Experimental technique for antiproton-nucleus annihilation cross section measurements at low energy

H. Aghai–Khozani1, M. Corradini2,3, R. Hayano4, M. Hori1,4, M. Leali2,3, E. Lodi–Rizzini2, V. Mascagna2,3,a, Y. Murakami4, M. Prest5,6, L. Solazzi7,3, E. Vallazza8, L. Venturelli2,3, and H. Yamada4 1Max-Planck-Institut für Quantenoptik, D-85748 Garching, Germany 2Dipartimento di Ingegneria dell’Informazione, Università degli Studi di Brescia, I-25123 Brescia, Italy 3Istituto Nazionale di Fisica Nucleare Pavia, I-27100 Pavia, Italy 4Department of Physics, University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan 5Dipartimento di Scienza e Alta Tecnologia, Università degli Studi dell’Insubria, I-22100 Como, Italy 6Istituto Nazionale di Fisica Nucleare Milano Bicocca, I-20126 Milano, Italy 7Dipartimento di Ingegneria Meccanica e Industriale, Università degli Studi di Brescia, I-25123 Brescia, Italy 8Istituto Nazionale di Fisica Nucleare Trieste, I-34127 Trieste, Italy


Introduction
Nowdays the only source of antiprotons in the ∼ MeV energy range is the CERN Antiproton Decelerator (AD) [1].The CERN Proton Synchrotron protons are extracted towards the AD target where the emerging antiprotons are selected, collected and injected into the storage ring at 3.5 GeV.Then a deceleration down to 5.3 MeV is performed and eventually the ps are delivered to the experimental areas in a bunch every ∼2 minutes.A bunch lasts 200-400 ns and contains around 3×10 7 antiprotons.Since the first circulation of antiprotons in AD, the ASACUSA Collaboration (Atomic Spectroscopy A vacuum chamber contains the target held by a movable arm.Several scintillation detectors surround the chamber, and the apparatus is put within two movable beam profile monitors based on Gas Electron Multipliers (GEM45 and GEM47).The GEM47 is left on the beam axis position in order to stop the antiproton bunch.
And Collisions Using Slow Anti-protons) has performed experiments in different physics subjects: laser spectroscopy of antiprotonic helium atoms [3][4][5], microwave spectroscopy of antihydrogen ground-state hyper-fine structure [7][8][9], and nuclear collision cross section measurements of p at low energies (∼0.1-5MeV) [10][11][12][13][14][15].The existing experimental data of the annihilation cross section of antinucleons on nuclei show more than one discrepancy in the low energy region.For instance, if the same optical models which well describe the behavior of antineutrons and antiprotons at higher energies is used, the low energy extrapolated values are smaller than the measured ones; another strangeness is that the data for the antineutron seem to be well fitted by a function with a 1/p 2 dependence [6] which is common for low energy antiprotons, or more generally for charged particles (where a focusing effect due to the Coulomb attraction by the target nucleus is expected).
In order to provide useful data to solve these puzzles, the ASACUSA Collaboration has performed the measurement of the antiproton annihilation on carbon nucleus at 5.3 MeV, and this paper describes the experimental apparatus and technique used to achieve this goal.

The apparatus
A scheme of the experimental apparatus is shown in Fig. 1.Two cylindrical chambers (diameters of 120 cm and 60 cm, lengths of 170 cm and 130 cm, respectively) are connected toghether and put between two retractable Gas Electron Multipliers beam monitors (GEM45 and GEM47) used during the beam tuning to align the antiproton beam axis to the apparatus.Moreover, the second one (GEM47) is left in the on-axis position to dump the beam at the end.Two Cherenkov counters, one at the target level, another one at the end of the beamline, are used to monitor the time structure and the intensity of the beam (see Sec. 2.3).

The target system
The two cylinders are made of stainless steel and form a unique vacuum chamber ∼3 m long wich is emptied at 3-4×10 −7 mbar.Inside the chambers 4 movable arms are available, each one supporting a target holder which can be moved both in the radial direction (i.e. in and out of the beam axis) and in the longitudinal one (parallel to the beam axis).The target holders, which consist in rings with an internal diameter of 11 cm and an external one of 13 cm, are made of carbon fiber to match the target material.
The carbon targets are Diamond-like Carbon (DLC) self-supporting foils manufactured by Micromatter 1 .They have been produced with thicknesses of 700 nm and 1000 nm, according to the manufacturer with a precision of 2-3% and a (relative) uniformity of the target surface smaller than 2%.Two out of four of the target holders are used for a target, a third one is kept without any target (for evaluating the background) and the last holder has a different diameter to evaluate the elastic scattering contribution (see Sec. 3).

The scintillating detector
The scintillators surrounding the target chamber consist in 9 planes, 7 with an area of ∼1 m2 and 2 with an area of ∼0.5 m 2 .A variable number of scintillating bars ranging from 30 to 62 bars forms each plane.The bars are provided by FNAL 2 and have a length of 96 cm and a section of 1.5×1.9cm 2 (see Fig. 2).They scintillator material is Polystyrene Dow Styron 663 W + 1% PPO + 0.03% POPOP and each adjustable gain (8 bit).Then the chain is split in two parts: the first one, preserving the information on the analog pulse amplitude, consists of a slow shaper and a sample&hold circuit; the second one is composed by a fast shaper and a discriminator.The analog outputs are multiplexed with a clock of 5 MHz and digitized by the ADC on the board.The digital output, following the electronic scheme already used in the previous cross section measurements by the ASACUSA collaboration [16][17][18][19], is sampled by means of a 300 MHz clock of the FPGAs to obtain the time information of each hit.The FEB can be operated using either the analog ouputs or the digital one, respectively the "analog mode" and the "digital mode".In the present experiment the analog mode is used only for testing purposes, while the actual acquisition scheme makes use of the digital one following the scheme shown in when the signals are respectively below/above the discriminator threshold.The so formed bit string contains the information needed for the analysis, that is the instant the hit occured (starting point of the bit string) and its length (the time the signal is over threshold which is assumed to be proportional to the pulse height).

The beam monitor detectors
Two Cherenkov detectors have been used in two different positions (Fig. 1): the first one at the target level is used to study the time structure of the bunch (the destructive measurement is taken putting the target holder frame on the beam axis in order to completely stop the beam); in order to have a reliable monitor of the p beam intensity on a shot by shot basis, a second detector is put at the end of the beamline.The two Cherenkov detectors are made of lead fluoride crystals with a refractive index of n=1.89.The counters contain 5 crystals each one with a size of 30 mm×30 mm×160 mm.Avalanche photodiodes by Hamamatsu (S8664-1010) with a ∼5 µm depletion layer were used to reduce background events caused by direct hits of charged particles at the depletion layers.

Measurement technique
The measurement technique is sketched in Fig. 5.Each antiproton bunch travels along the beamline crossing the thin carbon target where some antiprotons undergo in-flight annihilation.Despite the fact that this is the physical quantity to be measured, this number is expected to be a tiny fraction of the incoming ps, 10 −6 or 10 −5 depending on the used model.A fraction of antiprotons can also be elastically scattered by the target nuclei towards the lateral walls and eventually annihilate.
In both cases, the emerging annihilation charged products (mainly pions) are detected by the scintillation detectors placed close to the vacuum chamber containing the target, but the detection does not occur at the very same time since the pions from the annihilation of the scattered antiprotons are expected later in time (because of the time-of-flight of the antiprotons from the target to the lateral walls).Given the antiproton speed (3 cm/ns at 5.  Sec. 2), the time delay is at least 20 ns, and this is a crucial parameter for the data analysis.However, the vast majority of the ps cross at all with the target and travel to the end wall undisturbed where a measurement of the bunch intensity is performed by the Cherenkov detector described in Sec. 2 and, at the same time, the scintillating bars counters saturate.To perform the measurement, a fiducial time window corresponding to the first 20 ns of the annihilations in the target region is selected and the hits detected in this window are counted.In order to compute the annihilation cross section, the number of incoming antiprotons must be taken into account.A normalization on a shot by shot basis is done by means of the already mentioned Cherenkov detector (Sec.2), while the absolute number of incoming ps is determined using a second ring-shaped frame (similar to the target support) as shown in Fig. 6.Being both the Coulomb scat- tering cross-section and the fraction of solid angle seen by the center of the target well known, it is possible to calibrate the beam monitor detector (see a more detailed desctiption in [12,20]).The final used formula for the annihilation cross section of antiprotons on carbon is: where N target is the number of events detected when the target is placed on the beam axis, N ring the ones when the "second ring" is added and σ scatt the Coulomb multiple scattering cross section integrated over the solid angle intercepted by the "second ring".

Conclusion
The apparatus and technique to perform the measurement of the annihilation cross section of 5.3 MeV antiprotons on a carbon nucleus have been described.The experiment has been performed at the CERN Antiproton Decelerator in November 2015, the preliminary results have been published [21]; for a full analysis discussione see [22,23].

aFigure 1 .
Figure1.Experimental layout.A vacuum chamber contains the target held by a movable arm.Several scintillation detectors surround the chamber, and the apparatus is put within two movable beam profile monitors based on Gas Electron Multipliers (GEM45 and GEM47).The GEM47 is left on the beam axis position in order to stop the antiproton bunch.

Figure 2 .Figure 3 .
Figure 2. Right top: detail of the edge of one scintillating bar used in the detector: a hole along the bar axis is grooved to host the WLS fiber, while the one visible in the upper part is filled with glue once the fiber has been inserted.Right bottom: the 3×3 cm 2 aluminum plate used to match the fibers with the multianode PMT window.Left: a scintillator plane with 62 bars before closing the detector; the darkened WLS fibers are visibile.

Fig. 4 .Figure 4 .
Figure 4. Details of the digital acquisition scheme.The output consists in a 512 bit string for each channel representing its time-over-threshold status.

Figure 5 .
Figure 5. Scheme of the measurement technique (not to scale): the different moments of the experiment are shown as time increases from the left to the right: (left) the antiproton bunch (50 ns long) is sent into the vacuum chamber, (middle) it crosses the carbon target and the first charged particles arising from the annihilation in the target and in the lateral walls are detected, (right) eventually the bunch arrives at the chamber end.See the text for a more detailed description.

Figure 6 .
Figure 6.Layou of the "second frame" technique to measure the incoming antiproton number.A known fraction of scattered ps is intercepted by the ring and the annihilations are detected (with the ones in the target) in dedicated runs.
3 MeV) and the chamber diameter (60 cm, see