How to properly treat potassium-iron oxide catalysts

That, or possibly rust jacking.

http://www.cbiconsultingllc.com/the-power-of-rust/

Quote

We have known for many years that as rust forms on metal installed in masonry or concrete, the rust expands up to seven times the original thickness of the metal. The force of this expansion, if confined, can reach as high as 9,000psi.

http://onlinepubs.trb.org/onlinepubs/nchrp/nchrp_rpt_333.pdf

Quote

CORROSION EFFECTS [...] 4. Introduction of unintended movement According to one study, built-up corrosion product in constricted areas ("pack rust") can generate pressures up to 10,000 psi. This pressure can bend or move bridge components with damaging effects.

Curbina

I'm not 100% sure but it might also be the red iron oxide precursor material itself expanding as it reacts with potassium oxide (after CO2 has been evolved from the initial potassium carbonate) especially at high temperature. I recall reading that it hosts potassium in an "expanded, Rydberg-like" form. I don't know if this can be intuitively figured out from lattice parameters:

https://www.materialsproject.org/materials/mp-24972/

https://www.materialsproject.org/materials/mp-759749/

EDIT: here is a related excerpt from http://dx.doi.org/10.1088/0953-8984/19/27/276206 :

Refs 28 and 29 are:

[28] http://dx.doi.org/10.1023/A:1019013504729 (on Researchgate here)

[29] http://dx.doi.org/10.1021/ja0542901

EDIT2: here's another worded differently from http://dx.doi.org/10.1016/j.ijms.2010.12.008

With possible relevance to the latest tests I made, in some papers Holmlid suggests that the iron oxide catalyst material, at least when heated in a vacuum, looks dark (in the visible range) due to the formation of Rydberg states and Rydberg matter upon thermal excitation. At the same time, it appears it is a very efficient emitter in the infrared.

Here is an excerpt from http://dx.doi.org/10.1016/j.icarus.2005.09.004 :

It would be interesting (as well as surprising) if the reversible darkening effect I observed with my fresh samples was due to this (i.e. Rydberg atom and matter formation on the surface). At the very least it would imply that these states not as difficult to produce as often thought.

Also, if the material looks "cold" in the visible range, but "hot" in the infrared, this could mean that infrared probes or cameras may show higher temperatures than what could be measured with thermocouples. Perhaps some notorious results in the LENR field could even be explained this way.

EDIT: for the sake of keeping a record of my own tests, below are more photos related to the tests I did with last catalyst-resistor sample I made, until destruction.

Spoiler tag added to avoid flooding the thread with more photos and dilute away the discussion about more technical/theoretical matters.

EDIT2: here is another photo sequence (from photos taken on 2020-08-29) showing the color change from dark to red again upon cooling. I marked the time of the photo too here. Except for the last photo, camera settings and lighting conditions were about the same.

(Out of attachment space in the previous post so I had to make a new one)

I just finished assembling together into a composite image the numerous photos I took during the first test with the last resistor-catalyst I made, showing the surface appearance progression with heating. The image is 3436x2287 pixels large.

During the test I took photos both with and without flashlight enabled. It turns out that when it was almost black-looking, the resistor was already starting to get somewhat incandescent (about 600 °C). However, under low-light conditions / high ISO settings the camera-phone seemed to make it look brighter than it actually was by naked eye, relatively to the surroundings.

Alan Smith

That's certainly a possibility; perhaps even more so if there are excited (Rydberg) species or clusters formed on the surface of the pellet that emit strongly in the IR upon de-excitation (as theoretically proposed). These might not be just from potassium, but also adsorbed gaseous species like nitrogen (source). The emission of excited potassium species from the bulk would facilitate that of other adsorbed atoms and molecules on the surface.

I am not entirely sure if this is supposed to also occur in the atmosphere (under perhaps slightly oxygen-deprived conditions in my case), but as both this property and the reversible darkening in the visible range appear to have been related (by different authors) to the thermal emission of excited species from the catalyst, it might be.

Holmlid has proposed that the emission of excited species with temperature is a general property of non-metal surfaces like carbon or many metal-oxides, so there could be a correlation with the difficulties in pyrometrically determining the correct temperature of heated ceramics (which has caused many controversies in the past in the LENR world).

So, I was wondering if temperature measurements with a standard pyrometer on materials specifically designed to emit or diffuse out such states in large amounts would be able to clearly show that an anomaly on this regard exists.

With spot pyrometers there is the potential issue that the red laser spot does not necessarily represent the true measurement area, so if the measuring distance is incorrect (too distant), lower temperatures may be shown due to the instrument taking into account a larger viewing angle than that covered by the heated surface.

Lighter atoms/molecules and smaller clusters might shift such selective emission to higher frequencies though, so if processes similar to those described earlier are at play and involve H or H₂ in desorption, they could fall in the NIR or even visible range and possibly look cooler to the pyrometer and hot (or hotter) to the eyes. I don't have a complete understanding of the subject though, so this could be wrong.

This diagram is from fig.1 in https://arxiv.org/abs/physics/0607193

Quote

In Fig. 1, the different states involved in the stimulated emission process are shown. The RM clusters are formed by thermal desorption from the emitter, in excitation levels n = 20-30 [9,26]. The high excitation level of the clusters is decreased by efficient stimulated emission in the IR [8,9], and the cluster rotation is cooled down by stimulated emission in the RF as shown in the figure.

EDIT: I came to realize however that the same process may also be used to justify candoluminescence, which typically involves heating suitable metal-oxide materials in a hydrocarbon flame.

A few days ago I uploaded a more complete video of the gif I also posted here, showing the pellet turning dark reversibly upon heating.

I tried extracting video frames from the original video at regular intervals and calculating the average perceived brightness of a small area corresponding to the pellet after cooling (about halfway in the accelerated video above), which is related to the color. The samples from which brightness is calculated look like this:

It appears that the calculated brightness (color) decays exponentially after power is removed. The time on the X axis below is relative to the selected starting frame. Note that the Y axis is inverted.

I used suggestions from this link: https://stackoverflow.com/ques…e-brightness-using-python

I recently got a bench DC power and been trying with several things with it.

In the various tests performed, I've also revisited the previously mentioned process for forming the (presumably) catalytically-active green potassium–iron oxide layers on steel surfaces, which in short basically only involves applying a concentrated KOH solution on the surface and heating it up. I could now do it in a more controlled way.

Some time ago I found that the process apparently works also with ferritic stainless steel (FeCr) and even Kanthal A-1 alloy (FeCrAl alloy). In the latter case, it looks like the application of KOH solution (2M concentration) on the surface and heat interferes with the formation of the protective Al₂O₃ layer, and eventually a surface rust layer, probably mixed with various chromium oxides, forms. This is interesting because actual catalysts are formed by Fe₂O₃–Cr₂O₃–K₂O so the oxides formed with these ferritic alloys may approximate those catalysts better than just plain steel as I initially used.

Unusual colors (for Kanthal alloy, at least) could then be seen, as in the photos below. The green color in principle could be chromium oxide, but it immediately reverts to brown when wetted, so I think it is mostly the potassium–iron oxide compound discussed in this thread.

In a way, the process with Joule-heated wires is similar to the one used by Celani et al to form porous oxide structures on CuNi wires.

SindreZG

Besides testing my new adjustable power supply, I am just exploring the application on different surface types of the same method previously used to obtain on plain steel (99% iron) the known catalytically active K-Fe oxide oxide phase. Since such phase has an olive-green color and easily decomposes at room temperature with moisture (properties often mentioned in the related catalyst literature), it is easy to visually detect its presence, at least at high enough concentrations.

I also find interesting that stainless steel alloys that normally oxidize very slowly can rather easily corrode and form relatively thick oxide films when heated after applying a concentrated alkali (potassium, but sodium also works) hydroxide solution to their surface.

A related high-temperature oxidation method by Joule heating wires composed of different alloys (e.g. CuNi / constantan) has been used by other researchers like Francesco Celani to obtain LENR-active materials. Here, I'm doing it in a slightly different way with Fe-based alloys.

I am writing about these crude tests mostly for the sake of keeping a public record of what I have been informally doing and discussing elsewhere. There's not much else besides this.

In retrospect, an obvious observation, but heating with flowing air, especially if damp, will of course accelerate the oxidation process, making the outer olive-green layer acquire a more intense color (higher concentration). As the moisture in the airflow will be decomposed to oxidize the underlying surface, some hydrogen will be evolved too and possibly be retained within the porous oxide structure formed.

Recently I have been making tests with a semi-closed vessel and noticed that the resistance of Kanthal A-1 wires wires treated similarly to previously described would decrease noticeably upon admission of organic compounds like acetone or ethanol.

Acetone in particular would initially combust on the wire, producing a heat burst, but then, as oxygen gas got used up, it decomposed on its surface, causing carbon deposition and presumably hydrogen evolution. I think such hydrogen is what is causing this resistance decrease. At the same time, as the wire gets dark, incandescence becomes dim, perhaps due to the wire acting as a more efficient infrared radiator.

Since the cell is not sealed and air can come in (as well as hydrogen and acetone go out), the wires would eventually acquire back their oxidized color. As this happened, resistance slowly increased back to the starting values.

I tried this with a brand-new bare Kanthal wire and although to a much lesser extent, the effects were still visible (resistance decrease and incandescence dimming).

I then attempted this with a "powered catalyst pellet", which in this case involved embedding the same Kanthal heater in a 75% Fe2O3–25% K2CO3 paste and calcining it to high temperatures to form a hard ceramic pellet similar to actual industrial catalysts, only much larger (approx. 60x6 mm) and with an internal heat source.

Upon admission of about 1.5 ml acetone in the cell (involving temporarily opening it), the previously resistance decrease effect turned out to be much larger, causing a repeatable drop from 5.7 Ohm to about 3.6 Ohm. This drop was repeatable. Notice how turning power off in the second half of the diagram apparently did not affect the trend in resistance.

In the process (still keeping the catalyst at a constant 13V with the recently acquired power supply), the powered pellet first got covered in carbon, but eventually, possibly corresponding to the resistance minima in the above graph, the carbon apparently slowly started combusting and eventually mostly disappear. Resistance would then return back to roughly the initial values.

After a few mishaps following such testing, eventually the ceramic portion broke in half, although the wire didn't. After a period at room temperature the material looks red.

It's not clear how much this effect is related to hydrogen absorption, but it's interesting that it could be easily observed with this kind of material.

Alan Smith

Electrical conduction certainly occurs not just through the heating wire but also through the ceramic material into which the coil is embedded. More than just the thin layer on top however I think it might be the ceramic material itself becoming more conductive, possibly due to hydrogen-caused effects.