RU2741793C2 - Neutral particle beam based injector based on negative ions - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: neutral particle beam based on negative ions contains an ion source, an accelerator and a neutraliser in order to form a beam of neutral particles of approximately 5 MW with energy of approximately 0.50–1.0 MeV. Ions formed by ion source, before being accelerated before injection into high energy accelerator by means of electrostatic pre-accelerator based on multi-aperture grid, which is used to draw ion beams from plasma and accelerate to some fraction of required beam energy. Beam from the ion source passes through a pair of deflecting magnets which enable the beam to displace along the axis before reaching the high energy accelerator. After acceleration to full energy, the beam enters the neutraliser, in which it is partially converted into a beam of neutral particles. Remaining types of ions are separated by means of a magnet and directed to converters of electrostatic energy. Beam of neutral particles passes through the shut-off valve and enters the plasma chamber.

EFFECT: technical result is increase of injector energy efficiency.

15 cl, 17 dwg

Description

Technology area

The subject matter described herein generally relates to neutral particle beam injectors, and more particularly to a negative ion neutral beam injector.

Prior art

In fact, to this day, neutral particle beams used in fusion research, etching, material handling, sterilization, and other applications are formed from positive ions. The positive ions of the hydrogen isotope are drawn and accelerated from the gas-discharge plasma by means of electrostatic fields. Immediately after the grounded plane of the accelerator, they enter the gas cell in which they undergo both charge exchange reactions in order to obtain reactions based on ionization by electrons and impact ionization for additional containment. Since the charge exchange cross section decreases much more rapidly with increasing energy than the ionization cross section, the fraction of equilibrium neutral particles in a thick gaseous cell begins to drop rapidly at energies above 60 keV for hydrogen particles. For applications of a beam of neutral particles based on hydrogen isotope ions that require energies much higher than this, it is necessary to form and accelerate negative ions and then convert them into neutral particles in a thin gaseous element, which can lead to a fraction of neutral particles of about 60% over a wide range energies up to several MeV. Even higher fractions of neutral particles can be obtained if a plasma or photonic element is used to convert high energy negative ion beams into neutral particles. In the case of a photonic element, in which the photon energy exceeds the electron affinity of hydrogen, the proportion of neutral particles can be almost 100%. It should be noted that the idea of using negative ions in accelerator physics was first formulated by Alvarez more than 50 years ago [1].

Since beams of neutral particles for excitation and heating by current in large thermonuclear devices of the future, as well as some applications in modern devices, require energies significantly exceeding the range available when using positive ions, in recent years beams of neutral particles based on negative ions have been developed. Nevertheless, the beam currents attained to date are much lower than the beam currents generated in a quite conventional way by means of positive ion sources. The physical reason for the lower productivity of negative ion sources with respect to the beam current is the low electron affinity of hydrogen, which is only 0.75 eV. Consequently, it is much more difficult to form negative hydrogen ions than their positive equivalents. It is also quite difficult for newborn negative ions to reach the stretch region without collisions with high energy electrons, which, with a very high probability, lead to the loss of an excess loosely bound electron. Pulling H ^- ions out of plasma to form a beam is similarly more difficult than for H ⁺ ions, since negative ions are accompanied by a much higher electron current, unless containment measures are in place. Since the cross section for collisional stripping of an electron from an H ^- ion in order to form an atom significantly exceeds the cross section for H ⁺ ions in order to obtain an electron from a hydrogen molecule, the fraction of ions converted to neutral particles during acceleration can be significant if the density of the gas pipeline along the path of the accelerator is not minimized by operating the ion source at low pressure. Ions prematurely neutralized during acceleration form a low energy residue and, in general, have greater divergence than ions that experience full acceleration potential.

Neutralization of a beam of accelerated negative ions can be performed in a gas target with an efficiency of approximately 60%. The use of plasma and photonic targets makes it possible to further increase the efficiency of neutralizing negative ions. The overall energy efficiency of the injector can be increased by recuperating the energy of the ion species remaining in the beam after passing through the neutralizer.

A schematic diagram of a high power neutral particle beam injector for the ITER tokamak, which is also typical for other considered systems of magnetic plasma confinement in a reactor, is shown in Fig. 3 [2]. The basic components of the injector are a high-current source of negative ions, an ion accelerator, a neutralizer, a magnetic separator of the charged component of the recharged beam with ion receivers / recuperators.

To maintain the required vacuum conditions in the injector, a high vacuum pumping system is typically used with large isolation valves that cut off the beam flow from the plasma device and / or provide access to the main elements of the injector. Beam parameters are measured using retractable calorimetric targets as well as non-destructive optical methods. Formation of powerful beams of neutral particles requires the use of an appropriate power source.

According to the principle of formation, sources of negative ions can be divided into the following groups:

- sources of volumetric formation (plasma), in which ions are formed in the volume of the plasma;

- sources of surface formation, in which ions are formed on the surface of electrodes or special targets;

- surface plasma sources, in which ions are formed on the surfaces of electrodes interacting with plasma particles, which were developed by the Novosibirsk group [3]; and

- sources of charge exchange, in which negative ions are formed as a result of charge exchange of beams of accelerated positive ions on various targets.

To form plasma in modern bulk sources of Н ^- ions, similar to the source of positive ions, arc discharges with thermionic filaments or hollow cathodes, as well as radio frequency discharges in hydrogen, are used. To improve the confinement of electrons during the discharge and to reduce the hydrogen density in the gas discharge chamber, which is important for sources of negative ions, discharges in a magnetic field are used. Systems with an external magnetic field (ie, with Penning geometry or magnetron geometry of electrodes, with electron oscillations in a longitudinal magnetic field of a "reflective" discharge) and systems with a peripheral magnetic field (multipolar) are widely used. A sectional view of a discharge chamber with a peripheral magnetic field developed for a jet injector of a beam of neutral particles is shown in Fig. 4 [3]. The magnetic field at the periphery of the plasma box is formed by means of permanent magnets installed on its outer surface. The magnets are placed in rows in which the direction of magnetization is constant or changes in displacement order, so that the magnetic field lines have the geometry of linear or staggered protrusions near the wall.

The use of systems with a multi-pole magnetic field at the periphery of plasma chambers, in particular, makes it possible for systems to maintain dense plasma in the source at a reduced working gas pressure in the chamber up to 1-4 Pa (without cesium) and up to 0.3 Pa in systems with cesium [4 ]. Such a decrease in the hydrogen density in the discharge chamber, in particular, is of importance for high-current multi-aperture giant ion sources, which are being developed for use in research in the field of thermonuclear fusion.

Currently, ion sources based on surface-plasma formation are considered to be the most suitable for the formation of high-current negative ion beams.

In ion sources based on surface plasma formation, ions are formed by interaction between particles having sufficient energy and a surface with a low work function. This effect can be enhanced by alkaline coating of the bombarded surface. There are two main processes, namely thermodynamically equilibrium surface ionization, in which a slow atom or molecule colliding with a surface is emitted back as a positive or negative ion after an average residence time, and a nonequilibrium (kinetic) atomic-surface interaction, in which negative Ions are formed by sputtering, shock desorption (as opposed to thermal desorption, in which thermal particles are desorbed), or reflection with an alkali metal coating. In the process of thermodynamically equilibrium ionization, the adsorbed particles are detached from the surface under conditions of thermal equilibrium. The ionization coefficient of particles leaving the surface is determined using the Saha formula and is assumed to be very small ~ 0.02%.

. Поле картины распределения зарядов усиливает энергетический уровень дополнительного электрона относительно энергетических уровней электронов в металле. Начиная с некоторого критического расстояния, уровень дополнительного электрона становится выше верхнего энергетического уровня электронов в металле, и резонансное туннелирование возвращает электрон от уходящего иона обратно в металл. В случае если частица отрывается достаточно быстро, коэффициент отрицательной ионизации предположительно является довольно высоким для поверхности с низкой работой выхода, которая может предоставляться посредством нанесения покрытия из щелочного металла, в частности, цезия.The processes of nonequilibrium kinetic surface ionization are presumably much more efficient at the surface and have a rather low work function, comparable to the electron affinity of a negative ion. During this process, the negative ion is detached from the surface, overcoming the subsurface barrier using the kinetic energy obtained from the primary particle. Near the surface, the energy level of an extra electron is below the upper Fermi level of electrons in a metal, and this level can very easily be dealt with by tunneling electrons out of the metal. During ion movement from the surface, it overcomes the potential barrier formed by the mirror charge

... The field of the charge distribution pattern enhances the energy level of an additional electron relative to the energy levels of electrons in the metal. Starting from a certain critical distance, the level of the additional electron rises above the upper energy level of electrons in the metal, and resonant tunneling returns the electron from the leaving ion back to the metal. If the particle is peeled off quickly enough, the negative ionization coefficient is presumably quite high for a surface with a low work function that can be provided by coating with an alkali metal, in particular cesium.

=0,67. Следует отметить, что работа выхода на вольфрамовых поверхностях имеет минимальное значение с покрытием Cs в 0,6 монослоев (на поверхности вольфрамового кристалла 110).It has been shown experimentally that the degree of negative ionization of hydrogen particles detached from this surface with a reduced work function can reach

= 0.67. It should be noted that the work function on tungsten surfaces has a minimum value with a Cs coating of 0.6 monolayers (on the 110 tungsten crystal surface).

For the development of sources of negative hydrogen ions, it is important that the integral yield of negative ions is sufficiently high, K ^- = 9-25%, for collisions of hydrogen atoms and positive ions with energies of 3-25 eV with surfaces with a low work function, such as Mo + Cs , W + Cs [5]. In particular (see Fig. 5), when bombarding a cised molybdenum surface with Franck-Condon atoms with an energy exceeding 2 eV, the integral conversion efficiency into H ^- ions can reach K ^- ~ 8%.

In surface plasma sources (SPS) [3], the formation of negative ions is realized due to kinetic surface ionization, namely, the processes of sputtering, desorption or reflection on electrodes in contact with the gas-discharge plasma. Low work function special emitter electrodes are used in the SPS to improve negative ion formation. As a rule, the addition of a small amount of cesium to the discharge makes it possible to obtain an increase in brightness and intensity in the collector of Hˉ beams. The introduction of cesium atoms into the discharge significantly reduces the concomitant flux of electrons pulled out with negative ions.

In SPS, gas-discharge plasma performs several functions, namely, it forms intense streams of particles that bombard the electrodes; the plasma sheath adjacent to the electrode forms the acceleration of the ions, thereby increasing the energy of the bombarding particles; negative ions, which are formed in electrodes with a negative potential, are accelerated by the potential of the plasma sheath and penetrate through the plasma layer into the stretching region without significant destruction. Intensive formation of negative ions with rather high efficiencies of power and gas utilization was obtained in various modifications of SPS under conditions of a "dirty" gas discharge and intense bombardment of electrodes.

Several SPS sources are designed for large fusion devices such as the LHD, JT-60U and the international (ITER) tokamak.

Typical features of these sources can be understood when considering the stellarator injector LHD [4], which is shown in Fig.6 [4, 6]. Arc plasma is generated in a large magnetic multi-pole vane enclosure chamber with a volume of ~ 100 liters. Twenty-four tungsten filaments support a 3 kA, ~ 80 V arc at a hydrogen pressure of approximately 0.3-0.4 Pa. An external magnetic filter with a maximum field at the center of ~ 50 G provides the electron density and temperature drop in the pull-out region near the plasma electrode. A positive bias of the plasma electrode (~ 10 V) reduces the accompanying electron flow. Negative ions are formed on a plasma electrode coated with an optimal cesium layer. External cesium furnaces (three for one source) equipped with pneumatic valves feed the distributed introduction of cesium atoms. The formation of negative ions reaches a maximum at an optimum temperature of the plasma electrode of 200-250 ^o C. The plasma electrode is thermally insulated and its temperature is determined by means of a plasma discharge of power loads.

A four-electrode multi-aperture ion-optical system that is used in the LHD ion source is shown in Fig. 7 [6]. Negative ions are drawn through 770 radiation apertures with a diameter of 1.4 cm. The apertures occupy an area of 25x125 cm ² on the plasma electrode. Small permanent magnets are embedded in the pulling grid between the apertures to deflect the co-pulling electrons from the beam onto the pulling electrode wall. An additional electronic retention grid, installed behind the pulling grid, traps secondary electrons backscattered or emitted from the walls of the pulling electrodes. A multi-slot grounded grid with high transparency is used in the ion source. This reduces the area of intersection of the beams, thereby increasing the voltage holding capacity and lowering the gas pressure in the gaps by a factor of 2.5 with a corresponding decrease in the beam stripping losses. Both the pull electrode and the ground electrode are water cooled.

The introduction of cesium atoms into a multipoint source provides a 5-fold increase in the current of extracted negative ions and a linear increase in the yield of H ^- ions over a wide range of discharge powers and pressures when filled with hydrogen. Other important advantages of the introduction of cesium atoms are a ~ 10-fold decrease in the jointly drawn electron current and a significant decrease in the hydrogen pressure during the discharge to 0.3 Pa.

Multipoint sources in LHD typically deliver approximately 30 A ion current with a current density of 30 mA / cm ² in 2 second pulses [6]. The main problems for LHD ion sources are blocking of cesium, which is injected into the arc chamber, by sputtering tungsten from the filaments, and a decrease in the high voltage holding capacity when operating in long pulse mode at high power levels.

The LHD negative ion neutral particle injector has two ion sources interacting with hydrogen at a nominal beam energy of 180 keV. Each injector achieves a nominal injection power of 5 MW for a pulse of 128 seconds, so each ion source provides a beam of neutral particles of 2.5 MW. 8A and B show an LHD neutral beam injector. The focal length of the ion source is 13 m and the pivot point of the two sources is 15.4 m lower. The injection port is approximately 3 m long, with the narrowest part having a diameter of 52 cm and a length of 68 cm.

Ion sources with RF plasma shapers and the formation of negative ions on a cesium-coated plasma electrode are developed at IPP Garching. RF shapers produce cleaner plasma so there is no tungsten blocking of cesium in these sources. Steady-state pulling of a negative ion beam pulse with a beam current of 1 A, an energy of ~ 20 kV, and a duration of 3600 seconds was demonstrated by IPP in 2011.

At present, high-energy neutral particle beam injectors, which are being developed for the next stage of thermonuclear devices, such as, for example, the ITER tokamak, do not demonstrate stable operation at the required energy of 1 MeV and operation in a steady-state or continuous-wave mode ( CW) at a sufficiently high current. Therefore, there is a need to develop feasible solutions if problems can be solved that prevent the achievement of target beam parameters, such as, for example, a beam energy in the range of 500-1000 keV, an effective current density in neutral particles of the main port of a reservoir of 100-200 A / m ³ , the power per neutral beam injector is approximately 5-20 MW, the pulse duration is 1000 seconds, and the gas loads introduced by the beam injector are less than 1-2% of the beam current. It should be noted that achieving this goal becomes much less costly if the negative ion current in the injector module is reduced to an extraction ion current of up to 8-10 A, compared to an ion extraction current of 40 A for an ITER beam. A stepwise reduction in the extracted current and beam power should lead to dramatic changes in the design of key elements of the ion source in the form of an injector and a high-energy accelerator, so that much more elaborate technologies and approaches become applicable, which increases the reliability of the injector. Therefore, in the current situation, an extractable current of 8-10 A per module is proposed, on the assumption that the required extraction power of injection can be obtained using several injector modules that form beams with low divergence and high current density.

The performance of surface plasma sources is fairly well documented, and multiple ion sources in operation today produce continuous scalable ion beams in excess of 1 A or higher. Until now, the main parameters of neutral particle beam injectors, such as beam power and pulse duration, are rather far from those required for the injector under consideration. The current state of development of these injectors can be understood from Table 1.

Table 1 TAE ITER JT-60U LHD IPP CEA-JAERI Current density (A / m ² ) 200 D ^-
280 H ^- 100 D ^- 350 H ^- 230 D ^-
330 H ^- 216 D ^-
195 H ^- Beam energy (keV) 1000 H ^- 1000 D ^-
100 H ^- 365 186 9 25 Pulse duration (sec) ≥1000 3600 D ^-
3 H ^- 19 ten <6 five
1000 The ratio of the number of electrons to the number of ions one ~ 0.25 <1 <1 <1 Pressure (pa) 0.3 0.3 0.26 0.3 0.3 0.35 Comments Combined numbers have not yet been reached, full-scale experiments are being carried out at IPP Garching - the source of long pulses (MANITU) currently provides 1 A / 20 kV for 3600 seconds at D ^- Filament source Filament source RF source, partial pull, test bench known as BATMAN, operates at 2 A / 20 kV for ~ 6 sec Source KamabokoIII (JAERI) on MANTIS (CEA)

Therefore, it is desirable to provide an improved neutral beam injector.

Summary of the invention

The embodiments provided herein are directed to systems and methods for a negative ion neutral beam injector. A negative ion neutral beam injector contains an ion source, an accelerator and a neutralizer in order to form a neutral particle beam of approximately 5 MW with an energy of approximately 0.50-1.0 MeV. The ion source is located in a vacuum tank and forms a 9 A negative ion beam. Ions generated by the ion source are pre-accelerated to 120 kV prior to injection into the high-energy accelerator by an electrostatic pre-accelerator based on a multi-aperture grid in the ion source, which is used to pull the ion beams out of the plasma and accelerate to a certain fraction of the required beam energy. The 120 keV beam from the ion source passes through a pair of deflecting magnets that allow the beam to be displaced axially before entering the high energy accelerator. After acceleration to full energy, the beam enters a neutralizer, in which it is partially converted into a beam of neutral particles. The remaining types of ions are separated by means of a magnet and sent to electrostatic energy converters. The beam of neutral particles passes through the shut-off valve and enters the plasma chamber.

The elevated temperature of the plasma formers and the inner walls of the plasma box of the ion source (150-200 ° C) is maintained to prevent the accumulation of cesium on their surfaces. A distribution manifold is provided to deliver cesium directly to the surface of the plasma gratings and not to the plasma. This is in contrast to existing ion sources that feed cesium directly into the plasma discharge chamber.

The magnetic field used to deflect the co-elongated electrons in the ion elongation and pre-acceleration regions is generated by external magnets, rather than by magnets embedded in the mesh body as in prior designs. The absence of built-in "low-temperature" magnets in the grids makes it possible to heat them up to higher temperatures. Prior designs often use magnets embedded in the mesh casing, which often results in a significant reduction in the drawn beam current and prevents high temperature operation as well as adequate heating / cooling performance.

The high voltage accelerator is not directly connected to the ion source, but is separated from the ion source by a transition zone (LEBT) with deflecting magnets, vacuum pumps and cesium traps. The transition zone intercepts and removes most of the co-flowing particles, including electrons, photons and neutral particles from the beam, evacuates the gas released from the ion source and prevents it from reaching the high voltage accelerator, prevents cesium from flowing out of the ion source and penetrating the high voltage accelerator , prevents electrons and neutral particles produced by stripping negative ions from entering the high voltage accelerator. In prior designs, the ion source is directly coupled to the high voltage accelerator, which often makes the high voltage accelerator susceptible to gas, charged particles, and cesium flowing out of and into the ion source.

The deflection magnets in the LEBT deflect and focus the beam along the axis of the accelerator and thereby compensate for any displacement and deflection of the beam during transport through the magnetic field of the ion source. The displacement between the axes of the pre-accelerator and the high-voltage accelerator reduces the flow of co-flowing particles into the high-voltage accelerator and prevents highly accelerated particles (positive ions and neutral particles) from flowing back into the pre-accelerator and the ion source. The focusing of the beam also contributes to the homogeneity of the beam entering the accelerator compared to systems based on a multi-aperture grid.

The neutralizer includes a plasma neutralizer and a photoneutralizer. The plasma neutralizer is based on a multipoint plasma confinement system with permanent magnets of strong magnetic fields on the walls. The photonic converter is a photon trap based on a cylindrical resonator with walls with a high degree of reflection and pumped out using lasers with high efficiency. These neutralizer technologies have never been considered for commercial neutral beam injectors.

Other systems, methods, features, and advantages of the exemplary embodiments will become apparent to those skilled in the art upon examination of the accompanying drawings and detailed description.

Brief Description of Drawings

The details of the exemplary embodiments, including structure and mode of operation, may be revealed in part by examination of the accompanying drawings, in which like reference numerals refer to like parts. The components in the drawings do not have to be drawn to scale, instead the emphasis is on illustrating the principles of the invention. Moreover, all illustrations are intended to convey general ideas, and relative sizes, shapes, and other detailed attributes may be illustrated schematically rather than literally or accurately.

1 is a top view of a schematic diagram of a negative ion neutral beam injector.

FIG. 2 is a cross-sectional isometric view of the negative ion neutral beam injector shown in FIG. 1. FIG.

3 is a top view of a high power neutral particle injector for an ITER tokamak.

4 is an isometric cross-sectional view of a discharge chamber with a peripheral multi-pole magnetic field for a neutral particle beam jet injector.

5 is a graph showing the cumulative yield of negative ions formed by bombarding a Mo + Cs surface with neutral H atoms and positive molecular H as a function of incident flux energy. The yield is improved by using DC curing as compared to only preliminary surface curing.

6 is a top view of a negative ion source for an LHD.

7 is a schematic view of a multi-aperture ionic optical system for an LHD source.

8A and B are top and side views of an LHD neutral beam injector.

Fig. 9 is a cross-sectional view of an ion source.

10 is a cross-sectional view of a source of low energy hydrogen atoms.

11 is a graph showing the trajectories of H ^- ions in the low energy path.

12 is an isometric view of an accelerator.

13 is a diagram showing the trajectories of an ion in an accelerating tube.

14 is an isometric view of a triplet of quadrupole lenses.

15 is a diagram showing a top view (a) and a side view (b) of ion trajectories in a high energy beam transport line accelerator.

16 is an isometric view of an arrangement of plasma targets.

Fig. 17 is a diagram showing the results of two-dimensional calculations of ion beam deceleration in the recuperator.

It should be noted that elements of like structures or functions are generally represented by like reference numerals for purposes of illustration throughout the drawings. It should also be noted that the drawings are only intended to facilitate describing the preferred embodiments.

Description of preferred embodiments of the invention

Each of the additional features and ideas disclosed below can be used alone or in combination with other features and ideas to provide a new negative ion neutral beam injector. Specific examples of the embodiments described herein are described in more detail below, these examples employing many of these additional features and ideas, either individually or in combination, with reference to the accompanying drawings. This detailed description is intended only to educate those skilled in the art in additional detail for practicing preferred aspects of the teachings of the present invention, and is not intended to limit the scope of the invention. Therefore, combinations of features and steps disclosed in the following detailed description may be optional in order to practice the invention in its broadest sense, and instead are studied simply to specifically describe typical examples of the present ideas.

Moreover, various features of the typical examples and dependent claims may be combined in ways that are not specifically and explicitly listed to provide additional useful embodiments of the present teachings. In addition, it should be explicitly noted that all the features disclosed in the description and / or the claims are intended to be disclosed separately and independently from each other for the purposes of the original disclosure, as well as for the purpose of limiting the claimed subject matter, regardless of the arrangements of the features in the embodiments. implementation and / or in the claims. It should also be noted that all ranges of values or group designators disclose every possible intermediate value or intermediate object for purposes of the original disclosure as well as for purposes of limiting the claimed subject matter.

The embodiments provided herein are directed to a novel negative ion neutral beam injector with an energy of preferably about 500-1000 keV and a high overall energy efficiency. A preferred arrangement of an embodiment of a negative ion neutral beam injector 100 is illustrated in FIGS. 1 and 2. As illustrated, the injector 100 includes an ion source 110, a check valve 120, deflecting magnets 130 for deflecting the low energy beam line, a support insulator 140 , high energy accelerator 150, check valve 160, neutralizing tube (shown schematically) 170, separating magnet (shown schematically) 180, check valve 190, panels 200 and 202 for evacuation, vacuum tank 210 (which is part of the vacuum reservoir 250, explained below), cryosorption pumps 220 and triplet quadrupole lenses 230. Injector 100, as noted above, contains an ion source 110, an accelerator 150 and a neutralizer 170 in order to generate a beam of neutral particles of about 5 MW with an energy of about 0.50-1 , 0 MeV. The ion source 110 is located in the vacuum tank 210 and forms a 9 A negative ion beam. The vacuum tank 210 is shifted to -880 kV, i.e. relative to the ground, and is mounted on insulating supports 140 inside a tank 240 of a larger diameter filled with SF ₆ gas. The ions generated by the ion source are pre-accelerated to 120 kV prior to injection into the high-energy accelerator 150 by the multi-aperture mesh electrostatic pre-accelerator 111 (see FIG. 9) in the ion source 110, which is used to pull the ion beams out of the plasma and accelerate to some fraction of the required beam energy. The 120 keV beam from the ion source 110 passes through a pair of deflecting magnets 130 which allow the beam to be offset off axis prior to entering the high energy accelerator 150. The evacuation panels 202, shown between the deflection magnets 130, include a baffle and a cesium trap.

It is assumed that the gas efficiency of the ion source 110 is approximately 30%. The projected negative ion beam current of 9-10 A corresponds to a gas injection of 6-7 l ∙ Torr / s in the 110 ion source. The neutral gas flowing out of the ion source 110 increases its average pressure in the pre-accelerator 111 to about 2x10 ^-4 Torr. At this pressure, the neutral gas results in a ~ 10% loss for stripping the ion beam in the preaccelerator 111. Between the deflecting magnets 130, neutral particle dumps (not shown) are provided, which are a consequence of the primary negative ion beam. Also provided are vents (not shown) for positive ions flowing back from the high energy accelerator 150. The differential pumping low energy beam transport area 205 from the pumping panels 200 is used immediately after pre-acceleration to reduce the gas pressure to ~ 10 ^-6 Torr before it reaches the high energy accelerator 150. This introduces an additional beam loss of ~ 5%, but since this occurs at a low pre-acceleration energy, the power loss is relatively small. The recharge loss in the high energy accelerator 150 is below 1% at a background pressure of 10 ^-6 Torr.

After being accelerated to a total energy of 1 MeV, the beam enters the neutralizer 170, where it is partially converted into a beam of neutral particles. The remaining types of ions are separated by the magnet 180 and sent to electrostatic energy converters (not shown). The beam of neutral particles passes through the check valve 190 and enters the plasma chamber 270.

The vacuum reservoir 250 is divided into two sections. One section contains a pre-accelerator 111 and a low energy beam line 205 in a first vacuum tank 210. The other section houses a high energy beam line 265, a neutralizer 170 and charged particle energy converters / recuperators in a second vacuum tank 255. The sections of the vacuum tank 250 are connected through a chamber 260 c high energy booster tube 150 inside.

The first vacuum tank 210 is the vacuum boundary of the pre-accelerator 111 and the low energy beam line 205, and SF ₆ is pressurized in the larger tank or external reservoir 240 to isolate the high voltage. Vacuum tanks 210 and 255 act as support structures for indoor equipment such as magnets 130, cryosorption pumps 220, etc. The removal of heat from the internal heat transfer components must be carried out by means of cooling tubes, which must have insulation breaks in the case of the first vacuum tank 210, which is displaced to -880 kV.

Ion source

A schematic diagram of an ion source 110 is shown in FIG. 9. The ion source includes: electrostatic multi-aperture pre-accelerating grids 111, ceramic insulators 112, RF plasma formers 113, permanent magnets 114, plasma box 115, channels and headers 116 for cooling water and gas valves 117. In the ion source 110, a cesium surface plasma pre-accelerating grids 111 is used to convert the positive ions and neutral atoms generated by the plasma shapers 113 into negative ions in the plasma expansion volume (the volume between the shapers 113 and the grids 111 indicated by the parenthesis labeled "PE" in FIG. 9) with retention in the form of a magnetic multi-pole vane, as provided by permanent magnets 114.

The positive bias voltage for receiving electrons in the plasma pre-acceleration grids 111 is applied to optimized conditions for negative ion formation. Geometric shaping of the apertures 111B in the plasma pre-accelerating grids 111 is used to focus the H ^- ions into the pulling grid apertures 111B. A small transverse magnetic filter formed by external permanent magnets 114 is used to reduce the temperature of the electrons scattered from the shaper region or the PE region of the plasma emitter of the plasma box 115 to the pulling region ER of the plasma box 115. The electrons in the plasma are reflected from the pulling region ER by field of a small transverse magnetic filter formed by external permanent magnets 114. Ions are accelerated to 120 kV before injection into high energy accelerator 150 by plasma grids 111 of electrostatic multi-aperture pre-accelerator in ion source 110. Before being accelerated to high energy, the ion beam has a diameter of approximately 35 cm. The ion source 110 must therefore generate 26 mA / cm ² in the 111B apertures assuming 33% transparency in the plasma grids 111 of the preaccelerator. Compared to previously obtained values, this represents a reasonably reasonable projection for the ion source 110.

The plasma that enters the plasma box 115 is formed by an array of plasma shapers 113 mounted on the rear flange 115A of the plasma box, which is preferably a water-cooled cylindrical copper chamber (700 mm in diameter by 170 mm in length). The open end of the plasma box 115 is limited by the plasma grids 111 of the acceleration and extension system pre-accelerator.

It is assumed that negative ions should be formed on the surface of the plasma grids 111, which are covered with a thin layer of cesium. Cesium is introduced into plasma box 115 using a cesium supply system (not shown in FIG. 9).

The ion source 110 is surrounded by permanent magnets 114 such that it forms a linear tip configuration to confine plasma and primary electrons. Columns 114A of magnets on the cylindrical wall of the plasma box 115 are connected in the rear flange 115A by rows of magnets 114B, which also have a linear-tapered configuration. A magnetic filter near the plane of the plasma grids 111 divides the plasma box 115 into a PE plasma emitter and an ER extraction region. The filter magnets 114C are mounted in the flange 111A adjacent to the plasma grids 111 to provide a transverse magnetic field (B = 107 Gs at the center) that serves to prevent the high energy primary electrons emanating from the ion former 113 from reaching the pull-out region ER. However, positive ions and low energy electrons can be scattered through the filter in the ER pulling region.

The electrode-based drawing and pre-acceleration system 111 comprises five electrodes 111C, 111D, 111E, 111F, and 111G, each of which has 142 holes or apertures 111B formed orthogonally therein and used to provide a negative ion beam. The pull apertures 111B have a diameter of 18 mm, so that the total ion pull area of these 142 pull apertures is approximately 361 cm ² . The density of the negative ion current is 25 mA / cm ² , and it is required to form an ion beam of 9 A. The magnetic field of the 114C magnets in the filter enters the gaps between the electrostatic pulling and pre-accelerating grids 111 to deflect the co-drawn electrons into special slots in the inner surface of the apertures 111B in pulling electrodes 111C, 111D and 111E. The magnetic field of the magnets in the magnetic filter 114C together with the magnetic field of the additional magnets 114D provides deflection and interception of electrons co-pulled with negative ions. The additional magnets 114D include an array of magnets installed between the electrode holders 111F and 111G of the accelerator grid downstream of the pulling grid containing pulling electrodes 111C, 111D, and 111E. The third grid electrode 111E, which accelerates negative ions to 120 keV, is positively biased away from the grounded grid electrode 111D to reflect backflowing positive ions entering the pre-accelerating grid.

Plasma drivers 113 include two alternatives, namely, an RF plasma driver and an atomic arc driver. BINP's developed arc-based plasma-arc generator is used in an atomic shaper. A special feature of an arc-based plasma generator is the formation of a directed plasma jet. The ions in the expanding jet move without collisions and, due to acceleration through the fall of the ambipolar plasma potential, receive energies of ~ 5-20 eV. The plasma jet can be directed at an inclined molybdenum or tantalum surface of the transducer (see 320 in Fig. 10), on which a stream of hydrogen atoms is formed as a result of neutralization and reflection of the jet. The energy of hydrogen atoms can be increased beyond the initial 5–20 eV by negatively displacing the converter relative to the plasma box 115. Experiments on obtaining intense fluxes of atoms with such a converter were carried out at the Budker Institute in 1982-1984.

In Fig. 10, the designed arrangement of the low energy atom source 300 is shown as including a gas valve 310, a cathode insert 312, an electrical outlet to a heater 314, cooling water manifolds 316, an electron emitter 318 LaB6, and an ion-to-atom converter 320. In experiments, formed a stream of hydrogen atoms with an equivalent current of 20-25 A and an energy ranging from 20 eV to 80 eV, with an efficiency of more than 50%.

Such a source can be used in a negative ion source to supply atoms with energy optimized for efficient formation of negative ions on the cesium surface of the plasma grids 111.

Low energy beam transport line

Ions H ^- formed and preliminarily accelerated to an energy of 120 keV by the ion source 110 when passing along the low-energy beam transport line 205 are displaced perpendicular to their direction of motion by 440 mm with a deflection by the peripheral magnetic field of the ion source 110 and by the magnetic field of two special wedge-shaped deflecting magnets 130. This negative ion beam bias in the low energy beam transport line 205 (as illustrated in FIG. 11) is provided to separate the regions of the ion source 110 and the high energy accelerator 150. This bias is used to prevent the penetration of fast atoms resulting from the stripping of the H ^- beam on the residual hydrogen in the accelerating tube 150, to reduce the fluxes of cesium and hydrogen from the ion source 110 to the accelerating tube 150, as well as to retard the flux of secondary ions from the accelerating tube 150 to the ion source 110. 11 shows the computed trajectories of H ^- ions in the low energy beam transport line.

High energy beam path

The low energy beam emanating from the low energy beam line enters a conventional electrostatic multi-aperture accelerator 150 shown in FIG. 12.

The results of calculating the acceleration of the negative ion beam by 9 A, taking into account the space charge fraction, are shown in Fig. 13. Ions are accelerated from an energy of 120 keV to 1 MeV. The accelerating potential across tube 150 is 880 kV, and the potential step between the electrodes is 110 kV.

Calculations show that the field strength does not exceed 50 kV / cm in the optimized accelerating tube 150 at the electrodes in the zones of possible electron discharge.

After acceleration, the beam passes through a triplet 230 of commercial conventional quadrupole lenses 231, 232 and 233 (FIG. 14), which are used to compensate for slight beam defocusing at the exit of accelerating tube 150 and to form a preferred beam size at the exit port. The triplet 230 is installed in the vacuum tank 255 of the high energy beam transport line 265. Each of the quadrupole lenses 231, 232, and 233 includes a traditional array of quadrupole electromagnets that generate the usual magnetic focusing fields found in all modern conventional particle accelerators.

The computed trajectories of a 9 A negative ion beam with a 12 eV transverse temperature in accelerating tube 150, quadrupole lenses 230, and high energy beam transport line 265 are shown in FIG. 15. The calculation corresponds to the beam outside of its focusing point.

The calculated diameter of a beam of neutral particles with an equivalent current of 6A after the neutralizer at a distance of 12.5 m at half-height of the radial profile is 140 mm, and 95% of the beam current is in a circle with a diameter of 180 mm.

Neutralization

The photodetachment neutralizer 170 selected for the beam system achieves greater than 95% ion beam stripping. The neutralizer 170 includes an array of xenon lamps and a cylindrical light trap with highly reflective walls to provide the required photon density. Chilled mirrors with a reflectance greater than 0.99 are used to provide a wall power flux of approximately 70 kW / cm ² . Alternatively, a plasma neutralizer can be used instead using conventional technology, but at the cost of a slight decrease in efficiency. However, the neutralization efficiency of ~ 85% of the plasma element is quite sufficient if the energy recovery system has an efficiency> 95% as predicted.

Plasma in the plasma neutralizer is held in a cylindrical chamber 175 with a multi-pole magnetic field on the walls, which is formed by an array of permanent magnets 172. A general view of the holding device is shown in Fig.16. The neutralizer 170 includes cooling water collectors 171, permanent magnets 172, cathode assemblies 173, and LaB6 cathodes 174.

The cylindrical chamber 175 is 1.5-2 m long and has holes at the ends for the beam to pass. Plasma is formed by using several cathode assemblies 173 installed in the center of the containment chamber 175. The working gas is supplied near the center of the device 170. In experiments with the prototype of such a plasma neutralizer 170, it should be noted that the confinement of electrons by means of multi-pole magnetic fields 172 on the walls is quite good and much better retention of plasma ions. In order to equalize the loss of ions and electrons, a significant negative potential develops in the plasma, so that the ions are effectively confined by the electric field.

A sufficiently long holding of the plasma results in a relatively low level of discharge power required to maintain a plasma density of approximately 10 ¹³ cm ^-3 in neutralizer 170.

Energy recovery

There are objective reasons for achieving high power efficiency in our conditions. First of all, these are the following: a relatively small ion beam current and scattering at low energy. In the considered scheme, when using plasma or vaporous metal targets, one can expect that the residual ion current should be ~ 3 A after the neutralizer. These streams of diverted ions with a positive or negative charge must be deflected through the deflecting magnet 180 to two energy recuperators, one for positive and negative ions, respectively. Numerical simulations of the deceleration of these residual beams of diverted ions are carried out, typically with energies of 1 MeV and 3A in direct converters in recuperators without space charge compensation. The direct converter converts a significant part of the energy contained in the residual ion beam directly into electricity and supplies the rest of the energy as high quality heat for inclusion in the heat cycle. Direct converters correspond to the design of an electrostatic multi-aperture moderator, as a result of which successive sections of charged electrodes form longitudinal breakdown fields and absorb the kinetic energy of ions.

FIG. 17 shows the results of two-dimensional calculations of the ion beam deceleration in the converter. From the presented calculations, it follows that the deceleration of an ion beam with an energy of 1 MeV to an energy of 30 keV is quite feasible, so that a value of the recovery coefficient of 96-97% can be obtained.

Previous efforts to develop high power neutral ion beam injectors based on negative ions have been analyzed to reveal critical issues still preventing the achievement of steady state injectors of ~ 1 MeV and a power of several MW. Of the most important, we highlight the following:

- Control of the cesium layer as well as losses and redeposition (temperature control, etc.)

- Optimization of surface formation of negative ions for pulling

- Separation of co-flowing electrons

- Inhomogeneity of the ion current profile in the plasma grid due to internal magnetic fields

- Low ion current density

- Accelerators are becoming more sophisticated and many new technologies are still being developed (low voltage withstand capability, large insulators, etc.)

- Reverse flow of positive ions

- Advanced neutralizer technologies (plasma, photons) are not demonstrated under relevant conditions

- Energy conversion is not well developed

- Blocking the beam in the tract

The innovative solutions to the problems provided for in this document can be grouped according to the system with which they are connected, namely negative ion source, pull / acceleration, neutralizer, energy converters, etc.

1.0 110 negative ion source:

1.1. An elevated temperature of the inner walls of the plasma box 115 and plasma shapers 113 (150-200 ° C) is maintained to prevent the accumulation of cesium on their surfaces.

Elevated temperature:

- prevents uncontrolled release of cesium due to desorption / sputtering and reducing its penetration into the ionic optical system (111 grids),

- reduces the absorption and recombination of hydrogen atoms in the cesium layer on the walls,

- reduces the consumption and poisoning of cesium.

To achieve this, a high temperature fluid is circulated through all components. Surface temperature is additionally stabilized through active feedback control, i.e. heat is removed or added during CW and transient operation. In contrast to this approach, all other existing and planned beam injectors use passive systems with water cooling and thermal breakdowns between the cooling tubes and hot electrode bodies.

1.2. Cesium is fed through the distribution manifold directly to the surface of the plasma grids 111, and not to the plasma. Supply of cesium through the distribution manifold:

- provides a controlled and distributed supply of cesium during the entire time of beam activation,

- prevents cesium deficiency, typically due to blocking by plasma,

- reduces the release of cesium from the plasma after its accumulation and unblocking during long pulses.

In contrast, existing ion sources supply cesium directly to the discharge chamber.

2.0 Pre-accelerator 111 (100 keV):

2.1. The magnetic field used to deflect the co-elongated electrons in the stretching and pre-acceleration regions of the ions is generated by external magnets, rather than by magnets embedded in the mesh body, as in previous designs:

- the magnetic field lines in the high voltage gaps between the grids are completely concave in the direction of the negatively biased grids, i.e. in the direction of the plasma grid in the pulling gap and in the direction of the pulling grid in the preliminary accelerating gap. The concavity of the magnetic field lines in the direction of negatively biased grids prevents local Penning traps in high voltage gaps and trapping / multiplication of co-extracting electrons, which can occur in configurations with built-in magnets.

- the electrodes of the ion optical system (IOS) (111 grids) without built-in "low-temperature" NIB-magnets can be heated to an elevated temperature (150-200 ° C) and provide the possibility of heat removal during long pulses by using a hot (100-150 ° C) liquids.

- the absence of built-in magnets leaves free space between the radiation apertures of the grids and allows the introduction of channels for more efficient heating / cooling of the electrodes.

In contrast, prior designs use magnets embedded in the mesh body. This leads to the creation of static magnetoelectric traps in high voltage gaps, which trap and increase the co-extracting electrons. This can lead to a significant decrease in the extracted beam current. This also prevents high temperature operation as well as proper heating / cooling performance, which is critical for long pulse operation.

2.2. The elevated temperature of all electrodes of the ionic optical system (grid 111) (150-200 ° C) is always maintained to prevent the accumulation of cesium on their surfaces and to increase the intensity of the high voltage pulling and pre-accelerating gaps. In contrast, in traditional designs, the electrodes are water cooled. The electrodes have elevated temperatures because there is thermal breakdown between the cooling tubes and the electrode bodies and there is no active feedback.

2.3. The initial heating of grids 111 at startup and heat removal during the beam activation phase is performed by passing a hot liquid with a controlled temperature through the internal channels in the grids 111.

2.4. The gas is additionally pumped from the pre-accelerating gap through the side space and large holes in the mesh holders to reduce the gas pressure along the beam line and delay the stripping of negative ions and the formation / multiplication of secondary particles in the gaps.

2.5. The inclusion of positively biased grids 111 is used to repel back flowing positive ions.

3.0 High voltage accelerator 150 (1 MeV):

3.1. The high voltage accelerator 150 is not directly connected to the ion source, but is separated from the ion source by a transition zone (low energy beam transport line - LEBT 205) with deflecting magnets 130, vacuum pumps and cesium traps. Transition zone:

- intercepts and removes most of the co-flowing particles, including electrons, photons and neutral particles from the beam,

- evacuates the gas released from the ion source 110 and prevents it from reaching the high voltage accelerator 150,

- prevents leakage of cesium from the ion source 110 and penetration into the high voltage accelerator 150,

- prevents electrons and neutral particles formed by stripping negative ions from entering the high voltage accelerator 150.

In the prior art, the ion source is directly connected to the high voltage accelerator. This makes the high voltage accelerator susceptible to gas, charged particles, and cesium leaking out of and into the ion source. This strong interference reduces the voltage holding capacity of the high voltage accelerator.

3.2. Deflection magnets 130 in LEBT 205 deflect and focus the beam along the axis of the accelerator. Deflection magnets 130:

- compensate for all displacements and deflections of the beam during transport through the magnetic field of the ion source 110,

- the displacement between the axes of the pre-accelerator and the high-voltage accelerator 111 and 150 reduces the flow of co-flowing particles into the high-voltage accelerator 150 and prevents highly accelerated particles (positive ions and neutral particles) from flowing back into the pre-accelerator 111 and the ion source 110.

In contrast, prior systems do not have a physical separation between acceleration stages and, as a consequence, do not allow for axial displacements as shown in this document.

3.3. The low-energy beam line magnets 205 focus the beam at the input of the single-aperture accelerator 150:

- Focusing the beam contributes to the homogeneity of the beam entering the accelerator 150, compared to systems based on a multi-aperture grid.

3.4. Application of a single-aperture accelerator:

- simplifies system alignment and beam focusing

- helps to pump gas and remove secondary particles from the high energy accelerator 150

- reduces the loss of the beam at the electrodes of the high energy accelerator 150.

3.5. Magnetic lenses 230 are used after acceleration to compensate for refocusing in accelerator 150 and generate a quasi-parallel beam.

In traditional designs, there is no means for beam focusing and deflection except for the accelerator itself.

4.0. Neutralizer 170:

4.1. Plasma neutralizer based on a multipoint plasma confinement system with permanent magnets of strong fields on the walls;

- increases the efficiency of neutralization,

- minimizes the total losses of the neutral particle beam injector.

These technologies have never been considered for commercial neutral beam injectors.

4.2. Photon neutralizer is a photon trap based on a cylindrical resonator with walls with a high degree of reflection and pumping out using lasers with high efficiency:

- additionally increases the efficiency of neutralization,

- additionally minimizes the total losses of the neutral particle beam injector.

These technologies have never been considered for commercial neutral beam injectors.

5.0. Recuperators:

5.1. Application of the recuperator (s) of residual ion energy:

- increases the overall efficiency of the injector.

In contrast, recuperation is not expected at all in traditional designs.

Bibliographic list

[1.] L. W. Alvarez, Rev. Sci. Instrum. 22, 705 (1951)

[2.] R. Hemsworth et al., Rev. Sc. Instrum., Volume 67, p. 1120 (1996)

[3.] Capitelli M. and Gorse C., IEEE Trans on Plasma Sci, 33, no. 6, pp. 1832-1844 (2005)

[4.] Hemsworth R. S., Inoue T., IEEE Trans on Plasma Sci, 33, no. 6, pp. 1799-1813 (2005)

[5.] B. Rasser, J. van Wunnik and J. Los, Surf. Sci. 118 (1982) p. 697 (1982)

[6.] Y. Okumura, H. Hanada, T. Inoue et al. AIP Conf. Proceedings # 210, New York, pp. 169-183 (1990)

[7.] O. Kaneko, Y. Takeiri, K. Tsumori, Y. Oka and M. Osakabe et al., "Engineering prospects of negative-ion-based neutral beam injection system from high power operation for the large helical device" , Nucl. Fus., Volume 43, pp. 692-699, 2003

Although the invention is susceptible to various modifications and alternative forms, specific examples thereof are shown in the drawings and described in detail herein. All references are definitely contained in this document. However, it should be understood that the invention is not limited to the specific forms or methods disclosed, but rather that the invention is intended to encompass all modifications, equivalents, and alternatives falling within the spirit and scope of the appended claims.

Claims (17)

1. Negative ion particle beam injector, comprising: an ion source configured to generate a beam of negative ions and including a pre-accelerator, the pre-accelerator comprising an electrostatic array having a plurality of electrodes, the electrostatic array is positively biased to repel flowing back positive ions, and a high-energy accelerator spatially separated from the ion source. 2. An injector according to claim 1, comprising a neutralizer associated with a high energy accelerator. 3. The injector of claim 1, wherein each of the plurality of electrodes has a plurality of holes. 4. The injector of claim 3, wherein the plurality of holes are configured to focus and pass negative ions to form a negative ion beam. 5. An injector according to claim 3 or 4, wherein at least one of the plurality of electrodes is positively biased to pre-accelerate negative ions in the negative ion beam. 6. Injector according to any one of paragraphs. 4 or 5, further comprising a transition zone between the pre-accelerator and the high-energy accelerator. 7. The injector of claim 6, wherein the transition zone comprises a low energy beam transport line. 8. The injector of claim 6, wherein the transition zone comprises a pair of deflecting magnets, the pair of deflecting magnets configured to allow the negative ion beam from the pre-accelerator to displace orthogonally to the direction of travel before entering the high energy accelerator. 9. Injector according to any one of paragraphs. 1-8, in which the ion source includes a plasma box and plasma drivers. 10. An injector according to claim 9, wherein the inner walls of the plasma box are configured to maintain an elevated temperature of 150-200 ° C. 11. Injector according to any one of paragraphs. 1-8, further comprising a distribution manifold for direct supply of cesium to the electrostatic pre-accelerator grid. 12. Injector according to any one of paragraphs. 1-11, in which the pre-accelerator includes external magnets configured to deflect the co-elongated electrons in the ion elongation and pre-acceleration regions. 13. Injector according to any one of paragraphs. 1-12, further comprising a pumping system for pumping gas out of the pre-acceleration interval. 14. The injector of claim 7, wherein the low energy beam transport line includes cesium traps. 15. Injector according to any one of paragraphs. 7 or 14, wherein the low energy beam transport line includes deflecting magnets that are configured to deflect the beam perpendicular to its direction of travel and focus the beam along the axis of the high energy accelerator.

Images