Towards particle accelerators in chip size

Instead of using kilometre-long large-scale research facilities, physicists want to accelerate electrons in micrometre-sized structures in the future - and thus much more cheaply and efficiently than today. A team from Erlangen and Darmstadt has now made an important breakthrough, which they have published in the renowned scientific journal Nature: It uses ultra-short laser pulses to lead elementary particles through tiny channels.

Guiding structure for electrons

Guiding structure for particles Foto: FAU

Professor Peter Hommelhoff, Chair of Laser Physics at FAU. Foto: FAU

Professor Peter Hommelhoff, Chair of Laser Physics at FAU. Foto: FAU

Particle accelerators are essential tools in research areas such as biology, materials science and particle physics. Researchers are always looking for more powerful ways of accelerating particles to improve existing equipment and increase capacities for experiments. One such powerful technology is dielectric laser acceleration (DLA). In this approach, particles are accelerated in the optical near-field which is created when ultra-short laser pulses are focused on a nanophotonic structure. Using this method, researchers from the Chair of Laser Physics at Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), which is associated with the Max Planck Institute for the Science of Light, have succeeded in guiding electrons through a vacuum channel, an essential component of particle accelerators. The basic design of the photonic nanostructure channel was developed by cooperation partner TU Darmstadt. They have now published their joint findings in the journal Nature.

35 times more effective thanks to laser acceleration

As charged particles tend to move further away from each other as they spread, all accelerator technologies face the challenge of keeping the particles within the required spatial and time boundaries. As a result, particle accelerators can be up to ten kilometres long, and entail years of preparation and construction before they are ready for use, not to mention the major investments involved. Dielectric laser acceleration, or DLA, uses ultra-fast laser technology and advances in semi-conductor production to potentially minimise these accelerators to merely a few millimetres or centimetres in size.

A promising approach: Experiments have already demonstrated that DLA exceeds currently used technologies by at least 35 times. This means that the length of a potential accelerator could be reduced by the same factor. Until now, however, it was unclear whether these figures could be scaled up for longer and longer structures.

A team of physicists led by Prof. Dr. Peter Hommelhoff from the Chair of Laser Physics at FAU has taken a major step forward towards adapting DLA for use in fully-functional accelerators. Their work is the first to set out a scheme which can be used to guide electron pulses over long distances.

Electrons race through colonnades of oval pillars

The scheme, known as ‘alternating phase focusing’ (APF) is a method taken from the early days of accelerator theory. A fundamental law of physics means that focusing charged particles in all three dimensions at once – width, height and depth – is impossible. However, this can be avoided by alternately focusing the electrons in different dimensions. First of all, electrons are focused using a modulated laser beam, then they ‘drift’ through another short passage where no forces act on them, before they are finally accelerated, which allows them to be guided forward.

In their experiment, the scientists from FAU and TU Darmstadt incorporated a colonnade of oval pillars with short gaps at regular intervals, resulting in repeating macro cells. Each macro cell either has a focusing or defocusing effect on the particles, depending on the delay between the incident laser, the electron, and the gap which creates the drifting section. This setup allows precise electron phase space control at the optical or femto-second ultra-timescale (a femto-second corresponds to a millionth of a billionth of a second). In the experiment, shining a laser on the structure shows an increase in the beam current through the structure. If a laser is not used, the electrons are not guided and gradually crash into the walls of the channel.  ‘It’s very exciting,’ says FAU physicist Johannes Illmer, co-author of the publication. ‘By way of comparison, the large Hadron collider at CERN uses 23 of these cells in a 2450 metre long curve. Our nanostructure uses five similar-acting cells in just 80 micrometres.’

The next step is to move into the third dimension

‘The results are extremely significant, but for us it is really just an interim step,’ explains Dr. Roy Shiloh, ‘and our final goal is clear: we want to create a fully-functional accelerator – on a microchip.’

Work in this area is being driven by the international ‘accelerator on a chip’ (ACHIP) collaboration, of which the authors are members. The collaboration has already proven that, in theory, APF can be adjusted to achieve acceleration of electron beams. Complex, three-dimensional APF setups could therefore form the basis for the particle accelerator technology of the future. ‘We have to capture the electrons in all three dimensions if we are to be able to accelerate them over longer distances without any losses,’ explains Dr. Uwe Niedermayer from TU Darmstadt, and co-author of the publication.

The scientists have published their results in Nature. The research was funded by the Gordon and Betty Moore Foundation (#GBMF4744), the ERC projects NearFieldAtto (#616823) and AccelOnChip (#884217) as well as the BMBF projects 05K19WEB and 05K19RDE.


Original publication:

“Electron phase space control in photonic chip-based particle acceleration", Nature, doi 10.1038/s41586-021-03812-9



Chair of Laser Physics
Dr. Roy Shiloh
Phone: +49 9131 85 27211

Johannes Illmer M.Sc.
Phone: +49 9131 85 27211

Prof. Dr. Peter Hommelhoff
Phone: +49 9131 85 27090



Edda Fischer

Head of Communication and Marketing
Phone: +49 (0)9131 7133 805


MPL Newsletter

Stay up-to-date with MPL’s latest research via our Newsletter. 

Current issue: Newsletter No 32 - October/November 2023

Click here to view previous issues.

MPL Research Centers and Schools