SEARCH · Engineering Papers
Results for “Materials Science”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Assessment of fine-tuned large language models for real-world chemistry and material science applications
The current generation of large language models (LLMs) has limited chemical knowledge. Recently, it has been shown that these LLMs can learn and predict chemical properties through fine-tuning. Using natural language to train machine learning models opens doors to a wider chemical audience, as field-specific featurization techniques can be omitted. In this work, we explore the potential and limitations of this approach. We studied the performance of fine-tuning three open-source LLMs (GPT-J-6B, Llama-3.1-8B, and Mistral-7B) for a range of different chemical questions. We benchmark their performances against “traditional” machine learning models and find that, in most cases, the fine-tuning approach is superior for a simple classification problem. Depending on the size of the dataset and the type of questions, we also successfully address more sophisticated problems. The most important conclusions of this work are that, for all datasets considered, their conversion into an LLM fine-tuning training set is straightforward and that fine-tuning with even relatively small datasets leads to predictive models. These results suggest that the systematic use of LLMs to guide experiments and simulations will be a powerful technique in any research study, significantly reducing unnecessary experiments or computations.
Correction: Atomate2: modular workflows for materials science
[This corrects the article DOI: 10.1039/D5DD00019J.].
Mining experimental data from materials science literature with large language models: an evaluation study
Not Available
Erratum: Review—Materials Science Predictions of Thermal Runaway in Layered Metal-Oxide Cathodes: A Review of Thermodynamics [ J. Electrochem. Soc. , 167, 090543 (2020)]
It was found that corrections are required for the published article cited in the title of this document. A revised Table VII is shown below, in which the values for S (entropy of the liquid solvents) have been corrected for consistency with the original sources. No corrections were required for the enthalpies, but they are included for completeness. The cited references and table footnotes found in the originally published version of Table VII remain applicable.
Nucleation, Growth, and Aggregation of Todorokite Nanoparticles from Both Geochemical and Materials Science Perspectives (Final Technical Report)
A vast amount of manganese (Mn) nodules exists in ocean sediments, covering about 30% of the ocean floor. Tunneled and layered Mn(IV) oxides, hereafter called TMOs and LMOs, are the primary mineral phases hosting Mn. Owing to their unique structure and properties, these Mn oxides are highly enriched in transition metals, platinum group metals, and rare earth elements, up to several percentages in weight. Thus, many metals are stored in these nodules and the nodules have thus been increasingly considered for their potential as a source of strategic metals for national security applications and the energy industry. The nodules also play an important role in trace metal biogeochemical cycling in the ocean because they control the partition and distribution of trace metals in oceanic sediments. Layered and tunneled structures differ in their mineralogical and chemical properties. For example, TMOs contain fewer metals compared to LMOs in ocean nodules. Therefore, the relative proportion of the two types of minerals in ocean nodules determines the metal sequestration capacities of the nodules, hence the amount of metals stored in the nodules and metal cycling in oceans. The overall objective of this study is to investigate the mechanisms of layered to tunneled structural transform driven by redox reactions. Specifically, we focus on the redox reactions between Mn(IV) and Mn(II) because they often coexist during Mn redox cycling.
Interatomic Potentials for Materials Science and Beyond; Advances in Machine Learned Spectral Neighborhood Analysis Potentials.
Abstract not provided.
Gaussian process and Bayesian optimization - Bridging the gap between theory and practice in materials science.
Abstract not provided.
Solving Stochastic Inverse Problems for Property-Structure Relationships in Computational Materials Science.
Abstract not provided.
Interatomic Potentials for Materials Science and Beyond; Advances in Machine Learned Spectral Neighborhood Analysis Potentials.
Abstract not provided.
Versatile Gaussian process and Bayesian optimization for multiscale computational materials science applications.
Abstract not provided.
Advanced Gaussian process regression and optimization for computational materials science applications.
Abstract not provided.
Enabling the fourth-paradigm of materials science through versatile Gaussian process and Bayesian optimization.
Abstract not provided.
Versatile Gaussian Process and Bayesian Optimization for Multiscale Computational Materials Science Applications.
Abstract not provided.
Quantifying Uncertainties in Limited-Angle Tomography - Applications to Dynamic Materials Science Experiments
Explore the source record for details and available documents.
Modular and Scalable Solutions of Machine Learned Models (for Material Science and Beyond)
Explore the source record for details and available documents.
Toward automated design of optimized high energy density material science experiments on the Z Machine
Explore the source record for details and available documents.