Trying to understand HILIC..

[Mattias]

Hi,

I am quite new to HILIC and I am testing a HILIC column right now (Acquity BEH HILIC) for a basic peptide. I know the basics but is trying to understand the following behavior:

- The retention increases when I increase pH of the mobile phase
- The retention increases when I decrease the ion strength (highest retention with no buffer at all)
- Using TFA in the mobile phase is not good, very little retention

Everything tested at 85% ACN in the mobile phase (isocratic). Sample is dissolved in 90% ACN in water.

Good peak symmetry and plate count I think. Around 25.000 plates on a 150 mm column.

[Gerhard Kratz]

Hi Mattias,
HydrophilicInteractionChromatography. I go with a very simple explanation. It is a partition Chromatography between the mobile phase and a water layer on the surface. Beside the pH you can tune the retention time by different water contents. Retention time for charged compounds will be reversed, neutral compounds will mostly stay stable. HILIC is not normal phase. High performance Liquid Chromatography is the title of a HPLC book by Uwe Neue who worked for Waters. If you have a chance to get access to this book that will give you valuable information on HILIC. Or Google. Good luck

[Mattias]

Thank you Gerhard,

I remember Uwe from his time at this forum, he really knew his chromatography!

I do not have this book, but will see if I can order it. I understand the general principle of HILIC, but cannot explain why (for instance) pH plays a big role in my separation.

My compound has several positive charges at neutral pH. But still the retention goes up with pH. I would have guessed the opposite, or at least no effect.

[HPLCaddict]

My compound has several positive charges at neutral pH. But still the retention goes up with pH. I would have guessed the opposite, or at least no effect.

BEH HILIC is AFAIK a plain silica column. If you increase pH of your mobile phase, you may deprotonate silanols on the silica, so the stationary phase has a negative charge. Because of this increased polarity, the adsorbed water layer will get bigger and therefore increase retention...and if your analyte has positive charges, you'll get ionic interactions which will increase retention,too.
Just a very speculative guess
Would be interesting to see what happens if you use a neutral HILIC column, such as the BEH amide.

[Andy Alpert]

Addressing your various questions:

1) BEH HILIC is uncoated BEH material, I believe. That's a hybrid between silica and silica hydride. It contains fewer silanols and the ones it does have are more isolated and less acidic. Waters has contended that it is uncharged below a pH or around 9, but empirical evidence suggests that in fact it does acquire some negative charge in the same pH range as silica: pH 4. The ease with which an immobilized ionizable group acquires charge is affected by the charge on neighboring groups. Therefore, a titration curve of a material like silica or a BEH material would be a continuum rather than a curve with a sharp inflection point. That accounts for the continuous increase in retention of your basic peptide with pH.

2) Further on that retention: Retention on a HILIC column is a combination of hydrophilic interaction (the partitioning into the immobilized aqueous layer that Gerhard mentioned) plus any electrostatic effects that may be present. In your case, this involves electrostatic attraction between your basic peptide and the (-) charged silanols of the BEH material.

3) Salt shields electrostatic effects in chromatography, both attractive and repulsive. In your case, it shields the electrostatic attraction, which is why retention of your basic peptide decreases as salt increases. What's left is the retention by pure hydrophilic interaction.

4) Basic residues are normally the most hydrophilic residues in peptides. TFA tenaciously forms neutral, somewhat hydrophobic ion pairs with them, rendering them all but invisible to a HILIC column. That's why retention decreased so much. If this bothers you, then use formic acid instead (if you don't need to monitor absorbance at low wavelengths) or a salt. Salts that are quite soluble in HILIC mobile phases and which are transparent at low wavelengths include triethylammonium phosphate and the sodium salt of methylphosphonic acid. Be warned that these exhibit very different selectivity in HILIC of peptides. Phosphate promotes retention of basic peptides and suppresses retention of acidic ones while phosphonate has the opposite effect.

Per HPLC addict: Applying a neutral, hydrophilic coating tends to suppress the charge effects of the stationary phase. BEH Amide is in that category, as are various diol and pentahydroxy materials, our own lovely PolyHYDROXYETHYL A, and so on.

If you'd like to read more on the subject, then try the following papers:

1) Anal. Chem. 80 (2008) 62-76.

2) Anal. Chem. 87 (2015) 4704-4711.

[Mattias]

Thank you very much for you insight in the subject!! It was exactly what I needed And it makes perfectly sense too.

The method needs to be MS-compatible, so I am stuck with acetate or formate more or less. I could maybe run with just acetonitrile and water, but I understand that there may be a risk that I will see differences between column batches (depending on the electrostatic properties of the column).

I have learnt the hard way that this type of separation need a much longer equilibration time, and that fast gradients are not optimal. Not sure how temperature affects the efficiency of the column, have only tried 40°C yet. Same with flow rate, only tried 0.4 ml/min.

[Andy Alpert]

Matthias:

You are correct; HILIC does require longer reeequilibration times than do other modes of chromatography. Do use a formate or acetate salt. Under no circumstances should you run a basic peptide in HILIC with no electrolyte in either the mobile phase or the sample solvent. A number of HILIC stationary phases (especially those with "zwitterionic" ligand) act as high-capacity cation-exchange materials when there is less than 20 mM salt present. Leaving out salt entirely would cause your basic peptides' elution time to wander all over the map, since the charge state of the coating would be greatly affected by any residual salt that had titrated the charged groups in the stationary phase and which was now slowly being leached out, with no salt in the mobile phase to replace it.

Optimization of the variables for a MS-based detection method can't take into account just the conditions that would result in a convenient MS analysis. The requirements of the separation step must also be taken into account.

[Mattias]

Hi,

I am only using 5mM acetate at the moment, but it seems to work. The retention times are stable.

Higher temperature seems to give more retention (!), this is also opposite to reversed-phase!