Article
Chemistry, Multidisciplinary
Daniil Bash, Yongqiang Cai, Vijila Chellappan, Swee Liang Wong, Xu Yang, Pawan Kumar, Jin Da Tan, Anas Abutaha, Jayce J. W. Cheng, Yee-Fun Lim, Siyu Isaac Parker Tian, Zekun Ren, Flore Mekki-Berrada, Wai Kuan Wong, Jiaxun Xie, Jatin Kumar, Saif A. Khan, Qianxao Li, Tonio Buonassisi, Kedar Hippalgaonkar
Summary: The study introduces a rapid machine learning-driven automated flow mixing setup for thin film preparation, which, combined with high-throughput experiments, accelerates materials and process optimization. This approach presents a robust machine-learning driven high-throughput experimental scheme that can effectively understand, optimize, and design new materials and composites.
ADVANCED FUNCTIONAL MATERIALS
(2021)
Article
Engineering, Environmental
Lei Tao, Jinlong He, Nuwayo Eric Munyaneza, Vikas Varshney, Wei Chen, Guoliang Liu, Ying Li
Summary: This study explores the discovery of high-performance polyimides using machine learning and molecular dynamics simulations. A comprehensive library of over 8 million hypothetical polyimides is built, and multiple machine learning models are established to predict the thermal and mechanical properties of polyimides. Through the screening of machine learning models, three novel polyimides with superior properties are discovered and validated through molecular dynamics simulations and experiments. This study provides an efficient approach to expedite the discovery of novel polymers.
CHEMICAL ENGINEERING JOURNAL
(2023)
Article
Engineering, Multidisciplinary
Siwei Song, Yi Wang, Fang Chen, Mi Yan, Qinghua Zhang
Summary: This study presents a methodology that combines domain knowledge, machine learning algorithms, and experiments to accelerate the discovery of novel energetic materials. The established high-throughput virtual screening system allows for the rapid selection of candidate molecules with promising properties and desirable crystal packing modes from a large molecular space. Experimental results confirm the effectiveness of the proposed methodology.
Article
Chemistry, Multidisciplinary
Jiaxin Fan, Wenxian Li, Sean Li, Jack Yang
Summary: Ammonia has gained attention as a carrier for hydrogen usage in the hydrogen economy, but new synthesis techniques are needed to overcome high energy consumption. Chemical looping ammonia synthesis (CLAS) is a promising approach, but ideal redox materials are yet to be discovered. This study screens 1699 bicationic redox pairs using the MP database and employs machine learning to broaden the search for potential redox materials. Bicationic compounds containing alkali/alkaline-earth metals and transition metal/metalloid elements show promise in CLAS.
Article
Materials Science, Multidisciplinary
Z. Liu, M. Jiang, T. Luo
Summary: This study demonstrates the use of transfer learning to improve machine learning models for fast screening of semiconductor candidates with desirable thermal conductivity (TC), utilizing a large low-fidelity dataset as a proxy task to improve models trained on high-fidelity but small data. The transfer learning models show improved accuracy and have potential implications for materials informatics.
MATERIALS TODAY PHYSICS
(2022)
Article
Nanoscience & Nanotechnology
Sauradeep Majumdar, Seyed Mohamad Moosavi, Kevin Maik Jablonka, Daniele Ongari, Berend Smit
Summary: This study developed a database of around 20,000 hypothetical MOFs, visualizing and quantifying their diversity using machine learning techniques. The addition of these structures improved the overall diversity metrics of the databases, especially in terms of the chemistry of metal nodes. Evaluations using grand-canonical Monte Carlo simulations showed that many of these diverse structures outperformed benchmark materials in post-combustion carbon capture and hydrogen storage applications.
ACS APPLIED MATERIALS & INTERFACES
(2021)
Article
Nanoscience & Nanotechnology
Guokui Zheng, Yanle Li, Xu Qian, Ge Yao, Ziqi Tian, Xingwang Zhang, Liang Chen
Summary: Exploring high-activity, selective, and stable electrocatalysts is crucial for electrocatalytic ammonia synthesis. Transition-metal-doped Au-based single-atom alloys (SAAs) were identified as promising candidates for nitrogen reduction reaction (NRR) due to their ability to lower free energy barriers. Initial screening revealed Mo and W-doped systems as having the best activity in terms of limiting potential.
ACS APPLIED MATERIALS & INTERFACES
(2021)
Article
Nanoscience & Nanotechnology
Mao Wang, Qisong Xu, Hongjian Tang, Jianwen Jiang
Summary: This study applies machine learning models to the development of pervaporation (PV) membranes and successfully predicts the performance of PV membranes and screens out 20 polymers for PV separation through the collection of experimental data and the development of two types of models.
ACS APPLIED MATERIALS & INTERFACES
(2022)
Article
Chemistry, Physical
Jiewei Cheng, Tingwei Li, Yongyi Wang, Ahmed H. Ati, Qiang Sun
Summary: In this study, a multiscale workflow was proposed to computationally screen a large number of MXene compounds for hydrogen storage under near ambient conditions. By training neural networks and validating through simulations, it was found that ScYC exhibits a high hydrogen storage capacity.
APPLIED SURFACE SCIENCE
(2023)
Article
Agriculture, Multidisciplinary
Samiul Haque, Edgar Lobaton, Natalie Nelson, G. Craig Yencho, Kenneth Pecota, Russell Mierop, Michael W. Kudenov, Mike Boyette, Cranos M. Williams
Summary: The quality variation in horticultural crops significantly affects market value, but objectively characterizing subjective crop quality characteristics at production-scale remains challenging. Using sweetpotato as a case study, researchers introduced a high-throughput computer vision algorithm for quantifying shape and size characteristics, demonstrating the potential for industrial crop production big data analytics.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Article
Chemistry, Multidisciplinary
Renchao Che, Zhengchen Wu, Bin Quan, Ruixuan Zhang, Huiran Zhang, Jincang Zhang, Wencong Lu
Summary: An unprecedented machine learning-based forecasting system was constructed to directly predict the process conditions of carbonyl iron/ferrosoferric oxide hybrids with enhanced microwave absorption (MA) performance. High-throughput screening and inverse projection based on pattern recognition recommended a series of excellent MA materials with a high performance correlation coefficient of up to 0.9844. After manual selection, the optimal hybrid showed enhancements of maximum absorption efficiency and bandwidth by 207% and 360%, respectively, compared to the original database. The standardized machine learning forecasting system greatly shortened the research cycle to a few weeks compared to several months of manual orthogonal experiment.
ADVANCED FUNCTIONAL MATERIALS
(2023)
Article
Mathematics, Applied
Wensi Wu, Christophe Bonneville, Christopher Earls
Summary: This study presents a strategy based on (constrained) Bayesian optimization to alleviate the burden of high fidelity fluid-structure interaction simulations in terms of time and economic considerations. By gradually conducting Bridge Simulations, we are able to achieve fast convergence towards a working admissible design for an underwater, unmanned, autonomous vehicle (UUAV) sail plane.
FINITE ELEMENTS IN ANALYSIS AND DESIGN
(2021)
Article
Materials Science, Multidisciplinary
Zhihao Feng, Yifan Cheng, Alexandra Khlyustova, Aasim Wani, Trevor Franklin, Jeffrey D. Varner, Andrew L. Hook, Rong Yang
Summary: Amphiphilic copolymers (AP) are a class of antibiofouling materials that can be optimized by tuning their chemistry and composition. However, the large design space associated with AP makes optimization laborious. To address this, a machine learning approach is used to accurately predict biofilm formation on a library of AP, expanding the accessible design space and identifying best-performing candidates for experimental validation.
ADVANCED MATERIALS TECHNOLOGIES
(2023)
Article
Nanoscience & Nanotechnology
Diptendu Roy, Shyama Charan Mandal, Biswarup Pathak
Summary: This study demonstrated an efficient approach for screening new product-selective catalysts using machine learning algorithm and microstructure model, leading to the discovery of seven active catalysts for CO2 to methanol formation.
ACS APPLIED MATERIALS & INTERFACES
(2021)
Article
Chemistry, Physical
Markus Hussner, Richard Adam Pacalaj, Gerhard Olaf Mueller-Dieckert, Chao Liu, Zhisheng Zhou, Nahdia Majeed, Steve Greedy, Ivan Ramirez, Ning Li, Seyed Mehrdad Hosseini, Christian Uhrich, Christoph Josef Brabec, James Robert Durrant, Carsten Deibel, Roderick Charles Ian MacKenzie
Summary: The evaluation of organic solar cell materials based solely on power conversion efficiency has hindered progress. This study introduces a machine learning technique that can extract additional performance parameters from light-dark JV curves, enabling faster material analysis.
ADVANCED ENERGY MATERIALS
(2023)
Article
Computer Science, Software Engineering
Matt Benatan, Kia Ng
JOURNAL OF VISUAL LANGUAGES AND COMPUTING
(2015)
Article
Chemistry, Medicinal
James L. McDonagh, Ardita Shkurti, David J. Bray, Richard L. Anderson, Edward O. Pyzer-Knapp
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2019)
Article
Multidisciplinary Sciences
Laura-Jayne Gardiner, Rachel Rusholme-Pilcher, Josh Colmer, Hannah Rees, Juan Manuel Crescente, Anna Paola Carrieri, Susan Duncan, Edward O. Pyzer-Knapp, Ritesh Krishna, Anthony Hall
Summary: Machine learning is used to predict complex circadian gene expression patterns in Arabidopsis, classifying genes and revealing potential regulatory mechanisms without experimental work or prior knowledge. Model interpretation helps optimize sampling strategies and accurately predict circadian time, providing insight into biological processes and experimental design.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)
Article
Multidisciplinary Sciences
Edward O. Pyzer-Knapp, Linjiang Chen, Graeme M. Day, Andrew I. Cooper
Summary: The study introduces a new method using parallel Bayesian optimization to acquire ESF data, providing the same level of insight at a lower computational cost and paving the way for high-throughput virtual screening. By greatly reducing the opportunity risk associated with the choice of system, this approach achieved a significant computational speedup in a specific research scenario.
Article
Computer Science, Artificial Intelligence
Alessandro Varsi, Simon Maskell, Paul G. Spirakis
Summary: In this paper, a novel parallel redistribution algorithm for Distributed Memory with a time complexity of O(log2N) is proposed. Empirical results indicate that this new approach outperforms the traditional method.
Article
Chemistry, Physical
Edward O. Pyzer-Knapp, Jed W. Pitera, Peter W. J. Staar, Seiji Takeda, Teodoro Laino, Daniel P. Sanders, James Sexton, John R. Smith, Alessandro Curioni
Summary: New tools and technologies enable new ways of working in materials science, enhancing each stage of the discovery cycle through automation and simulation. The impact of these technologies is amplified when used in conjunction with each other as powerful, heterogeneous workflows.
NPJ COMPUTATIONAL MATERIALS
(2022)
Article
Chemistry, Medicinal
David E. Graff, Matteo Aldeghi, Joseph A. Morrone, Kirk E. Jordan, Edward O. Pyzer-Knapp, Connor W. Coley
Summary: In this study, a technique called design space pruning (DSP) is proposed to mitigate inference costs in model-guided optimization by removing poor-performing candidates. The experimental results show that this technique can effectively reduce overhead costs without sacrificing performance.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Chemistry, Medicinal
Matteo Aldeghi, David E. Graff, Nathan Frey, Joseph A. Morrone, Edward O. Pyzer-Knapp, Kirk E. Jordan, Connor W. Coley
Summary: In molecular discovery and drug design, analyzing the roughness of molecular property landscapes can guide chemical space navigation.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Chemistry, Multidisciplinary
Adam D. Clayton, Edward O. Pyzer-Knapp, Mark Purdie, Martin F. Jones, Alexandre Barthelme, John Pavey, Nikil Kapur, Thomas W. Chamberlain, A. John Blacker, Richard A. Bourne
Summary: The optimization of multistep chemical syntheses is crucial for rapid development of new pharmaceuticals. A continuous flow platform was developed to automate the optimization of telescoped reactions. By integrating Bayesian optimization techniques, an 81% overall yield was achieved in just 14 hours, and a favorable competing pathway for the desired product was identified.
ANGEWANDTE CHEMIE-INTERNATIONAL EDITION
(2023)
Article
Chemistry, Medicinal
David E. Graff, Matteo Aldeghi, Joseph A. Morrone, Kirk E. Jordan, Edward O. Pyzer-Knapp, Connor W. Coley
Summary: In this study, a technique called design space pruning (DSP) is proposed to reduce inference costs in model-guided optimization. By permanently removing poor-performing candidates from consideration, DSP achieves significant reductions in overhead costs while maintaining similar performance to baseline optimization.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Engineering, Electrical & Electronic
Alessandro Varsi, Jack Taylor, Lykourgos Kekempanos, Edward Pyzer Knapp, Simon Maskell
IEEE SIGNAL PROCESSING LETTERS
(2020)
Article
Computer Science, Hardware & Architecture
E. O. Pyzer-Knapp
IBM JOURNAL OF RESEARCH AND DEVELOPMENT
(2018)