The advantage of high intensity high brightness positron beam that we will achieve over the other facilities is result of the combined gain of high current, high quality JLab FEEL beam and our design, which separates incident beam of electrons and low energy positrons. Only thin rotating converter will be exposed to the high power electron beam. Because of its thickness, diameter, and high speed of rotation the power generated in that converter (just about 2kW) will be removed by the thermal radiation process. There is no need for additional cooling. These calculations are done carefully and they are based on study performed by Saclay group. Produced positrons will be separated from the high energy electrons. Electrons will be directed to the beam dump and only positrons will be channeled to the moderator. This will be done by two quads and a solenoid.
The positrons will be produced in the 0.5 mm tungsten converter, which is moving with the velocity of 100 m/s on the rotating wheel of 50 cm diameter. Electron beam of 1 mA will be focused to the 0.10 mm spot. After pass through the converter electrons will be deflected to the beam dump. Positrons with average energy of 30 MeV will be focused to the 180o magnetic bend, which provides the momentum selection of +/-5% (+/-1 MeV) and the energy-phase alignment. A combiner magnet in front of the linac serves for an electron beam of 15 MeV and a positron beam of 30 MeV. A beam splitter after the linac separates an electron beam of 120 MeV, positron beam of 30 MeV, and a positron beam of 150 MeV
The production of positrons with electron beam of hundred MeV is well understood and is calculated by means of GEANT code, see for instance our recent worki. The plot on fig. 2 shows the number of positrons per 1 MeV bin in kinetic energy produced in 0.5 mm tungsten converter for 2.5 106 incident electrons of 120 MeV.
Basic beam parameters at full intensity:
Considerations below are for 120 MeV electron beam with intensity of 1 mA.
With 120 MeV beam and current of 1 ma the beam intensity (with a conservative approach for a moderator efficiency of 4x10-4) is 8x1010 which is for a factor 100 better than anywhere else. However, even more important is the brightness of the beam. At NCSU and other institution the beam phase is of order of several cm, but because of the high quality of JLAB FEL beam the diameter of the positron beam at the proposed center will be 1 mm. It results in an additional factor of 1000 in the brightness of the beam, so the brightness will be for a factor of 100 (beam intensity) x 1000 (size of the beam) = 105 higher at the proposed center than it is for instance at NCSU. However, we will use further in the text for this factor 104 as a safe estimate.
Please note that we are also conservative in our predictions for the beam intensity using a safety factor in our calculations. For instance for the moderator efficiency we are using 4x10−4 ref26 even the efficiency of 10-3 are reported27. In addition we also did not take into account that FEL will be upgraded to 600 MeV which will also result in higher positron intensity than it was in the proposal (because of easy separation secondary and primary beams). In addition the bam at JLAB is with well defined pulsed structure which may be an additional advantage for some experiments.
Figure 1. The first stage beam arrangement. C - converter; AB achromatic bend; SS – strong solenoid, BD - electron beam dump, M - moderator, WS - weak solenoid, AT - annihilation target, D1, D2 - detectors of 511 keV gammas.
Emittance of the positron beam for 120 MeV electron beam and 30 MeV positron beam after 0.5 mm tungsten converter
Transverse size of the incident electron beam on the converter foil is sx = 20 m (much tighter focus doesn’t help, however it requires a very large raster frequency or fast rotation due to high power heat density in the converter) for 0.5 mm thickness of the converter and 30 MeV positron beam energy leads to the positron beam source transverse size of sx = 25 m (a thickness of the radiator – 0.5 mm product with a positron typical angle – 0.03 rad and 20 m electron beam spot size). Emittance of the produced positron beam is normalized = sx xbeam = 45 m.
Transport of the positron beam and its intensity
Electron beam will be deflected to the local dump (total power of 150 kW) by the FEL bend magnet. Positrons with energy 30 +/- 1.5 MeV will be transported through the 180 degree bend to the injection point of the linac for production of high energy 150 MeV positrons or will be redirected to the experimental area for the study of the bulk material properties. Transport line will have focusing quads or solenoids, which to match the FEL lattice.
Figure 2. Energy distribution of the positrons produced by 120 MeV electrons in the W converter integrated over all positrons angles. Number of positrons is given per the positron energy interval of 1 MeV for the total number of incident electrons of 2.5 106.
Heat load on the converter, needs for raster, temperature profile
Experimental test by Saclay group demonstrated reliable operation for heat power density of 1-2 kW/cm2 on the tungsten target. We had estimate the total power deposited in the 0.5 mm thickness tungsten converter by 1 mA electron beam at 120 MeV incident energy as of 1.7 kW, so the target area of 1 cm2 is sufficient. Such area will provide sufficient cooling by means of the converter body radiation (in fact radiation is stronger than SB law predicts).
For the wheel rotating with the speed of 100 m/s the instant temperature rise on the beam trail is 350o C. The flywheel diameter of 30 cm and for plane of the wheel perpendicular to beam direction we need the heat to be distributed in the circular strip of 0.1 mm width. It is larger than desirable 2 x 0.02 mm width of the beam spot, however in time of just one revolution of the wheel the heat propagates of 0.6 mm. As result the effective area of the target is about 12 cm2.
Therefore the raster (and anti-raster) could be avoided for the beam intensity of 1 mA. The converter parameters: thickness of tungsten is 0.5 mm, the diameter is 30 cm, the frequency of rotation is 6000 rpm, an axis is parallel to the beam direction.
Magnetic optics elements near the flywheel converter, induction heat
The conductor moving relatively to the magnetic elements of the positron optics could get a heat load due to very large induction currents. As an estimate we can use the field value of 1 kG with an area of 2x2 cm2 oriented normal to the plane of flywheel. Such situation is possible for the case of solenoid focusing. For the conditions of above presented flywheel an estimated power is about 4 kW over area 200 cm2. So, it should not melt the converter, because power density is low, but this heat load needs to be considered in detailed design of the wheel. Some reduction of the field could be achieved for example by means of the thin iron shield.
End User Positron Beam Design
The analyses of surfaces and depth-profile of thin films that can vary from several nanometer to a few microns require energy of the implanted monoenergetic low-energy positrons which is tunable from several eV to few keV. It will be provided by the secondary bemaline that consists of the moderator, magnetic and electrostatic lenses, and a magnetic solenoid that will accelerate the positron beam to the energy required by the desired penetrated depth for a particular experiment and transport it to the experimental area. A schematic of 4 m long secondary bemline developed at our collaborators at NCSU positron facility is shown on the picture 3a and 3b. Very similar secondary beamline will be developed for Jlab positron facility.
Figure 3. A) Schematic of 0.5m long moderator with accelerating extraction lenses and 4m long solenoid assembly attached to it. B) The photograph of the built device shown on the schematic on the left, which is developed and it is in use at NCSU.
The secondary beamline also includes several beam switches which are basically Helmholtz lenses, gate valves, ion pump, turbo pump, sample insertion chamber, and x-y translator.
The small fraction of monoenergetic positrons leaving the moderator will be separated from the unmoderated fraction before they will be utilized in experiments. This separation will take place in the beam guidance system by an energy filter, by applying external magnetic fields perpendicular to the beam direction or by the use of bent solenoids. The positrons beams will be then diverted to different end stations using switchyards and electrostatic elements. Experiments that will be actively pursued include high resolution measurements of electron momentum by angular correlation techniques, positron diffraction and holography, positron induced desorption from the surface, and positron induced Auger electron spectroscopy. These techniques have been developed and tested at other facilities and will be attached to the various end stations.
Three kinds of the positron beams will be available for the users:
Unfortunately thermally produced few eV monoenergetic positron beam can not directly be applied in the spectroscopic coincidence experiments. The conventional setup with start and stop detectors is not applicable in a slow-positron-beam system due to the strength of the source, since the start -quanta in the source cannot be any more correlated with annihilation events in the sample. Furthermore, the time of flight is much longer than the lifetime in the specimen.
Thus, to make measurement of the positron lifetime possible we will provide pulsed beam and supply the start pulse by a specially designed bunching system. Positrons produced in the linac target will be captured into a "stretcher," which will reduce their energy and intensify the pulses. They will be then sent through a series of bunchers, which will shorten the pulses, and remoderators, which will cool the pulses further.
We will also design the beamline to obtain a clean data collection time window of about 50 ns that will be not disturbed by annihilations from other sources choosing the length of the optical columns to produce all annihilations from moderators, the magnetic grid, accelerating grids, and stopper in a 10 ns window before the target annihilations.
The experienced JLab FEL engineers will also design beamline to compensate for the beam spot shifts from residual ac magnetic fields and a dc background fields and to reduce the shift of the final focus spot and degradation of the spot size.