Search for Extra Scalars Produced in Association with Muon Pairs at the ILC
SSearch for Extra Scalars Produced in Associationwith Muon Pairs at the ILC
Yan Wang (cid:5)∗ , Jenny List (cid:5) , Mikael Berggren (cid:5) (cid:5)
DESY, Notkestraße 85, 22607 Hamburg, Germany ∗ IHEP, 19B Yuquan Road, Shijingshan District, Beijing, ChinaE-mail: [email protected] on behalf of the International Large Detector concept group
We study the search for an extra scalar S boson produced in association with the Z boson at theInternational Linear Collider (ILC). The study is performed at center-of-mass energies of 250GeV and 500 GeV based on the full simulation of the International Large Detector (ILD). Inorder to be as model-independent as possible, the analysis uses the recoil technique, in particularwith the Z boson decaying into a pair of muons. As a result, exclusion cross-section limits aregiven in terms of a scale factor k with respect to the Standard Model Higgs-strahlung processcross section. These predicted results, covering all possible searching regions of the extra scalarsat the 250 GeV ILC and the 500 GeV ILC, can be interpreted independently of the decay modesof the S boson. Talk presented at the International Workshop on Future Linear Colliders (LCWS2018),Arlington, Texas, 22-26 October 2018.C18-10-22. a r X i v : . [ h e p - e x ] F e b earch for Extra Scalars at the ILC
1. Introduction
The motivation of our study is to find a new scalar S boson in the SZZ coupling since oneor more extra scalars are predicted in many new physics models. However, the properties of 125GeV scalar measured at the LHC is very similar to the Standard Model (SM) Higgs boson [1].As a result, the new scalar’s coupling will be highly suppressed [2]. Furthermore, the LEP/LHCconstraints on the extra scalars always rely on the model details. Thus, a more precise analysiswith model-independent assumptions to a scalar with the small coupling is preferred. Although theOPAL collaboration has searched for light scalars (less than 100 GeV) in a model-independent wayat LEP, the results are limited by the low luminosity [3]. The International Linear Collider (ILC) isa proposed electron-positron linear collider, whose luminosity will be over a thousand times higherthan that of LEP, which makes the recoil mass technique more accurate to find such extra scalars[4]. And the ILC has higher center-of-mass energies, which will cover more searching regions forthe extra scalar. A preliminary version of this analysis has been reported at LCWS2017 [5] andICHEP2018 [6]. Thus, only the updates from ICHEP2018 is summarized in this contribution.
2. Event Generation and Detector Simulation
The signal is e + e − → S + Z production, where the Z boson decays to a pair of muons. Thedecay branching ratios of S are fixed as same as the 125 GeV Higgs boson, but no use wouldbe made of this fact. As SM backgrounds, bremsstrahlung and initial state radiation (ISR) areexplicitly considered for all events. The event samples are generated with 100% left-handed andright-handed beam polarization, using the Whizard 1.95 Monte Carlo (MC) event generator [7].Then the samples are reweighted with beam polarizations of ±
80% for the electron beam and ±
30% for the positron beam.The event samples are generated, simulated and reconstructed for different center-of-massenergies ( √ s =
250 GeV and 500 GeV). In 250 GeV cases, we use the same setting as the samplesgenerated in the context of the ILD Detailed Baseline Design document [8]. The fractions ofintegrated luminosity 2000 fb − are dedicated to the four sign combinations ( − + , + − , ++ , −− ) =( , , , ) . The signal benchmark points are chosen as every 5 GeV in the range of10 ≤ M S ≤
160 GeV (totally 30 signal benchmark points). In 500 GeV cases, we use the samplesgenerated in the context of the ILD Design Report [9], the fractions of integrated luminosity 4000fb − are dedicated to ( − + , + − , ++ , −− ) = ( , , , ) . Totally 48 signal benchmarkpoints are chosen in the range of 10 ≤ M S ≤
410 GeV. Event reconstruction has been performedusing the PandoraPFA algorithm to reconstruct individual final state particles, so-called ParticleFlow Objects (PFOs).
Firstly, a pair of oppositely charged muons is selected by minimizing the following χ func-tion: χ ( M µ + µ − , M rec ) = ( M µ + µ − − M Z ) σ M µ + µ − + ( M rec − M S ) σ M rec , (2.1)1 earch for Extra Scalars at the ILC where M µ + µ − and M rec are the invariant mass and the recoil mass of the muon pair, and σ M µ + µ − and σ M rec are determined by a Gaussian fit to the generator-level distributions of M µ + µ − and M rec . Then, the bremsstrahlung and final state radiation photons from the muon are combinedwith the muon.Background events are rejected by firstly considering kinematic variables only relied on muons(and the reconstructed Z boson): the invariant mass and transverse momentum of the muon pair, aswell as the polar angle of the missing momentum. Then, a BDTG is trained using 6 input variablesbased on TMVA [10]: muon pair invariant mass, the polar angle of each muon, the polar angleof the muon pair, the opening angle of the muon pair, and the π − ( φ µ + − φ µ − ) , where φ µ ± isthe azimuthal angles of the muons with respect to the beam line. Finally, taking into account theISR photon return effects, the two fermion background can be further rejected by ISR energy vetocuts. With these cuts, no information on the decay of S is needed, thus the expected results will bemodel-independent. The recoil mass distributions are obtained after these cuts (Figure 1). Figure 1:
The recoil mass distributions for signal and backgrounds at the 250 GeV center of mass energy.These distributions are before the ISR veto cuts.
3. Results
A likelihood analysis is applied for calculating 2 σ expected exclusion limits on k with a bin-by-bin comparison between the signal and background recoil mass histograms for each benchmarkpoints, where k is defined as k = σ SZ σ H SM Z ( m H SM = m S ) , (3.1)and k is the 2 σ exclusion limits for the cross section scale factor k hereinafter.In Figure 2, the ILC results at 250 GeV are compared with the LEP results directly. The redpoints are 2 σ exclusion limits for (cid:82) Ldt = − and √ S =
250 GeV at the ILC, while the redline was obtained with the recoil mass method by the OPAL Collaboration [3] at LEP with about2 earch for Extra Scalars at the ILC
Figure 2:
The directly comparison between the LEP and ILC simulation results.
Figure 3:
The comparison between the theoretical LEP extroplation results and ILC expected results. earch for Extra Scalars at the ILC − data in total. Also shown with the blue line is the model-dependent results from LEP,combining measurements by ALEPH, DELPHI, L3, and OPAL [11], in which the decay modes ofthe scalars were utilized. In general, the ILC exclusion limits will reach 10 − , and are one or twoorders better than the OPAL recoil results and even better than the LEP traditional results.In Figure 3, the ILC theoretical predictions, which are extrapolated from the LEP measure-ments with fixed (variable) scalar width, are compared with the ILC simulation results [12]. Thetheoretical predictions combine S µ + µ − and Se + e − channels, while the ILC simulation resultsonly use S µ + µ − channel, but divide the results by √ k = k exp ( S µ + µ − ) / √ − luminosity with P ( e − , e + ) = ( − , + ) . From the figure, theILC simulation results agree to the theoretical predictions with fixed scalar width in the low massregion. However, the theoretical predictions extrapolate the expected background events in an in-terval around the Z pole region [12], so there is no Z pole peak in the theoretical curves; at the sametime, the theoretical predictions don’t include SM Higgs background, as a result, they are betterthan simulation results in the high mass region. Figure 4:
The effects for ISR photon veto cuts and the reconstruction efficiency.
In Figure 4, a comparison is given among the PFO and their corresponding MC simulationinputs (MCtruth). The red and magenta points are the results with/without the ISR photon vetocuts on the level of PFOs, because the two fermion background can be efficiently discarded byconsidering the ISR effects. Thus, the ISR photon reconstructed efficiency will affect the resultsignificantly. The black points are the results after photon veto cuts using MCtruths with the detec-tor simulation, which reflect the best searching capability. And the difference between MCtruthsand PFOs results shows we can improve the results with better photon reconstructions.In Figure. 5, we show the preliminary exclusion limits for the 500 GeV ILC. In the low massregion, the 500 GeV results are worse than the 250 GeV cases mainly due to the suppressed cross4 earch for Extra Scalars at the ILC
Figure 5:
Preliminary Final Exclusion Limits for 250/500 GeV ILC. sections, while they cover a larger searching region. Especially when M S <
300 GeV, k is in theorder of 10 − , which could set strong model-independent constraints for the extra scalars.
4. Conclusions
By applying the recoil technique, the potential of the ILC to search for scalars has been inves-tigated at √ S =
250 GeV and 500 GeV, with the full simulation of the ILD concept. The method isoptimized to be independent of the scalar decay modes. 2 σ expected exclusion limits for the crosssection scale factor k are shown for scalar mass from 10 GeV to 160 GeV when √ S =
250 GeVand from 10 GeV to 410 GeV when √ S = Acknowledgements
We would like to thank the LCC generator working group and the ILD software workinggroup for providing the simulation and reconstruction tools and producing the Monte Carlo samplesused in this study. This work has benefited from computing services provided by the ILC VirtualOrganization, supported by the national resource providers of the EGI Federation and the OpenScience GRID. We are grateful for the support from Collaborative Research Center SFB676 of theDeutsche Forschungsgemeinschaft (DFG), Particles, Strings and the Early Universe, project B1.Y.W. is supported by the China Postdoctoral Science Foundation under Grant No. 2016M601134,and an International Postdoctoral Exchange Fellowship Program between the Office of the NationalAdministrative Committee of Postdoctoral Researchers of China (ONACPR) and DESY.5 earch for Extra Scalars at the ILC
References [1] The ATLAS and CMS collaboration, "Measurements of the Higgs boson production and decay ratesand constraints on its couplings from a combined ATLAS and CMS analysis of the LHC pp collisiondata at √ S = √ s = 250 GeV at the ILC", arXiv:1801.08164 [hep-ex].[6] Y. Wang, J. List,and M. Berggren, "Search for Light Scalars Produced in Association with a Z bosonat the 250 GeV stage of the ILC", ICHEP2018 Proceeding, PoS(ICHEP2018) 630.[7] W. Kilian, T. Ohl and J. Reuter, "WHIZARD: Simulating Multi-Particle Processes at LHC and ILC",Eur. Phys. J. C , 1742 (2011).[8] H. Abramowicz et al., "The International Linear Collider Technical Design Report",ILC-REPORT-2013-040.[9] ILD Detector Collaboration, "ILD Design Report" (2019).[10] TMVA home page https://root.cern/tmva.[11] R. Barate et al., "Search for the standard model Higgs boson at LEP", Phys. Lett. B565 (2003) p.61-75.[12] P. Drechsel, G. Moortgat-Pick, and G. Weiglein, "Sensitivity of the ILC to light Higgs masses",arXiv:1801.09662[hep-ph]., 1742 (2011).[8] H. Abramowicz et al., "The International Linear Collider Technical Design Report",ILC-REPORT-2013-040.[9] ILD Detector Collaboration, "ILD Design Report" (2019).[10] TMVA home page https://root.cern/tmva.[11] R. Barate et al., "Search for the standard model Higgs boson at LEP", Phys. Lett. B565 (2003) p.61-75.[12] P. Drechsel, G. Moortgat-Pick, and G. Weiglein, "Sensitivity of the ILC to light Higgs masses",arXiv:1801.09662[hep-ph].