Understanding the proton radius puzzle: Nuclear structure corrections in light muonic atoms

Transcription

1 Canada s national laboratory for particle and nuclear physics Laboratoire national canadien pour la recherche en physique nucléaire et en physique des particules Understanding the proton radius puzzle: Nuclear structure corrections in light muonic atoms Click icon to add picture Click to edit the outline text format Second Outline Level Click to edit the outline text format Third Outline Level Second Outline Level Fourth Outline Level Third Outline Level Click to edit the outline text format Second Outline Fourth Level Fifth Outline Outline Level Level Third Outline Sixth Fifth LevelOutline Level Level Fourth Seventh Sixth Outline Outline Level Outline Level Level Seventh Fifth Outline Outline Level Level Eighth Eighth Sixth Outline Outline Outline Level Level Level Seventh Outline Ninth In Accelerating collaboration Outline Science for CanadaLevelTagline, LevelSubtitle, with: if any if any Eighth Outline Level Un accélérateur de la démarche scientifique canadienne Nir Barnea, Javier Hernandez, Chen Ji and Nir Nevo Dinur Ninth Outline LevelName Position TRIUMF Owned and operated as a joint venture by a consortium of Canadian universities via a contribution through the National Research Council Canada Click icon to add picture Click icon to add picture Propriété d un consortium d universités canadiennes, géré en co-entreprise à partir d une contribution administrée par le Conseil national de recherches Canada

2 The Proton Charge Radius e r Pic from Pohl et al., Ann.Rev.Nucl.Part.Sci. 63 (2013) r 2 p = 6 dg dq 2 Q 2 =0 CODATA2010: rp= (51)fm Mohr, et al. Rev. Mod Phys. (2012) 2

3 How Small is the Proton? In 2010, proton charge radius was determined in muonic Hydrogen at PSI from spectroscopy measurements of the Lamb-shift: Pohl et al. (2010): rp= (67) fm Antognini et al. (2013): rp= (39) fm Proton Radius Puzzle! μ r 3

4 Origin of the discrepancy? Experimental results may be wrong: Multiple independent electron-proton experiments agree, and the muonic hydrogen experiment looks more convincing than any of the electron-proton experiments Pohl, Gilman, Miller, Pachucki, Ann.Rev.Nucl.Part.Sci. 63 (2013) Electron scattering experiments are done at finite Q 2, maybe not small enough Dispersion analysis: global fit of n and p give rp= 0.84(1) with Lorenz, Hammer, Meissner, EPJA (2012) Exotic hadronic structures? Birse, McGovern EPJA (2012) vs Miller PLB (2013) New physics beyond standard model? New force carrier, e.g. dark photon, that couples differently with e and µ Yavin, Pospelov, Carlson etc... 4

5 New Experiments to Shed Light on the Puzzle Higher precision electron scattering experiments Q 2 from 10-4 GeV 2 to 10-2 GeV 2 MUSE collaboration (2016) measure e ± p and µ ± p to reduce systematic errors CREMA collaboration currently measuring Lamb shift in light muonic atoms: Deuterium, Helions µ- µ- 5

6 Charge Radius From the Lamb-Shift Extracting the radius from measurements requires theoretical input 2S 2P E LS = QED + pol + m3 r 12 (Z )4 hr 2 i m 4 r 24 (Z )5 hr 3 i (2) in a Z expansion up to 5 th order 6

7 Charge Radius From the Lamb-Shift Extracting the radius from measurements requires theoretical input 2S 2P E LS = QED + pol + m3 r 12 (Z )4 hr 2 i m 4 r 24 (Z )5 hr 3 i (2) QED corrections vacuum polarizations lepton self energy relativistic recoil 6

8 Charge Radius From the Lamb-Shift Extracting the radius from measurements requires theoretical input 2S 2P E LS = QED + pol + m3 r 12 (Z )4 hr 2 i m 4 r 24 (Z )5 hr 3 i (2) Nuclear structure corrections Elastic corrections: Finite size 7

9 Charge Radius From the Lamb-Shift Extracting the radius from measurements requires theoretical input 2S 2P E LS = QED + pol + m3 r 12 (Z )4 hr 2 i m 4 r 24 (Z )5 hr 3 i (2) Nuclear structure corrections Elastic corrections: Finite size Zemach moment 7

10 Charge Radius From the Lamb-Shift Extracting the radius from measurements requires theoretical input 2S 2P E LS = QED + pol + m3 r 12 (Z )4 hr 2 i m 4 r 24 (Z )5 hr 3 i (2) Nuclear structure corrections Elastic corrections: Finite size Zemach moment Inelastic corrections: nuclear polarization pol 7

11 Charge Radius From the Lamb-Shift Extracting the radius from measurements requires theoretical input 2S 2P E LS = QED + pol + m3 r 12 (Z )4 hr 2 i m 4 r 24 (Z )5 hr 3 i (2) Nuclear polarizability corrections r µ- Dipole excitation Stronger Coulomb - reduced energy CM 8

12 Charge Radius From the Lamb-Shift Extracting the radius from measurements requires theoretical input 2S 2P E LS = QED + pol + m3 r 12 (Z )4 hr 2 i m 4 r 24 (Z )5 hr 3 i (2) Nuclear polarizability corrections CM r µ- Dipole excitation Stronger Coulomb - reduced energy The distorted charge distribution follows the orbiting μ like a tide Nuclear response function 8

13 The Muonic Atom System µ- H µ = p2 2m r Z r CM r R a Perturbative potential: correction to the bulk Coulomb Using perturbation theory at second order one obtains the expression for up to order pol 9

14 Contributions to pol Non relativistic terms Take non-relativistic kinetic energy in muon propagator Neglect Coulomb force in the intermediate state p Expand the muon matrix elements in 2mr! R R 0 P ' m3 r(z ) 5 12 r 2mr! apple p R R 0 2 2mr! R R m r! 4 10 R R0 4 (0) (1) (2) R R 0 virtual distance traveled by the proton between the two-photon exchange Uncertainty principle R R 0 1 p 2mN! p 2mr! R R 0 r mr m N =0.17 e.g. for µ- 4 He 10

15 Contributions to pol Non relativistic terms (0) / R R 0 2 dominant term, related to the energy-weighted integral of the dipole response function (0) D1 = 2 m3 r 9 (Z ) 5 Z 1! th d! r 2mr! S D 1 (!) (1) / R R 0 3 contains a part that cancels the Zemach moment elastic contribution (2) / R R 0 4 cf. Pachucki (2011) Friar (2013) leads to energy-weighted integrals of three different response functions S R 2(!),S Q (!),S D1D3 (!) 11

16 Contributions to pol Relativistic terms Take the relativistic kinetic energy in muon propagator Separate in longitudinal and transverse term Related to the dipole response function (0) L(T ) = 2m3 r 9 (Z )5 Coulomb term Z 1! th d! K L(T )! m r S D1 (!) Consider the Coulomb force in the intermediate states Naively it is a (0) C (Z )6 corrections, but actually logarithmically enhanced (0) C (Z )5 log(z ) Related to the dipole response function Friar (1977), Pachucki (2011) 12

17 Contributions to pol Finite Nucleon Size Corrections In point nucleon limit V = ZX i 1 r R i R i CM r µ- Consider finite nucleon size by including charge distributions V = ZX Z i dr 0 n p (R 0 R i ) r R 0 NX Z j dr 0 n n (R 0 R j ) r R 0 Low-q approximation of the nucleon form factors G E p (q) ' 1 G E n (q) ' hr 2 pi 6 q2 hr2 ni 6 q2 13

18 Nuclear Polarizability Corrections The accuracy of the extracted radius depends on the accuracy of the nuclear polarizability 2S 2P E LS = QED + pol + m3 r 12 (Z )4 hr 2 i m 4 r 24 (Z )5 hr 3 i (2) Experimental requirement pol must be known with 5% accuracy To estimate the nuclear polarizability one needs information on the excitations of the nucleus nuclear response function Theoretically calculate it Extract it from data 14

19 Previous Work Simple potential models µ- 12 C (Square-well) Rosenfelder 83 µ-d (Yamaguchi) Lu & Rosenfelder 93 From experimental photo-absorption cross section µ- 4 He Bernabeu & Karlskog 74; Rinker 76; Friar 77 pol = 3.1 mev± 20% µ-d Carlson, Gorchtein, Vanderhagen % Uncertainty Zero-range expansion (pion-less EFT) µ-d Friar 2013 Accuracy roughly estimated ~ 2% State-of-the-art potentials µ-d (AV14) Leidemann & Rosenfelder 95 (AV18) Pachucki 2011 Accuracy < 2% or less Ab-initio calculations of the nuclear polarization with stateof-the-art potentials 15

20 Ab-initio Theory Tools s 1 r 2 s 2... Traditional Nuclear Physics J µ = J µ N + J µ NN +... r 1 s A AV18+UIX,..., J2 m + m r A H i = E i i N H = T + V NN + V 3N +... N High precision two-nucleon potentials: well constraint on NN phase shifts Effective Field Theory N 2 LO, N 3 LO... N N N N two-body currents (or MEC) subnuclear d.o.f. J µ consistent with V r J = i[v, ] Three nucleon forces: less known, constraint on A>2 observables S(!) / h f J µ 0 i 2 Exact Initial state & Final state in the continuum at different energies and for different A 16

21 Traditional Potentials AV18 fitted to 1787 pp & 2514 np scattering data for Elab=350 MeV with 2 =1.04 Urbana, UIX V ijk = V 2 P ijk + V R ijk Illinois IL2 +Vijk 2 S + Vijk 3 R 17

22 Chiral EFT Potentials Separation of scales H( )=T + V NN ( )+V 3N ( )+V 4N ( )+... NN 3N 3N 4N 4N Limited resolution at low energy Effective Field Theory of low-energy QCD Nuclear Forces are built from systematic expansions in Q/ Coupling constants fit to nuclear data V NN >V 3N >V 4N 18

23 Lorentz Integral Transform Method Efros, et al., JPG.: Nucl.Part.Phys. 34 (2007) R459 Reduce the continuum problem to a bound-state problem S(!) R( ) = L(, )= d where f 2 f Ô 0 (Ef E 0 ) J µ R( ) S(!) ( ) 2 + is obtained solving = 2 < 1 (H E 0 + i ) = J Ô µ 0 Due to imaginary part the solution is unique h i Since is finite, has bound state asymptotic behaviour inversion L(, ) S(!) R( ) The exact final state interaction is included in the continuum rigorously! 19

24 Hyperspherical Harmonics (r 1, r 2,...,r A ) = (R CM ) ( 1, 2,..., A 1) Recursive definition of hyper-spherical coordinates 1,..., A 1 2 = 0 = A R CM, A i=1 r 2 i = A 1 i=1 2 i b In nuclear physics we are able to use HH for A up to 6 and 7 Antisymmetrization by Barnea and Novoselsky (1998) 20

25 4 He Photo-absorption cross section 5 4 Arkatov et al. (1979) Shima et al. (2005) Nilsson et al. (2005) Nakayama et al. (2007) Tornow et al. (2012) σ γ [mb] ω [MeV] 21

26 4 He Photo-absorption cross section 5 4 Arkatov et al. (1979) Shima et al. (2005) Nilsson et al. (2005) Nakayama et al. (2007) Tornow et al. (2012) AV18+UIX (2006) NN(N 3 LO)+3N(N 2 LO) (2007) σ γ [mb] ω [MeV] Theory can be more precise than experiment 21

27 Nuclear Polarizability Corrections in µ 4 He + We performed the first ab-initio calculations of the nuclear polarizability corrections in muonic Helium Ji, Nevo Dinur, S.B. and Barnea, PRL 111, (2013) Few-body methods: Lorenz Integral Transform method Hyperspherical Harmonics expansion Hamiltonians: AV18+UIX and EFT NN(N3LO)+3N(N2LO) (CD=1 and CE=-0.029) We used the difference from the two Hamiltonians to estimate the uncertainty in nuclear physics Lanczos algorithm: the nuclear polarization is like an energy-dependent sum rule This can be calculated directly without first obtaining the response function and the convergence is fast if is smooth Nevo Dinur, Ji, S.B. and Barnea, arxiv: g(!) 22

28 Nuclear Polarizability Corrections in µ 4 He + Systematic convergence from (0) (2) to The difference between the two potentials for is 5.5% pol Uncertainty from nuclear physics ± 5.5% p 2!±4% CD=1 and CE=

29 Nuclear Polarizability Corrections in µ 4 He + The work is not yet finished... 24

30 Nuclear Polarizability Corrections in µ 4 He + Numerical accuracy HH expansion % δ pol AV18 + UIX 2N(N 3 LO) + 3N(N 2 LO) % K max 25

31 Nuclear Polarizability Corrections in µ 4 He + Atomic Physics Uncertainty (Z ) 6 effects (beyond second order perturbation theory) Relativistic and Coulomb effects to multipoles other than dipole Higher order nuclear size effects Combined they give an additional 3-4 % 26

32 Nuclear Polarizability Corrections in µ 4 He + Error Budget Nuclear Physics 4% Numerical Accuracy 0.4% Atomic Physics 4% Total 6% Dramatic improvement from pervious work based on experimental data Bernabeu & Karlskog 74; Rinker 76; Friar 77 pol = 3.1 mev± 20% We almost meet the experimental requirement (5% accuracy) 27

33 Possible Refinements To further understand the uncertainty in nuclear physics Use chiral EFT at different orders to track the convergence At a fixed order vary the cutoff to assess the theoretical error Epelbaum, Gloeckle, Meissner, NPA (2005) We will first apply this analysis to μd 28

34 Nuclear Polarizability Corrections in µd Previous Work with AV18 total error very small and mostly coming from atomic physics 28

35 Nuclear Polarizability Corrections in µd with Javier Hernandez Expand on the HO basis N max =2n + ` Can reproduce published observables with AV18 and N3LO Quantity N3LO[a] This Work Experiment: E (9) p hr 2 i Qd 0.285* 0.285* (3) r 1.978** 1.978** (89) * Including MEC and Relativistic Corrections in the amount of fm Pd **Including MEC and relativistic Corrections [a] D.R Entem, et al Physical Review C 68, (R) (2003) 30

36 Nuclear Polarizability Corrections in µd with Javier Hernandez Magnetic correction M = 1 3 m3 r 5 gp 4m p g n 2 Z 1! th d! r! 2m r S M1 (!) MEC are important in magnetic transitions S M1!h f j q 0 i Appearance of two-body currents in Chiral EFT Operator LO NLO N2LO N3LO N4LO j IA-NR OPE IA-RC OPE TPE CT IA-NR IA-RC OPE TPE N m + N N m N 31

37 Nuclear Polarizability Corrections in µd Hernandez, Ji, S.B., Nevo-Dinur, Barnea, in preparation Error Budget Nuclear Physics 1.1% Atomic Physics 0.95% Total 1.5% Preliminary More solid estimate of the total error, that includes the nuclear physics error More accurate than evaluation based on experimental data µ- D Carlson, Gorchtein, Vanderhagen (2014) 7% Uncertainty 33

38 Lamb shifts in muonic atoms Raise interesting questions about lepton symmetry Connect nuclear, atomic and particle physics μ4 He + Conclusions and Outlook We have performed the first ab-initio calculation obtaining Much more accurate than earlier estimates based on experimental data μ D Based on EFT expansion we provide more solid estimates of uncertainty Much more accurate than estimates based on experimental data pol = 2.47meV ± 6% Our calculations are key to the charge radius extractions of the CREMA experiment Future Investigate μ 3 He + Narrow uncertainty in nuclear physics (including MEC, higher order chiral forces, explore correlations of observables) 34

39 Canada s national laboratory for particle and nuclear physics Laboratoire national canadien pour la recherche en physique nucléaire et en physique des particules Thank you! TRIUMF: Alberta British Columbia Calgary Carleton Guelph Manitoba McMaster Montréal Northern British Columbia Queen s Regina Saint Mary s Simon Fraser Toronto Victoria Winnipeg York Merci Owned and operated as a joint venture by a consortium of Canadian universities via a contribution through the National Research Council Canada Propriété d un consortium d universités canadiennes, géré en co-entreprise à partir d une contribution administrée par le Conseil national de recherches Canada