rexresearch.com
Randy CURRY
Atmospheric Plasma Generator
Atmospheric Plasma Generator
Timothy Wall ( Missouri University News ) : Plasma Device Developed at MU Could Revolutionize Energy Generation and Storage
US2013057151 // WO2012173864 // SYSTEMS AND METHODS TO GENERATE A SELF-CONFINED HIGH DENSITY AIR PLASMA [ PDF ]
US2013062314 // DIELECTRIC LOADED FLUIDS FOR HIGH VOLTAGE SWITCHING [ PDF ]
US2007197051 // WO2012145122 // HIGH DIELECTRIC CONSTANT COMPOSITE MATERIALS AND METHODS OF MANUFACTURE [ PDF ]
US7390984 // WO2005038838 // High power liquid dielectric switch [ PDF ]
Research paper: doi: 10.1109/PLASMA.2012.6383564 – “Investigation of a toroidal air plasma under atmospheric conditions”
http://www.extremetech.com/extreme/153630-open-air-plasma-device-could-revolutionize-energy-generation-us-navys-weaponry
April 17, 2013
Sebastian Anthony : "Open-air
plasma device could revolutionize energy generation, US
Navy’s weaponry"
http://www.youtube.com/watch?v=6BfU-wMwL2U&feature=player_embedded
April 15, 2013
http://munews.missouri.edu/news-releases/2013/0415-plasma-device-developed-at-mu-could-revolutionize-energy-generation-and-storage/
MU News Bureau -- University of Missouri
University of Missouri researcher Randy Curry and his team have developed a method of creating and controlling plasma that could revolutionize American energy generation and storage.
Plasma Device Developed at MU Could
Revolutionize Energy Generation and Storage
Story Contact(s): Timothy Wall, walltj@missouri.edu, 573-882-3346
COLUMBIA, Mo. — University of Missouri engineer Randy Curry and his team have developed a method of creating and controlling plasma that could revolutionize American energy generation and storage. Besides liquid, gas and solid, matter has a fourth state, known as plasma. Fire and lightning are familiar forms of plasma. Life on Earth depends on the energy emitted by plasma produced during fusion reactions within the sun. However, Curry warns that without federal funding of basic research, America will lose the race to develop new plasma energy technologies. The basic research program was originally funded by the Office of Naval Research, but continued research has been funded by MU.
Curry’s device launches a ring of plasma as far as two feet. The plasma doesn’t emit radiation, and it is completely safe for humans to be in the same room with it, although the plasma reaches a temperature hotter than the surface of the sun. The secret to Curry’s success was developing a way to make the plasma form its own self-magnetic field, which holds it together while it travels through the air.
“Launching plasma in open air is the ‘Holy Grail’ in the field of physics,” said Curry, professor of electrical and computer engineering in the University of Missouri’s College of Engineering. “Creating plasma in a vacuum tube surrounded by powerful electromagnets is no big deal; dozens of labs can do that. Our innovation allows the plasma to hold itself together while it travels through regular air without any need for containment.”
The plasma device at MU could be enlarged to handle much larger amounts of energy, according to Curry. With sufficient funding, they could develop a system within three to five years that would also be considerably smaller. He noted that they used old technologies to build the current prototype of the plasma-generating machine. Using newer, miniaturized parts, he suggests they could shrink the device to the size of a bread box.
“We have a world-class team at MU’s Center for Physical & Power Electronics, but that team will evaporate without funding,” Curry said. “Department of Defense funding for basic research led to our plasma innovation. The sequester’s funding cuts threaten America’s ability to compete in the future of energy technology. Not only will research not be advanced, a new generation of Americans won’t be trained to take the reins of American engineering leadership.”
Curry is the Logan Distinguished Professor of Electrical & Computer Engineering and Director of the Center for Physical & Power Electronics.
WO2012173864 [ PDF ]
US2013057151
SYSTEMS AND METHODS TO GENERATE A SELF-CONFINED HIGH DENSITY AIR PLASMA
Inventor(s): CURRY RANDY D
Abstract -- This disclosure relates to methods and devices for generating electron dense air plasmas at atmospheric pressures. In particular, this disclosure relate to self-contained toroidal air plasmas. Methods and apparatuses have been developed for generating atmospheric toroidal air plasmas. The air plasmas are self-confining, can be projected, and do not require additional support equipment once formed.
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application claims priority to U.S. Provisional Application No. 61/498,281 , entitled "Systems and Methods to Generate a High Density Air Plasma," filed on June 17, 201 1 , which is incorporated by reference in its entirety.
FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
[0002] This invention was made with government support under grant number N00014-08-1 -0266 by Office of Naval Research (Agency). The government has certain rights in the invention.
FIELD OF THE INVENTION
[0003] The present invention relates to a method and apparatus for generating self-sustaining air plasmas at atmospheric pressures.
BACKGROUND OF THE INVENTION
[0004] An air plasma is an electrically conductive state of matter composed of ions, electrons, radicals, and other neutral species formed at atmospheric pressure that exist in an independent state. Air plasmas may be used in a variety of applications, such as nonlethal weapons, fusion, plasma processing, propulsion, disinfection applications, and Shockwave mitigation.
[0005] However, current plasma sources have been unable to generate an air plasma with an electron density sufficient to protect against the consequences of the overpressure caused by a Shockwave at atmospheric pressure. Furthermore, current plasma sources have been unable to generate self-containing or self-confining air plasmas that have lengthy lifetimes without the use of expensive and unwieldy support equipment or large magnets. Therefore, there remains a need for a versatile, scalable, and repeatable method and apparatus to generate air plasmas.
SUMMARY OF THE INVENTION
[0006] The present invention relates to a method and an apparatus for generating self-confined and self-stabilized air plasmas at atmospheric pressures. In particular, the method and apparatus generate toroidal air plasmas (TAPs) at atmospheric pressure having an electron density sufficient for a number of applications. The method and apparatus may be configured to generate TAPs at a high repetition rate.
[0007] The method includes generating a self-contained air plasma at an atmospheric pressure. The air plasma is generated in a first ignition region and restricted in radial expansion. The method also includes applying a high voltage pulse to the air plasma in a secondary ignition region to heat and accelerate the air plasma away from the second ignition region. Heating the air plasma causes the air plasma to expand and become self-contained.
[0008] The apparatus for generating a self-contained air plasma at an atmospheric pressure includes a primary ignition region that includes a first shielding material defining a first cavity, that may be elongated or another configuration, to contain a plasma source. The apparatus also includes an ignition device to generate the air plasma from the plasma source and a secondary ignition region that includes a second shielding material defining a second region, that may be elongated or another configuration, wherein the second cavity is in fluid communication with the first cavity to receive the air plasma. In one embodiment, the second region is defined, at least in part, by a wire mesh that allows a current to be discharged through the air therein and form a plasma discharge.
[0009] The apparatus includes a high voltage circuit that includes at least one capacitor and is in communication with a voltage source in order to apply a high voltage pulse to the air plasma. The high voltage pulse heats and accelerates the air plasma away from the apparatus to form the self-contained air plasma at the atmospheric pressure. In various other embodiments, the plasma source is at least one member of a group consisting of an exploding wire, an explosive, a puffed gas plasma, a hollow cathode plasma, a hypervelocity plasma source, a railgun, a microwave-driven plasma source, or other compact plasma source that can be directed into the second region. The plasma source may also be provided by a one or more laser-induced plasma channels.
[0010] In another embodiment, a method for generating a self- contained air plasma at an atmospheric pressure includes applying a first high voltage pulse across a wire to explode the wire and generate the air plasma in a first ignition region located between an anode and a cathode. The method also includes restricting radial expansion of the air plasma, such that the air plasma travels parallel to a longitudinal axis of the wire to a second ignition region between the cathode and an accelerator electrode. A second high voltage pulse is applied across the cathode and the accelerator electrode to heat the air plasma, wherein heating the air plasma causes the air plasma to expand, accelerate, and form a toroidal structure. The method also includes discharging the self-contained toroidal air plasma from the second ignition region at the atmospheric pressure.
[0011] The method further includes providing rigid electrically insulating materials between the anode and the cathode, as well as between the cathode and the accelerator electrode. The insulating materials define cavities, which may be elongated. The elongated cavity between the anode and the cathode receives the wire and restricts the radial expansion of the air plasma. The cavity between the cathode and the accelerator electrode allows the air plasma to expand. Both cavities may have generally cylindrical or spiral configurations. The cavities may have equal or different diameters and may be configured to increase or decrease the diameter of the toroidal plasma. In addition, the cavities may be configured to increase or decrease the velocity of the toroidal plasma.
[0012] In another embodiment, a method for generating a self- contained air plasma at an atmospheric pressure includes generating the air plasma in a first ignition region, directing a velocity of expansion of the air plasma out of the first region, and imparting energy to the air plasma in a secondary ignition region, wherein the imparted energy causes the air plasma to expand, accelerate out of the second ignition region, and become self-contained. Alternately, the method may include restricting radial expansion of the air plasma.
[0013] In various embodiments, the wire has a gauge in the range between 00 AWG and 80 AWG. In other embodiments, the first high voltage pulse is between 10kV and 50kV and has a duration between 0.1 s and 200 ms, while the second high voltage pulse is between 100V and 300V or up to many thousands of volts and has a duration between 1 ns and 1000 ms.
[0014] In another embodiment, an apparatus for generating a self- contained air plasma at an atmospheric pressure includes a first shielding material positioned between an anode and a semi-permeable cathode in a primary ignition region. The first shielding material has a first longitudinal cavity to contain a conductive wire extended between and in communication with the anode and the cathode. The apparatus also includes a primary high voltage circuit with at least one voltage source and at least one capacitor. The primary high voltage circuit is in communication with the anode and the cathode to apply a first high voltage pulse across the wire causing it to explode and generate the air plasma. The first longitudinal cavity restricting radial expansion of the air plasma.
[0015] The apparatus also includes a secondary ignition region defined by a second shielding material positioned between the cathode and a semi-permeable electrode. The second shielding material has a second longitudinal cavity extending between the cathode and the electrode wherein the second longitudinal cavity is in fluid communication with the first longitudinal cavity to receive the air plasma. The apparatus also includes a secondary high voltage circuit with at least one other capacitor that is in communication the voltage source. The secondary high voltage circuit further communicates with the cathode and the electrode to apply a second high voltage pulse across the gap between the cathode and the electrode, wherein the second high voltage pulse further heats and accelerates the air plasma as it traverses the electrode to form the self-contained air plasma at the atmospheric pressure.
[0016] In various embodiments, the self-contained air plasma may be formed by a laser induced plasma and subsequently heated by a laser, a microwave pulse, or any means for imparting energy. The plasma formed in air is self-confined by electrostatic or electromagnetic fields and interactions. As such, the air plasma inherently has a long lifetime. The self-confined air plasma may have a lifetime on the order of milliseconds to multiple seconds or even minutes.
[0017] The density of the plasma may be increased by using a pressurization system that may increase the pressure in the apparatus to a range between 1 ATM- 2000 ATM or higher. In addition, the air within and/or around the apparatus may be modified to optimize the size and electron density of the generated air plasmas. For example, the air within and/or around the apparatus may include one or more gas mixtures or gases seeded with nanoparticles or various chemical compounds.
[0018] In various embodiments, the self-contained air plasmas have an electron density of at least 10<10>/cm<3> and may be as high as 10<19>/cm<3>. In addition, the geometry of the apparatus leads the air plasma to form a toroidal structure. DESCRIPTION OF FIGURES
[0019] FIG. 1 depicts an embodiment of a toroidal air plasma generator.
[0020] FIG. 2 is a photograph of one embodiment of the air plasma generation apparatus.
[0021] FIG. 3 is a side-view photograph of one embodiment of the air plasma generation apparatus
[0022] FIG. 4. is a schematic layout of a primary high-voltage circuit according to one embodiment.
[0023] FIG. 5 is a high-speed image of a toroidal air plasma according to one embodiment.
[0024] FIGS. 6A and 6B are photographs providing a cross sectional view of the formation of a toroidal air plasma according to one embodiment.
[0025] FIG. 7 is a flowchart depicting a method to form a toroidal air plasma according to one embodiment.
DETAILED DESCRIPTION OF THE INVENTION
[0026] The present invention relates to the generation of high- density air plasmas at atmospheric pressure that are sustainable for a sufficient duration and have an electron density sufficient to be used in a variety of applications. As used herein, an air plasma at atmospheric pressure refers to an air plasma having pressures substantially equal to the surrounding atmosphere. In addition, air plasmas at atmospheric pressure do not require specialized high- pressure or low-pressure vessels. In one aspect, the geometry of the air plasma generating apparatus gives rise to the shape and the self-containing nature of the air plasma. Once formed, the air plasmas are self-containing and do not require additional support equipment. For example, the air plasma generator may be configured to generate a toroidal air plasma (TAP). A TAP is an air plasma having a substantially toroidal shape.
[0027] For example, the generated air plasmas may be used for shock wave mitigation, used as fusion sources for Tritium-Tritium or Deuterium- Tritium reactions or any other advanced fusion cycle, or plasma capacitors. In addition, the generated air plasmas may be used in nonlethal applications, including but not limited to electroshock weapons, such as a Taser. The air plasmas may also be used for a number of industrial applications, including but not limited to: plasma surface modification including semiconductor processing, polymer modification, directed energy applications, microwave generation, energy storage and generation, UV generation for semiconductor manufacturing, plasma chaff, surface disinfection, and microwave channeling at a distance. The air plasmas may also be used as an ignition source for turbines, combustion engines, and rocket engines. The generated plasmas may also be used in other applications, for example, the generated air plasmas may be precursors to ball lightning.
The Air Plasma Generator Apparatus
[0028] An embodiment of an air plasma generation apparatus 100 that generates a toroidal air plasma (TAP) is shown in FIGS. 1 -3. The apparatus 100 includes an TAP generator 102 that is in electrical communication with a primary high-voltage circuit 104 and a secondary circuit 106.
[0029] The TAP generator 102 is capable of generating a TAP discharge, generally indicated as 130, that has a finite duration. According to one embodiment, the TAP generator 102 uses an exploding wire 108 to form the TAP discharge 130.
[0030] As shown, the exploding wire 108 may be formed of a single strand of wire positioned within the TAP generator 102. Alternately, the exploding wire 108 may consist of a single stand of wire that is woven or looped back and forth within the TAP generator 102, such that multiple lengths of the wire may be exploded simultaneously. In various other embodiments, the exploding wire 108 may consist of multiple stands of distinct or looped wires.
[0031] By way of example and not limitation, the exploding wire 106 may be a 40-gauge copper wire; however, any suitable wire that heats and vaporizes in air may be used. In other examples, the exploding wire 108 may be any gauge of wire ranging from 00 AWG to 80 AWG. In addition, the exploding wire 108 may be a solid wire, a plated wire, a wire that is doped with other materials, or a wire-clad in another material. The exploding wire 108 is suspended between an anode 1 10 and a cathode 1 12. To ignite the exploding wire 108, a high voltage current is applied across the anode 1 10 and a cathode 1 12 and through the wire 108. In various embodiments, the high voltage current superheats at least a portion of the exploding wire 108, thereby causing it to expand explosively.
[0032] The anode 1 10 and the cathode 1 12 define a primary ignition region 1 14 in which the exploding wire 108 is ignited. The primary ignition region 1 14 also includes a non-conductive primary shielding material 1 16 that fills a portion of the space between the anode 1 10 and the cathode 1 12. The primary shielding material 1 16 has a thickness equal to the spacing between the anode 1 10 and the cathode 1 12. In one example, the primary shielding material 1 16 may have a thickness between 5 cm and 20 cm; however, other thickness and spacing distances may be used. In one embodiment, the primary shielding material 1 16 defines an primary elongated cavity 1 18 that receives the exploding wire 108. The diameter of the elongated cavity is larger than the diameter of the exploding wire, such that the exploding wire 108 does not contact the primary shielding material 1 16, thereby allowing the exploding wire 1 14 to ignite in air at atmospheric pressure. The primary elongated cavity 1 18 restricts the radial expansion of air, as indicated by 120, within the elongated cavity following the explosion from the exploding wire 108. Restriction the radial expansion 120 of the air, along with the momentum from the explosion directs the velocity of expanding air out of the primary ignition region 1 14.
[0033] The composition of the exploding wire 108 may also contribute to the formation of the air plasma. By way of example and not limitation, the explosion of the wire 108 generates Shockwaves of electrons, ions, plasmas, UV waves and/or metal particles, as well as a number of other conditions, which may augment the formation of the TAP discharge 130. The exploding wire 108 also generates a pressure pulse that imparts momentum to the gas molecules in a secondary ignition region 122 of the TAP generator 102. Similarly, the exploding wire 108 imparts energy and momentum to the TAP discharge 130 within the secondary ignition region 122.
[0034] In one embodiment, the primary elongated cavity 1 18 is generally cylindrical. In another embodiment, the primary elongated cavity 1 18 has a spiral configuration. Similarly, other configurations of the primary elongated cavity 1 18 may be used; however, in all embodiments, the TAP discharge 130 from the exploding wire 108 is substantially restricted to axial acceleration along the axis of the central axis of the elongated cavity in order to generate boundary conditions that help form and shape the TAP discharge 130 in the secondary ignition region 122.
[0035] The secondary ignition region 122 is defined, in part, by the cathode 1 12 and an accelerator electrode 124. In one embodiment, the cathode 1 12 and the accelerator electrode 124 are a semi-permeable materials, such as but not limited to a mesh or screen, such that the TAP discharge 130 may traverse the cathode and the accelerator electrode. By way of example and not limitation, the accelerator electrode 124 may be composed of stainless steel or any other semi-permeable conductive material.
[0036] The secondary ignition region 122 includes a secondary shielding material 126. The secondary shielding material 126 is non-conductive and may have the same composition as the primary shielding material 1 16. Alternately, the secondary shielding material 126 may have a different
composition than the primary shielding material 1 16.
[0037] In one embodiment, secondary shielding material 126 has a thickness equal to the spacing between the cathode 1 12 and the accelerator electrode 124. In one example, the secondary shielding material 126 has a thickness ranging between approximately 2 mm and 2 cm depending upon the distance between the cathode 1 12 and the accelerator electrode 124; however other thickness and spacing distances may be used. The secondary shielding material 126 also defines a secondary cavity 128 that is axially aligned with the primary elongated cavity 1 18 of the primary shielding material 1 16.
[0038] In one embodiment, the diameter of the secondary cavity 128 is greater than the diameter of the primary elongated cavity 1 18 to allow the TAP discharge 130 to expand as it travels through or, alternately, is formed in and by the secondary ignition region 122. In another embodiment, the diameter of the secondary cavity 128 may be equal to or less than the diameter of the primary elongated cavity 1 18. Similarly, the length of the secondary cavity may be greater than, equal to, or less than the length of the first elongated cavity. In various other embodiments, the secondary ignition region 122 has multiple cavities that, optionally, may be aligned in parallel to one another and the primary elongated cavity 1 18.
[0039] While a single primary ignition region 1 14 and a single secondary ignition region 122 are shown in FIGS. 1 -3, in other embodiments multiple ignition regions may be used to further amplify the effects of the TAP discharge 130. For example, multiple plasma sources may be ignited in multiple primary ignition regions and/or multiple secondary ignition regions may be used to amplify, accelerate, augment, and/or shape the TAP discharge 130.
[0040] In various embodiments, the diameters of the primary and secondary cavities can be formed or otherwise configured to increase or decrease the diameter of the air plasma and to increase or decrease the velocity of the air plasma. The geometry of the self-contained air plasmas may also be enhanced through optimization of the air plasma generation apparatus 100 and the surrounding environment. For example, the TAP generator may be configured to generate stable plasmoids or spheres of plasma similar to ball lightning.
[0041] The TAP generator 102 is electrically connected to a primary high voltage circuit 106 that is configured to deliver a high-voltage pulse to the anode 1 10 and the cathode 1 12. The TAP generator 102 is also electrically connected to a secondary circuit 106 configured to discharge energy through the plasma in the secondary ignition region 122.
[0042] The primary high voltage circuit 106 includes one or more capacitor banks, one or more high voltage power sources, and one or more high- voltage switches, and suitable pulse generating circuitry to deliver a high-voltage pulse across the anode 1 10 and the cathode 1 12. In one embodiment, the primary high voltage circuit 106 includes a capacitor bank energized to between approximately 2 kV and approximately 100 kV to deliver a high voltage pulse having a duration between about 10 ns and 200 ms pulse through the anode 1 10 and the cathode 1 12 to the exploding wire 108. In this embodiment, the anode 1 10 is solid or a semi-permeable conductor while the cathode 1 12 is semipermeable conductor.
[0043] As shown in FIG. 4, a particular embodiment of the primary high voltage circuit 106 is an RLC circuit 400 that includes a number of resistors 402A-C, one or more inductors 404, and one or more capacitors or capacitor banks 406. The primary high voltage circuit 106 also includes as a power source 408, a three-plate pressurized air gap switch 410, a lead 412 connected to the anode 1 10, another lead 414 connected to the cathode 1 12, and additional protection and safety circuitry, including but not limited to switches and diodes, generally indicated as 416. [0044] In one embodiment, the power source 408 is a direct current (DC) power source that supplies approximately 30kV to the primary high voltage circuit 106. The capacitor bank 406 has a capacitance of approximately 1 1 F to store and release approximately 4.4 kJ generate a 6 kA, 46 s current pulse (full-
width half maximum) through the wire 108, causing the wire to explode. The inductor 404 is typically an 1 1 .77 [mu][Eta] air-core inductor. The inductor 404 and a 5.5 [Omega] aqueous-electrolyte shaping resistor 402A are used to shape the current pulse.
[0045] The circuit inductance and resistance are both variable parameters that affect the amount of current and energy delivered to and deposited into the wire 108. To determine the effects of circuit inductance on the current pulse delivered to the wire 108, the air core inductor 404, was replaced in various embodiments with other inductors having inductance values of 0.6 [mu][Eta] and 27.5 [mu][Eta]. Similarly, in other embodiments, the aqueous-electrolyte resistor was replaced with resistors having resistances of approximately 20 [Omega] to approximately 300 [iota][tau][iota][Omega]. Non aqueous-electrolyte resistors may also be used.
[0046] When varying the inductance of the primary high-voltage circuit 104, a shaping resistor 402A with a resistance of approximately 5.2 [Omega] was used. Likewise, the inductor 404 had a resistance of approximately 1 1 .77 [mu][Eta] when the resistance of resistor 402A was varied.
[0047] The current pulse generated by the primary high-voltage circuit 104 with a typical 1 1 .77 [mu][Eta] inductor 404 and a typical 5.2 [Omega] shaping resistor 402A delivers approximately 6 kA with a pulse width of approximately 46.08 [mu][epsilon]. It was observed that the peak and width of the current pulse varied with changes in inductance. For example, when the inductor 404 had an inductance of approximately 27.5 [mu][Eta] the current pulse delivered to the wire 104 had a peak current of approximately 5.48 kA and a pulse width of approximately 53.55 [mu][epsilon]. While the current pulse generated when the inductor 404 had an inductance of 0.6 [mu][Eta] results in higher current (approximately 6.88kA) delivered in a smaller pulse width (approximately 35.9 s). As expected in view of traditional circuit theory, it was observed that the current pulse decreases in amplitude yet spreads in pulse width as the inductance increases. Further, it was observed that varying the inductance of the primary high-voltage circuit 104 did not result in a significant change in the height or duration of the TAP discharge 130. Similarly, no significant effect was observed in the distance traveled data by the TAP discharge 130. As such, the inductance of the primary high-voltage circuit 104 may be varied according to the desired application of the air plasma generation apparatus 100 without diminishing the generated TAPs.
[0048] Conversely, it was determined that varying the resistance in the primary high-voltage circuit 104, did however, affect the generated TAPs. For example, the current pulse from a typical configuration of the primary high- voltage circuit, where the resistance of the shaping resistor 402A is
approximately 5.2 [Omega], is approximately 6 kA with a pulse width of approximately 46.08 s. The current pulse, when the resistor 402A has a resistance of approximately 20 [Omega], however, reaches a peak of only about 2.02 kA with a pulse width of approximately 130.85 s.
[0049] Further, by removing the typical aqueous-electrolyte resistor 402A from the circuit and directly connecting the inductor 404 to the anode 1 10, through lead 412, resulted in a stray resistance of approximately 300 [iota][tau][iota][Omega]. In this configuration, the primary high-voltage circuit 104 is underdamped, rather than the typical overdamped configuration. As such, the resultant current oscillates about four times in approximately 288 s while reaching a peak of approximately 23.6 kA.
[0050] Changing the resistance of the resistor 402A yields appreciable differences in the size and the duration of the TAP discharge 130. For example, when the resistor 402A has a resistance of approximately 20 [Omega] the TAP discharge 130 has a shorter duration and smaller diameter when compared to a shaping resistance of approximately 5.2 [Omega]. Further, when the resistor 402A is removed or otherwise reduced to yield a resistance of approximately 300 [eta][tau][iota][Omega], the TAP discharge 130 is approximately twice as large in diameter and has a longer duration when compared to TAP discharges with a 5.2 [Omega] resistor. In additionally, the TAP discharge 130 generated with a 300 [iota][tau][iota][Omega] resistor for the shaping resistor 402A travels approximately twice as far as the TAP discharges generated using a 20 [Omega] resistor or a 5.2 [Omega] resistor for the shaping resistor. In this configuration, additional energy has been deposited into the TAP discharge 130 formed by the exploding wire 108. This results in an increase in the volume and duration of the TAP discharge 130 and may be caused, at least in part by the reduction in dampening of the primary high-voltage circuit 104.
[0051] Preferably, the secondary circuit 106 includes a capacitor bank charged to a voltage suitable for heating the TAP discharge 130. For example, when the secondary high voltage circuit 106 is charged to between 100V and 300V, the TAP discharge 130 entering the secondary ignition region 122 completes a circuit between the cathode 1 12 and the accelerator electrode 124. The energy imparted by the secondary high voltage circuit 106 enhances the duration and velocity of the TAP discharge 130. In one embodiment, the secondary high voltage circuit 106 is connected to the same high voltage power source as the primary high voltage circuit 106. In another embodiment, the secondary high voltage circuit 106 is powered by another high voltage source. In yet another embodiment, the primary high voltage circuit 106 and the secondary high voltage circuit 106 may be incorporated into a single high voltage system.
[0052] By way of example and not limitation, the secondary circuit 106 may include a secondary 8.8 mF electrolytic capacitor bank 132 that is charged to approximately 250 V to heat the plasma in the secondary ignition region 122. The post-explosion heating has been shown to enhance both the size and duration of the TAP discharge 130.
[0053] The additional heating provided by the secondary circuit 106 also plays a role in forming the toroidal shape of the TAP discharge 130. For example, the elongated cavity 128 defined by the secondary shielding material 126 allows for the plasma generated by the explosion of the wire 104 to expand. During expansion, when the area between the cathode 1 12 and the accelerator anode 124 is filled with plasma, the secondary capacitor bank 132 discharges stored energy through the plasma. In one embodiment, a 400 A current drawn by the plasma from the secondary capacitor bank 132 has a pulse width of approximately 4 ms. After the discharge from the secondary capacitor bank 132, the bulk of the TAP discharge 130 detaches from a portion 134 of the discharge that remains in the secondary ignition region 122 and exits from the TAP generator 102, as shown in FIG. 5. After the bulk of the TAP discharge 130 has separated from the remaining portion, the capacitor bank 132 may continue to discharge and energize the remaining plasma in the TAP generator 102.
[0054] A cross sectional view of the evolution of the toroidal structure 500 of the TAP discharge 130 is shown in FIG. 5. For approximately the first millisecond after ignition, the discharge 130 is still expanding from the secondary ignition region 122 and has a very homogeneous profile.
Approximately 1 .5 ms after ignition, the toroidal shape begins to form. These two images illustrate the toroidal shape of the discharge at 6 ms and 7 ms after ignition. FIG. 5 also shows the remaining discharge 134 within the secondary ignition region 122.
[0055] In one embodiment, the TAP discharge 130 can last up to 15 ms while travelling approximately 30 cm from the TAP generator 102. In other embodiments, the TAP discharge 130 may have a lifetime in the range of milliseconds to multiple seconds and multiple minutes. The toroidal structure 400 of the TAP discharge 103 may expand to approximately 12 cm in diameter. In other embodiments, the toroidal structure 400 may expand to other diameters including those less than or greater than 12 cm. The electron density of the TAP is preferably at least 10<10>/cm<3> and may be as high as 10<19>/cm<3>. In various embodiments, the electron density is determined to be approximately 10<14>-10<15>/cm<3> based upon the measured current passing through the plasma while it is in the secondary ignition region 122.
[0056] The density of the plasma may be increased by using a pressurization system (not shown) that may increase the pressure in the apparatus to a range between 1 ATM- 2000 ATM or higher. In addition, the air within and/or around the apparatus may be modified to optimize the size and electron density of the generated air plasmas. For example, the air within and/or around the apparatus may include one or more gas mixtures or gases seeded with nanoparticles or various chemical compounds.
[0057] In various embodiments, the radial expansion 120 of the shock wave and heat generated by the explosion of the wire 108 is confined within the primary and secondary cavities 1 18 and 128, respectively. The discharge from the exploding wire 108 is thus dissipated, predominantly, through axial expansion along the axis of the primary elongated cavity 1 18 and the secondary cavity 128. This imparts hydrodynamic effects upon the TAP discharge 130 and therefore, the geometry of the TAP generator 102 lends itself to the self-containing characteristics of the TAP discharge 130.
[0058] The combined effects of the initial axial expansion from the exploding wire 108 and the secondary excitation in the secondary ignition region 122 result in the formation of the toroidal structure 400. In various other embodiments, the secondary ignition region 122 may have any geometry that can transfer energy into the TAP discharge 130. In these embodiments, the temperature and subsequent absorption and emission of light by the TAP discharge 130 can be tailored to specific requirements based upon the geometry of the secondary ignition region 122. The duration and amount of energy delivered to the plasma in the secondary ignition region 122 can be optimized to generate characteristics of the TAP discharge 130 that are required for the desired application. For example, by increasing the energy imparted to the TAP discharge in the secondary ignition region 122, the lifetime of the TAP discharge may be extended from milliseconds to minutes thereby allowing the long-range projection of the plasma.
[0059] Although the TAP generator 102 has been described using the exploding wire 108 as the initial plasma source, other plasma sources may be used. By way of example and not limitation, other plasma sources include explosives, puffed gas plasmas, hollow cathode plasmas, microwave driven sources, high power laser arrays, railguns, hypervelocity plasma accelerators, and any other plasma source that has a high repetition rate to generate ionized particles. In these embodiments, the plasma source is activated by a suitable activation device corresponding to the plasma course. For example, an activation device for an explosive is a detonator, while an activation device for a microwave driven source is a microwave generator.
[0060] In another example, one or more lasers is used to form or further heat the TAP discharge 130. For example, a laser may be used to form a laser-induced air plasma in the primary ignition region 1 14. Alternately, a laser may be used to heat a plasma discharge within the secondary ignition region 122.
[0061] In various embodiments, the air plasma generation apparatus 100 is configured for single or multi-shot operation. As such, the air plasma generation apparatus 100 may generate a single or multiple self- contained air plasmas at a high rate of repetition.
The Toroidal Air Plasma
[0062] The TAP discharge 130 has a very homogenous profile immediately after the ignition of the exploding wire 108 as it expands from the first primary elongated cavity 1 18. In one embodiment, the TAP discharge 130 begins to take on the toroidal structure 400 approximately 1 .5 ms after ignition. The toroidal structure 400 of the TAP discharge 200 is shown at approximately 6 ms and approximately 7 ms after ignition in FIGS. 5A and 5B, respectively. FIGS. 6A and 6B also show the secondary ignition 600 of the TAP discharge 130 within the secondary ignition region 122. When the TAP discharge 130 exits the TAP generator 102, the discharge has a circulating current or field reversal that generates a self-magnetic field as well as a rotating plasma region on the minor radius of the toroid structure 400. The self-magnetic field confines the TAP discharge 130 and significantly increases the lifetime of the TAP discharge to effectively produce a self-sustaining TAP discharge by reducing interactions that may recombine molecules of the air plasma with atmospheric gas molecules.
[0063] In various embodiments, the TAP discharge can be sustained for approximately 2-30 ms and may travel approximately 10-40 cm away from the TAP generator 102 at up to 200 m/s. The toroidal shape 500 may expand up to approximately 12 cm in diameter. The electron density of the TAP discharge 130 is approximately 10<14>-10<15>/cm<3> as determined by the measured current passing through the TAP discharge 130 during the secondary heating of the discharge in the secondary ignition region 122. In various other embodiments, the TAP discharge 130 is scalable to higher energies, densities and can be used for a number of advanced applications.
[0064] For example, 1 kilojoule to 1 gigajoule or higher of energy may be imparted to the TAP discharge 130 in the secondary ignition region 122. Increasing the energy will increase the lifetime of the TAP discharge 130 from an order of milliseconds to minutes allowing for the long-range projection of the TAP discharge.
[0065] FIG. 7 is a flowchart illustrating one embodiment of a method 700 for generating a TAP discharge 130. At step 702, a first high voltage pulse is applied across the anode 1 10, the cathode 1 12, and the exploding wire 108 in the primary ignition region 1 14. The first high voltage causes the wire to explode thereby producing the TAP discharge 130. At step 704, the radial expansion of the AP discharge is restricted such that the TAP discharge travels along the longitudinal axis of the wire to a second ignition region defined by the cathode 1 12 and the accelerator electrode 124. [0066] In the second ignition region 122, a second high voltage pulse is applied across the cathode 1 12 and the accelerator electrode 124 to further heat and expand the TAP discharge 130, at step 706. Within the secondary ignition region 122, the TAP discharge becomes self-sustaining and takes on the toroid structure 200. At step 708, the self-contained TAP discharge is discharged from the second ignition region 122, wherein it may be used to mitigate the effects of a shock wave or another propagating wave.
Example Method for Generating a Toroidal Air Plasma
[0067] By way of example and not limitation, an exemplary method for generating a TAP discharge, such as the discharge 130 is provided. The primary high voltage circuit 104 of the air plasma generation apparatus 100 included an 1 1 F capacitor bank energized to approximately 30kV to deliver a 4 kA pulse for a duration of approximately 200 s pulse through two strands of 40 AWG silver-plated copper wire 108 within the TAP generator 102. The anode 1 10 connected to the wire 108 was a copper screen while the cathode 1 12 was a stainless steel screen. The primary shielding material 1 16 was a polycarbonate material having a thickness of approximately 10 cm and the elongated cavity 1 18 had a diameter of approximately 1 .25 cm.
[0068] The secondary circuit 106 used an 8.8 mF electrolytic capacitor bank 132 charged to 250V to heat the TAP discharge 130. The secondary primary shielding material 126 was plastic approximately 7 mm thick and defined another elongated cavity 128 with a diameter of approximately 3 cm. The secondary circuit 106 discharged approximately 400A into the TAP discharge 130 over approximately 4 ms. The TAP discharge 130 exiting the TAP generator 102 has an electron density of approximately 10<16>-10<17>/cm<3> as determined by the measured current that passed through the discharge during the secondary heating. [0069] It will be appreciated that the device and method of the present invention are capable of being incorporated in the form of a variety of embodiments, only a few of which have been illustrated and described above. The invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive and the scope of the invention is, therefore indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.
US2013062314 [ PDF ]
DIELECTRIC LOADED FLUIDS FOR HIGH VOLTAGE SWITCHING
Inventor: CURRY RANDY D // YECKEL CHRISTOPHER [US] (+1)
Applicant: UNIV MISSOURI [US]
Abstract -- This disclosure relates to methods and systems to reduce high voltage breakdown jitters in liquid dielectric switches. In particular, dielectric liquids have been produced that contain a suspension of nanoparticles and a surfactant to reduce the breakdown jitter. In one embodiment, the suspended nanoparticles are Barium Strontium Titanate (BST) nanoparticles.
FIELD OF THE INVENTION
[0001] The invention generally relates to a dielectric fluid that can be used to reduce the voltage breakdown jitter in high voltage switches and spark gaps. The invention further relates to methods of reducing the voltage breakdown jitter in high voltage switches and spark gaps, various methods of preparing the dielectric fluid, and high voltage switches incorporating the dielectric fluid. Generally, the dielectric fluid is pressurized, filtered and may include a nanoparticle suspension.
BACKGROUND OF THE INVENTION
[0002] The design and characterization of dielectrics is critical for optimum high-voltage switch performance. Previous attempts to optimize switch performance, such as the tuning of the oil pressure and flow rates for oil dielectrics, have dramatically reduced the rep-rate self-break jitter by eliminating breakdown byproducts.
[0003] During self-break operation, however, switches using oil dielectrics may still demonstrate erratic breakdown patterns. Some switches have shown percentages of standard deviation from the mean breakdown as high as 20%. This is problematic for electrical loads that require uniform pulse repetition. Therefore, there remains a need to further reduce the voltage breakdown jitter in high voltage switches and spark gaps.
SUMMARY OF THE INVENTION
[0004] The present disclosure relates to systems and methods for reducing the voltage breakdown jitter in high voltage switches and other applications. In one embodiment, a method for preparing a dielectric fluid to reduce voltage breakdown jitter in a high voltage spark gap includes providing a dielectric fluid, sparging the dielectric fluid with dry nitrogen, and adding a plurality of nanoparticles to the dielectric fluid. The method further includes adding a surfactant to the dielectric fluid, sonicating the dielectric fluid, filtering the dielectric fluid, and pressurizing the dielectric fluid.
[0005] In various embodiments, the dielectric fluid is a polyolefin having a chemical formula of C6H32, such as 1-Hexadecane or a hydrocarbon-based coolant fluid, such as NYCODIEL. The method may further include sparging the dielectric fluid with dry nitrogen to reduce the water content of a dielectric oil to less than approximately 20 ppm or less.
[0006] In one embodiment, the plurality of nanoparticles have a dielectric constant ranging from 20-6000 and may be composed of Barium Strontium Titanate (BST) nanoparticles. The nanoparticles range in diameter from about 50 nm to about 250 nm and may be added to obtain a concentration in the dielectric fluid ranging between 0.1% and 10% by weight. In various embodiments, the ratio of the dielectric constant of the plurality of nanoparticles to the dielectric constant of the dielectric fluid is at least 3:1, 10:1, 2000:1, or greater.
[0007] In another embodiment, the surfactant is added to obtain a concentration in the dielectric fluid ranging between 0.1% and 10% by weight. In one embodiment of the method, the dielectric fluid is pressurized to between about 10 psig and 2,500 psig. In another embodiment, the dielectric fluid is pressurized to approximately atmospheric pressure.
[0008] In another embodiment, a method for reducing voltage breakdown jitter in a high voltage switch includes providing two or more electrodes separated by a gap and providing a pressurized dielectric suspension between the electrodes. The dielectric suspension includes a dielectric fluid, a surfactant to reduce a breakdown voltage of the dielectric fluid and remove carbon from at least one of the two or more electrodes, and a plurality of nanoparticles to enhance the formation of streamers thereby reducing an electrode gap.
[0009] The method also includes applying a voltage across the two or more electrodes to form an electric field between the two or more electrodes. The reduction of the breakdown voltage and the formation of streamers enhance the electric field between the two or more electrodes to reduce the voltage breakdown jitter in the gap. In one embodiment, the voltage breakdown jitter is decreased by 10% or greater. In another embodiment, the voltage breakdown jitter is decreased by a factor of 2 or greater. For example, the jitter may be decreased by a factor ranging between 2 and 3. In yet another embodiment, the voltage breakdown jitter is decreased by a factor of 10 or greater. In other embodiments, the gap is between 5 [mu]m and 2000 [mu]m, while the applied voltage ranges from about 2 kV to about 10 MV.
[0010] In one embodiment, a dielectric to reduce jitter in a high voltage spark gap between an anode and a cathode includes a pressurized hydrocarbon-based fluid having a first dielectric constant and a plurality of nanoparticles having a second dielectric constant, where the ratio of the second dielectric constant to the first dielectric constant is approximately 3 to 1. The dielectric also includes a surfactant to suspend the plurality the nanoparticles in the fluid, to reduce the voltage breakdown of the fluid, and to remove carbon particles from the surface of an electrode. The high voltage spark gap may be a component of a high voltage switch. In one embodiment, the high voltage switch is selected from a group consisting of a laser-triggered switch, an electrically triggered trigatron, or an electric substation switch. In various embodiments, the shape of the plurality of nanoparticles may be tailored to induce specific breakdown characteristics.
[0011] In another embodiment, an electric switch system includes at least two electrodes separated by a gap and a dielectric liquid including a surfactant and a suspension of nanoparticles. The switch system may also include a high-pressure pump and a filter system. The liquid may be continuously pumped through the gap.
DESCRIPTION OF FIGURES
[0012] FIG. 1 depicts an exemplary high-voltage switch geometry.
[0013] FIG. 2 depicts an exemplary alignment of dipoles in a dielectric according to one embodiment.
[0014] FIG. 3 depicts a three-dimensional model of an electrode system according to one embodiment.
[0015] FIG. 4 depicts a two-dimensional cross-sectional view of the three-dimensional model of FIG. 3.
[0016] FIG. 5 depicts an embodiment of an electric switch system according to one embodiment.
[0017] FIG. 6 depicts a circuit schematic of an embodiment of an electric switch according to one embodiment.
[0018] FIG. 7 is a graph depicting the effect of a single particle's distance from the cathode on the average cathode electric field according to one embodiment.
[0019] FIG. 8 is a graph depicting the average breakdown electric field in a dielectric liquid with an increasing concentration of nanoparticles according to one embodiment.
[0020] FIG. 9 is a flowchart depicting a method for reducing voltage breakdown jitter in a high voltage switch system according to one embodiment.
[0021] FIG. 10 is a flowchart depicting a method for preparing a dielectric fluid to reduce voltage breakdown jitter in a high voltage spark gap according to one embodiment.
DETAILED DESCRIPTION OF THE INVENTION
[0022] The present disclosure relates to a dielectric fluid that reduces the voltage breakdown jitter in high voltage switches and spark gaps. In various embodiments, the dielectric fluid can be used in flowing or pressurized nonflowing spark gaps. In one aspect, the pressurized fluid reduces the size of the vapor bubble resulting from the arc in the liquid. An isostatic pressure is applied reducing the size of the vapor bubble. The isostatic pressure also increases the voltage breakdown jitter of the switch. The pressure inherently reduces microbubbles that result from the formation of electrons emitted from the cathode surface. Typically, the emitted electrons result from high electrical field levels at the surface of the cathode that are caused by electrode pitting and the electrode gap spacing.
[0023] In a high pressure spark gap, a minimum of 2 electrodes are utilized for the switch. By way of example, and not limitation, exemplary high voltage switches are disclosed in U.S. Pat. Nos. 7,312,412 and 7,390,984 to Cravey et al., both of which are hereby incorporated by reference in their entireties. In various embodiments, the switch operates at voltages ranging from approximately 2 kV to approximately 10 MV, with subsequent currents that range from about 100 A to about 5 MA.
[0024] Various testing to identify one or more embodiments of the invention were conducted and presented in "Electrostatic Field Simulation Study of Nanoparticles Suspended in Synthetic Insulating Oil," IEEE Transactions on Plasma Science, Vol. 38 Issue 10, pp. 2514-2519, herein incorporated by reference in its entirety. The dielectric fluid has applications to laser-triggered and electrically-triggered trigatron designs, and other triggered spark gaps that utilize a fluid. Spark gaps that utilize fluids of this type may one day be used in the electrical industry to replace SF6, a gas that is believed to destroy ozone. Other potential applications include use within various components of electric substations, directed energy systems, or pulsed power systems.
[0025] A dielectric fluid 100, according to one embodiment, is shown in FIG. 1 as flowing through a gap 102 between a set of electrodes 104 and 106, (e.g. an anode and a cathode) for an exemplary switch 108. In one embodiment, the dielectric fluid 100 is an oil. Preferably, the dielectric fluid 100 is a polyolefin having a chemical formula of C6H32 or a hydrocarbon-based coolant fluid. In one embodiment, the dielectric fluid 100 is 1-Hexadecene. In another embodiment, the dielectric fluid 100 is NYCODIEL, a coolant fluid produced by Nyco. In various other embodiments, the dielectric fluid 100 is any dielectric liquid, including a poly-alpha-olefin (PAO), which is suitable for use in a spark gap. By way of example and not limitation, the dielectric fluid 100 may be a silicone-based oil or water. In other examples, the dielectric fluid 100 may include one or more oils.
Nanoparticles
[0026] In one embodiment, the dielectric fluid 100 includes a number of particles, such as a nanoparticle 300, as shown in FIGS. 3 and 4, that act as field enhancements. The nanoparticle 300 presents an artificial geometry that allows the electric field to be enhanced in the fluid. The nanoparticle 300 also enhances the formation of streamers in the dielectric fluid 100. The streamers in turn, effectively close the electrode gap 102 to form an arc, when a voltage is applied to the electrodes 104 and 106.
[0027] The nanoparticle 300 may be composed of any material having a high dielectric constant, including but not limited to ceramics. Preferably, the nanoparticle 300 is a Barium-Strontium-Titanate (BST) nanoparticle. In various embodiments, the nanoparticle 300 has a dielectric constant that ranges from 20 to 6000. The dielectric constant of an exemplary BST nanoparticle is approximately 2000. In another embodiment, the nanoparticle 300 has a dielectric constant that is at least 3 times greater than the dielectric constant of the dielectric fluid 100. In a preferred embodiment, the nanoparticle 300 has a dielectric constant that is at least 10 times greater than the dielectric constant of the dielectric fluid 100.
[0028] In one embodiment, the nanoparticles 300 within the dielectric fluid 100 have uniform composition and dielectric constants. In another embodiment, the dielectric fluid 100 may include a distribution of nanoparticles 300 having mixed compositions and/or dielectric constants. In various embodiments, the nanoparticles 300 are added to the dielectric fluid 100 to achieve a final concentration that ranges from approximately 0.1% weight by volume (w/v) to 10% w/v or higher.
[0029] The size of the nanoparticle 300 can be selected to provide for the optimum field enhancement to reduce the voltage breakdown jitter. Preferably, the nanoparticles 300 range in diameter from 50 nm to about 250 nm. The nanoparticles 300 may, however, agglomerate to form clusters that may be several microns in diameter.
[0030] In one embodiment, the nanoparticles 300 are polar molecules that will form dipoles 200 in an electric field 202, as the charges separate with a magnitude proportional to the electric field. As shown in FIG. 2, dipoles 200 will align within the dielectric liquid 100. The dipoles 200 will exhibit a polarization charge at its extreme edges. If the net charge on a nanoparticle 300 is zero, as under direct current conditions, then the electric field 202 will not cause it to drift. The polarization magnitude of the surface charge on the nanoparticle 300 under the influence of an electric field is proportional to its dielectric constant. This relationship is defined by the equation: D=[element of]E, where D is the electric displacement field and [element of] is the dielectric constant ([element of]0[element of]r). D is in units of C/m<2 >and defines the electric flux density associated with the nanoparticle 300. The dielectric constant of a BST nanoparticle 300 is around 2000.
[0031] FIG. 3 depicts a three-dimensional model of an electrode system according to one embodiment, while FIG. 4 depicts a two-dimensional cross-sectional view of the three-dimensional model. The dielectric fluid 100 was modeled as a three-dimensional homogeneous background of material with a constant relative permittivity into which a random distribution of fixed-diameter particles could be introduced. The particles have a different relative permittivity than the background. Conditions such as particle size, particle density, and distance of a single particle from the electrodes were investigated numerically. The simulation suggests that nanoparticle density is a critical parameter, with increased concentrations corresponding to increased average electric field within the simulated dielectric sample.
[0032] The results of the electric field models, shown in FIGS. 3-4 indicate that the nanoparticles 300 introduce a field enhancement effect on the cathode 106 surface and within the bulk of the dielectric fluid 100. In one example, the high dielectric constant associated with a nanoparticle 300 of BST is approximately 2000, and therefore excludes electric fields 202 and concentrates them within the dielectric fluid 100 having a dielectric constant of approximately 2. This produces an electric field enhancement at the BST-fluid interface of approximately 1000. A BST nanoparticle 300 near the cathode 106 surface will increase the local electric field 202 in the fluid 100 and increase the probability of electron emission from the cathode. The particle-generated electric fields 202 in the bulk of the fluid 100 may help guide the breakdown streamer across the gap 102.
[0033] In one embodiment, the particle shape and/or particle morphology may determine the breakdown characteristics of the system. For example, the particles interact with one or more electrode surface field enhancements that are present at the electrode surface. Repeated discharges may cause pitting and other deformations (other field enhancements) on the surface of the electrodes 104 and 106. The deformations increase the probability that a subsequent breakdown will initiate on or near the field enhancement site. Typically, field enhancements on the electrode surface are either hemispherical or conical. Various simulations conducted to study the interactions between the nanoparticles 300 and various field enhancements indicate that an increasing particle concentration increases the chance that a particle is located near the electrode surface, contributing to an average field increase across the surface of the electrode. This effect allows electrons to be emitted at lower switch voltages.
[0034] The results from the simulations also indicate that the interactions between the electrode field enhancements and the high dielectric nanoparticles 300 may contribute to a reduction in jitter. For example, jitter values as low as about 7% have been recorded with nanoparticle suspension and optimized gap-spacing. In one embodiment, the nanoparticles 300 polarize in the applied electric field and disrupt any electric fields at the peaks of the field enhancements while increasing the electric fields at the non-enhanced portions of the electrode surface. Overall, the nanoparticles 300 may function to combine to "smooth" the fields on the electrode surface, thereby reducing jitter in the system. As such, the particular shape and/or morphology of the nanoparticles 300 may be tailored to induce specific breakdown characteristics of the system. For example, the nanoparticles 300 may be spherical, cubic, or irregularly shaped, among others.
[0035] The amount of nanoparticles within the dielectric fluid 100 as well as the nanoparticle size, shape, and/or morphology help determine the reduction in the electric field hold-off value in the fluid switching media. Measured values indicate that the electric field hold-off of the fluid can be reduced by over 50% if required. The surfactant also can be used to reduce the voltage breakdown. All of these factors allow the electrode spacing to be increased thereby reducing the effect of the electrode pitting and inherently reducing the jitter of the voltage across the electrodes.
Surfactant
[0036] In order to suspend the nanoparticles 300 in the dielectric fluid 100, a surfactant is added to the fluid. In a preferred embodiment, the surfactant is a polar additive. By way of example and not limitation, the surfactant may be oleic acid, alkylbenzene-sulfonic acid, phosphate ester, or other surfactants that are suitable for use in a spark gap. In various embodiments, the surfactant is added to the dielectric fluid 100 to achieve an initial concentration that ranges from approximately 0.1% w/v up to about 10% w/v.
[0037] The polarization of the surfactant thereby aids to further reduce the voltage breakdown within the gap 102. In both flowing and non-flowing embodiments of the dielectric fluid 100, the polarization of the surfactant can be used to "tune" or selectively modify the voltage hold-off of the gap 102 without changing the length of the gap. Moreover, in embodiments where the dielectric fluid flows through the gap 102, the surfactant aids in the removal of carbon particles from surfaces of the electrodes 104 and 106, thus further reducing the voltage breakdown jitter. The surfactant also mitigates the agglomeration of the nanoparticles 300.
[0038] In one aspect, the surfactant additive, in some instances, may produce a significant decrease in breakdown jitter. For example, simulations have shown that a dielectric fluid, such as PAO, without additives will decrease jitter to some extent; however, one or more additives may be added to decrease the voltage breakdown.
Electric Switch System
[0039] An electric switch system 500 according to one embodiment is shown in FIG. 5. The electric switch system 500 includes a high-pressure heating system 502, a filtration system 504, and a switch 506.
[0040] The high pressure system includes a pump 508 and a hydraulic translator 510 to pressurize and pump the dielectric fluid 100 within the switch 506. In one embodiment, the pump 508 is a hand pump. In other embodiments, the pump 508 may be an electrically-powered pump.
[0041] The filtration system 504 includes at least one valve 512, a pump 514, and at least one filter 516. In one embodiment, the valve 512 is an isolation valve that can isolate portions of the filtration system 504 from the high-pressure system 502 and the switch 506.
[0042] The pump 514 can be used to pump the dielectric fluid 100 through the switch 506 and through the filter 516. In various embodiments, the pump 514 may pump the dielectric fluid 100 to be filtered before, during and/or after triggering of the switch 506.
[0043] The filter 516 filters the dielectric fluid 100. The filter 516 removes large undesirable nanoparticle agglomerates. In various embodiments, the filter 516 has pore diameters ranging from as large as 5 [mu]m to as little as 200 nm. In another embodiment, a series of filters may be used to filter the dielectric fluid 100.
[0044] The switch 506 is high-voltage switch having spaced electrodes, similar to the electrodes 104 and 106. In one aspect, the switch 506 also includes a trigger mechanism for operating the switch. The trigger may be a trigatron, a laser pulse, a microwave pulse, or a series injection. By way of example and not limitation, the switch 506 may be a high-voltage system capable of generating a 250-kV voltage pulse across electrodes 104-106 and through the dielectric liquid 100 with a rise-time of 1.6 [mu]s.
[0045] In one embodiment, as shown in FIG. 6, the switch 506 includes a Marx-generator circuit 600 that rings a peaking capacitor 602 through a linear inductor 604. The voltage developed on the peaking capacitor 602 is simultaneously applied to the electrodes 104 and 106 of the switch 506. The values shown in FIG. 6 for the various components of the switch 506 are provided for illustration only and do not limit the configuration of the switch.
[0046] By way of example and not limitation, the electrodes may be planar 1-inch diameter stainless steel electrodes that may have smooth or rough surfaces. In one embodiment, the electrode gap 102 is externally adjustable. In a preferred embodiment, the gap 102 ranges from 5 [mu]m to 50,000 [mu]m. In various other embodiments, other gap 102 distances may be used. Typically, the gap 102 is set such that the voltage breaks down at 20-100% of the attainable charge voltage.
[0047] The switch 506 is configured to apply a voltage across the electrodes 104 and 106 that rises at the rate of approximately 150 kV/[mu]s. In one embodiment, the switch 506 is further configured for self-breakdown operation. Further, in operation, the switch 506 is suited for pressures up to and including 2,500 psig or greater.
[0048] In various embodiments, the dielectric fluid 100 may be circulated through and/or within high voltage switches or spark gaps. By way of example and not limitation, the dielectric fluid may be pumped through or otherwise agitated in the gap 102 between the electrodes 104 and 106, thereby causing continuous motion in the fluid in the gap. The fluid 100 may be circulated by a fluid pump or a vibrating device in communication with the fluid.
[0049] FIG. 7 is a graph depicting the effect of a single particle's distance from the cathode on the average cathode electric field strength according to one embodiment. If the nanoparticles 300 are in close proximity to each other or an electrode, the polarization charge can create the nonlinear electric fields shown in FIG. 4. FIG. 7 provides the results of an exemplary simulation showing the change in the average electric field strength corresponding to a position for a nanoparticle. During the simulation, the nanoparticle was modeled as a 100 nm nanoparticle and the electrodes are modeled as flat electrodes having a 16 [mu]m diameter and 8 [mu]m gap of separation.
[0050] Electromagnetic theory suggests that the high dielectric constant of the nanoparticles is a result of their ability to polarize into dipoles in an applied field. This allows a nanoparticle near or on the surface of the cathode to increase the local electric field and act as a field enhancement to draw electrons out of the cathode and into collision with the neutral chains of oil molecules eventually resulting in a breakdown event.
[0051] The simulations also suggest that the polarized nanoparticles 300 may change the electric fields within the bulk of the dielectric fluid 100 to act as a path for an ionized streamer to propagate through. The local nonlinear electric fields generated by the polarized nanoparticles 300 may guide the elongating ionized streamer from the cathode electrode 106 to the anode electrode 104 thereby producing more predictable breakdown events.
[0052] Various experimental tests, conducted without filtering the dielectric fluid 100, have shown a marked decrease in the average breakdown electric field with increasing concentrations of nanoparticles 300. It is thought that this may be due to the large extraction fields generated by the nanoparticles on the surface of the cathode and in the bulk of the oil. FIG. 8 is a graph depicting the average breakdown electric field in a dielectric liquid with an increasing concentration of nanoparticles 300 according to one embodiment, where an inline filter of 5 [mu]m, similar to the filter 516, was used for these tests.
[0053] These measurements were gathered using a switch, similar to the switch 506, where the flat electrode gap spacing was set at 0.18 cm. The flat electrode geometry was used to generate a uniform electric field across the gap and mitigate the effect of field enhancement. During the test cycle 150 shots were taken with 75 shots taken per pressure. The oil utilized was a hydrogenated 1-decene polyalphaolefin.
[0054] The decrease of the electric breakdown strength in oil can be counteracted by a minimal increase in the gap spacing of the switch, so it should not be considered a detrimental effect. It is included in this section to show that the nanoparticles have a realizable effect on the breakdown. The decrease in the breakdown strength is even less significant considering that the tests in FIG. 4 were completed with no pre-filtering of the oil. After pre-filtering, it is predicted that there is less than 5% of the original nanoparticles remain in suspension.
[0055] FIG. 9 depicts a method, indicated generally as 900 for reducing voltage breakdown jitter in a high voltage switch system, such as the switch system 500. At step 902, a switch having two or more electrodes 104-106 separated by a gap 102 is selected. At step 904, a pressurized dielectric suspension is pumped between the electrodes. The dielectric suspension includes the dielectric fluid 100, a surfactant to reduce a breakdown voltage of the dielectric fluid and to remove carbon from at least one of the two or more electrodes 104-106. The dielectric suspension also includes a plurality of nanoparticles, 300 held in suspension due at least in part to the surfactant, to enhance the formation of streamers thereby effectively reducing the electrode gap 102.
[0056] A voltage is applied across the two or more electrodes to form an electric field, at step 906. Within the electric field, the surfactant helps to reduce the breakdown voltage and the nanoparticles 300 encourage the formation of streamers that enhance the electric field between the electrodes 104-106. In one embodiment, the combination of the dielectric liquid 100, surfactant, and nanoparticles 300 work in combination to reduce the voltage breakdown jitter in the gap 102.
[0057] FIG. 10 depicts a method, indicated generally as 1000 for preparing a dielectric fluid to reduce voltage breakdown jitter in a high voltage spark gap. At step 1002, a dielectric oil to serve as the dielectric fluid 100 is prepared. The dielectric oil is sparged with dry nitrogen at step 1004. In one embodiment, the dielectric oil is sparged before being pumped into in the switch. The dielectric oil is sparged to reduce the water concentration to a uniform level of around 20 ppm or less.
[0058] At step 1006, the plurality of nanoparticles 300 are added to the dielectric oil, while at step 1008 the surfactant is added to the dielectric oil. The oil, nanoparticles 300, and surfactant combination is sonicated at step 1010 to break up the nanoparticles and disperses them throughout the oil to maintain an adequate suspension of the nanoparticles in the oil. The oil, nanoparticles 300, and the surfactant are then filtered at step 1012 to sufficiently remove the large nanoparticles agglomerates from the oil. In various embodiments, multiple filtering steps are used. For example, pre-filtering is necessary to sufficiently remove the large nanoparticles agglomerates from the oil so that it can be passed through the filter to remove carbon during a rep-rate operation. In this example, the oil is filtered by a vacuum filtration system that starts with a large pore size filter and uses other small-pore size filters. The oil and suspended additives are then pressurized at step 1014 to reduce the vapor bubble size and/or eliminate microcavities that result from an arc traveling through the oil.
Example
Dielectric Liquid Preparation Method
[0059] By way of example and not limitation, an exemplary method used to prepare the dielectric oil suspension is provided. An initial evaluation of a number of dielectric fluids was conducted to evaluate both the average electric field strength at breakdown over a number of breakdowns. The evaluation also considered the percent standard deviation of the electric field, which was defined as the ratio of the standard deviation to the mean value of the breakdown electric field for a sample of breakdown at approximately 20-80% of the attainable charge voltage. The selected dielectric fluids were military-grade decene-based poly-[alpha]-olefins and synthetic hydrocarbons having a controlled dielectric constant. The selected oils were first pre-filtered through a 0.45-[mu]m filter nitrocellulose filter to remove macro particle contaminants. Next BST particles and a chemical surfactant were added at with concentrations of 5% and 1% by weight relative to the volume of the suspension, respectively. An ultrasonic liquid processor was activated after the addition of the BST particles for a period of time, which then caused the particles to break apart and dispersed them throughout the oil. The oil was then filtered by a vacuum filtration system that had a filter pore size of 5 [mu]m. The pore size was gradually decreased until the nanoparticle oil passed easily through it. In this example, the oil was then passed through pre-filters having pore sizes that range from 200 nm to 5000 nm. In addition, the oil suspension was subjected to inline filtering using filter pore sizes ranging from 200 nm to 5000 nm. Before being placed in the switch system, the oil was sparged with dry nitrogen for 5 to 10 minutes to reduce the water concentration to a uniform level of around 20 ppm or less. The oil and nanoparticle suspension was then placed in the switch and tested. By way of example, and not limitation the simulated gap was approximately 8 [mu]m which corresponded to an experimental gap spacing of approximately 1.8 cm.
Experimental Results
[0060] Table 1 presents experimental results using a variety of dielectric fluid configurations, filter configurations, and the corresponding breakdown jitter results. These tests were performed with an electric switch system in the same configuration as the system of the Electrostatic Field Simulation Study of Nanoparticles Suspended in Synthetic Insulating Oil publication. The first tests were done with hexadecane oil to establish a baseline.
[0061] The most desired results we obtained were from the NYCO oil pre-filtered through the 5-[mu]m filter and filtered through a 5-[mu]m filter during testing. The value of the percent standard deviation (PSD) was 7.61% at 600-psi. The second best PSD of 9.80% was found with the NYCO oil pre-filtered with the 1-[mu]m filter and filtered through a 1-[mu]m filter during testing. After several test cycles at 300 and 600 psi, it was determined that lower PSD values were being found at the higher pressure.
[0062] One of the NYCO oil samples was contaminated with water saturated Diala AX transformer oil and was included in the table to show how increased water content can influence the results. The water was removed through sparging and retested with significantly better results.
[0000]
TABLE I
Breakdown Results of Two Oils with Nanoparticle Suspension
Percent
Standard
Mean Breakdown Deviation Pre-Filter Inline Filter Water
Oil Name Voltage (kV) (PSD) Size Size Clogged (ppm)
300-psi 600-psi 300-psi 600-psi
Hexadecane 177.0 188.4 14.88% 11.03% none 1-um no 20
No Nano
Hexadecane 5% 176.4 186.0 17.28% 14.53% none 5-um yes 22
BST 1% Surfactant*
Hexadecane 5% 163.2 173.8 9.57% 9.93% 1-um 1-um yes 18
BST 1% Surfactant
Nyco MIL-PRF- 176.7 185.5 13.36% 13.55% none 1-um no 22
87252 No Nano
Nyco MIL-PRF- 162.2 180.4 13.30% 7.61% 5-um 5-um no 21
87252 5% BST 1%
Surfactant
Nyco MIL-PRF- 155.2 173.4 13.05% 9.80% 1-um 1-um yes 20
87252 5% BST 1%
Surfactant
600-psi 1000-psi 600-psi 1000-psi
Nyco MIL-PRF- 146.0 164.8 16.97% 14.05% 0.2-um< > .45-um< > yes 68**
87252 5% BST 1%
Surfactant
Nyco MIL-PRF- 183.7 196.7 14.68% 10.25% 0.2-um< > 5-um no 17
87252 5% BST 1%
Surfactant***
Nyco MIL-PRF- 162.5 172.2 13.40% 10.24% .45-um< > 5-um no 23
87252 5% BST 1%
Surfactant
Nyco MIL-PRF- 176.8 187.6 14.21% 11.10% 1-um 5-um no 22
87252 5% BST 1%
Surfactant
*All percentages are by weight
**This sample was contaminated by water saturated DialaAX transformer oil before testing
***This is the contaminated sample resparged and retested
[0063] It will be appreciated that the device and method of the present invention are capable of being incorporated in the form of a variety of embodiments, only a few of which have been illustrated and described above. The invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive and the scope of the invention is, therefore indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.
US7987068
EXPLOSIVE DEVICE COUNTERMEASURES
US2007197051 [ PDF ]
WO2012145122
HIGH DIELECTRIC CONSTANT COMPOSITE MATERIALS AND METHODS OF MANUFACTURE
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application claims priority to U.S. Provisional Application No. 61/466,604, filed on March 23, 201 1 , and entitled "High Dielectric Constant Composite Material," which is incorporated herein by reference in its entirety.
FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
[0002] This invention was made with government support under Grant No. N000-14-08-1 -0267 from the Office of Naval Research. The Government has certain rights to this invention.
FIELD OF THE INVENTION
[0003] The invention relates generally to the field of composite materials, and, more particularly, to a composite material with a high dielectric constant. The composite materials may be used in a variety of applications requiring high-dielectric materials, including antennas, capacitors, and high voltage insulators, among others. The invention further relates to methods of manufacturing high dielectric constant composite materials and devices incorporating the composite material.
BACKGROUND OF THE INVENTION
[0004] Ceramics are often used in applications that require materials with high dielectric constants, such as in capacitors and energy storage devices.
Conventional ceramic materials, however, are typically brittle and susceptible to fracturing under tensile and torsion stresses. Additionally, conventional ceramic materials exhibit a low dielectric strength, limiting their application in high voltage, high power, or high energy storage systems. [0005] Existing efforts to compensate for the inherent brittle nature and low dielectric strength of ceramic material rely on incorporating epoxies or other polymeric macromolecules into a mixture with high dielectric constant ceramic particles. Existing efforts however, do not achieve the ceramic particle packing fraction required to achieve a high effective dielectric constant for the composite. The current efforts also produce composites containing voids, which decrease the dielectric constant and the dielectric strength of the composite.
[0006] Therefore, there remains a need for dielectric materials having high dielectric constants at a range of high frequencies that also possess a high dielectric strength and have robust mechanical properties and strengths.
SUMMARY OF THE INVENTION
[0007] The present invention relates to composite materials having a high dielectric constant and high dielectric strength, methods of producing the composite materials, as well as various devices and structures, such as antennas, capacitors, and high-voltage insulator assemblies incorporating the composite materials. In one embodiment, the high dielectric constant composite material includes a distribution of high dielectric constant ceramic particles and a polymeric material that is mixed with the particles and polymerized in-situ. In various embodiments, the dielectric constant is greater than 20.
[0008] The ceramic particles may be of a single particle size ranging between about 2 nm to about 220 [mu][iota][tau][iota]. Alternately, the distribution of high dielectric constant ceramic particles may be a bimodal distribution, a trimodal distribution, a quadmodal distribution, or higher. The distribution of the high dielectric constant ceramic particles has a first volume fraction of 50% or greater and the polymeric material has a second volume fraction of 50% or less.
[0009] The diameter of the ceramic particles of a largest distribution of the trimodal distribution may be between about 10 [mu][iota][tau][iota] and about 220 [mu][iota][tau][iota], the diameter of the ceramic particles of the intermediary distribution may be between about 500 nm and about 5 [mu][eta][tau][iota], and the diameter of the ceramic particles of a smallest distribution of the trimodal distribution may be less than about 500 nm, and as small as 2 nm. In one embodiment, the trimodal distribution includes at least one first ceramic particle having a first diameter in a first range between 10 [mu][iota][tau][iota] and 220 [mu][iota][tau][iota], at least one second ceramic particle having a second diameter in a second range between 500 nm and 5 [mu][iota][tau][iota], and at least one third ceramic particle having a third diameter in a third range between 50 nm and 500 nm. In another embodiment, the trimodal distribution includes at least one first ceramic particle having a first diameter in a first range between 0.5 [mu][iota][tau][iota] and 3 [mu][iota][tau][iota], at least one second ceramic particle having a second diameter in a second range between 65 [mu][iota][tau][iota] and 150 [mu][iota][tau][iota], and at least one third ceramic particle having a third diameter in a third range between 150 [mu][iota][tau][iota] and 500 [mu][iota][tau][iota].
[0010] Any high dielectric particles, including ceramic particles may be used in the composite material. In various embodiments, the ceramic particles are perovskites, including compounds thereof. More specifically, the perovskites may include barium titanate, strontium titanate, barium strontium titanate, lead zirconate titanate, lead magnesium niobate-lead titanate, and combinations thereof. Further, the surface portions of each of the ceramic particles that are not in contact with the surface of another ceramic particle are in contact with the polymeric material or a liquid filler.
[0011] The polymeric material substantially fills a void space between two or more of the high dielectric constant ceramic particles and, in one embodiment, the polymeric material binds directly to the surface of the high dielectric constant ceramic particles. In one embodiment, the polymeric material is an inorganic-organic coupling agent. Specifically, the binder material may be a polysilsesquioxane formed from coupling agents including silanes, titanates, zirconates, or combinations thereof. For example, the silane coupling agent may be any trialkoxysilane, including those selected from a group consisting of vinyltrimethoxysilane, triethoxyvinylsilane, aminopropytriethoxysilane, or combinations thereof.
[0012] During fabrication, the composite material is compressed in a die press according to one embodiment. In various embodiments, the compression may further facilitate contact and binding between the polymeric material and the ceramic particles. A precursor of the polymeric material may be mixed with the distribution of high dielectric constant ceramic particles before compression in the die press. Further, the precursor may be polymerized and cross-linked in-situ by at least one of heat, a chemical catalyst, or ultraviolet light.
[0013] In various embodiments, the composite material further includes a dielectric fluid that may be separate and distinct from the ceramic particle and polymeric material mixture. The dielectric fluid or a mixture of dielectric fluids may be incorporated into the composite material after the polymerization of the polymeric material. Optionally, the dielectric fluid may also have a high dielectric constant. The dielectric fluid may be selected from a group consisting of water, an alkylene carbonate, an oil, or combinations thereof and the dielectric fluid may include silane, titanate, zirconate, or combinations thereof.
[0014] After polymerization or other formation of the polymer binder, the dielectric fluid is impregnated into the composite material to fill or displace any voids remaining in the composite material, thereby increasing both the dielectric constant and the dielectric strength of the composite material. The composite material may be submerged and/or bathed in the dielectric fluid. Alternately, the dielectric fluid may be forced into the pores of the ceramic material using a vacuum or other pressurized system.
[0015] A method for manufacturing a composite material having a high dielectric constant includes mixing a ceramic powder having a distribution of ceramic particles with a liquid polymer precursor into a paste, placing the paste into a die, compressing the paste, and polymerizing the polymer precursor to form a polymer binder that binds directly to ceramic particles of the ceramic powder distribution. The method may further include compacting the ceramic powder distribution such that the compacted distribution has a packing factor of at least 80%, and impregnating the composite material with a dielectric liquid to fill voids in the composite material and eliminate air from the composite material. In one embodiment, the dielectric fluid is injected or forced into the pores of the ceramic material by a pressurized system, such as a vacuum. [0016] In various embodiments, the paste is compressed by a pressure of about 30 tons per square inch in a die press. Polymerizing the polymer precursor to form a high dielectric constant polymer further includes at least one of heating the die containing the pressed paste for at least thirty minutes and cooling the high dielectric constant composite material, providing a chemical catalyst, or exposing the polymer precursor to ultraviolet light. After polymerization, the method may include removing the composite material from the die, machining the composite material into a desired shape, sanding the composite material, and applying one or more electrodes to the composite material.
[0017] In one embodiment, an antenna assembly includes a composite material having a high dielectric constant, where the composite material further includes a distribution of high dielectric constant ceramic particles and a polymeric material. The antenna assembly also includes conductive sheets or wires that may include copper. The antenna assembly may be a helical antenna, dielectric resonator antenna, or any other suitable antenna. Typically, antennas incorporating the high dielectric constant composite material disclosed herein may be fabricated with smaller dimensions relative to similar conventional antennas.
[0018] In another embodiment, the high dielectric constant composite material may be incorporated into a capacitive or high-energy storage device. The capacitive device has a high dielectric constant and high dielectric strength. In addition, the capacitive device includes two or more electrodes separated by the high dielectric constant composite material. The capacitive device may be formed as a single layer or a multi-layered structure, and the capacitive device may be used for high-density energy storage and filtering. In addition, the capacitive device incorporating the high dielectric constant composite material may be used as a substitute for conventional capacitors in a variety of applications. BRIEF DESCRIPTION OF THE FIGURES
[0019] FIGS. 1 A-C are sectional views of a high dielectric constant composite material according to various embodiments.
[0020] FIG. 2 is a sectional view of a high dielectric constant composite material according to one embodiment.
[0021] FIG. 3 is a sectional view of a high dielectric constant composite material according to one embodiment.
[0022] FIG. 4 is a sectional view of a high dielectric constant composite material mixture according to one embodiment.
[0023] FIGS. 5A-C depict a sequence of functionalizing and binding particles of the high dielectric constant composite material according to one
embodiment.
[0024] FIG. 6 is a plan view of a machined high dielectric constant composite material according to one embodiment.
[0025] FIG. 7 is a plan view of a machined high dielectric constant composite material according to one embodiment.
[0026] FIGS. 8A-B are plan views of a machined high dielectric constant composite material according to one embodiment.
[0027] FIG. 9 is a plan view of a machined high dielectric constant composite material according to one embodiment.
[0028] FIG. 10 is a graph illustrating a dielectric constant range and a loss tangent range for a high dielectric constant composite material according to one embodiment.
[0029] FIG. 1 1 is a graph illustrating a dielectric constant range and a loss tangent range for a high dielectric constant composite material according to one embodiment. [0030] FIG. 12 is a graph illustrating a dielectric constant range and a loss tangent range for a high dielectric constant composite material according to one embodiment.
DETAILED DESCRIPTION OF THE INVENTION
[0031] The present invention relates to a composite material having a high dielectric constant. In particular, the present invention relates to composites that have a high dielectric constant over a range of high frequencies, a high dielectric strength, and that can be formed or machined into complex geometries. The ceramic composite material may be used to manufacture antennas, radio frequency transmission components, microwave transmission components, high energy density capacitors, high-voltage insulators, and other applications that may benefit from a composite material with a high dielectric constant. As used herein, a high dielectric constant refers to a material having a dielectric constant of about 20 or greater.
[0032] In various embodiments, the high dielectric constant composite material includes a distribution of ceramic particles ranging from about 2 nm to about 2000 [mu][iota][tau][iota]. Other embodiments include distributions of ceramic particles ranging from about 10 nm to about 1000 [mu][iota][tau][iota], 50 nm to about 500 [mu][iota][tau][iota], or 200nm to about 220 [mu][iota][tau][iota]. In other embodiments, the high dielectric constant composite material contains ceramic particles of a uniform size. Similarly, the high dielectric constant composite material may contain a bimodal, trimodal, or greater distribution of particles. Preferably, the high dielectric constant composite material has a trimodal distribution of ceramic particles. In all embodiments, the ceramic particle distribution is configured to increase the ceramic packing factor (i.e. the fraction of the volume in the composite structure that is occupied by the ceramic particles), the dielectric constant, and other beneficial properties of the ceramic composite material.
[0033] In one embodiment, the high dielectric constant composite material is manufactured using in-situ polymerization in which polymer precursors are mixed and compressed with the ceramic particles prior to the polymerization and cross-linking of the precursors. In-situ polymerization further increases the packing factor of the ceramic composites. In addition, in-situ polymerization allows for the direct bonding between the polymer and the surfaces of the inorganic ceramic particles. Typically, the polymer is a substantially insulating or non-conductive dielectric polymer that does not contain any metal. As such, the matrix of non-metallic dielectric polymer formed after polymerization allows for higher breakdown voltages and extends the lifetime of the composite material.
[0034] In one embodiment, the high dielectric constant composite material may be impregnated with a dielectric fluid, and preferably a fluid with a high dielectric constant. The dielectric fluid fills any remaining voids in the composite material to eliminate dead space occupied by air, thereby improving the dielectric properties of the composite material.
[0035] The high dielectric constant composites disclosed herein are suitable for use with applications requiring machinable dielectric material having robust mechanical properties, a high dielectric strength, complex shapes, or low temperature forming environments.
The Ceramic Particles
[0036] Referring now to FIGS. 1A-C, various embodiments of the particle distributions for high dielectric constant composite materials 100A-C are depicted. As shown the composite materials 100A-C include a distribution of at least one variety of dielectric particles 102-106 that are bound together by a polymer binder 108. For illustrative purposes, the particles 102-106 are shown as having a regular pattern of distribution in FIGS.1A-C. In various embodiments, the particles 102-106 may be randomly arranged to result in random particle packing. Despite the random
arrangement of the particles 102-106, the particles are dispersed evenly throughout the composite material by stirring and mixing, such that the particle density and other properties of the composite, including the dielectric constant and dielectric strength, are uniform throughout the material. [0037] The composite material may be made using particles of any high dielectric constant. Preferably, the particles 102-106 are ceramic, including but not limited to barium titanate (BaTiOs), strontium titanate (SrTiOs), or barium strontium titanate (BST) particles with the formula Ba(i - X)SrxTiO3. The BST particles may be incorporated as independent particles or they may be included in the composite material 100 as a constituent of a BaTiO3 particle composite. Other suitable materials for the composite particles include lead zirconate titanate (PZT) a ceramic perovskite with the formula Pb[ZrxTii-x]O3, where 0<=x<1 ). In addition, other perovskite materials, non- perovskite materials, and non-linear ferroelectric or anti-ferroelectric materials may also be used.
[0038] Perovskite particles are desired, as they possess a number of characteristics, such as magnetoresistance, robust dielectric properties, and an inherent polarity of the lattice structure, particularly when the unit cell has a tetragonal structure. In addition, the flexibility of bond angles inherent in the perovskite structure permit a variety of molecular distortions, including Jahn-Teller distortions, which may be desired in certain electronic applications.
[0039] Further, the particles 102-106 may be refractory ceramic particles, non-refractory ceramic particles, or combinations thereof. The ceramic particles 102- 106 may be chosen such that the Curie temperature of the particles is much different than the contemplated temperature of operation for the composite materials 100A-C to ensure relative stability in the permittivity of the composite materials. For example, the particles 102-106 having a Curie temperature that differs from the operating temperature of the final composite material by about 20[deg.] C or greater may be chosen. Alternatively, the ceramic particles 102-106 may be chosen such that the Curie temperature of the particles is at or near the contemplated operating temperature for the composite materials 100A-C to maximize the dielectric constant of the composite materials. Furthermore, ceramic particles 102-106 having varied Curie temperatures may be selected in order to broaden the temperature range at which the composite material's dielectric strength and dielectric constant are at peak values or otherwise stable. [0040] Although shown as spherical in FIGS. 1A-C, the ceramic particles 102-106 may have a variety of shapes, including but not limited to spherical and irregular shapes. FIGS. 2-3 depict composite materials 100D and 100E having distributions of cubic and spherical particles, respectively. FIG. 4 depicts a mixture 200 of particles 102 and 104 that may be used to manufacture composite material, similar to the composite material 100B. Irregular shapes are preferred for all particles 102-106, as irregular shapes typically result in high packing densities, which in turn yield a higher dielectric constant for the composite materials 100A-E. As the particles 102-106 may be any shape, the term diameter as used when referencing particle size, may refer to a nominal dimension of the particles. For example, the diameter of irregular shaped particles may refer to a mean diameter of the particle. Similarly, the diameter of predominately-cubic particles may refer to an edge length of the particle. For other particle shapes, such as elliptical, the diameter may refer to the greatest axial or transverse length.
[0041] The composite material 100A, as shown in FIG. 1A, has a trimodal distribution of the dielectric particles 102-106 that range in diameter from less than approximately 2 nm to greater than approximately 2000 [mu][iota][tau][iota]. The particles 102-106 in each size range may be uniform. Alternately, the particles within each size range may be any of a variety of sizes within the size range. Similarly, the particles 102-106 within each size range (i.e. small, intermediate, and large) and across each size range may be the same material or be of different materials and compositions. For example, a combination of barium titanate, strontium titanate, and barium strontium titanate particles may be used within each size range and across size ranges.
[0042] The distribution and range of particle sizes are selected to increase the ceramic packing factor and the dielectric constant of the composites 100A-C. While ceramic particles of any size may be used, the ranges for the smallest particles 102 may be limited by manufacturing limits and the desire to keep the small particles from agglomerating, as agglomerations of small particles will degrade the dielectric characteristics of the composite materials 100A-C. In one embodiment, the small particles 102 are approximately 50 nm in diameter; however, smaller particles as small as 2 nm may be used. Conversely, particles larger than 50 nm can be used as the small particles 102. It is preferably, however, that the smallest achievable nanoparticles are used to maximize packing of the particles.
[0043] Typically, the largest particles 106 are limited only by practical considerations relating to the thickness of the final composite material 100A-C. For example, it is preferred that any single large particle 106 is less than or equal to about 10% of the total thickness of the composite material 100A-C. For example, a composite material that is approximately 2-2.5mm thick may be fabricated using large particles 106 approximately 220 [mu][iota][tau][iota] in size. Similarly, thin composite films may be made using large particles 106 that have smaller dimensions.
[0044] In one embodiment, the size range for the intermediate particles 104 is calculated such that the range of intermediate particle sizes is separated by a common factor from the size range for the smallest particles 102 and the size range for the large particles 106. For example, when using 50 nm small particles 102 and 50 [mu][iota][tau][iota] large particles 106, the ratio of the large particle size to the small particle sizes is 1000. To determine the size range of the intermediate particles 104, the square root of 1000 (approximately 31 .6) is used as factor to determine an intermediate particle size of approximately 1 .58 [mu][iota][tau][iota] (i.e. 50 nm x 31 .6 or 50 [mu][iota][tau][iota] / 31 .6). In other embodiments, the size range for the intermediate particles 104 may be closer to the size range of the large particles 106 or closer to the size range of the small particles 102.
[0045] The ceramic particles 102-106 within each desired size distribution may be purchased commercially, or alternately, produced by milling large ceramic particles into the desired sizes. In various embodiments, the distribution of particle sizes may be optimized by using formulas similar to those used in various concrete or explosive manufacturing processes. In addition, the ceramic particles may be sintered before use.
[0046] By way of example and not limitation, the composite material 100A may contain a trimodal distribution of ceramic particles consisting of
BaTiO3 large particles 106 having diameters between about 40 [mu][iota][tau][iota] and about 220 [mu][iota][tau][iota], BaTiO3 intermediate particles 104 with diameters between about 500 nm and about 5 [mu][iota][tau][iota]. and small particles 102 composed of BaTiO3 or BST with diameters less than about 500 nm. In one embodiment, the smallest particles may have diameters of approximately 2 nm or less.
[0047] Preferably, the ratio of each size distribution as well as the volume fraction for each particle size is calculated to achieve the highest packing factor and to minimize the volume of the polymer binder 108 within the composite materials 100A-C. As used herein, the volume fraction refers to the volume of a constituent (e.g. particle(s) or polymer) divided by the volume of all the constituents of the mixture. For example, in one embodiment of the composite material 100A, having a trimodal distribution of particles, the ratio of the large particles 106, the intermediate particles 104, and the small particles 102, respectively, is approximately 65:25:10 wherein the large particles constitute the largest proportion of the composite mass and volume. In another example, ratios of approximately 65-80 % large particles, approximately 15-20% intermediate particles, and approximately 5-15% small particles may be used. While any other ratio may be used, it is desirable that the ratios for the distribution of particles is determined based on the sizes of the particles and the void fraction for each distribution of the particles. As such, the largest particles 106 typically have the largest proportion of the mass and volume, while the intermediately sized particles have the second largest proportion.
[0048] In various other embodiments, the composite material may be composed of particles within a single size range as shown in FIG. 1 C, or the composite material may have a bimodal distribution of particles as shown in FIG. 1 B. Other distributions, including quadmodal or greater may be used. Further, the sizes of the particles used in forming the composite materials need not be in "adjacent" size ranges. For example, a composite material may be formed using a mixture of large particles 106 and small particles 102, as shown in FIG. 4.
[0049] In various embodiments, the surfaces of the ceramic particles 102- 106 are functionalized to increase the direct bonding between the particles and the polymer binder. There are a multitude of ways in which the surface can be functionalized. For example, each particle 102-106 may be functionalized by hydroxylating the surface 300 to introduce a hydroxyl (-OH) group 302, as shown in FIGS. 5A-C. The hydroxyl groups 302 may be introduced by treating the particles102- 106 with hydrogen peroxide; however, any other suitable method for hydroxylating the particle surfaces without unduly modifying the particles may be used.
[0050] Once the surfaces of the ceramic particles 102-106 have been hydroxylated, silanes and zirconates, for example, may be used to functionalize the surface of the ceramic particles. By way of example, and not limitation, a heated solution of a silane-based polymeric precursor 304, such as the byproduct of vinyltrimethoxysilane and water, 306 may function as both a surface treatment to functionalize and bond directly to the particle surfaces and a binding material to form a highly cross-linked structure 308 between all of the ceramic surfaces 102-106.
The Polymer Binder
[0051] The polymer binder 108 may be any dielectric polymer, including but not limited to cyanate esters or polycyanurates. Preferably, the polymer binder 108 is a polymer material that can be irreversibly cured, although other polymer materials may be used. The polymer binder 108 may also be any polymer or polymer precursor that, preferably, has a small molecular size, is capable of binding directly to the particle surfaces, and is capable of forming highly cross-linked polymer networks. Alternately, other embodiments may use polymers or polymer precursors that do not necessarily have these characteristics. These include gelling polymers and cellulose-based polymers, among others.
[0052] In one embodiment, the polymer binder 108 is formed from a polymeric precursor that is polymerized in-situ to bind directly to the ceramic particles 102-106. In other embodiments, the polymer binder 108 is formed from the melting of a gelling polymeric precursor, such as agar, that swells in the presence of certain solvents, such as water. In these embodiments, the polymer binder 108 is formed when the mixture of the gelling polymeric precursor, the solvent, and the particles 102-106 is cooled and the solvent is removed.
[0053] Preferably, the polymer binder 108 or polymeric precursors do not contain metal particles. A composite material having a non-metallic polymer matrix allows for higher breakdown voltages and a longer life span of the composite material.
[0054] When using a polymeric precursor, the polymer precursor is mixed with the distribution of ceramic particles 102-106 and allowed to penetrate through the distribution to contact the surface of each of the particles. For example, a silane-based polymer precursor, such as a trialkoxysilane and more specifically vinyltrimethoxysilane, tnethoxyvinylsilane, aminopropytriethoxysilane, or combinations thereof, may be used to form the polymer binder 108. In addition, other silane, zirconate or titanate-based coupling agents or polymer precursors that polymerize to form polysilsesquioxanes may be used. Preferably, the polymer precursors have physical and chemical characteristics that allow them to penetrate through the mixture to the particle surfaces 300 and to bind directly to the surfaces of the particles 102-106. Preferably, the polymer precursor has a viscosity low enough to flow between the ceramic particles and coat the particle surfaces 300. Under pressure and optionally, heat, applied during formation of the composite materials 100A-E, the polymer binder 108 and/or the polymeric precursor penetrates the nanometer and sub-nanometer level imperfections on the ceramic particle surfaces 300. The polymer binder 104 therefore eliminates any air voids at the surface boundaries of the particles 102-106. By way of example and not limitation, silane, titanate, and/or zirconate-based polymeric precursors are used to facilitate direct binding between the ceramic particle surfaces 300 and to form the highly cross-linked polymer matrix, such as the matrix 308. [0055] In various embodiments, the dielectric constant of the final composite material 100A-E can be optimized for use based upon the selected polymer binder 104. In one example, a desirable silane-based polymeric precursor has a silane concentration higher than typical silane-based polymer precursors such that there is no need for an additional coupling agent to bind the inorganic ceramic particles 102-106. The elimination of extraneous binding additives, such as additional polymeric
macromolecules further improves the ceramic packing factor, thereby increasing the dielectric constant of the composite materials 100A-E.
[0056] In one embodiment, the polymer binder 108 is composed of a cyanoresin. The cyanoresin binder will yield a composite material 100A-E, having a dielectric constant that may vary from between about 10 to about 55 at a frequency of hundreds of MHz. For example, the polymer binder may be a commercially available product such as CR-S manufactured by Shin-Etsu Chemical Co., Ltd of Tokyo, Japan. CR-S is a cyanoethylated cellulose polymer having a high dielectric constant, good strength and is machinable.
[0057] In another embodiment, the dielectric constant of the composite materials 100A-E is increased by using a gelling polymeric precursor, preferably agarose, which is typically in the form of agar. The gelling polymer may be a biopolymer that is naturally derived or synthetically produced. The gelling polymeric precursor is melted in a solvent and compacted into a composite mold with the distribution of ceramic particles 102-106. The composite mold is then cooled under pressure allowing the gelling polymeric precursor to swell in the presence of the solvent. The composite mold is subsequently reheated to a temperature below the melting temperature of the polymeric precursor to remove the solvent, thereby forming the polymer binder 108. As a result, the volume of the polymer binder 108 is reduced from that of the previously swollen polymeric precursor without changing the overall structure of the composite material.
[0058] The reduced volume of the polymer binder 108 may create voids or cavities around the ceramic particles 102-106. A dielectric liquid is then impregnated within the composite material to fill and displace all of the cavities or voids within the composite material and to surround the surfaces 300 of the ceramic particles 102-106. Other polymeric precursors may also be used, including gelatin, carrageen, or any other precursor for a polymer binder 108 that will swell in the presence of the solvent but will not absorb and swell in the presence of the dielectric liquid.
The Dielectric Liquid
[0059] The dielectric liquid is a filler fluid that may be impregnated into the composite material to penetrate into and fill any remaining voids within the composite material. The dielectric fluid also coats, at the nanometer and sub-nanometer level, any exposed surfaces 300 of the ceramic particles. The addition of the dielectric liquid can increase both the dielectric constant and the dielectric strength of the composite by displacing voids in the material, reducing the porosity of the composite material, and enhancing the boundary interface of the ceramic particles. The dielectric fluid can be impregnated into the formed composite material by submerging the material in the dielectric fluid. The dielectric fluid may further infuse into the composite material via capillary action alone, or may be forced into the material with the aid of a vacuum or other pressurized system.
[0060] In one embodiment, the dielectric fluid has a high dielectric constant. For example, the high dielectric constant may be, but is not limited to, water, glycerine, and alkylene carbonates such as ethylene carbonate, propylene carbonate, glycerine carbonate, butylene carbonate, and combinations thereof. Conversely, a dielectric fluid having a low dielectric constant may be used to increase the dielectric strength of the composite while, to a lesser extent, increasing the dielectric constant of the composite material. Example dielectric fluids having low dielectric constants include oils, such as those used in electrical insulators. Preferably, the dielectric liquid has a high dielectric constant, low viscosity, low dielectric losses, and a low evaporation rate.
[0061] The dielectric fluid may be impregnated into composite materials formed by the in-situ polymerization of the polymer binder, as well as those formed from the gelling polymeric precursor. In addition, composite materials that are infused with the dielectric liquid may optionally be coated with any suitable gas-impermeable material to prevent or reduce evaporation of the dielectric liquid. For example, the composite material may be coated with silane, titanate, and/or zirconate-based polymers.
[0062] In another embodiment, a composite material having a dielectric constant of approximately 100 at frequencies between 100 and 1000 MHz may be made using silane, titanate, and zirconate chemicals that form polysilsesquioxanes after in-situ polymerization. The polymer precursors containing these compounds and the resulting polysilsesquioxanes have a small molecule size and are suitable for binding directly with the ceramic particle surface and forming highly cross-linked polymer networks.
[0063] In yet another embodiment, a composite material having a dielectric constant of approximately 550 at frequencies between 100 and 1000 MHz may be made using a gelling polymeric precursor that swells in the presence of water. Preferably, the gelling polymeric precursor swells in the presence of water but not in the presence of alkylene carbonates. Further, the preferred gelling polymeric precursor can be melted before mixing with the ceramic particles 102-106, has a high strength when dried, has a high melting temperature, and is a biomaterial.
Fabricating the Composite Material
[0064] As previously described, the ceramic particles 102-106 may be obtained commercially or may be formed during a milling process. In one embodiment, where the ceramic particles are formed by milling, a solvent such as methyl ethyl ketone is used to prevent agglomeration of the particles. Additional solvents, including but not limited to acetone, methyl propyl ketone, polymethylmethacrylate, or dichloromethylene may also be used. In addition, a surfactant may be added to the milling process to prevent agglomeration. By way of example and not limitation, the surfactant may be oleic acid, alkylbenzene-sulfonic acid, or phosphate ester. Preferably, the surfactant is a polar additive. In addition, the surfactant may act to functionalize the surfaces of the ceramic particles to enhance bonding with the polymer binder. [0065] In various embodiments, the ceramic particles may be sintered to produce hardened particles before mixing with the polymeric precursors or binders. The sintered particles have reduced porosity which will decrease voids within the particles, increase the particle packing fraction of the particles, and prevent or at least minimize the infiltration of the polymeric precursors or binders into the particles.
[0066] In one embodiment, a mixture of the ceramic particles and a polymer precursor containing silane, titanate, or zirconate, such as vinyltrimethoxysilane or triethoxyvinylsilane, is further mixed with water and prepared at a light boil while stirring. For example, the mixture may be prepared at a temperature between about 100[deg.]C and about 200[deg.]C. Heating the mixture increases the rate at which the surfaces of the ceramic particles are functionalized and increases the rate of formation for the polysilsesquioxane network.
[0067] The composite material is further formed by dry pressing in a die to shape the material. For example, a paste composed of the ceramic particles 102-106, the polymer precursors, and any surfactants are pressed in a die. Typically, the die is pressed using mechanical or hydraulic means to compact the paste to the shape of the die. In other embodiments, the composite material may be formed by isostatic pressing, where the paste is isostatically pressed using a flexible membrane acting as a mold. While pressing the composite paste is the preferred method of fabrication, the
composite materials may also be formed by spin coating or solution casting.
[0068] In various embodiments, the composite material is pressed at a pressure from about 100 pound per square inch (PSI) to well over 30 metric tons per square inch. The composite material 100 may be formed by bidirectional orunidirectional pressing technique, although other pressing techniques may be used. In one embodiment, bidirectional pressing eliminates a density gradient that may form in the composite material and provides a uniform dielectric constant throughout the composite material. Conversely, if a gradient density is desired, the density of particles can be varied throughout the composite material by implementing a unidirectional pressing process to fabricate the composite material. [0069] Subsequent to or concurrent with pressing the paste, the die may be heated to facilitate in-situ polymerization of the polymer precursors, thereby forming the polymer binder 104. In other embodiments, the polymer precursors may be polymerized or cured using ultraviolet (UV) radiation or by the addition of a catalyst to bring about chemical polymerization. In one embodiment, the polymer binder 108 binds directly to the surface of the ceramic particles 102-106 and forms a highly-crosslinked structure. As such, there is no need for additional polymers to function as binders.
[0070] In various embodiments, the dielectric fluid having a high dielectric constant is added after polymerization of the polymer precursor to fill any voids that may remain between the polymer binder 108 and ceramic particles 102-106. In particular, the filler liquid improves the boundary interface at the nanometer and sub-nanometer level on the surface 300 of the ceramic particles 102-106.
[0071] The final composite materials may be coated to prevent the evaporation of an infused dielectric liquid and to reduce the porosity of the composite material. The composite materials may then be cut, machined, and or sanded to the desired size. Additional finishing may include the addition of one or more electrodes to the final composite material. By way of example, the electrodes may be platinum, gold, or any suitable conducting material. In one embodiment, the electrodes are sputtered directly onto the composite material to eliminate any air gaps between the electrodes and the composite material.
[0072] In various embodiments, the in-situ polymerization process allows for particle packing factors of at least 80% to be achieved and correspondingly higher dielectric constants to be observed. The in-situ polymerization process also allows the loss tangent and the dielectric constant to be simultaneously tuned to desired requirements as determined by the final application of the composite material 100. The polysilsesquioxane used to bind the matrix of particles eliminates the need for a high viscosity polymer and epoxy to bind the particles.
The Composite Material
[0073] Examples of machined high dielectric constant composite materials are shown in FIGS. 6-9. As shown in FIG. 6, the composite materials 100A-E may be cut and machined into a solid disc 400. A ruler 402 is depicted to provide a sense of scale; however, the composite materials may be fabricated to any size and configuration. FIG. 7 is a photograph of a composite material that has been machined into substrate 500 for receiving additional elements. FIGS. 8A depicts a composite material that has been machined into a number of annular discs 600A-G. As shown in FIG. 8B, the annular discs 600A-G may be mounted onto a rod 602 and secured together by fasteners 604A-B. FIG. 9 depicts a composite material that has been fabricated as a disc 700. In addition, single thin layer sheets or multiple-layer composite capacitors may also be fabricated using the high dielectric composite materials.
[0074] As shown in FIG. 10, a profile 1000 of the dielectric constant for a composite material having an average dielectric constant of approximately 100 actually varies from approximately 108 to approximately 90 or +/-10% over a frequency range from 200 MHz to 4.5 GHz. As shown, a profile 1002 of the loss tangent over the entire frequency range is less than 0.12 and below 500 MHz, the loss tangent is less than 0.05. As previously described, the loss tangent and dielectric constant may be predetermined by the selected particle mixture, the heat and pressure applied during fabrication, and the period over which the heat and pressure are applied. In one embodiment, a loss tangent below 0.08 at 2 GHz has been achieved, while in other embodiments, loss tangents less than 0.001 are possible. The low loss tangent significantly expands the variety of applications for using the high dielectric constant composite material.
[0075] The range of dielectric constants and loss tangents for a composite material having an average dielectric constant of approximately 45 is shown in FIG. 1 1 . As shown, the profile 1 100 of the dielectric constant varies from approximately 49 to approximately 39 over the frequency range from 200 MHz to 4.5 GHz. The profile 1 102 of the loss tangent over the entire frequency range is less than 0.14 and below 500 MHz, the loss tangent is less than 0.08. FIG. 12 depicts the profile 1200 of the dielectric constant range and loss tangent range for a composite material having a peak dielectric constant of approximately 550. As shown, the dielectric constant varies from approximately 550 to approximately 400 over the frequency range from 200 MHz to 4.5 GHz. The profile 1202 of the loss tangent over the entire frequency range is less than 0.35 and below 500 MHz, the loss tangent is less than approximately 0.125. It is believed that the actual dielectric constant of the composite materials, as shown in FIGS. 10-12 may be approximately 10% higher than shown due, in part, to imperfections in the methods of measurement and surface imperfections in the tested composite materials and the testing apparatus.
[0076] In operation, the electric field of the high dielectric constant composite materials has been found to be in the range of about 1 -1 .84 MV/cm for single breakdowns. With variations to the mixing techniques, as disclosed herein, the electric field breakdown characteristics can be improved to well over 4 MV/cm, in some embodiments. Therefore, the high dielectric constant composite materials may have an energy storage density of over 200 J/cm<3> and up to 1000 J/cm<3>. In one embodiment, the high dielectric constant composite material has an energy density at room
temperature of approximately 20 J/cm<3>. As such, low loss capacitors, either single layer or multilayer, can be fabricated using the in situ polymerization process disclosed herein. Further, the composite material has been produced having a dielectric constant over 550 at 200 MHz and as high as 56000 at 20 KHz.
Exemplary uses for the Composite Material
[0077] The high dielectric constant composite materials 100A-E, 400, 500, 600A-G, and 700 are suitable for a variety of applications, including but not limited to antennas, capacitors, high-energy storage devices, and high-voltage insulators. The high frequency properties of the high dielectric constant composite materials make them an ideal material for radio frequency and microwave transmission components, including but not limited to antennas and microwave substrates. The high dielectric constant materials can also be used for tuning microwave cavities, high energy density capacitors, and high frequency capacitors. By way of example and not limitation, the high dielectric composite material is suitable for use in high power antennas. As such, the size of dielectric loaded antennas can be minimized by incorporating the high dielectric constant composite materials of the present disclosure. In particular, the composite materials may be incorporated into helical antennas, dielectric resonator antennas, or any antenna which may benefit from and whose dimensions may be reduced because of the inclusion of the composite material. For example, according to one embodiment, antennas incorporating the high dielectric constant composite disclosed herein may be fabricated with a six to ten-fold reduction in size. In other embodiments, the antennas incorporating the high dielectric constant composite material may be reduced even further to dimensions approaching the Chu-Harrington limit (Chu-limit).
[0078] Typically, one or more dimensions of an antenna are a function of the wavelength of the electromagnetic wave propagating in the antenna material. As such, the antenna size for a given frequency is approximately proportional to one over the square root of the dielectric constant. In another aspect, the composite material 100 can be used to fabricate antennas, including broadband antennas, approximately one- tenth the size or less of conventional antennas that use an air dielectric.
[0079] In addition to antennas, the composite material may be used to improve the function and shrink the dimensions of conventional capacitors.
Capacitance between two points is proportional to the dielectric constant of the material separating those two points. By incorporating a composite material with a high dielectric constant, the area of the electrodes required for a given capacitance is reduced in proportion to the increase in the dielectric constant over a conventional dielectric. Therefore, capacitors incorporating the high dielectric constant composite material can be made more compact than traditional capacitors. Additionally, the energy stored in a capacitor increases by the square of the voltage across it. By incorporating the high dielectric constant material with a high dielectric strength compared to traditional high dielectric constant ceramic materials, the energy capable of being stored in the same volume is increased by the square of the increase of the dielectric strength over the conventional high dielectric constant material. Since the energy density of a capacitor is proportional to the dielectric constant and the square of the electric field within the dielectric, the energy density in capacitors incorporating the composite material with a high dielectric constant and a high dielectric strength can be orders of magnitude greater than those achieved with conventional low dielectric constant materials and conventional high dielectric constant ceramics.
[0080] The composite materials may also be incorporated into high- voltage insulators. The composite materials allow for greater control and field shaping of electrical fields. Field shaping may further allow the dimensions of the insulators, in particular the length, to be greatly decreased.
High Dielectric Constant Composite Material Preparation Methods
Example 1
[0081] By way of example and not limitation, an exemplary method used to prepare the high dielectric constant composite material with a verified dielectric constant of between about 75-140 at frequencies ranging between 200 MHz and 4.5 GHz is provided. Initially, approximately 250 ml of water and 250 ml of triethoxyvinylsilane were poured into an open beaker in a fume hood. The liquids were mixed with a magnetic stirrer at a temperature just below the boiling point of the mixture. The mixture was removed from the heat source and stirring was suspended when the mixture became miscible.
[0082] The solution was cooled to room temperature for immediate use. Alternately, the solution may be chilled for long-term storage. While the solution was cooling to room temperature, a mixture of ceramic powder was prepared. The mixture was composed of 65 wt% BaTiO3 particles with diameters between 65 [mu][iota][tau][iota] and 150 [mu][iota][tau][iota], 25 wt% BaTiO3 particles with diameters between 0.5 [mu][iota][tau][iota] and 3 [mu][iota][tau][iota], and 10 wt% BaTiO3 and BST particles with diameters less than 100 nm. The ceramic powder was mixed with a mixer mill for approximately two minutes. Caution was exercised to avoid excessive mixing time that could have resulted in milling the larger particles down to smaller sizes.
[0083] The miscible solution cooled to room temperature was added to the powder mixture, while manually stirring. The miscible solution was added until a paste formed. The paste had a light consistency but was viscous enough to hold its shape. The paste was then placed in a dry pressing die and packed lightly to ensure an even distribution of the material.
[0084] The die was placed on a lab press stand and the press was activated to apply force to the top and the bottom pistons of the die assembly.
Approximately, 30 metric tons per square inch was applied to the die. The force was applied to the die, while any excess fluid that emerged from the die was removed. While, the force was still applied, the die was heated to a temperature of approximately 100[deg.] F by a heating element.
[0085] The temperature of the die was held at the desired level for approximately thirty minutes, and then the heating element was deactivated, while the die and pressed composite were allowed to cool to room temperature. After cooling, the composite was removed from the die, machined, and sanded to the desired dimensions. Lastly, electrodes were applied to the composite material 100 and a number of tests were performed to determine the dielectric constant and loss tangent of the composite material.
Example 2
[0086] By way of example and not limitation, another exemplary method is provided for producing a composite material with a dielectric constant between 400 and 600 at frequencies between 200 MHz and 4.5 GHz. Initially, a mixture of perovskite ceramic particles was prepared. The mixture was composed of 72 wt% BaTiO3 particles with diameters between 65 [mu][iota][tau][iota] and 150 [mu][iota][tau][iota], 21 wt% BaTiO3 particles with diameters between 0.5 [mu][iota][tau][iota] and 3 [mu][iota][tau][iota], and 7 wt% BaTiO3 and BST particles with diameters less than 100 nm. The ceramic powder was mixed with a mixer mill for approximately two minutes. Caution was exercised to avoid excessive mixing that could have resulted in milling the larger particles down to smaller sizes.
[0087] Agar, primarily as agarose, was used as the polymer binder. While, stirring, the agar was heated and melted at a high concentration in water. While heated, the agar and water solution was slowly added to and mixed with the ceramic particles in a pre-heated container. The mixture was then heated and pressed in a pre-heated die. After pressing, the compacted composite material was allowed to cool while still under pressure.
[0088] The agar within the composite material was further dried by heating the compacted composite material to temperature below the melting temperature of the agar binder, thereby allowing any remaining water to evaporate. The agar binder, no longer swollen with water, had a reduced volume, thereby creating a number of channels within the composite material. Despite the reduction in volume of the polymer binder, the structural integrity of the highly packed ceramic particles prevents the overall structure of the composite material from contracting. The composite material was then impregnated with a dielectric fluid to fill any voids that remained between the polymer binder and ceramic particle surfaces by submerging the composite in the dielectric fluid. The dielectric fluid was a mixture of akylene carbonates consisting of 50% ethylene carbonate and 50% propylene carbonate.
[0089] It will be appreciated that the materials, devices, and methods of the present invention are capable of being incorporated in the form of a variety of embodiments, only a few of which have been illustrated and described above. The invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive and the scope of the invention is, therefore indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.
US7390984
WO2005038838
High power liquid dielectric switch
Method and apparatus for switching high power at high repetition rates. The apparatus is preferably a switch utilizing a pressurized flowing dielectric. The pressurized dielectric suppresses growth of dielectric breakdown byproducts, such as large bubbles and breakdown contamination, enabling lower dielectric flow rates to remove the byproducts. In addition to the advantage of lower flow rates, and thus smaller and lighter pumping means, the switch can switch high energies (up to megajoules) at fast repetition rates, up to thousands of pulses per second. The switch is preferably triggered to reduce jitter. The switch can also be used to remove water from oil.
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims the benefit of the filing of U.S. Provisional Patent Application Ser. No. 60/690,223, entitled "High Power Liquid Dielectric Switch," filed on Jun. 13, 2005. This application is also a continuation-in-part application of U.S. patent application Ser. No. 10/870,381, entitled "High Power Liquid Dielectric Switch", filed on Jun. 17, 2004, which claims the benefit of the filing of U.S. Provisional Patent Application Ser. No. 60/479,405, entitled "Development of High Power, High Pressure, Rep-Rate, Liquid Dielectric Switches," filed on Jun. 17, 2003. The specification and claims of all of these references are incorporated herein by reference.
GOVERNMENT RIGHTS
[0002] The U.S. Government has a paid-up license in this invention and the right in limited circumstances to require the patent owner to license others on reasonable terms as provided for by the terms of United States Air Force, Air Force Research Lab under Contract No. USAF F33615-01-C-2191.
BACKGROUND OF THE INVENTION
[0003] 1. Field of the Invention (Technical Field)
[0004] The present invention relates to a high power electric switch which has an ultra short rise time and can be fired at a repetition rate from less than a pulse per second to more than 20,000 pulses per second and can switch joules to megajoules of energy per pulse with switch rise times of less than a nanosecond, yet switch pulse widths ranging from picoseconds to milliseconds.
[0005] 2. Background Art
[0006] Note that the following discussion is given for more complete background of the scientific principles and is not to be construed as an admission that such concepts are prior art for patentability determination purposes.
[0007] Large scale pulse power systems, such as accelerators, fusion accelerators, medical accelerators, high power microwave systems, and other high voltage or pulse power systems require the switching of very high power (megawatt) loads, for example from one Joule to megajoules per pulse, and high repetition rates, for example from less than one pulse per second to 20,000 pulses per second. Early studies at moderate pressures have shown breakdown strength in liquids to be a function of pressure up to at least 350 psi (see K. C. Kao and J. B. Higham, "The effects of hydrostatic pressure, temperature, and voltage duration on the electric strengths of hydrocarbon liquids," J. Electrochem. Soc., vol. 108, no. 6, pp. 522-528, June 1961). Pressurized flowing dielectric switches which can switch several hundred kilovolts are known in the art. However, such switches which operate at or near atmospheric pressure require substantial dielectric flow rates of 10-1000 liters per second (l/sec) when they are used to switch multikilojoule pulses. In 1992, subnanosecond rise time, kilohertz rep-rate oil switches were built and demonstrated that could operate at up to 290 kV at 200 pps and at 170 kV with a rep-rate of 1000 pps. The demonstrated rise time into a 97 [Omega] resistive load was 280 ps. The modulator system, which utilized near atmospheric medium pressure oil switches, transferred a peak energy of 50 J per pulse (R. Curry et al., "The Development and Testing of Subnanosecond-Rise, Kilohertz Oil Switches for The Generation of High-Frequency Impulses", IEEE Transactions on Plasma Science, Vol. 20, No. 3, June 1992, pp. 383-392, incorporated herein by reference) and demonstrated significant improvement in the breakdown jitter of liquid switches. These oil switches utilized transformer oil at pressures ranging from 1 atmosphere up to 100 psig. The flow rate geometries used in the switches included cross flow, or axial flow in switches that had a near uniform and enhanced electrode geometry. However, these switches were unable to switch kilojoules of energy for they were limited by residual bubbles at a flow rate of 1.6-7.57 l/sec at a repetition rate of over 100 pulses per second (pps).
[0008] Single-shot work on high pressure liquid switches examined the effects of pressure upon breakdown voltage (see J. Leckbee, R. Curry, K. McDonald, R. Cravey, and A. Grimmis, "An advanced model of a high pressure liquid dielectric switch for directed energy applications," in Proc. IEEE 14th Int'l. Pulse Power Conf., 2003, pp. 1389-1393, and J. Leckbee, R. Curry, K. McDonald, P. Norgard, R. Cravey, G. Anderson and S. Heidger, "Design and testing of a high pressure, rep-rate, liquid dielectric switch for directed energy applications," in Proc. IEEE 26th Int'l. Power Modulator Conf., 2004, pp. 193-196).
[0009] When a high voltage pulse is applied to a flowing dielectric switch, once the switch breakdown voltage is reached, a streamer is launched and subsequent avalanche ionization and breakdown of the dielectric results. The arc then ionizes the dielectric medium and a gas bubble is formed between the electrodes. As the hydraulic or hydrostatic pressure is increased, the bubble size decreases. It is known that above a critical pressure for certain liquids, no bubbles are formed by charge injection (R. Kattan et al., "Formation of Vapor Bubbles in Non-polar Liquids Initiated by Current Pulses", IEEE Transactions on Electrical Insulation Vol. 26, No. 4, August 1991, pp 656-662, incorporated herein by reference). However, below a given operating or critical pressure the diameter of the bubble expands well beyond the electrode separation distance. The gas bubble grows and subsequently collapses, oscillating, until it finally rapidly degenerates into both suspended micro-bubbles and discharge byproducts (principally hydrocarbons) that encompass a large volume, if not the entirety, of the switch housing and electrode region.
[0010] Liquid dielectric insulated switches cannot sustain high voltages when gas bubbles, dissolved gases, and hydrocarbon byproducts are present because arcing or pre-firing is uncontrollably self-initiated. This also prevents recovery of the switch if voltage were reapplied before the entire volume of liquid in the switch could be exchanged, thus reducing the required achievable repetition rate because of the enormous liquid flow rates that would otherwise be required. Consequently, the repetition rate attainable by present day low-pressure liquid dielectric switches which transfer 100 J-1 MJ is typically limited to much less than one pulse per second, thereby eliminating them from addressing the high average power requirements of many crucial applications. This phenomenon occurs in all known liquid dielectric media suitable for pulse power switching applications, including water, water-glycol solutions, transformer oil, polyalphaolefin (PAO), and other synthetic dielectrics.
[0011] Thus there is a need for a kilovolt to megavolt capable, multijoule to megajoule range high power switch with high repetition rate operation, minimized dielectric media flow volume requirements with maximized local flow velocity in the vicinity of the electrodes; minimized electrode erosion; and reduced byproduct formation. There is also need for a compact switch with reduced acoustic impulse and a reduced EMI signature, and with enhanced reliability due to the inhibition of the access and/or adherence of the discharge byproducts to the switch housing solid insulators. The ability of the switch to utilize fluids such as PAO or other synthetic or natural dielectrics that are compatible with existing airframe and aerospace systems is a major advantage, allowing the switch to be integrated with an existing airframe hydraulic system, thereby reducing the volume of support equipment required for directed energy systems.
SUMMARY OF THE INVENTION (DISCLOSURE OF THE INVENTION)
[0012] The present invention is an electric switch comprising at least two electrodes and a flowing liquid dielectric having a pressure greater than approximately 100 psig, wherein the switch is capable of switching greater than approximately 1 joule, or preferably greater than approximately 50 joules, or more preferably greater than approximately one kilojoule, or most preferably greater than approximately one megajoule. The switch is preferably capable of switching greater than approximately five kilovolts, or more preferably greater than approximately 50 kilovolts, or yet more preferably greater than approximately one megavolt, or most preferably greater than approximately 5 megavolts. The switch preferably has a repetition rate of greater than approximately one pulse per second (pps), or more preferably greater than approximately 10 pps, or even more preferably greater than approximately 100 pps, or yet more preferably greater than approximately 1000 pps, or most preferably greater than approximately 10,000 pps. The dielectric preferably has a flow rate of less than approximately 100 liters per second, or more preferably less than approximately 20 liters per second, or even more preferably less than approximately 2 liters per second, or most preferably less than approximately 0.2 liters per second.
[0013] The dielectric is preferably de-aerated and preferably comprises a synthetic lubricant, optionally hydraulic fluid. The dielectric most preferably comprises polyalphaolefin (PAO). The distance between the at least two electrodes is preferably variable. Each electrode preferably has at least one opening which enables the dielectric to flow between an interior and exterior of each electrode. The dielectric preferably enters the switch cavity through the opening in a first electrode and exits the switch cavity through the opening in a second electrode. Each electrode is optionally substantially hemispherical. Alternatively, a first electrode is partially surrounded by a second electrode. The switch optionally comprises at least one cylindrical flow channel, which preferably provides a flow of the dielectric around the first electrode. The switch preferably comprises an outer coaxial return and a dielectric flow system. The dielectric flow system is optionally integrated with the switch, or alternatively comprises an airframe hydraulic system, or alternatively comprises a stand alone pump cart or system.
[0014] The invention is also a method for switching comprising the steps of providing at least two electrodes, pressurizing a liquid dielectric to a pressure greater than about 100 psig, flowing the liquid dielectric between the electrodes, inducing a voltage drop between the electrodes of at least a breakdown voltage of the dielectric, breaking down the dielectric between the electrodes, thereby providing an electrical path between the electrodes; and switching greater than approximately 1 joule. The breaking down step is preferably performed at a rate of at least approximately one pps, or more preferably at least approximately 10 pps, or even more preferably at least approximately 100 pps, or yet more preferably at least approximately 1000 pps, or most preferably at least approximately 10,000 pps. The dielectric flows at a rate of preferably less than approximately 100 liters per second, or more preferably less than approximately 20 liters per second, or even more preferably less than approximately 2 liters per second, or most preferably less than approximately 0.2 liters per second. The present method is for switching preferably greater than approximately 50 joules, or more preferably greater than approximately one kilojoule, or most preferably greater than approximately one megajoule. The method is also for switching preferably greater than approximately five kilovolts, or more preferably greater than approximately 50 kilovolts, or even more preferably greater than approximately one megavolt, or most preferably greater than approximately 5 megavolts.
[0015] The method preferably further comprises the step of de-aerating the dielectric, and preferably further comprises the step of varying a distance between the electrodes. The dielectric preferably flows out of a first opening in a first electrode and into a second opening in a second electrode. The method preferably further comprises the step of partially surrounding the first electrode with the second electrode, wherein the dielectric is preferably flowed around the first electrode. The method further preferably comprises the step of removing breakdown contamination, optionally comprising bubbles, from between the electrodes. The method further preferably comprises the step of lowering an inductance of a switch comprising the electrodes and the dielectric, preferably by partially surrounding the switch with an outer coaxial return.
[0016] The invention is also an electric switch comprising at least two electrodes, a flowing liquid dielectric having a pressure greater than approximately 100 psig; and a trigger for operating the switch. The trigger preferably comprises an element selected from the group consisting of a trigatron, a laser pulse, a microwave pulse, and series injection. The trigatron is preferably disposed substantially between the electrodes. The trigger is alternatively operated by adjusting a pressure of the dielectric. The switch optionally further comprises at least one additive in the dielectric for reducing the dielectric strength of the dielectric. The switch preferably further comprises an element for controlling flow of said dielectric. The element is preferably cylindrically disposed around one of the electrodes. The switch is preferably capable of switching greater than approximately 1 joule.
[0017] The invention is further a method for switching comprising the steps of providing at least two electrodes, pressurizing a liquid dielectric to a pressure greater than about 100 psig, flowing the liquid dielectric between the electrodes; and triggering a dielectric breakdown between the electrodes, thereby providing an electrical path between the electrodes. The method preferably further comprises the step of switching greater than approximately 1 joule. The triggering step is preferably performed at a desired voltage or desired time and optionally comprises operating a trigatron. The triggering step optionally comprises raising a pressure of the dielectric, thereby increasing a gap between the electrodes to a first gap value, providing a voltage difference between the electrodes, and lowering the pressure of the switch, thereby decreasing the gap between the electrodes to a second gap value. The voltage difference is preferably not large enough to cause dielectric breakdown at the first gap value but is preferably large enough to cause dielectric breakdown at the second gap value. The method preferably further comprises the step of controlling a flow of the dielectric, preferably comprising employing a flow shaping element.
[0018] The invention is also a method for reducing water content in a dielectric, the method comprising the steps of providing at least two electrodes, pressurizing a liquid dielectric containing water to a pressure greater than about 100 psig, flowing the liquid dielectric between the electrodes, and triggering a dielectric breakdown between the electrodes, thereby removing at least some of the water from the dielectric. The dielectric preferably comprises a fluid selected from the group consisting of a synthetic lubricant, hydraulic fluid, and polyalphaolefin (PAO).
[0019] The invention is yet further a switch comprising at least two electrodes, a flowing liquid dielectric having a pressure greater than approximately 100 psig, and a flow shaping element for controlling the flow of the liquid dielectric. The switch alternatively comprises at least two electrodes, a flowing liquid dielectric having a pressure greater than approximately 100 psig, and a dielectric additive for reducing the dielectric strength of the dielectric.
[0020] An object of the present invention is to provide a high power switch capable of achieving high repetition rates.
[0021] Another object of the invention is to provide a high pressure oil switching technology that results in low switching jitter and long electrode lifetime.
[0022] An advantage of the switch of the present invention is its lower dielectric flow rate, which permits the use of a small, lightweight flow recirculating system, and increases the achievable repetition rate.
[0023] A further advantage is the compatibility of the present switch with existing hydraulic fluids and airframe hydraulic systems, thus optionally eliminating the need for a separate dielectric flow system.
[0024] Other objects, advantages and novel features, and further scope of applicability of the present invention will be set forth in part in the detailed description to follow, taken in conjunction with the accompanying drawings, and in part will become apparent to those skilled in the art upon examination of the following, or may be learned by practice of the invention. The objects and advantages of the invention may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.
BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The accompanying drawings, which are incorporated into and form a part of the specification, illustrate several embodiments of the present invention and, together with the description, serve to explain the principles of the invention. The drawings are only for the purpose of illustrating a preferred embodiment of the invention and are not to be construed as limiting the invention. In the drawings:
[0026] FIG. 1 is a cutaway view of an inline switch of the present invention;
[0027] FIG. 2 is a cutaway view of a coaxial switch of the present invention;
[0028] FIGS. 3A and 3B are graphs depicting experimental results showing the variation of switching voltage and carbon byproduct region size with time according to Example 1 of the present invention;
[0029] FIGS. 4A-4C are graphs depicting experimental results showing the variation of breakdown voltage, maximum bubble radius, and bubble oscillation period with pressure according to Example 1 of the present invention;
[0030] FIG. 5 is a cutaway diagram of a preferred embodiment of the switch geometry;
[0031] FIG. 6 is a basic circuit diagram showing pulse modulator to the left of the transformer and the water PFL to the right of the transformer;
[0032] FIG. 7 depicts a typical discharge waveform recorded at the anode (11.4 kV/div vertical resolution, 20 ns/div horizontal resolution);
[0033] FIG. 8 shows cross sectional and end views of a flow shaping element of an embodiment of the present invention;
[0034] FIG. 9 shows the flow shaping element of FIG. 8 installed in the switch;
[0035] FIG. 10 shows electric field strength at breakdown versus switch inlet pressure for a net volumetric flow rate of 0.379 L.s<-1 > (6 gpm);
[0036] FIG. 11 shows electric field strength as a function of switch inlet pressure for a net volumetric flow rate of 0.568 L.s<-1 > (9 gpm);
[0037] FIG. 12 shows five shot bursts at 1 pps (71.4 kV/div vertical resolution, 500 ns/div horizontal resolution);
[0038] FIG. 13 shows a type 304 stainless steel electrode following 250,000 shots with original oil flow configuration. The width of the arc band is approximately 1.7 cm;
[0039] FIG. 14 shows an Elektro-Metall (formerly Schwarzkopf) K-33 electrode following 4000 shots with enhanced oil flow configuration. The width of the main spot (top to bottom) is approximately 0.7 cm;
[0040] FIG. 15 shows a second type 304 stainless steel electrode after 150,000 shots with enhanced oil flow configuration. The width of the arc band is approximately 1.0 cm; and
[0041] FIG. 16 depicts a trigatron-triggered switch of the present invention.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0042] The present invention is a liquid dielectric switch able to switch from hundreds of kilovolts to megavolts and thousands of kiloamperes, with discharge times ranging from a picosecond to a few milliseconds or less, operated at pressures ranging from 1 psig to 8000 psig.
[0043] As used throughout the specification and claims, "breakdown contamination" means discharge, carbon, hydrocarbon and/or electrode byproducts, byproducts, debris, debris cloud, bubbles, micro-bubbles, and the like.
[0044] The switch preferably uses liquid dielectric pressures on the order of about 10 psig, and more preferably about 50 psig, and more preferably about 100 psig, and most preferably about 1000 psig or more, thereby either preventing gas bubble formation or dramatically reducing bubble size, which enables rapid reabsorption of the bubbles by the fluid. Although a bubble of significant size is not generated at high enough pressures, a debris cloud containing discharge byproducts (principally carbon and electrode byproducts) expands from the discharge site and, if not removed, eventually fills a significant portion of the switch volume. However, pressurization minimizes this issue, reducing the volume of contaminated dielectric fluid, allowing the byproducts to be rapidly swept out of the inter-electrode gap with a minimum of flow and replacing them with fresh, uncontaminated flowing insulating dielectric material. That is, the velocity of the dielectric media flow in the vicinity of the electrodes easily exceeds the expansion velocity of the debris cloud, thus sweeping the debris away from the electrodes and into a field-free region prior to the next charge cycle. The combination of elimination of the large gas bubble expansion, reabsorption of the micro-bubbles, and a smaller debris cloud means a dramatically lower flow rate may be used. This enables a switch to recover in less time when operated above a threshold pressure, thus enabling higher repetition rates and higher power operation. In addition, use of the present invention is advantageous over the existing art even in low power and/or low repetition rate applications. This is because of the much lower dielectric flow rate required to sweep bubbles and debris out from between the electrodes. Not only is this easier to implement, but also it enables the use of smaller, lower power, and lower weight pumps, which is especially advantageous for aerospace applications.
[0045] Operation of the switch of the present invention at high pressures preferably results in undersaturation of the flowing dielectric, providing an advantage over other switches known in the art. For de-aerated, pressurized liquid dielectrics, the gas desorbed by the arc breaks up into microbubbles and then is partially reabsorbed into the liquid dielectric on a millisecond time scale. The amount of gas desorbed is also significantly less in undersaturated solutions, facilitating much quicker voltage recovery of the switch. In contrast, for prior art switches which operate at or near atmospheric pressure, the liquid dielectrics are normally saturated with gas, which greatly lengthens the reabsorption time of the desorbed gas, and thus the recovery time of the switch, thereby lowering the achievable repetition rate.
[0046] Several electrode geometries, including but not limited to axial electrodes or radial electrodes, may be employed to optimize the flow, discharge, and electrode erosion properties of the switch. One preferred embodiment, an inline switch with radial insulator, is depicted in FIG. 1. A high voltage pulse is applied to input electrode 10, which is preferably supported by high voltage insulator 20, which is preferably designed to operate at high voltage and high pressures simultaneously. An electric field is generated between the input electrode 10 and output electrode 30. Electrodes 10, 30 are preferably substantially hemispherical. Switch cavity 40 is filled with liquid dielectric at a pressure significantly higher than atmospheric pressure, preferably between about 1000 and 2000 psig. The electric field causes the liquid dielectric between input electrode 10 and output electrode 30 to break down, enabling current to flow between the electrodes, thereby closing the switch. The current flow and resulting plasma causes the high pressure dielectric to form debris comprising carbon and other byproducts between the electrodes. Flowing dielectric enters the switch through inlet 50, preferably flows through hollow electrodes 10, 30 in the direction indicated by the arrow, and exits the switch via outlet 60. The dielectric preferably flows through a recirculating system (not pictured), more fully described below. This flow sweeps the debris out from the center of the electrodes. The breakdown voltage of the switch is determined by the electrode spacing, which is preferably adjusted by moving output electrode 30. Output electrode 30 preferably comprises a threaded assembly to facilitate this adjustment. Sight ports 70, 70' are preferably used to view and/or record the breakdown process and clearing time of the debris.
[0047] FIG. 2 depicts a switch of the present invention having coaxial switch geometry. High voltage is applied to input electrode 100, which are preferably designed to be replaceable. Output electrode 110 is isolated from input electrode 100 by the high pressure liquid dielectric in breakdown region 120, which is preferably contained by annular high pressure insulator 130. One or more field shapers 140 are preferably used to control the electric field distribution across insulator 130. Gap adjuster 150 is preferably threaded and is preferably used to adjust the gap spacing between input electrode 100 and output electrode 110. Micro-bubbles and breakdown contaminants in the liquid dielectric formed by the conducting plasma, which occurs during breakdown of the dielectric, are swept out of breakdown region 120 by the flowing dielectric.
[0048] Multiple dielectric flow paths may be utilized. Axial flow dielectric preferably enters the switch through axial flow inlet 160 and enters cavity 180 through input electrode channel 170. Cylindrical flow dielectric optionally enters the switch through cylindrical flow inlet 200 and enters cavity 180 through cylindrical flow channel 210. A plurality, preferably twelve, of cylindrical flow inlets 200 and cylindrical flow channels 210, preferably circumferentially arranged around the switch, and preferably evenly spaced, may be employed. The cylindrically flowing dielectric facilitates the removal from breakdown region 120 of the carbon and other breakdown byproducts. Dielectric from cavity 180 exits the switch through output electrode 110 via dielectric outlet 190. The axial flow topology may optionally operate in a "jet pump" mode, whereby the axial flow dielectric from entering cavity 180 from channel 170 has a high enough flow rate so that the dielectric media surrounding the electrodes, optionally partly comprising dielectric entering cavity 180 from cylindrical flow channels 210, is "pulled" radially into gap cavity 180 and exits axially through outlet 190. In the jet pump mode the flow may be provided by the main dielectric pumping system or alternatively by a separate, smaller system used solely for pumping the axial flow dielectric.
[0049] Fast switching times are accomplished preferably by utilizing outer coaxial return 220, which is substantially cylindrically disposed about the switch, thereby reducing the overall inductance of the switch. Outer coaxial return 220 is preferably connected via an electric load to the output electrode assembly.
[0050] As depicted in the above embodiments, the liquid dielectric media flow may enter the gap between the electrodes either radially or axially, or by a combination thereof, but preferably exits the gap axially in order to transport the discharge the byproduct debris cloud into the interior of the electrode, which is the nearest electric field-free region, prior to application of the next voltage pulse. The electrodes may be hollow or may optionally consist of machined electrodes that allow on axis flow and subsequent removal of the byproducts. Porous electrode surfaces may optionally be used to prevent boundary layers from forming.
[0051] In all embodiments of the present invention, the voltage breakdown of the system is preferably monitored in real time. In order to compensate for electrode erosion, the gap spacing of the electrodes is preferably adjusted to increase or decrease the breakdown voltage until the desired value is reached. The adjustment system preferably comprises a mechanical or electrical system, preferably comprising piezoelectric actuators, and optionally comprising a feedback system. The adjustment can alternatively be made manually. The electrodes preferably comprise a metal or other conducting material with low erosion rates, including but not limited to stainless steel, tungsten composites, tungsten-copper matrices, single crystal tungsten, and other synthetic materials that have a low erosion rate. Directed flow electrodes that allow the flow to be reduced and direct the byproducts into a field free region may alternatively be employed.
[0052] The flowing dielectric used in the high pressure switch of the present invention may comprise transformer oil, water, water-glycol mixtures, synthetic oils such as hydraulic fluid, or any other dielectric with desirable insulation characteristics which can be pressurized. Some transformer oils, which have been traditionally used for high voltage switches, are not compatible with high pressure, flowing pumping systems. Water and other natural or synthetic dielectrics also may freeze and require external heaters or separate pumping systems as well as additives such as antifreeze. The preferable use of synthetic lubricants in the present switch further enhances its capabilities because of these lubricants' greater voltage hold-off capability and reduced formation of byproducts. The tested performance of the synthetic fluids also increase the electric breakdown field of the switch allowing the electrode spacing to be reduced and lowering the inductance of the switch, the switch losses and the flow rate due to the reduced volume of liquid between the electrodes.
[0053] A preferred synthetic oil is polyalphaolefin (PAO), which has a higher flashpoint, is compatible with current airframe systems, and has a superior viscosity than that of transformer oil. The measured breakdown voltage, 1.1-1.25 MV/cm, of de-aerated PAO was found to be comparable or superior to that of transformer oils operated at pressures in the range of 1000-2000 psig. For various pulse charge times the breakdown field may be in the range of 200 kV/cm up to 10 MV/cm. The utilization of flowing dielectrics that are compatible with existing airframe and aerospace hydraulic systems (that is, fluids that are currently used as hydraulic fluid in airframe systems), including but not limited to PAO, is a novel aspect of the present invention. Such fluids have not been used as dielectrics in the past. The PAO solution both lubricates the hydraulic system and provides the dielectric strength required to hold off voltage and achieve the low inductance required by directed energy systems. In certain applications the switch may be directly integrated with the airframe hydraulic system, thus eliminating the need for a separate dielectric flow system, along with its attendant weight, complexity, and cost.
[0054] The switch or the present invention preferably utilizes electrode configurations which permit the control of the location of the discharge and the ability to move the discharge location to different areas on the electrode surface, thereby minimizing localized electrode erosion. These electrode configurations are also preferably optimized to minimize global dielectric media flow volume requirements while maximizing the flow velocity in the critical area of the electrodes, thereby rapidly sweeping the discharge byproducts into a field free region.
[0055] The switch also preferably comprises an integrated flow system and preferably utilizes a design prohibiting discharge byproducts from accessing and adhering to the switch housing solid insulators. The switch preferably is operated with a hydraulic recirculating dielectric media flow system, preferably comprising a pump for pressurization of the system and a reservoir. The pump may comprise an onboard hydraulic pump; alternatively an actuator may be used to pressurize the switch. The flow system preferably comprises one or more accumulators and particulate filters, which actively filter out carbon particle byproducts, including but not limited to micron sized particles, and allow continuous flow of the dielectric through the switch. The filters preferably include a particulate filter and/or a coalescing filter for removal of water and/or particles as desired. During normal operation of the switch, gases are introduced into the liquid dielectric by the arc and the subsequent arc byproducts. Therefore the liquid dielectric should preferably be de-aerated prior to and during operation of the switch. A de-aeration system, comprising one or more de-aeration stages, which consists of a vacuum pump and a reservoir is therefore preferably integrated into the pumping system. In one embodiment, the flow is preferably pulsed on and off to reduce the power required for the hydraulic system. This recirculating system may be integrated with the switch, or alternatively comprises either an existing on-board hydraulic system, such as that employed on an aircraft, or a stand alone pump cart or system.
[0056] In pulsed power systems, or in other systems where low jitter is required, a triggered switch may be used, whereby a high voltage or trigger pulse is applied to the switch and the switch self-breaks upon command. The switch may be triggered by application of a high voltage trigger pulse, a laser pulse, a microwave pulse, series injection, or other means that introduces UV, electron avalanches or bubbles into the electrode gap and results in the switch breaking down, thereby triggering the switch with low jitter. If an electrical pulse is used to trigger the switch the switch may incorporate a midplane or a third electrode, as in a trigatron. FIG. 16 shows trigatron 400, which is placed so that there is a gap between it and both electrodes. The trigatron may be used on either the cathode or anode side of the switch, depending on the switch polarity. The high voltage electrical pulse starts streamers in the gap and introduces UV radiation and an electron avalanche which triggers the switch. In this embodiment the gas bubble introduced into the electrode gap provides a dielectric mismatch and an ionization path in the high electric field which exists in the gap between the electrodes. The initiation of the avalanche and subsequent ionization of the dielectric triggers the switch. The operation of an untriggered switch has high jitter, because without a trigger breakdown may occur at one value in one instance, for example 100 kV, but another value (for example 105 kV) during a different shot. Or, if the voltage is held constant, the time that the switch triggers will vary. By triggering the switch, the exact voltage or time can be chosen, thus reducing jitter.
[0057] A pressure-induced triggering scheme may also be used. Increasing or decreasing the pressure in the switch housing preferably causes an increase and/or decrease in electrode gap spacing. For example, increasing fluid pressure expands insulator 130, which expands output electrode 110 thereby increasing the gap between the electrodes. So one method may be to increase the pressure, charge up switch, and remove the pressure, at which point the electrodes get closer, and switch closes (breakdown occurs). This technique can be used to trigger the switch as well as adjust the breakdown voltage of the gap due to the variation in gap spacing.
[0058] It may also be desirable for oil additives to be added to reduce the dielectric strength of the liquid, enabling larger electrode gap spacing and increasing voltage hold-off. In general, examples of these additives include but are not limited to nanoparticles, solids, liquids, or any additive that can influence the breakdown behavior of the high pressure liquid dielectric.
EXAMPLE 1
[0059] A test stand comprising a switch of the present invention was constructed that has an output impedance of 4.4 [Omega] and produces a 70 ns pulse. The switch was designed for the following requirements: switched voltage: 250-1000 kV; current: 50-250 kA; risetime: <50 ns; charge transfer: 0.5 Coulombs/pulse; switched energy: 250-1000 Joules per pulse; pressure: up to 3000 psig; jitter: <<50 ns; repetition rate: 50-150 pps; pulse width (duration): 50-500 ns; and lifetime: 10<7> -10<8 > pulses. These parameters were chosen because these are the requirements specified for potential directed energy systems.
[0060] The switch incorporated adjustable electrodes, allowing the electrode separation to be adjusted from 0.1 to 1 cm. Optical viewports were also integrated with the design allowing both the framing and high speed camera diagnostics to be integrated into the test stand, for characterization of bubble formation and byproduct expansion velocity. The single shot switch of the present example typically switched a 100 ns, 270-325 kV, 100 kA pulse into a 1.6 [Omega] load, with an energy per pulse delivered to the load of approximately 1 kJ. A graph of switching voltage vs. time for one experiment is shown in FIG. 3A. Both transformer oil and synthetic lubricants, such as PAO, were used in the experiments. For a 0.2 cm electrode gap the calculated arc inductance of the switch was 3 nH (15 nH/cm*0.2 cm). For a 1.6 [Omega] discharge load the 10-90% inductive rise time of the switch was 3.8 ns, while the calculated 10-90% risetime of the switch was 10-11 ns, which is an order of magnitude less than the risetime for the rest of the circuit (thus the switch was not the limiting factor). The electrodes comprised a copper tungsten composite (K3); however, any conductive material may be used. The electrodes had a diameter of 3.81 cm (1.5 inches), although other sizes and/or shapes may be used.
[0061] High speed optical diagnostics were used to observe the formation of bubbles and other byproducts. FIG. 3B graphs the radius of the region containing carbon byproducts vs. time after pulse at 2000 psig, showing that this region expands rapidly for about the first two milliseconds, with a modest expansion velocity of about 12.5 cm/s after that. This indicates that a 300 kV switch for use at or near this pressure, which can switch kilojoules of energy per pulse with a repetition rate of 100 pps, requires only a modest flow rate of 1-2 l/sec, which is almost a factor of 10 reduction from the atmospheric pressure switches known in the prior art. Concurrently the rise time of such a switch pulse charged in 1-1.2 microseconds will have a rise time of 10-11 nanoseconds or less, and allow kilojoules per pulse to be transferred at 100-200 pps. In addition, high speed photography showed that the bubbles and byproducts were swept out of the inter-electrode region in a short enough time to enable a repetition rate of at least hundreds of pulses per second. Thus the technology is scaleable to the goal of 1 MV and 100 pps operation, since only a modest 3-7 l/sec flow-rate will be required for such a switch.
[0062] Experiments were conducted from atmospheric pressure up to 13.8 MPa (2000 psig). The voltage breakdown of the switch versus pressure, for a 0.2 cm electrode gap, is shown in FIG. 4A. The data correspond to a breakdown electric field varying between 1.1-1.25 MV/cm. Each data point represents the statistical average of ten breakdowns using unconditioned electrodes. The variation, about +-10%, decreased to about +-6.5% after conditioning of the electrodes. The error bars represent one standard deviation for each data set at the test pressure. The curve fit shown is a second order polynomial least squares approximation. The data shown in FIG. 4A indicate that the breakdown strength increases by 25-30% from atmospheric pressure to 10.3 MPa (1500 psig). Thus the use of high pressures increases the switching voltage, in agreement with earlier experiments. In addition, there is indicated an optimal pressure for which maximum voltage breakdown occurs. The maximum bubble radius and bubble oscillation period vs. pressure are shown in FIGS. 4B and 4C, clearly demonstrating the advantage of operating the switch at high pressures.
EXAMPLE 2
[0063] An improved high power, high pressure flowing oil switch for gigawatt, repetitive applications was constructed and tested. The switch is of the present invention is typically operated at test pressures to 17.24 MPa (2500 psi), flow rates to 0.72 L.s<-1 > (11.4 gpm), charge voltages to -300 kV and discharge energies to 275 J per pulse at 20 pps. An examination of the electrodes after 250,000 shots with the original design led to the design of an insert device which resulted in higher performance fluid flow within the switch. The flow shaper-enhanced switch was tested for 150,000 shots.
[0064] Typical operating parameters are presented in Table 1. A cross section of the switch geometry is shown in FIG. 5. As illustrated in the drawing, oil flows around the cathode electrode and into a contoured anode throat section. The switch gap spacing is a function of the operating pressure and preferably increases with pressure. Gap spacing is preferably set to 1.02 mm, with an estimated error of less than +-15 [mu]m, while the switch is under atmospheric pressure.
TABLE 1
TEST STAND PARAMETERS
Charge Voltage -300 kV
Pulse Current 28.75 kA
PFL Impedance 4.8 [Omega]
Pulse Risetime 16 ns
Pulse Length 70 ns
Repetition Rate 20 pps
Test StandPulse Generator
[0065] The pulse generator used for testing the high pressure switch concept under repetitive pulse conditions is a 4.8 [Omega], 70 ns water pulse forming line (PFL). The water PFL was pulse charged to a maximum of -300 kV in 2.5 [mu]s through a pulse transformer. A capacitor-based pulse modulator was used to pulse charge the PFL. The modulator consisted of a hydrogen thyratron, a capacitor bank, and a snubber network as shown in FIG. 6, although other circuit configurations may be used. The capacitor bank was charged up to 26 kV, storing 273 J. Twelve 50 [Omega] cables 15.25 m in length were used to provide 70 ns of time-isolation between the PFL and the load resistor.
[0066] The charge and discharge voltages were monitored with a pair of D-dot probes. The probes were installed in the outer wall of the cylindrical metal structure that surrounds the switch. A liquid tight fit was made via Swagelok fittings and the output was fed into a passive integrator. A typical output voltage waveform is presented in FIG. 7 which shows a 10-90 rise time of about 16 ns. Hydraulic Power Supply
[0067] Hydraulic power was provided by a portable hydraulic pumping unit. The pump utilized generated flow rates up to 0.72 L.s<-1 > at pressures in excess of 17.24 MPa. The pump had an adjustable stroke compensation which was used to adjust volumetric flow at various pressures. The oil used for test purposes was an electrically insulating, thermally conductive, synthetic olefin-based liquid (PAO).
[0068] Fluid diagnostics used on the test stand included a pair of analog pressure gages and various flow rate sensors. Pressure in and across the switch was monitored with a pair of ENFM bourdon-tube pressure gages that had pressure resolution to 10 psig. Both the inlet pressure and the outlet pressure were recorded, however only the inlet pressure is reported herein. Overall volumetric flow rate was monitored in the oil return line with a viscosity compensated turbine sensor manufactured by Cox Instruments. A pair of Hedland variable area flow sensors monitored relative flow rates in the two oil lines that fed the high pressure switch.
[0069] High performance filters were installed to increase the cleanliness of the oil both entering and leaving the switch. The filter elements are rated to retain more than 99.9% of particles 0.45 [mu]m or larger that are suspended in the oil passed through the media. Test Results
[0070] The high pressure switch was tested under both single shot and repetitive conditions over a range of pressures, flow rates and temperatures. The single shot work examined the statistical nature of breakdown voltage, electric field strength, and jitter under typical test conditions.
[0071] The high pressure switch geometry is a pin in hole type geometry as indicated in FIG. 5. In the original geometry oil is forced to flow around the cathode and down through the center of the anode. In the present test a flow shaping element was introduced that reduced the cross sectional area of the oil path, thereby reducing turbulence and effectively eliminating eddy flows near the stressed region of the switch. FIG. 8 shows a cross section and an end view of the flow shaping element or flow straightener. As shown in FIG. 9, flow shaping element 500 is preferably disposed approximately cylindrically around input electrode 100. The flow straightener preferably forces the flow to equalize across the switch volume, preventing circulation of the fluid azimuthally and radially. By shaping the flow cross section, the flow shaper also controls the velocity of the liquid dielectric in the switch. The addition of two flow sections in the switch produces a venturi or jet pump on-axis pumping action to break up boundary layers or stagnation layers that are inherent to flowing geometries. Flow through the on-axis nozzle or orifice can be controlled separately or independently from the flow through the shell of the switch. Use of this flow element enables high repetition rates for the switch.
[0072] The gap spacing was generally set at 1.016 mm and the electrodes had a pressure dependent field enhancement factor of about 11.7 at 13.79 MPa and 11.0 at 17.24 MPa. The peak field stress expected at 250 kV and 13.79 MPa was approximately 2.3 MV.cm<-1> . Pressure drop across the switch varied, depending on the flow rate, from 69 kPa (10 psi) at the lowest flow rates to 207 kPa (30 psi) at the highest flow rates. Single Shot Tests
[0073] An analysis of the single shot switch performance was undertaken to define a hold-off strength for oil under test conditions. Tests were performed at six pressures and two volumetric flow rates. At each combination of pressure and flow a sample of 50 shots was recorded, each shot separated in time by greater than 45 seconds. The tests were performed using 304 stainless steel electrodes following 140,000 shots of electrode break in and conditioning. The D-dot probe adjacent to the charge electrode was used to record the waveforms. Post-processing was performed to reconstruct the actual charge waveform and generate an electric field strength value.
[0074] Plots of the electric field strength at breakdown as a function of pressure for two volumetric flow rates are shown in FIGS. 10 and 11. The graphs show a solid line representing the linear least-squares fit to the data. The dashed line represents the boundaries of a 90% confidence interval of the data at each pressure. The small slope of the solid lines combined with the relative width of the confidence intervals suggest that the electric field strength at breakdown is not strongly affected by pressure over the range of pressures reported.
[0075] Breakdown jitter was noted earlier to be an important performance parameter for the high pressure switch of the present invention. The data plotted in FIG. 10 shows no strong correlation between jitter over the range of pressures reported. The 1 [sigma] jitter for the data reported is +-9.7% at 13.79 MPa and +-10.0% at 17.24 MPa. Examination of the 90% confidence interval data in FIG. 11 does show some correlation between breakdown jitter and pressure, with jitter decreasing as a function of pressure. The 1 [sigma] jitter is +-11.4% at 13.79 MPa and +-8.1% at 17.24 MPa. These results indicate a reduction of around +-3.3% over the range of pressures examined. Repetition Rate Tests
[0076] The switch was tested under repetitive conditions for several hundred thousand shots. The repetition rate tests were conducted to establish the relationship between oil pressure, volumetric flow rate, breakdown hold-off jitter and recovery, and electrode wear. Waveforms were photographed to obtain information about total jitter and mean breakdown electric field strength under various rep rates. FIG. 12 shows the results of operation at 1 pps and at constant pressure, constant flow rate and constant oil temperature. The time jitter in the figure, measured from the leading edge of the pulse train to the trailing edge of the pulse train, is approximately 125 ns or 5% of the time to peak. Qualitative analysis of the repetition rate data over 1000 shot bursts indicate that jitter is within the same order of magnitude as the single shot jitter. The switch has been tested at up to 22 pps and the results are nearly identical to single shot and 1 pps results.
[0077] Electrode lifetime tests were conducted under rep rate conditions. Oil pressure was generally kept around 13.79 MPa (2000 psig), flow rates varied from 0.379 L.s<-1> to 0.681 L.s<-1 > and temperatures were kept between 18[deg.] C. and 32[deg.] C. Tests were performed at repetition rates between 1 pps and 20 pps, with the majority of the tests taking place at 15 pps and in 1000 shot bursts.
[0078] Repetition rate testing was performed with two distinctly different fluid path designs. The original switch concept had a geometry that resembled that shown in FIG. 5. In the interest of improved switching performance a second design was implemented to modify the oil path. The design was meant to reduce random swirling that was predicted by a computational fluid dynamics simulation of the original geometry. The flow shaper design featured a constrained oil path with less cross sectional area to increase the average oil velocity and vanes in the fluid path to establish a swirl-free velocity profile within the electrically stressed regions of the switch.
[0079] The original switch design was run with 304 stainless steel electrodes. More than 250,000 shots were taken in the course of this first test series. When the switch was removed from the test stand the discharge pattern was recorded. The discharge is supposed to occur within a band on the side of the pin electrode. The arc sites were uniformly distributed over a band that was approximately 1.7 cm wide. A photograph of the discharge band is shown in FIG. 13.
[0080] A set of K-33 sintered copper-tungsten electrodes was fabricated and installed in the hopes of increased performance over the stainless steel electrodes. In addition to the electrode material change the flow shaper was installed. After less than 4000 shots under rep rate conditions rapid increases in system pressure were observed during operation. The switch was removed and the electrodes were inspected. The wear pattern was noted to be very localized, showing signs consistent with spallation. The pressure variations observed during operation were apparently a result of pieces of the electrode getting caught in a down-stream needle valve. A photograph of the damage is shown in FIG. 14.
[0081] A second set of 304 stainless steel electrodes were installed with the flow shaper and lifetime tests were restarted. After approximately 140,000 shots the single shot tests already reported were performed. Following the single shot tests the electrodes were removed. The wear pattern that had developed was more distinct and defined than that seen on the first 304 stainless steel electrodes. The width of the discharge band had decreased by 41% to about 1.0 cm. A photograph of the wear pattern is shown in FIG. 15. These tests indicate that the electrode lifetime may be greater than 10<7 > shots.
[0082] Throughout all of the tests described herein the dielectric oil was not flushed and replenished. Water content was expected to be around 200 to 300 ppm since the oil reservoir is open to the atmosphere and tests were performed in a region notorious for high humidity. Analysis unexpectedly revealed that water content was around 13 ppm. Thus during operation, the switch of the present invention removes water from the oil or hydrocarbon fluid used due to the high energy plasma arc which is produced. The present invention can therefore be used to remove water from the hydraulic fluid, PAO, or flowing dielectric with out chemically changing the oil. Further, this method can be used to remove water from such fluids, or related fluids, even when the use of a high power switch is not required. One example of such use is in an aircraft, where the fluid comprises hydraulic fluid. If water exists in any appreciable concentration in the fluid, corrosion of the hydraulic components can occur. The present invention can prevent that from happening.
[0083] As noted in Leckbee et al's 2004 paper, significant quantities of carbon are generated during discharge under high pressure. Following the first 250,000 shots the oil return line filter was examined and found to be jet black from all of the carbon deposits. Based on this examination the filter was changed between the first 304 stainless steel electrode tests and the K-33 electrode tests.
[0084] The preceding examples can be repeated with similar success by substituting the generically or specifically described operating conditions of this invention for those used in the preceding components.
[0085] Although the invention has been described in detail with particular reference to these preferred embodiments, other embodiments can achieve the same results. Variations and modifications of the present invention will be obvious to those skilled in the art and it is intended to cover in the appended claims all such modifications and equivalents. The entire disclosures of all references, applications, patents, and publications cited above are hereby incorporated by reference.