Relatively little emphasis has been placed on the role of material properties in cold fusion experiments. Characterization of the anode and cathode compositions, microstructures, homogeneity, etc. generally has not been carried out and has not played a significant role in the discussions directed towards understanding cold fusion. These materials parameters have, however, been involved in discussions directed towards explaining difficulties associated with reproducibility in these experiments. Most experimental protocols have utilized Pd cathodes for the electrolytic charging, and Ti (or Ti alloys) for gas charging experiments, following the examples of the initial experiments by Fleischmann and Pons [FLE], and Jones [JON-1]. In addition, most electrolytic charging experiments have utilized Pt anodes, with the important exception being those at Texas A&M for which tritium (T) generation has been reported and which utilize Ni anodes [PAC].
It has become increasingly common for the D/Pd ratio attained to be reported in the electrolysis experiments. Measurements of the D concentration are often based on the weight gain of the cathode during electrolysis. Values of D/Pd in the range 0.7 to 1.1 have been reported for experiments which report both positive and negative cold fusion results. The gravimetric method of measurement is often employed and is generally reliable for D/Pd values in the range of 0.6 to 1, as long as the loss of D is minimized between the period of charging and the weighing. Most investigators do not report the details of their procedures. Precautions taken to avoid loss of D, such as the storage of specimens at low temperature, are not discussed. Consequently, in the absence of other difficulties, the reported D/Pd values are lower limits to the values in the operating cells. An additional difficulty in the electrolytic charging experiments is that deposits on the cathode or "diffusion" of Li into the cathode must be accounted for properly before the gravimetric results can be considered accurate. In the majority of the reported results, it has not been shown that suitable procedures have been carried out.
In a few cases the cathodes have been analyzed for T, He3, and He4 after electrolysis, with negative results [WIL, HOL, THO, LEW]. In most experiments these analyses have not been carried out despite the importance of these fusion products in establishing the presence or absence of cold fusion. Reports of diffusion of Li into the cathodes have been made [PAC] but the basis for these reports has generally not been specified. In one case [WIL], the presence of Li up to several micrometers into the Pd cathode was detected by Secondary Ion Mass Spectrometry (SIMS) but the authors suggest that this was the result of surface cracking and contamination of the crack surfaces by the electrolyte. They also detected other components of the electrolyte on the cathode surface.
Even less materials characterization has been carried out for the gas charging experiments. Often mixtures of metals have been used and they have been characterized as "turnings", "sintered powder", "mossy solids" etc.; which have very limited meaning with respect to the material's character. The D/Pd ratios have generally not been determined nor has sufficient information been given to allow an estimate to be made. In some cases, the pressure of the D2 gas above the material has been measured and shown to decrease during period when the material was to have absorbed D. In other experiments, D absorption has not been demonstrated and there is no evidence that the experiments have been carried out on metals which contain D. It is important to determine the D/metal ratio in these experiments as absorption of D in Ti and its alloys is a very surface sensitive process. Exposure to D2 gas does not assure that significant absorption of D occurs.
As discussions of the success or failure of cold fusion experiments have often been ascribed to properties of the cathode materials, it is useful to consider the various materials parameters suggested to be significant. In the following, these will be considered in light of what is known of the behavior of D(H) in Pd and Ti.
The suggestion has been made that as a result of the high fugacity, confinement of D in Pd during electrolysis results in a very small D-D distance. Both the alpha and beta phases of the Pd-H(D) system are solid solutions. As discussed in Appendix IV.A, of all the configurations considered, the smallest distance between interstitial sites occupied by D in Pd is 0.17 nm; which is large compared to the D-D distance in the D2 molecule, 0.074 nm. Significant dual occupancy of interstitial sites by D is not in accord with the linearity of the lattice expansion measured by X-rays [PEI) or with theoretical treatments of the Pd-D system [NOR, RIC]. Furthermore, multiple occupancy of vacancy sites [BES] leads to an increased D-D spacing. Thus even under the high D/Pd values attained by electrolytic charging, the D-D spacing remains much too large for any fusion reaction.
The possibility of a successful electrolytic cold fusion experiment has been ascribed to the very high deuterium "confinement pressures" achieved at the cathode at the electrolytic overpotentials applied. Fleischmann and Pons [FLE) discuss equivalent D2 pressures of about 8x1026 atm. corresponding to their overpotential of 0.8 V. As discussed in Appendix IV.A, the overpotentials do result in a high deuterium fugacity, i.e. a high chemical potential of hydrogen relative to the standard state at one atmosphere, at the Pd surface. However, the non-ideality of D2 gas and the loss of deuterium by D2 bubble formation at the surface results in a much decreased equivalent D2 pressure. The relation between fugacity and
equivalent pressure is not easily obtained as it depends sensitively on the surface structure and impurity concentration. However, at steady state, i.e. after long charging times, the D in the specimen is at the same chemical potential as that at the surface. Hence the H(D)/Pd values obtained by cathodic charging may be compared to those obtained by gas charging [BARO]. Upon making this comparison (Appendix IV.A), it is seen that the compositions attained by cathodic charging correspond to a very moderate D2 pressure of 15x1O3 atmospheres. While high fugacities are produced by the overpotentials used in cathodic charging, confinement pressures of the order discussed by Fleischmann are not present.
The suggestion has been made that successful experiments require very long charging times. Assuming that this is related to materials issues, diffusion is the important materials parameter to consider. As discussed in Appendix IV.A, the diffusivity of D in Pd is sufficiently large at 300°K so that equilibrium can be attained in a 1 cm thick sheet of Pd in times of the order of 7 days. In the 1 mm thick cathodes often used, the time to attain equilibrium at 300°K is of the order of 2 hrs. Longer time may be required as a consequence of surface impedances to diffusion which result from impurities plated out from the electrolyte (Pd) or from surface oxides (Ti). These times are very much shorter than the very long times which, it is often claimed, are required for successful cold fusion experiments.
The suggestion [PON] that a particular form of the cathode material, e.g. cast material, must be used for successful cold fusion experiments is not in accord with the known behavior of Pd. Annealing the material to minimize the dislocation concentration should not be necessary. The very large inhomogeneous deformations which accompany cathodic charging of Pd, under the conditions used, result in high dislocation densities even in a well annealed material [HO]. Furthermore, the suggestion that vacuum annealing is necessary to remove the interstitial impurities [HUG) does not seem to be in accord with the conditions under which cold fusion is claimed. At the high D/Pd values, the presence of a few interstitial or dislocation trapping sites is insignificant compared to the high D concentrations.
Deposition of various materials from the anode and from the electrolyte (leached from the container) onto the cathode has clearly occurred in many of the experiments. Analysis [WIL] has revealed Pt (probably from the anode), Si (probably leached from the glass container), Cu, Zn, Fe, Pb, and other trace elements. Surface contamination of the cathode may affect the rates of reactions and therefore the effective fugacity. In general, surface poisons increase the fugacity by decreasing the rate of D2 gas formation. This is indicated by the high D/Pd values attained and by the observation, reported in several experiments, that the loss of D from the cathode after electrolysis was slow. Thus surface contamination probably had the effect of increasing the D/Pd values and the fugacities attained.
"Dendritic" growths were reported on the surfaces of the cathodes in the Texas A&M experiments which utilized Ni anodes [PAC). This has led to the suggestion that these dendrites are of significance for the formation of tritium during electrolysis, possibly "due to increased electric fields at the dendrite tips" [BOC, PAC]. Other experiments which reported these "dendrites" have not reported generation of tritium [WIL]. It is difficult to understand the importance of these "dendrites" in cold fusion. (These dendritic growths are not uncommon in plating experiments.) The field at the "dendrite tip" is the voltage across the double layer divided by the thickness of the layer. Since the total voltage across the double layer at the cathode is of the order of several volts, high fields can be attained if the double layer thickness is small. However, acceleration of D+ ions across this double layer will only result in energies of several eV (the energy determined by the voltage drop at the cathode), hardly enough to cause fusion.
The conditions of the gas charging experiments are likely to result in hydride formation in the Ti-D system and formation of the beta phase in the Pd-D system. The process of gas charging is well understood, as is the process of discharging of D into the gas phase from hydrides. Anomalous behavior which may lead to fusion and the formation of fusion products is not expected in these processes. Fracture of the brittle hydride phases is likely and has been well studied [SHI-1] . These fracture processes are of low energy and occur at a fraction of the speed of sound in the hydride. While acceleration of charged particles has been suggested to occur during fracture, this phenomenon remains speculative. Palladium deuteride and titanium deuteride are metallic conductors and hence large fields are unlikely to form during fracture. An alternative view has been expressed [MAY].
The behavior of H(D) in Pd and Ti is sufficiently well understood both experimentally and theoretically to answer many of the speculations about the properties of H(D) interstitials under the conditions for which cold fusion has been reported. None of the known behavior is consistent with the possibility of overcoming the coulombic repulsion between the D ion cores. The confinement pressures and energies of D in Pd and Ti are far from the regimes which are required for a reasonable possibility of cold fusion based on our present understanding. No aspect of the known behavior of D in Pd or Ti is compatible with formation of energetic neutrons, tritium or helium.