the role of TFA on Reverse phase chromatography?

[dodora]

Hi,

I would like to know the role of the TFA (trifluoroacetic acid : CF3COOH) used on reverse phase chromatography. It's is often used with a concentration of 0.1% on the mobile phase.
How does it improve resolution and sharpen the peaks? Does it form ion pair? Why?

(This is the unique role that i find on internet until now , but i still ignoring the details).

i would like to use this chromatography to check the purity of an ferrocenic amid that has an amine (NH2) group at the end of its chain, and my friends tell me they use to add TFA to the mobile phase but they ignore the reason.

Please help me. Please forgive my english, i don't use to write in english.

Best regards

[dodora]

Hi again,

sorry i forget to say that by "reverse phase chromatography" i meant chromatography that uses C8 or C18 column.
i didn't mean the exchange ions chromatogrphy that uses ion-pair-agent.

Sorry,

Best regards

[tom jupille]

Nothing in HPLC has only one effect

The two most easily describable effects of TFA in the system you are proposing are:

1. pH control (somewhere around 2.1 if memory serves). This is sufficiently low to avoid problems with partial ionization of most analytes and to suppress ionization of residual silanols on the stationary phase (this is probably the major reason for the peak sharpening).

2. Ion-pair. TFA is a weak ion-pairing reagent.

And, by the way, your English is just fine!

[dodora]

Thank you Tom for your answer.

1." to avoid problems with partial ionization of most analytes " : this means that we want to have a total ionisation of the analyte ? which means that with a pH of approximately 2.1 all the NH2 groups will be ionised and takes their acid form that is NH3+ , and thanks to that, it become easy to detect them (and then to detect the ferrocenic amid) ? Because , in HPLC , we can't detect all analytes , can we? So we use some reagent that react with the analytes to make its detection possible, use we?

2. "TFA is a weak ion-pairing reagent" : i find this role of TFA only in the ion-exchange chromatogrphy. i don't see how it can be an ion-pairing reagent in my case : i have a free NH2 group at the end of the ferrocenic amid'chain. Can u help me?

For a moment i thought that in presence of TFA , i would have a cation and an anion : R-NH3+ , CF3COO- , that why it becomes possible to detect the amid by HPLC.

This is my personal idea, i m not sure at all.

Thank you very much for helping me.

Best regards.

[tom jupille]

." to avoid problems with partial ionization of most analytes " : this means that we want to have a total ionisation of the analyte ?

In general, you want to start with either total ionization or total suppression. Partial ionization runs the risk of robustness problems, since the degree of ionization can change with small variations in pH.

which means that with a pH of approximately 2.1 all the NH2 groups will be ionised and takes their acid form that is NH3+ ,

Yes

and thanks to that, it become easy to detect them (and then to detect the ferrocenic amid) ?

No, the comment did not refer to detection but rather to potential variations in retention.

Because , in HPLC , we can't detect all analytes , can we? So we use some reagent that react with the analytes to make its detection possible, use we?

"post-column reaction" detection is used in HPLC, but usually for compounds which cannot be conveniently detected any other way (which does include many amines). That's not what the use of TFA is about.

"TFA is a weak ion-pairing reagent"

In one "mental model", the CF3 group is somewhat hydrophobic, so presumably some of the TFA sticks to the surface of your reversed-phase packing. Your cation can then interact with the carboxyl group via an ion-exchange type of mechanism. CF3 is not very hydrophobic, and the carboxyl group is only partially ionized, which is why TFA is a "weak" ion pairing reagent. The effect has been well documented with peptides, in which case the relative retention of basic peptides increases with increasing TFA concentration.

For a moment i thought that in presence of TFA , i would have a cation and an anion : R-NH3+ , CF3COO- , that why it becomes possible to detect the amid by HPLC.

You do, and the other "mental model" is that these two neutralize each other to form a relatively hydrophobic "ion pair" which can be retained on a reversed-phase column. As with the comment about pH, however, this has nothing to do with detection, but rather with retention.

[Kostas Petritis]

I agree with everything except from the statement "and the carboxyl group is only partially ionized". With a pKa of approximately 0.23, I would say that the carboxyl group is almost fully ionized...

[dodora]

Thank you Tom for all information and for being so patient. Thank you kostas for your answer and information about pka.

So, the analyte may have a bad retention. This is why we try to improve its retention using TFA.

To resume, thanks to the relative hydrophibicity of the CF3 group , TFA has an affinity with my C18 column.... that 's why he comes on the surface of my stationary phase. Then, my amine group comes to react with the free carboxy group of the TFA. By this why, i get :

--C18
CF3-COO,H3N-R

By this why we improve the retention of the analyte !

Just a laste question :
With a pH2 , the TFA and the NH2 group are certainly totaly ionised.
My mobile phase , contain already ionised TFA : CF3COO-,H+
so when the séparation starts, CF3COO-,H+ go instantaneously to the surface of my column. Then ,the analyte R-NH2 comes to the surface and get the proton H+ to become R-NH3+ and then keep in touch with CF3COO-. Since the TFA is a weak ion-pair agent, the ion-pair is then liberated from the column and eluted to be observed at the end.
Is this ta probable mecanism?

Thank you very much Thank you

[Kostas Petritis]

The ion-pairing mechanism is not just on and off, it is kind of an equilibrium; your ion pair will "spend time" in your stationary and mobile phase back and forth. The amount of time that spends in the stationary over the mobile phase it will kind of determine it's retention time. Also the ion pair it will also break and reform so instead of the whole ion-pair partitioning between the mobile phase and stationary phase a lot of times you will have a dynamic ion exchange mechanism where the TFA will not leave the stationary phase but the ion pair will break, your analyte will enter again the mobile phase and reform another ion-pair later and so on...

[HW Mueller]

Also, the equilibrium concentration of the ion pair should be extremely low so, as Tom already told you, it should not play a role in detection.

[tom jupille]

With a pKa of approximately 0.23, I would say that the carboxyl group is almost fully ionized...

You;re right, Kostas. I sit corrected!

[dodora]

THANK YOU VERY MUCH to all : Tom, Kostas and HW Mueller
i m very happy to know all these details that i did not find anywhere.
i have inderstood all your answers. THANK YOU

[mohan_2008]

I agree both with Costas and Tom:

Yes, with a pretty low pKa as 0.23 (empirically, three fluorine atoms exerting electronwithdrawing field effect on the "carbonyl carbon" thus making the carboxylic protons more acidic) - hence Costas is correct.

However, it is considered as a Weak ion pairing agent (compare to a sulfonate sulfur, as in a sulfonate ion pair, which exerts direct electron withdrawing affect on the sulfonate proton) - hence is much stronger than TFA - hence Tom is correct.

A strong or a weak ion pair comparison is only valid, as relative to each other.

[dodora]

thankyou very much for confirming answers.
best rigards.