Sie sind auf Seite 1von 11

THE ALICE TPC

JOACHIM BAECHLER European Organization for Nuclear Research, CERN, Department of Physics 1211 Geneva 23, Switzerland

for the ALICE collaboration

The TPC is the main tracking detector in the central region of the ALICE experiment. This paper describes the main components of the ALICE TPC and addresses the specific technological challenges. Furthermore, the results of comprehensive system tests are summarized.

1.

Introduction

ALICE [1] is one of the four experiments presently under construction for the Large Hadron Collider (LHC) at CERN in Geneva. ALICE is a general-purpose heavy-ion experiment designed to study the physics of strongly interacting nuclear matter and the quark-gluon plasma in nucleus-nucleus collisions. At a maximum cms energy of 5.5 TeV per nucleon in Pb-Pb collisions, up to 8000 charged particles per unit rapidity are produced within a central collision. Owing to the enormous particle multiplicity per event, very specific requirements are made on the performance of the detectors, the electronics and the data acquisition. The ALICE TPC, operated in a solenoid of 0.5 T, is the main detector of the ALICE experiment for tracking, momentum measurement and particle identification of charged particles. In conjunction with the Transition Radiation Detector (TRD), the Inner Tracking System (ITS), Cherenkov counters (HMPID) and Multi-gap Resistive Plate Chambers for time of flight (TOF), the TPC will provide identification of leptonic and hadronic particles in the momentum range from 0.5 to 10 GeV/c.

1

2

The specific requirements on the TPC for hadronic physics are :

Good two track separation

dE/dx resolution of better than 10%

Track matching with 85-95% efficiency, using inner and outer detectors.

In combination with the ITS and the TRD, the TPC needs to satisfy the following criteria for electron physics:

Track finding efficiency for tracks with p t > 1GeV/c of > 90% .

Momentum resolution for electrons of 2 % with a momentum of ~ 4GeV/c.

Rate capability for central collisions to be at least 200Hz.

1.1. TPC – Main Components

The ALICE TPC is composed of a cylindrical gas volume (barrel), divided into two half volumes of equal size, separated by a 30μm thick HV electrode to generate the drift field. It has an inner radius of about 80 cm and an outer radius of about 280 cm, with an overall length of 500 cm in the beam direction (Figures1 and 2). At both ends of the barrel, conventional multi-wire proportional chambers with pad readout are mounted into end plates with 18 trapezoidal sectors each.

mounted into end plates with 18 trapezoidal sectors each. Figure 1. Layout of the ALICE TPC

Figure 1. Layout of the ALICE TPC showing the central electrode, the field cage and the end plates supporting the readout chambers.

3

The front-end (FE) electronic cards with an optical interface to the DAQ are connected to the backplane of the readout chambers via ~8cm long flat Kapton cables.

of the readout chambers via ~8cm long flat Kapton cables. Figure 2. Photography of the ALICE

Figure 2. Photography of the ALICE TPC positioned in the clean room with readout chambers mounted, in preparation for electronics installation.

1.2. Field Cage

To keep the structural material (mass) at a minimum, the field cage is composed of two outer cylinders (outer containment vessel, outer field cage vessel) and two inner cylinders (inner field cage vessel, inner containment vessel). The drift gas is confined in the outer and inner field cage vessel; the volume created between the containment- and field cage vessels is filled with CO 2 and serves as an electrical insulator. At the center of the of the field cage the central electrode of 30μm thickness is strung and mechanically supported by the outer and inner field cage cylinders, to generate the drift field in the two half volumes. Two opposite axial potential degraders with potential strips create a uniform drift field on either side of the central electrode. A unique feature of this field cage is the method to remove the potential strips from the field cage surface and wind them instead around support rods held away from the surface. This avoids instabilities and discharges over the huge surface of the field cage cylinder walls.

4

In total, 18 such axial rods, in line with the 18 sectors of the endplate, support the potential strips over a length of 250 cm in either direction. This is shown in Figures 3 and 4.

cm in either direction. This is shown in Figures 3 and 4. Figures 3 and 4.
cm in either direction. This is shown in Figures 3 and 4. Figures 3 and 4.

Figures 3 and 4. TPC field cage prototype with the six support rods and potential strips (left), and the ALICE TPC field cage (right) with 18 support rods plus strips; the TPC is mounted vertically on one endplate with an opening for a readout chamber.

To ensure minimal multiple scattering and low secondary particle production, the material budget of the TPC field cage components is kept as low as possible. Furthermore, the oxygen content of the drift gas has to be maintained below 5ppm in order to assure good detector efficiency. Consequently, we use, wherever possible, only low-mass composite material providing sufficient mechanical rigidity and the required leak tightness. After assembly, the field cage was leak-tested with dummy chambers in the end plates. An oxygen level of 5ppm was measured with one volume exchange of gas per day. It is a major challenge to guarantee the mechanical stability and mounting precision of 250µm for the central electrode and the readout planes within the entire assembly. With detailed photogrammetry, before and after chamber installation, it was shown that the entire detector stayed within the required mechanical tolerances.

1.3. Readout Chambers

The ALICE TPC readout [1] is based on conventional multi-wire proportional counters with cathode pad readout. The segmentation of the readout pad plane was chosen to optimize, in the high multiplicity environment of central Pb-Pb collisions, the momentum and dE/dx resolution. The radial dependence of the track density led to two different types of readout chambers and a radial segmentation of the readout plane. The active area varies radially from 84.1 to 132.1 cm and from 134.6 to 246.6 cm for the inner and outer chamber, respectively. The total active area is about 32.5 m 2 . The readout

5

chambers are made of standard wire planes, i.e. a grid of anode wires above the pad plane, followed by a cathode plane and gating wire plane facing the drift volume. To account for the different pad sizes chosen, the wire geometry is different for the inner and outer readout chambers. The general wire layout was tuned with Garfield (electrostatic simulation program [2]) to obtain the maximum charge coupling to the pads. As a consequence, the field wires between the anode wires were suppressed. The wire pad geometry was optimized [2] based on the specific event topology of the particle spectra and densities as they are expected to occur in central Pb-Pb collision. This optimization procedure led to three different pad sizes of 4 x 7.5 mm 2 in the inner readout chambers and 6 x 10 mm 2 and 6 x 15mm 2 in the outer readout chambers.

1.4. Gas Mixture

The requirements on momentum and position resolution in the high multiplicity environment, as well as on minimum interaction processes inside the detector, determined the choice of gas for the ALICE TPC.

detector, determined the choice of gas for the ALICE TPC. Figure 4. Comparison of gas gain
detector, determined the choice of gas for the ALICE TPC. Figure 4. Comparison of gas gain
detector, determined the choice of gas for the ALICE TPC. Figure 4. Comparison of gas gain
detector, determined the choice of gas for the ALICE TPC. Figure 4. Comparison of gas gain
detector, determined the choice of gas for the ALICE TPC. Figure 4. Comparison of gas gain
detector, determined the choice of gas for the ALICE TPC. Figure 4. Comparison of gas gain
detector, determined the choice of gas for the ALICE TPC. Figure 4. Comparison of gas gain

Figure 4. Comparison of gas gain as a function of the anode wire voltage for a Ne-CO 2 and a Ne- CO 2 -N 2 gas mixture. The N 2 admixture allows a stable operation with a higher gain.

The Ne-CO 2 (90%-10%) mixture provides low diffusion, low Z and large ion mobility. It was therefore chosen as the optimum gas mixture for our application. From further optimization work [3] it was found that a small addition of nitrogen (to improve quenching) to the Ne-CO 2 gas, results in higher operational stability of the readout chamber at the expense of a tolerable reduction in drift speed for constant E/p (Figure 4).

6

The physics results presented here were obtained with a Ne-CO 2 -N 2 (85.7-9.5-4.8) mixture. Unfortunately, the drift velocities in both the Ne-CO 2 and Ne-CO 2 -N 2 gas mixtures are not saturated at E-fields of 400 V/cm. Therefore, small variations in temperature and gas mixture have a significant bearing on the drift velocity and herewith produce distortions in the space coordinates. To limit these distortions to the intrinsic space resolution of the TPC of < 1000μm (limited by diffusion) the temperature gradient [4] needs to be smaller than 0.1K. These constraints require that the TPC be operated at high drift fields of > 400V/cm to achieve acceptable drift times of 88μs. Therefore, the HV on the central electrode is 100 kV.

1.5. Front-End Electronics

The front-end electronics [5] has to read out the charge detected by the 570132 pads from the cathode plane of the readout chambers. Each single pad is red out by a charge sensitive preamp/shaper, a 10-bit 10 MHz low power ADC, and an ASIC that contains a digital filter for tail cancellation and baseline subtraction as well as zero-suppression circuits and a multi event buffer. The charge induced on a single pad is amplified and integrated via a low input impedance amplifier. The continuously sensitive charge amplifier is followed by a semi-gaussian pulse shaper of second order (Figure 5). The output of the amplifier/shaper chip is fed into the ALICE TPC Read Out (ALTRO) chip containing 16 channels. Upon arrival of the first level trigger the data stream is stored in a memory. Once the second level trigger (“accept” or “reject”) is received, the latest data stream will either be frozen in the data memory, until its complete readout has taken place, or discarded.

until its complete readout has taken place, or discarded. Figure 5. Schematic of the ALICE TPC

Figure 5. Schematic of the ALICE TPC readout electronics

7

The ALTRO chip contains 16 times the TSA 1001 ADC from ST Microelectronics in the form of 8 ADC IP blocks. This ADC has differential inputs and can work up to a clock frequency of 20 MHz with a resolution of 10 bits, while consuming only 12 mW. After digitization a Baseline Correction Unit removes systematic perturbations of the baseline by subtracting a pattern stored in memory. Between triggers, a self-calibration feature can correct any baseline shifts due to temperature shifts or low frequency perturbations. The tail cancellation filter is a three-stage IIR digital filter that removes the long complex tail of the detector signal, thus narrowing the charge clusters to improve identification. It also removes the undershoot that typically distorts the signal amplitude when pile-up occurs. A second correction of the baseline is performed based on the moving average of the samples that fall within a certain acceptance window. The algorithm removes non-systematic perturbations of the baseline. Zero suppression finally removes all the data that is below a certain threshold, except for a specified number of pre- and post-samples around each peak.

1.6. TPC Test Facility

We have set up a comprehensive test facility in which every individual component of the ALICE TPC can be tested, either separately or jointly with other parts of the detector system. A cosmic ray scintillator array allowed to trigger on single particles or on cosmic showers. Figure 6 shows the signals generated on a single pad by a cosmic shower event and the effect of the ALTRO filter on the baseline. Since the local occupancy is similar to that in global central Pb-Pb collisions a detailed study and optimization of the baseline filters could be performed. In spring 2004, a final “integral test” was performed in the PS-T10 test beam facility at CERN. An inner readout chamber (IROC) with the final electronics – DAQ chain was operated using the newly proposed gas mixture. The purpose of this test was to check the performance of the detector in combination with the entire readout chain, including the readout software. Thanks to the smooth and stable operation of the complete system, we were able to extract from the data the Bethe-Bloch-curve in a momentum range from 1.0 up to 7GeV/c.

8

8 Figure 6: High-multiplicity cosmic ray event in the prototype TPC. Signal on a single pad,
8 Figure 6: High-multiplicity cosmic ray event in the prototype TPC. Signal on a single pad,

Figure 6: High-multiplicity cosmic ray event in the prototype TPC. Signal on a single pad, before ALTRO action (upper), and after ALTRO action (lower).

Because of the high performance DAQ system with a data throughput of 100 MB/s it was possible to take events even without zero suppression allowing the study of the original signal characteristics with sufficient statistics. In total we have accumulated about 6’000 non-zero suppressed events for each momentum setting.

650 650 650 600 600 600 550 550 550 500 500 500 450 450 450
650
650 650
600
600 600
550 550
550
500 500
500
450
450 450
400 400
400
350 350
350
300
300
300
0 0
0
20
20
20
40
40
40
60
60
60
80
80
80
100
100
100
120
120
120
140
140
140
L
L L
(cm)
(cm)
(cm)
drift
drift
drift
Figure 7: Space point resolution in the pad plane (x-y) as a function of drift length.
σ
σ σ
m)μ(
m)μ(
m)μ(
φr
φr
φr

m)μ(

σ z

9

800 700 600 500 400 300 0 20 40 60 80 100 120 140 (cm)
800
700
600
500
400
300
0
20
40
60
80
100
120
140
(cm)
L drift

Figure 8: Space point resolution in the z direction (drift direction) as a function of drift distance.

Furthermore, we have successfully tested the ALTRO baseline restoration algorithm and could compare online-corrected data with uncorrected ones and hence prove the reliability of the online data reduction. The space resolution in the pad plane (x-y) and the drift direction (z) is shown in Figures 7 and 8, respectively.

direction (z) is shown in Figures 7 and 8, respectively. Figure 9. dE/dx beam data: Triangles

Figure 9. dE/dx beam data: Triangles correspond to measurements with the three-component mixture Ne-CO 2 -N 2 . The solid line is the mean energy loss curve (for Ar-CH 4 (90/10) [1]), superimposed on the data.

10

The results of the dE/dx measurements obtained in 2004 with the new three-component gas mixture Ne-CO 2 -N 2 (85.7-9.5-4.8) during the test beam run at CERN are shown in Figure 9.

Conclusions

The ALICE TPC was designed to cope with the extreme instantaneous particle densities produced in heavy-ion collisions at the LHC. Although conventional in its concept, this detector system incorporates innovative and state of the art technologies, from the mechanical structure to the readout and data processing chain. Carefully optimized components and extensive test programs performed in parallel with production are to ensure that the TPC can be operated in a stable and safe mode. Integral leak tests with argon, for example, have demonstrated the excellent gas tightness of the entire field cage system allowing the running of the TPC with less than 5ppm oxygen at nominal gas flow. On the other hand, extensive surveys of the TPC after chamber installation showed mechanical tolerances of the field cage elements being below 200µm with respect to the readout chambers. The integral tests performed in 2004 with the TPC system demonstrated that the complete electronics and DAQ chain, in combination with an Inner Readout Chamber, work according to specification [6]. Furthermore, the dE/dx curve taken over an extensive momentum range was derived from the newly proposed three-component gas mixture Ne-CO 2 -N 2 . The dE/dx resolution was found to be better than 9% for the inner readout chamber, the position resolution in the pad- and drift direction confirmed the design values. The comparison of the measured energy loss [6] with the predictions from the models developed by Bichsel [7] is still underway. The understanding thereof could potentially lead to an improvement of the particle identification capability of the TPC.

11

References

1. ALICE Collaboration, ALICE TDR 7, CERN/LHCC 2000-001

2. R. Veenhof, “Calculations for the ALICE TPC Read-Out”,

3. C. Garabatos, “The ALICE TPC” , Proc. 10th International Vienna

Conference on Instrumentation, Vienna, Austria, NIM in physics research, Section A, Accelerators, Spectrometers, detectors and associated equipment 531

(1):197-200.

4. T. C. Meyer (ALICE collaboration) 2004, Proc. Of the 42nd Workshop of the

INFN ELOISATRON Project , Innovative Detectors For Super-Colliders, Erice, Italy, 28 Sept -4 Oct 2003.

5. L. Musa et al., The ALICE TPC Front End Electronics, Proc. of the IEEE Nuclear Science Symposium, October 2003, Portland

6. P. Christiansen et al.,``Performance Test of the ALICE TPC IROC Module at

the PS'', ALICE-INT-2005-029.

7. H. Bichsel, “Particle identification at STAR-TPC with ionization

measurements”, prepared for 8 th International Conference on Advanced Technology and Particle Physics (ICATPP 2003): Astroparticle, Particle, Space Physics, Detectors and Medical Physics Applications, Como, Italy, 6-10 Oct. 2003.