Open-air hot powder cell—costs and planning

  • If I were to build a hot cell as depicted in the diagram below, how much could it cost just in materials and driving equipment, and what safety measures would be worth investing on with such experimentation?

    Roughly speaking the outer tubular body would be about 8–10 cm high.

    My suspicion is that the main difficulty and likely what could drive costs up would be having an internal cartridge heater and cables rated for the temperatures required. Thermocouples, other sensors and related controllers and loggers have been omitted in the diagram but would probably also affect costs significantly. Catalyst costs on the other hand would be minimal as long as no rare/precious elements or isotopes are used.

    The optional other oxides mentioned would be for example ZrO2, TiO2, SiO2, CeO2, etc.

    From past observations in semi-electrolytic experiments at low temperatures and narrow-gap electrodes, passing a current directly through the catalyst (composed of a similar material) could lower the operating temperature range to 300–350 °C, but actual feasibility with the configuration above would have to be verified in practice.

  • Cerium oxide (purity perhaps a bit questionable) can be obtained from glass repair shops cheap, (free even sometimes if you don’t need much and are polite).

    Cable extensions for the heater can be made with Kanthal wire significantly larger in diameter than the element thickness so that normal wire can be connected at a cooler position. Usually about 15 cm of extension is plenty. These can be crimped on with stainless steel, uninsulated butt connectors, which can be made from thin tubing or the real thing ordered from an electrical supplier. (Use a good quality crimp tool.) Ceramic weave sleeves are available to cover the bare Kanthal, and this sleeve material can sometimes be scavenged from halogen light fixtures.

  • I will keep in mind glass repair shops as a low-cost CeO2 (and SiO2) source. By association that reminded me that ordinary beach sand would be a zero-cost (albeit low-quality) source for SiO2, CaCO2 (which would become CaO upon heating) and various organic impurities which shouldn't be of significant concern, but the sodium chloride would have to be washed off beforehand.

    Thanks for the tips about the wiring. Besides how to design it so that it will not fall apart in the above configuration, what I was not sure of is mainly if affordable cartridge heaters suitable for sustained temperatures up to 800 °C and short-term peaks up to 900 °C exist.

    It's possible though that such a high-temperature heater might not be that required. Upon heating, the main constituents of the catalyst material Fe2O3 and K2CO3 transform into an unstable green compound called potassium ferrite (KFeO2) which among other things is also (especially upon doping) a good fast ion conductor / solid electrolyte. It is conductive at high temperatures and the heater could be potentially turned off if a large current is passed directly through the catalyst. In practice so far I've only made short energetic fizzes through narrowly-spaced electrodes with it however, so it's not clear if it could work as I think here.

    Some random examples retrieved from Google Scholar on the potassium ion conductivity of KFeO2 :

  • The basic set up can probably be built for around €100 or less. To drive the heater you would need a PID thermostat, SSR, (control relay) and K-type thermocouple. Ebay is offering this trio for around €12 for example. You would need a longer-reach TC than this but you can buy 100mm or longer K-types for €5 plus post.…6e4f95:g:b5AAAOSw0eNdBwSF

    A 240v cartridge heater is also cheap - these types normally come with glass-fibre insulated wires as well.…d4c744:g:KNIAAOSwEjFXeh00

    You might also want to power this via a triac voltage control - this means you can control maximum temperature via the PID/SSR/TC combo but also the heating rate -these are ludicrously cheap, for example.…9047dc:g:tnwAAOSwDOpbzWaJ

    For a reactor containment I suggest you use a stainless-steel vacuum flask stripped of any plastic parts. Like one of these - 630ml capacity should be enough.…id=p2047675.c100008.m2219

    If you want more heat-resisting wire, then something like this should do…id=p2045573.c100281.m3567

    And for connectors, something like this.…faa6f1:g:d2wAAOSwBLlVJjaI

    Any wiring problems I would be pleased to help. A final thought- these parts are so cheap, I would buy 2 of everything if possible- especially the heater and TC;s you want. Cheap TC's (even stainless steel k-types tend to die after long exposures to 800C plus. In the subject of TC's btw, I am not sure what the effect of molten Potassium Carbonate on the sheathing would be.

    AND - No- you cannot add water to molten salts at anything over 200C without risking a steam explosion. And even below 200C I would be very cautious.

  • Steam explosion is indeed a risk if there is a phase change with increasing temperature. That can be minimized according to the salt concentration and how fast the temperature will increase around the phase change temperature.

    If you plan such ingeresting experiment, I can participate financially. Just PM me.

  • Alan Smith

    Many thanks for the parts list. At about 100 € that's much less than I thought, but I very rarely purchase parts on ebay. So those typical Chinese cartridge heaters are fine for 800 °C continuous (under oxidizing conditions), aren't they? I didn't think they would be up to the task.

    As for the active material, by the time iron oxide and potassium carbonate in those proportions have reached 800 °C, most (if not all) the potassium carbonate should have reacted with the iron oxide, forming a solid material. On its own K2CO3 melts at 891 °C, so melting should not be a concern, but excess water might convert some of the potassium ferrite to Fe2O3 and KOH, the latter being liquid above 360 °C and caustic at all temperatures.

    More than water traces possibly violently turning to steam as they come in contact with the hot material (which I guess could be defined as an explosion) a possible concern that I have is that heating the solid material too quickly might cause it to explode from the pressure of the internally trapped water, like it can sometimes happen with carelessly fired clay or with viscous magmas.


    The catalyst material should be solid at the operating temperatures of interest and at least up to 1200 °C, but the phase change of water from liquid to gaseous state could cause issues. To minimize them ideally it would be supplied already in the gaseous state, but I imagine this would require pre-heating or a vacuum.

    Similarly formulated catalysts are often used in the petrochemical industry with steam and hydrocarbons, at operating temperatures in the 600 °C range. Ceria addition to K/Fe oxide catalysts is for example studied here (in a different application), so I'm not really making anything novel:

    The main difference from typical industrial environments would be heating patterns and the high-temperature electrolytic arrangement when used.

    If costs are in the order of what Alan Smith is suggesting above, at least to materially arrange the experiment—monitoring and detection equipment might be a different story—there's no need for financial contribution, but thanks for the offer anyway.

  • More than water traces possibly violently turning to steam as they come in contact with the hot material (which I guess could be defined as an explosion) a possible concern that I have is that heating the solid material too quickly might cause it to explode from the pressure of the internally trapped water, like it can sometimes happen with carelessly fired clay or with viscous magmas.

    That is the reason I suggest you use a Triac voltage controller in line with the PID- you can 'bake out' any trapped water by keeping the heater voltage down below (say) 180V - the slower rate allows time for water to evaporate safely and also prevents the formation of 'super-hot' spots up against the heater wall. Those heaters are glow bright red at 800C.

  • I think you would also find some foamed alumina fire-bricks useful. These are sold for building pizza ovens and ceramic kilns. Lightweight and easy to cut and shape with woodworking tools, you will need to shape one to hold the hemispherical base of the vacuum flask and possibly another to cut for a lid. If you are good with hand tools you might be able to get a base and a lid from one brick.

  • Alan Smith

    For the cost it would be a pity not to add a Triac voltage controller, although it would make determining input energy more difficult when used. On the other hand passing a current through the catalyst might too be complex on this regard as its resistance will vary dynamically depending on temperature and other operating conditions.

    Regarding water, I argue that it must be present in some amounts inside the catalyst's pore system, but not so much that it ruins the material, and of course not in dangerous amounts for the temperatures used.

    Interestingly some authors propose that acidified water in certain materials like clays (etc) may form dense hydrogen clusters which would release large amounts of energy upon formation, possibly explaining some natural phenomena inside the Earth. See Mayer and Reitz here and a related excerpt below:…y_Generation_in_the_Earth

    As for the thermal insulation, my original plan was having a "fluid" setup that allowed reusing the parts with different configurations (and lower costs) as much as possible, so I thought I could use refractory mats rather than bricks to be drilled and shaped with appropriate tools which I might not have yet. But nothing has been decided yet nor fixed in stone.

    I'm aware that you have published tips on shaping alumina bricks for insulated tube furnaces in this presentation:…sign-and-build-final-pdf/

  • Wyttenbach

    I haven't considered them so far, but upon reflection, graphite crucibles would slowly burn in the oxidizing conditions at relatively high temperatures I plan using, which could be both a good a bad thing. The worst is that the continously evolved CO2 could poison the activated K-Fe oxide catalyst material, which is very sensitive to this gas according to the literature. Excerpt below from :

    I do realize that I listed graphite in the diagram in the opening post, but I believe that at the concentrations that I plan using it would also turn into CO2 and probably disappear within short periods of operation. It would reduce somewhat the catalyst material in the process, which can help forming its active ferrite state, which at least at lower temperatures requires that some of the iron oxide is reduced. This isn't so necessary at temperatures in the order of 800 °C or more.

    Under different conditions a layer (possibly monolayer to a few atoms thick) of graphite on the catalytic material is reported to be useful though, and I'm aware that Leif Holmlid thinks it's very important for the activity of his catalysts (which are fundamentally of the same type I described so far in this thread) in producing [ultra-] dense hydrogen—under a hydrogen atmosphere in a vacuum.

  • Dr Richard

    The underlying idea is that as water molecules diffuse through the tunneled nanostucture (pore system) of the catalyst held at high (magmatic?) temperatures, possibly with the help of doping with selected oxides improving its acidic nature (on this regard see Wikipedia: Solid acid), some of the protons in the H3O+ ions will be converted to what could be considered Holmlid's ultra-dense hydrogen, with immediate release of condensation energy inside the material (in the order 1–2 keV per H-H pair, possibly more according to Mayer and Reitz—3.7 keV).

    Practically speaking I do not expect much to occur until electrolysis is applied and water more directly dissociated, although it is suggested that the often explosive effects of water absorbed into clay-like materials heated to high temperatures might be at least partially due to the condensation energy of such dense form of hydrogen.

    Due to the open nature of the cell using D2O would be rather wasteful. However, according to Holmlid's suggestions, deuterium might condense more easily to the ultra-dense state and release more energy in the process due to the shorter interatomic distances on average than protium and D+D fusion possibilty, so it could help improving any result if used. From Holmlid's publications, the spontaneous meson-emitting nuclear reaction capability should be about the same as with protium, but most of the energy from these reactions would be be deposited away from the reactor itself, essentially almost not showing as excess heat unless serious steps towards thermalizing the particles are attempted.

    I have to clarify that this is mostly a brainstorming thread on a possible experiment I might or might not do and I have not decided anything yet. There are safety issues to be considered besides costs.

    • From Holmlid's publications, the spontaneous meson-emitting nuclear reaction capability should be about the same as with protium, but most of the energy from these reactions would be be deposited away from the reactor itself, essentially not showing as excess heat unless serious steps towards thermalizing the particles are attempted.
    • Build a Mizuno-type reactor around the hot powder cell to initiate muon-catalysed fusion reactions (or simply replace the heater in Mizuno's reactor with a UDH muon-generating hot powder cell). I think the infra-red radiation from the internal heater acts in a similar way to the Nd-YAG laser pulses to stimulate meson emission, which yielded sustained 3KW excess heat in Mizuno's R20 - the UDD perhaps formed by trace metal oxides. The failed replication by Deneum, transients observed by Zhang may have been due to lack of metal oxide catalysts/high D pressure - making the KFeO2 an integral part of the reactor in close apposition to the IR heater source might realise much larger excess heat power output.
  • Wyttenbach

    Definitely containing costs is a major priority. My initial idea was even using disposable stainless steel cookware—this is the degree of cost reduction we're dealing about.

    Dr Richard

    For the same reasons as above, a sealed high-vacuum cell with gaseous hydrogen (D or p or both) is off-limits for me as well. In any case, graphite (possibly with metal carbides) or surface oxide impurities could be a reason for the variability between various Mizuno-type experiments, as according to Holmlid, if you think his model is applicable to them, the H Rydberg matter conversion process will not occur easily from pure metal surfaces (including internal surfaces like nanopores, nanotunnels, cracks, etc). Other factors may be at play however.

  • Just a suggestion - maybe more appropriate for somebody doing a full Mizuno replication since the equipment is v expensive - but the principle of surrounding the internal heater with a KFeO2 catalyst for UDD synthesis & meson emission then inducing -muon catalysed fusion in the Pd/Ni mesh seems somewhat attractive as a testable working hypothesis.:)

  • can , i have some excess stainless steel tubing available od 10mm and 25mm (rated to 1100°c) and i am located in Austria. As an efficient heat source a ceramic glow plug could be used to be threaded in pipe fittings. It can easily handle 1000°c

  • gerold.s

    I might actually be interested more in materials and equipment than direct economic support as basically I have little idea of where to get suitable quality parts besides large mainsteam shops. However since what I'm proposing in this thread might not necessarily work as I expect, I'd like to avoid an excessively custom and potentially expensive, over-engineered approach. Glow plugs might be convenient to use, though.

    This said, on a related note Holmlid has sometimes stressed (albeit not strongly enough in the published work) that strong magnetic fields have to be avoided as they are known to remove and prevent the further formation of the so-called long ultra-dense hydrogen clusters, so in this respect stainless steel tubing (normally non-magnetic) might be desirable. A similar quenching effect is reported for electric fields. Protium might possibly be more sensitive to this than deuterium.


    For these reasons, it's possible that in these tubular reactors the condensation of hydrogen into the ultra-dense form and subsequent energy release might mostly occur during the "off" periods if they are long enough, hence the pulsed DC approach suggested in the opening diagram.

    The main reason why the ultra-dense long clusters seem to be so easily disrupted despite suggestions of having a bonding energy per H-H pair in the keV range is that they are actually still highly excited and haven't released their condensation energy yet to the surrounding environment. See here section 9:…02-4896/ab1276#psab1276s9


    [...] The transfer from H(1) to H(0) is quite complex, since the energy given off by the H(0) cluster formation will mainly be taken up as rotational energy in the clusters. Due to their super properties, they will not easily transfer or lose this energy to the surroundings. This process is included in figure 12, and this figure indicates that the higher state H(1) will be reformed, if the excess condensation energy cannot be removed. Thus, the spontaneous condensation to H(0) is normally a slow process.

    I believe however that a weak and permanent uniform magnetic field might actually help the condensation of the highly polarizable Rydberg atoms emitted from the catalyst into ordinary Rydberg matter (the precursor to the ultra-dense form).