new thread on pH inside a sulfonated column

[Bill Tindall]

Chris stated in an earlier post that the pH inside the stationary phase of a sulfonated styrene (protonated state) column was 0 because the "concentration" available acid groups was high(a number was quoted but I dare not risk leaving this note to find it for I may not find my way back and I hate retyping). For purposes of furthering this discussion I would not approach this problem from the concept of pH as a result of an available hydrogen ion concentration in a solvent. These are aqueous acid/base chemistry concepts and it is always perilous extrapolating these concepts to a partially aqueous or nonaqueous situation.

In the context of absorbing a basic species/analyte into the stationary phase and wondering what will happen I would think of the stationary phase as a solvent with some acidic properties( the solvent/stationary phase has some pKa). Only a tiny fraction of the hydrogen ions in this resin will be dissociated. It will not be the fraction of dissociated hydrogen ions that determines the fate of a base in the resin, nor the total amount of available hydrogen ions in the resin. Rather, it will be the relative acidity of the analyte and the resin "solvent" that determines what happens. And it will not be the relative acidity of these species in water, but in the resin matrix, that is important.

Think of the case of sulfuric acid as a solvent. There is almost no concentration of hydrogen ions in this solvent. One does not consider how acidic this solvent is by considering how many available hydrogen ions are in a liter or kg of sulfuric acid, or the concentration of dissociated hydrogen ions. Rather its acidity is determined by what this solvent can protonate(which is almost anything!). Same for acetic acid as a solvent (which probably has about the same hydrogen ion concentration as sulfuric acid, but can't protonate as weak a basic species as sulfuric acid.)

I don't know enough about the resin to speculate further. Chris, does the resin contain a lot of water? If so its dielectric may be high enough that ion pairing is insignificant. In such a case a Hammet function can be used to determine the acidity of the resin . One would absorb a test base into the resin and spectrophotometrically determine the ratio of acid and base from of test base. From this ratio and pKb of test base the acidity of the solvent is determined.

Without doing the experiment I have no idea what the acidity of the resin might be. (It probably is known because similar resins are used as a proton source in organic synthesis.) It could be significantly more or less than a 1 M solution of strong acid in water. I would bet that it is quite a bit less, but that is mostly a guess.

[Chris Pohl]

Bill,

First off, let me say that I have tremendous respect for your understanding of pH and the use of buffers as it applies to chromatography. So it makes me a bit uneasy being at odds with you regarding this topic but let me try to address a few of the points you raised:

1). I agree totally that one should think about the cation exchange resin used in ion exclusion as a homogeneous solvent rather than thinking in terms of fixed sites and mobile cations. Enhanced retention of certain anionic species such as benzoic acid can only be explained in terms of this sort of "solubility parameter" thinking. Unfortunately, there is no tabulated reference data on solubility parameters for ionic polyelectrolytes so one can only work from inference using inductive reasoning when pursuing this line of analysis.

2). I'm not sure I understand why you would predict only a tiny percentage of the functional groups would be ionized. This, after all, is a function of several critical parameters such as the dissociation constant of polystyrene sulfonic acid and the activity coefficient of the stationary phase. It is true that the general observation for a given anionic species is that the apparent pKa range of the polyelectrolyte is significantly broadened when compared with the pKa of the monomeric species. Generally for an anionic species the polyelectrolyte will exhibit a pKa starting with roughly 1.5 pH units below that of the monomeric species and extending roughly 3 pH units above that of the monomeric species, but one can not come to any judgment regarding the effective pH in the stationary phase without knowing the pKa of the monomeric species. Unfortunately, while pKa data is available for some inorganic species with negative pKa values, there is almost no data which can be trusted for organic acids of this sort. Thus, one cannot have a solid basis for making a prediction by pKa alone. Nonetheless, my earlier prediction of a pH value around zero was based on the following generalization: one can estimate the pKa of a organic analogue of an inorganic polyprotic acid by simply assuming that the organic substituent replaces the highest pKa ionizable group on the inorganic polyprotic acid. For example, consider the pKa values for phosphoric acid: 2.15, 7.20 and 12.38. The pKa values for phenylphosphonic acid are: 1.83 and 7.07. In the case of both pKa values they are fairly close to that of phosphoric acid but a bit lower due to the inductive effect of the benzene ring. If I use the same method to estimate the pKa of styrene sulfonic acid, I predict the pKa to be around -3 matching that of the first pKa of sulfuric acid. If the pKa of styrene sulfonic acid is really as low as this, then most of the functional groups within the resin should be fully dissociated.

Under these circumstances, I think that perhaps the best approach is to yield to experimental data. Since one cannot readily perform pH measurements inside in ion exchange resin particle, I did the next best thing and measured the pH of 30% polystyrene sulfonic acid. I had to recalibrate my pH meter to work at such a low pH but after recalibrating at pH 1, I got a measured value for this polymeric acid solution of pH 0.50. The effective functional group concentration in this polymer is somewhat different from that in ion exclusion resin (the water content in our model polymer solution is 70% whereas a typical ion exclusion resin water content is around 50%). Making allowances for this difference, one would expect the measured pH for ion exclusion resin to fall in the range of around 0.25. While this isn't precisely what I predicted in my earlier post on this topic, I think it's reasonable supporting evidence for my contention that one should consider such resins to be effectively a very concentrated acid solution (with a pH around 0) within which the most carboxylic acids are fully protonated.

3). The polarity of the environment is most certainly too polar to be conducive to the formation of ion pairs, so the Hammet function could be used to determine the acidity of the resin. However, I'm not so sure that one can use either anionic or cationic species as the test probes in this case as electrostatic effects may well compromise the results.

[Bill Tindall]

Your view of the resin situation is correct based on its water content. My response was based on the flawed premis that the resin consisted of much less water than what you reported in your response. Indeed, at 50% water this system can be thought of, to a first approximation, as a solution of very stong acid in water, which will result in substantial hydrogen ion dissociation and low pH as you originally stated.

Only at low water level would the issues I raised be relevant. But in either case the resin is a powerful hydrogen ion donor and would therefore protonate anything of even modest basicity, including water. (which no doubt is why it is used in organic synthesis as a proton donor)

[HW Mueller]

Bill, what´s the evidence for low dissociation of H2SO4? Near zero conductivity, like H2O?
Anyway, I was thinking it wouldn´t make too much difference whether the sulfonates are dissociated or not, as hydrogen ion exchange (or other ions) would be extremely rapid anyway (as it is in pure H2O or H2SO4)

Chris,
wouldn´t benzoic be retained as benzoic not benzoate?

[Chris Pohl]

Hans,

You are correct. The species which permeates the ion exclusion resin would be benzoic acid regardless of whether or not the analyte injected was benzoic acid or a benzoate salt. My point in mentioning that specific analyte is that it is retained well beyond the total permeation volume which implies an enhancement of the retention factor based on a favorable solubility parameter in the stationary phase.

On the other question you posed for Bill, I believe he was referring to the ionization state of "anhydrous" sulfuric acid. In that case, it is essentially neutral in spite of the extremely low pKa of sulfuric acid.

[Bill Tindall]

Hans, yes even in nonaqueous solvent the resin is a proton donor and it will protonate anything that is a stronger base than bisulfate. In its soggy wet state (I had no idea these resins sopped up so much water. No wonder columns packed with the resin don't like solvent changes) it will most likely be the H3O+ that does the protonating.

Cris is precisely correct on the sulfuric acid. Pure sulfuric acid has only itself to protonate. So if a sulfuric acid molecule protonates another, bisulfate is formed. Because acid (H+ solvated) equals conjugate base (bisulfate) that is the definition of neutral in any solvent. I don't know what this dissociation constant is(autoprotolysis constant), but it will not be large. It is 10e-14 for water and 10e-14.5 for glacial acetic acid. Yes, it would be measured by conductivity and I am sure it has been.

[HW Mueller]

Actually my question was stated too primitively. I was thinking along the line of wether H2SO4 existed as H+SO4- (ionized), or as HSO4- & H3SO4+ (dissociated), or mainly as H2SO4. But this is really an academic question as H+ certainly exchanges very fast, so that one would expect the sulfonate resins to exchange very fast also, no matter what their physical state looked like.