Your interpretation of this EDS analysis

[EngineerDave]

A recent EDS analysis of a dark black contaminant powder a customer brought in, indicated approximately 45% oxygen by concentration and approximately 40% Fe by concentration. The remainder was mostly carbon at 12%. This powder was found dispersed in the equipment in question.

If you were to call it an iron oxide powder (it was also black, so possibly black iron oxide) however wouldn't the ratios be a bit different? By mass FeO wouldn't one expect a higher concentration of Fe due to the higher atomic weight? Black iron oxide is Fe3O4 but also (3 X 55.847 for iron) is much greater than 4 x 16 for oxygen.

It seems like a simple chemistry question. Sorry if it is trivial, it's only my 3rd week back in the world of metallurgy.

Any advice would be greatly appreciated.

 

[Metalguy]

1. Is it magnetic? Fe3O4 is.

2. Lots of C. Cast iron metal? Or is the C from little tiny bodies--MIC? Any S compounds?

What were the materials involved and the environment? These are almost mandatory but frequently not provided.

 

[acr2003]

Hi EngineerDave,
I totally agree with Metalguy questions. If you could provide the answers it would be easier to analize.
I have a question though. Is this powder material a subproduct of a process? I am asking that as it resembles a bit that you might have metal dusting, which is process that occurs in iron alloys in oxidizing environments and the corrosion product formed is a dark powder containing O, Fe and C.
Please, provide the type of atmosphere and the type of alloy the equipment was made and also, if possible, the temperature in which this equipment is submmited to for a further disscussion.
acr2003

 

[Christian62]

Good morning,
EDS is a nice tool for elemental analysis but it can't say what is bonded to what! The Oxygen could be almost entirely bonded to the C (organic contaminant) and that would solve your question. You need at least an XPS analysis and if you have money a SIMS. These will really give you a clear view! There is more tools for organic analysis like I.R. but your sample may not be suitable. Hope it helps

 

[kclim]

I'm with Christian, I think you'll need to perform different analyses, although I was thinking more on the lines of XRD - which would help identify mineral compounds. Raman spectroscopy may be useful too.

As far as black powders go, many finely divided materials have a black powdery appearance, even copper metal if I remember correctly - so I wouldn't put too much stock on that.

 

[hacksaw]

iron oxide cange range from ferrous to ferroso-ferric to ferric. commonly a mix

 

[stanweld]

I also agree with Christian. Additional analytical methods are required. EDS analysis for both O and C can be quite misleading.

 

[TEV]

Several good comments by others here. XPS or XRD will help you determine the chemical (not elemental) identity of your powder. But knowledge of the source and examination of the technical literature may make that unnecessary.

Another comment. Was your EDS done in the SEM? I have used graphite SEM stubs for imaging powder. The very low secondary electrom emission of graphite results in excellect image contrast between the powder and the substrate. Any possibility that this was done in your SEM?

Stanweld's comment about accuracy of EDS for analysis of C and O is also pertinent. These elements are near the minimum atomic number range for EDS. Their fluorescent x-rays are "soft" and their x-ray emission is relatively low.

 

[EngineerDave]

Those are some great ideas. I performed the analysis with the EDS unit of the SEM. As one stated, it did provide good contrast between the powder and the stage, or as my co-worker stated "That looks cool!"

I will keep these ideas in mind when we get a similar sample in the future. As the nature of a test lab goes, you only spend a little time with each item that comes in the door. The plus side is that you get to see so many different products (and failures unfortunately) as well.