New Hydrogen Energy with the “Patterson Power Cell” by Dennis Cravens, Ph.D.* Abstract The anomalous heat generation from hydrogen introduction into a metal lattice has a colorful and interesting history. Although the last seven years has seen a wide range of claims and counter-cwms, only a few embodiments have yielded a reproducible pattern and utility. The background of the Patterson Power Cell will be discussed during the symposium. The most important feature of the cell is that the electrodes have a uniform and large surface area composed of small spheres. These beads are composed of a plastic inner core plated with three layers: nickel, paha@ and nickel. Small research celkhave been produced with give I to 10 watts of anomalous or “excess heat” with fractions of watts of input electrical power. General background is given in preparation for a work session held in conjunction with the symposium Introduction In 1989, Stanley Pons and Martin Fleischmann announced that they could achieve what called “cold fusion” (CF) by introducing large ratios of deuterium into palladium by electrolysis(Fleischmarm,1989). After an inunediate flurry of activities their results were called ‘s now into question as others had difficulties in reproducing the claimed effects. This was primarily due to 1) lack of patience on the part of some researchers and 2) preconceived ideas of how the process must work. The basic signature of the effect is that power is released as heat at levels greater than the electrical power delivered to the electrolysis. It takes about 4 weeks before the excess heat (actually power) appears in a 4 mm Pd rod. In the first few months, some noted researchers tried to produce the effect but in most cases they did not load (i.e. place deuterium into the palladium) slowly over long times. Loading a piece of palladium initially at high current seems to produce cracks and other detrimental conditions within the metal lattice which prevents the effect from appearing. Large pieces of metal can develop large internal stresses as they are loaded and the diffusion into the metal decreases almost exponential with depth. It is a historical curiosity that the largest laboratories used large power supplies, large palladium samples and tried to load quickly to produce negative reports within months of the announcements. Such early negative reports’ have hampered the advancements in this area. Likewise, during the first several years, most researchers assumed that neutrons must be emitted at high levels’ from the events. During that time, it was assumed that the reaction must be a D + D reaction. It is well known that in high energy work that the D + D reaction has nuclear channels leading to equal production of T + p and He + n. By calculating the energy claimed to be produced within the cells, the levels of neutrons from the He + n channel can be calculated assuming rates known from high energy work. Yet the neutrons (Okamoto 1993) were not to be found in levels anywhere near the expected levels (I OE- 12 below expectations). Some tritium was found (Bockris, 1992) but at levels also below expectations (I OE-6). Most high energy physicists believed that if the reactions were somehow taking place in the small cell that they must proceed by the assumed D +D reactions and in the exact nuclear channels they observed in hot plasma. The low and doubtful levels of the neutrons and tritiums coming from the cells caused most high energy physicists to brand CF as a total mistake. Evidence has been mounting over the late year, or so that the heat generating reactions are not proceeding via D + D or even D + Li events. Instead they are proceeding through a H + Metal reaction (here H can be H- I or H-2 - D). In other words, the fusion is not between two hydrogen nuclei but between the hydrogen and metal nuclei. We may be burning the Pd or Ni metal within the electrodes. This of course presents problems of it own. The main problems are 1) how is the protori/deuteron getting inside the nucleus circumventing the Coulomb barrier and 2) why do the products of the reaction give no appreciable radioactive signatures. The Coulomb potential seems almost insurmountable unless there are additional forces at work to superimpose over that potential. Backround The most direct and common indicator that something is going on inside a hydrogen loaded piece of metal is that of anomalous heat. This has now been observed in a wide number of cases and methods. These include the use of Pd and Ni in electrolysis of heavy and natural water solutions, the use of molten salt electrolytes as discharges, metals in gaseous environments, and metals in acoustic fields. E. Storins has a review of these and other embodiments (Storms, 1996). There has been more study on the original Pd /heavy water electrolysis than any other. It appears from the work of McKubre (I 994), that large hydrogen to metal ratios are required for robust observation’of the effect. This means that the effect will not proceed until the hydrogen (meant to include both H-? and H- 1) has been loaded inside the metal at very large ratios. Additionally, the metal must be selected (or discovered) to have certain characteristics. The P@ requirement is that it does not unduly “sw@l! l” upon loading. legh levels of volume expansion on hydrogen loading seems to indicate production of internal voids, cracks, etc. This has been one of the more frustrating aspect of the typical Pd/ D electrolysis systems- that it is highly specific metal production batch and machining dependent. In the present work we will be working with normal water solutions and Nickel and palladium as our metals. The nickel-normal water methods seem to be less prone to the exact conditions of the metal. They..Oo, however, have their own variations from the Pd/D based effects. First, they seem to pref6r a lower current density whereas the Pd prefers a high current density. Like Pd the Mi/water systems do seem to have a positive temperature coefficient. This means that the reaction is accelerated at elevated temperatures. That is an important point for commercialization. Patterson Power The purpose of this paper is twofold: to describe the thermal energy measurements of the Patterson Power Cell (PPC) and outline overall experimental designs of such systems as ground work for nuclear measurements of subsequent nuclear studies. There are three basic measurements required to calculate the energy balance within the present flow design: 1) voltage and current of the electrolytic current 2) the flow rate of the electrolyte and 3) the temperature differential of the flow. Additional minor measurements are 4) thermal heat loss of the cell 5) the room temperature, 6) specific heat of the electrolyte, 7) gas evolution by the cell, 8) chemical potential changes within the system , and 9) pumping impedance of the cell. These minor measurements do not impact directly the qualitative restdts of excess power and by proper relative measurement and protocol they do not affect the conclusion of the apparent excess heat but only its absolute magnitude. As the experimental r! esults are questioned it is important to keep in mind how various factors affect the reality of the excess heat and the absolute magnitude of the effect. V,Ihen referring to power and energy the term cell is limited to that region of space enclosing just the cell and its insulation and not the support systems. The term system is meant to mclude the cell and all supporting devices such as the ptunp and pm-heater. The overall system is depicted in figure 1. ThevoltageandcurrentsaremeasuredtowithinO.5%vAth’Iraceablemeters”. Dueto the electrochemical interactions within the cell, they do not follow ohm’s law but have a resistance dependent on the applied current, concentration of the electrolyte, the temperature and other factors. Normal operating ranges are from 2 to 200 ohms. The voltage measurement are made with high impedance meter at the input leads to the cell. There is a trade off between low electrical resistance of leads and themw loss out of the cell by the leads. Short 22 gauge Pt wires are used for contact leads within the cell assembly but these are connect to a crimped stainless steel tubes which give electrical conduction through the end plugs but limit the thermal losses down the leads. Anyone trying to duplicate the design should consider , el@cm, thermal, and chemical factors. The electrical input to the cell is the most accurate measurement (in the constant DC current case) and is know to well within 1% and ! often to 0.01%. Typical input powers to the small cell ( 1. 9cm diameter research design) are 0. 0 1 0 to 0. 50 watts. The thermal output of the cell is calculated by knowing the flow rate, specific heat of the solution and the change of temperature of the flowing electrolyte. The flow rate is held constant by a constant volume pump in the laboratory models and by a fixed peristalic pump and synchronous motor in the material study models and some demonstration models. The flow rate is checked by time and periodic volume measurements. This has been found to be constant to within 2% when measured over I niinute intervals and spot checked several times daily. Total electrolyte volume between temperature probes is typically 30 mi/@n. It is important that the calibraticin runs be done at the same flow rates as the cell will be experience during the testing. The calibration constant ( slope of delta T vs. Watts) is dependent on the flow rate. Various cells dffer, but in the 1. 9cm design it is often observed that the calibration constant varies widely at low flow rates (below 20 mi/@n) but is f! airly consistent above 25 ml/min. Tubing, connections, insulation and air flow can affect the calibration constant and must be uniform between calibration and experimental runs. The temperature differential between the inlet and outlet to the cell is the most critical and sensitive measurement. Its calibration and experimental controls will be discussed in more detail. Original designs uses stainless steel temperature probe wells. That design was abandoned when it was found that the wells contributed substantially to impurities when searching for nuclear products. It was all so determined that they conducted room temperature to the probes causing spurious signals and that they artificially lowered the inlet temperature reading causing an over estimate of the thermal output at elevated working temperature. The “research cells” now being produced have a probe well within the end plugs. These are useful in relative measurements and when searching for nuclear products since the design mininiizes the exposure of the electrolyte to impurities. In present designs for heat measurements, the temp e probes consist of Nylon tubing which extends 4 cm into the flowing electrolyte. They enter from a sealed “T” and the well is coaxial with the inlet and outlet tubing. This design allows for rapid exchange of thermocouples and checking with RID’s and thermistors. A common mistake has been when others try to mqocate the inlet thermocouple too close to the cell’s bead bed. In that case the thez7nocoup)e will read an weighted average of the bead bed and the -inlet electrolyte flow. Normally the temperature of the inlet electrolyte is read at a position about 2 to 4 cm below the inlet to avoid the registration of heat from the bead bed. Only minimum temperature measurements where attempted to keep contamination down in the nuclear ash studies. The specific heat of the electrolyte (2.OMolar lithium sulfate in normal water) was measured to be about 0.98 . The specific heat of the solution is temperature dependent but is thought to be good well within I 00/a over the entire range of the experiment. The heat calibration assures that the relative measurements between active and inactive cells can be made without precise measurement of the specific heat. It is factored into the cell’s calibration constant if the same worldng fluids are used for both the experimental run and the calibration. If the absolute specific heat of the solution is required it can be compared to a nin with pure water. In the calorimetric studies, a loop of Teflon insulted tungsten wire was used as a calibration resistor within the cell’s housing and placed below the cathode connection plate. The ceff’s heat loss can then be found. The magnitude of the present heat generation is above unity even without factoring in the heat loss of the cell. However, to get absolute heat production the heat loss from the cell needs to be determined. The exact heat loss is dependent on the insulation and environment surrounding the cell. Typical values range from 5 to 10% loss in dewer systems, 10 to 20% in foam insulated systems and 15 to 30% in open air systems. Most current work is done with polyethylene, foam insulated systems and have a heat loss of about 15% at normal working conditions. The calibration of the cell is done by observing the temperature differential developed at various resistive heat settings. The cell normally deposits about 85% of the heat to the electrolyte and about 15% to the ! environment. Cell design The cell is made from a 1. 9cm ID quartz tube and is 4 cm in length. A general diagram is given in figure 2. The heat producing reaction takes place within the bed of micro spheres. This bed is less than one cm in length and occupies about 2 cubic cm of beads. Between I and IO watts of thermal energy is produced in this volume. The cylindrical design minimizes the thermal loss at the perimeter of the cell. It also allows for compression of the bead bed from the end plugs. This compression helps retain the u@forrn pacidng within the cell. The cell is sealed by 0-ring and the compression adjusted by means of plates and screws on the outside of the cell. The bottom of the bead bed contacts a nickel plate with is connected to the negative terminal of a DC constant current supply. The plate has an array of holes for the uniform flow of electrolyte through the cell. It is important that the fluid flow through the cell be uniform or “hot spots” will develop and will disrupt the loading and activity of the beads. The top plate is made of titanium or platinum. There is a nylon separator between the top positive plate and bead bed. Care is taken to make sure that both the anode (positive terminal) and the separator allow for both uniform fluid flow and uninterrupted gas flow. The oxygen at the anode is of special concern and the anode plate is slightly indented in the center to prevent oxygen build up which would greatly increase the cell’s resistance. The heart of the PPC is the specially prepared beads. These are produced (Patterson, US patents) by first making a cross linked polymer bead. Some of the sputtered beads use a glass core but the plastic has special properties which seem to help the system. These beads are affonated and then flashed coated with copper to achieve an intimate bond between the plastic and metal. The beads are then covered with about 6 microns of nickel. A layer of Pd is then added to thickness of 0.98 microns. Finally, the beads are plated with about I microns of nickel. During the plating or sputtering, special care is taken to assure a uniform coating of metal at each Page i3& l@r-navonal Forum on New Science step of the process, The exact conditions of the plating seem to affect the ffiw utility and performance of the beads. It has been reported that individuals using substandard layering techniques produce beads whose layers dimupt during loading or fail to load at all. The beads are stored under water (pure or with ammonia added) to keep the surf aces clean and fresh. The beads are spooned into the cell wet until they are to a depth of about 8 beads thick. It is easier to handle and pack the beads when they are wet. They are then w@ed down with water. This sets the beads. They have a close packed array configuration when property packed, A nonconductive nylon mesh is then placed on the bead bed and gently pressed down onto the bead bed. The nylon mesh (500 niicron ) was chosen to be fine enough to prevent passage of the bead (about 0.8 mm or 30 mesh) but coarse enough to allow for easy passage of the gas bubbles freed during electrolysis. Nylon is superior to most plastics since it is “wettable” and allows for better bubble flow. An 0-@ is placed on the mesh. Its purpose is to prevent the beads from making direct electrical contact with the positive electrode at the perimeter of the mesh and cell wall. When properly assembled the cell should not be conductive without the presence of the electrolyte. This prevents the current from being! carried by electrons (as in metallic conduction) and assures that the current is carried by only ionic species such as H+. Above the nylon separator mesh, a salt bridge is placed. This is prepared from a special ion exchange resin which has its positive sftes occupied by lithium. it fitnctions to keep the elecnic fields at the anode uniforta increase the effective surface area of the anode, and decrease the chance of the metallic layers on the beads from making direct electrical contact with the anode. Without the salt bridge the beads sometime melt through the nylon and short out the cell. Loading Initially the beads have a low hydrogen to metal ratio. It is critical that the first loading be done slowly at a low m=erit. legh levels of mffent densities during the first loading m result form a rapid expansion of the metal lattice and the metal dislodging from the surface of the beads. It is preferred that the initial loading be done in the range of 0. 0 1 to 0. IO amps per square 0 centimeters of the ceu’s cross section. During this period the hydrogen ions are loaded into the metal layers. This causes the ceff’s resistance to change. When loaded at constant current and temperature, the voltage across the cell typically increases by about I oo/a. It then slowly decreases by a few percent. It is assumed that this reflects the change in resistance of the metal layers due to hydrogen uptake. This takes between 4 and 12 hours for the present chemically deposited beads. With sputtered bead having layers only about 1000 Angstroms thick (instead of microns) this loading pr! oceeds much faster. The loading should be done at around room temperature. General Results Once the cell has been packed with beads and the beads electrolytically loaded with hydrogen, observations of the excess heat can be undertaken. Many different experimental runs and configurations have been conducted. The following is a typical but sppcific experimental run. The cell was loaded at 0.020 Amps for 6 hours at 23 degrees C. The current to the cell was set to 0.02OAmps (constant current setting)through out the experiment. The voltage was measured was3.llto3.27voltsduringtheentireninafterloading. TlfismeansthataboutO.06wattswas applied during the experiment. The 0.50 molar lithium sulfate electrolyte was pumped through the cell at a rate of 34.0 ml/min. The pressure measured at the pump at 2 psi and this represents the maximum pressure in the system with most of the pressure drop across the filter before it entered the cell. Before the heat generation measurements, the cell’s inlet temperature was slowly heated to 65 degrees C at a rate of about 10 degrees per hour. This heat was applied to the electrolyte by placing a IO foot length of tubing loca! ted between the fflter and the cell into a water bath and changing the bath’s temperature. A comparison of the inlet and outlet temperature of the cell showed a differential of 1.9 degrees ( 70.8 - 68.9 ). This corresponds to a “raw” ther7nal energy gain of 4.4 watts (specific heat of solution * flow rate ). This level was maintained for 3 days before the experiment was terminated by the researcher. Only about 85% of the heat generated in the cell is recovered by the electrolyte the remaining 15 % lost to the @orunent. Tlfis means about 5 watts of heat has been produced between the inlet and the outlet. This must be a result of some internal energy conversion since the cell is operating above room temperature (20 C) and the second law of thermodynamics would prevent flow from the room to the cell. On the other hand, only 0.06 watts of electrical energy is being supplied to the cell. Also, of that approximately 3 volts delivered, only 1.5 volts is available for Joule heating since about 1.5 volts is used to dissociate the water into the hydrogen and oxygen gas b! eing generated by the cell. For comparison, consider that the cell contains about 32 mg of Ni and 21.6 mg of Pd in the case of the sputtered micro spheres produced at University of Illinois. The most realistic exothermic reactions possible would be less than 590 J (140 cal) if all the metal was consumed by chemical events. At a modest power output of about I Watt, the heat could not persist for more than 590 seconds or about 10 minutes. Clearly the experiment above was producing total energy yields at levels thousands of times greater than the total chemical energy available within the bead bed. A range of considerations have been leveled at the heat generation. These will be dealt with at length in the work session if requested. Conclusion The Patterson Power Cell is unique in that it has been found on several occasions by individuals other than the inventor to produce anonwous heat or heat (more accurately power as a heat flux) in excess over the electrical power delivered to the cell. This anomalous heat is at levels orders of magnitude greater than can be accounted for by chemical events within the cell. Research is continuing to verify, quantify, and understand the origins of these effects. The author is placing ceds into the hands of skilled individuals and laboratories to be characterize the nature of these observations. The technology is patented and Clean Energy Technology, Inc (of DWias, Pac3c W @onal Forum on New.SGien@ Texas) is pursuing commercialization of these cells. References Bockris, JO’K ChenC., Hodko, D., Nfinevski,Z., “Tritium and Helium Production in Palladium Electrodes”, Third International Conference on Cold Fusion, October 21-25, 1992, Hagoya Japan. Page 23 I Fleischmann NV, Pons,S, 1989, “ Electrochemically Induced Nuclear Fusion of Deuterium”. J. Electroanal. Chem., vol 261, page 301 Okamoto,M. Y., YoshinagaM., and Kusunold, “ Excess Heat Generation, Voltage Deviation, and Neutron Emission in D20-LiOD systems”, Proc. Fourth International Conference on Cold Fusion, Lahaina, Maui, Dec. 6-9, EPRI TR-104188 vol 1, page 3 Mckubre,M. C. H, et al, “Loading, Calorimetric and Nuclear investigation of the D/Pd Systems”, Proc. Fourth Intemati., Conf., on Cold Fusion, Lahaina, Maui, Dec6, 1993, EPRI TR- 1 04188 vol 1, page 5. Patterson, J. US Patents “System for Electrolysis” US 5,494,559, 1996, see also 9 4,943,355 5,036,031, 5,318675, and 5,372,688.