For its 4th iteration, the Smilei user & training workshop is hosted in Prague (Czech Republic) by ELI-beamlines and CTU. Lunches will be provided on site.
Programme
Organizers
Links
Acknowledgements: This workshop is supported by ELI beamlines, Plas@Par and Appel
Smilei is celebrating its 10th anniversary since the first written line of code, and its 5th anniversary since its reference paper publication. We will review its current capabilities, new and future features, the recent achievements, as well as the participating people and its position in the landscape of PIC codes.
Smilei is a c++ code at its core, but it presents many more aspects: documentation, installation procedures, input files, post-processing, support, training, teaching capabilities, quality, etc. In this presentation, we will give an overview of this community-driven project.
An intense laser pulse can create a thin plasma surface capable of reflecting the laser field, a phenomenon commonly referred to as a 'plasma mirror.' Within this plasma, electrons are influenced by the laser field, displaying oscillatory motion. At high laser intensities, the oscillatory electron motion becomes highly nonlinear due to the relativistic effects, resulting in the generation of high harmonics (HHG). Given the difficulty of directly measuring the ultrafast dynamics of electrons driven by such intense laser fields, a detailed analysis of HHG generation mechanisms has heavily relied on theoretical approaches. Two prominent generation mechanisms, Coherent Wake Emission (CWE) and the Relativistic Oscillating Mirror (ROM) model, explain these nonlinear phenomena. In this study, we present results from numerical simulations conducted using the particle-in-cell simulation code, Smilei. Our investigation focuses on assessing the influence of laser properties (intensity, beam shape, wavefront, and chirp) as well as plasma properties (scale length and thickness) on HHG from a liquid plasma mirror [1]. Our findings indicate that the amplitude of plasma oscillations increases linearly with laser intensity. Additionally, we observed that positively chirped pulses tend to generate HHG more efficiently in the CWE regime. Furthermore, adjusting the laser beam focus slightly away from the target surface leads to a reduction in the divergence of the harmonic beam. These results provide insights into identifying optimal conditions for maximizing HHG yield, thereby guiding the development of an intense attosecond light source in the EUV and X-ray wavelength range.
[1] Yang Hwan Kim et al., “High-harmonic generation from a flat liquid-sheet plasma mirror,” Nature Communications 14, 2328 (2023).
Historically, dark matter searches have primarily focused on hunting for effects from two-to-two scattering. However, given that the visible universe is primarily composed of plasmas governed by collective effects, there is great potential to explore similar effects in the dark sector. Recent semi-analytic work has shown that new areas of parameter space for dark U(1) and millicharged models can be probed through the observation of collisionless shock formation in astrophysical dark plasmas, a nonlinear process that requires simulation. Here, I will show the initial results from simulating such warm, non-relativistic pair plasmas within the Smelei framework and highlight the impacts in dark matter phenomenology.
Particle-in-cell (PIC) code SMILEI enables various approaches and options how to initialize numerical particles or track them during simulation run. These features of SMILEI were employed for several studies of laser-driven ion acceleration with the aim to study ion acceleration mechanisms or interpret experimental results. The possibility to start the simulation with particles’ positions and weights defined in external HDF5 file for each particle was used for the interpretation of experimental results on multi-MeV alpha particle source via proton-boron fusion [V. Istokskaia et al., Comm. Physics 6, 27 (2023)]. In this study, hydrodynamic (HD) simulations with FLASH code were firstly used to calculate densities of preplasma created by picosecond prepulse. Then, Python script was developed to create macroparticles for PIC simulation with various numbers and numerical weights in cells depending on the density of plasma in each cell. This approach of variable numbers and weights of numerical particles at the same time enables to model laser pulse interaction with plasma in a large range of densities in the simulation box in a reasonable computational time. The particle tracking including accelerating fields experienced by ions with relatively high final energies was used in 3D simulation of laser-driven ion acceleration from near critical Gaussian plasma density profile [J. Psikal et al., Plasma Phys. Control. Fus. 63, 064002 (2021)]. Here, the tracking was realized in two subsequent simulation runs as it is impossible to track all particles due to their huge numbers in 3D simulation. In the first run, ID numbers of high energy ions were recorded at the end of simulation. After their random selection, low number of particles was tracked from the beginning of their acceleration in the repeated (second) simulation run from a restart point.
The Extreme Light Infrastructure Nuclear Physics (ELI NP) facility in Măgurele, Romania, runs the currently most powerful laser in the world, capable of delivering two 10 PW pulses each minute. It stands at the forefront of cutting-edge research in laser-driven particle acceleration and high-energy nuclear physics. We present a selected part of our simulation effort where the SMILEI capabilities and the PIC approach in general are tested at the edge regime, in particular:
1. Interaction of a 10 PW pulse with a nickel target: simulations for an ongoing experimental campaign aimed at the measurement of the plasma state evolution within a picosecond after the target irradiation. Roles of field ionisation, collisions and QED effects are discussed.
2. Monochromatic ions from the nanostructured peeler: high-energy, quasi-monochromatic ion beam can be generated by shooting a PW-class laser pulse at the narrow side of a tape target. We further develop the scheme by placing a carbon nanostructure at the non-irradiated edge of the tape to tune it to deliver monochromatic proton and carbon bunches with a narrow energy spread suitable for hotly anticipated medical applications.
3. Ultrabright Gamma Sources: The generation of ultrabright gamma sources is a key research area at ELI NP, with implications for fundamental nuclear physics and medical imaging. The newly implemented NewParticles diagnostic provides a new insight into the process of photon emission.
We will provide feedback to developers and potentially suggest additional ideas for further development.
Amongst the different features and boundaries encountered around comets, one remains of particular interest to the plasma community: the diamagnetic cavity. Crossed for the first time at 1P/Halley during the Giotto flyby in 1986 and later met more than 700 times by the ESA Rosetta spacecraft around Comet 67P/Churyumov-Gerasimenko, this region, almost free of any magnetic field, surrounds nuclei of active comets. However, previous observations and modelling of this part of the coma have not yet provided a definitive answer as to the origin of such a cavity and on its border, the diamagnetic cavity boundary layer. We investigate which forces and equilibrium might be at play and balance the magnetic pressure at this boundary down to the spatial and temporal scales of the electrons in the 1D collisionless case thanks to Smilei. In addition, we scrutinise assumptions made in magneto-hydrodynamic and hybrid simulations of this environment and check for their validity at these scales.
Electromagnetic waves emitted in the solar wind and corona during type III solar radio bursts are studied owing to data provided by PIC simulations computed using the 2D/3V version of the code SMILEI. In a 2D simulation box modeling a plasma with random density fluctuations of average level $\Delta N$ of several percent of the ambient plasma, an electron beam generates Langmuir wave turbulence which in turn radiates electromagnetic waves at the plasma frequency and its harmonics. Several challenging tasks have to be achieved simultaneously : (i) to reduce the numerical noise below $\Delta N$, (ii) to use a box involving both electrostatic and electromagnetic scales, and (iii) to compute very long time series of fields and densities in order to identify the low frequency waves participating in the processes of electromagnetic wave generation.
Nowadays, there is a growing interest in the polarization characteristics of electromagnetic emissions, which are crucial for understanding the processes generating them and diagnosing the solar wind and coronal plasmas. The conducted research consists in modeling virtual satellites moving in a 2D simulation box and recording waveforms of fields and particle densities. Several methods to analyze the waveforms recorded have been implemented, tested, used and compared, enabling us to identify the wave modes emitted (in particular, the electromagnetic waves emitted at frequencies $\omega_p$ and $2\omega_p$) and to determine their polarization characteristics (sense and ellipticity). Statistical studies using 256 virtual satellites have been performed to determine the distributions of ellipticity as a function of time, as well as of magnetic field amplitude and average level of density fluctuations of the ambient plasma. Results obtained show a good agreement with space observations.
We combine the scattered field formalism with Smilei's particle-in-cell code, using the PrescribedField block to model the relativistic dynamics of laser plasmas in complex field configurations. Despite the strong nonlinearity of the interaction, we provide theoretical justifications for the applicability of this method, supported by numerical analysis. We then apply this method to describe electron acceleration under real experimental conditions in a subcritical plasma from tightly focused linear and radially polarized beams. According to our simulations and their analysis, this approach looks very promising for optimizing laser-induced relativistic electron beams at ultra-high intensities.
The interaction of an ultra-intense (>$10^{18}$ W/cm$^2$) laser pulse with a suitable target can result in the acceleration of particles, like protons, and the generation of secondary radiation, like high-energy photons and positrons. Concerning solid targets, the Target Normal Sheath Acceleration (TNSA) [1] is surely the most assessed configuration to accelerate the protons naturally present as impurities on the target surface. To optimise the process, non-conventional targets capable of increasing laser-matter coupling have been developed. One can achieve this goal by growing on top of a solid metallic film a nanostructured foam having an average density in the near-critical regime for the laser [2]. This configuration is referred to as Double Layer Target (DLT). Experimental campaigns confirmed the enhancement of the energy of the protons generated with DLTs when comparing them to conventional targets [3]. Moreover, numerical studies have proposed such a configuration to efficiently generate high-energy (MeV) photons via Non-Linear Inverse Compton Scattering (NICS) [4] and these photons could initiate pair production via the non-linear Breit-Wheeler process during the interaction itself. Particle in-cell (PIC) tools are amongst the most assessed methods to study the interaction of a laser with a target and their integration with Monte Carlo (MC) modules allows accounting for ionisation, photon generation, pair production and collisions. The open-source PIC code SMILEI [5] offers the possibility both to integrate the foam nanostructure using external files and to use some specific MC modules to account for all the aforementioned processes. This contribution aims to present the results of simulations performed in SMILEI of different possible uses of DLTs. The study of the interaction of different TW-class lasers with DLTs is first presented to evaluate their exploitation as compact proton sources. The relatively low intensity of such lasers raises the relevance of the effects of the nanostructure of the foams [6] and of the ionisation processes in laser propagation. A campaign of 2D and 3D simulations to study these effects was performed. The results show that the DLTs are effective in increasing and optimising the maximum energy of the protons (achieving tens of MeV for tens of TW lasers or activating the acceleration for sub-TW ones) and that the ionisation process becomes influential when the laser intensity is relatively low. Secondly, the use of TW/PW-class lasers as sources of high-energy photons and for pair production is considered. Our results show the relevance of DLTs for efficient photon and pair production and the possibility to optimise these processes acting on target parameters like foam density and thickness and solid layer composition.
[1] M. Passoni et al 2010 New J. Phys. 12 045012
[2] M Passoni et al 2020 Plasma Phys. Control. Fusion 62 014022
[3] M. Passoni et al. 2016 Phys. Rev. Accel. Beams 19 061301
[4] M. Galbiati et al. 2023 Front. Phys. 11 1117543
[5] J. Derouillat et al. 2018 Comput. Phys. Commun. 222 351-373
[6] L. Fedeli et al. 2018 Sci. Rep. 8 3834
Capturing the target behavior during a high-intensity laser-solid interaction is crucial to understanding the interplay of fundamental processes such as ionization and plasma kinetics. Moreover, predicting and controlling the pre-plasma evolution produced by the laser rising edge is key for enhancing, for instance, the laser-driven ion beam quality [1]. By monitoring the dynamics of the initial stage of the interaction, we report on a detailed visualization of the pre-plasma evolution. Nanometer-thin diamond-like carbon foils are shown to transition from solid to plasma during the laser rising edge with intensities $I\le 10^{16}$ W/cm$^2$. Single-shot near-infrared broadband probe pulse [2] transmission measurements evidence sub-picosecond dynamics of an expanding plasma with densities above $10^{23} ~\rm cm^{-3}$ (about 100 times the critical plasma density). The unique complementarity of a solid-state interaction model and a kinetic plasma description using the SMILEI PIC code provides deep insight into the interplay of initial ionization, collisions, and expansion [3].
References
[1] S. Keppler et al., Phys. Rev. Res, 4, 013065 (2022).
[2] I. Tamer et al., Opt. Express, 28, 19034, (2020).
[3] Y. Azamoum et al., under review, arXiv:2309.00303 (2023).
We use particle-in-cell simulations performed with Smilei to investigate the autoresonant wakefield excitation. The kinetic simulation reveals significant fluid nonlinearities of the laser self-consistent evolution occur under high plasma density, invalidating the fluid model of quasi-static approximation. However, when considering low underdense plasma, a remarkable agreement emerges between the fluid model and kinetic simulation results. In this regime, the frequency chirp offers effective control over wave amplitude and self-injection of particles. Optimal laser and plasma parameters are identified for amplifying the wakefield to the point of wave-breaking, enabling acceleration of particles via a high-gradient electric field, and $\sim$30 pC charge of the high-energy ($\sim 250$ MeV) electrons is expected to be obtained over $\sim3.5$ mm acceleration length. This versatile and efficient acceleration scheme holds promise for a wide range of applications, from tens to hundreds of MeV energies, making it worthy of investigation as a potentially attractive option for various industrial and medical applications.