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 T_c. 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 E_hull values for approximately 150k materials for training. Materials with E_hull 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.

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∙cm^-2. 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.