Issue link: https://iconnect007.uberflip.com/i/1137649
100 SMT007 MAGAZINE I JULY 2019 as the hole wall of a drilled panel. This cre- ates electrical connectivity through the plated holes. Then, the electrolytic copper plating pro- cess uses an electrical current to electrodeposit copper from a copper sulfate solution onto the copper-clad production panel. Electroplating inherently prefers to plate onto protruding fea- tures, edges, or corners; therefore, the slight- est surface protrusion on the base copper foil or a minor imperfection co-deposited during the electroless copper process will attract more copper plating and grow in size to become a visually detectable nodule after electroplating. Nodules that were observed optically under 40X magnification were cross-sectioned in an attempt to determine the process at which the nodule began to form. Given the small size of the nodules and knowing that the nodule starts with an even smaller surface imperfection that grows through electroplating, cross-sectioning down to the absolute center of the nodule was difficult to achieve and objective evidence of the nodule's origin could not be found all the time. Figure 3 shows an example of a cross- sectioned nodule. In this case, it seemed to be clear that the nodule was formed before the Ni/Pd/Au plating process, and, therefore, nod- ule formation would have likely happened dur- ing electroless or electrolytic copper plating. Using a fishbone diagram to define poten- tial root causes of the nodules, a short list of potential root causes was identified: copper foil imperfections, electroless copper co-deposited particles, and electroplated copper co-depos- ited particles. Each potential root cause was explored in-depth to prove or disprove that it was the source of—or contributed to—the for- mation of nodules. A random sample of copper foil was visu- ally inspected under 100X magnification and no significant signs of imperfections were found. Using a controlled production lot, a light mechanical scrubbing process was performed on the laminated foil through a horizontal con - veyorized machine, but the number of boards that were rejected for nodules remained close to the average from previous production lots. Con- sequently, this potential root cause was elim- inated early into the investigation. The same mechanical scrubbing process was performed on panels that had been processed through electroless copper. In this case, the results were noticeable. The density of nodule occurrences was reduced but there were still boards being rejected for single nodules located at wire- bond locations. And in some production lots, the frequency of the single nodules was exces - sive enough to still have a significant impact on yield. However, an overall improvement had been made, and this mechanical scrubbing pro- cess became a standard process. After an extended period of processing boards with the mechanical scrubbing pro- cess after electroless copper, an opportunity to test the electrolytic copper plating process arose when an annual preventative mainte- nance procedure was completed on one of the three plating tanks on the automated plating line. The procedure involves: • A complete carbon treatment of the copper sulfate based chemistry to remove organic molecules that have leached out from the dry-film photoresist • Emptying and cleaning all copper anode baskets to remove copper sludge • Replacing all of the anode bags • Dummy plating after chemical additions to form a uniform copper oxide film over the anodes Panels were plated in the new tank and a normal tank, and the number of nodules was significantly reduced on the panels plated in Figure 3: Cross-sectioned view of a nodule as an example.