This is probably a real simple question, but I've been struggling with it today. I have three columns which have been individually tested for HETP and asymmetry. The three columns are then put in series and re-tested. Is there a relationship between the individual Asymmetry values and the combined one? The individual values have been ~1.9, 2.1, 1.8 and the combined value is ~4.6 (these are large production columns, so the As values are terrible!). Any suggestions would be appreciated.

Connecting in series? Are you creating voids that would lead to the Asymmetry problem?

Unfortunately due to the size of this SEC step, we are forced to run three large columns in series. We are sensitive to the fact that a majority of the band spreading is due to the multiple column configuration and plumbing in- between. We have minimized void volume as much as possible, but this is still an issue as you point out. The only thing I am trying to determine is if I can relate the individual column Asymmetries to the combined series asymmetry? Or, as you mention, does the void space dominate this situation?

The real issue is the asymmetry of the peaks. As you add columns in series, the plate count improves, and so does the symmetrical portion of your peak. However, you got some asymmetric portion, and this will become more pronounced, if the symmetrical portion of your peak improves, while the assymmetrical portion of your peak does not.

The real question is the reason for the asymmetry. Why do you have an asymmetry of around 2.0 to start with? If it is caused by flow nonuniformity or improper sample distribution, it is not impossible that it simply adds up.

Actually, you can get the theoreticians to do triple mental headstands on this question...

Once upon a time an innocent sample enters a chromatographic column “the first one” and doesn’t know what is going on therein “no body knows actually”. But in no time some part of its “brain” or whatever it was, realized that it is not a simple sample, but an important chromatographic band “peak of interest”. And this peak of interest is leaving that first column with an asymmetry of 1.8. So it already knows that it is an object of importance, but starts to wander why somebody, so badly likes to make an ugly unshaped “asymmetric” dude out of that Importance, when forced him into another tube “chromatographic column”, and, with no patient into a third one.
I do not know either.

Hello John, you mentioned that this is a kind of production size SEC method. Is it possible for you to change the column to a more "high resolution" column! There are some new SEC column materials available, and there is a chance than to improve your asymmetry factor. With smaler particle size these materials give a much better performance. And perhaps you can get a better separation on just one column. Please don't hesitate to e-mail me: g.kratz@tosohbiosep.de!

John,
My estimate is that the asymmetry would add as the square root of the sum of the squares at best.
That is, you would expect 3.4 if your column to column connections were perfect.
This type of estimate assumes gaussian behavior, maybe not a good idea as asymmetry is not gaussian. Your value seems real world to me, not some thing that you can fix easily. Better columns or chromatography probably fixes the problem if the asymmetry troubles you.
Jim