Chromatography Forum

We are considering running a polar compound with a carboxylic acid function at about pH 7 with a phosphate buffer containing a tetrabutyl ammonium ion pair reagent. Is it possible that the phosphate counterion (Na+, K+ or NH4+) can have any bearing (and how might they be different?) on retention by associating with our compound? We'd be injecting about 0.2-0.5 µmol.

I wouldn't worry that much about the phosphate coutnerion associating with your compound (I wouldn't expect them to have to high of affinity for your compound). However, the phosphate itself will compete with your compound of interest for the available tetrabutyl ammonium ion pair reagent sites. And here is where it get's tricky because the highest the concentration of phosphate, the highest the concentration of ion-pairing reagent that is going to be adsorbed in your column with the potential of higher retention for your compound.

Anyway from an emperical point of view, I saw that I was getting better results (retention/separation) without any buffer ions in it, but there are a lot of points that can be made here about if you should use or not (but if you need to have your pH at pH 7 then you should buffer with phosphate.

For my applications I was either going pretty acid with perfluorocarboxylic acids or pretty basic with alkylamines (hexylamine, heptylamine etc.) and high pH resistant columns.

Hi Kostas,
Thanks for posting.

From your reply, however, there are a couple of things I don't seem to be grasping. Firstly, if there is going to be some sort of competition between the phosphate and analyte anions for the positively charged tetrabutylammonium adsorbed on the surface, isn't there a similar competition between the phosphate-counterion and the ion-pair cation reagent for my carboxy anion? Surely if my compound can associate with tetrabutylammonium on the surface it can associate with ammonium in the MP, for example. Or is it that there is not much difference in the affinities of the given cations to make a difference in retention..

Secondly, I didn't understand your third sentence which suggested the higher my phosphate buffer concentration, the more tetrabutylammonium gets adsorbed on the surface. Why is that?

Regards

JA,

I will first answer your second question:
The amount of your ion pairing reagent fixed in your stationary phase will depend on the amount of salt (counter ion) in your mobile phase. I will suggest to read the Bartha et al article in J. Chromatogr. 291, 1984, 91-102 Studies in reversed-phase ion-pair chromatography III. The effect of counter ion concentration which is a great article which will answer all the question you post here.

There are actually several studies that showed from adsorption isotherms of the pairing ion that in the presence of extra salt, a higher adsorption of the pairing ion is aboserved at the same ion pairing reagent concentration. The authors of the above article showed that the adsorption is linearly related to the counter ion cocnetnration and virtually indepepndent of the ionic strength.

Now about your first question, ion-pairing occured in the mobile phase is insignificant when compared with the one in the stationary phase (i.e. dynamic ion exchange. For example if you add different concentrations of TFA and you keep the concentration of a stronger ion pairing reagent like tridecafluorooctanoic acid constant the retention time of your analytes (assuming the same pH) will be about the same. In other words, your analytes retention will be almost be affected from the concentration of your stronger ion pairing reagent than your much weaker ion pairing reagent.

If I remember well from the article of Patthy in J. Chromatogr A 1994, 660, 17-23 (Explanation of the selectivity differences between reversed-phase ion pair chromatographic systems containing trifluoroacetate or heptafluorobutyrate as pairing ion), when you have already a dynamic ion exchange mechanism going on in your system, the contribution of the other mechanism in the chromatographic retention time is well below 20% (the 20% is for the same ion pairing reagent effect in the retention other than dynamic ion exchange, I would expect that mineral ions of the same charge as your ion pairing reagent won't have almost any effect)

Wow, I managed to make a whole paragraph one sentence . Anyway, is the above clear enough?

Kostas, thanks, your reply is perfectly clear. I'm sure the two articles you referenced would satisfy my immediate interest, the only thing is they are relatively old and I don't have immediate access.

Therefore having not read read the references, I am in a way trying to pursue a layman's explanation.

Despite you stating that retention will be about the same as long as the concentration of the stronger ion pairing reagent stays the same, my brain doesn't comprehend why having an excess of some other, very similar, cation doesn't diminish the interaction between compound & strong IPR. Taking 5 mM TBA-phosphate and 20 mM dibasic ammonium phosphate, for example, I've got an 8-fold excess of ammonium over tetrabutylammonium. Doesn't my compound therefore spend at least 8 times longer associated with ammonium in the MP and hence not being partitioned onto the surface when paired with TBA? Am I to completely disregard the interaction between my compound and ammonium/TBA in the MP because the concentrations are just far too low in the eluent stream?

If anyone else gets it, to save Kostas getting mad at me please refer to my second sentence

JA,

Your question is perfectly eligible . Here are some more things that are not considered in your assumptions:

1) It may take hours for your system to be equilibrated. What happens during equilibration? Your ion-pairing reagent is dynamically modifying your C18 column so in reality you might have much more ion-pairing reagent to interact with your analyte than ammonium (how about 100 fold more ion pairing reagent than your ammonium anions?).

2) You have to consider the "association constants" (don't ask me if I have any values for this).

3) If ever you have ammonium in excess, depending on your pH, this might not be fully ionized, unlike your ion pairing reagent that is always ionized and ready to interact with your analytes

Finally, if you contact me by e-mail I could send you the electronic version of the articles...

........and that´s why some people spend a lot of time to create columns which obviate ion pairing to a large extend.
(Kostas, please don´t take that as a call for a ban on TFA)

Hans,

I think that Vlad from SIELC would like to shake hands with you and concratulate you, for your off-topic message .

But as you mention it, I feel like I have to reply, that mixed columns is a good idea but I have personally haven't use them yet so I can not talk from personal experience.

Now, if you have a good understanding of how ion-pairing chromatography works I think that you can develop nice methods with this mode of chromatography.

I will also post a link in one of the previous posts where Vlad was asking (in a pretty strong way, so his message was sensored by the administrator) why people are still using ion-pairing chromatography which should have been obsolete by now...

I replied at that one with some of my thoughts so here it is:

http://www.sepsci.com/chromforum/viewto ... =2273#2273

PS: I hope I am not opening a can of worms...

Kostas,

Don't worry I will keep my mouse shut!...I am just a bit surprised that you did not try it yet-as I can see it from your resume/ publication that you are an expert.

regards,

Vlad

Vlad,

I really do not have anything against your columns and I would love to work with them in the first opportunity. It is just that what I am working on right now (separation of complex mixtures of peptides and proteins) can not really benefit from your columns (i.e. we need > 300 A pore size and some other reasons that have nothing to do with the column capabilities).

I will probably start again with small molecule analysis in the future, and at this point I would probably start evaluating your columns as well.

Regards,

Kostas

Kostas,

We have columns for peptides and proteins too (300A and up). It is our new Promix line of columns. Check our website if you have time.

We did a few methods for our customers for peptides 3-50KD and fished out a lot of peaks buy using multi mode approach on one column.

here is the link for comaprison of Promix and Vydac columns:
http://sielc.com/Technology_Peptides_Al ... ivity.html
http://sielc.com/Technology_Peptides_Promix.html
http://www.sielc.com/Technology_Peptide ... Types.html

Whoa, Kostas, thought you were long enough with us to know that I was certainly not thinking of mixed up mode, but rather NP like the old TLC guys did, now called HILIC, or "Aqu. " type columns.
I must admit, though, that I also have not tried the mixed ones..... it´s difficult to justify spending money for "playing" with columns in my position, and my offer to test such was not taken up.

Kostas,
In response to your three points made in the post of November 3rd:

1) When the system is equilibrated, I presume this means that the rate of adsoption & desorption of IPR onto the stat. phase is constant. So, for a unit volume along the length of the column I have the same number of TBA ions on the surface as in the eluent above it? If I assume that ammonium doesn't adsorb then does it follow that I can work out the number of ammonium ions in the eluent for this unit column volume from the concentration in my MP. I could try to do the same for TBA in my MP but I have no idea what contribution the TBA equilibrium makes to this number. Am I going off on a complete tangent here?

2) Would it be fair to assume the association constant for ammonium > TBA, due to greater charge density (smaller ion and not 'diluted' by the electron-donating character of alkyl groups). I however, also have no numbers.

3) Our aqueous portion is at pH 7. I think I recall reading that the pKa of nitrogen bases can decrease when you add organic modifier so perhaps the ammonium isn't fully ionised in the MP and therefore not very good at forming an ion pair

JA

In regard to your questions to Kostas, let me take a stab at answering them:

1). The interaction between the analyte and the ion pairing reagent at the stationary phase surface is largely electrostatic. The higher the concentration of ion pair reagent at the surface, the greater the retention of the counter ion. Conversely, the more the counter ion allows collapse of the double layer, the higher the concentration of ion pair reagent at the stationary phase surface. However, a co-ion will have a minimal effect on the surface concentration of an identically charged ion pair reagent. Since the concentration of ammonium ion at the stationary phase surface will be virtually zero (because it has virtually no affinity for the stationary phase), it will play no role in the retention process. Theoretically, increasing the ionic strength could have a slight effect on the amount of the amount of adsorbed ion pair reagent but as Kostas pointed out, this effect is small. Since the amount of adsorbed ammonium ion will be negligible, the concentration in the mobile phase will be uneffected by the presence of the stationary phase, even transiently.

2). I think a better description would be that the association constant for either ammonium ion or tetrabutylammonium ion would be negligible in the mobile phase unless conditions are sufficient to induce ion pair formation in which case the tetrabutylammonium ion would have the higher association constant. In no case will the ammonium ion have a meaningfully higher association constant.

3). Ammonium ion is incapable of forming ion pairs under the conditions of reversed phase chromatography.

JA,

Chris covered it pretty well. Just want to add a couple of things,

1) You seem to think that after equilibration you will have the same amount of ion-pairing reagent in the stationary phase as in the mobile phase. This is not true, actually you will have much higher amounts in the stationary phase than in the mobile phase. Only for molecules that are virtually unretained in the stationary phase (such as the ammonium ions), you can have the same amount in the mobile phase as in the stationary phase...

2) In case that this is the first time that you hear about the (electrical) double layer (from Chris post), you might find more from the article from Liu and Cantwell, Anal Chem 1991, 63, 2032.

Hans,

I apologize for not getting what you were implying. However, your post said: "that´s why some people spend a lot of time to create columns which obviate ion pairing to a large extend". Normal phase has been there for ever and solubility issues might arise so I didn't think of it. Furthermore, I would think that HILIC or aqua type columns is not a very good alternative for ion pairing chromatography. There is only one column that is really advertized as a good alternative of ion-pairing chromatography, so that is why I made that connection....

Vlad,

These columns must be new as I didn't see them last time I checked.

The other reason I mentioned (the one that has nothing to do with your column capabilities) is that we have developed a predictive model for peptide retention time prediction (see Petritis et al. Anal Chem. 2003, 75, 1039-1048) that directly or indirectly we use for our proteomic efforts. The model works pretty well for reversed phase columns peptide separations where TFA/formic acid/acetic acid has been used as additive. So for that part we are kind of stuck on this type of columns (as your columns will give alternative selectivities which will make the above model "a priori" not usable).

However, I am about to start a study (in a couple of months) that I could probably make use of your column, so at that moment I might contact you.

Unlike maybe Hans, I always like (or I am still in position to) playing around with new columns, but I learned to try to minimize the time between getting the columns and using them for something. As there is nothing for free, when people gives you a column they seek in general for somekind of feedback.

In general, if the person is in the sales, this feedback is:

a) if you are university or national laboratory for something exciting that they can use for their catalogues, and ultimately if you can publish something
b) if you are industry to order a massive amount of their column

If within 6 months you do not provide something (and it happened once to me, but it was a situation out of my hands) some people can get frustrated.

If the person is in the R&D, he/she probably forgets about it, and somehow extremetly happy if ever something nice comes out of it...