Run 5 of our 2nd attempt to replicate the Mizuno-Ohmori Incandescent W Excess Heat Effect - 10JUN99
For Run 5 we removed the mechanical stirrer used on Runs 3 and 4 and replaced it with a magnetic stirrer. A standard teflon-coated magnetic stirring bar was placed in the bottom of the cell.
This photo shows the arrangement of the stirring drive motor, a small DC motor with a large ceramic magnet mounted on its shaft. The calorimeter enclosure was modified to provide space for the motor but a 12.6 mm thick polystyrene foam insulation barrier was maintained between the motor and the experiment.
This small decrease in the overall insulation quality of the calorimeter enclosure did not noticeably affect the heat recovery.
We used a new cathode (of our own manufacture) and fresh 0.1M K2CO3 electrolyte for Run 5. Because the mechanical stirrer was gone we were again able to fully collect the gas from Run 5 and we employed a simple ice-cooled condenser to collect most of the water vapor from the gas.
Results:
The color legend and vertical scales for this plot are as follows:
Pin (0-200 watts)
Pout (0-200 watts)
Tcell (0-100° C)
Vcell (0-200V)
Icell (0-5A)
Tin (39-41° C)
The horizontal scale covers 3 hours.
The run begins with the usual equilibration period. At 0.66 hours we began to apply voltage to the cell, stopping briefly at 30 volts to allow the cell temperature (Tcell) to exceed 70° C. At 140 volts, the cathode suddenly burst into bright incandescence so we just left the voltage (Vcell) at about 143 volts for most of the run. Note that the cell temperature (Tcell) was 90-91° C during this time. As you can see from the close approach of the Pout and Pin traces, no excess heat signal was observed at this voltage. At about 1.5 hours we turned the voltage down into the "hysteresis loop" region to about 100 volts...just above the voltage where the gas sheath usually collapses. We left the voltage at 100 volts long enough (about 15-20 minutes) to see that again Pout was very nearly equal to Pin...no excess heat. We then proceeded to lower the voltage slowly until the gas sheath collapsed (marked by a sudden increase in Pin) and then continued on down to a low voltage that resulted in a Pin of 10 watts. By periodic adjustment of the voltage as the cell cooled we held Pin at 10 watts until Pout was also reading about 10 watts and then turned off the cell power altogether.
Gas Measurements:
This photo shows the ice-cooled condenser (left) and the gas-flowrate apparatus employed during this experiment.
In ordinary electrolysis, one expects about 0.19 cc/sec of gas to be produced for each ampere of current through the cell (assuming the gas is collected at 25° C). By assuming that the voltage delivered to the cell was constant, we were able to calculate the average cell current from the displayed values of cell power and cell voltage on the Clarke-Hess power analyzer. (NOTE: The Clarke-Hess displays the RMS current directly but that quantity is not what is desired for the gas flow calculations).
During the warm-up period at the start of the run when the cell voltage was held at about 30, the observed gas flow was 1.15 times higher than that predicted from the average current.
During the 143 volt plateau, when the cathode was incandescent (bright orange with dancing white flashes around the edges), the observed gas flow was 2.5 - 2.8 times higher than predicted from the average current.
During the 100 volt plateau, when the cathode was not visibly hot but was still covered with dancing purplish sparks, the observed gas flow was 2.5 times higher than expected from the average current.
During the 10 watt plateau at the end of the run, the observed gas flow was 1.05 times higher than expected from the average current.
Condenser Measurements:
The water collected in the condenser was carefully weighed at 0.515 grams. The evaporation of this much water over the hour-long hot portion of the run would remove only about 0.3 watts from the cell.
After the run, the collected water was evaporated to dryness and only 0.002 grams of a whitish residue was observed.
This close-up photo was taken during the 143 volt plateau with the camera held right up to the small viewport in the calorimeter enclosure. The left half of the cathode can be seen through a gap in the coils of Cu tubing that surround the cell. The orange color in this photo is quite true. A hint of the dancing white flashes that populate the edges of the cathode can be seen on the left edge in this photo. The dark vertical object that obscures the right side of the cathode is a vertical section of the Cu tubing (see the top photo in this report for clarification).
Discussion:
The magnetic stirrer did not enable the excess heat effect. Next we will try one of Mizuno's cathodes.
The gas flow measurements were a little surprising. In this lab we have observed on many occasions that the gas flow from an electrolysis cell operating at normal voltage and current is around 1.05 times higher than expected from the current. We always thought the extra gas was water vapor coming from the cell. However, we observed the same level of excess in the latter part of this run with an ice-cooled condenser before the gas flow apparatus. Furthermore, an even greater excess (i.e. 1.15) was observed at the first of the run when the voltage was at 30 volts, still far below the threshold for sparking at the cathode.
As we showed in our first series of experiments on this cell, the very large excesses during cathode incandescence and sparking are probably due to hydrogen gas that is orphaned by oxidation of the W cathode.
Comment, questions, and suggestions are welcome.