A Hybrid Machine Learning Framework for Predicting Hydrogen Storage Capacities in Metal Hydrides: Unsupervised Feature Learning with Deep Neural Networks
Satadeep Bhattacharjee, Pritam Das, Swetarekha Ram, Seung-Cheol Lee
In this study, we present a sophisticated hybrid machine-learning framework that significantly improves the accuracy of predicting hydrogen storage capacities in metal hydrides. This is a critical challenge due to the scarcity of experimental data and the complexity of high-dimensional feature spaces. Our approach employs the power of unsupervised learning through the use of a state-of-the-art autoencoder. This autoencoder is trained on elemental descriptors obtained from Mendeleev software, enabling the extraction of a meaningful and lower dimensional latent space from the input data. This latent representation serves as the basis for our deep multi-layer perceptron (MLP) model, which consists of five layers and shows good precision in predicting hydrogen storage capacities. Furthermore, our results show very good agreement with the results of density functional theory (DFT). In addition to addressing the limitations caused by limited and unevenly distributed data in the field of hydrogen storage materials, we also focus on discovering new materials that show promising opportunities for hydrogen storage. These materials were identified using both feature-based approaches and predictions generated by a large language model. Finally, our investigation into the effectiveness of transferring weights from the autoencoder to the MLP, in addition to the latent features, suggests that while this strategy slightly improves model performance indicated by a slightly higher R$^2$ value and lower RMSE, it emphasizes the intricate challenge of adapting pre-trained weights for specific supervised tasks.
High-Entropy Hydrides for Fast and Reversible Hydrogen Storage at Room Temperature: Binding-Energy Engineering via First-Principles Calculations and Experiments
Core-Shell Confinement Blocks Hydride Formation: The Impact of Surface Oxides on Hydrogen Sorption in Nanoporous FeTi
Montanuniversität Leoben(莱奥本矿业大学);Erich Schmid Institute of Materials Science, Austrian Academy of Sciences(奥地利科学院埃里希·施密德材料科学研究所);Foundation for Research and Technology – Hellas(希腊研究与技术基金会);Deutsches Elektronen-Synchrotron DESY(德国电子同步辐射中心);Helmholtz-Zentrum Hereon(亥姆霍兹海雷恩中心)
Physically Interpretable Descriptors Drive the Materials Design of Metal Hydrides for Hydrogen Storage
Advanced Institute for Materials Research (WPI-AIMR), Tohoku University(东北大学先进物质研究院);Institute of Fluid Science, Tohoku University(东北大学流体科学研究所);Department of Materials Engineering, The University of Tokyo(东京大学材料工程系);Institute for Materials Research (IMR), Tohoku University(东北大学材料研究所)
A comprehensive first principles investigation of A$_2$BH$_6$ type (A= Li,Na, and K; B= Al, and Si) double perovskite hydrides for high capacity hydrogen storage
Physical properties of transition metal hydride superconductors Mg2TmH6 (Tm = Rh, Pd, Ir, Pt) by first-principles calculations
Mawlana Bhashani Science and Technology University(毛拉纳·巴沙尼科技大学);University of Rajshahi(拉杰沙希大学);Rajshahi University of Engineering and Technology(拉杰沙希工程与技术大学)
Machine learning to explore high-entropy alloys with desired enthalpy for room-temperature hydrogen storage: Prediction of density functional theory and experimental data
Chemical transformation of MgH2/V2O5 composite to Mg-V-O rock salt and its influence on the electrochemical Li conversion and hydrogen storage characteristics of MgH2
University of Aveiro(阿威罗大学);LASI - Intelligent Systems Associate Laboratory(智能系统联合实验室);International Iberian Nanotechnology Laboratory (INL)(伊比利亚纳米技术国际实验室)
A generative machine learning model for designing metal hydrides applied to hydrogen storage
Mathematics and Statistics, Louisiana Tech University(数学与统计学系,拉法叶技术大学);Louisiana Tech University(拉法叶技术大学);Institute for Micromanufacturing, Louisiana Tech University(微制造研究所,拉法叶技术大学);National Energy Technology Laboratory, United States Department of Energy, Pittsburgh, PA 15236(美国能源部国家能源技术实验室,匹兹堡,PA 15236)
Ambient and Pressure Dependent Superconductivity with Hydrogen Storage Potential in Quaternary Hydride LiMgZr2H12: A Comprehensive First-principles Insights
Department of Physics, University of Rajshahi(拉杰沙希大学物理系)
A unified descriptor framework for hydrogen storage capacity and equilibrium pressure in interstitial hydrides
Advanced Institute for Materials Research (WPI-AIMR), Tohoku University(东北大学先进材料研究所);Unprecedented-scale Data Analytics Center, Tohoku University(东北大学超大规模数据分析中心);Frontier Research Institute for Interdisciplinary Sciences (FRIS), Tohoku University(东北大学前沿交叉科学研究所);Department of Materials Engineering, The University of Tokyo(东京大学材料工程系);Institute for Materials Research, Tohoku University(东北大学物质材料研究机构);Institute of Fluid Science, Tohoku University(东北大学流体科学研究所)
Machine learning via artificial neural networks coupled with density functional theory and experiments for thermodynamic optimization of high-entropy alloys for hydrogen storage at room temperature
WPI, International Institute for Carbon-Neutral Energy Research (WPI-I2CNER), Kyushu University(九州大学);Department of Automotive Science, Graduate School of Integrated Frontier Sciences, Kyushu University(九州大学);Institute for Materials Science, University of Stuttgart(斯图加特大学)
Materials for hydrogen-based energy storage: Past, recent progress and future outlook
氢化镁在固体储氢材料方面占有最大的份额。为IEA Task 32 Hydrogen Based Energy Storage做出贡献的国际镁专家组最近发表了两篇综述论文,介绍了该小组的活动重点是氢化镁基材料和用于氢和能量存储的镁基化合物。这篇综述文章不仅概述了镁基复合材料的基本方面及其应用的最新活动,而且还介绍了该主题的历史概述,并概述了未来的发展。重点介绍了Mg-H体系在极压下的理论和实验研究、MgH 2体系的动力学和热力学研究、纳米结构研究、新型镁基化合物和新型复合材料研究以及镁基储氢体系中的催化作用研究。最后,热能存储和放大氢存储系统容纳MgH 2。
Hydrogen uptake and hydride formation in Al$_x$CoCrFeNi high-entropy alloys: First-principles, universal-potential, and experimental study
Ruhr-Universität Bochum(波鸿鲁尔大学);Max-Planck-Institut for Sustainable Materials(马克斯·普朗克可持续材料研究所);University of Stuttgart(斯图加特大学);Deutsches Elektronen-Synchrotron DESY(德国电子同步辐射中心);Goethe University Frankfurt(法兰克福歌德大学);Federal Institute for Materials Research and Testing (BAM)(联邦材料研究与测试研究所)
Synthesis and Characterization of TiFe0.7-xMn0.3Vx (x=0.05, and 0.1) and Ti1-yTayFe0.7Mn0.3 (y=0.2, and 0.4) Nanostructured Metal Hydrides for Low Temperature Applications
Modeling the thermodynamics of the FeTi hydrogenation under para-equilibrium: an ab-initio and experimental study
与替代的金属间化合物相比,基于FeTi的金属间化合物由于有利的可逆性、良好的吸附动力学和相对低的成本而作为固定的贮氢材料最近重新引起注意。采用OpenCalphad软件对(FeTi)$_{1-x}$ H $_{x}$(0 $\leq x \leq$ 1)体系的热力学进行了评估,作为模拟FeTi基合金氢化的关键基础。从我们的实验压力-组成-等温线(PCI)曲线,以及利用密度泛函理论(DFT)的第一性原理计算获得新的热力学数据。根据文献信息和\emph{ab-initio}调查的批判性分析,仔细选择了热力学相模型。关键的热力学性质,如离解压力,形成的热力学和相图计算与我们进行的实验和文献报道的数据吻合良好。这项工作提供了一个初步的观点,可以扩展到更高阶的热力学评估,并随后使设计新的FeTi为基础的双金属氧化物。此外,评估的热力学数据可以作为动力学模型和氢化物微观结构模拟的关键输入。
Structural phase stability, electronic structure, magnetic properties and chemical bonding analysis of transition metal ammine borohydrides with amphoteric hydrogen for hydrogen storage