RevModPhys.85.751

Have been investigated as a trade off approach as

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: mentioned previously, special target materials may be used to produce plasmas with density close to nc (for laser wavelengths  $ 1 m) in order to enhance the generation of hot electrons which drive TNSA (see Sec. IV.D). Gas jet targets have also been used both with  $ 1 m lasers to investigate ion acceleration in underdense plasmas (see Sec. IV.D) and with CO2 lasers ( $ 10 m) for studies of RPA and CSA in moderately overdense plasmas (see Secs. IV.A and IV.B). Apart from the possibility to vary the background density, using flowing gas jets as targets is of interest because they enable the interaction with a pure proton plasma and are suitable for high-repetition rate operation as needed for most foreseen applications (see Sec. V). D. Particle-in-cell simulations The PIC method (Dawson, 1983; Birdsall and Langdon, 1991), mentioned in the Introduction, is the most widely used approach to the kinetic simulation of plasmas. The PIC method provides a solution to the Maxwell-Vlasov system using a Lagrangian approach, with fields and currents allocated on a fixed grid and the phase space represented by an ensemble of computational particles. Thus, the PIC method is mostly appropriate to describe collisionless laser-plasma interaction dynamics, although models are available to implement either collisions [see, e.g., Fiuza et al. (2011), and references therein] or ionization [see, e.g., Petrov, Davis, and Petrova (2009), and references therein]. PIC simulations of laser interaction with solid-density plasmas at peak densities typically exceeding 102 nc are a very demanding task even when the most powerful supercomputers are used. As a minimum requirement, one has to resolve temporal scales $!À1 and spatial scales $c=!p p where !p $ n1=2 . Thus, when approaching parameters of a e real experiment, relevant lengths such as the laser beam waist may correspond to thousands of grid points in each spatial direction and typical dynamic times to thousands of time steps. In addition, kinetic effects such as the generation of hot electron tails in the distribution function and large density variations need very large numbers of particles to be properly resolved. For these reasons, realistic 3D simulations with proper resolution are typically beyond computational possibilities. These constrains result, in most of the cases, in either using a reduced dimensionality or in relaxing the actual parameters to some extent, e.g., by assuming relatively low densities or short scales. For some peculiar problems, development of hybrid modeling may be appropriate, as discussed in Sec. III.C.4. Despite the above mentioned limitations, several groups have been able to perform large-scale 3D simulations relevant to ion acceleration regimes such as, e.g., TNSA (Pukhov, 2001), RPA (Esirkepov et al., 2004; Tamburini et al., 2012), and BOA (Yin et al., 2011a). The use of parallel supercomputers has also allowed extended multiparametric studies (see Fig. 8) aimed to infer sca...
View Full Document

This document was uploaded on 09/28/2013.

Ask a homework question - tutors are online