Experimental evaluation of anodic bonding process using Taguchi method for maximum interfacial fracture toughness

Abstract

Anodic bonding quality has been quantitatively evaluated in terms of interfacial fracture toughness. In the theoretical analysis, the interfacial fracture toughness at the interface of an anodically bonded two dissimilar material layers has been analyzed and related with energy release rate. The energy release rate at an anodically bonded silicon-to-glass interface has been investigated for 81 different bonding conditions: three conditions for each of the four parameters, including bonding load, bonding temperature, anodic potential and bonding time. Taguchi method has been used to reduce the number of experiments required for the bonding quality evaluation, thus resulting in 9 experiment cases out of 81 possible cases. Interfacial fracture toughness has been measured from the specially designed blade test specimen, where an 80 /spl mu/m-thick aluminum blade is pre-inserted and bonded between the 535 /spl mu/m-thick silicon plate and the 517 /spl mu/m-thick Pyrex #7740 glass. The process condition, resulting in the maximum energy release rate, has been found. The influence of bonding process conditions on the interfacial fracture toughness has been quantified and discussed. It is found that the bonding temperature is the most dominant factor influencing the anodic bonding strength. Other process parameters, such as bonding load, anodic potential and bonding time, contribute weakly to the bonding strength, although they influence the speed and area of anodic bonding.