Historical research guide · 0399 · 2005
Comments on “Thermal behavior of polarized Pd/D electrodes prepared by co-deposition”
Kirk L. Shanahan
In one minute
Kirk Shanahan, a chemist at the Savannah River National Laboratory, wrote this short paper in Thermochimica Acta as a comment on the co-deposition experiment of Szpak, Mosier-Boss, Miles and Fleischmann, which reported excess heat from a palladium and deuterium electrode. Szpak’s team had set recombination aside: the gas volumes they collected matched the Faradaic prediction to better than one per cent, and Fleischmann and Pons had calculated that a recombination hot spot would carry about six nanowatts, too little for an infrared camera to see. Shanahan’s reply is that this answers a different reaction. He is not proposing electrochemical oxygen reduction; he is proposing ordinary burning of entrained deuterium and oxygen in bubbles at the electrode, below the liquid surface, and he works out that a single millimetre bubble of a stoichiometric mix releases about three and a half millijoules — milliwatts, not nanowatts. He then shows how a calorimeter with two regions of different heat-capture efficiency shifts its calibration constant when heat moves between them, with no new heat source involved.
Connection
Chapter 12 follows lattice confinement fusion and the calorimetry every claim in it rests on, and chapter 1 is the evidence ladder. This exchange is the ladder working: it names the exact measurements — the calorimeter’s true noise level, and a replicated heat-after-death run — that would decide between a chemical and a nuclear reading of the same data.
From the source
The references Szpak and colleagues cite against recombination deal with electrochemical oxygen reduction mediated by dissolved oxygen, a reaction Shanahan agrees is not significant here; what he proposes instead is the simple burning of hydrogen and oxygen to form water at the electrodes, below the electrolyte surface, in entrained bubbles — a process needing only fuel, oxidiser and an ignition source, with no electrochemistry involved.Section 2.1, SMMF’s objection to recombination
A bubble one millimetre across at 350 K holds 1.82 × 10⁻⁸ moles of gas; at an optimum two-to-one hydrogen–oxygen mix and the 285.8 kilojoule per mole heat of formation of water, one bubble releases about 0.00347 joules — milliwatts per bubble rather than nanowatts — so 100 to 300 bubbles give 0.35 to 1.05 joules, and the count of short-lived hot spots per infrared video frame in the Szpak group’s own photographs runs from zero to a few thousand, which Shanahan reads as consistent with that arithmetic.Section 2.1, bubble calculation and hot-spot count
Modelling the calorimeter as two regions rather than one — heat capture efficiency 99.9 percent in the first and 90 percent in the second, with the input power split evenly between them at calibration — shows that moving the split to 75 and 25 percent changes the calibration constant while adding no heat source at all; the apparent excess power then follows directly from the ratio of the calibration and shifted constants, and Shanahan’s earlier reanalysis of closed-cell data from Storms found that a calibration-constant variation of plus or minus 2.5 percent accounted for apparent excess up to about 0.8 watts.Section 2.2, Model of a heterogeneous calorimeter/closed cell
In the reported open-cell run all three Joule-heater pulses register as peaks in the excess-enthalpy curve, and since a Joule heater cannot produce the Fleischmann–Pons–Hawkins effect Shanahan reads those peaks as a calibration response; he adds that the measured heavy-water consumption of 7.7 cubic centimetres against a computed 7.2 is a 6.5 percent deviation, and that the excess rather than deficit of collected water points to entrainment of electrolyte droplets in the gas stream.Section 2.3, The open cell case
Taking the applied current together with the thermoneutral voltage puts the thermal energy available from recombination at about 1.54 times the current in amperes; at the 0.3 amperes sustained for an appreciable time that is about 0.5 watts, against the reported average apparent excess of about 0.27 watts — so on Shanahan’s own accounting the recombination extent needed would be at most about 50 percent, and probably much less.Section 2.3, thermoneutral-voltage estimate
What Shanahan says the field should do next: determine the calorimeter’s true noise level rather than its baseline fluctuation, since uncontrolled steady-state shifts add a non-random noise component; replicate the positive-feedback and heat-after-death observations, which so far are single examples; and characterise the cathode surface state with modern surface science, for which platinum electrodes are the more promising route because platinum forms no hydride and so removes bulk loading from the question.Section 2.4 and Section 3, Conclusions
The way in
https://doi.org/10.1016/j.tca.2004.11.007How to cite it
Kirk L. Shanahan (2005) Comments on “Thermal behavior of polarized Pd/D electrodes prepared by co-deposition” doi:10.1016/j.tca.2004.11.007