Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
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