Injection solvent for reverse phase

[kouroshh1]

Can someone please explain to me the effect of injection solvent on the peak shape and resolution? My mobile phase is starting with 95 % water and 5 % Methanol for 3 minutes then it goes through a linear gradient. Should I dissolve my standard in 5 % methanol before injecting it into an LC system?

[HPLC chemist]

An organic injection solvent like Methanol, ACN, IPA... will sharpen the peak and bring it closer to the solvent front. This is true in 'reverse phase' (C18) but untrue in HILIC (different retention mechanism). In this case because the target molecule is 'water loving' the increased addition of organic (and lowering of %water) will retard the elution.

[HPLC chemist]

If your molecule loves Methanol, fine! Once injected onto the column it is just going to sit there until the gradient composition (%organic) allows it to be solubilized and thus elute.

[mattmullaney]

Hi Again, kouroshh1,

Here are three links that you can peruse:

http://www.chromatographyonline.com/rol ... -solvent-0

https://www.crawfordscientific.com/tech ... ts-in-hplc

http://www.restek.com/Pages/faq_lc

Like many things in chromatography...it depends on the situation at hand. If you inject a small volume (relatively speaking) of a strongly-eluting solvent containing your analyte(s) of interest, the peak shape and retention factor may be okay...and there is a chance that if you do this, it will also Not be Okay.

Got to try it out and see.

General recommendation for gradient chromatography is to use for the sample solvent the initial mobile phase composition of the gradient program...or a weaker-eluting solvent than the initial mobile phase composition.

Best Wishes!

[mattmullaney]

OOPS!!!

First two links didn't seem to work well, I'll try once again:

http://www.chromatographyonline.com/rol ... -solvent-0

http://www.chromacademy.com/chromatogra ... -HPLC.html

To see the second link, you may have to sign up for a "Lite" membership at ChromAcademy. IMHO, it is worthwhile to do so.

[HPLCaddict]

An organic injection solvent like Methanol, ACN, IPA... will sharpen the peak

No.

If your molecule loves Methanol, fine! Once injected onto the column it is just going to sit there until the gradient composition (%organic) allows it to be solubilized and thus elute.

No.

Remember, chromatography is a series of equillibration steps of the analyte between stationary and mobile phase. If your injection solvent is significantly different from the (initial) mobile phase, there is a severe disruption of the equillibrium, because initially the analyte "sees" only the injection solvent and not the mobile phase. This can lead to all sorts of peak broadening and peak shape issues.
Using pure methanol or the like as sample solvent in reversed-phase is a no-go from a chromatographic point of view - but sometimes you don't have much options because of solubility issues.
Generally, you cannot predict if you'll see any artifacts from using a certain solvent - it heavily depends on all sorts of variables like column dimensions, injection volume, dead-volume of the system, flow-rate etc.

The best thing to do from chromatographic theory is to use the mobile phase (in case of gradients: the initial mobile phase) as sample solvent, so there will be no equillibrium issues.
The best thing to do from a practical point of view is to use a weaker (in terms of elution power) sample solvent in order to "compress" the peak at the head of the column. In the case of reversed-phase this means using a solvent which is more aqueous than the initial mobile phase. In the case of HILIC it's the other way round, you should use more organic in the sample solvent, because here water is the strong eluent.
If you MUST use a stronger solvent, limit the injection volume as low as possible a warch out for any peak shape issues.
An remember, organic/aqueous ratio is not the only possible difference, pH and salt concentration differences might also have influences (especially in HILIC).

[lmh]

Yes, listen to HPLC-addict on this one, they're right. Viewed simplistically, the analyte sees only its local environment. If you inject it in a high percentage of methanol or acetonitrile, which are strongly eluting solvents in reverse phase, it will not bind to the column, because its local environment is strongly eluting.

Of course the injection solvent will mix by diffusion with the running buffer (the column is a giant and effective mixer). Meanwhile if there is any slight binding between the analyte and the column, the analyte will be delayed relative to the injection solvent, and the two will start to separate, allowing the analyte to bind more strongly.

In the utter extreme case that you inject a very large volume of sample in a very strongly eluting solvent, the dilution-effect will never be enough, and much of the analyte will pass straight through the column without binding at all.

In the less extreme case, it will spread out into the column. The analyte near the tail end of the plug of injection solvent will quickly mix with following running buffer and bind at the start of the column, while analyte in the middle of the plug will move much further in. In the early-eluting peaks, the analyte never really stops, it just moves (almost isocratically, even if you're running a gradient), and thus emerges as a spread-out and misshapen peak. Late-eluting peaks tend to be less affected, because they are more strongly retained by the column, and therefore require less dilution by the running buffer before they behave themselves and bind.

Summary: if you inject in the same solvent as the running buffer, you are safe (unless you do something really silly, like inject a truly enormous volume in isocratic chromatography where the analyte is already mobile). If you inject in a solvent that is weaker than the running buffer, you're probably OK. If you inject in a solvent that is stronger than the running buffer, then keep the volume down, so the spreading effect is reduced to a minimum; the likely symptoms of a problem will be bad peak-shape and width of early peaks, and possibly unreliable retention times of these peaks.

[mattmullaney]

Hi Again, kouroshh1,

I concur with the last two gentlemen...I probably submitted too many links to read. The links state the same things we are all saying, without the brevity of thought.