Machine-learning-assisted thin-film growth: Bayesian optimization in molecular beam epitaxy of SrRuO3 thin films

  title={Machine-learning-assisted thin-film growth: Bayesian optimization in molecular beam epitaxy of SrRuO3 thin films},
  author={Yuki K. Wakabayashi and Takuma Otsuka and Yoshiharu Krockenberger and Hiroshi Sawada and Yoshitaka Taniyasu and Hideki Yamamoto},
  journal={APL Materials},
Materials informatics exploiting machine learning techniques, e.g., Bayesian optimization (BO), has the potential to offer high-throughput optimization of thin-film growth conditions through incremental updates of machine learning models in accordance with newly measured data. Here, we demonstrated BO-based molecular beam epitaxy (MBE) of SrRuO3, one of the most-intensively studied materials in the research field of oxide electronics, mainly owing to its unique nature as a ferromagnetic metal… 

Figures from this paper

Machine-learning-assisted fabrication: Bayesian optimization of laser-induced graphene patterning using in-situ Raman analysis
Abstract The control of the physical, chemical, and electronic properties of laser-induced graphene (LIG) is crucial in the fabrication of flexible electronic devices. However, the optimization of
Accelerated HKUST-1 Thin-Film Property Optimization Using Active Learning.
The understanding of high-quality MOF thin-film formation using solution shearing is improved by correlating the processing parameter sets and their corresponding film coverage by incorporating an active learning approach into the solution-sheared HKUST-1 thin- film-processing parameters.
Toward Precise n-Type Doping Control in MOVPE-Grown β-Ga2O3 Thin Films by Deep-Learning Approach
In this work, we train a hybrid deep-learning model (fDNN, Forest Deep Neural Network) to predict the doping level measured from the Hall Effect measurement at room temperature and to investigate the
Application of Bayesian optimization for high-performance TiOx/SiOy/c-Si passivating contact
Abstract We report on the application of Bayesian optimization (BO), which could accelerate the time-intensive process optimization of many parameters, to fabrication of the high-performance titanium
Single-domain perpendicular magnetization induced by the coherent O 2p -Ru 4d hybridized state in an ultra-high-quality SrRuO3 film
We investigated the Ru 4d and O 2p electronic structure and magnetic properties of an ultra-high-quality SrRuO3 film on SrTiO3 grown by machine-learning-assisted molecular beam epitaxy. The high
Wide-Range Epitaxial Strain Control of Electrical and Magnetic Properties in High-Quality SrRuO3 Films
Epitaxial strain in 4d ferromagnet SrRuO3 films is directly linked to the physical properties through the strong coupling between lattices, electrons, and spins. It provides an excellent opportunity
Machine learning-driven optimization in powder manufacturing of Ni-Co based superalloy
Abstract The process parameters in powder manufacturing must be optimized to produce high-quality powders with desired sizes depending on the use. Machine learning-driven optimization was applied to
Emergence of Weyl fermions in an epitaxial ferromagnetic oxide
Magnetic Weyl fermions, which occur in magnets, have novel transport phenomena related to pairs of Weyl nodes, and they are, of both, scientific and technological interest, with the potential for use
Machines for Materials and Materials for Machines: Metal-Insulator Transitions and Artificial Intelligence
In this perspective, we discuss the current and future impact of artificial intelligence and machine learning for the purposes of better understanding phase transitions, particularly in correlated
High-mobility two-dimensional carriers from surface Fermi arcs in magnetic Weyl semimetal films
High-mobility two-dimensional carriers originating from surface Fermi arcs in magnetic Weyl semimetals are highly desired for accessing exotic quantum transport phenomena and for topological


Accelerated discovery of metallic glasses through iteration of machine learning and high-throughput experiments
This work trains a machine learning model on previously reported observations, parameters from physiochemical theories, and makes it synthesis method–dependent to guide high-throughput experiments to find a new system of metallic glasses in the Co-V-Zr ternary, and provides a quantitatively accurate, synthesis method-sensitive predictor for metallic glasses that improves performance with use and thus promises to greatly accelerate discovery of many new metallic glasses.
Molecular beam epitaxy : from research to mass production
This multi-contributor handbook discusses Molecular Beam Epitaxy (MBE), an epitaxial deposition technique which involves laying down layers of materials with atomic thicknesses on to substrates. It
Enhanced metallic properties of SrRuO3 thin films via kinetically controlled pulsed laser epitaxy
Metal electrodes are a universal element of all electronic devices. Conducting SrRuO3 (SRO) epitaxial thin films have been extensively used as electrodes in complex-oxide heterostructures due to good
Machine-Learning-Assisted Development and Theoretical Consideration for the Al2Fe3Si3 Thermoelectric Material.
Preliminary results suggest that Al2Fe3Si3 can have a low thermal conductivity and the calculated formation energies of point defects suggest that the antisite defects between Al and Si are likely to cause the Al andSi off-stoichiometries in Al2 Fe3 Si3.
High‐throughput and fully automated system for molecular‐beam epitaxy
A new high‐throughput and fully automated system for molecular‐beam epitaxy was developed by employing multiple substrate mount pallets and automatic pallet transfer mechanisms. The geometrical
Machine learning modeling of superconducting critical temperature
A team led by Valentin Stanev from the University of Maryland at College Park and including researchers from Duke University and NIST develops several machine learning schemes to model the critical temperature (Tc) of over 12,000 known superconductors and candidate materials.
Multi-source MBE with high-precision rate control system as a synthesis method sui generis for multi-cation metal oxides
Abstract Molecular beam epitaxy (MBE) has been an excellent tool to design artificial heterostructures in the research field of semiconductors by means of an alternate stacking of existing lattices.
Predictions of new AB O 3 perovskite compounds by combining machine learning and density functional theory
We apply machine learning (ML) methods to a database of 390 experimentally reported $AB{\mathrm{O}}_{3}$ compounds to construct two statistical models that predict possible new perovskite materials
Control of the growth and domain structure of epitaxial SrRuO3 thin films by vicinal (001) SrTiO3 substrates
We report the effect of both miscut angle (α) and miscut direction (β) of vicinal substrates on the epitaxial growth and domain structure of isotropic metallic oxide SrRuO3 thin films. The thin films
Machine learning with systematic density-functional theory calculations: Application to melting temperatures of single- and binary-component solids
A combination of systematic density-functional theory (DFT) calculations and machine learning techniques has a wide range of potential applications. This study presents an application of the