High-entropy rare earth aluminates are promising candidates for thermal/environmental barrier coatings (T/EBCs), while the immense compositional space presents significant challenges for traditional experimental discovery. To addressed this issue, artificial neural network (ANN), support vector machine (SVM), and random forest classification (RFC) are employed as three machine learning models to predict the single-phase formation ability of (RE^Ⅰ_1/4RE^Ⅱ_1/4RE^Ⅲ_1/4RE^Ⅳ_1/4)₄Al₂O₉ materials. Both ANN and RFC models achieve the optimal validation accuracy, demonstrating their outstanding ability to capture complex patterns from the dataset. SHapley Additive exPlanations (SHAP) analysis is utilized to interpret the contribution of feature descriptors, revealing the significant impact of factors like average ionic radius ( ) on phase stability. According to the prediction results of machine learning, three representative ceramic samples are selected and single-phase monoclinic crystal structures and uniform elemental distribution are confirmed by X-ray diffraction and scanning electron microscope. The synthesized ceramics exhibit quasi-ductile behavior with enhanced damage tolerance combined with lower thermal conductivity, thus making them promising candidates for next-generation T/EBCs.

Passive radiative cooling (PRC) is a promising way to alleviate the global energy crisis by reflecting sunlight and dissipating heat through the atmospheric transparent window (ATW). Despite possessing a wide bandgap and complex phonon modes, the PRC performance of Mg₂Al₄Si₅O₁₈ is limited by phonon-polariton resonance. Herein, phonon engineering is integrated with bandgap engineering to design and synthesize a series of Mg₂Al₄Si₅O₁₈:xY³⁺ (x = 0%, 2.5%, 5%, 7.5%, and 10%) ceramics with excellent PRC performance. Density functional theory (DFT) identifies that Y³⁺ doping effectively suppresses phonon-polariton resonance and widens the bandgap, synergistically enhancing the PRC performance. The as-prepared samples exhibit high ATW emissivity (94.39%–98.39%) and high reflectivity (89.52%–94.77%) in the 0.4–2.5 μm range. Furthermore, the “cooling glass” coating successfully achieves a maximum temperature reduction of 16.5 °C and an average net radiative cooling power of 113.1 W·m⁻². Y³⁺ doping enhances ATW emissivity by inducing lattice distortion, which reduces symmetry and alters the dipole moment while boosting reflectivity in the visible and near-infrared (vis-NIR) regions by preserving the wide bandgap through the introduction of optically inert elements. This work synergistically integrates the advantages of high performance, low cost, and environmental friendliness, offering a highly promising ceramic material solution for large-scale radiative cooling applications.