WO2019043070A2

WO2019043070A2 - Linear accelerating structure for protons

Info

Publication number: WO2019043070A2
Application number: PCT/EP2018/073259
Authority: WO
Inventors: Steven Hunt
Original assignee: Alceli Limited
Priority date: 2017-08-29
Filing date: 2018-08-29
Publication date: 2019-03-07
Also published as: GB201716879D0; GB2566118A; WO2019043070A3; EP3845038A2; GB201713889D0; GB2566118B

Abstract

A linear particle accelerating structure is provided for accelerating protons using radio frequency, RF, radiation, the structure comprising at least two coupled resonant cavity units. Each resonant cavity unit comprising a corresponding resonant cavity with an input port and an output port for a beam of protons; and an RF input to receive an RF signal; an RF phase shifter to control the phase of the RF signal. The particle accelerating structure further comprises at least one RF phase shifter to adjust, in one or more of the respective cavities, a phase of the RF signal to expose the protons to an accelerating portion of the electric field for a first duration and to expose the protons to a decelerating portion of the electric field for a second duration, wherein the first duration is greater than the second duration and greater than zero and wherein the second duration is greater than or equal to zero.

Description

Linear Accelerating Structure For Protons

FIELD OF THE INVENTION

The present invention relates to an apparatus, method and computer program for a linear accelerating structure for accelerating protons and a particle accelerator comprising at least one of the linear accelerating structures.

BACKGROUND OF THE INVENTION

Particle acceleration is used in many applications, including High Energy and Nuclear Physics, Neutron Scattering, Synchrotron light sources, Isotope production and Medical Accelerators. At very low energy, high Direct Current (DC) voltages can be used to provide a voltage difference between metal plates that can directly accelerate charged particles such as protons directed between the metal plates. Using a DC voltage may be practical up to energies of a few keV, but becomes inefficient at higher energies. To generate higher energies of several MeV, an alternating electromagnetic field at high frequencies may be used for particle acceleration, for example using operating frequencies in a radio range of the electromagnetic spectrum. A variety of different configurations of radio frequency (RP) structures may be used. The alternating field particle accelerating systems may be referred to as RF systems regardless of their actual operating frequency.

Cyclic accelerators such as cyclotrons and synchrotrons guide charged particles in a circular path multiple times through the same electric fields. Cyclotrons are relatively small circular accelerators where the orbit of the particles increases as the energy increases. Synchrotrons are larger than cyclotrons for producing the same particle energy. In a synchrotron, an orbit of a particle may be fixed by increasing the magnetic field in bending magnets which deflect the particles. During each orbit, the particle energy is increased by RP amplification. By way of contrast, Linear Accelerators have RF accelerating structures arranged in a line to direct charged particles through a sequence of accelerating fields, which eliminates the need for bending magnets to keep the particles in a circular orbit. High efficiency, high power radio frequency amplifiers such as klystrons may be used to generate radio frequencies. Klystrons are normally operated continuously in DC mode - even when a beam is not present, they may be fully powered - and as a result, energy may be wasted. Absorption of the energy by the accelerator cavity may result in excess heat being generated. This excess heat, coupled with the high power used by the klystron means water cooling may be needed to cool the system down and to counteract the heating effect. Water cooling of the linear accelerator may increase the complexity in the design, increase running costs and reduce reliability of the system.

As a result of the above discussed examples, there is a need for an efficient linear accelerator which uses less power and generates less excess heat.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention are further described hereinafter with reference to the accompanying drawings, in which:

Figure 1 schematically illustrates a cross-sectional view of a linear accelerating structure for accelerating protons, including two coupled resonant cavities;

Figure 2 schematically illustrates a block diagram of the linear accelerating structure for accelerating protons, including two coupled resonant cavities in which an application of the radio frequency (RF) radiation to the cavities is illustrated in more detail;

Figure 3 schematically illustrates a flow chart of an example method performed to accelerate protons in the linear accelerating structure of Figure 1 and Figure 2 using RF radiation; and

Figure 4A schematically illustrates a variation of an on-axis electric field strength in Mega Volts (MV) against distance in metres along the accelerator axis, as would be seen by a proton travelling through an accelerator structure comprising three identical cavities;

Figure 4B schematically illustrates a variation of an on-axis electric field strength in Mega Volts (MV) against distance in metres along the accelerator axis, as would be seen by a proton travelling through a previously known drift tube linear (DTL) accelerator structure;

Figure 4C is a graph which schematically illustrates a variation of an on-axis electric field strength in Mega Volts (MV) against distance in metres along the accelerator axis, as would be seen by a proton travelling through an accelerator structure according to an example embodiment comprising three identical resonant cavities of the accelerator structure;

Figure 4D schematically illustrates an accelerator structure comprising three resonant structures according to an example embodiment;

Figure 4E schematically illustrates a variation of electric field as seen by the hadron in a three resonant cavity accelerator structure of an embodiment showing curves for four different phase alignments of the hadron implemented by phase shifting;

Figure 5 schematically illustrates an example timing diagram to indicate when each of, the beam of protons and the hadron-accelerating RF radiation is on for accelerating protons in a low duty cycle;

Figure 6 schematically illustrates a block diagram of the particle accelerator including the linear accelerating structure of Figure 1 and Figure 2; and

Figure 7 schematically illustrates an example loop coupler for coupling RF radiation to a resonant cavity

DETAILED DESCRIPTION OF THE INVENTION

Examples relate to a linear accelerating structure which comprises at least two coupled resonant cavities each for accelerating particles for less than one RF cycle. Warm linear accelerators commonly use a single tank accelerator, which may accelerate particles over multiple RF cycles. Such accelerators are not superconducting and generally kept above room temperature.

Although the present technique pertains to the acceleration of protons, the technique may be used to accelerate any charged particle such as hadrons or carbon ions and may not be limited to acceleration of protons. The invention defined herein and embodiments thereof are defined in relation to acceleration of protons. However, in the alternative, the technique may in the alternative be applied to any charged particle, such as hadrons or carbon ions, preferably hadrons and as such the invention may, in the alternative, be directed to hadrons or carbon ions (e.g. by substituting the term 'proton' with 'hadron' in definition and description of the invention herein, including in the claims).

Furthermore, embodiments described herein may, where the context allows, refer instead to hadrons (or carbon ions).

Particles may be accelerated in either standing wave or travelling wave accelerating structures. Proton accelerators are often standing wave accelerating structures, in which the walls of the cavity along the direction of travel of the protons form nodes of the standing wave. To accelerate particles, an electric field may be arranged along a direction of beam propagation. If the electric field is driven by a periodically varying voltage such as a sinusoidal voltage or a Gaussian voltage or some other waveform, the polarity of the electric field serves to accelerate charged particles only half of the time, say for 180° of a 360° cycle, and for the remaining time the field (voltage) decelerates the charged particles of the beam.

Particles are often accelerated in "bunches" consisting of collections of particles with spaces in between the bursts. A bunch may refer to a collection of particles under the influence of one RF cycle. Each burst of particles that is injected into an RF structure may have finite (non-zero) duration and a finite (non-zero) spread of energies around a target energy. The presence of a bunch of particles in an accelerating cavity can be timed so that it is "synchronous" with an accelerating portion of the electric field.

Some phases of the sinusoidal electric field are better than others for achieving stable particle acceleration. For example, a phase of 0° corresponding to a peak positive electric field (voltage) when the bunch is in the centre of the cavity may provide maximum acceleration but may not be a stable point for acceleration of particle bunches, whereas a phase of a number of degrees (e.g. 10°, 20° or 30° below a maximum), corresponding to a rising portion of the sinusoidal curve, close to the peak may provide a more stable point for acceleration. Thus the presence of a particle bunch in the cavity may be aligned with, for example, a so called "synchronous phase" < _S, such as upon entry to the cavity to efficiently accelerate the bunch of charged particles through the cavity.

The transit time of a charged particle through a previously known tank type linear accelerator systems such as a drift tube linear accelerator (DTL) would be several RF wavelengths long. In such known systems the bunches are shielded from the decelerating portion of the electric field by shielding tubes called drift tubes. Particle acceleration occurs only in the gaps between the drift tubes, to coincide with the accelerating RF field portion. During one half of an alternating electric (voltage) field cycle, particles in gaps between the drift tubes experience an accelerating field, whereas during the remaining half of the alternating field cycle when the field is in the opposite direction and would act to decelerate particles, the particles pass through the drift tubes where they are shielded from the decelerating electric field.

As the particles travel along a linear accelerator of the known type, the drift tubes and gaps increase in length to directly compensate for the increase in speed because the particles should spend the same amount of time passing through successive gaps and the same amount of time in successive tubes, the time being dependent on the duration for which the electric field is an accelerating and then decelerating field, which in turn depends on the RF frequency. If the particles are injected into the accelerating region to coincide with a stable point in the field for particle acceleration,, then an effect of "phase stability" can be exploited, where if a particle in the bunch arrives too soon in the accelerating gap as the electric field rises, it will not experience as a high a field as it should and will receive less of a velocity boost than the particle in the centre of the bunch arriving later, but a particle that arrives late , when the field magnitude is higher is accelerated more giving a net effect of phase stability where the charged particles are kept in phase with the field in each accelerating region.

The RF radiation in each resonant cavity may be independently controllable by an RF phase shifter and RF amplifier to allow RF radiation with a particle accelerating amplitude to be radiated within each cavity when the beam of protons is present. In this way, each bunch of protons within the beam is arranged to be synchronous with the phase of the accelerating electric field (or equivalently the sinusoidal voltage). A beam comprising "bunches" of protons may travel through the resonant cavities. The bunches may be formed via the action of the RF radiation on particles injected into the structure. As a bunch of protons passes through each cavity, the protons may be subject to the accelerating, but not the decelerating field of the RF radiation. The accelerating field (but not the decelerating field) may have a field amplitude corresponding to an accelerating voltage amplitude. The protons within the cavity may be subjected to an overall accelerating force in their direction of travel as a result of the accelerating component of the RF radiation in the resonant cavity and the RF radiation will thus transfer energy to the protons accelerating them along the cavities.

The independent control of the phase of the RF radiation in each cavity allows the cavities to be placed close together because the phase of the RF radiation in each cavity can be controlled as the particles propagate and accelerate along the structure comprising a plurality of contiguous resonant cavities such that the protons arriving in each successive cavity are synchronous with the RF radiation applied to the respective cavity. Without this control of phase, the cavities would have to be spaced further apart, allowing for the so-called 'drift space', to ensure by the time the particle reaches the next cavity it again sees an accelerating but not a decelerating field.

The power needed by a cavity scales with the square of the RF field needed. For example, by halving the RF field per cavity only a quarter of the power is needed. Although this may double the number of cavities needed to achieve a given RF field (given voltage) and may double the length of the accelerator, the total power needed may be halved. Thus, by increasing the number of cavities, the power per cavity can be reduced which simplifies the engineering of a cavity and the cost of amplification.

According to the present technique, less power is required for each resonant cavity relative to that required by a single tank linear accelerator for the same acceleration and as a result individual lower power amplifiers can be used for each resonant cavity in the structure. The lower power per resonant cavity may allow a solid-state amplifier to be used for each resonant cavity as opposed to a single klystron that may be used for the higher power in a previously known single tank accelerator such as a drift tube linear accelerator (DTL). The drift tubes in a DTL, which shield the particles from the decelerating field, typically require water cooling which increases the cost and complexity of the design. Also, individual lower power solid-state amplifiers per cavity typically generate less excess heat than klystrons. The design of the linear accelerating structure of the embodiments in which successive stages of hadron acceleration are provided by phase control in successive accelerating chambers and the use of solid-state amplifiers allow the linear accelerating structure to be air cooled. Without water cooling, the complexity of the design and the cost of both the manufacturing and maintenance may be reduced.

Figure 1 is a cross-sectional view of an example of a linear accelerating structure 100 for use in a particle accelerator for accelerating protons. The linear accelerating structure 100 comprises a first resonant cavity unit 130a and a second resonant cavity unit 130b.

The first resonant cavity unit 130a comprises the first resonant cavity 102a with a first hadron input port 114a and a first hadron output port 116a, a first RF phase shifter 104a, a first RF solid state amplifier 106a and a first radiation coupler 108a.

The second resonant cavity unit 130b comprises the second resonant cavity 102b with a second hadron input port 114b and a second hadron output port 116b, a second RF phase shifter 104b, a second RF solid state amplifier 106b and a second radiation coupler 108b.

For the purposes of this example there are two resonant cavity units 130a; 130b, however the linear accelerating structure may consist of any number of resonant cavity units each with their own resonant cavity having a respective input port and a output port for the protons, a RF phase shifter, a RF solid state amplifier and a radiation coupler. In alternative embodiments there may not be a one to one relationship between phase shifters and resonant cavities or between RF amplifiers and resonant cavities. For example, one RF amplifier or one RF phase shifter may be shared between two or more cavities. The word "cavity" is used herein to describe a container and the space encompassed by it, unless the context clearly indicates otherwise. In other examples, a hierarchy of phase shifters may be provided to control phases of RF signals entering the resonant cavity units. In an arrangement having two resonant cavities cavity is only one of the resonant cavities may use an RF phase shifter, and yet the phase of RF radiation could be adjusted in a second resonant cavity based on a given RF phase in a first resonant cavity and using a determined hadron trajectory and timing. Embodiments allow for a phase adjustment of RF radiation on one cavity relative to a phase in another cavity as seen by a hadron on its trajectory through the two cavities such that the charged hadron is exposed to an accelerating portion of the electric field for longer than it is exposed to a decelerating portion of the electric field in its path through the accelerating structure, without any shielding (such as by a drift tube) of the hadron from a decelerating electric field in the resonant cavity. In some embodiments the protons may not be exposed (or "may not see") to a decelerating field. In such embodiments the protons have a zero duration exposure to a decelerating portion of the electric field. Individual cavity units within the accelerating structure allow for ease of manufacture and maintenance, as each resonant cavity may be identical and may be suitable for mass production in a factory separately before assembling in a modular way on site. The multiple cavity units 130a; 130b may be coupled, for example by being detachably coupled, vacuum brazed or welded together.

Alternatively, the cavity units may not be identical and the individual cavity units allow for flexibility in the design of the overall accelerating structure.

A beam of protons 112 is input into the first hadron input port 114a and travels through the first resonant cavity 102a to the first hadron output port 116a. The beam of protons 112 may continue to propagate across a gap 118 between the first hadron output port 116a and the second hadron input port 114b, 102b and may be input to the second hadron input port 114b. The hadron beam 112 comprising successive particle bunches propagates through the second resonant cavity 102b and out of the second hadron output port 116b. The input and output ports may be located at any position on the resonant cavity and not necessarily in the central region of each cavity as shown in Figure 7, and the beam of protons may take any path from the input port through the resonant cavity, as directed by the accelerating RF radiation, before reaching the output port.

The first resonant cavity 102a may have different dimensions to the second resonant cavity 102b, or the first and second resonant cavities may have the same dimensions. The width of each resonant cavity 102a; 102b may be less than or equal to a distance travelled by a hadron in half a period of the RF radiation applied to the cavities through the radiation couplers 108a; 108b, the distance travelled depending on, for example, on an input energy of the hadron. The first resonant cavity 102a is separated from the second resonant cavity 102b by the gap 118 which may have a width much less than the distance travelled by the particle in half a period of the RF radiation. From an alternative perspective, in some embodiments, a distance between cavity centres is less than a distance travelled by a proton on one period of the RF radiation. The cavities may be separated by a distance to minimise or at least reduce the leakage of the RF radiation from one cavity to the next. The small gap 118 between the cavities allows the accelerating structure to have a more compact design even if the structure comprises multiple resonant cavity units.

The resonant cavities and couplers may be composed of materials not prone to radio activation, for example, copper such as high purity oxygen- free copper. Cyclotrons in contrast require large amounts of steel, which, combined with inefficient extraction inherent in their design, may cause the structure to become radio-active over time, posing problems for disposal at end of life. The cavities 102a, 102b may be considered a 'warm' design as they may not be superconducting. Warm cavities may be simpler to design, build, operate and maintain than superconducting cavities. Although superconducting cavities require less RF energy to produce the accelerating RF field, warm cavities may not require expenditure of energy associated with liquid helium for cooling. The resonant cavities may be designed to have high "shunt impedance" which may result in a lower RF power consumption. Shunt impedance Shunt impedance R = |V|² / P, where V is the voltage experienced by the particle (vector component along beam trajectory). The power dissipation is inversely proportional to the shunt impedance.

The protons may be accelerated by the RF radiation in the resonant cavities.

The RF radiation is sent from an RF source 120 which may be shared between the resonant cavities or optionally, each resonant cavity may have its own individual RF source. An example of an RF source may have a frequency of around 800MHz. However, any frequency of RF radiation can be used that is compatible to the dimensions of each cavity, to achieve resonance, i.e. a standing wave in the electric field within the cavity. For example, if the RF frequency used to achieve resonance in one cavity is to be doubled, the height of the cavity can be halved to achieve resonance. Each resonant cavity may have a designated RF phase shifter 104a; 104b and RF amplifier 106a; 106b to individually control the RF signal received from the RF source 120 for each resonant cavity 102a; 102b. The RF phase shifter 104a; 104b may be any component which can change the phase of an RF signal sent from the RF source. The RF phase shifters may be mass produced integrated circuits. The design may not require fast RF feedback on the phase or amplitude of the RF radiation. The individual phase control using the RF phase shifter may allow shielded drift spaces as present in a DTL between cavities to be eliminated.

The RF amplifier 106a; 106b may be a solid-state amplifier, such as a single or multiple transistor amplifier, that increases the amplitude of the RF signal. Each RF amplifier may amplify the RF signal for its respective resonant cavity, such that the amplitudes applied to each resonant cavity are independently controlled. The successive acceleration of proton bunches by the two or more resonant cavities may reduce the power required per cavity. For example, a power of only 600W per cavity can be used to achieve an average on-axis accelerating field of nearly lMV/m in each cavity. The RF amplifiers may be selectively turned on only when the beam is present in the cavity, however a klystron would require to be powered even when the beam is not present to maintain stability. RF power levels used to perform particle acceleration in a linear accelerator may not be sustainable continuously, Semiconductor amplifiers are quick to switch on and may be configured to provide power for only when the protons are present, which may be as low as only 1% of the time, for example, meaning the protons may only be accelerated for ten milliseconds every second. Thus improved power efficiency may be achieved relative to the use of a klystron. A "duty cycle" may be defined as a percentage of time during operation of the accelerator when the beam is on. For example, by using, for example, a 1% duty cycle when the amplifier provides 600W when the beam is present the average RF power may be reduced to 6 watts per cavity reducing heating in the amplifier and cavity, and reducing the power costs. The distributed amplification using the individual RF amplifiers may avoid high power transmission and allow simpler engineering for the accelerating structure. Thus the structure may be conveniently air cooled. Alternatively, a thermoelectric heat pump may be used to perform thermoelectric cooling.

Each amplified and phase-shifted RF signal may be coupled into each respective resonant cavity using a radiation coupler 108a; 108b. The radiation coupler may be a loop coupler, an RF port from a waveguide or any coupler capable of radiating the phase-shifted and amplified RF signal into the resonant cavity. The loop size, rotation angle and depth of the loop coupler may be determined based on the impedance of the cavity to optimise or at least increase the transmission and reduce reflection of the RF radiation. The RF radiation in at least one of the cavities can be monitored using one or more sensors 110a; 110b. The sensor may be any sensor for detecting RF radiation, for example a second radiation coupler may be used to monitor the field amplitude and phase.

Measurement feedback from at least one of the sensors may be used by at least one of the phase shifters to, in some embodiments, maintain synchronism with the incoming hadron bunches. However, according to the present technique, the phase adjustments relative to the hadron bunch timings mean that some embodiments do not maintain synchronism and yet still produce a net acceleration for hadron bunches. Each linear accelerating structure 100 may accelerate the protons with input energies of anywhere from 5MeV to 250MeV for protons, however the input energy is not limited to this energy range. The output energy of each structure may be determined by the RF power applied per cavity and the number of cavities in a structure. The output energy of beam of protons leaving the linear accelerator may be varied within the non-limiting range of, for example, 80Mev to 250MeV for protons by adjusting the RF power in each cavity as desired.

Quadrupole magnets may be used to adjust the focus of the beam and may be positioned at regular intervals between the structures. For example, one or more quadrupole may be placed between each, or with a spacing of more than one, structure. The quadrupoles may be any electro -magnetic quadrupoles which use electric current to excite the magnets to the required field, permanent magnet quadrupoles, or a combination of permanent and electro -magnetic quadrupoles. The beam 112 may be steered in at least one of the horizontal and vertical direction using corrector magnets. The corrector magnets may be placed anywhere along the linear accelerator.

Figure 2 is a block diagram to schematically illustrate a path of the RF signal in the linear accelerating structure 100, in contrast to Figure 1 which shows a path of the beam of protons. An RF signal is initially transmitted from an RF source, which may be an individual RF source for each resonant cavity unit or may be a shared RF source for the at least two resonant cavity units 230a; 230b. The RF signal is input into an RF phase shifter 204a; 204b and the phase of the RF radiation can be independently controlled and adjusted for each resonant cavity. The phase of the RF radiation is changed by the phase shifter to contain, at least predominantly, the accelerating portion (up to half the period) of the RF cycle of the RF radiation in the resonant cavity. The RF signal is then amplified by an RF amplifier 206a; 206b which may be a solid-state amplifier such as a single transistor amplifier. The amplification an phase shifting may be performed in a different order in some embodiments. The RF signal is coupled into the resonant cavities 202a; 202b using a radiation coupler and the RF signal can radiate in the resonant cavity to accelerate incoming protons. The RF radiation inside each resonant cavity can then be monitored using a sensor 210a; 210b which may be connected, detachably or otherwise, to the respective resonant cavity. The phase and amplitude of the RF signal may be changed according to the measurements (i.e. based on feedback) of the sensor using at least one of the RF phase shifter 204a; 204b and the RF amplifier 206a; 206b respectively. The phase may be changed to ensure that an at least predominantly accelerating portion of the RF cycle is contained within the resonant cavity during the time in which the bunch of protons are present in the resonant cavity, avoiding undesirable deceleration from the decelerating portion of the RF cycle. The amplifier may change an amplitude of the RF signal to ensure the RF radiation has an amplitude sufficient to accelerate the protons to a target velocity. The velocity of the protons should increase as the hadrons pass though successive resonant cavities to achieve a target terminal velocity on output from the final cavity. As in the embodiment of Figure 1, processing circuitry (not shown) may be provided to control a phases of RF radiation in different ones of the resonant cavities 202a, 202b such that the hadron sees predominantly a accelerating field in its beam path through the linear accelerator. The phase adjustments may be determined based on a calculation indicating when a charged particle will be present (e.g. at a predetermined point) in a resonant cavity and the phase adjustments may be implemented using one or more RF phase shifters. In some embodiments each resonant cavity has a corresponding RF phase shifter.

Figure 3 is a flow chart schematically illustrating a method 300 performed to accelerate protons in the linear accelerating structure 100. At element 302, an RF signal of a selected frequency (for example around 800MHz) is received from the RF source. At element 304, an RF phase shifter controls the phase of the RF signal. At element 306, the RF signal is amplified using an RF solid state amplifier. At element 308, the phase-shifted and amplified RF signal is radiated into a resonant cavity using a radiation coupler. At element 310, a beam of protons is passed through an input port of the resonant cavity and out through an output port of the resonant cavity. The beam of protons is accelerated by the at least predominantly accelerating portion of the RF radiation in the resonant cavity so that the beam is output at a higher velocity than a beam received via the hadron input port of the respective cavity. At element 312, the RF radiation in the resonant cavity is monitored using a sensor. The sensor measurements may determine if and how the phase of the signal should be changed to improve the acceleration of the protons in the resonant cavity to meet target hadron velocity or acceleration criteria. Using the feedback from the sensor, elements 304-312 can be repeated. The method 300 may be performed by circuitry. The circuitry may include machine-readable instructions which may be stored on a machine-readable storage medium. One or more processors may execute the instructions on the machine- readable storage medium to carry out the steps of method 300 as described above.

Figure 4A is a graph schematically illustrating a variation of an on- axis electric field strength in Mega Volts (MV) against distance in metres along the accelerator axis, as would be seen by a charged hadron travelling through an accelerator structure comprising three identical cavities where all cavities may have the same phase shift. The cavities of the accelerator are resonant cavities so the RF wavelength is selected to allow for resonance given the cavity dimensions. The electric field in the cavities corresponds mainly to a given mode of oscillation.

It can be seen from Figure 4A that a first half of the electric field waveform, which spans the first cavity, corresponds to a positive voltage whilst a second half of the electric field waveform, spanning the second cavity, has a negative voltage, In the third cavity, the positive half of the electric field waveform repeats. Note that the maximum field along the beam path through the centre of the cavity depends on the detailed design of the cavity and may approximate to a normal distribution, but the intensity of the field varies over time with a sinusodal function. Thus the field seen by the proton as it traverses the cavity in time, may be a combination of those functions. Figure 4A illustrates a varying electric field strength from a frame of reference of the hadron (proton) travelling through the accelerator structure.

Due to variations in the sign, magnitude and rate of change of the electric field created by standing waves in the resonant cavities, over a full RF period of the RF radiation, the charged hadron may see an accelerating force in one half of the RF period, but may see a decelerating force in a remaining half of the RF period. In the Figure 4A example, the electric field seen by the hadron as it travels along the axis of the first encountered cavity is an accelerating field but the electric field seen by the hadron as it travels through the second encountered cavity is a decelerating field. In the third encountered cavity, the electric field profile of the first encountered cavity is repeated, so the hadron is again accelerated by the field in the third encountered cavity. Figure 4A is just one non-limiting illustrative example.

For an even number of cavities a net effect of the electric field on a hadron passing through where the standing waves in each cavity have the same RF phase may be zero transfer of energy. However, for an odd number of cavities, such as the three illustrated in Figure 4A, the hadron experiences a net acceleration due to passing through two accelerating, but only one decelerating portion of the electric field. In the Figure 4A example the electric field profile corresponds to three adjacent identical cavities all at the same RF phase. The width of each cavity may correspond to the distance travelled by the protons at a particular energy in the time of 1/2 of the RF period.

Figure 4B is a graph schematically illustrating a variation of an on- axis electric field strength in Mega Volts (MV) against distance in metres along the accelerator axis, as would be seen by a charged hadron travelling through a previously known drift tube linear (DTL) accelerator structure having an accelerating gap for the first 4cm along the, axis, a drift tube shielding the hadron from the decelerating portion of the RF radiation in the region from 4cm to 8cm along the axis and a further accelerating gap in the axis region from 8cm to 12 cm. Similarly to the Figure 4A example, the phase of the RF signal in all three regions from 0cm through to 12cm is the same along the hadron trajectory through the structure. In the DTL, there is a single resonant cavity in contrast to the multiple resonant cavities according to example embodiments. The trajectory of the hadron along the axis in the DTL arrangement of Figure 4B is shielded from the electric field that would otherwise decelerate the hadrons by drift tubes. This has the effect that in the region between 4cm and 8cm along the axis, the hadron sees no electric field, either accelerating or decelerating.

Figure 4C is a graph which schematically illustrates a variation of an on-axis electric field strength in Mega Volts (MV) against distance in metres along the accelerator axis, as would be seen by a charged hadron travelling through an accelerator structure according to an example embodiment comprising three identical resonant cavities of the accelerator structure. A corresponding three cavity accelerator structure is shown in Figure 4D.

In the arrangement of Figure 4D, the RF radiation waveform 410 is the RF radiation as applied to each of the resonant cavities 402a-c by each respective radiation coupler as described in the previous examples. The RF radiation has a wavelength, RF. AS described in the previous examples the phase and the amplitude of the RF radiation in each resonant cavity may be separately controlled by a respective RF phase shifter and RF amplifier before entering the resonant cavity via the radiation coupler. However, for a given cavity width is determined by the objective of achieving resonance.

As previously described, the resonant cavities 402a-c in the linear accelerating structure 400 of Figure 4D may be separated by a gap 418 with a width of less than the distance travelled by an accelerated particle at a given energy (eg an input energy) in half a period of the RF radiation applied in the cavity. The period is fixed, but the distance thus specified may vary from cavity to cavity, similarly to drift tube length increasing along the beam trajectory. The width of each of the resonant cavities 402a-c themselves may also be less than the distance travelled by the particle in half a period of the RF radiation because a net hadron acceleration can be achieved by adjusting the RF phase. The RF radiation may be shifted in phase by the phase shifter to allow a positive part of the RF cycle to be contained within the resonant cavity at the relevant time, to coincide with the presence of a particle bunch in the cavity.

According to the present technique, a decelerating portion of the RF radiation is at least partly eliminated in the middle cavity (4cm through to 8cm) by controlling a phase of the RF signal on input to the second cavity so that an accelerating portion of the electric field coincides with an arrival of the charged hadron in the middle cavity.

In this example, the phase of the middle cavity is controlled so that when the hadron arrives in the middle of the central cavity (as measured along the hadron trajectory through the structure), the electric field is at an identical phase as the hadron experienced when it was in the middle of the first cavity. This may correspond to a synchronous phase. For example, a point at the peak of the positive electric field (denoted 0°) may be chosen as a field strength to coincide with the hadron being present in the middle of the cavity. In other examples, the phases in the cavities may be controlled such that the hadron sees a -10°, -20° or - 30° electric field, corresponding to a rising portion of the positive field. This is schematically illustrated in Figure 4E, which shows that when the phase is changed from zero degrees to -10° (curve 492) , -20°(curve 494) , -30° (curve 496), when the proton is the centre of the cavity for the phases other than 0 °, the electric field is lower (because it is before the peak) and therefore the curve corresponding to the electric filed strength seen by the proton moves to the right and the peak is later in position of the proton (also time) moving progressively through curves 492, 494, 496. Note that the -10°, -20°, -30° examples, the field is mostly positive although the field is slightly more negative with a bigger (more negative) phase change at the beginning of each cavity. In other examples, the phases in different cavities may differ with respect to each other.

By adjusting the phase of the RF radiation in the resonant cavity using a phase shifter according to example embodiments, the time at which an accelerating force is present in the cavity is adjusted. In other words, rather than the particles having to 'wait' in drift space for an accelerating field in a single tank resonant cavity which forms a DTL, the accelerating field is timed to be present in a cavity to coincide with passage of a bunch of hadrons through the cavity, deliberately eliminating at least some of the decelerating portion of the RF filed via the phase shift. DTLs may have multiple accelerating gaps and multiple drift tubes but a single RF phase throughout. Figure 5 schematically illustrates example timing of a beam of protons 502 through the linear accelerator cavities of Figure 1 relative to a timing of powering the RF radiation 504. In this example, the protons are protons, which are accelerated with a low duty cycle, although the timings are not shown to scale. In some examples the duty cycle may be as low as 1%. The RF source may apply RF radiation to the group of cavities at a time to coincide with the passage of the charged particle beam through the structure. The charged particles of the beam may comprise a sequence of bunches as described earlier. The electric field forms a standing wave (stationary wave) in each cavity, which is a time-varying field with nodes at the cavity walls. The fundamental frequency of the standing wave depends on the dimensions of the cavity. Although the illustrated standing wave portions in graph 410 have the same half wavelength, the RF radiation and associated electric field may be set to have different parameters (phase and amplitude) in each cavity, provided that resonance can be achieved in the respective cavity. Standing waves are characteristic patterns associated with resonance. The standing wave is set up by combination of RF waves moving in opposite directions along the closed chamber space and is a time-varying field although the wave is stationary at some nodes in space.

In contrast to previously known drift tube linear accelerators, where a single phase, amplitude and frequency of RF radiation is applied for the full trajectory of a particle bunch through the structure, according to the present technique, a phase of the electric field in each successive cavity may be set so that upon arrival in each successive cavity by a charged hadron bunch (for example, a proton bunch) , the bunch is synchronous with the electric field in the respective cavity. There is no need to accelerate the hadron bunch for a full 180° accelerating phase of the sinusoidal electric field waveform although this may be done, but instead acceleration may be applied to the beam for only a part of the accelerating phase.

Instead of providing drift tubes to shield the bunches of protons from the decelerating portion of the sinusoidal electric field waveform, the phase of the RF radiation in each cavity can simply be adjusted so that the accelerating portion of the field coincides with arrival of a bunch of hadrons in the cavity. Thus the phase of the RF radiation in each cavity is adjusted so that hadrons passing through the cavity experience predominantly an accelerating force from the RF field. Thus the phase shifting of the RF signals is performed to expose the protons to an accelerating portion of the electric field for a first duration and to expose the protons to a decelerating portion of the electric field for a second duration, wherein the first duration is greater than the second duration and the first duration is greater than zero. The second duration in some examples may be zero or close to zero. This means that the drift tube sections of the linear accelerator between successive accelerating "gaps" in previously known linear accelerators can be eliminated according to the present technique. This provides for a more compact design because the separation between adjacent cavities is not constrained by having to allow for drift tubes between successive accelerating regions to shield the hadrons from a decelerating portion of the electric field. Instead, according to

embodiments, a separation between adjacent accelerating cavities may be less than half of a distance covered by incoming accelerated particles (at the relevant incoming velocity) in a time corresponding to half an RF wavelength.

The timing of the beam of protons 502 entering the accelerating structure may be made to coincide with the time the RF radiation 504 is caused to be present by turning on the RF sources and optionally perform amplification of the RF signal to a given resonant cavity. This may allow the RF amplifiers and/or the RF sources to only be powered when the RF radiation is required to accelerate the protons and it may be turned off when there are no protons present in the structures. This may save power compared to a single resonant tank structure which typically requires the RF radiation to be continuously present in the cavity and/or the RF source to be constantly powered. In the example, for a less than 1% duty cycle, the duration of time 506 that the beam of protons is on may vary between Ιμβ and 10 ms. A 1% duty cycle for the proton beam means that the duration of time the beam is off 508 may be greater than or equal to 100 times the duration of time the beam is on 506. For example, if the beam is on for 1 then this suggests the beam may be off for a duration 508 of time greater than 100 μβ, although these numbers are non-limiting examples.

The duration of time the RF radiation is on 510 may coincide with the times when the proton beam is on, but the RF-on duration 510 may be longer than the duration of time that the proton beam is on 506 for each "beam on" period. The accelerating electric field builds up over time when the RF radiation is introduced into the cavities via the loop couplers 208a, 208b (see Figure 2), so switching the RF radiation on shortly before each period when the particle beam is on allows the RF field to reach optimum (or at least a good) strength and to be stable by the time the protons arrive. The RF field is switched off in Figure 5 only slightly after (e.g. a variable time period of between ΙΟμβ and 1ms after) the proton beam is present.

In some examples, the duration 510 of time the RF radiation is on may be any time larger than 10 μβ. The time for a proton to transit through several accelerating regions in a particle accelerator depends on the energy of the proton, which increases as the hadron is accelerated along the series of accelerating regions. The RF source 120 and/or RF amplifier 106a, 106b may be turned on for a duration of time 510 based on the duration of time in which the particle beam is turned on 506 so that the accelerating effect of the RF energy is experienced by particles of the beam.

Figure 6 schematically illustrates a linear particle accelerator 620, where the particle accelerator 620 may comprise one or more linear accelerating structures 600 (e.g. corresponding to the structure 100 in Figure 1) as described in the above examples and Figures. A linear particle accelerator 620 may further comprise a particle source 602, Low Energy Beam Transport (LEBT) 604, Radio Frequency Quadrupole (RFQ) 606, Medium Energy Beam Transport (MEBT) 608, High Energy Beam Transport (HEBT) 610 and beam delivery system 612.

The particle source 602 may provide a source of protons, for example, a proton source. The beam of protons 616 may be output at an energy of, for example, between 25keV and lOOkeV. A hadron source may have an input of gas, such as hydrogen gas for protons, which may be heated to a temperature high enough to break the bonds of atoms. The protons may then be extracted from the gas by applying a high accelerating DC voltage. At this stage, the beam may be contaminated by a number of types of charged particles and the desired particles may be selected by setting a magnetic field to bend the beam and passing the particles through a slit. Only the particles with the requisite charge to mass ratio will pass through this slit. These particles may form the beam of protons 616 to be accelerated.

The Low Energy Beam Transport (LEBT) 604 may guide the beam 616 between the particle source 602 and the RFQ 606. The LEBT 604 may include elements for at least one of: beam focussing, moving the beam in at least one of the horizontal and vertical direction(s), segmenting the beam 616 into discrete pulses, measuring beam characteristics and matching the beam phase space to the RFQ. The LEBT 604 may operate at an energy matching the output energy of the source.

The Radio Frequency Quadrupole (RFQ) 606 may accelerate the beam of protons 616, for example the energy of the protons may increase from between 25keV and lOOKeV to an energy between 3MeV and 20MeV, although examples are not limited to this energy increase. The beam of protons 616 may be switched On' either electronically or mechanically. For example, the beam of protons may only be On' for lOOus every 10ms.

The Medium Energy Beam Transport (MEBT) 608 may guide the beam from the RFQ 606 to the linear accelerating structures 600. The linear accelerating structures 600 may accelerate the protons from the output energy of the RFQ up to a selected energy which may be between 80MeV and 230MeV, for example. The selected energy may be user-configurable. Quadrupoles of the RFQ 606 may be used to focus and defocus the beam 616 at regular intervals between the linear accelerating structures, and the beam size may be kept small. Corrector magnets may steer the beam in at least one of the horizontal and vertical direction. The beam of protons 616 throughout at least a part of the system may be monitored using one or more diagnostic methods which may involve computer programs. The High Energy Beam Transport (HEBT) 610 may guide the beam output from the linear accelerating structures 600 to the beam delivery system 612. The HEBT 610 may comprise a plurality of quadrupoles for focusing the beam 616, bending magnets to steer the beam and beam diagnostic elements to measure beam properties. The HEBT 610 may operate at energies matching the output energy range of the collection of linear accelerating structures 600. The beam diagnostics may measure at least one of the position, size, energy and intensity of the beam 616 and may determine that these characteristics meet a required value for operation of the particle accelerator. Beam position may normally be measured by beam position monitors (BPMs) which may include RF capacitive pickups. The proton beam 616 may induce an electrical signal as it passes an RF capacitive pickup. A plurality of pickups may be placed around a beam pipe containing the proton beam 616. The pickups may be equally spaced and relative intensities may provide beam position in horizontal and vertical planes. Optionally, the pickups may provide a total (sum) signal proportional to beam intensity (beam current). Alternatively, wire scanners, destructive scintillation measurements or other methods may be used instead of or in addition to BPMs.

Beam size may be measured with a wire scanner that may have a single wire or multiple wires. The wire scanner may be mechanically moved through the beam 616, and the induced current may be measured over time to provide a beam profile. Alternatively, the beam size may be measured by acquiring an image from a scintillation screen of the beam on a charged coupled device, CCD, camera and the image may be analysed digitally. A scintillation screen may be mounted at, for example, 45 degrees to the beam 616 and may be moved into and out of the beam and the image may be viewed on a camera. The camera may be a CCD camera and may be vertically mounted.

Beam energy may be measured using a time-of-flight (TOF) monitor.

Beam intensity (or beam current) may be measured using a fast beam current monitor. The particle accelerator 620 of Figure 6 may further comprise at least one vacuum pump (not shown) or other negative pressure pump. The vacuum pump may be used to make a vacuum in the RF cavities and throughout the linear accelerator to reduce the beam-gas interaction which may disturb the beam. With a relatively low beam current there may be a low probability of beam-gas interactions and a lower standard of vacuum may be used. In some embodiments, a vacuum better than lxl 0^"6 mbar may be used, however at the particle source a vacuum less than lxlO^"7 mbar may be used. A primary vacuum line may be constructed separate from (e.g. below) the linear accelerating structures 600 and may connect some or all of the accelerating structures. This may reduce the number of vacuum pumps required to create a vacuum in the order of 1 x 10^"6 mbar (for example) in the accelerating structures. Alternatively, instead of a separate primary vacuum line, multiple individual vacuum pumps may be used. A combination of turbo -molecular pumps, oil-free backing pumps or any other suitable pump may be used.

The particle accelerator 620 may further comprise an accelerator control system 614. The accelerator control system 614 may be used to control the accelerator and may integrate other accelerator components to correctly function together. The accelerator control system may include circuitry to operate the particle accelerator and to control the linear accelerating structures to perform the method 300 as described above. The control system 614 may use industry standard programmable logic controllers (PLCs). In one embodiment, these may cover 95% of the requirements for control and monitoring. Faster signals may be acquired using oscilloscopes. Due to the use of CCD images, these fast signals may only be used for machine setup and not for operation of the machine. At least a portion of the control system 614 may be implemented by machine-readable instructions of a computer program.

The particle accelerator may be used, for example, for proton therapy in the treatment of tumours. The particle accelerator 620 as described above and illustrated in Figure 6 may further comprise a beam delivery system. The beam delivery system may ensure the correct beam dose at the correct energy may be provided to the correct position as determined by a patient treatment plan. The particle accelerator is not limited to medical use but can be used in a variety of different settings for particle acceleration such as for high energy and nuclear physics, neutron scattering and isotope production. The particle accelerator may be used to accelerate protons for use in proton therapy for medical treatment, for example for the radiological treatment of cancer. The individual resonant cavities may allow simple manufacturing and maintenance of the linear accelerating structure within the particle accelerator, which may reduce the manufacturing cost of an accelerator and make each proton therapy treatment cheaper. The distributed amplification of the RF signal, using solid state amplifiers, may allow water cooling to not be used which may further reduce the manufacturing and maintenance costs and reduce the complexity of the design. By way of contrast, the klystrons of previously known systems run as a "class A" amplifier, and as such is not as efficient as semi-conductor amplifiers as implemented in example embodiments, which can run at comparatively higher efficiencies. The linear accelerating structures may be used in a linear particle accelerator to accelerate protons as opposed to a cyclotron. Using careful design and selection of materials, for example by minimising or at least reducing the amount of steel and maximising or at least increasing the use of copper, the linear particle accelerator made up of linear accelerating structures may reduce prompt (i.e initial rather than residual) radiation and induced radiation.

Figure 7 schematically illustrates a loop coupler for use in example embodiments. The component 710 on the right-hand side of Figure 7 fits into the top left hand part 720 and is screwed in with four small bolts and epoxied in. The assembly comprising the illustrated components 710, 720 can then be joined to the middle bit in the left hand with a seal (shown) and clamp (not shown). For example, the loop coupler may be used as the radiation coupler 108a, 108b in the Figure 1 example. The loop coupler may be made of copper wire mounted on a KF vacuum fitting to allow quick replacement of the loop coupler. A "KF" vacuum fitting is an International Standards Organisation (ISO) standard quick release flange is known by the names Quick Flange (QF), Klein Flange (KF) or NW, sometimes also as DN. KF flanges are made with a chamfered back surface that may be attached with a circular clamp and an elastomeric "o-ring" that may be mounted in a metal centring ring. The KF flanges may come in standard sizes. KF flanges may be cheaper and easier to implement than other flanges (e.g. Conflat, CF, flanges) that may be needed if a higher vacuum (less pressure) is to be maintained.

According to some example embodiments, the phase of the RF radiation in each cavity may be adjusted with respect to a timing of when a particle bunch is present in the respective cavity to increase an exposure of the particle bunch and individual particles belonging to the particle bunch to an accelerating electric field portion of the RF radiation and to reduce an exposure of the particle bunch to a decelerating electric field portion of the RF radiation.

According to some example embodiments, a linear particle accelerating structure is provided for accelerating protons using radio frequency, RF, radiation, the linear particle accelerating structure comprising:

a group of at least two coupled resonant cavity units, each resonant cavity unit having an input port and an output port for a beam of protons and having an RF input to receive an RF signal to generate an electric field in the resonant cavity to accelerate the charged proton; and

processing circuitry to determine a time taken by a charged proton of the beam of protons to travel between a predetermined point of each of the two coupled resonant cavity units;

the processing circuitry being arranged to use at least one phase shifter to control relative phases of RF signals supplied to a first one of the coupled resonant cavity units and to a second one of the coupled resonant cavity units depending on the determined travel time, such that between travelling from a centre of the first one of the resonant cavity units to a centre of the second one of the resonant cavity units, the charged proton is exposed to an accelerating portion of the electric field for longer than it is exposed to a decelerating portion of the electric field. According to other example embodiments, to increase kinetic energy of protons in the range of a few MeV and a few hundred MeV, an efficient accelerating structure is provided comprising individually powered warm resonant RF cavities with individual phase control for one or more of plurality of the cavities so allow a proton travelling through two or more cavities to be exposed to a higher proportion of an accelerating phase of an electric field than its exposure to a decelerating portion of the electric field. Thus, instead of shielding protons from a decelerating electric filed of a resonant cavity using drift tubes as in a DTL, example embodiments perform a phase shift of RF radiation to at least partly avoid exposure of the protons to a decelerating portion of the electric field in the resonant cavity.

A distance between cavity centres of adjacent cavities may be less than a total the distance travelled by the protons being accelerated for one period (360°) of the RF radiation at a given input energy. Adjacent cavities may have the same or different RF frequencies. The phase adjustment may be performed for each cavity, taking account of an arrival time of a charged proton in that cavity, for example based on a calculated arrival time of the charged proton at a centre of the cavity, to ensure that when the charged proton travels through each cavity, the phase of the RF radiation is such that the charged particle experiences an overall accelerating rather than a decelerating electric field and is stable. This phase adjustment by individual phase shifters in respective cavities so that a phase of the RF radiation may be adjusted in one cavity relative to a phase of RF radiation in another cavity of the structure may eliminate the need for a drift space between cavities where otherwise, without shielding by a drift tube or without the phase shifting in the cavities, the protons would experience a decelerating field

Controlling an RF phase in each resonant cavity thus allows cavities to be closer together and therefore reduces structure length relative to a structure that is arranged to shield the protons from a decelerating phase of the RF field between accelerating "gaps".

Furthermore, controlling phase in each or at least one of a plurality of cavities of the accelerating structure means acceleration from each cavity can be more readily adjusted. For example, the phase can be adjusted so that a reference particle arriving at the centre of the cavity corresponds to a particular electric field strength of the time-varying electric field strength in the cavity induced in the cavity by the RF radiation. It may be arranged to perform a phase shift such that the proton in a cavity is exposed to a peak field strength or a close to peak field strength and is exposed to proportionally more of a proton accelerating phase of the RF cycle than a proton decelerating phase of the RF cycle.

The ability to control the RF phase in each cavity of some example embodiments such that a phase differs between at least two cavities of the accelerator structure may allow particles of different atomic weight and charge to be accelerated. By way of contrast a DTL has a fixed spacing of (accelerating) gaps and (shielding drifts), and as such only works for one charge to mass ratio of particles. In particular, a particle with double the mass but the same charge would not be accelerated as much, so would be slower to travel through the DTL and as a consequence the DTL spacing of gaps and drifts would be wrong.

In some embodiments, different ones of the plurality of resonant cavities may operate at different frequencies by synchronising two or more RF clocks, thus reducing crosstalk (interference) between cavities.

The accelerator "lattice" design, comprising a single structure having a plurality of cavities, has a low emittance growth reducing losses and therefore prompt and induced radiation.

An ability to control the energy of the protons or other protons produced by adjusting amplifier power and/or phase in each resonant cavity may allow a final beam energy to be adjusted without the use of a degrader. Protons are controlled to be in linear accelerating structure for only low percentage of time (low duty cycle) for a given charged proton acceleration session, when the linear accelerating structure is in operation.

RF radiation can be tuned down or off when protons are not in the linear accelerating structure to save power and reduce heating of structure.

Having protons or other hadrons in the accelerator structure for only when they are needed reduces prompt radiation and induced radiation.

A main power supply to each RF solid state amplifier may only provide full power when the respective amplifier needs it, such as when a charged proton bunch is passing through the cavity. This may allow for a power supply that has much lower average than peak power consumption saving money at purchase and during operation.

Low heating produced due to low RF power and a low duty cycle may eliminate or at least reduce need for water cooling of the cavity and associated amplifier.

Low electric field strength in each cavity relative to, for example a drift tube LINAC to achieve the same particle acceleration, means manufacturing tolerances can be relaxed and cost of manufacture reduced. This is because high electric field strengths can result in sparking and arcing of fields, which

components may have to be precisely machined to reduce or withstand.

Low electric field strength in each cavity means less surface preparation of the cavities is may be undertaken, thus reducing cost of

manufacture.

Low average charged proton beam current associated with the lower duty cycle may mean less beam gas scattering thus better beam quality.

The reduced beam scattering may mean that the level of vacuum pumping performed to achieve efficient operation can be reduced.

The low average proton beam current of at least some embodiments results in less beam gas scattering, which means that higher vacuum pressures can be used, which in turn may reduce cost and maintenance needed on an associated vacuum system.

The low average proton beam current may mean less beam gas scattering and therefore less prompt and induced radiation.

The use of RF frequencies of around 800 MHz in some example embodiments may allow the use of mass-produced low cost RF amplifier chips used for cellular phone base station and digital TV.

RF frequencies of approximately 800MHz, for example, can allow a good balance to be achieved between efficiency (higher frequency produces more acceleration per unit power) and complexity of engineering (higher frequencies require greater control of manufacturing tolerances). The use of individual amplifiers for two or more of a plurality of cavities has an associated benefit that the loss of one or more amplifiers can be compensated for by the other amplifiers if appropriate. Thus, in some embodiments only a non-zero subset of the resonant cavities have amplifiers. Similarly, only a non-zero subset of the resonant cavities may have phase shifters.

Implementing the charged particle accelerator as a linear accelerator rather than circular machine such as a cyclotron or a synchrotron can help to reduce particle losses that can generate harmful radiation.

In some example embodiments, the proportion of steel or iron used in fabrication of the cavities and components may be reduced, which in turn reduces radio activation.

Using a comparatively low electric field per cavity of the accelerator structure to produce the same proton acceleration may make a longer machine in total but may reduce the total power needed. This is because to achieve a given voltage, VT, for example, two cavities may be used each using a field strength of (VT/2), but adding up to the same total voltage VT. However, since RF power needed is given by P=V2/Z, where Z is impedance (akin to resistance in a DC circuit), each cavity having field strength (VT/2) requires only ¼ of the power of a single cavity operating at VT. In this example, an overall power reduction of ½ may be achieved by implementing two cavities instead of one. The total voltage of the structure may provide a measure of overall particle acceleration.

Some examples may use mainly permanent magnet quadrupoles rather than electro-magnets to focus the charged proton beam. This can reduce the power for beam focusing and therefore may reduce the cost to operate the accelerating structure.

An arrangement comprising the accelerator elements and settings (RF, magnets, etc.) to optimise the transport and acceleration of the protons with minimum losses may be referred to as an accelerator lattice. Although previously known systems use electro -magnetic quadrupoles that can be adjusted to change proton beam focusing and de-focusing magnetic fields, some embodiments may use magnet quadrupoles, for which some beam movement may be obtained by moving the magnetic quadrupole position longitudinally.

Having the possibility of adjusting the longitudinal position of the quadrupoles according to some example embodiments can allow some level of lattice (accelerator structure) adjustment and optimisation without using variable energy quadrupoles.

Accelerator structures according to some example embodiments may be long and thin, meaning that accelerators built from these multi-cavity structures can be placed in shipping containers and assembled on site in a more modular way allowing for less construction from scratch on site. The improved ability to transport the accelerator structure in shipping containers allows the machine to be built and tested more readily before being moved to a final installation site. The ability to use shipping containers may also allow linear accelerators according to some embodiments to be re-sited at minimum or reduced cost and may reduce a cost to remove facility at end of life.

Embodiments which implement a loop coupler made of copper wire mounted on a KF vacuum fitting may have the benefit of allowing easy rotation of the loop to tune the coupling (before being under vacuum).

Some embodiments use copper rather than steel vacuum fittings on the end of at least some resonant cavity units of the structure to reduce prompt and induced radiation produced when operating the accelerator. Using copper rather than steel vacuum fittings on the end of each structure also simplifies fabrication because a copper to copper joint may be easier to implement reliably than a copper to steel joint.

Some embodiments may use aluminium rather than steel supports for the accelerating structures because this may also reduce the level of prompt and induced radiation produced in the accelerator facility

Some embodiments may use of co-axial cables rather than waveguides to transmit power from the RF amplifier to the associated resonant cavity. This can reduce manufacturing costs. Such use of co-axial cables rather than waveguides to transmit the power from the RF amplifier to the resonant cavity may also simplify and speed up replacement of the power connections if needed. The use of comparatively low power per resonant cavity (relative to a drift tube LINAC for example) is one aspect that makes it easier to use coaxial cables cheap mass- produced products such as coaxial cables.

Some embodiments use similar fittings for a monitoring pickup (e.g. the RF sensors 110a, 110b of Figure 1) to each cavity for the purpose of as for the RF power input loop (radiation couples 108a, 108b). Conveniently, the loop can be rotated during setup to achieve the right impedance. By using similar fittings for a monitoring pickup (sensors) to each cavity as for the RF power input loop

(radiation coupler), the components can be easily interchanged or replaced.

For example, it is efficient for a characteristic impedance of the RF input loop coupler to be designed or adjusted to match an impedance of the resonant cavity. This can facilitate achieving a maximum power transfer to the cavity and minimum power reflected back to the amplifier. One way of performing impedance adjustment of this type is to change a penetration of the loop coupler into the cavity, the size of the loop, and the angle that the loop forms compared to the cavity axis (for example by rotating the loop).

Using water rather than concrete for radiation shielding (as opposed to using water for cooling) around the accelerator may reduce the amount of radio- activation of shielding materials and may further allow for easier disposal at end of life and may allow quick periodic replacement of primary shielding material to reducing even further the activation level of the shielding material. Using water rather than concrete for radiation shielding around the accelerator may also more readily facilitate changing the composition of the shielding for instance by the addition of boron.

Where functional units have been described as circuitry, the circuitry may be general purpose processor circuitry configured by program code to perform specified processing functions. The circuitry may also be configured by modification to the processing hardware. Configuration of the circuitry to perform a specified function may be entirely in hardware, entirely in software or using a combination of hardware modification and software execution. Program instructions may be used to configure logic gates of general purpose or special- purpose processor circuitry to perform a processing function.

Circuitry may be implemented, for example, as a hardware circuit comprising custom Very Large Scale Integrated, VLSI, circuits or gate arrays, off- the-shelf semiconductors such as logic chips, transistors, or other discrete components. Circuitry may also be implemented in programmable hardware devices such as field programmable gate arrays, FPGA, programmable array logic, programmable logic devices, A System on Chip, SoC, or the like.

Machine-readable program instructions may be provided on a transitory medium such as a transmission medium or on a non-transitory medium such as a storage medium. Such machine-readable instructions (computer program code) may be implemented in a high level procedural or object oriented

programming language. However, the program(s) may be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language, and combined with hardware implementations. Program instructions may be executed on a single processor or on two or more processors in a distributed manner.

For the purposes of the description, a phrase in the form "A / B" or in the form "A and/or B" means (A), (B), or (A and B). For the purposes of the description, a phrase in the form "at least one of A, B, and C" means (A), (B), (C), (A and B), (A and C), (B and C), or (A, B and C).

EXAMPLES

The following examples pertain to the present technique.

Some examples provide a linear particle accelerating structure for accelerating protons using radio frequency, RF, radiation, the linear particle accelerating structure comprising a group of at least two coupled resonant cavity units. Each resonant cavity unit comprises a resonant cavity with an input port and an output port for a beam of protons; an RF input to receive an RF signal; an RF phase shifter to control the phase of the RF signal; an RF solid state amplifier to amplify the RF signal; and a radiation coupler to radiate the RF signal into the cavity, wherein the RF phase shifter is to control the phase of the RF radiation to accelerate the protons for less than one full RF radiation cycle in their path through the respective cavity and the width of resonant cavity is less than the distance travelled by the particles in one period of the RF radiation.

Further example embodiments provide a linear particle accelerating structure for accelerating protons using radio frequency, RF, radiation, the linear particle accelerating structure comprising at least two coupled resonant cavity units each resonant cavity unit comprising:

a resonant cavity having an input port and an output port for a beam of protons;

an RF input to receive an RF signal; and

a radiation coupler to radiate the RF signal into the cavity;

wherein the particle accelerating structure further comprises at least one RF phase shifter to adjust, in one or more of the respective cavities, a phase of the RF signal to expose the protons to an accelerating portion of the electric field for a first duration and to expose the protons to a decelerating portion of the electric field for a second duration, wherein the first duration is greater than the second duration and greater than zero and wherein the second duration is greater than or equal to zero..

The accelerating structure may comprise a plurality of separate but coupled resonant cavity units which makes manufacturing and maintenance of the accelerating structure easier than for a single resonant tank drift tube accelerator. The resonant cavity units allow for flexibility of design as the cavities can be different sizes and may be positioned as close to each other or as far apart from each other in the accelerating structure as desired. Additionally, the resonant cavity units may be identical which allows for simple manufacture. Each resonant cavity unit may have its own designated RF phase shifter and the RF radiation in the resonant cavity may be controlled as desired to make the accelerating portion of the RF radiation be confined in the resonant cavity. Each resonant cavity unit may have its own designated RF amplifier which means less power may be required per amplifier as opposed to having one amplifier for an entire single resonant tank drift tube accelerator. The lower power amplifier could be a solid-state amplifier which may not give off excess heat and there may be no requirement for water cooling of the structure or amplifier. This may reduce the complexity of the design and the costs of manufacture and maintain the accelerating structure.

The width of the resonant cavity may be less than the distance travelled by the particles in one period of the RF radiation which means that not more than one cycle of RF radiation may be present in the cavity during the transit of the particles. This may distinguish the present technique from a single tank drift tube accelerator with a length corresponding to multiple cycles of RF during the transit of the particles, and the decelerating portions are shielded by drift tubes.

A further example of the present technique includes that only the less than one full RF radiation cycle comprises at least predominantly an accelerating portion of the RF radiation cycle to accelerate the protons. The accelerating portion of the RF radiation cycle may comprise a portion of the RF radiation cycle with a positive field amplitude.

A further example of the present technique includes the two coupled resonant cavity units to be separated by a distance less than the distance travelled by the particle in half a period of the RF radiation. This allows the possibility of positioning the resonant cavities close together and this may reduce the length of the overall linear accelerating structure. This may be particularly useful in places with confined space. In some examples an inter-cavity distance between a proton output port of a first one of the resonant cavity units and a proton input of a second, adjacent one of the cavity units is less than a distance travelled by the charged protons during one half period of the RF radiation at a proton velocity upon output from the proton output port of the first resonant cavity unit.

In some examples a distance between centres of adjacent resonant cavity units is less than a distance travelled by a proton in one period of the RF radiation.

A further example of the present technique includes the application of the RF radiation to the group of cavities, e.g. an accelerating structure, at a time to coincide with a passage of the charged particle beam through the structure. The application of the RF radiation may coincide with an arrival of the charged particle beam at a first-encountered resonant cavity of the group. The RF amplification, although not necessarily the RF amplifier, may turned off when the protons are not present in the structure. The timing of the RF radiation may allow the RF radiation to only be turned on when necessary to accelerate the particles, as opposed to being turned on continuously. This may save power and may prevent overheating of the system.

A further example of the present technique includes an RF source. The RF source may be turned off when no protons are present in the structure.

A further example of the present technique includes the beam of protons or carbon ions. Protons and carbon ions are may be used in medical treatments, however the present technique is not limited in this aspect and the beam may comprise any particle such as protons or ions such as carbon ions.

A further example of the present technique includes a sensor to monitor characteristics of the RF radiation in the cavity. The characteristics of the RF radiation may include at least one of a phase and an amplitude of the RF radiation. Feedback from the sensor may be used to determine if the phase and amplitude of the RF radiation can be changed to improve the acceleration of the protons. The RF phase shifter may be arranged to control the phase of the RF radiation based on at least one of the characteristics of the RF radiation measured by the sensor. The RF amplifier may be arranged to control the amplitude of the RF radiation based on at least one of the characteristics of the RF radiation measured by the sensor.

A further example of the present technique includes that each resonant cavity is composed of a non-superconducting material. A further example of the present technique includes that each resonant cavity is composed of a material resistant to radio activation. One such example is copper.

A further example of the present technique is when the respective ones of the resonant cavity units are non-detachably coupled (e.g. welded or bonded together). A further example of the present technique includes the radiation coupler comprising at least one of a loop coupler or an RF port from a waveguide.

An aspect of the present technique is a particle accelerator comprising a particle source to output a beam of particles, an energy accelerating device to increase the energy of the beam and one or more linear particle accelerating structures. The particle source may output protons at an energy (if protons) between 25keV and lOOkeV. The accelerating device may be a radio frequency quadrupole. The low power, low loss, low radio-activation, ease of manufacture, and distributed amplification make the particle accelerator ideal for use in proton therapy in a medical setting, however the present technique is not limited in this respect. The particle accelerator including the linear particle accelerating structure can be implemented for any suitable purpose, including but not limited to high energy and nuclear physics, neutron scattering, Isotope production and medical accelerators.

An aspect of the present technique is a method for accelerating protons using radio frequency, RF, radiation, the method may comprise: receiving an RF signal; controlling the phase of the RF signal; amplifying the RF signal; radiating the RF signal into a resonant cavity; and passing a beam of protons through the resonant cavity, wherein the controlling the phase of the RF radiation includes accelerating the protons for less than one full RF radiation cycle in their path through the cavity and wherein the width of the resonant cavity is less than the distance taken by the protons to travel during one period of the RF radiation.

A further aspect of the present technique includes a method for accelerating protons using RF radiation wherein the less than one full RF radiation cycle may comprise at least predominantly an accelerating portion of the RF radiation cycle to at least predominantly accelerate the protons.

Example embodiments provide a method for accelerating protons using radio frequency, RF, radiation, the method comprising:

receiving one or more RF signals; and

radiating the one or more RF signals into respective resonant cavities of at least two coupled resonant cavities; passing a beam of protons through the at least two coupled resonant cavities;

adjusting a phase of an RF signal, in one or more of the respective resonant cavities, to expose protons of the beam of protons to an accelerating portion of the electric field for a first duration and to expose the protons to a decelerating portion of the electric field for a second duration, wherein the first duration is greater than the second duration and greater than zero and wherein the second duration is greater than or equal to zero..

A further aspect of the present technique includes a method for accelerating protons using RF radiation including monitoring the RF radiation in the cavity.

An aspect of the present technique is a computer-readable storage medium comprising machine-readable instructions which, when executed by a processor, cause the processor to carry out the steps of the method for accelerating protons using RF radiation as described above.

Claims

CLAIMS:

1. A linear particle accelerating structure (100) for accelerating protons using radio frequency, RF, radiation, the linear particle accelerating structure comprising at least two coupled resonant cavity units (130a; 130b; 230a; 230b), each resonant cavity unit comprising:

a resonant cavity (102a; 102b; 202a; 202b) having an input port (114a; 114b) and an output port (116a; 116b) for a beam of protons (112);

an RF input to receive an RF signal; and

a radiation coupler (108a; 108b; 208a; 208b) to radiate the RF signal into the cavity;

wherein the particle accelerating structure further comprises at least one RF phase shifter (104a; 104b; 204a; 204b) to adjust, in one or more of the respective cavities, a phase of the RF signal, to expose the protons to an accelerating portion of the electric field for a first duration and to expose the protons to a decelerating portion of the electric field for a second duration, wherein the first duration is greater than the second duration and greater than zero and wherein the second duration is greater than or equal to zero.

2. The linear particle accelerating structure of claim 1 , wherein the less than one full RF radiation cycle comprises at least predominantly an accelerating portion of the RF radiation cycle to at least predominantly accelerate the protons.

3. The linear particle accelerating structure of claim 2, comprising at least one solid-state amplifier to amplify the RF signal of corresponding to a respective radiation coupler.

4. The linear particle accelerator of any claims 1 to 3, wherein a distance between centres of adjacent resonant cavity units is less than a distance travelled by a proton in one period of the RF radiation.

5. The linear particle accelerating structure of any claims 1 to 4, wherein the RF radiation is applied to the group of cavities at a time to coincide with a passage of the charged particle beam through the structure.

6. The linear particle accelerating structure of any one of claims 1 to 5, wherein the RF amplification is turned off when the protons are not present in the structure.

7. The linear particle accelerating structure of any one of claims 1 to 6, wherein each resonant cavity unit comprises a corresponding RF phase shifter.

8. The linear particle accelerating structure of claim 7, comprising a controller to turn off the at least one RF source when no protons are present in the structure.

9. The linear particle accelerating structure of any one of claims 1 to 8, wherein at least one of the resonant cavity units comprises: a sensor (110a; 110b; 210a; 210b) to monitor characteristics of the RF radiation in the cavity.

10. The linear particle accelerating structure of claim 9, wherein the characteristics of the RF radiation monitored by the sensor include at least one of a phase and amplitude of the RF radiation.

11. The linear particle accelerating structure of claim 9 or 10, wherein the RF phase shifter is arranged to control the phase of the RF radiation in the cavity unit is based on at least one of the characteristics of the RF radiation measured by the sensor such that a phase of the RF radiation is synchronous with protons of the beam entering the cavity unit.

12. The linear particle accelerating structure of claim 9 or 10, wherein the RF amplifier is arranged to control the amplitude of the RF radiation based on at least one of the characteristics of the RF radiation measured by the sensor.

13. The linear particle accelerating structure of any one of claims 1 to 12, wherein each resonant cavity is composed of a non-superconducting material.

14. The linear particle accelerating structure of any one of claims 1 to 13, wherein each resonant cavity composed of a material resistant to radio activation.

15. The linear particle accelerating structure of any one of claims 1 to 14, wherein the respective ones of the at least two resonant cavity units are non- detachably coupled.

16. The linear particle accelerating structure of any one of claims 1 to 15, wherein the radiation coupler comprises at least one of a loop coupler or an RF port from a waveguide.

17. A particle accelerator comprising :

a particle source to output a beam of protons;

an energy accelerating device to increase the energy of the beam; and the linear particle accelerating structure (100; 400) of any one of claims 1 to 16.

18. The particle accelerator of claim 17, wherein the particles accelerated comprise protons and the particle source is arranged to output protons at an energy of between 25keV and lOOkeV.

19. The particle accelerator of claim 17 or claim 18, wherein the energy accelerating device used before the main accelerating structures is a radio frequency quadrupole.

20. A method (300) for accelerating protons using radio frequency, RF, radiation, the method comprising: receiving one or more RF

radiating the one or more RF signals into respective resonant cavities of at least two coupled resonant cavities;

passing a beam of protons through the at least two coupled resonant cavities;

adjusting a phase of an RF signal, in one or more of the respective resonant cavities, to expose protons of the beam of protons to an accelerating portion of the electric field for a first duration and exposing the protons to a decelerating portion of the electric field for a second duration, wherein the first duration is greater than the second duration and greater than zero and wherein the second duration is greater than or equal to zero.

21. The method for accelerating protons using RF radiation of claim 20, wherein the less than one full RF radiation cycle comprises at least predominantly an accelerating portion of the RF radiation cycle to at least predominantly accelerate the protons.

22. The method for accelerating protons using RF radiation of claim 20, the method further comprising:

monitoring (312) the RF radiation in the cavity.

23. A computer-readable storage medium comprising instructions which, when executed by a processor, cause the processor to carry out the steps of the method of any one of claims 20 to 22