Erzeugung intensiver Protonenpulse durch Kompression

Size: px

Start display at page:

Download "Erzeugung intensiver Protonenpulse durch Kompression"

Stephanie Carroll
6 years ago
Views:

1 Erzeugung intensiver Protonenpulse durch Kompression L.P. Chau, M. Droba, O. Meusel, D. Noll, U. Ratzinger, C. Wiesner MAMI-Seminar, Mainz, Germany 2010/05/06

2 Outlines: o FRANZ: Overview & Status o Bunch Compressor: Design & Optimization Concepts: Geometry 5MHz Kicker Single Bunch Beam Dynamics Merging Scenario Rebuncher Cavities Magnet Design Beam Dynamics with Realistic Field Distributions o Outlook 2

Frankfurt Neutron Source at the Stern-Gerlach-Zentrum FRANZ

Measurement of neutron capture cross sections, nuclear

3 Frankfurt Neutron Source at the Stern-Gerlach-Zentrum FRANZ Development of new accelerator concepts for intense proton and ion beams. High intensity proton beam 7 Li target Neutron beam Measurement of neutron capture cross sections, nuclear astrophysics, material research and detector development. Development of high power targets. Expertise of IAP in linac design Technology and knowledge transfer from Karlsruhe. P. Fischer, A. Schempp 4-rod RFQ U. Ratzinger IH-DTL 4π BaF 2 Detector 3

dissolved. Technology and knowledge transfer to GSI and Goethe-University.

4 Who is FRANZ? Continuation and extension of the physics programme started around 1980 at the 3.7 MV Van-de-Graaff Laboratory at FZ Karlsruhe. FRANZ Käppeler Franz Käppeler retired since 2008: his research group is dissolved. Technology and knowledge transfer to GSI and Goethe-University. FRANZ will increase the intensities by about 3 orders of magnitude. Allows investigations on radioactive samples with N about : Sub-milligramm sample of short-lived isotopes. 4

5 Comparison with existing intense neutron sources Facility FZ Karlsruhe DANCE, Los Alamos n_tof, CERN GELINA, Geel ORELA, Oak Ridge Elbe Dresden FRANZ, Frankfurt Neutron flux at sample position ) [cm -2 s -1 ] Repetition Rate [Hz] Flight path [m] Pulse Width [ns] Neutron energy range [kev] th , th th th (500) ) Integrated flux between 1 kev and 100 kev only 5

6 Planed astrophysical Experiments at FRANZ The Frankfurt neutron source will provide the highest neutron flux in the astrophysically relevant kev region (1 500 kev) worldwide. Factor of 1000 higher than at FZK!!! Neutron capture measurements of small cross sections: Big Bang nucleosynthesis: 1 H(n, ) Neutron poisons for the s-process: 12 C(n, ), 16 O(n, ), 22 Ne(n, ). ToF measurements of medium mass nuclei for the weak s-process. Neutron capture measurements with small sample masses: Radio-isotopes for -ray astronomy 59 Fe(n, ) and 60 Fe(n, ) Branch point nuclei, e.g. 85 Kr(n, ), 95 Zr(n, ), 147 Pm(n, ), 154 Eu(n, ), 155 Eu(n, ), 153 Gd(n, ), 185 W(n, ) Ref.: René Reifarth (GSI / U. Frankfurt) 6

7 Frankfurt Neutron Source FRANZ dc extraction & transport cw operation I b ~ 3-30 ma pulsed operation, rep. rate 250 khz, t = 1ns <I b > ~ 2 ma Activation Mode Neutrons: n / cm 2 s Compressor Mode Neutrons: 10 7 n / cm 2 l=0.8m Operation Modes of the Frankfurt Neutron Source FRANZ. 7

$Nörenberg Operation mode dc Ion species / fraction Protons / 90 % Discharge power 10 12 kw Extraction current 200 ma Extraction voltage 62 kv Extraction field strength 5 kv/mm Beam energy 120 kev$

8 Volume-Type Ion Source K. Volk, R. Nörenberg Operation mode dc Ion species / fraction Protons / 90 % Discharge power kw Extraction current 200 ma Extraction voltage 62 kv Extraction field strength 5 kv/mm Beam energy 120 kev Input emittance (norm. rms) 0.07 p mm mrad Aspect ratio 0.2 Cross-sectional view of the ion source Filament driven discharge: high brilliance ion beam Filter magnet: high proton fraction First beam tests are running. K. Volk, W. Schweizer 8

Ion Beam Extraction v 2eU m p simulated beam extraction using a pentode system extracted beam current with 3% noise (simulated) n p I 1 A I I K peak 3/ 2 2pe v r 4 p 0 2q U I0 b proton

9 Ion Beam Extraction v 2eU m p simulated beam extraction using a pentode system extracted beam current with 3% noise (simulated) n p I 1 A I I K peak 3/ 2 2pe v r 4 p 0 2q U I0 b proton density n p = m -3 gen. Perveance K = compression ratio = 1 Pentode extraction system: extraction properties independent to end energy Space charge compensation adjustable. 9

= 100 ns rep. rate = 250 khz scheme of LEBT section aperture 100 mm, B z = 0.

10 Low Energy Beam Transport C. Wiesner, O. Meusel dc extraction & transport W b,max = 120 kev I b,max = 200 ma pulsed beam operation t min = 100 ns rep. rate = 250 khz scheme of LEBT section aperture 100 mm, B z = 0.6 T Double telescopic system. Solenoidal transport section to provide space charge compensation. Pulsed Wien-filter: DC-beam => 100ns macro bunches with 250kHz repetition. 10

11 Pulsed Wien-Filter: E B-Kicker C. Wiesner scheme of LEBT section Kicker Frequency Solenoid2 Solenoid3 Dipole C-magnet U max Beam Energy 250 khz 422 mt 537 mt 60 mt 300 mt ± 6 kv 120 kev Static magnetic field, temporally compensated by electric field. C-magnet : deflection enhancement. Fast switch array (MOSFET) + nano-crystaline tape wound core. 11

section Kicker Frequency 250 khz Solenoid2 Dipole Pulsed

C-magnet U max 300 mt ± 6 kv DC-Beam Half-Pipe Septum Beam

Switch Array Static magnetic field, temporally compensated

12 Pulsed Wien-Filter: E B-Kicker C. Wiesner C-Magnet Beam Dump Solenoid3 RFQ scheme of LEBT section Kicker Frequency 250 khz Solenoid2 Dipole Pulsed Beam Solenoid2 422 mt Solenoid3 537 mt Dipole 60 mt C-magnet U max 300 mt ± 6 kv DC-Beam Half-Pipe Septum Beam Energy 120 kev Deflector incl. Switch Array Static magnetic field, temporally compensated by electric field. C-magnet : deflection enhancement. Fast switch array (MOSFET) + nano-crystaline tape wound core. 12

13 Radio Frequency Quadrupole - RFQ A. Schempp / NTG company RFQ test module RFQ technical design Beam dynamics Design studies Construction in progress Power test at scaled model proton source LEBT RFQ 13

14 Focusing, Compression and Acceleration d = 0.51 m l RFQ = 1.7 m l IH = 0.6 m Proton Beam 100ns P 150kW P 45kW W b = 120 kev f 0 = 175 MHz W b = 700 kev W b = 2 MeV RFQ: Micro bunching + pre-acceleration. RFQ-IH combination (coupled cavities): one power amplifier, shorter drift space. IH-cavity: main acceleration, KONUS dynamics. 14

Metz H 111 H 211 8 gap and internal msq triplet output beam energy 2MeV IH-cavity: main

15 IH-DTL and CH-Rebuncher final energy 2 MeV U. Ratzinger, M. Heilmann energy variation ± 0.2 MeV H. Podlech, A. Metz H 111 H gap and internal msq triplet output beam energy 2MeV IH-cavity: main acceleration. CH type cavity 4gap KONUS (Combined Zero degree structure) dynamics: high efficiency acceleration. CH-cavity: rebunching and variation of end energy (activation mode). 15

16 FRANZ: Requirements at the Target 175MHz-DTL N bunch = 9 rep.rate = 250kHz E ~ 2.0MeV I = 150mA s 0ns ΔT = ns => ΔT rms 1ns A (beam at target) < 3x3cm 2 Makro Bunch Mikro Bunch I (per pulse) 8A 100ns ΔW < ± 5% 16

17 High Power Target at FZ Karlsruhe and KALLA-Laboratory Intensity D. Petrich, F. Käppeler r = 10 mm y profile x / mm transverse beam profile (simulated) target prototype for beam power up to 6 kw avg. power ~ 4 kw peak power ~ 20 MW Scaled prototype was built and tested. D. Petrich et al., Nucl. Instr. Methods A 596, (2008) 17

18 4πBaF 2 Detector Array 4p colorimetrical measurements: good energy resolution. high granularity (#42): signal to noise ratio. fast timing ( < 1ns) to achieve acceptable TOF resolution. suppression of other-reaction-signals E n (kev) 80 cm flight path 10 cm inner detector radius E max = 200 kev E min = 1 kev prompt flash other reactions (n, ) on sample TOF (ns) M. Heil et al., Nucl. Instr. Methods A459, 229(2001). R. Reifarth et al., Publication of Astrophysical Society of Australia, 26, (2009). 18

19 Systems Perspective source is constructed LEBT vacuum tests RFQ test module First Beam 2010 compressor design high power target test detector reassembled 19

20 Outlines: o FRANZ: Overview & Status o Bunch Compressor: Design & Optimization Concepts: Geometry 5MHz Kicker Single Bunch Beam Dynamics Merging Scenario Rebuncher Cavities Magnet Design Beam Dynamics with Realistic Field Distributions o Outlook 20

21 Outlines: o FRANZ: Overview & Status o Bunch Compressor: Design & Optimization Concepts: Geometry 5MHz Kicker Single Bunch Beam Dynamics Merging Scenario Rebuncher Cavities Magnet Design Beam Dynamics with Realistic Field Distributions o Outlook 21

Realistic particle distribution after the LINAC Beam size r 0 1cm Average current over one RF-period: I = 150mA Center energy of the Bunch: W = 2MeV Velocity/speed of light : = 0.

22 Realistic particle distribution after the LINAC Beam size r 0 1cm Average current over one RF-period: I = 150mA Center energy of the Bunch: W = 2MeV Velocity/speed of light : = Charge per micro bunch: Q bunch = 0.85nC Number of protons per Bunch: N proton = Path length(1. traj.) : L 4m Electric field on the surface of the bunch: E 0 = 76.4kV/m Acceleration (proton on surface): a = m/s 2 Potential Energy (proton on surface): W pot =763.9 ev Max. velocity due to space charge forces: v max = m/s Significant Space charge forces! => max. Beam size after 4m drift: r 1 5cm

23 Types of Bunch Compression 23

24 Estimation: Additional Energy Spread for Typ1 BC 24

881 MeV DE n Variable driver beam energy : Adjustable Neutron

25 Reasonable Energy Spread for Driver Beam E p = 2 ± 0.2 MeV H 3Li 4Be 0n 1, 646 MeV Threshold: MeV DE n Variable driver beam energy : Adjustable Neutron Spectra + TOF high resolution energy dependent Neutron capture cross sections. 25

26 Combination of both Bunch Compression Concepts 26

27 Estimation: Trajectory Separation due to Energy Variation 27

28 Estimation: Trajectory Separation due to Energy Variation 28

29 Estimation: Trajectory Separation due to Energy Variation 29

30 Mobley Type Bunch Compressor Mobley-Buncher: ( A-Proton beams) Kicker Separation of the micro bunches Bending system (1 Dipole) weak focusing path length differences longitudinal compression 30

31 Mobley Type Bunch Compressor Mobley-Buncher: ( A-Proton beams) Kicker Separation of the micro bunches Bending system (1 Dipole) weak focusing path length differences longitudinal compression Studied Systems: 1 Dipole: variable shape or gradient 2 Dipoles: rectangular magnets 3 Dipoles: 2 homogenous + 1 gradient => insufficient separation => big gradient for every trajectory => ( s/ a) const => longitudinal defocusing 31

Mobley Type Bunch Compressor Mobley-Buncher: ( A-Proton beams) Kicker Separation of the micro bunches Bending system (1 Dipole) weak focusing path length differences

separation of the trajectories momentum exchange in trans.

32 Mobley Type Bunch Compressor Mobley-Buncher: ( A-Proton beams) Kicker Separation of the micro bunches Bending system (1 Dipole) weak focusing path length differences longitudinal compression Improvements for 150mA Proton beams: 2 main dipoles (Gradient) more parameters for beam dynamics 2 auxiliary dipoles (homogeneous) linear separation of the trajectories momentum exchange in trans. plane 2 rebuncher cavities longitudinal Beam dynamics Studied Systems: 1 Dipole: variable shape or gradient 2 Dipoles: rectangular magnets 3 Dipoles: 2 homogenous + 1 gradient => insufficient separation => big gradient for every trajectory => ( s/ a) const => longitudinal defocusing 32

33 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design Drifts, bends, angles <LORASR> LINAC: Rebuncher, Quadrupoles LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution <kicker_traj2> 33

34 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design Drifts, bends, angles <LORASR> LINAC: Rebuncher, Quadrupoles LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution <kicker_traj2> <c4_h>: Geometrical parameters trajectories, angles, distances magnetic fields, gradients shape of the dipoles 34

35 Bunch Compressor: Design / Optimization Cycles Geometry: g9_5x <c4_h> Δ α max = deg B 1 = mT Geometry-design Drifts, bends, angles B 2 = ± 98.4mT <kicker_traj2> <c4_h>: Geometrical parameters trajectories, angles, distances magnetic fields, gradients shape of the dipoles 35

36 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design Drifts, bends, angles <LORASR> LINAC: Rebuncher, Quadrupoles LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution <kicker_traj2> <c4_h>: Geometrical parameters trajectories, angles, distances magnetic fields, gradients shape of the dipoles <kicker_traj2>: kicker properties max. deflection angles max. voltages shape of the plates 36

37 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design Drifts, bends, angles <LORASR> LINAC: Rebuncher, Quadrupoles LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution <kicker_traj2> Kicker-shape, U max, Φ 0 <transp.: kicker> 37

38 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design Drifts, bends, angles <LORASR> LINAC: Rebuncher, Quadrupoles LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution <kicker_traj2> Kicker-shape, U max, Φ 0 <transp.: kicker> <transp.:kicker>: Particle in Cell (PIC) dynamic lattice finite differences Poisson solver U(t) + kicker-shape Particle distribution 38

39 Bunch Compressor: Design / Optimization Cycles Bunch (1) Bunch (5) Bunch (9) <c4_h> Geometry-design Drifts, bends, angles <LORASR> LINAC: Rebuncher, Quadrupoles LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution <kicker_traj2> Kicker-shape, U max, Φ 0 <transp.:kicker>: Particle in Cell (PIC) dynamic lattice finite differences Poisson solver U(t) + kicker-shape <transp.: kicker> Particle distribution L = 20cm U max = 250kV f = 5MHz d min = 18mm α 1 = 29.5deg α 2 = 11.5deg a = 1.2m -1 I = 150mA N particle 10k N grid = 60x60x60 Δ x stepsize = 1mm Δ t calc+plot 7s 39

40 Designs Studies for the 5MHz-Kicker Bunch (1) Bunch (5) Bunch (9) Analytical estimations: - RLC-Resonator - Geometrical parameters: diameter, lengths coil geometry number of windings - Inductance L and Capacitance C - RF-properties: f, Q 0, P <LORASR> cv L = 20cm U max = 250kV f = 5MHz d min = 18mm α 1 = 29.5deg α 2 = 11.5deg a = 1.2m -1 I = 150mA N particle 10k N grid = 60x60x60 Δ x stepsize = 1mm Δ t calc+plot 7s 40

41 Designs Studies for the 5MHz-Kicker Analytical estimations: - RLC-Resonator - Geometrical parameters: diameter, lengths coil geometry number of windings - Inductance L and Capacitance C - RF-properties: f, Q 0, P Scaled Model (analytical): Number of turns 8 Length of the coil 200 mm Radius 150 mm Effective Resistivity Effective Inductance 12.9 H Effective Capacitance 28.8 pf Frequency 9.09 MHz Intrinsic Quality Factor 2986 Shunt impedance 4.4 M Power 250kV 14.8 kw 41

Designs Studies for the 5MHz-Kicker Analytical estimations: - RLC-Resonator - Geometrical parameters: diameter, lengths coil geometry number of windings - Inductance L and Capacitance C -

42 Designs Studies for the 5MHz-Kicker Analytical estimations: - RLC-Resonator - Geometrical parameters: diameter, lengths coil geometry number of windings - Inductance L and Capacitance C - RF-properties: f, Q 0, P Numerical estimation: - 3D Cavity designs with CST Studios Scaled Model (analytical): Number of turns 8 Length of the coil 200 mm Radius 150 mm Effective Resistivity Effective Inductance 12.9 H Effective Capacitance 28.8 pf Frequency 9.09 MHz Intrinsic Quality Factor 2986 Shunt impedance 4.4 M Power 250kV 14.8 kw 42

Measurement at scaled kicker model P f Bidirectional Coupler P r P f AMP

Perturbation Capacitor method => shunt impedance W(t) E(t) 2 => t = 2 t L t L

factor Q 0 = Q L (1+ ) : intrinsic quality factor P on r = P off r <=> =1

43 Measurement at scaled kicker model P f Bidirectional Coupler P r P f AMP Scaled cavity P f P r P t Generator Powermeter P t : U(t) = U 0 exp(-t/t) E(t) Perturbation Capacitor method => shunt impedance W(t) E(t) 2 => t = 2 t L t L : decay time of the stored energy in the cavity Q L = t L : loaded quality factor Q 0 = Q L (1+ ) : intrinsic quality factor P on r = P off r <=> =1 (critical coupling) => Q 0 =2Q L f R per p 2p U P 2 1 L( C DC) 2Q0 C C ( f DC f ) per

44 Measurement at scaled kicker model Analytical MWS Measured (Powermeter) Effective Inductance H Effective Capacitance pf Frequency MHz Intrinsic Quality Factor Shunt impedance M Good agreement for the Inductance Stray Capacitance underestimated => higher frequency 60% of the calculated intrinsic quality factor can be reached Measurements with network analyzer give comparable results Analytic formulas are good enough for first shot estimations big loops ( mm 2 ) are needed for critical coupling => mechanical problems + RF-properties of the loop alternative coupling methods ( capacitive, galvanic ) have to be investigated 44

45 Outlines: o FRANZ: Overview & Status o Bunch Compressor: Design & Optimization Concepts: Geometry 5MHz Kicker Single Bunch Beam Dynamics Merging Scenario Rebuncher Cavities Magnet Design Beam Dynamics with Realistic Field Distributions o Outlook 45

46 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design <kicker_traj2> Drifts, bends, angles <LORASR> Single-Bunch Beam dynamics LINAC: Rebuncher, Quadrupoles Requirement LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution Kicker-shape, U max, Φ 0 <transp.: kicker> Particle distribution Particle distribution <LORASR> 46

47 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design <kicker_traj2> Drifts, bends, angles <LORASR> Single-Bunch Beam dynamics LINAC: Rebuncher, Quadrupoles Requirement LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution Kicker-shape, U max, Φ 0 <transp.: kicker> Particle distribution 16 Parameters: 3 3 Quadrupoles 2 2 Edge Angles 3x Rebunchers Particle distribution <LORASR> 47

48 Alternative Approach for Optimization 16 parameters for every trajectory: 1 calculation several minutes brute force solution unworkable! smart algorithm like Particle Swarm - Optimization (PSO) 48

49 Bunch Compressor: Envelopes(95%) bunch(1) IH + CH dip1 dip2 Multi-aperture Rebuncher: (87.5MHz, 100kV) reb2:(87.5mhz, 120kV) quadrupoles triplets LORASR 49

50 Bunch Compressor: Envelopes(95%) all bunches LORASR 50

51 Single Bunch Projections at the Target Bunch(1) Bunch(5) Rebuncher(2) f/[mhz] U eff /[kv] ε x,rms /[π mm mrad] % ε y,rms /[π mm mrad] % ε z,rms /[π kev ns] % Rebuncher(2) f/[mhz] U eff /[kv] ε x,rms /[π mm rad] % ε y,rms /[π mm rad] % ε z,rms /[π kev ns] % Bunch(9) Single Bunch Beam Dynamics: N bunch = 9 ΔT = ns => ΔT 1ns Rebuncher(2) f/[mhz] U eff /[kv] ε x,rms /[π mm mrad] % A (beam at target) < 3x3cm 2 I (per pulse) 8A ε y,rms /[πmm mrad] % ε z,rms /[π kev ns] % ΔW < ± 5% 51

52 Alternative Method: Particle-Swarm-Optimization o N points in search space (e.g. in Â 16 ) position x = set of parameter in Â 16 velocity v = change of the position in one iteration o Initialize x and v with random values o Iteration: x x v i 1 i i 1 Ü v v c empirical parameters: w = , c 1 =2, c 2 =2 random diagonal matrix: A, B o Collective Memory : Local best result on path in search space Global best result in all iterations o Collective movement of the x s in search space like fish swarm chasing a bait A ( xbest, local xi ) c2 B ( xbest, x i 1 i 1 global i ) 52

53 Optimization of the central trajectory 53

54 Neighboring Trajectories 54

55 Preliminary Results Transverse beam dynamics: Longest trajectory => hardest to optimize Longitudinal beam dynamics: Shortest trajectory => hardest to optimize Result vectors near by each other 55

56 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design <kicker_traj2> Drifts, bends, angles <LORASR> Single-Bunch Beam dynamics LINAC: Rebuncher, Quadrupoles Requirement LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution Kicker-shape, U max, Φ 0 <transp.: kicker> Particle distribution 16 Parameters: 3 3 Quadrupoles 2 2 Edge Angles 3x Rebunchers Particle distribution <LORASR> 56

57 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design <kicker_traj2> Drifts, bends, angles <LORASR> Single-Bunch Beam dynamics LINAC: Rebuncher, Quadrupoles Requirement LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution Particle distribution Kicker-shape, U max, Φ 0 <transp.: kicker> Particle distribution 16 Parameters: 3 3 Quadrupoles 2 2 Edge Angles 3x Rebunchers Particle distribution Multi-Bunch Beam dynamics <LORASR> <transp.: merge> 57

58 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design <kicker_traj2> Drifts, bends, angles <LORASR> Single-Bunch Beam dynamics LINAC: Rebuncher, Quadrupoles Requirement LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution Particle distribution Kicker-shape, U max, Φ 0 <transp.: kicker> Particle distribution Particle distribution Multi-Bunch Beam dynamics <LORASR> <transp.: merge> 58

59 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design <kicker_traj2> Drifts, bends, angles <LORASR> Single-Bunch Beam dynamics LINAC: Rebuncher, Quadrupoles Requirement LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution Particle distribution Kicker-shape, U max, Φ 0 <transp.: kicker> <transp.:merge>: Particle in Cell (PIC) dynamic lattice finite differences Poisson solver Particle distribution <LORASR> Particle distribution Multi-Bunch Beam dynamics <transp.: merge> 59

Bunch Compressor: Design / Optimization Cycles L <c4_h>

9x150mA N particle 90k N grid = 100x100x100 Δ x stepsize = 1mm Δ t

:merge>: Particle in Cell (PIC) dynamic lattice finite differences

: kicker> merge Particle Projections at target distribution

Rebuncher, Quadrupoles Requirement Particle distribution LINAC:

60 Bunch Compressor: Design / Optimization Cycles L <c4_h> Geometry-design <kicker_traj2> Kicker-shape, U max, Φ 0 = 35cm I = 9x150mA N particle 90k N grid = 100x100x100 Δ x stepsize = 1mm Δ t calc+plot 50s <transp.:merge>: Particle in Cell (PIC) dynamic lattice finite differences Poisson solver Drifts, bends, angles <transp.: kicker> merge Particle Projections at target distribution projections <LORASR> Single-Bunch Beam dynamics <LORASR> LINAC: Rebuncher, Quadrupoles Requirement Particle distribution LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution Particle distribution Multi-Bunch Beam dynamics <transp.: merge> 60

61 Merge: Projections at the Target Requirements 61

62 Outlines: o FRANZ: Overview & Status o Bunch Compressor: Design & Optimization Concepts: Geometry 5MHz Kicker Single Bunch Beam Dynamics Merging Scenario Rebuncher Cavities Magnet Design Beam Dynamics with Realistic Field Distributions o Outlook 62

63 Rebuncher Cavities D. U eff = 120kV P 7.7kW R p 1.9MW Broad-Gap-Rebuncher D. l/4-cavity longitudinal Beam Dynamics U eff = kV, P 11kW, R p 1.3MW Multi-Aperture-Rebuncher Energy variation of the final Pulse 63

64 Outlines: o FRANZ: Overview & Status o Bunch Compressor: Design & Optimization Concepts: Geometry 5MHz Kicker Single Bunch Beam Dynamics Merging Scenario Magnet Design Beam Dynamics with Realistic Field Distributions o Outlook 64

65 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design <kicker_traj2> Drifts, bends, angles <LORASR> Single-Bunch Beam dynamics LINAC: Rebuncher, Quadrupoles Requirement LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution Particle distribution Kicker-shape, U max, Φ 0 Magnet-Design <transp.: kicker> Particle distribution 16 Parameters: 3 3 Quadrupoles 2 2 Edge Angles 3x Rebunchers Particle distribution Multi-Bunch Beam dynamics <CST:EMST> <LORASR> <transp.: merge> 65

66 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design <kicker_traj2> Drifts, bends, angles <LORASR> Single-Bunch Beam dynamics LINAC: Rebuncher, Quadrupoles Requirement LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution Particle distribution Kicker-shape, U max, Φ 0 Magnet-Design <transp.: kicker> Particle distribution 16 Parameters: 3 3 Quadrupoles 2 2 Edge Angles 3x Rebunchers Particle distribution Multi-Bunch Beam dynamics <CST:EMST> <LORASR> <transp.: merge> 66

67 Preliminary Magnet Design 849mm Geometry: g9_5x 660mm Δ α max = deg B 1 = mT B 2 = ± 98.4mT 1117mm 660mm 67

Preliminary Magnet Design 849mm 660mm Dipole(1) B 1 /[mt] 515 g/[mm] 60 N I/*A+ 10420 A coil /[mm 2 ] 50 150 A wind /[mm 2 ] 7 7 N 153 I/[A]

68 Preliminary Magnet Design 849mm 660mm Dipole(1) B 1 /[mt] 515 g/[mm] 60 N I/*A A coil /[mm 2 ] A wind /[mm 2 ] 7 7 N 153 I/[A] 68 Geometry: g9_5x 1117mm 660mm Dipole(2) B 2 /[mt] 650 g/[mm] 60 N I/*A A coil /[mm 2 ] A wind /[mm 2 ] 7 7 N 408 I/[A]

69 B / [ T ] Field along the first Trajectory large fringing field Connected fringing field region => Effects of fringing fields on beam dynamics? 69

70 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design <kicker_traj2> Drifts, bends, angles <LORASR> Single-Bunch Beam dynamics LINAC: Rebuncher, Quadrupoles Requirement LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution Particle distribution Kicker-shape, U max, Φ 0 Magnet-Design <transp.: kicker> Particle distribution 16 Parameters: 3 3 Quadrupoles 2 2 Edge Angles 3x Rebunchers Particle distribution Multi-Bunch Beam dynamics <CST:EMST> <LORASR> <transp.: merge> 70

71 Bunch Compressor: Design / Optimization Cycles <c4_h> Geometry-design <kicker_traj2> Drifts, bends, angles <LORASR> Single-Bunch Beam dynamics LINAC: Rebuncher, Quadrupoles Requirement LINAC: (Ratzinger, Heilmann) Geometry, KONUS dynamics, Voltage distribution, quadrupole sizes, particle distribution Particle distribution Kicker-shape, U max, Φ 0 Magnet-Design <transp.: kicker> Particle distribution 16 Parameters: 3 3 Quadrupoles 2 2 Edge Angles 3x Rebunchers Particle distribution Multi-Bunch Beam dynamics <CST:EMST> <LORASR> 3d-field distribution <transp.: dipole> <transp.: merge> 71

72 Beam Dynamics: paraxial Approach vs. realistic Fields The complete effect of the fringing field is applied in one instantaneous kick in the transverse planes: x x x( x 0 k, 0) g y y y ( y Dy 0 k,, K) Dx Comparison with realistic field distribution: 0mA: real field dist. vs. matrix Parameters of the first dipole: Fringing field Integral K = Edge angle entrance = [deg] Edge angle exit = [deg] Magnetic field B 0 = [mt] Gap g = 60.0 [mm] 150mA: real field dist. vs. matrix real field dist.: 0mA vs. 150mA 72

73 Beam Dynamics: paraxial Approach vs. realistic Fields Center motion significantly changed by large fringing field. paraxial approach over estimate the emittance growth in transverse plane. bigger emittance growth in long. plane with realistic fields. field enhancement nearby the edges due to saturation Insufficiently described by first order matrix formalism. 73

74 Improvement in Magnet Design Iron: m r,max =13k, B s = 1.6T Non magnetic vacuum chamber Vacoflux50: m r,max =15k, B s =2.2T Modular pole face Additional return yoke 74

75 Improvement in Magnet Design K=1.034 K=0.098 Without additional return yoke Center motion Transport: paraxial approach vs. realistic field 75

76 Summery Improvement of the Mobley bunch compressor for high current applications: - Several A beam current => 150mA per micro bunch - Additional dipoles => transverse beam dynamics - Rebuncher cavities => longitudinal beam dynamics Single Bunch beam dynamics + merging scenario => fulfills the requirements Smart Algorithm for multi dimensional optimization applied on Bunch Compressor Code Development: space charge dominated transport in realistic external Fields first step for technical realization of the bunch Compressor Kicker: design studies + numerical studies + measurements at scaled model => Results in good agreement with analytical and numerical estimations Dipoles: numerical studies with CST:EMS => Realistic field distributions <=> beam dynamics => Technical realization of the hardwares Cavities: feasible design with CST:MWS => optimization of the power consumptions 76

77 Outlook: Improve magnet design, magnet with gradient Beam dynamics: Front to end simulation Kicker + dipoles + rebunchers + merging scenario 77

78 Thank you for your attention. on behalf of: M. Droba, O. Meusel, D. Noll, U. Ratzinger, C. Wiesner IAP, Goethe University Frankfurt acknowledgment: Franz Käppeler(FZK) R. Reifarth (GSI / U. Frankfurt) M. Heil (GSI) Y. Nie, H. Podlech, A. Schempp, S. Schmidt IAP, Goethe University Frankfurt LINAC-AG AG-Schempp NNP-AG

79 79

80 Bunch Compressor: geometrical parameters g9_5x B 1 = [T] α max = 25.69[deg] < α > = 3.21[deg] al tp dr1 dr2 dr3 b1 b2 bet1 bet2 R2 B2 d_x1 d_x2 d_x3 d_x4 d_x5 d_a d_x1p d_tp psi11 psi12 psi21 psi22 [deg] [m] [m] [m] [m] [m] [m] [deg] [deg] [m] [T] [m] [m] [m] [m] [m] [deg] [m] [m] [deg] [deg] [deg] [deg]

Proton LINAC for the Frankfurt Neutron Source FRANZ

Proton LINAC for the Frankfurt Neutron Source FRANZ - IAEA - International Topical Meeting on Nuclear Research Applications and Utilization of Accelerators Oliver Meusel 4-8 May 2009 Vienna, Austria Motivation