I have performed an experiment where I had some test tubes with the reagents and different amounts of catalyst. I started the reactions and measured the time they took.

Then I calculated the rate of reaction and plotted it in function of the amount of catalyst (in mL) in each test tube.

However I can't decide between a logarithmic fit line and a linear one (though the logarithmic one has a higher $R^{2}$). The data points are below.

If it helps, the reaction we used was the one of potassium permanganate with oxalic acid. We also added sulfuric acid and the catalyst used was manganese sulphate.

What would make more sense?

enter image description here

0 4e-05

0.2 7.22e-05

0.4 0.000136

0.6 0.000159

0.8 0.000183

1 0.0002

  • $\begingroup$ Neither of the choices makes sense without a mechanistic reason to believe that. What I mean is that it is very tricky to fit experimental data to infer the order of the reaction (in this case wrt the catalyst). If on the other hand you have a reaction mechanism that predicts a first order dependence, then it makes absolute sense tot fit a linear curve $\endgroup$ – Michiel Oct 30 '14 at 9:47
  • 2
    $\begingroup$ One other thing: do NOT use $R^2$ to decide whether one fit is better than another. It is not meant for that. The only thing $R^2$ tests is how well the current fit (linear, logarithmic, whatever) does as compared to just averaging the values (i.e. assuming a constant value). $\endgroup$ – Michiel Oct 30 '14 at 9:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.