HIMB Muon Conversion with high-Z target at - Andreas Knecht Paul Scherrer Institute 8. 4. 2021 HIMB Physics Case Workshop Zoom - PSI Indico
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Muon Conversion with high-Z target at HIMB Andreas Knecht Paul Scherrer Institute 8. 4. 2021 HIMB Physics Case Workshop Zoom
The predictions may, however, deviate byawayfactors from the2–3 at ons and muons at given detectors. grader which is used to remove pions from the beam (see momentum. Figure 1 shows the Sect. 2.3). Because of the large momentum band behind higher Z values. probability of 52 MeV/c pionsandand the degrader muons the many turns to crossmake the muons a CH in the2 SINDRUM II experiment As a result of the two-body final state3.1the Muonelectrons beam moderator of given thickness. transport solenoid no beam focus exists inside the spec- trometer. StillAs beamcan be seen particles fromreturn periodically thetofig- the produced in µ − e conversion are mono-energetic and their The 590 MeV proton beam ure has a the mean range differs by about a factor two and range time structure of 0.3 ns solenoid axis, i.e. every ≈ 65 cm for a typical momentum energy is given by: wide bursts every 19.75 ns. The πE5 secondary beam line after the degrader of 35 MeV/c and for this reason the gold straggling can be neglected so perfect separation should be extracts particles emitted in backward direction from the target was made in the form of a 65 cm long tube. expected with an 8 mm thick CH2 degrader. In practice the Eµe = mµ c2 − Bµ (Z) − R(A) , (1) distributions are broadened by the finite momentum band Measurementwhere of muon Bµ (Z) conversion is the atomic binding on gold energy of the muon and transmitted by the beam line. Settings of the beam mag- R(A) is the atomic recoil energy for a muonic atom with nets and slits have to be carefully adjusted to minimize the Removal of pions atomicfrom number beam mass number A. In first approxima- tails in the pion range distribution. At the cost of ≈ 30% Z and through degrader and tion Bµ (Z) ∝ Z 2 and subsequent R(A) ∝ A−1 . For muonic gold Bµ = loss in muon intensity the pion stops in the target could transport 10.08 MeV and R = 0.025 MeV give Eµe = 95.56 MeV. be suppressed to the required level. A simulation using the solenoid measured range distribution shows that about one in 106 pions cross the moderator. Since these particles are rela- Some residual2.2background Muon inducedevents background from tively slow 99.9% of them decay before reaching the target pions observed through timing (mostly which Fig. 2. Plan view of the experiment. Theis 1 MWsituated some 590 MeV proton beam 10extracted m further downstream. from the PSI ring cyclotron hits the 40 mm car- Muon344 decay in orbit (MIO) constitutes an intrinsic bon production back- target (top left of RPC the figure). may The πE5 produce beam line background transports secondary even particles (π,when taking µ, e) emitted place in the backward from degrader) The SINDRUM ground source which can only be suppresseda)with II Collaboration: direction to a A degrader sufficient shows the search situated for at µ the in measured momentum dispersion − e conversion entrance of the moderator. a in transport muonic solenoid The at the position of the gold connected axially first slitresulting to the SINDRUM electrons system. The momentum II spectrometer. havefrom≈the10% was calculated Inset flight time through the channel and the distributions show the increase when opening one side of the slit. Inset b) shows a cross section electron energy resolution. The process predominantly of the beam observedre-at the position of the beam focus 2 sults in electrons with energy EMIO below mµ c /2, the typically twice larger for positrons than for electrons [25]. kinematic endpoint in free muon decay, with a steeply Simulation shows that decay in flight is dominantly π− → falling high-energy component reaching up to Eµe . By e− ν e shortly before the pion would have reached the mod- using a magnetic spectrometer the vast majority of MIO erator. O(10) background events are expected with a flat electrons can be kept away from the tracking detectors still energy distribution between 80 and 100 MeV. The decay maintaining a ≈ 50% acceptance in the region of interest chain π− → µ− ν µ , µ− → e− νµ ν e has a similar yield but the around 95 MeV. In the endpoint region the MIO rate varies energy distribution falls steeply and is for all energies neg- as (Eµe − EMIO )5 and a resolution of 1–2 MeV (FWHM) ligible compared to MIO. As is illustrated in Fig. 10 there is indeed a beam corre- is sufficient to keep MIO background under control (see lated signal which is strongly peaked in the forward direc- Sect. 6.4 below). Since the MIO endpoint rises at lower Z tion and which does contain more electrons than positrons. great care has to be taken to avoid low-Z contaminations The observed time spread is ≈ 10 ns explained mainly by in andFig. 10. Prompt around beam induced background. Electron and the momentum spread in the beam. the target. positron events were selected with total momentum above Another background source is due to radiative muon To cope with π − induced background two event classes 87 MeV/c but outside the main signal region 92.5–95.5 MeV/c. capture Panel(RMC) µ− (A, Z) a: spectrometer → γ(A, timing Z −to relative 1)∗the νµ 50.6 after MHz which cy- have been introduced based on the values of polar angle the photon − e+ edifferences and rf phase: clotron creates rf signal. an Time pair either internally with respect to both (Dalitz the Fig. 1. Fraction of pions and muons with a momentum of + − ora|tfunction pair) previous or through γ →next and the e erf bucket pair production in the are incremented. target. Shown is – Class 52 MeV/c 1 contains that cross eventsa CH with cos θ < 0.4as 2 moderator rf − 10 ns|of >the mod- the difference The RMC endpoint between can be thekept distributions below Ewithµe forcosselected θ > 0.4 and iso- erator 4.5 nsthickness. which areGEANT practically [23]free of pion induced back- simulation cos θ < −0.4. Panel b: cos θ distribution. Shown is the distri- ground. bution corresponding to the phase enhancement by the pion – Class 2 contains events with cos θ > 0.4 and |trf − induced events. The arrows indicate the region populated by 10 ns| < 4.5 ns which are contaminated by pion induced pion induced events. See the text for a discussion of the nature background. of these events Eur. Phys. J. C 47, 337–346 (2006) 6.4 Single-event sensitiviy Andreas Knecht cosmic background can be distinguished. Most common 2
SINDRUM II experiment 340 The SINDRUM II Collaboration: A search for µ − e conversion in muonic gold Fig. 3. The SINDRUM II spec- trometer. Typical trajectories of a beam muon and a hypo- thetical conversion electron are indicated 3.2 SINDRUM II spectrometer DC2 is made of low-density foam sandwiched between alu- minized Kapton foils and has a density of only 35 mg/cm2 . Figure 3 shows a vertical cross section through the SIN- To stabilize its position DC1 is kept at an overpressure of DRUM II spectrometer in the configuration used for this 1 mbar relative to DC2. The aluminum on the inside of the experiment. Beam is entering from the solenoid (A in wall is divided into 4.4 mm wide Eur. Phys. J. C 47, 337–346 (2006) helical strips which al- Fig. 3) on the left. The gold target (B) with a radius of lows 3-dimensional track reconstruction. There are separate Andreas Knecht ≈ 20 mm and a wall thickness of 75 mg/cm2 was produced strips for the upstream and downstream halves of the de-3
Why high-Z? COMET TDR Scaling of branching ratio of muon conversion as a function of Z differs depending on underlying physics -> allows to understand underlying physics in the case of a discovery High-Z atoms not ideal for pulsed beams due to short lifetimes Andreas Knecht 4
Muon conversion at HIMB Muon rates: SINDRUM II: ~ 2e6 μ-/s at 52 MeV/c (not written explicitly, but estimated from beam power at that time and given muon stops on target) 346 The SINDRUM II Collaboration: A search for µ − e conversion in muonic gold HIMB: ~ 1e8 μ-/s at 40 MeV/c (highest momentum with good transport efficiency) capture probability: Sensitivity: 11. SINDRUM Collaboration, U. Bellgardt et al B 299, 1 (1988) SINDRUM II: Bµe Au < 7 × 10−13 90% C.L. (4) 12. W. Bertl et al., Prepared for International Conference on High-Energy Physics (HEP HIMB: Can This increase limit the sensitivity is more stringent by aof factor by two orders magnitude 100 if backgrounds dapest, Hungary,can beJulreduced 12–18 2001. accordingly than the best previous limit on a heavy target [26]. It is the 13. A. van der Schaaf, J. Phys. G 29, 1503 (2003 final result of the research program on rare π and µ decays 14. T. Suzuki et al., Phys. Rev. C 35, 2212 (1987 with the SINDRUM I and II spectrometers at PSI. The 15. A. Czarnecki, W.J. Marciano, K. Melnikov search for LFV in rare muon decays is continued at PSI by Proc. 549, 938 (2002) the MEG collaboration [27] aiming at a sensitivity of 10−13 16. T.S. Kosmas, I.E. Lagaris, J. Phys. G 28, 290 17. R. Kitano, M. Koike, Y. Okada, Phys. Rev. for the µ → eγ decay. (2002) [arXiv:hep-ph/0203110] 18. T.S. Kosmas, J.D. Vergados, O. Civitarese Nucl. Phys. A 570, 637 (1994) References 19. E.A. Hermes, H.P. Wirtz, F. Rosenbaum, N Meth. A 413, 185 (1998) 1. R.D. McKeown, P. Vogel, Phys. Rept. 394, 315 (2004) 20. M. Grossmann-Handschin et al., Nucl. Instr 2. A. van der Schaaf, Prog. Part. Nucl. Phys. 31, 1 (1993) 327, 378 (1993) Andreas Knecht 3. T.S. Kosmas, G.K. Leontaris, J.D. Vergados, Prog. Part. 5 21. B. Robert-Tissot, ”Analyse de Transition 197
Muon conversion at HIMB Pion contamination: The SINDRUM II Collaboration: A searc HIMB: Can also use degrader at 40 MeV/c (less efficient compared to 50 MeV/c, but due to lower momentum ~10x less pions in the beam) Add bend in transport section after degrader to reduce backgrounds coming from moderator With better detector, active target, and the above background reduction, can probably keep pion backgrounds low enough to capitalise on higher muon rate Fig. 11. Momentum distributions of electrons and positrons for the two event classes. Measured distributions are compared with the results of simulations of muon decay in orbit and µ − e conversion Eur. Phys. J. C 47, 337–346 (2006) decay in orbit (MIO) which is the dominant source of 40 MeV/c background. Figure 11 shows momentum spectra of elec- trons and positrons for the two event classes introduced 3% Δp/p in Sect. 6.3. In general the electron distribution of sam- ple 1 is well described by muon decay in orbit. Whereas no events are observed with energies expected for µ − e conversion at higher energy an electron and a positron event have been found. Since cosmic ray background con- Andreas Knecht tains much more electrons than positrons these events6 are most likely caused by pions. In sample 2 the elec-
Mu2e/COMET Mu2e and COMET will perform their measurements with expected muon rates of around 1e10 μ-/s They will improve the sensitivity over SINDRUM II by a factor 104 Changes in expected branching ratio as a function of Z are on the level of
You can also read