changing peaks

[bobb]

I'm doing RP-HPLC using a C18 packed column. My mobile phases are H2O with phosphoric acid to pH 3.65, and methanol with 5% THF. I've been using the same method for a few months but over time I'm noticing some changes in peak shapes. The injection peak has slowly split over time (the first half is a nice sharp peak, the second half tails off) and the other peaks have been growing more broad. I've tried replacing the guard column various times, with no noticeable change on any given replacement. I've also tried flushing the column in a reversed orientation with 50% methanol, also to no avail. I then tried replacing the main column with a brand new column of the same stationary phase, but the peaks were even more broad (I don't know if the new column is just faulty or what). Any suggestions?

Thanks

[haiedc]

I'm doing RP-HPLC using a C18 packed column. My mobile phases are H2O with phosphoric acid to pH 3.65, and methanol with 5% THF. I've been using the same method for a few months but over time I'm noticing some changes in peak shapes. The injection peak has slowly split over time (the first half is a nice sharp peak, the second half tails off) and the other peaks have been growing more broad. I've tried replacing the guard column various times, with no noticeable change on any given replacement. I've also tried flushing the column in a reversed orientation with 50% methanol, also to no avail. I then tried replacing the main column with a brand new column of the same stationary phase, but the peaks were even more broad (I don't know if the new column is just faulty or what). Any suggestions?

Thanks

Do the retention times change?
Do you use gradient
Could you post the chromatogram so it'll be easier for others to diagnose.

[danko]

Hi Bobb,
In addition to Haiedc’ quite relevant questions, I’d like to turn your attention towards the “bufferâ€

[haiedc]

I concur with Danko about the pH meter. Just recently, we have very big problem of agonist analysis and finally (after nearly a week) it was because of the pH meter. It showed something that was 1 unit higher than it was.

[bobb]

haiedc and danko,

Thanks for your responses.

I'm trying to figure out how to post the chromatograms, I'm really new on this forum and I'm not sure how its done. The retention times have shifted a couple minutes left. I checked the pH meter and it is working correctly. I'm running a gradient starting with 10% MeOH/THF increasing to 15% at 2 minutes, then to 40% at 20 min, 90% at 20.1, then back to 10% at 22.1 to re-equilibrate.

As far as mobile phase preparation I haven't changed my method and all the chemicals seem to be ok.

One thing I didn't mention is that I changed the detector from detecting at 345nm to 365nm, but would that cause the problems I'm seeing? [/quote]

[haiedc]

haiedc and danko,

Thanks for your responses.

I'm trying to figure out how to post the chromatograms, I'm really new on this forum and I'm not sure how its done. The retention times have shifted a couple minutes left. I checked the pH meter and it is working correctly. I'm running a gradient starting with 10% MeOH/THF increasing to 15% at 2 minutes, then to 40% at 20 min, 90% at 20.1, then back to 10% at 22.1 to re-equilibrate.

As far as mobile phase preparation I haven't changed my method and all the chemicals seem to be ok.

One thing I didn't mention is that I changed the detector from detecting at 345nm to 365nm, but would that cause the problems I'm seeing?

[/quote]

My suspicion is the proportional valve. Your pump may not deliver correctly the mobile phase composition. Also leak or contaminated at the sampling valve.

[amyweaver29]

Hello! I am new in the forum and I am glad I come upon this thread. My daughter is trying on this in school. She somehow got the same as Bobb. Now, we are both interested how to get this right.

Looking forward to hear from you all. God bless,

[danko]

Anyweaver29,

Welcome to the forum.

As you can see we’ve been through the general trouble-shooting tools, but the definite solution still lies ahead. The following is applicable for you as well as Bobb:

More details are needed – such as mobile phase composition and preparation protocol. And then sample (analyte) type i.e. ionizable, potentially pKa? Sample solvent and injection volume?
Finally, chromatograms are typically at least the half of the solution.

Best Regards

[tom jupille]

The procedure for displaying chromatograms is here:
viewtopic.php?t=2617

There is a general diagnostic procedure on our web site:
http://www.lcresources.com/resources/TSWiz

In the specific case of a suspected proportioning problem, the diagnostic is the "step test":
http://www.lcresources.com/resources/TSWiz/hs450.htm

[bobb]

Here are the chromatograms. I tried the suggested proportioning valve diagnostic and it checked out ok. I have yet to test the sampling valve. The sample solvent is 50% methanol and the injection volume is 100uL. I tried 50uL also, but it didn't affect the peak shapes. I haven't changed the mobile phase preparation protocol, so I'm assuming it's not the cause.

The small final peak is systemic, the peak between 8 and 12 minutes is of primary interest, it is the glutathione-AFB1 adduct and I'm looking for the pKa but have not yet found it. As you can see there has been a left shift and peak distortion, especially of the 8-12 minute peak and the injection peak.

[danko]

Hi again,

Please don’t take it as a arrogant remark, but â€

[Uwe Neue]

Now that we know what your analyte is, we also know that the pH control is definitely not sufficient. I do not need to find the pKa of the beast, but from the structure I can tell that all ionic activity comes from the glutathione, and I only nneed to look uo the pKa of that entity to have a good guess of what is going on. You definitely and absolutely do not have sufficient pH control, and must adjust the pH to get peaks that look like the things at the end of your chromatogram.

I am also not sure what the composition of the sample solvent is. There could be an issue as well. Please specify. Also specify column dimensions and flow rates. It makes our troubleshooting much easier if we know what we are dealing with. This is why the nature of your analyte with the chromatograms were such an important part of your response.

[HW Mueller]

Since the system peak at the end is the only one that didn´t change, and assuming that not all peaks are pH dependent, I would think that serious changes in mobile phase composition, besides pH instability, are behind this. What is AFB 1? In any case, I concur that your "primary interest" peak is not chromatographing correctly. How did this blob integrate over the "few months"?

[bobb]

Thanks everyone for your help. I took danko's suggestion of making a 20mM potassium phosphate buffer pH 2.4 and substituting the former mobile phase of H2O to pH 3.65 without a salt for buffering with the new buffer. The improvement is very encouraging.

The sample solvent is 50% methanol 50% water. The column is 150mm x 4.6mm with a C18 5u stationary phase. Flow rate is 1ml/min.

The sample is an injection of a biological assay in which Aflatoxin B1 (AFB1) is converted into an epoxide and then conjugated with glutathione. The intermediate epoxide can be either an endo or exo conformation but is preferentially exo. In previous studies when both are present and conjugated with glutathione the 12 min peak can be split, but In this case the 12 min peak looks alot like the last peak (the systemic peak) so I'm wondering if the peak splitting is due to something else? Do I need a stronger buffer than 20mM or a different pH?

HW Mueller pointed out that there might be changes in mobile phase composition ( besides pH instability). In view of the fact that I make these fresh each time before running samples, how would I address such an issue?

[danko]

Good to see an improvement. And yes, I believe there’s a room for a further optimization.

Here; a couple of suggestions:

1. Increase the buffer concentration to 30 – 40 mM keeping the pH constant. Preferably stepwise e.g. 30 and then 40.

2. Secondly (or firstly, but not simultaneously with the buffer concentration increase) reduce the methanol concentration in the sample solution to let’s say 20 %, but be sure the sample is still soluble i.e. beware of analyte precipitation.

And then let’s see the potential additional improvement

Best Regards