Feb 2 2016
Laser-plasma accelerators (LPAs) are referred to as “tabletop” since, as exhibited by the novel BELLA accelerator at the U.S. Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab), they can increase electron beams to multibillion electron-volt energies (GeVs) in a few centimeters — a distance several times shorter than conventional accelerators.
Members of the BELLA Center staging experiment team, from left, are Eric Esarey, Wim Leemans, Jeroen van Tilborg, Carlo Benedetti, Kelly Swanson, Anthony Gonsalves, Joost Daniels, Sven Steinke, and Kei Nakamura. Not pictured are Cameron Geddes, Carl Schroeder, Nicholas Matlis, and Brian Shaw. (Photo credit: Roy Kaltschmidt/Berkeley Lab)
Beyond the small distance, the energy gain stalls and laser pulse weakens. LPAs will have to shed the tabletop image if they are to compete with traditional colliders, such as 30-kilometer-long electron-positron linear colliders or circular proton colliders 100 kilometers in circumference, with electron-volt energies in the trillions (TeVs). By coupling a hundred LPAs in series, where each LPA is run by a BELLA-class laser in series, and accelerating a well-formed beam from one stage to the other, high energies can be realized.
Long before planning began for BELLA, we’d set our sights on staging as the way to achieve energies needed for compact particle colliders, free-electron lasers, and other tools of future science.
Wim Leemans, Director of Berkeley Lab’s Accelerator Technology and Applied Physics Division (ATAP), Director of the BELLA Center.
Due to the daunting technical hurdles, such as keeping electron beams having dimensions measured in millionths of a meter and laser pulses measured in femtoseconds, Leemans says, “Lots of people told us we’d never be able to do it.”
In an experiment with many scientific firsts, Leemans and his collaborators demonstrate that an electron beam can be accelerated by a laser pulse and can be coupled to another accelerator, allowing another laser pulse to accelerate the electron beam to higher energy — an original breakthrough in advanced accelerator science.
The results have been published in the Feb. 1 issue of Nature.
A billion electron-volts wouldn’t matter, what mattered was stability. You don’t want to spend three-quarters of your day tuning your beam injector, with no time left to do an experiment.
Sven Steinke, Lead Author
The solution was to employ two types of LPA. The more advanced but finicky is a discharge capillary, which is a sapphire with a thin tube through it in horizontal direction. The tube is filled with hydrogen gas, and a strong electrical discharge ionizes the gas, separating each electron from its nucleus and generating a plasma. The plasma is heated by the discharge arc and a laser waveguide is formed. This waveguide is a cylindrical channel of a thinner plasma at the core; the inward laser pulse drives through the channel, similar to a water speedboat and accelerates the free electrons.
A jet of supersonic gas, with a diameter of few hundred micrometers, is the second type of LPA. The laser pulse drills into the gas and forms a plasma by ionizing the gas, accelerating the free electrons.
The team chose the gas jet, conceptually simple yet has beam energies of more than a hundred million electron-volts, for Stage 1, the beam injector. For Stage 2, the choice was the powerful discharge capillary, like the kind used in BELLA.
A major challenge was how to introduce the second laser pulse, using a mirror, within the few-millimeter space that lay between the two stages. The electron beam had to pass through a hole in the mirror, and the reflected laser pulse had to follow close behind. To focus enough power to accelerate the beam, the laser focus has to be very close to the mirror, which would result in the mirror being blown into pieces.
We decided from the beginning of the project that instead of worrying about blowing up the mirror, we’d blow it up with every shot.
Wim Leemans, Director of Berkeley Lab’s Accelerator Technology and Applied Physics Division (ATAP), Director of the BELLA Center
They initially developed a prototype mirror made up of water film but settled for much more robust VHS tape.
Video cassette players may not be in fashion today, but the VHS tape is stretch-resistant, thin, and can be run for long periods of time. The electron beam pierces the tape virtually untouched. On the other side, in a fraction of a second prior to the laser pulse penetrating the tape, it ionizes the tape surface to form a dense and perfectly flat plasma: an efficient mirror.
Steinke, whose dissertation was concerning plasma mirrors and joined the BELLA Center after being a postdoc at the Max Born Institute in Berlin, was responsible for characterizing the mirror system for performing the staging experiment. The earlier plasma mirrors were made out of expensive solid optics but for different purposes. Steinke and Leemans agree: “This was the first use of a continuous, high-repetition-rate, disposable plasma mirror.”
The staging system was set for its first test. The first laser pulse generated an electron beam in the gas-jet LPA, and passed through the tape, while the second laser pulse was reflected by the plasma mirror. Both the laser pulse and the electron beam entered the stage 2 capillary. However, no beam exited.
We were stunned. Suddenly there were four or five of us sitting around scribbling on the backs of envelopes.
Jeroen van Tilborg, Member of the BELLA Center
ATAP scientists had employed discharge capillaries for injecting and accelerating electrons for more than a decade, but this was the first instance that someone shot an external electron beam into one. Nobody had dealt with the impact of the powerful discharge current: the current ionizes the gas so that an optical waveguide is created through the plasma, and a strong magnetic field is created, which can break apart a pre-existing electron beam.
While analyzing the problem, Van Tilborg realized that the pulsed magnetic field could be a better plasma lens. This kind of fast-acting lens can have many applications, for instance it could conditioning existing free-electron laser beams. Focusing the injector beam of the staging experiment could be its immediate application.
The final configuration included plasma lens, gas-jet injector, discharge capillary second stage, plasma mirror, and diagnostics displayed energy gains of around a hundred million electron-volts for major sections of the electron beam.
The experiment was successful because of on-the-job discoveries and continuous feedback between computer modeling and experimental observations. The efficient INF&RNO code for modeling laser and plasma interactions, running on a Cray supercomputer at DOE’s National Energy Research Scientific Computing Center (NERSC) at Berkeley Lab, is able to quickly simulate a day’s experimental data. It could enable exploring the intricacies of laser timing and simulating the ragged but energetic beam from the gas jet.
Through matching to the experimental observations, simulation can see everything. We can see how the laser beam is behaving and understand which electrons are the ones being accelerated.
Carlo Benedetti, BELLA Center’s simulation team
The principlehas been proven by successfully coupling two independent laser-plasma accelerators for the first time. The real thing comes next.
“We’re ready for staging BELLA,” says Leemans, using two charge-capillary LPAs. “We’ll split the BELLA laser beam,” capable of a quadrillion watts (a petawatt) per 40-femtosecond pulse every second.
The first stage should bring up the beam to about 5 GeV. We will do the bunch transport with our capillary lens and play around with the timing of the second pulse. We should come out of the second stage with 10 GeV. And, while in the staging experiment we’re only trapping about three or four percent of the electrons available, in BELLA we’ll be able to trap 100 percent of the charge.
Wim Leemans, Director of Berkeley Lab’s Accelerator Technology and Applied Physics Division (ATAP), Director of the BELLA Center.
Steinke says, “BELLA is much simpler. The effects of the tape on beam quality should be much less, and the beam is much ‘stiffer,’ easier to handle.”
Van Tilborg concurs: “At 5 GeV per stage there may be no problem. The higher energy saves you.”
Many groups around the world are working on different aspects of LPA development, and I am confident that we will see the first applications of LPAs in the coming decade. As with all new technologies, the nature of those applications may surprise us.
James Symons, Associate Laboratory Director for Physical Sciences, Berkeley Lab.
Challenges exist, but the era of compact accelerators that can achieve high energies is finally upon us.
This research work received support from DOE’s Office of Science, the National Science Foundation, and DOE’s National Nuclear Security Administration. The computational resources of the National Energy Research Scientific Computing Center (NERSC), a DOE Office of Science User Facility, have also played a crucial role in this study.