From PGSC Wiki
Jump to: navigation, search

The Alpha Magnetic Spectrometer was conceived by Sam Ting to search for dark matter, the origin of cosmic rays, and to measure antinuclei [1]. A prototype, AMS-01, was flown on the space shuttle Discovery during flight STS-91 to gather background data, test detector performance, and adjust operating parameters. The data gathered during this flight was used to search for antihelium nuclei.

The flight began on 3 June 1998 and continued for 184 hours. The shuttle altitude varied between 320 and 390 km and the latitude was between ± 51.7°. Except during the docking with the MIR space station and the descent, the z-axis of the detector was kept pointing within 45° of the zenith.

Detector Design[edit]

The AMS is the first large acceptance magnetic spectrometer to operate in outer space, which is background free, to conduct a high statistics study of cosmic particles [2]. The detector consists of a permanent magnet, tracker, time of flight hodoscopes, Cerenkov counter, and anticoincidence counters. Every component of the detector had to undergo space qualification tests including acceleration, vibration, thermal vacuum, electromagnetic interference and radiation. The magnet was tested up to 17.7g in a centrifuge, and electronics were tested in heavy ion beams.


The magnet was made from 1.9 tons of Nd-Fe-B in the shape of a cylindrical shell with inner diameter 1115 mm and length 800 mm. The center of the magnetic field had strength 0.14 Tesla and the analyzing power BL2 was 0.14 Tm2.


The tracker was made of six planes of double sided silicon microstrip detectors. This device measured the rigidity by measuring the deflection and the magnitude of the charge by measuring multiple energy loss. The tracker volume consisted of less than 3% of a radiation length of material. The resolution of the tracker was 10 μm in the bending direction and 30 &mum; in the other directions.


The time-of-flight hodoscope consists of 4 layers of scintillator paddles. Each layer has 14 paddles, which measure 110 mm in width, 10 mm in thickness, and between 720 and 1360 mm in length. The paddles overlap by 5mm to improve hermiticity. There are two layers above and two layers below the magnet, and the layers on either side are orthogonal to each other. The TOF can determine the magnitude of the charge using the pulse height in addition to the time of flight measurement. The TOF has a timing resolution of 105 psec which results in a Β resolution of 2.4%.

Anti-Coincidence Counter[edit]

The Anti-Coincidence scintillation Counters (ACC) cover the inner surface of the magnet to reject particles that pass through the magnet walls or support structures.

Data Analysis[edit]

Event Reconstruction[edit]

Using the TOF, the direction of the particle and the velocity Β were measured. The particle rigidity, R, is measured by calculating the deflection in the magnetic field, which requires at least four hits in the tracker and the agreement of the two tracking algorithms. The magnitude of the charge was determined using dE/dx in the TOF and tracker.

Background Rejection[edit]

The backgrounds to the antihelium search are electrons, protons, and helium nuclei. The detector response to these backgrounds was studying using in flight data, test beam data, and Monte Carlo studies. The first rejection point was to select events with |Z| = 2 to reduce the contamination from |Z| = 1 events that might look like higher charge events. The probability of the wrong charge magnitude being assigned by the TOF and the tracker was less than 10-7.

The second rejection factor is to determine the sign of the charge. This is clearly designed to distinguish helium from antihelium. To determine the sign, the particle direction is determined from the TOF. This information, combined with the deflection, allows the charge sign determination. AMS always assigned the direction correctly. To make this determination, it was important to remove tracks with kinks, which result from a nuclear recoil. These kinks are removed by calculating the rigidity in the first and last three points separately and cutting if the difference is large. Events with collinear delta rays can shift single points causing the rigidity to be calculated incorrectly. These events are removed by rejecting an excess of energy within 5 mm of the track.

Finally, a compatibility function from measurements of velocity, rigidy, and energy loss was constructed for helium and antihelium. A cut on low compatibility removes the remaining background. After all of the cuts, AMS-01 found 2.86 x 106 helium events up to a rigidity of 140 GV. No antihelium events were observed.


The AMS-01 measurement puts a limit on the ratio of antihelium to helium flux. The incident rigidity spectrum for the helium is extracted from the measured rigidity spectrum by deconvoluting it with a resolution function that was created using GEANT to simulate the detector response. Another small correction for the difference between absorption cross sections was made. By assuming the antihelium rigidity spectrum has the same shape as helium, the ratio of antihelium to helium flux was found to be less than 1.1 x 10-6.


1. J. Alcaraz, et al. Physics Letters B 461 (1999) 387-396.

2. M. Alguilar et al. Physics Reports 366 (2002) 331-405.