This is getting most interesting. Moller threads together some ideas derived [perhaps not correctly] from Langmuir and builds a cell. Naudin runs tests on the cell and finds interesting apparently OU heat anomalies. The cell is an ideal black box and we don't get to peek inside to see what is going on, or to test our speculations about what is going on.
Also, reading Moller's notes and speculations are one thing, reading the Langmuir paper is another, and Naudin's tests are still another thing, all loosely connected. It can be difficult to keep track of the pea under the moving shells. So far there are several candidates, to wit: 1) Dissociation-recombination, loosely related to the performance of plasma torches. 2) ZPE as the primary energy source for 1) 3) LENR reactions based on creation of a nuclear reactive sites on the filament 4) BLP reactions based on autocatalysis of H as in 2H+H, production of hydrinos, and further catalysis cascades Peter Gluck has made calculations of the total Wh of energy realized from the charge of H with BLP reactions, which are less than claimed by some of Naudin's runs. However, Peter's calculations were first based on H(1/2) as the end product, whereas H(1/11), even H(1/16) have been seen in BLP spectra. In none of these cases was the gas as dense as MAHG, nor were the BLP observations for closed cells. So we may safely speculate that with hydrogen only, in a closed cell, a very large amount of energy could be released by BLP reactions already observed and reported. Yes, these reactions could go to completion, as Peter and Jones have observed, but none of Naudin's runs have gone on long enough to test this. Two hours just isn't enough. Nor are two hours enough to rule out LENR reactions,. which are more energetic than BLP reactions on a per-atom basis. Ed's conjecture assumes the existence of an electrically accelerated plasma. Moller's notes discuss the establishment of a plasma with the shell as anode and the tungsten wire cage as cathode. However, the Naudin experimental setups show no trace of a high voltage supply, only low voltage pulses to the filament from half-wave-rectified 50 Hz or a 12 V battery through a semiconductor switch. BLP has reported intense plasmas from cells using tungsten heaters to dissociate H with the presence K+ ions from dissociated catalysts; these experiments did not show the 2H+H reaction discussed by Phillips in later papers. I am not aware of any BLP paper testing the MAHG conditions. Mills' work has been distant in 'parameter space', but the atomic reactions have been shown. Tungsten at the cited temperatures is a rich electron emitter. Naudin's flow calorimetry is in the right direction, but the energy calculations are based on quite small temperature differentials across the cell. Naudin makes no statement or claim for the accuracy of the temperature differential. His tests for "efficiency" show a large spread for different runs. Thus none of his data are reliable enough, or runs long enough, to test either the LENR or BLP conjectures above. The most recent tests have been with a slower coolant flow and higher delta T, which is good. He has also plotted the delta T with time. The system has significant thermal mass and thermal delays, so when the temperature-time plots show many irregularities, as they do, they point to very large fluctuations going on inside the cell. The Langmuir 1912 paper is based on the observations of the conduction of heat away from heated tungsten wires in various gases. This general subject is of interest in designing incandescent lamps, whose life is extended by inhibiting evaporation of the tungsten wire by a gas fill, but that fill may also cool the wire by conduction, requiring more power to reach incandescence. Above a certain temperature, the conduction of hydrogen increases very rapidly, and Langmuir investigated. Langmuir attributes the increased energy loss to dissociation of the hydrogen at or near the filament surface, convection/conduction away from the surface, and recombination at some greater distance. He lays out a mathematical derivation of the rate of dissociation as a function of temperature, which is plotted in Moller's paper on Langmuir. According to the Langmuir equations, the dissociation rate in the vicinity of 1-2000K is very low, yet the heater in the BLP plasma cells operates in this range. It should be noted that the BLP thermally driven cell is intended only to show the catalysis plasma and was built using available laboratory 'quartzware' as much as possible. Langmuir's calculation does not take into account any catalytic properties of the hot tungsten surface itself, which may have a profound effect on the actual behavior. Naudin shows an increase in efficiency with decreasing drive pulse width. I haven't attempted to estimate how fast the filament will heat up, but it could reach a useful temperature very quickly. So you get a burst of dissociation of H2 molecules at or near the filament surface. The dissociation event is effectively instant, and the dissociated atoms move away. Langmuir poses the question of ionization of the dissociated H atoms, and comes down firmly on the side of no ionization, no plasma. [the BLP plasmas come from different reactions altogether]. Thus there is no merit in investing heating energy beyond a certain point. *IF* recombination yields *excess* energy, then it can occur at the tungsten surface between heating pulses. If this is true, then increasing the repetition rate of short pulses should show changes in the thermal performance of the cell. So far, Naudin has not made such tests to my knowledge. The heat pulses could also launch cascades of BLP H/hydrino catalysis reactions. Such are my thoughts to date. Mike Carrell