Retention of Pyridine N-Oxides on HPLC

[schlenkline]

I am currently trying to analyze reaction mixture containing a series of compounds, all of which are pyridine N-oxides. However, I am running into trouble getting of these compounds to retain on the current system we have.

I am using an Waters XBridge BEH C18 Column. The test compounds I work with elute with the solvent front at 98% Aqueous Mobile Phase

Mobile Phase A: Water (0.1%/0.05% A-/HA Formate Buffer)
Mobile Phase B: ACN (0.1%/0.05% A-/HA Formate Buffer)

Most of my attempts have been running the system at 0.75 - 1.5 mL/min.

I was wondering if there were any options to gaining retention using this column. I'm trying a round of buffers to see if this effects the system, but I don't suspect it will (no readily labile hydrogens on these compounds).

Will trying a different mobile phases, changing flow rate, temperature, ect really effect retention or should I simply look into getting a new column (and if so, do you have any recommendations?) I'm relatively new to LC method development and been trying to read up as much as I can, but could certainly use some guidence.

I've provided a link for one of the compounds I am looking at below.
http://www.sigmaaldrich.com/catalog/pro ... &region=US

[Vlad Orlovsky]

you will experience poor peak shape and no retention at the conditions you are trying. If you decide to go with RP you will need very low organic and pH above 8. You can also try HILIC instead of RP. I spent sometime analyzing something similar and had to abandon my approach.

[krickos]

Hi

Limited experiance on N-oxides, however som 5-6 years ago we exchanged a GC procedure with a LC procedure where one compund was an N-oxide, there was a very good correlation between the procedures. In our case the N-oxide had a side function that allowed good retention on a "normal RP C18 system".

Perhaps switching to GC instead could be an option, if the target molecules are suitenble for GC injection?

[trammells]

Hi,

I work with similar compounds all the time. I use a Thermo Hypercarb Column with 10 mM Ammonium acetate + .1% formic acid in A and .1% formic acid in ACN in B. Nicotinamide and methyl nicotinamide are retained on the column for quite some time. I would imagine that purine oxides would have similar retention. You can also use basic conditions with some salt and ammonium hydroxide. In my experience with HILIC separations, nicotinamide and nicotinic acid elute fairly early.

[Vlad Orlovsky]

Pyridine oxides behaves nothing like nicotinamide and methylnicotinamide. Nicotinamide is basic in nature, if you are talking about N-methyl nicotinamide it is basic too (in terms of cation-exchange interaction). Pyridine N-oxides are neutral (you can call them zwitter-ionic if you want), more polar and have no way to interact by ion-exchange mechanism.

[carls]

These compounds are very polar so HILIC is likely the best approach. You could try this approach if you currently have a bare silica column. Remember to remove the hexane (if present) by flushing slowly with IPA (10 CV) prior to going to a HILIC mobile phase.

There are many HILIC specific columns available as well (google)

[Vlad Orlovsky]

regular HILIC produces a horrible peak shape for pyridine N-oxides

[carls]

regular HILIC produces a horrible peak shape for pyridine N-oxides

Any ideas as to why? They should be simple polar compounds. What interaction(s) would cause this behavior? Were they well retained?

[Vlad Orlovsky]

we had proprietary N-Oxides and tried a lot of different things, had to give up. I have no idea why peak shape was bad. Only at higher pH (above peak shape was relatively okay.

[schlenkline]

Thanks for the input. Vlad seems to be right, I can get some limited retention with a low organic content in some higher pH buffers. Specifically, I get retention in 1% B with an Ammonium Hydroxide buffer.

Still, the retention is limited to pyridinium N-Oxides containing non-polar substituents. Small, unsubstituted derivatives are not retained at all.