CLIC stands for the Compact Linear Collider. Despite a main accelerator of 42 km in length, the accelerator is “compact” due to its high accelerating gradient of 100 MV/m. To achieve the same centre-of-mass energy with LHC acceleration (5 MV/m) would require a distance of 840 km! Or alternatively, 700 km of LEP2 acceleration (6 MV/m).

When particles change direction (as they must in a circular collider) they emit photons and lose energy. This effect, called synchrotron radiation, can be avoided by accelerating particles in a straight line. The challenge of linear acceleration is to achieve a very high acceleration gradient, because, unlike in circular machines, the particles pass through a linear accelerator only once.

The luminosity of the accelerator scales as the wall-plug-to-beam efficiency. So one needs at the same time a high-gradient acceleration and an efficient energy transfer. The use of high-frequency RF maximizes the electric field in the RF cavities for a given stored energy. However, standard RF sources scale unfavorably to high frequencies, both in maximum delivered power and in efficiency. A way to overcome such a drawback is to use standard low-frequency RF sources to accelerate the drive beam and use it to produce RF power at high frequency. The drive beam is therefore used for intermediate energy storage.

In a classical approach, the linear accelerators used to accelerate the beams would be powered by Radio Frequency (RF) power supplies, called klystrons. In the CLIC acceleration scheme, the klystrons are replaced with an intense particle beam, called the drive beam. The kinetic energy in the drive beam is converted into RF power, which in turn is used to accelerate the main beams for collision. This scheme is called two-beam acceleration.

An intense beam of electrons is accelerated to a comparatively low energy (2.4 GeV) using conventional klystrons. This ‘drive beam’ is injected into a series of Power Extraction and Transfer Structures (PETS), which decelerate the dense beam and extract its energy. This energy is fed via an RF field in a waveguide to a second beam, which is much less intense. Since there are far fewer particles in this ‘main beam’, each one is accelerated to higher energy.

Each 21 km main linac has 10,380 two-beam modules. Each two-beam module contains up to four PETS. Each PETS generates the RF power for two accelerating structures. CLIC therefore has 71,460 PETS and some 143,000 accelerating cavities in total. The LHC uses 8 accelerating cavities per beam.

CLIC accelerating structures are designed and built to run very stably at a very high accelerating gradient (100 MV/m). The structures are built to micron-level tolerances to ensure that the beam quality is not degraded by beam-to-structure misalignment effects.

Each spent drive beam will already have lost 90% of its power in the PETS. After each of the 48 decelerating sectors the drive beam must be bent away from the linac, leaving enough space for a new drive beam to be injected to the next sector. The old beam is bent away using a dipole magnet, into a beam dump about 20 m away. The power to be absorbed per dump is 0.5 MW.

The two main CLIC linacs are each divided into 24 sectors. Each sector is 878 m long, and contains around 3000 accelerating structures. A fresh drive beam is injected into each sector to accelerate the main beam. At the end of the sector, the spent drive beam is dumped and a new drive beam is injected into the next sector. Each sector accelerates the main beam by 62 GeV.

Designed to be a high luminosity, high energy linear collider, CLIC will inevitably need high power. Compared to an accelerator using superconducting technology, CLIC nevertheless has very low power consumption in stand-by or "waiting-for-beam" mode. A preliminary analysis of the overall CLIC energy consumption per year for the various stages shows that the first stage of CLIC would be similar to LHC, and the second stage similar to the total CERN energy consumption. However, work is on-going in several domains (overall re-baselining, permanent magnets, air-handling etc.) to further reduce the anticipated power consumption of CLIC.

After the LHC, currently due to complete data-taking in 2030.

It is presently assumed that CLIC will be built underground, near to CERN in the area close to Geneva.