Any explanation for "reverse fronting" peak shape?

[Tony]

Hello,
Can anyone give insight into what may cause a "reverse fronting" peak shape? I am analysing an uncharged (i.e. no acidic or basic groups)alkene by non-aqueous reverse-phase HPLC using a 100 x 3mm hypercarb PGC column. The mobile phase is 10% ethanol 90% dichloromethane, with the alkene injected in a small amount of ethanol, and pumped at 0.5 mL/min. The column is maintained at 30 degrees C. Detection is by APCI-MS.

What I mean by "reverse fronting" (for lack of a better term) is that the peak looks like an overloaded peak, but in reverse; i.e. with the bulge on the tail of the peak. The front of the peak is shaped perfectly, rising smoothly to the apex. Following the apex is a symmetrical tail falling to about 20% of the peak height. At this point the peak bulges out (about another peak width!) but then continues to the baseline in a symmetrical fashion, with no marked tailing.

There are no obvious voids in the lines connecting the injector to the column or the column to the detector. The alkene is pure, so I'm not looking at an impurity as a trailing shoulder (the mass signature is also identical across the whole peak). I know the interaction mechanisms between solutes and PGC phases are somewhat unusual and difficult to explain compared to silica-based C18 - but perhaps this isn't the cause and I am missing something obvious....?

Any help or advice would be appreciated
Thanks
Tony

[Uwe Neue]

How about making up the sample in the mobile phase?

[Tony]

I could do that, but I thought EtOH should be weaker than DCM, and therefore not a problem?

[maris]

In NP mode EtOH is stronger!

[HW Mueller]

Like you said, Hypercarb does some strange things, among them seems to be that it is extremely sensitive to mobile phase changes. Thus I would take Uwe´s suggestion very seriously. Also, is your olefin a geometric isomer mixture?
(maris, you mean RP?).

[Tony]

The olefin is documented to be a single isomer, but I will try several other related compounds to see if I get the same problem. I will also try injecting in mobile phase, and let you know what happens.

Thanks
Tony

[Tony]

OK, I've tried dissolving in the mobile phase (90%DCM 10%EtOH), and I get the exact same strange reverse-fronting peak shape, also for other olefins I have tried. I'm at a loss to explain what is going on,

[dblux]

Have you tried to replace precolumn (in case of mechanical impurities on inlet frit) ?

[HW Mueller]

What happens if you use MeOH, mabe with small amounts of H2O, or other solvents? I would get rid of the dichloromethane, it probably contains a stabilizer, is very volatile .....ugly.

[Tony]

OK, I have removed the guard, and there is no effect on the chromatography. I have also tried a different hypercarb column, with the same result, so I don't think it is a problem with column degradation or a mechanical column fault.

As for different solvents - here's something interesting. The strong solvents THF, hexane, or MTBE do not elute the olefins from the column (at least over a 1 h period). However, toluene does, and very effectively - everything comes off in <3min. With toluene, the peak shape is also better (although it's hard to judge with these preliminary runs - I will have to decrease the % toluene to get later retention and see if peak shape is improved).

I suspect the elutropic stength of toluene with the hypercarb material has something to do with strong pi electron interaction between toluene and the PGC phase (in addition to toluene's nature as a strong hydrophobic solvent). If it was a purely strong solvent effect, then I would have expected hexane to elute the olefins. Any other theories as to why this might be happening would be most welcome.

Tony

[HW Mueller]

On "theory": You may get out an earlier discussion on Hypercarb. What other functional groups does that olefin have?

[tom jupille]

To amplify a bit on HW's comment, do a search of the forum (search function is near the upper left of the screen, just below the big "Chromatography Forum" logo). I think there were a couple of earlier threads about Hypercarb.

[Tony]

Tom,

I've done a search through the forum archives and looked at all the hypercarb threads, but there's no further information shedding light on peak shape. I have since come across a few literature references where this problem is noted, but generally accepted as a sacrifice one has to make if resolution of isomers using Hypercarb is a priority. I've also spoken with a tech person at Hypersil, who confimed toluene and chloroform are usually the strongest hypercarb solvents, and that peak tailing is a problem with hydrophobic analytes due to strong dispersive effects.

I have noticed that some people make an addition of equimolar triethylamine:formic acid to the mobile phase to improve ELSD response, but no comment is made on how this affects peak shape. I might give this a try, but I can't see how these additives would affect the peak shape for my non-polar olefinic analytes, which differ in degress of unsaturation but otherwise have no acidic or basic functional groups.

thanks
Tony

[HW Mueller]

Tony, what do you mean by "strong dispersive effects"? A wide dispersion of hydrophobic interactions types?

[Tony]

I'm just using the term as used by others in the literature when they describe how PGC phases interact with hydrophobic solutes. My interpretation is that "dispersion" refers to the solute spreading out on the stationary phase, with more spreading and interaction leading to more peak tailing.