I would also add one very important definition - selectivity, which is different for different analytical methods. General formula for analytical signal is " S = ka Ca + k1 C1 + k2 C2 + k3 C3 + k4 C4 + …". where S is a total analytical signal, k is slope, C is concentration, a is analyte, 1,2,3,4 are interfering compounds. In non-selective methods, we measure a total signal, which includes signals of analyte and interfering compounds. In selective methods, ka is much higher than k1, k2, etc. Non-selective methods can only provide proper accuracy when Ca is much higher than C1, C2, etc. In chromatographic methods, we separate the total signal to signals of individual compounds. After students understand this, they will better understand the difference between different analytical methods.

From my point of view, the "Analytical Signal" = f (Concentration)" concept must be well understood before studying different methods of quantitative chemical analysis. In qualitative analysis, analytical signals and the way of their processing are different. Student should understand that analytical signals are different for different methods. In this case, they will better understand the general pattern of an analysis: 1) measure a suitable signal and 2) convert it to an analyte concentration or presence.

A vacuum UV detector for GC is a good example. For many years it was forgotten. In last few years it started to be very interesting and attractive concept.

Same FID for HPLC.

Check also term Ice Chromatography.

regards,

GB

I have found that there is a misconception regarding error of measurement and uncertainty of measurement. Even though GUM states they are not the same and shows a diagram which indicates that the error of measurement is expanded by the uncertainty, analysts are replacing the error of measurement with the calculated estimated measure of uncertainty.

I am agree whit Bulab Kenessov, but would add, that the first thing we must understand is the difference between, "analitycal signal" and "Analitycal response", concepts like a "Anlyte", "Linearity", "Presicion" "Accuracy", etc, must be well understood.

I would also add one very important definition - selectivity, which is different for different analytical methods. General formula for analytical signal is " S = ka Ca + k1 C1 + k2 C2 + k3 C3 + k4 C4 + …". where S is a total analytical signal, k is slope, C is concentration, a is analyte, 1,2,3,4 are interfering compounds. In non-selective methods, we measure a total signal, which includes signals of analyte and interfering compounds. In selective methods, ka is much higher than k1, k2, etc. Non-selective methods can only provide proper accuracy when Ca is much higher than C1, C2, etc. In chromatographic methods, we separate the total signal to signals of individual compounds. After students understand this, they will better understand the difference between different analytical methods.

Please look at the following below links which may help you in your analysis:

http://pqri.org/wp-content/uploads/2015/11/Norwood.Integration_Toxicology.pdf

https://pubs.acs.org/doi/abs/10.1021/ed500679k

https://files.eric.ed.gov/fulltext/ED558533.pdf

Thanks

In analytical measurement most of the time we think that uncertainty is the combination of so many things but if you go through various uncertainty measurement guidelines than you will found that uncertainty is not beyond the calculation formula.. Which means directly related to the components which we are using during the calculation.

What I am seeing is that reports give a weight without the accuracy or precision (ie the error of measurement) but instead give the estimated measure of uncertainty and then state, when challenged, that emu and error are the same. As I said above, GUM and other guides differentiate between error and uncertainty. There is no argument that a weight is obtained, but the reliability of that weight, ie, the range, is wrong. This needs to better conveyed to scientists.

I would like to add another important dimension of analysis here, i.e. the possibility of interaction of an analyte with other components present in the sample. These interactions can sometimes drastically affect analysis results and this is a tricky business, specially when the sample matrix is complex and/or unknown. For e.g. the chemical interferences in AAS. Analysis of different analytes in wastewater etc. Prior information about the sample or its composition comes in handy besides one could adopt techniques like spiking.