ALICE through the phase transition

30 October 2000

While proton-proton collisions will be the principal diet
of CERN’s LHC machine, heavy-ion collisions will also
be on the menu. The ALICE experiment will be ready
and waiting.

The ALICE collaboration of some 900 physicists is
preparing a detector to study lead-ion collisions at
CERN’s Large Hadron Collider (LHC) starting in 2005.
By observing the results of such collisions, ALICE will
continue CERN’s pioneering investigations through the
phase transition from ordinary nuclear matter to the
state of matter as it is believed to have existed when the
universe was just a few microseconds old.

ALICE detector is being built around the existing
magnet of the L3 detector, which is currently taking its
last data at the Large Electron Positron collider (LEP),
CERN’s current flagship accelerator. ALICE has the
same Russian-doll structure as most collider detectors,
but with the addition of a dedicated spectrometer at one
end. This will have the job of reconstructing the
particles produced in lead-lead collisions that decay into

The overall design of the muon
spectrometer was finalized in 1999, and was the subject
of one of the collaboration’s technical design reports
(TDRs) – the documents that mark the transition from
R&D to production phases for all the LHC’s
experiments. A full-scale prototype of the muon
spectrometer’s momentum-analysing dipole magnet was
completed in April. The spectrometer’s sensitive
elements, 10 multi-wire proportional chambers with a
total of around a million read-out channels, have been
the subject of test-beam studies this

Particle tracking

Another of
ALICE’s TDRs concerns the experiment’s inner
tracking system (ITS). This is the innermost layer of the
detector, responsible for tracking emerging particles
where their density will be at its highest.

physicists have been working with colleagues from
fellow LHC experiment LHCb to develop silicon pixel
chips for the inner two layers of the ITS. The result is a
chip with 50 ¥ 425 mm cells; a prototype detector
based on this chip is being tested this year. The ITS has
six layers, all using silicon technology, and about 10
million digital and 2 million analogue readout channels
to digest the huge number of particles produced in LHC
lead-ion collisions. The collaboration has opted for a
hybrid ITS structure combining sensors, electronics and
mechanical support. Beam tests so far have indicated
that the ITS should achieve position resolution better
than 20 mm.

Surrounding the ITS
is the core of the ALICE detector – its time projection
chamber (TPC). The TPC is a gas-filled detector with an
electric field applied across it. Electrons liberated by
ionization of the gas caused by a passing charged
particle drift in the electric field and are detected at the
end of the chamber. By measuring the arrival time of
these electrons, the TPC reconstructs the path of the
original charged particles. Several of the TPC’s most
crucial elements have been extensively tested over
recent years, and its largest element – a gas-filled
cylinder made of composite structures similar to those
used in space applications – will enter the production
phase early next year.

The TPC in turn will be
inside a transition radiation detector (TRD) that will be
used to identify electrons and positrons. Several physical
processes of interest to lead-ion physics give rise to
electron-positron pairs. These include pairs directly
produced in the initial stages of the collision or pairs
produced as the result of the decays of heavier particles.
The TRD, in conjunction with the ITS and TPC, will be
able to identify the sources of electron pairs. It will work
by measuring the radiation emitted when charged
particles cross the boundary between two media with
different refractive indices.Unlike a conventional “Russian doll” collider
detector, several of ALICE’s subdetectors will cover
only a part of the full solid angle surrounding the
collisions. Among these are the experiment’s
high-momentum particle identification system (HMPID),
its photon spectrometer (PHOS) and four detectors
designed to measure particles emerging very close to
the beam direction – the zero-degree calorimeters
(ZDCs), the CASTOR small-angle calorimeter, the
photon multiplicity detector (PMD), and a
forward-charged particle multiplicity detector

A prototype HMPID was put through its
paces in 1998 using particles produced when a 350 GeV
pion beam from the proton synchrotron struck a
beryllium target. This allowed the collaboration to test
both the detector’s performance and the
pattern-recognition programs that will identify individual
particles. That prototype was then shipped to
Brookhaven in the US where it is now operational in
the STAR detector at the laboratory’s Relativistic Heavy
Ion Collider. ALICE, meanwhile, has started
construction of its full-scale HMPID detector.

PHOS will be made of lead tungstate crystals and will
sample emerging photons over a limited area. This
means that a relatively modest number of 18 000
crystals is needed. These will be supplied from
production plants in China and Russia. For comparison,
the LHC’s CMS experiment, whose electromagnetic
calorimeter is based on the same crystals, will require
around 80 000. By summer 2000, ALICE crystal
production was getting under way with several hundred
crystals having been produced in Russia.

One of
the small-angle detector systems will serve primarily for
triggering purposes – telling the electronics when an
interesting collision has taken place. The ZDCs will be
placed about 100 m away on each side of the main
ALICE detector to measure the energy carried by
particles emerging very close to the beam

The CASTOR calorimeter will be
placed about 16 m away from the interaction point on
the opposite side to the muon spectrometer. It will have
a sandwich structure of quartz fibre planes separated by
tungsten plates. CASTOR’s job will be to search for
exotic particles.

The FMD will consist of a
number of discs, each divided into sensitive pads, placed
at varying distances from the interaction point so as to
cover the largest possible area. It will count the number
of charged particles in the forward region and provide
information for the experiment’s trigger.

PMD will be embedded within the main detector,
attached to the magnet return yoke at 5.8 m from the
interaction point opposite the muon spectrometer. It will
be used primarily with the FMD to measure the ratio of
photons to charged particles emerging on an
event-by-event basis. This will give ALICE physicists
information about event shapes and fluctuations in the
forward region. The PMD has a honeycomb geometry,
the cells of which are 8 mm deep with a surface of
about 1 cm2. Copper walls separate the
cells in order to prevent signals from blowing up by
confining low-energy electrons to a single cell. A 96-cell
prototype has been successfully tested in beams at

Time of flight

Another critical
measurement for ALICE is the time of flight of
emerging particles. Conventional time-of-flight detectors
use fast scintillator detectors with coarse granularity.
For ALICE, where a very fine granularity is required,
this would have presented a very costly option and
several alternatives were studied.

The one that the
collaboration has adopted is multigap resistive plate
chambers (MRPCs). These consist of a series of
gas-filled gaps separated by high-resistivity plates. A
strong electric field across the gaps gives rise to electron
avalanches when charged particles pass through the
chambers. The design of ALICE’s MRPCs involves
optimizing the gap size – a small gap gives a faster
response but a large gap gives a stronger signal.
Multiple gaps allow for a smaller gap size since the
signal can be integrated over several gaps. Extensive
tests in 1999 with varying gap size and using different
material for the resistive plates gave very encouraging
results with time resolution better than 80 ps at more
than 95% efficiency – easily competitive with classical
scintillator detectors. Further tests aimed at finalizing the
detector design are under way.

collaboration is currently putting the finishing touches to
its last few TDRs. With these completed, the entire
detector blueprint will be in place and the collaboration
expects its detector to be in full-scale production by the
end of the year – right on schedule to be ready for the
LHC’s first lead-ion collisions in 2005.

bright-rec iop pub iop-science physcis connect