1. The ALFA project in ATLAS Antwerpen 25/10/07 Per Grafstrom

2. 2 ATLAS FORWARD DETECTORS

3. 3

4. 4 Purpose of ALFA Additional handle on the luminosity  ALFA = Absolute Luminosity For ATLAS Measurement of  tot and elastic scattering parameters Tag proton for single diffraction

5. 5 Luminosity measurements-why? Cross sections for “Standard “ processes  t-tbar production  W/Z production  ……. Theoretically known to better than 10% ……will improve in the future New physics manifesting in deviation of  x BR relative the Standard Model predictions Important precision measurements  Higgs production  x BR  tan  measurement for MSSM Higgs  …….

6. 6 Relative precision on the measurement of  H  BR for various channels, as function of m H, at  L dt = 300 fb –1. The dominant uncertainty is from Luminosity: 10% (open symbols), 5% (solid symbols). (ATLAS-TDR-15, May 1999) Higgs coupling tan  measurement Examples Systematic error dominated by luminosity (ATLAS Physics TDR )

7. 7 Elastic scattering as a handle on luminosity optical theorem: forward elastic rate + total inelastic rate:  needs large |η| coverage to get a good measurement of the inelastic rate- otherwise rely on MC in unmeasured regions  Use  tot measured by others (TOTEM)  Combine machine luminosity with optical theorem luminosity from Coulomb Scattering ATLAS pursuing all options

8. 8 Absolute vs relative measurement STRATEGY: 1. Measure the luminosity with most precise method at optimal conditions 2. Calibrate luminosity monitor with this measurement, which can then be used at different conditions Relative Methods :  LUCID (dedicated luminosity monitor)  BCM  Min. Bias Scintillators  Tile/LAr Calorimeters

9. 9 Elastic scattering at small angles Measure elastic rate dN/dt down to the Coulomb interference region (à la UA4). |t|~0.00065 GeV 2 or Θ ~ 3.5 microrad. This requires (apart from special beam optics) to place detectors ~1.5 mm from LHC beam axis to operate detectors in the secondary vacuum of a Roman Pot spatial resolution s x = s y well below 100 micron (goal 30 micron ) no significant inactive edge (< 100 m icron)

10. 10 Elastic scattering All very simplified – we need Electromagnetic form factor Proper treatment of the Coloumb-hadron interference phase t- dependence of rho and phase non-exponential behaviour -t dependence of the slope Saturation effects

11. 11  tot vs  s and fit to (lns)   =1.0 )  =2.2  (best fit) The total cross section  Alan Valery Mishka

12. 12 The ρ parameter ρ = Re F(0)/Im F(0) linked to the total cross section via dispersion relations ρ is sensitive to the total cross section beyond the energy at which ρ is measured  predictions of  tot beyond LHC energies is possible Inversely :Are dispersion relations still valid at LHC energies? (Figures from Compete collaboration)

13. 13 The b-parameter or the forward peak The b-parameter for lt l<.1 GeV 2 “Old” language : shrinkage of the forward peak b(s)  2  ’ log s ;  ’ the slope of the Pomeron trajectory ;  ’  0.25 GeV 2 Not simple exponential dependence of local slope Structure of small oscillations?

14. 14 Single Diffraction RP IP 240m RP IP 240m RP ZDC 140m LUCID ZDC 140m LUCID ATLAS 17m elastic scattering single diffraction

15. 15 Forward detectors

16. 16 Trigger conditions For the special run (~100 hrs, L=10 27 cm -2 s -1 ) 1. ALFA trigger  coincidence signal left-right arm (elastic trigger)  each arm must have a coincidence between 2 stations  rate about 30 Hz 2. LUCID trigger  coincidence left-right arm (luminosity monitoring)  single arm signal: one track in one tube 3. ZDC trigger  single arm signal: energy deposit > 1 TeV (neutrons) 4. Single diffraction trigger  ALFA.AND.(LUCID.OR.ZDC)  central ATLAS detector not considered for now (MBTS good candidate) For the special run (~100 hrs, L=10 27 cm -2 s -1 ) 1. ALFA trigger  coincidence signal left-right arm (elastic trigger)  each arm must have a coincidence between 2 stations  rate about 30 Hz 2. LUCID trigger  coincidence left-right arm (luminosity monitoring)  single arm signal: one track in one tube 3. ZDC trigger  single arm signal: energy deposit > 1 TeV (neutrons) 4. Single diffraction trigger  ALFA.AND.(LUCID.OR.ZDC)  central ATLAS detector not considered for now (MBTS good candidate)

17. 17 Event generation and simulation PYTHIA6.4 modified elastic with coulomb- and ρ-term single diffraction PHOJET1.1 elastic & single diffraction beam properties at IP1 size of the beam spot σ x,y beam divergence σ ’ x,y momentum dispersion beam transport MadX tracking IP1  RP high β * optics V6.5 including apertures ALFA simulation track reconstruction t-spectrum ξ-spectrum luminosity determination single diffraction L1 filter LUCID & ZDC pre-selection elastic scattering (Work of Hasko Stenzel-Giessen)

18. 18 Single diffraction: trigger conditions Efficiency [%]PythiaPhojet Preselection ξ<0.297.194.8 ZDC [E>1 TeV]51.538.7 LUCID [1 track]45.157.3 [Central ATLAS E> 100 GeV] 24.938.7 Total preselection7574 RP selection ALFA (Relative to preselection) 60.154.2 Total acceptance44.940.1

19. 19 Hit pattern in ALFA hit pattern for 10 M SD events simulated with PYTHIA + MADX for the beam transport Dispersion

20. 20 acceptance for t and ξ global acceptance: PYTHIA 45 % PHOJET 40.1 % global acceptance: PYTHIA 45 % PHOJET 40.1 %

21. 21 Feedthrough for trigger photodetectors Kapton flat cable motherboard MAPMT + VD + RO cards

22. 22 The fiber tracker

23. 23 ALFA 2007: a full scale detection module 23 MAPMTs 10x2 for fiber detector 3x1 for overlap detector Frame from the 2006 TB Base plate similar to the 2006 version, but with central fixation for fiber plates and 1 free slot for triggers feed-through New design for the fiber plates support 10-2-64 fiber plates: New substrates design 3 overlaps fiber plates: New substrates design Trigger scintillators:

24. 24 Roman Pot Concept

25. 25

26. 26

27. 27

28. 28 FE electronics

29. 29 Test Beam campaigns at DESY and at CERN

30. 30 DESY test beam results

31. 31 The test beam at DESY the validity of the chosen detector concept with MAPMT readout the baseline fibre Kuraray SCSF-78 0.5 mm2 square expected photoelectric yield ~4 low optical cross-talk good spatial resolution high track reconstruction efficiency No or small inactive edge Technology appears fully appropriate for the proposed measurement. Conclusions from DESY test beam

32. 32 Test beam at CERN

33. 33 Test Beam at CERN

34. 34 Time line Mechanics  Prototype tested  Full production launched  Delivery end February 2008 Detector  A number of small prototypes tested  Construction of one full detector started (1/8 of total system)  Production start after validation spring 2008.  Full detector in 2009 Electronics  Prototypes tested  Electronics corresponding to one full detector by end 2007  All electronics by end 2008

35. 35 Back up

36. 36 Simulation of the LHC set-up elastic generator PYTHIA6.4 with coulomb- and ρ-term SD+DD non-elastic background, no DPE beam properties at IP1 size of the beam spot σ x,y beam divergence σ ’ x,y momentum dispersion beam transport MadX tracking IP1  RP high β * optics V6.5 including apertures ALFA simulation track reconstruction t-spectrum luminosity determination later: GEANT4 simulation

37. 37Acceptance Global acceptance = 67% at yd=1.5 mm, including losses in the LHC aperture. Require tracks 2(R)+2(L) RP’s. distance of closest approach to the beam Detectors have to be operated as close as possible to the beam in order to reach the coulomb region! -t=6·10 -4 GeV 2

38. 38 L from a fit to the t-spectrum inputfiterrorcorrelation L8.10 10 26 8.151 10 26 1.77 % σ tot 101.5 mb101.14 mb0.9%-99% B18 Gev -2 17.93 Gev -2 0.3% 57% ρ0.150.1434.3%89% Simulating 10 M events, running 100 hrs fit range 0.00055-0.055 large stat.correlation between L and other parameters

39. 39 Simulation of elastic scattering t reconstruction: hit pattern for 10 M elastic events simulated with PYTHIA + MADX for the beam transport  special optics  parallel-to-point focusing  high β*

40. 40 t- and ξ-resolution: PYTHIA vs PHOJET t- and ξ-resolution: PYTHIA vs PHOJET Good agreement between PYTHIA and PHOJET for the reolutions

41. 41 reconstruction bias reconstruction bias True and reconstructed values are in average slightly shifted  needs to be corrected some differences observed at small t True and reconstructed values are in average slightly shifted  needs to be corrected some differences observed at small t

42. 42 Introduction – physics case single diffraction pp  X+p:  complements the elastic scattering program  measurement of cross section and differential distributions  fundamental measurement, tuning of models, background determination  special detectors ALFA+LUCID+ZDC  high β* optics  same special run as for luminosity calibration single diffraction pp  X+p:  complements the elastic scattering program  measurement of cross section and differential distributions  fundamental measurement, tuning of models, background determination  special detectors ALFA+LUCID+ZDC  high β* optics  same special run as for luminosity calibration

43. 43 resolution for t and ξ main contribution to the resolution t: vertex smearing, beam divergence (small t), det. resolution (large t) ξ: vertex smearing and detector resolution main contribution to the resolution t: vertex smearing, beam divergence (small t), det. resolution (large t) ξ: vertex smearing and detector resolution

44. 44 Systematic uncertainties generator difference, model dependence  acceptance, detector corrections ± 5-10% beam conditions, optical functions, alignment  ± 2% (based on results for elastic scattering) background (being estimated)  double diffraction  minimum bias  beam halo DD ≈ 2 %, MB ≈ 0.5 %, beam halo + DD/MB 1-2% luminosity  ± 3%, very best possible luminosity determination, at calibration point! statistical uncertainty small, expect 1.6-2.3 M accepted events generator difference, model dependence  acceptance, detector corrections ± 5-10% beam conditions, optical functions, alignment  ± 2% (based on results for elastic scattering) background (being estimated)  double diffraction  minimum bias  beam halo DD ≈ 2 %, MB ≈ 0.5 %, beam halo + DD/MB 1-2% luminosity  ± 3%, very best possible luminosity determination, at calibration point! statistical uncertainty small, expect 1.6-2.3 M accepted events

45. 45 Conclusion & outlook A measurement of single diffraction with ATLAS appears to be possible, however it won’t be a precision measurement in contrast to elastic scattering.  Combination ALFA, LUCID and ZDC  Special running conditions  measurement of cross section and t-, ξ-distribution  not a precision measurement, 10% systematic uncertainty achievable?  goal: improve model predictions and background estimates for central diffraction This first pilot study must be pursued and confirmed by full simulation and systematic studies involving the LUCID and ZDC communities. The option of including the MBTS for tagging the diffractive system should be investigated. A measurement of single diffraction with ATLAS appears to be possible, however it won’t be a precision measurement in contrast to elastic scattering.  Combination ALFA, LUCID and ZDC  Special running conditions  measurement of cross section and t-, ξ-distribution  not a precision measurement, 10% systematic uncertainty achievable?  goal: improve model predictions and background estimates for central diffraction This first pilot study must be pursued and confirmed by full simulation and systematic studies involving the LUCID and ZDC communities. The option of including the MBTS for tagging the diffractive system should be investigated.

46. 46 Systematic errors Background subtraction ~ 1 %

47. 47

48. 48 Luminosity transfer 10 27 -10 34 cm -2 sec -1 Bunch to bunch resolution  we can consider luminosity / bunch  ~ 2 x10 -4 interactions per bunch to 20 interactions/bunch  Required dynamic range of the detector ~ 20 Required background  < 2 x10 -4 interactions per bunch  main background from beam-gas interactions  Dynamic vacuum difficult to estimate but at low luminosity we will be close to the static vacuum.  Assume static vacuum  beam gas ~ 10 -7 interactions /bunch/m  We are in the process to perform MC calculation to see how much of this will affect LUCID

49. 49

50. 50t-resolution The t-resolution is dominated by the divergence of the incoming beams. σ’=0.23 µrad ideal case real world

51. 51