Review
Chemistry, Physical
Austin D. Sendek, Brandi Ransom, Ekin D. Cubuk, Lenson A. Pellouchoud, Jagjit Nanda, Evan J. Reed
Summary: This article discusses the application of machine learning-based battery modeling in the small data domain. Through the case study of ionic conductivity modeling, the working principle of ML battery modeling and methods for judging model performance are analyzed in depth. Recommendations for building models in the small data regime are provided, and promising future directions in battery design modeling are identified.
ADVANCED ENERGY MATERIALS
(2022)
Article
Chemistry, Multidisciplinary
Flavio Carsughi
Summary: In this research, a simplified polydispersion analysis (SPA) method is proposed for analyzing small-angle scattering (SAS) data. By straightforward interpolation of SAS data, the size distribution function (SDF) of polydisperse inhomogeneities can be determined without advanced computational skills. The method is tested against simulated and experimental data with excellent results, offering new opportunities for scientists to investigate polydisperse systems using SAS technique.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Physical
Abigail Barclay, Birthe B. Kragelund, Lise Arleth, Martin Cramer Pedersen
Summary: Recent advances in protein expression protocols have allowed for the investigation of structurally complex and disordered biomolecules using small-angle scattering experiments. A modeling scheme has been developed that combines classical form factor based modeling with spherical harmonics-based formulation to accurately calculate scattering profiles from these complex samples. The scheme can account for flexible domains and other structurally elaborate components, and we demonstrate its utility through a case study on a growth hormone receptor membrane protein. We also explore how the scattering profiles vary under different contrasts and discuss the implications for data modeling.
JOURNAL OF COLLOID AND INTERFACE SCIENCE
(2023)
Article
Biochemical Research Methods
Ye Wang, Tathagata Bhattacharya, Yuchao Jiang, Xiao Qin, Yue Wang, Yunlong Liu, Andrew J. Saykin, Li Chen
Summary: With the development and decreasing cost of next-generation sequencing technologies, the study of the human microbiome has become a rapidly expanding research field with various clinical applications. Building a prediction model for clinical outcomes based on microbiome data is essential for improving prediction performance. The phylogenetic tree represents a unique correlation structure of microbiome and can be important for this purpose.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Automation & Control Systems
Nishtha Hooda, Jasgurpreet Singh Chohan, Ruchika Gupta, Raman Kumar
Summary: In this study, artificial intelligence systems are integrated with mechanical systems to reduce manufacturing time and cost of products. The random forest machine learning model is used to predict and validate the optimum deposition angle in Fused Deposition Modeling, achieving a prediction accuracy of 94.57%, significantly better than other methods. This robust model efficiently predicts the optimum deposition angle for any geometry, enhancing the applicability of digitally manufactured products.
Article
Engineering, Civil
Kexin Li, Mandar Chitre
Summary: Acoustic propagation models are widely used in various oceanic and underwater applications. This paper proposes a data-aided physics-based approach for high-frequency acoustic propagation modeling, which is not only data-efficient but also capable of incorporating varying degrees of environmental knowledge and generalizing well to extrapolation beyond the data collection area.
IEEE JOURNAL OF OCEANIC ENGINEERING
(2023)
Article
Physics, Multidisciplinary
Mehmet Salti, Emel Ciger, Evrim Ersin Kangal, Bilgin Zengin
Summary: In this study, we redesigned the generalized pressure dark energy model using a caloric framework and employed machine learning techniques to analyze the cosmic Hubble parameter. The optimized model parameters were obtained using a genetic neural network algorithm and the most recent observational measurements. Additionally, we addressed the issue of calculating errors on the optimized parameter values using the Fisher Information Matrix algorithm. The results showed good agreement with observational data and provided additional cosmological insights.
Article
Multidisciplinary Sciences
Jordan Venderley, Krishnanand Mallayya, Michael Matty, Matthew Krogstad, Jacob Ruff, Geoff Pleiss, Varsha Kishore, David Mandrus, Daniel Phelan, Lekhanath Poudel, Andrew Gordon Wilson, Kilian Weinberger, Puspa Upreti, Michael Norman, Stephan Rosenkranz, Raymond Osborn, Eun-Ah Kim
Summary: Researchers have developed an unsupervised machine learning method that can automatically extract charge density wave order parameters and detect intraunit cell ordering and its fluctuations from X-ray diffraction measurements. The method has been successfully applied to different materials and has provided valuable insights when connected to physical principles.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2022)
Review
Chemistry, Multidisciplinary
Hadas Shalit Peleg, Anat Milo
Summary: The chemistry community is experiencing a surge in scientific discoveries in organic chemistry with the support of machine learning. However, the use of machine learning techniques is often limited by small datasets in experimental organic chemistry. This article highlights the limitations of small data in machine learning and emphasizes the impact of bias and variance on constructing reliable predictive models. It aims to raise awareness of these potential pitfalls and provides an introductory guideline for good practice. Ultimately, the article stresses the great value of statistical analysis of small data and advocates for a holistic data-centric approach in chemistry.
ANGEWANDTE CHEMIE-INTERNATIONAL EDITION
(2023)
Article
Agronomy
Joao Vasco Silva, Joost van Heerwaarden, Pytrik Reidsma, Alice G. Laborte, Kindie Tesfaye, Martin K. van Ittersum
Summary: The performance of statistical and machine learning methods in explaining and predicting crop yield variability was assessed in this study. The results showed that big data from farmers' fields can to some extent explain on-farm yield variability, but not predict it across time and space.
FIELD CROPS RESEARCH
(2023)
Article
Chemistry, Analytical
Yabin Yu, Ying Liu, Jiawei Chen, Dong Jiang, Zilong Zhuang, Xiaoli Wu
Summary: This study investigates the method of detecting bolt looseness in timber structures using deep learning and machine vision technology. Through training models and testing, it achieves the goal of low cost and high accuracy in multi-bolted connections.
Article
Computer Science, Artificial Intelligence
Mathieu Doucet, Richard K. Archibald, William T. Heller
Summary: Neutron reflectometry is a powerful tool for studying thin films at nanoscale, and using a neural network to predict thin film structures can provide accurate results and improve efficiency in traditional fitting methods. The stability of neural network predictions against statistical fluctuations of measured reflectivity profiles shows promise for further development in more complex thin film systems.
MACHINE LEARNING-SCIENCE AND TECHNOLOGY
(2021)
Article
Chemistry, Multidisciplinary
Herman Rull, Markus Fischer, Stefan Kuhn
Summary: This study demonstrates a novel machine learning model that can achieve accurate prediction of F-19 and C-13 NMR chemical shifts of small molecules in specific solvents, even with relatively limited data.
JOURNAL OF CHEMINFORMATICS
(2023)
Article
Biochemical Research Methods
Zhenxing Wu, Minfeng Zhu, Yu Kang, Elaine Lai-Han Leung, Tailong Lei, Chao Shen, Dejun Jiang, Zhe Wang, Dongsheng Cao, Tingjun Hou
Summary: A study on learning QSAR models using various ML algorithms for 14 public datasets showed that rbf-SVM, rbf-GPR, XGBoost, and DNN generally perform better than other algorithms. SVM and XGBoost are recommended for regression learning on small datasets, while XGBoost is an excellent choice for large datasets. Ensemble models integrating multiple algorithms can improve prediction accuracy.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Chemistry, Multidisciplinary
Jules Schleinitz, Maxime Langevin, Yanis Smail, Benjamin Wehnert, Laurence Grimaud, Rodolphe Vuilleumier
Summary: This study focuses on the application of machine learning in predicting synthetic yields and builds a dataset based on organic reaction publications. The study finds that including optimization data improves prediction accuracy and emphasizes the impact of publication constraints on the exploration of chemical space by the synthetic community.
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
(2022)
Article
Materials Science, Ceramics
Ryuhei Motokawa, Koji Kaneko, Yojiro Oba, Takayuki Nagai, Yoshihiro Okamoto, Taishi Kobayashi, Takayuki Kumada, William T. Heller
Summary: This study investigates the impact of additives on the nanoscopic structure of borosilicate glasses. The results show that Na2O decreases the glass melting point and increases the spacing of SiO2 and B2O3-rich domains, while CaO/ZnO additives induce void structures in the glass, which can be suppressed by Li2O.
JOURNAL OF NON-CRYSTALLINE SOLIDS
(2022)
Article
Chemistry, Multidisciplinary
William T. Heller, Changwoo Do
Summary: This study investigates the impact of two water-miscible ionic liquids on the temperature-dependent self-assembly of triblock copolymers in aqueous solution. The results show that both ionic liquids lower the temperature of structural transitions, with one of the liquids having a stronger effect. Additionally, the self-assembled structures of the copolymer do not change significantly.
Article
Polymer Science
Monojoy Goswami, Oluwagbenga Oare Iyiola, Wei Lu, Kunlun Hong, Piotr Zolnierczuk, Laura-Roxana Stingaciu, William T. Heller, Omar Taleb, Bobby G. Sumpter, Daniel T. Hallinan Jr
Summary: The structure and dynamics of a block copolymer were studied to understand the effect of an interfacial block on chain dynamics. It was found that the interfacial rubbery block exhibited slower dynamics and confined layered morphologies, while the chain-end rubbery block dispersed in the rubbery matrix showed faster dynamics. The dynamical slowing at the interface was observed only at length scales larger than the characteristic segmental length, and the disparity between interfacial and chain-end dynamics increased with increasing length.
Article
Instruments & Instrumentation
J. F. Ankner, R. Ashkar, J. F. Browning, T. R. Charlton, M. Doucet, C. E. Halbert, F. Islam, A. Karim, E. Kharlampieva, S. M. Kilbey II, J. Y. Y. Lin, M. D. Phan, G. S. Smith, S. A. Sukhishvili, R. Thermer, G. M. Veith, E. B. Watkins, D. Wilson
Summary: The Quite Intense Kinetics Reflectometer (QIKR) is a versatile neutron reflectometer that can be used to analyze the composition and distribution of matter near interfaces. By harnessing the increased brilliance of the Spallation Neutron Source Second Target Station, QIKR is able to collect specular and off-specular reflectivity data faster than existing machines, allowing for real-time observations of time-dependent processes.
REVIEW OF SCIENTIFIC INSTRUMENTS
(2023)
Article
Chemistry, Multidisciplinary
Minkyu Kim, Moon Jong Han, Hansol Lee, Paraskevi Flouda, Daria Bukharina, Kellina J. Pierce, Katarina M. Adstedt, Madeline L. Buxton, Young Hee Yoon, William T. Heller, Srikanth Singamaneni, Vladimir V. Tsukruk
Summary: In this study, a method for the synthesis of chiral metal-organic frameworks (MOFs) from achiral precursors was reported by utilizing chiral nematic cellulose-derived bio-templates. It was demonstrated that chiral MOFs, specifically zeolitic imidazolate frameworks (ZIF), can be grown from regular precursors on twisted bundles of cellulose nanocrystals, resulting in a tetragonal crystal structure with chiral space group of P4(1). The templated chiral ZIF also exhibited enantioselective recognition and chiral sensing abilities.
ANGEWANDTE CHEMIE-INTERNATIONAL EDITION
(2023)
Article
Biochemistry & Molecular Biology
Frank Heinrich, Catherine E. Thomas, John J. Alvarado, Rebecca Eells, Alyssa Thomas, Mathieu Doucet, Kindra N. Whitlatch, Manish Aryal, Mathias Losche, Thomas E. Smithgall
Summary: The HIV-1 Nef protein is essential for viral infectivity, replication, and immune escape. It interacts with host cell signaling proteins and intracellular trafficking pathways to carry out its functions. Previous studies have shown that Nef forms homodimers at the plasma membrane, which are crucial for its activities.
JOURNAL OF MOLECULAR BIOLOGY
(2023)
Article
Chemistry, Physical
Uvinduni I. Premadasa, Vera Bocharova, Lu Lin, Anne-Caroline Genix, William T. Heller, Robert L. Sacci, Ying-Zhong Ma, Nikki A. Thiele, Benjamin Doughty
Summary: Liquid/liquid (L/L) interfaces play a crucial yet poorly understood role in various complex chemical phenomena, acting as gatekeepers to function. By using surface-specific vibrational sum frequency generation combined with neutron and X-ray scattering methods, we track the transport of DOP and DEHPA ligands in solvent extraction at buried oil/aqueous interfaces away from equilibrium. Our results show evidence of dynamic interfacial restructuring at low ligand concentrations, contrary to expectations. These findings provide new insights into interfacially controlled chemical transport at L/L interfaces, presenting potential avenues to design selective kinetic separations.
JOURNAL OF PHYSICAL CHEMISTRY B
(2023)
Article
Chemistry, Multidisciplinary
Yingzhen Ma, Christian Heil, Gergely Nagy, William T. Heller, Yaxin An, Arthi Jayaraman, Bhuvnesh Bharti
Summary: This study investigates the effects of salinity and temperature on the adsorption of a nonionic surfactant onto hydrophilic nanoparticles. The results show that both salinity and temperature increase the amount of surfactant adsorbed onto the nanoparticles and lead to the aggregation of the nanoparticles. Furthermore, the study demonstrates the non-monotonic changes in viscosity for the surfactant-nanoparticle mixture with increasing temperature and salinity and relates these observations to the aggregated state of the nanoparticles.
Article
Polymer Science
Mingqiu Hu, Xindi Li, William T. Heller, Wim Bras, Javid Rzayev, Thomas P. Russell
Summary: In this study, grazing-incidence small-angle neutron scattering (GISANS) was used to investigate the depth-dependent orientation of self-assembled morphologies of block copolymers. The orientation of the microphase-separated structure differed at the polymer-air and polymer-substrate interfaces, as revealed by GISANS. The conversion of a hydrophobic-hydrophobic block copolymer to a hydrophilic-hydrophobic block copolymer significantly increased the segmental interaction parameter and caused microphase separation of the copolymers.
Article
Chemistry, Physical
Katie L. Browning, Andrew S. Westover, James F. Browning, Mathieu Doucet, Robert L. Sacci, Gabriel M. Veith
Summary: This study demonstrates the measurement of the interface between Li metal and the solid electrolyte LiPON using neutron reflectometry and in situ electrochemistry, which confirms the presence of a thin interphase less than 7 nm thick. The interphase is found to be a chemical gradient consisting of a Li-rich layer that gradually blends into pure LiPON. The ability to create ideal solid-solid interphases thinner than 10 nm holds significant importance in facilitating the adoption of high efficiency next generation solid state batteries, as well as providing a more general methodology for studying buried solid-solid interfaces across applications.
ACS ENERGY LETTERS
(2023)
Article
Polymer Science
Zhan Chen, Christian Steinmetz, Mingqiu Hu, E. Bryan Coughlin, Hanyu Wang, William T. Heller, Wim Bras, Thomas P. Russell
Summary: The study demonstrates that star block copolymers (s-BCPs) can accumulate at the interface between two immiscible homopolymers and promote adhesion. The molecular weight and number of s-BCPs at the interface affect the width and morphology of the interface. Lower molecular weight s-BCPs generate wider interfaces, while higher molecular weight s-BCPs inhibit phase separation. Additionally, s-BCPs are more efficient in promoting adhesion compared to linear block copolymers.
Article
Multidisciplinary Sciences
Yan N. Fang, Artem M. Rumyantsev, Angelika E. Neitzel, Heyi Liang, William T. Heller, Paul F. Nealey, Matthew V. Tirrell, Juan J. de Pablo
Summary: Polyelectrolyte complexation plays a crucial role in materials science and biology. Small-angle scattering experiments have revealed structural similarities between polyelectrolyte complexes (PECs) and semidilute solutions of neutral polymers. The positional correlations between polyanion and polycation charges have been investigated using small-angle neutron scattering profiles, and the results indicate the existence of Coulomb repulsions between polyanion fragments and their attractions to polycations. The addition of salt leads to the disappearance of these correlations, reverting the scattering functions to an Ornstein-Zernike form.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2023)
Article
Chemistry, Multidisciplinary
Sarah J. Blair, Mathieu Doucet, Valerie A. Niemann, Kevin H. Stone, Melissa E. Kreider, James F. Browning, Candice E. Halbert, Hanyu Wang, Peter Benedek, Eric J. McShane, Adam C. Nielander, Alessandro Gallo, Thomas F. Jaramillo
Summary: One way to improve the electrochemical synthesis of ammonia through the Li-mediated N-2 reduction reaction (Li-NRR) is by cycling the current between open-circuit conditions and periods of applied current density. In this study, the dynamics of the electrode-electrolyte interface under Li-NRR conditions during current cycling were investigated using in situ time-resolved neutron reflectometry and grazing-incidence synchrotron X-ray diffraction. The observations suggest that the benefits associated with returning to open-circuit conditions may be related to the concomitant loss of Li-containing species from a thin layer at the cathode surface into a porous SEI layer that becomes filled with electrolyte or dissolves.
ENERGY & ENVIRONMENTAL SCIENCE
(2023)
Article
Chemistry, Multidisciplinary
Dustin Eby, Mikolaj Jakowski, Valeria Lauter, Mathieu Doucet, Panchapakesan Ganesh, Miguel Fuentes-Cabrera, Rajeev Kumar
Summary: Diblock copolymers undergo microphase separation due to repulsive interactions between dissimilar monomers. In thin films, additional effects make microphase separation more complicated. Physics-based models have been used to extract interaction parameters, but they are time-intensive and prone to errors. This work develops an alternative method using neural networks to extract parameters from neutron scattering length density profiles, paving the way for automated analysis using machine learning tools.
Article
Computer Science, Interdisciplinary Applications
Addi Malviya-Thakur, David E. Bernholdt, William F. Godoy, Gregory R. Watson, Mathieu Doucet, Mark A. Coletti, David M. Rogers, Marshall McDonnell, Jay Jay Billings, Barney Maccabe
Summary: Research software engineers (RSE) play a vital role in scientific discoveries by leading the development of applications and tools that enable supercomputers to process vast volumes of data. This article describes the mission, culture, and practices of RSE teams at Oak Ridge National Laboratory (ORNL) and highlights the importance of team dynamics, work ethics, and evolving skill sets for scientific innovation.
COMPUTING IN SCIENCE & ENGINEERING
(2022)