Abstract

The identification of relevant adhesion properties is important for the development of microelectronic systems, e.g., microprocessors, light-emitting diodes, etc. This is best done on an actual device, rather than on dedicated specimens, because such properties are highly specimen specific. This paper discusses how the full-field identification method of integrated digital image correlation (IDIC) can be used to identify three mixed-mode cohesive zone parameters of a flexible, organic light-emitting diode (OLED) from a single mechanical test. A tri-axial, micro-mechanical testing setup is employed for the in situ, on-device delamination experiment, employing optical microscopy.The microscopic magnification, required for capturing the relevant kinematics, prevents the experimentally imposed loading conditions to be captured within the field of view. Virtual experiments are firstly conducted to optimize the local boundary conditions therefore required by the finite element simulation within the IDIC-framework, and which constitutes the cohesive zone model. The virtual experiments also allow for selecting an adequate load-case for the actual experiment, in order to maximize the kinematic sensitivity of IDIC with respect to the three mixed-mode cohesive zone parameters, and identify them accurately. The image residual, minimized in IDIC, together with the convergence behavior towards the associated global minimum, provide adequate measures for assessing the trustworthiness of the solution.The optimized load-case is subsequently used to trigger delamination of a 50 [μm] thick barrier stack from its substrate in a real in situ experiment on a flexible OLED. The optimized boundary conditions from the virtual experiments are used in a corresponding simulation for identifying the three mixed-mode cohesive zone parameters by IDIC. The optimization scheme within IDIC was found to be robust against deviating initial guesses, and consistently converged to a unique solution for the cohesive zone parameters.

Full Text
Paper version not known

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