[eb51111]

Hi,

For Mg determination by acetylene flame AAS, with standards in the range of 0.2 - 1 ppm Mg, is there a measureable change in absolute error when measuring a sample of 0.3 and one of 0.7 - or does the error only increase after a very large increase in concentration such that atomisation would be incomplete (as I’m not an AAS expert).

Many thanks

[hypervalent_iodine]

It depends a lot on how you have prepared your samples.

[eb51111]

Just assume that the samples are all prepared with all the same (unavoidable, small) systematic error - what I mean is does the error in the ppm reading on the instrument increase more than a negligible amount when roughly doubling the concentration of sample - as it does with some other techniques, or is AAS quite versatile over large ranges?

Thanks again

[Raayou]

I greatly recommend that you must include a check sample as a QC  of known concentration within your samples to understand what accutaly is the problem. Miss  alignment of the burner or poor flame optimization of the signal could also be the cause of this.

[hypervalent_iodine]

My own experience is that it can vary with the metal you are looking at. I am unfamiliar with Mg, but I have seen for example that Mn is often more susceptible to error from the AAS that I have used than is something like Cu. There are a lot of things that are machine and lamp specific that could contribute to systematic error, so it’s hard to say how significant the problem may be for you. As mentioned by the other poster, you would probably do well to perform some sort of QC.