Graphite Column

[s]

What are the main differences between a graphite column (e.g. Hypercarb) and a C18 column (e.g. Phenomenex Luna(2))?
So far, I understand that the graphite column is more hydrophobic than (any?) C18 column but I also heard that it has anion capabilities? Any other qualities?
How about using a 100% aqueous solvent for the PGC? Would that provide me with any problems in terms of a "deweeting" problem?
Does anyone have a reference or a link to an article concerning this subject?

Thank you all in advance for your help.

[Newman768]

The retaining mechanism of a graphite column can't be compared with those of rp columns.
With a rp column the retaining mechanism is based on interaction with the hydrophobic functional groups (like C18) and hydrophilic silanols.
On a graphit column there are only interactions between the electrons within the graphit and and electron pairs from O, N, aromatic rings, ... of the analytes.

That means one can't say that a graphit column is more hydrophobic than a C18. The interaction mechanism of a graphit column is too different to be compared with a c18. What one can say (and see) is that compounds with lots of electron pairs are more likely retained on a graphit column than on c18 (like compounds with two or more aromatic systems, carbonic acids like 2,4-dichlorophenoxy acetic acid).
I have never read that a graphit column can't be used with 100% aqueous solvent and I doubt that something like dewetting can happen with a graphit column.

More information you can find on the websites of Thermo Electron (www.thermo.com), ZirChrom (www.zrichrom.com) and in some articles of LC/GC.

[Chris Pohl]

One additional point worth mentioning concerning Hypercarb is the well-documented presence of anion exchange sites. While the specific source of these anion exchange sites has not been identified, a number of researchers have documented the presence of such sites. At a conference that I attended last week one researcher demonstrated that these anion exchange sites appear to have interesting redox properties. Treatment of Hypercarb with hydrogen peroxide more than doubled the anion exchange capacity.

[HW Mueller]

Chris, that sounds interesting. Do you remember whether the anion exchange was attained without acid (better: H3O+) being present? The for-forrunner of Thermo used to indicate that a charge transfer gave rise to the ion exchanger properties. (Thus we analyzed oxalate with TFA/H2O mobile phase, incidentally, the only time that I was ever able to get something useful out orf TFA. Charge transfer probably referred to H+ formally attaching to pi-electrons). Also, did the H2O2 oxidize saturated carbon to aromatic type pi-bonds? Probably not known?

Newman, there is no substance that doesn´t do van der Waal´s type interactions, so some parallelism does exist.

[Kostas Petritis]

The porous graphitic carbon is indeed a very interesting material.

First of all, although I agree that is difficult to compare the graphite column with the alkyl bonded ones, you can still find people that says that graphite column is more hydrophobic. This perception came after the studies of Knox and Kaur (J Chromatogr. 352 (1986) 3-25 where they saw that they need to use 95% methanol in order to achieve comparable retention as in an ODS Hypersil which was used only 50%. However, the analysed compounds were phenol, anisole, p-cresol, phenetole and 3,5-xylenol...

Actually the graphitic column has been used in almost all the modes of chromatography, including normal phase and as a cation exchanger (and as mentioned before as anion exchanger).

HWM, retention was achieved with 100% water. Actually the acids were used as electronic competitors in order to elute the anionic compounds out of the column (Elfakir et al. J. Chromatogr. A 829 (1998)193-199).

What is even more amazing is that the column is able to retain inorganic cations as well (remember that there are not residual silanols in this support). Under 100% water there was infinitive retention for Cu, and Pb was quite retained. The authors optimized their conditions to separate Ni, Cd, Pb and Cu (Merly et al. J. Chromatogr. A 804 (1998) 187-192).

It is worth to say (as Chris mentioned) that they haven't identify the reasons of the anion or cation exchange capacity of the graphitic column.

Other interesting capabilities of this support is it's exceptional selectivities towards isomers.

Another one is the affinity toward certain polar groups. For example it was found that the retention of mono-di- and trihydric phenols increased with the number of OH groups in the molecule whereas with ODS silica gel adn PRP-1 they decreased as one would have expected... Knox and Ross called it "polar retentio effect by graphite" I do not know if this term still stands nowadays.

When we used ion-pairing chromatography for the analysis of underivatized amino acids with PGC, we observed totally different selectivities between PGC and C18 for the same ion-pairing reagent. Even the selectivities of isomers like Ile and Leu where inversed (J. Chromatogr. A 870 (2000) 245-254, J. Chromatogr. A 896 2000 253-263).

These carbon columns, being conductive have been also been used under electrochemically modulated liquid chromatography (Ting and Porter, Anal. Chem. 69 (1997) 675-678).

Chris, I think that the work of the researcher you mentioned has just been published (Shibukawa et al. Evaluation of the syrface charge properties of porous graphtitic carbon stationary phases treated with reodx agents, Analyst 129 2004, 623-628).

About the dewetting problem, I have never heard or seen a problem with these material.

That's all I can think for this support.

SorenSJ, if you search in general a reference describing this material, there are two exaustive in the subject book chapter in advances in chromatography Vol 37 page 73-119 and 122-162 that covers everything up to 1997.

I used to collect every article was out there about the porous graphitic carbon material (up to 2000). I still make a copy if I find something on my way but do not try to get everything anymore...

[Newman768]

Most of my information I get at a seminar about Thermo hplc columns some years ago. My impression was, that even the people from Thermo leading this seminar can't say what exactly was going on inside the hypercarb and why.
Compared to the fundament data you collected those infos I heard there were like playing with matchbox cars compared to driving formula 1 racing cars.

[s]

I would like to thank you all for your very interesting replies - they are very helpful for me in my future work.

[Kostas Petritis]

Thanks Andreas

Kostas

[Chris Pohl]

HWM:

I'm not quite sure what you mean by anion exchange without hydronium being present but if you mean can you retain anions at a relatively high pH values, M. Shibikawa at the conference I recently attended (perhaps this data can also be found in the Shibikawa reference cited above by Kostas Petritis) showed me some data comparing retention of aromatic anions versus pH. He showed significant retention even at pH 10 although retention was highest at pH 2.

Regarding the charge transfer hypothesis you mentioned, I believe this explanation has been largely abandoned. The ability to measure and alter the magnitude of both anion and cation exchange capacity suggests an entirely different mechanism. I think a far more likely explanation is that the pyrolysis process used to produce the material results in minute levels of contamination with polar ionizable substituents which are responsible for this retention. The question in my mind after hearing the most recent data is: what sort of it anion exchange site is produced by an oxidation process? The only candidate retention sites for anions that I am aware of will result in diminished retention upon oxidation rather than enhanced retention.

[HW Mueller]

Thanks Kostas and Chris,
this got my interest up now, I didn´t do anything with those columns for at least 8 years (their multiple interaction possibility was a bit bothersome with biological fluid samples), looked in my cabinet and found one unused one. I lend two more to a friend, it will be interesting to compare all of these with my radioactive ions (usually relatively simple matrix).
All I could think of on the oxidation is that more aromatic pi-electrons are produced... but if the charge transfer is unlikely and this stuff still held anions at pH = 10 there must be something weird going on?