ABSTRACT
Megasonic Removal of Flux Entrained in Narrow Gaps in Packaging Structures
In electronic packaging, Solder Paste Reflow process is routinely used to attach capacitors on a chip to copper pads formed in solder resist film. During cooling after reflow step, flux solidifies around the tin-based solder and in the gap (~tens of microns) between the chip capacitor and solder resist (Figure 1). A deflux step is necessary to remove the solidified flux in the gap and is commonly done using liquid formulations. The kinetics of removal is often enhanced by irradiating the formulation with ultrasonic waves. Switching from ultrasonic to megasonic frequency waves can offer improved benefits such as better mass transfer due to higher liquid streaming and reduced cavitation implosion.
The objective of the work to be presented was to optimize megasonic removal of flux trapped in the gap between C0204 chip capacitors [0.6 (L) x 0.3 (W) x ~0.35 (T) mm] mounted on chips and packaging substrates. Flux removal tests were conducted in a slightly alkaline (pH ~ 8.0 to 9) semi-aqueous formulation that comprised alkyleneglycol alkyl ethers, water (<10%) and organic inhibitors in the presence of a megasonic (925 kHz) field. Specifically, the tests were conducted in a flow through MegBowl® (~500 ml) expressly modified to expose flat pieces (27 (L) x 17 (W) mm) of a chip package to megasonic field at different incident angles (0 to 90 deg.). Key variables investigated were megasonic power, pulsing of the megasonic field, distance between the sample and megasonic source, angle between the sample and megasonic transducer, and temperature of the cleaning formulation. To determine the efficacy of flux removal, the chips were mechanically removed and the region under the chips was examined using an optical microscope at a magnification of 50X to determine the extent of remaining flux, if any. Representative photographs showing good and poor flux removal under the chip capacitors are presented in Figure 2.
To run the cleaning tests on reflowed solder structures, a simple DOE matrix was constructed using power density, angle, time and temperature as key variables. Optimal cleaning was achieved at an angle of zero (sample facing the transducer), a cleaning time of five minutes at 500C, and a power density of 2 W/cm2. If the location of the test sample was in the near-field region of the megasonic transducer, cleaning was not significantly influenced by the distance between the sample and the transducer. A mechanism for the removal of flux has been developed from kinetic data obtained on flux coated flat substrates.
BIOGRAPHY
Dr. Srini Raghavan is a Professor Emeritus in MSE at the University of Arizona, an Adjunct Professor in the School of Engineering at Arizona State University, and the Director of New Process Development at ProSys Megasonics. He obtained his Ph.D. from the University of California at Berkeley and has published extensively in wet processing of materials relevant to semiconductor device fabrication.