Chemistry Lab Demonstrations: Candy Chromatography

*For more cool stories, pictures, and videos of chemistry demonstrations, click here*

Last lab of the semester today. Next week is the lab final and checkout. This week the students practiced column chromatography. They purified their crude product mixture from last week’s nitration lab. I’ve talked about the theory behind column chromatography before, so I won’t rehash it here in any detail. Suffice it to say that different organic compounds have differing affinities for a stationary phase versus a mobile phase. These differing affinities allow for one compound of interest to be separated from a mixture through the use of column chromatography. Students were aided this week in that their product was bright yellow. They could physically watch it run down the column, then only collect the yellow fractions.

Last lab of the semester means last demo of the semester.  This one’s a do-it-yourself demo, if you’d like.  You can separate the colors contained in M&M shells (or Skittles, or Reese’s Pieces, or Sharpies, etc) through chromatography.  I got my M&M proceedure here.  If you’re interested, other proceedures are available here, and here.   Basic rundown: put drops of water on wax paper, and put a piece of candy on each drop.  Allow for the water to strip the color off the colorful candy shell.  Cut a coffee filter into a rectangle.  Use a toothpick to spot each color onto the coffee filter.  Put the coffee filter into a 1% solution of table salt and allow the water to rise through the coffee filter.  Watch the colors separate like magic!

Couple’a observations I noticed.  Quite interestingly… the stationary phase matters.  A lot.  I started by spotting the colors on my silica gel TLC plates .  I was quite disappointed because the red and yellow both travelled with the solvent front and there was little separation.  I tried several different solvents… no luck.  I also noticed that according to the websites I was looking at, red should have travel the shortest distance.  Then I switched over to filter paper, and all of a sudden I got the results I was expecting.  Who knew?  Also, you should put a crease in the coffee filter before placing it in the solvent.  The paper will start to buckle and it will droop and fall over if it is not creased first.  The more distance you give the colors to separate, the better the results.  I used the largest filter paper we had, and ran the chromatograph twice to get the results shown.

Pop quiz, hot shot: Do you know what the difference between Red 40 and Red 40 Lake are?  I didn’t either.  Turns out… nothing.  At least, not as far as the compound responsible for the hue is concerned.  It’s all in the formulation:

Color additives are available for use in food as either “dyes” or “lakes”.

Dyes dissolve in water, but are not soluble in oil. Dyes are manufactured as powders, granules, liquids or other special purpose forms. They can be used in beverages, dry mixes, baked goods, confections, dairy products, pet foods and a variety of other products. Dyes also have side effects which lakes do not, including the fact that large amounts of dyes ingested can color stools.

Lakes are the combination of dyes and insoluble material. Lakes tint by dispersion. Lakes are not oil soluble, but are oil dispersible. Lakes are more stable than dyes and are ideal for coloring products containing fats and oils or items lacking sufficient moisture to dissolve dyes. Typical uses include coated tablets, cake and donut mixes, hard candies and chewing gums, lipsticks, soaps, shampoos, talc, etc.

There are 5 food coloring agents in M&Ms: Red 40, Yellow 5, Yellow 6, Blue 1, and Blue 2.  As you might expect, green separates into blue and yellow, but surprising the red and yellow of the orange M&M do not separate.  Rather, there is one orange spot with a larger Rf than red.  Brown separates to blue, red and orange.   But it looks like the blue in the blue M&M is a different blue than the blue in the green and brown M&M.

I’ve got lots of pictures from my experience (click for larger).  Note how poorly silica works and how different the Rf’s are between silica and filter paper.  the video is of separating components of felt tip pens, but it’s also neat.

There are no more demos planned, since the lab course is over.  Hope you enjoyed my miniseries.




  1. Nice, I wonder why, if filter paper works good in some cases, we don’t have something on the lab consisting of a disposable paper column :).

    Seems like orange doesn’t consist of the yellow and red, since brown seperated itself in blue orange and red.

    • Seems like orange doesn’t consist of the yellow and red

      yeah… I noticed that. I wonder why that is. There’s only one red dye in the M&Ms, but 2 blues and 2 yellows. I wonder what the reason is behind not having the red and yellow separate… The brown does look like it might have a ring of yellow around the orange at the same Rf.

  2. Though I did hear about the use of filter paper as the stationary phase (especially the researchers working in the filed of natural products (Oils and fats), urs is really interesting. Thanx, for the detailed info.

  3. As a long time (nearly 30 years) practitioner of TLC techniques, I can tell you that there are many different stationary phases that may be used and a huge range of solvent mixtures. Your lack of sucess in separating some colorant mixtures vs others is more due to a lack of assuming the same solid/liquid phase combination would well work for all colorants.

    I once was doing some analytics work for a researcher that sent me a series of purified compounds which he had dilegently purified by column chromatography on silica. Micro analysis and several other properties seemed to suggest pure materials, but NMR, and MS seemed to indicate mixtures. Switch from Silica to Alumina on TLC, I was able to tease out the mixture into their component materials… There are a lot of solid supports available for TLC work. I have used both normal and reverse phase solid supports. You might try your experiment again sometime with colorless gelatin in a tube. Place the colorant on the top and just let them sit for a week or two. This is an interesting variation. Add an electrode at the bottom and top and apply a small DC electrical potential and you might see some interesting thing as well. (basis of electrophoresis).

    One last thing you may want to note. I assume you are a graduate student. TLC is becoming a lost art and many graduate students, and industry researchers have become too dependent of HPLC and LC/MS. Good techniques but probably over used, and expensive habits to feed (Acetonitrile shortage comes to mind). TLC is cheap and with experence and some simple guidlines can lead to quick identifcation of mixtures, and separation conditions that can be used directly on bulk materials via flash and medium pressure chromatography.

    I am glad you find it so enjoyable to teach labs, and find interesting demonstrations to engage your students. I loved to teach laboratory technique, and this continued as I worked in industry. Some pretty well educated well connected PhD’s actually learned a thing or two from this older BS degreed lab rat.

    • Your lack of sucess in separating some colorant mixtures vs others is more due to a lack of assuming the same solid/liquid phase combination would well work for all colorants.

      quite right… the dyes are all anions (except blue 2 which is neutral and probably why it did retain on silica) I did assume the colorants would behave on silica the same way my organic products behave on silica. This is clearly a false assumption. After realizing they’re anions, it made a bit more sense. I wonder what would have happened if I’d protonated each dye before I ran the TLC…

      More often than not, when I have trouble separating things by TLC (which means trouble separating them by column) the first thing I try is different solvents. Ethyl acetate:hexanes is always the standard, but if that doesn’t work, I’ll usually turn to ether:hexanes first. I do have one column I’ve been running lately that works best in an acetone:hexanes solvent system. I’ll sometimes deactivate the silica with triethyl amine, but I rarely change the stationary phase for my separations.

      I am a grad student. I don’t have a HPLC or LC/MS at my disposal that would rival the triviality of running a TLC. So I’m actually less familiar with optimizing HPLC parameters than TLC. I’ve run a few 2D TLCs in my day, along with extensive solvent screens, but again, I don’t usually mess with stationary phase – probably because we buy the silica gel, glass-backed TLC plates in bulk and I’d have to make a plate with another stationary phase by hand.

      • How do you deactivate your silica with tea? And what are the effects? I’ve never done this before.

        • This probably goes to answering your other question, below, too.

          Silica gel is acidic. In the presence of acid-labile groups, reactions can take place on silica. Silica gel is notorious for knocking off extremely acid-sensitive protecting groups on occasion.

          If this happens, you can deactivate the silica by doping the eluent with TEA. This will buffer the acidity of the silica gel and your acid-labile groups stand a better chance of surviving column chromatography.

          To do so, simply add ~3% TEA or so to your solvent mixture before you pack your column.

  4. Sander – with regards to your comment about disposible paper column. The problem may have more to do with loading capacity. Many years ago one use to be able to buy rolls of chromatography paper for use in TLC. We use to suspend the paper from clips in the development tank. I have seen the use of clear gelatin in a column, and have used Silica, Alumina, and Florosil, as solid phases myself. One may also modify the separation properties of of these materials by chemical modification of their surfaces or by simple exposure of these materials to dilute water, bases, and acids in solvents, drying them and then using them in separations.

    • “One may also modify the separation properties of of these materials by chemical modification of their surfaces or by simple exposure of these materials to dilute water, bases, and acids in solvents, drying them and then using them in separations.”

      Any examples? 🙂 would like to learn something about this.

  5. Is it the caffeine in tea that fixes the silica plate, or will herbal teas also work? I would prefer to work with green or herbal tea. They’re healthier.


  6. Axmanam,
    Before reading your post, I had never heard of using candy for chromotraphy. I think it is a great idea. Thank you for clarifying how you had some glitches in the procedure, which seemed very helpful. Using wax paper and coffee filters did seem a little odd, but maybe the author decided to use materials you can find at home. I have always had good luck with filter paper and even paper towels work well sometimes. Another hint you might try is to fold/roll a piece of filter paper and stick into the center of the filter paper you have the dye on. Then place the rolled filter paper in the solvent. This allows only the rolled filter paper to get wet and helps prevent the dye filter paper to get soaked. Overall, thanks for the great procedure (I am planning on trying it sometime in my Chemistry class) and thanks for the great advice!

    • Glad you found it useful. I’m sure the authors did intend the setup to be “home chemist”-friendly. The stiffness of the filter paper definitely helped, so if you’ve got it, I’d use it 🙂

  7. can i use a laboratory filter paper as my coffee filter. I think its cheaper to buy laboratory paper than that of pre-cut coffee filter paper. thanks

    • yes, you may. In fact, good filter paper is thicker and sturdier than coffee filters. DIY chemists at home may not have access to lab filter paper though.

  8. As long as it is not contaminated, it should work. I am not sure about the quality of your coffee though. I would think the lab filter is less porous than the regular coffee filter.

  9. Pingback: Experiment 3: TLC | Organic Chemistry

Leave a Reply

Your email address will not be published. Required fields are marked *