Note: These experiments are performed over 10 years ago (end of 2000). My new experiments which I am performing now will be listed soon.
Burning sugar mixed with KClO3. Pictures taken with 0.5 seconds
Do not perform such experiments, unless you are an experienced chemist.
Potassium chlorate (KClO3) is a strong oxidizer. Do not
heat or rub it with combustibles, like carbon, sugar, etc. With Phosphorus
or even sulphur in may ignite easily. Avoid contact with acids as
R 9,20/22; S 13 16 27
Potassium chlorate KClO3 (also called Potassii Chloras or
Kalii Chloras or Chloras Kalicus/Potassicus) is the potassium salt
of chloric acid and is an oxidizer, used in chemistry to produce oxygen, but
also used in fireworks. It is obtainable at chemicals supply shops, such as Merck
or other large suppliers. It used to be available at the local chemist but this
is getting harder and harder. With rather simple means you can make this
yourself by electrolysis. Due to its poor solubility of 5 % in cold water,
compared to sodium chlorate and chlorides, it is to prepare rather purely.
item to be needed
where to obtain
Jar or beaker of about 0.5 liters
This will be the electrolysis cell
available in household
0.5 - 1 liter Beaker (preferably made of Schott Duran or compatible
Needed for heating the solution
Lab supply shop or ebay
A stainless steel pan may also be OK.
Funnel and fine filters
to filter out the insoluble substances and carbon anode dust
The fine lab filters (such as Mackerey-Nagel MN 640d) are best,
they cost no more than $10 per 100 in lab supply stores
When coffee filters are used, use at least three at at time.
Platinum anode, but due to the high price, carbon rod(s) are a good
alternative. Better are MMO anodes
Available in welding supply shops, MMO anode os ebay for $50
Dismantling a flat 4.5 battery may also do the job, but the
MnO2 inthe cells may cause hard to remove stains.
Laundry or other nonmetal (plastic or wood) clamps
To attach the electrodes to the jar, preventing
Household or hardwarestore
Stainless steel spoon or other strips
DC Power supply for 5-6 Volts and at least 5 Amps. A large PV
solar panel will do fine in sunlight. Note higher voltages have an adverse effect and corrode the anode more quickly.
Auto supply, electronics stores, computer repair shop or ebay
A cheap battery charger does the job very well, or better a computer power supply which supllies the required 5 V
DC Ammeter (when not already in the supply) with range of at least
the value the power supply can supply, e.g. 10 Amps.
Check the current.
Electronics store or eBay
Thermometer with 100oC range, preferably made of glass
or plastic. Electronic ones with a stainless steel sensor are OK as
The old power supply (an 8 Amp. car battery charger) was finally deceased. It was too unstable and drew almost the same power from the mains (110W) when running 5V 8A than the new PS with 5.5V 18A which consumes 130W.
Here I used the new setup using a 5V 40A power supply for $30 from Amazon. For the cell I used the same 600ml 'pindakaas' (peanut butter) jar with the MMO anode and the stainless steel cathode in it. I let it run for several hours.
I used to use a taller plastic jar but when reaching 100°C it became soft, so glass is a better option.
Finally I got 62 grams in the first session and 77 grams dry purified KClO3 in the next session.
In that latter session I used another power supply: a variable voltage 3-24V, 20A for $50 from Amazon. I started with 7V, 15-18A and ended up with 5.5V 16A. When the solution gets hotter, it conducts better. After the first two hours, it was 85-90°C.
These experiments I still did with the good old carbon rod method while I am waiting for delivery of the MMO anode I ordered. An extensive discussion on this subject (MMO anodes) is on the Sciencemadness forum. I perform all these experiments in the fumehood due to some escaping chlorine gas. When the process goes well not much Cl2 gas escapes, a mere 'swimming pool smell' is detectable inside the fumehood.
Starting with the same setuo as 10 years earlier using a 500ml "Calvé Pindakaas" peanut butter jar but now with the 16mm carbon rod and stainless steel sheet sealed into the lid and sealed with caulking agent (Yes it stinks to acetic acid CH3COOH during the caulking action...). And a PVC hose to release the gases (H2 and Cl2 and water vapor) into a water jar. After one hour of running at 10 A (voltage was lowering from 5.5 to 4.1 volts) the temperature of the bath raised to 60°C. The ammeter displayed on the picture is an 80/1 scaled ammeter so when it shows 800A it is actually 10A.
Next day I checked the pH with Mackerey-Nagel pH strips. I could estimate it is about seven so OK.However the strips look somewhat bleached due to the chlorine gas dissolved in the water. I dissolved some more JozoVitaal in the heated solution and poured it again into the peanut butter jar to start another session. I ran 9-10 amps again and the voltage was 3.6 volts. It seems that the higher the temperature, the lower the voltage.
After the two hours run I thought that the power supply deceased ..... no voltage was appearing from the supply and the fuse was OK. So I filtered the solution and some crystals appeared probably KClO3. Later on it turned out that the power supply worked again: a thermal safety turned off when the transformer overheated.
The following day I saw clear crystals on the bottom of the piss like solution. I put the glass container in the fridge and allowed it to cool further. Then while running a third session for 5 hours with newly added 'JozoVitaal' I poured off the solution and put the crystals ito a filter and rinsed them with ice-cold water to rinse off the remaining Na+ and Cl-, ad allowed it to dry for another re-crystallization to purify it.
I noticed that the anode wear is remarkably small, probably to the correct pH and presence of K2Cr2O7.
After a few more electrolysis sessions of 4 hours I collected all the (wet) crystals and dissolved in just ats much water that everyihing just dissolved at boiling. I let it cool outdoors (4°C) overnight and the next morning I poured off the still slightly yellowish liquid presumably NaClO3, KCl and NaCl dissolved. Then again I added som fresh water and again let it boil and then cooldown. After cooling I rinsed the crystals with cold water as the flame test after the first recrystallizaion turned out too sodium-yellow. After the second recrystallization and rinding again the flame color was OK. So I let it dry and scraped it out of the evaporation dish , rubbed it with mortar and pestle. Result 69g. I tested a little with sugar and with a lighter it catched (purplish pink) fire immediately. So is was OK.
Electrolyzing an alkali chloride solution results in the following reactions:
anode: 2 Cl- -> Cl2 + 2e
cathode: 2 H2O + 2e -> 2OH- + H2
Now the trick: In commercial plants, chlorine gas and caustic soda NaOH is
produced this way. A diaphragm should be put to prevent intermixing the
OH- and Cl2, otherwise the OH- will react with
the chlorine by:
2OH- + Cl2 -> 2 ClO- + H2
thus generating hypochlorite. The first target for making chlorate however is
hypochlorite, to be oxidized to chlorate, so this process must generate
hypochlorite and therefore mixing should take place.
When the solution heats up by heat loss (because the voltage is usually
higher than required to yield the needed electrochemical energy), the
hypochlorite will be oxidized to chlorate by:
3 ClO- -> ClO3- + 2 Cl-
and therefore the chloride ions will react again with the OH-. So
the total reaction (helped by the electric energy) is:
2Cl- + 3 H2O -> 2 ClO3- + 3
The oxidation state of the chlorine will be from -1 (Cl-) to +5
(ClO3-), which requires 6 electrons per Cl- ion.
Theoretically one mol electrons is equivalent by the physical constant of
Faraday which is 96560 Coulombs (Ampere-seconds), which is nearly 27
Ampere-hours per electron (mole). To oxidize one mole of Cl-
to ClO3- costs 6 * 27 = 162 Ah. For one mole KCl (74.5
grams) to KClO3 (122.5 grams) one needs 162 Ah in theory. In practice
it is more, estimate about 200 Ah.
An extensive explanation of the reactions taking place and the pH control is described in this document.
The voltage over the cell should be at best 3.5-4 volts with a temperature of 55-60°C. When the solution is cold it should be no more that 5-5.5 volts which will decrease when temperature rises.
The temperature is at best 50-70°C.
The pH shoud be slightly under 7. When you don't have a lab pH meter, a soil pH meter for gardeners is OK. Normally, the pH tends to shift in the more basic range above 8 which results in poor performance and more anode erosion. Add a knifetip of K2Cr2O7 per 500ml to the solution and if you don't have this, use a few drops of diluted HCl (available in the hardware store as 'muriatic acid' or 'Salzsäure' in German), but in the latter case check the pH again as it should get not too acid (i.e. under 6).
When you meet these requirements even carbon / graphite anodes will erode very slowly and therefor last long.
How to set up
NOTE: Perform the electrolysis outdoors or under a
well ventilated hood, because flammable hydrogen gas and some toxic chlorine
are generated. Use a fused power supply, to prevent fire at short-circuiting.
Most battery chargers have 10 or 20 Amp fuses.
Attach copper wire to the carbon rod, on a place which is not to be
immersed, as copper will be attacked otherwise. It should be tightly connected
as 10 amps will pass through it. Clamp it to the jar. Attach the stainless steel
cathode on the other side.
Dissolve around 100 grams of KCl (or diet salt which consists of about 50%
NaCl and 50 % KCl) in 0.3-0.6 liters of water and let it boil. Pour this into
the jar (carefully, as it will break otherwise at the quick temperature change).
Put the thermometer in the jar. Newer diet salt consists of even 67% KCl and 33% NaCl, which is even better.
Connect the anode with the red (+) clamp of the supply and the cathode with the
black (or white) (-) clamp. Set the ammeter in series with the cell (with
correct polarity). Set the power supply in the lowest possible current and check
whether everything is connected well and no short-circuiting can take place.
Run the electrolysis
Now turn on power. When the current is low, turn it up (e.g. by turning on
'quick charging' and/or 12 V instead of 6 V). NOTE: Check that the voltage with the load on does not exceed 6 volts, otherwise the anode might erode far more quickly, particularly when you are using an MMO anode. A computer power supply supplies 5 volts with 10-20 amperes. When you are using a car battery charger, you can use a power controller and lower this until the voltage is about 5 volts.
Watch the current and temperature. The current should not exceed the rating of
the supply and the best temperature is between 50 and 70oC.
Let it run for several hours at the optimum current and temperature, after
determining these by experience. Check the water level and, when needed, add
some water, as something evaporates. For 0.3-0.4 liter of solution about 240-360
Ampere-hours are needed, thus 24-26 h for 10 A continuous current. Note that the
carbon rod will erode and the solution will become black because of small carbon
particles.You can reduce this by keeping the voltage below 4.5 volts and the pH not in the basic range (i.e. keep is alightly below 7).
When you interrupt it (e.g. you do not want to let it run while you are not at home) remove the electrodes from the cell.
Collecting and cleaning the crystals
As KClO3 has a poor solubility in cold water, but a good one in hot
water, while other salts (like NaCl, KCl and NaClO3) dissolve much
better in cold water, one can let crystallize out the KClO3 and
obtain it rather purely.
Source: W. Ostwald: Grundlinien der Anorganischen Chemie (Fundamentals
of Inorganic Chemistry) Steinkopff Verlag Dresden, Germany, 1919.
After it is finished, pour the liquid in a beaker or stainless steel
pan which should be heated up to boiling point. Prepare another heat-resistant
beaker or pan with a filter in a funnel on it. When you have coffee filters, use
two or three filters at a time. When the liquid boils, pour it carefully in the
filter. A clear liquid should appear from the low end of the funnel. It can be
yellowish colored, because of oxidation of the iron in the cathode to yellow
Fe3+ by some chlorine. When all liquid is passed through the funnel,
let it cool down to as low as possible, possibly put it in the freezer or fridge
after it has air-cooled. Do not allow to let the solution to freeze, as the
container may crack.
Now large white crystals should appear on the bottom of the beaker. Pour gently
off the liquid into another container while holding back the crystals. The
liquid can be mixed with a fresh KCl solution for a new electolysis session.
Pour some ice-cold water (from the fridge or freezer) over the crystals (no more
volume than the crystals) to rinse off other salt solution.
Pour off any water and put the beaker with the crystals on a heat source and
heat is with a very soft flame. Allow any water to evaporate. When the
crystals are dry, scrape it from the beaker and collect it on an old newspaper
or a cardboard. Then put them together into a mortar and crush the crystals.
A good test for the purity is :
Flame test. Put an inert titanium or magnesia stick (stainless steel is also OK) in the KClO3 slurry after rinsing. Hold it in a colorless flame and no yellow Na color should appear. Even small traces of Na will color the flame yellow. The color should be lilac like.
Dissolve a few crystals in AgNO3 solution. Only a slight precipitation (or none at all) should appear. When there is more precipitation, too much chloride ions are in the solution.
You can recrystallize it by dissolving it in boiling water just enough water to dissolve everything. Then allow it to cool and put it into the fridge or freezer or outdoors in cold winter weather. After cooling pour off the water and dry the crystals. Now they should be more pure.
Note: Do not mix it with any combustible stuff
while rubbing it in a mortar,as a spontaneous reaction might occur !
The powder is the chlorate ready to use.
Last update: 2018 January 20
Powered, developed under Linux and Mac and hosted under Linux.
This site is best viewed with Firefox, Safari, Chrome and Opera. Using landscape mode on a smartphone gives a better view.