Chemical Forums

Chemistry Forums for Students => Analytical Chemistry Forum => Topic started by: moltres on December 21, 2018, 04:11:41 AM

Title: Plot of Conc vs Abs with large intercept. What could be the cause?
Post by: moltres on December 21, 2018, 04:11:41 AM
With a UV-Vis spectrophotometer, I took the absorbance of standard solutions in triplicate over a certain concentration range and I obtained Absorbance values of ~0.1 to 0.5. I plotted concentration versus absorbance as per Lambert-Beer's law and obtained an equation for a straight line with R2 of >0.9. I also, however, acquired an intercept or constant of +0.218.

I struggled to find answers but according to my read through some of the literature and comments from colleagues, large intercepts such as the one I obtained can arise from instrumental variations but I am not entirely sure or convinced.

What could be the cause of a large intercept?
Title: Re: Plot of Conc vs Abs with large intercept. What could be the cause?
Post by: Borek on December 21, 2018, 04:24:45 AM
Have you measured the blind?

How does the curve look on the plot? It can have a nice correlation coefficient yet be obviously non-linear.
Title: Re: Plot of Conc vs Abs with large intercept. What could be the cause?
Post by: moltres on December 21, 2018, 04:45:59 AM
Yes, the absorbance values were obtained versus a solvent blank (water). And yes, my coefficient of determination is R2 high >0.9.

I obtained 0.001 for a water blank vs. blank so I'm still not sure why I have such an intercept.

Edit: Unfortunately, I can't show the calibration curve here because I am prohibited to do so.
Title: Re: Plot of Conc vs Abs with large intercept. What could be the cause?
Post by: Babcock_Hall on December 21, 2018, 07:11:04 AM
Are the residuals randomly or non-randomly distributed?
Title: Re: Plot of Conc vs Abs with large intercept. What could be the cause?
Post by: Corribus on December 21, 2018, 10:04:20 AM
I don't really understand why you'd be prohibited. At this point it's just numbers with no identifying information attached.

It's very hard to troubleshoot a problem without seeing the spread of the data.