Chromatography Forum

Hi everyone,

I have a question that is bugging me. Lets say I have a T-union between my HPLC column and MS detector (ESI) which splits my effluent, so that 10-100% of the effluent reaches MS detector.

Since ESI is a concentration dependent detector, should I get the same sensitivity in all cases, regardles if 10% or 90% of the effluent reaches the MS? The concentration does not change, only the flow rate changes, right? Therefore, peak hight should stay the same but peak widths (area) should vary?

If someone could elaborate on that I would really appreciate it.

For the sake of the argument, let's assume the ideal case where ionization is optimized in every case of split ratio.

Cheers

Is there really no one who could help me understand or direct me to some literature where I could find some answers?

I thank you for your help in advance.

Cheers
A

I will agree that ESI depends mainly on the concentration. Nevertheless this is only an approximation since the ionization process is more complex. Such surface-reactions are really good for some surprises.

Please keep also in mind, that ionization efficiency with ESI is usually better when a lower flow is used.

But as usual, it depends always on your individual experiment. It is simply not possibly to give you here general recommendations. Sorry.

Regards
Klaus

No. Although the ESI does exhibit concentration-dependent tendencies the MS is highly affected by flow rates. Assuming that you have a 1 mL/minute flow that you split anywhere from 10% to 100%, you will get significant improvement in sensitivity with higher split ratios (less of the material directed into the ESI/MS.) This is the same thing that happens when you move to smaller diameter columns; the flow rates drop significantly as the diameter of the column decreases and yet, conversely, the sensitivity (in terms of LOD) goes way UP with decreasing column diameter.

Thank you for your thoughts guys.

I know the sensitivity increases in ESI when going to narrow bore columns (reducing flow rate). I was just wondering if the increase in sensitivity when doing the split (reducing flow into MS) might be countered by a lower amount of sample (absolute) reaching the MS. Since ESI is a concentration dependent detector (at least at "high" flow rates), this should not be the case.

Thank you again for your help.

Cheers

Decreasing flow without increasing concentration doesn't improve sensitivity unless your source is already operating at non-optimal flows. If you are running a typical modern electrospray source at 200 microL per min, then doing a post-column split to get it down to 20 will probably just give you very broad peaks and longer delay times. If you're running it at 1mL/min because the manufacturer says you can (because they had to make it do this or the instrument wouldn't compete with other manufacturers...), and it's struggling to dry the solvent adequately, then a post-column split will help.
But if you're doing this, you'll get a much better increase in sensitivity by going down to smaller-bore columns and running them at the lower flow without the split.
The split obviously doesn't change the concentration, and that is the fundamental issue. Protein people do nano-flow because they can then use a very narrow column and inject tiny absolute amounts of peptides but still achieve fairly good molar concentrations in the tiny volumes that they put into the spray chamber.

Imh, thank you also for your imput.

The thing is that "they" want me to run an LC-MS method with a high aqueous percentage at 1.3 mL/min and no changes are allowed. I dont feel comfortable flooding my MS with this. Thats why I did the split and wondered what kind of results can I expect (sensitivity wise) apart from broader peaks. If not an increase in sensitivity at least I didnt want it to drop (get worse) with the split.

I usually use 2.1 mm i.d. colums for MS, but in this case, this was a no-go from the start.

Cheers

Then you're doing exactly the right thing (in my view). If you're running at that flow, splitting really shouldn't cause any problems to your peak shape because the flow to the MS will still be far too fast to allow broadening, and the concentration is just the same so the height shouldn't change. Compared to a theoretical LC-MS that could handle 1.3mL/min your result will be almost unchanged, but real LC-MS systems obviously can't handle that flow rate, and you're quite right, would suffer horrible reduced sensitivity, noise, and maintenance issues because of filling up with buffer. I love 2mm columns too. It's a pity "they" don't share your opinion on that one!

The thing is that "they" want me to run an LC-MS method with a high aqueous percentage at 1.3 mL/min and no changes are allowed.

In this case, post-column split at 1:4 - 1:6 (static) is the only way to go...

This is an interesting topic. How would one physically set up a T-union with a known split ratio?

This is an interesting topic. How would one physically set up a T-union with a known split ratio?

Split ratio calculation (optimization length of restriction capillary) based on volumetric, gravimetric or pressure drop measurements.

This is an interesting topic. How would one physically set up a T-union with a known split ratio?

Split ratio calculation (optimization length of restriction capillary) based on volumetric, gravimetric or pressure drop measurements.

Theoretical calculations of a split ratio are merely an approximation at best, since the backpressure of an ESI (or any other ion source) can vary greatly with experimental conditions.

I usually take one piece of narrow bore capillary, cut it to an approximately appropriate length and incorporate one end into the T-link and the other into waste. Then I measure the "waste" flow rate (stop watch + measuring cilinder) and shorten the capillary again if the ratio is too small.

There are also split flow regulators on the market but they cost a lot of $$$.

Yes, all flow splitters are basically a pair of resistances. If you buy a nice expensive fixed-ratio splitter, you're buying a pair of resistances that someone has carefully calibrated to give the desired split-ratio .... without an MS on the end! If you add an MS with extra resistance, the split ratio will be different. You might just as well make up your own with thin tubing and a tee-piece.
If you buy a variable splitter, you have the convenience of a knob to turn until you have the desired flow. Even these, however, are still just resistors, so if you determine your flow by detaching the tube from the MS and measuring the volume that comes out in a minute, as soon as you attach the tube again, the flow drops.
It is quite difficult to assess the split-ratio of a splitter going into two destructive detectors, or where the available flow to measure is very large compared to the other (e.g. MS and fraction collector, where you really want to know whether 100uL/min or 120uL/min is going to the MS, so you can assess the delay time, but the flow you can measure is that to the fraction collector, which is 4900uL/min or 4880uL/min, hard to measure with adequate precision).

I usually take one piece of narrow bore capillary...

Usually, I do the same also, but once i used backpressure measurements (restriction capillary ending vs ESI ending) and its worked fine.