Somebody has some information about a HPLC method for the assay of cetylpyridinium chloride which does not use ion pairing agents?

Does pH influence its retention (quaternary pyridinium salt)?

There is an example chromatogram on page 284 of Ion Chromatography by Joachim Weiss (second English edition) of a typical commercial material containing quaternary alkyl pyridinium cationic species. The column used was an IonPac NS1 from Dionex (a polymeric reversed phase column). The mobile phase was 76% acetonitrile with 20mM HCl (if you are using a stainless steel instrument I would recommend using methanesulfonic acid at the same concentration instead). The flow rate was 1 ml per minute.

In answer to your other question, pH should have no direct effect on the retention of a quaternary species.

A small correction to Chris's post: if you are using a silica-based reversed-phase column, the retention of the quaternary amine will increase with pH due to the involvement of silanols.

We do it routinely: XTerra RP18, 10 mM KH2PO4, Gradient ACN/buffer 50 to 80% ACN in 10 minutes. For isocratic method 60% ACN is a good starting point. Detection: 254 nm.
We tried several C18 columns for this compound, but only XTerra gives acceptable tailing in this buffer (in 60% ACN assymetry is 3,80, less than 1,5in gradient).
I cannot use high pH due to matrix composition.

Uwe, we see extreme effects of the type you mention with positively charged metal complexes (with Atlantis, Prodigy.....).

I am not yet sure about this: Negatively charged complexes seem to "see" positive ions at a pH of ~2. Do silanols get protonated (or something else)?
Darn, I have to ask again, any refs?

I vaguely remember an old publication that claimed that silica bcomes positively charged at very acidic pH (below 2), but I don't recall the reference, nor any details (10+ years after reading it, this is not a surprise).

Check this link. It decribes the retention of quaternary amine on Primesep B column. You do not have to use ELSD detector because your cetylpyridinium chloride is UV active.

http://allsep.com/makeChr.php?chr=Chr_043

Main problem with HPLC of quaternary amines is their strong interaction with residual silanols. Simple approach which solves this problem without having complex mobile phase is to use Primesep B column which has embedded amino groups (shielding silanol) along with hydrophobic alkyl chain. Mobile phase of MeCN/H2O/TFA allows to obtain symmetrical peaks for hydrophobic quaternary amines such as cetylpyridine. See example of tetrabutylammonium retention at this conditions: http://allsep.com/makeChr.php?chr=Chr_043

On modern stationary phases, there is no need to panick about an increase in the retention of a quaternary amine with pH. The increase in retention due to ion-exchange is rather moderate on Symmetry (pK around 7) and non-existant with XTerra (pK in the alkaline range, around 10, and a virtually non-existing population of silanols).

My suggestion would be to use a high-quality C18, with pH control, and run a water acetonitrile gradient. Depending on your detector, you may use a phosphate buffer at pH 2.5 or formic acid (around the same pH). You will have no difficulty finding the peak. If you need an isocratic method, you may develop it from there.

High quality C18 seems does not solve all problems of analyzing quaternary amines.
Just compare two chromatograms.
http://allsep.com/makeCmp.php?cmp=Cmp_110

It is usually not difficult to generate bad chromatograms on competitive materials.

We admit that we know our products much better then any other columns, and it might be that conditions we used for our comparison were not absolutely perfect for other columns we have tried.
To get the fair comparison picture, we'll be happy to post a similar separation of
Quaternary amines on Waters column performed by Waters
scientist (or any other brand by any scientist).
Tell us what you think of the idea.

As in my previous post: this here is to both Uwe and Zelechonok. The goal of this forum is to give HPLC users advice on how to solve problems, and not to advertise.

Thank you for your replies! I'm currently validating a HPLC method for lidocaine the other active principle in the formulation. Sample preparation is done in basic alcohol/water solvent + NH4CO3 (for removing mucoadhesive excipient). Eluent is basic acetonitrile-water pH 8 on Symmetry C18 from 20% acetonitrile to 80% acetonitrile - 220 nm. Do you think such a gradient/column is also feasible with cetylpyridinium chloride ?

Reply to vojtech: A tailing factor of 3,8 is outside the specs of our validation protocol. A tailing factor < 2 is a must. Will try your suggestion. I have a XTerra 100 mm column in lab.

I think the gradient method will work fine. You do not need to go to pH 8, and I would not do so since I expect that silica-based column work longer at lower pH. For 220 nm, I would use phosphate at pH 7.

To A.Mouse:
Thank you for your input, but I do not think that stating the pKa's of different materials amounted to advertisements. It was rather good advice on how to approach a method. Otherwise, I agree with your suggestions.
best regards
Uwe