Future particle accelerators could be more powerful and consume less energy thanks to new software being developed by ATAP researchers.

Today’s large-scale particle accelerators consume vast amounts of electricity, creating substantial carbon footprints. Now, researchers from the Accelerator Technology & Applied Physics (ATAP) Division at Berkeley Lab are developing advanced modeling and simulation tools to create new accelerators that are potentially more powerful and more energy efficient than current designs.

Realizing the full potential of such accelerators, however, will require “a much better understanding of energy transfer and loss during the acceleration processes,” says Remi Lehe, a Research Scientist at ATAP’s Accelerator Modeling Program (AMP) who is part of the team developing new software for modeling the performance of plasma-based particle colliders.

“Particle physicists are always looking to design increasingly powerful accelerators, which drives up the costs of building and running new accelerators and produces even larger environmental impacts as more and more power is consumed.”

While power consumption accounts for a significant share of the operating costs of current accelerators, he notes that “only a small fraction of this power is eventually transferred to the particle beams, with most of it lost as heat.”

So, to meet the need for more powerful colliders while at the same time identifying ways to reduce power consumption, Lehe is working with colleagues at AMP and ATAP’s Berkeley Lab Laser Accelerator (BELLA) Center to develop advanced simulation and modeling software for systematically evaluating the performance of different accelerator designs.

“By combining a number of tools already developed by AMP,” explains Lehe, “we aim to create a new framework to study and optimize the performance of new accelerator designs while also maximizing their energy efficiency.”

The research, he adds, will use and extend the capabilities of WarpX, a Particle-in-Cell (PIC) simulation code. WarpX was developed under the DOE’s Exascale Computing Project by an international team of researchers led by Jean-Luc Vay, head of AMP, working with Berkeley Lab’s Applied Mathematics & Computational Research Division, as well as other key organizations.

Awarded the 2022 ACM Gordon Bell Prize, WarpX can leverage large-scale high-performance computing resources to model the transport and acceleration of particle beams in both laser-plasma accelerators and conventional colliders. WarpX also incorporates advanced mathematical algorithms, some of which were derived by Lehe and collaborators, to increase the speed and accuracy of the simulations.

Simulation showing a laser pulse (red) driving an accelerating plasma wakefield (blue) with an electron bunch (white) being accelerated in the wakefield. (Credit: Berkeley Lab/Remi Lehe)

Lehe, who has been a core developer of WarpX since its inception, says that using the capabilities of this code will help to identify ways for recapturing and reusing the energy of the accelerated beams.

“For example, in laser-driven accelerators, it may be possible to use a second laser pulse to capture the energy left in the plasma after accelerating the beam. Similarly, the energy of the final, post-collision beam could be recovered with a laser pulse in an additional plasma stage.”

The work, he notes, supports the approach adopted by the US Department of Energy (DOE), which has identified energy efficiency as a key performance measure for future accelerators. Lehe’s approach takes a different path to that of ATAP Research Scientist Marlene Turner’s Early Career Research Program project, “Energy Recycling for a Green Plasma Based Collider,” which focuses on experiments addressing energy recovery in plasma stages.

In addition to this edict from DOE, a draft report recently published by the Snowmass 2021 Collider Implementation Task Force—an international group of experts (which includes Turner) that evaluates accelerators for performance, technology readiness, schedule, cost, and environmental impact—stressed the importance of power consumption considerations as critical to the design of new accelerators, and recommended that research on energy efficiency “be given high-priority.”

“We are also building machine learning and optimization tools that can combine with WarpX simulations and guide us towards the most efficient accelerator designs. This could lead to significant enhancements in the capabilities and energy performance of new accelerators,” says Lehe.

The work described here is supported by the DOE’s Scientific Discovery Through Advanced Computing (SciDAC) program and Office of High Energy Physics. It includes researchers from AMP, BELLA, Argonne National Laboratory, Stanford Linear Accelerator Center, Fermi National Accelerator Laboratory, and the University of California Los Angeles.

“With his unique set of combined skills in high-performance computing, plasma physics, and inventiveness,” noted Jean-Luc Vay, “Remi is in an ideal position to leverage the power of supercomputing to further our understanding of energy dynamics in plasma-based accelerators, which will be key to maturing this new acceleration technology for many important applications.”


Learn More

  1. Ferran Pousa, A., Jalas, S., Kirchen, M., Martinez de la Ossa, M., Thevenet, M., Hudson, S., Larson, J., Huebl, A., Vay, J.-L., and Lehe, R. “Bayesian optimization of laser-plasma accelerators assisted by reduced physical models,” arXiv:2212.12551v1 [physics.acc-ph], submitted 23 December, 2022, https://doi.org/10.48550/arXiv.2212.12551
  2. Roser, T., Brinkmann, R., Cousineau, S., Denisov, D., Gessner, S., Gourlay, S., Lebrun, P., Narain, M., Oide, K., Raubenheimer, T., Seeman, J., Shiltsev, V., Strait, J., Turner, M., and Wang, L.-T. “Report of the Snowmass’21 Collider Implementation Task Force,” 2022, https://indico.fnal.gov/event/54953/sessions/20614/attachments/156153/205983/ITFreportDRAFT-July19.pdf
  3. Shapoval, O., Lehe, R., Thevenet, M., Zoni, E., Zhao, Y., and Vay, J.-L. “Overcoming timestep limitations in boosted-frame particle-in-cell simulations of plasma-based acceleration,” Physical Review E 104, 2021, https://doi.org/10.1103/PhysRevE.104.055311