Chlorates

1. Background
2. Motivation
3. Chlorate Cell
4. Recrystallization of Potassium Chlorate
5. Run Log

Background

In 1787, Frenchman Claude Louis Berthollet became the first person to make potassium chlorate by bubbling chlorine (isolated just 13 years prior by Scheele) into a solution of potassium hydroxide. Since its discovery, Berthollet's salt, as it is sometimes called, has been the most common and useful of the chlorates and is now prepared on an industrial scale electrochemically. It is a strong oxidizer used in the production of matches, fireworks, dyes, disinfectants, and chlorine dioxide, an important bleaching agent.

All chlorates can be prepared by passing chlorine through alkali as Berthollet did, by thermally decomposing the hypochlorite, or by electrolyzing the chloride. The first method is generally regarded as the most inefficient and is also probably the worst method for the amateur chemist since large amounts of chlorine gas are required.

3Cl₂ + 6KOH --> KClO₃ + 5KCl + 3H₂O

It is best that the hydroxide solution be hot and concentrated as the disproportination of the hypochlorite is really taking place. Cold solutions, therefore, yield mostly hypochlorite. As is clearly shown in the above equation, five-sixths of the potash lye is converted to the less exciting potassium chloride. Worse yet, potassium hydroxide is a valuable chemical in its own right and is something that I do not want to waste. Hence, if I ever produce chlorates with this method, I will first make sodium chlorate with caustic soda and then convert it to the potassium form through a displacement reaction with potassium chloride. If I really feel like being cheap I can bubble chlorine into a suspension of lime giving me calcium chlorate which can be converted to potcrate in an analogous manner. Although very doable, I have much better options.

Instead of essentially synthesizing hypochlorite in situ to make the chlorate, one can start straight from it. Dilute solutions of sodium hypochlorite and impure solid calcium hypochlorite are sold as household bleach and pool chlorinating agents, respectively. The hypochlorite is boiled down until chloride begins to precipitate out. When sodium hypchlorite is used the following reaction occurs.

3NaClO --> 2NaCl + NaClO₃

A satatured solution of potassium chloride is then added to form a white precipitate of potassium chlorate. Now this method is still a bit unresourceful because two-thirds of your starting material is converted to chloride, but at least you do not have to waste precious hydroxide ion. Also, most household bleaches are no more than 6% NaClO by weight with a good deal of NaCl also thrown in there so you can go through a lot of bleach quickly.

The seemingly most elegant way to reach the chlorate ion is to oxidize the chloride ion with electricity. The reaction mechanisms for this process are quite complex and are still the subject of much debate. They can be best summarized by the following diagram.

It takes six electrons to oxidize six chloride ions to three molecules of chlorine gas. These molecules immediately dissolve and react with water to form the three hypochlorite ions. These three ions can then disproportionate (autoxidize) to form one chlorate ion and two chloride ions. However, when near the anode, they can also receive three more electrons and react with water to form the same products plus oxygen. These two underlying mechanisms are in constant competition and since one requires nine electrons while the other requires six electrons to produce one chlorate ion, it is advantageous to promote the latter. In industry, this can be down by ensuring a rapid flow rate of the solution and minimizing the immersion depth of the electrodes.

To some extent, chlorate is also reduced at the cathode and this is prevented by adding a small amount of dichromate or fluoride ion to the solution. Since I do not currently have easy access to either of these chemicals, I will not use them and this mereley means that my cell will be less efficient.

Throughout operation of the cell, some chlorine gas will escape from solution, resulting in the formation of hydroxide ion. This increases the pH of the cell, and should be readjusted to a value of about 6 with hydrochloric acid to maintain maximum efficiency.

A variety of materials can be used as electrodes, but they must not react with the solution they are in. For both electrodes, platinum is of course the best option as it hardly corrodes at all, but it is way too expensive. Steel, graphite, copper, and a host of other materials can be used as cathodes because the metals are protected from oxidation by the surrounding negative charge. Unsubmerged parts of the electrode, however, can react with the nascent chlorine gas, but these impurities are largely removed during the filtration process. Choices for an adequate anode are much more limited. Graphite is the most common choice for the amateur because it is most attainable, but it corrodes rather fast so it must be replaced frequently. Lead dioxide is the second most popular choice and is much more resistant to corrosion, but you have to deal with nasty lead compounds and figure out a way to mold the material onto a substrate. Iron (II,III) oxide and maganese (IV) oxide are rarely used and have similar problems. Industrial cells now typically use dimensionally stable anodes, which most frequently are titanium rods coated with a thin layer of ruthenium (IV) oxide. Since titanium by itself does not suffice, this route is out of the question for almost all home operations. Plastics like polypropylene and polyvinyl chloride are common materials that are suitable for the container of the cell.

The temperature of most cells runs anywhere from 60-80C. As temperature increases, resistance decreases thereby increasing current flow, but the anode corrosion rate increases. Graphite rods corrode so quickly that they are usually operated around 40-50C. At temperatures higher than 80C, the vapor pressure of water becomes too high and evaporation of the solution and chlorine desorption are promoted.

The required amount of electricity that must go through the cell to convert a certain amount of chloride to chlorate can be calculated, assuming 100% efficiency. However, not all the chloride can be converted to chlorate because at low chloride concentrations (about 10% w/w), chlorate ions begin oxidizing to perchlorate ions when compatible anodes are used. When a graphite anode is used, it begins to disintegrate at a very rapid rate and reduces any perchlorate formed.

Once a solution of mostly sodium chlorate is formed and electrolysis is completed, the solution is filtered to remove electrode corrosion products. It is then boiled to destroy residual hypochlorite and a saturated solution of potassium chloride is added to precipitate potassium chlorate. The solution is cooled and the raw potcrate can be washed and furthered purified by recrystallization.

Motivation

Potassium chlorate has quite a number of uses for the home experimenter as it is a strong oxidizer. When mixed with sugar, it forms a nice rocket fuel. This mixture can be magically ingited by a drop of concentrated sulfuric acid, which if completely free from sodium ion contamination, burns with a beautiful lilac flame. Molten potassium chlorate violently reacts with practically anything organic and the classic demonstration of dropping a gummy bear into a test tube of it heated to about 150C testifies to this. Potassium chlorate can also be used as an oxidizer in organic chemistry. For example, a mixture of hydrochloric acid and potassium chlorate react with napthalene to form napthalene dichloride.

Chlorate Cell

I am happy to report that my first real shot at producing potassium chlorate was a success and proved to generate much more chlorate than the decomposition of bleach method which I had used twice before to produce no more than 10 grams. Power to the cell came from a computer power supply. Getting the supply working was pretty easy. All I had to do was bunch up all of the similar colored wires together and hook one green wire and one black wire up to a switch. The supply came installed with a 500mA fuse which I replaced with a 5A one. I ran 12 volts in the cell even though I knew this was excessive.

For the first 46.8 hours, I used the design pictured to the left. The electrolyte was made by dissolving 80 grams of table salt in 200mL of boiling water. I then let the solution cool down to room temperature and filtered the formed crystals. So my electrolyte probably consisted of about 70g of salt in 200mL of water given that at 25C, 35.9g of salt dissolve in 100mL of water. Throughout the entire run, graphite rods from lantern batteries were used as the anode and a stainless steel three inch long screw was used as the cathode. The electrodes in this design were about five centimeters apart, but I moved them closer occasionally when the current dropped. I put the whole thing in a large cold water bath to regulate the current and keep the temperature around 100F so that the graphite anode would not corrode too quickly. I had a straw in there to vent away exit gasses and this acted to some extent as a reflux condenser so that too much water did not evaporate. I also had a metal thermometer in there, but after a couple of days its stem corroded due to the action of chlorine gas and it now no longer works. Without the thermometer, I did not worry about regulating the temperature too much and as a result the anode sometimes eroded very quickly. After running the cell using the food container, I switched to a smaller prescription medicine container.

The cell was run for eight days, but not constantly because I had to stop it often to make repairs and modifications. A few times I ran the cell during night and this was not the best of ideas because I could not monitor the current at all. During these runs, the anode would completely shred leaving me with a final reading of something along the lines of 0.01 amps. Averaging the initial reading with this reading probably is not very accurate because at lower current levels, the anode erodes less quickly and thus the cell most likely operated at extremely low currents for the vast majority of these night runs. This is possibly one reason for my mediocre yield. 1-3 amps was generally put through the cell. In total it was run for 115.9 hours at an average of 2.11 amps, for a grand total of 9.14 faradays. Only 50g of the 70g of salt in the 200mL electrolyte could have potentially been transformed into chlorate because at this point the concentration of chloride ion would be at the 10% perchlorate threshold. It takes 6 moles of electrons to make one mole of chlorate so since 50g of salt is about 0.86 moles, this corresponds to about 5.13 faradays of theoretical needed charge. So that means that the amount of charge I put in corresponded to a 56% efficiency rate.

I initially was trying to improve cell efficiency by controling the pH, but it turned out to be too much of a pain and I did not want to mess up my pH meter by having it immersed in graphite sludge. Hence, I just guessed and occasionally squirted ten to fifteen drops of 0.6M hydrochloric acid into the cell once or twice a day.

By the end of the run, the cell was jet black with lots of graphite sludge in it. After electrolysis was completed, I filtered and carefully decanted the solution about four times, but this did not get rid of collodial graphite particles. The solution was yellow due to iron impurities from the cathode. I boiled the solution down a bit to destroy any hypochlorite. I then prepared a solution of 65g of potassium chloride in about 250mL of water. I added this to the solution and continued to boil down until I saw a crust begin to form. I then stopped boiling and covered the beaker with aluminum foil and let it cool to room temperature before placing it in the freezer. Needle-like crystals of what is presumably mostly potassium chlorate grew out of the yellow-brown solution.

Next, I filtered and mashed the yield up and set it to dry in the toaster oven for about an hour and a half. The powder is yellow-green and I suppose these discolorations are due to the iron and chromium impurities in the stainless steel cathode. I tested some of the product with silver nitrate solution and a definite precipitate of silver chloride formed. The chlorate is obviously rather impure, and later I should recrystallize it until it burns with a lilac flame. The total impure yield was 48.2 grams which is a 46% yield. This is not too good of a percentage yield but now I have a lot of chlorate on hand. My present purposes do not require that I use too much of it so I am more than content with these results.

Recrystallization of Potassium Chlorate

I recrystallized some of my potassium chlorate today and washed what precipitated out with water. During boiling, the solution was olive green and even more graphite particulates came out. After drying, I obtained a nearly perfect white powder with only the slightest tint of green in some areas. The test with silver chloride tested positive for chloride ions but now the potassium chlorate burns with a purple flame.

Run Log