Part 1 - Mixing
Written by Bob Higgins on 9 May 2014
For those of us following reports from LENR researchers working in the Ni-H domain, it is becoming increasingly apparent that nano-scale features may be desirable for the reaction. The problem is that nanopowders, by themselves, are fragile. As David Nagel points out, the second hurdle to overcome in commercializing LENR will be to make the LENR materials durable enough to last for months or years while producing energy. Thus, it is desirable to attach nano-scale features to a more durable macro-structure. This blarticle [blog-article] summarizes the process I used to add nano-scale features to a macro-scale metal powder, but it could also be used with other macro-scale metal configurations.
Starting materials for this example are Hunter Chemical’s AH50 nickel powder (3-8 micron particles) and Fe2O3 nanopowder (Alfa Aesar 44895 iron III oxide 20-40 nm particles). Figure 1 shows an SEM of Hunter’s AH50 powder. Notice its large “external” surface area (as opposed to an internal sponge area like Raney Ni).
Figure 1: Hunter Chemical’s SEM of AH50 carbonyl nickel powder and TEM of Fe2O3 (courtesy Nanophase)
In Figure 1, note the ~100x difference in scale between the Ni micrograph and the Fe2O3 micrograph. The intent is to “activate” or catalyze as much of this external AH50 surface area as is possible using the Fe2O3 nanopowder having nearly spherical particles. The technique chosen is much like how a baker coats donut holes with powdered sugar: put both in a box and shake – well almost.
I should point out that the nanopowder must be handled in a dry (<8% RH) glove box of some kind for 2 reasons. First, the nanopowder would clump via hydrophilic and chemical bonding when exposed to humidity (think of dirt clods). “Clumping” would interfere with the mixing action and prevent spread of the nanopowder across the AH50’s surface area. Second, care must be taken not to inhale the nanopowder, which easily lofts into the air. The Fe2O3 is substantially less toxic than fully reduced nano-iron – one can easily be sickened by breathing even small amounts of reduced nano-metals (vs. oxides).
I cannot stress this enough: for your safety do not handle nanopowders with your unprotected hands or in open air where the nanopowder could be inhaled! Nanopowder will go right through breathing masks, cloth gloves, and even skin.
MFMP team member Alan Goldwater reported on his testing of a method for doing this in this blog, however, my mixing method begins with placing equal volumes of AH50 micro-powder and Fe2O3 nanopowder in a 100ml square HDPE bottle (Cole Parmer EW-06019-72; see Figure 2) while inside a dry glove box.
Figure 2: Cole Parmer square HDPE bottle for mixing (EW-06019-72)
First I add 40ml of the AH50 nickel powder (~64g) to the bottle and then 40ml of Fe2O3 on top (only 6.5g). The bottle is plugged, capped, and wired closed (to keep the cap from unscrewing). This bottle with the 2 powders is placed in a 125mm diameter rubber tumbler jar ( http://www.thumlerstumbler.com/rotary.html, tumbler and jar ~$100). The rubber jar keeps the HDPE bottle from wearing during rotary tumbling. The square HDPE bottle tumbles and bounces inside the rubber jar and provides mixing with no added ball tumbling media. The jar is tumbled for 24 hours. While an inexpensive rotary tumbler is shown, any rotary tumbler may be used, but the rubber tumbling jar is highly desirable. The tumble mixing is done completely DRY.
Figure 3: HDPE bottle in tumbler and inexpensive Thumbler’s Tumbler
When the HDPE bottle is removed from the rubber jar, the first surprise is that there is only 40ml of powder in the jar (Figure 4). Where did the other 40ml of powder go?
Figure 4: HDPE with tumbled powder mix
The SEM image of the mixed powder (Figure 5) reveals that the nanopowder has covered the external surface of the AH50 nickel powder (that’s where the nanopowder went).
Figure 5: Tumble-mixed AH50 nickel and iron oxide nanopowder
The mixed nanopowder should be unloaded in the dry box as well – there is still free nanopowder that can loft and be inhaled until after thermo-chemical processing.
While the Fe2O3 is ON the surface area of the Ni particles, it has not added nano-features useful in catalyzing a LENR reaction. That’s where thermo-chemical processing (TCP) comes in. By TCP, I mean heating the mixed powder in the presence of a flowing process gas (I.E. H2, O2, or Ar) at elevated temperature(s) and in cycles. TCP will be used to permanently attach the nanopowder to the Ni particles, to reduce the size of the nano-scale features, and to activate the nano-sites as LENR catalysts. TCP will be described in Part 2 of this series of blarticles.
Part 2 - Thermo-Chemical Processing
Written by Bob Higgins on 10 May 2014
In Part 1 of this series of blarticles, I described mixing a nano-scale powder (Alfa Aesar Fe2O3) with a micro-scale Ni powder (Hunter Chemical AH50) to spread the nano-particles across the high surface area of the Ni micro-particles. In this Part 2, I will describe the Thermo-Chemical Processing (TCP) that I use to anchor these particles to the Ni surface and activate them for possible LENR reaction.
The micrograph in Figure 5 in Part 1 above, showed the almost smooth “plastering over” of the high surface area of the Ni particles with the Fe2O3 nanopowder. The only portion of the Ni particles that can still be seen in this micrograph are the points of each flower-like bud. Why did the Fe2O3 nanopowder stick?
My best guess is that the Ni micro-powder had adsorbed moisture on its surface with an H-O-H attached to a surface nickel oxide oxygen atom as …-Ni-Ni-O-H-O-H. When the Fe2O3 is added, a loose bond comes from the dangling H atom as
…-Ni-Ni-O-H-O-H-O-Fe-O-Fe-O-Fe-O-…. Depending on the initial humidity, there could be longer chains of H-O-H-O-H … between the two surfaces.
Hydrophilic bonding is used commercially to bond flat glass plates together, for example to make hermetic crystal packages or optical interferometer components. Just take two clean, flat plates of glass, wet them, place them together, and heat. Initially each surface would look something like
…-Si-O-Si-O with a dangling oxide on the surface. The water chain between them forms
When heated, H-O-H groups drop out of the sandwich until you are left with only
and, at that stage the glass surfaces are permanently bonded. This also occurs in nature, agglomerating smaller oxides particles into larger clusters, and is one reason why nanopowder is not found in nature on the Earth.
By thermo-chemical processing (TCP), I mean heating the mixed powder in the presence of a flowing process gas (such as H2, O2, or Ar) at elevated temperature(s), and possibly in cycles. The set of temperatures, process gas species, and flow rate vs. time is a “processing profile”.
My vision is that the TCP profile should be designed to partly reduce the Fe2O3 nanopowder and allow it to partially alloy with the surface of the Ni powder – firmly attaching the nano-particle to the Ni surface. Reduced/alloyed Fe2O3 will leave nano-scale features on the surface of the Ni. Additionally, partly reduced Fe2Ox is a known H2 splitting catalyst. If you believe that nano-cracks are needed to stimulate LENR, then it may be desirable to go through cycles of reduction and oxidation (redox). Whenever iron is oxidized, it grows. [Ever see a rusty nail grow?] After reduction and alloying, subsequent oxidation may cause the iron to grow and wedge open a nano-crack. If you believe that LENR needs nano-scale magnetic domains for a BEC to form … well, Ni-Fe alloys have a high/very high relative magnetic permeability (µr) that may be favorable for BEC formation.
- Partly reduced Fe2O3-x is a known H2 splitting catalyst
- Cyclic alloying/redox of Fe2O3 nano-sites could induce nano-cracks [for E. Storms]
- Ni-Fe alloy spots have high µr which could help form magnetic domains to support BECs [for Y. Kim]
Figure 6 shows a schematic and photos of my TCP system. It consists of a porcelain 100ml crucible in a sealed stainless steel canister, inside a small box furnace (ThermoScientific Blue M Muffle Furnace with 150x150x230mm chamber), plumbed to the outside with coaxial fittings to provide for the flowing process gas and a cover gas. The H2, Ar (large), and O2 bottles are seen in the photo in Figure 5B behind the furnace anchored to the wall.
Figure 6: TCP system
Inert “cover” gas (Ar) is flowed around the canister inside the furnace box to insure any H2 leakage will not form an explosive mix (the furnace is only rated for inert gas). Manual valves and flow gauges are used to implement the processing profile in the present system.
In the future, computer coordinated temperature, gas selection, and mass flow control would be far more desirable. This development is currently underway.
The inert cover gas and the process gas mix in the chimney (at the top), where the gas is cooled in a coiled copper tube and exhausted to atmosphere.
While the described TCP vision is pretty simple, reality is more complex. Nickel oxide strips easily – before the oxide is stripped from the iron. Even at 400°C, the clean nickel particles will begin to sinter together. The Ni particle surface area is already coated with the nano-catalyst; as long as the sintered Ni form that results still remains porous enough for the H2 to get to the nano-coated area, it will retain its activity. Also, at the nano-scale, alloying and melting happen at about half the normal macro-scale particle temperature. This suggests processing at temperatures below ~750°C to prevent melting of the nano-scale features.
Figure 7 is a graphic representation of a processing profile that I found produced desirable features in the resulting nano-activated nickel.
Figure 7: Processing profile
When this profile was used, sintering of the nickel did occur and the result in the crucible was 1 particle – a seemingly solid chunk that came from the crucible in one piece (see Figure 8A&B). However, this chunk is brittle because it sintered only at the particle edges, and produced a solid body only connected internally by a porous 3D web. This chunk was readily pulverized in a homemade steel pulverizer (Figure 8C).
Figure 8: Thermo-Chemically Processed powder
Figure 8C shows relatively large powder pieces which were screened and re-pulverized until a mean particle size of ~250 microns diameter (but with a large spread) was achieved. Reports suggest that a mean particle size <15 microns is desirable. The normal way to reach smaller particle size is to take the pulverized powder and put it in a ball mill (ceramic jar with large ceramic balls, on a rotary tumbler) for 24-48 hours (this has not yet been done).
Other experimental processing profiles have been tried. The ones having longer reduction in H2 produced chunks that were too malleable to be pulverized. For this technique, it seems better to leave more oxide during the TCP, pulverize the powder, mill it to the desired particle size, and do any further reduction in-situ in the test cell.
Figure 9 shows a micrograph of the powder produced by TCP with the Figure 7 profile. 9A is a wide view. For those that follow descriptions provided by Andrea Rossi for his LENR powder, could the larger scale features seen in 9A be the “tubercules” he spoke of “growing”? [He never said that these would be nano features, only that you had to “grow” them.] Figure 9B is a zoom of the center of 9A showing Ni dendrites growing 100-300 nm long. The dendrites are only an observation about the material – there is no indication that these participate in LENR.
Figure 9: SEM of Thermo-Chemically Processed powder
Despite seeming to be a solid chunk, this material is highly porous – ambient H2 would be able to get to the activated area. Also, I can say from personal experience pulverizing this material – it is mechanically robust. If it also proves to be an active LENR material, it stands a good chance of answering David Nagel’s challenge to make the LENR powder durable.
Early testing of this powder showed what appeared to be LENR outbursts. To find out more, read my full paper available at: Surface processing of Carbonyl Nickel. Future information on testing of this powder and similar replications will be posted to MFMP’s Blog.