In Situ Observation of Liquid Solder Alloys and Solid Substrate Reactions Using High-Voltage Transmission Electron Microscopy

Abstract

The complex reaction between liquid solder alloys and solid substrates has been studied ex-situ in a few studies, utilizing creative setups to “freeze” the reactions at different stages during the reflow soldering process. However, full understanding of the dynamics of the process is difficult due to the lack of direct observation at micro- and nano-meter resolutions. In this study, high voltage transmission electron microscopy (HV-TEM) is employed to observe the morphological changes that occur in Cu6Sn5 between a Sn-3.0 wt%Ag-0.5 wt%Cu (SAC305) solder alloy and a Cu substrate in situ at temperatures above the solidus of the alloy. This enables the continuous surveillance of rapid grain boundary movements of Cu6Sn5 during soldering and increases the fundamental understanding of reaction mechanisms in solder solid/liquid interfaces.

Laser-Assisted Micro-Solder Bumping for Copper and Nickel–Gold Pad Finish

“…A prominent difference was seen on the Cu30/100-SAC interface (compared to the same sample after bonding and temperature cycling), where more copper was consumed to enhance the Cu 3 Sn layer. Meanwhile, the Cu diffusion to Cu-Sn-Pd creates vacancies leading to Kirkendall voids, as also observed by [ 46 ]. The SEM micrographs show an uneven Cu 3 Sn-layer thickness across the interface; the maximum thickness was about 1.5 µm.…”

In Situ Observation of the Ga- and Cu-Based Substrate Reaction by Synchrotron Microradiography

“…Over 420 s of heating at 100 °C from 637 to 1057 s, the morphology of the substrate surface has changed very little due to the consumption of Cu6Ni and the formation of IMCs. The change in the substrate surface morphology has also been observed for the solid–liquid reaction between SAC305 and Cu substrates . However, the slow change of the substrate proves that at 100 °C, the reaction rate between Ga and Cu6Ni is much lower than that at 150 °C.…”

“…The IMC layer uniformly forms between the interface of Ga and the Cu6Ni substrate, and as the temperature increases, the IMC formation speed increases significantly, and the growth is in the direction of the Cu6Ni substrate which is similar to the observation for SAC/Cu joints observed under in situ TEM. 26 It can be seen from Supplementary Video 1 that the IMC growth rate at 104 °C is much slower than that at 150 °C. The large bright spots are bubbles that formed when the Ga melted, and some bubbles along the substrate are flux bubbles.…”

“…The change in the substrate surface morphology has also been observed for the solid−liquid reaction between SAC305 and Cu substrates. 26 However, the slow change of the substrate proves that at 100 °C, the reaction rate between Ga and Cu6Ni is much lower than that at 150 °C. This indicates that the IMC growth rate is heavily dependent on the reaction temperature during heating.…”