Deep Neural Network‐Driven Development of Coking‐ and Sulfur‐Resistant Nanostructured NiGa Alloy‐Based Anodes for Solid Oxide Fuel Cells
Advanced Energy Materials · 2026-10 · 16 (40)
DOI: 10.1002/aenm.71569 ↗Kurz gefasst
Fachartikel in der Zeitschrift Advanced Energy Materials, online erschienen am 09.09.2026. Verfasst von Haipeng Zhang und 13 weiteren Autor:innen. Die Arbeit nennt 60 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Solid oxide fuel cells (SOFCs) that directly convert natural gas into electricity are promising high‐efficiency energy‐conversion devices, yet conventional Ni‐based anodes suffer from severe carbon deposition and sulfur poisoning, leading to rapid electrode deactivation. Although binary alloying effectively regulates surface reaction pathways and improves anode stability, its development largely relies on experience‐driven trial‐and‐error screening. Here, we propose a rational screening framework that integrates microkinetic analysis with deep neural networks to identify the Ni‐Ga alloy from a vast binary alloy chemical space. Density functional theory confirms that such a Ni‐Ga alloy combines efficient methane activation with intrinsic resistance to carbon deposition. Guided by these predictions, we further construct a nanocomposite electrode with Gd 0.1 Ce 0.9 O 1.95 (GDC)‐encapsulated Ni‐Ga alloys (NiGa‐GDC). Benefiting from alloy‐induced electronic structure modulation and a nanocomposite electrode architecture, the NiGa‐GDC nanocomposite anode exhibits exceptional electrochemical performance and coking resistance. In nearly dry methane fuel (3% H 2 O), a corresponding electrolyte‐supported single cell achieves both high peak power density and durable operation, far outperforming the counterpart Ni‐GDC nanocomposite anode. Moreover, the NiGa‐GDC anode demonstrates remarkable resistance to sulfur poisoning. This study provides a new strategy for developing highly active and durable nickel‐based nanocomposite anodes for the direct utilization of natural gas.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Zhang, H., Wang, H., Lin, C., Chen, Z., Qian, J., Cheng, Z., Long, D., Guan, C., Ai, N., Jiang, S. P., Tian, D., Lin, Z., Chen, K., & Zhang, J. (2026). Deep Neural Network‐Driven Development of Coking‐ and Sulfur‐Resistant Nanostructured NiGa Alloy‐Based Anodes for Solid Oxide Fuel Cells. Advanced Energy Materials, 16(40). https://doi.org/10.1002/aenm.71569
Vancouver
Zhang H, Wang H, Lin C, Chen Z, Qian J, Cheng Z, et al. Deep Neural Network‐Driven Development of Coking‐ and Sulfur‐Resistant Nanostructured NiGa Alloy‐Based Anodes for Solid Oxide Fuel Cells. Advanced Energy Materials. 2026;16(40). doi:10.1002/aenm.71569
BibTeX
@article{Zhang2026,
author = {Zhang, Haipeng and Wang, Hao and Lin, Chenghui and Chen, Zhiyi and Qian, Jiaqi and Cheng, Zixiang and Long, Dewu and Guan, Chengzhi and Ai, Na and Jiang, San Ping and Tian, Dong and Lin, Zijing and Chen, Kongfa and Zhang, Jiujun},
title = {Deep Neural Network‐Driven Development of Coking‐ and Sulfur‐Resistant Nanostructured NiGa Alloy‐Based Anodes for Solid Oxide Fuel Cells},
journal = {Advanced Energy Materials},
year = {2026},
volume = {16},
number = {40},
doi = {10.1002/aenm.71569}
}
EUVIMED