Estimated reading time: 1 minute

Copper Discoloration and Other Concerns with OSP

Yes, it is about process. And that includes equipment operating parameters as well as chemistry. Getting the OSP process to perform as it is intended requires attention to both. This month's "Trouble in Your Tank" delves into one of the most irritating issues with respect to OSP: discoloration (read "oxidation") on critical circuit features. Why did this occur? Is there more than one cause? Will solderability be compromised? These are just a few of the questions raised when such a situation occurs. An example of this occurrence is shown in Figure 1.

Discoloration and even oxidation of the copper may occur for one of several reasons. Some of those listed may be more or less obvious than the others (Table 1). Some of these causes can be both equipment and chemistry related.

Figure 2 shows another example of discoloration on the copper pad most likely related to oxidation of the underlying copper. Further analysis of this situation pointed to two possible causes. The first was very low thickness of the OSP coating. In this case the thickness was less than the minimum recommended by the supplier (measured OSP thickness 0.08 micron versus recommended 0.20–0.25 micron). Certainly, the low thickness is cause for concern. With such a low thickness, the OSP would not effectively protect the copper from oxidation. However, what other process variations could be the cause for the low thickness? One possibility is the micro-etch step. It has been stated in the literature that a clean active copper surface is necessary in order to allow for a more uniform deposit of the OSP film. In addition, one should avoid an overly aggressive microetch as there are concerns about insufficient thickness of the organic film in deep crevasses.

Surface roughness does have an impact on OSP film thickness and is directly connected to solderability.

Read The Full Article Here

Editor's Note: This article originally appeared in the April issue of The PCB Magazine.