calibration data

[Sophie-DR]

Dear Charlotte, this might be a very easy question... I am wondering what type of data is the calibration file suppose to contain... Is it for example for Li:

Calibration of Li = concentration of Li in NIST612 / concentration of Ca43 in NIST612

How do I choose which repliacte of NIST612 to use to create the calibration file?

Thank you for your help, Sophie

[charlottesirot]

Dear Sophie,

You have to distinguish the standard files, that are all the files representing analysis of the standards (NIST or others), from the calibration file (only one file) that contained "[X]/[internal standard]" (X being a chemical element, so exactly what you mentioned) for all the elements that are included in the standard you use and that you need in your analysis.

So you will first reduce all the standard analysis and then the calibration file will give you the equivalence cps/sec to the ratio of concentration (don't worry, you will do everything with the elementR software, it will guide you on the different steps of this procedure).

For more details you can look into the documentation that is included within the package (you can find it on gitHub too).

Does it make sense to you ? Don't hesitate to tell if it is still unclear

[Sophie-DR]

Dear Charlotte,

thank you for your answer and clarifications. Your explanations totally make sense to me. You are right I was confusing the standards measured and the calibration file based on the theoretical concentration of the NIST612.

How do you make the calibration file? Where do you get the NIST612 calibration values for all the elements you have in your analysis? Do you take into account the uncertainty of the values of the concentration of each element in the NIST612?

For example, I found the values for Li : = 42/85002 = 0.000494106 which is not what you have in your calibration file (0.000488429) How come? I lost the source of the values I have from the NIST612... What is your source?

I guess the value is the same for several isotopes of the same elements.

I used several standards (NIST610, NIES, FEBS), could I use the other standards as well to make the calibration file?

[charlottesirot]

Good that it gets clearer.

I use to create my calibration file based on the literature. You can use the standard that you want (NIST610, NIES, FEBS) but for the moment, you can only use one type of standard for the whole session. That means that you can have as many standard analyses as you want in the session (I usually analyze one standard every 3 samples) but they all have to be from the same standard material.

However, something that you can do, is to use the other standard analyses as reference materials, i.e. analysis that will be use to valid your experiment. For example, let's take a session of 3 analysis of NIST 612, 2 of NIST 610 and 3 of regular samples. You will consider the 3 NIST 612 as standards (it will be use to calibrate the cps/sec -> concentration conversion), then filter the data obtained through the analyses of the 2 NIST 610 as regular samples and then compare the obtained values to the literature. In that case you will have an idea of the correctness of your ICPMS analyses. I don't know if it is clear...:D ??? (Unfortunately elementR is not able to use the uncertainty of the calibration file for the moment but it would be an idea of implementation :) )

Regarding the resolution, I developed the software on a device of 1600*900. This resolution (and above) will be enough to work comfortably on elementR

Let me know if you have any other question :)

[Sophie-DR]

All clear!

You might have missed some questions above: For example, I found the values for Li : = 42/85002 = 0.000494106 which is not what you have in your calibration file (0.000488429) How come? I lost the source of the values I have from the NIST612... What is your source of the values of NIST612 you used?

I guess the value is the same for several isotopes of the same element.

[charlottesirot]

Hello Sophie,

We use Pearce et al 1997, but it's only our personal choice. With elementR, you can use the calibration data that you considers as the best for your need.

Does it answer your question?