Every electric-powered system whether it is a computer, phone, car, or clock possesses at least one packaged microor nano-electronic device. Traditional methods of mechanical quality assurance for packaged devices are… Click to show full abstract
Every electric-powered system whether it is a computer, phone, car, or clock possesses at least one packaged microor nano-electronic device. Traditional methods of mechanical quality assurance for packaged devices are outlined in standards such as MIL STD 883, SAE AS6171/5, among others and are primarily destructive in nature. For a device with a bond wire interconnect architecture, which account for roughly 90% of circulated devices [1-6] the device’s packaging is removed and the bond wires are pulled until failure before a statistical inference is made on whether to accept the device that will ultimately be integrated. Among other shortcomings, this practice introduces uncertainty since the device that is being integrated is different from the tested device. A non-destructive testing alternative centered around the device that will be integrated has been shown for a bond wire pull test [7]. In summary, realistic geometries of the internal components of the device are captured with 3D X-ray tomography, segmented, reverse engineered to a CAD-editable format, and brought to finite element modeling to perform the mechanical test in simulation space. The advantages of this non-destructive approach include: reusability and assurance of the tested circuit that will be integrated, simultaneous testing of multiple components under different conditions, testing bond wires with fine pitch/shared leads, significantly faster testing in long duration dynamic tests (e.g. fatigue, creep, etc.), creating extremely unique test environments that could be prohibitively expensive to recreate (e.g. extreme pressure, extreme temperature, etc.), and performing parametric studies to answer design questions shortcutting the iterative prototyping process [7]. The results of this testing alternative requires sufficient validation based in physical testing. In this study, a novel in-situ mechanical tester was developed and used to perform the bond wire mechanical test in-situ of a FIB-SEM system. The data gathered from this physical testing was used to find the physical principal strains of the bond wire through digital image correlation (DIC). These principal strain values were used as validation criteria for the simulation models developed with the non-destructive workflow.
               
Click one of the above tabs to view related content.