Prediction of MOF Adsorption Isotherms and High-Throughput Screening Using a Physics-Informed Membranes Graph Neural Network
- 1School of Chemical Engineering, Ocean and Life Sciences
- 2Dalian University of Technology
Abstract
Metal-organic frameworks (MOFs) are prime CO 2 -capture candidates, but grand canonical Monte Carlo (GCMC) screening is costly when full isotherms, not single uptakes, are required; machine learning surrogates rarely resolve the continuous pressure response. We report a physics-informed pressure-aware multi-task graph neural network (PI-MGNN) to predict CO 2 uptake at arbitrary pressure and reconstruct isotherms from structure. From ~130,000 candidates, PCA and K-means select 5,023 representative MOFs, each simulated by GCMC at five pressures spanning 0.1-5.0 MPa, enforcing a continuous pressure response. Crystal graphs are processed by decoupled geometric and electrostatic graph-attention branches with auxiliary supervision, while a Gaussian-expanded pressure query is fused by cross-attention. PI-MGNN attains R 2 = 0.977 and MAE = 0.291 mol kg -1 ; isotherm MAEs are 0.428/0.312/0.330 mol kg -1 for high/medium/low-uptake MOFs, and cross-attention improves accuracy by 6.3%. Attention weights recover the pressure-resolved physics, enabling accurate and interpretable high-throughput screening.
Related
- Moringa oleifera Alginate-Based Beads for Adsorption of Selected Heavy Metals: Preparation, Kinetics, Isotherms, and Reusability Studies
- Beyond Materials Rankings: When Adsorption or Membranes Win for CO2/CH4 Separation
- Microscale high-throughput screening-assisted discovery of a photo/copper dual catalytic system for benzylic C(sp2)–C(sp3) Suzuki–Miyaura coupling: data mining and mechanistic insights
- Yield Prediction and Condition Recommendation on High-Throughput Experimentation Datasets: Encoding Effects and Noise Ceilings
- A Fluorescence‐Based High‐Throughput Screening Method for EGFR Inhibitors Using an EGFR‐Targeted Environment‐Sensitive Fluorescent Probe
BibTeX
@misc{linfeng2026prediction,
title = {{Prediction of MOF Adsorption Isotherms and High-Throughput Screening Using a Physics-Informed Membranes Graph Neural Network}},
author = {Li Linfeng and Xingshuo Fan and Ning Zhang},
howpublished = {ChemRxiv},
year = {2026},
month = oct,
doi = {10.26434/chemrxiv.15010295/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010295/v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Prediction of MOF Adsorption Isotherms and High-Throughput Screening Using a Physics-Informed Membranes Graph Neural Network
AU - Li Linfeng
AU - Xingshuo Fan
AU - Ning Zhang
PY - 2026
DA - 2026/10/12
PB - ChemRxiv
DO - 10.26434/chemrxiv.15010295/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010295/v1
AB - Metal-organic frameworks (MOFs) are prime CO 2 -capture candidates, but grand canonical Monte Carlo (GCMC) screening is costly when full isotherms, not single uptakes, are required; machine learning surrogates rarely resolve the continuous pressure response. We report a physics-informed pressure-aware multi-task graph neural network (PI-MGNN) to predict CO 2 uptake at arbitrary pressure and reconstruct isotherms from structure. From ~130,000 candidates, PCA and K-means select 5,023 representative MOFs, each simulated by GCMC at five pressures spanning 0.1-5.0 MPa, enforcing a continuous pressure response. Crystal graphs are processed by decoupled geometric and electrostatic graph-attention branches with auxiliary supervision, while a Gaussian-expanded pressure query is fused by cross-attention. PI-MGNN attains R 2 = 0.977 and MAE = 0.291 mol kg -1 ; isotherm MAEs are 0.428/0.312/0.330 mol kg -1 for high/medium/low-uptake MOFs, and cross-attention improves accuracy by 6.3%. Attention weights recover the pressure-resolved physics, enabling accurate and interpretable high-throughput screening.
ER -