Hi all,
I am working on the adsorbtion isotherm for 2-phenyl ethanol (PE) and I use a HPLC with a Zorbax XDB C8 column. The solvents that I use are water and methanol (50:50). The problem is that with every injected volume from the same PE solution, in the same conditions, the retention time variates in a quite big range (6 to 10 seconds). I did some tests without the column and the retention time was constant, so I think the problem is with the column. Did anyone experience the same problem? What should be done in this case?
Thank you,
Catalin

Your pump is broken or your system leaking. Most likely a check valve issue.

The system functions okay without the column, with no variations in pressure. The pump was checked and it is okay.

Thanks,
Catalin

If you are not using premixed mobile phase,maybe your there is valve that's leaking, so the content of methanol in the mobile phase is changing. You won't see this without column, maybe you can use premixed mobile phase.

Definitely an equipment issue, we've had no issues at my lab with similar analytes.

Catalin,

A six-to-ten second shift is not that large within an analysis. Basically, you are stating that your analyte is shifting from RT X min to RT X +/- 0.10-0.16 min.

Is your retention time both increasing and decreasing within your analysis, not just moving in one direction? If the RT is moving up and down within the same run, then I would agree perhaps there is a valve problem if you are using on-line mixing. If it simply moving in one direction, you may have a problem with your column, perhaps it is aging.

I would check the validation report (if there is one) for this method and see what kind of RT variation was seen during development/validation. Perhaps this shift is not large enough to warrant concern.

I hope this helps. Please provide more information and we can give more help.

David

Are you controlling the column temperature?

Hi again,
First of all, thank you for your answers. And now back to the issue...
I have made about 5 consecutive injections for the same solution and the retention times goes down linearly. The difference between the first RT and the last RT is about 0.2 min. At one experiment the RT went up linearly but never up and down within the same consecutive injections.
I am using on-line mixing so I will check the valves again. The column tempreture is controled and constant at 20°C.
Thanks again for the input and I hope to hear from you soon.

Regards,
Catalin

I have seen stronger shifts of this type which were caused by nonequilibrium of the column. The analysis was simply started too early.

some systems takes a long time before they start to give consistent data. I believe aged column takes even more time to equilibrate? Have you seen a change in column performance (tailing factor, plate count)? Just determine RT precision at the end of a long run and compare it to the first 5 injections.

What do you mean when you say "a long run" ? For each injection I allowed a time of 20 minutes to follow the UV response.

Regards,
Catalin

The retention time varies around the value of 15 min and the height and area of the peak is aprox. constant for all experiments. Shouldn't they seriously vary if the column is not in equilibrium?
I forgot to mention also that I am new to the field of Chromatography, so please excuse some stupid questions that I might ask.

Regards,
Catalin

By long run I ment e.g. 30 consecutive injections (20 min each) with the same parameters. just leave it running over night then record the RT's from first 10 injections and compare them to the last 10 inj. This will certainly tell you if it is an equilibration issua. Relative standard deviation is widely used to determine system suitability. 1-2% should be acceptable depending on the method. Also you should follow the trends to be able to identify problems.

It's normal to see repeatable area and height while RT is fluctuating. You still need constant RT because it's usually the only parameter you base your identification on.

By the way do you have a cooler in your column thermostat? Atleast agilent systems has this option. I'm asking this because if you have only a heater in the thermostat and it is set at 20 C you won't actually get 20 C when the room temperature is above it?? With only heating option you should set the temperature well above room temperature (35-40) to get data independent from room temp variability.

Also it's a good practise to:

-purge all lines carefully
-equilibrate column for 30 min with initial solvent composition
-inject 2-3 blanks to ensure stable baseline

before starting anything

Hello Anon #1:

I'm using the XDB-C8 with two or three different gradient methods. One advantage of the column is the constant retention time, even compared to some month old chroms.
If you're running other methods on the same system and get more stable retention, maybe the shift of retention is caused by the method or the compound.

Phenols are sometimes difficult to analyse. They're acidic and will react with methanol, if the chance is given. So using acetonitril instead of methanol and adding a litte acid to the mobile phase (pH around 3 should be okay) may lead to more stable retention.

How do you control the temperature. Ordinary column oven, water bath, block heater?? To maintain T = 20centigrade could be problematic using a column oven if the ambient T would be higher. Van't Hoff would suggest decrease in retention for neutrals with increasing T!!

Retention time changes? What about retention factor?

PE is not a phenol but as the name would suggest an alcohol. It thus should not undergo acid-base reactions in an unbuffered sytem. I guess under strongly alkaline conditions we rather would get a carb-anion than a dissociation of the alcohol group.

Dear Catalin,

when your system's okay and you're running constantly at 20°C there are quiet a lot of possibilities for retention shift.
One assumption (after all I'm running out of time to write more):

- The tube which is connected to the column isn't inside the column cooler/heater and lab temperature can vary a few decrease:
-> When temperature inside the lab rises, the temperature of the mobile phase is rising too. This leads to decreasing retention times even when the column is always at the same temperature.
Which also means that nearly every morning the retention time is the same as the mornings before.

Hi again to all,

First of all, thank you for all your input. We managed to get around this problem by changing to a new YMC C18 column. But only with this we didn't solve the problem, because it seems that phenylethanol has a poor solubility in water. So, we decided also to switch to a 70:30 (methanol:water) ratio. Now it works okay... i think :).

Thank you all,
Catalin

Question to Andreas and others...

I have a similar observation in my lab with my retention time occasionally. I cannot control the temp in my lab (the lab adjacent to me controls it) and therefore the temp is up and down all day! If I have fluctuating RT but still obtaine acceptable assay results, does this cause difficulty for validation purposes? Meaning, my accuracy results are accurate, but the precision in RT may not be. Is this considered acceptable? No luck with me trying to get them to keep the room at a constant temp...also some of my other methods require a very specific column temp...and I can only deviate 1 or 2 degrees. Thanks

If your SOP has no limits for RT derivation you can validate a method, which shows acceptable assay results and "normal" RT fluctuation.
IF the RT fluctuation was caused by changing ambient temperature mention this in the validation report.
Normally RT drift is rising with the time a sequence is running (without a termostat used). With an over night run it is possible that RT goes up and down more than 5%, because ambient temperature may vary 5 to 10 °C in a lab without extensive solar radiation.