Org Prep Daily

May 16, 2007

On Using Glass TLC plates

Filed under: procedures — milkshake @ 2:07 pm

We buy Merck silica TLC glass plates (250um, with UV indicator), 20×20 cm. We get them from EMD through VWR, for about $90 per pack of 25 (after discount). I divide each plate into 3 strips of equal size (about 6.5 x20cm) and I cut the strips into individual TLCs as I momentarily need them, the width being determined by the number of reactions (fractions, standards) that I need to run on the TLC side-by-side.  

Some people hate cutting their glass-backed TLCs and buy pre-cut ones (expensive and one-size only) or use aluminum-backed plates (incompatible with several good TLC stains). I found that cutting the glass plates is not hard to do and that crappy, worn or damaged diamond cutter is the most common reason for frustrating results. For example, Chemglass sells acceptable diamond cutters for reasonable price. (One has to clean the grease off the tip with some organic solvent and fix the wiggly headpiece with diamond tip onto the handle  – by loosening the screw, rotating the handle and re-fastening it). Silicon-carbide disc “pizza-cutters” can be also pretty good – I have been sceptical about them but now I use a plate-cutting contraption that has Si-C disc in it and it works fine.

One common beginners mistake is to double-cut or push hard on the cutter. If the used diamond cutter is sharp, a hair-thin, nearly invisible cut made with one light slide works lot better than wide a trench dug in the glass through a screeching effort. Also, it is important to use some straight solid ruler that does not bend sideways, for cutting/breaking the plate. (A nice plastic one has a Y-shaped profile and marks with different scales.)  

If you find yourself next to Home Depot, you should buy a polished marble tile for your TLCs;  it is not expensive (something like $5 a piece). It serves as a perfectly flat surface with a sharp edge. Also, the marble (but not granite!) is quite soft – so your cutter slips off the glass plate, the impact on the tile will not damage the diamond tip. The tile is also a good place where to store couple of pre-cut TLC glass strips (facing down). But the tile has to be a real marble, not a glazed ceramic.

[I stole my marble tile at a construction site in Boston and I carried it with me ever since, through couple of jobs, to San Francisco and then back East again.]



    Great suppliers for all things glassworking.

    Comment by Uncle Al — May 16, 2007 @ 2:22 pm

  2. What tlc stains are incompatible with aluminium-backed plates?

    Comment by petr — May 16, 2007 @ 2:47 pm

  3. anything that is based on mineralization with a strong acid – if you heat-up your Al-backed TLC plate dipped in cerium-ammonium-vanadate stain or anisaldehyde stain, the present sulfuric acid will foam on the Al backing foil and the silica layer will peal off…

    Comment by milkshake — May 16, 2007 @ 2:50 pm

  4. Not to mention the truly ghetto nature of the Al-backed species- even with a sharp razor, there is almost always a slight curl that sends your spots running on the diagonal, which means you’ll have to rerun it, which well, negates the entire “cost effective” argument!

    Comment by Jose — May 16, 2007 @ 5:27 pm

  5. Tell me a good way to cut quartz slides and I’ll buy you a cake. The stuff is such a pain in the ass.

    Comment by excimer — May 16, 2007 @ 5:35 pm

  6. Ive never had trouble with silica peeling off alumininium plates. Im mostly a carbohydrate chemist at present, and I stain all my compounds by charring with 10% sulfuric acid with no problems. Glass or Al is just a matter of personal preference, and what you are used to working with. The secret to cutting the Al ones so the spots run straight is to score them, not cut, and gently bend along the score backwards and forwards until the metal breaks. Use a ruler, and press with the razor ( I use a Stanley knife) on the silica side as hard as you would to cut paper. Cutting a useable plate with scissors is almost impossible in my experience.

    Comment by Handles — May 16, 2007 @ 6:09 pm

  7. I don’t have any problems with scissori.. uhm, cutting Al plates with scissors. You can also cut the angles off. Most often we use permanganate stain, not the basic stuff + heat, just sat. aq. KMnO4, and P-Mo acid.

    Comment by liquidcarbon — May 16, 2007 @ 6:29 pm

  8. I’ve seen several people who had trouble with glass-backed plates because they were cutting the silica gel rather than the glass backing. Usually people in my lab use a sheet of paper (I like cardstock for this) on the benchtop as a cutting surface. Do you think that could be damaging the glasscutters? I’m not entirely sure what kind we’re using, but they have a little wheel in them.

    Comment by Ψ*Ψ — May 16, 2007 @ 9:10 pm

  9. I have never been able to run decent TLCs with Al plates cut with scissors.
    However this kind of paper cutter does miracles:

    You won’t believe it if you don’t try it.

    Comment by Spiro — May 16, 2007 @ 9:15 pm

  10. I have been using the same glass cutter for seven years. It cost $5 at the drug store.

    Comment by Milo — May 16, 2007 @ 10:13 pm

  11. Oh, I average a 97% cut rate (3 per 100 TLC plate don’t snap off correctly),so it is a good cutter 🙂

    Comment by Milo — May 16, 2007 @ 10:14 pm

  12. Just get one of these:

    We had one in my grad lab that’s going on 10 yrs old at this point. The scribers lasted about 2 yrs for a group of 30-40 chemists. I don’t know what they go for now, but then it was about $120 for the cutter and $20 per scriber, not exactly bargain basement, but considering the usage, a good deal.

    Not so great if you want to size plate on the fly, but serviceable if you want to cut a stock. We’ve got one in my post-doc lab as well, works great, haven’t had below 100% cut rate in months (I usually cut one 20×20 plate into one of three sizes: 1×5 (investigative), 2×5 (comparative/column scoping), and 5×5 (column/2D)).

    Comment by Zinc — May 17, 2007 @ 12:47 am

  13. I had issues also with alumina backed plates until I learned at Hatter said, the secret is to score them and them bend them until they break cleanly. As far a stains, the only one I’m having issues with is KMnO4, but I think that’s because I’m an idiot and can’t follow a recipe…. And on the subject of stains, Vanillin is incredible.. wonderful smell and a great stain… So far I’m having little luck with ‘Magic’ stain.. probably again a recipe issue… Anyone care to send me idiot proof directions for either? I’ve tried recipes found on the web to no avail… And I mean, first year grad student idiot proof….

    Comment by Javaslinger — May 17, 2007 @ 2:46 am

  14. whoops…

    Comment by Javaslinger — May 17, 2007 @ 2:47 am

  15. when you use the paper cutter to cut the Al backed plate, place the plate face down gently against the cutter instead of the other way around. use one stroke instead of back and forth.

    Comment by pc — May 17, 2007 @ 11:03 am

  16. Jslinger-

    Seebach’s Stain, aka “Magic”

    2.5 g PMA
    1 g Ce(SO4)2
    6 mL conc H2SO4
    94 mL H2O

    Just like std PMA, it needs to age for a week or two before it works well (equilibria? solubility?)
    Funny aside, when I moved into my lab in grad school, there was a bottle labeled just “Magic”- I used it for a year until in ran out, and then I spent hours walking around the dept until I could find anyone who knew what the hell was in it! Great stain- with that, permag, vanillin, and ninhydrin, you can cover 95% of everything-

    Comment by Jose — May 17, 2007 @ 12:18 pm

  17. Would it work with cerium nitrate instead of sulfate?

    Comment by liquidcarbon — May 17, 2007 @ 4:39 pm

  18. Hmmm, I’d check the solubility of both salts if you can find it- if they are pretty close, I’d suspect the stain would work. No guarantees though!

    Comment by Jose — May 17, 2007 @ 6:41 pm

  19. Al-backed plates are fine, i’ve only ever found one stain that destroyed the plate (“soviet spray”: spray with conc HClO4, evaporate on a heat plate, spray with malachite green/ammonium molybate). but i cut them as required off 20x5cm strips, with a knife+stainless steel ruler, so you don’t get curl. scissors are no good

    Comment by kiwi — May 17, 2007 @ 9:30 pm

  20. one thing i love about aromatic chemistry: no TLC plate staining

    Comment by Ψ*Ψ — May 18, 2007 @ 12:15 am

  21. #20-Totally agree, sometimes it will give beau-ti-ful shinny blue light under long wave too!

    Comment by Taitauwai — May 18, 2007 @ 5:02 am

  22. What’s the best method for recovering your product from the silica after running a prep TLC? I’ve already scraped the bands into separate containers.

    Comment by Javaslinger — May 22, 2007 @ 8:37 am

  23. I don’t know – maybe pouring the dry powder to a Pastuer pippette with a small plug of cotton and eluting the silica with some polar solvent mix, like chloroform-methanol 10:1 ?

    Comment by milkshake — May 22, 2007 @ 12:13 pm

  24. Here is what I have used to extract prepTLC bands.

    1) Scrape off the desired band using a flat eXacto blade.
    2) Use a flat stopper or round bottom flask and crush the silica into powder
    3) Elute with 15% MeOH/EtOAc. I use about 50-125 ml for 5-100 mg extracts.

    Works great with very good mass recovery.

    Comment by MKULTRA — October 30, 2007 @ 4:48 pm

  25. On the subject of TLC carrier material, we almost exclusively use aluminium plates in our workgroup. We do have some glass plates, but everybody is just too lazy to cut them. You can usually cut the aluminium plates with good scissors, although sometimes bits will break off. It can help if you gently bend back the edges if they didn’t come out flat from cutting, and/or if you cut the corners off of the bottom of the plate. I have no idea why, but I have the “superstition” that if you open a fresh box of TLC plates they get cut better than if you use some that have been laying open (as in, not sealed anymore) for a few months.

    As for stains, we generally use the PMA and Seebach stain, as well as the pink peroxide stain (because we happen to work with peroxides a lot).
    I have a question though concerning stains: A while ago I tried to stain some rather simple lactones with PMA or Seebach’s reagent, without much luck. I tried to run e-caprolactone that we still had for a test and noticed that i needed a rather high amount to even get a faint spot from it on TLC. Also, simple ketones like for example cyclohexanone don’t really work well with these stains. Do you have an idea of a good stain that will make simple lactones visible, without having to use high concentrations of them? (if you have a stain for things like the described cyclohexanone too that would be nice as well, but the lactones are important for me right now) I also tried a stain specific for lactones, consisting of hydroxylamine and iron chloride, among other things, but either I didn’t prepare this one right or it only worked in high concentrations as well.

    Another question I have about UV absorption: The compounds I use usually don’t absorb UV very much. Sometimes they do have for example a phenyl ring somewhere and absorb uv light. I have made the experience that when you change a group that I (with my limited experience) consider unrelated to the UV absorbing group (like a keto group that is 2-4 C atoms away from the phenyl to an acetal group) the UV absorption can change quite drastically. Do you have any explanation (and maybe also a prediction) for that?
    I dunno if you will still look into this, but if you reply, thanks in advance anyway 🙂

    Comment by Xing — October 25, 2008 @ 8:29 am

  26. Cyclohexanone is volatile so you will have difficulty with any detection system that requires heating to work. I would use 2,4-dinitrophenylhydrazine stain (a saturated solution in 1% HCl)

    For lactone detection you can try a first spray with a solution of hydroxylamine (free base) in methanol, followed after while by diluted solution of Fe(III) salt, the produced hydroxamic acids stain pink with Fe(III). I have no experience with this system.

    Comment by milkshake — October 27, 2008 @ 4:15 pm

  27. We have a pretty wide selections of stains…but most commenly used is permanganate…shows up pretty much everything.

    Comment by Jay — December 2, 2008 @ 6:01 pm

  28. I have cut glass backed TLC plates and pulled TLC spotters out of used pasteur pipets for years and years as long I was in the US. I became quite an expert at doing both.

    My favorite stain is vanillin – beautiful smell, though chemistry wise, I like the CAM (magic) stain. KMnO4 works great. My only issue is that on dipping the plate in the KMnO4 solution, I used to wipe it in a paper towel. After drying, the towel used to look like a soiled diaper!

    Comment by Vinod Jairaj — March 31, 2009 @ 7:35 am

  29. Before you’ll recover your product from the silica after running a prep TLC, which stain (of course beside UV light) would be non invasive? So I have some lactone with amide and ketone group, which is not visible in UV and I have to purify it by preparative TLC, because I have very small amount…

    Comment by AHL — January 12, 2011 @ 9:36 am

    • unfortunately this is a difficult problem. If your compound can survive iodine vapor chamber it would be probably the way to go (but I would still use iodine chamber for the shortest possible time so as to not to have the purified product full of iodine). Most other methods of staining are far more destructive.

      Also if your compound gets detected by basic KMnO4 stain (without heating) maybe with a razor blade you can scrape in a tiny detection corridor border on the left and right side of the plate and develop it in such a way that the stain does not spread past this margin, that would tell you the position of the band – but it sacrifices a part of the material…

      One more crazy idea: perhaps you can also try toluene or xylene vapor chamber – xylene vapors will saturate the plate and should completely dim its UV fluorescence. Once out of chamber, the plate fluorescence will get restored over time as xylene evaporates away but xylene will be retained in the zone with compound for a longer time so the zone with a compound should stay dim under UV for a longer time.

      Comment by milkshake — January 12, 2011 @ 5:21 pm

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Blog at

%d bloggers like this: