journal article·2025-12Open access

A computational framework for interface design using lattice matching, machine learning potentials, and active learning: A case study on LaCoO3/La2NiO4

Guangchen Liu, Songge Yang, Yu Zhong

Abstract

Interfaces play a crucial role in determining the stability and functional properties of materials, yet their complex nature poses significant challenges for computational modeling. While Density Functional Theory (DFT) offers accurate insights into interfacial behavior, its high computational cost limits its applicability to large-scale studies. To address this, a scalable computational framework integrating lattice matching algorithms, machine learning potentials, and active learning is developed to enable efficient and accurate interface design. The LaCoO 3 /La 2 NiO 4 interface is employed as a case study to demonstrate the framework's capabilities in systematically exploring interface stability, geometry, and composition. Customized interface structures are generated via lattice matching, and atomic simulations are accelerated using machine learning potentials. Key geometric factors—including interface area, shape, and thickness—are systematically evaluated, while active learning is used to optimize the exploration of interfacial orientations and compositions, enabling efficient identification of low-energy configurations. A predictive model is established, linking interfacial characteristics to stability with DFT-level fidelity at reduced computational cost. This framework offers a versatile, data-driven approach for complex interface modeling and design.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Machine Learning in Materials Science · Magnetic and transport properties of perovskites and related materials · Electronic and Structural Properties of Oxides

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (51)

Show source-provided references
1. Surface and interface engineering of nanoarrays toward advanced electrodes and electrochemical energy storage devices10.1002/adma.202004959 ↗
2. Interface engineering for organic electronics10.1002/adfm.200902236 ↗
3. Surface and interface control in nanoparticle catalysis10.1021/acs.chemrev.9b00220 ↗
4. Strain engineering of 2D materials: issues and opportunities at the interface
5. Interface-assisted synthesis of 2D materials: trend and challenges10.1021/acs.chemrev.8b00056 ↗
6. Interfaces and interphases in multicomponent materials: past, present, future10.1016/j.eurpolymj.2004.10.035 ↗
7. Electrode polarization and charge transport at solid interfaces10.1103/PhysRevB.80.184301 ↗
8. Chemical reactivity descriptor for the oxide-electrolyte interface in Li-ion batteries10.1021/acs.jpclett.7b01655 ↗
9. Surface and interface engineering of electrode materials for lithium‐ion batteries10.1002/adma.201402962 ↗
10. Electrodeposition and mechanical stability at lithium-solid electrolyte interface during plating in solid-state batteries
11. Efficient alloy design of Sr-modified A356 alloys driven by computational thermodynamics and machine learning10.1016/j.jmst.2021.09.061 ↗
12. Boosting for concept design of casting aluminum alloys driven by combining computational thermodynamics and machine learning techniques
13. A machine learning accelerated distributed task management system (Malac-Distmas) and its application in high-throughput CALPHAD computations aiming at efficient alloy design10.1016/j.apmate.2021.09.005 ↗
14. Accelerated discovery of high-performance Al-Si-Mg-Sc casting alloys by integrating active learning with high-throughput CALPHAD calculations10.1080/14686996.2023.2196242 ↗
15. High-entropy materials design by integrating the first-principles calculations and machine learning: a case study in the al-co-cr-fe-ni system
16. Accelerating high-throughput searches for new alloys with active learning of interatomic potentials10.1016/j.commatsci.2018.09.031 ↗
17. Neural network potentials: a concise overview of methods10.1146/annurev-physchem-082720-034254 ↗
18. Accessing thermal conductivity of complex compounds by machine learning interatomic potentials10.1103/PhysRevB.100.144308 ↗
19. Efficient machine-learning based interatomic potentialsfor exploring thermal conductivity in two-dimensional materials
20. Improving accuracy of interatomic potentials: more physics or more data? A case study of silica10.1016/j.mtcomm.2018.11.008 ↗
21. Active learning of linearly parametrized interatomic potentials10.1016/j.commatsci.2017.08.031 ↗
22. Lattice dynamics simulation using machine learning interatomic potentials10.1016/j.commatsci.2019.109333 ↗
23. Machine-learning interatomic potentials enable first-principles multiscale modeling of lattice thermal conductivity in graphene/borophene heterostructures10.1039/D0MH00787K ↗
24. Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning10.1103/PhysRevB.99.064114 ↗
25. Towards universal neural network potential for material discovery applicable to arbitrary combination of 45 elements10.1038/s41467-022-30687-9 ↗
26. DeePMD-kit: a deep learning package for many-body potential energy representation and molecular dynamics10.1016/j.cpc.2018.03.016 ↗
27. Schnet–a deep learning architecture for molecules and materials10.1063/1.5019779 ↗
28. ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost10.1039/C6SC05720A ↗
29. A universal graph deep learning interatomic potential for the periodic table10.1038/s43588-022-00349-3 ↗
30. Solid oxide fuel cell electrolyte review10.1016/0272-8842(95)00101-8 ↗
31. A systematic ab initio study of vacancy formation energy, diffusivity, and ionic conductivity of Ln2NiO4+δ (Ln= La, Nd, Pr)10.1016/j.jpowsour.2023.233200 ↗
32. Crystal structure study of Perovskite-type LaCoO3 electro-catalyst synthesized by Pechini method
33. Defect chemistry and surface properties of LaCoO310.1039/b002168g ↗
34. A comparative study of the Ruddlesden-popper series, Lan+ 1NinO3n+ 1 (n= 1, 2 and 3), for solid-oxide fuel-cell cathode applications10.1016/j.ssi.2006.05.005 ↗
35. Surface structures and defect properties of pure and doped La2NiO410.1039/b102678j ↗
36. Lattice match: an application to heteroepitaxy10.1063/1.333084 ↗
37. The atomic simulation environment—a Python library for working with atoms
38. On the limited memory BFGS method for large scale optimization10.1007/BF01589116 ↗
39. R. Fletcher, Practical Methods of Optimization, John Wiley & Sons2000.10.1002/9781118723203 ↗
40. Sur la loi de répartition de certaines grandeurs géographiques
41. Pytorch: an imperative style, high-performance deep learning library
42. Optuna: a next-generation hyperparameter optimization framework
43. Tree-structured parzen estimator: understanding its algorithm components and their roles for better empirical performance
44. Neural networks for machine learning lecture 6a overview of mini-batch gradient descent
45. Online algorithms and stochastic approximations
46. Adam: a method for stochastic optimization
47. On finite‐size effects in computer simulations using the Ewald potential10.1063/1.470721 ↗
48. Hansen · 2013
49. Adhesion at diamond/metal interfaces: a density functional theory study10.1063/1.3277013 ↗
50. First-principles calculation of W/WC interface: atomic structure, stability and electronic properties10.1016/j.apsusc.2014.10.119 ↗
51. The austenite/ɛ martensite interface: a first-principles investigation of the fcc Fe (111)/hcp Fe (0001) system10.1016/j.apsusc.2012.06.059 ↗