Historical research guide · 0756 · 2017
Observation of Excess Heat by Activated Metal and Deuterium Gas
Tadahiko Mizuno
In one minute
Tadahiko Mizuno lines the inside of a twenty-kilogram steel reactor with nickel mesh, cleans it until a mass spectrometer says the gas is clean, sputters palladium onto it with a glow discharge, fills the vessel with deuterium and switches on an ordinary heater. Then he measures, very carefully, how much heat comes out. His calorimeter is a stream of air through an insulated box: multiply the air flow by the temperature rise and you have the power, with a control reactor of the same design run at the same input for comparison. The control returns what you put in. The treated reactor does not. At 100 watts in, the output climbs to 180; at 248 watts in, it reaches 480, nearly double. The excess grows steeply with reactor temperature — about 100 watts at 100 degrees Celsius, 480 watts at 250 — and the surface photographs show why he is fussy about cleanliness: the reaction lives on finely divided nickel, evenly dusted with palladium.
Connection
Chapter 12 asks whether hydrogen’s isotopes will react inside a metal lattice rather than inside a plasma, and this is one of the most fully instrumented gas-loaded reports of it: a matched control reactor, a measured heat-recovery curve, an activation energy read off the temperature dependence, and electron micrographs of the surface that produced the heat. Its temperature scaling is the interesting part — the effect gets larger as the reactor gets hotter, which is the opposite of what a mundane artefact usually does.
From the source
The reactant is nickel mesh — two 300 by 300 millimetre pieces of 99.9 per cent nickel, 23 grams in total with 0.31 square metres of surface — pressed against the inner wall of a cruciform stainless-steel reactor of 2740 cubic centimetres and 20.3 kilograms, with a palladium discharge electrode, a ceramic heater and deuterium gas. Heat is measured by air flow calorimetry against an identical control reactor.Sections 2.1 to 2.3 and 2.6; Figs. 1 to 4
The control reactor calibrates cleanly: 100 watts supplied for 71.46 kiloseconds gives 7.15 megajoules in and an estimated 7.05 megajoules out, an output-to-input ratio of 0.986, with the shortfall attributed to radiation from the box. The same control was run at 80, 120 and 248 watts and returned its input each time.Section 3.1, Figs. 18 and 27
After the activation treatment the same reactor at 100 watts of input reaches 180 watts of output — 8.25 megajoules in against an estimated 15.19 megajoules out, a ratio of 1.841 — and at 248 watts of input for 22 kiloseconds the output reaches 480 watts, a ratio of 1.953, which is twice the input power.Section 3.1, Figs. 24 and 28
The excess power rises with reactor temperature: about 10 to 20 watts near room temperature, 100 watts at 100 degrees Celsius, 315 watts at 200 degrees and 480 watts at 250 degrees. Plotted against reciprocal absolute temperature the relationship is linear between 100 and 523 degrees, giving a reaction activation energy of 0.165 electronvolts, and the author extrapolates to kilowatt-scale excess heat near 700 degrees. Excess heat per unit of reactant was several tens of watts per gram, or 1 to 10 watts per square centimetre.Section 3.1, Figs. 29 and 30; Section 4, Discussion, opening paragraph
Four conditions are named as necessary: activate the sample surface into a fine structure and add a surface-modifying metal, cover that surface with activated deuteride, remove impurities from the gas, and control the reactor temperature and deuterium pressure. Electron micrographs show unprocessed nickel produces no excess heat at all, while a roughened surface uniformly dusted with palladium and free of oxide, nitride and carbon does — the smaller the particles and the more evenly the palladium is spread, the more heat appears.Section 3.2, Figs. 31 to 34; Section 4, Discussion, list of four factors
The fuel accounting is offered as a rough consistency check and nothing more: a thirty-day run at a typical 300 watts is about 260 megajoules, and if the reaction were deuterium-deuterium fusion consuming all the gas, roughly 12 cubic centimetres at standard conditions would be needed against the 20 that were used. The author states plainly that the calculation is speculation and that more data are needed to confirm it or to explain it.Section 4, Discussion, second paragraph
The way in
https://doi.org/10.70923/001c.72456How to cite it
Tadahiko Mizuno (2017) Observation of Excess Heat by Activated Metal and Deuterium Gas doi:10.70923/001c.72456