Sciencemadness Discussion Board

EFFECT OF RIPPLE CURRENT ON ELECTROLYSIS

DerAlte - 29-6-2007 at 20:39

I don't want to start a new topic if it's been covered, but a search didn't reveal any germane stuff. Maybe I didn't use the right keywords.

My question is, have any of the experienced users of electolysis any practical experience of whether it makes a significant difference to smooth the supply. I have done a fair number of elecrolyses but always used a full wave rectified supply without any smoothing, as a typical car battery chrger does. It obviously is good enough for PbO2 deposition in that case.

If anyone has any web-accesible research papers on this subject I would love to read them.

I remember that a superimposed AC has been quoted as beneficicial in some cases, for polishing metal deposition. I am more interested in the effects of raw, unsmoothed supply vs. smoothed. As an electronic engineer (retired) I am capable of and do build my own supplies.

Regards,

DerAlte

Rosco Bodine - 29-6-2007 at 22:17

Definitely ripple can make a difference in scenarios where the range of the ripple voltage can cover the different redox potentials which apply to an electrolyte that can produce more than one product depending on the applied voltage . You can have a mixed product whose proportion is dependent upon the waveform .

Think about how metals are refined from mixed metal electrolytes , being separated according to their
reduction potentials ....and you will see this is how
either a single pure metal may be plated out using
a pure smooth DC , or an alloy of a specific composition
can be plated using some variety of baseline shifted
AC waveform . It is actually AC signal at the anode that never reverses polarity with respect to ground .


Even where only one product is possible , in some cases I think the yield may actually increase from use of an unfiltered or even a controllably pulsed DC which can cause the electrodes to hum at high current levels.....if they can physically withstand the punishment . You might see the yield go up a bit in some reactions as compared with using a smooth DC . It probably has a depolarizing effect and a physical resonance effect , sort of like a fluidized bed being vibrated ....keeps the reacting components agitated and colliding with some added physical and electrical energy .

not_important - 29-6-2007 at 22:18

Check the old book on electrochemistry in the library area of the site, it has some mentions of effects of ripple on certain reactions.

I've seen, as you, that actual reversal of polarity can be desirable for some operations including electroplating and electrorefining. I've also read descriptions of reactions where ripple appeared to be helpful, and others where they had to resort to a battery (1920s tech). I suspect that you'll find some reactions may care, that the low voltage periods give rise to other reactions and products, while in other cases it helps through depolarisation or giving time for diffusion to take effect.

DerAlte - 29-6-2007 at 22:36

Thanks!

@ Roscoe - excellent point about redox potential - should have occurred to me! Maybe Alzheimers has set in.

@ not_important, I'll look for that book. Incidentally, as regards a battery, you can always use a lead acid accumulator across your crude full wave rectified output- far easier than an inductor or trying to find a several farad capacitor.

Regards,

DerAlte

12AX7 - 30-6-2007 at 05:50

Incidentially, a series resistor, or a more constant-current supply, will avoid some of the problems associated with ripple. Given a constant current, the cell voltage is dependent on the cell's effective resistance. If the supply is lightly filtered and the cell supplied through a resistor, current will vary up and down and, if it works out right, will not go so high or low as to bring in side reactions.

By the way, at this current level, an inductor is a very good idea. Get a microwave oven transformer, cut it open, remove the windings and shunts and replace with 8 AWG wire (or whatever is sufficient for your current draw) to fill the space. Add a heavy piece of cardboard or plastic to gap the iron core by 1/16 to 1/8", clamp back together and you are done. The voltage output will be as high as usual when lightly loaded, but will drop substantially under a higher load. Ripple will remain about constant. The important thing here is the inductor stabilizes current, while capacitors stabilize voltage. Cells have a steep V-I characteristic, so going after voltage is less effective.

Tim

DerAlte - 30-6-2007 at 09:59

@ 12AX7 - I normally use auto headlight bulbs, series or parallel, or a sidelightlight bulb for lower current, to regulate RMS current by the variation of resistivity with temp of the filament, which gives fair regulation.

Reading your post triggered long distant memories from my tube days in electronics. Swinging choke! I Googled it and found a million articles. You might enjoy this one:

http://www.geocities.com/SiliconValley/Pines/5440/supply.htm...

Regards,

DerAlte

dann2 - 30-6-2007 at 15:41

Hello DerAlte,

A problem with ripply voltage/current supply is trying to measure the exact charge that has entered the cell. A simple current meter made to respond to DC will not measure a 'ripply DC. It will be close enough but not exact (if that is important). I don't know how close it will be... perhaps you can inform me?
Even if you do measure the exact charge (amps and time) entering the cell (use, say, a true RMS meter) with a ripply supply you will still be left with the niggling question of what is the 'effective' charge that has entered cell in relation to the reaction you want to happen. ie. different products for different (instantanous) anode current density and voltage.

The above is not too important if you are trying to simply make a product but if you are trying to measure current efficeincy a ripply supply complicates everything and leaves results that you obtain possibly not relevant to someone else using a good stable supply.

I aften though about putting an oscilloscope between a ripply (unsmoothed DC) and a Chlorate or Perchlorate cell and get a picture of the current (and the voltage too) going into cell. It would surly be quite pulsed in nature (more that the unsmoothed DC) because of the V/I charistic of the cell.

Regarding the redox voltages associated with certain reactions it is really the current density on the anode that will dictate what is happening. Voltages that the products see (at anode interface) are hard to measure. The Voltage accross the cell means nothing when quoted on its own as all the voltage may be lost in the bulk (anode and cathode far apart) or the electrolyte may be more dilute that the next guys cell or bad connection etc. Also the redox voltages are measures in all sorts of weird and wonderful ways (of which I know nothing). They are often refered to a 'saturated calomel Electrode' or a 'hydrogen electyrode' etc. They will not be directly applicable to your (or my) set up with (say) a Ti cathode.



Ramble over.

Dann2

12AX7 - 30-6-2007 at 17:23

Well, V = I*R, so average voltage is average current. All meters read DC through a filter (perhaps 20Hz or lower cutoff frequency), so ripple is ignored.

This isn't a problem. Thousands of electronic measurements involve knowing the DC voltage with some AC signal present. And for that matter, most AC meters read the average absolute value of the signal (adjusted to give the RMS value for a sine wave), so you can also measure the ripple voltage independent of the DC voltage present.

Note that ripple on the voltage across the cell is a potentially different problem, because a cell is non-ohmic.

Speaking of redox voltage, I noticed my cell gives around 1 to 1.5 volts when my power supply is turned off. Likely, the chlorine absorbed in the porous graphite anode is capable of generating a few miliamperes at a voltage near the reduction potential of the same reaction.

Tim

DerAlte - 30-6-2007 at 18:12

Nice ramble, Dann2. With regard to RMS and average, I should have said average as 12AX7 suggests. Only a hot wire ammeter responds to RMS and they're rare these days. A Dc mmeter reads average direct current.

Who cares about current efficiency except bulk or commercial operations? Or power efficiency? In my case I don't even care about eating up carbon in chlorate manufacture. (Though I'd like to try those PbO2 anodes you guys are trying so earnestly to make.)

I agree about redox potentials - we tend to make too much of them, but they do provide a guide to tell you whether a reaction will be spontaneous and a few other things.

I was wondering whether the diverse experiences people have had making chlorates (and especially perchlorates) actually depend upon the waveshape delivered to the cells. My personal experience says the electrode disposition has a lot to do with the success of the method. I use the sloping inverted gutter cathode (iron) with a close Carbon rod underslung. Others swear by other configurations but I've tried most and they just don't work as well, or chew up rods in bizzare ways. For perchlorate from chlorate, using carbon, I use side by side. It's not carbon or current efficient but it works.

The chlorate electrolysis depends on straight chemical action as well as electrolytic. It's important to mix the stream from the anode with the surrounding liquid and to have the temp as high as possible consistent with carbon erosion. My ClO3 cells tend to run hot (C. 50C). My test is nasal - if I smell chlorine rather than hypochlorite things are not right!

I always use sodium chloride sometimes with some KCl, converting to KClO3 by 'double displacement' later. If I have dichromate I use it (or a touch of permanganate); not convinced it's necessary but haven't done definitive tests.

ph is vital - never let the cell get much above 7.0 - this inhibits the disproportionation of the HClO produced. Nor too acidic. If you don't have a pH meter you have to guess. I just add a drop of HCl every so often and nose it for chlorine or ClO2.. I suppose monitoring the current could tell you something but I just make sure it is roughly constant throughout. Rough and ready until the final purification.

What started me off was thinking about anodic oxidation of MnO2. I intend to use a divided cell to avoid the hope-for KnO4- ions from getting ( mechanically) near the cathode and getting reduced by the H2 produced there. Then it struck me that a supply that causes the voltage to go to zero might let the ions creep back over the barrier and was hence considering a well filtered DC supply instead of the brute force full wave bridge. Any thoughts, anyone?

Regards

DerAlte

Rosco Bodine - 30-6-2007 at 20:01

@DerAlte

Hmmmm...redox potentials are absolutely relevant in
electrochemistry , it's like the "zener voltage" for
that particular reaction . You try to push more voltage
across it and the current and the reaction rate will increase to try to regulate the voltage , like a chemically
based linear regulator of sorts , ideally dissipating the
redox products as a result ....along with whatever heat
is the result of the deviation from 100% conversion efficiency .

For "proof" of reaction true current rate and Faradays passed
(1F = 26.80 Ah) , a standard hydrogen evolution cell can be placed in series with the separate cell in which an experiment is being conducted , and the volume of hydrogen evolved is actually a meter for the actual current passed .

A pH color indicator that could stay intact and functional
in a perchlorate cell environment would be a nice find .

Regarding linear PS ripple ,
A big old filter capacitor the size of a beer can is your friend , and a six pack is even better:P . For a chaser ,
a good old linear regulator stage to level things out .
But for a linear , it looks like about a pound per Amp is
about a minimum on what the end project will weigh as
a piece of hardware . So you end up needing to put
your project case on casters instead of rubber feet ,
if you get ambitious about the current capacity .

Myself and 12AX7 and Twospoons were hashing things out for awhile on how to most elegantly go about adapting a computer ATX switching PS to get some acceptable laboratory power supply function , and keep the weight and expense down .

I scratched my head for about a month , and came up with something that may / should work ( taking bets? )
and posted it in another thread . Here's the schematic
for Rosco's revolutionary , I wore out my calculator keypad on this totally original design .

http://www.sciencemadness.org/talk/viewthread.php?action=att...

from this thread
http://www.sciencemadness.org/talk/viewthread.php?tid=3885

So far this has proven to be a mental exercise as I have been too ass deep in alligators to have time to build and test the design . Anyway , it's definitely thinking outside
the box if you are intrigued by such things and have time to solder .

It's sort of like that TV commercial ....I'm not really a
technician ....but I stayed at a Holiday Inn last night
so I feel confident improvising :P

[Edited on 30-6-2007 by Rosco Bodine]

12AX7 - 30-6-2007 at 20:43

Already gave ample reasons concerning said circuit. It will not work.

A much simpler circuit, ranging from a variac and LC filter, to a fixed transformer with a more common linear regulator, you know the kind using 2N3055s, will work much better (which isn't really saying much, since "at all" is automatically 'much better' than "not" at all).

Rosco Bodine - 30-6-2007 at 20:56

Quote:
Originally posted by 12AX7
Already gave ample reasons concerning said circuit. It will not work.


Actually no , ......you didn't give precise or adequate explanation why it would not work , so let another person about three times your age and experience take a look at it . And point out to *them* what you say is the fault in the output transistor driver stage , which I would guess is what you still say doesn't work , even though that is the least controversial arrangement and is used in integrated form in commercially manufactured linear regulator chips . Your problem really is one of not invented here .
Quote:

A much simpler circuit, ranging from a variac and LC filter, to a fixed transformer with a more common linear regulator, you know the kind using 2N3055s, will work much better (which isn't really saying much, since "at all" is automatically 'much better' than "not" at all).


On your planet maybe , where the laws of physics and mathematics are different from here on earth .

DerAlte - 30-6-2007 at 21:20

You folks bring back memories! I used to be an old analog man. I’ve done almost everything from DC to light (no lasers, sad to say – well, only semiconductor IR types in the piddle watt region). I started out analog and went digital and theoretical – don’t ask me to design a digital circuit other than simple logic, but I’ll design you a digital filter and the code for a DSP. Well, I could once…. I’ve made circulators at 60 Ghz, wideband amps 0-20 Mhz, been in missiles, satcom microwave com and radar and things unmentionable. But I never made a decent switching power supply only old rectify and filter jobs, my first a 350v anode supply at age 14. Nearly electrocuted myself! 12AX7, you must be older than you sound from your inventive nature – if you know what a 12AX7 is; like a 12AT7, yes?. Enough retrospective BS and OT stuff.

I browsed the thread and scanned the cct diagram. My first reaction was, these guys are full of BS – it said voltage and current regulated! Should have said or, as I found out looking to the left. Touches I loved – using #14 Cu wire as a current measuring shunt; an array of 50A MOSFETs, and a programmable Zener ( ignoramus and out of touch that I am now, didn’t even know they existed!) A mere 50A, select your voltage range or current. Handy! No idea whether it would work, give me a week or two!

However, Rosco, I did not mean that redox potentials have no relevance in electrolysis. They obviously do. But there’s a tendency to rely too much on the Standard potentials from a table without taking into account the full formulation of Nernst WRT the effects of concentration, activity factor, and concentration and temperature . It becomes complicated. I think Dann2 rightly implied that such things are for the professional Chem eng and professional – the amateur can get by without them. I happened to be a theoretician by nature and a lover of mathematics but your average amateur is not. Sure you can measure current and calculate faradays but all that really matters in the end is did you get adequate product. The idea of a filter the size of a six pack has a strange appeal… or maybe it’s the six pack.

Regards,
DerAlte

Rosco Bodine - 30-6-2007 at 21:32

DerAlte has been tapping tubes with a screwdriver handle to find the culprit , run to the convenience store
where there was no ATM , but of course there was a
tube tester and a cabinet of replacements underneath
for those late night TV repairs :D

Probably even knows an Indian chief test pattern and
a newfangled nixie tube counter readout when he sees one :D

DerAlte - 30-6-2007 at 21:38

Yep! - DerAlte

Rosco Bodine - 30-6-2007 at 22:06

Been to the shoe store where they have the X-ray fitting
tester ? For those youngsters that never saw one ....
you could stand on the machine and look downward through the rubber surrounded mask on top so you could see the live X-ray image of your feet inside your prospective new shoes ....just to make sure the fit
passed a scientific examination :D

dann2 - 1-7-2007 at 05:19

Hello all,
Quote:
Originally posted by DerAlte


I was wondering whether the diverse experiences people have had making chlorates (and especially perchlorates) actually depend upon the waveshape delivered to the cells. My personal experience says the electrode disposition has a lot to do with the success of the method. I use the sloping inverted gutter cathode (iron) with a close Carbon rod underslung. Others swear by other configurations but I've tried most and they just don't work as well, or chew up rods in bizzare ways. For perchlorate from chlorate, using carbon, I use side by side. It's not carbon or current efficient but it works.


Regarding wave shape in (Per)Chlorate making.:This is a weak argument and not one that is illuminating to the problem but...I have never seen or heard of a study, patent or industrial set up that has suggested that some weird and wonderful waveshape will give you more effeciency. Therefor I am inclined to think that there is no magic or sweet waveshape. Steady DC is best (if you have it of course). There will be no complications to worry about.
All the big industrial set ups use a simply (boring) side by side arrangement of anodes and cathodes (usually flat plates). It always fasinates me why people come up with the inverted guttering (cathode) over the goughing rod when it comes to carbon anodes. The first time I saw this was back in the ninties.

When making Perchlorates with carbon, what current densitys, salt concentrations, temperature etc etc do you use?
How much Perchlorate do you get for the amount to carbon that gets used up.
Quote:


The chlorate electrolysis depends on straight chemical action as well as electrolytic. It's important to mix the stream from the anode with the surrounding liquid and to have the temp as high as possible consistent with carbon erosion. My ClO3 cells tend to run hot (C. 50C). My test is nasal - if I smell chlorine rather than hypochlorite things are not right!


pH controlling Chlorate makers are thin on the ground:D
If you smell Chlorine it means that the pH is lower that what a non pH controlled cell runs at (about pH 8- 9).
If you smell hypochloride, what does that mean?
Quote:

ph is vital - never let the cell get much above 7.0 - this inhibits the disproportionation of the HClO produced. Nor too acidic. If you don't have a pH meter you have to guess. I just add a drop of HCl every so often and nose it for chlorine or ClO2.. I suppose monitoring the current could tell you something but I just make sure it is roughly constant throughout. Rough and ready until the final purification.



How much acid do you add and how often?
Is it not a pain adding acid every few hours (or more often)?

It is by no means essential to keep pH controlled. You just get more current effeciency when you do controll pH.
Then there is the question of Carbon usage. When you control pH you get more current effeciency and therefor you get less carbon being eaten directly per gram Chlorate produced.
But perpaps you get far less carbon being eaten (along with the reduction from increased current effeciency) as reactions that take place in the bulk of the solution in a pH controlled cell do not now have to happen at the surface to the Graphite anode.
Perhaps theses reactions (the one's that take place in the soluton bulk in a pH controlled cell) are causing alot of Carbon erosion as they (or their equivalent) have to take place at the carbon surface??

Perhaps all Graphite set ups should be controlling pH in order to control the black mess.
Industrial set ups have very impressive carbon erosion per KG Chlorate make. They would put amateur cells to shame as far as black mess is concerned.


This is my take on Cell chem. etc.

http://www.geocities.com/CapeCanaveral/Campus/5361/chlorate/...

Regards,

Dann2

http://www.geocities.com/CapeCanaveral/Campus/5361/chlorate/...

DerAlte - 1-7-2007 at 09:23

@ Dann2 - I am afraid it is some years since I did the chlorate/perchlorate thing. I wish I could give some figures but my efforts were very amateurish and on a very small scale. Made about 100g chlorate and 50g perchlorate. I am not into pyrotechnics except to produce the occasional colored flame on July 4.

Most of the rods I used were out of large cells from an old camping lantern - the old really big type. Not sure they exist any more. And the rods all got eaten up. I used about 3v for chlorate across cell terminals, IIRC, current density unknown but high; I upped it to do the perchlorate with even higher current density. Cell size was only about 0.5 liter.

I am certainly no expert at this! I used sodium because I had a layer of salt at the bottom of the cell to keep the NaCl saturated. KCl is too soluble to do this, and harder to electrolyse, in my experience. But getting rid of Na from any KClOx made this way is next to impossible - not a good idea for pyrotechnical use.

My experience was that if you get chlorine, the cell is too acidic. If you can't smell hypochlorite, it's too alkaline. That crude! It tends to drift alkaline, so I'd add a drip or so of HCl until a slight whiff of Cl2 was noted - done with the power off and electrodes out of solution. I'd check every few hours, although I sometimes ran overnight. Tedious over a 4-5 day total run. Very crude but I do have a more sensitive nose than the average person. When I said I "Controlled" the ph at 7.0, I had no pH meter. Ishould have said tried to control, by this methodology.

WRT ripple, I agree that using a smoothed DC is always sound practice. Rosco pointed out above that in cases where you are trying to separate ions close in Redox potential it's probably essential. However, if only two ions are dominant, an anion and a cation, does it really matter?

@ Rosco - been there and done that! I can even remember the bombing in London in 1941-2. A VI landed close to our house in 44. The Jerries (now don't get chauvinstic, DerAlte) killed one grandmather in 1942. I am a true antique.

Regards,

DerAlte.

Rosco Bodine - 1-7-2007 at 10:07

Evidently the potential is also quite relevant to survival of the particular anode material being used , as being discussed over in the perchlorate thread where Xenoid
is experimenting with MnO2 anodes .

What voltage makes perchlorate needs to be well below what voltage depassivates the anode material and starts
it corroding away .

The Germans had a lot of nerve taking on Russia ,
and then the U.S. too ....but once they had pissed off the Lancaster clan ....it was all over for them :P