Press release of December 19, 2019

Laser-plasma accelerators without limits

HZDR physicists present novel laser-driven electron accelerator scheme

Particle physicists harbor secret hopes of developing a new class of laser-plasma accelerators. They are small and inexpensive. But the maximum energy gain achievable per acceleration stage is limited. Or at least it was until now. A research team led by Dr. Alexander Debus from the Helmholtz-Zentrum Dresden-Rossendorf (HZDR) has developed a scheme based on theoretical considerations and large-scale simulations that may well approach the energy frontier. This would bring us one step closer to using compact particle accelerators in research and for medical and industrial applications. The scientists have published their findings in the journal Physical Review X (DOI: 10.1103/PhysRevX.9.031044).

New scheme for laser particle acceleration: Instead of shooting the laser into the plasma along its longitudinal axis, two separate, sideways-incident lasers enable higher particle energies, as well as unprecedented control over the acceleration process in the plasma.

New scheme for laser particle acceleration: Instead of shooting the laser into the plasma along its longitudinal axis, two separate, sideways-incident lasers enable higher particle energies, as well as unprecedented control over the acceleration process in the plasma.

Foto: HZDR / A. Debus


Physicists use particle accelerators to unravel the mysteries of matter. Biologists use them to generate X-ray lasers in order to watch cells in action. Medical scientists use the high-energy radiation to treat cancer, and engineers use it to create more economical components for power electronics. However, conventional accelerators are extremely large and costly. For example: an electron beam with 17.5 gigaelectronvolts (GeV) is needed to produce X-ray radiation at the European XFEL – and the accelerator has an overall length of around 1.7 kilometers.

Compact laser-plasma accelerators

“It is possible to build a plasma accelerator with the same electron energy that is more than 1,000 times or possibly even 10,000 times smaller,” asserted Alexander Debus. “What used to be kilometers are then only a matter of centimeters.” The technology has been intensively developed over the past two decades. Today, these compact devices are able to accelerate electrons up to 10 GeV in a single stage. However, that seems to be the end of the road. Laser-plasma accelerators are still several orders of magnitude away from energies for fundamental particle physics in the region of several teraelectronvolts (TeV), such as achieved by the Large Hadron Collider (LHC) using protons at CERN, the European Organization for Nuclear Research, in Switzerland.

Debus and his team want to change this. Their goal: “We want to accelerate electrons to the highest energies achievable on earth.” To do this, they have now developed a scheme that could transcend the frontiers. Instead of the usual procedure of laser propagation through the plasma by longitudinal injection, the researchers inject the energy laterally into a plane using two separate lasers. The scientists have checked and further refined their theoretical considerations using various simulations. To do this, they used the HZDR’s simulation software PIConGPU and Switzerland’s Piz Daint, Europe’s fastest supercomputer.

Surfing the plasma wave

For accelerating electrons using current plasma accelerators, an intense laser beam is targeted into a thin gas. The high intensity rips electrons from the gas atoms. They become ionized and a plasma is created. Within this plasma, the laser beam drives a charge density wave. If additional electrons are then shot into this wave, they surf the plasma wave directly behind the laser, just like a surfer riding an ocean wave towards the shore, gaining energy from the wave and thus accelerating. “But there are three basic problems that prevent maximum energy gain above 10 GeV within a single acceleration stage,” remarked Debus. “The dephasing, depletion and defocusing limits.”

For each of these limits, experimentally tested approaches on how to circumvent them already exist – however not for all limits combined. Even attempts to realize higher electron energies by means of several successive accelerator stages often lead to a deterioration of other beam parameters, such as the accelerated charge. Debus and his team are the first seeking to overcome all three performance limitations for laser-plasma accelerators – simultaneously and permanently.

“Of course, we are unable to suspend the laws of nature,” admitted the physicist. “But we propose using a different setup for accelerators so that the previously limiting laws are no longer an issue.” This setup is based on a complex, elaborately refined geometry: two high-intensity laser pulses are focused laterally into a gas – the accelerator medium. The laser pulses have to be cylindrically focused, while the laser pulse fronts are tilted at a certain angle. This causes the point of intersection of the two laser pulses in the plasma to move forward along the acceleration length at the vacuum speed of light.

This novel scheme offers several advantages for the future of laser-plasma accelerators. Not only does the approach enable this type of accelerator to be built longer and longer, resulting in ever greater energies far beyond the 10 GeV mark. It also enables lower GeV electron energies to be achieved with high plasma densities and simultaneously lower laser energy. This paves the way for increasing the electron beam repetition rate at the same total laser power.

Experimental implementation and simulations on an exaflops supercomputer

The next step will involve Debus and his team implementing their proposed approach in the laboratory. The aim is to create a functioning laser-plasma accelerator that can be used to verify their theoretical predictions. At the same time, they will continue to explore the theory behind this new class of accelerator. However, current computing resources are insufficient for simulating a 100-GeV to TeV-scale accelerator on a supercomputer. For this reason, the HZDR physicists now intend to use Frontier, the supercomputer that is scheduled to launch at the US Oak Ridge National Laboratory (ORNL) in 2021 as the world’s most powerful computer. The prospects are good, because the Dresden physicists, in collaboration with the University of Delaware, are one of eight teams that will gain early access to the Frontier supercomputer over the next two years.

More information about the three D’s of laser-plasma acceleration:

Dephasing can be translated freely with phase shift. This effect is due to the fact that the laser beams propagate in the plasma at a slower speed than in vacuum. However, the electrons are accelerated very close to the vacuum speed of light. As a result, they are faster than the laser-driven plasma wave from a certain acceleration length, i.e. the accelerated electrons escape eventually the wave. The electron beam must therefore be extracted before this limit is reached, hence setting an upper limit on the achievable electron energies.

The depletion limit is due to the current conventional setup of plasma particle accelerators. This is because, on the one hand, the achievable maximum energy gains of electrons increase with the length of the acceleration distance. On the other hand, the laser pulse energy decreases on its path directly through the plasma, and no longer suffices, from a certain length, to continue accelerating the electrons.

The path of the laser through the plasma along its longitudinal axis is also the reason for defocusing. After all, as the distance increases, it becomes more difficult to maintain a point focus of the beam. It fans out and can no longer drive the plasma wave. Electron acceleration stops.


A. Debus, R. Pausch, A. Huebl, K. Steiniger, R. Widera, T. E. Cowan, U. Schramm, M. Bussmann: Circumventing the dephasing and depletion limits of laser-wakefield acceleration, in Physical Review X, 2019 (DOI-Link: 10.1103/PhysRevX.9.031044)

More information:

Dr. Alexander Debus
Institute of Radiation Physics at the HZDR
Phone: +49 351 260 2619

Media contact:

Dr. Christine Bohnet | Spokesperson and Head of HZDR Communication
Phone: +49 351 260 2450 or +49 160 969 288 56