About TFA

[mdyo]

Hi everybody,

Many HPLC methods use TFA in the composition of mobile phases.

Can anyone thankfully help me to know the cases in which we need to use TFA and when we should not use it?

What other compunds can replace it?

Thank you in advance

[Mark Tracy]

TFA can provide a couple of benefits. First it is a strong (but noncorrosive) acid that can protonate residual silanols on the column and also protonate bases. Second it is a mild ion-pairing agent that can alter the selectivity of basic substances. Third, it can often improve peak shape for bases. Fourth, it is volatile.

It has problems too. It is expensive, especially UV/spectro grades. It causes noisy baselines at low UV wavelengths, even when you buy high-purity grades. It can give strange baseline effects, especially below 230 nm. It suppresses ion formation in electrospray MS. It must be stored carefully or it goes bad. Pure TFA attacks many polymers, including the ones used to make bottle caps.

Many methods specify 0.1% TFA in the mobile phase. This is something of a holdover from days gone by when reversed-phase columns had relatively high residual silanol activity. Modern columns with their lower silanol activity can gain most of the the same benefits with as little as 0.01% TFA.

Often you can substitute phosphoric or sulfuric acid. They have better UV characteristics, and do not ion-pair. Baselines are much nicer. They are nonvolatile. They also suppress electrospray ionization. They are cheap. Not as effective at improving peak shape.

Formic acid is sometimes a substitute. It is not as strong an acid. It is volatile and inexpensive. UV cutoff is similar, but baseline noise and anomalies are usually less. It usually gives much better electrospray ionization. It does not ion-pair. It is less effective at improving peak shape.

Methanesulfonic and perchloric acids are also substitutes, but are less commonly used. Hydrochloric acid is rarely used because it is corrosive to stainless steel HPLC components.

I usually try phosphoric or formic acids first, and only use TFA if the results are not satisfactory. I try to use the least amount that does the job.

[mdyo]

Thank you very much Mark! That was really what I want to know about TFA

But if TFA protonate basic compunds, doesn't that lead to elute basic compunds faster or maybe causing them not to be retained?

[Uwe Neue]

Excellent answer, Mark, nothing to add...

[Mark Tracy]

Basic compounds do elute earlier in acid than in neutral, but you get enough retention in the great majority of cases. For instance pseudo-ephedrine elutes with k' = 2 in 15% acetonitrile and 0.1% TFA. Actually, TFA gives slightly later elutions than formic acid because of the ion-pairing effect.

One other thing, TFA is also a mild wetting agent, so your reversed-phase column can tolerate a lower percent organic without risk of dewetting.

[mdyo]

Thank you very much Mark!

[HW Mueller]

Yes, neat summary, Mark.
But, as mentioned before, TFA in my hands had almost been only trouble (one exception: The sep of oxalic using Hypercarb). Thus I wonder whether TFA doesn´t disturb peak shape more often than it improves it? In other words, I would put emphasis on Mark´s statement of using TFA as sort of a last resort.

[adam]

Mark

Can you suggest some references where such issues are discussed. If none exist maybe you should author a paper on the subject ("A step-wise approach to buffer selection in HPLC" - or something like that).

[syx]

It must be stored carefully or it goes bad. Pure TFA attacks many polymers, including the ones used to make bottle caps.

How is the recommended condition to store TFA?
How could be know that TFA goes bad? From its appearance?

[Mark Tracy]

One option is to buy sealed 1-mL ampules (Pierce for example). Bad TFA is usually brown. Another way to tell is the absorbance of 0.1% TFA should be <0.35 AU at 214 nm, 1 cm path.

I have seen one bottle cap that does stand the exposure: made from glass-filled PTFE, lined with pure PTFE, and no adhesives used. It's the cap on my 5-year-old bottle of Pierce Sequanal-grade TFA. Fluka uses some kind of glue to hold the liner in their bottle cap, and once you open the bottle, the glue starts to ooze brown gunk.

One more option is to buy pre-mixed eluents; Fisher sells them, but I haven't tried them.

[Kostas Petritis]

I will vote for the sealed 1-mL ampules (we buy them from Sigma). If budgeting is not an issue, it is the way to go... no worries about storing...

[Bill Tindall]

Mark, I must take issue with your statement that TFA results in better peak shapes than phosphoric acid. My experience is entirely with protonating strong organic acids, for example terephthalic acid. Phosphoric acid is the preferred acid for protonating these kind of analytes and there is no shortcoming of peak shape so long as sufficient acid is added to suppress organic acid dissociation. TFA, nasty stuff that it is, may be better for other applications, but not this example.

Perhaps you could be more specific for where you have seen that TFA results in better peak shapes?

[KarenJ]

Can some one also explain why application notes often recommend using TFA in mobile phase for protein or peptide separations? I think that 0.1% TFA in an acetonitrile/water mobile phase seems to be commonly used.

Thanks,

KarenJ

[tom jupille]

Can some one also explain why application notes often recommend using TFA in mobile phase for protein or peptide separations? I think that 0.1% TFA in an acetonitrile/water mobile phase seems to be commonly used.

There is evidence that TFA acts as a weak ion-pairing reagent, and it's a good denaturing agent.

[Mark Tracy]

BIll,
My comments about TFA giving better peak shapes were in the context of peptides and organic bases. I'm sorry for not making that clear. In your case of organic acids, I agree that phosphoric acid is equally good at controlling peak shape and clearly better in most other respects.