Pub Date : 2026-01-20DOI: 10.1016/j.mtphys.2026.102027
Fan Ye , Fei Liang , Jian-Wei Zhong , Guan-Ling Li , Xin-Gui Tang
Due to the traditional von Neumann architecture, computing power is regarded as one of the key constraints in the era of artificial intelligence (AI). The use of non-volatile ferroelectric memristors to simulate the characteristics of biological synapses has been validated as a viable approach for mimicking the human brain's execution of large-scale complex computational tasks. The neuromorphic computing potential of various emerging materials and devices has attracted widespread research interest. Among these, BaTiO3, a ferroelectric perovskite, is an ideal candidate due to its distinct advantages and exceptional performance. In this paper, the recent research progress on BaTiO3-based ferroelectric memristors is reviewed, including ferroelectric tunnel junctions (FTJs) and ferroelectric diodes (FDs). And the optimization schemes for synaptic behavior of the corresponding devices are discussed. Finally, the efficient application of two neural network architectures based on artificial neural networks (ANNs) and convolutional neural networks (CNNs) is introduced, illuminating the development prospects for next-generation BaTiO3-based memristive artificial intelligence.
{"title":"Research progress of BaTiO3-based ferroelectric memristors for artificial synapse and neuromorphic computing","authors":"Fan Ye , Fei Liang , Jian-Wei Zhong , Guan-Ling Li , Xin-Gui Tang","doi":"10.1016/j.mtphys.2026.102027","DOIUrl":"10.1016/j.mtphys.2026.102027","url":null,"abstract":"<div><div>Due to the traditional von Neumann architecture, computing power is regarded as one of the key constraints in the era of artificial intelligence (AI). The use of non-volatile ferroelectric memristors to simulate the characteristics of biological synapses has been validated as a viable approach for mimicking the human brain's execution of large-scale complex computational tasks. The neuromorphic computing potential of various emerging materials and devices has attracted widespread research interest. Among these, BaTiO<sub>3</sub>, a ferroelectric perovskite, is an ideal candidate due to its distinct advantages and exceptional performance. In this paper, the recent research progress on BaTiO<sub>3</sub>-based ferroelectric memristors is reviewed, including ferroelectric tunnel junctions (FTJs) and ferroelectric diodes (FDs). And the optimization schemes for synaptic behavior of the corresponding devices are discussed. Finally, the efficient application of two neural network architectures based on artificial neural networks (ANNs) and convolutional neural networks (CNNs) is introduced, illuminating the development prospects for next-generation BaTiO<sub>3</sub>-based memristive artificial intelligence.</div></div>","PeriodicalId":18253,"journal":{"name":"Materials Today Physics","volume":"61 ","pages":"Article 102027"},"PeriodicalIF":9.7,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-20DOI: 10.1016/j.mtphys.2026.102026
T. Crozier-Bioud , S. Jouen , S. Rolere , V. Nachbaur , M. Ollivier , L. Favergeon , S. Luca
The Powder Injection Molding (PIM) process offers the possibility to obtain dense near-net shape NdFeB parts with magnetic properties similar to the conventional powder metallurgy counterparts. Understanding the structural evolution of the magnets with various carbon contamination coming from the organic binders will offer the possibility to improve the process. Here, the evolution of the structure of the NdFeB magnets with various carbon contamination within the PIM process was investigated. It was found that the carbon contaminates the Nd-rich phases forming a NdOxCy phase. For carbon concentrations higher than 5300 ppm wt., a tetragonal neodymium carbide phase was formed at the expense of the metallic Nd-dhcp phase. Moreover, 23 % of boron atoms are substituted in the main Nd2Fe14B magnetic phase, forming a Nd2Fe14B0.77C0.23 phase.
{"title":"Exploring the structural evolution of the NdFeB magnets with various carbon contamination in the PIM process","authors":"T. Crozier-Bioud , S. Jouen , S. Rolere , V. Nachbaur , M. Ollivier , L. Favergeon , S. Luca","doi":"10.1016/j.mtphys.2026.102026","DOIUrl":"10.1016/j.mtphys.2026.102026","url":null,"abstract":"<div><div>The Powder Injection Molding (PIM) process offers the possibility to obtain dense near-net shape NdFeB parts with magnetic properties similar to the conventional powder metallurgy counterparts. Understanding the structural evolution of the magnets with various carbon contamination coming from the organic binders will offer the possibility to improve the process. Here, the evolution of the structure of the NdFeB magnets with various carbon contamination within the PIM process was investigated. It was found that the carbon contaminates the Nd-rich phases forming a NdO<sub>x</sub>C<sub>y</sub> phase. For carbon concentrations higher than 5300 ppm wt., a tetragonal neodymium carbide phase was formed at the expense of the metallic Nd-<em>dhcp</em> phase. Moreover, 23 % of boron atoms are substituted in the main Nd<sub>2</sub>Fe<sub>14</sub>B magnetic phase, forming a Nd<sub>2</sub>Fe<sub>14</sub>B<sub>0.77</sub>C<sub>0.23</sub> phase.</div></div>","PeriodicalId":18253,"journal":{"name":"Materials Today Physics","volume":"61 ","pages":"Article 102026"},"PeriodicalIF":9.7,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014822","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-19DOI: 10.1016/j.mtphys.2026.102013
Tao Xiong , Chenhao Ren , Hailong He, Chunping Niu, Youqun Li, Feng Jiang, Yi Wu, Mingzhe Rong
N-type bismuth telluride-based materials are more sensitive to texturing, which often leads to limited electrical transport properties in samples fabricated via conventional hot-press processing routes. Although low thermal conductivity can be achieved, the insufficient electrical properties significantly restrict the further enhancement of the module's output power. Hence, we have innovatively introduced a rotary melting technique that utilizes a dynamic centrifugal field for microstructural control, thereby achieving co-optimization of both electrical conductivity and Seebeck coefficient in n-type Bi2Te3-based materials (PF = 40 μWcm−1K−2). In addition, a substantial reduction in lattice thermal conductivity was successfully achieved, as full-spectrum phonon scattering centers were constructed by high densities of dislocations, twin boundaries, and linear nanoscale secondary phases. As a result, the peak and average values of zT for the Cu0.002Bi2Te2.7Se0.3 + 4 wt% Te sample were measured to be 1.2 and 1.13, respectively. A single-leg thermoelectric power generation module assembled with the optimized n-type material demonstrated a conversion efficiency of 5.1 % at ΔT = 200 K. Moreover, the TE cooling module, constructed using the n-type material in combination with a self-fabricated p-type material, achieved a temperature difference of 82 K at a hot-side temperature of 350 K. Both properties metrics surpass those of commercially available TE modules.
{"title":"Advanced thermoelectric cooling performance in BiTeSe alloys through composition tuning under dynamic melting fields","authors":"Tao Xiong , Chenhao Ren , Hailong He, Chunping Niu, Youqun Li, Feng Jiang, Yi Wu, Mingzhe Rong","doi":"10.1016/j.mtphys.2026.102013","DOIUrl":"10.1016/j.mtphys.2026.102013","url":null,"abstract":"<div><div>N-type bismuth telluride-based materials are more sensitive to texturing, which often leads to limited electrical transport properties in samples fabricated via conventional hot-press processing routes. Although low thermal conductivity can be achieved, the insufficient electrical properties significantly restrict the further enhancement of the module's output power. Hence, we have innovatively introduced a rotary melting technique that utilizes a dynamic centrifugal field for microstructural control, thereby achieving co-optimization of both electrical conductivity and Seebeck coefficient in n-type Bi<sub>2</sub>Te<sub>3</sub>-based materials (<em>PF</em> = 40 μWcm<sup>−1</sup>K<sup>−2</sup>). In addition, a substantial reduction in lattice thermal conductivity was successfully achieved, as full-spectrum phonon scattering centers were constructed by high densities of dislocations, twin boundaries, and linear nanoscale secondary phases. As a result, the peak and average values of <em>zT</em> for the Cu<sub>0.002</sub>Bi<sub>2</sub>Te<sub>2.7</sub>Se<sub>0.3</sub> + 4 wt% Te sample were measured to be 1.2 and 1.13, respectively. A single-leg thermoelectric power generation module assembled with the optimized n-type material demonstrated a conversion efficiency of 5.1 % at Δ<em>T</em> = 200 K. Moreover, the TE cooling module, constructed using the n-type material in combination with a self-fabricated p-type material, achieved a temperature difference of 82 K at a hot-side temperature of 350 K. Both properties metrics surpass those of commercially available TE modules.</div></div>","PeriodicalId":18253,"journal":{"name":"Materials Today Physics","volume":"61 ","pages":"Article 102013"},"PeriodicalIF":9.7,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-19DOI: 10.1016/j.mtphys.2026.102028
Sobia Nisar , Ghulam Dastgeer , Muhammad Wajid Zulfiqar , Hammad Ghazanfar , Muneeb Ahmad , Muhammad Rabeel , Aamir Rasheed , Muhammad Imran , Deok-kee Kim
Diabetes mellitus requires accurate and continuous glucose monitoring for early diagnosis, effective disease management, and prevention of severe complications. However, conventional glucose sensors often suffer from limited stability, frequent calibration, and invasive operation, which restrict their long-term and real-time use. In recent years, two-dimensional (2D) materials integrated into electrochemical, optical, and field-effect transistor (FET) platforms have emerged as promising alternatives due to their high surface area, tunable electronic properties, and excellent bio-interfacing capability. To address existing gaps in understanding and comparison, this review presents a systematic framework that categorizes glucose sensors according to both sensing mechanisms (electrochemical, optical, and FET-based) and application formats (invasive, non-invasive, and wearable). Beyond static classification, we analyze recent temporal trends in material selection, device architecture, and sensing performance, highlighting the evolution from graphene-based systems toward transition metal dichalcogenides (TMDCs) and MXene-enabled platforms for flexible and biofluid-compatible sensing. A critical comparison of enzyme immobilization and surface functionalization strategies is also provided to clarify their influence on sensitivity, stability, and reproducibility. Finally, key challenges related to long-term stability, selectivity, and device integration are discussed, and emerging directions are outlined to support the development of reliable, flexible, and high-performance glucose sensors for real-time and personalized healthcare.
{"title":"2D material-based smart sensors for efficient and non-invasive glucose monitoring","authors":"Sobia Nisar , Ghulam Dastgeer , Muhammad Wajid Zulfiqar , Hammad Ghazanfar , Muneeb Ahmad , Muhammad Rabeel , Aamir Rasheed , Muhammad Imran , Deok-kee Kim","doi":"10.1016/j.mtphys.2026.102028","DOIUrl":"10.1016/j.mtphys.2026.102028","url":null,"abstract":"<div><div>Diabetes mellitus requires accurate and continuous glucose monitoring for early diagnosis, effective disease management, and prevention of severe complications. However, conventional glucose sensors often suffer from limited stability, frequent calibration, and invasive operation, which restrict their long-term and real-time use. In recent years, two-dimensional (2D) materials integrated into electrochemical, optical, and field-effect transistor (FET) platforms have emerged as promising alternatives due to their high surface area, tunable electronic properties, and excellent bio-interfacing capability. To address existing gaps in understanding and comparison, this review presents a systematic framework that categorizes glucose sensors according to both sensing mechanisms (electrochemical, optical, and FET-based) and application formats (invasive, non-invasive, and wearable). Beyond static classification, we analyze recent temporal trends in material selection, device architecture, and sensing performance, highlighting the evolution from graphene-based systems toward transition metal dichalcogenides (TMDCs) and MXene-enabled platforms for flexible and biofluid-compatible sensing. A critical comparison of enzyme immobilization and surface functionalization strategies is also provided to clarify their influence on sensitivity, stability, and reproducibility. Finally, key challenges related to long-term stability, selectivity, and device integration are discussed, and emerging directions are outlined to support the development of reliable, flexible, and high-performance glucose sensors for real-time and personalized healthcare.</div></div>","PeriodicalId":18253,"journal":{"name":"Materials Today Physics","volume":"61 ","pages":"Article 102028"},"PeriodicalIF":9.7,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146000685","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-17DOI: 10.1016/j.mtphys.2026.102025
Xianteng Zhou , Chaokun Guo , Zhen Yang , Yuanji Xu , Hongquan Song , De-Ye Lin , Fuyang Tian
The synergistic optimization of ultra-high hardness and low thermal conductivity in high entropy carbides is achieved by adjusting the concentration of ordered carbon vacancies. By using machine-learning interatomic potentials integrated with molecular dynamics simulations and calculations, we elucidate how lattice distortion, carbon vacancies and grain boundaries regulate the mechanical response and thermal transport of (NbTaZr)C. The results reveal that lattice distortion drives anomalous CNb bond rupture, enabling edge dislocation nucleation. Carbon vacancies reduce the critical resolved shear stress(CRSS) and induce localized amorphization, thereby enhancing intrinsic plasticity. The preferential segregation of carbon vacancies at grain boundaries optimizes stress redistribution, mitigating stress concentration while enhancing both yield strength and strain. Carbon vacancies markedly suppress lattice thermal transport capability via increasing vibrational localization and scattering between phonons and defects, whereas the ordering of carbon vacancies partially enhances lattice thermal conductivity through low-frequency phonon delocalization. Pronounced lattice distortion and Anderson localization collectively intensify electron scattering, thereby reducing electronic thermal conductivity, whereas the ordered carbon vacancies facilitate the delocalization of electrons, leading to a modest increase of electronic thermal conductivity. The interfacial thermal conductance(ITC) decreases due to impaired phonon mode matching and strengthened localization. We establish carbon vacancy-mediated strategies for concurrently tuning mechanical and thermal transport in multi-principal carbide ceramics.
{"title":"Vacancy-induced mechanism on deformation and thermal conductivity in medium-entropy carbides with typical grain boundaries","authors":"Xianteng Zhou , Chaokun Guo , Zhen Yang , Yuanji Xu , Hongquan Song , De-Ye Lin , Fuyang Tian","doi":"10.1016/j.mtphys.2026.102025","DOIUrl":"10.1016/j.mtphys.2026.102025","url":null,"abstract":"<div><div>The synergistic optimization of ultra-high hardness and low thermal conductivity in high entropy carbides is achieved by adjusting the concentration of ordered carbon vacancies. By using machine-learning interatomic potentials integrated with molecular dynamics simulations and <span><math><mrow><mi>a</mi><mi>b</mi></mrow></math></span> <span><math><mrow><mi>i</mi><mi>n</mi><mi>i</mi><mi>t</mi><mi>i</mi><mi>o</mi></mrow></math></span> calculations, we elucidate how lattice distortion, carbon vacancies and grain boundaries regulate the mechanical response and thermal transport of (NbTaZr)C. The results reveal that lattice distortion drives anomalous C<img>Nb bond rupture, enabling edge dislocation nucleation. Carbon vacancies reduce the critical resolved shear stress(CRSS) and induce localized amorphization, thereby enhancing intrinsic plasticity. The preferential segregation of carbon vacancies at grain boundaries optimizes stress redistribution, mitigating stress concentration while enhancing both yield strength and strain. Carbon vacancies markedly suppress lattice thermal transport capability via increasing vibrational localization and scattering between phonons and defects, whereas the ordering of carbon vacancies partially enhances lattice thermal conductivity through low-frequency phonon delocalization. Pronounced lattice distortion and Anderson localization collectively intensify electron scattering, thereby reducing electronic thermal conductivity, whereas the ordered carbon vacancies facilitate the delocalization of electrons, leading to a modest increase of electronic thermal conductivity. The interfacial thermal conductance(ITC) decreases due to impaired phonon mode matching and strengthened localization. We establish carbon vacancy-mediated strategies for concurrently tuning mechanical and thermal transport in multi-principal carbide ceramics.</div></div>","PeriodicalId":18253,"journal":{"name":"Materials Today Physics","volume":"61 ","pages":"Article 102025"},"PeriodicalIF":9.7,"publicationDate":"2026-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145993398","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-15DOI: 10.1016/j.mtphys.2026.102023
Kun Zhang , Zixue Luo , Jinlin Song , Qiang Cheng
Enhanced near-field radiative heat transfer (NFRHT), mediated by the waveguide modes, holds great promise for advanced thermal management and energy conversion. While extensive researches have focused on the infinite structures, the role of finite-size effect remains insufficiently unexplored. Here, we investigate the NFRHT between two nanoparticles above a finite-size rectangular waveguide, revealing a profound influence of the width of the waveguide on the heat transfer. Based on the fluctuating-surface current formulation and the boundary element method, we demonstrate that the enhancement factor exhibits a non-monotonic dependence on the width, peaking at a specific value, where the heat conductance can be enhanced by over two orders of magnitude compared to the vacuum case. This optimal enhancement is attributed to the strong coupling between the localized surface resonances (LSR) of the nanoparticles and the degenerate corner- and edge-modes (ss and sa modes) of the waveguide. Furthermore, we show that the shape of the nanoparticle is critical, with the sharp features (e.g., cubes, pyramids) enable a superior enhancement over the spherical nanoparticles, as their LSR frequencies align with the waveguide modes, facilitating a stronger coupling. The distances between nanoparticles and between nanoparticles and the waveguide are also key tuning parameters, with an identified inflection point in enhancement at d ≈ 2 μm linked to the finite propagation lengths of the waveguide modes. Our work provides fundamental insights into the manipulation of NFRHT via finite-size effects and mode engineering, with implications for the design of nanoscale thermal devices.
{"title":"Giant enhancement of near-field radiative heat transfer enabled by a finite-size waveguide","authors":"Kun Zhang , Zixue Luo , Jinlin Song , Qiang Cheng","doi":"10.1016/j.mtphys.2026.102023","DOIUrl":"10.1016/j.mtphys.2026.102023","url":null,"abstract":"<div><div>Enhanced near-field radiative heat transfer (NFRHT), mediated by the waveguide modes, holds great promise for advanced thermal management and energy conversion. While extensive researches have focused on the infinite structures, the role of finite-size effect remains insufficiently unexplored. Here, we investigate the NFRHT between two nanoparticles above a finite-size rectangular waveguide, revealing a profound influence of the width of the waveguide on the heat transfer. Based on the fluctuating-surface current formulation and the boundary element method, we demonstrate that the enhancement factor exhibits a non-monotonic dependence on the width, peaking at a specific value, where the heat conductance can be enhanced by over two orders of magnitude compared to the vacuum case. This optimal enhancement is attributed to the strong coupling between the localized surface resonances (LSR) of the nanoparticles and the degenerate corner- and edge-modes (<em>ss</em> and <em>sa</em> modes) of the waveguide. Furthermore, we show that the shape of the nanoparticle is critical, with the sharp features (e.g., cubes, pyramids) enable a superior enhancement over the spherical nanoparticles, as their LSR frequencies align with the waveguide modes, facilitating a stronger coupling. The distances between nanoparticles and between nanoparticles and the waveguide are also key tuning parameters, with an identified inflection point in enhancement at <em>d</em> ≈ 2 μm linked to the finite propagation lengths of the waveguide modes. Our work provides fundamental insights into the manipulation of NFRHT via finite-size effects and mode engineering, with implications for the design of nanoscale thermal devices.</div></div>","PeriodicalId":18253,"journal":{"name":"Materials Today Physics","volume":"61 ","pages":"Article 102023"},"PeriodicalIF":9.7,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145995368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-15DOI: 10.1016/j.mtphys.2026.102022
Umair Haider , Gul Rahman , Imran Shakir , M.S. Al-Buriahi , Norah Alomayrah , Imen Kebaili
We implement a reliable and generalizable multistep workflow that leverages supervised machine learning algorithms to construct accurate, data-driven models for predicting the work function (WF) of 4000 MMXT-type MXenes. Among the tested models, the random forest regressor demonstrates excellent performance, achieving a mean absolute error of 0.03 eV on the training set and 0.09 eV on the test set. Remarkably, through recursive feature elimination and hyperparameter tuning, the model attains even higher accuracy with only ten key descriptors, reducing the test MAE to 0.02 eV. The optimized model is employed to predict the properties of 150 unexplored MXenes for applications in catalysis (86 MXenes), electronics (38 MXenes), and energy storage (26 MXenes). The low-WF energy-storage candidates are dominated by nitride- and halide-terminated species, often incorporating early transition metals or rare-earth elements such as Y, Sc, and Hf. The intermediate-WF window contains compositions with balanced metallic and semiconducting features, such as , , and . The high-WF catalytic group is characterized by carbide-, oxide-, and chalcogen-terminated MXenes enriched in Mo, Zr, Hf, Ti, and Cr, e.g., , , and . These predictions provide a robust starting point for experimental validation, supporting both multifunctional design and application-specific optimization of MXenes.
{"title":"Accelerated discovery of MM′XT2 MXenes for catalysis, electronics, and energy storage using supervised machine learning","authors":"Umair Haider , Gul Rahman , Imran Shakir , M.S. Al-Buriahi , Norah Alomayrah , Imen Kebaili","doi":"10.1016/j.mtphys.2026.102022","DOIUrl":"10.1016/j.mtphys.2026.102022","url":null,"abstract":"<div><div>We implement a reliable and generalizable multistep workflow that leverages supervised machine learning algorithms to construct accurate, data-driven models for predicting the work function (WF) of 4000 MM<span><math><msup><mrow></mrow><mrow><mo>′</mo></mrow></msup></math></span>XT<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span>-type MXenes. Among the tested models, the <em>random forest regressor</em> demonstrates excellent performance, achieving a mean absolute error of 0.03 eV on the training set and 0.09 eV on the test set. Remarkably, through recursive feature elimination and hyperparameter tuning, the model attains even higher accuracy with only ten key descriptors, reducing the test MAE to 0.02 eV. The optimized model is employed to predict the properties of 150 unexplored MXenes for applications in catalysis (86 MXenes), electronics (38 MXenes), and energy storage (26 MXenes). The low-WF energy-storage candidates are dominated by nitride- and halide-terminated species, often incorporating early transition metals or rare-earth elements such as Y, Sc, and Hf. The intermediate-WF window contains compositions with balanced metallic and semiconducting features, such as <span><math><msub><mrow><mi>TiZrNCl</mi></mrow><mrow><mn>2</mn></mrow></msub></math></span>, <span><math><msub><mrow><mi>VMoNCl</mi></mrow><mrow><mn>2</mn></mrow></msub></math></span>, and <span><math><msub><mrow><mi>TiScCF</mi></mrow><mrow><mn>2</mn></mrow></msub></math></span>. The high-WF catalytic group is characterized by carbide-, oxide-, and chalcogen-terminated MXenes enriched in Mo, Zr, Hf, Ti, and Cr, e.g., <span><math><msub><mrow><mi>MnNbCS</mi></mrow><mrow><mn>2</mn></mrow></msub></math></span>, <span><math><msub><mrow><mi>CrTiCCl</mi></mrow><mrow><mn>2</mn></mrow></msub></math></span>, and <span><math><msub><mrow><mi>ZrMoCCl</mi></mrow><mrow><mn>2</mn></mrow></msub></math></span>. These predictions provide a robust starting point for experimental validation, supporting both multifunctional design and application-specific optimization of MXenes.</div></div>","PeriodicalId":18253,"journal":{"name":"Materials Today Physics","volume":"61 ","pages":"Article 102022"},"PeriodicalIF":9.7,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145993423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-14DOI: 10.1016/j.mtphys.2026.102021
Yaran Shi , Ze Yang , Xiaohui Li , Zhouzhou Wang , Xue Dong , Wenzhu Cao , Chenchen Wei , Zhixuan Huang , Zijun Sun , Yan Jiang , Ying Yu
Sn-based aqueous acidic batteries (SnAABs) as a new type of non-toxicity, acid-resistant, and ease of recycling batteries, face the challenges of inhomogeneous Sn deposition and excessive hydrogen evolution reaction (HER) in acidic electrolytes, leading to its fast failure. Herein, current density has been identified as a key parameter for tuning Sn2+ nucleation and mass-transfer processes simultaneously on Sn anode. Both low current densities (LCD) and high current densities (HCD) resulted in poor plating/stripping stability due to inhomogeneous deposition and excessive HER. The optimal stability was achieved at moderate current densities (MCD), which balanced the nucleation and mass-transfer processes. As such, the Sn symmetrical cell exhibited stable cycling for 1000 h with a voltage polarization of 47 mV at the MCD, which remarkably surpassed the performances under the LCD (110 h) and the HCD (68 h). This work provides fundamental and practical insights for designing highly stable metal anodes.
{"title":"Balancing nucleation and mass-transfer processes through regulating current density for stable aqueous Sn anode batteries","authors":"Yaran Shi , Ze Yang , Xiaohui Li , Zhouzhou Wang , Xue Dong , Wenzhu Cao , Chenchen Wei , Zhixuan Huang , Zijun Sun , Yan Jiang , Ying Yu","doi":"10.1016/j.mtphys.2026.102021","DOIUrl":"10.1016/j.mtphys.2026.102021","url":null,"abstract":"<div><div>Sn-based aqueous acidic batteries (SnAABs) as a new type of non-toxicity, acid-resistant, and ease of recycling batteries, face the challenges of inhomogeneous Sn deposition and excessive hydrogen evolution reaction (HER) in acidic electrolytes, leading to its fast failure. Herein, current density has been identified as a key parameter for tuning Sn<sup>2+</sup> nucleation and mass-transfer processes simultaneously on Sn anode. Both low current densities (LCD) and high current densities (HCD) resulted in poor plating/stripping stability due to inhomogeneous deposition and excessive HER. The optimal stability was achieved at moderate current densities (MCD), which balanced the nucleation and mass-transfer processes. As such, the Sn symmetrical cell exhibited stable cycling for 1000 h with a voltage polarization of 47 mV at the MCD, which remarkably surpassed the performances under the LCD (110 h) and the HCD (68 h). This work provides fundamental and practical insights for designing highly stable metal anodes.</div></div>","PeriodicalId":18253,"journal":{"name":"Materials Today Physics","volume":"61 ","pages":"Article 102021"},"PeriodicalIF":9.7,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145962499","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-13DOI: 10.1016/j.mtphys.2026.102018
Xinhang Guo , Haobo Yang , Yuhan Yao , Te-Huan Liu , Ronggui Yang , Xin Qian
Coupled transport of light and heat in multilayered nanostructures is pivotal for accurate pump-probe thermal characterizations. Conventional thermoreflectance measurements involve a transducer film deposited onto the sample surface, and the optical heating is typically treated as a surface heat flux. However, sensitive in-plane thermal conductivity measurements and nonintrusive characterizations require optically thin transducers or even the removal of transducers. Despite the recent advances in transducerless thermoreflectance techniques, the modeling of light absorption and heat generation involves simplifications that limit their application to a few specific cases, such as bulk absorptive materials or absorptive thin films on transparent substrates. In this work, we derive an optothermal transfer matrix formalism that enables unified analysis of propagation and absorption of light, generation and conduction of heat, as well as the multilayer thermoreflectance effects. The thermal conductivity could be severely underestimated if the surface heating model is naively used when the light can penetrate across the top layer. Based on the coupled optothermal transfer matrix model, thermal conductivity can be robustly measured for the three following cases: (a) bulk samples coated with optically thin metal transducer films in which the surface heat flux boundary condition breaks down; (b) absorptive semiconductor film on transparent substrates where the thermoreflectance cannot be regarded as a surface effect; and (c) absorptive semiconductor film on absorptive substrates with multilayer absorption and thermoreflectance effects. This work enables the use of ultrathin transducer layers and convenient modeling of coupled light propagation and heat conduction in multilayered structures and devices.
{"title":"Light propagation and heat conduction across multilayered structures in pump-probe thermoreflectance","authors":"Xinhang Guo , Haobo Yang , Yuhan Yao , Te-Huan Liu , Ronggui Yang , Xin Qian","doi":"10.1016/j.mtphys.2026.102018","DOIUrl":"10.1016/j.mtphys.2026.102018","url":null,"abstract":"<div><div>Coupled transport of light and heat in multilayered nanostructures is pivotal for accurate pump-probe thermal characterizations. Conventional thermoreflectance measurements involve a transducer film deposited onto the sample surface, and the optical heating is typically treated as a surface heat flux. However, sensitive in-plane thermal conductivity measurements and nonintrusive characterizations require optically thin transducers or even the removal of transducers. Despite the recent advances in transducerless thermoreflectance techniques, the modeling of light absorption and heat generation involves simplifications that limit their application to a few specific cases, such as bulk absorptive materials or absorptive thin films on transparent substrates. In this work, we derive an optothermal transfer matrix formalism that enables unified analysis of propagation and absorption of light, generation and conduction of heat, as well as the multilayer thermoreflectance effects. The thermal conductivity could be severely underestimated if the surface heating model is naively used when the light can penetrate across the top layer. Based on the coupled optothermal transfer matrix model, thermal conductivity can be robustly measured for the three following cases: (a) bulk samples coated with optically thin metal transducer films in which the surface heat flux boundary condition breaks down; (b) absorptive semiconductor film on transparent substrates where the thermoreflectance cannot be regarded as a surface effect; and (c) absorptive semiconductor film on absorptive substrates with multilayer absorption and thermoreflectance effects. This work enables the use of ultrathin transducer layers and convenient modeling of coupled light propagation and heat conduction in multilayered structures and devices.</div></div>","PeriodicalId":18253,"journal":{"name":"Materials Today Physics","volume":"61 ","pages":"Article 102018"},"PeriodicalIF":9.7,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145962506","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-13DOI: 10.1016/j.mtphys.2026.102019
Zhi-Yuan Qiu , Zheng-Tang Liu , Qi-Jun Liu
Exploring new types of high-temperature superconductors has always been a central topic in condensed matter physics and materials science. This study breaks through the traditional element substitution strategy and takes boron-carbon compound MB4C4 as the parent structure, innovatively proposing a new material design paradigm of "structural isomer engineering (SIE)". Through first-principles calculations, it systematically studies the structural isomers MC4B4 formed after the positions of B and C atoms are swapped. We conducted a high-throughput screening of 59 compounds and found that only SiC4B4 and BeC4B4 can maintain kinetic stability at normal pressure. SiC4B4 is an electron semiconductor similar to diamond, while BeC4B4 is predicted to be a new type of high-temperature superconductor with a superconducting transition temperature (TC) of up to 87.52 K. Particularly noteworthy is that its TC exhibits remarkable robustness compared to the original BeB4C4 (∼76–83 K). The analysis of the electronic structure reveals that the difference in physical properties is attributed to the degree of electron filling in the framework. The in-depth microscopic mechanism study indicates that the high-temperature superconductivity of BeC4B4 stems from an unprecedented "dynamic donor-skeleton framework coupling" mechanism. The strong electron-phonon coupling (EPC) ( = 1.76) is mainly contributed by the high-frequency collective vibration mode driven by the light Be2+ ions, which efficiently couples with the delocalized electrons of the electron-deficient B-C framework. Based on this, we constructed a two-dimensional design descriptor centered on "donor-skeleton coupling degree " and "electron filling degree (Ntot)", successfully explaining the superconducting trend of MB4C4 and its isomer systems, and providing a universal blueprint for systematically searching for new high-temperature superconductors driven by dynamic ionic coupling in three-dimensional rigid covalent frameworks.
{"title":"Structural isomer engineering to create novel high TC: Predicting the dynamic donor-skeleton coupling mechanism of superconducting BeC4B4","authors":"Zhi-Yuan Qiu , Zheng-Tang Liu , Qi-Jun Liu","doi":"10.1016/j.mtphys.2026.102019","DOIUrl":"10.1016/j.mtphys.2026.102019","url":null,"abstract":"<div><div>Exploring new types of high-temperature superconductors has always been a central topic in condensed matter physics and materials science. This study breaks through the traditional element substitution strategy and takes boron-carbon compound MB<sub>4</sub>C<sub>4</sub> as the parent structure, innovatively proposing a new material design paradigm of \"structural isomer engineering (SIE)\". Through first-principles calculations, it systematically studies the structural isomers MC<sub>4</sub>B<sub>4</sub> formed after the positions of B and C atoms are swapped. We conducted a high-throughput screening of 59 compounds and found that only SiC<sub>4</sub>B<sub>4</sub> and BeC<sub>4</sub>B<sub>4</sub> can maintain kinetic stability at normal pressure. SiC<sub>4</sub>B<sub>4</sub> is an electron semiconductor similar to diamond, while BeC<sub>4</sub>B<sub>4</sub> is predicted to be a new type of high-temperature superconductor with a superconducting transition temperature (<em>T</em><sub><em>C</em></sub>) of up to 87.52 K. Particularly noteworthy is that its <em>T</em><sub><em>C</em></sub> exhibits remarkable robustness compared to the original BeB<sub>4</sub>C<sub>4</sub> (∼76–83 K). The analysis of the electronic structure reveals that the difference in physical properties is attributed to the degree of electron filling in the framework. The in-depth microscopic mechanism study indicates that the high-temperature superconductivity of BeC<sub>4</sub>B<sub>4</sub> stems from an unprecedented \"dynamic donor-skeleton framework coupling\" mechanism. The strong electron-phonon coupling (EPC) (<span><math><mrow><mi>λ</mi></mrow></math></span> = 1.76) is mainly contributed by the high-frequency collective vibration mode driven by the light Be<sup>2+</sup> ions, which efficiently couples with the delocalized electrons of the electron-deficient B-C framework. Based on this, we constructed a two-dimensional design descriptor centered on \"donor-skeleton coupling degree <span><math><mrow><mo>(</mo><msub><mi>Γ</mi><mi>X</mi></msub><mo>)</mo></mrow></math></span>\" and \"electron filling degree (<em>N</em><sub><em>tot</em></sub>)\", successfully explaining the superconducting trend of MB<sub>4</sub>C<sub>4</sub> and its isomer systems, and providing a universal blueprint for systematically searching for new high-temperature superconductors driven by dynamic ionic coupling in three-dimensional rigid covalent frameworks.</div></div>","PeriodicalId":18253,"journal":{"name":"Materials Today Physics","volume":"61 ","pages":"Article 102019"},"PeriodicalIF":9.7,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145956514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","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