WO1990013128A1

WO1990013128A1 - Enhancing nuclear fusion rate in a solid

Info

Publication number: WO1990013128A1
Application number: PCT/US1990/002171
Authority: WO
Inventors: Mario Rabinowitz; David H. Worledge
Original assignee: Electric Power Research Institute, Inc.
Priority date: 1989-04-25
Filing date: 1990-04-20
Publication date: 1990-11-01

Abstract

A method of producing power by nuclear fusion by introducing fusible nuclei into a solid carrier material (20).

Description

ENHANCING NUCLEAR FUSION RATE IN A SOLID BACKGROUND OF THE INVENTION

This invention relates generally to nuclear fusion in the solid state, and more particularly to techniques for increasing the fusion rate of solid solutions of low atomic weight nuclei in a solid metal lattice with special

applications to the generation of electrical power.

Nuclear fusion is an ideal source of energy since one of the potential fuels, deuterium, occurs in vast amounts in the oceans. In addition there is relatively little radioactivity associated with fusion compared with nuclear fission as an energy source. Because of the

conversion of mass to energy, substantially more energy is produced than the energy input into the system.

Much work has been done for over three decades on high temperature plasma controlled fusion. However, the achievement of sustained controlled fusion in high- temperature plasmas still seems remote. Moreover, the apparatus is expensive and cumbersome.

In another approach, called muon catalyzed fusion, the electrons orbiting the hydrogen nuclei are replaced by muons. This decreases the orbit size, i.e. the radius of the hydrogen isotope, by the ratio of muon mass to the electron mass, which is a factor of approximately 200. This reduction in the width of the Coulomb barrier by a factor of 200 increases the tunneling probability by many orders of magnitude for fusing the nuclei of hydrogen isotopes (H) such as deuterium and/or tritium. Muonic fusion occurs at low temperatures. However, this method presents great difficulties in the production of the muons, their capture by the fusion reaction products, and to a lesser extent the very short half-life of the muons (of the order of

microseconds). In March, 1989, Professors Stanley Pons and Martin Fleischmann ("P&F") announced their achievement of sustained controlled fusion at room temperature in a palladium (Pd) electrolytic cell using heavy water (deuterium oxide) as the electrolyte. Liquid solutions of deuterium oxide and of tritium oxide have densities comparable to that of the deuterium in the palladium in the P&F demonstration of fusion. Yet fusion has not been seen in these liquids. In addition palladium and other metals such as platinum (Pt) have been used to purify hydrogen and its isotopes from other gases as the hydrogen isotopes move readily through windows of these metals, but other gases do not - not even helium. Fusion has not been observed in these

circumstances.

Therefore for these reasons, hitherto unconsidered physical mechanisms must be present for the fusion to occur at the observed levels. The implementation of and

improvement upon these mechanisms is the basis for our invention as described herein in its many embodiments.

SUMMARY OF THE INVENTION

The present invention provides techniques,

applicable individually or in combination, for enhancing the fusion rate of light isotopes in a solid carrier material. One aspect of the present invention contemplates increasing the collision frequency of fusible isotopes in a solid carrier by reducing the number of degrees of freedom of the isotope motion. In one variation, the nuclei are

constrained to essentially one-dimensional motion by forming the carrier material as thin filaments or providing thin filamentary channels within the carrier material. In another variation, the nuclei are constrained to essentially two-dimensional motion by forming the carrier material in thin layers. This reduction in dimensionality may be accomplished by a variety of techniques such as vapor deposition, sputtering, and ion bombardment. Alternatively, it is possible to use hydrogen absorbing material having naturally occurring channels or planes.

The collision frequency of the fusible nuclei may also be increased by increasing the velocity of the fusible nuclei. This is accomplished by making the channels or planes more conducting than other parts of the solid, and applying electrical current so as to result in local

heating.

A further aspect of the invention contemplates increasing the fusion rate by increasing the tunneling probability of fusible nuclei through the Coulomb barrier. One way of accomplishing this is to increase the effective mass of electrons in the solid material, as for example, by selecting a heavy fermion material. Another way is to decrease the effective mass of fusible nuclei such as hydrogen isotope ions, as for example by using solids with relatively large lattice spacing (say greater than about 3A), as for example by selecting a heavy fermion material.

A further aspect of the invention utilizes

superconducting materials to provide quasi-particles

(electron pairs) which provide both increased electron mass and increased electron charge to increase the tunneling probability of fusible nuclei through the Coulomb barrier.

A further aspect of the invention deals with the initiation and/or acceleration of the fusion reaction by magnetic compression to increase the density of the fusible nuclei.

A further understanding of the nature and advantages of the present invention may be realized by reference to the remaining portions of the specification and the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

Figs. 1A and 1B show a microcircuit technique for manufacturing material having channels;

Figs. 2A and 2B show lateral and longitudinal channels in a hydrogen dissolving or absorbing solid (HAS); Fig. 3 shows radial channels in a HAS;

Figs. 4A and 4B show a conducting tube, and the magnetic field generated when a current flows in it;

Figs. 5A and 5B show a tube structure and the magnetic fields generated by currents flowing through the components;

Fig. 6 shows a channeled crystal lattice;

Fig. 7 shows a planar pathways crystal lattice;

Fig. 8 shows a planar sandwich structure in a HAS; Fig. 9 shows a slurry of HAS whiskers in a steam generating device operating in connection with a DC source;

Fig. 10 shows a slurry of HAS whiskers in a steam generating device operating by induction;

Figs. 11A and 11B are side and top cross-sectional views of a multi-rod HAS structure immersed in heavy water for producing heat;

Fig. 12 is a schematic depiction of a periodic potential (depicted in one-dimension) of a solid lattice in which the ionized hydrogen nuclei find themselves;

Fig. 13 is a schematic one-dimensional depiction of the Coulomb barrier that is tunneled in the fusion process;

Fig. 14 shows the band structure for electrons in a HAS as related to their effective mass;

Fig. 15 shows band structure for hydrogen ions in a hydrogen dissolving solid as related to their effective mass; and

Fig. 16 shows a heavy or quasi-electron between two hydrogen nuclei.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Overview

The present invention contemplates a number of techniques which may be practiced individually or in

combination. To provide an overall perspective, these are presented in this Overview section. There are four essential items for sustained controlled nuclear fusion: (1) Collision Frequency; (2) Tunneling Probability; (3) Fusion Probability; and (4)

Sustaining the reaction. The power delivered by nuclear fusion (fusion rate) is proportional to the product of the first three items. The fourth item relates to prevention of poisoning the reactions and replenishment of fuel (deuterium and tritium) . This invention relates to method and

apparatus for improving items (1), (2), and (4).

Basic Fusion Reactions. The basic deuterondeuteron fusion reactions are as follows:

1. H² + H² → H³ + H¹ + 4 MeV

2. H² + H² → He³ + n + 3.3 MeV

3. H² + H² → He⁴ + 24 MeV

At high energies (i.e. at high temperatures), reactions 1 and 2 occur with about equal probability and reaction 3 occurs only about one-millionth of the time in high temperature plasmas. Reaction 3 does not occur in free space.

P&F have reported a 4-watt output. This would seem to imply 10¹² to 10¹³ fusions per second, but the measured neutron and other radioactivity levels seem to imply a rate many orders of magnitude lower. One possible reason is that the fusion rate is indeed low and the

additional energy is being supplied by co-existing

exothermic chemical reactions. Two other possible reasons are much more exciting, and if correct, have great practical import. One possibility is that at the low temperatures (energies) of the P&F experiment, reaction 1 occurs at a much higher frequency than reaction 2. This would explain why the power output is so much higher than the

radioactivity count -- simply because this reaction does not produce neutrons, and reaction 2 is greatly suppressed in comparison. Reaction 1 may be favored at low energy of the fusing particles because this provides time for mutual polarization of two approaching deuterons. Since the center-of-mass does not coincide with the center-of-charge for the deuteron, at low energy the Coulomb repulsion between the two protons in the two deuterons approaching on a collision course will orient the nuclei so the neutrons are facing each other and the protons are as far apart as possible. This favors reaction 1. The other possibility is that reaction 3 is also favored at low energy in a solid.

In any event the desirability of reactions that do not produce neutrons is manifest since neutrons and the energy they carry are lost to the system unless there is a great deal of shielding. Additionally, the neutrons can produce undesirable radioactivity in the ambient

environment.

If H³ and He³ are available, the following

reactions can take place:

4. H² + H³ → He⁴ + n + 17.6 MeV

5. H² + He³ → He⁴ + H¹ + 18.3 MeV

A rare reaction that occurs in stars but might occur in a solid is: 6. H¹ + H¹ → H² + e⁺ + v

Reaction 5 is particularly desirable since it is also neutronless, and the end products are charged. Reactions 1, 3, 5, and 6 allow the possibility of direct energy

conversion to electricity, and would thus avoid the penalty of Carnot conversion efficiency in a heat cycle.

The probability for reaction 3, and others that may not be seen in the plasma state, is enhanced due to the increased probability for a many-body collision per reaction in the solid state which increases the number of reactions that can conserve energy and momentum compared with two body collisions. Three other reactions for fusing particles may be possible. These are:

7 . H¹ + H² → He³ + γ + 5. 4 MeV

8. H¹ + H³ → He⁴ + γ + 20 MeV

9. H³ + H³ → He⁴ + n+n+10 MeV

Decreasing the Effective Isotope Mass. The fusion rate is extremely sensitive to the tunneling probability (coefficient). The present invention contemplates changing the effective hydrogen isotope mass, which has the greatest impact as even small variations in the relevant variables can substantially change the tunneling coefficient.

Tunneling is considered to be a quantum mechanical

phenomenon in which a particle whose energy is less than the potential energy of a barrier can nevertheless be found on the other side of the barrier. For fusion to occur, the hydrogen isotope nuclei (e.g.deuterons, and/or tritons) must overcome the repulsive Coulomb barrier and fuse together. One aspect of the invention relates to the discovery that it is possible for the effective mass of the fusing particles in a solid to differ from the mass of these particles in free space sufficiently to increase the tunneling

coefficient by many orders of magnitude.

Due to the periodic potential of the lattice ions in a solid and the quantum mechanical wave nature of the particles that move freely in this lattice, it is possible for the effective mass of the particle in the solid to differ from its mass in free space (free particle mass). In the case of electrons in a solid, the effective mass may be either less or greater than that for a free electron. The same seems to be true of deuterons and/or tritons in a solid as related to the band structure of these particles. This is related to the periodic potential of the lattice in terms of its shape, spacing, and depth for these particles. When the effective mass of the hydrogen isotope is less than the free mass, this increases the tunnel coefficient by many, many orders of magnitude. When the effective mass outside the nuclear Coulomb barrier is greater than the free mass, and the effective mass inside the nuclear Coulomb barrier is less than or equal to the free mass, the tunneling

coefficient is also increased. Thus this aspect of the invention relates to producing the proper effective mass of the fusing particles in the solid, as influenced by the solid's lattice structure. This will be described in greater detail below.

Enhancing the Collision Frequency. The fusion rate is proportional to the collision frequency of the fusing particles. One aspect of the invention contemplates enhancing the collision frequency by decreasing the degrees of freedom in the solid so that the fusing particle (e.g. deuteron or triton) is confined essentially to two or one dimensional motion in the solid (i.e., to planes or

channels). In three dimensions, the collision frequency per particle is F₃ = ncv, where n is the number density, c is the cross section, and v is the mean thermal velocity. This number should be roughly the same in the liquid state and in ordinary solid solution such as in Pd. Decreasing the dimensionality (degrees of freedom) decreases the number of ways potentially colliding particles can miss each other. The two-dimensional case can be viewed as similar to the collision of marbles on a table top. The table top need not be flat as long as the marbles are confined to moving on its surface. The one-dimensional case can be viewed as similar to marbles constrained to move in a tube. The tube need not be straight; even with a sinuous tube, marbles cannot avoid collision.

The actual solid lattice is much more complicated than this simplified marble example because on the atomic and sub-atomic scale the world is quantum-mechanical, and because it is difficult on this scale to achieve only two degrees of freedom in the planar case, and only one degree of freedom in the channel case. However, for the purposes of numerically illustrating that a significant increase may be achieved, it is helpful to consider some very simple equations as crude approximations to the real situation. In two dimensions the collision frequency is F₂ = n^2/3c^1/2v. In one dimension the collision frequency is F₁ = n^1/3v.

Depending on the particular values of n and c when the particles are confined to planar motion in the solid, F₂ can be several orders of magnitude larger than F₃.

Similarly, F₁ can be roughly many orders of magnitude larger than F₃. Not all the volume of the solid is available to the absorbed particles which increases the effective number density. Similarly, not all the volume of the channels or planes is available to the particles, also increasing the effective number density. This increase in the number density, combined with the increase in cross section caused by the reduced dimensionality substantially increases the collision frequency (rate). The fusion rate is proportional to the collision frequency, and is thus greatly enhanced. Increasing the Effective Electron Mass. In certain solids, especially the heavy Fermion metals, the effective mass of the electrons is substantially higher than in free space due to the periodicity, shape, spacing, and depth of the lattice's ionic potential wells. The effective mass arises because of the quantum mechanical wave nature of the electrons in this environment, and the physics of the wave's group velocity. Also important is the fact that an electron is coupled both to the lattice and the ensemble of other electrons in the solid by the Coulomb interaction so that it does not act like a free particle.

It appears that one may be able to utilize the increased effective mass of electrons in a solid to great advantage. In Pd the electron's effective mass may not be too much greater than for a free electron. However, in some materials, such as the heavy Fermion metals, the electron's effective mass can be 100 times greater than for a free electron. Even a factor of 2 can be very important and be amplified by many orders of magnitude in tunneling and fusion probability.

In accord with the Heisenberg Uncertainty Principle, the greater effective electron mass permits greater localization of electrons between hydrogen isotope nuclei. Thus, the Coulomb barrier is decreased in width, and the nuclei may be brought much closer together (by a factor approximately inversely proportional to the increase in the electron's effective mass). This increases the tunneling probability by many orders of magnitude. In simple terms, any increase in the electron effective mass helps a lot; and if the effective mass is increased by a factor of 100 or more, the nuclei are close enough together that the nuclear force can help overcome the Coulomb repulsion.

Electron Pairs. Palladium is not a

superconductor, but palladium hydride, Pd_XH_Y, is a

superconductor at about 10K. This material has an inverse isotope effect because of the zero point vibrational energy of the interstitial hydrogen atoms. So, contrary to the standard BCS theory, the heavier isotopes of hydrogen have higher transition temperatures, rather than lower ones.

Thus, palladium deuteride, Pd_XD_Y, has a transition

temperature of about 12K, and palladium tritide, Pd_XT_Y, has a temperature of about 13K. According to the Rabinowitz theory as described in Quantum-Gas Model Estimate for Wide Range of Superconducting Temperatures (Intl. J. Theo. Phys. vol 28, No. 2, pp. 137-145, 1989), fine filaments of Pd_XH_Y may be superconducting at close to room temperature. Fine filaments on the order of 100A in diameter of Pd_XD_Y and

Pd_XT_Y may be superconducting above room temperature, at about 324K and 351K respectively. Quasi-two-dimensional Pd_XD_Y and Pd_XT_Y will have appreciably higher critical temperatures than their three-dimensional forms. As described in the Rabinowitz paper, highly compressed hydrogen in three dimensions may be a superconductor at ~ 300K; and quasi-one- and quasi-two-dimensional hydrogen and its isotopes as may exist in the solid of this invention may also be superconducting at even higher temperatures.

As in the previous descriptions of effective isotope mass tunneling enhancement and increasing the effective electron mass, electron pairing as occurs in high- temperature superconductors can greatly enhance the

tunneling probability (coefficient). For large energy gaps in high temperature superconductors, the separation between electron pairs (coherence length) can be as small as atomic dimensions. Unlike the conventional BCS superconductors where the coherence length is on the order of 10³A and the pairing is in phase space rather than real space, it is possible for the coherence length to be on the order of 1A (since the coherence is inversely proportional to the energy gap) and the pairing to be in real space for high

temperature superconductors.

Since nuclear dimensions are considerably smaller than 1A, electron pairs must be able to come closer together than 1A to enhance tunneling between nuclei. One might expect the Coulomb repulsion to put a limit on the nearness of approach of the electrons in a pair. Let us consider the statistical nature of the physics and fluctuations in the coherence length. In the presence of a supercurrent, as the velocity of the electrons gets larger (increase in current density) the magnetic force of attraction between the electrons can reduce the Coulomb repulsion (in the limit as the electron velocity in the lab frame approaches the velocity of light, the two forces cancel). Even though in the center-of-mass frame, the electrons have opposite momenta, they can have in common a large velocity in the lab frame at high current density. Although individual

electrons obey Fermi-Dirac statistics, electron pairs obey Bose-Einstein statistics. Statistically, a small number of electron pairs with very high velocities can be close enough together and get between two approaching hydrogen isotope nuclei to enhance the fusion probability by many orders of magnitude. Even for electron pairs with lower velocity, as they encounteir approaching hydrogen isotope nuclei, the attractive Coulomb interaction to the nuclei will bring the electron pair closer together, which in turn will greatly increase the tunneling coefficient for the nuclei to fuse.

Thus it is possible for the electrons to be close enough together that the combination of two electron charges and two electron masses will act to doubly screen the

Coulomb barrier of two nuclei as well as reduce the barrier width. The effect will be to reduce both the height and width of the Coulomb barrier, thus greatly enhancing the fusion rate.

Illustrative Embodiments

The fusion enhancing embodiments of this invention can be constructed in various sizes, shapes, and

configurations. Figs. 1A and 1B provide a schematic

depiction of a microcircuit technique for manufacturing channels. The process is similar to that used in the manufacture of integrated circuits. A number of layers 10 of a substrate material 12 have outer faces on which open semicircular channels 15 are cut. When the substrate material itself is not the desired material for direct contact with the hydrogen isotopes, the cut channels are overlaid with the desired material by electrolytic, vapor deposition, or sputtering techniques, etc.

Materials such as Pd, Pt, Ti, Zr, and heavy

Fermion metals can be deposited in the channels as thin films 11 or be used directly as the substrate material 12. These and other materials will be described and discussed below. The final channel 15 has dimensions of atomic size, say less than 15A diameter and preferably on the order of 1A. The completed channels can be checked by a variety of non-destructive diagnostic techniques such as with the electron tunneling microscope. In fact, this instrument can be utilized both for making and checking the channels. The channels can also be machined with microarcs. Layers 10 are then bonded with the open channels on facing surfaces registered to provide a structure 17 having cylindrical channels 18.

Alternatively, channels may be produced directly in the desired bulk material by ion bombardment using proton, alpha, or other ion beams. If hydrogen isotope beams such as deuteron and triton beams are available, the channels may be cut and loaded with the hydrogen isotopes in the electrode by this ion bombardment. The unloaded

(unfilled) channels formed by ion bombardment may be

increased in diameter by standard acid etching or leaching techniques. The rod may be used as the cathode electrode in an electrolytic cell, or in non-electrolytic apparatus as described in some embodiments of this invention. In

addition to the manufacture of channels, as already

described, naturally occurring crystalline materials can provide channel and/or planar structure as will be described below.

Fig. 2A shows a completed rod structure 20 having channels 22 running in a direction lateral to the long axis of the rod. Fig. 2B shows a rod 25 having channels 27 running substantially in the lengthwise direction of the rod. The rod can be made of materials like LaNi₅, LiH, BeH₂, LiBeH₃, LiBeH₅, Li₂BeH₄, LiBe₂H₅, etc. (H represents the hydrogen isotopes). The other alkali metals, such as Na and K, may replace Li. A less hazardous material would be LiMgH₃ as would other light metal hydrides such as LiAlH₄. LiBeH₃, Li₂BeH₄ and probably LiMgH₃ have the desired perovskite-like structure. The issues of toxicity,

reactivity, oxidation, etc. may be avoided by means of a thin coating of an inert material over the rod which is thin enough to permit ingress of hydrogen isotopes. This would be desirable for LaNi₅ which is a remarkable absorber of hydrogen isotopes. It can absorb and desorb large amounts of hydrogen isotopes rapidly at convenient temperatures and pressures. For all the HAS, desorption of normal hydrogen as a preliminary step to adsorption of deuterium and/or tritium greatly speeds up the absorption process.

Fig. 3 shows a rod 30 with radial channels 32.

The directional advantage of the channels relates to the overall geometry and electric field configuration when the hydrogen isotope containing electrode is used in an

environment in which currents are set up in preferential directions in the rod used as an electrode such as in an electrolytic cell. The radial channels are preferred when the configuration is radial. The lateral or longitudinal channels are preferred when the configurations are lateral or longitudinal with respect to the electric field.

Figs. 4A and 4B relate to method and apparatus for initiating and/or accelerating the fusion reaction. Fig. 4A is an oblique view of a hollow cylinder 35 having inner and outer radii R₁ and R₂, thereby defining a central channel 37. A current is shown flowing into the page. Fig. 4B is a plot of the magnitude of the magnetic fields as a function of distance from the axis. The channel radius R₁ is of atomic dimensions on the order of angstroms (A). The radius R₂ is substantially bigger than R₁. Very high current densities may be produced by very high short-duration current pulses. In these short pulses, the material can carry many orders of magnitude higher current density than it can in steady state. The current pulse causes both heating of the fusible particles (deuterons and tritons), and magnetic constriction of the channel 37.

When R₂ is also microscopically small, the

magnetic field produced even at extremely high current densities (10⁷ amp/cm²) plays a secondary role to the extremely high power density that is produced by the Joule heating of the current. The ionized hydrogen isotopes dissolved in a solid or in the atomic sized channel 37 may be thought of as a highly-density cold plasma. In this situation the current pulse produces rapid heating of the plasma to a temperature on the order of 10³K and pressures on the order of 10³K atmospheres before there is a substantial decrease in the number density of hydrogen isotope plasma ("HIP"), and before the cylinder permits expansion. This increased kinetic energy of the ions is sufficient to accelerate the fusion process, when the ions are below the threshold for fusion prior to the current pulse.

The power density (per unit volume) is equal to the resistivity of the cylinder times the square of the current density. If cylinder 35 is superconducting, its resistivity is negligible while the current pulse is rising until the critical current density is reached, and then the resistivity rises sharply. Use of superconductivity

materials for the cylinder enables current densities of the order of 10⁶ to 10⁷ amp/cm² to be attained before there is appreciable heating. Thus expansion effects can be

minimized, and a sharper power heating pulse can be attained than just inherent in the sharpness of the current pulse itself since the transition from the superconducting state to the normal state is very sharp. The superconductor can serve two functions, as it can also be a source of quasiparticles of double charge and double mass to assist in tunneling as described above. This is applicable when the radius R₁ of channel 37 is of atomic dimensions. Table 1 lists some exceptional metallic superconducting materials, as well as each material's transition temperature T_c and second critical field H_c2. Many of the alloys are of the previously mentioned naturally channeled beta-tungsten structure, especially the Nb₃ alloys. (H_c2 is listed for 4.2K, and decreases as the temperature increases.)

Table 2 lists some ceramic oxide superconductors that may also be used. The ones with the higher transition temperatures are preferable with respect to ease of meeting the cooling requirements. The higher temperature ceramic oxides all have very high second critical fields. In thin film form, as would be used in the small cylinders, they have high critical current densities which exceed 10⁶ amp/cm². The metallic superconductors have critical current densities as high as 10⁷ amp/cm².

For small cylinders, the power density for producing high temperatures and high pressures can be controlled by the magnitude of the current pulse and its frequency of application. For a lattice work of small superconducting channels as previously described, the local current density in the cylinder can be very high even though the overall current density is quite low e.g. of the order of only 1 amp/cm².

For superconducting operation, the cylinder must be cooled. Table 3 lists some coolants that may be used. Care must be exercised in using some of the coolants, such as liquid oxygen and liquid hydrogen, which are potentially explosive. Liquid hydrogen can serve both as a coolant and as a source of hydrogen isotopes for insertion into the appropriate solid in molecular, atomic, or ionic form. When atomic hydrogen is placed in a conducting solid, it may get ionized in the solid if the conduction band of the solid is not filled. As already mentioned above, palladium hydride is a superconductor for all the isotopes of hydrogen; and hydrogen and its isotopes in the solid may be in metallic and possibly even superconducting form.

Figs. 5A and 5B shows an embodiment that provides better control of the heating and magnetic constriction of the HIP. A small hollow cylinder 40 has inner and outer radii R₁, and R₂ to define a central channel 42 of radius R₁ (as in the case shown in Fig. 4A). Cylinder 40 is

surrounded by a hollow cylinder 43 of thermally and

electrically insulating material having inner and outer radii R₂ and R₃. Cylinder 43 is surrounded by a relatively large hollow cylinder 45 having inner and outer radii R₃ and R₄. To produce magnetic constriction of the HIP in the atomic dimension channel 42, a current pulse I₂ is produced in cylinder 45 many orders of magnitude greater than is possible in small cylinder 40. The concurrent current pulse I₁ in cylinder 40 can thus be separately tailored to produce a maximum temperature subsequent to the maximum pressure produced by the current pulse in the outer cylinder. Thus in igniting the cold plasma the outward kinetic pressure in the channel need not act to negate the inward magnetic pressure. The magnetic force produced is proportional to the square of the current I₂. The compression caused by the magnetic field serves to increase the HIP number density and by decreasing the channel diameter increases the collision frequency, particularly by making the channel more closely approximate one-dimensionality. Fig. 5B is a plot of magnetic field, and shows the relative magnitudes of the magnetic fields, but not to scale, from the two cylinders.

Fig. 6 shows an example of a naturally occurring crystalline material 46 that has a structure defining channels 47. These may be used in the practice of this invention in the various channeling embodiments. Materials which have the beta-tungsten (wolfram) structure, A-15 compounds possess this property. Many superconducting materials fall in this category, especially the Nb₃ alloys in Table 1. Many non-superconducting materials also fall in this category.

In a face-centered cubic lattice such Pd, deuterons see a large repulsive potential at each of the metal ions with periodic potential wells at the positions of the octahedral interstitial sites. A deuteron of thermal energy with effective mass in the lattice equal to .02 of its free mass, has a reduced DeBroglie wavelength, L, of about 10 ^-7 cm, and about 10^-9 cm for the free mass. This is a measure of the deuteron localization. In either case, in tunneling from one interstitial site to an adjacent one (potential depth ~ 1-2 eV) the particles are extremely unlikely to take paths differing laterally by more than L because such paths are inhibited by the tunneling

probability. In addition to the advantages of quasi-one- dimensional motion described above, the relative velocity, on average, will be twice the deuteron velocity. This has a significant effect on increasing the tunneling probability for fusion. One aspect of the invention is to promote this quasi-one-dimensionality by proper choice of materials, or to produce it synthetically.

Fig. 7 shows an example of a naturally occurring crystalline material 48 that has a "planar-like" structure having planar channels 49. In this context, a channel is considered planar if its width is greater than twice its height (typical height on the order of 1-15A). These may be used in the practice of this invention in the various planar embodiments. Materials like the dichalcogenides,

intercalated graphite, the ceramic oxide superconductors, and Perovskites in general possess this property. Almost all of the superconductors in Table 2 have Perovskite-like structure. Many superconducting materials fall in this category in addition to the recent high temperature

superconductors. Many non-superconducting materials also fall in this category.

Fig. 8 illustrates a planar-like sandwich structure 50 synthesized by vapor deposition, sputtering, etc., of materials like Pd, Pt, Ti, Zr, zeolites, heavy Fermion metals, the materials listed in conjunction with Figs. 2, 3, and 17, etc., in layers 52. By use of masking techniques, supports 55 are laid down alternately between layers 52 to leave planar-like spaces 57 of atomic thickness to accommodate the hydrogen isotopes. A trestlework support structure may be added to provide mechanical strength for the sandwich structure 34.

Fig. 9 shows a vertical schematic cross-section of a fusion tank 60 for the generation of electricity by means of a Carnot cycle (heat cycle). A source of particles 62 small filaments or hollow cylinders) of hydrogen absorbing solid, HAS is mixed with heavy water 63 (D₂O and/or T₂O) to form a slurry 64. The slurry is then circulated inside tank 60 with net flow through a hollow central cathodic electrode 70 that has a large communicating hole 72, and smaller side communicating holes 75. The central electrode is connected to the negative terminal of a DC power supply 77, whose positive terminal 78 is connected to the tank 60 which serves as the anode. The cathodic electrode 70 is attached to an electrical insulator 80 and communicates electrically through the electrolyte slurry 64 with the anode. An ordinary fluid 82, such as water in an annular chamber, surrounds the fusion tank 60 and is in good thermal contact with the tank. By heat exchange, fluid 82 goes to high temperature and then circulates to an electrical generating device. After performing work, the cooled fluid 85 returns to the annular chamber.

There are several advantages to having the HAS in fine filamentary form. First, this enables a slurry to be formed so that HAS particles may be circulated and HAS particles removed as they become defective. Second, the smallness of particles permits them to be quickly charged with the hydrogen isotopes. Third, the very smallness lets most of the fusion energy out of the particles, rather than permitting large dissipation of the fusion energy inside the particles which would damage the lattice more quickly.

Fig. 10 shows a device 90 for generating electricity which is similar to that of Fig. 9, except that a time-varying power source such as a pulsed power source 92 couples to the electrolytic slurry by means of a coil 95. The HAS particles 97, in solid or hollow form are mixed with heavy water 98 to form a slurry 100 which circulates in the tank. Use of a central hollow electrode 102 with

communicating holes 105 is optional in this induced voltage case, though it does help to direct the flow. As before, a fluid 107 like water surrounds the tank and is heated indirectly.

Figs. 11A and 11B are schematic side and top cross-sectional views of a heat producing fusion device 110 containing an electrolyte of heavy water 112, and a mat-like arrangement of HAS filaments 115 which may be solid, but are preferably channeled on an atomic scale as described above. The mat arrangement is the cathode and is connected to the negative terminal of a power supply 117. The positive terminal is connected to the tank 118. An electrical connection 120 is made between mats. Electrical connection 120 also provides support structure for the filaments or rods 115 in a given mat. The advantages of this structure are: (1) a large surface to volume ratio is provided for fast loading of the hydrogen isotopes in the HAS; (2) the thinness of the filaments or rods minimizes lattice damage in them from the energetic fusion products; and (3) the mats, being modular, may easily be removed for inspection and replacement.

Fig. 12 is a simplified periodic potential

(depicted in one-dimension) of a hydrogen absorbing solid (HAS) lattice in which the ionized hydrogen isotope nuclei find themselves. The potential is depicted as being square as an approximation to the real potential. The real

potential may be anisotropic, and may have a different periodicity in different directions. The potential barriers of potential energy (height) V_o have a width b, and form a well between them of width a. As shown a, b, and V_o are depicted as being constant; however, they may vary not only in different directions, but even in the same direction, and yet have a general periodicity about them.

Within a periodic potential, charged particles respond dynamically as if they had an effective mass, m^* =

²/[d²E/dk²], Equation (1) where

is Planck's constant/2π, E is the energy of the particle, and k = the particle's momentum/

is called the wave vector. This is true for electrons, protons,

deuterons, tritons, or other charged particles free to move about in the solid lattice. The effective mass is an inertial mass, and is not related to the energy equivalence of the particle, E = mc².

Although the following does not purport to be an accurate calculation of the deuteron or triton ion effective mass, M^*, the magnitude to be expected can be estimated from the simple one-dimensional Kronig-Penney model as

illustrated in Fig. 13.

M^* ~

²/Ea², Equation (2) where E represents the eigenstate energies for the ions.

Electrons and tritons are Fermions (particles of half- integer spin). Since only two Fermions (of opposite spin) of the same species can occupy the same state, the energy levels for the electrons and tritons will be higher than for the deuterons which are Bosons (particles of integer spin). As can be seen from Equation (2), or more complicated equations related to Equation (1) or the density of states, the effective mass of the deuteron is 1/50 of the free deuteron mass for thermal energy (E~1/40 eV) and for a well width a ~ 4A. Therefore, the periodicity a+b of the

periodic potential of the HAS as related to the lattice spacing should be large rather than small, and the well width a should be large and the width b of the equivalent square potential barrier small. This will result in a reduced effective mass of the fusing particles. The

tunneling probability through the Coulomb barrier goes up many orders of magnitude as the effective reduced mass of the tunneling particles goes down. The reduced mass of the tunneling particles is equal to the product of the two effective masses divided by the sum of the two effective masses.

Fig. 13 is a one-dimensional illustration of the three-dimensional Coulomb barrier through which the

particles must tunnel in order to fuse. The particles have an energy E which is one of the factors that determines the width, w, of the barrier to be tunneled. The barrier is of height B at the radius R of the nuclear well. Just as a reduced mass for the tunneling particle greatly enhances fusion, so does an increased mass for the electron(s) between the fusing particles, as this serves primarily to decrease the barrier width w, and also the barrier height B. The following descriptions will teach both how the effective mass of the electrons is increased, and how the effective mass of the fusing particles is decreased.

Fig. 14 shows the conduction band structure for electrons in a hydrogen absorbing solid (HAS) as related to their effective mass. Energy E of the electrons is plotted as a function of the electrons' wave vector k. As Equation (1) shows, the effective mass of the electrons is low in the region 1(e) of the low-mass minima because of its steepness. At zero or low current densities, the electron drift

velocity is small and the electrons will remain in region 1(e) of the low-mass minima. As the current density

increases, the drift velocities increase, increasing the electron momentum and hence k. Thus it is possible for electrons to transfer to region 2(e) of the high-mass minima, where their effective mass is higher than their free mass. Such transfer can occur by scattering of electrons which have high enough energy to scatter to region 2(e); or it can occur directly due to the gain in energy as the drift velocity is increased. This is similar to what happens in the Gunn effect in materials like GaAs in which a microwave travelling through the material is amplified.

Fig. 15 shows a similar plot to that of Fig. 14, for deuterons or other fusing particles (which might even include lithium). However, there is an important

distinction. For these particles the low-mass region 1(d) occurs at high k, i.e. large momentum or drift velocity. These particles find themselves in the high-mass region 2(d) at zero or low current densities. As the current density increases, the drift velocity and hence k increases bringing them to the low-mass region 1(d) for similar reasons to those given for the electron change in region. In region 1(d), the effective mass is lower than the free mass of the fusing particles. Since the velocities for these heavy particles are much lower than for electrons, it should be possible to also thermally excite these particles to the low-mass region 1(d). With low-effective mass, the fusion probability for these particles increases tremendously.

Thus, those materials in which the effective mass of the fusing particles is less than their free mass and/or in which the effective mass of the electrons is greater than their free mass will greatly enhance the fusion rate. When both effects occur, an even greater enhancement is achieved.

Fig. 16 shows a moment in time when a heavy or quasi-electron is between two deuterons with an arbitrary orientation of the protons and neutrons. The effective mass of the electron makes it heavier than the free mass. The mass m^* may also represent the paired mass of two electrons of effective charge e^* = 2e (twice the free electron

charge), which also increases the tunneling probability.

This larger effective mass, m^*, can be achieved in many materials, but is particularly striking in the heavy Fermion metals in which m^* ~ 10-100 times the free electron mass. The large effective mass results from the strong coupling between the conduction electrons and the local f-electron moment fluctuations characteristic of these materials.

These heavy electrons are almost as heavy as the muon, and move freely in the metal since they are in the conduction band. Thus these metals differ considerably from ordinary metals in their normal, antiferromagnetic, and

superconducting states. Table 4 lists the three-dimensional ordering temperature, T, for some heavy Fermion metals in these states.

The electronic specific heat per mole at low temperatures is substantially higher for the heavy Fermion metals than for Pd, and for example is as much as 162 times higher in the case of CeAl₃. The heavy Fermions have an electronic specific heat per mole that is 10² to 10³ times higher than for most metals. The Curie-Weiss temperature (related to the magnetic susceptibility) is 284K for UCu₅ compared with only 86K for Pd. The electronic resistivity at ambient temperature (~300K) is ~10 times larger than Pd's 10 micro-ohm-cm. As opposed to conventional metals, with few exceptions (UPt₃ and UAl₂) the electronic resistivity at ambient temperature decreases as the temperature is

increased. In addition to the desirable heavy electron mass, these metals have a large lattice spacing (a) which reduces the fusing particles effective mass as previously described. Other metals with lattice spacing in excess of 3A can also reduce the effective mass of the fusing

particles. However, the heavy Fermion metals can be even more effective in enhancing fusion.

While the invention has been described with reference to many embodiments, the descriptions are

illustrative of the invention and are not to be construed as limiting the invention. Thus, various modifications and applications may occur to those skilled in the art without departing from the true spirit and scope of the invention as defined by the appended claims.

Claims

1. A method of producing power by nuclear fusion, comprising the steps of:

introducing fusible nuclei into a solid carrier material; and

restricting the motion of the fusible nuclei to fewer than three dimensions.

2. The method of claim 1 wherein the solid carrier material is in the form of at least one filament or hollow tube.

3. The method of claim 1 wherein the solid carrier material is of a planar structure.

4. In a method of producing power by nuclear fusion wherein fusible nuclei are introduced into a solid carrier material in the presence of an electric field, the improvement comprising the step of:

restricting the motion of the fusible nuclei to fewer than three dimensions.

5. In a method of producing power by nuclear fusion wherein fusible nuclei are introduced into a solid carrier material, the improvement wherein the solid carrier material is characterized by a lattice spacing of greater than three angstroms.

6. The improvement of claim 5 wherein the lattice spacing is greater than four angstroms.

7. The improvement of claim 5 wherein the solid carrier material is a heavy fermion metal.

8. In a method of producing power by nuclear fusion wherein fusible nuclei are introduced into a solid carrier material, the improvement wherein the solid carrier material is a heavy fermion metal.

9. In a method of producing power by nuclear fusion wherein fusible nuclei are introduced into a solid carrier material, the improvement wherein the solid carrier material is a light metal hydride.

10. A method of producing power by nuclear fusion comprising the steps of:

introducing fusible nuclei into a solid carrier material; and

heating the solid carrier material.

11. The method of claim 10 wherein the solid carrier material is in the form of a filament or hollow tube.

12. The method of claim 10 wherein said heating step comprises applying a current pulse.

13. The method of claim 12, and further comprising the step of magnetically compressing the fusible nuclei.

14. In a method of producing power by nuclear fusion wherein fusible nuclei are introduced into a solid carrier material, the improvement wherein the carrier material is in a superconducting state.

15. In a method of producing power by nuclear fusion wherein fusible nuclei are introduced into a solid carrier material, the improvement wherein the material is in the form of filaments or hollow tubes having an outer diameter less than approximately 100 angstroms, the solid carrier material being a superconductive material.

16. The improvement of claim 15 wherein the

material is maintained at a temperature above the bulk

critical temperature for the material.

17. A method of producing power by nuclear fusion comprising the steps of:

providing a slurry of filaments or hollow tubes of hydrogen absorbing solid material and a liquid containing hydrogen isotopes; and

applying an electric field to the slurry.

18. A method of producing power by nuclear fusion comprising the steps of:

applying a time varying magnetic field to the slurry.

19. The method of claim 18 wherein the time varying magnetic field is a pulsed field.

20. A method of producing power by nuclear fusion comprising the steps of:

supporting a mat of hollow tubes or filaments of hydrogen absorbing solid material in a liquid containing hydrogen isotopes; and

applying an electric field.

21. A method of producing power by nuclear fusion comprising the steps of:

applying a time varying magnetic field.