Abstract

In the market of the mobile cover glass, the development of chemically strengthened glass is focused on the drop resistance improvement. The cover glass which protects the display panel is a typical brittle material that the micro cracks tends to be occurred underneath a glass surface by physical impacts. The micro cracks tends to be propagated by tensile stress and it is known as a general procedures on glass breakage. In purpose of the cover glass strength improvement, compressive stress is applied to the glass surface using chemical strengthening method by ion exchange. However, since the central area of the glass is relatively subjected to tensile stress, the glass is instantly broken when the propagated cracks are reached on the tensile stress exerted area. The glass specifications have been managed for maintaining the strength using parameters of compressive stress (CS), depth of compression (DOC), and central tension (CT). However, there was no method to judge the most effective factor for impact drop resistance. In order to solve this problem, we developed a machine learning program for cover glass that can predict the breakage height based on stress and mapped evaluation data based on the measured stress profile data and the actual drop breakage simulated evaluation. Especially, eXplainable AI (XAI) method is used to identify the effective factors on drop breakage height. And the test and measurement data were applied to various chemically strengthened glass for analysis.In this study, it was possible to predict the drop breakage height only by measuring the stress profile of chemically strengthened glass. The feature value of chemically strengthened that has the most effect on the drop breakage resistance on rough surfaces was identified to be stress value at 30um in the case of alkali aluminosilicate glass. And it was proved that the drop breakage height was improved by 25% only by increasing the compressive stress at 30um depth.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call