UV excited optical and temperature sensing properties of Sr3Gd(PO4)3: Er3+ phosphors: Machine learning predicted colour perception

The Er3+ activated Sr3Gd(1−x)(PO4)3 phosphors were synthesized via modified citrate sol-gel combustion method. Powder X-ray diffraction was adopted for phase examination. Field emission scanning electron microscopy was employed for morphology and elemental mapping. They exhibit Er3+ emission bands related to (2H11/2,4S3/2, 4F9/2) → 4I15/2 transitions up on 376 nm UV excitation. They produce potential greenish-yellow emission of colour purity of 94.71% with Commission International de I’Eclairage (CIE) coordinates (0.289, 0.698). The Er3+ doping was optimized as 1% for prominent luminescence. The CIE coordinates were predicted to assess their effectiveness through three different machine learning algorithms such as the Random Forest regression, Extreme Gradient Boosting regression and Light Gradient Boosting Machine regression. The findings show that all the three models have considerable accuracy for prediction with overall test root square, R2 ≥ 0.96285 and provide useful information for choosing appropriate CIE coordinates for real-world applications. The quenching temperature was found higher than 300 °C showing their good thermal stability with 0.235 eV activation energy. The optimal phosphor shows a maximum absolute sensitivity of 0.00464 K− 1 at 573 K and relative sensitivity of 0.934%K− 1 at 373 K. The Sr3Gd(PO4)3: 1%Er3+phosphor possess excellent optical as well as temperature sensing properties.