The Hydrophobicity and retention time of a compound

[s]

Is it possible to directly say something about the hydrophobicity/hydrophilicity of a compound when looking at the retention time of the compound on a reversed phase column like C18? Or do other parameters like pH and temperature have so much influence on that issue as to hinder a short and precise conclusion?

[Kostas Petritis]

Well...

You can always say something, especially if the compound is very retained in C18. Temperature does not influence that much (speaking about temperatures close to the ambient one...) but pH might inlfluence compounds retention a lot. Maybe you can have a better idea of it's hydrophobicity and the role of the pH if you do three gradients from high aqueous to high organic with acidic, neutral and basic mobile phase pH (although things might get tricky in the case of a zwitteronic molecule with a lot of pKa's).

Kostas

PS: I assume that you talk for a compound that you have no idea of it's structure, otherwise there are other ways to calculate it much more precisely...

[s]

Kostas, thank you for your quick reply.....

Actually, I do know the structures of my compounds (maybe I should have mentioned this at the beginning...) but I am also interested in a more general assessment about whether you could directly say something about the hydrophobicity of a compound exclusive from reversed phase chromatography data.

But now you have raised my interest about calculating the hydrophobicity. How do you do or should the question rather be: does a calculating-tool exist somewhere on net?

I'm sorry for these rather "basic" questions but I am quite new at this topic so every bit of new information is new knowledge for me.

[Kostas Petritis]

Soren,

Yes you can use reversed phase chromatography to calculate the hydrophobicity of your compounds and there are several references dealing with this (i.e. I have seen papers calculating the hydrophobicity of derivatized amino acids etc...).

There are also a lot of software tools that can calculate the logP. Most of them you just design the molecule and they can calculate the logP. I would dare to say that the smaller the molecule the better the calculation (i.e. it is difficult for the software to calculate the hydrophobicity of very large molecules).

Kostas

[tom jupille]

Actually, the first-order retention mechanism in reversed-phase is based on hydrophobicity. The pH effect on retention comes primarily from the fact that the ionized form of an analyte is more hydrophilic and the neutral form more hydrophobic.

In the real world, "secondary" retention effects can complicate life by holding some molecules a bit more tightly or keeping other molecules away. These effects are generally small relative to hydrophobicity, but they do provide a mechanism for separating molecules of similar overall hydrophobicity (arguably the reason reversed-phase works for as many things as it does!).

If you are seeing large differences in retention, then it is usually safe to say that the more retained compound is more hydrophobic. For close peaks, that's a dangerous generalization, as Kostas implied.

Probably the most generally used measure of hydrophobicity is the octanol-water partition coefficient. Reversed-phase retention is used to estimate it. A quick Google search came up with at least a couple of on-line calculators:

http://www.pirika.com/chem/TCPEE/LOGKOW/ourlogKow.htm

http://www.molinspiration.com/services/logp.html

[Uwe Neue]

There have been publications by several authors on the subject of getting a measure of hydrophobicity from retention in reversed-phase chromatography. There is a good summary article by Klara Valko in the Handbook of Analytical Separations, Volume 1 (Elsevier 2000). You may also look at Anal Chem 69 (1997) 2022 or Anal Chem 70 (1998), 4228 for more references and additional information.

[Kostas Petritis]

In addition of what Uwe suggested you can look in the more recent reference : Kaliszan et al J. Chromatogr. A 965 (2002) 117. The authors in addition of the hydrophobicity they are able to calculate the pka of their analytes (with different gradients) from gradient RPLC.

I agree with Tom's statements, however I will promt you in one of the previous postings by TOT that you can find here: http://www.sepsci.com/chromforum/viewtopic.php?t=867

In his example, for the same percentage of MeOH:water his molecule has a retention of 4.6 min at pH 2.5 and more than 30 min at pH 5.5 (he didn't wait more than 30 min... who knows when that molecule would be eluted). The experimental hydrophobicity of that particular molecule would be very much pH dependant.

Furthermore, the logP and pKa values given by different software are good approximations but I wouldn't count solely on them for my method development.