Chromatography Forum

Hello,

I am currently trying to separate nicotinamide riboside on a Thermo Scientific Hypercarb column with the following conditions: A: .05% ammonium hydroxide in 10 mM ammonium acetate and B: .05% ammonium hydroxide in acetonitrile, a flow rate of .2 mL/min, and a gradient from 5% B to 20% from 2 to 15 minutes followed by 20% to 60% from 15 to 27 minutes. I would like to stick with a pH of 8 or above to avoid conversion of other metabolites of interest to their oxidized forms. The problem is that upon going to a more basic pH the retention time of NR shifted to very high organic percentile (greater than 40%) and the peak shape became intolerable with a width of greater than two minutes and incredible tailing. I have tried varying the salt content without any effect as increasing the eluting strength of solvent B by adding more isopropanol (also with no effect). I know that the pH has greatly affected retention here as I originally separated compounds with a pH of 4 to 4.5. This led to a much early and neater retention time.

Does anyone have advice for an ion pairing agent that may help or general LC advice with this problem?

In some cases, HyperCarb exhibits behavior similar to an anion-exchange material. For example, it's significantly more retentive of phosphopeptides than is C-18 silica. This retention is antagonized by high salt concentrations. To elute a peptide with more than one phosphate, organic solvent alone does not suffice. You also need a significant amount of base. 200 mM NH4OH works pretty well, but 200 mM pyrrolidine is somewhat more effective. In your case, then, try a higher concentration of NH4OH or else use pyrrolidine. Incidentally, none of this behavior is explained by the ionization state of nicotinamide (pKa = 3.3).

your compound is hydrophilic quaternary amine. You can use mixed-mode cation-exchange similar to analysis of these quaternary compounds:
http://www.sielc.com/Application-HPLC-S ... isc-R.html
http://www.sielc.com/Application-Separa ... mines.html

Vlad is correct. I was mistaken about the pKa of this compound; it's 12, not 3.3. That said, it still wouldn't hurt to assess the results of using a much higher concentration of NH4OH in the eluting solvent.

i don't think that it is correct to state pKa of quats as they are always charged, I usualkly call it indefinite pKa >14

Hello,

I appreciate the advice regarding that column but we are trying to stay with the same column because we have acceptable separation of other analytes. I have tried increasing the concentration of NH4OH to 200 mM; however, the signal disappeared at that point on the mass spectrometer I am using. I tried running in both positive and negative ion mode and could not detect NR. I am trying other methods I have located in the literature and should have the results of those by Monday. I really appreciate your interest in my problem and value your insights.

Update: I have essentially given up hope of eluting NR in a reasonable amount of time and with good peak shape in basic conditions and have therefore moved on to a two separation scheme. All but NR will be analyzed with the basic pH and NR, plus some other metabolites that I know I can add to the list will be separated with the other.

I agree wholeheartedly that increasing the salt concentration to 200 mM would likely solve the LC problem. However, I am running in-line with a mass spectrometer. The amount of ion suppression would be far too high and the system would dirty quickly. Thank you all for your advice.