Home » Plasmas » Plasmas

A large spectrum of plasmas exists naturally or artificially. Their creation processes, their control and the understanding of their underlying physics are tighly bound to facilities and their up-to-date associated technologies and to numerical simulations of the physical processes at any scale from microscopic to macroscopic scales as well. Here below a non comprehensive list of  those plasmas found is given in various contexts.


LIBS : Laser Induced Breakdown Spectroscopy.  The chemical analysis of a solid material sample can be performed par ablation induced by a laser pulse. This is the so-called LIBS technique, whose acronym stands for for Laser Induced Breakdown Spectroscopy.

Crédit : P.Stroppa – CEA/DSM/Saclay, 10/2005

Plasma reactor : The plasma reactors are used for deposit of multiple layers on thin substrates by cathode spray.
Crédit : P.Stroppa – CEA/DSM/Ganil-Caen

Plasma torche : Plasma torches refer to facilities which via an oscillating magnetic field allow to keep a plasma area at temperatures in excess of 3000 Celsius degrees. These facilities allow to deal with powder-shape waste, muds … made up with intricate combination of chemicals which imped their proccesing by conventional methods such as cement process. They allow to spout nanometer-scale powders or to make a deposit on a material as in the illustration with an Argon plasma.

Crédit : F. Vrignaud – CEA/Le Ripault, 2010

Medical applications

Development of a method of dose calculation for the radiotherapy: The precise and fast algorithms of the dose deposition are necessary for the improvement of the plans of treatment in the radiotherapy. The deterministic model of transport of particles based on the resolution of the equation of Fokker-Planck proposes a better quality of calculation in the heterogeneous tissues such as bones or lungs compared the commercial software “Pencil Beam”. Figure shows the isodose deposited by electrons (red) propagating through the trunk bones showed in gray.

Credit : B Dubroca – CELIA, 2012

ISeult : the most powerful MRI magnet in the world
Since May 2017, a giant magnet  is operated within the research lab NeuroSpin located on the CEA site of Paris-Saclay. It is the main part of MRI (magnetic resonance imaging) with very high space resolution dedicated to the human brain imaging. This exceptional large facility has been designed by CEA engineers and researchers at CEA and the 130-ton weight magnet with 180 kms of supercoductiong cable has been built by Alstom-General Electric – Belfort. It produces a 11.7 teslas magnetic induction over a volume for whole human body imaging.

Crédit photo P.Briet © CEA

Plasmas for inertial confinement fusion

Development of the numerical methods in hydrodynamics of plasmas: The numerical simulations in the domain of inertial fusion are performed with the Lagrangian method. It requires improvements of robustness in order to describe the flows with large vorticity and shear. The figure shows the performance of a new method of grid regularization ReALE (reconnection based arbitrary Lagrangian Eulerian) which allows to build an evolutionary mesh by using the polygons of Voronoï. It adapts itself naturally to big distortions of the flow that take place for example in the development of Kelvin-Helmholtz instability.

J Breil, S. Galera, P-H Maire, CELIA, Computers & Fluids 46, 161 (2011)

Kinetic simulations of stimulated Raman backscattering for shock-ignition: Shock ignition of the inertial fusion operates in the domain of laser intensities where the parametric processes may contribute to the laser energy absorption and hot electron generation. These processes can be described in full kinetic simulations using high performance computers. Figure shows the angular and spectral distribution of the scattered laser light due to the Raman instability.

C. Riconda, S. Weber, V.T. Tikhonchuk, A. Heron, LULI-CELIA-CPHT, Phys. Plasmas 18, 092701 (2011)

Interpretation of shock-ignition experiment on OMEGA laser facility: In experiments conducted on the Omega laser facility we studied the effect of converging shock on the shell compression and the neutron yield. 40 uniformly distributed beams were used to compress a cryogenic Deuterium shell and 20 beams have been focused on a dodecahedron to launch the choc. 2D hydrodynamic simulations show the density and temperature profile of the target stagnation, Rayleigh-Taylor instability modulations seen in the figure were measured experimentally and compared to the code. Agreement on the shape of the modulation is clear however, however the simulations overestimate the shell compression.

W. Theobald et al., Phys, Plasmas 19, 102706 (2012)

Laser smoothing with an underdense foam: In the direct drive approach to inertial confinement fusion, inhomogeneities in the laser beam intensity may seed pressure perturbations and induce hydrodynamic instabilities that undermine the target performance. The use a low density foam on the target front-side enables laser beam smoothing and reduces the hydrodynamic instability growth. Figure presents an optical streak image providing a direct measurement of the ionization front induced by the laser beam in a foam.

Ph. Nicolaet al., CELIA, Phys. Plasmas 19, 113105 (2012)

Magnetized plasmas and magnetic confinement fusion

Instantaneous value of the pressure from numerical simulations of turbulence on tokamak edge: on the picture, the « hole » in the middle shows the plasma center excluded from the simulation. Top: without transport barrier; bottom : with barrier. Microscopic instabilities cannot be ruled out due to gradients of density, temperature and pressure. Those instabilities induce a micro-turbulence in which eddies (convention cells) increase drammatically transport of both matter and heat from the plasma centre towards the edge. This turbulent transport causes the plasma confinement to be spoiled. Nevertheless, Self-organisation of the turbulence can strongly improve the confinement by large scale flow! This self-organization mechanism is systematically use to reach an improved confinement regime by a thin layer located at the plasma edge where turbulence is strongly reduced. This layer is coined transport barrier. But unfortunately this layer is not stable bur relaxes quasi-periodically. In experiments for ITER scale, in order to maintain the heat flux associated to these relaxations within acceptable limits for the wall, control is required. Such a control has been experimentally demonstrated by using a magnetic perturbation brought from outside which create structures called magnetic islands.

Crédit : P. Beyer – Aix-Marseille Université

Dense plasmas

Aluminum plasma: Snapshot of expanded aluminum at 1.4 g/cm3 and 10000K computed with a quantum molecular dynamics code.

Crédit : J. Clerouin – CEA – 2012

Natural plasmas

Heliomagnetism: Numerical simulation of heliomagnetism. (Top) the magnetic field lines in the solar corona, the background representing the radial component of the magnetic field close to the convective area. (Bottom) the longitudinal component of the magnetic field. Stars are big hot fluid spheres, spinning and showing turbulence with numerous convection phenomena and lying within a self-induced magnetic field.

Crédit : A.S.Brun – CEA/DSM/IRFU


Superconducting coil: Within the context of magnetic fusion, one plasma confinment scenario is based upon a toric confinement via a magnetic field generated by non planar coils, according to the stellerator concept. On the picture, a superconducting coil of the W7X syterellator before its asssembly on the facility in Greiswald in Germany.

Crédit : CEA/DSM/Saclay/IRFU

Tokamak wall: In tokamaks, such as the Tore Supra facility in Cadarache (France) in the illustration, used to confine and heat plasmas by means of magnetic confinement, a toric chamber is used which is made of metal or carbon fibers able to withstand large surface density of thermal power.

Crit :P.Stroppa – CEA/DSM/Cadarache, 2009

Deuterium ice injector: Injector of ice pellets in Tore Supra tokamak. This injector is aimed at maintaining the plasma density during a few minutes, by throwing deuterium ice pellets in the plasma.

Crédit : P.Stroppa – CEA/DSM/Cadarache

1 MeV neutral beam injector for ITER: 1-MeV Neutral Beam Injector (NBI) for Plasmas Heating and Current Drive. The extraction of negative ions from the plasma source dedicated to support the European experimental developments around the ITER-Beutral Beam Injector system has been simulated by the ONIX (Orsay Negative Ion eXtraction) code. ONIX code simulates the transport of the negative ions in the plasma source and their extraction by the external electric field. The plasma efficiently screens the extraction electric field and some negative ions are not extracted (on right), compared to the scenario without plasma (on left).

Crédit : LPGP / CNRS-UniversitParis-Sud

Modeling of laser-induced damage in optical materials by nanosecond pulses: High power laser consist of optical materials which may damage due to intense laser flux. In order to increase the damage resistance, physical processes responsible for laser-induced damage are studied. In particular, a sub-sonic absorption front is shown to be induced due to the coupling between defect formation and heat transfer. The evolution of the position of the absorption front with respect to time is shown in the figure.

G. Duchateau, M.D. Feit, and S.G. Demos, J. Appl. Phys 111, 093106 (2012)

Cooling of superconducting magnets: The HELIOS loop (Helium loop for high load smoothing) is developed to study how the variable thermal loads received by the the cryogenic systems of the future fusion reactors JT60-SA and ITER can be smoothed. For the japanese facility JT60-SA, the goal is to absorb around 12 kW peak power with a 6 kW average cooling power. The HELIOS facility runs on forced convection of supercritical Helium at 4.4 K temperature and under 5 bars pressure.

Credit : CEA/DSM/Grenoble/INAC, 2010

X-ray source: By impinging a laser piulse on a target, a plasma is created and can be transformed into a source of extreme UV Exulite radiation used for photolithography purpose.

Crédit : P.Stroppa – CEA/DSM/Saclay, 10/2005


Superconducting magnets: the magnetic field si generated by making a current circulate in superconductiing coils. Her eshown, the magnetic conception of the 90 mm double aperture quadrupole fo the Hilumi LHC project. The coils are made with NbTi, the frettage clamps with austenitic steel and the head with steel. The shematic drawing shows the magnetic field amplitude within normal use conditions.

Crédit : CEA/DSM/Saclay/IRFU

Superconducting cavities: they show an electromagnetic field which accelerate the beam with optimized efficiency. They are made with  Niobium and inserted intoa cryomodule where they are held at 4 K (-269 °C) temperature, even 2 K, by liquid Helium. The electromagnetic field is injected through a RF power coupler.

Crédit : CEA/DSM/Saclay/IRFU

Steaming of a superconducting cavity: the performances of the superconducting cavities are directly impacted by the purity of its Niobium surface. This quality can be improved by steaming the cavity, i.e. by heating it up to 120° under ultrahigh vacuum conditions during 48 hours.

Credit: CEA/DSM/Saclay/IRFU

High intensity proton injector: Testing materials able to confine the plasma inside the future fusion reactors requires high-current particle accelerators. The facility IPHI (Injector of Protons at High Intensity) is a prototype of the low energy part (up to 3 MeV) of a high power accelerator of light ions. This demonstrator is aimed at showing the faisability of such accelerators by focussing on a delicate issue, namely the injector, which needs to deliver a continous proton beam with outstanding optical qualities. The main component of this injector is a 6-m long accelerating cavity denoted FRQ (Radio Frequency Quadrupole).

Crédit : V. Hennion – CEA/DSM/Saclay/IRFU

SPIRAL2 facility: this 2nd generation facility is aimed at producing radioactive ions. It is currently in the implementation phase at GANIL at Caen, Normandy (Grand accélérateur national d’ions lourds). The resulting ion beam is a stable high intensity beam which can be sent on target. The beam-target collisions are expcted to produce exotic nuclei, scarcely visible on earth.

Crédit : CEA/DSM/Saclay/IRFU

Beam display in 2D phase spaces: the beam is a cold non neutral plasma with a relativistic velocity. The beam dynamics is the brahc of physics which tries to model how evolves its momentum distribution function along the accelerator axis  within the 6D-geometry phase space (3 for particle positions, 3 pour their momentum).

Crédit : CEA/DSM/Saclay/IRFU

ECR (Electron Cyclotron Resonance) source: a plasma is created and accelerated by a radiofrequency electromagnetic field in a gas. The particle orbits are extended and synchronized with the RF field by a static magnetic field. The beam is extracted through a small hole via a potential difference held between the plasma chamber wall and the extraction electrodes.

Crédit : CEA/DSM/Saclay/IRFU

Ultrahigh laser intensities

Accélérateur de particules par onde plasma: like any medium, the plasma can be crossed by waves.  More specifically longitudinal waves driven by electrons only, ions kept fixed. These waves show a sequence of areas where electrons are compressed and dilated, which induces an electric field oriented along the wave propagation direction. In cas when one electron differs from the others with a low relative velocity regarding the phase velocity of the wave, the electron can be located in a accelerating area of the electric field. This electron is going to surf as a usual wave surfer who feels a push along the wave propagation of the wave, the usual  swimmers oscillating localling. One such wave can be created in the wake  of a laser pulse:  this founds the concept of plasma driven particle accelerator. Whether the accelerating wave slows down like the swell close to the shore (resp. the plasma wave is amplified by a laser wave), the set of water droplets (resp. the electrons) will have their movement orbits cross each other. The swell (resp. the plasma wave) cannot propagate anymore: this is the so-called wave-breaking.

Credit : A. Leblanc

Proton acceleration: Numerical simulation of proton acceleration by impinging a ultrashort (150 fs) ultra high iradiance (7 1019 W/cm²) laser pulse on a 5 mm long hydrogen plasma with uniform 1022 cm-3 density. The ion density is displayed as a function of space-time (x-t), with a 2-decade log scale. The laser wave propagation direction is aligned on the red arrow. The linear polarization (left) induces a strong electron heating (thermal energy 2.1 MeV) ; a soliton deacting into 3 oscillatins is seen to propagate up to the time when the target is dispersed by electron-driven ion expansion. With circular polarization (right), the electrons are heated up to 120 keV only. The laser pulse pushed the ions, as a piston. When the laser pulse shuts down, the ions which lay inside the target stay immobile, whereas the other ions produce a monokinetic jet which crossed the target.

Credit: G. Bonnaud – CEA/INSTN, 2010

X-ray amplification by Raman effect in a free electron laser: The free electron lasers (FEL) are efficient and tunable X-ray sources. But their size and associated high cost make their use difficult. A new scheme, much more compact is proposed. Here, the relativistic electron beam comes from the acceleration by the electric field of a plasma wave generated in its wake by an ultrashort intense lase pulse, and the X-radiation is created by means of an optical grating, created by making two laser pulses interfere.

I.A. Andriyash et al., Phys. Rev. Lett. 2012

Synchronized harmonics and electron generation: When an ultraintense laser pulse (in excess of 1019W/cm²) impinged upon a solid target, the target is immediately ionized. The created plasma behaves like a mirror reflecting the incident laser pulse. Actually the incident laser field make this electronic mirror quiver; as a result the reflected field is modified by Doppler effect, for every optical cycle. His spectrum is broaden toward XUV domain via high order harmonics (bottom image). Meanwhile electrons leave the plasma miroir with relativistic energies (top) towards vacuum on the edage of the reflected pulse (left). The plasma mirrors appear to be the source of both XUV pulses, and multiMeV electron beams and to be well fitted to probe the matter by two synchronized particle and light beams.

Credit: CEA/Saclay/LIDyL

Relativistic laboratory astrophysics: Laser-induced electron-positron pair collisions : Particle-in-cell simulation of the collision between two e e+ pair jets (of Lorentz factor ≃500)  issued from solid foils irradiated at extreme laer intensities (1024 Wcm-2). A fast-growing magnetic filamentation instability is triggered in the overlap region, which decelerates and thermalizes the incoming jets, notably through strong synchrotron emission. (a) Experimental setup; (b) Density of the rightward-moving positrons 1021  cm-3; (c) Magnetostatic fluctuations 104 T. Crédit : L. Gremillet – CEA/DAM/DPTA

Ultrashort pulse laser: When focussed on tiny spots, powerful laser pulses allow to achieve very high irradiances (about 1020 times the peak sun irradiance on earth). Any piece of matter located in the light spot becomes a plasma where extreme phenomena take place: acceleration of electrons and ions up to tenths of MeV energy, hard X ray radiation, nuclear reactions, …. To achieve such irradiances, the laser power nedds to be amplified in excess of tenths even hundreds of terawatts by means of optical compressors, like in the illustration with the UHI10 laser at CEA/Saclay. This compressor encompases optical gratings, to shorten the light pulses by a factor 10 000 (from nanosecond to subpicosecond durations) and consequently increase by the same factor the delivered power.

Credit:  P.Stroppa – CEA/DSM/Saclay, 10/2005

Attosecond flash lighthouse : electromagnetic fields within ultraintense light pulses induce a relativistic dynamics to electrons et give rise to many questions: how to gauge this dynamics? How to generate and control intense and ultrashort light sources (100 attoseconds = 10-16 s or below). Theoreticians and experimentalists of the High inensity physics (PHI) headed by F. Quéré try to answer these questions via the UHI100 laser (100TW – 20 femtoseconds) and numerical simulations on supercomputers. As illustrated by simulation results, a solid target impinged upon by a light pulse becomes a dense plasma which behaves  like a mirror reflecting the incident light. The laser-plasma interaction generates an harmonics spectrum of the incident light associated in the time domain to a sequence of attosecond (atts) pulses. The plasma mirror (bottom) appears to bend due to the inhomogeneous radiation pressure at he laser focus and focuses the harmonics. In order to isolate a single atts pulse, more useful to time-resolved experiments, we shaped the space-time profile of the incident laser pulse (top left) so that the wavefronts turn in time and generate as pulses within different directions (top right for the reflected pulse spectrum). This ultrashort light pulse is one of three parts of the European project Extreme Light Infrastructure ; actually a X-UV atts light in under construction in Hungary.

Credit: CEA/Saclay/LIDyL

Electron-positron pair production by ultra-intense laser pulses : The figures display the results from a particle-in-cell simulation of the interaction of a 30 fs duration, 1024 Wcm-2 intensity laser pulse with a 1 μm thick Al target (with a 3 μm density ramp): evolution of the laser electric field (Ey), ion density (ni) and positron density (n+). The positron generation proceeds via two stages: the synchrotron emission by relativistic electrons oscillating in the intense laser field produces a copious number of γ-ray photons (carrying up to 50% of the laser energy); through their subsequent interaction with the laser, these energetic photons can decay into e e+ pairs via the Breit-Whheeler process. Crédit : L. Gremillet – CEA/DAM/DPTA