Themed collection Machine Learning and Data Science in Materials Design

11 items
Editorial

Machine learning and data science in materials design: a themed collection

Guest Editors Andrew Ferguson and Johannes Hachmann introduce this themed collection of papers.

Graphical abstract: Machine learning and data science in materials design: a themed collection
Open Access Communication

Can machine learning identify the next high-temperature superconductor? Examining extrapolation performance for materials discovery

Traditional machine learning (ML) metrics overestimate model performance for materials discovery.

Graphical abstract: Can machine learning identify the next high-temperature superconductor? Examining extrapolation performance for materials discovery
From the themed collection: MSDE most-read Q1 2019
Paper

Correlative analysis of metal organic framework structures through manifold learning of Hirshfeld surfaces

We demonstrate the use of non-linear manifold learning methods to map the connectivity and extent of similarity between diverse metal–organic framework (MOF) structures in terms of their surface areas by taking into account both crystallographic and electronic structure information.

Graphical abstract: Correlative analysis of metal organic framework structures through manifold learning of Hirshfeld surfaces
Paper

A simple constrained machine learning model for predicting high-pressure-hydrogen-compressor materials

Here we present the results of using techno-economic analysis as constraints for machine learning guided studies of new metal hydride materials.

Graphical abstract: A simple constrained machine learning model for predicting high-pressure-hydrogen-compressor materials
From the themed collection: Industry R&D collection
Paper

Tuning the molecular weight distribution from atom transfer radical polymerization using deep reinforcement learning

Combination of deep reinforcement learning and atom transfer radical polymerization gives precise in silico control on polymer molecular weight distributions.

Graphical abstract: Tuning the molecular weight distribution from atom transfer radical polymerization using deep reinforcement learning
Paper

Classification of spatially resolved molecular fingerprints for machine learning applications and development of a codebase for their implementation

Direct mapping between material structures and properties for various classes of materials is often the ultimate goal of materials researchers.

Graphical abstract: Classification of spatially resolved molecular fingerprints for machine learning applications and development of a codebase for their implementation
Paper

Enriched optimization of molecular properties under constraints: an electrochromic example

We present a deterministic optimization procedure of molecular properties that ensures diverse coverage of the given chemical compound search space.

Graphical abstract: Enriched optimization of molecular properties under constraints: an electrochromic example
Paper

Understanding structural adaptability: a reactant informatics approach to experiment design

The structural and electronic adaptability of a vanadium selenite framework is determined using cheminformatics data and machine learning algorithms.

Graphical abstract: Understanding structural adaptability: a reactant informatics approach to experiment design
Paper

Molecular dynamics simulations and PRISM theory study of solutions of nanoparticles and triblock copolymers with solvophobic end blocks

Hybrid materials composed of inorganic nanoparticles (NPs) and amphiphilic block copolymers (BCPs) combine desirable properties of NPs with the rich phase behavior of BCPs, making them attractive for use in biomaterials, responsive materials for sensing, active materials in robotics, etc.

Graphical abstract: Molecular dynamics simulations and PRISM theory study of solutions of nanoparticles and triblock copolymers with solvophobic end blocks
Paper

Deep learning for chemical reaction prediction

We describe a deep learning-based system for predicting chemical reactions and identifying experimentally-observed masses.

Graphical abstract: Deep learning for chemical reaction prediction
From the themed collection: MSDE most-read Q1 2019
Paper

Statistical models are able to predict ionic liquid viscosity across a wide range of chemical functionalities and experimental conditions

Herein we present a method of developing predictive models of viscosity for ionic liquids (ILs) using publicly available data in the ILThermo database and the open-source software toolkits PyChem, RDKit, and SciKit-Learn.

Graphical abstract: Statistical models are able to predict ionic liquid viscosity across a wide range of chemical functionalities and experimental conditions
11 items

About this collection

From MSDE

Guest Edited by Professor Andrew Ferguson (University of Illinois at Urbana-Champaign, USA) and Professor Johannes Hachmann (University at Buffalo, USA), this collection showcases the latest research leveraging data analytics and machine learning approaches to guide the design of hard, soft, and biological materials with tailored properties, function and behaviour.

Data-driven modelling and machine learning have opened new paths to the understanding, engineering, and design of materials. Physical laws define the fundamental connection between materials chemistry and emergent structure and function, and theoretical or numerical models based on these laws provide a route to quantitative predictions.

The inverse problem, however, is far more challenging: reverse engineering a material with particular functionality or behaviour. Approaches and tools from data sciences and machine learning are proving enormously successful in addressing this task, ultimately informing experimental synthesis and characterization efforts of novel materials, compounds, and reactions.

Spotlight

Advertisements