Pub Date : 2026-10-15DOI: 10.1186/s40580-026-00581-x
Juho Ha, Hyunki Kim, Seohyeon Jang, Andi Haryanto, Jungmin Yoo, Yeonghye Yun, Sunbeom Ahn, Gyeong Ho Han, Myoung Hwan Oh, Soo Young Kim, Chan Woo Lee, Inho Nam, Sang Hyun Ahn
Copper sulfides typically favor formate production in CO2 electrolysis, yet their behavior under membrane electrode assembly (MEA) conditions remains largely uncharted. In this study, CuxS/Cu electrodes with tunable morphologies, including spheres, films, polygons, and cubes, were synthesized, and the surface sulfur content was systematically tuned through electrodeposition followed by controlled sulfidation. Performance evaluation in MEA systems reveals that sulfidation markedly enhances C2H4 selectivity and partial current density (PCD). For spherical Cu electrodes, sulfidation to a S/Cu surface ratio of 0.33 resulted in a 1.76-fold increase in C2H4 PCD, from 58.2 to 102.5 mA/cm2 at 3.2 V. In situ Raman spectroscopy suggests that sulfur coordination stabilizes bridge-bonded CO and promotes the formation of the CH2CHO intermediate, which is consistent with the enhanced C2H4 activity. Density functional theory calculations reveal that Cu-rich sulfide surfaces, including Cu2S and Cu31S16, stabilize *OCCO intermediates, suggesting a thermodynamically favorable C–C coupling reaction pathway through synergistic Cu–S coordination. These findings identify surface sulfur incorporation as a structural motif that actively promotes CO dimerization and enhances C2H4 selectivity under MEA conditions.
{"title":"Sulfur-incorporated copper catalysts for enhanced ethylene production in membrane electrode assembly CO2 electrolysis","authors":"Juho Ha, Hyunki Kim, Seohyeon Jang, Andi Haryanto, Jungmin Yoo, Yeonghye Yun, Sunbeom Ahn, Gyeong Ho Han, Myoung Hwan Oh, Soo Young Kim, Chan Woo Lee, Inho Nam, Sang Hyun Ahn","doi":"10.1186/s40580-026-00581-x","DOIUrl":"10.1186/s40580-026-00581-x","url":null,"abstract":"<div><p>Copper sulfides typically favor formate production in CO<sub>2</sub> electrolysis, yet their behavior under membrane electrode assembly (MEA) conditions remains largely uncharted. In this study, Cu<sub>x</sub>S/Cu electrodes with tunable morphologies, including spheres, films, polygons, and cubes, were synthesized, and the surface sulfur content was systematically tuned through electrodeposition followed by controlled sulfidation. Performance evaluation in MEA systems reveals that sulfidation markedly enhances C<sub>2</sub>H<sub>4</sub> selectivity and partial current density (PCD). For spherical Cu electrodes, sulfidation to a S/Cu surface ratio of 0.33 resulted in a 1.76-fold increase in C<sub>2</sub>H<sub>4</sub> PCD, from 58.2 to 102.5 mA/cm<sup>2</sup> at 3.2 V. In situ Raman spectroscopy suggests that sulfur coordination stabilizes bridge-bonded CO and promotes the formation of the CH<sub>2</sub>CHO intermediate, which is consistent with the enhanced C<sub>2</sub>H<sub>4</sub> activity. Density functional theory calculations reveal that Cu-rich sulfide surfaces, including Cu<sub>2</sub>S and Cu<sub>31</sub>S<sub>16</sub>, stabilize *OCCO intermediates, suggesting a thermodynamically favorable C–C coupling reaction pathway through synergistic Cu–S coordination. These findings identify surface sulfur incorporation as a structural motif that actively promotes CO dimerization and enhances C<sub>2</sub>H<sub>4</sub> selectivity under MEA conditions.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":712,"journal":{"name":"Nano Convergence","volume":"13 1","pages":""},"PeriodicalIF":13.8,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s40580-026-00581-x.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149189812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-10-14DOI: 10.1186/s40580-026-00583-9
Zayakhuu Gerelkhuu, Angela Serra, Seung-Geun Park, Ishita Virmani, Aline Maria Zigiotto de Medeiros, Ariadna Acedo-Terrades, Fotini Nikiforou, Achilleas Karakoltzidis, Seong-Kwan Kim, Hyeonsoo Eom, Martin Himly, Dimitris G. Mintis, Vasileios Minadakis, Dimitra-Danai Varsou, Spyros Karakitsios, Dimosthenis A. Sarigiannis, Giusy del Giudice, Marcella Torres Maia, Michele Fratello, Anastasios G. Papadiamantis, Panagiotis Isigonis, Anna Pohl, Sang Hee Lee, Byung-Tae Lee, Yang-won Suh, Candida Lorusso, Francesco Dondero, Steffi Friedrichs, Antreas Afantitis, Haralambos Sarimveis, Eva-Kathrin Schillinger, Peter Wick, Diego Stéfani Teodoro Martinez, Iseult Lynch, Thomas E. Exner, Tommaso Serchi, Tae Hyun Yoon, Dario Greco
Safe and Sustainable by Design (SSbD) is emerging as a transformative framework for integrating safety and sustainability with functionality in nanotechnology. This review provides a unified perspective on SSbD, bridging its scientific foundations, global policy developments, and industrial implementation. We highlight how advances in mechanistic understanding of nanomaterial–biological interactions enable earlier and more predictive safety assessment, supporting informed material design. At the policy level, evolving regulatory and governance approaches across regions are shaping the integration of sustainability and safety requirements. From an industrial perspective, SSbD is driving innovation in material substitution, process optimization, and lifecycle-oriented design. Despite this progress, challenges remain in standardization, data integration, and alignment between regulatory frameworks and industrial practice. This work positions SSbD as a central paradigm for enabling responsible innovation and accelerating the transition toward safer and more sustainable nanotechnologies.
{"title":"Safe and sustainable by design (SSbD) for nanotechnology: scientific foundations, global policy perspectives, and industrial practice","authors":"Zayakhuu Gerelkhuu, Angela Serra, Seung-Geun Park, Ishita Virmani, Aline Maria Zigiotto de Medeiros, Ariadna Acedo-Terrades, Fotini Nikiforou, Achilleas Karakoltzidis, Seong-Kwan Kim, Hyeonsoo Eom, Martin Himly, Dimitris G. Mintis, Vasileios Minadakis, Dimitra-Danai Varsou, Spyros Karakitsios, Dimosthenis A. Sarigiannis, Giusy del Giudice, Marcella Torres Maia, Michele Fratello, Anastasios G. Papadiamantis, Panagiotis Isigonis, Anna Pohl, Sang Hee Lee, Byung-Tae Lee, Yang-won Suh, Candida Lorusso, Francesco Dondero, Steffi Friedrichs, Antreas Afantitis, Haralambos Sarimveis, Eva-Kathrin Schillinger, Peter Wick, Diego Stéfani Teodoro Martinez, Iseult Lynch, Thomas E. Exner, Tommaso Serchi, Tae Hyun Yoon, Dario Greco","doi":"10.1186/s40580-026-00583-9","DOIUrl":"10.1186/s40580-026-00583-9","url":null,"abstract":"<div><p>Safe and Sustainable by Design (SSbD) is emerging as a transformative framework for integrating safety and sustainability with functionality in nanotechnology. This review provides a unified perspective on SSbD, bridging its scientific foundations, global policy developments, and industrial implementation. We highlight how advances in mechanistic understanding of nanomaterial–biological interactions enable earlier and more predictive safety assessment, supporting informed material design. At the policy level, evolving regulatory and governance approaches across regions are shaping the integration of sustainability and safety requirements. From an industrial perspective, SSbD is driving innovation in material substitution, process optimization, and lifecycle-oriented design. Despite this progress, challenges remain in standardization, data integration, and alignment between regulatory frameworks and industrial practice. This work positions SSbD as a central paradigm for enabling responsible innovation and accelerating the transition toward safer and more sustainable nanotechnologies.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":712,"journal":{"name":"Nano Convergence","volume":"13 1","pages":""},"PeriodicalIF":13.8,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s40580-026-00583-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149178630","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-10-08DOI: 10.1186/s40580-026-00582-w
Subin Lee, Joohee Lee, Saleem Sidra, Sieon Jung, Myong Yong Choi, Do Hwan Kim, Myung Hwa Kim, Youngmi Lee
Reducing the Ru content while maintaining high oxygen evolution reaction (OER) activity remains a major challenge. In this study, single phase Ru0.30Cr0.31Mn0.39O2 nanofibers were prepared through electrospinning process followed by temperature-regulated heat treatment. The temperature regulation process induces a unique formation of a hollow tube-in-tube structure while simultaneously influencing crystallinity and defect distribution. Among the prepared samples, the material treated at 400 °C shows the most favorable performance for OER. This behavior is associated with improved transport properties and increased accessibility of catalytically active sites. The catalyst reaches an overpotential of only 211 mV at 10 mA cm− 2 and maintains stable operation in alkaline electrolyte. The mass-normalized activity and turnover frequency (TOFmetal) are significantly 293- and 104-fold higher than those of commercial RuO2, respectively, highlighting its superior catalytic efficiency with reduced Ru content. Theoretical calculations based on density functional theory suggest that the incorporation of Cr and Mn modulates the electronic structure of RuO2, particularly near the Fermi level, and adjusts the d-band center position. These changes lower the energy barrier associated with the O* to OOH* step. The results thus demonstrate that simultaneous control of composition and morphology can improve OER performance in alkaline media.
Graphical abstract
在保持高析氧反应（OER）活性的同时降低Ru含量仍然是一个主要的挑战。本研究采用静电纺丝法制备了单相Ru0.30Cr0.31Mn0.39O2纳米纤维，并进行了温度调节热处理。温度调节过程导致了独特的中空管中管结构的形成，同时影响了结晶度和缺陷分布。在所制备的样品中，400℃处理的材料表现出最有利的OER性能。这种行为与改善的运输性质和增加的催化活性位点的可及性有关。该催化剂在10 mA cm−2下的过电位仅为211 mV，在碱性电解质中保持稳定的工作状态。质量归一化活性和周转频率（TOFmetal）分别比商用RuO2高293倍和104倍，表明其在降低Ru含量的情况下具有优异的催化效率。基于密度泛函理论的理论计算表明，Cr和Mn的掺入调节了RuO2的电子结构，特别是在费米能级附近，并调整了d带中心位置。这些变化降低了与O*到OOH*步骤相关的能垒。结果表明，同时控制组成和形貌可以改善碱性介质中的OER性能。图形抽象
{"title":"Tube-in-tube hollow RuCrMn oxide nanofibers for noble-metal-efficient oxygen evolution via Cr/Mn-regulated Ru active sites","authors":"Subin Lee, Joohee Lee, Saleem Sidra, Sieon Jung, Myong Yong Choi, Do Hwan Kim, Myung Hwa Kim, Youngmi Lee","doi":"10.1186/s40580-026-00582-w","DOIUrl":"10.1186/s40580-026-00582-w","url":null,"abstract":"<div><p>Reducing the Ru content while maintaining high oxygen evolution reaction (OER) activity remains a major challenge. In this study, single phase Ru<sub>0.30</sub>Cr<sub>0.31</sub>Mn<sub>0.39</sub>O<sub>2</sub> nanofibers were prepared through electrospinning process followed by temperature-regulated heat treatment. The temperature regulation process induces a unique formation of a hollow tube-in-tube structure while simultaneously influencing crystallinity and defect distribution. Among the prepared samples, the material treated at 400 °C shows the most favorable performance for OER. This behavior is associated with improved transport properties and increased accessibility of catalytically active sites. The catalyst reaches an overpotential of only 211 mV at 10 mA cm<sup>− 2</sup> and maintains stable operation in alkaline electrolyte. The mass-normalized activity and turnover frequency (TOF<sub>metal</sub>) are significantly 293- and 104-fold higher than those of commercial RuO<sub>2</sub>, respectively, highlighting its superior catalytic efficiency with reduced Ru content. Theoretical calculations based on density functional theory suggest that the incorporation of Cr and Mn modulates the electronic structure of RuO<sub>2</sub>, particularly near the Fermi level, and adjusts the d-band center position. These changes lower the energy barrier associated with the O* to OOH* step. The results thus demonstrate that simultaneous control of composition and morphology can improve OER performance in alkaline media.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":712,"journal":{"name":"Nano Convergence","volume":"13 1","pages":""},"PeriodicalIF":13.8,"publicationDate":"2026-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s40580-026-00582-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149132993","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-10-01DOI: 10.1186/s40580-026-00579-5
Jong Won Baek, Ju-Hyun Jung, Hyeong-Seok Kim, Junho Kim, Dong-Hyun Kim, Jangyup Son, Sangyul Ha
As back-end-of-line (BEOL) scaling pushes porous low-k dielectrics toward their reliability limits, nonporous dielectrics that combine a low dielectric constant (k) with mechanical, thermal, and interfacial robustness are increasingly needed. Among nonporous low-k candidates, amorphous boron nitride (aBN) has attracted attention for advanced BEOL interconnects owing to its low-k value (k ≈ 2) and high mechanical robustness, as reflected by a Young’s modulus of ~50 GPa. In this review, we discuss aBN from a BEOL process-integration perspective, focusing on the origin of its low-k behavior, process-dependent structure–property relationships, and its potential roles in BEOL interconnects and hybrid-bonding interfaces. Finally, we discuss remaining issues and future directions for validating aBN as a nonporous low-k dielectric/interface material under BEOL and hybrid-bonding integration.
{"title":"Amorphous boron nitride as a low-k dielectric for next-generation BEOL interconnects","authors":"Jong Won Baek, Ju-Hyun Jung, Hyeong-Seok Kim, Junho Kim, Dong-Hyun Kim, Jangyup Son, Sangyul Ha","doi":"10.1186/s40580-026-00579-5","DOIUrl":"10.1186/s40580-026-00579-5","url":null,"abstract":"<div><p>As back-end-of-line (BEOL) scaling pushes porous low-<i>k</i> dielectrics toward their reliability limits, nonporous dielectrics that combine a low dielectric constant (<i>k</i>) with mechanical, thermal, and interfacial robustness are increasingly needed. Among nonporous low-<i>k</i> candidates, amorphous boron nitride (aBN) has attracted attention for advanced BEOL interconnects owing to its low-<i>k</i> value (<i>k</i> ≈ 2) and high mechanical robustness, as reflected by a Young’s modulus of ~50 GPa. In this review, we discuss aBN from a BEOL process-integration perspective, focusing on the origin of its low-<i>k</i> behavior, process-dependent structure–property relationships, and its potential roles in BEOL interconnects and hybrid-bonding interfaces. Finally, we discuss remaining issues and future directions for validating aBN as a nonporous low-<i>k</i> dielectric/interface material under BEOL and hybrid-bonding integration.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":712,"journal":{"name":"Nano Convergence","volume":"13 1","pages":""},"PeriodicalIF":13.8,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13631203/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149102519","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-09-30DOI: 10.1186/s40580-026-00580-y
Myeong-Jun Lee, Minkyu Shin, Joungpyo Lim, Ha Rim Ahn, Dain Jung, Shan Liu, Chenzhong Li, Jin-Ha Choi, Jeong-Woo Choi
CRISPR-based, amplification-free colorimetric sensing has emerged as a powerful approach for exosomal miRNA (Exo-miR) analysis, offering high specificity, simple operation, and direct visual readout suitable for point-of-care diagnostics. However, the sensitivity of existing systems is limited because conventional colorimetric probes generate inherently weak signals, making the reliable detection of ultralow-abundance Exo-miRs difficult without amplification. Here, we firstly report an amplification-free colorimetric biosensor based on a horseradish peroxidase (HRP)/gold nanoparticle (Au)/molybdenum disulfide (MoS2) nanobiohybrid catalyst (HRP/Au/MoS2) that generates intense chromogenic signals via synergistic catalytic interactions. In this nanoparticle, MoS2 provides a high-surface-area scaffold with intrinsic peroxidase-like activity, Au nanoparticles facilitate efficient electron transfer, and HRP contributes to high catalytic turnover. Together, these features markedly enhance the oxidation of 3,3′,5,5′-tetramethylbenzidine (TMB) and the resulting blue coloration. Using this signal-amplifying HRP/Au/MoS2, a magnet-based sensing system was constructed by immobilizing the HRP/Au/MoS2 onto magnetic nanobeads (MNBs) through biotinylated single-stranded RNA linkers (MNB–ssRNA–HRP/Au/MoS2). This construction was then integrated into a CRISPR/Cas13a assay, in which the target miR-155 activates Cas13a, which cleaves the ssRNA linkers, releasing the HRP/Au/MoS2 into solution for colorimetric readout. The resulting biosensor exhibits a detection range of 1 fM–1 nM (LOD = 0.82 fM), and single-nucleotide specificity without nucleic acid amplification. Furthermore, our system enables detection of Exo-miR-155 in both cell-derived exosomes and clinical samples, distinguishing cancer patients from healthy donors. Therefore, this platform represents a promising approach for highly sensitive, selective, simple, and time-efficient detection of Exo-miR, which can be used for liquid biopsy-based early cancer detection.
{"title":"Amplification-free colorimetric detection of exosomal miRNA using a CRISPR/Cas13a-activated HRP/Au/MoS2 nanobiohybrid catalyst","authors":"Myeong-Jun Lee, Minkyu Shin, Joungpyo Lim, Ha Rim Ahn, Dain Jung, Shan Liu, Chenzhong Li, Jin-Ha Choi, Jeong-Woo Choi","doi":"10.1186/s40580-026-00580-y","DOIUrl":"10.1186/s40580-026-00580-y","url":null,"abstract":"<div><p>CRISPR-based, amplification-free colorimetric sensing has emerged as a powerful approach for exosomal miRNA (Exo-miR) analysis, offering high specificity, simple operation, and direct visual readout suitable for point-of-care diagnostics. However, the sensitivity of existing systems is limited because conventional colorimetric probes generate inherently weak signals, making the reliable detection of ultralow-abundance Exo-miRs difficult without amplification. Here, we firstly report an amplification-free colorimetric biosensor based on a horseradish peroxidase (HRP)/gold nanoparticle (Au)/molybdenum disulfide (MoS<sub>2</sub>) nanobiohybrid catalyst (HRP/Au/MoS<sub>2</sub>) that generates intense chromogenic signals via synergistic catalytic interactions. In this nanoparticle, MoS<sub>2</sub> provides a high-surface-area scaffold with intrinsic peroxidase-like activity, Au nanoparticles facilitate efficient electron transfer, and HRP contributes to high catalytic turnover. Together, these features markedly enhance the oxidation of 3,3′,5,5′-tetramethylbenzidine (TMB) and the resulting blue coloration. Using this signal-amplifying HRP/Au/MoS<sub>2</sub>, a magnet-based sensing system was constructed by immobilizing the HRP/Au/MoS<sub>2</sub> onto magnetic nanobeads (MNBs) through biotinylated single-stranded RNA linkers (MNB–ssRNA–HRP/Au/MoS<sub>2</sub>). This construction was then integrated into a CRISPR/Cas13a assay, in which the target miR-155 activates Cas13a, which cleaves the ssRNA linkers, releasing the HRP/Au/MoS<sub>2</sub> into solution for colorimetric readout. The resulting biosensor exhibits a detection range of 1 fM–1 nM (LOD = 0.82 fM), and single-nucleotide specificity without nucleic acid amplification. Furthermore, our system enables detection of Exo-miR-155 in both cell-derived exosomes and clinical samples, distinguishing cancer patients from healthy donors. Therefore, this platform represents a promising approach for highly sensitive, selective, simple, and time-efficient detection of Exo-miR, which can be used for liquid biopsy-based early cancer detection.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":712,"journal":{"name":"Nano Convergence","volume":"13 1","pages":""},"PeriodicalIF":13.8,"publicationDate":"2026-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13627550/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149102596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-09-22DOI: 10.1186/s40580-026-00574-w
Sunghyun Kim, Hanbin Cho, Taejee Kim, Huseok Lee, Sungjae An, Sungju Ryu
Image classification on edge devices requires models that maintain accuracy while reducing inference cost. This paper reviews the development of image classification models from convolutional neural networks to Vision Transformer models and examines their relevance to edge artificial intelligence applications. To improve Vision Transformer inference efficiency without redesigning the backbone, we propose a selective token merging method that identifies low-significance image tokens and applies token merging mainly to those tokens. Experimental results on CIFAR-10 using ViT-Tiny show that the proposed method improves the accuracy-throughput tradeoff compared with standard Token Merging. In particular, selectively merging the least significant token group with a layer-dependent schedule achieves 97.41% accuracy and 19.04 images/s, compared with 97.22% accuracy and 18.97 images/s for standard Token Merging. Additional ImageNet-1 K evaluations using ViT-Tiny, ViT-Small, and ViT-Base show that selective merging in an shallow-layer window preserves more of the baseline accuracy than standard ToMe while retaining an inference speedup. These results show that token significance can guide more accurate and efficient token reduction for edge-oriented image classification.
{"title":"Review and optimization of image classification models for edge AI applications","authors":"Sunghyun Kim, Hanbin Cho, Taejee Kim, Huseok Lee, Sungjae An, Sungju Ryu","doi":"10.1186/s40580-026-00574-w","DOIUrl":"10.1186/s40580-026-00574-w","url":null,"abstract":"<div><p>Image classification on edge devices requires models that maintain accuracy while reducing inference cost. This paper reviews the development of image classification models from convolutional neural networks to Vision Transformer models and examines their relevance to edge artificial intelligence applications. To improve Vision Transformer inference efficiency without redesigning the backbone, we propose a selective token merging method that identifies low-significance image tokens and applies token merging mainly to those tokens. Experimental results on CIFAR-10 using ViT-Tiny show that the proposed method improves the accuracy-throughput tradeoff compared with standard Token Merging. In particular, selectively merging the least significant token group with a layer-dependent schedule achieves 97.41% accuracy and 19.04 images/s, compared with 97.22% accuracy and 18.97 images/s for standard Token Merging. Additional ImageNet-1 K evaluations using ViT-Tiny, ViT-Small, and ViT-Base show that selective merging in an shallow-layer window preserves more of the baseline accuracy than standard ToMe while retaining an inference speedup. These results show that token significance can guide more accurate and efficient token reduction for edge-oriented image classification.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":712,"journal":{"name":"Nano Convergence","volume":"13 1","pages":""},"PeriodicalIF":13.8,"publicationDate":"2026-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s40580-026-00574-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149023504","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-09-22DOI: 10.1186/s40580-026-00577-7
Su Bin Lee, Anara Molkenova, Seung Yun Yang, Ki Su Kim
The continuously growing demand for efficient and non-invasive treatment has accelerated the development of near infrared (NIR) responsive phototherapy, a therapeutic modality that leverages the biological transparent window (700–1850 nm) to enable deeper tissue penetration compared to ultraviolet (UV)/visible light approaches. NIR-responsive two-dimensional (2D) nanomaterials have emerged as particularly promising platforms for advanced photomedicine, attributed to high surface-to-volume ratio, tunable electronic structure, and broadband optical absorption. 2D nanomaterials can be broadly classified into elemental materials, layered nonmetal compounds, metal-based compounds, and 2D organic frameworks, each exhibiting distinct physicochemical and photonic features that enable tailored photothermal and photodynamic therapy applications. Their structural versatility and surface functionality further facilitate multimodal NIR theranostic integration. This review systematically investigates recent advances in NIR-responsive 2D nanomaterial-mediated phototherapy, with particular emphasis on their photomedical performance. Furthermore, the existing critical challenges, including biocompatibility, long-term biosafety, controlled degradation, NIR penetration depth limitations, and clinical translation barriers have been overviewed. Finally, future perspectives are proposed to guide the development of the next-generation NIR-responsive 2D nanomaterials for photomedicine.
{"title":"2D nanomaterials for near infrared responsive photomedicine: a review","authors":"Su Bin Lee, Anara Molkenova, Seung Yun Yang, Ki Su Kim","doi":"10.1186/s40580-026-00577-7","DOIUrl":"10.1186/s40580-026-00577-7","url":null,"abstract":"<div><p>The continuously growing demand for efficient and non-invasive treatment has accelerated the development of near infrared (NIR) responsive phototherapy, a therapeutic modality that leverages the biological transparent window (700–1850 nm) to enable deeper tissue penetration compared to ultraviolet (UV)/visible light approaches. NIR-responsive two-dimensional (2D) nanomaterials have emerged as particularly promising platforms for advanced photomedicine, attributed to high surface-to-volume ratio, tunable electronic structure, and broadband optical absorption. 2D nanomaterials can be broadly classified into elemental materials, layered nonmetal compounds, metal-based compounds, and 2D organic frameworks, each exhibiting distinct physicochemical and photonic features that enable tailored photothermal and photodynamic therapy applications. Their structural versatility and surface functionality further facilitate multimodal NIR theranostic integration. This review systematically investigates recent advances in NIR-responsive 2D nanomaterial-mediated phototherapy, with particular emphasis on their photomedical performance. Furthermore, the existing critical challenges, including biocompatibility, long-term biosafety, controlled degradation, NIR penetration depth limitations, and clinical translation barriers have been overviewed. Finally, future perspectives are proposed to guide the development of the next-generation NIR-responsive 2D nanomaterials for photomedicine. </p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":712,"journal":{"name":"Nano Convergence","volume":"13 1","pages":""},"PeriodicalIF":13.8,"publicationDate":"2026-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s40580-026-00577-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149023494","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-09-11DOI: 10.1186/s40580-026-00576-8
Pravin S. Pawar, Rahul K. Yadav, Vishesh Manjunath, Anil V. Ghule, Gee Yeong Kim, Jaeyeong Heo
Two-terminal (2T) monolithic CIGS/perovskite tandem solar cells exploit complementary light-absorption characteristics to approach the theoretical Shockley–Queisser (SQ) efficiency limit (~ 45%), with recently reported highest power conversion efficiency (PCE) of 31.09% and a certified PCE of 30.57%. Nevertheless, surface and interface defects remain the critical loss mechanism, inducing non-radiative recombination, reducing open-circuit voltage (VOC), and accelerating degradation pathways. This review provides a systematic analysis of surface and interface defect chemistries and engineering strategies across 2T CIGS/perovskite architectures. We elucidate defect formation mechanisms at the CIGS/perovskite heterojunction and individual layer surfaces, encompassing point defects, grain boundaries, and structural imperfections, and their impact on device performance. Mitigation strategies are critically evaluated, including surface passivation, interface band-alignment engineering, compositional defect tolerance, and interconnection layer optimization. The coupled relationships between defect density, environmental degradation (moisture penetration, thermal cycling), and device reliability are analyzed systematically. Finally, we identify fundamental bottlenecks and research directions toward high efficiencies (≥ 32%), commercial scalability, and long-term operational stability in 2T CIGS/perovskite tandem solar cells.
{"title":"Strategies for mitigating surface and interface defects in CIGS/perovskite tandem solar cells: a comprehensive review","authors":"Pravin S. Pawar, Rahul K. Yadav, Vishesh Manjunath, Anil V. Ghule, Gee Yeong Kim, Jaeyeong Heo","doi":"10.1186/s40580-026-00576-8","DOIUrl":"10.1186/s40580-026-00576-8","url":null,"abstract":"<div><p>Two-terminal (2T) monolithic CIGS/perovskite tandem solar cells exploit complementary light-absorption characteristics to approach the theoretical Shockley–Queisser (SQ) efficiency limit (~ 45%), with recently reported highest power conversion efficiency (PCE) of 31.09% and a certified PCE of 30.57%. Nevertheless, surface and interface defects remain the critical loss mechanism, inducing non-radiative recombination, reducing open-circuit voltage (V<sub>OC</sub>), and accelerating degradation pathways. This review provides a systematic analysis of surface and interface defect chemistries and engineering strategies across 2T CIGS/perovskite architectures. We elucidate defect formation mechanisms at the CIGS/perovskite heterojunction and individual layer surfaces, encompassing point defects, grain boundaries, and structural imperfections, and their impact on device performance. Mitigation strategies are critically evaluated, including surface passivation, interface band-alignment engineering, compositional defect tolerance, and interconnection layer optimization. The coupled relationships between defect density, environmental degradation (moisture penetration, thermal cycling), and device reliability are analyzed systematically. Finally, we identify fundamental bottlenecks and research directions toward high efficiencies (≥ 32%), commercial scalability, and long-term operational stability in 2T CIGS/perovskite tandem solar cells.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":712,"journal":{"name":"Nano Convergence","volume":"13 1","pages":""},"PeriodicalIF":13.8,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s40580-026-00576-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148927507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-09-10DOI: 10.1186/s40580-026-00575-9
Junsu Yu, Hwiho Hwang, Hyungjin Kim, Woo Young Choi
Energy-efficient neural computing is increasingly limited not by computational throughput but by memory access and data movement. In-memory computing architectures offer a promising solution by collocating storage and computation, yet their practical realization remains constrained by static power dissipation, thermal challenges, and limited integration density in existing memory devices. Here, we demonstrate a vertically stackable charge-trap-flash (CTF)-based memcapacitor array that combines the high integration density of 3D NAND technology with the charge-domain computation. A continuous in-situ-doped N+ bottom readout electrode is introduced beneath the lightly doped active layer, decoupling lateral readout resistance from depletion-based capacitance modulation and alleviating the resistance–depletion trade-off of the previous architecture. The fabricated 24 × 48 memcapacitor array operates through transient charge displacement and demonstrates highly uniform array-level characteristics (σ/µ < 0.37%), reliable 16-state closed-loop weight programming, and linear charge-domain VMM with an error below 0.227%. When scaled toward modern 3D NAND dimensions, a verification–inference electrostatic mismatch introduces systematic VMM error, which is reduced by up to 84.6% through an intercell trapped-charge scheme in TCAD simulations. Geometric scaling based on the CV2 relation projects femtojoule-level intrinsic cell read energy at scaled dimensions. A hybrid hardware–software spiking neural network evaluation, in which only the final 24 × 10 layer is mapped to measured arrays, achieves 88.01% CIFAR-10 accuracy compared with 88.17% in software. These results establish vertically stackable memcapacitors as a scalable charge-domain computing platform combining reliable array-level operation with high-density vertical integration.
{"title":"Capacitive in-memory vector–matrix multiplication computing with charge-trap memcapacitors","authors":"Junsu Yu, Hwiho Hwang, Hyungjin Kim, Woo Young Choi","doi":"10.1186/s40580-026-00575-9","DOIUrl":"10.1186/s40580-026-00575-9","url":null,"abstract":"<div><p>Energy-efficient neural computing is increasingly limited not by computational throughput but by memory access and data movement. In-memory computing architectures offer a promising solution by collocating storage and computation, yet their practical realization remains constrained by static power dissipation, thermal challenges, and limited integration density in existing memory devices. Here, we demonstrate a vertically stackable charge-trap-flash (CTF)-based memcapacitor array that combines the high integration density of 3D NAND technology with the charge-domain computation. A continuous in-situ-doped N<sup>+</sup> bottom readout electrode is introduced beneath the lightly doped active layer, decoupling lateral readout resistance from depletion-based capacitance modulation and alleviating the resistance–depletion trade-off of the previous architecture. The fabricated 24 × 48 memcapacitor array operates through transient charge displacement and demonstrates highly uniform array-level characteristics (σ/µ < 0.37%), reliable 16-state closed-loop weight programming, and linear charge-domain VMM with an error below 0.227%. When scaled toward modern 3D NAND dimensions, a verification–inference electrostatic mismatch introduces systematic VMM error, which is reduced by up to 84.6% through an intercell trapped-charge scheme in TCAD simulations. Geometric scaling based on the <i>CV</i><sup>2</sup> relation projects femtojoule-level intrinsic cell read energy at scaled dimensions. A hybrid hardware–software spiking neural network evaluation, in which only the final 24 × 10 layer is mapped to measured arrays, achieves 88.01% CIFAR-10 accuracy compared with 88.17% in software. These results establish vertically stackable memcapacitors as a scalable charge-domain computing platform combining reliable array-level operation with high-density vertical integration.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":712,"journal":{"name":"Nano Convergence","volume":"13 1","pages":""},"PeriodicalIF":13.8,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1186/s40580-026-00575-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148917699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-08-31DOI: 10.1186/s40580-026-00572-y
Sang Hun Lee, Tae Hyeon Kim, Jae Wook Song, Na Yeon Kim, Ye-Jin Jo, Tae-Hyung Kim, Bong-Gun Ju, Bong Geun Chung
Wearable wound dressing is a promising approach for treating wound infection and promoting healing process. However, conventional wearable dressing methods have the drawback of poor long-term stability due to water loss. Herein, a novel hydrogel patch is fabricated by incorporating phosphoric acid and glycerol to enhance the water retention capacity. This hydrogel exhibits an excellent water retention rate of 97.7% after 7 days under ambient conditions (25 ℃, 50% relative humidity). Consequently, the stretchability, adhesion, and electrical conductivity required for wearable dressing are stably maintained. The antibiotic loaded into the dressing is effectively delivered via iontophoresis, which is demonstrated through in vitro drug release experiments, ex vivo porcine skin deposition tests, and in vivo mouse models. Therefore, this multi-functional hydrogel patch can be a promising strategy for developing an efficient drug delivery and wound healing platform using anti-drying conductive hydrogels and iontophoresis-based transdermal drug delivery.
{"title":"Dehydration-resistant multi-functional hydrogel-based iontophoretic drug delivery for wound healing","authors":"Sang Hun Lee, Tae Hyeon Kim, Jae Wook Song, Na Yeon Kim, Ye-Jin Jo, Tae-Hyung Kim, Bong-Gun Ju, Bong Geun Chung","doi":"10.1186/s40580-026-00572-y","DOIUrl":"10.1186/s40580-026-00572-y","url":null,"abstract":"<div><p>Wearable wound dressing is a promising approach for treating wound infection and promoting healing process. However, conventional wearable dressing methods have the drawback of poor long-term stability due to water loss. Herein, a novel hydrogel patch is fabricated by incorporating phosphoric acid and glycerol to enhance the water retention capacity. This hydrogel exhibits an excellent water retention rate of 97.7% after 7 days under ambient conditions (25 ℃, 50% relative humidity). Consequently, the stretchability, adhesion, and electrical conductivity required for wearable dressing are stably maintained. The antibiotic loaded into the dressing is effectively delivered via iontophoresis, which is demonstrated through in vitro drug release experiments, ex vivo porcine skin deposition tests, and in vivo mouse models. Therefore, this multi-functional hydrogel patch can be a promising strategy for developing an efficient drug delivery and wound healing platform using anti-drying conductive hydrogels and iontophoresis-based transdermal drug delivery.</p><h3>Graphical abstract</h3><div><figure><div><div><picture><source><img></source></picture></div></div></figure></div></div>","PeriodicalId":712,"journal":{"name":"Nano Convergence","volume":"13 1","pages":""},"PeriodicalIF":13.8,"publicationDate":"2026-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13530066/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148863363","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号