Machine learning analysis and discovery of novel blue-cyan-emitting Na18Ca13Mg5(PO4)18:Eu2+ phosphor

Currently, data-driven methodologies for discovering advanced luminescent materials are receiving considerable focus, driven by the potential to expedite material development timelines and provide multi-faceted insights into material properties. This study implements machine learning to develop a predictive model for emission wavelengths in Eu²⁺-activated A^IB^IIPO₄-type phosphors, thereby advancing rational design strategies. Furthermore, the model's performance is rigorously evaluated by synthesizing a novel Na₁₈Ca₁₃Mg₅(PO₄)₁₈:Eu²⁺ phosphor and comparing its emitting wavelength with the value predicted by the model, showing satisfactory agreement. The compound Na₁₈Ca₁₃Mg₅(PO₄)₁₈ exhibits a glaserite-type crystal structure and belongs to the R-3m space group of the Trigonal crystal system. The Eu²⁺-activated Na₁₈Ca₁₃Mg₅(PO₄)₁₈ phosphor demonstrates intense broadband absorption spanning 250–400 nm in the near-ultraviolet region, emitting bright blue-cyan luminescence under 365 nm excitation. The high Ra (93.3) and low correlated color temperature (CCT: 5096 K) values of the as-fabricated w-LED lamp indicate that Na₁₈Ca₁₃Mg₅(PO₄)₁₈:Eu²⁺ can act as a promising blue-cyan-emitting phosphor for full-spectrum healthy lighting applications.