Transition-Metal Interlink Neural Network: Machine Learning of 2D Metal-Organic Frameworks with High Magnetic Anisotropy

Machine Learning-Assisted Exploration of Intrinsically Spin-Ordered Two-Dimensional (2D) Nanomagnets

The existence of spontaneous spin-ordering in two-dimensional (2D) nanomagnets holds significant importance due to their several unique and promising properties that distinguish them from conventional 2D materials. In recent times, machine learning (ML) has emerged as a powerful tool for effectively exploring and identifying the optimal 2D materials for specific applications or properties within a limited span of time. Here, we have introduced ML-accelerated approaches to specifically estimate the properties, such as the HSE bandgap and magnetoanisotropic energy (MAE) of 2D magnetic materials. Supervised ML algorithms were employed to derive the descriptors that are capable of predicting the properties of intrinsic 2D magnetic materials. Furthermore, the feature selection score is also calculated to reduce the feature space complexity and improve the model accuracy. The input features were obtained from the C2DB database, and models were constructed using linear regression, Lasso, decision tree, random forest, XG Boost, and support vector machine algorithms. The random forest model predicted the HSE band gaps with an unprecedented low root-mean-square error (RMSE) of 0.22 eV, while the linear regression gives the best fit with RMSEs of 0.25 and 0.22 meV for the MAE(x) and MAE(y), respectively. Therefore, the integration of interpretable analytical models with density functional theory offers a swift and reliable approach for uncovering the properties of intrinsic 2D magnetic materials. This collaborative methodology not only ensures speed in analysis but also enriches the material space.

When Machine Learning Meets 2D Materials: A Review

The availability of an ever-expanding portfolio of 2D materials with rich internal degrees of freedom (spin, excitonic, valley, sublattice, and layer pseudospin) together with the unique ability to tailor heterostructures made layer by layer in a precisely chosen stacking sequence and relative crystallographic alignments, offers an unprecedented platform for realizing materials by design. However, the breadth of multi-dimensional parameter space and massive data sets involved is emblematic of complex, resource-intensive experimentation, which not only challenges the current state of the art but also renders exhaustive sampling untenable. To this end, machine learning, a very powerful data-driven approach and subset of artificial intelligence, is a potential game-changer, enabling a cheaper - yet more efficient - alternative to traditional computational strategies. It is also a new paradigm for autonomous experimentation for accelerated discovery and machine-assisted design of functional 2D materials and heterostructures. Here, the study reviews the recent progress and challenges of such endeavors, and highlight various emerging opportunities in this frontier research area.

Rational Design of Full-Color Fluorescent C3N Quantum Dots

Carbon-based quantum dots (QDs) exhibit unique photoluminescence due to size-dependent quantum confinement, giving rise to fascinating full-color emission properties. Accurate emission calculations using time-dependent density functional theory are a time-costing and expensive process. Herein, we employed an artificial neural network (ANN) combined with statistical learning to establish the relationship between geometrical/electronic structures of ground states and emission wavelength for C3N QDs. The emission energy of these QDs can be doubly modulated by size and edge effects, which are governed by the number of C4N2 rings and the CH group, respectively. Moreover, these two structural characteristics also determine the phonon vibration mode of C3N QDs to harmonize the emission intensity and lifetime of hot electrons in the electron-hole recombination process, as indicated by nonadiabatic molecular dynamics simulation. These computational results provide a general approach to atomically precise design the full-color fluorescent carbon-based QDs with targeted functions and high performance.

Understanding the Magnetic Microstructure through Experiments and Machine Learning Algorithms

Advanced machine learning techniques have unfurled their applications in various interdisciplinary areas of research and development. This paper highlights the use of image regression algorithms based on advanced neural networks to understand the magnetic properties directly from the magnetic microstructure. In this study, Co/Pd multilayers have been chosen as a reference material system that displays maze-like magnetic domains in pristine conditions. Irradiation of Ar+ ions with two different energies (50 and 100 keV) at various fluences was used as an external perturbation to investigate the modification of magnetic and structural properties from a state of perpendicular magnetic anisotropy to the vicinity of the spin reorientation transition. Magnetic force microscopy revealed domain fragmentation with a smaller periodicity and weaker magnetic contrast up to the fluence of 1014 ions/cm2. Further increases in the ion fluence result in the formation of feather-like domains with a variation in local magnetization distribution. The experimental results were complemented with micromagnetic simulations, where the variations of effective magnetic anisotropy and exchange constant result in qualitatively similar changes in magnetic domains, as observed experimentally. Importantly, a set of 960 simulated domain images was generated to train, validate, and test the convolutional neural network (CNN) that predicts the magnetic properties directly from the domain images with a high level of accuracy (maximum 93.9%). Our work has immense importance in promoting the applications of image regression methods through the CNN in understanding integral magnetic properties obtained from the microscopic features subject to change under external perturbations.