Perchlorate cleansing - SUCCESS!
Entry posted by Swede
4861 views
Do you like the new board better? Worse?
65 members have voted
This is going to be a long and rather dry blog entry, but I believe it to be important for those seeking the electrolytic production of potassium perchlorate, producing a product clean enough to be called "Pyro Ready."
At last, a real victory in this challenging task - taking perchlorate from a cell, and ridding it of remaining potassium chloride and chlorate. No batch of perchlorate can be considered pure until it is rid of ALL detectable chlorate. Chlorate remnants render the entire batch unusable in the sense that sulfur sensitivities, and the other drawbacks associated with chlorate, will remain... admittedly at a lower level. Of course, even the finest, purest, most expensive batch of commercial potassium perchlorate is going to have some PPB (Parts Per Billion) of chlorate, but it was decided, as a personal goal, to create perchlorate that tests as "clean" as the perchlorate purchased from suppliers like Thunder Valley and Skylighter.
In a previous blog, I determined that a test consisting of N-Phenylanthranilic Acid and Sulfuric Acid would detect chlorate to exceptionally low levels. Commercial perchlorate tested white, while even the most minute chlorate contamination (a pinhead in a liter of water) showed a distinct coloration, usually a muddy purple. The test is excellent and I consider it to be nearly infallible.
I took a small harvest from a Pt-anode perchlorate cell, which was initially stocked with potassium chlorate from the T-Cell, my giant chlorate cell. I did attempt a previous cleanup using sodium sulfite, but this undoubtedly introduced sodium into the product; not desirable for pyrotechnics. Based upon Tentacle's recommendation, I decided to try Potassium Metabisulfite to reduce residual chlorate, and also keep sodium out of the product.
20 grams of dirty perchlorate was dissolved in 100 ml distiled water, heated over a hot plate.
Theory: In an acid environment, potassium metabisulphite dissociates according to the following:
K2S2O5 + 2 H+ --> 2 K+ + H2O + 2 SO2
Of the products of the previous dissociation, the sulphur dioxide reduces the residual chlorate, according to this postulated process, thanks to Mumbles:
SO2 + H2O ---> 3 H2SO3 + ClO3 ---> Cl- + 3 SO4(2-)
The chlorate is definitely preferentially reduced by the sulfur dioxide gas
Five drops of concentrated HCl was added to acidify. From a known mass, potassium metabisulphite was added in portions, with the NPAA test conducted between each, using a separate test tube. Hot water was added as needed to keep the volume of the perchlorate at 100 ml.
After 9 small additions, the perc tested clean. Total mass of potassium metabisulphite used was 4.23 grams, which seems like quite a lot for a 20 gram batch of perchlorate, but which was probably due to poor technique rather than the need for a large mass of bisulfite. For example, one thing I noticed initially was that direct additions of the metabisulfite powder resulted in instantaneous bubbling of SO2 gas to the atmosphere. with little chance at contact. About halfway through, I went with slow additions of aqueous metabisulfite using a pipette held near the bottom of the beaker, and added quite slowly. This gave the gas maximum contact
The array of test tubes showed the gradual cleanup of the perchlorate.
The liquor was neutralized with KOH, allowed to cool slowly, and then refrigerated, to recover the potassium perchlorate. The remnants of the cleanup gave me a bit of concern. Looking at the equations, the reduction of the chlorate ions would ultimately yield KCl and K2SO4, both potential contaminants of the final product. I decided to see exactly what was in my perchlorate.
There are several ways to check for both chloride and sulfate ions present in an aqueous solution. Silver nitrate (AgNO3), when added to a solution with chloride, yields an unmistakable white precipitate. As for the sulfate ions, the addition of free Pb++ ions will produce a precipitate of insoluble lead sulfate, also white.
The test began... 5 grams of the "cleaned" perchlorate was dissolved in 300 ml of distilled water, with a bit of heating to help it along, as potassium perchlorate is not very soluble; 8 grams per liter at 0 degrees, and at best 200 grams per liter at 100 degrees C.
The first two test were for residual chloride. My Hach test strips showed no chloride, but this test was not very sensitive.
The next test was to use silver nitrate. The sample is on the left in the test tube, while the silver nitrate is on the right. A white precipitate indicates the presence of chloride ion.
The results:
To verify the sensitivity of the silver nitrate test, I added the remaining silver nitrate to tap water. There was an instant cloudiness; the left test tube is the perchlorate sample, while the right test tube contains tap water:
No chloride that I could see! The next step had me a bit more worried... the presence of significant K2SO4 in the perchlorate. I mixed a very small sample of saturated lead nitrate in distilled water, and added several drops to the test tube sample. Once again, there was NO visible precipitate, or discoloration of any kind.
Without expensive lab equipment, I was fairly confident that these simple tests confirmed that the perchlorate was of good quality. One final test was conducted...
The perchlorate was ground in a mortar and pestle, and a common perchlorate composition, with well-known characteristics, was carefully created, approx 1/2 gram.
This comp was tested by the simple expedient of burning about 1/10th of a gram at a time, in the open. I had the camera set up to capture MPEG, but something happened, and the movies did not turn out. Rest assured, it performed as expected.
To summarize: This small bit of potassium perchlorate was produced by the electrochemical oxidation of potassium chlorate. Cleanup of 20 grams took 4.23 grams of potassium metabisulfite, but I suspect at least 75% of the necessary SO2 gas was flashed to the atmosphere with the initial addition of the powder to the BOILING solution of perchlorate.
Subsequent additions were made with aqueous metabisulfite, rather than powder, by slowly pipetting the metabisulfite solution into a hot (but not boiling) perchlorate solution, at the bottom of the beaker. SO2 gas is quite soluble, but follows an inverse solubility, with more dissolving at cooler temperatures. This means there is a tradeoff... you need heat to dissolve potassium perchlorate, yet SO2 gas prefers cooler temps. Once again, the use of sodium vs. potassium salts may make this an easier process overall.
One other thing I noticed... stored potassium metabisulfite stinks a bit, indicating atmospheric moisture may cause some dissociation and release of sulfur dioxide gas, so keep your metabisulfite well-stored. I double-bagged mine and store it in a good poly jar.
Given this modest experiment, subsequent cleanup attempts, then should probably proceed as follows:
1) Calculate the amount of water needed to dissolve the perchlorate at 100 degrees; use 50% more water. 200 grams of potassium perchlorate would require 1.5 liters. Heat, but do not quite boil, the potassium perchlorate solution until all of the perc has dissolved.
2) Prepare a solution that contains 10% by weight potassium metabisulfite, relative to the amount of perchlorate you wish to clean. In other words, if you have 500 grams of perchlorate, start with 50 grams of potassium metabisulfite. This is cheaply obtained from home brew shops, and similar.
3) Add enough HCl to gently acidify the perchlorate solution, perhaps 15 drops per liter. Using a pipette, and with stirring, gradually add 1/2 of the potassium metabisulfite solution to the perchlorate. Allow it to stand for several minutes with continued stirring. Test the solution for chlorate. If still dirty, add additional metabisulfite.
4) Once clean, the solution needs to be neutralized, or made very slightly basic. The caustic of choice is KOH. Since the reduction byproducts are acidic, it requires more KOH on a Normal basis than the initial added HCl would indicate. Slowly add saturated KOH, dropwise, to the solution, and test with pH paper or a pH probe. When nearing neutral, it doesn't take much additional base to overshoot, so use caution. If you make it too basic, add a drop or two of HCl to bring it back to neutral. A Magnetic stirrer can be put to good use in this phase.
5) Boil the solution down until the quantity of water is equivalent to the necessary mass of water for the amount of perchlorate used. Potassium Perchlorate dissolves at a rate of 200 to 220 grams per liter at 100 degrees C. An easy way to do this is to boil down until the very first crystals appear, then remove from heat. Allow it to slowly cool, and once at room temperature (there will already be significant crystals), refrigerate. Harvest the crystallized perchlorate by the usual means (decant + filtration) when the solution is at approximately 5 degrees C. Wash the perchlorate with ice water, then cold ethanol + water, 50:50. The remaining potassium chloride and potassium sulfate remain dissolved, and are discarded. Spread the perchlorate out to dry, then ball mill as desired.
Enjoy! 
2 Comments
Recommended Comments