Quantify by capillary column

Chromatography Forum: GC Archives: Quantify by capillary column
Top of pagePrevious messageNext messageBottom of pageLink to this message  By Pierre Leblanc on Thursday, February 14, 2002 - 06:15 am:

I am a young scientist and I need a help from expperienced analyst. my question is when I use capillary column to quantify , should I use the hight or the surface of the peak and why?

Top of pagePrevious messageNext messageBottom of pageLink to this message  By Bruce Freeman on Thursday, February 14, 2002 - 12:56 pm:

The answer depends upon a number of factors. Peak height can be more precise, but that statement comes with considerable qualification.

Typically, quatitation in chromatography is done by the "external standard" technique, in which you quantitatively compare a peak in the chromatogram of your sample to the corresponding peak in the chromatogram of the standard. If the peak shapes are identical in sample and standard, then peak height can be the better method. Otherwise, peak area must be used. Matrix effects can often skew or otherwise affect the shape of a peak. Peak width is a good quick way of gauging peak shape -- any difference in width between sample and standard means that peak height is an unreliable measure.

Then, too, is the question of how the peak height is measured. In the bad old days this was done with a ruler -- or by reading from the chart paper scale. This was a great deal more accurate than planimetry or "cut and weigh" techniques for measuring peak area. We've come a long way from those days with electronic measurements, but note that peak height has to be measured at the peak apex.

Beckman Process Instruments Division used to market a "peak picker" that would record the exact peak maximum for any peak. It was an analog circuit that digitized the result after capture. This was ideal. However you won't find that anymore. ADC's used on modern instruments cannot be expected to happen to take a measurement at the exact moment of the peak apex. For measuring area, this introduces little error, providing sampling rate is fast enough -- like ten (signal) samples per peak or so. But for peak height measurements, ten samples per peak means you're almost guaranteed to miss the peak apex.

(When capillary columns were first introduced, what data systems there were couldn't handle the necessary acquisition speed to measure ten samples per peak. Hence, peak height measurements briefly came to the fore once again. That is no longer an issue, however.)

Chromatographic data systems make up for this deficiency in sampling by back-calculating the peak apex using a quadratic fit, which is a reasonable approximation to the shape of a typical chromatographic peak. Most likely, this is good enough, unless the algorithm in your data system is faulty. However, the data system probably assumes a Gaussian peak shape. If your peaks are not symmetrical, then this approximation might be faulty.

In general, peak height measurements excel in any situation in which peak area cannot be measured accurately (e.g., due to baseline noise or peak overlap), but in which the peak shape is good and the baseline can be reconstructed with good confidence..

Bottom line: Use peak areas most of the time. Use peak heights when there is an overriding reason to do so.

Posting is currently disabled in this topic. Contact your discussion moderator for more information.