FLUKA applications: from LHC to hadrontherapy

Transcription

1 FLUKA applications: from LHC to hadrontherapy A. Ferrari, for the FLUKA Collaboration CERN, Geneva, Switzerland

2 CERN: from low energies to LHC LINAC 4 (3-160 MeV).. PS (20 GeV).. SPS (400 GeV) LHC (4 TeV p, 1.6 TeV/n Pb ions, ultimately 7 TeV and 2.7 TeV/n) Energy deposition (quenching, damage) Radiation damage (electronics, insulation) Shielding Secondary beam line design May, 29th 2013 Activation Residual dose rates (maintenance) Waste disposal All problems treated with FLUKA calculations in collaboration between accelerator and RP groups Alfredo Ferrari, MCNEG13 2

3 Examples at LHC, 3.5 TeV p beam Controlled quench test with a wire scanner: mimics beam losses with a thin wire in the beam, measure dose in the downstream region, compare with simulations Search for UFO s: Unidentified Falling Objects causing random beam losses and beam dumping: use simulations to guess position/type comparing patterns at beam loss monitors In stable conditions, measure the effect of collision debris around the beam-beam interaction point: TeV in center of mass!!! May, 29th 2013 Alfredo Ferrari, MCNEG13 3

4 As a cold magnet quenches (on purpose) Controlled beam losses induced with beam wire scanner (30 μm carbon fiber) Beam losses from 3.5 TeV p beam measured by Beam Loss Monitors (BLM) in various positions (30-50 m from wire scanner) May, 29th 2013 Alfredo Ferrari, MCNEG13 4

5 Hunting for Unidentified Falling Objects Dose in BLMs along the beam pipe for various hypothetical UFO locations, compared with data May, 29th 2013 Alfredo Ferrari, MCNEG13 5

6 Collision debris IP BLM pattern along IR5 triplet Q3 May, 29th 2013 BLM dose per collision assuming CMS luminosity measurement and 73.5 mb proton-proton crosssection (from TOTEM) Alfredo Ferrari, MCNEG13 6

7 Radiation to electronics Betatron cleaning insertion (IR7) [cm -2 ] [K. Roedd et al] Single Event Upsets in 2010 May, 29th 2013 Alfredo Ferrari, MCNEG13 7

8 Down at only 25 GeV: PS beam on the AD target For a pulse of protons: Measured : mgy/pulse FLUKA : mgy/pulse DOSE: 50 ± 15 Gy (meas.) vs. 60 Gy (FLUKA) HEH fluence/thermal neutron fluence (R factor): 5 ± 40% (meas.) vs. 5 ± 10% (FLUKA) May, 29th 2013 Alfredo Ferrari, MCNEG13 8

9 CNGS muon monitors TBID 2.7m 43.4m 100m 1095m 18m 5m 67m 5m p + C (interactions) π +, K + (decay in flight) µ + + ν µ Flight path to Gran Sasso : 732 km. Whole neutrino beam line simulated with FLUKA Muon monitors: a check of neutrino production May, 29th 2013 Alfredo Ferrari, MCNEG13 9

10 Muon pit instrumentation 60cm 270cm 11.25cm LHC type Beam Loss Monitors 2 x 37 fixed Stainless steel cylinder monitors (Ionization Al electrodes, 0.5cm Chambers) separation N 2 gas filling 2 x 1 movable monitor May, 29th 2013 Alfredo Ferrari, MCNEG13 10

11 Pion and Kaon production data (NA49) 158 GeV/c K + Histos: FLUKA Symbols: NA49 Phase space of interest for CNGS 158 GeV/c π + Histos: FLUKA Symbols: NA49 π + (left) and K + (right) yield as a function of p T for different X F bins for 158 GeV/c p on C (π) or p (K) May, 29th 2013 Alfredo Ferrari, MCNEG13 11

12 Results at Muon pits: data vs MC First muon pit 2 nd muon pit: MC :open circles (ρ=2.3 g/cm 3 ), open squares (ρ=2.5 g/cm 3 ), Exp. full circles Included in MC: effect of earth magnetic field (in the 1 km long decay tunnel) Experimental uncertainties: detector calibration, density of the rock in between the two pits (67 m) May, 29th 2013 Alfredo Ferrari, MCNEG13 12

13 May, 29th 2013 Alfredo Ferrari, MCNEG13 13

14 Fluka vs hadrontherapy, present: HIT, CNAO, Used for generating p, 12 C dose vs depth databases then used for TP Med Phys Group CNAO Depth-dose-distributions of CNAO From A.Mairani, Varenna2012 in water wo/with RiFi for the 147 energies in the initial phase of the operation May, 29th 2013 Alfredo Ferrari, MCNEG13 14

15 12 C beams on water: May, 29th 2013 Alfredo Ferrari, MCNEG13 15

16 Lateral-depth dose CNAO Mairani et al to be published at different water equivalent depths (WED) for MeV/u Both the experimental data and the MC results are renormalized to the maximum at each depth. May, 29th 2013 Alfredo Ferrari, MCNEG13 16

17 Some recent bibliography: PMB PMB PMB PMB PMB Int.J.Rad.Biol JINST A Monte Carlo-based treatment planning tool for proton therapy, PMB in press A Monte Carlo-based treatment planning tool for ion beam therapy, JRR in press May, 29th 2013 Alfredo Ferrari, MCNEG13 17

18 Characterizing a 12 C SOBP in HIT I Max % difference FLUKA DATA =1.5% At the depths of 3 cm and 7 cm the fragments contribution is 7% and 15% of the total dose. Mairani et al, submitted to PMB May, 29th 2013 Alfredo Ferrari, MCNEG13 18

19 Characterizing a 12 C SOBP in HIT II primary 12 C ion beams simulated. The average CPU time per primary particle history is 10 ms on a GHz Intel Core Duo processor. May, 29th 2013 Alfredo Ferrari, MCNEG13 19 Mairani et al, submitted to PMB

20 *PMB55, 4273 (201 0) α = Biologically Oriented Scoring in FLUKA* For each energy deposition i, FLUKA interpolates from the external database provided by the user the α D,i and β D,i parameters for the specific ion with a certain charge at a certain energy. Then FLUKA sums up properly the mixed radiation effect applying the Kellerer and Rossi theory of dual radiation action: α D, id i β D, i Di Then the average biological parameters can be calculated at the end of the FLUKA run: 2 α D, i D i β D, i D i and D = Di D For example the cell survival can be calculated: S = β = with D exp( α D 29/5/13 MCNEG13, Alfredo Ferrari 20 β D 2 )

21 Example of database for bio scoring The user can provide α D and β D for low LET reference radiation and for the components of the mixed radiation field. The database should be provided by the users based on experimental data or on radiological models. 29/5/13 MCNEG13, Alfredo Ferrari 21

22 270 MeV/u 12 C ions on V79 phantom α = α D, i D i D Is there any interest in using this approach for antiproton beams? And if so how reliable are the radiobiological parameters for low energy fragments, like those typical in the peak region? β = D β D, i D i 2 S = exp( α D β D 2 ) MCNEG13, Alfredo Ferrari 22 29/5/13

23 Development and Application of a Monte Carlo based TP tool for Ion Beam Therapy A. Mairani 1,2, G. Battistoni 3,T.T. Böhlen 4,5 S. Brons 2, M. Dosanjh 4, A. Ferrari 4, Th. Haberer 2, K. Parodi 6, A. Schiavi 7, V. Patera 7 1 CNAO, Pavia, Italy 2 HIT, Heidelberg, Germany 3 INFN Sezione di Milano, Milan, Italy 4 CERN, Geneva, Switzerland 5 Medical Radiation Physics, Karolinska Institutet and Stockholm University, Stockholm, Sweden 6 LMU Munich, Germany 7 La Sapienza Università di Roma, Rome, Italy

24 2 Fields p patient case with RBE= 1.1 Biological Dose distribution for a patient calculation using fixed RBE = 1.1 with pencil beams (2 opposing fields). Prescribed dose 2 Gy (RBE). MC calculation of RBE-weighted dose matrixes (5k histories per pencil) = 10 h (20 CPUs, 10CPUs per field) Optimization time = 1 h (1 CPU) 11 h Application at CNAO with the exact nozzle geometry Re-check a given plan suggesting a better solution

25 Field 1 Field 2 Dose normalized to2 Gy (RBE) DVH for PTV

26 May, 29th 2013 Alfredo Ferrari, MCNEG13 26

27 (p,d), (n,d) reactions: Excitation functions C(p,x) 11 C,left, O(p,x) 15 O, right: now deuteron formation at low energies is treated directly and no longer through coalescence (Data: CSISRS, NNDC, blue Fluka2011.2, red Fluka2013.0) May, 29th 2013 Alfredo Ferrari, MCNEG13 27

28 β + production profile in PMMA (134 MeV p) Green: (FLUKA) proton fluence folded with median cross section exp. Data Red: (FLUKA) proton fluence folded with upper limit of exp. Data Blue: (FLUKA) proton fluence folded with the lower limit of exp. Data Purple: directly computed by FLUKA present models May, 29th 2013 Alfredo Ferrari, MCNEG13 28

29 GANIL: 90 deg photon yields by 95 MeV/n 12 C in PMMA Photon yield Blue: Fluka Red: data Green: dose profile E γ > 2 MeV, within few ns from spill Z (mm) Z (mm) [sketch and exp. data taken from F. Le Foulher et al IEEE TNS 57 (2009), E. Testa et al, NIMB 267 (2009) 993. exp. data have been reevaluated in 2012 with substantial corrections] May, 29th 2013 Alfredo Ferrari, MCNEG13 29

30 Photon yields by 160 MeV p in PMMA Pb Collimator NaI detector Absolute comparison PMMA target Alfredo Ferrari, MCNEG13 Schematic layout (dimensions mm) from J.Smeets et al., IBA Energy spectrum of photons after background subtraction (collimator open collimator closed) for 160 MeV p on PMMA. FLUKA red line, data black line (J.Smeets et al., IBA, ENVISION WP3) May, 29th

31 May, 29th 2013 Alfredo Ferrari, MCNEG13 31

32 TEPC response* as a function of impact parameter FLUKA sim. and measurements of the mean imparted energy ε in the TEPC cavity vs. the impact parameter b for Ne ions at 210MeV/n and Fe ions at 360MeV/n Simulations with a δ-ray threshold of 150 ev and 1 kev are shown. The unrestricted LET times the chord-length in the cavity is also shown: this gives the geometrical response. Ions which pass in the wall close to the cavity surface produce delta-rays with energies large enough to penetrate in the cavity. Source: Taddei et al May, 29th 2013 Alfredo Ferrari, 32 MCNEG13 *PMB (201 1 ) No delta ray equilibrium in the cavity!

33 TEPC response* for ions of = LET *PMB (201 1 ) Frequency distribution of imparted energy ε in a TEPC for Ne ions at 210MeV/n and for Si ions at 780MeV/n. Both exp. Data (Guetersloh 2004) and simulations are normalized to the primary ions producing at least a 3 kev ε in the cavity May, 29th 2013 Alfredo Ferrari, MCNEG13 33

34 May, 29th 2013 Alfredo Ferrari, MCNEG13 34

35 GCR/SEP doses in open space: setup GOLEM voxel phantom with organ segmentation thanks to M.Zankl, GSF Capsule like enclosure with variable Al thickness Uniform and isotropic Galactic Cosmic Rays (GCR s) and Solar Energetic Particles (SEP) open space spectra Scoring (each organ): Organ Dose (GCR: Gy/day, SEP: Gy/event) Organ Dose Equivalent (GCR: Sv/day, SEP: Sv/event), using the ICRP60 Q(LET) relationship Effective Dose (whole body) using the newest ICRP recommendations (ICRP 1-3, 2007) (GCR: Sv/day, SEP: Sv/event) May, 29th 2013 Alfredo Ferrari, MCNEG13 35

36 GCR: open space doses after 1 g/cm 2 Al: Red Bone Marrow (RBM) Organ Dose: 0.378±0.006 mgy/day (Uncollided contr.: 0.206±0.002 mgy/day) Organ Dose Equiv.: 1.26±0.03 msv/day (Uncollided contr.: 0.72±0.02 msv/day) The contributions of the various Z groups include the primary ion contributions and those of all products generated in their interactions May, 29th 2013 Alfredo Ferrari, MCNEG13 36

37 Open space doses (RBM) due to GCR s vs Al thickness The RBM dose equivalent decrease is only due to a lowering in Ion Quality Factors due to fragmentation in the spacecraft wall May, 29th 2013 Alfredo Ferrari, MCNEG13 37

38 Solar Energetic Particle (SEP) events Integrated fluence: up to (nucleon/cm 2 ), E > 1 MeV / n Large variations in spectra Variable composition: mostly protons ( 90%) and α s (~9%), but ions up to Iron are not negligible Variable duration, from hours to days Rise time from minutes to hours Dose equivalent up to ~Sv, highly dependent on organ, shielding, and SEP intensity/spectrum Unpredictable (from Mewaldt, ICRC2005) Nightmare scenarios for (manned) missions beyond Earth low orbits May, 29th 2013 Alfredo Ferrari, MCNEG13 38

39 SEP: the hard event of 20 Jan 2005 o Integrated flux: (nucleon/cm 2 ), E > 1 MeV / n o The hardest spectrum after the famous February 1956 event o Detectable increase in ground level muons above 5 GeV!!! o Very fast rise time ( minutes) Optimal benchmark for a(n) (almost) worst case scenario (from Mewaldt, ICRC2005) May, 29th 2013 Alfredo Ferrari, MCNEG13 39

40 20-Sep-2005: open space doses after 1 g/cm 2 Al, Skin Organ Dose: 1.363±0.004 Gy (Uncollided contr.: 1.250±0.004 Gy) Organ Dose Equivalent: 6.16±0.03 Sv (Uncollided contr.: 5.61±0.02 Sv) Whole body Dose Eq.: 1.83±0.05 Sv SEP 28-Oct-2003: Whole body Dose Eq.: 4.9±0.1 Sv!! May, 29th 2013 Alfredo Ferrari, MCNEG13 40

41 May, 29th 2013 Alfredo Ferrari, MCNEG13 41

42 ITER geometry: divertor cassette [FLUKA ray tracer by D. Pastor] Alfredo Ferrari, MCNEG13 May, 29th

43 Ambient residual dose rate Results expressed in µsv/h. Statistical errors vary from 2% to 30 % (around the CS part). Full operational scenario (20 years) compared with 40 pulses within 1 week of operation. Various cooling times from 1 second to 300 years. 2 vertical cut for the ambient residual dose rate map: -Through the ports -Through the TF coil 1 second of cooling 50 years of cooling Alfredo Ferrari, MCNEG13 May, 29th

44 Ambient residual dose rate: one blanket module ~5 Sv/h Blanket module between 2 equatorial ports. Results expressed in µsv/h. Statistical errors < 10%. Full operational scenario of 20 years. Various cooling times from 1 second to 300 years. 4 meters of distance all around the blanket module. Vertical cut averaged over 50 cm. 10 years ~300 Sv/h ~30 Sv/h 1 hour 11.6 days Alfredo Ferrari, MCNEG13 May, 29th

45 Ambient residual dose rate: one divertor cassette Results expressed in µsv/h. Statistical errors < 10%. Full operational scenario of 20 years. Various cooling times from 1 second to 300 years. 4 meters of distance all around the divertor cassette. Vertical cut averaged over 40 cm. ~0.4 Sv/h 10 years ~200 Sv/h ~3 Sv/h 1 hour 11.6 days May, 29th

46 DPA calculations: blanket module Max. 3 DPA blanket Vacuum vessel 10%<err.<20% Stat. error > 20% Stat. error < 10% May, 29th 2013 Alfredo Ferrari, 46 MCNEG13

47 May, 29th 2013 Alfredo Ferrari, MCNEG13 47

48 WARM DIPOLE DETERIORATION v h s beam 1 TCP.D C B 6L7.B1 beam 1 entering kgy > 500 kgy kgy kgy [J. Trummer et al.] 250kGy 250kGy (0.5mm x 0.5mm ~3 MGy transverse resolution) May, 29th 2013 Alfredo Ferrari, MCNEG13 48

49 WARM QUADRUPOLE DETERIORATION v h s beam 1 TCP.D C B 6L7.B1 beam 1 entering 25.7 kgy 59.6 kgy fallen off (487.3 kgy) x kgy [J. Trummer et al.] beam 1 pointing outwards ~ MGy 60kGy x May, 29th 2013 Alfredo Ferrari, MCNEG13 49

50 0.5 MW ON THE COLLIMATORS [R.W. Assmann et al.] measured: LSS/Q11 = FLUKA: LSS/Q11 = May, 29th 2013 Alfredo Ferrari, MCNEG13 50

51 Examples at LHC, 3.5TeV p beam May, 29th 2013 Alfredo Ferrari, MCNEG13 51

52 UP TOWARDS HIGH LUMINOSITY cm -2 s fb -1 C2 TAS Q1 C1 Q2 Q3 Beam screen with W absorbers at mid-planes 800W in the innermost strands 3.7mm, 10 mm x stainless steel, tungsten no beam screen May, 29th 2013 Alfredo Ferrari, MCNEG13 52

53 TEPC simulations Comparison of FLUKA simulation with data from TEPC irradiations (PMB (2011) May, 29th 2013 Alfredo Ferrari, MCNEG13 53

54 GCR: open space doses after 1 g/cm 2 Al: Whole Body Dose Equivalent vs Effective Dose The effective dose definition is clearly nonsense for relativistic heavy ions; in particular the w R =20 for α s and all heavy ions is largely overestimating their contribution May, 29th 2013 Alfredo Ferrari, MCNEG13 54

55 GCR: Moon shelter doses after 30 g/cm 2 regolith Red Bone Marrow (RBM) Organ Dose: 0.246±0.006 mgy/day (Uncollided contr.: 0.070±0.002 mgy/day) Organ Dose Equiv.: 0.54±0.03 msv/day (Uncollided contr.: 0.10±0.1 msv/day) May, 29th 2013 Alfredo Ferrari, MCNEG13 55

56 GCR: Moon shelter doses (RBM) vs regolith thickness May, 29th 2013 Alfredo Ferrari, MCNEG13 56

57 GCR/SEP doses for lunar shelter: setup Phantom: same as for spacecraft case Shelter like geometry with variable thickness Lunar regolith shield Spectra: same as for spacecraft case Scoring: same as for spacecraft case May, 29th 2013 Alfredo Ferrari, MCNEG13 57

58 20-Jan-2005 SEP: Moon shelter doses (skin) vs regolith thickness At 2 g/cm 2 Organ dose Equivalent: 2.27±0.06 Sv Organ Dose: 0.54±0.01 Gy (Uncollided contr.: 1.95±0.06 Sv) (Uncollided contr.: 0.47±0.01 Gy) May, 29th 2013 Alfredo Ferrari, MCNEG13 58

59 20-Sep-2005: Moon shelter after 30 g/cm 2 regolith, Red Bone Marrow (RBM) Organ Dose: 0.117±0.004 Gy (Uncollided contr.: 0.048±0.002 Gy) Organ Dose Equivalent: 0.27±0.02 Sv (Uncollided contr.: 0.087±0.003 Sv) Whole body Dose Eq.: 0.28±0.02 Sv SEP 28-Oct-2003: Whole body Dose Eq.: 1.7±0.1 Sv!! May, 29th 2013 Alfredo Ferrari, MCNEG13 59

60 12 C beams: details May, 29th 2013 Alfredo Ferrari, MCNEG13 60

61 20 Ne beams: May, 29th 2013 Alfredo Ferrari, MCNEG13 61