Platonic representation of foundation machine learning interatomic potentials
Z. Li, A. Walsh. Platonic representation of foundation machine learning interatomic potentials. arXiv:2512.05349, 2025.
CV
Education
- Ph.D. in Physical Chemistry, Peking University, China, 2012-2017
- Visiting Ph.D. Student, Research Laboratory of Electronics, MIT, USA, 2016
- B.S. in Chemistry, Shandong University, China, 2008-2012
Work Experience
- 2023-present: Research Fellow (AI in Science)
- Imperial College London, Department of Materials
- Schmidt Sciences funded fellowship
- Supervisor: Prof. Aron Walsh
- 2020-2023: Research Associate
- Imperial College London, Department of Materials
- Research on perovskite solar cells and machine learning for materials
- Supervisor: Prof. Aron Walsh
- 2017-2020: Post-doctorate
- Soochow University, Department of Physics
- Machine learning for halide perovskite materials design
- Supervisor: Prof. Wan-Jian Yin
Research Skills
- Theoretical Modelling
- Electronic properties: band structures, photoabsorption, carrier transport (VASP, Wannier90)
- Vibrational phonon properties: thermodynamics of chalcogenide perovskites (VASP, Phonopy)
- Molecular dynamics: ML force fields for phase transitions
- Machine Learning for Materials
- Optimization: Reinforcement learning, Bayesian optimization, Evolutionary algorithms
- ML methods: Crystal graph neural networks, VAE, scikit-learn
- Tools: Materials Project, Matminer, Matbench, M3Gnet, CHGnet, MACE
- Programming
- Languages: Python, PyTorch
- High-throughput: Atomate2-MongoDB pipeline
Software Development
- PlatonicRep: Platonic representation analysis of ML interatomic potentials
- NH3SOFC: Ammonia solid oxide fuel cell simulation toolkit
- Polytype: Perovskite polytypes predictions
- ChalcogenideGrowth: Chalcogenide perovskite growth modelling
Publications
Thermodynamics of Sulfur Vacancy Formation in the Chalcogenide Perovskite BaZrS3
Z. Li, A. Walsh. Thermodynamics of Sulfur Vacancy Formation in the Chalcogenide Perovskite BaZrS3. J. Phys. Chem. C, 2025.
Dual oxidation suppression in lead-free perovskites for low-threshold and long-lifespan lasing
Y. Li, Z. Li, Y. Han, et al. Dual oxidation suppression in lead-free perovskites for low-threshold and long-lifespan lasing. Adv. Mater., 2418931, 2025.
Has generative artificial intelligence solved inverse materials design?
H. Park, Z. Li, A. Walsh. Has generative artificial intelligence solved inverse materials design? Matter, 7, 2355-2367, 2024.
Stoichiometry-tunable synthesis and magnetic property exploration of two-dimensional chromium selenides
F. Cui, K. He, S. Wu, H. Zhang, Y. Lu, Z. Li, et al. Stoichiometry-tunable synthesis and magnetic property exploration of two-dimensional chromium selenides. ACS Nano, 18, 6276-6285, 2024.
A hole-selective self-assembled monolayer for both efficient perovskite and organic solar cells
M. Li#, Z. Li#, M. Liu, et al. A hole-selective self-assembled monolayer for both efficient perovskite and organic Solar Cells. Langmuir, 40, 4772-4778, 2024.
Molecularly engineered self-assembled monolayers as effective hole-selective layers for organic solar cells
M. Li#, Z. Li#, H. Fu, et al. Molecularly engineered self-assembled monolayers as effective hole-selective layers for organic solar cells. ACS Appl. Energy Mater., 7, 1306-1312, 2024.
Structural dynamics toolkit for metal halide perovskites
L. Xia, W. Baldwin, J. Klarbring, Z. Li, G. Csányi, A. Walsh. Structural dynamics toolkit for metal halide perovskites. J. Phys. Chem. C, 127, 19141-19151, 2023.
From cubic to hexagonal: electronic trends across metal halide perovskite polytypes
Z. Li, J.-S. Park, A. M. Ganose, A. Walsh. From cubic to hexagonal: electronic trends across metal halide perovskite polytypes. J. Phys. Chem. C, 127, 12695-12701, 2023.
Inhomogeneous Defect Distributions in Mixed-Polytype Metal Halide Perovskites
Y. Woo, Z. Li, Y.-K. Jung, J.-S. Park, A. Walsh. Inhomogeneous Defect Distributions in Mixed-Polytype Metal Halide Perovskites. ACS Energy Lett., 8, 356-360, 2023.
Lone pair driven anisotropy in antimony chalcogenide semiconductors
X. Wang, Z. Li, S. R. Kavanagh, A. M. Ganose, A. Walsh. Lone pair driven anisotropy in antimony chalcogenide semiconductors. Phys. Chem. Chem. Phys., 24, 7195-7202, 2022.
Secondary bonding channel design induces intercalation pseudocapacitance toward ultrahigh-capacity and high-rate organic electrodes
Z. Hu, X. Zhao, Z. Li, et al. Secondary bonding channel design induces intercalation pseudocapacitance toward ultrahigh-capacity and high-rate organic electrodes. Adv. Mater., 2104039, 2021.
Evolutionary exploration of polytypism in lead halide perovskites
Z. Li, J.-S. Park, A. Walsh. Evolutionary exploration of polytypism in lead halide perovskites. Chem. Sci., 12, 12165-12173, 2021.
Efficient potential-tuning strategy through p-type doping for designing cathodes with ultrahigh energy density
Z. Wang, D. Wang, Z. Zou, T. Song, D. Ni, Z. Li, et al. Efficient potential-tuning strategy through p-type doping for designing cathodes with ultrahigh energy density. Natl Sci. Rev., 7, 11, 1768-1775, 2020.
Hexagonal stacking faults act as hole-blocking layers in lead halide perovskites
J.-S. Park, Z. Li, J. N. Wilson, W.-J. Yin, A. Walsh. Hexagonal stacking faults act as hole-blocking layers in lead halide perovskites. ACS Energy Lett., 5, 2231-2233, 2020.
Bayesian optimization based on a unified figure of merit for accelerated materials screening: a case study of halide perovskites
X. Chen, C. Wang, Z. Li, Z. Hou, W.-J. Yin. Bayesian optimization based on a unified figure of merit for accelerated materials screening: a case study of halide perovskites. Sci. China Mater., 63, 1024-1035, 2020.
Coordination assembly of 2D ordered organic metal chalcogenides with widely tunable electronic band gaps
Y. Li, X. Jiang, Z. Fu, Q. Huang, G.-E. Wang, W.-H. Deng, C. Wang, Z. Li, W.-J. Yin, B. Chen, G. Xu. Coordination assembly of 2D ordered organic metal chalcogenides with widely tunable electronic band gaps. Nat. Comm., 11, 261, 2020.
Confining MOF-derived SnSe Nanoplatelets in nitrogen-doped graphene cages via direct CVD for durable sodium ion storage
C. Lu#, Z. Li#, Z. Xia, et al. Confining MOF-derived SnSe Nanoplatelets in nitrogen-doped graphene cages via direct CVD for durable sodium ion storage. Nano Res., 12, 3051-3058, 2019.
PECVD-derived graphene nanowall/lithium composite anodes towards highly stable lithium metal batteries
Z. Hu#, Z. Li#, Z. Xia, et al. PECVD-derived graphene nanowall/lithium composite anodes towards highly stable lithium metal batteries. Energy Storage Mater., 22, 29-29, 2019.
Plasmon-free surface-enhanced raman spectroscopy using metallic 2D materials
X. Song, Y. Wang, F. Zhao, et al. Plasmon-free surface-enhanced raman spectroscopy using metallic 2D materials. ACS Nano, 13(7), 8312-8319, 2019.
Large-area synthesis of superclean graphene via selective etching of amorphous carbon with carbon dioxide
J. Zhang#, K. Jia#, L. Lin#, et al. Large-area synthesis of superclean graphene via selective etching of amorphous carbon with carbon dioxide. Angew. Chem. Int. Ed., 58, 1-6, 2019.
Copper-containing carbon feedstock for growing super-clean graphene
K. Jia#, J. Zhang#, L. Lin#, Z. Li, et al. Copper-containing carbon feedstock for growing super-clean graphene. J. Am. Chem. Soc., 141, 7670-7674, 2019.
Superhydrophilic graphdiyne accelerates interfacial mass/electron transportation to boost electrocatalytic and photoelectrocatalytic water oxidation activity
J. Li, X. Gao, Z. Li, et al. Superhydrophilic graphdiyne accelerates interfacial mass/electron transportation to boost electrocatalytic and photoelectrocatalytic water oxidation activity. Adv. Funct. Mater., 1808079, 2019.
Thermodynamic stability landscape of halide double perovskite via high-throughput computing and machine learning
Z. Li, Q. Xu, Q. Sun, Z. Hou, W.-J. Yin. Thermodynamic stability landscape of halide double perovskite via high-throughput computing and machine learning. Adv. Funct. Mater., 1807280, 2019.
Oxide perovskites, double perovskites and derivatives for electrocatalysis, photocatalysis, and photovoltaics
W.-J. Yin#, B. Weng#, J. Ge#, Q. Sun, Z. Li, Y. Yan. Oxide perovskites, double perovskites and derivatives for electrocatalysis, photocatalysis, and photovoltaics. Energy Environ. Sci., 12, 442-462, 2019.
Perspective of machine learning in material design
Z. Sun, Z. Li, G. Chen, Q. Xu, Z. Hou, W.-J. Yin. Perspective of machine learning in material design. Chin. Sci. Bulletin, 2019.
Rationalizing perovskites data for machine learning and materials design
Q. Xu, Z. Li, M. Liu, W.-J. Yin. Rationalizing perovskites data for machine learning and materials design. J. Phys. Chem. Lett., 9, 6948-6954, 2018.
Low-temperature heteroepitaxy of 2D PbI2/Graphene for large-area flexible photodetectors
J. Zhang, Y. Huang, Z. Tan, et al. Low-temperature heteroepitaxy of 2D PbI2/Graphene for large-area flexible photodetectors. Adv. Mater., 1803194, 2018.
Nanostructured Bi2S3 encapsulated within three-dimensional N-doped graphene as active and flexible anodes for sodium-ion batteries
C. Lu#, Z. Li#, L. Yu, et al. Nanostructured Bi2S3 encapsulated within three-dimensional N-doped graphene as active and flexible anodes for sodium-ion batteries. Nano Res., 11(9), 4614-4626, 2018.
Recent progress in Pb-free stable inorganic double halide perovskites
Z. Li, W.-J. Yin. Recent progress in Pb-free stable inorganic double halide perovskites. J. Semicond., 39(7), 2018.
Growth of defect-engineered graphene on manganese oxides for Li-ion storage
K. Chen, F. Zhang, J. Sun, Z. Li, et al. Growth of defect-engineered graphene on manganese oxides for Li-ion storage. Energy Storage Mater., 12, 110-118, 2018.
Monitoring local strain vector in atomic-layered MoSe2 by second-harmonic generation
J. Liang#, J. Zhang#, Z. Li#, et al. Monitoring local strain vector in atomic-layered MoSe2 by second-harmonic generation. Nano Lett., 17(12), 7539-7543, 2017.
Anisotropic carrier mobility in two-dimensional materials with tilted Dirac cones: theory and application
T. Cheng#, H. Lang#, Z. Li, Z. Liu, Z. Liu. Anisotropic carrier mobility in two-dimensional materials with tilted Dirac cones: theory and application. Phys. Chem. Chem. Phys., 19, 23942-23950, 2017.
Movement of Dirac points and band gaps in graphyne under rotating strain
Z. Li, Z. Liu, Z. Liu. Movement of Dirac points and band gaps in graphyne under rotating strain. Nano Res., 10(6), 2005-2020, 2017.
Architecture of β-graphdiyne contained thin film using modified Glaser Hay coupling reaction for enhanced photocatalytic property of TiO2
J. Li#, Z. Xie#, Y. Xiong, Z. Li, et al. Architecture of β-graphdiyne contained thin film using modified Glaser Hay coupling reaction for enhanced photocatalytic property of TiO2. Adv. Mater., 1700421, 2017.
Chemical vapor deposition growth of linked carbon monolayers with acetylenic scaffoldings on silver foil
R. Liu, X. Gao, J. Zhou, H. Xu, Z. Li, et al. Chemical vapor deposition growth of linked carbon monolayers with acetylenic scaffoldings on silver foil. Adv. Mater., 1604665, 2017.
Raman spectra and corresponding strain effects in graphyne and graphdiyne
S. Zhang, J. Wang, Z. Li, R. Zhao, L. Tong, Z. Liu, J. Zhang, Z. Liu. Raman spectra and corresponding strain effects in graphyne and graphdiyne. J. Phys. Chem. C, 120(19), 10606-10613, 2016.
Intrinsic carrier mobility of Dirac cones: the limitations of deformation potential theory
Z. Li, J. Wang, Z. Liu. Intrinsic carrier mobility of Dirac cones: the limitations of deformation potential theory. J. Chem. Phys., 141(14), 144107/1-9, 2014.
Talks
2025 1st InnoCORE Symposium: AI for Energy System - from Materials to Automatization
Invited Talk at UNIST, South Korea
Machine Learning for Science 2025
Invited Talk at World Young Scientist Summit, Wenzhou, China
AI for Discovery and Research Automation
Invited Talk at Nature Communications & Nature Machine Intelligence Virtual Conference, Virtual
AI4Matter
Invited Talk at Nanyang Technological University, Singapore
Metastable transition metal dichalcogenides from machine learning force fields
Oral Presentation at CIMTEC 2024, Montecatini Terme, Italy
Defect control and ab initio thermodynamics for synthesizing chalcogenide perovskite
Oral Presentation at E-MRS 2023 Spring Meeting, Strasbourg, France
Towards chemical accuracy for materials thermodynamics
Invited Talk at CPE Workshop on Artificial Intelligence in Soft Electronic Materials, London, UK
Data-driven materials design
Invited Talk at Imperial College London, London, UK
Atomic scale insights into metal halide perovskite
Invited Talk at Snow Mountain Summit, London, UK
Evolutionary exploration of halide perovskite polytypes
Oral Presentation at ICAMD2021, Jeju, South Korea
Teaching
- Imperial-X and AIMS AI in Science Workshop - “Reinforcement learning for energy materials design”, 4 hrs, 30 students, 2025
- Leading Teaching Assistant - “Machine learning for materials science” with Prof. Aron Walsh, 20 hrs, 100 students, 2024-2025
- Guest Lecturer - “Reinforcement learning for energy materials design”, 1 hr, 50 students, 2024
- Course Designer - “Machine learning for materials science” with Prof. Aron Walsh, 2023-2024
Awards
- Imperial-AIMS Education Fund, Imperial College London (2024)
- AI for Science Postdoctoral Fellowship, Imperial College London (2023)
- ICAMD2021 Young Researcher Award & Best Oral Presentation Prize (2021)
- May-Fourth Scholarship, Peking University (2016)
- Graduate Student Scholarship for International Exchange, Peking University (2016)
- Guang-hua Scholarship, Peking University (2015)
- HORIBA Scientific Scholarship, Peking University (2014)
Service
- Reviewer for journals under ACS, RSC, and IOP
- Guest Editor for Energies
- Invited Reviewer for ICLR 2023
- Co-organizer: AIMS-Imperial Workshop in Agriculture, Health and Energy, Cameroon (2024)