Page 1 of 1
How to define or calculate the peak width of a flat-top peak
Posted: Mon Jul 20, 2009 6:57 pm
by mingh
How can we define the peak width of a flat-top peak? Is there any equation or program? Should we use peak width at half maximum or baseline width?
Posted: Mon Jul 20, 2009 9:45 pm
by skunked_once
If you have a flat-topped peak, most likely your detector is overloaded and you are above its maximum signal range. In this case measuring peak area accurately is impossible.
Posted: Mon Jul 27, 2009 7:16 pm
by Ron
To be more precise, you could determine the area of a truncated peak, but that would not be the true area produced from that quantity of analyte. So even if you could determine the area of a truncated peak, your calculated value of the quantitative result would be incorrect.
Posted: Tue Jul 28, 2009 8:02 am
by Peter Apps
GC software will probably not work well if you ask it to give you width at half height of a flat-topped peak but I do not anticipate any problems if you ask for baseline width. The problem with baseline width is that it can be sensitive to minor changes in peak shape unless you set the integration parameters to give robust detection of peak start and end.
I'm interested in why you are working with flat-topped peaks ?
Peter
Posted: Tue Jul 28, 2009 6:43 pm
by tom jupille
Why do you need to know the width?
And how accurately do you need to know it?
As others have pointed out, if you have a flat-top peak, you are well outside the linear range of your system. Width at half-height is meaningless because you have no way of knowing the true maximum height (your peak is truncated). Width at baseline is determined by drawing tangents to the inflection points. With a truncated peak, you cannot accurately identify those inflection points. That said, a manual measurement with a pencil and straight edge can give you a crude estimate. Depending on your purpose, that may be good enough.
Posted: Thu Sep 17, 2009 1:50 pm
by bisonium
Hi!
If you are looking at fluorescence (FLD?) there sometimes are what the manufaturers (Waters) call "sticky settings" to use when setting up a new method. One of these settings is for example pmt/100, pmt/1000 or pmt left normal. These settings make it possible to see saturated peaks as if they were not saturated, the FLD is somewhat less sensitive.
Could this perhaps help you out?