At the frontier of human genius

An underground ring 27 km in circumference, running between Switzerland and France, within which the LHC should make it possible to recreate the conditions that occur at the origin of matter. © CERN
An underground ring 27 km in circumference, running between Switzerland and France, within which the LHC should make it possible to recreate the conditions that occur at the origin of matter. © CERN
Simulation of the disintegration of a four-muon Higgs boson isolated in the CMS detector of the LHC at CERN. The tracks indicate that the particles were produced by the collision of a pair of very highprecision energy protons. The energy deposited by the particles in the detector is shown in blue. © CERN
Simulation of the disintegration of a four-muon Higgs boson isolated in the CMS detector of the LHC at CERN. The tracks indicate that the particles were produced by the collision of a pair of very highprecision energy protons. The energy deposited by the particles in the detector is shown in blue. © CERN
ALICE detector © CERN
ALICE detector © CERN
10 September 2008, awaiting the launch of the LHC. The successful operation was disrupted nine days later. © CERN
10 September 2008, awaiting the launch of the LHC. The successful operation was disrupted nine days later. © CERN

After 20 years of development, the Large Hadron Collider (LHC) was launched on 10 September 2008. Just nine days later, a faulty electrical connection resulted in helium escape and a forced shutdown. It could be the summer of 2009 or even later before it starts up again. The excitement among physicists remains palpable nevertheless, as we discovered behind the scenes at the site of this new jewel of high technology.

With 10 000 scientists and more than 80 contributing countries, the European Organisation for Nuclear Research (CERN) leads the way in the field of particle physics. Its latest marvel, the Large Hadron Collider - LHC - almost defies superlatives. The 27km ring, site of future 14 TeV(1) collisions, makes it the largest and most powerful particle accelerator in the world.

It is within this structure that protons will collide until they produce the conditions needed for the creation of the primary matter of which the universe was made, some 13.7 billion years ago. As the media often reports, Higgs boson (2) is the most eagerly awaited star in this scientific saga. Demonstrating its existence - as first evoked by Brout and Englert in the 1960s and then independently by Higgs - would not only be a major scientific develop - ment but also a technological feat that would owe a great deal to the creativity and excellence of the teams at CERN.

The magnet that came in from the cold

Upon entering the LHC, particles have a relatively low energy level (0.45 TeV) which then increases with each revolution as a result interacting with a magnetic field. It takes more than a million revolutions to arrive at the maximum energy of 7 TeV, something approaching the speed of light. But how, over a distance of approximately 10 billion kilometres, is it possible to exercise perfect control over their trajectory? This is the task of around 1 600 superconductor magnets which produce a cons - training magnetic field that develops along with the particle energy. ‘Superconductors' mean that the cables that form the magnet coils give no resistance to the current that passes through them If they did, the LHC's electricity demands would be 40 times greater than they are. These electrical currents produce the magnetic fields required to maintain the particles in orbit. As Roberto Saban, LHC's head of hardware commissioning, explains: "For the niobium-titanium alloy used in making these cables to become superconducting, the magnets must be brought to and maintained at a temperature of 1.9 K (or -271 °C). That makes the LHC the coldest place in the universe!" To achieve this, the magnets are plunged into a superfluid helium bath. This superfluidity enables the heat to be evacuated as efficiently as possible. The magnets are surrounded by cryostats - akin to steel vacuum flasks - that insulate them from the ambient heat. But superconductivity is not a stable state and it can disappear as a result of local heating of the cable. As the energy stored in each sector is enough to heat and melt a 1.5-tonne block of copper, electronic systems monitor the magnets to detect any resistive transition and, if they do, trigger the magnet protection procedure.

Spy materials

Bundles of particles flow in opposing directions within an almost perfect vacuum. As they collide they produce a myriad of elementary particles. To identify them, the collisions occur within four huge detectors: ALICE, ATLAS, CMS and LHCb.

The onion-like structure of each detector is made up of several layers, each designed for carrying out specific tasks. Many technologies are concentrated in these strata. As Yves Schultz, head of the ALICE experiment, explains: "Closest to the point of impact, the trajectograph consists of silicon sensors that are the result of many years of development.

A single 1.8 cm² chip contains over 8 000 pixels that are linked to miniature cables. This degree of granularity makes it possible to identify the track of each of the particles created in the collision. To avoid the information gathered being polluted by background electromagnetic noise, the signals are converted into light pulses that are conducted by optical fibre to the computer acquisition systems. In all, more than 10 million cells sensitive to the particle passage surround the point of impact at a distance of under 50 cm." ATLAS and CMS, the two multifunctional LHC giants, are also packed full of technological marvels. "In addition to the trajectograph there is also usually an electromagnetic calorimeter that measures the energy of the emitted electrons and photons," says Daniel Denegri, who has headed the CMS experiment over the past 10 years. "In our case, this role is assured principally by approximately 80 000 lead tungstate crystals. 10 years of effort went into their production and delivery. You can imagine the painstaking work involved in subsequently assembling this immense chandelier."

Not just Europe

"CERN could not have taken on this task alone," continues Daniel Denegri. "Around 10 countries were involved in developing the electromagnetic calorimeter: Russia and China for the crystals, the United States, France, Italy and Switzerland for the electronics and mechanics. In fact, the LHC became a global project because although CERN - through the intermediary of the Member States - invested 6 billion Swiss francs in the LHC, 10 % of the total cost was borne by non-Member States.

Everybody benefits from it. We have the necessary funds to build structures more rapidly and the institutions in the contributing countries benefit from access to the experiments." Currently, non-EU researchers make up around a third of the CERN staff. "We of course adapt our demands in line with the expertise of each Member State. We recently concluded agreements with Egypt to manufacture components with a technology that requires no prior research. This international cooperation also gives us access to some very interesting recycled materials. Part of the iron used by the CMS, for example, is taken from former Russian vessels and the brass used in the hadronic calorimeter comes from artillery cartridges of the Russian fleet! Perhaps best of all, the detector stands on earthquake-resistant feet made in the United States and rests on a structure coming from Japan, with the result that the detector can withstand earthquakes of up to 7 on the Richter scale."

Painstaking work

Patrick Fassnacht, who heads the ATLAS detector, points out that each component must meet the same quality criteria irrespective of its origin. "Our experiment is equipped with several thousand muon chambers. Whether they come from Israel or Germany, they must present similar technical performances." He also stresses the huge contrast between the size of the instruments and the precision sought. "You have to remember that ATLAS weighs 7 000 tonnes and each instrument has to be positioned very precisely, down to the very last millimetre. The technicians, sometimes working more than 30 metres above the ground, fit devices aimed at achieving a precision measured in terms of the micron or nanosecond!

We also have exceptionally difficult constraints. If there is a fault at the heart of the detector we must be able to gain access to correct it. That is why our engineers have designed a totally mobile architecture that enables each element to be moved on air cushions.

These hundreds of tonnes of equipment are also moved while remaining connected to utilities (electricity, gas, etc.), especially the cyrogenic system for the magnets and calorimeters, which must avoid any heating and therefore any expansion of the enclosed gases. Otherwise we would be forced to allow them to escape - and replacing them would generate major costs."

100 000 processors

With approximately 40 million collisions a second, CERN physicists are unlikely to ever run out of data to analyse. Fortunately, because the scientists know approximately what they are looking for they have developed automatic data acquisition systems that are activated immediately when a potentially interesting event is detected. In this way the information stored will be ‘limited' to the equivalent of 3 million CD-ROM disks a year. "CERN has neither the financial means nor the space to hold the 100 000 processors needed to annually process 15 million gigabytes," stresses Frédéric Hemmer, deputy chief of CERN's IT department. "This is why, along with our partners, we created the LHC Computing Grid, which enables us to share the work between 150 institutions in 40 countries." This international structure combines the calculating power of thousands of computers.

Upstream, at CERN itself, is Tier-0. This transmits data to the 11 national centres of Tier-1 that are available continuously and guarantee permanent recording. After pre-processing these centres redistribute the information to Tier-2 and Tier-3 where researchers are charged with processing the data. "The end-user does not see the complexity of the system as he has access to all data, whatever its location. This latter point is both the strength and the weakness of the Grid," explains Frédéric Hemmer, "because the scientist is not always aware of the number of calculations or transfers that his research will generate. We are, however, very confident as the network is continuing to expand. A number of private companies are already offering us access to their processors when they are operating at under-capacity."

>All of that... for what?

So what is the ultimate purpose of this high-tech infrastructure? To verify fundamental theoretical models and, as the icing on the cake, to discover the famous Higgs boson. "But unless there is a sudden breakthrough, I expect another two or three years before we achieve results of this kind," predicts physicist Patrick Fassnacht. "We will first verify what we know already and then after about a year when the detector performances have been mastered we will get to grips with some of the finer elements of physics. Hadronic machines(3) such as the LHC have a great potential for discovering new information, but it is the old electron accelerators such as the LEP (Large Electron-Positron Collider) that permit a more detailed validation of theories.

This is why the successor to the LHC will certainly be an electron-positron collider. We must forget the days when we made two or three revolutionary discoveries a year. We have now reached such levels of complexity that we must learn to be patient."

Marie-Françoise Lefèvre

  1. 1 TeV (= 1 teraelectronvolt = 1012 eV = 1.6x10-7 joules) corresponds to approximately the energy of a mosquito in flight. It is applied here to a proton with a mass a billion times smaller.
  2. See the special issue of RTD info, February 2007, A matter of life.
  3. The hadrons are a class of elementary particles that are ranked among the protons.


Read more

Research in everyday life

CERN pursues an active communication policy on several fronts - from comics on the ALICE experiment to the most advanced scientific texts on the development of detectors. But the knowledge transfer does not end there, as the applications that stem from these years of scientific exploration are far from being limited to particle physics. In medicine, a number of techniques have benefited from the knowledge acquired. Hadrotherapy uses carbon ions and protons for the treatment of tumours, while positron-emission tomography has become a commonplace scanner technique.

Many industrial partners are also improving their production processes by applying the quality control measures required by CERN.

The World Wide Web, which has revolutionised society, was created by an engineer working at CERN in response to the need for its physicists to exchange large quantities of information in different formats. The LHC Computing Grid has also had its imitators: this pooling of computing power is used by biochemistry, for example, to test the interaction between complex molecules and proteins.

Processors ‘experiment' on the basis of theoretical models to arrive at every possible molecular combination, thereby reducing the need for in vitro experiments. In regard to avian flu alone, 20% of the compounds generated in this way have an effectiveness superior to Tamiflu.