Positron-Annihilation Lifetime Spectroscopy for Materials Science Andreas Wagner Nuclear Physics Division Institute of Radiation Physics

Transcription

1 Positron-Annihilation Lifetime Spectroscopy for Materials Science Andreas Wagner Nuclear Physics Division Institute of Radiation Physics

2 Outline Motivation Positron Annihilation & Materials Science Positron sources From reactors to radio-isotopes to LINACs Pulsed beams: annihilation lifetime spectroscopy for thin films, bulk materials, fluids, and gases MePS the Monoenergetic Positron Source GiPS the Gamma-induced Positron Source Some apps porous glasses w/ D. Enke (Univ. Leipzig, D) low-k dielectrics w/ A. Uedono (Univ. Tsukuba, JP) MnSi skyrmion lattices w/ C. Hugenschmidt (TU Munich & FRM) 3D tomography w/ some medical physicsts Courtesy: R. Krause-Rehberg / M. Butterling Page 2

3 Fate of positrons in matter Positrons slow down to thermal energies in 3-10 ps. After diffusing inside the matter positrons are trapped in vacancies or defects. Kinetics results in trapping rates about 10 5 times larger than annihilation rates: defect concentrations down to 10-7 are accessible. e + metals (screening) insulators voids para-ps ortho-ps free trapped self capture self spin-flip 2 g 2 g 2 g 2 g 3 g 2 g 0.2 ns 0.5 ns ns 1.0 ns 142 ns 1.0 ns Page 3

4 Typical types of defects in condensed matter dislocations (cluster) vacancies grain boundaries decorated vacancies - Local electron density changes at defects - Lack of positively charged nucleus causes trapping impurities Page 4

5 Positron trapping model defect-free bulk κ defect defect λ bulk λ defect annihilation radiation dn dt dn N dt bulk defect annih I I ( I1 e I I bulk 1 2 bulk defect defect I t 1 bulk 2 n 1 defect I 2 defect defect 2 defect ) n e defect bulk defect t 2 defect n bulk reduced bulk lifetime defect lifetime - Measurement of decay lifetime is characteristic for defect type - Intensity is sensitive to defect concentration Page 5

6 Sensitivity limits STM/AFM optical microscopy TEM optical microscopy n- scatt. TEM x-ray scattering Positron Annihilation Lifetime Spectroscopy n- scatt STM/AFM x-ray scattering Positron Annihilation Lifetime Spectroscopy Page 6

7 Depth distributions for mono-energetic e + 1 kev 3 kev 5 kev 10 kev Positron depth in Si Positron diffusion can be described as random-walk problem with dominant scattering from longitudinal acoustic phonons (in semiconductors and with T < 300 K). Positron trapping or positron annihilation stops further diffusion. Internal electric fields may significantly change diffusion (Fokker-Planck). surface bulk film substrate Page 7

8 Sources? Isotopes, reactors, accelerators! Production of positrons in weak interactions (mediated by W s) neutron (u d d) ν e e + 27 Al(p,n) 27 Si(β + ν e, 4.2 s) 27 Al W + Sumitomi Heavy Industries Cyclotron 18 MeV protons, 50 µa beam current (u d u) proton 22 Na(β + ν e, 2.6 a) 22 Ne Page 8

9 Sources? Isotopes, reactors, accelerators! Production of positrons through electromagnetic interactions (photons) Use intense source of photons for pair production Capture-neutron gamma-rays from reactor 113 Cd(n,γ) 114 Cd FRMII Munich KUR Kyoto TU Delft Mc Master North Carolina e - e + Bremsstrahlung from electron accelerators (AIST/Tsukuba, HZDR Dresden) or upcoming high power lasers (Changsha/Wuhan, Osaka/Hyogo, ELI-NP Bucuresti) e - γ e - AIST, Tsukuba, Japan ELBE, Dresden Page 9

10 Positrons from accelerators ~ 2.8 GHz / 1 µs / 100 Hz / 10 µa NC-LINAC in bunched mode converter linear storage buncher A buncher B moderator chopper e - e + 1 µs 10 ev 3 ms 5 ns 2 ns 250 ps sample 38 ns / 26 MHz SC-LINAC in CW mode EPOS facility Page 10 converter magnetic transport buncher e - moderator chopper e + 10 ps 2 kev 2 ns 250 ps sample

11 Positrons from accelerators 1.6 ma, 40 MeV (64 kw) CW electron accelerator coherent IR-radiation µm coh. THz radiation 100 µm 3 mm Bremsstrahlung 16 MeV Gamma-induced Positrons electrons 34MeV radiation biology detector tests pulsed, mono-energetic positrons kev SPONSOR: mono-energetic positrons kev from 22 Na Page 11

12 Superconducting Accelerator Structures Superconducting niobium resonators developed by TESLA Technology Collaboration. Used in FLASH, XFEL, mode High-frequency standing-wave mode with 1.3 GHz corresponding to unit cell length of 11.5 cm. E 250 kev I 1mA P 250 W 0, K LHe tank with LN 2 -cold shield and super-insulation E 20 MeV I 1mA P 20 kw 1 Page 12 H. Büttig, et al., Nucl. Instr. Meth. B 704 (2013) 7

13 The Mono-energetic Positron Source MePS Flux: e + /s on sample max 500 V/cm max 2.5 kv/cm 4 ns chopper 78 MHz buncher sample 20 cm Pb m concrete shielding SC-LINAC beam 30 MeV, 0.1 ma MHz repetition rate 615 ns spacing 2 kev magnetic transport system post-accelerator kv 30% HPGe detector for DBS BaF 2 detector for PALS Page 13

14 Converter issues Accelerators can produce intense and pulsed slow positron beams. LINear ACcelerators have some advantage due to their high beam power and time structure. A) normal conducting LINAC nelbe photoneutron E ~ 50 MeV source: LPb loop (450 C) I peak ~ 100 ma beam power can take 50 kw power t bunch ~ 1 µs f rep ~ 100 Hz 500 W B) superconducting LINAC (HZDR) E ~ 35 MeV I average ~ 100 µa beam power f rep ~ 1-10 MHz 3500 W in ~ 1cm 3 sophisticated converter designs and heavy shielding needed converter e - Page 14 moderator e + e + stack of 50 x 100 µm thick W foils Water-cooled electron to bremsstrahlung converter A.R. Junghans, et al., Nuclear Data Sheets 119 (2014) 349

15 The Mono-energetic Positron Source MePS DBS (HPGe) PALS (BaF 2, plastic scint.) Page 15

16 MePS PALS timing resolution Si YSZ Lifetimes of Silicon (218 ps) and Yttria-stabilized Zirkonia can be reproduced, timing resolution is 230 ps with an additional (and stable) lifetime component of 720 ps. Signal to noise ratio: Page 16

17 Experiments at MePS porous glasses D. Enke, G. Dornberg (U Leipzig): Porous glasses feature a tunable pore width and adjustable surface properties for membranes, chemo-sensors, drug delivery, optical coatings, etc. glass beads thin film thin film 20 µm Stimulated phase separation in sodium borosilicate glass into silica and an alkali borate phase generates sponge-like porous structures with unknown porosity. H. Uhlig, G. Adouane, C. Bluhm, S. Zieger, R. Krause-Rehberg, D. Enke, J. Porous Materials 23 (2016) 139 Page 17

18 Experiments at MePS porous glasses Annihilation lifetime experiments reveal depth dependent pore size distributions nm 2.6 nm 1.8 nm 1.3 nm 1.0 nm Pore sizes according to modified Tao-Eldrup model by K. Wada and T. Hyodo, JPhysG 443 (2013) Page 18

19 Experiments at MePS porous glasses Effect of tempering -> unexpected reduction of porosity. 5.0 nm 2.6 nm 1.8 nm 1.3 nm 1.0 nm Page 19

20 MePS low-k SiO 2 Nano-porous organo-silicate glasses Depending on the specific treatment processes nano-porous structures can be stabilized or not. A. Uedono et al., Appl. Surf. Science 368 (2016) 272 Page 20

21 Future work in films In-situ AIDA Gas loading reactor 20 bar K e-beam evap. Fe, Co, Ni, Au, Ag, Cr, Pt, Mo Duoplasmotron N, O, H < 30 kev 2 ma Climate chamber Raman spectrometer XPS analysis unit Motivation Hydrogen loading and deloading process, e.g. Hao et al, J. Phys. Chem. Lett. 1 (2010) 2968 Hydrogen storage capacity: decoration of defects, e.g. J. Cizek et al., Phys. Rev. B 79 (2009) Magnetic properties and open volume defects in Fe 60 Al 40 M.O.Liedke, et al., J. Appl. Phys. 117 (2015) Page 21

22 Future work XPS analysis and PALS chamber UHV system at the 6 MV ion beam for H-depth profiling High-voltage cage on top of MePS Page 22

23 @KEK

24 What about bulk materials, fluids, gases, or biological stuff? converter magnetic transport buncher e - moderator chopper e + 10 ps 2 ns 100 ps sample radiator e - 10 ps γ e + e - 10 ps sample GiPS The Gamma-induced Positron annihilation Spectroscopy Page 24

25 Positron production using electron-bremsstrahlung Positron beams for material research E I σ e e f t 16 MeV 900µA 26 MHz 10 ps F.A. Selim, et al. Nucl. Instr. Meth. A 495 (2002) 154 M. Butterling, et al., Nucl. Instr. Meth. B 269 (2011) 2623 Y. Taira, et al., Rad. Phys. Chem. 95 (2014) 30 Annihilation Lifetime Spectroscopy (Coincidence) Doppler Broadening Age-momentum Correlation photon beam 2 cm diameter 10 8 cm -2 s -1 Nb foil: 10-3 X 0 studies performed so far: - animal tissue - semicondudtors, alloys (ZnO, MnSi) - (neutron-activated) reactor materials - water, glycerol from 10 C to 100 C Page 25

26 Positrons: backgnd for nuclear physics exp ts Kapton at 16 MeV electron energy Hard bremsstrahlung produces a huge amount of positrons via pair production inside the target material. High-energy photons act as a volume source of positrons throughout the entire volume. Seite 26

27 science cases porous glasses D. Enke (U Leipzig) Si nano-channels R. Brusa (U & INFN Trento, I) low-k dielectrics A. Uedono (U & AIST Tsukuba, JP) low-k dielectrics E. Zschech (TU & IKTS Dresden) low-k dielectrics N. Köhler (TU & ENAS Chemnitz) reactor pressure vessel steel F. Bergner (FWI) ZnO luminescence and scintillation F. A. Selim (U Bowling Green, OH) MnSi skyrmion lattices C. Hugenschmidt (TU & FRM Munich) 19 nm MnSi shows spontaneously formed and topologically stable magnetic vortices. Defect identifications and concentrations in Mn 1+x Si published in: M. Reiner, et al., Scientific Reports 6(2016)29109 S. Mühlbauer, C. Pfleiderer, et al., Science 323(2009)915 Page 27 Page 2

28 science cases porous glasses D. Enke, U Leipzig Si nano-channels R. Brusa, U & INFN Trento, I low-k dielectrics A. Uedono, U & AIST Tsukuba, JP low-k dielectrics E. Zschech, TU & Fraunhofer IKTS Dresden low-k dielectrics N. Köhler, TU & Fraunhofer ENAS Dresden reactor pressure vessel steel F. Bergner, HZDR ZnO luminescence and scintillation F. A. Selim, U Bowling Green, OH MnSi skyrmion lattices C. Hugenschmidt, TU & FRM Munich Green luminescence in ZnO results from defects (both Zn and O vacancies) passivated by hydrogen. Very fast scintillation lifetimes obtained! Published in: J. Ji, et al., Scientific Reports 6(2016)31238 Page 28 Page 2

29 Intrinsic radioactive materials: RPV steels ~ 1 µs / 100 Hz 38 ns / 26 MHz conventional LINAC mode pulsed RF, highest energy typically pile-up problems F.A. Selim, D.P. Wells, J.F. Harmon, et al. Nucl. Instr. Meth. A 495 (2002) 154 SC-LINAC in CW mode highest average power high yield and low pile-up High resolution lifetime spectrum with signal to noise ratios of better than 10 5 :1 using gamma-gamma coincidence techniques for background reduction. Lifetime spectra are free from artefacts. Long lifetimes reveal atomic defects caused by neutron-induced damage. Can (and how) defects be removed by thermal annealing? Page 30

30 Physics with GiPS: RPV steel Reactor vessel steel becomes brittle due to neutroninduced defects like open-volume defects seeding cracks. Collaboration with Reactor Safety Division at HZDR. To be published Preferential formation of double vacancies Thermal annealing (290 C) not sufficient to remove defects! Page 31

31 Physics with GiPS: Kapton Annihilation lifetime in Kapton has been under debate for quite some time - > get a measurement without source correction. Kapton applied cuts on Germanium and BaF 2 detector energy signal reduce background from interactions outside the sample consistent single positron lifetime of (381 ± 1) ps two components show larger χ 2 Page 32

32 Physics with GiPS: Fluids Conventional lifetime measurements: dissolve 22 Na and dispose it afterwards Positrons from bremsstrahlung homogeneously distributed, sharp time stamp Target is temperature-stabilized, continuously circulated, degassed, dry-nitrogen flushed. Kapton tube target fluid Positron Physics Ortho-Positronium (o-ps) in a fluid forms a bubble given by its zero-point energy and the surface tension. o-ps 142 ns e + e - We know estimate the change of the o-ps pick-off annihilation lifetime with temperature in a bubble created by the o-ps itself. p-ps R.A. Ferell, Phys. Rev., 108,167, 1957 S.J. Tao, J. Chem. Phys., 56,5499, 1972 M. Eldrup et al., Chem. Phys., 63,51, ps Page 33

33 Physics with GiPS: Fluids 2 U r E stationary Schrödinger eqn. 2mPs Rr Rr R0 j kr Ansatz: spherical Bessel fct. l sin kr j0kr 1 st non-trivial solution kr h E0 zero-point energy 2 2 8mPsr0 4mer0 2 E 4 r fluid e + e - r m r r a surf E e E r A 16m e 40 2 m e surf 2 0 c m c A Page 34

34 Physics with GiPS: Positron(-ium) Chemistry Experiments with water are in variance with a simple bubble-type model. Extension: chemical reactions between radiolysis products of the slowing-down of the positron Ps chemistry. Courtesy: Maik Butterling, S V.Stepanov Radicals are positron scavengers which reduce annihilation lifetimes. Extended bubble model including chemistry [S.V. Stepanov et al., Mat. Sci. Forum 607] and energetics of hydrated e + aq and e - aq (what are the relaxation times of un-/polar media?) Chemistry of radiolysis directly accessible since the probe creates the ionization itself Page 35

35 Towards imaging of defects / porosity Material failures impose a significant threat to the integrity and the safety of technical systems. A thorough understanding of the microscopic origin and the development of defects requires advanced methods. early days and of material the quests analysis of today Page 36

36 Motivation Establish a non-destructive and non-intrusive method which allows for spatially resolved positron-lifetime spectroscopy. Reconstruct PET-like images plus positron annihilation lifetime. Possible Applications (list not complete): Porosimetry Medicine in-beam positron lifetime spectroscopy during hard x-ray tumor therapy Engineering pre-failure diagnostics of micro fractures fuel rod inspection APS Physics & Society Newsletter R. Hargraves, R. Moir Page 37

37 Prerequisites Intense source of positrons with deep penetration (cm) 15 MeV X-rays Accurate time-stamping of positron creation (<10 ps) CW LINAC Position-sensitive positron detectors (mm) Siemens LSO PET Time-resolution for lifetime spectroscopy (~100 ps) in-house (physics) Efficient data acquisition in-house (physics) 3-D image reconstruction in-house (medicine) e - sample e - Gamma-induced Positron Spectroscopy Page 38

38 Towards 2/3-D positron lifetime tomography Two position-sensitive photon detectors with 169 elements each LSO-based commercial 13x13 PET pixel detector 4 mm x 4 mm x 20 mm LSO crystals Siemens Lutetium oxyorthosilicate Lu 2 SiO 5 :Ce courtesy: university hospital Dresden Each crystal array read out using 4 PMT Summed PMT signal -> gamma energy Correlation of individual PMT signals -> position Positron annihilation time given by sum over all 8 PMT involved Page 39

39 3D reconstruction Bremsstrahlung beam γ γ 3D tomography applied for the first time using bulk volume positron production. Target is rotated in 2 deg. steps and the image is reconstructed using a cubical (30 mm) 3 voxel space and back-projection algorithm. PTFE Cu Fe Al Page 44

40 Maximum Likelihood Expectation Maximization Aluminium Copper Steel Iterative method for image reconstruction based on a algorithm developed in PET [L.A. Shepp, Y. Vardi, IEEE-MI 2 (1982) 113]. Solves the inversion problem numerically where one has a system matrix M, an a-priori unknown source distribution s and a measured distribution r. step 1 Mˆ s r all prompt long The system matrix has a size of 13 2 x 13 2 x 180 x 30 3 = 138 x Page 45

41 MLEM Aluminium Copper Steel short gate long gate step 2 all prompt long Page 46 2 nd Iteration

42 MLEM Aluminium Copper Steel short gate long gate step 5 all prompt long Page 47 5 th Iteration

43 MLEM Aluminium Copper Steel short gate long gate step 10 all prompt long Page th Iteration

44 MLEM Aluminium Copper Steel short gate long gate step 20 all prompt long Page th Iteration

45 MLEM Aluminium Copper Steel short gate long gate all prompt prompt / all Page 50

46 MLEM Aluminium Copper Steel Gating on positron lifetimes with 225 ps timing resolution. Now the Al is clearly discriminated against the surrounding Teflon. all prompt prompt / all Page 51

47 Lifetime-sensitive analysis B-field coil Courtesy: Jochen Wosnitza Cut through the record coil which reached 91.4 T peak field. Coil is fed by the world s largest capacitor bank w/ 50 MJ stored energy. Page 52

48 Copper Tomography: B-field coil y Teflon Zylon z x 48 h measurement time, 316 GB, 1.6 G events 324 M filtered coincidences Page 53

49 Lifetime-sensitive analysis: B-field coil Page 54

50 Lifetime-sensitive analysis: B-field coil τ = 1.48 ns Now, we select specific voxels and determine the annihilation lifetimes for spatially separated regions. Since the voxel is identified as an ensemble over all possible lines-of-response between two detector crystals, the lifetime distribution is a convolution as well. Some real physics questions needed Page 55

51 Credits Helmholtz-Zentrum Dresden-Rossendorf W. Anwand, M. Butterling #, T. E. Cowan*, J. Ehrler *, M.O. Liedke, K. Potzger, T.T. Trinh *, A.W. * also at Technische Universität Dresden Martin-Luther-Univ. Halle-Wittenberg R. Krause-Rehberg, A. G. Attalah, M. Elsayed, E. Hirschmann, M. John, M. Jungmann, A. Müller # now at TU Delft and many collaborators S.V. Stepanov, D.S.Zvezhinskiy (ITEP, MEPhI) e + chemistry C. Hugenschmidt, M. Reiner (TU Munich) skyrmion-lattice MnSi F. Selim (U Bowling Green) ZnO luminescence M. Kraatz, E. Zschech (Fraunhofer IKTS) low-k dielectrics A. Uedono (Univ. Tsukuba) low-k dielectrics R. Brusa, L. Ravelli (Univ. Trento) nano-si for anti-matter D. Enke, G. Dornberg (Univ. Leipzig) porous glasses for catalysts Funding provided by BMBF (05K2013) and the Helmholtz Energy Materials Characterization Platform Page 56

52 Positron Studies of Defects 2017 domo arigatou gozaimasu Page 57