Chromatography Forum

Hi All,

I have been thinking about the point of adding buffer at the same concentration, e.g. 1 g/L AmAc and 1 mL/L acetic acid, to both mobile phases rather than just the aqueous part- This pertains to HILIC and RP. I keep reading that it is important to maintain constant ionic strenght. However, don't we have a gradient in apparant pH anyway, as the mobile phase becomes more aqueous (HILIC) or more organic (RP).

Thinking about HILIC - doesn't the buffer enrich in the water layer anyway? And couldn't there therefore be an equally logical argument for only including buffer in MPA?

Does anyone have empirical evidence for improved results when the buffer is added to both mobile phases (e.g. 5% ACN in water and 5% water in ACN for reversed phase or 95% ACN and 1:1 ACN:water for HILIC.

We have good methods in the lab that use both cases. One has 10mM AmAc pH 5 in both A (95% ACN) and B (50:50) whereas another method has 25mM ammonium formate pH adjusted to 3.6 just in the aqueous part of the eluent.

Interested to hear your thoughts!

Arne

I'm retired, after 4 decades with chromatography.

We NEVER added buffer to the organic phase in an aqueous-organic system. We didn't use HILIC.

However, when using refractive index detector, we did pre-mix (and add buffer if necessary) to the mobile phase, then filtered.

Hi Arne,

I think you are right on pretty much all of your statements, but it is more of a "is it feasible/worthy to do this?". let me put a silly example here: you have MPA as a PBS and MPB as ACN, as far as i am aware, it should not be possible to dissolve phosphate salts in ACN unless you are pulling some ionic liquid stuff, so it would not be something very feasible.

Discussing about HILIC, hey that's some dangerous territory right there, i would avoid running HILIC gradients, HILIC requires quite some time to equilibrate. you may need less eq time if using no more than 20% water as mobile phase and be able to even reduce it more with pseudo-equilibration states... but this is something I would advise not to do simply because the hydration layer needs stabilization... so by running gradients you have a less robust method going on.

This being said, i can also tell that you need to have a clear understanding of what exactly you need when using the same additive loading on both phases. If FMB is 100% organic or close to it you are probably not going to be able to have the salt quantity you would have on FMA. And even if you have water on FMB it might not be worth to have same amounts if your goal is to maintain microspecies the way they are (because the addition of organic solvents changes pH) or even maintain ionic strength. I take that it is way more common to just increase salt quantity in FMA until ionic strength is not an issue (tbh traditional RP or even HILIC should not be considering ionic interactions as actual contributions even though they can be around). For this i have two advices: 1-) when adding more salt to FMA, be aware that this is going to be mixed with organic solvents so you can only do this after you confirm the salt compatibility with the proportion of water/organic you will be using do some maths in order to assure there is not going to be a single moment where there is too much salt for the certain composition of water/organic, avoid salt precipitation (salting out) in your system at all cost! 2-) If you are running mixed mode you might really need to have some ionic strength stabilization (not mandatorry though). go ahead, but first confirm the salt concentration you will apply can be used in both the water/organic mixtures you are going to be using as FMA and FMB.

I will go ahead and bring something up that may or may not be related to this topic. something that crosses my way far more is the addition of acids in both mobile phases. everything i have said above apply to this kind of additive, the thing is that i had different problems related to those.

Case study #1 - FMA had 0,1% acid and FMB didnt, there was some reproducible baseline anomalies just a few minutes after the the gradient start in the chromatogram (due to dwell volume), fixed it by having acid in the same concentration on both sides. Case study #2 - the same thing but it was a chaotropic acid, retention was all over the place. having it on both MPs rendered a very good chromatogram!

bottom line is that you could experience that while handling a salt and potentially solve it by having some quantity on both sides. but once more, it is very very VERY IMPORTANT to confirm both sides can handle the amount of salt you are using in order to avoid salt precipitation on your system!!

With applications I have done, we've only added to the aqueous portion and didn't see much of a change in the chromatogram compared to methods we were basing our analysis on. If you are concerned about changes in the gradient between the A and B solvents many papers/methods have kept the buffer in the C solvent and that is always at 5% or so. The gradient is then run between A and B with C constant and it keeps the same ionic strength across the gradient.

From the perspective of reverse phase, if your method actually cares whether it's running in 0.1% formic acid or 0.07% formic acid, you're probably inviting problems of reproducibility and robustness. Reverse phase nearly always works fine with additive just in the aqueous solvent.
Hilic is totally weird. All sorts of mechanisms are going on, before you even start thinking about zic-Hilics. It's a real pity that Andy Alpert isn't around to comment. What I would say is that if you have a Hilic method that works, don't try to change it to make it better just because theoretically it feels like it ought to be better to make the phases different to what they currently are. By all means do this as an experiment to see what difference it makes, but don't go into it expecting a logical improvement. Your method may be working for reasons that you don't yet appreciate!

From the perspective of reverse phase, if your method actually cares whether it's running in 0.1% formic acid or 0.07% formic acid, you're probably inviting problems of reproducibility and robustness.

Great point !!!