Extinction Coefficient from Time series data

[lee403]

I have some time series data of the absorbance of Br₂ formation using UV Vis spectroscopy and I need to figure out the extinction coefficient/ absorptivity.
The overall reaction is
BrO₃^-+5Br^- +6H⁺-->3Br₂+3H₂O
which is expcted to go to completion
I know that the equation relating absorbance to concentration is
A=εcl
and I have times series A measurements and can calculate the initial concentrations of BrO₃, Br ^-and the expected concentration of Br₂ from the solutions I made. I just need to find ε.

I first attempted to plot the absorbance v. time and find the slope where it was most linear but I don't know how valid this approach is.

 

[Borek]

I first attempted to plot the absorbance v. time and find the slope where it was most linear but I don't know how valid this approach is.

It is about as good as it can be.

I would record an additional point after waiting for some time to make sure the amount of Br₂ produced is just stoichiometric. That would give a good calibration point.

Besides, if all you are after is a time series (for kinetic measurements), all you are interested in is the rate of changes - are you sure you need absolute values for that?

 

[lee403]

Maybe I'm thinking about this wrong. I need to know the extinction coefficient for Br₂ because in another reaction I measure its loss over time. So I dd an initial run for the formation of Br₂ in order to determine its extinction coefficient. Then in a second run I added a compound that reacts with it and measured the absorbance again. I am interested in the rate of loss so does that mean the extinction coefficient from the initial run is not an exact value?

 

[Borek]

If you are using the same cuvette, wavelength and the same instrument you don't need extinction coefficient but a calibration curve. Linear regression on the data is typically a way to go.

What I don't get about your setup is why you use time series instead of just making a series of samples of different concentrations?