I have a solid sample that contains Co, Mn, Al, C, and O. The sample mass is ~200 mg. I need to determine what the composition of this sample is. I've looked at atomic adsorption spectroscopy, but that method is unable to detect C or O. I've heard mention that inductively coupled plasma mass spectrometry (ICP-MS) might work, but I am looking for additional potential solutions. I have access to a large number of instruments (university instrument sharing agreement). It is fine if the sample is destroyed as part of the analysis.

  • $\begingroup$ How much sample do you have, what accuracy do you need, which equipment do you have, what is the sample made of (is it organic or inorganic sample? If you have enough sample (100 mg) elemental analysis service would work. C, H, N, O, and metals are quantified using classical techniques. If you have small amounts of the sample you have to use several different $\endgroup$ – sixtytrees Jul 20 '16 at 21:55
  • 1
    $\begingroup$ Are those all the elements present? X-ray fluorescense might be an idea, although it can have trouble with light elements. TGA could help you to quantify carbon, or carbonate.Is your sample very very likely to be completely oxidized? $\endgroup$ – Phanz Jun 8 '17 at 7:23

It very much depends on element concentrations and the accuracy required for the total assay, but, I would do the metals by ICP-AES or AA and C by a classical combustion technique. ICP-MS would be unnecessarily expensive and since you are looking at compositional levels rather than low trace levels, it just wouldn't buy you anything.

If you know oxygen is the only other element present at compositional (i.e. percent) levels then don't bother with the finickier oxygen analysis and just determine it by difference. If carbon is present at at least a few percent then you only need a few mg, leaving enough to do multiple digestions for the metals analysis if desired.

If you want to do direct oxygen analysis, it can be done by classical pyrolysis techniques, though this presumably high-metals matrix would require a special technique (Mertz) and this analysis could have relative uncertainties of greater than 10%.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.