Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userMachine Learning Enabled Prediction of Cathode Materials for Zn ion Batteries
131
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Rechargeable Zn batteries with aqueous electrolytes have been considered as promising alternative energy storage technology, with various advantages such as low cost, high volumetric capacity, environmentally friendly, and high safety. However, a lack of reliable cathode materials has largely pledged their applications. Herein, we developed a machine learning (ML) based approach to predict cathodes with high capacity (>150 mAh/g) and high voltage (>0.5V). We screened over ~130,000 inorganic materials from the Materials Project database and applied the crystal graph convolutional neural network (CGCNN) based ML approach with data from the AFLOW database. The combination of these two could not only screen cathode materials that match well with the experimental data but also predict new promising candidates for further experimental validations. We hope this study could spur further interests in ML-based advanced theoretical tools for battery materials discovery.
rate research
Read More
The large-scale search for high-performing candidate 2D materials is limited to calculating a few simple descriptors, usually with first-principles density functional theory calculations. In this work, we alleviate this issue by extending and generalizing crystal graph convolutional neural networks to systems with planar periodicity, and train an ensemble of models to predict thermodynamic, mechanical, and electronic properties. To demonstrate the utility of this approach, we carry out a screening of nearly 45,000 structures for two largely disjoint applications: namely, mechanically robust composites and photovoltaics. An analysis of the uncertainty associated with our methods indicates the ensemble of neural networks is well-calibrated and has errors comparable with those from accurate first-principles density functional theory calculations. The ensemble of models allows us to gauge the confidence of our predictions, and to find the candidates most likely to exhibit effective performance in their applications. Since the datasets used in our screening were combinatorically generated, we are also able to investigate, using an innovative method, structural and compositional design principles that impact the properties of the structures surveyed and which can act as a generative model basis for future material discovery through reverse engineering. Our approach allowed us to recover some well-accepted design principles: for instance, we find that hybrid organic-inorganic perovskites with lead and tin tend to be good candidates for solar cell applications.
Lattice constants such as unit cell edge lengths and plane angles are important parameters of the periodic structures of crystal materials. Predicting crystal lattice constants has wide applications in crystal structure prediction and materials property prediction. Previous work has used machine learning models such as neural networks and support vector machines combined with composition features for lattice constant prediction and has achieved a maximum performance for cubic structures with an average $R^2$ of 0.82. Other models tailored for special materials family of a fixed form such as ABX3 perovskites can achieve much higher performance due to the homogeneity of the structures. However, these models trained with small datasets are usually not applicable to generic lattice parameter prediction of materials with diverse compositions. Herein, we report MLatticeABC, a random forest machine learning model with a new descriptor set for lattice unit cell edge length ($a,b,c$) prediction which achieves an R2 score of 0.979 for lattice parameter $a$ of cubic crystals and significant performance improvement for other crystal systems as well. Source code and trained models can be freely accessed at https://github.com/usccolumbia/MLatticeABC
Bismuth has recently attracted interest in connection with Na-ion battery anodes due to its high volumetric capacity. It reacts with Na to form Na$_3$Bi which is a prototypical Dirac semimetal with a nontrivial electronic structure. Density-functional-theory based first-principles calculations are playing a key role in understanding the fascinating electronic structure of Na$_3$Bi and other topological materials. In particular, the strongly-constrained-and-appropriately-normed (SCAN) meta-generalized-gradient-approximation (meta-GGA) has shown significant improvement over the widely used generalized-gradient-approximation (GGA) scheme in capturing energetic, structural, and electronic properties of many classes of materials. Here, we discuss the electronic structure of Na$_3$Bi within the SCAN framework and show that the resulting Fermi velocities and {it s}-band shift around the $Gamma$ point are in better agreement with experiments than the corresponding GGA predictions. SCAN yields a purely spin-orbit-coupling (SOC) driven Dirac semimetal state in Na$_3$Bi in contrast with the earlier GGA results. Our analysis reveals the presence of a topological phase transition from the Dirac semimetal to a trivial band insulator phase in Na$_{3}$Bi$_{x}$Sb$_{1-x}$ alloys as the strength of the SOC varies with Sb content, and gives insight into the role of the SOC in modulating conduction properties of Na$_3$Bi.
In this work, we develop a combined convolutional neural networks (CNNs) and finite element method (FEM) to examine the effective thermal properties of composite phase change materials (CPCMs) consisting of paraffin and copper foam. In this approach, first the CPCM microstructures are modeled using FEM and next the image dataset with corresponding thermal properties is created. The image dataset is subsequently used to train and test the CNN performance, which is then compared with the performance of a popular network architecture for image classification tasks. The predicted thermal properties are employed to define the properties of the CPCM material of a battery pack. The heat generation and electrochemical response of a Li-ion cell during the charging/discharging is simulated by applying Newman battery model. Thermal management is achieved by the latent heat of paraffin, with copper foam for enhancing the thermal conductivity. The multiscale model is finally developed using FEM to investigate the effectiveness of the thermal management of the battery pack. In these models the thermal properties estimated by the FEM and the CNN are employed to define the CPCM materials properties of a battery pack. Our results confirm that the model developed on the basis of a CNN can evaluate the effectiveness of the battery packs thermal management system with an excellent accuracy in comparison with the original FEM models.
Geometric information such as the space groups and crystal systems plays an important role in the properties of crystal materials. Prediction of crystal system and space group thus has wide applications in crystal material property estimation and structure prediction. Previous works on experimental X-ray diffraction (XRD) and density functional theory (DFT) based structure determination methods achieved outstanding performance, but they are not applicable for large-scale screening of materials compositions. There are also machine learning models using Magpie descriptors for composition based material space group determination, but their prediction accuracy only ranges between 0.638 and 0.907 in different kinds of crystals. Herein, we report an improved machine learning model for predicting the crystal system and space group of materials using only the formula information. Benchmark study on a dataset downloaded from Materials Project Database shows that our random forest models based on our new descriptor set, achieve significant performance improvements compared with previous work with accuracy scores ranging between 0.712 and 0.961 in terms of space group classification. Our model also shows large performance improvement for crystal system prediction. Trained models and source code are freely available at url{https://github.com/Yuxinya/SG_predict}
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
