Sensitivity in UHPLC vs Classical HPLC

[Jimi]

Hello to all Ultra-High Pressure Experts

I would like to pose a question about the effect on sensitivity, on going from 'conventional HPLC' to 'Ultra High Pressure HPLC'.

Now I understand that it breaks down to two competing factors:

- The smaller particle diameter, and the shorter column lengths usually used for UHPLC, tend to promote sharper peaks and, hence, better sensitivity.

- The need to go to a smaller size detection cell (in UV), as your reduce the ID of the column, tends to promote worse sensitivity. There has to be some 'pathlength penalty'.

But we are having difficulty finding any publications that discuss the combination of these factors more quantitatively, or at least with some experimental data.

I guess the 'pathlength penalty' would vary somewhat from vendor to vendor, depending on exactly what size detector cells they make.

Any thoughts - or recommended reading - are very much welcome.

Thanks

[HPLCaddict]

Concerning pathlength of the flow-cell, this is an issue mainly with the "first-generation" UHPLCs. Back then, the most direct way to reduce flow-cell volume was used - decreasing the path-length. So, 5mm or 3mm flow-cells were usually used instead of the "traditional" 10mm path length, possibly leading to lower sensitivity.
In the meantime, technology has advanced. Contemporary UHPLCs usually use flow-cells based on total reflection which makes it possible to have the standard 10mm path length combined with very low volumes. Therefore, the "path-length issue" may be considered resolved - at least with some contemporary UHPLCs.

[tom jupille]

This will sound nit-picky, but it really matters. Are you defining sensitivity in terms of concentration injected (eg. ug/mL) or in terms of mass-on-column (e.g, 5 ng)?

If you are thinking mass-on-column, then it's pretty much as stated. Assuming you scale the column length to the particle size, HPLC and UHPLC will give about the same plate number, so the sensitivity will be inversely proportional to the column volume (e.g., going from a 100 x 4.6mm, 3.5-micron to a 50 x 2.1 mm 1.7 micron will give you 10x lower LOQ) and approximately directly proportional to the flow cell path length.

If you're thinking concentration, then it's all over the map. Injection volume is an issue if the diluent is equal to or stronger than the mobile phase. If the diluent is substantially weaker than the mobile phase, then you may still be able to get away with a large injection volume.

[Jimi]

Hmm....yeah....I don't know. I don't think it's true that the pathlength problem has been solved. I am familiar with the 'reflection technology' (or whatever is the proper term) that makes it possible to have a long and thin detection cell, so the pathlength is longer than it otherwise would be. But it stands to reason that whatever pathlength they're able to make work with a 1 mm ID column, they could use an even longer one with a 2 mm ID column, and longer still with a 4 mm ID column. So I still think there has to be some pathlength penalty when going to a smaller ID column.

Tom: I am referring to concentration sensitivity, e.g. with a UV detector. My inherent assumption is that whatever is the maximum volume one can inject in a given situation, it would scale with column diameter squared. So the concentration on-column and in the detector will be the same. So this is why I focus on the 'pathlength penalty' issue, because I think that's what it comes down to.

By the way, I posed the question in terms of UPLC but, more generally, this comes up any time you want to reduce the column ID. For example, if someone decides to go from a 4.6 mm column to a 2 mm column (both in conventional HPLC mode) in order to save solvent and reduce waste, this issue comes up as well.