is sodium lauryl sulfate the culprit?

[mohan_2008]

We are noticing a significant spike in pressure as soon as we inject a small volume of the vehicle formulation diluted 1:1 with the diluent.

The equilibration pressure was around 150 bar with the mobile phase, and as soon upon injection - it shoots to 210 bar.

Chromatography is unacceptable with significant tailing and peak distortions.

The Vehicle contains - Sodium lauryl sulfate.

Mobile phase: 60% Water: 40% Methanol: 0.1% phosphoric acid (Isocratic mode)
Flowrate: 1.5 mL/min
Column: Zorbax SB C8 4.6X75 mm; 3.5uM
Diluent: same as mobile phase
Column temp: 40C
Wavelength: 220 nm
Run time: 20 minutes
Injection volume: 75 uL
Working concentration of sample: 0.0015 mg/mL

What is the problem with this spike in pressure. Obviously, it is doing something to the chromatography.

1. Is SDS absorbing at this wavelength. If so, is it significant.
SDS is part of the vehicle formulation.
2. Is the flow rate too high that creates foaming issues with the SDS surfactant in the column that leads to this back pressure.

I need your expert advice to address this situation.

Thank you in advance.

[HW Mueller]

Your diluent does not cause any cloudiness (very fine precipitation or aggregate formation), whatever? If not, the injected solution must be considerably more viscous than the mobile phase. Or is your sample cloudy from the start?
Foaming is caused by lowering the surface tension of water and then shaking/mixing with air/gas.
Why do you want to know about light absorption of SDS? There shouldn´t be much, unless you have micelles (doubt that), which can scatter light.

[lmh]

SDS is however a very strong ion pairing reagent, and if you keep injecting significant amounts of it onto a reverse phase column, you may end up with some strange chromatography and retention time drifts.

[mohan_2008]

SDS is injected in small amounts. However, with repeated injections - I am looking at the worst case scenario - assuming a deposited SDS on the stationary phase (as it doesn't go off easily) if it can create shifts in wavelength absorption.

[HW Mueller]

If SDS would get hung up in the column (doubt all of it would stay there) it couldn´t get detected or interfere/interact with any other substance in the detector. One can imagine that large amounts of SDS going through the detector cell with analyte might cause the bathochromic or hypsochromic shifts I mentioned elsewhere. I doubt that you could see any ion pairs. I am talking about concentration below the critical micelle conc. No idea what it would look like if a considerable amount of analyte were enclosed in micelles.

[mohan_2008]

Thanks Mr. Mueller,

I checked with one of the experts (has a significant expertise as related to Ion pair chromatography). He has published several papers on Ion pair chromatography. According to him, it is an extremely arduous task trying to get rid off the SDS from the column - completely!

It is too late that we can re-validate the method, but we are asked to develop a rinse procedure for the SDS from the column.

My question is: Can I use a 1% Phosphoric acid as a part of the rinse to get rid of SDS - a little worried about using 1% since it may hydrolyze the stationary phase (column - Zorbax SB C8 ).

Are you aware of any potential rinse procedures. I appreciate it.

[sfe-co2]

It's generally quite difficult to rid a column of ion-pairing agent. However, you may wish to look into using uric acid. The usual practice is to use as little as possible, and I would always use a guard column too.

Do you always inject 75 uL into the same column? Could explain your observed peak distortion.

Cheers.

[HW Mueller]

There have been some discussions on SDS and RP columns. If I remember there was some conclusion that it is attached reversibly, therefore should be removable. I don´t remember what suggestions were made in this regard, but imagine that some patience is required, certainly a basic medium (maybe to get some exclusion.....) with a good measure of organic modifyer might do the trick.
But, what does SDS sticking to the column have to do with resistance of flow?
I mentioned several times that I have used a mixture of LiDS and dithiothreitol to get rid of very stubborn protein deposits. Don´t recall to have any pressure problems, nor any retention problems of analytes (after the treatment), as a matter of fact I did this to successfully restore the columns after proteins + ? had changed the retention characteristics. Rinsing took a very long time to get rid of the dithiothreiitol absorption (UV).
LiDS was used since its solubility is higher (in H2O, so less precipitation problems) and it appears to have less tendency to precipitate some proteins (some people had warned me regarding precipitation of proteins by SDS, and, sure enough, I have seen it).

[danko]

Could it be the case, that the analyte itself is causing the pressure spikes? The reason for using SDS as a part of the vehicle is probably the solubility properties of the analyte. So, it make me think; if the analyte’s solubility in you mobile phase is poor (without SDS) that might be the explanation of the pressure spike upon injection. One has to keep in mind that the analyte and the solvent (typically) are rapidly separated from each other, as soon as they reach the stationary phase.

Best Regards

[Consumer Products Guy]

Our liquid soaps, body washes, and laundry detergents all contain sodium laureth sulfate, which contains a decent amount of sodium lauryl sulfate as not all of the lauryl alcohol gets ethoxylated. We inject solutions of those every day on RP-18 columns for various assays, seems it's too ionic to be retained. We've done this for decades, don't seem to have any issues with any of our assays.

[sfe-co2]

With a high injection plug (75 uL), one would need to be mindful to the possibility of viscosity mismatches between the injected solvent and the mobile phase, even if the the column is jacketed at 40C using a 60% water-based mobile phase on a 75 mm long column (3.5uM dp).

Is this the first time you run this method? When it shoots to 210 bar, does the pressure remain there? remain there?

[mohan_2008]

This is a method developed by one of our peers.

Yes,

When I rinsed this column with rinse solvents - the pressure falls to about 150 bar. However, with a first sample injection - jacks up to 210 bar and stays there.

It stays at that pressure and doesn't seem to equilibrate to the initial 150 bar before injection. You have a valid question - I am just curious to know what could be the reason?

I believe you are right, sfe. It makes a reasonable assumption of a viscosity mismatch between injected solvent and the mobile phase - which explains a pressure spike upon injection.

We are trying to increase the sample dilution - however, staying within the established linear working range. This dilution should try to lower the viscosity more or less.

[HW Mueller]

Now this is confusing:
"jacks up to 210 bar and stays there."
So not a spike?

and

"It stays at that pressure and doesn't seem to equilibrate to the initial 150 bar before injection."
Does this mean it stays at 210 bar until the next injection, after the first injection produced 210 bar? What does the second injection produce?

[mohan_2008]

I apologize for the confusion.

Initially, when I start running the mobile phase on the column - column equilibration pressure is 150 bar.

This is steady and upon injecting standards (made in the diluent - no SDC)
there is no change in pressure after injection. It stays at 150 bar.

However, with the very first sample injection (with the vehicle SDS diluted 1:1 with the diluent) the pressure rapidly shoots to 210 bar - and just stays there. It doesn't equilibrate back to 150 bar even for several subsequent injections.

Once it is at 210 bar - the following sample injections do not seem to have any effect - and the pressure remains constant at 210 bar.

[HW Mueller]

Just wonder how many times you did this cycle: Injecting sample, pressure goes to 210, stays at 210 for further sample injections, then when column is washed it goes back to 150 bar, injecting sample it goes to 210 again . . . . .
I can understand that this happens once (injected some dirt only with first sample), but repeating this?
Incidentally a spike is a sudden rise and fall of a parameter.