Particle Acceleration in Multiple Dissipation Regions - PowerPoint PPT Presentation

1 / 23
About This Presentation
Title:

Particle Acceleration in Multiple Dissipation Regions

Description:

Matern class PSD with index a=1.5. Threshold current jc=encs exceeded in 5% of volume ... Proton lifecycle in the presence of an adiabtic invariant Py ... – PowerPoint PPT presentation

Number of Views:65
Avg rating:3.0/5.0
Slides: 24
Provided by: lif87
Category:

less

Transcript and Presenter's Notes

Title: Particle Acceleration in Multiple Dissipation Regions


1
Particle Acceleration in Multiple Dissipation
Regions
  • Kaspar Arzner (PSI/ETHZ)
  • Loukas Vlahos, Heinz Isliker, Fabio Lepreti
    (AUTH)
  • Bernard Knaepen, Daniele Carati, Nicolas Denewet
    (ULB)

2
(No Transcript)
3
Acceleration environment
hot spots with E.jgt0
  • True MHD turbulence Gauss proxy

4
Gauss Proxy for Dissipation Regions
magnetic field lines
Sample trajectory
  • B0 20G, dBx dBy 20G, dBz 100G
  • Outer scales
  • lxly1km, lz20km
  • Matern class PSD with index a1.5
  • Threshold current jcencs exceeded in 5 of
    volume

5
Simulations at PSI
  • Merlin cluster
  • 56 nodes
  • Myrinet / Ethernet
  • code f90
  • MPI / MPICH

6
Diagnostics Anomaleous Diffusion
  • What quantity to consider? P mgv is a good
    candidate since it is unbounded and incremented
    by the equation of motion.
  • For initially Maxwellian protons (T106K) we find
  • EW / ED lt 103 ordinary diffusion.
  • EW / ED gt 5 .103 subdiffusion.

7
EW / ED 2.102 ltP2gt t
100 keV
EW / ED 105 ltP2gt t0.8
time
8
Proton acceleration efficiency
200 ms
Terminal Proton Energy
rms dissipative field
9
Proton lifecycle in the presence of an adiabtic
invariant Py
Reason ltvyEygt lt(Ay - Py)DAygt f(e,Py)
10
Electron Acceleration
sample
mean
11
Distribution of Energy Jumps
Simulation
Best-fit Levy stable distribution
12
Close-up dominant B field direction is z
(x,y) Brown motion OK z jump/drift
dominated
13
Also existing, and definitifely not Brownian
Adiabatic/bounded trajectories
14
Velocity vs. Waiting time
Prob(Dt) f(Dt L/v)
15
Shape of f(Dt v / L)
Exponential 1D Poisson nearest-neighbour
distance. (1D because of ltdBz2gt 10 ltdBx,y2gt)
16
Summary
  • Conversion of electromagnetic energy to other
    forms implies j.E gt 0. The j-parallel component
    of E is responsible for acceleration.
  • Present study super-Dreicer test particles in a
    Gauss proxy for strong resistive MHD turbulence
  • Cross - (rms) field diffusion
  • (rms) field-aligned jumps
  • Fokker-Planck may be insufficient

References
Arzner, K. Vlahos, L., ApJ (2004) 605, L69
17
(No Transcript)
18
(No Transcript)
19
-gt Montroll-Weiss type jump model
(Vlahos, Isliker, Lepreti 2003)
P(v,t -gt v,t) Pv(v-v) Pt(t-t)/t(v)/t(v)
t(v) collision time with dissipation regions
20
Some exact Results (last-but-one slide!)
(q1/t )
21
Resumee
  • Super-Dreicer electrons perform energy jumps
    when dissipation regions are encountered
  • On macroscopic (observable!) time scales, their
    orbit is thus a jump process rather than Brownian
    motion with continuous sample paths.

22
EW / ED 105
t 0.8
23
Normal Diffusion
100 keV
Initially maxwellian protons with T 106 K.
Write a Comment
User Comments (0)
About PowerShow.com