removing/avoiding dissolved silica when extractingTLC plates

[PaulD]

RE: Any clues or hints to removing or avoiding dissolved silica from TLC plate extracts?

We analyze anionic oligosaccharides that need aquous buffers for good resolution - typically silica gel G TLC plates developed with n-butanol/acetic acid/water in 3:1:1 to 1:1:1: ratios give great resolution of the various size sugars as well as some epimers. We do semi-preparative clean-up of some compounds by eluting the silica w/ target bands with water after air-drying the Al-backed plate ...we cannot use straight MeOH, EtOH or non-polar solvents to extract these sugars for solubility reasons. However, we typically get some silica residue in our final extracted samples (even though extensively centrifuged before lyophilization) & sometimes this hurts our enzyme-based steps later on.

As a 'semi-solution to the problem', we have added a Na-based buffer (phosphate or Cl salt) to resuspend the dried residues hoping that insoluble Na silicate would be formed which we could remove later by centrifugation, filtration...it works to a certain degree, but sometimes we need salt-free conditions, or its effectiveness is sporadic etc. (The amount of silica removed may be controlled by how long we dry the plate > the water extraction may be more acidic if not as dry, etc, but not tracked down).

We are sure the insoluble residue is not our sugars & our solvents are HPLC grade. Did not if Si by elemental test, etc

Any clues or hints to removing or avoiding dissolved silica?

Thank you for your time & consideration,

All the best,

Paul

[Uwe Neue]

If indeed the problem is a dissolution of the silica, you can minimize it my going to slightly acidic pH when you do the reextraction, maybe with some acetic acid...

[Bruce Hamilton]

Assuming that you have confirmed it's dissolved silica, rather than suspended fine particulates ( 0.2um filter ), binding agent, or reaction with aluminium backing ( try some glass plates ), you could also consider the duration and agitation of your re-extraction.

Techniques that include ultrasonication or heat are likely to increase the dissolution of silica over time. You should minimise the aggressiveness and duration of the re-extraction.

Also, it may be worth trying another brand or type of TLC plate, and talking to the manufacturer of your plates about the issue.

Please keep having fun

Bruce Hamilton

[PaulD]

Thanks for the thoughts thus far, but:

a) water during extraction step should be on acidic-side since TLC plate, after development in an AcOOH solvent, is just air-dried mildly - sure it would be vinegary-pungent if took a big whiff! (but could spike water w/ AcOOH, too, for good measure) ,

b) never too aggressive extracting silica - often just cut the Al backing as a strip w/ prep target band, put in a tube, add water for 3-8 min contact time w/ rolling-around action, pour off water (we know most target compound comes off since has a color or is fluorescent) - no heating or ultrasonication, & collect the liquid extract. Centrifuge (15,000xg, 4-7 min)to clarify & lyophilize the supernatant....often some insoluble particulates from re-suspension, but we re-centrifuge again....we know more silica (or substance x) there because when we add buffer or protein solution, we often get a new cloud of inorganic debri formation, &

c) always use Merk plates, I thought were good - not using Wal-Mart brand, etc....

any more ideas?

[Bruce Hamilton]

In my opinion, it's unlikely to be silica from the plates with that mild extraction. I'd still try a 0.2um filter, just to see if the material is soluble or insoluble.

Have you performed blanks by taking two other portions of your developed plates, one from above solvent front, and one below the solvent front ( with no obvious compounds present ), and extracted those?.

If you have, and the deposit is present in both cases, I'd be inclined to try glass plates or send the precipitate for analysis. A small amount ignited in a crucible would also give an indication if inorganic salts are present.

Please keep having fun,

Bruce Hamilton

[HW Mueller]

We have done 100reds of glass prep. plate extrations without even a hint of silica dissolution. If silica dissolved so easily that you could see a precipitate (under the conditions described) silica would be quite useless. My guess is that you are doing inappropriate filtration/centrifugation, or that you are introducung the precipitate somehow. The check mentioned by Bruce, namely to extract a blank portion is essential. Merck might give you info under what conditions the binder could cause problems. We have not seen any here either, though it´s imaginable that some extreme pH could be detrimental?