🔬 physics

Dependence of self-injected bunch parameters on the plasma density gradient and laser pulse amplitude at LWFA in a conical plasma channel

This paper uses WarpX numerical simulations to investigate how the charge, energy, length, area, and transverse emittance of self-injected electron bunches in laser wakefield acceleration depend on the laser pulse amplitude and the longitudinal plasma density gradient within a conical channel.

Original authors: D. S. Bondar, V. I. Maslov, I. N. Onishchenko

Published 2026-09-04

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: D. S. Bondar, V. I. Maslov, I. N. Onishchenko

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the quest to build smaller, more powerful machines for exploring the fundamental nature of matter, scientists have long sought ways to accelerate particles to incredible speeds without the massive, room-sized structures that currently define the field. One promising approach involves using a pulse of intense laser light to create a wake in a cloud of gas, much like a boat creates a wake in water. As the laser moves through this ionized gas, known as plasma, it pushes electrons aside, leaving behind a trail of positive charge that acts as a powerful accelerator. If the conditions are just right, some of the electrons from the gas itself get caught in this wake and are swept along, gaining tremendous energy in a very short distance. This process, called self-injection, is attractive because it eliminates the need for a separate, complex machine to feed electrons into the accelerator. However, controlling exactly how many electrons get caught and how fast they go is difficult, as the process is extremely sensitive to the shape of the laser and the density of the gas.

Researchers in Ukraine and Germany recently investigated how to better control this self-injection process by shaping the environment through which the laser travels. Instead of using a uniform tube of gas, they simulated a setup where the channel narrows like a cone and the gas becomes denser as the laser moves forward. Using a powerful computer program that tracks the movement of individual particles, they tested how changing the strength of the laser and the steepness of the gas density increase affected the resulting bunch of accelerated electrons. Their goal was to find the precise combination of conditions that would trap the most electrons and give them the highest speed without causing the bunch to break apart.

The team modeled a laser pulse moving through a plasma channel that starts wide and gradually narrows, while the density of the gas inside increases from the entrance to the exit. They ran thousands of simulations, varying the intensity of the laser and the rate at which the gas density rose. They discovered that the narrowing channel acts like a lens, compressing the laser light and making it more intense as it travels, which helps create a stronger wake. However, the density of the gas played a critical role in determining whether the trapped electrons would stay together or be lost. When the gas density increased too sharply, the wake created by the laser shrank so much that it crushed the back of the electron bunch, causing a portion of the particles to be lost.

The simulations revealed a clear "sweet spot" for these conditions. When the gas density increased moderately along the channel, the electrons stayed trapped in the accelerating phase of the wake for longer, gaining more speed and forming a larger, more energetic bunch. Specifically, when the density at the exit was three times higher than at the entrance, and the laser was set to a specific high intensity, the researchers observed the best results. In this optimal scenario, the self-injected bunch reached a length of 5.10 and carried a charge of 32.1 microcoulombs per meter. The electrons in this bunch achieved a mean longitudinal momentum of 111.9 times the rest mass of an electron, a significant increase compared to what was achieved in a uniform channel.

Crucially, the study showed that pushing the gas density even higher did not yield better results. When the density at the exit was increased to four times the starting value, the wake bubble contracted too violently. This excessive compression caused the rear wall of the wake to collapse into the bunch, stripping away electrons and reducing the overall momentum of the group. Similarly, if the laser was too strong without the right density gradient, the bunch would also break apart. The researchers found that the most effective range for the laser intensity was between 2.8 and 3.6 times a specific reference value, provided the density gradient was moderate. In these conditions, the bunch remained stable, with a transverse emittance—a measure of how tightly the beam is focused—staying below 4.2 times 10 to the negative 2 millimeter-milliradians.

The findings suggest that carefully tuning the shape of the plasma channel and the density of the gas can significantly improve the quality of electron beams generated by lasers. By avoiding the extremes where the wake either fails to form properly or collapses under its own pressure, scientists can produce beams that are both powerful and well-structured. The simulations indicate that with the right combination of a narrowing channel and a rising density gradient, it is possible to create self-injected bunches that are larger, faster, and more stable than those produced in uniform environments. This work provides a roadmap for designing future experiments where the goal is to generate high-quality electron beams for scientific research without the need for massive external injectors.

Technical Summary: Dependence of Self-Injected Bunch Parameters on Plasma Density Gradient and Laser Pulse Amplitude in LWFA

Problem Statement
Laser Wakefield Acceleration (LWFA) offers high-gradient acceleration of charged particles, particularly through the self-injection mechanism where background plasma electrons are trapped and accelerated by the wakefield without an external injector. While self-injected bunches possess unique parameters (charge, energy, emittance), their characteristics are highly sensitive to laser and plasma conditions. Previous studies have utilized shaped plasma density profiles and laser pulse shaping to control these parameters. However, specific investigations into the limiting configurations of these controls were lacking. Specifically, the interplay between the longitudinal plasma density gradient, laser amplitude, and the geometry of a conical plasma channel had not been fully explored regarding the stability of the self-injected bunch. The primary challenge addressed is identifying the limits of density inhomogeneity gradients where the beneficial increase in longitudinal momentum is offset by strong wake bubble deformation, leading to the destruction of the bunch at the bubble's rear wall.

Methodology
The study employs numerical simulations using the fully relativistic Particle-In-Cell (PIC) code WarpX in a 2D3V geometry. The simulation models a conical plasma channel where the radius decreases linearly from $4.0 c/\omega_{pe}$ to $2.16 c/\omega_{pe}$ over a propagation distance. The plasma density is configured in four distinct cases:

Homogeneous: Constant density $n_e = 2.61 \times 10^{19} \text{ cm}^{-3}$ .
Inhomogeneous: Linear longitudinal density ramps increasing from $n_e$ at the entrance to maximum values of $2n_e$ , $3n_e$ , and $4n_e$ at the channel exit.

The laser driver parameters include a wavelength of $\lambda=800$ nm, a waist $w_0=3.9 c/\omega_{pe}$ , and a duration of 21.7 fs (3 $\sigma$ ). The normalized laser amplitude ( $a_0$ ) is varied systematically across six values: 2.0, 2.4, 2.8, 3.2, 3.6, and 4.0. The self-injected bunch parameters are analyzed at $t=238$ fs, corresponding to the moment the wake bubble head reaches the channel exit ( $z=76\text{--}90 c/\omega_{pe}$ ).

Key Results
The simulations reveal a complex interdependence between the laser amplitude, density gradient, and the resulting bunch parameters:

Laser Energy Density: The conical channel geometry induces radial compression of the laser pulse, increasing the peak laser energy density on the system axis by a factor of 2.1 compared to a cylindrical channel of constant radius.
Wake Bubble Dynamics: In the homogeneous case, increasing $a_0$ $a_{0}$ causes the wake bubble to expand until it is constrained by the channel walls, leading to the shortening of the bubble and the loss of its rear section. In inhomogeneous cases ( $2n_e$ $2 n_{e}$ , $3n_e$ $3 n_{e}$ , $4n_e$ $4 n_{e}$ ), the increasing plasma density shortens the plasma wavelength, reducing the bubble radius.
- At gradients of $2n_e$ and $3n_e$ , the bubble remains smaller than the channel, keeping the bunch in the accelerating phase.
- At the $4n_e$ gradient, the bubble contracts excessively, causing part of the bunch to be destroyed at the rear wall.
Longitudinal Momentum ( $p_z$ ): The mean longitudinal momentum increases with both laser amplitude and density gradient up to a limit.
- In the homogeneous case, $p_z$ reaches $96.7 m_ec$ at $a_0=4.0$ .
- In the $3n_e$ case, $p_z$ reaches a maximum of $126.5 m_ec$ at $a_0=4.0$ .
- At $4n_e$ , the gain in momentum is reduced ( $123 m_ec$ ) due to the excessive contraction of the bubble and subsequent bunch destruction.
Bunch Charge, Length, and Area: These parameters exhibit threshold behavior.
- For moderate gradients ( $2n_e$ , $3n_e$ ), charge, length, and area increase with $a_0$ . For instance, at $3n_e$ , the charge increases by a factor of 3.13 when $a_0$ goes from 2.0 to 4.0 (while at $2n_e$ the increase is 2.84 times).
- In the homogeneous case and at the $4n_e$ gradient, these parameters peak (e.g., at $a_0=2.8$ ) and then decline as $a_0$ increases further, indicating the destruction of the bunch tail.
Emittance: The transverse emittance ( $\varepsilon_x$ ) remains low, not exceeding $4.2 \times 10^{-2} \text{ mm}\cdot\text{mrad}$ across all tested configurations.

Optimal Configuration
The study identifies an optimal operating range where the density gradient and laser amplitude compensate for each other to maintain the bunch in the accelerating phase without significant deformation. The optimal configuration is found at a density gradient of $3n_e$ and a laser amplitude of $a_0=3.6$ . Under these conditions, the self-injected bunch achieves:

Charge: $32.1 \text{ \mu C/m}$
Mean Longitudinal Momentum: $111.9 m_ec$
Length: $5.10 \text{ \mu m}$
Area: $5.3 \text{ \mu m}^2$
Transverse Emittance: $1.6 \times 10^{-2} \text{ mm}\cdot\text{mrad}$

Significance and Claims
The paper claims to have investigated the specific limits of density inhomogeneity gradients and laser amplitudes that define the stability of self-injected bunches in conical channels. The primary contribution is the identification of the "limiting" density gradient ( $4n_e$ ) where the beneficial effects of density tapering are negated by bubble over-contraction and bunch destruction. The results demonstrate that a moderate longitudinal density gradient (up to $3n_e$ ) combined with appropriate laser amplitude can significantly enhance the longitudinal momentum and charge of self-injected bunches compared to homogeneous channels, while maintaining low emittance. The findings regarding momentum enhancement at $3n_e$ are noted to be consistent with experimental data showing energy increases in up-ramp plasma profiles. The study establishes that the combined influence of amplitude changes, density gradients, and conical channel effects must be carefully balanced to avoid the deformation of the wake bubble structure that leads to bunch degradation.

More like this