JP2858188B2 - Beam energy control method for high frequency linear accelerator - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a high frequency linear accelerator for accelerating a charged particle beam such as an ion beam, and more particularly to a beam energy control method for a high frequency linear accelerator.

[0002]

2. Description of the Related Art Among ion accelerators for accelerating an ion beam, a drift tube type linear accelerator such as an Alvaret type linear accelerator, a Vidley type linear accelerator, an IH type linear accelerator, etc. Accelerators and RFQ linear accelerators are known.

The RFQ type linear accelerator shown in FIG. 1 has a conductive outer cylinder portion 10 whose both ends are closed by plates, and a beam acceleration axis is defined along the central axis of the conductive outer cylinder portion 10. . Beam passing holes are provided in the center of the plates at both ends. Also, four conductive wings 11 are provided in the conductive outer cylinder 10 so as to surround the beam acceleration axis.
a to 11d are positioned at an angular interval of 90 degrees. In FIG. 1, it is assumed that the ion beam enters from the left plate and exits through the right plate.

[0004] On the beam acceleration axis side of each of the wings 11a to 11d, a waveform is formed along the beam acceleration axis, and this cycle becomes gradually longer from the entrance side to the exit side. As shown in FIG. 2, the waveforms of the wings at positions opposite to each other, that is, at positions 180 degrees apart, for example, 1
1a and 11c or 11b and 11d have the same phase. On the other hand, the waveforms at the wings adjacent to each other, for example, 11a and 11b have a phase difference of 180 degrees. When a high-frequency power of a predetermined frequency is introduced into the cavity formed by the conductive outer cylinder portion 10 and the wing portion, an in-phase high-frequency voltage is applied to two opposing wing portions, and the adjacent wing portion is applied. When high frequency voltages having opposite phases are applied to the portion, a quadrupole accelerated focused electric field is formed along the beam acceleration axis.

In this configuration, a half cycle of the waveform forms a unit cell which is a repeating unit of the acceleration structure (see FIG. 2).

As shown in FIG. 3, the Alvaret-type linear accelerator has a conductive outer cylinder portion 10a whose both ends are closed by plates, similarly to the RFQ-type linear accelerator, and has a beam acceleration axis along its central axis. Is stipulated. A plurality of stems 12 protrude at intervals on the inner surface of the conductive outer cylindrical portion 10a in the beam acceleration axis direction, and the stems 12 are arranged along a straight line. In this case, each stem 12
Are arranged such that the interval between the stems 12 gradually increases from the incident side of the ion beam toward the exit side.

[0007] Electrodes 13 called drift tubes are attached to the stems 12, respectively. In each electrode 13, a quadrupole magnet is provided so as to surround the beam acceleration axis. When high-frequency power of a predetermined frequency is introduced into the cavity formed by the conductive outer cylinder portion 10a, the stem 12, and the electrodes, a high-frequency voltage is applied between adjacent electrodes, and charged particles on the beam acceleration axis are applied. Is accelerated. In this Alvaret-type linear accelerator, as shown in FIG. 4, a region sandwiched between adjacent electrode center positions is called a unit cell.

[0008] As shown in FIG.
As in FIGS. 1 and 3, a conductive outer cylinder portion 10b having both ends closed by a plate is provided, and a beam passage hole is provided in the center of the plate at both ends. Conductive outer cylinder 10
b, a pair of conductive metal plates 15a and 15b project in the beam acceleration axis direction from positions facing each other,
The stems 16a and 16b are alternately attached to the conductive metal plates 15a and 15b, respectively. Further, these stems 16a and 16b are electrically connected to electrodes (drift tubes) 17a and 17b each containing a four-pole magnet. These electrodes 17a and 17
The length b and the electrode gap are configured to gradually increase toward the emission side. These conductive outer cylinder 10b, electrodes 17a, 17b, conductive metal plate 15a,
When high-frequency power of a predetermined frequency is introduced into the cavity formed by 15b and stems 16a and 16b, a high-frequency voltage is applied between adjacent electrodes, and the ion beam on the beam acceleration axis is sequentially accelerated. In this accelerator as well, as shown in FIG. 6, a region sandwiched between adjacent electrode center positions is called a unit cell.

A description of the structure of the Vidley-type linear accelerator is omitted, but the interconnection between the electrodes and the stem and the unit cell are the same as those of the IH-type linear accelerator.

Although these linear accelerators have the advantage of continuously accelerating an ion beam having a large value, the linear accelerator has a limitation that the energy of the output beam from the output side is constant. . For this reason, the linear accelerator is actually used only for extremely limited applications, such as a physical experiment apparatus that does not need to change the emission energy, or a synchrotron injector.

On the other hand, it has been considered to apply these linear accelerators to, for example, an ion implanter for implanting ions into a semiconductor. In such an ion implanter, it is essential that ions of different energies can be implanted into the semiconductor. As described above, since continuously changing the emission energy is indispensable for implanting ions into the semiconductor, various techniques for continuously changing the emission energy in the linear accelerator as described above have been proposed. ing.

For example, a Vidley-type linear accelerator capable of changing the output energy by making the resonance frequency variable has been studied and put into practical use at RIKEN. Similar attempts have been made in the RFQ linear accelerator.

[0013]

However, these methods for changing the resonance frequency require the provision of a circuit for widely changing the resonance frequency of the linear accelerator and the frequency of the high-frequency power source. The configuration of the linac is complicated. Therefore, it is practically difficult to commercialize the linear accelerator as a general-purpose device applicable to an ion implanter or the like by this method.

Further, in the RFQ type linear accelerator, 4
It has also been proposed to change the output energy in steps by changing the high-frequency voltage applied to the wings. However, as such, the conventional RF
In the case of the Q type, when the high-frequency voltage is reduced, the output energy is greatly reduced in a step-like manner. Therefore, not only is it impossible to continuously vary the output energy, but also the changing output energy cannot be arbitrarily selected. There is.

On the other hand, when the high-frequency voltage is changed by the conventional drift tube type linear accelerator, the charged particles are either accelerated to a predetermined energy or are emitted without incident acceleration without incident energy. Is. Hereinafter, the change of the acceleration characteristic when the high frequency voltage is changed by the conventional drift tube type linear accelerator will be described in detail.

In a high-frequency linear accelerator, a synchronization phase is known as a representative parameter representing acceleration characteristics. Synchronous phase refers to particles that are completely synchronized with the high-frequency phase, in other words, among charged particles that are being accelerated, the velocity of each cell over all cells matches the high-frequency phase velocity in that cell. This is the high frequency phase when particles (referred to as synchronous particles) pass through the electrode gap. From the viewpoint of phase stability described later, in order to accelerate charged particles, the synchronous phase Φs must be set in a range between -90 degrees and 0 degrees. In the conventional drift tube type linear accelerator, the electrode length, the inter-electrode length, the high-frequency voltage, and the like are defined so that the synchronous phase Φs has a constant value over all cells within a range of 25 to 30 degrees. . When a high frequency of a predetermined frequency and a predetermined power is introduced into the drift tube type linear accelerator configured as described above, a high frequency voltage Vo (n) is applied to the electrode gap of each cell. The energy gain ΔWs (n) when the synchronous particle passes through the electrode gap of the n-th cell is given by Expression 1.

[0017]

(Equation 1)

In equation (1), T is the transit time factor, e is the number of charges of the charged particles, and Vo is the high frequency voltage between the electrodes.

Here, the behavior when charged particles positioned before and after the synchronous particle gradually pass through the electrode gap will be described with reference to FIG. The high-frequency phase when the particle A located slightly ahead of the synchronous particle passes through the electrode gap of the n-th cell is represented by Φ
_{Assuming A} (n), Equation 2 holds.

[0020]

(Equation 2)

Therefore, the energy gain ΔW of the particle A
_A (n) is given by Equation 3.

[0022]

(Equation 3)

As is apparent from Equation 3, since the energy gain ΔW _A (n) of the particle A in the n-th cell is smaller than the energy gain ΔWs (n) of the synchronous particle, the particle A repeats acceleration in each cell. As it approaches the synchronous particle.

Since the speed of the particle A is lower than that of the synchronous particle, the particle A will eventually be located behind the synchronous particle.

On the other hand, assuming that the high-frequency phase when the particle B located slightly behind the synchronous particle passes through the electrode gap of the n-th cell is Φ _B (n), the following equation 4 holds.

[0026]

(Equation 4)

Therefore, the energy gain ΔW _{B of the} particle _B
(n) is represented by the following equation (5).

[0028]

(Equation 5)

As is apparent from Equation 5, since the energy gain ΔW _B (n) of the particle B in the n-th cell is larger than the energy gain of the synchronous particle, the particle B becomes a synchronous particle as the acceleration in each cell is repeated. approach. Since the speed of the particle B is higher than that of the synchronous particle, the particle B will be located in front of the synchronous particle.

The particles located before and after the synchronous particle are thus accelerated to an energy which is on average equal to that of the synchronous particle while oscillating around the synchronous phase Φs. Such phase oscillation accompanied by energy oscillation is called synchrotron oscillation, and such acceleration characteristics are called phase stability. Further, a stable region of synchrotron oscillation expressed in a plane having an energy difference and a phase difference from the synchronous particle as coordinate axes is called a phase stable region.
Changes as shown in FIG. That is, when Φs <0, the phase stability region becomes smaller as Φs approaches 0.

On the other hand, when Φs ≧ 0, the energy gain of the particle located behind the synchronous particle is always smaller than the energy gain of the synchronous particle, so that the phase stable region does not exist and the charged particle beam is accelerated. That is impossible.

As described above, the conventional drift tube type linear accelerator is configured such that the synchronous phase Φs is constant over all cells. Here, attention is paid to a change in acceleration characteristics when the introduced high-frequency power is changed from a predetermined value. When the high-frequency power is changed from the predetermined value P ₀ to P ₁ and the high-frequency voltage applied between the electrodes of the n-th cell changes from the predetermined value V ₀ (n) to V ₁ (n),
₁ (n) is given by the following equation (6).

[0033]

(Equation 6)

As is apparent from Equation 6, the voltage between the electrodes drops at a constant rate in all the cells.

At this time, the synchronization phase of each cell is set to a predetermined value Φ.
When becomes [Phi _s1 from _s0, [Phi _s1 from the condition that the energy gain ΔWs synchronization particles in the n-th cell (n) must be constant regardless of the high-frequency voltage, varies according to the following equation (7).

[0036]

(Equation 7)

That is, when the high frequency voltage is increased, the phase stable region increases, and when the high frequency voltage is decreased, the phase stable region decreases.

Now, let us consider a case where high-frequency power such that V ₁ (n) = V ₀ (n) cos Φ _s0 is introduced. From Equation 7,
Since Φ _s1 = 0 ° and phase stability is lost in all cells, the charged particle beam does not accelerate and reaches the emission side with incident energy.

Similarly, when V ₁ (n) <V ₀ (n) cos Φ _s0 , the phase stability is lost in all cells, and the charged particle beam is not accelerated. Therefore, the high-frequency voltage V ₁ (n) is set to a predetermined value V by the conventional drift tube type linear accelerator.
When the acceleration characteristic is changed from ₀ (n), V ₁ (n)> V
_{In the case of 0} (n) cos Φ _s0 , acceleration can be performed up to a predetermined energy in the same manner as when a predetermined high-frequency voltage is applied.
_{When 1} (n) ≤ V ₀ (n) cosΦ _s0 , it is divided into two types: it is not accelerated and is emitted with incident energy.
The emitted energy cannot be changed.

An object of the present invention is to change the output energy of an ion beam to a continuous or discontinuous value very easily, and to be applicable to various linear accelerators using the principle of phase stability. It is an object of the present invention to provide a method for controlling the output energy of a linear accelerator.

[0041]

According to the present invention, there is provided an entrance side, an exit side, and a plurality of cells disposed between the entrance side and the exit side, and charged particles incident from the entrance side. In a high-frequency linear accelerator that sequentially accelerates a beam based on the principle of phase stability by a high-frequency voltage applied to the electrodes of each cell, the synchronization phase of each cell is determined by the length of each cell and the high-frequency voltage between the electrodes Is changed from the incident side to the emission side, and by changing the high-frequency voltage, a method of controlling the emission energy of the high-frequency linear accelerator that makes the emission energy of the charged particles variable is obtained.

[0042]

According to the present invention, the cell length and the high-frequency voltage are defined in advance so that the synchronization phase in a plurality of cells arranged from the incident side to the exit side of the charged particle beam is sequentially changed from the entrance side to the exit side. By changing the amount of high-frequency power to be introduced, the output energy can be varied continuously or discontinuously.

[0043]

EXAMPLES 9 is a diagram showing a first embodiment of the drift tube linac of the present invention, it is introduced a predetermined high frequency power P ₀ in the linear accelerator, a predetermined high frequency voltage between the electrodes of each cell The dependency of the synchronization phase Φ _s0 (n) and the beam energy W (n) on the number of cells n in the n-th cell when V ₀ (n) is applied is indicated by solid lines.

As is apparent from the figure, in this embodiment of the present invention, the synchronization phase, which is conventionally constant regardless of the number of cells, is changed so as to continuously increase with the number of cells. Specifically, in the present invention, the cell length, the high-frequency voltage between the electrodes, and the like are defined in advance so that the synchronization phase has such dependence on the number of cells. With such a configuration, when a predetermined high-frequency power P ₀ is introduced and a predetermined high-frequency voltage V ₀ (n) is applied between the electrodes of each cell, phase stability is obtained over all cells, and charging is performed. particle beam is accelerated in the extraction energy W _out from the incident energy W _in, and emitted.

Next, a description will be given of an operation method in the case where the beam is emitted with the beam energy W (n ') in the n'th cell with reference to FIG. Now, when the high frequency power to be introduced is changed from a predetermined value P ₀ to P ₁ , the high frequency voltage applied between the electrodes of each cell changes from a predetermined value V ₀ (n ′) to V ₁ (n ′). And change to At this time, focusing on the change in phase stability in the n'th cell, the synchronous phase Φ _s1 (n ') changes according to Equation 8.

[0046]

(Equation 8)

Therefore, as the introduced high-frequency power decreases, the synchronization phase φ _s1 (n ′) gradually increases, and the phase stable region decreases.

Further, the high frequency power P ₁ is calculated as P ₁ = P ₀ (cos
Φ _s0 (n ')) When reduced to the value given by ² , cos Φ
_s1 (n ') = 0, and the phase stability in the n'th cell is lost. At this time, the synchronization phase over all cells shows cell dependence as shown by the broken line in FIG. That is, between the n'th cell from the incident side, the synchronization phase increases to 0 with an increase in the number of cells, and phase stability is obtained in this region.

On the other hand, the region on the emission side from the n'th cell has no synchronous phase and has no phase stability. Therefore, the charged particle beam incident from the incident side is accelerated in the phase stable region of the n'th cell from the incident side, and n '
The beam energy W (n ') in the cell is reached. on the other hand,
Thereafter, in a region without phase stability, the charged particle beam is not accelerated, and the beam energy W in the n'th cell is not increased.
As (n '), the light reaches the emission side and is emitted.

According to the above principle, a drift tube type linear structure is constructed such that the synchronization phase when a predetermined high-frequency power is introduced increases as the number of cells increases from the input side to the output side. In the accelerator, by specifying the high-frequency power to be introduced to an appropriate value, it is possible to make a phase stable region from the incident side to any cell, and to make this arbitrary cell transition a region without phase stability, A charged particle beam having beam energy in this arbitrary cell is emitted from the emission side.

FIG. 10 is a view showing a second embodiment of the drift tube type linear accelerator of the present invention, in which a predetermined high frequency power P ₀ is introduced into the linear accelerator, and a predetermined high frequency voltage V is applied between the electrodes of each cell. _This shows the dependence of the synchronization phase Φ _s and the beam energy W (n) on the number n of cells in each cell when ₀ (n) is applied. As can be seen, in the embodiment of the present invention, in the region of the n _1-th cell from the incident side of the synchronous phase [Phi _s, at a constant value [Phi _{s1, 2-th} to n (n ₁ +1) th cell In the region of the cell _No. , a constant value Φ _s2 , and in the region from the (n ₂ +1) -th cell to the emission side, a constant value Φ _s3. It has been changed to increase. Specifically, in this embodiment of the present invention, as the number of cells increases,
The length of each cell, the high-frequency voltage, and the like are defined in advance so that the synchronization phase takes a plurality of values and increases discontinuously.

Considering the change in the synchronous phase and phase stability when the introduced high-frequency power is changed in the linear accelerator configured as described above, as in the case of the first embodiment, the acceleration characteristics It can easily be inferred that the following changes.

[0053] When the high-frequency power to be introduced is greater than P ₀ cos ² [Phi _s3 from the predetermined value P ₀ or less, all the cells on the output side from the incident side becomes a phase stable region, the charged particle beam of the beam energy W _{ou t} Is emitted.

When the high frequency power to be introduced is equal to or less than P ₀ cos ² Φ _s3 and larger than P ₀ cos ² Φ _s2 , n ₂
The region up to the third cell is a phase stable region, and a charged particle beam having a beam energy W (n ₂ ) is emitted.

When the introduced high-frequency power is equal to or less than P ₀ cos ² Φ _s2 and larger than P ₀ cos ² Φ _s1 , the phase from the incident side to the n _1st cell becomes a phase stable region, and the beam energy W (n ₁ ) Is emitted.

As described above, the configuration is such that the synchronization phase when a predetermined high-frequency power is introduced increases to a plurality of discontinuous values as the number of cells increases from the input side to the output side. In the drift tube type linear accelerator, by specifying the high-frequency power to be introduced to an appropriate value, from the incident side to any cell forming a discontinuous point of the synchronous phase is set as a phase stable region, A region without phase stability can be set, and as a result, a charged particle beam having a beam energy in this arbitrary cell can be emitted from the emission side.

FIG. 11 is a diagram showing an embodiment in which the present invention is applied to a cell area called an accelerator of the RFQ type linear accelerator of the present invention. A predetermined high-frequency power P ₀ is introduced into the linear accelerator, and The dependency of the synchronization phase Φ _s0 (n) and the beam energy W (n) on the number n of cells in each cell when a predetermined high-frequency voltage V ₀ is applied between the conductive wings is shown by a solid line. .

In the conventional RFQ linear accelerator, the dependency of the synchronization phase Φ _s0 (n) on the number n of cells when a predetermined high-frequency power is introduced, as shown in FIG. To efficiently capture and bunch the incident charged particle beam, increase it slowly from -90 degrees,
It is configured to be constant in a cell region called an accelerator that accelerates charged particles in earnest. In the example shown in FIG. 12, the accelerator area is the cell after the n _A- th cell, and the synchronization phase here is constant at Φ _s0 (n _A ).

For this reason, in the conventional RFQ type linear accelerator, when the high-frequency power to be introduced is P ₀ cos ² Φ _s0 (n _A ), the synchronous phase Φ is set over all cells constituting the accelerator.
_{Since s} (n ') = 0 degrees, when a high-frequency power larger than P ₀ cos ² Φ _s0 (n _A ) is introduced, all cells from the entrance side to the exit side become a phase stable region and have a predetermined energy W _out
Is emitted, but P ₀ cos ² Φ
_s0 in the case of introducing (n _A) following high frequency power, there is no phase stability Te all cells Akuserereita region, emission energy of the charged particle beam is discontinuously greatly reduced below W (n _A) Therefore, a charged particle beam having an arbitrary energy cannot be obtained.

On the other hand, in the embodiment of the present invention shown in FIG.
Since the synchronous phase with the increase in the number of cells in Akuserereita region is changing to increase, charging with any beam energy W _out from W _in the same manner as the first embodiment for the drift tube linac The particle beam can be emitted from the emission side. That is, the introduced high-frequency power is changed from a predetermined value P ₀ to P ₀ cos ² Φ _s0 (n ′)
To the synchronization phase Φ in the n′th cell.
At the same time as _s1 (n ') = 0 degrees, the synchronization phase _.PHI.s1 (n) in each cell from the incident side to the exit side changes as shown by the broken line in FIG. At this time, the n'th cell from the incident side is a phase-stable region, and the cells after the n '+ 1th cell are regions without phase stability. The charged particle beam incident from the incident side is accelerated in the phase stable region of the n'th cell from the incident side,
The beam energy W (n ') in the n'th cell is reached. On the other hand, in the region after this, the charged particle beam is not accelerated,
The light beam reaches the light exit side with the beam energy W (n ') in the n'th cell and is emitted.

Therefore, in the RFQ type linear accelerator having the accelerator region configured so that the synchronization phase when a predetermined high-frequency power is introduced increases with the number of cells, the high-frequency power to be introduced is set to an appropriate value. Stipulates that the phase from the entrance side to any cell can be a phase-stable region, and at the same time, any cell after this cell can be a region without phase stability. A particle beam is emitted from the emission side.

[0062]

As described above, according to the present invention, the high-frequency linear accelerator configured so that the synchronization phase increases continuously or discontinuously as the number of cells increases from the entrance side to the exit side. Thus, the output energy can be easily changed over a wide range by changing the amount of high-frequency power to be introduced.

[Brief description of the drawings]

FIG. 1 is a partially cutaway view for explaining an example of an RFQ linac to which the present invention can be applied.

FIG. 2 is a diagram showing the vicinity of a conductive wing tip in more detail for explaining a unit cell of the RFQ linac shown in FIG. 1;

FIG. 3 is a partially cutaway view for explaining an example of an Alvaret-type linear accelerator to which the present invention can be applied.

FIG. 4 is a view for explaining a unit cell of the Alvaret-type linac shown in FIG. 3;

FIG. 5 is a diagram for explaining an example of an IH type linear accelerator to which the present invention can be applied.

FIG. 6 is a diagram for explaining unit cells of the IH type and the Videley type linear accelerator.

FIG. 7 is a diagram for explaining the principle of phase stability.

FIG. 8 is a diagram illustrating a change in a phase stable region with respect to a value of a synchronization phase.

FIG. 9 is a diagram for explaining a case where the present invention is applied to a drift tube type linear accelerator.

FIG. 10 is a diagram for explaining another case in which the present invention is applied to a drift tube type linear accelerator.

FIG. 11 is a diagram for explaining an operation when the present invention is applied to an accelerator area of an RFQ linear accelerator.

FIG. 12 is a diagram for explaining acceleration characteristics of a conventional RFQ linear accelerator.

[Explanation of symbols]

 10, 10a, 10b Conductive outer cylinder part 11a to 11d Wing part 12, 16a, 16b Stem 13, 17a, 17b Electrode

Claims (3)

(57) [Claims] An input side, an output side, and a plurality of cells disposed between the input side and the output side, wherein a charged particle beam incident from the incident side is applied to a high-frequency voltage applied to an electrode. Based on the principle of phase stability, in a beam energy control method of a high-frequency linear accelerator that sequentially accelerates, while changing the synchronization phase of each cell determined by the length and the high-frequency voltage of each cell from the entrance side to the exit side,
A beam energy control method for a high-frequency linear accelerator, characterized in that the emission energy of the charged particles is varied by changing the high-frequency voltage. 2. A method according to claim 1, wherein said synchronous phase and said high-frequency power are continuously changed. 3. The beam energy control method for a high frequency linear accelerator according to claim 1, wherein said synchronous phase and said high frequency power are changed discontinuously.