1. Radiation Damage at High Energies: Data and Theoretical Models (View from the Particle-Matter Interaction Side) FermilabAccelerator Physics Center Nikolai Mokhov Meeting of the EuCARD2 Collimation work package (WP11) GSI, Darmstadt, Germany December 4-5, 2014

2. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov Materials Under Irradiation Depending on material, level of energy deposition density and its time structure, one can face a variety of effects in materials under irradiation. Component damage (lifetime): Melting, thermal shocks and quasi-instantaneous damage Insulation property deterioration due to dose buildup Radiation damage to inorganic materials due to atomic displacements and helium production. Operational (performance): Superconducting magnet quench Single-event effects in electronics Detector performance deterioration Radioactivation, prompt dose and impact on environment 2

3. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov3 Fermilab Pbar Target Stack Damage In Tevatron Run I, evidence of external target damage sustained when the rotation mechanism failed for several months. With target rotated properly, no damage to target although the outer Ti sleeve showed signs of swelling. Nickel was used first year of Run II. After several months of operation at 4.5×10 12 ppp with  x =0.22 mm,  y = 0.16 mm, a region of damage about 2.5 mm wide developed on the target, with Ti jacket evaporated in that region and with a 15% drop in pbar yield. In Sep. 2002, switched to Inconel targets with their excellent high temperature tensile strength, although with relatively low thermal conductivity compared to Copper and Nickel. Short pulses with energy deposition density EDD in the range from 200 J/g (W), 600 J/g (Cu), ~1 kJ/g (Ni, Inconel) to ~15 kJ/g: thermal shocks resulting in fast ablation and slower structural changes, or melting.

4. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov4 Run I Target Damage (1993-1994) Re Al Ni Cu Target disks are 0.95cm thick, with two 0.32cm Copper cooling disks between. 1993: pRe Xe Iodine isotopes in AP0 service building revealed flaw in the building ventilation system – fixed. 1994: Damage at the exit of Ni target chord because target rotation had failed for a period of months and only vertical motion was available. Also ejection of Ni bits led to a contamination incident. The entire vault wall was then resealed to ensure that air exiting the target vault could only do so through the filter system.

5. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov5 Run II Target 7 Damage (2007) Inconel-600 disks after 4 months of operation with total 2.65×10 19 pot and RMS beam size  x ~0.18mm,  y ~0.22mm. A “consumable mode” to maximize pbar yield in the middle of Run II The observed was attributed to chemical oxidation of the damaged – by thermal shocks - target surface at the target chord beam exit. Because of tight limits established for vertical target positioning, the copper cooling disks emerged relatively undamaged, though distorted in some places.

6. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov6 Target and Jacket Lifetime Switch to Inconel-600 extended the service life of each target from weeks to months (×6 for the entire stack) with practically no decrease in pbar yield. Radioactive particles from a damaged target were also a problem amplified by radioactive Ti bits from the damaged jacket. In Run II, the Titanium jacket has been replaced by a thin-walled cylinder of Graphite, and then by a 6.35-mm thick Be jacket, both being nearly transparent to primary beam and products of its interactions with the target.

7. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov Tevatron Collimator Damage in 2003 Detailed modeling of dynamics of beam loss (STRUCT), energy deposition (MARS) and time evolution over 1.6 ms of the tungsten collimator ablation, fully explained what happened. 7 J/g Hole in 5-mm W 25-cm groove in SS 980-GeV p-beam

8. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov Hydrodynamics in Solid Materials 8 Pulses with EDD >15 kJ/g: hydrodynamic regime. First done for the 300-  s, 400-MJ, 20-TeV proton beams for the SSC graphite beam dump, steel collimators and tunnel-surrounding Austin Chalk by SSC-LANL Collaboration (D. Wilson, …, N. Mokhov, PAC93, p. 3090). Combining MARS ED calculations at each time step for a fresh material state and MESA/SPHINX hydrodynamics codes. The hole was drilled at the 7 cm/  s penetration rate. Axial density of graphite beam dump in 60  s after the spill start. Later, studies by N. Tahir et al with FLUKA+BIG2 codes for SPS & LHC These days we use MARS+FRONTIER. Tools are in hands

9. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov Interaction of Radiation with Organic Materials 9 A. Idesaki For given insulator and irradiation conditions radiation damage is proportional to energy deposition (dose)

10. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov Energy Deposition Modeling: Highly Accurate with Electronic and NIEL included 10 In majority of real-life complex applications, FLUKA and MARS15 energy deposition results coincide within a few % and agree with data. MARS15

11. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov DPA The dominant mechanism of structural damage of inorganic materials is displacement of atoms from their equilibrium position in a crystalline lattice due to irradiation with formation of interstitial atoms and vacancies in the lattice. Resulting deterioration of material critical properties is characterized as a function of displacements per atom (DPA). DPA is a strong function of projectile type, energy and charge as well as material properties including its temperature. OK at t= 0 followed by, e.g., kinetic Monte Carlo up to 10 4 s. 11

12. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov DPA Model in MARS15: Microscopic Modeling 12 NRT damage function: T d is displacement energy (~40 eV) E d is damage energy (~keV) Energy-dependent displacement efficiency k(T) by Stoller/Smirnov: All products of elastic and inelastic nuclear interactions as well as Coulomb elastic scattering (NIEL) of transported charged particles (hadrons, electrons, muons and heavy ions) from 1 keV to 10 TeV contribute to DPA in this model. For electromagnetic elastic (Coulomb) scattering, Rutherford cross section with Mott corrections and nuclear form factors are used.

13. Medium- and Low-E Neutron DPA Model in MARS15 and Optional Correction at Cryo Temperatures For neutrons from 10 -5 eV to 150 MeV: NJOY99+ENDF-VII database, for 393 nuclides. At T=4-6K, optional correction for experimental defect production efficiency η (Broeders, Konobeev, 2004), where η is a ratio of a number of single interstitial atom vacancy pairs (Frenkel pairs) produced in a material to the number of defects calculated using NRT model 13 Al EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov T = 4-6 K

14. DPA Code Intercomparision EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov14 2013 FLUKA, MARS15 and PHITS intercomparison for Mu2e SC coil hottest spot: 15% agreement M.J. Boschini et al., “ Nuclear and Non- Ionizing Energy-Loss for Coulomb Scattered Particles from Low Energy up to Relativistic Regime in Space Radiation Environment”, arXiv:1011.4822v6 [physics.space-ph] 10 Jan 2011 p + Pb

15. MARS15 vs Jun Modeling I. Jun, “Electron Nonionizing Energy Loss for Device Applications”, IEEE TRANSACTIONS ON NUCLEAR SCIENCE, VOL. 56, NO. 6, DECEMBER 2009 EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov Minimal proton transport cutoff energy in MARS is 1 keV p + Cp + Cup + Si 15

16. Defect Production vs Irradiation Type Electron irradiations create single Frenkel defects (point defects of interstitials and vacancies) Low-energy protons and light ions create similar defects as electron irradiations Heavy ions and neutrons produce cascade displacement damage Fusion neutrons create significant subcascades with very high energy PKAs Damage correlation by different types of irradiation should consider production of defects in single point defects and defect clusters EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 201416Rad. Damage at High Energies - N. Mokhov

17. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov Code Capabilities in DPA Modeling 1.Electron and heavy-ion beams: in most cases, DPA is dominated by electronic energy loss of a primary beam; FLUKA/MARS/PHITS/SRIM are doing quite well here. 2.Neutrons at E < 150 MeV: Point defects; ENDF/NJOY damage x-section libraries & processing – OK. 3.Protons: Medium energies: (1) and (2) – OK. 4.High-energy hadrons and heavy ions: nuclear interaction model dependent; most difficult; mixed (1) and (2) regimes; FLUKA/MARS can be OK even at very high energies. 17

18. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov Hydrogen and Helium Gas Production 18 At accelerators, radiation damage to structural materials is amplified by increased hydrogen and helium gas production for high-energy beams. In SNS-type beam windows, the ratio of He/atom to DPA is about 500 of that in fission reactors. These gases can lead to grain boundary embrittlement and accelerated swelling. In modern codes at intermediate energies, uncertainties on production of hydrogen are ~20%. For helium these could be up to 50%. D. Hilscher et al., J.Nucl.Mat, 296(2001)83C. Broeders, A. Konobeyev, FZKA 7197 (2006)

19. Nick Simos’ Rad.Damage Studies at BLIP with 100- 200 MeV protons 2-D carbonGraphite 3-D carbon Glidcop in both axial and transverse directions sees 40% reduction at ~1 dpa. 3-D CC (~ 0.2 dpa) conductivity reduces by 3.2. 2-D CC (~0.2 dpa) measured under irradiated conditions (to be compared with company data). Graphite (~0.2 dpa) conductivity reduces by 6. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov19

20. BLIP Data and Possible Explanations Observation: In proton irradiation, a threshold exists on carbon composites and graphite at ~0.2 DPA, lower than expected from reactor data (although there were indications of such a level for CTE in reactor data). Possible explanations: Contribution of Coulomb elastic scattering ~Z p 2, becomes important (dominant in some instances) for charged particle irradiation. DPA in BLIP tests was estimated with earlier versions of MCNPX known to underestimate (neglect) contribution of this process. Helium gas production adds, becoming increasingly important at high energies. Graphite, wrt some properties, is indeed less radiation-resistant than expected. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov20

21. FNAL Neutrino Yield and Target Degradation Observation: Neutrino yield has degraded by up to 10% after exposure of the target with 6e20 protons. Explanation: Radiation damage of the target material in the shower maximum region (z ~ 10-15 cm, r < 1-2 mm). Possible mechanisms: 1.Void creation due to DPA (a simple model tried by two groups). 2.Helium bubble production, quite substantial at 120 GeV. Trapped bubbles and helium pressure would reduce target density in the above region. 3.Graphite amorphization:  = 1.8 gcc porous carbon  ~ 1.5 gcc (according to M. Tomut). 4.Transmutation of carbon nuclei. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov21

22. Rad. Damage at High Energies - N. Mokhov 22 EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014 Link of Calculated Quantities to Observable Changes Link of calculated quantities (DPA, dose, fluence etc.) to observable changes in critical properties of materials in theoretical/modeling studies remains to be a dream Promising experiments  Low-energy heavy ions at GSI in very clean conditions (although, surface vs bulk damage?)  Studies at BNL: 200-MeV protons and fast neutrons at BLIP, and 28-MeV protons at TANDEM (BNL)  Kurchatov institute experiments  Neutrons at Kyoto reactor at room and cryo temperatures (remember the lower dose rate)  Counting DPA in Si-detectors  Measuring gas production Promising developments: kinetic Monte-Carlo Meanwhile, rely on phenomenology and correlations

23. Damage Correlation Parameters Irradiation particle, e.g. protons vs. neutrons Energy spectra Flux or dose rate (dpa/s) Fluence or dose (dpa) Irradiation temperature Transmutation (e.g. He, H) Pulsed irradiation vs. continuous irradiation EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 201423Rad. Damage at High Energies - N. Mokhov

24. 24 EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014 Correlations Observed at High Energies For the given insulator, environment and irradiation conditions, radiation damage is proportional to the energy deposition (dose) in the same (microscopic) region. In inorganic materials, DPA is proportional to  dE/dx for heavy ions in thin targets  Energy deposition in thin (in terms of in ) targets  Hadron (mainly neutron) fluence in extended targets Hydrogen/Helium production rate grows with energy up to ~10 GeV and is proportional to hadron fluence.

25. DPA, Helium and Energy Deposition Distributions in 1-m Graphite for 120-GeV Protons EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov Similar in this particular case (high energy, low-Z, small r), not in general 25

26. Rad. Damage at High Energies - N. Mokhov 26 EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014 FLUKA and MARS15 agree within 20% DPA follows neutron fluence profile in SC coils of IP1/5 inner triplet 7×7 TeV pp-collisions

27. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov27 Corrections to the reference NRT model: BCA, MD, BCA-MD (IOTA code at KIT) and various forms of defect production efficiency. Most recently by Nordlund: Kinetic Monte Carlo (up to 10 4 s cf to ns in MD): Individual defects, clusters, impurities, annealing Complete simulations: particle interaction and transport codes coupled to BCA+MD(+KMC) Recent Developments

28. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov M 28

29. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov M 29

30. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov30 Gas Production: p, d, t, 3 He and 4 He Nuclear models implemented in popular computer codes predict gas production cross-section with varying degrees of success depending on the energy of projectiles. The use of cross-sections evaluated using nuclear model calculations and measured data is one of the most reliable and flexible approach for advanced calculation of gas production rate, certainly at intermediate energies. KIT: 278 targets from 7 Li to 209 Bi; incident proton energies: 62, 90, 150, 600, 800, 1200 MeV

31. HB2014, East Lansing, Nov. 10-14, 2014Beam-Materials Interactions - N. Mokhov Quantify Material Performance via FOM 31 With added damage limit factor

32. Desperate Needs for Data Tolerable limits on maximum steady-state temperature of Al-6026 with respect to creep, swelling and fatigue without radiation and for pulsed irradiation and pulsed current; starting from room temperature: focusing horns, beam windows and beam dumps, first of all in neutrino experiments. Irradiation limits on graphite, graphite components, MoGR, CuCD etc. for variety of beams and irradiation conditions: targets, collimators, beam dumps (LHC in the first place) RRR degradation of Al and Cu at T ~ 4 K, including annealing: large superconducting solenoids (Mu2e, COMET etc.) and collider SC magnets. Correlation of data from reactors and proton, heavy ion and electron beams for the same or similar conditions. These are the multi-million dollar questions. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 201432Rad. Damage at High Energies - N. Mokhov

33. Summary (1) 1.Radiation damage by high energy protons includes displacement cascades and transmutation production of helium and hydrogen and other impurities. 2.DPA is the parameter commonly used to correlate displacement damage. However, the extent of radiation damage cannot be fully characterized by a single parameter. 3.Damage correlation between different types of irradiations should consider:  Primary recoil energy spectra  Displacement dose rate, DPA/s  Transmutation production rates, He/DPA and H/DPA  Kinetics of irradiation-induced defect production and accumulation behavior due to pulsed irradiation 4.Further quantification is needed on annealed vs non-annealed defects. 5.Develop an approach to accurate calculation of DPA induced by low-energy neutrons in compounds. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 201433Rad. Damage at High Energies - N. Mokhov

34. Summary (2) 6.Radiation damage by high-energy beams is more intense (for the same fluence) due to Coulomb elastic scattering ~Z p 2 ; codes which include this process, nuclear interactions, and same DPA model parameters agree quite well; radiation damage at high energies is amplified by intense helium gas production. 7.Move from occasional comparisons of calculated radiation- damage related quantities to a comprehensive code intercomparison with “standardized” DPA models and benchmarking at well defined irradiation conditions including temperature, dose rate, H 2 /He gas production, etc. The correlation factors - DPA/D, DPA/  n, N He /DPA – need to be generated, compared and related to reactor data. 8.Further weight experimental possibilities and promising theoretical and simulation developments towards link of DPA to changes in material critical properties. EuCARD2 WP11, GSI, Darmstadt, Dec. 4-5, 2014Rad. Damage at High Energies - N. Mokhov34