The Advanced Gamma-Tracking array AGATA

1. Introduction

High resolution g-ray spectroscopy using beams from the new GSI fragmentation facility will enable unique nuclear structure studies to be performed. At energies between 50 and 200 A·MeV single step Coulomb excitation and secondary fragmentation can be employed. The first process populates highly excited low-spin states up to the Giant resonances while the second leads to fragments at high spin, 30ħ [Jon98]. By slowing down the beam to energies around the Coulomb barrier the classical reaction types (transfer, deep-inelastic or compound nucleus reactions) become available with beam intensities of >10⁵ particles/s. By decay spectroscopy after implantation very rare beam species down to 10^-3 particles/s can be studied.

The development of a detection system suitable for these challenging experimental conditions started from the Euroball project [Sim97] and evolved into the concept of g-ray tracking as a result of the R&D performed by GRETA [Del99], MARS [Kro00] and the TMR network Gamma Ray Tracking Detectors [Lie01]. The segmented Ge detector arrays Exogam [Sim00], Miniball [Ebe97] and Vega [Ger98], are already going to exploit some of its benefits, but it is with the Advanced GAmma Tracking Array, AGATA, that the new concept is taken to reality. As a result, AGATA will be orders of magnitude more powerful than all current and near-future arrays, overcoming therefore many of their limitations. AGATA will be realized within a European collaboration and is intended to be employed in experimental campaigns at radioactive and stable beam facilities in Europe.

2. Requirements and features

The exotic beams from the new GSI facility impose the most stringent design criteria for the new g-ray spectrometer. They result from limited beam intensities, particularly for the most exotic nuclei, a wide range of beam velocities (from stopped to v/c»50%), high g-ray and particle backgrounds and g-ray multiplicities up to M_g=30, which are typical characteristics of the reactions. A 4p g-ray array with highest efficiency, selectivity and energy resolution is required which is capable of high event rates. The main properties of such an array are summarised in table 1. These features can only be achieved with a close packed arrangement of Ge detectors, i.e. a 4p shell built from large, highly segmented Ge crystals. The individual interaction points of the g quanta have to be disentangled (“tracked”) by tracking algorithms.

Table 1 Basic properties of AGATA

Photo-peak efficiency Peak-to-total ratio Angular resolution Event rates

Pph (Eγ = 1 MeV, Mγ = 1, b<50%) Pph (Eγ = 1 MeV, Mγ = 30, b<50%) Pph (Eγ = 10 MeV, Mγ = 1) P/T (Mγ = 1) Dqg (DE/E<1%) for Mγ = 1 for Mγ = 30

50% 25% 10% 60% < 1° 3 MHz 0.3 MHz

3. Set-up and design

3.1 Array configuration

The geometrical structure of AGATA is based on the geodesic tiling of a sphere with 12 regular pentagons and 180 hexagons. Owing to the symmetries of this specific bucky-ball construction three slightly different irregular hexagons are needed. To minimise inter-detector space while still preserving modularity, three hexagonal crystals (one of each type) are arranged in one cryostat. The pentagonal detectors are individually canned. Each Ge crystal is encapsulated and electronically divided into 36 segments. The inner radius of the array is 17 cm. The total solid angle covered by Ge material is close to 80% and the photo peak efficiency is as high as 50% for individual 1 MeV g rays.

Fig 1. Artist’s view of AGATA. The three slightly different hexagonal crystal shapes are shown in red, green and blue. Three crystals are packed in a cryostat, as indicated by the grey aluminium walls. The total number of such triple cluster detectors is 60. The pentagonal detectors are shown in light blue colour.

The total number of segments in the array is 6780. Together with pulse shape analysis, this provides unprecedented position sensitivity. Realistic simulations of the tracking performance indicate efficiencies of 50% for individual transitions and 25% for a cascade of 30 g rays at 1 MeV. A key feature of AGATA is the high precision for determining the emission direction of the detected g quanta of <1° (corresponding to an effective granularity of >5·10⁴!). This ensures an energy resolution better than 0.5% for transitions emitted by nuclei recoiling at velocities as high as 50% of the speed of light. This value is only a factor of two bigger than the intrinsic resolution of Ge detectors and is comparable with the values currently observed at 10 times smaller recoil velocity. Fig.2 demonstrates the large gain in sensitivity and selectivity obtained by AGATA.

Fig. 2 Simulated g-ray spectra comparing the response of AGATA (top) with that of a conventional Ge-array (bottom) assuming a fragmentation reaction at 100 MeV/A. The enormous gain in resolving power is obvious.

3.2 Detector unit

Fig. 3 shows an AGATA detector module. Each cryostat contains three 36-fold segmented Ge detectors of hexagonal, tapered shape (8 cm diameter before shaping, 10 cm length). The individual Ge crystals are encapsulated in a very thin Aluminium can – a new technology, developed in the framework of the Euroball and Miniball projects, which significantly improves the reliability of the detectors. The 111 preamplifiers consist of a cold part including the FET’s mounted inside the cryostat and a warm part behind the Ge detectors. Highly integrated digital pulse processing electronics are mounted in a second layer behind the preamplifiers. The data are transferred via a fibre-optic channel for further analysis. A central support frame is situated between the preamplifier and the pulse processing-section. The Ge detectors are cooled with liquid nitrogen contained in conical dewars.

Fig. 3 The AGATA Detector Module consisting of (1) three 36-fold segmented Ge detectors, (2) 111 preamplifiers, (3) a frame support, (4) possible position of digital pulse processing electronics, (5) fiber-optics read-out and (6) a LN₂ dewar. The target position (7) is indicated.

3.3 Electronics and Data Acquisition System

The AGATA electronics will work on the principle of sampling the preamplifier signals with fast ADCs to preserve the full signal-shape information. Digital processing will be used to extract energy, timing and spatial information on the position of the interaction points from the sampled data by pulse-shape analysis. It is planned to house part of the processing electronics directly at the detectors. The data will be transferred by high bandwidth fiber links from the experimental area to the data acquisition equipment.

In the AGATA Data Acquisition system the data will be time-stamped and the software triggering will be implemented. The software triggering is very flexible, for instance it collects infrequent events efficiently and will allow the construction of delayed coincidences without dead-time problems. The Event Builder will receive the data packets in parallel from the front-end detector electronics. It will perform all necessary functions of time-ordering, data-merging and gain-matching. User-defined data selection criteria will be applied to reduce the data volume by eliminating unwanted background events. Finally, formatted events will be written to the recording medium with a data rate up to 80 Mbyte/s. Data processing will involve several stages of pipelining and parallelism.

3.4 Tracking analysis

Tracking requires powerful computer algorithms that take into account the physical characteristics of the g-ray interactions in the detector. The development of such algorithms mainly follows two successful lines. In the so-called “clusterisation” method a preliminary identification of interaction-point clusters is followed by a comparison of all possible scattering angles within a cluster against the Compton scattering formula. The second approach starts from those points likely to be the last interaction and goes back, step by step, to the origin of the incident g ray. This "backtracking" method allows, in principle, to disentangle the interaction points of two g rays which enter the detector very close to one another. Furthermore, long-range scattering such as backscattering across the target region may also be recovered. The optimal tracking algorithm may be a combination of cluster recognition and backtracking both including features such as pair production and neutron rejection. For a simulated test case these tracking algorithms achieve a reconstruction efficiency of up to about 60%, depending on the assumed accuracy of the interaction positions in the detector system.

References

[Del99] M.A. Deleplanque et al., Nucl. Instr. Meth. A430 (1999) 292

[Ebe97] J. Eberth et al., Prog.Part.Nucl.Phys. Vol. 38 (1997) 29

[Ger98] J. Gerl et al., VEGA-Proposal, GSI report (1998)

[Jon98] M. de Jong, Nucl.Phys. A628 (1998) 479

[Kro00] Th. Kröll et al., Proceedings Bologna 2000, World Scientific (2001)

[Lie01] R.M. Lieder et al., Nucl. Phys. A682 (2001) 279c

[Pod00] Zs. Podolyák et al., Phys. Lett. B491 (2000) 225

[Sim97] J. Simpson, Z.Phys. A358 (1997) 139

[Sim00] J. Simpson et al., APH N.S. Heavy Ion Physics 11 (2000) 159