You'll put your eye out...

Please excuse me and correct me if you disagree or I am wrong as my chemistry knowledge is fair to Poor. I have been led to believe that the Indigo Carmine test is the most sensitive indicating chlorate at 1~2 parts ppm and so had not considered N-Phenylanthranilic acid.

I was very concerned about boiling the HCL also so used a little as I could and found that it was quite straight forward. Some of my batches that I re-crystallized and washed only once show a strong color change leading me to the conclusion that it exceeds 5ppm. I have 2 x batches that show only a just noticeable change or medium change so my question is should I try and destroy what remains given that this should indicate less than 5ppm . I tried to use less HCL- 2.5ml and 0.5ml of the indicator. I repeated several times and found the color shift the same by heating and adding my cell solution just before it gets to boiling. If it is too cool it seems to still indicate if you keep the heat on after you add the cell solution but I do not know how much this is affecting the accuracy. The worse thing I found about this is the toxic fumes.

One solution I had a nasty surprise when it reacted badly with the first drip of solution. If I am not sure how much chlorate there is now I dip a long pin tip in my cell solution and dip that in first so I get a rough indication if this is going to happen.

I have not worked with indigo carmine yet, but I plan on trying it as an indicator.

 

Couple of thoughts... this NPAA test seems to detect 10 ppm; I suspect it will do better. One way to get a handle on a perc batch is to create a STANDARD chlorate solution which can be used for comparison purposes. Decide what ppm is acceptable... and I have no idea what that number might be. Let's say it's 5 ppm. Dissolve 5 milligrams of chlorate in a liter of water, and there's your standard. When you do a test, do two of them side by side, one of them the 5 ppm standard, the other, unknown, and look at color intensity. This will give a rough idea if the unknown is above, or below, 5 ppm.

 

5 ppm may be completely acceptable for chlorate contamination. Picture 5 milligrams of potassium chlorate evenly distributed in a kilogram of perchlorate, that's 5 ppm, and it is not much at all.

 

I guess what I'm trying to say is, it is impossible to rid your perc of ALL (100%) chlorate. There will always be some, we just need to decide what's OK for Pyro use.

 

You also mentioned testing "cell solution." Are you using the test as a means to determine when perchlorate production begins? I can't think of any reason to be testing electrolyte for perchlorate... I think these tests are only for post-production cleanup of perchlorate.

 

If you decide to do further cleaning, you may want to combine all of your perch and do it all at the same time, if this is physically possible. I think you'll end up with less losses than if you worked each batch separately.

 

One other way to execute a recrystallization cleanup that would be a bit more thorough, but which would also lose more perchlorate, is to NOT take the recrystallization to 0 degrees C. Take it to maybe 5 degrees C. More of the contaminating chlorate will want to remain in solution at 5 degrees vs. 0, and of course more perc will remain in solution too, but what drops out will be exceptionally pure. It is probably those last few grams that drop out that carry the bulk of the contamination. Of course anything left over goes right back into a cell, so nothing gets wasted.

The Indigo Carmine test refers to using a sample of the electrolyzed cell solution but I am sorry I should have

said I made it from my Perch as I have as I have not run my cell for several months. I am keen to hear what is

considered safe as I have been under the assumption that you should try and remove any detectable amount.

What you are saying makes sense though because even 10 parts per million is very small.

I read and have tried one method of getting exceptionally pure perchlorate is to dissolve around 100mg/L and

do not let the temp go much below room temperature. It would mean a lot is left behind but as you said none is waisted. It was some time ago that I tried this and I did get results but I remember i had trouble getting the crystals to form. It sort of makes sense to me if you look at the solubilities. You should get back at least 75% if it was mostly perchlorate but almost none if it was all chlorate. I am interested in what you think of this at 20deg the solution will still hold a lot of Chlorate.

The perchlorate solubility takes a REAL dive below 20 C... The chlorate goes down, but not as dramatically. There is a solubility chart which you've probably seen:

 

http://www.5bears.com/perc/tpc07.jpg

 

While this is not a mutual solubility chart, it's still pretty useful. Up to about 20 degrees, the lines are close to parallel, but then they diverge, with the solubility of chlorate increasing faster than perchlorate. Maybe that is the region, around 30 to 40, that would be worth exploring. At 40 degrees, you'd have about 4 out of 22 grams/l perchlorate still dissolved, which is 18%. For the chlorate, it's 14 out of 58, or 24%. But we're talking about SUCH small bits of chlorate. I'd guess even at 20 celcius, that chlorate is going to remain in solution. In the filter, you've got pure perc that is wetted with a solution that contains your evil chlorate. If allowed to dry without a good washing, the chlorate might recrystallize.

 

I think the ultimate recrystallization method might work something like this - recrystallize to 20 C, decant, then filter. IMMEDIATELY while the perc is still wet, dump the filter contents into pure water at 1 or 2 degrees C. Almost none of the perc will dissolve. But the wetness that was on the perc crystals will immediately be massively diluted by the cold water, and get stripped off the perc crystals. Working quickly, you then vacuum filter that cold water with the suspended perc. Hit the perc crystals at the very end with methanol, ethanol, or iso alcohol, which will further wet and strip the crystals. They will then dry quickly.

 

If your contamination was 20 ppm before this, I'd have to guess it would be driven to 1 ppm easily, with the loss of 20% of the perc, give or take. That lost 20% goes back into your perc cell.

 

All of this ignores chemical purification methods. The time to do any of that would be with the 100 degree perc solution, prior to cooling.

I forgot to add, if you google "chlorate contamination of perchlorate" all you get is environmental whackos whining about perchlorate pollution. Googling just "chlorate contamination" brings up the articles we already are familiar with. I'd pay $100 to tour a perchlorate plant and pick the brains of some of the engineers.

 

There's very little literature out there. Iron (II) Sulphate is a reducing agent that can be used, but in what quantity? I think it might be worthwhile to run some small-scale tests whereby a small batch of commercially pure perc is intentionally contaminated with chlorate. Then, the Iron Sulphate method can be tried. The hot liquor can be periodically checked with NPAA or indigo carmine, and at the point where the chlorate contamination is reduced back to nil, there's your quantity of iron sulphate. There's enough work waiting to last for weeks of investigation.

Swede, your Blog has been one of the most interesting reads for me in a LONG time.

 

I understand less than 1/4 of it () but it's fascinating nonetheless.

 

Keep up the good work!!