In using an unretained compound for determining the "approximate" void volume of a column, does one measure the void volume at the point of first deflection from baseline for the resulting peak, or does one use the retention time of the peak, i.e. the midpoint of peak elution?

I am currently using uracil as a void volume probe for reverse phase C8 and C18 columns that are pure alkyl, that is no imbedded polar groups that could retain uracil.

I've always used the maximum of the peak. In practice, it doesn't matter too much so long as you are consistent (precision is more important than accuracy here!)

-- Tom Jupille / LC Resources Inc

I would agree. The center of the peak however is actually the void volume. I would say also that I have found acetone to be a good choice also.

One other possibility is injecting a solution of potassium nitrate. Guaranteed no retention in the column in reversed phase systems (I don't know about straight phase systems). The middle of the peak represents the unretained component.

I have also used KI. When I first read about this tip, I was a bit surprised. But it works well.

All molecules of all compounds only pass through the column when they are in the mobile phase. Different compounds spend different lengths of time in the stationary phase. All compounds should spend exactly the same time in the mobile phase. In a perfect world all molecules of any particular compound should come out together in a single band sharp spike. The peak shapes we see in all but unretained compounds are of increasing width because of band broadening effects. The difference between the first molecules and last molecules to elute in an unretained peak are due to random statistical factors from different paths the mobile phase can take. The time at which the peak reaches its highest point is the statistical average time at which most of the molecules of the unretained compound elutes. The top of the peak is therefore the time used to determine void volume.

The determination of the "unretained" peak is an interesting exercise, full of little complications and surprises. For the average practitioner, a simple and consistent choice is good enough. I used to use dihydroxyacetone. Most of my colleagues have settled for uracil. We always use the run-of-the-mill tip of the peak, as determined by standard software. For anything but sophisticated theoretical stuff, this is good enough.