======== Copyright protection begins with (and includes) this line ======== A TWIST OF RIBBON Terry B. Bollinger 2416 Branch Oaks Lane Flower Mound, Texas 75028 Home: (214) 539-3897 Originally: April 19, 1991 [Reissued May 2, 1992, with updated address, copyright, and acknow- ledgements. References from August 1991 have also been appended.] Copyright 1992 by Terry B. Bollinger. Please feel free to make any number of copies in any type of printed or electronic media, as long as such copies are complete and are provided free of charge to their recipients. Short quotes are also fine, provided that they are clearly identified as being from this document. PREFACE This is not a paper in the conventional sense of the word, nor is it intended to be. If appropriate, a full paper in conventional printed media will follow at some future date. This particular document might better be described as a Farfetch that got out of hand. (Farfetching is a "game" I once proposed for trying to solve difficult problems.) There are some key points I have been unable to eliminate during the self-critiquing phase of Farfetching, and this irritates me greatly. I thus think it is time to let other people have a look at these rather unusual ideas. Besides, they lead to some interesting (and testable) predictions. OVERVIEW This document proposes a method whereby certain classes of low-energy solid state systems may violate baryon conservation. The method is to induce heavy baryon-containing particles to jump from energy states that closely approximate stationary (standing wave) momentum Dirac delta functions into nearby energy states that closely approximate stationary spatial Dirac delta functions. Although similar in principle to the type of state transitions seen in small-scale atomic phenomena, these spatially large state transitions are interesting because they appear to be subject to the constraints of large-scale, unidirectional time flow. This is in sharp contrast to small-scale state transitions of atoms, where time is much more space-like in character and thus can support an essentially instantaneous flow of state change information throughout a wavefunction. If such transitions are embedded in conventional unidirectional time, they should require non-zero lengths of time to complete. It is proposed that during such transition periods there will exist briefly conditions that are roughly equivalent to stretching or distending the particles over macro- scopic distances. Such distances are far larger than those over which the strong force normally can act. This implies that such transient particle states will decompose into simpler particle sets that need only preserve long-range conservation values of mass, charge, and spin. Since such decompositions would occur over physically large distances, most of the mass of the particle would be translated into low-energy photon pairs (heat) along the physical length of the decomposition event. Neutrinos, electrons, or positrons would form at the site of the spatially localized goal state to resolve any residual spin and charge quantum numbers. It should be noted that such proposed decompositions would apply only to heavy, baryon-containing particles, since light particles (electrons) would be protected by the fact that there are no simpler particle sets into which they could decay and still conserve mass, charge, and spin. There appear to be at least two classes of solid state systems that could provide a test of this proposed mechanism. The first is high-density injection of monochromatic thermal neutrons down the major axis of highly anisotropic crystals such as graphite. The second is the introduction of fermion chemical species of hydrogen into transition metal hydrides that have been physically or electrochemically structured so as to provide pseudo-1D diffusion environments. Because high hydrogen mobility is the best indictor of potential band behavior, the best elemental medium for this latter case would be the metal palladium. Specifically, a single crystal of pure palladium that has been electrochemically structured along one of its three cubic crystal axis should provide a good approximation of the conditions needed to produce the hydrogen isotope baryon violation test case. Slow hydrogen charging would be needed in such a system to avoid alpha-beta phase transition disruption of the long-range crystal structure, since that structure is critical to the formation of pseudo-1D heavy particle bands. The best candidate fermion hydrogen species for baryon violation within such structures would be neutral deuterium atoms and protium cations (protons). Neutral deuterium should be somewhat more likely to form the necessary band structures because of its minimal, alloy-like chemical interaction with the surrounding metallic crystal lattice. For both neutron-in-graphite and hydrogen-in-metal system classes, the creation of adjacent Dirac delta-like states could be accomplished by inducing solitons or soliton-like regions into pseudo-1D heavy-particle band structures. Soliton distortions of continuum band states would create physically localized intergap states at the sites of the solitons, with the energies of those states being somewhat above those of the highest filled band states. A jump of a heavy particle from the highest filled band state to this intergap state then would meet the prerequisites for inducing a rapid transition from an extended momentum-like states into a localized space-like state. Experimental evidence developed for other purposes at Oak Ridge National Laboratories may provide some support for baryon violation in solid state systems. In particular, Oak Ridge has found that some palladium deuteride electrochemical systems do in fact appear to produce excess heat with no obvious by-products. Such instances could be examples of baryon violation for neutral atomic deuterium in palladium, which should produce only heat and neutrinos. A decay rate of roughly 10^10 deuterium atoms per second per Joule/second of net heat output would be required to explain such results. A comparable neutrino output of at least 10^10 neutrinos/second per Joule/second of heat output would then be expected to provide resolution of the residual 1/2 spins of neutral deuterium atoms. -------------- GO DELTA TO DELTA, THE DIRAC WAY For the past few months I have been looking at a specialized class of quantum state transitions, an investigation that led to this document. Specifically, what I have been looking at is a class of direct transitions from quantum states that approximate momentum-domain Dirac delta functions into quantum states that approximate spatial-domain Dirac delta functions. Although unusual, such Dirac to Dirac transitions are discussed indirectly in a large body of solid state physics literature. They can occur when a soliton -- a phase "twist" -- is induced into the continuum band states of a pseudo-1D metal such as polyacetylene. The creation of a soliton region results in the formation of a new, highly localized intergap state (the spatial Dirac state) that is somewhat higher in energy than the highest filled band state (the momentum Dirac state). Although close in energy, these two states have very different wavefunction properties. Such drastic state transitions interesting primarily because they appear to invoke some interesting temporal issues. Momentum Dirac states can be very large in their physical extent, as shown by the shared macroscopic boson states of liquid helium II and the delocalization of the (fermi) continuum state electrons found in metals. In the case of metals, such macroscopic delocalization of such states have energy implications that lead to the formation of band gaps. Specifically, the energy of the highest filled band states -- say those where lambda approaches twice the lattice unit -- are significantly lowered by the diffraction effects of the lattice, and so are no longer roughly equivalent in energy to their free-space counter- parts. They are "tied down" to the large-scale crystal lattice structure by an energy debt. Such states provide the familiar band-surface states that are readily observable in all sorts of electronic and physical phenomena. For example, the holes of semiconductor bands represent electrons that have been removed from such states. Since a "found" electron is itself a pretty good approximation of a spatial Dirac function state, an obvious question is why there should be anything novel about the soliton version of inducing an electron to enter the spatially localized intergap state. Would it not be just another example of finding a top-of-the band electron, much as in the case of semiconductor hole formation? The difference is that hole formation corresponds to finding a top-of-the- band electron at an unknown location in the lattice. In contrast, soliton formation attempts to "pull in" the highly delocalized, top-of-the-band electron state into a known, well-defined location. If soliton formation occurs too quickly, this "pulling in" of the delocalized wave function will occur too rapidly to allow proper integration of psi^2 probability over time. Viewed another way, the energy used to raise the electron to the intergap state must first be redistributed throughout the lattice to compensate for removal of the particle from an energetically preferable band state. A simpler way to describe to situation is to note that since extended band states are energetically tied to large-scale crystal lattices that exists in large-scale, unidirectional time, the band states themselves will also be embedded in large-scale, unidirectional time. Thus even though they are nominally quantum entities, such band states will be subject to many of the same time constraints that apply to large-scale objects. This is in sharp contrast to small-scale, highly localized state transitions of the type found in atoms, where time is highly space-like in nature and thus permits essentially instantaneous reconfigurations of wave states. The embedding of a state transition in large-scale time implies that the intermediate states that occur during state transitions should not be ignored; they will behave as real transient states with real physical consequences. EFFECTS ON FERMIONS In the case of an electron, these consequences are likely to be minimal, at least for the particles themselves. No matter how much stress a large- scale quantum transition event might place on an electron, there are no simpler particles into which it can decay and still preserve its long-range quantum numbers of mass, spin, and charge. For a more complex fermion such as a neutron, proton, or even a composite fermion such as a neutral deuterium atom, the issue is more complicated. Specifically, the strong forces that hold together a neutron or proton are not subject to any long-range conservation principles such as those that apply to charge, mass, and spin. (Because they fall off gradually with distance, the effects of charge, mass, and spin of any given particle are known throughout the universe, and must therefore be conserved without violation throughout the universe. In contrast, the strong force falls off very rapidly with distance, and thus is only locally conserved; it can be violated without creating an inherent contradiction in the structure of space.) If the temporal issues I have proposed for macroscopic state transitions are real, one consequence is that they may be able to differentiate between fundamental long-range conservation laws for charge, mass, and spin, and the local conservation laws that apply to the strong force and the weak interaction. If the short-range strong forces are "stretched" over too great a distance during a large-scale state transition event, the result may be a conversion to other particle types that meet long-range conservation laws while providing a more direct resolution to the distributed energy debts of the original delocalized band state. WHY IT SHOULDN'T HAPPEN One obvious objection to all of this speculation is that it cannot happen in the first place, since even the lightest stable baryons are many hundreds of times heavier than electrons, and thus cannot possibly form bands in the same fashion as electrons. As a physicist once noted when I described the heavy particle band idea to him, any "atom" that had, say, protons orbiting in the manner of electrons would be thousands of times smaller in radius than a metal atom. Such a construction would be quite impossible, since strong force interactions would take over at that range and preclude any electron-like behavior by the protons. This would seem to mean there is no way a large-state transition experiment could even be set up for baryons, let alone performed. WHY IT COULD HAPPEN However, it turns out that the electrons which form bands in metals are pi electrons, which do not compare directly to the highly localized sigma electrons found in situations such as covalent (single) bonding. If proton band behavior depended on protons behaving in a fashion similar to that of sigma electrons, then such bands would indeed be impossible to construct. But the pi electrons are free to wander over an entire crystal lattice, as in metallic bonds and the second bonds of double-bonded organic compounds. This is a much more plausible mode of behavior for heavy particles such as protons. Thus the first requirement for building a heavy-fermion quantum transition experiment is not that the heavy fermions orbit as though they were parts of atoms, but that they be relatively free to move through a some type of periodic lattice. Furthermore, the unit cell of that lattice should have a length comparable to the lambda (momentum) wavelength of the fermion. This turns out to be a fairly plausible requirement, one that appears to be possible in more than one type of physically realizable system. NEUTRON BANDS One of such system would be an intense beam of neutrons directed down the main axis of a large, high-quality graphite crystal. The asymmetry of the graphite crystal layers would encourage some degree of preferential diffraction back in the direction of the beam, giving at least a rough pseudo-1D structure. If the intensity of the beam was high enough and was centered strongly around a lambda value comparable to twice the distance between graphite layers, competition for momentum states should lead to some degree of band behavior -- assuming the crystal would not simply be vaporized by such a nasty level of bombardment! Although interesting, the difficulty of maintaining high levels of neutron bombardment without disrupting ordinary matter clearly make this a difficult experiment. HYDROGEN BANDS The second class of physical systems that appear to meet the minimal needs for a heavy-fermion state transition experiment are much messier and far more complex, but they are also much easier with which to work. It turns out that the isotopes of hydrogen -- protium, deuterium, and tritium -- are all remarkably mobile in various transition metals. These metals thus provide a possible basis for heavy fermion banding effects. There are many transition metals which demonstrate significant levels of such hydrogen mobility, including niobium, tantalum, and titanium. However, since mobility is the key starting point for band behavior, the best metal would be the one with the highest hydrogen mobility. That metal is palladium. THE QUANTUM DOT MODEL Some idea of whether the heavy-fermion band idea is plausible can be deduced by noting that the quantum wavefunctions of fermion hydrogen species in metallic lattices are quite similar mathematically to scaled- down versions of quantum dot semiconductor devices. Quantum dot devices are an experimental class of semiconductor devices that work by jumping electrons across distances of hundreds or thousands of Angstroms. As it turns out, there is a simple inverse relationship between the mass of an electron and the ideal distance over which they should be made to jump to produce a working quantum dot device. For example, if you could somehow make electron a couple of thousand times heavier (about mass of a proton), you would need to place the dots about an Angstrom apart. Such a spacing would place the devices right in the same ball park as the spacing between interstitial hydrogen sites in a metallic hydride crystal lattice. This simple relationship strongly suggests that the overall behaviors of metallic hydrides and quantum dot semiconductor devices should be rather similar. Quantum dot devices are dominated by Van Hove singularities, which are an extreme form of the energy/momentum band skewing I mentioned earlier. The density of states -- that is, the number of continuum states contributing electron charge to each dot in the quantum structure -- nominally goes to infinity in such a structure, although of course in any real system it will simply grow to be the (very large) number of electrons contributing across all the dots. Delocalization effects thus become quite important for understanding how such devices work. ATOMS OR PERTURBATIONS? If my quantum-dot to metal-hydride analogy is valid, then the "atoms" of hydrogen in metallic hydrides tend to behave oddly for a very good reason: They are not really singular atoms at all. They are instead composite nodes of a very large number of highly delocalized atoms, just as the second-level (pi) bonds in polyacetylene are charge density waves nodes composed of small perturbations of a very large number of continuum electron states. If you poke at these composite hydrogen band nodes hard enough, they will be forced to "decide" to become distinct, singular atoms of hydrogen. But until such poking occurs, such nodes are more accurately described as local perturbations of a continuum band of delocalized atoms. This composite atoms idea has some precedent, by the way. It has been known for a long time that a small volume of liquid helium II is composed not of individual atoms, but a continuum of atomic states delocalized over the entire liquid. The main difference in the case that I have just proposed for metallic hydrides is that the composite behavior of the continuum states would exhibit local perturbations that closely resemble localized atoms in mass and charge. SETTING UP THE EXPERIMENT If palladium and other metals do in fact meet the first criterion of band formation for the rapid quantum transition experiment, how might one go about setting up such an experiment? For reasons that have to do mostly with the ease of soliton formation, a pseudo-1D environment is preferable. This means that the palladium medium would need some type of anisotropic orientation along one of its three (cubic) symmetry axis. The easiest way to do this would be to apply a voltage differential across two opposite faces, since this would cause the hydrogen to move preferentially along that axis. The desired anisotropic effect might also be significantly enhanced by using integer ratio mixes of hydrogen and deuterium, such as a 1:1 protium/deuterium ratio. These isotopes travel at different rates as they move through palladium, and thus would tend to compete over time for distinct paths that would enhance pseudo-1D behavior for both isotopes. Loading of the crystal with hydrogen should be as slow as possible to minimize the disruptive effects of alpha/beta phase transition on the metallic crystal lattice. In the long run, a structurally imposed pseudo-1D environment would be preferable, since it would provide greater stability. Metallic compounds that create isolated 1D chains of palladium and other atoms are already known to exist, so such constructions appear to be quite plausible. SORTING THE SPECIES The hydrogen species most likely to participate in band formation are H+ (in which I use H to mean protium), neutral D (deuterium) atoms, and T+ (tritium cations). T+ is a rather poor candidate both because of its instability and high mass. The other possible species are the anion hydrogen species (H-, D-, and T-), and the boson species (H, D+ [, D-], and T). Even though two of the anion species (H- and T-) are fermions, the anion species probably can be eliminated because of their similarity to the neutral He configuration, which is extremely immobile in palladium. The boson species can delocalize, but cannot compete for states to create structured bands. This in turn means they cannot create intergap states when inverted at a soliton region, since the existence of the intergap state depends on inversion of the (filled) valence and (empty) conduction states. The structured band thus acts like a ribbon of finite width, and the soliton behaves like a twist that flips the ribbon over. I suspect that both the neutral and positively charged (cation) states are sufficiently mobile in palladium and other metals to lead to band behavior. Since band formation leads to a net lowering of energy, both the H+ and D species would tend to transition into band structures if allowed to move freely. (The idea of bands of neutral particles may sound a bit odd, but it really is not. Band formation is consequence of fermi statistics and the energy lowering due to lattice periodicity. Neither of these require that the particles be charged.) However, of the H+ and D possibilities, D should be a better candidate because its lack of charge would keep it from interacting as strongly with the surrounding metallic crystal lattice. A further complication is that there could be two metallic hydride species of neutral D. One would be an ortho version where the electron spins in the same direction as the nucleus, and the other a para version with opposing electron/nucleus spins. Such variations would give rise to distinct bands for each spin state of each species, but would not change the overall model for soliton formation. (As I mentioned, the hydrogen in metal scenario is quite a bit messier than the neutrons in graphite scenario.) It may sound odd to recommend a composite fermion such as D for an attempt to "stress" the strong forces, which are many, many orders of magnitude more powerful than the weak electromagnetic force that binds a single electron to a deuterium nucleus. But as best I can tell, the formation of a set of continuum states would have the effect of "hiding" the internal structure of the neutral D atom until it can reappear in a normal (vs. frequency) space, at which point it would be able to eject its electron with only a little prodding. But by that time it may be too late, since anything that will happen to the D atom will occur during the state transition itself. CREATING THE SOLITON REGION Actual creation of the soliton state could occur either naturally, as it does in polyacetylene, or as a result of a very slight (one lattice unit) separation or compression of the crystal structure, say at roughly the center. Polyacetylene soliton formation can occur via relaxation around an electron injected (or removed) by a donor (or acceptor) impurity, and roughly similar classes of injection mechanisms may also be possible for metallic hydrides. PROPOSED RESULTS As mentioned earlier, I would suggest that the consequences of building successful state transition mechanisms for heavy particle would for the most part be controlled by the long-range quantum conservation laws, with the energy of such an event distributed over the original physical extent of the delocalized particle state. One visual analogy would be a long piece of elastic ribbon that has been lightly tacked down along a very long bar. The attempt to suddenly create the spatial Dirac state would then be like attempting to pull in the ribbon suddenly by striking it very quickly in the middle. If the speed of the strike goes beyond some limit, its internal structure will begin to fray -- and once that occurs, the fraying will propagate outwards along the physical length wavefunction. In the case of neutral deuterium, the only fundamental quantities that would need to be conserved in such a fraying process would be spin and mass. The simplest particle for carrying off the spin would be an electron neutrino, while the simplest particles for carrying off the mass would be spin-cancelling pairs of photons. If one assumes that photon pairs would be generated along the length of the original delocalized state, the result would be a remarkably low average energy that would closely resemble ordinary heat. For the proton (H+) case, a positron would be needed to carry off the positive charge. This in turn would result in a classic positron-electron gamma ray emission. I should also note that these highly symmetrical cases are not the only ones possible. Depending primarily on the mass and charge distributions of the both original band structure states and the goal intergap states, many different types of particles could conceivably be produced. For example, since the end points of pseudo-1D band structures (at the crystal surface) are themselves similar to solitons in topology, there could be modes in which the goal states have large mass nodes at either end of the electro- chemically anisotropic crystal. A slight asymmetry in the mass distribu- tion between two such nodes could then lead to the production of neutrons at one of the band structure. Such neutrons would then combine with the ambient elements to form new isotopes, such as tritium via combination with ambient deuterium at the crystal surface. Continuous energy production would presumably occur by a sort of "candle wick" effect in which a stable soliton would remove fermions from the highest (valence) band state and try to move them into the intergap state, resulting in decay. New fermions would then "move up the queue" to also be disrupted by attempting the transition. To produce one Joule/second (watt) of energy in this way would require that roughly 10^10 deuterons decay per second (assuming that the energy carried off by neutrinos is negligible). Other than heat, the only major by-product would be the production of neutrinos at a rate of at least one per deuterium atom decay. ACKNOWLEDGEMENTS My thanks to Dr. Alan Salisbury for his outstanding technical and personal leadership of the Contel Technology Center, at which I worked on unrelated software issues during the early days of the "cold fusion" story. I would also like to thank my good email friend Dieter Britz and many other participants in the Internet sci.physics.fusion interest group for many interesting email conversations about chemistry and physics. -------------------------------------------------------------- Appendix A -- Palladium Hydride References The intent of my research when I collected the following list of palladium hydride references was to determine whether there is any type of plausible solid-state system that could (at least in principle) support the existence of heavy-particle fermi band solitons. This objective led quickly to the extensive multi-decade literature on palladium hydride and deuteride systems, but only rarely to the more recent "cold fusion" literature, which by comparison generally exhibited only a rather superficial knowledge of palldium hydride physics and chemistry. One exception to this was the Oak Ridge Scott et al paper, which was carefully done and appears to provide experimental evidence for the existence of highly unusual solid-state effects in some palladium hydride systems. Whether or not such effects are any connection to heavy particle band solitons is clearly very much open to debate, but it does at least provide a data point indicating sufficient unknown behavior in palladium hydride systems to permit the existence of some type of novel quantum effect. My net conclusion from this exploration of PdHx literature is that while past work does not provide any clear proof of the existence of heavy particle band solitons in palladium hydride systems, it does show that PdHx systems are sufficiently complex and poorly understood even today that such an effect could probably exist without necessarily having been recognized. Perhaps more importantly, data such as the detection of atomic hydrogen and deuterium tunnelling via the SPOTS NMR technique seem to provide strong hints that the critical prerequisite of band formation for atomic hydrogen isotopes is fairly plausible. In combination with predictive modeling, similar experimental methods might help prove or disprove the existence of heavy particle band and soliton phenomena in palladium hydride systems. The references themselves are in no particular order. Some of the most interesting ones come later in the list, for example. Many of them are cryogenic studies that look at ordering effects at low temperatures, rather than at room temperature. 1. O. Blaschko, R. Klemencic, and P. Weinzierl, "Lattice dynamics of beta-PdD0.78 at 85 K and ordering effects at 75 K," Physical Review B, Vol 24 No 3 (1 Aug 1981), pp 1552-1555. Abstract: The phonon dispersion relations of PdD0.78 have been measured by inelastic neutron scattering at temperatures slightly above and below the order-disorder transition. The acoustic modes are not sensitive to the ordering process and are somewhat higher in frequency than the correspond- ing modes in PdD0.63. The optic frequencies in the disordered phase of PdD0.78 are equal to those of PdD0.63. After ordering, a slight increase was observed in some transverse branches. Comments: A good discussion of some of the interesting vacancy ordering effects found in PdH compounds, which may be indicative of long-range coherency effects. 2. O. Blaschko, P. Fratzl, and R. Klemencic, "Model for the structural changes occurring at low temperatures in PdDx," Physical Review B, Vol 24 No 1 (1 July 1981), pp 277-282. Abstract: A structural model for the short-range-ordered state in PdDx, with 0.7 < x <= 0.78, observed at low temperatures, is proposed. The model describes the complicated diffuse intensity contours with "mixed" micro- domains consisting of cells of the NinMo type. A simple distortion, i.e., a relaxation of the D atoms towards vacant sites, is introduced in the "mixed" microdomain and accounts for the measured intensity asymmetry. Furthermore, a measurement of the intensities of the superlattice reflect- ions in the long-range-ordered Ni4Mo structure is presented, and its result is reproduced by calculation, when the same relaxation towards the vacancy is assumed. Finally an important Debye-Waller factor for the superlattice reflections is found. 3. O. Blaschko, P. Fratzl, and R. Klemencic, "Model for the structural changes occurring at low temperatures in PdDx; II. Extension to lower concentrations," Physical Review B, Vol 24 No 11 (1 Dec 1981), pp 6486-6490. Abstract: The concept of "mixed" microdomains presented in a previous paper for the description of the complicated diffuse intensity contours occurring in PdDx at low temperatures for x > 0.71 has been extended to describe the diffuse scattering at lower concentrations, i.e., near 0.65. A diffuse neutron scattering experiment in PdD0.69 is presented and the results in this intermediate-concentration range can easily be explained in the framework of the extended model. The model shows that the simple isointensity contours at lower concentrations can be described by "mixed" microdomains whose components are cells having a (100) mirror plane. The transition to the complicated contours for x > 0.71 occurs when in the composition of the "mixed" domain the contirbution of cells asymmetric relative to the (100) plane is progressively increased at higher concentrations. 4. O. Blaschko, "Fermi-surface imaging effect in the D short-range order of PdDx," Physical Review B, Vol 29 No 9 (1 May 1984), pp 5187-5189. Abstract: A Fermi-surface imaging effect is proposed to explain the concentration-dependent D short-range order in the vicinity of the (1,1/2,0) point of PdDx with 0.7 < x < 0.78. Within a rigid-band approximation the shape of the Fermi surface in PdD0.75 is deduced from that calculated by Gupta and Freeman for stoichiometric PdH. In PdD0.75 the obtained Fermi surface shows flat portions and 2kF vectors close to the (1,1/2,0) points. 5. O. Blaschko and J. Pleschiutschnig, "Chainlike hydrogen ordering in alpha-ScDx systems," Physical Review B, Vol 40 No 8 (15 Sept 1989-I), pp 5344-5349. Abstract: Hydrogen ordering is investigated in the alpha-ScDx system for two concentrations x = 0.19 and x = 0.33 by diffuse neutron-scattering techniques. The results reveal the formation of chains, consisting of pairs of hydrogen atoms located on second-neighbor tetrahedral interstitial sites along the hexagonal direction. The results are similar to those found previously in alpha-LuDx. The differences between alpha-ScDx and alpha-LuDx give evidence for a chain formation due to a lowering of coherency stresses. 6. T.E. Ellis, C.B. Satterthwaite, M.H. Mueller, and T.O. Brun, "Evidence for H (D) Ordering in PdHx (PdDx)," Physical Review Letters, Vol 42 No 7 (12 Feb 1979), pp 456-458. Abstract: Resistivity, annealing, and neutron-diffraction studies of PdHx and PdDx, provide evidence for ordering of the H and D atoms on the octahedral interstitial sublattice. Neutron diffraction results on a PdD0.76 single crystal indicate that the D atoms order on the fcc interstitial lattice with every fifth (4,2,0) plane vacant and all others filled, corresponding to the Ni4Mo structure. 7. Hector E. Avram and Robin L. Armstrong, "Low-Temperature Evolution of the Proton Magnetic Resonance Line Shape in beta-Phase Palladium Hydride," Journal of Low Temperature Physics, Vol 69 Nos 5/6 (1987), 391-400. Abstract: Measurement of the proton magnetic resonance absorption line shape in beta-phase palladium hydride indicate that there is no short-range ordering associated with the order-disorder phase transition that occurs in the neighborhood of 50 K. However, measurements of the line shape in the range 10-30 K show a change from a Gaussian to a non-Gaussian profile. This effect is not related to the order-disorder phase transition. A possible explanation in terms of thermally assisted tunneling is proposed. 8. Lawrence R. Pratt and J. Eckert, "Molecular dynamics of a dilute solution of hydrogen in palladium," Physical Review B, Vol 39 No 18 (15 June 1989-II), pp 13170-13174. Abstract: Molecular-dynamics results on a dilute solution of H in Pd are presented and compared with available incoherent inelastic neutron scattering results. The embedded-atom model adopted here does a good job of describing the H-Pd atomic forces probed by incoherent inelastic neutron scattering. The time correlation functions associated with the computed spectras are strongly damped and indicative of the anharmonicity that has been suggested as the principle contribution to the anomolous isotope dependence of the superconducting transition temperature in PdH. These results highlight the fact that the H-atom vibrations in Pd-H solutions are low-frequency, large-amplitude vibrations relative to the vibrations of H atoms in usual covalent interactions. The rms displacement of the H atom from its mean position in the center of the Pd octahedron compares favorably with the available neutron-diffraction results. 9. M.M. Beg and D.K. Ross, "The quasielastic scattering of cold neutrons from the beta-phase of palladium hydride," Journal of Physics C: Solid State Physics (Great Britain), Vol 3 (1970), pp 2487-2500. Abstract: The cold neutron time of flight technique has been used to study the diffusive motions of hydrogen in the beta-phase of palladium hydride. The quasielastic scattering was measured at 20 degrees C, 100 degrees C, 150 degrees C, and 200 degrees C. The data were found to be well predicted by Lorentzians broadened by the appropriate resolution function. In contrast to the results obtained by Skold and Nelim in the alpha-phase, the dependence of the broadening on momentum transfer was not as predicted by the Chudley-Elliot model for jumps between octahedral sites. It was, however, well fitted by a model which included the possibility of jumps between octahedral and tetrahedral interstitial sites as well as octahedral ones. The resultant diffusion coefficients were described by the expression D = 1.1 * 10^-3 exp ( -3380 / RT ) cm^2 s^-1. At 100 degrees C this agrees with the macroscopic diffusion coefficient measured by Wicke and Bohmholdt, but the temperature dependence of their results indicated the much higher value of 5700 +or- 200 cal for the activation energy. The predicted jump time extrapolated to 250 K, 10^-10 s, is a factor 50 less than the value obtained from the measurements of the spin-lattice time in the nmr experiments of Burger, and possible sources of this discrepancy are discussed. 10. Hector E Avram and Robin L Armstrong, "Effects of tunnelling on proton magnetic resonance transient responses in beta-phase palladium hydride," J. Phyics C: Solid State Phys., Vol 20 (1987), pp 6305-6314. Abstract: A comprehensive study of the effects of hydrogen atom tunnelling on proton magnetic resonance transient responses is reported for beta-phase palladium hydride. Data for beta-PdHx are presented for concentrations X = 0.80, 0.75 and 0.71 at 40 K and are characteristic of data taken over the temperature range 30 to 80 K. It is shown that the free-induction decay signal can be represented by the same empirical function for each concentration studied. The ratio of second to fourth moments deduced by fitting this function to the data is consistent with a random distribution of hydrogen atoms over the octahedral sites of the Pd lattice. However, the moments themselves are reduced from their rigid lattice values as a result of the tunnelling of the hydrogen atoms. A solid-echo study shows that the direct dipolar coupling between protons is much more important than indirect coupling. The presence of a Hahn echo is related to the distribution of Knight shifts in this non-stoichiometric alloy. The characteristics of this echo are documented. 11. C.J. Carlile and D.K. Ross, "An experimental verification of the Chudley-Elliot model for the diffusion of hydrogen in alpha-phase Pd/H," Solid State Communications (Great Britain), Vol 15 (1974), pp 1923-1927. Abstract: Measurements have been made of quasi-elastic neutron scattering from hydrogen in a single crystal of palladium in the alpha-phase. The energy broadening was obtained for Q vectors in the (200) direction at 573 K and in the (220) direction at 623 K out to a maximum Q value of 2.0 Angstrom^-1. These were found to be in good agreement with the Chudley- Elliot model for residence times of 4.31 and 2.84 ps respectively. 12. Herbert H. Johnson, "Hydrogen-vacancy trapping and detrapping kinetics," Scripta Metallurgica, Vol 23 (1989), pp 1203-1206. Comments: Lists hydrogen-vacancy relaxation times and diffusion activation energies for Al, Cu, Ni, Fe, and Pd. 13. P. Brill and J. Voitlaender, "Palladium and Proton Knight Shift in Palladium Hydride," Berichte der Bunsenges. physic. Chem., Vol 77 No 12 (1973), pp 1097-1103. Abstract: The nuclear magnetic resonance of ^105 PD has been observed in alpha-PdHn for the first time. The resonant field was measured at 75 degrees C as a function of hydrogen pressure p within the entire alpha phase region. A linear relationship between Knight shift and hydrogen concentration has been established by comparison with the p(n) absorption isotherm at 75 degrees C. Moreover, it has been found that the ^105 Pd Knight shift K and the susceptibility chi of alpha-PdHn are linearly related. From the K - chi dependence a value of -(1.80 +or- 014)% [sic] for the Knight shift in pure palladium metal at 75 degrees C and a value of -(1.49 +or- 015)% [sic] for PdH0.025 at 75 degrees C could be determined. This may be attributed to the decreasing 4d spin susceptibility within the alpha phase. The proton Knight shift in alpha-PdHn was also studied at 75 degrees C. Similarly to the ^105 Pd Knight shift for the ^1 H Knight shift in alpha-PdHn depends linearly on hydrogen concentration and susceptibility. Knight shifts between -(93 +or- 8) ppm for PdH0.014 and -(50 +or- 5) ppm for PdH0.035 relative to the bare proton have been obtained. The proton Knight shift K in beta-PdHn was observed at 0 degrees, 25 degrees, 50 degrees, and 75 degrees C under hydrogen pressures between 0 and 700 Torr. From a comparison of the p(K) isotherms with the corresponding p(n) isotherms the ^1 H Knight shift in beta PdHn has been obtained as a function of hydrogen concentration. In the neighbourhood of the phase transition the K(n) isotherms become temperature dependent and show a hysteresis. Within the pure beta phase the K(n) curves coincide and show no temperature dependence. For hydrogen concentrations greater than 0.70 the Knight shift is linearly related to the hydrogen content. With increasing hydrogen concentration the proton line shifts towards lower applied magnetic fields. For PdH0.62 a proton Knight shift of -(13.5 +or- 0.5) ppm and for PdH0.75 a Knight shift of -(2 +or- 0.5) ppm compared to the bare proton was obtained. This increasing Knight shift may be due to the vanishing d spin susceptibility and the increasing s electron density at the proton site. 14. J. Zbasnik and M. Mahnig, "The Electronic Structure of Beta-Phase Palladium Hydride," Z. Physik B, Vol 23 (1976), pp 15-21. Abstract: The electronic energy bands for some beta-phase palladium hydrides have been computed using the augmented plane wave method. The Xalpha exchange potential was used. It was found that the localized d-type states are insensitive, in a relative sense, to the hydrogen concentration, while the other states are constantly elevated in energy as the hydrogen concentration is decreased. Some differences are noticed when the present bands are compared to the photoemission spectra, but the computed Fermi level density of states is in good agreement with electronic specific heat and magnetic susceptibility measurements. 15. K Wyrzykowski, A Rodzik and B Baranowski, "Optical transmission and reflection of PdHx thin films," J. Phys.: Condensed Matter (United Kingdom), Vol 1 (1989), pp 2269-2277. Abstract: Transmission and reflection data (0.5-5eV) for 300 Angstrom PdHx thin films (x = 0.65, 0.85 and 0.92), hydrogenated under high-pressure conditions are presented and compared with similar data for pure palladium before and after hydrogen desorption. The main difference in transmission data consists of a pronounced peak around 1 eV in the transmission of PdHx compared with the structureless continuously increasing transmission in pure palladium. This confirms the theoretical expectation of Khan and Riedinger that the inter-band absorption edge in PdH occurs at about 1 eV. The reflectivity data were measurable for PdH0.65 only; samples with higher hydrogen contents (x = 0.85 and x = 0.92) exhibit a reflectivity below 0.5%, which is thus not quantitatively measurable by the method applied. This high reduction in reflectivity cannot be explained by the increase in surface roughness only. A possible contribution from the hydrogen itself is briefly discussed. 16. J.E. Worsham, Jr., M.K. Wilkinson and C.G. Shull, "Neutron Diffraction Observations on the Palladium-Hydrogen and Palladium-Deuterium Systems," J. Phys. Chem. Solids, Vol 3 (1957), pp 303-310. Abstract: Neutron-diffraction investigations on powdered samples have shown that both hydrogen and deuterium atoms in beta-phase Pd-H and Pd-D are located in the octahedral positions of the palladium lattice. Results obtained for samples with low gas concentration are inconclusive in determining the atomic positions in the alpha-phase, since at room temperature only a small amount of gas enters this phase. Although the vibrational amplitudes of hydrogen and deuterium are similar to those observed in other compounds, the total neutron scattering cross-section for hydrogen in this system is abnormally low, indicating that the protons are more nearly free than in the other hydrogen compounds. 17. Donald M. Nace and J.G. Aston, "Palladium Hydride. I. The Thermodynamic Properties of Pd2H between 273 and 345 degrees K," Journal of the American Chemical Society," Vol 79 No 14 (25 July 1957), pp 3619-3623. Abstract: The 30 degree absorption isotherm for hydrogen dissolved in a highly active palladium black has been measured and found to agree very closely with an equilibirum desorption isotherm obtained previously by Gillespie and Hall. Partial molar heats of desorption for the system palladium-hydrogen have been calculated from the isotherms. Partial molar heats of absorption and desorption of hydrogen from palladium hydride also have been measured in an adiabatic calorimeter. These values are used to calculate the free energy and the entropy change associated with the formation of Pd2H from the elements at 30 degrees. Agreement between the calorimetric heats and those calculated from the temperature coefficient of the equilibrium pressure establishes a reasonable criterion of equilibrium for the system. 18. Donald M. Nace and J.G. Aston, "Palladium Hydride. II. The Entropy of Pd2H at 0 degrees K," Journal of the American Chemical Society," Vol 79 No 14 (25 July 1957), pp 3623-3626. Abstract: The heat capacity contribution due to the absorbed hydrogen atoms in palladium hydride at a composition of nearly Pd2H was measured in a newly constructed adiabatic calorimeter over small temperature intervals from 16 to 340 degrees K. A residual entropy of 0.59 +or- 0.18 e.u. has been calculated from experimental data to exist at 0 degrees K. It is concluded that Pd2H does not approach a completely ordered state at low temperatures and so is not a true compound. The shape of the heat capacity curve indicates the probability of covalently bound hydrogen at low temperatures with a dissociation process occurring as the temperature is varied. More evidence for this model is to be given in the succeeding paper. 19. Donald M. Nace and J.G. Aston, "Palladium Hydride. III. Thermodynamic Study of Pd2D from 15 to 303 degrees K. Evidence for the Tetrahedral PdH4 Structure in Palladium Hydride," Journal of the American Chemical Society," Vol 79 No 14 (25 July 1957), pp 3627-3633. Abstract: The heat capacity of Pd2D has been measured from 15 to 152 degrees K and 280 to 300 degrees K for comparison with that of Pd2H to verify a theoretical conclusion based on the shape of the lambda transition in the latter. For completeness certain points in the isotherm and partial molal heats of absorption of deuterium in palladium have been measured. The heat capacities of Pd2H have been analyzed. Reasonable barrier heights and frequencies have been calculated to fit for a model based on PdH4.7Pd at low temperatures. To explain the heat capacities above 120 degrees K, a dissociation process is postulated which involves a decrease in the average number of hydrogen bonds as the temperature is raised. Hydrogen diffusion through palladium is conceived as a movement of hydrogen atoms by a continuous combination of (1) rotation about a parent palladium atom and (2) bonding to, and subsequent rotation about, a neighboring palladium atom. A new theory has been developed to explain the phase separation and hysteresis in the absorption isotherm. 20. J.G. Aston and Paul Mitacek, Jr., "Structure of Hydrides of Palladium," Nature, Vol 195 No 4836 (7 July 1962), pp 70-71. A complete ascii-form reprint of this short paper is given below: ..... "The recent determinations of activation energies for slow processes in palladium hydride [1,2] confirm new results obtained in this laboratory. "The heat capacities of PdHx recently completed for x = 0.5 and x = 0.75 are show in Fig. 1 along with points for other concentrations.[3,4] Between 35 degrees K and 85 degrees K the heat capacity is obviously independent of the hydrogen concentration up to 0.75 H/Pd ratio. 3 -| * * | * * Cp | * * 2 -| * * (cal. | * * * * * deg.^-1 | * mole^-1 1 -| * of H2) | * | * 0 -+------|------|------|------|------| 35 45 55 65 75 85 Temperature (degrees K) Fig. 1. Cp of a mole of hygrogen in block palladium samples of ratio 0.75, 0.50, 0.25 and 0.125 H/Pd and palldium powder sample of 0.50 H/Pd ratio between 35 degrees K and 85 degrees K. [TB -- Individual data points have been ommitted from the ascii graph.] "Warm drifts in the calorimetry were again noted. Fig. 2 presents a logarithmic plot of the warm drift versus the reciprocal temperature for the earlier and new data from the slope of which activation energies were obtained.[3] 6.5 -| | 5.5 -| * 0.75 H/Pd from 200K to 250K | * 4.5 -| / 0.125 H/Pd * 0.125 H/Pd from 150K to 200K -ln heat | / from - rate 3.5 -| * * 200K _ * | / / to - 2.5 -| * 250K _ * | / * 1.5 -+--*----|------|------|------| 3.5 4.5 5.5 6.5 7.5 s 1/T x 10^-3 (degrees K) Fig. 2. Plot of reciprocal of T versus ln rate of heat evolution to give activation energies for the three samples. "These results, summarized in Table 1, provide clues leading to a model consistent with modern ligand theory which leads dirctly to the observed zero point entropy of Pd2H. Neutron diffraction investigations place the hydrogen atoms in the octahedral positions of the palladium lattice at room temperature [5] and down to 20 degrees K (ref. 6). In the proposed model applied to the beta-phase, these hydrogens form sheets through hydrogen bonding with the neighboring palladium atoms in the lattice. A square planar arrangement of hydrogens about the palladium atoms in these sheets is favoured by resonance leading to the Waa term in the Lacher quasi- chemical treatment. The sheets can take right angle bends in any direction, but never intersect.[7] "Table 1. Activation Energies for Diffusion in PdHx Region of Drift x = 0.125 0.25 0.50 0.75 ---------------- -------- -------- -------- -------- 150 - 200 deg. K 2.9 kcal 2.9 kcal no drift no drift 200 - 250 deg. K 6.4 kcal 6.4 kcal no drift 5.5 kcal "Simple numerology shows that at the composition Pd2H all palladium atoms at the corners of the face-centered cubes are part of sheets, with hydrogen midway between each palladium atom in the sheets and with no hydrogen outside the sheets. Above 0.50 H/Pd ratio cross planes of hydrogen and palladium sheets begin to form. This explains the sudden change from metallic to non-metallic conduction around 0.50 H/Pd ratio noted by Schindler et al.[8] "In the alpha phase, hydrogen is randomly distibuted between corener palladium atoms which are not part of sheets. Thus at low temperatures, due to resonance stabilization of the beta-phase, the alpha-phase has a negligible concentration of hydrogen, as follows from the calculations of Lacher. Since resonance favours full occupation of the square planar positions, at low temperatures all vacancies in the sheets will be filled. "Below the heat capacity peak at 55 degrees K, PdH4 molecules librate about an axis through the palladium atom and perpendicular to the fully occupied planes. As the temperature increases this libration turns to hindered rotation (ring diffusion) and as rotation starts the potential barrier to rotation decreases (that is, the value of Waa decreases). This is a co-operative type phenomenon and gives the type of heat capacities presented in Fig. 1. Since the addition of more hydrogen to the sample mererly forms larger sheets the heat capacity effect of the hydrogen is concentration independent. "At temperatures where vacancies in the sheets exist, diffusion to neighbouring cells can occur through rotation (ring diffusion). Such a process, leading to filling of vacancies, is the cause of the warm drift starting at 150 degrees K in the 0.125 H/Pd and 0.25 H/Pd compositions. On rapid cooling, the system does not reach the equilibirum which favours less vacancies. Its slow establishment produces the warm drifts near 150 degrees K with activation energy of 2,900 calories per mole. At 0.50 H/Pd ratio and above the sheets of the beta-phase are filled and there can be no lack of equilibrium with respect to the filling process. By nuclear magnetic resonance methods, valid at equilibrium, Spalthoff [1] measured the activation energy of this neighbouring cell diffusion in a 0.64 H/Pd ratio sample at equilibrium conditions and found an activation energy of 2,400 +or- 300 calories. "The warm drifts at 200 degrees K and higher are due to long-range diffusion from alpha- to beta-phase or from one sheet to another. By rapid cooling the sample cannot establish the true equilibrium which requires less hydrogen in the alpha-phase. The slow return to equilibrium requires the long-range diffusion occurring above 200 degrees K with an activation energy near 6,000 calories. The 0.50 H/Pd ratio has no incomplete sheets (no beta-phase vacancies) and therefore could not exhibit a warm drift in this region. Above 0.50 H/Pd ratio the sheets are cross-linked to form a set of systems of sheets at right angles with no vacancies in any single system. Only with the resonance energy thus available can the extra repulsion be overcome. "The activation energy for long-range diffusion is the same as the measured activation energy (6-8 kcal) for diffusion of hydrogen through palladium, [9] as it should be. Rapid cooling produces lack of equilibrium with respect to the size of these units which is slowly restored by the long- range diffusion. "At the composition 0.50 H/Pd if certain hydrogens on palladium atoms rotate through pi/2 the sheets suffer a sharp bend by pi/2. Since this rotation can occur at random, at equilibrium the situation is equivalent numeralogically to that of ice. There are three non-participating face-centred palladium atoms to each corner one. Since the species analogous to water is PdH2 in each corner palladium atom, the basic unit is Pd4H2. In keeping with this it was found [10] that the zero point entropy of two Pd2H units in palldium black is 1.18 +or- 0.36 (2 x 0.59 +or- 0.18). The theoretical value is 0.82. "We thank the National Science Foundation for its financial support in connexion [sic] with this work. "J.G. Aston Paul Mitacek, Jun. "Cryogenic Laboratory College of Chemistry and Physics Pennsylvania State University "1 Spalthoff, W., Z. Phys. Chem., Vol 29, 258 (1961). 2 Burger, J.P., Paulis, N.J., and Hass, W.P.A., Physica, Vol 27, No 6, 514 (1961). 3 For the lower concentrations, see Mitacek, jun., P., and Aston, J.G., Nature, Vol 191, 271 (1961). 4 Nace, D.M., and Aston, J.G., J. Amer. Chem. Soc., Vol 79, 3619 (1957). 5 Warsham, J.G., Fr., Wilkinson, M.K., and Shull, C.G., J. Phys. Chem. Solids, Vol 3, 303 (1957). 6 Schindler, A.I. (unpublished ork). Schindler, from his intensities, has found evidence for hydrogen at tetrahedral sites in the neighborhood of the heat capacity peak. 7 Lacher, J.G., Proc. Roy. Soc., A, Vol 161, 525 (1937). 8 Schindler, A.I., Smith, R.J., and Kammer, W., Comm. 1, Intern. Inst. for Refrigeration, Copenhagen (August 1959). 9 Barrer, R.M., Diffusion in and through Solids (Cambridge, 1941). 10 Nace, D.M., and Aston, J.G., J. Amer. Chem. Soc., Vol 79, 3627 (1957)." ..... Comments: Although many of the concepts it contains are now out of date or incomplete, this early Aston/Mitacek paper represents an important early attempt to make theoretical sense out of the unusual data derived from palladium hydride systems. It is also rather amusing to note that the recent "cold fusion" controvery is not the first time that experiment- ers have had to explain unexpected heat outputs from palladium hydride systems. This and another Aston/Mitacek paper (see below) go into consid- erable detail to formulate an explanation for "heat drifts" that occur for certain Pd/H ratios at low temperatures. 21. J.G. Aston and Paul Mitacek, Jr., "Solutions of Hydrogen in Palladium," Advances in Chemistry Series, Vol 39 (Nonstoichiometric Compounds) (1963), pp 111-121. Abstract: The heat capacity curves of solutions of hydrogen in palladium of certain compositions are used to throw light on the structures in the palladium-hydrogen system. The warming rate of the solutions after heating has been used to identify two processes of diffusion. A peak in the heat capacity curve at 55 degrees K is connected with an anomoly in the resistivity for solutions with the hydrogen-palladium ratio greater than H/Pd = 0.5. This connection, together with a change from metallic conduction below this ratio to semiconduction behavior above, is used as the basis for a model which also explains the diffusion processes. Comments: Essentially an expanded version of the short 1962 Aston/Mitacek letter to Nature, which was given above in its entirety. 22. E.O. Wollan, J.W. Cable, and W.C. Koehler, "The Hydrogen Atom Positions in Face Centered Cubic Nickel Hydride," J. Phys. Chem. Solids, Vol 24 (1963), pp 1141-1143. Abstract: Neutron diffraction observations on the nickel-nickel hydride system show that the hydrogen atoms occupy the octahedral sites of the face centered cubic lattice. The hydrogen to nickel atom ratio in the hydride phase was found to be 0.6 +or- 0.1. In these respects nickel and palladium behave similarly. The systems differ, however, in the one important detail that the palladium hydride is stable at room temperature whereas the nickel hydride is unstable. Comments: Of only modest relevance to palladium hydride work; simply shows that many of the structural issues of palladium hydride are shared with other metallic hydride systems. 23. F.A., Lewis, The Palladium Hydrogen System, Academic Press, London / New York, 1967. See especially Chapter 9, "Investigations of Changes of Lattice Structure and of Some Other Physical and Electrical Properties as Functions of Hydrogen and Deuterium Contents," and Chapter 10, "Theoretical Models for the Pd/H and Related Systems." 24. R. Griessen, W.J. Venema, J.K. Jacobs, and F.D. Manchester, "Electronic States of Concentrated Pd-H Alloys from de Haas-van Alphen Measurements," pp 123-127 in Hydrides for Energy Storage (A.F. Andresen and A.J. Maeland, eds.), Pergamon Press; proceedings of an international symposium held in Geilo, Norway, 14-19 August 1977. Abstract of Griessen et al article: The results of de Haas-van Alphen measurements on palladium-hydrogen alloys, with hydrogen concentrations up to 30 at. % H in Pd, show that for H/Pd ~< 0.06 the hydride remains in the alpha-phase. The observed decrease in the area of various extremal cross- sections of the d-hole ellipsoids at X and L is found to be entirely due to the lattice expansion induced by the interstitial hydrogen, and not to the added hydrogen electrons. For H/Pd ~> 0.07 the alloy cannot be quenched in a metastable single phase state. This leads to a sharp drop in the Dingle temperature TD at H/Pd ~= 0.07. Both TD and the dHvA frequencies remain approximately constant at higher hydrogen concentra- tions, as expected for a two phase alloy with incoherent segregation. First paragraph of article: "The observation of oscillatory quantum effects has traditionally been restricted to high quality single crystals of pure metals, very dilute alloys and stoichiometric compounds. We present here measurements of quantum oscillatory effects in palldium-hydrogen alloys with hydrogen concentrations up to 30%. These are to our knowledge the first measurements of quantum oscillations in concentrated transition metal hydrides." 25. Earl L. Muetterties, ed., Transition Metal Hydrides; Marcel Dekker, Inc. (New York), 1971. See Palladium Hydride section, pp 21-23, and its six corresponding references on page 30 (refs 55-56, 59-62). Comments: A good short, general introduction to issues in palladium hydride chemistry. Four of the palladium references from this book are already included in this reference list. The other two references are: a. E.O. Wollan, J.W. Cable, and W.C. Koehler, J. Phys. Chem. Solids, Vol 24, 1141 (1963). b. T.R.P Gibb, Jr., in Progress in Inorganic Chemistry, Vol 3 (F.A. Cotton, ed.), Interscience, New York, 1962, pp 419-424." 26. C.D. Scott et al, "A Preliminary Investigation of Cold Fusion by Electrolysis of Heavy Water," Oak Ridge National Laboratory Report ORNL/TM-11322 (Nov 1989). Abstract: Several tests have been made with electrolytic cells utilizing 0.1 to 0.2 N LiOD in D2O as the electrolyte and a palladium cathode surrounded by a wire-wound platinum anode operating at cathode current densities of 100 to 600 mA/cm^2. The cathodes were swaged to diameters of 2.8 or 5.5 mm with 8.0 to 8.5 cm of active length and then annealed in some tests. The electrolyte temperature was controlled and heat was removed by flowing water in a cooling jacket, and the cell was insulated. Cooling water an electrolyte temperatures were determinted by thermocouples; neutron and gamma-ray spectra were measured; and the electrolyte was periodically analyzed for tritium. In one test, an internal wire coil of platinum coated with palladium black was used in a closed system to recombine the electrolytically generated D2 and O2 without release of any off-gas. The electrolyte was periodically sampled and electrolyte of the nominal concentration was added to replace the volume withdrawn; makeup D2O was also added, when required, in those experiments which did not include a recombiner. Neutron and gamma-ray spectra were recorded on magnetic media; temperatures, coolant flow rate, and voltages were recorded and, in the last two experiments, acquired by a computer data acquisition system. Tests up to 1000 h in duration were made, and in some experiments excess power was detected for periods of many hours, usually in the range of 5 to 15%. However, during one 12-h period, excess power of up to 50% was observed. On three separate occasions, the neutron count rate exceeded the background by three standard deviations; in addition, an apparent transient increase in tritium in the electrolyte by at least a factor of 25 occurred during one test. Text from section 4, Conclusions: "Preliminary tests of the electrolysis of D2O utilizing LiOD electrolytes and palladium electrodes have not confirmed the "cold fusion" phenomena. However, there have been several apparently anomalous neutron count rates, one unexplained 25-fold increase in tritium, and periods of many hours of apparent excess energy. None of these results has been precisely reproduced, nor can they be explained by conventional nuclear or chemical theory." Comments: The slight net energy production over a period of hundreds of hours for the closed-system experiment labeled CF-4 is notable for the rigor with which it was obtained. However, the paper provides very little insight into the conditions needed to produce such unusual excursions, since (by design) it makes no attempt to correlate heat production and other anomalies to any detailed theories about the complex structure and dynamic behavior of palladium hydrides. Without some specification of which detailed aspects of PdHx structure and dynamics are critical, easy reproduction of the Scott result is unlikely. For most of the references listed below this point I was unable to obtain copies of the papers. The references are given as they appear in Chemical Abstracts. 27. 112: 127527p Solubility of hydrogen, deuterium, and tritium in palladium metal. Powell, G. L.; Laesser, R. (Oak Ridge 4-12 Plant, Oak Ridge, TN USA). Report 1988, Y-2395; Order No. DE89005304, 41 pp. (Eng). Avail. Ntis. From Energy REs. Abstr. 1989, 14(6), Abstr. No. 11503. The soly. of H, D, and T in Pd was measured, described anal. over a wide temp. range (296 to 1460 K), and used to calc. enthalpies of reaction, isotope sepn. factors, and phase boundaries. [The next five intriguing references are all taken from a single issue (Vol. 164, No. 1) of Z. Phys. Chem (Munich), which is the English translation of the widely respected German physical chemistry journal. Although all but one of them deal with niobium, niobium hydroxide, and tantalum, they are all involved with the general issue of proton tunneling and hopping in metallic compounds.] 28. 112: 43126u Nonadiabatic tunneling and diffusion of hydrogen in niobium. Wipf, Helmut; Neumaier, Karl (Inst. Festkoerperphys., Tech. Hochsch. Darmstadt, D-6100 Darmstadt, Fed. Rep. Ger.). Z. Phys. Chem. (Munich) 1988 (Pub. 1989). 164(1), 953-62 (Eng). The tunneling and local diffusion of H interstitials in Nb(OH)x (0.0002 <= x <= 0.011) was investigated by neutron spectroscopy in the temp. range between 0.05 and 160 K. The expts. demonstrate the transition from a low-temp. tunneling behavior (T <= 10 K, inelastic spectra) to a f jump diffusion behaviour at elevated temps. (T >= 10 K, quasielastic spectra). Up to ~70 K, both tunneling and jump diffusion diffusion is controlled by nonadiabatic (direct) coupling of the H to conduction electrons. Above ~70 K, the jump diffusion behavior is increasingly dominated by interaction with phonons. 29. 112: 43127v Tunneling of trapped hydrogen in niobium: dynamics, density of states, and isotope effect. Morr, W.; Mueller, A.; Weiss, G.; Wipf, H. (Inst. Angew. Phys. II, Univ. Heidelberg, D-6900 Heidelberg, Fed. Rep. Ger.). Z. Phys. Chem. (Munich) 1988 (Pub. 1989). 164(1), 963-8 (Eng). We present ultrasonic studies on H tunneling systems in superconducting and normal conducting Nb. In the superconducting state an absorption max. is found around 3.5 K for H and 5 K for D at frequencies of 90 MHz. This is caused by a relaxation process where the relaxation rate of the tunneling systems is detd. by their interaction with thermally excited quasiparticles. Moreover, the shift of the absorption max. can be explained almost perfectly by the relaxation rate being proportional to the square of the resp. tunneling matrix element. 30. 112: 43128w Nuclear magnetic resonance (NMR) studies of hydrogen tunneling and trapping in niobium. Pfiz, T.; Messer, R.; Seeger, A. (Inst. Phys., Max-Planck-Inst. Metallforsch., D-7000 Stuttgart, 80 Fed. Rep. Ger.). Z. Phys. Chem. (Munich) 1988 (Pub. 1989). 164(1), 969-74 (Eng). Measurements of the spin-lattice relaxation rate Gamma1 of protons in the system NbOH were performed from 2 K up to 450 K. A relaxation max. at about 200 K is attributed to the reorientation of H trapped by O. The low-temp Gamma1 is compatible with recent neutron scattering expts. On the basis of the present measure- ments, a model for the OH complexes in Nb is proposed which incorporates the results of neutron-scattering as well as internal-friction expts. 31. 112: 43129x Diffusion coefficient of hydrogen in niobium and tantalum. Vargas, Patricio; Miranda, Lorenzo; Lagos, Miguel (Dep. Fis., Univ. Santiago, Santiago, Chile 5659). Z. Phys. Chem. (Munich) 1988 (Pub. 1989). 164(1), 975-83 (Eng). The current data on H diffusion in Ta at 15-550 K and in Nb at 135-400 K can be quant. explained by the small polaron theory. The exptl. data can be understood assuming ground-state to ground-state tunneling between interstitial sites with tetrahedral symmetry plus a thermally activated contribution due to tunneling between excited states having octahedral symmetry. The break of the diffusivity curve at T ~= 250 K follows naturally. It evidences the transition between the tetrahedral and octahedral hopping. For Ta the second break of the diffusivity curve at T ~= 20 K indicates the recovering of the ground-state hopping with tetrahedral symmetry. Below T ~= 10 K and T ~= 7 K for Nb the diffusion coeff. becomes independent of T. 32. 112: 43133u Measurements of hydrogen diffusion coefficient in the alpha-phase concentration range of the palladium-hydrogen system by the diffusion-elastic technique. Kufudakis, A.; Cermak, J.; Lewis, F.A. (Inst. Phys., 180 40 Prague, Czech.). Z. Phys. Chem. (Munich) 1988 (Pub. 1989). 164(1), 1013-18 (Eng). The diffusion-elastic technique (DE) was applied to measure the diffusivity D of H in the alpha-phase of the Pd/H sytem at 50 degrees and to investigate the influence of boundary and initial conditions on derived value of D. The agreement between exptl. curves and theor. shapes and coincidence with available data indicate a good fit of the model to the chosen exptl. regimes. High sensitivity of the DE-effect as compared to the Gorsky effect allows a close check of this fit which is significant for a correct interpretation of involved processes and results. The items below are references that I missed when I put together the original palladium reference list in July 1991. They are mostly references from Chemical Abstracts, and in contrast to the first batch of references they include several papers generated by the "cold fusion" controversy. 33. 110: 45312j Molecular-dynamics investigation of hydrogen hopping in palladium. Culvahouse, J. W.; Richards, Peter M. (Dept. Phys. Astron., Univ. Kansas, Lawrence, KS 66044 USA). Phys. Rev. B: Condens. Matter 1988, 38(14), 10020-31 (Eng). Diffusive hopping of hydrogen in PdHx has been reexamd. with mol.-dynamics (MD) simulations in order to clarify earlier results of M. J. Gillan (1986). The same values for the diffusion coeff. and width Gamma_q of the quasielastic neutron scattering functions were obtained when Gillan's potentials are used; but the results go beyond those of Gillan by performing a statistical anal. of the jumps. The anal. shows that over 95% of the jumps are uncorrelated and nearest neighbor, but the distribution of times between jumps is highly nonexponential. This latter feature accounts for the departure of Gamma_q, noted by gillan, from the Chudley-Elliott (CE) theory based on simple one-step nearest-neighbor hopping. A simple phys. model of the hopping as a two-step process can explain the results, and suggests they are highly sensitive to the H-Pd potential at short distances. A new MD simulation with a harder-core potential gives a Gamma_q in much better agreement with the CE theory. The non-CE behavior might be seen in real expts. on systems with suitably soft potentials. 34. 111: 84469f Hydrogen hopping rates and hydrogen-hydrogen interactions in palladium hydride. Nygren, L. A.; Leisure, R. G. (Dept. Phys., Colorado State Univ., Fort Collins, CO 80523 USA). Phys. Rev. B: Condens. Matter 1989, 39(11), 7611-19 (Eng). Ultrasonic-attenuation measurements were performed on single-crystal PdHx over the temp. range 80-300 K and for 0.64 <= x <= 0.76. The longitudinal and the two independent transverse modes that propagate along the (110) cryst. direction were investigated over a range of frequencies from 10 to 150 MHz. Evidence was found for more than one type of motion. The assumption of two simple relaxation processes was sufficient to give a reasonable fit to all of the data. The attempt frequencies for these two processes were found to depend somewhat on H concn. and acoustic mode, but the activation energies did not. The activation energies and approx. attempt frequencies were found to be 0.155 eV and 3 x 10^12 s^-1, and 0.254 eV and 10^14 s^-1. These values are suggestive, resp., of tunneling and classical barrier hopping. The relaxation strengths (deltaC's) derived from the fit do not satisfy the Cauchy relation, which means that central forces do not fully describe the H-H interactions. By restricting consideration to nearest- and next-nearest-neighbor H-H interactions, elements of the force-dipole tensor describing H-H interactions were found, including a noncentral force contribution. 35. 111: 183456f Palladium L3 absorption edge of palladium hydride (PdH0.6) films: evidence for hydrogen induced unoccupied states. Davoli, I.; Marcelli, A.; Fortunato, G.; D'Amico, A.; Coluzza, C.; Bianconi, A. (Dip. Mat. Fis., Univ. Camerino, 62032 Camerino, Italy). Solid State Commun. 1989, 71(5), 383-90 (Eng). Exptl. evidence for H-induced unoccupied states in the beta-phase of the PdH0.6 system is reported. The XANES spectroscopy at the L3 edge of Pd was used to probe the unoccupied states at the selected Pd site and for selected angular momentum (tI = 2 and tI = 0). The formation of he H1s Pd4d antibonding states which appear at 5.6 eV above the Fermi level as a new narrow band was obsd. in PdH0.6. In the beta-phase there is a set of unoccupied states induced by H, giving a peak in the L3 XANES at 0.7 +or- 0.2 eV above the Fermi level. Comments: This paper is mostly about how the electronic structure of Pd is affected by heavy loading with H, rather than about H itself. 33. 111: 69086k The origin of the inverse isotope effect in the palladium hydride and palladium deuteride (PdH and PdD) systems. Chen, Hua; Liu, Fusui (Phys. Dep., Peking Univ., Beijing, Peop. Rep. China). Phys. Lett. A 1989, 137(9), 485-8 (Eng). Using the Eliashberg equation in the Matsubara representation, the calcn. of superconductiong Tc for the PdH and PdD systems was performed under 2 schemes, and the inverse isotope effect was restudied. The inverse isotope effect derives from the harmonicity of the lattice in most cases. Other expts. support these results. 34. 77: 5259g (see also 6612z) Electronic energy bands of metal hydrides. Palladium and nickel hydride. Switendick, A. C. (Sandia Lab., Albuquerque, N. Mex.). Ber. Bunsenges. Phys. Chem. 1972, 76(6), 535-42 (Eng). Electronic energy band cacns. for transition metal hydrides provide a new theoretical understanding of these materials. The calcns. show when the proton model has relevance and when the anion model is more appropriate, the essential feature being the structure of the hydride. The structure in turn is detd. by the lowest energy state of the solid. The relative positions of the electronic energy levels are found to be a function of the interat. distances. The relative stability and structures ofthe metal hydrides are understood in terms of their band structures. The calcns. lead to a unifying new model, the "band model," for interpreting and understanding the electronic properties of these materials. Energy band calcns. of the Pd-H system for various structures are given. The calcns. were carried out with the H atoms occupying various interstitial positions in the fccd. Pd lattice to represent varying H contents for PdHx (specifically, for x = 0.25, 0.75, 1.0, and 2.0). In contrast to the simple proton model for this system, the calcns. are able to show why beta-phase PdH saturates at higher H concens. than the known concen. of holes in Pd bands. Calcns. for Ni hydride also show why the simple bandfilling concept fails. This result is in agreement with recent exptl. electronic specific heat measurements. Both systems demonstrate that the rigid band model is not applicable. Comments: This is an interesting but difficult paper. (I have a copy of it that is packed away at the moment.) The comments about proton (H+) and anion (H-) modeling are interesting because they introduce the idea of multiple species of hydrogen isotopes in palladium. My best recollection is that the "band model" that is mentioned in the abstract refers to conventional electronic bands, not heavy fermion bands. 35. 82: 4686h Diffusion of hydrogen in the beta-phase of palladium- hydrogen studied by small energy transfer neutron scattering. Nelin, G.; Skold, K. (R. Inst. technol., Studsvik, Swed.). Ab. Atomenergi, Stockholm [Rapp.] 1974, AE-493, 25 pp. (eng). The diffusion in beta-PdHx was studied by quasi-elastic neutron scattering. Diffusion occurs by jumps between adjacent octahedral interstitial sites. The obsd. integrated quasi-elastic intentsities are not described by a simple Debye-Waller factor. The alpha-beta phase transition was also studied. No dramatic changes in the scattering patterns were obsd. The diffusion mechanisms between low-concn. alpha-phase and high-concn. beta-phase are similar. D. Tomkin 36. 111: 179171s Diatomic hydrogen clusters and hydrogen diffusion in a metal. Reznik, A. I.; Tkachev, V. I.; Rudenko, N. V. (Fiz.-Mekh. Inst. im. Karpenko, Lvov, USSR). Fiz.-Khim. Mekh. Mater. 1989, 25(3), 7-12 (Russ). The existence of diat. H clusters in metals was proven theor. by calcg. the bond energy of H atoms. The theor. conclusions were confirmed in tests showing the increase in H diffusion rate in Ni, Pd, and Fe by irradn. with low-energy gamma-quanta as a result of the dissocn. of H clusters and increase in amt. of diffusion-mobile H atoms. 37. 112: 26130y Hydrogen interactions in palladium hydride (PdHn, 1 <= n <= 4). Wang, X. W.; Louie, Steven G.; Cohen, Marvin L. (Dep. Phys., Univ. California, Berkeley, CA 94720 USA). Phys. Rev. B: Condens. Matter 1989, 40(8), 5822-5 (Eng). By use of a first-principles pseudopotential total-energy approach, the electronic and structure properties of PdHn are studied with n varying from 1 to 4. The interaction between a pair of hyudrogen atoms in the octahedral cage of the Pd fcc. lattice is examd. in the light of recent claims of cold-fusion processes in metal hydrides. The calcn. shows that there is virtually no heavy d-electron screen effect for the hydrogen interactions at short distances. The sepn. between hydrogen atoms remains much larger than the value required for a significant fusion rate. However, our total-energy results indicate that, with a moderate chem. potential shift, a very high d. of hydrogen (n>=2) can be achieved inside the Pd crystal. Comments: The first-principles analysis of various PdHx loadings is well done and interesting. The following four papers are early "cold fusion" papers that looked at whether electronic screening effects might increase the background rate of D-D fusion in palladium. I recall being astonished at how seriously many people seemed to be taking the electronic screening idea at the time. (I.e., was palladium somehow supposed to mysteriously alter the simple, extremely well-proven QED behavior of electrons whenever they came into close proximity to H or D nuclei?) I'm bothering to list these beatings of a horse that was never born mostly because they provide useful backround information on the electronic structures of PdHx and PdDx. 38. 112: 25943d The two-stage screening model for impurity atoms in a host metal lattice from the view of quantum chemistry. Fritsche, H. G.; Mueller, H.; Opitz, C. (Sekt. Chem., Friedrich-Schiller-Univ., DDR-6900 Jena. Ger. Dem. Rep.). Z. Phys. Chem. (Munich) 1988 (Pub. 1989), 163(2), 535-40 (Eng). The EHMO and scattered-wave-Xalpha methods were applied to the Pd/H system as an example of interstitial at. H in transition metals for analyzing the charge distribution. The charge distribution, as postulated by the two-stage screening model of E. Wicke (1984), was indeed found. The electrons, donated by the hydrogen to the d-states of the metal, remain localized mainly at the Pd atoms neighboring the H atom. Two neighboring interstitial H atoms repell [sic] each other. 39. 111: 182148v Hydrogen-hydrogen/deuterium-deuterium bonding in palladium and the superconducting/electrochemical properties of PdHx/PdDx. Johnson, K. H.; Clougherty, D. P. (Massachusetts Inst. Technol., Cambridge, MA 02139 USA). Mod. Phys. Lett. B 1989, 3(10), 195-803 (Eng). A common quantum chem. origin of supercond. and anomalous electrochem. properties of H/D in Pd is proposed, based on the presense of an interstitial network of delocalized H-H/D-D s sigma-bonding MO at the Fermi energy. Dynamic Jahn-Teller oscillations of the p/d (optical phonons) promote e- pairing and supercond. at low temp., while they could be a principal source of electrochem. induced heat from Pd electrodes. Cold nuclear fusion in Pd might be promoted by such D-D oscillations, but at an extremely low rate and with insignificant heat prodn. Comments: I believe Dieter Britz's compendium also includes this abstract. 40. 112: 43755y Comment on "Cold fusion: how close can deuterium atoms come inside palladium?" Lam, Pui K.; Yu, Rici (Dep. Phys. Astron., Univ. Hawaii, Manoa, Honolulu, HI 96822 USA). Phyus. Rev. Lett. 1989, 63(17), 1895 (Eng). Total energy calcns. were performed on PdD1.25, assuming a structure where 2 D atoms occupy the central octahedron of a fcc. unit cell and a single D atom occupies the rest of the Oh sites. The equil. D-D distances are 0.9, 1.1, and 1.3 angstroms for the (100), (110), and (111) orientations, compared to 0.74 angstrom in free D2. There is an effective attractive Pd-D which pulls the D2 apart. The fusion process is unlikely to be enhanced in Pd. 41. 112: 43758b Pair interaction energy of hydrogen isotopes in metallic lattices. Estimate of fusion rates. Marchesoni, F.; Presilla, C.; Sacchetti, F. (Dip. Fis., Univ. Perugia, I-06100 Perugia, Italy). Europhys. Lett. 1989, 10(5), 493-8 (Eng). The fusion rates for the reactions pd, dd, and pp in a solid-state environment were detd. under the assumption that the fusion process is mediated by bound states of the reactant pair. An effective binding potential which accounts for lattice, electronic and pair interactions was envisaged to model plausible equil. and nonequil. conditions. Ests. for the upper-bound fusion rates in condensed matter lie below the actual exptl. limits. I am including this last reference mostly because Zuqia is the only "cold fusion" author I know of who invokes solitons as his mechanism. This paper does not mention the word "soliton" directly, but another paper by Zuqia (see Dieter Britz's compendium) makes it clear that the nonlinear effects Zuqia describes below are in fact solitons. As best I as I could understand them from his short paper, Zuqia solitons appear to be quite different from heavy-particle band solitions I have proposed. Zuqia solitons are particle waves mediated in some fashion by the palladium lattice, whereas my own proposed heavy particle band solitons are discontinuities in the actual wavefunctions of delocalized particles. 42. 112: 43761x A possible explanation of the room temperature nuclear fusion. Huang, Zuqia (Inst. Lower Nucl. Energy Phys., Beijing Norm. Univ., Peop. Rep. China). Beijing Shifan Daxue Xuebao, Ziran Kexueban 1989, (2), 43-4 (Eng). A possible phys. model is proposed. The d absorbed in the Pd lattice would form a sublattice in non-equil. state. They would oscillate vigorously and be coupled nonlinearly through sep. interactions with the Pd lattice. A cooperative effect thus induced would cause d energies to conc. on a few of them. The latter would be energetic enough to cause appreciable nuclear fusion. Comments: I once looked at the idea of whether band solitons might support another type of fusion reaction, but eventually concluded that if band solitons were truly capable of slamming fermions together with that much force, then polyacetylene with its rich supply of electron band solitons should be dangerously radioactive. Having never heard of any cases of polyacetylene radiation poisoning (and also because of other difficulties), I eventually concluded that it is highly unlikely that heavy particle solitons could be a direct cause of fusion reactions in solid media. ========= Copyright protection ends with (and includes) this line =========