hi all,

I heard a lot of people saying that be careful with water when using normal phase column, especially silica column. But why water is bad for silica column? long equilibrium problem? column stability? or other reasons?

I have seen papers using silica column with water/ACN isocratic mobile phase, the silica column nice chromatogram....
(J. Pharmaceutical & Biomedical analysis 23 (2000) 697-704)

What's the real story about water and silica column?

Any comments? Thanks!

Xu

Silica columns work very well in water. The most frequent application is hydrophilic interaction chromatography, a technique that uses mixtures of an organic solvent like acetonitrile with water. It is commonly used for the separation of sugars, but it is good for many different polar compounds. It can also be used as an ion-exchanger for the retention of basic compounds. Silanols are bad in reversed-phase, but when they are readily accessible, one can do wonderful chromatography on them.
The bad reputation of silica with water comes from normal phase chromatography with very hydrophobic mobile phases such as hexane. Then small differences in the water content on the surface of the silica can create large differences in retention. The water content on the surface in turn is difficult to control with hexane as the mobile phase. But with hydrophilic interaction chromatography, there is no problem at all since you are using mobile phases with a water content between 5 and 50%.
I think that silica columns with acetonitrile/water mobile phases will become really popular now with all the recent publications about this technique.

Some more comments, in addition to what Uwe said.
Of course, bare silica and not endcapped phases are not compartible with water containing eluents. THE chromatogram may be very nice, firstly. After several days it`ll be much worth, and after a week your packing will die)). Silica will be dissolved.
You may use water-AcN mixtures when using polar endcapped phases.
As for silica -- dont use water concentration higher than 1-2%. I resolved alprenolol and 5 similar compounds using hexane-isopropanol-water-Et3N-AcOH 70-30-2-0.2-0.8 on 2 different silicas. Standard deviation was less than 1%. Yes, equilibrating the column is time consuming (100 zero volumes), but it works!
Sure, when using unpolar hexane based eluents, uncontrolled water leds to uncontrolled retention.
Constantine

C.Sychov,
the old window panes in the European churches are thicker on the bottom due to the liquid nature of glas. They have, however, withstood the rain for centuries. Therefore, I would predict that your silica might cake up (in 500+ years) before it would dissolve, unless you use fairly strong base.... (Kidding a bit, window panes have a low surface area, thus low contact to water).

Maybe it was the 1st generation of silica columns which gave rise to some myths. In the late 60īs? we ruined one by using chloroform on it. The chloroform contained EtOH as stabilizer, the latter changed the columnīs performance so strongly that we junked the column. For the next column we used only chloroform which was treated with con H2SO4. The column now survived, but we, the students, almost did not: The chloroform developed phosgene. (Seems that I am repeating myself?)

Constantine,

You said the opposite of what I said. Silica is very well compatible with aqueous mobile phases, as are packings that are not endcapped. Silica works like a charm for HILIC, which is carried out with large fractions of water - 95% to 50% water. There have been recently several publications by Naidong Weng showing how silica can be used with advantage in LC/MS applications.

Hi,

I have not used HILIC mode much. Is there any advantage to the polar bonded phases over bare silica, I know that they hydrolize easily and can bleed? Is there any systematic method development procedure used to alter selectivity while using HILIC? Can you use column switching and load the flow through fraction of a RP column onto a HILIC column? Thanks for any help.

Amino columns bleed. Other polar surfaces may do well, but if silica works, why bother?
We have used off-line combinations of reversed-phase sample clean-up with HILIC separations. We have not yet made any attempts to do this on-line. On first glance, it should not be impossible.
For method development, you have the option of working with HILIC, and using different organic solvents such as acetonitrile and isopropanol. You can also manipulate the pH and/or take advantage of silica as an ion-exchanger for bases.

To Uwe I dont know what is HILIC. Maybe some links for me? :)
To HW Mueller :) What does " cake up" mean? ( on knowledge of English rather than on chromatography)))

This was, sort of, a joke that the particles will coalesce (stick together) by liquid flow before dissolving. (Assumes that, formally, silica gel is a liquid, like glass)

Constantine,

I had posted you an answer, but it got lost somehow.
HILIC is hydrophilic interaction chromatography. It works in the opposite way to reversed-phase: high retention with lots of organic solvent, low retention in water. Look at it as an extension of normal phase chromatography into the world of aqueous mobile phases. You can brush up on this in my book on "HPLC Columns".

Uwe, you mean so called "partition" mechanism?
Silica columns rapidly loose their efficiency due to swallowing, its my experience. Besides Separon silicas dont stand water conc upper than 2-3%.
Ok, I'll search literature, using "HILIC" as the key word.
As for messages, Yandex is russian server, and once one had troubles when sending messages from Germany / In numer of other cases it was all right.

Constantine,

While I do not understand what you mean by swallowing, I can guarantee you that silica columns perform very well with these mobile phases. Otherwise, we would not have introduced a product line based on this.

There is a lot of literature on the use of silica with aqueous mobile phases, both old and new. The retention mechanisms are understood. In addition, there is a renewed interest on this due to the improved sensitivity in LC/MS applications.

Best regards
Uwe