[synthon]

Hi,

I've had difficulty with ion exchange chromatography in the past, and again am having difficulty separating a tertiary amine (PROD) from it's substitution precursor, iminodiacetic acid (IDA).  I'm trying to wrap my head around what is happening on the column so I can match that with what to expect.  I keep getting unexpected results, but I feel like I understand what is supposed to happen pretty well.  I'm hoping someone can point out a misunderstanding or mistake in my procedure.

So, to run the column:

I prepare the gel by passing 6M HCl through to saturate the column with H+ ions and wash any other ions off the gel.  Then, I add the crude mixture at pH 3 and both IDA and PROD stick to the gel, while the negative ions from the rxn and adjusting pH (Cl-, Br-) pass through the column with the eluent. 

At this point, the eluent becomes acidic which I presume is due to the H+ that the amines and + ions (Na+) have displaced, and that the eluent pH returning to neutral signals that all + ions from the crude mixture are fixed to the column.  So, at this point I begin to elute with a NH4+ gradient, which causes the pH of the collected fractions to again become acidic as more H+ are displaced from the column by NH4+.

Soon enough, the collected fractions show a pH of ~9, as does the mobile phase, which I assume means that all H+ have been pushed off of the column.  Since that's the case, my amines (IDA and PROD) should have also been pushed off at this point, since amines bind more weakly to the column than H+, right?

Unfortunately, I can't find my product in the acidic fractions between starting the gradient and column saturation with the eluent.  Am I missing something?

Thanks for the long read!

[synthon]

Details: DOWEX 50X4 gel, H+ form, eluting with ~300 mL 0.1-0.2M NH3 before pH of fractions reaches 9.

[TheUnassuming]

Have you checked the fractions before you started your gradient?  I know in theory it shouldn't be there, but I've definitely had compounds run right through that should have stuck.   
Dowex H+ resin is just an immobilized sulfonic acid, so when you hit a pH of 9 that just means that essentially all of your resin is deprotonated and paired with an ammonium salt (your materials and NH4+).  The inductive effects on the tertiary amine make it more basic than ammonia, so you might have to up the concentration of NH4+ further to force the exchange of NH4 for your product.   
The other option of course is just to run 6N HCl through and get everything back. 
I have never had good luck with trying to separate small molecules by IEC.  But I find its handy for making salts of your compound.

[TheUnassuming]

You could also use a page from IEC of proteins and try to salt it off, though I've never tried that with a small molecule before.

[synthon]

Thanks for the tips.  I should have mentioned that I checked the fractions before the gradient as well, and also that this is was the second column I've run on the product.  For the first column I used 0.5M NH4+ and everything came off in the 4th fraction after the gradient started, so I tried to purify that fraction again using a more dilute gradient.

The main concern I was having was wondering whether the product could still be stuck to the column if the fractions were pH9, or if that meant everything must be washed off by then.  Thanks for explaining that part in detail, specifically.

Luckily, some of the product in the first few pH 9 fractions crystallized out overnight, making it obvious where the product is. 

I'm now running permanganate stains vs ninhydrin to see where the IDA is vs the tertiary product and hopefully there will be at least a couple pure fractions.  If not, I think keeping the gradient at 0.1M in a subsequent column may do the trick.

Thanks again!

[Babcock_Hall]

After adding HCl did you run some water through your column before adding your sample?

[synthon]

After adding HCl did you run some water through your column before adding your sample?

Yes, I've learned that the hard way as well from previous columns.  After washing with HCl, I'll typically run 3-5 CVs of water until the eluting pH is near neutral again.  Then load my sample and wash with water again for a CV or two, then start the gradient.

[Babcock_Hall]

I am having a difficult time understanding your conditions.  Are you eluting with ammonia or ammonium ions?

[synthon]

I'm eluting with ammonia, but since it's the NH4+ ions that are relevant to the elution, I used that notation for brevity.

[Babcock_Hall]

I picture what is happening on the column somewhat differently from you, but perhaps I am still not fully understanding what you did.  After the material of interest is bound and other stuff rinsed through, the ammonia is added IIUC.  Ammonia will react with H⁺ ions bound to the column and the ammonium ions are now bound to the resin.  At sufficient concentration ammonia will also raise the pH of medium sufficiently to deprotonate the material of interest (at least this is true for amino acids).  They will elute and ammonium ions that were produced in this acid-base reaction will take their place on the column.  BTW ammonium ions bind slightly more tightly to Dowex-50 than protons do, IIRC.

You wrote, "So, at this point I begin to elute with a NH4+ gradient, which causes the pH of the collected fractions to again become acidic as more H+ are displaced from the column by NH4+."  I acknowledge that my picture of what is happening does not predict or explain why the fractions again become acidic.

[synthon]

Correct, them relative affinities of cations to strong acid resin are as follows, according to http://www.forumsci.co.il/HPLC/ion_chrm.html:

Tl+ > Ag+ >  Cs+ >  Rb+ >  K+ >  NH4+ >   Na+ >  H+ >  Li+ >

I convinced myself that read <, even though I doubted it, thanks for the clarification. 

So NH4 easily displaces all H+, but should have a harder time to displace all IDA and PROD, since they should stick to the resin slightly better than NH4+ (and also becomes protonated earlier).  That means that the PROD (and IDA) shouldn't come off until all the H+ is pushed off the column by the gradient NH4+ and should show up in/near the final basic fractions, which is what was observed.

Unfortunately, the NMR for the material I collected looks impure, so perhaps I'll try it again with a more dilute eluent gradient.

ALSO:  The PROD should elute slightly after IDA, right?

[TheUnassuming]

ALSO:  The PROD should elute slightly after IDA, right?

Should, but might not elute far enough after to get separation. 

Are you limited to H+ IEC for any reason in particular?  Does your product have any free acid moieties?

[Babcock_Hall]

I only have time for a drive-by comment today.  A gradient of triethylammonium bicarbonate in IEX chromatography can be very effective in purifying nucleotides or amino acid derivatives.