Potassium Chlorate from Bleach

This post serves as the Video Companion to this video.

This experiment was also featured on Hack a Day!

In this simple experiment you can create potassium chlorate, a powerful oxidizer that finds use in amateur rocketry, a convenient source of oxygen, and the famous "screaming gummy bear" demo (among other things), from common household items with a minimum of effort. The tradeoff is that it's a very inefficient process and yields tend to be very low. Electrolysis is a far superior method, and is something I plan on trying out in the future.

To begin, I measured out 500mL of household bleach. This is about the minimum for this method to produce reasonable (but still quite low) yields. The active ingredient in bleach is sodium hypochlorite, and the higher the concentration in yours the better. Mine was listed as 8.25%.

The good news for chemists is that when shopping for household chemicals to use in experiments, the cheaper the better! The cheapest brands generally do not include extra fancy chemicals like detergents and fragrances, which is desirable for chemistry use.

This solution was then boiled on a hotplate to get rid of most of the water, and to drive the conversion of hypochlorite to chlorate (via a disproportionation reaction, with table salt as a byproduct):
3NaClO == 2NaCl + NaClO₃
This only occurs on boiling, so slow evaporation will not work in this experiment. Once crystals start to appear in the liquid, it is time to remove from heat and let cool.

These crystals are the sodium chloride byproduct, and need to be filtered off and can be discarded. The remaining solution contains sodium chlorate, as well as more sodium chloride.
Sodium chlorate is quite soluble, so to recover it and separate it from the sodium chloride it must be converted to potassium chlorate. This can be done with another common household chemical: sodium-free salt alternative. Mine was No Salt brand, which is almost entirely KCl.

For this experiment, we need enough KCl to convert all the sodium chlorate to potassium chlorate. This happens via a simple metathesis reaction:
NaClO₃ + KCl == KClO₃ + NaCl
Using 500mL of an 8.25% solution of sodium hypochlorite and the two equations above, I calculated that this reaction should only require 13.3g of KCl to go to completion. Other experimenters have recommended using a saturated solution of KCl that is equal in volume to the boiled bleach solution, so that is what I went with this time. Since my bleach ended up at about 200mL, I needed about 75g of KCl to make a saturated solution. The extra ingredients generally leave a solution of No Salt rather cloudy, so I always filter it before use.

Next I combined the solutions together. This should precipitate KClO₃ crystals immediately, but I did not see any at all! I think the culprit was using that huge excess of KCl solution. This means that there is a lot of extra water around for the products to dissolve in, so even though potassium chlorate is much less soluble there was still enough solution for it to dissolve in. I believe that if I had used the stoichiometric amount of KCl in the minimum amount of water, 13.3g in about 39mL of water at room temperature, I think I would have seen immediate crystallization. Certainly there is room for future experimentation here.
To get crystallization, I put the solution in the lab fridge over night. When I took the beaker out, there was a nice layer of white, plate-like crystals covering the bottom. I was able to recover 9 grams of product this way.

 22 grams of KClO₃ was the theoretical yield here, meaning I had only a 41% recovery. In an attempt to get more yield, I went ahead and boiled the remaining solution down to half its volume and placed it back in the fridge. I ended up with 23.5g of additional product, which by itself is greater than the maximum possible yield! That means that this second crop is very impure; perhaps I shouldn't have reduced the volume quite as much. These crystals looked much different too - they were much more compact, and some were needle-like rather than the nice plates from the first batch. Here is a comparison of the two, with the impure product on the right:

There's still a chance to recover good product from this second batch, and it would require a recrystallization or two. I have yet to try this, but if I do I will update the post. It honestly may not be worth the effort, considering how cheap the reagents were and how simple the process is.

Finally, I wanted to test the product to make sure it was in fact potassium chlorate. There is a simple and exciting test for this - combine some chlorate with half its weight of sugar, and add a drop or two of concentrated sulfuric acid. The mixture crackles and quickly bursts into beautiful lilac flames, making a lot of smoke as well. I tested 1 gram of my pure product combined with 0.5g of sugar, and two drops of acid.

See the video link at the top to see this reaction in action. It appears near the end.