WO1999037581A2

WO1999037581A2 - High efficiency glow discharge gaseous processing system for hydrogen peroxide production and other chemical processing of gases

Info

Publication number: WO1999037581A2
Application number: PCT/US1999/000911
Authority: WO
Inventors: William M. Moeny; David M. Barrett; Robert W. Bennett
Original assignee: Moeny William M; Barrett David M; Bennett Robert W
Priority date: 1998-01-22
Filing date: 1999-01-15
Publication date: 1999-07-29
Also published as: WO1999037581A3; AU2230999A

Abstract

A high efficiency plasma processing system for use in the generation of large quantities of gases, particularly hydrogen peroxide or ozone, for chemical commercial and industrial applications is based upon a homogeneous electrical glow discharge. In addition to the homogenous discharge, the high efficiency of ozone or other chemical production is achieved by exciting the discharge with high voltage pulses which have a voltage rate-of-rise in excess of 500 V/ns and incorporating an internal gas circulation system which maintains the temperature of the process gas to a level where the thermal decomposition of ozone, for example, is significantly reduced. A homogeneous and stable atmospheric glow discharge is developed by preionizing the discharge region with ultra-violet radiation, for example, immediately before the application of the main high voltage pulse across the electrodes of the discharge cell.

Description

HIGH EFFICIENCY GLOW DISCHARGE

GASEOUS PROCESSING SYSTEM FOR

HYDROGEN PEROXIDE PRODUCTION

AND OTHER CHEMICAL PROCESSING OF GASES

RELATED APPLICATIONS

This application is a continuation-in-part of application Serial No. 09/057,377, entitled High Efficiency Glow Discharge Gaseous Processing System for Ozone Production and Other Chemical Processing of Gases, filed on April 8,

1998, which claimed the benefit of the filing date of provisional application Serial No. 60/042,721 , also entitled High Efficiency Glow Discharge Gaseous Processing System for Ozone Production and Other Chemical Processing of Gases, filed April 8, 1997, the entire disclosures of which are hereby incorporated by reference. This application also claims the benefit of the filing of provisional patent application Serial No. 60/072,21 1 , entitled Glow Discharge Production of Hydrogen Peroxide and Other Chemicals, filed January 22, 1998, the entire disclosure of which also is incorporated herein by reference.

BACKGROUND OF THE INVENTION

(a) Field of the Invention

The invention relates generally to glow discharge methods and apparatus for processing gases into useful products. More particularly, it relates to a glow discharge apparatus having magnetic switching to provide fast rise-time voltage pulses, and employing radial discharge and high pressure in the production of commercially valuable product gases such as ozone and hydrogen peroxide. (b) Description of Related Art

Glow discharges provide a convenient means for processing gases or mixtures of gases to produce new compounds. The ability of the glow discharge to initiate specific chemical reactions, due to the control of the electron energy distribution in the weakly ionized plasma, is a major advantage over conventional thermal discharges. A glow discharge chemical processor can achieve many of the effects of thermal processing but does so by heating only the electron swarm and not the complete gas mixture.

Glow discharges comprised of weekly ionized plasmas are widely used for pumping pulsed C0₂ lasers and excimer lasers, for example. Glow discharges differ from other types of discharges in that the discharge is diffuse. Namely, the current in the discharge is uniformly distributed throughout a volume of gas to be processed or excited. The discharges are weakly ionized with typical ionization fractions of 0.1 to 1 PPM. Because the gas is very weakly ionized, almost all of the collisions by electrons will be with neutrals. As a result, the characteristics of the electron swarm are dominated by these collisions. Also, the average energy of the electrons is typically quite low, on the order of 1 electron volt. For many laser applications, glow discharges typically operate at atmospheric pressure, however, some operate at pressures as low as 0.1 atmospheres and others operate at pressures as high as 5 atmospheres. The free electrons and ions that conduct the current are created from within the discharge, either by an external source or by the discharge itself. The distinction between these two types of discharges, self-sustained and externally sustained, is important to the present invention. An externally-sustained discharge is much easier to operate because it can be kept much more stable.

A glow discharge is very complex, with hundreds of plasma chemistry interactions occurring simultaneously. The interactions can be grouped into four types of processes. The first type is electron attachment. As the electrons are driven through the gas by the electric field, an electron can be captured by a constituent of the gas. For example, if there is oxygen in the gas mixture, an oxygen molecule might capture an electron in the reaction 0₂ + e^" > 0₂ ^' . The rate of attachment is proportional to the number of electrons and is characterized by the term "b," which is the number of attachments per second per electron. As expressed in Equation 1 below, the loss of electrons due to attachment is equal to the electron number density times the number of attachments per electron per second.

Equation 1 dn_β/dt = -bn„

The second type of process occurring in the gas mixture is recombination, where an electron and an ion recombine. Recombination is characterized by the term "g," which is the number of recombinations that occur per second per electron per ion. Notably, in these weakly ionized discharges the ion number density is approximately the same as the electron number density. Thus, as expressed in Equation 2 below, the number of electrons lost per second is equal to the electron number density squared times the recombination coefficient:

Equation 2 dn„/dt = -gn_e ²

The third type of process occurring in the gas mixture is ionization. In this phenomena, an electron collides with and ionizes a neutral, thereby creating another free electron. The coefficient "a" for this process is the number of electrons created per second per electron. Equation 3 below expresses the ionization phenomena as a function of "a" :

Equation 3 dn_β/dt = an_β

The fourth type of process ocurring in the gas mixture is that an electron will collide with a neutral and will give up some energy to the neutral, thereby changing the energy state of the neutral, but neither ionizes the neutral nor is attached to it (recombination only occurs with ions). The electron number density changes with time according to whether the ionization, attachment, or recombination process dominates. This is described by Equation 4 below:

Equation 4 dn„/dt = an„ - bn„ - gn_β ²

Equation 4 can be used to describe two regimes of discharge operation. For most discharges of interest to chemical processing and gaseous lasers, recombination is not a significant process because the attachment rate is very high and dominates the discharge loss processes (for discharges operating near 1 atmosphere). In this case, whether the electron density increases or decreases depends upon whether the ionization coefficient, a, is greater than or less than the attachment coefficient, b. FIG. 1 is a plot of the relative attachment and ionization coefficients for a typical pulsed C0₂ laser mixture. As FIG. 1 shows, at a lower normalized electric field (the electric field normalized to the neutral number density, i.e., essentially the electric field per molecule), the attachment rate is much higher than the ionization rate. This is intuitively logical because a slow moving electron has a greater chance of being captured by a neutral than does a fast moving electron. Conversely, at high electric fields, the ionization rate is much higher than the attachment rate. This also is intuitive because the electron must be moving fast to release another electron from a neutral.

If a particular location in the discharge begins to avalanche, that is, the ionization coefficient becomes locally significantly larger than the attachment coefficient, then the number density of electrons will rise exponentially and the current will concentrate into one location. This creates an arc or spark in the discharge and the diffuse glow collapses into a single channel, causing the voltage across the discharge to decrease. This is commonly referred to as discharge collapse. The formation of such an arc terminates the diffuse nature of the discharge and terminates its usefulness. Discharge collapse is the principle phenomena leading to limitation on the amount of energy that can be put into the gas by the glow discharge. As mentioned, there are two primary regimes of operation of a glow discharge: externally-sustained and self-sustained. The externally-sustained discharge is the electric field regime where the attachment coefficient is larger than the ionization coefficient. As shown from Equation 4, without the external supply, the electron number density will continually decay and eventually, on a time scale set by the difference between the attachment and the ionization coefficients, the discharge will extinguish through lack of electrons. In an externally-sustained discharge, the external sustainer device makes up the electrons lost to attachment. An electron beam gun might be utilized to inject high energy electrons into the discharge region, as shown, for example, in U.S. Patent

No. 3,883,413. The high energy electrons from the gun collide with gas particles and produce low energy free electrons, "S" per second per cubic centimeter. The electric field then causes the low energy electrons to flow through the gas mixture producing the discharge.

Equation 5 dn„/dt = S + an_β - bn_β - gn 2

In a self-sustained discharge, the electrons are provided by the ionization process in the discharge. The electrons avalanche to a steady-state condition where the ionization and attachment rates are equal. Thus, a self-sustained discharge operates at a particular electric field where the attachment and ionization rates are equal. This electric field is commonly called the glow field or glow point. At the glow field, the electron number density is neither increasing nor decreasing. For a discharge to successfully operate at this point requires an external circuit control of some type, such as a pulse forming network. It is very difficult to operate a glow discharge in the self-sustained mode because the electric fields have to be very uniform and the discharge has to be uniformly initiated for it to operate without collapsing into an arc.

The foregoing explains the difference between the externally-sustained discharge and the self-sustained discharge. The externally-sustained discharge is easier to realize in practice because the electron number density is controlled by the external discharge sustainer device, such as an electron beam gun, and thus electron number density can be controlled somewhat independently of the electric field. The electric field can be kept well below the glow point, reducing the chance for discharge collapse due to avalanche runaway. The self-sustained discharge is much more difficult to operate because it is a balancing act; the discharge being balanced very carefully between excessive ionization and excessive attachment. It is especially difficult to do with a highly attaching gas mixture.

Externally-sustained and self-sustained discharges have traditionally been utilized for pumping molecular gas lasers such as pulsed electric C0₂ lasers and excimer lasers. With few exceptions, the self-sustained devices have all utilized linear electrodes, that is, where the discharge region is formed between two plane parallel, long, narrow electrodes to provide a long gain-path for the laser. The C0₂ laser gas mixture is typically composed of carbon dioxide, nitrogen, and helium. There are no strong attachers in this gas mixture. Self-sustained discharges in these mixtures have proven to be fairly reliable and commercially are used extensively. The introduction of a small amount of oxygen, however, can create significant discharge instability with these devices because of the high attachment rate of oxygen. Excimer lasers that utilize small quantities of halogens, such fluorine or chlorine have also been successfully operated using either externally-sustained or self-sustained discharges. These discharges typically operate with very short pulses and the total amount of halogen present is kept very small because the high attachment rate of the halogens tends to cause the discharge to be unstable, just as oxygen does.

It is possible to operate a glow discharge in a non-planer mode if the gas mixtures are fairly benign and thus, less susceptible to discharge instability. In a radial discharge mode, the electrodes are two concentric cylinders with the glow discharge operating between the cylinders. The discharge is termed "radial" because the electron flow is radial from one electrode to another. It is also known to make a radial glow discharge with an externally sustained discharge. Again, in known systems an externally sustained discharge provides a means of improving discharge stability. The use of a gas with very mild attaching characteristics, such as the pulsed C0₂ laser gas mixture, enhances stability in the radial discharge configuration. Radial glow discharge devices have been built, for example, at Los Alamos National Laboratory for the Antares fusion program; these devices utilized electron beam guns to sustain the discharge.

Ozone is an unstable triatomic compound of oxygen and is used in a number of applications, such as water treatment, industrial effluent treatment, and industrial bleaching operations. Since ozone is an unstable compound, decomposing back to diatomic oxygen after a short while, it is necessary to generate ozone at its point of use. Ozone is formed by using an electrical discharge to dissociate diatomic oxygen molecules into single atoms, which subsequently combine with other diatomic oxygen molecules to form ozone. In most conventional ozone generators, commonly referred to as ozonators, the electrical discharge is developed in a dielectric barrier discharge, also referred to as a silent discharge configuration. A silent discharge can be configured in a coaxial or planar geometry and typically consists of a pair of electrodes that are excited with a relatively high voltage alternating current waveform. One or perhaps both electrodes are covered with a dielectric material such as glass or ceramic. Typically, a sinusoidal or bipolar pulsed high voltage waveform is impressed across the electrode pair to generate an electric field in the discharge gap of the silent discharge configuration. When the breakdown electric field of the gas molecules between the electrodes is reached, the gas becomes partially ionized. Specifically, the ionization of the gas proceeds as a multiplicity of microdischarges are formed between the dielectric barrier and an electrode. As each microdischarge is formed, it conducts electrical charge through the gas volume that accumulates on the surface of the dielectric barrier. As charge continues to be conducted through the gas volume, the charge accumulates on the surface of the dielectric barrier and reduces the local electric field. As a result, the microdischarge is extinguished and another microdischarge is developed at another location. Thus, the multiplicity of microdischarges are semi-uniformly distributed throughout the gas volume.

Each microdischarge serves as a source of energetic electrons, which are necessary for the production of ozone. Electrons having sufficient energy ( > 6 eV), collide with diatomic oxygen molecules causing them to dissociate and form singlet oxygen atoms. The singlet atoms subsequently combine with other diatomic oxygen molecules to form ozone.

When a conventional silent discharge configuration is employed for the manufacture of ozone, the efficiency at which ozone is produced in practice is limited to approximately 12%. There are a number of factors that contribute to this low efficiency. First, in accordance with the above discussion, the discharge in a silent discharge configuration is formed as a group of individual microdischarges that are semi-uniformly distributed throughout the gas volume. As a result, the gas volume that is actually exposed to the electrical discharge at any given time is a fraction of the total gas volume that occupies the discharge region. Therefore, the dynamics necessary for the formation of ozone are restricted to a small portion of the total gas volume. Another factor contributing to the relatively low ozone production efficiency of silent discharges is that they are typically excited with high voltage waveforms that have a relatively low voltage- rate-of-rise (e.g., < 100 V/ns). At these values, the ignition voltage of the discharge gap is relatively low, resulting in the production of electrons having an average energy of less than 6 eV. Therefore, the majority of electrons do not attain the energy needed to dissociate diatomic oxygen. In addition, the ozone production efficiency of conventional silent discharge configurations is limited by the poor heat transfer characteristics of the geometry. Specifically, the rate at which ozone dissociates is accelerated as the temperature of the process gas is increased. In a conventional silent discharge configuration, the electrodes are typically water-cooled; however, the flow rate of the process gas through the discharge cell is relatively low. Therefore, because the principal heat transfer mechanism between the process gas and the wall-cooled electrode is conduction, a large electrode area is required to provide the required heat transfer. It has been determined that ozone can be generated with efficiencies exceeding 25%. Significantly improved ozone generation efficiencies may be achieved by exciting an electrical discharge with high voltage pulses having a voltage rate-of-rise in excess of 500 V/ns. In addition, using high voltage pulses having such a voltage-rate-of-rise enables ozone to be generated in a bulk glow discharge configuration, which does not require a dielectric barrier. As a result, a much higher discharge power density can be achieved so that for a given ozone production rate, a smaller discharge cell can be used. Although a number of experiments have demonstrated that it is possible to significantly increase the efficiency of ozone generation, the principal reason that such techniques have not been incorporated in the design of commercial ozonators is that, traditionally, it has been difficult to provide reliable and inexpensive sources of fast rise-time, short duration high voltage pulses. In the experiments referenced here, the high voltage pulses were generated with pulse forming networks that used spark gap switches or rotating high voltage switching devices. Although such switching devices proved to be adequate for experimental purposes, they are not practical in many industrial and commercial applications. Specifically, such switching devices suffer from electrode erosion problems, which seriously limits their reliability, durability, and economy. The severity of these problems increase significantly as the average power of the pulse generator increases over several kilowatts. Consequently, it is technically difficult and not cost effective to scale such switches to the power levels required for industrial ozonator equipment.

Some high performance pulsed C0₂ laser devices in use employ trace hydrocarbon additives to enhance discharge stability. As stated above, one of the principle requirements for successful initiation and control of a self-sustained pulsed discharge is uniform initiation of the discharge. This requires the creation of an initial background of electrons uniformly distributed across the discharge region and of a number density well above any non-uniformities caused by stray cosmic rays or other ionizing sources. By putting hydrocarbon photoelectron seedants into the gas mixture, it is possible to create a much higher density of photoelectrons than would be created without seedants. Seedants also can be used to depress the glow electric field by increasing the ionization rate at a specific electric field. The equilibrium point where the ionization rate matches the attachment rate can be reduced, thereby increasing discharge stability.

U.S. Patent No. 5,554,345, entitled "Ozone Generation Apparatus and Method" to Kitchenman, O. S. G., discloses a dielectric barrier apparatus for the generation of ozone that is an improvement upon the conventional dielectric barrier ozone generator. Kitchenman's system uses a conventional electrical supply having a frequency of at least 20 kHz. Kitchenman does not teach the use of recirculating the ozone to increase concentration while decreasing temperature rise and does not teach the use of magnetic switching or UV preionization for the bulk discharge for the generation of ozone. Furthermore, Kitchenman does not disclose the use of a bulk glow discharge for ozone generation.

U.S. Patent No. 5,370,846, entitled "Apparatus and Method for Generating High Concentration Ozone," to Yokomi, et al., shows an apparatus and method for diluting pure oxygen with another gas such as a noble gas, carbon dioxide, or nitrogen to improve the concentration of ozone. This method does not teach the use of magnetic switches for controlling the discharge and does not teach the use of a bulk glow discharge, but rather, teaches the use of a dielectric barrier discharge. Furthermore, Yokomi does not teach the use of recirculating the gas to control temperature nor does it teach the use of magnetic switches for generating the pulse to the electrodes.

U.S. Patent No. 3,883,413, entitled "Ozone Generator Using Pulsed Electron Beam and Decaying Electric Field" to Douglas-Hamilton, D. H., shows a system to generate ozone by irradiating oxygen with brief bursts of high-energy electrons from an electron beam to produce secondary electrons, which are accelerated in an electric field to form a glow discharge. The glow discharge is externally sustained. The electric field is pulsed on during the electron beam irradiation and thereafter pulsed off to maximize energy deposition by electrons and minimize energy deposition by ions. Douglas-Hamilton does not teach the use of UV preionization or any type of preionization since the electron beam provides all ionization. It does not teach the use of magnetic switches for the discharge, speaking only of producing a controllable electric field. Finally, Douglas-Hamilton does not provide a means for recirculating the gas to provide cooling. Dermott- Hamilton suggests the use of the externally-sustained discharge to form hydrogen peroxide. U.S. Patent No. 5, 105,097, entitled "Passive Magnetic Switch for Erecting

Multiple-Stage High-Pulse-Rate Voltage Multipliers" to Rothe, D. E., discloses the use of a voltage multiplication system utilizing a multiple-stage Marx bank voltage multiplier circuit for generating high output voltage pulses. This system utilizes magnetic switches for each stage of the Marx bank voltage multiplier. The approach used is quite different than the magnetic switching approach of the present invention. The present invention's magnetic modulator does not utilize the Marx bank approach of the patent of Rothe.

U.S. Patent No. 4,908,524, entitled "High Voltage Pulsed Power Converter" to Sojka, R. J., shows a high voltage pulsed power converter comprising a modified type E pulse-charging network in series with a non-linear magnetic switch. The present invention does not utilize an E-type pulse forming line for controlling the shape of the output pulse.

U.S. Patent No. 3,856,671 , entitled "Closed Loop Ozone Generating and Contacting System" by Lee, H., et al., indicates the use of a closed loop system for treating water with ozone and/or oxygen, thus conserving the high oxygen content gas. The system comprises a method of utilizing oxygen-enriched gas contacted with water to be treated. The excess oxygen and/or oxygen and nitrogen forms a recycled gas which is combined with a charge of air and flowed to another apparatus for recirculating the oxygen for subsequent use. The present invention utilizes a recirculation system for controlling the temperature rise in oxygen containing gaseous mixtures or other chemical mixtures that are being processed by a glow discharge.

In a paper presented on January 17, 1989 at a conference of the International Society for Optical Engineering (SPIE), entitled "High Pressure Pulsed Radial Glow Discharge C0₂ Laser," author W.M. Moeny, describes a pulsed C0₂ laser pumped by a UV-preionized glow discharge. The article does not disclose any chemical gas processing.

In February 1997, two papers were published describing a pulsed glow discharge for the production of ozone: "High-Pressure Pulsed Avalanche Discharges: Formulas for Required Preionization Density and Rate of

Homogeneity", by Nils Brenning, et al., IEEE Transactions of Plasma Science, Vol 25, No. 1 , February 1997, and "Numerical Modeling of Ozone Production in a Pulsed Homogeneous Discharge: A Parameter Study" by J. Olof Nilsson, et al, IEEE Transactions on Plasma Science, Vol 25, No. 1 , February 1997. Both of these papers describe the use of the volumetric pulsed glow discharge in oxygen as a method to produce ozone. Yamabe also describes using a double discharge method in mixtures of pure oxygen and oxygen and helium at atmospheric pressure to produce ozone. "Ozone Generation in 0₂/N₂ Mixtures Using Double Discharge in Atmospheric Diffuse Glow Discharge", Chobei Yamabe, et al., Proc. Japan. Symp. Plasma Chem, Vol 1 , 1988.

Thus, a need remains for a means and method for creating a stable, self- sustained, glow discharge in highly attaching gas mixtures, such as mixtures with large quantities of oxygen, for the production of commercially valuable products such as ozone and hydrogen peroxide. Against the foregoing background, the present invention was developed. Through the use of unique preionization techniques and careful electric field management, the present invention creates self-sustained discharges in highly attaching mixtures.

SUMMARY OF THE INVENTION The present invention overcomes the deficiencies associated with the prior art of gaseous chemical processing by providing a novel homogeneous bulk glow discharge apparatus and method.

In accordance with one aspect of the present invention, an apparatus for converting a process gas into a product gas includes anode and cathode electrodes that define a discharge volume between the electrodes such that the discharge volume contains at least some of the process gas. A magnetic pulse compression network, having at least one magnetic switch, is coupled to the electrodes and provides fast rise-time high voltage pulses to the electrodes and the discharge volume. The fast rise-time high voltage pulses, when impressed across the discharge volume, produce a self-sustaining glow discharge that converts at least some of the process gas into the product gas.

In some embodiments, the magnetic pulse compression network includes a coupled network of magnetic switches and charge storage elements that are arranged in stages so that the magnetic switches couple between the charge storage elements of the stages. Each of the stages receives a voltage pulse having a first rise-time and produces an output pulse having a second rise-time less than the first rise-time.

In other embodiments, the magnetic pulse compression network is adapted to impress a series of fast rise-time high voltage pulses across the discharge volume such that each successive pulse from the series of voltage pulses has less energy density than the previous pulse.

In other embodiments, a pre-ionizer pre-ionizes the process gas. In still other embodiments, a gas circulation system conveys process gas to the discharge volume and removes product gas from the discharge volume. A heat exchanger is preferably included to remove heat from the process gas and the product gas.

In accordance with another aspect of the present invention, a method of converting a process gas into a product gas conveys the process gas into a discharge volume that is defined between a cathode electrode and an anode electrode. The process gas within the discharge volume is pre-ionized. A magnetic modulator network having at least one magnetic switch generates a fast rise-time high voltage pulse and couples it to the electrodes. A self-sustaining glow discharge is produced in the discharge volume and at least some of the process gas is converted into the product gas.

In some embodiments, the process gas and the product gas are circulated through a heat exchanger. In other embodiments, the discharge volume is pressurized to a pressure greater than atmospheric pressure.

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

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:

FIG. 1 is a graphical representation of the known attachment and ionization rate coefficients divided by the electron drift velocity for a glow discharge in 0₂;

FIG. 2 is a block diagram of a high efficiency bulk discharge processing system according to one embodiment of the invention;

FIG. 3 is a graph depicting the temporal relationships between the preionizer voltage pulse, the main discharge voltage pulse and the electron number density (n_e) of the discharge according to the invention;

FIG. 4A is a top sectional view of the discharge assembly according to an embodiment of the invention including a plurality of rectangular ceramic blocks to form the dielectric barrier for the preionization subassembly;

FIG. 4B is a partial sectional view, taken along line X--X in Fig. 4A, of the embodiment shown in Fig. 4A, with a portion broken away to reveal certain internal components; FIG. 5A is a top sectional view of another embodiment of the invention, showing a discharge assembly incorporating a single piece of ceramic material to form the dielectric barrier for the preionization subassembly;

FIG. 5B is a partial sectional view, taken along line Y--Y of Fig. 5A, of the embodiment shown in Fig. 5A;

FIG. 6 is a diagram of the inventive apparatus with its accompanying process gas circulation system;

FIG. 7 is a cross sectional view of an alternative radial discharge embodiment of the invention, showing a concentrically arranged discharge electrode array and preionizers;

FIG. 8 is a is an axial sectional diagram of a portion of the embodiment shown in Fig. 7, illustrating the preionizers in a position transverse to the sides of the discharge volume;

FIG. 9 is a radial sectional view of the embodiment shown in Fig. 7, illustrating a radially directed gas flow;

FIG. 10 is diagram of a system, according to the invention, for the production of hydrogen peroxide;

FIG. 1 1 is a schematic drawing of a multistage magnetic pulse compression network used to supply fast rise-time, short duration, high voltage pulses to the discharge processing apparatus of the invention;

FIG. 12 is a graphical depiction of the voltage waveforms on each capacitor in the solid state switched magnetic modulator network, as well as the main high voltage pulse which is impressed across the discharge cell according to the present invention; and FIG. 13 is a block diagram of a timing and triggering circuit according to the invention.

DETAILED DESCRIPTION OF THE PRESENT INVENTION The present invention relates to an apparatus and method for efficiently processing a process gas, such as oxygen, into a product gas or compound, for example ozone, by a homogeneous electrical glow discharge. The glow discharge may be operated at about one atmosphere to produce ozone, for example. The present invention also permits the practical implementation of an above- atmospheric pressure glow discharge configuration for the production, from an oxygen/water vapor process gas mixture, of hydrogen peroxide at high efficiencies. This disclosure describes a new approach to glow discharge processing of gases for a multitude of purposes, while at the same time carefully controlling the temperature of the gases in order to control the overall process rates and functions. In this specification, the conversion of oxygen to ozone is used as an example of chemical process enabled and/or facilitated by this invention. Most of the principal features of the invention can be described by considering the oxygen to ozone conversion as an example of the inventive process. However, the present invention may be used to produce a wide variety of commercially useful gases and chemical products, such as hydrogen peroxide from water vapor and oxygen, for example. Broadly described, one embodiment of the invention includes a discharge assembly or cell, principally including a generally planar cathode electrode and a substantially planar anode electrode. A preionization assembly is connected to the discharge cell, and features an auxiliary electrode and a dielectric barrier proximate to the anode electrode of the discharge cell. Preionization is performed by providing ultraviolet (UV) light or radiation into the glow discharge volume. A power source, such as a DC source, powers a source of fast rise-time high voltage pulses, the source chiefly including a command resonance charge circuit in communication with a multi-stage magnetic pulse compression network, which in turn is in communication with the anode electrode and the auxiliary electrode. A modular timing and triggering circuit preferably is provided for controlling the voltage pulse source circuitry. Circulatory system components provide for the recirculation of process and product gases through the discharge cell and through a heat exchanger to maintain the gas flow temperature at desirable process temperatures. Another embodiment of the instant invention for chemical processing is a pulsed radial self-sustained glow discharge at or above atmospheric pressure; all other radial discharges have been operated either externally sustained, continuous operation, at much lower pressure, or for pulsed lasers. For example, it is known to operate pulsed C0₂ lasers at higher pressure, but such operations use an externally-sustained discharge. The early Antares program at Los Alamos national Laboratory ran 5 atmosphere pressure pulsed C0₂ lasers.

The radial glow discharge embodiment of the invention uses a self-sustained radial glow discharge, previously utilized only for pumping pulsed C0₂ laser plasmas. The anode and cathode discharge electrodes are concentrically mounted cylinders, with the discharge occurring in the annular volume between the cylinders. The combination of uniform UV preionization illuminating the discharge volume, coupled with a fast-rising, high voltage electric field, and a very carefully controlled electric field distribution enables the use of the radial glow discharge for gas processing in highly attaching gases. The discharge may be used to create ozone from oxygen or create hydrogen peroxide from mixtures of water and oxygen, for example. Other chemicals, such as hydrazine, acetylene, and cyanogen, can also be created with this unique discharge configuration.

The self-sustained discharge requires that electrons be initially created in the discharge volume (pre-ionization) in sufficient concentration to overcome electron non-uniformities caused by the passage of stray cosmic rays through the gas mixture. This typically requires preionization levels of approximately 10⁹ to 10" electrons per cubic centimeter. These electrons are then avalanched by the electric field to approximately 10¹² to 10¹³ electrons per cm³ to form the glow discharge which processes the gas (the neutral atom/molecule number density at atmospheric pressure is 2.55 x 10¹⁹ per cm³ at 15 degrees Celsius). A self- sustained discharge can use a UV source pre-ionizer such as an array of raised sparks or a dielectric barrier. With either of these two devices, the preionizer is utilized to flood the discharge volume with UV light, which creates photo-electrons that are accelerated by the main electric field to produce the desired discharge characteristics. For this disclosure and in the claims, a self-sustained glow discharge is a discharge wherein the electrons used for conduction are released from neutral and ionized atoms and molecules by the inelastic collision by other electrons, the ionizing electrons originating either from similar inelastic collisions or from electrons resulting from pre-ionization and accelerated by the electric field between the discharge electrodes.

An electron gun (e-gun) could also be utilized for preionization of a self- sustained discharge by creating 10⁹ to 10¹¹ secondary electrons per cm³ in the discharge. It should be recognized that this is two orders of magnitude less power required from the e-gun than if the discharge were externally sustained by the e- gun.

Another type of UV source can be utilized, called a "flashboard", wherein the sparks that create the ultraviolet light are formed between metal pads deposited on the surface of a dielectric material such as printed circuit boards. Another type, called a flashlamp, uses a discharge in low pressure gas such as xenon. The discharge occurs inside an envelope transparent to a significant part of the UV spectrum, such as quartz. The lamps can have high efficiencies, but are only suitable for preionizing certain types of gas mixtures that have a photoelectron process that falls within the transmitted spectrum of the contained discharge.

The radial glow discharge configuration is central to commercial applications of the invention because the geometry provides a means for processing all the gas that is flowing through the processor. All of the gas that flows through the apparatus must pass through the glow discharge. There are no electrode ends for the gas to flow around. In known linear devices, because the walls of the discharge chamber must be separated from the ends of the electrodes to avoid surface flashover on the walls, there is always space at the ends of the discharge electrodes through which the process gas can flow, impairing efficiency. Thus, for a commercial glow discharge gas processor, the radial glow discharge is advantageous.

The gas processing radial glow discharge is very different from a laser discharge because gases typical for industrial applications have very high attachment rates, such as the oxygen utilized for ozone and peroxide production, for example, and thus require different techniques to operate successfully. For the radial glow discharge, the preionizer can be placed in one of three locations. The preionizer can be placed behind the outer electrode, thus flooding the inner electrode with UV light. This is an attractive location because it provides plenty of room for the preionizer. The preionizer can also be located behind the inner electrode, flooding the outer electrode with ultraviolet light. This location is the preferred embodiment of this invention because the relative curvature of the two electrodes provides higher stability conditions when the outer electrode is used as a cathode. The UV light must always flood the cathode for highest performance, self-sustained discharge operation.

One version of the radial discharge embodiment employs sidelight emission, that is, sources of ultraviolet light, such as sparks between point electrodes, are arrayed on either side of the main discharge electrodes. When fired, the point electrodes flood the interior of the discharge with ultraviolet light and the discharge receives adequate preionization.

Another aspect of the inventive means and method for achieving high performance self-sustained discharges is to control the distribution of the photoelectrons while also controlling the distribution of the electric field. Locating the UV source either behind the inner or outer electrode is the preferred way to initiate the self-sustained discharge operation, and contributes to higher performance operation with greater reliability. The radial glow discharge embodiment of the invention can utilize two types of gas flow. One is axial gas flow, as shown in Fig. 7, where the gas flows between the electrodes parallel to the axis of the cylindrical electrodes. The second is for the gas to flow radially, either out from the inner electrode to the outer electrode or out of the outer electrode to the inner electrode (Fig. 9). This embodiment provides the capability of operating the discharge at high pulse repetition rates.

Control of the voltage in time across the electrodes is very important to successful operation of the self-sustained glow discharge for chemical processing. The various embodiments of the present invention may use either magnetic switches, solid-state, or gaseous switches. Magnetic switches are the preferred embodiment. Solid state switches, as well as gaseous switches such as spark gaps, thyratrons, pseudospark switches and other gas switches may also be utilized. Liquid switches, such as water and oil switches are also potential candidates for operating the pulsed self-sustained radial glow discharge for gaseous processing. A key characteristic of all of these switches is that they must produce a rapid electric field rise time to achieve a successful stable operation of the glow discharge for gas processing.

A block diagram of one embodiment of the present invention is shown in FIG. 2. The high efficiency system of the invention includes a containment vessel 1 1 , which houses the discharge assembly or cell (16-20), as well as the gas circulation loop 13, a magnetic modulator network 22, including a command resonant charge circuit 23 and a magnetic pulse compression network 21 , provides fast rise-time high voltage pulses to the discharge assembly. Preferably, a DC power supply 24 provides a constant voltage to the input of the modulator network 22, and a modulator timing and triggering circuit 25 controls the operation of the modulator network 22.

During the practice of the invention, a feedstock process gas stream 10 enters the containment vessel 1 1 of the processor. After the process gas has entered the containment vessel 1 1 , a fan 12 circulates the gas along a looped path 13 preferably including ducts for directing it through the discharge volume 14 into the heat exchanger 15 and returns it to the inlet side of the fan 12. The discharge volume 14 is defined between the anode 18 and the cathode 20. The gas within the discharge volume 14 between the electrodes is preionized by homogeneously illuminating the discharge volume with ultra-violet radiation. The ultra-violet radiation for the preionization is provided by an electrical discharge 16 that occurs between a surface of a dielectric barrier 17 and the anode screen 18. The anode screen 18 is a portion of the anode that is permeable or transparent to ultraviolet light. The ultra-violet radiation produced by the discharge 16 penetrates or passes through the anode screen 18 and into the discharge volume 14. The preionization process is initiated by the application of a high voltage pulse between the anode screen 18 and an auxiliary electrode 19 disposed in contact with the back of the dielectric barrier 17. After the discharge volume 14 has been preionized, the discharge volume is excited by the application of the main high voltage pulse, which is impressed across the anode 18 and the cathode 20 electrodes. A homogeneous and stable electrical discharge subsequently ensues in the discharge volume 14 and continues until the main high voltage pulse is terminated. The discharge preionization and excitation process is further illustrated in

FIG. 3 where the electron number density 40 of the discharge volume, the preionization voltage pulse 41 , and the main discharge voltage pulse 42 are graphed as a function of time. Initially, the electron number density is very low (^" 10³-10⁴ electrons/cm³). The preionization voltage pulse 41 is applied to the electrode 19 disposed on the back of the dielectric barrier 17. This causes the volume between the front surface of the dielectric barrier and the anode screen 18 to become partially ionized, thereby producing a significant amount of ultra-violet radiation that penetrates the anode screen 18 and enters and permeates the gas volume 14 between the principal electrodes 18, 20. The UV radiation causes electrons to be freed from the gas molecules within the discharge volume 14 such that the electron number density 40 of the gas volume 14 increases to the order of 10⁹ electrons/cm³, as shown in FIG. 3. The main high voltage pulse 42 is then impressed across the principal electrodes 18, 20, to act through the discharge volume 14. Preferably, the high voltage pulses are applied at a voltage rate-of-rise in excess of 500 V/ns. The rapid rise pulses cause a homogeneous electrical discharge to form between the anode 18 and cathode 20 electrodes, which in turn results in a further increase of the electron number density 40 within the discharge volume 14. The electron number density remains at the elevated level until the main voltage pulse 42 is terminated. Alternative embodiments of the electrode and preionization assemblies are described in greater detail in FIGS. 4A, 4B, 5A and 5B. FIGS. 4A and 4B illustrate a configuration which uses a plurality of ceramic blocks 50 as the dielectric barrier (analogous to barrier 17 in Fig. 2) for the preionization assembly. The cathode 51 and anode 52 electrodes (analogous to 18 and 20 in Fig. 2) are fabricated from an ozone resistant or other chemical resistant alloy, such as stainless steel or the like. The shapes of the cathode 51 and anode 52 are configured to maintain an uniform electric field within the discharge volume 14. Both the cathode 51 and anode 52 are mounted in respective ones of a pair of electrically non-conductive structures 53, which are used to rigidly support the electrodes 51 , 52 as well as to direct the gas flow through the discharge volume 14 in an efficient manner (i.e., serving as flow transition structures to match a flow cross-section of the discharge volume 14 to a flow cross-section in the ducts). The anode and auxiliary electrode assemblies are secured together with electrically non-conductive standoffs 54, which are used to maintain the precise spacing between the anode 52 and cathode 51 electrodes without interfering significantly with the process gas flow through the volume 14.

Continued reference is made to Figs. 4A and 4B. The anode electrode assembly includes the anode body 52 and the anode screen 18. The anode screen 18, preferably, is a relatively thin piece of perforated stainless steel or other conductive material attached to the anode body 52. Immediately above the anode screen 18 are mounted the dielectric blocks 50 which serve as the dielectric barrier (generally analogous to barrier 17 in Fig. 2). In this embodiment, the dielectric barrier features a plurality of preferably rectangular dielectric blocks 50 mounted end-to-end in order to span the length of the anode electrode assembly. Specifically, each dielectric block 50 is fashioned from a material such as, but not limited to, ceramic or polymer, having a dielectric constant greater than that of air.

The back surface of the dielectric material is plated with an electrically conductive material, thereby forming the preionization or auxiliary electrode 55 to which the preionization voltage pulse is applied. Central to the auxiliary electrode 55 is an electrically conductive stud 56 which enables the connection of the output of the modulator 22 to the preionization electrode 55. The surface, with exception of the front face 57, of each dielectric block 50 is preferably coated with a high-voltage insulating material 58 to prevent electrical breakdown across the surface of the dielectric blocks 50. The face 57 of each dielectric block 50 is appropriately spaced from the anode screen 18 by electrically non-conductive spacers 59 as shown in FIG. 4A. When a high voltage pulse is applied to the input stud 56, most of the voltage appears across gap 60 between the face 57 of the dielectric block 50 and the anode screen 18. As a result, the gas in gap 60 is at least partially ionized, thereby producing the required ultra-violet radiation 61 , which penetrates the anode screen 18 and impinges upon the gas molecules within the discharge volume 14. The main discharge is subsequently formed when a high voltage pulse is applied between the anode 52 and cathode 51 electrodes.

An alternative embodiment of the preionization and electrode assemblies is illustrated in FIGS. 5A and 5B. In this embodiment, the anode and cathode structures are similar to those shown in FIGS. 4A and 4B; however, in this configuration, the dielectric barrier is formed with a single piece of machinable dielectric such as, but not limited to, ceramic material 70 which spans the length of the anode electrode assembly. Specifically, the ceramic (for example) dielectric material 70 is machined into a U-shape as indicated by FIGS. 5A and 5B. An electrically conducting film (not shown) is deposited on the bottom of the ceramic well 71 to which the high voltage preionization pulse is applied. Alternatively, the preionization electrode 72 may be formed by placing an electrically conducting plate on the bottom of the ceramic well 71 and securing it in place with a backfill of epoxy material 73. The face 74 of the ceramic material 70 is appropriately spaced from the anode screen 18 by electrically non-conductive spacers 75 which span the length of the preionizer assembly. When a high voltage pulse is applied to the input connections 77 of the preionizer assembly, most of the voltage appears across the gap 76 between the face 74 of the ceramic material 70 and the anode screen 18. As a result, the gas in gap 76 is partially ionized and a significant amount of ultra-violet radiation 62 is produced. The ultra-violet radiation penetrates the anode screen 18 as shown in FIG. 5A and impinges upon the gas molecules within the discharge volume 14. The main discharge is subsequently formed when a high voltage pulse is impressed across the anode 52 and cathode 51 electrodes.

It will be immediately appreciated that alternative means for providing ultraviolet light may be provided. Various closed discharge sources, such as an off-the-shelf flashlamp, may be employed in lieu of the auxiliary electrode and dielectric barrier, and connected to a trigger circuit in the modulator 22. A detailed schematic view of the gas circulation system and discharge assembly is shown in FIG. 6. The gas circulation system and the discharge assembly are both housed within the containment vessel 1 1 . An average flow of process gas 85 is passed through the containment vessel 1 1 and exhausted from the exit port 80. As a result, the gas pressure within the containment vessel 1 1 is maintained at a constant value. In this implementation, a cross-flow fan 81 is used to circulate the process gas around the flow loop defined by the flow ducting 82 and the discharge assembly 83, which is generally comprised of the anode and cathode assemblies substantially as described. As a gas volume enters the discharge volume 14, the gas is preionized with ultra-violet radiation generated by the preionization assembly 88, and is subsequently ionized by the application of the main discharge voltage pulse across the anode 52 and cathode 51 electrode assemblies. The electrical discharge that develops within the discharge volume 14 serves as a homogeneous source of energetic electrons. The energetic electrons collide, for example, with diatomic oxygen molecules, a portion of which dissociate to form single oxygen atoms that subsequently recombine with other diatomic oxygen molecules to form ozone. As a result, the average ozone concentration of the process gas 85 within the containment vessel 1 1 increases.

To preserve the product gas concentration at the highest levels possible, it is desirable to maintain the temperature of the process gas at near room temperature, so that the rate of thermal decomposition of the product gas, such as ozone, is significantly reduced. In other processes, the product gas may need to be maintained at some other selected temperature. This is accomplished in the present invention by passing the flow of gas 85 that exits the discharge volume 14 through a gas-to-liquid heat exchanger 86. The heat exchanger 86 is designed to remove heat from the process gas resulting from the electrical discharge. The exchanger 86 may be a tube-and-fin type exchanger, or a surface conduction exchanger. In one alternative, the heat energy is transferred to a liquid cooling medium such as water, chilled water, or propylene glycol and water, or other suitable heat transfer fluid within the exchanger 86. After the gas 85 flows through the heat exchanger 86, it is directed into the inlet side 87 of the fan assembly 81. The fan 81 preferably is a cross-flow fan, so that the fan assembly 81 allows an uniform flow to be developed along the length of the discharge assembly 83. In addition, the motor, which powers the fan 81 , is easily mounted on the exterior of the containment vessel 1 1 , thereby eliminating the need to protect the motor from the effects of the ozone. A magnetic coupler, or other means of transmission, transmits power from the exteriorly mounted motor to the internally disposed cross-flow fan 81. Alternately, a group of axial fans can be used to circulate the process gas 85; however, the motors which power the axial fans disposed within the containment vessel 1 1 preferably are of the sealed type to protect them from the generated ozone or other produced gas.

Many different gases may be chemically processed using the technique described above. Hydrogen peroxide is one example of how a well-controlled electrical discharge with average electron energy at high levels can induce chemical changes in a gas by pumping the associating molecules and atoms to excited metastable states. With other chemical processes, the same issues of uniform excitation and processing of the gas for high efficiency and control of the temperature are important, as for ozone. In the present disclosure, hydrogen peroxide and ozone are used as primary examples of chemical processing of gases by the use of over-voltage, uniform bulk glow discharges. The invention includes bulk glow discharge, and such discharge in conjunction with multiprocess cooling systems to maintain a low gas temperature to avoid causing degradation of the product or stimulating their reaction to go in a direction not desired.

It is known to make a radial glow discharge with an externally sustained discharge. The externally sustained discharge provides a means of improving the discharge stability. The use of a gas with very mild attaching characteristics, such as the pulsed C0₂ laser gas mixture, contributes to enabling stable operation in the radial discharge configuration. In contrast, an alternative embodiment of the inventive apparatus, for use in chemical gas processing, provides a pulsed radial self-sustained glow discharge operated at near atmospheric pressure. All other radial discharges have been operated either externally sustained, continuously (non-pulsed), at much lower pressure, or in lasers. As shown in FIG. 7, the invention includes the operation of a glow discharge in a non-planer radial electrode assembly 160. In this embodiment, two concentric, generally cylindrical electrodes 162, 166 define therein the glow discharge volume 164, as shown in Fig. 7. The electrodes 162, 166 are cylinders, but feature curved contours at the corners where the cylinder wall meets the cylinder base, as shown by Fig. 8 The glow discharge operates in the discharge volume 164 between the inner electrode 162 (preferably the anode) and the outer electrode 166 (preferably the cathode). The discharge is "radial" because the electron motion is radial from one electrode (e.g., 166) to another (e.g., 162). The process gas mixture is preionized by a preionizer 168 located radially outward from, and behind, the cathode electrode 166 in relation to the discharge volume 164.

For the radial discharge embodiment, the preionizer 168 may be disposed in one of three alternative locations. As shown in Fig. 7, the preionizer 168 may be placed radially outward behind the outer electrode 166, thus flooding the inner electrode 162 with UV light upon actuation. This is an attractive location because it allows ample room for the preionizer 168. Preferably, however, the preionizer 168 is located radially inward from the inner electrode 162, thereby flooding the outer electrode 166 with ultraviolet light. Locating the preionizer 168 centrally within the inner electrode 162 is preferred because the relative curvature of the two electrodes 162, 166 provides greater discharge stability when the outer electrode 166 is used as a cathode, and the UV light floods the cathode 166.

Alternatively, the preionizer may involve sidelight emission, that is, sources of ultraviolet light, such as sparks between point electrodes, arrayed transversely on either or both sides of the main discharge electrodes 162, 166. FIG. 8, which is a sectional diagram of the discharge volume 164 and electrodes 162, 166 of the radial glow discharge, illustrates this alternative embodiment. The cathode 166 and anode 162 are shown with the discharge volume 164 between them. The preionizers 168, 168' are separate sources of UV light, such as spark electrodes, and are shown located to one, preferably both, sides of the electrodes 162, 166. Upon actuation, the point electrodes 168, 168' flood the discharge volume 164 transversely, rather than radially, with ultraviolet light to preionize the discharge. An aspect of the invention for achieving high performance self-sustained discharges is to control the distribution of the photoelectrons while simultaneously controlling the distribution of the electric field. Locating the UV source either radially inward behind the inner electrode 162 or radially outward behind the outer electrode 166 is the preferred embodiment for initiating the self-sustained discharge operation and contributes to higher performance operation with greater reliability. Also, the radial discharge can be preionized by UV flashlamps, for gas mixtures that have a photoelectron process that falls within the transmitted spectrum of the flashlamp envelope.

The radial glow discharge embodiment may utilize either of two types of gas flow. Axial gas flow, as shown in FIG. 7, involves gas flow between the electrodes 162, 166 generally parallel to their common axis. Alternatively, as depicted in FIG. 9, the gas may flow radially, either radially outward from the inner electrode 162 through the outer electrode 166 or radially inward from the outer electrode 166 through the inner electrode 162. As shown by the directional arrow of FIG. 9, the gas 137 preferably flows radially out from near the preionizer 1 68, through the inner electrode 162 provided with openings (not shown) to facilitate the gas flow, through the discharge volume 164, and out through the outer electrode 166, which also has openings to facilitate the gas flow. This embodiment offers the capability of operating at high pulse repetition rates. By keeping the gas flow path short, the gas heated by the discharge can be quickly removed in preparation for the next shot. The foregoing radial glow discharge embodiment 160 is particularly advantageous for commercial applications because the geometry provides a means for discharge processing all the gas that is flowing through the apparatus. All of the gas that flows through the radial glow discharge apparatus must pass through the discharge volume 164. Thus, radial glow discharge processing offers advantageous economy and efficiency in commercial applications. Moreover, the gas processing radial glow discharge embodiment 160 of the invention is distinguishable from a laser discharge because gases typical for an industrial chemical processing applications have very high attachment rates such as the oxygen utilized for ozone and peroxide production, and requires processes distinct from a laser discharge to operate successfully. Three different types of electric arc UV sources are suitable for the radial glow discharge embodiment. The first type utilizes an arc between small diameter electrodes 1 68, 168' elevated away from the discharge electrodes 1 62, 1 66, as shown in FIG. 8. This type of preionizer 168 can also be located behind either of the electrodes 1 62, 166. The second type utilizes a dielectric barrier similar to that illustrated in FIG. 4A. A third type is a flashboard UV source, wherein the sparks that create the ultraviolet light are formed between metal pads deposited on the surface of a dielectric material, such as printed circuit board material.

The radial discharge can be preionized by preionizers other than UV sources, including x-ray preionizers. These preionizers emit x-rays that penetrate and ionize the process gas or gas mixture. They typically have low conversion efficiency, but are useful in some applications. They have the advantage of being able to penetrate a discharge electrode and thus do not require a thin foil window (as in e-beam preionizer), or a screen window (as in a UV preionizer). The radial discharge can also be preionized by an electron beam. In this embodiment, the e- beam operates at perhaps 1 -2 orders of magnitude in power density less than required to sustain the discharge. Either the x-ray or e-beam preionizers can be located centrally inside the inner electrode 1 62, outside the outer electrode 166, or axially offset from the side of the electrodes (Fig. 8).

FIG. 10 illustrates additional aspects of the invention for self-sustained pulsed glow discharge used to produce hydrogen peroxide. The closed cycle hydrogen peroxide generator system 170 operates at elevated pressure. The system is closed, with suitable conduits, piping, and the like for providing recirculation of gases. In the production of hydrogen peroxide, the system 170 preferably operates at between about 3 and about 10 atmospheres to promote high hydrogen peroxide conversion efficiency. It has been determined that operating the discharge at elevated pressure provides substantially increased yields of hydrogen peroxide. In the practice of this embodiment of the invention, a process gas mixture of water vapor, oxygen, and ozone flows through a conduit into the glow discharge processing container 172. The process gas mixture is processed in the container 172 by the glow discharge, which is powered by the electrical pulsed power supply 178, to produce a product gas stream mixture hydrogen peroxide, water vapor, oxygen and ozone. The glow discharge in the processing container 172 may be generated by linear electrodes; preferably, however, the processing container surrounds a radial discharge electrode assembly such as the electrode assembly 160 depicted in FIG. 7, or alternatively the electrode assembly shown in FIGS. 8 or 9.

The product gas mixture is transported via another closed conduit from the processing container 172 to a tank 180 containing liquid water at elevated temperature. The tank 180 for holding water acts as a hydrogen peroxide condenser. The conduit which transports product gas from the container 172 to the tank 172 is provided, upon entry into the container, with any suitable means known for bubbling the product gas into the water. The temperature of the water held in the condenser tank 180 is preferably between about 40°C and about 80^CC, as the water temperature should be cool enough to promote condensation of the hydrogen peroxide constituent of the incoming product gas stream, and warm enough to produce the desired fraction of water vapor in the gas mixture exiting the tank 180 for recirculation as process gas to the discharge container 172. Preferably, the water in tank 180 is maintained at approximately 75°C. To minimize pumping losses, the water in the tank 180 is maintained at the similar elevated pressure as the pressure in discharge processing container 172. As the product gas stream is bubbled through the water in the tank 180 the constituent hydrogen peroxide condenses in the water, since the hydrogen peroxide condenses at 150°C. The "exit gases" remaining after the extraction of hydrogen peroxide include a mixture of oxygen, ozone, and water vapor gas. Additional oxygen from oxygen source 183 and water from an external water source 182 are injected, by suitable conduit means, into the water tank 180 to compensate for the peroxide condensed out of the product gas stream and extracted from the water bath.

The exit gases are vented from the tank 180 and then recirculated, again via any suitable conduit means in the art, to the glow discharge processing container 172. The peroxide and liquid water mixture is continuously removed from the water tank 180 through a pressure regulating valve 186 to maintain the high pressure in the system 170. The concentration, by weight, of the hydrogen peroxide in the water is a function of the amount of water injected relative to water extracted. The peroxide concentration can be controlled from about 1 % up to the safe handling limit of about 70% hydrogen peroxide in water. Other known and suitable means, such as mechanical or pressure gas pumps, may be provided as means for transporting the product gas to the tank 180, and for recirculating gases from the tank 180 through the conduit leading back to the discharge processing container 172. Notably, the only additional basic equipment, besides that previously discussed, required for commercial production of hydrogen peroxide is a system for packaging the peroxide. There are no carbon-containing fluids in the system, so there will be no hydrocarbons to remove. There may be trace amounts of nitrates that will have to be removed from the water, depending upon the seedants used for the discharge and the quality requirements of the end user.

As discussed previously, it is known to use glow discharges to create ozone. However, prior instances of self-sustained glow discharges to produce chemicals other than ozone are not generally known, probably because the discharge stability issues are very difficult to overcome in practice in highly attaching gas mixtures. Utilizing an apparatus constructed in accordance with the present invention having plane parallel electrodes, and with UV preionization, discharge processing tests have been conducted at both ambient temperature and elevated temperature. Hydrogen peroxide was extracted as a liquid from the product gas stream by a chiller to condense the hydrogen peroxide and water vapor. The discharge differed from most laser discharges in that the concentration of highly attaching gas (in this case, both the oxygen and the water vapor) was very high.

In one implementation, a process gas mixture containing 39% oxygen, 58.5% nitrogen, and 2.5% water vapor was processed by a pair of linear electrodes spaced 5 mm apart and defining a discharge volume of 0.1 liter. The electrodes were specially shaped and contoured, as is commonly known in the art, to eliminate field enhancement and prevent discharge collapse. Fast rise-time high voltage pulses having a peak voltage of approximately 25 kV, a rate of rise in excess of 500 V/nsec, and a pulse width of less than 100 nsec were applied to the electrodes. The product gas was condensed, producing concentrations of 30 mg of hydrogen peroxide per liter of water extracted.

The present invention provides an improved self-sustained glow discharge stability in gas mixtures suitable for gas chemical processing through the use of various additives to the process gas mixture. In the case of ozone or peroxide manufacture, however, hydrocarbon additives are not appropriate because the presence of high concentrations of oxygen makes the use of hydrocarbon seedants impractical. Certain monatomic and polyatomic additives are stable in the presence of oxygen, and yet provide increased photoelectron yield and reduced glow voltage. These additives are typically, but not necessarily, noble gases, and include nitrogen, helium, argon, neon, krypton, and xenon. Other monatomics and polyatomics are also suitable for photoelectron seedants and ionization rate enhancers. For example, in glow discharge measurements to create hydrogen peroxide from mixtures of water and oxygen, nitrogen proved to be a valuable seedant for enhancing UV photoionization processes and improving discharge stability.

The precise control of the voltage in time across the electrodes 1 62, 1 66 is key to successful operation of the self-sustained glow discharge for chemical processing. The radial glow discharge gas processor can utilize magnetic switches, solid-state, or gaseous switches. Magnetic switches, as described herein above, are the preferred embodiment. Magnetic switches are typically custom designed for each application using commercially available materials that may be obtained from Ceramic Magnetics, Inc., for example. In general, the design of magnetic switches is commonly known, and is discussed in detail, for example, in "Magnetic Switches and Circuits" by W. C. Nunnally, Pulsed Power Lecture Series Number 25, Plasma Switching Laboratory, Texas 79409, September 1981 , which is incorporated herein by reference in its entirety.

In accordance with the present invention, the rate of rise of input voltage to the magnetic switch determines the material and dimensions for the core. The core material and dimensions are selected so that the switch core is saturated (i.e., the relative permeability has been reduced to near 1 ) by leakage current when the input voltage is substantially equal to its peak value. Aligning the peak input voltage with the saturation point of the magnetic switch results in a rapid transfer of energy through the switch because the impedance of the saturated switch is substantially near zero. By way of example only, an input signal having a rise time of 65 nsec applied to a magnetic switch having four cores made of CMD-10 material from Ceramic Magnetics, Inc., core dimensions of 9 cm O.D./3 cm

I.D./2.54 cm thick, and wound in a single turn coaxial configuration, generated an output signal with a voltage rate of rise less than 50 nsec.

Solid state switches can also be used for switching the gas mixture such as GTOs and SCRs. The third category includes gaseous switches such as spark gaps, thyratrons, pseudospark switches (for example as disclosed in co-pending Application Serial No. 08/890,485) and other gas switches. Liquid switches, such as water and oil switches are also potential candidates for operating the pulsed self-sustained radial glow discharge for gaseous processing. A key characteristic of all the aforementioned switches is that they must produce a rapid rise time in the electric field to achieve a successful stable operation of the glow discharge for gas chemical processing.

Combined reference is made to FIGS. 2, 6, and 7. It is highly preferable to provide fast rise-time high-voltage pulses to the preionization assemblies 88, 168 and main discharge assemblies 83, 160. Specifically, it is preferable to excite the main discharge with high voltage pulses having rate-of-voltage-rises in excess of

500 V/ns. Such a pulse is provided to the discharge assembly 83, 160 by a solid state modulator 22 featuring a switched magnetic pulse compression network 21 . The modulator 22 processes direct current (DC) power, provided by the power supply 24, into a continuous train of fast-rise-time high-voltage pulses of relatively short duration. The modulator 22 principally consists of two circuits, for example, a command resonant charge circuit 23, and a multi-stage magnetic pulse compression network 21.

A detailed schematic of an example magnetic modulator circuit 22 is shown in FIG. 1 1. A DC power supply 24, which may be as simple as a transformer and a full wave rectifier bridge, is used to charge filter capacitor 101 to a DC voltage that is typically between 600 V and 1000 V. The transfer of energy from the filter capacitor 101 to the first capacitor 108 in the magnetic pulse compression network 21 is controlled by a solid state switch, such as an insulating gate bipolar transistor (IGBT) 102 which, when turned on, enables energy to be transferred from the filter capacitor 101 to capacitor 108 in the magnetic pulse compression network. Charging inductor 105 determines the rate at which energy is transferred to first capacitor 108. The precise amount of energy which is transferred to first capacitor 108 is determined by the conduction period of IGBT 102. The voltage to which first capacitor 108 is charged is controlled by a feedback circuit 150B in the modulator timing and triggering circuit 25, which controls the conduction time of the IGBT 102. The feedback circuit 150B uses voltage signals derived from current monitor 106 and voltage divider 107 to determine the conduction time of the IGBT 102. The charging sequence for first capacitor 108 begins with a trigger pulse which turns on the IGBT 102, which in turn causes current to begin to flow from filter capacitor 101 , though the charging inductor 105, and into first capacitor 108. When the charge current and the voltage level on capacitor 108 reaches the appropriate level, the feedback control circuit terminates the IGBT drive pulse and the IGBT 102 is turned off. Once the IGBT 102 is turned off, the energy stored in charging inductor 105, by virtue of the current flowing through the inductor, is transferred to capacitor 108 by f ree- wheeling diode 104. After the IGBT 102 has been turned off, free-wheeling current continues to flow until the energy which is stored in the charging inductor 105 has been transferred to first capacitor 108.

After first capacitor 108 has been charged to the desired voltage, silicon controlled rectifier (SCR) 109 is triggered and the energy stored in capacitor 108 is transferred to second capacitor 1 1 1 over a period determined by the values of capacitors 108, 1 1 1 and the saturated inductance of magnetic anode assist 1 10. Alternately, an IGBT device may be used in place of SCR 109. In addition to determining the transfer time between first capacitor 108 and capacitor 1 1 1 , the magnetic anode assist 1 10 provides a time delay (several hundreds of ns) for the onset of current flow, so that SCR 109 is in full conduction before it conducts heavily.

The energy transfer from first capacitor 108 to second capacitor 1 1 1 is graphically illustrated in FIG. 12. In the figure, all of the voltage waveforms have been normalized with respect to their peak values. Referring jointly to FIGS. 1 1 and 12, it is seen that when SCR 109 is triggered at t=to, the voltage 130 on first capacitor 108 begins to decrease as the voltage 131 on second capacitor 1 1 1 increases, indicating that the energy stored in capacitor 108 is being transferred to capacitor 1 1 1. Saturable inductor 1 12 prevents current from flowing into the primary of the step-up transformer 1 13 until the voltage 131 on capacitor 1 1 1 has reached its peak at t =t When voltage 131 has reached its peak, saturable inductor 1 12 saturates at t = t_{1 r} and capacitor 1 1 1 is discharged into the primary of the step-up transformer 1 13. This results in the energy that is stored in capacitor 11 1 being transferred to capacitor 1 14 as indicated by voltage waveform 132 in FIG. 12. Because a step-up transformer is interposed between capacitor 111 and capacitor 1 14, capacitor 1 14 is charged to a substantially higher voltage

(approximately 25-30 kV) consistent with the turns ratio of transformer 1 13. Saturable inductor 1 15 prevents current flow until capacitor 1 14 has reached its full charge voltage at t = t₂. At t=t₂, saturable inductor 1 15 saturates and the energy stored in capacitor 1 14 is transferred to capacitor 1 16, as indicated by voltage waveforms 132 and 133 shown in FIG. 12. Again, saturable inductor 1 17 is designed to prevent current flow until the voltage on capacitor 1 16 has reached it peak at t = t₃. Saturable inductor 1 17 incorporates a secondary winding which is used to excite the preionization assembly 122. Capacitor 1 19 is placed between the secondary winding 1 18 and the preionization assembly 122 and is used to control the amount of current which is sourced to the preionization assembly 88. Continued reference is made to FIG. 1 1. In this embodiment, the voltage pulse is applied to the preionization assembly before the main high voltage pulse is applied across the anode 124 or 162 and cathode 125 or 166 electrodes. At t=t₃, saturable inductor 1 17 saturates and the energy stored in capacitor 1 16 is discharged into the primary of step-up transformer 120. This causes the energy stored in capacitor 1 16 to be transferred to the peaking capacitor 121 , which results in the fast rise-time voltage waveform 134 being impressed across the anode 124 or 162 and cathode 125 or 166 electrodes. The value of the peaking capacitor 121 is tuned so that the appropriate voltage rise-time and amplitude is developed across the electrodes 124, 162 or 125, 166 so as to maintain a uniform and stable discharge between the electrodes of the discharge cell. The invention, however, is not limited to this circuit. This circuit is for illustration purposes only. Other equivalent magnetic compression pulse circuits can accomplish a similar function.

The magnetic modulator network 21 is designed to deliver high voltage pulses to the electrode assembly 83, 160 (Figs. 6 and 7) which have durations of 100-200 ns. In order to achieve the desired ozone, hydrogen peroxide, or other product chemical yield, the high voltage pulses must be applied across the electrodes 51 , 52, 162, 166 at a specific repetition rate. To the first order, the average power delivered to the discharge, and therefore the gas chemical production rate, is directly proportional to the repetition rate at which the high voltage pulses are delivered to the discharge.

The modulator timing and triggering circuit 25 includes timing and triggering circuitry 150A and the feedback circuit 150B. The timing and triggering circuitry 150A provides a continuous stream of triggering pulses to the IGBT 102 and the SCR 109 within the magnetic modulator network 22. An external analog signal

(Vin) controls the repetition rate (i.e., frequency) of the triggering pulses provided by the timing and triggering circuitry 150A to the IGBT 102 and the SCR 109. The feedback circuit 150B controls the conduction time of the IGBT 102 by modulating the timing of a reset pulse 149 that is provided to the timing and triggering circuitry 150A. In accordance with the present invention, the conduction time of the IGBT 102 may be varied to increase or decrease the amount of energy that is stored in the first capacitor 108 at the beginning of each pulse compression cycle. Using the current monitor 106 and the voltage divider 107, the feedback circuit 150B monitors the instantaneous energy that will be stored in the first capacitor 108. This instantaneous energy is compared to a desired energy level input that is represented by an analog voltage (Vref) applied to the feedback circuit 1 50B. When the instantaneous energy level reaches the desired energy level, the feedback circuit 150B sends the reset pulse 149 to the timing and triggering circuit 150A, thereby terminating conduction through the IGBT 102. Preferably, as Vref is increased, the reset pulse 149 interval increases, which in turn increases the conduction time for IGBT 102 and the energy transferred to the first capacitor

108.

Thus, the both the frequency and the energy of the high voltage pulses applied to the discharge volume can be controlled by properly varying the analog input signals Vref and Vin to the timing and triggering circuit 25. Those skilled in the art will recognize that these analog input signals may be provided by manually adjusted external voltage supplies or signal generators, or by automatically adjusted voltages supplied by a microprocessor, a micro-controller, a computer system, or the like.

Now turning to FIG. 13 a more detailed block diagram of an example timing and triggering circuit is shown. A voltage-to-frequency convertor 140 is used to generate the master timing signal for the modulator system. The repetition rate of the pulse train which is generated by the voltage-to-frequency convertor 140 is controlled by an analog input signal. The pulse train generated by the voltage- to-frequency converter 140 is then used to trigger the first monostable multivibrator 141 , which upon each trigger input generates a pulse 142 of fixed duration, which in turn is used to drive the pulse amplifier 143 used to provide a high current pulse to the gate of SCR 109 in the magnetic pulse compression network 21. In addition, pulse 142 is employed as the input trigger to a second monostable multivibrator 144. Upon each trigger pulse the multivibrator 144 generates pulse 145 which is used to determine the delay between the triggering of SCR 109 and the turn on of IGBT 102 in the command charge circuit 23.

Pulse 145 is also used to trigger a third monostable multivibrator, turning on the IGBT 102 by generating pulse 147, used to drive the IGBT gate driver circuit 148. Pulse 147 remains high, and therefore the IGBT 102 remains turned on, until monostable multivibrator 146 is reset by the reset pulse 149 derived from the feedback circuit 150B.

The feedback circuit 150B calculates the instantaneous energy stored in the charging inductor 105 and capacitor 108, both of which are located in the magnetic modulator network shown in FIG. 1 1 , and compares it to the energy which is to be stored in capacitor 108 for a given pulse. In this manner, the voltage to which capacitor 108 is charged can be precisely controlled.

Specifically, IGBT 102 in the command charge circuit is turned on to charge the first capacitor 108. This causes a current to flow from the DC power supply 24 through IGBT 102 and charging inductor 105 into capacitor 108 in the magnetic pulse compression network 21 . Turning off IGBT 102 causes the energy which is stored in charge inductor 105 to be transferred to capacitor 108 through the free-wheeling diode 104. Regulation of the voltage on capacitor 108 is achieved by turning off IGBT 102 in the command charge circuit 23 at the instant that the sum of the instantaneous energies stored in charging inductor 105 and capacitor 108 yield the desired final charge voltage upon capacitor 108. This is accomplished by calculating the instantaneous energies stored in charging inductor 105 and capacitor 108, summing them, and comparing that sum to the final energy which is desired to be stored in capacitor 108 of the magnetic pulse compression network 21.

The instantaneous energy stored in capacitor 108 is calculated by squaring a voltage signal which is proportional to the instantaneous voltage on capacitor 108 with the analog multiplier circuit 153. Likewise, the instantaneous energy stored in charging inductor 105 is calculated by squaring a voltage which is proportional to the instantaneous current which is flowing through charging inductor 105 with analog multiplier circuit 155. The squared voltage signal which is proportional to the charging current is then multiplied by a fixed gain by amplifier 154 to account for the specific values of charging inductor 105 as well as capacitor 108. The desired final energy to which capacitor 108 is to be charged is calculated by squaring a reference voltage with analog multiplier circuit 156. The squared voltage signal proportional to the voltage on capacitor 108 and the gain adjusted squared voltage signal which is proportional to the charging current are then summed in summing amplifier 152, the output of which is compared to the squared reference voltage. When the sum of the squared voltage signal and the gain adjusted charging current signal exceeds the square of the reference voltage, the output of the voltage comparator 151 generates a reset signal which is then used to reset monostable multivibrator 146 which causes the IGBT drive pulse to be disabled. As a result, IGBT 102 in the command charge circuit is turned off, preventing additional current from flowing from the filter capacitor to capacitor 108 of the magnetic pulse compression network 21. However, since energy remains in charging inductor 105 when IGBT 102 is turned off, the current which is flowing through the charging inductor is permitted to flow through the free-wheeling diode 104 until the energy stored in the charging inductor has been transferred to capacitor 108. Again, this circuit is an example. Other control schemes are possible and within the scope of this invention.

There is a maximum amount of energy per unit volume that can be loaded into a bulk volume of gas with a glow discharge without creating a discharge collapse. This maximum is called the energy loading arc limit. The magnetic modulator network 22 of the invention enhances the total energy that is loaded into a given volume of gas mixture. One of the mechanisms that leads to discharge instability is the creation of super-elastic electrons. Super-elastic electrons are electrons that collide with excited neutrals and, instead of loosing energy to the neutral, they gain energy. These electrons make the discharge much more susceptible to collapse and formation of an arc, and are one of the factors that contributes to the energy loading arc limit.

In the present invention, a multi-pulsing process adds energy to promote the chemistry processes without exceeding the discharge stability limits. Thus, a given volume of gas is processed with a glow discharge pulse to below the energy loading arc limit. Before the energy loading limit is exceeded, the voltage is removed from the discharge. Once the voltage is removed, the process gas constituents relax to a near-Boltzman distribution of the energy among the energy states of the gas. Then a second pulse is applied to the same gas that allows more energy to be loaded into the same gas mixture. Again, the voltage is removed prior to initiation of discharge runaway and collapse. Subsequent pulses can be applied until the gas is no longer capable of accepting additional energy. Then the volume of gas is moved out of the discharge volume and a second bulk volume of gas is moved into the discharge and the process repeated. One notable aspect of the invention is the determination that subsequent pulses must be reduced in energy compared to the first pulse, in order for multi- pulsing to be effective. That is, once the gas volume has been processed by a discharge pulse to a particular energy, the next discharge pulse must typically be of less energy to avoid arcing. Even though the second pulse is of reduced energy, the total amount of energy that can be loaded into the gas before it is moved out of the discharge region can be substantially more than could be imposed with a single pulse.

In general, the energy of successive pulses can be controlled (i.e., reduced) by varying the energy stored in the first capacitor 108 of the magnetic modulator network 22 (shown in FIG. 1 1 ). There are several manners in which the energy of successive pulses can be controlled. In one approach, the analog reference voltage (Vref) applied to the feedback circuit 150B may steadily reduced to accomplish a controlled droop of the peak voltage on the first capacitor 108 over successive pulses. In another approach, the DC power supply 24 may be selected so that it operates in a downward sloping portion of its voltage regulation characteristic, thereby reducing the voltage on filter capacitor 101 over successive pulses. In yet another approach, a burst mode pulse forming network could be used to provide a burst of pulses with decreasing voltage. Use of a burst mode pulse forming network with the present invention would require modifications to the magnetic modulator 22 that could be accomplished by one of ordinary skill in the art. A more detailed discussion of burst mode pulse forming networks can be found in a paper entitled "Conceptual Design of Pulse Generators for Driving Recirculating Induction Accelerators," by W. M. Money, M. G. White, and F. W. White, 1991 IEEE Pulsed Power Conference, pages 943-944, which is hereby incorporated by reference in its entirety.

Industrial Applicability The invention is further illustrated by the following non-limiting examples, where useful product gases are produced according to the apparatus and processes of the invention. One industrial application of the present invention is to generate ozone for municipal water applications. By the invention, the generator is constructed so as to produce approximately 10,000 pounds of ozone per day. The produced ozone is then used to treat water in a municipal water supply. Ozone is exceptionally beneficial for treating water because it leaves no residue as does chlorine. in a second industrial application, hydrocarbon feedstock such as methane from natural gas is fed into the generator along with a mixture of other additive gas for processing into new chemicals. These gases are then processed by the glow discharge system of the invention where new chemical combinations are initiated by the discharge. The resulting new products are then extracted out of the gas stream and sold as new product.

In another application of the invention, large quantities of hydrogen peroxide may be comparatively economically produced for use in other downstream chemical processes, as an oxidizer, or disinfectant, or the like.

In another industrial application of the invention is in manufacturing ozone for the bleaching of cloth. Ozone is very useful as a cloth-bleaching agent, particularly to produce the "stone washed look" in blue jeans. In this application, a medium-sized ozone generator, producing approximately 100 pounds per day, creates ozone in a gas mixture that includes oxygen. The streams of the ozone- bearing gas are directed into the spots of cloth that the manufacturer desires to have bleached. Typically, the ozone is consumed by the cloth process bleaching and the resulting gas can then be vented to the atmosphere since it contains mostly oxygen with very little ozone.

Other industrial applications for both ozone and hydrogen peroxide include pulp bleaching in paper processing, and purifying water in municipal swimming pools. Again, moderate to large size ozone generators based on the present invention can be utilized to supply the ozone for these other applications.

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.

Claims

CLAIMSWhat is claimed is:

1 . An apparatus for converting a process gas into a product gas, the apparatus comprising: an anode electrode; a cathode electrode; a discharge volume defined between the cathode electrode and the anode electrode, the discharge volume adapted to contain at least some of the process gas; and a magnetic pulse compression network coupled to the anode electrode and the cathode electrode, the magnetic pulse compression network having at least one magnetic switch and adapted to impress fast rise-time high voltage pulses across the discharge volume, whereby the fast rise-time high voltage pulses produce a self-sustaining glow discharge in the discharge volume, and the self- sustaining glow discharge converts at least some of the process gas into the product gas.

2. The apparatus of claim 1 , wherein the magnetic pulse compression network further comprises: a coupled network of magnetic switches and charge storage elements; the coupled network of magnetic switches and charge storage elements being arranged in stages so that the magnetic switches couple between the charge storage elements of the stages; and each of the stages being adapted to receive a voltage pulse having a first rise time and to produce an output voltage pulse having a second rise time less than the first rise time.

3. The apparatus of claim 2, wherein at least some of the charge storage elements are capacitors.

4. The apparatus of claim 2, wherein at least some of the magnetic switches are saturable inductors.

5. The apparatus of claim 2, further comprising at least one voltage step up transformer coupled to at least one of the stages.

6. The apparatus of claim 1 , wherein at least one magnetic switch is coupled to a charge storage device and is adapted to delay current flow from the charge storage device until a pre-determined quantity of electrical charge is stored in the charge storage device.

7. The apparatus of claim 6, wherein the charge storage device is a capacitor.

8. The apparatus of claim 1 , wherein the magnetic pulse compression network is further adapted to impress a series of fast rise-time high voltage pulses across the discharge volume such that each successive pulse from the series of voltage pulses has less energy density than the previous pulse.

9. The apparatus of claim 1 , wherein at least one magnetic switch is a saturable inductor.

10. The apparatus of claim 1 , wherein the magnetic pulse compression network comprises at least one gaseous switch.

1 1 . The apparatus of claim 1 , wherein the magnetic pulse compression network comprises at least one vacuum insulated switch.

12. The apparatus of claim 1 , wherein the magnetic pulse compression network comprises at least one solid state switch.

13. The apparatus of claim 1 , wherein the fast rise-time high voltage pulses have a rate of rise greater than or substantially equal to five-hundred volts per nanosecond.

14. The apparatus of claim 1 , wherein the anode and cathode electrodes have substantially planar geometries.

15. The apparatus of claim 1 , wherein the anode and cathode electrodes have substantially cylindrical geometries.

16. The apparatus of claim 15, wherein the electrodes are concentrically disposed.

17. The apparatus of claim 1 , further comprising a pre-ionizer.

18. The apparatus of claim 17, wherein the pre-ionizer is adapted to emit ultraviolet radiation into the discharge volume.

19. The apparatus of claim 17, wherein the pre-ionizer is adapted to emit an electron beam into the discharge volume.

20. The apparatus of claim 17, wherein the pre-ionizer is adapted to emit x-rays into the discharge volume.

21. The apparatus of claim 17, wherein the pre-ionizer is coupled to the magnetic pulse compression network.

22. The apparatus of claim 21 , wherein the magnetic pulse compression network is further adapted to provide high voltage pulses to the pre-ionizer.

23. The apparatus of claim 1 , further comprising a charge resonant command circuit coupled to the magnetic pulse compression network and adapted to provide one or more pulses having a pre-determined energy level to the magnetic pulse compression network.

24. The apparatus of claim 1 , further comprising a timing and triggering circuit adapted to control pulse formation in the charge resonant command circuit and the magnetic pulse compression network.

25. The apparatus of claim 1 , further comprising a gas circulation system adapted to convey process gas to the discharge volume and to remove product gas and process gas from the discharge volume.

26. The apparatus of claim 25, wherein the gas circulation system includes a heat exchanger adapted to remove heat from the process gas and the product gas.

27. The apparatus of claim 1 , further comprising a containment vessel encapsulating at least the discharge volume.

28. The apparatus of claim 27, wherein the pressure within the containment vessel may be greater than atmospheric pressure.

29. The apparatus of claim 1 , further comprising a condenser adapted to condense the product gas.

30. The apparatus of claim 29, wherein the condenser comprises a pressurized closed tank for holding liquid water at an elevated temperature substantially between forty degrees Celsius and eighty degrees Celsius.

31 . The apparatus of claim 29, further comprising a pressure regulating valve adapted to remove the condensed product gas from the condenser.

32. A method of converting a process gas into a product gas comprising the steps of: conveying the process gas into a discharge volume, the discharge volume being defined between a cathode electrode and an anode electrode; pre-ionizing the process gas in the discharge volume; generating a fast rise-time high voltage pulse using a magnetic modulator network having at least one magnetic switch; coupling the fast rise-time high voltage pulse to the anode and cathode electrodes; generating a self-sustaining glow discharge in the discharge volume; and converting at least some of the process gas into the product gas.

33. The method of claim 32, further comprising the step of circulating the process gas and the product gas through a heat exchanger.

34. The method of claim 32, further comprising the step of pressurizing the discharge volume to a pressure greater than atmospheric pressure.

35. The method of claim 32, wherein the product gas is ozone.

36. The method of claim 35, wherein the process gas contains oxygen.

37. The method of claim 36, wherein the process gas contains at least one stabilizing gas.

38. The method of claim 37, wherein the stabilizing gas comprises at least one selected from nitrogen, helium, argon, krypton, and xenon.

39. The method of claim 32, wherein the product gas is hydrogen peroxide.

40. The method of claim 39, wherein the process gas includes water vapor and oxygen.

41. The method of claim 40, wherein the process gas contains at least one stabilizing gas.

42. The method of claim 41 , wherein the stabilizing gas comprises at least one selected from the group consisting of nitrogen, helium, argon, krypton, and xenon.

43. The method of claim 39, further comprising the step of condensing the hydrogen peroxide gas.

44. The method of claim 32, further comprising the step of recirculating at least some of the product gas to the discharge volume.