The Bucket Cell - Start to Finish

[WSM]

 

I wonder if our concerns about sodium contamination in the final product are overblown? I did a calculation based on the relative solubilities of sodium perchlorate and potassium perchlorate; I believe the potassium salt is about one percent (1%) the solubility of the sodium salt.

At that rate, a simple wash with cold distilled water after vacuum filtering the KClO₄, should leave the final product pure enough to use, as is, when dried and powdered. When used, I expect there'll be no sodium contamination to speak of.

WSM

 

The jury is still out on this one. Maybe multiple washings with chilled distilled water will be required.

 

I should know soon and report it here...

 

WSM

[WSM]

Yeah, I think so too. The problem with using heatsinks is they're so bulky, they might short out electrically if the leads are too close to each other. I'm sure the problem can be circumvented with proper placement of the right configuration of heatsink. If they're too small, fans can help but I prefer keeping things as simple as possible (making the tubular leads large enough and filling them with high conductivity filler should prevent them from heating too much in the first place).

We'll see...

WSM

 

This is proving to be true! I've learned the larger diameter titanium tubing filled with large diameter, bare copper wire plus lead-free solder works well to increase the conductivity and reduce the heating seen in smaller diameter leads.

 

In my experimental sodium perchlorate cell using a lead dioxide anode, I used 3/8" diameter CP titanium leads, filled as described, and the heating seen was greatly diminished from what it would have otherwise been.

 

It's a lot of work, to be sure, preparing the leads this way; but the results are proving it to be worth the effort.

 

WSM

[pyrojig]

Awesome!!! Did you by chance probe the electrode shanks for a temp? I ran a heat probe on the last set and I believe that the temp was closer to 90C* or so. I was really getting worried that the fittings would not hold for long at those temps . They did discolor and show a little damage , but surprisingly held;/ gotta love Kynar.

I am glad to see progress with the electrode overheating prob. The heat should stay in the cell where it can do some work ,instead of burning off as wasted energy outside the cell. Soon we'll have to do another set of BCA pics with the new electrodes and upgrades etc.

[WSM]

Awesome!!! Did you by chance probe the electrode shanks for a temp? I ran a heat probe on the last set and I believe that the temp was closer to 90C* or so. I was really getting worried that the fittings would not hold for long at those temps . They did discolor and show a little damage , but surprisingly held;/ gotta love Kynar.

I am glad to see progress with the electrode overheating prob. The heat should stay in the cell where it can do some work ,instead of burning off as wasted energy outside the cell. Soon we'll have to do another set of BCA pics with the new electrodes and upgrades etc.

 

I did measure the heat of the leads, and they were well within tolerances (though one was a bit hotter than I would like). I think a little fine tuning can keep things well within specs.

 

I'm looking forward to seeing your progress with the bucket cell.

 

WSM

Edited June 28, 2015 by WSM

[Mumbles]

 

The jury is still out on this one. Maybe multiple washings with chilled distilled water will be required.

 

I should know soon and report it here...

 

WSM

 

 

I'm just thinking out loud here. I'm curious if you think going through the trouble of using chilled water is worth it for perchlorate. I can understand it with the relatively more soluble chlorate. However with perchlorate, the difference between 0C and 20C is about 0.75g/100mL vs. 1.5g/100mL. Sodium chlorate and perchlorate are at least 100x as soluble as potassium perchlorate.

 

I was also thinking that the one of the possibilities is sodium being trapped in the crystals being more difficult to remove, as opposed to contaminants just clinging to the surface. Stirring the crude crop of product with water for a while might do a better job at extracting this sodium. From there, a few water washes might get rid of the rest. This may be where the temperature of water could make a difference, since you'd probably be using more of it.

 

I've also been trying to think of a good, simple spot check to see when you're done washing. A flame test could certainly work. I was actually thinking about measuring the conductivity of the washing eluate. In theory it should plateau out, when you're only removing the amount of potassium perchlorate that is soluble.

[WSM]

I'm just thinking out loud here. I'm curious if you think going through the trouble of using chilled water is worth it for perchlorate. I can understand it with the relatively more soluble chlorate. However with perchlorate, the difference between 0C and 20C is about 0.75g/100mL vs. 1.5g/100mL. Sodium chlorate and perchlorate are at least 100x as soluble as potassium perchlorate.

I was also thinking that the one of the possibilities is sodium being trapped in the crystals being more difficult to remove, as opposed to contaminants just clinging to the surface. Stirring the crude crop of product with water for a while might do a better job at extracting this sodium. From there, a few water washes might get rid of the rest. This may be where the temperature of water could make a difference, since you'd probably be using more of it.

I've also been trying to think of a good, simple spot check to see when you're done washing. A flame test could certainly work. I was actually thinking about measuring the conductivity of the washing eluate. In theory it should plateau out, when you're only removing the amount of potassium perchlorate that is soluble.

 

Most of this post is reasonable and well thought out.

 

As to the sodium being trapped in the crystals; this may not be the case. Unlike chlorate crystals which form slowly with lots of interstitial spaces to catch and hold contaminants, my experience with converting sodium chlorate to potassium perchlorate by metathesis seems to show that the exchange is almost instantaneous, forming very fine crystals, even from cold solutions. I suspect most, if not all contaminants will be on the surface of the fine, white crystals formed.

 

If there are contaminants held within the crystalline matrices, that's a case for re-crystallization, which I believe the potassium perchlorate industry avoids as unnecessary.

 

I may be wrong, but my argument makes sense to me (and if I am wrong, I'll freely admit it).

 

I, too, was considering a spot test for sodium, but am resorting to flame tests since the sodium lines are so prominent and strong. I suspect the smallest portion of sodium, if present, would mask the weak violet spectral lines of potassium.

 

I like the way you think, Mumbles. Thanks for your input.

 

WSM

Edited June 30, 2015 by WSM

[Mumbles]

My thoughts on trapping sodium inside the crystal had more to do with the speed at which it's being precipitated, not necessarily the size of the crystals. Being so instantaneous, things could get caught sort of in essence because it can't get out of the way fast enough. I guess there's only one way to find out. Keep us posted.

[Arthur]

The text by Schumacher does address the "rate of crystallisation" issue

[WSM]

My thoughts on trapping sodium inside the crystal had more to do with the speed at which it's being precipitated, not necessarily the size of the crystals. Being so instantaneous, things could get caught sort of in essence because it can't get out of the way fast enough. I guess there's only one way to find out. Keep us posted.

 

I will... Actually I need to anyway for Part Eleven of my Homegrown Oxidizers series. Thanks for the encouragement.

 

WSM

[WSM]

The text by Schumacher does address the "rate of crystallisation" issue

 

Yes, I believe that's true.

 

When I do testing on my potassium perchlorate, I'll know whether I need to modify my approach or continue what I'm doing.

 

WSM

[WSM]

As to the sodium being trapped in the crystals; this may not be the case. Unlike chlorate crystals which form slowly with lots of interstitial spaces to catch and hold contaminants, my experience with converting sodium chlorate to potassium perchlorate by metathesis seems to show that the exchange is almost instantaneous, forming very fine crystals, even from cold solutions. I suspect most, if not all contaminants will be on the surface of the fine, white crystals formed.

WSM

 

Oops! This should have read, ...converting sodium perchlorate to potassium perchlorate by metathesis...

 

I'm surprised nobody called me on that mistake. I guess either I'm getting more respect that I deserve, or I'm using too many big words and nobody is daring to challenge me? The worst possibility would be, this topic is so dry and boring that nobody really cares...

 

The most likely reason, it's fireworks season and folks have better things to do than read this stuff . (Yeah, that's it )

 

WSM

[WSM]

 

Oops! This should have read, ...converting sodium perchlorate to potassium perchlorate by metathesis...

I'm surprised nobody called me on that mistake. I guess either I'm getting more respect that (?) I deserve, or I'm using too many big words and nobody is daring to challenge me? The worst possibility would be, this topic is so dry and boring that nobody really cares...

The most likely reason, it's fireworks season and folks have better things to do than read this stuff . (Yeah, that's it )

WSM

 

Darned spellcheck missed that one. It should have read, "than I deserve". I really should double check what I write so I accurately say what I mean !

 

WSM

[pyrojig]

Your all good. Dont be so hard on your self. I am a lazy typer at times , and clerical errors are excepted on the site ( just this time ) . But if you ever find a direct route from Naclo3 to Kclo4 then by all means post it .

[WSM]

Your all good. Dont be so hard on your self. I am a lazy typer at times , and clerical errors are excepted on the site ( just this time ) . But if you ever find a direct route from Naclo3 to Kclo4 then by all means post it .

 

They say, "If you can dream it, you can do it.", but in this case I don't see any such direct route. Maybe, sometime in the future will reveal a method we're not seeing at this point, that will allow a direct route. We'll see...

 

WSM

[Arthur]

The Graphite substrate lead dioxide electrode is said to be good for one process to completion BUT the electrode is impossible to buy and hard to make with toxic by-products, it's also not of infinite life, SO til I can buy a GSLD electrode at Asda/Walmart I'll assume that the sodium chlorate cell is best, followed by chlorate cleaning, followed by Ti substrate perc cell then an addition of KCl to ppt out the Kperc.

 

I think some people are going to rely on using either K or Na chlorate for process simplicity, but need to remember the limitations of the chlorate based formulary, in particular the sulphur incompatibility.

[WSM]

The Graphite substrate lead dioxide electrode is said to be good for one process to completion BUT the electrode is impossible(?) to buy and hard to make with toxic by-products, it's also not of infinite life, SO til I can buy a GSLD electrode at Asda/Walmart I'll assume that the sodium chlorate cell is best, followed by chlorate cleaning, followed by Ti substrate perc cell then an addition of KCl to ppt out the Kperc.

 

If my understanding is correct, GSLD electrodes can be had from China if someone places a minimum order and is willing to sell the surplus. The life of the GSLD anodes is very good if properly handled (same with LD on titanium).

 

Due to the lack of availability of commercially made GSLD anodes (which would make a sodium chloride to sodium perchlorate cell possible), I've resigned myself to two-step perchlorate manufacture,

 

1) sodium chlorate manufacture, and

2) sodium perchlorate manufacture followed by metathesis with KCl, with purification steps in between.

 

WSM

Edited August 2, 2015 by WSM

[WSM]

The Graphite substrate lead dioxide electrode is said to be good for one process to "completion" BUT the electrode is impossible to buy and hard to make with toxic by-products, it's also not of infinite life, SO til I can buy a GSLD electrode at Asda/Walmart I'll assume that the sodium chlorate cell is best, followed by chlorate cleaning, followed by Ti substrate perc cell then an addition of KCl to ppt out the Kperc.

I think some people are going to rely on using either K or Na chlorate for process simplicity, but need to remember the limitations of the chlorate based formulary, in particular the sulphur incompatibility.

 

I just re-read this post and several assumptions are stated. The first to catch my attention is using a GSLD as a "one-shot" anode to run from chloride to "completion". Several steps are still required, even if one could go directly from sodium chloride to sodium perchlorate.

• purification to remove residual chlorate and other unwanted ions
• metathesis with KCl to convert sodium perchlorate to potassium perchlorate
• filtration and washing to remove residual sodium ions from the potassium salt
• handling the wash water and filtrate to prepare for its reuse in a chlorate cell

I may be forgetting other steps, as well.

 

Lacking the GSLD anodes to try the "direct route", I'll continue my research with platinum and LD anodes and a two step conversion between chlorides and perchlorates (NaCl -> NaClO₃ and then NaClO₃ -> NaClO₄). It's not perfect, but so far it certainly works. Maybe better than I hoped for. Truly, the solubilities work quite well for such a system (no wonder industry does it this way, and so can we if we take the necessary steps to do so).

 

WSM

Edited August 29, 2015 by WSM

[pyrojig]

(GSLD:)

I know it is probably a pointless , toxic process making your own( of which one could buy one on ebay already made). It makes me wonder if the paint and bake method on a graphite substrate would be something /1st worth the effort, and second , something a amateur with limited material ( lab apparatus) could make. The " One Shot " sys seems attractive , but i have my doubts on CE , and it being simple as stated above.

[Arthur]

One of the Patents I've read said that the rate of precipitation in the metathesis stage was determined by the rate of temperature fall, and the degree of metathesis by the quantity of K salt added. KClO₄ ppts out before KClO₃ Apparently the two are rapidly identifiable as they sequentially drop out.

[WSM]

(GSLD:)

I know it is probably a pointless , toxic process making your own( of which one could buy one on ebay already made). It makes me wonder if the paint and bake method on a graphite substrate would be something /1st worth the effort, and second , something a amateur with limited material ( lab apparatus) could make. The " One Shot " sys seems attractive , but i have my doubts on CE , and it being simple as stated above.

 

I would try making GSLD by an electrochemical deposition of the material rather than a "paint and bake" method. The proper setup would be a well organized and stocked lab.

 

Swede's LD on titanium worked because he studied it out well and took the time and effort to do it right. We learned the MMO substrate needs to be completely covered with MMO to provide a solid base for the LD to adhere to (as witnessed by the heavy LD nodules dropping off the cut edges of the MMO, but everywhere else sticking solidly).

 

Making the sodium perchlorate directly from sodium chloride depends on using GSLD, plus a few additives, and I won't attempt it without using GSLD. Since I have MMO anodes for making sodium chlorate, and LD and platinum anodes for making perchlorate; I'll stick to the two step process for making perchlorate for now. If I score some GSLD, I'll try my hand at the "one-step" process; at least as an experiment.

 

Whether or not making sodium perchlorate directly from sodium chloride is "simple", remains to be seen. We'll see for ourselves when we ever get some GSLD to try...

 

WSM

[WSM]

One of the Patents I've read said that the rate of precipitation in the metathesis stage was determined by the rate of temperature fall, and the degree of metathesis by the quantity of K salt added. KClO₄ ppts out before KClO₃ Apparently the two are rapidly identifiable as they sequentially drop out.

 

The temperature level makes sense, plus slowly adding the potassium chloride should also help (if the KCl port doesn't jam). It seems the potassium perchlorate is so insoluble (especially compared to the other things in the mix), that it instantly drops out of solution. This is why I suspect the sodium salts in solution or clinging to the small potassium perchlorate crystals can (and should) be washed off with distilled water while they're still in the Buchner funnel with the vacuum on.

 

As to using fractional crystallization to separate potassium perchlorate and potassium chlorate, it would make sense if I hadn't already destroyed the residual chlorate (in the sodium perchlorate) with SO₂ generated from sodium metabisulfite solution before the metathesis with KCl. We may be discussing apples and oranges here.

 

WSM

[pyrojig]

 

it would make sense if I hadn't already destroyed the residual chlorate (in the sodium perchlorate) with SO2 generated from sodium metabisulfite solution before the metathesis with KCl. We may be discussing apples and oranges here.

@WSM: guess we may have to make more .

[WSM]

@WSM: guess we may have to make more .

We definitely need to make more. The bulk of the sodium perchlorate is sitting, waiting to be processed. In the mean time, I'm also preparing to set up a sodium chlorate cell to make feed stock for subsequent runs of the perchlorate cells.

 

I expect we'll learn a lot and trail blaze for others to follow in our footsteps.

 

I got 100 pounds of salt (sodium chloride) for less than ten cents a pound (<$0.20 US/kilogram). It's pure without additives and should work well. I'm giving serious consideration to running the sodium chlorate cell as a bucket cell; we'll see...

 

WSM

Edited September 15, 2015 by WSM

[Arthur]

IMO the professional method

electrolyses NaCl to say 90% completion then ppts out chlorate but not necessarily all of it

The chlorate goes on to a perc cell and the chloride and some chlorate goes back as returned electrolyte for more chloride. -It doesn't matter that some chlorate goes back to the chlorate cell input.

The chlorate is then electrolysed to say 90% completion to ease the life of the electrodes

Then KCl is added as a cold solution to ppt out say 90 - 95% of the perc any residual chlorate being returned to the process with NaCl added.

 

destroying chlorate risks leaving liquor that cannot be recycled.

[WSM]

IMO the professional method

electrolyses NaCl to say 90% completion then ppts out chlorate but not necessarily all of it

The chlorate goes on to a perc cell and the chloride and some chlorate goes back as returned electrolyte for more chloride. -It doesn't matter that some chlorate goes back to the chlorate cell input.

The chlorate is then electrolysed to say 90% completion to ease the life of the electrodes

Then KCl is added as a cold solution to ppt out say 90 - 95% of the perc any residual chlorate being returned to the process with NaCl added.

destroying chlorate risks leaving liquor that cannot be recycled.

 

Well said, Arthur.

 

As to destroying the chlorate, it would create a risk IF left unaltered. I'd propose purifying the used liquor by removing unwanted species before recharging with additional chloride to run it in a chlorate cell.

 

WSM

Edited September 16, 2015 by WSM