making potassium (per) chlorate

[WSM]

I imagine if you can work with molten glass it would seal well on a platinized wire. This would be ideal if the glass tubing were tough enough to take some handling without breaking. I think I'll stick to spot-welding tubular titanium and sealing it in the cell with PVDF compression fittings.

WSM

 

I double checked with a friend who works in a neon sign shop, and borosilicate glass (as is Pyrex) does use platinum wire (and I believe platinized wire would work) to stick to the melted glass for electrodes. Borosilicate has many handling advantages but it takes a certain skill set to handle and form the molten glass and cool it so it doesn't break from internal stresses. Once mastered, a lot of nifty things can be made with it.

 

WSM

Edited September 12, 2012 by WSM

[ANFO]

If you go through Swede's blogs, you'll come across his mention of round leads through PVDF compression fittings. That revelation by Swede opened the door for me to progress my work without the aggravating salt creep with the subsequent power lead corrosion problems.

 

I've tried it and it works well. The difficult parts are sourcing the titanium tubing to use for leads, forming them and spot-welding them to the electrodes, filling them with a low melting point conductive metal, etc. Once all that is managed, it's completely possible to make a productive cell without the mess.

 

It's not easy, but it is possible.

 

I think Swede is getting ready to offer some electrodes with tubular leads in the Agora; stay tuned...

 

WSM

 

I've already seen this on Swedes blog, and the exact same idea of tube through compression fitting had occurred to me already, except I thought of using solid bar, then cutting a 3-4cm recess into the bottom of the bar and drilling two holes through the recessed bit. Then you can slide plate or MMO into the recess (which must be tight obviously) and secure with titianuim bolts. What you get is exactly like Swede's round necked electrode, but doesn't need to be filled with tin, doesn't require spot welding and is perfectly balanced as the plate is attached dead center in the bar. The only thing would be conductivity with the MMO, but a snug fit and two titanium screws should do it, at least for lower amperage.

Edited September 12, 2012 by ANFO

[Swede]

Solid Ti would work great. The cross section is more than adequate compared to the skinny cross section of rectangular sheet straps. The only drawback that I can think of would be the cost, since solid Ti bar is much more expensive than tube.

 

Ditto the Ti bolts, either with nuts or tapping the solid Ti round bar shank. A Ti washer under the head of the bolt would help with surface contact. The beauty of this is that you can use a variety of anodes by simply unscrewing the old one. Neat concept! All that would need to be done is to aggressively scrape and clean the portions of the anode and shank where contact is made prior to assembly.

[WSM]

Solid Ti would work great. The cross section is more than adequate compared to the skinny cross section of rectangular sheet straps. The only drawback that I can think of would be the cost, since solid Ti bar is much more expensive than tube.

Ditto the Ti bolts, either with nuts or tapping the solid Ti round bar shank. A Ti washer under the head of the bolt would help with surface contact. The beauty of this is that you can use a variety of anodes by simply unscrewing the old one. Neat concept! All that would need to be done is to aggressively scrape and clean the portions of the anode and shank where contact is made prior to assembly.

 

I think the tubular leads have one distinct advantage over round, solid titanium leads; if the tubular leads are filled with tin, they're more conductive. Tin is about five times more conductive than titanium, so the leads will be cooler as current goes through. This will help prevent heat damage to the polymer compression fittings.

 

WSM

Edited September 13, 2012 by WSM

[ANFO]

As nice as this concept is, unfortunately I'm on quite a tight budget (a student one), so I'm really going to try for a cheaper option (sealant). I'm sure I can make it work, even if I have to take three viton o rings, put them on the electrode and solvent melt them into a plug of some kind with solvent. My cell is going to be the best of the budget class

[Swede]

ANFO, can you elaborate on what you are trying to do with the viton? Is the goal the sealing of the interface between the electrode leads and the cell lid? Or is it to protect something INSIDE the cell?

 

For a basic sealant, RTV silicone, the silicone sealant in a tube, will last for a couple of runs, and when it needs to be replenished, it scrapes off with a razor pretty easily. Likewise, good old hot glue works OK for sealing, but it too will degrade. Worse, it softens with the heat of the electrodes. Of the two, silicone is better and longer lasting. If you do it, don't be shy, goop it on there in healthy amounts.

 

I like the idea of a solvated viton. I've got plenty of viton to play with, as well as methylene chloride, acetone, MEK, and a bunch of other solvents that might be worth a shot. If I can get some to dissolve, a heavy viscosity would work best. I'll give it a try tomorrow.

[Arthur]

There's always Polish suppliers, and one other supplier in the UK (like oliver brown) who sells chlorates at a good price. Also shipment from Poland to the UK does not pass through customs.

 

In the EU there are still lots of checkpoints! there is certainly customs checking at the borders of some countries within the EU, Largely it's for drugs and illegal articles but most pyro chems are white powders too.

 

http://www.onlinemetals.com/merchant.cfm?pid=12629&step=4&showunits=inches&id=322&top_cat=1353

Any interest -titanium supplies but at a price

[Arthur]

ANFO The particular properties of this electrolysis mean that at times hydtogen and chlorine and possible oxygen are evolved and if you want big output there is usually a big chlorine gas stream to cope with.

If you have an open cell then that's where the chlorine will be, corroding your cell and killing any plants and people nearby.

If you have a closed cell with a ducted vent then you can let the chlorine go where it's least harmful BUT the backpressure and spray in the cell mean that all the openings need to be properly sealed for the gas to flow out of the vent. -so good O rings are needed and the pipe may need changing every few months.

 

There has been a cell failure thought to be due to the glass continer eroding and plating onto the electrodes.

Swede's idea of the bucket cell is good. Get a snap top bucket of 5 - 50 litres and seal an electrode holder onto it's lid. finer points of sealing help a lot otherwise you get salt creep out of all the joints and that can corrode the electrode hangers and the cables

Edited September 13, 2012 by Arthur

[ANFO]

I'm fully aware of the gasses involved, and the redox reactions taking place. My cell body is going to be HDPE chlorine bucket (I mentioned it further up), which includes an airtight lid naturally.

 

I'm just trying to find the best ways of sealing up the areas where the electrodes enter the lid.

 

ANFO, can you elaborate on what you are trying to do with the viton? Is the goal the sealing of the interface between the electrode leads and the cell lid? Or is it to protect something INSIDE the cell?

 

For a basic sealant, RTV silicone, the silicone sealant in a tube, will last for a couple of runs, and when it needs to be replenished, it scrapes off with a razor pretty easily. Likewise, good old hot glue works OK for sealing, but it too will degrade. Worse, it softens with the heat of the electrodes. Of the two, silicone is better and longer lasting. If you do it, don't be shy, goop it on there in healthy amounts.

 

I like the idea of a solvated viton. I've got plenty of viton to play with, as well as methylene chloride, acetone, MEK, and a bunch of other solvents that might be worth a shot. If I can get some to dissolve, a heavy viscosity would work best. I'll give it a try tomorrow.

 

I was just planning to use it on the interface, I didn't know that silicon would hold up. But still, viton would be much more durable, and could have numerous applications elsewhere in the cell.

 

From what I gather, latent solvents like acetone might not do the job, but MEK or other active solvents should work.

 

While I'm posting, I thought I should mention that I spent about three hours looking into the power supply and amperage side of the cell. I gathered a lot of information on how to work out the resistance of the cell, an thus the amperage before even connecting it up. The resistance of the KCl cell can't be measured with a simple multimeter because the resistances are way below 1 ohm. So I found this:

 

http://upload.wikimedia.org/math/1/b/8/1b80b5da1c69cec0245b22e0ed5c1bd6.png

 

R being the resistance of the cell, L being the distance between the electrodes in meters, A being the surface area of the plates and p being the specific resistance of the solution. Unfortunately, I can't find the specific resistance of concentrated KCl solution. All I found was that it is 0.097ohm/m at 1mol/L, which doesn't help seeing as concentrated is at 4.6 mol/L and more.

 

Anyway, with this one can calculate the current your cell will draw exactly at certain electrode sizes and distances, all I need to do is calculate 'p' for concentrated KCl which will be easy enough. I was just working on this seeing as in the very distant future I want to build a 135 amp linear power supply from a MOT and some high amp bridge rectifiers

[Mumbles]

KCl solution is actually one of the things they use to calibrate things off of. Resistivity is going to vary greatly with temperature. I'm sending you a paper that might help. I only had a quick glance through it, but it sounds like they took all of the historical data of resistivity and tabulated it and calculated a more correct formula or something like that to allow for one to calculate the resistivity at any temperature and concentration.

[Arthur]

Most peoplr seem to use a ATX computer power supply or a switch mode PSU, with high current 50Hz PSUs the capacitance needed to smooth the supply to a reasonable level is too big, impossibly big unless you have a three phase supply. SMPSUs are much better at handling the power factor correction needed so you only pay for the power consumed. Certainly platinum electrodes don't like voltage ripple.

[WSM]

KCl solution is actually one of the things they use to calibrate things off of. Resistivity is going to vary greatly with temperature. I'm sending you a paper that might help. I only had a quick glance through it, but it sounds like they took all of the historical data of resistivity and tabulated it and calculated a more correct formula or something like that to allow for one to calculate the resistivity at any temperature and concentration.

 

The paper sounds like something many of us could benefit from. Could you post it here? Thanks.

 

WSM

[Mumbles]

It's a published paper, but I'm not sure how they feel about me freely posting it, since they do charge for access. I didn't read through it very carefully, but if there is a useful formula in there, I'm sure it will be posted soon.

 

http://www.springerlink.com/content/2k611w47g72k6273/

[WSM]

I think the tubular leads have one distinct advantage over round, solid titanium leads; if the tubular leads are filled with tin, they're more conductive. Tin is about five times more conductive than titanium, so the leads will be cooler as current goes through. This will help prevent heat damage to the polymer compression fittings.

WSM

 

I found a chart that lists the conductivity of several metals:

 

http://www.eddy-current.com/condres.htm

 

In this list pure copper is 100 and titanium is 3.1. Tin is 15.0 and 95:5 solder (Tin:silver) is 16.6, so it's plain to see a tin-filled titanium tube will be more conductive than solid titanium. I wonder if a copper wire in a tin filled titanium tube wouldn't be better?!

 

WSM

[Arthur]

Copper wold help if you could be certain that the titanium tube was hermetically sealed as even a trace of copper chlorate would be risky.

[taiwanluthiers]

if formation of copper chlorate is so bad, how come copper salts are safe to use in chlorate blue stars? is it because potassium chlorate doesn't form a double decomposition reaction with copper salts?

[WSM]

Copper wold help if you could be certain that the titanium tube was hermetically sealed as even a trace of copper chlorate would be risky.

 

I heat and flatten the titanium tube repeatedly till it is flat and no air passes through before I spot-weld the electrode to it. Then I fill the tube with low-melting point metal. After all this is done I treat the electrode with HCl to remove any contaminants. When all this is done, I'm fairly certain my electrodes are good.

 

I think a copper wire in the solder fill would help, too.

 

WSM

Edited September 14, 2012 by WSM

[ANFO]

Most peoplr seem to use a ATX computer power supply or a switch mode PSU, with high current 50Hz PSUs the capacitance needed to smooth the supply to a reasonable level is too big, impossibly big unless you have a three phase supply. SMPSUs are much better at handling the power factor correction needed so you only pay for the power consumed. Certainly platinum electrodes don't like voltage ripple.

 

Would a higher frequency require less capacitance?

 

EDIT: Nevermind I've found the formula.

Edited September 14, 2012 by ANFO

[Swede]

The notion of a Cu core in a Ti tube is a good one. I'd do it by stripping a section of stranded copper, possibly even the welding cable, and painting on a very light coating of rosin flux, definitely not the acid flux normally used for plumbing.

 

The Ti tube should be mechanically abraded inside (dremel, or round file) aggressively. This removes oxides and exposes a coarse surface, as any molten tin or solder is going to create a mechanical attachment to the Ti... Ti, like Al, does not "solder" in the normal sense.

 

Next, I'm thinking that the Cu should be cut such that when tapped into the pre-formed (pinched closed at the bottom) Ti tube, the bottom of the Cu wire bundle spreads out to fill the bottom cavity area, leaving maybe 25mm at the top NOT occupied by Cu. This area needs to see only the tin so that a stainless steel stud can be installed. With the Cu tapped in firmly with a dowel, in goes the tin or solder, and then I'd maintain a molten heat for a period of time so that flux and/or crud has at least a modest chance to rise to the top.

 

Drill and tap the tin, install a SS stud, and it'd be ready to go. I think this will work great and also cost a lot less than filling the tube with tin or solder.

[Arthur]

Copper sealed in a tube is a great idea, exposed it's a bad idea.

 

Higher frequenciy power supplies use much smaller capacitors to reduce the DC ripple, High current power supplies are careful about the power factor too.

[ANFO]

It turns out that those chlorine buckets are very thick polypropylene, not HDPE. Which is a bit of a let down

 

Copper sealed in a tube is a great idea, exposed it's a bad idea.

 

Higher frequenciy power supplies use much smaller capacitors to reduce the DC ripple, High current power supplies are careful about the power factor too.

 

Arthur, would it be possible to wire two SMPSU's in parallel to increase the current output (say two 30 amp computer PSU's for a possible output of 60 amps), provided they were wired on the same voltage obviously.

Edited September 14, 2012 by ANFO

[pdfbq]

... as even a trace of copper chlorate would be risky.

Are yiu sure about this Arthur?

[patsroom]

Are yiu sure about this Arthur?

There are a few chlorate(s) that are very unstable. And it does not take much to create a unpleasant surprise which could ruin much more than your day................Pat

[WSM]

It turns out that those chlorine buckets are very thick polypropylene, not HDPE. Which is a bit of a let down

 

I think the polypropylene would hold up better. It's rated for somewhat higher temperatures than PVC, too. I think it should be tested.

 

WSM

Edited September 15, 2012 by WSM

[WSM]

Copper sealed in a tube is a great idea, exposed it's a bad idea.

 

I agree.

 

When I use a titanium tube for an electrode lead I first tap the top end with a special tap that's made for putting machine threads in titanium or other hard metals. Next, I heat the other end to red-hot and press, several times till it's pressed flat and no air can be blown through the tube. Then I spot weld the electrode on the flattened end. Next, I heat the tube with a torch and melt the solder in, followed by chasing the threads with the same tap used in the first place.

 

I attach the power lead with a stainless steel machine screw and two stainless washers (the power lead is terminated with a ring lug which is locked in place between the two washers under the machine screw). I get a very good electrical connection this way.

 

 

This photo shows some electrodes with the type of terminations I'm describing. They're labor intensive but effective.

 

WSM