Identification of polymer materials in electronic packages including counterfeit prevention

Epoxy-based underfill materials are widely used in microelectronic packaging to reduce coefficient of thermal expansion (CTE) mismatch between the organic substrate and the silicon chip and thermal stresses on the solder joints. The lack of information about the underfill material type together with its material properties can be a real hindrance to researchers when they deal with simulations involved with various underfills. Therefore, getting the correct material properties of underfill materials can enhance the reliability of simulation results to achieve the optimal solution, which significantly reduces experiment costs and time. A novel identification method for cured underfill materials from assembled packages is presented in this paper. Because the underfill materials are tiny size, insoluble in organic solvent and hard to harvest issues, the Fourier-transform infrared spectroscopy microscope Attenuated total reflectance (FT-IR Microscope ATR) have been proposed as a proper method to detect the underfill materials from assembled package. The fingerprint region of each material spectrum is chosen to apply the chemometrics to build the discriminant models. The soft independent modeling of class analogy (SIMCA) method is used to create a classification model that exhibits a high discrimination power ratio. By increasing the number of training sets and the confidence limit, the SIMCA model showed almost 100% accuracy on identification of UF1230, EP1641, SMT88U, and SMC-375TGSF5, which indicates its superior ability to discriminate underfill material with low risk of misclassification. In this study, the method of collecting dispensed or cured underfill material data from assembled packages is also presented.