Award DE-SC0012485 MUSEing on the Proton Radius Puzzle W.J. Briscoe and E.J. Downie on behalf of the MUSE Collaboration Awards PHY-1309130,1314148,1614850,1714833
2 Outline u Why are we puzzled: è What is a radius? How do we measure it? è Electron scattering measurements è The source of all the trouble: muonic spectroscopy measurements u Are we still puzzled? è Possible explanations è What are we doing now? u The MUon proton Scattering Experiment (MUSE) u Conclusions
3 Proton Radius Problem u The Proton Radius Puzzle (PRP) has attracted a lot of interest! u Not only interesting in itself: è Tests our theoretical understanding of proton è Radius of proton - dominant uncertainty in many QED processes u What exactly is the puzzle?
The Proton Radius u What is a radius? How do we measure it? u Classical physics: u Non-relativistic quantum mechanics: u Relativistic quantum mechanics: Electron Scattering Atomic Energy Levels Fit form factor trend with q 2, to data, find slope as q 2 0 NRQM: finite size of proton perturbs energies of s states - rp <<<< ratomic, so effect proportional to ψ 2 a(r=0). 4
The Proton Radius as a Function of Time From Pohl, Gilman, Miller, Pachucki review, arxiv:1301.0905, AnnRevNPS, modified 5
6 Electron Scattering Measurements u Classical Rosenbluth separation u Measure the reduced cross section at several values of ε (angle/beam energy combination) while keeping Q 2 fixed. u Linear fit to get intercept and slope.
7 Electron Scattering Measurements 1950s u Fit to RMS radius Stanford 1956 u R.W. McAllister and R. Hofstadter, Phys. Rev. 102, 851 (1956)
8 Electron Scattering Measurements u Bernauer et al. PRL 105 (2010), 242001: world's largest data set u Fit functional forms to data rather than Rosenbluth separation u Zhan et al. PLB 705 (2011) 59-64: Polarisation measurements to get G E /G M, valuable over a large Q 2 range u Fit(Jlab + world Bernauer) gives radius compatible with Bernauer
9 Time Evolution of the Radius from ep Data CODATA Zhan et al. (JLab) Bernauer et al. (Mainz) Older ep Data
Components of the Hydrogen Energy Levels n=3 n=2 n=1 2P3/2 2S1/2, 2P1/2-43.5 GHz 1S1/2 2S1/2 2P1/2 8.2 Ghz F=1 F=0 1.4 GHz F=1 0.15MHz 0.014% of the Lamb Shift! 1.2 MHz Bohr Dirac Lamb Darwin Term Spin-Orbit Relativity QED F=0 HFS Proton Size 10
11 Time Evolution of the Radius from Hydrogen Lamb Shift CODATA H-Lamb Data
12 Time Evolution of the Radius from Hydrogen Lamb Shift and ep CODATA Zhan et al. (JLab) Bernauer et al. (Mainz) Older ep Data H-Lamb Data
Why Measure with µh? S-Orbital P-Orbital u While lepton is inside proton, attractive potential is lower u Average potential reduced the longer lepton spends inside proton u Strongly affects S orbitals, much less so P, so SP transitions change u Probability for lepton to be inside proton = volume of proton / volume of atom: u m µ =~205m e is µh is ~205 3 ~ 8 million times more sensitive to r P Orbitals: http://chemistry.umeche.maine.edu/chy251/quantum.html 13
Pictures: R. Pohl 14 Mechanics of Measuring with µh u Simple, but technically challenging! u Form µh*(n~14) by firing muon beam on 1mbar H 2 target u 99% decay to 1S, giving out fast γ pulse u 1% decay to longer-lived 2S state u Excited to 2P state by tuned laser & decay with release of delayed γ u Vary laser frequency to find transition peak 2P to 2S ΔE r p
Mechanics of Measuring with µh Pictures: R. Pohl 15
The Proton Radius from Excitation Spectrum u Take ratio of delayed to prompt as a function of laser frequency: Randolf Pohl et al., Nature 466, 213 (2010): 0.84184 ± 0.00067 fm 5σ off 2006 CODATA 16
17 Time Evolution of the Lamb Shift Measurements & ep Data Pohl et al Zhan et al. (JLab) Bernauer et al. (Mainz) Older ep Data H-Lamb Data CODATA.
18 Curiouser & Curiouser... u Aldo Antognini et al. Science 339, 417 (2013) u Further analysis of data taken in Pohl measurement u Magnetic radius agrees with e - scattering data u Electric radius in agreement with Pohl 0.84087 ± 0.00039 fm u 7.9σ from 2010 CODATA u Analysis gives:
We are still puzzled! 19
Why do the Muon and Electron give Different Proton Radii? u Assuming the experimental results are not bad, what are viable theoretical explanations of the Radius Puzzle? u Novel Beyond Standard Model Physics: Pospelov, Yavin, Carlson,...: the electron is measuring an EM radius, the muon measures an (EM+BSM) radius u Novel Hadronic Physics: G. Miller: currently unconstrained correction in proton polarizibility affects µ, but not e (effect ml 4 ) u Basically everything else suggested has been ruled out - missing atomic physics, structures in form factors, anomalous 3rd Zemach radius,... u See Trento Workshops on PRP for more details: http://www.mpq.mpg.de/~rnp/wiki/pmwiki.php/main/workshoptrento (2012) http://www.ectstar.eu/node/1659 (2016) 2 20
How do we Resolve the Radius Puzzle? u New data needed to test that the e and µ are really different, and the implications of novel BSM and hadronic physics è BSM: scattering modified for Q 2 up to m 2 BSM (typically expected to be MeV to 10s of MeV), enhanced parity violation è Hadronic: enhanced 2γ exchange effects u Experiments include: è Light muonic atoms for radius comparison in heavier systems è Redoing atomic hydrogen è Redoing electron scattering at lower Q 2 è Muon scattering! 2 21
22 How do we Resolve the Radius Puzzle? u New data needed to test that the e and µ are really different, and the implications of novel BSM and hadronic physics è BSM: scattering modified for Q 2 up to m 2 BSM (typically expected to be MeV to 10s of MeV), enhanced parity violation è Hadronic: enhanced 2γ exchange effects u Experiments include: MUSE tests these è Redoing atomic hydrogen è Light muonic atoms for radius comparison in heavier systems è Redoing electron scattering at lower Q 2 è Muon scattering! The MUSE Experiment.
23 MUSE Experiment u Simultaneous measurement of e + /µ + e - /µ - at beam momenta of 115, 153, 210 MeV/c in πm1 channel at PSI allows: è Determination of two photon effects è Test of lepton universality è Simultaneous determination of proton radius in both ep and µp scattering
Paul Scherrer Institute πm1 Beam u 590 MeV proton beam, 2.2mA, 1.3 MW beam, 50.6 MHz RF frequency u World's most powerful proton beam è Secondary e ±, µ ±, π ± in pim1 beamline u Separate out particle species by timing relative to beam RF u Cut as many pions as possible, trigger on e ±, µ ± 2 24
MUSE Experiment θ 20 o 100 o Q 2 0.002-0.07 GeV 2 3.3 MHz total beam flux 2-15% µ's 10-98% e's 0-80% π's u Low beam flux è Large angle, non-magnetic detectors u Secondary beam è Tracking of beam particles to target u Mixed beam è Identification of beam particle in trigger 25
26 SiPM Detector (TAU, Rutgers, PSI) u Provides timing vs. beam RF for trigger PID, precise offline analysis u Distinguishes muon scattering and decay using Δt with SPS u Thin (2mm) BC-404 read out by Hamamatsu S13360-3075PE SiPMs u 16 per plane, double ended readout, 2-4 planes depending on beam u Extensively prototyped, exceeds requirements
u Efficiency as required, working on readout speed 27 GEM Chambers (Hampton) Inefficiency Data from MUSE September 2015 Test beam time Efficiency u Set of 3 10cm x 10cm GEM detectors built for & run in OLYMPUS
28 Veto Detector (USC) u Annular 8-element veto detector surrounding target entrance window u Eliminate upstream scattering & beam decays
29 Liquid Hydrogen Target (U.Mich.) u Cell prototyping complete u Design safety review complete u Construction by U.Mich., PSI, CREARE
30 Straw Tube Tracker (HUJI, Temple) u Based on PANDA straw tube tracker design u 2 chambers with 5 planes each in x and y, 2850 straws
31 Scattered Particle Scintilators & Beam Monitor (USC) u Two planes on each side of beam u 92 bars, double-ended readout u 55 ps achieved
Back Wall Scintillator Time Resolution 32 55 ps average resolution
Readout TRB3: trb.gsi.de DAQ system (GWU & Montgomery College) u 3000 TDC, 500 ADC channels u TRB3-based read-out u Mesytec MQDC-32 ADCs mostly for timing correction Measured timing differences from 2014 PSI test beam time Trigger (Rutgers) u TRB3 FPGA-based, accept e ±, µ ±, reject π ± u SiPM PID && Scattered Particle (LUT) && NOT(veto) u PID determined by time between RF pulse and SiPM Relative timing test, σ = 32 ps 33
Mechanical Assembly (ANL & PSI) u Rotating table u Retractable beam tracker u Dedicated alignment procedures 34
Mechanical Assembly (ANL & PSI) 35
36 Several Beam Tests Time of flight relative to RF time (Fall 2012) Beam spot with GEM May 23, 2013
37 Simulations (USC) u Particle vertex and scattering angle reconstruction meet MUSE requirements u Background from target walls and windows can be cleanly eliminated or subtracted u Simulations verified by test data
TOF Beam Momentum Measurement t simulation t experiment = Δ(t Xcm - t 0cm ) In the simulation ü Use realistic beam profile ü Match to experimental time resolution ü Match to experimental particle flux 38
TOF Beam Momentum Measurement Consistent beam momenta were extracted from muon and pion spectra Good agreement between simulation and data, no evidence of beam tail from collimation p(π) p(µ) with Δp / p < 0.3% Preliminary results meet specifications 39
40 MUSE Error Budget Scintillator efficiency 0.1% Solid angle 0.1% Beam momentum offset u MUSE measuring relative cross sections u Point-to-point uncertainties, most important 0.1% Theta offset 0.2% Multiple scattering 0.15% Muon decay in flight 0.1% Radiative corrections 0.1% µ; 0.5% e Target wall subtraction 0.3% Beam PID mis-id 0.1% u Uncertainties mostly well controlled: largest from angle and radiative corrections. u Have six settings and two independent detectors, consistency check u Multiple calibration measurements / simulations planned
41 Projected Sensitivity for MUSE u Cross sections to < 1% stat. for backward µ, <<1% for forward e and µ, absolute 2%, point-to-point relative uncertainties to a few x 10-3 u Individual radius extractions from e ±, µ ± each to 0.01 fm +Q 4 +Q 6 linear
Projected Sensitivity for MUSE u Compare e± xsecs and µ± xsecs for TPE. Charge average to eliminate TPE to 0.01 fm u From e/µ xsec ratios: extract e-µ radius difference with minimal truncation error to 0.005 fm u If no difference, extract radius to 0.007 fm (2nd-order fit) *Note: MUSE point arbitrarily put at r p =0.875 fm 42
*Note: Difference in MUSE determined entirely by MUSE. Other differences are taken with respect to Antognini muonic hydrogen radius. 43 Projected Sensitivity for MUSE u Charge radius extraction limited by systematics, fit uncertainties u Many uncertainties are common to all extractions in the experiments, cancel in e+/e-, µ+/µ-, and µ/e comparisons u MUSE suited to verify 5.6σ effect (CODATA 2014) with even higher significance u R e - R µ = 0.034±0.006 fm (5.6σ), MUSE: δr = 0.005 fm (~7σ) Uncertainties on radius difference ~0.005 fm (stat.) ~0.1 fm (syst.)
44 MUSE Membership u 59 MUSE collaborators from 24 institutions in 5 countries u Funded by 5 agencies u TDR submitted to arxiv
Conclusion u Many efforts underway to explain the Proton Radius Puzzle! u Since 1 st MUSE proposal in February 2012, very many test beam times: è Determine beam line properties u Prototyped most of needed technology for the experiment u R & D from NSF, DOE, BSF; NSF construction funding (May '16) enables: è Funding & construction 2016 2017 è Production running 2018 2019 (2x 6 months) u MUSE will be the first muon scattering measurement with the required precision to address the PRP! 45
Anticipated MUSE Results u Extract radius from ep and µp form factors u Error on radius difference ~0.005fm u Error on radius ~0.01fm u MUSE will: è Verify the effect è Compare form factors è Compare cross sections è Test two photon effect è Solve the PRP?