space charge in high current lower energy electron bunches · space charge in high current lower...
TRANSCRIPT
Space Charge in High CurrentLower Energy Electron Bunches
Seminar DESY, 29.05.2007, S. Becker,T. Eichner, M. Fuchs, F. Grüner, U. Schramm,
R. Weingartner, D. Habs
● Experimental Status
● Self acceleration:
– Visual explanation
– Renormalization of the electron mass
● GPT: The applied calculation method
● Retardation
● Simulating 2 GeV, 1 nC
– Point 2 Point and Poisson-Solver still differ
– Convergence of Point 2 Point
● Simulation Benchmark: Energy conservation assuming mass renormalization
Experimental Status: Beam transport
Quadrupoles:
- Aperture 6 mm (5 mm effective)- Field gradient 503 T/m
Measurements at MaMi Mainz:
Experimental Status: Undulator
Test Undulator:
- 60 Periods- 5 mm period length- 0.9 T field on axis at a gap of 2.5 mm
1st Harmonic 3rd Harmonic
Measurements at MaMi Mainz:
laser pulse
~ 1 µm only!!~ nC charge
typical length scale = plasma wavelength
Experimental Status: Electron acceleration
Electron acceleration at MQP:
Experimental Status: Electron acceleration
● σx = σ
y = σ
z = 1 µm
● Charge: 1.25 nC -> I = 150 kA● Initial distribution: Gaussian
“Extreme” case:
The Regime:
Self Acceleration
Self Acceleration
Considering a beam expansion due to emittance > 0, neglecting space charge.Beam evolves “Pythagorean”Particles have uncorrelated energies
s= 2
2⋅L
Considering a beam propagation with no emittance, neglecting space charge.
Self Acceleration
z ' = ⋅ zDue to relativistic contraction, the longitudinal dimension becomes larger:
Considering a Coulomb explosion: Assuming x = y = z in the lab frame.
= 260
“Explosion” is transversal dominated.
Self Acceleration
Considering a beam expansion with no emittance, respecting space charge.
Electrons off axishave to gain speed,hence gain energyto “keep up” with theelectrons on axis.
Self Acceleration
Quantitative approach: Compressing electrons correspondsto a gain of potential energy,which can be expressed in terms of a mass:
E = m c2
me
The self field can be considered as a mass itself,which “dissolves” as the beam expands:
m f
E f = m f c2 =U '
Momentum before Expansion:
Momentum after Expansion:
Total energy in theelectrons' mean rest frame:
E ' = E0E f = m0c2E f = m0m f c
2
E ' = m' c2 = ' m0 c2
Potential energy leads to a mass of the field: E f = m f c2
Mean Lorentz factor due to Coulomb explosion:
' = 1m f
m0
Electron's rest frame:
E = 0m' c2 = 0m0c
2⋅1m f
m0
E = 0 ' m0c2 = m0c
2
= 0⋅ '
Laboratory frame: Electron bunch leaves the plasma with 0
Total energy:
after potential energy “released”Kinetic energy gainsby a factor of : '
after potential energy “released”
Mass Renormalization
(All energies E are normalized on one electron)
Potential energy in average is estimated to be of the order of 74 keV / electron
Self Acceleration: Lab Frame
Space charge (in the mean rest frame) acts transversally.
Where does the longitudinal momentum come from in the Laboratory Frame?
F E=d pdt
∥I
B =0I
2R
F res
GPT
● GPT adapts time steps dynamically -> energy conservation● Calculation precision is defined by relative momentum changes
stepwise● Space charge is considered point to point using Lorentz
transformation. Calculation time t ~ N²
For equation of motion:
Interaction in rest frame:
From Lorentz transformations:
Discussion of limitations due toRetardation Effects
"Event horizon"
Going to the electron bunch's mean rest frame:
~ 1
r2
The contribution of close neighbors is strongest
Assuming the bunch is “born” instantaneously- fields did not “spread” yet:
Discussion of limitations due toRetardation Effects
Possibilities of respecting retardation in calculations:
Interaction is described in terms of
Possible solution:Interaction only on case of " r' < Event Horizon "
Simulating 2 GeV, 1 nC
time=2e-009
Avg(G) = 4014.32 Std(G) = 12.0781
0.599582 0.599584 0.599586 0.599588GPT z
3960
3970
3980
3990
4000
4010
4020
4030
4040
4050
4060
4070
4080
G
Result of Poisson-Solver:after propagation of 60cm:
=4014/=0.003
E=2 GeV=3880 /=0.001
Initial distribution:
time=2e-009
Avg(G) = 4026.51 Std(G) = 17.1934
0.599582 0.599584 0.599586 0.599588GPT z
3980
3990
4000
4010
4020
4030
4040
4050
4060
4070
4080
G
=4027 /=0.0044
Result of Point2Pointafter propagation of 60cm:
Simulating 2 GeV, 1 nC
Simulation of designedbeam line for E = 2 GeV
Triplet focusing-Waist at approx. 5m
0 . 0 0 . 5 1 . 0 1 . 5 2 . 0 2 . 5 3 . 0 3 . 5 4 . 0 4 . 5 5 . 0 5 . 5
G P T a v g z
0
2 0
4 0
6 0
8 0
stdG
z
3.000 macro particles1.8%
0 . 0 0 . 5 1 . 0 1 . 5 2 . 0 2 . 5 3 . 0 3 . 5 4 . 0 4 . 5 5 . 0 5 . 5
G P T a v g z
0
2 0
4 0
6 0
8 0
std
G
z
10.000 macro particles 1.1%
Due to adaptive time steps,scaling is worse than
t ~ n2
Did not reach convergence
-> End of Point 2 Point ?!
0 . 0 0 . 5 1 . 0 1 . 5 2 . 0 2 . 5 3 . 0 3 . 5 4 . 0 4 . 5 5 . 0 5 . 5
G P T a v g z
0
2 0
4 0
6 0
8 0
std
G
Simulating 2 GeV, 1 nC
GPT-Poisson solver:
Energy spread decreases!
TODO: Try other solvers:
ASTRA
Simulation Benchmark: Energy conservationassuming mass renormalization
"Benchmark test" for the "extreme case": 130 MeV 1.2 nC: σx = σ
y = σ
z = 1 µm
0 1 0 2 0 3 0 4 0 5 00 , 0 0
0 , 0 1
0 , 0 2
0 , 0 3
0 , 0 4
0 , 0 5
0 , 0 6
0 , 0 7
0 , 0 8
Ek i n
Up o t
Up o t
+ Ek i n
E/
Me
V
t / a . u .
R e s t - F r a m e P 2 P
Mean rest frame:
0 , 0 0 , 2 0 , 4 0 , 6 0 , 8 1 , 00
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 4 0
1 6 0
E/
Me
V
z / m
Ek i n
Up o t
Up o t
+ Ek i n
P 2 P
Laboratory frame:
Point 2 Point:
simulation Benchmark: Energy conservationassuming mass renormalization
"Benchmark test" for 2 GeV 1.0 nC: σx = σ
y = σ
z = 1 µm
0 , 0 0 , 5 1 , 0 1 , 5 2 , 02 0 6 0
2 0 6 5
2 0 7 0
2 0 7 5
2 0 8 0
E/
Me
V
z / m
P 2 P
Ek i n
+ Ep o t
Ek i n
Point 2 Point
0 1 2 3 4 5
2 0 6 2
2 0 6 4
2 0 6 6
2 0 6 8
2 0 7 0
2 0 7 2
2 0 7 4
2 0 7 6
2 0 7 8
2 0 8 0
2 0 8 2
Ek i n
+ Ep o t
Ek i n
E/
Me
V
z / m
P o i s s o n - S o l v e r
Poisson-Solver
U ' = 180
∑i≠ j
qi q j∣r ' i−r ' j∣
rx '=r xr y '=r yr z '=z r x
pz=zm0 cz '≈ pz /m0 c
simulation Benchmark: Energy conservationassuming mass renormalization
Calculation of potential energy:Lorentz Transformation into mean beam rest frame:
E pot = zU '
Summary
Thank you !!!
● Experimental Status
● Self acceleration:
– Visual explanation
– Renormalization of the electron mass
● GPT: The applied calculation method
● Retardation
● Simulating 2 GeV, 1 nC
– Point 2 Point and Poisson-Solver still differ
– Convergence of Point 2 Point
● Simulation Benchmark: Energy conservation assuming mass renormalization