Physics-Informed Machine Learning for Fast Screening High Hydrogen Storage MOFs with Monotonicity Constraints
Xuanjie Chen1, Chunjian Pan1, Shaojun Ren2
1College of Automation Engineering, Shanghai University of Electric Power, Shanghai 200090, P.R. China.
Journal of Chemical Theory and Computation
|May 7, 2025
Summary
Physics-Informed Neural Networks (PINNs) offer a physically consistent approach to screening Metal-Organic Frameworks (MOFs) for efficient hydrogen storage. This method successfully identified MOFs meeting U.S. Department of Energy targets for a carbon-neutral economy.
Area of Science:
- Materials Science
- Computational Chemistry
- Energy Storage
Background:
- Hydrogen is a key alternative energy carrier for a carbon-neutral future.
- Current hydrogen storage methods face efficiency and cost challenges.
- Metal-Organic Frameworks (MOFs) show promise for hydrogen storage due to high surface area and tunability.
Purpose of the Study:
- To develop a physically consistent machine learning model for screening MOFs for hydrogen storage.
- To identify high-capacity MOFs that meet specific hydrogen storage targets.
- To overcome limitations of traditional machine learning models lacking physical consistency.
Main Methods:
- Development of a Physics-Informed Neural Network (PINN) incorporating crystallographic-property relationships.
- Integration of monotonic relationships into the neural network architecture.
- Validation of top-performing MOFs using Grand Canonical Monte Carlo (GCMC) simulations.
Main Results:
- PINN successfully screened MOFs for high hydrogen storage capacity.
- Identified MOFs exhibited meaningful crystallographic features via heatmap analysis.
- Two experimentally synthesized MOFs, LADQEM_CSD17 and LADQEM01_CSD17, met U.S. DOE 2025 onboard hydrogen storage targets.
Conclusions:
- PINNs provide a physically consistent and effective method for MOF screening in hydrogen storage.
- The identified MOFs represent viable candidates for advanced hydrogen energy systems.
- This approach accelerates the discovery of materials for efficient and cost-effective hydrogen storage.
Related Concept Videos
00:26Hydrogen Bonds
Hydrogen bonds are weak attractions between atoms that have formed other chemical bonds. One of these atoms is electronegative, like oxygen, and has a partial negative charge. The other is a hydrogen atom that has bonded with another electronegative atom and has a partial positive charge.
Hydrogen Bonds Control the World!
Because hydrogen has very weak electronegativity when it binds with a strongly electronegative atom, such as oxygen or nitrogen, electrons in the bond are unequally shared.
Hydrogen Bonds Control the World!
Because hydrogen has very weak electronegativity when it binds with a strongly electronegative atom, such as oxygen or nitrogen, electrons in the bond are unequally shared.
00:52¹H NMR of Conformationally Flexible Molecules: Temporal Resolution
At room temperature, the chair conformer of cyclohexane undergoes rapid ring flipping between two equivalent chair conformers at a rate of approximately 105 times per second. These two chair conformers are in equilibrium. The rapid ring flipping results in the interconversion of the axial proton to an equatorial proton and an equatorial to the axial proton. Such interconversions are too rapid and cannot be detected on the NMR timescale. Hence, the NMR spectrometer cannot distinguish between the...
01:15¹H NMR of Conformationally Flexible Molecules: Variable-Temperature NMR
The axial and equatorial protons in cyclohexane can be distinguished by performing a variable-temperature NMR experiment. In this process, except for one proton, the remaining eleven protons are replaced by deuterium. The deuterium substitution avoids the possible peak splitting caused by the spin-spin coupling between the adjacent protons. The remaining proton flips between the axial and equatorial positions.
01:04Hydrogen Bonds
A hydrogen bond is formed when a weakly positive hydrogen atom already bonded to one electronegative atom (for example, the oxygen in the water molecule) is attracted to another electronegative atom from another polar molecule, such as water (H2O), hydrogen fluoride (HF), or ammonia (NH3). The huge electronegativity difference between the H atom (2.1) and the atom to which it is bonded (4.0 for an F atom, 3.5 for an O atom, or 3.0 for an N atom), combined with the very small size of an H atom...
01:26Constraints and Statical Determinacy
In structural engineering, the equilibrium of a system is not only determined by its equations of equilibrium but also with the help of constraints. Constraints refer to restrictions on the motion of a system. The proper combinations of constraints can minimize the total number of constraints needed to maintain a system in mechanical equilibrium. When this happens, the system is said to be statically determinate. For such systems, the unknown reaction supports can be estimated using equilibrium...
01:29Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...