1. Status and Results of the UA9 Crystal Collimation Experiment at CERN-SPS Walter Scandale (CERN, LAL, INFN) for the UA9 collaboration Mini-workshop UA9, LAL Paris - 19 December 2012

2. Outline  Collimation in high-energy high-intensity colliders  Crystal collimation and the UA9 experiment  UA9 layout in 2012  UA9 results in 2012  Toward the installation in LHC  Conclusions 2 Crystal length: 2 mm Bending angle: 176 urad Bent planes: Torsion: ~ 1 urad / mm Amorphous layer: < 1 um Mis-cut angle: ~ 100 urad Silicon strip crystal

3. Mini-workshop UA9, LAL Paris - 19 December 2012 In any accelerator there are particle loss (they produce a intensity lifetime). Functions of collimation  Use collimators for o Radioprotection intercept lost hadrons at well defined locations, (no or minimal losses and activation at other locations!) o SC magnet integrity prevent beam loss-induced quenches of super-conducting magnets. o Physics reach (maximum signal/background) control background from beam halo in experiments o Hardware integrity set passive machine protection (beam lost first at collimators)

4. Mini-workshop UA9, LAL Paris - 19 December 2012 Principle of Beam Collimation Secondary halo p p e  Primary collimators Core Diffusion processes Shower Beam propagation Impact parameter ≤ 1  m Sensitive equipment Primary halo (p) e  Shower p Tertiary halo Secondary collimators... Two or multi stage cleaning...

5. Mini-workshop UA9, LAL Paris - 19 December 2012 Secondary and Tertiary Beam Halo Primary collimators Secondary collimators Protection devices Cold aperture Strategy: Primary collimators are closest. Secondary collimators are next. Absorbers for protection just outside secondary halo before cold aperture. Relies on good knowledge and control of orbit and beta around the ring!

6. Mini-workshop UA9, LAL Paris - 19 December 2012 Collimating with small gaps Machine impedance increases while closing collimators. impedance limit LHC will operate at the impedance limit with collimators closed! LHC beam will be physically quite close to collimator material and collimators are long (up to 1.2 m)! Precision positioning Risk of damage to collimators! Beam electro-magnetic fields interact with the collimator material!

7. Mini-workshop UA9, LAL Paris - 19 December 2012 7  Mechanically bent crystal instead of amorphous primary deflector.  Particles are subjected to a coherent interaction (channeling):  reduced loss rate close to the crystal  reduced probability of diffractive events and ion fragmentation/dissociation BUT  small angular acceptance 2 ×θ c depending on the beam energy  localization of the losses on a single absorber, thanks to large deflection angle Crystal collimation concept Beam Absorber Mask Sensitive equipment Halo & showers 6 σ Normalized aperture Deflected halo Strip crystal MCS ~ 3.6 urad (7 TeV) Amorphous (1 m CFC) Crystal (Channeling) (3 mm Si) optimal ~ 40 urad (7 TeV) ch ~ α bend R.W. Assmann, S. Redaelli, W. Scandale, “Optics studies for a possible crystal- based collimation system for the LHC”, EPAC 2006 Crystal + goniometer ??? Multiple scattering (multi-turn halo) De-channeled particles UA9 MISSION: investigate bent crystals as primary collimators in hadron colliders. Energyθ c [μrad] 120 GeV18.26 270 GeV7.30 450 GeV9.42 3.5 TeV3.38 7 TeV2.39

8. Mini-workshop UA9, LAL Paris - 19 December 2012 8 UA9 layout in 2012 8 Scraper (W, 10 cm) Scraper Beam Collimator Deflected beam Crystal + goniometer Absorber Scraper (W, 10 cm) Collimator Roman Pot (Medipix) Roman Pots (Medipix) BLM ~ 45 m, Δμ = 60 ∘ ~ 67 m, Δμ = 90 ∘ ~ 60 m, Δμ = 90 ∘ ~ 45 m, Δμ = 60 ∘ Scraper BLM Roman Pot (Medipix) Crystal collimation system (in two stages) With instrumentation for loss rate and efficiency measurement High-dispersion area for measurements on off- momentum halo Non-dispersive area for measurements “far” from the collimation system (graphite, 1 m) (W, 60 cm)

9. Mini-workshop UA9, LAL Paris - 19 December 2012 BLM 9 Loss rate reduction in the crystal area Protons Crystal aperture: 9 σ Absorber aperture: 13.5 σ data simulation data simulation  Large loss reduction factor from “amorphous” to “channeling” orientation  5÷20x reduction for protons 3÷7x reduction for Pb ions  Best performance using a 1 mm long crystal  Small discrepancy data/simulation  Mis-cut crystal modeled in simulation Perfect crystal Crystal with mis-cut Pb ions Crystal aperture: 3.5 σ Absorber aperture: 7 σ Beam Collimator Crystal + goniometer Absorber BLM (graphite, 1 m) (W, 60 cm) Roman Pots (Medipix)

10. Mini-workshop UA9, LAL Paris - 19 December 2012 10 Extraction efficiency  Efficiency = N deflected / N crystal  Assumptions:  the number of particles intercepted by a moving object is proportional to the loss rate downstream the object  N deflected is proportional to the losses when intercepting the whole deflected beam  N crystal is proportional to the losses when the collimator is the primary aperture  efficiency for protons: 70÷80%  efficiency for Pb ions: 50÷70% Extracted beam observed with the Medipix Beam Collimator Crystal + goniometer Absorber BLM (graphite, 1 m) (W, 60 cm) Roman Pots (Medipix)

11. Mini-workshop UA9, LAL Paris - 19 December 2012 Deflected beam 11 Reduction of the off-momentum halo population Protons Pb ions Scraper aperture > 30 σ Beam Collimator Crystal + goniometer Absorber Scraper (W, 10 cm) Roman Pots (Medipix) BLM Roman Pot (Medipix)  excellent correlation with the losses observed at high-dispersion and at the crystal  2÷6x reduction for protons (less than in crystal region)  3÷10x reduction for Pb ions (equal to crystal region reduction)

12. Mini-workshop UA9, LAL Paris - 19 December 2012 Clearance between the crystal and the absorber 12 Pb ions Scraper aperture > 30 σ clearance = 7 mm (9.5 σ) clearance = 1.2 mm (1.5 σ)  According to simulation, losses in dispersive area  are due particles back-scattered by the absorber  depend on the difference in aperture between crystal and absorber (= clearance)  To minimize the loss rate in the dispersive area for ions and protons the optimal clearance is ~ 7 mm Beam Collimator Crystal + goniometer Absorber Scraper (W, 10 cm) Roman Pots (Medipix) BLM Roman Pot (Medipix) clearance

13. Mini-workshop UA9, LAL Paris - 19 December 2012 Loss maps 13 UA9 installation Amorphous Channeling UA9 HD area Reduction factor crystal Sextant 6Sextant 1Sextant 2Sextant 3Sextant 4 Loss maps are beam loss monitor rates registered along the accelerator.  Loss map measurement is not trivial in UA9:  the SPS BLM system is optimized for high-intensity operation in pulsed mode  UA9 operates at low intensity and low loss rate.  Loss map measurement in 2011:  intensity increased from 1 bunch (I = 1.15 x 10 11 ) to 48 bunches,  Clear reduction of the losses in the Sextant 6 next to the experiment  Loss map measurement in 2012:  total intensity: 3.3 x 10 13, 4 x 72 bunches with 25 ns spacing  Loss reduction in the entire ring (in the reliable BLMs)

14. Mini-workshop UA9, LAL Paris - 19 December 2012  new scraper upstream the crystal  linear scans of the scraper with crystal in channeling and amorphous position  BLM signal close to detector baseline  BLM rate reduced with Pb-ions Halo profile “far from the crystal” 14 Absorber aperture Pb ions BLM [a.u.] Scraper position [σ] Scraper (W, 10 cm) Beam Collimator Deflected beam Crystal + goniometer Absorber Scraper (W, 10 cm) Roman Pot (Medipix) Roman Pots (Medipix) BLM

15. Mini-workshop UA9, LAL Paris - 19 December 2012 Toward installation in LHC  In September 2011, a letter of intents was presented to the LHCC, asking to extend UA9 to the LHC:  new experiment ( LUA9 ) recommended by the LHCC and accepted by the accelerator directorate  goals: demonstrate the extraction of the beam halo in the LHC measure the possible improvements with respect to standard collimation  the UA9 Collaboration together with the LHC collimation team (leader S. Redaelli) will conduct the test. 15

16. Mini-workshop UA9, LAL Paris - 19 December 2012 Toward LHC: layout 16  Layout of the LUA9 experiment  only one beam (beam 1)  two crystal (horizontal and vertical)  two detector stations (horizontal and vertical) o Mini-Roman pot with a segmented detector o Cherencov detector in vacuum  initially/later: o test at steady/ramping energy o standard collimation system in-place/retracted  Crystal located close to the primary collimators (see arrow):  extracted beam absorbed by a secondary collimator  highest radiation area, tight space allowance H & V crystals

17. Mini-workshop UA9, LAL Paris - 19 December 2012 Toward LHC: R&D for a goniometer  Acceptance for channeling defined by the critical angle θ c = √2U 0 /E = 2.4 μrad (7TeV)  Goniometer accuracy must be smaller than angular acceptance:  Possible solutions  mechanical goniometer à la “SPS (IHEP, Russia)” has resolution < 10 urad  mechanical device developed by industrial partner CINEL for the SPS: static resolution and accuracy meet the LHC specifications test on going to assess accuracy in dynamic regime  piezoelectric device under development in collaboration with industrial partner ATTOCUBE. 17  Total angular range : >10 mrad  “Resolution”: <0.1 μrad  “Accuracy”: < 1 μrad  Maximum tilt inaccuracy: < 1 μrad  Linear resolution: 5 μm  Total linear range: 40 mm

18. Mini-workshop UA9, LAL Paris - 19 December 2012 M1: Angular deflection ( 144 mm stroke) M2: Angular and linear displacement (88 mm stroke) M1 & M2: Linear displacement Mirror holder M1 M2 Autocollimator CINEL goniometer for the SPS M1 test - M2 at restM2 test - M1 at rest Repeatability0.22 μrad0.96 μrad Backlash0.11 μrad0.68 μrad Resolution (10 steps) Average = 0.998 μrad Std Dev = 0.078 μrad Average = 0.978 μrad Std Dev = 0.185 μrad Dynamical overshoot LHC Motor500% Nanotec Motor200%

19. Mini-workshop UA9, LAL Paris - 19 December 2012 Crystal damage 19  Radiation resistance:  IHEP U-70 (Biryukov et al, NIMB 234, 23-30): 70 GeV protons, 50 ms bunch of 10 14 p every 9.6 s, several minutes irradiation, channeling efficiency unchanged  NA48 (Biino et al, CERN-SL-96-30-EA): 450 GeV protons, 2.4 s spill of 5 x 10 12 p every 14.4 s, one year irradiation, channeling efficiency reduced by 30%  LHC: 7 TeV protons, 3 x 10 14 p per fill  test HRMT16-UA9CRY under approval at HiRadMat facility: 440 GeV protons, max 288 bunches, 1.7 x 10 11 protons per bunch intensity comparable with worst accident scenario in LHC (asynchronous beam dump) Simulation with only beam energy and silicon heat capacity): ΔT = 5 K per bunch, T melting after ~ 280 bunches The experiment has been performed and no sign of damage were observed in the irradiated crystal

20. Mini-workshop UA9, LAL Paris - 19 December 2012 Summary of the UA9 findings  Test with extracted beams at CERN North Area (~ 3÷5 weeks per year):  Crystal – beam interactions  Measurement of crystal properties before installation in CERN-SPS  Prototype crystal collimation system installed in CERN-SPS (~ 5 days per year):  2009  First results on the SPS beam collimation with bent crystals(Physics Letters B, vol. 692, no. 2, pp. 78–82).  2010  Comparative results on collimation of the SPS beam of protons and Pb ions with bent crystals (Physics Letters B, vol. 703, no. 5, pp. 547–551).  2011  Direct measurement of a strong reduction of the off-momentum halo in crystal assisted collimation of the SPS beam(Physics Letters B, 714(2-5), 231–236).  2012  Direct observation of the halo population reduction far from the crystal, SPS loss maps, optimized apertures for collimation system elements, … (data taking still on-going)  Working for future installation of a prototype system in LHC  2006  First of a crystal-assisted collimation layout (Assmann, Redaelli, Scandale EPAC2006).  2008  Medipix in a Roman pot  2009  Cherencov quartz detector in the primary vacuum  2011  Letter of Intent (CERN-LHCC-2011-007 / LHCC-I-019 10/06/2011)  2012  First goniometer industrially produced suited for the LHC requirements. 20

21. Mini-workshop UA9, LAL Paris - 19 December 2012 Conclusion  The UA9 experiment is studying the possibility to use crystals as primary obstacle in collimation systems.  Test beam measurements demonstrate the possibility to efficiently deflect particles at high angles using bent crystals.  Using a prototype crystal collimation system in the CERN-SPS: collimation of the beam reliably obtained for proton and lead ion beams losses in the collimation system and in the closest high dispersion area reduced when using a crystal target instead of an amorphous one new measurements to estimate loss reduction in the whole accelerator ring and to optimize the parameters of the system  The team is preparing the installation of a minimal crystal collimation system in the LHC. 21

22. Mini-workshop UA9, LAL Paris - 19 December 2012 22 Publications & Acknowledgments 1. W. Scandale et al., Strong reduction of the off-momentum halo in crystal assisted collimation of the SPS beam. Phys. Lett. B, 714 (2012), 231–236. The EN/STI group at CERN is providing extraordinary support to UA9. The BE/OP-BI-RF groups carefully prepare the SPS to allow for our measurements. We specially thank our funding agencies, reference Committees and Referees.