Small, but remote: a miniature linear particle accelerator, which set a new record

The principle “more means more powerful”, which is familiar to us, has long been established in many sectors of society, including in science and technology. However, in modern realities, the practical implementation of the saying “small and distant” is becoming more and more common. This is manifested both in computers that previously occupied a whole room, and now they are placed in the palm of a child, and in particle accelerators. Yes, remember the Large Hadron Collider (LHC), the impressive dimensions of which (26,659 meters in length) are literally indicated in its name? So, this is already in the past according to the scientists from DESY, who have developed a miniature version of the accelerator, which in terms of performance is not inferior to its full-sized predecessor. Moreover, the mini accelerator even set a new world record among terahertz accelerators, doubling the energy of the implanted electrons. How was the miniature accelerator developed, what were its main principles of operation and what did practical experiments show? This will help us to know the report of the research group. Go.

The basis of the study

According to Dongfang Zhang and his colleagues at DESY (German Electronic Synchrotron), who developed the mini-accelerator, ultrafast sources of electrons play an incredibly important role in the life of modern society. Many of them are manifested in medicine, the development of electronics and in scientific research. The biggest problem of the current linear accelerators using radio frequency generators is their high cost, complexity of the infrastructure and impressive appetites for power consumption. And such shortcomings severely limit the availability of such technologies for a wider range of users.
')
These obvious problems are an excellent incentive to develop devices whose dimensions will not cause horror, as well as the degree of power consumption.

Among the relative novelties in this industry can be identified terahertz accelerators, which have a number of "buns":

• it is expected that short waves and short pulses of terahertz radiation will significantly increase the breakdown threshold * caused by the field, which will increase the acceleration gradients;

Electrical breakdown * - a sharp increase in current when a voltage is applied above the critical one.

• the presence of efficient methods for generating high-field terahertz radiation allows for internal synchronization between electrons and excitation fields;
• Classical methods can be used to create such devices, but their cost, production time and size will be greatly reduced.

Scientists believe that their terahertz accelerator on a millimeter scale is a compromise between conventional accelerators, which are now available, and micro accelerators, which are being developed, but have many drawbacks because of their very small size.

Researchers do not deny that the technology of terahertz acceleration has been under development for some time. However, in their opinion, in this area there are still a lot of aspects that have not been studied, tested or implemented.

In their work, which we are considering today, scientists demonstrate the capabilities of STEAM ( segmented terahertz electron accelerator and manipulator ) - a segmented terahertz electron accelerator and manipulator. STEAM allows you to reduce the length of the electron beam to subpicosecond duration, thereby providing femtosecond control over the acceleration phase.

It was possible to achieve an acceleration field of 200 MV / m (MV - megavolt), which leads to a record terahertz acceleration of> 70 keV (kiloelectronvolt) from the implanted 55-keV electron beam. Thus, accelerated electrons up to 125 keV were obtained.

Device structure and its implementation

Image №1: scheme of the device under study.


Image No. 1-2: a is a diagram of the developed 5-layer segmented structure, b is the ratio of the calculated acceleration and the direction of propagation of electrons.

Electron beams (55 keV) are generated from an electron gun * and embedded in a terahertz STEAM-buncher (beam compressor), after which they are transferred to STEAM-linac ( linear accelerator * ).

Electron gun * is a device for generating an electron beam of the required configuration and energy.

A linear accelerator * is an accelerator in which charged particles pass a structure only once, which distinguishes a linear accelerator from a cyclic accelerator (for example, LHC).

Both STEAM devices receive terahertz pulses from a single near-infrared (NIR) laser, which also triggers the photocathode of an electron gun, which leads to internal synchronization between electrons and accelerating fields. Ultraviolet pulses for photoemission on a photocathode are generated through two successive stages of SHG * of the main wavelength of near infrared light. This process converts a laser pulse with a wavelength of 1020 nm, first to 510 nm and then to 255 nm.

SHG * (generation of the second optical harmonic) - the process of combining photons with the same frequency during interaction with nonlinear material, which leads to the formation of new photons with double energy and frequency, as well as two times shorter wavelength.

The remaining part of the NIR laser beam is divided into 4 beams, which are used to generate four single-cycle terahertz pulses by generating the difference in intra-pulse frequencies.

Two terahertz pulses are then delivered to each STEAM device through symmetric horny structures, which direct the terahertz energy to the interaction region across the direction of electron propagation.

When electrons enter each of the STEAM devices, they are exposed to the electrical and magnetic components of the Lorentz force * .

The Lorentz force * is the force with which an electromagnetic field acts on a charged particle.

In this case, the electric field is responsible for acceleration and deceleration, and the magnetic field causes lateral deviations.


Image number 2

As we see in images 2a and 2b , inside each STEAM device, the terahertz beams are separated across thin metal sheets into several layers of different thickness, each of which acts as a waveguide transporting part of the total energy to the interaction region. Also in each layer there are dielectric plates to match the arrival time of the terahertz wave front * with the front of electrons.

Wavefront * - the surface to which the wave reached.

Both STEAM devices operate in electric mode, that is, so as to produce an imposition of an electric field and suppression of the magnetic field at the center of the interaction region.

In the first device, the electrons are timed to pass through the intersection of the zero * terahertz field, where the time gradients of the electric field are maximized, and the average field is minimized.

Zero crossing * is the point where there is no voltage.

This configuration causes the acceleration of the tail of the electron beam and the slowing of its head, which leads to ballistic longitudinal focusing ( 2a and 2c ).

In the second device, the synchronization of the electron and terahertz radiation is established so that the electron beam experiences only the negative cycle of the terahertz electric field. Such a configuration results in pure continuous acceleration ( 2b and 2d ).

A laser with NIR radiation resembles a cryogenically cooled Yb: YLF system, which emits optical pulses with a duration of 1.2 ps and an energy of 50 mJ at a wavelength of 1020 nm and a repetition frequency of 10 Hz. And terahertz pulses with a center frequency of 0.29 terahertz (a period of 3.44 ps) are generated by the method of an inclined pulse front.

Only 2 x 50 nJ of terahertz energy was used to power the STEAM-buncher (beam compressor), and 2 x 15 mJ was required for the STEAM-linac (linear accelerator).

The diameter of the inlet and outlet openings of both STEAM devices is 120 microns.

The beam compressor is designed with three layers of the same height (0. 225 mm), which are equipped with fused silica plates (ϵ _r = 4.41) with a length of 0.42 and 0.84 mm to control time synchronization. Equal heights of compressor layers reflect the fact that acceleration does not occur ( 2c ).

But in a linear accelerator, the heights are already different - 0.225, 0.225 and 0.250 mm (+ plates of fused silica 0.42 and 0.84 mm). An increase in the height of the layer explains the increase in electron velocity during acceleration.

Scientists note that the number of layers is directly responsible for the functionality of each of the two devices. To achieve a higher degree of acceleration, for example, more layers and a different configuration of heights are needed to optimize the interaction.

The results of practical experiments

First of all, the researchers recall that in traditional radio frequency accelerators, the effect of the time span of the implanted electron beam on the properties of the accelerated beam is due to a change in the electric field experienced during the interaction by different electrons inside the beam, arriving at different times. Thus, it can be assumed that fields with a large gradient and beams with a longer duration will lead to a greater energy spread. Introduced beams of long duration can also lead to higher emittance values * .

Emittance * is the phase space occupied by an accelerated charged particle beam.

In the case of a terahertz accelerator, the excitation field period is about 200 times shorter. Consequently, the strength * of the supported field will be 10 times higher.

The electric field strength * is an indicator of the electric field, equal to the ratio of the force applied to a fixed point charge placed at a given field point to the magnitude of that charge.

Thus, in a terahertz accelerator, the field gradients experienced by electrons can be several orders of magnitude higher than in a conventional device. The time scale at which the curvature of the field is noticeable will be much less. From this it follows that the duration of the introduced electron beam will have a more pronounced effect.

Scientists in practice decided to check these theories. To do this, they introduced electron beams of different durations, which was controlled by compression using the first STEAM device (STEAM-buncher).


Image number 3

In the case when the compressor was not connected to a power source, electron beams (55 keV) with a charge of 1 fKl (femtocoulon) passed about 300 mm from the electron gun to the linear accelerator device (STEAM-linac). These electrons could expand under the action of space charge forces up to a duration of more than 1000 fs (femtoseconds).

With such a duration, the electron beam occupied about 60% of the half-wave of the accelerating field with a frequency of 1.7 ps, which led to the energy spectrum after acceleration with a peak of 115 keV and half-width of the energy distribution of more than 60 keV ( 3a ).

To compare these results with the expected ones, the situation of electron propagation through a linear accelerator was modeled, when the electrons were out of sync (i.e., do not coincide with) with respect to the optimal introduction time. Calculations of this situation showed that the increase in the electron energy is very dependent on the moment of introduction up to the subpicosecond time scale ( 3b ). That is, with optimal tuning, the electron will experience a full half-period of acceleration of terahertz radiation in each layer ( 3c ).

If electrons arrive at different times, then they experience less acceleration in the first layer, from which they need more time to pass. Then out of sync is enhanced in the next layers, from which an undesirable slowdown occurs ( 3d ).

In order to minimize the negative effect of the time span of the electron beam, the first STEAM device operated in the compression mode. The electron beam duration at the linear accelerator was optimized to a minimum of ~ 350 fs (half width) by adjusting the terahertz energy supplied to the compressor and switching the linear accelerator to the hatching mode ( 4b ).


Image number 4

The minimum duration of the beam was set in accordance with the duration of the UV pulse of the photocathode, the duration of which was ~ 600 fs. Also the important role was played by the distance between the compressor and the strip, which limited the speed of condensation by speed. Taken together, these measures ensure the femtosecond accuracy of the injection phase at the acceleration stage.

Image 4a shows that the spread of the energy of a compressed electron beam after optimized acceleration in a linear accelerator decreases by a factor of ~ 4 compared to an uncompressed one. Due to acceleration, the energy spectrum of the compressed beam shifts towards higher energies, in contrast to the uncompressed beam. The peak of the energy spectrum after acceleration is about 115 keV, and the high-energy tail reaches about 125 keV.

These indicators, according to a modest statement of scientists, are a new record of acceleration (before acceleration was 70 keV) in the terahertz range.

But in order to reduce the energy spread ( 4a ), it is necessary to achieve an even shorter beam.


Image number 5

In the case of an uncompressed injected beam, the parabolic dependence of the beam size on the current reveals the transverse emittance in the horizontal and vertical directions: ε _{x, n} = 1.703 mm * mrad and ε _{y, n} = 1.491 mm * mrad ( 5a ).

Compression, in turn, improved the transverse emittance 6 times to ε _{x, n} = 0.285 mm * mrad (horizontal) and ε _{y, n} = 0.246 mm * mrad (vertical).

It is worth noting that the degree of emittance reduction is approximately twice that of the reduction in the beam duration, which is a measure of the nonlinearity of the interaction dynamics with time when electrons experience strong focusing and defocusing of the magnetic field during acceleration ( 5b and 5c ).

Image 5b shows that electrons introduced at the optimum time experience the entire half-period of the acceleration of the electric field. But electrons that arrive before or after the optimal point in time experience less acceleration and even partial deceleration. Such electrons as a result receive less energy, roughly speaking.

A similar situation is observed when exposed to a magnetic field. Electrons introduced at optimum time experience a symmetrical number of positive and negative magnetic fields. If the introduction of electrons occurred before the optimal time, then there were more positive fields and less negative ones. In the case of the introduction of electrons later than the optimal time, it is less positive and more negative ( 5c ). And such deviations lead to the fact that the electron can deviate to the left, right, up or down, depending on its position relative to the axis, which leads to an increase in the transverse momentum corresponding to the focusing or defocusing of the beam.

For more detailed acquaintance with the nuances of the study I recommend to look into the report of scientists and additional materials to it.

Epilogue

Summing up, the accelerator's productivity will increase in the case of a decrease in the duration of the electron beam. In this work, the achievable beam duration was limited by the installation geometry. But, in theory, the beam duration can reach less than 100 fs.

Also, scientists note that the beam quality can be further improved by reducing the height of the layers and increasing their number. However, this method is not without problems, in particular increasing the complexity of the production device.

This work is the initial stage of a more extensive and detailed study of a miniature version of a linear accelerator. Despite the fact that the tested version already shows excellent results, which can fairly be called a record, there is still a lot of work.

Thank you for your attention, stay curious and have a good working week, guys! :)

Thank you for staying with us. Do you like our articles? Want to see more interesting materials? Support us by placing an order or recommending to friends, 30% discount for Habr's users on a unique analogue of the entry-level servers that we invented for you: The whole truth about VPS (KVM) E5-2650 v4 (6 Cores) 10GB DDR4 240GB SSD 1Gbps from $ 20 or how to share the server? (Options are available with RAID1 and RAID10, up to 24 cores and up to 40GB DDR4).

Dell R730xd 2 times cheaper? Only we have 2 x Intel TetraDeca-Core Xeon 2x E5-2697v3 2.6GHz 14C 64GB DDR4 4x960GB SSD 1Gbps 100 TV from $ 199 in the Netherlands! Dell R420 - 2x E5-2430 2.2Ghz 6C 128GB DDR3 2x960GB SSD 1Gbps 100TB - from $ 99! Read about How to build an infrastructure building. class c using servers Dell R730xd E5-2650 v4 worth 9000 euros for a penny?