Page 1 of 1

why gc give low assay result where hPLC gives right result

Posted: Sun Jul 03, 2011 8:11 am
by helal1703
I been asked a question in an interview and that is
same product is analysed by HPLC and GC.But HPLC result is ok but GC is giving low result suppose 93% where it should be give result 98%.Which are the possible reason behind it.
I answered probable proper extraction solvent is not been used that why result is low.but interview panel wasnt happy with my answer.

please help me to get right answer

regards
Helal

Re: why gc give low assay result where hPLC gives right resu

Posted: Sun Jul 03, 2011 5:34 pm
by 5.3ctwptr
Was sample extraction suggested or implied? Assuming the Standard is prepared in solvent and is pure analyst gets 100% recovery, but sample matrix may interfere and recovery is low. If method validated and extraction is required, then analyst error is the answer. If method validated for A/R @ 98.0%-102.0% what about the Standard bracket? Did it pass? A failure may suggest an instrumental error. What are the rest of the details to the question? If sample is pure and method validated std works, bracket pass then analyst error. The answer is in the details. What are the details?

Re: why gc give low assay result where hPLC gives right resu

Posted: Mon Jul 04, 2011 4:57 am
by Don_Hilton
Two thoughts come to mind
If they were looking to see your understanding of differences between GC and LC techniques, they may have been looking for you to describe something about thermal degredation in the inlet - but as pointed out above, the details matter, because if you are using an isotope diltuion method, losses will be equal, and quatitation will work correctly.

Second point is are we looking at internal standard calibration or external? If it is internal standard what differences are there between the internal stanard and the compounds of interest? Is the value to be determined percent area in an entire sample? (discrimination in the inlet could be a problem.) As pointed out - there are some details needed.