Understanding the three-dimensional multiscale porous microstructures by applying deep neural networks

• Project/Area Number: 21K14090
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 19020:Thermal engineering-related
• Research Institution: The University of Tokyo
• Principal Investigator: SCIAZKO Anna 東京大学, 生産技術研究所, 特任助教 (30898945)
• Project Period (FY): 2021-04-01 – 2023-03-31
• Project Status: Completed (Fiscal Year 2022)
• Budget Amount: ¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000); Fiscal Year 2022: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000); Fiscal Year 2021: ¥2,990,000 (Direct Cost: ¥2,300,000、Indirect Cost: ¥690,000)
• Keywords: solid oxide fuel cell / machine learning / 3D microstructure / FIB-SEM / GAN network / semantic segmentation / super-resolution / artificial structure / porous material / porous electrode

Outline of Research at the Start

The multiscale porous media found interest in many fields of engineering. In particular, the porous electrode microstructure determines the performances of fuel cells and batteries. The multi-sized pore design is beneficial as large pores enhance the gas transport and nano-pores increase active reaction area. Here, the characterization methods of multi-sized porous media based on deep neural networks will be proposed. The focus is to provide high resolution large volume 3D characterization, fabricate synthetic 3D structure from single 2D image and correlate microstructure with performance.

Outline of Final Research Achievements

An automated microstructure analysis framework using machine learning was developed. The framework was built to improve three-dimensional (3D) microscopy analysis by super resolution and semantic segmentation algorithms. The developed framework can contribute in shortening actual measurement time up to 8 times and data post-processing time by two orders of magnitude. Furthermore, a novel method with generative network was proposed to create an artificial 3D microstructure model from a single two-dimensional image. Additionally, the generative network trained with microstructure datasets, can fabricate realistic microstructure models with predefined properties and gradients. The automated segmentation framework was used to characterize gadolinium doped ceria ceria-based solid oxide fuel (SOFC) anodes with the controlled properties (porosity, material composition, and particle size) and it enables to reconstruct SOFC anode with carbon deposition.

Academic Significance and Societal Importance of the Research Achievements

The porous media found interest in many fields of engineering. Particularly, the porous electrode microstructure determines the performances of fuel cells and batteries. This study proposed comprehensive framework for analyzing porous media microstructures based on machine learning methods.

Research Products

• [Journal Article] 3D microstructures of solid oxide fuel cell Ni-YSZ anodes with carbon deposition (2023)
  • Author(s): Sciazko Anna、Komatsu Yosuke、Nakamura Akiko、Ouyang Zhufeng、Hara Toru、Shikazono Naoki
  • Journal Title: Chemical Engineering Journal Volume: 460 Pages: 141680-141680
  • DOI: 10.1016/j.cej.2023.141680
  • Peer Reviewed / Int'l Joint Research
• [Journal Article] Effects of mass fraction of La0.9Sr0.1Cr0.5Mn0.5O3-δ and Gd0.1Ce0.9O2-δ composite anodes for nickel free solid oxide fuel cells (2022)
  • Author(s): Sciazko Anna、Komatsu Yosuke、Yokoi Ryosuke、Shimura Takaaki、Shikazono Naoki
  • Journal Title: Journal of the European Ceramic Society Volume: 42 Issue: 4 Pages: 1556-1567
  • DOI: 10.1016/j.jeurceramsoc.2021.11.039
  • Peer Reviewed / Int'l Joint Research
• [Journal Article] Operando observations of active three phase boundary of patterned nickel - yttria stabilized zirconia electrode in solid oxide cell (2022)
  • Author(s): Ouyang, Z., Komatsu, Y., Sciazko, A., Onishi, J., Nishimura, K. and Shikazono, N.
  • Journal Title: J. Power Sources Volume: 529 Pages: 231228-231228
  • DOI: 10.1016/j.jpowsour.2022.231228
  • Peer Reviewed / Int'l Joint Research
• [Presentation] Machine Learning Methods for Predicting Microstructural Changes in Solid Oxide Cell Electrodes (2022)
  • Author(s): Sciazko, A., Yamagishi, R., Komatsu, Y., Ouyang, Z., Onishi, J., Nishimura, K., Shikazono, N.
  • Organizer: Materials Science and Technology 2022 (MS&T22), Pittsburgh, Pennsylvania, USA
  • Int'l Joint Research / Invited
• [Presentation] 機械学習による SOFC微細構造変化の予測 (2022)
  • Author(s): シチョンシコ アンナ，小松 洋介，山岸 鈴奈，鹿園 直毅
  • Organizer: 第26回動力・エネルギー技術シンポジウム講演論文集，佐賀
  • Invited
• [Presentation] Microstructures of Ni-GDC electrodes with carbon deposition (2022)
  • Author(s): Sciazko, A., Komatsu, Y., Nakamura, A., Sunada, Y., Ouyang, Z., Hara, T. and Shikazono, N.
  • Organizer: 15th European SOFC & SOE Forum, B1104, Lucerne, Switzerland
  • Int'l Joint Research
• [Presentation] Synthesizing Electrode Microstructures with Predefined Spatial Gradients By Conditional Generative Adversarial Networks (2022)
  • Author(s): Yamagishi, R., Sciazko, A., Komatsu, Y., and Shikazono, N.
  • Organizer: Proc. 241th ECS meeting, I06-1083, Vancouver, Canada
  • Int'l Joint Research
• [Presentation] 析出還元法によるナノNi-GDCアノードの発電特性と微細構造 (2022)
  • Author(s): Sciazko A., 砂田祐輔，小松洋介，鹿園直毅
  • Organizer: 第31回SOFC研究発表会講演要旨集
• [Presentation] Prediction of Microstructure Evolutions in Solid Oxide Cell Electrodes with Unsupervised Image-to-Image Translation Networks (2022)
  • Author(s): Sciazko, A., Komatsu, Y. and Shikazono, N.
  • Organizer: 46th International Conference and Expo on Advanced Ceramics and Composites (ICACC2022)
  • Int'l Joint Research
• [Presentation] NiO Reduction Prediction with Unsupervised Image-to-Image Translation Network (2021)
  • Author(s): Sciazko A., Komatsu, Y., Shikazono, N.
  • Organizer: 第30回SOFC研究発表会
• [Presentation] 3-D microstructures of Ni-YSZ anodes with carbon deposition (2021)
  • Author(s): Sciazko, A, Komatsu, Y., Ouyang, Z. and Shikazono, N.
  • Organizer: 6th Asian SOFC Symposium and Exhibition
  • Int'l Joint Research
• [Presentation] Degradation predictions of solid oxide cells through Electrochemical Impedance Spectra with Long Short-Term Memory Neural Network (2021)
  • Author(s): Yamagishi, R., Sciazko, A., Komatsu, Y., Nishimura, K. and Shikazono, N.
  • Organizer: 6th Asian SOFC Symposium and Exhibition
  • Int'l Joint Research
• [Presentation] Unsupervised Generative Adversarial Network for 3-D Microstructure Synthesis from 2-D Image (2021)
  • Author(s): Sciazko, A., Komatsu, Y. and Shikazono, N.
  • Organizer: ECS Transactions, Volume 103, Number 1
  • Int'l Joint Research
• [Presentation] 非等方な解像度を有するFIB-SEMデータセットからの電極構造再構築 (2021)
  • Author(s): SCIAZKO ANNA, 小松洋介, 志村敬彬, 鹿園直毅
  • Organizer: 第25回 動力・エネルギー技術シンポジウム