NSF Award
0539355
This Small Business Innovation Research (SBIR) Phase I project will extend the capabilities of magnetic current imaging to detect open failures in semiconductor packages with an accuracy of 30 microns. The largest, most difficult problem encountered in packages is electrical opens due to increasing shrinking and complex technology in leading edge designs. The only available techniques are time domain reflectometry with practical resolution of 1-2 mm and time consuming layer-by-layer deprocessing. There is a critical need for a faster, non-destructive and more reliable technique capable of locating opens at a level commensurate with todays package level wiring, 30 micron pitch and able to extend to future wiring pitches approaching 10 micron. It is proposed to use magnetic current imaging with a SQUID sensor to solve this critical need. Two approaches will be attempted: one is to increase the frequency detection limit of the equipment to detect high frequency signals at the defect location. The second is to use the SQUID to detect radio frequency signal emitted by the defect at higher frequency signals. It is expected that one or both of these approaches to be able to detect opens with a resolution of about 30 micron.<br/><br/>Commercially, the semiconductor industry has a critical need for localization of buried open defects in packages with resolution below 30 microns. The proposed technology is targeting a 30 micron resolution to fill a known gap in fault isolation technology for packaging. For the semiconductor companies it will enable the packaging manufacturing sites to isolate open defects and improve their manufacturing processes to minimize or eliminate these defects quickly so that they can get high quality reliable products to market faster. For the nation, it means faster introduction of advanced electronics that will have a broad impact across all industries and ultimately improve quality of life and labor productivity.
