The First Direct Observation of Positronium Hyperfine Splitting (Ps-HFS)

Transcription

1 The First Direct Observation of Positronium Hyperfine Splitting (Ps-HFS) Taikan Suehara (ICEPP, U. Tokyo, Presenter) T. Yamazaki, A. Miyazaki (U. Tokyo) with contributions from G. Akimoto, A. Ishida, T. Namba, S. Asai, T. Kobayashi, H. Saito (U. Tokyo) T. Idehara, I. Ogawa, Y. Urushizaki, S. Sabchevski (U. Fukui & BAS) Taikan Suehara et al., MIT, 2011/7/25 page 1

2 Positronium (Ps) positron (e + ) + electron(e - ) Ps is the bound state of e - and e + The lightest Hydrogen-like atom A good target for the precise QED test Free from hadronic interaction Simplest particle-antiparticle system Taikan Suehara et al., MIT, 2011/7/25 page 2

3 Spin states of Ps (o-ps, p-ps) ortho-positronium (o-ps) : 75% e + e - spin triplet o-ps 3γ (, 5γ, ) α 3 process Continuous energy spectrum, long lifetime of 142 ns o-ps para-positronium (p-ps) : 25% spin singlet p-ps 2γ (, 4γ, ) e + e - α 2 process p-ps Two back-to-back 511keV monochromatic γ rays, short lifetime of ns Taikan Suehara et al., MIT, 2011/7/25 page 3

4 Hyperfine splitting: o-ps p-ps Ground state HFS: GHz (λ=1.5mm, mm wave) Could be examined by stimulated transition Much larger than 1 H (1.3 GHz) Good target for precise bound-state QED test Taikan Suehara et al., MIT, 2011/7/25 page 4

5 Inconsistency of Ps-HFS Experimental results give 3.5 σ (15 ppm) smaller HFS value! -> need to investigate Taikan Suehara et al., MIT, 2011/7/25 page 5

6 Theoretical interpretations: new particle? Unknown light particle coupled to electron or photon can shift HFS value from QED calculation (as muon g-2) ex.1) particle X coupled to photon o-ps energy level can be affected by X via o-ps γ* oscillation (p-ps is not affected) ex.2) particle a 0 coupled to electron p-ps (or o-ps) energy level can be affected by a 0 via Ps a 0 oscillation (depending on a 0 spin) Taikan Suehara et al., MIT, 2011/7/25 page 6

7 or experimental mis-consideration? Energy level of Ps Unknown systematic effects may affect the experimental results All past experiments used indirect method Measure mix (Zeeman splitting with static B) instead of HFS and calculate HFS with mix e Controlling B in ppm level is very difficult and it may cause some unknown systematic errors Taikan Suehara et al., MIT, 2011/7/25 page 7

8 Our method: direct observation HFS value is inconsistent between experiment & theory All former experiments employed indirect method Direct measurement! The largest issue: high density radiation in 203 GHz o-ps -> p-ps transition is M1: lifetime is > 10 7 sec o-ps decay: lifetime is 142 nsec (10 9 times shorter) -> very strong radiation is needed for stimulated emission (impossible in 80 s) But recent evolutions of THz technology enable it Taikan Suehara et al., MIT, 2011/7/25 page 8

9 Keys for the direct observation: Accumulation of high-power 203 GHz radiation at the positronium forming area 1. High power sub-thz source: gyrotron 2. Gaussian mode converter 3. Fabry-Perot resonator Optimizing the positron source, detectors and shieldings for good signal selection Taikan Suehara et al., MIT, 2011/7/25 page 9

10 1. High power sub-thz gyrotron 2m Gyrotron FU CW Fukui Output window Collector of electrons 7.4T solenoid Cavity at center Electron gun Gyrotron utilizes cyclotron motion of electrons to resonate a cavity inside the solenoid [ characteristics ] 100 GHz 1 THz High power (used as heaters for nuclear fusion) Continuous / pulse Possibly frequency tunable/sweep 203GHz, 300W long pulse (60ms / 5Hz, duty 30%) gyrotron was developed for this study Taikan Suehara et al., MIT, 2011/7/25 page 10

11 2. Gaussian mode converter Gyrotron outputs TE 03 waveguide mode Need to convert to TEM 00 Gaussian mode to be coupled to optical resonator Long-focus parabolic mirror 1 Long-focus parabolic mirror 2 Gyrotron output (quasi-) TE 03 After converter TEM 00 Main parabolic mirror 550 mm To resonator Efficiency: ~30% Step-cut waveguide Beamsplitter Power monitors(input/reflection) Taikan Suehara et al., MIT, 2011/7/25 page 11

12 3. Fabry-Perot resonator One-dimensional resonator(optics) High power density (optical focusing) Freely changing resonant frequency Gold thinfilm mesh (1µm thick, 200µm width, 360µm period) is depleted on the quartz 99% reflection, ~0.7% transmission With a piezo stage, cavity length is precisely controlled (~100nm) to maintain maximum resonance With 130 mm cavity length, 6-7 kw power accumulation has been confirmed (Finesse: ~600) Taikan Suehara et al., MIT, 2011/7/25 page 12

13 Setup of source/detector Gyrotron power Resonator(1.9 atm N 2 0.1atm iso-c 4 H 10 ) 22 Na positron source(1mbq) β-tag plastic scintillator 100 µm thick Start time tagging 1.5 x 2 in. LaBr 3 crystal scintillators ~4% 511 kev, fast Energy & stop time tagging Lead shield Copper mirror Taikan Suehara et al., MIT, 2011/7/25 page 13

14 Overview Source / holder Chamber / resonator Taikan Suehara et al., MIT, 2011/7/25 page 14

15 Timing spectrum Prompt event (p-ps, annihilation) counts / 0.5ns / sec With power Without power o-ps decay + HFS transition lifetimer 135nsec (fit result) Pileup events time [ns] No difference between power on/off events Taikan Suehara et al., MIT, 2011/7/25 page 15

16 Pileup rejection The most significant bkg.: pileup short gate long gate (power off) Without pileup rejection With pileup rejection short gate long gate Cutting pileup events by comparing QDC value with short/long gate Pileup is reduced by a factor of 3 Taikan Suehara et al., MIT, 2011/7/25 page 16

17 Energy cut & result Livetime: 14 hours (power on) With power Without power Power on power off 7σ deviation is observed between on/off events Preliminary transition rate: 38.7 ± 5.6 mhz (Stat. error only) Taikan Suehara et al., MIT, 2011/7/25 page 17

18 Comparing to MC Power off Power on Good agreement between Data & MC (not perfect) Transition rate is consistent between Data & MC Taikan Suehara et al., MIT, 2011/7/25 page 18

19 Next step: frequency-tunable gyrotron HFS measurement needs freq scan Frequency-tunable gyrotron (gyro- BWO) is now under testing -> First direct HFS measurement O(100 ppm) in 1-2 years With positron beam, O(10 ppm) measurement is possible (3-5 years) Frequency-tunable gyrotron can also be used for exotic light particle searches (axions, paraphotons ) Transition probability Now! HFS freq. Assembled gyro-bwo Taikan Suehara et al., MIT, 2011/7/25 page 19

20 Summary Positronium HFS is inconsistent between QED calculation and experiment - need to be examined We are performing direct transition experiment Indication of HFS transition is seen in latest data - now confirming Frequency-tunable gyrotron can lead to precise measurement of ~100 ppm (ultimately 10 ppm) Taikan Suehara et al., MIT, 2011/7/25 page 20

21 backup Taikan Suehara et al., MIT, 2011/7/25 page 21

22 Gyrotron Our Target 22

23 Gyrotron Electrons emitted from the electron gun are accelerated and move in a circle in the magnetic field and go into the cavity If we tune the magnetic field such that the electron cyclotron frequency Ω = eb/mγ is slightly smaller than the cavity resonant frequency ω = c χ R mn π L, the energy of the cyclotron motion is converted to the energy of the EM wave in the resonant cavity l Resonant cavity SC magnet Electron gun EM wave 23

24 Gyrotron We are developing a frequency-tunable gyrotron (reflective gyro-bwo, Backward -Wave Oscillator). Inside this type of gyrotron, BW reflected at the entrance of the cavity interacts with electrons and resonant condition is changed as 2 2 ω + β z ω ωc = Ω = eb mγ ω : output EM wave frequency ω c : cavity resonant frequency Ω : cycrotron frequency β z : axial velocity of electrons can change output frequency continuously by tuning magnetic field strength. 24

25 Gyrotron Blue : Forward Wave Red : Backward Wave Gyrotron e - Output radiation High power Gyro-BWO e - cavity Output radiation Frequency-tunable Reflective gyro-bwo e - Output radiation Backward Wave is reflected at the entrance of the cavity High power & Frequency-tunable 25

26 Mode Conversion Gyrotron output = TE 03 waveguide mode Fabry-Pérot cavity mode = Gaussian beam (= TEM 00 ) have to convert gyrotron output to Gaussian beam in order to couple the sub-thz radiation with the Fabry-Pérot cavity TE 03 (far-field, calculation) Gaussian beam (calculation) 26

27 Mode Converter Mode conversion is based on geometric optics. Step-cut waveguide and the first parabolic mirror are key components. Step-cut waveguide Top View R R Parabolic Mirror f Step-cut P waveguide Parabolic Mirror light path polarization wave front The first ジャイロトロン parabolic mirror converts TE 03 wave to polarized plain wave because OPR = OP R from the definition of parabola. O P 27

28 Piezoelectric Stage - NANO CONTROL TS102 - Long stroke : 15mm - resolution : 10nm - withstand load : 2kg - size : 60mm x 60mm x 24mm 28

29 Pyroelectric Detector - SPECTRUM DETECTOR SPC-9H, SPH-49 - Lithium Tantalate (LiTaO3) crystal, which generates electricity when heated (pyroelectricity) - diameter : 9mm - Max average power : 500mW - size : 60mm x 60mm x 24mm 29

30 Fabry-Pérot resonant cavity Coupling of the input beam to the cavity can be estimated from decrease of the reflection power The number of round-trips inside the cavity is inversely proportional to the sharpness of the resonance Reflection power 67% decrease Transmission power Approximate cavity gain is Γ = 1.1μm (FWHM) Power accumulated in the Fabry-Pérot cavity is

31 PVC (PolyVinyl Chloride) International Journal of Infrared and Millimeter Waves 28, 363 (2007) - Small refractive index and large absorption coefficient n = 1.65, α = 1~2 cm -1 31

32 Absolute Accumulated Power PVC measurement measures temperature increase [K]. have to calibrate to obtain power [W] Water is a total absorption calorimeter and its heat capacity is well known. P [W] = 4.2 V [cc] T [K] t [sec] D. R. We obtained calibration constant for PVC by comparing PVC measurement [K] with water mesurement [W]

33 Absolute Accumulated Power monitored transmission power of the Fabry-Pérot cavity during DAQ Calibration constant from output of the pyroelectric detector to accumulated power in the cavity was obtained as follows PVC P in [W] Cu mirror hole φ0.6 mm P tr [V] pyroelectric detector Calibration constant : C [W/V] = P in [W] / P tr [V]

34 Gas (N 2 :i-c 4 H 10 = 1.9atm:0.1atm) i-c 4 H 10 has 3 good points and 2 bad points 3 good points - kill slow e + (at least 0.1atm of nitrogen is necessary) - high stopping power - high Ps formation probability 2 bad points - high pickoff probability - absorb 203GHz radiation 34

35 LaBr 3 (Ce) LaBr 3 (Ce) crystal scintillator (diameter 1.5 inch, thickness 2 inch) Good energy resolution (FWHM=4%@511 kev) distinguish 2γ and 3γ short-time constant 16 ns high statistics Good time resolution (FWHM=200 ps@511 kev) La (Z = 57), Br (Z = 35) counts / kev FWHM 511 kev energy [kev] 35