finding Void time(To)

[amanda]

I am looking for a simple way to find the void time(To). I am using a C18 column, 5micron particles size, 10cm long, 4.6mm id, methanol water (70:30) at 2.06ml/min. Please suggest me an easy way to find out the void time(To).....thanks in advance

[Consumer Products Guy]

Inject some acetone or DMF, they have UV absorbance and won't be retained.

[HW Mueller]

We had an extensive discussion not too long ago which indicated that this is not really simple. I have been using T2O (tritium water) for years, quite good for reverse phase columns, but apparently not for HILIC. Just last week I got T2O retention with one mobile phase (20% H2O!) on a ZIC-HILIC (zwitterionic phase).

[tom jupille]

HW, that's interesting but I guess we should not be surprised. My understanding of the mechanism of HILIC is that it involves partition between the mobile phase and a water-rich layer at the stationary phase surface (someone correct me if I'm wrong!). Looking at T20 behavior as a function of %H2O in the mobile phase might give some interesting insight into the transition from HILIC to reversed-phase behavior.

[Einar Ponten]

I agree, this is interesting!

Toluene is a void marker on a HILIC column. Sometime sodium nitrate may be used on a RP column, but what is valid for a AQ phase or a mixed-mode column?

Hans, have you seen resolution between D2O and T2O..?

[HW Mueller]

First of all, I noticed that I didn´t make clear, above, that I meant the discussion on tm in this forum.
Tom, here some more detail: 20% aqu acid or base (low concentration)+80%EtOH give a rt which must be at the tm of the ZIC, while 20% NH4OAc pH4.8, 0.075M + 80% ACN has a tm+1min.
I am still trying to find a "handle" with which I can grab this for comprehension.
Einar, I can´t detect D2O. (Maybe as a negative peak in UV, but my negative peaks are are not sortet out yet either).

(The biggest problem here is that it is difficult to explain what I am doing to the meds)

[tom jupille]

Hans, you have too many variable for me to deal with

What I was getting at is that if you assume that HILIC is driven by the formation of a water-rich layer on the stationary phase surface, and if you assume that T2O behaves just like H2O and is small enough to fully permeate the stationary phase, then the retention of T2O should provide a measure of how much of the mobile phase water is being bound to the stationary phase. Tracking that as a function of mobile phase composition for a particular column might provide a measure of the transition from RP (or whatever mechanism applies in highly aqueous mobile phases) to HILIC which is more meaninful than looking at retention of other molecules. It might also give useful insight into organic solvent selection for HILIC separations (EtOH is protic, ACN is aprotic; if you decrease the EtOH concentration even further, do you eventually see retention for T2O.

The problem is real life is that as analytical chemists, we are in a "service" business, and finding time to explore some of these questions is hard to do.

[HW Mueller]

My problem is that I am hung up by the thought that the ratio,
mobile H2O/stationary H20
should be so huge that the retention is obliterated.

[tom jupille]

My problem is that I am hung up by the thought that the ratio, mobile H2O/stationary H20 should be so huge that the retention is obliterated.

Actually, I had more of a conceptual problem the other way around (visualizing that a significant fraction of the water could be "bound").

But being very simplistic about things, let's assume a uniform cylindrical pore of 10 nm diameter. And let's also assume that there is no preferential binding of water to the stationary phase surface. That would make the mobile/stationary H20 ratio essentially infinity; the T20 would have 100% of the pore volume acessible as mobile phase and would elute at t0.

Now change the conditions so that you get exactly a monolayer of water sticking to the stationary phase surface. Taking the diameter of a water molecule as ≈ 0.3 nm, that means we have reduced the effective "mobile" pore diameter from 10 nm to 9.4 nm (there are two water molecules per diameter) and reduced the cross-sectional area from 100 nm^2 to 88 nm^2. The mobile/bound ratio is now 88/12 ≈ 7.

Stick a double layer of water to the surface and you reduce the cross-sectional area to ≈ 77 nm^2, giving a mobile/bound ratio of 77/23 ≈ 3. A triple layer of water gives a mobile/bound ratio of ≈ 2

and so on . . .

I have probably messed up the math, but on an "order of magnitude" basis, it looks like the ratios are not necessarily "huge" in either direction.

[DR]

This sounds like a G. Guichon type problem...
Is he a lurker here? http://www.chem.utk.edu/guiochon.html

[sfe-co2]

Hi Amanda,

I recall Uracil being used by some column manufacturers as a void marker, especially on C18 phases. However, I've used Benzamide before but think Uracil is better since it is slightly less retained than Benzamide.

Hope this helps.

[Tobias Jonsson]

This thread is taking a direction far beyond what amanda was asking for, but since I agree it is very interesting I’ll continue…

Hans, the retention time of T2O in 80% ethanol cannot correspond to the dead volume on ZIC-HILIC since water have retention and forms a bound water layer under these HILIC conditions (just as Tom wrote), but it would be really nice if you could demonstrate separation of H2O and T2O when both have retention... T2O should be slightly more hydrophilic, right?

Tom and Hans, I am, on the other hand, intrigued by the fact that HILIC is postulated to be a partitioning mechanism, whereas all plots of log k vs %H2O are strikingly non-linear. Instead, plots of log k vs log %H2O are almost linear for all types of compounds [for a compilation see the recent review in J. Sep. Sci., 29 (2006) 1784-1821], similarly to what is observed in normal phase chromatography.

The question is thus if the bound water layer acts as bulk water into which retained solutes partition, or if the bound water layer instead form a surface onto which retained solutes adsorb while replacing solvent molecules? Or perhaps combinations of this...

[HW Mueller]

Tom, thanks, just the jolt I needed. Even though I stated elswhere that I am for junking some dimensions, units... I confused nm and A in a very rough in head estimation of your calc type, though I also considered (over the thumb...) the intra particle volume. Anyway, the retention is really proof that your calcs can not be too far off.

Tobias, the discussion nicely shows how difficult it can be to determine the tm (to), maybe showuing Amanda that simple methods might not be so hot.
My problem was not with retention of water per se, but rather with that stationary (retained H2O) and mobile H2O ratio.
As with D2O there is a problem of detection with H2O.
Thanks for the ref, have to look at it.

DR, the last sentence above is also meant for you.

[rc_12321]

We generally use uracil for this purpose.

http://home.planet.nl/~skok/techniques/ ... html#table

[HW Mueller]

We have been into urcil before, it´s far from perfect.