that initially appears aperiodic. In simple terms, the team identified a “small repeating unit” that can represent an enormous array of seemingly different nanostructures. As a result, it is no longer necessary to calculate the entire large-area metasurface directly. Even for a device containing more than one billion nanostructures, its overall properties can be analyzed by calculating only a small unit consisting of several tens of nanostructures. Based on this framework, the research team fabricated actual metasurface devices and confirmed that light was controlled in the predicted directions. The method also enabled the researchers to clearly distinguish light traveling in the intended direction from unwanted diffraction and scattering channels, allowing metasurface performance to be analyzed rapidly and accurately. Because the underlying principle is based on wave physics, the framework is not limited to light and can potentially be extended to systems involving sound, electrons, and mechanical vibrations. The approach is therefore expected to provide a powerful foundation for faster and more precise development of next-generation optical devices, including AR and VR components and high-performance metalenses. Professor Junsuk Rho said, “By uncovering hidden order within structures that appear disordered, this framework provides a powerful tool for accelerating the development of next-generation optical devices.” This work was supported by the Ministry of Science and ICT of Korea and the National Research Foundation of Korea through research programs including the Mid-Career Researcher Program and the Sejong Science Fellowship. ▶️ DOI: https://doi.org/10.1038/s41467-026-74150-5 1. Moiré pattern: A Moiré pattern is a larger-scale interference pattern that emerges when two regular patterns with slightly different orientations or spacings are overlaid.

Sound Only You Can Hear Is Becoming a Reality

[POSTECH research team develops a highly directional speaker by integrating metamaterials with a single ultrasonic transducer] The kind of technology that can deliver sound to a single listener—like something out of a spy film—may soon become a reality. Researchers at Pohang University of Science and Technology have developed a speaker system that can direct speech to a specific person. The study, published in Nature Communications on May 22, 2026 under the title “Dual-domain metamaterials co-integrated with a compact ultrasonic transducer for highly directional audio generation”, is available at https://doi.org/10.1038/s41467-026-73604-0. Light and sound behave differently. Light from a flashlight or laser travels in a straight line, whereas sound readily spreads in all directions—that is why voices can be heard over a wall. The question, “Could sound, like light, be sent only where we want it?” led to the team’s highly directional speaker, MiPAL1). By modulating ultrasound with audible content, the system can deliver sound only within a targeted area. A representative technology is the parametric array loudspeaker (PAL2)), which uses the nonlinear interaction of two ultrasonic waves in air to generate and concentrate audible sound in a specific direction. Compared with conventional speakers, PALs can confine sound to a much narrower region. Traditional systems, however, require tens to hundreds of ultrasonic emitters to be precisely arranged and individually controlled, making them expensive and structurally complex. More recent PAL designs have simplified the hardware, but the vibrating diaphragm can also excite unwanted vibration modes, allowing sound to leak sideways and undermining directivity. The team addressed this problem by integrating two types of metamaterials3) into a single ultrasonic transducer. An acoustic metasurface was mounted at the front. Acting like a lens for sound, it reshapes the irregular ultrasonic wavefronts radiated by the diaphragm into a straight, narrow beam, much like a laser. Elastic meta-units were placed at the rear. Acting more like a sound barrier, they suppress unwanted vibrations at selected frequencies and keep sound from leaking to the sides. In effect, the design uses a double safeguard: it focuses the sound at the front and blocks parasitic noise at the back. As a result, the single-element device directed sound over a frequency range of 500 Hz to 10 kHz—more than four octaves4), covering a large portion of the audible spectrum. The researchers also conducted an experiment simulating aircraft seating, where different announcements or music might need to reach individual passengers without disturbing their neighbors. After playing real broadcast and music signals and measuring sound intensity at different angles, they observed a clear improvement after the elastic meta-units were added. Another key advantage is ease of fabrication and adaptation. The metamaterial structures are modular and can be produced using 3D printing, allowing the device to be readily redesigned for different seat spacing, angles and other operating conditions. Once commercialized, the technology could let passengers on planes or trains listen to personalized announcements or content without disturbing the person next to them. In museums and exhibition halls, explanations could be heard only by visitors standing in front of a particular display. The “sound only I can hear” once seen only in films may soon become part of everyday life. “By co-designing the metamaterials and the ultrasonic device as a single system, we overcame the limitations of cost, complexity and design flexibility at the same time,” said Professor Junsuk Rho. “Research on acoustic and elastic metamaterials has largely remained at the stage of validating principles and demonstrating concepts. This study is the first to integrate metamaterials into functional device and demonstrate their potential for practical use.” The study was conducted by a POSTECH research team comprising Professor Junsuk Rho of the Departments of Mechanical Engineering, Chemical Engineering, and Electrical Engineering and the Graduate School of Convergence Science and Technology; Professor Wonkyu Moon of the Department of Mechanical Engineering; Dr. Woongji Kim; and doctoral student Beomseok Oh. It was supported by the POSCO-POSTECH-RIST Convergence Research Center program and the Researcher Program of the Ministry of Science and ICT. ▶️ DOI: https://doi.org/10.1038/s41467-026-73604-0 1. MiPAL: Metamaterials-integrated Parametric Array Loudspeaker. MiPAL integrates acoustic and elastic metamaterials with a compact ultrasonic transducer to produce highly directional audible sound. 2. Parametric Array Loudspeaker, PAL: A speaker that uses the nonlinear properties of air. When two intense ultrasonic waves with nearby frequencies are radiated together, an audible wave is generated at their difference frequency. Unlike a conventional speaker, a PAL can form a highly directional sound source by concentrating sound into a narrow beam. 3. Metamaterial: An engineered material made from periodically arranged artificial structures not found in nature. Its geometry enables control of waves—such as light, sound and vibration—in ways that ordinary materials cannot. This study uses both acoustic metamaterials, which control sound, and elastic metamaterials, which control structural vibration. 4. Octave: An interval in pitch in which frequency doubles. The device’s operating range of 500 Hz to 10 kHz spans more than four octaves and covers a large portion of the frequencies people hear in everyday speech and audio.

Develops a 'Breathing Lung' Inspired by Soap Bubbles

[Professor Sungjune Jung's team at POSTECH reproduces alveolar movement in an artificial lung model that can operates stably over 240,000 breaths] Almost everyone has blown a soap bubble through a straw at least once. Inspired by this thin soap bubble film, an “artificial lung that breathes” like a real lung has been developed. A research team led by Professor Sungjune Jung of the Department of Materials Science and Engineering at POSTECH has developed an ultrathin artificial lung that reproduces the movement of the alveoli, the air sacs inside the lung, while also making it possible to observe the responses of lung cells. This artificial lung operated stably through approximately 240,000 breaths and also reproduced the lung’s response to influenza virus. The study was published on 30 July 2026 (local time) in Advanced Materials, one of the most prestigious journals in materials science The human lung moves tens of thousands of times a day. When we inhale, the lung expands; when we exhale, it contracts again. The alveoli, the small air sacs inside the lung, repeat this motion as well. The motion of the alveoli does not stop at simply exchanging air. It affects the growth and function of lung cells, and it serves as an important signal in the process by which inflammation and disease progress. To uncover the causes of lung disease and to confirm the efficacy of therapeutics, this motion must also be reproduced in the laboratory. However, it has been difficult to implement the actual motion of the lung with existing lung cell culture techniques, because cells are for the most part grown on flat, motionless surfaces. More recently, “lung-on-a-chip” technology, which implements the structure and motion of the lung within a chip, has also been developed, but because most such chips use a synthetic material (PDMS1)), there have been limits to how well they reproduce the characteristics of real lung tissue. Soft hydrogels, on the other hand, have had the problem of tearing easily during the fabrication of thin membranes. The research team found the answer in the “soap bubble.” Although a soap bubble is a thin film that the liquid forms by itself, it does not burst easily and moves freely. The team applied this to principle to a composite hydrogel made by mixing ] polymers and monomers. By dipping a mold of the desired shape into a hydrogel solution and then withdrawing it, they formed an extremely thin membrane within the mold. The ultrathin membrane produced in this way was soft and stretchable like the alveoli inside a native lung, yet strong enough to withstand hundreds of thousands of repeated motions. The team then also applied the principle by which humans actually breathe. When a person inhales, the diaphragm moves downward, the pressure inside the lung falls, and outside air flows in. The team developed a “breathing actuation system” that repeatedly generates this same pressure change beneath the ultrathin membrane. As a result, the ultrathin membrane stretched and contracted like a real lung, and operated stably through approximately 240,000 breaths. The interior of the artificial lung was likewise built to resemble a real lung. Using 3D bioprinting technology, a three-layer structure similar to the alveoli was implemented by stacking a “vascular cell layer,” a “basement membrane layer,” and an “epithelial cell layer” in sequence. In the artificial lung produced in this way, the motion of breathing was transmitted to the cells, and even the response to viral infection appeared similar to that of a real lung. In particular, influenza virus infection experiments newly revealed that the inflammatory response and the antiviral response of the lung differ according to respiratory motion. This study carries considerable significance in that it has opened the way to examining, in a dynamic environment very similar to the human lung, drug responses that are difficult to confirm through simple cell culture or animal testing alone. Professor Sungjune Jung said, “The ‘breathing artificial lung’ allows the rate and depth of breathing to be freely adjusted, so everything from a normal lung to a range of disease states can be reproduced in the laboratory,” adding, “This study will change the landscape of new drug development and lung disease research.” This research was conducted with the support of the Mid-Career Researcher Program of the National Research Foundation of Korea and the Bio Industrial Technology Development Program of the Ministry of Trade, Industry and Energy. ▶️ DOI: https://doi.org/10.1002/adma.74372 1. PDMS: polydimethylsiloxane, a soft silicone-based polymer material.

“Thinner Lenses, Brighter Colors”: Metalens Research Clears Two Hurdles for AR·VR Glasses

[From precision fabrication to scalable manufacturing, POSTECH develop technologies for commercializing metalenses published consecutively in Nature Communications] AR and VR glasses once seen only in science fiction may soon be realized not as bulky stacks of lenses, but as a single eyeglass-like optical element. That future is now coming closer to reality. A research team at Pohang University of Science and Technology has consecutively developed technologies for producing sharp full-color images using metalenses and published the findings in Nature Communications. The achievement is drawing attention because it addresses two major challenges for commercializing metalenses: high optical performance and scalable manufacturing. A metalens is an ultrathin lens made of densely arranged nanoscale structures that control the path of light as intended. Unlike conventional optics, which often require multiple stacked glass lenses, a metalens can perform similar functions in a single flat optical element, making it a key component for lightweight and compact devices such as AR and VR glasses. The challenge lies in achieving achromatic performance, which means focusing red, green, and blue light at the same point without color blur. Realizing this performance using low-refractive-index materials, which are inexpensive and suitable for mass production, has been extremely challenging. The research team solved this issue by controlling the height of nanoscale pillars known as meta-atoms, which form the metalens. While previous approaches mainly controlled light by adjusting the lateral width of meta-atoms, the team added height as a new design parameter. Just as buildings with the same footprint can create different cityscapes by varying their number of floors, meta-atoms built to different heights can precisely focus RGB light at the desired focal point. In the first study, the team directly fabricated complex three-dimensional nanopillars using two-photon lithography, a high-precision 3D printing technique. They built a database of various nanopillar designs and used an inverse-design approach to automatically select the most suitable structure for each position of the metalens. As a result, they realized a full-color achromatic metalens using relatively low-cost, low-refractive-index materials and confirmed sharp color imaging in experiments that integrated the metalens with an OLED display. The second study focused on mass production. The team fabricated a height-encoded nano-template using grayscale electron-beam lithography and applied it to nanoimprint lithography, a stamping-like replication process. Conventional nanoimprint methods are generally suited for repeatedly printing patterns with the same height, but this new technology enables structures with different heights to be replicated at once, improving both optical performance and production scalability. Although the two studies use different approaches, they share the same goal: overcoming the limitations of low-refractive-index materials and transforming three-dimensional metalenses from laboratory-scale demonstrations into technologies that can be manufactured for industrial applications. By securing both metalens performance and scalable replication strategies, the team has established two key pillars for the commercialization of metalens-based optics. Professor Junsuk Rho, who led the research, said, “These studies are significant because they demonstrate manufacturing technologies that could help bring full-color achromatic metalenses for AR and VR displays into practical industrial use through advanced nanofabrication. We expect the technologies to be broadly applicable not only to lighter and thinner AR glasses and next-generation displays, but also to various optical industries such as imaging systems and optical sensors.” This research was conducted by Professor Junsuk Rho’s team at POSTECH’s Department of Mechanical Engineering, Department of Chemical Engineering, Department of Electrical Engineering, and Graduate School of Convergence Science and Technology. The work was supported by the POSCO-POSTECH-RIST Convergence Research Center program funded by POSCO; National Research Foundation of Korea grants funded by the Ministry of Science and ICT of the Korean government; the Korea Planning & Evaluation Institute of Industrial Technology grant funded by the Ministry of Trade, Industry and Energy of the Korean government; and other related programs. ▶️ DOI: https://doi.org/10.1038/s41467-026-73103-2 ▶️ DOI: https://doi.org/10.1038/s41467-026-73940-1

No More Worries: All You Need Is Light for Skin Wounds

[POSTECH team led by Professor Sei Kwang Hahn develops a wearable patch that converts room light and sunlight into red light for phototherapy] Just as the Na’vi in the film Avatar heal wounds with light beneath the Tree of Souls, ambient light may now be harnessed to help treat skin wounds in humans. A research team led by Professor Sei Kwang Hahn, a master’s student Kyungjoo Jo, and Dr. Seong-Jong Kim in the Department of Materials Science and Engineering and the GraduateSchool of Convergence Science and Technology at POSTECH, has developed, in collaboration with Professor Hyemin Kim’s research team at Konkuk University, a wearable patch that simultaneously promotes wound closure and tissue regeneration using room light or sunlight, without batteries or an external electrical power source. This work has been published online in Advanced Materials, an international journal in materials science. Various technologies for treating wounds with light have been developed recently. When a wound is exposed to light at a specific wavelength, collagen molecules at the wound interface can be crosslinked, allowing the wound to be closed without conventional sutures or tissue adhesives. This technique is known as photochemical tissue bonding, or PTB1). Light can also regulate cellular metabolism and inflammatory responses, thereby promoting rapid tissue repair through a process known as photobiomodulation, or PBM2). These approaches are attractive because they may reduce scarring and reduce the need for invasive procedures. The challenge, however, is how to deliver the therapeutic light. Conventional phototherapeutic systems generally rely on externally powered devices such as LEDs, lasers, optical fibers, and batteries. Wearing bulky or tethered equipment for extended periods is particularly difficult on highly visible areas such as the face. To overcome this limitation, the research team explored a way to use the room light and sunlight that people routinely encounter in everyday life. At the heart of the technology are room light- and sunlight-activatable luminescent particles, or RSLPs3), which convert ambient light into the red light required for phototherapy. The researchers developed calcium–strontium sulfide/sulfate-based particles doped with europium (Eu) and thulium (Tm). By engineering defects and heterostructures within the particles, they enabled the particles to absorb light across a broad wavelength range. The resulting particles emit red light at approximately 640 nm, a wavelength suitable for wound treatment. Even when the surrounding light is briefly interrupted, the particles continue to emit an afterglow, much like glow-in-the-dark stars. The red light emitted by the particles activates the photosensitizer chlorin e6, or Ce64), inducing collagen crosslinking at the wound interface and helping seal incisional wounds. At the same time, the red light promotes fibroblast proliferation and cell migration, alleviates inflammation, and supports new tissue formation. In other words, a single light-emitting platform performs two therapeutic functions simultaneously: wound closure and tissue regeneration. The researchers uniformly embedded the particles within Ecoflex, a highly stretchable silicone elastomer, to create a soft patch that conforms closely to the skin. An adhesive hydrogel layer allows the patch to remain securely attached even as the skin moves, while its light-emitting performance is maintained when the patch is stretched or folded. Simply being exposed to room light or sunlight can therefore provide therapeutic red light without the need for LEDs or batteries. Animal-model experiments demonstrated clear therapeutic effects. When the patch was applied to incisional wounds for 12 hours, wound closure was accelerated and tissue regeneration was significantly enhanced. Circular wounds also healed more rapidly. In addition, a preliminary human evaluation involving the normal facial skin of healthy adults showed that wearing the patch for only 9 hours improved skin elasticity recovery and reduced visible redness. This technology points toward a future in which patients can continue postoperative wound care or skin-regeneration treatments as part of their everyday life, rather than remaining dependent on hospital-based equipment. “This study demonstrates that light readily available in the everyday environment can be used to promote wound closure and tissue regeneration simultaneously, without requiring an external electrical power source,” said Professor Sei Kwang Hahn. “With further development, this technology could become a next-generation wearable photomedicine platform that allows patients to conveniently continue wound care outside the hospital.” This research was supported by the BRIDGE Research Program, the KOBRA Project, the Basic Science Research Program, the Korea Medical Device Development Fund, the B-IRC Program, the Sejong Science Fellowship, and the National Research Facilities and Equipment Center Program of the Korea Basic Science Institute, funded by the Ministry of Science and ICT through the National Research Foundation of Korea. ▶️ DOI: https://doi.org/10.1002/adma.73965 1. PTB, Photochemical Tissue Bonding: A technique that uses light and a photosensitizer to form crosslinks between collagen molecules at a wound interface, thereby bonding the tissue together.. 2. PBM, Photobiomodulation: A light-based therapeutic approach that uses low-intensity light at specific wavelengths to regulate cellular metabolism, inflammatory responses, and tissue regeneration.. 3. RSLP(Room light- and sunlight-activatable luminescent particles): Luminescent particles that absorb room light and sunlight and convert them into therapeutic red light. In this study, the researchers used CaSr(S,SO₄):Eu,Tm-based particles. 4. Ce6(Chlorin e6): A photosensitizer that responds to red light. Upon light exposure, Ce6 can trigger photochemical reactions that induce collagen crosslinking.

“Acts as a ‘protective barrier’ in blood vessels, but disappears in tumor tissue”... Mussel protein opens a new path for

[POSTECH develops ‘stealth’ anticancer nanoparticles made from mussel adhesive proteins that 'lie in wait and attack only cancer cells'] Pancreatic cancer is considered one of the most lethal cancer because it is often diagnosed late and is difficult to treat. However, a South Korean research team has recently succeeded in developing “smart nanoparticles” that remain hidden in normal tissue but shed their protective coating and release anticancer drugs once they reach tumor tissue. This drug delivery technology is attracting attention for its potential to reduce the side effects of cancer treatment while significantly improving treatment efficacy. The findings of this study, conducted by a research team headed by Professor Hyung Joon Cha of the Department of Chemical Engineering and the Graduate School of Convergence Science and Technology at POSTECH, together with Ph.D. candidate Hyeokjun Lee of the Department of Chemical Engineering, were recently published in the online edition of Biomaterials, an international academic journal in the field of biomaterials. Pancreatic cancer is a notoriously difficult-to-treat disease because it is located deep within the body, making it hard to detect in its early stages, and it spreads rapidly to surrounding organs. While intravenous chemotherapy is the most widely used treatment, “gemcitabine”—a leading drug in this class—breaks down rapidly in the bloodstream and fails to reach the tumor tissue in sufficient quantities. Furthermore, it has the limitation of also attacking normal cells, causing severe systemic side effects. To address this issue, the research team drew inspiration from the principle that allows mussels to cling firmly even to wet rocks. Using molecular synthetic biotechnology, they mass-produced “mussel adhesive protein,” converted it into nanoparticles, encapsulated gemcitabine inside them, and coated the surface with a protective layer of the biocompatible polymer “polyethylene glycol (PEG1)).” Thanks to this protective layer, the nanoparticles can conceal their adhesive properties while traveling through the bloodstream, evading attacks from immune cells and circulating stably throughout the body. The key to this study is a “spatially controlled stimulus-response system” designed so that the protective coating is removed only in tumor tissue. The research team used a special peptide—which is cleaved only by “MMP22),” an enzyme secreted in particularly high amounts in pancreatic cancer tissue—as the link in the protective coating. As a result, the nanoparticles retain their protective coating while in the bloodstream, but once they reach tumor tissue, the coating is removed by the MMP2 enzyme. Once stripped of their protective coating, these “stealth” nanoparticles regain the strong adhesive properties of mussel adhesive protein, firmly attaching themselves to the tumor tissue. They then penetrate the cancer cells and continuously release the anticancer drug, selectively attacking only the cancer cells. When these nanoparticles were administered intravenously in an animal model of pancreatic cancer, the results were striking. Compared to cases where conventional anticancer drugs were administered alone or where standard nanoparticles were used, the accumulation and retention time of the nanoparticles within the tumor tissue increased by more than 60%. No systemic toxicity was observed, and tumor volume and weight were reduced by more than half compared to the group treated with conventional anticancer drugs alone. Histological analysis also confirmed widespread cancer cell death. This technology is significant because it not only enhances treatment efficacy by keeping the drug in the tumor tissue for a longer period but also reduces the amount of drug delivered to normal tissue, thereby minimizing systemic side effects—a major challenge in cancer treatment. The research team plans to further develop this platform into a next-generation targeted drug delivery technology that can be applied not only to pancreatic cancer but also to the treatment of various intractable solid tumors where drug delivery is difficult. Professor Cha Hyung Joon stated, “The drug delivery platform we have developed is a novel systemic treatment that, even when administered intravenously, is selectively activated only within tumor tissue to release the drug,” adding, “By reducing the side effects of cancer treatment while enhancing its efficacy, it will offer hope to patients with intractable solid tumors, including pancreatic cancer.” Meanwhile, this research was conducted with support from the National Research Laboratories (NRL) 2.0 program of the National Research Foundation of Korea (NRF) under the Ministry of Science and ICT, and the Support Program for Promoting the Dissemination of Outstanding R&D Achievements by the Commercializations Promotion Agency for R&D Outcomes (COMPA). ▶️ DOI: https://doi.org/10.1016/j.biomaterials.2026.124384 1. PEG: An abbreviation for polyethylene glycol, it is a polymer with excellent biocompatibility that is widely used in pharmaceuticals and cosmetics. When attached to the surface of a drug delivery system, it acts as a kind of “stealth shield,” helping the drug evade attack by immune cells and circulate in the bloodstream for an extended period. 2. MMP2 : An abbreviation for Matrix Metalloproteinase-2, it is a type of protease that cancer cells overexpress to invade and grow into surrounding tissues. In this study, it was used as a “chemical switch (trigger)” to selectively induce the removal of the protective coating (PEG) only when the nanoparticles reached pancreatic cancer tissue (tumor microenvironment).

Wealthier and More Populated Metropolitan Areas Respond More Strongly to Early Drought News by Saving Water

[POSTECH Professor Jonghun Kam’s research team, in collaboration with KRIHS, NDMI, Stanford University, and the National University of Singapore (NUS), presents AI-based, region-specific water conservation scenarios using drought news data] As climate change increases the risk of severe droughts, water resources management is an urgent challenge. Drought develops slowly, which makes the public aware of the ongoing drought. Governments rely on the media to communicate drought risks and encourage water conservation. However, it has remained unclear how much drought-related news actually contributes to water-saving behavior in different regions. A research team led by Professor Jonghun Kam at POSTECH used explainable artificial intelligence (XAI) to analyze how drought-related news coverage influenced household water conservation during the 2022-2023 drought in southwestern Korea. The study, titled “Spatiotemporal and Economic Impacts of Media on Water Conservation during Drought: An Explainable AI Approach,” was published in Water Resources Research, a leading international journal in water resources. Traditional hydrological models mainly focus on physical drought conditions, such as rainfall deficits and dam storage levels. However, drought response often depends on social factors, including public awareness, media coverage, and regional socioeconomic conditions. To capture interactions between natural and social systems, the research team developed AI models using household water use data, drought indices, dam storage levels, temperature, and drought-related news articles. The team first calculated monthly water conservation rates by comparing actual household water use with average water use in the same month over the previous five years. The results showed that metropolitan areas with higher population density and income levels tended to show stronger voluntary water conservation during drought. In Gwangju Metropolitan City, water conservation rates increased by up to about 10%, while smaller cities and rural areas in Jeollanam-do showed increases of only about 3%. Among several AI models, the N-BEATS model predicted monthly water conservation rates with more than 73% accuracy in both metropolitan and smaller regions. The AI-based scenario analysis showed that when drought-related news coverage increased during the early and caution stages of drought, household water conservation in Gwangju increased by about 14 percentage points compared with the baseline. This corresponded to approximately USD 0.82 million in reduced tap water production costs. In contrast, the same increase in news coverage led to only about four percentage point increase in smaller cities and rural areas, equivalent to approximately USD 0.3 million in cost savings. This suggests that the impact of media coverage differs by region. Metropolitan households may have flexible water use and respond sensitively to drought information, while rural households may have limited room to further reduce basic domestic water consumption. The study also found that the timing of media communication matters. News coverage had a stronger effect during the early and caution stages of drought, before conditions became severe. Once drought entered the alert stage, public concern and water conservation were already relatively high, leaving room for additional savings limited. This suggests that drought communication is effective when delivered early, rather than after the crisis has intensified. Existing drought contingency policies are from a top-bottom approach applying uniform water conservation targets across drought-affected regions. This study provides a scientific basis for setting region-specific water-saving targets by considering both regional conditions and drought stages. The first author, Eunmi Lee of POSTECH, said, “By using explainable AI, we are able to quantitatively reveal the hidden link between media coverage as a social factor and actual public water-saving behavior.” She added, “We hope that AI-based water resources research integrating environmental and social data will provide actionable information for developing effective, region-specific drought response strategies in climate crisis.” This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Ministry of Science and ICT (MSIT) (No. RS-2026-25470040) and by a grant (2022-MOIS63-001 (RS-2022-ND641011)) for Cooperative Research Method and Safety Management Technology in National Disaster, funded by the Ministry of Interior and Safety (MOIS, Korea) ▶️ DOI: https://doi.org/10.1029/2025WR042406