Chromatography Forum

Hi members,

I'm having trouble with fronting peaks on both analytes using a PFPP column. Both peaks front quite a bit. Analytes are ciprofloxacin and enrofloxacin with 15% ACN/85% 2% formic acid isocratic at 0.3ml/min. I'm using a short 50 x 3.2 mm column with 3 um particles. Last week I ran the same conditions with excellent peak shape, with no fronting. Also, I reduced the injection volume by 1/2 to see if it may have been overloading, but still no success. Any ideas?

What else changed during that week?

Temperature? New batch of mobile phase?

Yes, those were the only changes that I am aware of. The column is not thermostated in a column heater. The ambient temperature may have changed, but not by much. Also, a new batch of mobile phase was made, but I assume that it was made similarly.

"Also, a new batch of mobile phase was made, but I assume that it was made similarly."

If peak shape changed when the only other thing that changed was the batch of mobile phase, how good is that assumptiopn likely to be ?

Peter

A lot of times, we considered the column was dying if peaks was splitting under the same condition which gave good peaks. sorry

ym3142, if you have some interesting columns which died just after you created a fresh mobile phase, send them to me. I may be saving a lot of time and agony by not having to convince some meds of the necessity for a column investment.

HI, Hans,
I received your message. But do you mind letting me know the real purpose? I guess you were not intending to receive our dead column since I doublt that our company will allow me to send the company property away though the company may allow me to trash them.
btw, who is MEDS?

Medical doctors, physicians. He works in a hospital.

A lot of times, we considered the column was dying if peaks was splitting under the same condition which gave good peaks. sorry

One of the most common causes of peak splitting that I encounter is due to column exposure to a transitory mobile phase whilst standing or during startup, eg >95% water on some reverse phase columns, or a different pH solution.

The solvent in the method may not solve the problem, but flushing the column with suitable solvent for 15 minutes will usually eliminate the splitting.

If a column is slowly dying under the same conditions, it's usually worth investigating and fixing the cause before it dies. If a column suddenly dies, then resurrection is often possible. Pronouncing death should only occur when all vital signs remain extinguished after appropriate and economic resuscitation attempts .

Bruce Hamilton

ym3142, my guess simply was that the columns that you pronounced as dead are full of vibrant live. This was just a restatement of what Peter indicated. Bruce detailed this a bit more for you.

Peter, Bruce, and Hans,
Thanks for your input.
I was using Luna c18, mobile phase 60 Buffer( 0.96g EDTA/L) and 40 ACN. One of the injections was shouldering suddenly one day. I got nice peak in another column on the same LC(everything was the same). Thus I believe the first column was dead. So what you sugget me to wash the column with ? I tried 100% acetone , 0.5ml/min overnight, which did not help. I tried 90water , 10 ACN 60 min, and 10 water 90 ACN overnight, which did not work. Thanks

I have seen from time to time, a column suddenly develop head space for no obvious reason. The symptom is usually a split peak, with later peaks showing worse than early ones. When this happens, no amount of washing can restore the column. If you have some stationary phase (sacrifice a guard column) you can try to make a paste and force it into the void. If that does not work, your column is finished.

Column voids can cause peak shape distortion (fronting, shouldering, splitting) just like Mark suggested.
Another cause can be a partially blocked inlet frit. Back pressure usually does not change but peak distortion can occur.

best regards,

Peter, Bruce, and Hans,
Thanks for your input.
I was using Luna c18, mobile phase 60 Buffer( 0.96g EDTA/L) and 40 ACN. One of the injections was shouldering suddenly one day. I got nice peak in another column on the same LC(everything was the same). Thus I believe the first column was dead. So what you sugget me to wash the column with ? I tried 100% acetone , 0.5ml/min overnight, which did not help. I tried 90water , 10 ACN 60 min, and 10 water 90 ACN overnight, which did not work. Thanks

I would have initially tried changing the pH, either with 0.1% acetic acid or 0.1% TFA to remove the EDTA and friends first, then I would try the cleaning system recommended by Phenomenex . Presumably, you have a standard technique for removing the EDTA, and would use that before trying to clean the column.

As it was a sudden failure, I would consider either an unusual sample, or the column had encountered unusual conditions.

The normal protocol would be at 20 - 50% of normal flow, then at least 10 column volumes of each of the following:- 95:5 H2O/CH3CN, THF,
95:5 CH3CN:H2O. Then I'd try the either the normal test mixture or the samples again.

There was also a good article by Ron Majors in LC-GC ( I hope the link works ) called The Cleaning and Regeneration of Reverse Phase HPLC columns.
http://www.lcgceurope.com/lcgceurope/da ... rticle.pdf

Obviously, if your employer values your time highly, and columns cheaply, then regular replacement may be the preferred option. That doesn't mean the column was dead, just that your employer uses a different point on the cost/effort graph.

Bruce Hamilton

After many days we find out that you have EDTA on the column. Who knows what else you did besides using a new batch of mobile phase just before the column acted up. Maybe the 2nd column had a liquid phase in it which is compatible (no precip.) with your mobile phase, wheras the first column´s content was not compatible (maybe something precipitated out of the mobile phase on contact with the liquid in the column). You don´t tell us what happens if you wash backwards or what happens if you run the analysis backwards, etc. etc. etc. If you are not willing to formulate the problem precisely or are not approaching this systematically I can only repeat what I told Kati: Lifelong project....