Microsection vs. XRF Method for Plating Specification

A. Microsection is the referee method in all cases of plating thicknesses within the resolution of the light microscope, very thin layers, say 0.00010" or so, may need electron microscopy. You don't have to argue the calculation or theory, when you have a photograph in front of you. I assume that you are talking about plating thickness when you say "specification". Minimum plating thickness is easier to argue, since you tend to overestimate the thickness if the section is not perpendicular to the plating.

A. I agree that microsection is the referee method, but I must emphasize that the technique is important. In cases of dispute, it would be normal to have the sections prepared by a disinterested third party.

A. Larry,

The misreading from the XFR measurement could be a calibration problem with XRF machine.

1) Is the Base material used for calibration the same as plated sample?

2) What reading accuracy are you controlling at ± 10 to 20% between Cross section to XRF?

I come from a PCB fabrication background and in my experience a well prepared microsection gives an almost gospel account of the specimen problems and potential problems that even a well calibrated XRF may never detect. Having said that, poorly prepared microsection specimens can also give false information, e.g., signs of delamination, cracking and smearing can also be signs of poorly prepared specimens as well as potential real problems. Try mico-etching the microsection specimen; this will show the different plating processes under the microscope and, using a well calibrated VMS measurement system connected to the scope, measurement of all layers is possible and thus accurate readings are possible.

A. I recommend FISCHERSCOPE using the X-Ray Fluorescence test Method.

A. Hi Zaki. There are dozens of measurement techniques other than XRF, ranging from calipers, to magnetic "banana" gauges, to eddy current, to coulometric types which measure the time to electrochemically dissolve the coating. Thickness measurement machines can be of destructive or non-destructive type; they can be for use on magnetic or non magnetic substrates, and conductive or non-conductive substrates; the coatings can be magnetic or non-magnetic, and conductive or non-conductive. Very thick coatings can be measured with a micrometer whereas thin coatings cannot. Some methods work on pinhead size areas or smaller, some work only on relatively large flat surfaces; and of course the cost can range from the better part of a hundred thousand dollars down to promotional give-away items.

If you can tell us the particulars of the proposed application, people can probably provide two or three reasonable alternatives for you to consider. But a generic answer to a forum posting will not be sufficient to turn me into a thickness measurement expert, able to select the most appropriate method for every possible case, nor you either. Good luck.

Regards,

A. If the x-ray you're using is a older ROI based machine, there might be some problems but if it is a FP based machine the application is simple even that thin. Depending on your x-ray, there will be a limit to the thicknesses you can measure. If the Ni reaches about 500 - 600 µin, a simultaneous dual measurement might not be possible.

A. With your coatings being thin, you should also check your XRF device to see if it has mechanical filtering available as well as numerical filtering. Since you are trying to measure two elements with overlapping spectra you need to effectively separate the Ni from the Cu. The mechanical filter will minimize the effects of the underlying Cu. Then the numerical filter, through peak deconvolution will add back in the counts hidden in the overlapping areas to insure an accurate measurement. I would be wary of systems that are numerical filtering only for this application.