TThe Muon g-2 Experiment Overview and Status
J. L. Holzbauer ∗ University of MississippiE-mail: [email protected]
The Muon g-2 experiment at Fermilab will measure the anomalous magnetic moment of themuon to a precision of 140 parts per billion, which is a factor of four improvement over theprevious E821 measurement at Brookhaven. The experiment will also extend the search for themuon electric dipole moment (EDM) by approximately two orders of magnitude. Both of thesemeasurements are made by combining a precise measurement of the 1.45T storage ring magneticfield with an analysis of the modulation of the decay rate of the higher-energy positrons from the(anti-)muon decays recorded by 24 calorimeters and 3 straw tracking detectors. The current statusof the experiment as well as results from the initial beam delivery and commissioning run in thesummer of 2017 will be discussed.
The 19th International Workshop on Neutrinos from Accelerators-NUFACT201725-30 September, 2017Uppsala University, Uppsala, Sweden ∗ Speaker. c (cid:13) Copyright owned by the author(s) under the terms of the Creative CommonsAttribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). https://pos.sissa.it/ a r X i v : . [ h e p - e x ] D ec he Muon g-2 Experiment Overview and Status J. L. Holzbauer
1. Overview
The Muon g-2 experiment at Fermilab is designed to measure the anomalous magnetic momentof the muon to high precision [1]. It is an improved version of an experiment at BNL called E821,which measured the value to be approximately three standard deviations away from the theoreticalvalue [2]. Such an interesting possibility of new physics in a precision experiment makes the newmeasurement extremely important. The experiment is designed to make the measurement moreprecisely and potentially indicate whether or not there is really a sign of new physics in this sector.In Dirac theory, the g factor relates the spin and the magnetic moment of the muon and isexactly two [3]. However, higher order effects cause the value to be slightly different, and thisdeviation from two is known as the anomaly, namely a µ = (g-2)/2. If the anomaly varies fromthe standard model expectation, it can indicate new physics, such as a heavy W in certain SUSYmodels [4]. This anomaly is the value the Muon g-2 experiment is designed to measure, and alsocontributes to its name.The standard model value of a µ is estimated separately by theorists. It has several parts: quan-tum electrodynamics (QED) [5], electroweak (EW) [6], hadronic vacuum polarization (HVP) [7, 8]and hadronic light-by-light (HLbL) [9]. As can be seen in Table 1, the QED and EW parts areknown very well. The others, being hadronic terms, are less well known and are estimated usingvarious techniques, including data from other experiments (such as e+e- to hadrons), and latticecalculations.The E821 [2] measurement at BNL, corrected for updated constants [10] and given in [1], is a E µ = ± × − (54 ppm). The uncertainty is slightly larger than the standardmodel calculation uncertainty. It is expected that the current experiment will be able to measurethe value with an uncertainty approximately four times lower, 0.14ppm. If this happens and thecentral value remains the same, the deviation of a µ from the standard model expectation would beat least 5 sigma. Expected improvements to the understanding of the standard model value wouldgive a deviation of 8 sigma under these same assumptions [1]. Table 1:
Standard model components of the anomaly, taken directly from [1]. Two values are shown forHVP to reflect two recent estimates. The terms lo and ho indicate lower order and higher order, respectively.Other terms are defined in the text.
Values in 10 − unitsQED ( γ + l ) 116584718.951 ± ± ± ± ± ± ± ± ± ± H − LO ± H − HO ± other ( ± tot )Total SM [8] 116591828 ± H − LO ± H − HO ± other ( ± tot )1 he Muon g-2 Experiment Overview and Status J. L. Holzbauer
2. The Muon g-2 Experiment
Although the Muon g-2 experiment is located at Fermilab, the similarities in design to E821suggested the reuse of several parts from the BNL experiment. Thus, it was decided to movevarious components across the country, most notably the 15 ton cryostat ring, which was moved bybarge and truck, with the requirement that the superconducting coils not flex by more than 3mm. Inthe summer of 2015 the dipole magnet and cryo-system were assembled at Fermilab and achievedthe 1.45 Tesla field, indicating a successful transport.The muons for the storage ring are produced by the Fermilab accelerator complex. The sourceprotons are reused from the Tevatron operations, and they collide with an Inconel target to producepions. These pions enter a long delivery ring where they decay to muons and neutrinos beforeentering the Muon g-2 storage ring as a polarized positively charged muon beam. The extra timebetween the target and the storage ring helps to reduce beam contamination. The new experimentexpects a factor of 20 increase in muon statistics over the BNL experiment and which will reducethe statistical error to 0.1 ppm.In brief, the experiment consists of a cryo-system surrounding C-shaped dipole magnets,which contain rectangular vacuum chambers with electrostatic quadrupole plates and kicker platesinside, depending on the region of the ring. The magnets and quadrupole plates work in tandem tokeep the muons in the proper location after the kickers move the beam into a stable circular orbit.Other components include an inflector magnet to reduce the field as the beam enters the ring, colli-mators to control the beam, a beam monitoring system, and a trolley that rides around inside of thering under vacuum to measure the magnetic field with NMR probes. Detector systems include bothstraw trackers and lead crystal calorimeters with multiple readout channels. The dipole magnetsystem is C-shaped to allow less material between the calorimeters and the muon decay points.Most major components like the vacuum chambers for the storage ring, dipole magnets, andelectro-static quadrupole plates were reused from the E821 experiment. However, these compo-nents were disassembled and reassembled, including re-shimming the magnet to create a constantfield and realigning the quadrupole plates. Vacuum chambers were modified to include new sys-tems inside the storage ring. Many smaller components for these various sub-systems are new, andin some cases the assembly procedures are also different than those from E821. Other changes in-clude improved calorimeter resolution and calibration system, and the presence of a straw trackingsystem, which is completely new for the current experiment. The kicker plates were redesigned, aswere the collimators. Additionally, a thinner quadrupole plate was added near the region of beamentry, to reduce muon losses before the beam is kicked into orbit by the kickers, which are located90 degrees down the ring from the beam entry position.
3. The Measurement Procedure
There are two quantities that compose the measurement of a µ , namely ω a and ω p . The value ω a is the precession frequency of the muon with respect to the momentum, which is measured viahigh energy decay positrons, and ω p is the magnetic field (B) normalized to the proton lamourfrequency. The spin and cyclotron frequencies can be combined to give a µ in terms of ω a , B, andthe charge over mass ratio. This can then be rewritten in terms of ω a / ω p and the muon, proton2 he Muon g-2 Experiment Overview and Status J. L. Holzbauer magnetic moment ratio from hyperfine splitting, which is obtained elsewhere. Additionally, if onewere to write out the full form of ω a , one would notice an electric field term. This term will bezero with the appropriate choice of muon momenta, 3.09 GeV/c, known colloquially as the magicmomentum, and corresponding to a magic radius (which sets the radius of the storage ring). Ofcourse, not all muons will have this exact momentum, and this is accounted for as an analysisuncertainty. This experiment technique also makes alignment efforts particularly important.The value of ω p is obtained from measurements of the magnetic field, which is designed to beas constant as possible. There is a device called a trolley which has NMR probes mounted on it andtraverses the storage ring where the beam is located (when the beam is off). A second larger trolleywas also used to make measurements before the installation of the storage ring vacuum chambers,to have access to higher order field modes. Additionally, there are NMR probes permanently fixedto the tops and bottoms of the storage ring vacuum chambers. These can run when the beam is onor off and are used to relate the NMR trolley measurement to a beam-on measurement. Finally,absolute measurements are taken in a particular region of the ring using specialized probes, and therelative NMR measurements are related to this.The quantity ω a is obtained from high energy positron data. Calorimeters in 24 stations aroundthe ring record energy deposits from these particles and are used to make the so-called wiggle plot,which shows a varying number of events with time in an oscillatory pattern (in addition to thenormal decrease due to a finite number of particles decaying). Additional information from trackersystems and others help to reduce lost muon contamination and estimate various uncertainties onthe measurement.
4. 2017 Commissioning Run
In the summer of 2017, the Muon g-2 experiment had a short commissioning run to test thesystems and procedures, during which the accelerator complex ran with a reduced rate relative tothe physics run plan. On May 23, the first particles were delivered to the ring and beam splash wasseen in the calorimeters. The run ended July 7. Over the course of the run, the experiment wasable to achieve particles circulating through the full storage ring and demonstrated the operationsof all systems. In Figure 1 we see a distribution of energies detected in the calorimeters, with peakscorresponding to lost muons and protons. In Figure 2, a reconstructed track is shown in two spatialdimensions, compared to the magic radius location, of a positron that traversed several straws,again demonstrating an operational system. Finally, in Figure 3, the so-called wiggle plot is shownfor the commissioning data, along with a preliminary analysis fit. This distribution was createdfrom two weeks of data taken in June of 2017 and contains approximately 700,000 positrons. Thestatistics correspond to results similar to that of the older CERN-II experiment which precededE821 [11].Overall, the run was quite successful. The next run, which will be used for physics dataanalyses, will take place primarily in 2018. It will be preceded by a second commissioning phasethat began in November of 2017 as systems were restarted.3 he Muon g-2 Experiment Overview and Status
J. L. Holzbauer
Figure 1:
Energy distribution from June 2017 data recorded in the calorimeter. The peaks are from protonsand lost muons.
Figure 2:
One of the first tracks recorded by the tracker showing the hits from a single charged particle(likely a proton) through the straw trackers and the (wire)-track fit and the magic radius.
Figure 3:
Distribution of position counts from two weeks of data accumulated in June 2017. he Muon g-2 Experiment Overview and Status J. L. Holzbauer
5. Summary and Plans
The Muon g-2 experiment is entering an exciting data taking phase. The machine is assem-bled and operational, and capable of producing the various quantities needed to make the intendedmeasurement. It is expected that the new experiment will improve the uncertainty on the measuredvalue versus the previous measurement by a factor of four. With the data taken in 2018, the Muong-2 experiment has the potential to finally resolve the tension of the previous result and the standardmodel calculation.
References [1] Grange J et al. (Muon g-2) 2015 (
Preprint )[2] Bennett G et al. (Muon G-2 Collaboration) 2006
Phys.Rev.
D73
Preprint hep-ex/0602035 )[3] Dirac P 1928
Proc. R. Soc. (London)
A117, 610 and A118, 351 [4] Albrecht J et al. (Intensity Frontier Charged Lepton Working Group) 2013 Working Group Report:Charged Leptons
Proceedings, Community Summer Study 2013: Snowmass on the Mississippi(CSS2013): Minneapolis, MN, USA, July 29-August 6, 2013 ( Preprint ) URL https://inspirehep.net/record/1265506/files/arXiv:1311.5278.pdf [5] Aoyama T, Hayakawa M, Kinoshita T and Nio M 2012
Phys. Rev. Lett. (11) 111807 URL http://link.aps.org/doi/10.1103/PhysRevLett.109.111807 [6] Gnendiger C, Stockinger D and Stockinger-Kim H 2013
Phys. Rev.
D88
Preprint )[7] Davier M, Hoecker A, Malaescu B and Zhang Z 2011
Eur. Phys. J.
C71
Preprint )[8] Hagiwara K, Liao R, Martin A D, Nomura D and Teubner T 2011
J. Phys.
G38
Preprint )[9] Prades J, de Rafael E and Vainshtein A 2009
Adv. Ser. Direct. High Energy Phys. Preprint )[10] Mohr P J, Taylor B N and Newell D B 2008
Rev. Mod. Phys. (2) 633–730 URL http://link.aps.org/doi/10.1103/RevModPhys.80.633 [11] Farley F and Picasso E 1990 The cern (g-2) measurements Quantum Electrodynamics ed Kinoshita T(World Scientific) p 479ed Kinoshita T(World Scientific) p 479