Quote from magicsound

From an online description, I think the 100-lire coin is made from Cu-Ni coinage alloy of around 70% Cu / 30% Ni. The ferritic stainless steel alloy composition is unknown but would typically be at least 82% Fe (430 Alloy).

Now I see why you mentioned Cu-Ni. It looks like the one you linked was a newer series.

This is the coin I used, quoted from the same website to be composed of stainless steel:

https://www.leftovercurrency.c…-coin-minerva-large-type/

EDIT: I probably wouldn't have used the other coin type because it would have not been as large as I needed and I wanted to have mainly Fe-Cr-K oxides at the coin interface without involving significant amounts of other elements.

magicsound

To further clarify the underlying idea, the experiment was devised so that slight amounts of water (electrolyte) could be introduced from the opening on the top electrode and diffuse evenly through the gap, which surprisingly it generally did. So it was convenient that the bottom electrode was larger than the top (that coin is pretty large with a ~28 mm diameter), that the top electrode had a hole in the center and that they were both disk-shaped.

Since I don't have suitable equipment to make clean holes through hard alloys, a coin and a washer conveniently worked as the materials for the experiment. That they are both ferromagnetic incidentally made it easier to hold everything more or less together with a relatively strong magnet.

In a more professionally done experiment with the same "semi-wet" concept, larger sheets would be more convenient to use and allow to clamp everything together firmly in place. With square or rectangular sheets however there might have to be several openings for the water or electrolyte to diffuse properly everywhere.

Today I attempted an experiment with "wet" electrode plates of unknown steel alloy (still ferromagnetic) similar to another I described earlier in this thread, again with minimal gap. It appears that at a high current with the electrodes very close together there is minimal oxidation at the anode, which remains clean (and even gets cleaner in the process), while the cathode becomes black (this is what i used to observe in the tests described in the opening post). What looks like cavitation damage is visible at the inner surface of the anode.

EDIT: attached experiment notes and observations (WIP); the document also contains other photos.

How much of a "battery" effect should be expected from two stainless steel electrodes shaped and arranged as depicted in the previous post?

Today I found that after a period of operation the cell would continue producing gas for quite a while at a moderate rate. Although a video is no proof, I made a couple to show what I mean. The anode (left side) was not connected in both cases (I also tried later on to disconnect the cathode/- and it would be have the same). The second video was made a few minutes later; gas evolution was lower than in the first so the cell was discharging, somehow.

https://streamable.com/rox56 

https://streamable.com/heka6

(I'm aware that the videos are of rather poor quality and the environment is messy)

EDIT: to be fair, a possible skeptical explanation could be that it was outgassing previously electrolytically absorbed hydrogen rather than dissociating water.

magicsound

The electrolyte was indeed (ended up being) roughly 0.5M KOH + undefined amounts of K2CO3 (about 1/4 volume in 0.50-0.75mm granules of that 25ml jar). I could try again with a better defined (and new) electrolyte solution.

Before starting yesterday's tests I sanded off the remaining yellowish surface layer that didn't get removed the day earlier during electrolysis. It was not difficult to remove with 100-grit sandpaper; it must have been only microns-thick. Such layer made also more difficult to pass a current through the exposed portions of the metal pieces, so I figured that it was some kind of anodizing finish. More suitable technical terms for that probably exist.

The surfaces in the above photo have been intentionally left rough, which I thought would promote the formation of cavitation bubbles. I recall reading about this on a presentation by Moray B. King on water electrolyzers (here on page 105+).

After the test a significant amount of solid strongly ferromagnetic particles that would be easily attracted with a magnet were produced. In the photo below the magnet appeared to be picking also a lower amount of ferromagnetic iron oxides. I believe the ferromagnetic particles come from cavitation erosion.

Unlike other pieces of metal I have around the electrodes do not seem to easily rust after being put drying in a warm place, so the base alloy is probably still some kind of stainless steel.

During testing I found that magnet application would have some sort of effect on gas evolution. Sometimes it would get higher, even during off-power conditions. Upon closer inspection it appeared it promoted the formation of spikes on the cathode.

The rectangular piece on the right of the above photo is the bottom insulator which came off during operation. It looks like once electrolysis and cavitation start occurring at the interface of both electrodes it's not really that necessary anymore.

For the record, I repeated the test after cleaning the electrodes in diluted HCl and using a new electrolyte solution (2M KOH), but couldn't obtain the same results as yesterday. No thick layer was formed between the electrodes and very limited erosion occurred.

Details and photos in the spoiler tag.

So what I've been doing? Yesterday I prepared a 75 turns, 4cm height, 5cm radius coil (and roughly 1.1~1.2 Ohm) with an inner diameter about as large as the jar I've been using so far, using regular thick insulated electrical wire.

Since the electrodes are mildly ferromagnetic they will act as a magnetic core and concentrate the magnetic field generated by the coil. A current (under stabilized conditions) of about 3A when 12V is applied is enough to cause the jar to visibly shake. I've made a video today but unfortunately it's shaky and it might look as if I was moving the jar with my hands.

https://streamable.com/3x47s

According to this calculator the magnetic B-field for a relative magnetic permeability of 10 should be 0.07T. I think the magnetic permeability might be higher, or perhaps current is higher than I'm assuming.

I haven't observed any increase in Geiger counts while doing this, but on the other hand they coincidentally decreased once today I increased electrolyte conductivity and started applying a significantly higher current than I've been since yesterday. Only problem, in my case such reduction even though noticeable is within the very wide error margins for my Geiger readings in general. Also it's probably a coincidence since generally I get the lowest readings of the day at about that hour (rightmost dashed line) due to the periodic signal I am getting (which is lower than it previously was in the current Geiger counter location).

At the moment the jar is not operating / is turned off.

Depending on the electrolyte solution and the conditions of the electrode gap interesting reactions can occur. Thinking that a slightly acidic solution would promote rust/hematite formation I used 25 ml distilled water and literally one drop (roughly 0.06g) of 10% HCl. This made electrolysis and deposition occur differently than usual, making the solution turn black (probably due to iron chloride formation), the small gap to fill and a kind of "welding" reaction to occur as soon as the gap became wetted again. Next time I should probably try adding the same small drop of HCl only after an initial rust layer has formed (which it will eventually do with just distilled water).

https://streamable.com/wvc9h

Unfortunately still no increase in Geiger readings observed (again, if anything they appear to decrease).

To the best of my understanding below is what I think occurs, after some more testing.

As soon as even traces of HCl are introduced in the aqueous solution, the anode starts corroding quickly, likely more on the zones of highest current density, making the solution black very quickly (in the case of Fe, at least) from very fine metallic particulate (quite difficult to clean if it gets into anywhere porous). The deposition rate on the cathode increases significantly at the same time.

This allows for interesting effects. That material deposition increases to a very quick pace also means that sooner or later short-conduction paths will form between both electrodes if the gap is short enough (also at substantial fractions of a millimeter). If the power supply system is up to the task, these short conduction paths will turn the material into plasma and destroy the previously formed path.

In the early phases these short-circuit events transiently occur on multiple spots. From the AM radio I use to monitor the reaction I think this can be heard as a broadband noise event which can turn into a constant white noise stream after a while. However, later on depending on the conditions at the electrode can develop into a sort of welding rod effect and be difficult to recover from. When this happens the broadband radio noise stops as well.

The cathode wears up negligibly in the process, while the anode sees most of the damage.

This interesting effect could potentially be taken advantage of to obtain a sort of electrolytically-induced dusty plasma reaction. Locally the temperature would be very high (up to thousands of degrees), even if overall the environment would be within typical electrolytic temperatures.

During a short-circuit event electrolysis stops, of course.

Unfortunately I haven't seen any large change in Geiger readings that could be clearly associated with the experiment. The segmented red lines below show when:

1. I powered up electrolysis at 12V and distilled water (leaving the cell for the most part unattended at low load)
2. I added a few traces of HCl
3. I ended the experiment

Earlier today I tried swapping the electrodes (usually I use always the same ones as anode/cathode) to check out exactly how quickly the anode gets eroded in the process. It appears that it occurs very quickly. I haven't even added more HCl, just used the previously formed solution (probably very diluted FeCl3) which should be slightly acidic but not normally corrosive at this concentration.

I also added the previously made inductor in series with the circuit, but not with the jar within it. Instead I placed various ferromagnetic pieces on its center to increase its inductance. The main reason for this was to improve reliability as short transients would in this way hopefully be less likely to draw too much current from the power supply at once.

It appears to have partially worked in that I was able to sustain a current at 12V longer. In the process quite curiously the jar itself would emit screeching noises (both physically and through the AM radio) which I'm assuming were from cavitation and/or arc discharges that were more likely to be maintained longer. From the vibrations that could be felt through the tabletop I think it might have been cavitation. Do the photos show cavitation damage on the cathode or just plain anode erosion?

Interestingly, but not so unexpectedly, the area where tape was applied in order to increase spacing didn't get eroded at all.

To improve the process further I might have to add a few big capacitors in the circuit in order to deliver larger amounts of current during the various short-circuit events and further improve reliability.

The process appears to produce overall particles so fine that to a large extent (a chunk does) they cannot be all collected with a magnet and leave the solution pitch black . However I can't rule out that they are due to something else, like for example residues from the electrical tape.

I was thinking of using thin mica spacers normally used for electrically insulating MOSFETs from their heat sinks, which shouldn't easily break down with heat, although perhaps better options exist, also for keeping the entire assembly together with minimal costs and tools required.

There are reports of neutron emission from the cavitation of FeCl3 solutions so I'm moderately optimistic that once the bugs are solved perhaps something might be able to be observed even with the Geiger counter at my disposal. However this might end up requiring better/different detection methods than what I'm using at the moment.

http://aflb.ensmp.fr/AFLB-342/aflb342m669.pdf

(The authors observed an increased neutron background, but not gamma emission)

Wyttenbach

Since a large current is being passed (at least several amperes, although it could transiently be much higher if I add a few large capacitors), a relatively large inductor is (or can be) present in the circuit, wouldn't the collapsing magnetic field from the repeatedly formed and suddenly destroyed conduction paths at the electrode interface locally produce potentially large voltages? Of course the main disadvantage is that these wouldn't be easily controllable, but given enough running time perhaps ideal ones could be generated under the right conditions.

Alan Smith

Do you mean something like this?

Actually I was looking at insulating spacers in the order of 0.1-0.2mm of thickness. The mica ones I was thinking of getting (they're not expensive, I just don't like purchasing things unnecessarily) were 0.09mm thick.

It's indeed possible that Fe3O4 is a large fraction of the particles produced, but since they seem to be rather conductive and to come mainly from the electrode interface where equally active hydrogen is being produced, I thought it might not necessarily be the case. I haven't tried to isolate them yet, or at least the fraction that gets attracted through the jar to the hard disk neodymium magnets.

I had a lower grade of those scouring pads laying around and couldn't help trying out if a larger spacing would work better under my conditions where the power supply system just isn't capable enough. I didn't put them all over the electrode surface, but only as tiny strips of insulating spacers, leaving an empty space between both electrodes. After compressing them they become quite dense. After some time I added a tiny drop of HCl to increase solution conductivity since it seemed to be progressing rather slowly, the reaction increased exponentially.

The sparks on the exposed area occurred later in the process, but the hissing noise and vibrations started earlier. This is roughly what I observed early with a narrower gap, only apparently stronger this time.

Unfortunately the solution is so dark that it's impossible to tell what's going on there, but I think the same is occurring below as it is above the water level. I turned it off after it seemed to start becoming too energetic and a bit dangerous with the foaming occurring there. The noise was coming from the jar itself, it's not just the radio.

https://streamable.com/3vzp4

https://streamable.com/gvvb5

Unfortunately I ended up spilling the contents of the jar due to the unbalanced weight of the electrodes and the test had to end.

Even with the larger gap (slightly less than 1mm) I have problems with the PSU voltage sagging and not providing enough power to vaporize the some of the thicker conducting pathways forming between both electrodes, so eventually it either shuts down due to triggering its overload protections or overheats the wires without being able to restore impulsive conduction (mixed electrolysis-arc discharge?) status.

I haven't been able to observe any large and/or short term increase that could be associated with the ongoing reactions, but coincidentally CPM appeared to decrease twice as I started the experiment initially and as I added a tiny drop of HCl (first and second dashed lines). However it's too easy to read too much into these minute variations. The third dashed line denotes when I ended the tests, which haven't been continuously performed due to PSU issues. So, overall no significant change appears to have occurred.

Still, I think if improved this could have some potential to show interesting results.

EDIT: earlier I forgot to post this video which better demonstrates the screeching noise I was mentioning about:

https://streamable.com/ycfht