Researchers at ELI ALPS in Hungary have generated electron beams whose peak energy varied little across 10,000 consecutive laser shots. Their laser-plasma accelerator ran at 100 shots a second, producing a roughly 47 MeV beam while its peak energy fluctuated by 1.4% on a root-mean-square basis, according to a September 22 preprint.
How the laser feeds electrons into a plasma wave
A laser-wakefield accelerator sends an intense pulse through gas. The pulse turns the gas into plasma and drives an electric wave; electrons caught in that wave gain energy over a short distance. The challenge is to catch electrons in much the same way on every shot, so the output remains useful beyond a single demonstration.
The team focused extremely short, 70-millijoule laser pulses into a continuously flowing helium-and-argon gas jet. The laser uses optical parametric chirped-pulse amplification, which transfers energy to the pulse through nonlinear crystals while limiting heat buildup in the amplifier. A small amount of argon supplies electrons for the plasma wave to capture. The researchers' simulation suggests that the laser's self-focusing narrows the region where those electrons enter the wave, helping produce a concentrated energy spectrum.
Energy varied less than beam charge
In a 10,000-shot series, the beam's average peak energy was 47 MeV and its core charge, the electric charge carried by electrons in the measured central region of the beam, averaged 15 picocoulombs. Peak energy fluctuated by 1.4% root mean square, while charge fluctuated by 12% on the same measure. The authors estimate a 95% confidence upper limit of 10% for the beam's intrinsic relative energy spread after accounting for the spectrometer's resolution. These are measurements of the beam in this experimental setup, not a test of a medical device or other end use.
The 10,000 shots took about 100 seconds at the reported rate. Although the laser and gas setup can support faster operation, the accelerator itself was run at 100 Hz because the experiment lacked a dedicated high-power beam dump and faced radiation-protection limits. Its stability at 1,000 Hz has therefore not been demonstrated here. The simulation also needed less argon than the experiment to best reproduce the measurements, leaving part of the proposed injection mechanism unresolved.
