Energy Measurements by Means of Transition Radiation in novel Linacs
Advanced linear accelerator design may use Optical Transition Radiation (OTR) screens to measure beam spot size; for instance, such screens are foreseen in plasma based accelerators (EuPRAXIA@SPARC_LAB) or Compton machines (Gamma Beam Source@ELI-NP). Optical Transition Radiation angular distribution strongly depends on beam energy. Since OTR screens are typically placed in several positions along the Linac to monitor the beam envelope, one may perform a distributed energy measurement along the machine. Furthermore, a single shot energy measurement can be useful in plasma accelerators to measure shot to shot energy variations after the plasma interaction. Preliminary measurements of OTR angular distribution of about electrons have been performed at the SPARC_LAB facility. In this paper, we discuss the sensitivity of this measurement to beam divergence and others parameters, as well as the resolution required and the needed upgrades of conventional OTR diagnostics, using as an example the data collected at SPARC_LAB.
keywords:Compton Gamma Source, Optical Transition Radiation, Plasma acceleration, Energy measurement
The Gamma Beam Source Bacc2013 () (GBS) machine is an advanced source of up to Gamma rays based on Compton back-scattering, i.e. collision of an intense high power laser beam and a high brightness electron beam with maximum kinetic energy of about . The Linac will provide trains of bunches in each RF pulse, spaced by the same time interval needed to recirculate the laser pulse in a properly conceived and designed laser recirculator. Thus, the same laser pulse will collide with all the electron bunches in the RF pulse, before being dumped. The final design foresees trains of 32 electron bunches separated by , distributed along a RF pulse, with a repetition rate of .
In a typical monitor setup, the beam is imaged via Optical Transition Radiation (OTR) or YAG screen using standard lens optics, and the recorded intensity profile is a measure of the particle beam spot. In conjunction with other accelerator components, it will also be possible to perform various measurements on the beam, namely: its energy and energy spread (with a dipole), bunch length Fili2011 () (with an RF deflector), Twiss parameters Most2012 () (by means of the quadrupole scan technique) or in general 6D characterization on bunch phase space Cian2015 (). Such techniques are common in conventional ferrario2013sparc_lab () and unconventional Most2012 (); antici2012laser (); rossi2014external () high brightness Linacs. In this paper, we refer unconventional or novel Linacs to the plasma based accelerators (both beam and laser driven) and to the GBS machine. The reason why, in our opinion, the GBS machine can be defined as a novel Linac is due to the fact that it will produce high brightness multi-bunch pulses (bunch by bunch separation of ) that will be accelerated by a newly designed, and not yet fully characterized, C-Band accelerating structures alesini2017design (). Such schemes could pose different challenges in terms of beam stability that need to be measured by the appropriate diagnostics.
Since OTR screens are typically placed in several positions along the Linac to monitor the beam envelope, one may perform a distributed energy measurement along the machine. This will be useful, for instance, during the commissioning phase of the GBS in order to verify the correct functionality of the newly designed C-Band accelerating structures alesini2017design (), due to the fact that there are OTR screens after each accelerating module.
Furthermore, a single shot energy measurement can be useful in plasma accelerators to measure shot to shot energy variations after the plasma interaction (i.e. EuPRAXIA@SPARC_LAB ferrario2017eupraxia ()). In order to perform this measurement with ultra short beams (typical in plasma accelerators), one needs to take into account also the coherent OTR whose contribution is neglected in this paper. Moreover, for this type of beams, the use of dipoles to perform energy measurements could be critical, due to the high energy jitter.
Several techniques have been proposed to measure energy of a beam with high jitter using a spectrometer; for instance, in this study nakamura2008broadband () the proposed configuration (with a total length of 1 meter) foreseen one dipole and 2 scintillating screens that can measure beam energy in the range from to . Other studies sears2010high (); soloviev2011two () proposed schemes with both a permanent magnet and an electromagnetic spectrometer to increase the resolution in an energy range that goes from to . A simpler and more compact () scheme glinec2006absolute () is based on a permanent magnet spectrometer and 1 lanex screen for low charge beam in the energy range from to . The technique proposed in this paper, however, cover a wide range of energies (i.e. from to ) with a compact, cheap and already installed hardware (i.e. OTR screen, CCD sensor, lenses). Moreover, if a different range of energies or an improvement of resolution is needed, one can easily change “in air” optics without modifying in vacuum devices. This type of measurement meets also the requirement of having a compact Linac since it does not need any bending magnet.
This paper describes a theoretical concept of the OTR-based electron beam property measurements, followed by the experimental study using a 100-MeV class conventional accelerator (SPARC_Lab). Conclusions and outlook are presented as well.
Optical Transition Radiation screens are widely used for beam profile measurements, as well as in ELI-GBS marongiu2017optical (); cioeta2017spot (). The radiation is emitted when a charged particle beam crosses the boundary between two media with different optical properties. For beam diagnostic purposes the visible part of the radiation is used; an observation geometry in backward direction is chosen corresponding to the reflection of virtual photons at the screen which acts as a mirror.
The main advantages of OTR are the instantaneous emission process allowing fast single shot measurements, and the good linearity (neglecting coherent effects); indeed, the typical response time of the OTR is lumpkin1998time () while for a YAG screen is YAG_crytur (). The disadvantages are that the process of radiation generation is invasive, (i.e. a screen has to be inserted in the beam path and, unless a properly designed thin OTR foil is used, the beam got completely scattered when it passes through the screen), and that the radiation intensity is much lower in comparison to scintillation screens.
Another advantage of the OTR is the possibility to measure the beam energy by means of observation of its angular distribution (see figure 1); this technique has been proved feasible by many authors ginsburg1946radiation (); wartski1975thin (), also for low energy beams castellano1995analysis (). The angular distribution can be expressed by the well known formula ginsburg1946radiation ():
where is the frequency, is the solid angle, is the intensity of the radiation, is the electron charge, is the speed of light, is the vacuum permittivity and is the reflectivity of the screen; the peak of intensity is at with respect to the beam direction.
Due to the beam divergence, the angular distribution of the whole beam will be different from at the center: the ratio between the minimum and the maximum intensity is related to the beam divergence. A parameter called visibility can be defined as in Eq. (2): in analogy with the contrast function, the measurement with the OTR angular distribution can be reliably done if the visibility parameter is greater or equal to cianchi2016transverse ().
where is the complex error function and is the real part cianchi2016transverse ().
As it can be seen in Eq. (3), and depends on both divergence and energy of the beam. Equation 2, therefore, implicitly gives the range of beam energy and divergence over which this technique can be used: since for bigger energies the angular distribution narrows, the sensitivity to angular spread is higher than for low energy beams where the angular distribution is wide. For instance, for a beam energy of , the divergence must be below ; for a beam energy of , the divergence must be below , while for a beam energy of , the divergence must be below .
Moreover, the beam energy has an effect on the ability of a given optic system to resolve the angular distribution, since the angular distribution narrows as the energy increases; therefore, a change of the optic system (i.e. a bigger focal length) could be necessary.
|Data set 1||Data set 2|
|()||110.82 (0.07)||123.1 (0.04)|
|()||0.13 (0.002)||0.06 (0.0002)|
|()||108 (3)||120 (4)|
|()||0.52 (0.03)||1.1 (0.09)|
|()||0.66 (0.02)||1.04 (0.09)|
In this section we shown the application of the technique described in the previous section to the high brightness photoinjector of SPARC_LAB; we verified the feasibility of the technique for different values of charge, energy, divergence) and with different measurement setup (i.e. single shot and time integrated measurements).
Equation 3 was used to retrieve the beam energy and divergence for different machine working points. The first working point, called “Data set 1”, was characterized by lower charge, energy and divergence with respect to the second working point, called “Data set 2” (see Table 1). The optic layout used to observe the OTR angular distribution was the same for the different working points and it was reported in cianchi2016transverse ().
|Data set 1||()||()||()|
|Single Shot||105.35 (2.04)||0.72 (0.21)||0.74 (0.17)|
|10 shots||108.33 (1.53)||0.75 (0.09)||0.77 (0.08)|
|50 shots||109.87 (0.55)||0.72 (0.04)||0.78 (0.06)|
The measurements of the first working point, in the single shot configuration, were affected by a low Signal to Noise Ratio (SNR); the coefficient of determination of the fit (R-square) was while the uncertainty was around for the energy and below for the divergence.
A integration and a integration measurements were performed as well: the SNR was increased, as well as the goodness of fit. In the integration case, for instance, the R-square value became while the uncertainty became around for the energy and below for the divergence. The integration case, shown in figure 2, gave an R-square value of while the uncertainty was around for the energy and below for the divergence.
Also the accuracy of the measurement, calculated with respect to the values in table 1, increased: for the energy measurement, it went from of the single shot case to the of the integration case (in the integration case, the accuracy was ). For the divergence, instead, the accuracy remained around a value of (see Table 2).
|Data set 2||()||()||()|
|Single Shot||122.13 (2.04)||1.4 (0.1)||1.3 (0.1)|
|10 shots||123.66 (1.02)||1.3 (0.05)||1.2 (0.04)|
For the second working point, the measurements were done in the single shot configuration and with integration; in the first case, shown in figure 3, the R-square value was while the uncertainty was around for the energy and below for the divergence.
In the integration case, shown in figure 4, the R-square value was while the uncertainty was around for the energy and below for the divergence.
The accuracy was for the energy and around for the divergence in the single shot case, and it increased to for the energy and for the divergence in the integration case (see Table 3).
In order to perform a distributed energy measurement along the GBS, these results were promising: since the OTR intensity is linearly dependent on the charge and, due to the fact that the GBS bunch charge is , this uncertainty and accuracy results are expected for a beam energy around .
Furthermore, the beam energy has an effect on the OTR intensity and on the angular spread; the appropriate optics must be used in order to perform an accurate fit, putting enough points between the peaks and in the tails (a common rule of thumb is to acquire in the range ). This can be done changing the focal length (a bigger focal length implies a smaller field of view) or the sensor pixel size; in any case, the same optic system guarantees a wide range of energies (i.e. the one used in this experiment has a focal length of and it can measure energies between and but with an increased uncertainty). For lower energy (i.e. ), a smaller focal length must be chosen. In the latter case, the intensity is decreased and an intensifier becomes fundamental.
Finally, if a single shot measurement is needed, the uncertainty doubles with respect to the 1 second integration case both for the energy and the divergence.
4 Conclusion and outlook
The OTR could be a very useful diagnostic tool in order to measure the beam energy. Distributed energy measurements are foreseen especially to evaluate the performances of the accelerating structures at the ELI-GBS facility during the commissioning stage; indeed, the facility will be equipped with OTR diagnostic stations after each accelerating module. This measurement will be useful in particular for the evaluation of the newly designed C-Band structures alesini2017design ().
Furthermore, the energy measurement is foreseen not only for multi-bunch pulses, but also for a single bunch of the pulse train, using a gated camera system (i.e. Hamamatsu Orca4). In this case, the goal is to measure a single bunch within the pulse (i.e. first bunch of the first train, second bunch of the second train, etc.) and to evaluate the effects of the head bunch on the tail bunches; this could be done only in the commissioning stage, since this technique doesn’t have the required resolution to measure the in-spec energy jitter shot to shot.
The energy jitter shot to shot could also be evaluated after plasma interaction if the SNR is high enough (i.e. high energy, high charge). Indeed, the data analysis shows a strong dependence of the uncertainty to the SNR; also the accuracy of the measurement is affected by the SNR.
The main advantages of this technique are the use of diagnostics already in place (OTR screens) and its compactness since no dipole is needed. In case of a high energy jitter, this technique does not require any tuning due to its wide range of applicability ().
The experimental data shows that the uncertainty of the measurement is good enough (around 2%) and that, in the single shot configuration, it doubles with respect to the 1 second integration case. This is useful for plasma accelerated beams (i.e. EuPRAXIA@SPARC_LAB ferrario2017eupraxia ()).
This work was supported by the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 653782.
We wish to give a special thank to the SPARC_LAB group for their contribution with the data acquisition.
- (1) A. Bacci et al., “Electron linac design to drive bright Compton back-scattering gamma-ray sources,” Journal of Applied Physics, vol. 113, no. 19, p. 194508, 2013.
- (2) D. Filippetto et al., “Phase space analysis of velocity bunched beams,” Physical Review Special Topics-Accelerators and Beams, vol. 14, no. 9, p. 092804, 2011.
- (3) A. Mostacci et al., “Chromatic effects in quadrupole scan emittance measurements,” Physical Review Special Topics-Accelerators and Beams, vol. 15, no. 8, p. 082802, 2012.
- (4) A. Cianchi et al., “Six-dimensional measurements of trains of high brightness electron bunches,” Physical Review Special Topics-Accelerators and Beams, vol. 18, no. 8, p. 082804, 2015.
- (5) M. Ferrario et al., “Sparc_lab present and future,” Nuclear Instruments and Methods in Physics Research Section B, vol. 309, pp. 183–188, 2013.
- (6) P. Antici et al., “Laser-driven electron beamlines generated by coupling laser-plasma sources with conventional transport systems,” Journal of Applied Physics, vol. 112, no. 4, p. 044902, 2012.
- (7) A. R. Rossi et al., “The external-injection experiment at the sparc_lab facility,” Nuclear Instruments and Methods in Physics Research Section A, vol. 740, pp. 60–66, 2014.
- (8) D. Alesini et al., “Design of high gradient, high repetition rate damped c-band rf structures,” Physical Review Accelerators and Beams, vol. 20, no. 3, p. 032004, 2017.
- (9) M. Ferrario et al., “Eupraxia@sparc_lab: design study towards a compact fel facility at lnf,” these proceedings.
- (10) K. Nakamura et al., “Broadband single-shot electron spectrometer for gev-class laser-plasma-based accelerators,” Review of Scientific Instruments, vol. 79, no. 5, p. 053301, 2008.
- (11) C. M. Sears et al., “A high resolution, broad energy acceptance spectrometer for laser wakefield acceleration experiments,” Review of Scientific Instruments, vol. 81, no. 7, p. 073304, 2010.
- (12) A. Soloviev et al., “Two-screen single-shot electron spectrometer for laser wakefield accelerated electron beams,” Review of Scientific Instruments, vol. 82, no. 4, p. 043304, 2011.
- (13) Y. Glinec et al., “Absolute calibration for a broad range single shot electron spectrometer,” Review of scientific instruments, vol. 77, no. 10, p. 103301, 2006.
- (14) M. Marongiu et al., “Optical issues for the diagnostic stations for the eli-np compton gamma source,” in 8th Int. Particle Accelerator Conf.(IPAC’17), Copenhagen, Denmark, 2017, pp. 238–241.
- (15) F. Cioeta et al., “Spot size measurements in the eli-np compton gamma source,” in 5th Int. Beam Instrumentation Conf.(IBIC’16), Barcelona, Spain, 2017, pp. 533–536.
- (16) A. Lumpkin, “Time-resolved imaging for the aps linac beams.” Argonne National Lab., IL (US), Tech. Rep., 1998.
- (17) “Yag screen,” http://www.crytur.cz/materials/yagce/.
- (18) V. Ginsburg et al., “Radiation of a uniformly moving electron due to its transition from one medium into another,” Zhurnal eksperimentalnoi i teoreticheskoi fiziki, vol. 16, no. 1, pp. 15–28, 1946.
- (19) L. Wartski et al., “Thin films on linac beams as non-destructive devices for particle beam intenstty, profile, centertng and energy monitors,” IEEE Transactions on Nuclear Science, vol. 22, no. 3, pp. 1552–1557, 1975.
- (20) M. Castellano et al., “Analysis of optical transition radiation emitted by a 1 mev electron beam and its possible use as diagnostic tool,” Nuclear Instruments and Methods in Physics Research Section A, vol. 357, no. 2, pp. 231–237, 1995.
- (21) A. Cianchi et al., “Transverse emittance diagnostics for high brightness electron beams,” Nuclear Instruments and Methods in Physics Research Section A, 2016.