Brazilian researchers develop microchip for LHC

The Sampa chip will be used in an experiment that simulates a state of matter thought to have formed in the first microseconds after the Big Bang.

By Elton Alisson

Agência FAPESP – Researchers at the University of São Paulo’s Physics Institute (IF-USP) and Engineering School (Poli-USP), in collaboration with the Aeronautics Technology Institute (ITA) and the Gleb Wataghin Physics Institute at the University of Campinas (IFGW-UNICAMP), both also in São Paulo State, Brazil, are developing a microchip for use in an experiment at the world’s largest particle accelerator, the Large Hadron Collider (LHC), which is operated by the European Organization for Nuclear Research (CERN) in Switzerland.

The second version of the prototype chip, developed as part of the “Design of a signal acquisition and digital processing ASIC for time projection chamber of ALICE experiment” project, supported by FAPESP, will be completed this month (July 2015).

“The idea is to test this second version of the prototype in September and, if all goes well, to start production in 2016,” said Marcelo Gameiro Munhoz, a professor at IF-USP and a participant in the project.

Munhoz told Agência FAPESP that the chip, called Sampa, will be used in ALICE (A Large Ion Collider Experiment). A heavy-ion detector on the LHC ring, ALICE is one of the four large experiments at the CERN facility. It is an international collaboration involving 1,300 scientists at more than 100 research institutions in 30 countries around the world, including IF-USP.

The experiment will be upgraded in the years ahead so that it can be used to study rare phenomena involving particles produced in heavy-ion collisions beginning in 2020, when the collision production rate in the LHC will be increased from 500 hertz (Hz) to approximately 50 kilohertz (kHz).

“The LHC was shut down for a power upgrade in 2013 and is now restarting at a center-of-mass energy of 13 teraelectron-volts, up from 7-8 TeV previously,” Munhoz explained.

“The next planned shutdown is in 2018-19, when the collision rate will again be increased. This will also require an upgrade for ALICE, because without it the experiment’s detection system won’t be able to operate at the higher collision rate.”

One of the changes to be made to ALICE in the years ahead, according to Munhoz, involves the chips integrated into two of the detectors used by the experiment: the Time Projection Chamber (TPC), the main device for tracking and identifying charged particles, and the Muon Chamber, a forward spectrometer that detects muons, particles similar to electrons but 200 times heavier.

To be able to detect the huge number of heavy-ion collisions that will be produced in the LHC beginning in 2020, the chips connected to the TPC and Muon Chamber will need to operate continuously without a trigger, such as that used in the current system to detect events and record them for later analysis.

“The trigger fires a signal when the detector sees a particle collision. Normally the chips connected to the TPC and Muon Chamber start processing and storing data only when they receive this signal,” Munhoz said.

“With the higher collision rate, the chips will have to acquire data continuously, so there won’t be any need for a trigger to tell them when to start operating.”

Multiple functions

According to Munhoz, the Sampa chip is custom designed to read data from gas detectors like the TPC and Muon Chamber.

Each chamber is a cylinder filled with gas. When a particle passes through the cylinder, the gas is ionized as its electrons are stripped out by the particle.

Currently, a sensor located at the extremity of the detectors counts the electrons liberated from the gas and generates a charge pulse that is captured by a set of chips connected to the TPC and the Muon Chamber. The chips amplify and shape this signal.

Next, Munhoz explained, another group of chips transforms the signal into a stream of bits and pre-processes the digitized data to reduce the amount of information to be stored for later analysis by the researchers who participate in the experiment.

“The challenge for Sampa will be to integrate all these functions into a single electronic circuit instead of the several chips that do the work today,” he said. Eighty thousand chips will be produced to instrument the TPC and Muon Chamber detectors, at an estimated cost of US$1 million.

Each silicon chip will measure 9x9 mm and will be made in Taiwan because Brazil has no factories capable of producing chips to Sampa’s specifications. “Integrating several functions into a single chip requires advanced technology to fit everything onto a tiny device,” Munhoz said.

The US$1 million investment to produce the chips will represent 0.5% of the total cost of ALICE, which is projected to reach US$200 million, and will be the Brazilian researchers’ first contribution to the instrumentation for the experiment.

“We joined ALICE in 2006, and since then we’ve had access to the same data as any other member of the collaboration. We’ve also been doing plenty of joint research. But until now we hadn’t been able to collaborate on fabrication of the detectors,” Munhoz said.

The ALICE experiment is designed to study quark-gluon plasma, a state of matter thought to have formed in the first microseconds after the Big Bang, when the Universe was born.

Quarks and gluons are always confined in hadrons (such as protons, neutrons, and pi mesons) and have never been observed in isolation outside them.

In the extremely high temperatures generated by collisions in the LHC, similar to those just after the Big Bang, Munhoz explained, the bonds between quarks and gluons are broken and they are no longer confined in hadrons, enabling scientists to study the phenomenon of confinement, which remains a mystery in physics.

“Because ALICE is more interested in measuring the trajectories of the particles produced in collisions of nuclei, which are hundreds of times more numerous than proton collisions, it generates larger amounts of data per collision than the LHC’s other experiments, like ATLAS or CMS,” he said. In 2012, the ATLAS and CMS experiments proved the existence of the Higgs boson, the particle that explains why some elementary particles have mass.

“The Sampa chip will have to handle the even larger amount of data that will be generated by the experiment beginning in 2020,” Munhoz said.

According to him, other collider experiments are also interested in the Sampa chip. They include the Solenoidal Tracker At RHIC (STAR), one of the experiments attached to the Relativistic Heavy Ion Collider (RHIC) in the United States.

In addition to particle detectors, the chip is being studied for use in other applications, such as producing X-ray images “in color” (to show all wavelengths that are present) and measuring the neutrons emitted by nuclear reactors.

“Sampa is compact, so it’s very useful for instrumentalizing large detectors like those of the ALICE experiment as well as neutron detectors used in neutron imaging,” Munhoz said.