Dataset Information

Designing exceptional gas-separation polymer membranes using machine learning.

ABSTRACT: The field of polymer membrane design is primarily based on empirical observation, which limits discovery of new materials optimized for separating a given gas pair. Instead of relying on exhaustive experimental investigations, we trained a machine learning (ML) algorithm, using a topological, path-based hash of the polymer repeating unit. We used a limited set of experimental gas permeability data for six different gases in ~700 polymeric constructs that have been measured to date to predict the gas-separation behavior of over 11,000 homopolymers not previously tested for these properties. To test the algorithm's accuracy, we synthesized two of the most promising polymer membranes predicted by this approach and found that they exceeded the upper bound for CO2/CH4 separation performance. This ML technique, which is trained using a relatively small body of experimental data (and no simulation data), evidently represents an innovative means of exploring the vast phase space available for polymer membrane design.

SUBMITTER: Barnett JW

PROVIDER: S-EPMC7228755 | biostudies-literature | 2020 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Designing exceptional gas-separation polymer membranes using machine learning.

Barnett J Wesley JW Bilchak Connor R CR Wang Yiwen Y Benicewicz Brian C BC Murdock Laura A LA Bereau Tristan T Kumar Sanat K SK

Science advances 20200515 20

The field of polymer membrane design is primarily based on empirical observation, which limits discovery of new materials optimized for separating a given gas pair. Instead of relying on exhaustive experimental investigations, we trained a machine learning (ML) algorithm, using a topological, path-based hash of the polymer repeating unit. We used a limited set of experimental gas permeability data for six different gases in ~700 polymeric constructs that have been measured to date to predict the ...[more]

PMID: 32440545

Similar Datasets

Project description:To prevent greenhouse emissions into the atmosphere, separations like CO2/CH4 and CO2/N2 from natural gas, biogas, and flue gasses are crucial. Polymer membranes gained a key role in gas separations over the past decades, but these polymers are often not organized at a molecular level, which results in a trade-off between permeability and selectivity. In this work, the effect of molecular order and orientation in liquid crystals (LCs) polymer membranes for gas permeation is demonstrated. Using the self-assembly of polymerizable LCs to prepare membranes ensures control over the supramolecular organization and alignment of the building blocks at a molecular level. Robust freestanding LC membranes were fabricated that have various, distinct morphologies (isotropic, nematic cybotactic, and smectic C) and alignment (planar and homeotropic), while using the same chemical composition. Single gas permeation data show that the permeability decreases with increasing molecular order while the ideal gas selectivity of He and CO2 over N2 increases tremendously (36-fold for He/N2 and 21-fold for CO2/N2) when going from randomly ordered to the highly ordered smectic C morphology. The calculated diffusion coefficients showed a 10-fold decrease when going from randomly ordered membranes to ordered smectic C membranes. It is proposed that with increasing molecular order, the free volume elements in the membrane become smaller, which hinders gasses with larger kinetic diameters (Ar, N2) more than gasses with smaller kinetic diameters (He, CO2), inducing selectivity. Comparison of gas sorption and permeation performances of planar and homeotropic aligned smectic C membranes shows the effect of molecular orientation by a 3-fold decrease of the diffusion coefficient of homeotropic aligned smectic C membranes resulting in a diminished gas permeation and increased ideal gas selectivities. These results strongly highlight the importance of molecular order and orientation in LC polymer membranes for gas separation.

Dataset Information

Designing exceptional gas-separation polymer membranes using machine learning.

Publications

Designing exceptional gas-separation polymer membranes using machine learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets