The 2mrad horizontal crossing angle IR layout for the ILC Rob Appleby Daresbury Laboratory MDI workshop at SLAC - 07/01/05 D. Angal-Kalinin (DL), P. Bambade, B. Mouton (Orsay), O. Napoly, J. Payet (Saclay)

Could this be used for the ILC?

(cold ILC bunch-spacing  no multi-bunch kink instability) Rationale (cold ILC bunch-spacing  no multi-bunch kink instability) • only ~15% luminosity loss without crab-crossing (2 mrad) • correction possible without cavities exploiting the natural ’ in the local chromatic correction scheme used • no miniature SC final doublet needed • no strong electrostatic separators needed, and no septum • both beams only in last QD  more freedom in optics • negligible effects on physics • spent beam diagnostics may be easier compared to head-on (see CARE/ELAN Document 2004-20)

Possible doublet parameters for 0.5-1 TeV SC QD (r  35mm) 214-228 T/m warm QF (r ~ 10mm) 140-153 T/m 1-1.9m 3m < 2 mrad l*=4.1m ~ 6 mrad 1.3-2.3m  optical transfer R22 ~ 2.84 from IP to QD exit to beam diagnostics (consistent with global parameter sets as in TESLA TDR)

Tolerable beam power losses in SC QD LHC NbTi IR quads • gradient for 0.5 TeV : 215 T/m, • radius  35mm (effective = 31mm) • higher gradients are studied for LHC upgrades using NbTi(Ta), Nb3Sn Tolerable beam power losses in SC QD local & integral LHC spec: 0.4 mW/g & 5 W/m US-cold design parameters assumed at IP Ne  2 1010 per bunch x=543(489) nm y=5.7(4) nm z=0.3 mm Ebeam = 250(500) GeV x=15(24.2) mm y=0.4 mm

Beamstrahlung clearance at QF Calculated need of  0.5mrad around beam direction at IP in realistic beam conditions & beam pipe with r = 10mm in QF Horizontal cone half-opening angle Vertical cone half-opening angle CARE/ELAN document 2004-21 (A+B) 0.5 TeV doublet  c  1.7mrad 1 TeV doublet  c  1.6mrad

1 TeV doublet for Ne  2 1010 per bunch 0.5 TeV doublet Beam power losses in QD without beam size effect with beam size effect with beam size effect Compton tail Beamstrahlung tail 1 TeV doublet for Ne  2 1010 per bunch 0.5 TeV doublet 1 TeV doublet for Ne  1010 per bunch 0.5 TeV using 1 TeV doublet

Beamstrahlung extraction Simultaneous charged beam and beamstrahlung extraction feasible at 0.5 TeV (2mrad) Harder at 1 TeV - c=1.6mrad seems favoured (benefit from SCQ grad. R+D) incoming, outgoing and beamstrahlung close together at QF  plan common beampipe, with separation later

Clearance at QF for synchrotron radiation emitted in QD & QF x,y = core (1-) photons at QF (estimated from CS paras.) (x - c zQF)2  x2 + y2  y2 = 1  need to collimate to nx,yx,y nx,y=5.9, 49 • can increase c • specially shape QF magnet poles • use smaller beam pipe radius If the photon rate reflected off QF and transmitted back to the VD results in too tight collimation requirements (recent work by T. Murayama  OK) IP z

Luminosity loss without crab crossing L/L0 ~ 0.85 geometric formula  0.88 c[mrad]

Extraction line for disrupted beam Transport spent beam with controlled losses, both under disrupted and undisrupted conditions. Constant geometry up to 1 TeV  use 1 TeV doublet design Diagnostics on disrupted beam Compton polarimeter (need chicane and zero net bend) Energy measurement (energy chicane) WISRD-style spectrometer image beam Choose 10W in QD, requiring c=1.6mrad Study for 1 TeV machine and Ne=2 1010 (hardest case!) Extraction line geometry fixed: R22c=4.544mrad

Initial achromat to cancel dispersion Off-axis beam in QD produces horizontally dispersed beam Initial extraction line achromat designed to cancel this Horizontally focussing quad QFX produces  phase advance QFX very close to incoming beamline (50mm) use current-sheet quadrupole and 8m drift QD->QFX BPT=1.3T (g=26 T/m) aperture radius 5cm l=5.8m (1 TeV) Designed and costed for TESLA TDR (2001-21)

Initial achromat to cancel dispersion II QFX becomes weaker doublet; reduce over focussing Achromat completed by 14m drift and 2.5mrad bend Inherent flexibility - optimise strengths, lengths and apertures r Power loss for aperture r=5cm, scaling r For 1 TeV, lose 3.1kW for =1, 350W for =1.5 and 69kW for =0.5 Ne=1 1010, lose 10W; 500 GeV 80W (=1) Use Tungsten collimator  need to check VXD backsplash QFX

Linear dispersion for the 1 TeV lattice  point-focus and chicane "bend back" achromat

-functions for the 1 TeV lattice  point-focus and chicane achromat "bend back"

Some on-going studies • Increase BeamCAL aperture to 2cm to relax collimation? K. Büsser: rate in VXD increases - origin from inside of BeamCal or QD. Reduce by High Z material on BeamCAL and (early) QD surface? • Further study of post-IP beam transport beyond QFX to limit beam losses to acceptable levels and allow post-IP beam instrumentation. Possible reduction using chromatic correction scheme; inclusion of the detailed B field map in the half-quadrupole QFX • Feasibility of collimator for kW power loss for the case of TeV machine - backgrounds in the detector? • Beam dumps? incoming beam to -dump separation 0.62m, -dump to disrupted beam separation 0.54m