Deformation mechanism and accurate prediction method of aspheric glass molding process

Abstract

With the accelerating iteration speed of smartphones and the continuous development of emerging AR/VR wearable device technologies, the market demand for high-precision aspheric glass lens has surged. Glass molding is considered as the most efficient method for preparing aspheric glass lens, but the deformation mechanism of glass at high temperatures is extremely complex and difficult to accurately predict. If the molding process parameters are set unreasonably, it will cause internal stress fluctuation, lead to form error and fragmentation, reduce the yield of the lens, and even cause mold damage. In this study, the characterization of glass viscoelastic parameters and the friction mechanism at the glass-mold interface were analyzed, and a material parameter inversion optimization method to achieve accurate prediction and process optimization of aspheric glass lens molding was proposed. The uniaxial compression creep tests (CCT) were carried out and a viscoelastic fitting algorithm was proposed to obtain the viscoelastic parameters of the glass. Uniaxial compression creep simulation based on these parameters verified the accuracy of the fitting algorithm. The glass-mold interface friction coefficient tests were carried out, and the glass-mold interface friction mechanism was proposed. The aspheric glass molding simulations and experiments were conducted and the reasons for the significant deviation in the deformation curve were analyzed. A material parameter inversion optimization strategy based on aspheric glass molding experimental data was proposed, and the simulation curves of the optimized model matched the experimental data significantly better, with the sum of squared errors (SSE) lower than 0.2. The aspheric glass molding experiments were conducted using the optimized process parameters, and the form errors of the aspheric glass lens were within the design specifications. The result indicated that the optimization method can significantly improve the accuracy of molding prediction and product yield.