1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How to get calibration curve?

  1. Nov 24, 2015 #1
    I've just done an ion chromatography experiment and have normalized my data for a known sample which has a known concentration of 10ppm.

    I know that the first large peak is due to F- ions and so the area of the peak is propositional to the concentration of the ion.

    Given that I know the area, how can I get the calibration curve just by this information ?
     
  2. jcsd
  3. Nov 25, 2015 #2

    Borek

    User Avatar

    Staff: Mentor

    Assume signal would be zero if there were no F- at all, that gives you two pints, enough to draw a straight line.

    That's not the best approach in general, but better than nothing.
     
  4. Nov 25, 2015 #3
    Thanks, but I'm not sure I fully understand.

    This is my current data graph; https://i.gyazo.com/72d8f842cc504e924d1d0f106fb0266a.png
    The first large peak corresponds to F- ions.
    If I assumed the signal as 0 how would I plot a calibration curve ? Would it be possible for me to choose 2 points on the curve which produce a linear relationship ?


    Thank you in advance for your help :)
     
  5. Nov 25, 2015 #4

    Borek

    User Avatar

    Staff: Mentor

    You know the area of a peak that correspond to a 10 ppm concentration, yes?

    Assume 0 ppm would produce no peak - or a peak of area 0.

    That gives you two points for calibration curve - one for 0 ppm concentration and one for 10 ppm concentration.
     
  6. Nov 25, 2015 #5
    I see, thanks.

    To calculate the area, can I normalise my data so that the area under the curve is directly po
    Thanks. I've managed to do it and used excel to calculate the gradient, in the form y=mx+c
    One last question.

    If I run a sample with an unknown concentration of F-, if I know the area under the curve (y) and the gradient (m), knowing that c=0, would I just solve for (x) which would be my concentration of F- in the unknown sample ?
     
  7. Nov 25, 2015 #6

    Borek

    User Avatar

    Staff: Mentor

    You lost something here. You can use peak height instead of the area, typically these are related (and in the case of a nice curve - like the one you have - it is quite good approach).

    Yes.
     
  8. Nov 25, 2015 #7
    We've been told to use area, so would normalising be the right idea for this approach ?
     
  9. Nov 25, 2015 #8

    Borek

    User Avatar

    Staff: Mentor

    If you are told to use area - use area.
     
  10. Nov 25, 2015 #9
    Thanks for your help :)
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: How to get calibration curve?
Loading...