Deep-Learning Enhanced SrAl₂O₄: (Eu²⁺, Dy³⁺, Nd³⁺) Mechanoluminescence Film for Distributed Perception of Mechanical Deformation and Fracture

Mechanoluminescence (ML) sensor-derived distributing measurement urgently needs to overcome the trade-off between luminous intensity and afterglow duration. In this article, a strontium aluminate (SrAl₂O₄) based ML sensing candidate is controllably synthesized by solid-solution reaction of powdered precursors of SrCO₃ and Al₂O₃ under hybrid doping of rare earth cations (Eu²⁺, Dy³⁺, Nd³⁺) at 1400 °C. Compared with traditional SrAl₂O₄: Eu²⁺, SrAl₂O₄: (Eu²⁺, Dy³⁺, Nd³⁺) (SAOEDN) has demonstrated highly enhanced luminous intensity (over two orders increase), robust ML behavior (300 cycles), and tunable afterglow performance (50 to 325 s) after synergistic regulation of trap depth (from 0.2 to 0.88 eV). After in situ compounding of SAOEDN with epoxy resin matrix, a flexible ML sensing film is created for distributed detection of engineering strain distribution. The ML effect triggered by mechanical deformation presented an approximately linear dependence between strain and luminous intensity with a higher spatiotemporal resolution. As a result, the engineering strain field is reconstructed via a deep learning-derived image-to-image mapping process after eliminating the disturbance of afterglow. Moreover, the SAOEDN based ML film is capable of accurately detecting and capturing fracture propagation of engineering materials. It is suggested promising potential for distributed non-contact detection of stress and strain fields in engineering applications.