BACK to Contents

The claim for electrochemically charged palladium cells as prospective energy sources rests on reports by several groups of "excess heat" (or, more precisely, excess power) that cannot be accounted for in the thermal balance normally applied to water electrolysis. A number of other groups that have carried out calorimetric experiments In similar cells under a wide variety of conditions have found no excess heat.

The early chronology of reports on calorimetry in these electrochemical cells is outlined in Appendix 2.A. A summary of the results in published papers and manuscripts submitted or in press available to the Panel, is contained in Table 2.1. We also received a number of internal reports and private communications that were useful in assessing the question of excess heat. These varied widely with respect to number of trials, quality of data, number of control experiments, and types of cells and calorimeters (see Appendix 2.B). Many were considered by their contributors to be preliminary or interim reports, and in several, the results were ambiguous or inconclusive. These are listed in Table 2.2. Because it was difficult in most of these cases to examine the experimental procedures and results in detail, the quantitative results are not included in Table 2.2. The written reports from the national laboratories and others made available to the Panel are on file.

Among the issues the Panel addressed in examining published reports, assessing private communications, and in site visits were whether the power levels themselves are being accurately measured and whether the reactions being considered in these cells are, in fact, satisfying the chemical assumptions made. These heat measurements have been done with calorimetry that varied as to the technique and to the levels of precision and accuracy. In most cases calorimetric effects attributable to excess heat are small. The calorimetric measurements are difficult to make and may be subject to subtle errors arising from various experimental problems [SHE].

For the purposes of this report, the calorimetry is usefully differentiated as to whether the D2 and O2 gases are allowed to exit the cell completely unreacted (open cells) or are intentionally catalytically recombined to regenerate D2O and to recover the corresponding heat (closed cells).

In the case of open cells, where the gases are assumed to be vented without reaction, any output power (as heat) greater than the electrical input power minus the power equivalent of the D2O formation enthalpy [1.527 V (volts) x I (cell current)] is considered excess, a result reported by several groups. In closed cells with total recombination (and with a deuterium-charged Pd electrode), the total electrical power in and total heat power out would normally balance (as for Pt and Pd electrodes in light water). At present, most experimenters who have performed calorimetry with closed cells under strict recombination conditions have reported no excess heat. Another important point is that most of the reported excess heat measurements are actually power measurements, and the data in most experiments have not conclusively demonstrated that the total amount of energy produced (as heat and chemical energy) integrated over the whole period of tell operation exceeds the total electrical energy input.

Since the claimed excess heats have, in most cases, been of a magnitude significantly less than the 1.527 V x I factor itself, issues of calibration, reliability, and support of the assumptions of zero recombination in open cells are especially critical. The Panel's site visits have identified experimental


uncertainties, e.g., nonlinearities of the calibration in power output vs. temperature, time dependence of calibration, and doubtful accuracy of data acquisition relative to the magnitude of the effects asserted (Appendix 2.C). Even in laboratories that report excess heat, this effect, under apparently identical conditions, is not always reproducible. Moreover, in most cases where groups reporting excess heat have supplied complete cells or materials to other laboratories, excess heat effects have not been confirmed. In none of our visits to the different sites did we see an operating cell that was claimed to be producing excess heat at that time.

It is difficult to account for the widely divergent findings of excess heat. The failure to observe excess heat has been attributed by proponents to differences in materials (especially the Pd and D2O), size of electrodes, insufficient time of electrolysis and too small current densities, as well as to unknown effects of adventitious impurities or special, still unknown electrode surface conditions. However, examination of the results of those reporting positive effects (Table 2.1.A) shows that a wide variety of Pd from different sources has been used. There is a similar variation in the number of sources and batches of D2O. Current densities as low as 8 to 64 mA/cm2 have been reported to produce excess heat, and much higher current densities than these have been used in cells yielding negative results (Table 2.1.B). While very long electrolysis times have been required to produce excess heat effects in some laboratories, several have reported excess heat effects with electrodes of similar size that begin after a few hours of electrolysis. At this time we cannot find clear differences in materials, cells, or operating conditions that can account for the widely irreproducible behavior.

After assessing the reports from the different laboratories, considering the experimental difficulties and calibration problems, as well as a lack of consistency and reproducibility in observation of the excess heat phenomenon, we do not feel that the steady production of excess heat has been convincingly demonstrated.

However, there are reports of sporadic temperature "excursions" or "bursts" that apparently represent power outputs significantly larger than the input power. These events cannot be attributed to problems with accuracy or calibration alone and are presently not understood (see Appendix 2.D).


Table 2.1.A


(Published reports, manuscripts submitted or in press)

  Research Group Calorimeter, Cellb Pd type, sourcec Current Density, Voltage Excess Heat Controlsd Comments, References

1. Univ. of Utah Fleischmann, Pons et al. ISO A, open JM, rod, sheet, cube 8, 64, 512 mA/cm2, 3-10 V 5-111% (9 cells) none n,t [FLE]

2. Texas A&M Univ. Appleby, Srinivasan et. al. ISO-C (Tronac) Open Alfa, Wire 0.3, 0.6, 1.0 A/cm2, 3.4-5.6 V 6-30% H2O, Pt [APP]

3. Stanford Univ. Huggins, Gur et al. ISO A, Open Disks (arc melted) 10-1000 mA, 3-15 V 10-30% H2O [BEL]

4. Texas A&M Univ. Bockris et al. ISO A H&S, Rods (Ni anode) 100-500 mA/cm2 5-25% (3 of 10 cells) Pt [KAI]

5. U. Minnesota, Oriani et al. ISO-C, Open Jme, Rod 0.6-1.6 A/cm2 2-21% (2 of 3 cells) H2O [ORI]


Table 2.1.B


(Published reports, manuscripts submitted or in press)

NCAS EDITOR'S NOTE: This table was originally spread over two pages. In the interest of readability, the page break has been omitted.

  Research Group Calorimeter,
Pd type, sourcec Current Density, Voltage Excess Heat Controlsd Comments, References

1. U. British Columbia, Hayden et al. Flow, Closed Englehard, Bars to 2.2 A/cm2 none
(10 d)
Pt [HAY]

2. M.I.T., Wrighton et al. ISO B, Open JM/Aesar, Rods 69 mA/cm2, 2.9V none
(> 200 h)
H2O no t, n [ALBA]

3. Cal Tech, Lewis et al. ISO B, ISO C (Tronac), open Several Rods 72-140 mA/cm2 none
(5 cells)
H2O no t, n, He [LEW]

4. Naval Weapons Ctr. Miles et al. ISO, Open JM/Wesgo, Rods 100-200 mA/cm2, 3.5 V none
(1-10 d)
Pt, H2O no n, gammas

5. Sandia N.L. Roth et al. ISO A, Open Rod & Pd/Li alloy 320 mA/cm2 none
(17d, 36d)
  no n, t [ROT]

6. AT&T Bell Labs, Fleming, Law et al. ISO C (Setaram), closed&open Several wire, rod 16-512 mA/cm2, 2-10 V none
(1-40 d)
Pt, H2O [FLEM]

7. Argonne N.L. Redey et al. ISO B, ISO C, Open JM, Rod 15-500 mA/cm2 none
(460 h)
H2O [RED-1]

8. Free U. Berlin, Kreysa et al. ISO A, Open Rod, sheet 1.2 A, 9 V none
(circa 10 cells)
H2O no n, t, gammas, [KRE]

9. EG&G Idaho, Longhurst et al. ISO A, ISO C, Open Foil, wire 0.1ma-5.7A, 3.3-5.1 V none (> 20 cells) (120 h) H2O no n,t, gammas [LON-1]

10. Iowa St. U. Hill et al. open Rod 0.7-1 A/cm2 none   no n, gammas

11. U. Newcastle-upon-Tyne, Armstrong et al. flow, open Sheet, Cube 100 mA/cm2, 15-20 V none
(8 d, 2 cells)
Pt(H2O) [ARM]

12. Harwell Lab. Williams et al. ISO A, ISO C-type, Open JM, Rod, cast, Beads. Ribbons 20-530 mA/cm2, 3-15 V none
(11 cells, to 33 d)
H2O, Pt (14 cells) no n, t, gammas [WIL]

13. Chalk River Nuclear Labs, D.R. McCracken et al. Flow, closed JM, Rod 150-300 mA/cm2 none
(2 cells) (18d, 47 d)
  no n, t [MCC]

FOOTNOTES (Tables 2.1.A and 2.1.B):

Solutions were LiOD (usually 0.1 - 0.2 M) in D2O. Anodes were generally Pt.

Open cells: excess heat = measured heat + 1.53 I - power in.

Closed cells: excess heat = measured heat - power in.

% excess heat = (excess heat/power in) x 100

Sometimes % excess heat = excess heat/(power in - 1.53) 100 is used.

See Appendix 2B
JM, Johnson-Matthey; H&S, Hoover & Strong.

The Pd was obtained from manufacturer in various forms (drawn, cast. cold worked...) and subject to no further pretreatment, vacuum degassed, vacuum annealed, arc melted, preanodized, or recast, in both experiments finding excess heat and those yielding negative results.

H2O indicates substitution of H2O for D2O and usually LiOH for LiOD in cell . Pt indicates substitution Pt for Pd.
Electrolyte was 0.1 M LiOD reacted with excess D2SO4.

One Pd sample was a 0.1 cm rod obtained from Texas A&M University.

[[12]] [[13]]

Table 2.2

Internal and Unpublished Reports and
Private Communications on Calorimetric Results

(Experimental conditions generally the same as those in Table 2.1.)

A. Groups Observing Excess Heat or Bursts in Some Cells

  1. U. Landau - Case-Western Reserve University
  2. E. Yeager, B. Cahan, R. Adzic - Case-Western Reserve University
  3. D. Hutchinson - Oak Ridge National Laboratory
  4. C. Scott - Oak Ridge National Lab.
  5. A. Schoessow - University of Florida
  6. B. Liebert - University of Hawaii
  7. T. Droege - Batavia, Illinois
  8. G. Balding - Nytone Electronics
  9. H. T. Hall, Jr. - Novatek
  10. M. Wadsorth - University of Utah
  11. M. McKubre - Stanford Research Institute

B. Groups Reporting No Excess Heat

  1. C. Martin, B. Gannom, K. Marsh et al. - Texas A&M University
  2. S. Gottesfeld - Los Alamos National Laboratory
  3. S. Little et al. - Austin, TX
  4. 0. Bennion - Brigham Young University
  5. W. Ayers - Electron Transfer Technologies
  6. R.P. Allen, J.R. Morrey et al. - Battelle Northwest DOE Laboratory
  7. T. R. Jow, E. Plichta et al. - U.S. Army ETDL
  8. H. W. Randolph - Westinghouse Savannah River 'kite
  9. H. Weissman - Brookhaven National Laboratory
  10. P. Ross - Lawrence Berkeley Laboratory
  11. F. Wagner - General Motors Corp.
  12. J. C. Farmer - Lawrence Livermore National Laboratory
  13. E.L. Fuller, Jr. et al. - Oak Ridge National Laboratory


BACK to Contents