Session NO08: High Energy Density and Beam Physics

Mitigating the impacts of PIC noise in gradient-based device optimization

We have identified and mitigated issues arising from PIC noise sources in the computation of adjoint-based gradients of an objective function with respect to device geometry. The geometry can be parameterized by $N$ degrees of freedom, and the gradient of an objective function with respect to these parameters is called the 'shape gradient'. The brute-force finite difference approach for computing a shape gradient is too computationally expensive because it requires $N + 1$ evaluations of the objective function, and each evaluation of the objective function requires a full device simulation. However, a recently developed adjoint technique enables the efficient computation of the shape gradient for a particular objective function using only two simulations [1]. In its original formulation, this technique considered the steady state of a prescribed emission model with a fixed set of emitted particles. It was shown that existing particle-in-cell (PIC) codes can be used for these simulations. But, this inherently introduces discrete particle noise into the simulations, which propagates noise into the gradient calculations. A further issue occurs in the optimization of nanoscale vacuum electron devices where the emission depends self-consistently on the applied electric field, and thus the emitted particles change each timestep, introducing additional particle noise. This application is not possible with the original method. We have characterized these noise sources, and developed techniques to mitigate them. In particular, we have developed a procedure to enable the use of adjoint-based gradient calculations with self-consistent emission models.

[1]: Phys. Plasmas 26, 013109 (2019)   

Characterizing hot electrons in ensemble PIC simulations of high-intensity, laser-plasma interactions with machine learning.

 

In the interaction of high-intensity lasers with over-dense plasmas, the hot electron population dictates system dynamics, driving ion acceleration and x-ray generation, and fast ignition, for example. The primary method for modeling these systems are particle-in-cell simulations (PIC), where macroparticles approximate the kinetics of a distribution of particles. A problem shared with experiments, PIC simulations are inherently noisy given the statistical nature of the algorithm, particularly at lower fidelities. Individual electron energy distributions are easily analyzed with expert examination, but ‘by hand’ analysis is impractical for the hundreds of thousands of case examples produced by ensemble simulations, and this is exacerbated by noisy data. In this work, data from an ensemble of PIC simulations is analyzed and quantified using machine learning (ML) techniques to more reliably extract the hot electron temperature and other physical parameters. We present the dependency of the hot electron temperature and other electron sheath properties on various input parameters such as laser intensity and pulse length across a wide parameter space (~10^18-10^21 W/cm^2, 30-500 fs).

    

A Theoretical Model of Chromatic Emittance Evolution in Plasma-Based Accelerators

We present the first analytic theoretical model describing the chromatic transverse dynamics of an electron beam in a nonlinear plasma-based accelerator that can account for the evolution of the projected, longitudinal sliced, and energy sliced emittance. Beginning with a description of single particle motion, the evolution of the beam moments and centroid position for each slice is calculated. In our approach, the longitudinal dependence of energy gain due to the beam loading of the wake is included. This permits a 6D, slice-by-slice (energy or longitudinal) analytic prediction of the beam evolution at any point within an adiabatic plasma source for the first time. It includes effects from both transverse mismatch and transverse offsets of the beam. We show that the amount of beam emittance growth in plasma ramps is directly related to the integrated plasma density profile independent of the ramp shape, so long as the ramp is adiabatic. We also show how our theory can predict the optimal length for a high-brightness plasma injector stage for a given target beam energy and energy spread. Using our theoretical framework, researchers will be empowered to design emittance-preserving plasma accelerators with less reliance on expensive particle in cell simulations. In addition, they will be able to better predict and interpret the observed behavior of the beam, permitting informed parameter adjustments for performance optimization.   

Hydrodynamic Simulations of Plasma Accelerator Sources

Understanding the long-term dynamics of plasma sources is critical to improve several aspects of plasma acceleration including discharge control, laser guiding and repetition rate. Numerical simulations can provide helpful insight into the relevant dynamics, but they can be challenging. For the long-term dynamics covering thousands to billions of plasma periods the plasma is thermalized and can be described well by hydrodynamic simulations.

We propose a quasi-neutral single-fluid plasma model capturing long-term plasma dynamics relevant for plasma accelerators. The model uses two temperatures (for atoms and electrons, respectively) and the plasma composition is tracked via collisional reaction rates.

We will present simulation results capturing the full dynamics of hydrodynamic optical-field-ionized (HOFI) channels [R. J. Shalloo et al., Phys. Rev. E 97, 053203 (2018)], and show comparisons to measurements, and explore the effect of the main parameters on the channel properties.

    

Betatron motion in a laser wakefield accelerator using a customized dispersion-free particle-in-cell field solver

When laser fields overlap electrons located within the wake of a laser wakefield accelerator (LWFA), these electrons can attain large energies due to the combined effects of the accelerating wakefields and direct laser acceleration (DLA). Standard particle-in-cell (PIC) techniques (e.g., Yee field solver and Boris push) can lead to numerical errors in electron trajectories and energies. A customized finite-difference field solver has been developed that eliminates dispersion errors in the phase velocity of light waves and errors in the Lorentz force due to the staggering in time of electric and magnetic fields. In this work we examine the collective behavior of the accelerated electron bunch in real space and phasespace with various PIC field solvers. Comparison is made to experiment, showing better agreement when using the customized field solver. The LWFA and DLA mechanisms contribute similar amounts of energy to the highest-energy electrons. Finally, the equation of motion for accelerated electrons is numerically integrated with and without time-stagger error terms, confirming the PIC results and providing a tool for investigating sustained betatron resonance.   

Excitation of long solitary waves in a neutralizing ion beam

Ion beams are used in various engineering applications such as particle accelerators, ion-thrusters, and ion-implantations. The electrons are introduced to neutralize the beam and reduce their divergence along the axis. This is most economically achieved by introducing electrons via an external filament source. However, in recent numerical works by Lan et al., electrostatic-solitary-waves (ESWs) were observed for such a setup in a 2D ion beam where the ESWs slowed the beam neutralization process. We present a Particle-in-Cell[3] study of the beam neutralization in both 2D and 3D beams and show the formation of ESWs and their movement along the beam axis. Further, by performing a theoretical analysis, we will show the uniqueness of these ESWs found in ion beams where a non-Maxwellian electron distribution allows for ESWs of large lengths to form.   

Arbitrarily Structured Laser Pulses

Spatiotemporal control refers to a class of optical techniques for structuring a laser pulse with space-time dependent properties, including moving focal points, dynamic spot sizes, and evolving orbital angular momentum. These structured pulses have the potential to enhance a number of laser-plasma applications, including laser wakefield acceleration [1,2], photon acceleration [3], Raman amplification [4], and inertial confinement fusion [5]. Here we introduce the concept of arbitrarily structured laser (ASTRL) pulses which generalizes techniques for spatiotemporal control. The ASTRL formalism employs a superposition of prescribed pulses to create a desired electromagnetic field structure. Several examples will be presented to illustrate the versatility of ASTRL pulses to address a range of laser-plasma applications. For instance, the ASTRL concept allows for the construction of flying focus pulses that may eliminate dephasing while enabling controlled injection in a laser wakefield accelerator.

[1] Palastro et al., PRL 124, 134802 (2020)

[2] Palastro et al., Phys. Plasmas 28, 013109 (2021)

[3] Howard et al., PRL 123, 124801 (2019)

[4] Turnbull et al., PRL 120, 024801 (2018)

[5] Igumenshchev et al, PRL 110, 145001 (2013)

    

An Alternative Approach to Incorporating Laser Pulses in Particle-in-Cell Simulations

Numerical modeling of electromagnetic waves is a critical component of particle-in-cell simulation of laser–plasma interactions. Traditionally, laser pulses have been either launched from simulated antennas or initialized in their entirety in the computational domain. Relying on the electromagnetic field update to advance the laser pulse, however, imposes high computational expense in situations where the computational domain must cover many Rayleigh ranges. As an alternative, we demonstrate that laser pulses can be incorporated directly into the particle push provided that numerical dispersion is accounted for both in terms of the phase velocity and the ratio between the electric and magnetic fields. This approach can reduce the size of the computational domain and eliminate the need for multiboundary antennas, facilitating, for example, the modeling of flying-focus pulses.   

Parametric study of the current filamentation instability using laser wakefield accelerated electron beams

As the fidelity of electron beams produced via Laser Wakefield Acceleration (LWFA) improves over time, so too does the opportunity to use these beams for the application of studying relativistic plasma phenomena. Current filamentation instability (CFI), the formation of high current density filaments in a relativistic beam as it travels through a cold background plasma, is an astrophysical instability that may be observed in laboratory scales using beams from LWFA. An experiment was conducted at Colorado State University's ALEPH facility where we measured the dependence of the CFI growth rate on plasma parameters such as length, density, and ion species of the background plasma using electron beams generated via LWFA. The relativistic beam plasma from LWFA and the background plasma for CFI were controlled independently. Results from analysis of data taken on this experiment and trends relating the growth of beam filaments to the measured parameter space will be discussed. These results will be compared to Particle-in-Cell simulations and established theoretical frameworks.   

Characterization and Optimization of Bremsstrahlung X-ray Diodes

Investigations are underway at Sandia National Laboratories looking into characterization and optimization of bremsstrahlung x-ray diodes operating in the >10 MeV regime using the CHICAGO particle-in-cell code.  These diodes typically use a hollowed-cathode geometry to form a circular beam on a high Z metal target.  This paper will discuss driver target impedance matching, look at beam uniformity on target, estimate anode surface temperatures, and examine near and far-field dose predictions.  Other cathode geometries such as a solid hemispherical ball will be considered as well.  All geometries are designed to keep the anode surface temperature below 400^o C to prevent ion formation and beam pinching to the axis.  Anode target geometry will be discussed including the advantages and disadvantages of specific materials such as titanium and tantalum.  Comparisons will be made to previous Hermes III and RITS-6 experimental bremsstrahlung diodes data.   

    

Spatiotemporal optical vortices (STOVs) and relativistic optical guiding

We present 3D PIC simulations of the spatiotemporal development of self-guided laser pulses, both circularly and linearly polarized, in a plasma, providing the first evidence of relativistic spatiotemporal optical vortices (STOVs). In prior work studying filamentation of lower intensity (10¹³-10¹⁴ W/cm²) femtosecond pulses in atmosphere, we discovered that circulation of electromagnetic energy density in these self-guided pulses is mediated by the spontaneous formation of spatiotemporal optical vortices (STOVs), the phase circulation of which resides in spacetime [1]. In this work, at intensities 10¹⁹-10²⁰ W/cm², relativistic collapse of an intense laser pulse drives a plasma wave, which generates STOVs by nonlinear phase shear. These STOVs are seen to nucleate at multiple locations on the pulse and undergo vortex “reconnection,” evolving into vortex rings surrounding the pulse to which they are confined. After formation, phase circulation about STOVs is seen to dictate the local electromagnetic energy flux density and delimit self-focusing and diffraction, playing a critical role in relativistic self-guiding of pulses.

[1] N. Jhajj, I. Larkin, E. W. Rosenthal, S. Zahedpour, J. K. Wahlstrand, and H. M. Milchberg, "Spatiotemporal optical vortices," Phys. Rev. X 6, 031037 (2016).   

Towards AI-driven Experiments at PW-class Laser Facilities

Today, there are multiple high-intensity short-pulse lasers around the world that are capable of operating at high repetition rate (>1 Hz), representing an opportunity to accelerate the rate of scientific exploration by >3 orders of magnitude. In order to achieve this, diagnostics, targeting, laser control, and diagnostic analysis must all operate at commensurate rates. Machine learning provides a path for achieving this by utilizing fast surrogate models for each piece. While demonstrations of this technology for these purposes have been growing in number, they must now be integrated into a fully autonomous system through artificial intelligence.

Here, we present progress on this front by utilizing a physics-based ML model as the guide for experimental exploration and optimization by; proposing experimental samples, analyzing data, retraining the original model, and proposing new experimental samples with a “human in the loop”. The experiments were carried out at CSU’s ALEPH laser facility where the laser is controlled through spectral phase shaping and MeV-energy electrons and protons are measured with HRR diagnostics. The process was iterated multiple times in order to reduce the model uncertainty while searching for optimal laser settings to increase MeV particle production.

    

Structured targets at high-repetition rates for high-intensity laser-plasma interactions

The use of structured targets to improve specific experimental outcomes has been robustly studied in simulations and is being used more frequently for experiments as technology improves. These structures, including microtubes, cones, cone-wires, and curved targets, have proven useful for many applications but are limited in repetition rate due to the precision alignment required and high cost per target. With the rise of high repetition rate, ultra-intense laser systems that can operate above 100 Hz, there comes a need for solid density structured targets that can operate with these systems. In line with our previous work of using ethylene glycol sheet targets that can be 100s of nanometers thick that regenerate as the liquid flows, we demonstrate the generation of a dynamically shaped structured target that can be formed from these sheets after the irradiation of a 10¹⁶ Wcm^-2 short pulse laser. The structured target evolves over microseconds, forming a hollow channel, a cone, a cone-wire, and a curved surface with a wire as it evolves in time. The target can be implemented in a relativistic laser-plasma interaction by selecting the timing delay between a target shaping prepulse and a ultra-high intensity pulse, where the timing delay will determine the structure of the target when the main pulse arrives. The structures are highly repeatable in shape and position, operate above kHz repetition rates, and are very low cost, opening a path to structured target interactions at high repetition rates.

    

Identifying Trends in Self-Induced Relativistic Transparency in Plasmas with Ultrafast High Intensity Laser Pulses

A LaserNetUS experiment was conducted at the Scarlet Laser Facility (λ = 800 nm, 35 fs), investigating Relativistic Transparency (RT) in ultrafast high intensity laser plasma interactions. RT is the phenomenon where a high intensity laser heats the electrons in a classically overdense plasma to relativistic energies, so instead of the plasma being opaque to the light, it transmits through the plasma. The targets were 8CB liquid crystal films, with thicknesses varied between 20-200 nm and the laser intensity scanned over the range 10^19-10²¹ W/cm². Transmitted and reflected light were collected via scattering screens, and transmitted temporal and spectral properties were by a GRENOUILLE. We compared the results with 2D particle-in-cell (PIC) OSIRIS 4.0 simulations. We present our findings showing strong trends in transmitted light vs target thickness, as well as an unexpected spatial profile from the transmitted light   

700 nC electron bunches from intense laser-plasma interactions

The commissioning of multi-petawatt class laser facilities around the world is gathering pace. One of the primary motivations for these investments is the acceleration of high-quality, low-emittance electron bunches. In this work, the first conclusive computational evidence is provided that super-high charge electron beams (hundreds of nano-Coulombs) with emittance properties comparable to those required for forefront particle colliders are formed from the interaction of an intense laser pulse with over-dense plasma. On attosecond timescales, such bunches provide the dominant laser-plasma energy absorption mechanism. Bunch energies are predicted via the Zero Vector Potential model and compared to two-dimensional particle-in-cell (PIC) simulations over an unprecedentedly large parameter space: from non-relativistic laser intensities to the laser-QED regime and from the critical plasma density to well beyond solid density. These results have wide-ranging implications for future particle accelerator science and associated technologies.