Fermilab turns up the heat on electron cooling

29 May 2002

A technique that was first proposed by Gersh Budker in 1966 is being injected with new life by a team of physicists at Fermilab in the US. Kurt Riesselmann reports.


In February Sergei Nagaitsev and his group at the US Fermi National Accelerator Laboratory (Fermilab) reported a breakthrough. Working on an ambitious electron-cooling project, the team set a new world record for DC beam power – they maintained a continuous 3.5 MeV electron beam with a current of more than 500 mA for up to 8 h with only short interruptions.

These figures may not, at first sight, seem significant. After all, half an amp is the current flowing through a typical light bulb. However, the beam electrons travel at a much higher energy than those in an electric wire, leading to a record beam power of about 2 MW in the short prototype beamline.

Nagaitsev’s group aims to use an electron beam to cool antiprotons inside Fermilab’s 3 km Recycler antiproton storage ring and boost the luminosity of the laboratory’s Tevatron collider. When the electron-cooling system is complete, electrons and antiprotons will travel side by side in the Recycler. The electrons will absorb the excess heat of the antiprotons, shrinking the size of the antiproton beam. To be efficient, the electron beam must contain many more particles than the antiproton beam, requiring scientists to develop a high-current electron system.

The cooling process only consumes a fraction of the 2 MW beam power because scientists can recirculate the electrons and their power. The electrons start at the top of an 8 m high Pelletron accelerator – a Van de Graaff-type device developed by the National Electrostatics Corporation (NEC) – where they gain energy by travelling through a 3.5 MV electrostatic accelerating tube. They then pass through a loop and re-enter the Pelletron, where they are decelerated by traversing the electrostatic field in the opposite direction. A beam collector at the top of the Pelletron receives the electrons and supplies them for re-acceleration. Only a few electrons, about 20 in every million, are lost each trip. A 200 mA Pelletron-charging current is sufficient to ensure stable operation of the recirculation system and to restart beam recirculation within 20 seconds if the machine trips off. The Fermilab recirculation system is unique in sustaining such a high current with so little loss at an energy that is significantly more than a few hundred kilo-electron-volts.

A versatile machine

NEC, a Wisconsin-based company that received a Small Business Innovation Award in 1984 from the US Department of Energy, has made more than 140 Pelletrons and sold them in 38 countries. The machine gets its name from the chains of metal cylinders – pellets – that replace the belts of conventional Van de Graaff generators.

Pelletrons are used in applications such as surface analysis and doping of computer chips. But the machines are also valuable beyond the field of physics. The new security inspection system for the Channel Tunnel, for example, uses two Pelletrons to produce X-rays for scanning loaded trucks and containers. Pelletrons are also used for carbon dating in accelerator mass spectrometry.

Most Pelletrons operate as non-recirculating accelerators, typically featuring one-way beams of less than 50 mA. In contrast, Fermilab’s electron-cooling project relies on a continuous high-current beam, which can only be achieved through recirculation. “People in this business know how hard it is,” said project leader Nagaitsev. “Everybody is pushing the envelope. People working on related projects in the US and Europe are waiting for our results. Our success or failure means quite a bit at other laboratories.”

More collisions


With the help of electron cooling, Fermilab scientists will create a larger number of collisions inside the Tevatron. “The goal of our R&D project is simple – construct and commission an electron-cooling device that is ready to be moved to the Fermilab Recycler,” said Nagaitsev. A dedicated building to be located next to the Recycler is already being designed to house the electron-cooling equipment.

Nagaitsev’s team is currently working in a building more than a kilometre away from the Recycler. So far, electrons haven’t mingled with a single antiproton as the team is still making improvements on operating the Pelletron, producing an electron beam in stable mode for long periods of time. The Fermilab group plans to increase the beam energy to 4.3 MeV and the current to more than 1 A. So far, they have attained a 750 mA current for short periods of time.

The next step is to improve the quality of the electron beam as it travels through a special cooling section – initially without the presence of antiprotons. Only when that is achieved will the electron beam be used to cool antiprotons. “Depending on the efficiency of the Recycler,” said Nagaitsev, “maybe we can increase luminosity by a factor of two, maybe more.”

A test beamline with a nine-module cooling section is currently being incorporated into the Pelletron recirculation loop. This will enable the Fermilab team to study the electron beam carefully in the environment of the cooling section, determining the exact beam energy and the size of the high-current beam. Ultimately the electrons must travel parallel to the antiprotons, so the challenge is to put electron and antiproton beams on top of each other to within 50 mm.

In the final phase of the project, anticipated for 2003 or 2004, scientists will install the 20 m cooling section in the Recycler ring and send electrons and antiprotons through the cooling section at the same time. If everything works well, each antiproton will find itself surrounded by a cloud of electrons. Antiprotons going too fast will slow down as they bump against electrons in front of them. Antiprotons going too slow will speed up as electrons kick them from behind. With each collision, the lighter electrons will reduce the spread of energy within the antiproton beam. All of this will happen in a gentle way, since the masses of the particles make the collisions reminiscent of ping-pong balls bouncing off a bowling ball.

Cool idea


Gersh Budker first proposed the idea of electron cooling in 1966. It was first tested in 1974 at the Institute of Nuclear Physics in Novosibirsk, Russia, using proton beams. In 1976, David Cline, Peter McIntyre and Carlo Rubbia proposed using electron cooling for antiproton beams at Fermilab. Due to technical difficulties with cooling hot antiprotons, Fermilab turned to stochastic cooling, an alternative beam cooling technique developed at CERN by Simon van der Meer. The electron-cooling equipment went to the Indiana University Cyclotron Facility, where the equipment is still in use and provides electrons with a maximum energy of 300 keV.

CERN also developed electron-cooling systems, starting with the ICE ring in the late 1970s. CERN’s Low Energy Antiproton Ring (LEAR) used a 30 keV electron-cooling system from 1992 until 1996. Today, low-energy electron-cooling systems are used successfully at many facilities around the world. The Fermilab team is the first to develop the technique for electrons in the MeV range.


Potential applications for a recycled electron beam go beyond the world of particle physics, and the Fermilab result is attracting the attention of Free Electron Laser (FEL) builders around the world. FELs are powerful light sources that have many applications in molecular biology, materials science and chemistry. Rather than throwing away the electrons and their energy, recycling the beam could allow scientists to produce laser light with little electrical power input. Scientists at the University of California Santa Barbara have worked on Pelletron-driven FELs and beam recovery systems since the early 1980s using pulsed electron beams. The group has also looked at low-current continuous beam options, which were a precursor to the Fermilab project.

In particle physics, the future for electron cooling of antiproton beams looks bright. Stochastic cooling is limited, and to decrease beam temperature further, electron cooling is needed. Sergei Nagaitsev’s team has taken a big step in that direction. Some day, cooling antiprotons may be as easy as switching on a fridge. Although there is still a long way to go, it might be time to start chilling some champagne.

bright-rec iop pub iop-science physcis connect