regeneration of column after using ion-pairing reagent

[PHOBIUS]

Hello,
I would like to ask one question. I worked with HypurityAquastar C18 column and I have contamined it with tetrabutylammonium hydrogensulfate (it is an ion-pairing reagent). After using it for some time, retention loss was observed and the pressure went higher. How should I remove this ion-pairing reagent from the column without destroying it completely?
(sorry for my bloody english)

[wrighty]

as far as i know once the ion pairing reagent has entered the column it cannot be removed without damaging the column.

I dont use reverse phase HPLC such as this personally but i remember from a recent course that once a column has been desgnated as being used for ion-pairing it cannot be used for any thing else. this is because the reagent has interacted with the C18 chains and it is quite difficult to separate the two.

perhaps someone could correct me on this?

hope this helps phobius

steve

[PHOBIUS]

So I can modify the conditions to achieve retention again and it won´t get any worse?

[rhaefe]

there was a thread about this quite a while ago. Although I never tried "cleaning" a RP column used with IP reagents I recall Mark Tracy giving a procedure on how to do it. I do not remember all the details but I am sure you can find it if you search for it. It did involve a washing step with methanol...

[Mark Tracy]

Anionic IP agents are pretty easy to remove from C18 columns, but cationic IP agents are much harder because they stick by ion exchange to the silica substrate. Unfortunately, you have already seen loss of retention and elevated pressure; removing the IP will not bring back the lost bonded phase, nor remove the fouling from the head of the column.

You can try phosphate buffer at pH 2.5 mixed with 30% methanol to remove the Bu4N+. The concept is to protonate the silica substrate, provide some competitive cations and enough organic to elute the hydrophobic cations. To tell if you have removed it all, inject an acid/base/neutral test mixture and keep washing until the retention ratios stabilize. If you want to do something like this in the future, characterize the new column before exposing it to the IP agent so you know what it should look like after cleaning.

A fair bit of work. If you have to budget to devote a column to this application, it would be worth it.

[PHOBIUS]

Do you think, samples injected in 0.4 M TCA (trichloracetic acid) will destroy this column even more? btw, this column wasn´t new, it was already dedicated to my evil hands , so i can test it on the compound previously done without IP and I´ll see. I am just afraid that I won´t be able to use it on LC/MS application again (I am worried about IP eluting from it continuosly and messing the MS)

And one more thing: I would like to separate some compunds also LC/MS. I have read something about volatile IP - di-n-butylammonium acetate. Will this IP mess my new column again, or should it be OK ?

[Mark Tracy]

Yes, TCA will be harsh on the column, but that is why columns are sold by the "consumables" division! I'm guessing that you used TCA to precipitate plasma proteins; even after precipitation and filtration, those kind of sample are prone to foul columns.

I have not tried Bu2NH2+ as a IP agent. Being a weak base and less hydrophobic than the Bu4N+, it should be easier to remove. I'm not familiar with your column, but if it is a modern design with low silanol activity, it should be easy enough to clean. Actually, it may be more difficult to remove it from all the crevices and polymers in your HPLC pump than from the column if you need your system clean enough for LC/MS.

[Bruce Hamilton]

For decades, I've not worried about using IP reagents on reverse phase columns, I just assumed I was washing them off with my cleaning procedure, and used the same column for conventional RP and IP work.

When I started reading this forum I discovered that I shouldn't do that, but I still do, as my columns seem to last for years. However, I would strongly agree that if you can afford to dedicate columns to specific IP reagents, you should.

If, like me, you only need to run IP reagents once every few weeks or so, you should follow the manufacturer's recommendations for removing the ion-pair reagent, or develop your own...

My cleaning procedure is pretty simple, but does take time - seems to work for me, but certainly isn't optimal, as instrument availability was not critical for me. If your IP reagents are heavily laden with buffer, perhaps start the cleaning with a higher water content solvent, but 1:1 works for most compendial methods I've used.

The degasser rinse is only relevant if you have polymeric membrane degasser, as many modern instruments do. I've found that several IP reagents tend to form an equilibrium with the membrane that needs a reasonable solvent volume to remove. I work mainly with 250 x 4.6 mm columns at 1-2 ml/min, and the procedure is based on those columns.

Replace the ion-pair mobile phase bottle with 1:1 water / solvent ( CH3CN or MeOH ), rinse solvent lines and degasser membrane for at least 15 mins at 5 ml/min, then switch to column and flush for ~30 mins at the normal flow rate. Remember to flush/actuate the injector as well.

Replace first mobile phase with 95:5 water / solvent and rinse degasser for 10-15 minutes at 5 ml/min, then rinse column for 30-60 mins at normal flow rate.

Revert back to 1:1 water / solvent and flush for a further 30-60 mins at normal flows. Testing should show peaks should then have returned to conventional retention times, if not try more 95:5, or even 100:0 ...

Good luck, and please keep having fun,

Bruce Hamilton

[PHOBIUS]

Thanks for your advices. I decided to keep the column mentioned above for analysis by HPLC-UV only, i won´t use it for LC/MS. TCA is used mainly to precipitate proteins. OK. I use it for effective (and quick) extraction of some nucleotides. Another possibility is to do it with boilling ethanol, but the nucleotides I am working with are too unstable (especially at higher temperatures) I do not have much of choice. I hope, the column will "digest" at least a few samples in TCA. Let´s see how many samples do this column sustain.