Trouble resolving peaks

[TimB]

Hi Chromatography Forum,

Its been a while since I posted anything here, so hope everyone is having a good start to the New Year.

Below is a chromatogram and I'm trying to resolve the two peaks that are eluting, but I am encountering some difficulty:



The method conditions are as follows:
Mobile phase: 0.1% phosphoric acid in water (100%)
Flow: 1.2 mL/min
Isocratic for 20 min
Column: Waters Atlantis T3 250 x 4.6 (5um particle size)
Sample is dissolved in deionized water
UV detection, 190 nm

I have tried running a gradient with acetonitrile to no avail, I'm not sure if I am starting the gradient too late or not soon enough or if I shouldn't use ACN at all.

I'm kind of under pressure to figure this one out, so any help would be appreciated. Thanks,

Tim.

[willnatalie]

I'm assuming automated injection, and what is the size of injection and/or the concentration of the analyte?

[TimB]

I'm assuming automated injection, and what is the size of injection and/or the concentration of the analyte?

Forgot that bit of info. 5.0 uL injection and the sample size is 1.0 mg/mL.

[JGK]

What happens if you prepare the sample in the mobile phase and inject that?

do you get a similar chromatogram?

With a wavelength of 190 nm the analyte mustn't have much of a chromophore. Could you not try somenthing like ELS detection instead?

[TimB]

What happens if you prepare the sample in the mobile phase and inject that?

do you get a similar chromatogram?

With a wavelength of 190 nm the analyte mustn't have much of a chromophore. Could you not try somenthing like ELS detection instead?

You're correct, we don't have that much of a chromophore, and unfortunately UV is all I have to work with. (They've seriously squashed budgets and the hopes of getting an ELSD are zilch at this point).

The analyte in the above chromatogram is a carbodiimide, which converts to a urea compound in acidic conditions. I posted about a method I was working to improve a while back for analysis of that Urea product. (worked beautifully) This is essentially the opposite. The more I delved into it, the more I think the acidic mobile phase won't work.

Perhaps these two peaks are showing that conversion? I do believe the peaks at 2.1 and 4.0 are from the carbodiimide and the peak at about 7min is the hydrolized form (the urea).

So I don't think it would make sense to dilute the sample in mobile phase.

So now do I change the pH of the mobile phase and go to a dibasic potassium phosphate buffer at a pH of 6.8? What will that do at such a low wavelength?

Thanks.

[Uwe Neue]

First, dissolve the sample in your mobile phase with the phosphoric acid. If nothing is changing, then we should consider a conversion of your analyte during migration through the column. Could be degradation (less likely) or interconversion of stereoisomers.

[JMB]

TimB,

Enter "Fronting, Cause and Remedy" in the search box.

You will find a discussion from Dec. 2005 regarding peak fronting that may well be very similar to yours, i.e. chemical interconversions (and how to modulate them).

Please let us know if temp. UP/DOWN makes a difference. In any case, tell us what works/does not work so that we can learn from you.

Good Luck,

JMB

[HW Mueller]

The idea that the peak at 4 converts to the one at 7 seems the best (otherwise I would have trouble explaining the sharp rise at 4 and the sharp decline at 7). The stuff is stable at pH = 6.8?

[TimB]

Thanks for all the suggestions. I did what Uwe and JGK suggested and dissolved the sample in the mobile phase. I managed to separate the peaks, but could only see the second peak after subsequent injections. This convinced me that all of the carbodiimide was being converted to urea. I went back to my literature and racked my brain even more and thats when I realized two things:

1. In aqueous media at acidic pH (2-4 roughly), carbodiimides will hydrolyze to form the corresponding Urea. However, at neutral pH, carbodiimides are stable in solution. So the use of a 0.1% phosphoric acid mobile phase is completely out.

2. I also noticed that when I managed to separate out the compounds, peak one had a higher lambda max (214 nm as opposed to 190.6).

So after changing the wavelength, and changing the mobile phase to water/acn, I got a nice peak at about 3 min. I don't like that the peak elutes so soon, but I managed to run a few concentrations and get a nice linearity with an r2=0.9999.

Now I just have to play around with the mobile phase composition and solve some peak fronting issues, and I can finally look at this compound. Once again, my deepest gratitude for everyone on this board for the suggestions.

Tim