[walt3r_whit3]

Hello everyone,

I'm currently writing a report for a HPLC experiment i performed. Basically, it's late at night, it's due tomorrow and i have stress-induced confusion (i think)

Experimental Details:

I ran a sample containing 35ppm ethyl paraben (lets call this sample A), 35ppm propyl paraben (sample B) and a sample containing 35ppm ethyl paraben and 35ppm propyl paraben (sample C).

Then i ran an unknown (sample D) . From the retention times, it is clear that the two peaks in the chromatogram correspond to ethyl paraben and propyl paraben.

I have been asked to estimate the concentrations of compounds in the "unknown" mixture (sample D). We are using an old machine, so i have to manually calculate peaks areas and such. I'm a a bit unsure about how to estimate the concentration  with the data i have.

All i can think of is taking an average of the peak area for each compound. Then using this data as a sort of one-point calibration curve and performing a simple rise/run calculation with the peak area of the unknown to estimate the concentration.

Does anybody have any other ideas or clues? Does this seem correct?

Thank you for you time!!!!

[Borek]

If the peaks look nice (symmetrical, well shaped, no tails), you can even assume concentration is just proportional to the peak height. If you have only one calibration point, I doubt wasting time for peak area calculation will make your result more accurate.

[walt3r_whit3]

There is some peak tailing. Considering i have to manually measure the peaks which are cut off, which means i have to use a ruler to extend the lines onto another page, there is practically no more effort involed to use the simple equation H * W1/2.

So it is ok to assume that concentration is proportional to peak height/peak area and use these single points to estimate the unknowns?

[Arkcon]

Yes, this is commonly done for well behaving chromatographic analysis.  You can quantitate on one point, even for environmental purposes, or pharmaceutical analysis, if the method calls for that, and apparently, yours does.  Usually, in analytical chemistry, it is better to have two or more points and regress the line.  But you'll have to assume its acceptable in this case.

[walt3r_whit3]

thank you so much for the clarification!

in previous reports we usually construct 6 point calibration curves, which is straight forward, but sometimes the questions for the reports can seem vague and i never really know if i'm doing it right, except when i have a reference or vague idea of the concentrations to be expected. in this case, i have none.

thanks again!