Quantum Quenching of Radiation Losses in Short Laser Pulses

C. N. Harvey, A. Gonoskov, A. Ilderton, and M. Marklund

Phys. Rev. Lett. 118, 105004 – Published 10 March 2017

Abstract

Accelerated charges radiate, and therefore must lose energy. The impact of this energy loss on particle motion, called radiation reaction, becomes significant in intense-laser matter interactions, where it can reduce collision energies, hinder particle acceleration schemes, and is seemingly unavoidable. Here we show that this common belief breaks down in short laser pulses, and that energy losses and radiation reaction can be controlled and effectively switched off by appropriate tuning of the pulse length. This “quenching” of emission is impossible in classical physics, but becomes possible in QED due to the discrete nature of quantum emissions.

Received 4 August 2016

DOI:https://doi.org/10.1103/PhysRevLett.118.105004

Physics Subject Headings (PhySH)

Gamma-ray generation in plasmas Laser driven electron acceleration Radiation & particle generation in plasmas

Plasma PhysicsAccelerators & Beams

Authors & Affiliations

C. N. Harvey^1,*, A. Gonoskov^1,2,3,†, A. Ilderton^1,4,‡, and M. Marklund^1,§

¹Department of Physics, Chalmers University of Technology, SE-41296 Gothenburg, Sweden
²Institute of Applied Physics, Russian Academy of Sciences, Nizhny Novgorod 603950, Russia
³Lobachevsky State University of Nizhni Novgorod, Nizhny Novgorod 603950, Russia
⁴Centre for Mathematical Sciences, Plymouth University, PL4 8AA, United Kingdom

^*cnharvey@physics.org
^†arkady.gonoskov@chalmers.se
^‡anton.ilderton@plymouth.ac.uk
^§mattias.marklund@chalmers.se

Article Text (Subscription Required)

Click to Expand

Supplemental Material (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 118, Iss. 10 — 10 March 2017

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
(a) The no-emission probability $P_{0}$ in a one-cycle pulse as a function of initial electron energy and peak laser electric field ( $E$ ) in relativistic units $a_{0} = e E λ / 2 π m c^{2}$ (for electron charge $e$ , mass $m$ and $c$ is the speed of light). The laser has wavelength $λ = 0.82 μ m$ and spot size $w_{0} = 5 μ m$ . As the probability of not emitting is unity if, say, $a_{0} = 0$ (no laser), we restrict attention to the radiation dominated regime [10, 32] where the parameter $ε_{rad} = (2 / 3) α ω a_{0}^{2} γ / m \geq 1$ and significant recoil effects are expected: the color is therefore faded out for $ε_{rad} < 1$ . The optimal parameter region (blue) is then $a_{0} \in {100 \dots 200}$ and electron energies in the 100s of MeV. This region overlaps with that where $χ < 1$ : we work throughout in this regime, so that pair production need not be considered [17]. Also shown is the line $a_{0} = 2 γ$ above which classical RR causes reflection of the electron [33]. (b) Probability density of the electron energy loss in QED for initial electron energy 420 MeV and $a_{0} = 200$ . The corresponding peak intensity is $1.1 \times 1 0^{23} W / {cm}^{2}$ , slightly beyond the state of the art [34] but within reach of upcoming facilities [35, 36]. The average energy loss in QED (white line) grows more slowly than the classical prediction (pink line) of the Landau-Lifshitz equation [37] (“LL”; see Supplemental Material [18]). The bright yellow-white region in the lower left-hand corner shows that there is a high probability for the elections to pass through a short pulse without losing energy to emission. This is “quenching.”
Reuse & Permissions
Figure 2
Trajectories and energies of a single electron, incident from the right and passing through the center of an ultrashort, focused laser pulse of duration 0.5 cycles (1.4 fs, left) and one cycle (2.7 fs, right). Other parameters as in Fig. 1. The colored region shows the QED probability distribution, calculated from 1000 simulations with the same initial conditions. (For clarity any region containing $> 200$ electrons is colored as if it contained 200.) This distribution does not follow classical predictions which include RR, shown as a purple dashed line, but instead is visibly centered (white) on the Lorentz force curve shown with a dashed green line: this trajectory is by definition absent of RR. The lower panels show that, since the pulse is short, there is a high probability for the electrons not to emit until they are past the pulse peak. Even for the few electrons which subsequently emit, it is too late for their motion to be significantly affected.
Reuse & Permissions
Figure 3
Quenching in the collision of an electron bunch with laser pulses of different durations: the plots show the final electron distribution on a lanex screen 1 mm behind the focus. Classical predictions (top) show an asymmetric spatial distribution for short pulses. In the quantum results (bottom) the bright spot is formed by electrons which have traversed the pulse without significant emission or energy loss. The effect becomes less prominent as pulse length increases. For two cycles the classical distribution becomes symmetric, as one would expect, while in the quantum case stochastic transverse spreading [42] causes the bunch to spread out to a wider spot size than the classical bunch. The electron bunch initially had Gaussian distributions in energy, $420 \pm 0.35 MeV$ , and spacial position, $2 μ m$ FWHM. Laser intensity $a_{0} = 200$ , focal spot size $w_{0} = 5 μ m$ . (Parameters are similar to those achievable at ELI-NP [36].)
Reuse & Permissions
Figure 4
(a) The proposed experimental setup for observing quenching. (b) Lanex screens for 2 and 4 cycle pulses. Peak intensity $a_{0} = 50$ (50 TW total power), wavelength 820 nm, tightly focused using an $f / 1$ optic. A 0.32 T, 15 cm magnet is placed 1 m behind the slit, which deflects electrons in the $y$ direction according to their energy. The lanex screen is placed 50 cm behind the magnet and angled at 45°. The electron beam energy is 100 MeV (with 0.1% spread) and the beam is large enough so that precise synchronization would not be a concern in an experiment. The quantum distribution is very different from that predicted classically (“LL”), where electrons are confined to region 3. Quenched electrons, which have been deflected by the laser but not lost energy, will be delivered to region 1. Electrons can also appear in the nonclassical region 2, but due to stochastic spreading [42]: they have lost energy during the interaction. The boundary of region 1 is composed of the boundaries of the classical region 3 and the no-recoil Lorentz force prediction (“LF”).
Reuse & Permissions
Figure 5
3D QED PIC simulation results showing the density of electrons deflected in the $x$ direction by $> 0.5 °$ as a function of the $z y$ angle for pulse durations 1, 2, 4, 8, and 16 cycles (FWHM). The light gray zone ( $< 0.1 5 °$ ) demarks the classically forbidden region. The upper right inset shows the dependency of the total number of quenched electrons (from the gray region) on the number of cycles (right-hand scale, blue), and as a percent of all particles deflected by more than 0.5° in the $x$ direction (left-hand scale, green).
Reuse & Permissions

Physical Review Letters