peak purity

[jcmc]

I need to produce a COA that contains a peak purity analysis. My question is what values are typically determined in this type of analysis. I have Chromeleon v. 6.8 software. Does anybody know what peak purity match value is (I get a value of >900) and what it should be? I also see a peak purity index value that seems to be in wavelenght, but not at my lamda max?

[tom jupille]

"peak purity" is calculated in different ways by different data systems (and, in my opinion, are of dubious value because many coeluting peaks have very similar absorbance spectra). Most of them look for variations in the absorbance spectrum across the peak (and these variations are not necessarily at lambda max). I don't have any direct experience with Chromeleon, so I can't g.ive you any more specific help.

Best bet would be to contact your local Dionex tech service person

[HPLCCONSULT]

Most software packages calculate purity against a scale from 1 to 1,000 based on a 0.1% scale. This means that a "purity" of 990 would equal a purity of 99.0%. A value of 999 would give you 99.9% pure. Is pure 99 % ? Maybe 99.8% ? Who knows as it is up to you to define "pure" or "purity" for that matter. *BTW: Pretty sure that Chromeleon uses this same scale of 1 to 1,000.

However, I can easily set up the "Peak Purity" paramenters (or worse yet use the manufacter's default values!!!) to obtain any value for purity you would like. There are a large number of parameter's used to determine peak purity by UV/VIS. They also assume that you have excellent chromatography (which you should, but few actually do) and understand how to set up and use the different parameters correctly. It is a qualitative assessment of "purity" at best. Be careful...