Thoughts on the analytical equipment, setup – and personnel.

The majority of surface analysis I use is for troubleshooting, but it is also a critical tool in new product development. We use perhaps two dozen in research and development over the course of a few weeks. The range is quite significant, from x-ray fluorescence (XRF) to understand a metal-plated thickness to much more sophisticated methods, such as Auger for elemental identification.

Knowing the appropriate method to use becomes the greatest challenge. How many times have you submitted a sample for scanning electron microscopy (SEM) and electron dispersion spectroscopy (EDS) that resulted in not finding what you hypothesized or even giving a direction forward?

This month, we offer insight into understanding how the analysis method can be adjusted to gather more valuable information. To discuss all techniques would be beyond our scope, so I will highlight a few and hope to offer some clarity.

XRF is a tool used multiple times a day in circuit board fabrication houses. Each first article is measured by XRF to confirm proper metal-plated thickness. XRFs need to be calibrated daily. This is important to measurement accuracy. Some may not realize that XRF units need application programs to be set up for each substrate being plated. You may think that if you are measuring ENIG on a printed circuit board or a ceramic substrate, you can use the same program. That is not the case. During program setup, the substrate is used; largely different substrates will require different application programs. Using one application for all substrates will result in false thickness readings.

Phosphorus analysis in electroless nickel is difficult for many. I have found that digestion is the most accurate method for such analysis, although it takes a meticulous analyst, as there are many dilutions. Some use XRF or EDS for identification, all methods requiring a standard. XRF suppliers have specialized hardware and software for measuring phosphorus in the EN deposit; not all XRF units are the right equipment for this analysis. Some will give you a number based on the application setup, but this is not accurate. For EDS analysis, I suggest cross-sectioning the deposit and measuring through the section. It is more accurate than a top-down analysis.
Understand the capabilities and limitations of each surface analysis method. This becomes more detailed for more sophisticated techniques such as Auger, XPS and TOF-SIMS. Such methods may not be the best tool for troubleshooting defect panels, especially if the parts have been through an entire manufacturing process or worse, a PCB assembly that has traveled through manufacturing and assembly. You will find every piece of dirt or air contaminant that the board has encountered in its life.

I was recently asked to perform XPS analysis on a product that had been through assembly, product build and functional test to identify the elemental characteristics of the surface treatment. I cautioned the requestor about the instrument sensitivity, noting that using a tool after this much handling can give overwhelming data. Although hesitant, I agreed to move forward, anticipating what could be found. After multiple separate XPS analyses, for which I was distanced from the analyst, the data resulted in more questions than answers. A final set was run with abbreviated assembly and no functional testing. I had to lay out a significant amount of molecular information to a “new to this project” test facility. The final analysis resulted in conflicting information to the previous three, but made much more sense according to the chemical makeup of the coating. There were still pieces of the report not fully understood, but overall, it was a tedious, time-consuming road to understanding pieces of my puzzle were as hypothesized. The moral is that surface analysis on this level gets expensive very quickly, and consulting the experts first and laying all details out upfront is the key to success.

A beautiful display of this was presented at SMTA International this October. The experiment was a combination of mixed flowing gas to create creep corrosion and TOF-SIMS to identify the corrosion product. Some may think the creep corrosion product had been established years ago by Veale1 and confirmed later by Schueller2 and others. The presented material gave further insight into the materials that contribute to the initiation of the reaction. It really was an impressive body of work that will be valuable for the entire industry – well thought-out and executed. Money and resources well spent.

References

1. R. Veale, “Reliability of PCB Alternate Surface Finishes in Harsh Industrial Environments,” SMTAI, October 2005.
2. Randy Schueller, Ph.D., “Creep Corrosion on Lead-Free Printed Circuit Boards in High Sulfur Environments,” SMTAI, October 2007.

Lenora Toscano is final finish product manager at MacDermid (macdermid.com); This email address is being protected from spambots. You need JavaScript enabled to view it.. Her column appears quarterly.