SEM/EDS analysis in determining types of materials/coatings

[EngineerDave]

Using an SEM/EDS in identifying materials seems to be a bit of a stretch at times.

Since the accuracy of the EDS output is limited, how do you recommend to use this? Is it wiser to tell users that we can't positively identify a material with it, instead we can only give them a clue to what it can and can't be.

It is one thing to have a known and then find that your unknown sample is similar to the known. But when it is a true unknown (example surface contaminant, or unknown coating) what do you think?

 

[TEV]

Quantitative analysis by EDS requires standards and precise, reproducable conditions (take-off angles, distances, beam conditions, etc.). This can be difficult to achieve with an unknown such as a contaminant. I have been successful in confirming that a given sample is of a given alloy, however. The only time I got fooled was in mistaking stainless 303 for 316 because of the Mo/S peak overlaps! Should have caught it!

The value of EDS is the ability to do qualitative analysis of various samples to determine the possible origin of contaminants, corrosion products, etc.

 

[Christian62]

Hmm,
In all the contamination analysis that I did performed and contracted to get done I always stay very skeptical of any ONE source analysis. You have to get to the contaminant by different routes and they should merge, then you have an idea of what is your contaminant.
A) A max of info on the origin of the sample and what has been done with it (gathered dust, corroded under X/Y condition, etc)
B) Optical observations of the area of interest (with U.V. and different lighting if possible)
C)then EDS for fast first idea
D)Now you have to decide: do I what to I.D. the organics, the inorganics or both..

I could go on but you get the picture.

Hope it helps.

 

[acr2003]

EDS usually is a general overview of whatt you get in a material and for what kind of elements you have, though some EDS don't have proper windows for elents such as C or O, for example. I suggest you to do some semi-quantitative X-ray diffraction analysis, which will tell you what kind of phases you really have inside. EDS and x-ray diffraction are usually very good complementary techniques to determine what you really have in an alloy. If analyzing coatings, it will help you a lot. But you must care for the sample preparation and if possible, work with a good x-ray person, who will have the feeling for the analysis. The Rietveld method will help you with the quantitative part. If you need help for this kind of analysis, let me know.
acr2003