Uncertainty Quantification in Prediction of the In-Plane Young's Modulus of Thin Films With Fiber Texture

Electrodeposited thin films in MEMS devices often show fiber texture resulting in transverse isotropic, effective elastic properties. It is of interest to predict these elastic properties since they play a role in device performance. In addition to predicting effective material properties of the devices, we quantify the uncertainty in our predictions of these material properties for use in downstream simulations aimed at studies of performance, lifetime, or reliability. In this paper, we estimate the numerical value of the effective in-plane Young's modulus of thin nickel polycrystalline films using numerical simulation. We also examine the variability and sensitivity of the in-plane Young's modulus due to uncertainties in microstructure geometry, crystallographic texture, and numerical values of single-crystal elastic constants. The importance of accurate characterization of the texture is shown, as is the sensitivity of the effective in-plane Young's modulus to single-crystal elastic moduli.