Introducing MagBERT:A language model for magnesium textual data mining and analysis

Magnesium(Mg)based materials hold immense potential for various applications due to their lightweight and high strength-to-weight ratio.However,to fully harness the potential of Mg alloys,structured analytics are essential to gain valuable insights from centuries of accumulated knowledge.Efficient information extraction from the vast corpus of scientific literature is crucial for this purpose.In this work,we introduce MagBERT,a BERT-based language model specifically trained for Mg-based materials.Utilizing a dataset of approximately 370,000 abstracts focused on Mg and its alloys,MagBERT is designed to understand the intricate details and specialized terminology of this domain.Through rigorous evaluation,we demonstrate the effectiveness of MagBERT for information extraction using a fine-tuned named entity recognition(NER)model,named MagNER.This NER model can extract mechanical,microstructural,and processing properties related to Mg alloys.For instance,we have created an Mg alloy dataset that includes properties such as ductility,yield strength,and ultimate tensile strength(UTS),along with standard alloy names.The introduction of MagBERT is a novel advancement in the development of Mg-specific language models,marking a significant milestone in the discovery of Mg alloys and textual information extraction.By making the pre-trained weights of MagBERT publicly accessible,we aim to accelerate research and innovation in the field of Mg-based materials through efficient information extraction and knowledge discovery.

Trifunctional robust electrocatalysts based on 3D Fe/N-doped carbon nanocubes encapsulating Co4N nanoparticles for efficient battery-powered water electrolyzers

Herein,we have designed a highly active and robust trifunctional electrocatalyst derived from Prussian blue analogs,where Co_(4)N nanoparticles are encapsulated by Fe embedded in N-doped carbon nanocubes to synthesize hierarchically structured Co_(4)N@Fe/N-C for rechargeable zinc-air batteries and overall water-splitting electrolyzers.As confirmed by theoretical and experimental results,the high intrinsic oxygen reduction reaction,oxygen evolution reaction,and hydrogen evolution reaction activities of Co_(4)N@Fe/N-C were attributed to the formation of the heterointerface and the modulated local electronic structure.Moreover,Co_(4)N@Fe/N-C induced improvement in these trifunctional electrocatalytic activities owing to the hierarchical hollow nanocube structure,uniform distribution of Co_(4)N,and conductive encapsulation by Fe/N-C.Thus,the rechargeable zinc-air battery with Co_(4)N@Fe/N-C delivers a high specific capacity of 789.9 mAh g^(-1) and stable voltage profiles over 500 cycles.Furthermore,the overall water electrolyzer with Co_(4)N@Fe/N-C achieved better durability and rate performance than that with the Pt/C and IrO2 catalysts,delivering a high Faradaic efficiency of 96.4%.Along with the great potential of the integrated water electrolyzer powered by a zinc-air battery for practical applications,therefore,the mechanistic understanding and active site identification provide valuable insights into the rational design of advanced multifunctional electrocatalysts for energy storage and conversion.

Predictive modeling of critical temperatures in magnesium compounds using transfer learning

This study presents a transfer learning approach for discovering potential Mg-based superconductors utilizing a comprehensive target dataset.Initially,a large source dataset(Bandgap dataset)comprising approximately∼75k compounds is utilized for pretraining,followed by fine-tuning with a smaller Critical Temperature(T_(c))dataset containing∼300 compounds.Comparatively,there is a significant improvement in the performance of the transfer learning model over the traditional deep learning(DL)model in predicting Tc.Subsequently,the transfer learning model is applied to predict the properties of approximately 150k compounds.Predictions are validated computationally using density functional theory(DFT)calculations based on lattice dynamics-related theory.Moreover,to demonstrate the extended predictive capability of the transfer learning model for new materials,a pool of virtual compounds derived from prototype crystal structures from the Materials Project(MP)database is generated.T_(c) predictions are obtained for∼3600 virtual compounds,which underwent screening for electroneutrality and thermodynamic stability.An Extra Trees-based model is trained to utilize E_(hull)values to obtain thermodynamically stable materials,employing a dataset containing Ehull values for approximately 150k materials for training.Materials with Ehull values exceeding 5 meV/atom were filtered out,resulting in a refined list of potential Mg-based superconductors.This study showcases the effectiveness of transfer learning in predicting superconducting properties and highlights its potential for accelerating the discovery of Mg-based materials in the field of superconductivity.

Investigation of Domain Wall Velocity to External Fields for High-Quality Waveguide-Type Periodically Poled LiNbO_(3) Fabrication

The Ti-doped waveguide-type periodically poled LiNbO_(3)(PPLN)were fabricated and the dependence of domain wall velocity on an external field applied for domain inversion was investigated.The whole polarization reversal process was computer-controlled to regulate domain wall expansion at a feedback time shorter than 5μs.The coercive voltage and several values of excess voltage were applied on 500μm-thick wafers serially connected to a 1-MOhm external resistor which had an effect of the poling current reduction,i.e.the deceleration of domain wall expansion.The domain wall velocity is sensitive to the poling voltage,precisely speaking,to the excess voltage.The domain wall velocities were 28.70,16.02 and 5.75μm·s^(-1)under poling field of 23.5,22.0 and 21.0 kV·mm^(-1),respectively.Moreover,average duty cycle of PPLN is about 49.93%.

In-situ formation of MOF derived mesoporous Co3N/amorphous N-doped carbon nanocubes as an efficient electrocatalytic oxygen evolution reaction

The suitable materials,metal nitrides,are a promising class of electrocatalyst materials for a highly efficient oxygen evolution reaction (OER) because they exhibit superior intrinsic conductivity and have higher sustainability than oxide-based materials.To our knowledge,for the first time,we report a designable synthesis of three-dimensional (3D) and mesoporous Co3N@ amorphous N-doped carbon (AN-C) nanocubes (NCs) with well-controlled open-framework structures via monodispersed Co3[Co(CN)6]2 Prussian blue analogue (PBA) NC precursors using in situ nitridation and calcination processes.Co3N@AN-C NCs (2 h) demonstrate better OER activity with a remarkably low Tafel plot (69.6 mV-dec-1),low overpotential of 280 mV at a current density of 10 mA-crrf2.Additionally,excellent cycling stability in alkaline electrolytes was exhibited without morphological changes and voltage elevations,superior to most reported hierarchical structures of transition-metal nitride particles.The presented strategy for synergy effects of metal-organic frameworks (MOFs)-derived transition-metal nitrides-carbon hybrid nanostructures provides prospects for developing high-performance and advanced electrocatalyst materials.