Continuous Retention Time Decrease

[crillyr]

Hi. I'm having a problem with the RT of my peak decreasing gradually but continuously. I suspect phase collapse (or dewetting) or something similar but any research that I have done indicates that my mobile phase has a significant proportion of organic to avoid this phenomenon (~23% Methanol:~77% pH 2.5 buffer + 0.5% Acetic Acid).

My peak of interest is polar and, therefore, an ion-pairing reagent (1-octanesufonic acid sodium salt) is utilized. I realize the requirement for long column-mobile phase equilibration, but the retention time never settles and the column has mobile phase flowing through it all the time.

Column Temp: 40oC

Any help/suggestions would be great!

[Gerhard Kratz]

What is the pH of your buffer after you added Acetic acid 0,5%? And what is the pH value of your buffer with Acetic acid at 40°C? I guess it is below pH2! In that case you Strip of the Alkyl chains and you get more and more free -SiOH Groups on the surface. In Addition your ion pair reagent is doing a continous uncontrolled bonding on the surface. My recommendation is to use a more pH stable packing material and go to pH8-pH10. Beside pH stable silica based packing materials there are polymer based C18 columns available on the market which are sometimes helpfull. And try to skip the ion Pairing reagent. Good luck

[crillyr]

Thanks for the reply.

My description of the mobile phase may have lacked some clarity. In 1005ml total mobile phase volume, 770ml of phosphate buffer (containing ion-pairing reagent) is prepared to pH 2.5 to which 230ml of methanol and 5ml of glacial acetic acid is added. The buffer should keep the mobile phase pH quite close to 2.5. I would certainly not think it would be capable of moving it below pH2, however, the column I am using is trifunctionally aklyl-bonded and pH stable between 1-12 (apparently anyway).

I appreciate you taking the time to reply but I think I'm still searching for the answer...anyone?

[Gerhard Kratz]

No matter if a n-Octadecyltrichlorosilane or a n-Octadecyltrimethoxysilane is used for the bonding, statistically not all bonding will be on the silica surface. and pH stability is gained mostly by Special trimethylsilylating agents like Hexamethyldisilazane mixed with Trimethylchlorosilane.
Anyhow, is your mobile Phase heated to 40°C or do you use mobile Phase at room temperature? What is your flow rate and column Dimension? Did you ever messured pH in your mobile Phase at 40°C, for example with a lakmus paper?
If your column is in a column Thermostate at 40°C and your mobile Phase is at room temperature, estimated you use a Standard column Dimension of 4,6 x 250 and flow rate 1ml/min, you run a temperature Gradient, but more influence will have the pH Gradient, since pH is related to temperature!

[crillyr]

I measured the pH of the MP at 40oC and it's pH2.84. Again, I would suggest that this is irrelevant to the issue I have as the flow is constant from the mobile phase bottle through column to the detector, as is any temperature changes and, thus, minor pH changes. All injections uniformly go through this process.

The problem I am having appears to be related to column degradation. Over the course of 100 injections the retention time decreases by ~20 minutes and continues to decrease in subsequent injections to a lesser extent. I am searching for a reason for this degree of degradation which I have never before experienced, especially using conditions that do not appear to be out-of-the-ordinary?

Thanks

[HPLCaddict]

Just to be sure that it's the column which is changing and not the mobile phase: If you prepare a fresh batch of mobile phase, do the retention times stay the same (decreased) or do they return to the initial value?
The description of the column you're using sounds like an XBridge RP18, which should laugh about pH ~2.8. It should be absolutely stable at that pH, no matter if at 40°C or not.
Your analyte is basic in nature, correct?

[crillyr]

The mobile phase is changed regularly, as you would imagine, and the RT remains similar to the previous MP.

The analyte is actually a carboxylic acid

[HPLCaddict]

Your analyte is a carboxylic acid??? Then your mobile phase is generally not suited. No wonder you get weird results.

You should either
- work at a low pH (below the acid's pkA) WITHOUT any ion-pairing additive. Under these circumstances your acid should be protonated (i.e. neutral) and show decent retention under plain vanilla RP conditions.
- if the acid is too acidic (i.e. pkA too low) or if it's still too polar even in the protonated form, you might use an ion-pairing reagent to get retention. But under these circumstances your acid will be in the anionic form (since your mobile phase is above the acids pkA), so you need a CATIONIC ion-pairing reagent. Reagents of the sulfonic acid type are just not suited here. You should use one of the tetraalkylammonium type, e.g. tetrabutylammonium hydrogensulfate.

I'd prefer the method without ion-pairing, if possible

[pokemon]

Is it decreasing for all application or for particular one application?

[crillyr]

I appreciate your response but the method is from a highly respected and accepted source. The pka of the analyte is ~5.5, with the pH of the mobile phase low enough to ensure full protonation. Also, the ion-pair reagent is suitable, therefore, for cation analysis under these conditions.

[crillyr]

All associated peaks move with the same relativity to the analyte peak

[HPLCaddict]

I appreciate your response but the method is from a highly respected and accepted source. The pka of the analyte is ~5.5, with the pH of the mobile phase low enough to ensure full protonation. Also, the ion-pair reagent is suitable, therefore, for cation analysis under these conditions.

If your analyte is acidic with a pkA of ~5.5, it will be fully protonated and thus neutral at a pH of 2.8. It is NEUTRAL! Not cationic. So, what do you need an anionic ion-pairing reagent for?
Or does your analyte contain any additional basic groups, that are protonated under these conditions? For a purely acidic compound an anionic ion-pairing reagent like that sulfonate is quite senseless.
Did you ask your highly respected and accepted source for troubleshooting tips? Or does this refer to a pharmacopoeia ?

[crillyr]

The compound is indeed amphoteric containing both a tertiary amine and carboxylic acid portion.

You are also correct in assuming the source of the method...No chance of help!!!

[HPLCaddict]

Ah, OK.
Pharmacopoeial methods can be quite...innovative.
Can you disclose which compound and which method we're talking about?

[crillyr]

Unfortunately not friend. A sackable offense.

Ever heard of this problem I am having before?