Intra Beam scattering at BESSY II, MLS and BESSY VSR

Intra Beam scattering at BESSY II, MLS and BESSY VSR T. Mertens 1, T. Atkinson 1, J. Feikes 1, P. Goslawski 1, A. Jankowiak 1 J. G. Hwang 1, J. Li 1, D. Malyutin 1, Y. Petenev 1, M. Ries 1, I. Seiler 1 1 Helmholtz-Zentrum Berlin (HZB), Germany IPAC18 02.05.2018

Table of Contents 1. Introduction to Intra Beam Scattering 2. Motivation 3. Simulations 4. Comparing simulations with measurements 5. Conclusions T. Mertens, IPAC18, 02.05.2018 2

What is Intra Beam Scattering? Particle scattering Coulomb scattering within a bunch or beam Two types of collision effects : Single collision with large angle : When momentum is transfered from transverse to longitudinal plane this is amplified with relativistic γ, and if the particle lost we call it Touschek effect Multiple collisions with small angles : This causes beam size growth in all dimensions (like gas diffusion) and is referred to as IBS T. Mertens, IPAC18, 02.05.2018 3

Particle scattering Classical approach (1970 s) using Rutherford scattering (Le Duff,Piwinski,Sacherer,Mohl,...) Quantum Field Theory approach (1980 s) using Möller scattering Bjorken-Mtingwa b e e e Impact parameter b γ γ e e e e e Growth rates 1 τ = N (clog) γ 4 ε x ε y σ δ σ s T. Mertens, IPAC18, 02.05.2018 4

Coulomb Logarithm Coulomb Logarithm clog = log ( bmax b min ), with b max the maximum considered impact parameter and b min the minimum considered impact parameter. Minimum and maximum impact parameter not clearly defined For electrons a tail cut is often considered, with the motivation that the Gaussian tails are not well populated (increasing b min ) BESSY II : the clog changes from 21 without tail cut to 10 when a tail cut is applied MLS : the clog changes from 22 without tail cut to 11 when a tail cut is applied For BESSY II and MLS gives a factor two difference in growth rates T. Mertens, IPAC18, 02.05.2018 5

Motivation Growth rates 1 τ = N(clog) γ 4 ε x ε y σ δ σ s where γ is the relativistic γ, clog is the Coulomblog, N is the number of particles in the bunch, ε x, ε y are the transverse emittances, σ s is the bunch length and σ δ is the energy spread. Low emittance - short bunch machines Low emittance machines : ε i 1 τ Short bunch machines : σ s 1 τ (BESSY VSR 1.2 ps RMS zero-current) T. Mertens, IPAC18, 02.05.2018 6

Berlin light sources BESSY II Third generation light source Circumference : 240 m Energy : 1.7 GeV VSR : Triple RF system (0.5,1.5,1.75 MHz) Beam parameters ε x [nmrad] ε y [nmrad] σ s [mm] τx RAD [ms] τy RAD [ms] τ RAD [ms] σδ 2 BESSY II 7.5 0.057 5 7.8 7.7 3.8 MLS 36 0.18 7 13 11 5 BESSY VSR 7.5 0.057 0.4 7.8 7.7 3.8 T. Mertens, IPAC18, 02.05.2018 7

Berlin light sources Metrology Light Source (MLS) Third generation light source Circumference : 48 m Energy : range 50 630 MeV Ramped Beam parameters ε x [nmrad] ε y [nmrad] σ s [mm] τx RAD [ms] τy RAD [ms] τ RAD [ms] σδ 2 BESSY II 7.5 0.057 5 7.8 7.7 3.8 MLS 36 0.18 7 13 11 5 BESSY VSR 7.5 0.057 0.4 7.8 7.7 3.8 T. Mertens, IPAC18, 02.05.2018 7

MLS - IBS support beam size (um) 280 260 240 220 hor. beam size vert. beam size bunch length vert. excitation (a.u.) 11:57:20 11:57:50 11:58:20 11:58:50 11:59:20 11:59:50 time 28 26 24 22 rms bunch length (ps) T. Mertens, IPAC18, 02.05.2018 8

MLS - IBS support beam size (um) 280 260 240 220 hor. beam size vert. beam size bunch length vert. excitation (a.u.) 11:57:20 11:57:50 11:58:20 11:58:50 11:59:20 11:59:50 time Turning off the vertical beam excitation, reflects immediately in an increase in horizontal and longitudinal bunch size. Pointing towards IBS 28 26 24 22 rms bunch length (ps) T. Mertens, IPAC18, 02.05.2018 8

Motivation The questions are: MLS : Are the observed increase in emittances and bunch length caused by IBS? BESSY II: Can we observe IBS effects? BESSY VSR: Are expected IBS contributions to equilibrium beam sizes within acceptable limits? T. Mertens, IPAC18, 02.05.2018 9

Simulations ODE Evolution Equations dɛ i dt = 1 τɛ RAD (ɛ i ɛ i ) + ɛ i, ɛ i τɛ IBS i = ɛ x, y, σδ 2 (1) i where ɛ i are the radiation damping equilibrium beam sizes with quantum excitation, τɛ RAD i the radiation damping times and τɛ IBS i the IBS lifetimes. Particle Tracking 6D macro particle distribution tracking Turn-by-turn distribution update Physics routines applied sequentially IBS: longitudinal slicing of the bunch longitudinal density used in determining the kick amplitude applied to the particles T. Mertens, IPAC18, 02.05.2018 10

Comparing IBS models εx[m] 10 8 0.9 0.8 0.7 0.6 BESSY II - ε x evolution bjorken-mtingwa no tailcut conte-martini no tailcut ibspiwlattice no tailcut ibspiwsmooth no tailcut modpiwlattice no tailcut nagaitsevibslattice no tailcut 0.5 0.000 0.005 0.010 0.015 0.020 t[s] T. Mertens, IPAC18, 02.05.2018 11

Tail cut effects εx[m] 10 8 0.9 0.8 0.7 0.6 BESSY II - ε x evolution bjorken-mtingwa no tailcut bjorken-mtingwa tailcut conte-martini no tailcut conte-martini tailcut nagaitsevibslattice no tailcut nagaitsevibslattice tailcut 0.5 0.000 0.005 0.010 0.015 0.020 t[s] T. Mertens, IPAC18, 02.05.2018 12

BESSY II Simulation: vertical emittance from coupling (0.0075%) 6.2 Measurements conte-martini ODE 0.0060 0.0055 conte-martini ODE - tailcut conte-martini STE Plotted: equilibrium values for different currents ɛy[m] 10 11 6.0 5.8 5.6 5.4 σs[m] 0.0050 0.0045 0.0040 0.0035 0.0030 2.5 5.0 particles per bunch [1 10 10 ] 2.5 5.0 particles per bunch [1 10 10 ] T. Mertens, IPAC18, 02.05.2018 13

BESSY II Not IBS dominated. 6.2 Measurements conte-martini ODE 0.0060 0.0055 conte-martini ODE - tailcut conte-martini STE Impedance, CSR, Potential Well distortion ɛy[m] 10 11 6.0 5.8 5.6 σs[m] 0.0050 0.0045 0.0040 0.0035 5.4 0.0030 2.5 5.0 particles per bunch [1 10 10 ] 2.5 5.0 particles per bunch [1 10 10 ] T. Mertens, IPAC18, 02.05.2018 13

MLS Simulation: vertical emittance from coupling (0.005%) Plotted: equilibrium values for different currents ɛx[m] 10 8 5.50 5.25 5.00 4.75 4.50 4.25 4.00 0.0 2.5 Measurements conte-martini ODE ɛy[m] 10 10 6 5 4 3 conte-martini ODE - tailcut conte-martini STE σs[m] 10 3 7.2 7.0 6.8 6.6 2 0.0 2.5 0.0 2.5 particles per bunch [1 10 10 ] T. Mertens, IPAC18, 02.05.2018 14

MLS Measurements conte-martini ODE conte-martini ODE - tailcut conte-martini STE IBS dominated assuming ɛ y from coupling (0.5%). ɛx[m] 10 8 5.50 5.25 5.00 4.75 4.50 4.25 ɛy[m] 10 10 6 5 4 3 σs[m] 10 3 7.2 7.0 6.8 4.00 0.0 2.5 6.6 2 0.0 2.5 0.0 2.5 particles per bunch [1 10 10 ] T. Mertens, IPAC18, 02.05.2018 14

BESSY VSR conte-martini ODE - short conte-martini ODE - long ɛx[m] 10 8 1.4 1.2 1.0 ɛy[m] 10 10 1.0 0.8 σs[m] 10 3 3.0 2.5 2.0 1.5 0.8 0.6 1.0 0.6 2.5 5.0 0.5 0.4 2.5 5.0 particles per bunch [1 10 10 ] 2.5 5.0 T. Mertens, IPAC18, 02.05.2018 15

BESSY VSR conte-martini ODE - short conte-martini ODE - long IBS has only small ( 25%) impact assuming ɛ y from coupling (0.75%). ɛx[m] 10 8 current = 1 ma σ t = 1.2 ps RMS 1.4 1.2 1.0 0.8 0.6 2.5 5.0 ɛy[m] 10 10 1.0 0.8 0.6 σs[m] 10 3 3.0 2.5 2.0 1.5 1.0 0.5 0.4 2.5 5.0 particles per bunch [1 10 10 ] 2.5 5.0 T. Mertens, IPAC18, 02.05.2018 15

Conclusions Answering the questions: MLS : Are the observed increase in emittances and bunch length caused by IBS? Data consistent with IBS with some assumpations about coupling and vertical beam size. To be confirmed with further experiments. BESSY II: Can we observe IBS effects? No indication of IBS at BESSY II, dominated by other effects. BESSY VSR: Are expected IBS contributions to equilibrium beam sizes within acceptable limits? Simulations estimate an increase of 25% for the bunch length with the expected currents (1 ma) for the short bunches, assuming there are no other dominating effects that change the beam sizes. No IBS effects are expected for the long bunches. T. Mertens, IPAC18, 02.05.2018 16