We are in dispute in our lab over where exactly t0 is measured from. The USP/BP/EP state it's the apex of the first unretained peak. Our problem is that there often isn't an unretained peak. As a result we tend to use the first inflection of the baseline around the estimated solvent front (from John Dolan's equation). Does anyone have a standard procedure for calculating t0? We are considering introducing another injection as part of any suitability which would be of this unretained component, but aren't sure what to use - any ideas (this is for reverse phase HPLC using many different columns/mobile phases, so needs to be universal)?

Depends on what you want to do with the information. If you just want it to characterize retention behavior, then so long as you're consistent, using the basline inflection ("RI anomaly", "Schlieren peak", . . .) is probably OK. If you really need to get values for things like distribution coefficients, then the choice of marker is more important.

The usual choices for reversed-phase are things like uracil, nitrate (sodium nitrate or nitric acid), or azide (often added as a bacteriostat anyway).

A couple of decades ago, the group I was with was trying to get accurate t0 measurements in order to characterize columns. We found that we would get different values (consistent differences of a few percent) with different mobile phases (varied with both % organic and with the type of solvent used). The "hand-waving" explanations that were consistent with observed behavior included:
- change in the bonded-phase volume resulting from solvation differences (hence changing the volume of bulk mobile phase in the column)
- size exclusion effects (bigger molecules can't penetrate into all of the small pores)
- charge effects (molecules may be retained or excluded from small pores depending on their charge and the effective charge of the packing).

We couldn't justify the time to delve any further, but the fact is that t0 is not at all as constant as we'd like to believe. The extreme example, of course, applies in SEC, where analyte elutes at (its particular) t0.

Someone else may be able to comment more intelligently on the topic!

-- Tom Jupille / LC Resources Inc.

I agree with Tom that for most practical purposes the RI deflection or uracil is fine. For a more rigorous determination the column evaluation team of E19 used a MeOH/H2O mp and deuterated MeOH/ H20 (of the same composition) as an injection coupled w RI detection. The dead volume obtained was almost identical with the uracil value.

I agree with Tom. Depending on what is used for a "nonretained" marker the apparent t,o will vary more than the marker peak width.