Development of Scalable and Generalizable Machine Learned Force Field for Polymers

Mohanty, Shaswat; Stevenson, James; Browning, Andrea; Jacobson, Leif D.; Leswing, Karl; Afzal, Mohammad Atif Faiz

doi:10.26434/chemrxiv-2023-4z2bw

Cited by 5 publications

(6 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We performed MD simulations for all the structures at each experimental temperatures in the viscosity dataset to evaluate whether the inclusion of MD descriptors would improve ML models. For all simulations, we used the Schrödinger's Materials Science Suite (MSS) [46], which leverages the Desmond MD engine to rapidly speed up MD computations through GPU acceleration [7,47,48]. All molecules were parameterized with the OPLS4 force field [49].…”

Section: Classical Molecular Dynamics Simulationsmentioning

confidence: 99%

“…However, measuring a large number of experimental viscosities is challenging, costly, and limited based on the availability of compounds. Alternative to experiments, much effort has been invested in obtaining viscosity using physics-based modeling, such as molecular dynamics (MD) simulations [4,6,7]. Despite advancements in simulation procedures, estimating viscosities from MD is especially challenging for highly viscous systems greater than ∼5 cP and is computationally expensive, making MD simulations challenging to use for the high-throughput screening of viscosities.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Advancing Material Property Prediction: Using Physics-Informed Machine Learning Models for Viscosity

Chew,

Sender,

Kaplan

et al. 2024

Preprint

View full text Add to dashboard Cite

In materials science, accurately computing properties like viscosity, melting point, and glass transition temperatures solely through physics-based models is challenging. Data-driven machine learning (ML) also poses challenges in constructing ML models, especially in the material science domain where data is limited. To address this, we integrate physics-informed descriptors from molecular dynamics (MD) simulations to enhance the accuracy and interpretability of ML models. Our current study focuses on accurately predicting viscosity in liquid systems using MD descriptors. In this work, we curated a comprehensive dataset of over 4,000 small organic molecules’ viscosities from scientific literature, publications, and online databases. This dataset enabled us to develop quantitative structure–property relationships (QSPR) consisting of descriptor-based and graph neural network models to predict temperature-dependent viscosities for a wide range of viscosities with considerable accuracy. The QSPR models reveal that including MD descriptors improves prediction accuracies of experimental viscosities, particularly at the small data set scale of fewer than a thousand data points. Furthermore, feature importance tools reveal that intermolecular interactions captured by MD descriptors are most important for accurate viscosity predictions. Finally, the QSPR models can accurately capture the inverse relationship between viscosity and temperature for six battery-relevant solvents, some of which were not included in the original data set. Our research highlights the effectiveness of incorporating MD descriptors into QSPR models, which leads to improved accuracy for properties that are difficult to predict when using physics-based models alone or when limited data is available.

show abstract

Section: Classical Molecular Dynamics Simulationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Advancing Material Property Prediction: Using Physics-Informed Machine Learning Models for Viscosity

Chew,

Sender,

Kaplan

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Out of these, a total of 217,684 clusters were used to compute the DFT energies, atomic forces, and dipoles. All the DFT data generated in this work is shared on figshare platform [54]. All DFT calculations were performed at the ωB97X-D3BJ/def2-TZVPD level [27,28,55] with the electronic structure software package Psi4-1.3 [56].…”

Section: Dataset Preparation For Qrnn Trainingmentioning

confidence: 99%

“…All the DFT data generated in this work is available on the figshare platform [54]. We provide additional data on cluster sampling and modeling methods in the Supplementary appendices.…”

Section: Conflict Of Interestmentioning

confidence: 99%

Development of Scalable and Generalizable Machine Learned Force Field for Polymers

Mohanty,

Stevenson,

Browning

et al. 2023

Preprint

View full text Add to dashboard Cite

Understanding and predicting the properties of polymers is vital to developing tailored polymer molecules for desired applications. Classical force fields may fail to capture key properties, for example, the transport properties of certain polymer systems such as polyethylene glycol. As a solution, we present an alternative potential energy surface, a charge recursive neural network (QRNN) model trained on DFT calculations made on smaller atomic clusters that generalizes well to oligomers comprising larger atomic clusters or longer chains. We demonstrate the validity of the polymer QRNN workflow by modeling the oligomers of ethylene glycol. We apply two rounds of active learning (addition of new training clusters based on current model performance) and implement a novel model training approach that uses partial charges from a semi-empirical method. Our developed QRNN model for polymers produces stable molecular dynamics (MD) simulation trajectory and captures the dynamics of polymer chains as indicated by the striking agreement with experimental values. Our model allows working on much larger systems than allowed by DFT simulations, at the same time providing a more accurate force field than classical force fields which provides a promising avenue for large-scale molecular simulations of polymeric systems.

show abstract

“…Recently, GNN-based force fields have been used to predict MD trajectories using approaches such as SchNet [15], and DeepMD [11,16,17]. These force fields, when trained on high-quality ab initio data, can perform large-scale simulations at an accuracy comparable to that of ab initio model but only at a small fraction of the cost [18,19,20]. However, before GNN force fields (or any machine-learned force fields) can be confidently deployed, their transferability to configurations beyond the training data set must be established.…”

Section: Introductionmentioning

confidence: 99%

Evaluating the Transferability of Machine-Learned Force Fields for Material Property Modeling

Mohanty¹,

Yoo²,

Kang³

et al. 2023

Preprint

View full text Add to dashboard Cite

Machine-learned force fields have generated significant interest in recent years as a tool for molecular dynamics (MD) simulations, with the aim of developing accurate and efficient models that can replace classical interatomic potentials. However, before these models can be confidently applied to materials simulations, they must be thoroughly tested and validated. The existing tests on the radial distribution function and meansquared displacements are insufficient in assessing the transferability of these models.Here we present a more comprehensive set of benchmarking tests for evaluating the transferability of machine-learned force fields. We use a graph neural network (GNN)based force field coupled with the OpenMM package to carry out MD simulations for Argon as a test case. Our tests include computational X-ray photon correlation spectroscopy (XPCS) signals, which capture the density fluctuation at various length scales in the liquid phase, as well as phonon density-of-state in the solid phase and the liquid-solid phase transition behavior. Our results show that the model can accurately capture the behavior of the solid phase only when the configurations from the solid phase are included in the training dataset. This underscores the importance of appropriately selecting the training data set when developing machine-learned force fields. The tests presented in this work provide a necessary foundation for the development and application of machine-learned force fields for materials simulations.

show abstract

Development of Scalable and Generalizable Machine Learned Force Field for Polymers

Cited by 5 publications

References 29 publications

Advancing Material Property Prediction: Using Physics-Informed Machine Learning Models for Viscosity

Advancing Material Property Prediction: Using Physics-Informed Machine Learning Models for Viscosity

Development of Scalable and Generalizable Machine Learned Force Field for Polymers

Evaluating the Transferability of Machine-Learned Force Fields for Material Property Modeling

Contact Info

Product

Resources

About