Laser envelope model

[najlkin]

Hi, have you considered implementing an envelope model of laser? Fundamentally, the laser is described by the slowly varying envelope function modulated by the harmonic term. Only the equation for the envelope is then solved (explicitly or implicitly), easing the restriction on the time step given by the CFL condition of classical FDTD solvers. More about this topic can be found for example at the website of Smilei, which has an implementation of such model. I guess many people in LWFA would appreciate it, where the effect of the ponderomotive force remains almost the same when considering reasonably long pulses. However, as you may know, my reasons are somewhat different. I am mainly interested in hybrid methods combining continuum and particle approaches, where this may help to get from the microscale to larger scales, while still retaining most of the physics for low intensities of the laser.

[ax3l]

Hi @najlkin,

thanks for the idea! Yes, I saw laser envelope models since quite some time but didn't get to it. Would you be interested to work on this? We could support you on the endeavor.

cc @BeyondEspresso @PrometheusPi you might want to jump in

[najlkin]

Well, I am totally busy right now. I hoped that someone may has started working on that, because it is not related to the hybrid stuff only (or does someone want to start right now, guys?). However, it is the only way to leave the restrictive time stepping of PIC, so I will have to something about that when working on the hybrid code. It also does not make sense to take immediate fields for continuum models, because one could introduce aliasing effects, so average fields are needed instead. The direct interaction with the envelope is then the most natural description to my opinion. Anyway, it would be cool to have it in PIConGPU. Explicit or implicit that does not matter that much, I guess.

[PrometheusPi]

Hi @najlkin,

PIConGPU would definitely benefit from such a model approach. As @ax3l mentioned already, there are various solutions to this already around - but so far no one had time to implement them in PIConGPU.

Since I am also very interested in this kind of model, since it would ease my work on scattering effects in LWFA, I will assign me to this issue for now. I will have a look into including it in PIConGPU - but if someone wants to take over - please feel free to do so.

[Z10Frank]

Hello @najlkin and @PrometheusPi, in the Appendix of https://iopscience.iop.org/article/10.1088/1361-6587/ab49cf we describe in detail the implementation of the envelope model mentioned by @najlkin, maybe it can be useful if you decide to implement this particular model in PIConGPU. One of its advantages is that it is relatively easy to develop in a pre-existing PIC code structure.

Hope it can help

[sbastrakov]

Thanks for sharing @Z10Frank !

[najlkin]

Thanks @Z10Frank , looks interesting definitely and might come useful :wink: