Muon g-2 Precision Precession Steve Maxfield University of Liverpool sjm@hep.ph.liv.ac.uk 1 Stephen Maxfield Seminar Birmingham Oct2013
OUTLINE • What is it? • Why measure it (again)? • How? • Goals and how to achieve them: • Brief recap of technique • Upgrades! • Beam, detectors, field • Status and Conclusions † The material for this talk has been shamelessly stolen from many including: B. Lee Roberts, Leah Welty-Reiger, Mark Lancaster, Thomas Teubner, Chris Polly, Andreas Kronfeld , Ruth Van de Water…… 2 Stephen Maxfield Seminar Birmingham Oct2013
Magnetic Moments • Magnetic moment of elementary particles related to their spin by the “g - factor” Qe g S B S 2 m Larmor frequency A little history… 1924 Stern-Gerlach Magnetic moment of silver atom in it’s ground state is 1 Bohr magneton. (10%) …but not understood as spin 1/2 3 Stephen Maxfield Seminar Birmingham Oct2013
Spin ½? 1925/26 Uhlenbeck And Goldschmidt proposed electron spin to explain fine structure…. …but prediction off by factor of 2! Rescued by Thomas precession (1926) - relativistic kinematics effect (successive non-collinear boosts give rotation). 4 Stephen Maxfield Seminar Birmingham Oct2013
1928 g=2 0 i ieA m Non-relativistic reduction 2 p e 1 2 i L S B t 2 m 2 m g 2 g S 1 L 5 Stephen Maxfield Seminar Birmingham Oct2013
Greater Experimental Precision... …1947 ( Nafe , Nelson, Rabi)Hyperfine structure of H and D did not fit g=2…(It was a 5 sigma effect) 1948 Kusch and Foley : A precision measurement: g e =2(1 .00119 ± 0.00005 ) g 2 a An anomaly! Define 2 It takes QED to begin to explain the anomaly… a 0.001161 e 2 6 Stephen Maxfield Seminar Birmingham Oct2013
More QED… 2 3 4 5 g 1 C C C C C 1 2 3 4 5 2 Laporta,, Kinoshita et al. Rameddi Some very Even analytically… weird diagrams! 7 Stephen Maxfield Seminar Birmingham Oct2013
Status of electron g-2 Together with a succession of experiments exp 12 a 1159652180.73(28) 10 e 12 a 1.05 0.82 10 Ultra-precise agreement e Gives best value of E thy 12 a 1159652181.78(77) 10 e 8 Stephen Maxfield Seminar Birmingham Oct2013
a Standard Model Physics predicts electron magnetic moment anomaly at ppt level! But the story is different for the muon … It’s heavier More sensitive to more contributions… + … + + (Hadronic corrections only enter around 12 th decimal place in a e ) 9 Stephen Maxfield Seminar Birmingham Oct2013
QCD • QED well known • EW contributions also understood (only couple of loop accuracy needed) • Hadronic contributions are significant and the biggest source of uncertainty. Non-perturbative - cannot be calculated. Determined from experiment Low energy e + e - hadrons. + some lattice QCD for L-by-L contribution 10 Stephen Maxfield Seminar Birmingham Oct2013
e + e - hadrons new Older e + e - data new PHIPSI13 K + K - Rome, 2013 11 Stephen Maxfield Seminar Birmingham Oct2013
How the contributions stack up: Determination of hadronic contribution to muon g-2 has become an industry DHMZ HLMNT Paralleled by g- 2 measurements… 12 Stephen Maxfield Seminar Birmingham Oct2013
Experimental Determination of a . A succession of improving measurements FNAL GOES HERE …Details to follow! 13 Stephen Maxfield Seminar Birmingham Oct2013
The current state of the art: Not same precision as the electron but compensated by higher mass. Muon anomalous magnetic moment is sensitive to most of the standard model… and to new physics. A tantalising but inconclusive 3.3-3.6 s discrepancy 14 Stephen Maxfield Seminar Birmingham Oct2013
There is no shortage of interest in this intriguing result! Were it to persist… • Strong indicator of BSM physics … Loop contributions sensitive to new particles running round loop… 2 m better than e 40,000 m e 15 Stephen Maxfield Seminar Birmingham Oct2013
NP e.g. SUSY But broad spectrum of sensitivity in TeV mass range… flavour-conserving, CP-conserving, chirality a related to m flipping, loop-induced highly model dependent C Generically: 2 2 NP m m NP a O 1 M m NEW 16 Stephen Maxfield Seminar Birmingham Oct2013
NP a provides discriminating power… 17 Stephen Maxfield Seminar Birmingham Oct2013
NP …also can inform, low mass, below LHC reach… e.g. Dark photons: 18 Stephen Maxfield Seminar Birmingham Oct2013
How do we measure g-2? 19 Stephen Maxfield Seminar Birmingham Oct2013
Fortunes of Nature First make your muons … … from pions . Fortune of nature number 1 Parity violation delivers conveniently polarised muons: beam of polarised muons 20 Stephen Maxfield Seminar Birmingham Oct2013
Fortunes of Nature inject i nto a (very) uniform magnetic field… QeB Muon momentum turns with cyclotron frequency C m QeB QeB g 1 Spin turns with frequency S 2 m m Fortune of nature number 2: Direct dependence on the anomaly: an g 2 QeB QeB a immediate 3 orders of magnitude gain a S C over measuring in at-rest 2 m m experiments ! We need to measure and B a …and know very accurately? m 21 Stephen Maxfield Seminar Birmingham Oct2013
Actually measure: Normalise magnetic field to Larmor frequency of proton a p a a p p Measured from hyperfine structure of muonium: currently known to 120ppb † p † JPARC expt. to reduce this to ppb level 22 Stephen Maxfield Seminar Birmingham Oct2013
E821 at BROOKHAVEN 23 Stephen Maxfield Seminar Birmingham Oct2013
E821 21 Experimental erimental Te Technique chnique x c ≈ 77 mm 25ns bunch of 5 X 10 12 protons q ≈ 10 mrad π from AGS B·dl ≈ 0.1 Tm Pions Inflector p=3.1GeV/c B (1.45T) Target Injection orbit • Muon polarization Central orbit • Muon storage ring Storage • injection & kicking ring Kicker • focus with Electric Quadrupoles Modules • 24 electron calorimeters R=711.2cm d=9cm R R q 24 Electric Quadrupoles x c Stephen Maxfield Seminar Birmingham Oct2013
Magic • Vertical magnetic field – need vertical focussing to stop muons spiralling out of ring • Achieve using electrostatic dipoles • The E-field modifies the precession frequency: e 1 aB a E a 2 mc 1 • Unwelcome source of additional systematics Can be made to vanish for ‘magic’ . Extremely lucky that size of a makes • this possible! GeV 29.3 p 3.09 magic …but sadly, not every will be magic! Method pioneered by 3 rd CERN g-2 25 Stephen Maxfield Seminar Birmingham Oct2013
But how to measure a ? Parity violation again! • Highest energy e + emitted along direction of + spin • Use calorimeters to count e + above an energy threshold vs. t 26 Stephen Maxfield Seminar Birmingham Oct2013
…an iconic plot E 821 t / N t N e 1 A cos t 0 a “5 - parameter fit” 27 Stephen Maxfield Seminar Birmingham Oct2013
Measuring a …some reality Simulated for E989 High frequency modulation because muon bunch initially a doesn’t fill ring…decays as bunch spreads. This is good – can get p distribution of muons Expected for E989: c 149ns Bunch length 120ns at injection t / N t N e 1 A cos t 0 a N,A depend on energy 28 Stephen Maxfield Seminar Birmingham Oct2013
Many sources of systematic error. Particularly insidious are ‘early -to- late’ errors Example: Effect of pile up. a t 2 t t t t Time dependence in phase: 0 0 cos t t cos(( ) t ) a a 0 …but why should change? Things which change early to late in the fill can lead to a phase change in the accepted events direct bias to extracted a . 29 Stephen Maxfield Seminar Birmingham Oct2013
Higher energy positrons come from further away. If we get the energy wrong, we get the phase wrong. spin If we have pile-up, two low energy positrons fake a high energy positron. More pile-up early in the fill. Beam relaxation Vertical breathing 3 CBO terms Muons lost from ring 30 Stephen Maxfield Seminar Birmingham Oct2013
Recommend
More recommend