The collider of the future?

Why are two studies for one linear collider being conducted in parallel? This is far from a duplication of effort or a waste of resources, since the two studies reflect a complementary strategy aimed at providing the best technology for future physics. On Friday 12 June CERN hosted the first joint meeting between CLIC, ILC and the CERN management.

A view of the two beam lines in the CLIC experimental hall.

The International Linear Collider (ILC) and Compact Linear Collider (CLIC) studies both call for cutting-edge technologies. At first glance they may appear to be in competition, but they are in fact complementary and have a common objective – namely to propose a design , as soon as possible and at the lowest possible cost, for the linear accelerator best suited to taking over the baton of physics research at the high-energy frontier after the LHC.

If you’re looking for the answer to the question "which of the two?", you won’t find it here (or elsewhere!), simply because the answer will not be available before around 2012. "The LHC results will enable us to determine the energy range required by high-energy physics after the LHC and deduce which of the two technologies is the most suitable. In the meantime, it’s our duty to push both technologies as far as possible and make the most of the synergies they offer so as to meet the requirements of physics in an efficient way when the time comes," says Jean-Pierre Delahaye, who is in charge of the CLIC study.

The acceleration technology used by the ILC is based on the inherent qualities of superconducting cavities, which permit a highly efficient transfer of energy from radiofrequency fields to the beam, thereby reducing the electrical power consumption of a collider that has a beam power of several dozen MW. "Superconducting cavity technology is relatively "mature" as it has already been extensively used in numerous applications for accelerators. It has also greatly improved in recent years especially thanks to the tremendous R&D efforts deployed by the TESLA collaboration at DESY, " Delahaye explains. "The ILC collaboration has thus succeeded in producing cavities with an accelerating gradient of 35 MV/m, approximately 6 times that of the superconducting cavities used at LEP ten years ago. The problem that remains to be solved is that of the performance spread from one cavity to another, for even with identical manufacturing methods performances can vary significantly." Also, the accelerating gradient is limited by the intrinsic properties of the superconducting materials (see box). Even with a maximum gradient of 35 MV/m, the collider would need to be some 50 kilometres long to produce collisions of 1 TeV in the centre of mass. "Superconducting technology has the additional handicap of a relatively high cost per GeV since it requires special materials and cryogenic installations to keep them at very low temperatures," adds Jean-Pierre Delahaye. That’s why the ILC superconducting technology is viable but limited to the energy range up to 1 TeV.

Data sheet:

Energy range of collisions in the centre of mass (GeV)
200 to 1000
200 to 3000
Particles accelerated
Electrons and positrons
Electrons and positrons
Luminosity (cm-2sec-1) 2.1034 for 500 GeV/c 2.1034 for 3000 GeV/c
Nominal length (in km)
31 for 500 GeV/c
48 for 3000 GeV/c
Bunch frequency (in Hz)
Bunch spacing (in nanoseconds)
Number of particles per bunch 2. 1010 3.7. 109
Technology used to accelerate the particles
Superconducting RF cavities(1.3 GHz) at a temperature of 25K Two-beam acceleration scheme using high-frequency (12 GHz) RF cavities at room temperature
Maximum accelerating gradient MV/m
31.5 to 35
80 to 100
Estimated total cost (in millions of euros) 5310 + 13000 FTE-years at
500 GeV
Under evaluation
Cost per GeV (in millions of euros
Under evaluation
Conceptual/Technical designs available in (year)

The CLIC study, on the other hand, aims to develop a technology that will extend the energy range into the multi-TeV domain. It is based on an innovative technology, two-beam acceleration, which uses non-superconducting cavities at high frequency (12 GHz). The RF power used to generate high accelerating gradients is provided by a first high-intensity beam (known as the "drive beam") the energy of which is converted into the radiofrequency needed to accelerate the main beam (CERN Courier in September 2008 ). "CLIC technology had to be developed from scratch," says Jean-Pierre Delahaye. "Its main advantage is that it can generate high accelerating gradients, nominally 100 MV/m, making it possible to build a shorter and more compact machine that can therefore achieve higher collision energies. The second advantage is that, being shorter and not requiring cryogenic components, it has a lower cost per GeV. Conversely, it has called for technological developments in many areas, resulting in the construction of three test facilities, the latest of which, CTF3, is now practically complete. This project, which began 24 years ago in 1985, is now reaching the end of its feasibility study phase, the findings of which will be reported at the end of 2010 in a Conceptual Design Report (CDR) for a multi-TeV Linear Collider based on CLIC technology. "

In terms of collision energy, the ILC cannot go above 1 TeV, whereas CLIC, with the right design, could reach 3 TeV and beyond. "If the LHC experiments point to interesting physics at energies above 1 TeV, CLIC will be the only possible option because it is the only technology capable of reaching the multi-TeV energy range."

In the meantime, the two collaborations are working together to provide mutual support and ensure that both technologies are as mature as possible when decision time comes. As Jean-Pierre Delahaye points out, "Leaving aside the specific acceleration technology aspect, CLIC and ILC have much in common and face similar challenges. A number of years ago we came to the realisation that it was counter-productive for each community to develop its own study in isolation. So why not pool our resources on all common issues and take advantage of reciprocal knowledge and expertise to develop a better design for both? We thus identified seven key issues with strong synergies between the two studies and set up joint working groups that are co-managed by experts from both CLIC and ILC (see table). On 12 June, the leaders of CLIC and ILC had a first meeting, in the presence of the CERN management. At this meeting, the participants proposed a fundamental change in the ILC-CLIC collaboration. "We decided to move towards a collaboration not only on the technical level but also at management level. We have prepared a Memorandum of Understanding announcing the start-up of a joint venture to devise a strategy to promote and develop the scientific and technological preparations for the future linear collider that best meets the physics requirements, once these are identified."

At present, some 700 scientists from over 84 institutes in 12 different countries are participating in the ILC study, while 32 institutes from 17 countries are involved in the CLIC study. As Jean-Pierre Delahaye recalls "we created the CLIC/CTF3 collaboration in 2005 after evaluating the R&D work needed to develop CLIC technology and demonstrate its feasibility. Then we invited all laboratories worldwide, regardless of their involvement at CERN, to volunteer to take responsibility for one or more work packages. There were only two conditions: to convince us that a) they were capable of doing the work, and b) could bring their own resources to the table. CERN’s role is to host the collaboration and coordinate the work." Ten of CERN’s 20 Member States have volunteered, together with 7 non-Member States. The collaboration is run by a Management Board comprising one representative per member. This is very reminiscent of the collaboration model used for the construction of experiments but it’s the first time it has been used for an accelerator. "It is a testing ground for the possible future enlargement and globalization of CERN. And it works!" Jean-Pierre Delahaye concludes.

More information

The seven joint CLIC - ILC working groups

Civil Engineering & Conventional Facilities J. Osborne
V. Kuchler
C. Hauviller
J. Osborne
Cost & Schedule J. Carwardine
P. Garbincius
T. Shidara
K. Foraz
G. Riddone
P. Lebrun
3. Beam Delivery System (BDS) & Machine Detector Interface (MDI)

B. Parker
A. Seryi

D. Schulte
R. Tomas Garcia
L. Gatignon
Positron Generation (new)
J. Clarke
L. Rinolfi
Damping Rings (new)
M. Palmer
Y. Papaphilipou
Beam Dynamics A. Latina
K. Kubo

N. Walker
D. Schulte
Physics & Detectors F. Richard
S. Yamada
L. Linssen
D. Schlatter

Did you know?

The upgrade of LEP1 (50 GeV energy per beam) to LEP2 (100 GeV per beam) was made possible by replacing ordinary cavities with superconducting cavities. The accelerating gradient of the LEP2 cavities was 6 MV/metre. ILC has pushed this technology forward to obtain 35 MV/m. Above about 50 MV/m, the material from which the cavities are made, niobium, loses its superconducting properties. This value therefore constitutes the intrinsic operating limit of superconducting cavity technology for accelerators.

The machines used for accelerator physics have alternated between proton (and antiproton) machines, on the one hand, and electron (and positron) machines on the other. The former are called "hadron machines", owing to the fact that protons and their antiparticles belong to the family of hadrons. They are also known as "discovery machines". The latter, "lepton machines" are also referred to as "precision machines". This difference is due to the nature of hadrons, which themselves consist of smaller particles such as quarks, and leptons, which are elementary particles. Hadron accelerators such as the LHC can explore wide ranges of energy because in hadron collisions energy is distributed between the quarks (and other particles) so it is impossible to know the energy in the centre of mass in advance. Although the range of possible energies may be vast, accuracy is limited, precisely due to the inherent difficulty of evaluating the initial energy. The reverse is true for lepton machines such as CLIC or ILC. Here, the collision energy is known with great precision but it is a single value that has to be chosen on the basis of data supplied by the LHC.