Calorimeter

The requirement of hermeticity, which is a necessary condition to achieve good resolution of the measurement of the missing transverse momentum, is one of the key design components of the ATLAS calorimeter, which indeed has a coverage up to |⌘| = 4.9. Different technologies are used across different regions in pseudo-rapidity for the different calorimeter sub-detectors as Figure 1.7 shows. In particular the ATLAS Calorimeter is mainly divided in two types of Calorimeters, the Electro-Magnetic (EM) Calorimeter and the Hadronic Calorimeter.

Going radially from the interaction point to the outermost layer of the ATLAS exper-iment, the EM Calorimeter is located soon after the TRT. Over the |⌘| range where the

Figure 1.7: Cut-away view of the ATLAS calorimeter system

calorimeter is surrounding the ID, the EM calorimeter is finely segmented for precision mea-surements of electrons and photons. Those interact via bremsstrahlung process with the calorimeter material, generating a shower of electrons and positrons, detected in liquid Ar.

As the EM Calorimeter needs to provide the energy measurement with good resolution, it needs to have enough material to let the electron-positron shower be fully contained in the EM volume.

The hadronic calorimeter is located at outer radius with respect to the EM one and it is segmented more coarsely since it is mainly aimed at reconstructing jets and at measuring the missing transverse momentum.

An important design criterion came from the need of containing both the electromagnetic and hadronic showers of particles of energies around the TeV scale, since energy escaping the calorimeter results both in a significantly reduced energy resolution and in punch-through into the muon system. For the hadronic interactions the description is provided in terms of the absorption length . is defined as the distance that a particle spent into a material at which the probability of not being absorbed has dropped to 1/e.

The approximately 10 absorption length both in the barrel and in the end-caps are sufficient to provide very good resolution for high energy jets. The total thickness, including the outer support, is 11 at |⌘|=0; this has been shown by simulation and confirmed by test beam data to be sufficient to reduce punch-through into the muon system well below the irreducible level of prompt or in-flight decays into muons. The pseudo-rapidity coverage, granularity and segmentation in depth of the calorimeters are summarized in Table 1.4.

Electromagnetic Calorimeter The Electromagnetic Calorimeter is divided into a barrel (|⌘|<1.475) and two end-caps (1.375<|⌘|<3.2). Each end-cap calorimeter is mechanically divided in two coaxial wheels: an outer wheel covering the region 1.375<|⌘|<2.5 and an inner wheel covering 2.5<|⌘|<3.2. The EM calorimeter is based on a lead-LAr detector

⌘ coverage Granularity ( ⌘⇥ ) EM calorimeter barrel end-cap

Presampler |⌘| <1.54 1.5< |⌘| <1.8 0.025⇥0.1 Sampling 1 |⌘| <1.475 1.375<|⌘| <3.2 0.003⇥0.1^a

0.025⇥0.025^b 0.003 0.025⇥0.1^c

0.1⇥0.1^d Sampling 2 |⌘| <1.475 1.375<|⌘| <3.2 0.025⇥0.025

0.075⇥0.025^b 0.1⇥0.1^d Sampling 3 |⌘| <1.35 1.5< |⌘| <2.5 0.05⇥0.025 Tile calorimeter barrel extended barrel

Sampling 1-2 |⌘|< 1.0 0.8< |⌘| <1.7 0.1⇥0.1 Sampling 3 |⌘|< 1.0 0.8< |⌘| <1.7 0.2⇥0.1 Hadronic end-cap

Sampling 1-4 1.5<|⌘| < 3.2 0.1⇥0.1^e 0.2⇥0.2^d Forward

Sampling 1-3 3.1<|⌘| < 4.9 0.2⇥0.2

a|⌘|<1.4, ^b1.4<|⌘|<1.475,^c1.375 <|⌘|<2.5,^d2.5<|⌘|<3.2, ^e1.5<|⌘|<2.5

Table 1.4: Pseudo-rapidity coverage, longitudinal segmentation and granularity of the AT-LAS calorimeters.

Figure 1.8: A schematic of the barrel Electromagnetic calorimeter. The three layers of the calorimeter and the accordion layout of the lead absorbers and electrodes are shown. Each layer presents a different granularity as summarized in Table 1.4

with accordion-shaped kapton electrodes and lead absorption plates over its full coverage.

The liquid argon was chosen as an active medium because of its intrinsic radiation hardness and good energy resolution. The advantage of the accordion geometry is that it provides complete symmetry without azimuthal cracks. Over the region which is intended to be used for precision physics (|⌘|<2.5) the EM calorimeter is segmented in depth in three sections.

In addition, a presampler is used to recover the energy lost in dead material in front of the calorimeter. The total thickness of the EM calorimeter is 22 X0 in the barrel and 24 X0 in the end-caps The layout of the barrel is shown in Figure 1.8.

The first layer of the calorimeter, called the⌘-strip layer, is finely granulated in⌘in order to allow for a better separation between photons (which results in a single energy deposition) and neutral pions (which results into two very close deposits of energy from the ⇡⁰ ! decay).

The resolution achievable in the barrel EM calorimeter, according to test beam data, is:

(E)

E = 10%

pE(GeV) 0.17% (1.10)

where 10% is the stochastic term and 0.17% is the constant term. The energy response is also linear within ±0.1%. Similar results have been obtained for the end-cap EM calorimeters.

At the transition between the barrel and the end-cap calorimeters, at the boundary between the two cryostats, the amount of material in front of the calorimeter reaches a localized maximum of about 7 X0. For this reason, the region 1.37 < |⌘| < 1.52 is not used for

precision measurements involving photons and electrons.

Hadronic Calorimeters The hadronic calorimeters are subdivided in the tile calorime-ter, whose barrel covers the region |⌘|<1.0 and whose extended barrels cover the region 0.8<|⌘|<1.7, in the LAr hadronic end-cap calorimeters, which extend from |⌘| = 1.5 up to

|⌘| = 3.1 and finally the LAr forward calorimeter, which covers the pseudorapidity range up to |⌘| = 4.9. The tile calorimeter uses steel as the absorber and scintillating tiles as active material. Two sides of the scintillating tiles are read out by wavelength shifting fibres into two separate photomultiplier tubes. The energy response to isolated charged pions of the combined LAr and tile calorimeter was measured with test beams and turns out to be:

(E)

E = 52%

pE(GeV) 3% (1.11)

For the end-cap hadronic calorimeters LAr technology is used, as the EM calorimeter in the barrel region, but copper is used instead of lead as a passive material and a flat-plate geometry was chosen. The energy response to isolated pions can be condensed in the energy resolution:

(E)

E = 71%

pE(GeV) 1.5% (1.12)

Finally, the forward calorimeter is based again on LAr active material and uses copper as passive absorber material for the first layer and tungsten for the second and third layers.

As a result of test beams, the energy response to isolated pions is expressed by:

(E)

E = 94%

pE(GeV) 7.5% (1.13)