[Babcock_Hall]

Pagliano E et al. 2015 Analytica Chimica Acta 896:63-67. https://doi.org/10.1016/j.aca.2015.09.020
Tan A et al. 2011 J. Chromatography B 879:1954-1960. doi:10.1016/j.jchromb.2011.05.027
Quantitative Applications of Mass Spectrometry 2006 I. Lavagnini, F. Magno, R. Seraglia and P. Traldi John Wiley & Son. ISBN: 0-470-02516-6.

I am reading up on isotope dilution mass spectrometry (IDMS) for pedagogical purposes.  R is the area of the peak for the analyte over the area of the peak for the internal standard.  The standard often has 3-10 deuterium atoms.  My understanding is that R is plotted on the y-axis and mass or concentration is plotted on the x-axis.  I am aware that sometimes the calibration curve is not a straight line in practice, and I am including a couple of references for those who might be interested.  Suppose that our standard in IDMS had no contaminant that interferes with mass spectrum of the unknown.  Would the calibration curve be a straight line with a slope of one and with zero as the y-intercept?  The calibration curve for a derivative of testosterone in the third reference appears to have a slope of one.  See attached file.

[rjb]

Hi BC,

Interesting subject and not one I know a lot about!

What you are describing (i.e. area of analyte peak/IS from chromatogram) sounds more like a conventional internal standard calibration curve as opposed IDMS which (I would suspect) would plot something like isotope ratio (of H vs D) against ratio (H vs D) in the spike. Again, I know little of this subject and may have misunderstood your post, so I may well be very wrong but I would imagine isotope ratio (as opposed to chromatographic peak area) should be part of the plot.

A conventional IS curve should in theory produce a linear response throughout the analytical range and in an ideal world should have an R2 of one and intercept of zero! Of course that never happens and all the curves that I have seen are best described using a weighted approach as opposed to straight least residuals squared and never pass through zero (unless you force them to!). From what I gather IDMS might not produce a nice linear plot...

R

[Babcock_Hall]

Thanks for the reply.  My initial post was worded imperfectly.  I should have specified that the peak areas refer to peaks in a mass spectrum.  Does that make more sense?  Just to clarify one other matter, I would like to use one or more assumptions in order to simplify the topic as much as possible in a presentation.

[Babcock_Hall]

If I understand correctly, one cause of nonlinearity is that there is an impurity in the standard that appears at the mass of the analyte, or vice versa.