Issue link: https://iconnect007.uberflip.com/i/1288481
80 PCB007 MAGAZINE I SEPTEMBER 2020 When comparing the two tools, it seems clear that compromises become necessary. The investigation by a light microscope allows a comparably easy and fast examination of large areas with a tool that is available almost every- where. In contrast, the efforts to prepare SEM images of the cross-sections are higher. The tool is more expensive, the sample preparation (as well as the examination itself), consumes more time; on the other hand, the resolution is higher. Imaging with SEM allows us to detect smaller corrosion events, which might not be detectable with a light microscope evaluation. For both methods, there is the limitation that only a very small detail of the whole panel can be examined. To create a representative result that reflects the performance of the whole pan- el, the selection of investigation areas of the full panel is important. These areas should be selected from different locations on the pan- el, and the number of pads, or through-holes, which are investigated should be enough to al- low a representative picture of the panel. In the current version of the IPC-4552A, a description for the evaluation of ENIG corro- sion is contained, which leaves room for inter- pretation at the current stage. The target of the ongoing revision of the specification is to more clearly define the required number and loca- tions to be investigated and the rating to judge panels as acceptable or rejectable. The investi- gation is supposed to be done by a light micro- scope with a magnification of 1000x. For the plating process development and im- provement, it was found that a more systemat- ic comprehensive method is necessary to allow the judgment of gradations between "pass" and "fail"—hence a minimum of two locations of a panel where at least four PTHs are inves- tigated is selected. As the PTHs are more criti- cal than the pads in regard to the plating pro- cess conditions, such as agitation and solution flow, the focus in the corrosion evaluation is put on the study of the through-holes. Results show that the risk for excessive nickel corro- sion is higher in PTHs and, in particular, at the through-hole entrance where the solution flow is highest. To allow the most objective way of rating, which excludes the impact of the operator, a table has been implemented in the process de- velopment, where the corrosion is rated in re- gard to the depth in the nickel layer and the number of events. Different classes of corrosion depths are de- fined, which are specified as 0–20%, 21–40%, 41–100%, >100%, and surface corrosion (Ta- ble 3). Using this method allows a quantitative judgment of corrosion, which enables the en- gineer to compare different process conditions with statistical tools and relevance. Notice that 0–20% of corrosion includes all corrosion events, which penetrate the nick- el layer thickness up to 20%. Such corrosion events are usually rated as less critical because the attack on the nickel is low, and the im- pact on soldering and IMC formation is negli- gible. Further, 21–40% includes all those corro- sion marks, which penetrate the nickel by 21– 40% of the thickness. Such kind of corrosion is also expected to be less critical as long as the number of events is not too high because there is still enough volume left of the nickel layer thickness. More critical are those corro- sion events, where the penetration depths are between 41 and 100%, as this indicates that the corrosion event might penetrate the nick- el down to the copper layer (such as in Figure 1 at the bottom left image) so that the barrier function of the nickel layer is affected and cop- Table 3: Exemplary table to rate corrosion in process development.