Search for: All records

There has been a growing effort to replace manual extraction of data from research papers with automated data extraction based on natural language processing, language models, and recently, large language models (LLMs). Although these methods enable efficient extraction of data from large sets of research papers, they require a significant amount of up-front effort, expertise, and coding. In this work, we propose the method that can fully automate very accurate data extraction with minimal initial effort and background, using an advanced conversational LLM. consists of a set of engineered prompts applied to a conversational LLM that both identify sentences with data, extract that data, and assure the data’s correctness through a series of follow-up questions. These follow-up questions largely overcome known issues with LLMs providing factually inaccurate responses. can be applied with any conversational LLMs and yields very high quality data extraction. In tests on materials data, we find precision and recall both close to 90% from the best conversational LLMs, like GPT-4. We demonstrate that the exceptional performance is enabled by the information retention in a conversational model combined with purposeful redundancy and introducing uncertainty through follow-up prompts. These results suggest that approaches similar to , due to their simplicity, transferability, and accuracy are likely to become powerful tools for data extraction in the near future. Finally, databases for critical cooling rates of metallic glasses and yield strengths of high entropy alloys are developed using .

 

In laser powder bed fusion processes, keyholes are the gaseous cavities formed where laser interacts with metal, and their morphologies play an important role in defect formation and the final product quality. The in-situ X-ray imaging technique can monitor the keyhole dynamics from the side and capture keyhole shapes in the X-ray image stream. Keyhole shapes in X-ray images are then often labeled by humans for analysis, which increasingly involves attempting to correlate keyhole shapes with defects using machine learning. However, such labeling is tedious, time-consuming, error-prone, and cannot be scaled to large data sets. To use keyhole shapes more readily as the input to machine learning methods, an automatic tool to identify keyhole regions is desirable. In this paper, a deep-learning-based computer vision tool that can automatically segment keyhole shapes out of X-ray images is presented. The pipeline contains a filtering method and an implementation of the BASNet deep learning model to semantically segment the keyhole morphologies out of X-ray images. The presented tool shows promising average accuracy of 91.24% for keyhole area, and 92.81% for boundary shape, for a range of test dataset conditions in Al6061 (and one AliSi10Mg) alloys, with 300 training images/labels and 100 testing images for each trial. Prospective users may apply the presently trained tool or a retrained version following the approach used here to automatically label keyhole shapes in large image sets.

 

Electron counting can be performed algorithmically for monolithic active pixel sensor direct electron detectors to eliminate readout noise and Landau noise arising from the variability in the amount of deposited energy for each electron. Errors in existing counting algorithms include mistakenly counting a multielectron strike as a single electron event, and inaccurately locating the incident position of the electron due to lateral spread of deposited energy and dark noise. Here, we report a supervised deep learning (DL) approach based on Faster region-based convolutional neural network (R-CNN) to recognize single electron events at varying electron doses and voltages. The DL approach shows high accuracy according to the near-ideal modulation transfer function (MTF) and detector quantum efficiency for sparse images. It predicts, on average, 0.47 pixel deviation from the incident positions for 200 kV electrons versus 0.59 pixel using the conventional counting method. The DL approach also shows better robustness against coincidence loss as the electron dose increases, maintaining the MTF at half Nyquist frequency above 0.83 as the electron density increases to 0.06 e−/pixel. Thus, the DL model extends the advantages of counting analysis to higher dose rates than conventional methods.

 

Abstract The information content of atomic-resolution scanning transmission electron microscopy (STEM) images can often be reduced to a handful of parameters describing each atomic column, chief among which is the column position. Neural networks (NNs) are high performance, computationally efficient methods to automatically locate atomic columns in images, which has led to a profusion of NN models and associated training datasets. We have developed a benchmark dataset of simulated and experimental STEM images and used it to evaluate the performance of two sets of recent NN models for atom location in STEM images. Both models exhibit high performance for images of varying quality from several different crystal lattices. However, there are important differences in performance as a function of image quality, and both models perform poorly for images outside the training data, such as interfaces with large difference in background intensity. Both the benchmark dataset and the models are available using the Foundry service for dissemination, discovery, and reuse of machine learning models. 

Short-timescale atomic rearrangements are fundamental to the kinetics of glasses and frequently dominated by one atom moving significantly (a rearrangement), while others relax only modestly. The rates and directions of such rearrangements (or hops) are dominated by the distributions of activation barriers ( E act ) for rearrangement for a single atom and how those distributions vary across the atoms in the system. We have used molecular dynamics simulations of Cu 50 Zr 50 metallic glass below T g in an isoconfigurational ensemble to catalog the ensemble of rearrangements from thousands of sites. The majority of atoms are strongly caged by their neighbors, but a tiny fraction has a very high propensity for rearrangement, which leads to a power-law variation in the cage-breaking probability for the atoms in the model. In addition, atoms generally have multiple accessible rearrangement vectors, each with its own E act . However, atoms with lower E act (or higher rearrangement rates) generally explored fewer possible rearrangement vectors, as the low E act path is explored far more than others. We discuss how our results influence future modeling efforts to predict the rearrangement vector of a hopping atom. 

Transition metal dichalcogenides (TMDs), especially in two-dimensional (2D) form, exhibit many properties desirable for device applications. However, device performance can be hindered by the presence of defects. Here, we combine state of the art experimental and computational approaches to determine formation energies and charge transition levels of defects in bulk and 2D MX₂(M = Mo or W; X = S, Se, or Te). We perform deep level transient spectroscopy (DLTS) measurements of bulk TMDs. Simultaneously, we calculate formation energies and defect levels of all native point defects, which enable identification of levels observed in DLTS and extend our calculations to vacancies in 2D TMDs, for which DLTS is challenging. We find that reduction of dimensionality of TMDs to 2D has a significant impact on defect properties. This finding may explain differences in optical properties of 2D TMDs synthesized with different methods and lays foundation for future developments of more efficient TMD-based devices.