Silica and 100% Water?

[Alireza]

Hello everyone,
This is my first post, so please be patient with me!
I'm trying some TLC (silica) running with pure distilled water and it seems to work with my samples, I know that it may seems a little bit strange based on chrom. basics. This approach is mainly due to the fact the we want to emphasize on not using any organic solvent in the process and also keep the costs down ( well, see if we can).
I'm looking at a very polar fraction of the biomass that I'm working with and it seems it works as I have my polar fraction on the Rf=0.8. I elutemy fraction from silica band by scraping off silica and washing with water again and analyze the eluent with RP-HPLC or mass spec. So the tlc performs a kind of pretreatment fo my sample and that's what I expect from it.

My questions are:

1- What is your general idea about this? ( any hidden problem that I may don't know, unreliable or unacceptable results, loss of silica, difficulty in later steps..)

2- I'm goign to scale-up the process to silica column chromatography using same basics. is the silica is going to show the same results? Is it possible to prepare the and pre-elute the silica gel with pure water to pack the column? ( any other suggestion please let me know)

Thanks in advance

[Uwe Neue]

Your approach is not common, but it is fundamentally OK. Silica has some hydrophobic properties, but what is most likely the key ingredient in a fully aqueous mobile phase is the fact that it is a weak ion exchanger.

When you convert from TLC to a column, you may get differences, unless the same silica is used. There are differences between different silicas which depend on a range of different parameters. The simple thing is to just try, and worry about these things only, if it is not as straightforward as anticipated.

[Alireza]

Your approach is not common, but it is fundamentally OK. Silica has some hydrophobic properties, but what is most likely the key ingredient in a fully aqueous mobile phase is the fact that it is a weak ion exchanger.

When you convert from TLC to a column, you may get differences, unless the same silica is used. There are differences between different silicas which depend on a range of different parameters. The simple thing is to just try, and worry about these things only, if it is not as straightforward as anticipated.

then you suggest that I may have some ion exchange instead of pure absorption, well I did not know that, thanks. But what charged groups are involved in that?

[Uwe Neue]

Silanols are weak acids that would interact with basic functional groups of your analytes, such as amine functions. Silica also has Si-O-Si siloxane bridges, which have a weak hydrophobicity.

[Alireza]

in that case, could we assuem that the silane ligands would have charge in every application at neutral pH and with some water?

[Uwe Neue]

The pKa of silanols varies with the type of silica. But yes, at neutral pH, I expect that a lot of silanol groups are negatively charged.

[Alireza]

Thanks again, I've also heard about solubilizing of some of the binders in case of using water. is it true and if yes how can affect the operation?

[Uwe Neue]

If your TLC application works, don't worry about the binder. In a packed column, you do not have a binder. If the binder influences your application, you may get differences between the column and the TLC plate.

[Victor]

If this procedure works for you, then fine. However, it is possible that you may have problems with the reproducibility of your procedure. If you have ionic retention processes, these will be governed by the pH of the "solvent". Distilled water is unbuffered, and its pH is indeterminate (usually slightly acidic) but might be influenced by the sample itself or even the source of the distilled water. However, if your samples are very consistent and the quality of your distilled water is also consistent, things may still be ok. Changing the amount of sample (e.g. you talk about using a prep procedure) might just upset things though.

[Alireza]

If this procedure works for you, then fine. However, it is possible that you may have problems with the reproducibility of your procedure. If you have ionic retention processes, these will be governed by the pH of the "solvent". Distilled water is unbuffered, and its pH is indeterminate (usually slightly acidic) but might be influenced by the sample itself or even the source of the distilled water. However, if your samples are very consistent and the quality of your distilled water is also consistent, things may still be ok. Changing the amount of sample (e.g. you talk about using a prep procedure) might just upset things though.

Thanks to you victor as well, you're right. I'm thinking of the reproducibility may be an issue though I'm looking in a range that I think the main interaction is the adsorption, as I take my fraction not very far from the front ( in tlc of course) . However, using a buffer may resolve the issue I guess, doesn't it?
The other thing is the working life of silica ( or silanol activity) in case of using pure water, and especially its use at elevated temperatures. any idea?

[Uwe Neue]

Silica does have a limited solubility in water. However, at the particle size that you are considering, I am not very concerned about this.

If you set up a buffer, you will get better control of retention, and of dissolution. My suggestion would be to go to pH 3 or 4, if your separation will tolerate this.