e University of Manchester
October 15, 2018
Status and prospects for the LHCb upgrade
Andrea Contu on behalf of The LHCb collaboration INFN Sezione di Cagliari, Italyand CERN, Switzerland
High-precision measurements performed by the LHCb collaborationhave opened a new era in charm physics. Several crucial measurements,particularly in spectroscopy, rare decays and CP violation, can benefitfrom the increased statistical power of an upgraded LHCb detector. Theupgrade of LHCb detector, its software infrastructure, and the impact oncharm physics are discussed in detail.PRESENTED AT The 6 th International Workshop on Charm Physics(CHARM 2013)Manchester, UK, 31 August – 4 September, 2013 The workshop was supported by the University of Manchester, IPPP, STFC, and IOP a r X i v : . [ h e p - e x ] N ov Introduction
The LHC has performed excellently during its first years of operation allowing thefour main experiments to collect large data samples at unprecedented centre-of-massenergies. The LHCb detector outperformed its design specification and played acrucial role in the advancement of charm physics. The LHCb measurements rangefrom the charm cross-section at √ s = 7 TeV [1], to direct and indirect CP violation,neutral charm meson mixing, spectroscopy, and rare decays. These measurementsexploit the large charm cross-section at the LHC and the outstanding performance ofthe trigger and reconstruction system of LHCb, which allowed unprecedented charmyields to be available for precision analyses. Charm physics plays a crucial role in theLHCb upgrade programme as well, in which the sensitivity for several key observablesis expected to reach or exceed the theoretical precision. In this paper, the LHCbupgrade, both from the hardware and software point of view, is outlined. Prospectsfor charm physics in the LHCb upgrade era are discussed and extrapolations of theexpected sensitivities for several observables are listed.The scientific value of these advances has been recognised by the CERN researchboard, which approved the upgrade of LHCb to be part of the long-term exploitationof the LHC. The first running phase of the LHC, with pp centre of mass energy of 7 and 8 TeV,ended at the beginning of 2013. Currently, the LHC machine and the four experimentsare in a 18-months shutdown (LS1) for maintenance and consolidation. Data takingwill be resumed at the beginning of 2015 with a pp center of mass energy of 13–14 TeV.The spacing between consecutive proton bunches circulating in the accelerator isforeseen to go from 50 ns to the nominal 25 ns, effectively doubling the pp collisionrate. From the beginning of 2018 a second long shutdown (LS2) is expected to lastabout a year, followed by three years of running up to 2022, after which a luminosityupgrade of the LHC is foreseen. It is noted that this schedule is likely to evolve withtime. The LHCb detector [2] is a single-arm forward spectrometer covering the pseudora-pidity range 2 < η <
5, designed for the study of particles containing b or c quarks.The detector includes a high-precision tracking system consisting of a silicon-stripvertex detector surrounding the pp interaction region, a large-area silicon-strip detec-tor located upstream of a dipole magnet with a bending power of about 4 Tm, and1hree stations of silicon-strip detectors and straw drift tubes placed downstream. Thecombined tracking system provides a momentum measurement with relative uncer-tainty that varies from 0.4% at 5 GeV /c to 0.6% at 100 GeV /c , and impact parameterresolution of 20 µ m for tracks with large transverse momentum. Different types ofcharged hadrons are distinguished by information from two ring-imaging Cherenkovdetectors [3]. Photon, electron and hadron candidates are identified by a calorimetersystem consisting of scintillating-pad and preshower detectors, an electromagneticcalorimeter and a hadronic calorimeter. Muons are identified by a system composedof alternating layers of iron and multiwire proportional chambers [4].The upgraded LHCb detector is expected to be installed in 2018, during LS2,and is currently being designed to perform as well as or better than the currentone at a higher instantaneous luminosity. The physics goal for the upgrade is toreach a sensitivity at the level of the theoretical prediction (or better) in several keyobservables. Therefore, in order to keep the same level of performance in harsherconditions, improvements in the trigger, reconstruction strategy, and detector tech-nology are mandatory. The total integrated luminosity collected at the end of theLHCb upgrade data taking is expected to reach 70 fb − . The current trigger scheme is based on a multi-stage approach with a first level,hardware-based, trigger and two software levels that have access to the full eventinformation (see Figure 1(a)). The output rate of the first hardware-based triggerlevel, which uses information on transverse momentum, p T , and transverse energy E T , is limited by a maximum bandwidth of 1.1 MHz. At higher luminosity, thisconstraint would require using tighter p T and E T cuts in hadronic triggers in orderfor the computing infrastructure to cope with the increased event rate and size. Thiswill also cause the trigger efficiency for hadronic channels to deteriorate, as shownin Figure 2. On the other hand, events that are selected by muonic triggers will bemostly unaffected since the muon system is already capable of sustaining a higherinstantaneous luminosity to some extent. The effect is even more pronounced forcharm hadrons, which are produced at a lower p T than b hadrons. The inefficiencyin for the hadronic triggers will also affect the charm yield achievable.A new trigger strategy for the upgrade is being studied in which the first levelhardware trigger is completely removed and the events are sent directly to a softwaretrigger running on a larger and more powerful CPU farm, as shown in Figure 1(b).This new scheme is not affected by the “bandwidth bottleneck” after the first triggerlevel so that the event rate that can be processed and stored on disk depends onlyon the capabilities of the CPU farm. The final event output rate is expected to be afactor of four larger than the current one.2 a) Current trigger system (b) Trigger system in the up-grade Figure 1: Overview of current and planned LHCb trigger system.Figure 2: Trigger yield for several B decays as a function of the instantaneousluminosity in the current trigger scheme. B → π + π − is represented by blacksquares, B → φγ by red triangles, B s → J/ψφ by green upside-down trianglesand B s → D + s K − by blue circles. 3 .2 Tracking system and RICH upgrade One of the obvious effects of the increased instantaneous luminosity is a higher occu-pancy and radiation dose for all the subdetectors. Layout and technology improve-ments are needed to cope with the harsher conditions of the upgrade. In the following,the main changes introduced for the upgraded detector are described. Particular focusis given to the tracking system and the RICH detectors.At higher luminosity, the particle flux increases dramatically in the regions closeto the beam axis, therefore a major upgrade is foreseen for the whole LHCb trackingsystem.The current VELO is based on semicircular silicon-strip sensors arranged in tworows that close around the interaction regions during data taking. While the movinglayout will be kept, the baseline choice for the upgrade consists of silicon pixel sensorswith an aggressive micro-channel cooling system. The new VELO sensor layout andthe micro-channel cooling scheme are shown in Figure 3. The sensor choice is driventhe necessity to reduce the occupancy allowing for a faster track reconstruction andlow fake-track rate. (a) New VELO silicon pixel sensor layout (b) Micro-channel cooling technology
Figure 3: The VELO sensors in the upgraded LHCb.The current Trigger Tracker (TT) will be replaced by the Upstream Tracker (UT).The UT is currently being designed to have a lower material budget (less than 5% X ),and to have higher granularity and extended angular coverage compared with the TT.A comparison of the performance for tracks reconstructed using only informationfrom the VELO and the UT (TT), the so-called upstream tracks , in the current de-tector and in the upgrade scenario, is shown in Figure 4. It is noted that combinationof information from the upgraded VELO and UT tracking leads to a considerableimprovement in p T resolution compared with the current VELO+TT.The high track multiplicity in the central region also drives the upgrade of thecurrent downstream tracking stations, located between the dipole magnet and theRICH2 detector. Several detector technologies are currently under study, with thebaseline choice being the replacement of the entire inner tracking system (composed4igure 4: Transverse momentum resolution for tracks reconstructed using only infor-mation from the vertex detector and the upstream tracker. The performance of thecurrent LHCb detector is shown in black, and the baseline upgrade configuration ingreen.of a silicon strip tracker in the inner region and a straw tube outer tracker) witha design known as the Sci-Fi detector (see Figure 5). The Sci-Fi detector exploitsscintillating fibres as the active material. The scintillation light from the fibres isread-out by silicon-based photo-multipliers.Figure 5: Options for the replacement of the current downstream tracking stations.From left to right: replacement of the silicon-strip detector and straw tubes in thecentral region (outer straw tubes are kept), scintillating fibres detector only in thecentral region (outer straw tubes are kept), entire downstream tracking station usingscintillating fibres technology (baseline).The current RICH system is composed of two detectors, RICH1 and RICH2,located upstream and downstream of the dipole magnet, respectively. In order to5over a wide momentum range, three radiators are used: aerogel (solid) and C F (gaseous) in RICH1, and CF in RICH2. In the upgrade, due to increased occupancythe aerogel, which covers the low momentum range 1–10 GeV /c , will be removed.Moreover, the current Hybrid-PhotoDetectors will be replaced by Multi-Anode photo-multipliers which will require new front-end electronics. The optics of both RICH1and RICH2 will also be optimised. LHCb has a broad upgrade physics programme of which charm measurements are animportant part. The large charm production cross-section at √ s = 7 TeV, recentlymeasured at LHCb [1], is predicted to increase by a factor of 1.8 at √ s = 14 TeV.Exploratory studies indicate that improvements in the trigger strategy could providean increase of a factor two for the trigger efficiency on charm hadronic decays. Theimprovement is even more pronounced in multibody decays. In the upgrade era, thecharm signal yield is expected to increase by a factor of about 3.6 per fb − . Since theintegrated luminosity recorded per year is expected to also increase by a factor 3.5per year, the total charm yield per year could increase by one order of magnitude. Charm production and spectroscopy are very active areas of research in LHCb. Re-cent studies of double-charm production observed as double-charmonium, charmo-nium and open charm, and double open charm [5] can in principle be extended tosimultaneous charmonium and bottomonium production in the upgrade era. Thesearch for new D sJ states [6] will also benefit enormously from an increased statistics.Improvements are also expected in studies of χ c (1 , , production, J/ψ polarisation,and charmed and doubly charmed baryons.
Charm rare decays are very powerful means to search for new mediators and couplings.The current overview of for D decays is shown in Figure 6. LHCb results on D → µ + µ − [7] (see Figure 6) and multibody decays, such as D +( s ) → π + µ + µ − and D +( s ) → π − µ + µ + [8] and D → π + π − µ + µ − [9], are already available and improved previousmeasurements by one or two orders of magnitude. Multibody decays may proceed viaan intermediate resonance, e.g. D +( s ) → π + φ and then φ → µ + µ − . In this context therare decay searches mentioned above are for the non-resonant modes. However, theresonant modes are themselves of interest for an angular analyses. There is particular6 % C . L . γγ µ + µ ρ µ + µ π π + π e + e K + K e + e η µ + µ φ e + e π e + e φ µ + µ ω µ + µ + π K e + e ρ µ + µ K e + e ( ) * K µ + µ π + π µ + µ e + e π + π e + e µ + µ ( ) * K µ + µ K + K e + e + π K µ + µ π e + e K e + e ω µ + µ η Figure 6: Current limits on rare D decays [10].interest in the study of forward-backward asymmetries, T-odd correlations and near-resonance effects. Decay modes with intermediate resonances in the dimuon mass canalready be seen in the current LHCb data sample. The statistical precision requiredfor angular analyses is expected to be available at the end of the LHCb upgrade.It is noted that hadronic modes are a dangerous background to rare decay searches,having a branching fractions O (10 ) larger than typical predictions for electroweak D meson decays in the SM. While this background is greatly reduced with informationfrom the muon chambers, decays in flight of high momentum pions into muons caneasily mimic a genuine muon directly from a D decay in such a way that hadronicdecays become an irreducible background. Since the discriminating power is currentlyreaching a limit, improvements in the muon identification in the upgrade are one keyingredient for the progress in this area. Charm mixing is already established by a series of complementary measurementsalthough considerable improvements are still needed in the precision with which themixing parameters x and y are known. The LHCb collaboration, analysing datacollected during the 2011 run only, made the first single measurement to exclude theno-mixing hypothesis to a level above five standard deviations [11]. The analysis,based on the study of the time-dependent ratio between wrong- (WS) and right-sign (RS) D → K ∓ π ± , is a perfect demonstration of the LHCb’s statistical power.The updated analysis based on the complete Run 1 LHCb data sample (3 fb − ), alsocontains the most precise determination of the mixing parameters x (cid:48) and y (cid:48) and asearch for CP violation [12]. 7nother observable which give access to the mixing parameters is y CP , defined asthe ratio between the effective lifetime for decays to CP -even eigenstate ( K + K − or π + π − ) and Cabibbo-favoured decays to the CP -mixed final state K − π + . The currentmeasurement from LHCb, based on a small data sample collected in 2010, proves thefeasibility of the measurement at hadron machines [13]. An updated measurement,which uses the 2011 dataset, is in progress. The large yields available in the upgradewill allow a more refined treatment of backgrounds that will reduce the systematicuncertainty affecting the measurement.Other mixing measurement under study within the LHCb collaboration include: • x + y using the time integrated WS/RS ratio of D → K + µ − ν decays • Direct access to x and y via a time-dependent Dalitz plot measurement of D → K S hh decays • Access to x (cid:48)(cid:48) and y (cid:48)(cid:48) via a time-dependent WS/RS Dalitz plot measurement of D → K + π + π The sensitivities expected for several mixing observables, extrapolated to an in-tegrated luminosity of 50 fb − (note that the expected luminosity has increased sincethese estimates were made in Ref.[14]), are summarised in Table 1.Decay Observable Exp sensitivity [ × − ] (stat only) D → KK y CP D → ππ y CP D → K + π − x (cid:48) , y (cid:48) D → K S ππ x , y D → K + µ − ν R M = x + y − [14]. C P violation
As well as the
CP V search in the time-dependent wrong-sign D → K + π − decaymentioned previously, LHCb is carring out a search for indirect CP violation in thecharm sector through the measurement of A Γ [15]. The parameter A Γ , defined as theasymmetry between the effective lifetimes of D decays into a CP eigenstate, is analmost clean measurement of indirect CP violation and can expressed as A Γ = 12 ( A m + A d ) y cos φ − x sin φ ≈ − a indCP − a dirCP y CP , (1)8here A m = 1 − | q/p | , A d = 1 − | A f /A f | and φ is the relative CP violating phasebetween q/p and A f /A f . In Eq. 1 it is manifest that this measurements benefits froma precise determination of the mixing parameters x and y , which are expected to beconstrained at a 10 − level in the upgrade. Since the overall precision on A Γ at theend of the upgraded LHCb data-taking is expected to be better than 10 − , a precisionindependent measurement of the direct CP violating component is necessary to probethe SM prediction for A Γ which is set to about 10 − .In addition to the mixing parameters, D → K S h + h − decays which give also accessto CP violating quantities such as | q/p | and φ , making this a “golden-channel” for theLHCb upgrade. These parameters are accessible via a the time-dependent evolution inthe K S ππ Dalitz plane. Two strategies are possible: an unbinned, model-dependentmeasurement in which a full amplitude fit is performed, and a model-independentmeasurement that instead uses prior experimental measurements of the average strongphase difference in regions of the Dalitz-plot (e.g. from CLEOc and BESIII). Althoughsuch decays suffer from a relatively low reconstruction efficiency in LHCb, mainlydue to the K S long lifetime, precise measurements of x , y , q/p and φ can alreadybe performed with the existing data samples and will be greatly improved in theupgrade. C P violation
Measurements of direct CP violation are challenge for experiments at hadron col-liders. In fact, several sources of asymmetry can bias the measurement such as theproduction asymmetry present in proton-proton collisions. Moreover, analyses canbe affected by detection asymmetry biases. Therefore, independent measurements ofproduction and detection asymmetries are a crucial ingredient for direct CP viola-tion searches in charm. These measurements are currently being performed withinthe LHCb collaboration [16, 17, 18] and will be pursued in the upgrade phase.It is interesting to note that if detection and production asymmetries are small,observables can be constructed in which they cancel at the first order. This factis exploited in the measurement of ∆ A CP = A CP ( K + K − ) − A CP ( π + π − ) in prompt[19] and semileptonic [20] decays performed by LHCb. The improved detector andthe larger statistics of the LHCb upgrade are therefore vital to reduce the statisticaland systematic uncertainties and shed light on the still unclear picture of direct CP violation in the charm sector.The sensitivities for several direct CP violating observables are given in Table 2,assuming an integrated luminosity of 50 fb − .9ecay Observable Exp sensitivity [ × − ] (stat only) D → KK, ππ ∆ A CP D + → K S K + A CP D + → K − K + π + A CP D + → πππ x , y D + → hhπ CP V in phases (0 . − . ◦ D + → hhπ CP V in fractions 0 . − . CP observables with 50 fb − [14]. The LHCb detector is performing excellently and is already exceeding its design expec-tations confirming the feasibility of charm physics at hadron colliders. The collabora-tion is active in many complementary analysis in the charm sector, and in particularsub-percent measurements of several CP quantities are expected to be already avail-able before the upgrade and will reach or even exceed the current theoretical precisionafter the upgrade. In the upgrade era, these studies will be further improved thanksto the increased statistics and the improvements in the hardware and software in-frastructure. In addition, the upgraded LHCb detector has tremendous potential fornew measurements in charm rare decays, production and spectroscopy. In parallel,ongoing efforts are focused on reducing possible sources of systematic uncertaintiesthat may limit the LHCb scope. Further, detailed and information on the LHCbupgrade is reported in [21, 22]. Acknowledgements
The text below are the acknowledgements as approved by the collaboration board.Extending the acknowledgements to include individuals from outside the collaborationwho have contributed to the analysis should be approved by the EB and, if possible,be included in the draft of first circulation.We express our gratitude to our colleagues in the CERN accelerator departmentsfor the excellent performance of the LHC. We thank the technical and adminis-trative staff at the LHCb institutes. We acknowledge support from CERN andfrom the national agencies: CAPES, CNPq, FAPERJ and FINEP (Brazil); NSFC(China); CNRS/IN2P3 and Region Auvergne (France); BMBF, DFG, HGF andMPG (Germany); SFI (Ireland); INFN (Italy); FOM and NWO (The Netherlands);SCSR (Poland); MEN/IFA (Romania); MinES, Rosatom, RFBR and NRC “Kurcha-tov Institute” (Russia); MinECo, XuntaGal and GENCAT (Spain); SNSF and SER(Switzerland); NAS Ukraine (Ukraine); STFC (United Kingdom); NSF (USA). We10lso acknowledge the support received from the ERC under FP7. The Tier1 com-puting centres are supported by IN2P3 (France), KIT and BMBF (Germany), INFN(Italy), NWO and SURF (The Netherlands), PIC (Spain), GridPP (United King-dom). We are thankful for the computing resources put at our disposal by YandexLLC (Russia), as well as to the communities behind the multiple open source softwarepackages that we depend on.
References [1] R. Aaij et al. [LHCb Collaboration], Nucl. Phys. B (2013) 1[2] LHCb collaboration, A. A. Alves Jr. et al. , JINST (2008) S08005[3] M. Adinolfi et al. , Eur. Phys. J. C73 (2013) 2431,[4] A. A. Alves Jr. et al. , JINST (2013) P02022[5] R. Aaij et al. [LHCb Collaboration], JHEP (2012) 141[6] R. Aaij et al. [LHCb Collaboration], JHEP (2013) 145[7] R. Aaij et al. [LHCb Collaboration], Phys. Lett. B (2013) 15[8] R. Aaij et al. [LHCb Collaboration], Phys. Lett. B (2013) 203[9] R. Aaij et al. [LHCb Collaboration], arXiv:1310.2535 [hep-ex].[10] Y. Amhis et al. [Heavy Flavor Averaging Group Collaboration], arXiv:1207.1158[hep-ex].[11] R. Aaij et al. [LHCb Collaboration], Phys. Rev. Lett. (2013) 10, 101802[12] R. Aaij et al. (LHCb Collaboration), arXiv:1309.6534 [hep-ex].[13] R. Aaij et al. [LHCb Collaboration], JHEP (2012) 129[14] R. Aaij et al. [LHCb Collaboration], Eur. Phys. J. C (2013) 2373[15] R. Aaij et al. [LHCb Collaboration], arXiv:1310.7201 [hep-ex].[16] RAaij et al. [LHCb Collaboration], Phys. Lett. B (2012) 186[17] RAaij et al. [LHCb Collaboration], Phys. Lett. B (2013) 902[arXiv:1210.4112 [Unknown]].[18] R. Aaij et al. [LHCb Collaboration], LHCb-CONF-2013-0231119] R. Aaij et al. [LHCb Collaboration], LHCb-CONF-2013-003[20] R. Aaij et al. [LHCb Collaboration], Phys. Lett. B (2013) 33[21] CERN, Technical report, CERN-LHCC-2011-001. LHCC-I-018[22] I. Bediaga et al.et al.