Mitko, Sergey (113 Weegschaalstraat, CE Enschede, NL-7521, NL)
| CLAIMS 1. Apparatus for generating an electron beam which comprises an acceleration gap containing a gas at a pressure higher than 100 Pa and defined between a metallic cathode having a flat surface spaced by a uniform predetermined distance from a metallic anode, and means for applying a steep edged high voltage pulse by way of a transmission line to the acceleration gap, wherein the length of the acceleration gap is determined in dependence upon the peak voltage of the applied pulse, the rise time of the leading edge of the applied pulse and the capacitance of the acceleration gap in such manner as to generate a run-away current between the cathode and the anode . 2. Apparatus as claimed in claim 1, wherein the surface of the cathode is covered with a dielectric. 3. Apparatus as claimed in claim 2, wherein the dielectric comprises a polyimide film. 4. Apparatus as claimed in any preceding claim, wherein an ionisation source is placed close to the cathode to provide a controllable source of seed electrons. 5. Apparatus as claimed in any of claims 1 to 3, wherein the means for applying a steep edged high voltage pulse is operative to apply pulses repetitively. 6. Apparatus as claimed in claim 5, wherein the pulse repetition frequency is at or above 100 Hz 7. Apparatus as claimed in any preceding claim, wherein the impedance of the acceleration gap is matched to the impedance of the source of high voltage pulses. 8. Apparatus as claimed in claim 7, wherein the acceleration gap is designed with the minimum possible stray inductance and stray capacitance. 9. Apparatus as claimed in any preceding claim, wherein in order to extend the path traversed by the electron once a run-away current has been created, a second anode, connected to a high DV bias voltage, is provided on the opposite side of the first anode from the cathode. |
Field of the invention
The present invention relates to apparatus for generating an electron beam in the presence of gas at an elevated pressure.
Background of the invention
It has been proposed to use beams of electrons for sterilisation and for curing of printing ink. In such applications, it would be desirable to be able to expose a surface to be treated to an electron beam without the need for the process to be carried out under vacuum.
In order to accelerate electrons to the extent that they can perform useful work, they have to be provided with sufficient energy and this is not simple to achieve when the electrons are accelerated in the presence of a gas, such as air under atmospheric pressure. Within an evacuated envelope, it is easy to produce electrons and to supply them with small amounts of energy, say in the order of several electron-volts. This is what normally happens in a luminescent gas discharge lamps, neon lights, etc. After an initial breakdown takes place, electrons move slowly inside an evacuated tube. Any attempt to accelerate them further by applying more energy results in a higher current and warming up of the discharge tube, but not in highly energetic electrons.
The difficulty in accelerating electrons in a gas at an elevated pressure is that the electrons lose their energy through collisions with gas molecules faster than they gain it from an externally applied electric field. Furthermore, the conductivity of the ionised gaseous medium does not allow an increase of the applied accelerating voltage because the glow discharge will sooner or later undergo a transition to an arcing mode.
The term elevated pressure in the present context refers to a pressure higher than 100 Pa and reaching as high as 10 5 Pa, i.e. from about 1 Torr up to atmospheric pressure. For the reasons given above, most existing electron beam generating devices operate under high vacuum, i.e. at a pressure lower than 10 ~4 Pa or 10 ~6 Torr. Another reason for high vacuum operation is that the normally used thermal cathodes will immediately burn if the gas contains oxygen and the pressure exceeds a level of, say, 1 Pa or 10 ~2 Torr.
Object of the invention
The aim of the present invention is to provide an apparatus capable of efficient acceleration of electrons in non-vacuum conditions, preferably up to the atmospheric pressure .
Summary of the invention
According to the present invention, there is provided an apparatus for generating an electron beam which comprises an acceleration gap containing a gas at a pressure higher than 100 Pa and defined between a metallic cathode having a flat surface spaced by a uniform predetermined distance from a metallic anode, and means for applying a steep edged high voltage pulse by way of a transmission line to the acceleration gap, wherein the length of the acceleration gap is determined in dependence upon the peak voltage of the applied pulse, the rise time of the leading edge of the applied pulse and the capacitance of the acceleration gap in such manner as to generate a run-away current between the cathode and the anode . As electrons lose energy with each collision with a gas molecule, it is necessary for energy to be supplied to the electrons faster than they lose it. If this condition is satisfied, then the electrons will continue to gain energy as time goes on and they will finally reach the energy which will be determined by the peak externally applied voltage. This condition is easily fulfilled in vacuum accelerators. On the other hand, if the gas pressure rises to several Torr, the electrons will ordinarily lose the energy much faster than they will gain it in the applied field.
However, when the applied external voltage rises sufficiently fast, a different process takes place. During the time between collisions, each electron gains more energy than it loses in a single collision event. Therefore, when the next collision takes place, the electron will have had time to gain more energy than it loses in the collision, and the acceleration process will continue. This makes electrons less sensitive to the presence of gas molecules. By gaining the energy at a sufficiently high rate, the electrons will be able to move in a gas, as they still have the ability to accelerate. This is what is referred to herein as a run-away current .
In the present invention, in order to achieve a runaway current, it is vital for the electrons to be accelerated by supplying them with a vast amount of energy in a very short time and it is imperative for the applied voltage pulses to have a very steep edge.
The process of acceleration finds place in the area of cathode fall of the discharge, where electric field gradients are especially high. This phenomena was discussed by M. J. Druyvesteyn, in his article "The abnormal cathode fall of the glow discharge" ,Physica, 5, Issue 9, p.875-881 (1938) . When the above mentioned electric pulse with high rise time is applied to the electrode, an abnormal cathode fall region is formed and it is there that the electrons gain their energy.
An important advantage of the invention is its ability to generate an electron beam of large cross-section. Unlike smaller cathodes, the invention allows the acceleration gap to have an arbitrary shape and an area of several tens of centimetres in diameter.
In order to improve stability of the anomalous discharge and thus to secure stable and a reproducible acceleration regime, it is possible to cover the surface of the cathode with a dielectric. This will improve the stability of electron pulses and increase the homogeneity of the beam. Certain sorts of polyimide films have been found to provide excellent performance.
In order to operate the electron beam generator of the invention at elevated pressures, or in ambient air, it is beneficial not to rely on the spontaneous process of formation of initial seed electrons in the initial stage of the pulse, but to provide a controllable source of seed electrons. This can be done, for example, by placing in the close vicinity of the cathode an ionisation source, for example a UV light source, or a radioactive isotope, or an auxiliary weak discharge emitting electrons, ions and UV radiation .
It has however been found that the operation of the electron beam generator can be substantially simplified if it is operated in a pulse repetition mode. Fairly good levels of stability and reproducibility have been observed when the pulse repetition frequency is raised above 100 Hz.
In order to provide an efficient energy transfer the impedance of the electron acceleration gap should be matched to the impedance of the source of high voltage pulses. High impedance of the acceleration gap will impose limitations on the steepness of the voltage pulse, and should therefore be minimised. To avoid this, the acceleration gap is preferably designed with the minimum possible stray inductance and stray capacitance.
Another way to minimize the overall system impedance is to switch from a standard high voltage cable, with a 50 Ohm characteristic impedance, to strip line structures. This will effectively reduce the impedance of the system and simplify the generation of steep pulses of the accelerating voltage .
In order to extend the path traversed by the electron once a run-away current has been created, it is possible to provide a second anode beyond the first anode and to apply to it a high DC voltage. Using such a second anode, it has been possible to extend the path of the electron beam up to 10 mm but this is not believed to be the limit that can be reached.
Brief description of the drawings
The invention will now be described further, by way of example, with reference to the accompanying drawings, in which :
Figure 1 is a schematic diagram of an electron beam generating apparatus of the invention,
Figure 2 shows the variation of run-away current with anode separation when using a flat circular cathode having a diameter of 2.5 cm,
Figure 3 shows the variation of run-away current with anode separation when using a bullet shaped cathode, Figure 4 is a graph showing how electron beam energy per pulse varies with the charging voltage, Figure 5 is a graph showing the variation of electron bunching energy with switching time,
Figure 6 shows the dependence of the beam energy on the driving impedance, Figure 7 is a graph showing how the total electron bunch energy varies with electrode diameter, and
Figure 8 is a graph showing how the energy efficiency varies with separation between the cathode and the first anode.
Detailed description of the preferred embodiment
The apparatus in Figure 1 comprises a coaxial transmission line 10 of which one end is terminated in a load 16 having an impedance that matches the characteristic impedance of the transmission line. The opposite end of the transmission line 10 is connected to an electron beam acceleration gap defined between a metal cathode 18 having a dielectric coating such as Kapton and a first anode 20. A second anode 22 connected to a high voltage DC supply is provided beyond the first anode 20 to prolong the path traversed by the run-away electrons and extend the electrons into a working space. The gap between the cathode 18 and the first anode 20 and the gap between the first and second anodes 20,22 contain a gas, possibly air, at an elevated pressure, that is to say a pressure in excess of 100 Pa.
Near its ends, the transmission line 10 has two regions 12 from which the outer braid or sheath of the transmission line is removed. The sections of the outer sheath on opposite sides of the regions 12 are connected across a spark gap that is filled with pressurised air and has a breakdown voltage of around 9OkV.
The drive circuit of the pulser formed by the transmission line 10 comprises a DC voltage supply connected by way of a storage capacitor 32 to the primary winding 34 of a 1:5 pulse transformer of which the secondary coil 36 is connected to the outer sheath of the transmission line 10. The pulse transformer is constructed as a Tesla coil with an air core and a coupling coefficient of 0.6.
It should be mentioned that any suitable transmission line pulser, such as a strip line Blumlein pulser, may be used in place of the illustrated self-matched cable generator .
In operation, a charge stored in the capacitor 32 is rapidly discharge by triggering a thyratron 30 using a suitable control circuit (not shown) producing triggering pulses at the desired frequency. This creates a pulse in the winding 34 which is stepped up by the pulse transformer before it is applied to the transmission line 10. The sudden breakdown that the high voltage pulse causes within the spark gap 14 launches a steep voltage wave that propagates towards the acceleration gap where it induces gas discharge in the gas. The amplitude of this wave is many times larger than the static break-down voltage of the gap between the anode and the cathode and the resulting gas discharge serves as a source of high energy electrons.
The very rapidly changing voltage of the leading edge of the voltage pulse results in the run-away effect described previously provided that the correct relationship is achieved between the applied voltage, the rise time of the voltage, and the capacitance of the acceleration gap, which itself is dependent upon the area of the cathode, the separation between the cathode and the first anode and dielectric properties of the gas between them. Once a runaway current has been created, it can be extracted and the path of the electrons lengthened to achieve the desired work space by the application of a steady voltage to the second anode 22. The critical nature of the relationship between the various parameters discussed above is highlighted by the graphs of Figures 2 and 3. When a run-away current is created and its electrons collide with gas molecules, X-rays are emitted. The intensity of emitted X-rays is therefore a measure of the magnitude of the run-away current and in Figures 2 and 3 the readings in millivolts from an X-ray detector are plotted against the separation of the anode from the cathode.
It will be seen from Figure 2, where the cathode is flat and has a diameter of 2.5 cms that in all three experiments the peak of X-ray emission when applying a 3OkV pulse occurs at a separation of 1.15 mm. If a separation differing from this optimum value by as little as 0.35 mm is adopted, the X-ray emission drops by a factor of 10. It is therefore crucial for effective generation of a uniform beam of energetic electrons to be generated to ensure that the separation between the anode and the cathode is accurately matched to the other parameters discussed previously and that this separation is maintained uniform over the entire area of the electron beam.
This point is emphasised by Figure 3 in which the electrode used was bullet shaped with a radius of curvature of 3 mm. In this case, a very different result is obtained because only a small part of the cathode can be at the correct separation from the anode to achieve an optimum runaway current .
It is difficult to express the relationship between the various parameters in the form of a mathematical equation but an empirical understanding of the variation of the beam energy with the different parameters can be obtained from the experimental results displayed in Figures 4 to 8. In the legends of these drawings "Uch" represents the charging voltage, "ts" the switching time, "Z" the cable impedance, "a" the acceleration gap length between the cathode and the first anode and "D" the area of the electron beam.
From Figure 4 it is seen that the output energy is strongly dependent upon the charging voltage with a doubling of the charging voltage producing a ten fold increase in the output energy.
Figure 5 demonstrates the extreme importance of the switching time. An increase from 0.1 ns to 0.5 ns results in the bunch energy dropping from 2 mJ to about 0.6 mJ.
Figure 6 shows the dependence of the electron bunch energy on the driving impedance, with a drop in impedance resulting in an increased output energy.
In Figure 7, in which the dependence of bunch energy on the electrode diameter is illustrated, the results for charging voltages of 50 kV and 100 kV are shown. There is an optimum electrode diameter of around 2.5 cms at which the electron beam production efficiency reaches a maximum. This area varies only slightly with the charging voltage.
Figure 8 shows the variation of electron beam energy with the length of the acceleration gap. The energy efficiency of electron beam production using a 10 cm long piece of cable charged to 100 kV was calculated, though there is no necessity for the cable to be so short. It was found the optimum length of the acceleration gap is around 1 mm. Both below and above this length, the efficiency drops off rapidly.