Drive circuit for dielectric barrier discharge device and method for controlling discharge in dielectric barrier discharge

Abstract

Provided is a drive circuit for a dielectric barrier discharge device, the drive circuit comprising: a power supply connectable in use across a dielectric discharge gap, the dielectric discharge gap providing a capacitance; and an inductance between the power supply and the dielectric discharge gap when connected, thereby establishing a resonant tank in use, wherein power is supplied to the tank in, and only during, pulse trains, the pulse frequency of each pulse train being adjustable in use to a resonant frequency of the tank, the power supplied by each pulse train charges the tank and maintains the tank at a threshold at which discharge ignition occurs, and discharge ignition events per pulse train are limited to a maximum number of times based on the drive circuit being arranged in use to prohibit each pulse train from transferring power to the resonant tank after the maximum number of times has occurred.

Description

[Technical Field]

[0001] The present invention relates to resonant circuits, such as those used with dielectric barrier discharge devices. [Background technology]

[0002] Dielectric barrier discharge (DBD) devices, such as DBD reactors, can be used to remove undesirable substances from fluids, such as gases or liquids, passing through the reactor. These substances include hydrocarbons, nitrogen oxides (NOx), and sulfur oxides (SOx).

[0003] One application of DBD devices is the removal of substances from exhaust gases. In such applications, as in other applications, the gas passing through the device is at approximately atmospheric pressure. At near atmospheric pressure, DBD devices typically exhibit ignition / breakdown voltages of several kilovolts (kV) to tens of kV.

[0004] Electrically, DBD devices apply a capacitance of between about 10 nanofarads (nF) and about 100 nF to industrial-scale gas purification systems. Such devices are capable of receiving or accepting pulsed high voltages between their electrodes to initiate or trigger plasma ignition (also known as dielectric barrier discharge) between the electrodes.

[0005] Excitation of devices with high voltage slew rates (high dV / dt) and short pulse widths (approximately 100 nanoseconds (ns) to approximately 10 microseconds (μs)) results in higher reactor efficiency. This allows for increased reduction of contaminants in the gas passing through the reactor for a given amount of power. However, high power transfer is difficult to achieve efficiently due to the low power factor (PF) of such DBD devices, which is provided by the ratio of real power (P) to apparent power (S). High power transfer efficiency is intended to mean high efficiency, such as high conversion efficiency.

[0006] Available high voltage pulse power equipment for industrial scale systems typically uses a low voltage pulse generating unit with a peak output pulse voltage of about 400 volts (V) to about 1000 V, followed by a step-up transformer with a turns ratio of about 1:20 to about 1:40 to meet the required plasma ignition voltage level.

[0007] Due to the low PF of DBD devices, a large amount of reactive power is required to repeatedly cycle the voltage across the device, which results in a relatively small amount of real power actually being transferred to the plasma, posing a fundamental challenge to achieving high efficiency.

[0008] To illustrate this difficulty, a DBD device with an equivalent capacitance of 5 nF and an ignition voltage of 20 kilovolts (kV) requires a charge / discharge current of 100 amperes (A) to achieve a voltage rise time of at least 1 μs for the dielectric barrier discharge. As a result, for a 1:20 step-up transformer, a peak current of 2 kA must be handled by the power electronics of the pulse-generating unit used in the DBD device.

[0009] A further problem is that even if ignition to create a dielectric barrier discharge is achievable under these circumstances, the remaining energy stored in the capacitance of the DBD reactor after plasma ignition is not recovered. Instead, this energy is transferred to the pulse generating unit or the DBD. The power dissipated in the reactor itself. The resulting losses in the power semiconductors used in the pulse-generating unit and the winding losses in the step-up transformer result in insufficient power conversion efficiency when keeping the power electronics within safe operating temperatures, limiting the maximum achievable pulse repetition rate. To address this, it is necessary to limit the pulse repetition frequency (PRF) to values ​​of a few hundred hertz (Hz). However, this ultimately limits the average power delivered to the plasma, which is undesirable and ineffective.

[0010] There are known resonant power converter installations available that are often used to drive DBD devices with continuous, high-frequency alternating current (AC) (not pulsed), as opposed to devices that use repetitive cycles. Indeed, such systems are known for their good power conversion efficiency and high output voltage gain when operated near the resonant frequency. However, as discussed in the scientific literature and based on experimental evidence, continuous high-frequency AC excitation of DBD reactors typically reduces the effectiveness of pollutant reduction. This lack of effectiveness is due to the fact that fewer reactive species are produced by the decomposition caused by the excitation, and the flue gas is heated, dissipating power instead of being usable to drive further generation of reactive species.

[0011] Therefore, there is a need to address the low overall efficiency and limited average power transfer capability in DBD devices while protecting circuits from damage from high peak currents. Summary of the Invention

[0012] According to a first aspect, there is provided a drive circuit for (i.e. suitable for) a dielectric barrier discharge device, the circuit comprising: a power supply connectable in use across a dielectric discharge gap, the dielectric discharge gap providing a capacitance; and an inductance between the power supply and the dielectric discharge gap when connected, thereby establishing a resonant tank in use; wherein power is supplied to the tank in, and only during, pulse trains, the pulse frequency of each pulse train being adjustable in use to a resonant frequency of the tank; the power supplied by each pulse train charges the tank (at the dielectric discharge gap) and maintains the tank at a threshold where discharge ignition occurs; and wherein discharge ignition events per pulse train (such as discharge ignition events occurring during any one pulse train) are limited to a maximum number of times based on the drive circuit being arranged in use to prohibit each pulse train from transferring power to the resonant tank after the maximum number of times has occurred.

[0013] By supplying a pulse train of power to the resonant tank, an increasing amount of energy is stored in the resonant tank over the duration of each pulse train, also known as "charging" the resonant tank. A dielectric barrier discharge occurs when the potential difference across the gap exceeds a threshold (V th ) across the dielectric discharge gap. By adjusting the pulse frequency (intended to mean the period between individual pulses or the reciprocal of the cycle period of the pulses in the pulse train) of the pulse train to the resonant frequency of the tank, the charging process causes a rapid increase in the amplitude of the potential difference. This increases the potential difference amplitude to a threshold, for example, in less than 10 cycles, at which a dielectric barrier discharge occurs (also known as the "ignition threshold").

[0014] The use of the device of the first aspect provides a limit to the current application stress. The limit to the current application stress is achieved using such a device by increasing the potential difference relative to the threshold that occurs over several cycles (i.e., individual pulses) during the pulse train due to the resonant tank voltage gain, which results in reduced power losses in the drive circuit. In conventional pulsed plasma systems, the plasma discharge is provided by the use of a single pulse, which requires a high step-up transformer, resulting in higher currents and thereby increasing the current application stress on the primary winding.

[0015] Additionally, the power supply is protected from short circuits without the need for overcurrent detection due to the inductance of the resonant tank providing sufficient impedance to limit the current if the output terminals of the power supply are shorted, for example, due to a short fault in the dielectric barrier.

[0016] Additionally, by limiting the number of discharge ignition events, energy dissipation to simply heat or to the generation of less reactive species is reduced. Indeed, the inventors have found that by implementing such a hybrid of resonant AC and limited pulse excitation, effective pollutant abatement can be provided while also having high power conversion efficiency.

[0017] Overall, therefore, the device according to the first aspect achieves high efficiency power transfer to the dielectric barrier discharge device (through resonant operation) while limiting current application stress and protecting against short circuits in order to protect circuit components.

[0018] A dielectric discharge gap is intended to be the gap between the electrodes of a dielectric discharge device. This typically provides a capacitance due to the gap, with additional capacitance provided by the dielectric. Of course, when a drive circuit according to the first aspect is connected across the discharge gap, the edges / sides of this gap are provided by the electrodes, and therefore it is intended that the drive circuit be connected (i.e., electrically connected) to at least the electrodes so as to enable the drive circuit to supply current to the electrodes and establish a potential difference between the electrodes. In various examples, the drive circuit may be further connected across the dielectric discharge gap by being connected to a wire or cable connected to the electrodes that forms a closed circuit including the drive circuit and the dielectric discharge gap.

[0019] The cycle period of the power supplied by the resonant tank is intended to refer to the period of time it takes for the current and / or voltage to go through (only) a single oscillation cycle as determined by the frequency. In other words, it is intended to be the time it takes for the current and / or voltage to go through (only) a single wavelength.

[0020] Additionally, the term "discharge" is intended to mean any form of electrical discharge, such as a plasma-generating discharge. Typically, this refers to the emission and transfer of electricity in an applied electric field through a medium such as a gas. This is typically achieved by a flow of electrons in the form of a filament from one location to another or between two points. The electron flow is typically a transient flow of electrons in the form of a filament. This is intended to mean that the flow of electrons within the microdischarges / filaments during a discharge lasts only a short time per individual discharge ignition event. Of course, there can be many filaments over time if favorable conditions are maintained. The discharge allows for the transfer of electricity in an applied electric field through a gas.

[0021] The presence of a dielectric in a dielectric discharge gap typically does not produce an arc or spark (i.e., a discharge that generates a sustained current between the electrodes). Instead, it typically produces only a microdischarge, which typically lasts only for microseconds. This limits the amount of power required to provide a sustained discharge while providing the energy and constituents necessary to contribute to chemical reaction pathways for decomposing compounds in the medium through which the discharge passes.

[0022] By providing such a discharge, it is possible to cause the transfer of effective power to the medium by generating high-energy electrons that interact with the fluid. This is because when effective power is transferred to the medium, electrical energy is converted into chemical energy, allowing the decomposition of the medium or its components. This conversion occurs in circuits, electrodes, dielectrics, etc. This can cause losses due to several factors, such as losses in the electron beam and / or heating of the medium. While typically undesirable, such losses can be unavoidable in the process. Therefore, losses can be minimized to have the maximum production rate of high-energy electrons.

[0023] Turning to the process by which a discharge is triggered by the drive circuit according to the first embodiment, it can be considered that no discharge occurs initially before the ignition threshold is reached. This means that the gas in the discharge gap (e.g., between the electrodes) is not ionized, no discharge occurs, and, particularly relevant, no power is supplied to the gas. However, once the threshold is reached, a discharge occurs. This results in the formation of countless transient filaments (each representing a microdischarge) from a single point (e.g., some form of sub-macrostructure on the surface of the electrodes defining the sides of the discharge gap). The lifetime of each filament (i.e., the duration of its existence) is on the order of tens of nanoseconds. It is only during the lifetime of these transient microdischarges that high-energy electrons are formed in the discharge gap, allowing power to be supplied to the medium within the gap. The power supplied by the generated high-energy electrons is at an energy level sufficient to initiate chemical reactions, thereby initiating the decomposition of pollutants.

[0024] Maintaining the discharge gap at the voltage threshold indefinitely would cause charge buildup on the electrodes and dielectric barrier surfaces of the DBD device's dielectric discharge gap. This can be avoided by using pulses. Pulses can be thought of as limiting the amount of time the instantaneous voltage in the discharge gap is maintained at the ignition threshold to a period on the order of a few microseconds due to the alternating polarity provided by the pulse. This means that transient filaments can only be generated during this period. Therefore, the period during which microdischarges can occur can be thought of as limited to the amount of time the instantaneous voltage in the discharge gap is maintained at the ignition threshold, and the sum of these transient filaments can be considered a "macrodischarge" or "discharge event."

[0025] In view of the previous four paragraphs, the term "discharge ignition event" is therefore intended to be the initiation of a macro-discharge or discharge event, or in other words the start of a period when a micro-discharge in the form of a transient filament can occur, when a threshold is reached. This threshold is typically a voltage threshold, such as a voltage threshold in the dielectric discharge gap, for example in the form of a potential difference (e.g., ΔV) between the electrode / dielectric layer and the electrode defining the gap.

[0026] The pulse frequency of the pulse train that is adjustable to the tank's resonant frequency (which may also be referred to as the "resonant frequency") in use is intended to mean that the pulse frequency can be adjusted to one or more of several frequencies that can be considered resonant frequencies. These include theoretical resonant frequencies (i.e., frequencies calculated to be resonant frequencies when not considering real-world effects) or actually applicable resonant frequencies, such as frequencies that consider real-world effects that may include one or more of inductance and / or resistance, damping, or impedance in wiring and / or other components. Hence, the zero-voltage switching frequency, as described in further detail below.

[0027] The maximum number of discharge ignition events may typically be 1 to 5 events, such as 1 to 3 events, including 1 event (only), 2 events, or 3 events. By limiting to very few discharge events, the inventors have found that this results in the most energy-efficient and effective decomposition of contaminants. This is due to the energy transfer generated by the discharge ignition events limiting transfer to the medium in the discharge gap, thereby directing a higher proportion of the energy to cause decomposition of compounds within the medium.

[0028] The drive circuitry may further include a phase meter in communication with the tank and arranged, in use, to identify (such as by monitoring) a phase shift in the power supplied to the tank during each pulse train, the phase shift corresponding to the occurrence of a discharge ignition event, and the drive circuitry may further be arranged, in use, to determine when a maximum number of discharge ignition events has occurred based on the number of pulses in each pulse train since each respective discharge ignition event.

[0029] The inventors have found that such a phase shift represents the start of a discharge, and therefore it is possible to identify the number of discharge ignition events that occur from that point on (such as by counting or recognizing the number of pulses in the pulse train from that point onward). This means that it is possible to determine when a maximum number of discharge ignition events has been reached and stop further discharge ignition events from occurring. For example, an initial discharge ignition event can be detected by monitoring the voltage-current phase shift at the input to the resonant tank (such as the voltage-current phase shift measured at the H-bridge terminals; the relevance of the H-bridge is discussed in more detail below). While the resonant tank is charging (e.g., a rapid voltage rise), there is typically a phase shift (excited at resonance) that is close to zero. However, once a plasma is ignited as part of a discharge ignition event, there is typically a shift in the resonant frequency due to the increased capacitance applied by the "ignited" discharge gap. When monitored, this resonant frequency shift can be readily detected by monitoring the phase shift.

[0030] A phase meter (e.g., a phase detection unit) as described above may be provided by a controller, processor, microprocessor or microcontroller, or another such device capable of monitoring the phase of at least two signals.

[0031] In addition to, or instead of, using phase monitoring or a phase meter, each pulse train can have a pre-adjusted or optimized number of pulses (i.e., the number of pulses in the pulse train). Typically, it is possible to calculate or model how many pulses are needed to charge the resonant tank, and typically there is (only) a single discharge ignition event per pulse, or at least it is possible to calculate how many discharge ignition events will be triggered per pulse. This allows the number of pulses in the pulse train to be set to at least the maximum number of desired discharge ignition events plus the number of pulses needed to charge the tank. If such an approach is used, it will be appreciated that there may be additional pulses included in each pulse train, such as when pulses are used to discharge the resonant tank. These may also be included in the calculation of how many pulses are needed per pulse train if this approach is used.

[0032] The circuit may further include a power storage device connected across the power supply, arranged in use to receive and store power discharged from the tank (i.e., discharged power) after each pulse train (or after a maximum number of discharge ignition events have occurred). This provides a means for storing / recovering power in the circuit that would otherwise be lost due to energy dissipation in the resonant tank. This reduces energy loss between pulse trains and allows the stored energy to contribute to forming the next high voltage pulse train, resulting in improved efficiency.

[0033] Energy or power regeneration can be achieved by passive or active means. Typically, active means are used, such that the drive circuit is typically arranged, in use, to shift the phase of the pulse train (pulses within) by 180 degrees (°) after a maximum number of discharge ignition events have occurred. By implementing this mechanism, energy recovery can be achieved when passive means (and potentially any other active means) for energy recovery are not possible, such as by using a loosely coupled air-core transformer. This therefore This allows the efficiency gains achievable from energy recovery to be further achieved. The phase shift may be applied to the same number of pulses as those used in the pulse train to charge the resonant tank to threshold, but it is also possible to apply the phase shift to a different number of pulses. This maintains a similar power flow when charging and discharging the resonant tank.

[0034] The circuit may further include an inverter between the power source and the tank, the inverter being arranged to regulate the supply of power from the power source to the tank, in use. This allows the characteristics and properties of the power supplied to the resonant tank to be determined by components within the circuit rather than by any input to the circuit. This provides a greater amount of customisation and modification to be made than if this were determined by the power supplied at the circuit input.

[0035] The inverter may be any suitable type of inverter. Typically, the inverter is an H-bridge or half-bridge. This provides a simple mechanism for providing the inverter function while also allowing direct and easy control over the output from the inverter to achieve passive and / or active regeneration of the energy stored in the tank at the end of each pulse train.

[0036] When an H-bridge or half-bridge is used, the switches used in the bridge inverter may be any suitable switches, such as mechanical switches or power transistor switches. Typically, each switch in the inverter may be a silicon or silicon carbide (metal-oxide-semiconductor field-effect transistor, MOSFET) switch, a silicon insulated-gate bipolar transistor (IGBT) switch, or a gallium nitride power transistor (FET) switch. Silicon MOSFET switches typically have a blocking voltage of about 650 V, silicon carbide (SiC) MOSFET switches typically have a blocking voltage of about 1.2 kV, silicon IGBT switches typically have a blocking voltage of about 650 V or about 1.2 kV, and gallium nitride FET switches typically have a blocking voltage of about 650 V. It is also possible to achieve a higher blocking voltage bridge leg using a multilevel bridge leg with several low-voltage devices connected in series. However, a mechanism is typically required to ensure that the voltage is shared evenly among the switches, which makes the system more complex and less robust. This is why a two-level H-bridge is typically used in the drive circuit according to the first aspect. The use of such switches in inverters also makes it possible to keep components simple. Wide bandgap (WBG) semiconductors such as SiC and GaN are typically used due to their superior performance over Si-based power semiconductors.

[0037] The pulse frequency supplied to the resonant tank (such as the frequency of the voltage waveform when provided as a pulse train) may be exactly at the tank's resonant frequency, such as the frequency of the first harmonic (i.e., the fundamental or natural frequency), or near the resonant frequency, such as within the resonant frequency range. When higher harmonics are used, the resonant tank typically has a low-pass characteristic, which attenuates or weakens harmonics higher than the first harmonic. This is why the resulting current and voltage across the dielectric discharge gap are almost perfectly sinusoidal, even though the excitation is typically provided with a square waveform.

[0038] When an inverter using switches is used, such as an H-bridge or half-bridge inverter, the pulse frequency of each pulse train can be the zero voltage switching (ZVS) frequency. This is typically slightly higher than the exact resonant frequency of the tank, for example, about 5% to about 10% higher than the exact resonant frequency, but not more than about 10%, depending on the quality (Q) factor of the circuit. This reduces losses caused by switching and reduces the switching This reduces the electromagnetic interference (EMI) caused by inverter operation, thereby making the inverter more efficient and reducing the noise generated by the inverter.

[0039] The circuit may further include a transformer, the secondary winding of which forms part of the resonant tank, the transformer being a step-up transformer that increases the voltage input level by increasing the dielectric barrier discharge voltage level (i.e., V th (Note: The use of a transformer reduces the minimum voltage gain required in the resonant tank to achieve a 100% rated voltage.) Additionally, the use of a transformer reduces ground currents (currents flowing in the parasitic capacitance between the electrodes of the DBD device and any surrounding metallic housing), thereby reducing EMI. A transformer can be placed in a circuit with the primary winding forming part of the resonant tank instead of the secondary winding, but in arrangements where the secondary winding forms part of the resonant tank, the kilo-volt-ampere (kVA) rating of the transformer can be reduced. In such cases, the reactive power of the DBD device can be compensated.

[0040] If a transformer is used, the circuit may be arranged to short-circuit the primary transformer winding after each pulse train in use. When energy is being recovered / regenerated from the tank, the short-circuiting of the primary winding is typically applied after the energy is recovered, such as after each pulse train has elapsed. Shorting the primary winding reduces ringing that may be generated by the components that make up the resonant tank. When an inverter is used, shorting the primary winding of the transformer may be achieved in use by switching on the low or high side of the inverter. This eliminates the need for additional components in the circuit, thereby limiting the number of components.

[0041] The inductance of the resonant tank may be provided or contributed by one or more components, or may be provided by inductance in the wiring or cables between components in the circuit. At least a portion of the inductance (such as some or all of the inductance) may be provided by a transformer. This utilizes the typically undesirable characteristics of a transformer and allows those characteristics to be used as a contribution to the circuit's functionality. Any inductance provided by the transformer may be the transformer's leakage inductance (also called stray inductance). In some situations, this may allow the resonant tank to not also include an inductor as a specific component.

[0042] As described in more detail below, the transformer may be an air-core transformer. When an air-core transformer is used, it may have up to 60% magnetic coupling between its windings. The use of an air-core transformer, such as an air-core transformer with 60% magnetic coupling between its windings, increases the inductance that can be provided by the transformer, reducing the need for any additional inductance in the resonant tank. Furthermore, when using an air-core transformer, the resonant inductance, and therefore the resonant frequency of the resonant tank, can be adjusted by adjusting the distance between the primary winding (also called the transmit coil) and the secondary winding (also called the receive coil). This reduces the need for additional capacitors in the circuit, as is known to be done in existing systems, thereby reducing the number of components. This is achievable through the planar inductive power transfer that occurs when using an air-core transformer. Other configurations that allow for the implementation of an air-core transformer are also possible.

[0043] Air-core transformer windings have low coupling compared to other transformers (i.e., non-air-core or solid-core transformers). This allows the secondary (i.e., high-voltage) side of the transformer to oscillate freely when no voltage is applied from the primary (such as when all switches are off and the body diodes are not conducting). The measures for active energy recovery detailed above (i.e., 180° phase shift of some pulses) eliminate these oscillations and avoid power losses when air-core transformers are used.

[0044] The transformer may have a step-up ratio between the primary and secondary transformer windings of about 1:1 to about 1:10, for example about 1:5. By applying this arrangement, the following equation holds true, which is not generally true in known systems:

[0045]

number

[0046] For a dielectric barrier discharge ignition voltage threshold in DBD devices of approximately 20 kV, this means that a minimum resonant tank voltage gain of approximately 5 times is required for a boost ratio of approximately 1:5 when the input voltage to the drive circuit is approximately 800 V. This achieves an optimized balance between transformer boost and resonant tank voltage gain, significantly reducing the current stress on the drive circuit compared to conventional pulse power and resonant converter systems that primarily rely on high boost transformers (1:20 or greater) to achieve the required discharge voltage levels.

[0047] Until the discharge threshold is reached, there is minimal damping in the resonant tank. This is because there is no load on the resonant tank (such as power transfer to the medium in the discharge gap) during charging. Compared to known resonant systems, in such systems there is typically always a load, since there is a continuous or prolonged discharge that creates a load.

[0048] The lack of loading on the resonant tank of the drive circuit according to the first aspect results in a very high voltage gain (e.g., a gain with a Q factor greater than 50) compared to known systems. Unlike known systems, the achievable voltage gain of the resonant tank does not depend on the load (which, as noted, typically corresponds to the power transferred to the gas when a dielectric discharge occurs). Instead, it depends (only) on the parasitic resistance of the resonant tank (e.g., the parasitic resistance generated by the resistance of the magnetic material and electrodes).

[0049] Furthermore, the lack of load allows for more rapid charging and allows the pulse frequency of the pulse train to be as close as possible to the true resonant frequency of the tank (i.e., the theoretical resonant frequency, which typically does not take into account damping effects present in practice). This is because the amount of damping is so small that it only needs to be taken into account minimally when the pulse frequency is set. This increases energy transfer capability and makes the drive circuit more efficient.

[0050] If a transformer is present, the required dimensioning of the transformer's step-up turns ratio (i.e., the specifications set for the transformer's step-up turns ratio) also depends only on the parasitic resistance of the resonant tank. Similarly, if there is a load to consider, the dimensioning of the transformer's step-up turns ratio must also take this into account. This allows losses from the transformer to be kept to a minimum, thereby reducing the impact of using a transformer on the efficiency of the drive circuit compared to when the load needs to be considered.

[0051] Instead of, or in addition to, a transformer providing inductance, at least a portion of the inductance (such as part or all of the inductance) may be provided by an inductor. This provides a component designed to provide the inductance used, thereby optimizing the circuit. In situations where inductance is provided partially or entirely by an inductor and a transformer, each contributes to the inductance between the power source and the dielectric discharge gap, and thereby to the inductance of the resonant tank.

[0052] When separate transformers and inductors are provided, there are several possible circuit arrangements. One arrangement is for the inductor to be connected to the input to the resonant tank (such as the output of an inverter), which is then connected to the primary winding of a transformer, the secondary winding of which is then connected across the dielectric discharge gap. Another arrangement is for the input to the resonant tank to be connected to the primary winding of a transformer, the secondary winding of which is connected to an inductor connected in series with the dielectric discharge gap. In each of these arrangements, the leakage or stray inductance of the transformer contributes to the resonant inductance value (i.e., inductance) of the resonant tank. Naturally, if the resonant tank is placed after the transformer, the kVA rating of the transformer is reduced because the oscillating reactive power of the dielectric discharge device does not pass through the transformer.

[0053] Another arrangement is for the input to the resonant tank to be connected to the primary winding of a transformer, with the secondary winding of the transformer connected across the dielectric discharge gap. In this arrangement, because no separate inductor component is provided, the leakage or stray inductance of the transformer must be large enough to compensate for the loading on the dielectric discharge gap at the desired resonant frequency. This can be achieved with a transformer that has very low coupling between the windings, such as in the case of an air-core transformer (i.e., no magnetic core), as will be discussed in more detail below.

[0054] According to a second aspect, there is provided a system for providing a dielectric barrier discharge, the system comprising a dielectric barrier discharge device having at least two electrodes having a fluid gap therebetween that defines a dielectric discharge gap, a dielectric layer being located between the at least two electrodes, and a drive circuit according to the first aspect, wherein a power supply for the drive circuit is connected across the dielectric discharge gap.

[0055] The dielectric layer may be located between the electrodes, such as in the discharge gap, but may not contact the electrodes. Typically, at least one electrode may have a dielectric layer attached to it (or the dielectric layer when there is only a single dielectric layer).

[0056] Sub-macrostructures may be attached to at least one electrode. Applying sub-macrostructures to an electrode or dielectric portion (when a dielectric portion / layer is attached to the electrode) is a technically challenging process due to the need to maintain order within the structure and the difficulty of attaching the structure to the surface of the electrode or dielectric portion. Additionally, using sub-macrostructures implements a "plate-to-point" structure, which results in differences in the uniformity of the electric field strength because the electric field strength at the edge of the structure is higher than on (for example) the electrode, which typically has a larger area over which the electric field spreads. However, the inventors have found that using sub-macrostructures in a dielectric barrier discharge device allows less power to be used. This is because, during use, when an electric field is established between the anode and cathode, the structure's electric field emits electrons. Field emission increases the density of electrons in the gap between the anode and cathode. This saves power because more electrons are available to initiate chemical reactions. This, when used in physical applications, typically involves the classical electrostatic phenomenon of dielectric barrier discharge and a form of field emission, when classical and quantum processes are kept separate from each other. This is achieved by combining quantum phenomena of tunneling with quantum mechanics.

[0057] A structure connected to at least one electrode or dielectric portion / layer is intended to mean that at least one structure is connected to at least one electrode or dielectric. This means that two or more electrodes and / or dielectric portions can have one or more structures connected to them. Of course, there may be multiple structures, and each structure may be connected to one of the electrodes or dielectric portions; for example, all structures may be connected to only a single electrode or dielectric portion, or to one or more electrodes and / or dielectric portions having one or more structures connected to it. When a structure is connected to an electrode or dielectric portion, it is intended that the structure is connected only to its respective electrode or dielectric portion, and not also to another electrode or dielectric portion (when connected to an electrode).

[0058] The sub-macrostructures may be nanostructures. Nanostructures may be nanowires, nanotubes, or nanohorns of carbon, silicon, titanium oxide, or manganese oxide, or microneedles of stainless steel, aluminum, or titanium. The nanostructures may typically be carbon nanotubes (CNTs). CNTs have been found to be very good field emitters of electrons when exposed to an electric field. CNTs and other materials can generate large numbers of electrons at relatively low applied voltages due to their very high aspect ratios (typically 50 to 200 nanometers, or nm, diameters to 1 to 2 millimeters, or mm, lengths, i.e., aspect ratios of 5,000 to 40,000) and their low work functions (typically about 4 electron volts, or eV). The high aspect ratios cause large electric field enhancements at the tips of CNTs, of several volts per micrometer (also called a micron) (V / μm), achievable at low applied voltages. The minimum electric field strength required for field emission from CNTs is typically about 30 V / μm. This can be achieved by varying one or more of the following: the length of the CNTs, the diameter of the CNTs, the distance between the electrodes used to generate the electric field, and the applied voltage used to establish the electric field. If an array of CNTs is used, the density of the array can also be varied to vary the electric field strength, as the CNTs tend to shield each other.

[0059] The nanostructures can be multi-walled CNTs (MWNTs) or metallic single-walled CNTs (metallic SWNTs).

[0060] The structure may be electrically connected to at least one of the electrodes. Additionally or alternatively, the or each electrode to which the or each structure is electrically connected may be arranged to provide a cathode in use.

[0061] The nanostructures may have a length-to-width aspect ratio of at least 1,000 (i.e., 1,000 to 1). Nanostructures with an aspect ratio of at least 1,000 provide more efficient field emission than those with lower aspect ratios. The aspect ratio may be at least 5,000 or at least 10,000. Increasing the aspect ratio has been found to further increase the efficiency of field emission.

[0062] The electrodes may be of any suitable material to provide electrodes that allow an electric field to be established between them. Typically, the electrodes may be made from a conductive metal.

[0063] The dielectric portion may be connected to a first electrode (such as an anode) and the structure may be connected to a second electrode (such as a cathode). This allows the application of the dielectric portion and the structure to each electrode to be independent, meaning that the process for applying the dielectric portion to the electrode and the process for applying the structure to the electrode are independent of each other. This therefore simplifies the manufacturing process of the device and reduces the failure rate during manufacturing.

[0064] The use of the dielectric portion and the structure provides a synergistic effect of reducing the power and voltage required to establish a dielectric barrier discharge. In addition, the use of the dielectric portion allows the dielectric barrier discharge to become more controllable by reducing the amount of sparking, thereby reducing the amount of wear and damage caused by the dielectric barrier discharge. If the structure were used without the dielectric portion, a greater amount of sparking would limit the usefulness of the structure, as it is typically more susceptible to spark damage than other parts of the device. Conversely, if the dielectric were used without the structure, the density of electrons to initiate breakdown in the fluid passing between the electrodes would be lower, and therefore more energy would be required to achieve the same reduction efficiency. Therefore, the combined effect of using the dielectric and the structure has a greater benefit than the benefit provided by using each independently.

[0065] The dielectric portion may be one or more of mica, quartz, alumina (i.e., Al2O3), titania, barium titanate, fused silica, titania silicate, silicon nitride, hafnium oxide, or ceramic. The phrase "one or more" in this case is intended to mean a combination of two or more of two or more specified materials when two or more of these materials are used.

[0066] Typically, the dielectric portion is quartz because it is readily available, low cost, can be processed in large quantities, and has high resistance to thermal stress. Alternatively, the dielectric portion may be mica, which is beneficial because it has a slightly higher dielectric constant than other dielectric materials such as glass.

[0067] The system can further include a controller connected to the drive circuit, the controller configured to adjust, in use, the power supplied to the tank of the drive circuit based on inputs provided to the controller. This allows for modification of the power supplied to the resonant tank in use, providing the ability to make changes when parameters within the system change during use, causing changes in characteristics within the system. For example, changes in fluid passing between the electrodes can cause a change in the capacitance of the resonant tank, changing its resonant frequency. The controller can then be used to adjust the pulse frequency supplied to the resonant tank during a pulse train.

[0068] The controller may be arranged to adjust, in use, the pulse frequency (such as the frequency of the voltage or current waveform), and / or the pulse train frequency, and / or the number of pulses in a pulse train, and / or the number of pulse trains, and / or the pulse train repetition frequency. This provides a wide range of adjustments that can be made to allow the power provided to be adjusted to provide optimal dielectric barrier discharge generation during use of the system.

[0069] The inputs provided to the controller may include one or more relevant parameters. Typically, the inputs include the voltage and current at the output of the drive circuit, such as the output of an inverter. This allows the phase angle between the supplied voltage and current and the average phase of the pulse train to be calculated. This can be used to optimize the pulse frequency supplied in the pulse train. Thus, the controller may be arranged, in use, to determine (by "calculate") the phase difference between the voltage and the current. This may, of course, be determined by further components.

[0070] As mentioned above, this phase difference is used to detect the onset of the dielectric barrier discharge. A phase meter may also be used. Detecting this can enable identification of when a pulse train transitions from providing energy to recovering energy, for example, after a defined number of discharge ignition events. Also, as noted above, the occurrence of a dielectric barrier discharge in the discharge gap increases the effective capacitance. This results in a lower resonant frequency and therefore an increased measurable phase difference for a given drive frequency (e.g., the pulse frequency of the pulse train). With this in mind, it can be appreciated that the phase meter and controller of the drive circuit may be the same component. Alternatively, the controller and phase meter may be in communication with each other, or the controller may incorporate the phase meter, such that the phase meter is a component of the controller.

[0071] The drive circuit may include an inverter between the power source and the resonant tank of the drive circuit. In this case, the voltage and current may be supplied from the output of the inverter. This allows for a more granular (i.e., more precise) level of control of the output provided to the resonant tank than would be achievable if an AC power source were simply connected to the resonant tank to provide power, due to the higher frequencies achievable using the inverter. In addition, higher AC frequencies, such as those achievable using the inverter, can provide a shorter dielectric barrier discharge. This allows for simpler limitation of the maximum number of discharge ignition events and faster control than would be possible if a standard AC power source were used to maintain the efficiency gain achieved by limiting the number of discharge ignition events.

[0072] The controller may further be connected to the dielectric barrier discharge device, and the inputs include one or more properties of a fluid passing through the device in use, which allows the properties of the fluid to be taken into account when attempting to optimize the performance of the system.

[0073] The system may include a plurality of dielectric barrier discharge devices and a plurality of drive circuits, each drive circuit connected across the dielectric discharge gap of one or more dielectric barrier discharge devices, optionally with only a single power supply arranged in use to provide power to all the drive circuits, thereby enabling the system to be scaled to accommodate different volumes of fluid passing through the system, such as different sizes of engine exhaust gas to be cleaned.

[0074] According to a third aspect, there is provided a method of controlling a dielectric barrier discharge in a dielectric discharge device, the method comprising: supplying power to a resonant tank with a series of electrical pulse trains, wherein a pulse frequency of each pulse train is tuned to a resonant frequency of the tank, the resonant tank being connected across a gap between electrodes in the dielectric discharge device, a capacitance of the tank being provided by the dielectric discharge device, and the power supplied by each pulse train charges the tank and maintains the tank at a threshold at which discharge ignition occurs; providing a maximum number of discharge ignition events per pulse train by prohibiting each pulse train from transferring power to the resonant tank after the maximum number of discharge ignition events have occurred; and prohibiting power transfer to the tank between pulse trains.

[0075] The term "prohibit" is intended to mean passively or actively prohibiting the transmission of power to the tank, such as by not providing a path through which power can pass to the tank or by diverting the path to an alternate circuit, respectively.

[0076] As mentioned above, the maximum number of discharge ignition events may be between 1 and 5 events.

[0077] The method includes identifying a phase shift in the power supplied to the tank during each pulse train, the phase shift corresponding to the occurrence of a spark ignition event, and determining when a maximum number of spark ignition events will occur based on the number of pulses in the pulse train since each respective spark ignition event. This provides an accurate means for avoiding exceeding a maximum number of events.

[0078] Each electrical pulse train may be a voltage pulse train, by which it is intended to mean that the electrical pulse train may be provided by a voltage pulse train, such as a voltage waveform that may be used as an excitation waveform for the resonant tank and that may induce a current waveform in the resonant tank.

[0079] The method may further include modulating the pulse frequency, and / or the frequency of the pulse train, and / or the number of pulse trains in the series of electrical pulse trains, and / or the number of pulses in each pulse train. It is worth noting that the power frequency can be modulated by modulating the power or a component of the power, such as the voltage and / or current. The frequency of the power is twice the frequency of the voltage waveform contributing to the power (the frequency that the pulse frequency is intended to represent), which is the case in a typical power system. If the voltage and current are each sinusoidal, the power is the square of the sinusoidal waveform (i.e., Sin^2), and the spectral decomposition exhibits a fundamental frequency that is twice the excitation (i.e., voltage) frequency.

[0080] The modulation may be based on a phase difference in the characteristics of the power supplied to the resonant tank and / or one or more properties of the fluid passing through the device.

[0081] Power may be supplied to the resonant tank via a transformer, and the method further includes shorting a primary winding of the transformer between the repetitive pulse trains, which prevents (i.e., reduces) undesired oscillations between the magnetizing inductance of the transformer and the capacitance of the DBD reactor.

[0082] The pulse frequency of each pulse train supplied to the resonant tank may be set by switching circuitry between the power supply and the resonant tank.

[0083] For each pulse train, the resonant tank may be discharged (i.e., drained) after a maximum number of discharge ignition events have occurred. This may be achieved through active or passive regeneration. Under such circumstances, the method may further include storing the energy drained from the resonant tank by the discharge. Recovering energy in this manner significantly improves the energy efficiency of the method.

[0084] Typically, there is a time difference between the end of one pulse train and the start of the next pulse train. In other words, there may typically be a period of no pulses between the end of one pulse train and the start of the next pulse train, which allows one pulse train to be distinguished from the next pulse train and avoids any coincidence or overlap between consecutive pulse trains.

[0085] Exemplary circuits and methods of operating the exemplary circuits are described in detail below with reference to the accompanying drawings. [Brief explanation of the drawings]

[0086] [Figure 1] FIG. 1 shows an exemplary plot of voltage and current in a pulse train according to a prior art device.

[0087] [Figure 2] FIG. 1 shows a schematic diagram illustrating the principles of electron irradiation and dielectric barrier discharge scrubbing cleaning techniques in an exemplary dielectric barrier discharge device.

[0088] [Figure 3] FIG. 2 illustrates an exemplary plot of voltage, current, and power applied to an exemplary circuit.

[0089] [Figure 4] FIG. 10 shows an exemplary plot of voltage versus time comparing applied gap voltage to output voltage and a corresponding plot with a magnified portion of output current versus time.

[0090] [Figure 5] FIG. 1 illustrates an exemplary circuit.

[0091] [Figure 6] FIG. 10 illustrates a further exemplary circuit.

[0092] [Figure 7] FIG. 10 illustrates another exemplary circuit.

[0093] [Figure 8] FIG. 1 illustrates an exemplary method of operating an exemplary circuit.

[0094] [Figure 9] FIG. 10 illustrates an exemplary plot of a switching sequence over time and the resulting voltage over time.

[0095] [Figure 10] FIG. 10 illustrates an exemplary plot of voltage over time versus power transfer rate.

[0096] [Figure 11] FIG. 2 illustrates an example controller for an example circuit.

[0097] [Figure 12] FIG. 10 shows further exemplary plots of voltage and current over time during an exemplary pulse train.

[0098] [Figure 13] FIG. 10 illustrates a further exemplary controller.

[0099] [Figure 14a] FIG. 10 illustrates an exemplary plot of a switching sequence over time and the resulting voltage over time.

[0100] [Figure 14b]FIG. 10 is another diagram illustrating an example plot of a switching sequence over time and the resulting voltage over time.

[0101] [Figure 15] FIG. 10 illustrates an example plot of resonant tank input voltage and current and resulting DBD device voltage versus time without energy recovery.

[0102] [Figure 16] FIG. 10 illustrates an example plot of resonant tank input voltage and current and resulting DBD device voltage versus time with energy recovery. DETAILED DESCRIPTION OF THE INVENTION

[0103] When using DBD devices, a pulse system can be used to ignite the dielectric barrier discharge between the electrodes in the device. As mentioned above, available high-voltage pulse power equipment for industrial-scale DBD systems typically uses a low-voltage pulse-generating unit with a peak output pulse voltage of 400 V to 1000 V, followed by a step-up transformer with a turns ratio of 1:20 to 1:40 to meet the required dielectric barrier discharge voltage level.

[0104] The characteristic voltage and current waveforms of a single pulse using a conventional high-voltage pulse generator are shown in Figure 1. This shows two plots, one of voltage versus time and the other of current versus time, for a prior art single pulse generated using a high-voltage pulse modulator system used to charge large DBD devices.

[0105] The voltage plot shows that the pulse starts at 0 V and then rises to a peak of about 22 kV over about 1 microsecond (μs). The voltage then decreases from the peak to a level of about 12 kV over another 1.5 μs. The voltage decrease then slows to a linear decrease to 0 V over about 21 μs.

[0106] The dip from the peak is caused by a natural resonance between the DBD device and the transformer parasitics. The resonance initiates oscillations, which can be seen occurring in the dip from the peak. The pulse termination then stops the resonance and cuts off the supplied voltage. From that point on, therefore, a linear discharge occurs. If the pulse had not been terminated, a periodic waveform would be seen instead.

[0107] The corresponding current plot shows the current increasing from 0 A to a peak of about 90 A over about 0.5 μs, then dropping to about −40 A (negative 40 A) over about 1 μs, and then returning to 0 A over another about 1 μs.

[0108] The change in current occurs over the same period it takes for the voltage to pass through its peak and return to 12 kV. The dielectric barrier discharge begins at approximately the point when the voltage reaches its peak and ends when the voltage returns from the peak to 12 kV. The linear slope from this point back to 0 V is due to energy dissipation in the pulse-generating unit from the energy stored in the capacitance of the DBD device after the dielectric barrier discharge has occurred.

[0109] As mentioned above, the low power factor PF, determined from the ratio of real power to apparent power in DBD devices, i.e., the large amount of reactive power required to repeatedly cycle the voltage in the reactor and the relatively low amount of real power actually transferred to the plasma, poses a fundamental challenge to achieving high power transfer efficiency.

[0110]

number

[0111] To overcome this drawback, the inventors have developed exemplary devices, systems, and methods, which are described in detail below. Such devices can be used to scrub exhaust gases, such as the device disclosed in GB Patent No. 2010415.4, which is incorporated herein by reference. This device utilizes submacrofeatures, carbon nanotubes (CNTs), and a functionalized electrode having a dielectric portion. The submacrofeatures are exposed to an electric field, resulting in field emission of electrons from the CNTs and a dielectric barrier discharge between the dielectric and the counter electrode. The gas being scrubbed is then exposed to these electrons.

[0112] The phrase "functionalized electrode" is intended to mean an electrode that, in addition to acting as an electrode (i.e., as an anode and / or cathode), has one or more structures thereon, such as a coating, that have a functional aspect.

[0113] DBD Device Figure 2 shows a schematic diagram of the principle of this electron irradiation and dielectric barrier discharge scrubbing technique. Two electrodes, an anode 110 and a cathode 120, are positioned opposite each other. In this example, a dielectric portion 125 is positioned on the anode. This dielectric portion provides a coating over the entire surface of the anode.

[0114] The example of Figure 2 also includes a CNT 130 located between the anode 110 and the cathode 120. In this example, the CNT is electrically connected to the cathode. In other examples, other sub-macro features, such as microneedles or microneedle arrays, can be used in place of or in addition to one or more CNTs. These can function and operate in the same or similar manner as the CNTs are described below as functioning.

[0115] In use, the CNTs 130 or other sub-macro features emit electrons (e, e) in response to the presence of an electric field between the anode 110 and the cathode 120 when a potential difference is established between them. - The electric field between the anode and the cathode also induces a dielectric barrier discharge (in the form of a dielectric barrier discharge) between the dielectric portion 125 and the cathode 120.

[0116] The electrodes are coupled to the housing to position the dielectric portion 125 and CNTs 130 in proximity to a vessel 140 containing a gas (g) to be scrubbed so that the interior of the vessel can be exposed to field-emitted electrons and a dielectric barrier discharge.

[0117] For a compact arrangement, the anode 110 and / or cathode 120 can be mounted inside a container (such as a chimney), with the dielectric portion 125, CNT 130, and cathode surface each extending into the chimney, allowing the dielectric barrier discharge and electrons to traverse its cross section. However, many other arrangements are possible. For example, the dielectric portion and / or CNT and cathode surfaces can be located outside but close to the container, with a window (opening) in the container side allowing electron access and a surface where the dielectric barrier discharge can begin / end. Such an arrangement can be selected, for example, to make it easier to retrofit the device into an existing chimney or to facilitate maintenance of the dielectric and / or CNT portions of the device. The cathode and housing do not need to be located in the same location.

[0118] In industrial environments, for example, it may be more practical to use arrays of CNTs rather than individual CNTs. It may also be beneficial to provide multiple sets of anode-dielectric-cathode-CNT devices. Such larger-scale arrangements may be within chimneys, and it may be envisioned to have multiple sets of anode-dielectric-cathode-single CNTs, or a single set of anode-dielectric-cathode-CNT arrays.

[0119] Wavelet Pulse Train Using DBD devices such as those implementing the apparatus shown in FIG. 2, the inventors have developed a process for implementing high-frequency sinusoidal waveforms with varying amplitudes, similar to wavelet-type waveforms. In various examples, wavelets are generated by connecting an inductor in series with the DBD device, which provides capacitance. This forms a series resonant circuit, also known as a series resonant tank, that can be excited at the resonant frequency. When excited at the resonant frequency repeatedly over several cycles using bipolar voltage pulses, this This allows the DBD device to be excited at a high voltage slew rate with significantly reduced current stress, which lowers the peak power handled by the power electronics. The voltage gain achieved in the resonant tank therefore provides a high ignition voltage level for the DBD device instead of using a pulse transformer with a high turns ratio to provide the voltage gain. Therefore, the relevant characteristics of the resonant tank are the achievable voltage gain and the ability to compensate for the reactive power of the DBD device.

[0120] Applying several consecutive bipolar voltage pulses to form a pulse train allows for low power losses (demonstrated by the high efficiency described below) and higher pulse repetition frequencies to be applied, thus significantly increasing the average power transfer capability over systems using single pulses. By applying this process, for example, the pulse repetition frequency can be increased by at least 10 times over such systems. This can be achieved in combination with the use of silicon carbide semiconductor technology, as described in more detail below.

[0121] The repetition frequency of the pulse train is limited by the maximum operating temperature of the power electronics. Pulse power converter designs typically utilize a slow thermal response. This means that if a high pulse repetition frequency is used in a conventional pulse system, the peak power dissipated would be too great to remain within the safer operating temperature of the power electronics. This is avoided in the examples described herein by using pulse train modulation, as described below. Additionally, this is avoided by limiting the maximum number of discharge ignition events generated from a single pulse train and then allowing a period of time for cooling to occur before the next pulse train.

[0122] As described in connection with the examples provided herein, by implementing a pulse train of several consecutive bipolar voltage pulses, this is achieved while providing energy transfer with very high efficiency, such as approximately 90% or greater, even when the number of discharge ignition events is limited to one to five.

[0123] The use of successive bipolar voltage pulses produces three modes of operation induced in the DBD device, as shown in Figure 3. The first mode, occurring between 0 μs and time A in Figure 3, is the charging of the resonant circuit. This causes an increasing potential difference between the electrodes within the DBD device. As mentioned above, this is accomplished by applying successive bipolar voltage pulses at the resonant frequency of the resonant tank.

[0124] In the plot shown in Figure 3, this can be seen to be a sine wave at a consistent frequency with steadily increasing amplitude for both voltage and current. The resulting instantaneous power level of the rectified sine wave (as the square voltage multiplied by the sinusoidal inductor current) steadily increases in amplitude. The duration of the mode in the example shown in Figure 3 is approximately 2.5 voltage cycles, 2.5 current cycles, and 5 power cycles (one power cycle is the transition from zero to peak and back to zero). In this example, the current waveform leads the voltage waveform by approximately 90°.

[0125] The second mode occurs between time A and time B in the example plot of Figure 3. This mode occurs when the voltage reaches the ignition or breakdown voltage (V) that causes a dielectric barrier discharge between the electrodes of the DBD. th ) is reached. This delivers power to the plasma and should only last a few discharge cycles for most efficient contaminant reduction. During this mode, the voltage amplitude is V due to continuous excitation of the resonant tank at the resonant frequency. th The plot shows that the voltage and current continue as sinusoidal waves with a consistent frequency. The amplitude of the wave varies only slightly over the duration of this period (increasing until approximately the midpoint of the mode's duration, then beginning to decrease).

[0126] The example shown in Figure 3 is based on a DBD device with a capacitance of approximately 3.0 nF. The voltage peaks at approximately ±24 kV (positive-negative 24 kV) and the current is ±80 A. In other examples, the capacitance is approximately 1.0 nF, but may be approximately 45.0 nF or greater.

[0127] The voltage and current amplitude pattern remains the same for the instantaneous power, which remains a rectified sine wave. The peak instantaneous power is approximately 180 kilowatts (kW) for the example shown in Figure 3.

[0128] The duration of the second mode is about 1.5 voltage cycles, about 1.5 current cycles, and about 3 power cycles.

[0129] During the first and second modes, the resonant tank is excited by applying power. During the third mode, the excitation is stopped and the resonant tank is discharged by draining. In some instances, the tank is actively discharged by recovering energy from the tank. Passive discharge is also possible.

[0130] With excitation terminated and a discharge path provided, the voltage, current, and power decrease to zero in the third mode. The exemplary plot in Figure 3 shows the third mode after time B. The voltage and current follow sinusoidal waveforms with consistent frequencies, similar to the first and second modes. The power remains a rectified sinusoid. The amplitudes of the voltage and current decrease toward zero over a period of approximately 2.5 cycles for the voltage and 2.5 cycles for the current.

[0131] The power plot shown in Figure 3 is consistent with the resonant tank being passively discharged. This is evident in the inversion of the instantaneous power to become a rectified sine wave, but with negative peaks rather than positive peaks as in the first and second modes. The power amplitude decays to zero over approximately five cycles.

[0132] The three modes form a wavelet pulse power process in the form of a pulse train implemented by excitation of a resonant tank. The duration of the power transfer achieved using this process is determined by the length of time the excitation pulse train is supplied to the resonant tank. This is only one parameter of the excitation pulse train, determined by the circuit in which the pulse train is implemented. Figures 5, 6, and 7 show example circuits that can be used to implement one or more pulse trains.

[0133] An example of an excitation applied to the resonant tank is shown below in Figure 12. As can be seen from this figure, in various examples, the excitation takes the form of a square wave voltage waveform that includes a number of successive individual pulses that together form a pulse train. This induces a sinusoidal current in the resonant tank (the current waveform shown in Figure 12), providing the waveform in the DBD device shown in Figure 3.

[0134] While Figure 12 does not show the dielectric barrier discharge threshold or include detailed markings separating the first, second, and third modes, these figures show where the third mode begins. At time D in Figure 12, the voltage waveform can be seen to have a peak at its maximum positive value that has a shorter duration than the other peaks in the waveform. This occurs due to the transition from the second mode to the third mode. At this point, excitation is stopped, which means that voltage is no longer actively supplied to the resonant tank and DBD device.

[0135] Depending on the measures taken at that stage, such as whether active or passive energy recovery is used, this will cause a phase shift in the voltage waveform. In the simulation used to generate Figure 12, passive energy recovery was used, and therefore The change in the applied waveform is caused by the freewheeling of current in the H-bridge diodes. An alternative active energy recovery measure, applied in some instances, is a 180-degree phase shift, which results in the power being drained instead. These processes are described in more detail below along with an example inverter providing an H-bridge.

[0136] In various examples, the transition to the third mode in examples according to embodiments disclosed herein is applied after the maximum number of discharge ignition events. Some examples limit the maximum number of discharge ignition events to only one discharge ignition event or to a maximum of about five discharge ignition events. When only one discharge ignition event is used as the maximum number, or after the last discharge ignition event in a larger maximum number, the transition to the third mode is made shortly (e.g., immediately after) the maximum number of discharge ignition events occurs.

[0137] As for how an exemplary excitation applied to a DBD device is converted into a discharge, this is demonstrated by the plots shown in Figure 4, which shows a top plot and a bottom plot. The top plot is a plot of voltage versus time, and the bottom plot is a plot of current versus time.

[0138] The top plot in Figure 4 shows a solid line and a dashed line. The solid line is in the shape of a sine wave with a minimum at time zero. In this example, this line corresponds to the voltage applied across the DBD device. The dashed line is in the shape of a sine wave with the maximum and minimum peaks truncated to a plateau. Like the applied voltage curve, this is also in the shape of a minimum at time zero, which in this example corresponds to the voltage across the discharge gap.

[0139] The amplitude of the gap voltage is smaller than the applied voltage amplitude. As the applied voltage transitions positive, the gap voltage increases. After approximately one-eighth of the applied voltage cycle, the gap voltage becomes positive. Just before the end of the second one-eighth of the cycle, the gap voltage amplitude reaches a threshold. In Figure 4, this occurs at time α. This plateau is maintained until the applied voltage reaches a maximum value at time γ in Figure 4. At time γ, the process repeats itself, but the polarity is reversed, continuing to switch between positive and negative transitions as long as the applied voltage continues.

[0140] Compared to the first, second and third modes described above, the rise in gap voltage corresponds to, for example, a rise in voltage during the second mode after an initial drop in voltage during the second mode, from which it can be seen that discharge may occur during this period, and therefore the plateau in the gap voltage curve is due to the threshold voltage being reached.

[0141] The current plot in Figure 4 shows the current in the gap induced by the gap voltage. At time zero, it has a near-zero amplitude. It increases in the form of a sinusoidal wave. If the gap voltage does not reach the threshold voltage (such as when the plot in Figure 4 represents the voltage and current during the first or third modes), the sinusoidal wave continues uninterrupted, as indicated by the dashed line in the current plot in Figure 4. However, at time α, ignition occurs due to the threshold voltage being reached. This ionizes the medium in the discharge gap and initiates a discharge.

[0142] From time α, the gap current increases rapidly and peaks at time β, which corresponds to the zero crossing of the applied voltage. Time α is approximately the end of a quarter cycle of the applied voltage, so this is a very short period relative to the cycle of the current curve. From time β, the current decreases sinusoidally to zero at time γ, at which point it returns to its original shape and amplitude range. This cycle continues in parallel with the gap voltage and applied voltage.

[0143] As can be seen, the amplitude of the current simply increases to an amplified level.

[0144] The main current plot in FIG. 4 shows a continuous curve between time α and time γ. As mentioned above, this is the time when a discharge occurs. Therefore, this period can be considered the macrodischarge period, and time α is when the discharge ignition event occurs. However, as shown in the expanded portion of the current plot in FIG. 4, the current curve does not have a continuous shape. Instead, the curve is composed of many current spikes that are so close together that the curve appears continuous. Each spike represents a microdischarge, or transient filament, initiated from a single point on one of the electrodes (such as the submacrofeature 130 on electrode 120 shown in FIG. 2). Because the filaments provide a current path across the discharge gap, it is the connection each of these filaments provides between the opposing electrodes (one electrode 110, of course, has a dielectric layer 125 thereon, as shown in FIG. 2) that causes the current spike. These microdischarges ionize the medium in the gap, causing high-energy electrons to pass through the medium, thereby generating enough energy to promote chemical reactions that, for example, decompose contaminants in the medium.

[0145] Drive circuit structure A circuit diagram of an exemplary system suitable for providing a dielectric barrier discharge is shown generally at 1 in each of Figures 5, 6 and 7. The system includes a DBD device 10, also referred to as a DBD reactor.

[0146] The DBD reactor 10 is represented by a model in each of Figures 5, 6 and 7. The model, when in use, th The DBD device is shown in this model as being connected across the diode bridge.

[0147] The electrodes (specifically the gap between the electrodes, sometimes called the "dielectric discharge gap") and the dielectric barrier attached to one of the electrodes are represented by capacitors 12 in Figures 5, 6 and 7 because capacitance is the electrical functionality that the gap and dielectric barrier provide to the system when represented as a circuit.

[0148] The capacitance provided by the dielectric discharge gap is shown as being connected directly across the diode bridge. The capacitance provided by the dielectric barrier itself is shown as being connected to one end of the diode bridge in parallel with the capacitance provided by the gap. The other end of the capacitance provided by the dielectric barrier is not connected to the diode bridge; instead, it is connected to a drive circuit arranged to drive a dielectric barrier discharge across the gap between the electrodes.

[0149] As represented by the models in Figures 5, 6, and 7, the capacitance of the DBD device 10 is primarily determined by the capacitance of the medium (typically a gas such as air) in the dielectric discharge gap. This is because the medium typically has a dielectric constant of approximately 1, while the dielectric material typically has a dielectric constant significantly higher than 1, such as approximately 3 to 6 (measured at approximately 1 kHz and approximately 20 degrees Celsius). Because the medium and dielectric are connected in series, it is the smaller capacitance that dominates, and therefore, due to these relative dielectric constants, the effective capacitance of the DBD device is dominated by the medium.

[0150] Furthermore, the contribution from the capacitance of the medium in the gap is approximately constant and independent of the temperature of the composition of the medium in the gap. Thus, as explained in more detail below, the pulse trains used in the examples according to aspects disclosed herein limit the number of discharge ignition events to an extent that minimal change occurs to this capacitance, so that this "air gap" capacitance is approximately constant. However, the same cannot be said for known resonant systems. This is due to the extended nature of the discharge, which causes a shift in the capacitance of the medium, or the surface dielectric. Either the medium is of a different nature, as in the case where a barrier discharge device is used.

[0151] The drive circuit is shown at 20, 20' and 20" in Figures 5, 6 and 7 respectively. The drive circuit has a power supply 22 connected to an inverter 30. In the example of these figures, the power supply is provided by a DC power supply, which in the example shown is a DC link voltage source V dc- is.

[0152] 5 and 6, inverter 30 has a circuit loop connected across it with connections to the electrodes of DBD device 10 in series across the capacitance provided by the dielectric discharge gap and dielectric barrier, thereby closing the circuit loop connected across the inverter.

[0153] 7, inverter 30 has a transformer 50 connected across it. In this arrangement, connected across the inverter is a transformer primary 52. ​​The transformer secondary 54 has a connection to the electrodes of DBD device 10 that connect in series across the capacitance provided by the dielectric discharge gap and dielectric barrier.

[0154] The connection across the capacitance of the DBD device 10 in each of the examples of Figures 5, 6 and 7, and the ability to connect across this capacitance, makes the drive circuit 20 a separate circuit from the DBD device, and in some examples a separable circuit.

[0155] In the example shown in Figures 5 and 6, when the drive circuit 20, 20' is connected to the DBD device 10 as described above, a resonant tank 40 is formed between the inverter 30 and the capacitor 12 provided by the dielectric discharge gap and dielectric barrier. The inductance of the resonant tank is provided, in this example, by an inductor 42 connected in series with the capacitance. Some inductance is also provided by the wiring of the resonant tank. The inverter provides the power source for the resonant tank.

[0156] In the example shown in FIG. 7, when the drive circuit 20″ is connected to the DBD device 10 as described above, a resonant tank 40 is formed between the transformer 50 and the capacitance 12 provided by the dielectric discharge gap and the dielectric barrier. The inductance of the resonant tank is comprised of an inductor 42 in series with the transformer secondary 54 and an inductor L, designated 56. σ 7 as being connected in series with the transformer between the output from inverter 30 and the input to the transformer primary 52.

[0157] The transformer 50 shown in the example of FIG. 7 also includes an inductor L, referenced 58, connected in parallel with the transformer primary 52. m The magnetization induction is represented graphically by

[0158] In addition to providing step changes in voltage and current based on the turns ratio of the transformer 50, the transformer also provides galvanic isolation, which suppresses electromagnetic interference across the transformer from the inverter 30 to the resonant tank. A conventional magnetic core transformer can be used in various examples. In other examples, an air core transformer (ACT) can be used. Compared to a regular (i.e., magnetic core) transformer, an ACT can have very low coupling between windings (e.g., 40% instead of the 98% typical for magnetic core transformers). This results in higher leakage inductance than a regular transformer. However, this is advantageous because it allows several desirable functions for the drive circuit, namely, galvanic isolation for safety and EMI suppression (because the transformer provides a noise barrier), voltage step-up (as described in more detail below), and resonant inductance, to be incorporated entirely into a single component. In some instances, this is desirable. These functions can also be provided by a conventional transformer, although to a lesser extent in some instances.

[0159] Referring more particularly to inverter 30, in the example shown in Figures 5 and 7, the inverter is provided by an H-bridge. The H-bridge comprises two high-side switches S 1+ and S 2+ and two low-side switches S 1- and S 2- In the example shown in Figure 6, the inverter is provided by a half bridge, which has two switches 32 and two capacitors 34, and one high-side switch S 1+ and one low-side switch S 1- to provide.

[0160] 5 to 7, the switches 32 of the inverter 30 are provided by transistors. In the example shown in these figures, these are silicon carbide MOSFETs. In other examples, each switch may be provided by a MOSFET, such as an n-type MOSFET, a silicon MOSFET, or another type of electronic switch, such as an insulated gate bipolar transistor (IGBT), such as a silicon IGBT, a junction field effect transistor (IFET), a bipolar junction transistor (BJT), or a high electron mobility transistor (HEMT), such as a gallium nitride (GaN) HEMT.

[0161] In the examples shown in Figures 5 and 7, capacitor 24 is connected in parallel with inverter 30 and voltage source 22. This provides a DC link capacitance for drive circuit 20. In the example shown in Figure 6, this capacitance is provided by capacitor 34 of the half-bridge inverter.

[0162] Drive circuit function As shown in FIG. 8, a system is used to provide a train of electrical pulses to the resonant tank and inhibit power transfer to the resonant tank after the pulse train. There are also steps to modify the pulse train before a further pulse train is provided, recover energy from the resonant tank after a discharge ignition event, and modulate the power characteristics to store energy. While there are examples where energy recovery is not included in this process, energy recovery is typically included. However, the step of modulating the power characteristics is optional. Details of the process are described in more detail below, along with further details of the power modulation and energy recovery processes.

[0163] During use of the system 1, the power supplied to the DBD device 10 must be at least the dielectric barrier discharge voltage level (V th ) which is necessary to stimulate a dielectric barrier discharge across the discharge gap. The model circuits shown in Figures 5, 6 and 7 for DBD devices are thThe power absorbed by the DBD voltage source shown in these figures is V th and the current applied to the resonant tank (when the diode is conducting). Therefore, the voltage across the gap is V th When the voltage exceeds the V V (model) shown in the figure, the corresponding diode pair in the model circuit of the DBD device conducts. th This model represents the power transfer to the plasma. In this model, each time a dielectric barrier discharge occurs, the voltage across the gap increases by V th is clamped to

[0164] The power that supplies the dielectric barrier discharge voltage is supplied by the driver circuit 20 as a pulse train. The power provided by the pulse train is drawn from a DC link voltage source 22 at a level of approximately 800 V, which is supplied to the inverter 30. In other examples, the voltage provided by the DC link voltage source can be up to 900 V when using silicon carbide MOSFETs, or higher, such as 1.2 kV to 1.3 kV when using 1.7 kV rated silicon carbide transistors.

[0165] When using the example system shown in Figure 5 to initiate a pulse train, once power is drawn from the DC link voltage source 22, an H-bridge is used to excite the resonant tank 40. In this example, this is accomplished by the H-bridge outputting a 100% duty cycle square wave voltage for the duration of the first two modes of the pulse train (as described above in connection with Figure 3).

[0166] The switches 32 of the H-bridge are arranged to provide an output at a switching frequency adjusted to excite the resonant tank 40 at the tank's resonant frequency, so that only real power is processed by the H-bridge. To minimize switching losses, operation at a frequency slightly above the resonant frequency is feasible to achieve ZVS of the switches.

[0167] As discussed above in connection with FIG. 3, excitation of the resonant tank 40 occurs when the voltage level within the resonant tank 40 is V th When the electric field reaches the dielectric barrier discharge, it induces a dielectric barrier discharge, which transfers power to the plasma between the electrodes in the DBD device 10.

[0168] When the second mode of the pulse train is terminated, switch 32 is turned off. When using a transistor as in the examples shown in Figures 5-7, this can be done by either turning off the transistor apart from the transistor body diode (or external anti-parallel diode) being left active, or by reducing the bridge voltage (v) across inverter 30 to passively or actively recover, respectively, the remaining energy stored in resonant tank 40. FB ) is phase shifted by 180 degrees (°).

[0169] The recovered energy is transferred to DC link capacitor 24 (which corresponds to capacitor 34 of inverter 30 when example drive circuit 20′ shown in FIG. 6 is used instead of example drive circuit 20 shown in FIG. 5 or example drive circuit 20″ shown in FIG. 7). This is achieved by reversing the power flow through passive or active recovery as described in the previous paragraph, allowing this energy to contribute to the energy used for the next pulse train.

[0170] Passive power recovery is achieved by simply switching off the transistors in the inverter 30 at the end of the second mode (i.e., when the dielectric barrier discharge is terminated), as described above. Due to the arrangement of the circuit in an H-bridge or half-bridge, this eliminates all circuit paths through the transistors, leaving a path through the transistor body diodes (which provide connections across the transistors, as shown in Figures 5, 6, and 7). The connection of the resonant tank across the inverter to the diodes, as shown in Figures 5, 6, and 7, allows energy to flow through the diodes to the DC link capacitors 24, 34 when the transistors are switched off.

[0171] Instead, active power recovery is achieved by utilizing a transistor to provide a 180° phase shift at the output of the inverter 30 from the phase of the output in the second mode. Instead of allowing energy to flow to the DC link capacitors 24, 34, as occurs during passive power recovery, this drives energy into the DC link capacitors.

[0172] The quality factor (Q) of the resonant tank is the ratio of the dielectric discharge gap (v) to the bridge voltage at the resonant frequency. dbd ) voltage gain (i.e., Q=v dbd / v FB ) (for transformers without unit turns ratio, the quality factor is Q = v dbd / (v FB ( / n), where n is the turns ratio of the transformer. The total gain when using a transformer is also determined by the transformer step-up and resonant gain.) The effective voltage gain of the resonant tank is determined by the parasitic capacitance of the magnetic components. It is determined by the power losses imposed by the natural resistance and the wires connecting the electrodes of the DBD device, which provide damping to the circuit. Unlike known systems using resonant converters, in examples according to aspects disclosed herein, the effective voltage gain is not determined by the actual power being delivered to the plasma, since no discharge occurs during charging of the resonant tank. Thus, practical values ​​of Q greater than 40 allow for dielectric barrier discharge voltages in excess of 30 kV from a DC link input voltage of 800 V without the explicit need for a step-up transformer.

[0173] It can be appreciated, therefore, that when power is absorbed by the initiation of a discharge ignition event in a DBD device, the lower voltage gain can cause self-quenching effects due to the resulting damping and Q-factor shift. However, because only a small number of discharge ignition events (such as 1 to about 5 discharge ignition events) are required from each pulse train and there is sufficient momentum in the resonant tank (stored energy much greater than the energy absorbed by the discharge), this does not pose any practical challenges for examples according to embodiments disclosed herein. On the other hand, known resonant converters are configured for relatively low voltage gain resulting from continuous power absorption by the plasma and therefore require and are designed with a high step-up transformer turns ratio.

[0174] The voltage across the dielectric discharge gap is determined by the capacitance of the dielectric discharge gap, which is composed of the capacitance of the dielectric and the capacitance of the gap itself. In the examples of Figures 5, 6 and 7, the capacitance of the dielectric (C diel ) is typically the capacitance of the gap (C gap ) is much larger than C. diel is typically C gap This is also at least 10 times larger than the dielectric (V diel ) compared to the voltage across the gap (V gap ) gives the voltage ratio.

[0175] The energy recovery process can be applied in a corresponding manner using the example drive circuit 20' shown in Figure 6. When using the example drive circuit 20" shown in Figure 7, the same process can be used as can be applied to the example drive circuit 20 shown in Figure 5.

[0176] The power delivered by the DC link power supply is the power delivered to the drive circuit, averaged over the pulse train repetition interval. The energy exchanged between the DC link capacitor and the resonant tank during resonant tank charging, power transfer during the dielectric barrier discharge, and resonant tank discharge typically cause voltage ripple across the DC link capacitor. The interval during which power is transferred to the plasma by the dielectric barrier discharge also contributes to the DC link voltage ripple.

[0177] 7, transformer 50 provides a step-up ratio of between about 1:1 and 1:10. This step-up ratio, which is lower than that of conventional pulse power circuits (such as the exemplary step-up ratios described above), allows for limiting the current passing through transformer primary 52. ​​When a 1:1 ratio is used, it provides only galvanic isolation, instead of providing galvanic isolation and voltage boost when a higher step-up ratio, such as a 1:10 step-up ratio, is used.

[0178] The inductor 42 used in the drive circuit 20″ of FIG. 7 can be located on either the primary or secondary side of the transformer 50. However, as mentioned above, by placing the inductor on the secondary side (and therefore the high voltage side), the kVA rating of the transformer can be reduced. The reactive power of the DBD device 10 can then be compensated directly. Under such reactive load matching conditions, only real power is processed by the transformer.

[0179] The galvanic isolation imposed by the transformer 50 reduces ground currents, which are currents that flow in the parasitic capacitance between the electrodes of the DBD device 10 and any surrounding metallic housing. helps meet electromagnetic compatibility (EMC) limits.

[0180] The duration of each wavelet pulse train determines the number of dielectric barrier discharge ignition events. As can be seen in Figure 9, for a given V dc For the number of excitation periods n p(i.e., frequency cycle) is the effective duration of the wavelet pulse train and the V th This defines the number of dielectric barrier discharge ignition events after reaching . This therefore determines the amount of energy transferred to the plasma per pulse train.

[0181] The active power is adjusted by moving the bridge leg switching frequency away from the resonant frequency. This can be achieved by moving the switching frequency above the resonant frequency or by moving the switching frequency below the resonant frequency. This is achieved by adjusting the v FB and the bridge current i -FB and thus reducing the active power transferred to the DBD reactor.

[0182] By taking this approach, the high voltage gain is reduced and reactive power handling is increased. To maintain high voltage gain and minimize reactive power handling, in accordance with aspects of the present disclosure, inverter 30 can instead be arranged to provide excitation near the resonant frequency in use. This is known as v FB and i FB This is achieved by keeping the phase shift between near zero. The average power is adjusted by varying the repetition frequency of the wavelet pulse train (i.e., how often the wavelet pulse train is used to excite the resonant tank and cause a dielectric barrier discharge). This allows very high part-load efficiency to be achieved because the resonant tank is always operated at its resonance and therefore handles little or no reactive power.

[0183] As mentioned above, the length of the pulse train is variable. A pulse train of one duration can be seen in Figure 9. The pulse train shown in Figure 9 is a short pulse train, such as can be used with examples according to embodiments disclosed herein, to generate between two and four discharge ignition events.

[0184] In Figure 9, a pulse train is generated by an exemplary drive circuit such as that shown in Figure 5 or Figure 7. Of the two plots shown in this figure, one plot shows the states of switches 32 in an H-bridge inverter 30. These are either in the off state ("0" state) or the on state ("1" state). By operating these switches in pairs, the waveform pattern shown in the bottom plot of the figure can be generated in a DBD device.

[0185] The switch pair is S 2- S paired with switch 1+ Switch and S 2+ S paired with switch 1- During the first two modes of the pulse train, the switches in each pair (i.e., the two switches in each pair) are operated in phase, placing each switch in the same state as the other switch in the pair. During the first two modes of the pulse train, the pairs are operated out of phase, meaning that when the switches in one pair are in one state, the switches in the other pair are in the other state.

[0186] As in a conventional inverter, switch S 1-+ and S 1- There is a "dead time" or "interlock time" between the two switches, during which they are switched from one state to the other. This dead time is the period during which both switches are turned off. This period is typically a few hundred nanoseconds. This period serves as a safety interval to prevent the DC link power supply from being accidentally shorted out, which would cause a catastrophic failure in the system.

[0187] Switch vs. S 1+ and S 2- is turned on, and the switch pair S 1- and S 2+ Turn off By reversing the state, i.e., switch pair S 1+ and S 2-- is turned off, and the switch pair S 1-and S 2+ By turning on the switch pairs, this causes a negative voltage rise. By alternating this arrangement, a sinusoidal waveform like the one shown in the bottom plot of Figure 9 is generated, with the frequency of the waveform determined by the amount of time each switch pair is in the on and off states.

[0188] In Figure 9, each switch pair is operated for seven on-off cycles, S 1+ and S 2- The pair is the first pair to turn on, which lasts for about 1.75 cycles, a duration of about 40 μs, and a voltage of at least V th When the on-off cycling of the switch pair is stopped, a third mode of the pulse train occurs until the voltage returns to 0 V. Additionally, in the pulse trains shown in Figure 9, the first and third modes of each pulse train have approximately the same duration.

[0189] Figure 10 shows a mechanism for varying the amount of power delivered to the plasma. As mentioned above, a further mechanism for altering the amount of power delivered to the plasma is to vary the frequency of the pulse train (i.e., the number of pulse trains per unit time). This is achieved by varying the repetition frequency (f r ) Three different power transfer levels are shown in the three plots of Figure 10.

[0190] Each plot in Figure 10 shows a period of approximately 200 μs. At low power transfer rates, such as the bottom plot in Figure 10, there may be one pulse train, resulting in an f of approximately 5 kHz (equal to the reciprocal of 200 μs). r In the upper plot of Figure 10, f r is about 10 kHz (equal to the inverse of 100 μs) and the pulse train duration is about 40 μs. This second plot provides a moderate power transfer rate. A (very) high power transfer rate is exemplified by the top plot (third plot) of FIG. 10. In this third plot, f rThe frequency is approximately 18 kHz (equivalent to the inverse of 55 μs), and the pulse train duration is approximately 40 μs. In each of these three plots, the pulse trains are distinguishable from one another because of the discernible increase and subsequent decrease in voltage amplitude for each pulse train. In each pulse train, the voltage is at least V th When the voltage rises to V, a dielectric barrier discharge occurs. th When the temperature drops below this, the dielectric barrier discharge ceases.

[0191] Control and Feedback Parameters within system 1 may change over time and / or during use. For example, the effective capacitance of the reactor is affected by process parameters (such as temperature, humidity, gas flow rates and other properties). Therefore, a feedback mechanism for monitoring and responding is used with DBD reactor 10 and drive circuitry 20, 20', 20". This may be provided in the form of a controller, shown generally at 200 in FIG. 11, which is connected to the drive circuitry in use.

[0192] According to various examples, the controller can adjust the average power delivered to the DBD reactor 10. This can be accomplished by varying the number of pulses in the pulse train and / or the pulse repetition frequency (i.e., the repetition rate of the pulses within the pulse train) and / or the pulse train repetition frequency. In some examples, the controller can track the resonant frequency of the resonant tank. As discussed above, the resonant frequency can change depending on the condition of the fluid passing through the reactor and also changes when power is being transferred to the gas. The natural frequency can also be a damped or undampened natural frequency, which affects any frequency to which the tracked frequency may be compared. The resonant tank can be adjusted, such as by updating the frequency after each individual pulse in the pulse train. There are examples where the frequency of the input to the resonant tank can be adjusted within the duration of the pulse train. It is also possible for the frequency of the input to the resonant tank to be kept constant within the pulse train and only adjusted between successive pulse trains.

[0193] An exemplary monitoring and response process using controller 200 is as follows: Controller 200 includes a phase detection unit 210. The phase detection unit is connected to the output of inverter 30. This allows the phase detection unit to detect v FB- and i FB , and thereby feedback can be obtained by monitoring these parameters. From these measurements, the phase angle (φ) can be calculated by a phase detection unit. This unit then calculates the n p The phase angle is averaged over the excitation periods to obtain the pulse train average phase (<φ> w ) can be provided as output.

[0194] In some instances, the measurement of φ is taken as the voltage v that switches from negative to positive. FB Current i for the time FB This is achieved by detecting the point (e.g., time) of zero crossing (ZC) of the current. While it is possible to use ZC for voltage relative to current, because the voltage is generated by the switching action in inverter 30 determined by controller 200, such voltage ZC measurement may not be required since it can be reconstructed. There are other methods closely related to this and the use of current ZC that can be used directly as a means of feedback. Thus, phase control techniques such as those described herein can rely on ZC detection, but are not required.

[0195] As shown in Figure 12, φ is represented by a square waveform v FB The start time at time X of the zero crossing point of the current i FB The time window between time C and time D shows the pulse train averaging window (〈 〉 w ) is the period over which the phase angle is averaged. The period from time C to time D begins at the beginning of the pulse train (i.e., when the resonant tank begins to be excited). This period is the time it takes for the resonant tank to reach the firing voltage amplitude (V th) is reached (i.e., dielectric barrier discharge begins), allowing power transfer to occur. This period ends when excitation is stopped.

[0196] The excitation is stopped to stop the occurrence of discharge ignition events. This limits the number of discharge ignition events to a desired maximum number of discharge ignition events. In some examples, the point at which to stop excitation is determined based on the number of pulses in the pulse train compared to a preset number of pulses for the excitation period in the pulse train. However, in some other examples, instead of operating based on the number of pulses in some pulse arrangement, an arrangement is used that detects when a discharge ignition event has occurred. Detection of the first (and potentially subsequent) occurrence allows the number of discharge ignition events to occur over the next period to be known, calculated, or predicted to be one. This allows excitation to be stopped when the maximum number of discharge ignition events is reached, whether it be one, two, three, four, five, or another number of discharge ignition events.

[0197] To detect when a discharge ignition event occurs, phase shift detection occurs. In various examples, this is detection of the instantaneous phase, rather than the average phase as typically used when modulating the frequency of pulses in a pulse train to track the resonant frequency, as described above and below in connection with FIG. 11. This detected phase shift is the voltage-current phase shift measured at the H-bridge terminals. During charging of the resonant tank, the phase difference between the voltage and current at the terminals is close to zero. However, once a discharge ignition event occurs (i.e., plasma is ignited), there is a shift in the resonant frequency due to the increase in capacitance imposed by the "ignited" DBD device. This resonant frequency shift can be readily detected by monitoring the corresponding phase shift.

[0198] This monitoring can be done using the controller 200, in some examples, such as using the phase detection unit 210. As mentioned above, in such examples, this is connected to the inverter terminals.

[0199] In examples where the maximum number of discharge ignition events is one discharge ignition event, excitation is stopped when the first discharge ignition event is detected. In examples where the maximum number of discharge ignition events is higher (such as up to about five), excitation can then be stopped by counting the number of subsequent pulses and equating each pulse with, for example, one discharge ignition event. Alternatively, identifying further discharge ignition events can be achieved by continuing to monitor the phase and identifying when each discharge ignition event occurs by its effect on the voltage-current phase at the inverter terminals.

[0200] In various examples, the phase detection unit 210 is provided by analog circuitry, while in other examples, the phase detection unit is implemented digitally using a field programmable gate array (FPGA).

[0201] Using an FPGA, or another such digital implementation of phase detection unit 210, greater flexibility can be achieved than when analog circuitry is used, including the ability to change the controller by upgrading software and avoiding the need to design new physical circuitry and replace existing circuitry when an upgrade is desired.

[0202] The use of FPGA or analog circuitry also allows the phase angle to be calculated and fed through the controller 200 after each pulse cycle in the pulse train. Using FIG. 12 as an example, such a cycle would be FB A single cycle of a square wave and / or FB11 and provided in more detail below) to determine a new frequency setpoint, providing a higher performance system because it allows adjustments to be made to the pulse train during its duration. In contrast, using a pulse train averaging window only allows the PI controller to provide input for adjustments to the characteristics of the next pulse train, not the currently ongoing pulse train.

[0203] <φ> w Once calculated, it is compared by the controller 200 to a phase reference value (φ*). φ* is provided by the process control unit of the controller 200, shown at 220 in FIG. 11. It is derived from the properties of the gas passing through the DBD device 10. The properties shown in FIG. 11 are the amount of NOx, the amount of SOx, the amount of CH4, the percent humidity (% HO), the flow rate (liters per minute, l / min), and the temperature (°C), which in this example are provided as inputs to the process control unit. This provides further feedback by monitoring the properties and content of the gas passing through the DBD device. Although not shown in FIG. 11, the amount of nitrous oxide (NO) may also be included as an input to the process control unit.

[0204] 11, the quantities (such as the amounts of NOx, SOx, CH4, and / or NO) are provided in parts per million (ppm) in this example. In other examples, different units of measurement may be used.

[0205] The amounts of other components in the gas may also be monitored and provided as inputs, as indicated by the "..." notation as inputs to the process control unit in FIG.

[0206] The desired quantities of some or each of the component chemicals expected to be present in the gas are provided to the process control unit 200. The quantity input can then be compared to the desired quantities of each of the associated chemicals. Any differences between the quantity input and the desired quantities, and / or one or more of the quantity input and / or other gas properties, are then used to determine the output of the process control unit.

[0207] In the example shown in Figure 11, the output includes φ*, which represents the optimum phase angle, which is typically close to zero (such as about 0°), or about +5° to about +15° if zero voltage switching (ZVS) is applied.

[0208] <φ> w The output of the comparison between φ* and φ is the phase angle error (e φ ) This error is input to a compensator, shown in FIG. 11 as a proportional-integral (PI) controller 230. The PI controller calculates e φ Based on the frequency fluctuation (Δf s ) is calculated.

[0209] e φ Contributing factors that can be used in determining ω are the phase angle and the gain achievable based on how the inverter output frequency relative to the resonant frequency shifts the phase angle.

[0210] In drive systems according to various examples described herein, the gain factor (simple multiplication) achieved is typically between about 30 and about 50. This corresponds to a gain from about 800 V input at the DC link power supply 22 to about 30 kV above the dielectric barrier discharge threshold in the dielectric discharge gap. This corresponds to a gain of about 30 to about 34 decibels (dB).

[0211] The controller 200 calculates the nominal resonant frequency feedforward term (f s,ff ) based on the input to that unit, Δfs This is the frequency setpoint (f s *) is provided.

[0212] The process control unit 220 also determines f based on the unit inputs and the processing performed by the process control unit. r Set point (f r *) and n p Set point (n p *) is output. s *, f r * and n p * are provided by controller 200 to modulator unit 240. The modulator unit uses these to generate switching signals for the switches of inverter 30 to adjust the excitation supplied to resonant tank 40. When the inverter is an H-bridge, these are the switching signals for each of the four switches (as shown in the example controller of FIG. 11). When the inverter is a half-bridge, these are the switching signals for each of the two switches.

[0213] Typical applied switching frequencies in exemplary systems are from about 100 kHz to about 10 MHz. r * is typically in the range of about 100 Hz to 50 kHz. This latter parameter is also, in various examples, the speed at which the controller 200 is operated (i.e., the rate at which various parameters are used and updated by the controller). This places lower performance requirements on the controller than if higher operating speeds were used.

[0214] The system 1 can be used with several different sizes of gas flows, such as engines and boilers of various sizes. Therefore, there are examples where the exhaust gas purification system or other system employing the above-described drive circuits 20, 20', 20" and controller 200 is implemented in a modular manner.

[0215] In such an example, there are multiple DBD devices 10 connected in series along the gas flow. A drive circuit 20, 20', 20" is typically provided for each DBD device. A global controller 1000 can be implemented as shown in FIG. 13, which applies the same process and uses the same components as controller 200, as described in connection with FIG. 11. Inputs for phase detection are provided from each drive circuit. Gas characteristics are input to the global process control unit 1020. A modulator unit 240 is provided for each drive circuit to drive the switches for the inverters of each drive circuit. Thus, individual set points of the same type as those provided to modulator units 240 shown in FIG. 11 are provided from the global controller to each drive circuit, providing coordinated control of each drive circuit. The number of modulator units 240 is determined by the number of drive circuits, which therefore varies depending on the size of the gas flow being processed.

[0216] When multiple drive circuits are used, in some instances, a single DC power supply is arranged to power all of the drive circuits. In other instances, each drive circuit has its own DC power supply. In instances with a single DC power supply, a single AC / DC rectifier can supply DC power to each of the individual drives, thereby providing a single DC link power supply. As an exemplary implementation of each drive circuit having its own DC power supply, each drive circuit can be equipped with an individual AC / DC rectifier and a three-phase AC voltage source. In such instances, the DBD devices 10 are typically electrically connected in parallel, but may also be connected in series within the gas flow (i.e., sequentially along the gas flow path).

[0217] Of course, by having multiple driver circuits, various examples have multiple DBD devices, which are arranged in parallel, thereby increasing the overall capacitance of system 1 as the sum of the capacitances of each DBD device, thereby achieving capacitances of up to 45.0 nF, and in some cases 1.0 nF, for example.

[0218] optimization When system 1 is used in an application using a step-up transformer such as the example shown in FIG. 7, ringing may occur between magnetizing inductance 58 of transformer 50 and DBD device 10.

[0219] Ringing occurs in the timer intervals between pulse trains. This can be seen in Figure 14a as a wave between the two pulses in the bottom plot. This is due to standing waves that can establish in the circuit.

[0220] To minimize ringing, instead of turning all switches off between the end of the second mode of a pulse train and the start of the next pulse train, a "freewheeling" interval is introduced in some instances.

[0221] Such a freewheeling interval is shown in the top plot of Figure 14b, where the high-side switch S 1+ and S 2+ It can be seen that after the end of the third mode (i.e., the mode in which the resonant tank is discharged) of the first pulse train shown in the bottom plot of Figure 14b, is left in its on state until the start of the next pulse. This shorts out the transformer winding (i.e., applies a voltage of approximately 0 V). The response to this in System 1 is that while there is ringing between the two pulses shown in the bottom plot of Figure 14a, the ringing is minimized / damped, as can be seen by the absence of ringing between the two pulses shown in the bottom plot of Figure 14b.

[0222] The freewheeling interval is the time after the resonant tank is de-energized (i.e., the pulse train occurs). As mentioned above, this is initiated when the high-side switch is turned on and the low-side switch S 1- and S 2-Alternatively, the same result can be achieved by having the low-side switch in the on state and the high-side switch in the off phase.

[0223] In instances where an air core transformer is used, ringing also occurs when active energy recovery is not applied, as can be seen, for example, from the plot shown in FIG.

[0224] In Figure 15, three plots are shown. All plots have time in milliseconds as their x-axis. The top plot shows the voltage V at the inverter terminals (i.e., the terminals connected to the primary winding of the transformer) versus time. fb The central plot shows the corresponding current I at the inverter terminals versus time. fb The bottom plot shows the voltage across the discharge gap resulting from the voltage and current shown in the other two plots in this figure versus time.

[0225] 15 shows two pulse trains provided by the inverter. The first pulse train begins at approximately 9.00 ms. The pulse train has a square V fb The initiation of the pulse train causes charging in the resonant tank as seen by a ramp up in amplitude in the inverter terminal current and discharge gap voltage.

[0226] A discharge ignition event occurs in the discharge gap when the resonant tank is charged to a threshold voltage, which in the example shown in Figure 15 is approximately 10 kV.

[0227] The excitation is stopped immediately after this, depending on the maximum number of discharge ignition events desired. In the example shown in Figure 15, this number is between one and three discharge ignition events. The time at which the excitation is stopped is most clearly seen from the inverter terminal current plot, which shows a sudden drop in current amplitude from approximately 800 A during the discharge ignition event to a maximum peak of approximately 200 A in the next cycle. This occurs at approximately 9.02 ms, and charging to the threshold voltage takes approximately 9.01 ms.

[0228] As can be seen from the inverter terminal voltage and current plots, the next pulse train begins at approximately 9.11 ms. However, Figure 15 shows that the voltage at the inverter terminals and the discharge gap continues to oscillate. In fact, the amplitude of the voltage at the discharge gap is only reduced to about half the amplitude of the discharge threshold, or about 5 kV. However, this is only reduced by about 1 to 2 kV in the period between the end of excitation of the first pulse train and the beginning of the next pulse train.

[0229] Referring to Figure 16, this shows the same three plots as Figure 15 of inverter terminal voltage, inverter terminal current, and discharge gap voltage versus time. In the example shown in Figure 16, the inverter terminal plot shows that the pulse train begins at 8.00 ms. As can be seen from the inverter terminal current and discharge gap plots, the resonant tank charges from this time until approximately 8.01 ms. At approximately this point, the discharge threshold is reached and a discharge ignition event occurs.

[0230] After the maximum number of spark events has occurred, again one to three spark events in the example of FIG. 16, excitation is stopped. This occurs at approximately 8.02 ms. At this point, a 180° phase shift is applied to the inverter terminal voltage for a period of approximately 0.01 ms until approximately 8.03 ms. This forces the energy in the charged resonant tank to be expelled from the resonant tank. As discussed above, in various examples, this energy is then stored. The expulsion of energy from the resonant tank can also be seen from the inverter terminal current plot, which instead of showing the current having a sine wave (of varying amplitude) centered around 0 A, shows the current wave shifted negatively until the end of the voltage phase shift period.

[0231] This active energy recovery when using an air-core transformer can be seen in Figure 16 to reduce the ringing between the end of the phase shift period at approximately 8.03 ms and the beginning of the next pulse train at approximately 8.11 ms. This reduction is to an amplitude of approximately 1 kV at the discharge gap and approximately 50 V at the inverter terminals.

Claims

1. 1. A drive circuit for a dielectric barrier discharge device, the circuit comprising: a power supply connectable across the dielectric discharge gap in use, the dielectric discharge gap providing a capacitance; an inductance between said power supply and said dielectric discharge gap when connected, thereby establishing a resonant tank in use; an electrical storage device connected across said power supply and arranged, in use, to receive and store an electrical power discharge from said tank after each pulse train; Equipped with power is supplied to the tank in, and only during, pulse trains, in use, the pulse frequency of each pulse train being adjustable in use to a resonant frequency of the tank, the power supplied by each pulse train charges the tank and maintains it at a threshold at which spark ignition occurs, and spark ignition events per pulse train are limited to a maximum number of times based on the drive circuit being arranged in use to inhibit each pulse train from transferring power to the resonant tank after the maximum number of times has occurred; the drive circuit is arranged, in use, to shift the phase of the pulse train by 180 degrees (°) after the maximum number of discharge ignition events has occurred. Drive circuit.

2. The drive circuit of claim 1 , wherein the maximum number of discharge ignition events is between 1 and 5 events.

3. 3. The drive circuit of claim 1, further comprising a phase meter in communication with the tank and arranged, in use, to identify a phase shift in the power supplied to the tank during each pulse train, the phase shift corresponding to the occurrence of a discharge ignition event, the drive circuit being further arranged, in use, to determine when the maximum number of discharge ignition events has occurred based on the number of pulses in the respective pulse train since each respective discharge ignition event.

4. 4. A drive circuit as claimed in any one of claims 1 to 3, further comprising an inverter between the power supply and the tank, the inverter being arranged, in use, to regulate the supply of power from the power supply to the tank.

5. 5. The drive circuit of claim 4, wherein the inverter is an H-bridge or a half-bridge.

6. 6. The drive circuit of claim 5, wherein each switch of the inverter is a silicon carbide switch.

7. 7. A drive circuit according to claim 4, wherein the pulse frequency of each pulse train is the zero voltage switching frequency.

8. 8. A drive circuit as claimed in any one of claims 1 to 7, further comprising a transformer, the secondary winding of which forms part of the resonant tank, the transformer being a step-up transformer.

9. 9. A drive circuit as claimed in claim 8, wherein the circuit is arranged, in use, to short-circuit the primary transformer winding after each pulse train.

10. 10. A drive circuit as claimed in claim 9 when dependent on claim 5, wherein the primary transformer winding is shorted in use by switching on the low side or the high side of the inverter.

11. A drive circuit as claimed in any one of claims 8 to 10, wherein at least part of the inductance is provided by the transformer.

12. The drive circuit of claim 11 , wherein the inductance provided by the transformer is a leakage inductance of the transformer.

13. 13. The drive circuit according to claim 11 or 12, wherein the transformer is an air-core transformer.

14. 14. The drive circuit of claim 13, wherein the air core transformer has up to 60% magnetic coupling between windings.

15. 15. The drive circuit of claim 8, wherein the transformer has a step-up ratio between the primary and secondary transformer windings of about 1:1 to about 1:

10.

16. 16. A drive circuit according to any preceding claim, wherein at least part of the inductance is provided by an inductor.

17. 1. A system for providing a dielectric barrier discharge, the system comprising: a dielectric barrier discharge device having at least two electrodes with a fluid gap therebetween that defines a dielectric discharge gap, the dielectric barrier discharge device having a dielectric layer positioned between the at least two electrodes; 17. A drive circuit according to any one of claims 1 to 16, wherein the power supply of the drive circuit is connected across the dielectric discharge gap; A system comprising:

18. 18. The system of claim 17, further comprising a controller connected to the drive circuit, the controller arranged, in use, to adjust the power supplied to the tank of the drive circuit based on inputs provided to the controller.

19. The controller, in use, adjusts the pulse frequency and / or pulse train repetition frequency and / or number of pulse trains and / or number of pulses within a pulse train.

20. The system of claim 18, arranged to:

20. 20. The system of claim 18 or 19, wherein the inputs include the voltage and current at the output of the driver circuit.

21. 21. The system of claim 20, wherein the drive circuit comprises an inverter between a power supply of the drive circuit and a resonant tank, the voltage and current being supplied from an output of the inverter.

22. 22. A system according to claim 20 or 21, wherein the controller is arranged, in use, to determine a phase difference between the voltage and current.

23. 23. A system according to any one of claims 18 to 22, wherein the controller is further connected to the dielectric barrier discharge device and the input comprises one or more properties of a fluid passing through the device in use.

24. 24. The system of any one of claims 17 to 23, wherein the system comprises a plurality of dielectric barrier discharge devices and a plurality of drive circuits, each drive circuit connected across the dielectric discharge gap of one or more dielectric barrier discharge devices.

25. 25. A system as claimed in claim 24, wherein there is only a single power supply arranged in use to provide said power supply to all said drive circuits.

26. 1. A method for controlling a discharge in a dielectric discharge device, the method comprising: supplying power to a resonant tank with a series of electrical pulse trains, the pulse frequency of each pulse train being tuned to the resonant frequency of the tank, the resonant tank being connected across a gap between electrodes in a dielectric discharge device, the capacitance of the tank being provided by the dielectric discharge device, and the power provided by each pulse train charging the tank and maintaining it at a threshold at which discharge ignition occurs; providing a maximum number of discharge ignition events per pulse train by prohibiting each pulse train from transferring power to the resonant tank after the maximum number of discharge ignition events have occurred; inhibiting power transfer to the tank between pulse trains; Including, For each pulse train, the resonant tank is discharged after the maximum number of discharge ignition events have occurred, and the method includes storing energy discharged from the resonant tank by the discharge; The tank is discharged by shifting the phase of the power supplied by each of the pulse trains by 180°. method.

27. 27. The method of claim 26, wherein the maximum number of discharge ignition events is between 1 and 5 events.

28. identifying a phase shift in the power supplied to the tank during each pulse train, the phase shift corresponding to the occurrence of a spark ignition event; determining when the maximum number of discharge ignition events has occurred based on the number of pulses since each respective discharge ignition event; 28. The method of claim 26 or 27, further comprising:

29. 29. The method of any one of claims 26 to 28, wherein each electrical pulse train is a voltage pulse train.

30. 30. The method of any one of claims 26 to 29, further comprising modulating the pulse frequency, and / or the frequency of pulse trains, and / or the number of pulse trains in the series of electrical pulse trains, and / or the number of pulses in each pulse train.

31. 31. The method of claim 30, wherein the modulation is based on a phase difference in a characteristic of the power supplied to the resonant tank and / or one or more characteristics of a fluid passing through the device.

32. 32. The method of any one of claims 26 to 31, wherein power is supplied to the resonant tank via a transformer, the method further comprising shorting a primary winding of the transformer between pulse trains.

33. 33. A method according to any one of claims 26 to 32, wherein the pulse frequency of each pulse train supplied to the resonant tank is set by switching a circuit between a power supply and the resonant tank.

Images