Spatial and temporal beam profile monitor with nanosecond resolution for CERN's Linac4 and Superconducting Proton Linac

Abstract

The Linac4, now being developed at CERN, will provide 160-MeV ${\mathrm{H}}^{-}$ beams of high intensity $N=2\times {10}^{14}\mathrm{ions}{\mathrm{s}}^{-1}$. Before this beam can be injected into the CERN Proton Synchrotron Booster or future Superconducting Proton Linac for further acceleration, some sequences of 500-ps-long micro-bunches must be removed from it, using a beam chopper. These bunches, if left in the beam, would fall outside the longitudinal acceptance of the accelerators and make them radioactive. We developed a monitor to measure the time structure and spatial profile of this chopped beam, with respective resolutions $\mathrm{\Delta }t\sim 1\mathrm{ns}$ and $\mathrm{\Delta }x\sim 2\mathrm{mm}$. Its large active area $40\mathrm{mm}\times 40\mathrm{mm}$ and dynamic range also allows investigations of beam halos. The ion beam first struck a carbon foil, and secondary electrons emerging from the foil were accelerated by a series of parallel grid electrodes. These electrons struck a phosphor screen, and the resulting image of the scintillation light was guided to a thermoelectrically cooled, charge-coupled device camera. The time resolution was attained by applying high-voltage pulses of sub-nanosecond rise and fall times to the grids. The monitor has been tested with 700-ps-long UV laser pulses, and a 3-MeV proton beam. Its response over a wide range of beam intensities between $N_{\mathrm{e}}\sim 5$ and $4\times {10}^8$ electrons emitted from the foil per pulse was studied. The monitor can also be used to measure the profiles of antiproton beams in the future facilities of Facility for Low-energy Antiproton Ion Research (FLAIR) or Extra Low Energy Antiproton Ring (ELENA).

Introduction

The planned Linac4 facility [1], [2] of CERN will provide ${\mathrm{H}}^{-}$ beams of energy $E=160\mathrm{MeV}$ and a high intensity $N=2\times {10}^{14}\mathrm{ions}{\mathrm{s}}^{-1}$. This beam will be sequentially accelerated by the CERN Proton Synchrotron Booster (PSB), Proton Synchrotron (PS), and Super Proton Synchrotron (SPS), before being injected into the Large Hadron Collider (LHC) for proton–proton collisions at a center-of-mass energy $E=14\mathrm{TeV}$. Linac4 is important in achieving [3] the highest possible beam current ($I\sim 0.9\mathrm{A}$) in the LHC. The beam intensity delivered to LHC can be further increased by factor $\sim 2$, by replacing the PSB and PS with a 4–5-GeV Superconducting Proton Linac (SPL) [4] and a 50-GeV PS2 [3]. The first section of Linac4, which accelerates the beam to $E=3\mathrm{MeV}$ and chops its time structure (Fig. 1), is now being constructed at CERN. In this paper, we describe a monitor which characterizes the time evolution of the spatial profile of this beam. It is based on the imaging of secondary electrons emitted from a thin target foil.

In the planned design of Linac4 (Table 1, Fig. 1), a ${\mathrm{H}}^{-}$ beam of energy $E=45\mathrm{keV}$ is extracted from a duoplasmatron source and traverses a Low-Energy Beam Transport (LEBT). It is then accelerated to $E=3\mathrm{MeV}$ in a Radiofrequency Quadrupole (RFQ) which is excited at frequency $f_{\mathrm{e}}=352.2\mathrm{MHz}$. The beam exiting the RFQ output consists of macro-pulses of duration $\mathrm{\Delta }{t}_{\mathrm{m}}=400\mathrm{\mu }\mathrm{s}$ and current $I_{\mathrm{m}}=70\mathrm{mA}$ which arrive at a repetition rate $f_{\mathrm{r}}=2\mathrm{Hz}$, corresponding to an average beam current $I_{\mathrm{a}}=I_{\mathrm{m}}\mathrm{\Delta }{t}_{\mathrm{m}}{f}_{\mathrm{r}}\sim 60\mathrm{\mu }\mathrm{A}$. Each macro-pulse consists of a train of 500-ps-long micro-bunches that are spaced by intervals $f_{\mathrm{e}}^{-1}=2.8\mathrm{ns}$. Each macro-pulse and micro-bunch, respectively contain ${10}^{14}$ and ${10}^9$ ${\mathrm{H}}^{-}$ ions (Table 1). This beam is further accelerated to $E=160\mathrm{MeV}$ by traversing the Drift Tube Linac (DTL), Cell-Coupled Drift Tube Linac (CCDTL), and Pi-Mode Structure Linac (PIMS). It is then injected into the PSB using a charge-exchange method [5].

During the acceleration phase for Linac4 injection into PSB, a beam chopper [1], [6] positioned downstream of the RFQ removes 133 consecutive micro-bunches out of every 352 in the beam. It is here crucial to remove all the ions in these bunches, as the ions would otherwise fall outside the longitudinal acceptance of the PSB, strike its inner walls, and activate the accelerator.

The precise control of the transverse beam profile is also important for proper acceleration of the beam. At the high intensities of Linac4, a variety of (e.g., space-charge and resonant oscillation) effects can easily cause halos [7], [8], [9] to form around the core of the beam. These effects will be especially strong during the first stages of the acceleration to $E=3\mathrm{MeV}$ in the RFQ [1]. Simulations [10], [11] predict that the halo at the RFQ exit would constitute $⩾{10}^{-4}$ of the main beam. Unless this halo is properly diagnosed and removed, it will be accelerated with the main beam, eventually striking the walls of the accelerator, and thereby leading to further activation.

The monitor described herein will measure the spatial and time profiles of the 3-MeV beam, to diagnose whether or not they satisfy the above criteria on the beam chopping and halo. For this task a monitor of high time and spatial resolution ($\mathrm{\Delta }t\sim 1\mathrm{ns}$ and $\mathrm{\Delta }x\sim 2\mathrm{mm}$), large dynamic range ($>{10}^5$), and active area $(40\mathrm{mm}\times 40\mathrm{mm})$ is needed. The monitor must be robust, so that it can be routinely operated with minimum maintenance.

If Linac4 is to be used in conjunction with the future SPL, its beam quality must be even higher than that described above in terms of beam halo and timing structure [4]. The chopper frequency will here be increased, so that three consecutive micro-bunches out of every eight are removed. Experiments at the Spallation Neutron Source (SNS) linac of ORNL [9], [12] have revealed that any beam loss at such a facility must be kept below a value equivalent to power $P=1\mathrm{W}$ dissipated per meter of accelerator, if the activation is to be maintained within acceptable limits. Simulations show that this corresponds to less than $\sim {10}^3$ ions lost per micro-bunch, in the halo and chopped part of the Linac4 beam [1]. For the monitor to properly diagnose this, it must detect a weak pulse containing ${10}^3$ ions, without being blinded by the much stronger pulse of ${10}^9$ ions that lies nearby. This is difficult, since the two pulses are here temporally and spatially separated by only $f_{\mathrm{e}}^{-1}=2.8\mathrm{ns}$ and $\sim 10\mathrm{mm}$. In this paper, we also studied whether the monitor can be used in this SPL-quality beam.

Linac4 is in the early stages of construction, and ${\mathrm{H}}^{-}$ beams of comparable energy, intensity, and time structure were not readily available to us. We, therefore, tested the monitor by irradiating the target foil with UV laser pulses of short $(\mathrm{\Delta }t\sim 700\mathrm{ps})$ duration, and imaging the photoelectrons that were emitted, the laser intensity being adjusted to generate the electron flux ($N_{\mathrm{e}}\sim 4\times {10}^8$ emitted per micro-bunch) expected at Linac4. The monitor was also tested against 3-MeV proton beams of low intensity (between $N_{\mathrm{p}}=10$ and $6\times {10}^4$ particles per micro-bunch), which simulated the halo of the Linac4 beam.

Beam monitors based on secondary electron emission from wire electrodes [13] are now used at the Antiproton Decelerator (AD) of CERN to measure the profiles of beam pulses containing $N_{\overline{\mathrm{p}}}\sim {10}^6–{10}^7$ antiprotons, with energies between $E=100\mathrm{keV}$ [14] and 5 MeV [15]. These wire monitors [13] measured the spatial profile by intercepting $\sim 2\%$ of the beam, while allowing most of the antiprotons to pass through without degradation. These devices, however, are unable to measure low beam intensities $N_{\overline{\mathrm{p}}}<{10}^4$. The foil monitor described in this paper has a much higher sensitivity and dynamic range. It may thus be preferable to use them in some beamlines, at the future Extra Low Energy Antiproton Ring (ELENA) [16], [17] at CERN, or Facility for Low-energy Antiproton and Ion Research (FLAIR) [18] at GSI.