Scholarlink

<p>This study introduces an integrated design and optimization framework for AlCoCrFeNi-based composites reinforced with graphene nanoplatelets (GNPs), combining machine learning-driven predictive modelling with experimental validation to achieve uniform reinforcement distribution across a predefined microstructure. A comprehensive dataset was constructed incorporating ten elements (Al, Co, Cr, Fe, Ni, Cu, Mn, Ti, V, Mo) with carbon additions, where carbon specifically represents various reinforcement phases, including graphene nanoplatelets, enabling diverse carbon-based reinforcement strategies. A comparative analysis of five machine learning models was performed using six thermodynamic descriptors: valence electron concentration (VEC), atomic size difference (δ), electronegativity difference (Δχ), mixing enthalpy (ΔH<sub>mix</sub>), mixing entropy (ΔS<sub>mix</sub>) and Gibbs energy of mixing (ΔG<sub>mix</sub>) for the 750-sample dataset. Both random forest (RF) and extreme gradient boosting (XGBoost) demonstrated superior predictive accuracy (test accuracy: 0.94927; receiver-operating characteristic curve - area under the curve (ROC-AUC): 0.99446-0.99627; 10-fold cross-validation: 0.83315-0.85184), consistently identifying ΔH<sub>mix</sub> as the most critical feature influencing phase stability. X-ray diffraction confirmed dual-phase (FCC + BCC) structures in all samples, with BCC phase volume fraction increasing from 77.25% (base high entropy alloy (HEA)) to 85.198% (2 wt.% GNP), indicating GNP-induced BCC stabilization. Peak shifts and broadening suggested lattice strain from carbon dissolution and thermal mismatch. Energy-dispersive x-ray spectroscopy (EDS) mapping revealed uniform GNP dispersion and aluminium segregation, aiding BCC stability. Zener pinning analysis demonstrated increased pinning pressure and grain size reduction from 55&#xa0;μm to 50&#xa0;μm. GNP reinforcement enhances phase control and microstructural stability in HEA composites, making them promising candidates for structural applications.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Effect of GNP Reinforcement on Phase Stability and Microstructure of AlCoCrFeNi HEA Composites: A Data-Driven Approach

  • 期刊论文

  • 发表日期：

    2026年5月21日

查看全文
  • Renju C. Daniel,
  • Mohammed Mudassir,
  • C. N. Shyam Kumar,
  • R. Anand Sekhar

摘要

This study introduces an integrated design and optimization framework for AlCoCrFeNi-based composites reinforced with graphene nanoplatelets (GNPs), combining machine learning-driven predictive modelling with experimental validation to achieve uniform reinforcement distribution across a predefined microstructure. A comprehensive dataset was constructed incorporating ten elements (Al, Co, Cr, Fe, Ni, Cu, Mn, Ti, V, Mo) with carbon additions, where carbon specifically represents various reinforcement phases, including graphene nanoplatelets, enabling diverse carbon-based reinforcement strategies. A comparative analysis of five machine learning models was performed using six thermodynamic descriptors: valence electron concentration (VEC), atomic size difference (δ), electronegativity difference (Δχ), mixing enthalpy (ΔHmix), mixing entropy (ΔSmix) and Gibbs energy of mixing (ΔGmix) for the 750-sample dataset. Both random forest (RF) and extreme gradient boosting (XGBoost) demonstrated superior predictive accuracy (test accuracy: 0.94927; receiver-operating characteristic curve - area under the curve (ROC-AUC): 0.99446-0.99627; 10-fold cross-validation: 0.83315-0.85184), consistently identifying ΔHmix as the most critical feature influencing phase stability. X-ray diffraction confirmed dual-phase (FCC + BCC) structures in all samples, with BCC phase volume fraction increasing from 77.25% (base high entropy alloy (HEA)) to 85.198% (2 wt.% GNP), indicating GNP-induced BCC stabilization. Peak shifts and broadening suggested lattice strain from carbon dissolution and thermal mismatch. Energy-dispersive x-ray spectroscopy (EDS) mapping revealed uniform GNP dispersion and aluminium segregation, aiding BCC stability. Zener pinning analysis demonstrated increased pinning pressure and grain size reduction from 55 μm to 50 μm. GNP reinforcement enhances phase control and microstructural stability in HEA composites, making them promising candidates for structural applications.