1. Limited time only! Sign up for a free 30min personal 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!

Advantage/Purpose of Transmission vs Absorbance in calibration curve

  1. Feb 5, 2014 #1

    ~christina~

    User Avatar
    Gold Member

    1. The problem statement, all variables and given/known data

    The problem involves a calibration curve which was created and where % T vs concentration was favored for use over the usual Absorbance vs concentration and a question of why would this be used in favor over the latter relationship curve was presented.

    other details:
    1. % concentration is from 0.35-0.05 with a 0.1 interval between points. Total points on graph = 7
    2. Max transmitance occuring at the 0.05 % concentration is ~80%T
    3. Minimum transmittance at the 0.35% concentration is ~ 15%T

    Best fit line through the points is linear in relationship [exception for this is the highest concentration %T point which is somewhat above the line]

    2. Relevant equations

    [itex]A= abc[/itex]

    [itex]A=-log (T) [/itex]

    [itex]T= [/itex]Exiting light intensity / Incident light intensity = I / Io

    3. The attempt at a solution

    What I know:

    1. % Transmittance vs concentration plot usually has a negative sloping exponential relationship. (not talking about the one here which is linear in the range observed)
    2. Due to #1 above, it is preferential to plot calibrations curves using absorbance vs concentration. This is a result of the relationship A = abc which demonstrates that concentration is directly proportional to the absorbance of a solution. Generally it is linear within the working range of absorbance being ~ 0.10-0.82 from what I looked up.

    I was thinking that the reason that the % T vs concentration was used was that since the concentration range observed on the graph (0.05-0.3%) is small and as a result, in that area the relationship between %T and concentration is linear, it would not matter if one chose to use this calibration plot vs a abs vs concentration since the latter would be linear as well.

    But as to why % T was chosen over using Abs for the graph, I am unsure about since from what my thoughts on the subject tell me, they would be equivalent in this case (except for the positive slope in the abs vs concentration graph). In addition, all I could generally find around the web and in my text enlighten me on the benefits of using abs vs concentration as opposed to the cases in which %T would more favorable for use. :frown:

    I would appreciate any comments on whether my thoughts are correct or not.

    Thanks!
     
    Last edited: Feb 5, 2014
  2. jcsd
  3. Feb 5, 2014 #2

    Borek

    User Avatar

    Staff: Mentor

    Have you tried to plot A vs c?
     
  4. Feb 5, 2014 #3

    ~christina~

    User Avatar
    Gold Member

    I have converted the values based on approximation of the values given on the graph (just have the graph).

    Values based on my eyeball view:
    % T values (concentration) => 80% (0.05%), 60% (0.10%); 45% (0.15%), 32% (0.20%), 22% (0.25%), 17% (0.30%), 15% (0.35%)

    conversion:

    Abs value (concentration) => 0.0969 (0.05%), 0.2218 (0.10%); 0.346 (0.15%), 0.495 (0.20%), 0.657 (0.25%), 0.769 (0.30%), 0.824 (0.35%)

    Revelation: nothing much

    It looks the exact same as the transmittance except with a positive sloping graph when abs is plotted. Of course, the last value deviates from beers law but that is because it is at a higher concentration I guess.
     
  5. Feb 5, 2014 #4

    Borek

    User Avatar

    Staff: Mentor

    Strange. Neither plot looks completely linear, but T plot shows an obvious curvature, so I would prefer A plot. Unless some more advanced statistical analysis would point to T. Not that I have any idea what "more advanced" analysis could mean.

    attachment.php?attachmentid=66331&stc=1&d=1391636234.png
     

    Attached Files:

  6. Feb 5, 2014 #5

    ~christina~

    User Avatar
    Gold Member

    Hm, It looks exactly the same to me as it is plotted except they just used linear regression to find the best fit line which left the last point out. It was a bit straighter though.

    I just noticed something odd I didn't take notice of before for some reason. This may be why it was strange when you plotted it. The scale on the y axis is not spaced evenly. I measured the spacing btw the %T values and it is as follows:

    10-20%T (2cm), 20-30%T (1.2cm), 30-40%T (0.8cm), 40-50%T (0.65cm), 50-60%T(0.5cm), 60-70% (0.4cm), 70-80% (0.4cm), 80-90%T (0.35cm), 90-100%T (0.4cm)

    Now I'm even more perplexed about the problem.

    I'm not sure about the relationship between the numbers and isn't an axis usually spaced evenly? Is there any case where this would not happen like for my graph?

    Thanks Borek
     
    Last edited: Feb 5, 2014
  7. Feb 6, 2014 #6

    Borek

    User Avatar

    Staff: Mentor

    Sounds like a special graph paper - they come in many flavors, so that you don't have to convert your data. Google for logarithmic paper and you will see what I mean.
     
  8. Feb 6, 2014 #7

    ~christina~

    User Avatar
    Gold Member

    Ah, that makes sense since that is what it was. I haven't seen something like this before. This explains what I'm seeing and probably the question I guess.

    Thanks Borek!
     
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: Advantage/Purpose of Transmission vs Absorbance in calibration curve
Loading...