Gas-Conditioned Message Passing Graph Neural Network for Permeability Prediction in Polymeric Membranes
- 1McKetta Department of Chemical Engineering
- 2University of Texas at Austin
- 3Saudi Aramco
Abstract
Designing polymeric membranes for diverse gas-separation applications remains a critical challenge. Machine learning (ML) models can significantly accelerate this effort by capturing complex relationships between polymer-gas pairs and enabling permeability prediction over an extended chemical space of gases and polymers. In this work, we present a novel ML framework, the gas-conditioned message passing neural network (GC-MPNN), that jointly combines gas descriptors and polymer chemistry in a fused neural network model. Polymers are represented at the monomer level and encoded as molecular graphs, where node and edge information is propagated through message-passing layers. Six gas species (He, H2, N2, O2, CH4, and CO2) are represented using multiple descriptor sets spanning thermodynamic, kinetic, and electrostatic properties, along with one-hot encodings and combined feature sets that include kinetic plus thermodynamic descriptors and the full physicochemical descriptor set. For each descriptor combination, model hyperparameters are optimized using Bayesian optimization within a leave-one-gas-out (LOGO) validation framework. Test performance is then evaluated on the gas species excluded entirely from the corresponding training fold (out-of-training set) and compared across descriptor sets. For most gas species, the model achieves strong predictive performance (R2 ∼ 0.86–0.96), while CO2 remains more challenging in terms of absolute prediction accuracy (R2 ∼ 0.44). Active learning sampling further improves model performance by adding a few labeled samples from the test gas species into the training set. Among the three strategies tested, deep ensemble-based sampling and random sampling show strong improvement, with R2 reaching approximately ∼ 0.95 for all gas species, while evidential regression shows less consistent improvement. The model is further evaluated on a smaller polymer-gas dataset containing hydrocarbon species (C2H4, C2H6, C3H6, C3H8, and n-C4H10) using the GC-MPNN trained only on the six base gases, achieving moderate to strong Pearson correlations (r ∼ 0.77–0.89) for these out-of-training gases. Overall, this work presents a machine learning model that can learn transferable polymer-gas relationships from known gas species and extend them to previously unseen gases.
Related
- Nanostructure of PEGDA-PEG hydrogel membranes and how it controls their permeability
- Blood–Brain Barrier Permeability Prediction by Integrating Deep Learning Representations with Chemical Compound Knowledge
- ToSCA: Total Scattering Conditioned Variational Autoencoder for Atomic Structure Prediction from Pair Distribution Function Data
- Prediction of MOF Adsorption Isotherms and High-Throughput Screening Using a Physics-Informed Membranes Graph Neural Network
- Machine Learning Optimization of Polymeric Complex Coacervate Micelles Reveals Cell-Type-Specific Formulations for Enhanced mRNA Delivery
BibTeX
@misc{mukherjee2026gasconditioned,
title = {{Gas-Conditioned Message Passing Graph Neural Network for Permeability Prediction in Polymeric Membranes}},
author = {Krishnendu Mukherjee and Zidan Zhang and Mohammed Alshammasi and Mohammed G. Hashim and Hussain H. Naji and Zainab A. Aithan and Jihad A. Badra and Jalal Yagoubi and Hussain B. Tuwailib and Ali Hayek and Venkat Ganesan},
howpublished = {ChemRxiv},
year = {2026},
month = oct,
doi = {10.26434/chemrxiv.15010381/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010381/v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Gas-Conditioned Message Passing Graph Neural Network for Permeability Prediction in Polymeric Membranes
AU - Krishnendu Mukherjee
AU - Zidan Zhang
AU - Mohammed Alshammasi
AU - Mohammed G. Hashim
AU - Hussain H. Naji
AU - Zainab A. Aithan
AU - Jihad A. Badra
AU - Jalal Yagoubi
AU - Hussain B. Tuwailib
AU - Ali Hayek
AU - Venkat Ganesan
PY - 2026
DA - 2026/10/13
PB - ChemRxiv
DO - 10.26434/chemrxiv.15010381/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010381/v1
AB - Designing polymeric membranes for diverse gas-separation applications remains a critical challenge. Machine learning (ML) models can significantly accelerate this effort by capturing complex relationships between polymer-gas pairs and enabling permeability prediction over an extended chemical space of gases and polymers. In this work, we present a novel ML framework, the gas-conditioned message passing neural network (GC-MPNN), that jointly combines gas descriptors and polymer chemistry in a fused neural network model. Polymers are represented at the monomer level and encoded as molecular graphs, where node and edge information is propagated through message-passing layers. Six gas species (He, H2, N2, O2, CH4, and CO2) are represented using multiple descriptor sets spanning thermodynamic, kinetic, and electrostatic properties, along with one-hot encodings and combined feature sets that include kinetic plus thermodynamic descriptors and the full physicochemical descriptor set. For each descriptor combination, model hyperparameters are optimized using Bayesian optimization within a leave-one-gas-out (LOGO) validation framework. Test performance is then evaluated on the gas species excluded entirely from the corresponding training fold (out-of-training set) and compared across descriptor sets. For most gas species, the model achieves strong predictive performance (R2 ∼ 0.86–0.96), while CO2 remains more challenging in terms of absolute prediction accuracy (R2 ∼ 0.44). Active learning sampling further improves model performance by adding a few labeled samples from the test gas species into the training set. Among the three strategies tested, deep ensemble-based sampling and random sampling show strong improvement, with R2 reaching approximately ∼ 0.95 for all gas species, while evidential regression shows less consistent improvement. The model is further evaluated on a smaller polymer-gas dataset containing hydrocarbon species (C2H4, C2H6, C3H6, C3H8, and n-C4H10) using the GC-MPNN trained only on the six base gases, achieving moderate to strong Pearson correlations (r ∼ 0.77–0.89) for these out-of-training gases. Overall, this work presents a machine learning model that can learn transferable polymer-gas relationships from known gas species and extend them to previously unseen gases.
ER -