Jörg Schwartz, firstname.lastname@example.org
GENEVA – News coverage of the Large Hadron Collider (LHC) showed severe signs of overheating as its start time approached in September. Some tabloids predicted that the world would be sucked into a black hole created by the €3 billion physics supermachine, which has been set up to allow researchers to experimentally re-create – and understand – the conditions that existed just after our universe was created in the big bang. The theory is that this can be done by having two beams of subatomic particles called “hadrons” – either protons or lead ions – colliding head-on at very high energy.
The hadrons gain that energy when they travel lap by lap in opposite directions inside a circular accelerator, built into caverns 100 m underground near Geneva. The collisions will create the smallest known nuclear particles, the universe’s fundamental building blocks. Additionally, a wide range of other radiation is generated, but is well shielded from the public by the rocks of the Alps.
However, shortly after the European Organization for Nuclear Research’s (CERN) machine was switched on, it had to be switched off because a faulty electrical connection between two of the accelerator’s magnets caused helium to leak into the underground tunnel. The electromagnets guide the hadrons around the accelerator and must be superconducting, which requires chilling them with the liquid helium to about –271 °C – a temperature colder than that of outer space.
But while the LHC is being repaired (and news of it is cooling down), let’s take a step back and look at what is behind the machine and facilitating the experiments.
Tons of data
Key to understanding what is going on are about 150 million sensors delivering data 40 million times per second, out of which about 100 collisions of interest will be filtered in every second. The resulting data flow from all experiments is about 700 MB/s, or about 15 million GB per year, equal to a stack of CDs about 20 km tall (or 100,000 DVDs). The data is accessed and analysed by thousands of scientists worldwide, who are connected via a computing grid.
Another challenge is controlling the experiment – connecting all parts of the setup within the 27-km-long tunnel under the Alps with the control centre – considering the high levels of radiation being produced by the accelerator. The traditional approach of using coaxial cables is not viable because it lacks the bandwidth required; hence, light and fibres are the obvious choice.
Finding the fibre
Although using optical fibres under high radiation levels was not entirely new territory for CERN, it was an area to be readdressed. This was done by inviting suppliers to provide samples of single-mode fibre that could meet the requirements of the LHC application: Radiation-induced losses could not exceed 7 dB/km at 1310 nm after exposure, with a total ionizing dose of 100,000 gray from high-energy physics being the main requirement.
The performance of seven fibres (out of an initial selection of 12) was compared in a study conducted by Fraunhofer Institute INT of Euskirchen, Germany, and tests were run using the institute’s in-house 60Co as a gamma radiation source and also at CERN under high-energy physics radiation conditions.
“Radiation normally changes every property of the fibre,” said the institute’s Dr. Jochen Kuhnhenn, “but resistance can be built in.” This was confirmed by tests he and his team performed, which showed that there can be a huge impact, with radiation causing losses mounting at 35 dB/km, but also that there was one fibre that actually showed a saturation of theses losses versus radiation dose – at a level of less than 5 dB/km. Further details on the tests and the fibres, which also are likely to be relevant for space or nuclear applications, were published in IEEE Transactions on Nuclear Science, Vol. 55, No. 4, August 2008.
The winning fibre, submitted and manufactured by Fujikura Ltd. of Japan, was selected to supply 2500 km of the material, which now connects all of the LHC sensors. Although how this was achieved appears to be a bit of a secret, Kuhnhenn points out that the fibre doping plays a key role. It is a standard step when making the fibre, inducing the important refractive index difference between the cladding and core, which keeps the light well contained. The winning fibre used fluoride doping, whereas germanium-doped fibres performed the worst.
Because making such radiation-insensitive fibre takes a long time, you might have to take it slow sometimes when you want to offer fast communication.