A machine learning (ML) integrated workflow was utilized to guide the design of Al-Ni co-doping CoCrFe-based high-entropy alloys (HEAs) for improved corrosion resistance. ML directs the design of HEAs to chemical composition consisting of Co, Cr, Fe, Ni, Al with phase evolved from FCC plus BCC to single FCC. The corrosion behavior of CoCrFe-based HEAs with decreased Al-Ni was systematically investigated and the corrosion mechanism was elucidated. Co-doping Al-Ni content of 50 % resulted in pitting initiation due to the potential difference between two phases. The AlNi199 alloy exhibited superior corrosion resistance, whose corrosion current density is 6.7×10−8 A/cm2, pitting potential is 816.9 mV, and passivation region is 945.5 mV, respectively. The experimental results are in accordance with the ML perdition, indicating the superiority of ML in improving corrosion performance.

中文翻译：

---

机器学习辅助发现具有优异耐腐蚀性的相变Al-Ni共掺杂高熵合金


利用机器学习 (ML) 集成工作流程来指导 Al-Ni 共掺杂 CoCrFe 基高熵合金 (HEA) 的设计，以提高耐腐蚀性。 ML 将 HEA 的设计指导为由 Co、Cr、Fe、Ni、Al 组成的化学成分，相从 FCC 加 BCC 演变为单一 FCC。系统研究了 Al-Ni 含量降低的 CoCrFe 基 HEA 的腐蚀行为，并阐明了腐蚀机理。共掺杂 Al-Ni 含量为 50% 时，由于两相之间的电位差，导致点蚀萌生。 AlNi199合金表现出优异的耐腐蚀性能，其腐蚀电流密度为6.7×10−8 A/cm2，点蚀电位为816.9 mV，钝化区为945.5 mV。实验结果与ML结果一致，表明ML在改善腐蚀性能方面具有优越性。