Machine learning enabled discovery of application dependent design principles for two-dimensional materials
Problem
Large-scale searches for high-performing two-dimensional materials were limited by the high computational cost of first-principles density functional theory calculations.
Why it matters
2D materials are promising next-generation candidates for applications like mechanical strength and photovoltaics, but exhaustive DFT screening is too costly.
What the AI did
Extended crystal graph convolutional neural networks to planar periodic systems and trained an ensemble to predict thermodynamic, mechanical and electronic properties, then screened nearly 45,000 structures.
Results
Screened nearly 45,000 structures for mechanical strength and photovoltaics; approach recovered well-accepted design rules (e.g., lead/tin hybrid organic-inorganic perovskites)
Impact
The evidence supports ML-enabled large-scale screening that recovers known design rules for 2D materials.
Record
| Venue | Machine Learning Science and Technology |
|---|---|
| Published | 2020 |
| Work type | article |
| Authors | Victor Venturi, Holden L Parks, Zeeshan Ahmad, Venkatasubramanian Viswanathan |
| Citations | 18 (OpenAlex, at import time) |
| References | 47 |
| Model or method | Crystal graph convolutional neural network ensemble |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1088/2632-2153/aba002 |
| Open access | gold |
| Keywords | two-dimensional materials, crystal graph convolutional neural networks, materials screening, photovoltaics, machine learning |
Importance rating
significant — thousands to millions affected
Abstract states screening of nearly 45,000 structures for 2D materials relevant to next-generation applications like photovoltaics and mechanical strength, a genuinely new large-scale capability.
More in Engineering & Materials
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Scaling deep learning for materials discovery | 2023 | 1,471 |
8
|
| 2 | Application of Machine Learning in Material Synthesis and Property Prediction | 2023 | 165 |
7
|
| 3 | Machine learning-guided discovery of ionic polymer electrolytes for lithium metal batteries | 2023 | 115 |
7
|
| 4 | Machine Learning for Structural Health Monitoring of Aerospace Structures: A Review | 2025 | 44 |
7
|
| 5 | Power System Transient Stability Assessment Based on Machine Learning Algorithms and Grid Topology | 2023 | 42 |
7
|